W0530 22:18:25.521000 714796 torch/distributed/run.py:862] 
W0530 22:18:25.521000 714796 torch/distributed/run.py:862] *****************************************
W0530 22:18:25.521000 714796 torch/distributed/run.py:862] Setting OMP_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. 
W0530 22:18:25.521000 714796 torch/distributed/run.py:862] *****************************************
Config: {'project': 'thoughtformer', 'name': 'star-datacurr-only-4gpu-bs32-lr1e-5', 'only_eval': False, 'method': 'thoughtformer', 'data_curriculum': True, 'thought_curriculum': False, 'staging': 'adaptive', 'init_thought_stage': 1, 'init_data_stage': 1, 'patience': 5, 'staging_threshold': 0.9, 'c_thought': 1, 'max_latent_stage': 10, 'save_only_improve': False, 'model_id': 'Qwen/Qwen3-0.6B-Base', 'load_model_path': 'None', 'seed': 0, 'resume': 0, 'bf16': False, 'train_path': 'data/star_k10_L10_1000_train.json', 'val_path': 'data/star_k10_L10_100_valid.json', 'reset_optimizer': False, 'lr_schedule': 'none', 'batch_size_training': 4, 'eval_only_trained': False, 'eval_batch_size': 8, 'eval_every': 1, 'debug': False, 'gradient_accumulation_steps': 2, 'num_epochs': 100, 'lr': 1e-05, 'weight_decay': 0.01, 'group': 'star_datacurr_only', 'tags': ['star', 'k10', 'L10', 'data_curr_only', 'fixed_thought_10', 'fixed_lr', 'no_uniform'], 'run_type': 'pilot'}
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/c10d_logger.py:83: UserWarning: barrier(): using the device under current context. You can specify `device_id` in `init_process_group` to mute this warning.
  return func(*args, **kwargs)
[rank0]:[W530 22:18:31.109903417 ProcessGroupNCCL.cpp:5324] Guessing device ID based on global rank. This can cause a hang if rank to GPU mapping is heterogeneous. You can specify device_id in init_process_group()

Loading weights:   0%|          | 0/310 [00:00<?, ?it/s]
Loading weights:   0%|          | 0/310 [00:00<?, ?it/s]
Loading weights:   0%|          | 0/310 [00:00<?, ?it/s]
Loading weights:   0%|          | 0/310 [00:00<?, ?it/s]
Loading weights:  47%|████▋     | 147/310 [00:00<00:00, 1467.30it/s]
Loading weights:  50%|████▉     | 154/310 [00:00<00:00, 1526.30it/s]
Loading weights:  49%|████▉     | 152/310 [00:00<00:00, 1503.44it/s]
Loading weights:  48%|████▊     | 148/310 [00:00<00:00, 1476.45it/s]
Loading weights:  99%|█████████▉| 308/310 [00:00<00:00, 1539.67it/s]
Loading weights: 100%|██████████| 310/310 [00:00<00:00, 1549.26it/s]
Loading weights: 100%|██████████| 310/310 [00:00<00:00, 1550.63it/s]


Loading weights: 100%|██████████| 310/310 [00:00<00:00, 1536.51it/s]

Loading weights:  99%|█████████▉| 308/310 [00:00<00:00, 1540.37it/s]
Loading weights: 100%|██████████| 310/310 [00:00<00:00, 1538.91it/s]
Running FSDP on rank = 2, world size = 4
Running FSDP on rank = 0, world size = 4
Running FSDP on rank = 1, world size = 4
Running FSDP on rank = 3, world size = 4
FullyShardedDataParallel(
  (_fsdp_wrapped_module): Coconut(
    (base_causallm): Qwen3ForCausalLM(
      (model): Qwen3Model(
        (embed_tokens): Embedding(151672, 1024)
        (layers): ModuleList(
          (0-27): 28 x FullyShardedDataParallel(
            (_fsdp_wrapped_module): Qwen3DecoderLayer(
              (self_attn): Qwen3Attention(
                (q_proj): Linear(in_features=1024, out_features=2048, bias=False)
                (k_proj): Linear(in_features=1024, out_features=1024, bias=False)
                (v_proj): Linear(in_features=1024, out_features=1024, bias=False)
                (o_proj): Linear(in_features=2048, out_features=1024, bias=False)
                (q_norm): Qwen3RMSNorm((128,), eps=1e-06)
                (k_norm): Qwen3RMSNorm((128,), eps=1e-06)
              )
              (mlp): Qwen3MLP(
                (gate_proj): Linear(in_features=1024, out_features=3072, bias=False)
                (up_proj): Linear(in_features=1024, out_features=3072, bias=False)
                (down_proj): Linear(in_features=3072, out_features=1024, bias=False)
                (act_fn): SiLUActivation()
              )
              (input_layernorm): Qwen3RMSNorm((1024,), eps=1e-06)
              (post_attention_layernorm): Qwen3RMSNorm((1024,), eps=1e-06)
            )
          )
        )
        (norm): Qwen3RMSNorm((1024,), eps=1e-06)
        (rotary_emb): Qwen3RotaryEmbedding()
      )
      (lm_head): Linear(in_features=1024, out_features=151936, bias=False)
    )
    (embedding): Embedding(151672, 1024)
  )
)

Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):   3%|▎         | 29/1000 [00:01<00:55, 17.54 examples/s]
Map (num_proc=4):   9%|▉         | 90/1000 [00:01<00:14, 64.61 examples/s]
Map (num_proc=4):  15%|█▌        | 152/1000 [00:01<00:07, 120.90 examples/s]
Map (num_proc=4):  21%|██▏       | 214/1000 [00:01<00:04, 183.71 examples/s]
Map (num_proc=4):  28%|██▊       | 280/1000 [00:02<00:05, 128.26 examples/s]
Map (num_proc=4):  34%|███▍      | 342/1000 [00:02<00:03, 176.45 examples/s]
Map (num_proc=4):  40%|████      | 404/1000 [00:02<00:02, 230.71 examples/s]
Map (num_proc=4):  50%|████▉     | 497/1000 [00:03<00:01, 310.99 examples/s]
Map (num_proc=4):  56%|█████▌    | 559/1000 [00:03<00:02, 175.14 examples/s]
Map (num_proc=4):  62%|██████▏   | 621/1000 [00:03<00:01, 219.27 examples/s]
Map (num_proc=4):  71%|███████▏  | 714/1000 [00:04<00:00, 291.09 examples/s]
Map (num_proc=4):  78%|███████▊  | 780/1000 [00:04<00:01, 179.86 examples/s]
Map (num_proc=4):  84%|████████▍ | 842/1000 [00:04<00:00, 222.52 examples/s]
Map (num_proc=4):  90%|█████████ | 904/1000 [00:05<00:00, 270.86 examples/s]
Map (num_proc=4):  97%|█████████▋| 966/1000 [00:05<00:00, 322.16 examples/s]
Map (num_proc=4): 100%|██████████| 1000/1000 [00:05<00:00, 188.54 examples/s]

Map (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Map (num_proc=4):   0%|          | 29/10000 [00:02<13:47, 12.05 examples/s]
Map (num_proc=4):   1%|          | 121/10000 [00:02<02:41, 61.14 examples/s]
Map (num_proc=4):   2%|▏         | 212/10000 [00:02<01:22, 117.93 examples/s]
Map (num_proc=4):   3%|▎         | 305/10000 [00:02<00:54, 177.55 examples/s]
Map (num_proc=4):   4%|▍         | 391/10000 [00:03<00:38, 247.73 examples/s]
Map (num_proc=4):   4%|▍         | 450/10000 [00:03<00:32, 292.02 examples/s]
Map (num_proc=4):   5%|▌         | 509/10000 [00:03<00:29, 320.82 examples/s]
Map (num_proc=4):   6%|▌         | 593/10000 [00:03<00:23, 394.23 examples/s]
Map (num_proc=4):   7%|▋         | 673/10000 [00:03<00:21, 426.05 examples/s]
Map (num_proc=4):   7%|▋         | 731/10000 [00:03<00:20, 455.65 examples/s]
Map (num_proc=4):   8%|▊         | 789/10000 [00:03<00:19, 479.07 examples/s]
Map (num_proc=4):   9%|▉         | 877/10000 [00:03<00:16, 549.97 examples/s]
Map (num_proc=4):  10%|▉         | 988/10000 [00:03<00:13, 687.16 examples/s]
Map (num_proc=4):  11%|█         | 1107/10000 [00:04<00:10, 814.36 examples/s]
Map (num_proc=4):  12%|█▏        | 1197/10000 [00:04<00:10, 833.17 examples/s]
Map (num_proc=4):  13%|█▎        | 1298/10000 [00:04<00:09, 875.73 examples/s]
Map (num_proc=4):  14%|█▍        | 1417/10000 [00:04<00:09, 932.15 examples/s]
Map (num_proc=4):  16%|█▌        | 1558/10000 [00:04<00:08, 1023.77 examples/s]
Map (num_proc=4):  17%|█▋        | 1674/10000 [00:04<00:07, 1060.07 examples/s]
Map (num_proc=4):  18%|█▊        | 1788/10000 [00:04<00:08, 988.73 examples/s] 
Map (num_proc=4):  19%|█▉        | 1925/10000 [00:04<00:07, 1072.25 examples/s]
Map (num_proc=4):  21%|██        | 2066/10000 [00:04<00:07, 1084.27 examples/s]
Map (num_proc=4):  22%|██▏       | 2204/10000 [00:05<00:07, 1090.67 examples/s]
Map (num_proc=4):  23%|██▎       | 2318/10000 [00:05<00:06, 1102.49 examples/s]
Map (num_proc=4):  24%|██▍       | 2432/10000 [00:05<00:06, 1108.61 examples/s]
Map (num_proc=4):  26%|██▌       | 2602/10000 [00:05<00:05, 1266.15 examples/s]
Map (num_proc=4):  28%|██▊       | 2782/10000 [00:05<00:05, 1406.07 examples/s]
Map (num_proc=4):  30%|██▉       | 2961/10000 [00:05<00:04, 1448.16 examples/s]
Map (num_proc=4):  31%|███▏      | 3132/10000 [00:05<00:04, 1454.56 examples/s]
Map (num_proc=4):  34%|███▎      | 3365/10000 [00:05<00:04, 1579.51 examples/s]
Map (num_proc=4):  35%|███▌      | 3537/10000 [00:05<00:04, 1612.00 examples/s]
Map (num_proc=4):  37%|███▋      | 3720/10000 [00:06<00:03, 1592.76 examples/s]
Map (num_proc=4):  39%|███▉      | 3888/10000 [00:06<00:03, 1609.56 examples/s]
Map (num_proc=4):  41%|████      | 4058/10000 [00:06<00:03, 1628.61 examples/s]
Map (num_proc=4):  42%|████▏     | 4232/10000 [00:06<00:03, 1650.98 examples/s]
Map (num_proc=4):  44%|████▍     | 4398/10000 [00:06<00:03, 1645.93 examples/s]
Map (num_proc=4):  46%|████▌     | 4572/10000 [00:06<00:03, 1665.51 examples/s]
Map (num_proc=4):  47%|████▋     | 4744/10000 [00:06<00:03, 1673.84 examples/s]
Map (num_proc=4):  49%|████▉     | 4949/10000 [00:06<00:02, 1781.80 examples/s]
Map (num_proc=4):  52%|█████▏    | 5177/10000 [00:06<00:02, 1925.09 examples/s]
Map (num_proc=4):  54%|█████▍    | 5375/10000 [00:06<00:02, 1936.47 examples/s]
Map (num_proc=4):  56%|█████▌    | 5593/10000 [00:07<00:02, 1538.24 examples/s]
Map (num_proc=4):  58%|█████▊    | 5784/10000 [00:07<00:02, 1613.80 examples/s]
Map (num_proc=4):  60%|█████▉    | 5984/10000 [00:07<00:02, 1705.57 examples/s]
Map (num_proc=4):  62%|██████▏   | 6190/10000 [00:07<00:02, 1585.87 examples/s]
Map (num_proc=4):  64%|██████▎   | 6363/10000 [00:07<00:02, 1612.68 examples/s]
Map (num_proc=4):  65%|██████▌   | 6540/10000 [00:07<00:02, 1645.89 examples/s]
Map (num_proc=4):  67%|██████▋   | 6719/10000 [00:07<00:01, 1679.90 examples/s]
Map (num_proc=4):  69%|██████▉   | 6895/10000 [00:07<00:01, 1690.05 examples/s]
Map (num_proc=4):  71%|███████   | 7076/10000 [00:08<00:01, 1714.91 examples/s]
Map (num_proc=4):  73%|███████▎  | 7257/10000 [00:08<00:01, 1734.11 examples/s]
Map (num_proc=4):  74%|███████▍  | 7439/10000 [00:08<00:01, 1750.73 examples/s]
Map (num_proc=4):  76%|███████▋  | 7642/10000 [00:08<00:01, 1716.97 examples/s]
Map (num_proc=4):  78%|███████▊  | 7832/10000 [00:08<00:01, 1438.36 examples/s]
Map (num_proc=4):  80%|███████▉  | 7995/10000 [00:08<00:01, 1275.50 examples/s]
Map (num_proc=4):  81%|████████▏ | 8136/10000 [00:08<00:01, 1209.12 examples/s]
Map (num_proc=4):  83%|████████▎ | 8282/10000 [00:09<00:01, 1092.00 examples/s]
Map (num_proc=4):  84%|████████▍ | 8400/10000 [00:09<00:01, 1100.08 examples/s]
Map (num_proc=4):  85%|████████▌ | 8521/10000 [00:09<00:01, 1121.32 examples/s]
Map (num_proc=4):  86%|████████▋ | 8642/10000 [00:09<00:01, 1137.78 examples/s]
Map (num_proc=4):  88%|████████▊ | 8764/10000 [00:09<00:01, 1154.51 examples/s]
Map (num_proc=4):  89%|████████▉ | 8886/10000 [00:09<00:00, 1167.25 examples/s]
Map (num_proc=4):  90%|█████████ | 9010/10000 [00:09<00:00, 1177.65 examples/s]
Map (num_proc=4):  91%|█████████▏| 9133/10000 [00:09<00:00, 1187.89 examples/s]
Map (num_proc=4):  93%|█████████▎| 9270/10000 [00:10<00:00, 817.92 examples/s] 
Map (num_proc=4):  94%|█████████▍| 9391/10000 [00:10<00:00, 848.42 examples/s]
Map (num_proc=4):  95%|█████████▌| 9500/10000 [00:10<00:00, 662.37 examples/s]
Map (num_proc=4):  96%|█████████▌| 9589/10000 [00:10<00:00, 641.20 examples/s]
Map (num_proc=4):  97%|█████████▋| 9679/10000 [00:10<00:00, 627.65 examples/s]
Map (num_proc=4):  98%|█████████▊| 9770/10000 [00:10<00:00, 619.73 examples/s]
Map (num_proc=4):  99%|█████████▊| 9863/10000 [00:11<00:00, 616.02 examples/s]
Map (num_proc=4): 100%|█████████▉| 9956/10000 [00:11<00:00, 613.34 examples/s]
Map (num_proc=4): 100%|██████████| 10000/10000 [00:11<00:00, 860.78 examples/s]
wandb: [wandb.login()] Loaded credentials for https://api.wandb.ai from /home/ubuntu/.netrc.
wandb: Currently logged in as: seyedparsa to https://api.wandb.ai. Use `wandb login --relogin` to force relogin

Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):  17%|█▋        | 171/1000 [00:00<00:00, 1145.34 examples/s]
Map (num_proc=4):  17%|█▋        | 171/1000 [00:00<00:00, 1143.56 examples/s]
Map (num_proc=4):  17%|█▋        | 170/1000 [00:00<00:00, 1116.05 examples/s]
Map (num_proc=4):  84%|████████▍ | 843/1000 [00:00<00:00, 3475.34 examples/s]
Map (num_proc=4):  84%|████████▍ | 845/1000 [00:00<00:00, 3419.94 examples/s]
Map (num_proc=4):  84%|████████▍ | 842/1000 [00:00<00:00, 3264.71 examples/s]
Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2276.09 examples/s]

Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2230.47 examples/s]

Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2165.19 examples/s]
wandb: setting up run efv4go42

Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]wandb: Tracking run with wandb version 0.25.1
wandb: Run data is saved locally in /home/ubuntu/thoughtformer/wandb/run-20260530_221859-efv4go42
wandb: Run `wandb offline` to turn off syncing.
wandb: Syncing run star-datacurr-only-4gpu-bs32-lr1e-5_n1000_qwen3-0.6b-base_thoughtformer_lr1e-5_no-reset_dc_no-tc_adaptive-03b89ee9
wandb: ⭐️ View project at https://wandb.ai/seyedparsa/thoughtformer
wandb: 🚀 View run at https://wandb.ai/seyedparsa/thoughtformer/runs/efv4go42

============================================================
EPOCH 0/100 (thought_stage=10, data_stage=1)
============================================================
  thought_stage=10, c_thought=1, max_difficulty=1

Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:04, 2232.17 examples/s]
Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:04, 2155.01 examples/s]
Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:04, 2171.63 examples/s]
Filter (num_proc=4):  20%|██        | 2000/10000 [00:00<00:01, 4035.73 examples/s]
Map (num_proc=4):  18%|█▊        | 177/1000 [00:00<00:00, 1226.14 examples/s]
Filter (num_proc=4):  20%|██        | 2000/10000 [00:00<00:02, 3577.79 examples/s]
Filter (num_proc=4):  20%|██        | 2000/10000 [00:00<00:02, 3893.36 examples/s]
Filter (num_proc=4):  30%|███       | 3000/10000 [00:00<00:01, 5464.02 examples/s]
Map (num_proc=4):  76%|███████▋  | 765/1000 [00:00<00:00, 3188.53 examples/s]
Filter (num_proc=4):  30%|███       | 3000/10000 [00:00<00:01, 5259.27 examples/s]
Filter (num_proc=4):  50%|█████     | 5000/10000 [00:00<00:00, 9119.55 examples/s]
Filter (num_proc=4):  45%|████▌     | 4500/10000 [00:00<00:00, 7080.64 examples/s]
Filter (num_proc=4):  60%|██████    | 6000/10000 [00:00<00:00, 10983.25 examples/s]
Filter (num_proc=4):  75%|███████▌  | 7500/10000 [00:00<00:00, 12604.83 examples/s]
Filter (num_proc=4):  65%|██████▌   | 6500/10000 [00:00<00:00, 9501.77 examples/s]
Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 1842.96 examples/s]

Filter (num_proc=4):  80%|████████  | 8000/10000 [00:00<00:00, 12708.54 examples/s]
Filter (num_proc=4):  95%|█████████▌| 9500/10000 [00:01<00:00, 14019.95 examples/s]
Filter (num_proc=4):  85%|████████▌ | 8500/10000 [00:01<00:00, 11507.17 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 11666.71 examples/s]
Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 8143.47 examples/s]

Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 8074.84 examples/s] 

Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 8378.47 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 6910.26 examples/s]

Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:04, 2021.36 examples/s]
Filter (num_proc=4):  20%|██        | 2000/10000 [00:00<00:02, 3557.98 examples/s]
Filter (num_proc=4):  40%|████      | 4000/10000 [00:00<00:00, 7080.81 examples/s]
Filter (num_proc=4):  55%|█████▌    | 5500/10000 [00:00<00:00, 8872.48 examples/s]
Filter (num_proc=4):  70%|███████   | 7000/10000 [00:00<00:00, 9616.83 examples/s]
Filter (num_proc=4):  95%|█████████▌| 9500/10000 [00:01<00:00, 12838.89 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 7653.03 examples/s]

Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):  15%|█▌        | 151/1000 [00:00<00:03, 256.24 examples/s]
Map (num_proc=4):  40%|████      | 403/1000 [00:00<00:01, 450.13 examples/s]
Map (num_proc=4):  65%|██████▌   | 652/1000 [00:01<00:00, 563.70 examples/s]
Map (num_proc=4):  75%|███████▌  | 750/1000 [00:01<00:00, 621.08 examples/s]
Map (num_proc=4):  90%|█████████ | 905/1000 [00:01<00:00, 615.12 examples/s]
Map (num_proc=4): 100%|██████████| 1000/1000 [00:01<00:00, 664.06 examples/s]
Map (num_proc=4): 100%|██████████| 1000/1000 [00:01<00:00, 522.95 examples/s]

Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 1695.76 examples/s]
Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 1727.21 examples/s]
Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 1667.17 examples/s]
Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 1617.13 examples/s]
Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 3243.09 examples/s]

Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 3362.85 examples/s]

Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 3270.56 examples/s]

Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 3082.46 examples/s]

Map (num_proc=4):   0%|          | 0/100 [00:00<?, ? examples/s]
Map (num_proc=4):  25%|██▌       | 25/100 [00:00<00:00, 176.29 examples/s]
Map (num_proc=4):  75%|███████▌  | 75/100 [00:00<00:00, 321.75 examples/s]
Map (num_proc=4): 100%|██████████| 100/100 [00:00<00:00, 235.47 examples/s]

Training Epoch: 1:   0%|          | 0/31 [00:00<?, ?it/s]
Training Epoch: 1:   0%|          | 0/31 [00:00<?, ?it/s]
Training Epoch: 1:   0%|          | 0/31 [00:00<?, ?it/s]
Training Epoch: 1:   0%|          | 0/31 [00:00<?, ?it/s]logging training data
rank 0 | gpu 0 | alloc=3.09 GB | reserved=46.37 GB | max=35.24 GBrank 3 | gpu 3 | alloc=3.09 GB | reserved=46.37 GB | max=35.24 GB

rank 1 | gpu 1 | alloc=3.09 GB | reserved=46.38 GB | max=35.25 GB
rank 2 | gpu 2 | alloc=3.09 GB | reserved=46.38 GB | max=35.25 GB
Training Epoch: 1/100, batch 0/63 completed (loss: 8.2392:   0%|          | 0/31 [00:04<?, ?it/s]

Training Epoch: 1/100, batch 0/63 completed (loss: 8.4877:   0%|          | 0/31 [00:04<?, ?it/s]
Training Epoch: 1/100, batch 0/63 completed (loss: 8.0464:   0%|          | 0/31 [00:04<?, ?it/s]
Training Epoch: 1/100, batch 0/63 completed (loss: 8.4083:   0%|          | 0/31 [00:04<?, ?it/s]
Training Epoch: 1/100, batch 0/63 completed (loss: 8.2392:   3%|▎         | 1/31 [00:07<03:40,  7.33s/it]
Training Epoch: 1/100, batch 0/63 completed (loss: 8.4877:   3%|▎         | 1/31 [00:07<03:40,  7.33s/it]
Training Epoch: 1/100, batch 1/63 completed (loss: 8.2812:   3%|▎         | 1/31 [00:07<03:40,  7.33s/it]
Training Epoch: 1/100, batch 0/63 completed (loss: 8.4083:   3%|▎         | 1/31 [00:07<03:40,  7.34s/it]
Training Epoch: 1/100, batch 1/63 completed (loss: 8.9582:   3%|▎         | 1/31 [00:07<03:40,  7.33s/it]
Training Epoch: 1/100, batch 0/63 completed (loss: 8.0464:   3%|▎         | 1/31 [00:07<03:40,  7.34s/it]
Training Epoch: 1/100, batch 1/63 completed (loss: 7.7634:   3%|▎         | 1/31 [00:07<03:40,  7.34s/it]
Training Epoch: 1/100, batch 1/63 completed (loss: 8.4138:   3%|▎         | 1/31 [00:07<03:40,  7.34s/it]
Training Epoch: 1/100, batch 2/63 completed (loss: 7.2358:   3%|▎         | 1/31 [00:10<03:40,  7.34s/it]
Training Epoch: 1/100, batch 2/63 completed (loss: 8.0006:   3%|▎         | 1/31 [00:10<03:40,  7.33s/it]
Training Epoch: 1/100, batch 2/63 completed (loss: 7.7816:   3%|▎         | 1/31 [00:10<03:40,  7.33s/it]
Training Epoch: 1/100, batch 2/63 completed (loss: 7.7439:   3%|▎         | 1/31 [00:10<03:40,  7.34s/it]
Training Epoch: 1/100, batch 2/63 completed (loss: 7.7439:   6%|▋         | 2/31 [00:12<03:04,  6.35s/it]
Training Epoch: 1/100, batch 2/63 completed (loss: 8.0006:   6%|▋         | 2/31 [00:12<03:04,  6.35s/it]
Training Epoch: 1/100, batch 2/63 completed (loss: 7.2358:   6%|▋         | 2/31 [00:13<03:04,  6.36s/it]
Training Epoch: 1/100, batch 2/63 completed (loss: 7.7816:   6%|▋         | 2/31 [00:13<03:04,  6.36s/it]
Training Epoch: 1/100, batch 3/63 completed (loss: 7.5047:   6%|▋         | 2/31 [00:13<03:04,  6.36s/it]
Training Epoch: 1/100, batch 3/63 completed (loss: 7.9842:   6%|▋         | 2/31 [00:13<03:04,  6.35s/it]
Training Epoch: 1/100, batch 3/63 completed (loss: 7.9941:   6%|▋         | 2/31 [00:13<03:04,  6.36s/it]
Training Epoch: 1/100, batch 3/63 completed (loss: 7.6985:   6%|▋         | 2/31 [00:13<03:04,  6.35s/it]
Training Epoch: 1/100, batch 4/63 completed (loss: 6.7702:   6%|▋         | 2/31 [00:15<03:04,  6.35s/it]
Training Epoch: 1/100, batch 4/63 completed (loss: 6.6351:   6%|▋         | 2/31 [00:15<03:04,  6.36s/it]
Training Epoch: 1/100, batch 4/63 completed (loss: 6.6331:   6%|▋         | 2/31 [00:15<03:04,  6.36s/it]
Training Epoch: 1/100, batch 4/63 completed (loss: 7.2348:   6%|▋         | 2/31 [00:15<03:04,  6.35s/it]
Training Epoch: 1/100, batch 4/63 completed (loss: 6.6331:  10%|▉         | 3/31 [00:18<02:48,  6.03s/it]
Training Epoch: 1/100, batch 4/63 completed (loss: 7.2348:  10%|▉         | 3/31 [00:18<02:49,  6.04s/it]
Training Epoch: 1/100, batch 4/63 completed (loss: 6.7702:  10%|▉         | 3/31 [00:18<02:49,  6.04s/it]
Training Epoch: 1/100, batch 4/63 completed (loss: 6.6351:  10%|▉         | 3/31 [00:18<02:48,  6.03s/it]
Training Epoch: 1/100, batch 5/63 completed (loss: 6.9904:  10%|▉         | 3/31 [00:18<02:49,  6.04s/it]
Training Epoch: 1/100, batch 5/63 completed (loss: 6.4213:  10%|▉         | 3/31 [00:18<02:48,  6.03s/it]
Training Epoch: 1/100, batch 5/63 completed (loss: 6.5549:  10%|▉         | 3/31 [00:18<02:48,  6.03s/it]
Training Epoch: 1/100, batch 5/63 completed (loss: 6.5463:  10%|▉         | 3/31 [00:18<02:49,  6.04s/it]
Training Epoch: 1/100, batch 6/63 completed (loss: 5.2326:  10%|▉         | 3/31 [00:21<02:48,  6.03s/it]
Training Epoch: 1/100, batch 6/63 completed (loss: 5.2751:  10%|▉         | 3/31 [00:21<02:49,  6.04s/it]
Training Epoch: 1/100, batch 6/63 completed (loss: 5.5354:  10%|▉         | 3/31 [00:21<02:48,  6.03s/it]
Training Epoch: 1/100, batch 6/63 completed (loss: 5.0248:  10%|▉         | 3/31 [00:21<02:49,  6.04s/it]
Training Epoch: 1/100, batch 6/63 completed (loss: 5.2326:  13%|█▎        | 4/31 [00:24<02:39,  5.89s/it]
Training Epoch: 1/100, batch 6/63 completed (loss: 5.5354:  13%|█▎        | 4/31 [00:24<02:39,  5.89s/it]
Training Epoch: 1/100, batch 6/63 completed (loss: 5.2751:  13%|█▎        | 4/31 [00:24<02:39,  5.90s/it]
Training Epoch: 1/100, batch 6/63 completed (loss: 5.0248:  13%|█▎        | 4/31 [00:24<02:39,  5.90s/it]
Training Epoch: 1/100, batch 7/63 completed (loss: 5.3541:  13%|█▎        | 4/31 [00:24<02:39,  5.90s/it]
Training Epoch: 1/100, batch 7/63 completed (loss: 5.3605:  13%|█▎        | 4/31 [00:24<02:39,  5.89s/it]
Training Epoch: 1/100, batch 7/63 completed (loss: 5.2237:  13%|█▎        | 4/31 [00:24<02:39,  5.89s/it]
Training Epoch: 1/100, batch 7/63 completed (loss: 5.1932:  13%|█▎        | 4/31 [00:24<02:39,  5.90s/it]
Training Epoch: 1/100, batch 8/63 completed (loss: 2.9031:  13%|█▎        | 4/31 [00:27<02:39,  5.89s/it]
Training Epoch: 1/100, batch 8/63 completed (loss: 3.1456:  13%|█▎        | 4/31 [00:27<02:39,  5.89s/it]
Training Epoch: 1/100, batch 8/63 completed (loss: 3.5825:  13%|█▎        | 4/31 [00:27<02:39,  5.90s/it]
Training Epoch: 1/100, batch 8/63 completed (loss: 3.1999:  13%|█▎        | 4/31 [00:27<02:39,  5.90s/it]
Training Epoch: 1/100, batch 8/63 completed (loss: 3.1456:  16%|█▌        | 5/31 [00:30<02:31,  5.81s/it]
Training Epoch: 1/100, batch 8/63 completed (loss: 2.9031:  16%|█▌        | 5/31 [00:30<02:31,  5.81s/it]
Training Epoch: 1/100, batch 8/63 completed (loss: 3.5825:  16%|█▌        | 5/31 [00:30<02:31,  5.81s/it]
Training Epoch: 1/100, batch 8/63 completed (loss: 3.1999:  16%|█▌        | 5/31 [00:30<02:31,  5.81s/it]
Training Epoch: 1/100, batch 9/63 completed (loss: 3.6773:  16%|█▌        | 5/31 [00:30<02:31,  5.81s/it]
Training Epoch: 1/100, batch 9/63 completed (loss: 3.3105:  16%|█▌        | 5/31 [00:30<02:31,  5.81s/it]
Training Epoch: 1/100, batch 9/63 completed (loss: 3.1144:  16%|█▌        | 5/31 [00:30<02:31,  5.81s/it]
Training Epoch: 1/100, batch 9/63 completed (loss: 3.7113:  16%|█▌        | 5/31 [00:30<02:31,  5.81s/it]
Training Epoch: 1/100, batch 10/63 completed (loss: 2.2151:  16%|█▌        | 5/31 [00:32<02:31,  5.81s/it]
Training Epoch: 1/100, batch 10/63 completed (loss: 2.4314:  16%|█▌        | 5/31 [00:32<02:31,  5.81s/it]
Training Epoch: 1/100, batch 10/63 completed (loss: 1.9929:  16%|█▌        | 5/31 [00:32<02:31,  5.81s/it]
Training Epoch: 1/100, batch 10/63 completed (loss: 2.2499:  16%|█▌        | 5/31 [00:32<02:31,  5.81s/it]
Training Epoch: 1/100, batch 10/63 completed (loss: 2.4314:  19%|█▉        | 6/31 [00:35<02:24,  5.76s/it]
Training Epoch: 1/100, batch 10/63 completed (loss: 1.9929:  19%|█▉        | 6/31 [00:35<02:24,  5.76s/it]
Training Epoch: 1/100, batch 10/63 completed (loss: 2.2499:  19%|█▉        | 6/31 [00:35<02:24,  5.77s/it]
Training Epoch: 1/100, batch 10/63 completed (loss: 2.2151:  19%|█▉        | 6/31 [00:35<02:24,  5.76s/it]
Training Epoch: 1/100, batch 11/63 completed (loss: 2.3289:  19%|█▉        | 6/31 [00:35<02:24,  5.76s/it]
Training Epoch: 1/100, batch 11/63 completed (loss: 2.5682:  19%|█▉        | 6/31 [00:35<02:24,  5.76s/it]
Training Epoch: 1/100, batch 11/63 completed (loss: 2.4741:  19%|█▉        | 6/31 [00:35<02:24,  5.76s/it]
Training Epoch: 1/100, batch 11/63 completed (loss: 2.3253:  19%|█▉        | 6/31 [00:35<02:24,  5.77s/it]
Training Epoch: 1/100, batch 12/63 completed (loss: 0.9279:  19%|█▉        | 6/31 [00:38<02:24,  5.76s/it]
Training Epoch: 1/100, batch 12/63 completed (loss: 0.9864:  19%|█▉        | 6/31 [00:38<02:24,  5.76s/it]
Training Epoch: 1/100, batch 12/63 completed (loss: 0.9274:  19%|█▉        | 6/31 [00:38<02:24,  5.76s/it]
Training Epoch: 1/100, batch 12/63 completed (loss: 1.0207:  19%|█▉        | 6/31 [00:38<02:24,  5.77s/it]
Training Epoch: 1/100, batch 12/63 completed (loss: 0.9279:  23%|██▎       | 7/31 [00:41<02:17,  5.74s/it]
Training Epoch: 1/100, batch 12/63 completed (loss: 1.0207:  23%|██▎       | 7/31 [00:41<02:17,  5.74s/it]
Training Epoch: 1/100, batch 12/63 completed (loss: 0.9274:  23%|██▎       | 7/31 [00:41<02:17,  5.74s/it]
Training Epoch: 1/100, batch 12/63 completed (loss: 0.9864:  23%|██▎       | 7/31 [00:41<02:17,  5.74s/it]
Training Epoch: 1/100, batch 13/63 completed (loss: 1.0458:  23%|██▎       | 7/31 [00:41<02:17,  5.74s/it]
Training Epoch: 1/100, batch 13/63 completed (loss: 1.1906:  23%|██▎       | 7/31 [00:41<02:17,  5.74s/it]
Training Epoch: 1/100, batch 13/63 completed (loss: 0.9818:  23%|██▎       | 7/31 [00:41<02:17,  5.74s/it]
Training Epoch: 1/100, batch 13/63 completed (loss: 1.1507:  23%|██▎       | 7/31 [00:41<02:17,  5.74s/it]
Training Epoch: 1/100, batch 14/63 completed (loss: 1.0696:  23%|██▎       | 7/31 [00:44<02:17,  5.74s/it]
Training Epoch: 1/100, batch 14/63 completed (loss: 0.6835:  23%|██▎       | 7/31 [00:44<02:17,  5.74s/it]
Training Epoch: 1/100, batch 14/63 completed (loss: 0.4506:  23%|██▎       | 7/31 [00:44<02:17,  5.74s/it]
Training Epoch: 1/100, batch 14/63 completed (loss: 0.5291:  23%|██▎       | 7/31 [00:44<02:17,  5.74s/it]
Training Epoch: 1/100, batch 14/63 completed (loss: 1.0696:  26%|██▌       | 8/31 [00:47<02:11,  5.73s/it]
Training Epoch: 1/100, batch 14/63 completed (loss: 0.4506:  26%|██▌       | 8/31 [00:47<02:11,  5.73s/it]
Training Epoch: 1/100, batch 14/63 completed (loss: 0.6835:  26%|██▌       | 8/31 [00:47<02:11,  5.73s/it]
Training Epoch: 1/100, batch 14/63 completed (loss: 0.5291:  26%|██▌       | 8/31 [00:47<02:11,  5.73s/it]
Training Epoch: 1/100, batch 15/63 completed (loss: 0.5324:  26%|██▌       | 8/31 [00:47<02:11,  5.73s/it]
Training Epoch: 1/100, batch 15/63 completed (loss: 0.7466:  26%|██▌       | 8/31 [00:47<02:11,  5.73s/it]
Training Epoch: 1/100, batch 15/63 completed (loss: 0.6897:  26%|██▌       | 8/31 [00:47<02:11,  5.73s/it]
Training Epoch: 1/100, batch 15/63 completed (loss: 0.6313:  26%|██▌       | 8/31 [00:47<02:11,  5.73s/it]
Training Epoch: 1/100, batch 16/63 completed (loss: 0.2841:  26%|██▌       | 8/31 [00:49<02:11,  5.73s/it]
Training Epoch: 1/100, batch 16/63 completed (loss: 0.4698:  26%|██▌       | 8/31 [00:49<02:11,  5.73s/it]
Training Epoch: 1/100, batch 16/63 completed (loss: 0.3907:  26%|██▌       | 8/31 [00:49<02:11,  5.73s/it]
Training Epoch: 1/100, batch 16/63 completed (loss: 0.4278:  26%|██▌       | 8/31 [00:49<02:11,  5.73s/it]
Training Epoch: 1/100, batch 16/63 completed (loss: 0.3907:  29%|██▉       | 9/31 [00:52<02:05,  5.72s/it]
Training Epoch: 1/100, batch 16/63 completed (loss: 0.2841:  29%|██▉       | 9/31 [00:52<02:05,  5.72s/it]
Training Epoch: 1/100, batch 16/63 completed (loss: 0.4698:  29%|██▉       | 9/31 [00:52<02:05,  5.72s/it]
Training Epoch: 1/100, batch 16/63 completed (loss: 0.4278:  29%|██▉       | 9/31 [00:52<02:05,  5.72s/it]
Training Epoch: 1/100, batch 17/63 completed (loss: 0.383:  29%|██▉       | 9/31 [00:52<02:05,  5.72s/it] 
Training Epoch: 1/100, batch 17/63 completed (loss: 0.7143:  29%|██▉       | 9/31 [00:52<02:05,  5.72s/it]
Training Epoch: 1/100, batch 17/63 completed (loss: 0.4616:  29%|██▉       | 9/31 [00:52<02:05,  5.72s/it]
Training Epoch: 1/100, batch 17/63 completed (loss: 0.754:  29%|██▉       | 9/31 [00:52<02:05,  5.72s/it] 
Training Epoch: 1/100, batch 18/63 completed (loss: 0.4361:  29%|██▉       | 9/31 [00:55<02:05,  5.72s/it]
Training Epoch: 1/100, batch 18/63 completed (loss: 0.314:  29%|██▉       | 9/31 [00:55<02:05,  5.72s/it]
Training Epoch: 1/100, batch 18/63 completed (loss: 0.5938:  29%|██▉       | 9/31 [00:55<02:05,  5.72s/it]
Training Epoch: 1/100, batch 18/63 completed (loss: 0.2297:  29%|██▉       | 9/31 [00:55<02:05,  5.72s/it]
Training Epoch: 1/100, batch 18/63 completed (loss: 0.2297:  32%|███▏      | 10/31 [00:58<02:03,  5.86s/it]
Training Epoch: 1/100, batch 18/63 completed (loss: 0.314:  32%|███▏      | 10/31 [00:58<02:03,  5.86s/it]
Training Epoch: 1/100, batch 18/63 completed (loss: 0.4361:  32%|███▏      | 10/31 [00:58<02:03,  5.86s/it]
Training Epoch: 1/100, batch 18/63 completed (loss: 0.5938:  32%|███▏      | 10/31 [00:58<02:03,  5.86s/it]
Training Epoch: 1/100, batch 19/63 completed (loss: 0.3488:  32%|███▏      | 10/31 [00:58<02:03,  5.86s/it]
Training Epoch: 1/100, batch 19/63 completed (loss: 0.3048:  32%|███▏      | 10/31 [00:58<02:03,  5.86s/it]
Training Epoch: 1/100, batch 19/63 completed (loss: 0.3981:  32%|███▏      | 10/31 [00:58<02:03,  5.86s/it]
Training Epoch: 1/100, batch 19/63 completed (loss: 0.3049:  32%|███▏      | 10/31 [00:59<02:03,  5.86s/it]
Training Epoch: 1/100, batch 20/63 completed (loss: 0.2639:  32%|███▏      | 10/31 [01:02<02:03,  5.86s/it]
Training Epoch: 1/100, batch 20/63 completed (loss: 0.3526:  32%|███▏      | 10/31 [01:02<02:03,  5.86s/it]
Training Epoch: 1/100, batch 20/63 completed (loss: 0.442:  32%|███▏      | 10/31 [01:02<02:03,  5.86s/it] 
Training Epoch: 1/100, batch 20/63 completed (loss: 0.3401:  32%|███▏      | 10/31 [01:02<02:03,  5.86s/it]
Training Epoch: 1/100, batch 20/63 completed (loss: 0.2639:  35%|███▌      | 11/31 [01:05<01:59,  5.97s/it]
Training Epoch: 1/100, batch 20/63 completed (loss: 0.442:  35%|███▌      | 11/31 [01:05<01:59,  5.97s/it]
Training Epoch: 1/100, batch 20/63 completed (loss: 0.3401:  35%|███▌      | 11/31 [01:05<01:59,  5.97s/it]
Training Epoch: 1/100, batch 20/63 completed (loss: 0.3526:  35%|███▌      | 11/31 [01:05<01:59,  5.97s/it]
Training Epoch: 1/100, batch 21/63 completed (loss: 0.2297:  35%|███▌      | 11/31 [01:05<01:59,  5.97s/it]
Training Epoch: 1/100, batch 21/63 completed (loss: 0.3668:  35%|███▌      | 11/31 [01:05<01:59,  5.97s/it]
Training Epoch: 1/100, batch 21/63 completed (loss: 0.5972:  35%|███▌      | 11/31 [01:05<01:59,  5.97s/it]
Training Epoch: 1/100, batch 21/63 completed (loss: 0.3303:  35%|███▌      | 11/31 [01:05<01:59,  5.97s/it]
Training Epoch: 1/100, batch 22/63 completed (loss: 0.4362:  35%|███▌      | 11/31 [01:08<01:59,  5.97s/it]
Training Epoch: 1/100, batch 22/63 completed (loss: 0.2428:  35%|███▌      | 11/31 [01:08<01:59,  5.97s/it]
Training Epoch: 1/100, batch 22/63 completed (loss: 0.3231:  35%|███▌      | 11/31 [01:08<01:59,  5.97s/it]
Training Epoch: 1/100, batch 22/63 completed (loss: 0.3219:  35%|███▌      | 11/31 [01:08<01:59,  5.97s/it]
Training Epoch: 1/100, batch 22/63 completed (loss: 0.2428:  39%|███▊      | 12/31 [01:10<01:51,  5.89s/it]
Training Epoch: 1/100, batch 22/63 completed (loss: 0.4362:  39%|███▊      | 12/31 [01:10<01:52,  5.90s/it]
Training Epoch: 1/100, batch 22/63 completed (loss: 0.3231:  39%|███▊      | 12/31 [01:10<01:52,  5.90s/it]
Training Epoch: 1/100, batch 22/63 completed (loss: 0.3219:  39%|███▊      | 12/31 [01:10<01:52,  5.90s/it]
Training Epoch: 1/100, batch 23/63 completed (loss: 0.2058:  39%|███▊      | 12/31 [01:10<01:52,  5.90s/it]
Training Epoch: 1/100, batch 23/63 completed (loss: 0.7933:  39%|███▊      | 12/31 [01:10<01:52,  5.90s/it]
Training Epoch: 1/100, batch 23/63 completed (loss: 0.2304:  39%|███▊      | 12/31 [01:10<01:51,  5.89s/it]
Training Epoch: 1/100, batch 23/63 completed (loss: 0.411:  39%|███▊      | 12/31 [01:10<01:52,  5.90s/it] 
Training Epoch: 1/100, batch 24/63 completed (loss: 0.3126:  39%|███▊      | 12/31 [01:14<01:52,  5.90s/it]
Training Epoch: 1/100, batch 24/63 completed (loss: 0.2299:  39%|███▊      | 12/31 [01:14<01:51,  5.89s/it]
Training Epoch: 1/100, batch 24/63 completed (loss: 0.1941:  39%|███▊      | 12/31 [01:14<01:52,  5.90s/it]
Training Epoch: 1/100, batch 24/63 completed (loss: 0.1552:  39%|███▊      | 12/31 [01:14<01:52,  5.90s/it]
Training Epoch: 1/100, batch 24/63 completed (loss: 0.1941:  42%|████▏     | 13/31 [01:17<01:48,  6.01s/it]
Training Epoch: 1/100, batch 24/63 completed (loss: 0.3126:  42%|████▏     | 13/31 [01:17<01:48,  6.01s/it]
Training Epoch: 1/100, batch 24/63 completed (loss: 0.1552:  42%|████▏     | 13/31 [01:17<01:48,  6.01s/it]
Training Epoch: 1/100, batch 24/63 completed (loss: 0.2299:  42%|████▏     | 13/31 [01:17<01:48,  6.01s/it]
Training Epoch: 1/100, batch 25/63 completed (loss: 0.4215:  42%|████▏     | 13/31 [01:17<01:48,  6.01s/it]
Training Epoch: 1/100, batch 25/63 completed (loss: 0.3121:  42%|████▏     | 13/31 [01:17<01:48,  6.01s/it]
Training Epoch: 1/100, batch 25/63 completed (loss: 0.4012:  42%|████▏     | 13/31 [01:17<01:48,  6.01s/it]
Training Epoch: 1/100, batch 25/63 completed (loss: 0.0852:  42%|████▏     | 13/31 [01:17<01:48,  6.01s/it]
Training Epoch: 1/100, batch 26/63 completed (loss: 0.3602:  42%|████▏     | 13/31 [01:20<01:48,  6.01s/it]
Training Epoch: 1/100, batch 26/63 completed (loss: 0.1812:  42%|████▏     | 13/31 [01:20<01:48,  6.01s/it]
Training Epoch: 1/100, batch 26/63 completed (loss: 0.1762:  42%|████▏     | 13/31 [01:20<01:48,  6.01s/it]
Training Epoch: 1/100, batch 26/63 completed (loss: 0.2487:  42%|████▏     | 13/31 [01:20<01:48,  6.01s/it]
Training Epoch: 1/100, batch 26/63 completed (loss: 0.1812:  45%|████▌     | 14/31 [01:22<01:40,  5.94s/it]
Training Epoch: 1/100, batch 26/63 completed (loss: 0.3602:  45%|████▌     | 14/31 [01:22<01:40,  5.94s/it]
Training Epoch: 1/100, batch 26/63 completed (loss: 0.1762:  45%|████▌     | 14/31 [01:22<01:41,  5.94s/it]
Training Epoch: 1/100, batch 26/63 completed (loss: 0.2487:  45%|████▌     | 14/31 [01:22<01:41,  5.94s/it]
Training Epoch: 1/100, batch 27/63 completed (loss: 0.5085:  45%|████▌     | 14/31 [01:22<01:41,  5.94s/it]
Training Epoch: 1/100, batch 27/63 completed (loss: 0.363:  45%|████▌     | 14/31 [01:22<01:40,  5.94s/it] 
Training Epoch: 1/100, batch 27/63 completed (loss: 0.1617:  45%|████▌     | 14/31 [01:22<01:40,  5.94s/it]
Training Epoch: 1/100, batch 27/63 completed (loss: 0.2645:  45%|████▌     | 14/31 [01:22<01:41,  5.94s/it]
Training Epoch: 1/100, batch 28/63 completed (loss: 0.4611:  45%|████▌     | 14/31 [01:25<01:40,  5.94s/it]
Training Epoch: 1/100, batch 28/63 completed (loss: 0.381:  45%|████▌     | 14/31 [01:25<01:40,  5.94s/it]
Training Epoch: 1/100, batch 28/63 completed (loss: 0.4364:  45%|████▌     | 14/31 [01:25<01:41,  5.94s/it]
Training Epoch: 1/100, batch 28/63 completed (loss: 0.2733:  45%|████▌     | 14/31 [01:25<01:41,  5.94s/it]
Training Epoch: 1/100, batch 28/63 completed (loss: 0.4611:  48%|████▊     | 15/31 [01:28<01:34,  5.88s/it]
Training Epoch: 1/100, batch 28/63 completed (loss: 0.2733:  48%|████▊     | 15/31 [01:28<01:34,  5.88s/it]
Training Epoch: 1/100, batch 28/63 completed (loss: 0.381:  48%|████▊     | 15/31 [01:28<01:34,  5.88s/it]
Training Epoch: 1/100, batch 28/63 completed (loss: 0.4364:  48%|████▊     | 15/31 [01:28<01:34,  5.88s/it]
Training Epoch: 1/100, batch 29/63 completed (loss: 0.3735:  48%|████▊     | 15/31 [01:28<01:34,  5.88s/it]
Training Epoch: 1/100, batch 29/63 completed (loss: 0.3351:  48%|████▊     | 15/31 [01:28<01:34,  5.88s/it]
Training Epoch: 1/100, batch 29/63 completed (loss: 0.5413:  48%|████▊     | 15/31 [01:28<01:34,  5.88s/it]
Training Epoch: 1/100, batch 29/63 completed (loss: 0.4183:  48%|████▊     | 15/31 [01:28<01:34,  5.88s/it]
Training Epoch: 1/100, batch 30/63 completed (loss: 0.3973:  48%|████▊     | 15/31 [01:32<01:34,  5.88s/it]
Training Epoch: 1/100, batch 30/63 completed (loss: 0.4419:  48%|████▊     | 15/31 [01:32<01:34,  5.88s/it]
Training Epoch: 1/100, batch 30/63 completed (loss: 0.1895:  48%|████▊     | 15/31 [01:32<01:34,  5.88s/it]
Training Epoch: 1/100, batch 30/63 completed (loss: 0.197:  48%|████▊     | 15/31 [01:32<01:34,  5.88s/it] 
Training Epoch: 1/100, batch 30/63 completed (loss: 0.3973:  52%|█████▏    | 16/31 [01:34<01:29,  5.99s/it]
Training Epoch: 1/100, batch 30/63 completed (loss: 0.1895:  52%|█████▏    | 16/31 [01:34<01:29,  6.00s/it]
Training Epoch: 1/100, batch 30/63 completed (loss: 0.4419:  52%|█████▏    | 16/31 [01:34<01:29,  6.00s/it]
Training Epoch: 1/100, batch 30/63 completed (loss: 0.197:  52%|█████▏    | 16/31 [01:34<01:29,  6.00s/it]
Training Epoch: 1/100, batch 31/63 completed (loss: 0.4972:  52%|█████▏    | 16/31 [01:34<01:29,  5.99s/it]
Training Epoch: 1/100, batch 31/63 completed (loss: 0.3673:  52%|█████▏    | 16/31 [01:34<01:29,  6.00s/it]
Training Epoch: 1/100, batch 31/63 completed (loss: 0.4672:  52%|█████▏    | 16/31 [01:34<01:29,  6.00s/it]
Training Epoch: 1/100, batch 31/63 completed (loss: 0.4813:  52%|█████▏    | 16/31 [01:35<01:29,  6.00s/it]
Training Epoch: 1/100, batch 32/63 completed (loss: 0.1672:  52%|█████▏    | 16/31 [01:37<01:29,  6.00s/it]
Training Epoch: 1/100, batch 32/63 completed (loss: 0.3395:  52%|█████▏    | 16/31 [01:37<01:29,  5.99s/it]
Training Epoch: 1/100, batch 32/63 completed (loss: 0.1531:  52%|█████▏    | 16/31 [01:37<01:29,  6.00s/it]
Training Epoch: 1/100, batch 32/63 completed (loss: 0.2724:  52%|█████▏    | 16/31 [01:37<01:29,  6.00s/it]
Training Epoch: 1/100, batch 32/63 completed (loss: 0.1672:  55%|█████▍    | 17/31 [01:40<01:22,  5.93s/it]
Training Epoch: 1/100, batch 32/63 completed (loss: 0.1531:  55%|█████▍    | 17/31 [01:40<01:22,  5.93s/it]
Training Epoch: 1/100, batch 32/63 completed (loss: 0.3395:  55%|█████▍    | 17/31 [01:40<01:22,  5.93s/it]
Training Epoch: 1/100, batch 32/63 completed (loss: 0.2724:  55%|█████▍    | 17/31 [01:40<01:23,  5.93s/it]
Training Epoch: 1/100, batch 33/63 completed (loss: 0.1698:  55%|█████▍    | 17/31 [01:40<01:22,  5.93s/it]
Training Epoch: 1/100, batch 33/63 completed (loss: 0.3672:  55%|█████▍    | 17/31 [01:40<01:22,  5.93s/it]
Training Epoch: 1/100, batch 33/63 completed (loss: 0.315:  55%|█████▍    | 17/31 [01:40<01:22,  5.93s/it] 
Training Epoch: 1/100, batch 33/63 completed (loss: 0.392:  55%|█████▍    | 17/31 [01:40<01:23,  5.93s/it] 
Training Epoch: 1/100, batch 34/63 completed (loss: 0.4661:  55%|█████▍    | 17/31 [01:43<01:22,  5.93s/it]
Training Epoch: 1/100, batch 34/63 completed (loss: 0.2707:  55%|█████▍    | 17/31 [01:43<01:22,  5.93s/it]
Training Epoch: 1/100, batch 34/63 completed (loss: 0.2232:  55%|█████▍    | 17/31 [01:43<01:22,  5.93s/it]
Training Epoch: 1/100, batch 34/63 completed (loss: 0.2314:  55%|█████▍    | 17/31 [01:43<01:23,  5.93s/it]
Training Epoch: 1/100, batch 34/63 completed (loss: 0.2232:  58%|█████▊    | 18/31 [01:47<01:18,  6.07s/it]
Training Epoch: 1/100, batch 34/63 completed (loss: 0.4661:  58%|█████▊    | 18/31 [01:47<01:18,  6.07s/it]
Training Epoch: 1/100, batch 34/63 completed (loss: 0.2707:  58%|█████▊    | 18/31 [01:47<01:18,  6.07s/it]
Training Epoch: 1/100, batch 34/63 completed (loss: 0.2314:  58%|█████▊    | 18/31 [01:47<01:18,  6.07s/it]
Training Epoch: 1/100, batch 35/63 completed (loss: 0.4655:  58%|█████▊    | 18/31 [01:47<01:18,  6.07s/it]
Training Epoch: 1/100, batch 35/63 completed (loss: 0.4752:  58%|█████▊    | 18/31 [01:47<01:18,  6.07s/it]
Training Epoch: 1/100, batch 35/63 completed (loss: 0.5251:  58%|█████▊    | 18/31 [01:47<01:18,  6.07s/it]
Training Epoch: 1/100, batch 35/63 completed (loss: 0.3148:  58%|█████▊    | 18/31 [01:47<01:18,  6.07s/it]
Training Epoch: 1/100, batch 36/63 completed (loss: 0.3483:  58%|█████▊    | 18/31 [01:50<01:18,  6.07s/it]
Training Epoch: 1/100, batch 36/63 completed (loss: 0.4966:  58%|█████▊    | 18/31 [01:50<01:18,  6.07s/it]
Training Epoch: 1/100, batch 36/63 completed (loss: 0.2114:  58%|█████▊    | 18/31 [01:50<01:18,  6.07s/it]
Training Epoch: 1/100, batch 36/63 completed (loss: 0.2407:  58%|█████▊    | 18/31 [01:50<01:18,  6.07s/it]
Training Epoch: 1/100, batch 36/63 completed (loss: 0.4966:  61%|██████▏   | 19/31 [01:52<01:11,  5.98s/it]
Training Epoch: 1/100, batch 36/63 completed (loss: 0.2407:  61%|██████▏   | 19/31 [01:52<01:11,  5.98s/it]
Training Epoch: 1/100, batch 36/63 completed (loss: 0.2114:  61%|██████▏   | 19/31 [01:52<01:11,  5.98s/it]
Training Epoch: 1/100, batch 36/63 completed (loss: 0.3483:  61%|██████▏   | 19/31 [01:52<01:11,  5.98s/it]
Training Epoch: 1/100, batch 37/63 completed (loss: 0.2614:  61%|██████▏   | 19/31 [01:52<01:11,  5.98s/it]
Training Epoch: 1/100, batch 37/63 completed (loss: 0.3103:  61%|██████▏   | 19/31 [01:52<01:11,  5.98s/it]
Training Epoch: 1/100, batch 37/63 completed (loss: 0.7135:  61%|██████▏   | 19/31 [01:52<01:11,  5.98s/it]
Training Epoch: 1/100, batch 37/63 completed (loss: 0.3673:  61%|██████▏   | 19/31 [01:52<01:11,  5.98s/it]
Training Epoch: 1/100, batch 38/63 completed (loss: 0.2343:  61%|██████▏   | 19/31 [01:55<01:11,  5.98s/it]
Training Epoch: 1/100, batch 38/63 completed (loss: 0.5257:  61%|██████▏   | 19/31 [01:55<01:11,  5.98s/it]
Training Epoch: 1/100, batch 38/63 completed (loss: 0.27:  61%|██████▏   | 19/31 [01:55<01:11,  5.98s/it]  
Training Epoch: 1/100, batch 38/63 completed (loss: 0.285:  61%|██████▏   | 19/31 [01:55<01:11,  5.98s/it] 
Training Epoch: 1/100, batch 38/63 completed (loss: 0.2343:  65%|██████▍   | 20/31 [01:58<01:05,  5.93s/it]
Training Epoch: 1/100, batch 38/63 completed (loss: 0.5257:  65%|██████▍   | 20/31 [01:58<01:05,  5.93s/it]
Training Epoch: 1/100, batch 38/63 completed (loss: 0.285:  65%|██████▍   | 20/31 [01:58<01:05,  5.93s/it]
Training Epoch: 1/100, batch 38/63 completed (loss: 0.27:  65%|██████▍   | 20/31 [01:58<01:05,  5.93s/it]
Training Epoch: 1/100, batch 39/63 completed (loss: 0.2415:  65%|██████▍   | 20/31 [01:58<01:05,  5.93s/it]
Training Epoch: 1/100, batch 39/63 completed (loss: 0.321:  65%|██████▍   | 20/31 [01:58<01:05,  5.93s/it]
Training Epoch: 1/100, batch 39/63 completed (loss: 0.4077:  65%|██████▍   | 20/31 [01:58<01:05,  5.93s/it]
Training Epoch: 1/100, batch 39/63 completed (loss: 0.6745:  65%|██████▍   | 20/31 [01:58<01:05,  5.93s/it]
Training Epoch: 1/100, batch 40/63 completed (loss: 0.2451:  65%|██████▍   | 20/31 [02:02<01:05,  5.93s/it]
Training Epoch: 1/100, batch 40/63 completed (loss: 0.3933:  65%|██████▍   | 20/31 [02:02<01:05,  5.93s/it]
Training Epoch: 1/100, batch 40/63 completed (loss: 0.4131:  65%|██████▍   | 20/31 [02:02<01:05,  5.93s/it]
Training Epoch: 1/100, batch 40/63 completed (loss: 0.7015:  65%|██████▍   | 20/31 [02:02<01:05,  5.93s/it]
Training Epoch: 1/100, batch 40/63 completed (loss: 0.4131:  68%|██████▊   | 21/31 [02:04<01:00,  6.03s/it]
Training Epoch: 1/100, batch 40/63 completed (loss: 0.2451:  68%|██████▊   | 21/31 [02:04<01:00,  6.03s/it]
Training Epoch: 1/100, batch 40/63 completed (loss: 0.7015:  68%|██████▊   | 21/31 [02:04<01:00,  6.03s/it]
Training Epoch: 1/100, batch 40/63 completed (loss: 0.3933:  68%|██████▊   | 21/31 [02:04<01:00,  6.04s/it]
Training Epoch: 1/100, batch 41/63 completed (loss: 0.6396:  68%|██████▊   | 21/31 [02:05<01:00,  6.03s/it]
Training Epoch: 1/100, batch 41/63 completed (loss: 0.4301:  68%|██████▊   | 21/31 [02:05<01:00,  6.03s/it]
Training Epoch: 1/100, batch 41/63 completed (loss: 0.2984:  68%|██████▊   | 21/31 [02:05<01:00,  6.04s/it]
Training Epoch: 1/100, batch 41/63 completed (loss: 0.353:  68%|██████▊   | 21/31 [02:05<01:00,  6.03s/it] 
Training Epoch: 1/100, batch 42/63 completed (loss: 0.2541:  68%|██████▊   | 21/31 [02:07<01:00,  6.03s/it]
Training Epoch: 1/100, batch 42/63 completed (loss: 0.3738:  68%|██████▊   | 21/31 [02:07<01:00,  6.04s/it]
Training Epoch: 1/100, batch 42/63 completed (loss: 0.4852:  68%|██████▊   | 21/31 [02:07<01:00,  6.03s/it]
Training Epoch: 1/100, batch 42/63 completed (loss: 0.3603:  68%|██████▊   | 21/31 [02:07<01:00,  6.03s/it]
Training Epoch: 1/100, batch 42/63 completed (loss: 0.4852:  71%|███████   | 22/31 [02:10<00:53,  5.95s/it]
Training Epoch: 1/100, batch 42/63 completed (loss: 0.2541:  71%|███████   | 22/31 [02:10<00:53,  5.95s/it]
Training Epoch: 1/100, batch 42/63 completed (loss: 0.3738:  71%|███████   | 22/31 [02:10<00:53,  5.96s/it]
Training Epoch: 1/100, batch 42/63 completed (loss: 0.3603:  71%|███████   | 22/31 [02:10<00:53,  5.96s/it]
Training Epoch: 1/100, batch 43/63 completed (loss: 0.2247:  71%|███████   | 22/31 [02:10<00:53,  5.95s/it]
Training Epoch: 1/100, batch 43/63 completed (loss: 0.1901:  71%|███████   | 22/31 [02:10<00:53,  5.96s/it]
Training Epoch: 1/100, batch 43/63 completed (loss: 0.3:  71%|███████   | 22/31 [02:10<00:53,  5.95s/it]   
Training Epoch: 1/100, batch 43/63 completed (loss: 0.407:  71%|███████   | 22/31 [02:10<00:53,  5.96s/it] 
Training Epoch: 1/100, batch 44/63 completed (loss: 0.3202:  71%|███████   | 22/31 [02:13<00:53,  5.95s/it]
Training Epoch: 1/100, batch 44/63 completed (loss: 0.3808:  71%|███████   | 22/31 [02:13<00:53,  5.96s/it]
Training Epoch: 1/100, batch 44/63 completed (loss: 0.2201:  71%|███████   | 22/31 [02:13<00:53,  5.95s/it]
Training Epoch: 1/100, batch 44/63 completed (loss: 0.1389:  71%|███████   | 22/31 [02:13<00:53,  5.96s/it]
Training Epoch: 1/100, batch 44/63 completed (loss: 0.1389:  74%|███████▍  | 23/31 [02:16<00:47,  5.96s/it]
Training Epoch: 1/100, batch 44/63 completed (loss: 0.3202:  74%|███████▍  | 23/31 [02:16<00:47,  5.96s/it]
Training Epoch: 1/100, batch 44/63 completed (loss: 0.2201:  74%|███████▍  | 23/31 [02:16<00:47,  5.96s/it]
Training Epoch: 1/100, batch 44/63 completed (loss: 0.3808:  74%|███████▍  | 23/31 [02:16<00:47,  5.97s/it]
Training Epoch: 1/100, batch 45/63 completed (loss: 0.2177:  74%|███████▍  | 23/31 [02:16<00:47,  5.96s/it]
Training Epoch: 1/100, batch 45/63 completed (loss: 0.1395:  74%|███████▍  | 23/31 [02:16<00:47,  5.97s/it]
Training Epoch: 1/100, batch 45/63 completed (loss: 0.2454:  74%|███████▍  | 23/31 [02:16<00:47,  5.96s/it]
Training Epoch: 1/100, batch 45/63 completed (loss: 0.3967:  74%|███████▍  | 23/31 [02:16<00:47,  5.96s/it]
Training Epoch: 1/100, batch 46/63 completed (loss: 0.3656:  74%|███████▍  | 23/31 [02:20<00:47,  5.96s/it]
Training Epoch: 1/100, batch 46/63 completed (loss: 0.1981:  74%|███████▍  | 23/31 [02:20<00:47,  5.97s/it]
Training Epoch: 1/100, batch 46/63 completed (loss: 0.236:  74%|███████▍  | 23/31 [02:20<00:47,  5.96s/it] 
Training Epoch: 1/100, batch 46/63 completed (loss: 0.2721:  74%|███████▍  | 23/31 [02:20<00:47,  5.96s/it]
Training Epoch: 1/100, batch 46/63 completed (loss: 0.3656:  77%|███████▋  | 24/31 [02:22<00:42,  6.02s/it]
Training Epoch: 1/100, batch 46/63 completed (loss: 0.2721:  77%|███████▋  | 24/31 [02:22<00:42,  6.02s/it]
Training Epoch: 1/100, batch 46/63 completed (loss: 0.1981:  77%|███████▋  | 24/31 [02:22<00:42,  6.02s/it]
Training Epoch: 1/100, batch 46/63 completed (loss: 0.236:  77%|███████▋  | 24/31 [02:22<00:42,  6.03s/it]
Training Epoch: 1/100, batch 47/63 completed (loss: 0.2379:  77%|███████▋  | 24/31 [02:22<00:42,  6.03s/it]
Training Epoch: 1/100, batch 47/63 completed (loss: 0.5138:  77%|███████▋  | 24/31 [02:22<00:42,  6.02s/it]
Training Epoch: 1/100, batch 47/63 completed (loss: 0.3439:  77%|███████▋  | 24/31 [02:22<00:42,  6.02s/it]
Training Epoch: 1/100, batch 47/63 completed (loss: 0.2367:  77%|███████▋  | 24/31 [02:22<00:42,  6.02s/it]
Training Epoch: 1/100, batch 48/63 completed (loss: 0.3825:  77%|███████▋  | 24/31 [02:25<00:42,  6.03s/it]
Training Epoch: 1/100, batch 48/63 completed (loss: 0.3719:  77%|███████▋  | 24/31 [02:25<00:42,  6.02s/it]
Training Epoch: 1/100, batch 48/63 completed (loss: 0.1337:  77%|███████▋  | 24/31 [02:25<00:42,  6.02s/it]
Training Epoch: 1/100, batch 48/63 completed (loss: 0.1418:  77%|███████▋  | 24/31 [02:25<00:42,  6.02s/it]
Training Epoch: 1/100, batch 48/63 completed (loss: 0.1418:  81%|████████  | 25/31 [02:28<00:35,  5.96s/it]
Training Epoch: 1/100, batch 48/63 completed (loss: 0.3825:  81%|████████  | 25/31 [02:28<00:35,  5.96s/it]
Training Epoch: 1/100, batch 48/63 completed (loss: 0.1337:  81%|████████  | 25/31 [02:28<00:35,  5.96s/it]
Training Epoch: 1/100, batch 48/63 completed (loss: 0.3719:  81%|████████  | 25/31 [02:28<00:35,  5.96s/it]
Training Epoch: 1/100, batch 49/63 completed (loss: 0.4944:  81%|████████  | 25/31 [02:28<00:35,  5.96s/it]
Training Epoch: 1/100, batch 49/63 completed (loss: 0.2138:  81%|████████  | 25/31 [02:28<00:35,  5.96s/it]
Training Epoch: 1/100, batch 49/63 completed (loss: 0.5005:  81%|████████  | 25/31 [02:28<00:35,  5.96s/it]
Training Epoch: 1/100, batch 49/63 completed (loss: 0.159:  81%|████████  | 25/31 [02:28<00:35,  5.96s/it] 
Training Epoch: 1/100, batch 50/63 completed (loss: 0.4471:  81%|████████  | 25/31 [02:31<00:35,  5.96s/it]
Training Epoch: 1/100, batch 50/63 completed (loss: 0.717:  81%|████████  | 25/31 [02:31<00:35,  5.96s/it] 
Training Epoch: 1/100, batch 50/63 completed (loss: 0.5502:  81%|████████  | 25/31 [02:31<00:35,  5.96s/it]
Training Epoch: 1/100, batch 50/63 completed (loss: 0.2609:  81%|████████  | 25/31 [02:31<00:35,  5.96s/it]
Training Epoch: 1/100, batch 50/63 completed (loss: 0.5502:  84%|████████▍ | 26/31 [02:34<00:30,  6.05s/it]
Training Epoch: 1/100, batch 50/63 completed (loss: 0.2609:  84%|████████▍ | 26/31 [02:34<00:30,  6.05s/it]
Training Epoch: 1/100, batch 50/63 completed (loss: 0.4471:  84%|████████▍ | 26/31 [02:34<00:30,  6.06s/it]
Training Epoch: 1/100, batch 50/63 completed (loss: 0.717:  84%|████████▍ | 26/31 [02:34<00:30,  6.06s/it]
Training Epoch: 1/100, batch 51/63 completed (loss: 0.2594:  84%|████████▍ | 26/31 [02:35<00:30,  6.06s/it]
Training Epoch: 1/100, batch 51/63 completed (loss: 0.2728:  84%|████████▍ | 26/31 [02:35<00:30,  6.06s/it]
Training Epoch: 1/100, batch 51/63 completed (loss: 0.4848:  84%|████████▍ | 26/31 [02:35<00:30,  6.05s/it]
Training Epoch: 1/100, batch 51/63 completed (loss: 0.1166:  84%|████████▍ | 26/31 [02:35<00:30,  6.05s/it]
Training Epoch: 1/100, batch 52/63 completed (loss: 0.1171:  84%|████████▍ | 26/31 [02:38<00:30,  6.05s/it]
Training Epoch: 1/100, batch 52/63 completed (loss: 0.3661:  84%|████████▍ | 26/31 [02:38<00:30,  6.06s/it]
Training Epoch: 1/100, batch 52/63 completed (loss: 0.3202:  84%|████████▍ | 26/31 [02:38<00:30,  6.06s/it]
Training Epoch: 1/100, batch 52/63 completed (loss: 0.3144:  84%|████████▍ | 26/31 [02:38<00:30,  6.05s/it]
Training Epoch: 1/100, batch 52/63 completed (loss: 0.1171:  87%|████████▋ | 27/31 [02:40<00:24,  6.03s/it]
Training Epoch: 1/100, batch 52/63 completed (loss: 0.3144:  87%|████████▋ | 27/31 [02:40<00:24,  6.03s/it]
Training Epoch: 1/100, batch 52/63 completed (loss: 0.3202:  87%|████████▋ | 27/31 [02:40<00:24,  6.03s/it]
Training Epoch: 1/100, batch 52/63 completed (loss: 0.3661:  87%|████████▋ | 27/31 [02:40<00:24,  6.04s/it]
Training Epoch: 1/100, batch 53/63 completed (loss: 0.354:  87%|████████▋ | 27/31 [02:41<00:24,  6.03s/it] 
Training Epoch: 1/100, batch 53/63 completed (loss: 0.1587:  87%|████████▋ | 27/31 [02:41<00:24,  6.04s/it]
Training Epoch: 1/100, batch 53/63 completed (loss: 0.1327:  87%|████████▋ | 27/31 [02:41<00:24,  6.03s/it]
Training Epoch: 1/100, batch 53/63 completed (loss: 0.2128:  87%|████████▋ | 27/31 [02:41<00:24,  6.03s/it]
Training Epoch: 1/100, batch 54/63 completed (loss: 0.1362:  87%|████████▋ | 27/31 [02:43<00:24,  6.03s/it]
Training Epoch: 1/100, batch 54/63 completed (loss: 0.3423:  87%|████████▋ | 27/31 [02:43<00:24,  6.04s/it]
Training Epoch: 1/100, batch 54/63 completed (loss: 0.0616:  87%|████████▋ | 27/31 [02:43<00:24,  6.03s/it]
Training Epoch: 1/100, batch 54/63 completed (loss: 0.1983:  87%|████████▋ | 27/31 [02:43<00:24,  6.03s/it]
Training Epoch: 1/100, batch 54/63 completed (loss: 0.0616:  90%|█████████ | 28/31 [02:46<00:17,  5.97s/it]
Training Epoch: 1/100, batch 54/63 completed (loss: 0.1983:  90%|█████████ | 28/31 [02:46<00:17,  5.97s/it]
Training Epoch: 1/100, batch 54/63 completed (loss: 0.3423:  90%|█████████ | 28/31 [02:46<00:17,  5.97s/it]
Training Epoch: 1/100, batch 54/63 completed (loss: 0.1362:  90%|█████████ | 28/31 [02:46<00:17,  5.98s/it]
Training Epoch: 1/100, batch 55/63 completed (loss: 0.1754:  90%|█████████ | 28/31 [02:46<00:17,  5.98s/it]
Training Epoch: 1/100, batch 55/63 completed (loss: 0.3194:  90%|█████████ | 28/31 [02:46<00:17,  5.97s/it]
Training Epoch: 1/100, batch 55/63 completed (loss: 0.2395:  90%|█████████ | 28/31 [02:46<00:17,  5.97s/it]
Training Epoch: 1/100, batch 55/63 completed (loss: 0.4478:  90%|█████████ | 28/31 [02:46<00:17,  5.97s/it]
Training Epoch: 1/100, batch 56/63 completed (loss: 0.4938:  90%|█████████ | 28/31 [02:50<00:17,  5.98s/it]
Training Epoch: 1/100, batch 56/63 completed (loss: 0.205:  90%|█████████ | 28/31 [02:50<00:17,  5.97s/it] 
Training Epoch: 1/100, batch 56/63 completed (loss: 0.3967:  90%|█████████ | 28/31 [02:50<00:17,  5.97s/it]
Training Epoch: 1/100, batch 56/63 completed (loss: 0.3346:  90%|█████████ | 28/31 [02:50<00:17,  5.97s/it]
Training Epoch: 1/100, batch 56/63 completed (loss: 0.3967:  94%|█████████▎| 29/31 [02:53<00:12,  6.11s/it]
Training Epoch: 1/100, batch 56/63 completed (loss: 0.4938:  94%|█████████▎| 29/31 [02:53<00:12,  6.11s/it]
Training Epoch: 1/100, batch 56/63 completed (loss: 0.3346:  94%|█████████▎| 29/31 [02:53<00:12,  6.11s/it]
Training Epoch: 1/100, batch 56/63 completed (loss: 0.205:  94%|█████████▎| 29/31 [02:53<00:12,  6.11s/it]
Training Epoch: 1/100, batch 57/63 completed (loss: 0.1525:  94%|█████████▎| 29/31 [02:53<00:12,  6.11s/it]
Training Epoch: 1/100, batch 57/63 completed (loss: 0.7592:  94%|█████████▎| 29/31 [02:53<00:12,  6.11s/it]
Training Epoch: 1/100, batch 57/63 completed (loss: 0.2373:  94%|█████████▎| 29/31 [02:53<00:12,  6.11s/it]
Training Epoch: 1/100, batch 57/63 completed (loss: 0.201:  94%|█████████▎| 29/31 [02:53<00:12,  6.11s/it] 
Training Epoch: 1/100, batch 58/63 completed (loss: 0.3015:  94%|█████████▎| 29/31 [02:56<00:12,  6.11s/it]
Training Epoch: 1/100, batch 58/63 completed (loss: 0.3998:  94%|█████████▎| 29/31 [02:56<00:12,  6.11s/it]
Training Epoch: 1/100, batch 58/63 completed (loss: 0.4062:  94%|█████████▎| 29/31 [02:56<00:12,  6.11s/it]
Training Epoch: 1/100, batch 58/63 completed (loss: 0.1671:  94%|█████████▎| 29/31 [02:56<00:12,  6.11s/it]
Training Epoch: 1/100, batch 58/63 completed (loss: 0.3015:  97%|█████████▋| 30/31 [02:59<00:06,  6.02s/it]
Training Epoch: 1/100, batch 58/63 completed (loss: 0.3998:  97%|█████████▋| 30/31 [02:59<00:06,  6.02s/it]
Training Epoch: 1/100, batch 58/63 completed (loss: 0.1671:  97%|█████████▋| 30/31 [02:59<00:06,  6.03s/it]
Training Epoch: 1/100, batch 58/63 completed (loss: 0.4062:  97%|█████████▋| 30/31 [02:59<00:06,  6.03s/it]
Training Epoch: 1/100, batch 59/63 completed (loss: 0.4449:  97%|█████████▋| 30/31 [02:59<00:06,  6.03s/it]
Training Epoch: 1/100, batch 59/63 completed (loss: 0.382:  97%|█████████▋| 30/31 [02:59<00:06,  6.02s/it] 
Training Epoch: 1/100, batch 59/63 completed (loss: 0.2253:  97%|█████████▋| 30/31 [02:59<00:06,  6.02s/it]
Training Epoch: 1/100, batch 59/63 completed (loss: 0.2913:  97%|█████████▋| 30/31 [02:59<00:06,  6.03s/it]
Training Epoch: 1/100, batch 60/63 completed (loss: 0.2721:  97%|█████████▋| 30/31 [03:02<00:06,  6.03s/it]
Training Epoch: 1/100, batch 60/63 completed (loss: 0.5158:  97%|█████████▋| 30/31 [03:02<00:06,  6.02s/it]
Training Epoch: 1/100, batch 60/63 completed (loss: 0.2632:  97%|█████████▋| 30/31 [03:02<00:06,  6.02s/it]
Training Epoch: 1/100, batch 60/63 completed (loss: 0.1736:  97%|█████████▋| 30/31 [03:02<00:06,  6.03s/it]
Training Epoch: 1/100, batch 60/63 completed (loss: 0.1736: 100%|██████████| 31/31 [03:05<00:00,  6.05s/it]
Training Epoch: 1/100, batch 60/63 completed (loss: 0.5158: 100%|██████████| 31/31 [03:05<00:00,  6.05s/it]
Training Epoch: 1/100, batch 60/63 completed (loss: 0.2632: 100%|██████████| 31/31 [03:05<00:00,  6.05s/it]
Training Epoch: 1/100, batch 60/63 completed (loss: 0.2721: 100%|██████████| 31/31 [03:05<00:00,  6.05s/it]
Training Epoch: 1/100, batch 61/63 completed (loss: 0.1196: 100%|██████████| 31/31 [03:05<00:00,  6.05s/it]
Training Epoch: 1/100, batch 61/63 completed (loss: 0.6089: 100%|██████████| 31/31 [03:05<00:00,  6.05s/it]
Training Epoch: 1/100, batch 61/63 completed (loss: 0.2394: 100%|██████████| 31/31 [03:05<00:00,  6.05s/it]
Training Epoch: 1/100, batch 61/63 completed (loss: 0.2518: 100%|██████████| 31/31 [03:05<00:00,  6.05s/it]
Training Epoch: 1/100, batch 61/63 completed (loss: 0.2518: : 32it [03:07,  4.91s/it]                      
Training Epoch: 1/100, batch 61/63 completed (loss: 0.2394: : 32it [03:07,  4.91s/it]                      
Training Epoch: 1/100, batch 62/63 completed (loss: 0.542: : 32it [03:07,  4.91s/it] 
Training Epoch: 1/100, batch 62/63 completed (loss: 0.6108: : 32it [03:07,  4.91s/it]
Training Epoch: 1/100, batch 61/63 completed (loss: 0.6089: : 32it [03:07,  4.91s/it]                      
Training Epoch: 1/100, batch 62/63 completed (loss: 0.2626: : 32it [03:07,  4.91s/it]
Training Epoch: 1/100, batch 61/63 completed (loss: 0.1196: : 32it [03:07,  4.91s/it]                      
Training Epoch: 1/100, batch 62/63 completed (loss: 0.6307: : 32it [03:07,  4.91s/it]
Training Epoch: 1/100, batch 62/63 completed (loss: 0.542: : 32it [03:07,  5.86s/it]

Training Epoch: 1/100, batch 62/63 completed (loss: 0.2626: : 32it [03:07,  5.86s/it]

Training Epoch: 1/100, batch 62/63 completed (loss: 0.6108: : 32it [03:07,  5.86s/it]
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/c10d_logger.py:83: UserWarning: barrier(): using the device under current context. You can specify `device_id` in `init_process_group` to mute this warning.
  return func(*args, **kwargs)

Training Epoch: 1/100, batch 62/63 completed (loss: 0.6307: : 32it [03:07,  5.86s/it]
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
saving model + train state.
Warning: failed to upload checkpoint_1 to HF: (Request ID: Root=1-6a1b6324-2903403d18bc526b4fb5c26c;9de63419-5e0c-4a75-80ec-6d13a198ab8f)

403 Forbidden: You need to setup automatic credit recharge in order to upload more data. You can do so at /settings/billing..
Cannot access content at: https://huggingface.co/seyedparsa/thoughtformer-checkpoints.git/info/lfs/objects/batch.
Make sure your token has the correct permissions.
Warning: failed to upload train_state_1 to HF: (Request ID: Root=1-6a1b6328-66ace8d63ff6d54e347381a9;895952b3-0d77-492c-b876-bb907f34efed)

403 Forbidden: You need to setup automatic credit recharge in order to upload more data. You can do so at /settings/billing..
Cannot access content at: https://huggingface.co/seyedparsa/thoughtformer-checkpoints.git/info/lfs/objects/batch.
Make sure your token has the correct permissions.
[warn] HF upload failed for ckpt_1; kept locally, earlier ckpts deleted.

Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]
Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]
Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]eval loss 0.2619936785527638

Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]
[WRONG] Example 0 (diff=3)
  Expected : 'Claude'
  Extracted: 'Morgan'
  Generated: '### Morgan'

[WRONG] Example 1 (diff=8)
  Expected : 'Guy'
  Extracted: 'Irma'
  Generated: '### Irma'

[WRONG] Example 2 (diff=10)
  Expected : 'Elvis'
  Extracted: 'Shannon'
  Generated: '### Shannon'

Test Accuracy:   3%|▎         | 1/32 [00:06<03:27,  6.69s/it]
Test Accuracy:   3%|▎         | 1/32 [00:06<03:27,  6.70s/it]
Test Accuracy:   3%|▎         | 1/32 [00:06<03:27,  6.69s/it]
Test accuracy: 0.38:   3%|▎         | 1/32 [00:06<03:27,  6.69s/it]
Test accuracy: 0.62:   3%|▎         | 1/32 [00:06<03:27,  6.70s/it]
[CORRECT] Example 3 (diff=8)
  Expected : 'Coral'
  Extracted: 'Coral'
  Generated: '### Coral'
Test accuracy: 0.75:   3%|▎         | 1/32 [00:06<03:27,  6.69s/it]

[WRONG] Example 4 (diff=1)
  Expected : 'Patricia'
  Extracted: 'Evelyn'
  Generated: '### Evelyn'

Test Accuracy:   3%|▎         | 1/32 [00:06<03:22,  6.54s/it]
Test accuracy: 0.38:   3%|▎         | 1/32 [00:06<03:22,  6.54s/it]
Test accuracy: 0.75:   6%|▋         | 2/32 [00:13<03:17,  6.59s/it]
Test accuracy: 0.62:   6%|▋         | 2/32 [00:13<03:17,  6.60s/it]
Test accuracy: 0.69:   6%|▋         | 2/32 [00:13<03:17,  6.59s/it]
Test accuracy: 0.56:   6%|▋         | 2/32 [00:13<03:17,  6.60s/it]
Test accuracy: 0.38:   6%|▋         | 2/32 [00:13<03:17,  6.59s/it]
Test accuracy: 0.56:   6%|▋         | 2/32 [00:13<03:17,  6.59s/it]
Test accuracy: 0.38:   6%|▋         | 2/32 [00:13<03:15,  6.53s/it]
Test accuracy: 0.25:   6%|▋         | 2/32 [00:13<03:15,  6.53s/it]
Test accuracy: 0.56:   9%|▉         | 3/32 [00:19<03:13,  6.67s/it]
Test accuracy: 0.5:   9%|▉         | 3/32 [00:19<03:13,  6.67s/it] 
Test accuracy: 0.69:   9%|▉         | 3/32 [00:19<03:13,  6.67s/it]
Test accuracy: 0.58:   9%|▉         | 3/32 [00:19<03:13,  6.67s/it]
Test accuracy: 0.25:   9%|▉         | 3/32 [00:19<03:12,  6.63s/it]
Test accuracy: 0.42:   9%|▉         | 3/32 [00:19<03:12,  6.63s/it]
Test accuracy: 0.56:   9%|▉         | 3/32 [00:20<03:13,  6.68s/it]
Test accuracy: 0.58:   9%|▉         | 3/32 [00:20<03:13,  6.68s/it]
Test accuracy: 0.5:  12%|█▎        | 4/32 [00:26<03:06,  6.65s/it]
Test accuracy: 0.42:  12%|█▎        | 4/32 [00:26<03:05,  6.63s/it]
Test accuracy: 0.53:  12%|█▎        | 4/32 [00:26<03:06,  6.65s/it]
Test accuracy: 0.44:  12%|█▎        | 4/32 [00:26<03:05,  6.63s/it]
Test accuracy: 0.58:  12%|█▎        | 4/32 [00:26<03:06,  6.65s/it]
Test accuracy: 0.58:  12%|█▎        | 4/32 [00:26<03:06,  6.65s/it]
Test accuracy: 0.56:  12%|█▎        | 4/32 [00:26<03:06,  6.65s/it]
Test accuracy: 0.44:  12%|█▎        | 4/32 [00:26<03:06,  6.65s/it]
Test accuracy: 0.44:  16%|█▌        | 5/32 [00:32<02:55,  6.49s/it]
Test accuracy: 0.56:  16%|█▌        | 5/32 [00:32<02:55,  6.49s/it]
Test accuracy: 0.57:  16%|█▌        | 5/32 [00:32<02:55,  6.49s/it]
Test accuracy: 0.5:  16%|█▌        | 5/32 [00:32<02:55,  6.49s/it] 
Test accuracy: 0.53:  16%|█▌        | 5/32 [00:32<02:55,  6.49s/it]
Test accuracy: 0.55:  16%|█▌        | 5/32 [00:32<02:55,  6.49s/it]
Test accuracy: 0.44:  16%|█▌        | 5/32 [00:32<02:54,  6.48s/it]
Test accuracy: 0.47:  16%|█▌        | 5/32 [00:32<02:54,  6.48s/it]
Test accuracy: 0.57:  19%|█▉        | 6/32 [00:39<02:46,  6.39s/it]
Test accuracy: 0.52:  19%|█▉        | 6/32 [00:39<02:46,  6.39s/it]
Test accuracy: 0.5:  19%|█▉        | 6/32 [00:39<02:46,  6.39s/it]
Test accuracy: 0.48:  19%|█▉        | 6/32 [00:39<02:46,  6.39s/it]
Test accuracy: 0.55:  19%|█▉        | 6/32 [00:39<02:46,  6.39s/it]
Test accuracy: 0.47:  19%|█▉        | 6/32 [00:38<02:46,  6.39s/it]
Test accuracy: 0.56:  19%|█▉        | 6/32 [00:39<02:46,  6.39s/it]
Test accuracy: 0.54:  19%|█▉        | 6/32 [00:38<02:46,  6.39s/it]
Test accuracy: 0.52:  22%|██▏       | 7/32 [00:45<02:41,  6.46s/it]
Test accuracy: 0.5:  22%|██▏       | 7/32 [00:45<02:41,  6.46s/it] 
Test accuracy: 0.56:  22%|██▏       | 7/32 [00:45<02:41,  6.46s/it]
Test accuracy: 0.48:  22%|██▏       | 7/32 [00:45<02:41,  6.46s/it]
Test accuracy: 0.57:  22%|██▏       | 7/32 [00:45<02:41,  6.46s/it]
Test accuracy: 0.48:  22%|██▏       | 7/32 [00:45<02:41,  6.46s/it]
Test accuracy: 0.54:  22%|██▏       | 7/32 [00:45<02:41,  6.45s/it]
Test accuracy: 0.59:  22%|██▏       | 7/32 [00:45<02:41,  6.45s/it]
Test accuracy: 0.5:  25%|██▌       | 8/32 [00:51<02:33,  6.38s/it]
Test accuracy: 0.48:  25%|██▌       | 8/32 [00:51<02:33,  6.38s/it]
Test accuracy: 0.48:  25%|██▌       | 8/32 [00:51<02:33,  6.38s/it]
Test accuracy: 0.48:  25%|██▌       | 8/32 [00:51<02:33,  6.38s/it]
Test accuracy: 0.57:  25%|██▌       | 8/32 [00:51<02:33,  6.38s/it]
Test accuracy: 0.59:  25%|██▌       | 8/32 [00:51<02:32,  6.37s/it]
Test accuracy: 0.56:  25%|██▌       | 8/32 [00:51<02:33,  6.38s/it]
Test accuracy: 0.59:  25%|██▌       | 8/32 [00:51<02:32,  6.37s/it]
Test accuracy: 0.48:  28%|██▊       | 9/32 [00:58<02:29,  6.50s/it]
Test accuracy: 0.48:  28%|██▊       | 9/32 [00:58<02:29,  6.50s/it]
Test accuracy: 0.5:  28%|██▊       | 9/32 [00:58<02:29,  6.50s/it] 
Test accuracy: 0.49:  28%|██▊       | 9/32 [00:58<02:29,  6.50s/it]
Test accuracy: 0.59:  28%|██▊       | 9/32 [00:58<02:29,  6.49s/it]
Test accuracy: 0.58:  28%|██▊       | 9/32 [00:58<02:29,  6.49s/it]
Test accuracy: 0.56:  28%|██▊       | 9/32 [00:58<02:29,  6.50s/it]
Test accuracy: 0.54:  28%|██▊       | 9/32 [00:58<02:29,  6.50s/it]
Test accuracy: 0.49:  31%|███▏      | 10/32 [01:05<02:23,  6.53s/it]
Test accuracy: 0.51:  31%|███▏      | 10/32 [01:05<02:23,  6.53s/it]
Test accuracy: 0.5:  31%|███▏      | 10/32 [01:05<02:23,  6.53s/it]
Test accuracy: 0.54:  31%|███▏      | 10/32 [01:05<02:23,  6.53s/it]
Test accuracy: 0.5:  31%|███▏      | 10/32 [01:05<02:23,  6.53s/it]
Test accuracy: 0.58:  31%|███▏      | 10/32 [01:05<02:23,  6.53s/it]
Test accuracy: 0.55:  31%|███▏      | 10/32 [01:05<02:23,  6.53s/it]
Test accuracy: 0.55:  31%|███▏      | 10/32 [01:05<02:23,  6.53s/it]
Test accuracy: 0.5:  34%|███▍      | 11/32 [01:11<02:15,  6.43s/it]
Test accuracy: 0.51:  34%|███▍      | 11/32 [01:11<02:15,  6.43s/it]
Test accuracy: 0.55:  34%|███▍      | 11/32 [01:11<02:15,  6.43s/it]
Test accuracy: 0.55:  34%|███▍      | 11/32 [01:11<02:15,  6.43s/it]
Test accuracy: 0.55:  34%|███▍      | 11/32 [01:11<02:15,  6.43s/it]
Test accuracy: 0.53:  34%|███▍      | 11/32 [01:11<02:15,  6.43s/it]
Test accuracy: 0.51:  34%|███▍      | 11/32 [01:11<02:15,  6.43s/it]
Test accuracy: 0.49:  34%|███▍      | 11/32 [01:11<02:15,  6.43s/it]
Test accuracy: 0.55:  38%|███▊      | 12/32 [01:17<02:08,  6.43s/it]
Test accuracy: 0.51:  38%|███▊      | 12/32 [01:17<02:08,  6.43s/it]
Test accuracy: 0.53:  38%|███▊      | 12/32 [01:17<02:08,  6.43s/it]
Test accuracy: 0.53:  38%|███▊      | 12/32 [01:17<02:08,  6.43s/it]
Test accuracy: 0.53:  38%|███▊      | 12/32 [01:17<02:08,  6.43s/it]
Test accuracy: 0.51:  38%|███▊      | 12/32 [01:17<02:08,  6.43s/it]
Test accuracy: 0.49:  38%|███▊      | 12/32 [01:17<02:08,  6.43s/it]
Test accuracy: 0.46:  38%|███▊      | 12/32 [01:17<02:08,  6.43s/it]
Test accuracy: 0.46:  41%|████      | 13/32 [01:24<02:01,  6.42s/it]
Test accuracy: 0.46:  41%|████      | 13/32 [01:24<02:01,  6.42s/it]
Test accuracy: 0.53:  41%|████      | 13/32 [01:24<02:01,  6.42s/it]
Test accuracy: 0.51:  41%|████      | 13/32 [01:24<02:01,  6.42s/it]
Test accuracy: 0.55:  41%|████      | 13/32 [01:24<02:01,  6.42s/it]
Test accuracy: 0.53:  41%|████      | 13/32 [01:24<02:01,  6.42s/it]
Test accuracy: 0.51:  41%|████      | 13/32 [01:24<02:01,  6.42s/it]
Test accuracy: 0.51:  41%|████      | 13/32 [01:24<02:01,  6.42s/it]
Test accuracy: 0.51:  44%|████▍     | 14/32 [01:30<01:54,  6.35s/it]
Test accuracy: 0.51:  44%|████▍     | 14/32 [01:30<01:54,  6.35s/it]
Test accuracy: 0.55:  44%|████▍     | 14/32 [01:30<01:54,  6.35s/it]
Test accuracy: 0.5:  44%|████▍     | 14/32 [01:30<01:54,  6.35s/it] 
Test accuracy: 0.51:  44%|████▍     | 14/32 [01:30<01:54,  6.35s/it]
Test accuracy: 0.54:  44%|████▍     | 14/32 [01:30<01:54,  6.35s/it]
Test accuracy: 0.46:  44%|████▍     | 14/32 [01:30<01:54,  6.35s/it]
Test accuracy: 0.48:  44%|████▍     | 14/32 [01:30<01:54,  6.35s/it]
Test accuracy: 0.51:  47%|████▋     | 15/32 [01:36<01:48,  6.37s/it]
Test accuracy: 0.54:  47%|████▋     | 15/32 [01:36<01:48,  6.37s/it]
Test accuracy: 0.49:  47%|████▋     | 15/32 [01:36<01:48,  6.37s/it]
Test accuracy: 0.56:  47%|████▋     | 15/32 [01:36<01:48,  6.37s/it]
Test accuracy: 0.5:  47%|████▋     | 15/32 [01:36<01:48,  6.37s/it]
Test accuracy: 0.5:  47%|████▋     | 15/32 [01:36<01:48,  6.37s/it]
Test accuracy: 0.48:  47%|████▋     | 15/32 [01:36<01:48,  6.37s/it]
Test accuracy: 0.49:  47%|████▋     | 15/32 [01:36<01:48,  6.37s/it]
Test accuracy: 0.56:  50%|█████     | 16/32 [01:43<01:42,  6.38s/it]
Test accuracy: 0.54:  50%|█████     | 16/32 [01:43<01:42,  6.38s/it]
Test accuracy: 0.49:  50%|█████     | 16/32 [01:43<01:42,  6.38s/it]
Test accuracy: 0.5:  50%|█████     | 16/32 [01:43<01:42,  6.38s/it]
Test accuracy: 0.47:  50%|█████     | 16/32 [01:43<01:42,  6.38s/it]
Test accuracy: 0.49:  50%|█████     | 16/32 [01:43<01:42,  6.38s/it]
Test accuracy: 0.49:  50%|█████     | 16/32 [01:43<01:42,  6.38s/it]
Test accuracy: 0.49:  50%|█████     | 16/32 [01:43<01:42,  6.38s/it]
Test accuracy: 0.49:  53%|█████▎    | 17/32 [01:49<01:34,  6.33s/it]
Test accuracy: 0.47:  53%|█████▎    | 17/32 [01:49<01:34,  6.33s/it]
Test accuracy: 0.5:  53%|█████▎    | 17/32 [01:49<01:34,  6.33s/it] 
Test accuracy: 0.48:  53%|█████▎    | 17/32 [01:49<01:34,  6.33s/it]
Test accuracy: 0.54:  53%|█████▎    | 17/32 [01:49<01:34,  6.33s/it]
Test accuracy: 0.54:  53%|█████▎    | 17/32 [01:49<01:34,  6.33s/it]
Test accuracy: 0.49:  53%|█████▎    | 17/32 [01:49<01:34,  6.33s/it]
Test accuracy: 0.5:  53%|█████▎    | 17/32 [01:49<01:34,  6.33s/it] 
Test accuracy: 0.5:  56%|█████▋    | 18/32 [01:55<01:28,  6.35s/it]
Test accuracy: 0.49:  56%|█████▋    | 18/32 [01:55<01:28,  6.35s/it]
Test accuracy: 0.48:  56%|█████▋    | 18/32 [01:55<01:28,  6.35s/it]
Test accuracy: 0.54:  56%|█████▋    | 18/32 [01:55<01:28,  6.35s/it]
Test accuracy: 0.53:  56%|█████▋    | 18/32 [01:55<01:28,  6.35s/it]
Test accuracy: 0.47:  56%|█████▋    | 18/32 [01:55<01:28,  6.35s/it]
Test accuracy: 0.5:  56%|█████▋    | 18/32 [01:55<01:28,  6.35s/it]
Test accuracy: 0.51:  56%|█████▋    | 18/32 [01:55<01:28,  6.35s/it]
Test accuracy: 0.47:  59%|█████▉    | 19/32 [02:02<01:22,  6.36s/it]
Test accuracy: 0.47:  59%|█████▉    | 19/32 [02:02<01:22,  6.36s/it]
Test accuracy: 0.51:  59%|█████▉    | 19/32 [02:02<01:22,  6.36s/it]
Test accuracy: 0.51:  59%|█████▉    | 19/32 [02:02<01:22,  6.36s/it]
Test accuracy: 0.53:  59%|█████▉    | 19/32 [02:02<01:22,  6.36s/it]
Test accuracy: 0.49:  59%|█████▉    | 19/32 [02:02<01:22,  6.36s/it]
Test accuracy: 0.53:  59%|█████▉    | 19/32 [02:02<01:22,  6.36s/it]
Test accuracy: 0.49:  59%|█████▉    | 19/32 [02:02<01:22,  6.36s/it]
Test accuracy: 0.47:  62%|██████▎   | 20/32 [02:08<01:15,  6.32s/it]
Test accuracy: 0.46:  62%|██████▎   | 20/32 [02:08<01:15,  6.32s/it]
Test accuracy: 0.53:  62%|██████▎   | 20/32 [02:08<01:15,  6.32s/it]
Test accuracy: 0.49:  62%|██████▎   | 20/32 [02:08<01:15,  6.32s/it]
Test accuracy: 0.51:  62%|██████▎   | 20/32 [02:08<01:15,  6.32s/it]
Test accuracy: 0.49:  62%|██████▎   | 20/32 [02:08<01:15,  6.32s/it]
Test accuracy: 0.51:  62%|██████▎   | 20/32 [02:08<01:15,  6.32s/it]
Test accuracy: 0.51:  62%|██████▎   | 20/32 [02:08<01:15,  6.32s/it]
Test accuracy: 0.49:  66%|██████▌   | 21/32 [02:15<01:11,  6.53s/it]
Test accuracy: 0.51:  66%|██████▌   | 21/32 [02:15<01:11,  6.53s/it]
Test accuracy: 0.46:  66%|██████▌   | 21/32 [02:15<01:11,  6.53s/it]
Test accuracy: 0.49:  66%|██████▌   | 21/32 [02:15<01:11,  6.53s/it]
Test accuracy: 0.51:  66%|██████▌   | 21/32 [02:15<01:11,  6.53s/it]
Test accuracy: 0.46:  66%|██████▌   | 21/32 [02:15<01:11,  6.53s/it]
Test accuracy: 0.51:  66%|██████▌   | 21/32 [02:15<01:11,  6.53s/it]
Test accuracy: 0.51:  66%|██████▌   | 21/32 [02:15<01:11,  6.53s/it]
Test accuracy: 0.49:  69%|██████▉   | 22/32 [02:21<01:04,  6.49s/it]
Test accuracy: 0.48:  69%|██████▉   | 22/32 [02:21<01:04,  6.49s/it]
Test accuracy: 0.51:  69%|██████▉   | 22/32 [02:21<01:04,  6.49s/it]
Test accuracy: 0.5:  69%|██████▉   | 22/32 [02:21<01:04,  6.49s/it] 
Test accuracy: 0.51:  69%|██████▉   | 22/32 [02:21<01:04,  6.49s/it]
Test accuracy: 0.51:  69%|██████▉   | 22/32 [02:21<01:04,  6.49s/it]
Test accuracy: 0.46:  69%|██████▉   | 22/32 [02:21<01:04,  6.49s/it]
Test accuracy: 0.47:  69%|██████▉   | 22/32 [02:21<01:04,  6.49s/it]
Test accuracy: 0.48:  72%|███████▏  | 23/32 [02:28<00:58,  6.47s/it]
Test accuracy: 0.47:  72%|███████▏  | 23/32 [02:28<00:58,  6.47s/it]
Test accuracy: 0.48:  72%|███████▏  | 23/32 [02:28<00:58,  6.47s/it]
Test accuracy: 0.51:  72%|███████▏  | 23/32 [02:28<00:58,  6.47s/it]
Test accuracy: 0.51:  72%|███████▏  | 23/32 [02:28<00:58,  6.47s/it]
Test accuracy: 0.49:  72%|███████▏  | 23/32 [02:28<00:58,  6.47s/it]
Test accuracy: 0.5:  72%|███████▏  | 23/32 [02:28<00:58,  6.47s/it]
Test accuracy: 0.51:  72%|███████▏  | 23/32 [02:28<00:58,  6.47s/it]
Test accuracy: 0.51:  75%|███████▌  | 24/32 [02:35<00:52,  6.57s/it]
Test accuracy: 0.49:  75%|███████▌  | 24/32 [02:35<00:52,  6.57s/it]
Test accuracy: 0.51:  75%|███████▌  | 24/32 [02:35<00:52,  6.57s/it]
Test accuracy: 0.48:  75%|███████▌  | 24/32 [02:35<00:52,  6.57s/it]
Test accuracy: 0.52:  75%|███████▌  | 24/32 [02:35<00:52,  6.57s/it]
Test accuracy: 0.48:  75%|███████▌  | 24/32 [02:35<00:52,  6.57s/it]
Test accuracy: 0.49:  75%|███████▌  | 24/32 [02:34<00:52,  6.58s/it]
Test accuracy: 0.48:  75%|███████▌  | 24/32 [02:34<00:52,  6.58s/it]
Test accuracy: 0.48:  78%|███████▊  | 25/32 [02:41<00:45,  6.53s/it]
Test accuracy: 0.49:  78%|███████▊  | 25/32 [02:41<00:45,  6.53s/it]
Test accuracy: 0.48:  78%|███████▊  | 25/32 [02:41<00:45,  6.53s/it]
Test accuracy: 0.49:  78%|███████▊  | 25/32 [02:41<00:45,  6.53s/it]
Test accuracy: 0.48:  78%|███████▊  | 25/32 [02:41<00:45,  6.53s/it]
Test accuracy: 0.49:  78%|███████▊  | 25/32 [02:41<00:45,  6.53s/it]
Test accuracy: 0.52:  78%|███████▊  | 25/32 [02:41<00:45,  6.53s/it]
Test accuracy: 0.51:  78%|███████▊  | 25/32 [02:41<00:45,  6.53s/it]
Test accuracy: 0.51:  81%|████████▏ | 26/32 [02:47<00:38,  6.49s/it]
Test accuracy: 0.49:  81%|████████▏ | 26/32 [02:47<00:38,  6.49s/it]
Test accuracy: 0.51:  81%|████████▏ | 26/32 [02:47<00:38,  6.49s/it]
Test accuracy: 0.49:  81%|████████▏ | 26/32 [02:47<00:38,  6.49s/it]
Test accuracy: 0.48:  81%|████████▏ | 26/32 [02:47<00:38,  6.49s/it]
Test accuracy: 0.49:  81%|████████▏ | 26/32 [02:47<00:38,  6.49s/it]
Test accuracy: 0.49:  81%|████████▏ | 26/32 [02:47<00:38,  6.49s/it]
Test accuracy: 0.5:  81%|████████▏ | 26/32 [02:47<00:38,  6.49s/it] 
Test accuracy: 0.5:  84%|████████▍ | 27/32 [02:53<00:32,  6.41s/it]
Test accuracy: 0.49:  84%|████████▍ | 27/32 [02:54<00:32,  6.41s/it]
Test accuracy: 0.5:  84%|████████▍ | 27/32 [02:53<00:32,  6.41s/it]
Test accuracy: 0.49:  84%|████████▍ | 27/32 [02:54<00:32,  6.41s/it]
Test accuracy: 0.49:  84%|████████▍ | 27/32 [02:54<00:32,  6.41s/it]
Test accuracy: 0.49:  84%|████████▍ | 27/32 [02:54<00:32,  6.41s/it]
Test accuracy: 0.51:  84%|████████▍ | 27/32 [02:54<00:32,  6.41s/it]
Test accuracy: 0.51:  84%|████████▍ | 27/32 [02:54<00:32,  6.41s/it]
Test accuracy: 0.5:  88%|████████▊ | 28/32 [03:00<00:26,  6.57s/it]
Test accuracy: 0.49:  88%|████████▊ | 28/32 [03:01<00:26,  6.57s/it]
Test accuracy: 0.51:  88%|████████▊ | 28/32 [03:01<00:26,  6.57s/it]
Test accuracy: 0.5:  88%|████████▊ | 28/32 [03:01<00:26,  6.57s/it] 
Test accuracy: 0.5:  88%|████████▊ | 28/32 [03:00<00:26,  6.57s/it]
Test accuracy: 0.51:  88%|████████▊ | 28/32 [03:01<00:26,  6.57s/it]
Test accuracy: 0.49:  88%|████████▊ | 28/32 [03:01<00:26,  6.58s/it]
Test accuracy: 0.49:  88%|████████▊ | 28/32 [03:01<00:26,  6.58s/it]
Test accuracy: 0.5:  91%|█████████ | 29/32 [03:07<00:19,  6.53s/it]
Test accuracy: 0.51:  91%|█████████ | 29/32 [03:07<00:19,  6.53s/it]
Test accuracy: 0.49:  91%|█████████ | 29/32 [03:07<00:19,  6.53s/it]
Test accuracy: 0.51:  91%|█████████ | 29/32 [03:07<00:19,  6.53s/it]
Test accuracy: 0.5:  91%|█████████ | 29/32 [03:07<00:19,  6.53s/it]
Test accuracy: 0.49:  91%|█████████ | 29/32 [03:07<00:19,  6.53s/it]
Test accuracy: 0.5:  91%|█████████ | 29/32 [03:07<00:19,  6.53s/it]
Test accuracy: 0.49:  91%|█████████ | 29/32 [03:07<00:19,  6.53s/it]
Test accuracy: 0.5:  94%|█████████▍| 30/32 [03:13<00:12,  6.49s/it]
Test accuracy: 0.49:  94%|█████████▍| 30/32 [03:13<00:12,  6.49s/it]
Test accuracy: 0.49:  94%|█████████▍| 30/32 [03:13<00:12,  6.49s/it]
Test accuracy: 0.5:  94%|█████████▍| 30/32 [03:13<00:12,  6.49s/it]
Test accuracy: 0.49:  94%|█████████▍| 30/32 [03:13<00:12,  6.49s/it]
Test accuracy: 0.48:  94%|█████████▍| 30/32 [03:13<00:12,  6.49s/it]
Test accuracy: 0.51:  94%|█████████▍| 30/32 [03:13<00:12,  6.49s/it]
Test accuracy: 0.51:  94%|█████████▍| 30/32 [03:13<00:12,  6.49s/it]
Test accuracy: 0.51:  97%|█████████▋| 31/32 [03:20<00:06,  6.57s/it]
Test accuracy: 0.49:  97%|█████████▋| 31/32 [03:20<00:06,  6.57s/it]
Test accuracy: 0.5:  97%|█████████▋| 31/32 [03:20<00:06,  6.57s/it]
Test accuracy: 0.49:  97%|█████████▋| 31/32 [03:20<00:06,  6.57s/it]
Test accuracy: 0.51:  97%|█████████▋| 31/32 [03:20<00:06,  6.57s/it]
Test accuracy: 0.5:  97%|█████████▋| 31/32 [03:20<00:06,  6.57s/it]
Test accuracy: 0.48:  97%|█████████▋| 31/32 [03:20<00:06,  6.58s/it]
Test accuracy: 0.48:  97%|█████████▋| 31/32 [03:20<00:06,  6.58s/it]
Test accuracy: 0.5: 100%|██████████| 32/32 [03:23<00:00,  5.37s/it]
Test accuracy: 0.5: 100%|██████████| 32/32 [03:23<00:00,  5.37s/it]
Test accuracy: 0.51: 100%|██████████| 32/32 [03:23<00:00,  5.38s/it]
Test accuracy: 0.5: 100%|██████████| 32/32 [03:23<00:00,  5.38s/it] 
Test accuracy: 0.48: 100%|██████████| 32/32 [03:23<00:00,  5.38s/it]
Test accuracy: 0.49: 100%|██████████| 32/32 [03:23<00:00,  5.38s/it]
Test accuracy: 0.49: 100%|██████████| 32/32 [03:23<00:00,  5.38s/it]
Test accuracy: 0.49: 100%|██████████| 32/32 [03:23<00:00,  5.38s/it]
Test accuracy: 0.49: 100%|██████████| 32/32 [03:23<00:00,  6.35s/it]
Device 2: Cor=122, CoT=0, Total=250

Test accuracy: 0.5: 100%|██████████| 32/32 [03:23<00:00,  6.35s/it]
Device 1: Cor=126, CoT=0, Total=250

Test accuracy: 0.49: 100%|██████████| 32/32 [03:23<00:00,  6.35s/it]
Device 3: Cor=122, CoT=0, Total=250

Test accuracy: 0.5: 100%|██████████| 32/32 [03:23<00:00,  6.35s/it]
Device 0: Cor=125, CoT=0, Total=250

============================================================
EVALUATION RESULTS (Epoch 0)
============================================================
Config: data_curriculum=ON, thought_curriculum=OFF
thought_stage=10, data_stage=1, n_latent_tokens=10
Training on difficulties: 0-1

Overall accuracy: 495 / 1000 = 0.4950

Per-difficulty accuracy:
  diff=1: 63/100 = 0.6300 (trained)
  diff=2: 44/100 = 0.4400 (untrained)
  diff=3: 54/100 = 0.5400 (untrained)
  diff=4: 52/100 = 0.5200 (untrained)
  diff=5: 47/100 = 0.4700 (untrained)
  diff=6: 48/100 = 0.4800 (untrained)
  diff=7: 42/100 = 0.4200 (untrained)
  diff=8: 49/100 = 0.4900 (untrained)
  diff=9: 46/100 = 0.4600 (untrained)
  diff=10: 50/100 = 0.5000 (untrained)

Trained (diff 0-1): 63/100 = 0.6300
============================================================

  Staging check: acc_trained=0.6300, acc_diff1=0.6300 (threshold=0.9)
Warning: failed to upload checkpoint_1 to HF: (Request ID: Root=1-6a1b63fc-1b6961ae517a5552129a95dd;4736ce48-152a-4a0a-9ba7-35dd135dd824)

403 Forbidden: You need to setup automatic credit recharge in order to upload more data. You can do so at /settings/billing..
Cannot access content at: https://huggingface.co/seyedparsa/thoughtformer-checkpoints.git/info/lfs/objects/batch.
Make sure your token has the correct permissions.

============================================================
EPOCH 1/100 (thought_stage=10, data_stage=1)
============================================================
  thought_stage=10, c_thought=1, max_difficulty=1

Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):  15%|█▌        | 152/1000 [00:00<00:00, 1071.97 examples/s]
Map (num_proc=4):  15%|█▌        | 153/1000 [00:00<00:00, 1042.65 examples/s]
Map (num_proc=4):  15%|█▌        | 153/1000 [00:00<00:00, 1006.93 examples/s]
Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):  80%|████████  | 805/1000 [00:00<00:00, 3795.71 examples/s]
Map (num_proc=4):  81%|████████  | 810/1000 [00:00<00:00, 3641.38 examples/s]
Map (num_proc=4):  81%|████████▏ | 814/1000 [00:00<00:00, 3672.44 examples/s]
Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2756.98 examples/s]

Map (num_proc=4):  16%|█▌        | 157/1000 [00:00<00:01, 763.00 examples/s]
Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2208.41 examples/s]

Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2147.89 examples/s]

Map (num_proc=4):  82%|████████▏ | 819/1000 [00:00<00:00, 3155.62 examples/s]
Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2234.02 examples/s]

Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:04, 2072.69 examples/s]
Filter (num_proc=4):  20%|██        | 2000/10000 [00:00<00:02, 3772.30 examples/s]
Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:04, 2001.14 examples/s]
Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:04, 2081.88 examples/s]
Filter (num_proc=4):  30%|███       | 3000/10000 [00:00<00:01, 5103.52 examples/s]
Filter (num_proc=4):  20%|██        | 2000/10000 [00:00<00:02, 3776.48 examples/s]
Filter (num_proc=4):  20%|██        | 2000/10000 [00:00<00:02, 3858.31 examples/s]
Filter (num_proc=4):  50%|█████     | 5000/10000 [00:00<00:00, 8620.72 examples/s]
Filter (num_proc=4):  30%|███       | 3000/10000 [00:00<00:01, 4846.99 examples/s]
Filter (num_proc=4):  40%|████      | 4000/10000 [00:00<00:00, 7038.18 examples/s]
Filter (num_proc=4):  75%|███████▌  | 7500/10000 [00:00<00:00, 11901.86 examples/s]
Filter (num_proc=4):  60%|██████    | 6000/10000 [00:00<00:00, 10628.35 examples/s]
Filter (num_proc=4):  60%|██████    | 6000/10000 [00:00<00:00, 9995.35 examples/s]
Filter (num_proc=4):  90%|█████████ | 9000/10000 [00:01<00:00, 12410.65 examples/s]
Filter (num_proc=4):  80%|████████  | 8000/10000 [00:00<00:00, 11771.33 examples/s]
Filter (num_proc=4):  80%|████████  | 8000/10000 [00:00<00:00, 11880.70 examples/s]
Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:04, 1893.67 examples/s]
Filter (num_proc=4):  95%|█████████▌| 9500/10000 [00:01<00:00, 12098.92 examples/s]
Filter (num_proc=4):  20%|██        | 2000/10000 [00:00<00:02, 3473.54 examples/s]
Filter (num_proc=4):  95%|█████████▌| 9500/10000 [00:01<00:00, 11184.56 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 7669.01 examples/s]

Filter (num_proc=4):  30%|███       | 3000/10000 [00:00<00:01, 4758.19 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 7501.69 examples/s]

Filter (num_proc=4):  55%|█████▌    | 5500/10000 [00:00<00:00, 9101.49 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 7316.44 examples/s]

Filter (num_proc=4):  70%|███████   | 7000/10000 [00:01<00:00, 10111.89 examples/s]
Filter (num_proc=4):  95%|█████████▌| 9500/10000 [00:01<00:00, 11950.70 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 7192.61 examples/s]

Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):  13%|█▎        | 130/1000 [00:00<00:04, 191.66 examples/s]
Map (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:02, 360.62 examples/s]
Map (num_proc=4):  38%|███▊      | 383/1000 [00:01<00:01, 409.95 examples/s]
Map (num_proc=4):  50%|█████     | 500/1000 [00:01<00:00, 528.94 examples/s]
Map (num_proc=4):  63%|██████▎   | 631/1000 [00:01<00:00, 539.78 examples/s]
Map (num_proc=4):  75%|███████▌  | 750/1000 [00:01<00:00, 656.60 examples/s]
Map (num_proc=4):  88%|████████▊ | 885/1000 [00:01<00:00, 578.37 examples/s]
Map (num_proc=4): 100%|██████████| 1000/1000 [00:01<00:00, 663.22 examples/s]
Map (num_proc=4): 100%|██████████| 1000/1000 [00:02<00:00, 468.06 examples/s]

Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 1509.02 examples/s]
Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 1495.45 examples/s]
Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 1425.95 examples/s]
Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2902.42 examples/s]

Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2915.26 examples/s]

Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2766.69 examples/s]

Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 1014.74 examples/s]
Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 1975.50 examples/s]

Map (num_proc=4):   0%|          | 0/100 [00:00<?, ? examples/s]
Map (num_proc=4):  25%|██▌       | 25/100 [00:00<00:00, 114.59 examples/s]
Map (num_proc=4): 100%|██████████| 100/100 [00:00<00:00, 318.00 examples/s]
Map (num_proc=4): 100%|██████████| 100/100 [00:00<00:00, 178.73 examples/s]

Training Epoch: 2:   0%|          | 0/31 [00:00<?, ?it/s]
Training Epoch: 2:   0%|          | 0/31 [00:00<?, ?it/s]
Training Epoch: 2:   0%|          | 0/31 [00:00<?, ?it/s]
Training Epoch: 2:   0%|          | 0/31 [00:00<?, ?it/s]logging training data
rank 0 | gpu 0 | alloc=4.20 GB | reserved=57.00 GB | max=41.99 GB
rank 2 | gpu 2 | alloc=4.20 GB | reserved=57.00 GB | max=41.98 GB
rank 3 | gpu 3 | alloc=4.20 GB | reserved=57.00 GB | max=41.99 GB
rank 1 | gpu 1 | alloc=4.20 GB | reserved=51.44 GB | max=41.98 GB

Training Epoch: 2/100, batch 0/63 completed (loss: 0.1042:   0%|          | 0/31 [00:03<?, ?it/s]
Training Epoch: 2/100, batch 0/63 completed (loss: 0.2181:   0%|          | 0/31 [00:03<?, ?it/s]
Training Epoch: 2/100, batch 0/63 completed (loss: 0.2117:   0%|          | 0/31 [00:03<?, ?it/s]
Training Epoch: 2/100, batch 0/63 completed (loss: 0.255:   0%|          | 0/31 [00:03<?, ?it/s]
Training Epoch: 2/100, batch 0/63 completed (loss: 0.255:   3%|▎         | 1/31 [00:05<02:57,  5.92s/it]
Training Epoch: 2/100, batch 0/63 completed (loss: 0.2117:   3%|▎         | 1/31 [00:05<02:57,  5.91s/it]
Training Epoch: 2/100, batch 0/63 completed (loss: 0.2181:   3%|▎         | 1/31 [00:05<02:57,  5.92s/it]
Training Epoch: 2/100, batch 0/63 completed (loss: 0.1042:   3%|▎         | 1/31 [00:05<02:57,  5.92s/it]
Training Epoch: 2/100, batch 1/63 completed (loss: 0.1652:   3%|▎         | 1/31 [00:05<02:57,  5.92s/it]
Training Epoch: 2/100, batch 1/63 completed (loss: 0.1724:   3%|▎         | 1/31 [00:05<02:57,  5.91s/it]
Training Epoch: 2/100, batch 1/63 completed (loss: 0.1948:   3%|▎         | 1/31 [00:05<02:57,  5.92s/it]
Training Epoch: 2/100, batch 1/63 completed (loss: 0.1846:   3%|▎         | 1/31 [00:05<02:57,  5.92s/it]
Training Epoch: 2/100, batch 2/63 completed (loss: 0.2256:   3%|▎         | 1/31 [00:08<02:57,  5.92s/it]
Training Epoch: 2/100, batch 2/63 completed (loss: 0.1353:   3%|▎         | 1/31 [00:08<02:57,  5.92s/it]
Training Epoch: 2/100, batch 2/63 completed (loss: 0.3239:   3%|▎         | 1/31 [00:08<02:57,  5.91s/it]
Training Epoch: 2/100, batch 2/63 completed (loss: 0.2524:   3%|▎         | 1/31 [00:08<02:57,  5.92s/it]
Training Epoch: 2/100, batch 2/63 completed (loss: 0.3239:   6%|▋         | 2/31 [00:11<02:49,  5.85s/it]
Training Epoch: 2/100, batch 2/63 completed (loss: 0.2256:   6%|▋         | 2/31 [00:11<02:49,  5.85s/it]
Training Epoch: 2/100, batch 2/63 completed (loss: 0.1353:   6%|▋         | 2/31 [00:11<02:49,  5.85s/it]
Training Epoch: 2/100, batch 2/63 completed (loss: 0.2524:   6%|▋         | 2/31 [00:11<02:49,  5.86s/it]
Training Epoch: 2/100, batch 3/63 completed (loss: 0.1867:   6%|▋         | 2/31 [00:11<02:49,  5.85s/it]
Training Epoch: 2/100, batch 3/63 completed (loss: 0.3185:   6%|▋         | 2/31 [00:11<02:49,  5.85s/it]
Training Epoch: 2/100, batch 3/63 completed (loss: 0.3162:   6%|▋         | 2/31 [00:11<02:49,  5.85s/it]
Training Epoch: 2/100, batch 3/63 completed (loss: 0.234:   6%|▋         | 2/31 [00:11<02:49,  5.86s/it] 
Training Epoch: 2/100, batch 4/63 completed (loss: 0.1617:   6%|▋         | 2/31 [00:14<02:49,  5.85s/it]
Training Epoch: 2/100, batch 4/63 completed (loss: 0.426:   6%|▋         | 2/31 [00:14<02:49,  5.85s/it] 
Training Epoch: 2/100, batch 4/63 completed (loss: 0.147:   6%|▋         | 2/31 [00:14<02:49,  5.85s/it] 
Training Epoch: 2/100, batch 4/63 completed (loss: 0.2174:   6%|▋         | 2/31 [00:14<02:49,  5.86s/it]
Training Epoch: 2/100, batch 4/63 completed (loss: 0.1617:  10%|▉         | 3/31 [00:17<02:44,  5.88s/it]
Training Epoch: 2/100, batch 4/63 completed (loss: 0.147:  10%|▉         | 3/31 [00:17<02:44,  5.88s/it]
Training Epoch: 2/100, batch 4/63 completed (loss: 0.426:  10%|▉         | 3/31 [00:17<02:44,  5.88s/it]
Training Epoch: 2/100, batch 4/63 completed (loss: 0.2174:  10%|▉         | 3/31 [00:17<02:44,  5.88s/it]
Training Epoch: 2/100, batch 5/63 completed (loss: 0.1883:  10%|▉         | 3/31 [00:17<02:44,  5.88s/it]
Training Epoch: 2/100, batch 5/63 completed (loss: 0.1092:  10%|▉         | 3/31 [00:17<02:44,  5.88s/it]
Training Epoch: 2/100, batch 5/63 completed (loss: 0.3261:  10%|▉         | 3/31 [00:17<02:44,  5.88s/it]
Training Epoch: 2/100, batch 5/63 completed (loss: 0.37:  10%|▉         | 3/31 [00:17<02:44,  5.88s/it]  
Training Epoch: 2/100, batch 6/63 completed (loss: 0.1346:  10%|▉         | 3/31 [00:20<02:44,  5.88s/it]
Training Epoch: 2/100, batch 6/63 completed (loss: 0.445:  10%|▉         | 3/31 [00:20<02:44,  5.88s/it] 
Training Epoch: 2/100, batch 6/63 completed (loss: 0.2756:  10%|▉         | 3/31 [00:20<02:44,  5.88s/it]
Training Epoch: 2/100, batch 6/63 completed (loss: 0.263:  10%|▉         | 3/31 [00:20<02:44,  5.88s/it]
Training Epoch: 2/100, batch 6/63 completed (loss: 0.445:  13%|█▎        | 4/31 [00:23<02:35,  5.77s/it]
Training Epoch: 2/100, batch 6/63 completed (loss: 0.2756:  13%|█▎        | 4/31 [00:23<02:35,  5.78s/it]
Training Epoch: 2/100, batch 6/63 completed (loss: 0.1346:  13%|█▎        | 4/31 [00:23<02:35,  5.78s/it]
Training Epoch: 2/100, batch 6/63 completed (loss: 0.263:  13%|█▎        | 4/31 [00:23<02:35,  5.78s/it]
Training Epoch: 2/100, batch 7/63 completed (loss: 0.116:  13%|█▎        | 4/31 [00:23<02:35,  5.77s/it]
Training Epoch: 2/100, batch 7/63 completed (loss: 0.2863:  13%|█▎        | 4/31 [00:23<02:35,  5.78s/it]
Training Epoch: 2/100, batch 7/63 completed (loss: 0.1394:  13%|█▎        | 4/31 [00:23<02:35,  5.78s/it]
Training Epoch: 2/100, batch 7/63 completed (loss: 0.2751:  13%|█▎        | 4/31 [00:23<02:35,  5.78s/it]
Training Epoch: 2/100, batch 8/63 completed (loss: 0.2265:  13%|█▎        | 4/31 [00:26<02:35,  5.77s/it]
Training Epoch: 2/100, batch 8/63 completed (loss: 0.3792:  13%|█▎        | 4/31 [00:26<02:35,  5.78s/it]
Training Epoch: 2/100, batch 8/63 completed (loss: 0.2307:  13%|█▎        | 4/31 [00:26<02:35,  5.78s/it]
Training Epoch: 2/100, batch 8/63 completed (loss: 0.1761:  13%|█▎        | 4/31 [00:26<02:35,  5.78s/it]
Training Epoch: 2/100, batch 8/63 completed (loss: 0.1761:  16%|█▌        | 5/31 [00:29<02:32,  5.85s/it]
Training Epoch: 2/100, batch 8/63 completed (loss: 0.2307:  16%|█▌        | 5/31 [00:29<02:32,  5.86s/it]
Training Epoch: 2/100, batch 8/63 completed (loss: 0.2265:  16%|█▌        | 5/31 [00:29<02:32,  5.86s/it]
Training Epoch: 2/100, batch 8/63 completed (loss: 0.3792:  16%|█▌        | 5/31 [00:29<02:32,  5.86s/it]
Training Epoch: 2/100, batch 9/63 completed (loss: 0.1144:  16%|█▌        | 5/31 [00:29<02:32,  5.86s/it]
Training Epoch: 2/100, batch 9/63 completed (loss: 0.3161:  16%|█▌        | 5/31 [00:29<02:32,  5.86s/it]
Training Epoch: 2/100, batch 9/63 completed (loss: 0.1788:  16%|█▌        | 5/31 [00:29<02:32,  5.86s/it]
Training Epoch: 2/100, batch 9/63 completed (loss: 0.3794:  16%|█▌        | 5/31 [00:29<02:32,  5.85s/it]
Training Epoch: 2/100, batch 10/63 completed (loss: 0.3506:  16%|█▌        | 5/31 [00:32<02:32,  5.86s/it]
Training Epoch: 2/100, batch 10/63 completed (loss: 0.1879:  16%|█▌        | 5/31 [00:32<02:32,  5.86s/it]
Training Epoch: 2/100, batch 10/63 completed (loss: 0.2034:  16%|█▌        | 5/31 [00:32<02:32,  5.86s/it]
Training Epoch: 2/100, batch 10/63 completed (loss: 0.2785:  16%|█▌        | 5/31 [00:32<02:32,  5.85s/it]
Training Epoch: 2/100, batch 10/63 completed (loss: 0.3506:  19%|█▉        | 6/31 [00:35<02:25,  5.84s/it]
Training Epoch: 2/100, batch 10/63 completed (loss: 0.1879:  19%|█▉        | 6/31 [00:35<02:25,  5.84s/it]
Training Epoch: 2/100, batch 10/63 completed (loss: 0.2034:  19%|█▉        | 6/31 [00:35<02:26,  5.84s/it]
Training Epoch: 2/100, batch 10/63 completed (loss: 0.2785:  19%|█▉        | 6/31 [00:35<02:26,  5.85s/it]
Training Epoch: 2/100, batch 11/63 completed (loss: 0.1841:  19%|█▉        | 6/31 [00:35<02:26,  5.84s/it]
Training Epoch: 2/100, batch 11/63 completed (loss: 0.1538:  19%|█▉        | 6/31 [00:35<02:25,  5.84s/it]
Training Epoch: 2/100, batch 11/63 completed (loss: 0.1208:  19%|█▉        | 6/31 [00:35<02:25,  5.84s/it]
Training Epoch: 2/100, batch 11/63 completed (loss: 0.122:  19%|█▉        | 6/31 [00:35<02:26,  5.85s/it] 
Training Epoch: 2/100, batch 12/63 completed (loss: 0.2028:  19%|█▉        | 6/31 [00:37<02:25,  5.84s/it]
Training Epoch: 2/100, batch 12/63 completed (loss: 0.1678:  19%|█▉        | 6/31 [00:37<02:26,  5.84s/it]
Training Epoch: 2/100, batch 12/63 completed (loss: 0.4105:  19%|█▉        | 6/31 [00:37<02:25,  5.84s/it]
Training Epoch: 2/100, batch 12/63 completed (loss: 0.2274:  19%|█▉        | 6/31 [00:37<02:26,  5.85s/it]
Training Epoch: 2/100, batch 12/63 completed (loss: 0.4105:  23%|██▎       | 7/31 [00:40<02:20,  5.84s/it]
Training Epoch: 2/100, batch 12/63 completed (loss: 0.2274:  23%|██▎       | 7/31 [00:40<02:20,  5.84s/it]
Training Epoch: 2/100, batch 12/63 completed (loss: 0.1678:  23%|██▎       | 7/31 [00:40<02:20,  5.84s/it]
Training Epoch: 2/100, batch 12/63 completed (loss: 0.2028:  23%|██▎       | 7/31 [00:40<02:20,  5.84s/it]
Training Epoch: 2/100, batch 13/63 completed (loss: 0.5374:  23%|██▎       | 7/31 [00:40<02:20,  5.84s/it]
Training Epoch: 2/100, batch 13/63 completed (loss: 0.148:  23%|██▎       | 7/31 [00:40<02:20,  5.84s/it] 
Training Epoch: 2/100, batch 13/63 completed (loss: 0.2417:  23%|██▎       | 7/31 [00:40<02:20,  5.84s/it]
Training Epoch: 2/100, batch 13/63 completed (loss: 0.1018:  23%|██▎       | 7/31 [00:40<02:20,  5.84s/it]
Training Epoch: 2/100, batch 14/63 completed (loss: 0.1214:  23%|██▎       | 7/31 [00:43<02:20,  5.84s/it]
Training Epoch: 2/100, batch 14/63 completed (loss: 0.2595:  23%|██▎       | 7/31 [00:43<02:20,  5.84s/it]
Training Epoch: 2/100, batch 14/63 completed (loss: 0.2176:  23%|██▎       | 7/31 [00:43<02:20,  5.84s/it]
Training Epoch: 2/100, batch 14/63 completed (loss: 0.1678:  23%|██▎       | 7/31 [00:43<02:20,  5.84s/it]
Training Epoch: 2/100, batch 14/63 completed (loss: 0.2176:  26%|██▌       | 8/31 [00:46<02:15,  5.89s/it]
Training Epoch: 2/100, batch 14/63 completed (loss: 0.1214:  26%|██▌       | 8/31 [00:46<02:15,  5.89s/it]
Training Epoch: 2/100, batch 14/63 completed (loss: 0.1678:  26%|██▌       | 8/31 [00:46<02:15,  5.89s/it]
Training Epoch: 2/100, batch 14/63 completed (loss: 0.2595:  26%|██▌       | 8/31 [00:46<02:15,  5.90s/it]
Training Epoch: 2/100, batch 15/63 completed (loss: 0.3567:  26%|██▌       | 8/31 [00:46<02:15,  5.89s/it]
Training Epoch: 2/100, batch 15/63 completed (loss: 0.3952:  26%|██▌       | 8/31 [00:46<02:15,  5.90s/it]
Training Epoch: 2/100, batch 15/63 completed (loss: 0.1728:  26%|██▌       | 8/31 [00:46<02:15,  5.89s/it]
Training Epoch: 2/100, batch 15/63 completed (loss: 0.3955:  26%|██▌       | 8/31 [00:46<02:15,  5.89s/it]
Training Epoch: 2/100, batch 16/63 completed (loss: 0.1537:  26%|██▌       | 8/31 [00:49<02:15,  5.90s/it]
Training Epoch: 2/100, batch 16/63 completed (loss: 0.2556:  26%|██▌       | 8/31 [00:49<02:15,  5.89s/it]
Training Epoch: 2/100, batch 16/63 completed (loss: 0.2408:  26%|██▌       | 8/31 [00:49<02:15,  5.89s/it]
Training Epoch: 2/100, batch 16/63 completed (loss: 0.2159:  26%|██▌       | 8/31 [00:49<02:15,  5.89s/it]
Training Epoch: 2/100, batch 16/63 completed (loss: 0.1537:  29%|██▉       | 9/31 [00:52<02:08,  5.83s/it]
Training Epoch: 2/100, batch 16/63 completed (loss: 0.2556:  29%|██▉       | 9/31 [00:52<02:08,  5.83s/it]
Training Epoch: 2/100, batch 16/63 completed (loss: 0.2408:  29%|██▉       | 9/31 [00:52<02:08,  5.83s/it]
Training Epoch: 2/100, batch 16/63 completed (loss: 0.2159:  29%|██▉       | 9/31 [00:52<02:08,  5.83s/it]
Training Epoch: 2/100, batch 17/63 completed (loss: 0.2909:  29%|██▉       | 9/31 [00:52<02:08,  5.83s/it]
Training Epoch: 2/100, batch 17/63 completed (loss: 0.2179:  29%|██▉       | 9/31 [00:52<02:08,  5.83s/it]
Training Epoch: 2/100, batch 17/63 completed (loss: 0.1198:  29%|██▉       | 9/31 [00:52<02:08,  5.83s/it]
Training Epoch: 2/100, batch 17/63 completed (loss: 0.2648:  29%|██▉       | 9/31 [00:52<02:08,  5.83s/it]
Training Epoch: 2/100, batch 18/63 completed (loss: 0.2514:  29%|██▉       | 9/31 [00:55<02:08,  5.83s/it]
Training Epoch: 2/100, batch 18/63 completed (loss: 0.3306:  29%|██▉       | 9/31 [00:55<02:08,  5.83s/it]
Training Epoch: 2/100, batch 18/63 completed (loss: 0.2274:  29%|██▉       | 9/31 [00:55<02:08,  5.83s/it]
Training Epoch: 2/100, batch 18/63 completed (loss: 0.2747:  29%|██▉       | 9/31 [00:55<02:08,  5.83s/it]
Training Epoch: 2/100, batch 18/63 completed (loss: 0.3306:  32%|███▏      | 10/31 [00:58<02:03,  5.88s/it]
Training Epoch: 2/100, batch 18/63 completed (loss: 0.2747:  32%|███▏      | 10/31 [00:58<02:03,  5.88s/it]
Training Epoch: 2/100, batch 18/63 completed (loss: 0.2274:  32%|███▏      | 10/31 [00:58<02:03,  5.88s/it]
Training Epoch: 2/100, batch 18/63 completed (loss: 0.2514:  32%|███▏      | 10/31 [00:58<02:03,  5.88s/it]
Training Epoch: 2/100, batch 19/63 completed (loss: 0.2773:  32%|███▏      | 10/31 [00:58<02:03,  5.88s/it]
Training Epoch: 2/100, batch 19/63 completed (loss: 0.4372:  32%|███▏      | 10/31 [00:58<02:03,  5.88s/it]
Training Epoch: 2/100, batch 19/63 completed (loss: 0.2082:  32%|███▏      | 10/31 [00:58<02:03,  5.88s/it]
Training Epoch: 2/100, batch 19/63 completed (loss: 0.1852:  32%|███▏      | 10/31 [00:58<02:03,  5.88s/it]
Training Epoch: 2/100, batch 20/63 completed (loss: 0.5332:  32%|███▏      | 10/31 [01:01<02:03,  5.88s/it]
Training Epoch: 2/100, batch 20/63 completed (loss: 0.4443:  32%|███▏      | 10/31 [01:01<02:03,  5.88s/it]
Training Epoch: 2/100, batch 20/63 completed (loss: 0.2008:  32%|███▏      | 10/31 [01:01<02:03,  5.88s/it]
Training Epoch: 2/100, batch 20/63 completed (loss: 0.1715:  32%|███▏      | 10/31 [01:01<02:03,  5.88s/it]
Training Epoch: 2/100, batch 20/63 completed (loss: 0.4443:  35%|███▌      | 11/31 [01:04<01:58,  5.91s/it]
Training Epoch: 2/100, batch 20/63 completed (loss: 0.1715:  35%|███▌      | 11/31 [01:04<01:58,  5.92s/it]
Training Epoch: 2/100, batch 20/63 completed (loss: 0.5332:  35%|███▌      | 11/31 [01:04<01:58,  5.92s/it]
Training Epoch: 2/100, batch 20/63 completed (loss: 0.2008:  35%|███▌      | 11/31 [01:04<01:58,  5.92s/it]
Training Epoch: 2/100, batch 21/63 completed (loss: 0.435:  35%|███▌      | 11/31 [01:04<01:58,  5.92s/it] 
Training Epoch: 2/100, batch 21/63 completed (loss: 0.1184:  35%|███▌      | 11/31 [01:04<01:58,  5.92s/it]
Training Epoch: 2/100, batch 21/63 completed (loss: 0.19:  35%|███▌      | 11/31 [01:04<01:58,  5.91s/it]  
Training Epoch: 2/100, batch 21/63 completed (loss: 0.1574:  35%|███▌      | 11/31 [01:04<01:58,  5.92s/it]
Training Epoch: 2/100, batch 22/63 completed (loss: 0.1798:  35%|███▌      | 11/31 [01:07<01:58,  5.92s/it]
Training Epoch: 2/100, batch 22/63 completed (loss: 0.1618:  35%|███▌      | 11/31 [01:07<01:58,  5.91s/it]
Training Epoch: 2/100, batch 22/63 completed (loss: 0.1964:  35%|███▌      | 11/31 [01:07<01:58,  5.92s/it]
Training Epoch: 2/100, batch 22/63 completed (loss: 0.2367:  35%|███▌      | 11/31 [01:07<01:58,  5.92s/it]
Training Epoch: 2/100, batch 22/63 completed (loss: 0.1964:  39%|███▊      | 12/31 [01:10<01:51,  5.85s/it]
Training Epoch: 2/100, batch 22/63 completed (loss: 0.2367:  39%|███▊      | 12/31 [01:10<01:51,  5.85s/it]
Training Epoch: 2/100, batch 22/63 completed (loss: 0.1798:  39%|███▊      | 12/31 [01:10<01:51,  5.85s/it]
Training Epoch: 2/100, batch 22/63 completed (loss: 0.1618:  39%|███▊      | 12/31 [01:10<01:51,  5.86s/it]
Training Epoch: 2/100, batch 23/63 completed (loss: 0.2504:  39%|███▊      | 12/31 [01:10<01:51,  5.85s/it]
Training Epoch: 2/100, batch 23/63 completed (loss: 0.3297:  39%|███▊      | 12/31 [01:10<01:51,  5.86s/it]
Training Epoch: 2/100, batch 23/63 completed (loss: 0.2125:  39%|███▊      | 12/31 [01:10<01:51,  5.85s/it]
Training Epoch: 2/100, batch 23/63 completed (loss: 0.2081:  39%|███▊      | 12/31 [01:10<01:51,  5.85s/it]
Training Epoch: 2/100, batch 24/63 completed (loss: 0.4403:  39%|███▊      | 12/31 [01:13<01:51,  5.85s/it]
Training Epoch: 2/100, batch 24/63 completed (loss: 0.2129:  39%|███▊      | 12/31 [01:13<01:51,  5.86s/it]
Training Epoch: 2/100, batch 24/63 completed (loss: 0.1448:  39%|███▊      | 12/31 [01:13<01:51,  5.85s/it]
Training Epoch: 2/100, batch 24/63 completed (loss: 0.1811:  39%|███▊      | 12/31 [01:13<01:51,  5.85s/it]
Training Epoch: 2/100, batch 24/63 completed (loss: 0.4403:  42%|████▏     | 13/31 [01:16<01:48,  6.00s/it]
Training Epoch: 2/100, batch 24/63 completed (loss: 0.1448:  42%|████▏     | 13/31 [01:16<01:48,  6.00s/it]
Training Epoch: 2/100, batch 24/63 completed (loss: 0.2129:  42%|████▏     | 13/31 [01:16<01:48,  6.00s/it]
Training Epoch: 2/100, batch 24/63 completed (loss: 0.1811:  42%|████▏     | 13/31 [01:16<01:48,  6.01s/it]
Training Epoch: 2/100, batch 25/63 completed (loss: 0.2116:  42%|████▏     | 13/31 [01:16<01:48,  6.00s/it]
Training Epoch: 2/100, batch 25/63 completed (loss: 0.1791:  42%|████▏     | 13/31 [01:16<01:48,  6.00s/it]
Training Epoch: 2/100, batch 25/63 completed (loss: 0.1998:  42%|████▏     | 13/31 [01:16<01:48,  6.00s/it]
Training Epoch: 2/100, batch 25/63 completed (loss: 0.2974:  42%|████▏     | 13/31 [01:16<01:48,  6.01s/it]
Training Epoch: 2/100, batch 26/63 completed (loss: 0.2903:  42%|████▏     | 13/31 [01:19<01:48,  6.00s/it]
Training Epoch: 2/100, batch 26/63 completed (loss: 0.3223:  42%|████▏     | 13/31 [01:19<01:48,  6.00s/it]
Training Epoch: 2/100, batch 26/63 completed (loss: 0.1559:  42%|████▏     | 13/31 [01:19<01:48,  6.00s/it]
Training Epoch: 2/100, batch 26/63 completed (loss: 0.2007:  42%|████▏     | 13/31 [01:19<01:48,  6.01s/it]
Training Epoch: 2/100, batch 26/63 completed (loss: 0.2007:  45%|████▌     | 14/31 [01:22<01:40,  5.93s/it]
Training Epoch: 2/100, batch 26/63 completed (loss: 0.3223:  45%|████▌     | 14/31 [01:22<01:40,  5.93s/it]
Training Epoch: 2/100, batch 26/63 completed (loss: 0.2903:  45%|████▌     | 14/31 [01:22<01:40,  5.93s/it]
Training Epoch: 2/100, batch 26/63 completed (loss: 0.1559:  45%|████▌     | 14/31 [01:22<01:40,  5.93s/it]
Training Epoch: 2/100, batch 27/63 completed (loss: 0.2894:  45%|████▌     | 14/31 [01:22<01:40,  5.93s/it]
Training Epoch: 2/100, batch 27/63 completed (loss: 0.1683:  45%|████▌     | 14/31 [01:22<01:40,  5.93s/it]
Training Epoch: 2/100, batch 27/63 completed (loss: 0.2296:  45%|████▌     | 14/31 [01:22<01:40,  5.93s/it]
Training Epoch: 2/100, batch 27/63 completed (loss: 0.1634:  45%|████▌     | 14/31 [01:22<01:40,  5.93s/it]
Training Epoch: 2/100, batch 28/63 completed (loss: 0.2179:  45%|████▌     | 14/31 [01:25<01:40,  5.93s/it]
Training Epoch: 2/100, batch 28/63 completed (loss: 0.1982:  45%|████▌     | 14/31 [01:25<01:40,  5.93s/it]
Training Epoch: 2/100, batch 28/63 completed (loss: 0.372:  45%|████▌     | 14/31 [01:25<01:40,  5.93s/it] 
Training Epoch: 2/100, batch 28/63 completed (loss: 0.2692:  45%|████▌     | 14/31 [01:25<01:40,  5.93s/it]
Training Epoch: 2/100, batch 28/63 completed (loss: 0.372:  48%|████▊     | 15/31 [01:28<01:34,  5.92s/it]
Training Epoch: 2/100, batch 28/63 completed (loss: 0.2179:  48%|████▊     | 15/31 [01:28<01:34,  5.92s/it]
Training Epoch: 2/100, batch 28/63 completed (loss: 0.1982:  48%|████▊     | 15/31 [01:28<01:34,  5.92s/it]
Training Epoch: 2/100, batch 28/63 completed (loss: 0.2692:  48%|████▊     | 15/31 [01:28<01:34,  5.93s/it]
Training Epoch: 2/100, batch 29/63 completed (loss: 0.3274:  48%|████▊     | 15/31 [01:28<01:34,  5.92s/it]
Training Epoch: 2/100, batch 29/63 completed (loss: 0.1732:  48%|████▊     | 15/31 [01:28<01:34,  5.92s/it]
Training Epoch: 2/100, batch 29/63 completed (loss: 0.1955:  48%|████▊     | 15/31 [01:28<01:34,  5.92s/it]
Training Epoch: 2/100, batch 29/63 completed (loss: 0.1199:  48%|████▊     | 15/31 [01:28<01:34,  5.93s/it]
Training Epoch: 2/100, batch 30/63 completed (loss: 0.1498:  48%|████▊     | 15/31 [01:31<01:34,  5.92s/it]
Training Epoch: 2/100, batch 30/63 completed (loss: 0.2346:  48%|████▊     | 15/31 [01:31<01:34,  5.92s/it]
Training Epoch: 2/100, batch 30/63 completed (loss: 0.1543:  48%|████▊     | 15/31 [01:31<01:34,  5.92s/it]
Training Epoch: 2/100, batch 30/63 completed (loss: 0.319:  48%|████▊     | 15/31 [01:31<01:34,  5.93s/it] 
Training Epoch: 2/100, batch 30/63 completed (loss: 0.1498:  52%|█████▏    | 16/31 [01:34<01:29,  5.97s/it]
Training Epoch: 2/100, batch 30/63 completed (loss: 0.319:  52%|█████▏    | 16/31 [01:34<01:29,  5.97s/it]
Training Epoch: 2/100, batch 30/63 completed (loss: 0.1543:  52%|█████▏    | 16/31 [01:34<01:29,  5.98s/it]
Training Epoch: 2/100, batch 30/63 completed (loss: 0.2346:  52%|█████▏    | 16/31 [01:34<01:29,  5.98s/it]
Training Epoch: 2/100, batch 31/63 completed (loss: 0.1949:  52%|█████▏    | 16/31 [01:34<01:29,  5.98s/it]
Training Epoch: 2/100, batch 31/63 completed (loss: 0.1392:  52%|█████▏    | 16/31 [01:34<01:29,  5.97s/it]
Training Epoch: 2/100, batch 31/63 completed (loss: 0.2278:  52%|█████▏    | 16/31 [01:34<01:29,  5.98s/it]
Training Epoch: 2/100, batch 31/63 completed (loss: 0.2525:  52%|█████▏    | 16/31 [01:34<01:29,  5.97s/it]
Training Epoch: 2/100, batch 32/63 completed (loss: 0.1628:  52%|█████▏    | 16/31 [01:37<01:29,  5.98s/it]
Training Epoch: 2/100, batch 32/63 completed (loss: 0.4786:  52%|█████▏    | 16/31 [01:37<01:29,  5.97s/it]
Training Epoch: 2/100, batch 32/63 completed (loss: 0.1983:  52%|█████▏    | 16/31 [01:37<01:29,  5.98s/it]
Training Epoch: 2/100, batch 32/63 completed (loss: 0.1932:  52%|█████▏    | 16/31 [01:37<01:29,  5.97s/it]
Training Epoch: 2/100, batch 32/63 completed (loss: 0.4786:  55%|█████▍    | 17/31 [01:40<01:22,  5.90s/it]
Training Epoch: 2/100, batch 32/63 completed (loss: 0.1932:  55%|█████▍    | 17/31 [01:40<01:22,  5.90s/it]
Training Epoch: 2/100, batch 32/63 completed (loss: 0.1628:  55%|█████▍    | 17/31 [01:40<01:22,  5.90s/it]
Training Epoch: 2/100, batch 32/63 completed (loss: 0.1983:  55%|█████▍    | 17/31 [01:40<01:22,  5.91s/it]
Training Epoch: 2/100, batch 33/63 completed (loss: 0.2717:  55%|█████▍    | 17/31 [01:40<01:22,  5.91s/it]
Training Epoch: 2/100, batch 33/63 completed (loss: 0.1799:  55%|█████▍    | 17/31 [01:40<01:22,  5.90s/it]
Training Epoch: 2/100, batch 33/63 completed (loss: 0.1563:  55%|█████▍    | 17/31 [01:40<01:22,  5.90s/it]
Training Epoch: 2/100, batch 33/63 completed (loss: 0.4239:  55%|█████▍    | 17/31 [01:40<01:22,  5.90s/it]
Training Epoch: 2/100, batch 34/63 completed (loss: 0.1984:  55%|█████▍    | 17/31 [01:43<01:22,  5.91s/it]
Training Epoch: 2/100, batch 34/63 completed (loss: 0.2426:  55%|█████▍    | 17/31 [01:43<01:22,  5.90s/it]
Training Epoch: 2/100, batch 34/63 completed (loss: 0.2026:  55%|█████▍    | 17/31 [01:43<01:22,  5.90s/it]
Training Epoch: 2/100, batch 34/63 completed (loss: 0.3365:  55%|█████▍    | 17/31 [01:43<01:22,  5.90s/it]
Training Epoch: 2/100, batch 34/63 completed (loss: 0.2426:  58%|█████▊    | 18/31 [01:46<01:17,  6.00s/it]
Training Epoch: 2/100, batch 34/63 completed (loss: 0.3365:  58%|█████▊    | 18/31 [01:46<01:17,  6.00s/it]
Training Epoch: 2/100, batch 34/63 completed (loss: 0.2026:  58%|█████▊    | 18/31 [01:46<01:17,  6.00s/it]
Training Epoch: 2/100, batch 34/63 completed (loss: 0.1984:  58%|█████▊    | 18/31 [01:46<01:17,  6.00s/it]
Training Epoch: 2/100, batch 35/63 completed (loss: 0.373:  58%|█████▊    | 18/31 [01:46<01:17,  6.00s/it] 
Training Epoch: 2/100, batch 35/63 completed (loss: 0.3412:  58%|█████▊    | 18/31 [01:46<01:17,  6.00s/it]
Training Epoch: 2/100, batch 35/63 completed (loss: 0.1775:  58%|█████▊    | 18/31 [01:46<01:17,  6.00s/it]
Training Epoch: 2/100, batch 35/63 completed (loss: 0.153:  58%|█████▊    | 18/31 [01:46<01:17,  6.00s/it] 
Training Epoch: 2/100, batch 36/63 completed (loss: 0.1608:  58%|█████▊    | 18/31 [01:49<01:17,  6.00s/it]
Training Epoch: 2/100, batch 36/63 completed (loss: 0.1538:  58%|█████▊    | 18/31 [01:49<01:17,  6.00s/it]
Training Epoch: 2/100, batch 36/63 completed (loss: 0.3028:  58%|█████▊    | 18/31 [01:49<01:17,  6.00s/it]
Training Epoch: 2/100, batch 36/63 completed (loss: 0.1444:  58%|█████▊    | 18/31 [01:49<01:17,  6.00s/it]
Training Epoch: 2/100, batch 36/63 completed (loss: 0.3028:  61%|██████▏   | 19/31 [01:52<01:11,  5.96s/it]
Training Epoch: 2/100, batch 36/63 completed (loss: 0.1444:  61%|██████▏   | 19/31 [01:52<01:11,  5.96s/it]
Training Epoch: 2/100, batch 36/63 completed (loss: 0.1538:  61%|██████▏   | 19/31 [01:52<01:11,  5.96s/it]
Training Epoch: 2/100, batch 36/63 completed (loss: 0.1608:  61%|██████▏   | 19/31 [01:52<01:11,  5.96s/it]
Training Epoch: 2/100, batch 37/63 completed (loss: 0.3208:  61%|██████▏   | 19/31 [01:52<01:11,  5.96s/it]
Training Epoch: 2/100, batch 37/63 completed (loss: 0.1076:  61%|██████▏   | 19/31 [01:52<01:11,  5.96s/it]
Training Epoch: 2/100, batch 37/63 completed (loss: 0.2822:  61%|██████▏   | 19/31 [01:52<01:11,  5.96s/it]
Training Epoch: 2/100, batch 37/63 completed (loss: 0.373:  61%|██████▏   | 19/31 [01:52<01:11,  5.96s/it] 
Training Epoch: 2/100, batch 38/63 completed (loss: 0.1337:  61%|██████▏   | 19/31 [01:55<01:11,  5.96s/it]
Training Epoch: 2/100, batch 38/63 completed (loss: 0.2884:  61%|██████▏   | 19/31 [01:55<01:11,  5.96s/it]
Training Epoch: 2/100, batch 38/63 completed (loss: 0.2208:  61%|██████▏   | 19/31 [01:55<01:11,  5.96s/it]
Training Epoch: 2/100, batch 38/63 completed (loss: 0.1008:  61%|██████▏   | 19/31 [01:55<01:11,  5.96s/it]
Training Epoch: 2/100, batch 38/63 completed (loss: 0.2884:  65%|██████▍   | 20/31 [01:58<01:05,  5.95s/it]
Training Epoch: 2/100, batch 38/63 completed (loss: 0.2208:  65%|██████▍   | 20/31 [01:58<01:05,  5.95s/it]
Training Epoch: 2/100, batch 38/63 completed (loss: 0.1008:  65%|██████▍   | 20/31 [01:58<01:05,  5.95s/it]
Training Epoch: 2/100, batch 38/63 completed (loss: 0.1337:  65%|██████▍   | 20/31 [01:58<01:05,  5.95s/it]
Training Epoch: 2/100, batch 39/63 completed (loss: 0.2247:  65%|██████▍   | 20/31 [01:58<01:05,  5.95s/it]
Training Epoch: 2/100, batch 39/63 completed (loss: 0.3598:  65%|██████▍   | 20/31 [01:58<01:05,  5.95s/it]
Training Epoch: 2/100, batch 39/63 completed (loss: 0.0585:  65%|██████▍   | 20/31 [01:58<01:05,  5.95s/it]
Training Epoch: 2/100, batch 39/63 completed (loss: 0.2803:  65%|██████▍   | 20/31 [01:58<01:05,  5.95s/it]
Training Epoch: 2/100, batch 40/63 completed (loss: 0.2155:  65%|██████▍   | 20/31 [02:01<01:05,  5.95s/it]
Training Epoch: 2/100, batch 40/63 completed (loss: 0.4437:  65%|██████▍   | 20/31 [02:01<01:05,  5.95s/it]
Training Epoch: 2/100, batch 40/63 completed (loss: 0.2053:  65%|██████▍   | 20/31 [02:01<01:05,  5.95s/it]
Training Epoch: 2/100, batch 40/63 completed (loss: 0.2231:  65%|██████▍   | 20/31 [02:01<01:05,  5.95s/it]
Training Epoch: 2/100, batch 40/63 completed (loss: 0.4437:  68%|██████▊   | 21/31 [02:04<00:59,  5.98s/it]
Training Epoch: 2/100, batch 40/63 completed (loss: 0.2053:  68%|██████▊   | 21/31 [02:04<00:59,  5.98s/it]
Training Epoch: 2/100, batch 40/63 completed (loss: 0.2231:  68%|██████▊   | 21/31 [02:04<00:59,  5.98s/it]
Training Epoch: 2/100, batch 40/63 completed (loss: 0.2155:  68%|██████▊   | 21/31 [02:04<00:59,  5.98s/it]
Training Epoch: 2/100, batch 41/63 completed (loss: 0.2032:  68%|██████▊   | 21/31 [02:04<00:59,  5.98s/it]
Training Epoch: 2/100, batch 41/63 completed (loss: 0.1632:  68%|██████▊   | 21/31 [02:04<00:59,  5.98s/it]
Training Epoch: 2/100, batch 41/63 completed (loss: 0.3751:  68%|██████▊   | 21/31 [02:04<00:59,  5.98s/it]
Training Epoch: 2/100, batch 41/63 completed (loss: 0.1449:  68%|██████▊   | 21/31 [02:04<00:59,  5.98s/it]
Training Epoch: 2/100, batch 42/63 completed (loss: 0.1395:  68%|██████▊   | 21/31 [02:07<00:59,  5.98s/it]
Training Epoch: 2/100, batch 42/63 completed (loss: 0.1602:  68%|██████▊   | 21/31 [02:07<00:59,  5.98s/it]
Training Epoch: 2/100, batch 42/63 completed (loss: 0.3462:  68%|██████▊   | 21/31 [02:07<00:59,  5.98s/it]
Training Epoch: 2/100, batch 42/63 completed (loss: 0.2882:  68%|██████▊   | 21/31 [02:07<00:59,  5.98s/it]
Training Epoch: 2/100, batch 42/63 completed (loss: 0.2882:  71%|███████   | 22/31 [02:10<00:53,  5.93s/it]
Training Epoch: 2/100, batch 42/63 completed (loss: 0.3462:  71%|███████   | 22/31 [02:10<00:53,  5.93s/it]
Training Epoch: 2/100, batch 42/63 completed (loss: 0.1602:  71%|███████   | 22/31 [02:10<00:53,  5.93s/it]
Training Epoch: 2/100, batch 42/63 completed (loss: 0.1395:  71%|███████   | 22/31 [02:10<00:53,  5.94s/it]
Training Epoch: 2/100, batch 43/63 completed (loss: 0.2014:  71%|███████   | 22/31 [02:10<00:53,  5.93s/it]
Training Epoch: 2/100, batch 43/63 completed (loss: 0.2048:  71%|███████   | 22/31 [02:10<00:53,  5.94s/it]
Training Epoch: 2/100, batch 43/63 completed (loss: 0.2853:  71%|███████   | 22/31 [02:10<00:53,  5.93s/it]
Training Epoch: 2/100, batch 43/63 completed (loss: 0.2577:  71%|███████   | 22/31 [02:10<00:53,  5.93s/it]
Training Epoch: 2/100, batch 44/63 completed (loss: 0.2476:  71%|███████   | 22/31 [02:12<00:53,  5.94s/it]
Training Epoch: 2/100, batch 44/63 completed (loss: 0.2174:  71%|███████   | 22/31 [02:12<00:53,  5.93s/it]
Training Epoch: 2/100, batch 44/63 completed (loss: 0.2006:  71%|███████   | 22/31 [02:12<00:53,  5.93s/it]
Training Epoch: 2/100, batch 44/63 completed (loss: 0.4141:  71%|███████   | 22/31 [02:12<00:53,  5.93s/it]
Training Epoch: 2/100, batch 44/63 completed (loss: 0.2476:  74%|███████▍  | 23/31 [02:16<00:47,  5.97s/it]
Training Epoch: 2/100, batch 44/63 completed (loss: 0.2174:  74%|███████▍  | 23/31 [02:16<00:47,  5.97s/it]
Training Epoch: 2/100, batch 44/63 completed (loss: 0.4141:  74%|███████▍  | 23/31 [02:16<00:47,  5.97s/it]
Training Epoch: 2/100, batch 44/63 completed (loss: 0.2006:  74%|███████▍  | 23/31 [02:16<00:47,  5.98s/it]
Training Epoch: 2/100, batch 45/63 completed (loss: 0.2197:  74%|███████▍  | 23/31 [02:16<00:47,  5.98s/it]
Training Epoch: 2/100, batch 45/63 completed (loss: 0.3787:  74%|███████▍  | 23/31 [02:16<00:47,  5.97s/it]
Training Epoch: 2/100, batch 45/63 completed (loss: 0.2838:  74%|███████▍  | 23/31 [02:16<00:47,  5.97s/it]
Training Epoch: 2/100, batch 45/63 completed (loss: 0.2378:  74%|███████▍  | 23/31 [02:16<00:47,  5.97s/it]
Training Epoch: 2/100, batch 46/63 completed (loss: 0.3083:  74%|███████▍  | 23/31 [02:19<00:47,  5.98s/it]
Training Epoch: 2/100, batch 46/63 completed (loss: 0.2478:  74%|███████▍  | 23/31 [02:19<00:47,  5.97s/it]
Training Epoch: 2/100, batch 46/63 completed (loss: 0.1967:  74%|███████▍  | 23/31 [02:19<00:47,  5.97s/it]
Training Epoch: 2/100, batch 46/63 completed (loss: 0.316:  74%|███████▍  | 23/31 [02:19<00:47,  5.97s/it] 
Training Epoch: 2/100, batch 46/63 completed (loss: 0.316:  77%|███████▋  | 24/31 [02:22<00:42,  6.02s/it]
Training Epoch: 2/100, batch 46/63 completed (loss: 0.3083:  77%|███████▋  | 24/31 [02:22<00:42,  6.02s/it]
Training Epoch: 2/100, batch 46/63 completed (loss: 0.1967:  77%|███████▋  | 24/31 [02:22<00:42,  6.02s/it]
Training Epoch: 2/100, batch 46/63 completed (loss: 0.2478:  77%|███████▋  | 24/31 [02:22<00:42,  6.02s/it]
Training Epoch: 2/100, batch 47/63 completed (loss: 0.168:  77%|███████▋  | 24/31 [02:22<00:42,  6.02s/it] 
Training Epoch: 2/100, batch 47/63 completed (loss: 0.2276:  77%|███████▋  | 24/31 [02:22<00:42,  6.02s/it]
Training Epoch: 2/100, batch 47/63 completed (loss: 0.257:  77%|███████▋  | 24/31 [02:22<00:42,  6.02s/it] 
Training Epoch: 2/100, batch 47/63 completed (loss: 0.2111:  77%|███████▋  | 24/31 [02:22<00:42,  6.02s/it]
Training Epoch: 2/100, batch 48/63 completed (loss: 0.2636:  77%|███████▋  | 24/31 [02:25<00:42,  6.02s/it]
Training Epoch: 2/100, batch 48/63 completed (loss: 0.1659:  77%|███████▋  | 24/31 [02:25<00:42,  6.02s/it]
Training Epoch: 2/100, batch 48/63 completed (loss: 0.183:  77%|███████▋  | 24/31 [02:25<00:42,  6.02s/it] 
Training Epoch: 2/100, batch 48/63 completed (loss: 0.454:  77%|███████▋  | 24/31 [02:25<00:42,  6.02s/it] 
Training Epoch: 2/100, batch 48/63 completed (loss: 0.1659:  81%|████████  | 25/31 [02:28<00:35,  5.95s/it]
Training Epoch: 2/100, batch 48/63 completed (loss: 0.454:  81%|████████  | 25/31 [02:28<00:35,  5.96s/it]
Training Epoch: 2/100, batch 48/63 completed (loss: 0.183:  81%|████████  | 25/31 [02:28<00:35,  5.96s/it]
Training Epoch: 2/100, batch 48/63 completed (loss: 0.2636:  81%|████████  | 25/31 [02:28<00:35,  5.96s/it]
Training Epoch: 2/100, batch 49/63 completed (loss: 0.3689:  81%|████████  | 25/31 [02:28<00:35,  5.95s/it]
Training Epoch: 2/100, batch 49/63 completed (loss: 0.366:  81%|████████  | 25/31 [02:28<00:35,  5.96s/it] 
Training Epoch: 2/100, batch 49/63 completed (loss: 0.2634:  81%|████████  | 25/31 [02:28<00:35,  5.96s/it]
Training Epoch: 2/100, batch 49/63 completed (loss: 0.2667:  81%|████████  | 25/31 [02:28<00:35,  5.96s/it]
Training Epoch: 2/100, batch 50/63 completed (loss: 0.1525:  81%|████████  | 25/31 [02:31<00:35,  5.96s/it]
Training Epoch: 2/100, batch 50/63 completed (loss: 0.2595:  81%|████████  | 25/31 [02:31<00:35,  5.96s/it]
Training Epoch: 2/100, batch 50/63 completed (loss: 0.2569:  81%|████████  | 25/31 [02:31<00:35,  5.95s/it]
Training Epoch: 2/100, batch 50/63 completed (loss: 0.3669:  81%|████████  | 25/31 [02:31<00:35,  5.96s/it]
Training Epoch: 2/100, batch 50/63 completed (loss: 0.3669:  84%|████████▍ | 26/31 [02:34<00:30,  6.05s/it]
Training Epoch: 2/100, batch 50/63 completed (loss: 0.2595:  84%|████████▍ | 26/31 [02:34<00:30,  6.05s/it]
Training Epoch: 2/100, batch 50/63 completed (loss: 0.2569:  84%|████████▍ | 26/31 [02:34<00:30,  6.05s/it]
Training Epoch: 2/100, batch 50/63 completed (loss: 0.1525:  84%|████████▍ | 26/31 [02:34<00:30,  6.05s/it]
Training Epoch: 2/100, batch 51/63 completed (loss: 0.4279:  84%|████████▍ | 26/31 [02:34<00:30,  6.05s/it]
Training Epoch: 2/100, batch 51/63 completed (loss: 0.2675:  84%|████████▍ | 26/31 [02:34<00:30,  6.05s/it]
Training Epoch: 2/100, batch 51/63 completed (loss: 0.1922:  84%|████████▍ | 26/31 [02:34<00:30,  6.05s/it]
Training Epoch: 2/100, batch 51/63 completed (loss: 0.3219:  84%|████████▍ | 26/31 [02:34<00:30,  6.05s/it]
Training Epoch: 2/100, batch 52/63 completed (loss: 0.2729:  84%|████████▍ | 26/31 [02:37<00:30,  6.05s/it]
Training Epoch: 2/100, batch 52/63 completed (loss: 0.3787:  84%|████████▍ | 26/31 [02:37<00:30,  6.05s/it]
Training Epoch: 2/100, batch 52/63 completed (loss: 0.3112:  84%|████████▍ | 26/31 [02:37<00:30,  6.05s/it]
Training Epoch: 2/100, batch 52/63 completed (loss: 0.2505:  84%|████████▍ | 26/31 [02:37<00:30,  6.05s/it]
Training Epoch: 2/100, batch 52/63 completed (loss: 0.2729:  87%|████████▋ | 27/31 [02:40<00:23,  5.97s/it]
Training Epoch: 2/100, batch 52/63 completed (loss: 0.2505:  87%|████████▋ | 27/31 [02:40<00:23,  5.98s/it]
Training Epoch: 2/100, batch 52/63 completed (loss: 0.3112:  87%|████████▋ | 27/31 [02:40<00:23,  5.97s/it]
Training Epoch: 2/100, batch 52/63 completed (loss: 0.3787:  87%|████████▋ | 27/31 [02:40<00:23,  5.97s/it]
Training Epoch: 2/100, batch 53/63 completed (loss: 0.1534:  87%|████████▋ | 27/31 [02:40<00:23,  5.97s/it]
Training Epoch: 2/100, batch 53/63 completed (loss: 0.2738:  87%|████████▋ | 27/31 [02:40<00:23,  5.97s/it]
Training Epoch: 2/100, batch 53/63 completed (loss: 0.189:  87%|████████▋ | 27/31 [02:40<00:23,  5.97s/it] 
Training Epoch: 2/100, batch 53/63 completed (loss: 0.1698:  87%|████████▋ | 27/31 [02:40<00:23,  5.98s/it]
Training Epoch: 2/100, batch 54/63 completed (loss: 0.1764:  87%|████████▋ | 27/31 [02:43<00:23,  5.97s/it]
Training Epoch: 2/100, batch 54/63 completed (loss: 0.1555:  87%|████████▋ | 27/31 [02:43<00:23,  5.97s/it]
Training Epoch: 2/100, batch 54/63 completed (loss: 0.1578:  87%|████████▋ | 27/31 [02:43<00:23,  5.97s/it]
Training Epoch: 2/100, batch 54/63 completed (loss: 0.2174:  87%|████████▋ | 27/31 [02:43<00:23,  5.98s/it]
Training Epoch: 2/100, batch 54/63 completed (loss: 0.1578:  90%|█████████ | 28/31 [02:46<00:17,  5.98s/it]
Training Epoch: 2/100, batch 54/63 completed (loss: 0.1764:  90%|█████████ | 28/31 [02:46<00:17,  5.98s/it]
Training Epoch: 2/100, batch 54/63 completed (loss: 0.1555:  90%|█████████ | 28/31 [02:46<00:17,  5.98s/it]
Training Epoch: 2/100, batch 54/63 completed (loss: 0.2174:  90%|█████████ | 28/31 [02:46<00:17,  5.98s/it]
Training Epoch: 2/100, batch 55/63 completed (loss: 0.2792:  90%|█████████ | 28/31 [02:46<00:17,  5.98s/it]
Training Epoch: 2/100, batch 55/63 completed (loss: 0.0873:  90%|█████████ | 28/31 [02:46<00:17,  5.98s/it]
Training Epoch: 2/100, batch 55/63 completed (loss: 0.126:  90%|█████████ | 28/31 [02:46<00:17,  5.98s/it] 
Training Epoch: 2/100, batch 55/63 completed (loss: 0.1757:  90%|█████████ | 28/31 [02:46<00:17,  5.98s/it]
Training Epoch: 2/100, batch 56/63 completed (loss: 0.2295:  90%|█████████ | 28/31 [02:49<00:17,  5.98s/it]
Training Epoch: 2/100, batch 56/63 completed (loss: 0.217:  90%|█████████ | 28/31 [02:49<00:17,  5.98s/it]
Training Epoch: 2/100, batch 56/63 completed (loss: 0.2064:  90%|█████████ | 28/31 [02:49<00:17,  5.98s/it]
Training Epoch: 2/100, batch 56/63 completed (loss: 0.1438:  90%|█████████ | 28/31 [02:49<00:17,  5.98s/it]
Training Epoch: 2/100, batch 56/63 completed (loss: 0.2064:  94%|█████████▎| 29/31 [02:52<00:12,  6.02s/it]
Training Epoch: 2/100, batch 56/63 completed (loss: 0.217:  94%|█████████▎| 29/31 [02:52<00:12,  6.02s/it]
Training Epoch: 2/100, batch 56/63 completed (loss: 0.1438:  94%|█████████▎| 29/31 [02:52<00:12,  6.03s/it]
Training Epoch: 2/100, batch 56/63 completed (loss: 0.2295:  94%|█████████▎| 29/31 [02:52<00:12,  6.03s/it]
Training Epoch: 2/100, batch 57/63 completed (loss: 0.1685:  94%|█████████▎| 29/31 [02:52<00:12,  6.03s/it]
Training Epoch: 2/100, batch 57/63 completed (loss: 0.3687:  94%|█████████▎| 29/31 [02:52<00:12,  6.02s/it]
Training Epoch: 2/100, batch 57/63 completed (loss: 0.4419:  94%|█████████▎| 29/31 [02:52<00:12,  6.02s/it]
Training Epoch: 2/100, batch 57/63 completed (loss: 0.15:  94%|█████████▎| 29/31 [02:52<00:12,  6.03s/it]  
Training Epoch: 2/100, batch 58/63 completed (loss: 0.2289:  94%|█████████▎| 29/31 [02:55<00:12,  6.02s/it]
Training Epoch: 2/100, batch 58/63 completed (loss: 0.2194:  94%|█████████▎| 29/31 [02:55<00:12,  6.02s/it]
Training Epoch: 2/100, batch 58/63 completed (loss: 0.5227:  94%|█████████▎| 29/31 [02:55<00:12,  6.03s/it]
Training Epoch: 2/100, batch 58/63 completed (loss: 0.2323:  94%|█████████▎| 29/31 [02:55<00:12,  6.03s/it]
Training Epoch: 2/100, batch 58/63 completed (loss: 0.2289:  97%|█████████▋| 30/31 [02:58<00:05,  5.95s/it]
Training Epoch: 2/100, batch 58/63 completed (loss: 0.2194:  97%|█████████▋| 30/31 [02:58<00:05,  5.95s/it]
Training Epoch: 2/100, batch 58/63 completed (loss: 0.5227:  97%|█████████▋| 30/31 [02:58<00:05,  5.95s/it]
Training Epoch: 2/100, batch 58/63 completed (loss: 0.2323:  97%|█████████▋| 30/31 [02:58<00:05,  5.95s/it]
Training Epoch: 2/100, batch 59/63 completed (loss: 0.3684:  97%|█████████▋| 30/31 [02:58<00:05,  5.95s/it]
Training Epoch: 2/100, batch 59/63 completed (loss: 0.4976:  97%|█████████▋| 30/31 [02:58<00:05,  5.95s/it]
Training Epoch: 2/100, batch 59/63 completed (loss: 0.1351:  97%|█████████▋| 30/31 [02:58<00:05,  5.95s/it]
Training Epoch: 2/100, batch 59/63 completed (loss: 0.1858:  97%|█████████▋| 30/31 [02:58<00:05,  5.95s/it]
Training Epoch: 2/100, batch 60/63 completed (loss: 0.2118:  97%|█████████▋| 30/31 [03:00<00:05,  5.95s/it]
Training Epoch: 2/100, batch 60/63 completed (loss: 0.2232:  97%|█████████▋| 30/31 [03:00<00:05,  5.95s/it]
Training Epoch: 2/100, batch 60/63 completed (loss: 0.3938:  97%|█████████▋| 30/31 [03:00<00:05,  5.95s/it]
Training Epoch: 2/100, batch 60/63 completed (loss: 0.1325:  97%|█████████▋| 30/31 [03:00<00:05,  5.95s/it]
Training Epoch: 2/100, batch 60/63 completed (loss: 0.1325: 100%|██████████| 31/31 [03:04<00:00,  6.10s/it]
Training Epoch: 2/100, batch 60/63 completed (loss: 0.3938: 100%|██████████| 31/31 [03:04<00:00,  6.10s/it]
Training Epoch: 2/100, batch 60/63 completed (loss: 0.2232: 100%|██████████| 31/31 [03:04<00:00,  6.10s/it]
Training Epoch: 2/100, batch 60/63 completed (loss: 0.2118: 100%|██████████| 31/31 [03:04<00:00,  6.10s/it]
Training Epoch: 2/100, batch 61/63 completed (loss: 0.0816: 100%|██████████| 31/31 [03:04<00:00,  6.10s/it]
Training Epoch: 2/100, batch 61/63 completed (loss: 0.3536: 100%|██████████| 31/31 [03:04<00:00,  6.10s/it]
Training Epoch: 2/100, batch 61/63 completed (loss: 0.2287: 100%|██████████| 31/31 [03:04<00:00,  6.10s/it]
Training Epoch: 2/100, batch 61/63 completed (loss: 0.183: 100%|██████████| 31/31 [03:04<00:00,  6.10s/it] 
Training Epoch: 2/100, batch 61/63 completed (loss: 0.183: : 32it [03:06,  4.88s/it]                      
Training Epoch: 2/100, batch 61/63 completed (loss: 0.2287: : 32it [03:06,  4.89s/it]                      
Training Epoch: 2/100, batch 61/63 completed (loss: 0.0816: : 32it [03:06,  4.89s/it]                      
Training Epoch: 2/100, batch 61/63 completed (loss: 0.3536: : 32it [03:06,  4.89s/it]                      
Training Epoch: 2/100, batch 62/63 completed (loss: 0.389: : 32it [03:06,  4.89s/it] 
Training Epoch: 2/100, batch 62/63 completed (loss: 0.5474: : 32it [03:06,  4.89s/it]
Training Epoch: 2/100, batch 62/63 completed (loss: 0.146: : 32it [03:06,  4.89s/it] 
Training Epoch: 2/100, batch 62/63 completed (loss: 0.181: : 32it [03:06,  4.88s/it]
Training Epoch: 2/100, batch 62/63 completed (loss: 0.5474: : 32it [03:06,  5.83s/it]

Training Epoch: 2/100, batch 62/63 completed (loss: 0.389: : 32it [03:06,  5.83s/it]

Training Epoch: 2/100, batch 62/63 completed (loss: 0.146: : 32it [03:06,  5.83s/it]

Training Epoch: 2/100, batch 62/63 completed (loss: 0.181: : 32it [03:06,  5.84s/it]
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/c10d_logger.py:83: UserWarning: barrier(): using the device under current context. You can specify `device_id` in `init_process_group` to mute this warning.
  return func(*args, **kwargs)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
saving model + train state.
Warning: failed to upload checkpoint_2 to HF: (Request ID: Root=1-6a1b64cc-1dd601d508cb79416d032551;ca0462e9-b6c5-47e4-82fb-1b804bc427f8)

403 Forbidden: You need to setup automatic credit recharge in order to upload more data. You can do so at /settings/billing..
Cannot access content at: https://huggingface.co/seyedparsa/thoughtformer-checkpoints.git/info/lfs/objects/batch.
Make sure your token has the correct permissions.
Warning: failed to upload train_state_2 to HF: (Request ID: Root=1-6a1b64d0-65711dea3eea62d0635f3c51;76a55ab7-bf36-49c0-bf48-0c6367a29d39)

403 Forbidden: You need to setup automatic credit recharge in order to upload more data. You can do so at /settings/billing..
Cannot access content at: https://huggingface.co/seyedparsa/thoughtformer-checkpoints.git/info/lfs/objects/batch.
Make sure your token has the correct permissions.
Deleted old checkpoint: train_state_1
Deleted old checkpoint: checkpoint_1
[warn] HF upload failed for ckpt_2; kept locally, earlier ckpts deleted.

Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]
Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]
Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]eval loss 0.1995926180056163

Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]
[WRONG] Example 0 (diff=3)
  Expected : 'Claude'
  Extracted: 'Morgan'
  Generated: '### Morgan'

[WRONG] Example 1 (diff=8)
  Expected : 'Guy'
  Extracted: 'Irma'
  Generated: '### Irma'

Test Accuracy:   3%|▎         | 1/32 [00:06<03:25,  6.63s/it]
[CORRECT] Example 2 (diff=10)

Test accuracy: 0.5:   3%|▎         | 1/32 [00:06<03:25,  6.63s/it]  Expected : 'Elvis'
  Extracted: 'Elvis'
  Generated: '### Elvis'

Test Accuracy:   3%|▎         | 1/32 [00:06<03:25,  6.62s/it]
Test Accuracy:   3%|▎         | 1/32 [00:06<03:25,  6.62s/it]
Test accuracy: 0.5:   3%|▎         | 1/32 [00:06<03:25,  6.62s/it]
[CORRECT] Example 3 (diff=8)

Test accuracy: 0.75:   3%|▎         | 1/32 [00:06<03:25,  6.62s/it]  Expected : 'Coral'
  Extracted: 'Coral'
  Generated: '### Coral'

[CORRECT] Example 4 (diff=1)
  Expected : 'Patricia'
  Extracted: 'Patricia'
  Generated: '### Patricia'

Test Accuracy:   3%|▎         | 1/32 [00:06<03:20,  6.46s/it]
Test accuracy: 0.5:   3%|▎         | 1/32 [00:06<03:20,  6.46s/it]
Test accuracy: 0.75:   6%|▋         | 2/32 [00:13<03:17,  6.57s/it]
Test accuracy: 0.62:   6%|▋         | 2/32 [00:13<03:17,  6.57s/it]
Test accuracy: 0.5:   6%|▋         | 2/32 [00:13<03:17,  6.57s/it]
Test accuracy: 0.5:   6%|▋         | 2/32 [00:13<03:17,  6.57s/it]
Test accuracy: 0.5:   6%|▋         | 2/32 [00:13<03:17,  6.57s/it]
Test accuracy: 0.56:   6%|▋         | 2/32 [00:13<03:17,  6.57s/it]
Test accuracy: 0.5:   6%|▋         | 2/32 [00:12<03:14,  6.50s/it]
Test accuracy: 0.25:   6%|▋         | 2/32 [00:12<03:14,  6.50s/it]
Test accuracy: 0.56:   9%|▉         | 3/32 [00:19<03:13,  6.66s/it]
Test accuracy: 0.54:   9%|▉         | 3/32 [00:19<03:13,  6.66s/it]
Test accuracy: 0.62:   9%|▉         | 3/32 [00:19<03:13,  6.66s/it]
Test accuracy: 0.58:   9%|▉         | 3/32 [00:19<03:13,  6.66s/it]
Test accuracy: 0.25:   9%|▉         | 3/32 [00:19<03:12,  6.62s/it]
Test accuracy: 0.42:   9%|▉         | 3/32 [00:19<03:12,  6.62s/it]
Test accuracy: 0.5:   9%|▉         | 3/32 [00:19<03:13,  6.67s/it]
Test accuracy: 0.58:   9%|▉         | 3/32 [00:19<03:13,  6.67s/it]
Test accuracy: 0.58:  12%|█▎        | 4/32 [00:26<03:05,  6.63s/it]
Test accuracy: 0.62:  12%|█▎        | 4/32 [00:26<03:05,  6.63s/it]
Test accuracy: 0.58:  12%|█▎        | 4/32 [00:26<03:05,  6.63s/it]
Test accuracy: 0.59:  12%|█▎        | 4/32 [00:26<03:05,  6.63s/it]
Test accuracy: 0.54:  12%|█▎        | 4/32 [00:26<03:05,  6.64s/it]
Test accuracy: 0.42:  12%|█▎        | 4/32 [00:26<03:05,  6.61s/it]
Test accuracy: 0.62:  12%|█▎        | 4/32 [00:26<03:05,  6.64s/it]
Test accuracy: 0.47:  12%|█▎        | 4/32 [00:26<03:05,  6.61s/it]
Test accuracy: 0.62:  16%|█▌        | 5/32 [00:32<02:55,  6.48s/it]
Test accuracy: 0.62:  16%|█▌        | 5/32 [00:32<02:55,  6.48s/it]
Test accuracy: 0.59:  16%|█▌        | 5/32 [00:32<02:55,  6.48s/it]
Test accuracy: 0.65:  16%|█▌        | 5/32 [00:32<02:55,  6.48s/it]
Test accuracy: 0.62:  16%|█▌        | 5/32 [00:32<02:55,  6.48s/it]
Test accuracy: 0.57:  16%|█▌        | 5/32 [00:32<02:55,  6.48s/it]
Test accuracy: 0.47:  16%|█▌        | 5/32 [00:32<02:54,  6.47s/it]
Test accuracy: 0.5:  16%|█▌        | 5/32 [00:32<02:54,  6.47s/it] 
Test accuracy: 0.65:  19%|█▉        | 6/32 [00:38<02:46,  6.39s/it]
Test accuracy: 0.62:  19%|█▉        | 6/32 [00:38<02:46,  6.39s/it]
Test accuracy: 0.62:  19%|█▉        | 6/32 [00:38<02:46,  6.39s/it]
Test accuracy: 0.6:  19%|█▉        | 6/32 [00:38<02:46,  6.39s/it] 
Test accuracy: 0.57:  19%|█▉        | 6/32 [00:38<02:46,  6.39s/it]
Test accuracy: 0.56:  19%|█▉        | 6/32 [00:38<02:46,  6.39s/it]
Test accuracy: 0.5:  19%|█▉        | 6/32 [00:38<02:45,  6.38s/it]
Test accuracy: 0.52:  19%|█▉        | 6/32 [00:38<02:45,  6.38s/it]
Test accuracy: 0.56:  22%|██▏       | 7/32 [00:45<02:41,  6.45s/it]
Test accuracy: 0.59:  22%|██▏       | 7/32 [00:45<02:41,  6.45s/it]
Test accuracy: 0.62:  22%|██▏       | 7/32 [00:45<02:41,  6.45s/it]
Test accuracy: 0.64:  22%|██▏       | 7/32 [00:45<02:41,  6.45s/it]
Test accuracy: 0.52:  22%|██▏       | 7/32 [00:45<02:41,  6.44s/it]
Test accuracy: 0.6:  22%|██▏       | 7/32 [00:45<02:41,  6.45s/it]
Test accuracy: 0.57:  22%|██▏       | 7/32 [00:45<02:41,  6.44s/it]
Test accuracy: 0.55:  22%|██▏       | 7/32 [00:45<02:41,  6.45s/it]
Test accuracy: 0.64:  25%|██▌       | 8/32 [00:51<02:32,  6.37s/it]
Test accuracy: 0.62:  25%|██▌       | 8/32 [00:51<02:32,  6.37s/it]
Test accuracy: 0.57:  25%|██▌       | 8/32 [00:51<02:32,  6.37s/it]
Test accuracy: 0.58:  25%|██▌       | 8/32 [00:51<02:32,  6.37s/it]
Test accuracy: 0.59:  25%|██▌       | 8/32 [00:51<02:32,  6.37s/it]
Test accuracy: 0.56:  25%|██▌       | 8/32 [00:51<02:32,  6.37s/it]
Test accuracy: 0.55:  25%|██▌       | 8/32 [00:51<02:32,  6.37s/it]
Test accuracy: 0.55:  25%|██▌       | 8/32 [00:51<02:32,  6.37s/it]
Test accuracy: 0.62:  28%|██▊       | 9/32 [00:58<02:29,  6.50s/it]
Test accuracy: 0.6:  28%|██▊       | 9/32 [00:58<02:29,  6.50s/it] 
Test accuracy: 0.58:  28%|██▊       | 9/32 [00:58<02:29,  6.49s/it]
Test accuracy: 0.55:  28%|██▊       | 9/32 [00:58<02:29,  6.50s/it]
Test accuracy: 0.56:  28%|██▊       | 9/32 [00:58<02:29,  6.50s/it]
Test accuracy: 0.58:  28%|██▊       | 9/32 [00:58<02:29,  6.49s/it]
Test accuracy: 0.56:  28%|██▊       | 9/32 [00:58<02:29,  6.50s/it]
Test accuracy: 0.57:  28%|██▊       | 9/32 [00:58<02:29,  6.50s/it]
Test accuracy: 0.56:  31%|███▏      | 10/32 [01:05<02:23,  6.52s/it]
Test accuracy: 0.56:  31%|███▏      | 10/32 [01:05<02:23,  6.52s/it]
Test accuracy: 0.6:  31%|███▏      | 10/32 [01:05<02:23,  6.52s/it]
Test accuracy: 0.57:  31%|███▏      | 10/32 [01:05<02:23,  6.52s/it]
Test accuracy: 0.57:  31%|███▏      | 10/32 [01:05<02:23,  6.52s/it]
Test accuracy: 0.57:  31%|███▏      | 10/32 [01:05<02:23,  6.52s/it]
Test accuracy: 0.58:  31%|███▏      | 10/32 [01:04<02:23,  6.52s/it]
Test accuracy: 0.54:  31%|███▏      | 10/32 [01:04<02:23,  6.52s/it]
Test accuracy: 0.56:  34%|███▍      | 11/32 [01:11<02:14,  6.43s/it]
Test accuracy: 0.55:  34%|███▍      | 11/32 [01:11<02:14,  6.43s/it]
Test accuracy: 0.57:  34%|███▍      | 11/32 [01:11<02:14,  6.43s/it]
Test accuracy: 0.57:  34%|███▍      | 11/32 [01:11<02:14,  6.43s/it]
Test accuracy: 0.57:  34%|███▍      | 11/32 [01:11<02:14,  6.43s/it]
Test accuracy: 0.55:  34%|███▍      | 11/32 [01:11<02:14,  6.43s/it]
Test accuracy: 0.54:  34%|███▍      | 11/32 [01:11<02:14,  6.43s/it]
Test accuracy: 0.55:  34%|███▍      | 11/32 [01:11<02:14,  6.43s/it]
Test accuracy: 0.57:  38%|███▊      | 12/32 [01:17<02:08,  6.42s/it]
Test accuracy: 0.55:  38%|███▊      | 12/32 [01:17<02:08,  6.42s/it]
Test accuracy: 0.59:  38%|███▊      | 12/32 [01:17<02:08,  6.42s/it]
Test accuracy: 0.55:  38%|███▊      | 12/32 [01:17<02:08,  6.42s/it]
Test accuracy: 0.52:  38%|███▊      | 12/32 [01:17<02:08,  6.42s/it]
Test accuracy: 0.56:  38%|███▊      | 12/32 [01:17<02:08,  6.42s/it]
Test accuracy: 0.55:  38%|███▊      | 12/32 [01:17<02:08,  6.43s/it]
Test accuracy: 0.54:  38%|███▊      | 12/32 [01:17<02:08,  6.43s/it]
Test accuracy: 0.59:  41%|████      | 13/32 [01:24<02:01,  6.41s/it]
Test accuracy: 0.61:  41%|████      | 13/32 [01:24<02:01,  6.41s/it]
Test accuracy: 0.54:  41%|████      | 13/32 [01:24<02:01,  6.41s/it]
Test accuracy: 0.52:  41%|████      | 13/32 [01:24<02:01,  6.41s/it]
Test accuracy: 0.53:  41%|████      | 13/32 [01:24<02:01,  6.41s/it]
Test accuracy: 0.51:  41%|████      | 13/32 [01:24<02:01,  6.41s/it]
Test accuracy: 0.56:  41%|████      | 13/32 [01:23<02:01,  6.41s/it]
Test accuracy: 0.55:  41%|████      | 13/32 [01:23<02:01,  6.41s/it]
Test accuracy: 0.61:  44%|████▍     | 14/32 [01:30<01:54,  6.35s/it]
Test accuracy: 0.55:  44%|████▍     | 14/32 [01:30<01:54,  6.35s/it]
Test accuracy: 0.59:  44%|████▍     | 14/32 [01:30<01:54,  6.35s/it]
Test accuracy: 0.51:  44%|████▍     | 14/32 [01:30<01:54,  6.35s/it]
Test accuracy: 0.55:  44%|████▍     | 14/32 [01:30<01:54,  6.35s/it]
Test accuracy: 0.51:  44%|████▍     | 14/32 [01:30<01:54,  6.35s/it]
Test accuracy: 0.53:  44%|████▍     | 14/32 [01:30<01:54,  6.35s/it]
Test accuracy: 0.55:  44%|████▍     | 14/32 [01:30<01:54,  6.35s/it]
Test accuracy: 0.59:  47%|████▋     | 15/32 [01:36<01:48,  6.37s/it]
Test accuracy: 0.55:  47%|████▋     | 15/32 [01:36<01:48,  6.37s/it]
Test accuracy: 0.57:  47%|████▋     | 15/32 [01:36<01:48,  6.37s/it]
Test accuracy: 0.57:  47%|████▋     | 15/32 [01:36<01:48,  6.37s/it]
Test accuracy: 0.51:  47%|████▋     | 15/32 [01:36<01:48,  6.37s/it]
Test accuracy: 0.49:  47%|████▋     | 15/32 [01:36<01:48,  6.37s/it]
Test accuracy: 0.55:  47%|████▋     | 15/32 [01:36<01:48,  6.37s/it]
Test accuracy: 0.53:  47%|████▋     | 15/32 [01:36<01:48,  6.37s/it]
Test accuracy: 0.53:  50%|█████     | 16/32 [01:42<01:42,  6.38s/it]
Test accuracy: 0.49:  50%|█████     | 16/32 [01:43<01:42,  6.38s/it]
Test accuracy: 0.48:  50%|█████     | 16/32 [01:43<01:42,  6.38s/it]
Test accuracy: 0.52:  50%|█████     | 16/32 [01:42<01:42,  6.38s/it]
Test accuracy: 0.57:  50%|█████     | 16/32 [01:43<01:42,  6.38s/it]
Test accuracy: 0.57:  50%|█████     | 16/32 [01:43<01:42,  6.38s/it]
Test accuracy: 0.55:  50%|█████     | 16/32 [01:43<01:42,  6.38s/it]
Test accuracy: 0.58:  50%|█████     | 16/32 [01:43<01:42,  6.38s/it]
Test accuracy: 0.58:  53%|█████▎    | 17/32 [01:49<01:34,  6.32s/it]
Test accuracy: 0.48:  53%|█████▎    | 17/32 [01:49<01:34,  6.32s/it]
Test accuracy: 0.58:  53%|█████▎    | 17/32 [01:49<01:34,  6.32s/it]
Test accuracy: 0.55:  53%|█████▎    | 17/32 [01:49<01:34,  6.32s/it]
Test accuracy: 0.5:  53%|█████▎    | 17/32 [01:49<01:34,  6.32s/it] 
Test accuracy: 0.55:  53%|█████▎    | 17/32 [01:49<01:34,  6.32s/it]
Test accuracy: 0.52:  53%|█████▎    | 17/32 [01:49<01:34,  6.32s/it]
Test accuracy: 0.54:  53%|█████▎    | 17/32 [01:49<01:34,  6.32s/it]
Test accuracy: 0.58:  56%|█████▋    | 18/32 [01:55<01:28,  6.35s/it]
Test accuracy: 0.55:  56%|█████▋    | 18/32 [01:55<01:28,  6.35s/it]
Test accuracy: 0.59:  56%|█████▋    | 18/32 [01:55<01:28,  6.35s/it]
Test accuracy: 0.56:  56%|█████▋    | 18/32 [01:55<01:28,  6.35s/it]
Test accuracy: 0.5:  56%|█████▋    | 18/32 [01:55<01:28,  6.35s/it]
Test accuracy: 0.49:  56%|█████▋    | 18/32 [01:55<01:28,  6.35s/it]
Test accuracy: 0.54:  56%|█████▋    | 18/32 [01:55<01:28,  6.35s/it]
Test accuracy: 0.53:  56%|█████▋    | 18/32 [01:55<01:28,  6.35s/it]
Test accuracy: 0.59:  59%|█████▉    | 19/32 [02:02<01:22,  6.36s/it]
Test accuracy: 0.59:  59%|█████▉    | 19/32 [02:02<01:22,  6.36s/it]
Test accuracy: 0.56:  59%|█████▉    | 19/32 [02:02<01:22,  6.36s/it]
Test accuracy: 0.53:  59%|█████▉    | 19/32 [02:01<01:22,  6.36s/it]
Test accuracy: 0.55:  59%|█████▉    | 19/32 [02:02<01:22,  6.36s/it]
Test accuracy: 0.55:  59%|█████▉    | 19/32 [02:01<01:22,  6.36s/it]
Test accuracy: 0.49:  59%|█████▉    | 19/32 [02:02<01:22,  6.36s/it]
Test accuracy: 0.48:  59%|█████▉    | 19/32 [02:02<01:22,  6.36s/it]
Test accuracy: 0.55:  62%|██████▎   | 20/32 [02:08<01:15,  6.31s/it]
Test accuracy: 0.59:  62%|██████▎   | 20/32 [02:08<01:15,  6.31s/it]
Test accuracy: 0.56:  62%|██████▎   | 20/32 [02:08<01:15,  6.31s/it]
Test accuracy: 0.59:  62%|██████▎   | 20/32 [02:08<01:15,  6.31s/it]
Test accuracy: 0.55:  62%|██████▎   | 20/32 [02:08<01:15,  6.31s/it]
Test accuracy: 0.54:  62%|██████▎   | 20/32 [02:08<01:15,  6.31s/it]
Test accuracy: 0.48:  62%|██████▎   | 20/32 [02:08<01:15,  6.31s/it]
Test accuracy: 0.49:  62%|██████▎   | 20/32 [02:08<01:15,  6.31s/it]
Test accuracy: 0.49:  66%|██████▌   | 21/32 [02:15<01:11,  6.52s/it]
Test accuracy: 0.59:  66%|██████▌   | 21/32 [02:15<01:11,  6.52s/it]
Test accuracy: 0.56:  66%|██████▌   | 21/32 [02:15<01:11,  6.52s/it]
Test accuracy: 0.49:  66%|██████▌   | 21/32 [02:15<01:11,  6.52s/it]
Test accuracy: 0.58:  66%|██████▌   | 21/32 [02:15<01:11,  6.52s/it]
Test accuracy: 0.55:  66%|██████▌   | 21/32 [02:15<01:11,  6.52s/it]
Test accuracy: 0.54:  66%|██████▌   | 21/32 [02:15<01:11,  6.53s/it]
Test accuracy: 0.55:  66%|██████▌   | 21/32 [02:15<01:11,  6.53s/it]
Test accuracy: 0.58:  69%|██████▉   | 22/32 [02:21<01:04,  6.49s/it]
Test accuracy: 0.55:  69%|██████▉   | 22/32 [02:21<01:04,  6.49s/it]
Test accuracy: 0.57:  69%|██████▉   | 22/32 [02:21<01:04,  6.49s/it]
Test accuracy: 0.55:  69%|██████▉   | 22/32 [02:21<01:04,  6.49s/it]
Test accuracy: 0.49:  69%|██████▉   | 22/32 [02:21<01:04,  6.49s/it]
Test accuracy: 0.55:  69%|██████▉   | 22/32 [02:21<01:04,  6.49s/it]
Test accuracy: 0.49:  69%|██████▉   | 22/32 [02:21<01:04,  6.49s/it]
Test accuracy: 0.57:  69%|██████▉   | 22/32 [02:21<01:04,  6.49s/it]
Test accuracy: 0.49:  72%|███████▏  | 23/32 [02:28<00:58,  6.46s/it]
Test accuracy: 0.49:  72%|███████▏  | 23/32 [02:28<00:58,  6.46s/it]
Test accuracy: 0.57:  72%|███████▏  | 23/32 [02:27<00:58,  6.46s/it]
Test accuracy: 0.58:  72%|███████▏  | 23/32 [02:27<00:58,  6.46s/it]
Test accuracy: 0.55:  72%|███████▏  | 23/32 [02:28<00:58,  6.46s/it]
Test accuracy: 0.55:  72%|███████▏  | 23/32 [02:28<00:58,  6.46s/it]
Test accuracy: 0.57:  72%|███████▏  | 23/32 [02:28<00:58,  6.46s/it]
Test accuracy: 0.58:  72%|███████▏  | 23/32 [02:28<00:58,  6.46s/it]
Test accuracy: 0.55:  75%|███████▌  | 24/32 [02:34<00:52,  6.57s/it]
Test accuracy: 0.55:  75%|███████▌  | 24/32 [02:34<00:52,  6.57s/it]
Test accuracy: 0.49:  75%|███████▌  | 24/32 [02:34<00:52,  6.57s/it]
Test accuracy: 0.58:  75%|███████▌  | 24/32 [02:34<00:52,  6.57s/it]
Test accuracy: 0.48:  75%|███████▌  | 24/32 [02:34<00:52,  6.57s/it]
Test accuracy: 0.58:  75%|███████▌  | 24/32 [02:34<00:52,  6.57s/it]
Test accuracy: 0.58:  75%|███████▌  | 24/32 [02:34<00:52,  6.57s/it]
Test accuracy: 0.57:  75%|███████▌  | 24/32 [02:34<00:52,  6.57s/it]
Test accuracy: 0.58:  78%|███████▊  | 25/32 [02:41<00:45,  6.52s/it]
Test accuracy: 0.57:  78%|███████▊  | 25/32 [02:41<00:45,  6.52s/it]
Test accuracy: 0.55:  78%|███████▊  | 25/32 [02:41<00:45,  6.52s/it]
Test accuracy: 0.57:  78%|███████▊  | 25/32 [02:41<00:45,  6.52s/it]
Test accuracy: 0.55:  78%|███████▊  | 25/32 [02:41<00:45,  6.52s/it]
Test accuracy: 0.56:  78%|███████▊  | 25/32 [02:41<00:45,  6.52s/it]
Test accuracy: 0.48:  78%|███████▊  | 25/32 [02:41<00:45,  6.53s/it]
Test accuracy: 0.49:  78%|███████▊  | 25/32 [02:41<00:45,  6.53s/it]
Test accuracy: 0.57:  81%|████████▏ | 26/32 [02:47<00:38,  6.49s/it]
Test accuracy: 0.57:  81%|████████▏ | 26/32 [02:47<00:38,  6.49s/it]
Test accuracy: 0.49:  81%|████████▏ | 26/32 [02:47<00:38,  6.49s/it]
Test accuracy: 0.48:  81%|████████▏ | 26/32 [02:47<00:38,  6.49s/it]
Test accuracy: 0.55:  81%|████████▏ | 26/32 [02:47<00:38,  6.49s/it]
Test accuracy: 0.55:  81%|████████▏ | 26/32 [02:47<00:38,  6.49s/it]
Test accuracy: 0.56:  81%|████████▏ | 26/32 [02:47<00:38,  6.49s/it]
Test accuracy: 0.57:  81%|████████▏ | 26/32 [02:47<00:38,  6.49s/it]
Test accuracy: 0.55:  84%|████████▍ | 27/32 [02:53<00:32,  6.40s/it]
Test accuracy: 0.57:  84%|████████▍ | 27/32 [02:53<00:32,  6.40s/it]
Test accuracy: 0.54:  84%|████████▍ | 27/32 [02:53<00:32,  6.40s/it]
Test accuracy: 0.56:  84%|████████▍ | 27/32 [02:53<00:32,  6.40s/it]
Test accuracy: 0.48:  84%|████████▍ | 27/32 [02:53<00:32,  6.40s/it]
Test accuracy: 0.49:  84%|████████▍ | 27/32 [02:53<00:32,  6.40s/it]
Test accuracy: 0.57:  84%|████████▍ | 27/32 [02:53<00:32,  6.40s/it]
Test accuracy: 0.56:  84%|████████▍ | 27/32 [02:53<00:32,  6.40s/it]
Test accuracy: 0.56:  88%|████████▊ | 28/32 [03:00<00:26,  6.57s/it]
Test accuracy: 0.49:  88%|████████▊ | 28/32 [03:00<00:26,  6.57s/it]
Test accuracy: 0.56:  88%|████████▊ | 28/32 [03:00<00:26,  6.57s/it]
Test accuracy: 0.5:  88%|████████▊ | 28/32 [03:00<00:26,  6.57s/it] 
Test accuracy: 0.57:  88%|████████▊ | 28/32 [03:00<00:26,  6.57s/it]
Test accuracy: 0.55:  88%|████████▊ | 28/32 [03:00<00:26,  6.57s/it]
Test accuracy: 0.54:  88%|████████▊ | 28/32 [03:00<00:26,  6.58s/it]
Test accuracy: 0.54:  88%|████████▊ | 28/32 [03:00<00:26,  6.58s/it]
Test accuracy: 0.54:  91%|█████████ | 29/32 [03:07<00:19,  6.52s/it]
Test accuracy: 0.57:  91%|█████████ | 29/32 [03:07<00:19,  6.52s/it]
Test accuracy: 0.55:  91%|█████████ | 29/32 [03:07<00:19,  6.52s/it]
Test accuracy: 0.58:  91%|█████████ | 29/32 [03:07<00:19,  6.52s/it]
Test accuracy: 0.5:  91%|█████████ | 29/32 [03:07<00:19,  6.52s/it]
Test accuracy: 0.5:  91%|█████████ | 29/32 [03:07<00:19,  6.52s/it]
Test accuracy: 0.55:  91%|█████████ | 29/32 [03:07<00:19,  6.52s/it]
Test accuracy: 0.56:  91%|█████████ | 29/32 [03:07<00:19,  6.52s/it]
Test accuracy: 0.56:  94%|█████████▍| 30/32 [03:13<00:12,  6.49s/it]
Test accuracy: 0.55:  94%|█████████▍| 30/32 [03:13<00:12,  6.49s/it]
Test accuracy: 0.56:  94%|█████████▍| 30/32 [03:13<00:12,  6.49s/it]
Test accuracy: 0.54:  94%|█████████▍| 30/32 [03:13<00:12,  6.49s/it]
Test accuracy: 0.5:  94%|█████████▍| 30/32 [03:13<00:12,  6.49s/it]
Test accuracy: 0.58:  94%|█████████▍| 30/32 [03:13<00:12,  6.49s/it]
Test accuracy: 0.5:  94%|█████████▍| 30/32 [03:13<00:12,  6.49s/it]
Test accuracy: 0.57:  94%|█████████▍| 30/32 [03:13<00:12,  6.49s/it]
Test accuracy: 0.5:  97%|█████████▋| 31/32 [03:20<00:06,  6.58s/it]
Test accuracy: 0.56:  97%|█████████▋| 31/32 [03:20<00:06,  6.58s/it]
Test accuracy: 0.57:  97%|█████████▋| 31/32 [03:20<00:06,  6.58s/it]
Test accuracy: 0.5:  97%|█████████▋| 31/32 [03:20<00:06,  6.58s/it]
Test accuracy: 0.57:  97%|█████████▋| 31/32 [03:20<00:06,  6.58s/it]
Test accuracy: 0.56:  97%|█████████▋| 31/32 [03:20<00:06,  6.58s/it]
Test accuracy: 0.54:  97%|█████████▋| 31/32 [03:20<00:06,  6.59s/it]
Test accuracy: 0.54:  97%|█████████▋| 31/32 [03:20<00:06,  6.59s/it]
Test accuracy: 0.56: 100%|██████████| 32/32 [03:22<00:00,  5.38s/it]
Test accuracy: 0.56: 100%|██████████| 32/32 [03:22<00:00,  5.38s/it]
Test accuracy: 0.57: 100%|██████████| 32/32 [03:23<00:00,  5.38s/it]
Test accuracy: 0.57: 100%|██████████| 32/32 [03:23<00:00,  5.38s/it]
Test accuracy: 0.5: 100%|██████████| 32/32 [03:23<00:00,  5.38s/it]
Test accuracy: 0.5: 100%|██████████| 32/32 [03:23<00:00,  5.38s/it]
Test accuracy: 0.54: 100%|██████████| 32/32 [03:23<00:00,  5.38s/it]
Test accuracy: 0.55: 100%|██████████| 32/32 [03:23<00:00,  5.38s/it]
Test accuracy: 0.55: 100%|██████████| 32/32 [03:23<00:00,  6.35s/it]
Device 2: Cor=137, CoT=0, Total=250

Test accuracy: 0.57: 100%|██████████| 32/32 [03:23<00:00,  6.35s/it]
Device 1: Cor=143, CoT=0, Total=250

Test accuracy: 0.5: 100%|██████████| 32/32 [03:23<00:00,  6.35s/it]
Device 3: Cor=125, CoT=0, Total=250

Test accuracy: 0.56: 100%|██████████| 32/32 [03:23<00:00,  6.35s/it]
Device 0: Cor=140, CoT=0, Total=250

============================================================
EVALUATION RESULTS (Epoch 1)
============================================================
Config: data_curriculum=ON, thought_curriculum=OFF
thought_stage=10, data_stage=1, n_latent_tokens=10
Training on difficulties: 0-1

Overall accuracy: 545 / 1000 = 0.5450

Per-difficulty accuracy:
  diff=1: 72/100 = 0.7200 (trained)
  diff=2: 48/100 = 0.4800 (untrained)
  diff=3: 51/100 = 0.5100 (untrained)
  diff=4: 60/100 = 0.6000 (untrained)
  diff=5: 41/100 = 0.4100 (untrained)
  diff=6: 50/100 = 0.5000 (untrained)
  diff=7: 52/100 = 0.5200 (untrained)
  diff=8: 55/100 = 0.5500 (untrained)
  diff=9: 60/100 = 0.6000 (untrained)
  diff=10: 56/100 = 0.5600 (untrained)

Trained (diff 0-1): 72/100 = 0.7200
============================================================

  Staging check: acc_trained=0.7200, acc_diff1=0.7200 (threshold=0.9)
Warning: failed to upload checkpoint_2 to HF: (Request ID: Root=1-6a1b65a4-4657ae642789127437194cd4;905b11c1-8b34-48bd-85f4-8a89d2b63c2e)

403 Forbidden: You need to setup automatic credit recharge in order to upload more data. You can do so at /settings/billing..
Cannot access content at: https://huggingface.co/seyedparsa/thoughtformer-checkpoints.git/info/lfs/objects/batch.
Make sure your token has the correct permissions.

============================================================
EPOCH 2/100 (thought_stage=10, data_stage=1)
============================================================
  thought_stage=10, c_thought=1, max_difficulty=1

Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):  16%|█▌        | 157/1000 [00:00<00:00, 1075.36 examples/s]
Map (num_proc=4):  16%|█▌        | 157/1000 [00:00<00:00, 1024.74 examples/s]
Map (num_proc=4):  16%|█▌        | 157/1000 [00:00<00:00, 988.18 examples/s]
Map (num_proc=4):  81%|████████  | 812/1000 [00:00<00:00, 3543.02 examples/s]
Map (num_proc=4):  90%|█████████ | 905/1000 [00:00<00:00, 3673.55 examples/s]
Map (num_proc=4):  82%|████████▏ | 818/1000 [00:00<00:00, 3553.52 examples/s]
Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2667.13 examples/s]

Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2602.35 examples/s]

Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2121.37 examples/s]

Map (num_proc=4):  10%|█         | 102/1000 [00:00<00:02, 405.17 examples/s]
Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Map (num_proc=4):  89%|████████▊ | 886/1000 [00:00<00:00, 2632.87 examples/s]
Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 1942.20 examples/s]

Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:04, 2172.12 examples/s]
Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:04, 2125.10 examples/s]
Filter (num_proc=4):  20%|██        | 2000/10000 [00:00<00:02, 3936.30 examples/s]
Filter (num_proc=4):  30%|███       | 3000/10000 [00:00<00:01, 5391.57 examples/s]
Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:04, 1965.79 examples/s]
Filter (num_proc=4):  40%|████      | 4000/10000 [00:00<00:00, 7009.88 examples/s]
Filter (num_proc=4):  20%|██        | 2000/10000 [00:00<00:02, 3704.69 examples/s]
Filter (num_proc=4):  60%|██████    | 6000/10000 [00:00<00:00, 9740.61 examples/s]
Filter (num_proc=4):  30%|███       | 3000/10000 [00:00<00:01, 5191.36 examples/s]
Filter (num_proc=4):  65%|██████▌   | 6500/10000 [00:00<00:00, 10791.41 examples/s]
Filter (num_proc=4):  80%|████████  | 8000/10000 [00:00<00:00, 11494.55 examples/s]
Filter (num_proc=4):  90%|█████████ | 9000/10000 [00:00<00:00, 13171.70 examples/s]
Filter (num_proc=4):  60%|██████    | 6000/10000 [00:00<00:00, 10209.46 examples/s]
Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 11116.04 examples/s]
Filter (num_proc=4):  80%|████████  | 8000/10000 [00:00<00:00, 12021.95 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 8101.95 examples/s] 

Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 8070.02 examples/s]

Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 11473.83 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 7639.52 examples/s] 

Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:06, 1449.16 examples/s]
Filter (num_proc=4):  20%|██        | 2000/10000 [00:00<00:03, 2514.17 examples/s]
Filter (num_proc=4):  35%|███▌      | 3500/10000 [00:01<00:01, 4261.05 examples/s]
Filter (num_proc=4):  60%|██████    | 6000/10000 [00:01<00:00, 6938.31 examples/s]
Filter (num_proc=4):  85%|████████▌ | 8500/10000 [00:01<00:00, 9027.64 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 8866.19 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 5473.79 examples/s]

Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):  13%|█▎        | 131/1000 [00:01<00:10, 81.22 examples/s]
Map (num_proc=4):  25%|██▌       | 250/1000 [00:01<00:04, 169.13 examples/s]
Map (num_proc=4):  39%|███▊      | 387/1000 [00:02<00:03, 178.97 examples/s]
Map (num_proc=4):  50%|█████     | 500/1000 [00:02<00:01, 255.24 examples/s]
Map (num_proc=4):  64%|██████▎   | 637/1000 [00:03<00:01, 222.93 examples/s]
Map (num_proc=4):  75%|███████▌  | 750/1000 [00:03<00:00, 296.76 examples/s]
Map (num_proc=4):  88%|████████▊ | 884/1000 [00:04<00:00, 244.23 examples/s]
Map (num_proc=4): 100%|██████████| 1000/1000 [00:04<00:00, 316.18 examples/s]
Map (num_proc=4): 100%|██████████| 1000/1000 [00:04<00:00, 219.73 examples/s]

Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 1459.69 examples/s]
Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 1385.22 examples/s]
Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 1446.31 examples/s]
Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2842.19 examples/s]

Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2802.23 examples/s]

Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2686.95 examples/s]

Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 829.55 examples/s]
Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 1611.14 examples/s]

Map (num_proc=4):   0%|          | 0/100 [00:00<?, ? examples/s]
Map (num_proc=4):  25%|██▌       | 25/100 [00:00<00:00, 87.11 examples/s]
Map (num_proc=4):  75%|███████▌  | 75/100 [00:00<00:00, 193.08 examples/s]
Map (num_proc=4): 100%|██████████| 100/100 [00:00<00:00, 130.46 examples/s]

Training Epoch: 3:   0%|          | 0/31 [00:00<?, ?it/s]
Training Epoch: 3:   0%|          | 0/31 [00:00<?, ?it/s]
Training Epoch: 3:   0%|          | 0/31 [00:00<?, ?it/s]
Training Epoch: 3:   0%|          | 0/31 [00:00<?, ?it/s]logging training data
rank 2 | gpu 2 | alloc=4.20 GB | reserved=56.98 GB | max=41.98 GB
rank 3 | gpu 3 | alloc=4.20 GB | reserved=56.98 GB | max=41.99 GB
rank 1 | gpu 1 | alloc=4.20 GB | reserved=47.70 GB | max=41.98 GB
rank 0 | gpu 0 | alloc=4.20 GB | reserved=51.46 GB | max=41.99 GB

Training Epoch: 3/100, batch 0/63 completed (loss: 0.1507:   0%|          | 0/31 [00:03<?, ?it/s]
Training Epoch: 3/100, batch 0/63 completed (loss: 0.2112:   0%|          | 0/31 [00:03<?, ?it/s]
Training Epoch: 3/100, batch 0/63 completed (loss: 0.1285:   0%|          | 0/31 [00:03<?, ?it/s]
Training Epoch: 3/100, batch 0/63 completed (loss: 0.2767:   0%|          | 0/31 [00:03<?, ?it/s]
Training Epoch: 3/100, batch 0/63 completed (loss: 0.1507:   3%|▎         | 1/31 [00:05<02:58,  5.94s/it]
Training Epoch: 3/100, batch 0/63 completed (loss: 0.2112:   3%|▎         | 1/31 [00:05<02:58,  5.95s/it]
Training Epoch: 3/100, batch 0/63 completed (loss: 0.1285:   3%|▎         | 1/31 [00:05<02:58,  5.95s/it]
Training Epoch: 3/100, batch 0/63 completed (loss: 0.2767:   3%|▎         | 1/31 [00:05<02:59,  5.97s/it]
Training Epoch: 3/100, batch 1/63 completed (loss: 0.316:   3%|▎         | 1/31 [00:05<02:58,  5.95s/it] 
Training Epoch: 3/100, batch 1/63 completed (loss: 0.1457:   3%|▎         | 1/31 [00:05<02:58,  5.94s/it]
Training Epoch: 3/100, batch 1/63 completed (loss: 0.2317:   3%|▎         | 1/31 [00:05<02:58,  5.95s/it]
Training Epoch: 3/100, batch 1/63 completed (loss: 0.1328:   3%|▎         | 1/31 [00:06<02:59,  5.97s/it]
Training Epoch: 3/100, batch 2/63 completed (loss: 0.139:   3%|▎         | 1/31 [00:08<02:58,  5.95s/it]
Training Epoch: 3/100, batch 2/63 completed (loss: 0.5849:   3%|▎         | 1/31 [00:08<02:58,  5.94s/it]
Training Epoch: 3/100, batch 2/63 completed (loss: 0.0774:   3%|▎         | 1/31 [00:08<02:58,  5.95s/it]
Training Epoch: 3/100, batch 2/63 completed (loss: 0.1045:   3%|▎         | 1/31 [00:08<02:59,  5.97s/it]
Training Epoch: 3/100, batch 2/63 completed (loss: 0.5849:   6%|▋         | 2/31 [00:11<02:50,  5.89s/it]
Training Epoch: 3/100, batch 2/63 completed (loss: 0.0774:   6%|▋         | 2/31 [00:11<02:50,  5.89s/it]
Training Epoch: 3/100, batch 2/63 completed (loss: 0.139:   6%|▋         | 2/31 [00:11<02:51,  5.90s/it]
Training Epoch: 3/100, batch 2/63 completed (loss: 0.1045:   6%|▋         | 2/31 [00:11<02:51,  5.90s/it]
Training Epoch: 3/100, batch 3/63 completed (loss: 0.1078:   6%|▋         | 2/31 [00:11<02:51,  5.90s/it]
Training Epoch: 3/100, batch 3/63 completed (loss: 0.1392:   6%|▋         | 2/31 [00:11<02:50,  5.89s/it]
Training Epoch: 3/100, batch 3/63 completed (loss: 0.2488:   6%|▋         | 2/31 [00:11<02:50,  5.89s/it]
Training Epoch: 3/100, batch 3/63 completed (loss: 0.1474:   6%|▋         | 2/31 [00:11<02:51,  5.90s/it]
Training Epoch: 3/100, batch 4/63 completed (loss: 0.234:   6%|▋         | 2/31 [00:15<02:50,  5.89s/it] 
Training Epoch: 3/100, batch 4/63 completed (loss: 0.2417:   6%|▋         | 2/31 [00:15<02:50,  5.89s/it]
Training Epoch: 3/100, batch 4/63 completed (loss: 0.0945:   6%|▋         | 2/31 [00:15<02:51,  5.90s/it]
Training Epoch: 3/100, batch 4/63 completed (loss: 0.1722:   6%|▋         | 2/31 [00:15<02:51,  5.90s/it]
Training Epoch: 3/100, batch 4/63 completed (loss: 0.0945:  10%|▉         | 3/31 [00:17<02:47,  5.97s/it]
Training Epoch: 3/100, batch 4/63 completed (loss: 0.1722:  10%|▉         | 3/31 [00:17<02:47,  5.98s/it]
Training Epoch: 3/100, batch 4/63 completed (loss: 0.2417:  10%|▉         | 3/31 [00:17<02:47,  5.98s/it]
Training Epoch: 3/100, batch 4/63 completed (loss: 0.234:  10%|▉         | 3/31 [00:17<02:47,  5.98s/it]
Training Epoch: 3/100, batch 5/63 completed (loss: 0.1681:  10%|▉         | 3/31 [00:17<02:47,  5.98s/it]
Training Epoch: 3/100, batch 5/63 completed (loss: 0.3011:  10%|▉         | 3/31 [00:17<02:47,  5.98s/it]
Training Epoch: 3/100, batch 5/63 completed (loss: 0.1668:  10%|▉         | 3/31 [00:17<02:47,  5.97s/it]
Training Epoch: 3/100, batch 5/63 completed (loss: 0.2654:  10%|▉         | 3/31 [00:17<02:47,  5.98s/it]
Training Epoch: 3/100, batch 6/63 completed (loss: 0.331:  10%|▉         | 3/31 [00:20<02:47,  5.97s/it] 
Training Epoch: 3/100, batch 6/63 completed (loss: 0.1679:  10%|▉         | 3/31 [00:20<02:47,  5.98s/it]
Training Epoch: 3/100, batch 6/63 completed (loss: 0.128:  10%|▉         | 3/31 [00:20<02:47,  5.98s/it] 
Training Epoch: 3/100, batch 6/63 completed (loss: 0.1273:  10%|▉         | 3/31 [00:20<02:47,  5.98s/it]
Training Epoch: 3/100, batch 6/63 completed (loss: 0.331:  13%|█▎        | 4/31 [00:23<02:37,  5.82s/it]
Training Epoch: 3/100, batch 6/63 completed (loss: 0.1679:  13%|█▎        | 4/31 [00:23<02:37,  5.83s/it]
Training Epoch: 3/100, batch 6/63 completed (loss: 0.1273:  13%|█▎        | 4/31 [00:23<02:37,  5.83s/it]
Training Epoch: 3/100, batch 6/63 completed (loss: 0.128:  13%|█▎        | 4/31 [00:23<02:37,  5.83s/it]
Training Epoch: 3/100, batch 7/63 completed (loss: 0.1419:  13%|█▎        | 4/31 [00:23<02:37,  5.83s/it]
Training Epoch: 3/100, batch 7/63 completed (loss: 0.1385:  13%|█▎        | 4/31 [00:23<02:37,  5.83s/it]
Training Epoch: 3/100, batch 7/63 completed (loss: 0.0908:  13%|█▎        | 4/31 [00:23<02:37,  5.82s/it]
Training Epoch: 3/100, batch 7/63 completed (loss: 0.1912:  13%|█▎        | 4/31 [00:23<02:37,  5.83s/it]
Training Epoch: 3/100, batch 8/63 completed (loss: 0.1344:  13%|█▎        | 4/31 [00:26<02:37,  5.83s/it]
Training Epoch: 3/100, batch 8/63 completed (loss: 0.1266:  13%|█▎        | 4/31 [00:26<02:37,  5.83s/it]
Training Epoch: 3/100, batch 8/63 completed (loss: 0.1074:  13%|█▎        | 4/31 [00:26<02:37,  5.82s/it]
Training Epoch: 3/100, batch 8/63 completed (loss: 0.2162:  13%|█▎        | 4/31 [00:26<02:37,  5.83s/it]
Training Epoch: 3/100, batch 8/63 completed (loss: 0.2162:  16%|█▌        | 5/31 [00:29<02:33,  5.88s/it]
Training Epoch: 3/100, batch 8/63 completed (loss: 0.1074:  16%|█▌        | 5/31 [00:29<02:33,  5.89s/it]
Training Epoch: 3/100, batch 8/63 completed (loss: 0.1344:  16%|█▌        | 5/31 [00:29<02:33,  5.89s/it]
Training Epoch: 3/100, batch 8/63 completed (loss: 0.1266:  16%|█▌        | 5/31 [00:29<02:33,  5.89s/it]
Training Epoch: 3/100, batch 9/63 completed (loss: 0.2972:  16%|█▌        | 5/31 [00:29<02:33,  5.89s/it]
Training Epoch: 3/100, batch 9/63 completed (loss: 0.3208:  16%|█▌        | 5/31 [00:29<02:33,  5.89s/it]
Training Epoch: 3/100, batch 9/63 completed (loss: 0.3044:  16%|█▌        | 5/31 [00:29<02:33,  5.89s/it]
Training Epoch: 3/100, batch 9/63 completed (loss: 0.0754:  16%|█▌        | 5/31 [00:29<02:33,  5.88s/it]
Training Epoch: 3/100, batch 10/63 completed (loss: 0.2749:  16%|█▌        | 5/31 [00:32<02:33,  5.89s/it]
Training Epoch: 3/100, batch 10/63 completed (loss: 0.1432:  16%|█▌        | 5/31 [00:32<02:33,  5.89s/it]
Training Epoch: 3/100, batch 10/63 completed (loss: 0.2032:  16%|█▌        | 5/31 [00:32<02:33,  5.89s/it]
Training Epoch: 3/100, batch 10/63 completed (loss: 0.4174:  16%|█▌        | 5/31 [00:32<02:33,  5.88s/it]
Training Epoch: 3/100, batch 10/63 completed (loss: 0.1432:  19%|█▉        | 6/31 [00:35<02:26,  5.85s/it]
Training Epoch: 3/100, batch 10/63 completed (loss: 0.2032:  19%|█▉        | 6/31 [00:35<02:26,  5.85s/it]
Training Epoch: 3/100, batch 10/63 completed (loss: 0.4174:  19%|█▉        | 6/31 [00:35<02:26,  5.85s/it]
Training Epoch: 3/100, batch 10/63 completed (loss: 0.2749:  19%|█▉        | 6/31 [00:35<02:26,  5.85s/it]
Training Epoch: 3/100, batch 11/63 completed (loss: 0.2481:  19%|█▉        | 6/31 [00:35<02:26,  5.85s/it]
Training Epoch: 3/100, batch 11/63 completed (loss: 0.0971:  19%|█▉        | 6/31 [00:35<02:26,  5.85s/it]
Training Epoch: 3/100, batch 11/63 completed (loss: 0.2432:  19%|█▉        | 6/31 [00:35<02:26,  5.85s/it]
Training Epoch: 3/100, batch 11/63 completed (loss: 0.0844:  19%|█▉        | 6/31 [00:35<02:26,  5.85s/it]
Training Epoch: 3/100, batch 12/63 completed (loss: 0.0983:  19%|█▉        | 6/31 [00:38<02:26,  5.85s/it]
Training Epoch: 3/100, batch 12/63 completed (loss: 0.1018:  19%|█▉        | 6/31 [00:38<02:26,  5.85s/it]
Training Epoch: 3/100, batch 12/63 completed (loss: 0.1202:  19%|█▉        | 6/31 [00:38<02:26,  5.85s/it]
Training Epoch: 3/100, batch 12/63 completed (loss: 0.3286:  19%|█▉        | 6/31 [00:38<02:26,  5.85s/it]
Training Epoch: 3/100, batch 12/63 completed (loss: 0.0983:  23%|██▎       | 7/31 [00:40<02:18,  5.79s/it]
Training Epoch: 3/100, batch 12/63 completed (loss: 0.1018:  23%|██▎       | 7/31 [00:40<02:18,  5.78s/it]
Training Epoch: 3/100, batch 12/63 completed (loss: 0.3286:  23%|██▎       | 7/31 [00:40<02:18,  5.79s/it]
Training Epoch: 3/100, batch 12/63 completed (loss: 0.1202:  23%|██▎       | 7/31 [00:40<02:18,  5.79s/it]
Training Epoch: 3/100, batch 13/63 completed (loss: 0.2373:  23%|██▎       | 7/31 [00:40<02:18,  5.78s/it]
Training Epoch: 3/100, batch 13/63 completed (loss: 0.1639:  23%|██▎       | 7/31 [00:40<02:18,  5.79s/it]
Training Epoch: 3/100, batch 13/63 completed (loss: 0.3257:  23%|██▎       | 7/31 [00:40<02:18,  5.79s/it]
Training Epoch: 3/100, batch 13/63 completed (loss: 0.1091:  23%|██▎       | 7/31 [00:40<02:18,  5.79s/it]
Training Epoch: 3/100, batch 14/63 completed (loss: 0.1826:  23%|██▎       | 7/31 [00:43<02:18,  5.79s/it]
Training Epoch: 3/100, batch 14/63 completed (loss: 0.1528:  23%|██▎       | 7/31 [00:43<02:18,  5.79s/it]
Training Epoch: 3/100, batch 14/63 completed (loss: 0.108:  23%|██▎       | 7/31 [00:43<02:18,  5.78s/it] 
Training Epoch: 3/100, batch 14/63 completed (loss: 0.0821:  23%|██▎       | 7/31 [00:43<02:18,  5.79s/it]
Training Epoch: 3/100, batch 14/63 completed (loss: 0.108:  26%|██▌       | 8/31 [00:47<02:16,  5.95s/it]
Training Epoch: 3/100, batch 14/63 completed (loss: 0.0821:  26%|██▌       | 8/31 [00:47<02:16,  5.95s/it]
Training Epoch: 3/100, batch 14/63 completed (loss: 0.1528:  26%|██▌       | 8/31 [00:47<02:16,  5.95s/it]
Training Epoch: 3/100, batch 14/63 completed (loss: 0.1826:  26%|██▌       | 8/31 [00:47<02:16,  5.95s/it]
Training Epoch: 3/100, batch 15/63 completed (loss: 0.1284:  26%|██▌       | 8/31 [00:47<02:16,  5.95s/it]
Training Epoch: 3/100, batch 15/63 completed (loss: 0.1653:  26%|██▌       | 8/31 [00:47<02:16,  5.95s/it]
Training Epoch: 3/100, batch 15/63 completed (loss: 0.2387:  26%|██▌       | 8/31 [00:47<02:16,  5.95s/it]
Training Epoch: 3/100, batch 15/63 completed (loss: 0.0919:  26%|██▌       | 8/31 [00:47<02:16,  5.95s/it]
Training Epoch: 3/100, batch 16/63 completed (loss: 0.1905:  26%|██▌       | 8/31 [00:50<02:16,  5.95s/it]
Training Epoch: 3/100, batch 16/63 completed (loss: 0.127:  26%|██▌       | 8/31 [00:50<02:16,  5.95s/it] 
Training Epoch: 3/100, batch 16/63 completed (loss: 0.1176:  26%|██▌       | 8/31 [00:50<02:16,  5.95s/it]
Training Epoch: 3/100, batch 16/63 completed (loss: 0.187:  26%|██▌       | 8/31 [00:50<02:16,  5.95s/it] 
Training Epoch: 3/100, batch 16/63 completed (loss: 0.187:  29%|██▉       | 9/31 [00:52<02:09,  5.87s/it]
Training Epoch: 3/100, batch 16/63 completed (loss: 0.1176:  29%|██▉       | 9/31 [00:52<02:09,  5.88s/it]
Training Epoch: 3/100, batch 16/63 completed (loss: 0.127:  29%|██▉       | 9/31 [00:52<02:09,  5.88s/it]
Training Epoch: 3/100, batch 16/63 completed (loss: 0.1905:  29%|██▉       | 9/31 [00:52<02:09,  5.88s/it]
Training Epoch: 3/100, batch 17/63 completed (loss: 0.0958:  29%|██▉       | 9/31 [00:52<02:09,  5.88s/it]
Training Epoch: 3/100, batch 17/63 completed (loss: 0.1779:  29%|██▉       | 9/31 [00:52<02:09,  5.88s/it]
Training Epoch: 3/100, batch 17/63 completed (loss: 0.3441:  29%|██▉       | 9/31 [00:52<02:09,  5.88s/it]
Training Epoch: 3/100, batch 17/63 completed (loss: 0.2652:  29%|██▉       | 9/31 [00:52<02:09,  5.87s/it]
Training Epoch: 3/100, batch 18/63 completed (loss: 0.0914:  29%|██▉       | 9/31 [00:55<02:09,  5.88s/it]
Training Epoch: 3/100, batch 18/63 completed (loss: 0.1746:  29%|██▉       | 9/31 [00:55<02:09,  5.88s/it]
Training Epoch: 3/100, batch 18/63 completed (loss: 0.2065:  29%|██▉       | 9/31 [00:55<02:09,  5.88s/it]
Training Epoch: 3/100, batch 18/63 completed (loss: 0.2324:  29%|██▉       | 9/31 [00:55<02:09,  5.87s/it]
Training Epoch: 3/100, batch 18/63 completed (loss: 0.2065:  32%|███▏      | 10/31 [00:58<02:03,  5.87s/it]
Training Epoch: 3/100, batch 18/63 completed (loss: 0.2324:  32%|███▏      | 10/31 [00:58<02:03,  5.87s/it]
Training Epoch: 3/100, batch 18/63 completed (loss: 0.0914:  32%|███▏      | 10/31 [00:58<02:03,  5.87s/it]
Training Epoch: 3/100, batch 18/63 completed (loss: 0.1746:  32%|███▏      | 10/31 [00:58<02:03,  5.88s/it]
Training Epoch: 3/100, batch 19/63 completed (loss: 0.1163:  32%|███▏      | 10/31 [00:58<02:03,  5.88s/it]
Training Epoch: 3/100, batch 19/63 completed (loss: 0.1236:  32%|███▏      | 10/31 [00:58<02:03,  5.87s/it]
Training Epoch: 3/100, batch 19/63 completed (loss: 0.1295:  32%|███▏      | 10/31 [00:58<02:03,  5.87s/it]
Training Epoch: 3/100, batch 19/63 completed (loss: 0.1565:  32%|███▏      | 10/31 [00:58<02:03,  5.87s/it]
Training Epoch: 3/100, batch 20/63 completed (loss: 0.2164:  32%|███▏      | 10/31 [01:01<02:03,  5.87s/it]
Training Epoch: 3/100, batch 20/63 completed (loss: 0.0951:  32%|███▏      | 10/31 [01:01<02:03,  5.88s/it]
Training Epoch: 3/100, batch 20/63 completed (loss: 0.1736:  32%|███▏      | 10/31 [01:01<02:03,  5.87s/it]
Training Epoch: 3/100, batch 20/63 completed (loss: 0.1776:  32%|███▏      | 10/31 [01:01<02:03,  5.87s/it]
Training Epoch: 3/100, batch 20/63 completed (loss: 0.1776:  35%|███▌      | 11/31 [01:04<01:58,  5.92s/it]
Training Epoch: 3/100, batch 20/63 completed (loss: 0.1736:  35%|███▌      | 11/31 [01:04<01:58,  5.92s/it]
Training Epoch: 3/100, batch 20/63 completed (loss: 0.0951:  35%|███▌      | 11/31 [01:04<01:58,  5.92s/it]
Training Epoch: 3/100, batch 20/63 completed (loss: 0.2164:  35%|███▌      | 11/31 [01:04<01:58,  5.92s/it]
Training Epoch: 3/100, batch 21/63 completed (loss: 0.0988:  35%|███▌      | 11/31 [01:04<01:58,  5.92s/it]
Training Epoch: 3/100, batch 21/63 completed (loss: 0.1905:  35%|███▌      | 11/31 [01:04<01:58,  5.92s/it]
Training Epoch: 3/100, batch 21/63 completed (loss: 0.1955:  35%|███▌      | 11/31 [01:04<01:58,  5.92s/it]
Training Epoch: 3/100, batch 21/63 completed (loss: 0.1238:  35%|███▌      | 11/31 [01:04<01:58,  5.92s/it]
Training Epoch: 3/100, batch 22/63 completed (loss: 0.1148:  35%|███▌      | 11/31 [01:07<01:58,  5.92s/it]
Training Epoch: 3/100, batch 22/63 completed (loss: 0.0825:  35%|███▌      | 11/31 [01:07<01:58,  5.92s/it]
Training Epoch: 3/100, batch 22/63 completed (loss: 0.0885:  35%|███▌      | 11/31 [01:07<01:58,  5.92s/it]
Training Epoch: 3/100, batch 22/63 completed (loss: 0.202:  35%|███▌      | 11/31 [01:07<01:58,  5.92s/it] 
Training Epoch: 3/100, batch 22/63 completed (loss: 0.202:  39%|███▊      | 12/31 [01:10<01:51,  5.86s/it]
Training Epoch: 3/100, batch 22/63 completed (loss: 0.0885:  39%|███▊      | 12/31 [01:10<01:51,  5.86s/it]
Training Epoch: 3/100, batch 22/63 completed (loss: 0.1148:  39%|███▊      | 12/31 [01:10<01:51,  5.86s/it]
Training Epoch: 3/100, batch 22/63 completed (loss: 0.0825:  39%|███▊      | 12/31 [01:10<01:51,  5.86s/it]
Training Epoch: 3/100, batch 23/63 completed (loss: 0.2154:  39%|███▊      | 12/31 [01:10<01:51,  5.86s/it]
Training Epoch: 3/100, batch 23/63 completed (loss: 0.1511:  39%|███▊      | 12/31 [01:10<01:51,  5.86s/it]
Training Epoch: 3/100, batch 23/63 completed (loss: 0.1558:  39%|███▊      | 12/31 [01:10<01:51,  5.86s/it]
Training Epoch: 3/100, batch 23/63 completed (loss: 0.2229:  39%|███▊      | 12/31 [01:10<01:51,  5.86s/it]
Training Epoch: 3/100, batch 24/63 completed (loss: 0.2621:  39%|███▊      | 12/31 [01:13<01:51,  5.86s/it]
Training Epoch: 3/100, batch 24/63 completed (loss: 0.2402:  39%|███▊      | 12/31 [01:13<01:51,  5.86s/it]
Training Epoch: 3/100, batch 24/63 completed (loss: 0.1738:  39%|███▊      | 12/31 [01:13<01:51,  5.86s/it]
Training Epoch: 3/100, batch 24/63 completed (loss: 0.1557:  39%|███▊      | 12/31 [01:13<01:51,  5.86s/it]
Training Epoch: 3/100, batch 24/63 completed (loss: 0.1557:  42%|████▏     | 13/31 [01:16<01:46,  5.92s/it]
Training Epoch: 3/100, batch 24/63 completed (loss: 0.2621:  42%|████▏     | 13/31 [01:16<01:46,  5.92s/it]
Training Epoch: 3/100, batch 24/63 completed (loss: 0.2402:  42%|████▏     | 13/31 [01:16<01:46,  5.92s/it]
Training Epoch: 3/100, batch 24/63 completed (loss: 0.1738:  42%|████▏     | 13/31 [01:16<01:46,  5.92s/it]
Training Epoch: 3/100, batch 25/63 completed (loss: 0.1117:  42%|████▏     | 13/31 [01:16<01:46,  5.92s/it]
Training Epoch: 3/100, batch 25/63 completed (loss: 0.1328:  42%|████▏     | 13/31 [01:16<01:46,  5.92s/it]
Training Epoch: 3/100, batch 25/63 completed (loss: 0.1138:  42%|████▏     | 13/31 [01:16<01:46,  5.92s/it]
Training Epoch: 3/100, batch 25/63 completed (loss: 0.2233:  42%|████▏     | 13/31 [01:16<01:46,  5.92s/it]
Training Epoch: 3/100, batch 26/63 completed (loss: 0.2108:  42%|████▏     | 13/31 [01:19<01:46,  5.92s/it]
Training Epoch: 3/100, batch 26/63 completed (loss: 0.1995:  42%|████▏     | 13/31 [01:19<01:46,  5.92s/it]
Training Epoch: 3/100, batch 26/63 completed (loss: 0.1093:  42%|████▏     | 13/31 [01:19<01:46,  5.92s/it]
Training Epoch: 3/100, batch 26/63 completed (loss: 0.0974:  42%|████▏     | 13/31 [01:19<01:46,  5.92s/it]
Training Epoch: 3/100, batch 26/63 completed (loss: 0.0974:  45%|████▌     | 14/31 [01:22<01:40,  5.91s/it]
Training Epoch: 3/100, batch 26/63 completed (loss: 0.1995:  45%|████▌     | 14/31 [01:22<01:40,  5.90s/it]
Training Epoch: 3/100, batch 26/63 completed (loss: 0.1093:  45%|████▌     | 14/31 [01:22<01:40,  5.90s/it]
Training Epoch: 3/100, batch 26/63 completed (loss: 0.2108:  45%|████▌     | 14/31 [01:22<01:40,  5.90s/it]
Training Epoch: 3/100, batch 27/63 completed (loss: 0.2294:  45%|████▌     | 14/31 [01:22<01:40,  5.90s/it]
Training Epoch: 3/100, batch 27/63 completed (loss: 0.1713:  45%|████▌     | 14/31 [01:22<01:40,  5.90s/it]
Training Epoch: 3/100, batch 27/63 completed (loss: 0.282:  45%|████▌     | 14/31 [01:22<01:40,  5.90s/it] 
Training Epoch: 3/100, batch 27/63 completed (loss: 0.2951:  45%|████▌     | 14/31 [01:22<01:40,  5.91s/it]
Training Epoch: 3/100, batch 28/63 completed (loss: 0.214:  45%|████▌     | 14/31 [01:25<01:40,  5.90s/it] 
Training Epoch: 3/100, batch 28/63 completed (loss: 0.3016:  45%|████▌     | 14/31 [01:25<01:40,  5.90s/it]
Training Epoch: 3/100, batch 28/63 completed (loss: 0.1448:  45%|████▌     | 14/31 [01:25<01:40,  5.90s/it]
Training Epoch: 3/100, batch 28/63 completed (loss: 0.1335:  45%|████▌     | 14/31 [01:25<01:40,  5.91s/it]
Training Epoch: 3/100, batch 28/63 completed (loss: 0.214:  48%|████▊     | 15/31 [01:28<01:33,  5.85s/it]
Training Epoch: 3/100, batch 28/63 completed (loss: 0.1335:  48%|████▊     | 15/31 [01:28<01:33,  5.86s/it]
Training Epoch: 3/100, batch 28/63 completed (loss: 0.3016:  48%|████▊     | 15/31 [01:28<01:33,  5.86s/it]
Training Epoch: 3/100, batch 28/63 completed (loss: 0.1448:  48%|████▊     | 15/31 [01:28<01:33,  5.86s/it]
Training Epoch: 3/100, batch 29/63 completed (loss: 0.1111:  48%|████▊     | 15/31 [01:28<01:33,  5.85s/it]
Training Epoch: 3/100, batch 29/63 completed (loss: 0.2306:  48%|████▊     | 15/31 [01:28<01:33,  5.86s/it]
Training Epoch: 3/100, batch 29/63 completed (loss: 0.3241:  48%|████▊     | 15/31 [01:28<01:33,  5.86s/it]
Training Epoch: 3/100, batch 29/63 completed (loss: 0.0971:  48%|████▊     | 15/31 [01:28<01:33,  5.86s/it]
Training Epoch: 3/100, batch 30/63 completed (loss: 0.3568:  48%|████▊     | 15/31 [01:31<01:33,  5.86s/it]
Training Epoch: 3/100, batch 30/63 completed (loss: 0.1933:  48%|████▊     | 15/31 [01:31<01:33,  5.85s/it]
Training Epoch: 3/100, batch 30/63 completed (loss: 0.1234:  48%|████▊     | 15/31 [01:31<01:33,  5.86s/it]
Training Epoch: 3/100, batch 30/63 completed (loss: 0.2209:  48%|████▊     | 15/31 [01:31<01:33,  5.86s/it]
Training Epoch: 3/100, batch 30/63 completed (loss: 0.1234:  52%|█████▏    | 16/31 [01:34<01:29,  6.00s/it]
Training Epoch: 3/100, batch 30/63 completed (loss: 0.2209:  52%|█████▏    | 16/31 [01:34<01:30,  6.00s/it]
Training Epoch: 3/100, batch 30/63 completed (loss: 0.3568:  52%|█████▏    | 16/31 [01:34<01:29,  6.00s/it]
Training Epoch: 3/100, batch 30/63 completed (loss: 0.1933:  52%|█████▏    | 16/31 [01:34<01:30,  6.00s/it]
Training Epoch: 3/100, batch 31/63 completed (loss: 0.1396:  52%|█████▏    | 16/31 [01:34<01:29,  6.00s/it]
Training Epoch: 3/100, batch 31/63 completed (loss: 0.3454:  52%|█████▏    | 16/31 [01:34<01:29,  6.00s/it]
Training Epoch: 3/100, batch 31/63 completed (loss: 0.1971:  52%|█████▏    | 16/31 [01:34<01:30,  6.00s/it]
Training Epoch: 3/100, batch 31/63 completed (loss: 0.2586:  52%|█████▏    | 16/31 [01:34<01:30,  6.00s/it]
Training Epoch: 3/100, batch 32/63 completed (loss: 0.1646:  52%|█████▏    | 16/31 [01:37<01:30,  6.00s/it]
Training Epoch: 3/100, batch 32/63 completed (loss: 0.1733:  52%|█████▏    | 16/31 [01:37<01:29,  6.00s/it]
Training Epoch: 3/100, batch 32/63 completed (loss: 0.1393:  52%|█████▏    | 16/31 [01:37<01:29,  6.00s/it]
Training Epoch: 3/100, batch 32/63 completed (loss: 0.3955:  52%|█████▏    | 16/31 [01:37<01:30,  6.00s/it]
Training Epoch: 3/100, batch 32/63 completed (loss: 0.3955:  55%|█████▍    | 17/31 [01:40<01:22,  5.92s/it]
Training Epoch: 3/100, batch 32/63 completed (loss: 0.1393:  55%|█████▍    | 17/31 [01:40<01:22,  5.92s/it]
Training Epoch: 3/100, batch 32/63 completed (loss: 0.1733:  55%|█████▍    | 17/31 [01:40<01:22,  5.92s/it]
Training Epoch: 3/100, batch 32/63 completed (loss: 0.1646:  55%|█████▍    | 17/31 [01:40<01:22,  5.93s/it]
Training Epoch: 3/100, batch 33/63 completed (loss: 0.3508:  55%|█████▍    | 17/31 [01:40<01:22,  5.92s/it]
Training Epoch: 3/100, batch 33/63 completed (loss: 0.1312:  55%|█████▍    | 17/31 [01:40<01:22,  5.93s/it]
Training Epoch: 3/100, batch 33/63 completed (loss: 0.1202:  55%|█████▍    | 17/31 [01:40<01:22,  5.92s/it]
Training Epoch: 3/100, batch 33/63 completed (loss: 0.0819:  55%|█████▍    | 17/31 [01:40<01:22,  5.92s/it]
Training Epoch: 3/100, batch 34/63 completed (loss: 0.1087:  55%|█████▍    | 17/31 [01:43<01:22,  5.92s/it]
Training Epoch: 3/100, batch 34/63 completed (loss: 0.2111:  55%|█████▍    | 17/31 [01:43<01:22,  5.92s/it]
Training Epoch: 3/100, batch 34/63 completed (loss: 0.1962:  55%|█████▍    | 17/31 [01:43<01:22,  5.93s/it]
Training Epoch: 3/100, batch 34/63 completed (loss: 0.2837:  55%|█████▍    | 17/31 [01:43<01:22,  5.92s/it]
Training Epoch: 3/100, batch 34/63 completed (loss: 0.2111:  58%|█████▊    | 18/31 [01:46<01:17,  5.97s/it]
Training Epoch: 3/100, batch 34/63 completed (loss: 0.2837:  58%|█████▊    | 18/31 [01:46<01:17,  5.97s/it]
Training Epoch: 3/100, batch 34/63 completed (loss: 0.1962:  58%|█████▊    | 18/31 [01:46<01:17,  5.97s/it]
Training Epoch: 3/100, batch 34/63 completed (loss: 0.1087:  58%|█████▊    | 18/31 [01:46<01:17,  5.98s/it]
Training Epoch: 3/100, batch 35/63 completed (loss: 0.1388:  58%|█████▊    | 18/31 [01:46<01:17,  5.97s/it]
Training Epoch: 3/100, batch 35/63 completed (loss: 0.0894:  58%|█████▊    | 18/31 [01:46<01:17,  5.97s/it]
Training Epoch: 3/100, batch 35/63 completed (loss: 0.2303:  58%|█████▊    | 18/31 [01:46<01:17,  5.98s/it]
Training Epoch: 3/100, batch 35/63 completed (loss: 0.0844:  58%|█████▊    | 18/31 [01:46<01:17,  5.97s/it]
Training Epoch: 3/100, batch 36/63 completed (loss: 0.3237:  58%|█████▊    | 18/31 [01:49<01:17,  5.97s/it]
Training Epoch: 3/100, batch 36/63 completed (loss: 0.248:  58%|█████▊    | 18/31 [01:49<01:17,  5.98s/it] 
Training Epoch: 3/100, batch 36/63 completed (loss: 0.2571:  58%|█████▊    | 18/31 [01:49<01:17,  5.97s/it]
Training Epoch: 3/100, batch 36/63 completed (loss: 0.2041:  58%|█████▊    | 18/31 [01:49<01:17,  5.97s/it]
Training Epoch: 3/100, batch 36/63 completed (loss: 0.2571:  61%|██████▏   | 19/31 [01:52<01:11,  5.96s/it]
Training Epoch: 3/100, batch 36/63 completed (loss: 0.2041:  61%|██████▏   | 19/31 [01:52<01:11,  5.96s/it]
Training Epoch: 3/100, batch 36/63 completed (loss: 0.3237:  61%|██████▏   | 19/31 [01:52<01:11,  5.96s/it]
Training Epoch: 3/100, batch 36/63 completed (loss: 0.248:  61%|██████▏   | 19/31 [01:52<01:11,  5.97s/it]
Training Epoch: 3/100, batch 37/63 completed (loss: 0.1498:  61%|██████▏   | 19/31 [01:52<01:11,  5.96s/it]
Training Epoch: 3/100, batch 37/63 completed (loss: 0.1794:  61%|██████▏   | 19/31 [01:52<01:11,  5.97s/it]
Training Epoch: 3/100, batch 37/63 completed (loss: 0.4174:  61%|██████▏   | 19/31 [01:52<01:11,  5.96s/it]
Training Epoch: 3/100, batch 37/63 completed (loss: 0.197:  61%|██████▏   | 19/31 [01:52<01:11,  5.96s/it] 
Training Epoch: 3/100, batch 38/63 completed (loss: 0.1972:  61%|██████▏   | 19/31 [01:55<01:11,  5.96s/it]
Training Epoch: 3/100, batch 38/63 completed (loss: 0.3023:  61%|██████▏   | 19/31 [01:55<01:11,  5.97s/it]
Training Epoch: 3/100, batch 38/63 completed (loss: 0.1273:  61%|██████▏   | 19/31 [01:55<01:11,  5.96s/it]
Training Epoch: 3/100, batch 38/63 completed (loss: 0.2645:  61%|██████▏   | 19/31 [01:55<01:11,  5.96s/it]
Training Epoch: 3/100, batch 38/63 completed (loss: 0.1972:  65%|██████▍   | 20/31 [01:58<01:04,  5.90s/it]
Training Epoch: 3/100, batch 38/63 completed (loss: 0.2645:  65%|██████▍   | 20/31 [01:58<01:04,  5.90s/it]
Training Epoch: 3/100, batch 38/63 completed (loss: 0.3023:  65%|██████▍   | 20/31 [01:58<01:04,  5.90s/it]
Training Epoch: 3/100, batch 38/63 completed (loss: 0.1273:  65%|██████▍   | 20/31 [01:58<01:04,  5.90s/it]
Training Epoch: 3/100, batch 39/63 completed (loss: 0.1866:  65%|██████▍   | 20/31 [01:58<01:04,  5.90s/it]
Training Epoch: 3/100, batch 39/63 completed (loss: 0.1305:  65%|██████▍   | 20/31 [01:58<01:04,  5.90s/it]
Training Epoch: 3/100, batch 39/63 completed (loss: 0.1527:  65%|██████▍   | 20/31 [01:58<01:04,  5.90s/it]
Training Epoch: 3/100, batch 39/63 completed (loss: 0.1465:  65%|██████▍   | 20/31 [01:58<01:04,  5.90s/it]
Training Epoch: 3/100, batch 40/63 completed (loss: 0.1645:  65%|██████▍   | 20/31 [02:01<01:04,  5.90s/it]
Training Epoch: 3/100, batch 40/63 completed (loss: 0.3716:  65%|██████▍   | 20/31 [02:01<01:04,  5.90s/it]
Training Epoch: 3/100, batch 40/63 completed (loss: 0.131:  65%|██████▍   | 20/31 [02:01<01:04,  5.90s/it] 
Training Epoch: 3/100, batch 40/63 completed (loss: 0.3653:  65%|██████▍   | 20/31 [02:01<01:04,  5.90s/it]
Training Epoch: 3/100, batch 40/63 completed (loss: 0.1645:  68%|██████▊   | 21/31 [02:04<01:00,  6.01s/it]
Training Epoch: 3/100, batch 40/63 completed (loss: 0.3653:  68%|██████▊   | 21/31 [02:04<01:00,  6.01s/it]
Training Epoch: 3/100, batch 40/63 completed (loss: 0.3716:  68%|██████▊   | 21/31 [02:04<01:00,  6.01s/it]
Training Epoch: 3/100, batch 40/63 completed (loss: 0.131:  68%|██████▊   | 21/31 [02:04<01:00,  6.02s/it]
Training Epoch: 3/100, batch 41/63 completed (loss: 0.1542:  68%|██████▊   | 21/31 [02:04<01:00,  6.01s/it]
Training Epoch: 3/100, batch 41/63 completed (loss: 0.1792:  68%|██████▊   | 21/31 [02:04<01:00,  6.02s/it]
Training Epoch: 3/100, batch 41/63 completed (loss: 0.296:  68%|██████▊   | 21/31 [02:04<01:00,  6.01s/it] 
Training Epoch: 3/100, batch 41/63 completed (loss: 0.1736:  68%|██████▊   | 21/31 [02:04<01:00,  6.01s/it]
Training Epoch: 3/100, batch 42/63 completed (loss: 0.1738:  68%|██████▊   | 21/31 [02:07<01:00,  6.01s/it]
Training Epoch: 3/100, batch 42/63 completed (loss: 0.1436:  68%|██████▊   | 21/31 [02:07<01:00,  6.01s/it]
Training Epoch: 3/100, batch 42/63 completed (loss: 0.2116:  68%|██████▊   | 21/31 [02:07<01:00,  6.02s/it]
Training Epoch: 3/100, batch 42/63 completed (loss: 0.2002:  68%|██████▊   | 21/31 [02:07<01:00,  6.01s/it]
Training Epoch: 3/100, batch 42/63 completed (loss: 0.1436:  71%|███████   | 22/31 [02:10<00:53,  5.99s/it]
Training Epoch: 3/100, batch 42/63 completed (loss: 0.1738:  71%|███████   | 22/31 [02:10<00:53,  5.99s/it]
Training Epoch: 3/100, batch 42/63 completed (loss: 0.2116:  71%|███████   | 22/31 [02:10<00:53,  5.99s/it]
Training Epoch: 3/100, batch 42/63 completed (loss: 0.2002:  71%|███████   | 22/31 [02:10<00:53,  5.99s/it]
Training Epoch: 3/100, batch 43/63 completed (loss: 0.2448:  71%|███████   | 22/31 [02:10<00:53,  5.99s/it]
Training Epoch: 3/100, batch 43/63 completed (loss: 0.1333:  71%|███████   | 22/31 [02:10<00:53,  5.99s/it]
Training Epoch: 3/100, batch 43/63 completed (loss: 0.2184:  71%|███████   | 22/31 [02:10<00:53,  5.99s/it]
Training Epoch: 3/100, batch 43/63 completed (loss: 0.2682:  71%|███████   | 22/31 [02:10<00:53,  5.99s/it]
Training Epoch: 3/100, batch 44/63 completed (loss: 0.1339:  71%|███████   | 22/31 [02:13<00:53,  5.99s/it]
Training Epoch: 3/100, batch 44/63 completed (loss: 0.0689:  71%|███████   | 22/31 [02:13<00:53,  5.99s/it]
Training Epoch: 3/100, batch 44/63 completed (loss: 0.2278:  71%|███████   | 22/31 [02:13<00:53,  5.99s/it]
Training Epoch: 3/100, batch 44/63 completed (loss: 0.1314:  71%|███████   | 22/31 [02:13<00:53,  5.99s/it]
Training Epoch: 3/100, batch 44/63 completed (loss: 0.1339:  74%|███████▍  | 23/31 [02:16<00:47,  5.97s/it]
Training Epoch: 3/100, batch 44/63 completed (loss: 0.2278:  74%|███████▍  | 23/31 [02:16<00:47,  5.97s/it]
Training Epoch: 3/100, batch 44/63 completed (loss: 0.1314:  74%|███████▍  | 23/31 [02:16<00:47,  5.97s/it]
Training Epoch: 3/100, batch 44/63 completed (loss: 0.0689:  74%|███████▍  | 23/31 [02:16<00:47,  5.97s/it]
Training Epoch: 3/100, batch 45/63 completed (loss: 0.2895:  74%|███████▍  | 23/31 [02:16<00:47,  5.97s/it]
Training Epoch: 3/100, batch 45/63 completed (loss: 0.1459:  74%|███████▍  | 23/31 [02:16<00:47,  5.97s/it]
Training Epoch: 3/100, batch 45/63 completed (loss: 0.153:  74%|███████▍  | 23/31 [02:16<00:47,  5.97s/it] 
Training Epoch: 3/100, batch 45/63 completed (loss: 0.1263:  74%|███████▍  | 23/31 [02:16<00:47,  5.97s/it]
Training Epoch: 3/100, batch 46/63 completed (loss: 0.2422:  74%|███████▍  | 23/31 [02:19<00:47,  5.97s/it]
Training Epoch: 3/100, batch 46/63 completed (loss: 0.2949:  74%|███████▍  | 23/31 [02:19<00:47,  5.97s/it]
Training Epoch: 3/100, batch 46/63 completed (loss: 0.2751:  74%|███████▍  | 23/31 [02:19<00:47,  5.97s/it]
Training Epoch: 3/100, batch 46/63 completed (loss: 0.1779:  74%|███████▍  | 23/31 [02:19<00:47,  5.97s/it]
Training Epoch: 3/100, batch 46/63 completed (loss: 0.2751:  77%|███████▋  | 24/31 [02:22<00:42,  6.06s/it]
Training Epoch: 3/100, batch 46/63 completed (loss: 0.1779:  77%|███████▋  | 24/31 [02:22<00:42,  6.06s/it]
Training Epoch: 3/100, batch 46/63 completed (loss: 0.2949:  77%|███████▋  | 24/31 [02:22<00:42,  6.06s/it]
Training Epoch: 3/100, batch 46/63 completed (loss: 0.2422:  77%|███████▋  | 24/31 [02:22<00:42,  6.06s/it]
Training Epoch: 3/100, batch 47/63 completed (loss: 0.304:  77%|███████▋  | 24/31 [02:22<00:42,  6.06s/it] 
Training Epoch: 3/100, batch 47/63 completed (loss: 0.109:  77%|███████▋  | 24/31 [02:22<00:42,  6.06s/it] 
Training Epoch: 3/100, batch 47/63 completed (loss: 0.1346:  77%|███████▋  | 24/31 [02:22<00:42,  6.06s/it]
Training Epoch: 3/100, batch 47/63 completed (loss: 0.1361:  77%|███████▋  | 24/31 [02:22<00:42,  6.06s/it]
Training Epoch: 3/100, batch 48/63 completed (loss: 0.1679:  77%|███████▋  | 24/31 [02:25<00:42,  6.06s/it]
Training Epoch: 3/100, batch 48/63 completed (loss: 0.1303:  77%|███████▋  | 24/31 [02:25<00:42,  6.06s/it]
Training Epoch: 3/100, batch 48/63 completed (loss: 0.1877:  77%|███████▋  | 24/31 [02:25<00:42,  6.06s/it]
Training Epoch: 3/100, batch 48/63 completed (loss: 0.0776:  77%|███████▋  | 24/31 [02:25<00:42,  6.06s/it]
Training Epoch: 3/100, batch 48/63 completed (loss: 0.1877:  81%|████████  | 25/31 [02:28<00:35,  5.99s/it]
Training Epoch: 3/100, batch 48/63 completed (loss: 0.1303:  81%|████████  | 25/31 [02:28<00:35,  5.99s/it]
Training Epoch: 3/100, batch 48/63 completed (loss: 0.1679:  81%|████████  | 25/31 [02:28<00:35,  5.99s/it]
Training Epoch: 3/100, batch 48/63 completed (loss: 0.0776:  81%|████████  | 25/31 [02:28<00:35,  5.99s/it]
Training Epoch: 3/100, batch 49/63 completed (loss: 0.0875:  81%|████████  | 25/31 [02:28<00:35,  5.99s/it]
Training Epoch: 3/100, batch 49/63 completed (loss: 0.2924:  81%|████████  | 25/31 [02:28<00:35,  5.99s/it]
Training Epoch: 3/100, batch 49/63 completed (loss: 0.1746:  81%|████████  | 25/31 [02:28<00:35,  5.99s/it]
Training Epoch: 3/100, batch 49/63 completed (loss: 0.2508:  81%|████████  | 25/31 [02:28<00:35,  5.99s/it]
Training Epoch: 3/100, batch 50/63 completed (loss: 0.121:  81%|████████  | 25/31 [02:31<00:35,  5.99s/it] 
Training Epoch: 3/100, batch 50/63 completed (loss: 0.2146:  81%|████████  | 25/31 [02:31<00:35,  5.99s/it]
Training Epoch: 3/100, batch 50/63 completed (loss: 0.1368:  81%|████████  | 25/31 [02:31<00:35,  5.99s/it]
Training Epoch: 3/100, batch 50/63 completed (loss: 0.1941:  81%|████████  | 25/31 [02:31<00:35,  5.99s/it]
Training Epoch: 3/100, batch 50/63 completed (loss: 0.121:  84%|████████▍ | 26/31 [02:34<00:30,  6.09s/it]
Training Epoch: 3/100, batch 50/63 completed (loss: 0.1368:  84%|████████▍ | 26/31 [02:34<00:30,  6.09s/it]
Training Epoch: 3/100, batch 50/63 completed (loss: 0.2146:  84%|████████▍ | 26/31 [02:34<00:30,  6.09s/it]
Training Epoch: 3/100, batch 50/63 completed (loss: 0.1941:  84%|████████▍ | 26/31 [02:34<00:30,  6.10s/it]
Training Epoch: 3/100, batch 51/63 completed (loss: 0.1047:  84%|████████▍ | 26/31 [02:34<00:30,  6.09s/it]
Training Epoch: 3/100, batch 51/63 completed (loss: 0.3094:  84%|████████▍ | 26/31 [02:34<00:30,  6.09s/it]
Training Epoch: 3/100, batch 51/63 completed (loss: 0.3414:  84%|████████▍ | 26/31 [02:34<00:30,  6.09s/it]
Training Epoch: 3/100, batch 51/63 completed (loss: 0.1059:  84%|████████▍ | 26/31 [02:34<00:30,  6.10s/it]
Training Epoch: 3/100, batch 52/63 completed (loss: 0.2455:  84%|████████▍ | 26/31 [02:37<00:30,  6.09s/it]
Training Epoch: 3/100, batch 52/63 completed (loss: 0.201:  84%|████████▍ | 26/31 [02:37<00:30,  6.09s/it] 
Training Epoch: 3/100, batch 52/63 completed (loss: 0.2691:  84%|████████▍ | 26/31 [02:37<00:30,  6.09s/it]
Training Epoch: 3/100, batch 52/63 completed (loss: 0.343:  84%|████████▍ | 26/31 [02:37<00:30,  6.10s/it] 
Training Epoch: 3/100, batch 52/63 completed (loss: 0.201:  87%|████████▋ | 27/31 [02:40<00:24,  6.05s/it]
Training Epoch: 3/100, batch 52/63 completed (loss: 0.343:  87%|████████▋ | 27/31 [02:40<00:24,  6.05s/it]
Training Epoch: 3/100, batch 52/63 completed (loss: 0.2455:  87%|████████▋ | 27/31 [02:40<00:24,  6.05s/it]
Training Epoch: 3/100, batch 52/63 completed (loss: 0.2691:  87%|████████▋ | 27/31 [02:40<00:24,  6.05s/it]
Training Epoch: 3/100, batch 53/63 completed (loss: 0.1507:  87%|████████▋ | 27/31 [02:40<00:24,  6.05s/it]
Training Epoch: 3/100, batch 53/63 completed (loss: 0.1736:  87%|████████▋ | 27/31 [02:40<00:24,  6.05s/it]
Training Epoch: 3/100, batch 53/63 completed (loss: 0.1475:  87%|████████▋ | 27/31 [02:40<00:24,  6.05s/it]
Training Epoch: 3/100, batch 53/63 completed (loss: 0.3717:  87%|████████▋ | 27/31 [02:40<00:24,  6.05s/it]
Training Epoch: 3/100, batch 54/63 completed (loss: 0.1148:  87%|████████▋ | 27/31 [02:43<00:24,  6.05s/it]
Training Epoch: 3/100, batch 54/63 completed (loss: 0.1118:  87%|████████▋ | 27/31 [02:43<00:24,  6.05s/it]
Training Epoch: 3/100, batch 54/63 completed (loss: 0.3657:  87%|████████▋ | 27/31 [02:43<00:24,  6.05s/it]
Training Epoch: 3/100, batch 54/63 completed (loss: 0.3566:  87%|████████▋ | 27/31 [02:43<00:24,  6.05s/it]
Training Epoch: 3/100, batch 54/63 completed (loss: 0.1118:  90%|█████████ | 28/31 [02:46<00:17,  5.97s/it]
Training Epoch: 3/100, batch 54/63 completed (loss: 0.1148:  90%|█████████ | 28/31 [02:46<00:17,  5.97s/it]
Training Epoch: 3/100, batch 54/63 completed (loss: 0.3566:  90%|█████████ | 28/31 [02:46<00:17,  5.97s/it]
Training Epoch: 3/100, batch 54/63 completed (loss: 0.3657:  90%|█████████ | 28/31 [02:46<00:17,  5.97s/it]
Training Epoch: 3/100, batch 55/63 completed (loss: 0.0977:  90%|█████████ | 28/31 [02:46<00:17,  5.97s/it]
Training Epoch: 3/100, batch 55/63 completed (loss: 0.2792:  90%|█████████ | 28/31 [02:46<00:17,  5.97s/it]
Training Epoch: 3/100, batch 55/63 completed (loss: 0.1575:  90%|█████████ | 28/31 [02:46<00:17,  5.97s/it]
Training Epoch: 3/100, batch 55/63 completed (loss: 0.2048:  90%|█████████ | 28/31 [02:46<00:17,  5.97s/it]
Training Epoch: 3/100, batch 56/63 completed (loss: 0.1007:  90%|█████████ | 28/31 [02:49<00:17,  5.97s/it]
Training Epoch: 3/100, batch 56/63 completed (loss: 0.2159:  90%|█████████ | 28/31 [02:49<00:17,  5.97s/it]
Training Epoch: 3/100, batch 56/63 completed (loss: 0.1868:  90%|█████████ | 28/31 [02:49<00:17,  5.97s/it]
Training Epoch: 3/100, batch 56/63 completed (loss: 0.1572:  90%|█████████ | 28/31 [02:49<00:17,  5.97s/it]
Training Epoch: 3/100, batch 56/63 completed (loss: 0.2159:  94%|█████████▎| 29/31 [02:52<00:12,  6.05s/it]
Training Epoch: 3/100, batch 56/63 completed (loss: 0.1007:  94%|█████████▎| 29/31 [02:52<00:12,  6.05s/it]
Training Epoch: 3/100, batch 56/63 completed (loss: 0.1868:  94%|█████████▎| 29/31 [02:52<00:12,  6.05s/it]
Training Epoch: 3/100, batch 56/63 completed (loss: 0.1572:  94%|█████████▎| 29/31 [02:52<00:12,  6.05s/it]
Training Epoch: 3/100, batch 57/63 completed (loss: 0.168:  94%|█████████▎| 29/31 [02:52<00:12,  6.05s/it] 
Training Epoch: 3/100, batch 57/63 completed (loss: 0.2075:  94%|█████████▎| 29/31 [02:52<00:12,  6.05s/it]
Training Epoch: 3/100, batch 57/63 completed (loss: 0.1336:  94%|█████████▎| 29/31 [02:52<00:12,  6.05s/it]
Training Epoch: 3/100, batch 57/63 completed (loss: 0.0854:  94%|█████████▎| 29/31 [02:52<00:12,  6.05s/it]
Training Epoch: 3/100, batch 58/63 completed (loss: 0.2104:  94%|█████████▎| 29/31 [02:55<00:12,  6.05s/it]
Training Epoch: 3/100, batch 58/63 completed (loss: 0.2538:  94%|█████████▎| 29/31 [02:55<00:12,  6.05s/it]
Training Epoch: 3/100, batch 58/63 completed (loss: 0.1653:  94%|█████████▎| 29/31 [02:55<00:12,  6.05s/it]
Training Epoch: 3/100, batch 58/63 completed (loss: 0.1196:  94%|█████████▎| 29/31 [02:55<00:12,  6.05s/it]
Training Epoch: 3/100, batch 58/63 completed (loss: 0.2538:  97%|█████████▋| 30/31 [02:58<00:06,  6.06s/it]
Training Epoch: 3/100, batch 58/63 completed (loss: 0.1196:  97%|█████████▋| 30/31 [02:58<00:06,  6.06s/it]
Training Epoch: 3/100, batch 58/63 completed (loss: 0.1653:  97%|█████████▋| 30/31 [02:58<00:06,  6.06s/it]
Training Epoch: 3/100, batch 58/63 completed (loss: 0.2104:  97%|█████████▋| 30/31 [02:58<00:06,  6.06s/it]
Training Epoch: 3/100, batch 59/63 completed (loss: 0.2577:  97%|█████████▋| 30/31 [02:58<00:06,  6.06s/it]
Training Epoch: 3/100, batch 59/63 completed (loss: 0.2453:  97%|█████████▋| 30/31 [02:58<00:06,  6.06s/it]
Training Epoch: 3/100, batch 59/63 completed (loss: 0.194:  97%|█████████▋| 30/31 [02:58<00:06,  6.06s/it] 
Training Epoch: 3/100, batch 59/63 completed (loss: 0.1391:  97%|█████████▋| 30/31 [02:58<00:06,  6.06s/it]
Training Epoch: 3/100, batch 60/63 completed (loss: 0.1219:  97%|█████████▋| 30/31 [03:01<00:06,  6.06s/it]
Training Epoch: 3/100, batch 60/63 completed (loss: 0.1251:  97%|█████████▋| 30/31 [03:01<00:06,  6.06s/it]
Training Epoch: 3/100, batch 60/63 completed (loss: 0.2107:  97%|█████████▋| 30/31 [03:01<00:06,  6.06s/it]
Training Epoch: 3/100, batch 60/63 completed (loss: 0.251:  97%|█████████▋| 30/31 [03:01<00:06,  6.06s/it] 
Training Epoch: 3/100, batch 60/63 completed (loss: 0.251: 100%|██████████| 31/31 [03:04<00:00,  6.05s/it]
Training Epoch: 3/100, batch 60/63 completed (loss: 0.1219: 100%|██████████| 31/31 [03:04<00:00,  6.05s/it]
Training Epoch: 3/100, batch 60/63 completed (loss: 0.1251: 100%|██████████| 31/31 [03:04<00:00,  6.05s/it]
Training Epoch: 3/100, batch 60/63 completed (loss: 0.2107: 100%|██████████| 31/31 [03:04<00:00,  6.05s/it]
Training Epoch: 3/100, batch 61/63 completed (loss: 0.1858: 100%|██████████| 31/31 [03:04<00:00,  6.05s/it]
Training Epoch: 3/100, batch 61/63 completed (loss: 0.2753: 100%|██████████| 31/31 [03:04<00:00,  6.05s/it]
Training Epoch: 3/100, batch 61/63 completed (loss: 0.1229: 100%|██████████| 31/31 [03:04<00:00,  6.05s/it]
Training Epoch: 3/100, batch 61/63 completed (loss: 0.1685: 100%|██████████| 31/31 [03:04<00:00,  6.05s/it]
Training Epoch: 3/100, batch 61/63 completed (loss: 0.1858: : 32it [03:06,  4.84s/it]                      
Training Epoch: 3/100, batch 61/63 completed (loss: 0.1685: : 32it [03:06,  4.84s/it]                      
Training Epoch: 3/100, batch 61/63 completed (loss: 0.2753: : 32it [03:06,  4.84s/it]                      
Training Epoch: 3/100, batch 61/63 completed (loss: 0.1229: : 32it [03:06,  4.84s/it]                      
Training Epoch: 3/100, batch 62/63 completed (loss: 0.1942: : 32it [03:06,  4.84s/it]
Training Epoch: 3/100, batch 62/63 completed (loss: 0.0832: : 32it [03:06,  4.84s/it]
Training Epoch: 3/100, batch 62/63 completed (loss: 0.1148: : 32it [03:06,  4.84s/it]
Training Epoch: 3/100, batch 62/63 completed (loss: 0.2082: : 32it [03:06,  4.84s/it]
Training Epoch: 3/100, batch 62/63 completed (loss: 0.1148: : 32it [03:06,  5.84s/it]

Training Epoch: 3/100, batch 62/63 completed (loss: 0.1942: : 32it [03:06,  5.84s/it]

Training Epoch: 3/100, batch 62/63 completed (loss: 0.0832: : 32it [03:06,  5.84s/it]

Training Epoch: 3/100, batch 62/63 completed (loss: 0.2082: : 32it [03:07,  5.85s/it]
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/c10d_logger.py:83: UserWarning: barrier(): using the device under current context. You can specify `device_id` in `init_process_group` to mute this warning.
  return func(*args, **kwargs)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
saving model + train state.
Warning: failed to upload checkpoint_3 to HF: (Request ID: Root=1-6a1b667a-0273406047ae01017a7bd53e;e072a807-e338-43d4-bf8b-2ef825acfa3c)

403 Forbidden: You need to setup automatic credit recharge in order to upload more data. You can do so at /settings/billing..
Cannot access content at: https://huggingface.co/seyedparsa/thoughtformer-checkpoints.git/info/lfs/objects/batch.
Make sure your token has the correct permissions.
Warning: failed to upload train_state_3 to HF: (Request ID: Root=1-6a1b667d-23cb57b917feba531d936723;1db653d4-34f1-474d-8447-f7ce3084c7e2)

403 Forbidden: You need to setup automatic credit recharge in order to upload more data. You can do so at /settings/billing..
Cannot access content at: https://huggingface.co/seyedparsa/thoughtformer-checkpoints.git/info/lfs/objects/batch.
Make sure your token has the correct permissions.
Deleted old checkpoint: train_state_2
Deleted old checkpoint: checkpoint_2
[warn] HF upload failed for ckpt_3; kept locally, earlier ckpts deleted.

Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]
Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]
Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]eval loss 0.15297889496598924

Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]
Test Accuracy:   3%|▎         | 1/32 [00:06<03:25,  6.62s/it]
[WRONG] Example 0 (diff=3)
Test accuracy: 0.5:   3%|▎         | 1/32 [00:06<03:25,  6.62s/it]
Test Accuracy:   3%|▎         | 1/32 [00:06<03:25,  6.62s/it]

Test accuracy: 0.62:   3%|▎         | 1/32 [00:06<03:25,  6.62s/it]  Expected : 'Claude'
  Extracted: 'Morgan'
  Generated: '### Morgan'

Test Accuracy:   3%|▎         | 1/32 [00:06<03:25,  6.62s/it]
[WRONG] Example 1 (diff=8)
  Expected : 'Guy'
  Extracted: 'Irma'
  Generated: '### Irma'

Test accuracy: 0.5:   3%|▎         | 1/32 [00:06<03:25,  6.62s/it]
[WRONG] Example 2 (diff=10)
  Expected : 'Elvis'
  Extracted: 'Shannon'
  Generated: '### Shannon'

[WRONG] Example 3 (diff=8)
  Expected : 'Coral'
  Extracted: 'Claude'
  Generated: '### Claude'

[WRONG] Example 4 (diff=1)
  Expected : 'Patricia'
  Extracted: 'Evelyn'
  Generated: '### Evelyn'

Test Accuracy:   3%|▎         | 1/32 [00:06<03:20,  6.48s/it]
Test accuracy: 0.12:   3%|▎         | 1/32 [00:06<03:20,  6.48s/it]
Test accuracy: 0.62:   6%|▋         | 2/32 [00:13<03:16,  6.55s/it]
Test accuracy: 0.5:   6%|▋         | 2/32 [00:13<03:16,  6.55s/it]
Test accuracy: 0.5:   6%|▋         | 2/32 [00:13<03:16,  6.55s/it]
Test accuracy: 0.75:   6%|▋         | 2/32 [00:13<03:16,  6.55s/it]
Test accuracy: 0.69:   6%|▋         | 2/32 [00:13<03:16,  6.55s/it]
Test accuracy: 0.44:   6%|▋         | 2/32 [00:13<03:16,  6.55s/it]
Test accuracy: 0.12:   6%|▋         | 2/32 [00:12<03:14,  6.49s/it]
Test accuracy: 0.19:   6%|▋         | 2/32 [00:12<03:14,  6.49s/it]
Test accuracy: 0.75:   9%|▉         | 3/32 [00:19<03:12,  6.65s/it]
Test accuracy: 0.75:   9%|▉         | 3/32 [00:19<03:12,  6.65s/it]
Test accuracy: 0.69:   9%|▉         | 3/32 [00:19<03:12,  6.65s/it]
Test accuracy: 0.71:   9%|▉         | 3/32 [00:19<03:12,  6.65s/it]
Test accuracy: 0.19:   9%|▉         | 3/32 [00:19<03:11,  6.61s/it]
Test accuracy: 0.29:   9%|▉         | 3/32 [00:19<03:11,  6.61s/it]
Test accuracy: 0.44:   9%|▉         | 3/32 [00:19<03:13,  6.66s/it]
Test accuracy: 0.5:   9%|▉         | 3/32 [00:19<03:13,  6.66s/it] 
Test accuracy: 0.5:  12%|█▎        | 4/32 [00:26<03:05,  6.62s/it]
Test accuracy: 0.5:  12%|█▎        | 4/32 [00:26<03:05,  6.62s/it]
Test accuracy: 0.29:  12%|█▎        | 4/32 [00:26<03:04,  6.60s/it]
Test accuracy: 0.75:  12%|█▎        | 4/32 [00:26<03:05,  6.62s/it]
Test accuracy: 0.38:  12%|█▎        | 4/32 [00:26<03:04,  6.60s/it]
Test accuracy: 0.72:  12%|█▎        | 4/32 [00:26<03:05,  6.62s/it]
Test accuracy: 0.71:  12%|█▎        | 4/32 [00:26<03:05,  6.62s/it]
Test accuracy: 0.69:  12%|█▎        | 4/32 [00:26<03:05,  6.62s/it]
Test accuracy: 0.38:  16%|█▌        | 5/32 [00:32<02:54,  6.46s/it]
Test accuracy: 0.4:  16%|█▌        | 5/32 [00:32<02:54,  6.46s/it] 
Test accuracy: 0.69:  16%|█▌        | 5/32 [00:32<02:54,  6.47s/it]
Test accuracy: 0.5:  16%|█▌        | 5/32 [00:32<02:54,  6.47s/it]
Test accuracy: 0.72:  16%|█▌        | 5/32 [00:32<02:54,  6.47s/it]
Test accuracy: 0.68:  16%|█▌        | 5/32 [00:32<02:54,  6.47s/it]
Test accuracy: 0.57:  16%|█▌        | 5/32 [00:32<02:54,  6.47s/it]
Test accuracy: 0.7:  16%|█▌        | 5/32 [00:32<02:54,  6.47s/it] 
Test accuracy: 0.68:  19%|█▉        | 6/32 [00:38<02:45,  6.38s/it]
Test accuracy: 0.65:  19%|█▉        | 6/32 [00:38<02:45,  6.38s/it]
Test accuracy: 0.57:  19%|█▉        | 6/32 [00:38<02:45,  6.38s/it]
Test accuracy: 0.6:  19%|█▉        | 6/32 [00:38<02:45,  6.38s/it] 
Test accuracy: 0.7:  19%|█▉        | 6/32 [00:38<02:45,  6.38s/it]
Test accuracy: 0.67:  19%|█▉        | 6/32 [00:38<02:45,  6.38s/it]
Test accuracy: 0.4:  19%|█▉        | 6/32 [00:38<02:45,  6.37s/it]
Test accuracy: 0.46:  19%|█▉        | 6/32 [00:38<02:45,  6.37s/it]
Test accuracy: 0.67:  22%|██▏       | 7/32 [00:45<02:41,  6.44s/it]
Test accuracy: 0.66:  22%|██▏       | 7/32 [00:45<02:41,  6.44s/it]
Test accuracy: 0.6:  22%|██▏       | 7/32 [00:45<02:40,  6.44s/it]
Test accuracy: 0.61:  22%|██▏       | 7/32 [00:45<02:40,  6.44s/it]
Test accuracy: 0.46:  22%|██▏       | 7/32 [00:45<02:40,  6.43s/it]
Test accuracy: 0.52:  22%|██▏       | 7/32 [00:45<02:40,  6.43s/it]
Test accuracy: 0.65:  22%|██▏       | 7/32 [00:45<02:41,  6.44s/it]
Test accuracy: 0.64:  22%|██▏       | 7/32 [00:45<02:41,  6.44s/it]
Test accuracy: 0.64:  25%|██▌       | 8/32 [00:51<02:32,  6.36s/it]
Test accuracy: 0.66:  25%|██▌       | 8/32 [00:51<02:32,  6.36s/it]
Test accuracy: 0.52:  25%|██▌       | 8/32 [00:51<02:32,  6.36s/it]
Test accuracy: 0.61:  25%|██▌       | 8/32 [00:51<02:32,  6.36s/it]
Test accuracy: 0.53:  25%|██▌       | 8/32 [00:51<02:32,  6.36s/it]
Test accuracy: 0.62:  25%|██▌       | 8/32 [00:51<02:32,  6.36s/it]
Test accuracy: 0.66:  25%|██▌       | 8/32 [00:51<02:32,  6.36s/it]
Test accuracy: 0.62:  25%|██▌       | 8/32 [00:51<02:32,  6.36s/it]
Test accuracy: 0.62:  28%|██▊       | 9/32 [00:58<02:29,  6.49s/it]
Test accuracy: 0.66:  28%|██▊       | 9/32 [00:58<02:29,  6.49s/it]
Test accuracy: 0.62:  28%|██▊       | 9/32 [00:58<02:29,  6.49s/it]
Test accuracy: 0.64:  28%|██▊       | 9/32 [00:58<02:29,  6.49s/it]
Test accuracy: 0.53:  28%|██▊       | 9/32 [00:58<02:29,  6.49s/it]
Test accuracy: 0.56:  28%|██▊       | 9/32 [00:58<02:29,  6.49s/it]
Test accuracy: 0.62:  28%|██▊       | 9/32 [00:58<02:29,  6.50s/it]
Test accuracy: 0.62:  28%|██▊       | 9/32 [00:58<02:29,  6.50s/it]
Test accuracy: 0.62:  31%|███▏      | 10/32 [01:05<02:23,  6.52s/it]
Test accuracy: 0.64:  31%|███▏      | 10/32 [01:04<02:23,  6.52s/it]
Test accuracy: 0.62:  31%|███▏      | 10/32 [01:05<02:23,  6.52s/it]
Test accuracy: 0.59:  31%|███▏      | 10/32 [01:05<02:23,  6.52s/it]
Test accuracy: 0.62:  31%|███▏      | 10/32 [01:04<02:23,  6.52s/it]
Test accuracy: 0.61:  31%|███▏      | 10/32 [01:05<02:23,  6.52s/it]
Test accuracy: 0.56:  31%|███▏      | 10/32 [01:04<02:23,  6.52s/it]
Test accuracy: 0.52:  31%|███▏      | 10/32 [01:04<02:23,  6.52s/it]
Test accuracy: 0.59:  34%|███▍      | 11/32 [01:11<02:14,  6.42s/it]
Test accuracy: 0.61:  34%|███▍      | 11/32 [01:11<02:14,  6.42s/it]
Test accuracy: 0.58:  34%|███▍      | 11/32 [01:11<02:14,  6.42s/it]
Test accuracy: 0.6:  34%|███▍      | 11/32 [01:11<02:14,  6.42s/it] 
Test accuracy: 0.52:  34%|███▍      | 11/32 [01:11<02:14,  6.42s/it]
Test accuracy: 0.52:  34%|███▍      | 11/32 [01:11<02:14,  6.42s/it]
Test accuracy: 0.62:  34%|███▍      | 11/32 [01:11<02:14,  6.42s/it]
Test accuracy: 0.6:  34%|███▍      | 11/32 [01:11<02:14,  6.42s/it] 
Test accuracy: 0.58:  38%|███▊      | 12/32 [01:17<02:08,  6.42s/it]
Test accuracy: 0.54:  38%|███▊      | 12/32 [01:17<02:08,  6.42s/it]
Test accuracy: 0.52:  38%|███▊      | 12/32 [01:17<02:08,  6.42s/it]
Test accuracy: 0.6:  38%|███▊      | 12/32 [01:17<02:08,  6.42s/it]
Test accuracy: 0.53:  38%|███▊      | 12/32 [01:17<02:08,  6.42s/it]
Test accuracy: 0.6:  38%|███▊      | 12/32 [01:17<02:08,  6.42s/it]
Test accuracy: 0.6:  38%|███▊      | 12/32 [01:17<02:08,  6.42s/it]
Test accuracy: 0.55:  38%|███▊      | 12/32 [01:17<02:08,  6.42s/it]
Test accuracy: 0.53:  41%|████      | 13/32 [01:23<02:01,  6.40s/it]
Test accuracy: 0.52:  41%|████      | 13/32 [01:23<02:01,  6.40s/it]
Test accuracy: 0.54:  41%|████      | 13/32 [01:23<02:01,  6.40s/it]
Test accuracy: 0.55:  41%|████      | 13/32 [01:23<02:01,  6.40s/it]
Test accuracy: 0.54:  41%|████      | 13/32 [01:23<02:01,  6.40s/it]
Test accuracy: 0.6:  41%|████      | 13/32 [01:23<02:01,  6.40s/it]
Test accuracy: 0.54:  41%|████      | 13/32 [01:23<02:01,  6.40s/it]
Test accuracy: 0.61:  41%|████      | 13/32 [01:23<02:01,  6.40s/it]
Test accuracy: 0.61:  44%|████▍     | 14/32 [01:30<01:54,  6.34s/it]
Test accuracy: 0.52:  44%|████▍     | 14/32 [01:30<01:54,  6.34s/it]
Test accuracy: 0.59:  44%|████▍     | 14/32 [01:30<01:54,  6.34s/it]
Test accuracy: 0.53:  44%|████▍     | 14/32 [01:30<01:54,  6.34s/it]
Test accuracy: 0.54:  44%|████▍     | 14/32 [01:30<01:54,  6.34s/it]
Test accuracy: 0.54:  44%|████▍     | 14/32 [01:30<01:54,  6.34s/it]
Test accuracy: 0.54:  44%|████▍     | 14/32 [01:30<01:54,  6.34s/it]
Test accuracy: 0.54:  44%|████▍     | 14/32 [01:30<01:54,  6.34s/it]
Test accuracy: 0.54:  47%|████▋     | 15/32 [01:36<01:48,  6.36s/it]
Test accuracy: 0.54:  47%|████▋     | 15/32 [01:36<01:48,  6.36s/it]
Test accuracy: 0.55:  47%|████▋     | 15/32 [01:36<01:48,  6.36s/it]
Test accuracy: 0.53:  47%|████▋     | 15/32 [01:36<01:48,  6.36s/it]
Test accuracy: 0.59:  47%|████▋     | 15/32 [01:36<01:48,  6.36s/it]
Test accuracy: 0.58:  47%|████▋     | 15/32 [01:36<01:48,  6.36s/it]
Test accuracy: 0.53:  47%|████▋     | 15/32 [01:36<01:48,  6.36s/it]
Test accuracy: 0.51:  47%|████▋     | 15/32 [01:36<01:48,  6.36s/it]
Test accuracy: 0.55:  50%|█████     | 16/32 [01:42<01:41,  6.37s/it]
Test accuracy: 0.51:  50%|█████     | 16/32 [01:42<01:41,  6.37s/it]
Test accuracy: 0.58:  50%|█████     | 16/32 [01:42<01:41,  6.37s/it]
Test accuracy: 0.54:  50%|█████     | 16/32 [01:42<01:41,  6.37s/it]
Test accuracy: 0.56:  50%|█████     | 16/32 [01:42<01:41,  6.37s/it]
Test accuracy: 0.49:  50%|█████     | 16/32 [01:42<01:41,  6.37s/it]
Test accuracy: 0.53:  50%|█████     | 16/32 [01:42<01:41,  6.37s/it]
Test accuracy: 0.55:  50%|█████     | 16/32 [01:42<01:41,  6.37s/it]
Test accuracy: 0.55:  53%|█████▎    | 17/32 [01:49<01:34,  6.32s/it]
Test accuracy: 0.54:  53%|█████▎    | 17/32 [01:49<01:34,  6.32s/it]
Test accuracy: 0.56:  53%|█████▎    | 17/32 [01:49<01:34,  6.32s/it]
Test accuracy: 0.54:  53%|█████▎    | 17/32 [01:49<01:34,  6.32s/it]
Test accuracy: 0.56:  53%|█████▎    | 17/32 [01:49<01:34,  6.32s/it]
Test accuracy: 0.56:  53%|█████▎    | 17/32 [01:49<01:34,  6.32s/it]
Test accuracy: 0.49:  53%|█████▎    | 17/32 [01:49<01:34,  6.32s/it]
Test accuracy: 0.51:  53%|█████▎    | 17/32 [01:49<01:34,  6.32s/it]
Test accuracy: 0.56:  56%|█████▋    | 18/32 [01:55<01:28,  6.34s/it]
Test accuracy: 0.58:  56%|█████▋    | 18/32 [01:55<01:28,  6.34s/it]
Test accuracy: 0.54:  56%|█████▋    | 18/32 [01:55<01:28,  6.34s/it]
Test accuracy: 0.56:  56%|█████▋    | 18/32 [01:55<01:28,  6.34s/it]
Test accuracy: 0.53:  56%|█████▋    | 18/32 [01:55<01:28,  6.34s/it]
Test accuracy: 0.57:  56%|█████▋    | 18/32 [01:55<01:28,  6.34s/it]
Test accuracy: 0.51:  56%|█████▋    | 18/32 [01:55<01:28,  6.34s/it]
Test accuracy: 0.51:  56%|█████▋    | 18/32 [01:55<01:28,  6.34s/it]
Test accuracy: 0.58:  59%|█████▉    | 19/32 [02:01<01:22,  6.35s/it]
Test accuracy: 0.57:  59%|█████▉    | 19/32 [02:01<01:22,  6.35s/it]
Test accuracy: 0.51:  59%|█████▉    | 19/32 [02:01<01:22,  6.35s/it]
Test accuracy: 0.51:  59%|█████▉    | 19/32 [02:01<01:22,  6.35s/it]
Test accuracy: 0.57:  59%|█████▉    | 19/32 [02:01<01:22,  6.35s/it]
Test accuracy: 0.53:  59%|█████▉    | 19/32 [02:01<01:22,  6.35s/it]
Test accuracy: 0.57:  59%|█████▉    | 19/32 [02:01<01:22,  6.35s/it]
Test accuracy: 0.52:  59%|█████▉    | 19/32 [02:01<01:22,  6.35s/it]
Test accuracy: 0.57:  62%|██████▎   | 20/32 [02:08<01:15,  6.30s/it]
Test accuracy: 0.56:  62%|██████▎   | 20/32 [02:08<01:15,  6.30s/it]
Test accuracy: 0.57:  62%|██████▎   | 20/32 [02:08<01:15,  6.30s/it]
Test accuracy: 0.55:  62%|██████▎   | 20/32 [02:08<01:15,  6.30s/it]
Test accuracy: 0.52:  62%|██████▎   | 20/32 [02:08<01:15,  6.30s/it]
Test accuracy: 0.52:  62%|██████▎   | 20/32 [02:08<01:15,  6.30s/it]
Test accuracy: 0.51:  62%|██████▎   | 20/32 [02:07<01:15,  6.30s/it]
Test accuracy: 0.52:  62%|██████▎   | 20/32 [02:07<01:15,  6.30s/it]
Test accuracy: 0.52:  66%|██████▌   | 21/32 [02:15<01:11,  6.51s/it]
Test accuracy: 0.56:  66%|██████▌   | 21/32 [02:15<01:11,  6.51s/it]
Test accuracy: 0.52:  66%|██████▌   | 21/32 [02:15<01:11,  6.51s/it]
Test accuracy: 0.55:  66%|██████▌   | 21/32 [02:15<01:11,  6.51s/it]
Test accuracy: 0.52:  66%|██████▌   | 21/32 [02:14<01:11,  6.51s/it]
Test accuracy: 0.51:  66%|██████▌   | 21/32 [02:14<01:11,  6.51s/it]
Test accuracy: 0.55:  66%|██████▌   | 21/32 [02:15<01:11,  6.51s/it]
Test accuracy: 0.55:  66%|██████▌   | 21/32 [02:15<01:11,  6.51s/it]
Test accuracy: 0.55:  69%|██████▉   | 22/32 [02:21<01:04,  6.48s/it]
Test accuracy: 0.52:  69%|██████▉   | 22/32 [02:21<01:04,  6.48s/it]
Test accuracy: 0.55:  69%|██████▉   | 22/32 [02:21<01:04,  6.48s/it]
Test accuracy: 0.54:  69%|██████▉   | 22/32 [02:21<01:04,  6.48s/it]
Test accuracy: 0.52:  69%|██████▉   | 22/32 [02:21<01:04,  6.48s/it]
Test accuracy: 0.55:  69%|██████▉   | 22/32 [02:21<01:04,  6.48s/it]
Test accuracy: 0.51:  69%|██████▉   | 22/32 [02:21<01:04,  6.48s/it]
Test accuracy: 0.51:  69%|██████▉   | 22/32 [02:21<01:04,  6.48s/it]
Test accuracy: 0.52:  72%|███████▏  | 23/32 [02:27<00:58,  6.46s/it]
Test accuracy: 0.55:  72%|███████▏  | 23/32 [02:27<00:58,  6.45s/it]
Test accuracy: 0.52:  72%|███████▏  | 23/32 [02:27<00:58,  6.46s/it]
Test accuracy: 0.54:  72%|███████▏  | 23/32 [02:27<00:58,  6.45s/it]
Test accuracy: 0.54:  72%|███████▏  | 23/32 [02:27<00:58,  6.46s/it]
Test accuracy: 0.55:  72%|███████▏  | 23/32 [02:27<00:58,  6.46s/it]
Test accuracy: 0.51:  72%|███████▏  | 23/32 [02:27<00:58,  6.46s/it]
Test accuracy: 0.52:  72%|███████▏  | 23/32 [02:27<00:58,  6.46s/it]
Test accuracy: 0.54:  75%|███████▌  | 24/32 [02:34<00:52,  6.56s/it]
Test accuracy: 0.55:  75%|███████▌  | 24/32 [02:34<00:52,  6.56s/it]
Test accuracy: 0.52:  75%|███████▌  | 24/32 [02:34<00:52,  6.56s/it]
Test accuracy: 0.54:  75%|███████▌  | 24/32 [02:34<00:52,  6.56s/it]
Test accuracy: 0.55:  75%|███████▌  | 24/32 [02:34<00:52,  6.56s/it]
Test accuracy: 0.52:  75%|███████▌  | 24/32 [02:34<00:52,  6.56s/it]
Test accuracy: 0.52:  75%|███████▌  | 24/32 [02:34<00:52,  6.56s/it]
Test accuracy: 0.53:  75%|███████▌  | 24/32 [02:34<00:52,  6.56s/it]
Test accuracy: 0.52:  78%|███████▊  | 25/32 [02:41<00:45,  6.51s/it]
Test accuracy: 0.51:  78%|███████▊  | 25/32 [02:41<00:45,  6.51s/it]
Test accuracy: 0.53:  78%|███████▊  | 25/32 [02:40<00:45,  6.51s/it]
Test accuracy: 0.54:  78%|███████▊  | 25/32 [02:40<00:45,  6.51s/it]
Test accuracy: 0.55:  78%|███████▊  | 25/32 [02:41<00:45,  6.51s/it]
Test accuracy: 0.54:  78%|███████▊  | 25/32 [02:41<00:45,  6.51s/it]
Test accuracy: 0.54:  78%|███████▊  | 25/32 [02:41<00:45,  6.51s/it]
Test accuracy: 0.55:  78%|███████▊  | 25/32 [02:41<00:45,  6.51s/it]
Test accuracy: 0.51:  81%|████████▏ | 26/32 [02:47<00:38,  6.48s/it]
Test accuracy: 0.54:  81%|████████▏ | 26/32 [02:47<00:38,  6.48s/it]
Test accuracy: 0.5:  81%|████████▏ | 26/32 [02:47<00:38,  6.48s/it] 
Test accuracy: 0.55:  81%|████████▏ | 26/32 [02:47<00:38,  6.48s/it]
Test accuracy: 0.54:  81%|████████▏ | 26/32 [02:47<00:38,  6.48s/it]
Test accuracy: 0.54:  81%|████████▏ | 26/32 [02:47<00:38,  6.48s/it]
Test accuracy: 0.54:  81%|████████▏ | 26/32 [02:47<00:38,  6.47s/it]
Test accuracy: 0.55:  81%|████████▏ | 26/32 [02:47<00:38,  6.47s/it]
Test accuracy: 0.5:  84%|████████▍ | 27/32 [02:53<00:31,  6.39s/it]
Test accuracy: 0.5:  84%|████████▍ | 27/32 [02:53<00:31,  6.39s/it]
Test accuracy: 0.55:  84%|████████▍ | 27/32 [02:53<00:31,  6.39s/it]
Test accuracy: 0.56:  84%|████████▍ | 27/32 [02:53<00:31,  6.39s/it]
Test accuracy: 0.54:  84%|████████▍ | 27/32 [02:53<00:31,  6.40s/it]
Test accuracy: 0.54:  84%|████████▍ | 27/32 [02:53<00:31,  6.40s/it]
Test accuracy: 0.54:  84%|████████▍ | 27/32 [02:53<00:31,  6.40s/it]
Test accuracy: 0.54:  84%|████████▍ | 27/32 [02:53<00:31,  6.40s/it]
Test accuracy: 0.5:  88%|████████▊ | 28/32 [03:00<00:26,  6.57s/it]
Test accuracy: 0.54:  88%|████████▊ | 28/32 [03:00<00:26,  6.57s/it]
Test accuracy: 0.56:  88%|████████▊ | 28/32 [03:00<00:26,  6.57s/it]
Test accuracy: 0.5:  88%|████████▊ | 28/32 [03:00<00:26,  6.57s/it]
Test accuracy: 0.54:  88%|████████▊ | 28/32 [03:00<00:26,  6.57s/it]
Test accuracy: 0.56:  88%|████████▊ | 28/32 [03:00<00:26,  6.57s/it]
Test accuracy: 0.54:  88%|████████▊ | 28/32 [03:00<00:26,  6.58s/it]
Test accuracy: 0.54:  88%|████████▊ | 28/32 [03:00<00:26,  6.58s/it]
Test accuracy: 0.5:  91%|█████████ | 29/32 [03:07<00:19,  6.53s/it]
Test accuracy: 0.5:  91%|█████████ | 29/32 [03:07<00:19,  6.53s/it]
Test accuracy: 0.54:  91%|█████████ | 29/32 [03:07<00:19,  6.52s/it]
Test accuracy: 0.55:  91%|█████████ | 29/32 [03:07<00:19,  6.52s/it]
Test accuracy: 0.56:  91%|█████████ | 29/32 [03:06<00:19,  6.53s/it]
Test accuracy: 0.56:  91%|█████████ | 29/32 [03:06<00:19,  6.53s/it]
Test accuracy: 0.54:  91%|█████████ | 29/32 [03:07<00:19,  6.53s/it]
Test accuracy: 0.54:  91%|█████████ | 29/32 [03:07<00:19,  6.53s/it]
Test accuracy: 0.54:  94%|█████████▍| 30/32 [03:13<00:12,  6.48s/it]
Test accuracy: 0.5:  94%|█████████▍| 30/32 [03:13<00:12,  6.48s/it]
Test accuracy: 0.54:  94%|█████████▍| 30/32 [03:13<00:12,  6.48s/it]
Test accuracy: 0.49:  94%|█████████▍| 30/32 [03:13<00:12,  6.48s/it]
Test accuracy: 0.55:  94%|█████████▍| 30/32 [03:13<00:12,  6.48s/it]
Test accuracy: 0.54:  94%|█████████▍| 30/32 [03:13<00:12,  6.48s/it]
Test accuracy: 0.56:  94%|█████████▍| 30/32 [03:13<00:12,  6.48s/it]
Test accuracy: 0.56:  94%|█████████▍| 30/32 [03:13<00:12,  6.48s/it]
Test accuracy: 0.54:  97%|█████████▋| 31/32 [03:20<00:06,  6.58s/it]
Test accuracy: 0.49:  97%|█████████▋| 31/32 [03:20<00:06,  6.58s/it]
Test accuracy: 0.54:  97%|█████████▋| 31/32 [03:20<00:06,  6.58s/it]
Test accuracy: 0.5:  97%|█████████▋| 31/32 [03:20<00:06,  6.58s/it] 
Test accuracy: 0.56:  97%|█████████▋| 31/32 [03:20<00:06,  6.58s/it]
Test accuracy: 0.56:  97%|█████████▋| 31/32 [03:20<00:06,  6.58s/it]
Test accuracy: 0.54:  97%|█████████▋| 31/32 [03:20<00:06,  6.59s/it]
Test accuracy: 0.55:  97%|█████████▋| 31/32 [03:20<00:06,  6.59s/it]
Test accuracy: 0.56: 100%|██████████| 32/32 [03:22<00:00,  5.38s/it]
Test accuracy: 0.57: 100%|██████████| 32/32 [03:22<00:00,  5.38s/it]
Test accuracy: 0.54: 100%|██████████| 32/32 [03:22<00:00,  5.38s/it]
Test accuracy: 0.54: 100%|██████████| 32/32 [03:22<00:00,  5.38s/it]
Test accuracy: 0.5: 100%|██████████| 32/32 [03:22<00:00,  5.38s/it]
Test accuracy: 0.5: 100%|██████████| 32/32 [03:22<00:00,  5.38s/it]
Test accuracy: 0.55: 100%|██████████| 32/32 [03:22<00:00,  5.38s/it]
Test accuracy: 0.55: 100%|██████████| 32/32 [03:22<00:00,  5.38s/it]
Test accuracy: 0.55: 100%|██████████| 32/32 [03:22<00:00,  6.34s/it]
Device 2: Cor=137, CoT=0, Total=250

Test accuracy: 0.5: 100%|██████████| 32/32 [03:22<00:00,  6.34s/it]
Device 3: Cor=125, CoT=0, Total=250

Test accuracy: 0.54: 100%|██████████| 32/32 [03:22<00:00,  6.34s/it]
Device 1: Cor=136, CoT=0, Total=250

Test accuracy: 0.57: 100%|██████████| 32/32 [03:22<00:00,  6.34s/it]
Device 0: Cor=142, CoT=0, Total=250

============================================================
EVALUATION RESULTS (Epoch 2)
============================================================
Config: data_curriculum=ON, thought_curriculum=OFF
thought_stage=10, data_stage=1, n_latent_tokens=10
Training on difficulties: 0-1

Overall accuracy: 540 / 1000 = 0.5400

Per-difficulty accuracy:
  diff=1: 85/100 = 0.8500 (trained)
  diff=2: 54/100 = 0.5400 (untrained)
  diff=3: 46/100 = 0.4600 (untrained)
  diff=4: 55/100 = 0.5500 (untrained)
  diff=5: 45/100 = 0.4500 (untrained)
  diff=6: 47/100 = 0.4700 (untrained)
  diff=7: 47/100 = 0.4700 (untrained)
  diff=8: 51/100 = 0.5100 (untrained)
  diff=9: 54/100 = 0.5400 (untrained)
  diff=10: 56/100 = 0.5600 (untrained)

Trained (diff 0-1): 85/100 = 0.8500
============================================================

  Staging check: acc_trained=0.8500, acc_diff1=0.8500 (threshold=0.9)
Warning: failed to upload checkpoint_3 to HF: (Request ID: Root=1-6a1b6751-0c6f91486a1fe1fa72cdb327;33f5d46b-d39e-4528-8be8-7857754870f7)

403 Forbidden: You need to setup automatic credit recharge in order to upload more data. You can do so at /settings/billing..
Cannot access content at: https://huggingface.co/seyedparsa/thoughtformer-checkpoints.git/info/lfs/objects/batch.
Make sure your token has the correct permissions.

============================================================
EPOCH 3/100 (thought_stage=10, data_stage=1)
============================================================
  thought_stage=10, c_thought=1, max_difficulty=1

Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):  15%|█▌        | 153/1000 [00:00<00:00, 988.66 examples/s]
Map (num_proc=4):  15%|█▌        | 153/1000 [00:00<00:00, 954.84 examples/s]
Map (num_proc=4):  16%|█▌        | 157/1000 [00:00<00:00, 984.30 examples/s]
Map (num_proc=4):  81%|████████  | 812/1000 [00:00<00:00, 3561.98 examples/s]
Map (num_proc=4):  82%|████████▏ | 821/1000 [00:00<00:00, 3616.93 examples/s]
Map (num_proc=4):  81%|████████▏ | 814/1000 [00:00<00:00, 3115.59 examples/s]
Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2574.60 examples/s]

Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2519.01 examples/s]

Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2545.81 examples/s]

Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Map (num_proc=4):  16%|█▌        | 159/1000 [00:00<00:01, 548.85 examples/s]
Map (num_proc=4):  81%|████████  | 812/1000 [00:00<00:00, 2458.93 examples/s]
Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:04, 2009.84 examples/s]
Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 1300.92 examples/s]

Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:04, 1987.80 examples/s]
Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:04, 1926.17 examples/s]
Filter (num_proc=4):  20%|██        | 2000/10000 [00:00<00:02, 3708.13 examples/s]
Filter (num_proc=4):  20%|██        | 2000/10000 [00:00<00:02, 3564.24 examples/s]
Filter (num_proc=4):  20%|██        | 2000/10000 [00:00<00:02, 3459.52 examples/s]
Filter (num_proc=4):  30%|███       | 3000/10000 [00:00<00:01, 5061.32 examples/s]
Filter (num_proc=4):  30%|███       | 3000/10000 [00:00<00:01, 4989.86 examples/s]
Filter (num_proc=4):  30%|███       | 3000/10000 [00:00<00:01, 4822.44 examples/s]
Filter (num_proc=4):  50%|█████     | 5000/10000 [00:00<00:00, 8531.62 examples/s]
Filter (num_proc=4):  45%|████▌     | 4500/10000 [00:00<00:00, 7158.48 examples/s]
Filter (num_proc=4):  45%|████▌     | 4500/10000 [00:00<00:00, 6897.74 examples/s]
Filter (num_proc=4):  75%|███████▌  | 7500/10000 [00:00<00:00, 11647.16 examples/s]
Filter (num_proc=4):  75%|███████▌  | 7500/10000 [00:00<00:00, 12108.08 examples/s]
Filter (num_proc=4):  90%|█████████ | 9000/10000 [00:01<00:00, 12368.88 examples/s]
Filter (num_proc=4):  70%|███████   | 7000/10000 [00:01<00:00, 10366.43 examples/s]
Filter (num_proc=4):  95%|█████████▌| 9500/10000 [00:01<00:00, 13240.67 examples/s]
Filter (num_proc=4):  85%|████████▌ | 8500/10000 [00:01<00:00, 10790.57 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 11279.59 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 7553.50 examples/s]

Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 7376.84 examples/s]

Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 7112.35 examples/s] 

Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:06, 1412.27 examples/s]
Filter (num_proc=4):  20%|██        | 2000/10000 [00:00<00:03, 2509.76 examples/s]
Filter (num_proc=4):  35%|███▌      | 3500/10000 [00:01<00:01, 4135.34 examples/s]
Filter (num_proc=4):  60%|██████    | 6000/10000 [00:01<00:00, 6969.61 examples/s]
Filter (num_proc=4):  80%|████████  | 8000/10000 [00:01<00:00, 9064.26 examples/s]
Filter (num_proc=4):  95%|█████████▌| 9500/10000 [00:01<00:00, 8568.51 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 5099.33 examples/s]

Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):  13%|█▎        | 134/1000 [00:01<00:07, 116.35 examples/s]
Map (num_proc=4):  25%|██▌       | 250/1000 [00:01<00:03, 227.89 examples/s]
Map (num_proc=4):  38%|███▊      | 385/1000 [00:01<00:02, 220.75 examples/s]
Map (num_proc=4):  50%|█████     | 500/1000 [00:02<00:01, 309.79 examples/s]
Map (num_proc=4):  63%|██████▎   | 634/1000 [00:02<00:01, 264.32 examples/s]
Map (num_proc=4):  75%|███████▌  | 750/1000 [00:02<00:00, 346.11 examples/s]
Map (num_proc=4):  89%|████████▉ | 888/1000 [00:03<00:00, 286.60 examples/s]
Map (num_proc=4): 100%|██████████| 1000/1000 [00:03<00:00, 361.25 examples/s]
Map (num_proc=4): 100%|██████████| 1000/1000 [00:03<00:00, 261.19 examples/s]

Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 1462.84 examples/s]
Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 1424.87 examples/s]
Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 1418.20 examples/s]
Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2772.43 examples/s]

Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2690.24 examples/s]

Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2789.23 examples/s]

Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 799.14 examples/s]
Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 1531.11 examples/s]

Map (num_proc=4):   0%|          | 0/100 [00:00<?, ? examples/s]
Map (num_proc=4):  25%|██▌       | 25/100 [00:00<00:00, 81.46 examples/s]
Map (num_proc=4):  75%|███████▌  | 75/100 [00:00<00:00, 191.21 examples/s]
Map (num_proc=4): 100%|██████████| 100/100 [00:00<00:00, 124.69 examples/s]

Training Epoch: 4:   0%|          | 0/31 [00:00<?, ?it/s]
Training Epoch: 4:   0%|          | 0/31 [00:00<?, ?it/s]
Training Epoch: 4:   0%|          | 0/31 [00:00<?, ?it/s]
Training Epoch: 4:   0%|          | 0/31 [00:00<?, ?it/s]logging training data
rank 2 | gpu 2 | alloc=4.20 GB | reserved=55.12 GB | max=41.98 GB
rank 3 | gpu 3 | alloc=4.20 GB | reserved=55.11 GB | max=41.99 GB
rank 1 | gpu 1 | alloc=4.20 GB | reserved=51.41 GB | max=41.98 GB
rank 0 | gpu 0 | alloc=4.20 GB | reserved=51.46 GB | max=41.99 GB

Training Epoch: 4/100, batch 0/63 completed (loss: 0.143:   0%|          | 0/31 [00:03<?, ?it/s]
Training Epoch: 4/100, batch 0/63 completed (loss: 0.1723:   0%|          | 0/31 [00:03<?, ?it/s]
Training Epoch: 4/100, batch 0/63 completed (loss: 0.1206:   0%|          | 0/31 [00:03<?, ?it/s]
Training Epoch: 4/100, batch 0/63 completed (loss: 0.1554:   0%|          | 0/31 [00:03<?, ?it/s]
Training Epoch: 4/100, batch 0/63 completed (loss: 0.1206:   3%|▎         | 1/31 [00:05<02:58,  5.96s/it]
Training Epoch: 4/100, batch 0/63 completed (loss: 0.1554:   3%|▎         | 1/31 [00:05<02:59,  5.97s/it]
Training Epoch: 4/100, batch 0/63 completed (loss: 0.1723:   3%|▎         | 1/31 [00:05<02:58,  5.96s/it]
Training Epoch: 4/100, batch 0/63 completed (loss: 0.143:   3%|▎         | 1/31 [00:05<02:58,  5.97s/it]
Training Epoch: 4/100, batch 1/63 completed (loss: 0.0701:   3%|▎         | 1/31 [00:06<02:58,  5.96s/it]
Training Epoch: 4/100, batch 1/63 completed (loss: 0.1481:   3%|▎         | 1/31 [00:06<02:58,  5.96s/it]
Training Epoch: 4/100, batch 1/63 completed (loss: 0.139:   3%|▎         | 1/31 [00:06<02:58,  5.97s/it]
Training Epoch: 4/100, batch 1/63 completed (loss: 0.0798:   3%|▎         | 1/31 [00:06<02:59,  5.97s/it]
Training Epoch: 4/100, batch 2/63 completed (loss: 0.1176:   3%|▎         | 1/31 [00:09<02:58,  5.96s/it]
Training Epoch: 4/100, batch 2/63 completed (loss: 0.1339:   3%|▎         | 1/31 [00:09<02:58,  5.97s/it]
Training Epoch: 4/100, batch 2/63 completed (loss: 0.1166:   3%|▎         | 1/31 [00:09<02:58,  5.96s/it]
Training Epoch: 4/100, batch 2/63 completed (loss: 0.0829:   3%|▎         | 1/31 [00:09<02:59,  5.97s/it]
Training Epoch: 4/100, batch 2/63 completed (loss: 0.1176:   6%|▋         | 2/31 [00:11<02:50,  5.88s/it]
Training Epoch: 4/100, batch 2/63 completed (loss: 0.1166:   6%|▋         | 2/31 [00:11<02:50,  5.88s/it]
Training Epoch: 4/100, batch 2/63 completed (loss: 0.0829:   6%|▋         | 2/31 [00:11<02:50,  5.89s/it]
Training Epoch: 4/100, batch 2/63 completed (loss: 0.1339:   6%|▋         | 2/31 [00:11<02:50,  5.89s/it]
Training Epoch: 4/100, batch 3/63 completed (loss: 0.1598:   6%|▋         | 2/31 [00:11<02:50,  5.89s/it]
Training Epoch: 4/100, batch 3/63 completed (loss: 0.1846:   6%|▋         | 2/31 [00:11<02:50,  5.88s/it]
Training Epoch: 4/100, batch 3/63 completed (loss: 0.1265:   6%|▋         | 2/31 [00:11<02:50,  5.88s/it]
Training Epoch: 4/100, batch 3/63 completed (loss: 0.1976:   6%|▋         | 2/31 [00:11<02:50,  5.89s/it]
Training Epoch: 4/100, batch 4/63 completed (loss: 0.1348:   6%|▋         | 2/31 [00:15<02:50,  5.88s/it]
Training Epoch: 4/100, batch 4/63 completed (loss: 0.1693:   6%|▋         | 2/31 [00:15<02:50,  5.88s/it]
Training Epoch: 4/100, batch 4/63 completed (loss: 0.1345:   6%|▋         | 2/31 [00:15<02:50,  5.89s/it]
Training Epoch: 4/100, batch 4/63 completed (loss: 0.1921:   6%|▋         | 2/31 [00:15<02:50,  5.89s/it]
Training Epoch: 4/100, batch 4/63 completed (loss: 0.1921:  10%|▉         | 3/31 [00:18<02:49,  6.04s/it]
Training Epoch: 4/100, batch 4/63 completed (loss: 0.1348:  10%|▉         | 3/31 [00:18<02:49,  6.04s/it]
Training Epoch: 4/100, batch 4/63 completed (loss: 0.1693:  10%|▉         | 3/31 [00:18<02:49,  6.05s/it]
Training Epoch: 4/100, batch 4/63 completed (loss: 0.1345:  10%|▉         | 3/31 [00:18<02:49,  6.05s/it]
Training Epoch: 4/100, batch 5/63 completed (loss: 0.096:  10%|▉         | 3/31 [00:18<02:49,  6.04s/it] 
Training Epoch: 4/100, batch 5/63 completed (loss: 0.111:  10%|▉         | 3/31 [00:18<02:49,  6.05s/it] 
Training Epoch: 4/100, batch 5/63 completed (loss: 0.1018:  10%|▉         | 3/31 [00:18<02:49,  6.05s/it]
Training Epoch: 4/100, batch 5/63 completed (loss: 0.1222:  10%|▉         | 3/31 [00:18<02:49,  6.04s/it]
Training Epoch: 4/100, batch 6/63 completed (loss: 0.147:  10%|▉         | 3/31 [00:20<02:49,  6.04s/it]
Training Epoch: 4/100, batch 6/63 completed (loss: 0.1361:  10%|▉         | 3/31 [00:20<02:49,  6.05s/it]
Training Epoch: 4/100, batch 6/63 completed (loss: 0.0958:  10%|▉         | 3/31 [00:20<02:49,  6.05s/it]
Training Epoch: 4/100, batch 6/63 completed (loss: 0.1318:  10%|▉         | 3/31 [00:20<02:49,  6.04s/it]
Training Epoch: 4/100, batch 6/63 completed (loss: 0.147:  13%|█▎        | 4/31 [00:23<02:38,  5.86s/it]
Training Epoch: 4/100, batch 6/63 completed (loss: 0.1318:  13%|█▎        | 4/31 [00:23<02:38,  5.86s/it]
Training Epoch: 4/100, batch 6/63 completed (loss: 0.0958:  13%|█▎        | 4/31 [00:23<02:38,  5.86s/it]
Training Epoch: 4/100, batch 6/63 completed (loss: 0.1361:  13%|█▎        | 4/31 [00:23<02:38,  5.86s/it]
Training Epoch: 4/100, batch 7/63 completed (loss: 0.079:  13%|█▎        | 4/31 [00:23<02:38,  5.86s/it]
Training Epoch: 4/100, batch 7/63 completed (loss: 0.17:  13%|█▎        | 4/31 [00:23<02:38,  5.86s/it]  
Training Epoch: 4/100, batch 7/63 completed (loss: 0.074:  13%|█▎        | 4/31 [00:23<02:38,  5.86s/it] 
Training Epoch: 4/100, batch 7/63 completed (loss: 0.1137:  13%|█▎        | 4/31 [00:23<02:38,  5.86s/it]
Training Epoch: 4/100, batch 8/63 completed (loss: 0.1272:  13%|█▎        | 4/31 [00:26<02:38,  5.86s/it]
Training Epoch: 4/100, batch 8/63 completed (loss: 0.1394:  13%|█▎        | 4/31 [00:26<02:38,  5.86s/it]
Training Epoch: 4/100, batch 8/63 completed (loss: 0.2312:  13%|█▎        | 4/31 [00:26<02:38,  5.86s/it]
Training Epoch: 4/100, batch 8/63 completed (loss: 0.0935:  13%|█▎        | 4/31 [00:26<02:38,  5.86s/it]
Training Epoch: 4/100, batch 8/63 completed (loss: 0.1272:  16%|█▌        | 5/31 [00:29<02:33,  5.89s/it]
Training Epoch: 4/100, batch 8/63 completed (loss: 0.2312:  16%|█▌        | 5/31 [00:29<02:33,  5.89s/it]
Training Epoch: 4/100, batch 8/63 completed (loss: 0.0935:  16%|█▌        | 5/31 [00:29<02:33,  5.89s/it]
Training Epoch: 4/100, batch 8/63 completed (loss: 0.1394:  16%|█▌        | 5/31 [00:29<02:33,  5.89s/it]
Training Epoch: 4/100, batch 9/63 completed (loss: 0.1269:  16%|█▌        | 5/31 [00:29<02:33,  5.89s/it]
Training Epoch: 4/100, batch 9/63 completed (loss: 0.0701:  16%|█▌        | 5/31 [00:29<02:33,  5.89s/it]
Training Epoch: 4/100, batch 9/63 completed (loss: 0.0763:  16%|█▌        | 5/31 [00:29<02:33,  5.89s/it]
Training Epoch: 4/100, batch 9/63 completed (loss: 0.1593:  16%|█▌        | 5/31 [00:29<02:33,  5.89s/it]
Training Epoch: 4/100, batch 10/63 completed (loss: 0.097:  16%|█▌        | 5/31 [00:32<02:33,  5.89s/it]
Training Epoch: 4/100, batch 10/63 completed (loss: 0.1185:  16%|█▌        | 5/31 [00:32<02:33,  5.89s/it]
Training Epoch: 4/100, batch 10/63 completed (loss: 0.2214:  16%|█▌        | 5/31 [00:32<02:33,  5.89s/it]
Training Epoch: 4/100, batch 10/63 completed (loss: 0.1848:  16%|█▌        | 5/31 [00:32<02:33,  5.89s/it]
Training Epoch: 4/100, batch 10/63 completed (loss: 0.1185:  19%|█▉        | 6/31 [00:35<02:26,  5.85s/it]
Training Epoch: 4/100, batch 10/63 completed (loss: 0.097:  19%|█▉        | 6/31 [00:35<02:26,  5.85s/it]
Training Epoch: 4/100, batch 10/63 completed (loss: 0.1848:  19%|█▉        | 6/31 [00:35<02:26,  5.85s/it]
Training Epoch: 4/100, batch 10/63 completed (loss: 0.2214:  19%|█▉        | 6/31 [00:35<02:26,  5.85s/it]
Training Epoch: 4/100, batch 11/63 completed (loss: 0.0631:  19%|█▉        | 6/31 [00:35<02:26,  5.85s/it]
Training Epoch: 4/100, batch 11/63 completed (loss: 0.1305:  19%|█▉        | 6/31 [00:35<02:26,  5.85s/it]
Training Epoch: 4/100, batch 11/63 completed (loss: 0.0943:  19%|█▉        | 6/31 [00:35<02:26,  5.85s/it]
Training Epoch: 4/100, batch 11/63 completed (loss: 0.1078:  19%|█▉        | 6/31 [00:35<02:26,  5.85s/it]
Training Epoch: 4/100, batch 12/63 completed (loss: 0.0951:  19%|█▉        | 6/31 [00:38<02:26,  5.85s/it]
Training Epoch: 4/100, batch 12/63 completed (loss: 0.1374:  19%|█▉        | 6/31 [00:38<02:26,  5.85s/it]
Training Epoch: 4/100, batch 12/63 completed (loss: 0.0757:  19%|█▉        | 6/31 [00:38<02:26,  5.85s/it]
Training Epoch: 4/100, batch 12/63 completed (loss: 0.1683:  19%|█▉        | 6/31 [00:38<02:26,  5.85s/it]
Training Epoch: 4/100, batch 12/63 completed (loss: 0.1683:  23%|██▎       | 7/31 [00:41<02:19,  5.83s/it]
Training Epoch: 4/100, batch 12/63 completed (loss: 0.0757:  23%|██▎       | 7/31 [00:41<02:19,  5.83s/it]
Training Epoch: 4/100, batch 12/63 completed (loss: 0.0951:  23%|██▎       | 7/31 [00:41<02:19,  5.83s/it]
Training Epoch: 4/100, batch 12/63 completed (loss: 0.1374:  23%|██▎       | 7/31 [00:41<02:19,  5.83s/it]
Training Epoch: 4/100, batch 13/63 completed (loss: 0.1471:  23%|██▎       | 7/31 [00:41<02:19,  5.83s/it]
Training Epoch: 4/100, batch 13/63 completed (loss: 0.2048:  23%|██▎       | 7/31 [00:41<02:19,  5.83s/it]
Training Epoch: 4/100, batch 13/63 completed (loss: 0.186:  23%|██▎       | 7/31 [00:41<02:19,  5.83s/it] 
Training Epoch: 4/100, batch 13/63 completed (loss: 0.1942:  23%|██▎       | 7/31 [00:41<02:19,  5.83s/it]
Training Epoch: 4/100, batch 14/63 completed (loss: 0.1304:  23%|██▎       | 7/31 [00:43<02:19,  5.83s/it]
Training Epoch: 4/100, batch 14/63 completed (loss: 0.1226:  23%|██▎       | 7/31 [00:43<02:19,  5.83s/it]
Training Epoch: 4/100, batch 14/63 completed (loss: 0.1773:  23%|██▎       | 7/31 [00:43<02:19,  5.83s/it]
Training Epoch: 4/100, batch 14/63 completed (loss: 0.1741:  23%|██▎       | 7/31 [00:43<02:19,  5.83s/it]
Training Epoch: 4/100, batch 14/63 completed (loss: 0.1741:  26%|██▌       | 8/31 [00:47<02:15,  5.88s/it]
Training Epoch: 4/100, batch 14/63 completed (loss: 0.1304:  26%|██▌       | 8/31 [00:47<02:15,  5.88s/it]
Training Epoch: 4/100, batch 14/63 completed (loss: 0.1226:  26%|██▌       | 8/31 [00:47<02:15,  5.88s/it]
Training Epoch: 4/100, batch 14/63 completed (loss: 0.1773:  26%|██▌       | 8/31 [00:47<02:15,  5.89s/it]
Training Epoch: 4/100, batch 15/63 completed (loss: 0.1786:  26%|██▌       | 8/31 [00:47<02:15,  5.89s/it]
Training Epoch: 4/100, batch 15/63 completed (loss: 0.1131:  26%|██▌       | 8/31 [00:47<02:15,  5.88s/it]
Training Epoch: 4/100, batch 15/63 completed (loss: 0.1329:  26%|██▌       | 8/31 [00:47<02:15,  5.88s/it]
Training Epoch: 4/100, batch 15/63 completed (loss: 0.1648:  26%|██▌       | 8/31 [00:47<02:15,  5.88s/it]
Training Epoch: 4/100, batch 16/63 completed (loss: 0.1315:  26%|██▌       | 8/31 [00:50<02:15,  5.88s/it]
Training Epoch: 4/100, batch 16/63 completed (loss: 0.0957:  26%|██▌       | 8/31 [00:50<02:15,  5.89s/it]
Training Epoch: 4/100, batch 16/63 completed (loss: 0.0766:  26%|██▌       | 8/31 [00:50<02:15,  5.88s/it]
Training Epoch: 4/100, batch 16/63 completed (loss: 0.0984:  26%|██▌       | 8/31 [00:50<02:15,  5.88s/it]
Training Epoch: 4/100, batch 16/63 completed (loss: 0.0984:  29%|██▉       | 9/31 [00:52<02:08,  5.82s/it]
Training Epoch: 4/100, batch 16/63 completed (loss: 0.0957:  29%|██▉       | 9/31 [00:52<02:08,  5.82s/it]
Training Epoch: 4/100, batch 16/63 completed (loss: 0.0766:  29%|██▉       | 9/31 [00:52<02:08,  5.82s/it]
Training Epoch: 4/100, batch 16/63 completed (loss: 0.1315:  29%|██▉       | 9/31 [00:52<02:08,  5.82s/it]
Training Epoch: 4/100, batch 17/63 completed (loss: 0.1915:  29%|██▉       | 9/31 [00:52<02:08,  5.82s/it]
Training Epoch: 4/100, batch 17/63 completed (loss: 0.1162:  29%|██▉       | 9/31 [00:52<02:08,  5.82s/it]
Training Epoch: 4/100, batch 17/63 completed (loss: 0.1446:  29%|██▉       | 9/31 [00:52<02:08,  5.82s/it]
Training Epoch: 4/100, batch 17/63 completed (loss: 0.2219:  29%|██▉       | 9/31 [00:52<02:08,  5.82s/it]
Training Epoch: 4/100, batch 18/63 completed (loss: 0.098:  29%|██▉       | 9/31 [00:55<02:08,  5.82s/it] 
Training Epoch: 4/100, batch 18/63 completed (loss: 0.0873:  29%|██▉       | 9/31 [00:55<02:08,  5.82s/it]
Training Epoch: 4/100, batch 18/63 completed (loss: 0.1307:  29%|██▉       | 9/31 [00:55<02:08,  5.82s/it]
Training Epoch: 4/100, batch 18/63 completed (loss: 0.0715:  29%|██▉       | 9/31 [00:55<02:08,  5.82s/it]
Training Epoch: 4/100, batch 18/63 completed (loss: 0.098:  32%|███▏      | 10/31 [00:58<02:01,  5.81s/it]
Training Epoch: 4/100, batch 18/63 completed (loss: 0.0715:  32%|███▏      | 10/31 [00:58<02:02,  5.81s/it]
Training Epoch: 4/100, batch 18/63 completed (loss: 0.1307:  32%|███▏      | 10/31 [00:58<02:02,  5.81s/it]
Training Epoch: 4/100, batch 18/63 completed (loss: 0.0873:  32%|███▏      | 10/31 [00:58<02:02,  5.81s/it]
Training Epoch: 4/100, batch 19/63 completed (loss: 0.1774:  32%|███▏      | 10/31 [00:58<02:02,  5.81s/it]
Training Epoch: 4/100, batch 19/63 completed (loss: 0.1697:  32%|███▏      | 10/31 [00:58<02:02,  5.81s/it]
Training Epoch: 4/100, batch 19/63 completed (loss: 0.1765:  32%|███▏      | 10/31 [00:58<02:01,  5.81s/it]
Training Epoch: 4/100, batch 19/63 completed (loss: 0.1375:  32%|███▏      | 10/31 [00:58<02:02,  5.81s/it]
Training Epoch: 4/100, batch 20/63 completed (loss: 0.0649:  32%|███▏      | 10/31 [01:01<02:02,  5.81s/it]
Training Epoch: 4/100, batch 20/63 completed (loss: 0.2103:  32%|███▏      | 10/31 [01:01<02:01,  5.81s/it]
Training Epoch: 4/100, batch 20/63 completed (loss: 0.1472:  32%|███▏      | 10/31 [01:01<02:02,  5.81s/it]
Training Epoch: 4/100, batch 20/63 completed (loss: 0.1676:  32%|███▏      | 10/31 [01:01<02:02,  5.81s/it]
Training Epoch: 4/100, batch 20/63 completed (loss: 0.1676:  35%|███▌      | 11/31 [01:04<01:56,  5.85s/it]
Training Epoch: 4/100, batch 20/63 completed (loss: 0.0649:  35%|███▌      | 11/31 [01:04<01:56,  5.84s/it]
Training Epoch: 4/100, batch 20/63 completed (loss: 0.1472:  35%|███▌      | 11/31 [01:04<01:56,  5.85s/it]
Training Epoch: 4/100, batch 20/63 completed (loss: 0.2103:  35%|███▌      | 11/31 [01:04<01:56,  5.85s/it]
Training Epoch: 4/100, batch 21/63 completed (loss: 0.1654:  35%|███▌      | 11/31 [01:04<01:56,  5.84s/it]
Training Epoch: 4/100, batch 21/63 completed (loss: 0.119:  35%|███▌      | 11/31 [01:04<01:56,  5.85s/it] 
Training Epoch: 4/100, batch 21/63 completed (loss: 0.0685:  35%|███▌      | 11/31 [01:04<01:56,  5.85s/it]
Training Epoch: 4/100, batch 21/63 completed (loss: 0.2907:  35%|███▌      | 11/31 [01:04<01:56,  5.85s/it]
Training Epoch: 4/100, batch 22/63 completed (loss: 0.1227:  35%|███▌      | 11/31 [01:07<01:56,  5.85s/it]
Training Epoch: 4/100, batch 22/63 completed (loss: 0.0598:  35%|███▌      | 11/31 [01:07<01:56,  5.85s/it]
Training Epoch: 4/100, batch 22/63 completed (loss: 0.1764:  35%|███▌      | 11/31 [01:07<01:56,  5.84s/it]
Training Epoch: 4/100, batch 22/63 completed (loss: 0.1738:  35%|███▌      | 11/31 [01:07<01:56,  5.85s/it]
Training Epoch: 4/100, batch 22/63 completed (loss: 0.0598:  39%|███▊      | 12/31 [01:10<01:50,  5.80s/it]
Training Epoch: 4/100, batch 22/63 completed (loss: 0.1738:  39%|███▊      | 12/31 [01:10<01:50,  5.80s/it]
Training Epoch: 4/100, batch 22/63 completed (loss: 0.1227:  39%|███▊      | 12/31 [01:10<01:50,  5.80s/it]
Training Epoch: 4/100, batch 22/63 completed (loss: 0.1764:  39%|███▊      | 12/31 [01:10<01:50,  5.80s/it]
Training Epoch: 4/100, batch 23/63 completed (loss: 0.0633:  39%|███▊      | 12/31 [01:10<01:50,  5.80s/it]
Training Epoch: 4/100, batch 23/63 completed (loss: 0.0572:  39%|███▊      | 12/31 [01:10<01:50,  5.80s/it]
Training Epoch: 4/100, batch 23/63 completed (loss: 0.1623:  39%|███▊      | 12/31 [01:10<01:50,  5.80s/it]
Training Epoch: 4/100, batch 23/63 completed (loss: 0.2394:  39%|███▊      | 12/31 [01:10<01:50,  5.80s/it]
Training Epoch: 4/100, batch 24/63 completed (loss: 0.1449:  39%|███▊      | 12/31 [01:13<01:50,  5.80s/it]
Training Epoch: 4/100, batch 24/63 completed (loss: 0.2443:  39%|███▊      | 12/31 [01:13<01:50,  5.80s/it]
Training Epoch: 4/100, batch 24/63 completed (loss: 0.1599:  39%|███▊      | 12/31 [01:13<01:50,  5.80s/it]
Training Epoch: 4/100, batch 24/63 completed (loss: 0.1187:  39%|███▊      | 12/31 [01:13<01:50,  5.80s/it]
Training Epoch: 4/100, batch 24/63 completed (loss: 0.1187:  42%|████▏     | 13/31 [01:16<01:47,  5.95s/it]
Training Epoch: 4/100, batch 24/63 completed (loss: 0.1449:  42%|████▏     | 13/31 [01:16<01:47,  5.95s/it]
Training Epoch: 4/100, batch 24/63 completed (loss: 0.1599:  42%|████▏     | 13/31 [01:16<01:47,  5.95s/it]
Training Epoch: 4/100, batch 24/63 completed (loss: 0.2443:  42%|████▏     | 13/31 [01:16<01:47,  5.95s/it]
Training Epoch: 4/100, batch 25/63 completed (loss: 0.1474:  42%|████▏     | 13/31 [01:16<01:47,  5.95s/it]
Training Epoch: 4/100, batch 25/63 completed (loss: 0.0931:  42%|████▏     | 13/31 [01:16<01:47,  5.95s/it]
Training Epoch: 4/100, batch 25/63 completed (loss: 0.1085:  42%|████▏     | 13/31 [01:16<01:47,  5.95s/it]
Training Epoch: 4/100, batch 25/63 completed (loss: 0.0806:  42%|████▏     | 13/31 [01:16<01:47,  5.95s/it]
Training Epoch: 4/100, batch 26/63 completed (loss: 0.2002:  42%|████▏     | 13/31 [01:19<01:47,  5.95s/it]
Training Epoch: 4/100, batch 26/63 completed (loss: 0.2854:  42%|████▏     | 13/31 [01:19<01:47,  5.95s/it]
Training Epoch: 4/100, batch 26/63 completed (loss: 0.1361:  42%|████▏     | 13/31 [01:19<01:47,  5.95s/it]
Training Epoch: 4/100, batch 26/63 completed (loss: 0.0926:  42%|████▏     | 13/31 [01:19<01:47,  5.95s/it]
Training Epoch: 4/100, batch 26/63 completed (loss: 0.1361:  45%|████▌     | 14/31 [01:22<01:39,  5.88s/it]
Training Epoch: 4/100, batch 26/63 completed (loss: 0.0926:  45%|████▌     | 14/31 [01:22<01:39,  5.88s/it]
Training Epoch: 4/100, batch 26/63 completed (loss: 0.2002:  45%|████▌     | 14/31 [01:22<01:39,  5.88s/it]
Training Epoch: 4/100, batch 26/63 completed (loss: 0.2854:  45%|████▌     | 14/31 [01:22<01:39,  5.88s/it]
Training Epoch: 4/100, batch 27/63 completed (loss: 0.1447:  45%|████▌     | 14/31 [01:22<01:39,  5.88s/it]
Training Epoch: 4/100, batch 27/63 completed (loss: 0.0767:  45%|████▌     | 14/31 [01:22<01:39,  5.88s/it]
Training Epoch: 4/100, batch 27/63 completed (loss: 0.1736:  45%|████▌     | 14/31 [01:22<01:39,  5.88s/it]
Training Epoch: 4/100, batch 27/63 completed (loss: 0.1375:  45%|████▌     | 14/31 [01:22<01:39,  5.88s/it]
Training Epoch: 4/100, batch 28/63 completed (loss: 0.0666:  45%|████▌     | 14/31 [01:25<01:39,  5.88s/it]
Training Epoch: 4/100, batch 28/63 completed (loss: 0.0748:  45%|████▌     | 14/31 [01:25<01:39,  5.88s/it]
Training Epoch: 4/100, batch 28/63 completed (loss: 0.1804:  45%|████▌     | 14/31 [01:25<01:39,  5.88s/it]
Training Epoch: 4/100, batch 28/63 completed (loss: 0.1798:  45%|████▌     | 14/31 [01:25<01:39,  5.88s/it]
Training Epoch: 4/100, batch 28/63 completed (loss: 0.0748:  48%|████▊     | 15/31 [01:28<01:34,  5.89s/it]
Training Epoch: 4/100, batch 28/63 completed (loss: 0.1798:  48%|████▊     | 15/31 [01:28<01:34,  5.89s/it]
Training Epoch: 4/100, batch 28/63 completed (loss: 0.1804:  48%|████▊     | 15/31 [01:28<01:34,  5.89s/it]
Training Epoch: 4/100, batch 28/63 completed (loss: 0.0666:  48%|████▊     | 15/31 [01:28<01:34,  5.89s/it]
Training Epoch: 4/100, batch 29/63 completed (loss: 0.13:  48%|████▊     | 15/31 [01:28<01:34,  5.89s/it]  
Training Epoch: 4/100, batch 29/63 completed (loss: 0.2206:  48%|████▊     | 15/31 [01:28<01:34,  5.89s/it]
Training Epoch: 4/100, batch 29/63 completed (loss: 0.1844:  48%|████▊     | 15/31 [01:28<01:34,  5.89s/it]
Training Epoch: 4/100, batch 29/63 completed (loss: 0.1378:  48%|████▊     | 15/31 [01:28<01:34,  5.89s/it]
Training Epoch: 4/100, batch 30/63 completed (loss: 0.0695:  48%|████▊     | 15/31 [01:31<01:34,  5.89s/it]
Training Epoch: 4/100, batch 30/63 completed (loss: 0.196:  48%|████▊     | 15/31 [01:31<01:34,  5.89s/it] 
Training Epoch: 4/100, batch 30/63 completed (loss: 0.2424:  48%|████▊     | 15/31 [01:31<01:34,  5.89s/it]
Training Epoch: 4/100, batch 30/63 completed (loss: 0.1496:  48%|████▊     | 15/31 [01:31<01:34,  5.89s/it]
Training Epoch: 4/100, batch 30/63 completed (loss: 0.196:  52%|█████▏    | 16/31 [01:34<01:29,  5.98s/it]
Training Epoch: 4/100, batch 30/63 completed (loss: 0.1496:  52%|█████▏    | 16/31 [01:34<01:29,  5.98s/it]
Training Epoch: 4/100, batch 30/63 completed (loss: 0.0695:  52%|█████▏    | 16/31 [01:34<01:29,  5.98s/it]
Training Epoch: 4/100, batch 30/63 completed (loss: 0.2424:  52%|█████▏    | 16/31 [01:34<01:29,  5.98s/it]
Training Epoch: 4/100, batch 31/63 completed (loss: 0.0797:  52%|█████▏    | 16/31 [01:34<01:29,  5.98s/it]
Training Epoch: 4/100, batch 31/63 completed (loss: 0.0823:  52%|█████▏    | 16/31 [01:34<01:29,  5.98s/it]
Training Epoch: 4/100, batch 31/63 completed (loss: 0.1094:  52%|█████▏    | 16/31 [01:34<01:29,  5.98s/it]
Training Epoch: 4/100, batch 31/63 completed (loss: 0.1165:  52%|█████▏    | 16/31 [01:34<01:29,  5.98s/it]
Training Epoch: 4/100, batch 32/63 completed (loss: 0.0966:  52%|█████▏    | 16/31 [01:37<01:29,  5.98s/it]
Training Epoch: 4/100, batch 32/63 completed (loss: 0.1385:  52%|█████▏    | 16/31 [01:37<01:29,  5.98s/it]
Training Epoch: 4/100, batch 32/63 completed (loss: 0.1313:  52%|█████▏    | 16/31 [01:37<01:29,  5.98s/it]
Training Epoch: 4/100, batch 32/63 completed (loss: 0.0678:  52%|█████▏    | 16/31 [01:37<01:29,  5.98s/it]
Training Epoch: 4/100, batch 32/63 completed (loss: 0.0678:  55%|█████▍    | 17/31 [01:40<01:22,  5.92s/it]
Training Epoch: 4/100, batch 32/63 completed (loss: 0.0966:  55%|█████▍    | 17/31 [01:40<01:22,  5.91s/it]
Training Epoch: 4/100, batch 32/63 completed (loss: 0.1385:  55%|█████▍    | 17/31 [01:40<01:22,  5.91s/it]
Training Epoch: 4/100, batch 32/63 completed (loss: 0.1313:  55%|█████▍    | 17/31 [01:40<01:22,  5.92s/it]
Training Epoch: 4/100, batch 33/63 completed (loss: 0.1283:  55%|█████▍    | 17/31 [01:40<01:22,  5.92s/it]
Training Epoch: 4/100, batch 33/63 completed (loss: 0.1559:  55%|█████▍    | 17/31 [01:40<01:22,  5.91s/it]
Training Epoch: 4/100, batch 33/63 completed (loss: 0.1243:  55%|█████▍    | 17/31 [01:40<01:22,  5.91s/it]
Training Epoch: 4/100, batch 33/63 completed (loss: 0.1602:  55%|█████▍    | 17/31 [01:40<01:22,  5.92s/it]
Training Epoch: 4/100, batch 34/63 completed (loss: 0.1544:  55%|█████▍    | 17/31 [01:43<01:22,  5.91s/it]
Training Epoch: 4/100, batch 34/63 completed (loss: 0.1799:  55%|█████▍    | 17/31 [01:43<01:22,  5.91s/it]
Training Epoch: 4/100, batch 34/63 completed (loss: 0.0982:  55%|█████▍    | 17/31 [01:43<01:22,  5.92s/it]
Training Epoch: 4/100, batch 34/63 completed (loss: 0.0945:  55%|█████▍    | 17/31 [01:43<01:22,  5.92s/it]
Training Epoch: 4/100, batch 34/63 completed (loss: 0.0945:  58%|█████▊    | 18/31 [01:46<01:18,  6.00s/it]
Training Epoch: 4/100, batch 34/63 completed (loss: 0.1799:  58%|█████▊    | 18/31 [01:46<01:18,  6.00s/it]
Training Epoch: 4/100, batch 34/63 completed (loss: 0.0982:  58%|█████▊    | 18/31 [01:46<01:18,  6.00s/it]
Training Epoch: 4/100, batch 34/63 completed (loss: 0.1544:  58%|█████▊    | 18/31 [01:46<01:18,  6.01s/it]
Training Epoch: 4/100, batch 35/63 completed (loss: 0.1519:  58%|█████▊    | 18/31 [01:46<01:18,  6.00s/it]
Training Epoch: 4/100, batch 35/63 completed (loss: 0.1504:  58%|█████▊    | 18/31 [01:46<01:18,  6.01s/it]
Training Epoch: 4/100, batch 35/63 completed (loss: 0.1063:  58%|█████▊    | 18/31 [01:46<01:18,  6.00s/it]
Training Epoch: 4/100, batch 35/63 completed (loss: 0.193:  58%|█████▊    | 18/31 [01:46<01:18,  6.00s/it] 
Training Epoch: 4/100, batch 36/63 completed (loss: 0.1116:  58%|█████▊    | 18/31 [01:49<01:18,  6.00s/it]
Training Epoch: 4/100, batch 36/63 completed (loss: 0.1555:  58%|█████▊    | 18/31 [01:49<01:18,  6.00s/it]
Training Epoch: 4/100, batch 36/63 completed (loss: 0.1151:  58%|█████▊    | 18/31 [01:49<01:18,  6.01s/it]
Training Epoch: 4/100, batch 36/63 completed (loss: 0.0552:  58%|█████▊    | 18/31 [01:49<01:18,  6.00s/it]
Training Epoch: 4/100, batch 36/63 completed (loss: 0.0552:  61%|██████▏   | 19/31 [01:52<01:11,  5.97s/it]
Training Epoch: 4/100, batch 36/63 completed (loss: 0.1555:  61%|██████▏   | 19/31 [01:52<01:11,  5.97s/it]
Training Epoch: 4/100, batch 36/63 completed (loss: 0.1116:  61%|██████▏   | 19/31 [01:52<01:11,  5.97s/it]
Training Epoch: 4/100, batch 36/63 completed (loss: 0.1151:  61%|██████▏   | 19/31 [01:52<01:11,  5.97s/it]
Training Epoch: 4/100, batch 37/63 completed (loss: 0.122:  61%|██████▏   | 19/31 [01:52<01:11,  5.97s/it] 
Training Epoch: 4/100, batch 37/63 completed (loss: 0.1573:  61%|██████▏   | 19/31 [01:52<01:11,  5.97s/it]
Training Epoch: 4/100, batch 37/63 completed (loss: 0.0647:  61%|██████▏   | 19/31 [01:52<01:11,  5.97s/it]
Training Epoch: 4/100, batch 37/63 completed (loss: 0.0854:  61%|██████▏   | 19/31 [01:52<01:11,  5.97s/it]
Training Epoch: 4/100, batch 38/63 completed (loss: 0.1:  61%|██████▏   | 19/31 [01:55<01:11,  5.97s/it]  
Training Epoch: 4/100, batch 38/63 completed (loss: 0.0691:  61%|██████▏   | 19/31 [01:55<01:11,  5.97s/it]
Training Epoch: 4/100, batch 38/63 completed (loss: 0.0431:  61%|██████▏   | 19/31 [01:55<01:11,  5.97s/it]
Training Epoch: 4/100, batch 38/63 completed (loss: 0.1043:  61%|██████▏   | 19/31 [01:55<01:11,  5.97s/it]
Training Epoch: 4/100, batch 38/63 completed (loss: 0.0431:  65%|██████▍   | 20/31 [01:57<01:05,  5.91s/it]
Training Epoch: 4/100, batch 38/63 completed (loss: 0.1043:  65%|██████▍   | 20/31 [01:57<01:05,  5.91s/it]
Training Epoch: 4/100, batch 38/63 completed (loss: 0.0691:  65%|██████▍   | 20/31 [01:57<01:05,  5.91s/it]
Training Epoch: 4/100, batch 38/63 completed (loss: 0.1:  65%|██████▍   | 20/31 [01:57<01:05,  5.92s/it]
Training Epoch: 4/100, batch 39/63 completed (loss: 0.1521:  65%|██████▍   | 20/31 [01:58<01:05,  5.91s/it]
Training Epoch: 4/100, batch 39/63 completed (loss: 0.17:  65%|██████▍   | 20/31 [01:58<01:05,  5.92s/it]
Training Epoch: 4/100, batch 39/63 completed (loss: 0.1172:  65%|██████▍   | 20/31 [01:58<01:05,  5.91s/it]
Training Epoch: 4/100, batch 39/63 completed (loss: 0.1657:  65%|██████▍   | 20/31 [01:58<01:05,  5.91s/it]
Training Epoch: 4/100, batch 40/63 completed (loss: 0.1744:  65%|██████▍   | 20/31 [02:01<01:05,  5.91s/it]
Training Epoch: 4/100, batch 40/63 completed (loss: 0.2609:  65%|██████▍   | 20/31 [02:01<01:05,  5.91s/it]
Training Epoch: 4/100, batch 40/63 completed (loss: 0.3143:  65%|██████▍   | 20/31 [02:01<01:05,  5.92s/it]
Training Epoch: 4/100, batch 40/63 completed (loss: 0.1026:  65%|██████▍   | 20/31 [02:01<01:05,  5.91s/it]
Training Epoch: 4/100, batch 40/63 completed (loss: 0.1026:  68%|██████▊   | 21/31 [02:04<01:00,  6.06s/it]
Training Epoch: 4/100, batch 40/63 completed (loss: 0.3143:  68%|██████▊   | 21/31 [02:04<01:00,  6.06s/it]
Training Epoch: 4/100, batch 40/63 completed (loss: 0.1744:  68%|██████▊   | 21/31 [02:04<01:00,  6.07s/it]
Training Epoch: 4/100, batch 40/63 completed (loss: 0.2609:  68%|██████▊   | 21/31 [02:04<01:00,  6.06s/it]
Training Epoch: 4/100, batch 41/63 completed (loss: 0.0983:  68%|██████▊   | 21/31 [02:04<01:00,  6.06s/it]
Training Epoch: 4/100, batch 41/63 completed (loss: 0.1288:  68%|██████▊   | 21/31 [02:04<01:00,  6.07s/it]
Training Epoch: 4/100, batch 41/63 completed (loss: 0.0603:  68%|██████▊   | 21/31 [02:04<01:00,  6.06s/it]
Training Epoch: 4/100, batch 41/63 completed (loss: 0.1816:  68%|██████▊   | 21/31 [02:04<01:00,  6.06s/it]
Training Epoch: 4/100, batch 42/63 completed (loss: 0.2608:  68%|██████▊   | 21/31 [02:07<01:00,  6.07s/it]
Training Epoch: 4/100, batch 42/63 completed (loss: 0.0691:  68%|██████▊   | 21/31 [02:07<01:00,  6.06s/it]
Training Epoch: 4/100, batch 42/63 completed (loss: 0.1289:  68%|██████▊   | 21/31 [02:07<01:00,  6.06s/it]
Training Epoch: 4/100, batch 42/63 completed (loss: 0.0328:  68%|██████▊   | 21/31 [02:07<01:00,  6.06s/it]
Training Epoch: 4/100, batch 42/63 completed (loss: 0.0328:  71%|███████   | 22/31 [02:10<00:53,  5.99s/it]
Training Epoch: 4/100, batch 42/63 completed (loss: 0.2608:  71%|███████   | 22/31 [02:10<00:53,  5.99s/it]
Training Epoch: 4/100, batch 42/63 completed (loss: 0.0691:  71%|███████   | 22/31 [02:10<00:53,  5.99s/it]
Training Epoch: 4/100, batch 42/63 completed (loss: 0.1289:  71%|███████   | 22/31 [02:10<00:53,  5.99s/it]
Training Epoch: 4/100, batch 43/63 completed (loss: 0.109:  71%|███████   | 22/31 [02:10<00:53,  5.99s/it] 
Training Epoch: 4/100, batch 43/63 completed (loss: 0.1896:  71%|███████   | 22/31 [02:10<00:53,  5.99s/it]
Training Epoch: 4/100, batch 43/63 completed (loss: 0.1913:  71%|███████   | 22/31 [02:10<00:53,  5.99s/it]
Training Epoch: 4/100, batch 43/63 completed (loss: 0.0378:  71%|███████   | 22/31 [02:10<00:53,  5.99s/it]
Training Epoch: 4/100, batch 44/63 completed (loss: 0.1301:  71%|███████   | 22/31 [02:13<00:53,  5.99s/it]
Training Epoch: 4/100, batch 44/63 completed (loss: 0.1902:  71%|███████   | 22/31 [02:13<00:53,  5.99s/it]
Training Epoch: 4/100, batch 44/63 completed (loss: 0.2094:  71%|███████   | 22/31 [02:13<00:53,  5.99s/it]
Training Epoch: 4/100, batch 44/63 completed (loss: 0.0521:  71%|███████   | 22/31 [02:13<00:53,  5.99s/it]
Training Epoch: 4/100, batch 44/63 completed (loss: 0.0521:  74%|███████▍  | 23/31 [02:16<00:47,  5.99s/it]
Training Epoch: 4/100, batch 44/63 completed (loss: 0.2094:  74%|███████▍  | 23/31 [02:16<00:47,  5.99s/it]
Training Epoch: 4/100, batch 44/63 completed (loss: 0.1902:  74%|███████▍  | 23/31 [02:16<00:47,  5.99s/it]
Training Epoch: 4/100, batch 44/63 completed (loss: 0.1301:  74%|███████▍  | 23/31 [02:16<00:47,  5.99s/it]
Training Epoch: 4/100, batch 45/63 completed (loss: 0.1015:  74%|███████▍  | 23/31 [02:16<00:47,  5.99s/it]
Training Epoch: 4/100, batch 45/63 completed (loss: 0.1559:  74%|███████▍  | 23/31 [02:16<00:47,  5.99s/it]
Training Epoch: 4/100, batch 45/63 completed (loss: 0.1294:  74%|███████▍  | 23/31 [02:16<00:47,  5.99s/it]
Training Epoch: 4/100, batch 45/63 completed (loss: 0.1246:  74%|███████▍  | 23/31 [02:16<00:47,  5.99s/it]
Training Epoch: 4/100, batch 46/63 completed (loss: 0.064:  74%|███████▍  | 23/31 [02:19<00:47,  5.99s/it] 
Training Epoch: 4/100, batch 46/63 completed (loss: 0.1704:  74%|███████▍  | 23/31 [02:19<00:47,  5.99s/it]
Training Epoch: 4/100, batch 46/63 completed (loss: 0.1415:  74%|███████▍  | 23/31 [02:19<00:47,  5.99s/it]
Training Epoch: 4/100, batch 46/63 completed (loss: 0.0969:  74%|███████▍  | 23/31 [02:19<00:47,  5.99s/it]
Training Epoch: 4/100, batch 46/63 completed (loss: 0.0969:  77%|███████▋  | 24/31 [02:22<00:42,  6.06s/it]
Training Epoch: 4/100, batch 46/63 completed (loss: 0.1704:  77%|███████▋  | 24/31 [02:22<00:42,  6.06s/it]
Training Epoch: 4/100, batch 46/63 completed (loss: 0.064:  77%|███████▋  | 24/31 [02:22<00:42,  6.06s/it]
Training Epoch: 4/100, batch 46/63 completed (loss: 0.1415:  77%|███████▋  | 24/31 [02:22<00:42,  6.07s/it]
Training Epoch: 4/100, batch 47/63 completed (loss: 0.1543:  77%|███████▋  | 24/31 [02:22<00:42,  6.06s/it]
Training Epoch: 4/100, batch 47/63 completed (loss: 0.2151:  77%|███████▋  | 24/31 [02:22<00:42,  6.06s/it]
Training Epoch: 4/100, batch 47/63 completed (loss: 0.1986:  77%|███████▋  | 24/31 [02:22<00:42,  6.07s/it]
Training Epoch: 4/100, batch 47/63 completed (loss: 0.0391:  77%|███████▋  | 24/31 [02:22<00:42,  6.06s/it]
Training Epoch: 4/100, batch 48/63 completed (loss: 0.1292:  77%|███████▋  | 24/31 [02:25<00:42,  6.07s/it]
Training Epoch: 4/100, batch 48/63 completed (loss: 0.091:  77%|███████▋  | 24/31 [02:25<00:42,  6.06s/it] 
Training Epoch: 4/100, batch 48/63 completed (loss: 0.1265:  77%|███████▋  | 24/31 [02:25<00:42,  6.06s/it]
Training Epoch: 4/100, batch 48/63 completed (loss: 0.2329:  77%|███████▋  | 24/31 [02:25<00:42,  6.06s/it]
Training Epoch: 4/100, batch 48/63 completed (loss: 0.2329:  81%|████████  | 25/31 [02:28<00:35,  5.98s/it]
Training Epoch: 4/100, batch 48/63 completed (loss: 0.091:  81%|████████  | 25/31 [02:28<00:35,  5.99s/it]
Training Epoch: 4/100, batch 48/63 completed (loss: 0.1292:  81%|████████  | 25/31 [02:28<00:35,  5.99s/it]
Training Epoch: 4/100, batch 48/63 completed (loss: 0.1265:  81%|████████  | 25/31 [02:28<00:35,  5.99s/it]
Training Epoch: 4/100, batch 49/63 completed (loss: 0.1764:  81%|████████  | 25/31 [02:28<00:35,  5.99s/it]
Training Epoch: 4/100, batch 49/63 completed (loss: 0.1031:  81%|████████  | 25/31 [02:28<00:35,  5.99s/it]
Training Epoch: 4/100, batch 49/63 completed (loss: 0.3324:  81%|████████  | 25/31 [02:28<00:35,  5.99s/it]
Training Epoch: 4/100, batch 49/63 completed (loss: 0.1842:  81%|████████  | 25/31 [02:28<00:35,  5.98s/it]
Training Epoch: 4/100, batch 50/63 completed (loss: 0.087:  81%|████████  | 25/31 [02:31<00:35,  5.99s/it] 
Training Epoch: 4/100, batch 50/63 completed (loss: 0.0478:  81%|████████  | 25/31 [02:31<00:35,  5.99s/it]
Training Epoch: 4/100, batch 50/63 completed (loss: 0.1262:  81%|████████  | 25/31 [02:31<00:35,  5.99s/it]
Training Epoch: 4/100, batch 50/63 completed (loss: 0.0888:  81%|████████  | 25/31 [02:31<00:35,  5.98s/it]
Training Epoch: 4/100, batch 50/63 completed (loss: 0.1262:  84%|████████▍ | 26/31 [02:34<00:30,  6.08s/it]
Training Epoch: 4/100, batch 50/63 completed (loss: 0.0888:  84%|████████▍ | 26/31 [02:34<00:30,  6.09s/it]
Training Epoch: 4/100, batch 50/63 completed (loss: 0.087:  84%|████████▍ | 26/31 [02:34<00:30,  6.08s/it]
Training Epoch: 4/100, batch 50/63 completed (loss: 0.0478:  84%|████████▍ | 26/31 [02:34<00:30,  6.08s/it]
Training Epoch: 4/100, batch 51/63 completed (loss: 0.0812:  84%|████████▍ | 26/31 [02:34<00:30,  6.08s/it]
Training Epoch: 4/100, batch 51/63 completed (loss: 0.2113:  84%|████████▍ | 26/31 [02:34<00:30,  6.08s/it]
Training Epoch: 4/100, batch 51/63 completed (loss: 0.1418:  84%|████████▍ | 26/31 [02:34<00:30,  6.08s/it]
Training Epoch: 4/100, batch 51/63 completed (loss: 0.072:  84%|████████▍ | 26/31 [02:34<00:30,  6.09s/it] 
Training Epoch: 4/100, batch 52/63 completed (loss: 0.1643:  84%|████████▍ | 26/31 [02:37<00:30,  6.08s/it]
Training Epoch: 4/100, batch 52/63 completed (loss: 0.1438:  84%|████████▍ | 26/31 [02:37<00:30,  6.08s/it]
Training Epoch: 4/100, batch 52/63 completed (loss: 0.219:  84%|████████▍ | 26/31 [02:37<00:30,  6.08s/it] 
Training Epoch: 4/100, batch 52/63 completed (loss: 0.2162:  84%|████████▍ | 26/31 [02:37<00:30,  6.09s/it]
Training Epoch: 4/100, batch 52/63 completed (loss: 0.2162:  87%|████████▋ | 27/31 [02:40<00:24,  6.05s/it]
Training Epoch: 4/100, batch 52/63 completed (loss: 0.1438:  87%|████████▋ | 27/31 [02:40<00:24,  6.04s/it]
Training Epoch: 4/100, batch 52/63 completed (loss: 0.1643:  87%|████████▋ | 27/31 [02:40<00:24,  6.04s/it]
Training Epoch: 4/100, batch 52/63 completed (loss: 0.219:  87%|████████▋ | 27/31 [02:40<00:24,  6.04s/it]
Training Epoch: 4/100, batch 53/63 completed (loss: 0.1678:  87%|████████▋ | 27/31 [02:40<00:24,  6.04s/it]
Training Epoch: 4/100, batch 53/63 completed (loss: 0.05:  87%|████████▋ | 27/31 [02:40<00:24,  6.04s/it]  
Training Epoch: 4/100, batch 53/63 completed (loss: 0.1249:  87%|████████▋ | 27/31 [02:40<00:24,  6.04s/it]
Training Epoch: 4/100, batch 53/63 completed (loss: 0.2191:  87%|████████▋ | 27/31 [02:40<00:24,  6.05s/it]
Training Epoch: 4/100, batch 54/63 completed (loss: 0.1611:  87%|████████▋ | 27/31 [02:43<00:24,  6.04s/it]
Training Epoch: 4/100, batch 54/63 completed (loss: 0.0966:  87%|████████▋ | 27/31 [02:43<00:24,  6.04s/it]
Training Epoch: 4/100, batch 54/63 completed (loss: 0.2632:  87%|████████▋ | 27/31 [02:43<00:24,  6.04s/it]
Training Epoch: 4/100, batch 54/63 completed (loss: 0.0888:  87%|████████▋ | 27/31 [02:43<00:24,  6.05s/it]
Training Epoch: 4/100, batch 54/63 completed (loss: 0.0888:  90%|█████████ | 28/31 [02:46<00:17,  5.98s/it]
Training Epoch: 4/100, batch 54/63 completed (loss: 0.2632:  90%|█████████ | 28/31 [02:46<00:17,  5.97s/it]
Training Epoch: 4/100, batch 54/63 completed (loss: 0.0966:  90%|█████████ | 28/31 [02:46<00:17,  5.97s/it]
Training Epoch: 4/100, batch 54/63 completed (loss: 0.1611:  90%|█████████ | 28/31 [02:46<00:17,  5.97s/it]
Training Epoch: 4/100, batch 55/63 completed (loss: 0.2617:  90%|█████████ | 28/31 [02:46<00:17,  5.97s/it]
Training Epoch: 4/100, batch 55/63 completed (loss: 0.0784:  90%|█████████ | 28/31 [02:46<00:17,  5.97s/it]
Training Epoch: 4/100, batch 55/63 completed (loss: 0.1172:  90%|█████████ | 28/31 [02:46<00:17,  5.97s/it]
Training Epoch: 4/100, batch 55/63 completed (loss: 0.0982:  90%|█████████ | 28/31 [02:46<00:17,  5.98s/it]
Training Epoch: 4/100, batch 56/63 completed (loss: 0.0926:  90%|█████████ | 28/31 [02:49<00:17,  5.97s/it]
Training Epoch: 4/100, batch 56/63 completed (loss: 0.1155:  90%|█████████ | 28/31 [02:49<00:17,  5.97s/it]
Training Epoch: 4/100, batch 56/63 completed (loss: 0.1505:  90%|█████████ | 28/31 [02:49<00:17,  5.97s/it]
Training Epoch: 4/100, batch 56/63 completed (loss: 0.1136:  90%|█████████ | 28/31 [02:49<00:17,  5.98s/it]
Training Epoch: 4/100, batch 56/63 completed (loss: 0.1155:  94%|█████████▎| 29/31 [02:52<00:12,  6.10s/it]
Training Epoch: 4/100, batch 56/63 completed (loss: 0.0926:  94%|█████████▎| 29/31 [02:52<00:12,  6.10s/it]
Training Epoch: 4/100, batch 56/63 completed (loss: 0.1136:  94%|█████████▎| 29/31 [02:52<00:12,  6.11s/it]
Training Epoch: 4/100, batch 56/63 completed (loss: 0.1505:  94%|█████████▎| 29/31 [02:52<00:12,  6.11s/it]
Training Epoch: 4/100, batch 57/63 completed (loss: 0.181:  94%|█████████▎| 29/31 [02:52<00:12,  6.10s/it] 
Training Epoch: 4/100, batch 57/63 completed (loss: 0.1474:  94%|█████████▎| 29/31 [02:52<00:12,  6.10s/it]
Training Epoch: 4/100, batch 57/63 completed (loss: 0.2428:  94%|█████████▎| 29/31 [02:52<00:12,  6.11s/it]
Training Epoch: 4/100, batch 57/63 completed (loss: 0.0701:  94%|█████████▎| 29/31 [02:52<00:12,  6.11s/it]
Training Epoch: 4/100, batch 58/63 completed (loss: 0.0942:  94%|█████████▎| 29/31 [02:55<00:12,  6.11s/it]
Training Epoch: 4/100, batch 58/63 completed (loss: 0.0971:  94%|█████████▎| 29/31 [02:55<00:12,  6.10s/it]
Training Epoch: 4/100, batch 58/63 completed (loss: 0.1991:  94%|█████████▎| 29/31 [02:55<00:12,  6.10s/it]
Training Epoch: 4/100, batch 58/63 completed (loss: 0.1656:  94%|█████████▎| 29/31 [02:55<00:12,  6.11s/it]
Training Epoch: 4/100, batch 58/63 completed (loss: 0.1991:  97%|█████████▋| 30/31 [02:58<00:06,  6.02s/it]
Training Epoch: 4/100, batch 58/63 completed (loss: 0.0971:  97%|█████████▋| 30/31 [02:58<00:06,  6.02s/it]
Training Epoch: 4/100, batch 58/63 completed (loss: 0.1656:  97%|█████████▋| 30/31 [02:58<00:06,  6.02s/it]
Training Epoch: 4/100, batch 58/63 completed (loss: 0.0942:  97%|█████████▋| 30/31 [02:58<00:06,  6.02s/it]
Training Epoch: 4/100, batch 59/63 completed (loss: 0.2251:  97%|█████████▋| 30/31 [02:58<00:06,  6.02s/it]
Training Epoch: 4/100, batch 59/63 completed (loss: 0.0836:  97%|█████████▋| 30/31 [02:58<00:06,  6.02s/it]
Training Epoch: 4/100, batch 59/63 completed (loss: 0.0882:  97%|█████████▋| 30/31 [02:58<00:06,  6.02s/it]
Training Epoch: 4/100, batch 59/63 completed (loss: 0.082:  97%|█████████▋| 30/31 [02:58<00:06,  6.02s/it] 
Training Epoch: 4/100, batch 60/63 completed (loss: 0.1726:  97%|█████████▋| 30/31 [03:01<00:06,  6.02s/it]
Training Epoch: 4/100, batch 60/63 completed (loss: 0.0797:  97%|█████████▋| 30/31 [03:01<00:06,  6.02s/it]
Training Epoch: 4/100, batch 60/63 completed (loss: 0.0784:  97%|█████████▋| 30/31 [03:01<00:06,  6.02s/it]
Training Epoch: 4/100, batch 60/63 completed (loss: 0.1003:  97%|█████████▋| 30/31 [03:01<00:06,  6.02s/it]
Training Epoch: 4/100, batch 60/63 completed (loss: 0.1003: 100%|██████████| 31/31 [03:04<00:00,  5.95s/it]
Training Epoch: 4/100, batch 60/63 completed (loss: 0.0784: 100%|██████████| 31/31 [03:04<00:00,  5.95s/it]
Training Epoch: 4/100, batch 60/63 completed (loss: 0.0797: 100%|██████████| 31/31 [03:04<00:00,  5.95s/it]
Training Epoch: 4/100, batch 60/63 completed (loss: 0.1726: 100%|██████████| 31/31 [03:04<00:00,  5.95s/it]
Training Epoch: 4/100, batch 61/63 completed (loss: 0.1431: 100%|██████████| 31/31 [03:04<00:00,  5.95s/it]
Training Epoch: 4/100, batch 61/63 completed (loss: 0.1883: 100%|██████████| 31/31 [03:04<00:00,  5.95s/it]
Training Epoch: 4/100, batch 61/63 completed (loss: 0.2211: 100%|██████████| 31/31 [03:04<00:00,  5.95s/it]
Training Epoch: 4/100, batch 61/63 completed (loss: 0.1224: 100%|██████████| 31/31 [03:04<00:00,  5.95s/it]
Training Epoch: 4/100, batch 61/63 completed (loss: 0.1431: : 32it [03:06,  4.88s/it]                      
Training Epoch: 4/100, batch 61/63 completed (loss: 0.1224: : 32it [03:06,  4.88s/it]                      
Training Epoch: 4/100, batch 61/63 completed (loss: 0.2211: : 32it [03:06,  4.88s/it]                      
Training Epoch: 4/100, batch 61/63 completed (loss: 0.1883: : 32it [03:06,  4.88s/it]                      
Training Epoch: 4/100, batch 62/63 completed (loss: 0.2383: : 32it [03:06,  4.88s/it]
Training Epoch: 4/100, batch 62/63 completed (loss: 0.0217: : 32it [03:06,  4.88s/it]
Training Epoch: 4/100, batch 62/63 completed (loss: 0.1026: : 32it [03:06,  4.88s/it]
Training Epoch: 4/100, batch 62/63 completed (loss: 0.074: : 32it [03:06,  4.88s/it] 
Training Epoch: 4/100, batch 62/63 completed (loss: 0.1026: : 32it [03:06,  5.84s/it]

Training Epoch: 4/100, batch 62/63 completed (loss: 0.0217: : 32it [03:06,  5.84s/it]

Training Epoch: 4/100, batch 62/63 completed (loss: 0.2383: : 32it [03:06,  5.84s/it]

Training Epoch: 4/100, batch 62/63 completed (loss: 0.074: : 32it [03:06,  5.84s/it]
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/c10d_logger.py:83: UserWarning: barrier(): using the device under current context. You can specify `device_id` in `init_process_group` to mute this warning.
  return func(*args, **kwargs)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
saving model + train state.
Warning: failed to upload checkpoint_4 to HF: (Request ID: Root=1-6a1b6826-0410168e294c292d53e4125a;c77089cc-5a9f-4c23-b717-dff31d81a9fe)

403 Forbidden: You need to setup automatic credit recharge in order to upload more data. You can do so at /settings/billing..
Cannot access content at: https://huggingface.co/seyedparsa/thoughtformer-checkpoints.git/info/lfs/objects/batch.
Make sure your token has the correct permissions.
Warning: failed to upload train_state_4 to HF: (Request ID: Root=1-6a1b6829-7ce27406050d209a01be0d91;031240be-1b50-4ad0-9acd-1c534e88032e)

403 Forbidden: You need to setup automatic credit recharge in order to upload more data. You can do so at /settings/billing..
Cannot access content at: https://huggingface.co/seyedparsa/thoughtformer-checkpoints.git/info/lfs/objects/batch.
Make sure your token has the correct permissions.
Deleted old checkpoint: checkpoint_3
Deleted old checkpoint: train_state_3
[warn] HF upload failed for ckpt_4; kept locally, earlier ckpts deleted.

Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]
Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]
Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]eval loss 0.08887466096452304

Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]
[WRONG] Example 0 (diff=3)
  Expected : 'Claude'
  Extracted: 'Morgan'
  Generated: '### Morgan'

[WRONG] Example 1 (diff=8)
  Expected : 'Guy'
  Extracted: 'Irma'
  Generated: '### Irma'

[CORRECT] Example 2 (diff=10)
  Expected : 'Elvis'
  Extracted: 'Elvis'
  Generated: '### Elvis'
Test Accuracy:   3%|▎         | 1/32 [00:06<03:26,  6.65s/it]

Test accuracy: 0.75:   3%|▎         | 1/32 [00:06<03:26,  6.65s/it]
Test Accuracy:   3%|▎         | 1/32 [00:06<03:25,  6.64s/it]
Test Accuracy:   3%|▎         | 1/32 [00:06<03:26,  6.65s/it]
[WRONG] Example 3 (diff=8)
Test accuracy: 0.62:   3%|▎         | 1/32 [00:06<03:25,  6.64s/it]
  Expected : 'Coral'
  Extracted: 'Claude'
  Generated: '### Claude'

Test accuracy: 0.5:   3%|▎         | 1/32 [00:06<03:26,  6.65s/it]
[CORRECT] Example 4 (diff=1)
  Expected : 'Patricia'
  Extracted: 'Patricia'
  Generated: '### Patricia'

Test Accuracy:   3%|▎         | 1/32 [00:06<03:21,  6.49s/it]
Test accuracy: 0.38:   3%|▎         | 1/32 [00:06<03:21,  6.49s/it]
Test accuracy: 0.62:   6%|▋         | 2/32 [00:13<03:16,  6.57s/it]
Test accuracy: 0.69:   6%|▋         | 2/32 [00:13<03:16,  6.57s/it]
Test accuracy: 0.5:   6%|▋         | 2/32 [00:13<03:17,  6.57s/it]
Test accuracy: 0.38:   6%|▋         | 2/32 [00:12<03:14,  6.50s/it]
Test accuracy: 0.75:   6%|▋         | 2/32 [00:13<03:16,  6.57s/it]
Test accuracy: 0.56:   6%|▋         | 2/32 [00:13<03:17,  6.57s/it]
Test accuracy: 0.81:   6%|▋         | 2/32 [00:13<03:16,  6.57s/it]
Test accuracy: 0.25:   6%|▋         | 2/32 [00:12<03:14,  6.50s/it]
Test accuracy: 0.81:   9%|▉         | 3/32 [00:19<03:13,  6.66s/it]
Test accuracy: 0.79:   9%|▉         | 3/32 [00:19<03:13,  6.66s/it]
Test accuracy: 0.69:   9%|▉         | 3/32 [00:19<03:13,  6.66s/it]
Test accuracy: 0.71:   9%|▉         | 3/32 [00:19<03:13,  6.66s/it]
Test accuracy: 0.25:   9%|▉         | 3/32 [00:19<03:11,  6.62s/it]
Test accuracy: 0.38:   9%|▉         | 3/32 [00:19<03:11,  6.62s/it]
Test accuracy: 0.56:   9%|▉         | 3/32 [00:19<03:13,  6.67s/it]
Test accuracy: 0.58:   9%|▉         | 3/32 [00:19<03:13,  6.67s/it]
Test accuracy: 0.71:  12%|█▎        | 4/32 [00:26<03:05,  6.62s/it]
Test accuracy: 0.58:  12%|█▎        | 4/32 [00:26<03:05,  6.62s/it]
Test accuracy: 0.72:  12%|█▎        | 4/32 [00:26<03:05,  6.62s/it]
Test accuracy: 0.38:  12%|█▎        | 4/32 [00:26<03:04,  6.60s/it]
Test accuracy: 0.56:  12%|█▎        | 4/32 [00:26<03:05,  6.62s/it]
Test accuracy: 0.47:  12%|█▎        | 4/32 [00:26<03:04,  6.60s/it]
Test accuracy: 0.79:  12%|█▎        | 4/32 [00:26<03:05,  6.62s/it]
Test accuracy: 0.75:  12%|█▎        | 4/32 [00:26<03:05,  6.62s/it]
Test accuracy: 0.56:  16%|█▌        | 5/32 [00:32<02:54,  6.47s/it]
Test accuracy: 0.6:  16%|█▌        | 5/32 [00:32<02:54,  6.47s/it] 
Test accuracy: 0.47:  16%|█▌        | 5/32 [00:32<02:54,  6.46s/it]
Test accuracy: 0.72:  16%|█▌        | 5/32 [00:32<02:54,  6.47s/it]
Test accuracy: 0.7:  16%|█▌        | 5/32 [00:32<02:54,  6.47s/it] 
Test accuracy: 0.75:  16%|█▌        | 5/32 [00:32<02:54,  6.47s/it]
Test accuracy: 0.52:  16%|█▌        | 5/32 [00:32<02:54,  6.46s/it]
Test accuracy: 0.73:  16%|█▌        | 5/32 [00:32<02:54,  6.47s/it]
Test accuracy: 0.73:  19%|█▉        | 6/32 [00:38<02:45,  6.38s/it]
Test accuracy: 0.7:  19%|█▉        | 6/32 [00:38<02:45,  6.38s/it]
Test accuracy: 0.71:  19%|█▉        | 6/32 [00:38<02:45,  6.38s/it]
Test accuracy: 0.67:  19%|█▉        | 6/32 [00:38<02:45,  6.38s/it]
Test accuracy: 0.52:  19%|█▉        | 6/32 [00:38<02:45,  6.37s/it]
Test accuracy: 0.6:  19%|█▉        | 6/32 [00:38<02:45,  6.38s/it]
Test accuracy: 0.58:  19%|█▉        | 6/32 [00:38<02:45,  6.37s/it]
Test accuracy: 0.62:  19%|█▉        | 6/32 [00:38<02:45,  6.38s/it]
Test accuracy: 0.71:  22%|██▏       | 7/32 [00:45<02:41,  6.45s/it]
Test accuracy: 0.67:  22%|██▏       | 7/32 [00:45<02:41,  6.45s/it]
Test accuracy: 0.7:  22%|██▏       | 7/32 [00:45<02:41,  6.45s/it] 
Test accuracy: 0.66:  22%|██▏       | 7/32 [00:45<02:41,  6.45s/it]
Test accuracy: 0.62:  22%|██▏       | 7/32 [00:45<02:41,  6.45s/it]
Test accuracy: 0.64:  22%|██▏       | 7/32 [00:45<02:41,  6.45s/it]
Test accuracy: 0.58:  22%|██▏       | 7/32 [00:45<02:41,  6.44s/it]
Test accuracy: 0.61:  22%|██▏       | 7/32 [00:45<02:41,  6.44s/it]
Test accuracy: 0.64:  25%|██▌       | 8/32 [00:51<02:32,  6.37s/it]
Test accuracy: 0.66:  25%|██▌       | 8/32 [00:51<02:32,  6.37s/it]
Test accuracy: 0.61:  25%|██▌       | 8/32 [00:51<02:32,  6.36s/it]
Test accuracy: 0.62:  25%|██▌       | 8/32 [00:51<02:32,  6.36s/it]
Test accuracy: 0.66:  25%|██▌       | 8/32 [00:51<02:32,  6.37s/it]
Test accuracy: 0.7:  25%|██▌       | 8/32 [00:51<02:32,  6.37s/it]
Test accuracy: 0.64:  25%|██▌       | 8/32 [00:51<02:32,  6.37s/it]
Test accuracy: 0.66:  25%|██▌       | 8/32 [00:51<02:32,  6.37s/it]
Test accuracy: 0.66:  28%|██▊       | 9/32 [00:58<02:29,  6.49s/it]
Test accuracy: 0.64:  28%|██▊       | 9/32 [00:58<02:29,  6.49s/it]
Test accuracy: 0.64:  28%|██▊       | 9/32 [00:58<02:29,  6.49s/it]
Test accuracy: 0.65:  28%|██▊       | 9/32 [00:58<02:29,  6.49s/it]
Test accuracy: 0.62:  28%|██▊       | 9/32 [00:58<02:29,  6.49s/it]
Test accuracy: 0.65:  28%|██▊       | 9/32 [00:58<02:29,  6.49s/it]
Test accuracy: 0.66:  28%|██▊       | 9/32 [00:58<02:29,  6.50s/it]
Test accuracy: 0.65:  28%|██▊       | 9/32 [00:58<02:29,  6.50s/it]
Test accuracy: 0.65:  31%|███▏      | 10/32 [01:05<02:23,  6.52s/it]
Test accuracy: 0.65:  31%|███▏      | 10/32 [01:05<02:23,  6.52s/it]
Test accuracy: 0.64:  31%|███▏      | 10/32 [01:05<02:23,  6.52s/it]
Test accuracy: 0.64:  31%|███▏      | 10/32 [01:05<02:23,  6.52s/it]
Test accuracy: 0.65:  31%|███▏      | 10/32 [01:05<02:23,  6.52s/it]
Test accuracy: 0.65:  31%|███▏      | 10/32 [01:04<02:23,  6.52s/it]
Test accuracy: 0.64:  31%|███▏      | 10/32 [01:05<02:23,  6.52s/it]
Test accuracy: 0.61:  31%|███▏      | 10/32 [01:04<02:23,  6.52s/it]
Test accuracy: 0.64:  34%|███▍      | 11/32 [01:11<02:14,  6.42s/it]
Test accuracy: 0.64:  34%|███▍      | 11/32 [01:11<02:14,  6.42s/it]
Test accuracy: 0.61:  34%|███▍      | 11/32 [01:11<02:14,  6.42s/it]
Test accuracy: 0.62:  34%|███▍      | 11/32 [01:11<02:14,  6.42s/it]
Test accuracy: 0.61:  34%|███▍      | 11/32 [01:11<02:14,  6.42s/it]
Test accuracy: 0.65:  34%|███▍      | 11/32 [01:11<02:14,  6.42s/it]
Test accuracy: 0.62:  34%|███▍      | 11/32 [01:11<02:14,  6.42s/it]
Test accuracy: 0.6:  34%|███▍      | 11/32 [01:11<02:14,  6.42s/it] 
Test accuracy: 0.61:  38%|███▊      | 12/32 [01:17<02:08,  6.41s/it]
Test accuracy: 0.58:  38%|███▊      | 12/32 [01:17<02:08,  6.41s/it]
Test accuracy: 0.6:  38%|███▊      | 12/32 [01:17<02:08,  6.41s/it]
Test accuracy: 0.61:  38%|███▊      | 12/32 [01:17<02:08,  6.41s/it]
Test accuracy: 0.62:  38%|███▊      | 12/32 [01:17<02:08,  6.41s/it]
Test accuracy: 0.62:  38%|███▊      | 12/32 [01:17<02:08,  6.41s/it]
Test accuracy: 0.65:  38%|███▊      | 12/32 [01:17<02:08,  6.41s/it]
Test accuracy: 0.6:  38%|███▊      | 12/32 [01:17<02:08,  6.41s/it] 
Test accuracy: 0.65:  41%|████      | 13/32 [01:24<02:01,  6.40s/it]
Test accuracy: 0.61:  41%|████      | 13/32 [01:23<02:01,  6.40s/it]
Test accuracy: 0.64:  41%|████      | 13/32 [01:24<02:01,  6.40s/it]
Test accuracy: 0.6:  41%|████      | 13/32 [01:23<02:01,  6.40s/it] 
Test accuracy: 0.6:  41%|████      | 13/32 [01:24<02:01,  6.40s/it]
Test accuracy: 0.58:  41%|████      | 13/32 [01:24<02:01,  6.40s/it]
Test accuracy: 0.6:  41%|████      | 13/32 [01:24<02:01,  6.40s/it]
Test accuracy: 0.58:  41%|████      | 13/32 [01:24<02:01,  6.40s/it]
Test accuracy: 0.6:  44%|████▍     | 14/32 [01:30<01:54,  6.34s/it]
Test accuracy: 0.58:  44%|████▍     | 14/32 [01:30<01:54,  6.34s/it]
Test accuracy: 0.58:  44%|████▍     | 14/32 [01:30<01:54,  6.34s/it]
Test accuracy: 0.6:  44%|████▍     | 14/32 [01:30<01:54,  6.34s/it]
Test accuracy: 0.6:  44%|████▍     | 14/32 [01:30<01:54,  6.34s/it]
Test accuracy: 0.61:  44%|████▍     | 14/32 [01:30<01:54,  6.34s/it]
Test accuracy: 0.64:  44%|████▍     | 14/32 [01:30<01:54,  6.34s/it]
Test accuracy: 0.62:  44%|████▍     | 14/32 [01:30<01:54,  6.34s/it]
Test accuracy: 0.58:  47%|████▋     | 15/32 [01:36<01:48,  6.36s/it]
Test accuracy: 0.62:  47%|████▋     | 15/32 [01:36<01:48,  6.36s/it]
Test accuracy: 0.58:  47%|████▋     | 15/32 [01:36<01:48,  6.36s/it]
Test accuracy: 0.6:  47%|████▋     | 15/32 [01:36<01:48,  6.36s/it] 
Test accuracy: 0.61:  47%|████▋     | 15/32 [01:36<01:48,  6.36s/it]
Test accuracy: 0.6:  47%|████▋     | 15/32 [01:36<01:48,  6.36s/it] 
Test accuracy: 0.6:  47%|████▋     | 15/32 [01:36<01:48,  6.36s/it]
Test accuracy: 0.58:  47%|████▋     | 15/32 [01:36<01:48,  6.36s/it]
Test accuracy: 0.58:  50%|█████     | 16/32 [01:42<01:41,  6.36s/it]
Test accuracy: 0.56:  50%|█████     | 16/32 [01:42<01:41,  6.36s/it]
Test accuracy: 0.6:  50%|█████     | 16/32 [01:42<01:41,  6.36s/it]
Test accuracy: 0.6:  50%|█████     | 16/32 [01:42<01:41,  6.36s/it]
Test accuracy: 0.58:  50%|█████     | 16/32 [01:42<01:41,  6.36s/it]
Test accuracy: 0.59:  50%|█████     | 16/32 [01:42<01:41,  6.36s/it]
Test accuracy: 0.61:  50%|█████     | 16/32 [01:42<01:41,  6.36s/it]
Test accuracy: 0.57:  50%|█████     | 16/32 [01:42<01:41,  6.36s/it]
Test accuracy: 0.57:  53%|█████▎    | 17/32 [01:49<01:34,  6.31s/it]
Test accuracy: 0.56:  53%|█████▎    | 17/32 [01:49<01:34,  6.31s/it]
Test accuracy: 0.57:  53%|█████▎    | 17/32 [01:49<01:34,  6.31s/it]
Test accuracy: 0.58:  53%|█████▎    | 17/32 [01:49<01:34,  6.31s/it]
Test accuracy: 0.61:  53%|█████▎    | 17/32 [01:49<01:34,  6.31s/it]
Test accuracy: 0.59:  53%|█████▎    | 17/32 [01:49<01:34,  6.31s/it]
Test accuracy: 0.62:  53%|█████▎    | 17/32 [01:49<01:34,  6.31s/it]
Test accuracy: 0.59:  53%|█████▎    | 17/32 [01:49<01:34,  6.31s/it]
Test accuracy: 0.62:  56%|█████▋    | 18/32 [01:55<01:28,  6.33s/it]
Test accuracy: 0.62:  56%|█████▋    | 18/32 [01:55<01:28,  6.33s/it]
Test accuracy: 0.59:  56%|█████▋    | 18/32 [01:55<01:28,  6.33s/it]
Test accuracy: 0.6:  56%|█████▋    | 18/32 [01:55<01:28,  6.33s/it] 
Test accuracy: 0.57:  56%|█████▋    | 18/32 [01:55<01:28,  6.33s/it]
Test accuracy: 0.56:  56%|█████▋    | 18/32 [01:55<01:28,  6.33s/it]
Test accuracy: 0.58:  56%|█████▋    | 18/32 [01:55<01:28,  6.33s/it]
Test accuracy: 0.58:  56%|█████▋    | 18/32 [01:55<01:28,  6.33s/it]
Test accuracy: 0.62:  59%|█████▉    | 19/32 [02:01<01:22,  6.35s/it]
Test accuracy: 0.58:  59%|█████▉    | 19/32 [02:01<01:22,  6.35s/it]
Test accuracy: 0.61:  59%|█████▉    | 19/32 [02:01<01:22,  6.35s/it]
Test accuracy: 0.58:  59%|█████▉    | 19/32 [02:01<01:22,  6.35s/it]
Test accuracy: 0.56:  59%|█████▉    | 19/32 [02:01<01:22,  6.35s/it]
Test accuracy: 0.6:  59%|█████▉    | 19/32 [02:01<01:22,  6.35s/it]
Test accuracy: 0.56:  59%|█████▉    | 19/32 [02:01<01:22,  6.35s/it]
Test accuracy: 0.59:  59%|█████▉    | 19/32 [02:01<01:22,  6.35s/it]
Test accuracy: 0.58:  62%|██████▎   | 20/32 [02:07<01:15,  6.30s/it]
Test accuracy: 0.56:  62%|██████▎   | 20/32 [02:08<01:15,  6.30s/it]
Test accuracy: 0.56:  62%|██████▎   | 20/32 [02:08<01:15,  6.30s/it]
Test accuracy: 0.59:  62%|██████▎   | 20/32 [02:07<01:15,  6.30s/it]
Test accuracy: 0.59:  62%|██████▎   | 20/32 [02:08<01:15,  6.30s/it]
Test accuracy: 0.61:  62%|██████▎   | 20/32 [02:08<01:15,  6.30s/it]
Test accuracy: 0.56:  62%|██████▎   | 20/32 [02:08<01:15,  6.30s/it]
Test accuracy: 0.61:  62%|██████▎   | 20/32 [02:08<01:15,  6.30s/it]
Test accuracy: 0.61:  66%|██████▌   | 21/32 [02:15<01:11,  6.51s/it]
Test accuracy: 0.6:  66%|██████▌   | 21/32 [02:15<01:11,  6.51s/it] 
Test accuracy: 0.59:  66%|██████▌   | 21/32 [02:14<01:11,  6.51s/it]
Test accuracy: 0.56:  66%|██████▌   | 21/32 [02:15<01:11,  6.51s/it]
Test accuracy: 0.55:  66%|██████▌   | 21/32 [02:15<01:11,  6.51s/it]
Test accuracy: 0.58:  66%|██████▌   | 21/32 [02:14<01:11,  6.51s/it]
Test accuracy: 0.56:  66%|██████▌   | 21/32 [02:15<01:11,  6.52s/it]
Test accuracy: 0.57:  66%|██████▌   | 21/32 [02:15<01:11,  6.52s/it]
Test accuracy: 0.57:  69%|██████▉   | 22/32 [02:21<01:04,  6.47s/it]
Test accuracy: 0.6:  69%|██████▉   | 22/32 [02:21<01:04,  6.48s/it]
Test accuracy: 0.57:  69%|██████▉   | 22/32 [02:21<01:04,  6.47s/it]
Test accuracy: 0.59:  69%|██████▉   | 22/32 [02:21<01:04,  6.48s/it]
Test accuracy: 0.55:  69%|██████▉   | 22/32 [02:21<01:04,  6.48s/it]
Test accuracy: 0.55:  69%|██████▉   | 22/32 [02:21<01:04,  6.48s/it]
Test accuracy: 0.58:  69%|██████▉   | 22/32 [02:21<01:04,  6.48s/it]
Test accuracy: 0.58:  69%|██████▉   | 22/32 [02:21<01:04,  6.48s/it]
Test accuracy: 0.55:  72%|███████▏  | 23/32 [02:27<00:58,  6.45s/it]
Test accuracy: 0.57:  72%|███████▏  | 23/32 [02:27<00:58,  6.45s/it]
Test accuracy: 0.55:  72%|███████▏  | 23/32 [02:27<00:58,  6.45s/it]
Test accuracy: 0.56:  72%|███████▏  | 23/32 [02:27<00:58,  6.45s/it]
Test accuracy: 0.59:  72%|███████▏  | 23/32 [02:27<00:58,  6.45s/it]
Test accuracy: 0.59:  72%|███████▏  | 23/32 [02:27<00:58,  6.45s/it]
Test accuracy: 0.58:  72%|███████▏  | 23/32 [02:27<00:58,  6.45s/it]
Test accuracy: 0.59:  72%|███████▏  | 23/32 [02:27<00:58,  6.45s/it]
Test accuracy: 0.56:  75%|███████▌  | 24/32 [02:34<00:52,  6.56s/it]
Test accuracy: 0.56:  75%|███████▌  | 24/32 [02:34<00:52,  6.56s/it]
Test accuracy: 0.55:  75%|███████▌  | 24/32 [02:34<00:52,  6.56s/it]
Test accuracy: 0.59:  75%|███████▌  | 24/32 [02:34<00:52,  6.56s/it]
Test accuracy: 0.56:  75%|███████▌  | 24/32 [02:34<00:52,  6.56s/it]
Test accuracy: 0.59:  75%|███████▌  | 24/32 [02:34<00:52,  6.56s/it]
Test accuracy: 0.59:  75%|███████▌  | 24/32 [02:34<00:52,  6.57s/it]
Test accuracy: 0.6:  75%|███████▌  | 24/32 [02:34<00:52,  6.57s/it] 
Test accuracy: 0.56:  78%|███████▊  | 25/32 [02:41<00:45,  6.51s/it]
Test accuracy: 0.56:  78%|███████▊  | 25/32 [02:41<00:45,  6.51s/it]
Test accuracy: 0.55:  78%|███████▊  | 25/32 [02:41<00:45,  6.51s/it]
Test accuracy: 0.56:  78%|███████▊  | 25/32 [02:41<00:45,  6.51s/it]
Test accuracy: 0.6:  78%|███████▊  | 25/32 [02:40<00:45,  6.51s/it]
Test accuracy: 0.6:  78%|███████▊  | 25/32 [02:40<00:45,  6.51s/it]
Test accuracy: 0.59:  78%|███████▊  | 25/32 [02:41<00:45,  6.51s/it]
Test accuracy: 0.58:  78%|███████▊  | 25/32 [02:41<00:45,  6.51s/it]
Test accuracy: 0.58:  81%|████████▏ | 26/32 [02:47<00:38,  6.47s/it]
Test accuracy: 0.59:  81%|████████▏ | 26/32 [02:47<00:38,  6.47s/it]
Test accuracy: 0.56:  81%|████████▏ | 26/32 [02:47<00:38,  6.47s/it]
Test accuracy: 0.55:  81%|████████▏ | 26/32 [02:47<00:38,  6.47s/it]
Test accuracy: 0.56:  81%|████████▏ | 26/32 [02:47<00:38,  6.47s/it]
Test accuracy: 0.54:  81%|████████▏ | 26/32 [02:47<00:38,  6.47s/it]
Test accuracy: 0.6:  81%|████████▏ | 26/32 [02:47<00:38,  6.47s/it]
Test accuracy: 0.6:  81%|████████▏ | 26/32 [02:47<00:38,  6.47s/it]
Test accuracy: 0.54:  84%|████████▍ | 27/32 [02:53<00:31,  6.39s/it]
Test accuracy: 0.54:  84%|████████▍ | 27/32 [02:53<00:31,  6.39s/it]
Test accuracy: 0.6:  84%|████████▍ | 27/32 [02:53<00:31,  6.39s/it]
Test accuracy: 0.6:  84%|████████▍ | 27/32 [02:53<00:31,  6.39s/it]
Test accuracy: 0.56:  84%|████████▍ | 27/32 [02:53<00:31,  6.39s/it]
Test accuracy: 0.59:  84%|████████▍ | 27/32 [02:53<00:31,  6.39s/it]
Test accuracy: 0.56:  84%|████████▍ | 27/32 [02:53<00:31,  6.39s/it]
Test accuracy: 0.58:  84%|████████▍ | 27/32 [02:53<00:31,  6.39s/it]
Test accuracy: 0.54:  88%|████████▊ | 28/32 [03:00<00:26,  6.56s/it]
Test accuracy: 0.58:  88%|████████▊ | 28/32 [03:00<00:26,  6.56s/it]
Test accuracy: 0.54:  88%|████████▊ | 28/32 [03:00<00:26,  6.56s/it]
Test accuracy: 0.58:  88%|████████▊ | 28/32 [03:00<00:26,  6.56s/it]
Test accuracy: 0.6:  88%|████████▊ | 28/32 [03:00<00:26,  6.56s/it]
Test accuracy: 0.6:  88%|████████▊ | 28/32 [03:00<00:26,  6.56s/it]
Test accuracy: 0.56:  88%|████████▊ | 28/32 [03:00<00:26,  6.57s/it]
Test accuracy: 0.56:  88%|████████▊ | 28/32 [03:00<00:26,  6.57s/it]
Test accuracy: 0.54:  91%|█████████ | 29/32 [03:07<00:19,  6.52s/it]
Test accuracy: 0.54:  91%|█████████ | 29/32 [03:07<00:19,  6.52s/it]
Test accuracy: 0.6:  91%|█████████ | 29/32 [03:06<00:19,  6.52s/it]
Test accuracy: 0.6:  91%|█████████ | 29/32 [03:06<00:19,  6.52s/it]
Test accuracy: 0.56:  91%|█████████ | 29/32 [03:07<00:19,  6.52s/it]
Test accuracy: 0.58:  91%|█████████ | 29/32 [03:07<00:19,  6.52s/it]
Test accuracy: 0.56:  91%|█████████ | 29/32 [03:07<00:19,  6.52s/it]
Test accuracy: 0.58:  91%|█████████ | 29/32 [03:07<00:19,  6.52s/it]
Test accuracy: 0.54:  94%|█████████▍| 30/32 [03:13<00:12,  6.47s/it]
Test accuracy: 0.56:  94%|█████████▍| 30/32 [03:13<00:12,  6.47s/it]
Test accuracy: 0.53:  94%|█████████▍| 30/32 [03:13<00:12,  6.47s/it]
Test accuracy: 0.55:  94%|█████████▍| 30/32 [03:13<00:12,  6.47s/it]
Test accuracy: 0.58:  94%|█████████▍| 30/32 [03:13<00:12,  6.47s/it]
Test accuracy: 0.6:  94%|█████████▍| 30/32 [03:13<00:12,  6.47s/it]
Test accuracy: 0.58:  94%|█████████▍| 30/32 [03:13<00:12,  6.47s/it]
Test accuracy: 0.6:  94%|█████████▍| 30/32 [03:13<00:12,  6.47s/it]
Test accuracy: 0.58:  97%|█████████▋| 31/32 [03:20<00:06,  6.57s/it]
Test accuracy: 0.53:  97%|█████████▋| 31/32 [03:20<00:06,  6.57s/it]
Test accuracy: 0.58:  97%|█████████▋| 31/32 [03:20<00:06,  6.57s/it]
Test accuracy: 0.54:  97%|█████████▋| 31/32 [03:20<00:06,  6.57s/it]
Test accuracy: 0.6:  97%|█████████▋| 31/32 [03:20<00:06,  6.57s/it]
Test accuracy: 0.6:  97%|█████████▋| 31/32 [03:20<00:06,  6.57s/it]
Test accuracy: 0.55:  97%|█████████▋| 31/32 [03:20<00:06,  6.57s/it]
Test accuracy: 0.56:  97%|█████████▋| 31/32 [03:20<00:06,  6.57s/it]
Test accuracy: 0.6: 100%|██████████| 32/32 [03:22<00:00,  5.36s/it]
Test accuracy: 0.6: 100%|██████████| 32/32 [03:22<00:00,  5.36s/it]
Test accuracy: 0.58: 100%|██████████| 32/32 [03:22<00:00,  5.37s/it]
Test accuracy: 0.58: 100%|██████████| 32/32 [03:22<00:00,  5.37s/it]
Test accuracy: 0.54: 100%|██████████| 32/32 [03:22<00:00,  5.37s/it]
Test accuracy: 0.56: 100%|██████████| 32/32 [03:22<00:00,  5.36s/it]
Test accuracy: 0.54: 100%|██████████| 32/32 [03:22<00:00,  5.37s/it]
Test accuracy: 0.56: 100%|██████████| 32/32 [03:22<00:00,  5.36s/it]
Test accuracy: 0.56: 100%|██████████| 32/32 [03:22<00:00,  6.34s/it]
Device 2: Cor=140, CoT=0, Total=250

Test accuracy: 0.58: 100%|██████████| 32/32 [03:22<00:00,  6.34s/it]
Device 1: Cor=145, CoT=0, Total=250

Test accuracy: 0.54: 100%|██████████| 32/32 [03:22<00:00,  6.34s/it]
Device 3: Cor=134, CoT=0, Total=250

Test accuracy: 0.6: 100%|██████████| 32/32 [03:22<00:00,  6.34s/it]
Device 0: Cor=151, CoT=0, Total=250

============================================================
EVALUATION RESULTS (Epoch 3)
============================================================
Config: data_curriculum=ON, thought_curriculum=OFF
thought_stage=10, data_stage=1, n_latent_tokens=10
Training on difficulties: 0-1

Overall accuracy: 570 / 1000 = 0.5700

Per-difficulty accuracy:
  diff=1: 97/100 = 0.9700 (trained)
  diff=2: 58/100 = 0.5800 (untrained)
  diff=3: 50/100 = 0.5000 (untrained)
  diff=4: 49/100 = 0.4900 (untrained)
  diff=5: 49/100 = 0.4900 (untrained)
  diff=6: 47/100 = 0.4700 (untrained)
  diff=7: 60/100 = 0.6000 (untrained)
  diff=8: 50/100 = 0.5000 (untrained)
  diff=9: 53/100 = 0.5300 (untrained)
  diff=10: 57/100 = 0.5700 (untrained)

Trained (diff 0-1): 97/100 = 0.9700
============================================================


*** ADAPTIVE STAGING: acc_trained=0.9700, acc_diff1=0.9700 >= 0.9 ***
*** Advancing to stage 2 ***

Warning: failed to upload checkpoint_4 to HF: (Request ID: Root=1-6a1b68fe-5ef298115eee74b433674140;1168b31d-8889-4b0c-af0d-2219279dadab)

403 Forbidden: You need to setup automatic credit recharge in order to upload more data. You can do so at /settings/billing..
Cannot access content at: https://huggingface.co/seyedparsa/thoughtformer-checkpoints.git/info/lfs/objects/batch.
Make sure your token has the correct permissions.

============================================================
EPOCH 4/100 (thought_stage=10, data_stage=2)
============================================================
  thought_stage=10, c_thought=1, max_difficulty=2

Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):  16%|█▌        | 156/1000 [00:00<00:00, 984.78 examples/s]
Map (num_proc=4):  16%|█▌        | 158/1000 [00:00<00:00, 981.49 examples/s]
Map (num_proc=4):  15%|█▌        | 153/1000 [00:00<00:00, 954.81 examples/s]
Map (num_proc=4):  82%|████████▏ | 819/1000 [00:00<00:00, 3578.02 examples/s]
Map (num_proc=4):  82%|████████▏ | 816/1000 [00:00<00:00, 3087.11 examples/s]
Map (num_proc=4):  82%|████████▏ | 816/1000 [00:00<00:00, 3152.10 examples/s]
Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2452.59 examples/s]

Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2473.10 examples/s]

Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2486.06 examples/s]

Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Map (num_proc=4):  16%|█▌        | 155/1000 [00:00<00:01, 511.84 examples/s]
Map (num_proc=4):  81%|████████  | 807/1000 [00:00<00:00, 2313.57 examples/s]
Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:04, 1982.14 examples/s]
Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:04, 2101.56 examples/s]
Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:04, 1816.37 examples/s]
Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 1230.89 examples/s]

Filter (num_proc=4):  20%|██        | 2000/10000 [00:00<00:02, 3446.26 examples/s]
Filter (num_proc=4):  30%|███       | 3000/10000 [00:00<00:01, 4963.57 examples/s]
Filter (num_proc=4):  30%|███       | 3000/10000 [00:00<00:01, 5296.69 examples/s]
Filter (num_proc=4):  30%|███       | 3000/10000 [00:00<00:01, 4786.27 examples/s]
Filter (num_proc=4):  60%|██████    | 6000/10000 [00:00<00:00, 9215.58 examples/s]
Filter (num_proc=4):  60%|██████    | 6000/10000 [00:00<00:00, 9425.86 examples/s]
Filter (num_proc=4):  45%|████▌     | 4500/10000 [00:00<00:00, 6922.96 examples/s]
Filter (num_proc=4):  80%|████████  | 8000/10000 [00:00<00:00, 11116.85 examples/s]
Filter (num_proc=4):  80%|████████  | 8000/10000 [00:00<00:00, 11299.30 examples/s]
Filter (num_proc=4):  70%|███████   | 7000/10000 [00:01<00:00, 10635.10 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 10889.54 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 11583.06 examples/s]
Filter (num_proc=4):  90%|█████████ | 9000/10000 [00:01<00:00, 12060.72 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 7777.41 examples/s] 

Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 8100.17 examples/s] 

Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 7138.46 examples/s]

Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:06, 1493.06 examples/s]
Filter (num_proc=4):  20%|██        | 2000/10000 [00:00<00:02, 2893.61 examples/s]
Filter (num_proc=4):  30%|███       | 3000/10000 [00:00<00:01, 4190.54 examples/s]
Filter (num_proc=4):  45%|████▌     | 4500/10000 [00:01<00:00, 6307.81 examples/s]
Filter (num_proc=4):  70%|███████   | 7000/10000 [00:01<00:00, 9957.61 examples/s]
Filter (num_proc=4):  85%|████████▌ | 8500/10000 [00:01<00:00, 10357.99 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 10972.54 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 5923.19 examples/s] 

Map (num_proc=4):   0%|          | 0/2000 [00:00<?, ? examples/s]
Map (num_proc=4):   6%|▋         | 130/2000 [00:01<00:15, 123.71 examples/s]
Map (num_proc=4):  25%|██▌       | 500/2000 [00:01<00:03, 455.28 examples/s]
Map (num_proc=4):  32%|███▏      | 635/2000 [00:01<00:03, 427.84 examples/s]
Map (num_proc=4):  50%|█████     | 1000/2000 [00:01<00:01, 668.76 examples/s]
Map (num_proc=4):  57%|█████▋    | 1136/2000 [00:02<00:01, 637.94 examples/s]
Map (num_proc=4):  75%|███████▌  | 1500/2000 [00:02<00:00, 836.46 examples/s]
Map (num_proc=4):  82%|████████▏ | 1636/2000 [00:02<00:00, 677.06 examples/s]
Map (num_proc=4): 100%|██████████| 2000/2000 [00:03<00:00, 856.99 examples/s]
Map (num_proc=4): 100%|██████████| 2000/2000 [00:03<00:00, 571.92 examples/s]

Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 1437.82 examples/s]
Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 1428.65 examples/s]
Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 1387.09 examples/s]
Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2769.78 examples/s]

Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2742.78 examples/s]

Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2698.17 examples/s]

Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 776.38 examples/s]
Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 1466.57 examples/s]

Map (num_proc=4):   0%|          | 0/200 [00:00<?, ? examples/s]
Map (num_proc=4):  25%|██▌       | 50/200 [00:00<00:00, 153.20 examples/s]
Map (num_proc=4):  75%|███████▌  | 150/200 [00:00<00:00, 368.34 examples/s]
Map (num_proc=4): 100%|██████████| 200/200 [00:00<00:00, 237.02 examples/s]

Training Epoch: 5:   0%|          | 0/62 [00:00<?, ?it/s]
Training Epoch: 5:   0%|          | 0/62 [00:00<?, ?it/s]
Training Epoch: 5:   0%|          | 0/62 [00:00<?, ?it/s]
Training Epoch: 5:   0%|          | 0/62 [00:00<?, ?it/s]logging training data
rank 3 | gpu 3 | alloc=4.20 GB | reserved=55.11 GB | max=41.99 GB
rank 2 | gpu 2 | alloc=4.20 GB | reserved=56.95 GB | max=41.98 GB
rank 1 | gpu 1 | alloc=4.20 GB | reserved=47.68 GB | max=41.98 GB
rank 0 | gpu 0 | alloc=4.20 GB | reserved=49.59 GB | max=41.99 GB

Training Epoch: 5/100, batch 0/125 completed (loss: 0.154:   0%|          | 0/62 [00:03<?, ?it/s]
Training Epoch: 5/100, batch 0/125 completed (loss: 0.3186:   0%|          | 0/62 [00:03<?, ?it/s]
Training Epoch: 5/100, batch 0/125 completed (loss: 0.157:   0%|          | 0/62 [00:03<?, ?it/s]
Training Epoch: 5/100, batch 0/125 completed (loss: 0.201:   0%|          | 0/62 [00:03<?, ?it/s]
Training Epoch: 5/100, batch 0/125 completed (loss: 0.157:   2%|▏         | 1/62 [00:06<06:27,  6.36s/it]
Training Epoch: 5/100, batch 0/125 completed (loss: 0.201:   2%|▏         | 1/62 [00:06<06:28,  6.37s/it]
Training Epoch: 5/100, batch 0/125 completed (loss: 0.154:   2%|▏         | 1/62 [00:06<06:28,  6.36s/it]
Training Epoch: 5/100, batch 0/125 completed (loss: 0.3186:   2%|▏         | 1/62 [00:06<06:28,  6.36s/it]
Training Epoch: 5/100, batch 1/125 completed (loss: 0.0968:   2%|▏         | 1/62 [00:06<06:28,  6.36s/it]
Training Epoch: 5/100, batch 1/125 completed (loss: 0.2888:   2%|▏         | 1/62 [00:06<06:28,  6.36s/it]
Training Epoch: 5/100, batch 1/125 completed (loss: 0.091:   2%|▏         | 1/62 [00:06<06:27,  6.36s/it]
Training Epoch: 5/100, batch 1/125 completed (loss: 0.2303:   2%|▏         | 1/62 [00:06<06:28,  6.37s/it]
Training Epoch: 5/100, batch 2/125 completed (loss: 0.5018:   2%|▏         | 1/62 [00:09<06:28,  6.36s/it]
Training Epoch: 5/100, batch 2/125 completed (loss: 0.1409:   2%|▏         | 1/62 [00:09<06:28,  6.36s/it]
Training Epoch: 5/100, batch 2/125 completed (loss: 0.5176:   2%|▏         | 1/62 [00:09<06:27,  6.36s/it]
Training Epoch: 5/100, batch 2/125 completed (loss: 0.173:   2%|▏         | 1/62 [00:09<06:28,  6.37s/it] 
Training Epoch: 5/100, batch 2/125 completed (loss: 0.5176:   3%|▎         | 2/62 [00:12<06:01,  6.03s/it]
Training Epoch: 5/100, batch 2/125 completed (loss: 0.173:   3%|▎         | 2/62 [00:12<06:02,  6.04s/it]
Training Epoch: 5/100, batch 2/125 completed (loss: 0.5018:   3%|▎         | 2/62 [00:12<06:01,  6.03s/it]
Training Epoch: 5/100, batch 2/125 completed (loss: 0.1409:   3%|▎         | 2/62 [00:12<06:02,  6.03s/it]
Training Epoch: 5/100, batch 3/125 completed (loss: 0.2174:   3%|▎         | 2/62 [00:12<06:01,  6.03s/it]
Training Epoch: 5/100, batch 3/125 completed (loss: 0.1826:   3%|▎         | 2/62 [00:12<06:01,  6.03s/it]
Training Epoch: 5/100, batch 3/125 completed (loss: 0.1586:   3%|▎         | 2/62 [00:12<06:02,  6.03s/it]
Training Epoch: 5/100, batch 3/125 completed (loss: 0.0826:   3%|▎         | 2/62 [00:12<06:02,  6.04s/it]
Training Epoch: 5/100, batch 4/125 completed (loss: 0.2628:   3%|▎         | 2/62 [00:15<06:02,  6.03s/it]
Training Epoch: 5/100, batch 4/125 completed (loss: 0.2287:   3%|▎         | 2/62 [00:15<06:01,  6.03s/it]
Training Epoch: 5/100, batch 4/125 completed (loss: 0.1049:   3%|▎         | 2/62 [00:15<06:01,  6.03s/it]
Training Epoch: 5/100, batch 4/125 completed (loss: 0.1734:   3%|▎         | 2/62 [00:15<06:02,  6.04s/it]
Training Epoch: 5/100, batch 4/125 completed (loss: 0.1734:   5%|▍         | 3/62 [00:18<05:55,  6.02s/it]
Training Epoch: 5/100, batch 4/125 completed (loss: 0.2287:   5%|▍         | 3/62 [00:18<05:55,  6.02s/it]
Training Epoch: 5/100, batch 4/125 completed (loss: 0.1049:   5%|▍         | 3/62 [00:18<05:55,  6.02s/it]
Training Epoch: 5/100, batch 4/125 completed (loss: 0.2628:   5%|▍         | 3/62 [00:18<05:55,  6.02s/it]
Training Epoch: 5/100, batch 5/125 completed (loss: 0.1614:   5%|▍         | 3/62 [00:18<05:55,  6.02s/it]
Training Epoch: 5/100, batch 5/125 completed (loss: 0.2842:   5%|▍         | 3/62 [00:18<05:55,  6.02s/it]
Training Epoch: 5/100, batch 5/125 completed (loss: 0.212:   5%|▍         | 3/62 [00:18<05:55,  6.02s/it] 
Training Epoch: 5/100, batch 5/125 completed (loss: 0.149:   5%|▍         | 3/62 [00:18<05:55,  6.02s/it] 
Training Epoch: 5/100, batch 6/125 completed (loss: 0.1795:   5%|▍         | 3/62 [00:21<05:55,  6.02s/it]
Training Epoch: 5/100, batch 6/125 completed (loss: 0.3674:   5%|▍         | 3/62 [00:21<05:55,  6.02s/it]
Training Epoch: 5/100, batch 6/125 completed (loss: 0.2344:   5%|▍         | 3/62 [00:21<05:55,  6.02s/it]
Training Epoch: 5/100, batch 6/125 completed (loss: 0.0874:   5%|▍         | 3/62 [00:21<05:55,  6.02s/it]
Training Epoch: 5/100, batch 6/125 completed (loss: 0.1795:   6%|▋         | 4/62 [00:23<05:39,  5.86s/it]
Training Epoch: 5/100, batch 6/125 completed (loss: 0.0874:   6%|▋         | 4/62 [00:23<05:39,  5.86s/it]
Training Epoch: 5/100, batch 6/125 completed (loss: 0.2344:   6%|▋         | 4/62 [00:23<05:39,  5.86s/it]
Training Epoch: 5/100, batch 6/125 completed (loss: 0.3674:   6%|▋         | 4/62 [00:23<05:40,  5.86s/it]
Training Epoch: 5/100, batch 7/125 completed (loss: 0.3074:   6%|▋         | 4/62 [00:23<05:39,  5.86s/it]
Training Epoch: 5/100, batch 7/125 completed (loss: 0.1029:   6%|▋         | 4/62 [00:23<05:40,  5.86s/it]
Training Epoch: 5/100, batch 7/125 completed (loss: 0.2582:   6%|▋         | 4/62 [00:23<05:39,  5.86s/it]
Training Epoch: 5/100, batch 7/125 completed (loss: 0.1424:   6%|▋         | 4/62 [00:23<05:39,  5.86s/it]
Training Epoch: 5/100, batch 8/125 completed (loss: 0.1188:   6%|▋         | 4/62 [00:26<05:40,  5.86s/it]
Training Epoch: 5/100, batch 8/125 completed (loss: 0.2848:   6%|▋         | 4/62 [00:26<05:39,  5.86s/it]
Training Epoch: 5/100, batch 8/125 completed (loss: 0.2386:   6%|▋         | 4/62 [00:26<05:39,  5.86s/it]
Training Epoch: 5/100, batch 8/125 completed (loss: 0.0757:   6%|▋         | 4/62 [00:26<05:39,  5.86s/it]
Training Epoch: 5/100, batch 8/125 completed (loss: 0.2848:   8%|▊         | 5/62 [00:29<05:35,  5.88s/it]
Training Epoch: 5/100, batch 8/125 completed (loss: 0.1188:   8%|▊         | 5/62 [00:29<05:35,  5.88s/it]
Training Epoch: 5/100, batch 8/125 completed (loss: 0.0757:   8%|▊         | 5/62 [00:29<05:35,  5.89s/it]
Training Epoch: 5/100, batch 8/125 completed (loss: 0.2386:   8%|▊         | 5/62 [00:29<05:35,  5.88s/it]
Training Epoch: 5/100, batch 9/125 completed (loss: 0.0712:   8%|▊         | 5/62 [00:29<05:35,  5.88s/it]
Training Epoch: 5/100, batch 9/125 completed (loss: 0.1917:   8%|▊         | 5/62 [00:29<05:35,  5.88s/it]
Training Epoch: 5/100, batch 9/125 completed (loss: 0.4406:   8%|▊         | 5/62 [00:29<05:35,  5.88s/it]
Training Epoch: 5/100, batch 9/125 completed (loss: 0.3356:   8%|▊         | 5/62 [00:29<05:35,  5.89s/it]
Training Epoch: 5/100, batch 10/125 completed (loss: 0.4037:   8%|▊         | 5/62 [00:32<05:35,  5.88s/it]
Training Epoch: 5/100, batch 10/125 completed (loss: 0.0684:   8%|▊         | 5/62 [00:32<05:35,  5.88s/it]
Training Epoch: 5/100, batch 10/125 completed (loss: 0.1695:   8%|▊         | 5/62 [00:32<05:35,  5.88s/it]
Training Epoch: 5/100, batch 10/125 completed (loss: 0.211:   8%|▊         | 5/62 [00:32<05:35,  5.89s/it]
Training Epoch: 5/100, batch 10/125 completed (loss: 0.4037:  10%|▉         | 6/62 [00:35<05:30,  5.90s/it]
Training Epoch: 5/100, batch 10/125 completed (loss: 0.0684:  10%|▉         | 6/62 [00:35<05:30,  5.90s/it]
Training Epoch: 5/100, batch 10/125 completed (loss: 0.1695:  10%|▉         | 6/62 [00:35<05:30,  5.90s/it]
Training Epoch: 5/100, batch 10/125 completed (loss: 0.211:  10%|▉         | 6/62 [00:35<05:30,  5.91s/it]
Training Epoch: 5/100, batch 11/125 completed (loss: 0.0998:  10%|▉         | 6/62 [00:35<05:30,  5.90s/it]
Training Epoch: 5/100, batch 11/125 completed (loss: 0.2409:  10%|▉         | 6/62 [00:35<05:30,  5.90s/it]
Training Epoch: 5/100, batch 11/125 completed (loss: 0.4149:  10%|▉         | 6/62 [00:35<05:30,  5.90s/it]
Training Epoch: 5/100, batch 11/125 completed (loss: 0.2724:  10%|▉         | 6/62 [00:35<05:30,  5.91s/it]
Training Epoch: 5/100, batch 12/125 completed (loss: 0.5368:  10%|▉         | 6/62 [00:38<05:30,  5.90s/it]
Training Epoch: 5/100, batch 12/125 completed (loss: 0.2652:  10%|▉         | 6/62 [00:38<05:30,  5.90s/it]
Training Epoch: 5/100, batch 12/125 completed (loss: 0.4259:  10%|▉         | 6/62 [00:38<05:30,  5.90s/it]
Training Epoch: 5/100, batch 12/125 completed (loss: 0.3537:  10%|▉         | 6/62 [00:38<05:30,  5.91s/it]
Training Epoch: 5/100, batch 12/125 completed (loss: 0.2652:  11%|█▏        | 7/62 [00:41<05:22,  5.87s/it]
Training Epoch: 5/100, batch 12/125 completed (loss: 0.3537:  11%|█▏        | 7/62 [00:41<05:22,  5.87s/it]
Training Epoch: 5/100, batch 12/125 completed (loss: 0.5368:  11%|█▏        | 7/62 [00:41<05:22,  5.87s/it]
Training Epoch: 5/100, batch 12/125 completed (loss: 0.4259:  11%|█▏        | 7/62 [00:41<05:23,  5.87s/it]
Training Epoch: 5/100, batch 13/125 completed (loss: 0.1162:  11%|█▏        | 7/62 [00:41<05:23,  5.87s/it]
Training Epoch: 5/100, batch 13/125 completed (loss: 0.2063:  11%|█▏        | 7/62 [00:41<05:22,  5.87s/it]
Training Epoch: 5/100, batch 13/125 completed (loss: 0.0998:  11%|█▏        | 7/62 [00:41<05:22,  5.87s/it]
Training Epoch: 5/100, batch 13/125 completed (loss: 0.0951:  11%|█▏        | 7/62 [00:41<05:22,  5.87s/it]
Training Epoch: 5/100, batch 14/125 completed (loss: 0.1198:  11%|█▏        | 7/62 [00:44<05:22,  5.87s/it]
Training Epoch: 5/100, batch 14/125 completed (loss: 0.2946:  11%|█▏        | 7/62 [00:44<05:23,  5.87s/it]
Training Epoch: 5/100, batch 14/125 completed (loss: 0.2175:  11%|█▏        | 7/62 [00:44<05:22,  5.87s/it]
Training Epoch: 5/100, batch 14/125 completed (loss: 0.0361:  11%|█▏        | 7/62 [00:44<05:22,  5.87s/it]
Training Epoch: 5/100, batch 14/125 completed (loss: 0.1198:  13%|█▎        | 8/62 [00:47<05:19,  5.91s/it]
Training Epoch: 5/100, batch 14/125 completed (loss: 0.0361:  13%|█▎        | 8/62 [00:47<05:19,  5.91s/it]
Training Epoch: 5/100, batch 14/125 completed (loss: 0.2175:  13%|█▎        | 8/62 [00:47<05:19,  5.91s/it]
Training Epoch: 5/100, batch 14/125 completed (loss: 0.2946:  13%|█▎        | 8/62 [00:47<05:19,  5.91s/it]
Training Epoch: 5/100, batch 15/125 completed (loss: 0.1556:  13%|█▎        | 8/62 [00:47<05:19,  5.91s/it]
Training Epoch: 5/100, batch 15/125 completed (loss: 0.1369:  13%|█▎        | 8/62 [00:47<05:19,  5.91s/it]
Training Epoch: 5/100, batch 15/125 completed (loss: 0.1683:  13%|█▎        | 8/62 [00:47<05:19,  5.91s/it]
Training Epoch: 5/100, batch 15/125 completed (loss: 0.1586:  13%|█▎        | 8/62 [00:47<05:19,  5.91s/it]
Training Epoch: 5/100, batch 16/125 completed (loss: 0.0911:  13%|█▎        | 8/62 [00:50<05:19,  5.91s/it]
Training Epoch: 5/100, batch 16/125 completed (loss: 0.1965:  13%|█▎        | 8/62 [00:50<05:19,  5.91s/it]
Training Epoch: 5/100, batch 16/125 completed (loss: 0.1752:  13%|█▎        | 8/62 [00:50<05:19,  5.91s/it]
Training Epoch: 5/100, batch 16/125 completed (loss: 0.1084:  13%|█▎        | 8/62 [00:50<05:19,  5.91s/it]
Training Epoch: 5/100, batch 16/125 completed (loss: 0.1084:  15%|█▍        | 9/62 [00:53<05:09,  5.85s/it]
Training Epoch: 5/100, batch 16/125 completed (loss: 0.1965:  15%|█▍        | 9/62 [00:53<05:09,  5.85s/it]
Training Epoch: 5/100, batch 16/125 completed (loss: 0.1752:  15%|█▍        | 9/62 [00:53<05:09,  5.85s/it]
Training Epoch: 5/100, batch 16/125 completed (loss: 0.0911:  15%|█▍        | 9/62 [00:53<05:10,  5.85s/it]
Training Epoch: 5/100, batch 17/125 completed (loss: 0.2312:  15%|█▍        | 9/62 [00:53<05:09,  5.85s/it]
Training Epoch: 5/100, batch 17/125 completed (loss: 0.1158:  15%|█▍        | 9/62 [00:53<05:09,  5.85s/it]
Training Epoch: 5/100, batch 17/125 completed (loss: 0.0349:  15%|█▍        | 9/62 [00:53<05:10,  5.85s/it]
Training Epoch: 5/100, batch 17/125 completed (loss: 0.2616:  15%|█▍        | 9/62 [00:53<05:09,  5.85s/it]
Training Epoch: 5/100, batch 18/125 completed (loss: 0.1168:  15%|█▍        | 9/62 [00:56<05:09,  5.85s/it]
Training Epoch: 5/100, batch 18/125 completed (loss: 0.0988:  15%|█▍        | 9/62 [00:56<05:10,  5.85s/it]
Training Epoch: 5/100, batch 18/125 completed (loss: 0.113:  15%|█▍        | 9/62 [00:56<05:09,  5.85s/it] 
Training Epoch: 5/100, batch 18/125 completed (loss: 0.1255:  15%|█▍        | 9/62 [00:56<05:09,  5.85s/it]
Training Epoch: 5/100, batch 18/125 completed (loss: 0.0988:  16%|█▌        | 10/62 [00:59<05:04,  5.85s/it]
Training Epoch: 5/100, batch 18/125 completed (loss: 0.1255:  16%|█▌        | 10/62 [00:59<05:04,  5.86s/it]
Training Epoch: 5/100, batch 18/125 completed (loss: 0.113:  16%|█▌        | 10/62 [00:59<05:04,  5.86s/it]
Training Epoch: 5/100, batch 18/125 completed (loss: 0.1168:  16%|█▌        | 10/62 [00:59<05:04,  5.86s/it]
Training Epoch: 5/100, batch 19/125 completed (loss: 0.37:  16%|█▌        | 10/62 [00:59<05:04,  5.86s/it]  
Training Epoch: 5/100, batch 19/125 completed (loss: 0.2009:  16%|█▌        | 10/62 [00:59<05:04,  5.85s/it]
Training Epoch: 5/100, batch 19/125 completed (loss: 0.2621:  16%|█▌        | 10/62 [00:59<05:04,  5.86s/it]
Training Epoch: 5/100, batch 19/125 completed (loss: 0.1538:  16%|█▌        | 10/62 [00:59<05:04,  5.86s/it]
Training Epoch: 5/100, batch 20/125 completed (loss: 0.2473:  16%|█▌        | 10/62 [01:02<05:04,  5.86s/it]
Training Epoch: 5/100, batch 20/125 completed (loss: 0.0669:  16%|█▌        | 10/62 [01:02<05:04,  5.86s/it]
Training Epoch: 5/100, batch 20/125 completed (loss: 0.4:  16%|█▌        | 10/62 [01:02<05:04,  5.85s/it]   
Training Epoch: 5/100, batch 20/125 completed (loss: 0.2055:  16%|█▌        | 10/62 [01:02<05:04,  5.86s/it]
Training Epoch: 5/100, batch 20/125 completed (loss: 0.2055:  18%|█▊        | 11/62 [01:05<05:01,  5.91s/it]
Training Epoch: 5/100, batch 20/125 completed (loss: 0.4:  18%|█▊        | 11/62 [01:05<05:01,  5.91s/it]
Training Epoch: 5/100, batch 20/125 completed (loss: 0.2473:  18%|█▊        | 11/62 [01:05<05:01,  5.91s/it]
Training Epoch: 5/100, batch 20/125 completed (loss: 0.0669:  18%|█▊        | 11/62 [01:05<05:01,  5.91s/it]
Training Epoch: 5/100, batch 21/125 completed (loss: 0.1068:  18%|█▊        | 11/62 [01:05<05:01,  5.91s/it]
Training Epoch: 5/100, batch 21/125 completed (loss: 0.1151:  18%|█▊        | 11/62 [01:05<05:01,  5.91s/it]
Training Epoch: 5/100, batch 21/125 completed (loss: 0.121:  18%|█▊        | 11/62 [01:05<05:01,  5.91s/it]
Training Epoch: 5/100, batch 21/125 completed (loss: 0.3245:  18%|█▊        | 11/62 [01:05<05:01,  5.91s/it]
Training Epoch: 5/100, batch 22/125 completed (loss: 0.0755:  18%|█▊        | 11/62 [01:07<05:01,  5.91s/it]
Training Epoch: 5/100, batch 22/125 completed (loss: 0.1349:  18%|█▊        | 11/62 [01:07<05:01,  5.91s/it]
Training Epoch: 5/100, batch 22/125 completed (loss: 0.3199:  18%|█▊        | 11/62 [01:07<05:01,  5.91s/it]
Training Epoch: 5/100, batch 22/125 completed (loss: 0.0578:  18%|█▊        | 11/62 [01:08<05:01,  5.91s/it]
Training Epoch: 5/100, batch 22/125 completed (loss: 0.0755:  19%|█▉        | 12/62 [01:10<04:52,  5.86s/it]
Training Epoch: 5/100, batch 22/125 completed (loss: 0.1349:  19%|█▉        | 12/62 [01:10<04:52,  5.86s/it]
Training Epoch: 5/100, batch 22/125 completed (loss: 0.0578:  19%|█▉        | 12/62 [01:10<04:52,  5.86s/it]
Training Epoch: 5/100, batch 22/125 completed (loss: 0.3199:  19%|█▉        | 12/62 [01:10<04:52,  5.86s/it]
Training Epoch: 5/100, batch 23/125 completed (loss: 0.252:  19%|█▉        | 12/62 [01:10<04:52,  5.86s/it] 
Training Epoch: 5/100, batch 23/125 completed (loss: 0.4445:  19%|█▉        | 12/62 [01:10<04:52,  5.86s/it]
Training Epoch: 5/100, batch 23/125 completed (loss: 0.1316:  19%|█▉        | 12/62 [01:10<04:52,  5.86s/it]
Training Epoch: 5/100, batch 23/125 completed (loss: 0.1937:  19%|█▉        | 12/62 [01:10<04:52,  5.86s/it]
Training Epoch: 5/100, batch 24/125 completed (loss: 0.3348:  19%|█▉        | 12/62 [01:13<04:52,  5.86s/it]
Training Epoch: 5/100, batch 24/125 completed (loss: 0.0587:  19%|█▉        | 12/62 [01:13<04:52,  5.86s/it]
Training Epoch: 5/100, batch 24/125 completed (loss: 0.0731:  19%|█▉        | 12/62 [01:13<04:52,  5.86s/it]
Training Epoch: 5/100, batch 24/125 completed (loss: 0.1052:  19%|█▉        | 12/62 [01:13<04:52,  5.86s/it]
Training Epoch: 5/100, batch 24/125 completed (loss: 0.1052:  21%|██        | 13/62 [01:17<04:52,  5.96s/it]
Training Epoch: 5/100, batch 24/125 completed (loss: 0.0731:  21%|██        | 13/62 [01:17<04:52,  5.96s/it]
Training Epoch: 5/100, batch 24/125 completed (loss: 0.0587:  21%|██        | 13/62 [01:17<04:52,  5.96s/it]
Training Epoch: 5/100, batch 24/125 completed (loss: 0.3348:  21%|██        | 13/62 [01:17<04:52,  5.96s/it]
Training Epoch: 5/100, batch 25/125 completed (loss: 0.1091:  21%|██        | 13/62 [01:17<04:52,  5.96s/it]
Training Epoch: 5/100, batch 25/125 completed (loss: 0.3343:  21%|██        | 13/62 [01:17<04:52,  5.96s/it]
Training Epoch: 5/100, batch 25/125 completed (loss: 0.2136:  21%|██        | 13/62 [01:17<04:52,  5.96s/it]
Training Epoch: 5/100, batch 25/125 completed (loss: 0.093:  21%|██        | 13/62 [01:17<04:52,  5.96s/it] 
Training Epoch: 5/100, batch 26/125 completed (loss: 0.3106:  21%|██        | 13/62 [01:19<04:52,  5.96s/it]
Training Epoch: 5/100, batch 26/125 completed (loss: 0.3235:  21%|██        | 13/62 [01:19<04:52,  5.96s/it]
Training Epoch: 5/100, batch 26/125 completed (loss: 0.1673:  21%|██        | 13/62 [01:19<04:52,  5.96s/it]
Training Epoch: 5/100, batch 26/125 completed (loss: 0.3008:  21%|██        | 13/62 [01:19<04:52,  5.96s/it]
Training Epoch: 5/100, batch 26/125 completed (loss: 0.3008:  23%|██▎       | 14/62 [01:22<04:43,  5.90s/it]
Training Epoch: 5/100, batch 26/125 completed (loss: 0.1673:  23%|██▎       | 14/62 [01:22<04:43,  5.90s/it]
Training Epoch: 5/100, batch 26/125 completed (loss: 0.3235:  23%|██▎       | 14/62 [01:22<04:43,  5.90s/it]
Training Epoch: 5/100, batch 26/125 completed (loss: 0.3106:  23%|██▎       | 14/62 [01:22<04:43,  5.90s/it]
Training Epoch: 5/100, batch 27/125 completed (loss: 0.0584:  23%|██▎       | 14/62 [01:22<04:43,  5.90s/it]
Training Epoch: 5/100, batch 27/125 completed (loss: 0.2258:  23%|██▎       | 14/62 [01:22<04:43,  5.90s/it]
Training Epoch: 5/100, batch 27/125 completed (loss: 0.2258:  23%|██▎       | 14/62 [01:22<04:43,  5.90s/it]
Training Epoch: 5/100, batch 27/125 completed (loss: 0.1549:  23%|██▎       | 14/62 [01:22<04:43,  5.90s/it]
Training Epoch: 5/100, batch 28/125 completed (loss: 0.1153:  23%|██▎       | 14/62 [01:25<04:43,  5.90s/it]
Training Epoch: 5/100, batch 28/125 completed (loss: 0.2083:  23%|██▎       | 14/62 [01:25<04:43,  5.90s/it]
Training Epoch: 5/100, batch 28/125 completed (loss: 0.1594:  23%|██▎       | 14/62 [01:25<04:43,  5.90s/it]
Training Epoch: 5/100, batch 28/125 completed (loss: 0.1493:  23%|██▎       | 14/62 [01:25<04:43,  5.90s/it]
Training Epoch: 5/100, batch 28/125 completed (loss: 0.1153:  24%|██▍       | 15/62 [01:28<04:34,  5.84s/it]
Training Epoch: 5/100, batch 28/125 completed (loss: 0.1493:  24%|██▍       | 15/62 [01:28<04:34,  5.84s/it]
Training Epoch: 5/100, batch 28/125 completed (loss: 0.2083:  24%|██▍       | 15/62 [01:28<04:34,  5.84s/it]
Training Epoch: 5/100, batch 28/125 completed (loss: 0.1594:  24%|██▍       | 15/62 [01:28<04:34,  5.85s/it]
Training Epoch: 5/100, batch 29/125 completed (loss: 0.3077:  24%|██▍       | 15/62 [01:28<04:34,  5.84s/it]
Training Epoch: 5/100, batch 29/125 completed (loss: 0.251:  24%|██▍       | 15/62 [01:28<04:34,  5.85s/it] 
Training Epoch: 5/100, batch 29/125 completed (loss: 0.0743:  24%|██▍       | 15/62 [01:28<04:34,  5.84s/it]
Training Epoch: 5/100, batch 29/125 completed (loss: 0.1526:  24%|██▍       | 15/62 [01:28<04:34,  5.84s/it]
Training Epoch: 5/100, batch 30/125 completed (loss: 0.2202:  24%|██▍       | 15/62 [01:31<04:34,  5.84s/it]
Training Epoch: 5/100, batch 30/125 completed (loss: 0.085:  24%|██▍       | 15/62 [01:31<04:34,  5.85s/it]
Training Epoch: 5/100, batch 30/125 completed (loss: 0.4666:  24%|██▍       | 15/62 [01:31<04:34,  5.84s/it]
Training Epoch: 5/100, batch 30/125 completed (loss: 0.1431:  24%|██▍       | 15/62 [01:31<04:34,  5.84s/it]
Training Epoch: 5/100, batch 30/125 completed (loss: 0.1431:  26%|██▌       | 16/62 [01:34<04:34,  5.96s/it]
Training Epoch: 5/100, batch 30/125 completed (loss: 0.2202:  26%|██▌       | 16/62 [01:34<04:34,  5.96s/it]
Training Epoch: 5/100, batch 30/125 completed (loss: 0.4666:  26%|██▌       | 16/62 [01:34<04:34,  5.96s/it]
Training Epoch: 5/100, batch 30/125 completed (loss: 0.085:  26%|██▌       | 16/62 [01:34<04:34,  5.96s/it]
Training Epoch: 5/100, batch 31/125 completed (loss: 0.107:  26%|██▌       | 16/62 [01:34<04:34,  5.96s/it] 
Training Epoch: 5/100, batch 31/125 completed (loss: 0.1042:  26%|██▌       | 16/62 [01:34<04:34,  5.96s/it]
Training Epoch: 5/100, batch 31/125 completed (loss: 0.0743:  26%|██▌       | 16/62 [01:34<04:34,  5.96s/it]
Training Epoch: 5/100, batch 31/125 completed (loss: 0.1886:  26%|██▌       | 16/62 [01:34<04:34,  5.96s/it]
Training Epoch: 5/100, batch 32/125 completed (loss: 0.1502:  26%|██▌       | 16/62 [01:37<04:34,  5.96s/it]
Training Epoch: 5/100, batch 32/125 completed (loss: 0.1502:  26%|██▌       | 16/62 [01:37<04:34,  5.96s/it]
Training Epoch: 5/100, batch 32/125 completed (loss: 0.1775:  26%|██▌       | 16/62 [01:37<04:34,  5.96s/it]
Training Epoch: 5/100, batch 32/125 completed (loss: 0.1524:  26%|██▌       | 16/62 [01:37<04:34,  5.96s/it]
Training Epoch: 5/100, batch 32/125 completed (loss: 0.1524:  27%|██▋       | 17/62 [01:40<04:25,  5.89s/it]
Training Epoch: 5/100, batch 32/125 completed (loss: 0.1502:  27%|██▋       | 17/62 [01:40<04:25,  5.89s/it]
Training Epoch: 5/100, batch 32/125 completed (loss: 0.1502:  27%|██▋       | 17/62 [01:40<04:25,  5.89s/it]
Training Epoch: 5/100, batch 32/125 completed (loss: 0.1775:  27%|██▋       | 17/62 [01:40<04:25,  5.89s/it]
Training Epoch: 5/100, batch 33/125 completed (loss: 0.2578:  27%|██▋       | 17/62 [01:40<04:25,  5.89s/it]
Training Epoch: 5/100, batch 33/125 completed (loss: 0.1754:  27%|██▋       | 17/62 [01:40<04:25,  5.89s/it]
Training Epoch: 5/100, batch 33/125 completed (loss: 0.0325:  27%|██▋       | 17/62 [01:40<04:25,  5.89s/it]
Training Epoch: 5/100, batch 33/125 completed (loss: 0.1087:  27%|██▋       | 17/62 [01:40<04:25,  5.89s/it]
Training Epoch: 5/100, batch 34/125 completed (loss: 0.1191:  27%|██▋       | 17/62 [01:43<04:25,  5.89s/it]
Training Epoch: 5/100, batch 34/125 completed (loss: 0.3751:  27%|██▋       | 17/62 [01:43<04:25,  5.89s/it]
Training Epoch: 5/100, batch 34/125 completed (loss: 0.2322:  27%|██▋       | 17/62 [01:43<04:25,  5.89s/it]
Training Epoch: 5/100, batch 34/125 completed (loss: 0.2838:  27%|██▋       | 17/62 [01:43<04:25,  5.89s/it]
Training Epoch: 5/100, batch 34/125 completed (loss: 0.2838:  29%|██▉       | 18/62 [01:46<04:19,  5.90s/it]
Training Epoch: 5/100, batch 34/125 completed (loss: 0.1191:  29%|██▉       | 18/62 [01:46<04:19,  5.90s/it]
Training Epoch: 5/100, batch 34/125 completed (loss: 0.3751:  29%|██▉       | 18/62 [01:46<04:19,  5.90s/it]
Training Epoch: 5/100, batch 34/125 completed (loss: 0.2322:  29%|██▉       | 18/62 [01:46<04:19,  5.90s/it]
Training Epoch: 5/100, batch 35/125 completed (loss: 0.1219:  29%|██▉       | 18/62 [01:46<04:19,  5.90s/it]
Training Epoch: 5/100, batch 35/125 completed (loss: 0.3976:  29%|██▉       | 18/62 [01:46<04:19,  5.90s/it]
Training Epoch: 5/100, batch 35/125 completed (loss: 0.3561:  29%|██▉       | 18/62 [01:46<04:19,  5.90s/it]
Training Epoch: 5/100, batch 35/125 completed (loss: 0.1358:  29%|██▉       | 18/62 [01:46<04:19,  5.90s/it]
Training Epoch: 5/100, batch 36/125 completed (loss: 0.1423:  29%|██▉       | 18/62 [01:49<04:19,  5.90s/it]
Training Epoch: 5/100, batch 36/125 completed (loss: 0.2122:  29%|██▉       | 18/62 [01:49<04:19,  5.90s/it]
Training Epoch: 5/100, batch 36/125 completed (loss: 0.2034:  29%|██▉       | 18/62 [01:49<04:19,  5.90s/it]
Training Epoch: 5/100, batch 36/125 completed (loss: 0.2169:  29%|██▉       | 18/62 [01:49<04:19,  5.90s/it]
Training Epoch: 5/100, batch 36/125 completed (loss: 0.2169:  31%|███       | 19/62 [01:52<04:16,  5.97s/it]
Training Epoch: 5/100, batch 36/125 completed (loss: 0.2034:  31%|███       | 19/62 [01:52<04:16,  5.97s/it]
Training Epoch: 5/100, batch 36/125 completed (loss: 0.2122:  31%|███       | 19/62 [01:52<04:16,  5.97s/it]
Training Epoch: 5/100, batch 36/125 completed (loss: 0.1423:  31%|███       | 19/62 [01:52<04:16,  5.97s/it]
Training Epoch: 5/100, batch 37/125 completed (loss: 0.1662:  31%|███       | 19/62 [01:52<04:16,  5.97s/it]
Training Epoch: 5/100, batch 37/125 completed (loss: 0.0755:  31%|███       | 19/62 [01:52<04:16,  5.97s/it]
Training Epoch: 5/100, batch 37/125 completed (loss: 0.1506:  31%|███       | 19/62 [01:52<04:16,  5.97s/it]
Training Epoch: 5/100, batch 37/125 completed (loss: 0.0696:  31%|███       | 19/62 [01:52<04:16,  5.97s/it]
Training Epoch: 5/100, batch 38/125 completed (loss: 0.2735:  31%|███       | 19/62 [01:55<04:16,  5.97s/it]
Training Epoch: 5/100, batch 38/125 completed (loss: 0.3272:  31%|███       | 19/62 [01:55<04:16,  5.97s/it]
Training Epoch: 5/100, batch 38/125 completed (loss: 0.1876:  31%|███       | 19/62 [01:55<04:16,  5.97s/it]
Training Epoch: 5/100, batch 38/125 completed (loss: 0.3299:  31%|███       | 19/62 [01:55<04:16,  5.97s/it]
Training Epoch: 5/100, batch 38/125 completed (loss: 0.1876:  32%|███▏      | 20/62 [01:58<04:08,  5.92s/it]
Training Epoch: 5/100, batch 38/125 completed (loss: 0.3299:  32%|███▏      | 20/62 [01:58<04:08,  5.92s/it]
Training Epoch: 5/100, batch 38/125 completed (loss: 0.3272:  32%|███▏      | 20/62 [01:58<04:08,  5.92s/it]
Training Epoch: 5/100, batch 38/125 completed (loss: 0.2735:  32%|███▏      | 20/62 [01:58<04:08,  5.92s/it]
Training Epoch: 5/100, batch 39/125 completed (loss: 0.5002:  32%|███▏      | 20/62 [01:58<04:08,  5.92s/it]
Training Epoch: 5/100, batch 39/125 completed (loss: 0.1538:  32%|███▏      | 20/62 [01:58<04:08,  5.92s/it]
Training Epoch: 5/100, batch 39/125 completed (loss: 0.1539:  32%|███▏      | 20/62 [01:58<04:08,  5.92s/it]
Training Epoch: 5/100, batch 39/125 completed (loss: 0.1849:  32%|███▏      | 20/62 [01:58<04:08,  5.92s/it]
Training Epoch: 5/100, batch 40/125 completed (loss: 0.1589:  32%|███▏      | 20/62 [02:01<04:08,  5.92s/it]
Training Epoch: 5/100, batch 40/125 completed (loss: 0.359:  32%|███▏      | 20/62 [02:01<04:08,  5.92s/it] 
Training Epoch: 5/100, batch 40/125 completed (loss: 0.5842:  32%|███▏      | 20/62 [02:01<04:08,  5.92s/it]
Training Epoch: 5/100, batch 40/125 completed (loss: 0.1308:  32%|███▏      | 20/62 [02:01<04:08,  5.92s/it]
Training Epoch: 5/100, batch 40/125 completed (loss: 0.1308:  34%|███▍      | 21/62 [02:04<04:07,  6.03s/it]
Training Epoch: 5/100, batch 40/125 completed (loss: 0.1589:  34%|███▍      | 21/62 [02:04<04:07,  6.04s/it]
Training Epoch: 5/100, batch 40/125 completed (loss: 0.5842:  34%|███▍      | 21/62 [02:04<04:07,  6.03s/it]
Training Epoch: 5/100, batch 40/125 completed (loss: 0.359:  34%|███▍      | 21/62 [02:04<04:07,  6.04s/it]
Training Epoch: 5/100, batch 41/125 completed (loss: 0.1285:  34%|███▍      | 21/62 [02:04<04:07,  6.04s/it]
Training Epoch: 5/100, batch 41/125 completed (loss: 0.0806:  34%|███▍      | 21/62 [02:04<04:07,  6.03s/it]
Training Epoch: 5/100, batch 41/125 completed (loss: 0.1978:  34%|███▍      | 21/62 [02:04<04:07,  6.04s/it]
Training Epoch: 5/100, batch 41/125 completed (loss: 0.2144:  34%|███▍      | 21/62 [02:04<04:07,  6.03s/it]
Training Epoch: 5/100, batch 42/125 completed (loss: 0.0791:  34%|███▍      | 21/62 [02:07<04:07,  6.04s/it]
Training Epoch: 5/100, batch 42/125 completed (loss: 0.1231:  34%|███▍      | 21/62 [02:07<04:07,  6.03s/it]
Training Epoch: 5/100, batch 42/125 completed (loss: 0.1099:  34%|███▍      | 21/62 [02:07<04:07,  6.04s/it]
Training Epoch: 5/100, batch 42/125 completed (loss: 0.3182:  34%|███▍      | 21/62 [02:07<04:07,  6.03s/it]
Training Epoch: 5/100, batch 42/125 completed (loss: 0.0791:  35%|███▌      | 22/62 [02:10<03:58,  5.96s/it]
Training Epoch: 5/100, batch 42/125 completed (loss: 0.1099:  35%|███▌      | 22/62 [02:10<03:58,  5.96s/it]
Training Epoch: 5/100, batch 42/125 completed (loss: 0.1231:  35%|███▌      | 22/62 [02:10<03:58,  5.96s/it]
Training Epoch: 5/100, batch 42/125 completed (loss: 0.3182:  35%|███▌      | 22/62 [02:10<03:58,  5.97s/it]
Training Epoch: 5/100, batch 43/125 completed (loss: 0.177:  35%|███▌      | 22/62 [02:10<03:58,  5.96s/it] 
Training Epoch: 5/100, batch 43/125 completed (loss: 0.147:  35%|███▌      | 22/62 [02:10<03:58,  5.96s/it] 
Training Epoch: 5/100, batch 43/125 completed (loss: 0.1188:  35%|███▌      | 22/62 [02:10<03:58,  5.96s/it]
Training Epoch: 5/100, batch 43/125 completed (loss: 0.2727:  35%|███▌      | 22/62 [02:10<03:58,  5.97s/it]
Training Epoch: 5/100, batch 44/125 completed (loss: 0.4225:  35%|███▌      | 22/62 [02:13<03:58,  5.96s/it]
Training Epoch: 5/100, batch 44/125 completed (loss: 0.1212:  35%|███▌      | 22/62 [02:13<03:58,  5.96s/it]
Training Epoch: 5/100, batch 44/125 completed (loss: 0.1989:  35%|███▌      | 22/62 [02:13<03:58,  5.96s/it]
Training Epoch: 5/100, batch 44/125 completed (loss: 0.1116:  35%|███▌      | 22/62 [02:13<03:58,  5.97s/it]
Training Epoch: 5/100, batch 44/125 completed (loss: 0.1989:  37%|███▋      | 23/62 [02:16<03:50,  5.91s/it]
Training Epoch: 5/100, batch 44/125 completed (loss: 0.1116:  37%|███▋      | 23/62 [02:16<03:50,  5.91s/it]
Training Epoch: 5/100, batch 44/125 completed (loss: 0.1212:  37%|███▋      | 23/62 [02:16<03:50,  5.91s/it]
Training Epoch: 5/100, batch 44/125 completed (loss: 0.4225:  37%|███▋      | 23/62 [02:16<03:50,  5.91s/it]
Training Epoch: 5/100, batch 45/125 completed (loss: 0.0741:  37%|███▋      | 23/62 [02:16<03:50,  5.91s/it]
Training Epoch: 5/100, batch 45/125 completed (loss: 0.3004:  37%|███▋      | 23/62 [02:16<03:50,  5.91s/it]
Training Epoch: 5/100, batch 45/125 completed (loss: 0.1067:  37%|███▋      | 23/62 [02:16<03:50,  5.91s/it]
Training Epoch: 5/100, batch 45/125 completed (loss: 0.1641:  37%|███▋      | 23/62 [02:16<03:50,  5.91s/it]
Training Epoch: 5/100, batch 46/125 completed (loss: 0.0699:  37%|███▋      | 23/62 [02:19<03:50,  5.91s/it]
Training Epoch: 5/100, batch 46/125 completed (loss: 0.3582:  37%|███▋      | 23/62 [02:19<03:50,  5.91s/it]
Training Epoch: 5/100, batch 46/125 completed (loss: 0.4092:  37%|███▋      | 23/62 [02:19<03:50,  5.91s/it]
Training Epoch: 5/100, batch 46/125 completed (loss: 0.1719:  37%|███▋      | 23/62 [02:19<03:50,  5.91s/it]
Training Epoch: 5/100, batch 46/125 completed (loss: 0.1719:  39%|███▊      | 24/62 [02:22<03:50,  6.06s/it]
Training Epoch: 5/100, batch 46/125 completed (loss: 0.0699:  39%|███▊      | 24/62 [02:22<03:50,  6.06s/it]
Training Epoch: 5/100, batch 46/125 completed (loss: 0.4092:  39%|███▊      | 24/62 [02:22<03:50,  6.06s/it]
Training Epoch: 5/100, batch 46/125 completed (loss: 0.3582:  39%|███▊      | 24/62 [02:22<03:50,  6.06s/it]
Training Epoch: 5/100, batch 47/125 completed (loss: 0.2109:  39%|███▊      | 24/62 [02:22<03:50,  6.06s/it]
Training Epoch: 5/100, batch 47/125 completed (loss: 0.0858:  39%|███▊      | 24/62 [02:22<03:50,  6.06s/it]
Training Epoch: 5/100, batch 47/125 completed (loss: 0.1616:  39%|███▊      | 24/62 [02:22<03:50,  6.06s/it]
Training Epoch: 5/100, batch 47/125 completed (loss: 0.1552:  39%|███▊      | 24/62 [02:22<03:50,  6.06s/it]
Training Epoch: 5/100, batch 48/125 completed (loss: 0.0962:  39%|███▊      | 24/62 [02:25<03:50,  6.06s/it]
Training Epoch: 5/100, batch 48/125 completed (loss: 0.5707:  39%|███▊      | 24/62 [02:25<03:50,  6.06s/it]
Training Epoch: 5/100, batch 48/125 completed (loss: 0.2826:  39%|███▊      | 24/62 [02:25<03:50,  6.06s/it]
Training Epoch: 5/100, batch 48/125 completed (loss: 0.0918:  39%|███▊      | 24/62 [02:25<03:50,  6.06s/it]
Training Epoch: 5/100, batch 48/125 completed (loss: 0.0918:  40%|████      | 25/62 [02:28<03:41,  5.99s/it]
Training Epoch: 5/100, batch 48/125 completed (loss: 0.2826:  40%|████      | 25/62 [02:28<03:41,  5.98s/it]
Training Epoch: 5/100, batch 48/125 completed (loss: 0.0962:  40%|████      | 25/62 [02:28<03:41,  5.98s/it]
Training Epoch: 5/100, batch 48/125 completed (loss: 0.5707:  40%|████      | 25/62 [02:28<03:41,  5.99s/it]
Training Epoch: 5/100, batch 49/125 completed (loss: 0.0856:  40%|████      | 25/62 [02:28<03:41,  5.99s/it]
Training Epoch: 5/100, batch 49/125 completed (loss: 0.1652:  40%|████      | 25/62 [02:28<03:41,  5.98s/it]
Training Epoch: 5/100, batch 49/125 completed (loss: 0.1386:  40%|████      | 25/62 [02:28<03:41,  5.98s/it]
Training Epoch: 5/100, batch 49/125 completed (loss: 0.1533:  40%|████      | 25/62 [02:28<03:41,  5.99s/it]
Training Epoch: 5/100, batch 50/125 completed (loss: 0.0975:  40%|████      | 25/62 [02:31<03:41,  5.98s/it]
Training Epoch: 5/100, batch 50/125 completed (loss: 0.0718:  40%|████      | 25/62 [02:31<03:41,  5.98s/it]
Training Epoch: 5/100, batch 50/125 completed (loss: 0.225:  40%|████      | 25/62 [02:31<03:41,  5.99s/it] 
Training Epoch: 5/100, batch 50/125 completed (loss: 0.1583:  40%|████      | 25/62 [02:31<03:41,  5.99s/it]
Training Epoch: 5/100, batch 50/125 completed (loss: 0.1583:  42%|████▏     | 26/62 [02:34<03:36,  6.01s/it]
Training Epoch: 5/100, batch 50/125 completed (loss: 0.0975:  42%|████▏     | 26/62 [02:34<03:36,  6.01s/it]
Training Epoch: 5/100, batch 50/125 completed (loss: 0.225:  42%|████▏     | 26/62 [02:34<03:36,  6.01s/it]
Training Epoch: 5/100, batch 50/125 completed (loss: 0.0718:  42%|████▏     | 26/62 [02:34<03:36,  6.01s/it]
Training Epoch: 5/100, batch 51/125 completed (loss: 0.1082:  42%|████▏     | 26/62 [02:34<03:36,  6.01s/it]
Training Epoch: 5/100, batch 51/125 completed (loss: 0.2867:  42%|████▏     | 26/62 [02:34<03:36,  6.01s/it]
Training Epoch: 5/100, batch 51/125 completed (loss: 0.1307:  42%|████▏     | 26/62 [02:34<03:36,  6.01s/it]
Training Epoch: 5/100, batch 51/125 completed (loss: 0.1113:  42%|████▏     | 26/62 [02:34<03:36,  6.01s/it]
Training Epoch: 5/100, batch 52/125 completed (loss: 0.1701:  42%|████▏     | 26/62 [02:37<03:36,  6.01s/it]
Training Epoch: 5/100, batch 52/125 completed (loss: 0.1824:  42%|████▏     | 26/62 [02:37<03:36,  6.01s/it]
Training Epoch: 5/100, batch 52/125 completed (loss: 0.3856:  42%|████▏     | 26/62 [02:37<03:36,  6.01s/it]
Training Epoch: 5/100, batch 52/125 completed (loss: 0.0834:  42%|████▏     | 26/62 [02:37<03:36,  6.01s/it]
Training Epoch: 5/100, batch 52/125 completed (loss: 0.1701:  44%|████▎     | 27/62 [02:40<03:31,  6.05s/it]
Training Epoch: 5/100, batch 52/125 completed (loss: 0.0834:  44%|████▎     | 27/62 [02:40<03:31,  6.05s/it]
Training Epoch: 5/100, batch 52/125 completed (loss: 0.1824:  44%|████▎     | 27/62 [02:40<03:31,  6.05s/it]
Training Epoch: 5/100, batch 52/125 completed (loss: 0.3856:  44%|████▎     | 27/62 [02:40<03:31,  6.05s/it]
Training Epoch: 5/100, batch 53/125 completed (loss: 0.2691:  44%|████▎     | 27/62 [02:40<03:31,  6.05s/it]
Training Epoch: 5/100, batch 53/125 completed (loss: 0.0165:  44%|████▎     | 27/62 [02:40<03:31,  6.05s/it]
Training Epoch: 5/100, batch 53/125 completed (loss: 0.1233:  44%|████▎     | 27/62 [02:40<03:31,  6.05s/it]
Training Epoch: 5/100, batch 53/125 completed (loss: 0.2548:  44%|████▎     | 27/62 [02:40<03:31,  6.05s/it]
Training Epoch: 5/100, batch 54/125 completed (loss: 0.2079:  44%|████▎     | 27/62 [02:43<03:31,  6.05s/it]
Training Epoch: 5/100, batch 54/125 completed (loss: 0.3021:  44%|████▎     | 27/62 [02:43<03:31,  6.05s/it]
Training Epoch: 5/100, batch 54/125 completed (loss: 0.0926:  44%|████▎     | 27/62 [02:43<03:31,  6.05s/it]
Training Epoch: 5/100, batch 54/125 completed (loss: 0.1663:  44%|████▎     | 27/62 [02:43<03:31,  6.05s/it]
Training Epoch: 5/100, batch 54/125 completed (loss: 0.1663:  45%|████▌     | 28/62 [02:46<03:22,  5.97s/it]
Training Epoch: 5/100, batch 54/125 completed (loss: 0.0926:  45%|████▌     | 28/62 [02:46<03:22,  5.97s/it]
Training Epoch: 5/100, batch 54/125 completed (loss: 0.3021:  45%|████▌     | 28/62 [02:46<03:22,  5.97s/it]
Training Epoch: 5/100, batch 54/125 completed (loss: 0.2079:  45%|████▌     | 28/62 [02:46<03:22,  5.97s/it]
Training Epoch: 5/100, batch 55/125 completed (loss: 0.102:  45%|████▌     | 28/62 [02:46<03:22,  5.97s/it] 
Training Epoch: 5/100, batch 55/125 completed (loss: 0.3061:  45%|████▌     | 28/62 [02:46<03:22,  5.97s/it]
Training Epoch: 5/100, batch 55/125 completed (loss: 0.1557:  45%|████▌     | 28/62 [02:46<03:22,  5.97s/it]
Training Epoch: 5/100, batch 55/125 completed (loss: 0.2307:  45%|████▌     | 28/62 [02:46<03:22,  5.97s/it]
Training Epoch: 5/100, batch 56/125 completed (loss: 0.1971:  45%|████▌     | 28/62 [02:49<03:22,  5.97s/it]
Training Epoch: 5/100, batch 56/125 completed (loss: 0.4607:  45%|████▌     | 28/62 [02:49<03:22,  5.97s/it]
Training Epoch: 5/100, batch 56/125 completed (loss: 0.1271:  45%|████▌     | 28/62 [02:49<03:22,  5.97s/it]
Training Epoch: 5/100, batch 56/125 completed (loss: 0.1622:  45%|████▌     | 28/62 [02:49<03:22,  5.97s/it]
Training Epoch: 5/100, batch 56/125 completed (loss: 0.4607:  47%|████▋     | 29/62 [02:52<03:19,  6.06s/it]
Training Epoch: 5/100, batch 56/125 completed (loss: 0.1622:  47%|████▋     | 29/62 [02:52<03:19,  6.06s/it]
Training Epoch: 5/100, batch 56/125 completed (loss: 0.1971:  47%|████▋     | 29/62 [02:52<03:19,  6.06s/it]
Training Epoch: 5/100, batch 56/125 completed (loss: 0.1271:  47%|████▋     | 29/62 [02:52<03:19,  6.06s/it]
Training Epoch: 5/100, batch 57/125 completed (loss: 0.1812:  47%|████▋     | 29/62 [02:52<03:19,  6.06s/it]
Training Epoch: 5/100, batch 57/125 completed (loss: 0.101:  47%|████▋     | 29/62 [02:52<03:19,  6.06s/it] 
Training Epoch: 5/100, batch 57/125 completed (loss: 0.1332:  47%|████▋     | 29/62 [02:52<03:19,  6.06s/it]
Training Epoch: 5/100, batch 57/125 completed (loss: 0.2732:  47%|████▋     | 29/62 [02:52<03:19,  6.06s/it]
Training Epoch: 5/100, batch 58/125 completed (loss: 0.2305:  47%|████▋     | 29/62 [02:55<03:19,  6.06s/it]
Training Epoch: 5/100, batch 58/125 completed (loss: 0.2572:  47%|████▋     | 29/62 [02:55<03:19,  6.06s/it]
Training Epoch: 5/100, batch 58/125 completed (loss: 0.2139:  47%|████▋     | 29/62 [02:55<03:19,  6.06s/it]
Training Epoch: 5/100, batch 58/125 completed (loss: 0.319:  47%|████▋     | 29/62 [02:55<03:19,  6.06s/it] 
Training Epoch: 5/100, batch 58/125 completed (loss: 0.2572:  48%|████▊     | 30/62 [02:58<03:11,  5.98s/it]
Training Epoch: 5/100, batch 58/125 completed (loss: 0.2139:  48%|████▊     | 30/62 [02:58<03:11,  5.97s/it]
Training Epoch: 5/100, batch 58/125 completed (loss: 0.319:  48%|████▊     | 30/62 [02:58<03:11,  5.98s/it]
Training Epoch: 5/100, batch 58/125 completed (loss: 0.2305:  48%|████▊     | 30/62 [02:58<03:11,  5.98s/it]
Training Epoch: 5/100, batch 59/125 completed (loss: 0.2441:  48%|████▊     | 30/62 [02:58<03:11,  5.98s/it]
Training Epoch: 5/100, batch 59/125 completed (loss: 0.1251:  48%|████▊     | 30/62 [02:58<03:11,  5.97s/it]
Training Epoch: 5/100, batch 59/125 completed (loss: 0.0726:  48%|████▊     | 30/62 [02:58<03:11,  5.98s/it]
Training Epoch: 5/100, batch 59/125 completed (loss: 0.2092:  48%|████▊     | 30/62 [02:58<03:11,  5.98s/it]
Training Epoch: 5/100, batch 60/125 completed (loss: 0.3063:  48%|████▊     | 30/62 [03:01<03:11,  5.98s/it]
Training Epoch: 5/100, batch 60/125 completed (loss: 0.1841:  48%|████▊     | 30/62 [03:01<03:11,  5.97s/it]
Training Epoch: 5/100, batch 60/125 completed (loss: 0.2253:  48%|████▊     | 30/62 [03:01<03:11,  5.98s/it]
Training Epoch: 5/100, batch 60/125 completed (loss: 0.1254:  48%|████▊     | 30/62 [03:01<03:11,  5.98s/it]
Training Epoch: 5/100, batch 60/125 completed (loss: 0.1254:  50%|█████     | 31/62 [03:04<03:03,  5.93s/it]
Training Epoch: 5/100, batch 60/125 completed (loss: 0.3063:  50%|█████     | 31/62 [03:04<03:03,  5.93s/it]
Training Epoch: 5/100, batch 60/125 completed (loss: 0.2253:  50%|█████     | 31/62 [03:04<03:03,  5.93s/it]
Training Epoch: 5/100, batch 60/125 completed (loss: 0.1841:  50%|█████     | 31/62 [03:04<03:03,  5.93s/it]
Training Epoch: 5/100, batch 61/125 completed (loss: 0.2342:  50%|█████     | 31/62 [03:04<03:03,  5.93s/it]
Training Epoch: 5/100, batch 61/125 completed (loss: 0.3579:  50%|█████     | 31/62 [03:04<03:03,  5.93s/it]
Training Epoch: 5/100, batch 61/125 completed (loss: 0.4561:  50%|█████     | 31/62 [03:04<03:03,  5.93s/it]
Training Epoch: 5/100, batch 61/125 completed (loss: 0.2401:  50%|█████     | 31/62 [03:04<03:03,  5.93s/it]
Training Epoch: 5/100, batch 62/125 completed (loss: 0.1282:  50%|█████     | 31/62 [03:07<03:03,  5.93s/it]
Training Epoch: 5/100, batch 62/125 completed (loss: 0.3217:  50%|█████     | 31/62 [03:07<03:03,  5.93s/it]
Training Epoch: 5/100, batch 62/125 completed (loss: 0.383:  50%|█████     | 31/62 [03:07<03:03,  5.93s/it] 
Training Epoch: 5/100, batch 62/125 completed (loss: 0.2856:  50%|█████     | 31/62 [03:07<03:03,  5.93s/it]
Training Epoch: 5/100, batch 62/125 completed (loss: 0.3217:  52%|█████▏    | 32/62 [03:10<03:01,  6.04s/it]
Training Epoch: 5/100, batch 62/125 completed (loss: 0.1282:  52%|█████▏    | 32/62 [03:10<03:01,  6.04s/it]
Training Epoch: 5/100, batch 62/125 completed (loss: 0.383:  52%|█████▏    | 32/62 [03:10<03:01,  6.04s/it]
Training Epoch: 5/100, batch 62/125 completed (loss: 0.2856:  52%|█████▏    | 32/62 [03:10<03:01,  6.04s/it]
Training Epoch: 5/100, batch 63/125 completed (loss: 0.1477:  52%|█████▏    | 32/62 [03:10<03:01,  6.04s/it]
Training Epoch: 5/100, batch 63/125 completed (loss: 0.0713:  52%|█████▏    | 32/62 [03:10<03:01,  6.04s/it]
Training Epoch: 5/100, batch 63/125 completed (loss: 0.1068:  52%|█████▏    | 32/62 [03:10<03:01,  6.04s/it]
Training Epoch: 5/100, batch 63/125 completed (loss: 0.1072:  52%|█████▏    | 32/62 [03:10<03:01,  6.04s/it]
Training Epoch: 5/100, batch 64/125 completed (loss: 0.0986:  52%|█████▏    | 32/62 [03:13<03:01,  6.04s/it]
Training Epoch: 5/100, batch 64/125 completed (loss: 0.172:  52%|█████▏    | 32/62 [03:13<03:01,  6.04s/it] 
Training Epoch: 5/100, batch 64/125 completed (loss: 0.2086:  52%|█████▏    | 32/62 [03:13<03:01,  6.04s/it]
Training Epoch: 5/100, batch 64/125 completed (loss: 0.2207:  52%|█████▏    | 32/62 [03:13<03:01,  6.04s/it]
Training Epoch: 5/100, batch 64/125 completed (loss: 0.2207:  53%|█████▎    | 33/62 [03:16<02:53,  5.97s/it]
Training Epoch: 5/100, batch 64/125 completed (loss: 0.172:  53%|█████▎    | 33/62 [03:16<02:53,  5.97s/it]
Training Epoch: 5/100, batch 64/125 completed (loss: 0.2086:  53%|█████▎    | 33/62 [03:16<02:53,  5.97s/it]
Training Epoch: 5/100, batch 64/125 completed (loss: 0.0986:  53%|█████▎    | 33/62 [03:16<02:53,  5.97s/it]
Training Epoch: 5/100, batch 65/125 completed (loss: 0.2187:  53%|█████▎    | 33/62 [03:16<02:53,  5.97s/it]
Training Epoch: 5/100, batch 65/125 completed (loss: 0.1841:  53%|█████▎    | 33/62 [03:16<02:53,  5.97s/it]
Training Epoch: 5/100, batch 65/125 completed (loss: 0.2689:  53%|█████▎    | 33/62 [03:16<02:53,  5.97s/it]
Training Epoch: 5/100, batch 65/125 completed (loss: 0.044:  53%|█████▎    | 33/62 [03:16<02:53,  5.97s/it] 
Training Epoch: 5/100, batch 66/125 completed (loss: 0.0633:  53%|█████▎    | 33/62 [03:19<02:53,  5.97s/it]
Training Epoch: 5/100, batch 66/125 completed (loss: 0.2312:  53%|█████▎    | 33/62 [03:19<02:53,  5.97s/it]
Training Epoch: 5/100, batch 66/125 completed (loss: 0.064:  53%|█████▎    | 33/62 [03:19<02:53,  5.97s/it] 
Training Epoch: 5/100, batch 66/125 completed (loss: 0.3393:  53%|█████▎    | 33/62 [03:19<02:53,  5.97s/it]
Training Epoch: 5/100, batch 66/125 completed (loss: 0.2312:  55%|█████▍    | 34/62 [03:22<02:49,  6.06s/it]
Training Epoch: 5/100, batch 66/125 completed (loss: 0.0633:  55%|█████▍    | 34/62 [03:22<02:49,  6.06s/it]
Training Epoch: 5/100, batch 66/125 completed (loss: 0.064:  55%|█████▍    | 34/62 [03:22<02:49,  6.06s/it]
Training Epoch: 5/100, batch 66/125 completed (loss: 0.3393:  55%|█████▍    | 34/62 [03:22<02:49,  6.07s/it]
Training Epoch: 5/100, batch 67/125 completed (loss: 0.2208:  55%|█████▍    | 34/62 [03:22<02:49,  6.06s/it]
Training Epoch: 5/100, batch 67/125 completed (loss: 0.4161:  55%|█████▍    | 34/62 [03:22<02:49,  6.06s/it]
Training Epoch: 5/100, batch 67/125 completed (loss: 0.1347:  55%|█████▍    | 34/62 [03:22<02:49,  6.06s/it]
Training Epoch: 5/100, batch 67/125 completed (loss: 0.2239:  55%|█████▍    | 34/62 [03:22<02:49,  6.07s/it]
Training Epoch: 5/100, batch 68/125 completed (loss: 0.0866:  55%|█████▍    | 34/62 [03:25<02:49,  6.06s/it]
Training Epoch: 5/100, batch 68/125 completed (loss: 0.0952:  55%|█████▍    | 34/62 [03:25<02:49,  6.06s/it]
Training Epoch: 5/100, batch 68/125 completed (loss: 0.4566:  55%|█████▍    | 34/62 [03:25<02:49,  6.06s/it]
Training Epoch: 5/100, batch 68/125 completed (loss: 0.3427:  55%|█████▍    | 34/62 [03:25<02:49,  6.07s/it]
Training Epoch: 5/100, batch 68/125 completed (loss: 0.4566:  56%|█████▋    | 35/62 [03:28<02:43,  6.05s/it]
Training Epoch: 5/100, batch 68/125 completed (loss: 0.0952:  56%|█████▋    | 35/62 [03:28<02:43,  6.05s/it]
Training Epoch: 5/100, batch 68/125 completed (loss: 0.3427:  56%|█████▋    | 35/62 [03:28<02:43,  6.05s/it]
Training Epoch: 5/100, batch 68/125 completed (loss: 0.0866:  56%|█████▋    | 35/62 [03:28<02:43,  6.05s/it]
Training Epoch: 5/100, batch 69/125 completed (loss: 0.6845:  56%|█████▋    | 35/62 [03:28<02:43,  6.05s/it]
Training Epoch: 5/100, batch 69/125 completed (loss: 0.1163:  56%|█████▋    | 35/62 [03:28<02:43,  6.05s/it]
Training Epoch: 5/100, batch 69/125 completed (loss: 0.2747:  56%|█████▋    | 35/62 [03:28<02:43,  6.05s/it]
Training Epoch: 5/100, batch 69/125 completed (loss: 0.2881:  56%|█████▋    | 35/62 [03:28<02:43,  6.05s/it]
Training Epoch: 5/100, batch 70/125 completed (loss: 0.4148:  56%|█████▋    | 35/62 [03:31<02:43,  6.05s/it]
Training Epoch: 5/100, batch 70/125 completed (loss: 0.1799:  56%|█████▋    | 35/62 [03:31<02:43,  6.05s/it]
Training Epoch: 5/100, batch 70/125 completed (loss: 0.406:  56%|█████▋    | 35/62 [03:31<02:43,  6.05s/it] 
Training Epoch: 5/100, batch 70/125 completed (loss: 0.2908:  56%|█████▋    | 35/62 [03:31<02:43,  6.05s/it]
Training Epoch: 5/100, batch 70/125 completed (loss: 0.1799:  58%|█████▊    | 36/62 [03:34<02:35,  5.98s/it]
Training Epoch: 5/100, batch 70/125 completed (loss: 0.4148:  58%|█████▊    | 36/62 [03:34<02:35,  5.98s/it]
Training Epoch: 5/100, batch 70/125 completed (loss: 0.2908:  58%|█████▊    | 36/62 [03:34<02:35,  5.98s/it]
Training Epoch: 5/100, batch 70/125 completed (loss: 0.406:  58%|█████▊    | 36/62 [03:34<02:35,  5.99s/it]
Training Epoch: 5/100, batch 71/125 completed (loss: 0.113:  58%|█████▊    | 36/62 [03:34<02:35,  5.98s/it] 
Training Epoch: 5/100, batch 71/125 completed (loss: 0.1893:  58%|█████▊    | 36/62 [03:34<02:35,  5.99s/it]
Training Epoch: 5/100, batch 71/125 completed (loss: 0.2895:  58%|█████▊    | 36/62 [03:34<02:35,  5.98s/it]
Training Epoch: 5/100, batch 71/125 completed (loss: 0.2713:  58%|█████▊    | 36/62 [03:34<02:35,  5.98s/it]
Training Epoch: 5/100, batch 72/125 completed (loss: 0.1516:  58%|█████▊    | 36/62 [03:37<02:35,  5.99s/it]
Training Epoch: 5/100, batch 72/125 completed (loss: 0.4728:  58%|█████▊    | 36/62 [03:37<02:35,  5.98s/it]
Training Epoch: 5/100, batch 72/125 completed (loss: 0.3204:  58%|█████▊    | 36/62 [03:37<02:35,  5.98s/it]
Training Epoch: 5/100, batch 72/125 completed (loss: 0.1456:  58%|█████▊    | 36/62 [03:37<02:35,  5.98s/it]
Training Epoch: 5/100, batch 72/125 completed (loss: 0.4728:  60%|█████▉    | 37/62 [03:40<02:32,  6.11s/it]
Training Epoch: 5/100, batch 72/125 completed (loss: 0.1516:  60%|█████▉    | 37/62 [03:40<02:32,  6.11s/it]
Training Epoch: 5/100, batch 72/125 completed (loss: 0.3204:  60%|█████▉    | 37/62 [03:40<02:32,  6.11s/it]
Training Epoch: 5/100, batch 72/125 completed (loss: 0.1456:  60%|█████▉    | 37/62 [03:40<02:32,  6.11s/it]
Training Epoch: 5/100, batch 73/125 completed (loss: 0.1027:  60%|█████▉    | 37/62 [03:40<02:32,  6.11s/it]
Training Epoch: 5/100, batch 73/125 completed (loss: 0.1729:  60%|█████▉    | 37/62 [03:40<02:32,  6.11s/it]
Training Epoch: 5/100, batch 73/125 completed (loss: 0.3034:  60%|█████▉    | 37/62 [03:40<02:32,  6.11s/it]
Training Epoch: 5/100, batch 73/125 completed (loss: 0.1212:  60%|█████▉    | 37/62 [03:40<02:32,  6.11s/it]
Training Epoch: 5/100, batch 74/125 completed (loss: 0.2111:  60%|█████▉    | 37/62 [03:43<02:32,  6.11s/it]
Training Epoch: 5/100, batch 74/125 completed (loss: 0.2288:  60%|█████▉    | 37/62 [03:43<02:32,  6.11s/it]
Training Epoch: 5/100, batch 74/125 completed (loss: 0.5485:  60%|█████▉    | 37/62 [03:43<02:32,  6.11s/it]
Training Epoch: 5/100, batch 74/125 completed (loss: 0.1264:  60%|█████▉    | 37/62 [03:43<02:32,  6.11s/it]
Training Epoch: 5/100, batch 74/125 completed (loss: 0.2111:  61%|██████▏   | 38/62 [03:46<02:24,  6.03s/it]
Training Epoch: 5/100, batch 74/125 completed (loss: 0.5485:  61%|██████▏   | 38/62 [03:46<02:24,  6.03s/it]
Training Epoch: 5/100, batch 74/125 completed (loss: 0.2288:  61%|██████▏   | 38/62 [03:46<02:24,  6.03s/it]
Training Epoch: 5/100, batch 74/125 completed (loss: 0.1264:  61%|██████▏   | 38/62 [03:46<02:24,  6.03s/it]
Training Epoch: 5/100, batch 75/125 completed (loss: 0.3787:  61%|██████▏   | 38/62 [03:46<02:24,  6.03s/it]
Training Epoch: 5/100, batch 75/125 completed (loss: 0.2205:  61%|██████▏   | 38/62 [03:46<02:24,  6.03s/it]
Training Epoch: 5/100, batch 75/125 completed (loss: 0.0998:  61%|██████▏   | 38/62 [03:46<02:24,  6.03s/it]
Training Epoch: 5/100, batch 75/125 completed (loss: 0.2101:  61%|██████▏   | 38/62 [03:46<02:24,  6.03s/it]
Training Epoch: 5/100, batch 76/125 completed (loss: 0.1916:  61%|██████▏   | 38/62 [03:49<02:24,  6.03s/it]
Training Epoch: 5/100, batch 76/125 completed (loss: 0.5462:  61%|██████▏   | 38/62 [03:49<02:24,  6.03s/it]
Training Epoch: 5/100, batch 76/125 completed (loss: 0.2919:  61%|██████▏   | 38/62 [03:49<02:24,  6.03s/it]
Training Epoch: 5/100, batch 76/125 completed (loss: 0.0715:  61%|██████▏   | 38/62 [03:49<02:24,  6.03s/it]
Training Epoch: 5/100, batch 76/125 completed (loss: 0.1916:  63%|██████▎   | 39/62 [03:52<02:19,  6.07s/it]
Training Epoch: 5/100, batch 76/125 completed (loss: 0.2919:  63%|██████▎   | 39/62 [03:52<02:19,  6.07s/it]
Training Epoch: 5/100, batch 76/125 completed (loss: 0.0715:  63%|██████▎   | 39/62 [03:52<02:19,  6.07s/it]
Training Epoch: 5/100, batch 76/125 completed (loss: 0.5462:  63%|██████▎   | 39/62 [03:52<02:19,  6.07s/it]
Training Epoch: 5/100, batch 77/125 completed (loss: 0.3553:  63%|██████▎   | 39/62 [03:52<02:19,  6.07s/it]
Training Epoch: 5/100, batch 77/125 completed (loss: 0.2365:  63%|██████▎   | 39/62 [03:52<02:19,  6.07s/it]
Training Epoch: 5/100, batch 77/125 completed (loss: 0.3226:  63%|██████▎   | 39/62 [03:52<02:19,  6.07s/it]
Training Epoch: 5/100, batch 77/125 completed (loss: 0.2322:  63%|██████▎   | 39/62 [03:52<02:19,  6.07s/it]
Training Epoch: 5/100, batch 78/125 completed (loss: 0.2534:  63%|██████▎   | 39/62 [03:55<02:19,  6.07s/it]
Training Epoch: 5/100, batch 78/125 completed (loss: 0.3015:  63%|██████▎   | 39/62 [03:55<02:19,  6.07s/it]
Training Epoch: 5/100, batch 78/125 completed (loss: 0.0975:  63%|██████▎   | 39/62 [03:55<02:19,  6.07s/it]
Training Epoch: 5/100, batch 78/125 completed (loss: 0.3003:  63%|██████▎   | 39/62 [03:55<02:19,  6.07s/it]
Training Epoch: 5/100, batch 78/125 completed (loss: 0.0975:  65%|██████▍   | 40/62 [03:58<02:13,  6.05s/it]
Training Epoch: 5/100, batch 78/125 completed (loss: 0.3015:  65%|██████▍   | 40/62 [03:58<02:13,  6.05s/it]
Training Epoch: 5/100, batch 78/125 completed (loss: 0.2534:  65%|██████▍   | 40/62 [03:58<02:13,  6.05s/it]
Training Epoch: 5/100, batch 78/125 completed (loss: 0.3003:  65%|██████▍   | 40/62 [03:58<02:13,  6.05s/it]
Training Epoch: 5/100, batch 79/125 completed (loss: 0.1474:  65%|██████▍   | 40/62 [03:58<02:13,  6.05s/it]
Training Epoch: 5/100, batch 79/125 completed (loss: 0.0805:  65%|██████▍   | 40/62 [03:58<02:13,  6.05s/it]
Training Epoch: 5/100, batch 79/125 completed (loss: 0.3038:  65%|██████▍   | 40/62 [03:58<02:13,  6.05s/it]
Training Epoch: 5/100, batch 79/125 completed (loss: 0.2533:  65%|██████▍   | 40/62 [03:58<02:13,  6.05s/it]
Training Epoch: 5/100, batch 80/125 completed (loss: 0.0739:  65%|██████▍   | 40/62 [04:01<02:13,  6.05s/it]
Training Epoch: 5/100, batch 80/125 completed (loss: 0.0637:  65%|██████▍   | 40/62 [04:01<02:13,  6.05s/it]
Training Epoch: 5/100, batch 80/125 completed (loss: 0.294:  65%|██████▍   | 40/62 [04:01<02:13,  6.05s/it] 
Training Epoch: 5/100, batch 80/125 completed (loss: 0.2859:  65%|██████▍   | 40/62 [04:01<02:13,  6.05s/it]
Training Epoch: 5/100, batch 80/125 completed (loss: 0.2859:  66%|██████▌   | 41/62 [04:04<02:05,  5.98s/it]
Training Epoch: 5/100, batch 80/125 completed (loss: 0.0739:  66%|██████▌   | 41/62 [04:04<02:05,  5.99s/it]
Training Epoch: 5/100, batch 80/125 completed (loss: 0.0637:  66%|██████▌   | 41/62 [04:04<02:05,  5.99s/it]
Training Epoch: 5/100, batch 80/125 completed (loss: 0.294:  66%|██████▌   | 41/62 [04:04<02:05,  5.99s/it]
Training Epoch: 5/100, batch 81/125 completed (loss: 0.3111:  66%|██████▌   | 41/62 [04:04<02:05,  5.99s/it]
Training Epoch: 5/100, batch 81/125 completed (loss: 0.1361:  66%|██████▌   | 41/62 [04:04<02:05,  5.99s/it]
Training Epoch: 5/100, batch 81/125 completed (loss: 0.3228:  66%|██████▌   | 41/62 [04:04<02:05,  5.99s/it]
Training Epoch: 5/100, batch 81/125 completed (loss: 0.0666:  66%|██████▌   | 41/62 [04:04<02:05,  5.98s/it]
Training Epoch: 5/100, batch 82/125 completed (loss: 0.3716:  66%|██████▌   | 41/62 [04:08<02:05,  5.99s/it]
Training Epoch: 5/100, batch 82/125 completed (loss: 0.3072:  66%|██████▌   | 41/62 [04:08<02:05,  5.99s/it]
Training Epoch: 5/100, batch 82/125 completed (loss: 0.0917:  66%|██████▌   | 41/62 [04:08<02:05,  5.99s/it]
Training Epoch: 5/100, batch 82/125 completed (loss: 0.2185:  66%|██████▌   | 41/62 [04:08<02:05,  5.98s/it]
Training Epoch: 5/100, batch 82/125 completed (loss: 0.3072:  68%|██████▊   | 42/62 [04:11<02:01,  6.08s/it]
Training Epoch: 5/100, batch 82/125 completed (loss: 0.0917:  68%|██████▊   | 42/62 [04:11<02:01,  6.08s/it]
Training Epoch: 5/100, batch 82/125 completed (loss: 0.2185:  68%|██████▊   | 42/62 [04:11<02:01,  6.08s/it]
Training Epoch: 5/100, batch 82/125 completed (loss: 0.3716:  68%|██████▊   | 42/62 [04:11<02:01,  6.08s/it]
Training Epoch: 5/100, batch 83/125 completed (loss: 0.3335:  68%|██████▊   | 42/62 [04:11<02:01,  6.08s/it]
Training Epoch: 5/100, batch 83/125 completed (loss: 0.1838:  68%|██████▊   | 42/62 [04:11<02:01,  6.08s/it]
Training Epoch: 5/100, batch 83/125 completed (loss: 0.2705:  68%|██████▊   | 42/62 [04:11<02:01,  6.08s/it]
Training Epoch: 5/100, batch 83/125 completed (loss: 0.4295:  68%|██████▊   | 42/62 [04:11<02:01,  6.08s/it]
Training Epoch: 5/100, batch 84/125 completed (loss: 0.1012:  68%|██████▊   | 42/62 [04:14<02:01,  6.08s/it]
Training Epoch: 5/100, batch 84/125 completed (loss: 0.0605:  68%|██████▊   | 42/62 [04:14<02:01,  6.08s/it]
Training Epoch: 5/100, batch 84/125 completed (loss: 0.1748:  68%|██████▊   | 42/62 [04:14<02:01,  6.08s/it]
Training Epoch: 5/100, batch 84/125 completed (loss: 0.2117:  68%|██████▊   | 42/62 [04:14<02:01,  6.08s/it]
Training Epoch: 5/100, batch 84/125 completed (loss: 0.2117:  69%|██████▉   | 43/62 [04:17<01:54,  6.05s/it]
Training Epoch: 5/100, batch 84/125 completed (loss: 0.1748:  69%|██████▉   | 43/62 [04:17<01:55,  6.05s/it]
Training Epoch: 5/100, batch 84/125 completed (loss: 0.1012:  69%|██████▉   | 43/62 [04:17<01:55,  6.05s/it]
Training Epoch: 5/100, batch 84/125 completed (loss: 0.0605:  69%|██████▉   | 43/62 [04:17<01:55,  6.06s/it]
Training Epoch: 5/100, batch 85/125 completed (loss: 0.2305:  69%|██████▉   | 43/62 [04:17<01:55,  6.06s/it]
Training Epoch: 5/100, batch 85/125 completed (loss: 0.2488:  69%|██████▉   | 43/62 [04:17<01:55,  6.05s/it]
Training Epoch: 5/100, batch 85/125 completed (loss: 0.1236:  69%|██████▉   | 43/62 [04:17<01:55,  6.05s/it]
Training Epoch: 5/100, batch 85/125 completed (loss: 0.1774:  69%|██████▉   | 43/62 [04:17<01:54,  6.05s/it]
Training Epoch: 5/100, batch 86/125 completed (loss: 0.2706:  69%|██████▉   | 43/62 [04:20<01:55,  6.05s/it]
Training Epoch: 5/100, batch 86/125 completed (loss: 0.1497:  69%|██████▉   | 43/62 [04:20<01:55,  6.05s/it]
Training Epoch: 5/100, batch 86/125 completed (loss: 0.2715:  69%|██████▉   | 43/62 [04:20<01:55,  6.06s/it]
Training Epoch: 5/100, batch 86/125 completed (loss: 0.1712:  69%|██████▉   | 43/62 [04:20<01:54,  6.05s/it]
Training Epoch: 5/100, batch 86/125 completed (loss: 0.1497:  71%|███████   | 44/62 [04:23<01:48,  6.04s/it]
Training Epoch: 5/100, batch 86/125 completed (loss: 0.2715:  71%|███████   | 44/62 [04:23<01:48,  6.04s/it]
Training Epoch: 5/100, batch 86/125 completed (loss: 0.1712:  71%|███████   | 44/62 [04:23<01:48,  6.04s/it]
Training Epoch: 5/100, batch 86/125 completed (loss: 0.2706:  71%|███████   | 44/62 [04:23<01:48,  6.04s/it]
Training Epoch: 5/100, batch 87/125 completed (loss: 0.1196:  71%|███████   | 44/62 [04:23<01:48,  6.04s/it]
Training Epoch: 5/100, batch 87/125 completed (loss: 0.23:  71%|███████   | 44/62 [04:23<01:48,  6.04s/it]  
Training Epoch: 5/100, batch 87/125 completed (loss: 0.0964:  71%|███████   | 44/62 [04:23<01:48,  6.04s/it]
Training Epoch: 5/100, batch 87/125 completed (loss: 0.2987:  71%|███████   | 44/62 [04:23<01:48,  6.04s/it]
Training Epoch: 5/100, batch 88/125 completed (loss: 0.1916:  71%|███████   | 44/62 [04:26<01:48,  6.04s/it]
Training Epoch: 5/100, batch 88/125 completed (loss: 0.0829:  71%|███████   | 44/62 [04:26<01:48,  6.04s/it]
Training Epoch: 5/100, batch 88/125 completed (loss: 0.2398:  71%|███████   | 44/62 [04:26<01:48,  6.04s/it]
Training Epoch: 5/100, batch 88/125 completed (loss: 0.2941:  71%|███████   | 44/62 [04:26<01:48,  6.04s/it]
Training Epoch: 5/100, batch 88/125 completed (loss: 0.2941:  73%|███████▎  | 45/62 [04:29<01:42,  6.06s/it]
Training Epoch: 5/100, batch 88/125 completed (loss: 0.2398:  73%|███████▎  | 45/62 [04:29<01:42,  6.06s/it]
Training Epoch: 5/100, batch 88/125 completed (loss: 0.0829:  73%|███████▎  | 45/62 [04:29<01:42,  6.06s/it]
Training Epoch: 5/100, batch 88/125 completed (loss: 0.1916:  73%|███████▎  | 45/62 [04:29<01:42,  6.06s/it]
Training Epoch: 5/100, batch 89/125 completed (loss: 0.0596:  73%|███████▎  | 45/62 [04:29<01:42,  6.06s/it]
Training Epoch: 5/100, batch 89/125 completed (loss: 0.2114:  73%|███████▎  | 45/62 [04:29<01:42,  6.06s/it]
Training Epoch: 5/100, batch 89/125 completed (loss: 0.2718:  73%|███████▎  | 45/62 [04:29<01:42,  6.06s/it]
Training Epoch: 5/100, batch 89/125 completed (loss: 0.124:  73%|███████▎  | 45/62 [04:29<01:42,  6.06s/it] 
Training Epoch: 5/100, batch 90/125 completed (loss: 0.1864:  73%|███████▎  | 45/62 [04:32<01:42,  6.06s/it]
Training Epoch: 5/100, batch 90/125 completed (loss: 0.0925:  73%|███████▎  | 45/62 [04:32<01:42,  6.06s/it]
Training Epoch: 5/100, batch 90/125 completed (loss: 0.1777:  73%|███████▎  | 45/62 [04:32<01:42,  6.06s/it]
Training Epoch: 5/100, batch 90/125 completed (loss: 0.1587:  73%|███████▎  | 45/62 [04:32<01:42,  6.06s/it]
Training Epoch: 5/100, batch 90/125 completed (loss: 0.1864:  74%|███████▍  | 46/62 [04:35<01:36,  6.03s/it]
Training Epoch: 5/100, batch 90/125 completed (loss: 0.0925:  74%|███████▍  | 46/62 [04:35<01:36,  6.03s/it]
Training Epoch: 5/100, batch 90/125 completed (loss: 0.1587:  74%|███████▍  | 46/62 [04:35<01:36,  6.03s/it]
Training Epoch: 5/100, batch 90/125 completed (loss: 0.1777:  74%|███████▍  | 46/62 [04:35<01:36,  6.03s/it]
Training Epoch: 5/100, batch 91/125 completed (loss: 0.0756:  74%|███████▍  | 46/62 [04:35<01:36,  6.03s/it]
Training Epoch: 5/100, batch 91/125 completed (loss: 0.1381:  74%|███████▍  | 46/62 [04:35<01:36,  6.03s/it]
Training Epoch: 5/100, batch 91/125 completed (loss: 0.1398:  74%|███████▍  | 46/62 [04:35<01:36,  6.03s/it]
Training Epoch: 5/100, batch 91/125 completed (loss: 0.1325:  74%|███████▍  | 46/62 [04:35<01:36,  6.03s/it]
Training Epoch: 5/100, batch 92/125 completed (loss: 0.498:  74%|███████▍  | 46/62 [04:38<01:36,  6.03s/it] 
Training Epoch: 5/100, batch 92/125 completed (loss: 0.246:  74%|███████▍  | 46/62 [04:38<01:36,  6.03s/it] 
Training Epoch: 5/100, batch 92/125 completed (loss: 0.3983:  74%|███████▍  | 46/62 [04:38<01:36,  6.03s/it]
Training Epoch: 5/100, batch 92/125 completed (loss: 0.1914:  74%|███████▍  | 46/62 [04:38<01:36,  6.03s/it]
Training Epoch: 5/100, batch 92/125 completed (loss: 0.246:  76%|███████▌  | 47/62 [04:41<01:31,  6.10s/it]
Training Epoch: 5/100, batch 92/125 completed (loss: 0.1914:  76%|███████▌  | 47/62 [04:41<01:31,  6.10s/it]
Training Epoch: 5/100, batch 92/125 completed (loss: 0.498:  76%|███████▌  | 47/62 [04:41<01:31,  6.10s/it]
Training Epoch: 5/100, batch 92/125 completed (loss: 0.3983:  76%|███████▌  | 47/62 [04:41<01:31,  6.10s/it]
Training Epoch: 5/100, batch 93/125 completed (loss: 0.2431:  76%|███████▌  | 47/62 [04:41<01:31,  6.10s/it]
Training Epoch: 5/100, batch 93/125 completed (loss: 0.3563:  76%|███████▌  | 47/62 [04:41<01:31,  6.10s/it]
Training Epoch: 5/100, batch 93/125 completed (loss: 0.3974:  76%|███████▌  | 47/62 [04:41<01:31,  6.10s/it]
Training Epoch: 5/100, batch 93/125 completed (loss: 0.2927:  76%|███████▌  | 47/62 [04:41<01:31,  6.10s/it]
Training Epoch: 5/100, batch 94/125 completed (loss: 0.1019:  76%|███████▌  | 47/62 [04:44<01:31,  6.10s/it]
Training Epoch: 5/100, batch 94/125 completed (loss: 0.095:  76%|███████▌  | 47/62 [04:44<01:31,  6.10s/it] 
Training Epoch: 5/100, batch 94/125 completed (loss: 0.4371:  76%|███████▌  | 47/62 [04:44<01:31,  6.10s/it]
Training Epoch: 5/100, batch 94/125 completed (loss: 0.2542:  76%|███████▌  | 47/62 [04:44<01:31,  6.10s/it]
Training Epoch: 5/100, batch 94/125 completed (loss: 0.1019:  77%|███████▋  | 48/62 [04:47<01:24,  6.06s/it]
Training Epoch: 5/100, batch 94/125 completed (loss: 0.095:  77%|███████▋  | 48/62 [04:47<01:24,  6.06s/it]
Training Epoch: 5/100, batch 94/125 completed (loss: 0.4371:  77%|███████▋  | 48/62 [04:47<01:24,  6.06s/it]
Training Epoch: 5/100, batch 94/125 completed (loss: 0.2542:  77%|███████▋  | 48/62 [04:47<01:24,  6.06s/it]
Training Epoch: 5/100, batch 95/125 completed (loss: 0.1466:  77%|███████▋  | 48/62 [04:47<01:24,  6.06s/it]
Training Epoch: 5/100, batch 95/125 completed (loss: 0.2726:  77%|███████▋  | 48/62 [04:47<01:24,  6.06s/it]
Training Epoch: 5/100, batch 95/125 completed (loss: 0.1749:  77%|███████▋  | 48/62 [04:47<01:24,  6.06s/it]
Training Epoch: 5/100, batch 95/125 completed (loss: 0.2733:  77%|███████▋  | 48/62 [04:47<01:24,  6.06s/it]
Training Epoch: 5/100, batch 96/125 completed (loss: 0.4112:  77%|███████▋  | 48/62 [04:50<01:24,  6.06s/it]
Training Epoch: 5/100, batch 96/125 completed (loss: 0.1117:  77%|███████▋  | 48/62 [04:50<01:24,  6.06s/it]
Training Epoch: 5/100, batch 96/125 completed (loss: 0.3376:  77%|███████▋  | 48/62 [04:50<01:24,  6.06s/it]
Training Epoch: 5/100, batch 96/125 completed (loss: 0.0585:  77%|███████▋  | 48/62 [04:50<01:24,  6.06s/it]
Training Epoch: 5/100, batch 96/125 completed (loss: 0.3376:  79%|███████▉  | 49/62 [04:53<01:17,  6.00s/it]
Training Epoch: 5/100, batch 96/125 completed (loss: 0.4112:  79%|███████▉  | 49/62 [04:53<01:17,  6.00s/it]
Training Epoch: 5/100, batch 96/125 completed (loss: 0.0585:  79%|███████▉  | 49/62 [04:53<01:17,  6.00s/it]
Training Epoch: 5/100, batch 96/125 completed (loss: 0.1117:  79%|███████▉  | 49/62 [04:53<01:17,  6.00s/it]
Training Epoch: 5/100, batch 97/125 completed (loss: 0.1238:  79%|███████▉  | 49/62 [04:53<01:17,  6.00s/it]
Training Epoch: 5/100, batch 97/125 completed (loss: 0.1924:  79%|███████▉  | 49/62 [04:53<01:17,  6.00s/it]
Training Epoch: 5/100, batch 97/125 completed (loss: 0.203:  79%|███████▉  | 49/62 [04:53<01:17,  6.00s/it] 
Training Epoch: 5/100, batch 97/125 completed (loss: 0.1003:  79%|███████▉  | 49/62 [04:53<01:17,  6.00s/it]
Training Epoch: 5/100, batch 98/125 completed (loss: 0.0652:  79%|███████▉  | 49/62 [04:56<01:17,  6.00s/it]
Training Epoch: 5/100, batch 98/125 completed (loss: 0.1524:  79%|███████▉  | 49/62 [04:56<01:17,  6.00s/it]
Training Epoch: 5/100, batch 98/125 completed (loss: 0.1008:  79%|███████▉  | 49/62 [04:56<01:17,  6.00s/it]
Training Epoch: 5/100, batch 98/125 completed (loss: 0.2791:  79%|███████▉  | 49/62 [04:56<01:17,  6.00s/it]
Training Epoch: 5/100, batch 98/125 completed (loss: 0.0652:  81%|████████  | 50/62 [04:59<01:12,  6.08s/it]
Training Epoch: 5/100, batch 98/125 completed (loss: 0.1008:  81%|████████  | 50/62 [04:59<01:12,  6.08s/it]
Training Epoch: 5/100, batch 98/125 completed (loss: 0.2791:  81%|████████  | 50/62 [04:59<01:12,  6.08s/it]
Training Epoch: 5/100, batch 98/125 completed (loss: 0.1524:  81%|████████  | 50/62 [04:59<01:12,  6.08s/it]
Training Epoch: 5/100, batch 99/125 completed (loss: 0.2684:  81%|████████  | 50/62 [04:59<01:12,  6.08s/it]
Training Epoch: 5/100, batch 99/125 completed (loss: 0.1076:  81%|████████  | 50/62 [04:59<01:12,  6.08s/it]
Training Epoch: 5/100, batch 99/125 completed (loss: 0.1338:  81%|████████  | 50/62 [04:59<01:12,  6.08s/it]
Training Epoch: 5/100, batch 99/125 completed (loss: 0.1427:  81%|████████  | 50/62 [04:59<01:12,  6.08s/it]
Training Epoch: 5/100, batch 100/125 completed (loss: 0.1142:  81%|████████  | 50/62 [05:02<01:12,  6.08s/it]
Training Epoch: 5/100, batch 100/125 completed (loss: 0.1647:  81%|████████  | 50/62 [05:02<01:12,  6.08s/it]
Training Epoch: 5/100, batch 100/125 completed (loss: 0.1754:  81%|████████  | 50/62 [05:02<01:12,  6.08s/it]
Training Epoch: 5/100, batch 100/125 completed (loss: 0.1464:  81%|████████  | 50/62 [05:02<01:12,  6.08s/it]
Training Epoch: 5/100, batch 100/125 completed (loss: 0.1647:  82%|████████▏ | 51/62 [05:05<01:06,  6.05s/it]
Training Epoch: 5/100, batch 100/125 completed (loss: 0.1754:  82%|████████▏ | 51/62 [05:05<01:06,  6.05s/it]
Training Epoch: 5/100, batch 100/125 completed (loss: 0.1142:  82%|████████▏ | 51/62 [05:05<01:06,  6.05s/it]
Training Epoch: 5/100, batch 100/125 completed (loss: 0.1464:  82%|████████▏ | 51/62 [05:05<01:06,  6.05s/it]
Training Epoch: 5/100, batch 101/125 completed (loss: 0.434:  82%|████████▏ | 51/62 [05:05<01:06,  6.05s/it] 
Training Epoch: 5/100, batch 101/125 completed (loss: 0.1056:  82%|████████▏ | 51/62 [05:05<01:06,  6.05s/it]
Training Epoch: 5/100, batch 101/125 completed (loss: 0.1034:  82%|████████▏ | 51/62 [05:05<01:06,  6.05s/it]
Training Epoch: 5/100, batch 101/125 completed (loss: 0.2097:  82%|████████▏ | 51/62 [05:05<01:06,  6.05s/it]
Training Epoch: 5/100, batch 102/125 completed (loss: 0.1481:  82%|████████▏ | 51/62 [05:08<01:06,  6.05s/it]
Training Epoch: 5/100, batch 102/125 completed (loss: 0.1889:  82%|████████▏ | 51/62 [05:08<01:06,  6.05s/it]
Training Epoch: 5/100, batch 102/125 completed (loss: 0.0925:  82%|████████▏ | 51/62 [05:08<01:06,  6.05s/it]
Training Epoch: 5/100, batch 102/125 completed (loss: 0.1185:  82%|████████▏ | 51/62 [05:08<01:06,  6.05s/it]
Training Epoch: 5/100, batch 102/125 completed (loss: 0.1185:  84%|████████▍ | 52/62 [05:11<01:01,  6.11s/it]
Training Epoch: 5/100, batch 102/125 completed (loss: 0.0925:  84%|████████▍ | 52/62 [05:11<01:01,  6.12s/it]
Training Epoch: 5/100, batch 102/125 completed (loss: 0.1481:  84%|████████▍ | 52/62 [05:11<01:01,  6.12s/it]
Training Epoch: 5/100, batch 102/125 completed (loss: 0.1889:  84%|████████▍ | 52/62 [05:11<01:01,  6.12s/it]
Training Epoch: 5/100, batch 103/125 completed (loss: 0.1579:  84%|████████▍ | 52/62 [05:11<01:01,  6.12s/it]
Training Epoch: 5/100, batch 103/125 completed (loss: 0.3051:  84%|████████▍ | 52/62 [05:11<01:01,  6.12s/it]
Training Epoch: 5/100, batch 103/125 completed (loss: 0.0789:  84%|████████▍ | 52/62 [05:11<01:01,  6.12s/it]
Training Epoch: 5/100, batch 103/125 completed (loss: 0.1842:  84%|████████▍ | 52/62 [05:11<01:01,  6.11s/it]
Training Epoch: 5/100, batch 104/125 completed (loss: 0.2139:  84%|████████▍ | 52/62 [05:14<01:01,  6.12s/it]
Training Epoch: 5/100, batch 104/125 completed (loss: 0.1146:  84%|████████▍ | 52/62 [05:14<01:01,  6.12s/it]
Training Epoch: 5/100, batch 104/125 completed (loss: 0.3159:  84%|████████▍ | 52/62 [05:14<01:01,  6.12s/it]
Training Epoch: 5/100, batch 104/125 completed (loss: 0.4259:  84%|████████▍ | 52/62 [05:14<01:01,  6.11s/it]
Training Epoch: 5/100, batch 104/125 completed (loss: 0.4259:  85%|████████▌ | 53/62 [05:17<00:54,  6.03s/it]
Training Epoch: 5/100, batch 104/125 completed (loss: 0.1146:  85%|████████▌ | 53/62 [05:17<00:54,  6.04s/it]
Training Epoch: 5/100, batch 104/125 completed (loss: 0.2139:  85%|████████▌ | 53/62 [05:17<00:54,  6.04s/it]
Training Epoch: 5/100, batch 104/125 completed (loss: 0.3159:  85%|████████▌ | 53/62 [05:17<00:54,  6.04s/it]
Training Epoch: 5/100, batch 105/125 completed (loss: 0.261:  85%|████████▌ | 53/62 [05:17<00:54,  6.04s/it] 
Training Epoch: 5/100, batch 105/125 completed (loss: 0.2983:  85%|████████▌ | 53/62 [05:17<00:54,  6.04s/it]
Training Epoch: 5/100, batch 105/125 completed (loss: 0.2324:  85%|████████▌ | 53/62 [05:17<00:54,  6.04s/it]
Training Epoch: 5/100, batch 105/125 completed (loss: 0.2301:  85%|████████▌ | 53/62 [05:17<00:54,  6.03s/it]
Training Epoch: 5/100, batch 106/125 completed (loss: 0.097:  85%|████████▌ | 53/62 [05:20<00:54,  6.04s/it] 
Training Epoch: 5/100, batch 106/125 completed (loss: 0.2106:  85%|████████▌ | 53/62 [05:20<00:54,  6.04s/it]
Training Epoch: 5/100, batch 106/125 completed (loss: 0.1546:  85%|████████▌ | 53/62 [05:20<00:54,  6.04s/it]
Training Epoch: 5/100, batch 106/125 completed (loss: 0.1885:  85%|████████▌ | 53/62 [05:20<00:54,  6.03s/it]
Training Epoch: 5/100, batch 106/125 completed (loss: 0.2106:  87%|████████▋ | 54/62 [05:23<00:48,  6.02s/it]
Training Epoch: 5/100, batch 106/125 completed (loss: 0.1546:  87%|████████▋ | 54/62 [05:23<00:48,  6.02s/it]
Training Epoch: 5/100, batch 106/125 completed (loss: 0.1885:  87%|████████▋ | 54/62 [05:23<00:48,  6.02s/it]
Training Epoch: 5/100, batch 106/125 completed (loss: 0.097:  87%|████████▋ | 54/62 [05:23<00:48,  6.02s/it]
Training Epoch: 5/100, batch 107/125 completed (loss: 0.1456:  87%|████████▋ | 54/62 [05:23<00:48,  6.02s/it]
Training Epoch: 5/100, batch 107/125 completed (loss: 0.2014:  87%|████████▋ | 54/62 [05:23<00:48,  6.02s/it]
Training Epoch: 5/100, batch 107/125 completed (loss: 0.241:  87%|████████▋ | 54/62 [05:23<00:48,  6.02s/it] 
Training Epoch: 5/100, batch 107/125 completed (loss: 0.2251:  87%|████████▋ | 54/62 [05:23<00:48,  6.02s/it]
Training Epoch: 5/100, batch 108/125 completed (loss: 0.4145:  87%|████████▋ | 54/62 [05:26<00:48,  6.02s/it]
Training Epoch: 5/100, batch 108/125 completed (loss: 0.2481:  87%|████████▋ | 54/62 [05:26<00:48,  6.02s/it]
Training Epoch: 5/100, batch 108/125 completed (loss: 0.3032:  87%|████████▋ | 54/62 [05:26<00:48,  6.02s/it]
Training Epoch: 5/100, batch 108/125 completed (loss: 0.2953:  87%|████████▋ | 54/62 [05:26<00:48,  6.02s/it]
Training Epoch: 5/100, batch 108/125 completed (loss: 0.4145:  89%|████████▊ | 55/62 [05:29<00:42,  6.09s/it]
Training Epoch: 5/100, batch 108/125 completed (loss: 0.3032:  89%|████████▊ | 55/62 [05:29<00:42,  6.09s/it]
Training Epoch: 5/100, batch 108/125 completed (loss: 0.2481:  89%|████████▊ | 55/62 [05:29<00:42,  6.09s/it]
Training Epoch: 5/100, batch 108/125 completed (loss: 0.2953:  89%|████████▊ | 55/62 [05:29<00:42,  6.09s/it]
Training Epoch: 5/100, batch 109/125 completed (loss: 0.1503:  89%|████████▊ | 55/62 [05:29<00:42,  6.09s/it]
Training Epoch: 5/100, batch 109/125 completed (loss: 0.2489:  89%|████████▊ | 55/62 [05:29<00:42,  6.09s/it]
Training Epoch: 5/100, batch 109/125 completed (loss: 0.3556:  89%|████████▊ | 55/62 [05:29<00:42,  6.09s/it]
Training Epoch: 5/100, batch 109/125 completed (loss: 0.12:  89%|████████▊ | 55/62 [05:29<00:42,  6.09s/it]  
Training Epoch: 5/100, batch 110/125 completed (loss: 0.334:  89%|████████▊ | 55/62 [05:32<00:42,  6.09s/it] 
Training Epoch: 5/100, batch 110/125 completed (loss: 0.2699:  89%|████████▊ | 55/62 [05:32<00:42,  6.09s/it]
Training Epoch: 5/100, batch 110/125 completed (loss: 0.2665:  89%|████████▊ | 55/62 [05:32<00:42,  6.09s/it]
Training Epoch: 5/100, batch 110/125 completed (loss: 0.3108:  89%|████████▊ | 55/62 [05:32<00:42,  6.09s/it]
Training Epoch: 5/100, batch 110/125 completed (loss: 0.2699:  90%|█████████ | 56/62 [05:35<00:36,  6.07s/it]
Training Epoch: 5/100, batch 110/125 completed (loss: 0.3108:  90%|█████████ | 56/62 [05:35<00:36,  6.07s/it]
Training Epoch: 5/100, batch 110/125 completed (loss: 0.2665:  90%|█████████ | 56/62 [05:35<00:36,  6.07s/it]
Training Epoch: 5/100, batch 110/125 completed (loss: 0.334:  90%|█████████ | 56/62 [05:35<00:36,  6.07s/it]
Training Epoch: 5/100, batch 111/125 completed (loss: 0.2485:  90%|█████████ | 56/62 [05:35<00:36,  6.07s/it]
Training Epoch: 5/100, batch 111/125 completed (loss: 0.1063:  90%|█████████ | 56/62 [05:35<00:36,  6.07s/it]
Training Epoch: 5/100, batch 111/125 completed (loss: 0.3246:  90%|█████████ | 56/62 [05:35<00:36,  6.07s/it]
Training Epoch: 5/100, batch 111/125 completed (loss: 0.2287:  90%|█████████ | 56/62 [05:35<00:36,  6.07s/it]
Training Epoch: 5/100, batch 112/125 completed (loss: 0.3513:  90%|█████████ | 56/62 [05:38<00:36,  6.07s/it]
Training Epoch: 5/100, batch 112/125 completed (loss: 0.0711:  90%|█████████ | 56/62 [05:38<00:36,  6.07s/it]
Training Epoch: 5/100, batch 112/125 completed (loss: 0.4562:  90%|█████████ | 56/62 [05:38<00:36,  6.07s/it]
Training Epoch: 5/100, batch 112/125 completed (loss: 0.1224:  90%|█████████ | 56/62 [05:38<00:36,  6.07s/it]
Training Epoch: 5/100, batch 112/125 completed (loss: 0.4562:  92%|█████████▏| 57/62 [05:41<00:30,  6.05s/it]
Training Epoch: 5/100, batch 112/125 completed (loss: 0.1224:  92%|█████████▏| 57/62 [05:41<00:30,  6.05s/it]
Training Epoch: 5/100, batch 112/125 completed (loss: 0.0711:  92%|█████████▏| 57/62 [05:41<00:30,  6.05s/it]
Training Epoch: 5/100, batch 112/125 completed (loss: 0.3513:  92%|█████████▏| 57/62 [05:41<00:30,  6.05s/it]
Training Epoch: 5/100, batch 113/125 completed (loss: 0.2807:  92%|█████████▏| 57/62 [05:41<00:30,  6.05s/it]
Training Epoch: 5/100, batch 113/125 completed (loss: 0.181:  92%|█████████▏| 57/62 [05:41<00:30,  6.05s/it] 
Training Epoch: 5/100, batch 113/125 completed (loss: 0.2977:  92%|█████████▏| 57/62 [05:41<00:30,  6.05s/it]
Training Epoch: 5/100, batch 113/125 completed (loss: 0.4135:  92%|█████████▏| 57/62 [05:41<00:30,  6.05s/it]
Training Epoch: 5/100, batch 114/125 completed (loss: 0.201:  92%|█████████▏| 57/62 [05:44<00:30,  6.05s/it]
Training Epoch: 5/100, batch 114/125 completed (loss: 0.1882:  92%|█████████▏| 57/62 [05:44<00:30,  6.05s/it]
Training Epoch: 5/100, batch 114/125 completed (loss: 0.114:  92%|█████████▏| 57/62 [05:44<00:30,  6.05s/it] 
Training Epoch: 5/100, batch 114/125 completed (loss: 0.4178:  92%|█████████▏| 57/62 [05:44<00:30,  6.05s/it]
Training Epoch: 5/100, batch 114/125 completed (loss: 0.114:  94%|█████████▎| 58/62 [05:47<00:24,  6.03s/it]
Training Epoch: 5/100, batch 114/125 completed (loss: 0.201:  94%|█████████▎| 58/62 [05:47<00:24,  6.03s/it]
Training Epoch: 5/100, batch 114/125 completed (loss: 0.4178:  94%|█████████▎| 58/62 [05:47<00:24,  6.03s/it]
Training Epoch: 5/100, batch 114/125 completed (loss: 0.1882:  94%|█████████▎| 58/62 [05:47<00:24,  6.03s/it]
Training Epoch: 5/100, batch 115/125 completed (loss: 0.159:  94%|█████████▎| 58/62 [05:47<00:24,  6.03s/it] 
Training Epoch: 5/100, batch 115/125 completed (loss: 0.2675:  94%|█████████▎| 58/62 [05:47<00:24,  6.03s/it]
Training Epoch: 5/100, batch 115/125 completed (loss: 0.3072:  94%|█████████▎| 58/62 [05:47<00:24,  6.03s/it]
Training Epoch: 5/100, batch 115/125 completed (loss: 0.4367:  94%|█████████▎| 58/62 [05:47<00:24,  6.03s/it]
Training Epoch: 5/100, batch 116/125 completed (loss: 0.3209:  94%|█████████▎| 58/62 [05:50<00:24,  6.03s/it]
Training Epoch: 5/100, batch 116/125 completed (loss: 0.2076:  94%|█████████▎| 58/62 [05:50<00:24,  6.03s/it]
Training Epoch: 5/100, batch 116/125 completed (loss: 0.1175:  94%|█████████▎| 58/62 [05:50<00:24,  6.03s/it]
Training Epoch: 5/100, batch 116/125 completed (loss: 0.0467:  94%|█████████▎| 58/62 [05:50<00:24,  6.03s/it]
Training Epoch: 5/100, batch 116/125 completed (loss: 0.0467:  95%|█████████▌| 59/62 [05:53<00:18,  6.03s/it]
Training Epoch: 5/100, batch 116/125 completed (loss: 0.3209:  95%|█████████▌| 59/62 [05:53<00:18,  6.03s/it]
Training Epoch: 5/100, batch 116/125 completed (loss: 0.2076:  95%|█████████▌| 59/62 [05:53<00:18,  6.03s/it]
Training Epoch: 5/100, batch 116/125 completed (loss: 0.1175:  95%|█████████▌| 59/62 [05:53<00:18,  6.03s/it]
Training Epoch: 5/100, batch 117/125 completed (loss: 0.2077:  95%|█████████▌| 59/62 [05:53<00:18,  6.03s/it]
Training Epoch: 5/100, batch 117/125 completed (loss: 0.3229:  95%|█████████▌| 59/62 [05:53<00:18,  6.03s/it]
Training Epoch: 5/100, batch 117/125 completed (loss: 0.2161:  95%|█████████▌| 59/62 [05:53<00:18,  6.03s/it]
Training Epoch: 5/100, batch 117/125 completed (loss: 0.1169:  95%|█████████▌| 59/62 [05:53<00:18,  6.03s/it]
Training Epoch: 5/100, batch 118/125 completed (loss: 0.1886:  95%|█████████▌| 59/62 [05:56<00:18,  6.03s/it]
Training Epoch: 5/100, batch 118/125 completed (loss: 0.0912:  95%|█████████▌| 59/62 [05:56<00:18,  6.03s/it]
Training Epoch: 5/100, batch 118/125 completed (loss: 0.203:  95%|█████████▌| 59/62 [05:56<00:18,  6.03s/it] 
Training Epoch: 5/100, batch 118/125 completed (loss: 0.161:  95%|█████████▌| 59/62 [05:56<00:18,  6.03s/it] 
Training Epoch: 5/100, batch 118/125 completed (loss: 0.161:  97%|█████████▋| 60/62 [06:00<00:12,  6.10s/it]
Training Epoch: 5/100, batch 118/125 completed (loss: 0.0912:  97%|█████████▋| 60/62 [06:00<00:12,  6.10s/it]
Training Epoch: 5/100, batch 118/125 completed (loss: 0.1886:  97%|█████████▋| 60/62 [06:00<00:12,  6.10s/it]
Training Epoch: 5/100, batch 118/125 completed (loss: 0.203:  97%|█████████▋| 60/62 [06:00<00:12,  6.10s/it]
Training Epoch: 5/100, batch 119/125 completed (loss: 0.2525:  97%|█████████▋| 60/62 [06:00<00:12,  6.10s/it]
Training Epoch: 5/100, batch 119/125 completed (loss: 0.3822:  97%|█████████▋| 60/62 [06:00<00:12,  6.10s/it]
Training Epoch: 5/100, batch 119/125 completed (loss: 0.3286:  97%|█████████▋| 60/62 [06:00<00:12,  6.10s/it]
Training Epoch: 5/100, batch 119/125 completed (loss: 0.3795:  97%|█████████▋| 60/62 [06:00<00:12,  6.10s/it]
Training Epoch: 5/100, batch 120/125 completed (loss: 0.3191:  97%|█████████▋| 60/62 [06:03<00:12,  6.10s/it]
Training Epoch: 5/100, batch 120/125 completed (loss: 0.3723:  97%|█████████▋| 60/62 [06:03<00:12,  6.10s/it]
Training Epoch: 5/100, batch 120/125 completed (loss: 0.2067:  97%|█████████▋| 60/62 [06:03<00:12,  6.10s/it]
Training Epoch: 5/100, batch 120/125 completed (loss: 0.3018:  97%|█████████▋| 60/62 [06:03<00:12,  6.10s/it]
Training Epoch: 5/100, batch 120/125 completed (loss: 0.3191:  98%|█████████▊| 61/62 [06:05<00:06,  6.01s/it]
Training Epoch: 5/100, batch 120/125 completed (loss: 0.2067:  98%|█████████▊| 61/62 [06:05<00:06,  6.01s/it]
Training Epoch: 5/100, batch 120/125 completed (loss: 0.3018:  98%|█████████▊| 61/62 [06:05<00:06,  6.02s/it]
Training Epoch: 5/100, batch 120/125 completed (loss: 0.3723:  98%|█████████▊| 61/62 [06:05<00:06,  6.01s/it]
Training Epoch: 5/100, batch 121/125 completed (loss: 0.4:  98%|█████████▊| 61/62 [06:05<00:06,  6.01s/it]   
Training Epoch: 5/100, batch 121/125 completed (loss: 0.2325:  98%|█████████▊| 61/62 [06:05<00:06,  6.01s/it]
Training Epoch: 5/100, batch 121/125 completed (loss: 0.357:  98%|█████████▊| 61/62 [06:05<00:06,  6.01s/it] 
Training Epoch: 5/100, batch 121/125 completed (loss: 0.4623:  98%|█████████▊| 61/62 [06:05<00:06,  6.02s/it]
Training Epoch: 5/100, batch 122/125 completed (loss: 0.2769:  98%|█████████▊| 61/62 [06:09<00:06,  6.01s/it]
Training Epoch: 5/100, batch 122/125 completed (loss: 0.2045:  98%|█████████▊| 61/62 [06:09<00:06,  6.01s/it]
Training Epoch: 5/100, batch 122/125 completed (loss: 0.2032:  98%|█████████▊| 61/62 [06:09<00:06,  6.01s/it]
Training Epoch: 5/100, batch 122/125 completed (loss: 0.087:  98%|█████████▊| 61/62 [06:09<00:06,  6.02s/it] 
Training Epoch: 5/100, batch 122/125 completed (loss: 0.2045: 100%|██████████| 62/62 [06:11<00:00,  6.01s/it]
Training Epoch: 5/100, batch 122/125 completed (loss: 0.2769: 100%|██████████| 62/62 [06:11<00:00,  6.01s/it]
Training Epoch: 5/100, batch 122/125 completed (loss: 0.087: 100%|██████████| 62/62 [06:11<00:00,  6.01s/it]
Training Epoch: 5/100, batch 122/125 completed (loss: 0.2032: 100%|██████████| 62/62 [06:11<00:00,  6.01s/it]
Training Epoch: 5/100, batch 123/125 completed (loss: 0.1328: 100%|██████████| 62/62 [06:11<00:00,  6.01s/it]
Training Epoch: 5/100, batch 123/125 completed (loss: 0.2982: 100%|██████████| 62/62 [06:11<00:00,  6.01s/it]
Training Epoch: 5/100, batch 123/125 completed (loss: 0.0771: 100%|██████████| 62/62 [06:11<00:00,  6.01s/it]
Training Epoch: 5/100, batch 123/125 completed (loss: 0.062: 100%|██████████| 62/62 [06:11<00:00,  6.01s/it]
Training Epoch: 5/100, batch 123/125 completed (loss: 0.2982: : 63it [06:15,  5.20s/it]                      
Training Epoch: 5/100, batch 123/125 completed (loss: 0.062: : 63it [06:15,  5.20s/it]                      
Training Epoch: 5/100, batch 123/125 completed (loss: 0.0771: : 63it [06:15,  5.20s/it]                      
Training Epoch: 5/100, batch 123/125 completed (loss: 0.1328: : 63it [06:15,  5.20s/it]                      
Training Epoch: 5/100, batch 124/125 completed (loss: 0.3475: : 63it [06:15,  5.20s/it]
Training Epoch: 5/100, batch 124/125 completed (loss: 0.1925: : 63it [06:15,  5.20s/it]
Training Epoch: 5/100, batch 124/125 completed (loss: 0.3191: : 63it [06:15,  5.20s/it]
Training Epoch: 5/100, batch 124/125 completed (loss: 0.2284: : 63it [06:15,  5.20s/it]
Training Epoch: 5/100, batch 124/125 completed (loss: 0.3475: : 63it [06:15,  5.96s/it]

Training Epoch: 5/100, batch 124/125 completed (loss: 0.3191: : 63it [06:15,  5.96s/it]

Training Epoch: 5/100, batch 124/125 completed (loss: 0.1925: : 63it [06:15,  5.96s/it]

Training Epoch: 5/100, batch 124/125 completed (loss: 0.2284: : 63it [06:15,  5.96s/it]
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/c10d_logger.py:83: UserWarning: barrier(): using the device under current context. You can specify `device_id` in `init_process_group` to mute this warning.
  return func(*args, **kwargs)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
saving model + train state.
Warning: failed to upload checkpoint_5 to HF: (Request ID: Root=1-6a1b6a8e-13546b656e9dd1a74e25275d;15956abf-ee9d-4e9f-a7c4-a44e8fe9bb9e)

403 Forbidden: You need to setup automatic credit recharge in order to upload more data. You can do so at /settings/billing..
Cannot access content at: https://huggingface.co/seyedparsa/thoughtformer-checkpoints.git/info/lfs/objects/batch.
Make sure your token has the correct permissions.
Warning: failed to upload train_state_5 to HF: (Request ID: Root=1-6a1b6a92-6e9af3d718c7a1a84f925af2;e2cc7bfa-32f0-4d5d-a6c8-7888b5c3b9c0)

403 Forbidden: You need to setup automatic credit recharge in order to upload more data. You can do so at /settings/billing..
Cannot access content at: https://huggingface.co/seyedparsa/thoughtformer-checkpoints.git/info/lfs/objects/batch.
Make sure your token has the correct permissions.
Deleted old checkpoint: train_state_4
Deleted old checkpoint: checkpoint_4
[warn] HF upload failed for ckpt_5; kept locally, earlier ckpts deleted.

Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]
Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]
Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]eval loss 0.1396448050553982

Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]
[WRONG] Example 0 (diff=3)
  Expected : 'Claude'
  Extracted: 'Morgan'
  Generated: '### Morgan'

Test Accuracy:   3%|▎         | 1/32 [00:06<03:25,  6.63s/it]
Test Accuracy:   3%|▎         | 1/32 [00:06<03:25,  6.63s/it]
Test Accuracy:   3%|▎         | 1/32 [00:06<03:25,  6.63s/it]
Test accuracy: 0.75:   3%|▎         | 1/32 [00:06<03:25,  6.63s/it]
Test accuracy: 0.38:   3%|▎         | 1/32 [00:06<03:25,  6.63s/it]
Test accuracy: 0.75:   3%|▎         | 1/32 [00:06<03:25,  6.63s/it]
[WRONG] Example 1 (diff=8)
  Expected : 'Guy'
  Extracted: 'Irma'
  Generated: '### Irma'

[WRONG] Example 2 (diff=10)
  Expected : 'Elvis'
  Extracted: 'Shannon'
  Generated: '### Shannon'

[CORRECT] Example 3 (diff=8)
  Expected : 'Coral'
  Extracted: 'Coral'
  Generated: '### Coral'

[CORRECT] Example 4 (diff=1)
  Expected : 'Patricia'
  Extracted: 'Patricia'
  Generated: '### Patricia'

Test Accuracy:   3%|▎         | 1/32 [00:06<03:25,  6.62s/it]
Test accuracy: 0.5:   3%|▎         | 1/32 [00:06<03:25,  6.62s/it]
Test accuracy: 0.75:   6%|▋         | 2/32 [00:13<03:18,  6.61s/it]
Test accuracy: 0.81:   6%|▋         | 2/32 [00:13<03:18,  6.61s/it]
Test accuracy: 0.75:   6%|▋         | 2/32 [00:13<03:18,  6.61s/it]
Test accuracy: 0.38:   6%|▋         | 2/32 [00:13<03:18,  6.61s/it]
Test accuracy: 0.88:   6%|▋         | 2/32 [00:13<03:18,  6.61s/it]
Test accuracy: 0.56:   6%|▋         | 2/32 [00:13<03:18,  6.61s/it]
Test accuracy: 0.5:   6%|▋         | 2/32 [00:13<03:15,  6.52s/it]
Test accuracy: 0.44:   6%|▋         | 2/32 [00:13<03:15,  6.52s/it]
Test accuracy: 0.81:   9%|▉         | 3/32 [00:20<03:16,  6.78s/it]
Test accuracy: 0.88:   9%|▉         | 3/32 [00:20<03:16,  6.78s/it]
Test accuracy: 0.75:   9%|▉         | 3/32 [00:20<03:16,  6.78s/it]
Test accuracy: 0.83:   9%|▉         | 3/32 [00:20<03:16,  6.78s/it]
Test accuracy: 0.44:   9%|▉         | 3/32 [00:20<03:15,  6.73s/it]
Test accuracy: 0.42:   9%|▉         | 3/32 [00:20<03:15,  6.73s/it]
Test accuracy: 0.56:   9%|▉         | 3/32 [00:20<03:16,  6.79s/it]
Test accuracy: 0.54:   9%|▉         | 3/32 [00:20<03:16,  6.79s/it]
Test accuracy: 0.54:  12%|█▎        | 4/32 [00:26<03:03,  6.55s/it]
Test accuracy: 0.83:  12%|█▎        | 4/32 [00:26<03:03,  6.55s/it]
Test accuracy: 0.75:  12%|█▎        | 4/32 [00:26<03:03,  6.55s/it]
Test accuracy: 0.59:  12%|█▎        | 4/32 [00:26<03:03,  6.55s/it]
Test accuracy: 0.75:  12%|█▎        | 4/32 [00:26<03:03,  6.55s/it]
Test accuracy: 0.69:  12%|█▎        | 4/32 [00:26<03:03,  6.55s/it]
Test accuracy: 0.42:  12%|█▎        | 4/32 [00:26<03:02,  6.52s/it]
Test accuracy: 0.47:  12%|█▎        | 4/32 [00:26<03:02,  6.52s/it]
Test accuracy: 0.75:  16%|█▌        | 5/32 [00:32<02:55,  6.49s/it]
Test accuracy: 0.69:  16%|█▌        | 5/32 [00:32<02:55,  6.49s/it]
Test accuracy: 0.75:  16%|█▌        | 5/32 [00:32<02:55,  6.49s/it]
Test accuracy: 0.59:  16%|█▌        | 5/32 [00:32<02:55,  6.49s/it]
Test accuracy: 0.65:  16%|█▌        | 5/32 [00:32<02:55,  6.49s/it]
Test accuracy: 0.6:  16%|█▌        | 5/32 [00:32<02:55,  6.49s/it] 
Test accuracy: 0.47:  16%|█▌        | 5/32 [00:32<02:54,  6.47s/it]
Test accuracy: 0.55:  16%|█▌        | 5/32 [00:32<02:54,  6.47s/it]
Test accuracy: 0.6:  19%|█▉        | 6/32 [00:39<02:46,  6.40s/it]
Test accuracy: 0.75:  19%|█▉        | 6/32 [00:39<02:46,  6.40s/it]
Test accuracy: 0.6:  19%|█▉        | 6/32 [00:39<02:46,  6.40s/it]
Test accuracy: 0.65:  19%|█▉        | 6/32 [00:39<02:46,  6.40s/it]
Test accuracy: 0.69:  19%|█▉        | 6/32 [00:39<02:46,  6.40s/it]
Test accuracy: 0.55:  19%|█▉        | 6/32 [00:38<02:46,  6.39s/it]
Test accuracy: 0.69:  19%|█▉        | 6/32 [00:39<02:46,  6.40s/it]
Test accuracy: 0.62:  19%|█▉        | 6/32 [00:38<02:46,  6.39s/it]
Test accuracy: 0.69:  22%|██▏       | 7/32 [00:45<02:38,  6.33s/it]
Test accuracy: 0.69:  22%|██▏       | 7/32 [00:45<02:38,  6.33s/it]
Test accuracy: 0.66:  22%|██▏       | 7/32 [00:45<02:38,  6.33s/it]
Test accuracy: 0.62:  22%|██▏       | 7/32 [00:45<02:38,  6.33s/it]
Test accuracy: 0.6:  22%|██▏       | 7/32 [00:45<02:38,  6.33s/it]
Test accuracy: 0.59:  22%|██▏       | 7/32 [00:45<02:38,  6.33s/it]
Test accuracy: 0.62:  22%|██▏       | 7/32 [00:45<02:38,  6.32s/it]
Test accuracy: 0.62:  22%|██▏       | 7/32 [00:45<02:38,  6.32s/it]
Test accuracy: 0.59:  25%|██▌       | 8/32 [00:51<02:33,  6.40s/it]
Test accuracy: 0.62:  25%|██▌       | 8/32 [00:51<02:33,  6.40s/it]
Test accuracy: 0.59:  25%|██▌       | 8/32 [00:51<02:33,  6.40s/it]
Test accuracy: 0.64:  25%|██▌       | 8/32 [00:51<02:33,  6.40s/it]
Test accuracy: 0.66:  25%|██▌       | 8/32 [00:51<02:33,  6.40s/it]
Test accuracy: 0.62:  25%|██▌       | 8/32 [00:51<02:33,  6.40s/it]
Test accuracy: 0.62:  25%|██▌       | 8/32 [00:51<02:33,  6.40s/it]
Test accuracy: 0.64:  25%|██▌       | 8/32 [00:51<02:33,  6.40s/it]
Test accuracy: 0.59:  28%|██▊       | 9/32 [00:58<02:29,  6.51s/it]
Test accuracy: 0.58:  28%|██▊       | 9/32 [00:58<02:29,  6.51s/it]
Test accuracy: 0.64:  28%|██▊       | 9/32 [00:58<02:29,  6.51s/it]
Test accuracy: 0.64:  28%|██▊       | 9/32 [00:58<02:29,  6.51s/it]
Test accuracy: 0.65:  28%|██▊       | 9/32 [00:58<02:29,  6.51s/it]
Test accuracy: 0.64:  28%|██▊       | 9/32 [00:58<02:29,  6.51s/it]
Test accuracy: 0.62:  28%|██▊       | 9/32 [00:58<02:29,  6.52s/it]
Test accuracy: 0.62:  28%|██▊       | 9/32 [00:58<02:29,  6.52s/it]
Test accuracy: 0.64:  31%|███▏      | 10/32 [01:04<02:21,  6.41s/it]
Test accuracy: 0.58:  31%|███▏      | 10/32 [01:04<02:21,  6.42s/it]
Test accuracy: 0.59:  31%|███▏      | 10/32 [01:04<02:21,  6.42s/it]
Test accuracy: 0.6:  31%|███▏      | 10/32 [01:04<02:21,  6.41s/it] 
Test accuracy: 0.65:  31%|███▏      | 10/32 [01:04<02:21,  6.42s/it]
Test accuracy: 0.62:  31%|███▏      | 10/32 [01:04<02:21,  6.41s/it]
Test accuracy: 0.68:  31%|███▏      | 10/32 [01:04<02:21,  6.42s/it]
Test accuracy: 0.61:  31%|███▏      | 10/32 [01:04<02:21,  6.41s/it]
Test accuracy: 0.61:  34%|███▍      | 11/32 [01:11<02:15,  6.45s/it]
Test accuracy: 0.58:  34%|███▍      | 11/32 [01:11<02:15,  6.45s/it]
Test accuracy: 0.59:  34%|███▍      | 11/32 [01:11<02:15,  6.46s/it]
Test accuracy: 0.68:  34%|███▍      | 11/32 [01:11<02:15,  6.46s/it]
Test accuracy: 0.6:  34%|███▍      | 11/32 [01:11<02:15,  6.45s/it]
Test accuracy: 0.58:  34%|███▍      | 11/32 [01:11<02:15,  6.46s/it]
Test accuracy: 0.67:  34%|███▍      | 11/32 [01:11<02:15,  6.46s/it]
Test accuracy: 0.58:  34%|███▍      | 11/32 [01:11<02:15,  6.45s/it]
Test accuracy: 0.67:  38%|███▊      | 12/32 [01:17<02:07,  6.37s/it]
Test accuracy: 0.58:  38%|███▊      | 12/32 [01:17<02:07,  6.37s/it]
Test accuracy: 0.66:  38%|███▊      | 12/32 [01:17<02:07,  6.37s/it]
Test accuracy: 0.58:  38%|███▊      | 12/32 [01:17<02:07,  6.37s/it]
Test accuracy: 0.58:  38%|███▊      | 12/32 [01:17<02:07,  6.37s/it]
Test accuracy: 0.58:  38%|███▊      | 12/32 [01:17<02:07,  6.37s/it]
Test accuracy: 0.58:  38%|███▊      | 12/32 [01:17<02:07,  6.37s/it]
Test accuracy: 0.57:  38%|███▊      | 12/32 [01:17<02:07,  6.37s/it]
Test accuracy: 0.66:  41%|████      | 13/32 [01:23<01:59,  6.32s/it]
Test accuracy: 0.58:  41%|████      | 13/32 [01:23<01:59,  6.32s/it]
Test accuracy: 0.6:  41%|████      | 13/32 [01:23<01:59,  6.32s/it] 
Test accuracy: 0.65:  41%|████      | 13/32 [01:23<01:59,  6.32s/it]
Test accuracy: 0.58:  41%|████      | 13/32 [01:23<01:59,  6.32s/it]
Test accuracy: 0.57:  41%|████      | 13/32 [01:23<01:59,  6.31s/it]
Test accuracy: 0.58:  41%|████      | 13/32 [01:23<01:59,  6.32s/it]
Test accuracy: 0.57:  41%|████      | 13/32 [01:23<01:59,  6.31s/it]
Test accuracy: 0.57:  44%|████▍     | 14/32 [01:29<01:54,  6.34s/it]
Test accuracy: 0.58:  44%|████▍     | 14/32 [01:30<01:54,  6.34s/it]
Test accuracy: 0.65:  44%|████▍     | 14/32 [01:30<01:54,  6.34s/it]
Test accuracy: 0.56:  44%|████▍     | 14/32 [01:29<01:54,  6.34s/it]
Test accuracy: 0.6:  44%|████▍     | 14/32 [01:30<01:54,  6.34s/it]
Test accuracy: 0.56:  44%|████▍     | 14/32 [01:30<01:54,  6.34s/it]
Test accuracy: 0.64:  44%|████▍     | 14/32 [01:30<01:54,  6.34s/it]
Test accuracy: 0.59:  44%|████▍     | 14/32 [01:30<01:54,  6.34s/it]
Test accuracy: 0.59:  47%|████▋     | 15/32 [01:36<01:47,  6.29s/it]
Test accuracy: 0.56:  47%|████▋     | 15/32 [01:36<01:46,  6.29s/it]
Test accuracy: 0.59:  47%|████▋     | 15/32 [01:36<01:47,  6.29s/it]
Test accuracy: 0.56:  47%|████▋     | 15/32 [01:36<01:47,  6.29s/it]
Test accuracy: 0.54:  47%|████▋     | 15/32 [01:36<01:46,  6.29s/it]
Test accuracy: 0.57:  47%|████▋     | 15/32 [01:36<01:47,  6.29s/it]
Test accuracy: 0.64:  47%|████▋     | 15/32 [01:36<01:47,  6.29s/it]
Test accuracy: 0.65:  47%|████▋     | 15/32 [01:36<01:47,  6.29s/it]
Test accuracy: 0.65:  50%|█████     | 16/32 [01:42<01:41,  6.32s/it]
Test accuracy: 0.59:  50%|█████     | 16/32 [01:42<01:41,  6.32s/it]
Test accuracy: 0.66:  50%|█████     | 16/32 [01:42<01:41,  6.32s/it]
Test accuracy: 0.59:  50%|█████     | 16/32 [01:42<01:41,  6.32s/it]
Test accuracy: 0.57:  50%|█████     | 16/32 [01:42<01:41,  6.32s/it]
Test accuracy: 0.55:  50%|█████     | 16/32 [01:42<01:41,  6.32s/it]
Test accuracy: 0.54:  50%|█████     | 16/32 [01:42<01:41,  6.32s/it]
Test accuracy: 0.53:  50%|█████     | 16/32 [01:42<01:41,  6.32s/it]
Test accuracy: 0.59:  53%|█████▎    | 17/32 [01:48<01:35,  6.34s/it]
Test accuracy: 0.59:  53%|█████▎    | 17/32 [01:48<01:35,  6.34s/it]
Test accuracy: 0.53:  53%|█████▎    | 17/32 [01:48<01:35,  6.34s/it]
Test accuracy: 0.66:  53%|█████▎    | 17/32 [01:48<01:35,  6.34s/it]
Test accuracy: 0.66:  53%|█████▎    | 17/32 [01:48<01:35,  6.34s/it]
Test accuracy: 0.55:  53%|█████▎    | 17/32 [01:48<01:35,  6.34s/it]
Test accuracy: 0.55:  53%|█████▎    | 17/32 [01:48<01:35,  6.34s/it]
Test accuracy: 0.54:  53%|█████▎    | 17/32 [01:48<01:35,  6.34s/it]
Test accuracy: 0.66:  56%|█████▋    | 18/32 [01:55<01:28,  6.29s/it]
Test accuracy: 0.55:  56%|█████▋    | 18/32 [01:55<01:28,  6.29s/it]
Test accuracy: 0.59:  56%|█████▋    | 18/32 [01:55<01:28,  6.29s/it]
Test accuracy: 0.65:  56%|█████▋    | 18/32 [01:55<01:28,  6.29s/it]
Test accuracy: 0.6:  56%|█████▋    | 18/32 [01:55<01:28,  6.29s/it] 
Test accuracy: 0.54:  56%|█████▋    | 18/32 [01:55<01:28,  6.29s/it]
Test accuracy: 0.56:  56%|█████▋    | 18/32 [01:55<01:28,  6.29s/it]
Test accuracy: 0.53:  56%|█████▋    | 18/32 [01:55<01:28,  6.29s/it]
Test accuracy: 0.65:  59%|█████▉    | 19/32 [02:01<01:22,  6.32s/it]
Test accuracy: 0.63:  59%|█████▉    | 19/32 [02:01<01:22,  6.32s/it]
Test accuracy: 0.6:  59%|█████▉    | 19/32 [02:01<01:22,  6.32s/it]
Test accuracy: 0.56:  59%|█████▉    | 19/32 [02:01<01:22,  6.32s/it]
Test accuracy: 0.58:  59%|█████▉    | 19/32 [02:01<01:22,  6.32s/it]
Test accuracy: 0.53:  59%|█████▉    | 19/32 [02:01<01:22,  6.32s/it]
Test accuracy: 0.57:  59%|█████▉    | 19/32 [02:01<01:22,  6.32s/it]
Test accuracy: 0.55:  59%|█████▉    | 19/32 [02:01<01:22,  6.32s/it]
Test accuracy: 0.57:  62%|██████▎   | 20/32 [02:07<01:15,  6.29s/it]
Test accuracy: 0.58:  62%|██████▎   | 20/32 [02:07<01:15,  6.29s/it]
Test accuracy: 0.57:  62%|██████▎   | 20/32 [02:07<01:15,  6.29s/it]
Test accuracy: 0.59:  62%|██████▎   | 20/32 [02:07<01:15,  6.29s/it]
Test accuracy: 0.55:  62%|██████▎   | 20/32 [02:07<01:15,  6.29s/it]
Test accuracy: 0.63:  62%|██████▎   | 20/32 [02:07<01:15,  6.29s/it]
Test accuracy: 0.55:  62%|██████▎   | 20/32 [02:07<01:15,  6.29s/it]
Test accuracy: 0.63:  62%|██████▎   | 20/32 [02:07<01:15,  6.29s/it]
Test accuracy: 0.55:  66%|██████▌   | 21/32 [02:14<01:10,  6.44s/it]
Test accuracy: 0.63:  66%|██████▌   | 21/32 [02:14<01:10,  6.44s/it]
Test accuracy: 0.54:  66%|██████▌   | 21/32 [02:14<01:10,  6.44s/it]
Test accuracy: 0.59:  66%|██████▌   | 21/32 [02:14<01:10,  6.44s/it]
Test accuracy: 0.62:  66%|██████▌   | 21/32 [02:14<01:10,  6.44s/it]
Test accuracy: 0.58:  66%|██████▌   | 21/32 [02:14<01:10,  6.44s/it]
Test accuracy: 0.57:  66%|██████▌   | 21/32 [02:14<01:10,  6.45s/it]
Test accuracy: 0.59:  66%|██████▌   | 21/32 [02:14<01:10,  6.45s/it]
Test accuracy: 0.59:  69%|██████▉   | 22/32 [02:21<01:04,  6.48s/it]
Test accuracy: 0.58:  69%|██████▉   | 22/32 [02:21<01:04,  6.48s/it]
Test accuracy: 0.58:  69%|██████▉   | 22/32 [02:20<01:04,  6.48s/it]
Test accuracy: 0.59:  69%|██████▉   | 22/32 [02:20<01:04,  6.48s/it]
Test accuracy: 0.62:  69%|██████▉   | 22/32 [02:21<01:04,  6.48s/it]
Test accuracy: 0.54:  69%|██████▉   | 22/32 [02:21<01:04,  6.48s/it]
Test accuracy: 0.62:  69%|██████▉   | 22/32 [02:21<01:04,  6.48s/it]
Test accuracy: 0.53:  69%|██████▉   | 22/32 [02:21<01:04,  6.48s/it]
Test accuracy: 0.53:  72%|███████▏  | 23/32 [02:27<00:57,  6.39s/it]
Test accuracy: 0.62:  72%|███████▏  | 23/32 [02:27<00:57,  6.39s/it]
Test accuracy: 0.55:  72%|███████▏  | 23/32 [02:27<00:57,  6.39s/it]
Test accuracy: 0.6:  72%|███████▏  | 23/32 [02:27<00:57,  6.39s/it] 
Test accuracy: 0.58:  72%|███████▏  | 23/32 [02:27<00:57,  6.39s/it]
Test accuracy: 0.58:  72%|███████▏  | 23/32 [02:27<00:57,  6.39s/it]
Test accuracy: 0.59:  72%|███████▏  | 23/32 [02:27<00:57,  6.39s/it]
Test accuracy: 0.59:  72%|███████▏  | 23/32 [02:27<00:57,  6.39s/it]
Test accuracy: 0.58:  75%|███████▌  | 24/32 [02:34<00:52,  6.57s/it]
Test accuracy: 0.57:  75%|███████▌  | 24/32 [02:34<00:52,  6.57s/it]
Test accuracy: 0.6:  75%|███████▌  | 24/32 [02:34<00:52,  6.58s/it]
Test accuracy: 0.55:  75%|███████▌  | 24/32 [02:34<00:52,  6.58s/it]
Test accuracy: 0.6:  75%|███████▌  | 24/32 [02:34<00:52,  6.58s/it]
Test accuracy: 0.55:  75%|███████▌  | 24/32 [02:34<00:52,  6.58s/it]
Test accuracy: 0.59:  75%|███████▌  | 24/32 [02:34<00:52,  6.58s/it]
Test accuracy: 0.59:  75%|███████▌  | 24/32 [02:34<00:52,  6.58s/it]
Test accuracy: 0.57:  78%|███████▊  | 25/32 [02:40<00:45,  6.52s/it]
Test accuracy: 0.59:  78%|███████▊  | 25/32 [02:40<00:45,  6.51s/it]
Test accuracy: 0.58:  78%|███████▊  | 25/32 [02:40<00:45,  6.52s/it]
Test accuracy: 0.58:  78%|███████▊  | 25/32 [02:40<00:45,  6.51s/it]
Test accuracy: 0.55:  78%|███████▊  | 25/32 [02:40<00:45,  6.52s/it]
Test accuracy: 0.6:  78%|███████▊  | 25/32 [02:40<00:45,  6.52s/it]
Test accuracy: 0.56:  78%|███████▊  | 25/32 [02:40<00:45,  6.52s/it]
Test accuracy: 0.6:  78%|███████▊  | 25/32 [02:40<00:45,  6.52s/it]
Test accuracy: 0.6:  81%|████████▏ | 26/32 [02:46<00:38,  6.42s/it]
Test accuracy: 0.56:  81%|████████▏ | 26/32 [02:46<00:38,  6.42s/it]
Test accuracy: 0.6:  81%|████████▏ | 26/32 [02:46<00:38,  6.42s/it]
Test accuracy: 0.55:  81%|████████▏ | 26/32 [02:46<00:38,  6.42s/it]
Test accuracy: 0.58:  81%|████████▏ | 26/32 [02:46<00:38,  6.42s/it]
Test accuracy: 0.58:  81%|████████▏ | 26/32 [02:46<00:38,  6.42s/it]
Test accuracy: 0.58:  81%|████████▏ | 26/32 [02:46<00:38,  6.41s/it]
Test accuracy: 0.59:  81%|████████▏ | 26/32 [02:46<00:38,  6.41s/it]
Test accuracy: 0.55:  84%|████████▍ | 27/32 [02:53<00:32,  6.41s/it]
Test accuracy: 0.6:  84%|████████▍ | 27/32 [02:53<00:32,  6.41s/it]
Test accuracy: 0.57:  84%|████████▍ | 27/32 [02:53<00:32,  6.41s/it]
Test accuracy: 0.59:  84%|████████▍ | 27/32 [02:53<00:32,  6.41s/it]
Test accuracy: 0.58:  84%|████████▍ | 27/32 [02:53<00:32,  6.41s/it]
Test accuracy: 0.56:  84%|████████▍ | 27/32 [02:53<00:32,  6.41s/it]
Test accuracy: 0.59:  84%|████████▍ | 27/32 [02:53<00:32,  6.41s/it]
Test accuracy: 0.59:  84%|████████▍ | 27/32 [02:53<00:32,  6.41s/it]
Test accuracy: 0.59:  88%|████████▊ | 28/32 [03:00<00:26,  6.58s/it]
Test accuracy: 0.57:  88%|████████▊ | 28/32 [03:00<00:26,  6.58s/it]
Test accuracy: 0.58:  88%|████████▊ | 28/32 [03:00<00:26,  6.58s/it]
Test accuracy: 0.59:  88%|████████▊ | 28/32 [03:00<00:26,  6.58s/it]
Test accuracy: 0.58:  88%|████████▊ | 28/32 [03:00<00:26,  6.58s/it]
Test accuracy: 0.59:  88%|████████▊ | 28/32 [03:00<00:26,  6.58s/it]
Test accuracy: 0.56:  88%|████████▊ | 28/32 [03:00<00:26,  6.58s/it]
Test accuracy: 0.57:  88%|████████▊ | 28/32 [03:00<00:26,  6.58s/it]
Test accuracy: 0.58:  91%|█████████ | 29/32 [03:06<00:19,  6.47s/it]
Test accuracy: 0.57:  91%|█████████ | 29/32 [03:06<00:19,  6.47s/it]
Test accuracy: 0.58:  91%|█████████ | 29/32 [03:06<00:19,  6.47s/it]
Test accuracy: 0.59:  91%|█████████ | 29/32 [03:06<00:19,  6.47s/it]
Test accuracy: 0.57:  91%|█████████ | 29/32 [03:06<00:19,  6.47s/it]
Test accuracy: 0.57:  91%|█████████ | 29/32 [03:06<00:19,  6.47s/it]
Test accuracy: 0.59:  91%|█████████ | 29/32 [03:06<00:19,  6.47s/it]
Test accuracy: 0.59:  91%|█████████ | 29/32 [03:06<00:19,  6.47s/it]
Test accuracy: 0.57:  94%|█████████▍| 30/32 [03:12<00:12,  6.44s/it]
Test accuracy: 0.59:  94%|█████████▍| 30/32 [03:12<00:12,  6.44s/it]
Test accuracy: 0.56:  94%|█████████▍| 30/32 [03:12<00:12,  6.44s/it]
Test accuracy: 0.6:  94%|█████████▍| 30/32 [03:12<00:12,  6.44s/it] 
Test accuracy: 0.57:  94%|█████████▍| 30/32 [03:12<00:12,  6.44s/it]
Test accuracy: 0.57:  94%|█████████▍| 30/32 [03:12<00:12,  6.44s/it]
Test accuracy: 0.59:  94%|█████████▍| 30/32 [03:12<00:12,  6.44s/it]
Test accuracy: 0.59:  94%|█████████▍| 30/32 [03:12<00:12,  6.44s/it]
Test accuracy: 0.57:  97%|█████████▋| 31/32 [03:19<00:06,  6.60s/it]
Test accuracy: 0.6:  97%|█████████▋| 31/32 [03:19<00:06,  6.60s/it]
Test accuracy: 0.58:  97%|█████████▋| 31/32 [03:19<00:06,  6.60s/it]
Test accuracy: 0.6:  97%|█████████▋| 31/32 [03:19<00:06,  6.60s/it]
Test accuracy: 0.59:  97%|█████████▋| 31/32 [03:19<00:06,  6.60s/it]
Test accuracy: 0.59:  97%|█████████▋| 31/32 [03:19<00:06,  6.60s/it]
Test accuracy: 0.56:  97%|█████████▋| 31/32 [03:19<00:06,  6.61s/it]
Test accuracy: 0.56:  97%|█████████▋| 31/32 [03:19<00:06,  6.61s/it]
Test accuracy: 0.59: 100%|██████████| 32/32 [03:22<00:00,  5.33s/it]
Test accuracy: 0.59: 100%|██████████| 32/32 [03:22<00:00,  5.33s/it]
Test accuracy: 0.6: 100%|██████████| 32/32 [03:22<00:00,  5.33s/it]
Test accuracy: 0.6: 100%|██████████| 32/32 [03:22<00:00,  5.33s/it]
Test accuracy: 0.58: 100%|██████████| 32/32 [03:22<00:00,  5.34s/it]
Test accuracy: 0.56: 100%|██████████| 32/32 [03:22<00:00,  5.33s/it]
Test accuracy: 0.58: 100%|██████████| 32/32 [03:22<00:00,  5.34s/it]
Test accuracy: 0.56: 100%|██████████| 32/32 [03:22<00:00,  5.33s/it]
Test accuracy: 0.6: 100%|██████████| 32/32 [03:22<00:00,  6.32s/it]
Test accuracy: 0.56: 100%|██████████| 32/32 [03:22<00:00,  6.32s/it]

Device 2: Cor=141, CoT=0, Total=250Device 1: Cor=149, CoT=0, Total=250


Test accuracy: 0.58: 100%|██████████| 32/32 [03:22<00:00,  6.32s/it]
Device 3: Cor=145, CoT=0, Total=250

Test accuracy: 0.59: 100%|██████████| 32/32 [03:22<00:00,  6.33s/it]
Device 0: Cor=147, CoT=0, Total=250

============================================================
EVALUATION RESULTS (Epoch 4)
============================================================
Config: data_curriculum=ON, thought_curriculum=OFF
thought_stage=10, data_stage=2, n_latent_tokens=10
Training on difficulties: 0-2

Overall accuracy: 582 / 1000 = 0.5820

Per-difficulty accuracy:
  diff=1: 95/100 = 0.9500 (trained)
  diff=2: 75/100 = 0.7500 (trained)
  diff=3: 55/100 = 0.5500 (untrained)
  diff=4: 48/100 = 0.4800 (untrained)
  diff=5: 47/100 = 0.4700 (untrained)
  diff=6: 46/100 = 0.4600 (untrained)
  diff=7: 57/100 = 0.5700 (untrained)
  diff=8: 52/100 = 0.5200 (untrained)
  diff=9: 58/100 = 0.5800 (untrained)
  diff=10: 49/100 = 0.4900 (untrained)

Trained (diff 0-2): 170/200 = 0.8500
============================================================

  Staging check: acc_trained=0.8500, acc_diff2=0.7500 (threshold=0.9)
Warning: failed to upload checkpoint_5 to HF: (Request ID: Root=1-6a1b6b6a-3e650fc92489ada6693915e4;63971a35-fa07-457c-8d55-f72f3bf656e4)

403 Forbidden: You need to setup automatic credit recharge in order to upload more data. You can do so at /settings/billing..
Cannot access content at: https://huggingface.co/seyedparsa/thoughtformer-checkpoints.git/info/lfs/objects/batch.
Make sure your token has the correct permissions.

============================================================
EPOCH 5/100 (thought_stage=10, data_stage=2)
============================================================
  thought_stage=10, c_thought=1, max_difficulty=2

Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):  16%|█▌        | 156/1000 [00:00<00:00, 1012.03 examples/s]
Map (num_proc=4):  16%|█▌        | 156/1000 [00:00<00:00, 943.15 examples/s]
Map (num_proc=4):  16%|█▌        | 156/1000 [00:00<00:00, 922.54 examples/s]
Map (num_proc=4):  81%|████████  | 811/1000 [00:00<00:00, 3626.02 examples/s]
Map (num_proc=4):  81%|████████  | 811/1000 [00:00<00:00, 3443.06 examples/s]
Map (num_proc=4):  15%|█▌        | 153/1000 [00:00<00:01, 786.09 examples/s]
Map (num_proc=4):  81%|████████  | 811/1000 [00:00<00:00, 3298.30 examples/s]
Map (num_proc=4):  81%|████████  | 810/1000 [00:00<00:00, 3037.38 examples/s]
Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2110.20 examples/s]

Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2003.09 examples/s]

Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2021.00 examples/s]

Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 1785.18 examples/s]

Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:04, 1989.51 examples/s]
Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:04, 2067.28 examples/s]
Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:04, 1957.98 examples/s]
Filter (num_proc=4):  20%|██        | 2000/10000 [00:00<00:02, 3868.91 examples/s]
Filter (num_proc=4):  30%|███       | 3000/10000 [00:00<00:01, 4913.14 examples/s]
Filter (num_proc=4):  30%|███       | 3000/10000 [00:00<00:01, 5305.00 examples/s]
Filter (num_proc=4):  30%|███       | 3000/10000 [00:00<00:01, 4903.75 examples/s]
Filter (num_proc=4):  60%|██████    | 6000/10000 [00:00<00:00, 9315.21 examples/s]
Filter (num_proc=4):  50%|█████     | 5000/10000 [00:00<00:00, 9002.32 examples/s]
Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:05, 1724.67 examples/s]
Filter (num_proc=4):  60%|██████    | 6000/10000 [00:00<00:00, 9189.93 examples/s]
Filter (num_proc=4):  75%|███████▌  | 7500/10000 [00:00<00:00, 10518.91 examples/s]
Filter (num_proc=4):  75%|███████▌  | 7500/10000 [00:00<00:00, 12064.57 examples/s]
Filter (num_proc=4):  80%|████████  | 8000/10000 [00:00<00:00, 11085.14 examples/s]
Filter (num_proc=4):  95%|█████████▌| 9500/10000 [00:01<00:00, 11939.64 examples/s]
Filter (num_proc=4):  90%|█████████ | 9000/10000 [00:01<00:00, 12753.23 examples/s]
Filter (num_proc=4):  20%|██        | 2000/10000 [00:00<00:02, 2979.45 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 11028.61 examples/s]
Filter (num_proc=4):  40%|████      | 4000/10000 [00:00<00:01, 5676.91 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 7772.31 examples/s]

Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 7739.28 examples/s] 

Filter (num_proc=4):  60%|██████    | 6000/10000 [00:01<00:00, 7257.67 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 6954.60 examples/s]

Filter (num_proc=4):  85%|████████▌ | 8500/10000 [00:01<00:00, 9529.22 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 8036.51 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 5913.21 examples/s]

Map (num_proc=4):   0%|          | 0/2000 [00:00<?, ? examples/s]
Map (num_proc=4):   7%|▋         | 134/2000 [00:00<00:11, 156.03 examples/s]
Map (num_proc=4):  25%|██▌       | 500/2000 [00:01<00:02, 528.26 examples/s]
Map (num_proc=4):  32%|███▏      | 634/2000 [00:01<00:02, 512.84 examples/s]
Map (num_proc=4):  50%|█████     | 1000/2000 [00:01<00:01, 761.02 examples/s]
Map (num_proc=4):  57%|█████▋    | 1134/2000 [00:01<00:01, 691.44 examples/s]
Map (num_proc=4):  75%|███████▌  | 1500/2000 [00:02<00:00, 885.08 examples/s]
Map (num_proc=4):  82%|████████▏ | 1637/2000 [00:02<00:00, 792.12 examples/s]
Map (num_proc=4): 100%|██████████| 2000/2000 [00:02<00:00, 959.90 examples/s]
Map (num_proc=4): 100%|██████████| 2000/2000 [00:03<00:00, 666.33 examples/s]

Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 1468.43 examples/s]
Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 1385.78 examples/s]
Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 1376.78 examples/s]
Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 1098.74 examples/s]
Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2802.56 examples/s]

Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2711.20 examples/s]

Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2666.02 examples/s]

Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2080.90 examples/s]

Map (num_proc=4):   0%|          | 0/200 [00:00<?, ? examples/s]
Map (num_proc=4):  25%|██▌       | 50/200 [00:00<00:00, 216.93 examples/s]
Map (num_proc=4):  75%|███████▌  | 150/200 [00:00<00:00, 450.22 examples/s]
Map (num_proc=4): 100%|██████████| 200/200 [00:00<00:00, 318.46 examples/s]

Training Epoch: 6:   0%|          | 0/62 [00:00<?, ?it/s]
Training Epoch: 6:   0%|          | 0/62 [00:00<?, ?it/s]
Training Epoch: 6:   0%|          | 0/62 [00:00<?, ?it/s]
Training Epoch: 6:   0%|          | 0/62 [00:00<?, ?it/s]logging training data
rank 2 | gpu 2 | alloc=4.20 GB | reserved=47.74 GB | max=41.98 GB
rank 3 | gpu 3 | alloc=4.20 GB | reserved=47.74 GB | max=41.99 GB
rank 0 | gpu 0 | alloc=4.20 GB | reserved=53.36 GB | max=41.99 GB
rank 1 | gpu 1 | alloc=4.20 GB | reserved=47.73 GB | max=41.98 GB

Training Epoch: 6/100, batch 0/125 completed (loss: 0.2001:   0%|          | 0/62 [00:03<?, ?it/s]
Training Epoch: 6/100, batch 0/125 completed (loss: 0.1102:   0%|          | 0/62 [00:03<?, ?it/s]
Training Epoch: 6/100, batch 0/125 completed (loss: 0.125:   0%|          | 0/62 [00:03<?, ?it/s]
Training Epoch: 6/100, batch 0/125 completed (loss: 0.0645:   0%|          | 0/62 [00:03<?, ?it/s]
Training Epoch: 6/100, batch 0/125 completed (loss: 0.1102:   2%|▏         | 1/62 [00:05<05:56,  5.85s/it]
Training Epoch: 6/100, batch 0/125 completed (loss: 0.2001:   2%|▏         | 1/62 [00:05<05:56,  5.84s/it]
Training Epoch: 6/100, batch 0/125 completed (loss: 0.0645:   2%|▏         | 1/62 [00:05<05:57,  5.86s/it]
Training Epoch: 6/100, batch 0/125 completed (loss: 0.125:   2%|▏         | 1/62 [00:05<05:56,  5.85s/it]
Training Epoch: 6/100, batch 1/125 completed (loss: 0.0531:   2%|▏         | 1/62 [00:05<05:56,  5.85s/it]
Training Epoch: 6/100, batch 1/125 completed (loss: 0.2016:   2%|▏         | 1/62 [00:05<05:56,  5.85s/it]
Training Epoch: 6/100, batch 1/125 completed (loss: 0.1383:   2%|▏         | 1/62 [00:05<05:56,  5.84s/it]
Training Epoch: 6/100, batch 1/125 completed (loss: 0.1293:   2%|▏         | 1/62 [00:05<05:57,  5.86s/it]
Training Epoch: 6/100, batch 2/125 completed (loss: 0.1436:   2%|▏         | 1/62 [00:08<05:56,  5.84s/it]
Training Epoch: 6/100, batch 2/125 completed (loss: 0.1464:   2%|▏         | 1/62 [00:08<05:56,  5.85s/it]
Training Epoch: 6/100, batch 2/125 completed (loss: 0.2207:   2%|▏         | 1/62 [00:08<05:56,  5.85s/it]
Training Epoch: 6/100, batch 2/125 completed (loss: 0.0885:   2%|▏         | 1/62 [00:08<05:57,  5.86s/it]
Training Epoch: 6/100, batch 2/125 completed (loss: 0.1436:   3%|▎         | 2/62 [00:11<05:48,  5.80s/it]
Training Epoch: 6/100, batch 2/125 completed (loss: 0.0885:   3%|▎         | 2/62 [00:11<05:48,  5.81s/it]
Training Epoch: 6/100, batch 2/125 completed (loss: 0.1464:   3%|▎         | 2/62 [00:11<05:48,  5.81s/it]
Training Epoch: 6/100, batch 2/125 completed (loss: 0.2207:   3%|▎         | 2/62 [00:11<05:48,  5.81s/it]
Training Epoch: 6/100, batch 3/125 completed (loss: 0.0794:   3%|▎         | 2/62 [00:11<05:48,  5.81s/it]
Training Epoch: 6/100, batch 3/125 completed (loss: 0.1471:   3%|▎         | 2/62 [00:11<05:48,  5.80s/it]
Training Epoch: 6/100, batch 3/125 completed (loss: 0.1991:   3%|▎         | 2/62 [00:11<05:48,  5.81s/it]
Training Epoch: 6/100, batch 3/125 completed (loss: 0.1571:   3%|▎         | 2/62 [00:11<05:48,  5.81s/it]
Training Epoch: 6/100, batch 4/125 completed (loss: 0.1363:   3%|▎         | 2/62 [00:14<05:48,  5.80s/it]
Training Epoch: 6/100, batch 4/125 completed (loss: 0.0532:   3%|▎         | 2/62 [00:14<05:48,  5.81s/it]
Training Epoch: 6/100, batch 4/125 completed (loss: 0.1463:   3%|▎         | 2/62 [00:14<05:48,  5.81s/it]
Training Epoch: 6/100, batch 4/125 completed (loss: 0.1799:   3%|▎         | 2/62 [00:14<05:48,  5.81s/it]
Training Epoch: 6/100, batch 4/125 completed (loss: 0.1463:   5%|▍         | 3/62 [00:17<05:42,  5.81s/it]
Training Epoch: 6/100, batch 4/125 completed (loss: 0.1363:   5%|▍         | 3/62 [00:17<05:42,  5.81s/it]
Training Epoch: 6/100, batch 4/125 completed (loss: 0.1799:   5%|▍         | 3/62 [00:17<05:42,  5.81s/it]
Training Epoch: 6/100, batch 4/125 completed (loss: 0.0532:   5%|▍         | 3/62 [00:17<05:42,  5.81s/it]
Training Epoch: 6/100, batch 5/125 completed (loss: 0.1595:   5%|▍         | 3/62 [00:17<05:42,  5.81s/it]
Training Epoch: 6/100, batch 5/125 completed (loss: 0.1712:   5%|▍         | 3/62 [00:17<05:42,  5.81s/it]
Training Epoch: 6/100, batch 5/125 completed (loss: 0.0905:   5%|▍         | 3/62 [00:17<05:42,  5.81s/it]
Training Epoch: 6/100, batch 5/125 completed (loss: 0.1549:   5%|▍         | 3/62 [00:17<05:42,  5.81s/it]
Training Epoch: 6/100, batch 6/125 completed (loss: 0.1478:   5%|▍         | 3/62 [00:20<05:42,  5.81s/it]
Training Epoch: 6/100, batch 6/125 completed (loss: 0.1398:   5%|▍         | 3/62 [00:20<05:42,  5.81s/it]
Training Epoch: 6/100, batch 6/125 completed (loss: 0.2232:   5%|▍         | 3/62 [00:20<05:42,  5.81s/it]
Training Epoch: 6/100, batch 6/125 completed (loss: 0.0839:   5%|▍         | 3/62 [00:20<05:42,  5.81s/it]
Training Epoch: 6/100, batch 6/125 completed (loss: 0.0839:   6%|▋         | 4/62 [00:23<05:37,  5.82s/it]
Training Epoch: 6/100, batch 6/125 completed (loss: 0.1478:   6%|▋         | 4/62 [00:23<05:37,  5.82s/it]
Training Epoch: 6/100, batch 6/125 completed (loss: 0.2232:   6%|▋         | 4/62 [00:23<05:37,  5.82s/it]
Training Epoch: 6/100, batch 6/125 completed (loss: 0.1398:   6%|▋         | 4/62 [00:23<05:37,  5.82s/it]
Training Epoch: 6/100, batch 7/125 completed (loss: 0.2313:   6%|▋         | 4/62 [00:23<05:37,  5.82s/it]
Training Epoch: 6/100, batch 7/125 completed (loss: 0.1692:   6%|▋         | 4/62 [00:23<05:37,  5.82s/it]
Training Epoch: 6/100, batch 7/125 completed (loss: 0.1613:   6%|▋         | 4/62 [00:23<05:37,  5.82s/it]
Training Epoch: 6/100, batch 7/125 completed (loss: 0.1561:   6%|▋         | 4/62 [00:23<05:37,  5.82s/it]
Training Epoch: 6/100, batch 8/125 completed (loss: 0.1017:   6%|▋         | 4/62 [00:26<05:37,  5.82s/it]
Training Epoch: 6/100, batch 8/125 completed (loss: 0.0858:   6%|▋         | 4/62 [00:26<05:37,  5.82s/it]
Training Epoch: 6/100, batch 8/125 completed (loss: 0.2516:   6%|▋         | 4/62 [00:26<05:37,  5.82s/it]
Training Epoch: 6/100, batch 8/125 completed (loss: 0.1509:   6%|▋         | 4/62 [00:26<05:37,  5.82s/it]
Training Epoch: 6/100, batch 8/125 completed (loss: 0.0858:   8%|▊         | 5/62 [00:28<05:28,  5.76s/it]
Training Epoch: 6/100, batch 8/125 completed (loss: 0.2516:   8%|▊         | 5/62 [00:28<05:28,  5.76s/it]
Training Epoch: 6/100, batch 8/125 completed (loss: 0.1509:   8%|▊         | 5/62 [00:28<05:28,  5.76s/it]
Training Epoch: 6/100, batch 8/125 completed (loss: 0.1017:   8%|▊         | 5/62 [00:28<05:28,  5.76s/it]
Training Epoch: 6/100, batch 9/125 completed (loss: 0.1329:   8%|▊         | 5/62 [00:28<05:28,  5.76s/it]
Training Epoch: 6/100, batch 9/125 completed (loss: 0.0932:   8%|▊         | 5/62 [00:28<05:28,  5.76s/it]
Training Epoch: 6/100, batch 9/125 completed (loss: 0.1856:   8%|▊         | 5/62 [00:28<05:28,  5.76s/it]
Training Epoch: 6/100, batch 9/125 completed (loss: 0.1688:   8%|▊         | 5/62 [00:28<05:28,  5.76s/it]
Training Epoch: 6/100, batch 10/125 completed (loss: 0.1378:   8%|▊         | 5/62 [00:31<05:28,  5.76s/it]
Training Epoch: 6/100, batch 10/125 completed (loss: 0.11:   8%|▊         | 5/62 [00:31<05:28,  5.76s/it] 
Training Epoch: 6/100, batch 10/125 completed (loss: 0.0794:   8%|▊         | 5/62 [00:31<05:28,  5.76s/it]
Training Epoch: 6/100, batch 10/125 completed (loss: 0.1743:   8%|▊         | 5/62 [00:31<05:28,  5.76s/it]
Training Epoch: 6/100, batch 10/125 completed (loss: 0.1743:  10%|▉         | 6/62 [00:34<05:23,  5.78s/it]
Training Epoch: 6/100, batch 10/125 completed (loss: 0.0794:  10%|▉         | 6/62 [00:34<05:23,  5.78s/it]
Training Epoch: 6/100, batch 10/125 completed (loss: 0.11:  10%|▉         | 6/62 [00:34<05:23,  5.78s/it]
Training Epoch: 6/100, batch 10/125 completed (loss: 0.1378:  10%|▉         | 6/62 [00:34<05:23,  5.78s/it]
Training Epoch: 6/100, batch 11/125 completed (loss: 0.128:  10%|▉         | 6/62 [00:34<05:23,  5.78s/it] 
Training Epoch: 6/100, batch 11/125 completed (loss: 0.1059:  10%|▉         | 6/62 [00:34<05:23,  5.78s/it]
Training Epoch: 6/100, batch 11/125 completed (loss: 0.1052:  10%|▉         | 6/62 [00:34<05:23,  5.78s/it]
Training Epoch: 6/100, batch 11/125 completed (loss: 0.1318:  10%|▉         | 6/62 [00:34<05:23,  5.78s/it]
Training Epoch: 6/100, batch 12/125 completed (loss: 0.1091:  10%|▉         | 6/62 [00:37<05:23,  5.78s/it]
Training Epoch: 6/100, batch 12/125 completed (loss: 0.1005:  10%|▉         | 6/62 [00:37<05:23,  5.78s/it]
Training Epoch: 6/100, batch 12/125 completed (loss: 0.1565:  10%|▉         | 6/62 [00:37<05:23,  5.78s/it]
Training Epoch: 6/100, batch 12/125 completed (loss: 0.1407:  10%|▉         | 6/62 [00:37<05:23,  5.78s/it]
Training Epoch: 6/100, batch 12/125 completed (loss: 0.1407:  11%|█▏        | 7/62 [00:40<05:19,  5.81s/it]
Training Epoch: 6/100, batch 12/125 completed (loss: 0.1005:  11%|█▏        | 7/62 [00:40<05:19,  5.81s/it]
Training Epoch: 6/100, batch 12/125 completed (loss: 0.1565:  11%|█▏        | 7/62 [00:40<05:19,  5.81s/it]
Training Epoch: 6/100, batch 12/125 completed (loss: 0.1091:  11%|█▏        | 7/62 [00:40<05:19,  5.81s/it]
Training Epoch: 6/100, batch 13/125 completed (loss: 0.1038:  11%|█▏        | 7/62 [00:40<05:19,  5.81s/it]
Training Epoch: 6/100, batch 13/125 completed (loss: 0.0686:  11%|█▏        | 7/62 [00:40<05:19,  5.81s/it]
Training Epoch: 6/100, batch 13/125 completed (loss: 0.1114:  11%|█▏        | 7/62 [00:40<05:19,  5.81s/it]
Training Epoch: 6/100, batch 13/125 completed (loss: 0.1497:  11%|█▏        | 7/62 [00:40<05:19,  5.81s/it]
Training Epoch: 6/100, batch 14/125 completed (loss: 0.0539:  11%|█▏        | 7/62 [00:43<05:19,  5.81s/it]
Training Epoch: 6/100, batch 14/125 completed (loss: 0.1866:  11%|█▏        | 7/62 [00:43<05:19,  5.81s/it]
Training Epoch: 6/100, batch 14/125 completed (loss: 0.1905:  11%|█▏        | 7/62 [00:43<05:19,  5.81s/it]
Training Epoch: 6/100, batch 14/125 completed (loss: 0.0993:  11%|█▏        | 7/62 [00:43<05:19,  5.81s/it]
Training Epoch: 6/100, batch 14/125 completed (loss: 0.0993:  13%|█▎        | 8/62 [00:46<05:13,  5.81s/it]
Training Epoch: 6/100, batch 14/125 completed (loss: 0.0539:  13%|█▎        | 8/62 [00:46<05:13,  5.81s/it]
Training Epoch: 6/100, batch 14/125 completed (loss: 0.1866:  13%|█▎        | 8/62 [00:46<05:13,  5.81s/it]
Training Epoch: 6/100, batch 14/125 completed (loss: 0.1905:  13%|█▎        | 8/62 [00:46<05:14,  5.82s/it]
Training Epoch: 6/100, batch 15/125 completed (loss: 0.3173:  13%|█▎        | 8/62 [00:46<05:13,  5.81s/it]
Training Epoch: 6/100, batch 15/125 completed (loss: 0.1836:  13%|█▎        | 8/62 [00:46<05:14,  5.82s/it]
Training Epoch: 6/100, batch 15/125 completed (loss: 0.0989:  13%|█▎        | 8/62 [00:46<05:13,  5.81s/it]
Training Epoch: 6/100, batch 15/125 completed (loss: 0.1685:  13%|█▎        | 8/62 [00:46<05:13,  5.81s/it]
Training Epoch: 6/100, batch 16/125 completed (loss: 0.0599:  13%|█▎        | 8/62 [00:49<05:13,  5.81s/it]
Training Epoch: 6/100, batch 16/125 completed (loss: 0.1209:  13%|█▎        | 8/62 [00:49<05:13,  5.81s/it]
Training Epoch: 6/100, batch 16/125 completed (loss: 0.2755:  13%|█▎        | 8/62 [00:49<05:14,  5.82s/it]
Training Epoch: 6/100, batch 16/125 completed (loss: 0.1727:  13%|█▎        | 8/62 [00:49<05:13,  5.81s/it]
Training Epoch: 6/100, batch 16/125 completed (loss: 0.1727:  15%|█▍        | 9/62 [00:52<05:13,  5.91s/it]
Training Epoch: 6/100, batch 16/125 completed (loss: 0.0599:  15%|█▍        | 9/62 [00:52<05:13,  5.91s/it]
Training Epoch: 6/100, batch 16/125 completed (loss: 0.1209:  15%|█▍        | 9/62 [00:52<05:13,  5.91s/it]
Training Epoch: 6/100, batch 16/125 completed (loss: 0.2755:  15%|█▍        | 9/62 [00:52<05:13,  5.91s/it]
Training Epoch: 6/100, batch 17/125 completed (loss: 0.0976:  15%|█▍        | 9/62 [00:52<05:13,  5.91s/it]
Training Epoch: 6/100, batch 17/125 completed (loss: 0.0794:  15%|█▍        | 9/62 [00:52<05:13,  5.91s/it]
Training Epoch: 6/100, batch 17/125 completed (loss: 0.1094:  15%|█▍        | 9/62 [00:52<05:13,  5.91s/it]
Training Epoch: 6/100, batch 17/125 completed (loss: 0.112:  15%|█▍        | 9/62 [00:52<05:13,  5.91s/it] 
Training Epoch: 6/100, batch 18/125 completed (loss: 0.1867:  15%|█▍        | 9/62 [00:55<05:13,  5.91s/it]
Training Epoch: 6/100, batch 18/125 completed (loss: 0.0922:  15%|█▍        | 9/62 [00:55<05:13,  5.91s/it]
Training Epoch: 6/100, batch 18/125 completed (loss: 0.1652:  15%|█▍        | 9/62 [00:55<05:13,  5.91s/it]
Training Epoch: 6/100, batch 18/125 completed (loss: 0.1816:  15%|█▍        | 9/62 [00:55<05:13,  5.91s/it]
Training Epoch: 6/100, batch 18/125 completed (loss: 0.1816:  16%|█▌        | 10/62 [00:58<05:03,  5.84s/it]
Training Epoch: 6/100, batch 18/125 completed (loss: 0.1867:  16%|█▌        | 10/62 [00:58<05:03,  5.83s/it]
Training Epoch: 6/100, batch 18/125 completed (loss: 0.0922:  16%|█▌        | 10/62 [00:58<05:03,  5.83s/it]
Training Epoch: 6/100, batch 18/125 completed (loss: 0.1652:  16%|█▌        | 10/62 [00:58<05:03,  5.84s/it]
Training Epoch: 6/100, batch 19/125 completed (loss: 0.0962:  16%|█▌        | 10/62 [00:58<05:03,  5.83s/it]
Training Epoch: 6/100, batch 19/125 completed (loss: 0.2054:  16%|█▌        | 10/62 [00:58<05:03,  5.84s/it]
Training Epoch: 6/100, batch 19/125 completed (loss: 0.0986:  16%|█▌        | 10/62 [00:58<05:03,  5.83s/it]
Training Epoch: 6/100, batch 19/125 completed (loss: 0.1822:  16%|█▌        | 10/62 [00:58<05:03,  5.84s/it]
Training Epoch: 6/100, batch 20/125 completed (loss: 0.0981:  16%|█▌        | 10/62 [01:01<05:03,  5.84s/it]
Training Epoch: 6/100, batch 20/125 completed (loss: 0.1929:  16%|█▌        | 10/62 [01:01<05:03,  5.83s/it]
Training Epoch: 6/100, batch 20/125 completed (loss: 0.2358:  16%|█▌        | 10/62 [01:01<05:03,  5.83s/it]
Training Epoch: 6/100, batch 20/125 completed (loss: 0.0488:  16%|█▌        | 10/62 [01:01<05:03,  5.84s/it]
Training Epoch: 6/100, batch 20/125 completed (loss: 0.0981:  18%|█▊        | 11/62 [01:04<04:57,  5.83s/it]
Training Epoch: 6/100, batch 20/125 completed (loss: 0.0488:  18%|█▊        | 11/62 [01:04<04:57,  5.83s/it]
Training Epoch: 6/100, batch 20/125 completed (loss: 0.2358:  18%|█▊        | 11/62 [01:04<04:57,  5.83s/it]
Training Epoch: 6/100, batch 20/125 completed (loss: 0.1929:  18%|█▊        | 11/62 [01:04<04:57,  5.83s/it]
Training Epoch: 6/100, batch 21/125 completed (loss: 0.2726:  18%|█▊        | 11/62 [01:04<04:57,  5.83s/it]
Training Epoch: 6/100, batch 21/125 completed (loss: 0.2061:  18%|█▊        | 11/62 [01:04<04:57,  5.83s/it]
Training Epoch: 6/100, batch 21/125 completed (loss: 0.0523:  18%|█▊        | 11/62 [01:04<04:57,  5.83s/it]
Training Epoch: 6/100, batch 21/125 completed (loss: 0.1305:  18%|█▊        | 11/62 [01:04<04:57,  5.83s/it]
Training Epoch: 6/100, batch 22/125 completed (loss: 0.1388:  18%|█▊        | 11/62 [01:07<04:57,  5.83s/it]
Training Epoch: 6/100, batch 22/125 completed (loss: 0.1489:  18%|█▊        | 11/62 [01:07<04:57,  5.83s/it]
Training Epoch: 6/100, batch 22/125 completed (loss: 0.0595:  18%|█▊        | 11/62 [01:07<04:57,  5.83s/it]
Training Epoch: 6/100, batch 22/125 completed (loss: 0.0897:  18%|█▊        | 11/62 [01:07<04:57,  5.83s/it]
Training Epoch: 6/100, batch 22/125 completed (loss: 0.0897:  19%|█▉        | 12/62 [01:10<04:56,  5.93s/it]
Training Epoch: 6/100, batch 22/125 completed (loss: 0.1489:  19%|█▉        | 12/62 [01:10<04:56,  5.93s/it]
Training Epoch: 6/100, batch 22/125 completed (loss: 0.1388:  19%|█▉        | 12/62 [01:10<04:56,  5.93s/it]
Training Epoch: 6/100, batch 22/125 completed (loss: 0.0595:  19%|█▉        | 12/62 [01:10<04:56,  5.93s/it]
Training Epoch: 6/100, batch 23/125 completed (loss: 0.0841:  19%|█▉        | 12/62 [01:10<04:56,  5.93s/it]
Training Epoch: 6/100, batch 23/125 completed (loss: 0.0408:  19%|█▉        | 12/62 [01:10<04:56,  5.93s/it]
Training Epoch: 6/100, batch 23/125 completed (loss: 0.0935:  19%|█▉        | 12/62 [01:10<04:56,  5.93s/it]
Training Epoch: 6/100, batch 23/125 completed (loss: 0.1639:  19%|█▉        | 12/62 [01:10<04:56,  5.93s/it]
Training Epoch: 6/100, batch 24/125 completed (loss: 0.1009:  19%|█▉        | 12/62 [01:13<04:56,  5.93s/it]
Training Epoch: 6/100, batch 24/125 completed (loss: 0.1463:  19%|█▉        | 12/62 [01:13<04:56,  5.93s/it]
Training Epoch: 6/100, batch 24/125 completed (loss: 0.33:  19%|█▉        | 12/62 [01:13<04:56,  5.93s/it]  
Training Epoch: 6/100, batch 24/125 completed (loss: 0.1997:  19%|█▉        | 12/62 [01:13<04:56,  5.93s/it]
Training Epoch: 6/100, batch 24/125 completed (loss: 0.1009:  21%|██        | 13/62 [01:16<04:49,  5.91s/it]
Training Epoch: 6/100, batch 24/125 completed (loss: 0.33:  21%|██        | 13/62 [01:16<04:49,  5.91s/it]
Training Epoch: 6/100, batch 24/125 completed (loss: 0.1997:  21%|██        | 13/62 [01:16<04:49,  5.91s/it]
Training Epoch: 6/100, batch 24/125 completed (loss: 0.1463:  21%|██        | 13/62 [01:16<04:49,  5.91s/it]
Training Epoch: 6/100, batch 25/125 completed (loss: 0.1575:  21%|██        | 13/62 [01:16<04:49,  5.91s/it]
Training Epoch: 6/100, batch 25/125 completed (loss: 0.0738:  21%|██        | 13/62 [01:16<04:49,  5.91s/it]
Training Epoch: 6/100, batch 25/125 completed (loss: 0.0766:  21%|██        | 13/62 [01:16<04:49,  5.91s/it]
Training Epoch: 6/100, batch 25/125 completed (loss: 0.2293:  21%|██        | 13/62 [01:16<04:49,  5.91s/it]
Training Epoch: 6/100, batch 26/125 completed (loss: 0.1128:  21%|██        | 13/62 [01:19<04:49,  5.91s/it]
Training Epoch: 6/100, batch 26/125 completed (loss: 0.1669:  21%|██        | 13/62 [01:19<04:49,  5.91s/it]
Training Epoch: 6/100, batch 26/125 completed (loss: 0.0766:  21%|██        | 13/62 [01:19<04:49,  5.91s/it]
Training Epoch: 6/100, batch 26/125 completed (loss: 0.155:  21%|██        | 13/62 [01:19<04:49,  5.91s/it] 
Training Epoch: 6/100, batch 26/125 completed (loss: 0.155:  23%|██▎       | 14/62 [01:22<04:45,  5.95s/it]
Training Epoch: 6/100, batch 26/125 completed (loss: 0.1128:  23%|██▎       | 14/62 [01:22<04:45,  5.95s/it]
Training Epoch: 6/100, batch 26/125 completed (loss: 0.0766:  23%|██▎       | 14/62 [01:22<04:45,  5.95s/it]
Training Epoch: 6/100, batch 26/125 completed (loss: 0.1669:  23%|██▎       | 14/62 [01:22<04:45,  5.95s/it]
Training Epoch: 6/100, batch 27/125 completed (loss: 0.2372:  23%|██▎       | 14/62 [01:22<04:45,  5.95s/it]
Training Epoch: 6/100, batch 27/125 completed (loss: 0.318:  23%|██▎       | 14/62 [01:22<04:45,  5.95s/it] 
Training Epoch: 6/100, batch 27/125 completed (loss: 0.1624:  23%|██▎       | 14/62 [01:22<04:45,  5.95s/it]
Training Epoch: 6/100, batch 27/125 completed (loss: 0.1183:  23%|██▎       | 14/62 [01:22<04:45,  5.95s/it]
Training Epoch: 6/100, batch 28/125 completed (loss: 0.1708:  23%|██▎       | 14/62 [01:25<04:45,  5.95s/it]
Training Epoch: 6/100, batch 28/125 completed (loss: 0.1162:  23%|██▎       | 14/62 [01:25<04:45,  5.95s/it]
Training Epoch: 6/100, batch 28/125 completed (loss: 0.1394:  23%|██▎       | 14/62 [01:25<04:45,  5.95s/it]
Training Epoch: 6/100, batch 28/125 completed (loss: 0.0987:  23%|██▎       | 14/62 [01:25<04:45,  5.95s/it]
Training Epoch: 6/100, batch 28/125 completed (loss: 0.1162:  24%|██▍       | 15/62 [01:28<04:38,  5.93s/it]
Training Epoch: 6/100, batch 28/125 completed (loss: 0.1708:  24%|██▍       | 15/62 [01:28<04:38,  5.93s/it]
Training Epoch: 6/100, batch 28/125 completed (loss: 0.0987:  24%|██▍       | 15/62 [01:28<04:38,  5.93s/it]
Training Epoch: 6/100, batch 28/125 completed (loss: 0.1394:  24%|██▍       | 15/62 [01:28<04:38,  5.93s/it]
Training Epoch: 6/100, batch 29/125 completed (loss: 0.0904:  24%|██▍       | 15/62 [01:28<04:38,  5.93s/it]
Training Epoch: 6/100, batch 29/125 completed (loss: 0.0691:  24%|██▍       | 15/62 [01:28<04:38,  5.93s/it]
Training Epoch: 6/100, batch 29/125 completed (loss: 0.1533:  24%|██▍       | 15/62 [01:28<04:38,  5.93s/it]
Training Epoch: 6/100, batch 29/125 completed (loss: 0.0737:  24%|██▍       | 15/62 [01:28<04:38,  5.93s/it]
Training Epoch: 6/100, batch 30/125 completed (loss: 0.0646:  24%|██▍       | 15/62 [01:31<04:38,  5.93s/it]
Training Epoch: 6/100, batch 30/125 completed (loss: 0.112:  24%|██▍       | 15/62 [01:31<04:38,  5.93s/it] 
Training Epoch: 6/100, batch 30/125 completed (loss: 0.1006:  24%|██▍       | 15/62 [01:31<04:38,  5.93s/it]
Training Epoch: 6/100, batch 30/125 completed (loss: 0.0723:  24%|██▍       | 15/62 [01:31<04:38,  5.93s/it]
Training Epoch: 6/100, batch 30/125 completed (loss: 0.0723:  26%|██▌       | 16/62 [01:33<04:32,  5.92s/it]
Training Epoch: 6/100, batch 30/125 completed (loss: 0.112:  26%|██▌       | 16/62 [01:33<04:32,  5.92s/it]
Training Epoch: 6/100, batch 30/125 completed (loss: 0.1006:  26%|██▌       | 16/62 [01:33<04:32,  5.92s/it]
Training Epoch: 6/100, batch 30/125 completed (loss: 0.0646:  26%|██▌       | 16/62 [01:33<04:32,  5.92s/it]
Training Epoch: 6/100, batch 31/125 completed (loss: 0.0838:  26%|██▌       | 16/62 [01:33<04:32,  5.92s/it]
Training Epoch: 6/100, batch 31/125 completed (loss: 0.0507:  26%|██▌       | 16/62 [01:33<04:32,  5.92s/it]
Training Epoch: 6/100, batch 31/125 completed (loss: 0.1284:  26%|██▌       | 16/62 [01:33<04:32,  5.92s/it]
Training Epoch: 6/100, batch 31/125 completed (loss: 0.0678:  26%|██▌       | 16/62 [01:33<04:32,  5.92s/it]
Training Epoch: 6/100, batch 32/125 completed (loss: 0.2304:  26%|██▌       | 16/62 [01:36<04:32,  5.92s/it]
Training Epoch: 6/100, batch 32/125 completed (loss: 0.388:  26%|██▌       | 16/62 [01:36<04:32,  5.92s/it] 
Training Epoch: 6/100, batch 32/125 completed (loss: 0.0592:  26%|██▌       | 16/62 [01:36<04:32,  5.92s/it]
Training Epoch: 6/100, batch 32/125 completed (loss: 0.2096:  26%|██▌       | 16/62 [01:36<04:32,  5.92s/it]
Training Epoch: 6/100, batch 32/125 completed (loss: 0.2096:  27%|██▋       | 17/62 [01:40<04:28,  5.97s/it]
Training Epoch: 6/100, batch 32/125 completed (loss: 0.0592:  27%|██▋       | 17/62 [01:40<04:29,  5.98s/it]
Training Epoch: 6/100, batch 32/125 completed (loss: 0.388:  27%|██▋       | 17/62 [01:40<04:29,  5.98s/it]
Training Epoch: 6/100, batch 32/125 completed (loss: 0.2304:  27%|██▋       | 17/62 [01:40<04:29,  5.98s/it]
Training Epoch: 6/100, batch 33/125 completed (loss: 0.1859:  27%|██▋       | 17/62 [01:40<04:29,  5.98s/it]
Training Epoch: 6/100, batch 33/125 completed (loss: 0.0647:  27%|██▋       | 17/62 [01:40<04:29,  5.98s/it]
Training Epoch: 6/100, batch 33/125 completed (loss: 0.0937:  27%|██▋       | 17/62 [01:40<04:29,  5.98s/it]
Training Epoch: 6/100, batch 33/125 completed (loss: 0.1486:  27%|██▋       | 17/62 [01:40<04:28,  5.97s/it]
Training Epoch: 6/100, batch 34/125 completed (loss: 0.0514:  27%|██▋       | 17/62 [01:42<04:29,  5.98s/it]
Training Epoch: 6/100, batch 34/125 completed (loss: 0.1984:  27%|██▋       | 17/62 [01:42<04:29,  5.98s/it]
Training Epoch: 6/100, batch 34/125 completed (loss: 0.2195:  27%|██▋       | 17/62 [01:42<04:29,  5.98s/it]
Training Epoch: 6/100, batch 34/125 completed (loss: 0.128:  27%|██▋       | 17/62 [01:42<04:28,  5.97s/it] 
Training Epoch: 6/100, batch 34/125 completed (loss: 0.1984:  29%|██▉       | 18/62 [01:45<04:20,  5.92s/it]
Training Epoch: 6/100, batch 34/125 completed (loss: 0.2195:  29%|██▉       | 18/62 [01:45<04:20,  5.93s/it]
Training Epoch: 6/100, batch 34/125 completed (loss: 0.128:  29%|██▉       | 18/62 [01:45<04:20,  5.93s/it]
Training Epoch: 6/100, batch 34/125 completed (loss: 0.0514:  29%|██▉       | 18/62 [01:45<04:20,  5.93s/it]
Training Epoch: 6/100, batch 35/125 completed (loss: 0.041:  29%|██▉       | 18/62 [01:45<04:20,  5.93s/it] 
Training Epoch: 6/100, batch 35/125 completed (loss: 0.0465:  29%|██▉       | 18/62 [01:45<04:20,  5.93s/it]
Training Epoch: 6/100, batch 35/125 completed (loss: 0.234:  29%|██▉       | 18/62 [01:45<04:20,  5.92s/it] 
Training Epoch: 6/100, batch 35/125 completed (loss: 0.2242:  29%|██▉       | 18/62 [01:45<04:20,  5.93s/it]
Training Epoch: 6/100, batch 36/125 completed (loss: 0.1965:  29%|██▉       | 18/62 [01:48<04:20,  5.93s/it]
Training Epoch: 6/100, batch 36/125 completed (loss: 0.0383:  29%|██▉       | 18/62 [01:48<04:20,  5.93s/it]
Training Epoch: 6/100, batch 36/125 completed (loss: 0.1003:  29%|██▉       | 18/62 [01:48<04:20,  5.92s/it]
Training Epoch: 6/100, batch 36/125 completed (loss: 0.1593:  29%|██▉       | 18/62 [01:48<04:20,  5.93s/it]
Training Epoch: 6/100, batch 36/125 completed (loss: 0.0383:  31%|███       | 19/62 [01:51<04:16,  5.97s/it]
Training Epoch: 6/100, batch 36/125 completed (loss: 0.1593:  31%|███       | 19/62 [01:51<04:16,  5.97s/it]
Training Epoch: 6/100, batch 36/125 completed (loss: 0.1003:  31%|███       | 19/62 [01:51<04:16,  5.97s/it]
Training Epoch: 6/100, batch 36/125 completed (loss: 0.1965:  31%|███       | 19/62 [01:51<04:16,  5.97s/it]
Training Epoch: 6/100, batch 37/125 completed (loss: 0.1549:  31%|███       | 19/62 [01:51<04:16,  5.97s/it]
Training Epoch: 6/100, batch 37/125 completed (loss: 0.1261:  31%|███       | 19/62 [01:51<04:16,  5.97s/it]
Training Epoch: 6/100, batch 37/125 completed (loss: 0.0942:  31%|███       | 19/62 [01:51<04:16,  5.97s/it]
Training Epoch: 6/100, batch 37/125 completed (loss: 0.0519:  31%|███       | 19/62 [01:51<04:16,  5.97s/it]
Training Epoch: 6/100, batch 38/125 completed (loss: 0.0384:  31%|███       | 19/62 [01:55<04:16,  5.97s/it]
Training Epoch: 6/100, batch 38/125 completed (loss: 0.2862:  31%|███       | 19/62 [01:55<04:16,  5.97s/it]
Training Epoch: 6/100, batch 38/125 completed (loss: 0.0895:  31%|███       | 19/62 [01:55<04:16,  5.97s/it]
Training Epoch: 6/100, batch 38/125 completed (loss: 0.1761:  31%|███       | 19/62 [01:55<04:16,  5.97s/it]
Training Epoch: 6/100, batch 38/125 completed (loss: 0.1761:  32%|███▏      | 20/62 [01:57<04:10,  5.97s/it]
Training Epoch: 6/100, batch 38/125 completed (loss: 0.2862:  32%|███▏      | 20/62 [01:57<04:10,  5.97s/it]
Training Epoch: 6/100, batch 38/125 completed (loss: 0.0384:  32%|███▏      | 20/62 [01:57<04:11,  5.98s/it]
Training Epoch: 6/100, batch 38/125 completed (loss: 0.0895:  32%|███▏      | 20/62 [01:57<04:11,  5.98s/it]
Training Epoch: 6/100, batch 39/125 completed (loss: 0.0995:  32%|███▏      | 20/62 [01:57<04:11,  5.98s/it]
Training Epoch: 6/100, batch 39/125 completed (loss: 0.1458:  32%|███▏      | 20/62 [01:57<04:11,  5.98s/it]
Training Epoch: 6/100, batch 39/125 completed (loss: 0.1086:  32%|███▏      | 20/62 [01:57<04:10,  5.97s/it]
Training Epoch: 6/100, batch 39/125 completed (loss: 0.0943:  32%|███▏      | 20/62 [01:57<04:10,  5.97s/it]
Training Epoch: 6/100, batch 40/125 completed (loss: 0.2027:  32%|███▏      | 20/62 [02:00<04:10,  5.97s/it]
Training Epoch: 6/100, batch 40/125 completed (loss: 0.219:  32%|███▏      | 20/62 [02:00<04:11,  5.98s/it] 
Training Epoch: 6/100, batch 40/125 completed (loss: 0.129:  32%|███▏      | 20/62 [02:00<04:11,  5.98s/it] 
Training Epoch: 6/100, batch 40/125 completed (loss: 0.161:  32%|███▏      | 20/62 [02:00<04:10,  5.97s/it] 
Training Epoch: 6/100, batch 40/125 completed (loss: 0.161:  34%|███▍      | 21/62 [02:03<04:02,  5.92s/it]
Training Epoch: 6/100, batch 40/125 completed (loss: 0.2027:  34%|███▍      | 21/62 [02:03<04:02,  5.92s/it]
Training Epoch: 6/100, batch 40/125 completed (loss: 0.129:  34%|███▍      | 21/62 [02:03<04:02,  5.92s/it]
Training Epoch: 6/100, batch 40/125 completed (loss: 0.219:  34%|███▍      | 21/62 [02:03<04:03,  5.93s/it]
Training Epoch: 6/100, batch 41/125 completed (loss: 0.3654:  34%|███▍      | 21/62 [02:03<04:02,  5.92s/it]
Training Epoch: 6/100, batch 41/125 completed (loss: 0.1536:  34%|███▍      | 21/62 [02:03<04:03,  5.93s/it]
Training Epoch: 6/100, batch 41/125 completed (loss: 0.049:  34%|███▍      | 21/62 [02:03<04:02,  5.92s/it] 
Training Epoch: 6/100, batch 41/125 completed (loss: 0.1794:  34%|███▍      | 21/62 [02:03<04:02,  5.92s/it]
Training Epoch: 6/100, batch 42/125 completed (loss: 0.2567:  34%|███▍      | 21/62 [02:06<04:03,  5.93s/it]
Training Epoch: 6/100, batch 42/125 completed (loss: 0.0766:  34%|███▍      | 21/62 [02:06<04:02,  5.92s/it]
Training Epoch: 6/100, batch 42/125 completed (loss: 0.2232:  34%|███▍      | 21/62 [02:06<04:02,  5.92s/it]
Training Epoch: 6/100, batch 42/125 completed (loss: 0.1319:  34%|███▍      | 21/62 [02:06<04:02,  5.92s/it]
Training Epoch: 6/100, batch 42/125 completed (loss: 0.2232:  35%|███▌      | 22/62 [02:10<04:02,  6.07s/it]
Training Epoch: 6/100, batch 42/125 completed (loss: 0.1319:  35%|███▌      | 22/62 [02:10<04:02,  6.07s/it]
Training Epoch: 6/100, batch 42/125 completed (loss: 0.0766:  35%|███▌      | 22/62 [02:10<04:02,  6.07s/it]
Training Epoch: 6/100, batch 42/125 completed (loss: 0.2567:  35%|███▌      | 22/62 [02:10<04:02,  6.07s/it]
Training Epoch: 6/100, batch 43/125 completed (loss: 0.1165:  35%|███▌      | 22/62 [02:10<04:02,  6.07s/it]
Training Epoch: 6/100, batch 43/125 completed (loss: 0.0635:  35%|███▌      | 22/62 [02:10<04:02,  6.07s/it]
Training Epoch: 6/100, batch 43/125 completed (loss: 0.1183:  35%|███▌      | 22/62 [02:10<04:02,  6.07s/it]
Training Epoch: 6/100, batch 43/125 completed (loss: 0.0544:  35%|███▌      | 22/62 [02:10<04:02,  6.07s/it]
Training Epoch: 6/100, batch 44/125 completed (loss: 0.1863:  35%|███▌      | 22/62 [02:13<04:02,  6.07s/it]
Training Epoch: 6/100, batch 44/125 completed (loss: 0.0912:  35%|███▌      | 22/62 [02:13<04:02,  6.07s/it]
Training Epoch: 6/100, batch 44/125 completed (loss: 0.0921:  35%|███▌      | 22/62 [02:13<04:02,  6.07s/it]
Training Epoch: 6/100, batch 44/125 completed (loss: 0.0601:  35%|███▌      | 22/62 [02:13<04:02,  6.07s/it]
Training Epoch: 6/100, batch 44/125 completed (loss: 0.0601:  37%|███▋      | 23/62 [02:15<03:53,  5.99s/it]
Training Epoch: 6/100, batch 44/125 completed (loss: 0.0921:  37%|███▋      | 23/62 [02:15<03:53,  5.99s/it]
Training Epoch: 6/100, batch 44/125 completed (loss: 0.1863:  37%|███▋      | 23/62 [02:15<03:53,  5.99s/it]
Training Epoch: 6/100, batch 44/125 completed (loss: 0.0912:  37%|███▋      | 23/62 [02:15<03:53,  6.00s/it]
Training Epoch: 6/100, batch 45/125 completed (loss: 0.0758:  37%|███▋      | 23/62 [02:15<03:53,  5.99s/it]
Training Epoch: 6/100, batch 45/125 completed (loss: 0.1666:  37%|███▋      | 23/62 [02:15<03:53,  5.99s/it]
Training Epoch: 6/100, batch 45/125 completed (loss: 0.0409:  37%|███▋      | 23/62 [02:15<03:53,  6.00s/it]
Training Epoch: 6/100, batch 45/125 completed (loss: 0.102:  37%|███▋      | 23/62 [02:15<03:53,  5.99s/it] 
Training Epoch: 6/100, batch 46/125 completed (loss: 0.3009:  37%|███▋      | 23/62 [02:18<03:53,  5.99s/it]
Training Epoch: 6/100, batch 46/125 completed (loss: 0.0594:  37%|███▋      | 23/62 [02:18<03:53,  5.99s/it]
Training Epoch: 6/100, batch 46/125 completed (loss: 0.0811:  37%|███▋      | 23/62 [02:18<03:53,  6.00s/it]
Training Epoch: 6/100, batch 46/125 completed (loss: 0.1797:  37%|███▋      | 23/62 [02:18<03:53,  5.99s/it]
Training Epoch: 6/100, batch 46/125 completed (loss: 0.1797:  39%|███▊      | 24/62 [02:21<03:47,  5.98s/it]
Training Epoch: 6/100, batch 46/125 completed (loss: 0.0594:  39%|███▊      | 24/62 [02:21<03:47,  5.98s/it]
Training Epoch: 6/100, batch 46/125 completed (loss: 0.3009:  39%|███▊      | 24/62 [02:21<03:47,  5.98s/it]
Training Epoch: 6/100, batch 46/125 completed (loss: 0.0811:  39%|███▊      | 24/62 [02:21<03:47,  5.98s/it]
Training Epoch: 6/100, batch 47/125 completed (loss: 0.1202:  39%|███▊      | 24/62 [02:21<03:47,  5.98s/it]
Training Epoch: 6/100, batch 47/125 completed (loss: 0.1413:  39%|███▊      | 24/62 [02:21<03:47,  5.98s/it]
Training Epoch: 6/100, batch 47/125 completed (loss: 0.0843:  39%|███▊      | 24/62 [02:21<03:47,  5.98s/it]
Training Epoch: 6/100, batch 47/125 completed (loss: 0.1701:  39%|███▊      | 24/62 [02:21<03:47,  5.98s/it]
Training Epoch: 6/100, batch 48/125 completed (loss: 0.0869:  39%|███▊      | 24/62 [02:24<03:47,  5.98s/it]
Training Epoch: 6/100, batch 48/125 completed (loss: 0.2042:  39%|███▊      | 24/62 [02:24<03:47,  5.98s/it]
Training Epoch: 6/100, batch 48/125 completed (loss: 0.1699:  39%|███▊      | 24/62 [02:24<03:47,  5.98s/it]
Training Epoch: 6/100, batch 48/125 completed (loss: 0.1476:  39%|███▊      | 24/62 [02:24<03:47,  5.98s/it]
Training Epoch: 6/100, batch 48/125 completed (loss: 0.1699:  40%|████      | 25/62 [02:28<03:43,  6.05s/it]
Training Epoch: 6/100, batch 48/125 completed (loss: 0.1476:  40%|████      | 25/62 [02:28<03:44,  6.06s/it]
Training Epoch: 6/100, batch 48/125 completed (loss: 0.2042:  40%|████      | 25/62 [02:28<03:44,  6.05s/it]
Training Epoch: 6/100, batch 48/125 completed (loss: 0.0869:  40%|████      | 25/62 [02:28<03:44,  6.06s/it]
Training Epoch: 6/100, batch 49/125 completed (loss: 0.0956:  40%|████      | 25/62 [02:28<03:43,  6.05s/it]
Training Epoch: 6/100, batch 49/125 completed (loss: 0.0913:  40%|████      | 25/62 [02:28<03:44,  6.06s/it]
Training Epoch: 6/100, batch 49/125 completed (loss: 0.4189:  40%|████      | 25/62 [02:28<03:44,  6.05s/it]
Training Epoch: 6/100, batch 49/125 completed (loss: 0.1255:  40%|████      | 25/62 [02:28<03:44,  6.06s/it]
Training Epoch: 6/100, batch 50/125 completed (loss: 0.0831:  40%|████      | 25/62 [02:31<03:44,  6.06s/it]
Training Epoch: 6/100, batch 50/125 completed (loss: 0.132:  40%|████      | 25/62 [02:31<03:44,  6.05s/it] 
Training Epoch: 6/100, batch 50/125 completed (loss: 0.0935:  40%|████      | 25/62 [02:31<03:43,  6.05s/it]
Training Epoch: 6/100, batch 50/125 completed (loss: 0.1696:  40%|████      | 25/62 [02:31<03:44,  6.06s/it]
Training Epoch: 6/100, batch 50/125 completed (loss: 0.0935:  42%|████▏     | 26/62 [02:33<03:35,  5.99s/it]
Training Epoch: 6/100, batch 50/125 completed (loss: 0.1696:  42%|████▏     | 26/62 [02:33<03:35,  5.99s/it]
Training Epoch: 6/100, batch 50/125 completed (loss: 0.0831:  42%|████▏     | 26/62 [02:33<03:35,  5.99s/it]
Training Epoch: 6/100, batch 50/125 completed (loss: 0.132:  42%|████▏     | 26/62 [02:33<03:35,  5.99s/it]
Training Epoch: 6/100, batch 51/125 completed (loss: 0.1743:  42%|████▏     | 26/62 [02:33<03:35,  5.99s/it]
Training Epoch: 6/100, batch 51/125 completed (loss: 0.1899:  42%|████▏     | 26/62 [02:33<03:35,  5.99s/it]
Training Epoch: 6/100, batch 51/125 completed (loss: 0.075:  42%|████▏     | 26/62 [02:33<03:35,  5.99s/it] 
Training Epoch: 6/100, batch 51/125 completed (loss: 0.2741:  42%|████▏     | 26/62 [02:33<03:35,  5.99s/it]
Training Epoch: 6/100, batch 52/125 completed (loss: 0.0681:  42%|████▏     | 26/62 [02:37<03:35,  5.99s/it]
Training Epoch: 6/100, batch 52/125 completed (loss: 0.1296:  42%|████▏     | 26/62 [02:37<03:35,  5.99s/it]
Training Epoch: 6/100, batch 52/125 completed (loss: 0.0414:  42%|████▏     | 26/62 [02:37<03:35,  5.99s/it]
Training Epoch: 6/100, batch 52/125 completed (loss: 0.061:  42%|████▏     | 26/62 [02:37<03:35,  5.99s/it] 
Training Epoch: 6/100, batch 52/125 completed (loss: 0.061:  44%|████▎     | 27/62 [02:40<03:33,  6.10s/it]
Training Epoch: 6/100, batch 52/125 completed (loss: 0.0681:  44%|████▎     | 27/62 [02:40<03:33,  6.10s/it]
Training Epoch: 6/100, batch 52/125 completed (loss: 0.1296:  44%|████▎     | 27/62 [02:40<03:33,  6.10s/it]
Training Epoch: 6/100, batch 52/125 completed (loss: 0.0414:  44%|████▎     | 27/62 [02:40<03:33,  6.10s/it]
Training Epoch: 6/100, batch 53/125 completed (loss: 0.1059:  44%|████▎     | 27/62 [02:40<03:33,  6.10s/it]
Training Epoch: 6/100, batch 53/125 completed (loss: 0.1531:  44%|████▎     | 27/62 [02:40<03:33,  6.10s/it]
Training Epoch: 6/100, batch 53/125 completed (loss: 0.1069:  44%|████▎     | 27/62 [02:40<03:33,  6.10s/it]
Training Epoch: 6/100, batch 53/125 completed (loss: 0.0646:  44%|████▎     | 27/62 [02:40<03:33,  6.10s/it]
Training Epoch: 6/100, batch 54/125 completed (loss: 0.15:  44%|████▎     | 27/62 [02:43<03:33,  6.10s/it]  
Training Epoch: 6/100, batch 54/125 completed (loss: 0.1519:  44%|████▎     | 27/62 [02:43<03:33,  6.10s/it]
Training Epoch: 6/100, batch 54/125 completed (loss: 0.1099:  44%|████▎     | 27/62 [02:43<03:33,  6.10s/it]
Training Epoch: 6/100, batch 54/125 completed (loss: 0.1288:  44%|████▎     | 27/62 [02:43<03:33,  6.10s/it]
Training Epoch: 6/100, batch 54/125 completed (loss: 0.1099:  45%|████▌     | 28/62 [02:46<03:24,  6.01s/it]
Training Epoch: 6/100, batch 54/125 completed (loss: 0.1288:  45%|████▌     | 28/62 [02:46<03:24,  6.01s/it]
Training Epoch: 6/100, batch 54/125 completed (loss: 0.1519:  45%|████▌     | 28/62 [02:46<03:24,  6.01s/it]
Training Epoch: 6/100, batch 54/125 completed (loss: 0.15:  45%|████▌     | 28/62 [02:46<03:24,  6.01s/it]
Training Epoch: 6/100, batch 55/125 completed (loss: 0.0746:  45%|████▌     | 28/62 [02:46<03:24,  6.01s/it]
Training Epoch: 6/100, batch 55/125 completed (loss: 0.0261:  45%|████▌     | 28/62 [02:46<03:24,  6.01s/it]
Training Epoch: 6/100, batch 55/125 completed (loss: 0.1029:  45%|████▌     | 28/62 [02:46<03:24,  6.01s/it]
Training Epoch: 6/100, batch 55/125 completed (loss: 0.0629:  45%|████▌     | 28/62 [02:46<03:24,  6.01s/it]
Training Epoch: 6/100, batch 56/125 completed (loss: 0.1515:  45%|████▌     | 28/62 [02:49<03:24,  6.01s/it]
Training Epoch: 6/100, batch 56/125 completed (loss: 0.3824:  45%|████▌     | 28/62 [02:49<03:24,  6.01s/it]
Training Epoch: 6/100, batch 56/125 completed (loss: 0.0404:  45%|████▌     | 28/62 [02:49<03:24,  6.01s/it]
Training Epoch: 6/100, batch 56/125 completed (loss: 0.0209:  45%|████▌     | 28/62 [02:49<03:24,  6.01s/it]
Training Epoch: 6/100, batch 56/125 completed (loss: 0.3824:  47%|████▋     | 29/62 [02:51<03:16,  5.94s/it]
Training Epoch: 6/100, batch 56/125 completed (loss: 0.0209:  47%|████▋     | 29/62 [02:51<03:16,  5.95s/it]
Training Epoch: 6/100, batch 56/125 completed (loss: 0.0404:  47%|████▋     | 29/62 [02:51<03:16,  5.94s/it]
Training Epoch: 6/100, batch 56/125 completed (loss: 0.1515:  47%|████▋     | 29/62 [02:51<03:16,  5.95s/it]
Training Epoch: 6/100, batch 57/125 completed (loss: 0.0682:  47%|████▋     | 29/62 [02:51<03:16,  5.95s/it]
Training Epoch: 6/100, batch 57/125 completed (loss: 0.3385:  47%|████▋     | 29/62 [02:51<03:16,  5.94s/it]
Training Epoch: 6/100, batch 57/125 completed (loss: 0.1657:  47%|████▋     | 29/62 [02:51<03:16,  5.94s/it]
Training Epoch: 6/100, batch 57/125 completed (loss: 0.1747:  47%|████▋     | 29/62 [02:51<03:16,  5.95s/it]
Training Epoch: 6/100, batch 58/125 completed (loss: 0.1495:  47%|████▋     | 29/62 [02:55<03:16,  5.94s/it]
Training Epoch: 6/100, batch 58/125 completed (loss: 0.1445:  47%|████▋     | 29/62 [02:55<03:16,  5.94s/it]
Training Epoch: 6/100, batch 58/125 completed (loss: 0.1477:  47%|████▋     | 29/62 [02:55<03:16,  5.95s/it]
Training Epoch: 6/100, batch 58/125 completed (loss: 0.2406:  47%|████▋     | 29/62 [02:55<03:16,  5.95s/it]
Training Epoch: 6/100, batch 58/125 completed (loss: 0.2406:  48%|████▊     | 30/62 [02:58<03:13,  6.04s/it]
Training Epoch: 6/100, batch 58/125 completed (loss: 0.1495:  48%|████▊     | 30/62 [02:58<03:13,  6.04s/it]
Training Epoch: 6/100, batch 58/125 completed (loss: 0.1477:  48%|████▊     | 30/62 [02:58<03:13,  6.04s/it]
Training Epoch: 6/100, batch 58/125 completed (loss: 0.1445:  48%|████▊     | 30/62 [02:58<03:13,  6.04s/it]
Training Epoch: 6/100, batch 59/125 completed (loss: 0.168:  48%|████▊     | 30/62 [02:58<03:13,  6.04s/it] 
Training Epoch: 6/100, batch 59/125 completed (loss: 0.0483:  48%|████▊     | 30/62 [02:58<03:13,  6.04s/it]
Training Epoch: 6/100, batch 59/125 completed (loss: 0.1106:  48%|████▊     | 30/62 [02:58<03:13,  6.04s/it]
Training Epoch: 6/100, batch 59/125 completed (loss: 0.0476:  48%|████▊     | 30/62 [02:58<03:13,  6.04s/it]
Training Epoch: 6/100, batch 60/125 completed (loss: 0.1479:  48%|████▊     | 30/62 [03:01<03:13,  6.04s/it]
Training Epoch: 6/100, batch 60/125 completed (loss: 0.1276:  48%|████▊     | 30/62 [03:01<03:13,  6.04s/it]
Training Epoch: 6/100, batch 60/125 completed (loss: 0.1093:  48%|████▊     | 30/62 [03:01<03:13,  6.04s/it]
Training Epoch: 6/100, batch 60/125 completed (loss: 0.0978:  48%|████▊     | 30/62 [03:01<03:13,  6.04s/it]
Training Epoch: 6/100, batch 60/125 completed (loss: 0.0978:  50%|█████     | 31/62 [03:03<03:05,  5.97s/it]
Training Epoch: 6/100, batch 60/125 completed (loss: 0.1479:  50%|█████     | 31/62 [03:03<03:05,  5.98s/it]
Training Epoch: 6/100, batch 60/125 completed (loss: 0.1093:  50%|█████     | 31/62 [03:03<03:05,  5.98s/it]
Training Epoch: 6/100, batch 60/125 completed (loss: 0.1276:  50%|█████     | 31/62 [03:03<03:05,  5.98s/it]
Training Epoch: 6/100, batch 61/125 completed (loss: 0.151:  50%|█████     | 31/62 [03:04<03:05,  5.98s/it] 
Training Epoch: 6/100, batch 61/125 completed (loss: 0.1929:  50%|█████     | 31/62 [03:04<03:05,  5.98s/it]
Training Epoch: 6/100, batch 61/125 completed (loss: 0.0437:  50%|█████     | 31/62 [03:04<03:05,  5.98s/it]
Training Epoch: 6/100, batch 61/125 completed (loss: 0.1559:  50%|█████     | 31/62 [03:04<03:05,  5.97s/it]
Training Epoch: 6/100, batch 62/125 completed (loss: 0.2204:  50%|█████     | 31/62 [03:06<03:05,  5.98s/it]
Training Epoch: 6/100, batch 62/125 completed (loss: 0.0767:  50%|█████     | 31/62 [03:06<03:05,  5.98s/it]
Training Epoch: 6/100, batch 62/125 completed (loss: 0.3746:  50%|█████     | 31/62 [03:06<03:05,  5.98s/it]
Training Epoch: 6/100, batch 62/125 completed (loss: 0.0988:  50%|█████     | 31/62 [03:06<03:05,  5.97s/it]
Training Epoch: 6/100, batch 62/125 completed (loss: 0.2204:  52%|█████▏    | 32/62 [03:09<02:59,  5.97s/it]
Training Epoch: 6/100, batch 62/125 completed (loss: 0.0767:  52%|█████▏    | 32/62 [03:09<02:58,  5.97s/it]
Training Epoch: 6/100, batch 62/125 completed (loss: 0.0988:  52%|█████▏    | 32/62 [03:09<02:59,  5.97s/it]
Training Epoch: 6/100, batch 62/125 completed (loss: 0.3746:  52%|█████▏    | 32/62 [03:09<02:58,  5.97s/it]
Training Epoch: 6/100, batch 63/125 completed (loss: 0.1114:  52%|█████▏    | 32/62 [03:09<02:58,  5.97s/it]
Training Epoch: 6/100, batch 63/125 completed (loss: 0.0657:  52%|█████▏    | 32/62 [03:09<02:59,  5.97s/it]
Training Epoch: 6/100, batch 63/125 completed (loss: 0.096:  52%|█████▏    | 32/62 [03:09<02:58,  5.97s/it] 
Training Epoch: 6/100, batch 63/125 completed (loss: 0.0474:  52%|█████▏    | 32/62 [03:10<02:59,  5.97s/it]
Training Epoch: 6/100, batch 64/125 completed (loss: 0.075:  52%|█████▏    | 32/62 [03:13<02:59,  5.97s/it] 
Training Epoch: 6/100, batch 64/125 completed (loss: 0.04:  52%|█████▏    | 32/62 [03:13<02:58,  5.97s/it]  
Training Epoch: 6/100, batch 64/125 completed (loss: 0.1234:  52%|█████▏    | 32/62 [03:13<02:58,  5.97s/it]
Training Epoch: 6/100, batch 64/125 completed (loss: 0.2747:  52%|█████▏    | 32/62 [03:13<02:59,  5.97s/it]
Training Epoch: 6/100, batch 64/125 completed (loss: 0.075:  53%|█████▎    | 33/62 [03:16<02:55,  6.04s/it]
Training Epoch: 6/100, batch 64/125 completed (loss: 0.04:  53%|█████▎    | 33/62 [03:16<02:55,  6.04s/it]
Training Epoch: 6/100, batch 64/125 completed (loss: 0.1234:  53%|█████▎    | 33/62 [03:16<02:55,  6.04s/it]
Training Epoch: 6/100, batch 64/125 completed (loss: 0.2747:  53%|█████▎    | 33/62 [03:16<02:55,  6.04s/it]
Training Epoch: 6/100, batch 65/125 completed (loss: 0.2693:  53%|█████▎    | 33/62 [03:16<02:55,  6.04s/it]
Training Epoch: 6/100, batch 65/125 completed (loss: 0.1524:  53%|█████▎    | 33/62 [03:16<02:55,  6.04s/it]
Training Epoch: 6/100, batch 65/125 completed (loss: 0.0939:  53%|█████▎    | 33/62 [03:16<02:55,  6.04s/it]
Training Epoch: 6/100, batch 65/125 completed (loss: 0.2276:  53%|█████▎    | 33/62 [03:16<02:55,  6.04s/it]
Training Epoch: 6/100, batch 66/125 completed (loss: 0.0589:  53%|█████▎    | 33/62 [03:19<02:55,  6.04s/it]
Training Epoch: 6/100, batch 66/125 completed (loss: 0.1277:  53%|█████▎    | 33/62 [03:19<02:55,  6.04s/it]
Training Epoch: 6/100, batch 66/125 completed (loss: 0.1223:  53%|█████▎    | 33/62 [03:19<02:55,  6.04s/it]
Training Epoch: 6/100, batch 66/125 completed (loss: 0.0878:  53%|█████▎    | 33/62 [03:19<02:55,  6.04s/it]
Training Epoch: 6/100, batch 66/125 completed (loss: 0.0878:  55%|█████▍    | 34/62 [03:21<02:47,  5.97s/it]
Training Epoch: 6/100, batch 66/125 completed (loss: 0.1277:  55%|█████▍    | 34/62 [03:21<02:47,  5.97s/it]
Training Epoch: 6/100, batch 66/125 completed (loss: 0.0589:  55%|█████▍    | 34/62 [03:21<02:47,  5.97s/it]
Training Epoch: 6/100, batch 66/125 completed (loss: 0.1223:  55%|█████▍    | 34/62 [03:21<02:47,  5.97s/it]
Training Epoch: 6/100, batch 67/125 completed (loss: 0.158:  55%|█████▍    | 34/62 [03:21<02:47,  5.97s/it] 
Training Epoch: 6/100, batch 67/125 completed (loss: 0.0483:  55%|█████▍    | 34/62 [03:21<02:47,  5.97s/it]
Training Epoch: 6/100, batch 67/125 completed (loss: 0.1746:  55%|█████▍    | 34/62 [03:21<02:47,  5.97s/it]
Training Epoch: 6/100, batch 67/125 completed (loss: 0.0828:  55%|█████▍    | 34/62 [03:21<02:47,  5.97s/it]
Training Epoch: 6/100, batch 68/125 completed (loss: 0.0827:  55%|█████▍    | 34/62 [03:24<02:47,  5.97s/it]
Training Epoch: 6/100, batch 68/125 completed (loss: 0.0288:  55%|█████▍    | 34/62 [03:24<02:47,  5.97s/it]
Training Epoch: 6/100, batch 68/125 completed (loss: 0.1438:  55%|█████▍    | 34/62 [03:24<02:47,  5.97s/it]
Training Epoch: 6/100, batch 68/125 completed (loss: 0.17:  55%|█████▍    | 34/62 [03:24<02:47,  5.97s/it]  
Training Epoch: 6/100, batch 68/125 completed (loss: 0.1438:  56%|█████▋    | 35/62 [03:28<02:44,  6.10s/it]
Training Epoch: 6/100, batch 68/125 completed (loss: 0.17:  56%|█████▋    | 35/62 [03:28<02:44,  6.10s/it]
Training Epoch: 6/100, batch 68/125 completed (loss: 0.0827:  56%|█████▋    | 35/62 [03:28<02:44,  6.10s/it]
Training Epoch: 6/100, batch 68/125 completed (loss: 0.0288:  56%|█████▋    | 35/62 [03:28<02:44,  6.10s/it]
Training Epoch: 6/100, batch 69/125 completed (loss: 0.0989:  56%|█████▋    | 35/62 [03:28<02:44,  6.10s/it]
Training Epoch: 6/100, batch 69/125 completed (loss: 0.297:  56%|█████▋    | 35/62 [03:28<02:44,  6.10s/it] 
Training Epoch: 6/100, batch 69/125 completed (loss: 0.1383:  56%|█████▋    | 35/62 [03:28<02:44,  6.10s/it]
Training Epoch: 6/100, batch 69/125 completed (loss: 0.0492:  56%|█████▋    | 35/62 [03:28<02:44,  6.10s/it]
Training Epoch: 6/100, batch 70/125 completed (loss: 0.0619:  56%|█████▋    | 35/62 [03:31<02:44,  6.10s/it]
Training Epoch: 6/100, batch 70/125 completed (loss: 0.1137:  56%|█████▋    | 35/62 [03:31<02:44,  6.10s/it]
Training Epoch: 6/100, batch 70/125 completed (loss: 0.1336:  56%|█████▋    | 35/62 [03:31<02:44,  6.10s/it]
Training Epoch: 6/100, batch 70/125 completed (loss: 0.1189:  56%|█████▋    | 35/62 [03:31<02:44,  6.10s/it]
Training Epoch: 6/100, batch 70/125 completed (loss: 0.0619:  58%|█████▊    | 36/62 [03:34<02:36,  6.02s/it]
Training Epoch: 6/100, batch 70/125 completed (loss: 0.1137:  58%|█████▊    | 36/62 [03:34<02:36,  6.02s/it]
Training Epoch: 6/100, batch 70/125 completed (loss: 0.1189:  58%|█████▊    | 36/62 [03:34<02:36,  6.02s/it]
Training Epoch: 6/100, batch 70/125 completed (loss: 0.1336:  58%|█████▊    | 36/62 [03:34<02:36,  6.02s/it]
Training Epoch: 6/100, batch 71/125 completed (loss: 0.0614:  58%|█████▊    | 36/62 [03:34<02:36,  6.02s/it]
Training Epoch: 6/100, batch 71/125 completed (loss: 0.1004:  58%|█████▊    | 36/62 [03:34<02:36,  6.02s/it]
Training Epoch: 6/100, batch 71/125 completed (loss: 0.0778:  58%|█████▊    | 36/62 [03:34<02:36,  6.02s/it]
Training Epoch: 6/100, batch 71/125 completed (loss: 0.075:  58%|█████▊    | 36/62 [03:34<02:36,  6.02s/it] 
Training Epoch: 6/100, batch 72/125 completed (loss: 0.1122:  58%|█████▊    | 36/62 [03:37<02:36,  6.02s/it]
Training Epoch: 6/100, batch 72/125 completed (loss: 0.2535:  58%|█████▊    | 36/62 [03:37<02:36,  6.02s/it]
Training Epoch: 6/100, batch 72/125 completed (loss: 0.0944:  58%|█████▊    | 36/62 [03:37<02:36,  6.02s/it]
Training Epoch: 6/100, batch 72/125 completed (loss: 0.0552:  58%|█████▊    | 36/62 [03:37<02:36,  6.02s/it]
Training Epoch: 6/100, batch 72/125 completed (loss: 0.0552:  60%|█████▉    | 37/62 [03:39<02:28,  5.96s/it]
Training Epoch: 6/100, batch 72/125 completed (loss: 0.2535:  60%|█████▉    | 37/62 [03:39<02:29,  5.96s/it]
Training Epoch: 6/100, batch 72/125 completed (loss: 0.1122:  60%|█████▉    | 37/62 [03:39<02:29,  5.96s/it]
Training Epoch: 6/100, batch 72/125 completed (loss: 0.0944:  60%|█████▉    | 37/62 [03:40<02:29,  5.97s/it]
Training Epoch: 6/100, batch 73/125 completed (loss: 0.3488:  60%|█████▉    | 37/62 [03:40<02:29,  5.96s/it]
Training Epoch: 6/100, batch 73/125 completed (loss: 0.1014:  60%|█████▉    | 37/62 [03:40<02:29,  5.96s/it]
Training Epoch: 6/100, batch 73/125 completed (loss: 0.0756:  60%|█████▉    | 37/62 [03:40<02:29,  5.97s/it]
Training Epoch: 6/100, batch 73/125 completed (loss: 0.0272:  60%|█████▉    | 37/62 [03:40<02:28,  5.96s/it]
Training Epoch: 6/100, batch 74/125 completed (loss: 0.3707:  60%|█████▉    | 37/62 [03:43<02:29,  5.97s/it]
Training Epoch: 6/100, batch 74/125 completed (loss: 0.087:  60%|█████▉    | 37/62 [03:43<02:29,  5.96s/it] 
Training Epoch: 6/100, batch 74/125 completed (loss: 0.1908:  60%|█████▉    | 37/62 [03:43<02:29,  5.96s/it]
Training Epoch: 6/100, batch 74/125 completed (loss: 0.0992:  60%|█████▉    | 37/62 [03:43<02:28,  5.96s/it]
Training Epoch: 6/100, batch 74/125 completed (loss: 0.1908:  61%|██████▏   | 38/62 [03:46<02:25,  6.05s/it]
Training Epoch: 6/100, batch 74/125 completed (loss: 0.0992:  61%|██████▏   | 38/62 [03:46<02:25,  6.05s/it]
Training Epoch: 6/100, batch 74/125 completed (loss: 0.3707:  61%|██████▏   | 38/62 [03:46<02:25,  6.05s/it]
Training Epoch: 6/100, batch 74/125 completed (loss: 0.087:  61%|██████▏   | 38/62 [03:46<02:25,  6.05s/it]
Training Epoch: 6/100, batch 75/125 completed (loss: 0.1933:  61%|██████▏   | 38/62 [03:46<02:25,  6.05s/it]
Training Epoch: 6/100, batch 75/125 completed (loss: 0.0695:  61%|██████▏   | 38/62 [03:46<02:25,  6.05s/it]
Training Epoch: 6/100, batch 75/125 completed (loss: 0.368:  61%|██████▏   | 38/62 [03:46<02:25,  6.05s/it] 
Training Epoch: 6/100, batch 75/125 completed (loss: 0.2185:  61%|██████▏   | 38/62 [03:46<02:25,  6.05s/it]
Training Epoch: 6/100, batch 76/125 completed (loss: 0.0962:  61%|██████▏   | 38/62 [03:49<02:25,  6.05s/it]
Training Epoch: 6/100, batch 76/125 completed (loss: 0.3247:  61%|██████▏   | 38/62 [03:49<02:25,  6.05s/it]
Training Epoch: 6/100, batch 76/125 completed (loss: 0.0779:  61%|██████▏   | 38/62 [03:49<02:25,  6.05s/it]
Training Epoch: 6/100, batch 76/125 completed (loss: 0.1693:  61%|██████▏   | 38/62 [03:49<02:25,  6.05s/it]
Training Epoch: 6/100, batch 76/125 completed (loss: 0.3247:  63%|██████▎   | 39/62 [03:52<02:17,  5.98s/it]
Training Epoch: 6/100, batch 76/125 completed (loss: 0.1693:  63%|██████▎   | 39/62 [03:52<02:17,  5.98s/it]
Training Epoch: 6/100, batch 76/125 completed (loss: 0.0779:  63%|██████▎   | 39/62 [03:52<02:17,  5.98s/it]
Training Epoch: 6/100, batch 76/125 completed (loss: 0.0962:  63%|██████▎   | 39/62 [03:52<02:17,  5.98s/it]
Training Epoch: 6/100, batch 77/125 completed (loss: 0.2634:  63%|██████▎   | 39/62 [03:52<02:17,  5.98s/it]
Training Epoch: 6/100, batch 77/125 completed (loss: 0.1995:  63%|██████▎   | 39/62 [03:52<02:17,  5.98s/it]
Training Epoch: 6/100, batch 77/125 completed (loss: 0.1201:  63%|██████▎   | 39/62 [03:52<02:17,  5.98s/it]
Training Epoch: 6/100, batch 77/125 completed (loss: 0.0693:  63%|██████▎   | 39/62 [03:52<02:17,  5.98s/it]
Training Epoch: 6/100, batch 78/125 completed (loss: 0.1032:  63%|██████▎   | 39/62 [03:55<02:17,  5.98s/it]
Training Epoch: 6/100, batch 78/125 completed (loss: 0.2585:  63%|██████▎   | 39/62 [03:55<02:17,  5.98s/it]
Training Epoch: 6/100, batch 78/125 completed (loss: 0.1828:  63%|██████▎   | 39/62 [03:55<02:17,  5.98s/it]
Training Epoch: 6/100, batch 78/125 completed (loss: 0.0232:  63%|██████▎   | 39/62 [03:55<02:17,  5.98s/it]
Training Epoch: 6/100, batch 78/125 completed (loss: 0.0232:  65%|██████▍   | 40/62 [03:58<02:13,  6.06s/it]
Training Epoch: 6/100, batch 78/125 completed (loss: 0.2585:  65%|██████▍   | 40/62 [03:58<02:13,  6.06s/it]
Training Epoch: 6/100, batch 78/125 completed (loss: 0.1032:  65%|██████▍   | 40/62 [03:58<02:13,  6.06s/it]
Training Epoch: 6/100, batch 78/125 completed (loss: 0.1828:  65%|██████▍   | 40/62 [03:58<02:13,  6.06s/it]
Training Epoch: 6/100, batch 79/125 completed (loss: 0.0657:  65%|██████▍   | 40/62 [03:58<02:13,  6.06s/it]
Training Epoch: 6/100, batch 79/125 completed (loss: 0.1108:  65%|██████▍   | 40/62 [03:58<02:13,  6.06s/it]
Training Epoch: 6/100, batch 79/125 completed (loss: 0.0529:  65%|██████▍   | 40/62 [03:58<02:13,  6.06s/it]
Training Epoch: 6/100, batch 79/125 completed (loss: 0.0349:  65%|██████▍   | 40/62 [03:58<02:13,  6.06s/it]
Training Epoch: 6/100, batch 80/125 completed (loss: 0.2583:  65%|██████▍   | 40/62 [04:01<02:13,  6.06s/it]
Training Epoch: 6/100, batch 80/125 completed (loss: 0.1254:  65%|██████▍   | 40/62 [04:01<02:13,  6.06s/it]
Training Epoch: 6/100, batch 80/125 completed (loss: 0.1357:  65%|██████▍   | 40/62 [04:01<02:13,  6.06s/it]
Training Epoch: 6/100, batch 80/125 completed (loss: 0.0901:  65%|██████▍   | 40/62 [04:01<02:13,  6.06s/it]
Training Epoch: 6/100, batch 80/125 completed (loss: 0.1254:  66%|██████▌   | 41/62 [04:04<02:06,  6.03s/it]
Training Epoch: 6/100, batch 80/125 completed (loss: 0.0901:  66%|██████▌   | 41/62 [04:04<02:06,  6.03s/it]
Training Epoch: 6/100, batch 80/125 completed (loss: 0.1357:  66%|██████▌   | 41/62 [04:04<02:06,  6.03s/it]
Training Epoch: 6/100, batch 80/125 completed (loss: 0.2583:  66%|██████▌   | 41/62 [04:04<02:06,  6.03s/it]
Training Epoch: 6/100, batch 81/125 completed (loss: 0.0356:  66%|██████▌   | 41/62 [04:04<02:06,  6.03s/it]
Training Epoch: 6/100, batch 81/125 completed (loss: 0.0792:  66%|██████▌   | 41/62 [04:04<02:06,  6.03s/it]
Training Epoch: 6/100, batch 81/125 completed (loss: 0.1431:  66%|██████▌   | 41/62 [04:04<02:06,  6.03s/it]
Training Epoch: 6/100, batch 81/125 completed (loss: 0.2338:  66%|██████▌   | 41/62 [04:04<02:06,  6.03s/it]
Training Epoch: 6/100, batch 82/125 completed (loss: 0.0571:  66%|██████▌   | 41/62 [04:07<02:06,  6.03s/it]
Training Epoch: 6/100, batch 82/125 completed (loss: 0.0855:  66%|██████▌   | 41/62 [04:07<02:06,  6.03s/it]
Training Epoch: 6/100, batch 82/125 completed (loss: 0.0933:  66%|██████▌   | 41/62 [04:07<02:06,  6.03s/it]
Training Epoch: 6/100, batch 82/125 completed (loss: 0.1159:  66%|██████▌   | 41/62 [04:07<02:06,  6.03s/it]
Training Epoch: 6/100, batch 82/125 completed (loss: 0.0855:  68%|██████▊   | 42/62 [04:10<01:59,  5.96s/it]
Training Epoch: 6/100, batch 82/125 completed (loss: 0.0571:  68%|██████▊   | 42/62 [04:10<01:59,  5.96s/it]
Training Epoch: 6/100, batch 82/125 completed (loss: 0.0933:  68%|██████▊   | 42/62 [04:10<01:59,  5.96s/it]
Training Epoch: 6/100, batch 82/125 completed (loss: 0.1159:  68%|██████▊   | 42/62 [04:10<01:59,  5.96s/it]
Training Epoch: 6/100, batch 83/125 completed (loss: 0.0993:  68%|██████▊   | 42/62 [04:10<01:59,  5.96s/it]
Training Epoch: 6/100, batch 83/125 completed (loss: 0.2827:  68%|██████▊   | 42/62 [04:10<01:59,  5.96s/it]
Training Epoch: 6/100, batch 83/125 completed (loss: 0.1677:  68%|██████▊   | 42/62 [04:10<01:59,  5.96s/it]
Training Epoch: 6/100, batch 83/125 completed (loss: 0.0658:  68%|██████▊   | 42/62 [04:10<01:59,  5.96s/it]
Training Epoch: 6/100, batch 84/125 completed (loss: 0.1402:  68%|██████▊   | 42/62 [04:13<01:59,  5.96s/it]
Training Epoch: 6/100, batch 84/125 completed (loss: 0.1309:  68%|██████▊   | 42/62 [04:13<01:59,  5.96s/it]
Training Epoch: 6/100, batch 84/125 completed (loss: 0.1106:  68%|██████▊   | 42/62 [04:13<01:59,  5.96s/it]
Training Epoch: 6/100, batch 84/125 completed (loss: 0.0967:  68%|██████▊   | 42/62 [04:13<01:59,  5.96s/it]
Training Epoch: 6/100, batch 84/125 completed (loss: 0.1106:  69%|██████▉   | 43/62 [04:16<01:55,  6.10s/it]
Training Epoch: 6/100, batch 84/125 completed (loss: 0.1309:  69%|██████▉   | 43/62 [04:16<01:55,  6.10s/it]
Training Epoch: 6/100, batch 84/125 completed (loss: 0.1402:  69%|██████▉   | 43/62 [04:16<01:55,  6.10s/it]
Training Epoch: 6/100, batch 84/125 completed (loss: 0.0967:  69%|██████▉   | 43/62 [04:16<01:55,  6.10s/it]
Training Epoch: 6/100, batch 85/125 completed (loss: 0.1814:  69%|██████▉   | 43/62 [04:16<01:55,  6.10s/it]
Training Epoch: 6/100, batch 85/125 completed (loss: 0.0683:  69%|██████▉   | 43/62 [04:16<01:55,  6.10s/it]
Training Epoch: 6/100, batch 85/125 completed (loss: 0.1063:  69%|██████▉   | 43/62 [04:16<01:55,  6.10s/it]
Training Epoch: 6/100, batch 85/125 completed (loss: 0.0407:  69%|██████▉   | 43/62 [04:16<01:55,  6.10s/it]
Training Epoch: 6/100, batch 86/125 completed (loss: 0.3488:  69%|██████▉   | 43/62 [04:19<01:55,  6.10s/it]
Training Epoch: 6/100, batch 86/125 completed (loss: 0.1427:  69%|██████▉   | 43/62 [04:19<01:55,  6.10s/it]
Training Epoch: 6/100, batch 86/125 completed (loss: 0.2609:  69%|██████▉   | 43/62 [04:19<01:55,  6.10s/it]
Training Epoch: 6/100, batch 86/125 completed (loss: 0.0801:  69%|██████▉   | 43/62 [04:19<01:55,  6.10s/it]
Training Epoch: 6/100, batch 86/125 completed (loss: 0.2609:  71%|███████   | 44/62 [04:22<01:48,  6.02s/it]
Training Epoch: 6/100, batch 86/125 completed (loss: 0.1427:  71%|███████   | 44/62 [04:22<01:48,  6.02s/it]
Training Epoch: 6/100, batch 86/125 completed (loss: 0.0801:  71%|███████   | 44/62 [04:22<01:48,  6.02s/it]
Training Epoch: 6/100, batch 86/125 completed (loss: 0.3488:  71%|███████   | 44/62 [04:22<01:48,  6.02s/it]
Training Epoch: 6/100, batch 87/125 completed (loss: 0.037:  71%|███████   | 44/62 [04:22<01:48,  6.02s/it] 
Training Epoch: 6/100, batch 87/125 completed (loss: 0.1135:  71%|███████   | 44/62 [04:22<01:48,  6.02s/it]
Training Epoch: 6/100, batch 87/125 completed (loss: 0.0648:  71%|███████   | 44/62 [04:22<01:48,  6.02s/it]
Training Epoch: 6/100, batch 87/125 completed (loss: 0.1531:  71%|███████   | 44/62 [04:22<01:48,  6.02s/it]
Training Epoch: 6/100, batch 88/125 completed (loss: 0.0913:  71%|███████   | 44/62 [04:25<01:48,  6.02s/it]
Training Epoch: 6/100, batch 88/125 completed (loss: 0.1348:  71%|███████   | 44/62 [04:25<01:48,  6.02s/it]
Training Epoch: 6/100, batch 88/125 completed (loss: 0.2053:  71%|███████   | 44/62 [04:25<01:48,  6.02s/it]
Training Epoch: 6/100, batch 88/125 completed (loss: 0.2826:  71%|███████   | 44/62 [04:25<01:48,  6.02s/it]
Training Epoch: 6/100, batch 88/125 completed (loss: 0.2826:  73%|███████▎  | 45/62 [04:28<01:42,  6.01s/it]
Training Epoch: 6/100, batch 88/125 completed (loss: 0.2053:  73%|███████▎  | 45/62 [04:28<01:42,  6.01s/it]
Training Epoch: 6/100, batch 88/125 completed (loss: 0.0913:  73%|███████▎  | 45/62 [04:28<01:42,  6.01s/it]
Training Epoch: 6/100, batch 88/125 completed (loss: 0.1348:  73%|███████▎  | 45/62 [04:28<01:42,  6.02s/it]
Training Epoch: 6/100, batch 89/125 completed (loss: 0.4106:  73%|███████▎  | 45/62 [04:28<01:42,  6.01s/it]
Training Epoch: 6/100, batch 89/125 completed (loss: 0.0905:  73%|███████▎  | 45/62 [04:28<01:42,  6.01s/it]
Training Epoch: 6/100, batch 89/125 completed (loss: 0.1638:  73%|███████▎  | 45/62 [04:28<01:42,  6.02s/it]
Training Epoch: 6/100, batch 89/125 completed (loss: 0.3844:  73%|███████▎  | 45/62 [04:28<01:42,  6.01s/it]
Training Epoch: 6/100, batch 90/125 completed (loss: 0.1244:  73%|███████▎  | 45/62 [04:31<01:42,  6.02s/it]
Training Epoch: 6/100, batch 90/125 completed (loss: 0.1055:  73%|███████▎  | 45/62 [04:31<01:42,  6.01s/it]
Training Epoch: 6/100, batch 90/125 completed (loss: 0.1406:  73%|███████▎  | 45/62 [04:31<01:42,  6.01s/it]
Training Epoch: 6/100, batch 90/125 completed (loss: 0.0347:  73%|███████▎  | 45/62 [04:31<01:42,  6.01s/it]
Training Epoch: 6/100, batch 90/125 completed (loss: 0.0347:  74%|███████▍  | 46/62 [04:34<01:37,  6.09s/it]
Training Epoch: 6/100, batch 90/125 completed (loss: 0.1055:  74%|███████▍  | 46/62 [04:34<01:37,  6.09s/it]
Training Epoch: 6/100, batch 90/125 completed (loss: 0.1406:  74%|███████▍  | 46/62 [04:34<01:37,  6.09s/it]
Training Epoch: 6/100, batch 90/125 completed (loss: 0.1244:  74%|███████▍  | 46/62 [04:34<01:37,  6.09s/it]
Training Epoch: 6/100, batch 91/125 completed (loss: 0.1626:  74%|███████▍  | 46/62 [04:34<01:37,  6.09s/it]
Training Epoch: 6/100, batch 91/125 completed (loss: 0.1161:  74%|███████▍  | 46/62 [04:34<01:37,  6.09s/it]
Training Epoch: 6/100, batch 91/125 completed (loss: 0.1235:  74%|███████▍  | 46/62 [04:34<01:37,  6.09s/it]
Training Epoch: 6/100, batch 91/125 completed (loss: 0.0439:  74%|███████▍  | 46/62 [04:34<01:37,  6.09s/it]
Training Epoch: 6/100, batch 92/125 completed (loss: 0.2746:  74%|███████▍  | 46/62 [04:37<01:37,  6.09s/it]
Training Epoch: 6/100, batch 92/125 completed (loss: 0.0737:  74%|███████▍  | 46/62 [04:37<01:37,  6.09s/it]
Training Epoch: 6/100, batch 92/125 completed (loss: 0.1664:  74%|███████▍  | 46/62 [04:37<01:37,  6.09s/it]
Training Epoch: 6/100, batch 92/125 completed (loss: 0.3238:  74%|███████▍  | 46/62 [04:37<01:37,  6.09s/it]
Training Epoch: 6/100, batch 92/125 completed (loss: 0.3238:  76%|███████▌  | 47/62 [04:40<01:30,  6.00s/it]
Training Epoch: 6/100, batch 92/125 completed (loss: 0.2746:  76%|███████▌  | 47/62 [04:40<01:30,  6.00s/it]
Training Epoch: 6/100, batch 92/125 completed (loss: 0.0737:  76%|███████▌  | 47/62 [04:40<01:30,  6.00s/it]
Training Epoch: 6/100, batch 92/125 completed (loss: 0.1664:  76%|███████▌  | 47/62 [04:40<01:30,  6.01s/it]
Training Epoch: 6/100, batch 93/125 completed (loss: 0.0893:  76%|███████▌  | 47/62 [04:40<01:30,  6.01s/it]
Training Epoch: 6/100, batch 93/125 completed (loss: 0.2979:  76%|███████▌  | 47/62 [04:40<01:30,  6.00s/it]
Training Epoch: 6/100, batch 93/125 completed (loss: 0.0969:  76%|███████▌  | 47/62 [04:40<01:30,  6.00s/it]
Training Epoch: 6/100, batch 93/125 completed (loss: 0.1009:  76%|███████▌  | 47/62 [04:40<01:30,  6.00s/it]
Training Epoch: 6/100, batch 94/125 completed (loss: 0.2265:  76%|███████▌  | 47/62 [04:43<01:30,  6.01s/it]
Training Epoch: 6/100, batch 94/125 completed (loss: 0.1058:  76%|███████▌  | 47/62 [04:43<01:30,  6.00s/it]
Training Epoch: 6/100, batch 94/125 completed (loss: 0.2311:  76%|███████▌  | 47/62 [04:43<01:30,  6.00s/it]
Training Epoch: 6/100, batch 94/125 completed (loss: 0.0245:  76%|███████▌  | 47/62 [04:43<01:30,  6.00s/it]
Training Epoch: 6/100, batch 94/125 completed (loss: 0.0245:  77%|███████▋  | 48/62 [04:46<01:25,  6.08s/it]
Training Epoch: 6/100, batch 94/125 completed (loss: 0.1058:  77%|███████▋  | 48/62 [04:46<01:25,  6.08s/it]
Training Epoch: 6/100, batch 94/125 completed (loss: 0.2265:  77%|███████▋  | 48/62 [04:46<01:25,  6.07s/it]
Training Epoch: 6/100, batch 94/125 completed (loss: 0.2311:  77%|███████▋  | 48/62 [04:46<01:25,  6.08s/it]
Training Epoch: 6/100, batch 95/125 completed (loss: 0.1463:  77%|███████▋  | 48/62 [04:46<01:25,  6.08s/it]
Training Epoch: 6/100, batch 95/125 completed (loss: 0.2712:  77%|███████▋  | 48/62 [04:46<01:25,  6.08s/it]
Training Epoch: 6/100, batch 95/125 completed (loss: 0.0812:  77%|███████▋  | 48/62 [04:46<01:25,  6.07s/it]
Training Epoch: 6/100, batch 95/125 completed (loss: 0.2831:  77%|███████▋  | 48/62 [04:46<01:25,  6.08s/it]
Training Epoch: 6/100, batch 96/125 completed (loss: 0.3294:  77%|███████▋  | 48/62 [04:49<01:25,  6.07s/it]
Training Epoch: 6/100, batch 96/125 completed (loss: 0.2496:  77%|███████▋  | 48/62 [04:49<01:25,  6.08s/it]
Training Epoch: 6/100, batch 96/125 completed (loss: 0.0993:  77%|███████▋  | 48/62 [04:49<01:25,  6.08s/it]
Training Epoch: 6/100, batch 96/125 completed (loss: 0.0595:  77%|███████▋  | 48/62 [04:49<01:25,  6.08s/it]
Training Epoch: 6/100, batch 96/125 completed (loss: 0.2496:  79%|███████▉  | 49/62 [04:52<01:18,  6.04s/it]
Training Epoch: 6/100, batch 96/125 completed (loss: 0.0993:  79%|███████▉  | 49/62 [04:52<01:18,  6.04s/it]
Training Epoch: 6/100, batch 96/125 completed (loss: 0.3294:  79%|███████▉  | 49/62 [04:52<01:18,  6.04s/it]
Training Epoch: 6/100, batch 96/125 completed (loss: 0.0595:  79%|███████▉  | 49/62 [04:52<01:18,  6.04s/it]
Training Epoch: 6/100, batch 97/125 completed (loss: 0.3895:  79%|███████▉  | 49/62 [04:52<01:18,  6.04s/it]
Training Epoch: 6/100, batch 97/125 completed (loss: 0.1684:  79%|███████▉  | 49/62 [04:52<01:18,  6.04s/it]
Training Epoch: 6/100, batch 97/125 completed (loss: 0.0775:  79%|███████▉  | 49/62 [04:52<01:18,  6.04s/it]
Training Epoch: 6/100, batch 97/125 completed (loss: 0.2429:  79%|███████▉  | 49/62 [04:52<01:18,  6.04s/it]
Training Epoch: 6/100, batch 98/125 completed (loss: 0.072:  79%|███████▉  | 49/62 [04:55<01:18,  6.04s/it] 
Training Epoch: 6/100, batch 98/125 completed (loss: 0.1208:  79%|███████▉  | 49/62 [04:55<01:18,  6.04s/it]
Training Epoch: 6/100, batch 98/125 completed (loss: 0.1675:  79%|███████▉  | 49/62 [04:55<01:18,  6.04s/it]
Training Epoch: 6/100, batch 98/125 completed (loss: 0.0351:  79%|███████▉  | 49/62 [04:55<01:18,  6.04s/it]
Training Epoch: 6/100, batch 98/125 completed (loss: 0.1208:  81%|████████  | 50/62 [04:58<01:11,  5.97s/it]
Training Epoch: 6/100, batch 98/125 completed (loss: 0.072:  81%|████████  | 50/62 [04:58<01:11,  5.97s/it]
Training Epoch: 6/100, batch 98/125 completed (loss: 0.1675:  81%|████████  | 50/62 [04:58<01:11,  5.97s/it]
Training Epoch: 6/100, batch 98/125 completed (loss: 0.0351:  81%|████████  | 50/62 [04:58<01:11,  5.98s/it]
Training Epoch: 6/100, batch 99/125 completed (loss: 0.0999:  81%|████████  | 50/62 [04:58<01:11,  5.97s/it]
Training Epoch: 6/100, batch 99/125 completed (loss: 0.1049:  81%|████████  | 50/62 [04:58<01:11,  5.97s/it]
Training Epoch: 6/100, batch 99/125 completed (loss: 0.056:  81%|████████  | 50/62 [04:58<01:11,  5.97s/it] 
Training Epoch: 6/100, batch 99/125 completed (loss: 0.1213:  81%|████████  | 50/62 [04:58<01:11,  5.98s/it]
Training Epoch: 6/100, batch 100/125 completed (loss: 0.1368:  81%|████████  | 50/62 [05:01<01:11,  5.97s/it]
Training Epoch: 6/100, batch 100/125 completed (loss: 0.0926:  81%|████████  | 50/62 [05:01<01:11,  5.97s/it]
Training Epoch: 6/100, batch 100/125 completed (loss: 0.2145:  81%|████████  | 50/62 [05:01<01:11,  5.97s/it]
Training Epoch: 6/100, batch 100/125 completed (loss: 0.1642:  81%|████████  | 50/62 [05:01<01:11,  5.98s/it]
Training Epoch: 6/100, batch 100/125 completed (loss: 0.0926:  82%|████████▏ | 51/62 [05:04<01:07,  6.11s/it]
Training Epoch: 6/100, batch 100/125 completed (loss: 0.1642:  82%|████████▏ | 51/62 [05:04<01:07,  6.11s/it]
Training Epoch: 6/100, batch 100/125 completed (loss: 0.1368:  82%|████████▏ | 51/62 [05:04<01:07,  6.12s/it]
Training Epoch: 6/100, batch 100/125 completed (loss: 0.2145:  82%|████████▏ | 51/62 [05:04<01:07,  6.12s/it]
Training Epoch: 6/100, batch 101/125 completed (loss: 0.093:  82%|████████▏ | 51/62 [05:04<01:07,  6.11s/it] 
Training Epoch: 6/100, batch 101/125 completed (loss: 0.059:  82%|████████▏ | 51/62 [05:04<01:07,  6.12s/it] 
Training Epoch: 6/100, batch 101/125 completed (loss: 0.0874:  82%|████████▏ | 51/62 [05:04<01:07,  6.12s/it]
Training Epoch: 6/100, batch 101/125 completed (loss: 0.1221:  82%|████████▏ | 51/62 [05:04<01:07,  6.11s/it]
Training Epoch: 6/100, batch 102/125 completed (loss: 0.1086:  82%|████████▏ | 51/62 [05:07<01:07,  6.11s/it]
Training Epoch: 6/100, batch 102/125 completed (loss: 0.2548:  82%|████████▏ | 51/62 [05:07<01:07,  6.12s/it]
Training Epoch: 6/100, batch 102/125 completed (loss: 0.1226:  82%|████████▏ | 51/62 [05:07<01:07,  6.12s/it]
Training Epoch: 6/100, batch 102/125 completed (loss: 0.0986:  82%|████████▏ | 51/62 [05:07<01:07,  6.11s/it]
Training Epoch: 6/100, batch 102/125 completed (loss: 0.1086:  84%|████████▍ | 52/62 [05:10<01:00,  6.02s/it]
Training Epoch: 6/100, batch 102/125 completed (loss: 0.1226:  84%|████████▍ | 52/62 [05:10<01:00,  6.02s/it]
Training Epoch: 6/100, batch 102/125 completed (loss: 0.0986:  84%|████████▍ | 52/62 [05:10<01:00,  6.02s/it]
Training Epoch: 6/100, batch 102/125 completed (loss: 0.2548:  84%|████████▍ | 52/62 [05:10<01:00,  6.02s/it]
Training Epoch: 6/100, batch 103/125 completed (loss: 0.0427:  84%|████████▍ | 52/62 [05:10<01:00,  6.02s/it]
Training Epoch: 6/100, batch 103/125 completed (loss: 0.0584:  84%|████████▍ | 52/62 [05:10<01:00,  6.02s/it]
Training Epoch: 6/100, batch 103/125 completed (loss: 0.1094:  84%|████████▍ | 52/62 [05:10<01:00,  6.02s/it]
Training Epoch: 6/100, batch 103/125 completed (loss: 0.1484:  84%|████████▍ | 52/62 [05:10<01:00,  6.02s/it]
Training Epoch: 6/100, batch 104/125 completed (loss: 0.0902:  84%|████████▍ | 52/62 [05:13<01:00,  6.02s/it]
Training Epoch: 6/100, batch 104/125 completed (loss: 0.1165:  84%|████████▍ | 52/62 [05:13<01:00,  6.02s/it]
Training Epoch: 6/100, batch 104/125 completed (loss: 0.0561:  84%|████████▍ | 52/62 [05:13<01:00,  6.02s/it]
Training Epoch: 6/100, batch 104/125 completed (loss: 0.0388:  84%|████████▍ | 52/62 [05:13<01:00,  6.02s/it]
Training Epoch: 6/100, batch 104/125 completed (loss: 0.0388:  85%|████████▌ | 53/62 [05:16<00:54,  6.07s/it]
Training Epoch: 6/100, batch 104/125 completed (loss: 0.1165:  85%|████████▌ | 53/62 [05:16<00:54,  6.07s/it]
Training Epoch: 6/100, batch 104/125 completed (loss: 0.0561:  85%|████████▌ | 53/62 [05:16<00:54,  6.07s/it]
Training Epoch: 6/100, batch 104/125 completed (loss: 0.0902:  85%|████████▌ | 53/62 [05:16<00:54,  6.07s/it]
Training Epoch: 6/100, batch 105/125 completed (loss: 0.0834:  85%|████████▌ | 53/62 [05:16<00:54,  6.07s/it]
Training Epoch: 6/100, batch 105/125 completed (loss: 0.0802:  85%|████████▌ | 53/62 [05:16<00:54,  6.07s/it]
Training Epoch: 6/100, batch 105/125 completed (loss: 0.2453:  85%|████████▌ | 53/62 [05:16<00:54,  6.07s/it]
Training Epoch: 6/100, batch 105/125 completed (loss: 0.1659:  85%|████████▌ | 53/62 [05:16<00:54,  6.07s/it]
Training Epoch: 6/100, batch 106/125 completed (loss: 0.143:  85%|████████▌ | 53/62 [05:19<00:54,  6.07s/it] 
Training Epoch: 6/100, batch 106/125 completed (loss: 0.264:  85%|████████▌ | 53/62 [05:19<00:54,  6.07s/it] 
Training Epoch: 6/100, batch 106/125 completed (loss: 0.1107:  85%|████████▌ | 53/62 [05:19<00:54,  6.07s/it]
Training Epoch: 6/100, batch 106/125 completed (loss: 0.1401:  85%|████████▌ | 53/62 [05:19<00:54,  6.07s/it]
Training Epoch: 6/100, batch 106/125 completed (loss: 0.143:  87%|████████▋ | 54/62 [05:22<00:48,  6.04s/it]
Training Epoch: 6/100, batch 106/125 completed (loss: 0.1401:  87%|████████▋ | 54/62 [05:22<00:48,  6.04s/it]
Training Epoch: 6/100, batch 106/125 completed (loss: 0.264:  87%|████████▋ | 54/62 [05:22<00:48,  6.04s/it]
Training Epoch: 6/100, batch 106/125 completed (loss: 0.1107:  87%|████████▋ | 54/62 [05:22<00:48,  6.04s/it]
Training Epoch: 6/100, batch 107/125 completed (loss: 0.0481:  87%|████████▋ | 54/62 [05:22<00:48,  6.04s/it]
Training Epoch: 6/100, batch 107/125 completed (loss: 0.1679:  87%|████████▋ | 54/62 [05:22<00:48,  6.04s/it]
Training Epoch: 6/100, batch 107/125 completed (loss: 0.2907:  87%|████████▋ | 54/62 [05:22<00:48,  6.04s/it]
Training Epoch: 6/100, batch 107/125 completed (loss: 0.029:  87%|████████▋ | 54/62 [05:22<00:48,  6.04s/it] 
Training Epoch: 6/100, batch 108/125 completed (loss: 0.2531:  87%|████████▋ | 54/62 [05:25<00:48,  6.04s/it]
Training Epoch: 6/100, batch 108/125 completed (loss: 0.1884:  87%|████████▋ | 54/62 [05:25<00:48,  6.04s/it]
Training Epoch: 6/100, batch 108/125 completed (loss: 0.0464:  87%|████████▋ | 54/62 [05:25<00:48,  6.04s/it]
Training Epoch: 6/100, batch 108/125 completed (loss: 0.0903:  87%|████████▋ | 54/62 [05:25<00:48,  6.04s/it]
Training Epoch: 6/100, batch 108/125 completed (loss: 0.0903:  89%|████████▊ | 55/62 [05:28<00:42,  6.03s/it]
Training Epoch: 6/100, batch 108/125 completed (loss: 0.0464:  89%|████████▊ | 55/62 [05:28<00:42,  6.03s/it]
Training Epoch: 6/100, batch 108/125 completed (loss: 0.2531:  89%|████████▊ | 55/62 [05:28<00:42,  6.04s/it]
Training Epoch: 6/100, batch 108/125 completed (loss: 0.1884:  89%|████████▊ | 55/62 [05:28<00:42,  6.03s/it]
Training Epoch: 6/100, batch 109/125 completed (loss: 0.211:  89%|████████▊ | 55/62 [05:28<00:42,  6.03s/it] 
Training Epoch: 6/100, batch 109/125 completed (loss: 0.1131:  89%|████████▊ | 55/62 [05:28<00:42,  6.04s/it]
Training Epoch: 6/100, batch 109/125 completed (loss: 0.0662:  89%|████████▊ | 55/62 [05:28<00:42,  6.03s/it]
Training Epoch: 6/100, batch 109/125 completed (loss: 0.2328:  89%|████████▊ | 55/62 [05:28<00:42,  6.03s/it]
Training Epoch: 6/100, batch 110/125 completed (loss: 0.1026:  89%|████████▊ | 55/62 [05:32<00:42,  6.03s/it]
Training Epoch: 6/100, batch 110/125 completed (loss: 0.1785:  89%|████████▊ | 55/62 [05:32<00:42,  6.03s/it]
Training Epoch: 6/100, batch 110/125 completed (loss: 0.0659:  89%|████████▊ | 55/62 [05:32<00:42,  6.04s/it]
Training Epoch: 6/100, batch 110/125 completed (loss: 0.1474:  89%|████████▊ | 55/62 [05:32<00:42,  6.03s/it]
Training Epoch: 6/100, batch 110/125 completed (loss: 0.1474:  90%|█████████ | 56/62 [05:34<00:36,  6.06s/it]
Training Epoch: 6/100, batch 110/125 completed (loss: 0.1785:  90%|█████████ | 56/62 [05:34<00:36,  6.05s/it]
Training Epoch: 6/100, batch 110/125 completed (loss: 0.0659:  90%|█████████ | 56/62 [05:34<00:36,  6.06s/it]
Training Epoch: 6/100, batch 110/125 completed (loss: 0.1026:  90%|█████████ | 56/62 [05:34<00:36,  6.06s/it]
Training Epoch: 6/100, batch 111/125 completed (loss: 0.2547:  90%|█████████ | 56/62 [05:34<00:36,  6.05s/it]
Training Epoch: 6/100, batch 111/125 completed (loss: 0.2769:  90%|█████████ | 56/62 [05:34<00:36,  6.06s/it]
Training Epoch: 6/100, batch 111/125 completed (loss: 0.1357:  90%|█████████ | 56/62 [05:34<00:36,  6.06s/it]
Training Epoch: 6/100, batch 111/125 completed (loss: 0.2139:  90%|█████████ | 56/62 [05:35<00:36,  6.06s/it]
Training Epoch: 6/100, batch 112/125 completed (loss: 0.065:  90%|█████████ | 56/62 [05:38<00:36,  6.05s/it] 
Training Epoch: 6/100, batch 112/125 completed (loss: 0.1187:  90%|█████████ | 56/62 [05:38<00:36,  6.06s/it]
Training Epoch: 6/100, batch 112/125 completed (loss: 0.2029:  90%|█████████ | 56/62 [05:38<00:36,  6.06s/it]
Training Epoch: 6/100, batch 112/125 completed (loss: 0.1271:  90%|█████████ | 56/62 [05:38<00:36,  6.06s/it]
Training Epoch: 6/100, batch 112/125 completed (loss: 0.1187:  92%|█████████▏| 57/62 [05:40<00:30,  6.04s/it]
Training Epoch: 6/100, batch 112/125 completed (loss: 0.065:  92%|█████████▏| 57/62 [05:40<00:30,  6.04s/it]
Training Epoch: 6/100, batch 112/125 completed (loss: 0.1271:  92%|█████████▏| 57/62 [05:40<00:30,  6.05s/it]
Training Epoch: 6/100, batch 112/125 completed (loss: 0.2029:  92%|█████████▏| 57/62 [05:40<00:30,  6.05s/it]
Training Epoch: 6/100, batch 113/125 completed (loss: 0.2966:  92%|█████████▏| 57/62 [05:41<00:30,  6.04s/it]
Training Epoch: 6/100, batch 113/125 completed (loss: 0.0742:  92%|█████████▏| 57/62 [05:41<00:30,  6.04s/it]
Training Epoch: 6/100, batch 113/125 completed (loss: 0.0744:  92%|█████████▏| 57/62 [05:41<00:30,  6.05s/it]
Training Epoch: 6/100, batch 113/125 completed (loss: 0.1305:  92%|█████████▏| 57/62 [05:41<00:30,  6.05s/it]
Training Epoch: 6/100, batch 114/125 completed (loss: 0.0862:  92%|█████████▏| 57/62 [05:43<00:30,  6.04s/it]
Training Epoch: 6/100, batch 114/125 completed (loss: 0.0468:  92%|█████████▏| 57/62 [05:43<00:30,  6.05s/it]
Training Epoch: 6/100, batch 114/125 completed (loss: 0.0406:  92%|█████████▏| 57/62 [05:43<00:30,  6.04s/it]
Training Epoch: 6/100, batch 114/125 completed (loss: 0.3436:  92%|█████████▏| 57/62 [05:43<00:30,  6.05s/it]
Training Epoch: 6/100, batch 114/125 completed (loss: 0.0406:  94%|█████████▎| 58/62 [05:47<00:24,  6.08s/it]
Training Epoch: 6/100, batch 114/125 completed (loss: 0.0862:  94%|█████████▎| 58/62 [05:47<00:24,  6.08s/it]
Training Epoch: 6/100, batch 114/125 completed (loss: 0.3436:  94%|█████████▎| 58/62 [05:47<00:24,  6.08s/it]
Training Epoch: 6/100, batch 114/125 completed (loss: 0.0468:  94%|█████████▎| 58/62 [05:47<00:24,  6.08s/it]
Training Epoch: 6/100, batch 115/125 completed (loss: 0.1513:  94%|█████████▎| 58/62 [05:47<00:24,  6.08s/it]
Training Epoch: 6/100, batch 115/125 completed (loss: 0.1879:  94%|█████████▎| 58/62 [05:47<00:24,  6.08s/it]
Training Epoch: 6/100, batch 115/125 completed (loss: 0.2008:  94%|█████████▎| 58/62 [05:47<00:24,  6.08s/it]
Training Epoch: 6/100, batch 115/125 completed (loss: 0.1701:  94%|█████████▎| 58/62 [05:47<00:24,  6.08s/it]
Training Epoch: 6/100, batch 116/125 completed (loss: 0.3104:  94%|█████████▎| 58/62 [05:50<00:24,  6.08s/it]
Training Epoch: 6/100, batch 116/125 completed (loss: 0.1007:  94%|█████████▎| 58/62 [05:50<00:24,  6.08s/it]
Training Epoch: 6/100, batch 116/125 completed (loss: 0.0771:  94%|█████████▎| 58/62 [05:50<00:24,  6.08s/it]
Training Epoch: 6/100, batch 116/125 completed (loss: 0.1794:  94%|█████████▎| 58/62 [05:50<00:24,  6.08s/it]
Training Epoch: 6/100, batch 116/125 completed (loss: 0.1794:  95%|█████████▌| 59/62 [05:53<00:18,  6.05s/it]
Training Epoch: 6/100, batch 116/125 completed (loss: 0.3104:  95%|█████████▌| 59/62 [05:53<00:18,  6.05s/it]
Training Epoch: 6/100, batch 116/125 completed (loss: 0.0771:  95%|█████████▌| 59/62 [05:53<00:18,  6.05s/it]
Training Epoch: 6/100, batch 116/125 completed (loss: 0.1007:  95%|█████████▌| 59/62 [05:53<00:18,  6.05s/it]
Training Epoch: 6/100, batch 117/125 completed (loss: 0.0436:  95%|█████████▌| 59/62 [05:53<00:18,  6.05s/it]
Training Epoch: 6/100, batch 117/125 completed (loss: 0.0828:  95%|█████████▌| 59/62 [05:53<00:18,  6.05s/it]
Training Epoch: 6/100, batch 117/125 completed (loss: 0.1197:  95%|█████████▌| 59/62 [05:53<00:18,  6.05s/it]
Training Epoch: 6/100, batch 117/125 completed (loss: 0.0891:  95%|█████████▌| 59/62 [05:53<00:18,  6.05s/it]
Training Epoch: 6/100, batch 118/125 completed (loss: 0.2046:  95%|█████████▌| 59/62 [05:56<00:18,  6.05s/it]
Training Epoch: 6/100, batch 118/125 completed (loss: 0.0774:  95%|█████████▌| 59/62 [05:56<00:18,  6.05s/it]
Training Epoch: 6/100, batch 118/125 completed (loss: 0.1565:  95%|█████████▌| 59/62 [05:56<00:18,  6.05s/it]
Training Epoch: 6/100, batch 118/125 completed (loss: 0.2313:  95%|█████████▌| 59/62 [05:56<00:18,  6.05s/it]
Training Epoch: 6/100, batch 118/125 completed (loss: 0.0774:  97%|█████████▋| 60/62 [05:59<00:12,  6.03s/it]
Training Epoch: 6/100, batch 118/125 completed (loss: 0.2313:  97%|█████████▋| 60/62 [05:59<00:12,  6.03s/it]
Training Epoch: 6/100, batch 118/125 completed (loss: 0.2046:  97%|█████████▋| 60/62 [05:59<00:12,  6.03s/it]
Training Epoch: 6/100, batch 118/125 completed (loss: 0.1565:  97%|█████████▋| 60/62 [05:59<00:12,  6.03s/it]
Training Epoch: 6/100, batch 119/125 completed (loss: 0.252:  97%|█████████▋| 60/62 [05:59<00:12,  6.03s/it] 
Training Epoch: 6/100, batch 119/125 completed (loss: 0.0602:  97%|█████████▋| 60/62 [05:59<00:12,  6.03s/it]
Training Epoch: 6/100, batch 119/125 completed (loss: 0.1812:  97%|█████████▋| 60/62 [05:59<00:12,  6.03s/it]
Training Epoch: 6/100, batch 119/125 completed (loss: 0.1868:  97%|█████████▋| 60/62 [05:59<00:12,  6.03s/it]
Training Epoch: 6/100, batch 120/125 completed (loss: 0.1501:  97%|█████████▋| 60/62 [06:02<00:12,  6.03s/it]
Training Epoch: 6/100, batch 120/125 completed (loss: 0.0675:  97%|█████████▋| 60/62 [06:02<00:12,  6.03s/it]
Training Epoch: 6/100, batch 120/125 completed (loss: 0.0625:  97%|█████████▋| 60/62 [06:02<00:12,  6.03s/it]
Training Epoch: 6/100, batch 120/125 completed (loss: 0.3689:  97%|█████████▋| 60/62 [06:02<00:12,  6.03s/it]
Training Epoch: 6/100, batch 120/125 completed (loss: 0.3689:  98%|█████████▊| 61/62 [06:05<00:06,  6.11s/it]
Training Epoch: 6/100, batch 120/125 completed (loss: 0.1501:  98%|█████████▊| 61/62 [06:05<00:06,  6.11s/it]
Training Epoch: 6/100, batch 120/125 completed (loss: 0.0675:  98%|█████████▊| 61/62 [06:05<00:06,  6.11s/it]
Training Epoch: 6/100, batch 120/125 completed (loss: 0.0625:  98%|█████████▊| 61/62 [06:05<00:06,  6.11s/it]
Training Epoch: 6/100, batch 121/125 completed (loss: 0.089:  98%|█████████▊| 61/62 [06:05<00:06,  6.11s/it] 
Training Epoch: 6/100, batch 121/125 completed (loss: 0.2896:  98%|█████████▊| 61/62 [06:05<00:06,  6.11s/it]
Training Epoch: 6/100, batch 121/125 completed (loss: 0.196:  98%|█████████▊| 61/62 [06:05<00:06,  6.11s/it] 
Training Epoch: 6/100, batch 121/125 completed (loss: 0.0917:  98%|█████████▊| 61/62 [06:05<00:06,  6.11s/it]
Training Epoch: 6/100, batch 122/125 completed (loss: 0.2919:  98%|█████████▊| 61/62 [06:08<00:06,  6.11s/it]
Training Epoch: 6/100, batch 122/125 completed (loss: 0.0937:  98%|█████████▊| 61/62 [06:08<00:06,  6.11s/it]
Training Epoch: 6/100, batch 122/125 completed (loss: 0.0871:  98%|█████████▊| 61/62 [06:08<00:06,  6.11s/it]
Training Epoch: 6/100, batch 122/125 completed (loss: 0.0353:  98%|█████████▊| 61/62 [06:08<00:06,  6.11s/it]
Training Epoch: 6/100, batch 122/125 completed (loss: 0.2919: 100%|██████████| 62/62 [06:11<00:00,  6.07s/it]
Training Epoch: 6/100, batch 122/125 completed (loss: 0.0353: 100%|██████████| 62/62 [06:11<00:00,  6.07s/it]
Training Epoch: 6/100, batch 122/125 completed (loss: 0.0871: 100%|██████████| 62/62 [06:11<00:00,  6.07s/it]
Training Epoch: 6/100, batch 122/125 completed (loss: 0.0937: 100%|██████████| 62/62 [06:11<00:00,  6.08s/it]
Training Epoch: 6/100, batch 123/125 completed (loss: 0.1774: 100%|██████████| 62/62 [06:11<00:00,  6.07s/it]
Training Epoch: 6/100, batch 123/125 completed (loss: 0.0716: 100%|██████████| 62/62 [06:11<00:00,  6.08s/it]
Training Epoch: 6/100, batch 123/125 completed (loss: 0.2512: 100%|██████████| 62/62 [06:11<00:00,  6.07s/it]
Training Epoch: 6/100, batch 123/125 completed (loss: 0.0397: 100%|██████████| 62/62 [06:11<00:00,  6.07s/it]
Training Epoch: 6/100, batch 123/125 completed (loss: 0.1774: : 63it [06:14,  5.12s/it]                      
Training Epoch: 6/100, batch 123/125 completed (loss: 0.2512: : 63it [06:14,  5.13s/it]                      
Training Epoch: 6/100, batch 123/125 completed (loss: 0.0397: : 63it [06:14,  5.13s/it]                      
Training Epoch: 6/100, batch 123/125 completed (loss: 0.0716: : 63it [06:14,  5.13s/it]                      
Training Epoch: 6/100, batch 124/125 completed (loss: 0.1425: : 63it [06:14,  5.13s/it]
Training Epoch: 6/100, batch 124/125 completed (loss: 0.0632: : 63it [06:14,  5.13s/it]
Training Epoch: 6/100, batch 124/125 completed (loss: 0.1898: : 63it [06:14,  5.12s/it]
Training Epoch: 6/100, batch 124/125 completed (loss: 0.0992: : 63it [06:14,  5.13s/it]
Training Epoch: 6/100, batch 124/125 completed (loss: 0.1898: : 63it [06:14,  5.94s/it]

Training Epoch: 6/100, batch 124/125 completed (loss: 0.0632: : 63it [06:14,  5.94s/it]

Training Epoch: 6/100, batch 124/125 completed (loss: 0.1425: : 63it [06:14,  5.94s/it]

Training Epoch: 6/100, batch 124/125 completed (loss: 0.0992: : 63it [06:14,  5.94s/it]
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/c10d_logger.py:83: UserWarning: barrier(): using the device under current context. You can specify `device_id` in `init_process_group` to mute this warning.
  return func(*args, **kwargs)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
saving model + train state.
Warning: failed to upload checkpoint_6 to HF: (Request ID: Root=1-6a1b6cf7-6c5902db136352cd72523753;41af5cfb-63e0-47cb-9840-1b32a6dc0581)

403 Forbidden: You need to setup automatic credit recharge in order to upload more data. You can do so at /settings/billing..
Cannot access content at: https://huggingface.co/seyedparsa/thoughtformer-checkpoints.git/info/lfs/objects/batch.
Make sure your token has the correct permissions.
Warning: failed to upload train_state_6 to HF: (Request ID: Root=1-6a1b6cfb-669d189b15d6554817fa9871;05d5713a-5fa1-4428-8aaa-fc6cbd04f12f)

403 Forbidden: You need to setup automatic credit recharge in order to upload more data. You can do so at /settings/billing..
Cannot access content at: https://huggingface.co/seyedparsa/thoughtformer-checkpoints.git/info/lfs/objects/batch.
Make sure your token has the correct permissions.
Deleted old checkpoint: train_state_5
Deleted old checkpoint: checkpoint_5
[warn] HF upload failed for ckpt_6; kept locally, earlier ckpts deleted.

Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]
Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]
Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]eval loss 0.08041772819482364

Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]
[WRONG] Example 0 (diff=3)

Test Accuracy:   3%|▎         | 1/32 [00:06<03:25,  6.62s/it]  Expected : 'Claude'
Test Accuracy:   3%|▎         | 1/32 [00:06<03:25,  6.61s/it]
  Extracted: 'Morgan'
  Generated: '### Morgan'

Test accuracy: 0.62:   3%|▎         | 1/32 [00:06<03:25,  6.62s/it]
Test accuracy: 0.75:   3%|▎         | 1/32 [00:06<03:25,  6.61s/it]
[WRONG] Example 1 (diff=8)
  Expected : 'Guy'
  Extracted: 'Irma'
  Generated: '### Irma'

Test Accuracy:   3%|▎         | 1/32 [00:06<03:25,  6.62s/it]
Test accuracy: 0.75:   3%|▎         | 1/32 [00:06<03:25,  6.62s/it]
[WRONG] Example 2 (diff=10)
  Expected : 'Elvis'
  Extracted: 'Shannon'
  Generated: '### Shannon'

[WRONG] Example 3 (diff=8)
  Expected : 'Coral'
  Extracted: 'Claude'
  Generated: '### Claude'

[CORRECT] Example 4 (diff=1)
  Expected : 'Patricia'
  Extracted: 'Patricia'
  Generated: '### Patricia'

Test Accuracy:   3%|▎         | 1/32 [00:06<03:20,  6.47s/it]
Test accuracy: 0.38:   3%|▎         | 1/32 [00:06<03:20,  6.47s/it]
Test accuracy: 0.62:   6%|▋         | 2/32 [00:13<03:15,  6.52s/it]
Test accuracy: 0.75:   6%|▋         | 2/32 [00:13<03:15,  6.51s/it]
Test accuracy: 0.75:   6%|▋         | 2/32 [00:13<03:15,  6.52s/it]
Test accuracy: 0.81:   6%|▋         | 2/32 [00:13<03:15,  6.51s/it]
Test accuracy: 0.75:   6%|▋         | 2/32 [00:13<03:15,  6.52s/it]
Test accuracy: 0.81:   6%|▋         | 2/32 [00:13<03:15,  6.52s/it]
Test accuracy: 0.38:   6%|▋         | 2/32 [00:12<03:13,  6.45s/it]
Test accuracy: 0.5:   6%|▋         | 2/32 [00:12<03:13,  6.45s/it] 
Test accuracy: 0.81:   9%|▉         | 3/32 [00:20<03:15,  6.73s/it]
Test accuracy: 0.75:   9%|▉         | 3/32 [00:20<03:15,  6.73s/it]
Test accuracy: 0.81:   9%|▉         | 3/32 [00:20<03:15,  6.73s/it]
Test accuracy: 0.79:   9%|▉         | 3/32 [00:20<03:15,  6.73s/it]
Test accuracy: 0.5:   9%|▉         | 3/32 [00:19<03:14,  6.69s/it]
Test accuracy: 0.54:   9%|▉         | 3/32 [00:19<03:14,  6.69s/it]
Test accuracy: 0.75:   9%|▉         | 3/32 [00:20<03:15,  6.74s/it]
Test accuracy: 0.58:   9%|▉         | 3/32 [00:20<03:15,  6.74s/it]
Test accuracy: 0.58:  12%|█▎        | 4/32 [00:26<03:02,  6.52s/it]
Test accuracy: 0.62:  12%|█▎        | 4/32 [00:26<03:02,  6.52s/it]
Test accuracy: 0.79:  12%|█▎        | 4/32 [00:26<03:02,  6.52s/it]
Test accuracy: 0.54:  12%|█▎        | 4/32 [00:26<03:01,  6.50s/it]
Test accuracy: 0.75:  12%|█▎        | 4/32 [00:26<03:02,  6.52s/it]
Test accuracy: 0.56:  12%|█▎        | 4/32 [00:26<03:01,  6.50s/it]
Test accuracy: 0.75:  12%|█▎        | 4/32 [00:26<03:02,  6.52s/it]
Test accuracy: 0.66:  12%|█▎        | 4/32 [00:26<03:02,  6.52s/it]
Test accuracy: 0.62:  16%|█▌        | 5/32 [00:32<02:54,  6.47s/it]
Test accuracy: 0.57:  16%|█▌        | 5/32 [00:32<02:54,  6.47s/it]
Test accuracy: 0.66:  16%|█▌        | 5/32 [00:32<02:54,  6.47s/it]
Test accuracy: 0.62:  16%|█▌        | 5/32 [00:32<02:54,  6.47s/it]
Test accuracy: 0.75:  16%|█▌        | 5/32 [00:32<02:54,  6.47s/it]
Test accuracy: 0.73:  16%|█▌        | 5/32 [00:32<02:54,  6.47s/it]
Test accuracy: 0.56:  16%|█▌        | 5/32 [00:32<02:54,  6.46s/it]
Test accuracy: 0.62:  16%|█▌        | 5/32 [00:32<02:54,  6.46s/it]
Test accuracy: 0.57:  19%|█▉        | 6/32 [00:38<02:46,  6.39s/it]
Test accuracy: 0.6:  19%|█▉        | 6/32 [00:38<02:46,  6.39s/it] 
Test accuracy: 0.62:  19%|█▉        | 6/32 [00:38<02:46,  6.39s/it]
Test accuracy: 0.67:  19%|█▉        | 6/32 [00:38<02:46,  6.39s/it]
Test accuracy: 0.73:  19%|█▉        | 6/32 [00:38<02:46,  6.39s/it]
Test accuracy: 0.69:  19%|█▉        | 6/32 [00:38<02:46,  6.39s/it]
Test accuracy: 0.62:  19%|█▉        | 6/32 [00:38<02:45,  6.38s/it]
Test accuracy: 0.69:  19%|█▉        | 6/32 [00:38<02:45,  6.38s/it]
Test accuracy: 0.67:  22%|██▏       | 7/32 [00:45<02:37,  6.32s/it]
Test accuracy: 0.62:  22%|██▏       | 7/32 [00:45<02:37,  6.32s/it]
Test accuracy: 0.69:  22%|██▏       | 7/32 [00:44<02:37,  6.31s/it]
Test accuracy: 0.69:  22%|██▏       | 7/32 [00:45<02:37,  6.32s/it]
Test accuracy: 0.68:  22%|██▏       | 7/32 [00:44<02:37,  6.31s/it]
Test accuracy: 0.66:  22%|██▏       | 7/32 [00:45<02:37,  6.32s/it]
Test accuracy: 0.6:  22%|██▏       | 7/32 [00:45<02:37,  6.32s/it]
Test accuracy: 0.62:  22%|██▏       | 7/32 [00:45<02:37,  6.32s/it]
Test accuracy: 0.62:  25%|██▌       | 8/32 [00:51<02:33,  6.39s/it]
Test accuracy: 0.62:  25%|██▌       | 8/32 [00:51<02:33,  6.39s/it]
Test accuracy: 0.64:  25%|██▌       | 8/32 [00:51<02:33,  6.39s/it]
Test accuracy: 0.61:  25%|██▌       | 8/32 [00:51<02:33,  6.39s/it]
Test accuracy: 0.68:  25%|██▌       | 8/32 [00:51<02:33,  6.39s/it]
Test accuracy: 0.66:  25%|██▌       | 8/32 [00:51<02:33,  6.39s/it]
Test accuracy: 0.67:  25%|██▌       | 8/32 [00:51<02:33,  6.39s/it]
Test accuracy: 0.66:  25%|██▌       | 8/32 [00:51<02:33,  6.39s/it]
Test accuracy: 0.64:  28%|██▊       | 9/32 [00:58<02:29,  6.50s/it]
Test accuracy: 0.64:  28%|██▊       | 9/32 [00:58<02:29,  6.50s/it]
Test accuracy: 0.67:  28%|██▊       | 9/32 [00:58<02:29,  6.50s/it]
Test accuracy: 0.68:  28%|██▊       | 9/32 [00:58<02:29,  6.50s/it]
Test accuracy: 0.66:  28%|██▊       | 9/32 [00:58<02:29,  6.50s/it]
Test accuracy: 0.65:  28%|██▊       | 9/32 [00:58<02:29,  6.50s/it]
Test accuracy: 0.61:  28%|██▊       | 9/32 [00:58<02:29,  6.51s/it]
Test accuracy: 0.62:  28%|██▊       | 9/32 [00:58<02:29,  6.51s/it]
Test accuracy: 0.62:  31%|███▏      | 10/32 [01:04<02:21,  6.41s/it]
Test accuracy: 0.62:  31%|███▏      | 10/32 [01:04<02:21,  6.41s/it]
Test accuracy: 0.64:  31%|███▏      | 10/32 [01:04<02:21,  6.41s/it]
Test accuracy: 0.64:  31%|███▏      | 10/32 [01:04<02:21,  6.41s/it]
Test accuracy: 0.68:  31%|███▏      | 10/32 [01:04<02:21,  6.41s/it]
Test accuracy: 0.69:  31%|███▏      | 10/32 [01:04<02:21,  6.41s/it]
Test accuracy: 0.65:  31%|███▏      | 10/32 [01:04<02:21,  6.41s/it]
Test accuracy: 0.62:  31%|███▏      | 10/32 [01:04<02:21,  6.41s/it]
Test accuracy: 0.64:  34%|███▍      | 11/32 [01:11<02:15,  6.46s/it]
Test accuracy: 0.64:  34%|███▍      | 11/32 [01:11<02:15,  6.46s/it]
Test accuracy: 0.62:  34%|███▍      | 11/32 [01:11<02:15,  6.46s/it]
Test accuracy: 0.62:  34%|███▍      | 11/32 [01:10<02:15,  6.46s/it]
Test accuracy: 0.59:  34%|███▍      | 11/32 [01:11<02:15,  6.46s/it]
Test accuracy: 0.61:  34%|███▍      | 11/32 [01:10<02:15,  6.46s/it]
Test accuracy: 0.69:  34%|███▍      | 11/32 [01:11<02:15,  6.46s/it]
Test accuracy: 0.67:  34%|███▍      | 11/32 [01:11<02:15,  6.46s/it]
Test accuracy: 0.64:  38%|███▊      | 12/32 [01:17<02:07,  6.37s/it]
Test accuracy: 0.61:  38%|███▊      | 12/32 [01:17<02:07,  6.37s/it]
Test accuracy: 0.64:  38%|███▊      | 12/32 [01:17<02:07,  6.37s/it]
Test accuracy: 0.6:  38%|███▊      | 12/32 [01:17<02:07,  6.37s/it] 
Test accuracy: 0.59:  38%|███▊      | 12/32 [01:17<02:07,  6.37s/it]
Test accuracy: 0.58:  38%|███▊      | 12/32 [01:17<02:07,  6.37s/it]
Test accuracy: 0.67:  38%|███▊      | 12/32 [01:17<02:07,  6.37s/it]
Test accuracy: 0.65:  38%|███▊      | 12/32 [01:17<02:07,  6.37s/it]
Test accuracy: 0.64:  41%|████      | 13/32 [01:23<01:59,  6.32s/it]
Test accuracy: 0.63:  41%|████      | 13/32 [01:23<01:59,  6.32s/it]
Test accuracy: 0.58:  41%|████      | 13/32 [01:23<01:59,  6.31s/it]
Test accuracy: 0.58:  41%|████      | 13/32 [01:23<01:59,  6.31s/it]
Test accuracy: 0.65:  41%|████      | 13/32 [01:23<01:59,  6.32s/it]
Test accuracy: 0.6:  41%|████      | 13/32 [01:23<01:59,  6.31s/it]
Test accuracy: 0.66:  41%|████      | 13/32 [01:23<01:59,  6.32s/it]
Test accuracy: 0.59:  41%|████      | 13/32 [01:23<01:59,  6.31s/it]
Test accuracy: 0.63:  44%|████▍     | 14/32 [01:29<01:54,  6.34s/it]
Test accuracy: 0.63:  44%|████▍     | 14/32 [01:29<01:54,  6.34s/it]
Test accuracy: 0.66:  44%|████▍     | 14/32 [01:29<01:54,  6.34s/it]
Test accuracy: 0.67:  44%|████▍     | 14/32 [01:29<01:54,  6.34s/it]
Test accuracy: 0.59:  44%|████▍     | 14/32 [01:29<01:54,  6.34s/it]
Test accuracy: 0.58:  44%|████▍     | 14/32 [01:29<01:54,  6.34s/it]
Test accuracy: 0.59:  44%|████▍     | 14/32 [01:29<01:54,  6.34s/it]
Test accuracy: 0.57:  44%|████▍     | 14/32 [01:29<01:54,  6.34s/it]
Test accuracy: 0.63:  47%|████▋     | 15/32 [01:36<01:46,  6.29s/it]
Test accuracy: 0.63:  47%|████▋     | 15/32 [01:36<01:46,  6.29s/it]
Test accuracy: 0.67:  47%|████▋     | 15/32 [01:36<01:46,  6.29s/it]
Test accuracy: 0.57:  47%|████▋     | 15/32 [01:36<01:46,  6.29s/it]
Test accuracy: 0.67:  47%|████▋     | 15/32 [01:36<01:46,  6.29s/it]
Test accuracy: 0.57:  47%|████▋     | 15/32 [01:36<01:46,  6.29s/it]
Test accuracy: 0.59:  47%|████▋     | 15/32 [01:35<01:46,  6.29s/it]
Test accuracy: 0.57:  47%|████▋     | 15/32 [01:35<01:46,  6.29s/it]
Test accuracy: 0.67:  50%|█████     | 16/32 [01:42<01:41,  6.32s/it]
Test accuracy: 0.63:  50%|█████     | 16/32 [01:42<01:41,  6.32s/it]
Test accuracy: 0.67:  50%|█████     | 16/32 [01:42<01:41,  6.32s/it]
Test accuracy: 0.62:  50%|█████     | 16/32 [01:42<01:41,  6.32s/it]
Test accuracy: 0.57:  50%|█████     | 16/32 [01:42<01:41,  6.32s/it]
Test accuracy: 0.56:  50%|█████     | 16/32 [01:42<01:41,  6.32s/it]
Test accuracy: 0.57:  50%|█████     | 16/32 [01:42<01:41,  6.32s/it]
Test accuracy: 0.57:  50%|█████     | 16/32 [01:42<01:41,  6.32s/it]
Test accuracy: 0.62:  53%|█████▎    | 17/32 [01:48<01:35,  6.34s/it]
Test accuracy: 0.61:  53%|█████▎    | 17/32 [01:48<01:35,  6.34s/it]
Test accuracy: 0.57:  53%|█████▎    | 17/32 [01:48<01:35,  6.34s/it]
Test accuracy: 0.56:  53%|█████▎    | 17/32 [01:48<01:35,  6.34s/it]
Test accuracy: 0.59:  53%|█████▎    | 17/32 [01:48<01:35,  6.34s/it]
Test accuracy: 0.55:  53%|█████▎    | 17/32 [01:48<01:35,  6.34s/it]
Test accuracy: 0.67:  53%|█████▎    | 17/32 [01:48<01:35,  6.34s/it]
Test accuracy: 0.68:  53%|█████▎    | 17/32 [01:48<01:35,  6.34s/it]
Test accuracy: 0.61:  56%|█████▋    | 18/32 [01:54<01:28,  6.29s/it]
Test accuracy: 0.59:  56%|█████▋    | 18/32 [01:54<01:28,  6.29s/it]
Test accuracy: 0.61:  56%|█████▋    | 18/32 [01:54<01:28,  6.29s/it]
Test accuracy: 0.59:  56%|█████▋    | 18/32 [01:54<01:28,  6.29s/it]
Test accuracy: 0.55:  56%|█████▋    | 18/32 [01:54<01:28,  6.29s/it]
Test accuracy: 0.68:  56%|█████▋    | 18/32 [01:54<01:28,  6.29s/it]
Test accuracy: 0.56:  56%|█████▋    | 18/32 [01:54<01:28,  6.29s/it]
Test accuracy: 0.67:  56%|█████▋    | 18/32 [01:54<01:28,  6.29s/it]
Test accuracy: 0.61:  59%|█████▉    | 19/32 [02:01<01:22,  6.31s/it]
Test accuracy: 0.67:  59%|█████▉    | 19/32 [02:01<01:22,  6.31s/it]
Test accuracy: 0.6:  59%|█████▉    | 19/32 [02:01<01:22,  6.31s/it] 
Test accuracy: 0.59:  59%|█████▉    | 19/32 [02:01<01:22,  6.31s/it]
Test accuracy: 0.65:  59%|█████▉    | 19/32 [02:01<01:22,  6.31s/it]
Test accuracy: 0.61:  59%|█████▉    | 19/32 [02:01<01:22,  6.31s/it]
Test accuracy: 0.56:  59%|█████▉    | 19/32 [02:01<01:22,  6.32s/it]
Test accuracy: 0.57:  59%|█████▉    | 19/32 [02:01<01:22,  6.32s/it]
Test accuracy: 0.6:  62%|██████▎   | 20/32 [02:07<01:15,  6.28s/it]
Test accuracy: 0.65:  62%|██████▎   | 20/32 [02:07<01:15,  6.28s/it]
Test accuracy: 0.59:  62%|██████▎   | 20/32 [02:07<01:15,  6.28s/it]
Test accuracy: 0.64:  62%|██████▎   | 20/32 [02:07<01:15,  6.28s/it]
Test accuracy: 0.57:  62%|██████▎   | 20/32 [02:07<01:15,  6.28s/it]
Test accuracy: 0.57:  62%|██████▎   | 20/32 [02:07<01:15,  6.28s/it]
Test accuracy: 0.61:  62%|██████▎   | 20/32 [02:07<01:15,  6.28s/it]
Test accuracy: 0.61:  62%|██████▎   | 20/32 [02:07<01:15,  6.28s/it]
Test accuracy: 0.57:  66%|██████▌   | 21/32 [02:14<01:10,  6.44s/it]
Test accuracy: 0.61:  66%|██████▌   | 21/32 [02:14<01:10,  6.44s/it]
Test accuracy: 0.64:  66%|██████▌   | 21/32 [02:14<01:10,  6.44s/it]
Test accuracy: 0.58:  66%|██████▌   | 21/32 [02:14<01:10,  6.44s/it]
Test accuracy: 0.62:  66%|██████▌   | 21/32 [02:14<01:10,  6.44s/it]
Test accuracy: 0.61:  66%|██████▌   | 21/32 [02:14<01:10,  6.44s/it]
Test accuracy: 0.59:  66%|██████▌   | 21/32 [02:14<01:10,  6.44s/it]
Test accuracy: 0.6:  66%|██████▌   | 21/32 [02:14<01:10,  6.44s/it] 
Test accuracy: 0.6:  69%|██████▉   | 22/32 [02:20<01:04,  6.47s/it]
Test accuracy: 0.6:  69%|██████▉   | 22/32 [02:20<01:04,  6.47s/it]
Test accuracy: 0.61:  69%|██████▉   | 22/32 [02:20<01:04,  6.48s/it]
Test accuracy: 0.58:  69%|██████▉   | 22/32 [02:20<01:04,  6.48s/it]
Test accuracy: 0.61:  69%|██████▉   | 22/32 [02:20<01:04,  6.48s/it]
Test accuracy: 0.62:  69%|██████▉   | 22/32 [02:20<01:04,  6.48s/it]
Test accuracy: 0.58:  69%|██████▉   | 22/32 [02:20<01:04,  6.48s/it]
Test accuracy: 0.63:  69%|██████▉   | 22/32 [02:20<01:04,  6.48s/it]
Test accuracy: 0.6:  72%|███████▏  | 23/32 [02:27<00:57,  6.39s/it]
Test accuracy: 0.58:  72%|███████▏  | 23/32 [02:27<00:57,  6.39s/it]
Test accuracy: 0.58:  72%|███████▏  | 23/32 [02:27<00:57,  6.39s/it]
Test accuracy: 0.59:  72%|███████▏  | 23/32 [02:27<00:57,  6.39s/it]
Test accuracy: 0.61:  72%|███████▏  | 23/32 [02:26<00:57,  6.39s/it]
Test accuracy: 0.6:  72%|███████▏  | 23/32 [02:26<00:57,  6.39s/it] 
Test accuracy: 0.63:  72%|███████▏  | 23/32 [02:27<00:57,  6.39s/it]
Test accuracy: 0.62:  72%|███████▏  | 23/32 [02:27<00:57,  6.39s/it]
Test accuracy: 0.58:  75%|███████▌  | 24/32 [02:34<00:52,  6.56s/it]
Test accuracy: 0.59:  75%|███████▌  | 24/32 [02:34<00:52,  6.56s/it]
Test accuracy: 0.62:  75%|███████▌  | 24/32 [02:34<00:52,  6.56s/it]
Test accuracy: 0.58:  75%|███████▌  | 24/32 [02:34<00:52,  6.56s/it]
Test accuracy: 0.59:  75%|███████▌  | 24/32 [02:34<00:52,  6.56s/it]
Test accuracy: 0.62:  75%|███████▌  | 24/32 [02:34<00:52,  6.56s/it]
Test accuracy: 0.6:  75%|███████▌  | 24/32 [02:33<00:52,  6.57s/it]
Test accuracy: 0.6:  75%|███████▌  | 24/32 [02:33<00:52,  6.57s/it]
Test accuracy: 0.6:  78%|███████▊  | 25/32 [02:40<00:45,  6.51s/it]
Test accuracy: 0.61:  78%|███████▊  | 25/32 [02:40<00:45,  6.51s/it]
Test accuracy: 0.62:  78%|███████▊  | 25/32 [02:40<00:45,  6.51s/it]
Test accuracy: 0.61:  78%|███████▊  | 25/32 [02:40<00:45,  6.51s/it]
Test accuracy: 0.58:  78%|███████▊  | 25/32 [02:40<00:45,  6.51s/it]
Test accuracy: 0.59:  78%|███████▊  | 25/32 [02:40<00:45,  6.51s/it]
Test accuracy: 0.57:  78%|███████▊  | 25/32 [02:40<00:45,  6.51s/it]
Test accuracy: 0.6:  78%|███████▊  | 25/32 [02:40<00:45,  6.51s/it] 
Test accuracy: 0.57:  81%|████████▏ | 26/32 [02:46<00:38,  6.41s/it]
Test accuracy: 0.61:  81%|████████▏ | 26/32 [02:46<00:38,  6.41s/it]
Test accuracy: 0.6:  81%|████████▏ | 26/32 [02:46<00:38,  6.41s/it]
Test accuracy: 0.58:  81%|████████▏ | 26/32 [02:46<00:38,  6.41s/it]
Test accuracy: 0.61:  81%|████████▏ | 26/32 [02:46<00:38,  6.41s/it]
Test accuracy: 0.58:  81%|████████▏ | 26/32 [02:46<00:38,  6.41s/it]
Test accuracy: 0.61:  81%|████████▏ | 26/32 [02:46<00:38,  6.41s/it]
Test accuracy: 0.61:  81%|████████▏ | 26/32 [02:46<00:38,  6.41s/it]
Test accuracy: 0.58:  84%|████████▍ | 27/32 [02:52<00:31,  6.40s/it]
Test accuracy: 0.58:  84%|████████▍ | 27/32 [02:52<00:31,  6.40s/it]
Test accuracy: 0.59:  84%|████████▍ | 27/32 [02:52<00:31,  6.40s/it]
Test accuracy: 0.61:  84%|████████▍ | 27/32 [02:52<00:31,  6.40s/it]
Test accuracy: 0.57:  84%|████████▍ | 27/32 [02:53<00:31,  6.40s/it]
Test accuracy: 0.61:  84%|████████▍ | 27/32 [02:52<00:31,  6.40s/it]
Test accuracy: 0.61:  84%|████████▍ | 27/32 [02:52<00:31,  6.40s/it]
Test accuracy: 0.6:  84%|████████▍ | 27/32 [02:52<00:31,  6.40s/it] 
Test accuracy: 0.59:  88%|████████▊ | 28/32 [02:59<00:26,  6.57s/it]
Test accuracy: 0.61:  88%|████████▊ | 28/32 [02:59<00:26,  6.57s/it]
Test accuracy: 0.6:  88%|████████▊ | 28/32 [02:59<00:26,  6.57s/it]
Test accuracy: 0.6:  88%|████████▊ | 28/32 [02:59<00:26,  6.57s/it] 
Test accuracy: 0.6:  88%|████████▊ | 28/32 [02:59<00:26,  6.57s/it]
Test accuracy: 0.62:  88%|████████▊ | 28/32 [02:59<00:26,  6.57s/it]
Test accuracy: 0.57:  88%|████████▊ | 28/32 [02:59<00:26,  6.57s/it]
Test accuracy: 0.57:  88%|████████▊ | 28/32 [02:59<00:26,  6.57s/it]
Test accuracy: 0.6:  91%|█████████ | 29/32 [03:06<00:19,  6.46s/it]
Test accuracy: 0.57:  91%|█████████ | 29/32 [03:06<00:19,  6.46s/it]
Test accuracy: 0.59:  91%|█████████ | 29/32 [03:06<00:19,  6.46s/it]
Test accuracy: 0.56:  91%|█████████ | 29/32 [03:06<00:19,  6.46s/it]
Test accuracy: 0.62:  91%|█████████ | 29/32 [03:06<00:19,  6.46s/it]
Test accuracy: 0.6:  91%|█████████ | 29/32 [03:06<00:19,  6.46s/it]
Test accuracy: 0.62:  91%|█████████ | 29/32 [03:06<00:19,  6.46s/it]
Test accuracy: 0.6:  91%|█████████ | 29/32 [03:06<00:19,  6.46s/it]
Test accuracy: 0.56:  94%|█████████▍| 30/32 [03:12<00:12,  6.43s/it]
Test accuracy: 0.59:  94%|█████████▍| 30/32 [03:12<00:12,  6.43s/it]
Test accuracy: 0.56:  94%|█████████▍| 30/32 [03:12<00:12,  6.43s/it]
Test accuracy: 0.59:  94%|█████████▍| 30/32 [03:12<00:12,  6.43s/it]
Test accuracy: 0.62:  94%|█████████▍| 30/32 [03:12<00:12,  6.43s/it]
Test accuracy: 0.61:  94%|█████████▍| 30/32 [03:12<00:12,  6.43s/it]
Test accuracy: 0.6:  94%|█████████▍| 30/32 [03:12<00:12,  6.43s/it]
Test accuracy: 0.6:  94%|█████████▍| 30/32 [03:12<00:12,  6.43s/it]
Test accuracy: 0.59:  97%|█████████▋| 31/32 [03:19<00:06,  6.60s/it]
Test accuracy: 0.6:  97%|█████████▋| 31/32 [03:19<00:06,  6.60s/it]
Test accuracy: 0.61:  97%|█████████▋| 31/32 [03:19<00:06,  6.60s/it]
Test accuracy: 0.6:  97%|█████████▋| 31/32 [03:19<00:06,  6.60s/it] 
Test accuracy: 0.6:  97%|█████████▋| 31/32 [03:19<00:06,  6.60s/it]
Test accuracy: 0.62:  97%|█████████▋| 31/32 [03:19<00:06,  6.60s/it]
Test accuracy: 0.56:  97%|█████████▋| 31/32 [03:19<00:06,  6.60s/it]
Test accuracy: 0.56:  97%|█████████▋| 31/32 [03:19<00:06,  6.60s/it]
Test accuracy: 0.62: 100%|██████████| 32/32 [03:21<00:00,  5.32s/it]
Test accuracy: 0.62: 100%|██████████| 32/32 [03:21<00:00,  5.32s/it]
Test accuracy: 0.56: 100%|██████████| 32/32 [03:21<00:00,  5.32s/it]
Test accuracy: 0.6: 100%|██████████| 32/32 [03:21<00:00,  5.33s/it]
Test accuracy: 0.56: 100%|██████████| 32/32 [03:21<00:00,  5.32s/it]
Test accuracy: 0.6: 100%|██████████| 32/32 [03:21<00:00,  5.33s/it]
Test accuracy: 0.6: 100%|██████████| 32/32 [03:21<00:00,  5.33s/it]
Test accuracy: 0.6: 100%|██████████| 32/32 [03:21<00:00,  5.33s/it]
Test accuracy: 0.56: 100%|██████████| 32/32 [03:21<00:00,  6.31s/it]
Device 2: Cor=141, CoT=0, Total=250

Test accuracy: 0.6: 100%|██████████| 32/32 [03:21<00:00,  6.31s/it]
Device 1: Cor=151, CoT=0, Total=250

Test accuracy: 0.6: 100%|██████████| 32/32 [03:21<00:00,  6.31s/it]
Device 3: Cor=150, CoT=0, Total=250

Test accuracy: 0.62: 100%|██████████| 32/32 [03:21<00:00,  6.31s/it]
Device 0: Cor=154, CoT=0, Total=250

============================================================
EVALUATION RESULTS (Epoch 5)
============================================================
Config: data_curriculum=ON, thought_curriculum=OFF
thought_stage=10, data_stage=2, n_latent_tokens=10
Training on difficulties: 0-2

Overall accuracy: 596 / 1000 = 0.5960

Per-difficulty accuracy:
  diff=1: 97/100 = 0.9700 (trained)
  diff=2: 91/100 = 0.9100 (trained)
  diff=3: 54/100 = 0.5400 (untrained)
  diff=4: 51/100 = 0.5100 (untrained)
  diff=5: 44/100 = 0.4400 (untrained)
  diff=6: 53/100 = 0.5300 (untrained)
  diff=7: 54/100 = 0.5400 (untrained)
  diff=8: 43/100 = 0.4300 (untrained)
  diff=9: 56/100 = 0.5600 (untrained)
  diff=10: 53/100 = 0.5300 (untrained)

Trained (diff 0-2): 188/200 = 0.9400
============================================================


*** ADAPTIVE STAGING: acc_trained=0.9400, acc_diff2=0.9100 >= 0.9 ***
*** Advancing to stage 3 ***

Warning: failed to upload checkpoint_6 to HF: (Request ID: Root=1-6a1b6dd2-12fe65891c22ba5a6e89a8df;0124666f-2597-456a-b4f7-a8f2bc9c1025)

403 Forbidden: You need to setup automatic credit recharge in order to upload more data. You can do so at /settings/billing..
Cannot access content at: https://huggingface.co/seyedparsa/thoughtformer-checkpoints.git/info/lfs/objects/batch.
Make sure your token has the correct permissions.

============================================================
EPOCH 6/100 (thought_stage=10, data_stage=3)
============================================================
  thought_stage=10, c_thought=1, max_difficulty=3

Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):  16%|█▌        | 156/1000 [00:00<00:00, 988.43 examples/s]
Map (num_proc=4):  16%|█▌        | 156/1000 [00:00<00:00, 944.44 examples/s]
Map (num_proc=4):  16%|█▌        | 158/1000 [00:00<00:00, 959.66 examples/s]
Map (num_proc=4):  82%|████████▏ | 816/1000 [00:00<00:00, 3477.04 examples/s]
Map (num_proc=4):  81%|████████  | 811/1000 [00:00<00:00, 3339.60 examples/s]
Map (num_proc=4):  81%|████████▏ | 814/1000 [00:00<00:00, 3402.30 examples/s]
Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2470.48 examples/s]

Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2014.30 examples/s]

Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2036.52 examples/s]

Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Map (num_proc=4):  16%|█▌        | 158/1000 [00:00<00:01, 501.97 examples/s]
Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Map (num_proc=4):  82%|████████▏ | 819/1000 [00:00<00:00, 2183.01 examples/s]
Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:04, 2027.90 examples/s]
Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 1217.77 examples/s]

Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:04, 2025.23 examples/s]
Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:04, 1910.72 examples/s]
Filter (num_proc=4):  30%|███       | 3000/10000 [00:00<00:01, 5035.87 examples/s]
Filter (num_proc=4):  20%|██        | 2000/10000 [00:00<00:02, 3752.33 examples/s]
Filter (num_proc=4):  20%|██        | 2000/10000 [00:00<00:02, 3394.03 examples/s]
Filter (num_proc=4):  60%|██████    | 6000/10000 [00:00<00:00, 9325.49 examples/s]
Filter (num_proc=4):  40%|████      | 4000/10000 [00:00<00:00, 7170.28 examples/s]
Filter (num_proc=4):  40%|████      | 4000/10000 [00:00<00:00, 6231.79 examples/s]
Filter (num_proc=4):  75%|███████▌  | 7500/10000 [00:00<00:00, 10071.30 examples/s]
Filter (num_proc=4):  60%|██████    | 6000/10000 [00:00<00:00, 10113.17 examples/s]
Filter (num_proc=4):  75%|███████▌  | 7500/10000 [00:00<00:00, 11277.74 examples/s]
Filter (num_proc=4):  60%|██████    | 6000/10000 [00:01<00:00, 7690.79 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 11286.72 examples/s]
Filter (num_proc=4):  95%|█████████▌| 9500/10000 [00:01<00:00, 12910.03 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 7837.64 examples/s] 

Filter (num_proc=4):  85%|████████▌ | 8500/10000 [00:01<00:00, 10140.75 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 7728.17 examples/s]

Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 8623.99 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 6462.84 examples/s]

Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:06, 1376.89 examples/s]
Filter (num_proc=4):  20%|██        | 2000/10000 [00:00<00:03, 2489.44 examples/s]
Filter (num_proc=4):  30%|███       | 3000/10000 [00:01<00:02, 3476.59 examples/s]
Filter (num_proc=4):  45%|████▌     | 4500/10000 [00:01<00:01, 5448.29 examples/s]
Filter (num_proc=4):  70%|███████   | 7000/10000 [00:01<00:00, 7661.20 examples/s]
Filter (num_proc=4):  85%|████████▌ | 8500/10000 [00:01<00:00, 7984.92 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 8201.94 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:02<00:00, 4954.68 examples/s]

Map (num_proc=4):   0%|          | 0/3000 [00:00<?, ? examples/s]
Map (num_proc=4):   5%|▍         | 137/3000 [00:01<00:23, 121.85 examples/s]
Map (num_proc=4):  18%|█▊        | 548/3000 [00:01<00:04, 541.11 examples/s]
Map (num_proc=4):  25%|██▌       | 750/3000 [00:01<00:03, 589.51 examples/s]
Map (num_proc=4):  30%|██▉       | 886/3000 [00:01<00:03, 555.46 examples/s]
Map (num_proc=4):  43%|████▎     | 1297/3000 [00:01<00:01, 968.68 examples/s]
Map (num_proc=4):  50%|█████     | 1500/3000 [00:02<00:01, 880.20 examples/s]
Map (num_proc=4):  55%|█████▍    | 1637/3000 [00:02<00:01, 747.51 examples/s]
Map (num_proc=4):  68%|██████▊   | 2052/3000 [00:02<00:00, 1142.35 examples/s]
Map (num_proc=4):  75%|███████▌  | 2250/3000 [00:03<00:00, 986.15 examples/s] 
Map (num_proc=4):  80%|███████▉  | 2386/3000 [00:03<00:00, 824.96 examples/s]
Map (num_proc=4):  93%|█████████▎| 2803/3000 [00:03<00:00, 1217.80 examples/s]
Map (num_proc=4): 100%|██████████| 3000/3000 [00:03<00:00, 1040.15 examples/s]
Map (num_proc=4): 100%|██████████| 3000/3000 [00:04<00:00, 719.71 examples/s] 

Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 1472.38 examples/s]
Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 1444.58 examples/s]
Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 1383.87 examples/s]
Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2879.26 examples/s]

Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2803.68 examples/s]

Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2730.72 examples/s]

Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 761.15 examples/s]
Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 1454.91 examples/s]

Map (num_proc=4):   0%|          | 0/300 [00:00<?, ? examples/s]
Map (num_proc=4):  25%|██▌       | 75/300 [00:00<00:01, 214.03 examples/s]
Map (num_proc=4):  75%|███████▌  | 225/300 [00:00<00:00, 492.31 examples/s]
Map (num_proc=4): 100%|██████████| 300/300 [00:00<00:00, 334.43 examples/s]

Training Epoch: 7:   0%|          | 0/94 [00:00<?, ?it/s]
Training Epoch: 7:   0%|          | 0/94 [00:00<?, ?it/s]
Training Epoch: 7:   0%|          | 0/94 [00:00<?, ?it/s]
Training Epoch: 7:   0%|          | 0/94 [00:00<?, ?it/s]logging training data
rank 2 | gpu 2 | alloc=4.20 GB | reserved=47.74 GB | max=41.98 GBrank 3 | gpu 3 | alloc=4.20 GB | reserved=47.74 GB | max=41.99 GB

rank 1 | gpu 1 | alloc=4.20 GB | reserved=47.73 GB | max=41.98 GB
rank 0 | gpu 0 | alloc=4.20 GB | reserved=53.38 GB | max=41.99 GB

Training Epoch: 7/100, batch 0/188 completed (loss: 0.0413:   0%|          | 0/94 [00:03<?, ?it/s]
Training Epoch: 7/100, batch 0/188 completed (loss: 0.5091:   0%|          | 0/94 [00:03<?, ?it/s]
Training Epoch: 7/100, batch 0/188 completed (loss: 0.1456:   0%|          | 0/94 [00:03<?, ?it/s]
Training Epoch: 7/100, batch 0/188 completed (loss: 0.212:   0%|          | 0/94 [00:03<?, ?it/s]
Training Epoch: 7/100, batch 0/188 completed (loss: 0.212:   1%|          | 1/94 [00:06<09:18,  6.00s/it]
Training Epoch: 7/100, batch 0/188 completed (loss: 0.1456:   1%|          | 1/94 [00:05<09:17,  6.00s/it]
Training Epoch: 7/100, batch 0/188 completed (loss: 0.5091:   1%|          | 1/94 [00:05<09:17,  6.00s/it]
Training Epoch: 7/100, batch 0/188 completed (loss: 0.0413:   1%|          | 1/94 [00:05<09:17,  6.00s/it]
Training Epoch: 7/100, batch 1/188 completed (loss: 0.2253:   1%|          | 1/94 [00:06<09:17,  6.00s/it]
Training Epoch: 7/100, batch 1/188 completed (loss: 0.1738:   1%|          | 1/94 [00:06<09:17,  6.00s/it]
Training Epoch: 7/100, batch 1/188 completed (loss: 0.0666:   1%|          | 1/94 [00:06<09:17,  6.00s/it]
Training Epoch: 7/100, batch 1/188 completed (loss: 0.2752:   1%|          | 1/94 [00:06<09:18,  6.00s/it]
Training Epoch: 7/100, batch 2/188 completed (loss: 0.0264:   1%|          | 1/94 [00:09<09:17,  6.00s/it]
Training Epoch: 7/100, batch 2/188 completed (loss: 0.371:   1%|          | 1/94 [00:09<09:17,  6.00s/it] 
Training Epoch: 7/100, batch 2/188 completed (loss: 0.081:   1%|          | 1/94 [00:09<09:17,  6.00s/it] 
Training Epoch: 7/100, batch 2/188 completed (loss: 0.3055:   1%|          | 1/94 [00:09<09:18,  6.00s/it]
Training Epoch: 7/100, batch 2/188 completed (loss: 0.0264:   2%|▏         | 2/94 [00:11<09:10,  5.99s/it]
Training Epoch: 7/100, batch 2/188 completed (loss: 0.371:   2%|▏         | 2/94 [00:11<09:10,  5.99s/it]
Training Epoch: 7/100, batch 2/188 completed (loss: 0.3055:   2%|▏         | 2/94 [00:11<09:11,  6.00s/it]
Training Epoch: 7/100, batch 2/188 completed (loss: 0.081:   2%|▏         | 2/94 [00:11<09:11,  5.99s/it]
Training Epoch: 7/100, batch 3/188 completed (loss: 0.0798:   2%|▏         | 2/94 [00:12<09:11,  5.99s/it]
Training Epoch: 7/100, batch 3/188 completed (loss: 0.1957:   2%|▏         | 2/94 [00:12<09:10,  5.99s/it]
Training Epoch: 7/100, batch 3/188 completed (loss: 0.1802:   2%|▏         | 2/94 [00:12<09:10,  5.99s/it]
Training Epoch: 7/100, batch 3/188 completed (loss: 0.4461:   2%|▏         | 2/94 [00:12<09:11,  6.00s/it]
Training Epoch: 7/100, batch 4/188 completed (loss: 0.2067:   2%|▏         | 2/94 [00:15<09:10,  5.99s/it]
Training Epoch: 7/100, batch 4/188 completed (loss: 0.3064:   2%|▏         | 2/94 [00:15<09:10,  5.99s/it]
Training Epoch: 7/100, batch 4/188 completed (loss: 0.0806:   2%|▏         | 2/94 [00:15<09:11,  5.99s/it]
Training Epoch: 7/100, batch 4/188 completed (loss: 0.0415:   2%|▏         | 2/94 [00:15<09:11,  6.00s/it]
Training Epoch: 7/100, batch 4/188 completed (loss: 0.0415:   3%|▎         | 3/94 [00:17<08:59,  5.93s/it]
Training Epoch: 7/100, batch 4/188 completed (loss: 0.0806:   3%|▎         | 3/94 [00:17<08:59,  5.92s/it]
Training Epoch: 7/100, batch 4/188 completed (loss: 0.2067:   3%|▎         | 3/94 [00:17<08:59,  5.93s/it]
Training Epoch: 7/100, batch 4/188 completed (loss: 0.3064:   3%|▎         | 3/94 [00:17<08:59,  5.93s/it]
Training Epoch: 7/100, batch 5/188 completed (loss: 0.0816:   3%|▎         | 3/94 [00:17<08:59,  5.93s/it]
Training Epoch: 7/100, batch 5/188 completed (loss: 0.212:   3%|▎         | 3/94 [00:17<08:59,  5.92s/it] 
Training Epoch: 7/100, batch 5/188 completed (loss: 0.1134:   3%|▎         | 3/94 [00:17<08:59,  5.93s/it]
Training Epoch: 7/100, batch 5/188 completed (loss: 0.2095:   3%|▎         | 3/94 [00:17<08:59,  5.93s/it]
Training Epoch: 7/100, batch 6/188 completed (loss: 0.1878:   3%|▎         | 3/94 [00:20<08:59,  5.92s/it]
Training Epoch: 7/100, batch 6/188 completed (loss: 0.2134:   3%|▎         | 3/94 [00:20<08:59,  5.93s/it]
Training Epoch: 7/100, batch 6/188 completed (loss: 0.5274:   3%|▎         | 3/94 [00:20<08:59,  5.93s/it]
Training Epoch: 7/100, batch 6/188 completed (loss: 0.0777:   3%|▎         | 3/94 [00:20<08:59,  5.93s/it]
Training Epoch: 7/100, batch 6/188 completed (loss: 0.2134:   4%|▍         | 4/94 [00:23<08:58,  5.98s/it]
Training Epoch: 7/100, batch 6/188 completed (loss: 0.1878:   4%|▍         | 4/94 [00:23<08:58,  5.98s/it]
Training Epoch: 7/100, batch 6/188 completed (loss: 0.0777:   4%|▍         | 4/94 [00:23<08:58,  5.99s/it]
Training Epoch: 7/100, batch 6/188 completed (loss: 0.5274:   4%|▍         | 4/94 [00:23<08:58,  5.99s/it]
Training Epoch: 7/100, batch 7/188 completed (loss: 0.2566:   4%|▍         | 4/94 [00:23<08:58,  5.99s/it]
Training Epoch: 7/100, batch 7/188 completed (loss: 0.2522:   4%|▍         | 4/94 [00:23<08:58,  5.98s/it]
Training Epoch: 7/100, batch 7/188 completed (loss: 0.0406:   4%|▍         | 4/94 [00:23<08:58,  5.98s/it]
Training Epoch: 7/100, batch 7/188 completed (loss: 0.0958:   4%|▍         | 4/94 [00:23<08:58,  5.99s/it]
Training Epoch: 7/100, batch 8/188 completed (loss: 0.0854:   4%|▍         | 4/94 [00:26<08:58,  5.99s/it]
Training Epoch: 7/100, batch 8/188 completed (loss: 0.3491:   4%|▍         | 4/94 [00:26<08:58,  5.98s/it]
Training Epoch: 7/100, batch 8/188 completed (loss: 0.0895:   4%|▍         | 4/94 [00:26<08:58,  5.98s/it]
Training Epoch: 7/100, batch 8/188 completed (loss: 0.0477:   4%|▍         | 4/94 [00:26<08:58,  5.99s/it]
Training Epoch: 7/100, batch 8/188 completed (loss: 0.0895:   5%|▌         | 5/94 [00:29<08:42,  5.87s/it]
Training Epoch: 7/100, batch 8/188 completed (loss: 0.3491:   5%|▌         | 5/94 [00:29<08:42,  5.87s/it]
Training Epoch: 7/100, batch 8/188 completed (loss: 0.0854:   5%|▌         | 5/94 [00:29<08:42,  5.87s/it]
Training Epoch: 7/100, batch 8/188 completed (loss: 0.0477:   5%|▌         | 5/94 [00:29<08:42,  5.87s/it]
Training Epoch: 7/100, batch 9/188 completed (loss: 0.1175:   5%|▌         | 5/94 [00:29<08:42,  5.87s/it]
Training Epoch: 7/100, batch 9/188 completed (loss: 0.1237:   5%|▌         | 5/94 [00:29<08:42,  5.87s/it]
Training Epoch: 7/100, batch 9/188 completed (loss: 0.2561:   5%|▌         | 5/94 [00:29<08:42,  5.87s/it]
Training Epoch: 7/100, batch 9/188 completed (loss: 0.0256:   5%|▌         | 5/94 [00:29<08:42,  5.87s/it]
Training Epoch: 7/100, batch 10/188 completed (loss: 0.1159:   5%|▌         | 5/94 [00:32<08:42,  5.87s/it]
Training Epoch: 7/100, batch 10/188 completed (loss: 0.1468:   5%|▌         | 5/94 [00:32<08:42,  5.87s/it]
Training Epoch: 7/100, batch 10/188 completed (loss: 0.4282:   5%|▌         | 5/94 [00:32<08:42,  5.87s/it]
Training Epoch: 7/100, batch 10/188 completed (loss: 0.1653:   5%|▌         | 5/94 [00:32<08:42,  5.87s/it]
Training Epoch: 7/100, batch 10/188 completed (loss: 0.1468:   6%|▋         | 6/94 [00:35<08:34,  5.85s/it]
Training Epoch: 7/100, batch 10/188 completed (loss: 0.1159:   6%|▋         | 6/94 [00:35<08:34,  5.85s/it]
Training Epoch: 7/100, batch 10/188 completed (loss: 0.4282:   6%|▋         | 6/94 [00:35<08:34,  5.85s/it]
Training Epoch: 7/100, batch 10/188 completed (loss: 0.1653:   6%|▋         | 6/94 [00:35<08:34,  5.85s/it]
Training Epoch: 7/100, batch 11/188 completed (loss: 0.0485:   6%|▋         | 6/94 [00:35<08:34,  5.85s/it]
Training Epoch: 7/100, batch 11/188 completed (loss: 0.4818:   6%|▋         | 6/94 [00:35<08:34,  5.85s/it]
Training Epoch: 7/100, batch 11/188 completed (loss: 0.1359:   6%|▋         | 6/94 [00:35<08:34,  5.85s/it]
Training Epoch: 7/100, batch 11/188 completed (loss: 0.1571:   6%|▋         | 6/94 [00:35<08:34,  5.85s/it]
Training Epoch: 7/100, batch 12/188 completed (loss: 0.0981:   6%|▋         | 6/94 [00:38<08:34,  5.85s/it]
Training Epoch: 7/100, batch 12/188 completed (loss: 0.0907:   6%|▋         | 6/94 [00:38<08:34,  5.85s/it]
Training Epoch: 7/100, batch 12/188 completed (loss: 0.1521:   6%|▋         | 6/94 [00:38<08:34,  5.85s/it]
Training Epoch: 7/100, batch 12/188 completed (loss: 0.4032:   6%|▋         | 6/94 [00:38<08:34,  5.85s/it]
Training Epoch: 7/100, batch 12/188 completed (loss: 0.0907:   7%|▋         | 7/94 [00:41<08:35,  5.93s/it]
Training Epoch: 7/100, batch 12/188 completed (loss: 0.4032:   7%|▋         | 7/94 [00:41<08:35,  5.93s/it]
Training Epoch: 7/100, batch 12/188 completed (loss: 0.1521:   7%|▋         | 7/94 [00:41<08:35,  5.93s/it]
Training Epoch: 7/100, batch 12/188 completed (loss: 0.0981:   7%|▋         | 7/94 [00:41<08:36,  5.93s/it]
Training Epoch: 7/100, batch 13/188 completed (loss: 0.0927:   7%|▋         | 7/94 [00:41<08:36,  5.93s/it]
Training Epoch: 7/100, batch 13/188 completed (loss: 0.128:   7%|▋         | 7/94 [00:41<08:35,  5.93s/it] 
Training Epoch: 7/100, batch 13/188 completed (loss: 0.138:   7%|▋         | 7/94 [00:41<08:35,  5.93s/it] 
Training Epoch: 7/100, batch 13/188 completed (loss: 0.2174:   7%|▋         | 7/94 [00:41<08:35,  5.93s/it]
Training Epoch: 7/100, batch 14/188 completed (loss: 0.0288:   7%|▋         | 7/94 [00:44<08:36,  5.93s/it]
Training Epoch: 7/100, batch 14/188 completed (loss: 0.1128:   7%|▋         | 7/94 [00:44<08:35,  5.93s/it]
Training Epoch: 7/100, batch 14/188 completed (loss: 0.0758:   7%|▋         | 7/94 [00:44<08:35,  5.93s/it]
Training Epoch: 7/100, batch 14/188 completed (loss: 0.405:   7%|▋         | 7/94 [00:44<08:35,  5.93s/it] 
Training Epoch: 7/100, batch 14/188 completed (loss: 0.0758:   9%|▊         | 8/94 [00:47<08:28,  5.91s/it]
Training Epoch: 7/100, batch 14/188 completed (loss: 0.1128:   9%|▊         | 8/94 [00:47<08:28,  5.91s/it]
Training Epoch: 7/100, batch 14/188 completed (loss: 0.405:   9%|▊         | 8/94 [00:47<08:28,  5.91s/it]
Training Epoch: 7/100, batch 14/188 completed (loss: 0.0288:   9%|▊         | 8/94 [00:47<08:28,  5.91s/it]
Training Epoch: 7/100, batch 15/188 completed (loss: 0.3038:   9%|▊         | 8/94 [00:47<08:28,  5.91s/it]
Training Epoch: 7/100, batch 15/188 completed (loss: 0.3918:   9%|▊         | 8/94 [00:47<08:28,  5.91s/it]
Training Epoch: 7/100, batch 15/188 completed (loss: 0.1272:   9%|▊         | 8/94 [00:47<08:28,  5.91s/it]
Training Epoch: 7/100, batch 15/188 completed (loss: 0.4006:   9%|▊         | 8/94 [00:47<08:28,  5.91s/it]
Training Epoch: 7/100, batch 16/188 completed (loss: 0.0303:   9%|▊         | 8/94 [00:50<08:28,  5.91s/it]
Training Epoch: 7/100, batch 16/188 completed (loss: 0.1421:   9%|▊         | 8/94 [00:50<08:28,  5.91s/it]
Training Epoch: 7/100, batch 16/188 completed (loss: 0.3895:   9%|▊         | 8/94 [00:50<08:28,  5.91s/it]
Training Epoch: 7/100, batch 16/188 completed (loss: 0.2256:   9%|▊         | 8/94 [00:50<08:28,  5.91s/it]
Training Epoch: 7/100, batch 16/188 completed (loss: 0.3895:  10%|▉         | 9/94 [00:53<08:29,  5.99s/it]
Training Epoch: 7/100, batch 16/188 completed (loss: 0.1421:  10%|▉         | 9/94 [00:53<08:29,  5.99s/it]
Training Epoch: 7/100, batch 16/188 completed (loss: 0.2256:  10%|▉         | 9/94 [00:53<08:29,  5.99s/it]
Training Epoch: 7/100, batch 16/188 completed (loss: 0.0303:  10%|▉         | 9/94 [00:53<08:29,  5.99s/it]
Training Epoch: 7/100, batch 17/188 completed (loss: 0.0522:  10%|▉         | 9/94 [00:53<08:29,  5.99s/it]
Training Epoch: 7/100, batch 17/188 completed (loss: 0.0933:  10%|▉         | 9/94 [00:53<08:29,  5.99s/it]
Training Epoch: 7/100, batch 17/188 completed (loss: 0.3667:  10%|▉         | 9/94 [00:53<08:29,  5.99s/it]
Training Epoch: 7/100, batch 17/188 completed (loss: 0.0782:  10%|▉         | 9/94 [00:53<08:29,  5.99s/it]
Training Epoch: 7/100, batch 18/188 completed (loss: 0.1065:  10%|▉         | 9/94 [00:56<08:29,  5.99s/it]
Training Epoch: 7/100, batch 18/188 completed (loss: 0.4166:  10%|▉         | 9/94 [00:56<08:29,  5.99s/it]
Training Epoch: 7/100, batch 18/188 completed (loss: 0.0693:  10%|▉         | 9/94 [00:56<08:29,  5.99s/it]
Training Epoch: 7/100, batch 18/188 completed (loss: 0.6651:  10%|▉         | 9/94 [00:56<08:29,  5.99s/it]
Training Epoch: 7/100, batch 18/188 completed (loss: 0.0693:  11%|█         | 10/94 [00:59<08:15,  5.90s/it]
Training Epoch: 7/100, batch 18/188 completed (loss: 0.6651:  11%|█         | 10/94 [00:59<08:15,  5.90s/it]
Training Epoch: 7/100, batch 18/188 completed (loss: 0.1065:  11%|█         | 10/94 [00:59<08:16,  5.91s/it]
Training Epoch: 7/100, batch 18/188 completed (loss: 0.4166:  11%|█         | 10/94 [00:59<08:16,  5.91s/it]
Training Epoch: 7/100, batch 19/188 completed (loss: 0.7912:  11%|█         | 10/94 [00:59<08:16,  5.91s/it]
Training Epoch: 7/100, batch 19/188 completed (loss: 0.1408:  11%|█         | 10/94 [00:59<08:15,  5.90s/it]
Training Epoch: 7/100, batch 19/188 completed (loss: 0.1285:  11%|█         | 10/94 [00:59<08:16,  5.91s/it]
Training Epoch: 7/100, batch 19/188 completed (loss: 0.1941:  11%|█         | 10/94 [00:59<08:15,  5.90s/it]
Training Epoch: 7/100, batch 20/188 completed (loss: 0.3396:  11%|█         | 10/94 [01:02<08:16,  5.91s/it]
Training Epoch: 7/100, batch 20/188 completed (loss: 0.1132:  11%|█         | 10/94 [01:02<08:15,  5.90s/it]
Training Epoch: 7/100, batch 20/188 completed (loss: 0.2126:  11%|█         | 10/94 [01:02<08:16,  5.91s/it]
Training Epoch: 7/100, batch 20/188 completed (loss: 0.1851:  11%|█         | 10/94 [01:02<08:15,  5.90s/it]
Training Epoch: 7/100, batch 20/188 completed (loss: 0.1851:  12%|█▏        | 11/94 [01:05<08:10,  5.91s/it]
Training Epoch: 7/100, batch 20/188 completed (loss: 0.3396:  12%|█▏        | 11/94 [01:05<08:10,  5.91s/it]
Training Epoch: 7/100, batch 20/188 completed (loss: 0.1132:  12%|█▏        | 11/94 [01:05<08:10,  5.91s/it]
Training Epoch: 7/100, batch 20/188 completed (loss: 0.2126:  12%|█▏        | 11/94 [01:05<08:10,  5.91s/it]
Training Epoch: 7/100, batch 21/188 completed (loss: 0.0491:  12%|█▏        | 11/94 [01:05<08:10,  5.91s/it]
Training Epoch: 7/100, batch 21/188 completed (loss: 0.0603:  12%|█▏        | 11/94 [01:05<08:10,  5.91s/it]
Training Epoch: 7/100, batch 21/188 completed (loss: 0.3178:  12%|█▏        | 11/94 [01:05<08:10,  5.91s/it]
Training Epoch: 7/100, batch 21/188 completed (loss: 0.283:  12%|█▏        | 11/94 [01:05<08:10,  5.91s/it] 
Training Epoch: 7/100, batch 22/188 completed (loss: 0.0993:  12%|█▏        | 11/94 [01:08<08:10,  5.91s/it]
Training Epoch: 7/100, batch 22/188 completed (loss: 0.0563:  12%|█▏        | 11/94 [01:08<08:10,  5.91s/it]
Training Epoch: 7/100, batch 22/188 completed (loss: 0.0633:  12%|█▏        | 11/94 [01:08<08:10,  5.91s/it]
Training Epoch: 7/100, batch 22/188 completed (loss: 0.0802:  12%|█▏        | 11/94 [01:08<08:10,  5.91s/it]
Training Epoch: 7/100, batch 22/188 completed (loss: 0.0633:  13%|█▎        | 12/94 [01:11<08:11,  5.99s/it]
Training Epoch: 7/100, batch 22/188 completed (loss: 0.0802:  13%|█▎        | 12/94 [01:11<08:11,  5.99s/it]
Training Epoch: 7/100, batch 22/188 completed (loss: 0.0563:  13%|█▎        | 12/94 [01:11<08:11,  5.99s/it]
Training Epoch: 7/100, batch 22/188 completed (loss: 0.0993:  13%|█▎        | 12/94 [01:11<08:11,  5.99s/it]
Training Epoch: 7/100, batch 23/188 completed (loss: 0.1945:  13%|█▎        | 12/94 [01:11<08:11,  5.99s/it]
Training Epoch: 7/100, batch 23/188 completed (loss: 0.2953:  13%|█▎        | 12/94 [01:11<08:11,  5.99s/it]
Training Epoch: 7/100, batch 23/188 completed (loss: 0.1238:  13%|█▎        | 12/94 [01:11<08:11,  5.99s/it]
Training Epoch: 7/100, batch 23/188 completed (loss: 0.2453:  13%|█▎        | 12/94 [01:11<08:11,  5.99s/it]
Training Epoch: 7/100, batch 24/188 completed (loss: 0.0757:  13%|█▎        | 12/94 [01:14<08:11,  5.99s/it]
Training Epoch: 7/100, batch 24/188 completed (loss: 0.2341:  13%|█▎        | 12/94 [01:14<08:11,  5.99s/it]
Training Epoch: 7/100, batch 24/188 completed (loss: 0.0532:  13%|█▎        | 12/94 [01:14<08:11,  5.99s/it]
Training Epoch: 7/100, batch 24/188 completed (loss: 0.4774:  13%|█▎        | 12/94 [01:14<08:11,  5.99s/it]
Training Epoch: 7/100, batch 24/188 completed (loss: 0.0757:  14%|█▍        | 13/94 [01:17<08:03,  5.97s/it]
Training Epoch: 7/100, batch 24/188 completed (loss: 0.2341:  14%|█▍        | 13/94 [01:17<08:03,  5.97s/it]
Training Epoch: 7/100, batch 24/188 completed (loss: 0.4774:  14%|█▍        | 13/94 [01:17<08:03,  5.97s/it]
Training Epoch: 7/100, batch 24/188 completed (loss: 0.0532:  14%|█▍        | 13/94 [01:17<08:03,  5.97s/it]
Training Epoch: 7/100, batch 25/188 completed (loss: 0.3485:  14%|█▍        | 13/94 [01:17<08:03,  5.97s/it]
Training Epoch: 7/100, batch 25/188 completed (loss: 0.059:  14%|█▍        | 13/94 [01:17<08:03,  5.97s/it] 
Training Epoch: 7/100, batch 25/188 completed (loss: 0.1354:  14%|█▍        | 13/94 [01:17<08:03,  5.97s/it]
Training Epoch: 7/100, batch 25/188 completed (loss: 0.1452:  14%|█▍        | 13/94 [01:17<08:03,  5.97s/it]
Training Epoch: 7/100, batch 26/188 completed (loss: 0.0648:  14%|█▍        | 13/94 [01:20<08:03,  5.97s/it]
Training Epoch: 7/100, batch 26/188 completed (loss: 0.24:  14%|█▍        | 13/94 [01:20<08:03,  5.97s/it]  
Training Epoch: 7/100, batch 26/188 completed (loss: 0.1143:  14%|█▍        | 13/94 [01:20<08:03,  5.97s/it]
Training Epoch: 7/100, batch 26/188 completed (loss: 0.3001:  14%|█▍        | 13/94 [01:20<08:03,  5.97s/it]
Training Epoch: 7/100, batch 26/188 completed (loss: 0.1143:  15%|█▍        | 14/94 [01:23<07:55,  5.95s/it]
Training Epoch: 7/100, batch 26/188 completed (loss: 0.3001:  15%|█▍        | 14/94 [01:23<07:55,  5.95s/it]
Training Epoch: 7/100, batch 26/188 completed (loss: 0.0648:  15%|█▍        | 14/94 [01:23<07:55,  5.95s/it]
Training Epoch: 7/100, batch 26/188 completed (loss: 0.24:  15%|█▍        | 14/94 [01:23<07:56,  5.95s/it]
Training Epoch: 7/100, batch 27/188 completed (loss: 0.2809:  15%|█▍        | 14/94 [01:23<07:55,  5.95s/it]
Training Epoch: 7/100, batch 27/188 completed (loss: 0.0916:  15%|█▍        | 14/94 [01:23<07:56,  5.95s/it]
Training Epoch: 7/100, batch 27/188 completed (loss: 0.0634:  15%|█▍        | 14/94 [01:23<07:55,  5.95s/it]
Training Epoch: 7/100, batch 27/188 completed (loss: 0.0558:  15%|█▍        | 14/94 [01:23<07:55,  5.95s/it]
Training Epoch: 7/100, batch 28/188 completed (loss: 0.3553:  15%|█▍        | 14/94 [01:26<07:55,  5.95s/it]
Training Epoch: 7/100, batch 28/188 completed (loss: 0.1099:  15%|█▍        | 14/94 [01:26<07:55,  5.95s/it]
Training Epoch: 7/100, batch 28/188 completed (loss: 0.1328:  15%|█▍        | 14/94 [01:26<07:56,  5.95s/it]
Training Epoch: 7/100, batch 28/188 completed (loss: 0.0965:  15%|█▍        | 14/94 [01:26<07:55,  5.95s/it]
Training Epoch: 7/100, batch 28/188 completed (loss: 0.1099:  16%|█▌        | 15/94 [01:29<07:52,  5.98s/it]
Training Epoch: 7/100, batch 28/188 completed (loss: 0.0965:  16%|█▌        | 15/94 [01:29<07:52,  5.98s/it]
Training Epoch: 7/100, batch 28/188 completed (loss: 0.3553:  16%|█▌        | 15/94 [01:29<07:52,  5.98s/it]
Training Epoch: 7/100, batch 28/188 completed (loss: 0.1328:  16%|█▌        | 15/94 [01:29<07:52,  5.98s/it]
Training Epoch: 7/100, batch 29/188 completed (loss: 0.1358:  16%|█▌        | 15/94 [01:29<07:52,  5.98s/it]
Training Epoch: 7/100, batch 29/188 completed (loss: 0.0826:  16%|█▌        | 15/94 [01:29<07:52,  5.98s/it]
Training Epoch: 7/100, batch 29/188 completed (loss: 0.2537:  16%|█▌        | 15/94 [01:29<07:52,  5.98s/it]
Training Epoch: 7/100, batch 29/188 completed (loss: 0.0738:  16%|█▌        | 15/94 [01:29<07:52,  5.98s/it]
Training Epoch: 7/100, batch 30/188 completed (loss: 0.099:  16%|█▌        | 15/94 [01:32<07:52,  5.98s/it] 
Training Epoch: 7/100, batch 30/188 completed (loss: 0.1692:  16%|█▌        | 15/94 [01:32<07:52,  5.98s/it]
Training Epoch: 7/100, batch 30/188 completed (loss: 0.1569:  16%|█▌        | 15/94 [01:32<07:52,  5.98s/it]
Training Epoch: 7/100, batch 30/188 completed (loss: 0.223:  16%|█▌        | 15/94 [01:32<07:52,  5.98s/it] 
Training Epoch: 7/100, batch 30/188 completed (loss: 0.223:  17%|█▋        | 16/94 [01:35<07:44,  5.96s/it]
Training Epoch: 7/100, batch 30/188 completed (loss: 0.1569:  17%|█▋        | 16/94 [01:35<07:44,  5.96s/it]
Training Epoch: 7/100, batch 30/188 completed (loss: 0.1692:  17%|█▋        | 16/94 [01:35<07:44,  5.96s/it]
Training Epoch: 7/100, batch 30/188 completed (loss: 0.099:  17%|█▋        | 16/94 [01:35<07:44,  5.96s/it]
Training Epoch: 7/100, batch 31/188 completed (loss: 0.1173:  17%|█▋        | 16/94 [01:35<07:44,  5.96s/it]
Training Epoch: 7/100, batch 31/188 completed (loss: 0.203:  17%|█▋        | 16/94 [01:35<07:44,  5.96s/it] 
Training Epoch: 7/100, batch 31/188 completed (loss: 0.1601:  17%|█▋        | 16/94 [01:35<07:44,  5.96s/it]
Training Epoch: 7/100, batch 31/188 completed (loss: 0.3776:  17%|█▋        | 16/94 [01:35<07:44,  5.96s/it]
Training Epoch: 7/100, batch 32/188 completed (loss: 0.1173:  17%|█▋        | 16/94 [01:38<07:44,  5.96s/it]
Training Epoch: 7/100, batch 32/188 completed (loss: 0.0747:  17%|█▋        | 16/94 [01:38<07:44,  5.96s/it]
Training Epoch: 7/100, batch 32/188 completed (loss: 0.1226:  17%|█▋        | 16/94 [01:38<07:44,  5.96s/it]
Training Epoch: 7/100, batch 32/188 completed (loss: 0.0433:  17%|█▋        | 16/94 [01:38<07:44,  5.96s/it]
Training Epoch: 7/100, batch 32/188 completed (loss: 0.0433:  18%|█▊        | 17/94 [01:41<07:45,  6.04s/it]
Training Epoch: 7/100, batch 32/188 completed (loss: 0.0747:  18%|█▊        | 17/94 [01:41<07:45,  6.04s/it]
Training Epoch: 7/100, batch 32/188 completed (loss: 0.1173:  18%|█▊        | 17/94 [01:41<07:45,  6.05s/it]
Training Epoch: 7/100, batch 32/188 completed (loss: 0.1226:  18%|█▊        | 17/94 [01:41<07:45,  6.05s/it]
Training Epoch: 7/100, batch 33/188 completed (loss: 0.2977:  18%|█▊        | 17/94 [01:41<07:45,  6.05s/it]
Training Epoch: 7/100, batch 33/188 completed (loss: 0.1079:  18%|█▊        | 17/94 [01:41<07:45,  6.04s/it]
Training Epoch: 7/100, batch 33/188 completed (loss: 0.0911:  18%|█▊        | 17/94 [01:41<07:45,  6.05s/it]
Training Epoch: 7/100, batch 33/188 completed (loss: 0.1309:  18%|█▊        | 17/94 [01:41<07:45,  6.04s/it]
Training Epoch: 7/100, batch 34/188 completed (loss: 0.3682:  18%|█▊        | 17/94 [01:44<07:45,  6.05s/it]
Training Epoch: 7/100, batch 34/188 completed (loss: 0.1646:  18%|█▊        | 17/94 [01:44<07:45,  6.04s/it]
Training Epoch: 7/100, batch 34/188 completed (loss: 0.0639:  18%|█▊        | 17/94 [01:44<07:45,  6.05s/it]
Training Epoch: 7/100, batch 34/188 completed (loss: 0.1777:  18%|█▊        | 17/94 [01:44<07:45,  6.04s/it]
Training Epoch: 7/100, batch 34/188 completed (loss: 0.1646:  19%|█▉        | 18/94 [01:47<07:32,  5.96s/it]
Training Epoch: 7/100, batch 34/188 completed (loss: 0.1777:  19%|█▉        | 18/94 [01:47<07:32,  5.96s/it]
Training Epoch: 7/100, batch 34/188 completed (loss: 0.0639:  19%|█▉        | 18/94 [01:47<07:32,  5.96s/it]
Training Epoch: 7/100, batch 34/188 completed (loss: 0.3682:  19%|█▉        | 18/94 [01:47<07:32,  5.96s/it]
Training Epoch: 7/100, batch 35/188 completed (loss: 0.1111:  19%|█▉        | 18/94 [01:47<07:32,  5.96s/it]
Training Epoch: 7/100, batch 35/188 completed (loss: 0.0625:  19%|█▉        | 18/94 [01:47<07:32,  5.96s/it]
Training Epoch: 7/100, batch 35/188 completed (loss: 0.1335:  19%|█▉        | 18/94 [01:47<07:32,  5.96s/it]
Training Epoch: 7/100, batch 35/188 completed (loss: 0.1148:  19%|█▉        | 18/94 [01:47<07:32,  5.96s/it]
Training Epoch: 7/100, batch 36/188 completed (loss: 0.1234:  19%|█▉        | 18/94 [01:50<07:32,  5.96s/it]
Training Epoch: 7/100, batch 36/188 completed (loss: 0.1369:  19%|█▉        | 18/94 [01:50<07:32,  5.96s/it]
Training Epoch: 7/100, batch 36/188 completed (loss: 0.1547:  19%|█▉        | 18/94 [01:50<07:32,  5.96s/it]
Training Epoch: 7/100, batch 36/188 completed (loss: 0.089:  19%|█▉        | 18/94 [01:50<07:32,  5.96s/it] 
Training Epoch: 7/100, batch 36/188 completed (loss: 0.089:  20%|██        | 19/94 [01:53<07:29,  5.99s/it]
Training Epoch: 7/100, batch 36/188 completed (loss: 0.1547:  20%|██        | 19/94 [01:53<07:29,  5.99s/it]
Training Epoch: 7/100, batch 36/188 completed (loss: 0.1369:  20%|██        | 19/94 [01:53<07:29,  5.99s/it]
Training Epoch: 7/100, batch 36/188 completed (loss: 0.1234:  20%|██        | 19/94 [01:53<07:29,  5.99s/it]
Training Epoch: 7/100, batch 37/188 completed (loss: 0.0969:  20%|██        | 19/94 [01:53<07:29,  5.99s/it]
Training Epoch: 7/100, batch 37/188 completed (loss: 0.0762:  20%|██        | 19/94 [01:53<07:29,  5.99s/it]
Training Epoch: 7/100, batch 37/188 completed (loss: 0.2169:  20%|██        | 19/94 [01:53<07:29,  5.99s/it]
Training Epoch: 7/100, batch 37/188 completed (loss: 0.1468:  20%|██        | 19/94 [01:53<07:29,  5.99s/it]
Training Epoch: 7/100, batch 38/188 completed (loss: 0.1577:  20%|██        | 19/94 [01:56<07:29,  5.99s/it]
Training Epoch: 7/100, batch 38/188 completed (loss: 0.2012:  20%|██        | 19/94 [01:56<07:29,  5.99s/it]
Training Epoch: 7/100, batch 38/188 completed (loss: 0.1652:  20%|██        | 19/94 [01:56<07:29,  5.99s/it]
Training Epoch: 7/100, batch 38/188 completed (loss: 0.0702:  20%|██        | 19/94 [01:56<07:29,  5.99s/it]
Training Epoch: 7/100, batch 38/188 completed (loss: 0.2012:  21%|██▏       | 20/94 [01:59<07:25,  6.02s/it]
Training Epoch: 7/100, batch 38/188 completed (loss: 0.0702:  21%|██▏       | 20/94 [01:59<07:25,  6.02s/it]
Training Epoch: 7/100, batch 38/188 completed (loss: 0.1577:  21%|██▏       | 20/94 [01:59<07:25,  6.02s/it]
Training Epoch: 7/100, batch 38/188 completed (loss: 0.1652:  21%|██▏       | 20/94 [01:59<07:25,  6.02s/it]
Training Epoch: 7/100, batch 39/188 completed (loss: 0.1768:  21%|██▏       | 20/94 [01:59<07:25,  6.02s/it]
Training Epoch: 7/100, batch 39/188 completed (loss: 0.0561:  21%|██▏       | 20/94 [01:59<07:25,  6.02s/it]
Training Epoch: 7/100, batch 39/188 completed (loss: 0.1612:  21%|██▏       | 20/94 [01:59<07:25,  6.02s/it]
Training Epoch: 7/100, batch 39/188 completed (loss: 0.065:  21%|██▏       | 20/94 [01:59<07:25,  6.02s/it] 
Training Epoch: 7/100, batch 40/188 completed (loss: 0.2682:  21%|██▏       | 20/94 [02:02<07:25,  6.02s/it]
Training Epoch: 7/100, batch 40/188 completed (loss: 0.0991:  21%|██▏       | 20/94 [02:02<07:25,  6.02s/it]
Training Epoch: 7/100, batch 40/188 completed (loss: 0.2241:  21%|██▏       | 20/94 [02:02<07:25,  6.02s/it]
Training Epoch: 7/100, batch 40/188 completed (loss: 0.1891:  21%|██▏       | 20/94 [02:02<07:25,  6.02s/it]
Training Epoch: 7/100, batch 40/188 completed (loss: 0.2241:  22%|██▏       | 21/94 [02:05<07:13,  5.94s/it]
Training Epoch: 7/100, batch 40/188 completed (loss: 0.0991:  22%|██▏       | 21/94 [02:05<07:13,  5.94s/it]
Training Epoch: 7/100, batch 40/188 completed (loss: 0.1891:  22%|██▏       | 21/94 [02:05<07:14,  5.95s/it]
Training Epoch: 7/100, batch 40/188 completed (loss: 0.2682:  22%|██▏       | 21/94 [02:05<07:13,  5.94s/it]
Training Epoch: 7/100, batch 41/188 completed (loss: 0.1165:  22%|██▏       | 21/94 [02:05<07:13,  5.94s/it]
Training Epoch: 7/100, batch 41/188 completed (loss: 0.1386:  22%|██▏       | 21/94 [02:05<07:13,  5.94s/it]
Training Epoch: 7/100, batch 41/188 completed (loss: 0.3596:  22%|██▏       | 21/94 [02:05<07:13,  5.94s/it]
Training Epoch: 7/100, batch 41/188 completed (loss: 0.1598:  22%|██▏       | 21/94 [02:05<07:14,  5.95s/it]
Training Epoch: 7/100, batch 42/188 completed (loss: 0.0774:  22%|██▏       | 21/94 [02:08<07:13,  5.94s/it]
Training Epoch: 7/100, batch 42/188 completed (loss: 0.1971:  22%|██▏       | 21/94 [02:08<07:13,  5.94s/it]
Training Epoch: 7/100, batch 42/188 completed (loss: 0.1062:  22%|██▏       | 21/94 [02:08<07:13,  5.94s/it]
Training Epoch: 7/100, batch 42/188 completed (loss: 0.2816:  22%|██▏       | 21/94 [02:08<07:14,  5.95s/it]
Training Epoch: 7/100, batch 42/188 completed (loss: 0.2816:  23%|██▎       | 22/94 [02:11<07:15,  6.05s/it]
Training Epoch: 7/100, batch 42/188 completed (loss: 0.0774:  23%|██▎       | 22/94 [02:11<07:16,  6.06s/it]
Training Epoch: 7/100, batch 42/188 completed (loss: 0.1971:  23%|██▎       | 22/94 [02:11<07:15,  6.05s/it]
Training Epoch: 7/100, batch 42/188 completed (loss: 0.1062:  23%|██▎       | 22/94 [02:11<07:16,  6.06s/it]
Training Epoch: 7/100, batch 43/188 completed (loss: 0.1131:  23%|██▎       | 22/94 [02:11<07:16,  6.06s/it]
Training Epoch: 7/100, batch 43/188 completed (loss: 0.2746:  23%|██▎       | 22/94 [02:11<07:15,  6.05s/it]
Training Epoch: 7/100, batch 43/188 completed (loss: 0.2705:  23%|██▎       | 22/94 [02:11<07:16,  6.06s/it]
Training Epoch: 7/100, batch 43/188 completed (loss: 0.0463:  23%|██▎       | 22/94 [02:11<07:15,  6.05s/it]
Training Epoch: 7/100, batch 44/188 completed (loss: 0.0657:  23%|██▎       | 22/94 [02:14<07:15,  6.05s/it]
Training Epoch: 7/100, batch 44/188 completed (loss: 0.0809:  23%|██▎       | 22/94 [02:14<07:16,  6.06s/it]
Training Epoch: 7/100, batch 44/188 completed (loss: 0.0462:  23%|██▎       | 22/94 [02:14<07:16,  6.06s/it]
Training Epoch: 7/100, batch 44/188 completed (loss: 0.2058:  23%|██▎       | 22/94 [02:14<07:15,  6.05s/it]
Training Epoch: 7/100, batch 44/188 completed (loss: 0.2058:  24%|██▍       | 23/94 [02:17<07:05,  5.99s/it]
Training Epoch: 7/100, batch 44/188 completed (loss: 0.0809:  24%|██▍       | 23/94 [02:17<07:05,  5.99s/it]
Training Epoch: 7/100, batch 44/188 completed (loss: 0.0462:  24%|██▍       | 23/94 [02:17<07:05,  5.99s/it]
Training Epoch: 7/100, batch 44/188 completed (loss: 0.0657:  24%|██▍       | 23/94 [02:17<07:05,  5.99s/it]
Training Epoch: 7/100, batch 45/188 completed (loss: 0.2632:  24%|██▍       | 23/94 [02:17<07:05,  5.99s/it]
Training Epoch: 7/100, batch 45/188 completed (loss: 0.4982:  24%|██▍       | 23/94 [02:17<07:05,  5.99s/it]
Training Epoch: 7/100, batch 45/188 completed (loss: 0.2284:  24%|██▍       | 23/94 [02:17<07:05,  5.99s/it]
Training Epoch: 7/100, batch 45/188 completed (loss: 0.1546:  24%|██▍       | 23/94 [02:17<07:05,  5.99s/it]
Training Epoch: 7/100, batch 46/188 completed (loss: 0.1038:  24%|██▍       | 23/94 [02:20<07:05,  5.99s/it]
Training Epoch: 7/100, batch 46/188 completed (loss: 0.1964:  24%|██▍       | 23/94 [02:20<07:05,  5.99s/it]
Training Epoch: 7/100, batch 46/188 completed (loss: 0.3561:  24%|██▍       | 23/94 [02:20<07:05,  5.99s/it]
Training Epoch: 7/100, batch 46/188 completed (loss: 0.2353:  24%|██▍       | 23/94 [02:20<07:05,  5.99s/it]
Training Epoch: 7/100, batch 46/188 completed (loss: 0.2353:  26%|██▌       | 24/94 [02:23<06:58,  5.99s/it]
Training Epoch: 7/100, batch 46/188 completed (loss: 0.3561:  26%|██▌       | 24/94 [02:23<06:58,  5.99s/it]
Training Epoch: 7/100, batch 46/188 completed (loss: 0.1964:  26%|██▌       | 24/94 [02:23<06:58,  5.98s/it]
Training Epoch: 7/100, batch 46/188 completed (loss: 0.1038:  26%|██▌       | 24/94 [02:23<06:59,  5.99s/it]
Training Epoch: 7/100, batch 47/188 completed (loss: 0.0977:  26%|██▌       | 24/94 [02:23<06:59,  5.99s/it]
Training Epoch: 7/100, batch 47/188 completed (loss: 0.0537:  26%|██▌       | 24/94 [02:23<06:58,  5.98s/it]
Training Epoch: 7/100, batch 47/188 completed (loss: 0.0346:  26%|██▌       | 24/94 [02:23<06:58,  5.99s/it]
Training Epoch: 7/100, batch 47/188 completed (loss: 0.2041:  26%|██▌       | 24/94 [02:23<06:58,  5.99s/it]
Training Epoch: 7/100, batch 48/188 completed (loss: 0.2268:  26%|██▌       | 24/94 [02:26<06:58,  5.98s/it]
Training Epoch: 7/100, batch 48/188 completed (loss: 0.0955:  26%|██▌       | 24/94 [02:26<06:58,  5.99s/it]
Training Epoch: 7/100, batch 48/188 completed (loss: 0.2376:  26%|██▌       | 24/94 [02:26<06:59,  5.99s/it]
Training Epoch: 7/100, batch 48/188 completed (loss: 0.1208:  26%|██▌       | 24/94 [02:26<06:58,  5.99s/it]
Training Epoch: 7/100, batch 48/188 completed (loss: 0.0955:  27%|██▋       | 25/94 [02:29<06:56,  6.04s/it]
Training Epoch: 7/100, batch 48/188 completed (loss: 0.1208:  27%|██▋       | 25/94 [02:29<06:56,  6.04s/it]
Training Epoch: 7/100, batch 48/188 completed (loss: 0.2268:  27%|██▋       | 25/94 [02:29<06:56,  6.04s/it]
Training Epoch: 7/100, batch 48/188 completed (loss: 0.2376:  27%|██▋       | 25/94 [02:29<06:57,  6.04s/it]
Training Epoch: 7/100, batch 49/188 completed (loss: 0.1289:  27%|██▋       | 25/94 [02:29<06:56,  6.04s/it]
Training Epoch: 7/100, batch 49/188 completed (loss: 0.1555:  27%|██▋       | 25/94 [02:29<06:57,  6.04s/it]
Training Epoch: 7/100, batch 49/188 completed (loss: 0.3296:  27%|██▋       | 25/94 [02:29<06:56,  6.04s/it]
Training Epoch: 7/100, batch 49/188 completed (loss: 0.2192:  27%|██▋       | 25/94 [02:29<06:56,  6.04s/it]
Training Epoch: 7/100, batch 50/188 completed (loss: 0.0796:  27%|██▋       | 25/94 [02:32<06:56,  6.04s/it]
Training Epoch: 7/100, batch 50/188 completed (loss: 0.4337:  27%|██▋       | 25/94 [02:32<06:56,  6.04s/it]
Training Epoch: 7/100, batch 50/188 completed (loss: 0.1499:  27%|██▋       | 25/94 [02:32<06:57,  6.04s/it]
Training Epoch: 7/100, batch 50/188 completed (loss: 0.1502:  27%|██▋       | 25/94 [02:32<06:56,  6.04s/it]
Training Epoch: 7/100, batch 50/188 completed (loss: 0.1499:  28%|██▊       | 26/94 [02:35<06:45,  5.97s/it]
Training Epoch: 7/100, batch 50/188 completed (loss: 0.1502:  28%|██▊       | 26/94 [02:35<06:45,  5.97s/it]
Training Epoch: 7/100, batch 50/188 completed (loss: 0.0796:  28%|██▊       | 26/94 [02:35<06:46,  5.97s/it]
Training Epoch: 7/100, batch 50/188 completed (loss: 0.4337:  28%|██▊       | 26/94 [02:35<06:46,  5.97s/it]
Training Epoch: 7/100, batch 51/188 completed (loss: 0.1936:  28%|██▊       | 26/94 [02:35<06:46,  5.97s/it]
Training Epoch: 7/100, batch 51/188 completed (loss: 0.103:  28%|██▊       | 26/94 [02:35<06:46,  5.97s/it] 
Training Epoch: 7/100, batch 51/188 completed (loss: 0.0838:  28%|██▊       | 26/94 [02:35<06:45,  5.97s/it]
Training Epoch: 7/100, batch 51/188 completed (loss: 0.0876:  28%|██▊       | 26/94 [02:35<06:45,  5.97s/it]
Training Epoch: 7/100, batch 52/188 completed (loss: 0.0851:  28%|██▊       | 26/94 [02:38<06:46,  5.97s/it]
Training Epoch: 7/100, batch 52/188 completed (loss: 0.1808:  28%|██▊       | 26/94 [02:38<06:46,  5.97s/it]
Training Epoch: 7/100, batch 52/188 completed (loss: 0.0504:  28%|██▊       | 26/94 [02:38<06:45,  5.97s/it]
Training Epoch: 7/100, batch 52/188 completed (loss: 0.0973:  28%|██▊       | 26/94 [02:38<06:45,  5.97s/it]
Training Epoch: 7/100, batch 52/188 completed (loss: 0.0973:  29%|██▊       | 27/94 [02:41<06:46,  6.07s/it]
Training Epoch: 7/100, batch 52/188 completed (loss: 0.0504:  29%|██▊       | 27/94 [02:41<06:47,  6.07s/it]
Training Epoch: 7/100, batch 52/188 completed (loss: 0.1808:  29%|██▊       | 27/94 [02:41<06:47,  6.08s/it]
Training Epoch: 7/100, batch 52/188 completed (loss: 0.0851:  29%|██▊       | 27/94 [02:41<06:47,  6.08s/it]
Training Epoch: 7/100, batch 53/188 completed (loss: 0.1309:  29%|██▊       | 27/94 [02:41<06:47,  6.07s/it]
Training Epoch: 7/100, batch 53/188 completed (loss: 0.2524:  29%|██▊       | 27/94 [02:41<06:47,  6.08s/it]
Training Epoch: 7/100, batch 53/188 completed (loss: 0.2914:  29%|██▊       | 27/94 [02:41<06:47,  6.08s/it]
Training Epoch: 7/100, batch 53/188 completed (loss: 0.0839:  29%|██▊       | 27/94 [02:41<06:46,  6.07s/it]
Training Epoch: 7/100, batch 54/188 completed (loss: 0.1617:  29%|██▊       | 27/94 [02:44<06:47,  6.08s/it]
Training Epoch: 7/100, batch 54/188 completed (loss: 0.1983:  29%|██▊       | 27/94 [02:44<06:47,  6.08s/it]
Training Epoch: 7/100, batch 54/188 completed (loss: 0.2562:  29%|██▊       | 27/94 [02:44<06:47,  6.07s/it]
Training Epoch: 7/100, batch 54/188 completed (loss: 0.1156:  29%|██▊       | 27/94 [02:44<06:46,  6.07s/it]
Training Epoch: 7/100, batch 54/188 completed (loss: 0.1983:  30%|██▉       | 28/94 [02:47<06:35,  5.99s/it]
Training Epoch: 7/100, batch 54/188 completed (loss: 0.1156:  30%|██▉       | 28/94 [02:47<06:35,  5.99s/it]
Training Epoch: 7/100, batch 54/188 completed (loss: 0.2562:  30%|██▉       | 28/94 [02:47<06:35,  5.99s/it]
Training Epoch: 7/100, batch 54/188 completed (loss: 0.1617:  30%|██▉       | 28/94 [02:47<06:35,  5.99s/it]
Training Epoch: 7/100, batch 55/188 completed (loss: 0.2713:  30%|██▉       | 28/94 [02:47<06:35,  5.99s/it]
Training Epoch: 7/100, batch 55/188 completed (loss: 0.2024:  30%|██▉       | 28/94 [02:47<06:35,  5.99s/it]
Training Epoch: 7/100, batch 55/188 completed (loss: 0.1129:  30%|██▉       | 28/94 [02:47<06:35,  5.99s/it]
Training Epoch: 7/100, batch 55/188 completed (loss: 0.1584:  30%|██▉       | 28/94 [02:47<06:35,  5.99s/it]
Training Epoch: 7/100, batch 56/188 completed (loss: 0.0927:  30%|██▉       | 28/94 [02:50<06:35,  5.99s/it]
Training Epoch: 7/100, batch 56/188 completed (loss: 0.1324:  30%|██▉       | 28/94 [02:50<06:35,  5.99s/it]
Training Epoch: 7/100, batch 56/188 completed (loss: 0.1855:  30%|██▉       | 28/94 [02:50<06:35,  5.99s/it]
Training Epoch: 7/100, batch 56/188 completed (loss: 0.1386:  30%|██▉       | 28/94 [02:50<06:35,  5.99s/it]
Training Epoch: 7/100, batch 56/188 completed (loss: 0.0927:  31%|███       | 29/94 [02:53<06:26,  5.94s/it]
Training Epoch: 7/100, batch 56/188 completed (loss: 0.1386:  31%|███       | 29/94 [02:53<06:26,  5.95s/it]
Training Epoch: 7/100, batch 56/188 completed (loss: 0.1324:  31%|███       | 29/94 [02:53<06:26,  5.95s/it]
Training Epoch: 7/100, batch 56/188 completed (loss: 0.1855:  31%|███       | 29/94 [02:53<06:26,  5.95s/it]
Training Epoch: 7/100, batch 57/188 completed (loss: 0.2932:  31%|███       | 29/94 [02:53<06:26,  5.95s/it]
Training Epoch: 7/100, batch 57/188 completed (loss: 0.1658:  31%|███       | 29/94 [02:53<06:26,  5.94s/it]
Training Epoch: 7/100, batch 57/188 completed (loss: 0.163:  31%|███       | 29/94 [02:53<06:26,  5.95s/it] 
Training Epoch: 7/100, batch 57/188 completed (loss: 0.2571:  31%|███       | 29/94 [02:53<06:26,  5.95s/it]
Training Epoch: 7/100, batch 58/188 completed (loss: 0.163:  31%|███       | 29/94 [02:56<06:26,  5.95s/it]
Training Epoch: 7/100, batch 58/188 completed (loss: 0.146:  31%|███       | 29/94 [02:56<06:26,  5.95s/it] 
Training Epoch: 7/100, batch 58/188 completed (loss: 0.1711:  31%|███       | 29/94 [02:56<06:26,  5.94s/it]
Training Epoch: 7/100, batch 58/188 completed (loss: 0.1713:  31%|███       | 29/94 [02:56<06:26,  5.95s/it]
Training Epoch: 7/100, batch 58/188 completed (loss: 0.1711:  32%|███▏      | 30/94 [02:59<06:28,  6.08s/it]
Training Epoch: 7/100, batch 58/188 completed (loss: 0.1713:  32%|███▏      | 30/94 [02:59<06:29,  6.08s/it]
Training Epoch: 7/100, batch 58/188 completed (loss: 0.146:  32%|███▏      | 30/94 [02:59<06:29,  6.08s/it]
Training Epoch: 7/100, batch 58/188 completed (loss: 0.163:  32%|███▏      | 30/94 [02:59<06:29,  6.08s/it]
Training Epoch: 7/100, batch 59/188 completed (loss: 0.1113:  32%|███▏      | 30/94 [02:59<06:29,  6.08s/it]
Training Epoch: 7/100, batch 59/188 completed (loss: 0.0832:  32%|███▏      | 30/94 [02:59<06:29,  6.08s/it]
Training Epoch: 7/100, batch 59/188 completed (loss: 0.1336:  32%|███▏      | 30/94 [02:59<06:28,  6.08s/it]
Training Epoch: 7/100, batch 59/188 completed (loss: 0.2943:  32%|███▏      | 30/94 [02:59<06:29,  6.08s/it]
Training Epoch: 7/100, batch 60/188 completed (loss: 0.1366:  32%|███▏      | 30/94 [03:02<06:28,  6.08s/it]
Training Epoch: 7/100, batch 60/188 completed (loss: 0.2097:  32%|███▏      | 30/94 [03:02<06:29,  6.08s/it]
Training Epoch: 7/100, batch 60/188 completed (loss: 0.4183:  32%|███▏      | 30/94 [03:02<06:29,  6.08s/it]
Training Epoch: 7/100, batch 60/188 completed (loss: 0.1041:  32%|███▏      | 30/94 [03:02<06:29,  6.08s/it]
Training Epoch: 7/100, batch 60/188 completed (loss: 0.4183:  33%|███▎      | 31/94 [03:05<06:17,  6.00s/it]
Training Epoch: 7/100, batch 60/188 completed (loss: 0.2097:  33%|███▎      | 31/94 [03:05<06:17,  6.00s/it]
Training Epoch: 7/100, batch 60/188 completed (loss: 0.1041:  33%|███▎      | 31/94 [03:05<06:17,  6.00s/it]
Training Epoch: 7/100, batch 60/188 completed (loss: 0.1366:  33%|███▎      | 31/94 [03:05<06:18,  6.00s/it]
Training Epoch: 7/100, batch 61/188 completed (loss: 0.1629:  33%|███▎      | 31/94 [03:05<06:17,  6.00s/it]
Training Epoch: 7/100, batch 61/188 completed (loss: 0.0544:  33%|███▎      | 31/94 [03:05<06:17,  6.00s/it]
Training Epoch: 7/100, batch 61/188 completed (loss: 0.1436:  33%|███▎      | 31/94 [03:05<06:18,  6.00s/it]
Training Epoch: 7/100, batch 61/188 completed (loss: 0.1327:  33%|███▎      | 31/94 [03:05<06:17,  6.00s/it]
Training Epoch: 7/100, batch 62/188 completed (loss: 0.3856:  33%|███▎      | 31/94 [03:08<06:17,  6.00s/it]
Training Epoch: 7/100, batch 62/188 completed (loss: 0.4413:  33%|███▎      | 31/94 [03:08<06:17,  6.00s/it]
Training Epoch: 7/100, batch 62/188 completed (loss: 0.2417:  33%|███▎      | 31/94 [03:08<06:18,  6.00s/it]
Training Epoch: 7/100, batch 62/188 completed (loss: 0.115:  33%|███▎      | 31/94 [03:08<06:17,  6.00s/it] 
Training Epoch: 7/100, batch 62/188 completed (loss: 0.2417:  34%|███▍      | 32/94 [03:11<06:14,  6.05s/it]
Training Epoch: 7/100, batch 62/188 completed (loss: 0.115:  34%|███▍      | 32/94 [03:11<06:14,  6.05s/it]
Training Epoch: 7/100, batch 62/188 completed (loss: 0.3856:  34%|███▍      | 32/94 [03:11<06:14,  6.05s/it]
Training Epoch: 7/100, batch 62/188 completed (loss: 0.4413:  34%|███▍      | 32/94 [03:11<06:14,  6.05s/it]
Training Epoch: 7/100, batch 63/188 completed (loss: 0.0807:  34%|███▍      | 32/94 [03:11<06:14,  6.05s/it]
Training Epoch: 7/100, batch 63/188 completed (loss: 0.2668:  34%|███▍      | 32/94 [03:11<06:14,  6.05s/it]
Training Epoch: 7/100, batch 63/188 completed (loss: 0.2476:  34%|███▍      | 32/94 [03:11<06:14,  6.05s/it]
Training Epoch: 7/100, batch 63/188 completed (loss: 0.2457:  34%|███▍      | 32/94 [03:11<06:14,  6.05s/it]
Training Epoch: 7/100, batch 64/188 completed (loss: 0.046:  34%|███▍      | 32/94 [03:14<06:14,  6.05s/it] 
Training Epoch: 7/100, batch 64/188 completed (loss: 0.461:  34%|███▍      | 32/94 [03:14<06:14,  6.05s/it] 
Training Epoch: 7/100, batch 64/188 completed (loss: 0.1537:  34%|███▍      | 32/94 [03:14<06:14,  6.05s/it]
Training Epoch: 7/100, batch 64/188 completed (loss: 0.1065:  34%|███▍      | 32/94 [03:14<06:14,  6.05s/it]
Training Epoch: 7/100, batch 64/188 completed (loss: 0.461:  35%|███▌      | 33/94 [03:17<06:09,  6.06s/it]
Training Epoch: 7/100, batch 64/188 completed (loss: 0.1537:  35%|███▌      | 33/94 [03:17<06:09,  6.06s/it]
Training Epoch: 7/100, batch 64/188 completed (loss: 0.046:  35%|███▌      | 33/94 [03:17<06:09,  6.06s/it]
Training Epoch: 7/100, batch 64/188 completed (loss: 0.1065:  35%|███▌      | 33/94 [03:17<06:09,  6.06s/it]
Training Epoch: 7/100, batch 65/188 completed (loss: 0.1573:  35%|███▌      | 33/94 [03:17<06:09,  6.06s/it]
Training Epoch: 7/100, batch 65/188 completed (loss: 0.4412:  35%|███▌      | 33/94 [03:17<06:09,  6.06s/it]
Training Epoch: 7/100, batch 65/188 completed (loss: 0.352:  35%|███▌      | 33/94 [03:17<06:09,  6.06s/it]
Training Epoch: 7/100, batch 65/188 completed (loss: 0.1969:  35%|███▌      | 33/94 [03:17<06:09,  6.06s/it]
Training Epoch: 7/100, batch 66/188 completed (loss: 0.1142:  35%|███▌      | 33/94 [03:20<06:09,  6.06s/it]
Training Epoch: 7/100, batch 66/188 completed (loss: 0.5915:  35%|███▌      | 33/94 [03:20<06:09,  6.06s/it]
Training Epoch: 7/100, batch 66/188 completed (loss: 0.1727:  35%|███▌      | 33/94 [03:20<06:09,  6.06s/it]
Training Epoch: 7/100, batch 66/188 completed (loss: 0.1923:  35%|███▌      | 33/94 [03:20<06:09,  6.06s/it]
Training Epoch: 7/100, batch 66/188 completed (loss: 0.1142:  36%|███▌      | 34/94 [03:23<05:59,  5.98s/it]
Training Epoch: 7/100, batch 66/188 completed (loss: 0.1727:  36%|███▌      | 34/94 [03:23<05:58,  5.98s/it]
Training Epoch: 7/100, batch 66/188 completed (loss: 0.1923:  36%|███▌      | 34/94 [03:23<05:58,  5.98s/it]
Training Epoch: 7/100, batch 66/188 completed (loss: 0.5915:  36%|███▌      | 34/94 [03:23<05:58,  5.98s/it]
Training Epoch: 7/100, batch 67/188 completed (loss: 0.1428:  36%|███▌      | 34/94 [03:23<05:58,  5.98s/it]
Training Epoch: 7/100, batch 67/188 completed (loss: 0.1725:  36%|███▌      | 34/94 [03:23<05:58,  5.98s/it]
Training Epoch: 7/100, batch 67/188 completed (loss: 0.0889:  36%|███▌      | 34/94 [03:23<05:59,  5.98s/it]
Training Epoch: 7/100, batch 67/188 completed (loss: 0.2416:  36%|███▌      | 34/94 [03:23<05:58,  5.98s/it]
Training Epoch: 7/100, batch 68/188 completed (loss: 0.1244:  36%|███▌      | 34/94 [03:26<05:59,  5.98s/it]
Training Epoch: 7/100, batch 68/188 completed (loss: 0.124:  36%|███▌      | 34/94 [03:26<05:58,  5.98s/it] 
Training Epoch: 7/100, batch 68/188 completed (loss: 0.1145:  36%|███▌      | 34/94 [03:26<05:58,  5.98s/it]
Training Epoch: 7/100, batch 68/188 completed (loss: 0.272:  36%|███▌      | 34/94 [03:26<05:58,  5.98s/it] 
Training Epoch: 7/100, batch 68/188 completed (loss: 0.124:  37%|███▋      | 35/94 [03:29<05:56,  6.05s/it]
Training Epoch: 7/100, batch 68/188 completed (loss: 0.272:  37%|███▋      | 35/94 [03:29<05:56,  6.05s/it]
Training Epoch: 7/100, batch 68/188 completed (loss: 0.1244:  37%|███▋      | 35/94 [03:29<05:56,  6.05s/it]
Training Epoch: 7/100, batch 68/188 completed (loss: 0.1145:  37%|███▋      | 35/94 [03:29<05:56,  6.05s/it]
Training Epoch: 7/100, batch 69/188 completed (loss: 0.1925:  37%|███▋      | 35/94 [03:29<05:56,  6.05s/it]
Training Epoch: 7/100, batch 69/188 completed (loss: 0.3615:  37%|███▋      | 35/94 [03:29<05:56,  6.05s/it]
Training Epoch: 7/100, batch 69/188 completed (loss: 0.0587:  37%|███▋      | 35/94 [03:29<05:56,  6.05s/it]
Training Epoch: 7/100, batch 69/188 completed (loss: 0.2616:  37%|███▋      | 35/94 [03:29<05:56,  6.05s/it]
Training Epoch: 7/100, batch 70/188 completed (loss: 0.0325:  37%|███▋      | 35/94 [03:32<05:56,  6.05s/it]
Training Epoch: 7/100, batch 70/188 completed (loss: 0.018:  37%|███▋      | 35/94 [03:32<05:56,  6.05s/it] 
Training Epoch: 7/100, batch 70/188 completed (loss: 0.2037:  37%|███▋      | 35/94 [03:32<05:56,  6.05s/it]
Training Epoch: 7/100, batch 70/188 completed (loss: 0.128:  37%|███▋      | 35/94 [03:32<05:56,  6.05s/it] 
Training Epoch: 7/100, batch 70/188 completed (loss: 0.128:  38%|███▊      | 36/94 [03:35<05:46,  5.98s/it]
Training Epoch: 7/100, batch 70/188 completed (loss: 0.018:  38%|███▊      | 36/94 [03:35<05:46,  5.98s/it]
Training Epoch: 7/100, batch 70/188 completed (loss: 0.0325:  38%|███▊      | 36/94 [03:35<05:46,  5.98s/it]
Training Epoch: 7/100, batch 70/188 completed (loss: 0.2037:  38%|███▊      | 36/94 [03:35<05:46,  5.98s/it]
Training Epoch: 7/100, batch 71/188 completed (loss: 0.0637:  38%|███▊      | 36/94 [03:35<05:46,  5.98s/it]
Training Epoch: 7/100, batch 71/188 completed (loss: 0.0553:  38%|███▊      | 36/94 [03:35<05:46,  5.98s/it]
Training Epoch: 7/100, batch 71/188 completed (loss: 0.2047:  38%|███▊      | 36/94 [03:35<05:46,  5.98s/it]
Training Epoch: 7/100, batch 71/188 completed (loss: 0.2326:  38%|███▊      | 36/94 [03:35<05:46,  5.98s/it]
Training Epoch: 7/100, batch 72/188 completed (loss: 0.1065:  38%|███▊      | 36/94 [03:38<05:46,  5.98s/it]
Training Epoch: 7/100, batch 72/188 completed (loss: 0.16:  38%|███▊      | 36/94 [03:38<05:46,  5.98s/it]  
Training Epoch: 7/100, batch 72/188 completed (loss: 0.3015:  38%|███▊      | 36/94 [03:38<05:46,  5.98s/it]
Training Epoch: 7/100, batch 72/188 completed (loss: 0.2459:  38%|███▊      | 36/94 [03:38<05:46,  5.98s/it]
Training Epoch: 7/100, batch 72/188 completed (loss: 0.16:  39%|███▉      | 37/94 [03:41<05:40,  5.98s/it]
Training Epoch: 7/100, batch 72/188 completed (loss: 0.2459:  39%|███▉      | 37/94 [03:41<05:40,  5.98s/it]
Training Epoch: 7/100, batch 72/188 completed (loss: 0.1065:  39%|███▉      | 37/94 [03:41<05:40,  5.98s/it]
Training Epoch: 7/100, batch 72/188 completed (loss: 0.3015:  39%|███▉      | 37/94 [03:41<05:40,  5.98s/it]
Training Epoch: 7/100, batch 73/188 completed (loss: 0.2783:  39%|███▉      | 37/94 [03:41<05:40,  5.98s/it]
Training Epoch: 7/100, batch 73/188 completed (loss: 0.1157:  39%|███▉      | 37/94 [03:41<05:40,  5.98s/it]
Training Epoch: 7/100, batch 73/188 completed (loss: 0.1524:  39%|███▉      | 37/94 [03:41<05:40,  5.98s/it]
Training Epoch: 7/100, batch 73/188 completed (loss: 0.3229:  39%|███▉      | 37/94 [03:41<05:40,  5.98s/it]
Training Epoch: 7/100, batch 74/188 completed (loss: 0.1831:  39%|███▉      | 37/94 [03:44<05:40,  5.98s/it]
Training Epoch: 7/100, batch 74/188 completed (loss: 0.0489:  39%|███▉      | 37/94 [03:44<05:40,  5.98s/it]
Training Epoch: 7/100, batch 74/188 completed (loss: 0.4316:  39%|███▉      | 37/94 [03:44<05:40,  5.98s/it]
Training Epoch: 7/100, batch 74/188 completed (loss: 0.2148:  39%|███▉      | 37/94 [03:44<05:40,  5.98s/it]
Training Epoch: 7/100, batch 74/188 completed (loss: 0.0489:  40%|████      | 38/94 [03:47<05:35,  5.99s/it]
Training Epoch: 7/100, batch 74/188 completed (loss: 0.2148:  40%|████      | 38/94 [03:47<05:35,  5.99s/it]
Training Epoch: 7/100, batch 74/188 completed (loss: 0.1831:  40%|████      | 38/94 [03:47<05:35,  5.99s/it]
Training Epoch: 7/100, batch 74/188 completed (loss: 0.4316:  40%|████      | 38/94 [03:47<05:35,  5.99s/it]
Training Epoch: 7/100, batch 75/188 completed (loss: 0.235:  40%|████      | 38/94 [03:47<05:35,  5.99s/it] 
Training Epoch: 7/100, batch 75/188 completed (loss: 0.0828:  40%|████      | 38/94 [03:47<05:35,  5.99s/it]
Training Epoch: 7/100, batch 75/188 completed (loss: 0.0979:  40%|████      | 38/94 [03:47<05:35,  5.99s/it]
Training Epoch: 7/100, batch 75/188 completed (loss: 0.2436:  40%|████      | 38/94 [03:47<05:35,  5.99s/it]
Training Epoch: 7/100, batch 76/188 completed (loss: 0.0948:  40%|████      | 38/94 [03:50<05:35,  5.99s/it]
Training Epoch: 7/100, batch 76/188 completed (loss: 0.1216:  40%|████      | 38/94 [03:50<05:35,  5.99s/it]
Training Epoch: 7/100, batch 76/188 completed (loss: 0.0483:  40%|████      | 38/94 [03:50<05:35,  5.99s/it]
Training Epoch: 7/100, batch 76/188 completed (loss: 0.1035:  40%|████      | 38/94 [03:50<05:35,  5.99s/it]
Training Epoch: 7/100, batch 76/188 completed (loss: 0.1216:  41%|████▏     | 39/94 [03:53<05:26,  5.94s/it]
Training Epoch: 7/100, batch 76/188 completed (loss: 0.0483:  41%|████▏     | 39/94 [03:53<05:27,  5.95s/it]
Training Epoch: 7/100, batch 76/188 completed (loss: 0.1035:  41%|████▏     | 39/94 [03:53<05:27,  5.95s/it]
Training Epoch: 7/100, batch 76/188 completed (loss: 0.0948:  41%|████▏     | 39/94 [03:53<05:27,  5.95s/it]
Training Epoch: 7/100, batch 77/188 completed (loss: 0.1015:  41%|████▏     | 39/94 [03:53<05:27,  5.95s/it]
Training Epoch: 7/100, batch 77/188 completed (loss: 0.1376:  41%|████▏     | 39/94 [03:53<05:27,  5.95s/it]
Training Epoch: 7/100, batch 77/188 completed (loss: 0.1155:  41%|████▏     | 39/94 [03:53<05:26,  5.94s/it]
Training Epoch: 7/100, batch 77/188 completed (loss: 0.1557:  41%|████▏     | 39/94 [03:53<05:27,  5.95s/it]
Training Epoch: 7/100, batch 78/188 completed (loss: 0.1195:  41%|████▏     | 39/94 [03:56<05:26,  5.94s/it]
Training Epoch: 7/100, batch 78/188 completed (loss: 0.1919:  41%|████▏     | 39/94 [03:56<05:27,  5.95s/it]
Training Epoch: 7/100, batch 78/188 completed (loss: 0.0478:  41%|████▏     | 39/94 [03:56<05:27,  5.95s/it]
Training Epoch: 7/100, batch 78/188 completed (loss: 0.2931:  41%|████▏     | 39/94 [03:56<05:27,  5.95s/it]
Training Epoch: 7/100, batch 78/188 completed (loss: 0.1919:  43%|████▎     | 40/94 [03:59<05:27,  6.06s/it]
Training Epoch: 7/100, batch 78/188 completed (loss: 0.0478:  43%|████▎     | 40/94 [03:59<05:27,  6.06s/it]
Training Epoch: 7/100, batch 78/188 completed (loss: 0.1195:  43%|████▎     | 40/94 [03:59<05:27,  6.07s/it]
Training Epoch: 7/100, batch 78/188 completed (loss: 0.2931:  43%|████▎     | 40/94 [03:59<05:27,  6.06s/it]
Training Epoch: 7/100, batch 79/188 completed (loss: 0.3368:  43%|████▎     | 40/94 [03:59<05:27,  6.07s/it]
Training Epoch: 7/100, batch 79/188 completed (loss: 0.1448:  43%|████▎     | 40/94 [03:59<05:27,  6.06s/it]
Training Epoch: 7/100, batch 79/188 completed (loss: 0.0823:  43%|████▎     | 40/94 [03:59<05:27,  6.06s/it]
Training Epoch: 7/100, batch 79/188 completed (loss: 0.2173:  43%|████▎     | 40/94 [03:59<05:27,  6.06s/it]
Training Epoch: 7/100, batch 80/188 completed (loss: 0.1789:  43%|████▎     | 40/94 [04:02<05:27,  6.06s/it]
Training Epoch: 7/100, batch 80/188 completed (loss: 0.1161:  43%|████▎     | 40/94 [04:02<05:27,  6.07s/it]
Training Epoch: 7/100, batch 80/188 completed (loss: 0.0478:  43%|████▎     | 40/94 [04:02<05:27,  6.06s/it]
Training Epoch: 7/100, batch 80/188 completed (loss: 0.2843:  43%|████▎     | 40/94 [04:02<05:27,  6.06s/it]
Training Epoch: 7/100, batch 80/188 completed (loss: 0.1161:  44%|████▎     | 41/94 [04:05<05:17,  5.99s/it]
Training Epoch: 7/100, batch 80/188 completed (loss: 0.0478:  44%|████▎     | 41/94 [04:05<05:17,  5.99s/it]
Training Epoch: 7/100, batch 80/188 completed (loss: 0.1789:  44%|████▎     | 41/94 [04:05<05:17,  5.99s/it]
Training Epoch: 7/100, batch 80/188 completed (loss: 0.2843:  44%|████▎     | 41/94 [04:05<05:17,  5.99s/it]
Training Epoch: 7/100, batch 81/188 completed (loss: 0.545:  44%|████▎     | 41/94 [04:05<05:17,  5.99s/it] 
Training Epoch: 7/100, batch 81/188 completed (loss: 0.2349:  44%|████▎     | 41/94 [04:05<05:17,  5.99s/it]
Training Epoch: 7/100, batch 81/188 completed (loss: 0.0862:  44%|████▎     | 41/94 [04:05<05:17,  5.99s/it]
Training Epoch: 7/100, batch 81/188 completed (loss: 0.1379:  44%|████▎     | 41/94 [04:05<05:17,  5.99s/it]
Training Epoch: 7/100, batch 82/188 completed (loss: 0.0959:  44%|████▎     | 41/94 [04:08<05:17,  5.99s/it]
Training Epoch: 7/100, batch 82/188 completed (loss: 0.3167:  44%|████▎     | 41/94 [04:08<05:17,  5.99s/it]
Training Epoch: 7/100, batch 82/188 completed (loss: 0.1323:  44%|████▎     | 41/94 [04:08<05:17,  5.99s/it]
Training Epoch: 7/100, batch 82/188 completed (loss: 0.3296:  44%|████▎     | 41/94 [04:08<05:17,  5.99s/it]
Training Epoch: 7/100, batch 82/188 completed (loss: 0.1323:  45%|████▍     | 42/94 [04:11<05:08,  5.94s/it]
Training Epoch: 7/100, batch 82/188 completed (loss: 0.0959:  45%|████▍     | 42/94 [04:11<05:09,  5.94s/it]
Training Epoch: 7/100, batch 82/188 completed (loss: 0.3296:  45%|████▍     | 42/94 [04:11<05:08,  5.94s/it]
Training Epoch: 7/100, batch 82/188 completed (loss: 0.3167:  45%|████▍     | 42/94 [04:11<05:09,  5.94s/it]
Training Epoch: 7/100, batch 83/188 completed (loss: 0.0677:  45%|████▍     | 42/94 [04:11<05:09,  5.94s/it]
Training Epoch: 7/100, batch 83/188 completed (loss: 0.1716:  45%|████▍     | 42/94 [04:11<05:08,  5.94s/it]
Training Epoch: 7/100, batch 83/188 completed (loss: 0.0964:  45%|████▍     | 42/94 [04:11<05:09,  5.94s/it]
Training Epoch: 7/100, batch 83/188 completed (loss: 0.3258:  45%|████▍     | 42/94 [04:11<05:08,  5.94s/it]
Training Epoch: 7/100, batch 84/188 completed (loss: 0.4343:  45%|████▍     | 42/94 [04:14<05:09,  5.94s/it]
Training Epoch: 7/100, batch 84/188 completed (loss: 0.0951:  45%|████▍     | 42/94 [04:14<05:08,  5.94s/it]
Training Epoch: 7/100, batch 84/188 completed (loss: 0.1735:  45%|████▍     | 42/94 [04:14<05:09,  5.94s/it]
Training Epoch: 7/100, batch 84/188 completed (loss: 0.4363:  45%|████▍     | 42/94 [04:14<05:08,  5.94s/it]
Training Epoch: 7/100, batch 84/188 completed (loss: 0.0951:  46%|████▌     | 43/94 [04:17<05:10,  6.08s/it]
Training Epoch: 7/100, batch 84/188 completed (loss: 0.4343:  46%|████▌     | 43/94 [04:17<05:10,  6.08s/it]
Training Epoch: 7/100, batch 84/188 completed (loss: 0.1735:  46%|████▌     | 43/94 [04:17<05:10,  6.08s/it]
Training Epoch: 7/100, batch 84/188 completed (loss: 0.4363:  46%|████▌     | 43/94 [04:17<05:10,  6.08s/it]
Training Epoch: 7/100, batch 85/188 completed (loss: 0.2501:  46%|████▌     | 43/94 [04:17<05:10,  6.08s/it]
Training Epoch: 7/100, batch 85/188 completed (loss: 0.1178:  46%|████▌     | 43/94 [04:17<05:10,  6.08s/it]
Training Epoch: 7/100, batch 85/188 completed (loss: 0.1881:  46%|████▌     | 43/94 [04:17<05:10,  6.08s/it]
Training Epoch: 7/100, batch 85/188 completed (loss: 0.2465:  46%|████▌     | 43/94 [04:17<05:10,  6.08s/it]
Training Epoch: 7/100, batch 86/188 completed (loss: 0.3919:  46%|████▌     | 43/94 [04:20<05:10,  6.08s/it]
Training Epoch: 7/100, batch 86/188 completed (loss: 0.3883:  46%|████▌     | 43/94 [04:20<05:10,  6.08s/it]
Training Epoch: 7/100, batch 86/188 completed (loss: 0.1398:  46%|████▌     | 43/94 [04:20<05:10,  6.08s/it]
Training Epoch: 7/100, batch 86/188 completed (loss: 0.2225:  46%|████▌     | 43/94 [04:20<05:10,  6.08s/it]
Training Epoch: 7/100, batch 86/188 completed (loss: 0.1398:  47%|████▋     | 44/94 [04:23<05:00,  6.02s/it]
Training Epoch: 7/100, batch 86/188 completed (loss: 0.3919:  47%|████▋     | 44/94 [04:23<05:00,  6.02s/it]
Training Epoch: 7/100, batch 86/188 completed (loss: 0.2225:  47%|████▋     | 44/94 [04:23<05:00,  6.02s/it]
Training Epoch: 7/100, batch 86/188 completed (loss: 0.3883:  47%|████▋     | 44/94 [04:23<05:00,  6.02s/it]
Training Epoch: 7/100, batch 87/188 completed (loss: 0.1133:  47%|████▋     | 44/94 [04:23<05:00,  6.02s/it]
Training Epoch: 7/100, batch 87/188 completed (loss: 0.2051:  47%|████▋     | 44/94 [04:23<05:00,  6.02s/it]
Training Epoch: 7/100, batch 87/188 completed (loss: 0.06:  47%|████▋     | 44/94 [04:23<05:00,  6.02s/it]  
Training Epoch: 7/100, batch 87/188 completed (loss: 0.2224:  47%|████▋     | 44/94 [04:23<05:00,  6.02s/it]
Training Epoch: 7/100, batch 88/188 completed (loss: 0.2599:  47%|████▋     | 44/94 [04:26<05:00,  6.02s/it]
Training Epoch: 7/100, batch 88/188 completed (loss: 0.0978:  47%|████▋     | 44/94 [04:26<05:00,  6.02s/it]
Training Epoch: 7/100, batch 88/188 completed (loss: 0.0677:  47%|████▋     | 44/94 [04:26<05:00,  6.02s/it]
Training Epoch: 7/100, batch 88/188 completed (loss: 0.2596:  47%|████▋     | 44/94 [04:26<05:00,  6.02s/it]
Training Epoch: 7/100, batch 88/188 completed (loss: 0.2596:  48%|████▊     | 45/94 [04:29<04:56,  6.05s/it]
Training Epoch: 7/100, batch 88/188 completed (loss: 0.2599:  48%|████▊     | 45/94 [04:29<04:56,  6.05s/it]
Training Epoch: 7/100, batch 88/188 completed (loss: 0.0677:  48%|████▊     | 45/94 [04:29<04:56,  6.05s/it]
Training Epoch: 7/100, batch 88/188 completed (loss: 0.0978:  48%|████▊     | 45/94 [04:29<04:56,  6.05s/it]
Training Epoch: 7/100, batch 89/188 completed (loss: 0.1221:  48%|████▊     | 45/94 [04:29<04:56,  6.05s/it]
Training Epoch: 7/100, batch 89/188 completed (loss: 0.1206:  48%|████▊     | 45/94 [04:29<04:56,  6.05s/it]
Training Epoch: 7/100, batch 89/188 completed (loss: 0.1905:  48%|████▊     | 45/94 [04:29<04:56,  6.05s/it]
Training Epoch: 7/100, batch 89/188 completed (loss: 0.0872:  48%|████▊     | 45/94 [04:29<04:56,  6.05s/it]
Training Epoch: 7/100, batch 90/188 completed (loss: 0.0722:  48%|████▊     | 45/94 [04:32<04:56,  6.05s/it]
Training Epoch: 7/100, batch 90/188 completed (loss: 0.1198:  48%|████▊     | 45/94 [04:32<04:56,  6.05s/it]
Training Epoch: 7/100, batch 90/188 completed (loss: 0.1461:  48%|████▊     | 45/94 [04:32<04:56,  6.05s/it]
Training Epoch: 7/100, batch 90/188 completed (loss: 0.1348:  48%|████▊     | 45/94 [04:32<04:56,  6.05s/it]
Training Epoch: 7/100, batch 90/188 completed (loss: 0.1198:  49%|████▉     | 46/94 [04:35<04:52,  6.09s/it]
Training Epoch: 7/100, batch 90/188 completed (loss: 0.1461:  49%|████▉     | 46/94 [04:35<04:52,  6.09s/it]
Training Epoch: 7/100, batch 90/188 completed (loss: 0.0722:  49%|████▉     | 46/94 [04:35<04:52,  6.09s/it]
Training Epoch: 7/100, batch 90/188 completed (loss: 0.1348:  49%|████▉     | 46/94 [04:35<04:52,  6.09s/it]
Training Epoch: 7/100, batch 91/188 completed (loss: 0.1951:  49%|████▉     | 46/94 [04:35<04:52,  6.09s/it]
Training Epoch: 7/100, batch 91/188 completed (loss: 0.5647:  49%|████▉     | 46/94 [04:35<04:52,  6.09s/it]
Training Epoch: 7/100, batch 91/188 completed (loss: 0.0931:  49%|████▉     | 46/94 [04:35<04:52,  6.09s/it]
Training Epoch: 7/100, batch 91/188 completed (loss: 0.1708:  49%|████▉     | 46/94 [04:35<04:52,  6.09s/it]
Training Epoch: 7/100, batch 92/188 completed (loss: 0.177:  49%|████▉     | 46/94 [04:38<04:52,  6.09s/it] 
Training Epoch: 7/100, batch 92/188 completed (loss: 0.0994:  49%|████▉     | 46/94 [04:38<04:52,  6.09s/it]
Training Epoch: 7/100, batch 92/188 completed (loss: 0.059:  49%|████▉     | 46/94 [04:38<04:52,  6.09s/it] 
Training Epoch: 7/100, batch 92/188 completed (loss: 0.089:  49%|████▉     | 46/94 [04:38<04:52,  6.09s/it] 
Training Epoch: 7/100, batch 92/188 completed (loss: 0.059:  50%|█████     | 47/94 [04:41<04:43,  6.03s/it]
Training Epoch: 7/100, batch 92/188 completed (loss: 0.177:  50%|█████     | 47/94 [04:41<04:43,  6.02s/it]
Training Epoch: 7/100, batch 92/188 completed (loss: 0.0994:  50%|█████     | 47/94 [04:41<04:43,  6.03s/it]
Training Epoch: 7/100, batch 92/188 completed (loss: 0.089:  50%|█████     | 47/94 [04:41<04:43,  6.03s/it]
Training Epoch: 7/100, batch 93/188 completed (loss: 0.1741:  50%|█████     | 47/94 [04:41<04:43,  6.03s/it]
Training Epoch: 7/100, batch 93/188 completed (loss: 0.1196:  50%|█████     | 47/94 [04:41<04:43,  6.03s/it]
Training Epoch: 7/100, batch 93/188 completed (loss: 0.0766:  50%|█████     | 47/94 [04:41<04:43,  6.02s/it]
Training Epoch: 7/100, batch 93/188 completed (loss: 0.1961:  50%|█████     | 47/94 [04:41<04:43,  6.03s/it]
Training Epoch: 7/100, batch 94/188 completed (loss: 0.206:  50%|█████     | 47/94 [04:44<04:43,  6.03s/it] 
Training Epoch: 7/100, batch 94/188 completed (loss: 0.1301:  50%|█████     | 47/94 [04:44<04:43,  6.02s/it]
Training Epoch: 7/100, batch 94/188 completed (loss: 0.138:  50%|█████     | 47/94 [04:44<04:43,  6.03s/it] 
Training Epoch: 7/100, batch 94/188 completed (loss: 0.0674:  50%|█████     | 47/94 [04:44<04:43,  6.03s/it]
Training Epoch: 7/100, batch 94/188 completed (loss: 0.1301:  51%|█████     | 48/94 [04:47<04:40,  6.10s/it]
Training Epoch: 7/100, batch 94/188 completed (loss: 0.138:  51%|█████     | 48/94 [04:47<04:40,  6.10s/it]
Training Epoch: 7/100, batch 94/188 completed (loss: 0.0674:  51%|█████     | 48/94 [04:47<04:40,  6.10s/it]
Training Epoch: 7/100, batch 94/188 completed (loss: 0.206:  51%|█████     | 48/94 [04:47<04:40,  6.10s/it]
Training Epoch: 7/100, batch 95/188 completed (loss: 0.0762:  51%|█████     | 48/94 [04:47<04:40,  6.10s/it]
Training Epoch: 7/100, batch 95/188 completed (loss: 0.1353:  51%|█████     | 48/94 [04:47<04:40,  6.10s/it]
Training Epoch: 7/100, batch 95/188 completed (loss: 0.0972:  51%|█████     | 48/94 [04:47<04:40,  6.10s/it]
Training Epoch: 7/100, batch 95/188 completed (loss: 0.0624:  51%|█████     | 48/94 [04:47<04:40,  6.10s/it]
Training Epoch: 7/100, batch 96/188 completed (loss: 0.1298:  51%|█████     | 48/94 [04:51<04:40,  6.10s/it]
Training Epoch: 7/100, batch 96/188 completed (loss: 0.1935:  51%|█████     | 48/94 [04:51<04:40,  6.10s/it]
Training Epoch: 7/100, batch 96/188 completed (loss: 0.5239:  51%|█████     | 48/94 [04:51<04:40,  6.10s/it]
Training Epoch: 7/100, batch 96/188 completed (loss: 0.1343:  51%|█████     | 48/94 [04:51<04:40,  6.10s/it]
Training Epoch: 7/100, batch 96/188 completed (loss: 0.5239:  52%|█████▏    | 49/94 [04:53<04:33,  6.08s/it]
Training Epoch: 7/100, batch 96/188 completed (loss: 0.1343:  52%|█████▏    | 49/94 [04:53<04:33,  6.08s/it]
Training Epoch: 7/100, batch 96/188 completed (loss: 0.1298:  52%|█████▏    | 49/94 [04:53<04:33,  6.08s/it]
Training Epoch: 7/100, batch 96/188 completed (loss: 0.1935:  52%|█████▏    | 49/94 [04:53<04:33,  6.09s/it]
Training Epoch: 7/100, batch 97/188 completed (loss: 0.0887:  52%|█████▏    | 49/94 [04:54<04:33,  6.08s/it]
Training Epoch: 7/100, batch 97/188 completed (loss: 0.1585:  52%|█████▏    | 49/94 [04:54<04:33,  6.08s/it]
Training Epoch: 7/100, batch 97/188 completed (loss: 0.3737:  52%|█████▏    | 49/94 [04:54<04:33,  6.09s/it]
Training Epoch: 7/100, batch 97/188 completed (loss: 0.2653:  52%|█████▏    | 49/94 [04:54<04:33,  6.08s/it]
Training Epoch: 7/100, batch 98/188 completed (loss: 0.2945:  52%|█████▏    | 49/94 [04:56<04:33,  6.09s/it]
Training Epoch: 7/100, batch 98/188 completed (loss: 0.0896:  52%|█████▏    | 49/94 [04:56<04:33,  6.08s/it]
Training Epoch: 7/100, batch 98/188 completed (loss: 0.029:  52%|█████▏    | 49/94 [04:56<04:33,  6.08s/it] 
Training Epoch: 7/100, batch 98/188 completed (loss: 0.0712:  52%|█████▏    | 49/94 [04:56<04:33,  6.08s/it]
Training Epoch: 7/100, batch 98/188 completed (loss: 0.0712:  53%|█████▎    | 50/94 [04:59<04:24,  6.01s/it]
Training Epoch: 7/100, batch 98/188 completed (loss: 0.029:  53%|█████▎    | 50/94 [04:59<04:24,  6.01s/it]
Training Epoch: 7/100, batch 98/188 completed (loss: 0.2945:  53%|█████▎    | 50/94 [04:59<04:24,  6.01s/it]
Training Epoch: 7/100, batch 98/188 completed (loss: 0.0896:  53%|█████▎    | 50/94 [04:59<04:24,  6.01s/it]
Training Epoch: 7/100, batch 99/188 completed (loss: 0.4093:  53%|█████▎    | 50/94 [04:59<04:24,  6.01s/it]
Training Epoch: 7/100, batch 99/188 completed (loss: 0.0474:  53%|█████▎    | 50/94 [04:59<04:24,  6.01s/it]
Training Epoch: 7/100, batch 99/188 completed (loss: 0.2074:  53%|█████▎    | 50/94 [04:59<04:24,  6.01s/it]
Training Epoch: 7/100, batch 99/188 completed (loss: 0.1987:  53%|█████▎    | 50/94 [04:59<04:24,  6.01s/it]
Training Epoch: 7/100, batch 100/188 completed (loss: 0.0538:  53%|█████▎    | 50/94 [05:03<04:24,  6.01s/it]
Training Epoch: 7/100, batch 100/188 completed (loss: 0.3311:  53%|█████▎    | 50/94 [05:03<04:24,  6.01s/it]
Training Epoch: 7/100, batch 100/188 completed (loss: 0.1538:  53%|█████▎    | 50/94 [05:03<04:24,  6.01s/it]
Training Epoch: 7/100, batch 100/188 completed (loss: 0.0659:  53%|█████▎    | 50/94 [05:03<04:24,  6.01s/it]
Training Epoch: 7/100, batch 100/188 completed (loss: 0.1538:  54%|█████▍    | 51/94 [05:06<04:22,  6.11s/it]
Training Epoch: 7/100, batch 100/188 completed (loss: 0.3311:  54%|█████▍    | 51/94 [05:06<04:22,  6.11s/it]
Training Epoch: 7/100, batch 100/188 completed (loss: 0.0659:  54%|█████▍    | 51/94 [05:06<04:22,  6.11s/it]
Training Epoch: 7/100, batch 100/188 completed (loss: 0.0538:  54%|█████▍    | 51/94 [05:06<04:22,  6.11s/it]
Training Epoch: 7/100, batch 101/188 completed (loss: 0.0371:  54%|█████▍    | 51/94 [05:06<04:22,  6.11s/it]
Training Epoch: 7/100, batch 101/188 completed (loss: 0.4086:  54%|█████▍    | 51/94 [05:06<04:22,  6.11s/it]
Training Epoch: 7/100, batch 101/188 completed (loss: 0.3077:  54%|█████▍    | 51/94 [05:06<04:22,  6.11s/it]
Training Epoch: 7/100, batch 101/188 completed (loss: 0.2632:  54%|█████▍    | 51/94 [05:06<04:22,  6.11s/it]
Training Epoch: 7/100, batch 102/188 completed (loss: 0.1432:  54%|█████▍    | 51/94 [05:09<04:22,  6.11s/it]
Training Epoch: 7/100, batch 102/188 completed (loss: 0.1369:  54%|█████▍    | 51/94 [05:09<04:22,  6.11s/it]
Training Epoch: 7/100, batch 102/188 completed (loss: 0.1943:  54%|█████▍    | 51/94 [05:09<04:22,  6.11s/it]
Training Epoch: 7/100, batch 102/188 completed (loss: 0.1364:  54%|█████▍    | 51/94 [05:09<04:22,  6.11s/it]
Training Epoch: 7/100, batch 102/188 completed (loss: 0.1432:  55%|█████▌    | 52/94 [05:11<04:13,  6.02s/it]
Training Epoch: 7/100, batch 102/188 completed (loss: 0.1943:  55%|█████▌    | 52/94 [05:11<04:13,  6.02s/it]
Training Epoch: 7/100, batch 102/188 completed (loss: 0.1364:  55%|█████▌    | 52/94 [05:11<04:13,  6.03s/it]
Training Epoch: 7/100, batch 102/188 completed (loss: 0.1369:  55%|█████▌    | 52/94 [05:11<04:13,  6.02s/it]
Training Epoch: 7/100, batch 103/188 completed (loss: 0.3796:  55%|█████▌    | 52/94 [05:12<04:13,  6.02s/it]
Training Epoch: 7/100, batch 103/188 completed (loss: 0.097:  55%|█████▌    | 52/94 [05:12<04:13,  6.02s/it] 
Training Epoch: 7/100, batch 103/188 completed (loss: 0.0337:  55%|█████▌    | 52/94 [05:12<04:13,  6.02s/it]
Training Epoch: 7/100, batch 103/188 completed (loss: 0.3566:  55%|█████▌    | 52/94 [05:12<04:13,  6.03s/it]
Training Epoch: 7/100, batch 104/188 completed (loss: 0.3149:  55%|█████▌    | 52/94 [05:14<04:13,  6.02s/it]
Training Epoch: 7/100, batch 104/188 completed (loss: 0.2487:  55%|█████▌    | 52/94 [05:14<04:13,  6.02s/it]
Training Epoch: 7/100, batch 104/188 completed (loss: 0.1203:  55%|█████▌    | 52/94 [05:14<04:13,  6.02s/it]
Training Epoch: 7/100, batch 104/188 completed (loss: 0.1801:  55%|█████▌    | 52/94 [05:14<04:13,  6.03s/it]
Training Epoch: 7/100, batch 104/188 completed (loss: 0.2487:  56%|█████▋    | 53/94 [05:18<04:08,  6.06s/it]
Training Epoch: 7/100, batch 104/188 completed (loss: 0.1203:  56%|█████▋    | 53/94 [05:18<04:08,  6.06s/it]
Training Epoch: 7/100, batch 104/188 completed (loss: 0.1801:  56%|█████▋    | 53/94 [05:18<04:08,  6.06s/it]
Training Epoch: 7/100, batch 104/188 completed (loss: 0.3149:  56%|█████▋    | 53/94 [05:18<04:08,  6.06s/it]
Training Epoch: 7/100, batch 105/188 completed (loss: 0.1576:  56%|█████▋    | 53/94 [05:18<04:08,  6.06s/it]
Training Epoch: 7/100, batch 105/188 completed (loss: 0.5116:  56%|█████▋    | 53/94 [05:18<04:08,  6.06s/it]
Training Epoch: 7/100, batch 105/188 completed (loss: 0.0976:  56%|█████▋    | 53/94 [05:18<04:08,  6.06s/it]
Training Epoch: 7/100, batch 105/188 completed (loss: 0.3025:  56%|█████▋    | 53/94 [05:18<04:08,  6.06s/it]
Training Epoch: 7/100, batch 106/188 completed (loss: 0.082:  56%|█████▋    | 53/94 [05:21<04:08,  6.06s/it] 
Training Epoch: 7/100, batch 106/188 completed (loss: 0.0884:  56%|█████▋    | 53/94 [05:21<04:08,  6.06s/it]
Training Epoch: 7/100, batch 106/188 completed (loss: 0.079:  56%|█████▋    | 53/94 [05:21<04:08,  6.06s/it] 
Training Epoch: 7/100, batch 106/188 completed (loss: 0.1481:  56%|█████▋    | 53/94 [05:21<04:08,  6.06s/it]
Training Epoch: 7/100, batch 106/188 completed (loss: 0.079:  57%|█████▋    | 54/94 [05:24<04:01,  6.05s/it]
Training Epoch: 7/100, batch 106/188 completed (loss: 0.082:  57%|█████▋    | 54/94 [05:24<04:01,  6.05s/it]
Training Epoch: 7/100, batch 106/188 completed (loss: 0.1481:  57%|█████▋    | 54/94 [05:24<04:02,  6.05s/it]
Training Epoch: 7/100, batch 106/188 completed (loss: 0.0884:  57%|█████▋    | 54/94 [05:24<04:02,  6.05s/it]
Training Epoch: 7/100, batch 107/188 completed (loss: 0.4182:  57%|█████▋    | 54/94 [05:24<04:01,  6.05s/it]
Training Epoch: 7/100, batch 107/188 completed (loss: 0.221:  57%|█████▋    | 54/94 [05:24<04:01,  6.05s/it]
Training Epoch: 7/100, batch 107/188 completed (loss: 0.2496:  57%|█████▋    | 54/94 [05:24<04:02,  6.05s/it]
Training Epoch: 7/100, batch 107/188 completed (loss: 0.1491:  57%|█████▋    | 54/94 [05:24<04:02,  6.05s/it]
Training Epoch: 7/100, batch 108/188 completed (loss: 0.2289:  57%|█████▋    | 54/94 [05:27<04:02,  6.05s/it]
Training Epoch: 7/100, batch 108/188 completed (loss: 0.2298:  57%|█████▋    | 54/94 [05:27<04:01,  6.05s/it]
Training Epoch: 7/100, batch 108/188 completed (loss: 0.2642:  57%|█████▋    | 54/94 [05:27<04:01,  6.05s/it]
Training Epoch: 7/100, batch 108/188 completed (loss: 0.2362:  57%|█████▋    | 54/94 [05:27<04:02,  6.05s/it]
Training Epoch: 7/100, batch 108/188 completed (loss: 0.2298:  59%|█████▊    | 55/94 [05:30<03:53,  6.00s/it]
Training Epoch: 7/100, batch 108/188 completed (loss: 0.2289:  59%|█████▊    | 55/94 [05:30<03:53,  6.00s/it]
Training Epoch: 7/100, batch 108/188 completed (loss: 0.2362:  59%|█████▊    | 55/94 [05:30<03:53,  6.00s/it]
Training Epoch: 7/100, batch 108/188 completed (loss: 0.2642:  59%|█████▊    | 55/94 [05:30<03:53,  6.00s/it]
Training Epoch: 7/100, batch 109/188 completed (loss: 0.3329:  59%|█████▊    | 55/94 [05:30<03:53,  6.00s/it]
Training Epoch: 7/100, batch 109/188 completed (loss: 0.1027:  59%|█████▊    | 55/94 [05:30<03:53,  6.00s/it]
Training Epoch: 7/100, batch 109/188 completed (loss: 0.199:  59%|█████▊    | 55/94 [05:30<03:53,  6.00s/it] 
Training Epoch: 7/100, batch 109/188 completed (loss: 0.1808:  59%|█████▊    | 55/94 [05:30<03:53,  6.00s/it]
Training Epoch: 7/100, batch 110/188 completed (loss: 0.2612:  59%|█████▊    | 55/94 [05:33<03:53,  6.00s/it]
Training Epoch: 7/100, batch 110/188 completed (loss: 0.1511:  59%|█████▊    | 55/94 [05:33<03:53,  6.00s/it]
Training Epoch: 7/100, batch 110/188 completed (loss: 0.4464:  59%|█████▊    | 55/94 [05:33<03:53,  6.00s/it]
Training Epoch: 7/100, batch 110/188 completed (loss: 0.1357:  59%|█████▊    | 55/94 [05:33<03:53,  6.00s/it]
Training Epoch: 7/100, batch 110/188 completed (loss: 0.1357:  60%|█████▉    | 56/94 [05:36<03:51,  6.09s/it]
Training Epoch: 7/100, batch 110/188 completed (loss: 0.4464:  60%|█████▉    | 56/94 [05:36<03:51,  6.09s/it]
Training Epoch: 7/100, batch 110/188 completed (loss: 0.1511:  60%|█████▉    | 56/94 [05:36<03:51,  6.10s/it]
Training Epoch: 7/100, batch 110/188 completed (loss: 0.2612:  60%|█████▉    | 56/94 [05:36<03:51,  6.10s/it]
Training Epoch: 7/100, batch 111/188 completed (loss: 0.2349:  60%|█████▉    | 56/94 [05:36<03:51,  6.10s/it]
Training Epoch: 7/100, batch 111/188 completed (loss: 0.1835:  60%|█████▉    | 56/94 [05:36<03:51,  6.10s/it]
Training Epoch: 7/100, batch 111/188 completed (loss: 0.0354:  60%|█████▉    | 56/94 [05:36<03:51,  6.09s/it]
Training Epoch: 7/100, batch 111/188 completed (loss: 0.2126:  60%|█████▉    | 56/94 [05:36<03:51,  6.09s/it]
Training Epoch: 7/100, batch 112/188 completed (loss: 0.1029:  60%|█████▉    | 56/94 [05:39<03:51,  6.09s/it]
Training Epoch: 7/100, batch 112/188 completed (loss: 0.341:  60%|█████▉    | 56/94 [05:39<03:51,  6.10s/it] 
Training Epoch: 7/100, batch 112/188 completed (loss: 0.5157:  60%|█████▉    | 56/94 [05:39<03:51,  6.10s/it]
Training Epoch: 7/100, batch 112/188 completed (loss: 0.0421:  60%|█████▉    | 56/94 [05:39<03:51,  6.09s/it]
Training Epoch: 7/100, batch 112/188 completed (loss: 0.5157:  61%|██████    | 57/94 [05:42<03:44,  6.07s/it]
Training Epoch: 7/100, batch 112/188 completed (loss: 0.0421:  61%|██████    | 57/94 [05:42<03:44,  6.07s/it]
Training Epoch: 7/100, batch 112/188 completed (loss: 0.1029:  61%|██████    | 57/94 [05:42<03:44,  6.08s/it]
Training Epoch: 7/100, batch 112/188 completed (loss: 0.341:  61%|██████    | 57/94 [05:42<03:44,  6.07s/it]
Training Epoch: 7/100, batch 113/188 completed (loss: 0.206:  61%|██████    | 57/94 [05:42<03:44,  6.07s/it]
Training Epoch: 7/100, batch 113/188 completed (loss: 0.0757:  61%|██████    | 57/94 [05:42<03:44,  6.07s/it]
Training Epoch: 7/100, batch 113/188 completed (loss: 0.0865:  61%|██████    | 57/94 [05:42<03:44,  6.08s/it]
Training Epoch: 7/100, batch 113/188 completed (loss: 0.1189:  61%|██████    | 57/94 [05:42<03:44,  6.07s/it]
Training Epoch: 7/100, batch 114/188 completed (loss: 0.3156:  61%|██████    | 57/94 [05:45<03:44,  6.07s/it]
Training Epoch: 7/100, batch 114/188 completed (loss: 0.1822:  61%|██████    | 57/94 [05:45<03:44,  6.08s/it]
Training Epoch: 7/100, batch 114/188 completed (loss: 0.2456:  61%|██████    | 57/94 [05:45<03:44,  6.07s/it]
Training Epoch: 7/100, batch 114/188 completed (loss: 0.0837:  61%|██████    | 57/94 [05:45<03:44,  6.07s/it]
Training Epoch: 7/100, batch 114/188 completed (loss: 0.3156:  62%|██████▏   | 58/94 [05:48<03:37,  6.04s/it]
Training Epoch: 7/100, batch 114/188 completed (loss: 0.2456:  62%|██████▏   | 58/94 [05:48<03:37,  6.04s/it]
Training Epoch: 7/100, batch 114/188 completed (loss: 0.1822:  62%|██████▏   | 58/94 [05:48<03:37,  6.04s/it]
Training Epoch: 7/100, batch 114/188 completed (loss: 0.0837:  62%|██████▏   | 58/94 [05:48<03:37,  6.04s/it]
Training Epoch: 7/100, batch 115/188 completed (loss: 0.3243:  62%|██████▏   | 58/94 [05:48<03:37,  6.04s/it]
Training Epoch: 7/100, batch 115/188 completed (loss: 0.1359:  62%|██████▏   | 58/94 [05:48<03:37,  6.04s/it]
Training Epoch: 7/100, batch 115/188 completed (loss: 0.158:  62%|██████▏   | 58/94 [05:48<03:37,  6.04s/it] 
Training Epoch: 7/100, batch 115/188 completed (loss: 0.0845:  62%|██████▏   | 58/94 [05:48<03:37,  6.04s/it]
Training Epoch: 7/100, batch 116/188 completed (loss: 0.2129:  62%|██████▏   | 58/94 [05:51<03:37,  6.04s/it]
Training Epoch: 7/100, batch 116/188 completed (loss: 0.1303:  62%|██████▏   | 58/94 [05:51<03:37,  6.04s/it]
Training Epoch: 7/100, batch 116/188 completed (loss: 0.2278:  62%|██████▏   | 58/94 [05:51<03:37,  6.04s/it]
Training Epoch: 7/100, batch 116/188 completed (loss: 0.3679:  62%|██████▏   | 58/94 [05:51<03:37,  6.04s/it]
Training Epoch: 7/100, batch 116/188 completed (loss: 0.2129:  63%|██████▎   | 59/94 [05:54<03:33,  6.09s/it]
Training Epoch: 7/100, batch 116/188 completed (loss: 0.2278:  63%|██████▎   | 59/94 [05:54<03:33,  6.09s/it]
Training Epoch: 7/100, batch 116/188 completed (loss: 0.1303:  63%|██████▎   | 59/94 [05:54<03:33,  6.09s/it]
Training Epoch: 7/100, batch 116/188 completed (loss: 0.3679:  63%|██████▎   | 59/94 [05:54<03:33,  6.09s/it]
Training Epoch: 7/100, batch 117/188 completed (loss: 0.0938:  63%|██████▎   | 59/94 [05:54<03:33,  6.09s/it]
Training Epoch: 7/100, batch 117/188 completed (loss: 0.1627:  63%|██████▎   | 59/94 [05:54<03:33,  6.09s/it]
Training Epoch: 7/100, batch 117/188 completed (loss: 0.1475:  63%|██████▎   | 59/94 [05:54<03:33,  6.09s/it]
Training Epoch: 7/100, batch 117/188 completed (loss: 0.0929:  63%|██████▎   | 59/94 [05:54<03:33,  6.09s/it]
Training Epoch: 7/100, batch 118/188 completed (loss: 0.2068:  63%|██████▎   | 59/94 [05:57<03:33,  6.09s/it]
Training Epoch: 7/100, batch 118/188 completed (loss: 0.0856:  63%|██████▎   | 59/94 [05:57<03:33,  6.09s/it]
Training Epoch: 7/100, batch 118/188 completed (loss: 0.1251:  63%|██████▎   | 59/94 [05:57<03:33,  6.09s/it]
Training Epoch: 7/100, batch 118/188 completed (loss: 0.125:  63%|██████▎   | 59/94 [05:57<03:33,  6.09s/it] 
Training Epoch: 7/100, batch 118/188 completed (loss: 0.2068:  64%|██████▍   | 60/94 [06:00<03:24,  6.01s/it]
Training Epoch: 7/100, batch 118/188 completed (loss: 0.0856:  64%|██████▍   | 60/94 [06:00<03:24,  6.01s/it]
Training Epoch: 7/100, batch 118/188 completed (loss: 0.125:  64%|██████▍   | 60/94 [06:00<03:24,  6.01s/it]
Training Epoch: 7/100, batch 118/188 completed (loss: 0.1251:  64%|██████▍   | 60/94 [06:00<03:24,  6.01s/it]
Training Epoch: 7/100, batch 119/188 completed (loss: 0.3131:  64%|██████▍   | 60/94 [06:00<03:24,  6.01s/it]
Training Epoch: 7/100, batch 119/188 completed (loss: 0.5016:  64%|██████▍   | 60/94 [06:00<03:24,  6.01s/it]
Training Epoch: 7/100, batch 119/188 completed (loss: 0.4221:  64%|██████▍   | 60/94 [06:00<03:24,  6.01s/it]
Training Epoch: 7/100, batch 119/188 completed (loss: 0.2382:  64%|██████▍   | 60/94 [06:00<03:24,  6.01s/it]
Training Epoch: 7/100, batch 120/188 completed (loss: 0.2823:  64%|██████▍   | 60/94 [06:03<03:24,  6.01s/it]
Training Epoch: 7/100, batch 120/188 completed (loss: 0.1324:  64%|██████▍   | 60/94 [06:03<03:24,  6.01s/it]
Training Epoch: 7/100, batch 120/188 completed (loss: 0.065:  64%|██████▍   | 60/94 [06:03<03:24,  6.01s/it] 
Training Epoch: 7/100, batch 120/188 completed (loss: 0.175:  64%|██████▍   | 60/94 [06:03<03:24,  6.01s/it] 
Training Epoch: 7/100, batch 120/188 completed (loss: 0.2823:  65%|██████▍   | 61/94 [06:06<03:20,  6.09s/it]
Training Epoch: 7/100, batch 120/188 completed (loss: 0.175:  65%|██████▍   | 61/94 [06:06<03:20,  6.09s/it]
Training Epoch: 7/100, batch 120/188 completed (loss: 0.1324:  65%|██████▍   | 61/94 [06:06<03:20,  6.09s/it]
Training Epoch: 7/100, batch 120/188 completed (loss: 0.065:  65%|██████▍   | 61/94 [06:06<03:21,  6.09s/it]
Training Epoch: 7/100, batch 121/188 completed (loss: 0.2114:  65%|██████▍   | 61/94 [06:06<03:21,  6.09s/it]
Training Epoch: 7/100, batch 121/188 completed (loss: 0.0609:  65%|██████▍   | 61/94 [06:06<03:20,  6.09s/it]
Training Epoch: 7/100, batch 121/188 completed (loss: 0.1583:  65%|██████▍   | 61/94 [06:06<03:20,  6.09s/it]
Training Epoch: 7/100, batch 121/188 completed (loss: 0.1511:  65%|██████▍   | 61/94 [06:06<03:20,  6.09s/it]
Training Epoch: 7/100, batch 122/188 completed (loss: 0.3214:  65%|██████▍   | 61/94 [06:09<03:21,  6.09s/it]
Training Epoch: 7/100, batch 122/188 completed (loss: 0.1911:  65%|██████▍   | 61/94 [06:09<03:20,  6.09s/it]
Training Epoch: 7/100, batch 122/188 completed (loss: 0.2963:  65%|██████▍   | 61/94 [06:09<03:20,  6.09s/it]
Training Epoch: 7/100, batch 122/188 completed (loss: 0.1135:  65%|██████▍   | 61/94 [06:09<03:20,  6.09s/it]
Training Epoch: 7/100, batch 122/188 completed (loss: 0.1135:  66%|██████▌   | 62/94 [06:12<03:14,  6.07s/it]
Training Epoch: 7/100, batch 122/188 completed (loss: 0.2963:  66%|██████▌   | 62/94 [06:12<03:14,  6.07s/it]
Training Epoch: 7/100, batch 122/188 completed (loss: 0.1911:  66%|██████▌   | 62/94 [06:12<03:14,  6.07s/it]
Training Epoch: 7/100, batch 122/188 completed (loss: 0.3214:  66%|██████▌   | 62/94 [06:12<03:14,  6.07s/it]
Training Epoch: 7/100, batch 123/188 completed (loss: 0.196:  66%|██████▌   | 62/94 [06:12<03:14,  6.07s/it] 
Training Epoch: 7/100, batch 123/188 completed (loss: 0.0324:  66%|██████▌   | 62/94 [06:12<03:14,  6.07s/it]
Training Epoch: 7/100, batch 123/188 completed (loss: 0.1558:  66%|██████▌   | 62/94 [06:12<03:14,  6.07s/it]
Training Epoch: 7/100, batch 123/188 completed (loss: 0.2751:  66%|██████▌   | 62/94 [06:12<03:14,  6.07s/it]
Training Epoch: 7/100, batch 124/188 completed (loss: 0.0619:  66%|██████▌   | 62/94 [06:15<03:14,  6.07s/it]
Training Epoch: 7/100, batch 124/188 completed (loss: 0.0879:  66%|██████▌   | 62/94 [06:15<03:14,  6.07s/it]
Training Epoch: 7/100, batch 124/188 completed (loss: 0.0772:  66%|██████▌   | 62/94 [06:15<03:14,  6.07s/it]
Training Epoch: 7/100, batch 124/188 completed (loss: 0.1025:  66%|██████▌   | 62/94 [06:15<03:14,  6.07s/it]
Training Epoch: 7/100, batch 124/188 completed (loss: 0.0772:  67%|██████▋   | 63/94 [06:18<03:07,  6.05s/it]
Training Epoch: 7/100, batch 124/188 completed (loss: 0.0619:  67%|██████▋   | 63/94 [06:18<03:07,  6.05s/it]
Training Epoch: 7/100, batch 124/188 completed (loss: 0.1025:  67%|██████▋   | 63/94 [06:18<03:07,  6.05s/it]
Training Epoch: 7/100, batch 124/188 completed (loss: 0.0879:  67%|██████▋   | 63/94 [06:18<03:07,  6.05s/it]
Training Epoch: 7/100, batch 125/188 completed (loss: 0.2126:  67%|██████▋   | 63/94 [06:18<03:07,  6.05s/it]
Training Epoch: 7/100, batch 125/188 completed (loss: 0.1623:  67%|██████▋   | 63/94 [06:18<03:07,  6.05s/it]
Training Epoch: 7/100, batch 125/188 completed (loss: 0.2089:  67%|██████▋   | 63/94 [06:18<03:07,  6.05s/it]
Training Epoch: 7/100, batch 125/188 completed (loss: 0.2446:  67%|██████▋   | 63/94 [06:18<03:07,  6.05s/it]
Training Epoch: 7/100, batch 126/188 completed (loss: 0.2317:  67%|██████▋   | 63/94 [06:21<03:07,  6.05s/it]
Training Epoch: 7/100, batch 126/188 completed (loss: 0.0906:  67%|██████▋   | 63/94 [06:21<03:07,  6.05s/it]
Training Epoch: 7/100, batch 126/188 completed (loss: 0.1364:  67%|██████▋   | 63/94 [06:21<03:07,  6.05s/it]
Training Epoch: 7/100, batch 126/188 completed (loss: 0.2157:  67%|██████▋   | 63/94 [06:21<03:07,  6.05s/it]
Training Epoch: 7/100, batch 126/188 completed (loss: 0.2157:  68%|██████▊   | 64/94 [06:24<03:02,  6.07s/it]
Training Epoch: 7/100, batch 126/188 completed (loss: 0.2317:  68%|██████▊   | 64/94 [06:24<03:02,  6.08s/it]
Training Epoch: 7/100, batch 126/188 completed (loss: 0.0906:  68%|██████▊   | 64/94 [06:24<03:02,  6.08s/it]
Training Epoch: 7/100, batch 126/188 completed (loss: 0.1364:  68%|██████▊   | 64/94 [06:24<03:02,  6.08s/it]
Training Epoch: 7/100, batch 127/188 completed (loss: 0.0882:  68%|██████▊   | 64/94 [06:24<03:02,  6.08s/it]
Training Epoch: 7/100, batch 127/188 completed (loss: 0.2961:  68%|██████▊   | 64/94 [06:24<03:02,  6.08s/it]
Training Epoch: 7/100, batch 127/188 completed (loss: 0.2844:  68%|██████▊   | 64/94 [06:24<03:02,  6.08s/it]
Training Epoch: 7/100, batch 127/188 completed (loss: 0.1997:  68%|██████▊   | 64/94 [06:24<03:02,  6.07s/it]
Training Epoch: 7/100, batch 128/188 completed (loss: 0.0781:  68%|██████▊   | 64/94 [06:27<03:02,  6.08s/it]
Training Epoch: 7/100, batch 128/188 completed (loss: 0.117:  68%|██████▊   | 64/94 [06:27<03:02,  6.08s/it] 
Training Epoch: 7/100, batch 128/188 completed (loss: 0.2536:  68%|██████▊   | 64/94 [06:27<03:02,  6.08s/it]
Training Epoch: 7/100, batch 128/188 completed (loss: 0.0605:  68%|██████▊   | 64/94 [06:27<03:02,  6.07s/it]
Training Epoch: 7/100, batch 128/188 completed (loss: 0.0781:  69%|██████▉   | 65/94 [06:30<02:55,  6.05s/it]
Training Epoch: 7/100, batch 128/188 completed (loss: 0.117:  69%|██████▉   | 65/94 [06:30<02:55,  6.06s/it]
Training Epoch: 7/100, batch 128/188 completed (loss: 0.2536:  69%|██████▉   | 65/94 [06:30<02:55,  6.06s/it]
Training Epoch: 7/100, batch 128/188 completed (loss: 0.0605:  69%|██████▉   | 65/94 [06:30<02:55,  6.06s/it]
Training Epoch: 7/100, batch 129/188 completed (loss: 0.1894:  69%|██████▉   | 65/94 [06:30<02:55,  6.05s/it]
Training Epoch: 7/100, batch 129/188 completed (loss: 0.0869:  69%|██████▉   | 65/94 [06:30<02:55,  6.06s/it]
Training Epoch: 7/100, batch 129/188 completed (loss: 0.2064:  69%|██████▉   | 65/94 [06:30<02:55,  6.06s/it]
Training Epoch: 7/100, batch 129/188 completed (loss: 0.1454:  69%|██████▉   | 65/94 [06:30<02:55,  6.06s/it]
Training Epoch: 7/100, batch 130/188 completed (loss: 0.0391:  69%|██████▉   | 65/94 [06:33<02:55,  6.06s/it]
Training Epoch: 7/100, batch 130/188 completed (loss: 0.1771:  69%|██████▉   | 65/94 [06:33<02:55,  6.05s/it]
Training Epoch: 7/100, batch 130/188 completed (loss: 0.1489:  69%|██████▉   | 65/94 [06:33<02:55,  6.06s/it]
Training Epoch: 7/100, batch 130/188 completed (loss: 0.1596:  69%|██████▉   | 65/94 [06:33<02:55,  6.06s/it]
Training Epoch: 7/100, batch 130/188 completed (loss: 0.1489:  70%|███████   | 66/94 [06:36<02:50,  6.09s/it]
Training Epoch: 7/100, batch 130/188 completed (loss: 0.1596:  70%|███████   | 66/94 [06:37<02:50,  6.10s/it]
Training Epoch: 7/100, batch 130/188 completed (loss: 0.0391:  70%|███████   | 66/94 [06:36<02:50,  6.10s/it]
Training Epoch: 7/100, batch 130/188 completed (loss: 0.1771:  70%|███████   | 66/94 [06:36<02:50,  6.10s/it]
Training Epoch: 7/100, batch 131/188 completed (loss: 0.3904:  70%|███████   | 66/94 [06:37<02:50,  6.10s/it]
Training Epoch: 7/100, batch 131/188 completed (loss: 0.1767:  70%|███████   | 66/94 [06:37<02:50,  6.09s/it]
Training Epoch: 7/100, batch 131/188 completed (loss: 0.3821:  70%|███████   | 66/94 [06:37<02:50,  6.10s/it]
Training Epoch: 7/100, batch 131/188 completed (loss: 0.1483:  70%|███████   | 66/94 [06:37<02:50,  6.10s/it]
Training Epoch: 7/100, batch 132/188 completed (loss: 0.2069:  70%|███████   | 66/94 [06:39<02:50,  6.09s/it]
Training Epoch: 7/100, batch 132/188 completed (loss: 0.0456:  70%|███████   | 66/94 [06:39<02:50,  6.10s/it]
Training Epoch: 7/100, batch 132/188 completed (loss: 0.1245:  70%|███████   | 66/94 [06:39<02:50,  6.10s/it]
Training Epoch: 7/100, batch 132/188 completed (loss: 0.2482:  70%|███████   | 66/94 [06:39<02:50,  6.10s/it]
Training Epoch: 7/100, batch 132/188 completed (loss: 0.2069:  71%|███████▏  | 67/94 [06:43<02:44,  6.08s/it]
Training Epoch: 7/100, batch 132/188 completed (loss: 0.2482:  71%|███████▏  | 67/94 [06:43<02:44,  6.08s/it]
Training Epoch: 7/100, batch 132/188 completed (loss: 0.0456:  71%|███████▏  | 67/94 [06:43<02:44,  6.08s/it]
Training Epoch: 7/100, batch 132/188 completed (loss: 0.1245:  71%|███████▏  | 67/94 [06:43<02:44,  6.08s/it]
Training Epoch: 7/100, batch 133/188 completed (loss: 0.0442:  71%|███████▏  | 67/94 [06:43<02:44,  6.08s/it]
Training Epoch: 7/100, batch 133/188 completed (loss: 0.2153:  71%|███████▏  | 67/94 [06:43<02:44,  6.08s/it]
Training Epoch: 7/100, batch 133/188 completed (loss: 0.1007:  71%|███████▏  | 67/94 [06:43<02:44,  6.08s/it]
Training Epoch: 7/100, batch 133/188 completed (loss: 0.2102:  71%|███████▏  | 67/94 [06:43<02:44,  6.08s/it]
Training Epoch: 7/100, batch 134/188 completed (loss: 0.2059:  71%|███████▏  | 67/94 [06:45<02:44,  6.08s/it]
Training Epoch: 7/100, batch 134/188 completed (loss: 0.0619:  71%|███████▏  | 67/94 [06:45<02:44,  6.08s/it]
Training Epoch: 7/100, batch 134/188 completed (loss: 0.2152:  71%|███████▏  | 67/94 [06:45<02:44,  6.08s/it]
Training Epoch: 7/100, batch 134/188 completed (loss: 0.1195:  71%|███████▏  | 67/94 [06:46<02:44,  6.08s/it]
Training Epoch: 7/100, batch 134/188 completed (loss: 0.1195:  72%|███████▏  | 68/94 [06:48<02:36,  6.01s/it]
Training Epoch: 7/100, batch 134/188 completed (loss: 0.2152:  72%|███████▏  | 68/94 [06:48<02:36,  6.01s/it]
Training Epoch: 7/100, batch 134/188 completed (loss: 0.0619:  72%|███████▏  | 68/94 [06:48<02:36,  6.01s/it]
Training Epoch: 7/100, batch 134/188 completed (loss: 0.2059:  72%|███████▏  | 68/94 [06:48<02:36,  6.01s/it]
Training Epoch: 7/100, batch 135/188 completed (loss: 0.0496:  72%|███████▏  | 68/94 [06:48<02:36,  6.01s/it]
Training Epoch: 7/100, batch 135/188 completed (loss: 0.5445:  72%|███████▏  | 68/94 [06:48<02:36,  6.01s/it]
Training Epoch: 7/100, batch 135/188 completed (loss: 0.4328:  72%|███████▏  | 68/94 [06:48<02:36,  6.01s/it]
Training Epoch: 7/100, batch 135/188 completed (loss: 0.337:  72%|███████▏  | 68/94 [06:48<02:36,  6.01s/it] 
Training Epoch: 7/100, batch 136/188 completed (loss: 0.1441:  72%|███████▏  | 68/94 [06:51<02:36,  6.01s/it]
Training Epoch: 7/100, batch 136/188 completed (loss: 0.2833:  72%|███████▏  | 68/94 [06:51<02:36,  6.01s/it]
Training Epoch: 7/100, batch 136/188 completed (loss: 0.2154:  72%|███████▏  | 68/94 [06:51<02:36,  6.01s/it]
Training Epoch: 7/100, batch 136/188 completed (loss: 0.2524:  72%|███████▏  | 68/94 [06:52<02:36,  6.01s/it]
Training Epoch: 7/100, batch 136/188 completed (loss: 0.2833:  73%|███████▎  | 69/94 [06:55<02:32,  6.09s/it]
Training Epoch: 7/100, batch 136/188 completed (loss: 0.2154:  73%|███████▎  | 69/94 [06:55<02:32,  6.09s/it]
Training Epoch: 7/100, batch 136/188 completed (loss: 0.1441:  73%|███████▎  | 69/94 [06:55<02:32,  6.10s/it]
Training Epoch: 7/100, batch 136/188 completed (loss: 0.2524:  73%|███████▎  | 69/94 [06:55<02:32,  6.10s/it]
Training Epoch: 7/100, batch 137/188 completed (loss: 0.1179:  73%|███████▎  | 69/94 [06:55<02:32,  6.09s/it]
Training Epoch: 7/100, batch 137/188 completed (loss: 0.2949:  73%|███████▎  | 69/94 [06:55<02:32,  6.09s/it]
Training Epoch: 7/100, batch 137/188 completed (loss: 0.2668:  73%|███████▎  | 69/94 [06:55<02:32,  6.10s/it]
Training Epoch: 7/100, batch 137/188 completed (loss: 0.3154:  73%|███████▎  | 69/94 [06:55<02:32,  6.10s/it]
Training Epoch: 7/100, batch 138/188 completed (loss: 0.1649:  73%|███████▎  | 69/94 [06:58<02:32,  6.10s/it]
Training Epoch: 7/100, batch 138/188 completed (loss: 0.1397:  73%|███████▎  | 69/94 [06:58<02:32,  6.09s/it]
Training Epoch: 7/100, batch 138/188 completed (loss: 0.136:  73%|███████▎  | 69/94 [06:58<02:32,  6.09s/it] 
Training Epoch: 7/100, batch 138/188 completed (loss: 0.414:  73%|███████▎  | 69/94 [06:58<02:32,  6.10s/it] 
Training Epoch: 7/100, batch 138/188 completed (loss: 0.136:  74%|███████▍  | 70/94 [07:01<02:25,  6.07s/it]
Training Epoch: 7/100, batch 138/188 completed (loss: 0.414:  74%|███████▍  | 70/94 [07:01<02:25,  6.07s/it]
Training Epoch: 7/100, batch 138/188 completed (loss: 0.1397:  74%|███████▍  | 70/94 [07:01<02:25,  6.07s/it]
Training Epoch: 7/100, batch 138/188 completed (loss: 0.1649:  74%|███████▍  | 70/94 [07:01<02:25,  6.07s/it]
Training Epoch: 7/100, batch 139/188 completed (loss: 0.2078:  74%|███████▍  | 70/94 [07:01<02:25,  6.07s/it]
Training Epoch: 7/100, batch 139/188 completed (loss: 0.1318:  74%|███████▍  | 70/94 [07:01<02:25,  6.07s/it]
Training Epoch: 7/100, batch 139/188 completed (loss: 0.214:  74%|███████▍  | 70/94 [07:01<02:25,  6.07s/it] 
Training Epoch: 7/100, batch 139/188 completed (loss: 0.2015:  74%|███████▍  | 70/94 [07:01<02:25,  6.07s/it]
Training Epoch: 7/100, batch 140/188 completed (loss: 0.1399:  74%|███████▍  | 70/94 [07:04<02:25,  6.07s/it]
Training Epoch: 7/100, batch 140/188 completed (loss: 0.0786:  74%|███████▍  | 70/94 [07:04<02:25,  6.07s/it]
Training Epoch: 7/100, batch 140/188 completed (loss: 0.1603:  74%|███████▍  | 70/94 [07:04<02:25,  6.07s/it]
Training Epoch: 7/100, batch 140/188 completed (loss: 0.1981:  74%|███████▍  | 70/94 [07:04<02:25,  6.07s/it]
Training Epoch: 7/100, batch 140/188 completed (loss: 0.1603:  76%|███████▌  | 71/94 [07:07<02:19,  6.05s/it]
Training Epoch: 7/100, batch 140/188 completed (loss: 0.1981:  76%|███████▌  | 71/94 [07:07<02:19,  6.05s/it]
Training Epoch: 7/100, batch 140/188 completed (loss: 0.0786:  76%|███████▌  | 71/94 [07:07<02:19,  6.05s/it]
Training Epoch: 7/100, batch 140/188 completed (loss: 0.1399:  76%|███████▌  | 71/94 [07:07<02:19,  6.05s/it]
Training Epoch: 7/100, batch 141/188 completed (loss: 0.1017:  76%|███████▌  | 71/94 [07:07<02:19,  6.05s/it]
Training Epoch: 7/100, batch 141/188 completed (loss: 0.1501:  76%|███████▌  | 71/94 [07:07<02:19,  6.05s/it]
Training Epoch: 7/100, batch 141/188 completed (loss: 0.2011:  76%|███████▌  | 71/94 [07:07<02:19,  6.05s/it]
Training Epoch: 7/100, batch 141/188 completed (loss: 0.0571:  76%|███████▌  | 71/94 [07:07<02:19,  6.05s/it]
Training Epoch: 7/100, batch 142/188 completed (loss: 0.4095:  76%|███████▌  | 71/94 [07:10<02:19,  6.05s/it]
Training Epoch: 7/100, batch 142/188 completed (loss: 0.3076:  76%|███████▌  | 71/94 [07:10<02:19,  6.05s/it]
Training Epoch: 7/100, batch 142/188 completed (loss: 0.1464:  76%|███████▌  | 71/94 [07:10<02:19,  6.05s/it]
Training Epoch: 7/100, batch 142/188 completed (loss: 0.1983:  76%|███████▌  | 71/94 [07:10<02:19,  6.05s/it]
Training Epoch: 7/100, batch 142/188 completed (loss: 0.1983:  77%|███████▋  | 72/94 [07:13<02:13,  6.08s/it]
Training Epoch: 7/100, batch 142/188 completed (loss: 0.4095:  77%|███████▋  | 72/94 [07:13<02:13,  6.08s/it]
Training Epoch: 7/100, batch 142/188 completed (loss: 0.3076:  77%|███████▋  | 72/94 [07:13<02:13,  6.08s/it]
Training Epoch: 7/100, batch 142/188 completed (loss: 0.1464:  77%|███████▋  | 72/94 [07:13<02:13,  6.08s/it]
Training Epoch: 7/100, batch 143/188 completed (loss: 0.5734:  77%|███████▋  | 72/94 [07:13<02:13,  6.08s/it]
Training Epoch: 7/100, batch 143/188 completed (loss: 0.0725:  77%|███████▋  | 72/94 [07:13<02:13,  6.08s/it]
Training Epoch: 7/100, batch 143/188 completed (loss: 0.1169:  77%|███████▋  | 72/94 [07:13<02:13,  6.08s/it]
Training Epoch: 7/100, batch 143/188 completed (loss: 0.1428:  77%|███████▋  | 72/94 [07:13<02:13,  6.08s/it]
Training Epoch: 7/100, batch 144/188 completed (loss: 0.1345:  77%|███████▋  | 72/94 [07:16<02:13,  6.08s/it]
Training Epoch: 7/100, batch 144/188 completed (loss: 0.1392:  77%|███████▋  | 72/94 [07:16<02:13,  6.08s/it]
Training Epoch: 7/100, batch 144/188 completed (loss: 0.3421:  77%|███████▋  | 72/94 [07:16<02:13,  6.08s/it]
Training Epoch: 7/100, batch 144/188 completed (loss: 0.1869:  77%|███████▋  | 72/94 [07:16<02:13,  6.08s/it]
Training Epoch: 7/100, batch 144/188 completed (loss: 0.1345:  78%|███████▊  | 73/94 [07:19<02:07,  6.05s/it]
Training Epoch: 7/100, batch 144/188 completed (loss: 0.1392:  78%|███████▊  | 73/94 [07:19<02:07,  6.05s/it]
Training Epoch: 7/100, batch 144/188 completed (loss: 0.1869:  78%|███████▊  | 73/94 [07:19<02:07,  6.05s/it]
Training Epoch: 7/100, batch 144/188 completed (loss: 0.3421:  78%|███████▊  | 73/94 [07:19<02:07,  6.05s/it]
Training Epoch: 7/100, batch 145/188 completed (loss: 0.2956:  78%|███████▊  | 73/94 [07:19<02:07,  6.05s/it]
Training Epoch: 7/100, batch 145/188 completed (loss: 0.2517:  78%|███████▊  | 73/94 [07:19<02:07,  6.05s/it]
Training Epoch: 7/100, batch 145/188 completed (loss: 0.1352:  78%|███████▊  | 73/94 [07:19<02:07,  6.05s/it]
Training Epoch: 7/100, batch 145/188 completed (loss: 0.0741:  78%|███████▊  | 73/94 [07:19<02:07,  6.05s/it]
Training Epoch: 7/100, batch 146/188 completed (loss: 0.2154:  78%|███████▊  | 73/94 [07:22<02:07,  6.05s/it]
Training Epoch: 7/100, batch 146/188 completed (loss: 0.2386:  78%|███████▊  | 73/94 [07:22<02:07,  6.05s/it]
Training Epoch: 7/100, batch 146/188 completed (loss: 0.205:  78%|███████▊  | 73/94 [07:22<02:07,  6.05s/it] 
Training Epoch: 7/100, batch 146/188 completed (loss: 0.2021:  78%|███████▊  | 73/94 [07:22<02:07,  6.05s/it]
Training Epoch: 7/100, batch 146/188 completed (loss: 0.2021:  79%|███████▊  | 74/94 [07:25<02:02,  6.12s/it]
Training Epoch: 7/100, batch 146/188 completed (loss: 0.205:  79%|███████▊  | 74/94 [07:25<02:02,  6.12s/it]
Training Epoch: 7/100, batch 146/188 completed (loss: 0.2154:  79%|███████▊  | 74/94 [07:25<02:02,  6.12s/it]
Training Epoch: 7/100, batch 146/188 completed (loss: 0.2386:  79%|███████▊  | 74/94 [07:25<02:02,  6.12s/it]
Training Epoch: 7/100, batch 147/188 completed (loss: 0.5881:  79%|███████▊  | 74/94 [07:25<02:02,  6.12s/it]
Training Epoch: 7/100, batch 147/188 completed (loss: 0.293:  79%|███████▊  | 74/94 [07:25<02:02,  6.12s/it] 
Training Epoch: 7/100, batch 147/188 completed (loss: 0.3141:  79%|███████▊  | 74/94 [07:25<02:02,  6.12s/it]
Training Epoch: 7/100, batch 147/188 completed (loss: 0.0825:  79%|███████▊  | 74/94 [07:25<02:02,  6.12s/it]
Training Epoch: 7/100, batch 148/188 completed (loss: 0.1177:  79%|███████▊  | 74/94 [07:28<02:02,  6.12s/it]
Training Epoch: 7/100, batch 148/188 completed (loss: 0.153:  79%|███████▊  | 74/94 [07:28<02:02,  6.12s/it] 
Training Epoch: 7/100, batch 148/188 completed (loss: 0.1942:  79%|███████▊  | 74/94 [07:28<02:02,  6.12s/it]
Training Epoch: 7/100, batch 148/188 completed (loss: 0.2472:  79%|███████▊  | 74/94 [07:28<02:02,  6.12s/it]
Training Epoch: 7/100, batch 148/188 completed (loss: 0.2472:  80%|███████▉  | 75/94 [07:31<01:55,  6.09s/it]
Training Epoch: 7/100, batch 148/188 completed (loss: 0.1177:  80%|███████▉  | 75/94 [07:31<01:55,  6.09s/it]
Training Epoch: 7/100, batch 148/188 completed (loss: 0.153:  80%|███████▉  | 75/94 [07:31<01:55,  6.09s/it]
Training Epoch: 7/100, batch 148/188 completed (loss: 0.1942:  80%|███████▉  | 75/94 [07:31<01:55,  6.09s/it]
Training Epoch: 7/100, batch 149/188 completed (loss: 0.1127:  80%|███████▉  | 75/94 [07:31<01:55,  6.09s/it]
Training Epoch: 7/100, batch 149/188 completed (loss: 0.2141:  80%|███████▉  | 75/94 [07:31<01:55,  6.09s/it]
Training Epoch: 7/100, batch 149/188 completed (loss: 0.1959:  80%|███████▉  | 75/94 [07:31<01:55,  6.09s/it]
Training Epoch: 7/100, batch 149/188 completed (loss: 0.1789:  80%|███████▉  | 75/94 [07:31<01:55,  6.09s/it]
Training Epoch: 7/100, batch 150/188 completed (loss: 0.2865:  80%|███████▉  | 75/94 [07:34<01:55,  6.09s/it]
Training Epoch: 7/100, batch 150/188 completed (loss: 0.2287:  80%|███████▉  | 75/94 [07:34<01:55,  6.09s/it]
Training Epoch: 7/100, batch 150/188 completed (loss: 0.1147:  80%|███████▉  | 75/94 [07:34<01:55,  6.09s/it]
Training Epoch: 7/100, batch 150/188 completed (loss: 0.2191:  80%|███████▉  | 75/94 [07:34<01:55,  6.09s/it]
Training Epoch: 7/100, batch 150/188 completed (loss: 0.1147:  81%|████████  | 76/94 [07:37<01:48,  6.05s/it]
Training Epoch: 7/100, batch 150/188 completed (loss: 0.2191:  81%|████████  | 76/94 [07:37<01:48,  6.06s/it]
Training Epoch: 7/100, batch 150/188 completed (loss: 0.2865:  81%|████████  | 76/94 [07:37<01:48,  6.06s/it]
Training Epoch: 7/100, batch 150/188 completed (loss: 0.2287:  81%|████████  | 76/94 [07:37<01:49,  6.06s/it]
Training Epoch: 7/100, batch 151/188 completed (loss: 0.0616:  81%|████████  | 76/94 [07:37<01:48,  6.05s/it]
Training Epoch: 7/100, batch 151/188 completed (loss: 0.1719:  81%|████████  | 76/94 [07:37<01:48,  6.06s/it]
Training Epoch: 7/100, batch 151/188 completed (loss: 0.0808:  81%|████████  | 76/94 [07:37<01:49,  6.06s/it]
Training Epoch: 7/100, batch 151/188 completed (loss: 0.3326:  81%|████████  | 76/94 [07:37<01:48,  6.06s/it]
Training Epoch: 7/100, batch 152/188 completed (loss: 0.1207:  81%|████████  | 76/94 [07:40<01:48,  6.05s/it]
Training Epoch: 7/100, batch 152/188 completed (loss: 0.2472:  81%|████████  | 76/94 [07:40<01:49,  6.06s/it]
Training Epoch: 7/100, batch 152/188 completed (loss: 0.0535:  81%|████████  | 76/94 [07:40<01:48,  6.06s/it]
Training Epoch: 7/100, batch 152/188 completed (loss: 0.2043:  81%|████████  | 76/94 [07:40<01:48,  6.06s/it]
Training Epoch: 7/100, batch 152/188 completed (loss: 0.2043:  82%|████████▏ | 77/94 [07:43<01:42,  6.04s/it]
Training Epoch: 7/100, batch 152/188 completed (loss: 0.2472:  82%|████████▏ | 77/94 [07:43<01:42,  6.04s/it]
Training Epoch: 7/100, batch 152/188 completed (loss: 0.1207:  82%|████████▏ | 77/94 [07:43<01:42,  6.05s/it]
Training Epoch: 7/100, batch 152/188 completed (loss: 0.0535:  82%|████████▏ | 77/94 [07:43<01:42,  6.05s/it]
Training Epoch: 7/100, batch 153/188 completed (loss: 0.1037:  82%|████████▏ | 77/94 [07:43<01:42,  6.04s/it]
Training Epoch: 7/100, batch 153/188 completed (loss: 0.3324:  82%|████████▏ | 77/94 [07:43<01:42,  6.05s/it]
Training Epoch: 7/100, batch 153/188 completed (loss: 0.0839:  82%|████████▏ | 77/94 [07:43<01:42,  6.05s/it]
Training Epoch: 7/100, batch 153/188 completed (loss: 0.082:  82%|████████▏ | 77/94 [07:43<01:42,  6.04s/it] 
Training Epoch: 7/100, batch 154/188 completed (loss: 0.0654:  82%|████████▏ | 77/94 [07:46<01:42,  6.05s/it]
Training Epoch: 7/100, batch 154/188 completed (loss: 0.0975:  82%|████████▏ | 77/94 [07:46<01:42,  6.04s/it]
Training Epoch: 7/100, batch 154/188 completed (loss: 0.0902:  82%|████████▏ | 77/94 [07:46<01:42,  6.05s/it]
Training Epoch: 7/100, batch 154/188 completed (loss: 0.1289:  82%|████████▏ | 77/94 [07:46<01:42,  6.04s/it]
Training Epoch: 7/100, batch 154/188 completed (loss: 0.0975:  83%|████████▎ | 78/94 [07:49<01:36,  6.03s/it]
Training Epoch: 7/100, batch 154/188 completed (loss: 0.1289:  83%|████████▎ | 78/94 [07:49<01:36,  6.03s/it]
Training Epoch: 7/100, batch 154/188 completed (loss: 0.0654:  83%|████████▎ | 78/94 [07:49<01:36,  6.03s/it]
Training Epoch: 7/100, batch 154/188 completed (loss: 0.0902:  83%|████████▎ | 78/94 [07:49<01:36,  6.03s/it]
Training Epoch: 7/100, batch 155/188 completed (loss: 0.3222:  83%|████████▎ | 78/94 [07:49<01:36,  6.03s/it]
Training Epoch: 7/100, batch 155/188 completed (loss: 0.132:  83%|████████▎ | 78/94 [07:49<01:36,  6.03s/it] 
Training Epoch: 7/100, batch 155/188 completed (loss: 0.2797:  83%|████████▎ | 78/94 [07:49<01:36,  6.03s/it]
Training Epoch: 7/100, batch 155/188 completed (loss: 0.1406:  83%|████████▎ | 78/94 [07:49<01:36,  6.03s/it]
Training Epoch: 7/100, batch 156/188 completed (loss: 0.1064:  83%|████████▎ | 78/94 [07:52<01:36,  6.03s/it]
Training Epoch: 7/100, batch 156/188 completed (loss: 0.4028:  83%|████████▎ | 78/94 [07:52<01:36,  6.03s/it]
Training Epoch: 7/100, batch 156/188 completed (loss: 0.17:  83%|████████▎ | 78/94 [07:52<01:36,  6.03s/it]  
Training Epoch: 7/100, batch 156/188 completed (loss: 0.4088:  83%|████████▎ | 78/94 [07:52<01:36,  6.03s/it]
Training Epoch: 7/100, batch 156/188 completed (loss: 0.4088:  84%|████████▍ | 79/94 [07:55<01:31,  6.09s/it]
Training Epoch: 7/100, batch 156/188 completed (loss: 0.4028:  84%|████████▍ | 79/94 [07:55<01:31,  6.09s/it]
Training Epoch: 7/100, batch 156/188 completed (loss: 0.1064:  84%|████████▍ | 79/94 [07:55<01:31,  6.09s/it]
Training Epoch: 7/100, batch 156/188 completed (loss: 0.17:  84%|████████▍ | 79/94 [07:55<01:31,  6.09s/it]
Training Epoch: 7/100, batch 157/188 completed (loss: 0.1755:  84%|████████▍ | 79/94 [07:55<01:31,  6.09s/it]
Training Epoch: 7/100, batch 157/188 completed (loss: 0.2784:  84%|████████▍ | 79/94 [07:55<01:31,  6.09s/it]
Training Epoch: 7/100, batch 157/188 completed (loss: 0.1255:  84%|████████▍ | 79/94 [07:55<01:31,  6.09s/it]
Training Epoch: 7/100, batch 157/188 completed (loss: 0.0681:  84%|████████▍ | 79/94 [07:55<01:31,  6.09s/it]
Training Epoch: 7/100, batch 158/188 completed (loss: 0.4134:  84%|████████▍ | 79/94 [07:58<01:31,  6.09s/it]
Training Epoch: 7/100, batch 158/188 completed (loss: 0.1585:  84%|████████▍ | 79/94 [07:58<01:31,  6.09s/it]
Training Epoch: 7/100, batch 158/188 completed (loss: 0.2531:  84%|████████▍ | 79/94 [07:58<01:31,  6.09s/it]
Training Epoch: 7/100, batch 158/188 completed (loss: 0.1517:  84%|████████▍ | 79/94 [07:58<01:31,  6.09s/it]
Training Epoch: 7/100, batch 158/188 completed (loss: 0.1517:  85%|████████▌ | 80/94 [08:01<01:24,  6.02s/it]
Training Epoch: 7/100, batch 158/188 completed (loss: 0.4134:  85%|████████▌ | 80/94 [08:01<01:24,  6.01s/it]
Training Epoch: 7/100, batch 158/188 completed (loss: 0.1585:  85%|████████▌ | 80/94 [08:01<01:24,  6.02s/it]
Training Epoch: 7/100, batch 158/188 completed (loss: 0.2531:  85%|████████▌ | 80/94 [08:01<01:24,  6.02s/it]
Training Epoch: 7/100, batch 159/188 completed (loss: 0.5848:  85%|████████▌ | 80/94 [08:01<01:24,  6.02s/it]
Training Epoch: 7/100, batch 159/188 completed (loss: 0.2372:  85%|████████▌ | 80/94 [08:01<01:24,  6.02s/it]
Training Epoch: 7/100, batch 159/188 completed (loss: 0.3225:  85%|████████▌ | 80/94 [08:01<01:24,  6.01s/it]
Training Epoch: 7/100, batch 159/188 completed (loss: 0.2336:  85%|████████▌ | 80/94 [08:01<01:24,  6.02s/it]
Training Epoch: 7/100, batch 160/188 completed (loss: 0.1272:  85%|████████▌ | 80/94 [08:04<01:24,  6.02s/it]
Training Epoch: 7/100, batch 160/188 completed (loss: 0.1902:  85%|████████▌ | 80/94 [08:04<01:24,  6.01s/it]
Training Epoch: 7/100, batch 160/188 completed (loss: 0.07:  85%|████████▌ | 80/94 [08:04<01:24,  6.02s/it]  
Training Epoch: 7/100, batch 160/188 completed (loss: 0.2407:  85%|████████▌ | 80/94 [08:04<01:24,  6.02s/it]
Training Epoch: 7/100, batch 160/188 completed (loss: 0.1272:  86%|████████▌ | 81/94 [08:07<01:18,  6.02s/it]
Training Epoch: 7/100, batch 160/188 completed (loss: 0.2407:  86%|████████▌ | 81/94 [08:07<01:18,  6.02s/it]
Training Epoch: 7/100, batch 160/188 completed (loss: 0.1902:  86%|████████▌ | 81/94 [08:07<01:18,  6.02s/it]
Training Epoch: 7/100, batch 160/188 completed (loss: 0.07:  86%|████████▌ | 81/94 [08:07<01:18,  6.02s/it]
Training Epoch: 7/100, batch 161/188 completed (loss: 0.3831:  86%|████████▌ | 81/94 [08:07<01:18,  6.02s/it]
Training Epoch: 7/100, batch 161/188 completed (loss: 0.0656:  86%|████████▌ | 81/94 [08:07<01:18,  6.02s/it]
Training Epoch: 7/100, batch 161/188 completed (loss: 0.2173:  86%|████████▌ | 81/94 [08:07<01:18,  6.02s/it]
Training Epoch: 7/100, batch 161/188 completed (loss: 0.1717:  86%|████████▌ | 81/94 [08:07<01:18,  6.02s/it]
Training Epoch: 7/100, batch 162/188 completed (loss: 0.3858:  86%|████████▌ | 81/94 [08:10<01:18,  6.02s/it]
Training Epoch: 7/100, batch 162/188 completed (loss: 0.1914:  86%|████████▌ | 81/94 [08:10<01:18,  6.02s/it]
Training Epoch: 7/100, batch 162/188 completed (loss: 0.1236:  86%|████████▌ | 81/94 [08:10<01:18,  6.02s/it]
Training Epoch: 7/100, batch 162/188 completed (loss: 0.206:  86%|████████▌ | 81/94 [08:10<01:18,  6.02s/it] 
Training Epoch: 7/100, batch 162/188 completed (loss: 0.206:  87%|████████▋ | 82/94 [08:14<01:13,  6.10s/it]
Training Epoch: 7/100, batch 162/188 completed (loss: 0.1236:  87%|████████▋ | 82/94 [08:14<01:13,  6.10s/it]
Training Epoch: 7/100, batch 162/188 completed (loss: 0.1914:  87%|████████▋ | 82/94 [08:14<01:13,  6.10s/it]
Training Epoch: 7/100, batch 162/188 completed (loss: 0.3858:  87%|████████▋ | 82/94 [08:14<01:13,  6.10s/it]
Training Epoch: 7/100, batch 163/188 completed (loss: 0.0551:  87%|████████▋ | 82/94 [08:14<01:13,  6.10s/it]
Training Epoch: 7/100, batch 163/188 completed (loss: 0.1699:  87%|████████▋ | 82/94 [08:14<01:13,  6.10s/it]
Training Epoch: 7/100, batch 163/188 completed (loss: 0.2225:  87%|████████▋ | 82/94 [08:14<01:13,  6.10s/it]
Training Epoch: 7/100, batch 163/188 completed (loss: 0.3906:  87%|████████▋ | 82/94 [08:14<01:13,  6.10s/it]
Training Epoch: 7/100, batch 164/188 completed (loss: 0.3123:  87%|████████▋ | 82/94 [08:16<01:13,  6.10s/it]
Training Epoch: 7/100, batch 164/188 completed (loss: 0.3488:  87%|████████▋ | 82/94 [08:16<01:13,  6.10s/it]
Training Epoch: 7/100, batch 164/188 completed (loss: 0.1784:  87%|████████▋ | 82/94 [08:16<01:13,  6.10s/it]
Training Epoch: 7/100, batch 164/188 completed (loss: 0.3825:  87%|████████▋ | 82/94 [08:17<01:13,  6.10s/it]
Training Epoch: 7/100, batch 164/188 completed (loss: 0.3825:  88%|████████▊ | 83/94 [08:20<01:06,  6.08s/it]
Training Epoch: 7/100, batch 164/188 completed (loss: 0.1784:  88%|████████▊ | 83/94 [08:20<01:06,  6.08s/it]
Training Epoch: 7/100, batch 164/188 completed (loss: 0.3123:  88%|████████▊ | 83/94 [08:20<01:06,  6.08s/it]
Training Epoch: 7/100, batch 164/188 completed (loss: 0.3488:  88%|████████▊ | 83/94 [08:20<01:06,  6.08s/it]
Training Epoch: 7/100, batch 165/188 completed (loss: 0.1658:  88%|████████▊ | 83/94 [08:20<01:06,  6.08s/it]
Training Epoch: 7/100, batch 165/188 completed (loss: 0.1751:  88%|████████▊ | 83/94 [08:20<01:06,  6.08s/it]
Training Epoch: 7/100, batch 165/188 completed (loss: 0.0878:  88%|████████▊ | 83/94 [08:20<01:06,  6.08s/it]
Training Epoch: 7/100, batch 165/188 completed (loss: 0.1253:  88%|████████▊ | 83/94 [08:20<01:06,  6.08s/it]
Training Epoch: 7/100, batch 166/188 completed (loss: 0.3691:  88%|████████▊ | 83/94 [08:23<01:06,  6.08s/it]
Training Epoch: 7/100, batch 166/188 completed (loss: 0.1549:  88%|████████▊ | 83/94 [08:23<01:06,  6.08s/it]
Training Epoch: 7/100, batch 166/188 completed (loss: 0.2055:  88%|████████▊ | 83/94 [08:23<01:06,  6.08s/it]
Training Epoch: 7/100, batch 166/188 completed (loss: 0.2795:  88%|████████▊ | 83/94 [08:23<01:06,  6.08s/it]
Training Epoch: 7/100, batch 166/188 completed (loss: 0.1549:  89%|████████▉ | 84/94 [08:26<01:00,  6.10s/it]
Training Epoch: 7/100, batch 166/188 completed (loss: 0.2055:  89%|████████▉ | 84/94 [08:26<01:00,  6.10s/it]
Training Epoch: 7/100, batch 166/188 completed (loss: 0.3691:  89%|████████▉ | 84/94 [08:26<01:00,  6.10s/it]
Training Epoch: 7/100, batch 166/188 completed (loss: 0.2795:  89%|████████▉ | 84/94 [08:26<01:01,  6.10s/it]
Training Epoch: 7/100, batch 167/188 completed (loss: 0.0371:  89%|████████▉ | 84/94 [08:26<01:00,  6.10s/it]
Training Epoch: 7/100, batch 167/188 completed (loss: 0.1772:  89%|████████▉ | 84/94 [08:26<01:00,  6.10s/it]
Training Epoch: 7/100, batch 167/188 completed (loss: 0.5324:  89%|████████▉ | 84/94 [08:26<01:00,  6.10s/it]
Training Epoch: 7/100, batch 167/188 completed (loss: 0.1932:  89%|████████▉ | 84/94 [08:26<01:01,  6.10s/it]
Training Epoch: 7/100, batch 168/188 completed (loss: 0.074:  89%|████████▉ | 84/94 [08:29<01:00,  6.10s/it] 
Training Epoch: 7/100, batch 168/188 completed (loss: 0.152:  89%|████████▉ | 84/94 [08:29<01:00,  6.10s/it] 
Training Epoch: 7/100, batch 168/188 completed (loss: 0.071:  89%|████████▉ | 84/94 [08:29<01:00,  6.10s/it] 
Training Epoch: 7/100, batch 168/188 completed (loss: 0.3475:  89%|████████▉ | 84/94 [08:29<01:01,  6.10s/it]
Training Epoch: 7/100, batch 168/188 completed (loss: 0.071:  90%|█████████ | 85/94 [08:32<00:54,  6.09s/it]
Training Epoch: 7/100, batch 168/188 completed (loss: 0.152:  90%|█████████ | 85/94 [08:32<00:54,  6.09s/it]
Training Epoch: 7/100, batch 168/188 completed (loss: 0.074:  90%|█████████ | 85/94 [08:32<00:54,  6.09s/it]
Training Epoch: 7/100, batch 168/188 completed (loss: 0.3475:  90%|█████████ | 85/94 [08:32<00:54,  6.10s/it]
Training Epoch: 7/100, batch 169/188 completed (loss: 0.0798:  90%|█████████ | 85/94 [08:32<00:54,  6.09s/it]
Training Epoch: 7/100, batch 169/188 completed (loss: 0.1206:  90%|█████████ | 85/94 [08:32<00:54,  6.09s/it]
Training Epoch: 7/100, batch 169/188 completed (loss: 0.2505:  90%|█████████ | 85/94 [08:32<00:54,  6.09s/it]
Training Epoch: 7/100, batch 169/188 completed (loss: 0.2578:  90%|█████████ | 85/94 [08:32<00:54,  6.10s/it]
Training Epoch: 7/100, batch 170/188 completed (loss: 0.2672:  90%|█████████ | 85/94 [08:35<00:54,  6.09s/it]
Training Epoch: 7/100, batch 170/188 completed (loss: 0.0818:  90%|█████████ | 85/94 [08:35<00:54,  6.09s/it]
Training Epoch: 7/100, batch 170/188 completed (loss: 0.1952:  90%|█████████ | 85/94 [08:35<00:54,  6.09s/it]
Training Epoch: 7/100, batch 170/188 completed (loss: 0.2127:  90%|█████████ | 85/94 [08:35<00:54,  6.10s/it]
Training Epoch: 7/100, batch 170/188 completed (loss: 0.2672:  91%|█████████▏| 86/94 [08:38<00:48,  6.07s/it]
Training Epoch: 7/100, batch 170/188 completed (loss: 0.2127:  91%|█████████▏| 86/94 [08:38<00:48,  6.07s/it]
Training Epoch: 7/100, batch 170/188 completed (loss: 0.0818:  91%|█████████▏| 86/94 [08:38<00:48,  6.07s/it]
Training Epoch: 7/100, batch 170/188 completed (loss: 0.1952:  91%|█████████▏| 86/94 [08:38<00:48,  6.07s/it]
Training Epoch: 7/100, batch 171/188 completed (loss: 0.1514:  91%|█████████▏| 86/94 [08:38<00:48,  6.07s/it]
Training Epoch: 7/100, batch 171/188 completed (loss: 0.1798:  91%|█████████▏| 86/94 [08:38<00:48,  6.07s/it]
Training Epoch: 7/100, batch 171/188 completed (loss: 0.133:  91%|█████████▏| 86/94 [08:38<00:48,  6.07s/it] 
Training Epoch: 7/100, batch 171/188 completed (loss: 0.0707:  91%|█████████▏| 86/94 [08:38<00:48,  6.07s/it]
Training Epoch: 7/100, batch 172/188 completed (loss: 0.4317:  91%|█████████▏| 86/94 [08:41<00:48,  6.07s/it]
Training Epoch: 7/100, batch 172/188 completed (loss: 0.1508:  91%|█████████▏| 86/94 [08:41<00:48,  6.07s/it]
Training Epoch: 7/100, batch 172/188 completed (loss: 0.2816:  91%|█████████▏| 86/94 [08:41<00:48,  6.07s/it]
Training Epoch: 7/100, batch 172/188 completed (loss: 0.1427:  91%|█████████▏| 86/94 [08:41<00:48,  6.07s/it]
Training Epoch: 7/100, batch 172/188 completed (loss: 0.1427:  93%|█████████▎| 87/94 [08:44<00:42,  6.11s/it]
Training Epoch: 7/100, batch 172/188 completed (loss: 0.2816:  93%|█████████▎| 87/94 [08:44<00:42,  6.11s/it]
Training Epoch: 7/100, batch 172/188 completed (loss: 0.4317:  93%|█████████▎| 87/94 [08:44<00:42,  6.11s/it]
Training Epoch: 7/100, batch 172/188 completed (loss: 0.1508:  93%|█████████▎| 87/94 [08:44<00:42,  6.12s/it]
Training Epoch: 7/100, batch 173/188 completed (loss: 0.2589:  93%|█████████▎| 87/94 [08:44<00:42,  6.11s/it]
Training Epoch: 7/100, batch 173/188 completed (loss: 0.2086:  93%|█████████▎| 87/94 [08:44<00:42,  6.12s/it]
Training Epoch: 7/100, batch 173/188 completed (loss: 0.494:  93%|█████████▎| 87/94 [08:44<00:42,  6.11s/it] 
Training Epoch: 7/100, batch 173/188 completed (loss: 0.4923:  93%|█████████▎| 87/94 [08:44<00:42,  6.11s/it]
Training Epoch: 7/100, batch 174/188 completed (loss: 0.0313:  93%|█████████▎| 87/94 [08:47<00:42,  6.12s/it]
Training Epoch: 7/100, batch 174/188 completed (loss: 0.1543:  93%|█████████▎| 87/94 [08:47<00:42,  6.11s/it]
Training Epoch: 7/100, batch 174/188 completed (loss: 0.0742:  93%|█████████▎| 87/94 [08:47<00:42,  6.11s/it]
Training Epoch: 7/100, batch 174/188 completed (loss: 0.1008:  93%|█████████▎| 87/94 [08:47<00:42,  6.11s/it]
Training Epoch: 7/100, batch 174/188 completed (loss: 0.1008:  94%|█████████▎| 88/94 [08:50<00:36,  6.03s/it]
Training Epoch: 7/100, batch 174/188 completed (loss: 0.0313:  94%|█████████▎| 88/94 [08:50<00:36,  6.03s/it]
Training Epoch: 7/100, batch 174/188 completed (loss: 0.1543:  94%|█████████▎| 88/94 [08:50<00:36,  6.04s/it]
Training Epoch: 7/100, batch 174/188 completed (loss: 0.0742:  94%|█████████▎| 88/94 [08:50<00:36,  6.04s/it]
Training Epoch: 7/100, batch 175/188 completed (loss: 0.2741:  94%|█████████▎| 88/94 [08:50<00:36,  6.04s/it]
Training Epoch: 7/100, batch 175/188 completed (loss: 0.3571:  94%|█████████▎| 88/94 [08:50<00:36,  6.03s/it]
Training Epoch: 7/100, batch 175/188 completed (loss: 0.1534:  94%|█████████▎| 88/94 [08:50<00:36,  6.04s/it]
Training Epoch: 7/100, batch 175/188 completed (loss: 0.1585:  94%|█████████▎| 88/94 [08:50<00:36,  6.03s/it]
Training Epoch: 7/100, batch 176/188 completed (loss: 0.7536:  94%|█████████▎| 88/94 [08:53<00:36,  6.04s/it]
Training Epoch: 7/100, batch 176/188 completed (loss: 0.0848:  94%|█████████▎| 88/94 [08:53<00:36,  6.03s/it]
Training Epoch: 7/100, batch 176/188 completed (loss: 0.2149:  94%|█████████▎| 88/94 [08:53<00:36,  6.04s/it]
Training Epoch: 7/100, batch 176/188 completed (loss: 0.1134:  94%|█████████▎| 88/94 [08:53<00:36,  6.03s/it]
Training Epoch: 7/100, batch 176/188 completed (loss: 0.7536:  95%|█████████▍| 89/94 [08:56<00:30,  6.03s/it]
Training Epoch: 7/100, batch 176/188 completed (loss: 0.1134:  95%|█████████▍| 89/94 [08:56<00:30,  6.03s/it]
Training Epoch: 7/100, batch 176/188 completed (loss: 0.2149:  95%|█████████▍| 89/94 [08:56<00:30,  6.03s/it]
Training Epoch: 7/100, batch 176/188 completed (loss: 0.0848:  95%|█████████▍| 89/94 [08:56<00:30,  6.03s/it]
Training Epoch: 7/100, batch 177/188 completed (loss: 0.0584:  95%|█████████▍| 89/94 [08:56<00:30,  6.03s/it]
Training Epoch: 7/100, batch 177/188 completed (loss: 0.199:  95%|█████████▍| 89/94 [08:56<00:30,  6.03s/it] 
Training Epoch: 7/100, batch 177/188 completed (loss: 0.2134:  95%|█████████▍| 89/94 [08:56<00:30,  6.03s/it]
Training Epoch: 7/100, batch 177/188 completed (loss: 0.3888:  95%|█████████▍| 89/94 [08:56<00:30,  6.03s/it]
Training Epoch: 7/100, batch 178/188 completed (loss: 0.1026:  95%|█████████▍| 89/94 [08:59<00:30,  6.03s/it]
Training Epoch: 7/100, batch 178/188 completed (loss: 0.2306:  95%|█████████▍| 89/94 [08:59<00:30,  6.03s/it]
Training Epoch: 7/100, batch 178/188 completed (loss: 0.0707:  95%|█████████▍| 89/94 [08:59<00:30,  6.03s/it]
Training Epoch: 7/100, batch 178/188 completed (loss: 0.1956:  95%|█████████▍| 89/94 [08:59<00:30,  6.03s/it]
Training Epoch: 7/100, batch 178/188 completed (loss: 0.1956:  96%|█████████▌| 90/94 [09:02<00:24,  6.10s/it]
Training Epoch: 7/100, batch 178/188 completed (loss: 0.2306:  96%|█████████▌| 90/94 [09:02<00:24,  6.10s/it]
Training Epoch: 7/100, batch 178/188 completed (loss: 0.1026:  96%|█████████▌| 90/94 [09:02<00:24,  6.11s/it]
Training Epoch: 7/100, batch 178/188 completed (loss: 0.0707:  96%|█████████▌| 90/94 [09:02<00:24,  6.11s/it]
Training Epoch: 7/100, batch 179/188 completed (loss: 0.1183:  96%|█████████▌| 90/94 [09:02<00:24,  6.11s/it]
Training Epoch: 7/100, batch 179/188 completed (loss: 0.2396:  96%|█████████▌| 90/94 [09:02<00:24,  6.11s/it]
Training Epoch: 7/100, batch 179/188 completed (loss: 0.318:  96%|█████████▌| 90/94 [09:02<00:24,  6.10s/it] 
Training Epoch: 7/100, batch 179/188 completed (loss: 0.305:  96%|█████████▌| 90/94 [09:02<00:24,  6.10s/it] 
Training Epoch: 7/100, batch 180/188 completed (loss: 0.1568:  96%|█████████▌| 90/94 [09:05<00:24,  6.11s/it]
Training Epoch: 7/100, batch 180/188 completed (loss: 0.1564:  96%|█████████▌| 90/94 [09:05<00:24,  6.10s/it]
Training Epoch: 7/100, batch 180/188 completed (loss: 0.2779:  96%|█████████▌| 90/94 [09:05<00:24,  6.11s/it]
Training Epoch: 7/100, batch 180/188 completed (loss: 0.2956:  96%|█████████▌| 90/94 [09:05<00:24,  6.10s/it]
Training Epoch: 7/100, batch 180/188 completed (loss: 0.1564:  97%|█████████▋| 91/94 [09:08<00:18,  6.04s/it]
Training Epoch: 7/100, batch 180/188 completed (loss: 0.2779:  97%|█████████▋| 91/94 [09:08<00:18,  6.04s/it]
Training Epoch: 7/100, batch 180/188 completed (loss: 0.1568:  97%|█████████▋| 91/94 [09:08<00:18,  6.04s/it]
Training Epoch: 7/100, batch 180/188 completed (loss: 0.2956:  97%|█████████▋| 91/94 [09:08<00:18,  6.05s/it]
Training Epoch: 7/100, batch 181/188 completed (loss: 0.2257:  97%|█████████▋| 91/94 [09:08<00:18,  6.04s/it]
Training Epoch: 7/100, batch 181/188 completed (loss: 0.4989:  97%|█████████▋| 91/94 [09:08<00:18,  6.04s/it]
Training Epoch: 7/100, batch 181/188 completed (loss: 0.2223:  97%|█████████▋| 91/94 [09:08<00:18,  6.04s/it]
Training Epoch: 7/100, batch 181/188 completed (loss: 0.3063:  97%|█████████▋| 91/94 [09:08<00:18,  6.05s/it]
Training Epoch: 7/100, batch 182/188 completed (loss: 0.2179:  97%|█████████▋| 91/94 [09:11<00:18,  6.04s/it]
Training Epoch: 7/100, batch 182/188 completed (loss: 0.0933:  97%|█████████▋| 91/94 [09:11<00:18,  6.04s/it]
Training Epoch: 7/100, batch 182/188 completed (loss: 0.2069:  97%|█████████▋| 91/94 [09:11<00:18,  6.04s/it]
Training Epoch: 7/100, batch 182/188 completed (loss: 0.1376:  97%|█████████▋| 91/94 [09:11<00:18,  6.05s/it]
Training Epoch: 7/100, batch 182/188 completed (loss: 0.2179:  98%|█████████▊| 92/94 [09:14<00:12,  6.07s/it]
Training Epoch: 7/100, batch 182/188 completed (loss: 0.2069:  98%|█████████▊| 92/94 [09:14<00:12,  6.07s/it]
Training Epoch: 7/100, batch 182/188 completed (loss: 0.0933:  98%|█████████▊| 92/94 [09:14<00:12,  6.07s/it]
Training Epoch: 7/100, batch 182/188 completed (loss: 0.1376:  98%|█████████▊| 92/94 [09:14<00:12,  6.08s/it]
Training Epoch: 7/100, batch 183/188 completed (loss: 0.1098:  98%|█████████▊| 92/94 [09:14<00:12,  6.07s/it]
Training Epoch: 7/100, batch 183/188 completed (loss: 0.1361:  98%|█████████▊| 92/94 [09:14<00:12,  6.07s/it]
Training Epoch: 7/100, batch 183/188 completed (loss: 0.1526:  98%|█████████▊| 92/94 [09:14<00:12,  6.07s/it]
Training Epoch: 7/100, batch 183/188 completed (loss: 0.185:  98%|█████████▊| 92/94 [09:14<00:12,  6.08s/it] 
Training Epoch: 7/100, batch 184/188 completed (loss: 0.212:  98%|█████████▊| 92/94 [09:17<00:12,  6.07s/it] 
Training Epoch: 7/100, batch 184/188 completed (loss: 0.3275:  98%|█████████▊| 92/94 [09:17<00:12,  6.07s/it]
Training Epoch: 7/100, batch 184/188 completed (loss: 0.2587:  98%|█████████▊| 92/94 [09:17<00:12,  6.07s/it]
Training Epoch: 7/100, batch 184/188 completed (loss: 0.0972:  98%|█████████▊| 92/94 [09:17<00:12,  6.08s/it]
Training Epoch: 7/100, batch 184/188 completed (loss: 0.2587:  99%|█████████▉| 93/94 [09:20<00:06,  6.06s/it]
Training Epoch: 7/100, batch 184/188 completed (loss: 0.3275:  99%|█████████▉| 93/94 [09:20<00:06,  6.06s/it]
Training Epoch: 7/100, batch 184/188 completed (loss: 0.0972:  99%|█████████▉| 93/94 [09:20<00:06,  6.06s/it]
Training Epoch: 7/100, batch 184/188 completed (loss: 0.212:  99%|█████████▉| 93/94 [09:20<00:06,  6.06s/it]
Training Epoch: 7/100, batch 185/188 completed (loss: 0.1855:  99%|█████████▉| 93/94 [09:20<00:06,  6.06s/it]
Training Epoch: 7/100, batch 185/188 completed (loss: 0.031:  99%|█████████▉| 93/94 [09:20<00:06,  6.06s/it] 
Training Epoch: 7/100, batch 185/188 completed (loss: 0.3248:  99%|█████████▉| 93/94 [09:20<00:06,  6.06s/it]
Training Epoch: 7/100, batch 185/188 completed (loss: 0.2309:  99%|█████████▉| 93/94 [09:20<00:06,  6.06s/it]
Training Epoch: 7/100, batch 186/188 completed (loss: 0.1547:  99%|█████████▉| 93/94 [09:23<00:06,  6.06s/it]
Training Epoch: 7/100, batch 186/188 completed (loss: 0.0935:  99%|█████████▉| 93/94 [09:23<00:06,  6.06s/it]
Training Epoch: 7/100, batch 186/188 completed (loss: 0.2542:  99%|█████████▉| 93/94 [09:23<00:06,  6.06s/it]
Training Epoch: 7/100, batch 186/188 completed (loss: 0.1289:  99%|█████████▉| 93/94 [09:23<00:06,  6.06s/it]
Training Epoch: 7/100, batch 186/188 completed (loss: 0.2542: 100%|██████████| 94/94 [09:25<00:00,  5.73s/it]
Training Epoch: 7/100, batch 186/188 completed (loss: 0.1289: 100%|██████████| 94/94 [09:25<00:00,  5.73s/it]
Training Epoch: 7/100, batch 186/188 completed (loss: 0.1547: 100%|██████████| 94/94 [09:25<00:00,  5.73s/it]
Training Epoch: 7/100, batch 186/188 completed (loss: 0.0935: 100%|██████████| 94/94 [09:25<00:00,  5.73s/it]
Training Epoch: 7/100, batch 187/188 completed (loss: 0.0306: 100%|██████████| 94/94 [09:25<00:00,  5.73s/it]
Training Epoch: 7/100, batch 187/188 completed (loss: 0.4385: 100%|██████████| 94/94 [09:25<00:00,  5.73s/it]
Training Epoch: 7/100, batch 187/188 completed (loss: 0.2619: 100%|██████████| 94/94 [09:25<00:00,  5.73s/it]
Training Epoch: 7/100, batch 187/188 completed (loss: 0.1012: 100%|██████████| 94/94 [09:25<00:00,  5.73s/it]
Training Epoch: 7/100, batch 187/188 completed (loss: 0.4385: 100%|██████████| 94/94 [09:25<00:00,  6.02s/it]

Training Epoch: 7/100, batch 187/188 completed (loss: 0.0306: 100%|██████████| 94/94 [09:25<00:00,  6.02s/it]

Training Epoch: 7/100, batch 187/188 completed (loss: 0.2619: 100%|██████████| 94/94 [09:25<00:00,  6.02s/it]

Training Epoch: 7/100, batch 187/188 completed (loss: 0.1012: 100%|██████████| 94/94 [09:26<00:00,  6.02s/it]
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/c10d_logger.py:83: UserWarning: barrier(): using the device under current context. You can specify `device_id` in `init_process_group` to mute this warning.
  return func(*args, **kwargs)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
saving model + train state.
Warning: failed to upload checkpoint_7 to HF: (Request ID: Root=1-6a1b7024-07323b3963631df3055b72ea;35d5129c-ce58-4867-9ccf-ff049a61617d)

403 Forbidden: You need to setup automatic credit recharge in order to upload more data. You can do so at /settings/billing..
Cannot access content at: https://huggingface.co/seyedparsa/thoughtformer-checkpoints.git/info/lfs/objects/batch.
Make sure your token has the correct permissions.
Warning: failed to upload train_state_7 to HF: (Request ID: Root=1-6a1b7028-3ce4414f3aa0fc7d7b473eda;1c102bc5-8c26-4d23-a239-8c42548d96f3)

403 Forbidden: You need to setup automatic credit recharge in order to upload more data. You can do so at /settings/billing..
Cannot access content at: https://huggingface.co/seyedparsa/thoughtformer-checkpoints.git/info/lfs/objects/batch.
Make sure your token has the correct permissions.
Deleted old checkpoint: checkpoint_6
Deleted old checkpoint: train_state_6
[warn] HF upload failed for ckpt_7; kept locally, earlier ckpts deleted.

Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]
Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]
Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]eval loss 0.11478096638855181

Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]
[WRONG] Example 0 (diff=3)
  Expected : 'Claude'
  Extracted: 'Morgan'
  Generated: '### Morgan'

[WRONG] Example 1 (diff=8)
  Expected : 'Guy'
  Extracted: 'Irma'
  Generated: '### Irma'

Test Accuracy:   3%|▎         | 1/32 [00:07<03:38,  7.05s/it]
Test Accuracy:   3%|▎         | 1/32 [00:07<03:38,  7.05s/it]
Test accuracy: 0.38:   3%|▎         | 1/32 [00:07<03:38,  7.05s/it]
Test accuracy: 1.0:   3%|▎         | 1/32 [00:07<03:38,  7.05s/it]
[WRONG] Example 2 (diff=10)
Test Accuracy:   3%|▎         | 1/32 [00:07<03:38,  7.06s/it]
  Expected : 'Elvis'
  Extracted: 'Shannon'
  Generated: '### Shannon'

Test accuracy: 0.62:   3%|▎         | 1/32 [00:07<03:38,  7.06s/it]
[CORRECT] Example 3 (diff=8)
  Expected : 'Coral'
  Extracted: 'Coral'
  Generated: '### Coral'

[CORRECT] Example 4 (diff=1)
  Expected : 'Patricia'
  Extracted: 'Patricia'
  Generated: '### Patricia'

Test Accuracy:   3%|▎         | 1/32 [00:06<03:34,  6.90s/it]
Test accuracy: 0.38:   3%|▎         | 1/32 [00:06<03:34,  6.90s/it]
Test accuracy: 0.38:   6%|▋         | 2/32 [00:13<03:15,  6.53s/it]
Test accuracy: 0.62:   6%|▋         | 2/32 [00:13<03:15,  6.53s/it]
Test accuracy: 1.0:   6%|▋         | 2/32 [00:13<03:15,  6.53s/it]
Test accuracy: 0.5:   6%|▋         | 2/32 [00:13<03:15,  6.53s/it] 
Test accuracy: 0.75:   6%|▋         | 2/32 [00:13<03:15,  6.53s/it]
Test accuracy: 0.94:   6%|▋         | 2/32 [00:13<03:15,  6.53s/it]
Test accuracy: 0.38:   6%|▋         | 2/32 [00:13<03:13,  6.47s/it]
Test accuracy: 0.5:   6%|▋         | 2/32 [00:13<03:13,  6.47s/it] 
Test accuracy: 0.5:   9%|▉         | 3/32 [00:20<03:15,  6.73s/it]
Test accuracy: 0.94:   9%|▉         | 3/32 [00:20<03:15,  6.73s/it]
Test accuracy: 0.5:   9%|▉         | 3/32 [00:20<03:15,  6.73s/it]
Test accuracy: 0.92:   9%|▉         | 3/32 [00:20<03:15,  6.73s/it]
Test accuracy: 0.5:   9%|▉         | 3/32 [00:20<03:14,  6.69s/it]
Test accuracy: 0.54:   9%|▉         | 3/32 [00:20<03:14,  6.69s/it]
Test accuracy: 0.75:   9%|▉         | 3/32 [00:20<03:15,  6.74s/it]
Test accuracy: 0.62:   9%|▉         | 3/32 [00:20<03:15,  6.74s/it]
Test accuracy: 0.62:  12%|█▎        | 4/32 [00:26<03:04,  6.57s/it]
Test accuracy: 0.92:  12%|█▎        | 4/32 [00:26<03:04,  6.57s/it]
Test accuracy: 0.66:  12%|█▎        | 4/32 [00:26<03:04,  6.57s/it]
Test accuracy: 0.5:  12%|█▎        | 4/32 [00:26<03:04,  6.57s/it]
Test accuracy: 0.81:  12%|█▎        | 4/32 [00:26<03:04,  6.57s/it]
Test accuracy: 0.54:  12%|█▎        | 4/32 [00:26<03:03,  6.55s/it]
Test accuracy: 0.53:  12%|█▎        | 4/32 [00:26<03:04,  6.57s/it]
Test accuracy: 0.53:  12%|█▎        | 4/32 [00:26<03:03,  6.55s/it]
Test accuracy: 0.81:  16%|█▌        | 5/32 [00:32<02:53,  6.42s/it]
Test accuracy: 0.66:  16%|█▌        | 5/32 [00:32<02:53,  6.42s/it]
Test accuracy: 0.75:  16%|█▌        | 5/32 [00:32<02:53,  6.42s/it]
Test accuracy: 0.53:  16%|█▌        | 5/32 [00:32<02:53,  6.42s/it]
Test accuracy: 0.62:  16%|█▌        | 5/32 [00:32<02:53,  6.42s/it]
Test accuracy: 0.52:  16%|█▌        | 5/32 [00:32<02:53,  6.42s/it]
Test accuracy: 0.53:  16%|█▌        | 5/32 [00:32<02:52,  6.41s/it]
Test accuracy: 0.6:  16%|█▌        | 5/32 [00:32<02:52,  6.41s/it] 
Test accuracy: 0.62:  19%|█▉        | 6/32 [00:39<02:46,  6.40s/it]
Test accuracy: 0.75:  19%|█▉        | 6/32 [00:39<02:46,  6.41s/it]
Test accuracy: 0.65:  19%|█▉        | 6/32 [00:39<02:46,  6.40s/it]
Test accuracy: 0.71:  19%|█▉        | 6/32 [00:39<02:46,  6.41s/it]
Test accuracy: 0.52:  19%|█▉        | 6/32 [00:39<02:46,  6.41s/it]
Test accuracy: 0.58:  19%|█▉        | 6/32 [00:39<02:46,  6.41s/it]
Test accuracy: 0.6:  19%|█▉        | 6/32 [00:38<02:46,  6.40s/it]
Test accuracy: 0.67:  19%|█▉        | 6/32 [00:38<02:46,  6.40s/it]
Test accuracy: 0.65:  22%|██▏       | 7/32 [00:45<02:39,  6.38s/it]
Test accuracy: 0.58:  22%|██▏       | 7/32 [00:45<02:39,  6.38s/it]
Test accuracy: 0.68:  22%|██▏       | 7/32 [00:45<02:39,  6.38s/it]
Test accuracy: 0.61:  22%|██▏       | 7/32 [00:45<02:39,  6.38s/it]
Test accuracy: 0.67:  22%|██▏       | 7/32 [00:45<02:39,  6.37s/it]
Test accuracy: 0.71:  22%|██▏       | 7/32 [00:45<02:39,  6.38s/it]
Test accuracy: 0.64:  22%|██▏       | 7/32 [00:45<02:39,  6.38s/it]
Test accuracy: 0.68:  22%|██▏       | 7/32 [00:45<02:39,  6.37s/it]
Test accuracy: 0.68:  25%|██▌       | 8/32 [00:51<02:31,  6.30s/it]
Test accuracy: 0.64:  25%|██▌       | 8/32 [00:51<02:31,  6.30s/it]
Test accuracy: 0.7:  25%|██▌       | 8/32 [00:51<02:31,  6.30s/it] 
Test accuracy: 0.66:  25%|██▌       | 8/32 [00:51<02:31,  6.30s/it]
Test accuracy: 0.68:  25%|██▌       | 8/32 [00:51<02:31,  6.30s/it]
Test accuracy: 0.61:  25%|██▌       | 8/32 [00:51<02:31,  6.30s/it]
Test accuracy: 0.58:  25%|██▌       | 8/32 [00:51<02:31,  6.30s/it]
Test accuracy: 0.67:  25%|██▌       | 8/32 [00:51<02:31,  6.30s/it]
Test accuracy: 0.7:  28%|██▊       | 9/32 [00:58<02:30,  6.56s/it]
Test accuracy: 0.66:  28%|██▊       | 9/32 [00:58<02:30,  6.56s/it]
Test accuracy: 0.72:  28%|██▊       | 9/32 [00:58<02:30,  6.56s/it]
Test accuracy: 0.67:  28%|██▊       | 9/32 [00:58<02:30,  6.56s/it]
Test accuracy: 0.67:  28%|██▊       | 9/32 [00:58<02:30,  6.56s/it]
Test accuracy: 0.67:  28%|██▊       | 9/32 [00:58<02:30,  6.56s/it]
Test accuracy: 0.58:  28%|██▊       | 9/32 [00:58<02:31,  6.57s/it]
Test accuracy: 0.57:  28%|██▊       | 9/32 [00:58<02:31,  6.57s/it]
Test accuracy: 0.72:  31%|███▏      | 10/32 [01:04<02:21,  6.44s/it]
Test accuracy: 0.67:  31%|███▏      | 10/32 [01:04<02:21,  6.44s/it]
Test accuracy: 0.57:  31%|███▏      | 10/32 [01:04<02:21,  6.43s/it]
Test accuracy: 0.68:  31%|███▏      | 10/32 [01:04<02:21,  6.44s/it]
Test accuracy: 0.67:  31%|███▏      | 10/32 [01:04<02:21,  6.43s/it]
Test accuracy: 0.68:  31%|███▏      | 10/32 [01:04<02:21,  6.44s/it]
Test accuracy: 0.57:  31%|███▏      | 10/32 [01:04<02:21,  6.43s/it]
Test accuracy: 0.65:  31%|███▏      | 10/32 [01:04<02:21,  6.43s/it]
Test accuracy: 0.68:  34%|███▍      | 11/32 [01:11<02:14,  6.41s/it]
Test accuracy: 0.57:  34%|███▍      | 11/32 [01:11<02:14,  6.41s/it]
Test accuracy: 0.68:  34%|███▍      | 11/32 [01:11<02:14,  6.41s/it]
Test accuracy: 0.67:  34%|███▍      | 11/32 [01:11<02:14,  6.41s/it]
Test accuracy: 0.56:  34%|███▍      | 11/32 [01:11<02:14,  6.41s/it]
Test accuracy: 0.66:  34%|███▍      | 11/32 [01:11<02:14,  6.41s/it]
Test accuracy: 0.65:  34%|███▍      | 11/32 [01:10<02:14,  6.41s/it]
Test accuracy: 0.62:  34%|███▍      | 11/32 [01:10<02:14,  6.41s/it]
Test accuracy: 0.67:  38%|███▊      | 12/32 [01:17<02:07,  6.38s/it]
Test accuracy: 0.66:  38%|███▊      | 12/32 [01:17<02:07,  6.38s/it]
Test accuracy: 0.66:  38%|███▊      | 12/32 [01:17<02:07,  6.38s/it]
Test accuracy: 0.62:  38%|███▊      | 12/32 [01:17<02:07,  6.38s/it]
Test accuracy: 0.56:  38%|███▊      | 12/32 [01:17<02:07,  6.38s/it]
Test accuracy: 0.65:  38%|███▊      | 12/32 [01:17<02:07,  6.38s/it]
Test accuracy: 0.57:  38%|███▊      | 12/32 [01:17<02:07,  6.38s/it]
Test accuracy: 0.62:  38%|███▊      | 12/32 [01:17<02:07,  6.38s/it]
Test accuracy: 0.66:  41%|████      | 13/32 [01:23<01:59,  6.31s/it]
Test accuracy: 0.65:  41%|████      | 13/32 [01:23<01:59,  6.31s/it]
Test accuracy: 0.65:  41%|████      | 13/32 [01:23<01:59,  6.31s/it]
Test accuracy: 0.64:  41%|████      | 13/32 [01:23<01:59,  6.31s/it]
Test accuracy: 0.62:  41%|████      | 13/32 [01:23<01:59,  6.31s/it]
Test accuracy: 0.57:  41%|████      | 13/32 [01:23<01:59,  6.31s/it]
Test accuracy: 0.58:  41%|████      | 13/32 [01:23<01:59,  6.31s/it]
Test accuracy: 0.62:  41%|████      | 13/32 [01:23<01:59,  6.31s/it]
Test accuracy: 0.65:  44%|████▍     | 14/32 [01:29<01:53,  6.32s/it]
Test accuracy: 0.58:  44%|████▍     | 14/32 [01:29<01:53,  6.32s/it]
Test accuracy: 0.64:  44%|████▍     | 14/32 [01:29<01:53,  6.32s/it]
Test accuracy: 0.67:  44%|████▍     | 14/32 [01:29<01:53,  6.32s/it]
Test accuracy: 0.58:  44%|████▍     | 14/32 [01:29<01:53,  6.32s/it]
Test accuracy: 0.62:  44%|████▍     | 14/32 [01:29<01:53,  6.32s/it]
Test accuracy: 0.66:  44%|████▍     | 14/32 [01:29<01:53,  6.32s/it]
Test accuracy: 0.62:  44%|████▍     | 14/32 [01:29<01:53,  6.32s/it]
Test accuracy: 0.66:  47%|████▋     | 15/32 [01:36<01:47,  6.32s/it]
Test accuracy: 0.67:  47%|████▋     | 15/32 [01:36<01:47,  6.32s/it]
Test accuracy: 0.65:  47%|████▋     | 15/32 [01:36<01:47,  6.32s/it]
Test accuracy: 0.58:  47%|████▋     | 15/32 [01:36<01:47,  6.32s/it]
Test accuracy: 0.66:  47%|████▋     | 15/32 [01:36<01:47,  6.32s/it]
Test accuracy: 0.62:  47%|████▋     | 15/32 [01:36<01:47,  6.32s/it]
Test accuracy: 0.57:  47%|████▋     | 15/32 [01:36<01:47,  6.32s/it]
Test accuracy: 0.61:  47%|████▋     | 15/32 [01:36<01:47,  6.32s/it]
Test accuracy: 0.66:  50%|█████     | 16/32 [01:42<01:40,  6.27s/it]
Test accuracy: 0.57:  50%|█████     | 16/32 [01:42<01:40,  6.27s/it]
Test accuracy: 0.65:  50%|█████     | 16/32 [01:42<01:40,  6.27s/it]
Test accuracy: 0.65:  50%|█████     | 16/32 [01:42<01:40,  6.27s/it]
Test accuracy: 0.61:  50%|█████     | 16/32 [01:42<01:40,  6.27s/it]
Test accuracy: 0.58:  50%|█████     | 16/32 [01:42<01:40,  6.27s/it]
Test accuracy: 0.65:  50%|█████     | 16/32 [01:42<01:40,  6.27s/it]
Test accuracy: 0.61:  50%|█████     | 16/32 [01:42<01:40,  6.27s/it]
Test accuracy: 0.65:  53%|█████▎    | 17/32 [01:48<01:34,  6.30s/it]
Test accuracy: 0.58:  53%|█████▎    | 17/32 [01:48<01:34,  6.30s/it]
Test accuracy: 0.65:  53%|█████▎    | 17/32 [01:48<01:34,  6.30s/it]
Test accuracy: 0.67:  53%|█████▎    | 17/32 [01:48<01:34,  6.30s/it]
Test accuracy: 0.57:  53%|█████▎    | 17/32 [01:48<01:34,  6.30s/it]
Test accuracy: 0.61:  53%|█████▎    | 17/32 [01:48<01:34,  6.30s/it]
Test accuracy: 0.65:  53%|█████▎    | 17/32 [01:48<01:34,  6.30s/it]
Test accuracy: 0.62:  53%|█████▎    | 17/32 [01:48<01:34,  6.30s/it]
Test accuracy: 0.65:  56%|█████▋    | 18/32 [01:55<01:28,  6.30s/it]
Test accuracy: 0.67:  56%|█████▋    | 18/32 [01:55<01:28,  6.30s/it]
Test accuracy: 0.64:  56%|█████▋    | 18/32 [01:55<01:28,  6.30s/it]
Test accuracy: 0.62:  56%|█████▋    | 18/32 [01:54<01:28,  6.30s/it]
Test accuracy: 0.57:  56%|█████▋    | 18/32 [01:55<01:28,  6.30s/it]
Test accuracy: 0.67:  56%|█████▋    | 18/32 [01:55<01:28,  6.30s/it]
Test accuracy: 0.58:  56%|█████▋    | 18/32 [01:55<01:28,  6.30s/it]
Test accuracy: 0.62:  56%|█████▋    | 18/32 [01:54<01:28,  6.30s/it]
Test accuracy: 0.67:  59%|█████▉    | 19/32 [02:01<01:21,  6.26s/it]
Test accuracy: 0.64:  59%|█████▉    | 19/32 [02:01<01:21,  6.26s/it]
Test accuracy: 0.58:  59%|█████▉    | 19/32 [02:01<01:21,  6.26s/it]
Test accuracy: 0.62:  59%|█████▉    | 19/32 [02:01<01:21,  6.26s/it]
Test accuracy: 0.65:  59%|█████▉    | 19/32 [02:01<01:21,  6.26s/it]
Test accuracy: 0.58:  59%|█████▉    | 19/32 [02:01<01:21,  6.26s/it]
Test accuracy: 0.62:  59%|█████▉    | 19/32 [02:01<01:21,  6.26s/it]
Test accuracy: 0.63:  59%|█████▉    | 19/32 [02:01<01:21,  6.26s/it]
Test accuracy: 0.65:  62%|██████▎   | 20/32 [02:07<01:15,  6.28s/it]
Test accuracy: 0.58:  62%|██████▎   | 20/32 [02:07<01:15,  6.28s/it]
Test accuracy: 0.63:  62%|██████▎   | 20/32 [02:07<01:15,  6.28s/it]
Test accuracy: 0.62:  62%|██████▎   | 20/32 [02:07<01:15,  6.28s/it]
Test accuracy: 0.58:  62%|██████▎   | 20/32 [02:07<01:15,  6.28s/it]
Test accuracy: 0.63:  62%|██████▎   | 20/32 [02:07<01:15,  6.28s/it]
Test accuracy: 0.63:  62%|██████▎   | 20/32 [02:07<01:15,  6.29s/it]
Test accuracy: 0.64:  62%|██████▎   | 20/32 [02:07<01:15,  6.29s/it]
Test accuracy: 0.63:  66%|██████▌   | 21/32 [02:14<01:11,  6.49s/it]
Test accuracy: 0.58:  66%|██████▌   | 21/32 [02:14<01:11,  6.49s/it]
Test accuracy: 0.64:  66%|██████▌   | 21/32 [02:14<01:11,  6.49s/it]
Test accuracy: 0.61:  66%|██████▌   | 21/32 [02:14<01:11,  6.49s/it]
Test accuracy: 0.57:  66%|██████▌   | 21/32 [02:14<01:11,  6.49s/it]
Test accuracy: 0.64:  66%|██████▌   | 21/32 [02:14<01:11,  6.49s/it]
Test accuracy: 0.63:  66%|██████▌   | 21/32 [02:14<01:11,  6.49s/it]
Test accuracy: 0.64:  66%|██████▌   | 21/32 [02:14<01:11,  6.49s/it]
Test accuracy: 0.64:  69%|██████▉   | 22/32 [02:20<01:03,  6.38s/it]
Test accuracy: 0.57:  69%|██████▉   | 22/32 [02:20<01:03,  6.39s/it]
Test accuracy: 0.61:  69%|██████▉   | 22/32 [02:20<01:03,  6.39s/it]
Test accuracy: 0.63:  69%|██████▉   | 22/32 [02:20<01:03,  6.38s/it]
Test accuracy: 0.57:  69%|██████▉   | 22/32 [02:20<01:03,  6.39s/it]
Test accuracy: 0.61:  69%|██████▉   | 22/32 [02:20<01:03,  6.39s/it]
Test accuracy: 0.64:  69%|██████▉   | 22/32 [02:20<01:03,  6.39s/it]
Test accuracy: 0.64:  69%|██████▉   | 22/32 [02:20<01:03,  6.39s/it]
Test accuracy: 0.57:  72%|███████▏  | 23/32 [02:26<00:57,  6.34s/it]
Test accuracy: 0.63:  72%|███████▏  | 23/32 [02:26<00:57,  6.34s/it]
Test accuracy: 0.62:  72%|███████▏  | 23/32 [02:26<00:57,  6.34s/it]
Test accuracy: 0.61:  72%|███████▏  | 23/32 [02:26<00:57,  6.34s/it]
Test accuracy: 0.58:  72%|███████▏  | 23/32 [02:26<00:57,  6.34s/it]
Test accuracy: 0.64:  72%|███████▏  | 23/32 [02:26<00:57,  6.34s/it]
Test accuracy: 0.62:  72%|███████▏  | 23/32 [02:26<00:57,  6.34s/it]
Test accuracy: 0.64:  72%|███████▏  | 23/32 [02:26<00:57,  6.34s/it]
Test accuracy: 0.62:  75%|███████▌  | 24/32 [02:33<00:52,  6.53s/it]
Test accuracy: 0.61:  75%|███████▌  | 24/32 [02:33<00:52,  6.53s/it]
Test accuracy: 0.58:  75%|███████▌  | 24/32 [02:33<00:52,  6.53s/it]
Test accuracy: 0.62:  75%|███████▌  | 24/32 [02:33<00:52,  6.53s/it]
Test accuracy: 0.58:  75%|███████▌  | 24/32 [02:33<00:52,  6.53s/it]
Test accuracy: 0.62:  75%|███████▌  | 24/32 [02:33<00:52,  6.53s/it]
Test accuracy: 0.64:  75%|███████▌  | 24/32 [02:33<00:52,  6.54s/it]
Test accuracy: 0.64:  75%|███████▌  | 24/32 [02:33<00:52,  6.54s/it]
Test accuracy: 0.61:  78%|███████▊  | 25/32 [02:40<00:45,  6.48s/it]
Test accuracy: 0.58:  78%|███████▊  | 25/32 [02:40<00:45,  6.48s/it]
Test accuracy: 0.62:  78%|███████▊  | 25/32 [02:40<00:45,  6.48s/it]
Test accuracy: 0.62:  78%|███████▊  | 25/32 [02:40<00:45,  6.48s/it]
Test accuracy: 0.58:  78%|███████▊  | 25/32 [02:40<00:45,  6.48s/it]
Test accuracy: 0.64:  78%|███████▊  | 25/32 [02:40<00:45,  6.48s/it]
Test accuracy: 0.62:  78%|███████▊  | 25/32 [02:40<00:45,  6.48s/it]
Test accuracy: 0.64:  78%|███████▊  | 25/32 [02:40<00:45,  6.48s/it]
Test accuracy: 0.62:  81%|████████▏ | 26/32 [02:46<00:38,  6.38s/it]
Test accuracy: 0.62:  81%|████████▏ | 26/32 [02:46<00:38,  6.38s/it]
Test accuracy: 0.58:  81%|████████▏ | 26/32 [02:46<00:38,  6.38s/it]
Test accuracy: 0.62:  81%|████████▏ | 26/32 [02:46<00:38,  6.38s/it]
Test accuracy: 0.64:  81%|████████▏ | 26/32 [02:46<00:38,  6.38s/it]
Test accuracy: 0.62:  81%|████████▏ | 26/32 [02:46<00:38,  6.38s/it]
Test accuracy: 0.59:  81%|████████▏ | 26/32 [02:46<00:38,  6.38s/it]
Test accuracy: 0.64:  81%|████████▏ | 26/32 [02:46<00:38,  6.38s/it]
Test accuracy: 0.62:  84%|████████▍ | 27/32 [02:52<00:31,  6.36s/it]
Test accuracy: 0.62:  84%|████████▍ | 27/32 [02:52<00:31,  6.36s/it]
Test accuracy: 0.59:  84%|████████▍ | 27/32 [02:52<00:31,  6.36s/it]
Test accuracy: 0.64:  84%|████████▍ | 27/32 [02:52<00:31,  6.36s/it]
Test accuracy: 0.62:  84%|████████▍ | 27/32 [02:52<00:31,  6.36s/it]
Test accuracy: 0.62:  84%|████████▍ | 27/32 [02:52<00:31,  6.36s/it]
Test accuracy: 0.59:  84%|████████▍ | 27/32 [02:52<00:31,  6.36s/it]
Test accuracy: 0.64:  84%|████████▍ | 27/32 [02:52<00:31,  6.36s/it]
Test accuracy: 0.62:  88%|████████▊ | 28/32 [02:59<00:26,  6.55s/it]
Test accuracy: 0.64:  88%|████████▊ | 28/32 [02:59<00:26,  6.55s/it]
Test accuracy: 0.62:  88%|████████▊ | 28/32 [02:59<00:26,  6.55s/it]
Test accuracy: 0.59:  88%|████████▊ | 28/32 [02:59<00:26,  6.55s/it]
Test accuracy: 0.6:  88%|████████▊ | 28/32 [02:59<00:26,  6.55s/it] 
Test accuracy: 0.63:  88%|████████▊ | 28/32 [02:59<00:26,  6.55s/it]
Test accuracy: 0.62:  88%|████████▊ | 28/32 [02:59<00:26,  6.56s/it]
Test accuracy: 0.62:  88%|████████▊ | 28/32 [02:59<00:26,  6.56s/it]
Test accuracy: 0.62:  91%|█████████ | 29/32 [03:06<00:19,  6.49s/it]
Test accuracy: 0.6:  91%|█████████ | 29/32 [03:06<00:19,  6.49s/it]
Test accuracy: 0.62:  91%|█████████ | 29/32 [03:06<00:19,  6.49s/it]
Test accuracy: 0.62:  91%|█████████ | 29/32 [03:06<00:19,  6.49s/it]
Test accuracy: 0.59:  91%|█████████ | 29/32 [03:06<00:19,  6.49s/it]
Test accuracy: 0.63:  91%|█████████ | 29/32 [03:05<00:19,  6.49s/it]
Test accuracy: 0.62:  91%|█████████ | 29/32 [03:06<00:19,  6.49s/it]
Test accuracy: 0.63:  91%|█████████ | 29/32 [03:05<00:19,  6.49s/it]
Test accuracy: 0.62:  94%|█████████▍| 30/32 [03:12<00:12,  6.38s/it]
Test accuracy: 0.61:  94%|█████████▍| 30/32 [03:12<00:12,  6.38s/it]
Test accuracy: 0.59:  94%|█████████▍| 30/32 [03:12<00:12,  6.38s/it]
Test accuracy: 0.62:  94%|█████████▍| 30/32 [03:12<00:12,  6.38s/it]
Test accuracy: 0.63:  94%|█████████▍| 30/32 [03:12<00:12,  6.38s/it]
Test accuracy: 0.6:  94%|█████████▍| 30/32 [03:12<00:12,  6.38s/it] 
Test accuracy: 0.62:  94%|█████████▍| 30/32 [03:12<00:12,  6.38s/it]
Test accuracy: 0.63:  94%|█████████▍| 30/32 [03:12<00:12,  6.38s/it]
Test accuracy: 0.6:  97%|█████████▋| 31/32 [03:19<00:06,  6.57s/it]
Test accuracy: 0.62:  97%|█████████▋| 31/32 [03:19<00:06,  6.57s/it]
Test accuracy: 0.6:  97%|█████████▋| 31/32 [03:19<00:06,  6.57s/it]
Test accuracy: 0.62:  97%|█████████▋| 31/32 [03:19<00:06,  6.57s/it]
Test accuracy: 0.63:  97%|█████████▋| 31/32 [03:19<00:06,  6.57s/it]
Test accuracy: 0.64:  97%|█████████▋| 31/32 [03:19<00:06,  6.57s/it]
Test accuracy: 0.61:  97%|█████████▋| 31/32 [03:19<00:06,  6.57s/it]
Test accuracy: 0.6:  97%|█████████▋| 31/32 [03:19<00:06,  6.57s/it] 
Test accuracy: 0.64: 100%|██████████| 32/32 [03:21<00:00,  5.30s/it]
Test accuracy: 0.64: 100%|██████████| 32/32 [03:21<00:00,  5.30s/it]
Test accuracy: 0.62: 100%|██████████| 32/32 [03:21<00:00,  5.30s/it]
Test accuracy: 0.62: 100%|██████████| 32/32 [03:21<00:00,  5.30s/it]
Test accuracy: 0.6: 100%|██████████| 32/32 [03:21<00:00,  5.30s/it]
Test accuracy: 0.6: 100%|██████████| 32/32 [03:21<00:00,  5.30s/it]
Test accuracy: 0.61: 100%|██████████| 32/32 [03:21<00:00,  5.30s/it]
Test accuracy: 0.6: 100%|██████████| 32/32 [03:21<00:00,  5.30s/it]
Test accuracy: 0.61: 100%|██████████| 32/32 [03:21<00:00,  6.30s/it]
Device 2: Cor=152, CoT=0, Total=250

Test accuracy: 0.62: 100%|██████████| 32/32 [03:21<00:00,  6.30s/it]
Device 1: Cor=155, CoT=0, Total=250

Test accuracy: 0.6: 100%|██████████| 32/32 [03:21<00:00,  6.30s/it]
Device 3: Cor=151, CoT=0, Total=250

Test accuracy: 0.64: 100%|██████████| 32/32 [03:21<00:00,  6.30s/it]
Device 0: Cor=160, CoT=0, Total=250

============================================================
EVALUATION RESULTS (Epoch 6)
============================================================
Config: data_curriculum=ON, thought_curriculum=OFF
thought_stage=10, data_stage=3, n_latent_tokens=10
Training on difficulties: 0-3

Overall accuracy: 618 / 1000 = 0.6180

Per-difficulty accuracy:
  diff=1: 95/100 = 0.9500 (trained)
  diff=2: 86/100 = 0.8600 (trained)
  diff=3: 79/100 = 0.7900 (trained)
  diff=4: 52/100 = 0.5200 (untrained)
  diff=5: 46/100 = 0.4600 (untrained)
  diff=6: 52/100 = 0.5200 (untrained)
  diff=7: 56/100 = 0.5600 (untrained)
  diff=8: 49/100 = 0.4900 (untrained)
  diff=9: 52/100 = 0.5200 (untrained)
  diff=10: 51/100 = 0.5100 (untrained)

Trained (diff 0-3): 260/300 = 0.8667
============================================================

  Staging check: acc_trained=0.8667, acc_diff3=0.7900 (threshold=0.9)
Warning: failed to upload checkpoint_7 to HF: (Request ID: Root=1-6a1b7103-2dfe3cb82fda10ab480192b1;c7f1bd69-048a-42fc-b424-0aca7d16ea3a)

403 Forbidden: You need to setup automatic credit recharge in order to upload more data. You can do so at /settings/billing..
Cannot access content at: https://huggingface.co/seyedparsa/thoughtformer-checkpoints.git/info/lfs/objects/batch.
Make sure your token has the correct permissions.

============================================================
EPOCH 7/100 (thought_stage=10, data_stage=3)
============================================================
  thought_stage=10, c_thought=1, max_difficulty=3

Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):  16%|█▌        | 156/1000 [00:00<00:00, 986.19 examples/s]
Map (num_proc=4):  16%|█▌        | 157/1000 [00:00<00:00, 1000.58 examples/s]
Map (num_proc=4):  16%|█▌        | 156/1000 [00:00<00:00, 940.67 examples/s]
Map (num_proc=4):  82%|████████▏ | 817/1000 [00:00<00:00, 3518.27 examples/s]
Map (num_proc=4):  81%|████████  | 811/1000 [00:00<00:00, 3530.22 examples/s]
Map (num_proc=4):  81%|████████  | 811/1000 [00:00<00:00, 3475.80 examples/s]
Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2554.07 examples/s]

Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2534.88 examples/s]

Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2469.26 examples/s]

Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Map (num_proc=4):  16%|█▌        | 156/1000 [00:00<00:01, 525.98 examples/s]
Map (num_proc=4):  82%|████████▏ | 817/1000 [00:00<00:00, 2436.03 examples/s]
Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:04, 1993.84 examples/s]
Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:04, 1997.99 examples/s]
Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:04, 1975.97 examples/s]
Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 1254.72 examples/s]

Filter (num_proc=4):  20%|██        | 2000/10000 [00:00<00:02, 3646.76 examples/s]
Filter (num_proc=4):  20%|██        | 2000/10000 [00:00<00:02, 3710.85 examples/s]
Filter (num_proc=4):  20%|██        | 2000/10000 [00:00<00:02, 3586.13 examples/s]
Filter (num_proc=4):  30%|███       | 3000/10000 [00:00<00:01, 4902.61 examples/s]
Filter (num_proc=4):  40%|████      | 4000/10000 [00:00<00:00, 7026.34 examples/s]
Filter (num_proc=4):  40%|████      | 4000/10000 [00:00<00:00, 7044.19 examples/s]
Filter (num_proc=4):  50%|█████     | 5000/10000 [00:00<00:00, 8283.63 examples/s]
Filter (num_proc=4):  60%|██████    | 6000/10000 [00:00<00:00, 9919.88 examples/s]
Filter (num_proc=4):  55%|█████▌    | 5500/10000 [00:00<00:00, 8606.45 examples/s]
Filter (num_proc=4):  70%|███████   | 7000/10000 [00:00<00:00, 10779.82 examples/s]
Filter (num_proc=4):  80%|████████  | 8000/10000 [00:00<00:00, 11960.92 examples/s]
Filter (num_proc=4):  80%|████████  | 8000/10000 [00:00<00:00, 12012.71 examples/s]
Filter (num_proc=4):  90%|█████████ | 9000/10000 [00:01<00:00, 12550.64 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 11374.44 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 11781.49 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 7702.49 examples/s] 

Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 7474.94 examples/s]

Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 7661.30 examples/s] 

Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:06, 1397.71 examples/s]
Filter (num_proc=4):  20%|██        | 2000/10000 [00:00<00:03, 2438.04 examples/s]
Filter (num_proc=4):  35%|███▌      | 3500/10000 [00:01<00:01, 4192.19 examples/s]
Filter (num_proc=4):  60%|██████    | 6000/10000 [00:01<00:00, 6832.13 examples/s]
Filter (num_proc=4):  85%|████████▌ | 8500/10000 [00:01<00:00, 8635.55 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 8304.66 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 5137.20 examples/s]

Map (num_proc=4):   0%|          | 0/3000 [00:00<?, ? examples/s]
Map (num_proc=4):   4%|▍         | 133/3000 [00:00<00:19, 143.97 examples/s]
Map (num_proc=4):  18%|█▊        | 537/3000 [00:01<00:04, 615.48 examples/s]
Map (num_proc=4):  25%|██▌       | 750/3000 [00:01<00:03, 643.05 examples/s]
Map (num_proc=4):  34%|███▎      | 1010/3000 [00:01<00:02, 826.06 examples/s]
Map (num_proc=4):  50%|█████     | 1500/3000 [00:01<00:01, 1067.30 examples/s]
Map (num_proc=4):  59%|█████▉    | 1766/3000 [00:02<00:01, 1190.42 examples/s]
Map (num_proc=4):  75%|███████▌  | 2250/3000 [00:02<00:00, 1286.51 examples/s]
Map (num_proc=4):  84%|████████▍ | 2514/3000 [00:02<00:00, 1365.01 examples/s]
Map (num_proc=4):  98%|█████████▊| 2927/3000 [00:02<00:00, 1771.71 examples/s]
Map (num_proc=4): 100%|██████████| 3000/3000 [00:03<00:00, 880.14 examples/s] 

Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 1429.99 examples/s]
Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 1385.47 examples/s]
Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 1342.73 examples/s]
Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2802.75 examples/s]

Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2646.38 examples/s]

Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2571.47 examples/s]

Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 753.81 examples/s]
Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 1438.00 examples/s]

Map (num_proc=4):   0%|          | 0/300 [00:00<?, ? examples/s]
Map (num_proc=4):  25%|██▌       | 75/300 [00:00<00:01, 213.03 examples/s]
Map (num_proc=4):  75%|███████▌  | 225/300 [00:00<00:00, 519.14 examples/s]
Map (num_proc=4): 100%|██████████| 300/300 [00:00<00:00, 335.46 examples/s]

Training Epoch: 8:   0%|          | 0/94 [00:00<?, ?it/s]
Training Epoch: 8:   0%|          | 0/94 [00:00<?, ?it/s]
Training Epoch: 8:   0%|          | 0/94 [00:00<?, ?it/s]
Training Epoch: 8:   0%|          | 0/94 [00:00<?, ?it/s]logging training data
rank 3 | gpu 3 | alloc=4.20 GB | reserved=53.25 GB | max=41.99 GB
rank 1 | gpu 1 | alloc=4.20 GB | reserved=55.12 GB | max=41.98 GB
rank 2 | gpu 2 | alloc=4.20 GB | reserved=51.39 GB | max=41.98 GB
rank 0 | gpu 0 | alloc=4.20 GB | reserved=53.25 GB | max=41.99 GB

Training Epoch: 8/100, batch 0/188 completed (loss: 0.0507:   0%|          | 0/94 [00:03<?, ?it/s]
Training Epoch: 8/100, batch 0/188 completed (loss: 0.1094:   0%|          | 0/94 [00:03<?, ?it/s]
Training Epoch: 8/100, batch 0/188 completed (loss: 0.0703:   0%|          | 0/94 [00:03<?, ?it/s]
Training Epoch: 8/100, batch 0/188 completed (loss: 0.0788:   0%|          | 0/94 [00:03<?, ?it/s]
Training Epoch: 8/100, batch 0/188 completed (loss: 0.0703:   1%|          | 1/94 [00:06<09:38,  6.22s/it]
Training Epoch: 8/100, batch 0/188 completed (loss: 0.0788:   1%|          | 1/94 [00:06<09:40,  6.24s/it]
Training Epoch: 8/100, batch 0/188 completed (loss: 0.1094:   1%|          | 1/94 [00:06<09:40,  6.24s/it]
Training Epoch: 8/100, batch 0/188 completed (loss: 0.0507:   1%|          | 1/94 [00:06<09:40,  6.24s/it]
Training Epoch: 8/100, batch 1/188 completed (loss: 0.0856:   1%|          | 1/94 [00:06<09:40,  6.24s/it]
Training Epoch: 8/100, batch 1/188 completed (loss: 0.1968:   1%|          | 1/94 [00:06<09:38,  6.22s/it]
Training Epoch: 8/100, batch 1/188 completed (loss: 0.0737:   1%|          | 1/94 [00:06<09:40,  6.24s/it]
Training Epoch: 8/100, batch 1/188 completed (loss: 0.0884:   1%|          | 1/94 [00:06<09:40,  6.24s/it]
Training Epoch: 8/100, batch 2/188 completed (loss: 0.2129:   1%|          | 1/94 [00:09<09:40,  6.24s/it]
Training Epoch: 8/100, batch 2/188 completed (loss: 0.0607:   1%|          | 1/94 [00:09<09:38,  6.22s/it]
Training Epoch: 8/100, batch 2/188 completed (loss: 0.1537:   1%|          | 1/94 [00:09<09:40,  6.24s/it]
Training Epoch: 8/100, batch 2/188 completed (loss: 0.251:   1%|          | 1/94 [00:09<09:40,  6.24s/it] 
Training Epoch: 8/100, batch 2/188 completed (loss: 0.0607:   2%|▏         | 2/94 [00:12<09:13,  6.02s/it]
Training Epoch: 8/100, batch 2/188 completed (loss: 0.1537:   2%|▏         | 2/94 [00:12<09:13,  6.01s/it]
Training Epoch: 8/100, batch 2/188 completed (loss: 0.251:   2%|▏         | 2/94 [00:12<09:14,  6.03s/it]
Training Epoch: 8/100, batch 2/188 completed (loss: 0.2129:   2%|▏         | 2/94 [00:12<09:14,  6.02s/it]
Training Epoch: 8/100, batch 3/188 completed (loss: 0.1049:   2%|▏         | 2/94 [00:12<09:14,  6.02s/it]
Training Epoch: 8/100, batch 3/188 completed (loss: 0.0806:   2%|▏         | 2/94 [00:12<09:13,  6.02s/it]
Training Epoch: 8/100, batch 3/188 completed (loss: 0.1016:   2%|▏         | 2/94 [00:12<09:13,  6.01s/it]
Training Epoch: 8/100, batch 3/188 completed (loss: 0.1387:   2%|▏         | 2/94 [00:12<09:14,  6.03s/it]
Training Epoch: 8/100, batch 4/188 completed (loss: 0.1616:   2%|▏         | 2/94 [00:14<09:13,  6.02s/it]
Training Epoch: 8/100, batch 4/188 completed (loss: 0.1302:   2%|▏         | 2/94 [00:14<09:14,  6.02s/it]
Training Epoch: 8/100, batch 4/188 completed (loss: 0.1109:   2%|▏         | 2/94 [00:14<09:13,  6.01s/it]
Training Epoch: 8/100, batch 4/188 completed (loss: 0.1567:   2%|▏         | 2/94 [00:14<09:14,  6.03s/it]
Training Epoch: 8/100, batch 4/188 completed (loss: 0.1567:   3%|▎         | 3/94 [00:17<08:50,  5.83s/it]
Training Epoch: 8/100, batch 4/188 completed (loss: 0.1109:   3%|▎         | 3/94 [00:17<08:50,  5.83s/it]
Training Epoch: 8/100, batch 4/188 completed (loss: 0.1302:   3%|▎         | 3/94 [00:17<08:50,  5.83s/it]
Training Epoch: 8/100, batch 4/188 completed (loss: 0.1616:   3%|▎         | 3/94 [00:17<08:50,  5.84s/it]
Training Epoch: 8/100, batch 5/188 completed (loss: 0.0506:   3%|▎         | 3/94 [00:17<08:50,  5.83s/it]
Training Epoch: 8/100, batch 5/188 completed (loss: 0.0687:   3%|▎         | 3/94 [00:17<08:50,  5.84s/it]
Training Epoch: 8/100, batch 5/188 completed (loss: 0.0627:   3%|▎         | 3/94 [00:17<08:50,  5.83s/it]
Training Epoch: 8/100, batch 5/188 completed (loss: 0.0296:   3%|▎         | 3/94 [00:17<08:50,  5.83s/it]
Training Epoch: 8/100, batch 6/188 completed (loss: 0.252:   3%|▎         | 3/94 [00:20<08:50,  5.84s/it] 
Training Epoch: 8/100, batch 6/188 completed (loss: 0.1417:   3%|▎         | 3/94 [00:20<08:50,  5.83s/it]
Training Epoch: 8/100, batch 6/188 completed (loss: 0.1745:   3%|▎         | 3/94 [00:20<08:50,  5.83s/it]
Training Epoch: 8/100, batch 6/188 completed (loss: 0.0995:   3%|▎         | 3/94 [00:20<08:50,  5.83s/it]
Training Epoch: 8/100, batch 6/188 completed (loss: 0.252:   4%|▍         | 4/94 [00:23<08:43,  5.82s/it]
Training Epoch: 8/100, batch 6/188 completed (loss: 0.1745:   4%|▍         | 4/94 [00:23<08:44,  5.82s/it]
Training Epoch: 8/100, batch 6/188 completed (loss: 0.0995:   4%|▍         | 4/94 [00:23<08:45,  5.83s/it]
Training Epoch: 8/100, batch 6/188 completed (loss: 0.1417:   4%|▍         | 4/94 [00:23<08:44,  5.83s/it]
Training Epoch: 8/100, batch 7/188 completed (loss: 0.1801:   4%|▍         | 4/94 [00:23<08:44,  5.83s/it]
Training Epoch: 8/100, batch 7/188 completed (loss: 0.0348:   4%|▍         | 4/94 [00:23<08:44,  5.82s/it]
Training Epoch: 8/100, batch 7/188 completed (loss: 0.0488:   4%|▍         | 4/94 [00:23<08:43,  5.82s/it]
Training Epoch: 8/100, batch 7/188 completed (loss: 0.1126:   4%|▍         | 4/94 [00:23<08:45,  5.83s/it]
Training Epoch: 8/100, batch 8/188 completed (loss: 0.0572:   4%|▍         | 4/94 [00:26<08:43,  5.82s/it]
Training Epoch: 8/100, batch 8/188 completed (loss: 0.2201:   4%|▍         | 4/94 [00:26<08:44,  5.82s/it]
Training Epoch: 8/100, batch 8/188 completed (loss: 0.1106:   4%|▍         | 4/94 [00:26<08:44,  5.83s/it]
Training Epoch: 8/100, batch 8/188 completed (loss: 0.2917:   4%|▍         | 4/94 [00:26<08:45,  5.83s/it]
Training Epoch: 8/100, batch 8/188 completed (loss: 0.1106:   5%|▌         | 5/94 [00:29<08:44,  5.89s/it]
Training Epoch: 8/100, batch 8/188 completed (loss: 0.2917:   5%|▌         | 5/94 [00:29<08:44,  5.89s/it]
Training Epoch: 8/100, batch 8/188 completed (loss: 0.2201:   5%|▌         | 5/94 [00:29<08:44,  5.89s/it]
Training Epoch: 8/100, batch 8/188 completed (loss: 0.0572:   5%|▌         | 5/94 [00:29<08:44,  5.89s/it]
Training Epoch: 8/100, batch 9/188 completed (loss: 0.0638:   5%|▌         | 5/94 [00:29<08:44,  5.89s/it]
Training Epoch: 8/100, batch 9/188 completed (loss: 0.1248:   5%|▌         | 5/94 [00:29<08:44,  5.89s/it]
Training Epoch: 8/100, batch 9/188 completed (loss: 0.1299:   5%|▌         | 5/94 [00:29<08:44,  5.89s/it]
Training Epoch: 8/100, batch 9/188 completed (loss: 0.1132:   5%|▌         | 5/94 [00:29<08:44,  5.89s/it]
Training Epoch: 8/100, batch 10/188 completed (loss: 0.1118:   5%|▌         | 5/94 [00:32<08:44,  5.89s/it]
Training Epoch: 8/100, batch 10/188 completed (loss: 0.1409:   5%|▌         | 5/94 [00:32<08:44,  5.89s/it]
Training Epoch: 8/100, batch 10/188 completed (loss: 0.0771:   5%|▌         | 5/94 [00:32<08:44,  5.89s/it]
Training Epoch: 8/100, batch 10/188 completed (loss: 0.0942:   5%|▌         | 5/94 [00:32<08:44,  5.89s/it]
Training Epoch: 8/100, batch 10/188 completed (loss: 0.0942:   6%|▋         | 6/94 [00:35<08:32,  5.82s/it]
Training Epoch: 8/100, batch 10/188 completed (loss: 0.1118:   6%|▋         | 6/94 [00:35<08:32,  5.82s/it]
Training Epoch: 8/100, batch 10/188 completed (loss: 0.0771:   6%|▋         | 6/94 [00:35<08:32,  5.83s/it]
Training Epoch: 8/100, batch 10/188 completed (loss: 0.1409:   6%|▋         | 6/94 [00:35<08:32,  5.83s/it]
Training Epoch: 8/100, batch 11/188 completed (loss: 0.0899:   6%|▋         | 6/94 [00:35<08:32,  5.82s/it]
Training Epoch: 8/100, batch 11/188 completed (loss: 0.0783:   6%|▋         | 6/94 [00:35<08:32,  5.83s/it]
Training Epoch: 8/100, batch 11/188 completed (loss: 0.1381:   6%|▋         | 6/94 [00:35<08:32,  5.83s/it]
Training Epoch: 8/100, batch 11/188 completed (loss: 0.0621:   6%|▋         | 6/94 [00:35<08:32,  5.82s/it]
Training Epoch: 8/100, batch 12/188 completed (loss: 0.1558:   6%|▋         | 6/94 [00:38<08:32,  5.83s/it]
Training Epoch: 8/100, batch 12/188 completed (loss: 0.1188:   6%|▋         | 6/94 [00:38<08:32,  5.82s/it]
Training Epoch: 8/100, batch 12/188 completed (loss: 0.1014:   6%|▋         | 6/94 [00:38<08:32,  5.83s/it]
Training Epoch: 8/100, batch 12/188 completed (loss: 0.142:   6%|▋         | 6/94 [00:38<08:32,  5.82s/it] 
Training Epoch: 8/100, batch 12/188 completed (loss: 0.1014:   7%|▋         | 7/94 [00:41<08:32,  5.89s/it]
Training Epoch: 8/100, batch 12/188 completed (loss: 0.1558:   7%|▋         | 7/94 [00:41<08:32,  5.89s/it]
Training Epoch: 8/100, batch 12/188 completed (loss: 0.142:   7%|▋         | 7/94 [00:41<08:32,  5.90s/it]
Training Epoch: 8/100, batch 12/188 completed (loss: 0.1188:   7%|▋         | 7/94 [00:41<08:32,  5.90s/it]
Training Epoch: 8/100, batch 13/188 completed (loss: 0.1979:   7%|▋         | 7/94 [00:41<08:32,  5.89s/it]
Training Epoch: 8/100, batch 13/188 completed (loss: 0.1688:   7%|▋         | 7/94 [00:41<08:32,  5.90s/it]
Training Epoch: 8/100, batch 13/188 completed (loss: 0.2348:   7%|▋         | 7/94 [00:41<08:32,  5.89s/it]
Training Epoch: 8/100, batch 13/188 completed (loss: 0.06:   7%|▋         | 7/94 [00:41<08:32,  5.90s/it] 
Training Epoch: 8/100, batch 14/188 completed (loss: 0.0571:   7%|▋         | 7/94 [00:44<08:32,  5.89s/it]
Training Epoch: 8/100, batch 14/188 completed (loss: 0.1385:   7%|▋         | 7/94 [00:44<08:32,  5.89s/it]
Training Epoch: 8/100, batch 14/188 completed (loss: 0.11:   7%|▋         | 7/94 [00:44<08:32,  5.90s/it]  
Training Epoch: 8/100, batch 14/188 completed (loss: 0.075:   7%|▋         | 7/94 [00:44<08:32,  5.90s/it]
Training Epoch: 8/100, batch 14/188 completed (loss: 0.0571:   9%|▊         | 8/94 [00:47<08:26,  5.88s/it]
Training Epoch: 8/100, batch 14/188 completed (loss: 0.075:   9%|▊         | 8/94 [00:47<08:26,  5.89s/it]
Training Epoch: 8/100, batch 14/188 completed (loss: 0.1385:   9%|▊         | 8/94 [00:47<08:26,  5.89s/it]
Training Epoch: 8/100, batch 14/188 completed (loss: 0.11:   9%|▊         | 8/94 [00:47<08:26,  5.89s/it]
Training Epoch: 8/100, batch 15/188 completed (loss: 0.0819:   9%|▊         | 8/94 [00:47<08:26,  5.89s/it]
Training Epoch: 8/100, batch 15/188 completed (loss: 0.0562:   9%|▊         | 8/94 [00:47<08:26,  5.89s/it]
Training Epoch: 8/100, batch 15/188 completed (loss: 0.0713:   9%|▊         | 8/94 [00:47<08:26,  5.88s/it]
Training Epoch: 8/100, batch 15/188 completed (loss: 0.1685:   9%|▊         | 8/94 [00:47<08:26,  5.89s/it]
Training Epoch: 8/100, batch 16/188 completed (loss: 0.044:   9%|▊         | 8/94 [00:50<08:26,  5.88s/it] 
Training Epoch: 8/100, batch 16/188 completed (loss: 0.2957:   9%|▊         | 8/94 [00:50<08:26,  5.89s/it]
Training Epoch: 8/100, batch 16/188 completed (loss: 0.0779:   9%|▊         | 8/94 [00:50<08:26,  5.89s/it]
Training Epoch: 8/100, batch 16/188 completed (loss: 0.0701:   9%|▊         | 8/94 [00:50<08:26,  5.89s/it]
Training Epoch: 8/100, batch 16/188 completed (loss: 0.2957:  10%|▉         | 9/94 [00:53<08:20,  5.89s/it]
Training Epoch: 8/100, batch 16/188 completed (loss: 0.0779:  10%|▉         | 9/94 [00:53<08:21,  5.89s/it]
Training Epoch: 8/100, batch 16/188 completed (loss: 0.044:  10%|▉         | 9/94 [00:53<08:21,  5.89s/it]
Training Epoch: 8/100, batch 16/188 completed (loss: 0.0701:  10%|▉         | 9/94 [00:53<08:21,  5.90s/it]
Training Epoch: 8/100, batch 17/188 completed (loss: 0.1005:  10%|▉         | 9/94 [00:53<08:21,  5.89s/it]
Training Epoch: 8/100, batch 17/188 completed (loss: 0.0636:  10%|▉         | 9/94 [00:53<08:20,  5.89s/it]
Training Epoch: 8/100, batch 17/188 completed (loss: 0.2041:  10%|▉         | 9/94 [00:53<08:21,  5.89s/it]
Training Epoch: 8/100, batch 17/188 completed (loss: 0.0857:  10%|▉         | 9/94 [00:53<08:21,  5.90s/it]
Training Epoch: 8/100, batch 18/188 completed (loss: 0.1824:  10%|▉         | 9/94 [00:55<08:21,  5.89s/it]
Training Epoch: 8/100, batch 18/188 completed (loss: 0.0966:  10%|▉         | 9/94 [00:55<08:21,  5.89s/it]
Training Epoch: 8/100, batch 18/188 completed (loss: 0.0881:  10%|▉         | 9/94 [00:55<08:20,  5.89s/it]
Training Epoch: 8/100, batch 18/188 completed (loss: 0.162:  10%|▉         | 9/94 [00:55<08:21,  5.90s/it] 
Training Epoch: 8/100, batch 18/188 completed (loss: 0.162:  11%|█         | 10/94 [00:59<08:19,  5.95s/it]
Training Epoch: 8/100, batch 18/188 completed (loss: 0.0966:  11%|█         | 10/94 [00:59<08:19,  5.95s/it]
Training Epoch: 8/100, batch 18/188 completed (loss: 0.1824:  11%|█         | 10/94 [00:59<08:19,  5.95s/it]
Training Epoch: 8/100, batch 18/188 completed (loss: 0.0881:  11%|█         | 10/94 [00:59<08:19,  5.95s/it]
Training Epoch: 8/100, batch 19/188 completed (loss: 0.0696:  11%|█         | 10/94 [00:59<08:19,  5.95s/it]
Training Epoch: 8/100, batch 19/188 completed (loss: 0.1274:  11%|█         | 10/94 [00:59<08:19,  5.95s/it]
Training Epoch: 8/100, batch 19/188 completed (loss: 0.0401:  11%|█         | 10/94 [00:59<08:19,  5.95s/it]
Training Epoch: 8/100, batch 19/188 completed (loss: 0.1707:  11%|█         | 10/94 [00:59<08:19,  5.95s/it]
Training Epoch: 8/100, batch 20/188 completed (loss: 0.1925:  11%|█         | 10/94 [01:01<08:19,  5.95s/it]
Training Epoch: 8/100, batch 20/188 completed (loss: 0.1555:  11%|█         | 10/94 [01:01<08:19,  5.95s/it]
Training Epoch: 8/100, batch 20/188 completed (loss: 0.1344:  11%|█         | 10/94 [01:01<08:19,  5.95s/it]
Training Epoch: 8/100, batch 20/188 completed (loss: 0.1916:  11%|█         | 10/94 [01:02<08:19,  5.95s/it]
Training Epoch: 8/100, batch 20/188 completed (loss: 0.1344:  12%|█▏        | 11/94 [01:04<08:07,  5.87s/it]
Training Epoch: 8/100, batch 20/188 completed (loss: 0.1555:  12%|█▏        | 11/94 [01:04<08:07,  5.87s/it]
Training Epoch: 8/100, batch 20/188 completed (loss: 0.1916:  12%|█▏        | 11/94 [01:04<08:07,  5.87s/it]
Training Epoch: 8/100, batch 20/188 completed (loss: 0.1925:  12%|█▏        | 11/94 [01:04<08:07,  5.88s/it]
Training Epoch: 8/100, batch 21/188 completed (loss: 0.1151:  12%|█▏        | 11/94 [01:04<08:07,  5.87s/it]
Training Epoch: 8/100, batch 21/188 completed (loss: 0.1215:  12%|█▏        | 11/94 [01:04<08:07,  5.87s/it]
Training Epoch: 8/100, batch 21/188 completed (loss: 0.3037:  12%|█▏        | 11/94 [01:04<08:07,  5.88s/it]
Training Epoch: 8/100, batch 21/188 completed (loss: 0.0328:  12%|█▏        | 11/94 [01:04<08:07,  5.87s/it]
Training Epoch: 8/100, batch 22/188 completed (loss: 0.1741:  12%|█▏        | 11/94 [01:07<08:07,  5.87s/it]
Training Epoch: 8/100, batch 22/188 completed (loss: 0.1319:  12%|█▏        | 11/94 [01:07<08:07,  5.88s/it]
Training Epoch: 8/100, batch 22/188 completed (loss: 0.1013:  12%|█▏        | 11/94 [01:07<08:07,  5.87s/it]
Training Epoch: 8/100, batch 22/188 completed (loss: 0.0454:  12%|█▏        | 11/94 [01:07<08:07,  5.87s/it]
Training Epoch: 8/100, batch 22/188 completed (loss: 0.1013:  13%|█▎        | 12/94 [01:10<08:01,  5.88s/it]
Training Epoch: 8/100, batch 22/188 completed (loss: 0.0454:  13%|█▎        | 12/94 [01:10<08:01,  5.88s/it]
Training Epoch: 8/100, batch 22/188 completed (loss: 0.1741:  13%|█▎        | 12/94 [01:10<08:01,  5.88s/it]
Training Epoch: 8/100, batch 22/188 completed (loss: 0.1319:  13%|█▎        | 12/94 [01:10<08:01,  5.88s/it]
Training Epoch: 8/100, batch 23/188 completed (loss: 0.0234:  13%|█▎        | 12/94 [01:10<08:01,  5.88s/it]
Training Epoch: 8/100, batch 23/188 completed (loss: 0.1382:  13%|█▎        | 12/94 [01:10<08:01,  5.88s/it]
Training Epoch: 8/100, batch 23/188 completed (loss: 0.0557:  13%|█▎        | 12/94 [01:10<08:01,  5.88s/it]
Training Epoch: 8/100, batch 23/188 completed (loss: 0.0943:  13%|█▎        | 12/94 [01:10<08:01,  5.88s/it]
Training Epoch: 8/100, batch 24/188 completed (loss: 0.1788:  13%|█▎        | 12/94 [01:13<08:01,  5.88s/it]
Training Epoch: 8/100, batch 24/188 completed (loss: 0.2002:  13%|█▎        | 12/94 [01:13<08:01,  5.88s/it]
Training Epoch: 8/100, batch 24/188 completed (loss: 0.0875:  13%|█▎        | 12/94 [01:13<08:01,  5.88s/it]
Training Epoch: 8/100, batch 24/188 completed (loss: 0.0363:  13%|█▎        | 12/94 [01:13<08:01,  5.88s/it]
Training Epoch: 8/100, batch 24/188 completed (loss: 0.0363:  14%|█▍        | 13/94 [01:16<08:00,  5.94s/it]
Training Epoch: 8/100, batch 24/188 completed (loss: 0.1788:  14%|█▍        | 13/94 [01:16<08:00,  5.93s/it]
Training Epoch: 8/100, batch 24/188 completed (loss: 0.0875:  14%|█▍        | 13/94 [01:16<08:00,  5.94s/it]
Training Epoch: 8/100, batch 24/188 completed (loss: 0.2002:  14%|█▍        | 13/94 [01:16<08:01,  5.94s/it]
Training Epoch: 8/100, batch 25/188 completed (loss: 0.0462:  14%|█▍        | 13/94 [01:16<08:00,  5.94s/it]
Training Epoch: 8/100, batch 25/188 completed (loss: 0.1176:  14%|█▍        | 13/94 [01:16<08:01,  5.94s/it]
Training Epoch: 8/100, batch 25/188 completed (loss: 0.2677:  14%|█▍        | 13/94 [01:16<08:00,  5.93s/it]
Training Epoch: 8/100, batch 25/188 completed (loss: 0.19:  14%|█▍        | 13/94 [01:16<08:00,  5.94s/it]  
Training Epoch: 8/100, batch 26/188 completed (loss: 0.0852:  14%|█▍        | 13/94 [01:19<08:01,  5.94s/it]
Training Epoch: 8/100, batch 26/188 completed (loss: 0.0511:  14%|█▍        | 13/94 [01:19<08:00,  5.94s/it]
Training Epoch: 8/100, batch 26/188 completed (loss: 0.2584:  14%|█▍        | 13/94 [01:19<08:00,  5.93s/it]
Training Epoch: 8/100, batch 26/188 completed (loss: 0.0547:  14%|█▍        | 13/94 [01:19<08:00,  5.94s/it]
Training Epoch: 8/100, batch 26/188 completed (loss: 0.0511:  15%|█▍        | 14/94 [01:22<07:50,  5.88s/it]
Training Epoch: 8/100, batch 26/188 completed (loss: 0.0547:  15%|█▍        | 14/94 [01:22<07:50,  5.88s/it]
Training Epoch: 8/100, batch 26/188 completed (loss: 0.0852:  15%|█▍        | 14/94 [01:22<07:50,  5.88s/it]
Training Epoch: 8/100, batch 26/188 completed (loss: 0.2584:  15%|█▍        | 14/94 [01:22<07:50,  5.88s/it]
Training Epoch: 8/100, batch 27/188 completed (loss: 0.0303:  15%|█▍        | 14/94 [01:22<07:50,  5.88s/it]
Training Epoch: 8/100, batch 27/188 completed (loss: 0.0812:  15%|█▍        | 14/94 [01:22<07:50,  5.88s/it]
Training Epoch: 8/100, batch 27/188 completed (loss: 0.0957:  15%|█▍        | 14/94 [01:22<07:50,  5.88s/it]
Training Epoch: 8/100, batch 27/188 completed (loss: 0.0331:  15%|█▍        | 14/94 [01:22<07:50,  5.88s/it]
Training Epoch: 8/100, batch 28/188 completed (loss: 0.028:  15%|█▍        | 14/94 [01:25<07:50,  5.88s/it] 
Training Epoch: 8/100, batch 28/188 completed (loss: 0.0753:  15%|█▍        | 14/94 [01:25<07:50,  5.88s/it]
Training Epoch: 8/100, batch 28/188 completed (loss: 0.0452:  15%|█▍        | 14/94 [01:25<07:50,  5.88s/it]
Training Epoch: 8/100, batch 28/188 completed (loss: 0.1512:  15%|█▍        | 14/94 [01:25<07:50,  5.88s/it]
Training Epoch: 8/100, batch 28/188 completed (loss: 0.1512:  16%|█▌        | 15/94 [01:28<07:53,  5.99s/it]
Training Epoch: 8/100, batch 28/188 completed (loss: 0.0753:  16%|█▌        | 15/94 [01:28<07:53,  5.99s/it]
Training Epoch: 8/100, batch 28/188 completed (loss: 0.028:  16%|█▌        | 15/94 [01:28<07:53,  5.99s/it]
Training Epoch: 8/100, batch 28/188 completed (loss: 0.0452:  16%|█▌        | 15/94 [01:28<07:53,  5.99s/it]
Training Epoch: 8/100, batch 29/188 completed (loss: 0.0635:  16%|█▌        | 15/94 [01:28<07:53,  5.99s/it]
Training Epoch: 8/100, batch 29/188 completed (loss: 0.0463:  16%|█▌        | 15/94 [01:28<07:53,  5.99s/it]
Training Epoch: 8/100, batch 29/188 completed (loss: 0.0582:  16%|█▌        | 15/94 [01:28<07:53,  5.99s/it]
Training Epoch: 8/100, batch 29/188 completed (loss: 0.1253:  16%|█▌        | 15/94 [01:28<07:53,  5.99s/it]
Training Epoch: 8/100, batch 30/188 completed (loss: 0.0629:  16%|█▌        | 15/94 [01:31<07:53,  5.99s/it]
Training Epoch: 8/100, batch 30/188 completed (loss: 0.0693:  16%|█▌        | 15/94 [01:31<07:53,  5.99s/it]
Training Epoch: 8/100, batch 30/188 completed (loss: 0.0462:  16%|█▌        | 15/94 [01:31<07:53,  5.99s/it]
Training Epoch: 8/100, batch 30/188 completed (loss: 0.1331:  16%|█▌        | 15/94 [01:31<07:53,  5.99s/it]
Training Epoch: 8/100, batch 30/188 completed (loss: 0.1331:  17%|█▋        | 16/94 [01:34<07:42,  5.92s/it]
Training Epoch: 8/100, batch 30/188 completed (loss: 0.0462:  17%|█▋        | 16/94 [01:34<07:41,  5.92s/it]
Training Epoch: 8/100, batch 30/188 completed (loss: 0.0693:  17%|█▋        | 16/94 [01:34<07:42,  5.93s/it]
Training Epoch: 8/100, batch 30/188 completed (loss: 0.0629:  17%|█▋        | 16/94 [01:34<07:42,  5.93s/it]
Training Epoch: 8/100, batch 31/188 completed (loss: 0.0463:  17%|█▋        | 16/94 [01:34<07:42,  5.93s/it]
Training Epoch: 8/100, batch 31/188 completed (loss: 0.2273:  17%|█▋        | 16/94 [01:34<07:42,  5.93s/it]
Training Epoch: 8/100, batch 31/188 completed (loss: 0.1676:  17%|█▋        | 16/94 [01:34<07:41,  5.92s/it]
Training Epoch: 8/100, batch 31/188 completed (loss: 0.2412:  17%|█▋        | 16/94 [01:34<07:42,  5.92s/it]
Training Epoch: 8/100, batch 32/188 completed (loss: 0.0935:  17%|█▋        | 16/94 [01:37<07:42,  5.93s/it]
Training Epoch: 8/100, batch 32/188 completed (loss: 0.1159:  17%|█▋        | 16/94 [01:37<07:41,  5.92s/it]
Training Epoch: 8/100, batch 32/188 completed (loss: 0.058:  17%|█▋        | 16/94 [01:37<07:42,  5.93s/it] 
Training Epoch: 8/100, batch 32/188 completed (loss: 0.0872:  17%|█▋        | 16/94 [01:37<07:42,  5.92s/it]
Training Epoch: 8/100, batch 32/188 completed (loss: 0.0872:  18%|█▊        | 17/94 [01:40<07:33,  5.88s/it]
Training Epoch: 8/100, batch 32/188 completed (loss: 0.1159:  18%|█▊        | 17/94 [01:40<07:33,  5.88s/it]
Training Epoch: 8/100, batch 32/188 completed (loss: 0.0935:  18%|█▊        | 17/94 [01:40<07:33,  5.88s/it]
Training Epoch: 8/100, batch 32/188 completed (loss: 0.058:  18%|█▊        | 17/94 [01:40<07:33,  5.89s/it]
Training Epoch: 8/100, batch 33/188 completed (loss: 0.1067:  18%|█▊        | 17/94 [01:40<07:33,  5.88s/it]
Training Epoch: 8/100, batch 33/188 completed (loss: 0.0481:  18%|█▊        | 17/94 [01:40<07:33,  5.88s/it]
Training Epoch: 8/100, batch 33/188 completed (loss: 0.0918:  18%|█▊        | 17/94 [01:40<07:33,  5.89s/it]
Training Epoch: 8/100, batch 33/188 completed (loss: 0.1028:  18%|█▊        | 17/94 [01:40<07:33,  5.88s/it]
Training Epoch: 8/100, batch 34/188 completed (loss: 0.0641:  18%|█▊        | 17/94 [01:43<07:33,  5.89s/it]
Training Epoch: 8/100, batch 34/188 completed (loss: 0.1194:  18%|█▊        | 17/94 [01:43<07:33,  5.88s/it]
Training Epoch: 8/100, batch 34/188 completed (loss: 0.1421:  18%|█▊        | 17/94 [01:43<07:33,  5.88s/it]
Training Epoch: 8/100, batch 34/188 completed (loss: 0.0215:  18%|█▊        | 17/94 [01:43<07:33,  5.88s/it]
Training Epoch: 8/100, batch 34/188 completed (loss: 0.1194:  19%|█▉        | 18/94 [01:46<07:36,  6.01s/it]
Training Epoch: 8/100, batch 34/188 completed (loss: 0.0215:  19%|█▉        | 18/94 [01:46<07:36,  6.01s/it]
Training Epoch: 8/100, batch 34/188 completed (loss: 0.0641:  19%|█▉        | 18/94 [01:46<07:36,  6.01s/it]
Training Epoch: 8/100, batch 34/188 completed (loss: 0.1421:  19%|█▉        | 18/94 [01:46<07:36,  6.01s/it]
Training Epoch: 8/100, batch 35/188 completed (loss: 0.2521:  19%|█▉        | 18/94 [01:46<07:36,  6.01s/it]
Training Epoch: 8/100, batch 35/188 completed (loss: 0.1352:  19%|█▉        | 18/94 [01:46<07:36,  6.01s/it]
Training Epoch: 8/100, batch 35/188 completed (loss: 0.0177:  19%|█▉        | 18/94 [01:46<07:36,  6.01s/it]
Training Epoch: 8/100, batch 35/188 completed (loss: 0.1959:  19%|█▉        | 18/94 [01:46<07:36,  6.01s/it]
Training Epoch: 8/100, batch 36/188 completed (loss: 0.049:  19%|█▉        | 18/94 [01:49<07:36,  6.01s/it] 
Training Epoch: 8/100, batch 36/188 completed (loss: 0.1404:  19%|█▉        | 18/94 [01:49<07:36,  6.01s/it]
Training Epoch: 8/100, batch 36/188 completed (loss: 0.0724:  19%|█▉        | 18/94 [01:49<07:36,  6.01s/it]
Training Epoch: 8/100, batch 36/188 completed (loss: 0.1174:  19%|█▉        | 18/94 [01:49<07:36,  6.01s/it]
Training Epoch: 8/100, batch 36/188 completed (loss: 0.0724:  20%|██        | 19/94 [01:52<07:25,  5.94s/it]
Training Epoch: 8/100, batch 36/188 completed (loss: 0.1174:  20%|██        | 19/94 [01:52<07:25,  5.94s/it]
Training Epoch: 8/100, batch 36/188 completed (loss: 0.1404:  20%|██        | 19/94 [01:52<07:25,  5.94s/it]
Training Epoch: 8/100, batch 36/188 completed (loss: 0.049:  20%|██        | 19/94 [01:52<07:25,  5.94s/it]
Training Epoch: 8/100, batch 37/188 completed (loss: 0.2335:  20%|██        | 19/94 [01:52<07:25,  5.94s/it]
Training Epoch: 8/100, batch 37/188 completed (loss: 0.1159:  20%|██        | 19/94 [01:52<07:25,  5.94s/it]
Training Epoch: 8/100, batch 37/188 completed (loss: 0.1523:  20%|██        | 19/94 [01:52<07:25,  5.94s/it]
Training Epoch: 8/100, batch 37/188 completed (loss: 0.0497:  20%|██        | 19/94 [01:52<07:25,  5.94s/it]
Training Epoch: 8/100, batch 38/188 completed (loss: 0.1155:  20%|██        | 19/94 [01:55<07:25,  5.94s/it]
Training Epoch: 8/100, batch 38/188 completed (loss: 0.1009:  20%|██        | 19/94 [01:55<07:25,  5.94s/it]
Training Epoch: 8/100, batch 38/188 completed (loss: 0.0654:  20%|██        | 19/94 [01:55<07:25,  5.94s/it]
Training Epoch: 8/100, batch 38/188 completed (loss: 0.1096:  20%|██        | 19/94 [01:55<07:25,  5.94s/it]
Training Epoch: 8/100, batch 38/188 completed (loss: 0.1096:  21%|██▏       | 20/94 [01:58<07:23,  5.99s/it]
Training Epoch: 8/100, batch 38/188 completed (loss: 0.0654:  21%|██▏       | 20/94 [01:58<07:23,  5.99s/it]
Training Epoch: 8/100, batch 38/188 completed (loss: 0.1009:  21%|██▏       | 20/94 [01:58<07:23,  6.00s/it]
Training Epoch: 8/100, batch 38/188 completed (loss: 0.1155:  21%|██▏       | 20/94 [01:58<07:23,  6.00s/it]
Training Epoch: 8/100, batch 39/188 completed (loss: 0.0481:  21%|██▏       | 20/94 [01:58<07:23,  6.00s/it]
Training Epoch: 8/100, batch 39/188 completed (loss: 0.1765:  21%|██▏       | 20/94 [01:58<07:23,  5.99s/it]
Training Epoch: 8/100, batch 39/188 completed (loss: 0.152:  21%|██▏       | 20/94 [01:58<07:23,  6.00s/it] 
Training Epoch: 8/100, batch 39/188 completed (loss: 0.0376:  21%|██▏       | 20/94 [01:58<07:23,  5.99s/it]
Training Epoch: 8/100, batch 40/188 completed (loss: 0.1178:  21%|██▏       | 20/94 [02:01<07:23,  5.99s/it]
Training Epoch: 8/100, batch 40/188 completed (loss: 0.1756:  21%|██▏       | 20/94 [02:01<07:23,  6.00s/it]
Training Epoch: 8/100, batch 40/188 completed (loss: 0.0668:  21%|██▏       | 20/94 [02:01<07:23,  6.00s/it]
Training Epoch: 8/100, batch 40/188 completed (loss: 0.0406:  21%|██▏       | 20/94 [02:01<07:23,  5.99s/it]
Training Epoch: 8/100, batch 40/188 completed (loss: 0.1178:  22%|██▏       | 21/94 [02:04<07:19,  6.03s/it]
Training Epoch: 8/100, batch 40/188 completed (loss: 0.0406:  22%|██▏       | 21/94 [02:04<07:19,  6.03s/it]
Training Epoch: 8/100, batch 40/188 completed (loss: 0.0668:  22%|██▏       | 21/94 [02:04<07:19,  6.03s/it]
Training Epoch: 8/100, batch 40/188 completed (loss: 0.1756:  22%|██▏       | 21/94 [02:04<07:20,  6.03s/it]
Training Epoch: 8/100, batch 41/188 completed (loss: 0.1083:  22%|██▏       | 21/94 [02:04<07:20,  6.03s/it]
Training Epoch: 8/100, batch 41/188 completed (loss: 0.0934:  22%|██▏       | 21/94 [02:04<07:19,  6.03s/it]
Training Epoch: 8/100, batch 41/188 completed (loss: 0.098:  22%|██▏       | 21/94 [02:04<07:19,  6.03s/it] 
Training Epoch: 8/100, batch 41/188 completed (loss: 0.1238:  22%|██▏       | 21/94 [02:04<07:19,  6.03s/it]
Training Epoch: 8/100, batch 42/188 completed (loss: 0.1339:  22%|██▏       | 21/94 [02:07<07:19,  6.03s/it]
Training Epoch: 8/100, batch 42/188 completed (loss: 0.134:  22%|██▏       | 21/94 [02:07<07:19,  6.03s/it] 
Training Epoch: 8/100, batch 42/188 completed (loss: 0.0415:  22%|██▏       | 21/94 [02:07<07:20,  6.03s/it]
Training Epoch: 8/100, batch 42/188 completed (loss: 0.114:  22%|██▏       | 21/94 [02:07<07:19,  6.03s/it] 
Training Epoch: 8/100, batch 42/188 completed (loss: 0.0415:  23%|██▎       | 22/94 [02:10<07:08,  5.95s/it]
Training Epoch: 8/100, batch 42/188 completed (loss: 0.114:  23%|██▎       | 22/94 [02:10<07:09,  5.96s/it]
Training Epoch: 8/100, batch 42/188 completed (loss: 0.134:  23%|██▎       | 22/94 [02:10<07:08,  5.96s/it]
Training Epoch: 8/100, batch 42/188 completed (loss: 0.1339:  23%|██▎       | 22/94 [02:10<07:09,  5.96s/it]
Training Epoch: 8/100, batch 43/188 completed (loss: 0.0297:  23%|██▎       | 22/94 [02:10<07:08,  5.96s/it]
Training Epoch: 8/100, batch 43/188 completed (loss: 0.0734:  23%|██▎       | 22/94 [02:10<07:09,  5.96s/it]
Training Epoch: 8/100, batch 43/188 completed (loss: 0.0768:  23%|██▎       | 22/94 [02:10<07:08,  5.95s/it]
Training Epoch: 8/100, batch 43/188 completed (loss: 0.0526:  23%|██▎       | 22/94 [02:10<07:09,  5.96s/it]
Training Epoch: 8/100, batch 44/188 completed (loss: 0.0613:  23%|██▎       | 22/94 [02:13<07:08,  5.96s/it]
Training Epoch: 8/100, batch 44/188 completed (loss: 0.0453:  23%|██▎       | 22/94 [02:13<07:08,  5.95s/it]
Training Epoch: 8/100, batch 44/188 completed (loss: 0.1321:  23%|██▎       | 22/94 [02:13<07:09,  5.96s/it]
Training Epoch: 8/100, batch 44/188 completed (loss: 0.095:  23%|██▎       | 22/94 [02:13<07:09,  5.96s/it] 
Training Epoch: 8/100, batch 44/188 completed (loss: 0.095:  24%|██▍       | 23/94 [02:16<07:11,  6.08s/it]
Training Epoch: 8/100, batch 44/188 completed (loss: 0.0453:  24%|██▍       | 23/94 [02:16<07:11,  6.08s/it]
Training Epoch: 8/100, batch 44/188 completed (loss: 0.1321:  24%|██▍       | 23/94 [02:16<07:11,  6.08s/it]
Training Epoch: 8/100, batch 44/188 completed (loss: 0.0613:  24%|██▍       | 23/94 [02:16<07:11,  6.08s/it]
Training Epoch: 8/100, batch 45/188 completed (loss: 0.0523:  24%|██▍       | 23/94 [02:16<07:11,  6.08s/it]
Training Epoch: 8/100, batch 45/188 completed (loss: 0.033:  24%|██▍       | 23/94 [02:16<07:11,  6.08s/it] 
Training Epoch: 8/100, batch 45/188 completed (loss: 0.0533:  24%|██▍       | 23/94 [02:16<07:11,  6.08s/it]
Training Epoch: 8/100, batch 45/188 completed (loss: 0.0183:  24%|██▍       | 23/94 [02:16<07:11,  6.08s/it]
Training Epoch: 8/100, batch 46/188 completed (loss: 0.3239:  24%|██▍       | 23/94 [02:19<07:11,  6.08s/it]
Training Epoch: 8/100, batch 46/188 completed (loss: 0.0394:  24%|██▍       | 23/94 [02:19<07:11,  6.08s/it]
Training Epoch: 8/100, batch 46/188 completed (loss: 0.1706:  24%|██▍       | 23/94 [02:19<07:11,  6.08s/it]
Training Epoch: 8/100, batch 46/188 completed (loss: 0.0572:  24%|██▍       | 23/94 [02:19<07:11,  6.08s/it]
Training Epoch: 8/100, batch 46/188 completed (loss: 0.0394:  26%|██▌       | 24/94 [02:22<06:59,  5.99s/it]
Training Epoch: 8/100, batch 46/188 completed (loss: 0.0572:  26%|██▌       | 24/94 [02:22<06:59,  6.00s/it]
Training Epoch: 8/100, batch 46/188 completed (loss: 0.3239:  26%|██▌       | 24/94 [02:22<06:59,  5.99s/it]
Training Epoch: 8/100, batch 46/188 completed (loss: 0.1706:  26%|██▌       | 24/94 [02:22<06:59,  6.00s/it]
Training Epoch: 8/100, batch 47/188 completed (loss: 0.046:  26%|██▌       | 24/94 [02:22<06:59,  5.99s/it] 
Training Epoch: 8/100, batch 47/188 completed (loss: 0.0283:  26%|██▌       | 24/94 [02:22<06:59,  6.00s/it]
Training Epoch: 8/100, batch 47/188 completed (loss: 0.183:  26%|██▌       | 24/94 [02:22<06:59,  5.99s/it] 
Training Epoch: 8/100, batch 47/188 completed (loss: 0.0869:  26%|██▌       | 24/94 [02:22<06:59,  6.00s/it]
Training Epoch: 8/100, batch 48/188 completed (loss: 0.0938:  26%|██▌       | 24/94 [02:25<06:59,  5.99s/it]
Training Epoch: 8/100, batch 48/188 completed (loss: 0.168:  26%|██▌       | 24/94 [02:25<06:59,  5.99s/it]
Training Epoch: 8/100, batch 48/188 completed (loss: 0.0925:  26%|██▌       | 24/94 [02:25<06:59,  6.00s/it]
Training Epoch: 8/100, batch 48/188 completed (loss: 0.082:  26%|██▌       | 24/94 [02:25<06:59,  6.00s/it] 
Training Epoch: 8/100, batch 48/188 completed (loss: 0.0925:  27%|██▋       | 25/94 [02:28<06:56,  6.04s/it]
Training Epoch: 8/100, batch 48/188 completed (loss: 0.168:  27%|██▋       | 25/94 [02:28<06:56,  6.04s/it]
Training Epoch: 8/100, batch 48/188 completed (loss: 0.0938:  27%|██▋       | 25/94 [02:28<06:57,  6.04s/it]
Training Epoch: 8/100, batch 48/188 completed (loss: 0.082:  27%|██▋       | 25/94 [02:28<06:57,  6.04s/it]
Training Epoch: 8/100, batch 49/188 completed (loss: 0.1086:  27%|██▋       | 25/94 [02:28<06:57,  6.04s/it]
Training Epoch: 8/100, batch 49/188 completed (loss: 0.0962:  27%|██▋       | 25/94 [02:28<06:56,  6.04s/it]
Training Epoch: 8/100, batch 49/188 completed (loss: 0.0442:  27%|██▋       | 25/94 [02:28<06:56,  6.04s/it]
Training Epoch: 8/100, batch 49/188 completed (loss: 0.1686:  27%|██▋       | 25/94 [02:28<06:57,  6.04s/it]
Training Epoch: 8/100, batch 50/188 completed (loss: 0.1064:  27%|██▋       | 25/94 [02:32<06:57,  6.04s/it]
Training Epoch: 8/100, batch 50/188 completed (loss: 0.1832:  27%|██▋       | 25/94 [02:32<06:56,  6.04s/it]
Training Epoch: 8/100, batch 50/188 completed (loss: 0.0457:  27%|██▋       | 25/94 [02:32<06:56,  6.04s/it]
Training Epoch: 8/100, batch 50/188 completed (loss: 0.0402:  27%|██▋       | 25/94 [02:32<06:57,  6.04s/it]
Training Epoch: 8/100, batch 50/188 completed (loss: 0.0402:  28%|██▊       | 26/94 [02:34<06:52,  6.07s/it]
Training Epoch: 8/100, batch 50/188 completed (loss: 0.1832:  28%|██▊       | 26/94 [02:34<06:52,  6.07s/it]
Training Epoch: 8/100, batch 50/188 completed (loss: 0.0457:  28%|██▊       | 26/94 [02:34<06:52,  6.07s/it]
Training Epoch: 8/100, batch 50/188 completed (loss: 0.1064:  28%|██▊       | 26/94 [02:34<06:53,  6.07s/it]
Training Epoch: 8/100, batch 51/188 completed (loss: 0.0998:  28%|██▊       | 26/94 [02:34<06:53,  6.07s/it]
Training Epoch: 8/100, batch 51/188 completed (loss: 0.1948:  28%|██▊       | 26/94 [02:34<06:52,  6.07s/it]
Training Epoch: 8/100, batch 51/188 completed (loss: 0.077:  28%|██▊       | 26/94 [02:34<06:52,  6.07s/it] 
Training Epoch: 8/100, batch 51/188 completed (loss: 0.1453:  28%|██▊       | 26/94 [02:34<06:52,  6.07s/it]
Training Epoch: 8/100, batch 52/188 completed (loss: 0.0633:  28%|██▊       | 26/94 [02:37<06:52,  6.07s/it]
Training Epoch: 8/100, batch 52/188 completed (loss: 0.0826:  28%|██▊       | 26/94 [02:37<06:52,  6.07s/it]
Training Epoch: 8/100, batch 52/188 completed (loss: 0.0977:  28%|██▊       | 26/94 [02:37<06:53,  6.07s/it]
Training Epoch: 8/100, batch 52/188 completed (loss: 0.1448:  28%|██▊       | 26/94 [02:37<06:52,  6.07s/it]
Training Epoch: 8/100, batch 52/188 completed (loss: 0.0977:  29%|██▊       | 27/94 [02:40<06:41,  5.99s/it]
Training Epoch: 8/100, batch 52/188 completed (loss: 0.0633:  29%|██▊       | 27/94 [02:40<06:41,  5.99s/it]
Training Epoch: 8/100, batch 52/188 completed (loss: 0.1448:  29%|██▊       | 27/94 [02:40<06:41,  5.99s/it]
Training Epoch: 8/100, batch 52/188 completed (loss: 0.0826:  29%|██▊       | 27/94 [02:40<06:41,  6.00s/it]
Training Epoch: 8/100, batch 53/188 completed (loss: 0.1031:  29%|██▊       | 27/94 [02:40<06:41,  6.00s/it]
Training Epoch: 8/100, batch 53/188 completed (loss: 0.1026:  29%|██▊       | 27/94 [02:40<06:41,  5.99s/it]
Training Epoch: 8/100, batch 53/188 completed (loss: 0.0944:  29%|██▊       | 27/94 [02:40<06:41,  5.99s/it]
Training Epoch: 8/100, batch 53/188 completed (loss: 0.0864:  29%|██▊       | 27/94 [02:40<06:41,  5.99s/it]
Training Epoch: 8/100, batch 54/188 completed (loss: 0.026:  29%|██▊       | 27/94 [02:43<06:41,  6.00s/it] 
Training Epoch: 8/100, batch 54/188 completed (loss: 0.2033:  29%|██▊       | 27/94 [02:43<06:41,  5.99s/it]
Training Epoch: 8/100, batch 54/188 completed (loss: 0.0935:  29%|██▊       | 27/94 [02:43<06:41,  5.99s/it]
Training Epoch: 8/100, batch 54/188 completed (loss: 0.2378:  29%|██▊       | 27/94 [02:43<06:41,  5.99s/it]
Training Epoch: 8/100, batch 54/188 completed (loss: 0.026:  30%|██▉       | 28/94 [02:47<06:43,  6.12s/it]
Training Epoch: 8/100, batch 54/188 completed (loss: 0.2033:  30%|██▉       | 28/94 [02:47<06:43,  6.12s/it]
Training Epoch: 8/100, batch 54/188 completed (loss: 0.2378:  30%|██▉       | 28/94 [02:47<06:43,  6.12s/it]
Training Epoch: 8/100, batch 54/188 completed (loss: 0.0935:  30%|██▉       | 28/94 [02:47<06:43,  6.12s/it]
Training Epoch: 8/100, batch 55/188 completed (loss: 0.064:  30%|██▉       | 28/94 [02:47<06:43,  6.12s/it] 
Training Epoch: 8/100, batch 55/188 completed (loss: 0.0822:  30%|██▉       | 28/94 [02:47<06:43,  6.12s/it]
Training Epoch: 8/100, batch 55/188 completed (loss: 0.1116:  30%|██▉       | 28/94 [02:47<06:43,  6.12s/it]
Training Epoch: 8/100, batch 55/188 completed (loss: 0.0748:  30%|██▉       | 28/94 [02:47<06:43,  6.12s/it]
Training Epoch: 8/100, batch 56/188 completed (loss: 0.1832:  30%|██▉       | 28/94 [02:50<06:43,  6.12s/it]
Training Epoch: 8/100, batch 56/188 completed (loss: 0.2068:  30%|██▉       | 28/94 [02:50<06:43,  6.12s/it]
Training Epoch: 8/100, batch 56/188 completed (loss: 0.0687:  30%|██▉       | 28/94 [02:50<06:43,  6.12s/it]
Training Epoch: 8/100, batch 56/188 completed (loss: 0.1123:  30%|██▉       | 28/94 [02:50<06:43,  6.12s/it]
Training Epoch: 8/100, batch 56/188 completed (loss: 0.1123:  31%|███       | 29/94 [02:52<06:32,  6.04s/it]
Training Epoch: 8/100, batch 56/188 completed (loss: 0.1832:  31%|███       | 29/94 [02:52<06:32,  6.04s/it]
Training Epoch: 8/100, batch 56/188 completed (loss: 0.2068:  31%|███       | 29/94 [02:52<06:32,  6.04s/it]
Training Epoch: 8/100, batch 56/188 completed (loss: 0.0687:  31%|███       | 29/94 [02:52<06:32,  6.04s/it]
Training Epoch: 8/100, batch 57/188 completed (loss: 0.1252:  31%|███       | 29/94 [02:53<06:32,  6.04s/it]
Training Epoch: 8/100, batch 57/188 completed (loss: 0.0638:  31%|███       | 29/94 [02:53<06:32,  6.04s/it]
Training Epoch: 8/100, batch 57/188 completed (loss: 0.0251:  31%|███       | 29/94 [02:53<06:32,  6.04s/it]
Training Epoch: 8/100, batch 57/188 completed (loss: 0.1304:  31%|███       | 29/94 [02:53<06:32,  6.04s/it]
Training Epoch: 8/100, batch 58/188 completed (loss: 0.0528:  31%|███       | 29/94 [02:55<06:32,  6.04s/it]
Training Epoch: 8/100, batch 58/188 completed (loss: 0.1201:  31%|███       | 29/94 [02:55<06:32,  6.04s/it]
Training Epoch: 8/100, batch 58/188 completed (loss: 0.1104:  31%|███       | 29/94 [02:55<06:32,  6.04s/it]
Training Epoch: 8/100, batch 58/188 completed (loss: 0.0648:  31%|███       | 29/94 [02:55<06:32,  6.04s/it]
Training Epoch: 8/100, batch 58/188 completed (loss: 0.1201:  32%|███▏      | 30/94 [02:58<06:22,  5.97s/it]
Training Epoch: 8/100, batch 58/188 completed (loss: 0.0528:  32%|███▏      | 30/94 [02:58<06:22,  5.97s/it]
Training Epoch: 8/100, batch 58/188 completed (loss: 0.0648:  32%|███▏      | 30/94 [02:58<06:22,  5.97s/it]
Training Epoch: 8/100, batch 58/188 completed (loss: 0.1104:  32%|███▏      | 30/94 [02:58<06:22,  5.97s/it]
Training Epoch: 8/100, batch 59/188 completed (loss: 0.1163:  32%|███▏      | 30/94 [02:58<06:22,  5.97s/it]
Training Epoch: 8/100, batch 59/188 completed (loss: 0.0815:  32%|███▏      | 30/94 [02:58<06:22,  5.97s/it]
Training Epoch: 8/100, batch 59/188 completed (loss: 0.1639:  32%|███▏      | 30/94 [02:58<06:22,  5.97s/it]
Training Epoch: 8/100, batch 59/188 completed (loss: 0.2853:  32%|███▏      | 30/94 [02:58<06:22,  5.97s/it]
Training Epoch: 8/100, batch 60/188 completed (loss: 0.1772:  32%|███▏      | 30/94 [03:02<06:22,  5.97s/it]
Training Epoch: 8/100, batch 60/188 completed (loss: 0.0277:  32%|███▏      | 30/94 [03:02<06:22,  5.97s/it]
Training Epoch: 8/100, batch 60/188 completed (loss: 0.0838:  32%|███▏      | 30/94 [03:02<06:22,  5.97s/it]
Training Epoch: 8/100, batch 60/188 completed (loss: 0.0513:  32%|███▏      | 30/94 [03:02<06:22,  5.97s/it]
Training Epoch: 8/100, batch 60/188 completed (loss: 0.0277:  33%|███▎      | 31/94 [03:05<06:22,  6.07s/it]
Training Epoch: 8/100, batch 60/188 completed (loss: 0.0838:  33%|███▎      | 31/94 [03:05<06:22,  6.07s/it]
Training Epoch: 8/100, batch 60/188 completed (loss: 0.0513:  33%|███▎      | 31/94 [03:05<06:22,  6.07s/it]
Training Epoch: 8/100, batch 60/188 completed (loss: 0.1772:  33%|███▎      | 31/94 [03:05<06:22,  6.07s/it]
Training Epoch: 8/100, batch 61/188 completed (loss: 0.0798:  33%|███▎      | 31/94 [03:05<06:22,  6.07s/it]
Training Epoch: 8/100, batch 61/188 completed (loss: 0.0217:  33%|███▎      | 31/94 [03:05<06:22,  6.07s/it]
Training Epoch: 8/100, batch 61/188 completed (loss: 0.0425:  33%|███▎      | 31/94 [03:05<06:22,  6.07s/it]
Training Epoch: 8/100, batch 61/188 completed (loss: 0.0261:  33%|███▎      | 31/94 [03:05<06:22,  6.07s/it]
Training Epoch: 8/100, batch 62/188 completed (loss: 0.017:  33%|███▎      | 31/94 [03:08<06:22,  6.07s/it] 
Training Epoch: 8/100, batch 62/188 completed (loss: 0.0535:  33%|███▎      | 31/94 [03:08<06:22,  6.07s/it]
Training Epoch: 8/100, batch 62/188 completed (loss: 0.1895:  33%|███▎      | 31/94 [03:08<06:22,  6.07s/it]
Training Epoch: 8/100, batch 62/188 completed (loss: 0.0568:  33%|███▎      | 31/94 [03:08<06:22,  6.07s/it]
Training Epoch: 8/100, batch 62/188 completed (loss: 0.1895:  34%|███▍      | 32/94 [03:10<06:11,  6.00s/it]
Training Epoch: 8/100, batch 62/188 completed (loss: 0.0535:  34%|███▍      | 32/94 [03:10<06:11,  6.00s/it]
Training Epoch: 8/100, batch 62/188 completed (loss: 0.017:  34%|███▍      | 32/94 [03:10<06:11,  6.00s/it]
Training Epoch: 8/100, batch 62/188 completed (loss: 0.0568:  34%|███▍      | 32/94 [03:10<06:11,  6.00s/it]
Training Epoch: 8/100, batch 63/188 completed (loss: 0.0563:  34%|███▍      | 32/94 [03:10<06:11,  6.00s/it]
Training Epoch: 8/100, batch 63/188 completed (loss: 0.0389:  34%|███▍      | 32/94 [03:10<06:11,  6.00s/it]
Training Epoch: 8/100, batch 63/188 completed (loss: 0.1187:  34%|███▍      | 32/94 [03:10<06:11,  6.00s/it]
Training Epoch: 8/100, batch 63/188 completed (loss: 0.0599:  34%|███▍      | 32/94 [03:10<06:11,  6.00s/it]
Training Epoch: 8/100, batch 64/188 completed (loss: 0.0256:  34%|███▍      | 32/94 [03:13<06:11,  6.00s/it]
Training Epoch: 8/100, batch 64/188 completed (loss: 0.223:  34%|███▍      | 32/94 [03:13<06:11,  6.00s/it] 
Training Epoch: 8/100, batch 64/188 completed (loss: 0.2686:  34%|███▍      | 32/94 [03:13<06:11,  6.00s/it]
Training Epoch: 8/100, batch 64/188 completed (loss: 0.2228:  34%|███▍      | 32/94 [03:13<06:11,  6.00s/it]
Training Epoch: 8/100, batch 64/188 completed (loss: 0.2686:  35%|███▌      | 33/94 [03:17<06:08,  6.03s/it]
Training Epoch: 8/100, batch 64/188 completed (loss: 0.2228:  35%|███▌      | 33/94 [03:17<06:08,  6.03s/it]
Training Epoch: 8/100, batch 64/188 completed (loss: 0.223:  35%|███▌      | 33/94 [03:17<06:08,  6.03s/it]
Training Epoch: 8/100, batch 64/188 completed (loss: 0.0256:  35%|███▌      | 33/94 [03:17<06:08,  6.03s/it]
Training Epoch: 8/100, batch 65/188 completed (loss: 0.0216:  35%|███▌      | 33/94 [03:17<06:08,  6.03s/it]
Training Epoch: 8/100, batch 65/188 completed (loss: 0.069:  35%|███▌      | 33/94 [03:17<06:08,  6.03s/it] 
Training Epoch: 8/100, batch 65/188 completed (loss: 0.0397:  35%|███▌      | 33/94 [03:17<06:08,  6.03s/it]
Training Epoch: 8/100, batch 65/188 completed (loss: 0.2885:  35%|███▌      | 33/94 [03:17<06:08,  6.03s/it]
Training Epoch: 8/100, batch 66/188 completed (loss: 0.2082:  35%|███▌      | 33/94 [03:20<06:08,  6.03s/it]
Training Epoch: 8/100, batch 66/188 completed (loss: 0.0433:  35%|███▌      | 33/94 [03:20<06:08,  6.03s/it]
Training Epoch: 8/100, batch 66/188 completed (loss: 0.3778:  35%|███▌      | 33/94 [03:20<06:08,  6.03s/it]
Training Epoch: 8/100, batch 66/188 completed (loss: 0.0508:  35%|███▌      | 33/94 [03:20<06:08,  6.03s/it]
Training Epoch: 8/100, batch 66/188 completed (loss: 0.2082:  36%|███▌      | 34/94 [03:23<06:03,  6.06s/it]
Training Epoch: 8/100, batch 66/188 completed (loss: 0.0508:  36%|███▌      | 34/94 [03:23<06:03,  6.06s/it]
Training Epoch: 8/100, batch 66/188 completed (loss: 0.3778:  36%|███▌      | 34/94 [03:23<06:03,  6.06s/it]
Training Epoch: 8/100, batch 66/188 completed (loss: 0.0433:  36%|███▌      | 34/94 [03:23<06:03,  6.06s/it]
Training Epoch: 8/100, batch 67/188 completed (loss: 0.1938:  36%|███▌      | 34/94 [03:23<06:03,  6.06s/it]
Training Epoch: 8/100, batch 67/188 completed (loss: 0.0932:  36%|███▌      | 34/94 [03:23<06:03,  6.06s/it]
Training Epoch: 8/100, batch 67/188 completed (loss: 0.1226:  36%|███▌      | 34/94 [03:23<06:03,  6.06s/it]
Training Epoch: 8/100, batch 67/188 completed (loss: 0.059:  36%|███▌      | 34/94 [03:23<06:03,  6.06s/it] 
Training Epoch: 8/100, batch 68/188 completed (loss: 0.3116:  36%|███▌      | 34/94 [03:26<06:03,  6.06s/it]
Training Epoch: 8/100, batch 68/188 completed (loss: 0.057:  36%|███▌      | 34/94 [03:26<06:03,  6.06s/it] 
Training Epoch: 8/100, batch 68/188 completed (loss: 0.0679:  36%|███▌      | 34/94 [03:26<06:03,  6.06s/it]
Training Epoch: 8/100, batch 68/188 completed (loss: 0.0639:  36%|███▌      | 34/94 [03:26<06:03,  6.06s/it]
Training Epoch: 8/100, batch 68/188 completed (loss: 0.3116:  37%|███▋      | 35/94 [03:28<05:53,  5.99s/it]
Training Epoch: 8/100, batch 68/188 completed (loss: 0.057:  37%|███▋      | 35/94 [03:28<05:53,  5.99s/it]
Training Epoch: 8/100, batch 68/188 completed (loss: 0.0679:  37%|███▋      | 35/94 [03:28<05:53,  5.99s/it]
Training Epoch: 8/100, batch 68/188 completed (loss: 0.0639:  37%|███▋      | 35/94 [03:28<05:53,  5.99s/it]
Training Epoch: 8/100, batch 69/188 completed (loss: 0.248:  37%|███▋      | 35/94 [03:29<05:53,  5.99s/it]
Training Epoch: 8/100, batch 69/188 completed (loss: 0.1306:  37%|███▋      | 35/94 [03:29<05:53,  5.99s/it]
Training Epoch: 8/100, batch 69/188 completed (loss: 0.155:  37%|███▋      | 35/94 [03:29<05:53,  5.99s/it] 
Training Epoch: 8/100, batch 69/188 completed (loss: 0.0893:  37%|███▋      | 35/94 [03:29<05:53,  5.99s/it]
Training Epoch: 8/100, batch 70/188 completed (loss: 0.1041:  37%|███▋      | 35/94 [03:32<05:53,  5.99s/it]
Training Epoch: 8/100, batch 70/188 completed (loss: 0.0377:  37%|███▋      | 35/94 [03:32<05:53,  5.99s/it]
Training Epoch: 8/100, batch 70/188 completed (loss: 0.1398:  37%|███▋      | 35/94 [03:32<05:53,  5.99s/it]
Training Epoch: 8/100, batch 70/188 completed (loss: 0.1645:  37%|███▋      | 35/94 [03:32<05:53,  5.99s/it]
Training Epoch: 8/100, batch 70/188 completed (loss: 0.0377:  38%|███▊      | 36/94 [03:35<05:55,  6.12s/it]
Training Epoch: 8/100, batch 70/188 completed (loss: 0.1645:  38%|███▊      | 36/94 [03:35<05:55,  6.12s/it]
Training Epoch: 8/100, batch 70/188 completed (loss: 0.1041:  38%|███▊      | 36/94 [03:35<05:55,  6.12s/it]
Training Epoch: 8/100, batch 70/188 completed (loss: 0.1398:  38%|███▊      | 36/94 [03:35<05:55,  6.12s/it]
Training Epoch: 8/100, batch 71/188 completed (loss: 0.1378:  38%|███▊      | 36/94 [03:35<05:55,  6.12s/it]
Training Epoch: 8/100, batch 71/188 completed (loss: 0.068:  38%|███▊      | 36/94 [03:35<05:55,  6.12s/it] 
Training Epoch: 8/100, batch 71/188 completed (loss: 0.1169:  38%|███▊      | 36/94 [03:35<05:55,  6.12s/it]
Training Epoch: 8/100, batch 71/188 completed (loss: 0.0237:  38%|███▊      | 36/94 [03:35<05:55,  6.12s/it]
Training Epoch: 8/100, batch 72/188 completed (loss: 0.0578:  38%|███▊      | 36/94 [03:38<05:55,  6.12s/it]
Training Epoch: 8/100, batch 72/188 completed (loss: 0.0555:  38%|███▊      | 36/94 [03:38<05:55,  6.12s/it]
Training Epoch: 8/100, batch 72/188 completed (loss: 0.0619:  38%|███▊      | 36/94 [03:38<05:55,  6.12s/it]
Training Epoch: 8/100, batch 72/188 completed (loss: 0.0606:  38%|███▊      | 36/94 [03:38<05:55,  6.12s/it]
Training Epoch: 8/100, batch 72/188 completed (loss: 0.0578:  39%|███▉      | 37/94 [03:41<05:43,  6.03s/it]
Training Epoch: 8/100, batch 72/188 completed (loss: 0.0606:  39%|███▉      | 37/94 [03:41<05:43,  6.03s/it]
Training Epoch: 8/100, batch 72/188 completed (loss: 0.0619:  39%|███▉      | 37/94 [03:41<05:43,  6.03s/it]
Training Epoch: 8/100, batch 72/188 completed (loss: 0.0555:  39%|███▉      | 37/94 [03:41<05:43,  6.03s/it]
Training Epoch: 8/100, batch 73/188 completed (loss: 0.0276:  39%|███▉      | 37/94 [03:41<05:43,  6.03s/it]
Training Epoch: 8/100, batch 73/188 completed (loss: 0.0569:  39%|███▉      | 37/94 [03:41<05:43,  6.03s/it]
Training Epoch: 8/100, batch 73/188 completed (loss: 0.0893:  39%|███▉      | 37/94 [03:41<05:43,  6.03s/it]
Training Epoch: 8/100, batch 73/188 completed (loss: 0.1056:  39%|███▉      | 37/94 [03:41<05:43,  6.03s/it]
Training Epoch: 8/100, batch 74/188 completed (loss: 0.0748:  39%|███▉      | 37/94 [03:44<05:43,  6.03s/it]
Training Epoch: 8/100, batch 74/188 completed (loss: 0.0523:  39%|███▉      | 37/94 [03:44<05:43,  6.03s/it]
Training Epoch: 8/100, batch 74/188 completed (loss: 0.0811:  39%|███▉      | 37/94 [03:44<05:43,  6.03s/it]
Training Epoch: 8/100, batch 74/188 completed (loss: 0.0088:  39%|███▉      | 37/94 [03:44<05:43,  6.03s/it]
Training Epoch: 8/100, batch 74/188 completed (loss: 0.0088:  40%|████      | 38/94 [03:47<05:37,  6.02s/it]
Training Epoch: 8/100, batch 74/188 completed (loss: 0.0523:  40%|████      | 38/94 [03:47<05:37,  6.02s/it]
Training Epoch: 8/100, batch 74/188 completed (loss: 0.0811:  40%|████      | 38/94 [03:47<05:37,  6.02s/it]
Training Epoch: 8/100, batch 74/188 completed (loss: 0.0748:  40%|████      | 38/94 [03:47<05:37,  6.03s/it]
Training Epoch: 8/100, batch 75/188 completed (loss: 0.1077:  40%|████      | 38/94 [03:47<05:37,  6.02s/it]
Training Epoch: 8/100, batch 75/188 completed (loss: 0.0788:  40%|████      | 38/94 [03:47<05:37,  6.03s/it]
Training Epoch: 8/100, batch 75/188 completed (loss: 0.1014:  40%|████      | 38/94 [03:47<05:37,  6.02s/it]
Training Epoch: 8/100, batch 75/188 completed (loss: 0.0581:  40%|████      | 38/94 [03:47<05:37,  6.02s/it]
Training Epoch: 8/100, batch 76/188 completed (loss: 0.0321:  40%|████      | 38/94 [03:50<05:37,  6.02s/it]
Training Epoch: 8/100, batch 76/188 completed (loss: 0.0485:  40%|████      | 38/94 [03:50<05:37,  6.03s/it]
Training Epoch: 8/100, batch 76/188 completed (loss: 0.0225:  40%|████      | 38/94 [03:50<05:37,  6.02s/it]
Training Epoch: 8/100, batch 76/188 completed (loss: 0.1242:  40%|████      | 38/94 [03:50<05:37,  6.02s/it]
Training Epoch: 8/100, batch 76/188 completed (loss: 0.1242:  41%|████▏     | 39/94 [03:53<05:35,  6.10s/it]
Training Epoch: 8/100, batch 76/188 completed (loss: 0.0321:  41%|████▏     | 39/94 [03:53<05:35,  6.10s/it]
Training Epoch: 8/100, batch 76/188 completed (loss: 0.0225:  41%|████▏     | 39/94 [03:53<05:35,  6.09s/it]
Training Epoch: 8/100, batch 76/188 completed (loss: 0.0485:  41%|████▏     | 39/94 [03:53<05:35,  6.10s/it]
Training Epoch: 8/100, batch 77/188 completed (loss: 0.2104:  41%|████▏     | 39/94 [03:53<05:35,  6.09s/it]
Training Epoch: 8/100, batch 77/188 completed (loss: 0.0445:  41%|████▏     | 39/94 [03:53<05:35,  6.10s/it]
Training Epoch: 8/100, batch 77/188 completed (loss: 0.1252:  41%|████▏     | 39/94 [03:53<05:35,  6.10s/it]
Training Epoch: 8/100, batch 77/188 completed (loss: 0.0692:  41%|████▏     | 39/94 [03:53<05:35,  6.10s/it]
Training Epoch: 8/100, batch 78/188 completed (loss: 0.1232:  41%|████▏     | 39/94 [03:56<05:35,  6.10s/it]
Training Epoch: 8/100, batch 78/188 completed (loss: 0.1201:  41%|████▏     | 39/94 [03:56<05:35,  6.09s/it]
Training Epoch: 8/100, batch 78/188 completed (loss: 0.0879:  41%|████▏     | 39/94 [03:56<05:35,  6.10s/it]
Training Epoch: 8/100, batch 78/188 completed (loss: 0.0446:  41%|████▏     | 39/94 [03:56<05:35,  6.10s/it]
Training Epoch: 8/100, batch 78/188 completed (loss: 0.1232:  43%|████▎     | 40/94 [03:59<05:24,  6.01s/it]
Training Epoch: 8/100, batch 78/188 completed (loss: 0.1201:  43%|████▎     | 40/94 [03:59<05:24,  6.01s/it]
Training Epoch: 8/100, batch 78/188 completed (loss: 0.0879:  43%|████▎     | 40/94 [03:59<05:24,  6.01s/it]
Training Epoch: 8/100, batch 78/188 completed (loss: 0.0446:  43%|████▎     | 40/94 [03:59<05:24,  6.01s/it]
Training Epoch: 8/100, batch 79/188 completed (loss: 0.0864:  43%|████▎     | 40/94 [03:59<05:24,  6.01s/it]
Training Epoch: 8/100, batch 79/188 completed (loss: 0.0574:  43%|████▎     | 40/94 [03:59<05:24,  6.01s/it]
Training Epoch: 8/100, batch 79/188 completed (loss: 0.1315:  43%|████▎     | 40/94 [03:59<05:24,  6.01s/it]
Training Epoch: 8/100, batch 79/188 completed (loss: 0.1411:  43%|████▎     | 40/94 [03:59<05:24,  6.01s/it]
Training Epoch: 8/100, batch 80/188 completed (loss: 0.0816:  43%|████▎     | 40/94 [04:02<05:24,  6.01s/it]
Training Epoch: 8/100, batch 80/188 completed (loss: 0.0688:  43%|████▎     | 40/94 [04:02<05:24,  6.01s/it]
Training Epoch: 8/100, batch 80/188 completed (loss: 0.0435:  43%|████▎     | 40/94 [04:02<05:24,  6.01s/it]
Training Epoch: 8/100, batch 80/188 completed (loss: 0.0845:  43%|████▎     | 40/94 [04:02<05:24,  6.01s/it]
Training Epoch: 8/100, batch 80/188 completed (loss: 0.0688:  44%|████▎     | 41/94 [04:05<05:23,  6.10s/it]
Training Epoch: 8/100, batch 80/188 completed (loss: 0.0435:  44%|████▎     | 41/94 [04:05<05:23,  6.10s/it]
Training Epoch: 8/100, batch 80/188 completed (loss: 0.0816:  44%|████▎     | 41/94 [04:05<05:23,  6.10s/it]
Training Epoch: 8/100, batch 80/188 completed (loss: 0.0845:  44%|████▎     | 41/94 [04:05<05:23,  6.10s/it]
Training Epoch: 8/100, batch 81/188 completed (loss: 0.1046:  44%|████▎     | 41/94 [04:05<05:23,  6.10s/it]
Training Epoch: 8/100, batch 81/188 completed (loss: 0.061:  44%|████▎     | 41/94 [04:05<05:23,  6.10s/it] 
Training Epoch: 8/100, batch 81/188 completed (loss: 0.0597:  44%|████▎     | 41/94 [04:05<05:23,  6.10s/it]
Training Epoch: 8/100, batch 81/188 completed (loss: 0.0465:  44%|████▎     | 41/94 [04:05<05:23,  6.10s/it]
Training Epoch: 8/100, batch 82/188 completed (loss: 0.344:  44%|████▎     | 41/94 [04:08<05:23,  6.10s/it] 
Training Epoch: 8/100, batch 82/188 completed (loss: 0.1291:  44%|████▎     | 41/94 [04:08<05:23,  6.10s/it]
Training Epoch: 8/100, batch 82/188 completed (loss: 0.1279:  44%|████▎     | 41/94 [04:08<05:23,  6.10s/it]
Training Epoch: 8/100, batch 82/188 completed (loss: 0.057:  44%|████▎     | 41/94 [04:08<05:23,  6.10s/it] 
Training Epoch: 8/100, batch 82/188 completed (loss: 0.1291:  45%|████▍     | 42/94 [04:11<05:15,  6.07s/it]
Training Epoch: 8/100, batch 82/188 completed (loss: 0.344:  45%|████▍     | 42/94 [04:11<05:15,  6.07s/it]
Training Epoch: 8/100, batch 82/188 completed (loss: 0.057:  45%|████▍     | 42/94 [04:11<05:15,  6.07s/it]
Training Epoch: 8/100, batch 82/188 completed (loss: 0.1279:  45%|████▍     | 42/94 [04:11<05:15,  6.07s/it]
Training Epoch: 8/100, batch 83/188 completed (loss: 0.0715:  45%|████▍     | 42/94 [04:11<05:15,  6.07s/it]
Training Epoch: 8/100, batch 83/188 completed (loss: 0.1315:  45%|████▍     | 42/94 [04:11<05:15,  6.07s/it]
Training Epoch: 8/100, batch 83/188 completed (loss: 0.0404:  45%|████▍     | 42/94 [04:11<05:15,  6.07s/it]
Training Epoch: 8/100, batch 83/188 completed (loss: 0.0654:  45%|████▍     | 42/94 [04:11<05:15,  6.07s/it]
Training Epoch: 8/100, batch 84/188 completed (loss: 0.1064:  45%|████▍     | 42/94 [04:14<05:15,  6.07s/it]
Training Epoch: 8/100, batch 84/188 completed (loss: 0.0209:  45%|████▍     | 42/94 [04:14<05:15,  6.07s/it]
Training Epoch: 8/100, batch 84/188 completed (loss: 0.224:  45%|████▍     | 42/94 [04:14<05:15,  6.07s/it] 
Training Epoch: 8/100, batch 84/188 completed (loss: 0.031:  45%|████▍     | 42/94 [04:14<05:15,  6.07s/it] 
Training Epoch: 8/100, batch 84/188 completed (loss: 0.0209:  46%|████▌     | 43/94 [04:17<05:05,  5.99s/it]
Training Epoch: 8/100, batch 84/188 completed (loss: 0.031:  46%|████▌     | 43/94 [04:17<05:05,  6.00s/it]
Training Epoch: 8/100, batch 84/188 completed (loss: 0.224:  46%|████▌     | 43/94 [04:17<05:05,  6.00s/it]
Training Epoch: 8/100, batch 84/188 completed (loss: 0.1064:  46%|████▌     | 43/94 [04:17<05:05,  6.00s/it]
Training Epoch: 8/100, batch 85/188 completed (loss: 0.1666:  46%|████▌     | 43/94 [04:17<05:05,  6.00s/it]
Training Epoch: 8/100, batch 85/188 completed (loss: 0.0815:  46%|████▌     | 43/94 [04:17<05:05,  6.00s/it]
Training Epoch: 8/100, batch 85/188 completed (loss: 0.0731:  46%|████▌     | 43/94 [04:17<05:05,  5.99s/it]
Training Epoch: 8/100, batch 85/188 completed (loss: 0.1929:  46%|████▌     | 43/94 [04:17<05:05,  6.00s/it]
Training Epoch: 8/100, batch 86/188 completed (loss: 0.0909:  46%|████▌     | 43/94 [04:20<05:05,  6.00s/it]
Training Epoch: 8/100, batch 86/188 completed (loss: 0.0248:  46%|████▌     | 43/94 [04:20<05:05,  6.00s/it]
Training Epoch: 8/100, batch 86/188 completed (loss: 0.0872:  46%|████▌     | 43/94 [04:20<05:05,  5.99s/it]
Training Epoch: 8/100, batch 86/188 completed (loss: 0.1719:  46%|████▌     | 43/94 [04:20<05:05,  6.00s/it]
Training Epoch: 8/100, batch 86/188 completed (loss: 0.0872:  47%|████▋     | 44/94 [04:23<05:05,  6.10s/it]
Training Epoch: 8/100, batch 86/188 completed (loss: 0.0248:  47%|████▋     | 44/94 [04:23<05:05,  6.10s/it]
Training Epoch: 8/100, batch 86/188 completed (loss: 0.1719:  47%|████▋     | 44/94 [04:23<05:05,  6.10s/it]
Training Epoch: 8/100, batch 86/188 completed (loss: 0.0909:  47%|████▋     | 44/94 [04:23<05:05,  6.10s/it]
Training Epoch: 8/100, batch 87/188 completed (loss: 0.2451:  47%|████▋     | 44/94 [04:23<05:05,  6.10s/it]
Training Epoch: 8/100, batch 87/188 completed (loss: 0.0314:  47%|████▋     | 44/94 [04:23<05:05,  6.10s/it]
Training Epoch: 8/100, batch 87/188 completed (loss: 0.1418:  47%|████▋     | 44/94 [04:23<05:05,  6.10s/it]
Training Epoch: 8/100, batch 87/188 completed (loss: 0.0671:  47%|████▋     | 44/94 [04:23<05:05,  6.10s/it]
Training Epoch: 8/100, batch 88/188 completed (loss: 0.1051:  47%|████▋     | 44/94 [04:26<05:05,  6.10s/it]
Training Epoch: 8/100, batch 88/188 completed (loss: 0.0932:  47%|████▋     | 44/94 [04:26<05:05,  6.10s/it]
Training Epoch: 8/100, batch 88/188 completed (loss: 0.1844:  47%|████▋     | 44/94 [04:26<05:05,  6.10s/it]
Training Epoch: 8/100, batch 88/188 completed (loss: 0.2348:  47%|████▋     | 44/94 [04:26<05:05,  6.10s/it]
Training Epoch: 8/100, batch 88/188 completed (loss: 0.1844:  48%|████▊     | 45/94 [04:29<04:55,  6.02s/it]
Training Epoch: 8/100, batch 88/188 completed (loss: 0.1051:  48%|████▊     | 45/94 [04:29<04:54,  6.02s/it]
Training Epoch: 8/100, batch 88/188 completed (loss: 0.0932:  48%|████▊     | 45/94 [04:29<04:55,  6.02s/it]
Training Epoch: 8/100, batch 88/188 completed (loss: 0.2348:  48%|████▊     | 45/94 [04:29<04:55,  6.02s/it]
Training Epoch: 8/100, batch 89/188 completed (loss: 0.1278:  48%|████▊     | 45/94 [04:29<04:55,  6.02s/it]
Training Epoch: 8/100, batch 89/188 completed (loss: 0.0322:  48%|████▊     | 45/94 [04:29<04:54,  6.02s/it]
Training Epoch: 8/100, batch 89/188 completed (loss: 0.0245:  48%|████▊     | 45/94 [04:29<04:55,  6.02s/it]
Training Epoch: 8/100, batch 89/188 completed (loss: 0.0687:  48%|████▊     | 45/94 [04:29<04:55,  6.02s/it]
Training Epoch: 8/100, batch 90/188 completed (loss: 0.0808:  48%|████▊     | 45/94 [04:32<04:55,  6.02s/it]
Training Epoch: 8/100, batch 90/188 completed (loss: 0.3199:  48%|████▊     | 45/94 [04:32<04:54,  6.02s/it]
Training Epoch: 8/100, batch 90/188 completed (loss: 0.1722:  48%|████▊     | 45/94 [04:32<04:55,  6.02s/it]
Training Epoch: 8/100, batch 90/188 completed (loss: 0.0491:  48%|████▊     | 45/94 [04:32<04:55,  6.02s/it]
Training Epoch: 8/100, batch 90/188 completed (loss: 0.0491:  49%|████▉     | 46/94 [04:35<04:50,  6.05s/it]
Training Epoch: 8/100, batch 90/188 completed (loss: 0.1722:  49%|████▉     | 46/94 [04:35<04:50,  6.06s/it]
Training Epoch: 8/100, batch 90/188 completed (loss: 0.0808:  49%|████▉     | 46/94 [04:35<04:50,  6.06s/it]
Training Epoch: 8/100, batch 90/188 completed (loss: 0.3199:  49%|████▉     | 46/94 [04:35<04:50,  6.06s/it]
Training Epoch: 8/100, batch 91/188 completed (loss: 0.3809:  49%|████▉     | 46/94 [04:35<04:50,  6.06s/it]
Training Epoch: 8/100, batch 91/188 completed (loss: 0.0791:  49%|████▉     | 46/94 [04:35<04:50,  6.06s/it]
Training Epoch: 8/100, batch 91/188 completed (loss: 0.118:  49%|████▉     | 46/94 [04:35<04:50,  6.06s/it] 
Training Epoch: 8/100, batch 91/188 completed (loss: 0.231:  49%|████▉     | 46/94 [04:35<04:50,  6.05s/it] 
Training Epoch: 8/100, batch 92/188 completed (loss: 0.0987:  49%|████▉     | 46/94 [04:38<04:50,  6.06s/it]
Training Epoch: 8/100, batch 92/188 completed (loss: 0.1659:  49%|████▉     | 46/94 [04:38<04:50,  6.06s/it]
Training Epoch: 8/100, batch 92/188 completed (loss: 0.1507:  49%|████▉     | 46/94 [04:38<04:50,  6.06s/it]
Training Epoch: 8/100, batch 92/188 completed (loss: 0.2137:  49%|████▉     | 46/94 [04:38<04:50,  6.05s/it]
Training Epoch: 8/100, batch 92/188 completed (loss: 0.1659:  50%|█████     | 47/94 [04:41<04:46,  6.09s/it]
Training Epoch: 8/100, batch 92/188 completed (loss: 0.2137:  50%|█████     | 47/94 [04:41<04:46,  6.09s/it]
Training Epoch: 8/100, batch 92/188 completed (loss: 0.0987:  50%|█████     | 47/94 [04:41<04:46,  6.09s/it]
Training Epoch: 8/100, batch 92/188 completed (loss: 0.1507:  50%|█████     | 47/94 [04:41<04:46,  6.10s/it]
Training Epoch: 8/100, batch 93/188 completed (loss: 0.0232:  50%|█████     | 47/94 [04:41<04:46,  6.10s/it]
Training Epoch: 8/100, batch 93/188 completed (loss: 0.0507:  50%|█████     | 47/94 [04:41<04:46,  6.09s/it]
Training Epoch: 8/100, batch 93/188 completed (loss: 0.0401:  50%|█████     | 47/94 [04:41<04:46,  6.09s/it]
Training Epoch: 8/100, batch 93/188 completed (loss: 0.2011:  50%|█████     | 47/94 [04:42<04:46,  6.09s/it]
Training Epoch: 8/100, batch 94/188 completed (loss: 0.0644:  50%|█████     | 47/94 [04:44<04:46,  6.09s/it]
Training Epoch: 8/100, batch 94/188 completed (loss: 0.183:  50%|█████     | 47/94 [04:44<04:46,  6.09s/it] 
Training Epoch: 8/100, batch 94/188 completed (loss: 0.2674:  50%|█████     | 47/94 [04:44<04:46,  6.10s/it]
Training Epoch: 8/100, batch 94/188 completed (loss: 0.0322:  50%|█████     | 47/94 [04:44<04:46,  6.09s/it]
Training Epoch: 8/100, batch 94/188 completed (loss: 0.0644:  51%|█████     | 48/94 [04:47<04:36,  6.01s/it]
Training Epoch: 8/100, batch 94/188 completed (loss: 0.183:  51%|█████     | 48/94 [04:47<04:36,  6.01s/it]
Training Epoch: 8/100, batch 94/188 completed (loss: 0.2674:  51%|█████     | 48/94 [04:47<04:36,  6.01s/it]
Training Epoch: 8/100, batch 94/188 completed (loss: 0.0322:  51%|█████     | 48/94 [04:47<04:36,  6.01s/it]
Training Epoch: 8/100, batch 95/188 completed (loss: 0.1486:  51%|█████     | 48/94 [04:47<04:36,  6.01s/it]
Training Epoch: 8/100, batch 95/188 completed (loss: 0.1408:  51%|█████     | 48/94 [04:47<04:36,  6.01s/it]
Training Epoch: 8/100, batch 95/188 completed (loss: 0.0769:  51%|█████     | 48/94 [04:47<04:36,  6.01s/it]
Training Epoch: 8/100, batch 95/188 completed (loss: 0.1512:  51%|█████     | 48/94 [04:47<04:36,  6.01s/it]
Training Epoch: 8/100, batch 96/188 completed (loss: 0.0383:  51%|█████     | 48/94 [04:50<04:36,  6.01s/it]
Training Epoch: 8/100, batch 96/188 completed (loss: 0.2762:  51%|█████     | 48/94 [04:50<04:36,  6.01s/it]
Training Epoch: 8/100, batch 96/188 completed (loss: 0.0638:  51%|█████     | 48/94 [04:50<04:36,  6.01s/it]
Training Epoch: 8/100, batch 96/188 completed (loss: 0.1003:  51%|█████     | 48/94 [04:50<04:36,  6.01s/it]
Training Epoch: 8/100, batch 96/188 completed (loss: 0.0638:  52%|█████▏    | 49/94 [04:54<04:33,  6.09s/it]
Training Epoch: 8/100, batch 96/188 completed (loss: 0.1003:  52%|█████▏    | 49/94 [04:54<04:33,  6.09s/it]
Training Epoch: 8/100, batch 96/188 completed (loss: 0.0383:  52%|█████▏    | 49/94 [04:54<04:33,  6.09s/it]
Training Epoch: 8/100, batch 96/188 completed (loss: 0.2762:  52%|█████▏    | 49/94 [04:54<04:33,  6.09s/it]
Training Epoch: 8/100, batch 97/188 completed (loss: 0.0359:  52%|█████▏    | 49/94 [04:54<04:33,  6.09s/it]
Training Epoch: 8/100, batch 97/188 completed (loss: 0.1303:  52%|█████▏    | 49/94 [04:54<04:33,  6.09s/it]
Training Epoch: 8/100, batch 97/188 completed (loss: 0.0161:  52%|█████▏    | 49/94 [04:54<04:33,  6.09s/it]
Training Epoch: 8/100, batch 97/188 completed (loss: 0.0231:  52%|█████▏    | 49/94 [04:54<04:33,  6.09s/it]
Training Epoch: 8/100, batch 98/188 completed (loss: 0.0524:  52%|█████▏    | 49/94 [04:57<04:33,  6.09s/it]
Training Epoch: 8/100, batch 98/188 completed (loss: 0.1338:  52%|█████▏    | 49/94 [04:57<04:33,  6.09s/it]
Training Epoch: 8/100, batch 98/188 completed (loss: 0.2331:  52%|█████▏    | 49/94 [04:57<04:33,  6.09s/it]
Training Epoch: 8/100, batch 98/188 completed (loss: 0.1072:  52%|█████▏    | 49/94 [04:57<04:33,  6.09s/it]
Training Epoch: 8/100, batch 98/188 completed (loss: 0.0524:  53%|█████▎    | 50/94 [05:00<04:26,  6.06s/it]
Training Epoch: 8/100, batch 98/188 completed (loss: 0.2331:  53%|█████▎    | 50/94 [05:00<04:26,  6.06s/it]
Training Epoch: 8/100, batch 98/188 completed (loss: 0.1072:  53%|█████▎    | 50/94 [05:00<04:26,  6.06s/it]
Training Epoch: 8/100, batch 98/188 completed (loss: 0.1338:  53%|█████▎    | 50/94 [05:00<04:26,  6.06s/it]
Training Epoch: 8/100, batch 99/188 completed (loss: 0.1663:  53%|█████▎    | 50/94 [05:00<04:26,  6.06s/it]
Training Epoch: 8/100, batch 99/188 completed (loss: 0.1376:  53%|█████▎    | 50/94 [05:00<04:26,  6.06s/it]
Training Epoch: 8/100, batch 99/188 completed (loss: 0.2614:  53%|█████▎    | 50/94 [05:00<04:26,  6.06s/it]
Training Epoch: 8/100, batch 99/188 completed (loss: 0.0946:  53%|█████▎    | 50/94 [05:00<04:26,  6.06s/it]
Training Epoch: 8/100, batch 100/188 completed (loss: 0.0624:  53%|█████▎    | 50/94 [05:03<04:26,  6.06s/it]
Training Epoch: 8/100, batch 100/188 completed (loss: 0.1596:  53%|█████▎    | 50/94 [05:03<04:26,  6.06s/it]
Training Epoch: 8/100, batch 100/188 completed (loss: 0.064:  53%|█████▎    | 50/94 [05:02<04:26,  6.06s/it]
Training Epoch: 8/100, batch 100/188 completed (loss: 0.102:  53%|█████▎    | 50/94 [05:03<04:26,  6.06s/it]
Training Epoch: 8/100, batch 100/188 completed (loss: 0.0624:  54%|█████▍    | 51/94 [05:06<04:19,  6.04s/it]
Training Epoch: 8/100, batch 100/188 completed (loss: 0.064:  54%|█████▍    | 51/94 [05:06<04:19,  6.04s/it]
Training Epoch: 8/100, batch 100/188 completed (loss: 0.102:  54%|█████▍    | 51/94 [05:06<04:19,  6.04s/it]
Training Epoch: 8/100, batch 100/188 completed (loss: 0.1596:  54%|█████▍    | 51/94 [05:06<04:19,  6.04s/it]
Training Epoch: 8/100, batch 101/188 completed (loss: 0.0142:  54%|█████▍    | 51/94 [05:06<04:19,  6.04s/it]
Training Epoch: 8/100, batch 101/188 completed (loss: 0.1164:  54%|█████▍    | 51/94 [05:06<04:19,  6.04s/it]
Training Epoch: 8/100, batch 101/188 completed (loss: 0.0321:  54%|█████▍    | 51/94 [05:06<04:19,  6.04s/it]
Training Epoch: 8/100, batch 101/188 completed (loss: 0.0803:  54%|█████▍    | 51/94 [05:06<04:19,  6.04s/it]
Training Epoch: 8/100, batch 102/188 completed (loss: 0.0944:  54%|█████▍    | 51/94 [05:09<04:19,  6.04s/it]
Training Epoch: 8/100, batch 102/188 completed (loss: 0.1259:  54%|█████▍    | 51/94 [05:09<04:19,  6.04s/it]
Training Epoch: 8/100, batch 102/188 completed (loss: 0.1165:  54%|█████▍    | 51/94 [05:09<04:19,  6.04s/it]
Training Epoch: 8/100, batch 102/188 completed (loss: 0.0969:  54%|█████▍    | 51/94 [05:09<04:19,  6.04s/it]
Training Epoch: 8/100, batch 102/188 completed (loss: 0.0944:  55%|█████▌    | 52/94 [05:12<04:17,  6.12s/it]
Training Epoch: 8/100, batch 102/188 completed (loss: 0.0969:  55%|█████▌    | 52/94 [05:12<04:17,  6.12s/it]
Training Epoch: 8/100, batch 102/188 completed (loss: 0.1259:  55%|█████▌    | 52/94 [05:12<04:17,  6.12s/it]
Training Epoch: 8/100, batch 102/188 completed (loss: 0.1165:  55%|█████▌    | 52/94 [05:12<04:17,  6.12s/it]
Training Epoch: 8/100, batch 103/188 completed (loss: 0.1094:  55%|█████▌    | 52/94 [05:12<04:17,  6.12s/it]
Training Epoch: 8/100, batch 103/188 completed (loss: 0.2403:  55%|█████▌    | 52/94 [05:12<04:17,  6.12s/it]
Training Epoch: 8/100, batch 103/188 completed (loss: 0.0876:  55%|█████▌    | 52/94 [05:12<04:17,  6.12s/it]
Training Epoch: 8/100, batch 103/188 completed (loss: 0.3558:  55%|█████▌    | 52/94 [05:12<04:17,  6.12s/it]
Training Epoch: 8/100, batch 104/188 completed (loss: 0.0433:  55%|█████▌    | 52/94 [05:15<04:17,  6.12s/it]
Training Epoch: 8/100, batch 104/188 completed (loss: 0.1583:  55%|█████▌    | 52/94 [05:15<04:17,  6.12s/it]
Training Epoch: 8/100, batch 104/188 completed (loss: 0.1636:  55%|█████▌    | 52/94 [05:15<04:17,  6.12s/it]
Training Epoch: 8/100, batch 104/188 completed (loss: 0.1831:  55%|█████▌    | 52/94 [05:15<04:17,  6.12s/it]
Training Epoch: 8/100, batch 104/188 completed (loss: 0.0433:  56%|█████▋    | 53/94 [05:18<04:07,  6.04s/it]
Training Epoch: 8/100, batch 104/188 completed (loss: 0.1636:  56%|█████▋    | 53/94 [05:18<04:07,  6.04s/it]
Training Epoch: 8/100, batch 104/188 completed (loss: 0.1583:  56%|█████▋    | 53/94 [05:18<04:07,  6.04s/it]
Training Epoch: 8/100, batch 104/188 completed (loss: 0.1831:  56%|█████▋    | 53/94 [05:18<04:07,  6.04s/it]
Training Epoch: 8/100, batch 105/188 completed (loss: 0.0492:  56%|█████▋    | 53/94 [05:18<04:07,  6.04s/it]
Training Epoch: 8/100, batch 105/188 completed (loss: 0.0345:  56%|█████▋    | 53/94 [05:18<04:07,  6.04s/it]
Training Epoch: 8/100, batch 105/188 completed (loss: 0.0381:  56%|█████▋    | 53/94 [05:18<04:07,  6.04s/it]
Training Epoch: 8/100, batch 105/188 completed (loss: 0.0559:  56%|█████▋    | 53/94 [05:18<04:07,  6.04s/it]
Training Epoch: 8/100, batch 106/188 completed (loss: 0.0417:  56%|█████▋    | 53/94 [05:21<04:07,  6.04s/it]
Training Epoch: 8/100, batch 106/188 completed (loss: 0.0092:  56%|█████▋    | 53/94 [05:21<04:07,  6.04s/it]
Training Epoch: 8/100, batch 106/188 completed (loss: 0.0672:  56%|█████▋    | 53/94 [05:21<04:07,  6.04s/it]
Training Epoch: 8/100, batch 106/188 completed (loss: 0.289:  56%|█████▋    | 53/94 [05:21<04:07,  6.04s/it] 
Training Epoch: 8/100, batch 106/188 completed (loss: 0.289:  57%|█████▋    | 54/94 [05:24<04:03,  6.10s/it]
Training Epoch: 8/100, batch 106/188 completed (loss: 0.0672:  57%|█████▋    | 54/94 [05:24<04:03,  6.10s/it]
Training Epoch: 8/100, batch 106/188 completed (loss: 0.0092:  57%|█████▋    | 54/94 [05:24<04:03,  6.10s/it]
Training Epoch: 8/100, batch 106/188 completed (loss: 0.0417:  57%|█████▋    | 54/94 [05:24<04:04,  6.10s/it]
Training Epoch: 8/100, batch 107/188 completed (loss: 0.121:  57%|█████▋    | 54/94 [05:24<04:03,  6.10s/it] 
Training Epoch: 8/100, batch 107/188 completed (loss: 0.2884:  57%|█████▋    | 54/94 [05:24<04:03,  6.10s/it]
Training Epoch: 8/100, batch 107/188 completed (loss: 0.1371:  57%|█████▋    | 54/94 [05:24<04:04,  6.10s/it]
Training Epoch: 8/100, batch 107/188 completed (loss: 0.0779:  57%|█████▋    | 54/94 [05:24<04:03,  6.10s/it]
Training Epoch: 8/100, batch 108/188 completed (loss: 0.0673:  57%|█████▋    | 54/94 [05:27<04:03,  6.10s/it]
Training Epoch: 8/100, batch 108/188 completed (loss: 0.0964:  57%|█████▋    | 54/94 [05:27<04:04,  6.10s/it]
Training Epoch: 8/100, batch 108/188 completed (loss: 0.039:  57%|█████▋    | 54/94 [05:27<04:03,  6.10s/it]
Training Epoch: 8/100, batch 108/188 completed (loss: 0.1902:  57%|█████▋    | 54/94 [05:27<04:03,  6.10s/it]
Training Epoch: 8/100, batch 108/188 completed (loss: 0.039:  59%|█████▊    | 55/94 [05:30<03:56,  6.06s/it]
Training Epoch: 8/100, batch 108/188 completed (loss: 0.0964:  59%|█████▊    | 55/94 [05:30<03:56,  6.06s/it]
Training Epoch: 8/100, batch 108/188 completed (loss: 0.1902:  59%|█████▊    | 55/94 [05:30<03:56,  6.06s/it]
Training Epoch: 8/100, batch 108/188 completed (loss: 0.0673:  59%|█████▊    | 55/94 [05:30<03:56,  6.07s/it]
Training Epoch: 8/100, batch 109/188 completed (loss: 0.0332:  59%|█████▊    | 55/94 [05:30<03:56,  6.07s/it]
Training Epoch: 8/100, batch 109/188 completed (loss: 0.169:  59%|█████▊    | 55/94 [05:30<03:56,  6.06s/it] 
Training Epoch: 8/100, batch 109/188 completed (loss: 0.2181:  59%|█████▊    | 55/94 [05:30<03:56,  6.06s/it]
Training Epoch: 8/100, batch 109/188 completed (loss: 0.0653:  59%|█████▊    | 55/94 [05:30<03:56,  6.06s/it]
Training Epoch: 8/100, batch 110/188 completed (loss: 0.0545:  59%|█████▊    | 55/94 [05:33<03:56,  6.07s/it]
Training Epoch: 8/100, batch 110/188 completed (loss: 0.022:  59%|█████▊    | 55/94 [05:33<03:56,  6.06s/it]
Training Epoch: 8/100, batch 110/188 completed (loss: 0.1393:  59%|█████▊    | 55/94 [05:33<03:56,  6.06s/it]
Training Epoch: 8/100, batch 110/188 completed (loss: 0.1318:  59%|█████▊    | 55/94 [05:33<03:56,  6.06s/it]
Training Epoch: 8/100, batch 110/188 completed (loss: 0.1318:  60%|█████▉    | 56/94 [05:36<03:49,  6.03s/it]
Training Epoch: 8/100, batch 110/188 completed (loss: 0.022:  60%|█████▉    | 56/94 [05:36<03:49,  6.04s/it]
Training Epoch: 8/100, batch 110/188 completed (loss: 0.1393:  60%|█████▉    | 56/94 [05:36<03:49,  6.04s/it]
Training Epoch: 8/100, batch 110/188 completed (loss: 0.0545:  60%|█████▉    | 56/94 [05:36<03:49,  6.04s/it]
Training Epoch: 8/100, batch 111/188 completed (loss: 0.0601:  60%|█████▉    | 56/94 [05:36<03:49,  6.04s/it]
Training Epoch: 8/100, batch 111/188 completed (loss: 0.0656:  60%|█████▉    | 56/94 [05:36<03:49,  6.04s/it]
Training Epoch: 8/100, batch 111/188 completed (loss: 0.0534:  60%|█████▉    | 56/94 [05:36<03:49,  6.04s/it]
Training Epoch: 8/100, batch 111/188 completed (loss: 0.0726:  60%|█████▉    | 56/94 [05:36<03:49,  6.03s/it]
Training Epoch: 8/100, batch 112/188 completed (loss: 0.0545:  60%|█████▉    | 56/94 [05:39<03:49,  6.04s/it]
Training Epoch: 8/100, batch 112/188 completed (loss: 0.0559:  60%|█████▉    | 56/94 [05:39<03:49,  6.04s/it]
Training Epoch: 8/100, batch 112/188 completed (loss: 0.0405:  60%|█████▉    | 56/94 [05:39<03:49,  6.04s/it]
Training Epoch: 8/100, batch 112/188 completed (loss: 0.1265:  60%|█████▉    | 56/94 [05:39<03:49,  6.03s/it]
Training Epoch: 8/100, batch 112/188 completed (loss: 0.0405:  61%|██████    | 57/94 [05:42<03:43,  6.03s/it]
Training Epoch: 8/100, batch 112/188 completed (loss: 0.1265:  61%|██████    | 57/94 [05:42<03:43,  6.03s/it]
Training Epoch: 8/100, batch 112/188 completed (loss: 0.0559:  61%|██████    | 57/94 [05:42<03:43,  6.03s/it]
Training Epoch: 8/100, batch 112/188 completed (loss: 0.0545:  61%|██████    | 57/94 [05:42<03:43,  6.03s/it]
Training Epoch: 8/100, batch 113/188 completed (loss: 0.0492:  61%|██████    | 57/94 [05:42<03:43,  6.03s/it]
Training Epoch: 8/100, batch 113/188 completed (loss: 0.0396:  61%|██████    | 57/94 [05:42<03:43,  6.03s/it]
Training Epoch: 8/100, batch 113/188 completed (loss: 0.0201:  61%|██████    | 57/94 [05:42<03:43,  6.03s/it]
Training Epoch: 8/100, batch 113/188 completed (loss: 0.2888:  61%|██████    | 57/94 [05:42<03:43,  6.03s/it]
Training Epoch: 8/100, batch 114/188 completed (loss: 0.2606:  61%|██████    | 57/94 [05:45<03:43,  6.03s/it]
Training Epoch: 8/100, batch 114/188 completed (loss: 0.2212:  61%|██████    | 57/94 [05:45<03:43,  6.03s/it]
Training Epoch: 8/100, batch 114/188 completed (loss: 0.1393:  61%|██████    | 57/94 [05:45<03:43,  6.03s/it]
Training Epoch: 8/100, batch 114/188 completed (loss: 0.2643:  61%|██████    | 57/94 [05:45<03:43,  6.03s/it]
Training Epoch: 8/100, batch 114/188 completed (loss: 0.2606:  62%|██████▏   | 58/94 [05:48<03:36,  6.02s/it]
Training Epoch: 8/100, batch 114/188 completed (loss: 0.2643:  62%|██████▏   | 58/94 [05:48<03:36,  6.03s/it]
Training Epoch: 8/100, batch 114/188 completed (loss: 0.1393:  62%|██████▏   | 58/94 [05:48<03:36,  6.03s/it]
Training Epoch: 8/100, batch 114/188 completed (loss: 0.2212:  62%|██████▏   | 58/94 [05:48<03:37,  6.03s/it]
Training Epoch: 8/100, batch 115/188 completed (loss: 0.0467:  62%|██████▏   | 58/94 [05:48<03:37,  6.03s/it]
Training Epoch: 8/100, batch 115/188 completed (loss: 0.5391:  62%|██████▏   | 58/94 [05:48<03:36,  6.03s/it]
Training Epoch: 8/100, batch 115/188 completed (loss: 0.2113:  62%|██████▏   | 58/94 [05:48<03:36,  6.02s/it]
Training Epoch: 8/100, batch 115/188 completed (loss: 0.0234:  62%|██████▏   | 58/94 [05:48<03:36,  6.03s/it]
Training Epoch: 8/100, batch 116/188 completed (loss: 0.1635:  62%|██████▏   | 58/94 [05:51<03:36,  6.03s/it]
Training Epoch: 8/100, batch 116/188 completed (loss: 0.0229:  62%|██████▏   | 58/94 [05:51<03:37,  6.03s/it]
Training Epoch: 8/100, batch 116/188 completed (loss: 0.0377:  62%|██████▏   | 58/94 [05:51<03:36,  6.02s/it]
Training Epoch: 8/100, batch 116/188 completed (loss: 0.0832:  62%|██████▏   | 58/94 [05:51<03:36,  6.03s/it]
Training Epoch: 8/100, batch 116/188 completed (loss: 0.1635:  63%|██████▎   | 59/94 [05:54<03:30,  6.01s/it]
Training Epoch: 8/100, batch 116/188 completed (loss: 0.0229:  63%|██████▎   | 59/94 [05:54<03:30,  6.01s/it]
Training Epoch: 8/100, batch 116/188 completed (loss: 0.0377:  63%|██████▎   | 59/94 [05:54<03:30,  6.01s/it]
Training Epoch: 8/100, batch 116/188 completed (loss: 0.0832:  63%|██████▎   | 59/94 [05:54<03:30,  6.01s/it]
Training Epoch: 8/100, batch 117/188 completed (loss: 0.0764:  63%|██████▎   | 59/94 [05:54<03:30,  6.01s/it]
Training Epoch: 8/100, batch 117/188 completed (loss: 0.0808:  63%|██████▎   | 59/94 [05:54<03:30,  6.01s/it]
Training Epoch: 8/100, batch 117/188 completed (loss: 0.0611:  63%|██████▎   | 59/94 [05:54<03:30,  6.01s/it]
Training Epoch: 8/100, batch 117/188 completed (loss: 0.0322:  63%|██████▎   | 59/94 [05:54<03:30,  6.01s/it]
Training Epoch: 8/100, batch 118/188 completed (loss: 0.0288:  63%|██████▎   | 59/94 [05:57<03:30,  6.01s/it]
Training Epoch: 8/100, batch 118/188 completed (loss: 0.1721:  63%|██████▎   | 59/94 [05:57<03:30,  6.01s/it]
Training Epoch: 8/100, batch 118/188 completed (loss: 0.0981:  63%|██████▎   | 59/94 [05:57<03:30,  6.01s/it]
Training Epoch: 8/100, batch 118/188 completed (loss: 0.0685:  63%|██████▎   | 59/94 [05:57<03:30,  6.01s/it]
Training Epoch: 8/100, batch 118/188 completed (loss: 0.0288:  64%|██████▍   | 60/94 [06:00<03:25,  6.05s/it]
Training Epoch: 8/100, batch 118/188 completed (loss: 0.0685:  64%|██████▍   | 60/94 [06:00<03:25,  6.05s/it]
Training Epoch: 8/100, batch 118/188 completed (loss: 0.0981:  64%|██████▍   | 60/94 [06:00<03:25,  6.05s/it]
Training Epoch: 8/100, batch 118/188 completed (loss: 0.1721:  64%|██████▍   | 60/94 [06:00<03:25,  6.05s/it]
Training Epoch: 8/100, batch 119/188 completed (loss: 0.0345:  64%|██████▍   | 60/94 [06:00<03:25,  6.05s/it]
Training Epoch: 8/100, batch 119/188 completed (loss: 0.1303:  64%|██████▍   | 60/94 [06:00<03:25,  6.05s/it]
Training Epoch: 8/100, batch 119/188 completed (loss: 0.0881:  64%|██████▍   | 60/94 [06:00<03:25,  6.05s/it]
Training Epoch: 8/100, batch 119/188 completed (loss: 0.0363:  64%|██████▍   | 60/94 [06:00<03:25,  6.05s/it]
Training Epoch: 8/100, batch 120/188 completed (loss: 0.0501:  64%|██████▍   | 60/94 [06:03<03:25,  6.05s/it]
Training Epoch: 8/100, batch 120/188 completed (loss: 0.0841:  64%|██████▍   | 60/94 [06:03<03:25,  6.05s/it]
Training Epoch: 8/100, batch 120/188 completed (loss: 0.2724:  64%|██████▍   | 60/94 [06:03<03:25,  6.05s/it]
Training Epoch: 8/100, batch 120/188 completed (loss: 0.2497:  64%|██████▍   | 60/94 [06:03<03:25,  6.05s/it]
Training Epoch: 8/100, batch 120/188 completed (loss: 0.0841:  65%|██████▍   | 61/94 [06:06<03:19,  6.04s/it]
Training Epoch: 8/100, batch 120/188 completed (loss: 0.0501:  65%|██████▍   | 61/94 [06:06<03:19,  6.05s/it]
Training Epoch: 8/100, batch 120/188 completed (loss: 0.2497:  65%|██████▍   | 61/94 [06:06<03:19,  6.05s/it]
Training Epoch: 8/100, batch 120/188 completed (loss: 0.2724:  65%|██████▍   | 61/94 [06:06<03:19,  6.05s/it]
Training Epoch: 8/100, batch 121/188 completed (loss: 0.2551:  65%|██████▍   | 61/94 [06:06<03:19,  6.04s/it]
Training Epoch: 8/100, batch 121/188 completed (loss: 0.1413:  65%|██████▍   | 61/94 [06:06<03:19,  6.05s/it]
Training Epoch: 8/100, batch 121/188 completed (loss: 0.0574:  65%|██████▍   | 61/94 [06:06<03:19,  6.05s/it]
Training Epoch: 8/100, batch 121/188 completed (loss: 0.0215:  65%|██████▍   | 61/94 [06:06<03:19,  6.05s/it]
Training Epoch: 8/100, batch 122/188 completed (loss: 0.035:  65%|██████▍   | 61/94 [06:09<03:19,  6.05s/it] 
Training Epoch: 8/100, batch 122/188 completed (loss: 0.1503:  65%|██████▍   | 61/94 [06:09<03:19,  6.05s/it]
Training Epoch: 8/100, batch 122/188 completed (loss: 0.2184:  65%|██████▍   | 61/94 [06:09<03:19,  6.04s/it]
Training Epoch: 8/100, batch 122/188 completed (loss: 0.0415:  65%|██████▍   | 61/94 [06:09<03:19,  6.05s/it]
Training Epoch: 8/100, batch 122/188 completed (loss: 0.2184:  66%|██████▌   | 62/94 [06:12<03:15,  6.12s/it]
Training Epoch: 8/100, batch 122/188 completed (loss: 0.0415:  66%|██████▌   | 62/94 [06:12<03:15,  6.12s/it]
Training Epoch: 8/100, batch 122/188 completed (loss: 0.035:  66%|██████▌   | 62/94 [06:12<03:15,  6.12s/it]
Training Epoch: 8/100, batch 122/188 completed (loss: 0.1503:  66%|██████▌   | 62/94 [06:12<03:15,  6.12s/it]
Training Epoch: 8/100, batch 123/188 completed (loss: 0.0252:  66%|██████▌   | 62/94 [06:12<03:15,  6.12s/it]
Training Epoch: 8/100, batch 123/188 completed (loss: 0.018:  66%|██████▌   | 62/94 [06:12<03:15,  6.12s/it] 
Training Epoch: 8/100, batch 123/188 completed (loss: 0.0959:  66%|██████▌   | 62/94 [06:12<03:15,  6.12s/it]
Training Epoch: 8/100, batch 123/188 completed (loss: 0.0843:  66%|██████▌   | 62/94 [06:12<03:15,  6.12s/it]
Training Epoch: 8/100, batch 124/188 completed (loss: 0.1688:  66%|██████▌   | 62/94 [06:15<03:15,  6.12s/it]
Training Epoch: 8/100, batch 124/188 completed (loss: 0.0731:  66%|██████▌   | 62/94 [06:15<03:15,  6.12s/it]
Training Epoch: 8/100, batch 124/188 completed (loss: 0.0611:  66%|██████▌   | 62/94 [06:15<03:15,  6.12s/it]
Training Epoch: 8/100, batch 124/188 completed (loss: 0.0511:  66%|██████▌   | 62/94 [06:15<03:15,  6.12s/it]
Training Epoch: 8/100, batch 124/188 completed (loss: 0.0611:  67%|██████▋   | 63/94 [06:18<03:08,  6.09s/it]
Training Epoch: 8/100, batch 124/188 completed (loss: 0.0511:  67%|██████▋   | 63/94 [06:18<03:08,  6.09s/it]
Training Epoch: 8/100, batch 124/188 completed (loss: 0.1688:  67%|██████▋   | 63/94 [06:18<03:08,  6.09s/it]
Training Epoch: 8/100, batch 124/188 completed (loss: 0.0731:  67%|██████▋   | 63/94 [06:18<03:08,  6.09s/it]
Training Epoch: 8/100, batch 125/188 completed (loss: 0.1024:  67%|██████▋   | 63/94 [06:18<03:08,  6.09s/it]
Training Epoch: 8/100, batch 125/188 completed (loss: 0.183:  67%|██████▋   | 63/94 [06:18<03:08,  6.09s/it] 
Training Epoch: 8/100, batch 125/188 completed (loss: 0.0578:  67%|██████▋   | 63/94 [06:18<03:08,  6.09s/it]
Training Epoch: 8/100, batch 125/188 completed (loss: 0.0318:  67%|██████▋   | 63/94 [06:18<03:08,  6.09s/it]
Training Epoch: 8/100, batch 126/188 completed (loss: 0.0172:  67%|██████▋   | 63/94 [06:21<03:08,  6.09s/it]
Training Epoch: 8/100, batch 126/188 completed (loss: 0.0721:  67%|██████▋   | 63/94 [06:21<03:08,  6.09s/it]
Training Epoch: 8/100, batch 126/188 completed (loss: 0.1163:  67%|██████▋   | 63/94 [06:21<03:08,  6.09s/it]
Training Epoch: 8/100, batch 126/188 completed (loss: 0.0946:  67%|██████▋   | 63/94 [06:21<03:08,  6.09s/it]
Training Epoch: 8/100, batch 126/188 completed (loss: 0.0721:  68%|██████▊   | 64/94 [06:24<03:01,  6.06s/it]
Training Epoch: 8/100, batch 126/188 completed (loss: 0.1163:  68%|██████▊   | 64/94 [06:24<03:01,  6.06s/it]
Training Epoch: 8/100, batch 126/188 completed (loss: 0.0946:  68%|██████▊   | 64/94 [06:24<03:01,  6.06s/it]
Training Epoch: 8/100, batch 126/188 completed (loss: 0.0172:  68%|██████▊   | 64/94 [06:24<03:01,  6.06s/it]
Training Epoch: 8/100, batch 127/188 completed (loss: 0.1813:  68%|██████▊   | 64/94 [06:24<03:01,  6.06s/it]
Training Epoch: 8/100, batch 127/188 completed (loss: 0.1421:  68%|██████▊   | 64/94 [06:24<03:01,  6.06s/it]
Training Epoch: 8/100, batch 127/188 completed (loss: 0.0984:  68%|██████▊   | 64/94 [06:24<03:01,  6.06s/it]
Training Epoch: 8/100, batch 127/188 completed (loss: 0.0816:  68%|██████▊   | 64/94 [06:24<03:01,  6.06s/it]
Training Epoch: 8/100, batch 128/188 completed (loss: 0.0588:  68%|██████▊   | 64/94 [06:28<03:01,  6.06s/it]
Training Epoch: 8/100, batch 128/188 completed (loss: 0.0498:  68%|██████▊   | 64/94 [06:28<03:01,  6.06s/it]
Training Epoch: 8/100, batch 128/188 completed (loss: 0.0981:  68%|██████▊   | 64/94 [06:28<03:01,  6.06s/it]
Training Epoch: 8/100, batch 128/188 completed (loss: 0.065:  68%|██████▊   | 64/94 [06:28<03:01,  6.06s/it] 
Training Epoch: 8/100, batch 128/188 completed (loss: 0.0498:  69%|██████▉   | 65/94 [06:30<02:55,  6.06s/it]
Training Epoch: 8/100, batch 128/188 completed (loss: 0.065:  69%|██████▉   | 65/94 [06:30<02:55,  6.06s/it]
Training Epoch: 8/100, batch 128/188 completed (loss: 0.0588:  69%|██████▉   | 65/94 [06:30<02:55,  6.06s/it]
Training Epoch: 8/100, batch 128/188 completed (loss: 0.0981:  69%|██████▉   | 65/94 [06:30<02:55,  6.06s/it]
Training Epoch: 8/100, batch 129/188 completed (loss: 0.085:  69%|██████▉   | 65/94 [06:30<02:55,  6.06s/it] 
Training Epoch: 8/100, batch 129/188 completed (loss: 0.0367:  69%|██████▉   | 65/94 [06:30<02:55,  6.06s/it]
Training Epoch: 8/100, batch 129/188 completed (loss: 0.0402:  69%|██████▉   | 65/94 [06:30<02:55,  6.06s/it]
Training Epoch: 8/100, batch 129/188 completed (loss: 0.0481:  69%|██████▉   | 65/94 [06:30<02:55,  6.06s/it]
Training Epoch: 8/100, batch 130/188 completed (loss: 0.2932:  69%|██████▉   | 65/94 [06:34<02:55,  6.06s/it]
Training Epoch: 8/100, batch 130/188 completed (loss: 0.0706:  69%|██████▉   | 65/94 [06:34<02:55,  6.06s/it]
Training Epoch: 8/100, batch 130/188 completed (loss: 0.1794:  69%|██████▉   | 65/94 [06:34<02:55,  6.06s/it]
Training Epoch: 8/100, batch 130/188 completed (loss: 0.0553:  69%|██████▉   | 65/94 [06:34<02:55,  6.06s/it]
Training Epoch: 8/100, batch 130/188 completed (loss: 0.1794:  70%|███████   | 66/94 [06:36<02:49,  6.05s/it]
Training Epoch: 8/100, batch 130/188 completed (loss: 0.0553:  70%|███████   | 66/94 [06:36<02:49,  6.05s/it]
Training Epoch: 8/100, batch 130/188 completed (loss: 0.2932:  70%|███████   | 66/94 [06:36<02:49,  6.05s/it]
Training Epoch: 8/100, batch 130/188 completed (loss: 0.0706:  70%|███████   | 66/94 [06:36<02:49,  6.05s/it]
Training Epoch: 8/100, batch 131/188 completed (loss: 0.1312:  70%|███████   | 66/94 [06:36<02:49,  6.05s/it]
Training Epoch: 8/100, batch 131/188 completed (loss: 0.0559:  70%|███████   | 66/94 [06:36<02:49,  6.05s/it]
Training Epoch: 8/100, batch 131/188 completed (loss: 0.1099:  70%|███████   | 66/94 [06:36<02:49,  6.05s/it]
Training Epoch: 8/100, batch 131/188 completed (loss: 0.0263:  70%|███████   | 66/94 [06:36<02:49,  6.05s/it]
Training Epoch: 8/100, batch 132/188 completed (loss: 0.019:  70%|███████   | 66/94 [06:39<02:49,  6.05s/it] 
Training Epoch: 8/100, batch 132/188 completed (loss: 0.0606:  70%|███████   | 66/94 [06:39<02:49,  6.05s/it]
Training Epoch: 8/100, batch 132/188 completed (loss: 0.0525:  70%|███████   | 66/94 [06:39<02:49,  6.05s/it]
Training Epoch: 8/100, batch 132/188 completed (loss: 0.1958:  70%|███████   | 66/94 [06:39<02:49,  6.05s/it]
Training Epoch: 8/100, batch 132/188 completed (loss: 0.1958:  71%|███████▏  | 67/94 [06:43<02:45,  6.12s/it]
Training Epoch: 8/100, batch 132/188 completed (loss: 0.0525:  71%|███████▏  | 67/94 [06:43<02:45,  6.12s/it]
Training Epoch: 8/100, batch 132/188 completed (loss: 0.019:  71%|███████▏  | 67/94 [06:43<02:45,  6.12s/it]
Training Epoch: 8/100, batch 132/188 completed (loss: 0.0606:  71%|███████▏  | 67/94 [06:43<02:45,  6.12s/it]
Training Epoch: 8/100, batch 133/188 completed (loss: 0.4505:  71%|███████▏  | 67/94 [06:43<02:45,  6.12s/it]
Training Epoch: 8/100, batch 133/188 completed (loss: 0.0402:  71%|███████▏  | 67/94 [06:43<02:45,  6.12s/it]
Training Epoch: 8/100, batch 133/188 completed (loss: 0.1773:  71%|███████▏  | 67/94 [06:43<02:45,  6.12s/it]
Training Epoch: 8/100, batch 133/188 completed (loss: 0.2221:  71%|███████▏  | 67/94 [06:43<02:45,  6.12s/it]
Training Epoch: 8/100, batch 134/188 completed (loss: 0.2025:  71%|███████▏  | 67/94 [06:46<02:45,  6.12s/it]
Training Epoch: 8/100, batch 134/188 completed (loss: 0.1757:  71%|███████▏  | 67/94 [06:46<02:45,  6.12s/it]
Training Epoch: 8/100, batch 134/188 completed (loss: 0.0623:  71%|███████▏  | 67/94 [06:46<02:45,  6.12s/it]
Training Epoch: 8/100, batch 134/188 completed (loss: 0.0503:  71%|███████▏  | 67/94 [06:46<02:45,  6.12s/it]
Training Epoch: 8/100, batch 134/188 completed (loss: 0.0623:  72%|███████▏  | 68/94 [06:49<02:38,  6.09s/it]
Training Epoch: 8/100, batch 134/188 completed (loss: 0.0503:  72%|███████▏  | 68/94 [06:49<02:38,  6.09s/it]
Training Epoch: 8/100, batch 134/188 completed (loss: 0.1757:  72%|███████▏  | 68/94 [06:49<02:38,  6.09s/it]
Training Epoch: 8/100, batch 134/188 completed (loss: 0.2025:  72%|███████▏  | 68/94 [06:49<02:38,  6.09s/it]
Training Epoch: 8/100, batch 135/188 completed (loss: 0.2506:  72%|███████▏  | 68/94 [06:49<02:38,  6.09s/it]
Training Epoch: 8/100, batch 135/188 completed (loss: 0.0629:  72%|███████▏  | 68/94 [06:49<02:38,  6.09s/it]
Training Epoch: 8/100, batch 135/188 completed (loss: 0.0367:  72%|███████▏  | 68/94 [06:49<02:38,  6.09s/it]
Training Epoch: 8/100, batch 135/188 completed (loss: 0.1618:  72%|███████▏  | 68/94 [06:49<02:38,  6.09s/it]
Training Epoch: 8/100, batch 136/188 completed (loss: 0.0382:  72%|███████▏  | 68/94 [06:52<02:38,  6.09s/it]
Training Epoch: 8/100, batch 136/188 completed (loss: 0.0489:  72%|███████▏  | 68/94 [06:52<02:38,  6.09s/it]
Training Epoch: 8/100, batch 136/188 completed (loss: 0.0439:  72%|███████▏  | 68/94 [06:52<02:38,  6.09s/it]
Training Epoch: 8/100, batch 136/188 completed (loss: 0.0537:  72%|███████▏  | 68/94 [06:52<02:38,  6.09s/it]
Training Epoch: 8/100, batch 136/188 completed (loss: 0.0382:  73%|███████▎  | 69/94 [06:55<02:30,  6.00s/it]
Training Epoch: 8/100, batch 136/188 completed (loss: 0.0439:  73%|███████▎  | 69/94 [06:55<02:30,  6.01s/it]
Training Epoch: 8/100, batch 136/188 completed (loss: 0.0537:  73%|███████▎  | 69/94 [06:55<02:30,  6.01s/it]
Training Epoch: 8/100, batch 136/188 completed (loss: 0.0489:  73%|███████▎  | 69/94 [06:55<02:30,  6.01s/it]
Training Epoch: 8/100, batch 137/188 completed (loss: 0.1742:  73%|███████▎  | 69/94 [06:55<02:30,  6.01s/it]
Training Epoch: 8/100, batch 137/188 completed (loss: 0.0571:  73%|███████▎  | 69/94 [06:55<02:30,  6.00s/it]
Training Epoch: 8/100, batch 137/188 completed (loss: 0.1701:  73%|███████▎  | 69/94 [06:55<02:30,  6.01s/it]
Training Epoch: 8/100, batch 137/188 completed (loss: 0.0104:  73%|███████▎  | 69/94 [06:55<02:30,  6.01s/it]
Training Epoch: 8/100, batch 138/188 completed (loss: 0.0464:  73%|███████▎  | 69/94 [06:58<02:30,  6.01s/it]
Training Epoch: 8/100, batch 138/188 completed (loss: 0.1928:  73%|███████▎  | 69/94 [06:58<02:30,  6.00s/it]
Training Epoch: 8/100, batch 138/188 completed (loss: 0.0588:  73%|███████▎  | 69/94 [06:58<02:30,  6.01s/it]
Training Epoch: 8/100, batch 138/188 completed (loss: 0.1826:  73%|███████▎  | 69/94 [06:58<02:30,  6.01s/it]
Training Epoch: 8/100, batch 138/188 completed (loss: 0.0464:  74%|███████▍  | 70/94 [07:01<02:26,  6.09s/it]
Training Epoch: 8/100, batch 138/188 completed (loss: 0.1826:  74%|███████▍  | 70/94 [07:01<02:26,  6.09s/it]
Training Epoch: 8/100, batch 138/188 completed (loss: 0.1928:  74%|███████▍  | 70/94 [07:01<02:26,  6.09s/it]
Training Epoch: 8/100, batch 138/188 completed (loss: 0.0588:  74%|███████▍  | 70/94 [07:01<02:26,  6.09s/it]
Training Epoch: 8/100, batch 139/188 completed (loss: 0.2316:  74%|███████▍  | 70/94 [07:01<02:26,  6.09s/it]
Training Epoch: 8/100, batch 139/188 completed (loss: 0.0343:  74%|███████▍  | 70/94 [07:01<02:26,  6.09s/it]
Training Epoch: 8/100, batch 139/188 completed (loss: 0.2502:  74%|███████▍  | 70/94 [07:01<02:26,  6.09s/it]
Training Epoch: 8/100, batch 139/188 completed (loss: 0.1936:  74%|███████▍  | 70/94 [07:01<02:26,  6.09s/it]
Training Epoch: 8/100, batch 140/188 completed (loss: 0.031:  74%|███████▍  | 70/94 [07:04<02:26,  6.09s/it] 
Training Epoch: 8/100, batch 140/188 completed (loss: 0.0331:  74%|███████▍  | 70/94 [07:04<02:26,  6.09s/it]
Training Epoch: 8/100, batch 140/188 completed (loss: 0.1595:  74%|███████▍  | 70/94 [07:04<02:26,  6.09s/it]
Training Epoch: 8/100, batch 140/188 completed (loss: 0.0514:  74%|███████▍  | 70/94 [07:04<02:26,  6.09s/it]
Training Epoch: 8/100, batch 140/188 completed (loss: 0.031:  76%|███████▌  | 71/94 [07:07<02:19,  6.07s/it]
Training Epoch: 8/100, batch 140/188 completed (loss: 0.1595:  76%|███████▌  | 71/94 [07:07<02:19,  6.07s/it]
Training Epoch: 8/100, batch 140/188 completed (loss: 0.0514:  76%|███████▌  | 71/94 [07:07<02:19,  6.07s/it]
Training Epoch: 8/100, batch 140/188 completed (loss: 0.0331:  76%|███████▌  | 71/94 [07:07<02:19,  6.07s/it]
Training Epoch: 8/100, batch 141/188 completed (loss: 0.0513:  76%|███████▌  | 71/94 [07:07<02:19,  6.07s/it]
Training Epoch: 8/100, batch 141/188 completed (loss: 0.0376:  76%|███████▌  | 71/94 [07:07<02:19,  6.07s/it]
Training Epoch: 8/100, batch 141/188 completed (loss: 0.0362:  76%|███████▌  | 71/94 [07:07<02:19,  6.07s/it]
Training Epoch: 8/100, batch 141/188 completed (loss: 0.0471:  76%|███████▌  | 71/94 [07:07<02:19,  6.07s/it]
Training Epoch: 8/100, batch 142/188 completed (loss: 0.1029:  76%|███████▌  | 71/94 [07:10<02:19,  6.07s/it]
Training Epoch: 8/100, batch 142/188 completed (loss: 0.09:  76%|███████▌  | 71/94 [07:10<02:19,  6.07s/it]  
Training Epoch: 8/100, batch 142/188 completed (loss: 0.0294:  76%|███████▌  | 71/94 [07:10<02:19,  6.07s/it]
Training Epoch: 8/100, batch 142/188 completed (loss: 0.0456:  76%|███████▌  | 71/94 [07:10<02:19,  6.07s/it]
Training Epoch: 8/100, batch 142/188 completed (loss: 0.1029:  77%|███████▋  | 72/94 [07:13<02:14,  6.09s/it]
Training Epoch: 8/100, batch 142/188 completed (loss: 0.0456:  77%|███████▋  | 72/94 [07:13<02:14,  6.09s/it]
Training Epoch: 8/100, batch 142/188 completed (loss: 0.0294:  77%|███████▋  | 72/94 [07:13<02:14,  6.09s/it]
Training Epoch: 8/100, batch 142/188 completed (loss: 0.09:  77%|███████▋  | 72/94 [07:13<02:14,  6.10s/it]
Training Epoch: 8/100, batch 143/188 completed (loss: 0.1065:  77%|███████▋  | 72/94 [07:13<02:14,  6.10s/it]
Training Epoch: 8/100, batch 143/188 completed (loss: 0.0986:  77%|███████▋  | 72/94 [07:13<02:14,  6.09s/it]
Training Epoch: 8/100, batch 143/188 completed (loss: 0.1151:  77%|███████▋  | 72/94 [07:13<02:14,  6.09s/it]
Training Epoch: 8/100, batch 143/188 completed (loss: 0.0296:  77%|███████▋  | 72/94 [07:13<02:14,  6.09s/it]
Training Epoch: 8/100, batch 144/188 completed (loss: 0.045:  77%|███████▋  | 72/94 [07:16<02:14,  6.10s/it] 
Training Epoch: 8/100, batch 144/188 completed (loss: 0.1171:  77%|███████▋  | 72/94 [07:16<02:14,  6.09s/it]
Training Epoch: 8/100, batch 144/188 completed (loss: 0.1803:  77%|███████▋  | 72/94 [07:16<02:14,  6.09s/it]
Training Epoch: 8/100, batch 144/188 completed (loss: 0.0964:  77%|███████▋  | 72/94 [07:16<02:14,  6.09s/it]
Training Epoch: 8/100, batch 144/188 completed (loss: 0.1803:  78%|███████▊  | 73/94 [07:19<02:07,  6.07s/it]
Training Epoch: 8/100, batch 144/188 completed (loss: 0.0964:  78%|███████▊  | 73/94 [07:19<02:07,  6.07s/it]
Training Epoch: 8/100, batch 144/188 completed (loss: 0.045:  78%|███████▊  | 73/94 [07:19<02:07,  6.07s/it]
Training Epoch: 8/100, batch 144/188 completed (loss: 0.1171:  78%|███████▊  | 73/94 [07:19<02:07,  6.08s/it]
Training Epoch: 8/100, batch 145/188 completed (loss: 0.0582:  78%|███████▊  | 73/94 [07:19<02:07,  6.07s/it]
Training Epoch: 8/100, batch 145/188 completed (loss: 0.0266:  78%|███████▊  | 73/94 [07:19<02:07,  6.08s/it]
Training Epoch: 8/100, batch 145/188 completed (loss: 0.0439:  78%|███████▊  | 73/94 [07:19<02:07,  6.07s/it]
Training Epoch: 8/100, batch 145/188 completed (loss: 0.052:  78%|███████▊  | 73/94 [07:19<02:07,  6.07s/it] 
Training Epoch: 8/100, batch 146/188 completed (loss: 0.0236:  78%|███████▊  | 73/94 [07:22<02:07,  6.07s/it]
Training Epoch: 8/100, batch 146/188 completed (loss: 0.1612:  78%|███████▊  | 73/94 [07:22<02:07,  6.08s/it]
Training Epoch: 8/100, batch 146/188 completed (loss: 0.0896:  78%|███████▊  | 73/94 [07:22<02:07,  6.07s/it]
Training Epoch: 8/100, batch 146/188 completed (loss: 0.0692:  78%|███████▊  | 73/94 [07:22<02:07,  6.07s/it]
Training Epoch: 8/100, batch 146/188 completed (loss: 0.0236:  79%|███████▊  | 74/94 [07:25<02:01,  6.06s/it]
Training Epoch: 8/100, batch 146/188 completed (loss: 0.0692:  79%|███████▊  | 74/94 [07:25<02:01,  6.06s/it]
Training Epoch: 8/100, batch 146/188 completed (loss: 0.0896:  79%|███████▊  | 74/94 [07:25<02:01,  6.06s/it]
Training Epoch: 8/100, batch 146/188 completed (loss: 0.1612:  79%|███████▊  | 74/94 [07:25<02:01,  6.06s/it]
Training Epoch: 8/100, batch 147/188 completed (loss: 0.036:  79%|███████▊  | 74/94 [07:25<02:01,  6.06s/it] 
Training Epoch: 8/100, batch 147/188 completed (loss: 0.0124:  79%|███████▊  | 74/94 [07:25<02:01,  6.06s/it]
Training Epoch: 8/100, batch 147/188 completed (loss: 0.1375:  79%|███████▊  | 74/94 [07:25<02:01,  6.06s/it]
Training Epoch: 8/100, batch 147/188 completed (loss: 0.0101:  79%|███████▊  | 74/94 [07:25<02:01,  6.06s/it]
Training Epoch: 8/100, batch 148/188 completed (loss: 0.0557:  79%|███████▊  | 74/94 [07:28<02:01,  6.06s/it]
Training Epoch: 8/100, batch 148/188 completed (loss: 0.2272:  79%|███████▊  | 74/94 [07:28<02:01,  6.06s/it]
Training Epoch: 8/100, batch 148/188 completed (loss: 0.0919:  79%|███████▊  | 74/94 [07:28<02:01,  6.06s/it]
Training Epoch: 8/100, batch 148/188 completed (loss: 0.054:  79%|███████▊  | 74/94 [07:28<02:01,  6.06s/it] 
Training Epoch: 8/100, batch 148/188 completed (loss: 0.0919:  80%|███████▉  | 75/94 [07:31<01:56,  6.14s/it]
Training Epoch: 8/100, batch 148/188 completed (loss: 0.2272:  80%|███████▉  | 75/94 [07:31<01:56,  6.14s/it]
Training Epoch: 8/100, batch 148/188 completed (loss: 0.0557:  80%|███████▉  | 75/94 [07:31<01:56,  6.14s/it]
Training Epoch: 8/100, batch 148/188 completed (loss: 0.054:  80%|███████▉  | 75/94 [07:31<01:56,  6.14s/it]
Training Epoch: 8/100, batch 149/188 completed (loss: 0.054:  80%|███████▉  | 75/94 [07:31<01:56,  6.14s/it] 
Training Epoch: 8/100, batch 149/188 completed (loss: 0.2792:  80%|███████▉  | 75/94 [07:31<01:56,  6.14s/it]
Training Epoch: 8/100, batch 149/188 completed (loss: 0.0199:  80%|███████▉  | 75/94 [07:31<01:56,  6.14s/it]
Training Epoch: 8/100, batch 149/188 completed (loss: 0.0754:  80%|███████▉  | 75/94 [07:31<01:56,  6.14s/it]
Training Epoch: 8/100, batch 150/188 completed (loss: 0.0589:  80%|███████▉  | 75/94 [07:34<01:56,  6.14s/it]
Training Epoch: 8/100, batch 150/188 completed (loss: 0.0334:  80%|███████▉  | 75/94 [07:34<01:56,  6.14s/it]
Training Epoch: 8/100, batch 150/188 completed (loss: 0.1643:  80%|███████▉  | 75/94 [07:34<01:56,  6.14s/it]
Training Epoch: 8/100, batch 150/188 completed (loss: 0.0847:  80%|███████▉  | 75/94 [07:34<01:56,  6.14s/it]
Training Epoch: 8/100, batch 150/188 completed (loss: 0.0589:  81%|████████  | 76/94 [07:37<01:49,  6.10s/it]
Training Epoch: 8/100, batch 150/188 completed (loss: 0.0847:  81%|████████  | 76/94 [07:37<01:49,  6.10s/it]
Training Epoch: 8/100, batch 150/188 completed (loss: 0.0334:  81%|████████  | 76/94 [07:37<01:49,  6.09s/it]
Training Epoch: 8/100, batch 150/188 completed (loss: 0.1643:  81%|████████  | 76/94 [07:37<01:49,  6.10s/it]
Training Epoch: 8/100, batch 151/188 completed (loss: 0.0278:  81%|████████  | 76/94 [07:37<01:49,  6.10s/it]
Training Epoch: 8/100, batch 151/188 completed (loss: 0.0457:  81%|████████  | 76/94 [07:37<01:49,  6.10s/it]
Training Epoch: 8/100, batch 151/188 completed (loss: 0.0108:  81%|████████  | 76/94 [07:37<01:49,  6.09s/it]
Training Epoch: 8/100, batch 151/188 completed (loss: 0.1442:  81%|████████  | 76/94 [07:37<01:49,  6.10s/it]
Training Epoch: 8/100, batch 152/188 completed (loss: 0.4275:  81%|████████  | 76/94 [07:40<01:49,  6.10s/it]
Training Epoch: 8/100, batch 152/188 completed (loss: 0.0198:  81%|████████  | 76/94 [07:40<01:49,  6.09s/it]
Training Epoch: 8/100, batch 152/188 completed (loss: 0.0702:  81%|████████  | 76/94 [07:40<01:49,  6.10s/it]
Training Epoch: 8/100, batch 152/188 completed (loss: 0.0271:  81%|████████  | 76/94 [07:40<01:49,  6.10s/it]
Training Epoch: 8/100, batch 152/188 completed (loss: 0.0702:  82%|████████▏ | 77/94 [07:43<01:43,  6.07s/it]
Training Epoch: 8/100, batch 152/188 completed (loss: 0.0198:  82%|████████▏ | 77/94 [07:43<01:43,  6.06s/it]
Training Epoch: 8/100, batch 152/188 completed (loss: 0.0271:  82%|████████▏ | 77/94 [07:43<01:43,  6.07s/it]
Training Epoch: 8/100, batch 152/188 completed (loss: 0.4275:  82%|████████▏ | 77/94 [07:43<01:43,  6.06s/it]
Training Epoch: 8/100, batch 153/188 completed (loss: 0.0863:  82%|████████▏ | 77/94 [07:43<01:43,  6.07s/it]
Training Epoch: 8/100, batch 153/188 completed (loss: 0.3596:  82%|████████▏ | 77/94 [07:43<01:43,  6.06s/it]
Training Epoch: 8/100, batch 153/188 completed (loss: 0.1406:  82%|████████▏ | 77/94 [07:43<01:43,  6.06s/it]
Training Epoch: 8/100, batch 153/188 completed (loss: 0.0416:  82%|████████▏ | 77/94 [07:43<01:43,  6.07s/it]
Training Epoch: 8/100, batch 154/188 completed (loss: 0.1419:  82%|████████▏ | 77/94 [07:47<01:43,  6.07s/it]
Training Epoch: 8/100, batch 154/188 completed (loss: 0.2156:  82%|████████▏ | 77/94 [07:47<01:43,  6.06s/it]
Training Epoch: 8/100, batch 154/188 completed (loss: 0.0348:  82%|████████▏ | 77/94 [07:47<01:43,  6.06s/it]
Training Epoch: 8/100, batch 154/188 completed (loss: 0.045:  82%|████████▏ | 77/94 [07:47<01:43,  6.07s/it] 
Training Epoch: 8/100, batch 154/188 completed (loss: 0.1419:  83%|████████▎ | 78/94 [07:50<01:37,  6.10s/it]
Training Epoch: 8/100, batch 154/188 completed (loss: 0.2156:  83%|████████▎ | 78/94 [07:50<01:37,  6.09s/it]
Training Epoch: 8/100, batch 154/188 completed (loss: 0.045:  83%|████████▎ | 78/94 [07:50<01:37,  6.10s/it]
Training Epoch: 8/100, batch 154/188 completed (loss: 0.0348:  83%|████████▎ | 78/94 [07:50<01:37,  6.09s/it]
Training Epoch: 8/100, batch 155/188 completed (loss: 0.2212:  83%|████████▎ | 78/94 [07:50<01:37,  6.10s/it]
Training Epoch: 8/100, batch 155/188 completed (loss: 0.0465:  83%|████████▎ | 78/94 [07:50<01:37,  6.09s/it]
Training Epoch: 8/100, batch 155/188 completed (loss: 0.1068:  83%|████████▎ | 78/94 [07:50<01:37,  6.09s/it]
Training Epoch: 8/100, batch 155/188 completed (loss: 0.1306:  83%|████████▎ | 78/94 [07:50<01:37,  6.10s/it]
Training Epoch: 8/100, batch 156/188 completed (loss: 0.324:  83%|████████▎ | 78/94 [07:53<01:37,  6.09s/it] 
Training Epoch: 8/100, batch 156/188 completed (loss: 0.0543:  83%|████████▎ | 78/94 [07:53<01:37,  6.10s/it]
Training Epoch: 8/100, batch 156/188 completed (loss: 0.2905:  83%|████████▎ | 78/94 [07:53<01:37,  6.09s/it]
Training Epoch: 8/100, batch 156/188 completed (loss: 0.2941:  83%|████████▎ | 78/94 [07:53<01:37,  6.10s/it]
Training Epoch: 8/100, batch 156/188 completed (loss: 0.0543:  84%|████████▍ | 79/94 [07:56<01:31,  6.07s/it]
Training Epoch: 8/100, batch 156/188 completed (loss: 0.324:  84%|████████▍ | 79/94 [07:56<01:31,  6.07s/it]
Training Epoch: 8/100, batch 156/188 completed (loss: 0.2905:  84%|████████▍ | 79/94 [07:56<01:31,  6.07s/it]
Training Epoch: 8/100, batch 156/188 completed (loss: 0.2941:  84%|████████▍ | 79/94 [07:56<01:31,  6.07s/it]
Training Epoch: 8/100, batch 157/188 completed (loss: 0.1155:  84%|████████▍ | 79/94 [07:56<01:31,  6.07s/it]
Training Epoch: 8/100, batch 157/188 completed (loss: 0.0967:  84%|████████▍ | 79/94 [07:56<01:31,  6.07s/it]
Training Epoch: 8/100, batch 157/188 completed (loss: 0.1037:  84%|████████▍ | 79/94 [07:56<01:31,  6.07s/it]
Training Epoch: 8/100, batch 157/188 completed (loss: 0.1005:  84%|████████▍ | 79/94 [07:56<01:31,  6.07s/it]
Training Epoch: 8/100, batch 158/188 completed (loss: 0.1102:  84%|████████▍ | 79/94 [07:59<01:31,  6.07s/it]
Training Epoch: 8/100, batch 158/188 completed (loss: 0.0896:  84%|████████▍ | 79/94 [07:59<01:31,  6.07s/it]
Training Epoch: 8/100, batch 158/188 completed (loss: 0.1258:  84%|████████▍ | 79/94 [07:59<01:31,  6.07s/it]
Training Epoch: 8/100, batch 158/188 completed (loss: 0.0152:  84%|████████▍ | 79/94 [07:59<01:31,  6.07s/it]
Training Epoch: 8/100, batch 158/188 completed (loss: 0.1258:  85%|████████▌ | 80/94 [08:02<01:25,  6.09s/it]
Training Epoch: 8/100, batch 158/188 completed (loss: 0.0896:  85%|████████▌ | 80/94 [08:02<01:25,  6.09s/it]
Training Epoch: 8/100, batch 158/188 completed (loss: 0.1102:  85%|████████▌ | 80/94 [08:02<01:25,  6.09s/it]
Training Epoch: 8/100, batch 158/188 completed (loss: 0.0152:  85%|████████▌ | 80/94 [08:02<01:25,  6.09s/it]
Training Epoch: 8/100, batch 159/188 completed (loss: 0.014:  85%|████████▌ | 80/94 [08:02<01:25,  6.09s/it] 
Training Epoch: 8/100, batch 159/188 completed (loss: 0.0882:  85%|████████▌ | 80/94 [08:02<01:25,  6.09s/it]
Training Epoch: 8/100, batch 159/188 completed (loss: 0.0841:  85%|████████▌ | 80/94 [08:02<01:25,  6.09s/it]
Training Epoch: 8/100, batch 159/188 completed (loss: 0.0561:  85%|████████▌ | 80/94 [08:02<01:25,  6.09s/it]
Training Epoch: 8/100, batch 160/188 completed (loss: 0.1332:  85%|████████▌ | 80/94 [08:05<01:25,  6.09s/it]
Training Epoch: 8/100, batch 160/188 completed (loss: 0.0208:  85%|████████▌ | 80/94 [08:05<01:25,  6.09s/it]
Training Epoch: 8/100, batch 160/188 completed (loss: 0.0308:  85%|████████▌ | 80/94 [08:05<01:25,  6.09s/it]
Training Epoch: 8/100, batch 160/188 completed (loss: 0.0646:  85%|████████▌ | 80/94 [08:05<01:25,  6.09s/it]
Training Epoch: 8/100, batch 160/188 completed (loss: 0.0208:  86%|████████▌ | 81/94 [08:08<01:18,  6.06s/it]
Training Epoch: 8/100, batch 160/188 completed (loss: 0.0308:  86%|████████▌ | 81/94 [08:08<01:18,  6.06s/it]
Training Epoch: 8/100, batch 160/188 completed (loss: 0.0646:  86%|████████▌ | 81/94 [08:08<01:18,  6.06s/it]
Training Epoch: 8/100, batch 160/188 completed (loss: 0.1332:  86%|████████▌ | 81/94 [08:08<01:18,  6.07s/it]
Training Epoch: 8/100, batch 161/188 completed (loss: 0.2126:  86%|████████▌ | 81/94 [08:08<01:18,  6.07s/it]
Training Epoch: 8/100, batch 161/188 completed (loss: 0.0887:  86%|████████▌ | 81/94 [08:08<01:18,  6.06s/it]
Training Epoch: 8/100, batch 161/188 completed (loss: 0.0109:  86%|████████▌ | 81/94 [08:08<01:18,  6.06s/it]
Training Epoch: 8/100, batch 161/188 completed (loss: 0.0047:  86%|████████▌ | 81/94 [08:08<01:18,  6.06s/it]
Training Epoch: 8/100, batch 162/188 completed (loss: 0.2037:  86%|████████▌ | 81/94 [08:11<01:18,  6.06s/it]
Training Epoch: 8/100, batch 162/188 completed (loss: 0.1608:  86%|████████▌ | 81/94 [08:11<01:18,  6.07s/it]
Training Epoch: 8/100, batch 162/188 completed (loss: 0.3056:  86%|████████▌ | 81/94 [08:11<01:18,  6.06s/it]
Training Epoch: 8/100, batch 162/188 completed (loss: 0.173:  86%|████████▌ | 81/94 [08:11<01:18,  6.06s/it] 
Training Epoch: 8/100, batch 162/188 completed (loss: 0.1608:  87%|████████▋ | 82/94 [08:14<01:12,  6.05s/it]
Training Epoch: 8/100, batch 162/188 completed (loss: 0.2037:  87%|████████▋ | 82/94 [08:14<01:12,  6.05s/it]
Training Epoch: 8/100, batch 162/188 completed (loss: 0.173:  87%|████████▋ | 82/94 [08:14<01:12,  6.05s/it]
Training Epoch: 8/100, batch 162/188 completed (loss: 0.3056:  87%|████████▋ | 82/94 [08:14<01:12,  6.05s/it]
Training Epoch: 8/100, batch 163/188 completed (loss: 0.0927:  87%|████████▋ | 82/94 [08:14<01:12,  6.05s/it]
Training Epoch: 8/100, batch 163/188 completed (loss: 0.0582:  87%|████████▋ | 82/94 [08:14<01:12,  6.05s/it]
Training Epoch: 8/100, batch 163/188 completed (loss: 0.3474:  87%|████████▋ | 82/94 [08:14<01:12,  6.05s/it]
Training Epoch: 8/100, batch 163/188 completed (loss: 0.0826:  87%|████████▋ | 82/94 [08:14<01:12,  6.05s/it]
Training Epoch: 8/100, batch 164/188 completed (loss: 0.0522:  87%|████████▋ | 82/94 [08:17<01:12,  6.05s/it]
Training Epoch: 8/100, batch 164/188 completed (loss: 0.1782:  87%|████████▋ | 82/94 [08:17<01:12,  6.05s/it]
Training Epoch: 8/100, batch 164/188 completed (loss: 0.1046:  87%|████████▋ | 82/94 [08:17<01:12,  6.05s/it]
Training Epoch: 8/100, batch 164/188 completed (loss: 0.0619:  87%|████████▋ | 82/94 [08:17<01:12,  6.05s/it]
Training Epoch: 8/100, batch 164/188 completed (loss: 0.1046:  88%|████████▊ | 83/94 [08:20<01:07,  6.12s/it]
Training Epoch: 8/100, batch 164/188 completed (loss: 0.1782:  88%|████████▊ | 83/94 [08:20<01:07,  6.12s/it]
Training Epoch: 8/100, batch 164/188 completed (loss: 0.0522:  88%|████████▊ | 83/94 [08:20<01:07,  6.12s/it]
Training Epoch: 8/100, batch 164/188 completed (loss: 0.0619:  88%|████████▊ | 83/94 [08:20<01:07,  6.13s/it]
Training Epoch: 8/100, batch 165/188 completed (loss: 0.0224:  88%|████████▊ | 83/94 [08:20<01:07,  6.12s/it]
Training Epoch: 8/100, batch 165/188 completed (loss: 0.139:  88%|████████▊ | 83/94 [08:20<01:07,  6.12s/it] 
Training Epoch: 8/100, batch 165/188 completed (loss: 0.0437:  88%|████████▊ | 83/94 [08:20<01:07,  6.12s/it]
Training Epoch: 8/100, batch 165/188 completed (loss: 0.024:  88%|████████▊ | 83/94 [08:20<01:07,  6.13s/it] 
Training Epoch: 8/100, batch 166/188 completed (loss: 0.0359:  88%|████████▊ | 83/94 [08:23<01:07,  6.12s/it]
Training Epoch: 8/100, batch 166/188 completed (loss: 0.1408:  88%|████████▊ | 83/94 [08:23<01:07,  6.12s/it]
Training Epoch: 8/100, batch 166/188 completed (loss: 0.0711:  88%|████████▊ | 83/94 [08:23<01:07,  6.12s/it]
Training Epoch: 8/100, batch 166/188 completed (loss: 0.3713:  88%|████████▊ | 83/94 [08:23<01:07,  6.13s/it]
Training Epoch: 8/100, batch 166/188 completed (loss: 0.1408:  89%|████████▉ | 84/94 [08:26<01:00,  6.09s/it]
Training Epoch: 8/100, batch 166/188 completed (loss: 0.0359:  89%|████████▉ | 84/94 [08:26<01:00,  6.09s/it]
Training Epoch: 8/100, batch 166/188 completed (loss: 0.3713:  89%|████████▉ | 84/94 [08:26<01:00,  6.09s/it]
Training Epoch: 8/100, batch 166/188 completed (loss: 0.0711:  89%|████████▉ | 84/94 [08:26<01:00,  6.09s/it]
Training Epoch: 8/100, batch 167/188 completed (loss: 0.3281:  89%|████████▉ | 84/94 [08:26<01:00,  6.09s/it]
Training Epoch: 8/100, batch 167/188 completed (loss: 0.1728:  89%|████████▉ | 84/94 [08:26<01:00,  6.09s/it]
Training Epoch: 8/100, batch 167/188 completed (loss: 0.1216:  89%|████████▉ | 84/94 [08:26<01:00,  6.09s/it]
Training Epoch: 8/100, batch 167/188 completed (loss: 0.0648:  89%|████████▉ | 84/94 [08:26<01:00,  6.09s/it]
Training Epoch: 8/100, batch 168/188 completed (loss: 0.133:  89%|████████▉ | 84/94 [08:29<01:00,  6.09s/it] 
Training Epoch: 8/100, batch 168/188 completed (loss: 0.3413:  89%|████████▉ | 84/94 [08:29<01:00,  6.09s/it]
Training Epoch: 8/100, batch 168/188 completed (loss: 0.1447:  89%|████████▉ | 84/94 [08:29<01:00,  6.09s/it]
Training Epoch: 8/100, batch 168/188 completed (loss: 0.1573:  89%|████████▉ | 84/94 [08:29<01:00,  6.09s/it]
Training Epoch: 8/100, batch 168/188 completed (loss: 0.1447:  90%|█████████ | 85/94 [08:32<00:54,  6.06s/it]
Training Epoch: 8/100, batch 168/188 completed (loss: 0.133:  90%|█████████ | 85/94 [08:32<00:54,  6.06s/it]
Training Epoch: 8/100, batch 168/188 completed (loss: 0.3413:  90%|█████████ | 85/94 [08:32<00:54,  6.06s/it]
Training Epoch: 8/100, batch 168/188 completed (loss: 0.1573:  90%|█████████ | 85/94 [08:32<00:54,  6.06s/it]
Training Epoch: 8/100, batch 169/188 completed (loss: 0.2267:  90%|█████████ | 85/94 [08:32<00:54,  6.06s/it]
Training Epoch: 8/100, batch 169/188 completed (loss: 0.1283:  90%|█████████ | 85/94 [08:32<00:54,  6.06s/it]
Training Epoch: 8/100, batch 169/188 completed (loss: 0.0597:  90%|█████████ | 85/94 [08:32<00:54,  6.06s/it]
Training Epoch: 8/100, batch 169/188 completed (loss: 0.2891:  90%|█████████ | 85/94 [08:32<00:54,  6.06s/it]
Training Epoch: 8/100, batch 170/188 completed (loss: 0.1645:  90%|█████████ | 85/94 [08:35<00:54,  6.06s/it]
Training Epoch: 8/100, batch 170/188 completed (loss: 0.3:  90%|█████████ | 85/94 [08:35<00:54,  6.06s/it]   
Training Epoch: 8/100, batch 170/188 completed (loss: 0.0334:  90%|█████████ | 85/94 [08:35<00:54,  6.06s/it]
Training Epoch: 8/100, batch 170/188 completed (loss: 0.1697:  90%|█████████ | 85/94 [08:35<00:54,  6.06s/it]
Training Epoch: 8/100, batch 170/188 completed (loss: 0.3:  91%|█████████▏| 86/94 [08:38<00:48,  6.09s/it]
Training Epoch: 8/100, batch 170/188 completed (loss: 0.0334:  91%|█████████▏| 86/94 [08:38<00:48,  6.10s/it]
Training Epoch: 8/100, batch 170/188 completed (loss: 0.1697:  91%|█████████▏| 86/94 [08:38<00:48,  6.10s/it]
Training Epoch: 8/100, batch 170/188 completed (loss: 0.1645:  91%|█████████▏| 86/94 [08:38<00:48,  6.10s/it]
Training Epoch: 8/100, batch 171/188 completed (loss: 0.0176:  91%|█████████▏| 86/94 [08:38<00:48,  6.10s/it]
Training Epoch: 8/100, batch 171/188 completed (loss: 0.091:  91%|█████████▏| 86/94 [08:38<00:48,  6.09s/it]
Training Epoch: 8/100, batch 171/188 completed (loss: 0.1433:  91%|█████████▏| 86/94 [08:38<00:48,  6.10s/it]
Training Epoch: 8/100, batch 171/188 completed (loss: 0.1537:  91%|█████████▏| 86/94 [08:38<00:48,  6.10s/it]
Training Epoch: 8/100, batch 172/188 completed (loss: 0.0728:  91%|█████████▏| 86/94 [08:41<00:48,  6.09s/it]
Training Epoch: 8/100, batch 172/188 completed (loss: 0.0733:  91%|█████████▏| 86/94 [08:41<00:48,  6.10s/it]
Training Epoch: 8/100, batch 172/188 completed (loss: 0.1079:  91%|█████████▏| 86/94 [08:41<00:48,  6.10s/it]
Training Epoch: 8/100, batch 172/188 completed (loss: 0.2273:  91%|█████████▏| 86/94 [08:41<00:48,  6.10s/it]
Training Epoch: 8/100, batch 172/188 completed (loss: 0.0733:  93%|█████████▎| 87/94 [08:44<00:42,  6.07s/it]
Training Epoch: 8/100, batch 172/188 completed (loss: 0.0728:  93%|█████████▎| 87/94 [08:44<00:42,  6.08s/it]
Training Epoch: 8/100, batch 172/188 completed (loss: 0.1079:  93%|█████████▎| 87/94 [08:44<00:42,  6.08s/it]
Training Epoch: 8/100, batch 172/188 completed (loss: 0.2273:  93%|█████████▎| 87/94 [08:44<00:42,  6.08s/it]
Training Epoch: 8/100, batch 173/188 completed (loss: 0.0078:  93%|█████████▎| 87/94 [08:44<00:42,  6.08s/it]
Training Epoch: 8/100, batch 173/188 completed (loss: 0.1145:  93%|█████████▎| 87/94 [08:44<00:42,  6.08s/it]
Training Epoch: 8/100, batch 173/188 completed (loss: 0.0541:  93%|█████████▎| 87/94 [08:44<00:42,  6.07s/it]
Training Epoch: 8/100, batch 173/188 completed (loss: 0.0414:  93%|█████████▎| 87/94 [08:44<00:42,  6.08s/it]
Training Epoch: 8/100, batch 174/188 completed (loss: 0.7082:  93%|█████████▎| 87/94 [08:47<00:42,  6.07s/it]
Training Epoch: 8/100, batch 174/188 completed (loss: 0.1769:  93%|█████████▎| 87/94 [08:47<00:42,  6.08s/it]
Training Epoch: 8/100, batch 174/188 completed (loss: 0.1023:  93%|█████████▎| 87/94 [08:47<00:42,  6.08s/it]
Training Epoch: 8/100, batch 174/188 completed (loss: 0.2797:  93%|█████████▎| 87/94 [08:47<00:42,  6.08s/it]
Training Epoch: 8/100, batch 174/188 completed (loss: 0.1769:  94%|█████████▎| 88/94 [08:50<00:36,  6.09s/it]
Training Epoch: 8/100, batch 174/188 completed (loss: 0.2797:  94%|█████████▎| 88/94 [08:50<00:36,  6.09s/it]
Training Epoch: 8/100, batch 174/188 completed (loss: 0.1023:  94%|█████████▎| 88/94 [08:50<00:36,  6.09s/it]
Training Epoch: 8/100, batch 174/188 completed (loss: 0.7082:  94%|█████████▎| 88/94 [08:50<00:36,  6.09s/it]
Training Epoch: 8/100, batch 175/188 completed (loss: 0.1161:  94%|█████████▎| 88/94 [08:50<00:36,  6.09s/it]
Training Epoch: 8/100, batch 175/188 completed (loss: 0.1261:  94%|█████████▎| 88/94 [08:50<00:36,  6.09s/it]
Training Epoch: 8/100, batch 175/188 completed (loss: 0.171:  94%|█████████▎| 88/94 [08:50<00:36,  6.09s/it] 
Training Epoch: 8/100, batch 175/188 completed (loss: 0.1068:  94%|█████████▎| 88/94 [08:50<00:36,  6.09s/it]
Training Epoch: 8/100, batch 176/188 completed (loss: 0.0521:  94%|█████████▎| 88/94 [08:53<00:36,  6.09s/it]
Training Epoch: 8/100, batch 176/188 completed (loss: 0.5807:  94%|█████████▎| 88/94 [08:53<00:36,  6.09s/it]
Training Epoch: 8/100, batch 176/188 completed (loss: 0.0335:  94%|█████████▎| 88/94 [08:53<00:36,  6.09s/it]
Training Epoch: 8/100, batch 176/188 completed (loss: 0.0357:  94%|█████████▎| 88/94 [08:53<00:36,  6.09s/it]
Training Epoch: 8/100, batch 176/188 completed (loss: 0.0521:  95%|█████████▍| 89/94 [08:56<00:30,  6.06s/it]
Training Epoch: 8/100, batch 176/188 completed (loss: 0.0335:  95%|█████████▍| 89/94 [08:56<00:30,  6.06s/it]
Training Epoch: 8/100, batch 176/188 completed (loss: 0.0357:  95%|█████████▍| 89/94 [08:56<00:30,  6.06s/it]
Training Epoch: 8/100, batch 176/188 completed (loss: 0.5807:  95%|█████████▍| 89/94 [08:56<00:30,  6.06s/it]
Training Epoch: 8/100, batch 177/188 completed (loss: 0.023:  95%|█████████▍| 89/94 [08:56<00:30,  6.06s/it] 
Training Epoch: 8/100, batch 177/188 completed (loss: 0.1322:  95%|█████████▍| 89/94 [08:56<00:30,  6.06s/it]
Training Epoch: 8/100, batch 177/188 completed (loss: 0.036:  95%|█████████▍| 89/94 [08:56<00:30,  6.06s/it] 
Training Epoch: 8/100, batch 177/188 completed (loss: 0.3661:  95%|█████████▍| 89/94 [08:56<00:30,  6.06s/it]
Training Epoch: 8/100, batch 178/188 completed (loss: 0.0489:  95%|█████████▍| 89/94 [09:00<00:30,  6.06s/it]
Training Epoch: 8/100, batch 178/188 completed (loss: 0.2684:  95%|█████████▍| 89/94 [09:00<00:30,  6.06s/it]
Training Epoch: 8/100, batch 178/188 completed (loss: 0.3829:  95%|█████████▍| 89/94 [09:00<00:30,  6.06s/it]
Training Epoch: 8/100, batch 178/188 completed (loss: 0.0632:  95%|█████████▍| 89/94 [09:00<00:30,  6.06s/it]
Training Epoch: 8/100, batch 178/188 completed (loss: 0.3829:  96%|█████████▌| 90/94 [09:03<00:24,  6.11s/it]
Training Epoch: 8/100, batch 178/188 completed (loss: 0.0632:  96%|█████████▌| 90/94 [09:03<00:24,  6.11s/it]
Training Epoch: 8/100, batch 178/188 completed (loss: 0.0489:  96%|█████████▌| 90/94 [09:03<00:24,  6.11s/it]
Training Epoch: 8/100, batch 178/188 completed (loss: 0.2684:  96%|█████████▌| 90/94 [09:03<00:24,  6.11s/it]
Training Epoch: 8/100, batch 179/188 completed (loss: 0.052:  96%|█████████▌| 90/94 [09:03<00:24,  6.11s/it] 
Training Epoch: 8/100, batch 179/188 completed (loss: 0.0366:  96%|█████████▌| 90/94 [09:03<00:24,  6.11s/it]
Training Epoch: 8/100, batch 179/188 completed (loss: 0.1612:  96%|█████████▌| 90/94 [09:03<00:24,  6.11s/it]
Training Epoch: 8/100, batch 179/188 completed (loss: 0.0442:  96%|█████████▌| 90/94 [09:03<00:24,  6.11s/it]
Training Epoch: 8/100, batch 180/188 completed (loss: 0.0157:  96%|█████████▌| 90/94 [09:06<00:24,  6.11s/it]
Training Epoch: 8/100, batch 180/188 completed (loss: 0.0498:  96%|█████████▌| 90/94 [09:06<00:24,  6.11s/it]
Training Epoch: 8/100, batch 180/188 completed (loss: 0.233:  96%|█████████▌| 90/94 [09:06<00:24,  6.11s/it]
Training Epoch: 8/100, batch 180/188 completed (loss: 0.1873:  96%|█████████▌| 90/94 [09:06<00:24,  6.11s/it]
Training Epoch: 8/100, batch 180/188 completed (loss: 0.0498:  97%|█████████▋| 91/94 [09:09<00:18,  6.12s/it]
Training Epoch: 8/100, batch 180/188 completed (loss: 0.233:  97%|█████████▋| 91/94 [09:09<00:18,  6.12s/it]
Training Epoch: 8/100, batch 180/188 completed (loss: 0.1873:  97%|█████████▋| 91/94 [09:09<00:18,  6.12s/it]
Training Epoch: 8/100, batch 180/188 completed (loss: 0.0157:  97%|█████████▋| 91/94 [09:09<00:18,  6.12s/it]
Training Epoch: 8/100, batch 181/188 completed (loss: 0.0745:  97%|█████████▋| 91/94 [09:09<00:18,  6.12s/it]
Training Epoch: 8/100, batch 181/188 completed (loss: 0.0856:  97%|█████████▋| 91/94 [09:09<00:18,  6.12s/it]
Training Epoch: 8/100, batch 181/188 completed (loss: 0.0276:  97%|█████████▋| 91/94 [09:09<00:18,  6.12s/it]
Training Epoch: 8/100, batch 181/188 completed (loss: 0.2208:  97%|█████████▋| 91/94 [09:09<00:18,  6.12s/it]
Training Epoch: 8/100, batch 182/188 completed (loss: 0.181:  97%|█████████▋| 91/94 [09:12<00:18,  6.12s/it] 
Training Epoch: 8/100, batch 182/188 completed (loss: 0.1076:  97%|█████████▋| 91/94 [09:12<00:18,  6.12s/it]
Training Epoch: 8/100, batch 182/188 completed (loss: 0.0476:  97%|█████████▋| 91/94 [09:12<00:18,  6.12s/it]
Training Epoch: 8/100, batch 182/188 completed (loss: 0.1165:  97%|█████████▋| 91/94 [09:12<00:18,  6.12s/it]
Training Epoch: 8/100, batch 182/188 completed (loss: 0.181:  98%|█████████▊| 92/94 [09:15<00:12,  6.03s/it]
Training Epoch: 8/100, batch 182/188 completed (loss: 0.0476:  98%|█████████▊| 92/94 [09:15<00:12,  6.03s/it]
Training Epoch: 8/100, batch 182/188 completed (loss: 0.1076:  98%|█████████▊| 92/94 [09:15<00:12,  6.03s/it]
Training Epoch: 8/100, batch 182/188 completed (loss: 0.1165:  98%|█████████▊| 92/94 [09:15<00:12,  6.03s/it]
Training Epoch: 8/100, batch 183/188 completed (loss: 0.095:  98%|█████████▊| 92/94 [09:15<00:12,  6.03s/it] 
Training Epoch: 8/100, batch 183/188 completed (loss: 0.0555:  98%|█████████▊| 92/94 [09:15<00:12,  6.03s/it]
Training Epoch: 8/100, batch 183/188 completed (loss: 0.0451:  98%|█████████▊| 92/94 [09:15<00:12,  6.03s/it]
Training Epoch: 8/100, batch 183/188 completed (loss: 0.067:  98%|█████████▊| 92/94 [09:15<00:12,  6.03s/it] 
Training Epoch: 8/100, batch 184/188 completed (loss: 0.0695:  98%|█████████▊| 92/94 [09:18<00:12,  6.03s/it]
Training Epoch: 8/100, batch 184/188 completed (loss: 0.1553:  98%|█████████▊| 92/94 [09:18<00:12,  6.03s/it]
Training Epoch: 8/100, batch 184/188 completed (loss: 0.0925:  98%|█████████▊| 92/94 [09:18<00:12,  6.03s/it]
Training Epoch: 8/100, batch 184/188 completed (loss: 0.0814:  98%|█████████▊| 92/94 [09:18<00:12,  6.03s/it]
Training Epoch: 8/100, batch 184/188 completed (loss: 0.0814:  99%|█████████▉| 93/94 [09:21<00:06,  6.11s/it]
Training Epoch: 8/100, batch 184/188 completed (loss: 0.0925:  99%|█████████▉| 93/94 [09:21<00:06,  6.11s/it]
Training Epoch: 8/100, batch 184/188 completed (loss: 0.0695:  99%|█████████▉| 93/94 [09:21<00:06,  6.11s/it]
Training Epoch: 8/100, batch 184/188 completed (loss: 0.1553:  99%|█████████▉| 93/94 [09:21<00:06,  6.11s/it]
Training Epoch: 8/100, batch 185/188 completed (loss: 0.0349:  99%|█████████▉| 93/94 [09:21<00:06,  6.11s/it]
Training Epoch: 8/100, batch 185/188 completed (loss: 0.0197:  99%|█████████▉| 93/94 [09:21<00:06,  6.11s/it]
Training Epoch: 8/100, batch 185/188 completed (loss: 0.1586:  99%|█████████▉| 93/94 [09:21<00:06,  6.11s/it]
Training Epoch: 8/100, batch 185/188 completed (loss: 0.0912:  99%|█████████▉| 93/94 [09:21<00:06,  6.11s/it]
Training Epoch: 8/100, batch 186/188 completed (loss: 0.037:  99%|█████████▉| 93/94 [09:24<00:06,  6.11s/it] 
Training Epoch: 8/100, batch 186/188 completed (loss: 0.0174:  99%|█████████▉| 93/94 [09:24<00:06,  6.11s/it]
Training Epoch: 8/100, batch 186/188 completed (loss: 0.0532:  99%|█████████▉| 93/94 [09:24<00:06,  6.11s/it]
Training Epoch: 8/100, batch 186/188 completed (loss: 0.0386:  99%|█████████▉| 93/94 [09:24<00:06,  6.11s/it]
Training Epoch: 8/100, batch 186/188 completed (loss: 0.0174: 100%|██████████| 94/94 [09:26<00:00,  5.81s/it]
Training Epoch: 8/100, batch 186/188 completed (loss: 0.037: 100%|██████████| 94/94 [09:26<00:00,  5.81s/it]
Training Epoch: 8/100, batch 186/188 completed (loss: 0.0386: 100%|██████████| 94/94 [09:26<00:00,  5.82s/it]
Training Epoch: 8/100, batch 186/188 completed (loss: 0.0532: 100%|██████████| 94/94 [09:26<00:00,  5.82s/it]
Training Epoch: 8/100, batch 187/188 completed (loss: 0.0705: 100%|██████████| 94/94 [09:26<00:00,  5.82s/it]
Training Epoch: 8/100, batch 187/188 completed (loss: 0.2194: 100%|██████████| 94/94 [09:26<00:00,  5.81s/it]
Training Epoch: 8/100, batch 187/188 completed (loss: 0.0073: 100%|██████████| 94/94 [09:26<00:00,  5.81s/it]
Training Epoch: 8/100, batch 187/188 completed (loss: 0.2194: 100%|██████████| 94/94 [09:26<00:00,  6.03s/it]

Training Epoch: 8/100, batch 187/188 completed (loss: 0.0705: 100%|██████████| 94/94 [09:26<00:00,  6.03s/it]

Training Epoch: 8/100, batch 187/188 completed (loss: 0.0073: 100%|██████████| 94/94 [09:26<00:00,  6.03s/it]

Training Epoch: 8/100, batch 187/188 completed (loss: 0.0907: 100%|██████████| 94/94 [09:26<00:00,  5.82s/it]
Training Epoch: 8/100, batch 187/188 completed (loss: 0.0907: 100%|██████████| 94/94 [09:27<00:00,  6.03s/it]
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/c10d_logger.py:83: UserWarning: barrier(): using the device under current context. You can specify `device_id` in `init_process_group` to mute this warning.
  return func(*args, **kwargs)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
saving model + train state.
Warning: failed to upload checkpoint_8 to HF: (Request ID: Root=1-6a1b7356-2fd12493300125802380fe2b;14661ea1-1626-4d74-8706-218a3a97a914)

403 Forbidden: You need to setup automatic credit recharge in order to upload more data. You can do so at /settings/billing..
Cannot access content at: https://huggingface.co/seyedparsa/thoughtformer-checkpoints.git/info/lfs/objects/batch.
Make sure your token has the correct permissions.
Warning: failed to upload train_state_8 to HF: (Request ID: Root=1-6a1b735a-37e34c1a4ad5a094752fe9b0;e13b3edd-7985-49e9-aadc-cf7998963d1d)

403 Forbidden: You need to setup automatic credit recharge in order to upload more data. You can do so at /settings/billing..
Cannot access content at: https://huggingface.co/seyedparsa/thoughtformer-checkpoints.git/info/lfs/objects/batch.
Make sure your token has the correct permissions.
Deleted old checkpoint: checkpoint_7
Deleted old checkpoint: train_state_7
[warn] HF upload failed for ckpt_8; kept locally, earlier ckpts deleted.

Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]
Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]
Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]eval loss 0.05685582306039961

Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]
[CORRECT] Example 0 (diff=3)
  Expected : 'Claude'
  Extracted: 'Claude'
  Generated: '### Claude'

[WRONG] Example 1 (diff=8)

Test Accuracy:   3%|▎         | 1/32 [00:07<03:39,  7.08s/it]  Expected : 'Guy'
  Extracted: 'Irma'
Test Accuracy:   3%|▎         | 1/32 [00:07<03:39,  7.08s/it]
  Generated: '### Irma'

Test accuracy: 0.88:   3%|▎         | 1/32 [00:07<03:39,  7.08s/it]
Test accuracy: 1.0:   3%|▎         | 1/32 [00:07<03:39,  7.08s/it]
Test Accuracy:   3%|▎         | 1/32 [00:07<03:39,  7.08s/it]
[WRONG] Example 2 (diff=10)
  Expected : 'Elvis'
  Extracted: 'Shannon'
  Generated: '### Shannon'
Test accuracy: 0.5:   3%|▎         | 1/32 [00:07<03:39,  7.08s/it]

[CORRECT] Example 3 (diff=8)
  Expected : 'Coral'
  Extracted: 'Coral'
  Generated: '### Coral'

[CORRECT] Example 4 (diff=1)
  Expected : 'Patricia'
  Extracted: 'Patricia'
  Generated: '### Patricia'

Test Accuracy:   3%|▎         | 1/32 [00:06<03:34,  6.92s/it]
Test accuracy: 0.38:   3%|▎         | 1/32 [00:06<03:34,  6.92s/it]
Test accuracy: 0.5:   6%|▋         | 2/32 [00:13<03:16,  6.54s/it]
Test accuracy: 0.88:   6%|▋         | 2/32 [00:13<03:16,  6.53s/it]
Test accuracy: 0.56:   6%|▋         | 2/32 [00:13<03:16,  6.54s/it]
Test accuracy: 1.0:   6%|▋         | 2/32 [00:13<03:16,  6.53s/it]
Test accuracy: 0.81:   6%|▋         | 2/32 [00:13<03:16,  6.53s/it]
Test accuracy: 0.94:   6%|▋         | 2/32 [00:13<03:16,  6.53s/it]
Test accuracy: 0.38:   6%|▋         | 2/32 [00:13<03:13,  6.46s/it]
Test accuracy: 0.5:   6%|▋         | 2/32 [00:13<03:13,  6.46s/it] 
Test accuracy: 0.81:   9%|▉         | 3/32 [00:20<03:15,  6.73s/it]
Test accuracy: 0.94:   9%|▉         | 3/32 [00:20<03:15,  6.73s/it]
Test accuracy: 0.71:   9%|▉         | 3/32 [00:20<03:15,  6.73s/it]
Test accuracy: 0.88:   9%|▉         | 3/32 [00:20<03:15,  6.73s/it]
Test accuracy: 0.5:   9%|▉         | 3/32 [00:20<03:14,  6.69s/it]
Test accuracy: 0.58:   9%|▉         | 3/32 [00:20<03:14,  6.69s/it]
Test accuracy: 0.56:   9%|▉         | 3/32 [00:20<03:15,  6.74s/it]
Test accuracy: 0.54:   9%|▉         | 3/32 [00:20<03:15,  6.74s/it]
Test accuracy: 0.88:  12%|█▎        | 4/32 [00:26<03:04,  6.57s/it]
Test accuracy: 0.71:  12%|█▎        | 4/32 [00:26<03:04,  6.57s/it]
Test accuracy: 0.54:  12%|█▎        | 4/32 [00:26<03:04,  6.57s/it]
Test accuracy: 0.78:  12%|█▎        | 4/32 [00:26<03:04,  6.57s/it]
Test accuracy: 0.66:  12%|█▎        | 4/32 [00:26<03:04,  6.57s/it]
Test accuracy: 0.56:  12%|█▎        | 4/32 [00:26<03:04,  6.57s/it]
Test accuracy: 0.58:  12%|█▎        | 4/32 [00:26<03:03,  6.55s/it]
Test accuracy: 0.62:  12%|█▎        | 4/32 [00:26<03:03,  6.55s/it]
Test accuracy: 0.56:  16%|█▌        | 5/32 [00:32<02:53,  6.42s/it]
Test accuracy: 0.78:  16%|█▌        | 5/32 [00:32<02:53,  6.42s/it]
Test accuracy: 0.66:  16%|█▌        | 5/32 [00:32<02:53,  6.42s/it]
Test accuracy: 0.73:  16%|█▌        | 5/32 [00:32<02:53,  6.42s/it]
Test accuracy: 0.57:  16%|█▌        | 5/32 [00:32<02:53,  6.42s/it]
Test accuracy: 0.68:  16%|█▌        | 5/32 [00:32<02:53,  6.42s/it]
Test accuracy: 0.62:  16%|█▌        | 5/32 [00:32<02:53,  6.41s/it]
Test accuracy: 0.68:  16%|█▌        | 5/32 [00:32<02:53,  6.41s/it]
Test accuracy: 0.57:  19%|█▉        | 6/32 [00:39<02:46,  6.40s/it]
Test accuracy: 0.54:  19%|█▉        | 6/32 [00:39<02:46,  6.40s/it]
Test accuracy: 0.73:  19%|█▉        | 6/32 [00:39<02:46,  6.40s/it]
Test accuracy: 0.71:  19%|█▉        | 6/32 [00:39<02:46,  6.40s/it]
Test accuracy: 0.68:  19%|█▉        | 6/32 [00:39<02:46,  6.40s/it]
Test accuracy: 0.71:  19%|█▉        | 6/32 [00:39<02:46,  6.40s/it]
Test accuracy: 0.68:  19%|█▉        | 6/32 [00:38<02:46,  6.39s/it]
Test accuracy: 0.73:  19%|█▉        | 6/32 [00:38<02:46,  6.39s/it]
Test accuracy: 0.54:  22%|██▏       | 7/32 [00:45<02:39,  6.38s/it]
Test accuracy: 0.71:  22%|██▏       | 7/32 [00:45<02:39,  6.38s/it]
Test accuracy: 0.71:  22%|██▏       | 7/32 [00:45<02:39,  6.38s/it]
Test accuracy: 0.57:  22%|██▏       | 7/32 [00:45<02:39,  6.38s/it]
Test accuracy: 0.64:  22%|██▏       | 7/32 [00:45<02:39,  6.38s/it]
Test accuracy: 0.66:  22%|██▏       | 7/32 [00:45<02:39,  6.38s/it]
Test accuracy: 0.73:  22%|██▏       | 7/32 [00:45<02:39,  6.38s/it]
Test accuracy: 0.71:  22%|██▏       | 7/32 [00:45<02:39,  6.38s/it]
Test accuracy: 0.64:  25%|██▌       | 8/32 [00:51<02:31,  6.31s/it]
Test accuracy: 0.67:  25%|██▌       | 8/32 [00:51<02:31,  6.31s/it]
Test accuracy: 0.71:  25%|██▌       | 8/32 [00:51<02:31,  6.31s/it]
Test accuracy: 0.57:  25%|██▌       | 8/32 [00:51<02:31,  6.31s/it]
Test accuracy: 0.66:  25%|██▌       | 8/32 [00:51<02:31,  6.31s/it]
Test accuracy: 0.69:  25%|██▌       | 8/32 [00:51<02:31,  6.31s/it]
Test accuracy: 0.56:  25%|██▌       | 8/32 [00:51<02:31,  6.31s/it]
Test accuracy: 0.66:  25%|██▌       | 8/32 [00:51<02:31,  6.31s/it]
Test accuracy: 0.56:  28%|██▊       | 9/32 [00:58<02:30,  6.56s/it]
Test accuracy: 0.67:  28%|██▊       | 9/32 [00:58<02:30,  6.56s/it]
Test accuracy: 0.57:  28%|██▊       | 9/32 [00:58<02:30,  6.56s/it]
Test accuracy: 0.69:  28%|██▊       | 9/32 [00:58<02:30,  6.56s/it]
Test accuracy: 0.69:  28%|██▊       | 9/32 [00:58<02:30,  6.56s/it]
Test accuracy: 0.68:  28%|██▊       | 9/32 [00:58<02:30,  6.56s/it]
Test accuracy: 0.66:  28%|██▊       | 9/32 [00:58<02:31,  6.57s/it]
Test accuracy: 0.67:  28%|██▊       | 9/32 [00:58<02:31,  6.57s/it]
Test accuracy: 0.57:  31%|███▏      | 10/32 [01:04<02:21,  6.44s/it]
Test accuracy: 0.67:  31%|███▏      | 10/32 [01:04<02:21,  6.44s/it]
Test accuracy: 0.55:  31%|███▏      | 10/32 [01:04<02:21,  6.44s/it]
Test accuracy: 0.68:  31%|███▏      | 10/32 [01:04<02:21,  6.44s/it]
Test accuracy: 0.65:  31%|███▏      | 10/32 [01:04<02:21,  6.44s/it]
Test accuracy: 0.69:  31%|███▏      | 10/32 [01:04<02:21,  6.44s/it]
Test accuracy: 0.66:  31%|███▏      | 10/32 [01:04<02:21,  6.44s/it]
Test accuracy: 0.7:  31%|███▏      | 10/32 [01:04<02:21,  6.44s/it] 
Test accuracy: 0.55:  34%|███▍      | 11/32 [01:11<02:14,  6.41s/it]
Test accuracy: 0.65:  34%|███▍      | 11/32 [01:11<02:14,  6.41s/it]
Test accuracy: 0.7:  34%|███▍      | 11/32 [01:11<02:14,  6.41s/it]
Test accuracy: 0.57:  34%|███▍      | 11/32 [01:11<02:14,  6.41s/it]
Test accuracy: 0.66:  34%|███▍      | 11/32 [01:11<02:14,  6.41s/it]
Test accuracy: 0.68:  34%|███▍      | 11/32 [01:11<02:14,  6.41s/it]
Test accuracy: 0.66:  34%|███▍      | 11/32 [01:10<02:14,  6.41s/it]
Test accuracy: 0.65:  34%|███▍      | 11/32 [01:10<02:14,  6.41s/it]
Test accuracy: 0.66:  38%|███▊      | 12/32 [01:17<02:07,  6.38s/it]
Test accuracy: 0.68:  38%|███▊      | 12/32 [01:17<02:07,  6.38s/it]
Test accuracy: 0.57:  38%|███▊      | 12/32 [01:17<02:07,  6.38s/it]
Test accuracy: 0.66:  38%|███▊      | 12/32 [01:17<02:07,  6.38s/it]
Test accuracy: 0.67:  38%|███▊      | 12/32 [01:17<02:07,  6.38s/it]
Test accuracy: 0.54:  38%|███▊      | 12/32 [01:17<02:07,  6.38s/it]
Test accuracy: 0.65:  38%|███▊      | 12/32 [01:17<02:07,  6.38s/it]
Test accuracy: 0.65:  38%|███▊      | 12/32 [01:17<02:07,  6.38s/it]
Test accuracy: 0.66:  41%|████      | 13/32 [01:23<01:59,  6.31s/it]
Test accuracy: 0.67:  41%|████      | 13/32 [01:23<01:59,  6.31s/it]
Test accuracy: 0.67:  41%|████      | 13/32 [01:23<01:59,  6.31s/it]
Test accuracy: 0.54:  41%|████      | 13/32 [01:23<01:59,  6.31s/it]
Test accuracy: 0.67:  41%|████      | 13/32 [01:23<01:59,  6.31s/it]
Test accuracy: 0.57:  41%|████      | 13/32 [01:23<01:59,  6.31s/it]
Test accuracy: 0.65:  41%|████      | 13/32 [01:23<01:59,  6.31s/it]
Test accuracy: 0.63:  41%|████      | 13/32 [01:23<01:59,  6.31s/it]
Test accuracy: 0.57:  44%|████▍     | 14/32 [01:29<01:53,  6.32s/it]
Test accuracy: 0.67:  44%|████▍     | 14/32 [01:29<01:53,  6.32s/it]
Test accuracy: 0.67:  44%|████▍     | 14/32 [01:29<01:53,  6.32s/it]
Test accuracy: 0.63:  44%|████▍     | 14/32 [01:29<01:53,  6.32s/it]
Test accuracy: 0.68:  44%|████▍     | 14/32 [01:29<01:53,  6.32s/it]
Test accuracy: 0.59:  44%|████▍     | 14/32 [01:29<01:53,  6.32s/it]
Test accuracy: 0.68:  44%|████▍     | 14/32 [01:29<01:53,  6.32s/it]
Test accuracy: 0.64:  44%|████▍     | 14/32 [01:29<01:53,  6.32s/it]
Test accuracy: 0.59:  47%|████▋     | 15/32 [01:36<01:47,  6.32s/it]
Test accuracy: 0.68:  47%|████▋     | 15/32 [01:36<01:47,  6.32s/it]
Test accuracy: 0.56:  47%|████▋     | 15/32 [01:36<01:47,  6.32s/it]
Test accuracy: 0.68:  47%|████▋     | 15/32 [01:36<01:47,  6.32s/it]
Test accuracy: 0.64:  47%|████▋     | 15/32 [01:36<01:47,  6.32s/it]
Test accuracy: 0.68:  47%|████▋     | 15/32 [01:36<01:47,  6.32s/it]
Test accuracy: 0.68:  47%|████▋     | 15/32 [01:36<01:47,  6.32s/it]
Test accuracy: 0.63:  47%|████▋     | 15/32 [01:36<01:47,  6.32s/it]
Test accuracy: 0.56:  50%|█████     | 16/32 [01:42<01:40,  6.27s/it]
Test accuracy: 0.68:  50%|█████     | 16/32 [01:42<01:40,  6.27s/it]
Test accuracy: 0.68:  50%|█████     | 16/32 [01:42<01:40,  6.27s/it]
Test accuracy: 0.55:  50%|█████     | 16/32 [01:42<01:40,  6.27s/it]
Test accuracy: 0.66:  50%|█████     | 16/32 [01:42<01:40,  6.27s/it]
Test accuracy: 0.68:  50%|█████     | 16/32 [01:42<01:40,  6.27s/it]
Test accuracy: 0.63:  50%|█████     | 16/32 [01:42<01:40,  6.27s/it]
Test accuracy: 0.64:  50%|█████     | 16/32 [01:42<01:40,  6.27s/it]
Test accuracy: 0.64:  53%|█████▎    | 17/32 [01:48<01:34,  6.29s/it]
Test accuracy: 0.66:  53%|█████▎    | 17/32 [01:48<01:34,  6.29s/it]
Test accuracy: 0.68:  53%|█████▎    | 17/32 [01:48<01:34,  6.29s/it]
Test accuracy: 0.65:  53%|█████▎    | 17/32 [01:48<01:34,  6.29s/it]
Test accuracy: 0.55:  53%|█████▎    | 17/32 [01:48<01:34,  6.29s/it]
Test accuracy: 0.66:  53%|█████▎    | 17/32 [01:48<01:34,  6.29s/it]
Test accuracy: 0.68:  53%|█████▎    | 17/32 [01:48<01:34,  6.29s/it]
Test accuracy: 0.57:  53%|█████▎    | 17/32 [01:48<01:34,  6.29s/it]
Test accuracy: 0.57:  56%|█████▋    | 18/32 [01:55<01:28,  6.30s/it]
Test accuracy: 0.65:  56%|█████▋    | 18/32 [01:54<01:28,  6.30s/it]
Test accuracy: 0.66:  56%|█████▋    | 18/32 [01:55<01:28,  6.30s/it]
Test accuracy: 0.68:  56%|█████▋    | 18/32 [01:55<01:28,  6.30s/it]
Test accuracy: 0.67:  56%|█████▋    | 18/32 [01:55<01:28,  6.30s/it]
Test accuracy: 0.59:  56%|█████▋    | 18/32 [01:55<01:28,  6.30s/it]
Test accuracy: 0.68:  56%|█████▋    | 18/32 [01:55<01:28,  6.30s/it]
Test accuracy: 0.65:  56%|█████▋    | 18/32 [01:54<01:28,  6.30s/it]
Test accuracy: 0.67:  59%|█████▉    | 19/32 [02:01<01:21,  6.26s/it]
Test accuracy: 0.65:  59%|█████▉    | 19/32 [02:01<01:21,  6.26s/it]
Test accuracy: 0.66:  59%|█████▉    | 19/32 [02:01<01:21,  6.26s/it]
Test accuracy: 0.68:  59%|█████▉    | 19/32 [02:01<01:21,  6.26s/it]
Test accuracy: 0.59:  59%|█████▉    | 19/32 [02:01<01:21,  6.26s/it]
Test accuracy: 0.65:  59%|█████▉    | 19/32 [02:01<01:21,  6.26s/it]
Test accuracy: 0.67:  59%|█████▉    | 19/32 [02:01<01:21,  6.26s/it]
Test accuracy: 0.59:  59%|█████▉    | 19/32 [02:01<01:21,  6.26s/it]
Test accuracy: 0.59:  62%|██████▎   | 20/32 [02:07<01:15,  6.28s/it]
Test accuracy: 0.66:  62%|██████▎   | 20/32 [02:07<01:15,  6.28s/it]
Test accuracy: 0.67:  62%|██████▎   | 20/32 [02:07<01:15,  6.28s/it]
Test accuracy: 0.57:  62%|██████▎   | 20/32 [02:07<01:15,  6.28s/it]
Test accuracy: 0.67:  62%|██████▎   | 20/32 [02:07<01:15,  6.28s/it]
Test accuracy: 0.66:  62%|██████▎   | 20/32 [02:07<01:15,  6.28s/it]
Test accuracy: 0.65:  62%|██████▎   | 20/32 [02:07<01:15,  6.28s/it]
Test accuracy: 0.66:  62%|██████▎   | 20/32 [02:07<01:15,  6.28s/it]
Test accuracy: 0.66:  66%|██████▌   | 21/32 [02:14<01:11,  6.50s/it]
Test accuracy: 0.66:  66%|██████▌   | 21/32 [02:14<01:11,  6.50s/it]
Test accuracy: 0.67:  66%|██████▌   | 21/32 [02:14<01:11,  6.50s/it]
Test accuracy: 0.64:  66%|██████▌   | 21/32 [02:14<01:11,  6.50s/it]
Test accuracy: 0.68:  66%|██████▌   | 21/32 [02:14<01:11,  6.50s/it]
Test accuracy: 0.65:  66%|██████▌   | 21/32 [02:14<01:11,  6.50s/it]
Test accuracy: 0.57:  66%|██████▌   | 21/32 [02:14<01:11,  6.51s/it]
Test accuracy: 0.59:  66%|██████▌   | 21/32 [02:14<01:11,  6.51s/it]
Test accuracy: 0.59:  69%|██████▉   | 22/32 [02:20<01:03,  6.40s/it]
Test accuracy: 0.57:  69%|██████▉   | 22/32 [02:20<01:03,  6.40s/it]
Test accuracy: 0.65:  69%|██████▉   | 22/32 [02:20<01:04,  6.40s/it]
Test accuracy: 0.68:  69%|██████▉   | 22/32 [02:20<01:04,  6.40s/it]
Test accuracy: 0.64:  69%|██████▉   | 22/32 [02:20<01:04,  6.40s/it]
Test accuracy: 0.65:  69%|██████▉   | 22/32 [02:20<01:04,  6.40s/it]
Test accuracy: 0.68:  69%|██████▉   | 22/32 [02:20<01:04,  6.40s/it]
Test accuracy: 0.65:  69%|██████▉   | 22/32 [02:20<01:04,  6.40s/it]
Test accuracy: 0.57:  72%|███████▏  | 23/32 [02:27<00:57,  6.35s/it]
Test accuracy: 0.68:  72%|███████▏  | 23/32 [02:27<00:57,  6.35s/it]
Test accuracy: 0.65:  72%|███████▏  | 23/32 [02:27<00:57,  6.35s/it]
Test accuracy: 0.57:  72%|███████▏  | 23/32 [02:27<00:57,  6.35s/it]
Test accuracy: 0.68:  72%|███████▏  | 23/32 [02:27<00:57,  6.35s/it]
Test accuracy: 0.65:  72%|███████▏  | 23/32 [02:27<00:57,  6.35s/it]
Test accuracy: 0.65:  72%|███████▏  | 23/32 [02:26<00:57,  6.35s/it]
Test accuracy: 0.65:  72%|███████▏  | 23/32 [02:26<00:57,  6.35s/it]
Test accuracy: 0.57:  75%|███████▌  | 24/32 [02:33<00:52,  6.53s/it]
Test accuracy: 0.56:  75%|███████▌  | 24/32 [02:33<00:52,  6.53s/it]
Test accuracy: 0.68:  75%|███████▌  | 24/32 [02:33<00:52,  6.53s/it]
Test accuracy: 0.65:  75%|███████▌  | 24/32 [02:33<00:52,  6.53s/it]
Test accuracy: 0.68:  75%|███████▌  | 24/32 [02:33<00:52,  6.53s/it]
Test accuracy: 0.65:  75%|███████▌  | 24/32 [02:33<00:52,  6.53s/it]
Test accuracy: 0.65:  75%|███████▌  | 24/32 [02:33<00:52,  6.54s/it]
Test accuracy: 0.65:  75%|███████▌  | 24/32 [02:33<00:52,  6.54s/it]
Test accuracy: 0.68:  78%|███████▊  | 25/32 [02:40<00:45,  6.48s/it]
Test accuracy: 0.68:  78%|███████▊  | 25/32 [02:40<00:45,  6.48s/it]
Test accuracy: 0.65:  78%|███████▊  | 25/32 [02:40<00:45,  6.48s/it]
Test accuracy: 0.65:  78%|███████▊  | 25/32 [02:40<00:45,  6.48s/it]
Test accuracy: 0.63:  78%|███████▊  | 25/32 [02:40<00:45,  6.48s/it]
Test accuracy: 0.64:  78%|███████▊  | 25/32 [02:40<00:45,  6.48s/it]
Test accuracy: 0.56:  78%|███████▊  | 25/32 [02:40<00:45,  6.48s/it]
Test accuracy: 0.56:  78%|███████▊  | 25/32 [02:40<00:45,  6.48s/it]
Test accuracy: 0.56:  81%|████████▏ | 26/32 [02:46<00:38,  6.38s/it]
Test accuracy: 0.68:  81%|████████▏ | 26/32 [02:46<00:38,  6.38s/it]
Test accuracy: 0.63:  81%|████████▏ | 26/32 [02:46<00:38,  6.38s/it]
Test accuracy: 0.64:  81%|████████▏ | 26/32 [02:46<00:38,  6.38s/it]
Test accuracy: 0.57:  81%|████████▏ | 26/32 [02:46<00:38,  6.38s/it]
Test accuracy: 0.68:  81%|████████▏ | 26/32 [02:46<00:38,  6.38s/it]
Test accuracy: 0.63:  81%|████████▏ | 26/32 [02:46<00:38,  6.38s/it]
Test accuracy: 0.64:  81%|████████▏ | 26/32 [02:46<00:38,  6.38s/it]
Test accuracy: 0.68:  84%|████████▍ | 27/32 [02:52<00:31,  6.36s/it]
Test accuracy: 0.64:  84%|████████▍ | 27/32 [02:52<00:31,  6.36s/it]
Test accuracy: 0.63:  84%|████████▍ | 27/32 [02:52<00:31,  6.36s/it]
Test accuracy: 0.57:  84%|████████▍ | 27/32 [02:52<00:31,  6.36s/it]
Test accuracy: 0.69:  84%|████████▍ | 27/32 [02:52<00:31,  6.36s/it]
Test accuracy: 0.63:  84%|████████▍ | 27/32 [02:52<00:31,  6.36s/it]
Test accuracy: 0.56:  84%|████████▍ | 27/32 [02:52<00:31,  6.36s/it]
Test accuracy: 0.64:  84%|████████▍ | 27/32 [02:52<00:31,  6.36s/it]
Test accuracy: 0.63:  88%|████████▊ | 28/32 [02:59<00:26,  6.55s/it]
Test accuracy: 0.64:  88%|████████▊ | 28/32 [02:59<00:26,  6.55s/it]
Test accuracy: 0.69:  88%|████████▊ | 28/32 [02:59<00:26,  6.55s/it]
Test accuracy: 0.62:  88%|████████▊ | 28/32 [02:59<00:26,  6.55s/it]
Test accuracy: 0.69:  88%|████████▊ | 28/32 [02:59<00:26,  6.55s/it]
Test accuracy: 0.64:  88%|████████▊ | 28/32 [02:59<00:26,  6.55s/it]
Test accuracy: 0.56:  88%|████████▊ | 28/32 [02:59<00:26,  6.56s/it]
Test accuracy: 0.57:  88%|████████▊ | 28/32 [02:59<00:26,  6.56s/it]
Test accuracy: 0.69:  91%|█████████ | 29/32 [03:06<00:19,  6.49s/it]
Test accuracy: 0.68:  91%|█████████ | 29/32 [03:06<00:19,  6.49s/it]
Test accuracy: 0.62:  91%|█████████ | 29/32 [03:06<00:19,  6.49s/it]
Test accuracy: 0.63:  91%|█████████ | 29/32 [03:06<00:19,  6.49s/it]
Test accuracy: 0.57:  91%|█████████ | 29/32 [03:06<00:19,  6.49s/it]
Test accuracy: 0.56:  91%|█████████ | 29/32 [03:06<00:19,  6.49s/it]
Test accuracy: 0.64:  91%|█████████ | 29/32 [03:05<00:19,  6.49s/it]
Test accuracy: 0.65:  91%|█████████ | 29/32 [03:05<00:19,  6.49s/it]
Test accuracy: 0.56:  94%|█████████▍| 30/32 [03:12<00:12,  6.38s/it]
Test accuracy: 0.63:  94%|█████████▍| 30/32 [03:12<00:12,  6.38s/it]
Test accuracy: 0.68:  94%|█████████▍| 30/32 [03:12<00:12,  6.38s/it]
Test accuracy: 0.55:  94%|█████████▍| 30/32 [03:12<00:12,  6.38s/it]
Test accuracy: 0.65:  94%|█████████▍| 30/32 [03:12<00:12,  6.38s/it]
Test accuracy: 0.62:  94%|█████████▍| 30/32 [03:12<00:12,  6.38s/it]
Test accuracy: 0.68:  94%|█████████▍| 30/32 [03:12<00:12,  6.38s/it]
Test accuracy: 0.64:  94%|█████████▍| 30/32 [03:12<00:12,  6.38s/it]
Test accuracy: 0.68:  97%|█████████▋| 31/32 [03:19<00:06,  6.57s/it]
Test accuracy: 0.62:  97%|█████████▋| 31/32 [03:19<00:06,  6.57s/it]
Test accuracy: 0.68:  97%|█████████▋| 31/32 [03:19<00:06,  6.57s/it]
Test accuracy: 0.63:  97%|█████████▋| 31/32 [03:19<00:06,  6.57s/it]
Test accuracy: 0.64:  97%|█████████▋| 31/32 [03:19<00:06,  6.57s/it]
Test accuracy: 0.65:  97%|█████████▋| 31/32 [03:19<00:06,  6.57s/it]
Test accuracy: 0.55:  97%|█████████▋| 31/32 [03:19<00:06,  6.58s/it]
Test accuracy: 0.56:  97%|█████████▋| 31/32 [03:19<00:06,  6.58s/it]
Test accuracy: 0.65: 100%|██████████| 32/32 [03:21<00:00,  5.30s/it]
Test accuracy: 0.65: 100%|██████████| 32/32 [03:21<00:00,  5.30s/it]
Test accuracy: 0.56: 100%|██████████| 32/32 [03:21<00:00,  5.30s/it]
Test accuracy: 0.55: 100%|██████████| 32/32 [03:21<00:00,  5.30s/it]
Test accuracy: 0.63: 100%|██████████| 32/32 [03:21<00:00,  5.31s/it]
Test accuracy: 0.68: 100%|██████████| 32/32 [03:21<00:00,  5.31s/it]
Test accuracy: 0.63: 100%|██████████| 32/32 [03:21<00:00,  5.31s/it]
Test accuracy: 0.68: 100%|██████████| 32/32 [03:21<00:00,  5.31s/it]
Test accuracy: 0.55: 100%|██████████| 32/32 [03:21<00:00,  6.30s/it]
Device 2: Cor=138, CoT=0, Total=250

Test accuracy: 0.68: 100%|██████████| 32/32 [03:21<00:00,  6.30s/it]
Device 3: Cor=170, CoT=0, Total=250

Test accuracy: 0.63: 100%|██████████| 32/32 [03:21<00:00,  6.30s/it]
Device 1: Cor=157, CoT=0, Total=250

Test accuracy: 0.65: 100%|██████████| 32/32 [03:21<00:00,  6.30s/it]
Device 0: Cor=163, CoT=0, Total=250

============================================================
EVALUATION RESULTS (Epoch 7)
============================================================
Config: data_curriculum=ON, thought_curriculum=OFF
thought_stage=10, data_stage=3, n_latent_tokens=10
Training on difficulties: 0-3

Overall accuracy: 628 / 1000 = 0.6280

Per-difficulty accuracy:
  diff=1: 97/100 = 0.9700 (trained)
  diff=2: 97/100 = 0.9700 (trained)
  diff=3: 90/100 = 0.9000 (trained)
  diff=4: 49/100 = 0.4900 (untrained)
  diff=5: 37/100 = 0.3700 (untrained)
  diff=6: 55/100 = 0.5500 (untrained)
  diff=7: 53/100 = 0.5300 (untrained)
  diff=8: 48/100 = 0.4800 (untrained)
  diff=9: 54/100 = 0.5400 (untrained)
  diff=10: 48/100 = 0.4800 (untrained)

Trained (diff 0-3): 284/300 = 0.9467
============================================================


*** ADAPTIVE STAGING: acc_trained=0.9467, acc_diff3=0.9000 >= 0.9 ***
*** Advancing to stage 4 ***

Warning: failed to upload checkpoint_8 to HF: (Request ID: Root=1-6a1b7435-3c6484f25acda7746a774627;cee7dfce-1dd7-453b-be6d-d03af3214ad2)

403 Forbidden: You need to setup automatic credit recharge in order to upload more data. You can do so at /settings/billing..
Cannot access content at: https://huggingface.co/seyedparsa/thoughtformer-checkpoints.git/info/lfs/objects/batch.
Make sure your token has the correct permissions.

============================================================
EPOCH 8/100 (thought_stage=10, data_stage=4)
============================================================
  thought_stage=10, c_thought=1, max_difficulty=4

Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):  16%|█▌        | 156/1000 [00:00<00:00, 989.01 examples/s]
Map (num_proc=4):  16%|█▌        | 156/1000 [00:00<00:00, 962.36 examples/s]
Map (num_proc=4):  16%|█▌        | 156/1000 [00:00<00:00, 951.73 examples/s]
Map (num_proc=4):  81%|████████▏ | 813/1000 [00:00<00:00, 3485.33 examples/s]
Map (num_proc=4):  81%|████████  | 811/1000 [00:00<00:00, 3411.03 examples/s]
Map (num_proc=4):  81%|████████  | 812/1000 [00:00<00:00, 3388.59 examples/s]
Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2557.72 examples/s]

Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2499.47 examples/s]

Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2027.82 examples/s]

Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Map (num_proc=4):  16%|█▌        | 156/1000 [00:00<00:01, 503.55 examples/s]
Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Map (num_proc=4):  81%|████████▏ | 814/1000 [00:00<00:00, 2350.89 examples/s]
Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:04, 1994.21 examples/s]
Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:05, 1783.53 examples/s]
Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 1246.34 examples/s]

Filter (num_proc=4):  20%|██        | 2000/10000 [00:00<00:02, 3730.54 examples/s]
Filter (num_proc=4):  20%|██        | 2000/10000 [00:00<00:02, 3347.22 examples/s]
Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:05, 1797.81 examples/s]
Filter (num_proc=4):  30%|███       | 3000/10000 [00:00<00:01, 5076.93 examples/s]
Filter (num_proc=4):  20%|██        | 2000/10000 [00:00<00:02, 3407.80 examples/s]
Filter (num_proc=4):  30%|███       | 3000/10000 [00:00<00:01, 4572.26 examples/s]
Filter (num_proc=4):  60%|██████    | 6000/10000 [00:00<00:00, 10498.94 examples/s]
Filter (num_proc=4):  30%|███       | 3000/10000 [00:00<00:01, 4825.34 examples/s]
Filter (num_proc=4):  50%|█████     | 5000/10000 [00:00<00:00, 7781.83 examples/s]
Filter (num_proc=4):  75%|███████▌  | 7500/10000 [00:00<00:00, 11313.60 examples/s]
Filter (num_proc=4):  45%|████▌     | 4500/10000 [00:00<00:00, 6961.24 examples/s]
Filter (num_proc=4):  70%|███████   | 7000/10000 [00:01<00:00, 10181.54 examples/s]
Filter (num_proc=4):  95%|█████████▌| 9500/10000 [00:01<00:00, 12866.70 examples/s]
Filter (num_proc=4):  70%|███████   | 7000/10000 [00:01<00:00, 10660.80 examples/s]
Filter (num_proc=4):  90%|█████████ | 9000/10000 [00:01<00:00, 11916.75 examples/s]
Filter (num_proc=4):  85%|████████▌ | 8500/10000 [00:01<00:00, 11159.44 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 7625.92 examples/s]

Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 11194.17 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 7031.39 examples/s]

Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 7067.68 examples/s] 

Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:06, 1348.61 examples/s]
Filter (num_proc=4):  20%|██        | 2000/10000 [00:00<00:03, 2539.51 examples/s]
Filter (num_proc=4):  40%|████      | 4000/10000 [00:01<00:01, 5155.85 examples/s]
Filter (num_proc=4):  65%|██████▌   | 6500/10000 [00:01<00:00, 8141.16 examples/s]
Filter (num_proc=4):  85%|████████▌ | 8500/10000 [00:01<00:00, 8846.97 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 9206.13 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 5422.10 examples/s]

Map (num_proc=4):   0%|          | 0/4000 [00:00<?, ? examples/s]
Map (num_proc=4):   3%|▎         | 132/4000 [00:01<00:33, 114.13 examples/s]
Map (num_proc=4):  13%|█▎        | 537/4000 [00:01<00:06, 513.92 examples/s]
Map (num_proc=4):  20%|██        | 804/4000 [00:01<00:04, 791.20 examples/s]
Map (num_proc=4):  27%|██▋       | 1070/4000 [00:01<00:03, 832.45 examples/s]
Map (num_proc=4):  35%|███▍      | 1399/4000 [00:01<00:02, 1190.16 examples/s]
Map (num_proc=4):  41%|████      | 1643/4000 [00:02<00:01, 1203.68 examples/s]
Map (num_proc=4):  51%|█████     | 2041/4000 [00:02<00:01, 1317.42 examples/s]
Map (num_proc=4):  63%|██████▎   | 2537/4000 [00:02<00:00, 1688.26 examples/s]
Map (num_proc=4):  71%|███████   | 2844/4000 [00:02<00:00, 1834.76 examples/s]
Map (num_proc=4):  78%|███████▊  | 3113/4000 [00:02<00:00, 1543.18 examples/s]
Map (num_proc=4):  88%|████████▊ | 3537/4000 [00:03<00:00, 1786.11 examples/s]
Map (num_proc=4):  98%|█████████▊| 3900/4000 [00:03<00:00, 1982.63 examples/s]
Map (num_proc=4): 100%|██████████| 4000/4000 [00:03<00:00, 1005.74 examples/s]

Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 1450.25 examples/s]
Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 1446.99 examples/s]
Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 1371.93 examples/s]
Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2811.18 examples/s]

Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2784.73 examples/s]

Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2700.15 examples/s]

Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:01, 722.94 examples/s]
Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 1378.57 examples/s]

Map (num_proc=4):   0%|          | 0/400 [00:00<?, ? examples/s]
Map (num_proc=4):  25%|██▌       | 100/400 [00:00<00:01, 249.61 examples/s]
Map (num_proc=4):  75%|███████▌  | 300/400 [00:00<00:00, 630.87 examples/s]
Map (num_proc=4): 100%|██████████| 400/400 [00:00<00:00, 419.01 examples/s]

Training Epoch: 9:   0%|          | 0/125 [00:00<?, ?it/s]
Training Epoch: 9:   0%|          | 0/125 [00:00<?, ?it/s]
Training Epoch: 9:   0%|          | 0/125 [00:00<?, ?it/s]
Training Epoch: 9:   0%|          | 0/125 [00:00<?, ?it/s]logging training data
rank 2 | gpu 2 | alloc=4.20 GB | reserved=47.66 GB | max=41.98 GB
rank 3 | gpu 3 | alloc=4.20 GB | reserved=51.39 GB | max=41.99 GB
rank 1 | gpu 1 | alloc=4.20 GB | reserved=55.12 GB | max=41.98 GB
rank 0 | gpu 0 | alloc=4.20 GB | reserved=51.39 GB | max=41.99 GB

Training Epoch: 9/100, batch 0/250 completed (loss: 0.1891:   0%|          | 0/125 [00:03<?, ?it/s]
Training Epoch: 9/100, batch 0/250 completed (loss: 0.0558:   0%|          | 0/125 [00:03<?, ?it/s]
Training Epoch: 9/100, batch 0/250 completed (loss: 0.0391:   0%|          | 0/125 [00:03<?, ?it/s]
Training Epoch: 9/100, batch 0/250 completed (loss: 0.0245:   0%|          | 0/125 [00:03<?, ?it/s]
Training Epoch: 9/100, batch 0/250 completed (loss: 0.1891:   1%|          | 1/125 [00:06<12:48,  6.20s/it]
Training Epoch: 9/100, batch 0/250 completed (loss: 0.0391:   1%|          | 1/125 [00:06<12:50,  6.21s/it]
Training Epoch: 9/100, batch 0/250 completed (loss: 0.0245:   1%|          | 1/125 [00:06<12:53,  6.24s/it]
Training Epoch: 9/100, batch 0/250 completed (loss: 0.0558:   1%|          | 1/125 [00:06<12:51,  6.22s/it]
Training Epoch: 9/100, batch 1/250 completed (loss: 0.0576:   1%|          | 1/125 [00:06<12:50,  6.21s/it]
Training Epoch: 9/100, batch 1/250 completed (loss: 0.0752:   1%|          | 1/125 [00:06<12:48,  6.20s/it]
Training Epoch: 9/100, batch 1/250 completed (loss: 0.0586:   1%|          | 1/125 [00:06<12:51,  6.22s/it]
Training Epoch: 9/100, batch 1/250 completed (loss: 0.0273:   1%|          | 1/125 [00:06<12:53,  6.24s/it]
Training Epoch: 9/100, batch 2/250 completed (loss: 0.2549:   1%|          | 1/125 [00:09<12:48,  6.20s/it]
Training Epoch: 9/100, batch 2/250 completed (loss: 0.0547:   1%|          | 1/125 [00:09<12:51,  6.22s/it]
Training Epoch: 9/100, batch 2/250 completed (loss: 0.1863:   1%|          | 1/125 [00:09<12:50,  6.21s/it]
Training Epoch: 9/100, batch 2/250 completed (loss: 0.01:   1%|          | 1/125 [00:09<12:53,  6.24s/it]  
Training Epoch: 9/100, batch 2/250 completed (loss: 0.2549:   2%|▏         | 2/125 [00:12<12:17,  6.00s/it]
Training Epoch: 9/100, batch 2/250 completed (loss: 0.0547:   2%|▏         | 2/125 [00:12<12:17,  5.99s/it]
Training Epoch: 9/100, batch 2/250 completed (loss: 0.1863:   2%|▏         | 2/125 [00:12<12:17,  5.99s/it]
Training Epoch: 9/100, batch 2/250 completed (loss: 0.01:   2%|▏         | 2/125 [00:12<12:18,  6.00s/it]
Training Epoch: 9/100, batch 3/250 completed (loss: 0.017:   2%|▏         | 2/125 [00:12<12:17,  5.99s/it] 
Training Epoch: 9/100, batch 3/250 completed (loss: 0.1555:   2%|▏         | 2/125 [00:12<12:17,  6.00s/it]
Training Epoch: 9/100, batch 3/250 completed (loss: 0.0655:   2%|▏         | 2/125 [00:12<12:17,  5.99s/it]
Training Epoch: 9/100, batch 3/250 completed (loss: 0.166:   2%|▏         | 2/125 [00:12<12:18,  6.00s/it]
Training Epoch: 9/100, batch 4/250 completed (loss: 0.1285:   2%|▏         | 2/125 [00:14<12:17,  5.99s/it]
Training Epoch: 9/100, batch 4/250 completed (loss: 0.2517:   2%|▏         | 2/125 [00:14<12:17,  5.99s/it]
Training Epoch: 9/100, batch 4/250 completed (loss: 0.1239:   2%|▏         | 2/125 [00:14<12:17,  6.00s/it]
Training Epoch: 9/100, batch 4/250 completed (loss: 0.308:   2%|▏         | 2/125 [00:14<12:18,  6.00s/it]
Training Epoch: 9/100, batch 4/250 completed (loss: 0.2517:   2%|▏         | 3/125 [00:17<11:50,  5.82s/it]
Training Epoch: 9/100, batch 4/250 completed (loss: 0.1239:   2%|▏         | 3/125 [00:17<11:50,  5.83s/it]
Training Epoch: 9/100, batch 4/250 completed (loss: 0.308:   2%|▏         | 3/125 [00:17<11:51,  5.83s/it]
Training Epoch: 9/100, batch 4/250 completed (loss: 0.1285:   2%|▏         | 3/125 [00:17<11:51,  5.83s/it]
Training Epoch: 9/100, batch 5/250 completed (loss: 0.0539:   2%|▏         | 3/125 [00:17<11:50,  5.82s/it]
Training Epoch: 9/100, batch 5/250 completed (loss: 0.2687:   2%|▏         | 3/125 [00:17<11:51,  5.83s/it]
Training Epoch: 9/100, batch 5/250 completed (loss: 0.029:   2%|▏         | 3/125 [00:17<11:50,  5.83s/it] 
Training Epoch: 9/100, batch 5/250 completed (loss: 0.0317:   2%|▏         | 3/125 [00:17<11:51,  5.83s/it]
Training Epoch: 9/100, batch 6/250 completed (loss: 0.1232:   2%|▏         | 3/125 [00:20<11:50,  5.83s/it]
Training Epoch: 9/100, batch 6/250 completed (loss: 0.0694:   2%|▏         | 3/125 [00:20<11:50,  5.82s/it]
Training Epoch: 9/100, batch 6/250 completed (loss: 0.0365:   2%|▏         | 3/125 [00:20<11:51,  5.83s/it]
Training Epoch: 9/100, batch 6/250 completed (loss: 0.1778:   2%|▏         | 3/125 [00:20<11:51,  5.83s/it]
Training Epoch: 9/100, batch 6/250 completed (loss: 0.0694:   3%|▎         | 4/125 [00:23<11:43,  5.82s/it]
Training Epoch: 9/100, batch 6/250 completed (loss: 0.1232:   3%|▎         | 4/125 [00:23<11:43,  5.82s/it]
Training Epoch: 9/100, batch 6/250 completed (loss: 0.1778:   3%|▎         | 4/125 [00:23<11:44,  5.82s/it]
Training Epoch: 9/100, batch 6/250 completed (loss: 0.0365:   3%|▎         | 4/125 [00:23<11:44,  5.82s/it]
Training Epoch: 9/100, batch 7/250 completed (loss: 0.04:   3%|▎         | 4/125 [00:23<11:44,  5.82s/it]  
Training Epoch: 9/100, batch 7/250 completed (loss: 0.1081:   3%|▎         | 4/125 [00:23<11:43,  5.82s/it]
Training Epoch: 9/100, batch 7/250 completed (loss: 0.0169:   3%|▎         | 4/125 [00:23<11:43,  5.82s/it]
Training Epoch: 9/100, batch 7/250 completed (loss: 0.0898:   3%|▎         | 4/125 [00:23<11:44,  5.82s/it]
Training Epoch: 9/100, batch 8/250 completed (loss: 0.08:   3%|▎         | 4/125 [00:26<11:44,  5.82s/it]
Training Epoch: 9/100, batch 8/250 completed (loss: 0.016:   3%|▎         | 4/125 [00:26<11:43,  5.82s/it] 
Training Epoch: 9/100, batch 8/250 completed (loss: 0.0085:   3%|▎         | 4/125 [00:26<11:43,  5.82s/it]
Training Epoch: 9/100, batch 8/250 completed (loss: 0.0298:   3%|▎         | 4/125 [00:26<11:44,  5.82s/it]
Training Epoch: 9/100, batch 8/250 completed (loss: 0.08:   4%|▍         | 5/125 [00:29<11:45,  5.88s/it]
Training Epoch: 9/100, batch 8/250 completed (loss: 0.016:   4%|▍         | 5/125 [00:29<11:46,  5.88s/it]
Training Epoch: 9/100, batch 8/250 completed (loss: 0.0298:   4%|▍         | 5/125 [00:29<11:46,  5.88s/it]
Training Epoch: 9/100, batch 8/250 completed (loss: 0.0085:   4%|▍         | 5/125 [00:29<11:46,  5.89s/it]
Training Epoch: 9/100, batch 9/250 completed (loss: 0.505:   4%|▍         | 5/125 [00:29<11:46,  5.89s/it] 
Training Epoch: 9/100, batch 9/250 completed (loss: 0.1745:   4%|▍         | 5/125 [00:29<11:46,  5.88s/it]
Training Epoch: 9/100, batch 9/250 completed (loss: 0.0126:   4%|▍         | 5/125 [00:29<11:45,  5.88s/it]
Training Epoch: 9/100, batch 9/250 completed (loss: 0.4321:   4%|▍         | 5/125 [00:29<11:46,  5.88s/it]
Training Epoch: 9/100, batch 10/250 completed (loss: 0.1603:   4%|▍         | 5/125 [00:32<11:46,  5.89s/it]
Training Epoch: 9/100, batch 10/250 completed (loss: 0.3389:   4%|▍         | 5/125 [00:32<11:46,  5.88s/it]
Training Epoch: 9/100, batch 10/250 completed (loss: 0.1632:   4%|▍         | 5/125 [00:32<11:45,  5.88s/it]
Training Epoch: 9/100, batch 10/250 completed (loss: 0.1064:   4%|▍         | 5/125 [00:32<11:46,  5.88s/it]
Training Epoch: 9/100, batch 10/250 completed (loss: 0.3389:   5%|▍         | 6/125 [00:35<11:39,  5.88s/it]
Training Epoch: 9/100, batch 10/250 completed (loss: 0.1632:   5%|▍         | 6/125 [00:35<11:39,  5.88s/it]
Training Epoch: 9/100, batch 10/250 completed (loss: 0.1603:   5%|▍         | 6/125 [00:35<11:40,  5.88s/it]
Training Epoch: 9/100, batch 10/250 completed (loss: 0.1064:   5%|▍         | 6/125 [00:35<11:40,  5.89s/it]
Training Epoch: 9/100, batch 11/250 completed (loss: 0.0984:   5%|▍         | 6/125 [00:35<11:39,  5.88s/it]
Training Epoch: 9/100, batch 11/250 completed (loss: 0.1272:   5%|▍         | 6/125 [00:35<11:39,  5.88s/it]
Training Epoch: 9/100, batch 11/250 completed (loss: 0.3088:   5%|▍         | 6/125 [00:35<11:40,  5.88s/it]
Training Epoch: 9/100, batch 11/250 completed (loss: 0.179:   5%|▍         | 6/125 [00:35<11:40,  5.89s/it] 
Training Epoch: 9/100, batch 12/250 completed (loss: 0.2432:   5%|▍         | 6/125 [00:38<11:39,  5.88s/it]
Training Epoch: 9/100, batch 12/250 completed (loss: 0.1735:   5%|▍         | 6/125 [00:38<11:39,  5.88s/it]
Training Epoch: 9/100, batch 12/250 completed (loss: 0.0822:   5%|▍         | 6/125 [00:38<11:40,  5.88s/it]
Training Epoch: 9/100, batch 12/250 completed (loss: 0.2777:   5%|▍         | 6/125 [00:38<11:40,  5.89s/it]
Training Epoch: 9/100, batch 12/250 completed (loss: 0.1735:   6%|▌         | 7/125 [00:41<11:30,  5.85s/it]
Training Epoch: 9/100, batch 12/250 completed (loss: 0.2777:   6%|▌         | 7/125 [00:41<11:30,  5.85s/it]
Training Epoch: 9/100, batch 12/250 completed (loss: 0.0822:   6%|▌         | 7/125 [00:41<11:30,  5.85s/it]
Training Epoch: 9/100, batch 12/250 completed (loss: 0.2432:   6%|▌         | 7/125 [00:41<11:30,  5.85s/it]
Training Epoch: 9/100, batch 13/250 completed (loss: 0.0625:   6%|▌         | 7/125 [00:41<11:30,  5.85s/it]
Training Epoch: 9/100, batch 13/250 completed (loss: 0.3932:   6%|▌         | 7/125 [00:41<11:30,  5.85s/it]
Training Epoch: 9/100, batch 13/250 completed (loss: 0.0354:   6%|▌         | 7/125 [00:41<11:30,  5.85s/it]
Training Epoch: 9/100, batch 13/250 completed (loss: 0.1554:   6%|▌         | 7/125 [00:41<11:30,  5.85s/it]
Training Epoch: 9/100, batch 14/250 completed (loss: 0.1146:   6%|▌         | 7/125 [00:44<11:30,  5.85s/it]
Training Epoch: 9/100, batch 14/250 completed (loss: 0.0248:   6%|▌         | 7/125 [00:44<11:30,  5.85s/it]
Training Epoch: 9/100, batch 14/250 completed (loss: 0.4099:   6%|▌         | 7/125 [00:44<11:30,  5.85s/it]
Training Epoch: 9/100, batch 14/250 completed (loss: 0.0816:   6%|▌         | 7/125 [00:44<11:30,  5.85s/it]
Training Epoch: 9/100, batch 14/250 completed (loss: 0.1146:   6%|▋         | 8/125 [00:47<11:28,  5.89s/it]
Training Epoch: 9/100, batch 14/250 completed (loss: 0.0816:   6%|▋         | 8/125 [00:47<11:28,  5.89s/it]
Training Epoch: 9/100, batch 14/250 completed (loss: 0.0248:   6%|▋         | 8/125 [00:47<11:28,  5.89s/it]
Training Epoch: 9/100, batch 14/250 completed (loss: 0.4099:   6%|▋         | 8/125 [00:47<11:28,  5.89s/it]
Training Epoch: 9/100, batch 15/250 completed (loss: 0.0953:   6%|▋         | 8/125 [00:47<11:28,  5.89s/it]
Training Epoch: 9/100, batch 15/250 completed (loss: 0.0233:   6%|▋         | 8/125 [00:47<11:28,  5.89s/it]
Training Epoch: 9/100, batch 15/250 completed (loss: 0.0101:   6%|▋         | 8/125 [00:47<11:28,  5.89s/it]
Training Epoch: 9/100, batch 15/250 completed (loss: 0.2935:   6%|▋         | 8/125 [00:47<11:28,  5.89s/it]
Training Epoch: 9/100, batch 16/250 completed (loss: 0.0794:   6%|▋         | 8/125 [00:49<11:28,  5.89s/it]
Training Epoch: 9/100, batch 16/250 completed (loss: 0.033:   6%|▋         | 8/125 [00:49<11:28,  5.89s/it] 
Training Epoch: 9/100, batch 16/250 completed (loss: 1.0968:   6%|▋         | 8/125 [00:49<11:28,  5.89s/it]
Training Epoch: 9/100, batch 16/250 completed (loss: 0.0173:   6%|▋         | 8/125 [00:50<11:28,  5.89s/it]
Training Epoch: 9/100, batch 16/250 completed (loss: 0.0794:   7%|▋         | 9/125 [00:52<11:21,  5.88s/it]
Training Epoch: 9/100, batch 16/250 completed (loss: 1.0968:   7%|▋         | 9/125 [00:52<11:21,  5.88s/it]
Training Epoch: 9/100, batch 16/250 completed (loss: 0.0173:   7%|▋         | 9/125 [00:52<11:21,  5.88s/it]
Training Epoch: 9/100, batch 16/250 completed (loss: 0.033:   7%|▋         | 9/125 [00:52<11:21,  5.88s/it]
Training Epoch: 9/100, batch 17/250 completed (loss: 0.0793:   7%|▋         | 9/125 [00:53<11:21,  5.88s/it]
Training Epoch: 9/100, batch 17/250 completed (loss: 0.2313:   7%|▋         | 9/125 [00:53<11:21,  5.88s/it]
Training Epoch: 9/100, batch 17/250 completed (loss: 0.0203:   7%|▋         | 9/125 [00:53<11:21,  5.88s/it]
Training Epoch: 9/100, batch 17/250 completed (loss: 0.1751:   7%|▋         | 9/125 [00:53<11:21,  5.88s/it]
Training Epoch: 9/100, batch 18/250 completed (loss: 0.0627:   7%|▋         | 9/125 [00:56<11:21,  5.88s/it]
Training Epoch: 9/100, batch 18/250 completed (loss: 0.1911:   7%|▋         | 9/125 [00:56<11:21,  5.88s/it]
Training Epoch: 9/100, batch 18/250 completed (loss: 0.1435:   7%|▋         | 9/125 [00:56<11:21,  5.88s/it]
Training Epoch: 9/100, batch 18/250 completed (loss: 0.6378:   7%|▋         | 9/125 [00:56<11:21,  5.88s/it]
Training Epoch: 9/100, batch 18/250 completed (loss: 0.0627:   8%|▊         | 10/125 [00:58<11:20,  5.92s/it]
Training Epoch: 9/100, batch 18/250 completed (loss: 0.6378:   8%|▊         | 10/125 [00:59<11:20,  5.92s/it]
Training Epoch: 9/100, batch 18/250 completed (loss: 0.1435:   8%|▊         | 10/125 [00:58<11:20,  5.92s/it]
Training Epoch: 9/100, batch 18/250 completed (loss: 0.1911:   8%|▊         | 10/125 [00:58<11:20,  5.92s/it]
Training Epoch: 9/100, batch 19/250 completed (loss: 0.2675:   8%|▊         | 10/125 [00:59<11:20,  5.92s/it]
Training Epoch: 9/100, batch 19/250 completed (loss: 0.1559:   8%|▊         | 10/125 [00:59<11:20,  5.92s/it]
Training Epoch: 9/100, batch 19/250 completed (loss: 0.0573:   8%|▊         | 10/125 [00:59<11:20,  5.92s/it]
Training Epoch: 9/100, batch 19/250 completed (loss: 0.1043:   8%|▊         | 10/125 [00:59<11:20,  5.92s/it]
Training Epoch: 9/100, batch 20/250 completed (loss: 0.0146:   8%|▊         | 10/125 [01:02<11:20,  5.92s/it]
Training Epoch: 9/100, batch 20/250 completed (loss: 0.0685:   8%|▊         | 10/125 [01:02<11:20,  5.92s/it]
Training Epoch: 9/100, batch 20/250 completed (loss: 0.0085:   8%|▊         | 10/125 [01:02<11:20,  5.92s/it]
Training Epoch: 9/100, batch 20/250 completed (loss: 0.0654:   8%|▊         | 10/125 [01:02<11:20,  5.92s/it]
Training Epoch: 9/100, batch 20/250 completed (loss: 0.0146:   9%|▉         | 11/125 [01:04<11:12,  5.90s/it]
Training Epoch: 9/100, batch 20/250 completed (loss: 0.0085:   9%|▉         | 11/125 [01:04<11:12,  5.90s/it]
Training Epoch: 9/100, batch 20/250 completed (loss: 0.0654:   9%|▉         | 11/125 [01:04<11:12,  5.90s/it]
Training Epoch: 9/100, batch 20/250 completed (loss: 0.0685:   9%|▉         | 11/125 [01:04<11:12,  5.90s/it]
Training Epoch: 9/100, batch 21/250 completed (loss: 0.0841:   9%|▉         | 11/125 [01:04<11:12,  5.90s/it]
Training Epoch: 9/100, batch 21/250 completed (loss: 0.1661:   9%|▉         | 11/125 [01:04<11:12,  5.90s/it]
Training Epoch: 9/100, batch 21/250 completed (loss: 0.0421:   9%|▉         | 11/125 [01:04<11:12,  5.90s/it]
Training Epoch: 9/100, batch 21/250 completed (loss: 0.1001:   9%|▉         | 11/125 [01:04<11:12,  5.90s/it]
Training Epoch: 9/100, batch 22/250 completed (loss: 0.6201:   9%|▉         | 11/125 [01:07<11:12,  5.90s/it]
Training Epoch: 9/100, batch 22/250 completed (loss: 0.299:   9%|▉         | 11/125 [01:07<11:12,  5.90s/it] 
Training Epoch: 9/100, batch 22/250 completed (loss: 0.1223:   9%|▉         | 11/125 [01:07<11:12,  5.90s/it]
Training Epoch: 9/100, batch 22/250 completed (loss: 0.2477:   9%|▉         | 11/125 [01:07<11:12,  5.90s/it]
Training Epoch: 9/100, batch 22/250 completed (loss: 0.1223:  10%|▉         | 12/125 [01:10<11:06,  5.90s/it]
Training Epoch: 9/100, batch 22/250 completed (loss: 0.299:  10%|▉         | 12/125 [01:10<11:06,  5.89s/it]
Training Epoch: 9/100, batch 22/250 completed (loss: 0.2477:  10%|▉         | 12/125 [01:10<11:06,  5.90s/it]
Training Epoch: 9/100, batch 22/250 completed (loss: 0.6201:  10%|▉         | 12/125 [01:10<11:06,  5.90s/it]
Training Epoch: 9/100, batch 23/250 completed (loss: 0.2693:  10%|▉         | 12/125 [01:10<11:06,  5.89s/it]
Training Epoch: 9/100, batch 23/250 completed (loss: 0.2689:  10%|▉         | 12/125 [01:10<11:06,  5.90s/it]
Training Epoch: 9/100, batch 23/250 completed (loss: 0.0615:  10%|▉         | 12/125 [01:10<11:06,  5.90s/it]
Training Epoch: 9/100, batch 23/250 completed (loss: 0.0753:  10%|▉         | 12/125 [01:10<11:06,  5.90s/it]
Training Epoch: 9/100, batch 24/250 completed (loss: 0.2489:  10%|▉         | 12/125 [01:13<11:06,  5.90s/it]
Training Epoch: 9/100, batch 24/250 completed (loss: 0.5422:  10%|▉         | 12/125 [01:13<11:06,  5.90s/it]
Training Epoch: 9/100, batch 24/250 completed (loss: 0.4075:  10%|▉         | 12/125 [01:13<11:06,  5.89s/it]
Training Epoch: 9/100, batch 24/250 completed (loss: 0.061:  10%|▉         | 12/125 [01:13<11:06,  5.90s/it] 
Training Epoch: 9/100, batch 24/250 completed (loss: 0.2489:  10%|█         | 13/125 [01:16<11:05,  5.94s/it]
Training Epoch: 9/100, batch 24/250 completed (loss: 0.061:  10%|█         | 13/125 [01:16<11:05,  5.94s/it]
Training Epoch: 9/100, batch 24/250 completed (loss: 0.5422:  10%|█         | 13/125 [01:16<11:05,  5.94s/it]
Training Epoch: 9/100, batch 24/250 completed (loss: 0.4075:  10%|█         | 13/125 [01:16<11:05,  5.94s/it]
Training Epoch: 9/100, batch 25/250 completed (loss: 0.0272:  10%|█         | 13/125 [01:16<11:05,  5.94s/it]
Training Epoch: 9/100, batch 25/250 completed (loss: 0.0827:  10%|█         | 13/125 [01:16<11:05,  5.94s/it]
Training Epoch: 9/100, batch 25/250 completed (loss: 0.1423:  10%|█         | 13/125 [01:16<11:05,  5.94s/it]
Training Epoch: 9/100, batch 25/250 completed (loss: 0.053:  10%|█         | 13/125 [01:16<11:05,  5.94s/it]
Training Epoch: 9/100, batch 26/250 completed (loss: 0.1261:  10%|█         | 13/125 [01:19<11:05,  5.94s/it]
Training Epoch: 9/100, batch 26/250 completed (loss: 0.0451:  10%|█         | 13/125 [01:19<11:05,  5.94s/it]
Training Epoch: 9/100, batch 26/250 completed (loss: 0.2076:  10%|█         | 13/125 [01:19<11:05,  5.94s/it]
Training Epoch: 9/100, batch 26/250 completed (loss: 0.0657:  10%|█         | 13/125 [01:19<11:05,  5.94s/it]
Training Epoch: 9/100, batch 26/250 completed (loss: 0.0451:  11%|█         | 14/125 [01:22<10:58,  5.93s/it]
Training Epoch: 9/100, batch 26/250 completed (loss: 0.0657:  11%|█         | 14/125 [01:22<10:58,  5.93s/it]
Training Epoch: 9/100, batch 26/250 completed (loss: 0.2076:  11%|█         | 14/125 [01:22<10:59,  5.94s/it]
Training Epoch: 9/100, batch 26/250 completed (loss: 0.1261:  11%|█         | 14/125 [01:22<10:58,  5.94s/it]
Training Epoch: 9/100, batch 27/250 completed (loss: 0.4508:  11%|█         | 14/125 [01:22<10:59,  5.94s/it]
Training Epoch: 9/100, batch 27/250 completed (loss: 0.1092:  11%|█         | 14/125 [01:22<10:58,  5.94s/it]
Training Epoch: 9/100, batch 27/250 completed (loss: 0.1329:  11%|█         | 14/125 [01:22<10:58,  5.93s/it]
Training Epoch: 9/100, batch 27/250 completed (loss: 0.1216:  11%|█         | 14/125 [01:22<10:58,  5.93s/it]
Training Epoch: 9/100, batch 28/250 completed (loss: 0.1445:  11%|█         | 14/125 [01:25<10:59,  5.94s/it]
Training Epoch: 9/100, batch 28/250 completed (loss: 0.1608:  11%|█         | 14/125 [01:25<10:58,  5.93s/it]
Training Epoch: 9/100, batch 28/250 completed (loss: 0.071:  11%|█         | 14/125 [01:25<10:58,  5.94s/it] 
Training Epoch: 9/100, batch 28/250 completed (loss: 0.2996:  11%|█         | 14/125 [01:25<10:58,  5.93s/it]
Training Epoch: 9/100, batch 28/250 completed (loss: 0.1608:  12%|█▏        | 15/125 [01:28<10:57,  5.97s/it]
Training Epoch: 9/100, batch 28/250 completed (loss: 0.2996:  12%|█▏        | 15/125 [01:28<10:57,  5.97s/it]
Training Epoch: 9/100, batch 28/250 completed (loss: 0.071:  12%|█▏        | 15/125 [01:28<10:57,  5.97s/it]
Training Epoch: 9/100, batch 28/250 completed (loss: 0.1445:  12%|█▏        | 15/125 [01:28<10:57,  5.97s/it]
Training Epoch: 9/100, batch 29/250 completed (loss: 0.4207:  12%|█▏        | 15/125 [01:28<10:57,  5.97s/it]
Training Epoch: 9/100, batch 29/250 completed (loss: 0.1125:  12%|█▏        | 15/125 [01:28<10:57,  5.97s/it]
Training Epoch: 9/100, batch 29/250 completed (loss: 0.1496:  12%|█▏        | 15/125 [01:28<10:57,  5.97s/it]
Training Epoch: 9/100, batch 29/250 completed (loss: 0.3606:  12%|█▏        | 15/125 [01:28<10:57,  5.97s/it]
Training Epoch: 9/100, batch 30/250 completed (loss: 0.1751:  12%|█▏        | 15/125 [01:31<10:57,  5.97s/it]
Training Epoch: 9/100, batch 30/250 completed (loss: 0.1359:  12%|█▏        | 15/125 [01:31<10:57,  5.97s/it]
Training Epoch: 9/100, batch 30/250 completed (loss: 0.2152:  12%|█▏        | 15/125 [01:31<10:57,  5.97s/it]
Training Epoch: 9/100, batch 30/250 completed (loss: 0.1581:  12%|█▏        | 15/125 [01:31<10:57,  5.97s/it]
Training Epoch: 9/100, batch 30/250 completed (loss: 0.2152:  13%|█▎        | 16/125 [01:34<10:48,  5.95s/it]
Training Epoch: 9/100, batch 30/250 completed (loss: 0.1751:  13%|█▎        | 16/125 [01:34<10:47,  5.94s/it]
Training Epoch: 9/100, batch 30/250 completed (loss: 0.1581:  13%|█▎        | 16/125 [01:34<10:47,  5.94s/it]
Training Epoch: 9/100, batch 30/250 completed (loss: 0.1359:  13%|█▎        | 16/125 [01:34<10:47,  5.94s/it]
Training Epoch: 9/100, batch 31/250 completed (loss: 0.7915:  13%|█▎        | 16/125 [01:34<10:47,  5.94s/it]
Training Epoch: 9/100, batch 31/250 completed (loss: 0.1205:  13%|█▎        | 16/125 [01:34<10:48,  5.95s/it]
Training Epoch: 9/100, batch 31/250 completed (loss: 0.0744:  13%|█▎        | 16/125 [01:34<10:47,  5.94s/it]
Training Epoch: 9/100, batch 31/250 completed (loss: 0.4198:  13%|█▎        | 16/125 [01:34<10:47,  5.94s/it]
Training Epoch: 9/100, batch 32/250 completed (loss: 0.0351:  13%|█▎        | 16/125 [01:37<10:47,  5.94s/it]
Training Epoch: 9/100, batch 32/250 completed (loss: 0.1217:  13%|█▎        | 16/125 [01:37<10:48,  5.95s/it]
Training Epoch: 9/100, batch 32/250 completed (loss: 0.191:  13%|█▎        | 16/125 [01:37<10:47,  5.94s/it] 
Training Epoch: 9/100, batch 32/250 completed (loss: 0.027:  13%|█▎        | 16/125 [01:37<10:47,  5.94s/it] 
Training Epoch: 9/100, batch 32/250 completed (loss: 0.027:  14%|█▎        | 17/125 [01:40<10:41,  5.94s/it]
Training Epoch: 9/100, batch 32/250 completed (loss: 0.191:  14%|█▎        | 17/125 [01:40<10:41,  5.94s/it]
Training Epoch: 9/100, batch 32/250 completed (loss: 0.1217:  14%|█▎        | 17/125 [01:40<10:41,  5.94s/it]
Training Epoch: 9/100, batch 32/250 completed (loss: 0.0351:  14%|█▎        | 17/125 [01:40<10:41,  5.94s/it]
Training Epoch: 9/100, batch 33/250 completed (loss: 0.0451:  14%|█▎        | 17/125 [01:40<10:41,  5.94s/it]
Training Epoch: 9/100, batch 33/250 completed (loss: 0.0806:  14%|█▎        | 17/125 [01:40<10:41,  5.94s/it]
Training Epoch: 9/100, batch 33/250 completed (loss: 0.1526:  14%|█▎        | 17/125 [01:40<10:41,  5.94s/it]
Training Epoch: 9/100, batch 33/250 completed (loss: 0.1319:  14%|█▎        | 17/125 [01:40<10:41,  5.94s/it]
Training Epoch: 9/100, batch 34/250 completed (loss: 0.1568:  14%|█▎        | 17/125 [01:43<10:41,  5.94s/it]
Training Epoch: 9/100, batch 34/250 completed (loss: 0.0421:  14%|█▎        | 17/125 [01:43<10:41,  5.94s/it]
Training Epoch: 9/100, batch 34/250 completed (loss: 0.0264:  14%|█▎        | 17/125 [01:43<10:41,  5.94s/it]
Training Epoch: 9/100, batch 34/250 completed (loss: 0.2223:  14%|█▎        | 17/125 [01:43<10:41,  5.94s/it]
Training Epoch: 9/100, batch 34/250 completed (loss: 0.1568:  14%|█▍        | 18/125 [01:46<10:45,  6.03s/it]
Training Epoch: 9/100, batch 34/250 completed (loss: 0.2223:  14%|█▍        | 18/125 [01:46<10:45,  6.03s/it]
Training Epoch: 9/100, batch 34/250 completed (loss: 0.0421:  14%|█▍        | 18/125 [01:46<10:45,  6.03s/it]
Training Epoch: 9/100, batch 34/250 completed (loss: 0.0264:  14%|█▍        | 18/125 [01:46<10:45,  6.03s/it]
Training Epoch: 9/100, batch 35/250 completed (loss: 0.074:  14%|█▍        | 18/125 [01:46<10:45,  6.03s/it] 
Training Epoch: 9/100, batch 35/250 completed (loss: 0.0391:  14%|█▍        | 18/125 [01:46<10:45,  6.03s/it]
Training Epoch: 9/100, batch 35/250 completed (loss: 0.2041:  14%|█▍        | 18/125 [01:46<10:45,  6.03s/it]
Training Epoch: 9/100, batch 35/250 completed (loss: 0.0191:  14%|█▍        | 18/125 [01:46<10:45,  6.03s/it]
Training Epoch: 9/100, batch 36/250 completed (loss: 0.1123:  14%|█▍        | 18/125 [01:49<10:45,  6.03s/it]
Training Epoch: 9/100, batch 36/250 completed (loss: 0.2869:  14%|█▍        | 18/125 [01:49<10:45,  6.03s/it]
Training Epoch: 9/100, batch 36/250 completed (loss: 0.1669:  14%|█▍        | 18/125 [01:49<10:45,  6.03s/it]
Training Epoch: 9/100, batch 36/250 completed (loss: 0.0553:  14%|█▍        | 18/125 [01:49<10:45,  6.03s/it]
Training Epoch: 9/100, batch 36/250 completed (loss: 0.1123:  15%|█▌        | 19/125 [01:52<10:30,  5.95s/it]
Training Epoch: 9/100, batch 36/250 completed (loss: 0.0553:  15%|█▌        | 19/125 [01:52<10:30,  5.95s/it]
Training Epoch: 9/100, batch 36/250 completed (loss: 0.2869:  15%|█▌        | 19/125 [01:52<10:30,  5.95s/it]
Training Epoch: 9/100, batch 36/250 completed (loss: 0.1669:  15%|█▌        | 19/125 [01:52<10:31,  5.95s/it]
Training Epoch: 9/100, batch 37/250 completed (loss: 0.0511:  15%|█▌        | 19/125 [01:52<10:31,  5.95s/it]
Training Epoch: 9/100, batch 37/250 completed (loss: 0.0831:  15%|█▌        | 19/125 [01:52<10:30,  5.95s/it]
Training Epoch: 9/100, batch 37/250 completed (loss: 0.2262:  15%|█▌        | 19/125 [01:52<10:30,  5.95s/it]
Training Epoch: 9/100, batch 37/250 completed (loss: 0.0534:  15%|█▌        | 19/125 [01:52<10:30,  5.95s/it]
Training Epoch: 9/100, batch 38/250 completed (loss: 0.0967:  15%|█▌        | 19/125 [01:55<10:30,  5.95s/it]
Training Epoch: 9/100, batch 38/250 completed (loss: 0.0194:  15%|█▌        | 19/125 [01:55<10:30,  5.95s/it]
Training Epoch: 9/100, batch 38/250 completed (loss: 0.0459:  15%|█▌        | 19/125 [01:55<10:31,  5.95s/it]
Training Epoch: 9/100, batch 38/250 completed (loss: 0.2729:  15%|█▌        | 19/125 [01:55<10:30,  5.95s/it]
Training Epoch: 9/100, batch 38/250 completed (loss: 0.2729:  16%|█▌        | 20/125 [01:58<10:33,  6.03s/it]
Training Epoch: 9/100, batch 38/250 completed (loss: 0.0194:  16%|█▌        | 20/125 [01:58<10:33,  6.03s/it]
Training Epoch: 9/100, batch 38/250 completed (loss: 0.0967:  16%|█▌        | 20/125 [01:58<10:33,  6.03s/it]
Training Epoch: 9/100, batch 38/250 completed (loss: 0.0459:  16%|█▌        | 20/125 [01:58<10:32,  6.03s/it]
Training Epoch: 9/100, batch 39/250 completed (loss: 0.1377:  16%|█▌        | 20/125 [01:58<10:33,  6.03s/it]
Training Epoch: 9/100, batch 39/250 completed (loss: 0.0392:  16%|█▌        | 20/125 [01:58<10:32,  6.03s/it]
Training Epoch: 9/100, batch 39/250 completed (loss: 0.1444:  16%|█▌        | 20/125 [01:58<10:33,  6.03s/it]
Training Epoch: 9/100, batch 39/250 completed (loss: 0.1009:  16%|█▌        | 20/125 [01:58<10:33,  6.03s/it]
Training Epoch: 9/100, batch 40/250 completed (loss: 0.0409:  16%|█▌        | 20/125 [02:01<10:33,  6.03s/it]
Training Epoch: 9/100, batch 40/250 completed (loss: 0.0931:  16%|█▌        | 20/125 [02:01<10:32,  6.03s/it]
Training Epoch: 9/100, batch 40/250 completed (loss: 0.1095:  16%|█▌        | 20/125 [02:01<10:33,  6.03s/it]
Training Epoch: 9/100, batch 40/250 completed (loss: 0.0752:  16%|█▌        | 20/125 [02:01<10:33,  6.03s/it]
Training Epoch: 9/100, batch 40/250 completed (loss: 0.0409:  17%|█▋        | 21/125 [02:04<10:24,  6.00s/it]
Training Epoch: 9/100, batch 40/250 completed (loss: 0.1095:  17%|█▋        | 21/125 [02:04<10:24,  6.01s/it]
Training Epoch: 9/100, batch 40/250 completed (loss: 0.0752:  17%|█▋        | 21/125 [02:04<10:24,  6.01s/it]
Training Epoch: 9/100, batch 40/250 completed (loss: 0.0931:  17%|█▋        | 21/125 [02:04<10:24,  6.00s/it]
Training Epoch: 9/100, batch 41/250 completed (loss: 0.076:  17%|█▋        | 21/125 [02:04<10:24,  6.00s/it] 
Training Epoch: 9/100, batch 41/250 completed (loss: 0.1204:  17%|█▋        | 21/125 [02:04<10:24,  6.01s/it]
Training Epoch: 9/100, batch 41/250 completed (loss: 0.1244:  17%|█▋        | 21/125 [02:04<10:24,  6.00s/it]
Training Epoch: 9/100, batch 41/250 completed (loss: 0.3064:  17%|█▋        | 21/125 [02:04<10:24,  6.01s/it]
Training Epoch: 9/100, batch 42/250 completed (loss: 0.0492:  17%|█▋        | 21/125 [02:07<10:24,  6.01s/it]
Training Epoch: 9/100, batch 42/250 completed (loss: 0.0214:  17%|█▋        | 21/125 [02:07<10:24,  6.00s/it]
Training Epoch: 9/100, batch 42/250 completed (loss: 0.0252:  17%|█▋        | 21/125 [02:07<10:24,  6.00s/it]
Training Epoch: 9/100, batch 42/250 completed (loss: 0.0792:  17%|█▋        | 21/125 [02:07<10:24,  6.01s/it]
Training Epoch: 9/100, batch 42/250 completed (loss: 0.0214:  18%|█▊        | 22/125 [02:10<10:12,  5.94s/it]
Training Epoch: 9/100, batch 42/250 completed (loss: 0.0792:  18%|█▊        | 22/125 [02:10<10:12,  5.94s/it]
Training Epoch: 9/100, batch 42/250 completed (loss: 0.0492:  18%|█▊        | 22/125 [02:10<10:12,  5.94s/it]
Training Epoch: 9/100, batch 42/250 completed (loss: 0.0252:  18%|█▊        | 22/125 [02:10<10:12,  5.94s/it]
Training Epoch: 9/100, batch 43/250 completed (loss: 0.1044:  18%|█▊        | 22/125 [02:10<10:12,  5.94s/it]
Training Epoch: 9/100, batch 43/250 completed (loss: 0.0177:  18%|█▊        | 22/125 [02:10<10:12,  5.94s/it]
Training Epoch: 9/100, batch 43/250 completed (loss: 0.0378:  18%|█▊        | 22/125 [02:10<10:12,  5.94s/it]
Training Epoch: 9/100, batch 43/250 completed (loss: 0.24:  18%|█▊        | 22/125 [02:10<10:12,  5.94s/it]  
Training Epoch: 9/100, batch 44/250 completed (loss: 0.1191:  18%|█▊        | 22/125 [02:13<10:12,  5.94s/it]
Training Epoch: 9/100, batch 44/250 completed (loss: 0.3197:  18%|█▊        | 22/125 [02:13<10:12,  5.94s/it]
Training Epoch: 9/100, batch 44/250 completed (loss: 0.2026:  18%|█▊        | 22/125 [02:13<10:12,  5.94s/it]
Training Epoch: 9/100, batch 44/250 completed (loss: 0.1975:  18%|█▊        | 22/125 [02:13<10:12,  5.94s/it]
Training Epoch: 9/100, batch 44/250 completed (loss: 0.1191:  18%|█▊        | 23/125 [02:16<10:20,  6.09s/it]
Training Epoch: 9/100, batch 44/250 completed (loss: 0.1975:  18%|█▊        | 23/125 [02:16<10:21,  6.09s/it]
Training Epoch: 9/100, batch 44/250 completed (loss: 0.2026:  18%|█▊        | 23/125 [02:16<10:20,  6.09s/it]
Training Epoch: 9/100, batch 44/250 completed (loss: 0.3197:  18%|█▊        | 23/125 [02:16<10:21,  6.09s/it]
Training Epoch: 9/100, batch 45/250 completed (loss: 0.1478:  18%|█▊        | 23/125 [02:17<10:20,  6.09s/it]
Training Epoch: 9/100, batch 45/250 completed (loss: 0.0419:  18%|█▊        | 23/125 [02:17<10:21,  6.09s/it]
Training Epoch: 9/100, batch 45/250 completed (loss: 0.0508:  18%|█▊        | 23/125 [02:17<10:20,  6.09s/it]
Training Epoch: 9/100, batch 45/250 completed (loss: 0.1103:  18%|█▊        | 23/125 [02:17<10:21,  6.09s/it]
Training Epoch: 9/100, batch 46/250 completed (loss: 0.0658:  18%|█▊        | 23/125 [02:19<10:20,  6.09s/it]
Training Epoch: 9/100, batch 46/250 completed (loss: 0.2515:  18%|█▊        | 23/125 [02:19<10:20,  6.09s/it]
Training Epoch: 9/100, batch 46/250 completed (loss: 0.0669:  18%|█▊        | 23/125 [02:19<10:21,  6.09s/it]
Training Epoch: 9/100, batch 46/250 completed (loss: 0.0926:  18%|█▊        | 23/125 [02:19<10:21,  6.09s/it]
Training Epoch: 9/100, batch 46/250 completed (loss: 0.2515:  19%|█▉        | 24/125 [02:22<10:06,  6.00s/it]
Training Epoch: 9/100, batch 46/250 completed (loss: 0.0926:  19%|█▉        | 24/125 [02:22<10:06,  6.00s/it]
Training Epoch: 9/100, batch 46/250 completed (loss: 0.0658:  19%|█▉        | 24/125 [02:22<10:06,  6.01s/it]
Training Epoch: 9/100, batch 46/250 completed (loss: 0.0669:  19%|█▉        | 24/125 [02:22<10:06,  6.00s/it]
Training Epoch: 9/100, batch 47/250 completed (loss: 0.0661:  19%|█▉        | 24/125 [02:22<10:06,  6.01s/it]
Training Epoch: 9/100, batch 47/250 completed (loss: 0.1036:  19%|█▉        | 24/125 [02:22<10:06,  6.00s/it]
Training Epoch: 9/100, batch 47/250 completed (loss: 0.1279:  19%|█▉        | 24/125 [02:22<10:06,  6.00s/it]
Training Epoch: 9/100, batch 47/250 completed (loss: 0.0672:  19%|█▉        | 24/125 [02:22<10:06,  6.00s/it]
Training Epoch: 9/100, batch 48/250 completed (loss: 0.126:  19%|█▉        | 24/125 [02:25<10:06,  6.00s/it] 
Training Epoch: 9/100, batch 48/250 completed (loss: 0.0466:  19%|█▉        | 24/125 [02:25<10:06,  6.01s/it]
Training Epoch: 9/100, batch 48/250 completed (loss: 0.0643:  19%|█▉        | 24/125 [02:25<10:06,  6.00s/it]
Training Epoch: 9/100, batch 48/250 completed (loss: 0.1207:  19%|█▉        | 24/125 [02:25<10:06,  6.00s/it]
Training Epoch: 9/100, batch 48/250 completed (loss: 0.1207:  20%|██        | 25/125 [02:28<09:58,  5.98s/it]
Training Epoch: 9/100, batch 48/250 completed (loss: 0.0643:  20%|██        | 25/125 [02:28<09:58,  5.98s/it]
Training Epoch: 9/100, batch 48/250 completed (loss: 0.0466:  20%|██        | 25/125 [02:28<09:58,  5.99s/it]
Training Epoch: 9/100, batch 48/250 completed (loss: 0.126:  20%|██        | 25/125 [02:28<09:58,  5.99s/it]
Training Epoch: 9/100, batch 49/250 completed (loss: 0.0751:  20%|██        | 25/125 [02:28<09:58,  5.99s/it]
Training Epoch: 9/100, batch 49/250 completed (loss: 0.0309:  20%|██        | 25/125 [02:28<09:58,  5.98s/it]
Training Epoch: 9/100, batch 49/250 completed (loss: 0.2163:  20%|██        | 25/125 [02:28<09:58,  5.99s/it]
Training Epoch: 9/100, batch 49/250 completed (loss: 0.3009:  20%|██        | 25/125 [02:28<09:58,  5.98s/it]
Training Epoch: 9/100, batch 50/250 completed (loss: 0.1021:  20%|██        | 25/125 [02:32<09:58,  5.99s/it]
Training Epoch: 9/100, batch 50/250 completed (loss: 0.2259:  20%|██        | 25/125 [02:32<09:58,  5.98s/it]
Training Epoch: 9/100, batch 50/250 completed (loss: 0.0503:  20%|██        | 25/125 [02:32<09:58,  5.99s/it]
Training Epoch: 9/100, batch 50/250 completed (loss: 0.0906:  20%|██        | 25/125 [02:32<09:58,  5.98s/it]
Training Epoch: 9/100, batch 50/250 completed (loss: 0.0503:  21%|██        | 26/125 [02:34<10:00,  6.06s/it]
Training Epoch: 9/100, batch 50/250 completed (loss: 0.0906:  21%|██        | 26/125 [02:34<10:00,  6.06s/it]
Training Epoch: 9/100, batch 50/250 completed (loss: 0.2259:  21%|██        | 26/125 [02:34<10:00,  6.06s/it]
Training Epoch: 9/100, batch 50/250 completed (loss: 0.1021:  21%|██        | 26/125 [02:34<10:00,  6.06s/it]
Training Epoch: 9/100, batch 51/250 completed (loss: 0.3916:  21%|██        | 26/125 [02:34<10:00,  6.06s/it]
Training Epoch: 9/100, batch 51/250 completed (loss: 0.0945:  21%|██        | 26/125 [02:34<10:00,  6.06s/it]
Training Epoch: 9/100, batch 51/250 completed (loss: 0.2324:  21%|██        | 26/125 [02:34<10:00,  6.06s/it]
Training Epoch: 9/100, batch 51/250 completed (loss: 0.0989:  21%|██        | 26/125 [02:35<10:00,  6.06s/it]
Training Epoch: 9/100, batch 52/250 completed (loss: 0.057:  21%|██        | 26/125 [02:37<10:00,  6.06s/it] 
Training Epoch: 9/100, batch 52/250 completed (loss: 0.0512:  21%|██        | 26/125 [02:37<10:00,  6.06s/it]
Training Epoch: 9/100, batch 52/250 completed (loss: 0.2358:  21%|██        | 26/125 [02:37<10:00,  6.06s/it]
Training Epoch: 9/100, batch 52/250 completed (loss: 0.0515:  21%|██        | 26/125 [02:37<10:00,  6.06s/it]
Training Epoch: 9/100, batch 52/250 completed (loss: 0.057:  22%|██▏       | 27/125 [02:40<09:47,  5.99s/it]
Training Epoch: 9/100, batch 52/250 completed (loss: 0.2358:  22%|██▏       | 27/125 [02:40<09:47,  5.99s/it]
Training Epoch: 9/100, batch 52/250 completed (loss: 0.0512:  22%|██▏       | 27/125 [02:40<09:47,  5.99s/it]
Training Epoch: 9/100, batch 52/250 completed (loss: 0.0515:  22%|██▏       | 27/125 [02:40<09:47,  5.99s/it]
Training Epoch: 9/100, batch 53/250 completed (loss: 0.1293:  22%|██▏       | 27/125 [02:40<09:47,  5.99s/it]
Training Epoch: 9/100, batch 53/250 completed (loss: 0.1269:  22%|██▏       | 27/125 [02:40<09:47,  5.99s/it]
Training Epoch: 9/100, batch 53/250 completed (loss: 0.1092:  22%|██▏       | 27/125 [02:40<09:47,  5.99s/it]
Training Epoch: 9/100, batch 53/250 completed (loss: 0.091:  22%|██▏       | 27/125 [02:40<09:47,  5.99s/it] 
Training Epoch: 9/100, batch 54/250 completed (loss: 0.0722:  22%|██▏       | 27/125 [02:43<09:47,  5.99s/it]
Training Epoch: 9/100, batch 54/250 completed (loss: 0.1353:  22%|██▏       | 27/125 [02:43<09:47,  5.99s/it]
Training Epoch: 9/100, batch 54/250 completed (loss: 0.0942:  22%|██▏       | 27/125 [02:43<09:47,  5.99s/it]
Training Epoch: 9/100, batch 54/250 completed (loss: 0.1446:  22%|██▏       | 27/125 [02:43<09:47,  5.99s/it]
Training Epoch: 9/100, batch 54/250 completed (loss: 0.1353:  22%|██▏       | 28/125 [02:47<09:50,  6.08s/it]
Training Epoch: 9/100, batch 54/250 completed (loss: 0.0942:  22%|██▏       | 28/125 [02:47<09:50,  6.08s/it]
Training Epoch: 9/100, batch 54/250 completed (loss: 0.1446:  22%|██▏       | 28/125 [02:47<09:50,  6.08s/it]
Training Epoch: 9/100, batch 54/250 completed (loss: 0.0722:  22%|██▏       | 28/125 [02:47<09:50,  6.08s/it]
Training Epoch: 9/100, batch 55/250 completed (loss: 0.3027:  22%|██▏       | 28/125 [02:47<09:50,  6.08s/it]
Training Epoch: 9/100, batch 55/250 completed (loss: 0.1838:  22%|██▏       | 28/125 [02:47<09:50,  6.08s/it]
Training Epoch: 9/100, batch 55/250 completed (loss: 0.3472:  22%|██▏       | 28/125 [02:47<09:50,  6.08s/it]
Training Epoch: 9/100, batch 55/250 completed (loss: 0.207:  22%|██▏       | 28/125 [02:47<09:50,  6.08s/it] 
Training Epoch: 9/100, batch 56/250 completed (loss: 0.1918:  22%|██▏       | 28/125 [02:50<09:50,  6.08s/it]
Training Epoch: 9/100, batch 56/250 completed (loss: 0.0983:  22%|██▏       | 28/125 [02:50<09:50,  6.08s/it]
Training Epoch: 9/100, batch 56/250 completed (loss: 0.0788:  22%|██▏       | 28/125 [02:50<09:50,  6.08s/it]
Training Epoch: 9/100, batch 56/250 completed (loss: 0.3317:  22%|██▏       | 28/125 [02:50<09:50,  6.08s/it]
Training Epoch: 9/100, batch 56/250 completed (loss: 0.1918:  23%|██▎       | 29/125 [02:52<09:36,  6.01s/it]
Training Epoch: 9/100, batch 56/250 completed (loss: 0.0788:  23%|██▎       | 29/125 [02:52<09:36,  6.01s/it]
Training Epoch: 9/100, batch 56/250 completed (loss: 0.3317:  23%|██▎       | 29/125 [02:52<09:36,  6.01s/it]
Training Epoch: 9/100, batch 56/250 completed (loss: 0.0983:  23%|██▎       | 29/125 [02:52<09:37,  6.01s/it]
Training Epoch: 9/100, batch 57/250 completed (loss: 0.2267:  23%|██▎       | 29/125 [02:52<09:37,  6.01s/it]
Training Epoch: 9/100, batch 57/250 completed (loss: 0.0827:  23%|██▎       | 29/125 [02:52<09:36,  6.01s/it]
Training Epoch: 9/100, batch 57/250 completed (loss: 0.1597:  23%|██▎       | 29/125 [02:52<09:36,  6.01s/it]
Training Epoch: 9/100, batch 57/250 completed (loss: 0.0911:  23%|██▎       | 29/125 [02:52<09:36,  6.01s/it]
Training Epoch: 9/100, batch 58/250 completed (loss: 0.1552:  23%|██▎       | 29/125 [02:55<09:36,  6.01s/it]
Training Epoch: 9/100, batch 58/250 completed (loss: 0.0862:  23%|██▎       | 29/125 [02:55<09:36,  6.01s/it]
Training Epoch: 9/100, batch 58/250 completed (loss: 0.067:  23%|██▎       | 29/125 [02:55<09:37,  6.01s/it] 
Training Epoch: 9/100, batch 58/250 completed (loss: 0.0261:  23%|██▎       | 29/125 [02:55<09:36,  6.01s/it]
Training Epoch: 9/100, batch 58/250 completed (loss: 0.0862:  24%|██▍       | 30/125 [02:58<09:30,  6.00s/it]
Training Epoch: 9/100, batch 58/250 completed (loss: 0.0261:  24%|██▍       | 30/125 [02:58<09:30,  6.00s/it]
Training Epoch: 9/100, batch 58/250 completed (loss: 0.067:  24%|██▍       | 30/125 [02:58<09:30,  6.00s/it]
Training Epoch: 9/100, batch 58/250 completed (loss: 0.1552:  24%|██▍       | 30/125 [02:58<09:30,  6.00s/it]
Training Epoch: 9/100, batch 59/250 completed (loss: 0.0489:  24%|██▍       | 30/125 [02:58<09:30,  6.00s/it]
Training Epoch: 9/100, batch 59/250 completed (loss: 0.1377:  24%|██▍       | 30/125 [02:58<09:30,  6.00s/it]
Training Epoch: 9/100, batch 59/250 completed (loss: 0.205:  24%|██▍       | 30/125 [02:58<09:30,  6.00s/it] 
Training Epoch: 9/100, batch 59/250 completed (loss: 0.3377:  24%|██▍       | 30/125 [02:58<09:30,  6.00s/it]
Training Epoch: 9/100, batch 60/250 completed (loss: 0.0436:  24%|██▍       | 30/125 [03:02<09:30,  6.00s/it]
Training Epoch: 9/100, batch 60/250 completed (loss: 0.0178:  24%|██▍       | 30/125 [03:02<09:30,  6.00s/it]
Training Epoch: 9/100, batch 60/250 completed (loss: 0.4814:  24%|██▍       | 30/125 [03:02<09:30,  6.00s/it]
Training Epoch: 9/100, batch 60/250 completed (loss: 0.0221:  24%|██▍       | 30/125 [03:02<09:30,  6.00s/it]
Training Epoch: 9/100, batch 60/250 completed (loss: 0.4814:  25%|██▍       | 31/125 [03:05<09:31,  6.08s/it]
Training Epoch: 9/100, batch 60/250 completed (loss: 0.0178:  25%|██▍       | 31/125 [03:05<09:31,  6.08s/it]
Training Epoch: 9/100, batch 60/250 completed (loss: 0.0436:  25%|██▍       | 31/125 [03:05<09:31,  6.08s/it]
Training Epoch: 9/100, batch 60/250 completed (loss: 0.0221:  25%|██▍       | 31/125 [03:05<09:31,  6.08s/it]
Training Epoch: 9/100, batch 61/250 completed (loss: 0.1292:  25%|██▍       | 31/125 [03:05<09:31,  6.08s/it]
Training Epoch: 9/100, batch 61/250 completed (loss: 0.0719:  25%|██▍       | 31/125 [03:05<09:31,  6.08s/it]
Training Epoch: 9/100, batch 61/250 completed (loss: 0.1132:  25%|██▍       | 31/125 [03:05<09:31,  6.08s/it]
Training Epoch: 9/100, batch 61/250 completed (loss: 0.1792:  25%|██▍       | 31/125 [03:05<09:31,  6.08s/it]
Training Epoch: 9/100, batch 62/250 completed (loss: 0.6701:  25%|██▍       | 31/125 [03:08<09:31,  6.08s/it]
Training Epoch: 9/100, batch 62/250 completed (loss: 0.1148:  25%|██▍       | 31/125 [03:08<09:31,  6.08s/it]
Training Epoch: 9/100, batch 62/250 completed (loss: 0.0781:  25%|██▍       | 31/125 [03:08<09:31,  6.08s/it]
Training Epoch: 9/100, batch 62/250 completed (loss: 0.3812:  25%|██▍       | 31/125 [03:08<09:31,  6.08s/it]
Training Epoch: 9/100, batch 62/250 completed (loss: 0.1148:  26%|██▌       | 32/125 [03:10<09:17,  6.00s/it]
Training Epoch: 9/100, batch 62/250 completed (loss: 0.6701:  26%|██▌       | 32/125 [03:10<09:17,  6.00s/it]
Training Epoch: 9/100, batch 62/250 completed (loss: 0.3812:  26%|██▌       | 32/125 [03:10<09:17,  6.00s/it]
Training Epoch: 9/100, batch 62/250 completed (loss: 0.0781:  26%|██▌       | 32/125 [03:10<09:17,  6.00s/it]
Training Epoch: 9/100, batch 63/250 completed (loss: 0.5593:  26%|██▌       | 32/125 [03:11<09:17,  6.00s/it]
Training Epoch: 9/100, batch 63/250 completed (loss: 0.1682:  26%|██▌       | 32/125 [03:11<09:17,  6.00s/it]
Training Epoch: 9/100, batch 63/250 completed (loss: 0.2698:  26%|██▌       | 32/125 [03:11<09:17,  6.00s/it]
Training Epoch: 9/100, batch 63/250 completed (loss: 0.5141:  26%|██▌       | 32/125 [03:11<09:17,  6.00s/it]
Training Epoch: 9/100, batch 64/250 completed (loss: 0.0536:  26%|██▌       | 32/125 [03:13<09:17,  6.00s/it]
Training Epoch: 9/100, batch 64/250 completed (loss: 0.0104:  26%|██▌       | 32/125 [03:13<09:17,  6.00s/it]
Training Epoch: 9/100, batch 64/250 completed (loss: 0.6388:  26%|██▌       | 32/125 [03:13<09:17,  6.00s/it]
Training Epoch: 9/100, batch 64/250 completed (loss: 0.0388:  26%|██▌       | 32/125 [03:13<09:17,  6.00s/it]
Training Epoch: 9/100, batch 64/250 completed (loss: 0.0388:  26%|██▋       | 33/125 [03:17<09:16,  6.05s/it]
Training Epoch: 9/100, batch 64/250 completed (loss: 0.0104:  26%|██▋       | 33/125 [03:17<09:16,  6.05s/it]
Training Epoch: 9/100, batch 64/250 completed (loss: 0.6388:  26%|██▋       | 33/125 [03:17<09:16,  6.05s/it]
Training Epoch: 9/100, batch 64/250 completed (loss: 0.0536:  26%|██▋       | 33/125 [03:17<09:16,  6.05s/it]
Training Epoch: 9/100, batch 65/250 completed (loss: 0.1483:  26%|██▋       | 33/125 [03:17<09:16,  6.05s/it]
Training Epoch: 9/100, batch 65/250 completed (loss: 0.1279:  26%|██▋       | 33/125 [03:17<09:16,  6.05s/it]
Training Epoch: 9/100, batch 65/250 completed (loss: 0.1688:  26%|██▋       | 33/125 [03:17<09:16,  6.05s/it]
Training Epoch: 9/100, batch 65/250 completed (loss: 0.4219:  26%|██▋       | 33/125 [03:17<09:16,  6.05s/it]
Training Epoch: 9/100, batch 66/250 completed (loss: 0.1512:  26%|██▋       | 33/125 [03:20<09:16,  6.05s/it]
Training Epoch: 9/100, batch 66/250 completed (loss: 0.0625:  26%|██▋       | 33/125 [03:20<09:16,  6.05s/it]
Training Epoch: 9/100, batch 66/250 completed (loss: 0.2487:  26%|██▋       | 33/125 [03:20<09:16,  6.05s/it]
Training Epoch: 9/100, batch 66/250 completed (loss: 0.2032:  26%|██▋       | 33/125 [03:20<09:16,  6.05s/it]
Training Epoch: 9/100, batch 66/250 completed (loss: 0.1512:  27%|██▋       | 34/125 [03:23<09:08,  6.03s/it]
Training Epoch: 9/100, batch 66/250 completed (loss: 0.0625:  27%|██▋       | 34/125 [03:23<09:08,  6.03s/it]
Training Epoch: 9/100, batch 66/250 completed (loss: 0.2032:  27%|██▋       | 34/125 [03:23<09:08,  6.03s/it]
Training Epoch: 9/100, batch 66/250 completed (loss: 0.2487:  27%|██▋       | 34/125 [03:23<09:08,  6.03s/it]
Training Epoch: 9/100, batch 67/250 completed (loss: 0.2667:  27%|██▋       | 34/125 [03:23<09:08,  6.03s/it]
Training Epoch: 9/100, batch 67/250 completed (loss: 0.0484:  27%|██▋       | 34/125 [03:23<09:08,  6.03s/it]
Training Epoch: 9/100, batch 67/250 completed (loss: 0.0639:  27%|██▋       | 34/125 [03:23<09:08,  6.03s/it]
Training Epoch: 9/100, batch 67/250 completed (loss: 0.0803:  27%|██▋       | 34/125 [03:23<09:08,  6.03s/it]
Training Epoch: 9/100, batch 68/250 completed (loss: 0.1021:  27%|██▋       | 34/125 [03:26<09:08,  6.03s/it]
Training Epoch: 9/100, batch 68/250 completed (loss: 0.0601:  27%|██▋       | 34/125 [03:26<09:08,  6.03s/it]
Training Epoch: 9/100, batch 68/250 completed (loss: 0.0802:  27%|██▋       | 34/125 [03:26<09:08,  6.03s/it]
Training Epoch: 9/100, batch 68/250 completed (loss: 0.0883:  27%|██▋       | 34/125 [03:26<09:08,  6.03s/it]
Training Epoch: 9/100, batch 68/250 completed (loss: 0.0802:  28%|██▊       | 35/125 [03:28<08:56,  5.96s/it]
Training Epoch: 9/100, batch 68/250 completed (loss: 0.0601:  28%|██▊       | 35/125 [03:28<08:56,  5.96s/it]
Training Epoch: 9/100, batch 68/250 completed (loss: 0.0883:  28%|██▊       | 35/125 [03:28<08:56,  5.96s/it]
Training Epoch: 9/100, batch 68/250 completed (loss: 0.1021:  28%|██▊       | 35/125 [03:28<08:56,  5.96s/it]
Training Epoch: 9/100, batch 69/250 completed (loss: 0.1047:  28%|██▊       | 35/125 [03:28<08:56,  5.96s/it]
Training Epoch: 9/100, batch 69/250 completed (loss: 0.0677:  28%|██▊       | 35/125 [03:28<08:56,  5.96s/it]
Training Epoch: 9/100, batch 69/250 completed (loss: 0.3019:  28%|██▊       | 35/125 [03:28<08:56,  5.96s/it]
Training Epoch: 9/100, batch 69/250 completed (loss: 0.1886:  28%|██▊       | 35/125 [03:28<08:56,  5.96s/it]
Training Epoch: 9/100, batch 70/250 completed (loss: 0.0819:  28%|██▊       | 35/125 [03:32<08:56,  5.96s/it]
Training Epoch: 9/100, batch 70/250 completed (loss: 0.2516:  28%|██▊       | 35/125 [03:32<08:56,  5.96s/it]
Training Epoch: 9/100, batch 70/250 completed (loss: 0.1476:  28%|██▊       | 35/125 [03:32<08:56,  5.96s/it]
Training Epoch: 9/100, batch 70/250 completed (loss: 0.0452:  28%|██▊       | 35/125 [03:32<08:56,  5.96s/it]
Training Epoch: 9/100, batch 70/250 completed (loss: 0.1476:  29%|██▉       | 36/125 [03:35<09:03,  6.11s/it]
Training Epoch: 9/100, batch 70/250 completed (loss: 0.0452:  29%|██▉       | 36/125 [03:35<09:03,  6.11s/it]
Training Epoch: 9/100, batch 70/250 completed (loss: 0.2516:  29%|██▉       | 36/125 [03:35<09:03,  6.11s/it]
Training Epoch: 9/100, batch 70/250 completed (loss: 0.0819:  29%|██▉       | 36/125 [03:35<09:03,  6.11s/it]
Training Epoch: 9/100, batch 71/250 completed (loss: 0.0618:  29%|██▉       | 36/125 [03:35<09:03,  6.11s/it]
Training Epoch: 9/100, batch 71/250 completed (loss: 0.386:  29%|██▉       | 36/125 [03:35<09:03,  6.11s/it] 
Training Epoch: 9/100, batch 71/250 completed (loss: 0.2404:  29%|██▉       | 36/125 [03:35<09:03,  6.11s/it]
Training Epoch: 9/100, batch 71/250 completed (loss: 0.0474:  29%|██▉       | 36/125 [03:35<09:03,  6.11s/it]
Training Epoch: 9/100, batch 72/250 completed (loss: 0.1704:  29%|██▉       | 36/125 [03:38<09:03,  6.11s/it]
Training Epoch: 9/100, batch 72/250 completed (loss: 0.1939:  29%|██▉       | 36/125 [03:38<09:03,  6.11s/it]
Training Epoch: 9/100, batch 72/250 completed (loss: 0.118:  29%|██▉       | 36/125 [03:38<09:03,  6.11s/it] 
Training Epoch: 9/100, batch 72/250 completed (loss: 0.1547:  29%|██▉       | 36/125 [03:38<09:03,  6.11s/it]
Training Epoch: 9/100, batch 72/250 completed (loss: 0.1939:  30%|██▉       | 37/125 [03:41<08:50,  6.03s/it]
Training Epoch: 9/100, batch 72/250 completed (loss: 0.1704:  30%|██▉       | 37/125 [03:41<08:50,  6.03s/it]
Training Epoch: 9/100, batch 72/250 completed (loss: 0.118:  30%|██▉       | 37/125 [03:41<08:50,  6.03s/it]
Training Epoch: 9/100, batch 72/250 completed (loss: 0.1547:  30%|██▉       | 37/125 [03:41<08:50,  6.03s/it]
Training Epoch: 9/100, batch 73/250 completed (loss: 0.2554:  30%|██▉       | 37/125 [03:41<08:50,  6.03s/it]
Training Epoch: 9/100, batch 73/250 completed (loss: 0.3477:  30%|██▉       | 37/125 [03:41<08:50,  6.03s/it]
Training Epoch: 9/100, batch 73/250 completed (loss: 0.1206:  30%|██▉       | 37/125 [03:41<08:50,  6.03s/it]
Training Epoch: 9/100, batch 73/250 completed (loss: 0.133:  30%|██▉       | 37/125 [03:41<08:50,  6.03s/it] 
Training Epoch: 9/100, batch 74/250 completed (loss: 0.1474:  30%|██▉       | 37/125 [03:44<08:50,  6.03s/it]
Training Epoch: 9/100, batch 74/250 completed (loss: 0.0832:  30%|██▉       | 37/125 [03:44<08:50,  6.03s/it]
Training Epoch: 9/100, batch 74/250 completed (loss: 0.0474:  30%|██▉       | 37/125 [03:44<08:50,  6.03s/it]
Training Epoch: 9/100, batch 74/250 completed (loss: 0.2034:  30%|██▉       | 37/125 [03:44<08:50,  6.03s/it]
Training Epoch: 9/100, batch 74/250 completed (loss: 0.0474:  30%|███       | 38/125 [03:47<08:42,  6.00s/it]
Training Epoch: 9/100, batch 74/250 completed (loss: 0.1474:  30%|███       | 38/125 [03:47<08:42,  6.00s/it]
Training Epoch: 9/100, batch 74/250 completed (loss: 0.2034:  30%|███       | 38/125 [03:47<08:42,  6.00s/it]
Training Epoch: 9/100, batch 74/250 completed (loss: 0.0832:  30%|███       | 38/125 [03:47<08:42,  6.00s/it]
Training Epoch: 9/100, batch 75/250 completed (loss: 0.0995:  30%|███       | 38/125 [03:47<08:42,  6.00s/it]
Training Epoch: 9/100, batch 75/250 completed (loss: 0.1305:  30%|███       | 38/125 [03:47<08:42,  6.00s/it]
Training Epoch: 9/100, batch 75/250 completed (loss: 0.1609:  30%|███       | 38/125 [03:47<08:42,  6.00s/it]
Training Epoch: 9/100, batch 75/250 completed (loss: 0.0152:  30%|███       | 38/125 [03:47<08:42,  6.00s/it]
Training Epoch: 9/100, batch 76/250 completed (loss: 0.0446:  30%|███       | 38/125 [03:50<08:42,  6.00s/it]
Training Epoch: 9/100, batch 76/250 completed (loss: 0.18:  30%|███       | 38/125 [03:50<08:42,  6.00s/it]  
Training Epoch: 9/100, batch 76/250 completed (loss: 0.2714:  30%|███       | 38/125 [03:50<08:42,  6.00s/it]
Training Epoch: 9/100, batch 76/250 completed (loss: 0.0742:  30%|███       | 38/125 [03:50<08:42,  6.00s/it]
Training Epoch: 9/100, batch 76/250 completed (loss: 0.18:  31%|███       | 39/125 [03:53<08:44,  6.09s/it]
Training Epoch: 9/100, batch 76/250 completed (loss: 0.2714:  31%|███       | 39/125 [03:53<08:44,  6.10s/it]
Training Epoch: 9/100, batch 76/250 completed (loss: 0.0446:  31%|███       | 39/125 [03:53<08:44,  6.10s/it]
Training Epoch: 9/100, batch 76/250 completed (loss: 0.0742:  31%|███       | 39/125 [03:53<08:44,  6.10s/it]
Training Epoch: 9/100, batch 77/250 completed (loss: 0.0749:  31%|███       | 39/125 [03:53<08:44,  6.10s/it]
Training Epoch: 9/100, batch 77/250 completed (loss: 0.258:  31%|███       | 39/125 [03:53<08:44,  6.10s/it] 
Training Epoch: 9/100, batch 77/250 completed (loss: 0.3328:  31%|███       | 39/125 [03:53<08:44,  6.09s/it]
Training Epoch: 9/100, batch 77/250 completed (loss: 0.1066:  31%|███       | 39/125 [03:53<08:44,  6.10s/it]
Training Epoch: 9/100, batch 78/250 completed (loss: 0.0823:  31%|███       | 39/125 [03:56<08:44,  6.09s/it]
Training Epoch: 9/100, batch 78/250 completed (loss: 0.2614:  31%|███       | 39/125 [03:56<08:44,  6.10s/it]
Training Epoch: 9/100, batch 78/250 completed (loss: 0.0527:  31%|███       | 39/125 [03:56<08:44,  6.10s/it]
Training Epoch: 9/100, batch 78/250 completed (loss: 0.0888:  31%|███       | 39/125 [03:56<08:44,  6.10s/it]
Training Epoch: 9/100, batch 78/250 completed (loss: 0.0823:  32%|███▏      | 40/125 [03:59<08:31,  6.02s/it]
Training Epoch: 9/100, batch 78/250 completed (loss: 0.0527:  32%|███▏      | 40/125 [03:59<08:31,  6.02s/it]
Training Epoch: 9/100, batch 78/250 completed (loss: 0.2614:  32%|███▏      | 40/125 [03:59<08:31,  6.02s/it]
Training Epoch: 9/100, batch 78/250 completed (loss: 0.0888:  32%|███▏      | 40/125 [03:59<08:31,  6.02s/it]
Training Epoch: 9/100, batch 79/250 completed (loss: 0.1408:  32%|███▏      | 40/125 [03:59<08:31,  6.02s/it]
Training Epoch: 9/100, batch 79/250 completed (loss: 0.0772:  32%|███▏      | 40/125 [03:59<08:31,  6.02s/it]
Training Epoch: 9/100, batch 79/250 completed (loss: 0.1813:  32%|███▏      | 40/125 [03:59<08:31,  6.02s/it]
Training Epoch: 9/100, batch 79/250 completed (loss: 0.0649:  32%|███▏      | 40/125 [03:59<08:31,  6.02s/it]
Training Epoch: 9/100, batch 80/250 completed (loss: 0.1157:  32%|███▏      | 40/125 [04:02<08:31,  6.02s/it]
Training Epoch: 9/100, batch 80/250 completed (loss: 0.1273:  32%|███▏      | 40/125 [04:02<08:31,  6.02s/it]
Training Epoch: 9/100, batch 80/250 completed (loss: 0.0377:  32%|███▏      | 40/125 [04:02<08:31,  6.02s/it]
Training Epoch: 9/100, batch 80/250 completed (loss: 0.0309:  32%|███▏      | 40/125 [04:02<08:31,  6.02s/it]
Training Epoch: 9/100, batch 80/250 completed (loss: 0.0377:  33%|███▎      | 41/125 [04:05<08:35,  6.14s/it]
Training Epoch: 9/100, batch 80/250 completed (loss: 0.1157:  33%|███▎      | 41/125 [04:05<08:35,  6.14s/it]
Training Epoch: 9/100, batch 80/250 completed (loss: 0.1273:  33%|███▎      | 41/125 [04:05<08:35,  6.14s/it]
Training Epoch: 9/100, batch 80/250 completed (loss: 0.0309:  33%|███▎      | 41/125 [04:05<08:35,  6.14s/it]
Training Epoch: 9/100, batch 81/250 completed (loss: 0.144:  33%|███▎      | 41/125 [04:05<08:35,  6.14s/it] 
Training Epoch: 9/100, batch 81/250 completed (loss: 0.0758:  33%|███▎      | 41/125 [04:05<08:35,  6.14s/it]
Training Epoch: 9/100, batch 81/250 completed (loss: 0.1404:  33%|███▎      | 41/125 [04:05<08:35,  6.14s/it]
Training Epoch: 9/100, batch 81/250 completed (loss: 0.2262:  33%|███▎      | 41/125 [04:05<08:35,  6.14s/it]
Training Epoch: 9/100, batch 82/250 completed (loss: 0.1704:  33%|███▎      | 41/125 [04:08<08:35,  6.14s/it]
Training Epoch: 9/100, batch 82/250 completed (loss: 0.1894:  33%|███▎      | 41/125 [04:08<08:35,  6.14s/it]
Training Epoch: 9/100, batch 82/250 completed (loss: 0.2211:  33%|███▎      | 41/125 [04:08<08:35,  6.14s/it]
Training Epoch: 9/100, batch 82/250 completed (loss: 0.1942:  33%|███▎      | 41/125 [04:08<08:35,  6.14s/it]
Training Epoch: 9/100, batch 82/250 completed (loss: 0.2211:  34%|███▎      | 42/125 [04:11<08:22,  6.05s/it]
Training Epoch: 9/100, batch 82/250 completed (loss: 0.1894:  34%|███▎      | 42/125 [04:11<08:22,  6.05s/it]
Training Epoch: 9/100, batch 82/250 completed (loss: 0.1942:  34%|███▎      | 42/125 [04:11<08:22,  6.05s/it]
Training Epoch: 9/100, batch 82/250 completed (loss: 0.1704:  34%|███▎      | 42/125 [04:11<08:22,  6.05s/it]
Training Epoch: 9/100, batch 83/250 completed (loss: 0.0667:  34%|███▎      | 42/125 [04:11<08:22,  6.05s/it]
Training Epoch: 9/100, batch 83/250 completed (loss: 0.1787:  34%|███▎      | 42/125 [04:11<08:22,  6.05s/it]
Training Epoch: 9/100, batch 83/250 completed (loss: 0.2215:  34%|███▎      | 42/125 [04:11<08:22,  6.05s/it]
Training Epoch: 9/100, batch 83/250 completed (loss: 0.1492:  34%|███▎      | 42/125 [04:11<08:22,  6.05s/it]
Training Epoch: 9/100, batch 84/250 completed (loss: 0.0582:  34%|███▎      | 42/125 [04:14<08:22,  6.05s/it]
Training Epoch: 9/100, batch 84/250 completed (loss: 0.4013:  34%|███▎      | 42/125 [04:14<08:22,  6.05s/it]
Training Epoch: 9/100, batch 84/250 completed (loss: 0.33:  34%|███▎      | 42/125 [04:14<08:22,  6.05s/it]  
Training Epoch: 9/100, batch 84/250 completed (loss: 0.0627:  34%|███▎      | 42/125 [04:14<08:22,  6.05s/it]
Training Epoch: 9/100, batch 84/250 completed (loss: 0.0582:  34%|███▍      | 43/125 [04:17<08:10,  5.98s/it]
Training Epoch: 9/100, batch 84/250 completed (loss: 0.33:  34%|███▍      | 43/125 [04:17<08:10,  5.98s/it]
Training Epoch: 9/100, batch 84/250 completed (loss: 0.4013:  34%|███▍      | 43/125 [04:17<08:10,  5.98s/it]
Training Epoch: 9/100, batch 84/250 completed (loss: 0.0627:  34%|███▍      | 43/125 [04:17<08:10,  5.98s/it]
Training Epoch: 9/100, batch 85/250 completed (loss: 0.158:  34%|███▍      | 43/125 [04:17<08:10,  5.98s/it] 
Training Epoch: 9/100, batch 85/250 completed (loss: 0.2457:  34%|███▍      | 43/125 [04:17<08:10,  5.98s/it]
Training Epoch: 9/100, batch 85/250 completed (loss: 0.0331:  34%|███▍      | 43/125 [04:17<08:10,  5.98s/it]
Training Epoch: 9/100, batch 85/250 completed (loss: 0.1874:  34%|███▍      | 43/125 [04:17<08:10,  5.98s/it]
Training Epoch: 9/100, batch 86/250 completed (loss: 0.2524:  34%|███▍      | 43/125 [04:20<08:10,  5.98s/it]
Training Epoch: 9/100, batch 86/250 completed (loss: 0.2503:  34%|███▍      | 43/125 [04:20<08:10,  5.98s/it]
Training Epoch: 9/100, batch 86/250 completed (loss: 0.223:  34%|███▍      | 43/125 [04:20<08:10,  5.98s/it] 
Training Epoch: 9/100, batch 86/250 completed (loss: 0.1024:  34%|███▍      | 43/125 [04:20<08:10,  5.98s/it]
Training Epoch: 9/100, batch 86/250 completed (loss: 0.2503:  35%|███▌      | 44/125 [04:23<08:15,  6.12s/it]
Training Epoch: 9/100, batch 86/250 completed (loss: 0.223:  35%|███▌      | 44/125 [04:23<08:15,  6.12s/it]
Training Epoch: 9/100, batch 86/250 completed (loss: 0.2524:  35%|███▌      | 44/125 [04:23<08:15,  6.12s/it]
Training Epoch: 9/100, batch 86/250 completed (loss: 0.1024:  35%|███▌      | 44/125 [04:23<08:15,  6.12s/it]
Training Epoch: 9/100, batch 87/250 completed (loss: 0.1102:  35%|███▌      | 44/125 [04:23<08:15,  6.12s/it]
Training Epoch: 9/100, batch 87/250 completed (loss: 0.0761:  35%|███▌      | 44/125 [04:23<08:15,  6.12s/it]
Training Epoch: 9/100, batch 87/250 completed (loss: 0.1649:  35%|███▌      | 44/125 [04:23<08:15,  6.12s/it]
Training Epoch: 9/100, batch 87/250 completed (loss: 0.1259:  35%|███▌      | 44/125 [04:23<08:15,  6.12s/it]
Training Epoch: 9/100, batch 88/250 completed (loss: 0.1189:  35%|███▌      | 44/125 [04:26<08:15,  6.12s/it]
Training Epoch: 9/100, batch 88/250 completed (loss: 0.1192:  35%|███▌      | 44/125 [04:26<08:15,  6.12s/it]
Training Epoch: 9/100, batch 88/250 completed (loss: 0.1784:  35%|███▌      | 44/125 [04:26<08:15,  6.12s/it]
Training Epoch: 9/100, batch 88/250 completed (loss: 0.0709:  35%|███▌      | 44/125 [04:26<08:15,  6.12s/it]
Training Epoch: 9/100, batch 88/250 completed (loss: 0.1192:  36%|███▌      | 45/125 [04:29<08:03,  6.04s/it]
Training Epoch: 9/100, batch 88/250 completed (loss: 0.1189:  36%|███▌      | 45/125 [04:29<08:03,  6.04s/it]
Training Epoch: 9/100, batch 88/250 completed (loss: 0.0709:  36%|███▌      | 45/125 [04:29<08:03,  6.04s/it]
Training Epoch: 9/100, batch 88/250 completed (loss: 0.1784:  36%|███▌      | 45/125 [04:29<08:03,  6.05s/it]
Training Epoch: 9/100, batch 89/250 completed (loss: 0.1316:  36%|███▌      | 45/125 [04:29<08:03,  6.04s/it]
Training Epoch: 9/100, batch 89/250 completed (loss: 0.0467:  36%|███▌      | 45/125 [04:29<08:03,  6.04s/it]
Training Epoch: 9/100, batch 89/250 completed (loss: 0.2339:  36%|███▌      | 45/125 [04:29<08:03,  6.05s/it]
Training Epoch: 9/100, batch 89/250 completed (loss: 0.1405:  36%|███▌      | 45/125 [04:29<08:03,  6.04s/it]
Training Epoch: 9/100, batch 90/250 completed (loss: 0.1604:  36%|███▌      | 45/125 [04:32<08:03,  6.04s/it]
Training Epoch: 9/100, batch 90/250 completed (loss: 0.2524:  36%|███▌      | 45/125 [04:32<08:03,  6.04s/it]
Training Epoch: 9/100, batch 90/250 completed (loss: 0.4392:  36%|███▌      | 45/125 [04:32<08:03,  6.05s/it]
Training Epoch: 9/100, batch 90/250 completed (loss: 0.0571:  36%|███▌      | 45/125 [04:32<08:03,  6.04s/it]
Training Epoch: 9/100, batch 90/250 completed (loss: 0.4392:  37%|███▋      | 46/125 [04:35<07:55,  6.02s/it]
Training Epoch: 9/100, batch 90/250 completed (loss: 0.1604:  37%|███▋      | 46/125 [04:35<07:55,  6.02s/it]
Training Epoch: 9/100, batch 90/250 completed (loss: 0.2524:  37%|███▋      | 46/125 [04:35<07:55,  6.02s/it]
Training Epoch: 9/100, batch 90/250 completed (loss: 0.0571:  37%|███▋      | 46/125 [04:35<07:55,  6.02s/it]
Training Epoch: 9/100, batch 91/250 completed (loss: 0.3689:  37%|███▋      | 46/125 [04:35<07:55,  6.02s/it]
Training Epoch: 9/100, batch 91/250 completed (loss: 0.0966:  37%|███▋      | 46/125 [04:35<07:55,  6.02s/it]
Training Epoch: 9/100, batch 91/250 completed (loss: 0.2351:  37%|███▋      | 46/125 [04:35<07:55,  6.02s/it]
Training Epoch: 9/100, batch 91/250 completed (loss: 0.2752:  37%|███▋      | 46/125 [04:35<07:55,  6.02s/it]
Training Epoch: 9/100, batch 92/250 completed (loss: 0.0831:  37%|███▋      | 46/125 [04:39<07:55,  6.02s/it]
Training Epoch: 9/100, batch 92/250 completed (loss: 0.0421:  37%|███▋      | 46/125 [04:39<07:55,  6.02s/it]
Training Epoch: 9/100, batch 92/250 completed (loss: 0.1652:  37%|███▋      | 46/125 [04:39<07:55,  6.02s/it]
Training Epoch: 9/100, batch 92/250 completed (loss: 0.1923:  37%|███▋      | 46/125 [04:39<07:55,  6.02s/it]
Training Epoch: 9/100, batch 92/250 completed (loss: 0.0421:  38%|███▊      | 47/125 [04:41<07:55,  6.10s/it]
Training Epoch: 9/100, batch 92/250 completed (loss: 0.0831:  38%|███▊      | 47/125 [04:41<07:55,  6.10s/it]
Training Epoch: 9/100, batch 92/250 completed (loss: 0.1652:  38%|███▊      | 47/125 [04:41<07:55,  6.10s/it]
Training Epoch: 9/100, batch 92/250 completed (loss: 0.1923:  38%|███▊      | 47/125 [04:41<07:55,  6.10s/it]
Training Epoch: 9/100, batch 93/250 completed (loss: 0.1094:  38%|███▊      | 47/125 [04:41<07:55,  6.10s/it]
Training Epoch: 9/100, batch 93/250 completed (loss: 0.3191:  38%|███▊      | 47/125 [04:41<07:55,  6.10s/it]
Training Epoch: 9/100, batch 93/250 completed (loss: 0.0733:  38%|███▊      | 47/125 [04:41<07:55,  6.10s/it]
Training Epoch: 9/100, batch 93/250 completed (loss: 0.0481:  38%|███▊      | 47/125 [04:42<07:55,  6.10s/it]
Training Epoch: 9/100, batch 94/250 completed (loss: 0.119:  38%|███▊      | 47/125 [04:44<07:55,  6.10s/it] 
Training Epoch: 9/100, batch 94/250 completed (loss: 0.1727:  38%|███▊      | 47/125 [04:44<07:55,  6.10s/it]
Training Epoch: 9/100, batch 94/250 completed (loss: 0.1166:  38%|███▊      | 47/125 [04:44<07:55,  6.10s/it]
Training Epoch: 9/100, batch 94/250 completed (loss: 0.0919:  38%|███▊      | 47/125 [04:44<07:55,  6.10s/it]
Training Epoch: 9/100, batch 94/250 completed (loss: 0.1166:  38%|███▊      | 48/125 [04:47<07:43,  6.02s/it]
Training Epoch: 9/100, batch 94/250 completed (loss: 0.119:  38%|███▊      | 48/125 [04:47<07:43,  6.03s/it]
Training Epoch: 9/100, batch 94/250 completed (loss: 0.0919:  38%|███▊      | 48/125 [04:47<07:43,  6.03s/it]
Training Epoch: 9/100, batch 94/250 completed (loss: 0.1727:  38%|███▊      | 48/125 [04:47<07:44,  6.03s/it]
Training Epoch: 9/100, batch 95/250 completed (loss: 0.1033:  38%|███▊      | 48/125 [04:47<07:44,  6.03s/it]
Training Epoch: 9/100, batch 95/250 completed (loss: 0.3375:  38%|███▊      | 48/125 [04:47<07:43,  6.02s/it]
Training Epoch: 9/100, batch 95/250 completed (loss: 0.1589:  38%|███▊      | 48/125 [04:47<07:43,  6.03s/it]
Training Epoch: 9/100, batch 95/250 completed (loss: 0.2889:  38%|███▊      | 48/125 [04:47<07:43,  6.03s/it]
Training Epoch: 9/100, batch 96/250 completed (loss: 0.0856:  38%|███▊      | 48/125 [04:50<07:43,  6.02s/it]
Training Epoch: 9/100, batch 96/250 completed (loss: 0.1613:  38%|███▊      | 48/125 [04:50<07:44,  6.03s/it]
Training Epoch: 9/100, batch 96/250 completed (loss: 0.0606:  38%|███▊      | 48/125 [04:50<07:43,  6.03s/it]
Training Epoch: 9/100, batch 96/250 completed (loss: 0.3012:  38%|███▊      | 48/125 [04:50<07:43,  6.03s/it]
Training Epoch: 9/100, batch 96/250 completed (loss: 0.0856:  39%|███▉      | 49/125 [04:53<07:40,  6.06s/it]
Training Epoch: 9/100, batch 96/250 completed (loss: 0.0606:  39%|███▉      | 49/125 [04:53<07:40,  6.05s/it]
Training Epoch: 9/100, batch 96/250 completed (loss: 0.1613:  39%|███▉      | 49/125 [04:53<07:40,  6.05s/it]
Training Epoch: 9/100, batch 96/250 completed (loss: 0.3012:  39%|███▉      | 49/125 [04:53<07:40,  6.05s/it]
Training Epoch: 9/100, batch 97/250 completed (loss: 0.1856:  39%|███▉      | 49/125 [04:53<07:40,  6.05s/it]
Training Epoch: 9/100, batch 97/250 completed (loss: 0.2028:  39%|███▉      | 49/125 [04:53<07:40,  6.06s/it]
Training Epoch: 9/100, batch 97/250 completed (loss: 0.0274:  39%|███▉      | 49/125 [04:53<07:40,  6.05s/it]
Training Epoch: 9/100, batch 97/250 completed (loss: 0.2431:  39%|███▉      | 49/125 [04:53<07:40,  6.05s/it]
Training Epoch: 9/100, batch 98/250 completed (loss: 0.1973:  39%|███▉      | 49/125 [04:57<07:40,  6.05s/it]
Training Epoch: 9/100, batch 98/250 completed (loss: 0.0824:  39%|███▉      | 49/125 [04:57<07:40,  6.05s/it]
Training Epoch: 9/100, batch 98/250 completed (loss: 0.0726:  39%|███▉      | 49/125 [04:57<07:40,  6.06s/it]
Training Epoch: 9/100, batch 98/250 completed (loss: 0.0263:  39%|███▉      | 49/125 [04:57<07:40,  6.05s/it]
Training Epoch: 9/100, batch 98/250 completed (loss: 0.1973:  40%|████      | 50/125 [05:00<07:35,  6.07s/it]
Training Epoch: 9/100, batch 98/250 completed (loss: 0.0726:  40%|████      | 50/125 [05:00<07:35,  6.07s/it]
Training Epoch: 9/100, batch 98/250 completed (loss: 0.0824:  40%|████      | 50/125 [05:00<07:35,  6.07s/it]
Training Epoch: 9/100, batch 98/250 completed (loss: 0.0263:  40%|████      | 50/125 [05:00<07:35,  6.07s/it]
Training Epoch: 9/100, batch 99/250 completed (loss: 0.0545:  40%|████      | 50/125 [05:00<07:35,  6.07s/it]
Training Epoch: 9/100, batch 99/250 completed (loss: 0.1575:  40%|████      | 50/125 [05:00<07:35,  6.07s/it]
Training Epoch: 9/100, batch 99/250 completed (loss: 0.0721:  40%|████      | 50/125 [05:00<07:35,  6.07s/it]
Training Epoch: 9/100, batch 99/250 completed (loss: 0.0913:  40%|████      | 50/125 [05:00<07:35,  6.07s/it]
Training Epoch: 9/100, batch 100/250 completed (loss: 0.1738:  40%|████      | 50/125 [05:03<07:35,  6.07s/it]
Training Epoch: 9/100, batch 100/250 completed (loss: 0.12:  40%|████      | 50/125 [05:03<07:35,  6.07s/it] 
Training Epoch: 9/100, batch 100/250 completed (loss: 0.0327:  40%|████      | 50/125 [05:03<07:35,  6.07s/it]
Training Epoch: 9/100, batch 100/250 completed (loss: 0.0558:  40%|████      | 50/125 [05:03<07:35,  6.07s/it]
Training Epoch: 9/100, batch 100/250 completed (loss: 0.12:  41%|████      | 51/125 [05:06<07:27,  6.04s/it]
Training Epoch: 9/100, batch 100/250 completed (loss: 0.1738:  41%|████      | 51/125 [05:06<07:27,  6.04s/it]
Training Epoch: 9/100, batch 100/250 completed (loss: 0.0558:  41%|████      | 51/125 [05:06<07:27,  6.04s/it]
Training Epoch: 9/100, batch 100/250 completed (loss: 0.0327:  41%|████      | 51/125 [05:06<07:27,  6.04s/it]
Training Epoch: 9/100, batch 101/250 completed (loss: 0.0776:  41%|████      | 51/125 [05:06<07:27,  6.04s/it]
Training Epoch: 9/100, batch 101/250 completed (loss: 0.0849:  41%|████      | 51/125 [05:06<07:27,  6.04s/it]
Training Epoch: 9/100, batch 101/250 completed (loss: 0.1049:  41%|████      | 51/125 [05:06<07:27,  6.04s/it]
Training Epoch: 9/100, batch 101/250 completed (loss: 0.1167:  41%|████      | 51/125 [05:06<07:27,  6.04s/it]
Training Epoch: 9/100, batch 102/250 completed (loss: 0.232:  41%|████      | 51/125 [05:09<07:27,  6.04s/it] 
Training Epoch: 9/100, batch 102/250 completed (loss: 0.1585:  41%|████      | 51/125 [05:09<07:27,  6.04s/it]
Training Epoch: 9/100, batch 102/250 completed (loss: 0.0908:  41%|████      | 51/125 [05:09<07:27,  6.04s/it]
Training Epoch: 9/100, batch 102/250 completed (loss: 0.3188:  41%|████      | 51/125 [05:09<07:27,  6.04s/it]
Training Epoch: 9/100, batch 102/250 completed (loss: 0.232:  42%|████▏     | 52/125 [05:12<07:25,  6.10s/it]
Training Epoch: 9/100, batch 102/250 completed (loss: 0.3188:  42%|████▏     | 52/125 [05:12<07:25,  6.10s/it]
Training Epoch: 9/100, batch 102/250 completed (loss: 0.1585:  42%|████▏     | 52/125 [05:12<07:25,  6.10s/it]
Training Epoch: 9/100, batch 102/250 completed (loss: 0.0908:  42%|████▏     | 52/125 [05:12<07:25,  6.11s/it]
Training Epoch: 9/100, batch 103/250 completed (loss: 0.2286:  42%|████▏     | 52/125 [05:12<07:25,  6.10s/it]
Training Epoch: 9/100, batch 103/250 completed (loss: 0.1109:  42%|████▏     | 52/125 [05:12<07:25,  6.11s/it]
Training Epoch: 9/100, batch 103/250 completed (loss: 0.1981:  42%|████▏     | 52/125 [05:12<07:25,  6.10s/it]
Training Epoch: 9/100, batch 103/250 completed (loss: 0.1956:  42%|████▏     | 52/125 [05:12<07:25,  6.10s/it]
Training Epoch: 9/100, batch 104/250 completed (loss: 0.1614:  42%|████▏     | 52/125 [05:15<07:25,  6.11s/it]
Training Epoch: 9/100, batch 104/250 completed (loss: 0.0276:  42%|████▏     | 52/125 [05:15<07:25,  6.10s/it]
Training Epoch: 9/100, batch 104/250 completed (loss: 0.0738:  42%|████▏     | 52/125 [05:15<07:25,  6.10s/it]
Training Epoch: 9/100, batch 104/250 completed (loss: 0.0563:  42%|████▏     | 52/125 [05:15<07:25,  6.10s/it]
Training Epoch: 9/100, batch 104/250 completed (loss: 0.1614:  42%|████▏     | 53/125 [05:18<07:14,  6.03s/it]
Training Epoch: 9/100, batch 104/250 completed (loss: 0.0738:  42%|████▏     | 53/125 [05:18<07:14,  6.03s/it]
Training Epoch: 9/100, batch 104/250 completed (loss: 0.0276:  42%|████▏     | 53/125 [05:18<07:14,  6.03s/it]
Training Epoch: 9/100, batch 104/250 completed (loss: 0.0563:  42%|████▏     | 53/125 [05:18<07:14,  6.03s/it]
Training Epoch: 9/100, batch 105/250 completed (loss: 0.1854:  42%|████▏     | 53/125 [05:18<07:14,  6.03s/it]
Training Epoch: 9/100, batch 105/250 completed (loss: 0.0538:  42%|████▏     | 53/125 [05:18<07:14,  6.03s/it]
Training Epoch: 9/100, batch 105/250 completed (loss: 0.1064:  42%|████▏     | 53/125 [05:18<07:14,  6.03s/it]
Training Epoch: 9/100, batch 105/250 completed (loss: 0.0715:  42%|████▏     | 53/125 [05:18<07:14,  6.03s/it]
Training Epoch: 9/100, batch 106/250 completed (loss: 0.1573:  42%|████▏     | 53/125 [05:21<07:14,  6.03s/it]
Training Epoch: 9/100, batch 106/250 completed (loss: 0.0766:  42%|████▏     | 53/125 [05:21<07:14,  6.03s/it]
Training Epoch: 9/100, batch 106/250 completed (loss: 0.2495:  42%|████▏     | 53/125 [05:21<07:14,  6.03s/it]
Training Epoch: 9/100, batch 106/250 completed (loss: 0.2375:  42%|████▏     | 53/125 [05:21<07:14,  6.03s/it]
Training Epoch: 9/100, batch 106/250 completed (loss: 0.0766:  43%|████▎     | 54/125 [05:24<07:10,  6.06s/it]
Training Epoch: 9/100, batch 106/250 completed (loss: 0.1573:  43%|████▎     | 54/125 [05:24<07:10,  6.06s/it]
Training Epoch: 9/100, batch 106/250 completed (loss: 0.2495:  43%|████▎     | 54/125 [05:24<07:10,  6.07s/it]
Training Epoch: 9/100, batch 106/250 completed (loss: 0.2375:  43%|████▎     | 54/125 [05:24<07:10,  6.07s/it]
Training Epoch: 9/100, batch 107/250 completed (loss: 0.0862:  43%|████▎     | 54/125 [05:24<07:10,  6.06s/it]
Training Epoch: 9/100, batch 107/250 completed (loss: 0.0448:  43%|████▎     | 54/125 [05:24<07:10,  6.06s/it]
Training Epoch: 9/100, batch 107/250 completed (loss: 0.208:  43%|████▎     | 54/125 [05:24<07:10,  6.07s/it] 
Training Epoch: 9/100, batch 107/250 completed (loss: 0.0572:  43%|████▎     | 54/125 [05:24<07:10,  6.07s/it]
Training Epoch: 9/100, batch 108/250 completed (loss: 0.1734:  43%|████▎     | 54/125 [05:27<07:10,  6.06s/it]
Training Epoch: 9/100, batch 108/250 completed (loss: 0.1255:  43%|████▎     | 54/125 [05:27<07:10,  6.06s/it]
Training Epoch: 9/100, batch 108/250 completed (loss: 0.0745:  43%|████▎     | 54/125 [05:27<07:10,  6.07s/it]
Training Epoch: 9/100, batch 108/250 completed (loss: 0.0588:  43%|████▎     | 54/125 [05:27<07:10,  6.07s/it]
Training Epoch: 9/100, batch 108/250 completed (loss: 0.1255:  44%|████▍     | 55/125 [05:30<07:06,  6.10s/it]
Training Epoch: 9/100, batch 108/250 completed (loss: 0.1734:  44%|████▍     | 55/125 [05:30<07:06,  6.10s/it]
Training Epoch: 9/100, batch 108/250 completed (loss: 0.0745:  44%|████▍     | 55/125 [05:30<07:06,  6.10s/it]
Training Epoch: 9/100, batch 108/250 completed (loss: 0.0588:  44%|████▍     | 55/125 [05:30<07:06,  6.10s/it]
Training Epoch: 9/100, batch 109/250 completed (loss: 0.1931:  44%|████▍     | 55/125 [05:30<07:06,  6.10s/it]
Training Epoch: 9/100, batch 109/250 completed (loss: 0.042:  44%|████▍     | 55/125 [05:30<07:06,  6.10s/it] 
Training Epoch: 9/100, batch 109/250 completed (loss: 0.2233:  44%|████▍     | 55/125 [05:30<07:06,  6.10s/it]
Training Epoch: 9/100, batch 109/250 completed (loss: 0.118:  44%|████▍     | 55/125 [05:30<07:06,  6.10s/it] 
Training Epoch: 9/100, batch 110/250 completed (loss: 0.6164:  44%|████▍     | 55/125 [05:33<07:06,  6.10s/it]
Training Epoch: 9/100, batch 110/250 completed (loss: 0.1806:  44%|████▍     | 55/125 [05:33<07:06,  6.10s/it]
Training Epoch: 9/100, batch 110/250 completed (loss: 0.0488:  44%|████▍     | 55/125 [05:33<07:06,  6.10s/it]
Training Epoch: 9/100, batch 110/250 completed (loss: 0.2007:  44%|████▍     | 55/125 [05:33<07:06,  6.10s/it]
Training Epoch: 9/100, batch 110/250 completed (loss: 0.0488:  45%|████▍     | 56/125 [05:36<06:55,  6.02s/it]
Training Epoch: 9/100, batch 110/250 completed (loss: 0.6164:  45%|████▍     | 56/125 [05:36<06:55,  6.02s/it]
Training Epoch: 9/100, batch 110/250 completed (loss: 0.2007:  45%|████▍     | 56/125 [05:36<06:55,  6.02s/it]
Training Epoch: 9/100, batch 110/250 completed (loss: 0.1806:  45%|████▍     | 56/125 [05:36<06:55,  6.02s/it]
Training Epoch: 9/100, batch 111/250 completed (loss: 0.0618:  45%|████▍     | 56/125 [05:36<06:55,  6.02s/it]
Training Epoch: 9/100, batch 111/250 completed (loss: 0.0205:  45%|████▍     | 56/125 [05:36<06:55,  6.02s/it]
Training Epoch: 9/100, batch 111/250 completed (loss: 0.1304:  45%|████▍     | 56/125 [05:36<06:55,  6.02s/it]
Training Epoch: 9/100, batch 111/250 completed (loss: 0.1831:  45%|████▍     | 56/125 [05:36<06:55,  6.02s/it]
Training Epoch: 9/100, batch 112/250 completed (loss: 0.0286:  45%|████▍     | 56/125 [05:39<06:55,  6.02s/it]
Training Epoch: 9/100, batch 112/250 completed (loss: 0.1394:  45%|████▍     | 56/125 [05:39<06:55,  6.02s/it]
Training Epoch: 9/100, batch 112/250 completed (loss: 0.0688:  45%|████▍     | 56/125 [05:39<06:55,  6.02s/it]
Training Epoch: 9/100, batch 112/250 completed (loss: 0.0418:  45%|████▍     | 56/125 [05:39<06:55,  6.02s/it]
Training Epoch: 9/100, batch 112/250 completed (loss: 0.1394:  46%|████▌     | 57/125 [05:42<06:54,  6.10s/it]
Training Epoch: 9/100, batch 112/250 completed (loss: 0.0688:  46%|████▌     | 57/125 [05:42<06:54,  6.10s/it]
Training Epoch: 9/100, batch 112/250 completed (loss: 0.0286:  46%|████▌     | 57/125 [05:42<06:54,  6.10s/it]
Training Epoch: 9/100, batch 112/250 completed (loss: 0.0418:  46%|████▌     | 57/125 [05:42<06:54,  6.10s/it]
Training Epoch: 9/100, batch 113/250 completed (loss: 0.1368:  46%|████▌     | 57/125 [05:42<06:54,  6.10s/it]
Training Epoch: 9/100, batch 113/250 completed (loss: 0.125:  46%|████▌     | 57/125 [05:42<06:54,  6.10s/it] 
Training Epoch: 9/100, batch 113/250 completed (loss: 0.0293:  46%|████▌     | 57/125 [05:42<06:54,  6.10s/it]
Training Epoch: 9/100, batch 113/250 completed (loss: 0.6306:  46%|████▌     | 57/125 [05:42<06:54,  6.10s/it]
Training Epoch: 9/100, batch 114/250 completed (loss: 0.106:  46%|████▌     | 57/125 [05:45<06:54,  6.10s/it] 
Training Epoch: 9/100, batch 114/250 completed (loss: 0.0762:  46%|████▌     | 57/125 [05:45<06:54,  6.10s/it]
Training Epoch: 9/100, batch 114/250 completed (loss: 0.0606:  46%|████▌     | 57/125 [05:45<06:54,  6.10s/it]
Training Epoch: 9/100, batch 114/250 completed (loss: 0.2969:  46%|████▌     | 57/125 [05:45<06:54,  6.10s/it]
Training Epoch: 9/100, batch 114/250 completed (loss: 0.106:  46%|████▋     | 58/125 [05:48<06:47,  6.09s/it]
Training Epoch: 9/100, batch 114/250 completed (loss: 0.0606:  46%|████▋     | 58/125 [05:48<06:47,  6.09s/it]
Training Epoch: 9/100, batch 114/250 completed (loss: 0.2969:  46%|████▋     | 58/125 [05:48<06:47,  6.09s/it]
Training Epoch: 9/100, batch 114/250 completed (loss: 0.0762:  46%|████▋     | 58/125 [05:48<06:48,  6.09s/it]
Training Epoch: 9/100, batch 115/250 completed (loss: 0.0494:  46%|████▋     | 58/125 [05:48<06:48,  6.09s/it]
Training Epoch: 9/100, batch 115/250 completed (loss: 0.1111:  46%|████▋     | 58/125 [05:48<06:47,  6.09s/it]
Training Epoch: 9/100, batch 115/250 completed (loss: 0.2628:  46%|████▋     | 58/125 [05:48<06:47,  6.09s/it]
Training Epoch: 9/100, batch 115/250 completed (loss: 0.2526:  46%|████▋     | 58/125 [05:48<06:47,  6.09s/it]
Training Epoch: 9/100, batch 116/250 completed (loss: 0.1576:  46%|████▋     | 58/125 [05:51<06:47,  6.09s/it]
Training Epoch: 9/100, batch 116/250 completed (loss: 0.2411:  46%|████▋     | 58/125 [05:51<06:47,  6.09s/it]
Training Epoch: 9/100, batch 116/250 completed (loss: 0.0839:  46%|████▋     | 58/125 [05:51<06:48,  6.09s/it]
Training Epoch: 9/100, batch 116/250 completed (loss: 0.1759:  46%|████▋     | 58/125 [05:51<06:47,  6.09s/it]
Training Epoch: 9/100, batch 116/250 completed (loss: 0.0839:  47%|████▋     | 59/125 [05:54<06:42,  6.10s/it]
Training Epoch: 9/100, batch 116/250 completed (loss: 0.1759:  47%|████▋     | 59/125 [05:54<06:42,  6.10s/it]
Training Epoch: 9/100, batch 116/250 completed (loss: 0.1576:  47%|████▋     | 59/125 [05:54<06:42,  6.11s/it]
Training Epoch: 9/100, batch 116/250 completed (loss: 0.2411:  47%|████▋     | 59/125 [05:54<06:43,  6.11s/it]
Training Epoch: 9/100, batch 117/250 completed (loss: 0.0668:  47%|████▋     | 59/125 [05:54<06:42,  6.10s/it]
Training Epoch: 9/100, batch 117/250 completed (loss: 0.2741:  47%|████▋     | 59/125 [05:54<06:43,  6.11s/it]
Training Epoch: 9/100, batch 117/250 completed (loss: 0.2128:  47%|████▋     | 59/125 [05:54<06:42,  6.11s/it]
Training Epoch: 9/100, batch 117/250 completed (loss: 0.0284:  47%|████▋     | 59/125 [05:54<06:42,  6.10s/it]
Training Epoch: 9/100, batch 118/250 completed (loss: 0.2362:  47%|████▋     | 59/125 [05:57<06:42,  6.10s/it]
Training Epoch: 9/100, batch 118/250 completed (loss: 0.0112:  47%|████▋     | 59/125 [05:57<06:43,  6.11s/it]
Training Epoch: 9/100, batch 118/250 completed (loss: 0.0888:  47%|████▋     | 59/125 [05:57<06:42,  6.11s/it]
Training Epoch: 9/100, batch 118/250 completed (loss: 0.2415:  47%|████▋     | 59/125 [05:57<06:42,  6.10s/it]
Training Epoch: 9/100, batch 118/250 completed (loss: 0.0888:  48%|████▊     | 60/125 [06:00<06:37,  6.11s/it]
Training Epoch: 9/100, batch 118/250 completed (loss: 0.2362:  48%|████▊     | 60/125 [06:00<06:37,  6.12s/it]
Training Epoch: 9/100, batch 118/250 completed (loss: 0.2415:  48%|████▊     | 60/125 [06:00<06:37,  6.12s/it]
Training Epoch: 9/100, batch 118/250 completed (loss: 0.0112:  48%|████▊     | 60/125 [06:00<06:37,  6.12s/it]
Training Epoch: 9/100, batch 119/250 completed (loss: 0.1567:  48%|████▊     | 60/125 [06:00<06:37,  6.12s/it]
Training Epoch: 9/100, batch 119/250 completed (loss: 0.6974:  48%|████▊     | 60/125 [06:00<06:37,  6.12s/it]
Training Epoch: 9/100, batch 119/250 completed (loss: 0.1215:  48%|████▊     | 60/125 [06:00<06:37,  6.11s/it]
Training Epoch: 9/100, batch 119/250 completed (loss: 0.0955:  48%|████▊     | 60/125 [06:00<06:37,  6.12s/it]
Training Epoch: 9/100, batch 120/250 completed (loss: 0.2854:  48%|████▊     | 60/125 [06:03<06:37,  6.12s/it]
Training Epoch: 9/100, batch 120/250 completed (loss: 0.0541:  48%|████▊     | 60/125 [06:03<06:37,  6.12s/it]
Training Epoch: 9/100, batch 120/250 completed (loss: 0.0355:  48%|████▊     | 60/125 [06:03<06:37,  6.11s/it]
Training Epoch: 9/100, batch 120/250 completed (loss: 0.2334:  48%|████▊     | 60/125 [06:03<06:37,  6.12s/it]
Training Epoch: 9/100, batch 120/250 completed (loss: 0.2854:  49%|████▉     | 61/125 [06:06<06:25,  6.03s/it]
Training Epoch: 9/100, batch 120/250 completed (loss: 0.0541:  49%|████▉     | 61/125 [06:06<06:25,  6.03s/it]
Training Epoch: 9/100, batch 120/250 completed (loss: 0.2334:  49%|████▉     | 61/125 [06:06<06:25,  6.03s/it]
Training Epoch: 9/100, batch 120/250 completed (loss: 0.0355:  49%|████▉     | 61/125 [06:06<06:26,  6.03s/it]
Training Epoch: 9/100, batch 121/250 completed (loss: 0.1613:  49%|████▉     | 61/125 [06:06<06:25,  6.03s/it]
Training Epoch: 9/100, batch 121/250 completed (loss: 0.0914:  49%|████▉     | 61/125 [06:06<06:25,  6.03s/it]
Training Epoch: 9/100, batch 121/250 completed (loss: 0.0964:  49%|████▉     | 61/125 [06:06<06:26,  6.03s/it]
Training Epoch: 9/100, batch 121/250 completed (loss: 0.075:  49%|████▉     | 61/125 [06:06<06:25,  6.03s/it] 
Training Epoch: 9/100, batch 122/250 completed (loss: 0.0498:  49%|████▉     | 61/125 [06:09<06:25,  6.03s/it]
Training Epoch: 9/100, batch 122/250 completed (loss: 0.136:  49%|████▉     | 61/125 [06:09<06:25,  6.03s/it] 
Training Epoch: 9/100, batch 122/250 completed (loss: 0.3746:  49%|████▉     | 61/125 [06:09<06:26,  6.03s/it]
Training Epoch: 9/100, batch 122/250 completed (loss: 0.0236:  49%|████▉     | 61/125 [06:10<06:25,  6.03s/it]
Training Epoch: 9/100, batch 122/250 completed (loss: 0.136:  50%|████▉     | 62/125 [06:12<06:24,  6.10s/it]
Training Epoch: 9/100, batch 122/250 completed (loss: 0.3746:  50%|████▉     | 62/125 [06:12<06:24,  6.10s/it]
Training Epoch: 9/100, batch 122/250 completed (loss: 0.0498:  50%|████▉     | 62/125 [06:12<06:24,  6.10s/it]
Training Epoch: 9/100, batch 122/250 completed (loss: 0.0236:  50%|████▉     | 62/125 [06:13<06:24,  6.10s/it]
Training Epoch: 9/100, batch 123/250 completed (loss: 0.0435:  50%|████▉     | 62/125 [06:13<06:24,  6.10s/it]
Training Epoch: 9/100, batch 123/250 completed (loss: 0.0741:  50%|████▉     | 62/125 [06:13<06:24,  6.10s/it]
Training Epoch: 9/100, batch 123/250 completed (loss: 0.1625:  50%|████▉     | 62/125 [06:13<06:24,  6.10s/it]
Training Epoch: 9/100, batch 123/250 completed (loss: 0.1367:  50%|████▉     | 62/125 [06:13<06:24,  6.10s/it]
Training Epoch: 9/100, batch 124/250 completed (loss: 0.1574:  50%|████▉     | 62/125 [06:15<06:24,  6.10s/it]
Training Epoch: 9/100, batch 124/250 completed (loss: 0.0564:  50%|████▉     | 62/125 [06:15<06:24,  6.10s/it]
Training Epoch: 9/100, batch 124/250 completed (loss: 0.2596:  50%|████▉     | 62/125 [06:15<06:24,  6.10s/it]
Training Epoch: 9/100, batch 124/250 completed (loss: 0.0836:  50%|████▉     | 62/125 [06:15<06:24,  6.10s/it]
Training Epoch: 9/100, batch 124/250 completed (loss: 0.0564:  50%|█████     | 63/125 [06:18<06:16,  6.07s/it]
Training Epoch: 9/100, batch 124/250 completed (loss: 0.2596:  50%|█████     | 63/125 [06:18<06:16,  6.07s/it]
Training Epoch: 9/100, batch 124/250 completed (loss: 0.1574:  50%|█████     | 63/125 [06:18<06:16,  6.07s/it]
Training Epoch: 9/100, batch 124/250 completed (loss: 0.0836:  50%|█████     | 63/125 [06:19<06:16,  6.07s/it]
Training Epoch: 9/100, batch 125/250 completed (loss: 0.1423:  50%|█████     | 63/125 [06:19<06:16,  6.07s/it]
Training Epoch: 9/100, batch 125/250 completed (loss: 0.0621:  50%|█████     | 63/125 [06:19<06:16,  6.07s/it]
Training Epoch: 9/100, batch 125/250 completed (loss: 0.1831:  50%|█████     | 63/125 [06:19<06:16,  6.07s/it]
Training Epoch: 9/100, batch 125/250 completed (loss: 0.2465:  50%|█████     | 63/125 [06:19<06:16,  6.07s/it]
Training Epoch: 9/100, batch 126/250 completed (loss: 0.057:  50%|█████     | 63/125 [06:21<06:16,  6.07s/it] 
Training Epoch: 9/100, batch 126/250 completed (loss: 0.0679:  50%|█████     | 63/125 [06:21<06:16,  6.07s/it]
Training Epoch: 9/100, batch 126/250 completed (loss: 0.2321:  50%|█████     | 63/125 [06:21<06:16,  6.07s/it]
Training Epoch: 9/100, batch 126/250 completed (loss: 0.1173:  50%|█████     | 63/125 [06:21<06:16,  6.07s/it]
Training Epoch: 9/100, batch 126/250 completed (loss: 0.1173:  51%|█████     | 64/125 [06:25<06:08,  6.04s/it]
Training Epoch: 9/100, batch 126/250 completed (loss: 0.057:  51%|█████     | 64/125 [06:24<06:08,  6.04s/it]
Training Epoch: 9/100, batch 126/250 completed (loss: 0.2321:  51%|█████     | 64/125 [06:24<06:08,  6.05s/it]
Training Epoch: 9/100, batch 126/250 completed (loss: 0.0679:  51%|█████     | 64/125 [06:24<06:08,  6.05s/it]
Training Epoch: 9/100, batch 127/250 completed (loss: 0.0997:  51%|█████     | 64/125 [06:25<06:08,  6.05s/it]
Training Epoch: 9/100, batch 127/250 completed (loss: 0.1946:  51%|█████     | 64/125 [06:25<06:08,  6.04s/it]
Training Epoch: 9/100, batch 127/250 completed (loss: 0.1772:  51%|█████     | 64/125 [06:25<06:08,  6.05s/it]
Training Epoch: 9/100, batch 127/250 completed (loss: 0.1381:  51%|█████     | 64/125 [06:25<06:08,  6.04s/it]
Training Epoch: 9/100, batch 128/250 completed (loss: 0.0308:  51%|█████     | 64/125 [06:28<06:08,  6.04s/it]
Training Epoch: 9/100, batch 128/250 completed (loss: 0.2531:  51%|█████     | 64/125 [06:28<06:08,  6.05s/it]
Training Epoch: 9/100, batch 128/250 completed (loss: 0.2707:  51%|█████     | 64/125 [06:28<06:08,  6.05s/it]
Training Epoch: 9/100, batch 128/250 completed (loss: 0.2233:  51%|█████     | 64/125 [06:28<06:08,  6.04s/it]
Training Epoch: 9/100, batch 128/250 completed (loss: 0.0308:  52%|█████▏    | 65/125 [06:31<06:03,  6.06s/it]
Training Epoch: 9/100, batch 128/250 completed (loss: 0.2707:  52%|█████▏    | 65/125 [06:31<06:03,  6.06s/it]
Training Epoch: 9/100, batch 128/250 completed (loss: 0.2233:  52%|█████▏    | 65/125 [06:31<06:03,  6.06s/it]
Training Epoch: 9/100, batch 128/250 completed (loss: 0.2531:  52%|█████▏    | 65/125 [06:31<06:03,  6.06s/it]
Training Epoch: 9/100, batch 129/250 completed (loss: 0.1666:  52%|█████▏    | 65/125 [06:31<06:03,  6.06s/it]
Training Epoch: 9/100, batch 129/250 completed (loss: 0.2131:  52%|█████▏    | 65/125 [06:31<06:03,  6.06s/it]
Training Epoch: 9/100, batch 129/250 completed (loss: 0.1116:  52%|█████▏    | 65/125 [06:31<06:03,  6.06s/it]
Training Epoch: 9/100, batch 129/250 completed (loss: 0.0454:  52%|█████▏    | 65/125 [06:31<06:03,  6.06s/it]
Training Epoch: 9/100, batch 130/250 completed (loss: 0.1457:  52%|█████▏    | 65/125 [06:34<06:03,  6.06s/it]
Training Epoch: 9/100, batch 130/250 completed (loss: 0.1002:  52%|█████▏    | 65/125 [06:34<06:03,  6.06s/it]
Training Epoch: 9/100, batch 130/250 completed (loss: 0.0893:  52%|█████▏    | 65/125 [06:34<06:03,  6.06s/it]
Training Epoch: 9/100, batch 130/250 completed (loss: 0.405:  52%|█████▏    | 65/125 [06:34<06:03,  6.06s/it] 
Training Epoch: 9/100, batch 130/250 completed (loss: 0.1002:  53%|█████▎    | 66/125 [06:37<05:56,  6.05s/it]
Training Epoch: 9/100, batch 130/250 completed (loss: 0.0893:  53%|█████▎    | 66/125 [06:37<05:56,  6.05s/it]
Training Epoch: 9/100, batch 130/250 completed (loss: 0.1457:  53%|█████▎    | 66/125 [06:37<05:56,  6.05s/it]
Training Epoch: 9/100, batch 130/250 completed (loss: 0.405:  53%|█████▎    | 66/125 [06:37<05:56,  6.05s/it]
Training Epoch: 9/100, batch 131/250 completed (loss: 0.251:  53%|█████▎    | 66/125 [06:37<05:56,  6.05s/it] 
Training Epoch: 9/100, batch 131/250 completed (loss: 0.0619:  53%|█████▎    | 66/125 [06:37<05:56,  6.05s/it]
Training Epoch: 9/100, batch 131/250 completed (loss: 0.1722:  53%|█████▎    | 66/125 [06:37<05:56,  6.05s/it]
Training Epoch: 9/100, batch 131/250 completed (loss: 0.4354:  53%|█████▎    | 66/125 [06:37<05:56,  6.05s/it]
Training Epoch: 9/100, batch 132/250 completed (loss: 0.0787:  53%|█████▎    | 66/125 [06:40<05:56,  6.05s/it]
Training Epoch: 9/100, batch 132/250 completed (loss: 0.2325:  53%|█████▎    | 66/125 [06:40<05:56,  6.05s/it]
Training Epoch: 9/100, batch 132/250 completed (loss: 0.129:  53%|█████▎    | 66/125 [06:40<05:56,  6.05s/it]
Training Epoch: 9/100, batch 132/250 completed (loss: 0.2261:  53%|█████▎    | 66/125 [06:40<05:56,  6.05s/it]
Training Epoch: 9/100, batch 132/250 completed (loss: 0.2325:  54%|█████▎    | 67/125 [06:43<05:52,  6.08s/it]
Training Epoch: 9/100, batch 132/250 completed (loss: 0.0787:  54%|█████▎    | 67/125 [06:43<05:52,  6.08s/it]
Training Epoch: 9/100, batch 132/250 completed (loss: 0.2261:  54%|█████▎    | 67/125 [06:43<05:52,  6.08s/it]
Training Epoch: 9/100, batch 132/250 completed (loss: 0.129:  54%|█████▎    | 67/125 [06:43<05:52,  6.08s/it]
Training Epoch: 9/100, batch 133/250 completed (loss: 0.1569:  54%|█████▎    | 67/125 [06:43<05:52,  6.08s/it]
Training Epoch: 9/100, batch 133/250 completed (loss: 0.1266:  54%|█████▎    | 67/125 [06:43<05:52,  6.08s/it]
Training Epoch: 9/100, batch 133/250 completed (loss: 0.025:  54%|█████▎    | 67/125 [06:43<05:52,  6.08s/it] 
Training Epoch: 9/100, batch 133/250 completed (loss: 0.1189:  54%|█████▎    | 67/125 [06:43<05:52,  6.08s/it]
Training Epoch: 9/100, batch 134/250 completed (loss: 0.193:  54%|█████▎    | 67/125 [06:46<05:52,  6.08s/it]
Training Epoch: 9/100, batch 134/250 completed (loss: 0.1128:  54%|█████▎    | 67/125 [06:46<05:52,  6.08s/it]
Training Epoch: 9/100, batch 134/250 completed (loss: 0.0956:  54%|█████▎    | 67/125 [06:46<05:52,  6.08s/it]
Training Epoch: 9/100, batch 134/250 completed (loss: 0.2071:  54%|█████▎    | 67/125 [06:46<05:52,  6.08s/it]
Training Epoch: 9/100, batch 134/250 completed (loss: 0.0956:  54%|█████▍    | 68/125 [06:49<05:44,  6.05s/it]
Training Epoch: 9/100, batch 134/250 completed (loss: 0.1128:  54%|█████▍    | 68/125 [06:49<05:44,  6.05s/it]
Training Epoch: 9/100, batch 134/250 completed (loss: 0.2071:  54%|█████▍    | 68/125 [06:49<05:44,  6.05s/it]
Training Epoch: 9/100, batch 134/250 completed (loss: 0.193:  54%|█████▍    | 68/125 [06:49<05:44,  6.05s/it]
Training Epoch: 9/100, batch 135/250 completed (loss: 0.4488:  54%|█████▍    | 68/125 [06:49<05:44,  6.05s/it]
Training Epoch: 9/100, batch 135/250 completed (loss: 0.0421:  54%|█████▍    | 68/125 [06:49<05:44,  6.05s/it]
Training Epoch: 9/100, batch 135/250 completed (loss: 0.0964:  54%|█████▍    | 68/125 [06:49<05:44,  6.05s/it]
Training Epoch: 9/100, batch 135/250 completed (loss: 0.4338:  54%|█████▍    | 68/125 [06:49<05:44,  6.05s/it]
Training Epoch: 9/100, batch 136/250 completed (loss: 0.2897:  54%|█████▍    | 68/125 [06:52<05:44,  6.05s/it]
Training Epoch: 9/100, batch 136/250 completed (loss: 0.1478:  54%|█████▍    | 68/125 [06:52<05:44,  6.05s/it]
Training Epoch: 9/100, batch 136/250 completed (loss: 0.186:  54%|█████▍    | 68/125 [06:52<05:44,  6.05s/it] 
Training Epoch: 9/100, batch 136/250 completed (loss: 0.0341:  54%|█████▍    | 68/125 [06:52<05:44,  6.05s/it]
Training Epoch: 9/100, batch 136/250 completed (loss: 0.186:  55%|█████▌    | 69/125 [06:55<05:39,  6.07s/it]
Training Epoch: 9/100, batch 136/250 completed (loss: 0.1478:  55%|█████▌    | 69/125 [06:55<05:39,  6.07s/it]
Training Epoch: 9/100, batch 136/250 completed (loss: 0.2897:  55%|█████▌    | 69/125 [06:55<05:39,  6.07s/it]
Training Epoch: 9/100, batch 136/250 completed (loss: 0.0341:  55%|█████▌    | 69/125 [06:55<05:40,  6.07s/it]
Training Epoch: 9/100, batch 137/250 completed (loss: 0.1514:  55%|█████▌    | 69/125 [06:55<05:39,  6.07s/it]
Training Epoch: 9/100, batch 137/250 completed (loss: 0.3616:  55%|█████▌    | 69/125 [06:55<05:39,  6.07s/it]
Training Epoch: 9/100, batch 137/250 completed (loss: 0.2404:  55%|█████▌    | 69/125 [06:55<05:39,  6.07s/it]
Training Epoch: 9/100, batch 137/250 completed (loss: 0.0567:  55%|█████▌    | 69/125 [06:55<05:40,  6.07s/it]
Training Epoch: 9/100, batch 138/250 completed (loss: 0.0337:  55%|█████▌    | 69/125 [06:58<05:39,  6.07s/it]
Training Epoch: 9/100, batch 138/250 completed (loss: 0.164:  55%|█████▌    | 69/125 [06:58<05:39,  6.07s/it] 
Training Epoch: 9/100, batch 138/250 completed (loss: 0.1404:  55%|█████▌    | 69/125 [06:58<05:39,  6.07s/it]
Training Epoch: 9/100, batch 138/250 completed (loss: 0.0598:  55%|█████▌    | 69/125 [06:58<05:40,  6.07s/it]
Training Epoch: 9/100, batch 138/250 completed (loss: 0.0598:  56%|█████▌    | 70/125 [07:01<05:35,  6.09s/it]
Training Epoch: 9/100, batch 138/250 completed (loss: 0.1404:  56%|█████▌    | 70/125 [07:01<05:35,  6.10s/it]
Training Epoch: 9/100, batch 138/250 completed (loss: 0.0337:  56%|█████▌    | 70/125 [07:01<05:35,  6.10s/it]
Training Epoch: 9/100, batch 138/250 completed (loss: 0.164:  56%|█████▌    | 70/125 [07:01<05:35,  6.10s/it]
Training Epoch: 9/100, batch 139/250 completed (loss: 0.2247:  56%|█████▌    | 70/125 [07:01<05:35,  6.10s/it]
Training Epoch: 9/100, batch 139/250 completed (loss: 0.0442:  56%|█████▌    | 70/125 [07:01<05:35,  6.10s/it]
Training Epoch: 9/100, batch 139/250 completed (loss: 0.1741:  56%|█████▌    | 70/125 [07:01<05:35,  6.10s/it]
Training Epoch: 9/100, batch 139/250 completed (loss: 0.1227:  56%|█████▌    | 70/125 [07:01<05:35,  6.09s/it]
Training Epoch: 9/100, batch 140/250 completed (loss: 0.0976:  56%|█████▌    | 70/125 [07:04<05:35,  6.10s/it]
Training Epoch: 9/100, batch 140/250 completed (loss: 0.0997:  56%|█████▌    | 70/125 [07:04<05:35,  6.10s/it]
Training Epoch: 9/100, batch 140/250 completed (loss: 0.1788:  56%|█████▌    | 70/125 [07:04<05:35,  6.10s/it]
Training Epoch: 9/100, batch 140/250 completed (loss: 0.1309:  56%|█████▌    | 70/125 [07:04<05:35,  6.09s/it]
Training Epoch: 9/100, batch 140/250 completed (loss: 0.0997:  57%|█████▋    | 71/125 [07:07<05:24,  6.02s/it]
Training Epoch: 9/100, batch 140/250 completed (loss: 0.1788:  57%|█████▋    | 71/125 [07:07<05:24,  6.02s/it]
Training Epoch: 9/100, batch 140/250 completed (loss: 0.0976:  57%|█████▋    | 71/125 [07:07<05:24,  6.02s/it]
Training Epoch: 9/100, batch 140/250 completed (loss: 0.1309:  57%|█████▋    | 71/125 [07:07<05:24,  6.02s/it]
Training Epoch: 9/100, batch 141/250 completed (loss: 0.2077:  57%|█████▋    | 71/125 [07:07<05:24,  6.02s/it]
Training Epoch: 9/100, batch 141/250 completed (loss: 0.4183:  57%|█████▋    | 71/125 [07:07<05:24,  6.02s/it]
Training Epoch: 9/100, batch 141/250 completed (loss: 0.1517:  57%|█████▋    | 71/125 [07:07<05:24,  6.02s/it]
Training Epoch: 9/100, batch 141/250 completed (loss: 0.0631:  57%|█████▋    | 71/125 [07:07<05:24,  6.02s/it]
Training Epoch: 9/100, batch 142/250 completed (loss: 0.1648:  57%|█████▋    | 71/125 [07:10<05:24,  6.02s/it]
Training Epoch: 9/100, batch 142/250 completed (loss: 0.0819:  57%|█████▋    | 71/125 [07:10<05:24,  6.02s/it]
Training Epoch: 9/100, batch 142/250 completed (loss: 0.0297:  57%|█████▋    | 71/125 [07:10<05:24,  6.02s/it]
Training Epoch: 9/100, batch 142/250 completed (loss: 0.237:  57%|█████▋    | 71/125 [07:10<05:24,  6.02s/it] 
Training Epoch: 9/100, batch 142/250 completed (loss: 0.237:  58%|█████▊    | 72/125 [07:13<05:23,  6.10s/it]
Training Epoch: 9/100, batch 142/250 completed (loss: 0.1648:  58%|█████▊    | 72/125 [07:13<05:23,  6.10s/it]
Training Epoch: 9/100, batch 142/250 completed (loss: 0.0819:  58%|█████▊    | 72/125 [07:13<05:23,  6.10s/it]
Training Epoch: 9/100, batch 142/250 completed (loss: 0.0297:  58%|█████▊    | 72/125 [07:13<05:23,  6.10s/it]
Training Epoch: 9/100, batch 143/250 completed (loss: 0.2709:  58%|█████▊    | 72/125 [07:13<05:23,  6.10s/it]
Training Epoch: 9/100, batch 143/250 completed (loss: 0.2081:  58%|█████▊    | 72/125 [07:13<05:23,  6.10s/it]
Training Epoch: 9/100, batch 143/250 completed (loss: 0.2056:  58%|█████▊    | 72/125 [07:13<05:23,  6.10s/it]
Training Epoch: 9/100, batch 143/250 completed (loss: 0.1475:  58%|█████▊    | 72/125 [07:13<05:23,  6.10s/it]
Training Epoch: 9/100, batch 144/250 completed (loss: 0.0672:  58%|█████▊    | 72/125 [07:16<05:23,  6.10s/it]
Training Epoch: 9/100, batch 144/250 completed (loss: 0.1359:  58%|█████▊    | 72/125 [07:16<05:23,  6.10s/it]
Training Epoch: 9/100, batch 144/250 completed (loss: 0.5155:  58%|█████▊    | 72/125 [07:16<05:23,  6.10s/it]
Training Epoch: 9/100, batch 144/250 completed (loss: 0.121:  58%|█████▊    | 72/125 [07:16<05:23,  6.10s/it] 
Training Epoch: 9/100, batch 144/250 completed (loss: 0.5155:  58%|█████▊    | 73/125 [07:19<05:15,  6.07s/it]
Training Epoch: 9/100, batch 144/250 completed (loss: 0.1359:  58%|█████▊    | 73/125 [07:19<05:15,  6.07s/it]
Training Epoch: 9/100, batch 144/250 completed (loss: 0.0672:  58%|█████▊    | 73/125 [07:19<05:15,  6.07s/it]
Training Epoch: 9/100, batch 144/250 completed (loss: 0.121:  58%|█████▊    | 73/125 [07:19<05:15,  6.07s/it]
Training Epoch: 9/100, batch 145/250 completed (loss: 0.075:  58%|█████▊    | 73/125 [07:19<05:15,  6.07s/it] 
Training Epoch: 9/100, batch 145/250 completed (loss: 0.0265:  58%|█████▊    | 73/125 [07:19<05:15,  6.07s/it]
Training Epoch: 9/100, batch 145/250 completed (loss: 0.1521:  58%|█████▊    | 73/125 [07:19<05:15,  6.07s/it]
Training Epoch: 9/100, batch 145/250 completed (loss: 0.0691:  58%|█████▊    | 73/125 [07:19<05:15,  6.07s/it]
Training Epoch: 9/100, batch 146/250 completed (loss: 0.1863:  58%|█████▊    | 73/125 [07:22<05:15,  6.07s/it]
Training Epoch: 9/100, batch 146/250 completed (loss: 0.0464:  58%|█████▊    | 73/125 [07:22<05:15,  6.07s/it]
Training Epoch: 9/100, batch 146/250 completed (loss: 0.2329:  58%|█████▊    | 73/125 [07:22<05:15,  6.07s/it]
Training Epoch: 9/100, batch 146/250 completed (loss: 0.1148:  58%|█████▊    | 73/125 [07:22<05:15,  6.07s/it]
Training Epoch: 9/100, batch 146/250 completed (loss: 0.2329:  59%|█████▉    | 74/125 [07:25<05:08,  6.05s/it]
Training Epoch: 9/100, batch 146/250 completed (loss: 0.1148:  59%|█████▉    | 74/125 [07:25<05:08,  6.05s/it]
Training Epoch: 9/100, batch 146/250 completed (loss: 0.1863:  59%|█████▉    | 74/125 [07:25<05:08,  6.05s/it]
Training Epoch: 9/100, batch 146/250 completed (loss: 0.0464:  59%|█████▉    | 74/125 [07:25<05:08,  6.05s/it]
Training Epoch: 9/100, batch 147/250 completed (loss: 0.0685:  59%|█████▉    | 74/125 [07:25<05:08,  6.05s/it]
Training Epoch: 9/100, batch 147/250 completed (loss: 0.095:  59%|█████▉    | 74/125 [07:25<05:08,  6.05s/it] 
Training Epoch: 9/100, batch 147/250 completed (loss: 0.1496:  59%|█████▉    | 74/125 [07:25<05:08,  6.05s/it]
Training Epoch: 9/100, batch 147/250 completed (loss: 0.0739:  59%|█████▉    | 74/125 [07:25<05:08,  6.05s/it]
Training Epoch: 9/100, batch 148/250 completed (loss: 0.1053:  59%|█████▉    | 74/125 [07:28<05:08,  6.05s/it]
Training Epoch: 9/100, batch 148/250 completed (loss: 0.0267:  59%|█████▉    | 74/125 [07:28<05:08,  6.05s/it]
Training Epoch: 9/100, batch 148/250 completed (loss: 0.1381:  59%|█████▉    | 74/125 [07:28<05:08,  6.05s/it]
Training Epoch: 9/100, batch 148/250 completed (loss: 0.1072:  59%|█████▉    | 74/125 [07:28<05:08,  6.05s/it]
Training Epoch: 9/100, batch 148/250 completed (loss: 0.1381:  60%|██████    | 75/125 [07:31<05:03,  6.07s/it]
Training Epoch: 9/100, batch 148/250 completed (loss: 0.0267:  60%|██████    | 75/125 [07:31<05:03,  6.07s/it]
Training Epoch: 9/100, batch 148/250 completed (loss: 0.1053:  60%|██████    | 75/125 [07:31<05:03,  6.08s/it]
Training Epoch: 9/100, batch 148/250 completed (loss: 0.1072:  60%|██████    | 75/125 [07:31<05:03,  6.08s/it]
Training Epoch: 9/100, batch 149/250 completed (loss: 0.2602:  60%|██████    | 75/125 [07:31<05:03,  6.08s/it]
Training Epoch: 9/100, batch 149/250 completed (loss: 0.22:  60%|██████    | 75/125 [07:31<05:03,  6.07s/it]  
Training Epoch: 9/100, batch 149/250 completed (loss: 0.1194:  60%|██████    | 75/125 [07:31<05:03,  6.07s/it]
Training Epoch: 9/100, batch 149/250 completed (loss: 0.1403:  60%|██████    | 75/125 [07:31<05:03,  6.08s/it]
Training Epoch: 9/100, batch 150/250 completed (loss: 0.171:  60%|██████    | 75/125 [07:34<05:03,  6.08s/it] 
Training Epoch: 9/100, batch 150/250 completed (loss: 0.153:  60%|██████    | 75/125 [07:34<05:03,  6.07s/it] 
Training Epoch: 9/100, batch 150/250 completed (loss: 0.5255:  60%|██████    | 75/125 [07:34<05:03,  6.07s/it]
Training Epoch: 9/100, batch 150/250 completed (loss: 0.3585:  60%|██████    | 75/125 [07:34<05:03,  6.08s/it]
Training Epoch: 9/100, batch 150/250 completed (loss: 0.3585:  61%|██████    | 76/125 [07:37<04:56,  6.05s/it]
Training Epoch: 9/100, batch 150/250 completed (loss: 0.171:  61%|██████    | 76/125 [07:37<04:56,  6.05s/it]
Training Epoch: 9/100, batch 150/250 completed (loss: 0.153:  61%|██████    | 76/125 [07:37<04:56,  6.05s/it]
Training Epoch: 9/100, batch 150/250 completed (loss: 0.5255:  61%|██████    | 76/125 [07:37<04:56,  6.05s/it]
Training Epoch: 9/100, batch 151/250 completed (loss: 0.2041:  61%|██████    | 76/125 [07:37<04:56,  6.05s/it]
Training Epoch: 9/100, batch 151/250 completed (loss: 0.2038:  61%|██████    | 76/125 [07:37<04:56,  6.05s/it]
Training Epoch: 9/100, batch 151/250 completed (loss: 0.0709:  61%|██████    | 76/125 [07:37<04:56,  6.05s/it]
Training Epoch: 9/100, batch 151/250 completed (loss: 0.3221:  61%|██████    | 76/125 [07:37<04:56,  6.05s/it]
Training Epoch: 9/100, batch 152/250 completed (loss: 0.075:  61%|██████    | 76/125 [07:40<04:56,  6.05s/it] 
Training Epoch: 9/100, batch 152/250 completed (loss: 0.1259:  61%|██████    | 76/125 [07:40<04:56,  6.05s/it]
Training Epoch: 9/100, batch 152/250 completed (loss: 0.048:  61%|██████    | 76/125 [07:40<04:56,  6.05s/it] 
Training Epoch: 9/100, batch 152/250 completed (loss: 0.023:  61%|██████    | 76/125 [07:40<04:56,  6.05s/it] 
Training Epoch: 9/100, batch 152/250 completed (loss: 0.048:  62%|██████▏   | 77/125 [07:43<04:51,  6.07s/it]
Training Epoch: 9/100, batch 152/250 completed (loss: 0.075:  62%|██████▏   | 77/125 [07:43<04:51,  6.07s/it]
Training Epoch: 9/100, batch 152/250 completed (loss: 0.023:  62%|██████▏   | 77/125 [07:43<04:51,  6.07s/it]
Training Epoch: 9/100, batch 152/250 completed (loss: 0.1259:  62%|██████▏   | 77/125 [07:43<04:51,  6.07s/it]
Training Epoch: 9/100, batch 153/250 completed (loss: 0.0778:  62%|██████▏   | 77/125 [07:43<04:51,  6.07s/it]
Training Epoch: 9/100, batch 153/250 completed (loss: 0.5723:  62%|██████▏   | 77/125 [07:43<04:51,  6.07s/it]
Training Epoch: 9/100, batch 153/250 completed (loss: 0.1039:  62%|██████▏   | 77/125 [07:43<04:51,  6.07s/it]
Training Epoch: 9/100, batch 153/250 completed (loss: 0.0649:  62%|██████▏   | 77/125 [07:43<04:51,  6.07s/it]
Training Epoch: 9/100, batch 154/250 completed (loss: 0.3053:  62%|██████▏   | 77/125 [07:46<04:51,  6.07s/it]
Training Epoch: 9/100, batch 154/250 completed (loss: 0.1055:  62%|██████▏   | 77/125 [07:46<04:51,  6.07s/it]
Training Epoch: 9/100, batch 154/250 completed (loss: 0.028:  62%|██████▏   | 77/125 [07:46<04:51,  6.07s/it] 
Training Epoch: 9/100, batch 154/250 completed (loss: 0.1515:  62%|██████▏   | 77/125 [07:46<04:51,  6.07s/it]
Training Epoch: 9/100, batch 154/250 completed (loss: 0.1515:  62%|██████▏   | 78/125 [07:50<04:46,  6.09s/it]
Training Epoch: 9/100, batch 154/250 completed (loss: 0.3053:  62%|██████▏   | 78/125 [07:50<04:46,  6.09s/it]
Training Epoch: 9/100, batch 154/250 completed (loss: 0.1055:  62%|██████▏   | 78/125 [07:50<04:46,  6.09s/it]
Training Epoch: 9/100, batch 154/250 completed (loss: 0.028:  62%|██████▏   | 78/125 [07:50<04:46,  6.10s/it]
Training Epoch: 9/100, batch 155/250 completed (loss: 0.2184:  62%|██████▏   | 78/125 [07:50<04:46,  6.10s/it]
Training Epoch: 9/100, batch 155/250 completed (loss: 0.0866:  62%|██████▏   | 78/125 [07:50<04:46,  6.09s/it]
Training Epoch: 9/100, batch 155/250 completed (loss: 0.0522:  62%|██████▏   | 78/125 [07:50<04:46,  6.09s/it]
Training Epoch: 9/100, batch 155/250 completed (loss: 0.3784:  62%|██████▏   | 78/125 [07:50<04:46,  6.09s/it]
Training Epoch: 9/100, batch 156/250 completed (loss: 0.1296:  62%|██████▏   | 78/125 [07:52<04:46,  6.09s/it]
Training Epoch: 9/100, batch 156/250 completed (loss: 0.0984:  62%|██████▏   | 78/125 [07:52<04:46,  6.10s/it]
Training Epoch: 9/100, batch 156/250 completed (loss: 0.1042:  62%|██████▏   | 78/125 [07:52<04:46,  6.09s/it]
Training Epoch: 9/100, batch 156/250 completed (loss: 0.3136:  62%|██████▏   | 78/125 [07:52<04:46,  6.09s/it]
Training Epoch: 9/100, batch 156/250 completed (loss: 0.0984:  63%|██████▎   | 79/125 [07:55<04:36,  6.01s/it]
Training Epoch: 9/100, batch 156/250 completed (loss: 0.3136:  63%|██████▎   | 79/125 [07:55<04:36,  6.01s/it]
Training Epoch: 9/100, batch 156/250 completed (loss: 0.1296:  63%|██████▎   | 79/125 [07:55<04:36,  6.01s/it]
Training Epoch: 9/100, batch 156/250 completed (loss: 0.1042:  63%|██████▎   | 79/125 [07:55<04:36,  6.01s/it]
Training Epoch: 9/100, batch 157/250 completed (loss: 0.0154:  63%|██████▎   | 79/125 [07:55<04:36,  6.01s/it]
Training Epoch: 9/100, batch 157/250 completed (loss: 0.1381:  63%|██████▎   | 79/125 [07:55<04:36,  6.01s/it]
Training Epoch: 9/100, batch 157/250 completed (loss: 0.1018:  63%|██████▎   | 79/125 [07:55<04:36,  6.01s/it]
Training Epoch: 9/100, batch 157/250 completed (loss: 0.0505:  63%|██████▎   | 79/125 [07:55<04:36,  6.01s/it]
Training Epoch: 9/100, batch 158/250 completed (loss: 0.1383:  63%|██████▎   | 79/125 [07:59<04:36,  6.01s/it]
Training Epoch: 9/100, batch 158/250 completed (loss: 0.0321:  63%|██████▎   | 79/125 [07:59<04:36,  6.01s/it]
Training Epoch: 9/100, batch 158/250 completed (loss: 0.6772:  63%|██████▎   | 79/125 [07:59<04:36,  6.01s/it]
Training Epoch: 9/100, batch 158/250 completed (loss: 0.1191:  63%|██████▎   | 79/125 [07:59<04:36,  6.01s/it]
Training Epoch: 9/100, batch 158/250 completed (loss: 0.6772:  64%|██████▍   | 80/125 [08:02<04:34,  6.10s/it]
Training Epoch: 9/100, batch 158/250 completed (loss: 0.1191:  64%|██████▍   | 80/125 [08:02<04:34,  6.10s/it]
Training Epoch: 9/100, batch 158/250 completed (loss: 0.1383:  64%|██████▍   | 80/125 [08:02<04:34,  6.10s/it]
Training Epoch: 9/100, batch 158/250 completed (loss: 0.0321:  64%|██████▍   | 80/125 [08:02<04:34,  6.10s/it]
Training Epoch: 9/100, batch 159/250 completed (loss: 0.1356:  64%|██████▍   | 80/125 [08:02<04:34,  6.10s/it]
Training Epoch: 9/100, batch 159/250 completed (loss: 0.0974:  64%|██████▍   | 80/125 [08:02<04:34,  6.10s/it]
Training Epoch: 9/100, batch 159/250 completed (loss: 0.0549:  64%|██████▍   | 80/125 [08:02<04:34,  6.10s/it]
Training Epoch: 9/100, batch 159/250 completed (loss: 0.1836:  64%|██████▍   | 80/125 [08:02<04:34,  6.10s/it]
Training Epoch: 9/100, batch 160/250 completed (loss: 0.0867:  64%|██████▍   | 80/125 [08:05<04:34,  6.10s/it]
Training Epoch: 9/100, batch 160/250 completed (loss: 0.0611:  64%|██████▍   | 80/125 [08:05<04:34,  6.10s/it]
Training Epoch: 9/100, batch 160/250 completed (loss: 0.5612:  64%|██████▍   | 80/125 [08:05<04:34,  6.10s/it]
Training Epoch: 9/100, batch 160/250 completed (loss: 0.1286:  64%|██████▍   | 80/125 [08:05<04:34,  6.10s/it]
Training Epoch: 9/100, batch 160/250 completed (loss: 0.0611:  65%|██████▍   | 81/125 [08:08<04:27,  6.08s/it]
Training Epoch: 9/100, batch 160/250 completed (loss: 0.5612:  65%|██████▍   | 81/125 [08:08<04:27,  6.08s/it]
Training Epoch: 9/100, batch 160/250 completed (loss: 0.0867:  65%|██████▍   | 81/125 [08:08<04:27,  6.08s/it]
Training Epoch: 9/100, batch 160/250 completed (loss: 0.1286:  65%|██████▍   | 81/125 [08:08<04:27,  6.09s/it]
Training Epoch: 9/100, batch 161/250 completed (loss: 0.0798:  65%|██████▍   | 81/125 [08:08<04:27,  6.08s/it]
Training Epoch: 9/100, batch 161/250 completed (loss: 0.0737:  65%|██████▍   | 81/125 [08:08<04:27,  6.08s/it]
Training Epoch: 9/100, batch 161/250 completed (loss: 0.032:  65%|██████▍   | 81/125 [08:08<04:27,  6.08s/it] 
Training Epoch: 9/100, batch 161/250 completed (loss: 0.0668:  65%|██████▍   | 81/125 [08:08<04:27,  6.09s/it]
Training Epoch: 9/100, batch 162/250 completed (loss: 0.1112:  65%|██████▍   | 81/125 [08:11<04:27,  6.08s/it]
Training Epoch: 9/100, batch 162/250 completed (loss: 0.1661:  65%|██████▍   | 81/125 [08:11<04:27,  6.08s/it]
Training Epoch: 9/100, batch 162/250 completed (loss: 0.125:  65%|██████▍   | 81/125 [08:11<04:27,  6.08s/it] 
Training Epoch: 9/100, batch 162/250 completed (loss: 0.1283:  65%|██████▍   | 81/125 [08:11<04:27,  6.09s/it]
Training Epoch: 9/100, batch 162/250 completed (loss: 0.1283:  66%|██████▌   | 82/125 [08:14<04:19,  6.04s/it]
Training Epoch: 9/100, batch 162/250 completed (loss: 0.125:  66%|██████▌   | 82/125 [08:14<04:20,  6.05s/it]
Training Epoch: 9/100, batch 162/250 completed (loss: 0.1112:  66%|██████▌   | 82/125 [08:14<04:20,  6.05s/it]
Training Epoch: 9/100, batch 162/250 completed (loss: 0.1661:  66%|██████▌   | 82/125 [08:14<04:20,  6.05s/it]
Training Epoch: 9/100, batch 163/250 completed (loss: 0.0805:  66%|██████▌   | 82/125 [08:14<04:20,  6.05s/it]
Training Epoch: 9/100, batch 163/250 completed (loss: 0.1386:  66%|██████▌   | 82/125 [08:14<04:20,  6.05s/it]
Training Epoch: 9/100, batch 163/250 completed (loss: 0.1069:  66%|██████▌   | 82/125 [08:14<04:20,  6.05s/it]
Training Epoch: 9/100, batch 163/250 completed (loss: 0.0605:  66%|██████▌   | 82/125 [08:14<04:19,  6.04s/it]
Training Epoch: 9/100, batch 164/250 completed (loss: 0.2127:  66%|██████▌   | 82/125 [08:17<04:20,  6.05s/it]
Training Epoch: 9/100, batch 164/250 completed (loss: 0.3049:  66%|██████▌   | 82/125 [08:17<04:20,  6.05s/it]
Training Epoch: 9/100, batch 164/250 completed (loss: 0.12:  66%|██████▌   | 82/125 [08:17<04:20,  6.05s/it]  
Training Epoch: 9/100, batch 164/250 completed (loss: 0.0896:  66%|██████▌   | 82/125 [08:17<04:19,  6.04s/it]
Training Epoch: 9/100, batch 164/250 completed (loss: 0.2127:  66%|██████▋   | 83/125 [08:20<04:15,  6.08s/it]
Training Epoch: 9/100, batch 164/250 completed (loss: 0.3049:  66%|██████▋   | 83/125 [08:20<04:15,  6.08s/it]
Training Epoch: 9/100, batch 164/250 completed (loss: 0.0896:  66%|██████▋   | 83/125 [08:20<04:15,  6.08s/it]
Training Epoch: 9/100, batch 164/250 completed (loss: 0.12:  66%|██████▋   | 83/125 [08:20<04:15,  6.08s/it]
Training Epoch: 9/100, batch 165/250 completed (loss: 0.1656:  66%|██████▋   | 83/125 [08:20<04:15,  6.08s/it]
Training Epoch: 9/100, batch 165/250 completed (loss: 0.0429:  66%|██████▋   | 83/125 [08:20<04:15,  6.08s/it]
Training Epoch: 9/100, batch 165/250 completed (loss: 0.1572:  66%|██████▋   | 83/125 [08:20<04:15,  6.08s/it]
Training Epoch: 9/100, batch 165/250 completed (loss: 0.1001:  66%|██████▋   | 83/125 [08:20<04:15,  6.08s/it]
Training Epoch: 9/100, batch 166/250 completed (loss: 0.1231:  66%|██████▋   | 83/125 [08:23<04:15,  6.08s/it]
Training Epoch: 9/100, batch 166/250 completed (loss: 0.2419:  66%|██████▋   | 83/125 [08:23<04:15,  6.08s/it]
Training Epoch: 9/100, batch 166/250 completed (loss: 0.1437:  66%|██████▋   | 83/125 [08:23<04:15,  6.08s/it]
Training Epoch: 9/100, batch 166/250 completed (loss: 0.0885:  66%|██████▋   | 83/125 [08:23<04:15,  6.08s/it]
Training Epoch: 9/100, batch 166/250 completed (loss: 0.1231:  67%|██████▋   | 84/125 [08:26<04:07,  6.04s/it]
Training Epoch: 9/100, batch 166/250 completed (loss: 0.2419:  67%|██████▋   | 84/125 [08:26<04:07,  6.04s/it]
Training Epoch: 9/100, batch 166/250 completed (loss: 0.1437:  67%|██████▋   | 84/125 [08:26<04:07,  6.04s/it]
Training Epoch: 9/100, batch 166/250 completed (loss: 0.0885:  67%|██████▋   | 84/125 [08:26<04:07,  6.04s/it]
Training Epoch: 9/100, batch 167/250 completed (loss: 0.2097:  67%|██████▋   | 84/125 [08:26<04:07,  6.04s/it]
Training Epoch: 9/100, batch 167/250 completed (loss: 0.3611:  67%|██████▋   | 84/125 [08:26<04:07,  6.04s/it]
Training Epoch: 9/100, batch 167/250 completed (loss: 0.6088:  67%|██████▋   | 84/125 [08:26<04:07,  6.04s/it]
Training Epoch: 9/100, batch 167/250 completed (loss: 0.1013:  67%|██████▋   | 84/125 [08:26<04:07,  6.04s/it]
Training Epoch: 9/100, batch 168/250 completed (loss: 0.1241:  67%|██████▋   | 84/125 [08:29<04:07,  6.04s/it]
Training Epoch: 9/100, batch 168/250 completed (loss: 0.0337:  67%|██████▋   | 84/125 [08:29<04:07,  6.04s/it]
Training Epoch: 9/100, batch 168/250 completed (loss: 0.4054:  67%|██████▋   | 84/125 [08:29<04:07,  6.04s/it]
Training Epoch: 9/100, batch 168/250 completed (loss: 0.1071:  67%|██████▋   | 84/125 [08:29<04:07,  6.04s/it]
Training Epoch: 9/100, batch 168/250 completed (loss: 0.4054:  68%|██████▊   | 85/125 [08:32<04:03,  6.10s/it]
Training Epoch: 9/100, batch 168/250 completed (loss: 0.1241:  68%|██████▊   | 85/125 [08:32<04:03,  6.10s/it]
Training Epoch: 9/100, batch 168/250 completed (loss: 0.0337:  68%|██████▊   | 85/125 [08:32<04:03,  6.10s/it]
Training Epoch: 9/100, batch 168/250 completed (loss: 0.1071:  68%|██████▊   | 85/125 [08:32<04:04,  6.10s/it]
Training Epoch: 9/100, batch 169/250 completed (loss: 0.1536:  68%|██████▊   | 85/125 [08:32<04:03,  6.10s/it]
Training Epoch: 9/100, batch 169/250 completed (loss: 0.1396:  68%|██████▊   | 85/125 [08:32<04:03,  6.10s/it]
Training Epoch: 9/100, batch 169/250 completed (loss: 0.1:  68%|██████▊   | 85/125 [08:32<04:03,  6.10s/it]   
Training Epoch: 9/100, batch 169/250 completed (loss: 0.0675:  68%|██████▊   | 85/125 [08:32<04:04,  6.10s/it]
Training Epoch: 9/100, batch 170/250 completed (loss: 0.0757:  68%|██████▊   | 85/125 [08:35<04:03,  6.10s/it]
Training Epoch: 9/100, batch 170/250 completed (loss: 0.0625:  68%|██████▊   | 85/125 [08:35<04:03,  6.10s/it]
Training Epoch: 9/100, batch 170/250 completed (loss: 0.093:  68%|██████▊   | 85/125 [08:35<04:03,  6.10s/it] 
Training Epoch: 9/100, batch 170/250 completed (loss: 0.3128:  68%|██████▊   | 85/125 [08:35<04:04,  6.10s/it]
Training Epoch: 9/100, batch 170/250 completed (loss: 0.0625:  69%|██████▉   | 86/125 [08:38<03:58,  6.11s/it]
Training Epoch: 9/100, batch 170/250 completed (loss: 0.3128:  69%|██████▉   | 86/125 [08:38<03:58,  6.11s/it]
Training Epoch: 9/100, batch 170/250 completed (loss: 0.0757:  69%|██████▉   | 86/125 [08:38<03:58,  6.12s/it]
Training Epoch: 9/100, batch 170/250 completed (loss: 0.093:  69%|██████▉   | 86/125 [08:38<03:58,  6.12s/it]
Training Epoch: 9/100, batch 171/250 completed (loss: 0.0671:  69%|██████▉   | 86/125 [08:38<03:58,  6.11s/it]
Training Epoch: 9/100, batch 171/250 completed (loss: 0.089:  69%|██████▉   | 86/125 [08:38<03:58,  6.12s/it] 
Training Epoch: 9/100, batch 171/250 completed (loss: 0.1701:  69%|██████▉   | 86/125 [08:38<03:58,  6.12s/it]
Training Epoch: 9/100, batch 171/250 completed (loss: 0.1622:  69%|██████▉   | 86/125 [08:38<03:58,  6.11s/it]
Training Epoch: 9/100, batch 172/250 completed (loss: 0.0229:  69%|██████▉   | 86/125 [08:41<03:58,  6.11s/it]
Training Epoch: 9/100, batch 172/250 completed (loss: 0.0732:  69%|██████▉   | 86/125 [08:41<03:58,  6.12s/it]
Training Epoch: 9/100, batch 172/250 completed (loss: 0.0653:  69%|██████▉   | 86/125 [08:41<03:58,  6.12s/it]
Training Epoch: 9/100, batch 172/250 completed (loss: 0.2059:  69%|██████▉   | 86/125 [08:41<03:58,  6.11s/it]
Training Epoch: 9/100, batch 172/250 completed (loss: 0.0732:  70%|██████▉   | 87/125 [08:44<03:49,  6.04s/it]
Training Epoch: 9/100, batch 172/250 completed (loss: 0.2059:  70%|██████▉   | 87/125 [08:44<03:49,  6.04s/it]
Training Epoch: 9/100, batch 172/250 completed (loss: 0.0229:  70%|██████▉   | 87/125 [08:44<03:49,  6.04s/it]
Training Epoch: 9/100, batch 172/250 completed (loss: 0.0653:  70%|██████▉   | 87/125 [08:44<03:49,  6.04s/it]
Training Epoch: 9/100, batch 173/250 completed (loss: 0.044:  70%|██████▉   | 87/125 [08:44<03:49,  6.04s/it] 
Training Epoch: 9/100, batch 173/250 completed (loss: 0.0514:  70%|██████▉   | 87/125 [08:44<03:49,  6.04s/it]
Training Epoch: 9/100, batch 173/250 completed (loss: 0.0302:  70%|██████▉   | 87/125 [08:44<03:49,  6.04s/it]
Training Epoch: 9/100, batch 173/250 completed (loss: 0.1038:  70%|██████▉   | 87/125 [08:44<03:49,  6.04s/it]
Training Epoch: 9/100, batch 174/250 completed (loss: 0.1166:  70%|██████▉   | 87/125 [08:47<03:49,  6.04s/it]
Training Epoch: 9/100, batch 174/250 completed (loss: 0.108:  70%|██████▉   | 87/125 [08:47<03:49,  6.04s/it]
Training Epoch: 9/100, batch 174/250 completed (loss: 0.0682:  70%|██████▉   | 87/125 [08:47<03:49,  6.04s/it]
Training Epoch: 9/100, batch 174/250 completed (loss: 0.298:  70%|██████▉   | 87/125 [08:47<03:49,  6.04s/it] 
Training Epoch: 9/100, batch 174/250 completed (loss: 0.1166:  70%|███████   | 88/125 [08:50<03:47,  6.16s/it]
Training Epoch: 9/100, batch 174/250 completed (loss: 0.0682:  70%|███████   | 88/125 [08:50<03:47,  6.16s/it]
Training Epoch: 9/100, batch 174/250 completed (loss: 0.298:  70%|███████   | 88/125 [08:50<03:47,  6.16s/it]
Training Epoch: 9/100, batch 174/250 completed (loss: 0.108:  70%|███████   | 88/125 [08:50<03:47,  6.16s/it]
Training Epoch: 9/100, batch 175/250 completed (loss: 0.0637:  70%|███████   | 88/125 [08:50<03:47,  6.16s/it]
Training Epoch: 9/100, batch 175/250 completed (loss: 0.2596:  70%|███████   | 88/125 [08:50<03:47,  6.16s/it]
Training Epoch: 9/100, batch 175/250 completed (loss: 0.0428:  70%|███████   | 88/125 [08:50<03:47,  6.16s/it]
Training Epoch: 9/100, batch 175/250 completed (loss: 0.0814:  70%|███████   | 88/125 [08:51<03:47,  6.16s/it]
Training Epoch: 9/100, batch 176/250 completed (loss: 0.0207:  70%|███████   | 88/125 [08:53<03:47,  6.16s/it]
Training Epoch: 9/100, batch 176/250 completed (loss: 0.1589:  70%|███████   | 88/125 [08:53<03:47,  6.16s/it]
Training Epoch: 9/100, batch 176/250 completed (loss: 0.1903:  70%|███████   | 88/125 [08:53<03:47,  6.16s/it]
Training Epoch: 9/100, batch 176/250 completed (loss: 0.1179:  70%|███████   | 88/125 [08:53<03:47,  6.16s/it]
Training Epoch: 9/100, batch 176/250 completed (loss: 0.1589:  71%|███████   | 89/125 [08:56<03:37,  6.06s/it]
Training Epoch: 9/100, batch 176/250 completed (loss: 0.1179:  71%|███████   | 89/125 [08:56<03:37,  6.06s/it]
Training Epoch: 9/100, batch 176/250 completed (loss: 0.0207:  71%|███████   | 89/125 [08:56<03:38,  6.06s/it]
Training Epoch: 9/100, batch 176/250 completed (loss: 0.1903:  71%|███████   | 89/125 [08:56<03:37,  6.06s/it]
Training Epoch: 9/100, batch 177/250 completed (loss: 0.1851:  71%|███████   | 89/125 [08:56<03:37,  6.06s/it]
Training Epoch: 9/100, batch 177/250 completed (loss: 0.0965:  71%|███████   | 89/125 [08:56<03:38,  6.06s/it]
Training Epoch: 9/100, batch 177/250 completed (loss: 0.0197:  71%|███████   | 89/125 [08:56<03:37,  6.06s/it]
Training Epoch: 9/100, batch 177/250 completed (loss: 0.3656:  71%|███████   | 89/125 [08:56<03:37,  6.06s/it]
Training Epoch: 9/100, batch 178/250 completed (loss: 0.1501:  71%|███████   | 89/125 [08:59<03:37,  6.06s/it]
Training Epoch: 9/100, batch 178/250 completed (loss: 0.1524:  71%|███████   | 89/125 [08:59<03:37,  6.06s/it]
Training Epoch: 9/100, batch 178/250 completed (loss: 0.1577:  71%|███████   | 89/125 [08:59<03:38,  6.06s/it]
Training Epoch: 9/100, batch 178/250 completed (loss: 0.0325:  71%|███████   | 89/125 [08:59<03:37,  6.06s/it]
Training Epoch: 9/100, batch 178/250 completed (loss: 0.1501:  72%|███████▏  | 90/125 [09:02<03:30,  6.03s/it]
Training Epoch: 9/100, batch 178/250 completed (loss: 0.1524:  72%|███████▏  | 90/125 [09:02<03:30,  6.03s/it]
Training Epoch: 9/100, batch 178/250 completed (loss: 0.1577:  72%|███████▏  | 90/125 [09:02<03:30,  6.03s/it]
Training Epoch: 9/100, batch 178/250 completed (loss: 0.0325:  72%|███████▏  | 90/125 [09:02<03:31,  6.03s/it]
Training Epoch: 9/100, batch 179/250 completed (loss: 0.0661:  72%|███████▏  | 90/125 [09:02<03:30,  6.03s/it]
Training Epoch: 9/100, batch 179/250 completed (loss: 0.5671:  72%|███████▏  | 90/125 [09:02<03:30,  6.03s/it]
Training Epoch: 9/100, batch 179/250 completed (loss: 0.1438:  72%|███████▏  | 90/125 [09:02<03:30,  6.03s/it]
Training Epoch: 9/100, batch 179/250 completed (loss: 0.0974:  72%|███████▏  | 90/125 [09:02<03:31,  6.03s/it]
Training Epoch: 9/100, batch 180/250 completed (loss: 0.072:  72%|███████▏  | 90/125 [09:05<03:30,  6.03s/it] 
Training Epoch: 9/100, batch 180/250 completed (loss: 0.233:  72%|███████▏  | 90/125 [09:05<03:30,  6.03s/it] 
Training Epoch: 9/100, batch 180/250 completed (loss: 0.0528:  72%|███████▏  | 90/125 [09:05<03:30,  6.03s/it]
Training Epoch: 9/100, batch 180/250 completed (loss: 0.1108:  72%|███████▏  | 90/125 [09:05<03:31,  6.03s/it]
Training Epoch: 9/100, batch 180/250 completed (loss: 0.233:  73%|███████▎  | 91/125 [09:08<03:26,  6.08s/it]
Training Epoch: 9/100, batch 180/250 completed (loss: 0.0528:  73%|███████▎  | 91/125 [09:08<03:26,  6.08s/it]
Training Epoch: 9/100, batch 180/250 completed (loss: 0.072:  73%|███████▎  | 91/125 [09:08<03:26,  6.08s/it]
Training Epoch: 9/100, batch 180/250 completed (loss: 0.1108:  73%|███████▎  | 91/125 [09:08<03:26,  6.08s/it]
Training Epoch: 9/100, batch 181/250 completed (loss: 0.0906:  73%|███████▎  | 91/125 [09:08<03:26,  6.08s/it]
Training Epoch: 9/100, batch 181/250 completed (loss: 0.0457:  73%|███████▎  | 91/125 [09:08<03:26,  6.08s/it]
Training Epoch: 9/100, batch 181/250 completed (loss: 0.0612:  73%|███████▎  | 91/125 [09:08<03:26,  6.08s/it]
Training Epoch: 9/100, batch 181/250 completed (loss: 0.0792:  73%|███████▎  | 91/125 [09:09<03:26,  6.08s/it]
Training Epoch: 9/100, batch 182/250 completed (loss: 0.1195:  73%|███████▎  | 91/125 [09:11<03:26,  6.08s/it]
Training Epoch: 9/100, batch 182/250 completed (loss: 0.2233:  73%|███████▎  | 91/125 [09:11<03:26,  6.08s/it]
Training Epoch: 9/100, batch 182/250 completed (loss: 0.3593:  73%|███████▎  | 91/125 [09:11<03:26,  6.08s/it]
Training Epoch: 9/100, batch 182/250 completed (loss: 0.2221:  73%|███████▎  | 91/125 [09:11<03:26,  6.08s/it]
Training Epoch: 9/100, batch 182/250 completed (loss: 0.2233:  74%|███████▎  | 92/125 [09:14<03:18,  6.02s/it]
Training Epoch: 9/100, batch 182/250 completed (loss: 0.1195:  74%|███████▎  | 92/125 [09:14<03:18,  6.02s/it]
Training Epoch: 9/100, batch 182/250 completed (loss: 0.3593:  74%|███████▎  | 92/125 [09:14<03:18,  6.02s/it]
Training Epoch: 9/100, batch 182/250 completed (loss: 0.2221:  74%|███████▎  | 92/125 [09:14<03:18,  6.02s/it]
Training Epoch: 9/100, batch 183/250 completed (loss: 0.1108:  74%|███████▎  | 92/125 [09:14<03:18,  6.02s/it]
Training Epoch: 9/100, batch 183/250 completed (loss: 0.1618:  74%|███████▎  | 92/125 [09:14<03:18,  6.02s/it]
Training Epoch: 9/100, batch 183/250 completed (loss: 0.1591:  74%|███████▎  | 92/125 [09:14<03:18,  6.02s/it]
Training Epoch: 9/100, batch 183/250 completed (loss: 0.0886:  74%|███████▎  | 92/125 [09:14<03:18,  6.02s/it]
Training Epoch: 9/100, batch 184/250 completed (loss: 0.1966:  74%|███████▎  | 92/125 [09:17<03:18,  6.02s/it]
Training Epoch: 9/100, batch 184/250 completed (loss: 0.2153:  74%|███████▎  | 92/125 [09:17<03:18,  6.02s/it]
Training Epoch: 9/100, batch 184/250 completed (loss: 0.1155:  74%|███████▎  | 92/125 [09:17<03:18,  6.02s/it]
Training Epoch: 9/100, batch 184/250 completed (loss: 0.0746:  74%|███████▎  | 92/125 [09:17<03:18,  6.02s/it]
Training Epoch: 9/100, batch 184/250 completed (loss: 0.1966:  74%|███████▍  | 93/125 [09:21<03:15,  6.11s/it]
Training Epoch: 9/100, batch 184/250 completed (loss: 0.0746:  74%|███████▍  | 93/125 [09:21<03:15,  6.11s/it]
Training Epoch: 9/100, batch 184/250 completed (loss: 0.2153:  74%|███████▍  | 93/125 [09:21<03:15,  6.11s/it]
Training Epoch: 9/100, batch 184/250 completed (loss: 0.1155:  74%|███████▍  | 93/125 [09:21<03:15,  6.11s/it]
Training Epoch: 9/100, batch 185/250 completed (loss: 0.1818:  74%|███████▍  | 93/125 [09:21<03:15,  6.11s/it]
Training Epoch: 9/100, batch 185/250 completed (loss: 0.1233:  74%|███████▍  | 93/125 [09:21<03:15,  6.11s/it]
Training Epoch: 9/100, batch 185/250 completed (loss: 0.0873:  74%|███████▍  | 93/125 [09:21<03:15,  6.11s/it]
Training Epoch: 9/100, batch 185/250 completed (loss: 0.113:  74%|███████▍  | 93/125 [09:21<03:15,  6.11s/it] 
Training Epoch: 9/100, batch 186/250 completed (loss: 0.0497:  74%|███████▍  | 93/125 [09:24<03:15,  6.11s/it]
Training Epoch: 9/100, batch 186/250 completed (loss: 0.3492:  74%|███████▍  | 93/125 [09:24<03:15,  6.11s/it]
Training Epoch: 9/100, batch 186/250 completed (loss: 0.2629:  74%|███████▍  | 93/125 [09:24<03:15,  6.11s/it]
Training Epoch: 9/100, batch 186/250 completed (loss: 0.1441:  74%|███████▍  | 93/125 [09:24<03:15,  6.11s/it]
Training Epoch: 9/100, batch 186/250 completed (loss: 0.3492:  75%|███████▌  | 94/125 [09:27<03:08,  6.07s/it]
Training Epoch: 9/100, batch 186/250 completed (loss: 0.2629:  75%|███████▌  | 94/125 [09:27<03:08,  6.07s/it]
Training Epoch: 9/100, batch 186/250 completed (loss: 0.0497:  75%|███████▌  | 94/125 [09:27<03:08,  6.07s/it]
Training Epoch: 9/100, batch 186/250 completed (loss: 0.1441:  75%|███████▌  | 94/125 [09:27<03:08,  6.07s/it]
Training Epoch: 9/100, batch 187/250 completed (loss: 0.0866:  75%|███████▌  | 94/125 [09:27<03:08,  6.07s/it]
Training Epoch: 9/100, batch 187/250 completed (loss: 0.1828:  75%|███████▌  | 94/125 [09:27<03:08,  6.07s/it]
Training Epoch: 9/100, batch 187/250 completed (loss: 0.1366:  75%|███████▌  | 94/125 [09:27<03:08,  6.07s/it]
Training Epoch: 9/100, batch 187/250 completed (loss: 0.0357:  75%|███████▌  | 94/125 [09:27<03:08,  6.07s/it]
Training Epoch: 9/100, batch 188/250 completed (loss: 0.0863:  75%|███████▌  | 94/125 [09:30<03:08,  6.07s/it]
Training Epoch: 9/100, batch 188/250 completed (loss: 0.1351:  75%|███████▌  | 94/125 [09:30<03:08,  6.07s/it]
Training Epoch: 9/100, batch 188/250 completed (loss: 0.1287:  75%|███████▌  | 94/125 [09:30<03:08,  6.07s/it]
Training Epoch: 9/100, batch 188/250 completed (loss: 0.1377:  75%|███████▌  | 94/125 [09:30<03:08,  6.07s/it]
Training Epoch: 9/100, batch 188/250 completed (loss: 0.1351:  76%|███████▌  | 95/125 [09:33<03:01,  6.05s/it]
Training Epoch: 9/100, batch 188/250 completed (loss: 0.1287:  76%|███████▌  | 95/125 [09:33<03:01,  6.05s/it]
Training Epoch: 9/100, batch 188/250 completed (loss: 0.0863:  76%|███████▌  | 95/125 [09:33<03:01,  6.05s/it]
Training Epoch: 9/100, batch 188/250 completed (loss: 0.1377:  76%|███████▌  | 95/125 [09:33<03:01,  6.05s/it]
Training Epoch: 9/100, batch 189/250 completed (loss: 0.1103:  76%|███████▌  | 95/125 [09:33<03:01,  6.05s/it]
Training Epoch: 9/100, batch 189/250 completed (loss: 0.077:  76%|███████▌  | 95/125 [09:33<03:01,  6.05s/it] 
Training Epoch: 9/100, batch 189/250 completed (loss: 0.0845:  76%|███████▌  | 95/125 [09:33<03:01,  6.05s/it]
Training Epoch: 9/100, batch 189/250 completed (loss: 0.2229:  76%|███████▌  | 95/125 [09:33<03:01,  6.05s/it]
Training Epoch: 9/100, batch 190/250 completed (loss: 0.19:  76%|███████▌  | 95/125 [09:36<03:01,  6.05s/it]  
Training Epoch: 9/100, batch 190/250 completed (loss: 0.3096:  76%|███████▌  | 95/125 [09:36<03:01,  6.05s/it]
Training Epoch: 9/100, batch 190/250 completed (loss: 0.1181:  76%|███████▌  | 95/125 [09:36<03:01,  6.05s/it]
Training Epoch: 9/100, batch 190/250 completed (loss: 0.1224:  76%|███████▌  | 95/125 [09:36<03:01,  6.05s/it]
Training Epoch: 9/100, batch 190/250 completed (loss: 0.3096:  77%|███████▋  | 96/125 [09:39<02:58,  6.14s/it]
Training Epoch: 9/100, batch 190/250 completed (loss: 0.19:  77%|███████▋  | 96/125 [09:39<02:58,  6.14s/it]
Training Epoch: 9/100, batch 190/250 completed (loss: 0.1181:  77%|███████▋  | 96/125 [09:39<02:58,  6.14s/it]
Training Epoch: 9/100, batch 190/250 completed (loss: 0.1224:  77%|███████▋  | 96/125 [09:39<02:58,  6.14s/it]
Training Epoch: 9/100, batch 191/250 completed (loss: 0.0508:  77%|███████▋  | 96/125 [09:39<02:58,  6.14s/it]
Training Epoch: 9/100, batch 191/250 completed (loss: 0.153:  77%|███████▋  | 96/125 [09:39<02:58,  6.14s/it]
Training Epoch: 9/100, batch 191/250 completed (loss: 0.2249:  77%|███████▋  | 96/125 [09:39<02:58,  6.14s/it]
Training Epoch: 9/100, batch 191/250 completed (loss: 0.0889:  77%|███████▋  | 96/125 [09:39<02:58,  6.14s/it]
Training Epoch: 9/100, batch 192/250 completed (loss: 0.1745:  77%|███████▋  | 96/125 [09:42<02:58,  6.14s/it]
Training Epoch: 9/100, batch 192/250 completed (loss: 0.1016:  77%|███████▋  | 96/125 [09:42<02:58,  6.14s/it]
Training Epoch: 9/100, batch 192/250 completed (loss: 0.4572:  77%|███████▋  | 96/125 [09:42<02:58,  6.14s/it]
Training Epoch: 9/100, batch 192/250 completed (loss: 0.3086:  77%|███████▋  | 96/125 [09:42<02:58,  6.14s/it]
Training Epoch: 9/100, batch 192/250 completed (loss: 0.1016:  78%|███████▊  | 97/125 [09:45<02:49,  6.06s/it]
Training Epoch: 9/100, batch 192/250 completed (loss: 0.3086:  78%|███████▊  | 97/125 [09:45<02:49,  6.06s/it]
Training Epoch: 9/100, batch 192/250 completed (loss: 0.1745:  78%|███████▊  | 97/125 [09:45<02:49,  6.06s/it]
Training Epoch: 9/100, batch 192/250 completed (loss: 0.4572:  78%|███████▊  | 97/125 [09:45<02:49,  6.06s/it]
Training Epoch: 9/100, batch 193/250 completed (loss: 0.0494:  78%|███████▊  | 97/125 [09:45<02:49,  6.06s/it]
Training Epoch: 9/100, batch 193/250 completed (loss: 0.0736:  78%|███████▊  | 97/125 [09:45<02:49,  6.06s/it]
Training Epoch: 9/100, batch 193/250 completed (loss: 0.0862:  78%|███████▊  | 97/125 [09:45<02:49,  6.06s/it]
Training Epoch: 9/100, batch 193/250 completed (loss: 0.1214:  78%|███████▊  | 97/125 [09:45<02:49,  6.06s/it]
Training Epoch: 9/100, batch 194/250 completed (loss: 0.1815:  78%|███████▊  | 97/125 [09:48<02:49,  6.06s/it]
Training Epoch: 9/100, batch 194/250 completed (loss: 0.2165:  78%|███████▊  | 97/125 [09:48<02:49,  6.06s/it]
Training Epoch: 9/100, batch 194/250 completed (loss: 0.2139:  78%|███████▊  | 97/125 [09:48<02:49,  6.06s/it]
Training Epoch: 9/100, batch 194/250 completed (loss: 0.3324:  78%|███████▊  | 97/125 [09:48<02:49,  6.06s/it]
Training Epoch: 9/100, batch 194/250 completed (loss: 0.3324:  78%|███████▊  | 98/125 [09:51<02:43,  6.04s/it]
Training Epoch: 9/100, batch 194/250 completed (loss: 0.1815:  78%|███████▊  | 98/125 [09:51<02:43,  6.04s/it]
Training Epoch: 9/100, batch 194/250 completed (loss: 0.2139:  78%|███████▊  | 98/125 [09:51<02:43,  6.05s/it]
Training Epoch: 9/100, batch 194/250 completed (loss: 0.2165:  78%|███████▊  | 98/125 [09:51<02:43,  6.05s/it]
Training Epoch: 9/100, batch 195/250 completed (loss: 0.1541:  78%|███████▊  | 98/125 [09:51<02:43,  6.05s/it]
Training Epoch: 9/100, batch 195/250 completed (loss: 0.0783:  78%|███████▊  | 98/125 [09:51<02:43,  6.04s/it]
Training Epoch: 9/100, batch 195/250 completed (loss: 0.0747:  78%|███████▊  | 98/125 [09:51<02:43,  6.05s/it]
Training Epoch: 9/100, batch 195/250 completed (loss: 0.4296:  78%|███████▊  | 98/125 [09:51<02:43,  6.04s/it]
Training Epoch: 9/100, batch 196/250 completed (loss: 0.1051:  78%|███████▊  | 98/125 [09:54<02:43,  6.05s/it]
Training Epoch: 9/100, batch 196/250 completed (loss: 0.029:  78%|███████▊  | 98/125 [09:54<02:43,  6.05s/it] 
Training Epoch: 9/100, batch 196/250 completed (loss: 0.255:  78%|███████▊  | 98/125 [09:54<02:43,  6.04s/it] 
Training Epoch: 9/100, batch 196/250 completed (loss: 0.452:  78%|███████▊  | 98/125 [09:54<02:43,  6.04s/it] 
Training Epoch: 9/100, batch 196/250 completed (loss: 0.029:  79%|███████▉  | 99/125 [09:57<02:38,  6.11s/it]
Training Epoch: 9/100, batch 196/250 completed (loss: 0.452:  79%|███████▉  | 99/125 [09:57<02:38,  6.11s/it]
Training Epoch: 9/100, batch 196/250 completed (loss: 0.255:  79%|███████▉  | 99/125 [09:57<02:38,  6.11s/it]
Training Epoch: 9/100, batch 196/250 completed (loss: 0.1051:  79%|███████▉  | 99/125 [09:57<02:38,  6.11s/it]
Training Epoch: 9/100, batch 197/250 completed (loss: 0.117:  79%|███████▉  | 99/125 [09:57<02:38,  6.11s/it]
Training Epoch: 9/100, batch 197/250 completed (loss: 0.236:  79%|███████▉  | 99/125 [09:57<02:38,  6.11s/it] 
Training Epoch: 9/100, batch 197/250 completed (loss: 0.0567:  79%|███████▉  | 99/125 [09:57<02:38,  6.11s/it]
Training Epoch: 9/100, batch 197/250 completed (loss: 0.1647:  79%|███████▉  | 99/125 [09:57<02:38,  6.11s/it]
Training Epoch: 9/100, batch 198/250 completed (loss: 0.0566:  79%|███████▉  | 99/125 [10:00<02:38,  6.11s/it]
Training Epoch: 9/100, batch 198/250 completed (loss: 0.0861:  79%|███████▉  | 99/125 [10:00<02:38,  6.11s/it]
Training Epoch: 9/100, batch 198/250 completed (loss: 0.2061:  79%|███████▉  | 99/125 [10:00<02:38,  6.11s/it]
Training Epoch: 9/100, batch 198/250 completed (loss: 0.1269:  79%|███████▉  | 99/125 [10:00<02:38,  6.11s/it]
Training Epoch: 9/100, batch 198/250 completed (loss: 0.0861:  80%|████████  | 100/125 [10:03<02:31,  6.08s/it]
Training Epoch: 9/100, batch 198/250 completed (loss: 0.1269:  80%|████████  | 100/125 [10:03<02:32,  6.08s/it]
Training Epoch: 9/100, batch 198/250 completed (loss: 0.0566:  80%|████████  | 100/125 [10:03<02:32,  6.08s/it]
Training Epoch: 9/100, batch 198/250 completed (loss: 0.2061:  80%|████████  | 100/125 [10:03<02:32,  6.09s/it]
Training Epoch: 9/100, batch 199/250 completed (loss: 0.2411:  80%|████████  | 100/125 [10:03<02:32,  6.09s/it]
Training Epoch: 9/100, batch 199/250 completed (loss: 0.3013:  80%|████████  | 100/125 [10:03<02:32,  6.08s/it]
Training Epoch: 9/100, batch 199/250 completed (loss: 0.0215:  80%|████████  | 100/125 [10:03<02:31,  6.08s/it]
Training Epoch: 9/100, batch 199/250 completed (loss: 0.1079:  80%|████████  | 100/125 [10:03<02:32,  6.08s/it]
Training Epoch: 9/100, batch 200/250 completed (loss: 0.1373:  80%|████████  | 100/125 [10:06<02:32,  6.09s/it]
Training Epoch: 9/100, batch 200/250 completed (loss: 0.0519:  80%|████████  | 100/125 [10:06<02:31,  6.08s/it]
Training Epoch: 9/100, batch 200/250 completed (loss: 0.4083:  80%|████████  | 100/125 [10:06<02:32,  6.08s/it]
Training Epoch: 9/100, batch 200/250 completed (loss: 0.071:  80%|████████  | 100/125 [10:06<02:32,  6.08s/it] 
Training Epoch: 9/100, batch 200/250 completed (loss: 0.1373:  81%|████████  | 101/125 [10:09<02:26,  6.11s/it]
Training Epoch: 9/100, batch 200/250 completed (loss: 0.0519:  81%|████████  | 101/125 [10:09<02:26,  6.11s/it]
Training Epoch: 9/100, batch 200/250 completed (loss: 0.4083:  81%|████████  | 101/125 [10:09<02:26,  6.11s/it]
Training Epoch: 9/100, batch 200/250 completed (loss: 0.071:  81%|████████  | 101/125 [10:09<02:26,  6.11s/it]
Training Epoch: 9/100, batch 201/250 completed (loss: 0.0995:  81%|████████  | 101/125 [10:09<02:26,  6.11s/it]
Training Epoch: 9/100, batch 201/250 completed (loss: 0.0921:  81%|████████  | 101/125 [10:09<02:26,  6.11s/it]
Training Epoch: 9/100, batch 201/250 completed (loss: 0.0425:  81%|████████  | 101/125 [10:09<02:26,  6.11s/it]
Training Epoch: 9/100, batch 201/250 completed (loss: 0.0776:  81%|████████  | 101/125 [10:09<02:26,  6.11s/it]
Training Epoch: 9/100, batch 202/250 completed (loss: 0.0288:  81%|████████  | 101/125 [10:12<02:26,  6.11s/it]
Training Epoch: 9/100, batch 202/250 completed (loss: 0.1607:  81%|████████  | 101/125 [10:12<02:26,  6.11s/it]
Training Epoch: 9/100, batch 202/250 completed (loss: 0.0883:  81%|████████  | 101/125 [10:12<02:26,  6.11s/it]
Training Epoch: 9/100, batch 202/250 completed (loss: 0.1256:  81%|████████  | 101/125 [10:12<02:26,  6.11s/it]
Training Epoch: 9/100, batch 202/250 completed (loss: 0.0288:  82%|████████▏ | 102/125 [10:15<02:19,  6.08s/it]
Training Epoch: 9/100, batch 202/250 completed (loss: 0.1256:  82%|████████▏ | 102/125 [10:15<02:19,  6.08s/it]
Training Epoch: 9/100, batch 202/250 completed (loss: 0.1607:  82%|████████▏ | 102/125 [10:15<02:19,  6.08s/it]
Training Epoch: 9/100, batch 202/250 completed (loss: 0.0883:  82%|████████▏ | 102/125 [10:15<02:19,  6.08s/it]
Training Epoch: 9/100, batch 203/250 completed (loss: 0.1674:  82%|████████▏ | 102/125 [10:15<02:19,  6.08s/it]
Training Epoch: 9/100, batch 203/250 completed (loss: 0.094:  82%|████████▏ | 102/125 [10:15<02:19,  6.08s/it] 
Training Epoch: 9/100, batch 203/250 completed (loss: 0.1031:  82%|████████▏ | 102/125 [10:15<02:19,  6.08s/it]
Training Epoch: 9/100, batch 203/250 completed (loss: 0.1207:  82%|████████▏ | 102/125 [10:15<02:19,  6.08s/it]
Training Epoch: 9/100, batch 204/250 completed (loss: 0.1255:  82%|████████▏ | 102/125 [10:18<02:19,  6.08s/it]
Training Epoch: 9/100, batch 204/250 completed (loss: 0.2266:  82%|████████▏ | 102/125 [10:18<02:19,  6.08s/it]
Training Epoch: 9/100, batch 204/250 completed (loss: 0.0901:  82%|████████▏ | 102/125 [10:18<02:19,  6.08s/it]
Training Epoch: 9/100, batch 204/250 completed (loss: 0.1156:  82%|████████▏ | 102/125 [10:18<02:19,  6.08s/it]
Training Epoch: 9/100, batch 204/250 completed (loss: 0.1156:  82%|████████▏ | 103/125 [10:21<02:13,  6.05s/it]
Training Epoch: 9/100, batch 204/250 completed (loss: 0.1255:  82%|████████▏ | 103/125 [10:21<02:13,  6.05s/it]
Training Epoch: 9/100, batch 204/250 completed (loss: 0.2266:  82%|████████▏ | 103/125 [10:21<02:13,  6.05s/it]
Training Epoch: 9/100, batch 204/250 completed (loss: 0.0901:  82%|████████▏ | 103/125 [10:21<02:13,  6.05s/it]
Training Epoch: 9/100, batch 205/250 completed (loss: 0.027:  82%|████████▏ | 103/125 [10:21<02:13,  6.05s/it] 
Training Epoch: 9/100, batch 205/250 completed (loss: 0.1056:  82%|████████▏ | 103/125 [10:21<02:13,  6.05s/it]
Training Epoch: 9/100, batch 205/250 completed (loss: 0.0744:  82%|████████▏ | 103/125 [10:21<02:13,  6.05s/it]
Training Epoch: 9/100, batch 205/250 completed (loss: 0.1956:  82%|████████▏ | 103/125 [10:21<02:13,  6.05s/it]
Training Epoch: 9/100, batch 206/250 completed (loss: 0.1052:  82%|████████▏ | 103/125 [10:25<02:13,  6.05s/it]
Training Epoch: 9/100, batch 206/250 completed (loss: 0.2178:  82%|████████▏ | 103/125 [10:25<02:13,  6.05s/it]
Training Epoch: 9/100, batch 206/250 completed (loss: 0.1053:  82%|████████▏ | 103/125 [10:25<02:13,  6.05s/it]
Training Epoch: 9/100, batch 206/250 completed (loss: 0.0996:  82%|████████▏ | 103/125 [10:25<02:13,  6.05s/it]
Training Epoch: 9/100, batch 206/250 completed (loss: 0.1053:  83%|████████▎ | 104/125 [10:28<02:08,  6.13s/it]
Training Epoch: 9/100, batch 206/250 completed (loss: 0.1052:  83%|████████▎ | 104/125 [10:28<02:08,  6.13s/it]
Training Epoch: 9/100, batch 206/250 completed (loss: 0.2178:  83%|████████▎ | 104/125 [10:28<02:08,  6.13s/it]
Training Epoch: 9/100, batch 206/250 completed (loss: 0.0996:  83%|████████▎ | 104/125 [10:28<02:08,  6.13s/it]
Training Epoch: 9/100, batch 207/250 completed (loss: 0.1342:  83%|████████▎ | 104/125 [10:28<02:08,  6.13s/it]
Training Epoch: 9/100, batch 207/250 completed (loss: 0.1345:  83%|████████▎ | 104/125 [10:28<02:08,  6.13s/it]
Training Epoch: 9/100, batch 207/250 completed (loss: 0.1391:  83%|████████▎ | 104/125 [10:28<02:08,  6.13s/it]
Training Epoch: 9/100, batch 207/250 completed (loss: 0.0577:  83%|████████▎ | 104/125 [10:28<02:08,  6.13s/it]
Training Epoch: 9/100, batch 208/250 completed (loss: 0.174:  83%|████████▎ | 104/125 [10:31<02:08,  6.13s/it] 
Training Epoch: 9/100, batch 208/250 completed (loss: 0.3162:  83%|████████▎ | 104/125 [10:31<02:08,  6.13s/it]
Training Epoch: 9/100, batch 208/250 completed (loss: 0.0611:  83%|████████▎ | 104/125 [10:31<02:08,  6.13s/it]
Training Epoch: 9/100, batch 208/250 completed (loss: 0.3066:  83%|████████▎ | 104/125 [10:31<02:08,  6.13s/it]
Training Epoch: 9/100, batch 208/250 completed (loss: 0.0611:  84%|████████▍ | 105/125 [10:34<02:01,  6.08s/it]
Training Epoch: 9/100, batch 208/250 completed (loss: 0.174:  84%|████████▍ | 105/125 [10:34<02:01,  6.08s/it]
Training Epoch: 9/100, batch 208/250 completed (loss: 0.3162:  84%|████████▍ | 105/125 [10:34<02:01,  6.08s/it]
Training Epoch: 9/100, batch 208/250 completed (loss: 0.3066:  84%|████████▍ | 105/125 [10:34<02:01,  6.08s/it]
Training Epoch: 9/100, batch 209/250 completed (loss: 0.0839:  84%|████████▍ | 105/125 [10:34<02:01,  6.08s/it]
Training Epoch: 9/100, batch 209/250 completed (loss: 0.1546:  84%|████████▍ | 105/125 [10:34<02:01,  6.08s/it]
Training Epoch: 9/100, batch 209/250 completed (loss: 0.1043:  84%|████████▍ | 105/125 [10:34<02:01,  6.08s/it]
Training Epoch: 9/100, batch 209/250 completed (loss: 0.0621:  84%|████████▍ | 105/125 [10:34<02:01,  6.08s/it]
Training Epoch: 9/100, batch 210/250 completed (loss: 0.0825:  84%|████████▍ | 105/125 [10:37<02:01,  6.08s/it]
Training Epoch: 9/100, batch 210/250 completed (loss: 0.2146:  84%|████████▍ | 105/125 [10:37<02:01,  6.08s/it]
Training Epoch: 9/100, batch 210/250 completed (loss: 0.0917:  84%|████████▍ | 105/125 [10:37<02:01,  6.08s/it]
Training Epoch: 9/100, batch 210/250 completed (loss: 0.1227:  84%|████████▍ | 105/125 [10:37<02:01,  6.08s/it]
Training Epoch: 9/100, batch 210/250 completed (loss: 0.2146:  85%|████████▍ | 106/125 [10:40<01:55,  6.07s/it]
Training Epoch: 9/100, batch 210/250 completed (loss: 0.1227:  85%|████████▍ | 106/125 [10:40<01:55,  6.07s/it]
Training Epoch: 9/100, batch 210/250 completed (loss: 0.0825:  85%|████████▍ | 106/125 [10:40<01:55,  6.07s/it]
Training Epoch: 9/100, batch 210/250 completed (loss: 0.0917:  85%|████████▍ | 106/125 [10:40<01:55,  6.07s/it]
Training Epoch: 9/100, batch 211/250 completed (loss: 0.0802:  85%|████████▍ | 106/125 [10:40<01:55,  6.07s/it]
Training Epoch: 9/100, batch 211/250 completed (loss: 0.0659:  85%|████████▍ | 106/125 [10:40<01:55,  6.07s/it]
Training Epoch: 9/100, batch 211/250 completed (loss: 0.1342:  85%|████████▍ | 106/125 [10:40<01:55,  6.07s/it]
Training Epoch: 9/100, batch 211/250 completed (loss: 0.1583:  85%|████████▍ | 106/125 [10:40<01:55,  6.07s/it]
Training Epoch: 9/100, batch 212/250 completed (loss: 0.3005:  85%|████████▍ | 106/125 [10:43<01:55,  6.07s/it]
Training Epoch: 9/100, batch 212/250 completed (loss: 0.2901:  85%|████████▍ | 106/125 [10:43<01:55,  6.07s/it]
Training Epoch: 9/100, batch 212/250 completed (loss: 0.4512:  85%|████████▍ | 106/125 [10:43<01:55,  6.07s/it]
Training Epoch: 9/100, batch 212/250 completed (loss: 0.2861:  85%|████████▍ | 106/125 [10:43<01:55,  6.07s/it]
Training Epoch: 9/100, batch 212/250 completed (loss: 0.4512:  86%|████████▌ | 107/125 [10:46<01:49,  6.09s/it]
Training Epoch: 9/100, batch 212/250 completed (loss: 0.2861:  86%|████████▌ | 107/125 [10:46<01:49,  6.09s/it]
Training Epoch: 9/100, batch 212/250 completed (loss: 0.2901:  86%|████████▌ | 107/125 [10:46<01:49,  6.09s/it]
Training Epoch: 9/100, batch 212/250 completed (loss: 0.3005:  86%|████████▌ | 107/125 [10:46<01:49,  6.10s/it]
Training Epoch: 9/100, batch 213/250 completed (loss: 0.0646:  86%|████████▌ | 107/125 [10:46<01:49,  6.10s/it]
Training Epoch: 9/100, batch 213/250 completed (loss: 0.1985:  86%|████████▌ | 107/125 [10:46<01:49,  6.09s/it]
Training Epoch: 9/100, batch 213/250 completed (loss: 0.1565:  86%|████████▌ | 107/125 [10:46<01:49,  6.09s/it]
Training Epoch: 9/100, batch 213/250 completed (loss: 0.1053:  86%|████████▌ | 107/125 [10:46<01:49,  6.09s/it]
Training Epoch: 9/100, batch 214/250 completed (loss: 0.233:  86%|████████▌ | 107/125 [10:49<01:49,  6.09s/it] 
Training Epoch: 9/100, batch 214/250 completed (loss: 0.1067:  86%|████████▌ | 107/125 [10:49<01:49,  6.09s/it]
Training Epoch: 9/100, batch 214/250 completed (loss: 0.2754:  86%|████████▌ | 107/125 [10:49<01:49,  6.10s/it]
Training Epoch: 9/100, batch 214/250 completed (loss: 0.0795:  86%|████████▌ | 107/125 [10:49<01:49,  6.09s/it]
Training Epoch: 9/100, batch 214/250 completed (loss: 0.233:  86%|████████▋ | 108/125 [10:52<01:43,  6.06s/it]
Training Epoch: 9/100, batch 214/250 completed (loss: 0.0795:  86%|████████▋ | 108/125 [10:52<01:43,  6.07s/it]
Training Epoch: 9/100, batch 214/250 completed (loss: 0.2754:  86%|████████▋ | 108/125 [10:52<01:43,  6.07s/it]
Training Epoch: 9/100, batch 214/250 completed (loss: 0.1067:  86%|████████▋ | 108/125 [10:52<01:43,  6.07s/it]
Training Epoch: 9/100, batch 215/250 completed (loss: 0.2317:  86%|████████▋ | 108/125 [10:52<01:43,  6.07s/it]
Training Epoch: 9/100, batch 215/250 completed (loss: 0.2693:  86%|████████▋ | 108/125 [10:52<01:43,  6.07s/it]
Training Epoch: 9/100, batch 215/250 completed (loss: 0.4134:  86%|████████▋ | 108/125 [10:52<01:43,  6.06s/it]
Training Epoch: 9/100, batch 215/250 completed (loss: 0.098:  86%|████████▋ | 108/125 [10:52<01:43,  6.07s/it] 
Training Epoch: 9/100, batch 216/250 completed (loss: 0.3064:  86%|████████▋ | 108/125 [10:55<01:43,  6.06s/it]
Training Epoch: 9/100, batch 216/250 completed (loss: 0.0162:  86%|████████▋ | 108/125 [10:55<01:43,  6.07s/it]
Training Epoch: 9/100, batch 216/250 completed (loss: 0.4558:  86%|████████▋ | 108/125 [10:55<01:43,  6.07s/it]
Training Epoch: 9/100, batch 216/250 completed (loss: 0.0877:  86%|████████▋ | 108/125 [10:55<01:43,  6.07s/it]
Training Epoch: 9/100, batch 216/250 completed (loss: 0.3064:  87%|████████▋ | 109/125 [10:58<01:37,  6.07s/it]
Training Epoch: 9/100, batch 216/250 completed (loss: 0.0877:  87%|████████▋ | 109/125 [10:58<01:37,  6.08s/it]
Training Epoch: 9/100, batch 216/250 completed (loss: 0.4558:  87%|████████▋ | 109/125 [10:58<01:37,  6.08s/it]
Training Epoch: 9/100, batch 216/250 completed (loss: 0.0162:  87%|████████▋ | 109/125 [10:58<01:37,  6.08s/it]
Training Epoch: 9/100, batch 217/250 completed (loss: 0.0669:  87%|████████▋ | 109/125 [10:58<01:37,  6.08s/it]
Training Epoch: 9/100, batch 217/250 completed (loss: 0.1219:  87%|████████▋ | 109/125 [10:58<01:37,  6.07s/it]
Training Epoch: 9/100, batch 217/250 completed (loss: 0.1769:  87%|████████▋ | 109/125 [10:58<01:37,  6.08s/it]
Training Epoch: 9/100, batch 217/250 completed (loss: 0.1637:  87%|████████▋ | 109/125 [10:58<01:37,  6.08s/it]
Training Epoch: 9/100, batch 218/250 completed (loss: 0.0316:  87%|████████▋ | 109/125 [11:01<01:37,  6.08s/it]
Training Epoch: 9/100, batch 218/250 completed (loss: 0.717:  87%|████████▋ | 109/125 [11:01<01:37,  6.08s/it] 
Training Epoch: 9/100, batch 218/250 completed (loss: 0.3267:  87%|████████▋ | 109/125 [11:01<01:37,  6.07s/it]
Training Epoch: 9/100, batch 218/250 completed (loss: 0.3155:  87%|████████▋ | 109/125 [11:01<01:37,  6.08s/it]
Training Epoch: 9/100, batch 218/250 completed (loss: 0.717:  88%|████████▊ | 110/125 [11:04<01:30,  6.05s/it]
Training Epoch: 9/100, batch 218/250 completed (loss: 0.3267:  88%|████████▊ | 110/125 [11:04<01:30,  6.05s/it]
Training Epoch: 9/100, batch 218/250 completed (loss: 0.0316:  88%|████████▊ | 110/125 [11:04<01:30,  6.05s/it]
Training Epoch: 9/100, batch 218/250 completed (loss: 0.3155:  88%|████████▊ | 110/125 [11:04<01:30,  6.06s/it]
Training Epoch: 9/100, batch 219/250 completed (loss: 0.0678:  88%|████████▊ | 110/125 [11:04<01:30,  6.05s/it]
Training Epoch: 9/100, batch 219/250 completed (loss: 0.2068:  88%|████████▊ | 110/125 [11:04<01:30,  6.05s/it]
Training Epoch: 9/100, batch 219/250 completed (loss: 0.0994:  88%|████████▊ | 110/125 [11:04<01:30,  6.05s/it]
Training Epoch: 9/100, batch 219/250 completed (loss: 0.1227:  88%|████████▊ | 110/125 [11:04<01:30,  6.06s/it]
Training Epoch: 9/100, batch 220/250 completed (loss: 0.0704:  88%|████████▊ | 110/125 [11:07<01:30,  6.05s/it]
Training Epoch: 9/100, batch 220/250 completed (loss: 0.2975:  88%|████████▊ | 110/125 [11:07<01:30,  6.05s/it]
Training Epoch: 9/100, batch 220/250 completed (loss: 0.1519:  88%|████████▊ | 110/125 [11:07<01:30,  6.05s/it]
Training Epoch: 9/100, batch 220/250 completed (loss: 0.121:  88%|████████▊ | 110/125 [11:07<01:30,  6.06s/it] 
Training Epoch: 9/100, batch 220/250 completed (loss: 0.2975:  89%|████████▉ | 111/125 [11:10<01:24,  6.03s/it]
Training Epoch: 9/100, batch 220/250 completed (loss: 0.0704:  89%|████████▉ | 111/125 [11:10<01:24,  6.03s/it]
Training Epoch: 9/100, batch 220/250 completed (loss: 0.121:  89%|████████▉ | 111/125 [11:10<01:24,  6.03s/it]
Training Epoch: 9/100, batch 220/250 completed (loss: 0.1519:  89%|████████▉ | 111/125 [11:10<01:24,  6.03s/it]
Training Epoch: 9/100, batch 221/250 completed (loss: 0.1216:  89%|████████▉ | 111/125 [11:10<01:24,  6.03s/it]
Training Epoch: 9/100, batch 221/250 completed (loss: 0.1462:  89%|████████▉ | 111/125 [11:10<01:24,  6.03s/it]
Training Epoch: 9/100, batch 221/250 completed (loss: 0.273:  89%|████████▉ | 111/125 [11:10<01:24,  6.03s/it] 
Training Epoch: 9/100, batch 221/250 completed (loss: 0.0476:  89%|████████▉ | 111/125 [11:10<01:24,  6.03s/it]
Training Epoch: 9/100, batch 222/250 completed (loss: 0.0217:  89%|████████▉ | 111/125 [11:13<01:24,  6.03s/it]
Training Epoch: 9/100, batch 222/250 completed (loss: 0.2526:  89%|████████▉ | 111/125 [11:13<01:24,  6.03s/it]
Training Epoch: 9/100, batch 222/250 completed (loss: 0.1552:  89%|████████▉ | 111/125 [11:13<01:24,  6.03s/it]
Training Epoch: 9/100, batch 222/250 completed (loss: 0.1375:  89%|████████▉ | 111/125 [11:13<01:24,  6.03s/it]
Training Epoch: 9/100, batch 222/250 completed (loss: 0.1552:  90%|████████▉ | 112/125 [11:16<01:19,  6.12s/it]
Training Epoch: 9/100, batch 222/250 completed (loss: 0.2526:  90%|████████▉ | 112/125 [11:16<01:19,  6.12s/it]
Training Epoch: 9/100, batch 222/250 completed (loss: 0.0217:  90%|████████▉ | 112/125 [11:16<01:19,  6.12s/it]
Training Epoch: 9/100, batch 222/250 completed (loss: 0.1375:  90%|████████▉ | 112/125 [11:16<01:19,  6.12s/it]
Training Epoch: 9/100, batch 223/250 completed (loss: 0.1712:  90%|████████▉ | 112/125 [11:16<01:19,  6.12s/it]
Training Epoch: 9/100, batch 223/250 completed (loss: 0.2207:  90%|████████▉ | 112/125 [11:16<01:19,  6.12s/it]
Training Epoch: 9/100, batch 223/250 completed (loss: 0.0301:  90%|████████▉ | 112/125 [11:16<01:19,  6.12s/it]
Training Epoch: 9/100, batch 223/250 completed (loss: 0.357:  90%|████████▉ | 112/125 [11:16<01:19,  6.12s/it] 
Training Epoch: 9/100, batch 224/250 completed (loss: 0.093:  90%|████████▉ | 112/125 [11:19<01:19,  6.12s/it] 
Training Epoch: 9/100, batch 224/250 completed (loss: 0.0903:  90%|████████▉ | 112/125 [11:19<01:19,  6.12s/it]
Training Epoch: 9/100, batch 224/250 completed (loss: 0.0602:  90%|████████▉ | 112/125 [11:19<01:19,  6.12s/it]
Training Epoch: 9/100, batch 224/250 completed (loss: 0.0445:  90%|████████▉ | 112/125 [11:19<01:19,  6.12s/it]
Training Epoch: 9/100, batch 224/250 completed (loss: 0.0903:  90%|█████████ | 113/125 [11:22<01:12,  6.08s/it]
Training Epoch: 9/100, batch 224/250 completed (loss: 0.093:  90%|█████████ | 113/125 [11:22<01:12,  6.08s/it]
Training Epoch: 9/100, batch 224/250 completed (loss: 0.0602:  90%|█████████ | 113/125 [11:22<01:12,  6.08s/it]
Training Epoch: 9/100, batch 224/250 completed (loss: 0.0445:  90%|█████████ | 113/125 [11:22<01:12,  6.08s/it]
Training Epoch: 9/100, batch 225/250 completed (loss: 0.2566:  90%|█████████ | 113/125 [11:22<01:12,  6.08s/it]
Training Epoch: 9/100, batch 225/250 completed (loss: 0.3533:  90%|█████████ | 113/125 [11:22<01:12,  6.08s/it]
Training Epoch: 9/100, batch 225/250 completed (loss: 0.1928:  90%|█████████ | 113/125 [11:22<01:12,  6.08s/it]
Training Epoch: 9/100, batch 225/250 completed (loss: 0.3458:  90%|█████████ | 113/125 [11:22<01:12,  6.08s/it]
Training Epoch: 9/100, batch 226/250 completed (loss: 0.0941:  90%|█████████ | 113/125 [11:25<01:12,  6.08s/it]
Training Epoch: 9/100, batch 226/250 completed (loss: 0.102:  90%|█████████ | 113/125 [11:25<01:12,  6.08s/it] 
Training Epoch: 9/100, batch 226/250 completed (loss: 0.1459:  90%|█████████ | 113/125 [11:25<01:12,  6.08s/it]
Training Epoch: 9/100, batch 226/250 completed (loss: 0.0239:  90%|█████████ | 113/125 [11:25<01:12,  6.08s/it]
Training Epoch: 9/100, batch 226/250 completed (loss: 0.1459:  91%|█████████ | 114/125 [11:28<01:07,  6.09s/it]
Training Epoch: 9/100, batch 226/250 completed (loss: 0.102:  91%|█████████ | 114/125 [11:28<01:07,  6.10s/it]
Training Epoch: 9/100, batch 226/250 completed (loss: 0.0941:  91%|█████████ | 114/125 [11:28<01:07,  6.10s/it]
Training Epoch: 9/100, batch 226/250 completed (loss: 0.0239:  91%|█████████ | 114/125 [11:28<01:07,  6.10s/it]
Training Epoch: 9/100, batch 227/250 completed (loss: 0.1049:  91%|█████████ | 114/125 [11:28<01:07,  6.09s/it]
Training Epoch: 9/100, batch 227/250 completed (loss: 0.3157:  91%|█████████ | 114/125 [11:28<01:07,  6.10s/it]
Training Epoch: 9/100, batch 227/250 completed (loss: 0.1677:  91%|█████████ | 114/125 [11:28<01:07,  6.10s/it]
Training Epoch: 9/100, batch 227/250 completed (loss: 0.0582:  91%|█████████ | 114/125 [11:28<01:07,  6.10s/it]
Training Epoch: 9/100, batch 228/250 completed (loss: 0.2283:  91%|█████████ | 114/125 [11:31<01:07,  6.10s/it]
Training Epoch: 9/100, batch 228/250 completed (loss: 0.0527:  91%|█████████ | 114/125 [11:31<01:07,  6.09s/it]
Training Epoch: 9/100, batch 228/250 completed (loss: 0.1378:  91%|█████████ | 114/125 [11:31<01:07,  6.10s/it]
Training Epoch: 9/100, batch 228/250 completed (loss: 0.2129:  91%|█████████ | 114/125 [11:31<01:07,  6.10s/it]
Training Epoch: 9/100, batch 228/250 completed (loss: 0.1378:  92%|█████████▏| 115/125 [11:34<01:00,  6.07s/it]
Training Epoch: 9/100, batch 228/250 completed (loss: 0.2283:  92%|█████████▏| 115/125 [11:34<01:00,  6.07s/it]
Training Epoch: 9/100, batch 228/250 completed (loss: 0.0527:  92%|█████████▏| 115/125 [11:34<01:00,  6.07s/it]
Training Epoch: 9/100, batch 228/250 completed (loss: 0.2129:  92%|█████████▏| 115/125 [11:34<01:00,  6.07s/it]
Training Epoch: 9/100, batch 229/250 completed (loss: 0.0528:  92%|█████████▏| 115/125 [11:34<01:00,  6.07s/it]
Training Epoch: 9/100, batch 229/250 completed (loss: 0.2806:  92%|█████████▏| 115/125 [11:34<01:00,  6.07s/it]
Training Epoch: 9/100, batch 229/250 completed (loss: 0.1303:  92%|█████████▏| 115/125 [11:34<01:00,  6.07s/it]
Training Epoch: 9/100, batch 229/250 completed (loss: 0.2024:  92%|█████████▏| 115/125 [11:34<01:00,  6.07s/it]
Training Epoch: 9/100, batch 230/250 completed (loss: 0.1094:  92%|█████████▏| 115/125 [11:37<01:00,  6.07s/it]
Training Epoch: 9/100, batch 230/250 completed (loss: 0.0505:  92%|█████████▏| 115/125 [11:37<01:00,  6.07s/it]
Training Epoch: 9/100, batch 230/250 completed (loss: 0.1051:  92%|█████████▏| 115/125 [11:37<01:00,  6.07s/it]
Training Epoch: 9/100, batch 230/250 completed (loss: 0.1154:  92%|█████████▏| 115/125 [11:37<01:00,  6.07s/it]
Training Epoch: 9/100, batch 230/250 completed (loss: 0.1051:  93%|█████████▎| 116/125 [11:40<00:54,  6.05s/it]
Training Epoch: 9/100, batch 230/250 completed (loss: 0.1094:  93%|█████████▎| 116/125 [11:40<00:54,  6.05s/it]
Training Epoch: 9/100, batch 230/250 completed (loss: 0.1154:  93%|█████████▎| 116/125 [11:40<00:54,  6.05s/it]
Training Epoch: 9/100, batch 230/250 completed (loss: 0.0505:  93%|█████████▎| 116/125 [11:40<00:54,  6.06s/it]
Training Epoch: 9/100, batch 231/250 completed (loss: 0.1236:  93%|█████████▎| 116/125 [11:40<00:54,  6.05s/it]
Training Epoch: 9/100, batch 231/250 completed (loss: 0.0613:  93%|█████████▎| 116/125 [11:40<00:54,  6.05s/it]
Training Epoch: 9/100, batch 231/250 completed (loss: 0.2408:  93%|█████████▎| 116/125 [11:40<00:54,  6.06s/it]
Training Epoch: 9/100, batch 231/250 completed (loss: 0.1026:  93%|█████████▎| 116/125 [11:40<00:54,  6.05s/it]
Training Epoch: 9/100, batch 232/250 completed (loss: 0.2794:  93%|█████████▎| 116/125 [11:43<00:54,  6.06s/it]
Training Epoch: 9/100, batch 232/250 completed (loss: 0.1028:  93%|█████████▎| 116/125 [11:43<00:54,  6.05s/it]
Training Epoch: 9/100, batch 232/250 completed (loss: 0.2393:  93%|█████████▎| 116/125 [11:43<00:54,  6.05s/it]
Training Epoch: 9/100, batch 232/250 completed (loss: 0.1113:  93%|█████████▎| 116/125 [11:43<00:54,  6.05s/it]
Training Epoch: 9/100, batch 232/250 completed (loss: 0.2393:  94%|█████████▎| 117/125 [11:46<00:48,  6.08s/it]
Training Epoch: 9/100, batch 232/250 completed (loss: 0.2794:  94%|█████████▎| 117/125 [11:46<00:48,  6.08s/it]
Training Epoch: 9/100, batch 232/250 completed (loss: 0.1028:  94%|█████████▎| 117/125 [11:46<00:48,  6.08s/it]
Training Epoch: 9/100, batch 232/250 completed (loss: 0.1113:  94%|█████████▎| 117/125 [11:46<00:48,  6.08s/it]
Training Epoch: 9/100, batch 233/250 completed (loss: 0.4575:  94%|█████████▎| 117/125 [11:47<00:48,  6.08s/it]
Training Epoch: 9/100, batch 233/250 completed (loss: 0.1124:  94%|█████████▎| 117/125 [11:46<00:48,  6.08s/it]
Training Epoch: 9/100, batch 233/250 completed (loss: 0.0304:  94%|█████████▎| 117/125 [11:47<00:48,  6.08s/it]
Training Epoch: 9/100, batch 233/250 completed (loss: 0.2315:  94%|█████████▎| 117/125 [11:47<00:48,  6.08s/it]
Training Epoch: 9/100, batch 234/250 completed (loss: 0.1531:  94%|█████████▎| 117/125 [11:50<00:48,  6.08s/it]
Training Epoch: 9/100, batch 234/250 completed (loss: 0.135:  94%|█████████▎| 117/125 [11:50<00:48,  6.08s/it] 
Training Epoch: 9/100, batch 234/250 completed (loss: 0.0481:  94%|█████████▎| 117/125 [11:50<00:48,  6.08s/it]
Training Epoch: 9/100, batch 234/250 completed (loss: 0.0328:  94%|█████████▎| 117/125 [11:50<00:48,  6.08s/it]
Training Epoch: 9/100, batch 234/250 completed (loss: 0.135:  94%|█████████▍| 118/125 [11:52<00:42,  6.06s/it]
Training Epoch: 9/100, batch 234/250 completed (loss: 0.0481:  94%|█████████▍| 118/125 [11:52<00:42,  6.06s/it]
Training Epoch: 9/100, batch 234/250 completed (loss: 0.1531:  94%|█████████▍| 118/125 [11:52<00:42,  6.06s/it]
Training Epoch: 9/100, batch 234/250 completed (loss: 0.0328:  94%|█████████▍| 118/125 [11:52<00:42,  6.06s/it]
Training Epoch: 9/100, batch 235/250 completed (loss: 0.1002:  94%|█████████▍| 118/125 [11:53<00:42,  6.06s/it]
Training Epoch: 9/100, batch 235/250 completed (loss: 0.0414:  94%|█████████▍| 118/125 [11:53<00:42,  6.06s/it]
Training Epoch: 9/100, batch 235/250 completed (loss: 0.117:  94%|█████████▍| 118/125 [11:53<00:42,  6.06s/it] 
Training Epoch: 9/100, batch 235/250 completed (loss: 0.3162:  94%|█████████▍| 118/125 [11:53<00:42,  6.06s/it]
Training Epoch: 9/100, batch 236/250 completed (loss: 0.1525:  94%|█████████▍| 118/125 [11:56<00:42,  6.06s/it]
Training Epoch: 9/100, batch 236/250 completed (loss: 0.0329:  94%|█████████▍| 118/125 [11:56<00:42,  6.06s/it]
Training Epoch: 9/100, batch 236/250 completed (loss: 0.4085:  94%|█████████▍| 118/125 [11:56<00:42,  6.06s/it]
Training Epoch: 9/100, batch 236/250 completed (loss: 0.2424:  94%|█████████▍| 118/125 [11:56<00:42,  6.06s/it]
Training Epoch: 9/100, batch 236/250 completed (loss: 0.1525:  95%|█████████▌| 119/125 [11:59<00:36,  6.08s/it]
Training Epoch: 9/100, batch 236/250 completed (loss: 0.4085:  95%|█████████▌| 119/125 [11:59<00:36,  6.08s/it]
Training Epoch: 9/100, batch 236/250 completed (loss: 0.2424:  95%|█████████▌| 119/125 [11:59<00:36,  6.08s/it]
Training Epoch: 9/100, batch 236/250 completed (loss: 0.0329:  95%|█████████▌| 119/125 [11:59<00:36,  6.08s/it]
Training Epoch: 9/100, batch 237/250 completed (loss: 0.1208:  95%|█████████▌| 119/125 [11:59<00:36,  6.08s/it]
Training Epoch: 9/100, batch 237/250 completed (loss: 0.0364:  95%|█████████▌| 119/125 [11:59<00:36,  6.08s/it]
Training Epoch: 9/100, batch 237/250 completed (loss: 0.0283:  95%|█████████▌| 119/125 [11:59<00:36,  6.08s/it]
Training Epoch: 9/100, batch 237/250 completed (loss: 0.1644:  95%|█████████▌| 119/125 [11:59<00:36,  6.08s/it]
Training Epoch: 9/100, batch 238/250 completed (loss: 0.0126:  95%|█████████▌| 119/125 [12:02<00:36,  6.08s/it]
Training Epoch: 9/100, batch 238/250 completed (loss: 0.0568:  95%|█████████▌| 119/125 [12:02<00:36,  6.08s/it]
Training Epoch: 9/100, batch 238/250 completed (loss: 0.0454:  95%|█████████▌| 119/125 [12:02<00:36,  6.08s/it]
Training Epoch: 9/100, batch 238/250 completed (loss: 0.163:  95%|█████████▌| 119/125 [12:02<00:36,  6.08s/it] 
Training Epoch: 9/100, batch 238/250 completed (loss: 0.0126:  96%|█████████▌| 120/125 [12:05<00:30,  6.11s/it]
Training Epoch: 9/100, batch 238/250 completed (loss: 0.0454:  96%|█████████▌| 120/125 [12:05<00:30,  6.11s/it]
Training Epoch: 9/100, batch 238/250 completed (loss: 0.0568:  96%|█████████▌| 120/125 [12:05<00:30,  6.11s/it]
Training Epoch: 9/100, batch 238/250 completed (loss: 0.163:  96%|█████████▌| 120/125 [12:05<00:30,  6.11s/it]
Training Epoch: 9/100, batch 239/250 completed (loss: 0.2149:  96%|█████████▌| 120/125 [12:05<00:30,  6.11s/it]
Training Epoch: 9/100, batch 239/250 completed (loss: 0.1454:  96%|█████████▌| 120/125 [12:05<00:30,  6.11s/it]
Training Epoch: 9/100, batch 239/250 completed (loss: 0.1891:  96%|█████████▌| 120/125 [12:05<00:30,  6.11s/it]
Training Epoch: 9/100, batch 239/250 completed (loss: 0.1388:  96%|█████████▌| 120/125 [12:05<00:30,  6.11s/it]
Training Epoch: 9/100, batch 240/250 completed (loss: 0.1283:  96%|█████████▌| 120/125 [12:08<00:30,  6.11s/it]
Training Epoch: 9/100, batch 240/250 completed (loss: 0.0228:  96%|█████████▌| 120/125 [12:08<00:30,  6.11s/it]
Training Epoch: 9/100, batch 240/250 completed (loss: 0.227:  96%|█████████▌| 120/125 [12:08<00:30,  6.11s/it] 
Training Epoch: 9/100, batch 240/250 completed (loss: 0.0681:  96%|█████████▌| 120/125 [12:08<00:30,  6.11s/it]
Training Epoch: 9/100, batch 240/250 completed (loss: 0.227:  97%|█████████▋| 121/125 [12:11<00:24,  6.07s/it]
Training Epoch: 9/100, batch 240/250 completed (loss: 0.1283:  97%|█████████▋| 121/125 [12:11<00:24,  6.07s/it]
Training Epoch: 9/100, batch 240/250 completed (loss: 0.0228:  97%|█████████▋| 121/125 [12:11<00:24,  6.07s/it]
Training Epoch: 9/100, batch 240/250 completed (loss: 0.0681:  97%|█████████▋| 121/125 [12:11<00:24,  6.08s/it]
Training Epoch: 9/100, batch 241/250 completed (loss: 0.6345:  97%|█████████▋| 121/125 [12:11<00:24,  6.07s/it]
Training Epoch: 9/100, batch 241/250 completed (loss: 0.2443:  97%|█████████▋| 121/125 [12:11<00:24,  6.07s/it]
Training Epoch: 9/100, batch 241/250 completed (loss: 0.0716:  97%|█████████▋| 121/125 [12:11<00:24,  6.07s/it]
Training Epoch: 9/100, batch 241/250 completed (loss: 0.3991:  97%|█████████▋| 121/125 [12:11<00:24,  6.08s/it]
Training Epoch: 9/100, batch 242/250 completed (loss: 0.1734:  97%|█████████▋| 121/125 [12:14<00:24,  6.07s/it]
Training Epoch: 9/100, batch 242/250 completed (loss: 0.1948:  97%|█████████▋| 121/125 [12:14<00:24,  6.07s/it]
Training Epoch: 9/100, batch 242/250 completed (loss: 0.3917:  97%|█████████▋| 121/125 [12:14<00:24,  6.07s/it]
Training Epoch: 9/100, batch 242/250 completed (loss: 0.1728:  97%|█████████▋| 121/125 [12:14<00:24,  6.08s/it]
Training Epoch: 9/100, batch 242/250 completed (loss: 0.3917:  98%|█████████▊| 122/125 [12:17<00:18,  6.09s/it]
Training Epoch: 9/100, batch 242/250 completed (loss: 0.1948:  98%|█████████▊| 122/125 [12:17<00:18,  6.09s/it]
Training Epoch: 9/100, batch 242/250 completed (loss: 0.1734:  98%|█████████▊| 122/125 [12:17<00:18,  6.09s/it]
Training Epoch: 9/100, batch 242/250 completed (loss: 0.1728:  98%|█████████▊| 122/125 [12:17<00:18,  6.09s/it]
Training Epoch: 9/100, batch 243/250 completed (loss: 0.4217:  98%|█████████▊| 122/125 [12:17<00:18,  6.09s/it]
Training Epoch: 9/100, batch 243/250 completed (loss: 0.1559:  98%|█████████▊| 122/125 [12:17<00:18,  6.09s/it]
Training Epoch: 9/100, batch 243/250 completed (loss: 0.1094:  98%|█████████▊| 122/125 [12:17<00:18,  6.09s/it]
Training Epoch: 9/100, batch 243/250 completed (loss: 0.1609:  98%|█████████▊| 122/125 [12:17<00:18,  6.09s/it]
Training Epoch: 9/100, batch 244/250 completed (loss: 0.0755:  98%|█████████▊| 122/125 [12:20<00:18,  6.09s/it]
Training Epoch: 9/100, batch 244/250 completed (loss: 0.1468:  98%|█████████▊| 122/125 [12:20<00:18,  6.09s/it]
Training Epoch: 9/100, batch 244/250 completed (loss: 0.3993:  98%|█████████▊| 122/125 [12:20<00:18,  6.09s/it]
Training Epoch: 9/100, batch 244/250 completed (loss: 0.0983:  98%|█████████▊| 122/125 [12:20<00:18,  6.09s/it]
Training Epoch: 9/100, batch 244/250 completed (loss: 0.1468:  98%|█████████▊| 123/125 [12:23<00:12,  6.07s/it]
Training Epoch: 9/100, batch 244/250 completed (loss: 0.3993:  98%|█████████▊| 123/125 [12:23<00:12,  6.07s/it]
Training Epoch: 9/100, batch 244/250 completed (loss: 0.0755:  98%|█████████▊| 123/125 [12:23<00:12,  6.07s/it]
Training Epoch: 9/100, batch 244/250 completed (loss: 0.0983:  98%|█████████▊| 123/125 [12:23<00:12,  6.07s/it]
Training Epoch: 9/100, batch 245/250 completed (loss: 0.3965:  98%|█████████▊| 123/125 [12:23<00:12,  6.07s/it]
Training Epoch: 9/100, batch 245/250 completed (loss: 0.323:  98%|█████████▊| 123/125 [12:23<00:12,  6.07s/it] 
Training Epoch: 9/100, batch 245/250 completed (loss: 0.138:  98%|█████████▊| 123/125 [12:23<00:12,  6.07s/it] 
Training Epoch: 9/100, batch 245/250 completed (loss: 0.1324:  98%|█████████▊| 123/125 [12:23<00:12,  6.07s/it]
Training Epoch: 9/100, batch 246/250 completed (loss: 0.0842:  98%|█████████▊| 123/125 [12:26<00:12,  6.07s/it]
Training Epoch: 9/100, batch 246/250 completed (loss: 0.1675:  98%|█████████▊| 123/125 [12:26<00:12,  6.07s/it]
Training Epoch: 9/100, batch 246/250 completed (loss: 0.0841:  98%|█████████▊| 123/125 [12:26<00:12,  6.07s/it]
Training Epoch: 9/100, batch 246/250 completed (loss: 0.0575:  98%|█████████▊| 123/125 [12:26<00:12,  6.07s/it]
Training Epoch: 9/100, batch 246/250 completed (loss: 0.1675:  99%|█████████▉| 124/125 [12:29<00:06,  6.03s/it]
Training Epoch: 9/100, batch 246/250 completed (loss: 0.0841:  99%|█████████▉| 124/125 [12:29<00:06,  6.03s/it]
Training Epoch: 9/100, batch 246/250 completed (loss: 0.0842:  99%|█████████▉| 124/125 [12:29<00:06,  6.03s/it]
Training Epoch: 9/100, batch 246/250 completed (loss: 0.0575:  99%|█████████▉| 124/125 [12:29<00:06,  6.03s/it]
Training Epoch: 9/100, batch 247/250 completed (loss: 0.4499:  99%|█████████▉| 124/125 [12:29<00:06,  6.03s/it]
Training Epoch: 9/100, batch 247/250 completed (loss: 0.0549:  99%|█████████▉| 124/125 [12:29<00:06,  6.03s/it]
Training Epoch: 9/100, batch 247/250 completed (loss: 0.0536:  99%|█████████▉| 124/125 [12:29<00:06,  6.03s/it]
Training Epoch: 9/100, batch 247/250 completed (loss: 0.2681:  99%|█████████▉| 124/125 [12:29<00:06,  6.03s/it]
Training Epoch: 9/100, batch 248/250 completed (loss: 0.071:  99%|█████████▉| 124/125 [12:32<00:06,  6.03s/it] 
Training Epoch: 9/100, batch 248/250 completed (loss: 0.0311:  99%|█████████▉| 124/125 [12:32<00:06,  6.03s/it]
Training Epoch: 9/100, batch 248/250 completed (loss: 0.2945:  99%|█████████▉| 124/125 [12:32<00:06,  6.03s/it]
Training Epoch: 9/100, batch 248/250 completed (loss: 0.1762:  99%|█████████▉| 124/125 [12:32<00:06,  6.03s/it]
Training Epoch: 9/100, batch 248/250 completed (loss: 0.0311: 100%|██████████| 125/125 [12:35<00:00,  6.07s/it]
Training Epoch: 9/100, batch 248/250 completed (loss: 0.2945: 100%|██████████| 125/125 [12:35<00:00,  6.07s/it]
Training Epoch: 9/100, batch 248/250 completed (loss: 0.071: 100%|██████████| 125/125 [12:35<00:00,  6.07s/it]
Training Epoch: 9/100, batch 248/250 completed (loss: 0.1762: 100%|██████████| 125/125 [12:35<00:00,  6.07s/it]
Training Epoch: 9/100, batch 249/250 completed (loss: 0.1144: 100%|██████████| 125/125 [12:35<00:00,  6.07s/it]
Training Epoch: 9/100, batch 249/250 completed (loss: 0.0548: 100%|██████████| 125/125 [12:35<00:00,  6.07s/it]
Training Epoch: 9/100, batch 249/250 completed (loss: 0.0812: 100%|██████████| 125/125 [12:35<00:00,  6.07s/it]
Training Epoch: 9/100, batch 249/250 completed (loss: 0.1371: 100%|██████████| 125/125 [12:35<00:00,  6.07s/it]
Training Epoch: 9/100, batch 249/250 completed (loss: 0.0548: 100%|██████████| 125/125 [12:35<00:00,  6.05s/it]

Training Epoch: 9/100, batch 249/250 completed (loss: 0.0812: 100%|██████████| 125/125 [12:35<00:00,  6.05s/it]

Training Epoch: 9/100, batch 249/250 completed (loss: 0.1144: 100%|██████████| 125/125 [12:35<00:00,  6.05s/it]

Training Epoch: 9/100, batch 249/250 completed (loss: 0.1371: 100%|██████████| 125/125 [12:35<00:00,  6.05s/it]
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/c10d_logger.py:83: UserWarning: barrier(): using the device under current context. You can specify `device_id` in `init_process_group` to mute this warning.
  return func(*args, **kwargs)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
saving model + train state.
Warning: failed to upload checkpoint_9 to HF: (Request ID: Root=1-6a1b7743-293ad4eb193e1b5e222e5807;0c80145c-f582-42db-acad-9f9c68ef2c6b)

403 Forbidden: You need to setup automatic credit recharge in order to upload more data. You can do so at /settings/billing..
Cannot access content at: https://huggingface.co/seyedparsa/thoughtformer-checkpoints.git/info/lfs/objects/batch.
Make sure your token has the correct permissions.
Warning: failed to upload train_state_9 to HF: (Request ID: Root=1-6a1b7747-7e243eb15dbe425a205d0b80;56207c56-d63c-4c22-8b43-e21e6b38ef37)

403 Forbidden: You need to setup automatic credit recharge in order to upload more data. You can do so at /settings/billing..
Cannot access content at: https://huggingface.co/seyedparsa/thoughtformer-checkpoints.git/info/lfs/objects/batch.
Make sure your token has the correct permissions.
Deleted old checkpoint: checkpoint_8
Deleted old checkpoint: train_state_8
[warn] HF upload failed for ckpt_9; kept locally, earlier ckpts deleted.

Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]
Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]
Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]eval loss 0.09124311074614525

Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]
[CORRECT] Example 0 (diff=3)
  Expected : 'Claude'
  Extracted: 'Claude'
  Generated: '### Claude'

Test Accuracy:   3%|▎         | 1/32 [00:06<03:25,  6.62s/it]
Test Accuracy:   3%|▎         | 1/32 [00:06<03:25,  6.62s/it]
Test Accuracy:   3%|▎         | 1/32 [00:06<03:25,  6.62s/it]
Test accuracy: 0.88:   3%|▎         | 1/32 [00:06<03:25,  6.62s/it]
Test accuracy: 0.62:   3%|▎         | 1/32 [00:06<03:25,  6.62s/it]
[WRONG] Example 1 (diff=8)
Test accuracy: 0.62:   3%|▎         | 1/32 [00:06<03:25,  6.62s/it]
  Expected : 'Guy'
  Extracted: 'Irma'
  Generated: '### Irma'

[CORRECT] Example 2 (diff=10)
  Expected : 'Elvis'
  Extracted: 'Elvis'
  Generated: '### Elvis'

[WRONG] Example 3 (diff=8)
  Expected : 'Coral'
  Extracted: 'Claude'
  Generated: '### Claude'

[CORRECT] Example 4 (diff=1)
  Expected : 'Patricia'
  Extracted: 'Patricia'
  Generated: '### Patricia'

Test Accuracy:   3%|▎         | 1/32 [00:06<03:19,  6.45s/it]
Test accuracy: 0.75:   3%|▎         | 1/32 [00:06<03:19,  6.45s/it]
Test accuracy: 0.88:   6%|▋         | 2/32 [00:13<03:18,  6.63s/it]
Test accuracy: 0.62:   6%|▋         | 2/32 [00:13<03:18,  6.63s/it]
Test accuracy: 0.62:   6%|▋         | 2/32 [00:13<03:18,  6.63s/it]
Test accuracy: 0.81:   6%|▋         | 2/32 [00:13<03:18,  6.63s/it]
Test accuracy: 0.75:   6%|▋         | 2/32 [00:13<03:18,  6.63s/it]
Test accuracy: 0.56:   6%|▋         | 2/32 [00:13<03:18,  6.63s/it]
Test accuracy: 0.75:   6%|▋         | 2/32 [00:13<03:16,  6.55s/it]
Test accuracy: 0.69:   6%|▋         | 2/32 [00:13<03:16,  6.55s/it]
Test accuracy: 0.81:   9%|▉         | 3/32 [00:20<03:14,  6.69s/it]
Test accuracy: 0.79:   9%|▉         | 3/32 [00:20<03:14,  6.69s/it]
Test accuracy: 0.56:   9%|▉         | 3/32 [00:20<03:14,  6.69s/it]
Test accuracy: 0.58:   9%|▉         | 3/32 [00:20<03:14,  6.69s/it]
Test accuracy: 0.69:   9%|▉         | 3/32 [00:19<03:12,  6.65s/it]
Test accuracy: 0.71:   9%|▉         | 3/32 [00:19<03:12,  6.65s/it]
Test accuracy: 0.75:   9%|▉         | 3/32 [00:20<03:14,  6.70s/it]
Test accuracy: 0.79:   9%|▉         | 3/32 [00:20<03:14,  6.70s/it]
Test accuracy: 0.79:  12%|█▎        | 4/32 [00:26<03:03,  6.55s/it]
Test accuracy: 0.72:  12%|█▎        | 4/32 [00:26<03:03,  6.55s/it]
Test accuracy: 0.58:  12%|█▎        | 4/32 [00:26<03:03,  6.55s/it]
Test accuracy: 0.79:  12%|█▎        | 4/32 [00:26<03:03,  6.55s/it]
Test accuracy: 0.59:  12%|█▎        | 4/32 [00:26<03:03,  6.55s/it]
Test accuracy: 0.78:  12%|█▎        | 4/32 [00:26<03:03,  6.55s/it]
Test accuracy: 0.71:  12%|█▎        | 4/32 [00:26<03:02,  6.53s/it]
Test accuracy: 0.78:  12%|█▎        | 4/32 [00:26<03:02,  6.53s/it]
Test accuracy: 0.72:  16%|█▌        | 5/32 [00:32<02:54,  6.47s/it]
Test accuracy: 0.78:  16%|█▌        | 5/32 [00:32<02:54,  6.47s/it]
Test accuracy: 0.75:  16%|█▌        | 5/32 [00:32<02:54,  6.47s/it]
Test accuracy: 0.75:  16%|█▌        | 5/32 [00:32<02:54,  6.47s/it]
Test accuracy: 0.78:  16%|█▌        | 5/32 [00:32<02:54,  6.46s/it]
Test accuracy: 0.59:  16%|█▌        | 5/32 [00:32<02:54,  6.47s/it]
Test accuracy: 0.57:  16%|█▌        | 5/32 [00:32<02:54,  6.47s/it]
Test accuracy: 0.73:  16%|█▌        | 5/32 [00:32<02:54,  6.46s/it]
Test accuracy: 0.75:  19%|█▉        | 6/32 [00:38<02:45,  6.36s/it]
Test accuracy: 0.75:  19%|█▉        | 6/32 [00:38<02:45,  6.36s/it]
Test accuracy: 0.57:  19%|█▉        | 6/32 [00:38<02:45,  6.36s/it]
Test accuracy: 0.73:  19%|█▉        | 6/32 [00:38<02:45,  6.36s/it]
Test accuracy: 0.75:  19%|█▉        | 6/32 [00:38<02:45,  6.36s/it]
Test accuracy: 0.62:  19%|█▉        | 6/32 [00:38<02:45,  6.36s/it]
Test accuracy: 0.73:  19%|█▉        | 6/32 [00:38<02:45,  6.35s/it]
Test accuracy: 0.77:  19%|█▉        | 6/32 [00:38<02:45,  6.35s/it]
Test accuracy: 0.73:  22%|██▏       | 7/32 [00:45<02:38,  6.35s/it]
Test accuracy: 0.7:  22%|██▏       | 7/32 [00:45<02:38,  6.35s/it] 
Test accuracy: 0.62:  22%|██▏       | 7/32 [00:45<02:38,  6.35s/it]
Test accuracy: 0.77:  22%|██▏       | 7/32 [00:44<02:38,  6.35s/it]
Test accuracy: 0.59:  22%|██▏       | 7/32 [00:45<02:38,  6.35s/it]
Test accuracy: 0.75:  22%|██▏       | 7/32 [00:45<02:38,  6.35s/it]
Test accuracy: 0.77:  22%|██▏       | 7/32 [00:45<02:38,  6.35s/it]
Test accuracy: 0.73:  22%|██▏       | 7/32 [00:44<02:38,  6.35s/it]
Test accuracy: 0.7:  25%|██▌       | 8/32 [00:51<02:31,  6.32s/it]
Test accuracy: 0.77:  25%|██▌       | 8/32 [00:51<02:31,  6.32s/it]
Test accuracy: 0.59:  25%|██▌       | 8/32 [00:51<02:31,  6.32s/it]
Test accuracy: 0.72:  25%|██▌       | 8/32 [00:51<02:31,  6.32s/it]
Test accuracy: 0.73:  25%|██▌       | 8/32 [00:51<02:31,  6.32s/it]
Test accuracy: 0.61:  25%|██▌       | 8/32 [00:51<02:31,  6.32s/it]
Test accuracy: 0.73:  25%|██▌       | 8/32 [00:51<02:31,  6.32s/it]
Test accuracy: 0.72:  25%|██▌       | 8/32 [00:51<02:31,  6.32s/it]
Test accuracy: 0.72:  28%|██▊       | 9/32 [00:58<02:28,  6.46s/it]
Test accuracy: 0.71:  28%|██▊       | 9/32 [00:58<02:28,  6.46s/it]
Test accuracy: 0.73:  28%|██▊       | 9/32 [00:58<02:28,  6.46s/it]
Test accuracy: 0.75:  28%|██▊       | 9/32 [00:58<02:28,  6.46s/it]
Test accuracy: 0.72:  28%|██▊       | 9/32 [00:58<02:28,  6.46s/it]
Test accuracy: 0.71:  28%|██▊       | 9/32 [00:58<02:28,  6.46s/it]
Test accuracy: 0.61:  28%|██▊       | 9/32 [00:58<02:28,  6.46s/it]
Test accuracy: 0.64:  28%|██▊       | 9/32 [00:58<02:28,  6.46s/it]
Test accuracy: 0.71:  31%|███▏      | 10/32 [01:04<02:22,  6.49s/it]
Test accuracy: 0.64:  31%|███▏      | 10/32 [01:04<02:22,  6.49s/it]
Test accuracy: 0.75:  31%|███▏      | 10/32 [01:04<02:22,  6.49s/it]
Test accuracy: 0.71:  31%|███▏      | 10/32 [01:04<02:22,  6.49s/it]
Test accuracy: 0.6:  31%|███▏      | 10/32 [01:04<02:22,  6.49s/it] 
Test accuracy: 0.75:  31%|███▏      | 10/32 [01:04<02:22,  6.49s/it]
Test accuracy: 0.71:  31%|███▏      | 10/32 [01:04<02:22,  6.49s/it]
Test accuracy: 0.71:  31%|███▏      | 10/32 [01:04<02:22,  6.49s/it]
Test accuracy: 0.71:  34%|███▍      | 11/32 [01:10<02:14,  6.38s/it]
Test accuracy: 0.75:  34%|███▍      | 11/32 [01:10<02:13,  6.38s/it]
Test accuracy: 0.7:  34%|███▍      | 11/32 [01:10<02:14,  6.38s/it] 
Test accuracy: 0.6:  34%|███▍      | 11/32 [01:10<02:13,  6.38s/it]
Test accuracy: 0.75:  34%|███▍      | 11/32 [01:10<02:13,  6.38s/it]
Test accuracy: 0.57:  34%|███▍      | 11/32 [01:10<02:13,  6.38s/it]
Test accuracy: 0.71:  34%|███▍      | 11/32 [01:10<02:13,  6.38s/it]
Test accuracy: 0.68:  34%|███▍      | 11/32 [01:10<02:13,  6.38s/it]
Test accuracy: 0.7:  38%|███▊      | 12/32 [01:17<02:06,  6.31s/it]
Test accuracy: 0.68:  38%|███▊      | 12/32 [01:17<02:06,  6.31s/it]
Test accuracy: 0.57:  38%|███▊      | 12/32 [01:17<02:06,  6.31s/it]
Test accuracy: 0.75:  38%|███▊      | 12/32 [01:17<02:06,  6.31s/it]
Test accuracy: 0.55:  38%|███▊      | 12/32 [01:17<02:06,  6.31s/it]
Test accuracy: 0.72:  38%|███▊      | 12/32 [01:17<02:06,  6.31s/it]
Test accuracy: 0.68:  38%|███▊      | 12/32 [01:16<02:06,  6.31s/it]
Test accuracy: 0.67:  38%|███▊      | 12/32 [01:16<02:06,  6.31s/it]
Test accuracy: 0.68:  41%|████      | 13/32 [01:23<02:01,  6.38s/it]
Test accuracy: 0.72:  41%|████      | 13/32 [01:23<02:01,  6.38s/it]
Test accuracy: 0.67:  41%|████      | 13/32 [01:23<02:01,  6.38s/it]
Test accuracy: 0.55:  41%|████      | 13/32 [01:23<02:01,  6.38s/it]
Test accuracy: 0.72:  41%|████      | 13/32 [01:23<02:01,  6.38s/it]
Test accuracy: 0.56:  41%|████      | 13/32 [01:23<02:01,  6.38s/it]
Test accuracy: 0.67:  41%|████      | 13/32 [01:23<02:01,  6.38s/it]
Test accuracy: 0.64:  41%|████      | 13/32 [01:23<02:01,  6.38s/it]
Test accuracy: 0.72:  44%|████▍     | 14/32 [01:29<01:53,  6.31s/it]
Test accuracy: 0.67:  44%|████▍     | 14/32 [01:29<01:53,  6.31s/it]
Test accuracy: 0.72:  44%|████▍     | 14/32 [01:29<01:53,  6.31s/it]
Test accuracy: 0.69:  44%|████▍     | 14/32 [01:29<01:53,  6.31s/it]
Test accuracy: 0.56:  44%|████▍     | 14/32 [01:29<01:53,  6.31s/it]
Test accuracy: 0.64:  44%|████▍     | 14/32 [01:29<01:53,  6.31s/it]
Test accuracy: 0.56:  44%|████▍     | 14/32 [01:29<01:53,  6.31s/it]
Test accuracy: 0.64:  44%|████▍     | 14/32 [01:29<01:53,  6.31s/it]
Test accuracy: 0.69:  47%|████▋     | 15/32 [01:36<01:48,  6.35s/it]
Test accuracy: 0.68:  47%|████▋     | 15/32 [01:36<01:48,  6.35s/it]
Test accuracy: 0.72:  47%|████▋     | 15/32 [01:36<01:48,  6.35s/it]
Test accuracy: 0.56:  47%|████▋     | 15/32 [01:36<01:48,  6.35s/it]
Test accuracy: 0.7:  47%|████▋     | 15/32 [01:36<01:48,  6.35s/it] 
Test accuracy: 0.64:  47%|████▋     | 15/32 [01:35<01:48,  6.35s/it]
Test accuracy: 0.56:  47%|████▋     | 15/32 [01:36<01:48,  6.35s/it]
Test accuracy: 0.65:  47%|████▋     | 15/32 [01:35<01:48,  6.35s/it]
Test accuracy: 0.68:  50%|█████     | 16/32 [01:42<01:41,  6.35s/it]
Test accuracy: 0.68:  50%|█████     | 16/32 [01:42<01:41,  6.35s/it]
Test accuracy: 0.7:  50%|█████     | 16/32 [01:42<01:41,  6.35s/it]
Test accuracy: 0.56:  50%|█████     | 16/32 [01:42<01:41,  6.35s/it]
Test accuracy: 0.7:  50%|█████     | 16/32 [01:42<01:41,  6.35s/it]
Test accuracy: 0.65:  50%|█████     | 16/32 [01:42<01:41,  6.35s/it]
Test accuracy: 0.56:  50%|█████     | 16/32 [01:42<01:41,  6.35s/it]
Test accuracy: 0.64:  50%|█████     | 16/32 [01:42<01:41,  6.35s/it]
Test accuracy: 0.68:  53%|█████▎    | 17/32 [01:48<01:34,  6.29s/it]
Test accuracy: 0.68:  53%|█████▎    | 17/32 [01:48<01:34,  6.29s/it]
Test accuracy: 0.56:  53%|█████▎    | 17/32 [01:48<01:34,  6.29s/it]
Test accuracy: 0.57:  53%|█████▎    | 17/32 [01:48<01:34,  6.29s/it]
Test accuracy: 0.7:  53%|█████▎    | 17/32 [01:48<01:34,  6.29s/it]
Test accuracy: 0.64:  53%|█████▎    | 17/32 [01:48<01:34,  6.29s/it]
Test accuracy: 0.71:  53%|█████▎    | 17/32 [01:48<01:34,  6.29s/it]
Test accuracy: 0.65:  53%|█████▎    | 17/32 [01:48<01:34,  6.29s/it]
Test accuracy: 0.68:  56%|█████▋    | 18/32 [01:54<01:28,  6.31s/it]
Test accuracy: 0.69:  56%|█████▋    | 18/32 [01:54<01:28,  6.31s/it]
Test accuracy: 0.71:  56%|█████▋    | 18/32 [01:54<01:28,  6.31s/it]
Test accuracy: 0.71:  56%|█████▋    | 18/32 [01:54<01:28,  6.31s/it]
Test accuracy: 0.57:  56%|█████▋    | 18/32 [01:54<01:28,  6.31s/it]
Test accuracy: 0.65:  56%|█████▋    | 18/32 [01:54<01:28,  6.31s/it]
Test accuracy: 0.58:  56%|█████▋    | 18/32 [01:54<01:28,  6.31s/it]
Test accuracy: 0.67:  56%|█████▋    | 18/32 [01:54<01:28,  6.31s/it]
Test accuracy: 0.69:  59%|█████▉    | 19/32 [02:01<01:22,  6.32s/it]
Test accuracy: 0.67:  59%|█████▉    | 19/32 [02:01<01:22,  6.32s/it]
Test accuracy: 0.71:  59%|█████▉    | 19/32 [02:01<01:22,  6.32s/it]
Test accuracy: 0.7:  59%|█████▉    | 19/32 [02:01<01:22,  6.32s/it] 
Test accuracy: 0.58:  59%|█████▉    | 19/32 [02:01<01:22,  6.32s/it]
Test accuracy: 0.59:  59%|█████▉    | 19/32 [02:01<01:22,  6.32s/it]
Test accuracy: 0.67:  59%|█████▉    | 19/32 [02:01<01:22,  6.32s/it]
Test accuracy: 0.67:  59%|█████▉    | 19/32 [02:01<01:22,  6.32s/it]
Test accuracy: 0.67:  62%|██████▎   | 20/32 [02:07<01:15,  6.27s/it]
Test accuracy: 0.7:  62%|██████▎   | 20/32 [02:07<01:15,  6.27s/it]
Test accuracy: 0.66:  62%|██████▎   | 20/32 [02:07<01:15,  6.27s/it]
Test accuracy: 0.59:  62%|██████▎   | 20/32 [02:07<01:15,  6.27s/it]
Test accuracy: 0.7:  62%|██████▎   | 20/32 [02:07<01:15,  6.27s/it]
Test accuracy: 0.59:  62%|██████▎   | 20/32 [02:07<01:15,  6.27s/it]
Test accuracy: 0.67:  62%|██████▎   | 20/32 [02:07<01:15,  6.27s/it]
Test accuracy: 0.68:  62%|██████▎   | 20/32 [02:07<01:15,  6.27s/it]
Test accuracy: 0.66:  66%|██████▌   | 21/32 [02:14<01:11,  6.50s/it]
Test accuracy: 0.65:  66%|██████▌   | 21/32 [02:14<01:11,  6.50s/it]
Test accuracy: 0.59:  66%|██████▌   | 21/32 [02:14<01:11,  6.50s/it]
Test accuracy: 0.6:  66%|██████▌   | 21/32 [02:14<01:11,  6.50s/it] 
Test accuracy: 0.68:  66%|██████▌   | 21/32 [02:14<01:11,  6.50s/it]
Test accuracy: 0.66:  66%|██████▌   | 21/32 [02:14<01:11,  6.50s/it]
Test accuracy: 0.7:  66%|██████▌   | 21/32 [02:14<01:11,  6.51s/it]
Test accuracy: 0.7:  66%|██████▌   | 21/32 [02:14<01:11,  6.51s/it]
Test accuracy: 0.65:  69%|██████▉   | 22/32 [02:20<01:04,  6.46s/it]
Test accuracy: 0.66:  69%|██████▉   | 22/32 [02:20<01:04,  6.46s/it]
Test accuracy: 0.6:  69%|██████▉   | 22/32 [02:20<01:04,  6.46s/it]
Test accuracy: 0.7:  69%|██████▉   | 22/32 [02:20<01:04,  6.45s/it]
Test accuracy: 0.66:  69%|██████▉   | 22/32 [02:20<01:04,  6.46s/it]
Test accuracy: 0.59:  69%|██████▉   | 22/32 [02:20<01:04,  6.46s/it]
Test accuracy: 0.69:  69%|██████▉   | 22/32 [02:20<01:04,  6.45s/it]
Test accuracy: 0.65:  69%|██████▉   | 22/32 [02:20<01:04,  6.46s/it]
Test accuracy: 0.69:  72%|███████▏  | 23/32 [02:27<00:57,  6.36s/it]
Test accuracy: 0.66:  72%|███████▏  | 23/32 [02:27<00:57,  6.37s/it]
Test accuracy: 0.66:  72%|███████▏  | 23/32 [02:27<00:57,  6.37s/it]
Test accuracy: 0.69:  72%|███████▏  | 23/32 [02:27<00:57,  6.36s/it]
Test accuracy: 0.59:  72%|███████▏  | 23/32 [02:27<00:57,  6.37s/it]
Test accuracy: 0.6:  72%|███████▏  | 23/32 [02:27<00:57,  6.37s/it] 
Test accuracy: 0.65:  72%|███████▏  | 23/32 [02:26<00:57,  6.37s/it]
Test accuracy: 0.66:  72%|███████▏  | 23/32 [02:26<00:57,  6.37s/it]
Test accuracy: 0.69:  75%|███████▌  | 24/32 [02:34<00:52,  6.57s/it]
Test accuracy: 0.66:  75%|███████▌  | 24/32 [02:34<00:52,  6.57s/it]
Test accuracy: 0.66:  75%|███████▌  | 24/32 [02:34<00:52,  6.57s/it]
Test accuracy: 0.68:  75%|███████▌  | 24/32 [02:34<00:52,  6.57s/it]
Test accuracy: 0.6:  75%|███████▌  | 24/32 [02:34<00:52,  6.57s/it]
Test accuracy: 0.6:  75%|███████▌  | 24/32 [02:34<00:52,  6.57s/it]
Test accuracy: 0.66:  75%|███████▌  | 24/32 [02:33<00:52,  6.58s/it]
Test accuracy: 0.65:  75%|███████▌  | 24/32 [02:33<00:52,  6.58s/it]
Test accuracy: 0.66:  78%|███████▊  | 25/32 [02:40<00:45,  6.50s/it]
Test accuracy: 0.64:  78%|███████▊  | 25/32 [02:40<00:45,  6.50s/it]
Test accuracy: 0.6:  78%|███████▊  | 25/32 [02:40<00:45,  6.50s/it]
Test accuracy: 0.6:  78%|███████▊  | 25/32 [02:40<00:45,  6.50s/it]
Test accuracy: 0.65:  78%|███████▊  | 25/32 [02:40<00:45,  6.50s/it]
Test accuracy: 0.65:  78%|███████▊  | 25/32 [02:40<00:45,  6.50s/it]
Test accuracy: 0.68:  78%|███████▊  | 25/32 [02:40<00:45,  6.50s/it]
Test accuracy: 0.69:  78%|███████▊  | 25/32 [02:40<00:45,  6.50s/it]
Test accuracy: 0.64:  81%|████████▏ | 26/32 [02:46<00:38,  6.45s/it]
Test accuracy: 0.65:  81%|████████▏ | 26/32 [02:46<00:38,  6.45s/it]
Test accuracy: 0.69:  81%|████████▏ | 26/32 [02:46<00:38,  6.45s/it]
Test accuracy: 0.68:  81%|████████▏ | 26/32 [02:46<00:38,  6.45s/it]
Test accuracy: 0.65:  81%|████████▏ | 26/32 [02:46<00:38,  6.45s/it]
Test accuracy: 0.6:  81%|████████▏ | 26/32 [02:46<00:38,  6.45s/it]
Test accuracy: 0.59:  81%|████████▏ | 26/32 [02:46<00:38,  6.45s/it]
Test accuracy: 0.65:  81%|████████▏ | 26/32 [02:46<00:38,  6.45s/it]
Test accuracy: 0.65:  84%|████████▍ | 27/32 [02:52<00:31,  6.36s/it]
Test accuracy: 0.65:  84%|████████▍ | 27/32 [02:52<00:31,  6.36s/it]
Test accuracy: 0.59:  84%|████████▍ | 27/32 [02:52<00:31,  6.36s/it]
Test accuracy: 0.65:  84%|████████▍ | 27/32 [02:52<00:31,  6.36s/it]
Test accuracy: 0.59:  84%|████████▍ | 27/32 [02:52<00:31,  6.36s/it]
Test accuracy: 0.68:  84%|████████▍ | 27/32 [02:52<00:31,  6.36s/it]
Test accuracy: 0.68:  84%|████████▍ | 27/32 [02:52<00:31,  6.36s/it]
Test accuracy: 0.64:  84%|████████▍ | 27/32 [02:52<00:31,  6.36s/it]
Test accuracy: 0.65:  88%|████████▊ | 28/32 [02:59<00:26,  6.54s/it]
Test accuracy: 0.65:  88%|████████▊ | 28/32 [02:59<00:26,  6.54s/it]
Test accuracy: 0.59:  88%|████████▊ | 28/32 [02:59<00:26,  6.54s/it]
Test accuracy: 0.64:  88%|████████▊ | 28/32 [02:59<00:26,  6.54s/it]
Test accuracy: 0.6:  88%|████████▊ | 28/32 [02:59<00:26,  6.54s/it] 
Test accuracy: 0.65:  88%|████████▊ | 28/32 [02:59<00:26,  6.54s/it]
Test accuracy: 0.68:  88%|████████▊ | 28/32 [02:59<00:26,  6.54s/it]
Test accuracy: 0.68:  88%|████████▊ | 28/32 [02:59<00:26,  6.54s/it]
Test accuracy: 0.65:  91%|█████████ | 29/32 [03:06<00:19,  6.48s/it]
Test accuracy: 0.66:  91%|█████████ | 29/32 [03:06<00:19,  6.48s/it]
Test accuracy: 0.6:  91%|█████████ | 29/32 [03:06<00:19,  6.48s/it]
Test accuracy: 0.6:  91%|█████████ | 29/32 [03:06<00:19,  6.48s/it]
Test accuracy: 0.65:  91%|█████████ | 29/32 [03:06<00:19,  6.48s/it]
Test accuracy: 0.66:  91%|█████████ | 29/32 [03:06<00:19,  6.48s/it]
Test accuracy: 0.68:  91%|█████████ | 29/32 [03:06<00:19,  6.48s/it]
Test accuracy: 0.68:  91%|█████████ | 29/32 [03:06<00:19,  6.48s/it]
Test accuracy: 0.66:  94%|█████████▍| 30/32 [03:12<00:12,  6.44s/it]
Test accuracy: 0.65:  94%|█████████▍| 30/32 [03:12<00:12,  6.44s/it]
Test accuracy: 0.66:  94%|█████████▍| 30/32 [03:12<00:12,  6.44s/it]
Test accuracy: 0.6:  94%|█████████▍| 30/32 [03:12<00:12,  6.44s/it]
Test accuracy: 0.61:  94%|█████████▍| 30/32 [03:12<00:12,  6.44s/it]
Test accuracy: 0.65:  94%|█████████▍| 30/32 [03:12<00:12,  6.44s/it]
Test accuracy: 0.68:  94%|█████████▍| 30/32 [03:12<00:12,  6.44s/it]
Test accuracy: 0.67:  94%|█████████▍| 30/32 [03:12<00:12,  6.44s/it]
Test accuracy: 0.65:  97%|█████████▋| 31/32 [03:19<00:06,  6.56s/it]
Test accuracy: 0.65:  97%|█████████▋| 31/32 [03:19<00:06,  6.56s/it]
Test accuracy: 0.61:  97%|█████████▋| 31/32 [03:19<00:06,  6.56s/it]
Test accuracy: 0.61:  97%|█████████▋| 31/32 [03:19<00:06,  6.56s/it]
Test accuracy: 0.65:  97%|█████████▋| 31/32 [03:19<00:06,  6.56s/it]
Test accuracy: 0.66:  97%|█████████▋| 31/32 [03:19<00:06,  6.56s/it]
Test accuracy: 0.67:  97%|█████████▋| 31/32 [03:19<00:06,  6.56s/it]
Test accuracy: 0.67:  97%|█████████▋| 31/32 [03:19<00:06,  6.56s/it]
Test accuracy: 0.66: 100%|██████████| 32/32 [03:21<00:00,  5.27s/it]
Test accuracy: 0.66: 100%|██████████| 32/32 [03:21<00:00,  5.27s/it]
Test accuracy: 0.67: 100%|██████████| 32/32 [03:21<00:00,  5.27s/it]
Test accuracy: 0.65: 100%|██████████| 32/32 [03:21<00:00,  5.27s/it]
Test accuracy: 0.67: 100%|██████████| 32/32 [03:21<00:00,  5.27s/it]
Test accuracy: 0.61: 100%|██████████| 32/32 [03:21<00:00,  5.27s/it]
Test accuracy: 0.65: 100%|██████████| 32/32 [03:21<00:00,  5.27s/it]
Test accuracy: 0.61: 100%|██████████| 32/32 [03:21<00:00,  5.27s/it]
Test accuracy: 0.67: 100%|██████████| 32/32 [03:21<00:00,  6.30s/it]
Device 2: Cor=168, CoT=0, Total=250

Test accuracy: 0.61: 100%|██████████| 32/32 [03:21<00:00,  6.30s/it]
Device 3: Cor=153, CoT=0, Total=250

Test accuracy: 0.65: 100%|██████████| 32/32 [03:21<00:00,  6.30s/it]
Device 1: Cor=163, CoT=0, Total=250

Test accuracy: 0.66: 100%|██████████| 32/32 [03:21<00:00,  6.31s/it]
Device 0: Cor=165, CoT=0, Total=250

============================================================
EVALUATION RESULTS (Epoch 8)
============================================================
Config: data_curriculum=ON, thought_curriculum=OFF
thought_stage=10, data_stage=4, n_latent_tokens=10
Training on difficulties: 0-4

Overall accuracy: 649 / 1000 = 0.6490

Per-difficulty accuracy:
  diff=1: 93/100 = 0.9300 (trained)
  diff=2: 95/100 = 0.9500 (trained)
  diff=3: 92/100 = 0.9200 (trained)
  diff=4: 79/100 = 0.7900 (trained)
  diff=5: 42/100 = 0.4200 (untrained)
  diff=6: 44/100 = 0.4400 (untrained)
  diff=7: 46/100 = 0.4600 (untrained)
  diff=8: 48/100 = 0.4800 (untrained)
  diff=9: 59/100 = 0.5900 (untrained)
  diff=10: 51/100 = 0.5100 (untrained)

Trained (diff 0-4): 359/400 = 0.8975
============================================================

  Staging check: acc_trained=0.8975, acc_diff4=0.7900 (threshold=0.9)
Warning: failed to upload checkpoint_9 to HF: (Request ID: Root=1-6a1b7827-7a887c4f7328dd9350ba588e;145fac0f-7f72-49bf-93c4-8e3d582788b4)

403 Forbidden: You need to setup automatic credit recharge in order to upload more data. You can do so at /settings/billing..
Cannot access content at: https://huggingface.co/seyedparsa/thoughtformer-checkpoints.git/info/lfs/objects/batch.
Make sure your token has the correct permissions.

============================================================
EPOCH 9/100 (thought_stage=10, data_stage=4)
============================================================
  thought_stage=10, c_thought=1, max_difficulty=4

Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):  16%|█▌        | 156/1000 [00:00<00:00, 988.25 examples/s]
Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):  16%|█▌        | 156/1000 [00:00<00:00, 945.22 examples/s]
Map (num_proc=4):  16%|█▌        | 156/1000 [00:00<00:00, 935.87 examples/s]
Map (num_proc=4):  81%|████████  | 811/1000 [00:00<00:00, 3416.09 examples/s]
Map (num_proc=4):  81%|████████  | 811/1000 [00:00<00:00, 3368.15 examples/s]
Map (num_proc=4):  81%|████████  | 811/1000 [00:00<00:00, 3412.03 examples/s]
Map (num_proc=4):  16%|█▌        | 156/1000 [00:00<00:01, 743.73 examples/s]
Map (num_proc=4):  82%|████████▏ | 817/1000 [00:00<00:00, 2952.71 examples/s]
Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2041.21 examples/s]

Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2019.71 examples/s]

Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2046.76 examples/s]

Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 1663.53 examples/s]

Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:04, 1912.33 examples/s]
Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:04, 1984.04 examples/s]
Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:04, 1887.99 examples/s]
Filter (num_proc=4):  20%|██        | 2000/10000 [00:00<00:02, 3497.92 examples/s]
Filter (num_proc=4):  20%|██        | 2000/10000 [00:00<00:02, 3713.31 examples/s]
Filter (num_proc=4):  20%|██        | 2000/10000 [00:00<00:02, 3576.86 examples/s]
Filter (num_proc=4):  30%|███       | 3000/10000 [00:00<00:01, 4788.13 examples/s]
Filter (num_proc=4):  30%|███       | 3000/10000 [00:00<00:01, 5090.18 examples/s]
Filter (num_proc=4):  30%|███       | 3000/10000 [00:00<00:01, 4974.14 examples/s]
Filter (num_proc=4):  50%|█████     | 5000/10000 [00:00<00:00, 8145.56 examples/s]
Filter (num_proc=4):  50%|█████     | 5000/10000 [00:00<00:00, 8520.30 examples/s]
Filter (num_proc=4):  60%|██████    | 6000/10000 [00:00<00:00, 10313.44 examples/s]
Filter (num_proc=4):  75%|███████▌  | 7500/10000 [00:01<00:00, 11494.07 examples/s]
Filter (num_proc=4):  75%|███████▌  | 7500/10000 [00:00<00:00, 11159.57 examples/s]
Filter (num_proc=4):  75%|███████▌  | 7500/10000 [00:00<00:00, 11462.44 examples/s]
Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:05, 1703.65 examples/s]
Filter (num_proc=4):  90%|█████████ | 9000/10000 [00:01<00:00, 11928.91 examples/s]
Filter (num_proc=4):  90%|█████████ | 9000/10000 [00:01<00:00, 12192.23 examples/s]
Filter (num_proc=4):  95%|█████████▌| 9500/10000 [00:01<00:00, 12723.07 examples/s]
Filter (num_proc=4):  20%|██        | 2000/10000 [00:00<00:03, 2664.53 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 7580.67 examples/s]

Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 7271.44 examples/s]

Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 7333.97 examples/s]

Filter (num_proc=4):  35%|███▌      | 3500/10000 [00:00<00:01, 4659.17 examples/s]
Filter (num_proc=4):  55%|█████▌    | 5500/10000 [00:01<00:00, 7585.99 examples/s]
Filter (num_proc=4):  70%|███████   | 7000/10000 [00:01<00:00, 8936.73 examples/s]
Filter (num_proc=4):  85%|████████▌ | 8500/10000 [00:01<00:00, 9078.10 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 9511.00 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 6044.13 examples/s]

Map (num_proc=4):   0%|          | 0/4000 [00:00<?, ? examples/s]
Map (num_proc=4):   3%|▎         | 131/4000 [00:00<00:24, 158.83 examples/s]
Map (num_proc=4):  13%|█▎        | 538/4000 [00:00<00:05, 684.30 examples/s]
Map (num_proc=4):  20%|██        | 816/4000 [00:01<00:03, 1020.85 examples/s]
Map (num_proc=4):  27%|██▋       | 1084/4000 [00:01<00:02, 1039.63 examples/s]
Map (num_proc=4):  35%|███▍      | 1399/4000 [00:01<00:01, 1391.04 examples/s]
Map (num_proc=4):  42%|████▏     | 1677/4000 [00:01<00:01, 1539.82 examples/s]
Map (num_proc=4):  49%|████▉     | 1965/4000 [00:01<00:01, 1546.15 examples/s]
Map (num_proc=4):  56%|█████▌    | 2230/4000 [00:01<00:01, 1677.04 examples/s]
Map (num_proc=4):  63%|██████▎   | 2538/4000 [00:01<00:00, 1968.87 examples/s]
Map (num_proc=4):  70%|███████   | 2816/4000 [00:02<00:00, 1998.65 examples/s]
Map (num_proc=4):  77%|███████▋  | 3086/4000 [00:02<00:00, 1661.44 examples/s]
Map (num_proc=4):  85%|████████▌ | 3401/4000 [00:02<00:00, 1811.87 examples/s]
Map (num_proc=4):  95%|█████████▌| 3816/4000 [00:02<00:00, 2133.54 examples/s]
Map (num_proc=4): 100%|██████████| 4000/4000 [00:03<00:00, 1176.30 examples/s]

Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 1414.27 examples/s]
Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 1372.12 examples/s]
Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 1350.18 examples/s]
Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 1049.87 examples/s]
Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2756.45 examples/s]

Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2638.60 examples/s]

Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2548.26 examples/s]

Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 1980.50 examples/s]

Map (num_proc=4):   0%|          | 0/400 [00:00<?, ? examples/s]
Map (num_proc=4):  25%|██▌       | 100/400 [00:00<00:00, 341.39 examples/s]
Map (num_proc=4):  75%|███████▌  | 300/400 [00:00<00:00, 799.40 examples/s]
Map (num_proc=4): 100%|██████████| 400/400 [00:00<00:00, 556.72 examples/s]

Training Epoch: 10:   0%|          | 0/125 [00:00<?, ?it/s]
Training Epoch: 10:   0%|          | 0/125 [00:00<?, ?it/s]
Training Epoch: 10:   0%|          | 0/125 [00:00<?, ?it/s]
Training Epoch: 10:   0%|          | 0/125 [00:00<?, ?it/s]logging training data
rank 0 | gpu 0 | alloc=4.20 GB | reserved=47.74 GB | max=41.99 GB
rank 2 | gpu 2 | alloc=4.20 GB | reserved=47.69 GB | max=41.98 GB
rank 3 | gpu 3 | alloc=4.20 GB | reserved=47.73 GB | max=41.99 GB
rank 1 | gpu 1 | alloc=4.20 GB | reserved=49.58 GB | max=41.98 GB

Training Epoch: 10/100, batch 0/250 completed (loss: 0.0403:   0%|          | 0/125 [00:03<?, ?it/s]
Training Epoch: 10/100, batch 0/250 completed (loss: 0.215:   0%|          | 0/125 [00:03<?, ?it/s]
Training Epoch: 10/100, batch 0/250 completed (loss: 0.0249:   0%|          | 0/125 [00:03<?, ?it/s]
Training Epoch: 10/100, batch 0/250 completed (loss: 0.228:   0%|          | 0/125 [00:03<?, ?it/s]
Training Epoch: 10/100, batch 0/250 completed (loss: 0.228:   1%|          | 1/125 [00:06<12:42,  6.15s/it]
Training Epoch: 10/100, batch 0/250 completed (loss: 0.215:   1%|          | 1/125 [00:06<12:40,  6.13s/it]
Training Epoch: 10/100, batch 0/250 completed (loss: 0.0249:   1%|          | 1/125 [00:06<12:40,  6.13s/it]
Training Epoch: 10/100, batch 0/250 completed (loss: 0.0403:   1%|          | 1/125 [00:06<12:40,  6.14s/it]
Training Epoch: 10/100, batch 1/250 completed (loss: 0.1627:   1%|          | 1/125 [00:06<12:40,  6.13s/it]
Training Epoch: 10/100, batch 1/250 completed (loss: 0.0467:   1%|          | 1/125 [00:06<12:40,  6.13s/it]
Training Epoch: 10/100, batch 1/250 completed (loss: 0.1328:   1%|          | 1/125 [00:06<12:40,  6.14s/it]
Training Epoch: 10/100, batch 1/250 completed (loss: 0.1102:   1%|          | 1/125 [00:06<12:42,  6.15s/it]
Training Epoch: 10/100, batch 2/250 completed (loss: 0.0099:   1%|          | 1/125 [00:09<12:40,  6.13s/it]
Training Epoch: 10/100, batch 2/250 completed (loss: 0.0599:   1%|          | 1/125 [00:09<12:40,  6.13s/it]
Training Epoch: 10/100, batch 2/250 completed (loss: 0.0928:   1%|          | 1/125 [00:09<12:40,  6.14s/it]
Training Epoch: 10/100, batch 2/250 completed (loss: 0.0335:   1%|          | 1/125 [00:09<12:42,  6.15s/it]
Training Epoch: 10/100, batch 2/250 completed (loss: 0.0099:   2%|▏         | 2/125 [00:11<12:08,  5.92s/it]
Training Epoch: 10/100, batch 2/250 completed (loss: 0.0335:   2%|▏         | 2/125 [00:11<12:09,  5.93s/it]
Training Epoch: 10/100, batch 2/250 completed (loss: 0.0928:   2%|▏         | 2/125 [00:11<12:08,  5.93s/it]
Training Epoch: 10/100, batch 2/250 completed (loss: 0.0599:   2%|▏         | 2/125 [00:11<12:09,  5.93s/it]
Training Epoch: 10/100, batch 3/250 completed (loss: 0.174:   2%|▏         | 2/125 [00:11<12:08,  5.92s/it] 
Training Epoch: 10/100, batch 3/250 completed (loss: 0.1518:   2%|▏         | 2/125 [00:11<12:08,  5.93s/it]
Training Epoch: 10/100, batch 3/250 completed (loss: 0.1009:   2%|▏         | 2/125 [00:11<12:09,  5.93s/it]
Training Epoch: 10/100, batch 3/250 completed (loss: 0.1213:   2%|▏         | 2/125 [00:11<12:09,  5.93s/it]
Training Epoch: 10/100, batch 4/250 completed (loss: 0.2536:   2%|▏         | 2/125 [00:14<12:09,  5.93s/it]
Training Epoch: 10/100, batch 4/250 completed (loss: 0.084:   2%|▏         | 2/125 [00:14<12:08,  5.92s/it]
Training Epoch: 10/100, batch 4/250 completed (loss: 0.3146:   2%|▏         | 2/125 [00:14<12:08,  5.93s/it]
Training Epoch: 10/100, batch 4/250 completed (loss: 0.1946:   2%|▏         | 2/125 [00:14<12:09,  5.93s/it]
Training Epoch: 10/100, batch 4/250 completed (loss: 0.1946:   2%|▏         | 3/125 [00:17<12:05,  5.94s/it]
Training Epoch: 10/100, batch 4/250 completed (loss: 0.3146:   2%|▏         | 3/125 [00:17<12:04,  5.94s/it]
Training Epoch: 10/100, batch 4/250 completed (loss: 0.084:   2%|▏         | 3/125 [00:17<12:05,  5.94s/it]
Training Epoch: 10/100, batch 4/250 completed (loss: 0.2536:   2%|▏         | 3/125 [00:17<12:04,  5.94s/it]
Training Epoch: 10/100, batch 5/250 completed (loss: 0.0929:   2%|▏         | 3/125 [00:17<12:04,  5.94s/it]
Training Epoch: 10/100, batch 5/250 completed (loss: 0.0451:   2%|▏         | 3/125 [00:17<12:04,  5.94s/it]
Training Epoch: 10/100, batch 5/250 completed (loss: 0.103:   2%|▏         | 3/125 [00:17<12:05,  5.94s/it]
Training Epoch: 10/100, batch 5/250 completed (loss: 0.0717:   2%|▏         | 3/125 [00:17<12:05,  5.94s/it]
Training Epoch: 10/100, batch 6/250 completed (loss: 0.1362:   2%|▏         | 3/125 [00:20<12:04,  5.94s/it]
Training Epoch: 10/100, batch 6/250 completed (loss: 0.0432:   2%|▏         | 3/125 [00:20<12:04,  5.94s/it]
Training Epoch: 10/100, batch 6/250 completed (loss: 0.0331:   2%|▏         | 3/125 [00:20<12:05,  5.94s/it]
Training Epoch: 10/100, batch 6/250 completed (loss: 0.0365:   2%|▏         | 3/125 [00:20<12:05,  5.94s/it]
Training Epoch: 10/100, batch 6/250 completed (loss: 0.0365:   3%|▎         | 4/125 [00:23<11:44,  5.82s/it]
Training Epoch: 10/100, batch 6/250 completed (loss: 0.1362:   3%|▎         | 4/125 [00:23<11:43,  5.81s/it]
Training Epoch: 10/100, batch 6/250 completed (loss: 0.0432:   3%|▎         | 4/125 [00:23<11:43,  5.81s/it]
Training Epoch: 10/100, batch 6/250 completed (loss: 0.0331:   3%|▎         | 4/125 [00:23<11:43,  5.82s/it]
Training Epoch: 10/100, batch 7/250 completed (loss: 0.0725:   3%|▎         | 4/125 [00:23<11:43,  5.81s/it]
Training Epoch: 10/100, batch 7/250 completed (loss: 0.1308:   3%|▎         | 4/125 [00:23<11:43,  5.81s/it]
Training Epoch: 10/100, batch 7/250 completed (loss: 0.0609:   3%|▎         | 4/125 [00:23<11:43,  5.82s/it]
Training Epoch: 10/100, batch 7/250 completed (loss: 0.0737:   3%|▎         | 4/125 [00:23<11:44,  5.82s/it]
Training Epoch: 10/100, batch 8/250 completed (loss: 0.0784:   3%|▎         | 4/125 [00:26<11:43,  5.81s/it]
Training Epoch: 10/100, batch 8/250 completed (loss: 0.0923:   3%|▎         | 4/125 [00:26<11:43,  5.82s/it]
Training Epoch: 10/100, batch 8/250 completed (loss: 0.1467:   3%|▎         | 4/125 [00:26<11:43,  5.81s/it]
Training Epoch: 10/100, batch 8/250 completed (loss: 0.046:   3%|▎         | 4/125 [00:26<11:44,  5.82s/it] 
Training Epoch: 10/100, batch 8/250 completed (loss: 0.046:   4%|▍         | 5/125 [00:29<11:38,  5.82s/it]
Training Epoch: 10/100, batch 8/250 completed (loss: 0.0923:   4%|▍         | 5/125 [00:29<11:38,  5.82s/it]
Training Epoch: 10/100, batch 8/250 completed (loss: 0.1467:   4%|▍         | 5/125 [00:29<11:38,  5.82s/it]
Training Epoch: 10/100, batch 8/250 completed (loss: 0.0784:   4%|▍         | 5/125 [00:29<11:38,  5.82s/it]
Training Epoch: 10/100, batch 9/250 completed (loss: 0.0791:   4%|▍         | 5/125 [00:29<11:38,  5.82s/it]
Training Epoch: 10/100, batch 9/250 completed (loss: 0.1455:   4%|▍         | 5/125 [00:29<11:38,  5.82s/it]
Training Epoch: 10/100, batch 9/250 completed (loss: 0.0265:   4%|▍         | 5/125 [00:29<11:38,  5.82s/it]
Training Epoch: 10/100, batch 9/250 completed (loss: 0.0576:   4%|▍         | 5/125 [00:29<11:38,  5.82s/it]
Training Epoch: 10/100, batch 10/250 completed (loss: 0.1078:   4%|▍         | 5/125 [00:32<11:38,  5.82s/it]
Training Epoch: 10/100, batch 10/250 completed (loss: 0.037:   4%|▍         | 5/125 [00:32<11:38,  5.82s/it]
Training Epoch: 10/100, batch 10/250 completed (loss: 0.1341:   4%|▍         | 5/125 [00:32<11:38,  5.82s/it]
Training Epoch: 10/100, batch 10/250 completed (loss: 0.0409:   4%|▍         | 5/125 [00:32<11:38,  5.82s/it]
Training Epoch: 10/100, batch 10/250 completed (loss: 0.1341:   5%|▍         | 6/125 [00:35<11:39,  5.88s/it]
Training Epoch: 10/100, batch 10/250 completed (loss: 0.1078:   5%|▍         | 6/125 [00:35<11:39,  5.87s/it]
Training Epoch: 10/100, batch 10/250 completed (loss: 0.037:   5%|▍         | 6/125 [00:35<11:38,  5.87s/it]
Training Epoch: 10/100, batch 10/250 completed (loss: 0.0409:   5%|▍         | 6/125 [00:35<11:39,  5.88s/it]
Training Epoch: 10/100, batch 11/250 completed (loss: 0.0669:   5%|▍         | 6/125 [00:35<11:38,  5.87s/it]
Training Epoch: 10/100, batch 11/250 completed (loss: 0.0705:   5%|▍         | 6/125 [00:35<11:39,  5.87s/it]
Training Epoch: 10/100, batch 11/250 completed (loss: 0.1295:   5%|▍         | 6/125 [00:35<11:39,  5.88s/it]
Training Epoch: 10/100, batch 11/250 completed (loss: 0.2106:   5%|▍         | 6/125 [00:35<11:39,  5.88s/it]
Training Epoch: 10/100, batch 12/250 completed (loss: 0.1009:   5%|▍         | 6/125 [00:38<11:39,  5.87s/it]
Training Epoch: 10/100, batch 12/250 completed (loss: 0.0342:   5%|▍         | 6/125 [00:38<11:38,  5.87s/it]
Training Epoch: 10/100, batch 12/250 completed (loss: 0.0316:   5%|▍         | 6/125 [00:38<11:39,  5.88s/it]
Training Epoch: 10/100, batch 12/250 completed (loss: 0.0414:   5%|▍         | 6/125 [00:38<11:39,  5.88s/it]
Training Epoch: 10/100, batch 12/250 completed (loss: 0.0316:   6%|▌         | 7/125 [00:41<11:30,  5.86s/it]
Training Epoch: 10/100, batch 12/250 completed (loss: 0.1009:   6%|▌         | 7/125 [00:41<11:31,  5.86s/it]
Training Epoch: 10/100, batch 12/250 completed (loss: 0.0342:   6%|▌         | 7/125 [00:41<11:30,  5.86s/it]
Training Epoch: 10/100, batch 12/250 completed (loss: 0.0414:   6%|▌         | 7/125 [00:41<11:31,  5.86s/it]
Training Epoch: 10/100, batch 13/250 completed (loss: 0.022:   6%|▌         | 7/125 [00:41<11:30,  5.86s/it] 
Training Epoch: 10/100, batch 13/250 completed (loss: 0.0171:   6%|▌         | 7/125 [00:41<11:30,  5.86s/it]
Training Epoch: 10/100, batch 13/250 completed (loss: 0.027:   6%|▌         | 7/125 [00:41<11:31,  5.86s/it] 
Training Epoch: 10/100, batch 13/250 completed (loss: 0.0682:   6%|▌         | 7/125 [00:41<11:31,  5.86s/it]
Training Epoch: 10/100, batch 14/250 completed (loss: 0.0258:   6%|▌         | 7/125 [00:44<11:31,  5.86s/it]
Training Epoch: 10/100, batch 14/250 completed (loss: 0.0448:   6%|▌         | 7/125 [00:44<11:30,  5.86s/it]
Training Epoch: 10/100, batch 14/250 completed (loss: 0.1366:   6%|▌         | 7/125 [00:44<11:30,  5.86s/it]
Training Epoch: 10/100, batch 14/250 completed (loss: 0.0611:   6%|▌         | 7/125 [00:44<11:31,  5.86s/it]
Training Epoch: 10/100, batch 14/250 completed (loss: 0.0611:   6%|▋         | 8/125 [00:46<11:24,  5.85s/it]
Training Epoch: 10/100, batch 14/250 completed (loss: 0.1366:   6%|▋         | 8/125 [00:46<11:24,  5.85s/it]
Training Epoch: 10/100, batch 14/250 completed (loss: 0.0448:   6%|▋         | 8/125 [00:46<11:24,  5.85s/it]
Training Epoch: 10/100, batch 14/250 completed (loss: 0.0258:   6%|▋         | 8/125 [00:46<11:24,  5.85s/it]
Training Epoch: 10/100, batch 15/250 completed (loss: 0.0492:   6%|▋         | 8/125 [00:47<11:24,  5.85s/it]
Training Epoch: 10/100, batch 15/250 completed (loss: 0.0735:   6%|▋         | 8/125 [00:47<11:24,  5.85s/it]
Training Epoch: 10/100, batch 15/250 completed (loss: 0.0764:   6%|▋         | 8/125 [00:47<11:24,  5.85s/it]
Training Epoch: 10/100, batch 15/250 completed (loss: 0.1052:   6%|▋         | 8/125 [00:47<11:24,  5.85s/it]
Training Epoch: 10/100, batch 16/250 completed (loss: 0.0826:   6%|▋         | 8/125 [00:50<11:24,  5.85s/it]
Training Epoch: 10/100, batch 16/250 completed (loss: 0.1096:   6%|▋         | 8/125 [00:50<11:24,  5.85s/it]
Training Epoch: 10/100, batch 16/250 completed (loss: 0.0896:   6%|▋         | 8/125 [00:50<11:24,  5.85s/it]
Training Epoch: 10/100, batch 16/250 completed (loss: 0.0895:   6%|▋         | 8/125 [00:50<11:24,  5.85s/it]
Training Epoch: 10/100, batch 16/250 completed (loss: 0.0896:   7%|▋         | 9/125 [00:52<11:23,  5.89s/it]
Training Epoch: 10/100, batch 16/250 completed (loss: 0.0895:   7%|▋         | 9/125 [00:52<11:23,  5.89s/it]
Training Epoch: 10/100, batch 16/250 completed (loss: 0.1096:   7%|▋         | 9/125 [00:52<11:23,  5.89s/it]
Training Epoch: 10/100, batch 16/250 completed (loss: 0.0826:   7%|▋         | 9/125 [00:52<11:23,  5.89s/it]
Training Epoch: 10/100, batch 17/250 completed (loss: 0.0451:   7%|▋         | 9/125 [00:52<11:23,  5.89s/it]
Training Epoch: 10/100, batch 17/250 completed (loss: 0.04:   7%|▋         | 9/125 [00:52<11:23,  5.89s/it]  
Training Epoch: 10/100, batch 17/250 completed (loss: 0.0398:   7%|▋         | 9/125 [00:52<11:23,  5.89s/it]
Training Epoch: 10/100, batch 17/250 completed (loss: 0.0719:   7%|▋         | 9/125 [00:53<11:23,  5.89s/it]
Training Epoch: 10/100, batch 18/250 completed (loss: 0.0363:   7%|▋         | 9/125 [00:55<11:23,  5.89s/it]
Training Epoch: 10/100, batch 18/250 completed (loss: 0.1095:   7%|▋         | 9/125 [00:55<11:23,  5.89s/it]
Training Epoch: 10/100, batch 18/250 completed (loss: 0.0531:   7%|▋         | 9/125 [00:55<11:23,  5.89s/it]
Training Epoch: 10/100, batch 18/250 completed (loss: 0.1597:   7%|▋         | 9/125 [00:55<11:23,  5.89s/it]
Training Epoch: 10/100, batch 18/250 completed (loss: 0.1095:   8%|▊         | 10/125 [00:58<11:13,  5.86s/it]
Training Epoch: 10/100, batch 18/250 completed (loss: 0.1597:   8%|▊         | 10/125 [00:58<11:13,  5.86s/it]
Training Epoch: 10/100, batch 18/250 completed (loss: 0.0531:   8%|▊         | 10/125 [00:58<11:13,  5.86s/it]
Training Epoch: 10/100, batch 18/250 completed (loss: 0.0363:   8%|▊         | 10/125 [00:58<11:13,  5.86s/it]
Training Epoch: 10/100, batch 19/250 completed (loss: 0.0307:   8%|▊         | 10/125 [00:58<11:13,  5.86s/it]
Training Epoch: 10/100, batch 19/250 completed (loss: 0.0267:   8%|▊         | 10/125 [00:58<11:13,  5.86s/it]
Training Epoch: 10/100, batch 19/250 completed (loss: 0.0258:   8%|▊         | 10/125 [00:58<11:13,  5.86s/it]
Training Epoch: 10/100, batch 19/250 completed (loss: 0.0193:   8%|▊         | 10/125 [00:58<11:13,  5.86s/it]
Training Epoch: 10/100, batch 20/250 completed (loss: 0.2582:   8%|▊         | 10/125 [01:01<11:13,  5.86s/it]
Training Epoch: 10/100, batch 20/250 completed (loss: 0.0243:   8%|▊         | 10/125 [01:01<11:13,  5.86s/it]
Training Epoch: 10/100, batch 20/250 completed (loss: 0.0482:   8%|▊         | 10/125 [01:01<11:13,  5.86s/it]
Training Epoch: 10/100, batch 20/250 completed (loss: 0.0605:   8%|▊         | 10/125 [01:01<11:13,  5.86s/it]
Training Epoch: 10/100, batch 20/250 completed (loss: 0.2582:   9%|▉         | 11/125 [01:04<11:15,  5.93s/it]
Training Epoch: 10/100, batch 20/250 completed (loss: 0.0605:   9%|▉         | 11/125 [01:04<11:16,  5.93s/it]
Training Epoch: 10/100, batch 20/250 completed (loss: 0.0243:   9%|▉         | 11/125 [01:04<11:16,  5.93s/it]
Training Epoch: 10/100, batch 20/250 completed (loss: 0.0482:   9%|▉         | 11/125 [01:04<11:16,  5.93s/it]
Training Epoch: 10/100, batch 21/250 completed (loss: 0.1397:   9%|▉         | 11/125 [01:04<11:15,  5.93s/it]
Training Epoch: 10/100, batch 21/250 completed (loss: 0.0416:   9%|▉         | 11/125 [01:04<11:16,  5.93s/it]
Training Epoch: 10/100, batch 21/250 completed (loss: 0.0741:   9%|▉         | 11/125 [01:04<11:16,  5.93s/it]
Training Epoch: 10/100, batch 21/250 completed (loss: 0.0806:   9%|▉         | 11/125 [01:04<11:16,  5.93s/it]
Training Epoch: 10/100, batch 22/250 completed (loss: 0.0418:   9%|▉         | 11/125 [01:07<11:15,  5.93s/it]
Training Epoch: 10/100, batch 22/250 completed (loss: 0.0294:   9%|▉         | 11/125 [01:07<11:16,  5.93s/it]
Training Epoch: 10/100, batch 22/250 completed (loss: 0.0216:   9%|▉         | 11/125 [01:07<11:16,  5.93s/it]
Training Epoch: 10/100, batch 22/250 completed (loss: 0.0421:   9%|▉         | 11/125 [01:07<11:16,  5.93s/it]
Training Epoch: 10/100, batch 22/250 completed (loss: 0.0421:  10%|▉         | 12/125 [01:10<11:02,  5.86s/it]
Training Epoch: 10/100, batch 22/250 completed (loss: 0.0294:  10%|▉         | 12/125 [01:10<11:02,  5.86s/it]
Training Epoch: 10/100, batch 22/250 completed (loss: 0.0418:  10%|▉         | 12/125 [01:10<11:02,  5.86s/it]
Training Epoch: 10/100, batch 22/250 completed (loss: 0.0216:  10%|▉         | 12/125 [01:10<11:02,  5.86s/it]
Training Epoch: 10/100, batch 23/250 completed (loss: 0.0342:  10%|▉         | 12/125 [01:10<11:02,  5.86s/it]
Training Epoch: 10/100, batch 23/250 completed (loss: 0.0527:  10%|▉         | 12/125 [01:10<11:02,  5.86s/it]
Training Epoch: 10/100, batch 23/250 completed (loss: 0.0888:  10%|▉         | 12/125 [01:10<11:02,  5.86s/it]
Training Epoch: 10/100, batch 23/250 completed (loss: 0.0469:  10%|▉         | 12/125 [01:10<11:02,  5.86s/it]
Training Epoch: 10/100, batch 24/250 completed (loss: 0.0603:  10%|▉         | 12/125 [01:13<11:02,  5.86s/it]
Training Epoch: 10/100, batch 24/250 completed (loss: 0.1728:  10%|▉         | 12/125 [01:13<11:02,  5.86s/it]
Training Epoch: 10/100, batch 24/250 completed (loss: 0.0416:  10%|▉         | 12/125 [01:13<11:02,  5.86s/it]
Training Epoch: 10/100, batch 24/250 completed (loss: 0.0604:  10%|▉         | 12/125 [01:13<11:02,  5.86s/it]
Training Epoch: 10/100, batch 24/250 completed (loss: 0.0603:  10%|█         | 13/125 [01:16<10:56,  5.86s/it]
Training Epoch: 10/100, batch 24/250 completed (loss: 0.0604:  10%|█         | 13/125 [01:16<10:56,  5.86s/it]
Training Epoch: 10/100, batch 24/250 completed (loss: 0.0416:  10%|█         | 13/125 [01:16<10:56,  5.86s/it]
Training Epoch: 10/100, batch 24/250 completed (loss: 0.1728:  10%|█         | 13/125 [01:16<10:56,  5.86s/it]
Training Epoch: 10/100, batch 25/250 completed (loss: 0.0157:  10%|█         | 13/125 [01:16<10:56,  5.86s/it]
Training Epoch: 10/100, batch 25/250 completed (loss: 0.0421:  10%|█         | 13/125 [01:16<10:56,  5.86s/it]
Training Epoch: 10/100, batch 25/250 completed (loss: 0.1312:  10%|█         | 13/125 [01:16<10:56,  5.86s/it]
Training Epoch: 10/100, batch 25/250 completed (loss: 0.0105:  10%|█         | 13/125 [01:16<10:56,  5.86s/it]
Training Epoch: 10/100, batch 26/250 completed (loss: 0.0559:  10%|█         | 13/125 [01:19<10:56,  5.86s/it]
Training Epoch: 10/100, batch 26/250 completed (loss: 0.0484:  10%|█         | 13/125 [01:19<10:56,  5.86s/it]
Training Epoch: 10/100, batch 26/250 completed (loss: 0.1509:  10%|█         | 13/125 [01:19<10:56,  5.86s/it]
Training Epoch: 10/100, batch 26/250 completed (loss: 0.0224:  10%|█         | 13/125 [01:19<10:56,  5.86s/it]
Training Epoch: 10/100, batch 26/250 completed (loss: 0.0224:  11%|█         | 14/125 [01:22<11:01,  5.96s/it]
Training Epoch: 10/100, batch 26/250 completed (loss: 0.0559:  11%|█         | 14/125 [01:22<11:01,  5.96s/it]
Training Epoch: 10/100, batch 26/250 completed (loss: 0.1509:  11%|█         | 14/125 [01:22<11:01,  5.96s/it]
Training Epoch: 10/100, batch 26/250 completed (loss: 0.0484:  11%|█         | 14/125 [01:22<11:01,  5.96s/it]
Training Epoch: 10/100, batch 27/250 completed (loss: 0.1482:  11%|█         | 14/125 [01:22<11:01,  5.96s/it]
Training Epoch: 10/100, batch 27/250 completed (loss: 0.1269:  11%|█         | 14/125 [01:22<11:01,  5.96s/it]
Training Epoch: 10/100, batch 27/250 completed (loss: 0.2538:  11%|█         | 14/125 [01:22<11:01,  5.96s/it]
Training Epoch: 10/100, batch 27/250 completed (loss: 0.0122:  11%|█         | 14/125 [01:22<11:01,  5.96s/it]
Training Epoch: 10/100, batch 28/250 completed (loss: 0.0413:  11%|█         | 14/125 [01:25<11:01,  5.96s/it]
Training Epoch: 10/100, batch 28/250 completed (loss: 0.0574:  11%|█         | 14/125 [01:25<11:01,  5.96s/it]
Training Epoch: 10/100, batch 28/250 completed (loss: 0.01:  11%|█         | 14/125 [01:25<11:01,  5.96s/it]  
Training Epoch: 10/100, batch 28/250 completed (loss: 0.0659:  11%|█         | 14/125 [01:25<11:01,  5.96s/it]
Training Epoch: 10/100, batch 28/250 completed (loss: 0.0413:  12%|█▏        | 15/125 [01:28<10:55,  5.96s/it]
Training Epoch: 10/100, batch 28/250 completed (loss: 0.0659:  12%|█▏        | 15/125 [01:28<10:55,  5.96s/it]
Training Epoch: 10/100, batch 28/250 completed (loss: 0.0574:  12%|█▏        | 15/125 [01:28<10:55,  5.96s/it]
Training Epoch: 10/100, batch 28/250 completed (loss: 0.01:  12%|█▏        | 15/125 [01:28<10:56,  5.97s/it]
Training Epoch: 10/100, batch 29/250 completed (loss: 0.0361:  12%|█▏        | 15/125 [01:28<10:56,  5.97s/it]
Training Epoch: 10/100, batch 29/250 completed (loss: 0.2393:  12%|█▏        | 15/125 [01:28<10:55,  5.96s/it]
Training Epoch: 10/100, batch 29/250 completed (loss: 0.0418:  12%|█▏        | 15/125 [01:28<10:55,  5.96s/it]
Training Epoch: 10/100, batch 29/250 completed (loss: 0.0253:  12%|█▏        | 15/125 [01:28<10:55,  5.96s/it]
Training Epoch: 10/100, batch 30/250 completed (loss: 0.0595:  12%|█▏        | 15/125 [01:31<10:56,  5.97s/it]
Training Epoch: 10/100, batch 30/250 completed (loss: 0.0565:  12%|█▏        | 15/125 [01:31<10:55,  5.96s/it]
Training Epoch: 10/100, batch 30/250 completed (loss: 0.0858:  12%|█▏        | 15/125 [01:31<10:55,  5.96s/it]
Training Epoch: 10/100, batch 30/250 completed (loss: 0.0395:  12%|█▏        | 15/125 [01:31<10:55,  5.96s/it]
Training Epoch: 10/100, batch 30/250 completed (loss: 0.0858:  13%|█▎        | 16/125 [01:34<10:57,  6.03s/it]
Training Epoch: 10/100, batch 30/250 completed (loss: 0.0395:  13%|█▎        | 16/125 [01:34<10:57,  6.03s/it]
Training Epoch: 10/100, batch 30/250 completed (loss: 0.0565:  13%|█▎        | 16/125 [01:34<10:57,  6.03s/it]
Training Epoch: 10/100, batch 30/250 completed (loss: 0.0595:  13%|█▎        | 16/125 [01:34<10:57,  6.04s/it]
Training Epoch: 10/100, batch 31/250 completed (loss: 0.0314:  13%|█▎        | 16/125 [01:34<10:57,  6.03s/it]
Training Epoch: 10/100, batch 31/250 completed (loss: 0.0742:  13%|█▎        | 16/125 [01:34<10:57,  6.04s/it]
Training Epoch: 10/100, batch 31/250 completed (loss: 0.1311:  13%|█▎        | 16/125 [01:34<10:57,  6.03s/it]
Training Epoch: 10/100, batch 31/250 completed (loss: 0.1082:  13%|█▎        | 16/125 [01:34<10:57,  6.03s/it]
Training Epoch: 10/100, batch 32/250 completed (loss: 0.0355:  13%|█▎        | 16/125 [01:37<10:57,  6.04s/it]
Training Epoch: 10/100, batch 32/250 completed (loss: 0.0549:  13%|█▎        | 16/125 [01:37<10:57,  6.03s/it]
Training Epoch: 10/100, batch 32/250 completed (loss: 0.2235:  13%|█▎        | 16/125 [01:37<10:57,  6.03s/it]
Training Epoch: 10/100, batch 32/250 completed (loss: 0.1041:  13%|█▎        | 16/125 [01:37<10:57,  6.03s/it]
Training Epoch: 10/100, batch 32/250 completed (loss: 0.1041:  14%|█▎        | 17/125 [01:40<10:44,  5.97s/it]
Training Epoch: 10/100, batch 32/250 completed (loss: 0.2235:  14%|█▎        | 17/125 [01:40<10:44,  5.97s/it]
Training Epoch: 10/100, batch 32/250 completed (loss: 0.0549:  14%|█▎        | 17/125 [01:40<10:44,  5.97s/it]
Training Epoch: 10/100, batch 32/250 completed (loss: 0.0355:  14%|█▎        | 17/125 [01:40<10:44,  5.97s/it]
Training Epoch: 10/100, batch 33/250 completed (loss: 0.0204:  14%|█▎        | 17/125 [01:40<10:44,  5.97s/it]
Training Epoch: 10/100, batch 33/250 completed (loss: 0.0422:  14%|█▎        | 17/125 [01:40<10:44,  5.97s/it]
Training Epoch: 10/100, batch 33/250 completed (loss: 0.0208:  14%|█▎        | 17/125 [01:40<10:44,  5.97s/it]
Training Epoch: 10/100, batch 33/250 completed (loss: 0.0554:  14%|█▎        | 17/125 [01:40<10:44,  5.97s/it]
Training Epoch: 10/100, batch 34/250 completed (loss: 0.1041:  14%|█▎        | 17/125 [01:43<10:44,  5.97s/it]
Training Epoch: 10/100, batch 34/250 completed (loss: 0.1466:  14%|█▎        | 17/125 [01:43<10:44,  5.97s/it]
Training Epoch: 10/100, batch 34/250 completed (loss: 0.0244:  14%|█▎        | 17/125 [01:43<10:44,  5.97s/it]
Training Epoch: 10/100, batch 34/250 completed (loss: 0.0544:  14%|█▎        | 17/125 [01:43<10:44,  5.97s/it]
Training Epoch: 10/100, batch 34/250 completed (loss: 0.0544:  14%|█▍        | 18/125 [01:46<10:37,  5.96s/it]
Training Epoch: 10/100, batch 34/250 completed (loss: 0.1041:  14%|█▍        | 18/125 [01:46<10:37,  5.96s/it]
Training Epoch: 10/100, batch 34/250 completed (loss: 0.1466:  14%|█▍        | 18/125 [01:46<10:37,  5.96s/it]
Training Epoch: 10/100, batch 34/250 completed (loss: 0.0244:  14%|█▍        | 18/125 [01:46<10:37,  5.96s/it]
Training Epoch: 10/100, batch 35/250 completed (loss: 0.0269:  14%|█▍        | 18/125 [01:46<10:37,  5.96s/it]
Training Epoch: 10/100, batch 35/250 completed (loss: 0.034:  14%|█▍        | 18/125 [01:46<10:37,  5.96s/it] 
Training Epoch: 10/100, batch 35/250 completed (loss: 0.1279:  14%|█▍        | 18/125 [01:46<10:37,  5.96s/it]
Training Epoch: 10/100, batch 35/250 completed (loss: 0.1195:  14%|█▍        | 18/125 [01:46<10:37,  5.96s/it]
Training Epoch: 10/100, batch 36/250 completed (loss: 0.0206:  14%|█▍        | 18/125 [01:49<10:37,  5.96s/it]
Training Epoch: 10/100, batch 36/250 completed (loss: 0.0195:  14%|█▍        | 18/125 [01:49<10:37,  5.96s/it]
Training Epoch: 10/100, batch 36/250 completed (loss: 0.0401:  14%|█▍        | 18/125 [01:49<10:37,  5.96s/it]
Training Epoch: 10/100, batch 36/250 completed (loss: 0.0362:  14%|█▍        | 18/125 [01:49<10:37,  5.96s/it]
Training Epoch: 10/100, batch 36/250 completed (loss: 0.0206:  15%|█▌        | 19/125 [01:52<10:41,  6.05s/it]
Training Epoch: 10/100, batch 36/250 completed (loss: 0.0362:  15%|█▌        | 19/125 [01:52<10:41,  6.05s/it]
Training Epoch: 10/100, batch 36/250 completed (loss: 0.0195:  15%|█▌        | 19/125 [01:52<10:41,  6.05s/it]
Training Epoch: 10/100, batch 36/250 completed (loss: 0.0401:  15%|█▌        | 19/125 [01:52<10:41,  6.05s/it]
Training Epoch: 10/100, batch 37/250 completed (loss: 0.0077:  15%|█▌        | 19/125 [01:52<10:41,  6.05s/it]
Training Epoch: 10/100, batch 37/250 completed (loss: 0.0214:  15%|█▌        | 19/125 [01:52<10:41,  6.05s/it]
Training Epoch: 10/100, batch 37/250 completed (loss: 0.0283:  15%|█▌        | 19/125 [01:52<10:41,  6.05s/it]
Training Epoch: 10/100, batch 37/250 completed (loss: 0.0588:  15%|█▌        | 19/125 [01:52<10:41,  6.05s/it]
Training Epoch: 10/100, batch 38/250 completed (loss: 0.047:  15%|█▌        | 19/125 [01:55<10:41,  6.05s/it] 
Training Epoch: 10/100, batch 38/250 completed (loss: 0.0262:  15%|█▌        | 19/125 [01:55<10:41,  6.05s/it]
Training Epoch: 10/100, batch 38/250 completed (loss: 0.0754:  15%|█▌        | 19/125 [01:55<10:41,  6.05s/it]
Training Epoch: 10/100, batch 38/250 completed (loss: 0.0239:  15%|█▌        | 19/125 [01:55<10:41,  6.05s/it]
Training Epoch: 10/100, batch 38/250 completed (loss: 0.047:  16%|█▌        | 20/125 [01:58<10:26,  5.97s/it]
Training Epoch: 10/100, batch 38/250 completed (loss: 0.0239:  16%|█▌        | 20/125 [01:58<10:26,  5.97s/it]
Training Epoch: 10/100, batch 38/250 completed (loss: 0.0754:  16%|█▌        | 20/125 [01:58<10:26,  5.96s/it]
Training Epoch: 10/100, batch 38/250 completed (loss: 0.0262:  16%|█▌        | 20/125 [01:58<10:26,  5.97s/it]
Training Epoch: 10/100, batch 39/250 completed (loss: 0.0167:  16%|█▌        | 20/125 [01:58<10:26,  5.97s/it]
Training Epoch: 10/100, batch 39/250 completed (loss: 0.0483:  16%|█▌        | 20/125 [01:58<10:26,  5.96s/it]
Training Epoch: 10/100, batch 39/250 completed (loss: 0.0768:  16%|█▌        | 20/125 [01:58<10:26,  5.97s/it]
Training Epoch: 10/100, batch 39/250 completed (loss: 0.0427:  16%|█▌        | 20/125 [01:58<10:26,  5.97s/it]
Training Epoch: 10/100, batch 40/250 completed (loss: 0.0585:  16%|█▌        | 20/125 [02:01<10:26,  5.97s/it]
Training Epoch: 10/100, batch 40/250 completed (loss: 0.054:  16%|█▌        | 20/125 [02:01<10:26,  5.97s/it] 
Training Epoch: 10/100, batch 40/250 completed (loss: 0.0306:  16%|█▌        | 20/125 [02:01<10:26,  5.96s/it]
Training Epoch: 10/100, batch 40/250 completed (loss: 0.0699:  16%|█▌        | 20/125 [02:01<10:26,  5.97s/it]
Training Epoch: 10/100, batch 40/250 completed (loss: 0.0699:  17%|█▋        | 21/125 [02:04<10:30,  6.06s/it]
Training Epoch: 10/100, batch 40/250 completed (loss: 0.0585:  17%|█▋        | 21/125 [02:04<10:31,  6.07s/it]
Training Epoch: 10/100, batch 40/250 completed (loss: 0.054:  17%|█▋        | 21/125 [02:04<10:31,  6.07s/it]
Training Epoch: 10/100, batch 40/250 completed (loss: 0.0306:  17%|█▋        | 21/125 [02:04<10:31,  6.07s/it]
Training Epoch: 10/100, batch 41/250 completed (loss: 0.0437:  17%|█▋        | 21/125 [02:04<10:31,  6.07s/it]
Training Epoch: 10/100, batch 41/250 completed (loss: 0.0164:  17%|█▋        | 21/125 [02:04<10:31,  6.07s/it]
Training Epoch: 10/100, batch 41/250 completed (loss: 0.0182:  17%|█▋        | 21/125 [02:04<10:31,  6.07s/it]
Training Epoch: 10/100, batch 41/250 completed (loss: 0.2758:  17%|█▋        | 21/125 [02:04<10:30,  6.06s/it]
Training Epoch: 10/100, batch 42/250 completed (loss: 0.0189:  17%|█▋        | 21/125 [02:07<10:31,  6.07s/it]
Training Epoch: 10/100, batch 42/250 completed (loss: 0.0631:  17%|█▋        | 21/125 [02:07<10:31,  6.07s/it]
Training Epoch: 10/100, batch 42/250 completed (loss: 0.017:  17%|█▋        | 21/125 [02:07<10:31,  6.07s/it] 
Training Epoch: 10/100, batch 42/250 completed (loss: 0.0153:  17%|█▋        | 21/125 [02:07<10:30,  6.06s/it]
Training Epoch: 10/100, batch 42/250 completed (loss: 0.0189:  18%|█▊        | 22/125 [02:10<10:22,  6.05s/it]
Training Epoch: 10/100, batch 42/250 completed (loss: 0.0153:  18%|█▊        | 22/125 [02:10<10:23,  6.05s/it]
Training Epoch: 10/100, batch 42/250 completed (loss: 0.0631:  18%|█▊        | 22/125 [02:10<10:22,  6.05s/it]
Training Epoch: 10/100, batch 42/250 completed (loss: 0.017:  18%|█▊        | 22/125 [02:10<10:22,  6.05s/it]
Training Epoch: 10/100, batch 43/250 completed (loss: 0.0117:  18%|█▊        | 22/125 [02:10<10:22,  6.05s/it]
Training Epoch: 10/100, batch 43/250 completed (loss: 0.0378:  18%|█▊        | 22/125 [02:10<10:22,  6.05s/it]
Training Epoch: 10/100, batch 43/250 completed (loss: 0.0186:  18%|█▊        | 22/125 [02:10<10:22,  6.05s/it]
Training Epoch: 10/100, batch 43/250 completed (loss: 0.0296:  18%|█▊        | 22/125 [02:10<10:23,  6.05s/it]
Training Epoch: 10/100, batch 44/250 completed (loss: 0.0265:  18%|█▊        | 22/125 [02:13<10:22,  6.05s/it]
Training Epoch: 10/100, batch 44/250 completed (loss: 0.0196:  18%|█▊        | 22/125 [02:13<10:22,  6.05s/it]
Training Epoch: 10/100, batch 44/250 completed (loss: 0.031:  18%|█▊        | 22/125 [02:13<10:22,  6.05s/it] 
Training Epoch: 10/100, batch 44/250 completed (loss: 0.0462:  18%|█▊        | 22/125 [02:13<10:23,  6.05s/it]
Training Epoch: 10/100, batch 44/250 completed (loss: 0.031:  18%|█▊        | 23/125 [02:16<10:14,  6.02s/it]
Training Epoch: 10/100, batch 44/250 completed (loss: 0.0462:  18%|█▊        | 23/125 [02:16<10:14,  6.02s/it]
Training Epoch: 10/100, batch 44/250 completed (loss: 0.0196:  18%|█▊        | 23/125 [02:16<10:14,  6.02s/it]
Training Epoch: 10/100, batch 44/250 completed (loss: 0.0265:  18%|█▊        | 23/125 [02:16<10:14,  6.02s/it]
Training Epoch: 10/100, batch 45/250 completed (loss: 0.0513:  18%|█▊        | 23/125 [02:16<10:14,  6.02s/it]
Training Epoch: 10/100, batch 45/250 completed (loss: 0.0167:  18%|█▊        | 23/125 [02:16<10:14,  6.02s/it]
Training Epoch: 10/100, batch 45/250 completed (loss: 0.3408:  18%|█▊        | 23/125 [02:16<10:14,  6.02s/it]
Training Epoch: 10/100, batch 45/250 completed (loss: 0.2431:  18%|█▊        | 23/125 [02:16<10:14,  6.02s/it]
Training Epoch: 10/100, batch 46/250 completed (loss: 0.0407:  18%|█▊        | 23/125 [02:19<10:14,  6.02s/it]
Training Epoch: 10/100, batch 46/250 completed (loss: 0.2251:  18%|█▊        | 23/125 [02:19<10:14,  6.02s/it]
Training Epoch: 10/100, batch 46/250 completed (loss: 0.0458:  18%|█▊        | 23/125 [02:19<10:14,  6.02s/it]
Training Epoch: 10/100, batch 46/250 completed (loss: 0.0706:  18%|█▊        | 23/125 [02:19<10:14,  6.02s/it]
Training Epoch: 10/100, batch 46/250 completed (loss: 0.0706:  19%|█▉        | 24/125 [02:22<10:12,  6.07s/it]
Training Epoch: 10/100, batch 46/250 completed (loss: 0.0407:  19%|█▉        | 24/125 [02:22<10:12,  6.07s/it]
Training Epoch: 10/100, batch 46/250 completed (loss: 0.2251:  19%|█▉        | 24/125 [02:22<10:12,  6.07s/it]
Training Epoch: 10/100, batch 46/250 completed (loss: 0.0458:  19%|█▉        | 24/125 [02:22<10:13,  6.07s/it]
Training Epoch: 10/100, batch 47/250 completed (loss: 0.0103:  19%|█▉        | 24/125 [02:23<10:12,  6.07s/it]
Training Epoch: 10/100, batch 47/250 completed (loss: 0.0381:  19%|█▉        | 24/125 [02:23<10:13,  6.07s/it]
Training Epoch: 10/100, batch 47/250 completed (loss: 0.0241:  19%|█▉        | 24/125 [02:23<10:12,  6.07s/it]
Training Epoch: 10/100, batch 47/250 completed (loss: 0.0494:  19%|█▉        | 24/125 [02:23<10:12,  6.07s/it]
Training Epoch: 10/100, batch 48/250 completed (loss: 0.0567:  19%|█▉        | 24/125 [02:25<10:13,  6.07s/it]
Training Epoch: 10/100, batch 48/250 completed (loss: 0.0417:  19%|█▉        | 24/125 [02:25<10:12,  6.07s/it]
Training Epoch: 10/100, batch 48/250 completed (loss: 0.0928:  19%|█▉        | 24/125 [02:25<10:12,  6.07s/it]
Training Epoch: 10/100, batch 48/250 completed (loss: 0.0545:  19%|█▉        | 24/125 [02:25<10:12,  6.07s/it]
Training Epoch: 10/100, batch 48/250 completed (loss: 0.0545:  20%|██        | 25/125 [02:28<09:58,  5.99s/it]
Training Epoch: 10/100, batch 48/250 completed (loss: 0.0417:  20%|██        | 25/125 [02:28<09:58,  5.99s/it]
Training Epoch: 10/100, batch 48/250 completed (loss: 0.0567:  20%|██        | 25/125 [02:28<09:58,  5.99s/it]
Training Epoch: 10/100, batch 48/250 completed (loss: 0.0928:  20%|██        | 25/125 [02:28<09:59,  5.99s/it]
Training Epoch: 10/100, batch 49/250 completed (loss: 0.0428:  20%|██        | 25/125 [02:28<09:58,  5.99s/it]
Training Epoch: 10/100, batch 49/250 completed (loss: 0.1654:  20%|██        | 25/125 [02:28<09:59,  5.99s/it]
Training Epoch: 10/100, batch 49/250 completed (loss: 0.0549:  20%|██        | 25/125 [02:28<09:58,  5.99s/it]
Training Epoch: 10/100, batch 49/250 completed (loss: 0.1486:  20%|██        | 25/125 [02:28<09:58,  5.99s/it]
Training Epoch: 10/100, batch 50/250 completed (loss: 0.0308:  20%|██        | 25/125 [02:31<09:58,  5.99s/it]
Training Epoch: 10/100, batch 50/250 completed (loss: 0.1514:  20%|██        | 25/125 [02:31<09:59,  5.99s/it]
Training Epoch: 10/100, batch 50/250 completed (loss: 0.122:  20%|██        | 25/125 [02:31<09:58,  5.99s/it] 
Training Epoch: 10/100, batch 50/250 completed (loss: 0.0473:  20%|██        | 25/125 [02:31<09:58,  5.99s/it]
Training Epoch: 10/100, batch 50/250 completed (loss: 0.0473:  21%|██        | 26/125 [02:34<09:51,  5.98s/it]
Training Epoch: 10/100, batch 50/250 completed (loss: 0.1514:  21%|██        | 26/125 [02:34<09:51,  5.98s/it]
Training Epoch: 10/100, batch 50/250 completed (loss: 0.122:  21%|██        | 26/125 [02:34<09:51,  5.98s/it]
Training Epoch: 10/100, batch 50/250 completed (loss: 0.0308:  21%|██        | 26/125 [02:34<09:51,  5.98s/it]
Training Epoch: 10/100, batch 51/250 completed (loss: 0.0329:  21%|██        | 26/125 [02:34<09:51,  5.98s/it]
Training Epoch: 10/100, batch 51/250 completed (loss: 0.1145:  21%|██        | 26/125 [02:34<09:51,  5.98s/it]
Training Epoch: 10/100, batch 51/250 completed (loss: 0.0422:  21%|██        | 26/125 [02:34<09:51,  5.98s/it]
Training Epoch: 10/100, batch 51/250 completed (loss: 0.3906:  21%|██        | 26/125 [02:34<09:51,  5.98s/it]
Training Epoch: 10/100, batch 52/250 completed (loss: 0.0731:  21%|██        | 26/125 [02:37<09:51,  5.98s/it]
Training Epoch: 10/100, batch 52/250 completed (loss: 0.0271:  21%|██        | 26/125 [02:37<09:51,  5.98s/it]
Training Epoch: 10/100, batch 52/250 completed (loss: 0.0579:  21%|██        | 26/125 [02:37<09:51,  5.98s/it]
Training Epoch: 10/100, batch 52/250 completed (loss: 0.0176:  21%|██        | 26/125 [02:38<09:51,  5.98s/it]
Training Epoch: 10/100, batch 52/250 completed (loss: 0.0176:  22%|██▏       | 27/125 [02:40<09:49,  6.02s/it]
Training Epoch: 10/100, batch 52/250 completed (loss: 0.0731:  22%|██▏       | 27/125 [02:40<09:49,  6.02s/it]
Training Epoch: 10/100, batch 52/250 completed (loss: 0.0579:  22%|██▏       | 27/125 [02:40<09:49,  6.02s/it]
Training Epoch: 10/100, batch 52/250 completed (loss: 0.0271:  22%|██▏       | 27/125 [02:40<09:49,  6.02s/it]
Training Epoch: 10/100, batch 53/250 completed (loss: 0.0621:  22%|██▏       | 27/125 [02:40<09:49,  6.02s/it]
Training Epoch: 10/100, batch 53/250 completed (loss: 0.0146:  22%|██▏       | 27/125 [02:40<09:49,  6.02s/it]
Training Epoch: 10/100, batch 53/250 completed (loss: 0.0543:  22%|██▏       | 27/125 [02:40<09:49,  6.02s/it]
Training Epoch: 10/100, batch 53/250 completed (loss: 0.0664:  22%|██▏       | 27/125 [02:40<09:49,  6.02s/it]
Training Epoch: 10/100, batch 54/250 completed (loss: 0.0321:  22%|██▏       | 27/125 [02:43<09:49,  6.02s/it]
Training Epoch: 10/100, batch 54/250 completed (loss: 0.0385:  22%|██▏       | 27/125 [02:43<09:49,  6.02s/it]
Training Epoch: 10/100, batch 54/250 completed (loss: 0.1659:  22%|██▏       | 27/125 [02:43<09:49,  6.02s/it]
Training Epoch: 10/100, batch 54/250 completed (loss: 0.0159:  22%|██▏       | 27/125 [02:43<09:49,  6.02s/it]
Training Epoch: 10/100, batch 54/250 completed (loss: 0.0321:  22%|██▏       | 28/125 [02:46<09:37,  5.96s/it]
Training Epoch: 10/100, batch 54/250 completed (loss: 0.1659:  22%|██▏       | 28/125 [02:46<09:37,  5.96s/it]
Training Epoch: 10/100, batch 54/250 completed (loss: 0.0159:  22%|██▏       | 28/125 [02:46<09:38,  5.96s/it]
Training Epoch: 10/100, batch 54/250 completed (loss: 0.0385:  22%|██▏       | 28/125 [02:46<09:38,  5.96s/it]
Training Epoch: 10/100, batch 55/250 completed (loss: 0.0122:  22%|██▏       | 28/125 [02:46<09:37,  5.96s/it]
Training Epoch: 10/100, batch 55/250 completed (loss: 0.1696:  22%|██▏       | 28/125 [02:46<09:37,  5.96s/it]
Training Epoch: 10/100, batch 55/250 completed (loss: 0.0179:  22%|██▏       | 28/125 [02:46<09:38,  5.96s/it]
Training Epoch: 10/100, batch 55/250 completed (loss: 0.0656:  22%|██▏       | 28/125 [02:46<09:38,  5.96s/it]
Training Epoch: 10/100, batch 56/250 completed (loss: 0.0875:  22%|██▏       | 28/125 [02:49<09:37,  5.96s/it]
Training Epoch: 10/100, batch 56/250 completed (loss: 0.0291:  22%|██▏       | 28/125 [02:49<09:38,  5.96s/it]
Training Epoch: 10/100, batch 56/250 completed (loss: 0.0212:  22%|██▏       | 28/125 [02:49<09:37,  5.96s/it]
Training Epoch: 10/100, batch 56/250 completed (loss: 0.0297:  22%|██▏       | 28/125 [02:49<09:38,  5.96s/it]
Training Epoch: 10/100, batch 56/250 completed (loss: 0.0297:  23%|██▎       | 29/125 [02:52<09:41,  6.06s/it]
Training Epoch: 10/100, batch 56/250 completed (loss: 0.0291:  23%|██▎       | 29/125 [02:52<09:41,  6.06s/it]
Training Epoch: 10/100, batch 56/250 completed (loss: 0.0875:  23%|██▎       | 29/125 [02:52<09:42,  6.06s/it]
Training Epoch: 10/100, batch 56/250 completed (loss: 0.0212:  23%|██▎       | 29/125 [02:52<09:42,  6.07s/it]
Training Epoch: 10/100, batch 57/250 completed (loss: 0.0477:  23%|██▎       | 29/125 [02:53<09:42,  6.06s/it]
Training Epoch: 10/100, batch 57/250 completed (loss: 0.1113:  23%|██▎       | 29/125 [02:53<09:41,  6.06s/it]
Training Epoch: 10/100, batch 57/250 completed (loss: 0.0993:  23%|██▎       | 29/125 [02:53<09:42,  6.07s/it]
Training Epoch: 10/100, batch 57/250 completed (loss: 0.1179:  23%|██▎       | 29/125 [02:53<09:41,  6.06s/it]
Training Epoch: 10/100, batch 58/250 completed (loss: 0.0514:  23%|██▎       | 29/125 [02:55<09:41,  6.06s/it]
Training Epoch: 10/100, batch 58/250 completed (loss: 0.0304:  23%|██▎       | 29/125 [02:55<09:42,  6.06s/it]
Training Epoch: 10/100, batch 58/250 completed (loss: 0.0314:  23%|██▎       | 29/125 [02:55<09:42,  6.07s/it]
Training Epoch: 10/100, batch 58/250 completed (loss: 0.0126:  23%|██▎       | 29/125 [02:55<09:41,  6.06s/it]
Training Epoch: 10/100, batch 58/250 completed (loss: 0.0514:  24%|██▍       | 30/125 [02:58<09:29,  6.00s/it]
Training Epoch: 10/100, batch 58/250 completed (loss: 0.0126:  24%|██▍       | 30/125 [02:58<09:29,  6.00s/it]
Training Epoch: 10/100, batch 58/250 completed (loss: 0.0304:  24%|██▍       | 30/125 [02:58<09:30,  6.00s/it]
Training Epoch: 10/100, batch 58/250 completed (loss: 0.0314:  24%|██▍       | 30/125 [02:58<09:30,  6.01s/it]
Training Epoch: 10/100, batch 59/250 completed (loss: 0.0522:  24%|██▍       | 30/125 [02:58<09:29,  6.00s/it]
Training Epoch: 10/100, batch 59/250 completed (loss: 0.0154:  24%|██▍       | 30/125 [02:58<09:30,  6.00s/it]
Training Epoch: 10/100, batch 59/250 completed (loss: 0.0175:  24%|██▍       | 30/125 [02:58<09:30,  6.01s/it]
Training Epoch: 10/100, batch 59/250 completed (loss: 0.0033:  24%|██▍       | 30/125 [02:58<09:29,  6.00s/it]
Training Epoch: 10/100, batch 60/250 completed (loss: 0.181:  24%|██▍       | 30/125 [03:01<09:29,  6.00s/it] 
Training Epoch: 10/100, batch 60/250 completed (loss: 0.146:  24%|██▍       | 30/125 [03:01<09:30,  6.01s/it] 
Training Epoch: 10/100, batch 60/250 completed (loss: 0.1006:  24%|██▍       | 30/125 [03:01<09:30,  6.00s/it]
Training Epoch: 10/100, batch 60/250 completed (loss: 0.0227:  24%|██▍       | 30/125 [03:01<09:29,  6.00s/it]
Training Epoch: 10/100, batch 60/250 completed (loss: 0.181:  25%|██▍       | 31/125 [03:04<09:19,  5.96s/it]
Training Epoch: 10/100, batch 60/250 completed (loss: 0.0227:  25%|██▍       | 31/125 [03:04<09:20,  5.96s/it]
Training Epoch: 10/100, batch 60/250 completed (loss: 0.1006:  25%|██▍       | 31/125 [03:04<09:20,  5.96s/it]
Training Epoch: 10/100, batch 60/250 completed (loss: 0.146:  25%|██▍       | 31/125 [03:04<09:20,  5.96s/it]
Training Epoch: 10/100, batch 61/250 completed (loss: 0.0271:  25%|██▍       | 31/125 [03:04<09:19,  5.96s/it]
Training Epoch: 10/100, batch 61/250 completed (loss: 0.0364:  25%|██▍       | 31/125 [03:04<09:20,  5.96s/it]
Training Epoch: 10/100, batch 61/250 completed (loss: 0.1529:  25%|██▍       | 31/125 [03:04<09:20,  5.96s/it]
Training Epoch: 10/100, batch 61/250 completed (loss: 0.0865:  25%|██▍       | 31/125 [03:04<09:20,  5.96s/it]
Training Epoch: 10/100, batch 62/250 completed (loss: 0.0134:  25%|██▍       | 31/125 [03:07<09:20,  5.96s/it]
Training Epoch: 10/100, batch 62/250 completed (loss: 0.066:  25%|██▍       | 31/125 [03:07<09:20,  5.96s/it] 
Training Epoch: 10/100, batch 62/250 completed (loss: 0.1322:  25%|██▍       | 31/125 [03:07<09:19,  5.96s/it]
Training Epoch: 10/100, batch 62/250 completed (loss: 0.0453:  25%|██▍       | 31/125 [03:08<09:20,  5.96s/it]
Training Epoch: 10/100, batch 62/250 completed (loss: 0.1322:  26%|██▌       | 32/125 [03:11<09:24,  6.07s/it]
Training Epoch: 10/100, batch 62/250 completed (loss: 0.0453:  26%|██▌       | 32/125 [03:11<09:25,  6.08s/it]
Training Epoch: 10/100, batch 62/250 completed (loss: 0.066:  26%|██▌       | 32/125 [03:11<09:24,  6.07s/it]
Training Epoch: 10/100, batch 62/250 completed (loss: 0.0134:  26%|██▌       | 32/125 [03:11<09:25,  6.08s/it]
Training Epoch: 10/100, batch 63/250 completed (loss: 0.127:  26%|██▌       | 32/125 [03:11<09:25,  6.08s/it] 
Training Epoch: 10/100, batch 63/250 completed (loss: 0.0531:  26%|██▌       | 32/125 [03:11<09:24,  6.07s/it]
Training Epoch: 10/100, batch 63/250 completed (loss: 0.0146:  26%|██▌       | 32/125 [03:11<09:24,  6.07s/it]
Training Epoch: 10/100, batch 63/250 completed (loss: 0.0342:  26%|██▌       | 32/125 [03:11<09:25,  6.08s/it]
Training Epoch: 10/100, batch 64/250 completed (loss: 0.0172:  26%|██▌       | 32/125 [03:13<09:25,  6.08s/it]
Training Epoch: 10/100, batch 64/250 completed (loss: 0.0918:  26%|██▌       | 32/125 [03:13<09:24,  6.07s/it]
Training Epoch: 10/100, batch 64/250 completed (loss: 0.0128:  26%|██▌       | 32/125 [03:13<09:24,  6.07s/it]
Training Epoch: 10/100, batch 64/250 completed (loss: 0.0226:  26%|██▌       | 32/125 [03:14<09:25,  6.08s/it]
Training Epoch: 10/100, batch 64/250 completed (loss: 0.0128:  26%|██▋       | 33/125 [03:16<09:12,  6.00s/it]
Training Epoch: 10/100, batch 64/250 completed (loss: 0.0226:  26%|██▋       | 33/125 [03:16<09:12,  6.01s/it]
Training Epoch: 10/100, batch 64/250 completed (loss: 0.0172:  26%|██▋       | 33/125 [03:16<09:12,  6.00s/it]
Training Epoch: 10/100, batch 64/250 completed (loss: 0.0918:  26%|██▋       | 33/125 [03:16<09:12,  6.01s/it]
Training Epoch: 10/100, batch 65/250 completed (loss: 0.0235:  26%|██▋       | 33/125 [03:16<09:12,  6.01s/it]
Training Epoch: 10/100, batch 65/250 completed (loss: 0.0255:  26%|██▋       | 33/125 [03:16<09:12,  6.00s/it]
Training Epoch: 10/100, batch 65/250 completed (loss: 0.0333:  26%|██▋       | 33/125 [03:16<09:12,  6.00s/it]
Training Epoch: 10/100, batch 65/250 completed (loss: 0.0341:  26%|██▋       | 33/125 [03:16<09:12,  6.01s/it]
Training Epoch: 10/100, batch 66/250 completed (loss: 0.1688:  26%|██▋       | 33/125 [03:19<09:12,  6.00s/it]
Training Epoch: 10/100, batch 66/250 completed (loss: 0.0669:  26%|██▋       | 33/125 [03:19<09:12,  6.01s/it]
Training Epoch: 10/100, batch 66/250 completed (loss: 0.031:  26%|██▋       | 33/125 [03:19<09:12,  6.00s/it] 
Training Epoch: 10/100, batch 66/250 completed (loss: 0.0495:  26%|██▋       | 33/125 [03:19<09:12,  6.01s/it]
Training Epoch: 10/100, batch 66/250 completed (loss: 0.1688:  27%|██▋       | 34/125 [03:23<09:15,  6.10s/it]
Training Epoch: 10/100, batch 66/250 completed (loss: 0.0495:  27%|██▋       | 34/125 [03:23<09:15,  6.10s/it]
Training Epoch: 10/100, batch 66/250 completed (loss: 0.031:  27%|██▋       | 34/125 [03:23<09:15,  6.10s/it]
Training Epoch: 10/100, batch 66/250 completed (loss: 0.0669:  27%|██▋       | 34/125 [03:23<09:15,  6.10s/it]
Training Epoch: 10/100, batch 67/250 completed (loss: 0.0602:  27%|██▋       | 34/125 [03:23<09:15,  6.10s/it]
Training Epoch: 10/100, batch 67/250 completed (loss: 0.0352:  27%|██▋       | 34/125 [03:23<09:15,  6.10s/it]
Training Epoch: 10/100, batch 67/250 completed (loss: 0.0701:  27%|██▋       | 34/125 [03:23<09:15,  6.10s/it]
Training Epoch: 10/100, batch 67/250 completed (loss: 0.0506:  27%|██▋       | 34/125 [03:23<09:15,  6.10s/it]
Training Epoch: 10/100, batch 68/250 completed (loss: 0.0387:  27%|██▋       | 34/125 [03:26<09:15,  6.10s/it]
Training Epoch: 10/100, batch 68/250 completed (loss: 0.0141:  27%|██▋       | 34/125 [03:26<09:15,  6.10s/it]
Training Epoch: 10/100, batch 68/250 completed (loss: 0.0571:  27%|██▋       | 34/125 [03:26<09:15,  6.10s/it]
Training Epoch: 10/100, batch 68/250 completed (loss: 0.0198:  27%|██▋       | 34/125 [03:26<09:15,  6.10s/it]
Training Epoch: 10/100, batch 68/250 completed (loss: 0.0387:  28%|██▊       | 35/125 [03:29<09:06,  6.07s/it]
Training Epoch: 10/100, batch 68/250 completed (loss: 0.0198:  28%|██▊       | 35/125 [03:29<09:06,  6.07s/it]
Training Epoch: 10/100, batch 68/250 completed (loss: 0.0571:  28%|██▊       | 35/125 [03:29<09:06,  6.07s/it]
Training Epoch: 10/100, batch 68/250 completed (loss: 0.0141:  28%|██▊       | 35/125 [03:29<09:05,  6.07s/it]
Training Epoch: 10/100, batch 69/250 completed (loss: 0.0117:  28%|██▊       | 35/125 [03:29<09:06,  6.07s/it]
Training Epoch: 10/100, batch 69/250 completed (loss: 0.0147:  28%|██▊       | 35/125 [03:29<09:05,  6.07s/it]
Training Epoch: 10/100, batch 69/250 completed (loss: 0.0103:  28%|██▊       | 35/125 [03:29<09:06,  6.07s/it]
Training Epoch: 10/100, batch 69/250 completed (loss: 0.1443:  28%|██▊       | 35/125 [03:29<09:06,  6.07s/it]
Training Epoch: 10/100, batch 70/250 completed (loss: 0.008:  28%|██▊       | 35/125 [03:32<09:06,  6.07s/it] 
Training Epoch: 10/100, batch 70/250 completed (loss: 0.052:  28%|██▊       | 35/125 [03:32<09:05,  6.07s/it] 
Training Epoch: 10/100, batch 70/250 completed (loss: 0.153:  28%|██▊       | 35/125 [03:32<09:06,  6.07s/it] 
Training Epoch: 10/100, batch 70/250 completed (loss: 0.0288:  28%|██▊       | 35/125 [03:32<09:06,  6.07s/it]
Training Epoch: 10/100, batch 70/250 completed (loss: 0.008:  29%|██▉       | 36/125 [03:35<08:53,  6.00s/it]
Training Epoch: 10/100, batch 70/250 completed (loss: 0.0288:  29%|██▉       | 36/125 [03:35<08:53,  6.00s/it]
Training Epoch: 10/100, batch 70/250 completed (loss: 0.153:  29%|██▉       | 36/125 [03:35<08:53,  6.00s/it]
Training Epoch: 10/100, batch 70/250 completed (loss: 0.052:  29%|██▉       | 36/125 [03:35<08:53,  6.00s/it]
Training Epoch: 10/100, batch 71/250 completed (loss: 0.0665:  29%|██▉       | 36/125 [03:35<08:53,  6.00s/it]
Training Epoch: 10/100, batch 71/250 completed (loss: 0.0455:  29%|██▉       | 36/125 [03:35<08:53,  6.00s/it]
Training Epoch: 10/100, batch 71/250 completed (loss: 0.0538:  29%|██▉       | 36/125 [03:35<08:53,  6.00s/it]
Training Epoch: 10/100, batch 71/250 completed (loss: 0.0333:  29%|██▉       | 36/125 [03:35<08:53,  6.00s/it]
Training Epoch: 10/100, batch 72/250 completed (loss: 0.0931:  29%|██▉       | 36/125 [03:38<08:53,  6.00s/it]
Training Epoch: 10/100, batch 72/250 completed (loss: 0.1026:  29%|██▉       | 36/125 [03:38<08:53,  6.00s/it]
Training Epoch: 10/100, batch 72/250 completed (loss: 0.0441:  29%|██▉       | 36/125 [03:38<08:53,  6.00s/it]
Training Epoch: 10/100, batch 72/250 completed (loss: 0.0278:  29%|██▉       | 36/125 [03:38<08:53,  6.00s/it]
Training Epoch: 10/100, batch 72/250 completed (loss: 0.0441:  30%|██▉       | 37/125 [03:41<09:00,  6.14s/it]
Training Epoch: 10/100, batch 72/250 completed (loss: 0.1026:  30%|██▉       | 37/125 [03:41<09:00,  6.14s/it]
Training Epoch: 10/100, batch 72/250 completed (loss: 0.0931:  30%|██▉       | 37/125 [03:41<09:00,  6.14s/it]
Training Epoch: 10/100, batch 72/250 completed (loss: 0.0278:  30%|██▉       | 37/125 [03:41<09:00,  6.14s/it]
Training Epoch: 10/100, batch 73/250 completed (loss: 0.1916:  30%|██▉       | 37/125 [03:41<09:00,  6.14s/it]
Training Epoch: 10/100, batch 73/250 completed (loss: 0.1601:  30%|██▉       | 37/125 [03:41<09:00,  6.14s/it]
Training Epoch: 10/100, batch 73/250 completed (loss: 0.0318:  30%|██▉       | 37/125 [03:41<09:00,  6.14s/it]
Training Epoch: 10/100, batch 73/250 completed (loss: 0.1574:  30%|██▉       | 37/125 [03:41<09:00,  6.14s/it]
Training Epoch: 10/100, batch 74/250 completed (loss: 0.056:  30%|██▉       | 37/125 [03:44<09:00,  6.14s/it] 
Training Epoch: 10/100, batch 74/250 completed (loss: 0.0678:  30%|██▉       | 37/125 [03:44<09:00,  6.14s/it]
Training Epoch: 10/100, batch 74/250 completed (loss: 0.0254:  30%|██▉       | 37/125 [03:44<09:00,  6.14s/it]
Training Epoch: 10/100, batch 74/250 completed (loss: 0.2669:  30%|██▉       | 37/125 [03:44<09:00,  6.14s/it]
Training Epoch: 10/100, batch 74/250 completed (loss: 0.056:  30%|███       | 38/125 [03:47<08:46,  6.05s/it]
Training Epoch: 10/100, batch 74/250 completed (loss: 0.2669:  30%|███       | 38/125 [03:47<08:46,  6.05s/it]
Training Epoch: 10/100, batch 74/250 completed (loss: 0.0254:  30%|███       | 38/125 [03:47<08:46,  6.05s/it]
Training Epoch: 10/100, batch 74/250 completed (loss: 0.0678:  30%|███       | 38/125 [03:47<08:46,  6.06s/it]
Training Epoch: 10/100, batch 75/250 completed (loss: 0.0209:  30%|███       | 38/125 [03:47<08:46,  6.06s/it]
Training Epoch: 10/100, batch 75/250 completed (loss: 0.104:  30%|███       | 38/125 [03:47<08:46,  6.05s/it] 
Training Epoch: 10/100, batch 75/250 completed (loss: 0.0805:  30%|███       | 38/125 [03:47<08:46,  6.05s/it]
Training Epoch: 10/100, batch 75/250 completed (loss: 0.1815:  30%|███       | 38/125 [03:47<08:46,  6.05s/it]
Training Epoch: 10/100, batch 76/250 completed (loss: 0.1326:  30%|███       | 38/125 [03:50<08:46,  6.05s/it]
Training Epoch: 10/100, batch 76/250 completed (loss: 0.0377:  30%|███       | 38/125 [03:50<08:46,  6.05s/it]
Training Epoch: 10/100, batch 76/250 completed (loss: 0.0289:  30%|███       | 38/125 [03:50<08:46,  6.06s/it]
Training Epoch: 10/100, batch 76/250 completed (loss: 0.062:  30%|███       | 38/125 [03:50<08:46,  6.05s/it] 
Training Epoch: 10/100, batch 76/250 completed (loss: 0.062:  31%|███       | 39/125 [03:53<08:39,  6.05s/it]
Training Epoch: 10/100, batch 76/250 completed (loss: 0.0377:  31%|███       | 39/125 [03:53<08:40,  6.05s/it]
Training Epoch: 10/100, batch 76/250 completed (loss: 0.1326:  31%|███       | 39/125 [03:53<08:40,  6.05s/it]
Training Epoch: 10/100, batch 76/250 completed (loss: 0.0289:  31%|███       | 39/125 [03:53<08:41,  6.07s/it]
Training Epoch: 10/100, batch 77/250 completed (loss: 0.0889:  31%|███       | 39/125 [03:53<08:40,  6.05s/it]
Training Epoch: 10/100, batch 77/250 completed (loss: 0.0565:  31%|███       | 39/125 [03:53<08:41,  6.07s/it]
Training Epoch: 10/100, batch 77/250 completed (loss: 0.0343:  31%|███       | 39/125 [03:53<08:40,  6.05s/it]
Training Epoch: 10/100, batch 77/250 completed (loss: 0.1361:  31%|███       | 39/125 [03:53<08:39,  6.05s/it]
Training Epoch: 10/100, batch 78/250 completed (loss: 0.1905:  31%|███       | 39/125 [03:56<08:40,  6.05s/it]
Training Epoch: 10/100, batch 78/250 completed (loss: 0.0111:  31%|███       | 39/125 [03:56<08:40,  6.05s/it]
Training Epoch: 10/100, batch 78/250 completed (loss: 0.0871:  31%|███       | 39/125 [03:56<08:41,  6.07s/it]
Training Epoch: 10/100, batch 78/250 completed (loss: 0.0221:  31%|███       | 39/125 [03:56<08:39,  6.05s/it]
Training Epoch: 10/100, batch 78/250 completed (loss: 0.0111:  32%|███▏      | 40/125 [03:59<08:39,  6.11s/it]
Training Epoch: 10/100, batch 78/250 completed (loss: 0.0221:  32%|███▏      | 40/125 [03:59<08:39,  6.11s/it]
Training Epoch: 10/100, batch 78/250 completed (loss: 0.0871:  32%|███▏      | 40/125 [03:59<08:38,  6.11s/it]
Training Epoch: 10/100, batch 78/250 completed (loss: 0.1905:  32%|███▏      | 40/125 [03:59<08:39,  6.11s/it]
Training Epoch: 10/100, batch 79/250 completed (loss: 0.0504:  32%|███▏      | 40/125 [03:59<08:39,  6.11s/it]
Training Epoch: 10/100, batch 79/250 completed (loss: 0.1303:  32%|███▏      | 40/125 [03:59<08:38,  6.11s/it]
Training Epoch: 10/100, batch 79/250 completed (loss: 0.0585:  32%|███▏      | 40/125 [03:59<08:39,  6.11s/it]
Training Epoch: 10/100, batch 79/250 completed (loss: 0.327:  32%|███▏      | 40/125 [03:59<08:39,  6.11s/it] 
Training Epoch: 10/100, batch 80/250 completed (loss: 0.1476:  32%|███▏      | 40/125 [04:02<08:38,  6.11s/it]
Training Epoch: 10/100, batch 80/250 completed (loss: 0.0119:  32%|███▏      | 40/125 [04:02<08:39,  6.11s/it]
Training Epoch: 10/100, batch 80/250 completed (loss: 0.0441:  32%|███▏      | 40/125 [04:02<08:39,  6.11s/it]
Training Epoch: 10/100, batch 80/250 completed (loss: 0.0515:  32%|███▏      | 40/125 [04:02<08:39,  6.11s/it]
Training Epoch: 10/100, batch 80/250 completed (loss: 0.0441:  33%|███▎      | 41/125 [04:05<08:26,  6.03s/it]
Training Epoch: 10/100, batch 80/250 completed (loss: 0.0515:  33%|███▎      | 41/125 [04:05<08:26,  6.04s/it]
Training Epoch: 10/100, batch 80/250 completed (loss: 0.1476:  33%|███▎      | 41/125 [04:05<08:27,  6.04s/it]
Training Epoch: 10/100, batch 80/250 completed (loss: 0.0119:  33%|███▎      | 41/125 [04:05<08:27,  6.04s/it]
Training Epoch: 10/100, batch 81/250 completed (loss: 0.0222:  33%|███▎      | 41/125 [04:05<08:27,  6.04s/it]
Training Epoch: 10/100, batch 81/250 completed (loss: 0.0836:  33%|███▎      | 41/125 [04:05<08:26,  6.03s/it]
Training Epoch: 10/100, batch 81/250 completed (loss: 0.0588:  33%|███▎      | 41/125 [04:05<08:27,  6.04s/it]
Training Epoch: 10/100, batch 81/250 completed (loss: 0.0692:  33%|███▎      | 41/125 [04:05<08:26,  6.04s/it]
Training Epoch: 10/100, batch 82/250 completed (loss: 0.0202:  33%|███▎      | 41/125 [04:08<08:26,  6.03s/it]
Training Epoch: 10/100, batch 82/250 completed (loss: 0.022:  33%|███▎      | 41/125 [04:08<08:27,  6.04s/it] 
Training Epoch: 10/100, batch 82/250 completed (loss: 0.0022:  33%|███▎      | 41/125 [04:08<08:27,  6.04s/it]
Training Epoch: 10/100, batch 82/250 completed (loss: 0.0496:  33%|███▎      | 41/125 [04:08<08:26,  6.04s/it]
Training Epoch: 10/100, batch 82/250 completed (loss: 0.0202:  34%|███▎      | 42/125 [04:11<08:23,  6.07s/it]
Training Epoch: 10/100, batch 82/250 completed (loss: 0.0496:  34%|███▎      | 42/125 [04:11<08:23,  6.07s/it]
Training Epoch: 10/100, batch 82/250 completed (loss: 0.0022:  34%|███▎      | 42/125 [04:11<08:23,  6.07s/it]
Training Epoch: 10/100, batch 82/250 completed (loss: 0.022:  34%|███▎      | 42/125 [04:11<08:23,  6.07s/it]
Training Epoch: 10/100, batch 83/250 completed (loss: 0.0514:  34%|███▎      | 42/125 [04:11<08:23,  6.07s/it]
Training Epoch: 10/100, batch 83/250 completed (loss: 0.1591:  34%|███▎      | 42/125 [04:11<08:23,  6.07s/it]
Training Epoch: 10/100, batch 83/250 completed (loss: 0.0643:  34%|███▎      | 42/125 [04:11<08:23,  6.07s/it]
Training Epoch: 10/100, batch 83/250 completed (loss: 0.0167:  34%|███▎      | 42/125 [04:11<08:23,  6.07s/it]
Training Epoch: 10/100, batch 84/250 completed (loss: 0.0197:  34%|███▎      | 42/125 [04:14<08:23,  6.07s/it]
Training Epoch: 10/100, batch 84/250 completed (loss: 0.0165:  34%|███▎      | 42/125 [04:14<08:23,  6.07s/it]
Training Epoch: 10/100, batch 84/250 completed (loss: 0.0427:  34%|███▎      | 42/125 [04:14<08:23,  6.07s/it]
Training Epoch: 10/100, batch 84/250 completed (loss: 0.0458:  34%|███▎      | 42/125 [04:14<08:23,  6.07s/it]
Training Epoch: 10/100, batch 84/250 completed (loss: 0.0165:  34%|███▍      | 43/125 [04:17<08:20,  6.11s/it]
Training Epoch: 10/100, batch 84/250 completed (loss: 0.0197:  34%|███▍      | 43/125 [04:17<08:20,  6.10s/it]
Training Epoch: 10/100, batch 84/250 completed (loss: 0.0427:  34%|███▍      | 43/125 [04:17<08:20,  6.10s/it]
Training Epoch: 10/100, batch 84/250 completed (loss: 0.0458:  34%|███▍      | 43/125 [04:17<08:20,  6.11s/it]
Training Epoch: 10/100, batch 85/250 completed (loss: 0.2244:  34%|███▍      | 43/125 [04:17<08:20,  6.10s/it]
Training Epoch: 10/100, batch 85/250 completed (loss: 0.0837:  34%|███▍      | 43/125 [04:17<08:20,  6.10s/it]
Training Epoch: 10/100, batch 85/250 completed (loss: 0.0408:  34%|███▍      | 43/125 [04:17<08:20,  6.11s/it]
Training Epoch: 10/100, batch 85/250 completed (loss: 0.0905:  34%|███▍      | 43/125 [04:17<08:20,  6.11s/it]
Training Epoch: 10/100, batch 86/250 completed (loss: 0.2015:  34%|███▍      | 43/125 [04:20<08:20,  6.11s/it]
Training Epoch: 10/100, batch 86/250 completed (loss: 0.0068:  34%|███▍      | 43/125 [04:20<08:20,  6.10s/it]
Training Epoch: 10/100, batch 86/250 completed (loss: 0.0517:  34%|███▍      | 43/125 [04:20<08:20,  6.10s/it]
Training Epoch: 10/100, batch 86/250 completed (loss: 0.0072:  34%|███▍      | 43/125 [04:20<08:20,  6.11s/it]
Training Epoch: 10/100, batch 86/250 completed (loss: 0.0068:  35%|███▌      | 44/125 [04:23<08:08,  6.03s/it]
Training Epoch: 10/100, batch 86/250 completed (loss: 0.2015:  35%|███▌      | 44/125 [04:23<08:09,  6.04s/it]
Training Epoch: 10/100, batch 86/250 completed (loss: 0.0072:  35%|███▌      | 44/125 [04:23<08:09,  6.04s/it]
Training Epoch: 10/100, batch 86/250 completed (loss: 0.0517:  35%|███▌      | 44/125 [04:23<08:09,  6.04s/it]
Training Epoch: 10/100, batch 87/250 completed (loss: 0.1025:  35%|███▌      | 44/125 [04:23<08:09,  6.04s/it]
Training Epoch: 10/100, batch 87/250 completed (loss: 0.0548:  35%|███▌      | 44/125 [04:23<08:08,  6.03s/it]
Training Epoch: 10/100, batch 87/250 completed (loss: 0.0692:  35%|███▌      | 44/125 [04:23<08:09,  6.04s/it]
Training Epoch: 10/100, batch 87/250 completed (loss: 0.0391:  35%|███▌      | 44/125 [04:23<08:09,  6.04s/it]
Training Epoch: 10/100, batch 88/250 completed (loss: 0.0473:  35%|███▌      | 44/125 [04:26<08:09,  6.04s/it]
Training Epoch: 10/100, batch 88/250 completed (loss: 0.1321:  35%|███▌      | 44/125 [04:26<08:08,  6.03s/it]
Training Epoch: 10/100, batch 88/250 completed (loss: 0.0183:  35%|███▌      | 44/125 [04:26<08:09,  6.04s/it]
Training Epoch: 10/100, batch 88/250 completed (loss: 0.2448:  35%|███▌      | 44/125 [04:27<08:09,  6.04s/it]
Training Epoch: 10/100, batch 88/250 completed (loss: 0.0183:  36%|███▌      | 45/125 [04:30<08:13,  6.17s/it]
Training Epoch: 10/100, batch 88/250 completed (loss: 0.0473:  36%|███▌      | 45/125 [04:30<08:13,  6.17s/it]
Training Epoch: 10/100, batch 88/250 completed (loss: 0.2448:  36%|███▌      | 45/125 [04:30<08:13,  6.17s/it]
Training Epoch: 10/100, batch 88/250 completed (loss: 0.1321:  36%|███▌      | 45/125 [04:30<08:13,  6.17s/it]
Training Epoch: 10/100, batch 89/250 completed (loss: 0.0324:  36%|███▌      | 45/125 [04:30<08:13,  6.17s/it]
Training Epoch: 10/100, batch 89/250 completed (loss: 0.0622:  36%|███▌      | 45/125 [04:30<08:13,  6.17s/it]
Training Epoch: 10/100, batch 89/250 completed (loss: 0.0113:  36%|███▌      | 45/125 [04:30<08:13,  6.17s/it]
Training Epoch: 10/100, batch 89/250 completed (loss: 0.0878:  36%|███▌      | 45/125 [04:30<08:13,  6.17s/it]
Training Epoch: 10/100, batch 90/250 completed (loss: 0.0422:  36%|███▌      | 45/125 [04:33<08:13,  6.17s/it]
Training Epoch: 10/100, batch 90/250 completed (loss: 0.0511:  36%|███▌      | 45/125 [04:33<08:13,  6.17s/it]
Training Epoch: 10/100, batch 90/250 completed (loss: 0.0234:  36%|███▌      | 45/125 [04:33<08:13,  6.17s/it]
Training Epoch: 10/100, batch 90/250 completed (loss: 0.1337:  36%|███▌      | 45/125 [04:33<08:13,  6.17s/it]
Training Epoch: 10/100, batch 90/250 completed (loss: 0.0234:  37%|███▋      | 46/125 [04:36<07:59,  6.07s/it]
Training Epoch: 10/100, batch 90/250 completed (loss: 0.1337:  37%|███▋      | 46/125 [04:36<07:59,  6.07s/it]
Training Epoch: 10/100, batch 90/250 completed (loss: 0.0511:  37%|███▋      | 46/125 [04:36<07:59,  6.08s/it]
Training Epoch: 10/100, batch 90/250 completed (loss: 0.0422:  37%|███▋      | 46/125 [04:36<07:59,  6.08s/it]
Training Epoch: 10/100, batch 91/250 completed (loss: 0.111:  37%|███▋      | 46/125 [04:36<07:59,  6.08s/it] 
Training Epoch: 10/100, batch 91/250 completed (loss: 0.0402:  37%|███▋      | 46/125 [04:36<07:59,  6.07s/it]
Training Epoch: 10/100, batch 91/250 completed (loss: 0.0399:  37%|███▋      | 46/125 [04:36<07:59,  6.08s/it]
Training Epoch: 10/100, batch 91/250 completed (loss: 0.0712:  37%|███▋      | 46/125 [04:36<07:59,  6.07s/it]
Training Epoch: 10/100, batch 92/250 completed (loss: 0.0089:  37%|███▋      | 46/125 [04:39<07:59,  6.07s/it]
Training Epoch: 10/100, batch 92/250 completed (loss: 0.0807:  37%|███▋      | 46/125 [04:39<07:59,  6.08s/it]
Training Epoch: 10/100, batch 92/250 completed (loss: 0.0195:  37%|███▋      | 46/125 [04:39<07:59,  6.08s/it]
Training Epoch: 10/100, batch 92/250 completed (loss: 0.0695:  37%|███▋      | 46/125 [04:39<07:59,  6.07s/it]
Training Epoch: 10/100, batch 92/250 completed (loss: 0.0195:  38%|███▊      | 47/125 [04:42<07:51,  6.05s/it]
Training Epoch: 10/100, batch 92/250 completed (loss: 0.0089:  38%|███▊      | 47/125 [04:42<07:52,  6.05s/it]
Training Epoch: 10/100, batch 92/250 completed (loss: 0.0695:  38%|███▊      | 47/125 [04:42<07:51,  6.05s/it]
Training Epoch: 10/100, batch 92/250 completed (loss: 0.0807:  38%|███▊      | 47/125 [04:42<07:51,  6.05s/it]
Training Epoch: 10/100, batch 93/250 completed (loss: 0.0378:  38%|███▊      | 47/125 [04:42<07:51,  6.05s/it]
Training Epoch: 10/100, batch 93/250 completed (loss: 0.0791:  38%|███▊      | 47/125 [04:42<07:51,  6.05s/it]
Training Epoch: 10/100, batch 93/250 completed (loss: 0.0301:  38%|███▊      | 47/125 [04:42<07:52,  6.05s/it]
Training Epoch: 10/100, batch 93/250 completed (loss: 0.0452:  38%|███▊      | 47/125 [04:42<07:51,  6.05s/it]
Training Epoch: 10/100, batch 94/250 completed (loss: 0.0576:  38%|███▊      | 47/125 [04:45<07:51,  6.05s/it]
Training Epoch: 10/100, batch 94/250 completed (loss: 0.0296:  38%|███▊      | 47/125 [04:45<07:51,  6.05s/it]
Training Epoch: 10/100, batch 94/250 completed (loss: 0.1428:  38%|███▊      | 47/125 [04:45<07:52,  6.05s/it]
Training Epoch: 10/100, batch 94/250 completed (loss: 0.0327:  38%|███▊      | 47/125 [04:45<07:51,  6.05s/it]
Training Epoch: 10/100, batch 94/250 completed (loss: 0.1428:  38%|███▊      | 48/125 [04:48<07:49,  6.10s/it]
Training Epoch: 10/100, batch 94/250 completed (loss: 0.0327:  38%|███▊      | 48/125 [04:48<07:49,  6.10s/it]
Training Epoch: 10/100, batch 94/250 completed (loss: 0.0296:  38%|███▊      | 48/125 [04:48<07:50,  6.10s/it]
Training Epoch: 10/100, batch 94/250 completed (loss: 0.0576:  38%|███▊      | 48/125 [04:48<07:49,  6.10s/it]
Training Epoch: 10/100, batch 95/250 completed (loss: 0.0047:  38%|███▊      | 48/125 [04:48<07:49,  6.10s/it]
Training Epoch: 10/100, batch 95/250 completed (loss: 0.0067:  38%|███▊      | 48/125 [04:48<07:49,  6.10s/it]
Training Epoch: 10/100, batch 95/250 completed (loss: 0.0617:  38%|███▊      | 48/125 [04:48<07:50,  6.10s/it]
Training Epoch: 10/100, batch 95/250 completed (loss: 0.0953:  38%|███▊      | 48/125 [04:48<07:49,  6.10s/it]
Training Epoch: 10/100, batch 96/250 completed (loss: 0.0359:  38%|███▊      | 48/125 [04:51<07:49,  6.10s/it]
Training Epoch: 10/100, batch 96/250 completed (loss: 0.0399:  38%|███▊      | 48/125 [04:51<07:50,  6.10s/it]
Training Epoch: 10/100, batch 96/250 completed (loss: 0.1345:  38%|███▊      | 48/125 [04:51<07:49,  6.10s/it]
Training Epoch: 10/100, batch 96/250 completed (loss: 0.025:  38%|███▊      | 48/125 [04:51<07:49,  6.10s/it] 
Training Epoch: 10/100, batch 96/250 completed (loss: 0.1345:  39%|███▉      | 49/125 [04:54<07:38,  6.03s/it]
Training Epoch: 10/100, batch 96/250 completed (loss: 0.025:  39%|███▉      | 49/125 [04:54<07:38,  6.03s/it]
Training Epoch: 10/100, batch 96/250 completed (loss: 0.0359:  39%|███▉      | 49/125 [04:54<07:38,  6.03s/it]
Training Epoch: 10/100, batch 96/250 completed (loss: 0.0399:  39%|███▉      | 49/125 [04:54<07:38,  6.03s/it]
Training Epoch: 10/100, batch 97/250 completed (loss: 0.0372:  39%|███▉      | 49/125 [04:54<07:38,  6.03s/it]
Training Epoch: 10/100, batch 97/250 completed (loss: 0.0429:  39%|███▉      | 49/125 [04:54<07:38,  6.03s/it]
Training Epoch: 10/100, batch 97/250 completed (loss: 0.0341:  39%|███▉      | 49/125 [04:54<07:38,  6.03s/it]
Training Epoch: 10/100, batch 97/250 completed (loss: 0.0219:  39%|███▉      | 49/125 [04:54<07:38,  6.03s/it]
Training Epoch: 10/100, batch 98/250 completed (loss: 0.011:  39%|███▉      | 49/125 [04:57<07:38,  6.03s/it] 
Training Epoch: 10/100, batch 98/250 completed (loss: 0.0126:  39%|███▉      | 49/125 [04:57<07:38,  6.03s/it]
Training Epoch: 10/100, batch 98/250 completed (loss: 0.0183:  39%|███▉      | 49/125 [04:57<07:38,  6.03s/it]
Training Epoch: 10/100, batch 98/250 completed (loss: 0.028:  39%|███▉      | 49/125 [04:57<07:38,  6.03s/it] 
Training Epoch: 10/100, batch 98/250 completed (loss: 0.0183:  40%|████      | 50/125 [05:00<07:38,  6.11s/it]
Training Epoch: 10/100, batch 98/250 completed (loss: 0.028:  40%|████      | 50/125 [05:00<07:38,  6.11s/it]
Training Epoch: 10/100, batch 98/250 completed (loss: 0.011:  40%|████      | 50/125 [05:00<07:38,  6.11s/it]
Training Epoch: 10/100, batch 98/250 completed (loss: 0.0126:  40%|████      | 50/125 [05:00<07:39,  6.12s/it]
Training Epoch: 10/100, batch 99/250 completed (loss: 0.1244:  40%|████      | 50/125 [05:00<07:39,  6.12s/it]
Training Epoch: 10/100, batch 99/250 completed (loss: 0.1771:  40%|████      | 50/125 [05:00<07:38,  6.11s/it]
Training Epoch: 10/100, batch 99/250 completed (loss: 0.0135:  40%|████      | 50/125 [05:00<07:38,  6.11s/it]
Training Epoch: 10/100, batch 99/250 completed (loss: 0.0228:  40%|████      | 50/125 [05:00<07:38,  6.11s/it]
Training Epoch: 10/100, batch 100/250 completed (loss: 0.0037:  40%|████      | 50/125 [05:03<07:38,  6.11s/it]
Training Epoch: 10/100, batch 100/250 completed (loss: 0.049:  40%|████      | 50/125 [05:03<07:38,  6.11s/it]
Training Epoch: 10/100, batch 100/250 completed (loss: 0.0456:  40%|████      | 50/125 [05:03<07:39,  6.12s/it]
Training Epoch: 10/100, batch 100/250 completed (loss: 0.0084:  40%|████      | 50/125 [05:03<07:38,  6.11s/it]
Training Epoch: 10/100, batch 100/250 completed (loss: 0.049:  41%|████      | 51/125 [05:06<07:30,  6.09s/it]
Training Epoch: 10/100, batch 100/250 completed (loss: 0.0456:  41%|████      | 51/125 [05:06<07:30,  6.09s/it]
Training Epoch: 10/100, batch 100/250 completed (loss: 0.0084:  41%|████      | 51/125 [05:06<07:30,  6.09s/it]
Training Epoch: 10/100, batch 100/250 completed (loss: 0.0037:  41%|████      | 51/125 [05:06<07:30,  6.09s/it]
Training Epoch: 10/100, batch 101/250 completed (loss: 0.1464:  41%|████      | 51/125 [05:06<07:30,  6.09s/it]
Training Epoch: 10/100, batch 101/250 completed (loss: 0.0135:  41%|████      | 51/125 [05:06<07:30,  6.09s/it]
Training Epoch: 10/100, batch 101/250 completed (loss: 0.0125:  41%|████      | 51/125 [05:06<07:30,  6.09s/it]
Training Epoch: 10/100, batch 101/250 completed (loss: 0.0894:  41%|████      | 51/125 [05:06<07:30,  6.09s/it]
Training Epoch: 10/100, batch 102/250 completed (loss: 0.0116:  41%|████      | 51/125 [05:09<07:30,  6.09s/it]
Training Epoch: 10/100, batch 102/250 completed (loss: 0.051:  41%|████      | 51/125 [05:09<07:30,  6.09s/it] 
Training Epoch: 10/100, batch 102/250 completed (loss: 0.1159:  41%|████      | 51/125 [05:09<07:30,  6.09s/it]
Training Epoch: 10/100, batch 102/250 completed (loss: 0.0236:  41%|████      | 51/125 [05:09<07:30,  6.09s/it]
Training Epoch: 10/100, batch 102/250 completed (loss: 0.0236:  42%|████▏     | 52/125 [05:12<07:23,  6.07s/it]
Training Epoch: 10/100, batch 102/250 completed (loss: 0.1159:  42%|████▏     | 52/125 [05:12<07:23,  6.07s/it]
Training Epoch: 10/100, batch 102/250 completed (loss: 0.0116:  42%|████▏     | 52/125 [05:12<07:23,  6.07s/it]
Training Epoch: 10/100, batch 102/250 completed (loss: 0.051:  42%|████▏     | 52/125 [05:12<07:23,  6.08s/it]
Training Epoch: 10/100, batch 103/250 completed (loss: 0.3013:  42%|████▏     | 52/125 [05:12<07:23,  6.08s/it]
Training Epoch: 10/100, batch 103/250 completed (loss: 0.1221:  42%|████▏     | 52/125 [05:12<07:23,  6.07s/it]
Training Epoch: 10/100, batch 103/250 completed (loss: 0.0826:  42%|████▏     | 52/125 [05:12<07:23,  6.07s/it]
Training Epoch: 10/100, batch 103/250 completed (loss: 0.0647:  42%|████▏     | 52/125 [05:12<07:23,  6.07s/it]
Training Epoch: 10/100, batch 104/250 completed (loss: 0.1808:  42%|████▏     | 52/125 [05:15<07:23,  6.07s/it]
Training Epoch: 10/100, batch 104/250 completed (loss: 0.087:  42%|████▏     | 52/125 [05:15<07:23,  6.07s/it] 
Training Epoch: 10/100, batch 104/250 completed (loss: 0.0156:  42%|████▏     | 52/125 [05:15<07:23,  6.08s/it]
Training Epoch: 10/100, batch 104/250 completed (loss: 0.0763:  42%|████▏     | 52/125 [05:15<07:23,  6.07s/it]
Training Epoch: 10/100, batch 104/250 completed (loss: 0.087:  42%|████▏     | 53/125 [05:18<07:22,  6.15s/it]
Training Epoch: 10/100, batch 104/250 completed (loss: 0.0763:  42%|████▏     | 53/125 [05:18<07:23,  6.15s/it]
Training Epoch: 10/100, batch 104/250 completed (loss: 0.1808:  42%|████▏     | 53/125 [05:18<07:22,  6.15s/it]
Training Epoch: 10/100, batch 104/250 completed (loss: 0.0156:  42%|████▏     | 53/125 [05:18<07:23,  6.16s/it]
Training Epoch: 10/100, batch 105/250 completed (loss: 0.0512:  42%|████▏     | 53/125 [05:18<07:22,  6.15s/it]
Training Epoch: 10/100, batch 105/250 completed (loss: 0.0424:  42%|████▏     | 53/125 [05:18<07:23,  6.16s/it]
Training Epoch: 10/100, batch 105/250 completed (loss: 0.1334:  42%|████▏     | 53/125 [05:18<07:22,  6.15s/it]
Training Epoch: 10/100, batch 105/250 completed (loss: 0.0581:  42%|████▏     | 53/125 [05:18<07:23,  6.15s/it]
Training Epoch: 10/100, batch 106/250 completed (loss: 0.0984:  42%|████▏     | 53/125 [05:21<07:22,  6.15s/it]
Training Epoch: 10/100, batch 106/250 completed (loss: 0.0917:  42%|████▏     | 53/125 [05:21<07:23,  6.16s/it]
Training Epoch: 10/100, batch 106/250 completed (loss: 0.1371:  42%|████▏     | 53/125 [05:21<07:22,  6.15s/it]
Training Epoch: 10/100, batch 106/250 completed (loss: 0.0349:  42%|████▏     | 53/125 [05:21<07:23,  6.15s/it]
Training Epoch: 10/100, batch 106/250 completed (loss: 0.0917:  43%|████▎     | 54/125 [05:24<07:10,  6.07s/it]
Training Epoch: 10/100, batch 106/250 completed (loss: 0.1371:  43%|████▎     | 54/125 [05:24<07:10,  6.07s/it]
Training Epoch: 10/100, batch 106/250 completed (loss: 0.0349:  43%|████▎     | 54/125 [05:24<07:11,  6.07s/it]
Training Epoch: 10/100, batch 106/250 completed (loss: 0.0984:  43%|████▎     | 54/125 [05:24<07:10,  6.07s/it]
Training Epoch: 10/100, batch 107/250 completed (loss: 0.0406:  43%|████▎     | 54/125 [05:24<07:10,  6.07s/it]
Training Epoch: 10/100, batch 107/250 completed (loss: 0.0197:  43%|████▎     | 54/125 [05:24<07:10,  6.07s/it]
Training Epoch: 10/100, batch 107/250 completed (loss: 0.02:  43%|████▎     | 54/125 [05:24<07:10,  6.07s/it]  
Training Epoch: 10/100, batch 107/250 completed (loss: 0.1849:  43%|████▎     | 54/125 [05:24<07:11,  6.07s/it]
Training Epoch: 10/100, batch 108/250 completed (loss: 0.0635:  43%|████▎     | 54/125 [05:27<07:10,  6.07s/it]
Training Epoch: 10/100, batch 108/250 completed (loss: 0.0697:  43%|████▎     | 54/125 [05:27<07:10,  6.07s/it]
Training Epoch: 10/100, batch 108/250 completed (loss: 0.0437:  43%|████▎     | 54/125 [05:27<07:10,  6.07s/it]
Training Epoch: 10/100, batch 108/250 completed (loss: 0.0217:  43%|████▎     | 54/125 [05:27<07:11,  6.07s/it]
Training Epoch: 10/100, batch 108/250 completed (loss: 0.0217:  44%|████▍     | 55/125 [05:31<07:10,  6.15s/it]
Training Epoch: 10/100, batch 108/250 completed (loss: 0.0635:  44%|████▍     | 55/125 [05:31<07:10,  6.15s/it]
Training Epoch: 10/100, batch 108/250 completed (loss: 0.0437:  44%|████▍     | 55/125 [05:31<07:10,  6.15s/it]
Training Epoch: 10/100, batch 108/250 completed (loss: 0.0697:  44%|████▍     | 55/125 [05:31<07:10,  6.15s/it]
Training Epoch: 10/100, batch 109/250 completed (loss: 0.0093:  44%|████▍     | 55/125 [05:31<07:10,  6.15s/it]
Training Epoch: 10/100, batch 109/250 completed (loss: 0.2037:  44%|████▍     | 55/125 [05:31<07:10,  6.15s/it]
Training Epoch: 10/100, batch 109/250 completed (loss: 0.0061:  44%|████▍     | 55/125 [05:31<07:10,  6.15s/it]
Training Epoch: 10/100, batch 109/250 completed (loss: 0.0176:  44%|████▍     | 55/125 [05:31<07:10,  6.15s/it]
Training Epoch: 10/100, batch 110/250 completed (loss: 0.0963:  44%|████▍     | 55/125 [05:34<07:10,  6.15s/it]
Training Epoch: 10/100, batch 110/250 completed (loss: 0.0339:  44%|████▍     | 55/125 [05:34<07:10,  6.15s/it]
Training Epoch: 10/100, batch 110/250 completed (loss: 0.0081:  44%|████▍     | 55/125 [05:34<07:10,  6.15s/it]
Training Epoch: 10/100, batch 110/250 completed (loss: 0.0322:  44%|████▍     | 55/125 [05:34<07:10,  6.15s/it]
Training Epoch: 10/100, batch 110/250 completed (loss: 0.0322:  45%|████▍     | 56/125 [05:37<07:01,  6.11s/it]
Training Epoch: 10/100, batch 110/250 completed (loss: 0.0339:  45%|████▍     | 56/125 [05:37<07:01,  6.11s/it]
Training Epoch: 10/100, batch 110/250 completed (loss: 0.0963:  45%|████▍     | 56/125 [05:37<07:01,  6.11s/it]
Training Epoch: 10/100, batch 110/250 completed (loss: 0.0081:  45%|████▍     | 56/125 [05:37<07:01,  6.11s/it]
Training Epoch: 10/100, batch 111/250 completed (loss: 0.0803:  45%|████▍     | 56/125 [05:37<07:01,  6.11s/it]
Training Epoch: 10/100, batch 111/250 completed (loss: 0.132:  45%|████▍     | 56/125 [05:37<07:01,  6.11s/it] 
Training Epoch: 10/100, batch 111/250 completed (loss: 0.1941:  45%|████▍     | 56/125 [05:37<07:01,  6.11s/it]
Training Epoch: 10/100, batch 111/250 completed (loss: 0.1309:  45%|████▍     | 56/125 [05:37<07:01,  6.11s/it]
Training Epoch: 10/100, batch 112/250 completed (loss: 0.013:  45%|████▍     | 56/125 [05:40<07:01,  6.11s/it] 
Training Epoch: 10/100, batch 112/250 completed (loss: 0.0293:  45%|████▍     | 56/125 [05:40<07:01,  6.11s/it]
Training Epoch: 10/100, batch 112/250 completed (loss: 0.0427:  45%|████▍     | 56/125 [05:40<07:01,  6.11s/it]
Training Epoch: 10/100, batch 112/250 completed (loss: 0.0112:  45%|████▍     | 56/125 [05:40<07:01,  6.11s/it]
Training Epoch: 10/100, batch 112/250 completed (loss: 0.0112:  46%|████▌     | 57/125 [05:43<06:53,  6.08s/it]
Training Epoch: 10/100, batch 112/250 completed (loss: 0.0293:  46%|████▌     | 57/125 [05:43<06:53,  6.08s/it]
Training Epoch: 10/100, batch 112/250 completed (loss: 0.013:  46%|████▌     | 57/125 [05:43<06:53,  6.08s/it]
Training Epoch: 10/100, batch 112/250 completed (loss: 0.0427:  46%|████▌     | 57/125 [05:43<06:53,  6.08s/it]
Training Epoch: 10/100, batch 113/250 completed (loss: 0.0197:  46%|████▌     | 57/125 [05:43<06:53,  6.08s/it]
Training Epoch: 10/100, batch 113/250 completed (loss: 0.0236:  46%|████▌     | 57/125 [05:43<06:53,  6.08s/it]
Training Epoch: 10/100, batch 113/250 completed (loss: 0.0965:  46%|████▌     | 57/125 [05:43<06:53,  6.08s/it]
Training Epoch: 10/100, batch 113/250 completed (loss: 0.1078:  46%|████▌     | 57/125 [05:43<06:53,  6.08s/it]
Training Epoch: 10/100, batch 114/250 completed (loss: 0.0526:  46%|████▌     | 57/125 [05:46<06:53,  6.08s/it]
Training Epoch: 10/100, batch 114/250 completed (loss: 0.0998:  46%|████▌     | 57/125 [05:46<06:53,  6.08s/it]
Training Epoch: 10/100, batch 114/250 completed (loss: 0.098:  46%|████▌     | 57/125 [05:46<06:53,  6.08s/it] 
Training Epoch: 10/100, batch 114/250 completed (loss: 0.1505:  46%|████▌     | 57/125 [05:46<06:53,  6.08s/it]
Training Epoch: 10/100, batch 114/250 completed (loss: 0.098:  46%|████▋     | 58/125 [05:49<06:49,  6.11s/it]
Training Epoch: 10/100, batch 114/250 completed (loss: 0.1505:  46%|████▋     | 58/125 [05:49<06:49,  6.11s/it]
Training Epoch: 10/100, batch 114/250 completed (loss: 0.0526:  46%|████▋     | 58/125 [05:49<06:49,  6.12s/it]
Training Epoch: 10/100, batch 114/250 completed (loss: 0.0998:  46%|████▋     | 58/125 [05:49<06:50,  6.12s/it]
Training Epoch: 10/100, batch 115/250 completed (loss: 0.1051:  46%|████▋     | 58/125 [05:49<06:50,  6.12s/it]
Training Epoch: 10/100, batch 115/250 completed (loss: 0.0228:  46%|████▋     | 58/125 [05:49<06:49,  6.12s/it]
Training Epoch: 10/100, batch 115/250 completed (loss: 0.0046:  46%|████▋     | 58/125 [05:49<06:49,  6.11s/it]
Training Epoch: 10/100, batch 115/250 completed (loss: 0.0493:  46%|████▋     | 58/125 [05:49<06:49,  6.11s/it]
Training Epoch: 10/100, batch 116/250 completed (loss: 0.2639:  46%|████▋     | 58/125 [05:52<06:50,  6.12s/it]
Training Epoch: 10/100, batch 116/250 completed (loss: 0.0071:  46%|████▋     | 58/125 [05:52<06:49,  6.12s/it]
Training Epoch: 10/100, batch 116/250 completed (loss: 0.0033:  46%|████▋     | 58/125 [05:52<06:49,  6.11s/it]
Training Epoch: 10/100, batch 116/250 completed (loss: 0.0505:  46%|████▋     | 58/125 [05:52<06:49,  6.11s/it]
Training Epoch: 10/100, batch 116/250 completed (loss: 0.0033:  47%|████▋     | 59/125 [05:55<06:41,  6.09s/it]
Training Epoch: 10/100, batch 116/250 completed (loss: 0.0071:  47%|████▋     | 59/125 [05:55<06:41,  6.09s/it]
Training Epoch: 10/100, batch 116/250 completed (loss: 0.2639:  47%|████▋     | 59/125 [05:55<06:41,  6.09s/it]
Training Epoch: 10/100, batch 116/250 completed (loss: 0.0505:  47%|████▋     | 59/125 [05:55<06:41,  6.09s/it]
Training Epoch: 10/100, batch 117/250 completed (loss: 0.0414:  47%|████▋     | 59/125 [05:55<06:41,  6.09s/it]
Training Epoch: 10/100, batch 117/250 completed (loss: 0.0215:  47%|████▋     | 59/125 [05:55<06:41,  6.09s/it]
Training Epoch: 10/100, batch 117/250 completed (loss: 0.0174:  47%|████▋     | 59/125 [05:55<06:41,  6.09s/it]
Training Epoch: 10/100, batch 117/250 completed (loss: 0.0339:  47%|████▋     | 59/125 [05:55<06:41,  6.09s/it]
Training Epoch: 10/100, batch 118/250 completed (loss: 0.0378:  47%|████▋     | 59/125 [05:58<06:41,  6.09s/it]
Training Epoch: 10/100, batch 118/250 completed (loss: 0.0085:  47%|████▋     | 59/125 [05:58<06:41,  6.09s/it]
Training Epoch: 10/100, batch 118/250 completed (loss: 0.0175:  47%|████▋     | 59/125 [05:58<06:41,  6.09s/it]
Training Epoch: 10/100, batch 118/250 completed (loss: 0.029:  47%|████▋     | 59/125 [05:58<06:41,  6.09s/it] 
Training Epoch: 10/100, batch 118/250 completed (loss: 0.0175:  48%|████▊     | 60/125 [06:01<06:34,  6.06s/it]
Training Epoch: 10/100, batch 118/250 completed (loss: 0.029:  48%|████▊     | 60/125 [06:01<06:34,  6.07s/it]
Training Epoch: 10/100, batch 118/250 completed (loss: 0.0085:  48%|████▊     | 60/125 [06:01<06:34,  6.06s/it]
Training Epoch: 10/100, batch 118/250 completed (loss: 0.0378:  48%|████▊     | 60/125 [06:01<06:34,  6.07s/it]
Training Epoch: 10/100, batch 119/250 completed (loss: 0.011:  48%|████▊     | 60/125 [06:01<06:34,  6.07s/it] 
Training Epoch: 10/100, batch 119/250 completed (loss: 0.0637:  48%|████▊     | 60/125 [06:01<06:34,  6.06s/it]
Training Epoch: 10/100, batch 119/250 completed (loss: 0.2726:  48%|████▊     | 60/125 [06:01<06:34,  6.06s/it]
Training Epoch: 10/100, batch 119/250 completed (loss: 0.01:  48%|████▊     | 60/125 [06:01<06:34,  6.07s/it] 
Training Epoch: 10/100, batch 120/250 completed (loss: 0.0193:  48%|████▊     | 60/125 [06:04<06:34,  6.07s/it]
Training Epoch: 10/100, batch 120/250 completed (loss: 0.0346:  48%|████▊     | 60/125 [06:04<06:34,  6.06s/it]
Training Epoch: 10/100, batch 120/250 completed (loss: 0.0431:  48%|████▊     | 60/125 [06:04<06:34,  6.06s/it]
Training Epoch: 10/100, batch 120/250 completed (loss: 0.016:  48%|████▊     | 60/125 [06:04<06:34,  6.07s/it]
Training Epoch: 10/100, batch 120/250 completed (loss: 0.0431:  49%|████▉     | 61/125 [06:07<06:29,  6.09s/it]
Training Epoch: 10/100, batch 120/250 completed (loss: 0.016:  49%|████▉     | 61/125 [06:07<06:29,  6.09s/it]
Training Epoch: 10/100, batch 120/250 completed (loss: 0.0193:  49%|████▉     | 61/125 [06:07<06:29,  6.09s/it]
Training Epoch: 10/100, batch 120/250 completed (loss: 0.0346:  49%|████▉     | 61/125 [06:07<06:29,  6.09s/it]
Training Epoch: 10/100, batch 121/250 completed (loss: 0.0386:  49%|████▉     | 61/125 [06:07<06:29,  6.09s/it]
Training Epoch: 10/100, batch 121/250 completed (loss: 0.0209:  49%|████▉     | 61/125 [06:07<06:29,  6.09s/it]
Training Epoch: 10/100, batch 121/250 completed (loss: 0.0052:  49%|████▉     | 61/125 [06:07<06:29,  6.09s/it]
Training Epoch: 10/100, batch 121/250 completed (loss: 0.006:  49%|████▉     | 61/125 [06:07<06:29,  6.09s/it]
Training Epoch: 10/100, batch 122/250 completed (loss: 0.1203:  49%|████▉     | 61/125 [06:10<06:29,  6.09s/it]
Training Epoch: 10/100, batch 122/250 completed (loss: 0.1331:  49%|████▉     | 61/125 [06:10<06:29,  6.09s/it]
Training Epoch: 10/100, batch 122/250 completed (loss: 0.0266:  49%|████▉     | 61/125 [06:10<06:29,  6.09s/it]
Training Epoch: 10/100, batch 122/250 completed (loss: 0.0269:  49%|████▉     | 61/125 [06:10<06:29,  6.09s/it]
Training Epoch: 10/100, batch 122/250 completed (loss: 0.1203:  50%|████▉     | 62/125 [06:13<06:25,  6.13s/it]
Training Epoch: 10/100, batch 122/250 completed (loss: 0.0269:  50%|████▉     | 62/125 [06:13<06:26,  6.13s/it]
Training Epoch: 10/100, batch 122/250 completed (loss: 0.0266:  50%|████▉     | 62/125 [06:13<06:26,  6.13s/it]
Training Epoch: 10/100, batch 122/250 completed (loss: 0.1331:  50%|████▉     | 62/125 [06:13<06:26,  6.13s/it]
Training Epoch: 10/100, batch 123/250 completed (loss: 0.0307:  50%|████▉     | 62/125 [06:13<06:26,  6.13s/it]
Training Epoch: 10/100, batch 123/250 completed (loss: 0.2194:  50%|████▉     | 62/125 [06:13<06:25,  6.13s/it]
Training Epoch: 10/100, batch 123/250 completed (loss: 0.0749:  50%|████▉     | 62/125 [06:13<06:26,  6.13s/it]
Training Epoch: 10/100, batch 123/250 completed (loss: 0.02:  50%|████▉     | 62/125 [06:13<06:26,  6.13s/it]  
Training Epoch: 10/100, batch 124/250 completed (loss: 0.0492:  50%|████▉     | 62/125 [06:16<06:25,  6.13s/it]
Training Epoch: 10/100, batch 124/250 completed (loss: 0.0488:  50%|████▉     | 62/125 [06:16<06:26,  6.13s/it]
Training Epoch: 10/100, batch 124/250 completed (loss: 0.0414:  50%|████▉     | 62/125 [06:16<06:26,  6.13s/it]
Training Epoch: 10/100, batch 124/250 completed (loss: 0.0513:  50%|████▉     | 62/125 [06:16<06:26,  6.13s/it]
Training Epoch: 10/100, batch 124/250 completed (loss: 0.0513:  50%|█████     | 63/125 [06:19<06:20,  6.14s/it]
Training Epoch: 10/100, batch 124/250 completed (loss: 0.0492:  50%|█████     | 63/125 [06:19<06:20,  6.14s/it]
Training Epoch: 10/100, batch 124/250 completed (loss: 0.0414:  50%|█████     | 63/125 [06:19<06:20,  6.14s/it]
Training Epoch: 10/100, batch 124/250 completed (loss: 0.0488:  50%|█████     | 63/125 [06:19<06:20,  6.14s/it]
Training Epoch: 10/100, batch 125/250 completed (loss: 0.0462:  50%|█████     | 63/125 [06:19<06:20,  6.14s/it]
Training Epoch: 10/100, batch 125/250 completed (loss: 0.0599:  50%|█████     | 63/125 [06:19<06:20,  6.14s/it]
Training Epoch: 10/100, batch 125/250 completed (loss: 0.0111:  50%|█████     | 63/125 [06:19<06:20,  6.14s/it]
Training Epoch: 10/100, batch 125/250 completed (loss: 0.0508:  50%|█████     | 63/125 [06:19<06:20,  6.14s/it]
Training Epoch: 10/100, batch 126/250 completed (loss: 0.0102:  50%|█████     | 63/125 [06:22<06:20,  6.14s/it]
Training Epoch: 10/100, batch 126/250 completed (loss: 0.0884:  50%|█████     | 63/125 [06:22<06:20,  6.14s/it]
Training Epoch: 10/100, batch 126/250 completed (loss: 0.0195:  50%|█████     | 63/125 [06:22<06:20,  6.14s/it]
Training Epoch: 10/100, batch 126/250 completed (loss: 0.1339:  50%|█████     | 63/125 [06:22<06:20,  6.14s/it]
Training Epoch: 10/100, batch 126/250 completed (loss: 0.0195:  51%|█████     | 64/125 [06:25<06:09,  6.06s/it]
Training Epoch: 10/100, batch 126/250 completed (loss: 0.0102:  51%|█████     | 64/125 [06:25<06:09,  6.06s/it]
Training Epoch: 10/100, batch 126/250 completed (loss: 0.1339:  51%|█████     | 64/125 [06:25<06:10,  6.07s/it]
Training Epoch: 10/100, batch 126/250 completed (loss: 0.0884:  51%|█████     | 64/125 [06:25<06:09,  6.06s/it]
Training Epoch: 10/100, batch 127/250 completed (loss: 0.0162:  51%|█████     | 64/125 [06:25<06:09,  6.06s/it]
Training Epoch: 10/100, batch 127/250 completed (loss: 0.0236:  51%|█████     | 64/125 [06:25<06:09,  6.06s/it]
Training Epoch: 10/100, batch 127/250 completed (loss: 0.0276:  51%|█████     | 64/125 [06:25<06:09,  6.06s/it]
Training Epoch: 10/100, batch 127/250 completed (loss: 0.0142:  51%|█████     | 64/125 [06:25<06:10,  6.07s/it]
Training Epoch: 10/100, batch 128/250 completed (loss: 0.0317:  51%|█████     | 64/125 [06:29<06:09,  6.06s/it]
Training Epoch: 10/100, batch 128/250 completed (loss: 0.0262:  51%|█████     | 64/125 [06:29<06:09,  6.06s/it]
Training Epoch: 10/100, batch 128/250 completed (loss: 0.042:  51%|█████     | 64/125 [06:29<06:09,  6.06s/it] 
Training Epoch: 10/100, batch 128/250 completed (loss: 0.185:  51%|█████     | 64/125 [06:29<06:10,  6.07s/it] 
Training Epoch: 10/100, batch 128/250 completed (loss: 0.042:  52%|█████▏    | 65/125 [06:31<06:06,  6.10s/it]
Training Epoch: 10/100, batch 128/250 completed (loss: 0.185:  52%|█████▏    | 65/125 [06:31<06:06,  6.10s/it]
Training Epoch: 10/100, batch 128/250 completed (loss: 0.0317:  52%|█████▏    | 65/125 [06:31<06:06,  6.10s/it]
Training Epoch: 10/100, batch 128/250 completed (loss: 0.0262:  52%|█████▏    | 65/125 [06:31<06:06,  6.10s/it]
Training Epoch: 10/100, batch 129/250 completed (loss: 0.0591:  52%|█████▏    | 65/125 [06:31<06:06,  6.10s/it]
Training Epoch: 10/100, batch 129/250 completed (loss: 0.0143:  52%|█████▏    | 65/125 [06:31<06:06,  6.10s/it]
Training Epoch: 10/100, batch 129/250 completed (loss: 0.0453:  52%|█████▏    | 65/125 [06:31<06:06,  6.10s/it]
Training Epoch: 10/100, batch 129/250 completed (loss: 0.0299:  52%|█████▏    | 65/125 [06:32<06:06,  6.10s/it]
Training Epoch: 10/100, batch 130/250 completed (loss: 0.0694:  52%|█████▏    | 65/125 [06:35<06:06,  6.10s/it]
Training Epoch: 10/100, batch 130/250 completed (loss: 0.0573:  52%|█████▏    | 65/125 [06:35<06:06,  6.10s/it]
Training Epoch: 10/100, batch 130/250 completed (loss: 0.0165:  52%|█████▏    | 65/125 [06:35<06:06,  6.10s/it]
Training Epoch: 10/100, batch 130/250 completed (loss: 0.091:  52%|█████▏    | 65/125 [06:35<06:06,  6.10s/it] 
Training Epoch: 10/100, batch 130/250 completed (loss: 0.0165:  53%|█████▎    | 66/125 [06:38<06:01,  6.12s/it]
Training Epoch: 10/100, batch 130/250 completed (loss: 0.0573:  53%|█████▎    | 66/125 [06:38<06:01,  6.12s/it]
Training Epoch: 10/100, batch 130/250 completed (loss: 0.0694:  53%|█████▎    | 66/125 [06:38<06:01,  6.12s/it]
Training Epoch: 10/100, batch 130/250 completed (loss: 0.091:  53%|█████▎    | 66/125 [06:38<06:01,  6.12s/it]
Training Epoch: 10/100, batch 131/250 completed (loss: 0.0319:  53%|█████▎    | 66/125 [06:38<06:01,  6.12s/it]
Training Epoch: 10/100, batch 131/250 completed (loss: 0.0863:  53%|█████▎    | 66/125 [06:38<06:01,  6.12s/it]
Training Epoch: 10/100, batch 131/250 completed (loss: 0.0144:  53%|█████▎    | 66/125 [06:38<06:01,  6.12s/it]
Training Epoch: 10/100, batch 131/250 completed (loss: 0.0506:  53%|█████▎    | 66/125 [06:38<06:01,  6.12s/it]
Training Epoch: 10/100, batch 132/250 completed (loss: 0.0566:  53%|█████▎    | 66/125 [06:41<06:01,  6.12s/it]
Training Epoch: 10/100, batch 132/250 completed (loss: 0.0153:  53%|█████▎    | 66/125 [06:41<06:01,  6.12s/it]
Training Epoch: 10/100, batch 132/250 completed (loss: 0.0096:  53%|█████▎    | 66/125 [06:41<06:01,  6.12s/it]
Training Epoch: 10/100, batch 132/250 completed (loss: 0.008:  53%|█████▎    | 66/125 [06:41<06:01,  6.12s/it] 
Training Epoch: 10/100, batch 132/250 completed (loss: 0.0153:  54%|█████▎    | 67/125 [06:44<05:53,  6.09s/it]
Training Epoch: 10/100, batch 132/250 completed (loss: 0.0566:  54%|█████▎    | 67/125 [06:44<05:53,  6.09s/it]
Training Epoch: 10/100, batch 132/250 completed (loss: 0.0096:  54%|█████▎    | 67/125 [06:44<05:53,  6.09s/it]
Training Epoch: 10/100, batch 132/250 completed (loss: 0.008:  54%|█████▎    | 67/125 [06:44<05:53,  6.09s/it]
Training Epoch: 10/100, batch 133/250 completed (loss: 0.0011:  54%|█████▎    | 67/125 [06:44<05:53,  6.09s/it]
Training Epoch: 10/100, batch 133/250 completed (loss: 0.1298:  54%|█████▎    | 67/125 [06:44<05:53,  6.09s/it]
Training Epoch: 10/100, batch 133/250 completed (loss: 0.003:  54%|█████▎    | 67/125 [06:44<05:53,  6.09s/it] 
Training Epoch: 10/100, batch 133/250 completed (loss: 0.1296:  54%|█████▎    | 67/125 [06:44<05:53,  6.09s/it]
Training Epoch: 10/100, batch 134/250 completed (loss: 0.0056:  54%|█████▎    | 67/125 [06:47<05:53,  6.09s/it]
Training Epoch: 10/100, batch 134/250 completed (loss: 0.1131:  54%|█████▎    | 67/125 [06:47<05:53,  6.09s/it]
Training Epoch: 10/100, batch 134/250 completed (loss: 0.0214:  54%|█████▎    | 67/125 [06:47<05:53,  6.09s/it]
Training Epoch: 10/100, batch 134/250 completed (loss: 0.0045:  54%|█████▎    | 67/125 [06:47<05:53,  6.09s/it]
Training Epoch: 10/100, batch 134/250 completed (loss: 0.0045:  54%|█████▍    | 68/125 [06:50<05:50,  6.15s/it]
Training Epoch: 10/100, batch 134/250 completed (loss: 0.0056:  54%|█████▍    | 68/125 [06:50<05:50,  6.15s/it]
Training Epoch: 10/100, batch 134/250 completed (loss: 0.0214:  54%|█████▍    | 68/125 [06:50<05:50,  6.16s/it]
Training Epoch: 10/100, batch 134/250 completed (loss: 0.1131:  54%|█████▍    | 68/125 [06:50<05:50,  6.16s/it]
Training Epoch: 10/100, batch 135/250 completed (loss: 0.006:  54%|█████▍    | 68/125 [06:50<05:50,  6.16s/it] 
Training Epoch: 10/100, batch 135/250 completed (loss: 0.0226:  54%|█████▍    | 68/125 [06:50<05:50,  6.16s/it]
Training Epoch: 10/100, batch 135/250 completed (loss: 0.0619:  54%|█████▍    | 68/125 [06:50<05:50,  6.15s/it]
Training Epoch: 10/100, batch 135/250 completed (loss: 0.0101:  54%|█████▍    | 68/125 [06:50<05:50,  6.15s/it]
Training Epoch: 10/100, batch 136/250 completed (loss: 0.064:  54%|█████▍    | 68/125 [06:53<05:50,  6.16s/it]
Training Epoch: 10/100, batch 136/250 completed (loss: 0.0153:  54%|█████▍    | 68/125 [06:53<05:50,  6.15s/it]
Training Epoch: 10/100, batch 136/250 completed (loss: 0.0452:  54%|█████▍    | 68/125 [06:53<05:50,  6.16s/it]
Training Epoch: 10/100, batch 136/250 completed (loss: 0.0329:  54%|█████▍    | 68/125 [06:53<05:50,  6.15s/it]
Training Epoch: 10/100, batch 136/250 completed (loss: 0.064:  55%|█████▌    | 69/125 [06:56<05:39,  6.06s/it]
Training Epoch: 10/100, batch 136/250 completed (loss: 0.0452:  55%|█████▌    | 69/125 [06:56<05:39,  6.06s/it]
Training Epoch: 10/100, batch 136/250 completed (loss: 0.0153:  55%|█████▌    | 69/125 [06:56<05:39,  6.07s/it]
Training Epoch: 10/100, batch 136/250 completed (loss: 0.0329:  55%|█████▌    | 69/125 [06:56<05:39,  6.06s/it]
Training Epoch: 10/100, batch 137/250 completed (loss: 0.0053:  55%|█████▌    | 69/125 [06:56<05:39,  6.07s/it]
Training Epoch: 10/100, batch 137/250 completed (loss: 0.0109:  55%|█████▌    | 69/125 [06:56<05:39,  6.06s/it]
Training Epoch: 10/100, batch 137/250 completed (loss: 0.0688:  55%|█████▌    | 69/125 [06:56<05:39,  6.06s/it]
Training Epoch: 10/100, batch 137/250 completed (loss: 0.228:  55%|█████▌    | 69/125 [06:56<05:39,  6.06s/it] 
Training Epoch: 10/100, batch 138/250 completed (loss: 0.0061:  55%|█████▌    | 69/125 [06:59<05:39,  6.07s/it]
Training Epoch: 10/100, batch 138/250 completed (loss: 0.0801:  55%|█████▌    | 69/125 [06:59<05:39,  6.06s/it]
Training Epoch: 10/100, batch 138/250 completed (loss: 0.0862:  55%|█████▌    | 69/125 [06:59<05:39,  6.06s/it]
Training Epoch: 10/100, batch 138/250 completed (loss: 0.0019:  55%|█████▌    | 69/125 [06:59<05:39,  6.06s/it]
Training Epoch: 10/100, batch 138/250 completed (loss: 0.0862:  56%|█████▌    | 70/125 [07:02<05:34,  6.09s/it]
Training Epoch: 10/100, batch 138/250 completed (loss: 0.0061:  56%|█████▌    | 70/125 [07:02<05:34,  6.09s/it]
Training Epoch: 10/100, batch 138/250 completed (loss: 0.0801:  56%|█████▌    | 70/125 [07:02<05:35,  6.09s/it]
Training Epoch: 10/100, batch 138/250 completed (loss: 0.0019:  56%|█████▌    | 70/125 [07:02<05:35,  6.09s/it]
Training Epoch: 10/100, batch 139/250 completed (loss: 0.0497:  56%|█████▌    | 70/125 [07:02<05:34,  6.09s/it]
Training Epoch: 10/100, batch 139/250 completed (loss: 0.0191:  56%|█████▌    | 70/125 [07:02<05:35,  6.09s/it]
Training Epoch: 10/100, batch 139/250 completed (loss: 0.2089:  56%|█████▌    | 70/125 [07:02<05:34,  6.09s/it]
Training Epoch: 10/100, batch 139/250 completed (loss: 0.0162:  56%|█████▌    | 70/125 [07:02<05:35,  6.09s/it]
Training Epoch: 10/100, batch 140/250 completed (loss: 0.0167:  56%|█████▌    | 70/125 [07:05<05:34,  6.09s/it]
Training Epoch: 10/100, batch 140/250 completed (loss: 0.0809:  56%|█████▌    | 70/125 [07:05<05:34,  6.09s/it]
Training Epoch: 10/100, batch 140/250 completed (loss: 0.0961:  56%|█████▌    | 70/125 [07:05<05:35,  6.09s/it]
Training Epoch: 10/100, batch 140/250 completed (loss: 0.0256:  56%|█████▌    | 70/125 [07:05<05:35,  6.09s/it]
Training Epoch: 10/100, batch 140/250 completed (loss: 0.0809:  57%|█████▋    | 71/125 [07:08<05:30,  6.11s/it]
Training Epoch: 10/100, batch 140/250 completed (loss: 0.0961:  57%|█████▋    | 71/125 [07:08<05:30,  6.11s/it]
Training Epoch: 10/100, batch 140/250 completed (loss: 0.0167:  57%|█████▋    | 71/125 [07:08<05:30,  6.12s/it]
Training Epoch: 10/100, batch 140/250 completed (loss: 0.0256:  57%|█████▋    | 71/125 [07:08<05:30,  6.11s/it]
Training Epoch: 10/100, batch 141/250 completed (loss: 0.0368:  57%|█████▋    | 71/125 [07:08<05:30,  6.12s/it]
Training Epoch: 10/100, batch 141/250 completed (loss: 0.0082:  57%|█████▋    | 71/125 [07:08<05:30,  6.11s/it]
Training Epoch: 10/100, batch 141/250 completed (loss: 0.1477:  57%|█████▋    | 71/125 [07:08<05:30,  6.11s/it]
Training Epoch: 10/100, batch 141/250 completed (loss: 0.1728:  57%|█████▋    | 71/125 [07:08<05:30,  6.11s/it]
Training Epoch: 10/100, batch 142/250 completed (loss: 0.072:  57%|█████▋    | 71/125 [07:11<05:30,  6.12s/it] 
Training Epoch: 10/100, batch 142/250 completed (loss: 0.02:  57%|█████▋    | 71/125 [07:11<05:30,  6.11s/it]  
Training Epoch: 10/100, batch 142/250 completed (loss: 0.0937:  57%|█████▋    | 71/125 [07:11<05:30,  6.11s/it]
Training Epoch: 10/100, batch 142/250 completed (loss: 0.0164:  57%|█████▋    | 71/125 [07:11<05:30,  6.11s/it]
Training Epoch: 10/100, batch 142/250 completed (loss: 0.072:  58%|█████▊    | 72/125 [07:14<05:19,  6.04s/it]
Training Epoch: 10/100, batch 142/250 completed (loss: 0.0164:  58%|█████▊    | 72/125 [07:14<05:19,  6.03s/it]
Training Epoch: 10/100, batch 142/250 completed (loss: 0.0937:  58%|█████▊    | 72/125 [07:14<05:20,  6.04s/it]
Training Epoch: 10/100, batch 142/250 completed (loss: 0.02:  58%|█████▊    | 72/125 [07:14<05:20,  6.04s/it]
Training Epoch: 10/100, batch 143/250 completed (loss: 0.1851:  58%|█████▊    | 72/125 [07:14<05:20,  6.04s/it]
Training Epoch: 10/100, batch 143/250 completed (loss: 0.1789:  58%|█████▊    | 72/125 [07:14<05:20,  6.04s/it]
Training Epoch: 10/100, batch 143/250 completed (loss: 0.0315:  58%|█████▊    | 72/125 [07:14<05:19,  6.04s/it]
Training Epoch: 10/100, batch 143/250 completed (loss: 0.0151:  58%|█████▊    | 72/125 [07:14<05:19,  6.03s/it]
Training Epoch: 10/100, batch 144/250 completed (loss: 0.1214:  58%|█████▊    | 72/125 [07:17<05:20,  6.04s/it]
Training Epoch: 10/100, batch 144/250 completed (loss: 0.1466:  58%|█████▊    | 72/125 [07:17<05:20,  6.04s/it]
Training Epoch: 10/100, batch 144/250 completed (loss: 0.1065:  58%|█████▊    | 72/125 [07:17<05:19,  6.04s/it]
Training Epoch: 10/100, batch 144/250 completed (loss: 0.1273:  58%|█████▊    | 72/125 [07:17<05:19,  6.03s/it]
Training Epoch: 10/100, batch 144/250 completed (loss: 0.1273:  58%|█████▊    | 73/125 [07:20<05:20,  6.17s/it]
Training Epoch: 10/100, batch 144/250 completed (loss: 0.1065:  58%|█████▊    | 73/125 [07:20<05:20,  6.17s/it]
Training Epoch: 10/100, batch 144/250 completed (loss: 0.1466:  58%|█████▊    | 73/125 [07:20<05:21,  6.17s/it]
Training Epoch: 10/100, batch 144/250 completed (loss: 0.1214:  58%|█████▊    | 73/125 [07:20<05:21,  6.18s/it]
Training Epoch: 10/100, batch 145/250 completed (loss: 0.0326:  58%|█████▊    | 73/125 [07:21<05:20,  6.17s/it]
Training Epoch: 10/100, batch 145/250 completed (loss: 0.0213:  58%|█████▊    | 73/125 [07:21<05:21,  6.17s/it]
Training Epoch: 10/100, batch 145/250 completed (loss: 0.0854:  58%|█████▊    | 73/125 [07:21<05:21,  6.18s/it]
Training Epoch: 10/100, batch 145/250 completed (loss: 0.0124:  58%|█████▊    | 73/125 [07:21<05:20,  6.17s/it]
Training Epoch: 10/100, batch 146/250 completed (loss: 0.0008:  58%|█████▊    | 73/125 [07:23<05:21,  6.18s/it]
Training Epoch: 10/100, batch 146/250 completed (loss: 0.0765:  58%|█████▊    | 73/125 [07:23<05:21,  6.17s/it]
Training Epoch: 10/100, batch 146/250 completed (loss: 0.0057:  58%|█████▊    | 73/125 [07:23<05:20,  6.17s/it]
Training Epoch: 10/100, batch 146/250 completed (loss: 0.0154:  58%|█████▊    | 73/125 [07:23<05:20,  6.17s/it]
Training Epoch: 10/100, batch 146/250 completed (loss: 0.0057:  59%|█████▉    | 74/125 [07:26<05:10,  6.08s/it]
Training Epoch: 10/100, batch 146/250 completed (loss: 0.0154:  59%|█████▉    | 74/125 [07:26<05:10,  6.08s/it]
Training Epoch: 10/100, batch 146/250 completed (loss: 0.0765:  59%|█████▉    | 74/125 [07:26<05:10,  6.08s/it]
Training Epoch: 10/100, batch 146/250 completed (loss: 0.0008:  59%|█████▉    | 74/125 [07:26<05:10,  6.08s/it]
Training Epoch: 10/100, batch 147/250 completed (loss: 0.0634:  59%|█████▉    | 74/125 [07:26<05:10,  6.08s/it]
Training Epoch: 10/100, batch 147/250 completed (loss: 0.0041:  59%|█████▉    | 74/125 [07:26<05:10,  6.08s/it]
Training Epoch: 10/100, batch 147/250 completed (loss: 0.2267:  59%|█████▉    | 74/125 [07:26<05:10,  6.08s/it]
Training Epoch: 10/100, batch 147/250 completed (loss: 0.025:  59%|█████▉    | 74/125 [07:26<05:10,  6.08s/it] 
Training Epoch: 10/100, batch 148/250 completed (loss: 0.1468:  59%|█████▉    | 74/125 [07:29<05:10,  6.08s/it]
Training Epoch: 10/100, batch 148/250 completed (loss: 0.1801:  59%|█████▉    | 74/125 [07:29<05:10,  6.08s/it]
Training Epoch: 10/100, batch 148/250 completed (loss: 0.1377:  59%|█████▉    | 74/125 [07:29<05:10,  6.08s/it]
Training Epoch: 10/100, batch 148/250 completed (loss: 0.0546:  59%|█████▉    | 74/125 [07:29<05:10,  6.08s/it]
Training Epoch: 10/100, batch 148/250 completed (loss: 0.0546:  60%|██████    | 75/125 [07:32<05:03,  6.06s/it]
Training Epoch: 10/100, batch 148/250 completed (loss: 0.1801:  60%|██████    | 75/125 [07:32<05:03,  6.06s/it]
Training Epoch: 10/100, batch 148/250 completed (loss: 0.1468:  60%|██████    | 75/125 [07:32<05:03,  6.06s/it]
Training Epoch: 10/100, batch 148/250 completed (loss: 0.1377:  60%|██████    | 75/125 [07:32<05:03,  6.07s/it]
Training Epoch: 10/100, batch 149/250 completed (loss: 0.0195:  60%|██████    | 75/125 [07:32<05:03,  6.07s/it]
Training Epoch: 10/100, batch 149/250 completed (loss: 0.0292:  60%|██████    | 75/125 [07:32<05:03,  6.06s/it]
Training Epoch: 10/100, batch 149/250 completed (loss: 0.0319:  60%|██████    | 75/125 [07:32<05:03,  6.06s/it]
Training Epoch: 10/100, batch 149/250 completed (loss: 0.016:  60%|██████    | 75/125 [07:32<05:03,  6.06s/it] 
Training Epoch: 10/100, batch 150/250 completed (loss: 0.0017:  60%|██████    | 75/125 [07:36<05:03,  6.06s/it]
Training Epoch: 10/100, batch 150/250 completed (loss: 0.0043:  60%|██████    | 75/125 [07:36<05:03,  6.07s/it]
Training Epoch: 10/100, batch 150/250 completed (loss: 0.0183:  60%|██████    | 75/125 [07:36<05:03,  6.06s/it]
Training Epoch: 10/100, batch 150/250 completed (loss: 0.0491:  60%|██████    | 75/125 [07:36<05:03,  6.06s/it]
Training Epoch: 10/100, batch 150/250 completed (loss: 0.0043:  61%|██████    | 76/125 [07:39<05:00,  6.14s/it]
Training Epoch: 10/100, batch 150/250 completed (loss: 0.0491:  61%|██████    | 76/125 [07:39<05:00,  6.14s/it]
Training Epoch: 10/100, batch 150/250 completed (loss: 0.0017:  61%|██████    | 76/125 [07:39<05:00,  6.14s/it]
Training Epoch: 10/100, batch 150/250 completed (loss: 0.0183:  61%|██████    | 76/125 [07:39<05:01,  6.14s/it]
Training Epoch: 10/100, batch 151/250 completed (loss: 0.023:  61%|██████    | 76/125 [07:39<05:01,  6.14s/it] 
Training Epoch: 10/100, batch 151/250 completed (loss: 0.1586:  61%|██████    | 76/125 [07:39<05:00,  6.14s/it]
Training Epoch: 10/100, batch 151/250 completed (loss: 0.0548:  61%|██████    | 76/125 [07:39<05:00,  6.14s/it]
Training Epoch: 10/100, batch 151/250 completed (loss: 0.1212:  61%|██████    | 76/125 [07:39<05:00,  6.14s/it]
Training Epoch: 10/100, batch 152/250 completed (loss: 0.0602:  61%|██████    | 76/125 [07:42<05:00,  6.14s/it]
Training Epoch: 10/100, batch 152/250 completed (loss: 0.3459:  61%|██████    | 76/125 [07:42<05:01,  6.14s/it]
Training Epoch: 10/100, batch 152/250 completed (loss: 0.0057:  61%|██████    | 76/125 [07:42<05:00,  6.14s/it]
Training Epoch: 10/100, batch 152/250 completed (loss: 0.045:  61%|██████    | 76/125 [07:42<05:00,  6.14s/it] 
Training Epoch: 10/100, batch 152/250 completed (loss: 0.045:  62%|██████▏   | 77/125 [07:45<04:52,  6.10s/it]
Training Epoch: 10/100, batch 152/250 completed (loss: 0.0057:  62%|██████▏   | 77/125 [07:45<04:53,  6.11s/it]
Training Epoch: 10/100, batch 152/250 completed (loss: 0.3459:  62%|██████▏   | 77/125 [07:45<04:52,  6.10s/it]
Training Epoch: 10/100, batch 152/250 completed (loss: 0.0602:  62%|██████▏   | 77/125 [07:45<04:52,  6.10s/it]
Training Epoch: 10/100, batch 153/250 completed (loss: 0.0357:  62%|██████▏   | 77/125 [07:45<04:53,  6.11s/it]
Training Epoch: 10/100, batch 153/250 completed (loss: 0.081:  62%|██████▏   | 77/125 [07:45<04:52,  6.10s/it] 
Training Epoch: 10/100, batch 153/250 completed (loss: 0.1608:  62%|██████▏   | 77/125 [07:45<04:52,  6.10s/it]
Training Epoch: 10/100, batch 153/250 completed (loss: 0.0347:  62%|██████▏   | 77/125 [07:45<04:52,  6.10s/it]
Training Epoch: 10/100, batch 154/250 completed (loss: 0.0165:  62%|██████▏   | 77/125 [07:48<04:53,  6.11s/it]
Training Epoch: 10/100, batch 154/250 completed (loss: 0.0474:  62%|██████▏   | 77/125 [07:48<04:52,  6.10s/it]
Training Epoch: 10/100, batch 154/250 completed (loss: 0.0451:  62%|██████▏   | 77/125 [07:48<04:52,  6.10s/it]
Training Epoch: 10/100, batch 154/250 completed (loss: 0.0439:  62%|██████▏   | 77/125 [07:48<04:52,  6.10s/it]
Training Epoch: 10/100, batch 154/250 completed (loss: 0.0165:  62%|██████▏   | 78/125 [07:51<04:46,  6.10s/it]
Training Epoch: 10/100, batch 154/250 completed (loss: 0.0474:  62%|██████▏   | 78/125 [07:51<04:46,  6.10s/it]
Training Epoch: 10/100, batch 154/250 completed (loss: 0.0439:  62%|██████▏   | 78/125 [07:51<04:46,  6.10s/it]
Training Epoch: 10/100, batch 154/250 completed (loss: 0.0451:  62%|██████▏   | 78/125 [07:51<04:46,  6.10s/it]
Training Epoch: 10/100, batch 155/250 completed (loss: 0.0898:  62%|██████▏   | 78/125 [07:51<04:46,  6.10s/it]
Training Epoch: 10/100, batch 155/250 completed (loss: 0.0338:  62%|██████▏   | 78/125 [07:51<04:46,  6.10s/it]
Training Epoch: 10/100, batch 155/250 completed (loss: 0.0385:  62%|██████▏   | 78/125 [07:51<04:46,  6.10s/it]
Training Epoch: 10/100, batch 155/250 completed (loss: 0.0059:  62%|██████▏   | 78/125 [07:51<04:46,  6.10s/it]
Training Epoch: 10/100, batch 156/250 completed (loss: 0.0151:  62%|██████▏   | 78/125 [07:54<04:46,  6.10s/it]
Training Epoch: 10/100, batch 156/250 completed (loss: 0.0561:  62%|██████▏   | 78/125 [07:54<04:46,  6.10s/it]
Training Epoch: 10/100, batch 156/250 completed (loss: 0.0864:  62%|██████▏   | 78/125 [07:54<04:46,  6.10s/it]
Training Epoch: 10/100, batch 156/250 completed (loss: 0.0259:  62%|██████▏   | 78/125 [07:54<04:46,  6.10s/it]
Training Epoch: 10/100, batch 156/250 completed (loss: 0.0151:  63%|██████▎   | 79/125 [07:57<04:39,  6.09s/it]
Training Epoch: 10/100, batch 156/250 completed (loss: 0.0259:  63%|██████▎   | 79/125 [07:57<04:39,  6.09s/it]
Training Epoch: 10/100, batch 156/250 completed (loss: 0.0561:  63%|██████▎   | 79/125 [07:57<04:39,  6.09s/it]
Training Epoch: 10/100, batch 156/250 completed (loss: 0.0864:  63%|██████▎   | 79/125 [07:57<04:39,  6.09s/it]
Training Epoch: 10/100, batch 157/250 completed (loss: 0.1338:  63%|██████▎   | 79/125 [07:57<04:39,  6.09s/it]
Training Epoch: 10/100, batch 157/250 completed (loss: 0.0442:  63%|██████▎   | 79/125 [07:57<04:39,  6.09s/it]
Training Epoch: 10/100, batch 157/250 completed (loss: 0.2952:  63%|██████▎   | 79/125 [07:57<04:39,  6.09s/it]
Training Epoch: 10/100, batch 157/250 completed (loss: 0.0992:  63%|██████▎   | 79/125 [07:57<04:39,  6.09s/it]
Training Epoch: 10/100, batch 158/250 completed (loss: 0.1564:  63%|██████▎   | 79/125 [08:00<04:39,  6.09s/it]
Training Epoch: 10/100, batch 158/250 completed (loss: 0.2119:  63%|██████▎   | 79/125 [08:00<04:39,  6.09s/it]
Training Epoch: 10/100, batch 158/250 completed (loss: 0.0474:  63%|██████▎   | 79/125 [08:00<04:39,  6.09s/it]
Training Epoch: 10/100, batch 158/250 completed (loss: 0.0175:  63%|██████▎   | 79/125 [08:00<04:39,  6.09s/it]
Training Epoch: 10/100, batch 158/250 completed (loss: 0.2119:  64%|██████▍   | 80/125 [08:03<04:32,  6.06s/it]
Training Epoch: 10/100, batch 158/250 completed (loss: 0.0175:  64%|██████▍   | 80/125 [08:03<04:32,  6.06s/it]
Training Epoch: 10/100, batch 158/250 completed (loss: 0.0474:  64%|██████▍   | 80/125 [08:03<04:32,  6.06s/it]
Training Epoch: 10/100, batch 158/250 completed (loss: 0.1564:  64%|██████▍   | 80/125 [08:03<04:32,  6.06s/it]
Training Epoch: 10/100, batch 159/250 completed (loss: 0.0166:  64%|██████▍   | 80/125 [08:03<04:32,  6.06s/it]
Training Epoch: 10/100, batch 159/250 completed (loss: 0.0488:  64%|██████▍   | 80/125 [08:03<04:32,  6.06s/it]
Training Epoch: 10/100, batch 159/250 completed (loss: 0.0853:  64%|██████▍   | 80/125 [08:03<04:32,  6.06s/it]
Training Epoch: 10/100, batch 159/250 completed (loss: 0.0462:  64%|██████▍   | 80/125 [08:03<04:32,  6.06s/it]
Training Epoch: 10/100, batch 160/250 completed (loss: 0.0089:  64%|██████▍   | 80/125 [08:06<04:32,  6.06s/it]
Training Epoch: 10/100, batch 160/250 completed (loss: 0.0281:  64%|██████▍   | 80/125 [08:06<04:32,  6.06s/it]
Training Epoch: 10/100, batch 160/250 completed (loss: 0.1415:  64%|██████▍   | 80/125 [08:06<04:32,  6.06s/it]
Training Epoch: 10/100, batch 160/250 completed (loss: 0.2401:  64%|██████▍   | 80/125 [08:06<04:32,  6.06s/it]
Training Epoch: 10/100, batch 160/250 completed (loss: 0.0281:  65%|██████▍   | 81/125 [08:09<04:30,  6.14s/it]
Training Epoch: 10/100, batch 160/250 completed (loss: 0.2401:  65%|██████▍   | 81/125 [08:09<04:30,  6.14s/it]
Training Epoch: 10/100, batch 160/250 completed (loss: 0.0089:  65%|██████▍   | 81/125 [08:09<04:30,  6.14s/it]
Training Epoch: 10/100, batch 160/250 completed (loss: 0.1415:  65%|██████▍   | 81/125 [08:09<04:30,  6.14s/it]
Training Epoch: 10/100, batch 161/250 completed (loss: 0.0124:  65%|██████▍   | 81/125 [08:09<04:30,  6.14s/it]
Training Epoch: 10/100, batch 161/250 completed (loss: 0.015:  65%|██████▍   | 81/125 [08:09<04:30,  6.14s/it] 
Training Epoch: 10/100, batch 161/250 completed (loss: 0.0755:  65%|██████▍   | 81/125 [08:09<04:30,  6.14s/it]
Training Epoch: 10/100, batch 161/250 completed (loss: 0.0219:  65%|██████▍   | 81/125 [08:09<04:30,  6.14s/it]
Training Epoch: 10/100, batch 162/250 completed (loss: 0.1087:  65%|██████▍   | 81/125 [08:12<04:30,  6.14s/it]
Training Epoch: 10/100, batch 162/250 completed (loss: 0.0711:  65%|██████▍   | 81/125 [08:12<04:30,  6.14s/it]
Training Epoch: 10/100, batch 162/250 completed (loss: 0.0409:  65%|██████▍   | 81/125 [08:12<04:30,  6.14s/it]
Training Epoch: 10/100, batch 162/250 completed (loss: 0.0359:  65%|██████▍   | 81/125 [08:12<04:30,  6.14s/it]
Training Epoch: 10/100, batch 162/250 completed (loss: 0.0409:  66%|██████▌   | 82/125 [08:15<04:20,  6.07s/it]
Training Epoch: 10/100, batch 162/250 completed (loss: 0.1087:  66%|██████▌   | 82/125 [08:15<04:20,  6.06s/it]
Training Epoch: 10/100, batch 162/250 completed (loss: 0.0359:  66%|██████▌   | 82/125 [08:15<04:20,  6.07s/it]
Training Epoch: 10/100, batch 162/250 completed (loss: 0.0711:  66%|██████▌   | 82/125 [08:15<04:21,  6.07s/it]
Training Epoch: 10/100, batch 163/250 completed (loss: 0.1105:  66%|██████▌   | 82/125 [08:15<04:20,  6.06s/it]
Training Epoch: 10/100, batch 163/250 completed (loss: 0.077:  66%|██████▌   | 82/125 [08:15<04:20,  6.07s/it] 
Training Epoch: 10/100, batch 163/250 completed (loss: 0.0062:  66%|██████▌   | 82/125 [08:15<04:21,  6.07s/it]
Training Epoch: 10/100, batch 163/250 completed (loss: 0.1179:  66%|██████▌   | 82/125 [08:15<04:20,  6.07s/it]
Training Epoch: 10/100, batch 164/250 completed (loss: 0.1617:  66%|██████▌   | 82/125 [08:18<04:21,  6.07s/it]
Training Epoch: 10/100, batch 164/250 completed (loss: 0.0304:  66%|██████▌   | 82/125 [08:18<04:20,  6.06s/it]
Training Epoch: 10/100, batch 164/250 completed (loss: 0.0282:  66%|██████▌   | 82/125 [08:18<04:20,  6.07s/it]
Training Epoch: 10/100, batch 164/250 completed (loss: 0.0464:  66%|██████▌   | 82/125 [08:18<04:20,  6.07s/it]
Training Epoch: 10/100, batch 164/250 completed (loss: 0.0464:  66%|██████▋   | 83/125 [08:21<04:14,  6.06s/it]
Training Epoch: 10/100, batch 164/250 completed (loss: 0.0282:  66%|██████▋   | 83/125 [08:21<04:14,  6.06s/it]
Training Epoch: 10/100, batch 164/250 completed (loss: 0.0304:  66%|██████▋   | 83/125 [08:21<04:14,  6.06s/it]
Training Epoch: 10/100, batch 164/250 completed (loss: 0.1617:  66%|██████▋   | 83/125 [08:21<04:14,  6.06s/it]
Training Epoch: 10/100, batch 165/250 completed (loss: 0.2392:  66%|██████▋   | 83/125 [08:21<04:14,  6.06s/it]
Training Epoch: 10/100, batch 165/250 completed (loss: 0.005:  66%|██████▋   | 83/125 [08:21<04:14,  6.06s/it] 
Training Epoch: 10/100, batch 165/250 completed (loss: 0.056:  66%|██████▋   | 83/125 [08:21<04:14,  6.06s/it] 
Training Epoch: 10/100, batch 165/250 completed (loss: 0.0855:  66%|██████▋   | 83/125 [08:21<04:14,  6.06s/it]
Training Epoch: 10/100, batch 166/250 completed (loss: 0.0292:  66%|██████▋   | 83/125 [08:25<04:14,  6.06s/it]
Training Epoch: 10/100, batch 166/250 completed (loss: 0.0295:  66%|██████▋   | 83/125 [08:24<04:14,  6.06s/it]
Training Epoch: 10/100, batch 166/250 completed (loss: 0.1171:  66%|██████▋   | 83/125 [08:25<04:14,  6.06s/it]
Training Epoch: 10/100, batch 166/250 completed (loss: 0.0451:  66%|██████▋   | 83/125 [08:25<04:14,  6.06s/it]
Training Epoch: 10/100, batch 166/250 completed (loss: 0.1171:  67%|██████▋   | 84/125 [08:27<04:11,  6.13s/it]
Training Epoch: 10/100, batch 166/250 completed (loss: 0.0451:  67%|██████▋   | 84/125 [08:27<04:11,  6.13s/it]
Training Epoch: 10/100, batch 166/250 completed (loss: 0.0292:  67%|██████▋   | 84/125 [08:27<04:11,  6.12s/it]
Training Epoch: 10/100, batch 166/250 completed (loss: 0.0295:  67%|██████▋   | 84/125 [08:27<04:11,  6.13s/it]
Training Epoch: 10/100, batch 167/250 completed (loss: 0.2722:  67%|██████▋   | 84/125 [08:27<04:11,  6.13s/it]
Training Epoch: 10/100, batch 167/250 completed (loss: 0.0291:  67%|██████▋   | 84/125 [08:27<04:11,  6.13s/it]
Training Epoch: 10/100, batch 167/250 completed (loss: 0.1728:  67%|██████▋   | 84/125 [08:27<04:11,  6.12s/it]
Training Epoch: 10/100, batch 167/250 completed (loss: 0.1945:  67%|██████▋   | 84/125 [08:27<04:11,  6.13s/it]
Training Epoch: 10/100, batch 168/250 completed (loss: 0.0472:  67%|██████▋   | 84/125 [08:30<04:11,  6.13s/it]
Training Epoch: 10/100, batch 168/250 completed (loss: 0.0703:  67%|██████▋   | 84/125 [08:30<04:11,  6.13s/it]
Training Epoch: 10/100, batch 168/250 completed (loss: 0.007:  67%|██████▋   | 84/125 [08:30<04:11,  6.12s/it] 
Training Epoch: 10/100, batch 168/250 completed (loss: 0.1959:  67%|██████▋   | 84/125 [08:30<04:11,  6.13s/it]
Training Epoch: 10/100, batch 168/250 completed (loss: 0.0472:  68%|██████▊   | 85/125 [08:33<04:04,  6.10s/it]
Training Epoch: 10/100, batch 168/250 completed (loss: 0.1959:  68%|██████▊   | 85/125 [08:33<04:04,  6.10s/it]
Training Epoch: 10/100, batch 168/250 completed (loss: 0.0703:  68%|██████▊   | 85/125 [08:33<04:04,  6.10s/it]
Training Epoch: 10/100, batch 168/250 completed (loss: 0.007:  68%|██████▊   | 85/125 [08:33<04:04,  6.10s/it]
Training Epoch: 10/100, batch 169/250 completed (loss: 0.19:  68%|██████▊   | 85/125 [08:33<04:04,  6.10s/it]  
Training Epoch: 10/100, batch 169/250 completed (loss: 0.1221:  68%|██████▊   | 85/125 [08:33<04:04,  6.10s/it]
Training Epoch: 10/100, batch 169/250 completed (loss: 0.0166:  68%|██████▊   | 85/125 [08:33<04:04,  6.10s/it]
Training Epoch: 10/100, batch 169/250 completed (loss: 0.0217:  68%|██████▊   | 85/125 [08:33<04:04,  6.10s/it]
Training Epoch: 10/100, batch 170/250 completed (loss: 0.0163:  68%|██████▊   | 85/125 [08:36<04:04,  6.10s/it]
Training Epoch: 10/100, batch 170/250 completed (loss: 0.0234:  68%|██████▊   | 85/125 [08:36<04:04,  6.10s/it]
Training Epoch: 10/100, batch 170/250 completed (loss: 0.2677:  68%|██████▊   | 85/125 [08:36<04:04,  6.10s/it]
Training Epoch: 10/100, batch 170/250 completed (loss: 0.0067:  68%|██████▊   | 85/125 [08:36<04:04,  6.10s/it]
Training Epoch: 10/100, batch 170/250 completed (loss: 0.0163:  69%|██████▉   | 86/125 [08:40<03:58,  6.11s/it]
Training Epoch: 10/100, batch 170/250 completed (loss: 0.2677:  69%|██████▉   | 86/125 [08:40<03:58,  6.11s/it]
Training Epoch: 10/100, batch 170/250 completed (loss: 0.0234:  69%|██████▉   | 86/125 [08:40<03:58,  6.11s/it]
Training Epoch: 10/100, batch 170/250 completed (loss: 0.0067:  69%|██████▉   | 86/125 [08:40<03:58,  6.12s/it]
Training Epoch: 10/100, batch 171/250 completed (loss: 0.0457:  69%|██████▉   | 86/125 [08:40<03:58,  6.11s/it]
Training Epoch: 10/100, batch 171/250 completed (loss: 0.2907:  69%|██████▉   | 86/125 [08:40<03:58,  6.11s/it]
Training Epoch: 10/100, batch 171/250 completed (loss: 0.0165:  69%|██████▉   | 86/125 [08:40<03:58,  6.11s/it]
Training Epoch: 10/100, batch 171/250 completed (loss: 0.1347:  69%|██████▉   | 86/125 [08:40<03:58,  6.12s/it]
Training Epoch: 10/100, batch 172/250 completed (loss: 0.0503:  69%|██████▉   | 86/125 [08:43<03:58,  6.11s/it]
Training Epoch: 10/100, batch 172/250 completed (loss: 0.0393:  69%|██████▉   | 86/125 [08:43<03:58,  6.11s/it]
Training Epoch: 10/100, batch 172/250 completed (loss: 0.042:  69%|██████▉   | 86/125 [08:43<03:58,  6.11s/it] 
Training Epoch: 10/100, batch 172/250 completed (loss: 0.1422:  69%|██████▉   | 86/125 [08:43<03:58,  6.12s/it]
Training Epoch: 10/100, batch 172/250 completed (loss: 0.0393:  70%|██████▉   | 87/125 [08:46<03:51,  6.09s/it]
Training Epoch: 10/100, batch 172/250 completed (loss: 0.0503:  70%|██████▉   | 87/125 [08:46<03:51,  6.09s/it]
Training Epoch: 10/100, batch 172/250 completed (loss: 0.1422:  70%|██████▉   | 87/125 [08:46<03:51,  6.09s/it]
Training Epoch: 10/100, batch 172/250 completed (loss: 0.042:  70%|██████▉   | 87/125 [08:46<03:51,  6.09s/it]
Training Epoch: 10/100, batch 173/250 completed (loss: 0.0407:  70%|██████▉   | 87/125 [08:46<03:51,  6.09s/it]
Training Epoch: 10/100, batch 173/250 completed (loss: 0.0788:  70%|██████▉   | 87/125 [08:46<03:51,  6.09s/it]
Training Epoch: 10/100, batch 173/250 completed (loss: 0.0595:  70%|██████▉   | 87/125 [08:46<03:51,  6.09s/it]
Training Epoch: 10/100, batch 173/250 completed (loss: 0.1694:  70%|██████▉   | 87/125 [08:46<03:51,  6.09s/it]
Training Epoch: 10/100, batch 174/250 completed (loss: 0.023:  70%|██████▉   | 87/125 [08:49<03:51,  6.09s/it] 
Training Epoch: 10/100, batch 174/250 completed (loss: 0.0647:  70%|██████▉   | 87/125 [08:49<03:51,  6.09s/it]
Training Epoch: 10/100, batch 174/250 completed (loss: 0.1041:  70%|██████▉   | 87/125 [08:49<03:51,  6.09s/it]
Training Epoch: 10/100, batch 174/250 completed (loss: 0.0409:  70%|██████▉   | 87/125 [08:49<03:51,  6.09s/it]
Training Epoch: 10/100, batch 174/250 completed (loss: 0.1041:  70%|███████   | 88/125 [08:52<03:44,  6.07s/it]
Training Epoch: 10/100, batch 174/250 completed (loss: 0.023:  70%|███████   | 88/125 [08:52<03:44,  6.07s/it]
Training Epoch: 10/100, batch 174/250 completed (loss: 0.0409:  70%|███████   | 88/125 [08:52<03:44,  6.07s/it]
Training Epoch: 10/100, batch 174/250 completed (loss: 0.0647:  70%|███████   | 88/125 [08:52<03:44,  6.08s/it]
Training Epoch: 10/100, batch 175/250 completed (loss: 0.0117:  70%|███████   | 88/125 [08:52<03:44,  6.08s/it]
Training Epoch: 10/100, batch 175/250 completed (loss: 0.1938:  70%|███████   | 88/125 [08:52<03:44,  6.07s/it]
Training Epoch: 10/100, batch 175/250 completed (loss: 0.0042:  70%|███████   | 88/125 [08:52<03:44,  6.07s/it]
Training Epoch: 10/100, batch 175/250 completed (loss: 0.2979:  70%|███████   | 88/125 [08:52<03:44,  6.07s/it]
Training Epoch: 10/100, batch 176/250 completed (loss: 0.1166:  70%|███████   | 88/125 [08:55<03:44,  6.07s/it]
Training Epoch: 10/100, batch 176/250 completed (loss: 0.0079:  70%|███████   | 88/125 [08:55<03:44,  6.08s/it]
Training Epoch: 10/100, batch 176/250 completed (loss: 0.0606:  70%|███████   | 88/125 [08:55<03:44,  6.07s/it]
Training Epoch: 10/100, batch 176/250 completed (loss: 0.0107:  70%|███████   | 88/125 [08:55<03:44,  6.07s/it]
Training Epoch: 10/100, batch 176/250 completed (loss: 0.0107:  71%|███████   | 89/125 [08:58<03:42,  6.17s/it]
Training Epoch: 10/100, batch 176/250 completed (loss: 0.0606:  71%|███████   | 89/125 [08:58<03:42,  6.17s/it]
Training Epoch: 10/100, batch 176/250 completed (loss: 0.1166:  71%|███████   | 89/125 [08:58<03:42,  6.17s/it]
Training Epoch: 10/100, batch 176/250 completed (loss: 0.0079:  71%|███████   | 89/125 [08:58<03:42,  6.17s/it]
Training Epoch: 10/100, batch 177/250 completed (loss: 0.0735:  71%|███████   | 89/125 [08:58<03:42,  6.17s/it]
Training Epoch: 10/100, batch 177/250 completed (loss: 0.2413:  71%|███████   | 89/125 [08:58<03:42,  6.17s/it]
Training Epoch: 10/100, batch 177/250 completed (loss: 0.0686:  71%|███████   | 89/125 [08:58<03:42,  6.17s/it]
Training Epoch: 10/100, batch 177/250 completed (loss: 0.0907:  71%|███████   | 89/125 [08:58<03:42,  6.17s/it]
Training Epoch: 10/100, batch 178/250 completed (loss: 0.0255:  71%|███████   | 89/125 [09:01<03:42,  6.17s/it]
Training Epoch: 10/100, batch 178/250 completed (loss: 0.0101:  71%|███████   | 89/125 [09:01<03:42,  6.17s/it]
Training Epoch: 10/100, batch 178/250 completed (loss: 0.2789:  71%|███████   | 89/125 [09:01<03:42,  6.17s/it]
Training Epoch: 10/100, batch 178/250 completed (loss: 0.0471:  71%|███████   | 89/125 [09:01<03:42,  6.17s/it]
Training Epoch: 10/100, batch 178/250 completed (loss: 0.0101:  72%|███████▏  | 90/125 [09:04<03:34,  6.12s/it]
Training Epoch: 10/100, batch 178/250 completed (loss: 0.0471:  72%|███████▏  | 90/125 [09:04<03:34,  6.12s/it]
Training Epoch: 10/100, batch 178/250 completed (loss: 0.0255:  72%|███████▏  | 90/125 [09:04<03:34,  6.12s/it]
Training Epoch: 10/100, batch 178/250 completed (loss: 0.2789:  72%|███████▏  | 90/125 [09:04<03:34,  6.12s/it]
Training Epoch: 10/100, batch 179/250 completed (loss: 0.1271:  72%|███████▏  | 90/125 [09:04<03:34,  6.12s/it]
Training Epoch: 10/100, batch 179/250 completed (loss: 0.0203:  72%|███████▏  | 90/125 [09:04<03:34,  6.12s/it]
Training Epoch: 10/100, batch 179/250 completed (loss: 0.1173:  72%|███████▏  | 90/125 [09:04<03:34,  6.12s/it]
Training Epoch: 10/100, batch 179/250 completed (loss: 0.0387:  72%|███████▏  | 90/125 [09:04<03:34,  6.12s/it]
Training Epoch: 10/100, batch 180/250 completed (loss: 0.0135:  72%|███████▏  | 90/125 [09:07<03:34,  6.12s/it]
Training Epoch: 10/100, batch 180/250 completed (loss: 0.0269:  72%|███████▏  | 90/125 [09:07<03:34,  6.12s/it]
Training Epoch: 10/100, batch 180/250 completed (loss: 0.0499:  72%|███████▏  | 90/125 [09:07<03:34,  6.12s/it]
Training Epoch: 10/100, batch 180/250 completed (loss: 0.0636:  72%|███████▏  | 90/125 [09:07<03:34,  6.12s/it]
Training Epoch: 10/100, batch 180/250 completed (loss: 0.0269:  73%|███████▎  | 91/125 [09:10<03:27,  6.09s/it]
Training Epoch: 10/100, batch 180/250 completed (loss: 0.0135:  73%|███████▎  | 91/125 [09:10<03:26,  6.09s/it]
Training Epoch: 10/100, batch 180/250 completed (loss: 0.0636:  73%|███████▎  | 91/125 [09:10<03:27,  6.09s/it]
Training Epoch: 10/100, batch 180/250 completed (loss: 0.0499:  73%|███████▎  | 91/125 [09:10<03:27,  6.09s/it]
Training Epoch: 10/100, batch 181/250 completed (loss: 0.0394:  73%|███████▎  | 91/125 [09:10<03:27,  6.09s/it]
Training Epoch: 10/100, batch 181/250 completed (loss: 0.0199:  73%|███████▎  | 91/125 [09:10<03:26,  6.09s/it]
Training Epoch: 10/100, batch 181/250 completed (loss: 0.0164:  73%|███████▎  | 91/125 [09:10<03:27,  6.09s/it]
Training Epoch: 10/100, batch 181/250 completed (loss: 0.1732:  73%|███████▎  | 91/125 [09:10<03:27,  6.09s/it]
Training Epoch: 10/100, batch 182/250 completed (loss: 0.0228:  73%|███████▎  | 91/125 [09:13<03:27,  6.09s/it]
Training Epoch: 10/100, batch 182/250 completed (loss: 0.0448:  73%|███████▎  | 91/125 [09:13<03:26,  6.09s/it]
Training Epoch: 10/100, batch 182/250 completed (loss: 0.0243:  73%|███████▎  | 91/125 [09:13<03:27,  6.09s/it]
Training Epoch: 10/100, batch 182/250 completed (loss: 0.2363:  73%|███████▎  | 91/125 [09:13<03:27,  6.09s/it]
Training Epoch: 10/100, batch 182/250 completed (loss: 0.0243:  74%|███████▎  | 92/125 [09:16<03:21,  6.12s/it]
Training Epoch: 10/100, batch 182/250 completed (loss: 0.0448:  74%|███████▎  | 92/125 [09:16<03:22,  6.12s/it]
Training Epoch: 10/100, batch 182/250 completed (loss: 0.0228:  74%|███████▎  | 92/125 [09:16<03:22,  6.12s/it]
Training Epoch: 10/100, batch 182/250 completed (loss: 0.2363:  74%|███████▎  | 92/125 [09:16<03:22,  6.12s/it]
Training Epoch: 10/100, batch 183/250 completed (loss: 0.0275:  74%|███████▎  | 92/125 [09:16<03:21,  6.12s/it]
Training Epoch: 10/100, batch 183/250 completed (loss: 0.0641:  74%|███████▎  | 92/125 [09:16<03:22,  6.12s/it]
Training Epoch: 10/100, batch 183/250 completed (loss: 0.0454:  74%|███████▎  | 92/125 [09:16<03:22,  6.12s/it]
Training Epoch: 10/100, batch 183/250 completed (loss: 0.0655:  74%|███████▎  | 92/125 [09:16<03:22,  6.12s/it]
Training Epoch: 10/100, batch 184/250 completed (loss: 0.0825:  74%|███████▎  | 92/125 [09:19<03:22,  6.12s/it]
Training Epoch: 10/100, batch 184/250 completed (loss: 0.4045:  74%|███████▎  | 92/125 [09:19<03:22,  6.12s/it]
Training Epoch: 10/100, batch 184/250 completed (loss: 0.0772:  74%|███████▎  | 92/125 [09:19<03:21,  6.12s/it]
Training Epoch: 10/100, batch 184/250 completed (loss: 0.0053:  74%|███████▎  | 92/125 [09:19<03:22,  6.12s/it]
Training Epoch: 10/100, batch 184/250 completed (loss: 0.0772:  74%|███████▍  | 93/125 [09:22<03:15,  6.09s/it]
Training Epoch: 10/100, batch 184/250 completed (loss: 0.0053:  74%|███████▍  | 93/125 [09:22<03:15,  6.09s/it]
Training Epoch: 10/100, batch 184/250 completed (loss: 0.0825:  74%|███████▍  | 93/125 [09:22<03:15,  6.09s/it]
Training Epoch: 10/100, batch 184/250 completed (loss: 0.4045:  74%|███████▍  | 93/125 [09:22<03:15,  6.10s/it]
Training Epoch: 10/100, batch 185/250 completed (loss: 0.0682:  74%|███████▍  | 93/125 [09:22<03:15,  6.10s/it]
Training Epoch: 10/100, batch 185/250 completed (loss: 0.0215:  74%|███████▍  | 93/125 [09:22<03:15,  6.09s/it]
Training Epoch: 10/100, batch 185/250 completed (loss: 0.35:  74%|███████▍  | 93/125 [09:22<03:15,  6.09s/it]  
Training Epoch: 10/100, batch 185/250 completed (loss: 0.0142:  74%|███████▍  | 93/125 [09:22<03:15,  6.09s/it]
Training Epoch: 10/100, batch 186/250 completed (loss: 0.1613:  74%|███████▍  | 93/125 [09:25<03:15,  6.10s/it]
Training Epoch: 10/100, batch 186/250 completed (loss: 0.0485:  74%|███████▍  | 93/125 [09:25<03:15,  6.09s/it]
Training Epoch: 10/100, batch 186/250 completed (loss: 0.0124:  74%|███████▍  | 93/125 [09:25<03:15,  6.09s/it]
Training Epoch: 10/100, batch 186/250 completed (loss: 0.2425:  74%|███████▍  | 93/125 [09:25<03:15,  6.09s/it]
Training Epoch: 10/100, batch 186/250 completed (loss: 0.0124:  75%|███████▌  | 94/125 [09:29<03:10,  6.13s/it]
Training Epoch: 10/100, batch 186/250 completed (loss: 0.0485:  75%|███████▌  | 94/125 [09:29<03:10,  6.13s/it]
Training Epoch: 10/100, batch 186/250 completed (loss: 0.1613:  75%|███████▌  | 94/125 [09:29<03:10,  6.13s/it]
Training Epoch: 10/100, batch 186/250 completed (loss: 0.2425:  75%|███████▌  | 94/125 [09:29<03:10,  6.13s/it]
Training Epoch: 10/100, batch 187/250 completed (loss: 0.0145:  75%|███████▌  | 94/125 [09:29<03:10,  6.13s/it]
Training Epoch: 10/100, batch 187/250 completed (loss: 0.0512:  75%|███████▌  | 94/125 [09:29<03:10,  6.13s/it]
Training Epoch: 10/100, batch 187/250 completed (loss: 0.1545:  75%|███████▌  | 94/125 [09:29<03:10,  6.13s/it]
Training Epoch: 10/100, batch 187/250 completed (loss: 0.0459:  75%|███████▌  | 94/125 [09:29<03:10,  6.13s/it]
Training Epoch: 10/100, batch 188/250 completed (loss: 0.1879:  75%|███████▌  | 94/125 [09:31<03:10,  6.13s/it]
Training Epoch: 10/100, batch 188/250 completed (loss: 0.0398:  75%|███████▌  | 94/125 [09:31<03:10,  6.13s/it]
Training Epoch: 10/100, batch 188/250 completed (loss: 0.211:  75%|███████▌  | 94/125 [09:31<03:10,  6.13s/it] 
Training Epoch: 10/100, batch 188/250 completed (loss: 0.0533:  75%|███████▌  | 94/125 [09:32<03:10,  6.13s/it]
Training Epoch: 10/100, batch 188/250 completed (loss: 0.211:  76%|███████▌  | 95/125 [09:35<03:02,  6.09s/it]
Training Epoch: 10/100, batch 188/250 completed (loss: 0.0398:  76%|███████▌  | 95/125 [09:35<03:02,  6.09s/it]
Training Epoch: 10/100, batch 188/250 completed (loss: 0.0533:  76%|███████▌  | 95/125 [09:35<03:02,  6.09s/it]
Training Epoch: 10/100, batch 188/250 completed (loss: 0.1879:  76%|███████▌  | 95/125 [09:35<03:02,  6.09s/it]
Training Epoch: 10/100, batch 189/250 completed (loss: 0.1285:  76%|███████▌  | 95/125 [09:35<03:02,  6.09s/it]
Training Epoch: 10/100, batch 189/250 completed (loss: 0.0463:  76%|███████▌  | 95/125 [09:35<03:02,  6.09s/it]
Training Epoch: 10/100, batch 189/250 completed (loss: 0.0544:  76%|███████▌  | 95/125 [09:35<03:02,  6.09s/it]
Training Epoch: 10/100, batch 189/250 completed (loss: 0.0424:  76%|███████▌  | 95/125 [09:35<03:02,  6.09s/it]
Training Epoch: 10/100, batch 190/250 completed (loss: 0.0732:  76%|███████▌  | 95/125 [09:37<03:02,  6.09s/it]
Training Epoch: 10/100, batch 190/250 completed (loss: 0.0404:  76%|███████▌  | 95/125 [09:37<03:02,  6.09s/it]
Training Epoch: 10/100, batch 190/250 completed (loss: 0.1784:  76%|███████▌  | 95/125 [09:37<03:02,  6.09s/it]
Training Epoch: 10/100, batch 190/250 completed (loss: 0.0192:  76%|███████▌  | 95/125 [09:38<03:02,  6.09s/it]
Training Epoch: 10/100, batch 190/250 completed (loss: 0.1784:  77%|███████▋  | 96/125 [09:41<02:56,  6.07s/it]
Training Epoch: 10/100, batch 190/250 completed (loss: 0.0732:  77%|███████▋  | 96/125 [09:41<02:56,  6.07s/it]
Training Epoch: 10/100, batch 190/250 completed (loss: 0.0404:  77%|███████▋  | 96/125 [09:41<02:56,  6.07s/it]
Training Epoch: 10/100, batch 190/250 completed (loss: 0.0192:  77%|███████▋  | 96/125 [09:41<02:56,  6.07s/it]
Training Epoch: 10/100, batch 191/250 completed (loss: 0.1173:  77%|███████▋  | 96/125 [09:41<02:56,  6.07s/it]
Training Epoch: 10/100, batch 191/250 completed (loss: 0.0167:  77%|███████▋  | 96/125 [09:41<02:56,  6.07s/it]
Training Epoch: 10/100, batch 191/250 completed (loss: 0.0089:  77%|███████▋  | 96/125 [09:41<02:56,  6.07s/it]
Training Epoch: 10/100, batch 191/250 completed (loss: 0.0066:  77%|███████▋  | 96/125 [09:41<02:56,  6.07s/it]
Training Epoch: 10/100, batch 192/250 completed (loss: 0.0443:  77%|███████▋  | 96/125 [09:43<02:56,  6.07s/it]
Training Epoch: 10/100, batch 192/250 completed (loss: 0.1328:  77%|███████▋  | 96/125 [09:43<02:56,  6.07s/it]
Training Epoch: 10/100, batch 192/250 completed (loss: 0.0845:  77%|███████▋  | 96/125 [09:43<02:56,  6.07s/it]
Training Epoch: 10/100, batch 192/250 completed (loss: 0.0675:  77%|███████▋  | 96/125 [09:44<02:56,  6.07s/it]
Training Epoch: 10/100, batch 192/250 completed (loss: 0.1328:  78%|███████▊  | 97/125 [09:47<02:49,  6.07s/it]
Training Epoch: 10/100, batch 192/250 completed (loss: 0.0443:  78%|███████▊  | 97/125 [09:47<02:49,  6.07s/it]
Training Epoch: 10/100, batch 192/250 completed (loss: 0.0845:  78%|███████▊  | 97/125 [09:47<02:49,  6.07s/it]
Training Epoch: 10/100, batch 192/250 completed (loss: 0.0675:  78%|███████▊  | 97/125 [09:47<02:49,  6.07s/it]
Training Epoch: 10/100, batch 193/250 completed (loss: 0.1095:  78%|███████▊  | 97/125 [09:47<02:49,  6.07s/it]
Training Epoch: 10/100, batch 193/250 completed (loss: 0.0404:  78%|███████▊  | 97/125 [09:47<02:49,  6.07s/it]
Training Epoch: 10/100, batch 193/250 completed (loss: 0.1107:  78%|███████▊  | 97/125 [09:47<02:49,  6.07s/it]
Training Epoch: 10/100, batch 193/250 completed (loss: 0.117:  78%|███████▊  | 97/125 [09:47<02:49,  6.07s/it] 
Training Epoch: 10/100, batch 194/250 completed (loss: 0.0135:  78%|███████▊  | 97/125 [09:50<02:49,  6.07s/it]
Training Epoch: 10/100, batch 194/250 completed (loss: 0.0108:  78%|███████▊  | 97/125 [09:50<02:49,  6.07s/it]
Training Epoch: 10/100, batch 194/250 completed (loss: 0.0308:  78%|███████▊  | 97/125 [09:50<02:49,  6.07s/it]
Training Epoch: 10/100, batch 194/250 completed (loss: 0.2361:  78%|███████▊  | 97/125 [09:50<02:49,  6.07s/it]
Training Epoch: 10/100, batch 194/250 completed (loss: 0.0108:  78%|███████▊  | 98/125 [09:53<02:43,  6.05s/it]
Training Epoch: 10/100, batch 194/250 completed (loss: 0.2361:  78%|███████▊  | 98/125 [09:53<02:43,  6.05s/it]
Training Epoch: 10/100, batch 194/250 completed (loss: 0.0308:  78%|███████▊  | 98/125 [09:53<02:43,  6.05s/it]
Training Epoch: 10/100, batch 194/250 completed (loss: 0.0135:  78%|███████▊  | 98/125 [09:53<02:43,  6.05s/it]
Training Epoch: 10/100, batch 195/250 completed (loss: 0.1417:  78%|███████▊  | 98/125 [09:53<02:43,  6.05s/it]
Training Epoch: 10/100, batch 195/250 completed (loss: 0.1046:  78%|███████▊  | 98/125 [09:53<02:43,  6.05s/it]
Training Epoch: 10/100, batch 195/250 completed (loss: 0.0983:  78%|███████▊  | 98/125 [09:53<02:43,  6.05s/it]
Training Epoch: 10/100, batch 195/250 completed (loss: 0.0109:  78%|███████▊  | 98/125 [09:53<02:43,  6.05s/it]
Training Epoch: 10/100, batch 196/250 completed (loss: 0.055:  78%|███████▊  | 98/125 [09:56<02:43,  6.05s/it] 
Training Epoch: 10/100, batch 196/250 completed (loss: 0.0247:  78%|███████▊  | 98/125 [09:56<02:43,  6.05s/it]
Training Epoch: 10/100, batch 196/250 completed (loss: 0.0506:  78%|███████▊  | 98/125 [09:56<02:43,  6.05s/it]
Training Epoch: 10/100, batch 196/250 completed (loss: 0.1005:  78%|███████▊  | 98/125 [09:56<02:43,  6.05s/it]
Training Epoch: 10/100, batch 196/250 completed (loss: 0.1005:  79%|███████▉  | 99/125 [09:59<02:36,  6.03s/it]
Training Epoch: 10/100, batch 196/250 completed (loss: 0.055:  79%|███████▉  | 99/125 [09:59<02:36,  6.04s/it]
Training Epoch: 10/100, batch 196/250 completed (loss: 0.0247:  79%|███████▉  | 99/125 [09:59<02:36,  6.03s/it]
Training Epoch: 10/100, batch 196/250 completed (loss: 0.0506:  79%|███████▉  | 99/125 [09:59<02:36,  6.04s/it]
Training Epoch: 10/100, batch 197/250 completed (loss: 0.1597:  79%|███████▉  | 99/125 [09:59<02:36,  6.03s/it]
Training Epoch: 10/100, batch 197/250 completed (loss: 0.0684:  79%|███████▉  | 99/125 [09:59<02:36,  6.04s/it]
Training Epoch: 10/100, batch 197/250 completed (loss: 0.0534:  79%|███████▉  | 99/125 [09:59<02:36,  6.04s/it]
Training Epoch: 10/100, batch 197/250 completed (loss: 0.099:  79%|███████▉  | 99/125 [09:59<02:36,  6.03s/it] 
Training Epoch: 10/100, batch 198/250 completed (loss: 0.0821:  79%|███████▉  | 99/125 [10:02<02:36,  6.04s/it]
Training Epoch: 10/100, batch 198/250 completed (loss: 0.0951:  79%|███████▉  | 99/125 [10:02<02:36,  6.04s/it]
Training Epoch: 10/100, batch 198/250 completed (loss: 0.1287:  79%|███████▉  | 99/125 [10:02<02:36,  6.03s/it]
Training Epoch: 10/100, batch 198/250 completed (loss: 0.0402:  79%|███████▉  | 99/125 [10:02<02:36,  6.03s/it]
Training Epoch: 10/100, batch 198/250 completed (loss: 0.0951:  80%|████████  | 100/125 [10:05<02:32,  6.09s/it]
Training Epoch: 10/100, batch 198/250 completed (loss: 0.0821:  80%|████████  | 100/125 [10:05<02:32,  6.09s/it]
Training Epoch: 10/100, batch 198/250 completed (loss: 0.0402:  80%|████████  | 100/125 [10:05<02:32,  6.10s/it]
Training Epoch: 10/100, batch 198/250 completed (loss: 0.1287:  80%|████████  | 100/125 [10:05<02:32,  6.10s/it]
Training Epoch: 10/100, batch 199/250 completed (loss: 0.1425:  80%|████████  | 100/125 [10:05<02:32,  6.09s/it]
Training Epoch: 10/100, batch 199/250 completed (loss: 0.0168:  80%|████████  | 100/125 [10:05<02:32,  6.10s/it]
Training Epoch: 10/100, batch 199/250 completed (loss: 0.0438:  80%|████████  | 100/125 [10:05<02:32,  6.09s/it]
Training Epoch: 10/100, batch 199/250 completed (loss: 0.0265:  80%|████████  | 100/125 [10:05<02:32,  6.10s/it]
Training Epoch: 10/100, batch 200/250 completed (loss: 0.0221:  80%|████████  | 100/125 [10:08<02:32,  6.09s/it]
Training Epoch: 10/100, batch 200/250 completed (loss: 0.0043:  80%|████████  | 100/125 [10:08<02:32,  6.10s/it]
Training Epoch: 10/100, batch 200/250 completed (loss: 0.1581:  80%|████████  | 100/125 [10:08<02:32,  6.09s/it]
Training Epoch: 10/100, batch 200/250 completed (loss: 0.1183:  80%|████████  | 100/125 [10:08<02:32,  6.10s/it]
Training Epoch: 10/100, batch 200/250 completed (loss: 0.0221:  81%|████████  | 101/125 [10:11<02:25,  6.08s/it]
Training Epoch: 10/100, batch 200/250 completed (loss: 0.1183:  81%|████████  | 101/125 [10:11<02:25,  6.08s/it]
Training Epoch: 10/100, batch 200/250 completed (loss: 0.1581:  81%|████████  | 101/125 [10:11<02:25,  6.08s/it]
Training Epoch: 10/100, batch 200/250 completed (loss: 0.0043:  81%|████████  | 101/125 [10:11<02:25,  6.08s/it]
Training Epoch: 10/100, batch 201/250 completed (loss: 0.0203:  81%|████████  | 101/125 [10:11<02:25,  6.08s/it]
Training Epoch: 10/100, batch 201/250 completed (loss: 0.087:  81%|████████  | 101/125 [10:11<02:25,  6.08s/it] 
Training Epoch: 10/100, batch 201/250 completed (loss: 0.1588:  81%|████████  | 101/125 [10:11<02:25,  6.08s/it]
Training Epoch: 10/100, batch 201/250 completed (loss: 0.1887:  81%|████████  | 101/125 [10:11<02:25,  6.08s/it]
Training Epoch: 10/100, batch 202/250 completed (loss: 0.0377:  81%|████████  | 101/125 [10:14<02:25,  6.08s/it]
Training Epoch: 10/100, batch 202/250 completed (loss: 0.0379:  81%|████████  | 101/125 [10:14<02:25,  6.08s/it]
Training Epoch: 10/100, batch 202/250 completed (loss: 0.0652:  81%|████████  | 101/125 [10:14<02:25,  6.08s/it]
Training Epoch: 10/100, batch 202/250 completed (loss: 0.0413:  81%|████████  | 101/125 [10:14<02:25,  6.08s/it]
Training Epoch: 10/100, batch 202/250 completed (loss: 0.0652:  82%|████████▏ | 102/125 [10:17<02:21,  6.15s/it]
Training Epoch: 10/100, batch 202/250 completed (loss: 0.0413:  82%|████████▏ | 102/125 [10:17<02:21,  6.15s/it]
Training Epoch: 10/100, batch 202/250 completed (loss: 0.0377:  82%|████████▏ | 102/125 [10:17<02:21,  6.15s/it]
Training Epoch: 10/100, batch 202/250 completed (loss: 0.0379:  82%|████████▏ | 102/125 [10:17<02:21,  6.15s/it]
Training Epoch: 10/100, batch 203/250 completed (loss: 0.0214:  82%|████████▏ | 102/125 [10:17<02:21,  6.15s/it]
Training Epoch: 10/100, batch 203/250 completed (loss: 0.2122:  82%|████████▏ | 102/125 [10:17<02:21,  6.15s/it]
Training Epoch: 10/100, batch 203/250 completed (loss: 0.0496:  82%|████████▏ | 102/125 [10:17<02:21,  6.15s/it]
Training Epoch: 10/100, batch 203/250 completed (loss: 0.2599:  82%|████████▏ | 102/125 [10:17<02:21,  6.15s/it]
Training Epoch: 10/100, batch 204/250 completed (loss: 0.0167:  82%|████████▏ | 102/125 [10:20<02:21,  6.15s/it]
Training Epoch: 10/100, batch 204/250 completed (loss: 0.0501:  82%|████████▏ | 102/125 [10:20<02:21,  6.15s/it]
Training Epoch: 10/100, batch 204/250 completed (loss: 0.1005:  82%|████████▏ | 102/125 [10:20<02:21,  6.15s/it]
Training Epoch: 10/100, batch 204/250 completed (loss: 0.2448:  82%|████████▏ | 102/125 [10:20<02:21,  6.15s/it]
Training Epoch: 10/100, batch 204/250 completed (loss: 0.0501:  82%|████████▏ | 103/125 [10:23<02:14,  6.11s/it]
Training Epoch: 10/100, batch 204/250 completed (loss: 0.1005:  82%|████████▏ | 103/125 [10:23<02:14,  6.11s/it]
Training Epoch: 10/100, batch 204/250 completed (loss: 0.2448:  82%|████████▏ | 103/125 [10:23<02:14,  6.11s/it]
Training Epoch: 10/100, batch 204/250 completed (loss: 0.0167:  82%|████████▏ | 103/125 [10:23<02:14,  6.12s/it]
Training Epoch: 10/100, batch 205/250 completed (loss: 0.1361:  82%|████████▏ | 103/125 [10:23<02:14,  6.11s/it]
Training Epoch: 10/100, batch 205/250 completed (loss: 0.1899:  82%|████████▏ | 103/125 [10:23<02:14,  6.11s/it]
Training Epoch: 10/100, batch 205/250 completed (loss: 0.1285:  82%|████████▏ | 103/125 [10:23<02:14,  6.12s/it]
Training Epoch: 10/100, batch 205/250 completed (loss: 0.3361:  82%|████████▏ | 103/125 [10:23<02:14,  6.11s/it]
Training Epoch: 10/100, batch 206/250 completed (loss: 0.1172:  82%|████████▏ | 103/125 [10:26<02:14,  6.11s/it]
Training Epoch: 10/100, batch 206/250 completed (loss: 0.0298:  82%|████████▏ | 103/125 [10:26<02:14,  6.11s/it]
Training Epoch: 10/100, batch 206/250 completed (loss: 0.3396:  82%|████████▏ | 103/125 [10:26<02:14,  6.12s/it]
Training Epoch: 10/100, batch 206/250 completed (loss: 0.156:  82%|████████▏ | 103/125 [10:26<02:14,  6.11s/it] 
Training Epoch: 10/100, batch 206/250 completed (loss: 0.0298:  83%|████████▎ | 104/125 [10:29<02:07,  6.09s/it]
Training Epoch: 10/100, batch 206/250 completed (loss: 0.156:  83%|████████▎ | 104/125 [10:29<02:07,  6.09s/it]
Training Epoch: 10/100, batch 206/250 completed (loss: 0.1172:  83%|████████▎ | 104/125 [10:29<02:07,  6.09s/it]
Training Epoch: 10/100, batch 206/250 completed (loss: 0.3396:  83%|████████▎ | 104/125 [10:29<02:07,  6.09s/it]
Training Epoch: 10/100, batch 207/250 completed (loss: 0.1027:  83%|████████▎ | 104/125 [10:29<02:07,  6.09s/it]
Training Epoch: 10/100, batch 207/250 completed (loss: 0.024:  83%|████████▎ | 104/125 [10:29<02:07,  6.09s/it] 
Training Epoch: 10/100, batch 207/250 completed (loss: 0.1673:  83%|████████▎ | 104/125 [10:29<02:07,  6.09s/it]
Training Epoch: 10/100, batch 207/250 completed (loss: 0.0606:  83%|████████▎ | 104/125 [10:29<02:07,  6.09s/it]
Training Epoch: 10/100, batch 208/250 completed (loss: 0.0396:  83%|████████▎ | 104/125 [10:32<02:07,  6.09s/it]
Training Epoch: 10/100, batch 208/250 completed (loss: 0.079:  83%|████████▎ | 104/125 [10:32<02:07,  6.09s/it]
Training Epoch: 10/100, batch 208/250 completed (loss: 0.3236:  83%|████████▎ | 104/125 [10:32<02:07,  6.09s/it]
Training Epoch: 10/100, batch 208/250 completed (loss: 0.0216:  83%|████████▎ | 104/125 [10:32<02:07,  6.09s/it]
Training Epoch: 10/100, batch 208/250 completed (loss: 0.3236:  84%|████████▍ | 105/125 [10:35<02:02,  6.14s/it]
Training Epoch: 10/100, batch 208/250 completed (loss: 0.079:  84%|████████▍ | 105/125 [10:35<02:02,  6.14s/it]
Training Epoch: 10/100, batch 208/250 completed (loss: 0.0216:  84%|████████▍ | 105/125 [10:36<02:02,  6.14s/it]
Training Epoch: 10/100, batch 208/250 completed (loss: 0.0396:  84%|████████▍ | 105/125 [10:36<02:02,  6.14s/it]
Training Epoch: 10/100, batch 209/250 completed (loss: 0.1217:  84%|████████▍ | 105/125 [10:36<02:02,  6.14s/it]
Training Epoch: 10/100, batch 209/250 completed (loss: 0.0209:  84%|████████▍ | 105/125 [10:36<02:02,  6.14s/it]
Training Epoch: 10/100, batch 209/250 completed (loss: 0.0572:  84%|████████▍ | 105/125 [10:36<02:02,  6.14s/it]
Training Epoch: 10/100, batch 209/250 completed (loss: 0.0784:  84%|████████▍ | 105/125 [10:36<02:02,  6.14s/it]
Training Epoch: 10/100, batch 210/250 completed (loss: 0.008:  84%|████████▍ | 105/125 [10:38<02:02,  6.14s/it] 
Training Epoch: 10/100, batch 210/250 completed (loss: 0.0601:  84%|████████▍ | 105/125 [10:38<02:02,  6.14s/it]
Training Epoch: 10/100, batch 210/250 completed (loss: 0.0988:  84%|████████▍ | 105/125 [10:38<02:02,  6.14s/it]
Training Epoch: 10/100, batch 210/250 completed (loss: 0.2813:  84%|████████▍ | 105/125 [10:39<02:02,  6.14s/it]
Training Epoch: 10/100, batch 210/250 completed (loss: 0.0988:  85%|████████▍ | 106/125 [10:42<01:55,  6.10s/it]
Training Epoch: 10/100, batch 210/250 completed (loss: 0.008:  85%|████████▍ | 106/125 [10:42<01:55,  6.10s/it]
Training Epoch: 10/100, batch 210/250 completed (loss: 0.2813:  85%|████████▍ | 106/125 [10:42<01:55,  6.10s/it]
Training Epoch: 10/100, batch 210/250 completed (loss: 0.0601:  85%|████████▍ | 106/125 [10:42<01:56,  6.11s/it]
Training Epoch: 10/100, batch 211/250 completed (loss: 0.054:  85%|████████▍ | 106/125 [10:42<01:55,  6.10s/it]
Training Epoch: 10/100, batch 211/250 completed (loss: 0.1863:  85%|████████▍ | 106/125 [10:42<01:55,  6.10s/it]
Training Epoch: 10/100, batch 211/250 completed (loss: 0.2067:  85%|████████▍ | 106/125 [10:42<01:56,  6.11s/it]
Training Epoch: 10/100, batch 211/250 completed (loss: 0.1294:  85%|████████▍ | 106/125 [10:42<01:55,  6.10s/it]
Training Epoch: 10/100, batch 212/250 completed (loss: 0.1056:  85%|████████▍ | 106/125 [10:45<01:55,  6.10s/it]
Training Epoch: 10/100, batch 212/250 completed (loss: 0.063:  85%|████████▍ | 106/125 [10:45<01:55,  6.10s/it] 
Training Epoch: 10/100, batch 212/250 completed (loss: 0.0359:  85%|████████▍ | 106/125 [10:45<01:56,  6.11s/it]
Training Epoch: 10/100, batch 212/250 completed (loss: 0.1684:  85%|████████▍ | 106/125 [10:45<01:55,  6.10s/it]
Training Epoch: 10/100, batch 212/250 completed (loss: 0.063:  86%|████████▌ | 107/125 [10:48<01:49,  6.08s/it]
Training Epoch: 10/100, batch 212/250 completed (loss: 0.1056:  86%|████████▌ | 107/125 [10:48<01:49,  6.08s/it]
Training Epoch: 10/100, batch 212/250 completed (loss: 0.0359:  86%|████████▌ | 107/125 [10:48<01:49,  6.08s/it]
Training Epoch: 10/100, batch 212/250 completed (loss: 0.1684:  86%|████████▌ | 107/125 [10:48<01:49,  6.08s/it]
Training Epoch: 10/100, batch 213/250 completed (loss: 0.008:  86%|████████▌ | 107/125 [10:48<01:49,  6.08s/it] 
Training Epoch: 10/100, batch 213/250 completed (loss: 0.0213:  86%|████████▌ | 107/125 [10:48<01:49,  6.08s/it]
Training Epoch: 10/100, batch 213/250 completed (loss: 0.1287:  86%|████████▌ | 107/125 [10:48<01:49,  6.08s/it]
Training Epoch: 10/100, batch 213/250 completed (loss: 0.0675:  86%|████████▌ | 107/125 [10:48<01:49,  6.08s/it]
Training Epoch: 10/100, batch 214/250 completed (loss: 0.1001:  86%|████████▌ | 107/125 [10:51<01:49,  6.08s/it]
Training Epoch: 10/100, batch 214/250 completed (loss: 0.0215:  86%|████████▌ | 107/125 [10:51<01:49,  6.08s/it]
Training Epoch: 10/100, batch 214/250 completed (loss: 0.2959:  86%|████████▌ | 107/125 [10:51<01:49,  6.08s/it]
Training Epoch: 10/100, batch 214/250 completed (loss: 0.0455:  86%|████████▌ | 107/125 [10:51<01:49,  6.08s/it]
Training Epoch: 10/100, batch 214/250 completed (loss: 0.1001:  86%|████████▋ | 108/125 [10:54<01:43,  6.11s/it]
Training Epoch: 10/100, batch 214/250 completed (loss: 0.2959:  86%|████████▋ | 108/125 [10:54<01:43,  6.10s/it]
Training Epoch: 10/100, batch 214/250 completed (loss: 0.0215:  86%|████████▋ | 108/125 [10:54<01:43,  6.10s/it]
Training Epoch: 10/100, batch 214/250 completed (loss: 0.0455:  86%|████████▋ | 108/125 [10:54<01:43,  6.11s/it]
Training Epoch: 10/100, batch 215/250 completed (loss: 0.1377:  86%|████████▋ | 108/125 [10:54<01:43,  6.10s/it]
Training Epoch: 10/100, batch 215/250 completed (loss: 0.1413:  86%|████████▋ | 108/125 [10:54<01:43,  6.11s/it]
Training Epoch: 10/100, batch 215/250 completed (loss: 0.0899:  86%|████████▋ | 108/125 [10:54<01:43,  6.10s/it]
Training Epoch: 10/100, batch 215/250 completed (loss: 0.2969:  86%|████████▋ | 108/125 [10:54<01:43,  6.11s/it]
Training Epoch: 10/100, batch 216/250 completed (loss: 0.1875:  86%|████████▋ | 108/125 [10:57<01:43,  6.10s/it]
Training Epoch: 10/100, batch 216/250 completed (loss: 0.0578:  86%|████████▋ | 108/125 [10:57<01:43,  6.10s/it]
Training Epoch: 10/100, batch 216/250 completed (loss: 0.0465:  86%|████████▋ | 108/125 [10:57<01:43,  6.11s/it]
Training Epoch: 10/100, batch 216/250 completed (loss: 0.0304:  86%|████████▋ | 108/125 [10:57<01:43,  6.11s/it]
Training Epoch: 10/100, batch 216/250 completed (loss: 0.0578:  87%|████████▋ | 109/125 [11:00<01:37,  6.12s/it]
Training Epoch: 10/100, batch 216/250 completed (loss: 0.0465:  87%|████████▋ | 109/125 [11:00<01:37,  6.12s/it]
Training Epoch: 10/100, batch 216/250 completed (loss: 0.0304:  87%|████████▋ | 109/125 [11:00<01:37,  6.12s/it]
Training Epoch: 10/100, batch 216/250 completed (loss: 0.1875:  87%|████████▋ | 109/125 [11:00<01:37,  6.12s/it]
Training Epoch: 10/100, batch 217/250 completed (loss: 0.0164:  87%|████████▋ | 109/125 [11:00<01:37,  6.12s/it]
Training Epoch: 10/100, batch 217/250 completed (loss: 0.2269:  87%|████████▋ | 109/125 [11:00<01:37,  6.12s/it]
Training Epoch: 10/100, batch 217/250 completed (loss: 0.094:  87%|████████▋ | 109/125 [11:00<01:37,  6.12s/it] 
Training Epoch: 10/100, batch 217/250 completed (loss: 0.1172:  87%|████████▋ | 109/125 [11:00<01:37,  6.12s/it]
Training Epoch: 10/100, batch 218/250 completed (loss: 0.0415:  87%|████████▋ | 109/125 [11:03<01:37,  6.12s/it]
Training Epoch: 10/100, batch 218/250 completed (loss: 0.0613:  87%|████████▋ | 109/125 [11:03<01:37,  6.12s/it]
Training Epoch: 10/100, batch 218/250 completed (loss: 0.0562:  87%|████████▋ | 109/125 [11:03<01:37,  6.12s/it]
Training Epoch: 10/100, batch 218/250 completed (loss: 0.0399:  87%|████████▋ | 109/125 [11:03<01:37,  6.12s/it]
Training Epoch: 10/100, batch 218/250 completed (loss: 0.0415:  88%|████████▊ | 110/125 [11:06<01:31,  6.09s/it]
Training Epoch: 10/100, batch 218/250 completed (loss: 0.0613:  88%|████████▊ | 110/125 [11:06<01:31,  6.09s/it]
Training Epoch: 10/100, batch 218/250 completed (loss: 0.0562:  88%|████████▊ | 110/125 [11:06<01:31,  6.09s/it]
Training Epoch: 10/100, batch 218/250 completed (loss: 0.0399:  88%|████████▊ | 110/125 [11:06<01:31,  6.09s/it]
Training Epoch: 10/100, batch 219/250 completed (loss: 0.1171:  88%|████████▊ | 110/125 [11:06<01:31,  6.09s/it]
Training Epoch: 10/100, batch 219/250 completed (loss: 0.0288:  88%|████████▊ | 110/125 [11:06<01:31,  6.09s/it]
Training Epoch: 10/100, batch 219/250 completed (loss: 0.0302:  88%|████████▊ | 110/125 [11:06<01:31,  6.09s/it]
Training Epoch: 10/100, batch 219/250 completed (loss: 0.1392:  88%|████████▊ | 110/125 [11:06<01:31,  6.09s/it]
Training Epoch: 10/100, batch 220/250 completed (loss: 0.0875:  88%|████████▊ | 110/125 [11:09<01:31,  6.09s/it]
Training Epoch: 10/100, batch 220/250 completed (loss: 0.0725:  88%|████████▊ | 110/125 [11:09<01:31,  6.09s/it]
Training Epoch: 10/100, batch 220/250 completed (loss: 0.1497:  88%|████████▊ | 110/125 [11:09<01:31,  6.09s/it]
Training Epoch: 10/100, batch 220/250 completed (loss: 0.028:  88%|████████▊ | 110/125 [11:09<01:31,  6.09s/it] 
Training Epoch: 10/100, batch 220/250 completed (loss: 0.1497:  89%|████████▉ | 111/125 [11:12<01:25,  6.12s/it]
Training Epoch: 10/100, batch 220/250 completed (loss: 0.028:  89%|████████▉ | 111/125 [11:12<01:25,  6.11s/it]
Training Epoch: 10/100, batch 220/250 completed (loss: 0.0725:  89%|████████▉ | 111/125 [11:12<01:25,  6.11s/it]
Training Epoch: 10/100, batch 220/250 completed (loss: 0.0875:  89%|████████▉ | 111/125 [11:12<01:25,  6.11s/it]
Training Epoch: 10/100, batch 221/250 completed (loss: 0.1232:  89%|████████▉ | 111/125 [11:12<01:25,  6.12s/it]
Training Epoch: 10/100, batch 221/250 completed (loss: 0.042:  89%|████████▉ | 111/125 [11:12<01:25,  6.11s/it] 
Training Epoch: 10/100, batch 221/250 completed (loss: 0.0677:  89%|████████▉ | 111/125 [11:12<01:25,  6.11s/it]
Training Epoch: 10/100, batch 221/250 completed (loss: 0.2656:  89%|████████▉ | 111/125 [11:12<01:25,  6.11s/it]
Training Epoch: 10/100, batch 222/250 completed (loss: 0.2508:  89%|████████▉ | 111/125 [11:15<01:25,  6.11s/it]
Training Epoch: 10/100, batch 222/250 completed (loss: 0.0402:  89%|████████▉ | 111/125 [11:15<01:25,  6.12s/it]
Training Epoch: 10/100, batch 222/250 completed (loss: 0.0225:  89%|████████▉ | 111/125 [11:15<01:25,  6.11s/it]
Training Epoch: 10/100, batch 222/250 completed (loss: 0.0171:  89%|████████▉ | 111/125 [11:15<01:25,  6.11s/it]
Training Epoch: 10/100, batch 222/250 completed (loss: 0.0402:  90%|████████▉ | 112/125 [11:18<01:19,  6.08s/it]
Training Epoch: 10/100, batch 222/250 completed (loss: 0.0171:  90%|████████▉ | 112/125 [11:18<01:19,  6.08s/it]
Training Epoch: 10/100, batch 222/250 completed (loss: 0.0225:  90%|████████▉ | 112/125 [11:18<01:19,  6.09s/it]
Training Epoch: 10/100, batch 222/250 completed (loss: 0.2508:  90%|████████▉ | 112/125 [11:18<01:19,  6.09s/it]
Training Epoch: 10/100, batch 223/250 completed (loss: 0.0325:  90%|████████▉ | 112/125 [11:18<01:19,  6.08s/it]
Training Epoch: 10/100, batch 223/250 completed (loss: 0.0899:  90%|████████▉ | 112/125 [11:18<01:19,  6.09s/it]
Training Epoch: 10/100, batch 223/250 completed (loss: 0.046:  90%|████████▉ | 112/125 [11:18<01:19,  6.09s/it] 
Training Epoch: 10/100, batch 223/250 completed (loss: 0.1816:  90%|████████▉ | 112/125 [11:18<01:19,  6.08s/it]
Training Epoch: 10/100, batch 224/250 completed (loss: 0.0916:  90%|████████▉ | 112/125 [11:21<01:19,  6.09s/it]
Training Epoch: 10/100, batch 224/250 completed (loss: 0.0592:  90%|████████▉ | 112/125 [11:21<01:19,  6.09s/it]
Training Epoch: 10/100, batch 224/250 completed (loss: 0.1774:  90%|████████▉ | 112/125 [11:21<01:19,  6.08s/it]
Training Epoch: 10/100, batch 224/250 completed (loss: 0.061:  90%|████████▉ | 112/125 [11:21<01:19,  6.08s/it] 
Training Epoch: 10/100, batch 224/250 completed (loss: 0.1774:  90%|█████████ | 113/125 [11:24<01:13,  6.12s/it]
Training Epoch: 10/100, batch 224/250 completed (loss: 0.061:  90%|█████████ | 113/125 [11:24<01:13,  6.12s/it]
Training Epoch: 10/100, batch 224/250 completed (loss: 0.0916:  90%|█████████ | 113/125 [11:24<01:13,  6.12s/it]
Training Epoch: 10/100, batch 224/250 completed (loss: 0.0592:  90%|█████████ | 113/125 [11:24<01:13,  6.12s/it]
Training Epoch: 10/100, batch 225/250 completed (loss: 0.0137:  90%|█████████ | 113/125 [11:24<01:13,  6.12s/it]
Training Epoch: 10/100, batch 225/250 completed (loss: 0.0349:  90%|█████████ | 113/125 [11:24<01:13,  6.12s/it]
Training Epoch: 10/100, batch 225/250 completed (loss: 0.0978:  90%|█████████ | 113/125 [11:24<01:13,  6.12s/it]
Training Epoch: 10/100, batch 225/250 completed (loss: 0.0132:  90%|█████████ | 113/125 [11:24<01:13,  6.12s/it]
Training Epoch: 10/100, batch 226/250 completed (loss: 0.0908:  90%|█████████ | 113/125 [11:27<01:13,  6.12s/it]
Training Epoch: 10/100, batch 226/250 completed (loss: 0.0156:  90%|█████████ | 113/125 [11:27<01:13,  6.12s/it]
Training Epoch: 10/100, batch 226/250 completed (loss: 0.0737:  90%|█████████ | 113/125 [11:27<01:13,  6.12s/it]
Training Epoch: 10/100, batch 226/250 completed (loss: 0.1169:  90%|█████████ | 113/125 [11:27<01:13,  6.12s/it]
Training Epoch: 10/100, batch 226/250 completed (loss: 0.0156:  91%|█████████ | 114/125 [11:30<01:07,  6.13s/it]
Training Epoch: 10/100, batch 226/250 completed (loss: 0.1169:  91%|█████████ | 114/125 [11:30<01:07,  6.13s/it]
Training Epoch: 10/100, batch 226/250 completed (loss: 0.0737:  91%|█████████ | 114/125 [11:30<01:07,  6.13s/it]
Training Epoch: 10/100, batch 226/250 completed (loss: 0.0908:  91%|█████████ | 114/125 [11:30<01:07,  6.13s/it]
Training Epoch: 10/100, batch 227/250 completed (loss: 0.0525:  91%|█████████ | 114/125 [11:30<01:07,  6.13s/it]
Training Epoch: 10/100, batch 227/250 completed (loss: 0.0953:  91%|█████████ | 114/125 [11:30<01:07,  6.13s/it]
Training Epoch: 10/100, batch 227/250 completed (loss: 0.0364:  91%|█████████ | 114/125 [11:30<01:07,  6.13s/it]
Training Epoch: 10/100, batch 227/250 completed (loss: 0.0439:  91%|█████████ | 114/125 [11:31<01:07,  6.13s/it]
Training Epoch: 10/100, batch 228/250 completed (loss: 0.1685:  91%|█████████ | 114/125 [11:33<01:07,  6.13s/it]
Training Epoch: 10/100, batch 228/250 completed (loss: 0.0644:  91%|█████████ | 114/125 [11:33<01:07,  6.13s/it]
Training Epoch: 10/100, batch 228/250 completed (loss: 0.033:  91%|█████████ | 114/125 [11:33<01:07,  6.13s/it] 
Training Epoch: 10/100, batch 228/250 completed (loss: 0.0542:  91%|█████████ | 114/125 [11:33<01:07,  6.13s/it]
Training Epoch: 10/100, batch 228/250 completed (loss: 0.1685:  92%|█████████▏| 115/125 [11:36<01:01,  6.10s/it]
Training Epoch: 10/100, batch 228/250 completed (loss: 0.0542:  92%|█████████▏| 115/125 [11:36<01:01,  6.10s/it]
Training Epoch: 10/100, batch 228/250 completed (loss: 0.033:  92%|█████████▏| 115/125 [11:36<01:01,  6.11s/it]
Training Epoch: 10/100, batch 228/250 completed (loss: 0.0644:  92%|█████████▏| 115/125 [11:36<01:01,  6.11s/it]
Training Epoch: 10/100, batch 229/250 completed (loss: 0.4177:  92%|█████████▏| 115/125 [11:37<01:01,  6.11s/it]
Training Epoch: 10/100, batch 229/250 completed (loss: 0.0815:  92%|█████████▏| 115/125 [11:37<01:01,  6.11s/it]
Training Epoch: 10/100, batch 229/250 completed (loss: 0.0517:  92%|█████████▏| 115/125 [11:37<01:01,  6.10s/it]
Training Epoch: 10/100, batch 229/250 completed (loss: 0.0548:  92%|█████████▏| 115/125 [11:37<01:01,  6.10s/it]
Training Epoch: 10/100, batch 230/250 completed (loss: 0.0211:  92%|█████████▏| 115/125 [11:40<01:01,  6.11s/it]
Training Epoch: 10/100, batch 230/250 completed (loss: 0.1011:  92%|█████████▏| 115/125 [11:40<01:01,  6.10s/it]
Training Epoch: 10/100, batch 230/250 completed (loss: 0.1367:  92%|█████████▏| 115/125 [11:40<01:01,  6.11s/it]
Training Epoch: 10/100, batch 230/250 completed (loss: 0.2712:  92%|█████████▏| 115/125 [11:40<01:01,  6.10s/it]
Training Epoch: 10/100, batch 230/250 completed (loss: 0.2712:  93%|█████████▎| 116/125 [11:43<00:55,  6.15s/it]
Training Epoch: 10/100, batch 230/250 completed (loss: 0.0211:  93%|█████████▎| 116/125 [11:43<00:55,  6.14s/it]
Training Epoch: 10/100, batch 230/250 completed (loss: 0.1011:  93%|█████████▎| 116/125 [11:43<00:55,  6.15s/it]
Training Epoch: 10/100, batch 230/250 completed (loss: 0.1367:  93%|█████████▎| 116/125 [11:43<00:55,  6.15s/it]
Training Epoch: 10/100, batch 231/250 completed (loss: 0.0396:  93%|█████████▎| 116/125 [11:43<00:55,  6.15s/it]
Training Epoch: 10/100, batch 231/250 completed (loss: 0.1274:  93%|█████████▎| 116/125 [11:43<00:55,  6.15s/it]
Training Epoch: 10/100, batch 231/250 completed (loss: 0.0673:  93%|█████████▎| 116/125 [11:43<00:55,  6.14s/it]
Training Epoch: 10/100, batch 231/250 completed (loss: 0.0453:  93%|█████████▎| 116/125 [11:43<00:55,  6.15s/it]
Training Epoch: 10/100, batch 232/250 completed (loss: 0.1015:  93%|█████████▎| 116/125 [11:46<00:55,  6.15s/it]
Training Epoch: 10/100, batch 232/250 completed (loss: 0.0376:  93%|█████████▎| 116/125 [11:46<00:55,  6.15s/it]
Training Epoch: 10/100, batch 232/250 completed (loss: 0.0162:  93%|█████████▎| 116/125 [11:46<00:55,  6.14s/it]
Training Epoch: 10/100, batch 232/250 completed (loss: 0.0362:  93%|█████████▎| 116/125 [11:46<00:55,  6.15s/it]
Training Epoch: 10/100, batch 232/250 completed (loss: 0.1015:  94%|█████████▎| 117/125 [11:49<00:48,  6.10s/it]
Training Epoch: 10/100, batch 232/250 completed (loss: 0.0362:  94%|█████████▎| 117/125 [11:49<00:48,  6.11s/it]
Training Epoch: 10/100, batch 232/250 completed (loss: 0.0376:  94%|█████████▎| 117/125 [11:49<00:48,  6.11s/it]
Training Epoch: 10/100, batch 232/250 completed (loss: 0.0162:  94%|█████████▎| 117/125 [11:49<00:48,  6.11s/it]
Training Epoch: 10/100, batch 233/250 completed (loss: 0.0046:  94%|█████████▎| 117/125 [11:49<00:48,  6.11s/it]
Training Epoch: 10/100, batch 233/250 completed (loss: 0.1885:  94%|█████████▎| 117/125 [11:49<00:48,  6.11s/it]
Training Epoch: 10/100, batch 233/250 completed (loss: 0.4989:  94%|█████████▎| 117/125 [11:49<00:48,  6.10s/it]
Training Epoch: 10/100, batch 233/250 completed (loss: 0.0225:  94%|█████████▎| 117/125 [11:49<00:48,  6.11s/it]
Training Epoch: 10/100, batch 234/250 completed (loss: 0.1537:  94%|█████████▎| 117/125 [11:52<00:48,  6.10s/it]
Training Epoch: 10/100, batch 234/250 completed (loss: 0.0235:  94%|█████████▎| 117/125 [11:52<00:48,  6.11s/it]
Training Epoch: 10/100, batch 234/250 completed (loss: 0.0245:  94%|█████████▎| 117/125 [11:52<00:48,  6.11s/it]
Training Epoch: 10/100, batch 234/250 completed (loss: 0.063:  94%|█████████▎| 117/125 [11:52<00:48,  6.11s/it] 
Training Epoch: 10/100, batch 234/250 completed (loss: 0.1537:  94%|█████████▍| 118/125 [11:55<00:42,  6.08s/it]
Training Epoch: 10/100, batch 234/250 completed (loss: 0.063:  94%|█████████▍| 118/125 [11:55<00:42,  6.08s/it]
Training Epoch: 10/100, batch 234/250 completed (loss: 0.0235:  94%|█████████▍| 118/125 [11:55<00:42,  6.08s/it]
Training Epoch: 10/100, batch 234/250 completed (loss: 0.0245:  94%|█████████▍| 118/125 [11:55<00:42,  6.08s/it]
Training Epoch: 10/100, batch 235/250 completed (loss: 0.1232:  94%|█████████▍| 118/125 [11:55<00:42,  6.08s/it]
Training Epoch: 10/100, batch 235/250 completed (loss: 0.0205:  94%|█████████▍| 118/125 [11:55<00:42,  6.08s/it]
Training Epoch: 10/100, batch 235/250 completed (loss: 0.1596:  94%|█████████▍| 118/125 [11:55<00:42,  6.08s/it]
Training Epoch: 10/100, batch 235/250 completed (loss: 0.0744:  94%|█████████▍| 118/125 [11:55<00:42,  6.08s/it]
Training Epoch: 10/100, batch 236/250 completed (loss: 0.0414:  94%|█████████▍| 118/125 [11:58<00:42,  6.08s/it]
Training Epoch: 10/100, batch 236/250 completed (loss: 0.0298:  94%|█████████▍| 118/125 [11:58<00:42,  6.08s/it]
Training Epoch: 10/100, batch 236/250 completed (loss: 0.2353:  94%|█████████▍| 118/125 [11:58<00:42,  6.08s/it]
Training Epoch: 10/100, batch 236/250 completed (loss: 0.1254:  94%|█████████▍| 118/125 [11:58<00:42,  6.08s/it]
Training Epoch: 10/100, batch 236/250 completed (loss: 0.2353:  95%|█████████▌| 119/125 [12:01<00:36,  6.10s/it]
Training Epoch: 10/100, batch 236/250 completed (loss: 0.0414:  95%|█████████▌| 119/125 [12:01<00:36,  6.11s/it]
Training Epoch: 10/100, batch 236/250 completed (loss: 0.0298:  95%|█████████▌| 119/125 [12:01<00:36,  6.11s/it]
Training Epoch: 10/100, batch 236/250 completed (loss: 0.1254:  95%|█████████▌| 119/125 [12:01<00:36,  6.11s/it]
Training Epoch: 10/100, batch 237/250 completed (loss: 0.0611:  95%|█████████▌| 119/125 [12:01<00:36,  6.11s/it]
Training Epoch: 10/100, batch 237/250 completed (loss: 0.0076:  95%|█████████▌| 119/125 [12:01<00:36,  6.11s/it]
Training Epoch: 10/100, batch 237/250 completed (loss: 0.0183:  95%|█████████▌| 119/125 [12:01<00:36,  6.10s/it]
Training Epoch: 10/100, batch 237/250 completed (loss: 0.2659:  95%|█████████▌| 119/125 [12:01<00:36,  6.11s/it]
Training Epoch: 10/100, batch 238/250 completed (loss: 0.0552:  95%|█████████▌| 119/125 [12:04<00:36,  6.11s/it]
Training Epoch: 10/100, batch 238/250 completed (loss: 0.0794:  95%|█████████▌| 119/125 [12:04<00:36,  6.11s/it]
Training Epoch: 10/100, batch 238/250 completed (loss: 0.0522:  95%|█████████▌| 119/125 [12:04<00:36,  6.10s/it]
Training Epoch: 10/100, batch 238/250 completed (loss: 0.0905:  95%|█████████▌| 119/125 [12:04<00:36,  6.11s/it]
Training Epoch: 10/100, batch 238/250 completed (loss: 0.0552:  96%|█████████▌| 120/125 [12:07<00:30,  6.08s/it]
Training Epoch: 10/100, batch 238/250 completed (loss: 0.0522:  96%|█████████▌| 120/125 [12:07<00:30,  6.08s/it]
Training Epoch: 10/100, batch 238/250 completed (loss: 0.0905:  96%|█████████▌| 120/125 [12:07<00:30,  6.08s/it]
Training Epoch: 10/100, batch 238/250 completed (loss: 0.0794:  96%|█████████▌| 120/125 [12:07<00:30,  6.08s/it]
Training Epoch: 10/100, batch 239/250 completed (loss: 0.0137:  96%|█████████▌| 120/125 [12:07<00:30,  6.08s/it]
Training Epoch: 10/100, batch 239/250 completed (loss: 0.0316:  96%|█████████▌| 120/125 [12:07<00:30,  6.08s/it]
Training Epoch: 10/100, batch 239/250 completed (loss: 0.245:  96%|█████████▌| 120/125 [12:07<00:30,  6.08s/it] 
Training Epoch: 10/100, batch 239/250 completed (loss: 0.128:  96%|█████████▌| 120/125 [12:07<00:30,  6.08s/it] 
Training Epoch: 10/100, batch 240/250 completed (loss: 0.0823:  96%|█████████▌| 120/125 [12:10<00:30,  6.08s/it]
Training Epoch: 10/100, batch 240/250 completed (loss: 0.0475:  96%|█████████▌| 120/125 [12:10<00:30,  6.08s/it]
Training Epoch: 10/100, batch 240/250 completed (loss: 0.001:  96%|█████████▌| 120/125 [12:10<00:30,  6.08s/it]
Training Epoch: 10/100, batch 240/250 completed (loss: 0.1142:  96%|█████████▌| 120/125 [12:10<00:30,  6.08s/it]
Training Epoch: 10/100, batch 240/250 completed (loss: 0.0823:  97%|█████████▋| 121/125 [12:13<00:24,  6.15s/it]
Training Epoch: 10/100, batch 240/250 completed (loss: 0.0475:  97%|█████████▋| 121/125 [12:13<00:24,  6.15s/it]
Training Epoch: 10/100, batch 240/250 completed (loss: 0.001:  97%|█████████▋| 121/125 [12:13<00:24,  6.16s/it]
Training Epoch: 10/100, batch 240/250 completed (loss: 0.1142:  97%|█████████▋| 121/125 [12:13<00:24,  6.16s/it]
Training Epoch: 10/100, batch 241/250 completed (loss: 0.1502:  97%|█████████▋| 121/125 [12:13<00:24,  6.16s/it]
Training Epoch: 10/100, batch 241/250 completed (loss: 0.225:  97%|█████████▋| 121/125 [12:13<00:24,  6.15s/it] 
Training Epoch: 10/100, batch 241/250 completed (loss: 0.1205:  97%|█████████▋| 121/125 [12:13<00:24,  6.15s/it]
Training Epoch: 10/100, batch 241/250 completed (loss: 0.1493:  97%|█████████▋| 121/125 [12:13<00:24,  6.16s/it]
Training Epoch: 10/100, batch 242/250 completed (loss: 0.0839:  97%|█████████▋| 121/125 [12:16<00:24,  6.15s/it]
Training Epoch: 10/100, batch 242/250 completed (loss: 0.1204:  97%|█████████▋| 121/125 [12:16<00:24,  6.16s/it]
Training Epoch: 10/100, batch 242/250 completed (loss: 0.1081:  97%|█████████▋| 121/125 [12:16<00:24,  6.15s/it]
Training Epoch: 10/100, batch 242/250 completed (loss: 0.0332:  97%|█████████▋| 121/125 [12:16<00:24,  6.16s/it]
Training Epoch: 10/100, batch 242/250 completed (loss: 0.0839:  98%|█████████▊| 122/125 [12:19<00:18,  6.11s/it]
Training Epoch: 10/100, batch 242/250 completed (loss: 0.1204:  98%|█████████▊| 122/125 [12:19<00:18,  6.12s/it]
Training Epoch: 10/100, batch 242/250 completed (loss: 0.1081:  98%|█████████▊| 122/125 [12:19<00:18,  6.12s/it]
Training Epoch: 10/100, batch 242/250 completed (loss: 0.0332:  98%|█████████▊| 122/125 [12:19<00:18,  6.12s/it]
Training Epoch: 10/100, batch 243/250 completed (loss: 0.1101:  98%|█████████▊| 122/125 [12:19<00:18,  6.11s/it]
Training Epoch: 10/100, batch 243/250 completed (loss: 0.0994:  98%|█████████▊| 122/125 [12:19<00:18,  6.12s/it]
Training Epoch: 10/100, batch 243/250 completed (loss: 0.0516:  98%|█████████▊| 122/125 [12:19<00:18,  6.12s/it]
Training Epoch: 10/100, batch 243/250 completed (loss: 0.0975:  98%|█████████▊| 122/125 [12:19<00:18,  6.12s/it]
Training Epoch: 10/100, batch 244/250 completed (loss: 0.154:  98%|█████████▊| 122/125 [12:22<00:18,  6.12s/it] 
Training Epoch: 10/100, batch 244/250 completed (loss: 0.0996:  98%|█████████▊| 122/125 [12:22<00:18,  6.12s/it]
Training Epoch: 10/100, batch 244/250 completed (loss: 0.0815:  98%|█████████▊| 122/125 [12:22<00:18,  6.11s/it]
Training Epoch: 10/100, batch 244/250 completed (loss: 0.0477:  98%|█████████▊| 122/125 [12:22<00:18,  6.12s/it]
Training Epoch: 10/100, batch 244/250 completed (loss: 0.154:  98%|█████████▊| 123/125 [12:25<00:12,  6.09s/it]
Training Epoch: 10/100, batch 244/250 completed (loss: 0.0477:  98%|█████████▊| 123/125 [12:25<00:12,  6.09s/it]
Training Epoch: 10/100, batch 244/250 completed (loss: 0.0815:  98%|█████████▊| 123/125 [12:25<00:12,  6.09s/it]
Training Epoch: 10/100, batch 244/250 completed (loss: 0.0996:  98%|█████████▊| 123/125 [12:25<00:12,  6.10s/it]
Training Epoch: 10/100, batch 245/250 completed (loss: 0.0148:  98%|█████████▊| 123/125 [12:25<00:12,  6.10s/it]
Training Epoch: 10/100, batch 245/250 completed (loss: 0.1361:  98%|█████████▊| 123/125 [12:25<00:12,  6.09s/it]
Training Epoch: 10/100, batch 245/250 completed (loss: 0.1415:  98%|█████████▊| 123/125 [12:25<00:12,  6.09s/it]
Training Epoch: 10/100, batch 245/250 completed (loss: 0.0255:  98%|█████████▊| 123/125 [12:25<00:12,  6.09s/it]
Training Epoch: 10/100, batch 246/250 completed (loss: 0.0567:  98%|█████████▊| 123/125 [12:29<00:12,  6.09s/it]
Training Epoch: 10/100, batch 246/250 completed (loss: 0.0311:  98%|█████████▊| 123/125 [12:29<00:12,  6.10s/it]
Training Epoch: 10/100, batch 246/250 completed (loss: 0.1415:  98%|█████████▊| 123/125 [12:29<00:12,  6.09s/it]
Training Epoch: 10/100, batch 246/250 completed (loss: 0.0165:  98%|█████████▊| 123/125 [12:29<00:12,  6.09s/it]
Training Epoch: 10/100, batch 246/250 completed (loss: 0.1415:  99%|█████████▉| 124/125 [12:32<00:06,  6.12s/it]
Training Epoch: 10/100, batch 246/250 completed (loss: 0.0165:  99%|█████████▉| 124/125 [12:32<00:06,  6.12s/it]
Training Epoch: 10/100, batch 246/250 completed (loss: 0.0311:  99%|█████████▉| 124/125 [12:32<00:06,  6.12s/it]
Training Epoch: 10/100, batch 246/250 completed (loss: 0.0567:  99%|█████████▉| 124/125 [12:32<00:06,  6.13s/it]
Training Epoch: 10/100, batch 247/250 completed (loss: 0.1764:  99%|█████████▉| 124/125 [12:32<00:06,  6.12s/it]
Training Epoch: 10/100, batch 247/250 completed (loss: 0.0406:  99%|█████████▉| 124/125 [12:32<00:06,  6.12s/it]
Training Epoch: 10/100, batch 247/250 completed (loss: 0.1714:  99%|█████████▉| 124/125 [12:32<00:06,  6.13s/it]
Training Epoch: 10/100, batch 247/250 completed (loss: 0.3577:  99%|█████████▉| 124/125 [12:32<00:06,  6.12s/it]
Training Epoch: 10/100, batch 248/250 completed (loss: 0.0734:  99%|█████████▉| 124/125 [12:35<00:06,  6.12s/it]
Training Epoch: 10/100, batch 248/250 completed (loss: 0.0322:  99%|█████████▉| 124/125 [12:35<00:06,  6.13s/it]
Training Epoch: 10/100, batch 248/250 completed (loss: 0.0807:  99%|█████████▉| 124/125 [12:35<00:06,  6.12s/it]
Training Epoch: 10/100, batch 248/250 completed (loss: 0.0299:  99%|█████████▉| 124/125 [12:35<00:06,  6.12s/it]
Training Epoch: 10/100, batch 248/250 completed (loss: 0.0322: 100%|██████████| 125/125 [12:38<00:00,  6.09s/it]
Training Epoch: 10/100, batch 248/250 completed (loss: 0.0734: 100%|██████████| 125/125 [12:38<00:00,  6.09s/it]
Training Epoch: 10/100, batch 248/250 completed (loss: 0.0299: 100%|██████████| 125/125 [12:38<00:00,  6.09s/it]
Training Epoch: 10/100, batch 248/250 completed (loss: 0.0807: 100%|██████████| 125/125 [12:38<00:00,  6.09s/it]
Training Epoch: 10/100, batch 249/250 completed (loss: 0.1022: 100%|██████████| 125/125 [12:38<00:00,  6.09s/it]
Training Epoch: 10/100, batch 249/250 completed (loss: 0.0335: 100%|██████████| 125/125 [12:38<00:00,  6.09s/it]
Training Epoch: 10/100, batch 249/250 completed (loss: 0.1521: 100%|██████████| 125/125 [12:38<00:00,  6.09s/it]
Training Epoch: 10/100, batch 249/250 completed (loss: 0.0317: 100%|██████████| 125/125 [12:38<00:00,  6.09s/it]
Training Epoch: 10/100, batch 249/250 completed (loss: 0.0335: 100%|██████████| 125/125 [12:38<00:00,  6.07s/it]

Training Epoch: 10/100, batch 249/250 completed (loss: 0.1521: 100%|██████████| 125/125 [12:38<00:00,  6.07s/it]

Training Epoch: 10/100, batch 249/250 completed (loss: 0.1022: 100%|██████████| 125/125 [12:38<00:00,  6.07s/it]

Training Epoch: 10/100, batch 249/250 completed (loss: 0.0317: 100%|██████████| 125/125 [12:38<00:00,  6.07s/it]
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/c10d_logger.py:83: UserWarning: barrier(): using the device under current context. You can specify `device_id` in `init_process_group` to mute this warning.
  return func(*args, **kwargs)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
saving model + train state.
Warning: failed to upload checkpoint_10 to HF: (Request ID: Root=1-6a1b7b35-012b7a51201d45e14c33ba3e;d7b8e035-7891-472d-945a-f7cdfac0077c)

403 Forbidden: You need to setup automatic credit recharge in order to upload more data. You can do so at /settings/billing..
Cannot access content at: https://huggingface.co/seyedparsa/thoughtformer-checkpoints.git/info/lfs/objects/batch.
Make sure your token has the correct permissions.
Warning: failed to upload train_state_10 to HF: (Request ID: Root=1-6a1b7b39-1188d58f6ccbf90a7ab8cbd8;c4860322-1804-4428-b715-83f2293bbf12)

403 Forbidden: You need to setup automatic credit recharge in order to upload more data. You can do so at /settings/billing..
Cannot access content at: https://huggingface.co/seyedparsa/thoughtformer-checkpoints.git/info/lfs/objects/batch.
Make sure your token has the correct permissions.
Deleted old checkpoint: train_state_9
Deleted old checkpoint: checkpoint_9
[warn] HF upload failed for ckpt_10; kept locally, earlier ckpts deleted.

Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]
Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]
Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]eval loss 0.043101973682641986

Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]
[CORRECT] Example 0 (diff=3)
  Expected : 'Claude'
  Extracted: 'Claude'
  Generated: '### Claude'

[WRONG] Example 1 (diff=8)

Test Accuracy:   3%|▎         | 1/32 [00:06<03:25,  6.61s/it]  Expected : 'Guy'
Test Accuracy:   3%|▎         | 1/32 [00:06<03:24,  6.61s/it]
  Extracted: 'Irma'
  Generated: '### Irma'

Test accuracy: 0.62:   3%|▎         | 1/32 [00:06<03:25,  6.61s/it]
Test accuracy: 0.88:   3%|▎         | 1/32 [00:06<03:24,  6.61s/it]
Test Accuracy:   3%|▎         | 1/32 [00:06<03:24,  6.61s/it]
[WRONG] Example 2 (diff=10)
  Expected : 'Elvis'
  Extracted: 'Shannon'
  Generated: '### Shannon'

Test accuracy: 0.62:   3%|▎         | 1/32 [00:06<03:24,  6.61s/it]
[WRONG] Example 3 (diff=8)
  Expected : 'Coral'
  Extracted: 'Claude'
  Generated: '### Claude'

[CORRECT] Example 4 (diff=1)
  Expected : 'Patricia'
  Extracted: 'Patricia'
  Generated: '### Patricia'

Test Accuracy:   3%|▎         | 1/32 [00:06<03:19,  6.44s/it]
Test accuracy: 0.62:   3%|▎         | 1/32 [00:06<03:19,  6.44s/it]
Test accuracy: 0.62:   6%|▋         | 2/32 [00:13<03:18,  6.62s/it]
Test accuracy: 0.88:   6%|▋         | 2/32 [00:13<03:18,  6.62s/it]
Test accuracy: 0.62:   6%|▋         | 2/32 [00:13<03:18,  6.62s/it]
Test accuracy: 0.75:   6%|▋         | 2/32 [00:13<03:18,  6.62s/it]
Test accuracy: 0.81:   6%|▋         | 2/32 [00:13<03:18,  6.62s/it]
Test accuracy: 0.62:   6%|▋         | 2/32 [00:13<03:18,  6.62s/it]
Test accuracy: 0.62:   6%|▋         | 2/32 [00:13<03:16,  6.55s/it]
Test accuracy: 0.56:   6%|▋         | 2/32 [00:13<03:16,  6.55s/it]
Test accuracy: 0.81:   9%|▉         | 3/32 [00:20<03:14,  6.69s/it]
Test accuracy: 0.62:   9%|▉         | 3/32 [00:20<03:14,  6.69s/it]
Test accuracy: 0.71:   9%|▉         | 3/32 [00:20<03:14,  6.69s/it]
Test accuracy: 0.62:   9%|▉         | 3/32 [00:20<03:14,  6.69s/it]
Test accuracy: 0.56:   9%|▉         | 3/32 [00:19<03:12,  6.65s/it]
Test accuracy: 0.54:   9%|▉         | 3/32 [00:19<03:12,  6.65s/it]
Test accuracy: 0.75:   9%|▉         | 3/32 [00:20<03:14,  6.70s/it]
Test accuracy: 0.79:   9%|▉         | 3/32 [00:20<03:14,  6.70s/it]
Test accuracy: 0.79:  12%|█▎        | 4/32 [00:26<03:03,  6.55s/it]
Test accuracy: 0.71:  12%|█▎        | 4/32 [00:26<03:03,  6.56s/it]
Test accuracy: 0.62:  12%|█▎        | 4/32 [00:26<03:03,  6.56s/it]
Test accuracy: 0.78:  12%|█▎        | 4/32 [00:26<03:03,  6.55s/it]
Test accuracy: 0.72:  12%|█▎        | 4/32 [00:26<03:03,  6.56s/it]
Test accuracy: 0.62:  12%|█▎        | 4/32 [00:26<03:03,  6.56s/it]
Test accuracy: 0.54:  12%|█▎        | 4/32 [00:26<03:02,  6.53s/it]
Test accuracy: 0.59:  12%|█▎        | 4/32 [00:26<03:02,  6.53s/it]
Test accuracy: 0.72:  16%|█▌        | 5/32 [00:32<02:54,  6.47s/it]
Test accuracy: 0.62:  16%|█▌        | 5/32 [00:32<02:54,  6.47s/it]
Test accuracy: 0.78:  16%|█▌        | 5/32 [00:32<02:54,  6.47s/it]
Test accuracy: 0.75:  16%|█▌        | 5/32 [00:32<02:54,  6.47s/it]
Test accuracy: 0.62:  16%|█▌        | 5/32 [00:32<02:54,  6.47s/it]
Test accuracy: 0.75:  16%|█▌        | 5/32 [00:32<02:54,  6.47s/it]
Test accuracy: 0.59:  16%|█▌        | 5/32 [00:32<02:54,  6.46s/it]
Test accuracy: 0.6:  16%|█▌        | 5/32 [00:32<02:54,  6.46s/it] 
Test accuracy: 0.75:  19%|█▉        | 6/32 [00:38<02:45,  6.36s/it]
Test accuracy: 0.62:  19%|█▉        | 6/32 [00:38<02:45,  6.36s/it]
Test accuracy: 0.75:  19%|█▉        | 6/32 [00:38<02:45,  6.36s/it]
Test accuracy: 0.73:  19%|█▉        | 6/32 [00:38<02:45,  6.36s/it]
Test accuracy: 0.75:  19%|█▉        | 6/32 [00:38<02:45,  6.36s/it]
Test accuracy: 0.67:  19%|█▉        | 6/32 [00:38<02:45,  6.36s/it]
Test accuracy: 0.6:  19%|█▉        | 6/32 [00:38<02:45,  6.35s/it]
Test accuracy: 0.67:  19%|█▉        | 6/32 [00:38<02:45,  6.35s/it]
Test accuracy: 0.73:  22%|██▏       | 7/32 [00:45<02:38,  6.35s/it]
Test accuracy: 0.75:  22%|██▏       | 7/32 [00:45<02:38,  6.35s/it]
Test accuracy: 0.67:  22%|██▏       | 7/32 [00:44<02:38,  6.35s/it]
Test accuracy: 0.71:  22%|██▏       | 7/32 [00:45<02:38,  6.35s/it]
Test accuracy: 0.7:  22%|██▏       | 7/32 [00:45<02:38,  6.35s/it] 
Test accuracy: 0.67:  22%|██▏       | 7/32 [00:45<02:38,  6.35s/it]
Test accuracy: 0.64:  22%|██▏       | 7/32 [00:44<02:38,  6.35s/it]
Test accuracy: 0.64:  22%|██▏       | 7/32 [00:45<02:38,  6.35s/it]
Test accuracy: 0.7:  25%|██▌       | 8/32 [00:51<02:31,  6.32s/it]
Test accuracy: 0.71:  25%|██▌       | 8/32 [00:51<02:31,  6.32s/it]
Test accuracy: 0.64:  25%|██▌       | 8/32 [00:51<02:31,  6.32s/it]
Test accuracy: 0.64:  25%|██▌       | 8/32 [00:51<02:31,  6.32s/it]
Test accuracy: 0.73:  25%|██▌       | 8/32 [00:51<02:31,  6.32s/it]
Test accuracy: 0.72:  25%|██▌       | 8/32 [00:51<02:31,  6.32s/it]
Test accuracy: 0.64:  25%|██▌       | 8/32 [00:51<02:31,  6.32s/it]
Test accuracy: 0.67:  25%|██▌       | 8/32 [00:51<02:31,  6.32s/it]
Test accuracy: 0.72:  28%|██▊       | 9/32 [00:58<02:28,  6.46s/it]
Test accuracy: 0.73:  28%|██▊       | 9/32 [00:58<02:28,  6.46s/it]
Test accuracy: 0.74:  28%|██▊       | 9/32 [00:58<02:28,  6.46s/it]
Test accuracy: 0.74:  28%|██▊       | 9/32 [00:58<02:28,  6.46s/it]
Test accuracy: 0.67:  28%|██▊       | 9/32 [00:58<02:28,  6.46s/it]
Test accuracy: 0.67:  28%|██▊       | 9/32 [00:58<02:28,  6.46s/it]
Test accuracy: 0.64:  28%|██▊       | 9/32 [00:58<02:28,  6.46s/it]
Test accuracy: 0.67:  28%|██▊       | 9/32 [00:58<02:28,  6.46s/it]
Test accuracy: 0.74:  31%|███▏      | 10/32 [01:04<02:22,  6.48s/it]
Test accuracy: 0.74:  31%|███▏      | 10/32 [01:04<02:22,  6.48s/it]
Test accuracy: 0.67:  31%|███▏      | 10/32 [01:04<02:22,  6.48s/it]
Test accuracy: 0.69:  31%|███▏      | 10/32 [01:04<02:22,  6.48s/it]
Test accuracy: 0.74:  31%|███▏      | 10/32 [01:04<02:22,  6.48s/it]
Test accuracy: 0.65:  31%|███▏      | 10/32 [01:04<02:22,  6.48s/it]
Test accuracy: 0.67:  31%|███▏      | 10/32 [01:04<02:22,  6.48s/it]
Test accuracy: 0.66:  31%|███▏      | 10/32 [01:04<02:22,  6.48s/it]
Test accuracy: 0.74:  34%|███▍      | 11/32 [01:10<02:13,  6.38s/it]
Test accuracy: 0.65:  34%|███▍      | 11/32 [01:10<02:13,  6.38s/it]
Test accuracy: 0.69:  34%|███▍      | 11/32 [01:10<02:13,  6.38s/it]
Test accuracy: 0.66:  34%|███▍      | 11/32 [01:10<02:13,  6.37s/it]
Test accuracy: 0.72:  34%|███▍      | 11/32 [01:10<02:13,  6.38s/it]
Test accuracy: 0.65:  34%|███▍      | 11/32 [01:10<02:13,  6.38s/it]
Test accuracy: 0.7:  34%|███▍      | 11/32 [01:10<02:13,  6.38s/it] 
Test accuracy: 0.67:  34%|███▍      | 11/32 [01:10<02:13,  6.37s/it]
Test accuracy: 0.7:  38%|███▊      | 12/32 [01:16<02:06,  6.30s/it]
Test accuracy: 0.72:  38%|███▊      | 12/32 [01:16<02:06,  6.30s/it]
Test accuracy: 0.65:  38%|███▊      | 12/32 [01:16<02:06,  6.30s/it]
Test accuracy: 0.68:  38%|███▊      | 12/32 [01:16<02:06,  6.30s/it]
Test accuracy: 0.68:  38%|███▊      | 12/32 [01:16<02:06,  6.30s/it]
Test accuracy: 0.66:  38%|███▊      | 12/32 [01:16<02:06,  6.30s/it]
Test accuracy: 0.67:  38%|███▊      | 12/32 [01:16<02:06,  6.30s/it]
Test accuracy: 0.68:  38%|███▊      | 12/32 [01:16<02:06,  6.30s/it]
Test accuracy: 0.68:  41%|████      | 13/32 [01:23<02:00,  6.37s/it]
Test accuracy: 0.68:  41%|████      | 13/32 [01:23<02:00,  6.37s/it]
Test accuracy: 0.68:  41%|████      | 13/32 [01:23<02:00,  6.37s/it]
Test accuracy: 0.66:  41%|████      | 13/32 [01:23<02:00,  6.37s/it]
Test accuracy: 0.69:  41%|████      | 13/32 [01:23<02:00,  6.37s/it]
Test accuracy: 0.7:  41%|████      | 13/32 [01:23<02:00,  6.37s/it] 
Test accuracy: 0.66:  41%|████      | 13/32 [01:23<02:00,  6.37s/it]
Test accuracy: 0.66:  41%|████      | 13/32 [01:23<02:00,  6.37s/it]
Test accuracy: 0.69:  44%|████▍     | 14/32 [01:29<01:53,  6.30s/it]
Test accuracy: 0.7:  44%|████▍     | 14/32 [01:29<01:53,  6.30s/it]
Test accuracy: 0.71:  44%|████▍     | 14/32 [01:29<01:53,  6.30s/it]
Test accuracy: 0.71:  44%|████▍     | 14/32 [01:29<01:53,  6.30s/it]
Test accuracy: 0.66:  44%|████▍     | 14/32 [01:29<01:53,  6.30s/it]
Test accuracy: 0.66:  44%|████▍     | 14/32 [01:29<01:53,  6.30s/it]
Test accuracy: 0.66:  44%|████▍     | 14/32 [01:29<01:53,  6.30s/it]
Test accuracy: 0.66:  44%|████▍     | 14/32 [01:29<01:53,  6.30s/it]
Test accuracy: 0.71:  47%|████▋     | 15/32 [01:35<01:47,  6.31s/it]
Test accuracy: 0.71:  47%|████▋     | 15/32 [01:35<01:47,  6.31s/it]
Test accuracy: 0.66:  47%|████▋     | 15/32 [01:35<01:47,  6.31s/it]
Test accuracy: 0.69:  47%|████▋     | 15/32 [01:35<01:47,  6.31s/it]
Test accuracy: 0.71:  47%|████▋     | 15/32 [01:35<01:47,  6.31s/it]
Test accuracy: 0.66:  47%|████▋     | 15/32 [01:35<01:47,  6.31s/it]
Test accuracy: 0.65:  47%|████▋     | 15/32 [01:35<01:47,  6.31s/it]
Test accuracy: 0.66:  47%|████▋     | 15/32 [01:35<01:47,  6.31s/it]
Test accuracy: 0.69:  50%|█████     | 16/32 [01:42<01:41,  6.32s/it]
Test accuracy: 0.71:  50%|█████     | 16/32 [01:42<01:41,  6.32s/it]
Test accuracy: 0.65:  50%|█████     | 16/32 [01:42<01:41,  6.32s/it]
Test accuracy: 0.69:  50%|█████     | 16/32 [01:42<01:41,  6.32s/it]
Test accuracy: 0.72:  50%|█████     | 16/32 [01:42<01:41,  6.32s/it]
Test accuracy: 0.65:  50%|█████     | 16/32 [01:42<01:41,  6.32s/it]
Test accuracy: 0.66:  50%|█████     | 16/32 [01:42<01:41,  6.32s/it]
Test accuracy: 0.66:  50%|█████     | 16/32 [01:42<01:41,  6.32s/it]
Test accuracy: 0.69:  53%|█████▎    | 17/32 [01:48<01:33,  6.26s/it]
Test accuracy: 0.72:  53%|█████▎    | 17/32 [01:48<01:33,  6.26s/it]
Test accuracy: 0.65:  53%|█████▎    | 17/32 [01:48<01:33,  6.26s/it]
Test accuracy: 0.66:  53%|█████▎    | 17/32 [01:48<01:33,  6.26s/it]
Test accuracy: 0.72:  53%|█████▎    | 17/32 [01:48<01:33,  6.26s/it]
Test accuracy: 0.71:  53%|█████▎    | 17/32 [01:48<01:33,  6.26s/it]
Test accuracy: 0.65:  53%|█████▎    | 17/32 [01:48<01:33,  6.26s/it]
Test accuracy: 0.65:  53%|█████▎    | 17/32 [01:48<01:33,  6.26s/it]
Test accuracy: 0.72:  56%|█████▋    | 18/32 [01:54<01:28,  6.29s/it]
Test accuracy: 0.65:  56%|█████▋    | 18/32 [01:54<01:28,  6.29s/it]
Test accuracy: 0.73:  56%|█████▋    | 18/32 [01:54<01:28,  6.29s/it]
Test accuracy: 0.65:  56%|█████▋    | 18/32 [01:54<01:28,  6.29s/it]
Test accuracy: 0.67:  56%|█████▋    | 18/32 [01:54<01:28,  6.29s/it]
Test accuracy: 0.66:  56%|█████▋    | 18/32 [01:54<01:28,  6.29s/it]
Test accuracy: 0.71:  56%|█████▋    | 18/32 [01:54<01:28,  6.29s/it]
Test accuracy: 0.72:  56%|█████▋    | 18/32 [01:54<01:28,  6.29s/it]
Test accuracy: 0.73:  59%|█████▉    | 19/32 [02:01<01:21,  6.30s/it]
Test accuracy: 0.7:  59%|█████▉    | 19/32 [02:01<01:21,  6.30s/it] 
Test accuracy: 0.66:  59%|█████▉    | 19/32 [02:00<01:21,  6.30s/it]
Test accuracy: 0.67:  59%|█████▉    | 19/32 [02:01<01:21,  6.30s/it]
Test accuracy: 0.67:  59%|█████▉    | 19/32 [02:00<01:21,  6.30s/it]
Test accuracy: 0.66:  59%|█████▉    | 19/32 [02:01<01:21,  6.30s/it]
Test accuracy: 0.72:  59%|█████▉    | 19/32 [02:01<01:21,  6.30s/it]
Test accuracy: 0.71:  59%|█████▉    | 19/32 [02:01<01:21,  6.30s/it]
Test accuracy: 0.7:  62%|██████▎   | 20/32 [02:07<01:15,  6.25s/it]
Test accuracy: 0.66:  62%|██████▎   | 20/32 [02:07<01:15,  6.25s/it]
Test accuracy: 0.71:  62%|██████▎   | 20/32 [02:07<01:15,  6.25s/it]
Test accuracy: 0.69:  62%|██████▎   | 20/32 [02:07<01:15,  6.25s/it]
Test accuracy: 0.66:  62%|██████▎   | 20/32 [02:07<01:15,  6.25s/it]
Test accuracy: 0.7:  62%|██████▎   | 20/32 [02:07<01:15,  6.25s/it] 
Test accuracy: 0.67:  62%|██████▎   | 20/32 [02:07<01:15,  6.25s/it]
Test accuracy: 0.68:  62%|██████▎   | 20/32 [02:07<01:15,  6.25s/it]
Test accuracy: 0.69:  66%|██████▌   | 21/32 [02:14<01:11,  6.49s/it]
Test accuracy: 0.66:  66%|██████▌   | 21/32 [02:14<01:11,  6.49s/it]
Test accuracy: 0.68:  66%|██████▌   | 21/32 [02:14<01:11,  6.49s/it]
Test accuracy: 0.68:  66%|██████▌   | 21/32 [02:14<01:11,  6.49s/it]
Test accuracy: 0.67:  66%|██████▌   | 21/32 [02:14<01:11,  6.49s/it]
Test accuracy: 0.66:  66%|██████▌   | 21/32 [02:14<01:11,  6.49s/it]
Test accuracy: 0.7:  66%|██████▌   | 21/32 [02:14<01:11,  6.49s/it]
Test accuracy: 0.71:  66%|██████▌   | 21/32 [02:14<01:11,  6.49s/it]
Test accuracy: 0.68:  69%|██████▉   | 22/32 [02:20<01:04,  6.45s/it]
Test accuracy: 0.67:  69%|██████▉   | 22/32 [02:20<01:04,  6.45s/it]
Test accuracy: 0.68:  69%|██████▉   | 22/32 [02:20<01:04,  6.45s/it]
Test accuracy: 0.67:  69%|██████▉   | 22/32 [02:20<01:04,  6.45s/it]
Test accuracy: 0.66:  69%|██████▉   | 22/32 [02:20<01:04,  6.45s/it]
Test accuracy: 0.66:  69%|██████▉   | 22/32 [02:20<01:04,  6.45s/it]
Test accuracy: 0.71:  69%|██████▉   | 22/32 [02:20<01:04,  6.45s/it]
Test accuracy: 0.69:  69%|██████▉   | 22/32 [02:20<01:04,  6.45s/it]
Test accuracy: 0.69:  72%|███████▏  | 23/32 [02:26<00:57,  6.36s/it]
Test accuracy: 0.7:  72%|███████▏  | 23/32 [02:26<00:57,  6.36s/it] 
Test accuracy: 0.68:  72%|███████▏  | 23/32 [02:26<00:57,  6.36s/it]
Test accuracy: 0.69:  72%|███████▏  | 23/32 [02:26<00:57,  6.36s/it]
Test accuracy: 0.67:  72%|███████▏  | 23/32 [02:26<00:57,  6.36s/it]
Test accuracy: 0.68:  72%|███████▏  | 23/32 [02:26<00:57,  6.36s/it]
Test accuracy: 0.66:  72%|███████▏  | 23/32 [02:26<00:57,  6.36s/it]
Test accuracy: 0.66:  72%|███████▏  | 23/32 [02:26<00:57,  6.36s/it]
Test accuracy: 0.7:  75%|███████▌  | 24/32 [02:33<00:52,  6.57s/it]
Test accuracy: 0.69:  75%|███████▌  | 24/32 [02:33<00:52,  6.57s/it]
Test accuracy: 0.68:  75%|███████▌  | 24/32 [02:33<00:52,  6.57s/it]
Test accuracy: 0.68:  75%|███████▌  | 24/32 [02:33<00:52,  6.57s/it]
Test accuracy: 0.69:  75%|███████▌  | 24/32 [02:33<00:52,  6.57s/it]
Test accuracy: 0.68:  75%|███████▌  | 24/32 [02:33<00:52,  6.57s/it]
Test accuracy: 0.66:  75%|███████▌  | 24/32 [02:33<00:52,  6.57s/it]
Test accuracy: 0.66:  75%|███████▌  | 24/32 [02:33<00:52,  6.57s/it]
Test accuracy: 0.69:  78%|███████▊  | 25/32 [02:40<00:45,  6.50s/it]
Test accuracy: 0.68:  78%|███████▊  | 25/32 [02:40<00:45,  6.50s/it]
Test accuracy: 0.68:  78%|███████▊  | 25/32 [02:40<00:45,  6.50s/it]
Test accuracy: 0.66:  78%|███████▊  | 25/32 [02:40<00:45,  6.50s/it]
Test accuracy: 0.68:  78%|███████▊  | 25/32 [02:40<00:45,  6.50s/it]
Test accuracy: 0.66:  78%|███████▊  | 25/32 [02:40<00:45,  6.50s/it]
Test accuracy: 0.68:  78%|███████▊  | 25/32 [02:40<00:45,  6.50s/it]
Test accuracy: 0.69:  78%|███████▊  | 25/32 [02:40<00:45,  6.50s/it]
Test accuracy: 0.68:  81%|████████▏ | 26/32 [02:46<00:38,  6.46s/it]
Test accuracy: 0.68:  81%|████████▏ | 26/32 [02:46<00:38,  6.46s/it]
Test accuracy: 0.68:  81%|████████▏ | 26/32 [02:46<00:38,  6.46s/it]
Test accuracy: 0.67:  81%|████████▏ | 26/32 [02:46<00:38,  6.46s/it]
Test accuracy: 0.66:  81%|████████▏ | 26/32 [02:46<00:38,  6.45s/it]
Test accuracy: 0.65:  81%|████████▏ | 26/32 [02:46<00:38,  6.45s/it]
Test accuracy: 0.69:  81%|████████▏ | 26/32 [02:46<00:38,  6.46s/it]
Test accuracy: 0.68:  81%|████████▏ | 26/32 [02:46<00:38,  6.46s/it]
Test accuracy: 0.65:  84%|████████▍ | 27/32 [02:52<00:31,  6.36s/it]
Test accuracy: 0.68:  84%|████████▍ | 27/32 [02:52<00:31,  6.36s/it]
Test accuracy: 0.68:  84%|████████▍ | 27/32 [02:52<00:31,  6.36s/it]
Test accuracy: 0.64:  84%|████████▍ | 27/32 [02:52<00:31,  6.36s/it]
Test accuracy: 0.69:  84%|████████▍ | 27/32 [02:52<00:31,  6.36s/it]
Test accuracy: 0.69:  84%|████████▍ | 27/32 [02:52<00:31,  6.36s/it]
Test accuracy: 0.67:  84%|████████▍ | 27/32 [02:52<00:31,  6.36s/it]
Test accuracy: 0.68:  84%|████████▍ | 27/32 [02:52<00:31,  6.36s/it]
Test accuracy: 0.69:  88%|████████▊ | 28/32 [02:59<00:26,  6.54s/it]
Test accuracy: 0.68:  88%|████████▊ | 28/32 [02:59<00:26,  6.54s/it]
Test accuracy: 0.68:  88%|████████▊ | 28/32 [02:59<00:26,  6.54s/it]
Test accuracy: 0.64:  88%|████████▊ | 28/32 [02:59<00:26,  6.54s/it]
Test accuracy: 0.69:  88%|████████▊ | 28/32 [02:59<00:26,  6.54s/it]
Test accuracy: 0.65:  88%|████████▊ | 28/32 [02:59<00:26,  6.54s/it]
Test accuracy: 0.69:  88%|████████▊ | 28/32 [02:59<00:26,  6.55s/it]
Test accuracy: 0.68:  88%|████████▊ | 28/32 [02:59<00:26,  6.55s/it]
Test accuracy: 0.68:  91%|█████████ | 29/32 [03:06<00:19,  6.49s/it]
Test accuracy: 0.68:  91%|█████████ | 29/32 [03:06<00:19,  6.48s/it]
Test accuracy: 0.69:  91%|█████████ | 29/32 [03:06<00:19,  6.49s/it]
Test accuracy: 0.65:  91%|█████████ | 29/32 [03:05<00:19,  6.49s/it]
Test accuracy: 0.68:  91%|█████████ | 29/32 [03:06<00:19,  6.49s/it]
Test accuracy: 0.69:  91%|█████████ | 29/32 [03:06<00:19,  6.48s/it]
Test accuracy: 0.69:  91%|█████████ | 29/32 [03:06<00:19,  6.49s/it]
Test accuracy: 0.65:  91%|█████████ | 29/32 [03:05<00:19,  6.49s/it]
Test accuracy: 0.69:  94%|█████████▍| 30/32 [03:12<00:12,  6.43s/it]
Test accuracy: 0.68:  94%|█████████▍| 30/32 [03:12<00:12,  6.44s/it]
Test accuracy: 0.68:  94%|█████████▍| 30/32 [03:12<00:12,  6.43s/it]
Test accuracy: 0.69:  94%|█████████▍| 30/32 [03:12<00:12,  6.44s/it]
Test accuracy: 0.65:  94%|█████████▍| 30/32 [03:12<00:12,  6.44s/it]
Test accuracy: 0.68:  94%|█████████▍| 30/32 [03:12<00:12,  6.44s/it]
Test accuracy: 0.69:  94%|█████████▍| 30/32 [03:12<00:12,  6.44s/it]
Test accuracy: 0.64:  94%|█████████▍| 30/32 [03:12<00:12,  6.44s/it]
Test accuracy: 0.68:  97%|█████████▋| 31/32 [03:19<00:06,  6.55s/it]
Test accuracy: 0.69:  97%|█████████▋| 31/32 [03:19<00:06,  6.55s/it]
Test accuracy: 0.67:  97%|█████████▋| 31/32 [03:19<00:06,  6.55s/it]
Test accuracy: 0.64:  97%|█████████▋| 31/32 [03:18<00:06,  6.55s/it]
Test accuracy: 0.7:  97%|█████████▋| 31/32 [03:19<00:06,  6.55s/it] 
Test accuracy: 0.65:  97%|█████████▋| 31/32 [03:18<00:06,  6.55s/it]
Test accuracy: 0.68:  97%|█████████▋| 31/32 [03:19<00:06,  6.56s/it]
Test accuracy: 0.68:  97%|█████████▋| 31/32 [03:19<00:06,  6.56s/it]
Test accuracy: 0.65: 100%|██████████| 32/32 [03:21<00:00,  5.26s/it]
Test accuracy: 0.64: 100%|██████████| 32/32 [03:21<00:00,  5.26s/it]
Test accuracy: 0.68: 100%|██████████| 32/32 [03:21<00:00,  5.26s/it]
Test accuracy: 0.7: 100%|██████████| 32/32 [03:21<00:00,  5.26s/it]
Test accuracy: 0.67: 100%|██████████| 32/32 [03:21<00:00,  5.26s/it]
Test accuracy: 0.68: 100%|██████████| 32/32 [03:21<00:00,  5.26s/it]
Test accuracy: 0.7: 100%|██████████| 32/32 [03:21<00:00,  5.26s/it]
Test accuracy: 0.67: 100%|██████████| 32/32 [03:21<00:00,  5.26s/it]
Test accuracy: 0.68: 100%|██████████| 32/32 [03:21<00:00,  6.30s/it]
Device 2: Cor=169, CoT=0, Total=250

Test accuracy: 0.7: 100%|██████████| 32/32 [03:21<00:00,  6.30s/it]
Device 3: Cor=174, CoT=0, Total=250

Test accuracy: 0.67: 100%|██████████| 32/32 [03:21<00:00,  6.30s/it]
Device 1: Cor=168, CoT=0, Total=250

Test accuracy: 0.64: 100%|██████████| 32/32 [03:21<00:00,  6.30s/it]
Device 0: Cor=161, CoT=0, Total=250

============================================================
EVALUATION RESULTS (Epoch 9)
============================================================
Config: data_curriculum=ON, thought_curriculum=OFF
thought_stage=10, data_stage=4, n_latent_tokens=10
Training on difficulties: 0-4

Overall accuracy: 672 / 1000 = 0.6720

Per-difficulty accuracy:
  diff=1: 100/100 = 1.0000 (trained)
  diff=2: 99/100 = 0.9900 (trained)
  diff=3: 98/100 = 0.9800 (trained)
  diff=4: 92/100 = 0.9200 (trained)
  diff=5: 38/100 = 0.3800 (untrained)
  diff=6: 50/100 = 0.5000 (untrained)
  diff=7: 51/100 = 0.5100 (untrained)
  diff=8: 40/100 = 0.4000 (untrained)
  diff=9: 55/100 = 0.5500 (untrained)
  diff=10: 49/100 = 0.4900 (untrained)

Trained (diff 0-4): 389/400 = 0.9725
============================================================


*** ADAPTIVE STAGING: acc_trained=0.9725, acc_diff4=0.9200 >= 0.9 ***
*** Advancing to stage 5 ***

Warning: failed to upload checkpoint_10 to HF: (Request ID: Root=1-6a1b7c18-5635416a6c432b5f2b011e95;cb37aabe-dc31-4f1d-86a5-959703488ccb)

403 Forbidden: You need to setup automatic credit recharge in order to upload more data. You can do so at /settings/billing..
Cannot access content at: https://huggingface.co/seyedparsa/thoughtformer-checkpoints.git/info/lfs/objects/batch.
Make sure your token has the correct permissions.

============================================================
EPOCH 10/100 (thought_stage=10, data_stage=5)
============================================================
  thought_stage=10, c_thought=1, max_difficulty=5

Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):  16%|█▌        | 156/1000 [00:00<00:00, 968.80 examples/s]
Map (num_proc=4):  16%|█▌        | 158/1000 [00:00<00:00, 969.84 examples/s]
Map (num_proc=4):  16%|█▌        | 156/1000 [00:00<00:00, 914.50 examples/s]
Map (num_proc=4):  16%|█▌        | 157/1000 [00:00<00:01, 837.14 examples/s]
Map (num_proc=4):  82%|████████▏ | 815/1000 [00:00<00:00, 3405.23 examples/s]
Map (num_proc=4):  81%|████████▏ | 813/1000 [00:00<00:00, 3017.62 examples/s]
Map (num_proc=4):  81%|████████  | 811/1000 [00:00<00:00, 3288.13 examples/s]
Map (num_proc=4):  81%|████████  | 812/1000 [00:00<00:00, 3116.22 examples/s]
Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2055.01 examples/s]

Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2026.51 examples/s]

Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 1958.49 examples/s]

Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 1818.53 examples/s]

Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:04, 1882.97 examples/s]
Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:04, 1811.94 examples/s]
Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:04, 1811.27 examples/s]
Filter (num_proc=4):  20%|██        | 2000/10000 [00:00<00:02, 3453.00 examples/s]
Filter (num_proc=4):  20%|██        | 2000/10000 [00:00<00:02, 3380.78 examples/s]
Filter (num_proc=4):  20%|██        | 2000/10000 [00:00<00:02, 3304.92 examples/s]
Filter (num_proc=4):  30%|███       | 3000/10000 [00:00<00:01, 4764.42 examples/s]
Filter (num_proc=4):  30%|███       | 3000/10000 [00:00<00:01, 4667.51 examples/s]
Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:05, 1790.49 examples/s]
Filter (num_proc=4):  30%|███       | 3000/10000 [00:00<00:01, 4620.27 examples/s]
Filter (num_proc=4):  45%|████▌     | 4500/10000 [00:00<00:00, 6812.86 examples/s]
Filter (num_proc=4):  50%|█████     | 5000/10000 [00:00<00:00, 7904.22 examples/s]
Filter (num_proc=4):  45%|████▌     | 4500/10000 [00:00<00:00, 6756.23 examples/s]
Filter (num_proc=4):  20%|██        | 2000/10000 [00:00<00:02, 3129.85 examples/s]
Filter (num_proc=4):  75%|███████▌  | 7500/10000 [00:01<00:00, 11682.27 examples/s]
Filter (num_proc=4):  75%|███████▌  | 7500/10000 [00:01<00:00, 11259.09 examples/s]
Filter (num_proc=4):  70%|███████   | 7000/10000 [00:01<00:00, 10480.71 examples/s]
Filter (num_proc=4):  95%|█████████▌| 9500/10000 [00:01<00:00, 12791.15 examples/s]
Filter (num_proc=4):  40%|████      | 4000/10000 [00:00<00:01, 5675.50 examples/s]
Filter (num_proc=4):  95%|█████████▌| 9500/10000 [00:01<00:00, 12883.33 examples/s]
Filter (num_proc=4):  85%|████████▌ | 8500/10000 [00:01<00:00, 11015.19 examples/s]
Filter (num_proc=4):  65%|██████▌   | 6500/10000 [00:01<00:00, 9005.13 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 11378.81 examples/s]
Filter (num_proc=4):  80%|████████  | 8000/10000 [00:01<00:00, 10118.84 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 7107.31 examples/s]

Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 7097.23 examples/s]

Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 6987.82 examples/s] 

Filter (num_proc=4):  95%|█████████▌| 9500/10000 [00:01<00:00, 10417.00 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 6440.88 examples/s]

Map (num_proc=4):   0%|          | 0/5000 [00:00<?, ? examples/s]
Map (num_proc=4):   3%|▎         | 132/5000 [00:00<00:36, 132.60 examples/s]
Map (num_proc=4):  11%|█         | 541/5000 [00:01<00:07, 592.08 examples/s]
Map (num_proc=4):  16%|█▋        | 816/5000 [00:01<00:04, 902.43 examples/s]
Map (num_proc=4):  22%|██▏       | 1088/5000 [00:01<00:04, 932.94 examples/s]
Map (num_proc=4):  31%|███▏      | 1574/5000 [00:01<00:02, 1484.23 examples/s]
Map (num_proc=4):  41%|████▏     | 2070/5000 [00:01<00:01, 2027.41 examples/s]
Map (num_proc=4):  48%|████▊     | 2382/5000 [00:02<00:01, 1407.65 examples/s]
Map (num_proc=4):  61%|██████    | 3043/5000 [00:02<00:00, 2183.60 examples/s]
Map (num_proc=4):  68%|██████▊   | 3409/5000 [00:02<00:00, 2236.08 examples/s]
Map (num_proc=4):  76%|███████▋  | 3817/5000 [00:02<00:00, 1878.35 examples/s]
Map (num_proc=4):  87%|████████▋ | 4352/5000 [00:02<00:00, 2442.12 examples/s]
Map (num_proc=4):  94%|█████████▍| 4713/5000 [00:02<00:00, 2660.99 examples/s]
Map (num_proc=4): 100%|██████████| 5000/5000 [00:03<00:00, 1308.04 examples/s]

Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 1110.81 examples/s]
Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 1449.58 examples/s]
Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 1396.98 examples/s]
Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 1365.31 examples/s]
Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2821.97 examples/s]

Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2115.82 examples/s]

Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2697.53 examples/s]

Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2667.03 examples/s]

Map (num_proc=4):   0%|          | 0/500 [00:00<?, ? examples/s]
Map (num_proc=4):  25%|██▌       | 125/500 [00:00<00:00, 433.38 examples/s]
Map (num_proc=4):  75%|███████▌  | 375/500 [00:00<00:00, 959.38 examples/s]
Map (num_proc=4): 100%|██████████| 500/500 [00:00<00:00, 682.40 examples/s]

Training Epoch: 11:   0%|          | 0/156 [00:00<?, ?it/s]
Training Epoch: 11:   0%|          | 0/156 [00:00<?, ?it/s]
Training Epoch: 11:   0%|          | 0/156 [00:00<?, ?it/s]
Training Epoch: 11:   0%|          | 0/156 [00:00<?, ?it/s]logging training data
rank 3 | gpu 3 | alloc=4.20 GB | reserved=47.73 GB | max=41.99 GB
rank 0 | gpu 0 | alloc=4.20 GB | reserved=47.74 GB | max=41.99 GB
rank 1 | gpu 1 | alloc=4.20 GB | reserved=45.88 GB | max=41.98 GB
rank 2 | gpu 2 | alloc=4.20 GB | reserved=45.87 GB | max=41.98 GB

Training Epoch: 11/100, batch 0/313 completed (loss: 0.0904:   0%|          | 0/156 [00:03<?, ?it/s]
Training Epoch: 11/100, batch 0/313 completed (loss: 0.122:   0%|          | 0/156 [00:03<?, ?it/s]
Training Epoch: 11/100, batch 0/313 completed (loss: 0.0576:   0%|          | 0/156 [00:03<?, ?it/s]
Training Epoch: 11/100, batch 0/313 completed (loss: 0.1289:   0%|          | 0/156 [00:03<?, ?it/s]
Training Epoch: 11/100, batch 0/313 completed (loss: 0.1289:   1%|          | 1/156 [00:06<15:57,  6.18s/it]
Training Epoch: 11/100, batch 0/313 completed (loss: 0.0904:   1%|          | 1/156 [00:06<15:55,  6.17s/it]
Training Epoch: 11/100, batch 0/313 completed (loss: 0.0576:   1%|          | 1/156 [00:06<15:56,  6.17s/it]
Training Epoch: 11/100, batch 0/313 completed (loss: 0.122:   1%|          | 1/156 [00:06<15:59,  6.19s/it]
Training Epoch: 11/100, batch 1/313 completed (loss: 0.2083:   1%|          | 1/156 [00:06<15:55,  6.17s/it]
Training Epoch: 11/100, batch 1/313 completed (loss: 0.1758:   1%|          | 1/156 [00:06<15:59,  6.19s/it]
Training Epoch: 11/100, batch 1/313 completed (loss: 0.0836:   1%|          | 1/156 [00:06<15:56,  6.17s/it]
Training Epoch: 11/100, batch 1/313 completed (loss: 0.1246:   1%|          | 1/156 [00:06<15:57,  6.18s/it]
Training Epoch: 11/100, batch 2/313 completed (loss: 0.5583:   1%|          | 1/156 [00:09<15:55,  6.17s/it]
Training Epoch: 11/100, batch 2/313 completed (loss: 0.0979:   1%|          | 1/156 [00:09<15:59,  6.19s/it]
Training Epoch: 11/100, batch 2/313 completed (loss: 0.0199:   1%|          | 1/156 [00:09<15:56,  6.17s/it]
Training Epoch: 11/100, batch 2/313 completed (loss: 0.0627:   1%|          | 1/156 [00:09<15:57,  6.18s/it]
Training Epoch: 11/100, batch 2/313 completed (loss: 0.0627:   1%|▏         | 2/156 [00:12<15:21,  5.98s/it]
Training Epoch: 11/100, batch 2/313 completed (loss: 0.0199:   1%|▏         | 2/156 [00:11<15:19,  5.97s/it]
Training Epoch: 11/100, batch 2/313 completed (loss: 0.0979:   1%|▏         | 2/156 [00:12<15:18,  5.97s/it]
Training Epoch: 11/100, batch 2/313 completed (loss: 0.5583:   1%|▏         | 2/156 [00:12<15:19,  5.97s/it]
Training Epoch: 11/100, batch 3/313 completed (loss: 0.2223:   1%|▏         | 2/156 [00:12<15:19,  5.97s/it]
Training Epoch: 11/100, batch 3/313 completed (loss: 0.3824:   1%|▏         | 2/156 [00:12<15:18,  5.97s/it]
Training Epoch: 11/100, batch 3/313 completed (loss: 0.0966:   1%|▏         | 2/156 [00:12<15:19,  5.97s/it]
Training Epoch: 11/100, batch 3/313 completed (loss: 0.3199:   1%|▏         | 2/156 [00:12<15:21,  5.98s/it]
Training Epoch: 11/100, batch 4/313 completed (loss: 0.1711:   1%|▏         | 2/156 [00:14<15:18,  5.97s/it]
Training Epoch: 11/100, batch 4/313 completed (loss: 0.2689:   1%|▏         | 2/156 [00:14<15:19,  5.97s/it]
Training Epoch: 11/100, batch 4/313 completed (loss: 0.005:   1%|▏         | 2/156 [00:14<15:19,  5.97s/it] 
Training Epoch: 11/100, batch 4/313 completed (loss: 0.1415:   1%|▏         | 2/156 [00:14<15:21,  5.98s/it]
Training Epoch: 11/100, batch 4/313 completed (loss: 0.1415:   2%|▏         | 3/156 [00:18<15:16,  5.99s/it]
Training Epoch: 11/100, batch 4/313 completed (loss: 0.005:   2%|▏         | 3/156 [00:17<15:15,  5.98s/it]
Training Epoch: 11/100, batch 4/313 completed (loss: 0.2689:   2%|▏         | 3/156 [00:18<15:15,  5.98s/it]
Training Epoch: 11/100, batch 4/313 completed (loss: 0.1711:   2%|▏         | 3/156 [00:18<15:17,  6.00s/it]
Training Epoch: 11/100, batch 5/313 completed (loss: 0.1241:   2%|▏         | 3/156 [00:18<15:15,  5.98s/it]
Training Epoch: 11/100, batch 5/313 completed (loss: 0.0132:   2%|▏         | 3/156 [00:18<15:17,  6.00s/it]
Training Epoch: 11/100, batch 5/313 completed (loss: 0.3639:   2%|▏         | 3/156 [00:18<15:15,  5.98s/it]
Training Epoch: 11/100, batch 5/313 completed (loss: 0.0077:   2%|▏         | 3/156 [00:18<15:16,  5.99s/it]
Training Epoch: 11/100, batch 6/313 completed (loss: 0.4566:   2%|▏         | 3/156 [00:20<15:17,  6.00s/it]
Training Epoch: 11/100, batch 6/313 completed (loss: 0.0195:   2%|▏         | 3/156 [00:20<15:15,  5.98s/it]
Training Epoch: 11/100, batch 6/313 completed (loss: 0.0239:   2%|▏         | 3/156 [00:20<15:15,  5.98s/it]
Training Epoch: 11/100, batch 6/313 completed (loss: 0.2534:   2%|▏         | 3/156 [00:20<15:16,  5.99s/it]
Training Epoch: 11/100, batch 6/313 completed (loss: 0.2534:   3%|▎         | 4/156 [00:23<14:50,  5.86s/it]
Training Epoch: 11/100, batch 6/313 completed (loss: 0.0239:   3%|▎         | 4/156 [00:23<14:50,  5.86s/it]
Training Epoch: 11/100, batch 6/313 completed (loss: 0.4566:   3%|▎         | 4/156 [00:23<14:55,  5.89s/it]
Training Epoch: 11/100, batch 6/313 completed (loss: 0.0195:   3%|▎         | 4/156 [00:23<14:56,  5.90s/it]
Training Epoch: 11/100, batch 7/313 completed (loss: 0.1693:   3%|▎         | 4/156 [00:23<14:50,  5.86s/it]
Training Epoch: 11/100, batch 7/313 completed (loss: 0.0993:   3%|▎         | 4/156 [00:23<14:55,  5.89s/it]
Training Epoch: 11/100, batch 7/313 completed (loss: 0.1857:   3%|▎         | 4/156 [00:23<14:56,  5.90s/it]
Training Epoch: 11/100, batch 7/313 completed (loss: 0.011:   3%|▎         | 4/156 [00:23<14:50,  5.86s/it] 
Training Epoch: 11/100, batch 8/313 completed (loss: 0.0651:   3%|▎         | 4/156 [00:26<14:55,  5.89s/it]
Training Epoch: 11/100, batch 8/313 completed (loss: 0.3356:   3%|▎         | 4/156 [00:26<14:50,  5.86s/it]
Training Epoch: 11/100, batch 8/313 completed (loss: 0.0399:   3%|▎         | 4/156 [00:26<14:56,  5.90s/it]
Training Epoch: 11/100, batch 8/313 completed (loss: 0.0249:   3%|▎         | 4/156 [00:26<14:50,  5.86s/it]
Training Epoch: 11/100, batch 8/313 completed (loss: 0.0399:   3%|▎         | 5/156 [00:29<14:45,  5.87s/it]
Training Epoch: 11/100, batch 8/313 completed (loss: 0.3356:   3%|▎         | 5/156 [00:29<14:47,  5.88s/it]
Training Epoch: 11/100, batch 8/313 completed (loss: 0.0651:   3%|▎         | 5/156 [00:29<14:45,  5.87s/it]
Training Epoch: 11/100, batch 8/313 completed (loss: 0.0249:   3%|▎         | 5/156 [00:29<14:48,  5.88s/it]
Training Epoch: 11/100, batch 9/313 completed (loss: 0.0692:   3%|▎         | 5/156 [00:29<14:45,  5.87s/it]
Training Epoch: 11/100, batch 9/313 completed (loss: 0.1244:   3%|▎         | 5/156 [00:29<14:45,  5.87s/it]
Training Epoch: 11/100, batch 9/313 completed (loss: 0.3138:   3%|▎         | 5/156 [00:29<14:47,  5.88s/it]
Training Epoch: 11/100, batch 9/313 completed (loss: 0.0302:   3%|▎         | 5/156 [00:29<14:48,  5.88s/it]
Training Epoch: 11/100, batch 10/313 completed (loss: 0.0137:   3%|▎         | 5/156 [00:32<14:45,  5.87s/it]
Training Epoch: 11/100, batch 10/313 completed (loss: 0.0597:   3%|▎         | 5/156 [00:32<14:47,  5.88s/it]
Training Epoch: 11/100, batch 10/313 completed (loss: 0.0394:   3%|▎         | 5/156 [00:32<14:45,  5.87s/it]
Training Epoch: 11/100, batch 10/313 completed (loss: 0.2501:   3%|▎         | 5/156 [00:32<14:48,  5.88s/it]
Training Epoch: 11/100, batch 10/313 completed (loss: 0.0137:   4%|▍         | 6/156 [00:35<14:45,  5.90s/it]
Training Epoch: 11/100, batch 10/313 completed (loss: 0.0394:   4%|▍         | 6/156 [00:35<14:45,  5.90s/it]
Training Epoch: 11/100, batch 10/313 completed (loss: 0.0597:   4%|▍         | 6/156 [00:35<14:46,  5.91s/it]
Training Epoch: 11/100, batch 10/313 completed (loss: 0.2501:   4%|▍         | 6/156 [00:35<14:47,  5.92s/it]
Training Epoch: 11/100, batch 11/313 completed (loss: 0.1886:   4%|▍         | 6/156 [00:35<14:45,  5.90s/it]
Training Epoch: 11/100, batch 11/313 completed (loss: 0.2658:   4%|▍         | 6/156 [00:35<14:46,  5.91s/it]
Training Epoch: 11/100, batch 11/313 completed (loss: 0.2083:   4%|▍         | 6/156 [00:35<14:45,  5.90s/it]
Training Epoch: 11/100, batch 11/313 completed (loss: 0.0345:   4%|▍         | 6/156 [00:35<14:47,  5.92s/it]
Training Epoch: 11/100, batch 12/313 completed (loss: 0.0319:   4%|▍         | 6/156 [00:38<14:45,  5.90s/it]
Training Epoch: 11/100, batch 12/313 completed (loss: 0.0299:   4%|▍         | 6/156 [00:38<14:46,  5.91s/it]
Training Epoch: 11/100, batch 12/313 completed (loss: 0.06:   4%|▍         | 6/156 [00:38<14:45,  5.90s/it]  
Training Epoch: 11/100, batch 12/313 completed (loss: 0.0036:   4%|▍         | 6/156 [00:38<14:47,  5.92s/it]
Training Epoch: 11/100, batch 12/313 completed (loss: 0.0299:   4%|▍         | 7/156 [00:41<14:37,  5.89s/it]
Training Epoch: 11/100, batch 12/313 completed (loss: 0.06:   4%|▍         | 7/156 [00:41<14:37,  5.89s/it]
Training Epoch: 11/100, batch 12/313 completed (loss: 0.0036:   4%|▍         | 7/156 [00:41<14:38,  5.90s/it]
Training Epoch: 11/100, batch 12/313 completed (loss: 0.0319:   4%|▍         | 7/156 [00:41<14:37,  5.89s/it]
Training Epoch: 11/100, batch 13/313 completed (loss: 0.0507:   4%|▍         | 7/156 [00:41<14:37,  5.89s/it]
Training Epoch: 11/100, batch 13/313 completed (loss: 0.0625:   4%|▍         | 7/156 [00:41<14:37,  5.89s/it]
Training Epoch: 11/100, batch 13/313 completed (loss: 0.0144:   4%|▍         | 7/156 [00:41<14:37,  5.89s/it]
Training Epoch: 11/100, batch 13/313 completed (loss: 0.0316:   4%|▍         | 7/156 [00:41<14:38,  5.90s/it]
Training Epoch: 11/100, batch 14/313 completed (loss: 0.5944:   4%|▍         | 7/156 [00:44<14:37,  5.89s/it]
Training Epoch: 11/100, batch 14/313 completed (loss: 0.3566:   4%|▍         | 7/156 [00:44<14:37,  5.89s/it]
Training Epoch: 11/100, batch 14/313 completed (loss: 0.0514:   4%|▍         | 7/156 [00:44<14:37,  5.89s/it]
Training Epoch: 11/100, batch 14/313 completed (loss: 0.4077:   4%|▍         | 7/156 [00:44<14:38,  5.90s/it]
Training Epoch: 11/100, batch 14/313 completed (loss: 0.3566:   5%|▌         | 8/156 [00:47<14:21,  5.82s/it]
Training Epoch: 11/100, batch 14/313 completed (loss: 0.4077:   5%|▌         | 8/156 [00:47<14:21,  5.82s/it]
Training Epoch: 11/100, batch 14/313 completed (loss: 0.0514:   5%|▌         | 8/156 [00:47<14:21,  5.82s/it]
Training Epoch: 11/100, batch 14/313 completed (loss: 0.5944:   5%|▌         | 8/156 [00:47<14:21,  5.82s/it]
Training Epoch: 11/100, batch 15/313 completed (loss: 0.0153:   5%|▌         | 8/156 [00:47<14:21,  5.82s/it]
Training Epoch: 11/100, batch 15/313 completed (loss: 0.2195:   5%|▌         | 8/156 [00:47<14:21,  5.82s/it]
Training Epoch: 11/100, batch 15/313 completed (loss: 0.0165:   5%|▌         | 8/156 [00:47<14:21,  5.82s/it]
Training Epoch: 11/100, batch 15/313 completed (loss: 0.0086:   5%|▌         | 8/156 [00:47<14:21,  5.82s/it]
Training Epoch: 11/100, batch 16/313 completed (loss: 0.0799:   5%|▌         | 8/156 [00:50<14:21,  5.82s/it]
Training Epoch: 11/100, batch 16/313 completed (loss: 0.1197:   5%|▌         | 8/156 [00:50<14:21,  5.82s/it]
Training Epoch: 11/100, batch 16/313 completed (loss: 0.0681:   5%|▌         | 8/156 [00:50<14:21,  5.82s/it]
Training Epoch: 11/100, batch 16/313 completed (loss: 0.0066:   5%|▌         | 8/156 [00:50<14:21,  5.82s/it]
Training Epoch: 11/100, batch 16/313 completed (loss: 0.0681:   6%|▌         | 9/156 [00:53<14:23,  5.87s/it]
Training Epoch: 11/100, batch 16/313 completed (loss: 0.0066:   6%|▌         | 9/156 [00:53<14:23,  5.88s/it]
Training Epoch: 11/100, batch 16/313 completed (loss: 0.1197:   6%|▌         | 9/156 [00:53<14:23,  5.88s/it]
Training Epoch: 11/100, batch 16/313 completed (loss: 0.0799:   6%|▌         | 9/156 [00:53<14:23,  5.88s/it]
Training Epoch: 11/100, batch 17/313 completed (loss: 0.01:   6%|▌         | 9/156 [00:53<14:23,  5.88s/it]  
Training Epoch: 11/100, batch 17/313 completed (loss: 0.0613:   6%|▌         | 9/156 [00:53<14:23,  5.88s/it]
Training Epoch: 11/100, batch 17/313 completed (loss: 0.2415:   6%|▌         | 9/156 [00:53<14:23,  5.87s/it]
Training Epoch: 11/100, batch 17/313 completed (loss: 0.1614:   6%|▌         | 9/156 [00:53<14:23,  5.88s/it]
Training Epoch: 11/100, batch 18/313 completed (loss: 0.1087:   6%|▌         | 9/156 [00:55<14:23,  5.88s/it]
Training Epoch: 11/100, batch 18/313 completed (loss: 0.0827:   6%|▌         | 9/156 [00:55<14:23,  5.87s/it]
Training Epoch: 11/100, batch 18/313 completed (loss: 0.3366:   6%|▌         | 9/156 [00:55<14:23,  5.88s/it]
Training Epoch: 11/100, batch 18/313 completed (loss: 0.0191:   6%|▌         | 9/156 [00:55<14:23,  5.88s/it]
Training Epoch: 11/100, batch 18/313 completed (loss: 0.0827:   6%|▋         | 10/156 [00:58<14:16,  5.87s/it]
Training Epoch: 11/100, batch 18/313 completed (loss: 0.0191:   6%|▋         | 10/156 [00:58<14:16,  5.87s/it]
Training Epoch: 11/100, batch 18/313 completed (loss: 0.3366:   6%|▋         | 10/156 [00:58<14:16,  5.87s/it]
Training Epoch: 11/100, batch 18/313 completed (loss: 0.1087:   6%|▋         | 10/156 [00:58<14:16,  5.87s/it]
Training Epoch: 11/100, batch 19/313 completed (loss: 0.0465:   6%|▋         | 10/156 [00:58<14:16,  5.87s/it]
Training Epoch: 11/100, batch 19/313 completed (loss: 0.1056:   6%|▋         | 10/156 [00:58<14:16,  5.87s/it]
Training Epoch: 11/100, batch 19/313 completed (loss: 0.3721:   6%|▋         | 10/156 [00:58<14:16,  5.87s/it]
Training Epoch: 11/100, batch 19/313 completed (loss: 0.0324:   6%|▋         | 10/156 [00:59<14:16,  5.87s/it]
Training Epoch: 11/100, batch 20/313 completed (loss: 0.1827:   6%|▋         | 10/156 [01:01<14:16,  5.87s/it]
Training Epoch: 11/100, batch 20/313 completed (loss: 0.0817:   6%|▋         | 10/156 [01:01<14:16,  5.87s/it]
Training Epoch: 11/100, batch 20/313 completed (loss: 0.1098:   6%|▋         | 10/156 [01:01<14:16,  5.87s/it]
Training Epoch: 11/100, batch 20/313 completed (loss: 0.1506:   6%|▋         | 10/156 [01:01<14:16,  5.87s/it]
Training Epoch: 11/100, batch 20/313 completed (loss: 0.1506:   7%|▋         | 11/156 [01:04<14:18,  5.92s/it]
Training Epoch: 11/100, batch 20/313 completed (loss: 0.0817:   7%|▋         | 11/156 [01:04<14:17,  5.92s/it]
Training Epoch: 11/100, batch 20/313 completed (loss: 0.1098:   7%|▋         | 11/156 [01:04<14:17,  5.92s/it]
Training Epoch: 11/100, batch 20/313 completed (loss: 0.1827:   7%|▋         | 11/156 [01:04<14:18,  5.92s/it]
Training Epoch: 11/100, batch 21/313 completed (loss: 0.1328:   7%|▋         | 11/156 [01:05<14:17,  5.92s/it]
Training Epoch: 11/100, batch 21/313 completed (loss: 0.043:   7%|▋         | 11/156 [01:05<14:18,  5.92s/it] 
Training Epoch: 11/100, batch 21/313 completed (loss: 0.2037:   7%|▋         | 11/156 [01:05<14:17,  5.92s/it]
Training Epoch: 11/100, batch 21/313 completed (loss: 0.0539:   7%|▋         | 11/156 [01:05<14:18,  5.92s/it]
Training Epoch: 11/100, batch 22/313 completed (loss: 0.0466:   7%|▋         | 11/156 [01:07<14:18,  5.92s/it]
Training Epoch: 11/100, batch 22/313 completed (loss: 0.0169:   7%|▋         | 11/156 [01:07<14:17,  5.92s/it]
Training Epoch: 11/100, batch 22/313 completed (loss: 0.1573:   7%|▋         | 11/156 [01:07<14:17,  5.92s/it]
Training Epoch: 11/100, batch 22/313 completed (loss: 0.191:   7%|▋         | 11/156 [01:07<14:18,  5.92s/it] 
Training Epoch: 11/100, batch 22/313 completed (loss: 0.191:   8%|▊         | 12/156 [01:10<14:04,  5.86s/it]
Training Epoch: 11/100, batch 22/313 completed (loss: 0.1573:   8%|▊         | 12/156 [01:10<14:04,  5.86s/it]
Training Epoch: 11/100, batch 22/313 completed (loss: 0.0169:   8%|▊         | 12/156 [01:10<14:04,  5.87s/it]
Training Epoch: 11/100, batch 22/313 completed (loss: 0.0466:   8%|▊         | 12/156 [01:10<14:05,  5.87s/it]
Training Epoch: 11/100, batch 23/313 completed (loss: 0.2207:   8%|▊         | 12/156 [01:10<14:05,  5.87s/it]
Training Epoch: 11/100, batch 23/313 completed (loss: 0.1009:   8%|▊         | 12/156 [01:10<14:04,  5.86s/it]
Training Epoch: 11/100, batch 23/313 completed (loss: 0.0977:   8%|▊         | 12/156 [01:10<14:04,  5.87s/it]
Training Epoch: 11/100, batch 23/313 completed (loss: 0.0084:   8%|▊         | 12/156 [01:10<14:04,  5.86s/it]
Training Epoch: 11/100, batch 24/313 completed (loss: 0.1379:   8%|▊         | 12/156 [01:13<14:05,  5.87s/it]
Training Epoch: 11/100, batch 24/313 completed (loss: 0.2119:   8%|▊         | 12/156 [01:13<14:04,  5.86s/it]
Training Epoch: 11/100, batch 24/313 completed (loss: 0.028:   8%|▊         | 12/156 [01:13<14:04,  5.87s/it] 
Training Epoch: 11/100, batch 24/313 completed (loss: 0.1459:   8%|▊         | 12/156 [01:13<14:04,  5.86s/it]
Training Epoch: 11/100, batch 24/313 completed (loss: 0.028:   8%|▊         | 13/156 [01:16<14:01,  5.88s/it]
Training Epoch: 11/100, batch 24/313 completed (loss: 0.2119:   8%|▊         | 13/156 [01:16<14:01,  5.88s/it]
Training Epoch: 11/100, batch 24/313 completed (loss: 0.1459:   8%|▊         | 13/156 [01:16<14:01,  5.89s/it]
Training Epoch: 11/100, batch 24/313 completed (loss: 0.1379:   8%|▊         | 13/156 [01:16<14:01,  5.89s/it]
Training Epoch: 11/100, batch 25/313 completed (loss: 0.3443:   8%|▊         | 13/156 [01:16<14:01,  5.88s/it]
Training Epoch: 11/100, batch 25/313 completed (loss: 0.1089:   8%|▊         | 13/156 [01:16<14:01,  5.89s/it]
Training Epoch: 11/100, batch 25/313 completed (loss: 0.0378:   8%|▊         | 13/156 [01:16<14:01,  5.88s/it]
Training Epoch: 11/100, batch 25/313 completed (loss: 0.0895:   8%|▊         | 13/156 [01:16<14:01,  5.89s/it]
Training Epoch: 11/100, batch 26/313 completed (loss: 0.0361:   8%|▊         | 13/156 [01:19<14:01,  5.89s/it]
Training Epoch: 11/100, batch 26/313 completed (loss: 0.1063:   8%|▊         | 13/156 [01:19<14:01,  5.88s/it]
Training Epoch: 11/100, batch 26/313 completed (loss: 0.0557:   8%|▊         | 13/156 [01:19<14:01,  5.88s/it]
Training Epoch: 11/100, batch 26/313 completed (loss: 0.0676:   8%|▊         | 13/156 [01:20<14:01,  5.89s/it]
Training Epoch: 11/100, batch 26/313 completed (loss: 0.0676:   9%|▉         | 14/156 [01:22<14:08,  5.97s/it]
Training Epoch: 11/100, batch 26/313 completed (loss: 0.0557:   9%|▉         | 14/156 [01:22<14:08,  5.98s/it]
Training Epoch: 11/100, batch 26/313 completed (loss: 0.1063:   9%|▉         | 14/156 [01:22<14:08,  5.98s/it]
Training Epoch: 11/100, batch 26/313 completed (loss: 0.0361:   9%|▉         | 14/156 [01:22<14:08,  5.98s/it]
Training Epoch: 11/100, batch 27/313 completed (loss: 0.0285:   9%|▉         | 14/156 [01:22<14:08,  5.98s/it]
Training Epoch: 11/100, batch 27/313 completed (loss: 0.1033:   9%|▉         | 14/156 [01:22<14:08,  5.98s/it]
Training Epoch: 11/100, batch 27/313 completed (loss: 0.1004:   9%|▉         | 14/156 [01:22<14:08,  5.98s/it]
Training Epoch: 11/100, batch 27/313 completed (loss: 0.1753:   9%|▉         | 14/156 [01:22<14:08,  5.97s/it]
Training Epoch: 11/100, batch 28/313 completed (loss: 0.1195:   9%|▉         | 14/156 [01:25<14:08,  5.98s/it]
Training Epoch: 11/100, batch 28/313 completed (loss: 0.0747:   9%|▉         | 14/156 [01:25<14:08,  5.98s/it]
Training Epoch: 11/100, batch 28/313 completed (loss: 0.2253:   9%|▉         | 14/156 [01:25<14:08,  5.98s/it]
Training Epoch: 11/100, batch 28/313 completed (loss: 0.0016:   9%|▉         | 14/156 [01:25<14:08,  5.97s/it]
Training Epoch: 11/100, batch 28/313 completed (loss: 0.0016:  10%|▉         | 15/156 [01:28<14:00,  5.96s/it]
Training Epoch: 11/100, batch 28/313 completed (loss: 0.2253:  10%|▉         | 15/156 [01:28<14:00,  5.96s/it]
Training Epoch: 11/100, batch 28/313 completed (loss: 0.1195:  10%|▉         | 15/156 [01:28<14:00,  5.96s/it]
Training Epoch: 11/100, batch 28/313 completed (loss: 0.0747:  10%|▉         | 15/156 [01:28<14:01,  5.96s/it]
Training Epoch: 11/100, batch 29/313 completed (loss: 0.1226:  10%|▉         | 15/156 [01:28<14:01,  5.96s/it]
Training Epoch: 11/100, batch 29/313 completed (loss: 0.2304:  10%|▉         | 15/156 [01:28<14:00,  5.96s/it]
Training Epoch: 11/100, batch 29/313 completed (loss: 0.1229:  10%|▉         | 15/156 [01:28<14:00,  5.96s/it]
Training Epoch: 11/100, batch 29/313 completed (loss: 0.28:  10%|▉         | 15/156 [01:28<14:00,  5.96s/it]  
Training Epoch: 11/100, batch 30/313 completed (loss: 0.0158:  10%|▉         | 15/156 [01:31<14:01,  5.96s/it]
Training Epoch: 11/100, batch 30/313 completed (loss: 0.0255:  10%|▉         | 15/156 [01:31<14:00,  5.96s/it]
Training Epoch: 11/100, batch 30/313 completed (loss: 0.2384:  10%|▉         | 15/156 [01:31<14:00,  5.96s/it]
Training Epoch: 11/100, batch 30/313 completed (loss: 0.0696:  10%|▉         | 15/156 [01:31<14:00,  5.96s/it]
Training Epoch: 11/100, batch 30/313 completed (loss: 0.0696:  10%|█         | 16/156 [01:34<14:05,  6.04s/it]
Training Epoch: 11/100, batch 30/313 completed (loss: 0.2384:  10%|█         | 16/156 [01:34<14:06,  6.05s/it]
Training Epoch: 11/100, batch 30/313 completed (loss: 0.0158:  10%|█         | 16/156 [01:34<14:06,  6.05s/it]
Training Epoch: 11/100, batch 30/313 completed (loss: 0.0255:  10%|█         | 16/156 [01:35<14:06,  6.05s/it]
Training Epoch: 11/100, batch 31/313 completed (loss: 0.2805:  10%|█         | 16/156 [01:35<14:06,  6.05s/it]
Training Epoch: 11/100, batch 31/313 completed (loss: 0.1855:  10%|█         | 16/156 [01:35<14:06,  6.05s/it]
Training Epoch: 11/100, batch 31/313 completed (loss: 0.0352:  10%|█         | 16/156 [01:35<14:06,  6.05s/it]
Training Epoch: 11/100, batch 31/313 completed (loss: 0.0962:  10%|█         | 16/156 [01:35<14:05,  6.04s/it]
Training Epoch: 11/100, batch 32/313 completed (loss: 0.0506:  10%|█         | 16/156 [01:37<14:06,  6.05s/it]
Training Epoch: 11/100, batch 32/313 completed (loss: 0.0284:  10%|█         | 16/156 [01:37<14:06,  6.05s/it]
Training Epoch: 11/100, batch 32/313 completed (loss: 0.304:  10%|█         | 16/156 [01:37<14:06,  6.05s/it] 
Training Epoch: 11/100, batch 32/313 completed (loss: 0.1186:  10%|█         | 16/156 [01:37<14:05,  6.04s/it]
Training Epoch: 11/100, batch 32/313 completed (loss: 0.1186:  11%|█         | 17/156 [01:40<13:50,  5.97s/it]
Training Epoch: 11/100, batch 32/313 completed (loss: 0.304:  11%|█         | 17/156 [01:40<13:50,  5.97s/it]
Training Epoch: 11/100, batch 32/313 completed (loss: 0.0284:  11%|█         | 17/156 [01:40<13:50,  5.97s/it]
Training Epoch: 11/100, batch 32/313 completed (loss: 0.0506:  11%|█         | 17/156 [01:40<13:50,  5.97s/it]
Training Epoch: 11/100, batch 33/313 completed (loss: 0.661:  11%|█         | 17/156 [01:40<13:50,  5.97s/it]
Training Epoch: 11/100, batch 33/313 completed (loss: 0.0287:  11%|█         | 17/156 [01:40<13:50,  5.97s/it]
Training Epoch: 11/100, batch 33/313 completed (loss: 0.0367:  11%|█         | 17/156 [01:40<13:50,  5.97s/it]
Training Epoch: 11/100, batch 33/313 completed (loss: 0.0616:  11%|█         | 17/156 [01:40<13:50,  5.97s/it]
Training Epoch: 11/100, batch 34/313 completed (loss: 0.067:  11%|█         | 17/156 [01:43<13:50,  5.97s/it] 
Training Epoch: 11/100, batch 34/313 completed (loss: 0.045:  11%|█         | 17/156 [01:43<13:50,  5.97s/it] 
Training Epoch: 11/100, batch 34/313 completed (loss: 0.2067:  11%|█         | 17/156 [01:43<13:50,  5.97s/it]
Training Epoch: 11/100, batch 34/313 completed (loss: 0.221:  11%|█         | 17/156 [01:43<13:50,  5.97s/it] 
Training Epoch: 11/100, batch 34/313 completed (loss: 0.045:  12%|█▏        | 18/156 [01:46<13:43,  5.97s/it]
Training Epoch: 11/100, batch 34/313 completed (loss: 0.067:  12%|█▏        | 18/156 [01:46<13:43,  5.97s/it]
Training Epoch: 11/100, batch 34/313 completed (loss: 0.2067:  12%|█▏        | 18/156 [01:46<13:43,  5.97s/it]
Training Epoch: 11/100, batch 34/313 completed (loss: 0.221:  12%|█▏        | 18/156 [01:46<13:44,  5.97s/it]
Training Epoch: 11/100, batch 35/313 completed (loss: 0.0331:  12%|█▏        | 18/156 [01:46<13:43,  5.97s/it]
Training Epoch: 11/100, batch 35/313 completed (loss: 0.342:  12%|█▏        | 18/156 [01:46<13:43,  5.97s/it]
Training Epoch: 11/100, batch 35/313 completed (loss: 0.1834:  12%|█▏        | 18/156 [01:46<13:43,  5.97s/it]
Training Epoch: 11/100, batch 35/313 completed (loss: 0.0906:  12%|█▏        | 18/156 [01:46<13:44,  5.97s/it]
Training Epoch: 11/100, batch 36/313 completed (loss: 0.0565:  12%|█▏        | 18/156 [01:50<13:43,  5.97s/it]
Training Epoch: 11/100, batch 36/313 completed (loss: 0.4131:  12%|█▏        | 18/156 [01:50<13:43,  5.97s/it]
Training Epoch: 11/100, batch 36/313 completed (loss: 0.0187:  12%|█▏        | 18/156 [01:50<13:43,  5.97s/it]
Training Epoch: 11/100, batch 36/313 completed (loss: 0.0315:  12%|█▏        | 18/156 [01:50<13:44,  5.97s/it]
Training Epoch: 11/100, batch 36/313 completed (loss: 0.0315:  12%|█▏        | 19/156 [01:53<13:50,  6.07s/it]
Training Epoch: 11/100, batch 36/313 completed (loss: 0.0565:  12%|█▏        | 19/156 [01:53<13:50,  6.07s/it]
Training Epoch: 11/100, batch 36/313 completed (loss: 0.0187:  12%|█▏        | 19/156 [01:53<13:51,  6.07s/it]
Training Epoch: 11/100, batch 36/313 completed (loss: 0.4131:  12%|█▏        | 19/156 [01:53<13:51,  6.07s/it]
Training Epoch: 11/100, batch 37/313 completed (loss: 0.0458:  12%|█▏        | 19/156 [01:53<13:51,  6.07s/it]
Training Epoch: 11/100, batch 37/313 completed (loss: 0.0627:  12%|█▏        | 19/156 [01:53<13:50,  6.07s/it]
Training Epoch: 11/100, batch 37/313 completed (loss: 0.1447:  12%|█▏        | 19/156 [01:53<13:51,  6.07s/it]
Training Epoch: 11/100, batch 37/313 completed (loss: 0.3601:  12%|█▏        | 19/156 [01:53<13:50,  6.07s/it]
Training Epoch: 11/100, batch 38/313 completed (loss: 0.0799:  12%|█▏        | 19/156 [01:56<13:51,  6.07s/it]
Training Epoch: 11/100, batch 38/313 completed (loss: 0.656:  12%|█▏        | 19/156 [01:56<13:51,  6.07s/it] 
Training Epoch: 11/100, batch 38/313 completed (loss: 0.1836:  12%|█▏        | 19/156 [01:56<13:50,  6.07s/it]
Training Epoch: 11/100, batch 38/313 completed (loss: 0.036:  12%|█▏        | 19/156 [01:56<13:50,  6.07s/it] 
Training Epoch: 11/100, batch 38/313 completed (loss: 0.036:  13%|█▎        | 20/156 [01:58<13:34,  5.99s/it]
Training Epoch: 11/100, batch 38/313 completed (loss: 0.1836:  13%|█▎        | 20/156 [01:58<13:34,  5.99s/it]
Training Epoch: 11/100, batch 38/313 completed (loss: 0.0799:  13%|█▎        | 20/156 [01:58<13:34,  5.99s/it]
Training Epoch: 11/100, batch 38/313 completed (loss: 0.656:  13%|█▎        | 20/156 [01:58<13:34,  5.99s/it]
Training Epoch: 11/100, batch 39/313 completed (loss: 0.0357:  13%|█▎        | 20/156 [01:58<13:34,  5.99s/it]
Training Epoch: 11/100, batch 39/313 completed (loss: 0.0963:  13%|█▎        | 20/156 [01:58<13:34,  5.99s/it]
Training Epoch: 11/100, batch 39/313 completed (loss: 0.0912:  13%|█▎        | 20/156 [01:58<13:34,  5.99s/it]
Training Epoch: 11/100, batch 39/313 completed (loss: 0.1241:  13%|█▎        | 20/156 [01:58<13:34,  5.99s/it]
Training Epoch: 11/100, batch 40/313 completed (loss: 0.1738:  13%|█▎        | 20/156 [02:01<13:34,  5.99s/it]
Training Epoch: 11/100, batch 40/313 completed (loss: 0.0398:  13%|█▎        | 20/156 [02:01<13:34,  5.99s/it]
Training Epoch: 11/100, batch 40/313 completed (loss: 0.054:  13%|█▎        | 20/156 [02:01<13:34,  5.99s/it] 
Training Epoch: 11/100, batch 40/313 completed (loss: 0.0501:  13%|█▎        | 20/156 [02:02<13:34,  5.99s/it]
Training Epoch: 11/100, batch 40/313 completed (loss: 0.0501:  13%|█▎        | 21/156 [02:05<13:34,  6.03s/it]
Training Epoch: 11/100, batch 40/313 completed (loss: 0.054:  13%|█▎        | 21/156 [02:04<13:34,  6.03s/it]
Training Epoch: 11/100, batch 40/313 completed (loss: 0.0398:  13%|█▎        | 21/156 [02:04<13:34,  6.03s/it]
Training Epoch: 11/100, batch 40/313 completed (loss: 0.1738:  13%|█▎        | 21/156 [02:04<13:34,  6.03s/it]
Training Epoch: 11/100, batch 41/313 completed (loss: 0.311:  13%|█▎        | 21/156 [02:05<13:34,  6.03s/it]
Training Epoch: 11/100, batch 41/313 completed (loss: 0.026:  13%|█▎        | 21/156 [02:05<13:34,  6.03s/it] 
Training Epoch: 11/100, batch 41/313 completed (loss: 0.0165:  13%|█▎        | 21/156 [02:05<13:34,  6.03s/it]
Training Epoch: 11/100, batch 41/313 completed (loss: 0.0706:  13%|█▎        | 21/156 [02:05<13:34,  6.03s/it]
Training Epoch: 11/100, batch 42/313 completed (loss: 0.1039:  13%|█▎        | 21/156 [02:08<13:34,  6.03s/it]
Training Epoch: 11/100, batch 42/313 completed (loss: 0.192:  13%|█▎        | 21/156 [02:08<13:34,  6.03s/it]
Training Epoch: 11/100, batch 42/313 completed (loss: 0.0135:  13%|█▎        | 21/156 [02:08<13:34,  6.03s/it]
Training Epoch: 11/100, batch 42/313 completed (loss: 0.06:  13%|█▎        | 21/156 [02:08<13:34,  6.03s/it]  
Training Epoch: 11/100, batch 42/313 completed (loss: 0.06:  14%|█▍        | 22/156 [02:11<13:31,  6.05s/it]
Training Epoch: 11/100, batch 42/313 completed (loss: 0.1039:  14%|█▍        | 22/156 [02:11<13:30,  6.05s/it]
Training Epoch: 11/100, batch 42/313 completed (loss: 0.0135:  14%|█▍        | 22/156 [02:11<13:31,  6.05s/it]
Training Epoch: 11/100, batch 42/313 completed (loss: 0.192:  14%|█▍        | 22/156 [02:11<13:31,  6.06s/it]
Training Epoch: 11/100, batch 43/313 completed (loss: 0.1118:  14%|█▍        | 22/156 [02:11<13:30,  6.05s/it]
Training Epoch: 11/100, batch 43/313 completed (loss: 0.0406:  14%|█▍        | 22/156 [02:11<13:31,  6.06s/it]
Training Epoch: 11/100, batch 43/313 completed (loss: 0.1029:  14%|█▍        | 22/156 [02:11<13:31,  6.05s/it]
Training Epoch: 11/100, batch 43/313 completed (loss: 0.2277:  14%|█▍        | 22/156 [02:11<13:31,  6.05s/it]
Training Epoch: 11/100, batch 44/313 completed (loss: 0.2891:  14%|█▍        | 22/156 [02:14<13:31,  6.05s/it]
Training Epoch: 11/100, batch 44/313 completed (loss: 0.5122:  14%|█▍        | 22/156 [02:14<13:30,  6.05s/it]
Training Epoch: 11/100, batch 44/313 completed (loss: 0.0766:  14%|█▍        | 22/156 [02:14<13:31,  6.06s/it]
Training Epoch: 11/100, batch 44/313 completed (loss: 0.0126:  14%|█▍        | 22/156 [02:14<13:31,  6.05s/it]
Training Epoch: 11/100, batch 44/313 completed (loss: 0.0126:  15%|█▍        | 23/156 [02:16<13:16,  5.99s/it]
Training Epoch: 11/100, batch 44/313 completed (loss: 0.2891:  15%|█▍        | 23/156 [02:16<13:16,  5.99s/it]
Training Epoch: 11/100, batch 44/313 completed (loss: 0.5122:  15%|█▍        | 23/156 [02:16<13:16,  5.99s/it]
Training Epoch: 11/100, batch 44/313 completed (loss: 0.0766:  15%|█▍        | 23/156 [02:16<13:16,  5.99s/it]
Training Epoch: 11/100, batch 45/313 completed (loss: 0.1481:  15%|█▍        | 23/156 [02:16<13:16,  5.99s/it]
Training Epoch: 11/100, batch 45/313 completed (loss: 0.2141:  15%|█▍        | 23/156 [02:16<13:16,  5.99s/it]
Training Epoch: 11/100, batch 45/313 completed (loss: 0.1446:  15%|█▍        | 23/156 [02:16<13:16,  5.99s/it]
Training Epoch: 11/100, batch 45/313 completed (loss: 0.05:  15%|█▍        | 23/156 [02:17<13:16,  5.99s/it]  
Training Epoch: 11/100, batch 46/313 completed (loss: 0.2158:  15%|█▍        | 23/156 [02:20<13:16,  5.99s/it]
Training Epoch: 11/100, batch 46/313 completed (loss: 0.1118:  15%|█▍        | 23/156 [02:20<13:16,  5.99s/it]
Training Epoch: 11/100, batch 46/313 completed (loss: 0.0806:  15%|█▍        | 23/156 [02:20<13:16,  5.99s/it]
Training Epoch: 11/100, batch 46/313 completed (loss: 0.1078:  15%|█▍        | 23/156 [02:20<13:16,  5.99s/it]
Training Epoch: 11/100, batch 46/313 completed (loss: 0.1078:  15%|█▌        | 24/156 [02:23<13:27,  6.12s/it]
Training Epoch: 11/100, batch 46/313 completed (loss: 0.0806:  15%|█▌        | 24/156 [02:23<13:27,  6.12s/it]
Training Epoch: 11/100, batch 46/313 completed (loss: 0.2158:  15%|█▌        | 24/156 [02:23<13:27,  6.12s/it]
Training Epoch: 11/100, batch 46/313 completed (loss: 0.1118:  15%|█▌        | 24/156 [02:23<13:27,  6.12s/it]
Training Epoch: 11/100, batch 47/313 completed (loss: 0.3133:  15%|█▌        | 24/156 [02:23<13:27,  6.12s/it]
Training Epoch: 11/100, batch 47/313 completed (loss: 0.0911:  15%|█▌        | 24/156 [02:23<13:27,  6.12s/it]
Training Epoch: 11/100, batch 47/313 completed (loss: 0.0305:  15%|█▌        | 24/156 [02:23<13:27,  6.12s/it]
Training Epoch: 11/100, batch 47/313 completed (loss: 0.0817:  15%|█▌        | 24/156 [02:23<13:27,  6.12s/it]
Training Epoch: 11/100, batch 48/313 completed (loss: 0.0807:  15%|█▌        | 24/156 [02:26<13:27,  6.12s/it]
Training Epoch: 11/100, batch 48/313 completed (loss: 0.2456:  15%|█▌        | 24/156 [02:26<13:27,  6.12s/it]
Training Epoch: 11/100, batch 48/313 completed (loss: 0.0346:  15%|█▌        | 24/156 [02:26<13:27,  6.12s/it]
Training Epoch: 11/100, batch 48/313 completed (loss: 0.2918:  15%|█▌        | 24/156 [02:26<13:27,  6.12s/it]
Training Epoch: 11/100, batch 48/313 completed (loss: 0.2456:  16%|█▌        | 25/156 [02:29<13:09,  6.03s/it]
Training Epoch: 11/100, batch 48/313 completed (loss: 0.2918:  16%|█▌        | 25/156 [02:29<13:09,  6.03s/it]
Training Epoch: 11/100, batch 48/313 completed (loss: 0.0346:  16%|█▌        | 25/156 [02:29<13:09,  6.03s/it]
Training Epoch: 11/100, batch 48/313 completed (loss: 0.0807:  16%|█▌        | 25/156 [02:29<13:09,  6.03s/it]
Training Epoch: 11/100, batch 49/313 completed (loss: 0.0504:  16%|█▌        | 25/156 [02:29<13:09,  6.03s/it]
Training Epoch: 11/100, batch 49/313 completed (loss: 0.3921:  16%|█▌        | 25/156 [02:29<13:09,  6.03s/it]
Training Epoch: 11/100, batch 49/313 completed (loss: 0.176:  16%|█▌        | 25/156 [02:29<13:09,  6.03s/it] 
Training Epoch: 11/100, batch 49/313 completed (loss: 0.0289:  16%|█▌        | 25/156 [02:29<13:09,  6.03s/it]
Training Epoch: 11/100, batch 50/313 completed (loss: 0.1609:  16%|█▌        | 25/156 [02:32<13:09,  6.03s/it]
Training Epoch: 11/100, batch 50/313 completed (loss: 0.0884:  16%|█▌        | 25/156 [02:32<13:09,  6.03s/it]
Training Epoch: 11/100, batch 50/313 completed (loss: 0.0505:  16%|█▌        | 25/156 [02:32<13:09,  6.03s/it]
Training Epoch: 11/100, batch 50/313 completed (loss: 0.1287:  16%|█▌        | 25/156 [02:32<13:09,  6.03s/it]
Training Epoch: 11/100, batch 50/313 completed (loss: 0.1287:  17%|█▋        | 26/156 [02:35<13:06,  6.05s/it]
Training Epoch: 11/100, batch 50/313 completed (loss: 0.0505:  17%|█▋        | 26/156 [02:35<13:06,  6.05s/it]
Training Epoch: 11/100, batch 50/313 completed (loss: 0.1609:  17%|█▋        | 26/156 [02:35<13:06,  6.05s/it]
Training Epoch: 11/100, batch 50/313 completed (loss: 0.0884:  17%|█▋        | 26/156 [02:35<13:06,  6.05s/it]
Training Epoch: 11/100, batch 51/313 completed (loss: 0.0542:  17%|█▋        | 26/156 [02:35<13:06,  6.05s/it]
Training Epoch: 11/100, batch 51/313 completed (loss: 0.0487:  17%|█▋        | 26/156 [02:35<13:06,  6.05s/it]
Training Epoch: 11/100, batch 51/313 completed (loss: 0.0545:  17%|█▋        | 26/156 [02:35<13:06,  6.05s/it]
Training Epoch: 11/100, batch 51/313 completed (loss: 0.06:  17%|█▋        | 26/156 [02:35<13:06,  6.05s/it]  
Training Epoch: 11/100, batch 52/313 completed (loss: 0.1467:  17%|█▋        | 26/156 [02:38<13:06,  6.05s/it]
Training Epoch: 11/100, batch 52/313 completed (loss: 0.0715:  17%|█▋        | 26/156 [02:38<13:06,  6.05s/it]
Training Epoch: 11/100, batch 52/313 completed (loss: 0.0776:  17%|█▋        | 26/156 [02:38<13:06,  6.05s/it]
Training Epoch: 11/100, batch 52/313 completed (loss: 0.0927:  17%|█▋        | 26/156 [02:38<13:06,  6.05s/it]
Training Epoch: 11/100, batch 52/313 completed (loss: 0.0776:  17%|█▋        | 27/156 [02:41<13:03,  6.07s/it]
Training Epoch: 11/100, batch 52/313 completed (loss: 0.1467:  17%|█▋        | 27/156 [02:41<13:03,  6.07s/it]
Training Epoch: 11/100, batch 52/313 completed (loss: 0.0927:  17%|█▋        | 27/156 [02:41<13:03,  6.08s/it]
Training Epoch: 11/100, batch 52/313 completed (loss: 0.0715:  17%|█▋        | 27/156 [02:41<13:03,  6.08s/it]
Training Epoch: 11/100, batch 53/313 completed (loss: 0.0468:  17%|█▋        | 27/156 [02:41<13:03,  6.08s/it]
Training Epoch: 11/100, batch 53/313 completed (loss: 0.0211:  17%|█▋        | 27/156 [02:41<13:03,  6.07s/it]
Training Epoch: 11/100, batch 53/313 completed (loss: 0.1731:  17%|█▋        | 27/156 [02:41<13:03,  6.07s/it]
Training Epoch: 11/100, batch 53/313 completed (loss: 0.1257:  17%|█▋        | 27/156 [02:41<13:03,  6.08s/it]
Training Epoch: 11/100, batch 54/313 completed (loss: 0.443:  17%|█▋        | 27/156 [02:44<13:03,  6.07s/it] 
Training Epoch: 11/100, batch 54/313 completed (loss: 0.4215:  17%|█▋        | 27/156 [02:44<13:03,  6.07s/it]
Training Epoch: 11/100, batch 54/313 completed (loss: 0.2522:  17%|█▋        | 27/156 [02:44<13:03,  6.08s/it]
Training Epoch: 11/100, batch 54/313 completed (loss: 0.0177:  17%|█▋        | 27/156 [02:44<13:03,  6.08s/it]
Training Epoch: 11/100, batch 54/313 completed (loss: 0.4215:  18%|█▊        | 28/156 [02:47<12:47,  6.00s/it]
Training Epoch: 11/100, batch 54/313 completed (loss: 0.0177:  18%|█▊        | 28/156 [02:47<12:48,  6.00s/it]
Training Epoch: 11/100, batch 54/313 completed (loss: 0.443:  18%|█▊        | 28/156 [02:47<12:48,  6.00s/it]
Training Epoch: 11/100, batch 54/313 completed (loss: 0.2522:  18%|█▊        | 28/156 [02:47<12:48,  6.00s/it]
Training Epoch: 11/100, batch 55/313 completed (loss: 0.1279:  18%|█▊        | 28/156 [02:47<12:48,  6.00s/it]
Training Epoch: 11/100, batch 55/313 completed (loss: 0.0462:  18%|█▊        | 28/156 [02:47<12:47,  6.00s/it]
Training Epoch: 11/100, batch 55/313 completed (loss: 0.0547:  18%|█▊        | 28/156 [02:47<12:48,  6.00s/it]
Training Epoch: 11/100, batch 55/313 completed (loss: 0.0239:  18%|█▊        | 28/156 [02:47<12:48,  6.00s/it]
Training Epoch: 11/100, batch 56/313 completed (loss: 0.1293:  18%|█▊        | 28/156 [02:50<12:48,  6.00s/it]
Training Epoch: 11/100, batch 56/313 completed (loss: 0.3513:  18%|█▊        | 28/156 [02:50<12:48,  6.00s/it]
Training Epoch: 11/100, batch 56/313 completed (loss: 0.059:  18%|█▊        | 28/156 [02:50<12:47,  6.00s/it] 
Training Epoch: 11/100, batch 56/313 completed (loss: 0.0187:  18%|█▊        | 28/156 [02:50<12:48,  6.00s/it]
Training Epoch: 11/100, batch 56/313 completed (loss: 0.0187:  19%|█▊        | 29/156 [02:53<12:54,  6.10s/it]
Training Epoch: 11/100, batch 56/313 completed (loss: 0.1293:  19%|█▊        | 29/156 [02:53<12:54,  6.10s/it]
Training Epoch: 11/100, batch 56/313 completed (loss: 0.059:  19%|█▊        | 29/156 [02:53<12:54,  6.10s/it]
Training Epoch: 11/100, batch 56/313 completed (loss: 0.3513:  19%|█▊        | 29/156 [02:53<12:54,  6.10s/it]
Training Epoch: 11/100, batch 57/313 completed (loss: 0.1716:  19%|█▊        | 29/156 [02:53<12:54,  6.10s/it]
Training Epoch: 11/100, batch 57/313 completed (loss: 0.0284:  19%|█▊        | 29/156 [02:53<12:54,  6.10s/it]
Training Epoch: 11/100, batch 57/313 completed (loss: 0.1179:  19%|█▊        | 29/156 [02:53<12:54,  6.10s/it]
Training Epoch: 11/100, batch 57/313 completed (loss: 0.1129:  19%|█▊        | 29/156 [02:53<12:54,  6.10s/it]
Training Epoch: 11/100, batch 58/313 completed (loss: 0.1535:  19%|█▊        | 29/156 [02:56<12:54,  6.10s/it]
Training Epoch: 11/100, batch 58/313 completed (loss: 0.017:  19%|█▊        | 29/156 [02:56<12:54,  6.10s/it] 
Training Epoch: 11/100, batch 58/313 completed (loss: 0.2933:  19%|█▊        | 29/156 [02:56<12:54,  6.10s/it]
Training Epoch: 11/100, batch 58/313 completed (loss: 0.0429:  19%|█▊        | 29/156 [02:56<12:54,  6.10s/it]
Training Epoch: 11/100, batch 58/313 completed (loss: 0.0429:  19%|█▉        | 30/156 [02:59<12:45,  6.08s/it]
Training Epoch: 11/100, batch 58/313 completed (loss: 0.1535:  19%|█▉        | 30/156 [02:59<12:45,  6.08s/it]
Training Epoch: 11/100, batch 58/313 completed (loss: 0.017:  19%|█▉        | 30/156 [02:59<12:45,  6.08s/it]
Training Epoch: 11/100, batch 58/313 completed (loss: 0.2933:  19%|█▉        | 30/156 [02:59<12:45,  6.08s/it]
Training Epoch: 11/100, batch 59/313 completed (loss: 0.4599:  19%|█▉        | 30/156 [02:59<12:45,  6.08s/it]
Training Epoch: 11/100, batch 59/313 completed (loss: 0.1491:  19%|█▉        | 30/156 [02:59<12:45,  6.08s/it]
Training Epoch: 11/100, batch 59/313 completed (loss: 0.0303:  19%|█▉        | 30/156 [02:59<12:45,  6.08s/it]
Training Epoch: 11/100, batch 59/313 completed (loss: 0.0976:  19%|█▉        | 30/156 [02:59<12:45,  6.08s/it]
Training Epoch: 11/100, batch 60/313 completed (loss: 0.0499:  19%|█▉        | 30/156 [03:02<12:45,  6.08s/it]
Training Epoch: 11/100, batch 60/313 completed (loss: 0.053:  19%|█▉        | 30/156 [03:02<12:45,  6.08s/it] 
Training Epoch: 11/100, batch 60/313 completed (loss: 0.3016:  19%|█▉        | 30/156 [03:02<12:45,  6.08s/it]
Training Epoch: 11/100, batch 60/313 completed (loss: 0.0543:  19%|█▉        | 30/156 [03:02<12:45,  6.08s/it]
Training Epoch: 11/100, batch 60/313 completed (loss: 0.3016:  20%|█▉        | 31/156 [03:05<12:30,  6.00s/it]
Training Epoch: 11/100, batch 60/313 completed (loss: 0.0499:  20%|█▉        | 31/156 [03:05<12:30,  6.01s/it]
Training Epoch: 11/100, batch 60/313 completed (loss: 0.0543:  20%|█▉        | 31/156 [03:05<12:30,  6.01s/it]
Training Epoch: 11/100, batch 60/313 completed (loss: 0.053:  20%|█▉        | 31/156 [03:05<12:30,  6.01s/it]
Training Epoch: 11/100, batch 61/313 completed (loss: 0.0848:  20%|█▉        | 31/156 [03:05<12:30,  6.01s/it]
Training Epoch: 11/100, batch 61/313 completed (loss: 0.3242:  20%|█▉        | 31/156 [03:05<12:30,  6.01s/it]
Training Epoch: 11/100, batch 61/313 completed (loss: 0.0809:  20%|█▉        | 31/156 [03:05<12:30,  6.00s/it]
Training Epoch: 11/100, batch 61/313 completed (loss: 0.096:  20%|█▉        | 31/156 [03:05<12:30,  6.01s/it] 
Training Epoch: 11/100, batch 62/313 completed (loss: 0.0802:  20%|█▉        | 31/156 [03:08<12:30,  6.01s/it]
Training Epoch: 11/100, batch 62/313 completed (loss: 0.0932:  20%|█▉        | 31/156 [03:08<12:30,  6.01s/it]
Training Epoch: 11/100, batch 62/313 completed (loss: 0.0138:  20%|█▉        | 31/156 [03:08<12:30,  6.00s/it]
Training Epoch: 11/100, batch 62/313 completed (loss: 0.0804:  20%|█▉        | 31/156 [03:08<12:30,  6.01s/it]
Training Epoch: 11/100, batch 62/313 completed (loss: 0.0802:  21%|██        | 32/156 [03:11<12:41,  6.14s/it]
Training Epoch: 11/100, batch 62/313 completed (loss: 0.0804:  21%|██        | 32/156 [03:11<12:41,  6.14s/it]
Training Epoch: 11/100, batch 62/313 completed (loss: 0.0932:  21%|██        | 32/156 [03:11<12:40,  6.14s/it]
Training Epoch: 11/100, batch 62/313 completed (loss: 0.0138:  21%|██        | 32/156 [03:11<12:41,  6.14s/it]
Training Epoch: 11/100, batch 63/313 completed (loss: 0.0353:  21%|██        | 32/156 [03:11<12:41,  6.14s/it]
Training Epoch: 11/100, batch 63/313 completed (loss: 0.0865:  21%|██        | 32/156 [03:11<12:40,  6.14s/it]
Training Epoch: 11/100, batch 63/313 completed (loss: 0.0512:  21%|██        | 32/156 [03:11<12:41,  6.14s/it]
Training Epoch: 11/100, batch 63/313 completed (loss: 0.169:  21%|██        | 32/156 [03:11<12:41,  6.14s/it] 
Training Epoch: 11/100, batch 64/313 completed (loss: 0.0516:  21%|██        | 32/156 [03:14<12:41,  6.14s/it]
Training Epoch: 11/100, batch 64/313 completed (loss: 0.0928:  21%|██        | 32/156 [03:14<12:41,  6.14s/it]
Training Epoch: 11/100, batch 64/313 completed (loss: 0.1085:  21%|██        | 32/156 [03:14<12:40,  6.14s/it]
Training Epoch: 11/100, batch 64/313 completed (loss: 0.0547:  21%|██        | 32/156 [03:14<12:41,  6.14s/it]
Training Epoch: 11/100, batch 64/313 completed (loss: 0.0516:  21%|██        | 33/156 [03:17<12:22,  6.04s/it]
Training Epoch: 11/100, batch 64/313 completed (loss: 0.0547:  21%|██        | 33/156 [03:17<12:22,  6.04s/it]
Training Epoch: 11/100, batch 64/313 completed (loss: 0.0928:  21%|██        | 33/156 [03:17<12:22,  6.04s/it]
Training Epoch: 11/100, batch 64/313 completed (loss: 0.1085:  21%|██        | 33/156 [03:17<12:22,  6.04s/it]
Training Epoch: 11/100, batch 65/313 completed (loss: 0.2166:  21%|██        | 33/156 [03:17<12:22,  6.04s/it]
Training Epoch: 11/100, batch 65/313 completed (loss: 0.0226:  21%|██        | 33/156 [03:17<12:22,  6.04s/it]
Training Epoch: 11/100, batch 65/313 completed (loss: 0.0989:  21%|██        | 33/156 [03:17<12:22,  6.04s/it]
Training Epoch: 11/100, batch 65/313 completed (loss: 0.2647:  21%|██        | 33/156 [03:17<12:22,  6.04s/it]
Training Epoch: 11/100, batch 66/313 completed (loss: 0.0521:  21%|██        | 33/156 [03:20<12:22,  6.04s/it]
Training Epoch: 11/100, batch 66/313 completed (loss: 0.173:  21%|██        | 33/156 [03:20<12:22,  6.04s/it] 
Training Epoch: 11/100, batch 66/313 completed (loss: 0.0282:  21%|██        | 33/156 [03:20<12:22,  6.04s/it]
Training Epoch: 11/100, batch 66/313 completed (loss: 0.1271:  21%|██        | 33/156 [03:20<12:22,  6.04s/it]
Training Epoch: 11/100, batch 66/313 completed (loss: 0.1271:  22%|██▏       | 34/156 [03:23<12:19,  6.06s/it]
Training Epoch: 11/100, batch 66/313 completed (loss: 0.0282:  22%|██▏       | 34/156 [03:23<12:19,  6.06s/it]
Training Epoch: 11/100, batch 66/313 completed (loss: 0.0521:  22%|██▏       | 34/156 [03:23<12:19,  6.06s/it]
Training Epoch: 11/100, batch 66/313 completed (loss: 0.173:  22%|██▏       | 34/156 [03:23<12:19,  6.06s/it]
Training Epoch: 11/100, batch 67/313 completed (loss: 0.1002:  22%|██▏       | 34/156 [03:23<12:19,  6.06s/it]
Training Epoch: 11/100, batch 67/313 completed (loss: 0.2039:  22%|██▏       | 34/156 [03:23<12:19,  6.06s/it]
Training Epoch: 11/100, batch 67/313 completed (loss: 0.0446:  22%|██▏       | 34/156 [03:23<12:19,  6.06s/it]
Training Epoch: 11/100, batch 67/313 completed (loss: 0.052:  22%|██▏       | 34/156 [03:23<12:19,  6.06s/it] 
Training Epoch: 11/100, batch 68/313 completed (loss: 0.0851:  22%|██▏       | 34/156 [03:27<12:19,  6.06s/it]
Training Epoch: 11/100, batch 68/313 completed (loss: 0.1641:  22%|██▏       | 34/156 [03:27<12:19,  6.06s/it]
Training Epoch: 11/100, batch 68/313 completed (loss: 0.1338:  22%|██▏       | 34/156 [03:27<12:19,  6.06s/it]
Training Epoch: 11/100, batch 68/313 completed (loss: 0.1147:  22%|██▏       | 34/156 [03:27<12:19,  6.06s/it]
Training Epoch: 11/100, batch 68/313 completed (loss: 0.1147:  22%|██▏       | 35/156 [03:29<12:15,  6.08s/it]
Training Epoch: 11/100, batch 68/313 completed (loss: 0.0851:  22%|██▏       | 35/156 [03:29<12:15,  6.08s/it]
Training Epoch: 11/100, batch 68/313 completed (loss: 0.1641:  22%|██▏       | 35/156 [03:29<12:15,  6.08s/it]
Training Epoch: 11/100, batch 68/313 completed (loss: 0.1338:  22%|██▏       | 35/156 [03:29<12:15,  6.08s/it]
Training Epoch: 11/100, batch 69/313 completed (loss: 0.0282:  22%|██▏       | 35/156 [03:29<12:15,  6.08s/it]
Training Epoch: 11/100, batch 69/313 completed (loss: 0.137:  22%|██▏       | 35/156 [03:29<12:15,  6.08s/it] 
Training Epoch: 11/100, batch 69/313 completed (loss: 0.3617:  22%|██▏       | 35/156 [03:29<12:15,  6.08s/it]
Training Epoch: 11/100, batch 69/313 completed (loss: 0.0617:  22%|██▏       | 35/156 [03:29<12:15,  6.08s/it]
Training Epoch: 11/100, batch 70/313 completed (loss: 0.0524:  22%|██▏       | 35/156 [03:32<12:15,  6.08s/it]
Training Epoch: 11/100, batch 70/313 completed (loss: 0.3135:  22%|██▏       | 35/156 [03:32<12:15,  6.08s/it]
Training Epoch: 11/100, batch 70/313 completed (loss: 0.0828:  22%|██▏       | 35/156 [03:32<12:15,  6.08s/it]
Training Epoch: 11/100, batch 70/313 completed (loss: 0.0633:  22%|██▏       | 35/156 [03:32<12:15,  6.08s/it]
Training Epoch: 11/100, batch 70/313 completed (loss: 0.3135:  23%|██▎       | 36/156 [03:35<12:00,  6.00s/it]
Training Epoch: 11/100, batch 70/313 completed (loss: 0.0633:  23%|██▎       | 36/156 [03:35<12:00,  6.01s/it]
Training Epoch: 11/100, batch 70/313 completed (loss: 0.0828:  23%|██▎       | 36/156 [03:35<12:00,  6.00s/it]
Training Epoch: 11/100, batch 70/313 completed (loss: 0.0524:  23%|██▎       | 36/156 [03:35<12:00,  6.01s/it]
Training Epoch: 11/100, batch 71/313 completed (loss: 0.3341:  23%|██▎       | 36/156 [03:35<12:00,  6.00s/it]
Training Epoch: 11/100, batch 71/313 completed (loss: 0.1503:  23%|██▎       | 36/156 [03:35<12:00,  6.00s/it]
Training Epoch: 11/100, batch 71/313 completed (loss: 0.0761:  23%|██▎       | 36/156 [03:35<12:00,  6.01s/it]
Training Epoch: 11/100, batch 71/313 completed (loss: 0.1346:  23%|██▎       | 36/156 [03:35<12:00,  6.01s/it]
Training Epoch: 11/100, batch 72/313 completed (loss: 0.2579:  23%|██▎       | 36/156 [03:39<12:00,  6.01s/it]
Training Epoch: 11/100, batch 72/313 completed (loss: 0.2577:  23%|██▎       | 36/156 [03:39<12:00,  6.00s/it]
Training Epoch: 11/100, batch 72/313 completed (loss: 0.1133:  23%|██▎       | 36/156 [03:39<12:00,  6.00s/it]
Training Epoch: 11/100, batch 72/313 completed (loss: 0.0533:  23%|██▎       | 36/156 [03:39<12:00,  6.01s/it]
Training Epoch: 11/100, batch 72/313 completed (loss: 0.2577:  24%|██▎       | 37/156 [03:42<12:09,  6.13s/it]
Training Epoch: 11/100, batch 72/313 completed (loss: 0.1133:  24%|██▎       | 37/156 [03:42<12:09,  6.13s/it]
Training Epoch: 11/100, batch 72/313 completed (loss: 0.0533:  24%|██▎       | 37/156 [03:42<12:09,  6.13s/it]
Training Epoch: 11/100, batch 72/313 completed (loss: 0.2579:  24%|██▎       | 37/156 [03:42<12:09,  6.13s/it]
Training Epoch: 11/100, batch 73/313 completed (loss: 0.1983:  24%|██▎       | 37/156 [03:42<12:09,  6.13s/it]
Training Epoch: 11/100, batch 73/313 completed (loss: 0.0635:  24%|██▎       | 37/156 [03:42<12:09,  6.13s/it]
Training Epoch: 11/100, batch 73/313 completed (loss: 0.165:  24%|██▎       | 37/156 [03:42<12:09,  6.13s/it] 
Training Epoch: 11/100, batch 73/313 completed (loss: 0.1823:  24%|██▎       | 37/156 [03:42<12:09,  6.13s/it]
Training Epoch: 11/100, batch 74/313 completed (loss: 0.1025:  24%|██▎       | 37/156 [03:45<12:09,  6.13s/it]
Training Epoch: 11/100, batch 74/313 completed (loss: 0.106:  24%|██▎       | 37/156 [03:45<12:09,  6.13s/it]
Training Epoch: 11/100, batch 74/313 completed (loss: 0.1511:  24%|██▎       | 37/156 [03:45<12:09,  6.13s/it]
Training Epoch: 11/100, batch 74/313 completed (loss: 0.3031:  24%|██▎       | 37/156 [03:45<12:09,  6.13s/it]
Training Epoch: 11/100, batch 74/313 completed (loss: 0.1511:  24%|██▍       | 38/156 [03:48<11:54,  6.05s/it]
Training Epoch: 11/100, batch 74/313 completed (loss: 0.106:  24%|██▍       | 38/156 [03:48<11:54,  6.06s/it]
Training Epoch: 11/100, batch 74/313 completed (loss: 0.3031:  24%|██▍       | 38/156 [03:48<11:54,  6.06s/it]
Training Epoch: 11/100, batch 74/313 completed (loss: 0.1025:  24%|██▍       | 38/156 [03:48<11:54,  6.06s/it]
Training Epoch: 11/100, batch 75/313 completed (loss: 0.09:  24%|██▍       | 38/156 [03:48<11:54,  6.06s/it]  
Training Epoch: 11/100, batch 75/313 completed (loss: 0.0112:  24%|██▍       | 38/156 [03:48<11:54,  6.05s/it]
Training Epoch: 11/100, batch 75/313 completed (loss: 0.0475:  24%|██▍       | 38/156 [03:48<11:54,  6.06s/it]
Training Epoch: 11/100, batch 75/313 completed (loss: 0.1727:  24%|██▍       | 38/156 [03:48<11:54,  6.06s/it]
Training Epoch: 11/100, batch 76/313 completed (loss: 0.2248:  24%|██▍       | 38/156 [03:51<11:54,  6.06s/it]
Training Epoch: 11/100, batch 76/313 completed (loss: 0.3439:  24%|██▍       | 38/156 [03:51<11:54,  6.06s/it]
Training Epoch: 11/100, batch 76/313 completed (loss: 0.1246:  24%|██▍       | 38/156 [03:51<11:54,  6.05s/it]
Training Epoch: 11/100, batch 76/313 completed (loss: 0.3299:  24%|██▍       | 38/156 [03:51<11:54,  6.06s/it]
Training Epoch: 11/100, batch 76/313 completed (loss: 0.3299:  25%|██▌       | 39/156 [03:54<11:46,  6.04s/it]
Training Epoch: 11/100, batch 76/313 completed (loss: 0.1246:  25%|██▌       | 39/156 [03:54<11:46,  6.04s/it]
Training Epoch: 11/100, batch 76/313 completed (loss: 0.3439:  25%|██▌       | 39/156 [03:54<11:46,  6.04s/it]
Training Epoch: 11/100, batch 76/313 completed (loss: 0.2248:  25%|██▌       | 39/156 [03:54<11:47,  6.04s/it]
Training Epoch: 11/100, batch 77/313 completed (loss: 0.101:  25%|██▌       | 39/156 [03:54<11:46,  6.04s/it] 
Training Epoch: 11/100, batch 77/313 completed (loss: 0.0728:  25%|██▌       | 39/156 [03:54<11:46,  6.04s/it]
Training Epoch: 11/100, batch 77/313 completed (loss: 0.196:  25%|██▌       | 39/156 [03:54<11:47,  6.04s/it] 
Training Epoch: 11/100, batch 77/313 completed (loss: 0.2433:  25%|██▌       | 39/156 [03:54<11:46,  6.04s/it]
Training Epoch: 11/100, batch 78/313 completed (loss: 0.1966:  25%|██▌       | 39/156 [03:57<11:46,  6.04s/it]
Training Epoch: 11/100, batch 78/313 completed (loss: 0.2039:  25%|██▌       | 39/156 [03:57<11:46,  6.04s/it]
Training Epoch: 11/100, batch 78/313 completed (loss: 0.0897:  25%|██▌       | 39/156 [03:57<11:47,  6.04s/it]
Training Epoch: 11/100, batch 78/313 completed (loss: 0.0542:  25%|██▌       | 39/156 [03:57<11:46,  6.04s/it]
Training Epoch: 11/100, batch 78/313 completed (loss: 0.0542:  26%|██▌       | 40/156 [04:00<11:49,  6.12s/it]
Training Epoch: 11/100, batch 78/313 completed (loss: 0.0897:  26%|██▌       | 40/156 [04:00<11:49,  6.12s/it]
Training Epoch: 11/100, batch 78/313 completed (loss: 0.1966:  26%|██▌       | 40/156 [04:00<11:49,  6.12s/it]
Training Epoch: 11/100, batch 78/313 completed (loss: 0.2039:  26%|██▌       | 40/156 [04:00<11:50,  6.12s/it]
Training Epoch: 11/100, batch 79/313 completed (loss: 0.1608:  26%|██▌       | 40/156 [04:00<11:49,  6.12s/it]
Training Epoch: 11/100, batch 79/313 completed (loss: 0.0234:  26%|██▌       | 40/156 [04:00<11:50,  6.12s/it]
Training Epoch: 11/100, batch 79/313 completed (loss: 0.1056:  26%|██▌       | 40/156 [04:00<11:49,  6.12s/it]
Training Epoch: 11/100, batch 79/313 completed (loss: 0.072:  26%|██▌       | 40/156 [04:00<11:49,  6.12s/it] 
Training Epoch: 11/100, batch 80/313 completed (loss: 0.0669:  26%|██▌       | 40/156 [04:03<11:49,  6.12s/it]
Training Epoch: 11/100, batch 80/313 completed (loss: 0.1252:  26%|██▌       | 40/156 [04:03<11:49,  6.12s/it]
Training Epoch: 11/100, batch 80/313 completed (loss: 0.0187:  26%|██▌       | 40/156 [04:03<11:50,  6.12s/it]
Training Epoch: 11/100, batch 80/313 completed (loss: 0.0504:  26%|██▌       | 40/156 [04:03<11:49,  6.12s/it]
Training Epoch: 11/100, batch 80/313 completed (loss: 0.1252:  26%|██▋       | 41/156 [04:06<11:33,  6.03s/it]
Training Epoch: 11/100, batch 80/313 completed (loss: 0.0504:  26%|██▋       | 41/156 [04:06<11:34,  6.04s/it]
Training Epoch: 11/100, batch 80/313 completed (loss: 0.0187:  26%|██▋       | 41/156 [04:06<11:33,  6.03s/it]
Training Epoch: 11/100, batch 80/313 completed (loss: 0.0669:  26%|██▋       | 41/156 [04:06<11:34,  6.04s/it]
Training Epoch: 11/100, batch 81/313 completed (loss: 0.0154:  26%|██▋       | 41/156 [04:06<11:34,  6.04s/it]
Training Epoch: 11/100, batch 81/313 completed (loss: 0.0468:  26%|██▋       | 41/156 [04:06<11:33,  6.03s/it]
Training Epoch: 11/100, batch 81/313 completed (loss: 0.193:  26%|██▋       | 41/156 [04:06<11:33,  6.03s/it] 
Training Epoch: 11/100, batch 81/313 completed (loss: 0.2346:  26%|██▋       | 41/156 [04:06<11:34,  6.04s/it]
Training Epoch: 11/100, batch 82/313 completed (loss: 0.1795:  26%|██▋       | 41/156 [04:09<11:33,  6.03s/it]
Training Epoch: 11/100, batch 82/313 completed (loss: 0.289:  26%|██▋       | 41/156 [04:09<11:34,  6.04s/it] 
Training Epoch: 11/100, batch 82/313 completed (loss: 0.1535:  26%|██▋       | 41/156 [04:09<11:33,  6.03s/it]
Training Epoch: 11/100, batch 82/313 completed (loss: 0.0741:  26%|██▋       | 41/156 [04:09<11:34,  6.04s/it]
Training Epoch: 11/100, batch 82/313 completed (loss: 0.0741:  27%|██▋       | 42/156 [04:12<11:43,  6.18s/it]
Training Epoch: 11/100, batch 82/313 completed (loss: 0.289:  27%|██▋       | 42/156 [04:12<11:43,  6.18s/it]
Training Epoch: 11/100, batch 82/313 completed (loss: 0.1535:  27%|██▋       | 42/156 [04:12<11:44,  6.18s/it]
Training Epoch: 11/100, batch 82/313 completed (loss: 0.1795:  27%|██▋       | 42/156 [04:12<11:43,  6.17s/it]
Training Epoch: 11/100, batch 83/313 completed (loss: 0.1682:  27%|██▋       | 42/156 [04:12<11:44,  6.18s/it]
Training Epoch: 11/100, batch 83/313 completed (loss: 0.0174:  27%|██▋       | 42/156 [04:12<11:43,  6.18s/it]
Training Epoch: 11/100, batch 83/313 completed (loss: 0.2026:  27%|██▋       | 42/156 [04:12<11:43,  6.17s/it]
Training Epoch: 11/100, batch 83/313 completed (loss: 0.0215:  27%|██▋       | 42/156 [04:12<11:43,  6.18s/it]
Training Epoch: 11/100, batch 84/313 completed (loss: 0.0413:  27%|██▋       | 42/156 [04:15<11:43,  6.18s/it]
Training Epoch: 11/100, batch 84/313 completed (loss: 0.2828:  27%|██▋       | 42/156 [04:15<11:43,  6.17s/it]
Training Epoch: 11/100, batch 84/313 completed (loss: 0.0489:  27%|██▋       | 42/156 [04:15<11:44,  6.18s/it]
Training Epoch: 11/100, batch 84/313 completed (loss: 0.5567:  27%|██▋       | 42/156 [04:15<11:43,  6.18s/it]
Training Epoch: 11/100, batch 84/313 completed (loss: 0.0489:  28%|██▊       | 43/156 [04:18<11:27,  6.09s/it]
Training Epoch: 11/100, batch 84/313 completed (loss: 0.0413:  28%|██▊       | 43/156 [04:18<11:27,  6.09s/it]
Training Epoch: 11/100, batch 84/313 completed (loss: 0.5567:  28%|██▊       | 43/156 [04:18<11:28,  6.09s/it]
Training Epoch: 11/100, batch 84/313 completed (loss: 0.2828:  28%|██▊       | 43/156 [04:18<11:28,  6.09s/it]
Training Epoch: 11/100, batch 85/313 completed (loss: 0.0229:  28%|██▊       | 43/156 [04:18<11:28,  6.09s/it]
Training Epoch: 11/100, batch 85/313 completed (loss: 0.5401:  28%|██▊       | 43/156 [04:18<11:27,  6.09s/it]
Training Epoch: 11/100, batch 85/313 completed (loss: 0.0106:  28%|██▊       | 43/156 [04:18<11:27,  6.09s/it]
Training Epoch: 11/100, batch 85/313 completed (loss: 0.0398:  28%|██▊       | 43/156 [04:18<11:28,  6.09s/it]
Training Epoch: 11/100, batch 86/313 completed (loss: 0.2834:  28%|██▊       | 43/156 [04:21<11:27,  6.09s/it]
Training Epoch: 11/100, batch 86/313 completed (loss: 0.0493:  28%|██▊       | 43/156 [04:21<11:28,  6.09s/it]
Training Epoch: 11/100, batch 86/313 completed (loss: 0.0271:  28%|██▊       | 43/156 [04:21<11:27,  6.09s/it]
Training Epoch: 11/100, batch 86/313 completed (loss: 0.0978:  28%|██▊       | 43/156 [04:21<11:28,  6.09s/it]
Training Epoch: 11/100, batch 86/313 completed (loss: 0.0978:  28%|██▊       | 44/156 [04:24<11:18,  6.06s/it]
Training Epoch: 11/100, batch 86/313 completed (loss: 0.2834:  28%|██▊       | 44/156 [04:24<11:18,  6.06s/it]
Training Epoch: 11/100, batch 86/313 completed (loss: 0.0271:  28%|██▊       | 44/156 [04:24<11:18,  6.06s/it]
Training Epoch: 11/100, batch 86/313 completed (loss: 0.0493:  28%|██▊       | 44/156 [04:24<11:18,  6.06s/it]
Training Epoch: 11/100, batch 87/313 completed (loss: 0.0208:  28%|██▊       | 44/156 [04:24<11:18,  6.06s/it]
Training Epoch: 11/100, batch 87/313 completed (loss: 0.071:  28%|██▊       | 44/156 [04:24<11:18,  6.06s/it] 
Training Epoch: 11/100, batch 87/313 completed (loss: 0.0357:  28%|██▊       | 44/156 [04:24<11:18,  6.06s/it]
Training Epoch: 11/100, batch 87/313 completed (loss: 0.0351:  28%|██▊       | 44/156 [04:24<11:18,  6.06s/it]
Training Epoch: 11/100, batch 88/313 completed (loss: 0.0217:  28%|██▊       | 44/156 [04:28<11:18,  6.06s/it]
Training Epoch: 11/100, batch 88/313 completed (loss: 0.0626:  28%|██▊       | 44/156 [04:28<11:18,  6.06s/it]
Training Epoch: 11/100, batch 88/313 completed (loss: 0.0489:  28%|██▊       | 44/156 [04:28<11:18,  6.06s/it]
Training Epoch: 11/100, batch 88/313 completed (loss: 0.051:  28%|██▊       | 44/156 [04:28<11:18,  6.06s/it] 
Training Epoch: 11/100, batch 88/313 completed (loss: 0.051:  29%|██▉       | 45/156 [04:30<11:20,  6.13s/it]
Training Epoch: 11/100, batch 88/313 completed (loss: 0.0626:  29%|██▉       | 45/156 [04:30<11:21,  6.14s/it]
Training Epoch: 11/100, batch 88/313 completed (loss: 0.0217:  29%|██▉       | 45/156 [04:30<11:21,  6.14s/it]
Training Epoch: 11/100, batch 88/313 completed (loss: 0.0489:  29%|██▉       | 45/156 [04:30<11:21,  6.14s/it]
Training Epoch: 11/100, batch 89/313 completed (loss: 0.1512:  29%|██▉       | 45/156 [04:30<11:21,  6.14s/it]
Training Epoch: 11/100, batch 89/313 completed (loss: 0.0568:  29%|██▉       | 45/156 [04:30<11:21,  6.14s/it]
Training Epoch: 11/100, batch 89/313 completed (loss: 0.0906:  29%|██▉       | 45/156 [04:30<11:21,  6.14s/it]
Training Epoch: 11/100, batch 89/313 completed (loss: 0.3191:  29%|██▉       | 45/156 [04:30<11:20,  6.13s/it]
Training Epoch: 11/100, batch 90/313 completed (loss: 0.0338:  29%|██▉       | 45/156 [04:33<11:21,  6.14s/it]
Training Epoch: 11/100, batch 90/313 completed (loss: 0.0683:  29%|██▉       | 45/156 [04:33<11:21,  6.14s/it]
Training Epoch: 11/100, batch 90/313 completed (loss: 0.0366:  29%|██▉       | 45/156 [04:33<11:21,  6.14s/it]
Training Epoch: 11/100, batch 90/313 completed (loss: 0.1048:  29%|██▉       | 45/156 [04:33<11:20,  6.13s/it]
Training Epoch: 11/100, batch 90/313 completed (loss: 0.1048:  29%|██▉       | 46/156 [04:36<11:05,  6.05s/it]
Training Epoch: 11/100, batch 90/313 completed (loss: 0.0338:  29%|██▉       | 46/156 [04:36<11:05,  6.05s/it]
Training Epoch: 11/100, batch 90/313 completed (loss: 0.0683:  29%|██▉       | 46/156 [04:36<11:05,  6.05s/it]
Training Epoch: 11/100, batch 90/313 completed (loss: 0.0366:  29%|██▉       | 46/156 [04:36<11:05,  6.05s/it]
Training Epoch: 11/100, batch 91/313 completed (loss: 0.2757:  29%|██▉       | 46/156 [04:36<11:05,  6.05s/it]
Training Epoch: 11/100, batch 91/313 completed (loss: 0.1482:  29%|██▉       | 46/156 [04:36<11:05,  6.05s/it]
Training Epoch: 11/100, batch 91/313 completed (loss: 0.1873:  29%|██▉       | 46/156 [04:36<11:05,  6.05s/it]
Training Epoch: 11/100, batch 91/313 completed (loss: 0.1283:  29%|██▉       | 46/156 [04:36<11:05,  6.05s/it]
Training Epoch: 11/100, batch 92/313 completed (loss: 0.0194:  29%|██▉       | 46/156 [04:39<11:05,  6.05s/it]
Training Epoch: 11/100, batch 92/313 completed (loss: 0.0426:  29%|██▉       | 46/156 [04:39<11:05,  6.05s/it]
Training Epoch: 11/100, batch 92/313 completed (loss: 0.1686:  29%|██▉       | 46/156 [04:39<11:05,  6.05s/it]
Training Epoch: 11/100, batch 92/313 completed (loss: 0.3062:  29%|██▉       | 46/156 [04:39<11:05,  6.05s/it]
Training Epoch: 11/100, batch 92/313 completed (loss: 0.3062:  30%|███       | 47/156 [04:43<11:06,  6.12s/it]
Training Epoch: 11/100, batch 92/313 completed (loss: 0.1686:  30%|███       | 47/156 [04:43<11:06,  6.11s/it]
Training Epoch: 11/100, batch 92/313 completed (loss: 0.0426:  30%|███       | 47/156 [04:43<11:06,  6.12s/it]
Training Epoch: 11/100, batch 92/313 completed (loss: 0.0194:  30%|███       | 47/156 [04:43<11:06,  6.12s/it]
Training Epoch: 11/100, batch 93/313 completed (loss: 0.1681:  30%|███       | 47/156 [04:43<11:06,  6.12s/it]
Training Epoch: 11/100, batch 93/313 completed (loss: 0.053:  30%|███       | 47/156 [04:43<11:06,  6.11s/it] 
Training Epoch: 11/100, batch 93/313 completed (loss: 0.2645:  30%|███       | 47/156 [04:43<11:06,  6.12s/it]
Training Epoch: 11/100, batch 93/313 completed (loss: 0.0448:  30%|███       | 47/156 [04:43<11:06,  6.12s/it]
Training Epoch: 11/100, batch 94/313 completed (loss: 0.0416:  30%|███       | 47/156 [04:46<11:06,  6.11s/it]
Training Epoch: 11/100, batch 94/313 completed (loss: 0.162:  30%|███       | 47/156 [04:46<11:06,  6.12s/it] 
Training Epoch: 11/100, batch 94/313 completed (loss: 0.1065:  30%|███       | 47/156 [04:46<11:06,  6.12s/it]
Training Epoch: 11/100, batch 94/313 completed (loss: 0.1347:  30%|███       | 47/156 [04:46<11:06,  6.12s/it]
Training Epoch: 11/100, batch 94/313 completed (loss: 0.1347:  31%|███       | 48/156 [04:49<10:57,  6.08s/it]
Training Epoch: 11/100, batch 94/313 completed (loss: 0.162:  31%|███       | 48/156 [04:49<10:57,  6.08s/it]
Training Epoch: 11/100, batch 94/313 completed (loss: 0.1065:  31%|███       | 48/156 [04:49<10:56,  6.08s/it]
Training Epoch: 11/100, batch 94/313 completed (loss: 0.0416:  31%|███       | 48/156 [04:49<10:57,  6.09s/it]
Training Epoch: 11/100, batch 95/313 completed (loss: 0.3561:  31%|███       | 48/156 [04:49<10:57,  6.09s/it]
Training Epoch: 11/100, batch 95/313 completed (loss: 0.039:  31%|███       | 48/156 [04:49<10:57,  6.08s/it]
Training Epoch: 11/100, batch 95/313 completed (loss: 0.0432:  31%|███       | 48/156 [04:49<10:56,  6.08s/it]
Training Epoch: 11/100, batch 95/313 completed (loss: 0.4494:  31%|███       | 48/156 [04:49<10:57,  6.08s/it]
Training Epoch: 11/100, batch 96/313 completed (loss: 0.0295:  31%|███       | 48/156 [04:51<10:57,  6.09s/it]
Training Epoch: 11/100, batch 96/313 completed (loss: 0.3676:  31%|███       | 48/156 [04:51<10:57,  6.08s/it]
Training Epoch: 11/100, batch 96/313 completed (loss: 0.08:  31%|███       | 48/156 [04:51<10:56,  6.08s/it]  
Training Epoch: 11/100, batch 96/313 completed (loss: 0.2888:  31%|███       | 48/156 [04:52<10:57,  6.08s/it]
Training Epoch: 11/100, batch 96/313 completed (loss: 0.2888:  31%|███▏      | 49/156 [04:54<10:43,  6.01s/it]
Training Epoch: 11/100, batch 96/313 completed (loss: 0.0295:  31%|███▏      | 49/156 [04:54<10:43,  6.01s/it]
Training Epoch: 11/100, batch 96/313 completed (loss: 0.3676:  31%|███▏      | 49/156 [04:54<10:43,  6.01s/it]
Training Epoch: 11/100, batch 96/313 completed (loss: 0.08:  31%|███▏      | 49/156 [04:54<10:43,  6.01s/it]
Training Epoch: 11/100, batch 97/313 completed (loss: 0.1203:  31%|███▏      | 49/156 [04:54<10:43,  6.01s/it]
Training Epoch: 11/100, batch 97/313 completed (loss: 0.1146:  31%|███▏      | 49/156 [04:54<10:43,  6.01s/it]
Training Epoch: 11/100, batch 97/313 completed (loss: 0.0233:  31%|███▏      | 49/156 [04:54<10:43,  6.01s/it]
Training Epoch: 11/100, batch 97/313 completed (loss: 0.0152:  31%|███▏      | 49/156 [04:54<10:43,  6.01s/it]
Training Epoch: 11/100, batch 98/313 completed (loss: 0.0486:  31%|███▏      | 49/156 [04:58<10:43,  6.01s/it]
Training Epoch: 11/100, batch 98/313 completed (loss: 0.1376:  31%|███▏      | 49/156 [04:58<10:43,  6.01s/it]
Training Epoch: 11/100, batch 98/313 completed (loss: 0.09:  31%|███▏      | 49/156 [04:58<10:43,  6.01s/it]  
Training Epoch: 11/100, batch 98/313 completed (loss: 0.2474:  31%|███▏      | 49/156 [04:58<10:43,  6.01s/it]
Training Epoch: 11/100, batch 98/313 completed (loss: 0.09:  32%|███▏      | 50/156 [05:01<10:47,  6.11s/it]
Training Epoch: 11/100, batch 98/313 completed (loss: 0.2474:  32%|███▏      | 50/156 [05:01<10:47,  6.11s/it]
Training Epoch: 11/100, batch 98/313 completed (loss: 0.0486:  32%|███▏      | 50/156 [05:01<10:47,  6.11s/it]
Training Epoch: 11/100, batch 98/313 completed (loss: 0.1376:  32%|███▏      | 50/156 [05:01<10:47,  6.11s/it]
Training Epoch: 11/100, batch 99/313 completed (loss: 0.0339:  32%|███▏      | 50/156 [05:01<10:47,  6.11s/it]
Training Epoch: 11/100, batch 99/313 completed (loss: 0.0505:  32%|███▏      | 50/156 [05:01<10:47,  6.11s/it]
Training Epoch: 11/100, batch 99/313 completed (loss: 0.0773:  32%|███▏      | 50/156 [05:01<10:47,  6.11s/it]
Training Epoch: 11/100, batch 99/313 completed (loss: 0.144:  32%|███▏      | 50/156 [05:01<10:47,  6.11s/it] 
Training Epoch: 11/100, batch 100/313 completed (loss: 0.1405:  32%|███▏      | 50/156 [05:04<10:47,  6.11s/it]
Training Epoch: 11/100, batch 100/313 completed (loss: 0.0162:  32%|███▏      | 50/156 [05:04<10:47,  6.11s/it]
Training Epoch: 11/100, batch 100/313 completed (loss: 0.0668:  32%|███▏      | 50/156 [05:04<10:47,  6.11s/it]
Training Epoch: 11/100, batch 100/313 completed (loss: 0.0699:  32%|███▏      | 50/156 [05:04<10:47,  6.11s/it]
Training Epoch: 11/100, batch 100/313 completed (loss: 0.0668:  33%|███▎      | 51/156 [05:07<10:32,  6.02s/it]
Training Epoch: 11/100, batch 100/313 completed (loss: 0.0699:  33%|███▎      | 51/156 [05:07<10:32,  6.03s/it]
Training Epoch: 11/100, batch 100/313 completed (loss: 0.0162:  33%|███▎      | 51/156 [05:07<10:32,  6.02s/it]
Training Epoch: 11/100, batch 100/313 completed (loss: 0.1405:  33%|███▎      | 51/156 [05:07<10:32,  6.03s/it]
Training Epoch: 11/100, batch 101/313 completed (loss: 0.1308:  33%|███▎      | 51/156 [05:07<10:32,  6.03s/it]
Training Epoch: 11/100, batch 101/313 completed (loss: 0.1296:  33%|███▎      | 51/156 [05:07<10:32,  6.02s/it]
Training Epoch: 11/100, batch 101/313 completed (loss: 0.0494:  33%|███▎      | 51/156 [05:07<10:32,  6.02s/it]
Training Epoch: 11/100, batch 101/313 completed (loss: 0.0767:  33%|███▎      | 51/156 [05:07<10:32,  6.03s/it]
Training Epoch: 11/100, batch 102/313 completed (loss: 0.0331:  33%|███▎      | 51/156 [05:09<10:32,  6.02s/it]
Training Epoch: 11/100, batch 102/313 completed (loss: 0.0684:  33%|███▎      | 51/156 [05:09<10:32,  6.02s/it]
Training Epoch: 11/100, batch 102/313 completed (loss: 0.4548:  33%|███▎      | 51/156 [05:09<10:32,  6.03s/it]
Training Epoch: 11/100, batch 102/313 completed (loss: 0.0039:  33%|███▎      | 51/156 [05:10<10:32,  6.03s/it]
Training Epoch: 11/100, batch 102/313 completed (loss: 0.0039:  33%|███▎      | 52/156 [05:13<10:25,  6.02s/it]
Training Epoch: 11/100, batch 102/313 completed (loss: 0.0331:  33%|███▎      | 52/156 [05:13<10:25,  6.02s/it]
Training Epoch: 11/100, batch 102/313 completed (loss: 0.0684:  33%|███▎      | 52/156 [05:13<10:26,  6.02s/it]
Training Epoch: 11/100, batch 102/313 completed (loss: 0.4548:  33%|███▎      | 52/156 [05:13<10:26,  6.02s/it]
Training Epoch: 11/100, batch 103/313 completed (loss: 0.0285:  33%|███▎      | 52/156 [05:13<10:26,  6.02s/it]
Training Epoch: 11/100, batch 103/313 completed (loss: 0.0548:  33%|███▎      | 52/156 [05:13<10:26,  6.02s/it]
Training Epoch: 11/100, batch 103/313 completed (loss: 0.0694:  33%|███▎      | 52/156 [05:13<10:25,  6.02s/it]
Training Epoch: 11/100, batch 103/313 completed (loss: 0.0347:  33%|███▎      | 52/156 [05:13<10:25,  6.02s/it]
Training Epoch: 11/100, batch 104/313 completed (loss: 0.0731:  33%|███▎      | 52/156 [05:16<10:25,  6.02s/it]
Training Epoch: 11/100, batch 104/313 completed (loss: 0.4627:  33%|███▎      | 52/156 [05:16<10:26,  6.02s/it]
Training Epoch: 11/100, batch 104/313 completed (loss: 0.0671:  33%|███▎      | 52/156 [05:16<10:26,  6.02s/it]
Training Epoch: 11/100, batch 104/313 completed (loss: 0.0405:  33%|███▎      | 52/156 [05:16<10:25,  6.02s/it]
Training Epoch: 11/100, batch 104/313 completed (loss: 0.0405:  34%|███▍      | 53/156 [05:19<10:28,  6.11s/it]
Training Epoch: 11/100, batch 104/313 completed (loss: 0.4627:  34%|███▍      | 53/156 [05:19<10:28,  6.10s/it]
Training Epoch: 11/100, batch 104/313 completed (loss: 0.0671:  34%|███▍      | 53/156 [05:19<10:28,  6.11s/it]
Training Epoch: 11/100, batch 104/313 completed (loss: 0.0731:  34%|███▍      | 53/156 [05:19<10:28,  6.11s/it]
Training Epoch: 11/100, batch 105/313 completed (loss: 0.0598:  34%|███▍      | 53/156 [05:19<10:28,  6.11s/it]
Training Epoch: 11/100, batch 105/313 completed (loss: 0.0166:  34%|███▍      | 53/156 [05:19<10:28,  6.11s/it]
Training Epoch: 11/100, batch 105/313 completed (loss: 0.0696:  34%|███▍      | 53/156 [05:19<10:28,  6.10s/it]
Training Epoch: 11/100, batch 105/313 completed (loss: 0.0622:  34%|███▍      | 53/156 [05:19<10:28,  6.11s/it]
Training Epoch: 11/100, batch 106/313 completed (loss: 0.2413:  34%|███▍      | 53/156 [05:22<10:28,  6.11s/it]
Training Epoch: 11/100, batch 106/313 completed (loss: 0.0315:  34%|███▍      | 53/156 [05:22<10:28,  6.11s/it]
Training Epoch: 11/100, batch 106/313 completed (loss: 0.0069:  34%|███▍      | 53/156 [05:22<10:28,  6.10s/it]
Training Epoch: 11/100, batch 106/313 completed (loss: 0.0475:  34%|███▍      | 53/156 [05:22<10:28,  6.11s/it]
Training Epoch: 11/100, batch 106/313 completed (loss: 0.2413:  35%|███▍      | 54/156 [05:25<10:14,  6.03s/it]
Training Epoch: 11/100, batch 106/313 completed (loss: 0.0315:  35%|███▍      | 54/156 [05:25<10:15,  6.03s/it]
Training Epoch: 11/100, batch 106/313 completed (loss: 0.0475:  35%|███▍      | 54/156 [05:25<10:15,  6.03s/it]
Training Epoch: 11/100, batch 106/313 completed (loss: 0.0069:  35%|███▍      | 54/156 [05:25<10:15,  6.03s/it]
Training Epoch: 11/100, batch 107/313 completed (loss: 0.0987:  35%|███▍      | 54/156 [05:25<10:15,  6.03s/it]
Training Epoch: 11/100, batch 107/313 completed (loss: 0.1182:  35%|███▍      | 54/156 [05:25<10:15,  6.03s/it]
Training Epoch: 11/100, batch 107/313 completed (loss: 0.2055:  35%|███▍      | 54/156 [05:25<10:14,  6.03s/it]
Training Epoch: 11/100, batch 107/313 completed (loss: 0.4256:  35%|███▍      | 54/156 [05:25<10:15,  6.03s/it]
Training Epoch: 11/100, batch 108/313 completed (loss: 0.3745:  35%|███▍      | 54/156 [05:28<10:14,  6.03s/it]
Training Epoch: 11/100, batch 108/313 completed (loss: 0.0741:  35%|███▍      | 54/156 [05:28<10:15,  6.03s/it]
Training Epoch: 11/100, batch 108/313 completed (loss: 0.0836:  35%|███▍      | 54/156 [05:28<10:15,  6.03s/it]
Training Epoch: 11/100, batch 108/313 completed (loss: 0.0899:  35%|███▍      | 54/156 [05:28<10:15,  6.03s/it]
Training Epoch: 11/100, batch 108/313 completed (loss: 0.0899:  35%|███▌      | 55/156 [05:31<10:17,  6.12s/it]
Training Epoch: 11/100, batch 108/313 completed (loss: 0.0836:  35%|███▌      | 55/156 [05:31<10:17,  6.12s/it]
Training Epoch: 11/100, batch 108/313 completed (loss: 0.3745:  35%|███▌      | 55/156 [05:31<10:18,  6.12s/it]
Training Epoch: 11/100, batch 108/313 completed (loss: 0.0741:  35%|███▌      | 55/156 [05:31<10:18,  6.12s/it]
Training Epoch: 11/100, batch 109/313 completed (loss: 0.14:  35%|███▌      | 55/156 [05:31<10:17,  6.12s/it]  
Training Epoch: 11/100, batch 109/313 completed (loss: 0.0689:  35%|███▌      | 55/156 [05:31<10:18,  6.12s/it]
Training Epoch: 11/100, batch 109/313 completed (loss: 0.1782:  35%|███▌      | 55/156 [05:31<10:18,  6.12s/it]
Training Epoch: 11/100, batch 109/313 completed (loss: 0.0456:  35%|███▌      | 55/156 [05:31<10:17,  6.12s/it]
Training Epoch: 11/100, batch 110/313 completed (loss: 0.0029:  35%|███▌      | 55/156 [05:34<10:17,  6.12s/it]
Training Epoch: 11/100, batch 110/313 completed (loss: 0.1923:  35%|███▌      | 55/156 [05:34<10:18,  6.12s/it]
Training Epoch: 11/100, batch 110/313 completed (loss: 0.2503:  35%|███▌      | 55/156 [05:34<10:18,  6.12s/it]
Training Epoch: 11/100, batch 110/313 completed (loss: 0.0349:  35%|███▌      | 55/156 [05:34<10:17,  6.12s/it]
Training Epoch: 11/100, batch 110/313 completed (loss: 0.1923:  36%|███▌      | 56/156 [05:37<10:08,  6.08s/it]
Training Epoch: 11/100, batch 110/313 completed (loss: 0.0029:  36%|███▌      | 56/156 [05:37<10:08,  6.09s/it]
Training Epoch: 11/100, batch 110/313 completed (loss: 0.0349:  36%|███▌      | 56/156 [05:37<10:08,  6.09s/it]
Training Epoch: 11/100, batch 110/313 completed (loss: 0.2503:  36%|███▌      | 56/156 [05:37<10:08,  6.09s/it]
Training Epoch: 11/100, batch 111/313 completed (loss: 0.1227:  36%|███▌      | 56/156 [05:37<10:08,  6.09s/it]
Training Epoch: 11/100, batch 111/313 completed (loss: 0.0343:  36%|███▌      | 56/156 [05:37<10:08,  6.08s/it]
Training Epoch: 11/100, batch 111/313 completed (loss: 0.0083:  36%|███▌      | 56/156 [05:37<10:08,  6.09s/it]
Training Epoch: 11/100, batch 111/313 completed (loss: 0.3298:  36%|███▌      | 56/156 [05:37<10:08,  6.09s/it]
Training Epoch: 11/100, batch 112/313 completed (loss: 0.0994:  36%|███▌      | 56/156 [05:40<10:08,  6.08s/it]
Training Epoch: 11/100, batch 112/313 completed (loss: 0.0552:  36%|███▌      | 56/156 [05:40<10:08,  6.09s/it]
Training Epoch: 11/100, batch 112/313 completed (loss: 0.0727:  36%|███▌      | 56/156 [05:40<10:08,  6.09s/it]
Training Epoch: 11/100, batch 112/313 completed (loss: 0.3395:  36%|███▌      | 56/156 [05:40<10:08,  6.09s/it]
Training Epoch: 11/100, batch 112/313 completed (loss: 0.0727:  37%|███▋      | 57/156 [05:43<09:59,  6.05s/it]
Training Epoch: 11/100, batch 112/313 completed (loss: 0.3395:  37%|███▋      | 57/156 [05:43<09:59,  6.05s/it]
Training Epoch: 11/100, batch 112/313 completed (loss: 0.0552:  37%|███▋      | 57/156 [05:43<09:59,  6.05s/it]
Training Epoch: 11/100, batch 112/313 completed (loss: 0.0994:  37%|███▋      | 57/156 [05:43<09:59,  6.05s/it]
Training Epoch: 11/100, batch 113/313 completed (loss: 0.016:  37%|███▋      | 57/156 [05:43<09:59,  6.05s/it] 
Training Epoch: 11/100, batch 113/313 completed (loss: 0.0553:  37%|███▋      | 57/156 [05:43<09:59,  6.05s/it]
Training Epoch: 11/100, batch 113/313 completed (loss: 0.0714:  37%|███▋      | 57/156 [05:43<09:59,  6.05s/it]
Training Epoch: 11/100, batch 113/313 completed (loss: 0.1077:  37%|███▋      | 57/156 [05:43<09:59,  6.05s/it]
Training Epoch: 11/100, batch 114/313 completed (loss: 0.14:  37%|███▋      | 57/156 [05:46<09:59,  6.05s/it]  
Training Epoch: 11/100, batch 114/313 completed (loss: 0.1592:  37%|███▋      | 57/156 [05:46<09:59,  6.05s/it]
Training Epoch: 11/100, batch 114/313 completed (loss: 0.0806:  37%|███▋      | 57/156 [05:46<09:59,  6.05s/it]
Training Epoch: 11/100, batch 114/313 completed (loss: 0.0739:  37%|███▋      | 57/156 [05:46<09:59,  6.05s/it]
Training Epoch: 11/100, batch 114/313 completed (loss: 0.0806:  37%|███▋      | 58/156 [05:49<09:54,  6.07s/it]
Training Epoch: 11/100, batch 114/313 completed (loss: 0.0739:  37%|███▋      | 58/156 [05:49<09:55,  6.07s/it]
Training Epoch: 11/100, batch 114/313 completed (loss: 0.1592:  37%|███▋      | 58/156 [05:49<09:55,  6.07s/it]
Training Epoch: 11/100, batch 114/313 completed (loss: 0.14:  37%|███▋      | 58/156 [05:49<09:55,  6.07s/it]
Training Epoch: 11/100, batch 115/313 completed (loss: 0.1274:  37%|███▋      | 58/156 [05:49<09:55,  6.07s/it]
Training Epoch: 11/100, batch 115/313 completed (loss: 0.1654:  37%|███▋      | 58/156 [05:49<09:54,  6.07s/it]
Training Epoch: 11/100, batch 115/313 completed (loss: 0.1526:  37%|███▋      | 58/156 [05:49<09:55,  6.07s/it]
Training Epoch: 11/100, batch 115/313 completed (loss: 0.3843:  37%|███▋      | 58/156 [05:49<09:55,  6.07s/it]
Training Epoch: 11/100, batch 116/313 completed (loss: 0.0714:  37%|███▋      | 58/156 [05:52<09:54,  6.07s/it]
Training Epoch: 11/100, batch 116/313 completed (loss: 0.041:  37%|███▋      | 58/156 [05:52<09:55,  6.07s/it] 
Training Epoch: 11/100, batch 116/313 completed (loss: 0.1538:  37%|███▋      | 58/156 [05:52<09:55,  6.07s/it]
Training Epoch: 11/100, batch 116/313 completed (loss: 0.037:  37%|███▋      | 58/156 [05:52<09:55,  6.07s/it] 
Training Epoch: 11/100, batch 116/313 completed (loss: 0.037:  38%|███▊      | 59/156 [05:55<09:47,  6.06s/it]
Training Epoch: 11/100, batch 116/313 completed (loss: 0.0714:  38%|███▊      | 59/156 [05:55<09:48,  6.06s/it]
Training Epoch: 11/100, batch 116/313 completed (loss: 0.041:  38%|███▊      | 59/156 [05:55<09:48,  6.06s/it]
Training Epoch: 11/100, batch 116/313 completed (loss: 0.1538:  38%|███▊      | 59/156 [05:55<09:48,  6.06s/it]
Training Epoch: 11/100, batch 117/313 completed (loss: 0.2745:  38%|███▊      | 59/156 [05:55<09:48,  6.06s/it]
Training Epoch: 11/100, batch 117/313 completed (loss: 0.1978:  38%|███▊      | 59/156 [05:55<09:48,  6.06s/it]
Training Epoch: 11/100, batch 117/313 completed (loss: 0.043:  38%|███▊      | 59/156 [05:55<09:48,  6.06s/it] 
Training Epoch: 11/100, batch 117/313 completed (loss: 0.2766:  38%|███▊      | 59/156 [05:55<09:47,  6.06s/it]
Training Epoch: 11/100, batch 118/313 completed (loss: 0.035:  38%|███▊      | 59/156 [05:58<09:48,  6.06s/it] 
Training Epoch: 11/100, batch 118/313 completed (loss: 0.1971:  38%|███▊      | 59/156 [05:58<09:48,  6.06s/it]
Training Epoch: 11/100, batch 118/313 completed (loss: 0.0434:  38%|███▊      | 59/156 [05:58<09:48,  6.06s/it]
Training Epoch: 11/100, batch 118/313 completed (loss: 0.1953:  38%|███▊      | 59/156 [05:58<09:47,  6.06s/it]
Training Epoch: 11/100, batch 118/313 completed (loss: 0.1971:  38%|███▊      | 60/156 [06:01<09:44,  6.08s/it]
Training Epoch: 11/100, batch 118/313 completed (loss: 0.1953:  38%|███▊      | 60/156 [06:01<09:44,  6.08s/it]
Training Epoch: 11/100, batch 118/313 completed (loss: 0.0434:  38%|███▊      | 60/156 [06:01<09:44,  6.08s/it]
Training Epoch: 11/100, batch 118/313 completed (loss: 0.035:  38%|███▊      | 60/156 [06:01<09:43,  6.08s/it]
Training Epoch: 11/100, batch 119/313 completed (loss: 0.1695:  38%|███▊      | 60/156 [06:01<09:44,  6.08s/it]
Training Epoch: 11/100, batch 119/313 completed (loss: 0.2866:  38%|███▊      | 60/156 [06:01<09:44,  6.08s/it]
Training Epoch: 11/100, batch 119/313 completed (loss: 0.1859:  38%|███▊      | 60/156 [06:01<09:43,  6.08s/it]
Training Epoch: 11/100, batch 119/313 completed (loss: 0.2781:  38%|███▊      | 60/156 [06:01<09:44,  6.08s/it]
Training Epoch: 11/100, batch 120/313 completed (loss: 0.0273:  38%|███▊      | 60/156 [06:04<09:44,  6.08s/it]
Training Epoch: 11/100, batch 120/313 completed (loss: 0.4051:  38%|███▊      | 60/156 [06:04<09:44,  6.08s/it]
Training Epoch: 11/100, batch 120/313 completed (loss: 0.2032:  38%|███▊      | 60/156 [06:04<09:43,  6.08s/it]
Training Epoch: 11/100, batch 120/313 completed (loss: 0.0525:  38%|███▊      | 60/156 [06:04<09:44,  6.08s/it]
Training Epoch: 11/100, batch 120/313 completed (loss: 0.4051:  39%|███▉      | 61/156 [06:07<09:35,  6.05s/it]
Training Epoch: 11/100, batch 120/313 completed (loss: 0.0525:  39%|███▉      | 61/156 [06:07<09:35,  6.05s/it]
Training Epoch: 11/100, batch 120/313 completed (loss: 0.2032:  39%|███▉      | 61/156 [06:07<09:35,  6.05s/it]
Training Epoch: 11/100, batch 120/313 completed (loss: 0.0273:  39%|███▉      | 61/156 [06:07<09:35,  6.06s/it]
Training Epoch: 11/100, batch 121/313 completed (loss: 0.0221:  39%|███▉      | 61/156 [06:07<09:35,  6.06s/it]
Training Epoch: 11/100, batch 121/313 completed (loss: 0.21:  39%|███▉      | 61/156 [06:07<09:35,  6.05s/it]  
Training Epoch: 11/100, batch 121/313 completed (loss: 0.038:  39%|███▉      | 61/156 [06:07<09:35,  6.05s/it] 
Training Epoch: 11/100, batch 121/313 completed (loss: 0.1117:  39%|███▉      | 61/156 [06:07<09:35,  6.05s/it]
Training Epoch: 11/100, batch 122/313 completed (loss: 0.3896:  39%|███▉      | 61/156 [06:10<09:35,  6.05s/it]
Training Epoch: 11/100, batch 122/313 completed (loss: 0.1539:  39%|███▉      | 61/156 [06:10<09:35,  6.06s/it]
Training Epoch: 11/100, batch 122/313 completed (loss: 0.3189:  39%|███▉      | 61/156 [06:10<09:35,  6.05s/it]
Training Epoch: 11/100, batch 122/313 completed (loss: 0.8286:  39%|███▉      | 61/156 [06:10<09:35,  6.05s/it]
Training Epoch: 11/100, batch 122/313 completed (loss: 0.8286:  40%|███▉      | 62/156 [06:13<09:27,  6.03s/it]
Training Epoch: 11/100, batch 122/313 completed (loss: 0.3896:  40%|███▉      | 62/156 [06:13<09:27,  6.03s/it]
Training Epoch: 11/100, batch 122/313 completed (loss: 0.1539:  40%|███▉      | 62/156 [06:13<09:27,  6.03s/it]
Training Epoch: 11/100, batch 122/313 completed (loss: 0.3189:  40%|███▉      | 62/156 [06:13<09:27,  6.03s/it]
Training Epoch: 11/100, batch 123/313 completed (loss: 0.0906:  40%|███▉      | 62/156 [06:13<09:27,  6.03s/it]
Training Epoch: 11/100, batch 123/313 completed (loss: 0.0698:  40%|███▉      | 62/156 [06:13<09:27,  6.03s/it]
Training Epoch: 11/100, batch 123/313 completed (loss: 0.1062:  40%|███▉      | 62/156 [06:13<09:27,  6.03s/it]
Training Epoch: 11/100, batch 123/313 completed (loss: 0.042:  40%|███▉      | 62/156 [06:13<09:27,  6.03s/it] 
Training Epoch: 11/100, batch 124/313 completed (loss: 0.5677:  40%|███▉      | 62/156 [06:16<09:27,  6.03s/it]
Training Epoch: 11/100, batch 124/313 completed (loss: 0.043:  40%|███▉      | 62/156 [06:16<09:27,  6.03s/it] 
Training Epoch: 11/100, batch 124/313 completed (loss: 0.0116:  40%|███▉      | 62/156 [06:16<09:27,  6.03s/it]
Training Epoch: 11/100, batch 124/313 completed (loss: 0.4248:  40%|███▉      | 62/156 [06:16<09:27,  6.03s/it]
Training Epoch: 11/100, batch 124/313 completed (loss: 0.0116:  40%|████      | 63/156 [06:20<09:27,  6.11s/it]
Training Epoch: 11/100, batch 124/313 completed (loss: 0.4248:  40%|████      | 63/156 [06:20<09:28,  6.11s/it]
Training Epoch: 11/100, batch 124/313 completed (loss: 0.5677:  40%|████      | 63/156 [06:20<09:28,  6.11s/it]
Training Epoch: 11/100, batch 124/313 completed (loss: 0.043:  40%|████      | 63/156 [06:20<09:28,  6.12s/it]
Training Epoch: 11/100, batch 125/313 completed (loss: 0.1168:  40%|████      | 63/156 [06:20<09:27,  6.11s/it]
Training Epoch: 11/100, batch 125/313 completed (loss: 0.0265:  40%|████      | 63/156 [06:20<09:28,  6.12s/it]
Training Epoch: 11/100, batch 125/313 completed (loss: 0.0565:  40%|████      | 63/156 [06:20<09:28,  6.11s/it]
Training Epoch: 11/100, batch 125/313 completed (loss: 0.0397:  40%|████      | 63/156 [06:20<09:28,  6.11s/it]
Training Epoch: 11/100, batch 126/313 completed (loss: 0.0304:  40%|████      | 63/156 [06:23<09:27,  6.11s/it]
Training Epoch: 11/100, batch 126/313 completed (loss: 0.1654:  40%|████      | 63/156 [06:23<09:28,  6.12s/it]
Training Epoch: 11/100, batch 126/313 completed (loss: 0.0635:  40%|████      | 63/156 [06:23<09:28,  6.11s/it]
Training Epoch: 11/100, batch 126/313 completed (loss: 0.3489:  40%|████      | 63/156 [06:23<09:28,  6.11s/it]
Training Epoch: 11/100, batch 126/313 completed (loss: 0.0304:  41%|████      | 64/156 [06:26<09:20,  6.09s/it]
Training Epoch: 11/100, batch 126/313 completed (loss: 0.0635:  41%|████      | 64/156 [06:26<09:20,  6.09s/it]
Training Epoch: 11/100, batch 126/313 completed (loss: 0.3489:  41%|████      | 64/156 [06:26<09:20,  6.09s/it]
Training Epoch: 11/100, batch 126/313 completed (loss: 0.1654:  41%|████      | 64/156 [06:26<09:20,  6.09s/it]
Training Epoch: 11/100, batch 127/313 completed (loss: 0.2775:  41%|████      | 64/156 [06:26<09:20,  6.09s/it]
Training Epoch: 11/100, batch 127/313 completed (loss: 0.2759:  41%|████      | 64/156 [06:26<09:20,  6.09s/it]
Training Epoch: 11/100, batch 127/313 completed (loss: 0.0288:  41%|████      | 64/156 [06:26<09:20,  6.09s/it]
Training Epoch: 11/100, batch 127/313 completed (loss: 0.0643:  41%|████      | 64/156 [06:26<09:20,  6.09s/it]
Training Epoch: 11/100, batch 128/313 completed (loss: 0.132:  41%|████      | 64/156 [06:29<09:20,  6.09s/it] 
Training Epoch: 11/100, batch 128/313 completed (loss: 0.3942:  41%|████      | 64/156 [06:29<09:20,  6.09s/it]
Training Epoch: 11/100, batch 128/313 completed (loss: 0.328:  41%|████      | 64/156 [06:29<09:20,  6.09s/it] 
Training Epoch: 11/100, batch 128/313 completed (loss: 0.2335:  41%|████      | 64/156 [06:29<09:20,  6.09s/it]
Training Epoch: 11/100, batch 128/313 completed (loss: 0.328:  42%|████▏     | 65/156 [06:32<09:11,  6.06s/it]
Training Epoch: 11/100, batch 128/313 completed (loss: 0.2335:  42%|████▏     | 65/156 [06:32<09:11,  6.06s/it]
Training Epoch: 11/100, batch 128/313 completed (loss: 0.132:  42%|████▏     | 65/156 [06:32<09:11,  6.06s/it]
Training Epoch: 11/100, batch 128/313 completed (loss: 0.3942:  42%|████▏     | 65/156 [06:32<09:11,  6.06s/it]
Training Epoch: 11/100, batch 129/313 completed (loss: 0.0437:  42%|████▏     | 65/156 [06:32<09:11,  6.06s/it]
Training Epoch: 11/100, batch 129/313 completed (loss: 0.0311:  42%|████▏     | 65/156 [06:32<09:11,  6.06s/it]
Training Epoch: 11/100, batch 129/313 completed (loss: 0.198:  42%|████▏     | 65/156 [06:32<09:11,  6.06s/it]
Training Epoch: 11/100, batch 129/313 completed (loss: 0.0855:  42%|████▏     | 65/156 [06:32<09:11,  6.06s/it]
Training Epoch: 11/100, batch 130/313 completed (loss: 0.0594:  42%|████▏     | 65/156 [06:35<09:11,  6.06s/it]
Training Epoch: 11/100, batch 130/313 completed (loss: 0.0312:  42%|████▏     | 65/156 [06:35<09:11,  6.06s/it]
Training Epoch: 11/100, batch 130/313 completed (loss: 0.0528:  42%|████▏     | 65/156 [06:35<09:11,  6.06s/it]
Training Epoch: 11/100, batch 130/313 completed (loss: 0.0432:  42%|████▏     | 65/156 [06:35<09:11,  6.06s/it]
Training Epoch: 11/100, batch 130/313 completed (loss: 0.0312:  42%|████▏     | 66/156 [06:38<09:06,  6.08s/it]
Training Epoch: 11/100, batch 130/313 completed (loss: 0.0528:  42%|████▏     | 66/156 [06:38<09:06,  6.08s/it]
Training Epoch: 11/100, batch 130/313 completed (loss: 0.0594:  42%|████▏     | 66/156 [06:38<09:07,  6.08s/it]
Training Epoch: 11/100, batch 130/313 completed (loss: 0.0432:  42%|████▏     | 66/156 [06:38<09:07,  6.08s/it]
Training Epoch: 11/100, batch 131/313 completed (loss: 0.091:  42%|████▏     | 66/156 [06:38<09:06,  6.08s/it] 
Training Epoch: 11/100, batch 131/313 completed (loss: 0.0066:  42%|████▏     | 66/156 [06:38<09:06,  6.08s/it]
Training Epoch: 11/100, batch 131/313 completed (loss: 0.0691:  42%|████▏     | 66/156 [06:38<09:07,  6.08s/it]
Training Epoch: 11/100, batch 131/313 completed (loss: 0.1662:  42%|████▏     | 66/156 [06:38<09:07,  6.08s/it]
Training Epoch: 11/100, batch 132/313 completed (loss: 0.0208:  42%|████▏     | 66/156 [06:41<09:06,  6.08s/it]
Training Epoch: 11/100, batch 132/313 completed (loss: 0.0474:  42%|████▏     | 66/156 [06:41<09:07,  6.08s/it]
Training Epoch: 11/100, batch 132/313 completed (loss: 0.2213:  42%|████▏     | 66/156 [06:41<09:06,  6.08s/it]
Training Epoch: 11/100, batch 132/313 completed (loss: 0.084:  42%|████▏     | 66/156 [06:41<09:07,  6.08s/it] 
Training Epoch: 11/100, batch 132/313 completed (loss: 0.0208:  43%|████▎     | 67/156 [06:44<08:59,  6.06s/it]
Training Epoch: 11/100, batch 132/313 completed (loss: 0.2213:  43%|████▎     | 67/156 [06:44<08:59,  6.06s/it]
Training Epoch: 11/100, batch 132/313 completed (loss: 0.0474:  43%|████▎     | 67/156 [06:44<08:59,  6.06s/it]
Training Epoch: 11/100, batch 132/313 completed (loss: 0.084:  43%|████▎     | 67/156 [06:44<08:59,  6.06s/it]
Training Epoch: 11/100, batch 133/313 completed (loss: 0.042:  43%|████▎     | 67/156 [06:44<08:59,  6.06s/it] 
Training Epoch: 11/100, batch 133/313 completed (loss: 0.1098:  43%|████▎     | 67/156 [06:44<08:59,  6.06s/it]
Training Epoch: 11/100, batch 133/313 completed (loss: 0.2551:  43%|████▎     | 67/156 [06:44<08:59,  6.06s/it]
Training Epoch: 11/100, batch 133/313 completed (loss: 0.0429:  43%|████▎     | 67/156 [06:44<08:59,  6.06s/it]
Training Epoch: 11/100, batch 134/313 completed (loss: 0.138:  43%|████▎     | 67/156 [06:47<08:59,  6.06s/it]
Training Epoch: 11/100, batch 134/313 completed (loss: 0.0202:  43%|████▎     | 67/156 [06:47<08:59,  6.06s/it]
Training Epoch: 11/100, batch 134/313 completed (loss: 0.0112:  43%|████▎     | 67/156 [06:47<08:59,  6.06s/it]
Training Epoch: 11/100, batch 134/313 completed (loss: 0.0868:  43%|████▎     | 67/156 [06:47<08:59,  6.06s/it]
Training Epoch: 11/100, batch 134/313 completed (loss: 0.138:  44%|████▎     | 68/156 [06:50<08:59,  6.13s/it]
Training Epoch: 11/100, batch 134/313 completed (loss: 0.0202:  44%|████▎     | 68/156 [06:50<08:59,  6.13s/it]
Training Epoch: 11/100, batch 134/313 completed (loss: 0.0868:  44%|████▎     | 68/156 [06:50<08:59,  6.13s/it]
Training Epoch: 11/100, batch 134/313 completed (loss: 0.0112:  44%|████▎     | 68/156 [06:50<08:59,  6.13s/it]
Training Epoch: 11/100, batch 135/313 completed (loss: 0.1591:  44%|████▎     | 68/156 [06:50<08:59,  6.13s/it]
Training Epoch: 11/100, batch 135/313 completed (loss: 0.0177:  44%|████▎     | 68/156 [06:50<08:59,  6.13s/it]
Training Epoch: 11/100, batch 135/313 completed (loss: 0.0257:  44%|████▎     | 68/156 [06:50<08:59,  6.13s/it]
Training Epoch: 11/100, batch 135/313 completed (loss: 0.0516:  44%|████▎     | 68/156 [06:50<08:59,  6.13s/it]
Training Epoch: 11/100, batch 136/313 completed (loss: 0.0129:  44%|████▎     | 68/156 [06:53<08:59,  6.13s/it]
Training Epoch: 11/100, batch 136/313 completed (loss: 0.0845:  44%|████▎     | 68/156 [06:53<08:59,  6.13s/it]
Training Epoch: 11/100, batch 136/313 completed (loss: 0.619:  44%|████▎     | 68/156 [06:53<08:59,  6.13s/it] 
Training Epoch: 11/100, batch 136/313 completed (loss: 0.0987:  44%|████▎     | 68/156 [06:53<08:59,  6.13s/it]
Training Epoch: 11/100, batch 136/313 completed (loss: 0.619:  44%|████▍     | 69/156 [06:56<08:45,  6.04s/it]
Training Epoch: 11/100, batch 136/313 completed (loss: 0.0987:  44%|████▍     | 69/156 [06:56<08:45,  6.04s/it]
Training Epoch: 11/100, batch 136/313 completed (loss: 0.0845:  44%|████▍     | 69/156 [06:56<08:45,  6.04s/it]
Training Epoch: 11/100, batch 136/313 completed (loss: 0.0129:  44%|████▍     | 69/156 [06:56<08:45,  6.04s/it]
Training Epoch: 11/100, batch 137/313 completed (loss: 0.0554:  44%|████▍     | 69/156 [06:56<08:45,  6.04s/it]
Training Epoch: 11/100, batch 137/313 completed (loss: 0.0357:  44%|████▍     | 69/156 [06:56<08:45,  6.04s/it]
Training Epoch: 11/100, batch 137/313 completed (loss: 0.1192:  44%|████▍     | 69/156 [06:56<08:45,  6.04s/it]
Training Epoch: 11/100, batch 137/313 completed (loss: 0.219:  44%|████▍     | 69/156 [06:56<08:45,  6.04s/it] 
Training Epoch: 11/100, batch 138/313 completed (loss: 0.108:  44%|████▍     | 69/156 [06:59<08:45,  6.04s/it] 
Training Epoch: 11/100, batch 138/313 completed (loss: 0.5408:  44%|████▍     | 69/156 [06:59<08:45,  6.04s/it]
Training Epoch: 11/100, batch 138/313 completed (loss: 0.1023:  44%|████▍     | 69/156 [06:59<08:45,  6.04s/it]
Training Epoch: 11/100, batch 138/313 completed (loss: 0.0826:  44%|████▍     | 69/156 [06:59<08:45,  6.04s/it]
Training Epoch: 11/100, batch 138/313 completed (loss: 0.108:  45%|████▍     | 70/156 [07:02<08:43,  6.09s/it]
Training Epoch: 11/100, batch 138/313 completed (loss: 0.1023:  45%|████▍     | 70/156 [07:02<08:43,  6.09s/it]
Training Epoch: 11/100, batch 138/313 completed (loss: 0.0826:  45%|████▍     | 70/156 [07:02<08:44,  6.09s/it]
Training Epoch: 11/100, batch 138/313 completed (loss: 0.5408:  45%|████▍     | 70/156 [07:02<08:44,  6.10s/it]
Training Epoch: 11/100, batch 139/313 completed (loss: 0.0708:  45%|████▍     | 70/156 [07:02<08:44,  6.10s/it]
Training Epoch: 11/100, batch 139/313 completed (loss: 0.0477:  45%|████▍     | 70/156 [07:02<08:43,  6.09s/it]
Training Epoch: 11/100, batch 139/313 completed (loss: 0.0281:  45%|████▍     | 70/156 [07:02<08:43,  6.09s/it]
Training Epoch: 11/100, batch 139/313 completed (loss: 0.0549:  45%|████▍     | 70/156 [07:02<08:44,  6.09s/it]
Training Epoch: 11/100, batch 140/313 completed (loss: 0.0432:  45%|████▍     | 70/156 [07:05<08:44,  6.10s/it]
Training Epoch: 11/100, batch 140/313 completed (loss: 0.0488:  45%|████▍     | 70/156 [07:05<08:43,  6.09s/it]
Training Epoch: 11/100, batch 140/313 completed (loss: 0.1287:  45%|████▍     | 70/156 [07:05<08:43,  6.09s/it]
Training Epoch: 11/100, batch 140/313 completed (loss: 0.1027:  45%|████▍     | 70/156 [07:05<08:44,  6.09s/it]
Training Epoch: 11/100, batch 140/313 completed (loss: 0.0488:  46%|████▌     | 71/156 [07:08<08:40,  6.12s/it]
Training Epoch: 11/100, batch 140/313 completed (loss: 0.1027:  46%|████▌     | 71/156 [07:08<08:40,  6.12s/it]
Training Epoch: 11/100, batch 140/313 completed (loss: 0.0432:  46%|████▌     | 71/156 [07:08<08:40,  6.12s/it]
Training Epoch: 11/100, batch 140/313 completed (loss: 0.1287:  46%|████▌     | 71/156 [07:08<08:40,  6.13s/it]
Training Epoch: 11/100, batch 141/313 completed (loss: 0.1824:  46%|████▌     | 71/156 [07:08<08:40,  6.13s/it]
Training Epoch: 11/100, batch 141/313 completed (loss: 0.1063:  46%|████▌     | 71/156 [07:08<08:40,  6.12s/it]
Training Epoch: 11/100, batch 141/313 completed (loss: 0.3251:  46%|████▌     | 71/156 [07:08<08:40,  6.12s/it]
Training Epoch: 11/100, batch 141/313 completed (loss: 0.5034:  46%|████▌     | 71/156 [07:08<08:40,  6.12s/it]
Training Epoch: 11/100, batch 142/313 completed (loss: 0.1035:  46%|████▌     | 71/156 [07:11<08:40,  6.12s/it]
Training Epoch: 11/100, batch 142/313 completed (loss: 0.039:  46%|████▌     | 71/156 [07:11<08:40,  6.13s/it] 
Training Epoch: 11/100, batch 142/313 completed (loss: 0.0566:  46%|████▌     | 71/156 [07:11<08:40,  6.12s/it]
Training Epoch: 11/100, batch 142/313 completed (loss: 0.1749:  46%|████▌     | 71/156 [07:11<08:40,  6.12s/it]
Training Epoch: 11/100, batch 142/313 completed (loss: 0.0566:  46%|████▌     | 72/156 [07:14<08:30,  6.08s/it]
Training Epoch: 11/100, batch 142/313 completed (loss: 0.1035:  46%|████▌     | 72/156 [07:14<08:30,  6.08s/it]
Training Epoch: 11/100, batch 142/313 completed (loss: 0.1749:  46%|████▌     | 72/156 [07:14<08:30,  6.08s/it]
Training Epoch: 11/100, batch 142/313 completed (loss: 0.039:  46%|████▌     | 72/156 [07:14<08:31,  6.09s/it]
Training Epoch: 11/100, batch 143/313 completed (loss: 0.1087:  46%|████▌     | 72/156 [07:14<08:31,  6.09s/it]
Training Epoch: 11/100, batch 143/313 completed (loss: 0.1481:  46%|████▌     | 72/156 [07:14<08:30,  6.08s/it]
Training Epoch: 11/100, batch 143/313 completed (loss: 0.2182:  46%|████▌     | 72/156 [07:14<08:30,  6.08s/it]
Training Epoch: 11/100, batch 143/313 completed (loss: 0.0298:  46%|████▌     | 72/156 [07:14<08:30,  6.08s/it]
Training Epoch: 11/100, batch 144/313 completed (loss: 0.0522:  46%|████▌     | 72/156 [07:17<08:31,  6.09s/it]
Training Epoch: 11/100, batch 144/313 completed (loss: 0.1:  46%|████▌     | 72/156 [07:17<08:30,  6.08s/it]   
Training Epoch: 11/100, batch 144/313 completed (loss: 0.0483:  46%|████▌     | 72/156 [07:17<08:30,  6.08s/it]
Training Epoch: 11/100, batch 144/313 completed (loss: 0.0588:  46%|████▌     | 72/156 [07:17<08:30,  6.08s/it]
Training Epoch: 11/100, batch 144/313 completed (loss: 0.0588:  47%|████▋     | 73/156 [07:20<08:26,  6.10s/it]
Training Epoch: 11/100, batch 144/313 completed (loss: 0.0483:  47%|████▋     | 73/156 [07:20<08:26,  6.10s/it]
Training Epoch: 11/100, batch 144/313 completed (loss: 0.1:  47%|████▋     | 73/156 [07:20<08:26,  6.10s/it]
Training Epoch: 11/100, batch 144/313 completed (loss: 0.0522:  47%|████▋     | 73/156 [07:20<08:26,  6.10s/it]
Training Epoch: 11/100, batch 145/313 completed (loss: 0.1299:  47%|████▋     | 73/156 [07:20<08:26,  6.10s/it]
Training Epoch: 11/100, batch 145/313 completed (loss: 0.1133:  47%|████▋     | 73/156 [07:20<08:26,  6.10s/it]
Training Epoch: 11/100, batch 145/313 completed (loss: 0.1767:  47%|████▋     | 73/156 [07:20<08:26,  6.10s/it]
Training Epoch: 11/100, batch 145/313 completed (loss: 0.1706:  47%|████▋     | 73/156 [07:20<08:26,  6.10s/it]
Training Epoch: 11/100, batch 146/313 completed (loss: 0.0306:  47%|████▋     | 73/156 [07:24<08:26,  6.10s/it]
Training Epoch: 11/100, batch 146/313 completed (loss: 0.0845:  47%|████▋     | 73/156 [07:24<08:26,  6.10s/it]
Training Epoch: 11/100, batch 146/313 completed (loss: 0.1472:  47%|████▋     | 73/156 [07:24<08:26,  6.10s/it]
Training Epoch: 11/100, batch 146/313 completed (loss: 0.0244:  47%|████▋     | 73/156 [07:24<08:26,  6.10s/it]
Training Epoch: 11/100, batch 146/313 completed (loss: 0.0845:  47%|████▋     | 74/156 [07:26<08:17,  6.07s/it]
Training Epoch: 11/100, batch 146/313 completed (loss: 0.0244:  47%|████▋     | 74/156 [07:26<08:18,  6.07s/it]
Training Epoch: 11/100, batch 146/313 completed (loss: 0.0306:  47%|████▋     | 74/156 [07:26<08:18,  6.08s/it]
Training Epoch: 11/100, batch 146/313 completed (loss: 0.1472:  47%|████▋     | 74/156 [07:26<08:18,  6.07s/it]
Training Epoch: 11/100, batch 147/313 completed (loss: 0.0612:  47%|████▋     | 74/156 [07:26<08:18,  6.08s/it]
Training Epoch: 11/100, batch 147/313 completed (loss: 0.064:  47%|████▋     | 74/156 [07:26<08:18,  6.07s/it] 
Training Epoch: 11/100, batch 147/313 completed (loss: 0.2524:  47%|████▋     | 74/156 [07:26<08:17,  6.07s/it]
Training Epoch: 11/100, batch 147/313 completed (loss: 0.0399:  47%|████▋     | 74/156 [07:26<08:18,  6.07s/it]
Training Epoch: 11/100, batch 148/313 completed (loss: 0.132:  47%|████▋     | 74/156 [07:30<08:17,  6.07s/it] 
Training Epoch: 11/100, batch 148/313 completed (loss: 0.0266:  47%|████▋     | 74/156 [07:30<08:18,  6.08s/it]
Training Epoch: 11/100, batch 148/313 completed (loss: 0.1472:  47%|████▋     | 74/156 [07:30<08:18,  6.07s/it]
Training Epoch: 11/100, batch 148/313 completed (loss: 0.3945:  47%|████▋     | 74/156 [07:30<08:18,  6.07s/it]
Training Epoch: 11/100, batch 148/313 completed (loss: 0.3945:  48%|████▊     | 75/156 [07:33<08:13,  6.09s/it]
Training Epoch: 11/100, batch 148/313 completed (loss: 0.1472:  48%|████▊     | 75/156 [07:33<08:13,  6.09s/it]
Training Epoch: 11/100, batch 148/313 completed (loss: 0.0266:  48%|████▊     | 75/156 [07:33<08:13,  6.09s/it]
Training Epoch: 11/100, batch 148/313 completed (loss: 0.132:  48%|████▊     | 75/156 [07:33<08:13,  6.09s/it]
Training Epoch: 11/100, batch 149/313 completed (loss: 0.0385:  48%|████▊     | 75/156 [07:33<08:13,  6.09s/it]
Training Epoch: 11/100, batch 149/313 completed (loss: 0.0928:  48%|████▊     | 75/156 [07:33<08:13,  6.09s/it]
Training Epoch: 11/100, batch 149/313 completed (loss: 0.0474:  48%|████▊     | 75/156 [07:33<08:13,  6.09s/it]
Training Epoch: 11/100, batch 149/313 completed (loss: 0.0147:  48%|████▊     | 75/156 [07:33<08:13,  6.09s/it]
Training Epoch: 11/100, batch 150/313 completed (loss: 0.0625:  48%|████▊     | 75/156 [07:36<08:13,  6.09s/it]
Training Epoch: 11/100, batch 150/313 completed (loss: 0.0358:  48%|████▊     | 75/156 [07:36<08:13,  6.09s/it]
Training Epoch: 11/100, batch 150/313 completed (loss: 0.0278:  48%|████▊     | 75/156 [07:36<08:13,  6.09s/it]
Training Epoch: 11/100, batch 150/313 completed (loss: 0.0305:  48%|████▊     | 75/156 [07:36<08:13,  6.09s/it]
Training Epoch: 11/100, batch 150/313 completed (loss: 0.0305:  49%|████▊     | 76/156 [07:39<08:09,  6.12s/it]
Training Epoch: 11/100, batch 150/313 completed (loss: 0.0358:  49%|████▊     | 76/156 [07:39<08:09,  6.12s/it]
Training Epoch: 11/100, batch 150/313 completed (loss: 0.0278:  49%|████▊     | 76/156 [07:39<08:09,  6.12s/it]
Training Epoch: 11/100, batch 150/313 completed (loss: 0.0625:  49%|████▊     | 76/156 [07:39<08:09,  6.12s/it]
Training Epoch: 11/100, batch 151/313 completed (loss: 0.0964:  49%|████▊     | 76/156 [07:39<08:09,  6.12s/it]
Training Epoch: 11/100, batch 151/313 completed (loss: 0.0625:  49%|████▊     | 76/156 [07:39<08:09,  6.12s/it]
Training Epoch: 11/100, batch 151/313 completed (loss: 0.2683:  49%|████▊     | 76/156 [07:39<08:09,  6.12s/it]
Training Epoch: 11/100, batch 151/313 completed (loss: 0.7857:  49%|████▊     | 76/156 [07:39<08:09,  6.12s/it]
Training Epoch: 11/100, batch 152/313 completed (loss: 0.0992:  49%|████▊     | 76/156 [07:42<08:09,  6.12s/it]
Training Epoch: 11/100, batch 152/313 completed (loss: 0.1374:  49%|████▊     | 76/156 [07:42<08:09,  6.12s/it]
Training Epoch: 11/100, batch 152/313 completed (loss: 0.0395:  49%|████▊     | 76/156 [07:42<08:09,  6.12s/it]
Training Epoch: 11/100, batch 152/313 completed (loss: 0.0796:  49%|████▊     | 76/156 [07:42<08:09,  6.12s/it]
Training Epoch: 11/100, batch 152/313 completed (loss: 0.1374:  49%|████▉     | 77/156 [07:45<07:56,  6.04s/it]
Training Epoch: 11/100, batch 152/313 completed (loss: 0.0992:  49%|████▉     | 77/156 [07:45<07:56,  6.04s/it]
Training Epoch: 11/100, batch 152/313 completed (loss: 0.0796:  49%|████▉     | 77/156 [07:45<07:57,  6.04s/it]
Training Epoch: 11/100, batch 152/313 completed (loss: 0.0395:  49%|████▉     | 77/156 [07:45<07:56,  6.04s/it]
Training Epoch: 11/100, batch 153/313 completed (loss: 0.0853:  49%|████▉     | 77/156 [07:45<07:56,  6.04s/it]
Training Epoch: 11/100, batch 153/313 completed (loss: 0.1189:  49%|████▉     | 77/156 [07:45<07:56,  6.04s/it]
Training Epoch: 11/100, batch 153/313 completed (loss: 0.0785:  49%|████▉     | 77/156 [07:45<07:56,  6.04s/it]
Training Epoch: 11/100, batch 153/313 completed (loss: 0.1158:  49%|████▉     | 77/156 [07:45<07:57,  6.04s/it]
Training Epoch: 11/100, batch 154/313 completed (loss: 0.0228:  49%|████▉     | 77/156 [07:48<07:56,  6.04s/it]
Training Epoch: 11/100, batch 154/313 completed (loss: 0.1574:  49%|████▉     | 77/156 [07:48<07:56,  6.04s/it]
Training Epoch: 11/100, batch 154/313 completed (loss: 0.0154:  49%|████▉     | 77/156 [07:48<07:56,  6.04s/it]
Training Epoch: 11/100, batch 154/313 completed (loss: 0.0348:  49%|████▉     | 77/156 [07:48<07:57,  6.04s/it]
Training Epoch: 11/100, batch 154/313 completed (loss: 0.1574:  50%|█████     | 78/156 [07:51<07:56,  6.11s/it]
Training Epoch: 11/100, batch 154/313 completed (loss: 0.0348:  50%|█████     | 78/156 [07:51<07:57,  6.12s/it]
Training Epoch: 11/100, batch 154/313 completed (loss: 0.0154:  50%|█████     | 78/156 [07:51<07:57,  6.12s/it]
Training Epoch: 11/100, batch 154/313 completed (loss: 0.0228:  50%|█████     | 78/156 [07:51<07:57,  6.12s/it]
Training Epoch: 11/100, batch 155/313 completed (loss: 0.1593:  50%|█████     | 78/156 [07:51<07:56,  6.11s/it]
Training Epoch: 11/100, batch 155/313 completed (loss: 0.0178:  50%|█████     | 78/156 [07:51<07:57,  6.12s/it]
Training Epoch: 11/100, batch 155/313 completed (loss: 0.3459:  50%|█████     | 78/156 [07:51<07:57,  6.12s/it]
Training Epoch: 11/100, batch 155/313 completed (loss: 0.1832:  50%|█████     | 78/156 [07:51<07:57,  6.12s/it]
Training Epoch: 11/100, batch 156/313 completed (loss: 0.0437:  50%|█████     | 78/156 [07:54<07:57,  6.12s/it]
Training Epoch: 11/100, batch 156/313 completed (loss: 0.2692:  50%|█████     | 78/156 [07:54<07:57,  6.12s/it]
Training Epoch: 11/100, batch 156/313 completed (loss: 0.0132:  50%|█████     | 78/156 [07:54<07:56,  6.11s/it]
Training Epoch: 11/100, batch 156/313 completed (loss: 0.3758:  50%|█████     | 78/156 [07:54<07:57,  6.12s/it]
Training Epoch: 11/100, batch 156/313 completed (loss: 0.3758:  51%|█████     | 79/156 [07:57<07:48,  6.09s/it]
Training Epoch: 11/100, batch 156/313 completed (loss: 0.2692:  51%|█████     | 79/156 [07:57<07:48,  6.09s/it]
Training Epoch: 11/100, batch 156/313 completed (loss: 0.0132:  51%|█████     | 79/156 [07:57<07:48,  6.09s/it]
Training Epoch: 11/100, batch 156/313 completed (loss: 0.0437:  51%|█████     | 79/156 [07:57<07:48,  6.09s/it]
Training Epoch: 11/100, batch 157/313 completed (loss: 0.1965:  51%|█████     | 79/156 [07:57<07:48,  6.09s/it]
Training Epoch: 11/100, batch 157/313 completed (loss: 0.0104:  51%|█████     | 79/156 [07:57<07:48,  6.09s/it]
Training Epoch: 11/100, batch 157/313 completed (loss: 0.2196:  51%|█████     | 79/156 [07:57<07:48,  6.09s/it]
Training Epoch: 11/100, batch 157/313 completed (loss: 0.1315:  51%|█████     | 79/156 [07:57<07:48,  6.09s/it]
Training Epoch: 11/100, batch 158/313 completed (loss: 0.0192:  51%|█████     | 79/156 [08:00<07:48,  6.09s/it]
Training Epoch: 11/100, batch 158/313 completed (loss: 0.2609:  51%|█████     | 79/156 [08:00<07:48,  6.09s/it]
Training Epoch: 11/100, batch 158/313 completed (loss: 0.1277:  51%|█████     | 79/156 [08:00<07:48,  6.09s/it]
Training Epoch: 11/100, batch 158/313 completed (loss: 0.0116:  51%|█████     | 79/156 [08:00<07:48,  6.09s/it]
Training Epoch: 11/100, batch 158/313 completed (loss: 0.1277:  51%|█████▏    | 80/156 [08:03<07:40,  6.06s/it]
Training Epoch: 11/100, batch 158/313 completed (loss: 0.2609:  51%|█████▏    | 80/156 [08:03<07:40,  6.06s/it]
Training Epoch: 11/100, batch 158/313 completed (loss: 0.0116:  51%|█████▏    | 80/156 [08:03<07:40,  6.06s/it]
Training Epoch: 11/100, batch 158/313 completed (loss: 0.0192:  51%|█████▏    | 80/156 [08:03<07:40,  6.06s/it]
Training Epoch: 11/100, batch 159/313 completed (loss: 0.1637:  51%|█████▏    | 80/156 [08:03<07:40,  6.06s/it]
Training Epoch: 11/100, batch 159/313 completed (loss: 0.1054:  51%|█████▏    | 80/156 [08:03<07:40,  6.06s/it]
Training Epoch: 11/100, batch 159/313 completed (loss: 0.0424:  51%|█████▏    | 80/156 [08:03<07:40,  6.06s/it]
Training Epoch: 11/100, batch 159/313 completed (loss: 0.0435:  51%|█████▏    | 80/156 [08:03<07:40,  6.06s/it]
Training Epoch: 11/100, batch 160/313 completed (loss: 0.2054:  51%|█████▏    | 80/156 [08:06<07:40,  6.06s/it]
Training Epoch: 11/100, batch 160/313 completed (loss: 0.3842:  51%|█████▏    | 80/156 [08:06<07:40,  6.06s/it]
Training Epoch: 11/100, batch 160/313 completed (loss: 0.096:  51%|█████▏    | 80/156 [08:06<07:40,  6.06s/it] 
Training Epoch: 11/100, batch 160/313 completed (loss: 0.0628:  51%|█████▏    | 80/156 [08:06<07:40,  6.06s/it]
Training Epoch: 11/100, batch 160/313 completed (loss: 0.0628:  52%|█████▏    | 81/156 [08:09<07:40,  6.13s/it]
Training Epoch: 11/100, batch 160/313 completed (loss: 0.096:  52%|█████▏    | 81/156 [08:09<07:40,  6.13s/it]
Training Epoch: 11/100, batch 160/313 completed (loss: 0.2054:  52%|█████▏    | 81/156 [08:09<07:40,  6.14s/it]
Training Epoch: 11/100, batch 160/313 completed (loss: 0.3842:  52%|█████▏    | 81/156 [08:09<07:40,  6.14s/it]
Training Epoch: 11/100, batch 161/313 completed (loss: 0.0083:  52%|█████▏    | 81/156 [08:09<07:40,  6.14s/it]
Training Epoch: 11/100, batch 161/313 completed (loss: 0.126:  52%|█████▏    | 81/156 [08:09<07:40,  6.14s/it] 
Training Epoch: 11/100, batch 161/313 completed (loss: 0.0658:  52%|█████▏    | 81/156 [08:09<07:40,  6.13s/it]
Training Epoch: 11/100, batch 161/313 completed (loss: 0.064:  52%|█████▏    | 81/156 [08:09<07:40,  6.13s/it] 
Training Epoch: 11/100, batch 162/313 completed (loss: 0.025:  52%|█████▏    | 81/156 [08:12<07:40,  6.13s/it] 
Training Epoch: 11/100, batch 162/313 completed (loss: 0.1673:  52%|█████▏    | 81/156 [08:12<07:40,  6.14s/it]
Training Epoch: 11/100, batch 162/313 completed (loss: 0.2351:  52%|█████▏    | 81/156 [08:12<07:40,  6.14s/it]
Training Epoch: 11/100, batch 162/313 completed (loss: 0.0909:  52%|█████▏    | 81/156 [08:12<07:40,  6.13s/it]
Training Epoch: 11/100, batch 162/313 completed (loss: 0.0909:  53%|█████▎    | 82/156 [08:15<07:27,  6.05s/it]
Training Epoch: 11/100, batch 162/313 completed (loss: 0.025:  53%|█████▎    | 82/156 [08:15<07:27,  6.05s/it]
Training Epoch: 11/100, batch 162/313 completed (loss: 0.1673:  53%|█████▎    | 82/156 [08:15<07:27,  6.05s/it]
Training Epoch: 11/100, batch 162/313 completed (loss: 0.2351:  53%|█████▎    | 82/156 [08:15<07:27,  6.05s/it]
Training Epoch: 11/100, batch 163/313 completed (loss: 0.0926:  53%|█████▎    | 82/156 [08:15<07:27,  6.05s/it]
Training Epoch: 11/100, batch 163/313 completed (loss: 0.1162:  53%|█████▎    | 82/156 [08:15<07:27,  6.05s/it]
Training Epoch: 11/100, batch 163/313 completed (loss: 0.0325:  53%|█████▎    | 82/156 [08:15<07:27,  6.05s/it]
Training Epoch: 11/100, batch 163/313 completed (loss: 0.0543:  53%|█████▎    | 82/156 [08:15<07:27,  6.05s/it]
Training Epoch: 11/100, batch 164/313 completed (loss: 0.3672:  53%|█████▎    | 82/156 [08:18<07:27,  6.05s/it]
Training Epoch: 11/100, batch 164/313 completed (loss: 0.1189:  53%|█████▎    | 82/156 [08:18<07:27,  6.05s/it]
Training Epoch: 11/100, batch 164/313 completed (loss: 0.0421:  53%|█████▎    | 82/156 [08:18<07:27,  6.05s/it]
Training Epoch: 11/100, batch 164/313 completed (loss: 0.0279:  53%|█████▎    | 82/156 [08:18<07:27,  6.05s/it]
Training Epoch: 11/100, batch 164/313 completed (loss: 0.3672:  53%|█████▎    | 83/156 [08:21<07:23,  6.07s/it]
Training Epoch: 11/100, batch 164/313 completed (loss: 0.0279:  53%|█████▎    | 83/156 [08:21<07:23,  6.08s/it]
Training Epoch: 11/100, batch 164/313 completed (loss: 0.1189:  53%|█████▎    | 83/156 [08:21<07:23,  6.07s/it]
Training Epoch: 11/100, batch 164/313 completed (loss: 0.0421:  53%|█████▎    | 83/156 [08:21<07:23,  6.08s/it]
Training Epoch: 11/100, batch 165/313 completed (loss: 0.0177:  53%|█████▎    | 83/156 [08:21<07:23,  6.08s/it]
Training Epoch: 11/100, batch 165/313 completed (loss: 0.0793:  53%|█████▎    | 83/156 [08:21<07:23,  6.07s/it]
Training Epoch: 11/100, batch 165/313 completed (loss: 0.2428:  53%|█████▎    | 83/156 [08:21<07:23,  6.07s/it]
Training Epoch: 11/100, batch 165/313 completed (loss: 0.19:  53%|█████▎    | 83/156 [08:21<07:23,  6.08s/it]  
Training Epoch: 11/100, batch 166/313 completed (loss: 0.0693:  53%|█████▎    | 83/156 [08:24<07:23,  6.07s/it]
Training Epoch: 11/100, batch 166/313 completed (loss: 0.0295:  53%|█████▎    | 83/156 [08:24<07:23,  6.07s/it]
Training Epoch: 11/100, batch 166/313 completed (loss: 0.0699:  53%|█████▎    | 83/156 [08:24<07:23,  6.08s/it]
Training Epoch: 11/100, batch 166/313 completed (loss: 0.0564:  53%|█████▎    | 83/156 [08:24<07:23,  6.08s/it]
Training Epoch: 11/100, batch 166/313 completed (loss: 0.0295:  54%|█████▍    | 84/156 [08:27<07:18,  6.09s/it]
Training Epoch: 11/100, batch 166/313 completed (loss: 0.0699:  54%|█████▍    | 84/156 [08:27<07:18,  6.09s/it]
Training Epoch: 11/100, batch 166/313 completed (loss: 0.0564:  54%|█████▍    | 84/156 [08:27<07:18,  6.09s/it]
Training Epoch: 11/100, batch 166/313 completed (loss: 0.0693:  54%|█████▍    | 84/156 [08:27<07:18,  6.09s/it]
Training Epoch: 11/100, batch 167/313 completed (loss: 0.052:  54%|█████▍    | 84/156 [08:27<07:18,  6.09s/it] 
Training Epoch: 11/100, batch 167/313 completed (loss: 0.041:  54%|█████▍    | 84/156 [08:27<07:18,  6.09s/it] 
Training Epoch: 11/100, batch 167/313 completed (loss: 0.2584:  54%|█████▍    | 84/156 [08:27<07:18,  6.09s/it]
Training Epoch: 11/100, batch 167/313 completed (loss: 0.2756:  54%|█████▍    | 84/156 [08:27<07:18,  6.09s/it]
Training Epoch: 11/100, batch 168/313 completed (loss: 0.0605:  54%|█████▍    | 84/156 [08:30<07:18,  6.09s/it]
Training Epoch: 11/100, batch 168/313 completed (loss: 0.0248:  54%|█████▍    | 84/156 [08:30<07:18,  6.09s/it]
Training Epoch: 11/100, batch 168/313 completed (loss: 0.047:  54%|█████▍    | 84/156 [08:30<07:18,  6.09s/it]
Training Epoch: 11/100, batch 168/313 completed (loss: 0.3142:  54%|█████▍    | 84/156 [08:30<07:18,  6.09s/it]
Training Epoch: 11/100, batch 168/313 completed (loss: 0.047:  54%|█████▍    | 85/156 [08:33<07:07,  6.02s/it]
Training Epoch: 11/100, batch 168/313 completed (loss: 0.0248:  54%|█████▍    | 85/156 [08:33<07:07,  6.02s/it]
Training Epoch: 11/100, batch 168/313 completed (loss: 0.0605:  54%|█████▍    | 85/156 [08:33<07:07,  6.02s/it]
Training Epoch: 11/100, batch 168/313 completed (loss: 0.3142:  54%|█████▍    | 85/156 [08:33<07:07,  6.02s/it]
Training Epoch: 11/100, batch 169/313 completed (loss: 0.1709:  54%|█████▍    | 85/156 [08:33<07:07,  6.02s/it]
Training Epoch: 11/100, batch 169/313 completed (loss: 0.0615:  54%|█████▍    | 85/156 [08:33<07:07,  6.02s/it]
Training Epoch: 11/100, batch 169/313 completed (loss: 0.111:  54%|█████▍    | 85/156 [08:33<07:07,  6.02s/it] 
Training Epoch: 11/100, batch 169/313 completed (loss: 0.0466:  54%|█████▍    | 85/156 [08:33<07:07,  6.02s/it]
Training Epoch: 11/100, batch 170/313 completed (loss: 0.2219:  54%|█████▍    | 85/156 [08:36<07:07,  6.02s/it]
Training Epoch: 11/100, batch 170/313 completed (loss: 0.1594:  54%|█████▍    | 85/156 [08:36<07:07,  6.02s/it]
Training Epoch: 11/100, batch 170/313 completed (loss: 0.0036:  54%|█████▍    | 85/156 [08:36<07:07,  6.02s/it]
Training Epoch: 11/100, batch 170/313 completed (loss: 0.0566:  54%|█████▍    | 85/156 [08:36<07:07,  6.02s/it]
Training Epoch: 11/100, batch 170/313 completed (loss: 0.0036:  55%|█████▌    | 86/156 [08:39<07:07,  6.11s/it]
Training Epoch: 11/100, batch 170/313 completed (loss: 0.0566:  55%|█████▌    | 86/156 [08:40<07:07,  6.11s/it]
Training Epoch: 11/100, batch 170/313 completed (loss: 0.1594:  55%|█████▌    | 86/156 [08:39<07:07,  6.11s/it]
Training Epoch: 11/100, batch 170/313 completed (loss: 0.2219:  55%|█████▌    | 86/156 [08:39<07:07,  6.11s/it]
Training Epoch: 11/100, batch 171/313 completed (loss: 0.0456:  55%|█████▌    | 86/156 [08:40<07:07,  6.11s/it]
Training Epoch: 11/100, batch 171/313 completed (loss: 0.0955:  55%|█████▌    | 86/156 [08:40<07:07,  6.11s/it]
Training Epoch: 11/100, batch 171/313 completed (loss: 0.0653:  55%|█████▌    | 86/156 [08:40<07:07,  6.11s/it]
Training Epoch: 11/100, batch 171/313 completed (loss: 0.0318:  55%|█████▌    | 86/156 [08:40<07:07,  6.11s/it]
Training Epoch: 11/100, batch 172/313 completed (loss: 0.0986:  55%|█████▌    | 86/156 [08:43<07:07,  6.11s/it]
Training Epoch: 11/100, batch 172/313 completed (loss: 0.304:  55%|█████▌    | 86/156 [08:43<07:07,  6.11s/it] 
Training Epoch: 11/100, batch 172/313 completed (loss: 0.1293:  55%|█████▌    | 86/156 [08:43<07:07,  6.11s/it]
Training Epoch: 11/100, batch 172/313 completed (loss: 0.053:  55%|█████▌    | 86/156 [08:43<07:07,  6.11s/it] 
Training Epoch: 11/100, batch 172/313 completed (loss: 0.053:  56%|█████▌    | 87/156 [08:46<06:59,  6.08s/it]
Training Epoch: 11/100, batch 172/313 completed (loss: 0.1293:  56%|█████▌    | 87/156 [08:45<06:59,  6.08s/it]
Training Epoch: 11/100, batch 172/313 completed (loss: 0.0986:  56%|█████▌    | 87/156 [08:45<06:59,  6.08s/it]
Training Epoch: 11/100, batch 172/313 completed (loss: 0.304:  56%|█████▌    | 87/156 [08:45<06:59,  6.08s/it]
Training Epoch: 11/100, batch 173/313 completed (loss: 0.3382:  56%|█████▌    | 87/156 [08:46<06:59,  6.08s/it]
Training Epoch: 11/100, batch 173/313 completed (loss: 0.2771:  56%|█████▌    | 87/156 [08:46<06:59,  6.08s/it]
Training Epoch: 11/100, batch 173/313 completed (loss: 0.035:  56%|█████▌    | 87/156 [08:46<06:59,  6.08s/it] 
Training Epoch: 11/100, batch 173/313 completed (loss: 0.0104:  56%|█████▌    | 87/156 [08:46<06:59,  6.08s/it]
Training Epoch: 11/100, batch 174/313 completed (loss: 0.124:  56%|█████▌    | 87/156 [08:48<06:59,  6.08s/it] 
Training Epoch: 11/100, batch 174/313 completed (loss: 0.007:  56%|█████▌    | 87/156 [08:48<06:59,  6.08s/it] 
Training Epoch: 11/100, batch 174/313 completed (loss: 0.1401:  56%|█████▌    | 87/156 [08:48<06:59,  6.08s/it]
Training Epoch: 11/100, batch 174/313 completed (loss: 0.0324:  56%|█████▌    | 87/156 [08:48<06:59,  6.08s/it]
Training Epoch: 11/100, batch 174/313 completed (loss: 0.1401:  56%|█████▋    | 88/156 [08:52<06:52,  6.06s/it]
Training Epoch: 11/100, batch 174/313 completed (loss: 0.0324:  56%|█████▋    | 88/156 [08:52<06:52,  6.07s/it]
Training Epoch: 11/100, batch 174/313 completed (loss: 0.007:  56%|█████▋    | 88/156 [08:52<06:52,  6.07s/it]
Training Epoch: 11/100, batch 174/313 completed (loss: 0.124:  56%|█████▋    | 88/156 [08:52<06:52,  6.07s/it]
Training Epoch: 11/100, batch 175/313 completed (loss: 0.2268:  56%|█████▋    | 88/156 [08:52<06:52,  6.07s/it]
Training Epoch: 11/100, batch 175/313 completed (loss: 0.0826:  56%|█████▋    | 88/156 [08:52<06:52,  6.07s/it]
Training Epoch: 11/100, batch 175/313 completed (loss: 0.3232:  56%|█████▋    | 88/156 [08:52<06:52,  6.06s/it]
Training Epoch: 11/100, batch 175/313 completed (loss: 0.2323:  56%|█████▋    | 88/156 [08:52<06:52,  6.07s/it]
Training Epoch: 11/100, batch 176/313 completed (loss: 0.1229:  56%|█████▋    | 88/156 [08:55<06:52,  6.07s/it]
Training Epoch: 11/100, batch 176/313 completed (loss: 0.0689:  56%|█████▋    | 88/156 [08:55<06:52,  6.06s/it]
Training Epoch: 11/100, batch 176/313 completed (loss: 0.0688:  56%|█████▋    | 88/156 [08:55<06:52,  6.07s/it]
Training Epoch: 11/100, batch 176/313 completed (loss: 0.0604:  56%|█████▋    | 88/156 [08:55<06:52,  6.07s/it]
Training Epoch: 11/100, batch 176/313 completed (loss: 0.0604:  57%|█████▋    | 89/156 [08:58<06:50,  6.13s/it]
Training Epoch: 11/100, batch 176/313 completed (loss: 0.0689:  57%|█████▋    | 89/156 [08:58<06:50,  6.13s/it]
Training Epoch: 11/100, batch 176/313 completed (loss: 0.0688:  57%|█████▋    | 89/156 [08:58<06:50,  6.13s/it]
Training Epoch: 11/100, batch 176/313 completed (loss: 0.1229:  57%|█████▋    | 89/156 [08:58<06:50,  6.13s/it]
Training Epoch: 11/100, batch 177/313 completed (loss: 0.4022:  57%|█████▋    | 89/156 [08:58<06:50,  6.13s/it]
Training Epoch: 11/100, batch 177/313 completed (loss: 0.1104:  57%|█████▋    | 89/156 [08:58<06:50,  6.13s/it]
Training Epoch: 11/100, batch 177/313 completed (loss: 0.0296:  57%|█████▋    | 89/156 [08:58<06:50,  6.13s/it]
Training Epoch: 11/100, batch 177/313 completed (loss: 0.1104:  57%|█████▋    | 89/156 [08:58<06:50,  6.13s/it]
Training Epoch: 11/100, batch 178/313 completed (loss: 0.1594:  57%|█████▋    | 89/156 [09:01<06:50,  6.13s/it]
Training Epoch: 11/100, batch 178/313 completed (loss: 0.1999:  57%|█████▋    | 89/156 [09:01<06:50,  6.13s/it]
Training Epoch: 11/100, batch 178/313 completed (loss: 0.0204:  57%|█████▋    | 89/156 [09:01<06:50,  6.13s/it]
Training Epoch: 11/100, batch 178/313 completed (loss: 0.0311:  57%|█████▋    | 89/156 [09:01<06:50,  6.13s/it]
Training Epoch: 11/100, batch 178/313 completed (loss: 0.1594:  58%|█████▊    | 90/156 [09:04<06:38,  6.04s/it]
Training Epoch: 11/100, batch 178/313 completed (loss: 0.0311:  58%|█████▊    | 90/156 [09:04<06:38,  6.04s/it]
Training Epoch: 11/100, batch 178/313 completed (loss: 0.0204:  58%|█████▊    | 90/156 [09:04<06:38,  6.04s/it]
Training Epoch: 11/100, batch 178/313 completed (loss: 0.1999:  58%|█████▊    | 90/156 [09:04<06:38,  6.04s/it]
Training Epoch: 11/100, batch 179/313 completed (loss: 0.1958:  58%|█████▊    | 90/156 [09:04<06:38,  6.04s/it]
Training Epoch: 11/100, batch 179/313 completed (loss: 0.0655:  58%|█████▊    | 90/156 [09:04<06:38,  6.04s/it]
Training Epoch: 11/100, batch 179/313 completed (loss: 0.1642:  58%|█████▊    | 90/156 [09:04<06:38,  6.04s/it]
Training Epoch: 11/100, batch 179/313 completed (loss: 0.0386:  58%|█████▊    | 90/156 [09:04<06:38,  6.04s/it]
Training Epoch: 11/100, batch 180/313 completed (loss: 0.1254:  58%|█████▊    | 90/156 [09:07<06:38,  6.04s/it]
Training Epoch: 11/100, batch 180/313 completed (loss: 0.1104:  58%|█████▊    | 90/156 [09:07<06:38,  6.04s/it]
Training Epoch: 11/100, batch 180/313 completed (loss: 0.5127:  58%|█████▊    | 90/156 [09:07<06:38,  6.04s/it]
Training Epoch: 11/100, batch 180/313 completed (loss: 0.0168:  58%|█████▊    | 90/156 [09:07<06:38,  6.04s/it]
Training Epoch: 11/100, batch 180/313 completed (loss: 0.0168:  58%|█████▊    | 91/156 [09:10<06:35,  6.08s/it]
Training Epoch: 11/100, batch 180/313 completed (loss: 0.1254:  58%|█████▊    | 91/156 [09:10<06:35,  6.09s/it]
Training Epoch: 11/100, batch 180/313 completed (loss: 0.1104:  58%|█████▊    | 91/156 [09:10<06:35,  6.09s/it]
Training Epoch: 11/100, batch 180/313 completed (loss: 0.5127:  58%|█████▊    | 91/156 [09:10<06:35,  6.09s/it]
Training Epoch: 11/100, batch 181/313 completed (loss: 0.013:  58%|█████▊    | 91/156 [09:10<06:35,  6.09s/it] 
Training Epoch: 11/100, batch 181/313 completed (loss: 0.0893:  58%|█████▊    | 91/156 [09:10<06:35,  6.09s/it]
Training Epoch: 11/100, batch 181/313 completed (loss: 0.0149:  58%|█████▊    | 91/156 [09:10<06:35,  6.09s/it]
Training Epoch: 11/100, batch 181/313 completed (loss: 0.0092:  58%|█████▊    | 91/156 [09:10<06:35,  6.08s/it]
Training Epoch: 11/100, batch 182/313 completed (loss: 0.0344:  58%|█████▊    | 91/156 [09:13<06:35,  6.09s/it]
Training Epoch: 11/100, batch 182/313 completed (loss: 0.1265:  58%|█████▊    | 91/156 [09:13<06:35,  6.09s/it]
Training Epoch: 11/100, batch 182/313 completed (loss: 0.0587:  58%|█████▊    | 91/156 [09:13<06:35,  6.09s/it]
Training Epoch: 11/100, batch 182/313 completed (loss: 0.0119:  58%|█████▊    | 91/156 [09:13<06:35,  6.08s/it]
Training Epoch: 11/100, batch 182/313 completed (loss: 0.0344:  59%|█████▉    | 92/156 [09:16<06:30,  6.09s/it]
Training Epoch: 11/100, batch 182/313 completed (loss: 0.0119:  59%|█████▉    | 92/156 [09:16<06:30,  6.10s/it]
Training Epoch: 11/100, batch 182/313 completed (loss: 0.1265:  59%|█████▉    | 92/156 [09:16<06:30,  6.10s/it]
Training Epoch: 11/100, batch 182/313 completed (loss: 0.0587:  59%|█████▉    | 92/156 [09:16<06:30,  6.10s/it]
Training Epoch: 11/100, batch 183/313 completed (loss: 0.0286:  59%|█████▉    | 92/156 [09:16<06:30,  6.10s/it]
Training Epoch: 11/100, batch 183/313 completed (loss: 0.1376:  59%|█████▉    | 92/156 [09:16<06:30,  6.09s/it]
Training Epoch: 11/100, batch 183/313 completed (loss: 0.0622:  59%|█████▉    | 92/156 [09:16<06:30,  6.10s/it]
Training Epoch: 11/100, batch 183/313 completed (loss: 0.0955:  59%|█████▉    | 92/156 [09:16<06:30,  6.10s/it]
Training Epoch: 11/100, batch 184/313 completed (loss: 0.005:  59%|█████▉    | 92/156 [09:19<06:30,  6.10s/it] 
Training Epoch: 11/100, batch 184/313 completed (loss: 0.0037:  59%|█████▉    | 92/156 [09:19<06:30,  6.10s/it]
Training Epoch: 11/100, batch 184/313 completed (loss: 0.3559:  59%|█████▉    | 92/156 [09:19<06:30,  6.09s/it]
Training Epoch: 11/100, batch 184/313 completed (loss: 0.052:  59%|█████▉    | 92/156 [09:19<06:30,  6.10s/it] 
Training Epoch: 11/100, batch 184/313 completed (loss: 0.005:  60%|█████▉    | 93/156 [09:22<06:18,  6.01s/it]
Training Epoch: 11/100, batch 184/313 completed (loss: 0.3559:  60%|█████▉    | 93/156 [09:22<06:18,  6.02s/it]
Training Epoch: 11/100, batch 184/313 completed (loss: 0.052:  60%|█████▉    | 93/156 [09:22<06:18,  6.02s/it]
Training Epoch: 11/100, batch 184/313 completed (loss: 0.0037:  60%|█████▉    | 93/156 [09:22<06:18,  6.01s/it]
Training Epoch: 11/100, batch 185/313 completed (loss: 0.1826:  60%|█████▉    | 93/156 [09:22<06:18,  6.02s/it]
Training Epoch: 11/100, batch 185/313 completed (loss: 0.0816:  60%|█████▉    | 93/156 [09:22<06:18,  6.01s/it]
Training Epoch: 11/100, batch 185/313 completed (loss: 0.1256:  60%|█████▉    | 93/156 [09:22<06:18,  6.01s/it]
Training Epoch: 11/100, batch 185/313 completed (loss: 0.0443:  60%|█████▉    | 93/156 [09:22<06:18,  6.02s/it]
Training Epoch: 11/100, batch 186/313 completed (loss: 0.3292:  60%|█████▉    | 93/156 [09:25<06:18,  6.01s/it]
Training Epoch: 11/100, batch 186/313 completed (loss: 0.0796:  60%|█████▉    | 93/156 [09:25<06:18,  6.02s/it]
Training Epoch: 11/100, batch 186/313 completed (loss: 0.0356:  60%|█████▉    | 93/156 [09:25<06:18,  6.01s/it]
Training Epoch: 11/100, batch 186/313 completed (loss: 0.0905:  60%|█████▉    | 93/156 [09:25<06:18,  6.02s/it]
Training Epoch: 11/100, batch 186/313 completed (loss: 0.0796:  60%|██████    | 94/156 [09:28<06:21,  6.15s/it]
Training Epoch: 11/100, batch 186/313 completed (loss: 0.3292:  60%|██████    | 94/156 [09:28<06:21,  6.15s/it]
Training Epoch: 11/100, batch 186/313 completed (loss: 0.0905:  60%|██████    | 94/156 [09:28<06:21,  6.15s/it]
Training Epoch: 11/100, batch 186/313 completed (loss: 0.0356:  60%|██████    | 94/156 [09:28<06:21,  6.16s/it]
Training Epoch: 11/100, batch 187/313 completed (loss: 0.0934:  60%|██████    | 94/156 [09:28<06:21,  6.16s/it]
Training Epoch: 11/100, batch 187/313 completed (loss: 0.2539:  60%|██████    | 94/156 [09:28<06:21,  6.15s/it]
Training Epoch: 11/100, batch 187/313 completed (loss: 0.0742:  60%|██████    | 94/156 [09:28<06:21,  6.15s/it]
Training Epoch: 11/100, batch 187/313 completed (loss: 0.4747:  60%|██████    | 94/156 [09:28<06:21,  6.15s/it]
Training Epoch: 11/100, batch 188/313 completed (loss: 0.5805:  60%|██████    | 94/156 [09:31<06:21,  6.16s/it]
Training Epoch: 11/100, batch 188/313 completed (loss: 0.1052:  60%|██████    | 94/156 [09:31<06:21,  6.15s/it]
Training Epoch: 11/100, batch 188/313 completed (loss: 0.0674:  60%|██████    | 94/156 [09:31<06:21,  6.15s/it]
Training Epoch: 11/100, batch 188/313 completed (loss: 0.0183:  60%|██████    | 94/156 [09:31<06:21,  6.15s/it]
Training Epoch: 11/100, batch 188/313 completed (loss: 0.0183:  61%|██████    | 95/156 [09:34<06:10,  6.07s/it]
Training Epoch: 11/100, batch 188/313 completed (loss: 0.0674:  61%|██████    | 95/156 [09:34<06:10,  6.07s/it]
Training Epoch: 11/100, batch 188/313 completed (loss: 0.5805:  61%|██████    | 95/156 [09:34<06:10,  6.07s/it]
Training Epoch: 11/100, batch 188/313 completed (loss: 0.1052:  61%|██████    | 95/156 [09:34<06:10,  6.08s/it]
Training Epoch: 11/100, batch 189/313 completed (loss: 0.2068:  61%|██████    | 95/156 [09:34<06:10,  6.07s/it]
Training Epoch: 11/100, batch 189/313 completed (loss: 0.1149:  61%|██████    | 95/156 [09:34<06:10,  6.07s/it]
Training Epoch: 11/100, batch 189/313 completed (loss: 0.0619:  61%|██████    | 95/156 [09:34<06:10,  6.08s/it]
Training Epoch: 11/100, batch 189/313 completed (loss: 0.0129:  61%|██████    | 95/156 [09:34<06:10,  6.07s/it]
Training Epoch: 11/100, batch 190/313 completed (loss: 0.1772:  61%|██████    | 95/156 [09:37<06:10,  6.07s/it]
Training Epoch: 11/100, batch 190/313 completed (loss: 0.0418:  61%|██████    | 95/156 [09:37<06:10,  6.07s/it]
Training Epoch: 11/100, batch 190/313 completed (loss: 0.0461:  61%|██████    | 95/156 [09:37<06:10,  6.08s/it]
Training Epoch: 11/100, batch 190/313 completed (loss: 0.0149:  61%|██████    | 95/156 [09:37<06:10,  6.07s/it]
Training Epoch: 11/100, batch 190/313 completed (loss: 0.0461:  62%|██████▏   | 96/156 [09:40<06:03,  6.05s/it]
Training Epoch: 11/100, batch 190/313 completed (loss: 0.0149:  62%|██████▏   | 96/156 [09:40<06:03,  6.06s/it]
Training Epoch: 11/100, batch 190/313 completed (loss: 0.0418:  62%|██████▏   | 96/156 [09:40<06:03,  6.06s/it]
Training Epoch: 11/100, batch 190/313 completed (loss: 0.1772:  62%|██████▏   | 96/156 [09:40<06:03,  6.06s/it]
Training Epoch: 11/100, batch 191/313 completed (loss: 0.3222:  62%|██████▏   | 96/156 [09:40<06:03,  6.06s/it]
Training Epoch: 11/100, batch 191/313 completed (loss: 0.0241:  62%|██████▏   | 96/156 [09:40<06:03,  6.06s/it]
Training Epoch: 11/100, batch 191/313 completed (loss: 0.0323:  62%|██████▏   | 96/156 [09:40<06:03,  6.05s/it]
Training Epoch: 11/100, batch 191/313 completed (loss: 0.176:  62%|██████▏   | 96/156 [09:40<06:03,  6.06s/it] 
Training Epoch: 11/100, batch 192/313 completed (loss: 0.1379:  62%|██████▏   | 96/156 [09:44<06:03,  6.06s/it]
Training Epoch: 11/100, batch 192/313 completed (loss: 0.1776:  62%|██████▏   | 96/156 [09:44<06:03,  6.06s/it]
Training Epoch: 11/100, batch 192/313 completed (loss: 0.0587:  62%|██████▏   | 96/156 [09:44<06:03,  6.05s/it]
Training Epoch: 11/100, batch 192/313 completed (loss: 0.0317:  62%|██████▏   | 96/156 [09:44<06:03,  6.06s/it]
Training Epoch: 11/100, batch 192/313 completed (loss: 0.0317:  62%|██████▏   | 97/156 [09:46<06:01,  6.13s/it]
Training Epoch: 11/100, batch 192/313 completed (loss: 0.0587:  62%|██████▏   | 97/156 [09:46<06:01,  6.13s/it]
Training Epoch: 11/100, batch 192/313 completed (loss: 0.1379:  62%|██████▏   | 97/156 [09:46<06:01,  6.13s/it]
Training Epoch: 11/100, batch 192/313 completed (loss: 0.1776:  62%|██████▏   | 97/156 [09:46<06:01,  6.13s/it]
Training Epoch: 11/100, batch 193/313 completed (loss: 0.0491:  62%|██████▏   | 97/156 [09:46<06:01,  6.13s/it]
Training Epoch: 11/100, batch 193/313 completed (loss: 0.2603:  62%|██████▏   | 97/156 [09:46<06:01,  6.13s/it]
Training Epoch: 11/100, batch 193/313 completed (loss: 0.0267:  62%|██████▏   | 97/156 [09:46<06:01,  6.13s/it]
Training Epoch: 11/100, batch 193/313 completed (loss: 0.0427:  62%|██████▏   | 97/156 [09:47<06:01,  6.13s/it]
Training Epoch: 11/100, batch 194/313 completed (loss: 0.194:  62%|██████▏   | 97/156 [09:49<06:01,  6.13s/it] 
Training Epoch: 11/100, batch 194/313 completed (loss: 0.4204:  62%|██████▏   | 97/156 [09:49<06:01,  6.13s/it]
Training Epoch: 11/100, batch 194/313 completed (loss: 0.0489:  62%|██████▏   | 97/156 [09:49<06:01,  6.13s/it]
Training Epoch: 11/100, batch 194/313 completed (loss: 0.2722:  62%|██████▏   | 97/156 [09:49<06:01,  6.13s/it]
Training Epoch: 11/100, batch 194/313 completed (loss: 0.194:  63%|██████▎   | 98/156 [09:52<05:50,  6.05s/it]
Training Epoch: 11/100, batch 194/313 completed (loss: 0.2722:  63%|██████▎   | 98/156 [09:52<05:50,  6.05s/it]
Training Epoch: 11/100, batch 194/313 completed (loss: 0.4204:  63%|██████▎   | 98/156 [09:52<05:50,  6.05s/it]
Training Epoch: 11/100, batch 194/313 completed (loss: 0.0489:  63%|██████▎   | 98/156 [09:52<05:50,  6.05s/it]
Training Epoch: 11/100, batch 195/313 completed (loss: 0.0861:  63%|██████▎   | 98/156 [09:52<05:50,  6.05s/it]
Training Epoch: 11/100, batch 195/313 completed (loss: 0.0857:  63%|██████▎   | 98/156 [09:52<05:50,  6.05s/it]
Training Epoch: 11/100, batch 195/313 completed (loss: 0.1363:  63%|██████▎   | 98/156 [09:52<05:50,  6.05s/it]
Training Epoch: 11/100, batch 195/313 completed (loss: 0.2327:  63%|██████▎   | 98/156 [09:52<05:50,  6.05s/it]
Training Epoch: 11/100, batch 196/313 completed (loss: 0.2356:  63%|██████▎   | 98/156 [09:55<05:50,  6.05s/it]
Training Epoch: 11/100, batch 196/313 completed (loss: 0.0655:  63%|██████▎   | 98/156 [09:55<05:50,  6.05s/it]
Training Epoch: 11/100, batch 196/313 completed (loss: 0.0627:  63%|██████▎   | 98/156 [09:55<05:50,  6.05s/it]
Training Epoch: 11/100, batch 196/313 completed (loss: 0.232:  63%|██████▎   | 98/156 [09:55<05:50,  6.05s/it] 
Training Epoch: 11/100, batch 196/313 completed (loss: 0.0627:  63%|██████▎   | 99/156 [09:59<05:52,  6.18s/it]
Training Epoch: 11/100, batch 196/313 completed (loss: 0.232:  63%|██████▎   | 99/156 [09:59<05:52,  6.18s/it]
Training Epoch: 11/100, batch 196/313 completed (loss: 0.0655:  63%|██████▎   | 99/156 [09:59<05:52,  6.18s/it]
Training Epoch: 11/100, batch 196/313 completed (loss: 0.2356:  63%|██████▎   | 99/156 [09:59<05:52,  6.18s/it]
Training Epoch: 11/100, batch 197/313 completed (loss: 0.3219:  63%|██████▎   | 99/156 [09:59<05:52,  6.18s/it]
Training Epoch: 11/100, batch 197/313 completed (loss: 0.1366:  63%|██████▎   | 99/156 [09:59<05:52,  6.18s/it]
Training Epoch: 11/100, batch 197/313 completed (loss: 0.2444:  63%|██████▎   | 99/156 [09:59<05:52,  6.18s/it]
Training Epoch: 11/100, batch 197/313 completed (loss: 0.0322:  63%|██████▎   | 99/156 [09:59<05:52,  6.18s/it]
Training Epoch: 11/100, batch 198/313 completed (loss: 0.0272:  63%|██████▎   | 99/156 [10:02<05:52,  6.18s/it]
Training Epoch: 11/100, batch 198/313 completed (loss: 0.0585:  63%|██████▎   | 99/156 [10:02<05:52,  6.18s/it]
Training Epoch: 11/100, batch 198/313 completed (loss: 0.1197:  63%|██████▎   | 99/156 [10:02<05:52,  6.18s/it]
Training Epoch: 11/100, batch 198/313 completed (loss: 0.1747:  63%|██████▎   | 99/156 [10:02<05:52,  6.18s/it]
Training Epoch: 11/100, batch 198/313 completed (loss: 0.1197:  64%|██████▍   | 100/156 [10:05<05:40,  6.08s/it]
Training Epoch: 11/100, batch 198/313 completed (loss: 0.0585:  64%|██████▍   | 100/156 [10:05<05:40,  6.08s/it]
Training Epoch: 11/100, batch 198/313 completed (loss: 0.0272:  64%|██████▍   | 100/156 [10:05<05:40,  6.08s/it]
Training Epoch: 11/100, batch 198/313 completed (loss: 0.1747:  64%|██████▍   | 100/156 [10:05<05:40,  6.08s/it]
Training Epoch: 11/100, batch 199/313 completed (loss: 0.0218:  64%|██████▍   | 100/156 [10:05<05:40,  6.08s/it]
Training Epoch: 11/100, batch 199/313 completed (loss: 0.4074:  64%|██████▍   | 100/156 [10:05<05:40,  6.08s/it]
Training Epoch: 11/100, batch 199/313 completed (loss: 0.2503:  64%|██████▍   | 100/156 [10:05<05:40,  6.08s/it]
Training Epoch: 11/100, batch 199/313 completed (loss: 0.0197:  64%|██████▍   | 100/156 [10:05<05:40,  6.08s/it]
Training Epoch: 11/100, batch 200/313 completed (loss: 0.1328:  64%|██████▍   | 100/156 [10:08<05:40,  6.08s/it]
Training Epoch: 11/100, batch 200/313 completed (loss: 0.4055:  64%|██████▍   | 100/156 [10:08<05:40,  6.08s/it]
Training Epoch: 11/100, batch 200/313 completed (loss: 0.0488:  64%|██████▍   | 100/156 [10:08<05:40,  6.08s/it]
Training Epoch: 11/100, batch 200/313 completed (loss: 0.0581:  64%|██████▍   | 100/156 [10:08<05:40,  6.08s/it]
Training Epoch: 11/100, batch 200/313 completed (loss: 0.0488:  65%|██████▍   | 101/156 [10:11<05:30,  6.02s/it]
Training Epoch: 11/100, batch 200/313 completed (loss: 0.1328:  65%|██████▍   | 101/156 [10:11<05:30,  6.02s/it]
Training Epoch: 11/100, batch 200/313 completed (loss: 0.0581:  65%|██████▍   | 101/156 [10:11<05:30,  6.02s/it]
Training Epoch: 11/100, batch 200/313 completed (loss: 0.4055:  65%|██████▍   | 101/156 [10:11<05:31,  6.02s/it]
Training Epoch: 11/100, batch 201/313 completed (loss: 0.0289:  65%|██████▍   | 101/156 [10:11<05:30,  6.02s/it]
Training Epoch: 11/100, batch 201/313 completed (loss: 0.0284:  65%|██████▍   | 101/156 [10:11<05:30,  6.02s/it]
Training Epoch: 11/100, batch 201/313 completed (loss: 0.6036:  65%|██████▍   | 101/156 [10:11<05:31,  6.02s/it]
Training Epoch: 11/100, batch 201/313 completed (loss: 0.1562:  65%|██████▍   | 101/156 [10:11<05:30,  6.02s/it]
Training Epoch: 11/100, batch 202/313 completed (loss: 0.4159:  65%|██████▍   | 101/156 [10:14<05:31,  6.02s/it]
Training Epoch: 11/100, batch 202/313 completed (loss: 0.0513:  65%|██████▍   | 101/156 [10:14<05:30,  6.02s/it]
Training Epoch: 11/100, batch 202/313 completed (loss: 0.047:  65%|██████▍   | 101/156 [10:14<05:30,  6.02s/it] 
Training Epoch: 11/100, batch 202/313 completed (loss: 0.1231:  65%|██████▍   | 101/156 [10:14<05:30,  6.02s/it]
Training Epoch: 11/100, batch 202/313 completed (loss: 0.1231:  65%|██████▌   | 102/156 [10:17<05:30,  6.12s/it]
Training Epoch: 11/100, batch 202/313 completed (loss: 0.0513:  65%|██████▌   | 102/156 [10:17<05:30,  6.13s/it]
Training Epoch: 11/100, batch 202/313 completed (loss: 0.047:  65%|██████▌   | 102/156 [10:17<05:30,  6.13s/it]
Training Epoch: 11/100, batch 202/313 completed (loss: 0.4159:  65%|██████▌   | 102/156 [10:17<05:30,  6.13s/it]
Training Epoch: 11/100, batch 203/313 completed (loss: 0.0484:  65%|██████▌   | 102/156 [10:17<05:30,  6.13s/it]
Training Epoch: 11/100, batch 203/313 completed (loss: 0.2822:  65%|██████▌   | 102/156 [10:17<05:30,  6.13s/it]
Training Epoch: 11/100, batch 203/313 completed (loss: 0.1389:  65%|██████▌   | 102/156 [10:17<05:30,  6.13s/it]
Training Epoch: 11/100, batch 203/313 completed (loss: 0.0463:  65%|██████▌   | 102/156 [10:17<05:30,  6.12s/it]
Training Epoch: 11/100, batch 204/313 completed (loss: 0.3787:  65%|██████▌   | 102/156 [10:20<05:30,  6.13s/it]
Training Epoch: 11/100, batch 204/313 completed (loss: 0.2492:  65%|██████▌   | 102/156 [10:20<05:30,  6.13s/it]
Training Epoch: 11/100, batch 204/313 completed (loss: 0.0851:  65%|██████▌   | 102/156 [10:20<05:30,  6.13s/it]
Training Epoch: 11/100, batch 204/313 completed (loss: 0.0089:  65%|██████▌   | 102/156 [10:20<05:30,  6.12s/it]
Training Epoch: 11/100, batch 204/313 completed (loss: 0.3787:  66%|██████▌   | 103/156 [10:23<05:19,  6.03s/it]
Training Epoch: 11/100, batch 204/313 completed (loss: 0.0089:  66%|██████▌   | 103/156 [10:23<05:19,  6.04s/it]
Training Epoch: 11/100, batch 204/313 completed (loss: 0.0851:  66%|██████▌   | 103/156 [10:23<05:19,  6.04s/it]
Training Epoch: 11/100, batch 204/313 completed (loss: 0.2492:  66%|██████▌   | 103/156 [10:23<05:20,  6.04s/it]
Training Epoch: 11/100, batch 205/313 completed (loss: 0.0547:  66%|██████▌   | 103/156 [10:23<05:20,  6.04s/it]
Training Epoch: 11/100, batch 205/313 completed (loss: 0.0993:  66%|██████▌   | 103/156 [10:23<05:19,  6.04s/it]
Training Epoch: 11/100, batch 205/313 completed (loss: 0.1208:  66%|██████▌   | 103/156 [10:23<05:19,  6.03s/it]
Training Epoch: 11/100, batch 205/313 completed (loss: 0.1627:  66%|██████▌   | 103/156 [10:23<05:19,  6.04s/it]
Training Epoch: 11/100, batch 206/313 completed (loss: 0.1419:  66%|██████▌   | 103/156 [10:26<05:19,  6.03s/it]
Training Epoch: 11/100, batch 206/313 completed (loss: 0.0158:  66%|██████▌   | 103/156 [10:26<05:20,  6.04s/it]
Training Epoch: 11/100, batch 206/313 completed (loss: 0.2626:  66%|██████▌   | 103/156 [10:26<05:19,  6.04s/it]
Training Epoch: 11/100, batch 206/313 completed (loss: 0.1731:  66%|██████▌   | 103/156 [10:26<05:19,  6.04s/it]
Training Epoch: 11/100, batch 206/313 completed (loss: 0.1731:  67%|██████▋   | 104/156 [10:29<05:14,  6.05s/it]
Training Epoch: 11/100, batch 206/313 completed (loss: 0.1419:  67%|██████▋   | 104/156 [10:29<05:14,  6.06s/it]
Training Epoch: 11/100, batch 206/313 completed (loss: 0.0158:  67%|██████▋   | 104/156 [10:29<05:14,  6.05s/it]
Training Epoch: 11/100, batch 206/313 completed (loss: 0.2626:  67%|██████▋   | 104/156 [10:29<05:14,  6.06s/it]
Training Epoch: 11/100, batch 207/313 completed (loss: 0.2585:  67%|██████▋   | 104/156 [10:29<05:14,  6.06s/it]
Training Epoch: 11/100, batch 207/313 completed (loss: 0.1625:  67%|██████▋   | 104/156 [10:29<05:14,  6.06s/it]
Training Epoch: 11/100, batch 207/313 completed (loss: 0.1098:  67%|██████▋   | 104/156 [10:29<05:14,  6.05s/it]
Training Epoch: 11/100, batch 207/313 completed (loss: 0.586:  67%|██████▋   | 104/156 [10:29<05:14,  6.05s/it] 
Training Epoch: 11/100, batch 208/313 completed (loss: 0.2202:  67%|██████▋   | 104/156 [10:32<05:14,  6.06s/it]
Training Epoch: 11/100, batch 208/313 completed (loss: 0.0345:  67%|██████▋   | 104/156 [10:32<05:14,  6.06s/it]
Training Epoch: 11/100, batch 208/313 completed (loss: 0.2652:  67%|██████▋   | 104/156 [10:32<05:14,  6.05s/it]
Training Epoch: 11/100, batch 208/313 completed (loss: 0.1184:  67%|██████▋   | 104/156 [10:32<05:14,  6.05s/it]
Training Epoch: 11/100, batch 208/313 completed (loss: 0.0345:  67%|██████▋   | 105/156 [10:35<05:08,  6.05s/it]
Training Epoch: 11/100, batch 208/313 completed (loss: 0.2202:  67%|██████▋   | 105/156 [10:35<05:08,  6.05s/it]
Training Epoch: 11/100, batch 208/313 completed (loss: 0.2652:  67%|██████▋   | 105/156 [10:35<05:08,  6.05s/it]
Training Epoch: 11/100, batch 208/313 completed (loss: 0.1184:  67%|██████▋   | 105/156 [10:35<05:08,  6.05s/it]
Training Epoch: 11/100, batch 209/313 completed (loss: 0.0977:  67%|██████▋   | 105/156 [10:35<05:08,  6.05s/it]
Training Epoch: 11/100, batch 209/313 completed (loss: 0.1067:  67%|██████▋   | 105/156 [10:35<05:08,  6.05s/it]
Training Epoch: 11/100, batch 209/313 completed (loss: 0.3982:  67%|██████▋   | 105/156 [10:35<05:08,  6.05s/it]
Training Epoch: 11/100, batch 209/313 completed (loss: 0.0988:  67%|██████▋   | 105/156 [10:35<05:08,  6.05s/it]
Training Epoch: 11/100, batch 210/313 completed (loss: 0.2361:  67%|██████▋   | 105/156 [10:38<05:08,  6.05s/it]
Training Epoch: 11/100, batch 210/313 completed (loss: 0.0816:  67%|██████▋   | 105/156 [10:38<05:08,  6.05s/it]
Training Epoch: 11/100, batch 210/313 completed (loss: 0.2424:  67%|██████▋   | 105/156 [10:38<05:08,  6.05s/it]
Training Epoch: 11/100, batch 210/313 completed (loss: 0.0455:  67%|██████▋   | 105/156 [10:38<05:08,  6.05s/it]
Training Epoch: 11/100, batch 210/313 completed (loss: 0.2361:  68%|██████▊   | 106/156 [10:41<04:59,  5.98s/it]
Training Epoch: 11/100, batch 210/313 completed (loss: 0.0816:  68%|██████▊   | 106/156 [10:41<04:59,  5.98s/it]
Training Epoch: 11/100, batch 210/313 completed (loss: 0.2424:  68%|██████▊   | 106/156 [10:41<04:59,  5.98s/it]
Training Epoch: 11/100, batch 210/313 completed (loss: 0.0455:  68%|██████▊   | 106/156 [10:41<04:59,  5.99s/it]
Training Epoch: 11/100, batch 211/313 completed (loss: 0.0079:  68%|██████▊   | 106/156 [10:41<04:59,  5.98s/it]
Training Epoch: 11/100, batch 211/313 completed (loss: 0.5563:  68%|██████▊   | 106/156 [10:41<04:59,  5.98s/it]
Training Epoch: 11/100, batch 211/313 completed (loss: 0.3759:  68%|██████▊   | 106/156 [10:41<04:59,  5.98s/it]
Training Epoch: 11/100, batch 211/313 completed (loss: 0.1261:  68%|██████▊   | 106/156 [10:41<04:59,  5.99s/it]
Training Epoch: 11/100, batch 212/313 completed (loss: 0.2431:  68%|██████▊   | 106/156 [10:44<04:59,  5.98s/it]
Training Epoch: 11/100, batch 212/313 completed (loss: 0.0557:  68%|██████▊   | 106/156 [10:44<04:59,  5.98s/it]
Training Epoch: 11/100, batch 212/313 completed (loss: 0.1122:  68%|██████▊   | 106/156 [10:44<04:59,  5.98s/it]
Training Epoch: 11/100, batch 212/313 completed (loss: 0.1181:  68%|██████▊   | 106/156 [10:44<04:59,  5.99s/it]
Training Epoch: 11/100, batch 212/313 completed (loss: 0.0557:  69%|██████▊   | 107/156 [10:47<04:57,  6.07s/it]
Training Epoch: 11/100, batch 212/313 completed (loss: 0.1122:  69%|██████▊   | 107/156 [10:47<04:57,  6.07s/it]
Training Epoch: 11/100, batch 212/313 completed (loss: 0.2431:  69%|██████▊   | 107/156 [10:47<04:57,  6.07s/it]
Training Epoch: 11/100, batch 212/313 completed (loss: 0.1181:  69%|██████▊   | 107/156 [10:47<04:57,  6.08s/it]
Training Epoch: 11/100, batch 213/313 completed (loss: 0.2132:  69%|██████▊   | 107/156 [10:47<04:57,  6.07s/it]
Training Epoch: 11/100, batch 213/313 completed (loss: 0.3569:  69%|██████▊   | 107/156 [10:47<04:57,  6.07s/it]
Training Epoch: 11/100, batch 213/313 completed (loss: 0.0686:  69%|██████▊   | 107/156 [10:47<04:57,  6.07s/it]
Training Epoch: 11/100, batch 213/313 completed (loss: 0.0533:  69%|██████▊   | 107/156 [10:47<04:57,  6.08s/it]
Training Epoch: 11/100, batch 214/313 completed (loss: 0.1016:  69%|██████▊   | 107/156 [10:50<04:57,  6.07s/it]
Training Epoch: 11/100, batch 214/313 completed (loss: 0.3654:  69%|██████▊   | 107/156 [10:50<04:57,  6.07s/it]
Training Epoch: 11/100, batch 214/313 completed (loss: 0.7341:  69%|██████▊   | 107/156 [10:50<04:57,  6.07s/it]
Training Epoch: 11/100, batch 214/313 completed (loss: 0.1012:  69%|██████▊   | 107/156 [10:50<04:57,  6.08s/it]
Training Epoch: 11/100, batch 214/313 completed (loss: 0.1012:  69%|██████▉   | 108/156 [10:53<04:48,  6.01s/it]
Training Epoch: 11/100, batch 214/313 completed (loss: 0.1016:  69%|██████▉   | 108/156 [10:53<04:48,  6.01s/it]
Training Epoch: 11/100, batch 214/313 completed (loss: 0.7341:  69%|██████▉   | 108/156 [10:53<04:48,  6.01s/it]
Training Epoch: 11/100, batch 214/313 completed (loss: 0.3654:  69%|██████▉   | 108/156 [10:53<04:48,  6.02s/it]
Training Epoch: 11/100, batch 215/313 completed (loss: 0.024:  69%|██████▉   | 108/156 [10:53<04:48,  6.02s/it] 
Training Epoch: 11/100, batch 215/313 completed (loss: 0.1956:  69%|██████▉   | 108/156 [10:53<04:48,  6.01s/it]
Training Epoch: 11/100, batch 215/313 completed (loss: 0.0373:  69%|██████▉   | 108/156 [10:53<04:48,  6.01s/it]
Training Epoch: 11/100, batch 215/313 completed (loss: 0.0921:  69%|██████▉   | 108/156 [10:53<04:48,  6.01s/it]
Training Epoch: 11/100, batch 216/313 completed (loss: 0.2796:  69%|██████▉   | 108/156 [10:56<04:48,  6.01s/it]
Training Epoch: 11/100, batch 216/313 completed (loss: 0.0391:  69%|██████▉   | 108/156 [10:56<04:48,  6.01s/it]
Training Epoch: 11/100, batch 216/313 completed (loss: 0.2611:  69%|██████▉   | 108/156 [10:56<04:48,  6.02s/it]
Training Epoch: 11/100, batch 216/313 completed (loss: 0.5456:  69%|██████▉   | 108/156 [10:56<04:48,  6.01s/it]
Training Epoch: 11/100, batch 216/313 completed (loss: 0.0391:  70%|██████▉   | 109/156 [10:59<04:40,  5.96s/it]
Training Epoch: 11/100, batch 216/313 completed (loss: 0.5456:  70%|██████▉   | 109/156 [10:59<04:40,  5.96s/it]
Training Epoch: 11/100, batch 216/313 completed (loss: 0.2611:  70%|██████▉   | 109/156 [10:59<04:40,  5.97s/it]
Training Epoch: 11/100, batch 216/313 completed (loss: 0.2796:  70%|██████▉   | 109/156 [10:59<04:40,  5.97s/it]
Training Epoch: 11/100, batch 217/313 completed (loss: 0.0551:  70%|██████▉   | 109/156 [10:59<04:40,  5.97s/it]
Training Epoch: 11/100, batch 217/313 completed (loss: 0.3576:  70%|██████▉   | 109/156 [10:59<04:40,  5.97s/it]
Training Epoch: 11/100, batch 217/313 completed (loss: 0.0803:  70%|██████▉   | 109/156 [10:59<04:40,  5.96s/it]
Training Epoch: 11/100, batch 217/313 completed (loss: 0.1932:  70%|██████▉   | 109/156 [10:59<04:40,  5.96s/it]
Training Epoch: 11/100, batch 218/313 completed (loss: 0.2625:  70%|██████▉   | 109/156 [11:02<04:40,  5.97s/it]
Training Epoch: 11/100, batch 218/313 completed (loss: 0.5656:  70%|██████▉   | 109/156 [11:02<04:40,  5.97s/it]
Training Epoch: 11/100, batch 218/313 completed (loss: 0.2502:  70%|██████▉   | 109/156 [11:02<04:40,  5.96s/it]
Training Epoch: 11/100, batch 218/313 completed (loss: 0.0224:  70%|██████▉   | 109/156 [11:02<04:40,  5.96s/it]
Training Epoch: 11/100, batch 218/313 completed (loss: 0.0224:  71%|███████   | 110/156 [11:05<04:38,  6.05s/it]
Training Epoch: 11/100, batch 218/313 completed (loss: 0.2502:  71%|███████   | 110/156 [11:05<04:38,  6.05s/it]
Training Epoch: 11/100, batch 218/313 completed (loss: 0.5656:  71%|███████   | 110/156 [11:05<04:38,  6.05s/it]
Training Epoch: 11/100, batch 218/313 completed (loss: 0.2625:  71%|███████   | 110/156 [11:05<04:38,  6.05s/it]
Training Epoch: 11/100, batch 219/313 completed (loss: 0.5068:  71%|███████   | 110/156 [11:05<04:38,  6.05s/it]
Training Epoch: 11/100, batch 219/313 completed (loss: 0.1243:  71%|███████   | 110/156 [11:05<04:38,  6.05s/it]
Training Epoch: 11/100, batch 219/313 completed (loss: 0.0112:  71%|███████   | 110/156 [11:05<04:38,  6.05s/it]
Training Epoch: 11/100, batch 219/313 completed (loss: 0.103:  71%|███████   | 110/156 [11:05<04:38,  6.05s/it] 
Training Epoch: 11/100, batch 220/313 completed (loss: 0.009:  71%|███████   | 110/156 [11:08<04:38,  6.05s/it] 
Training Epoch: 11/100, batch 220/313 completed (loss: 0.4303:  71%|███████   | 110/156 [11:08<04:38,  6.05s/it]
Training Epoch: 11/100, batch 220/313 completed (loss: 0.1543:  71%|███████   | 110/156 [11:08<04:38,  6.05s/it]
Training Epoch: 11/100, batch 220/313 completed (loss: 0.1666:  71%|███████   | 110/156 [11:08<04:38,  6.05s/it]
Training Epoch: 11/100, batch 220/313 completed (loss: 0.009:  71%|███████   | 111/156 [11:11<04:29,  5.99s/it]
Training Epoch: 11/100, batch 220/313 completed (loss: 0.1543:  71%|███████   | 111/156 [11:11<04:29,  5.99s/it]
Training Epoch: 11/100, batch 220/313 completed (loss: 0.4303:  71%|███████   | 111/156 [11:11<04:29,  5.99s/it]
Training Epoch: 11/100, batch 220/313 completed (loss: 0.1666:  71%|███████   | 111/156 [11:11<04:29,  5.99s/it]
Training Epoch: 11/100, batch 221/313 completed (loss: 0.9362:  71%|███████   | 111/156 [11:11<04:29,  5.99s/it]
Training Epoch: 11/100, batch 221/313 completed (loss: 0.1757:  71%|███████   | 111/156 [11:11<04:29,  5.99s/it]
Training Epoch: 11/100, batch 221/313 completed (loss: 0.0291:  71%|███████   | 111/156 [11:11<04:29,  5.99s/it]
Training Epoch: 11/100, batch 221/313 completed (loss: 0.3121:  71%|███████   | 111/156 [11:11<04:29,  5.99s/it]
Training Epoch: 11/100, batch 222/313 completed (loss: 0.0189:  71%|███████   | 111/156 [11:14<04:29,  5.99s/it]
Training Epoch: 11/100, batch 222/313 completed (loss: 0.1121:  71%|███████   | 111/156 [11:14<04:29,  5.99s/it]
Training Epoch: 11/100, batch 222/313 completed (loss: 0.0227:  71%|███████   | 111/156 [11:14<04:29,  5.99s/it]
Training Epoch: 11/100, batch 222/313 completed (loss: 0.1634:  71%|███████   | 111/156 [11:14<04:29,  5.99s/it]
Training Epoch: 11/100, batch 222/313 completed (loss: 0.1634:  72%|███████▏  | 112/156 [11:17<04:26,  6.06s/it]
Training Epoch: 11/100, batch 222/313 completed (loss: 0.0189:  72%|███████▏  | 112/156 [11:17<04:26,  6.06s/it]
Training Epoch: 11/100, batch 222/313 completed (loss: 0.0227:  72%|███████▏  | 112/156 [11:17<04:26,  6.06s/it]
Training Epoch: 11/100, batch 222/313 completed (loss: 0.1121:  72%|███████▏  | 112/156 [11:17<04:26,  6.06s/it]
Training Epoch: 11/100, batch 223/313 completed (loss: 0.1373:  72%|███████▏  | 112/156 [11:17<04:26,  6.06s/it]
Training Epoch: 11/100, batch 223/313 completed (loss: 0.0456:  72%|███████▏  | 112/156 [11:17<04:26,  6.06s/it]
Training Epoch: 11/100, batch 223/313 completed (loss: 0.0545:  72%|███████▏  | 112/156 [11:17<04:26,  6.06s/it]
Training Epoch: 11/100, batch 223/313 completed (loss: 0.0452:  72%|███████▏  | 112/156 [11:17<04:26,  6.06s/it]
Training Epoch: 11/100, batch 224/313 completed (loss: 0.1038:  72%|███████▏  | 112/156 [11:20<04:26,  6.06s/it]
Training Epoch: 11/100, batch 224/313 completed (loss: 0.0195:  72%|███████▏  | 112/156 [11:20<04:26,  6.06s/it]
Training Epoch: 11/100, batch 224/313 completed (loss: 0.2663:  72%|███████▏  | 112/156 [11:20<04:26,  6.06s/it]
Training Epoch: 11/100, batch 224/313 completed (loss: 0.1509:  72%|███████▏  | 112/156 [11:20<04:26,  6.06s/it]
Training Epoch: 11/100, batch 224/313 completed (loss: 0.1509:  72%|███████▏  | 113/156 [11:23<04:20,  6.06s/it]
Training Epoch: 11/100, batch 224/313 completed (loss: 0.1038:  72%|███████▏  | 113/156 [11:23<04:20,  6.06s/it]
Training Epoch: 11/100, batch 224/313 completed (loss: 0.0195:  72%|███████▏  | 113/156 [11:23<04:20,  6.06s/it]
Training Epoch: 11/100, batch 224/313 completed (loss: 0.2663:  72%|███████▏  | 113/156 [11:23<04:20,  6.06s/it]
Training Epoch: 11/100, batch 225/313 completed (loss: 0.0117:  72%|███████▏  | 113/156 [11:23<04:20,  6.06s/it]
Training Epoch: 11/100, batch 225/313 completed (loss: 0.0626:  72%|███████▏  | 113/156 [11:23<04:20,  6.06s/it]
Training Epoch: 11/100, batch 225/313 completed (loss: 0.0304:  72%|███████▏  | 113/156 [11:23<04:20,  6.06s/it]
Training Epoch: 11/100, batch 225/313 completed (loss: 0.0567:  72%|███████▏  | 113/156 [11:23<04:20,  6.06s/it]
Training Epoch: 11/100, batch 226/313 completed (loss: 0.0573:  72%|███████▏  | 113/156 [11:26<04:20,  6.06s/it]
Training Epoch: 11/100, batch 226/313 completed (loss: 0.5784:  72%|███████▏  | 113/156 [11:26<04:20,  6.06s/it]
Training Epoch: 11/100, batch 226/313 completed (loss: 0.0613:  72%|███████▏  | 113/156 [11:26<04:20,  6.06s/it]
Training Epoch: 11/100, batch 226/313 completed (loss: 0.5876:  72%|███████▏  | 113/156 [11:26<04:20,  6.06s/it]
Training Epoch: 11/100, batch 226/313 completed (loss: 0.0573:  73%|███████▎  | 114/156 [11:29<04:11,  5.99s/it]
Training Epoch: 11/100, batch 226/313 completed (loss: 0.5784:  73%|███████▎  | 114/156 [11:29<04:11,  5.99s/it]
Training Epoch: 11/100, batch 226/313 completed (loss: 0.0613:  73%|███████▎  | 114/156 [11:29<04:11,  5.99s/it]
Training Epoch: 11/100, batch 226/313 completed (loss: 0.5876:  73%|███████▎  | 114/156 [11:29<04:11,  5.99s/it]
Training Epoch: 11/100, batch 227/313 completed (loss: 0.0883:  73%|███████▎  | 114/156 [11:29<04:11,  5.99s/it]
Training Epoch: 11/100, batch 227/313 completed (loss: 0.0258:  73%|███████▎  | 114/156 [11:29<04:11,  5.99s/it]
Training Epoch: 11/100, batch 227/313 completed (loss: 0.1683:  73%|███████▎  | 114/156 [11:29<04:11,  5.99s/it]
Training Epoch: 11/100, batch 227/313 completed (loss: 0.2562:  73%|███████▎  | 114/156 [11:29<04:11,  5.99s/it]
Training Epoch: 11/100, batch 228/313 completed (loss: 0.232:  73%|███████▎  | 114/156 [11:32<04:11,  5.99s/it] 
Training Epoch: 11/100, batch 228/313 completed (loss: 0.1861:  73%|███████▎  | 114/156 [11:32<04:11,  5.99s/it]
Training Epoch: 11/100, batch 228/313 completed (loss: 0.2047:  73%|███████▎  | 114/156 [11:32<04:11,  5.99s/it]
Training Epoch: 11/100, batch 228/313 completed (loss: 0.1229:  73%|███████▎  | 114/156 [11:32<04:11,  5.99s/it]
Training Epoch: 11/100, batch 228/313 completed (loss: 0.1229:  74%|███████▎  | 115/156 [11:35<04:11,  6.13s/it]
Training Epoch: 11/100, batch 228/313 completed (loss: 0.232:  74%|███████▎  | 115/156 [11:35<04:11,  6.13s/it]
Training Epoch: 11/100, batch 228/313 completed (loss: 0.1861:  74%|███████▎  | 115/156 [11:35<04:11,  6.13s/it]
Training Epoch: 11/100, batch 228/313 completed (loss: 0.2047:  74%|███████▎  | 115/156 [11:35<04:11,  6.14s/it]
Training Epoch: 11/100, batch 229/313 completed (loss: 0.0148:  74%|███████▎  | 115/156 [11:35<04:11,  6.13s/it]
Training Epoch: 11/100, batch 229/313 completed (loss: 0.1316:  74%|███████▎  | 115/156 [11:35<04:11,  6.14s/it]
Training Epoch: 11/100, batch 229/313 completed (loss: 0.0723:  74%|███████▎  | 115/156 [11:35<04:11,  6.13s/it]
Training Epoch: 11/100, batch 229/313 completed (loss: 0.6935:  74%|███████▎  | 115/156 [11:35<04:11,  6.13s/it]
Training Epoch: 11/100, batch 230/313 completed (loss: 0.0968:  74%|███████▎  | 115/156 [11:38<04:11,  6.13s/it]
Training Epoch: 11/100, batch 230/313 completed (loss: 0.1928:  74%|███████▎  | 115/156 [11:38<04:11,  6.14s/it]
Training Epoch: 11/100, batch 230/313 completed (loss: 0.0241:  74%|███████▎  | 115/156 [11:38<04:11,  6.13s/it]
Training Epoch: 11/100, batch 230/313 completed (loss: 0.1005:  74%|███████▎  | 115/156 [11:38<04:11,  6.13s/it]
Training Epoch: 11/100, batch 230/313 completed (loss: 0.0241:  74%|███████▍  | 116/156 [11:41<04:01,  6.04s/it]
Training Epoch: 11/100, batch 230/313 completed (loss: 0.1928:  74%|███████▍  | 116/156 [11:41<04:01,  6.04s/it]
Training Epoch: 11/100, batch 230/313 completed (loss: 0.1005:  74%|███████▍  | 116/156 [11:41<04:01,  6.04s/it]
Training Epoch: 11/100, batch 230/313 completed (loss: 0.0968:  74%|███████▍  | 116/156 [11:41<04:01,  6.05s/it]
Training Epoch: 11/100, batch 231/313 completed (loss: 0.0412:  74%|███████▍  | 116/156 [11:41<04:01,  6.04s/it]
Training Epoch: 11/100, batch 231/313 completed (loss: 0.0432:  74%|███████▍  | 116/156 [11:41<04:01,  6.05s/it]
Training Epoch: 11/100, batch 231/313 completed (loss: 0.0121:  74%|███████▍  | 116/156 [11:41<04:01,  6.04s/it]
Training Epoch: 11/100, batch 231/313 completed (loss: 0.0422:  74%|███████▍  | 116/156 [11:41<04:01,  6.04s/it]
Training Epoch: 11/100, batch 232/313 completed (loss: 0.1379:  74%|███████▍  | 116/156 [11:44<04:01,  6.04s/it]
Training Epoch: 11/100, batch 232/313 completed (loss: 0.0512:  74%|███████▍  | 116/156 [11:44<04:01,  6.05s/it]
Training Epoch: 11/100, batch 232/313 completed (loss: 0.1756:  74%|███████▍  | 116/156 [11:44<04:01,  6.04s/it]
Training Epoch: 11/100, batch 232/313 completed (loss: 0.1033:  74%|███████▍  | 116/156 [11:44<04:01,  6.04s/it]
Training Epoch: 11/100, batch 232/313 completed (loss: 0.0512:  75%|███████▌  | 117/156 [11:47<03:56,  6.08s/it]
Training Epoch: 11/100, batch 232/313 completed (loss: 0.1379:  75%|███████▌  | 117/156 [11:47<03:57,  6.08s/it]
Training Epoch: 11/100, batch 232/313 completed (loss: 0.1033:  75%|███████▌  | 117/156 [11:47<03:57,  6.08s/it]
Training Epoch: 11/100, batch 232/313 completed (loss: 0.1756:  75%|███████▌  | 117/156 [11:47<03:57,  6.08s/it]
Training Epoch: 11/100, batch 233/313 completed (loss: 0.0575:  75%|███████▌  | 117/156 [11:47<03:57,  6.08s/it]
Training Epoch: 11/100, batch 233/313 completed (loss: 0.1841:  75%|███████▌  | 117/156 [11:47<03:57,  6.08s/it]
Training Epoch: 11/100, batch 233/313 completed (loss: 0.0895:  75%|███████▌  | 117/156 [11:47<03:56,  6.08s/it]
Training Epoch: 11/100, batch 233/313 completed (loss: 0.4868:  75%|███████▌  | 117/156 [11:47<03:57,  6.08s/it]
Training Epoch: 11/100, batch 234/313 completed (loss: 0.2681:  75%|███████▌  | 117/156 [11:51<03:57,  6.08s/it]
Training Epoch: 11/100, batch 234/313 completed (loss: 0.0609:  75%|███████▌  | 117/156 [11:51<03:56,  6.08s/it]
Training Epoch: 11/100, batch 234/313 completed (loss: 0.1592:  75%|███████▌  | 117/156 [11:51<03:57,  6.08s/it]
Training Epoch: 11/100, batch 234/313 completed (loss: 0.052:  75%|███████▌  | 117/156 [11:51<03:57,  6.08s/it] 
Training Epoch: 11/100, batch 234/313 completed (loss: 0.052:  76%|███████▌  | 118/156 [11:54<03:52,  6.11s/it]
Training Epoch: 11/100, batch 234/313 completed (loss: 0.2681:  76%|███████▌  | 118/156 [11:54<03:52,  6.11s/it]
Training Epoch: 11/100, batch 234/313 completed (loss: 0.0609:  76%|███████▌  | 118/156 [11:54<03:52,  6.11s/it]
Training Epoch: 11/100, batch 234/313 completed (loss: 0.1592:  76%|███████▌  | 118/156 [11:54<03:52,  6.11s/it]
Training Epoch: 11/100, batch 235/313 completed (loss: 0.1265:  76%|███████▌  | 118/156 [11:54<03:52,  6.11s/it]
Training Epoch: 11/100, batch 235/313 completed (loss: 0.1117:  76%|███████▌  | 118/156 [11:54<03:52,  6.11s/it]
Training Epoch: 11/100, batch 235/313 completed (loss: 0.1011:  76%|███████▌  | 118/156 [11:54<03:52,  6.11s/it]
Training Epoch: 11/100, batch 235/313 completed (loss: 0.1039:  76%|███████▌  | 118/156 [11:54<03:52,  6.11s/it]
Training Epoch: 11/100, batch 236/313 completed (loss: 0.0676:  76%|███████▌  | 118/156 [11:57<03:52,  6.11s/it]
Training Epoch: 11/100, batch 236/313 completed (loss: 0.0743:  76%|███████▌  | 118/156 [11:57<03:52,  6.11s/it]
Training Epoch: 11/100, batch 236/313 completed (loss: 0.0975:  76%|███████▌  | 118/156 [11:57<03:52,  6.11s/it]
Training Epoch: 11/100, batch 236/313 completed (loss: 0.1004:  76%|███████▌  | 118/156 [11:57<03:52,  6.11s/it]
Training Epoch: 11/100, batch 236/313 completed (loss: 0.0975:  76%|███████▋  | 119/156 [11:59<03:43,  6.04s/it]
Training Epoch: 11/100, batch 236/313 completed (loss: 0.0743:  76%|███████▋  | 119/156 [11:59<03:43,  6.04s/it]
Training Epoch: 11/100, batch 236/313 completed (loss: 0.1004:  76%|███████▋  | 119/156 [11:59<03:43,  6.04s/it]
Training Epoch: 11/100, batch 236/313 completed (loss: 0.0676:  76%|███████▋  | 119/156 [11:59<03:43,  6.04s/it]
Training Epoch: 11/100, batch 237/313 completed (loss: 0.1329:  76%|███████▋  | 119/156 [11:59<03:43,  6.04s/it]
Training Epoch: 11/100, batch 237/313 completed (loss: 0.5051:  76%|███████▋  | 119/156 [11:59<03:43,  6.04s/it]
Training Epoch: 11/100, batch 237/313 completed (loss: 0.2491:  76%|███████▋  | 119/156 [11:59<03:43,  6.04s/it]
Training Epoch: 11/100, batch 237/313 completed (loss: 0.1707:  76%|███████▋  | 119/156 [11:59<03:43,  6.04s/it]
Training Epoch: 11/100, batch 238/313 completed (loss: 0.0859:  76%|███████▋  | 119/156 [12:02<03:43,  6.04s/it]
Training Epoch: 11/100, batch 238/313 completed (loss: 0.0521:  76%|███████▋  | 119/156 [12:02<03:43,  6.04s/it]
Training Epoch: 11/100, batch 238/313 completed (loss: 0.1132:  76%|███████▋  | 119/156 [12:02<03:43,  6.04s/it]
Training Epoch: 11/100, batch 238/313 completed (loss: 0.043:  76%|███████▋  | 119/156 [12:02<03:43,  6.04s/it] 
Training Epoch: 11/100, batch 238/313 completed (loss: 0.1132:  77%|███████▋  | 120/156 [12:06<03:40,  6.11s/it]
Training Epoch: 11/100, batch 238/313 completed (loss: 0.0859:  77%|███████▋  | 120/156 [12:06<03:40,  6.12s/it]
Training Epoch: 11/100, batch 238/313 completed (loss: 0.043:  77%|███████▋  | 120/156 [12:06<03:40,  6.12s/it]
Training Epoch: 11/100, batch 238/313 completed (loss: 0.0521:  77%|███████▋  | 120/156 [12:06<03:40,  6.12s/it]
Training Epoch: 11/100, batch 239/313 completed (loss: 0.3181:  77%|███████▋  | 120/156 [12:06<03:40,  6.12s/it]
Training Epoch: 11/100, batch 239/313 completed (loss: 0.3337:  77%|███████▋  | 120/156 [12:06<03:40,  6.12s/it]
Training Epoch: 11/100, batch 239/313 completed (loss: 0.3733:  77%|███████▋  | 120/156 [12:06<03:40,  6.11s/it]
Training Epoch: 11/100, batch 239/313 completed (loss: 0.1245:  77%|███████▋  | 120/156 [12:06<03:40,  6.12s/it]
Training Epoch: 11/100, batch 240/313 completed (loss: 0.2618:  77%|███████▋  | 120/156 [12:09<03:40,  6.12s/it]
Training Epoch: 11/100, batch 240/313 completed (loss: 0.1081:  77%|███████▋  | 120/156 [12:09<03:40,  6.11s/it]
Training Epoch: 11/100, batch 240/313 completed (loss: 0.0335:  77%|███████▋  | 120/156 [12:09<03:40,  6.12s/it]
Training Epoch: 11/100, batch 240/313 completed (loss: 0.0374:  77%|███████▋  | 120/156 [12:09<03:40,  6.12s/it]
Training Epoch: 11/100, batch 240/313 completed (loss: 0.0374:  78%|███████▊  | 121/156 [12:12<03:32,  6.08s/it]
Training Epoch: 11/100, batch 240/313 completed (loss: 0.0335:  78%|███████▊  | 121/156 [12:12<03:32,  6.08s/it]
Training Epoch: 11/100, batch 240/313 completed (loss: 0.2618:  78%|███████▊  | 121/156 [12:12<03:32,  6.08s/it]
Training Epoch: 11/100, batch 240/313 completed (loss: 0.1081:  78%|███████▊  | 121/156 [12:12<03:32,  6.08s/it]
Training Epoch: 11/100, batch 241/313 completed (loss: 0.3361:  78%|███████▊  | 121/156 [12:12<03:32,  6.08s/it]
Training Epoch: 11/100, batch 241/313 completed (loss: 0.2118:  78%|███████▊  | 121/156 [12:12<03:32,  6.08s/it]
Training Epoch: 11/100, batch 241/313 completed (loss: 0.0373:  78%|███████▊  | 121/156 [12:12<03:32,  6.08s/it]
Training Epoch: 11/100, batch 241/313 completed (loss: 0.0367:  78%|███████▊  | 121/156 [12:12<03:32,  6.08s/it]
Training Epoch: 11/100, batch 242/313 completed (loss: 0.0215:  78%|███████▊  | 121/156 [12:15<03:32,  6.08s/it]
Training Epoch: 11/100, batch 242/313 completed (loss: 0.1009:  78%|███████▊  | 121/156 [12:15<03:32,  6.08s/it]
Training Epoch: 11/100, batch 242/313 completed (loss: 0.1006:  78%|███████▊  | 121/156 [12:15<03:32,  6.08s/it]
Training Epoch: 11/100, batch 242/313 completed (loss: 0.0935:  78%|███████▊  | 121/156 [12:15<03:32,  6.08s/it]
Training Epoch: 11/100, batch 242/313 completed (loss: 0.0215:  78%|███████▊  | 122/156 [12:18<03:26,  6.06s/it]
Training Epoch: 11/100, batch 242/313 completed (loss: 0.1006:  78%|███████▊  | 122/156 [12:18<03:26,  6.06s/it]
Training Epoch: 11/100, batch 242/313 completed (loss: 0.0935:  78%|███████▊  | 122/156 [12:18<03:26,  6.06s/it]
Training Epoch: 11/100, batch 242/313 completed (loss: 0.1009:  78%|███████▊  | 122/156 [12:18<03:26,  6.06s/it]
Training Epoch: 11/100, batch 243/313 completed (loss: 0.0623:  78%|███████▊  | 122/156 [12:18<03:26,  6.06s/it]
Training Epoch: 11/100, batch 243/313 completed (loss: 0.1542:  78%|███████▊  | 122/156 [12:18<03:26,  6.06s/it]
Training Epoch: 11/100, batch 243/313 completed (loss: 0.246:  78%|███████▊  | 122/156 [12:18<03:26,  6.06s/it] 
Training Epoch: 11/100, batch 243/313 completed (loss: 0.0481:  78%|███████▊  | 122/156 [12:18<03:26,  6.06s/it]
Training Epoch: 11/100, batch 244/313 completed (loss: 0.2598:  78%|███████▊  | 122/156 [12:21<03:26,  6.06s/it]
Training Epoch: 11/100, batch 244/313 completed (loss: 0.0271:  78%|███████▊  | 122/156 [12:21<03:26,  6.06s/it]
Training Epoch: 11/100, batch 244/313 completed (loss: 0.4016:  78%|███████▊  | 122/156 [12:21<03:26,  6.06s/it]
Training Epoch: 11/100, batch 244/313 completed (loss: 0.3216:  78%|███████▊  | 122/156 [12:21<03:26,  6.06s/it]
Training Epoch: 11/100, batch 244/313 completed (loss: 0.2598:  79%|███████▉  | 123/156 [12:24<03:21,  6.10s/it]
Training Epoch: 11/100, batch 244/313 completed (loss: 0.0271:  79%|███████▉  | 123/156 [12:24<03:21,  6.10s/it]
Training Epoch: 11/100, batch 244/313 completed (loss: 0.3216:  79%|███████▉  | 123/156 [12:24<03:21,  6.10s/it]
Training Epoch: 11/100, batch 244/313 completed (loss: 0.4016:  79%|███████▉  | 123/156 [12:24<03:21,  6.10s/it]
Training Epoch: 11/100, batch 245/313 completed (loss: 0.2481:  79%|███████▉  | 123/156 [12:24<03:21,  6.10s/it]
Training Epoch: 11/100, batch 245/313 completed (loss: 0.0744:  79%|███████▉  | 123/156 [12:24<03:21,  6.10s/it]
Training Epoch: 11/100, batch 245/313 completed (loss: 0.5522:  79%|███████▉  | 123/156 [12:24<03:21,  6.10s/it]
Training Epoch: 11/100, batch 245/313 completed (loss: 0.1149:  79%|███████▉  | 123/156 [12:24<03:21,  6.10s/it]
Training Epoch: 11/100, batch 246/313 completed (loss: 0.0984:  79%|███████▉  | 123/156 [12:27<03:21,  6.10s/it]
Training Epoch: 11/100, batch 246/313 completed (loss: 0.1752:  79%|███████▉  | 123/156 [12:27<03:21,  6.10s/it]
Training Epoch: 11/100, batch 246/313 completed (loss: 0.2675:  79%|███████▉  | 123/156 [12:27<03:21,  6.10s/it]
Training Epoch: 11/100, batch 246/313 completed (loss: 0.068:  79%|███████▉  | 123/156 [12:27<03:21,  6.10s/it] 
Training Epoch: 11/100, batch 246/313 completed (loss: 0.1752:  79%|███████▉  | 124/156 [12:30<03:12,  6.02s/it]
Training Epoch: 11/100, batch 246/313 completed (loss: 0.068:  79%|███████▉  | 124/156 [12:30<03:12,  6.02s/it]
Training Epoch: 11/100, batch 246/313 completed (loss: 0.2675:  79%|███████▉  | 124/156 [12:30<03:12,  6.03s/it]
Training Epoch: 11/100, batch 246/313 completed (loss: 0.0984:  79%|███████▉  | 124/156 [12:30<03:12,  6.03s/it]
Training Epoch: 11/100, batch 247/313 completed (loss: 0.132:  79%|███████▉  | 124/156 [12:30<03:12,  6.03s/it] 
Training Epoch: 11/100, batch 247/313 completed (loss: 0.3799:  79%|███████▉  | 124/156 [12:30<03:12,  6.02s/it]
Training Epoch: 11/100, batch 247/313 completed (loss: 0.1192:  79%|███████▉  | 124/156 [12:30<03:12,  6.03s/it]
Training Epoch: 11/100, batch 247/313 completed (loss: 0.3022:  79%|███████▉  | 124/156 [12:30<03:12,  6.02s/it]
Training Epoch: 11/100, batch 248/313 completed (loss: 0.0839:  79%|███████▉  | 124/156 [12:33<03:12,  6.03s/it]
Training Epoch: 11/100, batch 248/313 completed (loss: 0.2884:  79%|███████▉  | 124/156 [12:33<03:12,  6.02s/it]
Training Epoch: 11/100, batch 248/313 completed (loss: 0.2742:  79%|███████▉  | 124/156 [12:33<03:12,  6.03s/it]
Training Epoch: 11/100, batch 248/313 completed (loss: 0.0645:  79%|███████▉  | 124/156 [12:33<03:12,  6.02s/it]
Training Epoch: 11/100, batch 248/313 completed (loss: 0.0645:  80%|████████  | 125/156 [12:36<03:08,  6.07s/it]
Training Epoch: 11/100, batch 248/313 completed (loss: 0.0839:  80%|████████  | 125/156 [12:36<03:08,  6.07s/it]
Training Epoch: 11/100, batch 248/313 completed (loss: 0.2742:  80%|████████  | 125/156 [12:36<03:08,  6.07s/it]
Training Epoch: 11/100, batch 248/313 completed (loss: 0.2884:  80%|████████  | 125/156 [12:36<03:08,  6.07s/it]
Training Epoch: 11/100, batch 249/313 completed (loss: 0.0173:  80%|████████  | 125/156 [12:36<03:08,  6.07s/it]
Training Epoch: 11/100, batch 249/313 completed (loss: 0.1628:  80%|████████  | 125/156 [12:36<03:08,  6.07s/it]
Training Epoch: 11/100, batch 249/313 completed (loss: 0.2344:  80%|████████  | 125/156 [12:36<03:08,  6.07s/it]
Training Epoch: 11/100, batch 249/313 completed (loss: 0.0898:  80%|████████  | 125/156 [12:36<03:08,  6.07s/it]
Training Epoch: 11/100, batch 250/313 completed (loss: 0.2748:  80%|████████  | 125/156 [12:39<03:08,  6.07s/it]
Training Epoch: 11/100, batch 250/313 completed (loss: 0.1164:  80%|████████  | 125/156 [12:39<03:08,  6.07s/it]
Training Epoch: 11/100, batch 250/313 completed (loss: 0.0619:  80%|████████  | 125/156 [12:39<03:08,  6.07s/it]
Training Epoch: 11/100, batch 250/313 completed (loss: 0.0354:  80%|████████  | 125/156 [12:39<03:08,  6.07s/it]
Training Epoch: 11/100, batch 250/313 completed (loss: 0.0619:  81%|████████  | 126/156 [12:42<03:02,  6.10s/it]
Training Epoch: 11/100, batch 250/313 completed (loss: 0.2748:  81%|████████  | 126/156 [12:42<03:02,  6.10s/it]
Training Epoch: 11/100, batch 250/313 completed (loss: 0.0354:  81%|████████  | 126/156 [12:42<03:02,  6.10s/it]
Training Epoch: 11/100, batch 250/313 completed (loss: 0.1164:  81%|████████  | 126/156 [12:42<03:02,  6.10s/it]
Training Epoch: 11/100, batch 251/313 completed (loss: 0.0874:  81%|████████  | 126/156 [12:42<03:02,  6.10s/it]
Training Epoch: 11/100, batch 251/313 completed (loss: 0.08:  81%|████████  | 126/156 [12:42<03:02,  6.10s/it]  
Training Epoch: 11/100, batch 251/313 completed (loss: 0.1203:  81%|████████  | 126/156 [12:42<03:02,  6.10s/it]
Training Epoch: 11/100, batch 251/313 completed (loss: 0.5404:  81%|████████  | 126/156 [12:42<03:02,  6.10s/it]
Training Epoch: 11/100, batch 252/313 completed (loss: 0.0338:  81%|████████  | 126/156 [12:45<03:02,  6.10s/it]
Training Epoch: 11/100, batch 252/313 completed (loss: 0.0635:  81%|████████  | 126/156 [12:45<03:02,  6.10s/it]
Training Epoch: 11/100, batch 252/313 completed (loss: 0.1266:  81%|████████  | 126/156 [12:45<03:02,  6.10s/it]
Training Epoch: 11/100, batch 252/313 completed (loss: 0.256:  81%|████████  | 126/156 [12:45<03:02,  6.10s/it] 
Training Epoch: 11/100, batch 252/313 completed (loss: 0.0635:  81%|████████▏ | 127/156 [12:48<02:54,  6.02s/it]
Training Epoch: 11/100, batch 252/313 completed (loss: 0.256:  81%|████████▏ | 127/156 [12:48<02:54,  6.02s/it]
Training Epoch: 11/100, batch 252/313 completed (loss: 0.0338:  81%|████████▏ | 127/156 [12:48<02:54,  6.02s/it]
Training Epoch: 11/100, batch 252/313 completed (loss: 0.1266:  81%|████████▏ | 127/156 [12:48<02:54,  6.02s/it]
Training Epoch: 11/100, batch 253/313 completed (loss: 0.1155:  81%|████████▏ | 127/156 [12:48<02:54,  6.02s/it]
Training Epoch: 11/100, batch 253/313 completed (loss: 0.0821:  81%|████████▏ | 127/156 [12:48<02:54,  6.02s/it]
Training Epoch: 11/100, batch 253/313 completed (loss: 0.3752:  81%|████████▏ | 127/156 [12:48<02:54,  6.02s/it]
Training Epoch: 11/100, batch 253/313 completed (loss: 0.0245:  81%|████████▏ | 127/156 [12:48<02:54,  6.02s/it]
Training Epoch: 11/100, batch 254/313 completed (loss: 0.4215:  81%|████████▏ | 127/156 [12:51<02:54,  6.02s/it]
Training Epoch: 11/100, batch 254/313 completed (loss: 0.0478:  81%|████████▏ | 127/156 [12:51<02:54,  6.02s/it]
Training Epoch: 11/100, batch 254/313 completed (loss: 0.13:  81%|████████▏ | 127/156 [12:51<02:54,  6.02s/it]  
Training Epoch: 11/100, batch 254/313 completed (loss: 0.2681:  81%|████████▏ | 127/156 [12:51<02:54,  6.02s/it]
Training Epoch: 11/100, batch 254/313 completed (loss: 0.0478:  82%|████████▏ | 128/156 [12:54<02:52,  6.15s/it]
Training Epoch: 11/100, batch 254/313 completed (loss: 0.4215:  82%|████████▏ | 128/156 [12:54<02:52,  6.15s/it]
Training Epoch: 11/100, batch 254/313 completed (loss: 0.2681:  82%|████████▏ | 128/156 [12:54<02:52,  6.15s/it]
Training Epoch: 11/100, batch 254/313 completed (loss: 0.13:  82%|████████▏ | 128/156 [12:54<02:52,  6.15s/it]
Training Epoch: 11/100, batch 255/313 completed (loss: 0.1341:  82%|████████▏ | 128/156 [12:54<02:52,  6.15s/it]
Training Epoch: 11/100, batch 255/313 completed (loss: 0.0317:  82%|████████▏ | 128/156 [12:54<02:52,  6.15s/it]
Training Epoch: 11/100, batch 255/313 completed (loss: 0.2765:  82%|████████▏ | 128/156 [12:54<02:52,  6.15s/it]
Training Epoch: 11/100, batch 255/313 completed (loss: 0.1267:  82%|████████▏ | 128/156 [12:54<02:52,  6.15s/it]
Training Epoch: 11/100, batch 256/313 completed (loss: 0.1261:  82%|████████▏ | 128/156 [12:57<02:52,  6.15s/it]
Training Epoch: 11/100, batch 256/313 completed (loss: 0.2226:  82%|████████▏ | 128/156 [12:57<02:52,  6.15s/it]
Training Epoch: 11/100, batch 256/313 completed (loss: 0.1452:  82%|████████▏ | 128/156 [12:57<02:52,  6.15s/it]
Training Epoch: 11/100, batch 256/313 completed (loss: 0.0546:  82%|████████▏ | 128/156 [12:57<02:52,  6.15s/it]
Training Epoch: 11/100, batch 256/313 completed (loss: 0.0546:  83%|████████▎ | 129/156 [13:00<02:43,  6.06s/it]
Training Epoch: 11/100, batch 256/313 completed (loss: 0.1261:  83%|████████▎ | 129/156 [13:00<02:43,  6.06s/it]
Training Epoch: 11/100, batch 256/313 completed (loss: 0.1452:  83%|████████▎ | 129/156 [13:00<02:43,  6.06s/it]
Training Epoch: 11/100, batch 256/313 completed (loss: 0.2226:  83%|████████▎ | 129/156 [13:00<02:43,  6.06s/it]
Training Epoch: 11/100, batch 257/313 completed (loss: 0.2381:  83%|████████▎ | 129/156 [13:00<02:43,  6.06s/it]
Training Epoch: 11/100, batch 257/313 completed (loss: 0.0826:  83%|████████▎ | 129/156 [13:00<02:43,  6.06s/it]
Training Epoch: 11/100, batch 257/313 completed (loss: 0.0455:  83%|████████▎ | 129/156 [13:00<02:43,  6.06s/it]
Training Epoch: 11/100, batch 257/313 completed (loss: 0.105:  83%|████████▎ | 129/156 [13:00<02:43,  6.06s/it] 
Training Epoch: 11/100, batch 258/313 completed (loss: 0.0372:  83%|████████▎ | 129/156 [13:03<02:43,  6.06s/it]
Training Epoch: 11/100, batch 258/313 completed (loss: 0.0459:  83%|████████▎ | 129/156 [13:03<02:43,  6.06s/it]
Training Epoch: 11/100, batch 258/313 completed (loss: 0.0343:  83%|████████▎ | 129/156 [13:03<02:43,  6.06s/it]
Training Epoch: 11/100, batch 258/313 completed (loss: 0.1251:  83%|████████▎ | 129/156 [13:03<02:43,  6.06s/it]
Training Epoch: 11/100, batch 258/313 completed (loss: 0.0459:  83%|████████▎ | 130/156 [13:06<02:37,  6.04s/it]
Training Epoch: 11/100, batch 258/313 completed (loss: 0.1251:  83%|████████▎ | 130/156 [13:06<02:37,  6.05s/it]
Training Epoch: 11/100, batch 258/313 completed (loss: 0.0372:  83%|████████▎ | 130/156 [13:06<02:37,  6.05s/it]
Training Epoch: 11/100, batch 258/313 completed (loss: 0.0343:  83%|████████▎ | 130/156 [13:06<02:37,  6.05s/it]
Training Epoch: 11/100, batch 259/313 completed (loss: 0.2285:  83%|████████▎ | 130/156 [13:06<02:37,  6.05s/it]
Training Epoch: 11/100, batch 259/313 completed (loss: 0.3402:  83%|████████▎ | 130/156 [13:06<02:37,  6.05s/it]
Training Epoch: 11/100, batch 259/313 completed (loss: 0.0841:  83%|████████▎ | 130/156 [13:06<02:37,  6.04s/it]
Training Epoch: 11/100, batch 259/313 completed (loss: 0.1748:  83%|████████▎ | 130/156 [13:06<02:37,  6.05s/it]
Training Epoch: 11/100, batch 260/313 completed (loss: 0.1338:  83%|████████▎ | 130/156 [13:10<02:37,  6.05s/it]
Training Epoch: 11/100, batch 260/313 completed (loss: 0.1639:  83%|████████▎ | 130/156 [13:10<02:37,  6.05s/it]
Training Epoch: 11/100, batch 260/313 completed (loss: 0.0389:  83%|████████▎ | 130/156 [13:10<02:37,  6.04s/it]
Training Epoch: 11/100, batch 260/313 completed (loss: 0.054:  83%|████████▎ | 130/156 [13:10<02:37,  6.05s/it] 
Training Epoch: 11/100, batch 260/313 completed (loss: 0.054:  84%|████████▍ | 131/156 [13:13<02:32,  6.12s/it]
Training Epoch: 11/100, batch 260/313 completed (loss: 0.1639:  84%|████████▍ | 131/156 [13:13<02:32,  6.12s/it]
Training Epoch: 11/100, batch 260/313 completed (loss: 0.0389:  84%|████████▍ | 131/156 [13:13<02:32,  6.12s/it]
Training Epoch: 11/100, batch 260/313 completed (loss: 0.1338:  84%|████████▍ | 131/156 [13:13<02:32,  6.12s/it]
Training Epoch: 11/100, batch 261/313 completed (loss: 0.0586:  84%|████████▍ | 131/156 [13:13<02:32,  6.12s/it]
Training Epoch: 11/100, batch 261/313 completed (loss: 0.2188:  84%|████████▍ | 131/156 [13:13<02:32,  6.12s/it]
Training Epoch: 11/100, batch 261/313 completed (loss: 0.3878:  84%|████████▍ | 131/156 [13:13<02:32,  6.12s/it]
Training Epoch: 11/100, batch 261/313 completed (loss: 0.0864:  84%|████████▍ | 131/156 [13:13<02:32,  6.12s/it]
Training Epoch: 11/100, batch 262/313 completed (loss: 0.403:  84%|████████▍ | 131/156 [13:16<02:32,  6.12s/it] 
Training Epoch: 11/100, batch 262/313 completed (loss: 0.3793:  84%|████████▍ | 131/156 [13:16<02:32,  6.12s/it]
Training Epoch: 11/100, batch 262/313 completed (loss: 0.0668:  84%|████████▍ | 131/156 [13:16<02:32,  6.12s/it]
Training Epoch: 11/100, batch 262/313 completed (loss: 0.2985:  84%|████████▍ | 131/156 [13:16<02:32,  6.12s/it]
Training Epoch: 11/100, batch 262/313 completed (loss: 0.2985:  85%|████████▍ | 132/156 [13:18<02:25,  6.05s/it]
Training Epoch: 11/100, batch 262/313 completed (loss: 0.403:  85%|████████▍ | 132/156 [13:18<02:25,  6.05s/it]
Training Epoch: 11/100, batch 262/313 completed (loss: 0.0668:  85%|████████▍ | 132/156 [13:18<02:25,  6.05s/it]
Training Epoch: 11/100, batch 262/313 completed (loss: 0.3793:  85%|████████▍ | 132/156 [13:18<02:25,  6.05s/it]
Training Epoch: 11/100, batch 263/313 completed (loss: 0.0593:  85%|████████▍ | 132/156 [13:18<02:25,  6.05s/it]
Training Epoch: 11/100, batch 263/313 completed (loss: 0.2144:  85%|████████▍ | 132/156 [13:18<02:25,  6.05s/it]
Training Epoch: 11/100, batch 263/313 completed (loss: 0.2314:  85%|████████▍ | 132/156 [13:18<02:25,  6.05s/it]
Training Epoch: 11/100, batch 263/313 completed (loss: 0.2828:  85%|████████▍ | 132/156 [13:19<02:25,  6.05s/it]
Training Epoch: 11/100, batch 264/313 completed (loss: 0.3521:  85%|████████▍ | 132/156 [13:21<02:25,  6.05s/it]
Training Epoch: 11/100, batch 264/313 completed (loss: 0.0389:  85%|████████▍ | 132/156 [13:21<02:25,  6.05s/it]
Training Epoch: 11/100, batch 264/313 completed (loss: 0.0841:  85%|████████▍ | 132/156 [13:21<02:25,  6.05s/it]
Training Epoch: 11/100, batch 264/313 completed (loss: 0.1118:  85%|████████▍ | 132/156 [13:21<02:25,  6.05s/it]
Training Epoch: 11/100, batch 264/313 completed (loss: 0.0841:  85%|████████▌ | 133/156 [13:25<02:20,  6.10s/it]
Training Epoch: 11/100, batch 264/313 completed (loss: 0.3521:  85%|████████▌ | 133/156 [13:25<02:20,  6.10s/it]
Training Epoch: 11/100, batch 264/313 completed (loss: 0.1118:  85%|████████▌ | 133/156 [13:25<02:20,  6.10s/it]
Training Epoch: 11/100, batch 264/313 completed (loss: 0.0389:  85%|████████▌ | 133/156 [13:25<02:20,  6.10s/it]
Training Epoch: 11/100, batch 265/313 completed (loss: 0.1293:  85%|████████▌ | 133/156 [13:25<02:20,  6.10s/it]
Training Epoch: 11/100, batch 265/313 completed (loss: 0.1226:  85%|████████▌ | 133/156 [13:25<02:20,  6.10s/it]
Training Epoch: 11/100, batch 265/313 completed (loss: 0.1046:  85%|████████▌ | 133/156 [13:25<02:20,  6.10s/it]
Training Epoch: 11/100, batch 265/313 completed (loss: 0.2008:  85%|████████▌ | 133/156 [13:25<02:20,  6.10s/it]
Training Epoch: 11/100, batch 266/313 completed (loss: 0.2303:  85%|████████▌ | 133/156 [13:28<02:20,  6.10s/it]
Training Epoch: 11/100, batch 266/313 completed (loss: 0.2368:  85%|████████▌ | 133/156 [13:28<02:20,  6.10s/it]
Training Epoch: 11/100, batch 266/313 completed (loss: 0.0883:  85%|████████▌ | 133/156 [13:28<02:20,  6.10s/it]
Training Epoch: 11/100, batch 266/313 completed (loss: 0.0382:  85%|████████▌ | 133/156 [13:28<02:20,  6.10s/it]
Training Epoch: 11/100, batch 266/313 completed (loss: 0.0382:  86%|████████▌ | 134/156 [13:31<02:13,  6.08s/it]
Training Epoch: 11/100, batch 266/313 completed (loss: 0.2368:  86%|████████▌ | 134/156 [13:31<02:13,  6.08s/it]
Training Epoch: 11/100, batch 266/313 completed (loss: 0.2303:  86%|████████▌ | 134/156 [13:31<02:13,  6.08s/it]
Training Epoch: 11/100, batch 266/313 completed (loss: 0.0883:  86%|████████▌ | 134/156 [13:31<02:13,  6.09s/it]
Training Epoch: 11/100, batch 267/313 completed (loss: 0.0858:  86%|████████▌ | 134/156 [13:31<02:13,  6.08s/it]
Training Epoch: 11/100, batch 267/313 completed (loss: 0.2274:  86%|████████▌ | 134/156 [13:31<02:13,  6.08s/it]
Training Epoch: 11/100, batch 267/313 completed (loss: 0.0488:  86%|████████▌ | 134/156 [13:31<02:13,  6.09s/it]
Training Epoch: 11/100, batch 267/313 completed (loss: 0.0481:  86%|████████▌ | 134/156 [13:31<02:13,  6.08s/it]
Training Epoch: 11/100, batch 268/313 completed (loss: 0.1584:  86%|████████▌ | 134/156 [13:34<02:13,  6.09s/it]
Training Epoch: 11/100, batch 268/313 completed (loss: 0.0234:  86%|████████▌ | 134/156 [13:34<02:13,  6.08s/it]
Training Epoch: 11/100, batch 268/313 completed (loss: 0.1084:  86%|████████▌ | 134/156 [13:34<02:13,  6.08s/it]
Training Epoch: 11/100, batch 268/313 completed (loss: 0.0608:  86%|████████▌ | 134/156 [13:34<02:13,  6.08s/it]
Training Epoch: 11/100, batch 268/313 completed (loss: 0.1084:  87%|████████▋ | 135/156 [13:37<02:06,  6.02s/it]
Training Epoch: 11/100, batch 268/313 completed (loss: 0.1584:  87%|████████▋ | 135/156 [13:37<02:06,  6.02s/it]
Training Epoch: 11/100, batch 268/313 completed (loss: 0.0608:  87%|████████▋ | 135/156 [13:37<02:06,  6.02s/it]
Training Epoch: 11/100, batch 268/313 completed (loss: 0.0234:  87%|████████▋ | 135/156 [13:37<02:06,  6.02s/it]
Training Epoch: 11/100, batch 269/313 completed (loss: 0.3866:  87%|████████▋ | 135/156 [13:37<02:06,  6.02s/it]
Training Epoch: 11/100, batch 269/313 completed (loss: 0.2183:  87%|████████▋ | 135/156 [13:37<02:06,  6.02s/it]
Training Epoch: 11/100, batch 269/313 completed (loss: 0.0545:  87%|████████▋ | 135/156 [13:37<02:06,  6.02s/it]
Training Epoch: 11/100, batch 269/313 completed (loss: 0.0883:  87%|████████▋ | 135/156 [13:37<02:06,  6.02s/it]
Training Epoch: 11/100, batch 270/313 completed (loss: 0.1599:  87%|████████▋ | 135/156 [13:40<02:06,  6.02s/it]
Training Epoch: 11/100, batch 270/313 completed (loss: 0.1309:  87%|████████▋ | 135/156 [13:40<02:06,  6.02s/it]
Training Epoch: 11/100, batch 270/313 completed (loss: 0.1626:  87%|████████▋ | 135/156 [13:40<02:06,  6.02s/it]
Training Epoch: 11/100, batch 270/313 completed (loss: 0.3042:  87%|████████▋ | 135/156 [13:40<02:06,  6.02s/it]
Training Epoch: 11/100, batch 270/313 completed (loss: 0.3042:  87%|████████▋ | 136/156 [13:43<02:02,  6.12s/it]
Training Epoch: 11/100, batch 270/313 completed (loss: 0.1309:  87%|████████▋ | 136/156 [13:43<02:02,  6.12s/it]
Training Epoch: 11/100, batch 270/313 completed (loss: 0.1599:  87%|████████▋ | 136/156 [13:43<02:02,  6.12s/it]
Training Epoch: 11/100, batch 270/313 completed (loss: 0.1626:  87%|████████▋ | 136/156 [13:43<02:02,  6.13s/it]
Training Epoch: 11/100, batch 271/313 completed (loss: 0.088:  87%|████████▋ | 136/156 [13:43<02:02,  6.13s/it] 
Training Epoch: 11/100, batch 271/313 completed (loss: 0.0837:  87%|████████▋ | 136/156 [13:43<02:02,  6.12s/it]
Training Epoch: 11/100, batch 271/313 completed (loss: 0.0602:  87%|████████▋ | 136/156 [13:43<02:02,  6.12s/it]
Training Epoch: 11/100, batch 271/313 completed (loss: 0.14:  87%|████████▋ | 136/156 [13:43<02:02,  6.12s/it]  
Training Epoch: 11/100, batch 272/313 completed (loss: 0.0301:  87%|████████▋ | 136/156 [13:46<02:02,  6.12s/it]
Training Epoch: 11/100, batch 272/313 completed (loss: 0.2345:  87%|████████▋ | 136/156 [13:46<02:02,  6.12s/it]
Training Epoch: 11/100, batch 272/313 completed (loss: 0.1016:  87%|████████▋ | 136/156 [13:46<02:02,  6.13s/it]
Training Epoch: 11/100, batch 272/313 completed (loss: 0.1668:  87%|████████▋ | 136/156 [13:46<02:02,  6.12s/it]
Training Epoch: 11/100, batch 272/313 completed (loss: 0.2345:  88%|████████▊ | 137/156 [13:49<01:54,  6.04s/it]
Training Epoch: 11/100, batch 272/313 completed (loss: 0.1668:  88%|████████▊ | 137/156 [13:49<01:54,  6.04s/it]
Training Epoch: 11/100, batch 272/313 completed (loss: 0.0301:  88%|████████▊ | 137/156 [13:49<01:54,  6.04s/it]
Training Epoch: 11/100, batch 272/313 completed (loss: 0.1016:  88%|████████▊ | 137/156 [13:49<01:54,  6.04s/it]
Training Epoch: 11/100, batch 273/313 completed (loss: 0.2876:  88%|████████▊ | 137/156 [13:49<01:54,  6.04s/it]
Training Epoch: 11/100, batch 273/313 completed (loss: 0.3281:  88%|████████▊ | 137/156 [13:49<01:54,  6.04s/it]
Training Epoch: 11/100, batch 273/313 completed (loss: 0.0837:  88%|████████▊ | 137/156 [13:49<01:54,  6.04s/it]
Training Epoch: 11/100, batch 273/313 completed (loss: 0.3813:  88%|████████▊ | 137/156 [13:49<01:54,  6.04s/it]
Training Epoch: 11/100, batch 274/313 completed (loss: 0.0307:  88%|████████▊ | 137/156 [13:52<01:54,  6.04s/it]
Training Epoch: 11/100, batch 274/313 completed (loss: 0.153:  88%|████████▊ | 137/156 [13:52<01:54,  6.04s/it] 
Training Epoch: 11/100, batch 274/313 completed (loss: 0.1001:  88%|████████▊ | 137/156 [13:52<01:54,  6.04s/it]
Training Epoch: 11/100, batch 274/313 completed (loss: 0.1373:  88%|████████▊ | 137/156 [13:52<01:54,  6.04s/it]
Training Epoch: 11/100, batch 274/313 completed (loss: 0.1001:  88%|████████▊ | 138/156 [13:55<01:49,  6.07s/it]
Training Epoch: 11/100, batch 274/313 completed (loss: 0.153:  88%|████████▊ | 138/156 [13:55<01:49,  6.07s/it]
Training Epoch: 11/100, batch 274/313 completed (loss: 0.1373:  88%|████████▊ | 138/156 [13:55<01:49,  6.07s/it]
Training Epoch: 11/100, batch 274/313 completed (loss: 0.0307:  88%|████████▊ | 138/156 [13:55<01:49,  6.07s/it]
Training Epoch: 11/100, batch 275/313 completed (loss: 0.0552:  88%|████████▊ | 138/156 [13:55<01:49,  6.07s/it]
Training Epoch: 11/100, batch 275/313 completed (loss: 0.1374:  88%|████████▊ | 138/156 [13:55<01:49,  6.07s/it]
Training Epoch: 11/100, batch 275/313 completed (loss: 0.0638:  88%|████████▊ | 138/156 [13:55<01:49,  6.07s/it]
Training Epoch: 11/100, batch 275/313 completed (loss: 0.3025:  88%|████████▊ | 138/156 [13:55<01:49,  6.07s/it]
Training Epoch: 11/100, batch 276/313 completed (loss: 0.0474:  88%|████████▊ | 138/156 [13:58<01:49,  6.07s/it]
Training Epoch: 11/100, batch 276/313 completed (loss: 0.1335:  88%|████████▊ | 138/156 [13:58<01:49,  6.07s/it]
Training Epoch: 11/100, batch 276/313 completed (loss: 0.2736:  88%|████████▊ | 138/156 [13:58<01:49,  6.07s/it]
Training Epoch: 11/100, batch 276/313 completed (loss: 0.0957:  88%|████████▊ | 138/156 [13:58<01:49,  6.07s/it]
Training Epoch: 11/100, batch 276/313 completed (loss: 0.1335:  89%|████████▉ | 139/156 [14:01<01:43,  6.10s/it]
Training Epoch: 11/100, batch 276/313 completed (loss: 0.0957:  89%|████████▉ | 139/156 [14:01<01:43,  6.10s/it]
Training Epoch: 11/100, batch 276/313 completed (loss: 0.0474:  89%|████████▉ | 139/156 [14:01<01:43,  6.10s/it]
Training Epoch: 11/100, batch 276/313 completed (loss: 0.2736:  89%|████████▉ | 139/156 [14:01<01:43,  6.10s/it]
Training Epoch: 11/100, batch 277/313 completed (loss: 0.0885:  89%|████████▉ | 139/156 [14:01<01:43,  6.10s/it]
Training Epoch: 11/100, batch 277/313 completed (loss: 0.0509:  89%|████████▉ | 139/156 [14:01<01:43,  6.10s/it]
Training Epoch: 11/100, batch 277/313 completed (loss: 0.1712:  89%|████████▉ | 139/156 [14:01<01:43,  6.10s/it]
Training Epoch: 11/100, batch 277/313 completed (loss: 0.1122:  89%|████████▉ | 139/156 [14:01<01:43,  6.10s/it]
Training Epoch: 11/100, batch 278/313 completed (loss: 0.0826:  89%|████████▉ | 139/156 [14:04<01:43,  6.10s/it]
Training Epoch: 11/100, batch 278/313 completed (loss: 0.2521:  89%|████████▉ | 139/156 [14:04<01:43,  6.10s/it]
Training Epoch: 11/100, batch 278/313 completed (loss: 0.0974:  89%|████████▉ | 139/156 [14:04<01:43,  6.10s/it]
Training Epoch: 11/100, batch 278/313 completed (loss: 0.1104:  89%|████████▉ | 139/156 [14:04<01:43,  6.10s/it]
Training Epoch: 11/100, batch 278/313 completed (loss: 0.0974:  90%|████████▉ | 140/156 [14:07<01:36,  6.04s/it]
Training Epoch: 11/100, batch 278/313 completed (loss: 0.2521:  90%|████████▉ | 140/156 [14:07<01:36,  6.04s/it]
Training Epoch: 11/100, batch 278/313 completed (loss: 0.0826:  90%|████████▉ | 140/156 [14:07<01:36,  6.04s/it]
Training Epoch: 11/100, batch 278/313 completed (loss: 0.1104:  90%|████████▉ | 140/156 [14:07<01:36,  6.04s/it]
Training Epoch: 11/100, batch 279/313 completed (loss: 0.1791:  90%|████████▉ | 140/156 [14:07<01:36,  6.04s/it]
Training Epoch: 11/100, batch 279/313 completed (loss: 0.0868:  90%|████████▉ | 140/156 [14:07<01:36,  6.04s/it]
Training Epoch: 11/100, batch 279/313 completed (loss: 0.055:  90%|████████▉ | 140/156 [14:07<01:36,  6.04s/it] 
Training Epoch: 11/100, batch 279/313 completed (loss: 0.0813:  90%|████████▉ | 140/156 [14:07<01:36,  6.04s/it]
Training Epoch: 11/100, batch 280/313 completed (loss: 0.2359:  90%|████████▉ | 140/156 [14:10<01:36,  6.04s/it]
Training Epoch: 11/100, batch 280/313 completed (loss: 0.083:  90%|████████▉ | 140/156 [14:10<01:36,  6.04s/it] 
Training Epoch: 11/100, batch 280/313 completed (loss: 0.1102:  90%|████████▉ | 140/156 [14:10<01:36,  6.04s/it]
Training Epoch: 11/100, batch 280/313 completed (loss: 0.0966:  90%|████████▉ | 140/156 [14:10<01:36,  6.04s/it]
Training Epoch: 11/100, batch 280/313 completed (loss: 0.2359:  90%|█████████ | 141/156 [14:13<01:31,  6.11s/it]
Training Epoch: 11/100, batch 280/313 completed (loss: 0.1102:  90%|█████████ | 141/156 [14:13<01:31,  6.11s/it]
Training Epoch: 11/100, batch 280/313 completed (loss: 0.083:  90%|█████████ | 141/156 [14:13<01:31,  6.11s/it]
Training Epoch: 11/100, batch 280/313 completed (loss: 0.0966:  90%|█████████ | 141/156 [14:13<01:31,  6.12s/it]
Training Epoch: 11/100, batch 281/313 completed (loss: 0.042:  90%|█████████ | 141/156 [14:13<01:31,  6.11s/it] 
Training Epoch: 11/100, batch 281/313 completed (loss: 0.5311:  90%|█████████ | 141/156 [14:13<01:31,  6.11s/it]
Training Epoch: 11/100, batch 281/313 completed (loss: 0.1297:  90%|█████████ | 141/156 [14:13<01:31,  6.11s/it]
Training Epoch: 11/100, batch 281/313 completed (loss: 0.175:  90%|█████████ | 141/156 [14:13<01:31,  6.12s/it] 
Training Epoch: 11/100, batch 282/313 completed (loss: 0.2053:  90%|█████████ | 141/156 [14:16<01:31,  6.11s/it]
Training Epoch: 11/100, batch 282/313 completed (loss: 0.1454:  90%|█████████ | 141/156 [14:16<01:31,  6.11s/it]
Training Epoch: 11/100, batch 282/313 completed (loss: 0.3312:  90%|█████████ | 141/156 [14:16<01:31,  6.11s/it]
Training Epoch: 11/100, batch 282/313 completed (loss: 0.08:  90%|█████████ | 141/156 [14:16<01:31,  6.12s/it] 
Training Epoch: 11/100, batch 282/313 completed (loss: 0.1454:  91%|█████████ | 142/156 [14:19<01:25,  6.10s/it]
Training Epoch: 11/100, batch 282/313 completed (loss: 0.3312:  91%|█████████ | 142/156 [14:19<01:25,  6.10s/it]
Training Epoch: 11/100, batch 282/313 completed (loss: 0.2053:  91%|█████████ | 142/156 [14:19<01:25,  6.10s/it]
Training Epoch: 11/100, batch 282/313 completed (loss: 0.08:  91%|█████████ | 142/156 [14:19<01:25,  6.10s/it]
Training Epoch: 11/100, batch 283/313 completed (loss: 0.0914:  91%|█████████ | 142/156 [14:19<01:25,  6.10s/it]
Training Epoch: 11/100, batch 283/313 completed (loss: 0.1289:  91%|█████████ | 142/156 [14:19<01:25,  6.10s/it]
Training Epoch: 11/100, batch 283/313 completed (loss: 0.2703:  91%|█████████ | 142/156 [14:19<01:25,  6.10s/it]
Training Epoch: 11/100, batch 283/313 completed (loss: 0.1496:  91%|█████████ | 142/156 [14:19<01:25,  6.10s/it]
Training Epoch: 11/100, batch 284/313 completed (loss: 0.0321:  91%|█████████ | 142/156 [14:22<01:25,  6.10s/it]
Training Epoch: 11/100, batch 284/313 completed (loss: 0.1125:  91%|█████████ | 142/156 [14:22<01:25,  6.10s/it]
Training Epoch: 11/100, batch 284/313 completed (loss: 0.1136:  91%|█████████ | 142/156 [14:22<01:25,  6.10s/it]
Training Epoch: 11/100, batch 284/313 completed (loss: 0.0522:  91%|█████████ | 142/156 [14:22<01:25,  6.10s/it]
Training Epoch: 11/100, batch 284/313 completed (loss: 0.0321:  92%|█████████▏| 143/156 [14:25<01:19,  6.12s/it]
Training Epoch: 11/100, batch 284/313 completed (loss: 0.0522:  92%|█████████▏| 143/156 [14:26<01:19,  6.11s/it]
Training Epoch: 11/100, batch 284/313 completed (loss: 0.1136:  92%|█████████▏| 143/156 [14:25<01:19,  6.12s/it]
Training Epoch: 11/100, batch 284/313 completed (loss: 0.1125:  92%|█████████▏| 143/156 [14:26<01:19,  6.12s/it]
Training Epoch: 11/100, batch 285/313 completed (loss: 0.2029:  92%|█████████▏| 143/156 [14:26<01:19,  6.12s/it]
Training Epoch: 11/100, batch 285/313 completed (loss: 0.1854:  92%|█████████▏| 143/156 [14:26<01:19,  6.12s/it]
Training Epoch: 11/100, batch 285/313 completed (loss: 0.1856:  92%|█████████▏| 143/156 [14:26<01:19,  6.12s/it]
Training Epoch: 11/100, batch 285/313 completed (loss: 0.0161:  92%|█████████▏| 143/156 [14:26<01:19,  6.11s/it]
Training Epoch: 11/100, batch 286/313 completed (loss: 0.2294:  92%|█████████▏| 143/156 [14:29<01:19,  6.12s/it]
Training Epoch: 11/100, batch 286/313 completed (loss: 0.1801:  92%|█████████▏| 143/156 [14:29<01:19,  6.12s/it]
Training Epoch: 11/100, batch 286/313 completed (loss: 0.059:  92%|█████████▏| 143/156 [14:29<01:19,  6.12s/it] 
Training Epoch: 11/100, batch 286/313 completed (loss: 0.3408:  92%|█████████▏| 143/156 [14:29<01:19,  6.11s/it]
Training Epoch: 11/100, batch 286/313 completed (loss: 0.059:  92%|█████████▏| 144/156 [14:32<01:13,  6.14s/it]
Training Epoch: 11/100, batch 286/313 completed (loss: 0.2294:  92%|█████████▏| 144/156 [14:32<01:13,  6.14s/it]
Training Epoch: 11/100, batch 286/313 completed (loss: 0.1801:  92%|█████████▏| 144/156 [14:32<01:13,  6.14s/it]
Training Epoch: 11/100, batch 286/313 completed (loss: 0.3408:  92%|█████████▏| 144/156 [14:32<01:13,  6.14s/it]
Training Epoch: 11/100, batch 287/313 completed (loss: 0.0505:  92%|█████████▏| 144/156 [14:32<01:13,  6.14s/it]
Training Epoch: 11/100, batch 287/313 completed (loss: 0.0502:  92%|█████████▏| 144/156 [14:32<01:13,  6.14s/it]
Training Epoch: 11/100, batch 287/313 completed (loss: 0.288:  92%|█████████▏| 144/156 [14:32<01:13,  6.14s/it] 
Training Epoch: 11/100, batch 287/313 completed (loss: 0.0594:  92%|█████████▏| 144/156 [14:32<01:13,  6.14s/it]
Training Epoch: 11/100, batch 288/313 completed (loss: 0.1574:  92%|█████████▏| 144/156 [14:35<01:13,  6.14s/it]
Training Epoch: 11/100, batch 288/313 completed (loss: 0.3326:  92%|█████████▏| 144/156 [14:35<01:13,  6.14s/it]
Training Epoch: 11/100, batch 288/313 completed (loss: 0.197:  92%|█████████▏| 144/156 [14:35<01:13,  6.14s/it] 
Training Epoch: 11/100, batch 288/313 completed (loss: 0.1702:  92%|█████████▏| 144/156 [14:35<01:13,  6.14s/it]
Training Epoch: 11/100, batch 288/313 completed (loss: 0.1574:  93%|█████████▎| 145/156 [14:38<01:06,  6.05s/it]
Training Epoch: 11/100, batch 288/313 completed (loss: 0.1702:  93%|█████████▎| 145/156 [14:38<01:06,  6.05s/it]
Training Epoch: 11/100, batch 288/313 completed (loss: 0.197:  93%|█████████▎| 145/156 [14:38<01:06,  6.06s/it]
Training Epoch: 11/100, batch 288/313 completed (loss: 0.3326:  93%|█████████▎| 145/156 [14:38<01:06,  6.06s/it]
Training Epoch: 11/100, batch 289/313 completed (loss: 0.1387:  93%|█████████▎| 145/156 [14:38<01:06,  6.05s/it]
Training Epoch: 11/100, batch 289/313 completed (loss: 0.058:  93%|█████████▎| 145/156 [14:38<01:06,  6.06s/it] 
Training Epoch: 11/100, batch 289/313 completed (loss: 0.2398:  93%|█████████▎| 145/156 [14:38<01:06,  6.06s/it]
Training Epoch: 11/100, batch 289/313 completed (loss: 0.2046:  93%|█████████▎| 145/156 [14:38<01:06,  6.05s/it]
Training Epoch: 11/100, batch 290/313 completed (loss: 0.0635:  93%|█████████▎| 145/156 [14:41<01:06,  6.06s/it]
Training Epoch: 11/100, batch 290/313 completed (loss: 0.1287:  93%|█████████▎| 145/156 [14:41<01:06,  6.05s/it]
Training Epoch: 11/100, batch 290/313 completed (loss: 0.2165:  93%|█████████▎| 145/156 [14:41<01:06,  6.06s/it]
Training Epoch: 11/100, batch 290/313 completed (loss: 0.1201:  93%|█████████▎| 145/156 [14:41<01:06,  6.05s/it]
Training Epoch: 11/100, batch 290/313 completed (loss: 0.0635:  94%|█████████▎| 146/156 [14:44<01:00,  6.08s/it]
Training Epoch: 11/100, batch 290/313 completed (loss: 0.2165:  94%|█████████▎| 146/156 [14:44<01:00,  6.08s/it]
Training Epoch: 11/100, batch 290/313 completed (loss: 0.1287:  94%|█████████▎| 146/156 [14:44<01:00,  6.08s/it]
Training Epoch: 11/100, batch 290/313 completed (loss: 0.1201:  94%|█████████▎| 146/156 [14:44<01:00,  6.08s/it]
Training Epoch: 11/100, batch 291/313 completed (loss: 0.2814:  94%|█████████▎| 146/156 [14:44<01:00,  6.08s/it]
Training Epoch: 11/100, batch 291/313 completed (loss: 0.1424:  94%|█████████▎| 146/156 [14:44<01:00,  6.08s/it]
Training Epoch: 11/100, batch 291/313 completed (loss: 0.0221:  94%|█████████▎| 146/156 [14:44<01:00,  6.08s/it]
Training Epoch: 11/100, batch 291/313 completed (loss: 0.2191:  94%|█████████▎| 146/156 [14:44<01:00,  6.08s/it]
Training Epoch: 11/100, batch 292/313 completed (loss: 0.1647:  94%|█████████▎| 146/156 [14:47<01:00,  6.08s/it]
Training Epoch: 11/100, batch 292/313 completed (loss: 0.026:  94%|█████████▎| 146/156 [14:47<01:00,  6.08s/it] 
Training Epoch: 11/100, batch 292/313 completed (loss: 0.0965:  94%|█████████▎| 146/156 [14:47<01:00,  6.08s/it]
Training Epoch: 11/100, batch 292/313 completed (loss: 0.1244:  94%|█████████▎| 146/156 [14:47<01:00,  6.08s/it]
Training Epoch: 11/100, batch 292/313 completed (loss: 0.0965:  94%|█████████▍| 147/156 [14:50<00:54,  6.10s/it]
Training Epoch: 11/100, batch 292/313 completed (loss: 0.1647:  94%|█████████▍| 147/156 [14:50<00:54,  6.10s/it]
Training Epoch: 11/100, batch 292/313 completed (loss: 0.026:  94%|█████████▍| 147/156 [14:50<00:54,  6.11s/it]
Training Epoch: 11/100, batch 292/313 completed (loss: 0.1244:  94%|█████████▍| 147/156 [14:50<00:54,  6.10s/it]
Training Epoch: 11/100, batch 293/313 completed (loss: 0.0778:  94%|█████████▍| 147/156 [14:50<00:54,  6.11s/it]
Training Epoch: 11/100, batch 293/313 completed (loss: 0.0569:  94%|█████████▍| 147/156 [14:50<00:54,  6.10s/it]
Training Epoch: 11/100, batch 293/313 completed (loss: 0.0567:  94%|█████████▍| 147/156 [14:50<00:54,  6.10s/it]
Training Epoch: 11/100, batch 293/313 completed (loss: 0.0842:  94%|█████████▍| 147/156 [14:50<00:54,  6.10s/it]
Training Epoch: 11/100, batch 294/313 completed (loss: 0.0852:  94%|█████████▍| 147/156 [14:53<00:54,  6.10s/it]
Training Epoch: 11/100, batch 294/313 completed (loss: 0.2567:  94%|█████████▍| 147/156 [14:53<00:54,  6.11s/it]
Training Epoch: 11/100, batch 294/313 completed (loss: 0.2016:  94%|█████████▍| 147/156 [14:53<00:54,  6.10s/it]
Training Epoch: 11/100, batch 294/313 completed (loss: 0.1061:  94%|█████████▍| 147/156 [14:53<00:54,  6.10s/it]
Training Epoch: 11/100, batch 294/313 completed (loss: 0.1061:  95%|█████████▍| 148/156 [14:56<00:48,  6.07s/it]
Training Epoch: 11/100, batch 294/313 completed (loss: 0.2567:  95%|█████████▍| 148/156 [14:56<00:48,  6.07s/it]
Training Epoch: 11/100, batch 294/313 completed (loss: 0.0852:  95%|█████████▍| 148/156 [14:56<00:48,  6.07s/it]
Training Epoch: 11/100, batch 294/313 completed (loss: 0.2016:  95%|█████████▍| 148/156 [14:56<00:48,  6.07s/it]
Training Epoch: 11/100, batch 295/313 completed (loss: 0.3395:  95%|█████████▍| 148/156 [14:56<00:48,  6.07s/it]
Training Epoch: 11/100, batch 295/313 completed (loss: 0.0457:  95%|█████████▍| 148/156 [14:56<00:48,  6.07s/it]
Training Epoch: 11/100, batch 295/313 completed (loss: 0.2048:  95%|█████████▍| 148/156 [14:56<00:48,  6.07s/it]
Training Epoch: 11/100, batch 295/313 completed (loss: 0.1618:  95%|█████████▍| 148/156 [14:56<00:48,  6.07s/it]
Training Epoch: 11/100, batch 296/313 completed (loss: 0.118:  95%|█████████▍| 148/156 [14:59<00:48,  6.07s/it] 
Training Epoch: 11/100, batch 296/313 completed (loss: 0.1391:  95%|█████████▍| 148/156 [14:59<00:48,  6.07s/it]
Training Epoch: 11/100, batch 296/313 completed (loss: 0.1301:  95%|█████████▍| 148/156 [14:59<00:48,  6.07s/it]
Training Epoch: 11/100, batch 296/313 completed (loss: 0.1287:  95%|█████████▍| 148/156 [14:59<00:48,  6.07s/it]
Training Epoch: 11/100, batch 296/313 completed (loss: 0.1391:  96%|█████████▌| 149/156 [15:02<00:42,  6.09s/it]
Training Epoch: 11/100, batch 296/313 completed (loss: 0.1301:  96%|█████████▌| 149/156 [15:02<00:42,  6.09s/it]
Training Epoch: 11/100, batch 296/313 completed (loss: 0.118:  96%|█████████▌| 149/156 [15:02<00:42,  6.09s/it]
Training Epoch: 11/100, batch 296/313 completed (loss: 0.1287:  96%|█████████▌| 149/156 [15:02<00:42,  6.09s/it]
Training Epoch: 11/100, batch 297/313 completed (loss: 0.3209:  96%|█████████▌| 149/156 [15:02<00:42,  6.09s/it]
Training Epoch: 11/100, batch 297/313 completed (loss: 0.1362:  96%|█████████▌| 149/156 [15:02<00:42,  6.09s/it]
Training Epoch: 11/100, batch 297/313 completed (loss: 0.0497:  96%|█████████▌| 149/156 [15:02<00:42,  6.09s/it]
Training Epoch: 11/100, batch 297/313 completed (loss: 0.0544:  96%|█████████▌| 149/156 [15:02<00:42,  6.09s/it]
Training Epoch: 11/100, batch 298/313 completed (loss: 0.1696:  96%|█████████▌| 149/156 [15:05<00:42,  6.09s/it]
Training Epoch: 11/100, batch 298/313 completed (loss: 0.0523:  96%|█████████▌| 149/156 [15:05<00:42,  6.09s/it]
Training Epoch: 11/100, batch 298/313 completed (loss: 0.0897:  96%|█████████▌| 149/156 [15:05<00:42,  6.09s/it]
Training Epoch: 11/100, batch 298/313 completed (loss: 0.0922:  96%|█████████▌| 149/156 [15:05<00:42,  6.09s/it]
Training Epoch: 11/100, batch 298/313 completed (loss: 0.1696:  96%|█████████▌| 150/156 [15:08<00:36,  6.07s/it]
Training Epoch: 11/100, batch 298/313 completed (loss: 0.0523:  96%|█████████▌| 150/156 [15:08<00:36,  6.07s/it]
Training Epoch: 11/100, batch 298/313 completed (loss: 0.0922:  96%|█████████▌| 150/156 [15:08<00:36,  6.07s/it]
Training Epoch: 11/100, batch 298/313 completed (loss: 0.0897:  96%|█████████▌| 150/156 [15:08<00:36,  6.08s/it]
Training Epoch: 11/100, batch 299/313 completed (loss: 0.0793:  96%|█████████▌| 150/156 [15:08<00:36,  6.07s/it]
Training Epoch: 11/100, batch 299/313 completed (loss: 0.1101:  96%|█████████▌| 150/156 [15:08<00:36,  6.07s/it]
Training Epoch: 11/100, batch 299/313 completed (loss: 0.1171:  96%|█████████▌| 150/156 [15:08<00:36,  6.08s/it]
Training Epoch: 11/100, batch 299/313 completed (loss: 0.5642:  96%|█████████▌| 150/156 [15:08<00:36,  6.07s/it]
Training Epoch: 11/100, batch 300/313 completed (loss: 0.2507:  96%|█████████▌| 150/156 [15:11<00:36,  6.07s/it]
Training Epoch: 11/100, batch 300/313 completed (loss: 0.0266:  96%|█████████▌| 150/156 [15:11<00:36,  6.07s/it]
Training Epoch: 11/100, batch 300/313 completed (loss: 0.1541:  96%|█████████▌| 150/156 [15:11<00:36,  6.08s/it]
Training Epoch: 11/100, batch 300/313 completed (loss: 0.0295:  96%|█████████▌| 150/156 [15:11<00:36,  6.07s/it]
Training Epoch: 11/100, batch 300/313 completed (loss: 0.1541:  97%|█████████▋| 151/156 [15:14<00:30,  6.11s/it]
Training Epoch: 11/100, batch 300/313 completed (loss: 0.2507:  97%|█████████▋| 151/156 [15:14<00:30,  6.11s/it]
Training Epoch: 11/100, batch 300/313 completed (loss: 0.0295:  97%|█████████▋| 151/156 [15:14<00:30,  6.11s/it]
Training Epoch: 11/100, batch 300/313 completed (loss: 0.0266:  97%|█████████▋| 151/156 [15:14<00:30,  6.11s/it]
Training Epoch: 11/100, batch 301/313 completed (loss: 0.0493:  97%|█████████▋| 151/156 [15:14<00:30,  6.11s/it]
Training Epoch: 11/100, batch 301/313 completed (loss: 0.2566:  97%|█████████▋| 151/156 [15:14<00:30,  6.11s/it]
Training Epoch: 11/100, batch 301/313 completed (loss: 0.02:  97%|█████████▋| 151/156 [15:14<00:30,  6.11s/it]  
Training Epoch: 11/100, batch 301/313 completed (loss: 0.1274:  97%|█████████▋| 151/156 [15:14<00:30,  6.11s/it]
Training Epoch: 11/100, batch 302/313 completed (loss: 0.052:  97%|█████████▋| 151/156 [15:17<00:30,  6.11s/it] 
Training Epoch: 11/100, batch 302/313 completed (loss: 0.2124:  97%|█████████▋| 151/156 [15:17<00:30,  6.11s/it]
Training Epoch: 11/100, batch 302/313 completed (loss: 0.1484:  97%|█████████▋| 151/156 [15:17<00:30,  6.11s/it]
Training Epoch: 11/100, batch 302/313 completed (loss: 0.0371:  97%|█████████▋| 151/156 [15:17<00:30,  6.11s/it]
Training Epoch: 11/100, batch 302/313 completed (loss: 0.0371:  97%|█████████▋| 152/156 [15:20<00:24,  6.07s/it]
Training Epoch: 11/100, batch 302/313 completed (loss: 0.052:  97%|█████████▋| 152/156 [15:20<00:24,  6.07s/it]
Training Epoch: 11/100, batch 302/313 completed (loss: 0.1484:  97%|█████████▋| 152/156 [15:20<00:24,  6.07s/it]
Training Epoch: 11/100, batch 302/313 completed (loss: 0.2124:  97%|█████████▋| 152/156 [15:20<00:24,  6.07s/it]
Training Epoch: 11/100, batch 303/313 completed (loss: 0.1735:  97%|█████████▋| 152/156 [15:20<00:24,  6.07s/it]
Training Epoch: 11/100, batch 303/313 completed (loss: 0.1335:  97%|█████████▋| 152/156 [15:20<00:24,  6.07s/it]
Training Epoch: 11/100, batch 303/313 completed (loss: 0.0607:  97%|█████████▋| 152/156 [15:20<00:24,  6.07s/it]
Training Epoch: 11/100, batch 303/313 completed (loss: 0.1982:  97%|█████████▋| 152/156 [15:20<00:24,  6.07s/it]
Training Epoch: 11/100, batch 304/313 completed (loss: 0.0539:  97%|█████████▋| 152/156 [15:23<00:24,  6.07s/it]
Training Epoch: 11/100, batch 304/313 completed (loss: 0.2355:  97%|█████████▋| 152/156 [15:23<00:24,  6.07s/it]
Training Epoch: 11/100, batch 304/313 completed (loss: 0.0551:  97%|█████████▋| 152/156 [15:23<00:24,  6.07s/it]
Training Epoch: 11/100, batch 304/313 completed (loss: 0.1008:  97%|█████████▋| 152/156 [15:23<00:24,  6.07s/it]
Training Epoch: 11/100, batch 304/313 completed (loss: 0.2355:  98%|█████████▊| 153/156 [15:26<00:18,  6.05s/it]
Training Epoch: 11/100, batch 304/313 completed (loss: 0.1008:  98%|█████████▊| 153/156 [15:26<00:18,  6.05s/it]
Training Epoch: 11/100, batch 304/313 completed (loss: 0.0539:  98%|█████████▊| 153/156 [15:26<00:18,  6.05s/it]
Training Epoch: 11/100, batch 304/313 completed (loss: 0.0551:  98%|█████████▊| 153/156 [15:26<00:18,  6.05s/it]
Training Epoch: 11/100, batch 305/313 completed (loss: 0.0459:  98%|█████████▊| 153/156 [15:26<00:18,  6.05s/it]
Training Epoch: 11/100, batch 305/313 completed (loss: 0.1689:  98%|█████████▊| 153/156 [15:26<00:18,  6.05s/it]
Training Epoch: 11/100, batch 305/313 completed (loss: 0.2222:  98%|█████████▊| 153/156 [15:26<00:18,  6.05s/it]
Training Epoch: 11/100, batch 305/313 completed (loss: 0.049:  98%|█████████▊| 153/156 [15:26<00:18,  6.05s/it] 
Training Epoch: 11/100, batch 306/313 completed (loss: 0.0863:  98%|█████████▊| 153/156 [15:29<00:18,  6.05s/it]
Training Epoch: 11/100, batch 306/313 completed (loss: 0.0473:  98%|█████████▊| 153/156 [15:29<00:18,  6.05s/it]
Training Epoch: 11/100, batch 306/313 completed (loss: 0.1751:  98%|█████████▊| 153/156 [15:29<00:18,  6.05s/it]
Training Epoch: 11/100, batch 306/313 completed (loss: 0.0514:  98%|█████████▊| 153/156 [15:29<00:18,  6.05s/it]
Training Epoch: 11/100, batch 306/313 completed (loss: 0.1751:  99%|█████████▊| 154/156 [15:32<00:12,  6.11s/it]
Training Epoch: 11/100, batch 306/313 completed (loss: 0.0514:  99%|█████████▊| 154/156 [15:32<00:12,  6.12s/it]
Training Epoch: 11/100, batch 306/313 completed (loss: 0.0473:  99%|█████████▊| 154/156 [15:32<00:12,  6.12s/it]
Training Epoch: 11/100, batch 306/313 completed (loss: 0.0863:  99%|█████████▊| 154/156 [15:32<00:12,  6.12s/it]
Training Epoch: 11/100, batch 307/313 completed (loss: 0.5007:  99%|█████████▊| 154/156 [15:33<00:12,  6.11s/it]
Training Epoch: 11/100, batch 307/313 completed (loss: 0.0427:  99%|█████████▊| 154/156 [15:33<00:12,  6.12s/it]
Training Epoch: 11/100, batch 307/313 completed (loss: 0.1737:  99%|█████████▊| 154/156 [15:33<00:12,  6.12s/it]
Training Epoch: 11/100, batch 307/313 completed (loss: 0.0252:  99%|█████████▊| 154/156 [15:33<00:12,  6.12s/it]
Training Epoch: 11/100, batch 308/313 completed (loss: 0.2675:  99%|█████████▊| 154/156 [15:36<00:12,  6.11s/it]
Training Epoch: 11/100, batch 308/313 completed (loss: 0.4671:  99%|█████████▊| 154/156 [15:36<00:12,  6.12s/it]
Training Epoch: 11/100, batch 308/313 completed (loss: 0.113:  99%|█████████▊| 154/156 [15:36<00:12,  6.12s/it] 
Training Epoch: 11/100, batch 308/313 completed (loss: 0.0603:  99%|█████████▊| 154/156 [15:36<00:12,  6.12s/it]
Training Epoch: 11/100, batch 308/313 completed (loss: 0.0603:  99%|█████████▉| 155/156 [15:38<00:06,  6.09s/it]
Training Epoch: 11/100, batch 308/313 completed (loss: 0.2675:  99%|█████████▉| 155/156 [15:38<00:06,  6.09s/it]
Training Epoch: 11/100, batch 308/313 completed (loss: 0.113:  99%|█████████▉| 155/156 [15:38<00:06,  6.09s/it]
Training Epoch: 11/100, batch 308/313 completed (loss: 0.4671:  99%|█████████▉| 155/156 [15:38<00:06,  6.09s/it]
Training Epoch: 11/100, batch 309/313 completed (loss: 0.1125:  99%|█████████▉| 155/156 [15:39<00:06,  6.09s/it]
Training Epoch: 11/100, batch 309/313 completed (loss: 0.0546:  99%|█████████▉| 155/156 [15:39<00:06,  6.09s/it]
Training Epoch: 11/100, batch 309/313 completed (loss: 0.0752:  99%|█████████▉| 155/156 [15:39<00:06,  6.09s/it]
Training Epoch: 11/100, batch 309/313 completed (loss: 0.2969:  99%|█████████▉| 155/156 [15:39<00:06,  6.09s/it]
Training Epoch: 11/100, batch 310/313 completed (loss: 0.0158:  99%|█████████▉| 155/156 [15:41<00:06,  6.09s/it]
Training Epoch: 11/100, batch 310/313 completed (loss: 0.0207:  99%|█████████▉| 155/156 [15:41<00:06,  6.09s/it]
Training Epoch: 11/100, batch 310/313 completed (loss: 0.282:  99%|█████████▉| 155/156 [15:41<00:06,  6.09s/it] 
Training Epoch: 11/100, batch 310/313 completed (loss: 0.1257:  99%|█████████▉| 155/156 [15:41<00:06,  6.09s/it]
Training Epoch: 11/100, batch 310/313 completed (loss: 0.0158: 100%|██████████| 156/156 [15:44<00:00,  6.06s/it]
Training Epoch: 11/100, batch 310/313 completed (loss: 0.0207: 100%|██████████| 156/156 [15:44<00:00,  6.06s/it]
Training Epoch: 11/100, batch 310/313 completed (loss: 0.1257: 100%|██████████| 156/156 [15:44<00:00,  6.06s/it]
Training Epoch: 11/100, batch 310/313 completed (loss: 0.282: 100%|██████████| 156/156 [15:44<00:00,  6.06s/it]
Training Epoch: 11/100, batch 311/313 completed (loss: 0.5154: 100%|██████████| 156/156 [15:45<00:00,  6.06s/it]
Training Epoch: 11/100, batch 311/313 completed (loss: 0.068: 100%|██████████| 156/156 [15:45<00:00,  6.06s/it] 
Training Epoch: 11/100, batch 311/313 completed (loss: 0.373: 100%|██████████| 156/156 [15:45<00:00,  6.06s/it]
Training Epoch: 11/100, batch 311/313 completed (loss: 0.1403: 100%|██████████| 156/156 [15:45<00:00,  6.06s/it]
Training Epoch: 11/100, batch 311/313 completed (loss: 0.373: : 157it [15:47,  4.94s/it]                       
Training Epoch: 11/100, batch 311/313 completed (loss: 0.068: : 157it [15:47,  4.94s/it]                       
Training Epoch: 11/100, batch 311/313 completed (loss: 0.1403: : 157it [15:47,  4.94s/it]                       
Training Epoch: 11/100, batch 311/313 completed (loss: 0.5154: : 157it [15:47,  4.94s/it]                       
Training Epoch: 11/100, batch 312/313 completed (loss: 0.1554: : 157it [15:47,  4.94s/it]
Training Epoch: 11/100, batch 312/313 completed (loss: 0.0958: : 157it [15:47,  4.94s/it]
Training Epoch: 11/100, batch 312/313 completed (loss: 0.2008: : 157it [15:47,  4.94s/it]
Training Epoch: 11/100, batch 312/313 completed (loss: 0.0049: : 157it [15:47,  4.94s/it]
Training Epoch: 11/100, batch 312/313 completed (loss: 0.2008: : 157it [15:47,  6.03s/it]

Training Epoch: 11/100, batch 312/313 completed (loss: 0.0958: : 157it [15:47,  6.03s/it]

Training Epoch: 11/100, batch 312/313 completed (loss: 0.1554: : 157it [15:47,  6.03s/it]

Training Epoch: 11/100, batch 312/313 completed (loss: 0.0049: : 157it [15:47,  6.04s/it]
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/c10d_logger.py:83: UserWarning: barrier(): using the device under current context. You can specify `device_id` in `init_process_group` to mute this warning.
  return func(*args, **kwargs)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
saving model + train state.
Warning: failed to upload checkpoint_11 to HF: (Request ID: Root=1-6a1b7fe4-0b29783635ad7e4022d98452;49d4bba8-ee60-44b6-be73-8897721fc992)

403 Forbidden: You need to setup automatic credit recharge in order to upload more data. You can do so at /settings/billing..
Cannot access content at: https://huggingface.co/seyedparsa/thoughtformer-checkpoints.git/info/lfs/objects/batch.
Make sure your token has the correct permissions.
Warning: failed to upload train_state_11 to HF: (Request ID: Root=1-6a1b7fe8-26b1317d24d3c1b52f8e4c21;bba9a65e-bf56-444d-97b5-24503f8c3a6a)

403 Forbidden: You need to setup automatic credit recharge in order to upload more data. You can do so at /settings/billing..
Cannot access content at: https://huggingface.co/seyedparsa/thoughtformer-checkpoints.git/info/lfs/objects/batch.
Make sure your token has the correct permissions.
Deleted old checkpoint: train_state_10
Deleted old checkpoint: checkpoint_10
[warn] HF upload failed for ckpt_11; kept locally, earlier ckpts deleted.

Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]
Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]
Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]eval loss 0.0605329415993765

Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]
[CORRECT] Example 0 (diff=3)
  Expected : 'Claude'
  Extracted: 'Claude'
  Generated: '### Claude'

Test Accuracy:   3%|▎         | 1/32 [00:06<03:24,  6.58s/it]
Test Accuracy:   3%|▎         | 1/32 [00:06<03:24,  6.58s/it]
Test accuracy: 0.88:   3%|▎         | 1/32 [00:06<03:24,  6.58s/it]
[WRONG] Example 1 (diff=8)

Test accuracy: 1.0:   3%|▎         | 1/32 [00:06<03:24,  6.58s/it]  Expected : 'Guy'
  Extracted: 'Irma'
  Generated: '### Irma'

Test Accuracy:   3%|▎         | 1/32 [00:06<03:24,  6.58s/it]
Test accuracy: 0.62:   3%|▎         | 1/32 [00:06<03:24,  6.58s/it]
[CORRECT] Example 2 (diff=10)
  Expected : 'Elvis'
  Extracted: 'Elvis'
  Generated: '### Elvis'

[WRONG] Example 3 (diff=8)
  Expected : 'Coral'
  Extracted: 'Claude'
  Generated: '### Claude'

[CORRECT] Example 4 (diff=1)
  Expected : 'Patricia'
  Extracted: 'Patricia'
  Generated: '### Patricia'

Test Accuracy:   3%|▎         | 1/32 [00:06<03:19,  6.42s/it]
Test accuracy: 0.75:   3%|▎         | 1/32 [00:06<03:19,  6.42s/it]
Test accuracy: 1.0:   6%|▋         | 2/32 [00:12<03:09,  6.33s/it]
Test accuracy: 0.88:   6%|▋         | 2/32 [00:12<03:09,  6.33s/it]
Test accuracy: 0.62:   6%|▋         | 2/32 [00:12<03:09,  6.33s/it]
Test accuracy: 0.94:   6%|▋         | 2/32 [00:12<03:09,  6.33s/it]
Test accuracy: 0.81:   6%|▋         | 2/32 [00:12<03:09,  6.33s/it]
Test accuracy: 0.56:   6%|▋         | 2/32 [00:12<03:09,  6.33s/it]
Test accuracy: 0.75:   6%|▋         | 2/32 [00:12<03:07,  6.26s/it]
Test accuracy: 0.75:   6%|▋         | 2/32 [00:12<03:07,  6.26s/it]
Test accuracy: 0.56:   9%|▉         | 3/32 [00:19<03:15,  6.73s/it]
Test accuracy: 0.94:   9%|▉         | 3/32 [00:19<03:15,  6.73s/it]
Test accuracy: 0.58:   9%|▉         | 3/32 [00:19<03:15,  6.73s/it]
Test accuracy: 0.88:   9%|▉         | 3/32 [00:19<03:15,  6.73s/it]
Test accuracy: 0.75:   9%|▉         | 3/32 [00:19<03:14,  6.69s/it]
Test accuracy: 0.71:   9%|▉         | 3/32 [00:19<03:14,  6.69s/it]
Test accuracy: 0.81:   9%|▉         | 3/32 [00:19<03:15,  6.74s/it]
Test accuracy: 0.79:   9%|▉         | 3/32 [00:19<03:15,  6.74s/it]
Test accuracy: 0.58:  12%|█▎        | 4/32 [00:26<03:01,  6.50s/it]
Test accuracy: 0.79:  12%|█▎        | 4/32 [00:26<03:01,  6.50s/it]
Test accuracy: 0.88:  12%|█▎        | 4/32 [00:26<03:01,  6.50s/it]
Test accuracy: 0.59:  12%|█▎        | 4/32 [00:26<03:01,  6.50s/it]
Test accuracy: 0.78:  12%|█▎        | 4/32 [00:26<03:01,  6.50s/it]
Test accuracy: 0.75:  12%|█▎        | 4/32 [00:26<03:01,  6.50s/it]
Test accuracy: 0.71:  12%|█▎        | 4/32 [00:25<03:01,  6.48s/it]
Test accuracy: 0.78:  12%|█▎        | 4/32 [00:25<03:01,  6.48s/it]
Test accuracy: 0.75:  16%|█▌        | 5/32 [00:32<02:53,  6.44s/it]
Test accuracy: 0.59:  16%|█▌        | 5/32 [00:32<02:53,  6.44s/it]
Test accuracy: 0.78:  16%|█▌        | 5/32 [00:32<02:53,  6.44s/it]
Test accuracy: 0.75:  16%|█▌        | 5/32 [00:32<02:53,  6.44s/it]
Test accuracy: 0.6:  16%|█▌        | 5/32 [00:32<02:53,  6.44s/it] 
Test accuracy: 0.75:  16%|█▌        | 5/32 [00:32<02:53,  6.44s/it]
Test accuracy: 0.78:  16%|█▌        | 5/32 [00:32<02:53,  6.43s/it]
Test accuracy: 0.77:  16%|█▌        | 5/32 [00:32<02:53,  6.43s/it]
Test accuracy: 0.75:  19%|█▉        | 6/32 [00:38<02:46,  6.40s/it]
Test accuracy: 0.6:  19%|█▉        | 6/32 [00:38<02:46,  6.40s/it]
Test accuracy: 0.75:  19%|█▉        | 6/32 [00:38<02:46,  6.40s/it]
Test accuracy: 0.62:  19%|█▉        | 6/32 [00:38<02:46,  6.40s/it]
Test accuracy: 0.75:  19%|█▉        | 6/32 [00:38<02:46,  6.40s/it]
Test accuracy: 0.77:  19%|█▉        | 6/32 [00:38<02:46,  6.39s/it]
Test accuracy: 0.79:  19%|█▉        | 6/32 [00:38<02:46,  6.40s/it]
Test accuracy: 0.81:  19%|█▉        | 6/32 [00:38<02:46,  6.39s/it]
Test accuracy: 0.62:  22%|██▏       | 7/32 [00:44<02:37,  6.31s/it]
Test accuracy: 0.79:  22%|██▏       | 7/32 [00:44<02:37,  6.31s/it]
Test accuracy: 0.75:  22%|██▏       | 7/32 [00:44<02:37,  6.31s/it]
Test accuracy: 0.66:  22%|██▏       | 7/32 [00:44<02:37,  6.31s/it]
Test accuracy: 0.8:  22%|██▏       | 7/32 [00:44<02:37,  6.31s/it] 
Test accuracy: 0.71:  22%|██▏       | 7/32 [00:44<02:37,  6.31s/it]
Test accuracy: 0.81:  22%|██▏       | 7/32 [00:44<02:37,  6.30s/it]
Test accuracy: 0.82:  22%|██▏       | 7/32 [00:44<02:37,  6.30s/it]
Test accuracy: 0.66:  25%|██▌       | 8/32 [00:51<02:31,  6.32s/it]
Test accuracy: 0.71:  25%|██▌       | 8/32 [00:51<02:31,  6.32s/it]
Test accuracy: 0.8:  25%|██▌       | 8/32 [00:51<02:31,  6.32s/it]
Test accuracy: 0.66:  25%|██▌       | 8/32 [00:51<02:31,  6.32s/it]
Test accuracy: 0.75:  25%|██▌       | 8/32 [00:51<02:31,  6.32s/it]
Test accuracy: 0.83:  25%|██▌       | 8/32 [00:51<02:31,  6.32s/it]
Test accuracy: 0.82:  25%|██▌       | 8/32 [00:51<02:31,  6.31s/it]
Test accuracy: 0.83:  25%|██▌       | 8/32 [00:51<02:31,  6.31s/it]
Test accuracy: 0.83:  28%|██▊       | 9/32 [00:58<02:29,  6.51s/it]
Test accuracy: 0.75:  28%|██▊       | 9/32 [00:58<02:29,  6.51s/it]
Test accuracy: 0.75:  28%|██▊       | 9/32 [00:58<02:29,  6.51s/it]
Test accuracy: 0.82:  28%|██▊       | 9/32 [00:58<02:29,  6.51s/it]
Test accuracy: 0.83:  28%|██▊       | 9/32 [00:57<02:29,  6.51s/it]
Test accuracy: 0.85:  28%|██▊       | 9/32 [00:57<02:29,  6.51s/it]
Test accuracy: 0.66:  28%|██▊       | 9/32 [00:58<02:29,  6.52s/it]
Test accuracy: 0.68:  28%|██▊       | 9/32 [00:58<02:29,  6.52s/it]
Test accuracy: 0.68:  31%|███▏      | 10/32 [01:04<02:20,  6.40s/it]
Test accuracy: 0.75:  31%|███▏      | 10/32 [01:04<02:20,  6.40s/it]
Test accuracy: 0.77:  31%|███▏      | 10/32 [01:04<02:20,  6.40s/it]
Test accuracy: 0.68:  31%|███▏      | 10/32 [01:04<02:20,  6.40s/it]
Test accuracy: 0.82:  31%|███▏      | 10/32 [01:04<02:20,  6.40s/it]
Test accuracy: 0.85:  31%|███▏      | 10/32 [01:04<02:20,  6.40s/it]
Test accuracy: 0.79:  31%|███▏      | 10/32 [01:04<02:20,  6.40s/it]
Test accuracy: 0.82:  31%|███▏      | 10/32 [01:04<02:20,  6.40s/it]
Test accuracy: 0.79:  34%|███▍      | 11/32 [01:10<02:13,  6.38s/it]
Test accuracy: 0.68:  34%|███▍      | 11/32 [01:10<02:13,  6.38s/it]
Test accuracy: 0.77:  34%|███▍      | 11/32 [01:10<02:13,  6.38s/it]
Test accuracy: 0.77:  34%|███▍      | 11/32 [01:10<02:13,  6.38s/it]
Test accuracy: 0.7:  34%|███▍      | 11/32 [01:10<02:13,  6.38s/it] 
Test accuracy: 0.75:  34%|███▍      | 11/32 [01:10<02:13,  6.38s/it]
Test accuracy: 0.82:  34%|███▍      | 11/32 [01:10<02:13,  6.38s/it]
Test accuracy: 0.82:  34%|███▍      | 11/32 [01:10<02:13,  6.38s/it]
Test accuracy: 0.75:  38%|███▊      | 12/32 [01:16<02:07,  6.36s/it]
Test accuracy: 0.77:  38%|███▊      | 12/32 [01:16<02:07,  6.36s/it]
Test accuracy: 0.7:  38%|███▊      | 12/32 [01:16<02:07,  6.36s/it]
Test accuracy: 0.82:  38%|███▊      | 12/32 [01:16<02:07,  6.36s/it]
Test accuracy: 0.73:  38%|███▊      | 12/32 [01:16<02:07,  6.36s/it]
Test accuracy: 0.74:  38%|███▊      | 12/32 [01:16<02:07,  6.36s/it]
Test accuracy: 0.72:  38%|███▊      | 12/32 [01:16<02:07,  6.36s/it]
Test accuracy: 0.8:  38%|███▊      | 12/32 [01:16<02:07,  6.36s/it] 
Test accuracy: 0.74:  41%|████      | 13/32 [01:23<02:00,  6.35s/it]
Test accuracy: 0.72:  41%|████      | 13/32 [01:23<02:00,  6.35s/it]
Test accuracy: 0.73:  41%|████      | 13/32 [01:23<02:00,  6.35s/it]
Test accuracy: 0.74:  41%|████      | 13/32 [01:23<02:00,  6.35s/it]
Test accuracy: 0.73:  41%|████      | 13/32 [01:23<02:00,  6.35s/it]
Test accuracy: 0.75:  41%|████      | 13/32 [01:23<02:00,  6.35s/it]
Test accuracy: 0.8:  41%|████      | 13/32 [01:23<02:00,  6.35s/it]
Test accuracy: 0.78:  41%|████      | 13/32 [01:23<02:00,  6.35s/it]
Test accuracy: 0.74:  44%|████▍     | 14/32 [01:29<01:53,  6.33s/it]
Test accuracy: 0.73:  44%|████▍     | 14/32 [01:29<01:53,  6.33s/it]
Test accuracy: 0.78:  44%|████▍     | 14/32 [01:29<01:53,  6.33s/it]
Test accuracy: 0.75:  44%|████▍     | 14/32 [01:29<01:53,  6.33s/it]
Test accuracy: 0.74:  44%|████▍     | 14/32 [01:29<01:53,  6.33s/it]
Test accuracy: 0.75:  44%|████▍     | 14/32 [01:29<01:53,  6.33s/it]
Test accuracy: 0.77:  44%|████▍     | 14/32 [01:29<01:53,  6.33s/it]
Test accuracy: 0.77:  44%|████▍     | 14/32 [01:29<01:53,  6.33s/it]
Test accuracy: 0.75:  47%|████▋     | 15/32 [01:35<01:47,  6.33s/it]
Test accuracy: 0.74:  47%|████▋     | 15/32 [01:35<01:47,  6.33s/it]
Test accuracy: 0.75:  47%|████▋     | 15/32 [01:35<01:47,  6.33s/it]
Test accuracy: 0.77:  47%|████▋     | 15/32 [01:35<01:47,  6.33s/it]
Test accuracy: 0.73:  47%|████▋     | 15/32 [01:35<01:47,  6.33s/it]
Test accuracy: 0.77:  47%|████▋     | 15/32 [01:35<01:47,  6.33s/it]
Test accuracy: 0.77:  47%|████▋     | 15/32 [01:35<01:47,  6.33s/it]
Test accuracy: 0.77:  47%|████▋     | 15/32 [01:35<01:47,  6.33s/it]
Test accuracy: 0.75:  50%|█████     | 16/32 [01:42<01:41,  6.33s/it]
Test accuracy: 0.73:  50%|█████     | 16/32 [01:42<01:41,  6.33s/it]
Test accuracy: 0.75:  50%|█████     | 16/32 [01:42<01:41,  6.33s/it]
Test accuracy: 0.72:  50%|█████     | 16/32 [01:42<01:41,  6.33s/it]
Test accuracy: 0.77:  50%|█████     | 16/32 [01:42<01:41,  6.33s/it]
Test accuracy: 0.77:  50%|█████     | 16/32 [01:42<01:41,  6.33s/it]
Test accuracy: 0.77:  50%|█████     | 16/32 [01:42<01:41,  6.33s/it]
Test accuracy: 0.77:  50%|█████     | 16/32 [01:42<01:41,  6.33s/it]
Test accuracy: 0.72:  53%|█████▎    | 17/32 [01:48<01:34,  6.33s/it]
Test accuracy: 0.77:  53%|█████▎    | 17/32 [01:48<01:34,  6.33s/it]
Test accuracy: 0.75:  53%|█████▎    | 17/32 [01:48<01:34,  6.33s/it]
Test accuracy: 0.71:  53%|█████▎    | 17/32 [01:48<01:34,  6.33s/it]
Test accuracy: 0.78:  53%|█████▎    | 17/32 [01:48<01:34,  6.33s/it]
Test accuracy: 0.77:  53%|█████▎    | 17/32 [01:48<01:34,  6.33s/it]
Test accuracy: 0.75:  53%|█████▎    | 17/32 [01:48<01:34,  6.33s/it]
Test accuracy: 0.76:  53%|█████▎    | 17/32 [01:48<01:34,  6.33s/it]
Test accuracy: 0.78:  56%|█████▋    | 18/32 [01:54<01:27,  6.27s/it]
Test accuracy: 0.77:  56%|█████▋    | 18/32 [01:54<01:27,  6.27s/it]
Test accuracy: 0.71:  56%|█████▋    | 18/32 [01:54<01:27,  6.27s/it]
Test accuracy: 0.75:  56%|█████▋    | 18/32 [01:54<01:27,  6.27s/it]
Test accuracy: 0.72:  56%|█████▋    | 18/32 [01:54<01:27,  6.27s/it]
Test accuracy: 0.75:  56%|█████▋    | 18/32 [01:54<01:27,  6.27s/it]
Test accuracy: 0.76:  56%|█████▋    | 18/32 [01:54<01:27,  6.27s/it]
Test accuracy: 0.76:  56%|█████▋    | 18/32 [01:54<01:27,  6.27s/it]
Test accuracy: 0.75:  59%|█████▉    | 19/32 [02:01<01:21,  6.29s/it]
Test accuracy: 0.77:  59%|█████▉    | 19/32 [02:01<01:21,  6.29s/it]
Test accuracy: 0.74:  59%|█████▉    | 19/32 [02:01<01:21,  6.29s/it]
Test accuracy: 0.72:  59%|█████▉    | 19/32 [02:01<01:21,  6.29s/it]
Test accuracy: 0.76:  59%|█████▉    | 19/32 [02:01<01:21,  6.29s/it]
Test accuracy: 0.72:  59%|█████▉    | 19/32 [02:01<01:21,  6.29s/it]
Test accuracy: 0.76:  59%|█████▉    | 19/32 [02:00<01:21,  6.29s/it]
Test accuracy: 0.78:  59%|█████▉    | 19/32 [02:00<01:21,  6.29s/it]
Test accuracy: 0.74:  62%|██████▎   | 20/32 [02:07<01:15,  6.30s/it]
Test accuracy: 0.72:  62%|██████▎   | 20/32 [02:07<01:15,  6.30s/it]
Test accuracy: 0.74:  62%|██████▎   | 20/32 [02:07<01:15,  6.30s/it]
Test accuracy: 0.72:  62%|██████▎   | 20/32 [02:07<01:15,  6.30s/it]
Test accuracy: 0.78:  62%|██████▎   | 20/32 [02:07<01:15,  6.30s/it]
Test accuracy: 0.77:  62%|██████▎   | 20/32 [02:07<01:15,  6.30s/it]
Test accuracy: 0.76:  62%|██████▎   | 20/32 [02:07<01:15,  6.30s/it]
Test accuracy: 0.76:  62%|██████▎   | 20/32 [02:07<01:15,  6.30s/it]
Test accuracy: 0.72:  66%|██████▌   | 21/32 [02:14<01:11,  6.47s/it]
Test accuracy: 0.74:  66%|██████▌   | 21/32 [02:14<01:11,  6.47s/it]
Test accuracy: 0.73:  66%|██████▌   | 21/32 [02:14<01:11,  6.47s/it]
Test accuracy: 0.74:  66%|██████▌   | 21/32 [02:14<01:11,  6.47s/it]
Test accuracy: 0.77:  66%|██████▌   | 21/32 [02:14<01:11,  6.47s/it]
Test accuracy: 0.77:  66%|██████▌   | 21/32 [02:14<01:11,  6.47s/it]
Test accuracy: 0.76:  66%|██████▌   | 21/32 [02:14<01:11,  6.47s/it]
Test accuracy: 0.77:  66%|██████▌   | 21/32 [02:14<01:11,  6.47s/it]
Test accuracy: 0.73:  69%|██████▉   | 22/32 [02:20<01:04,  6.42s/it]
Test accuracy: 0.74:  69%|██████▉   | 22/32 [02:20<01:04,  6.42s/it]
Test accuracy: 0.76:  69%|██████▉   | 22/32 [02:20<01:04,  6.42s/it]
Test accuracy: 0.72:  69%|██████▉   | 22/32 [02:20<01:04,  6.42s/it]
Test accuracy: 0.77:  69%|██████▉   | 22/32 [02:20<01:04,  6.42s/it]
Test accuracy: 0.78:  69%|██████▉   | 22/32 [02:20<01:04,  6.42s/it]
Test accuracy: 0.77:  69%|██████▉   | 22/32 [02:20<01:04,  6.42s/it]
Test accuracy: 0.77:  69%|██████▉   | 22/32 [02:20<01:04,  6.42s/it]
Test accuracy: 0.72:  72%|███████▏  | 23/32 [02:26<00:57,  6.39s/it]
Test accuracy: 0.77:  72%|███████▏  | 23/32 [02:26<00:57,  6.39s/it]
Test accuracy: 0.76:  72%|███████▏  | 23/32 [02:26<00:57,  6.39s/it]
Test accuracy: 0.73:  72%|███████▏  | 23/32 [02:26<00:57,  6.39s/it]
Test accuracy: 0.76:  72%|███████▏  | 23/32 [02:26<00:57,  6.39s/it]
Test accuracy: 0.76:  72%|███████▏  | 23/32 [02:26<00:57,  6.39s/it]
Test accuracy: 0.78:  72%|███████▏  | 23/32 [02:26<00:57,  6.40s/it]
Test accuracy: 0.78:  72%|███████▏  | 23/32 [02:26<00:57,  6.40s/it]
Test accuracy: 0.76:  75%|███████▌  | 24/32 [02:33<00:52,  6.59s/it]
Test accuracy: 0.76:  75%|███████▌  | 24/32 [02:33<00:52,  6.59s/it]
Test accuracy: 0.73:  75%|███████▌  | 24/32 [02:33<00:52,  6.59s/it]
Test accuracy: 0.76:  75%|███████▌  | 24/32 [02:33<00:52,  6.59s/it]
Test accuracy: 0.76:  75%|███████▌  | 24/32 [02:33<00:52,  6.59s/it]
Test accuracy: 0.73:  75%|███████▌  | 24/32 [02:33<00:52,  6.59s/it]
Test accuracy: 0.78:  75%|███████▌  | 24/32 [02:33<00:52,  6.60s/it]
Test accuracy: 0.77:  75%|███████▌  | 24/32 [02:33<00:52,  6.60s/it]
Test accuracy: 0.76:  78%|███████▊  | 25/32 [02:40<00:45,  6.46s/it]
Test accuracy: 0.75:  78%|███████▊  | 25/32 [02:40<00:45,  6.46s/it]
Test accuracy: 0.73:  78%|███████▊  | 25/32 [02:40<00:45,  6.46s/it]
Test accuracy: 0.73:  78%|███████▊  | 25/32 [02:40<00:45,  6.46s/it]
Test accuracy: 0.76:  78%|███████▊  | 25/32 [02:40<00:45,  6.46s/it]
Test accuracy: 0.75:  78%|███████▊  | 25/32 [02:40<00:45,  6.46s/it]
Test accuracy: 0.77:  78%|███████▊  | 25/32 [02:39<00:45,  6.46s/it]
Test accuracy: 0.77:  78%|███████▊  | 25/32 [02:39<00:45,  6.46s/it]
Test accuracy: 0.75:  81%|████████▏ | 26/32 [02:46<00:38,  6.43s/it]
Test accuracy: 0.73:  81%|████████▏ | 26/32 [02:46<00:38,  6.43s/it]
Test accuracy: 0.73:  81%|████████▏ | 26/32 [02:46<00:38,  6.43s/it]
Test accuracy: 0.75:  81%|████████▏ | 26/32 [02:46<00:38,  6.43s/it]
Test accuracy: 0.77:  81%|████████▏ | 26/32 [02:46<00:38,  6.43s/it]
Test accuracy: 0.77:  81%|████████▏ | 26/32 [02:46<00:38,  6.43s/it]
Test accuracy: 0.75:  81%|████████▏ | 26/32 [02:46<00:38,  6.43s/it]
Test accuracy: 0.75:  81%|████████▏ | 26/32 [02:46<00:38,  6.43s/it]
Test accuracy: 0.75:  84%|████████▍ | 27/32 [02:52<00:32,  6.41s/it]
Test accuracy: 0.73:  84%|████████▍ | 27/32 [02:52<00:32,  6.41s/it]
Test accuracy: 0.75:  84%|████████▍ | 27/32 [02:52<00:32,  6.41s/it]
Test accuracy: 0.73:  84%|████████▍ | 27/32 [02:52<00:32,  6.41s/it]
Test accuracy: 0.75:  84%|████████▍ | 27/32 [02:52<00:32,  6.41s/it]
Test accuracy: 0.75:  84%|████████▍ | 27/32 [02:52<00:32,  6.41s/it]
Test accuracy: 0.77:  84%|████████▍ | 27/32 [02:52<00:32,  6.41s/it]
Test accuracy: 0.77:  84%|████████▍ | 27/32 [02:52<00:32,  6.41s/it]
Test accuracy: 0.73:  88%|████████▊ | 28/32 [02:59<00:26,  6.54s/it]
Test accuracy: 0.75:  88%|████████▊ | 28/32 [02:59<00:26,  6.54s/it]
Test accuracy: 0.73:  88%|████████▊ | 28/32 [02:59<00:26,  6.54s/it]
Test accuracy: 0.74:  88%|████████▊ | 28/32 [02:59<00:26,  6.54s/it]
Test accuracy: 0.77:  88%|████████▊ | 28/32 [02:59<00:26,  6.54s/it]
Test accuracy: 0.77:  88%|████████▊ | 28/32 [02:59<00:26,  6.54s/it]
Test accuracy: 0.75:  88%|████████▊ | 28/32 [02:59<00:26,  6.54s/it]
Test accuracy: 0.75:  88%|████████▊ | 28/32 [02:59<00:26,  6.54s/it]
Test accuracy: 0.73:  91%|█████████ | 29/32 [03:05<00:19,  6.48s/it]
Test accuracy: 0.74:  91%|█████████ | 29/32 [03:05<00:19,  6.48s/it]
Test accuracy: 0.74:  91%|█████████ | 29/32 [03:05<00:19,  6.48s/it]
Test accuracy: 0.77:  91%|█████████ | 29/32 [03:05<00:19,  6.48s/it]
Test accuracy: 0.74:  91%|█████████ | 29/32 [03:05<00:19,  6.48s/it]
Test accuracy: 0.78:  91%|█████████ | 29/32 [03:05<00:19,  6.48s/it]
Test accuracy: 0.75:  91%|█████████ | 29/32 [03:05<00:19,  6.48s/it]
Test accuracy: 0.75:  91%|█████████ | 29/32 [03:05<00:19,  6.48s/it]
Test accuracy: 0.75:  94%|█████████▍| 30/32 [03:12<00:12,  6.44s/it]
Test accuracy: 0.74:  94%|█████████▍| 30/32 [03:12<00:12,  6.44s/it]
Test accuracy: 0.75:  94%|█████████▍| 30/32 [03:12<00:12,  6.44s/it]
Test accuracy: 0.75:  94%|█████████▍| 30/32 [03:12<00:12,  6.44s/it]
Test accuracy: 0.74:  94%|█████████▍| 30/32 [03:12<00:12,  6.44s/it]
Test accuracy: 0.73:  94%|█████████▍| 30/32 [03:12<00:12,  6.44s/it]
Test accuracy: 0.78:  94%|█████████▍| 30/32 [03:12<00:12,  6.44s/it]
Test accuracy: 0.78:  94%|█████████▍| 30/32 [03:12<00:12,  6.44s/it]
Test accuracy: 0.73:  97%|█████████▋| 31/32 [03:19<00:06,  6.56s/it]
Test accuracy: 0.75:  97%|█████████▋| 31/32 [03:19<00:06,  6.56s/it]
Test accuracy: 0.78:  97%|█████████▋| 31/32 [03:18<00:06,  6.56s/it]
Test accuracy: 0.74:  97%|█████████▋| 31/32 [03:19<00:06,  6.56s/it]
Test accuracy: 0.75:  97%|█████████▋| 31/32 [03:19<00:06,  6.56s/it]
Test accuracy: 0.78:  97%|█████████▋| 31/32 [03:18<00:06,  6.56s/it]
Test accuracy: 0.75:  97%|█████████▋| 31/32 [03:19<00:06,  6.57s/it]
Test accuracy: 0.75:  97%|█████████▋| 31/32 [03:19<00:06,  6.57s/it]
Test accuracy: 0.78: 100%|██████████| 32/32 [03:21<00:00,  5.27s/it]
Test accuracy: 0.78: 100%|██████████| 32/32 [03:21<00:00,  5.27s/it]
Test accuracy: 0.75: 100%|██████████| 32/32 [03:21<00:00,  5.27s/it]
Test accuracy: 0.75: 100%|██████████| 32/32 [03:21<00:00,  5.27s/it]
Test accuracy: 0.74: 100%|██████████| 32/32 [03:21<00:00,  5.27s/it]
Test accuracy: 0.74: 100%|██████████| 32/32 [03:21<00:00,  5.27s/it]
Test accuracy: 0.75: 100%|██████████| 32/32 [03:21<00:00,  5.27s/it]
Test accuracy: 0.74: 100%|██████████| 32/32 [03:21<00:00,  5.27s/it]
Test accuracy: 0.74: 100%|██████████| 32/32 [03:21<00:00,  6.30s/it]
Device 2: Cor=185, CoT=0, Total=250

Test accuracy: 0.75: 100%|██████████| 32/32 [03:21<00:00,  6.30s/it]
Device 3: Cor=188, CoT=0, Total=250

Test accuracy: 0.74: 100%|██████████| 32/32 [03:21<00:00,  6.30s/it]
Device 1: Cor=186, CoT=0, Total=250

Test accuracy: 0.78: 100%|██████████| 32/32 [03:21<00:00,  6.30s/it]
Device 0: Cor=194, CoT=0, Total=250

============================================================
EVALUATION RESULTS (Epoch 10)
============================================================
Config: data_curriculum=ON, thought_curriculum=OFF
thought_stage=10, data_stage=5, n_latent_tokens=10
Training on difficulties: 0-5

Overall accuracy: 753 / 1000 = 0.7530

Per-difficulty accuracy:
  diff=1: 97/100 = 0.9700 (trained)
  diff=2: 96/100 = 0.9600 (trained)
  diff=3: 98/100 = 0.9800 (trained)
  diff=4: 96/100 = 0.9600 (trained)
  diff=5: 91/100 = 0.9100 (trained)
  diff=6: 52/100 = 0.5200 (untrained)
  diff=7: 55/100 = 0.5500 (untrained)
  diff=8: 55/100 = 0.5500 (untrained)
  diff=9: 60/100 = 0.6000 (untrained)
  diff=10: 53/100 = 0.5300 (untrained)

Trained (diff 0-5): 478/500 = 0.9560
============================================================


*** ADAPTIVE STAGING: acc_trained=0.9560, acc_diff5=0.9100 >= 0.9 ***
*** Advancing to stage 6 ***

Warning: failed to upload checkpoint_11 to HF: (Request ID: Root=1-6a1b80cc-6eb663257bba71cc351fed23;6a30aa55-11f0-48db-85e1-1306f9c780ce)

403 Forbidden: You need to setup automatic credit recharge in order to upload more data. You can do so at /settings/billing..
Cannot access content at: https://huggingface.co/seyedparsa/thoughtformer-checkpoints.git/info/lfs/objects/batch.
Make sure your token has the correct permissions.

============================================================
EPOCH 11/100 (thought_stage=10, data_stage=6)
============================================================
  thought_stage=10, c_thought=1, max_difficulty=6

Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):  16%|█▌        | 156/1000 [00:00<00:00, 984.27 examples/s]
Map (num_proc=4):  16%|█▌        | 156/1000 [00:00<00:00, 948.65 examples/s]
Map (num_proc=4):  16%|█▌        | 156/1000 [00:00<00:00, 929.34 examples/s]
Map (num_proc=4):  82%|████████▏ | 815/1000 [00:00<00:00, 3559.38 examples/s]
Map (num_proc=4):  15%|█▌        | 152/1000 [00:00<00:01, 790.96 examples/s]
Map (num_proc=4):  81%|████████  | 812/1000 [00:00<00:00, 3449.03 examples/s]
Map (num_proc=4):  81%|████████▏ | 813/1000 [00:00<00:00, 2960.56 examples/s]
Map (num_proc=4):  81%|████████  | 809/1000 [00:00<00:00, 3050.66 examples/s]
Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2060.07 examples/s]

Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 1999.80 examples/s]

Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 1997.65 examples/s]

Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 1776.55 examples/s]

Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:04, 1916.24 examples/s]
Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:04, 1826.39 examples/s]
Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:04, 1855.87 examples/s]
Filter (num_proc=4):  20%|██        | 2000/10000 [00:00<00:02, 3483.46 examples/s]
Filter (num_proc=4):  20%|██        | 2000/10000 [00:00<00:02, 3408.29 examples/s]
Filter (num_proc=4):  20%|██        | 2000/10000 [00:00<00:02, 3435.69 examples/s]
Filter (num_proc=4):  30%|███       | 3000/10000 [00:00<00:01, 4921.51 examples/s]
Filter (num_proc=4):  30%|███       | 3000/10000 [00:00<00:01, 4746.78 examples/s]
Filter (num_proc=4):  30%|███       | 3000/10000 [00:00<00:01, 4688.66 examples/s]
Filter (num_proc=4):  50%|█████     | 5000/10000 [00:00<00:00, 8517.01 examples/s]
Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:05, 1720.70 examples/s]
Filter (num_proc=4):  50%|█████     | 5000/10000 [00:00<00:00, 7977.71 examples/s]
Filter (num_proc=4):  50%|█████     | 5000/10000 [00:00<00:00, 8041.12 examples/s]
Filter (num_proc=4):  75%|███████▌  | 7500/10000 [00:00<00:00, 11600.22 examples/s]
Filter (num_proc=4):  20%|██        | 2000/10000 [00:00<00:02, 3025.72 examples/s]
Filter (num_proc=4):  75%|███████▌  | 7500/10000 [00:01<00:00, 11529.86 examples/s]
Filter (num_proc=4):  70%|███████   | 7000/10000 [00:01<00:00, 10524.51 examples/s]
Filter (num_proc=4):  90%|█████████ | 9000/10000 [00:01<00:00, 12365.26 examples/s]
Filter (num_proc=4):  90%|█████████ | 9000/10000 [00:01<00:00, 11864.41 examples/s]
Filter (num_proc=4):  85%|████████▌ | 8500/10000 [00:01<00:00, 10908.40 examples/s]
Filter (num_proc=4):  40%|████      | 4000/10000 [00:00<00:01, 5711.95 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 11247.36 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 7407.42 examples/s]

Filter (num_proc=4):  65%|██████▌   | 6500/10000 [00:01<00:00, 8667.05 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 7152.10 examples/s]

Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 7102.18 examples/s] 

Filter (num_proc=4):  85%|████████▌ | 8500/10000 [00:01<00:00, 9189.44 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 8484.25 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 6122.34 examples/s]

Map (num_proc=4):   0%|          | 0/6000 [00:00<?, ? examples/s]
Map (num_proc=4):   2%|▏         | 138/6000 [00:00<00:35, 164.19 examples/s]
Map (num_proc=4):   9%|▉         | 551/6000 [00:00<00:07, 700.57 examples/s]
Map (num_proc=4):  14%|█▍        | 826/6000 [00:01<00:04, 1036.41 examples/s]
Map (num_proc=4):  18%|█▊        | 1102/6000 [00:01<00:04, 981.20 examples/s]
Map (num_proc=4):  29%|██▉       | 1725/6000 [00:01<00:02, 1715.67 examples/s]
Map (num_proc=4):  38%|███▊      | 2278/6000 [00:01<00:01, 2371.98 examples/s]
Map (num_proc=4):  44%|████▍     | 2638/6000 [00:02<00:02, 1483.52 examples/s]
Map (num_proc=4):  61%|██████▏   | 3689/6000 [00:02<00:00, 2591.85 examples/s]
Map (num_proc=4):  69%|██████▉   | 4169/6000 [00:02<00:00, 1946.68 examples/s]
Map (num_proc=4):  87%|████████▋ | 5236/6000 [00:02<00:00, 3072.59 examples/s]
Map (num_proc=4):  96%|█████████▋| 5783/6000 [00:03<00:00, 2179.18 examples/s]
Map (num_proc=4): 100%|██████████| 6000/6000 [00:03<00:00, 1500.57 examples/s]

Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 1073.59 examples/s]
Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 1392.48 examples/s]
Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 1362.71 examples/s]
Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 1336.66 examples/s]
Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 1994.70 examples/s]

Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2684.42 examples/s]

Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2641.95 examples/s]

Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2642.97 examples/s]

Map (num_proc=4):   0%|          | 0/600 [00:00<?, ? examples/s]
Map (num_proc=4):  23%|██▎       | 138/600 [00:00<00:00, 523.78 examples/s]
Map (num_proc=4):  48%|████▊     | 285/600 [00:00<00:00, 836.97 examples/s]
Map (num_proc=4):  75%|███████▌  | 450/600 [00:00<00:00, 1097.39 examples/s]
Map (num_proc=4): 100%|██████████| 600/600 [00:00<00:00, 792.82 examples/s] 

Training Epoch: 12:   0%|          | 0/187 [00:00<?, ?it/s]
Training Epoch: 12:   0%|          | 0/187 [00:00<?, ?it/s]
Training Epoch: 12:   0%|          | 0/187 [00:00<?, ?it/s]
Training Epoch: 12:   0%|          | 0/187 [00:00<?, ?it/s]logging training data
rank 2 | gpu 2 | alloc=4.20 GB | reserved=55.12 GB | max=41.98 GB
rank 0 | gpu 0 | alloc=4.20 GB | reserved=51.39 GB | max=41.99 GB
rank 1 | gpu 1 | alloc=4.20 GB | reserved=51.39 GB | max=41.98 GB
rank 3 | gpu 3 | alloc=4.20 GB | reserved=55.11 GB | max=41.99 GB

Training Epoch: 12/100, batch 0/375 completed (loss: 0.0503:   0%|          | 0/187 [00:03<?, ?it/s]
Training Epoch: 12/100, batch 0/375 completed (loss: 0.0736:   0%|          | 0/187 [00:03<?, ?it/s]
Training Epoch: 12/100, batch 0/375 completed (loss: 0.057:   0%|          | 0/187 [00:03<?, ?it/s]
Training Epoch: 12/100, batch 0/375 completed (loss: 0.0909:   0%|          | 0/187 [00:03<?, ?it/s]
Training Epoch: 12/100, batch 0/375 completed (loss: 0.0909:   1%|          | 1/187 [00:06<20:27,  6.60s/it]
Training Epoch: 12/100, batch 0/375 completed (loss: 0.057:   1%|          | 1/187 [00:06<20:25,  6.59s/it]
Training Epoch: 12/100, batch 0/375 completed (loss: 0.0503:   1%|          | 1/187 [00:06<20:25,  6.59s/it]
Training Epoch: 12/100, batch 0/375 completed (loss: 0.0736:   1%|          | 1/187 [00:06<20:25,  6.59s/it]
Training Epoch: 12/100, batch 1/375 completed (loss: 0.0613:   1%|          | 1/187 [00:06<20:25,  6.59s/it]
Training Epoch: 12/100, batch 1/375 completed (loss: 0.2855:   1%|          | 1/187 [00:06<20:25,  6.59s/it]
Training Epoch: 12/100, batch 1/375 completed (loss: 0.0876:   1%|          | 1/187 [00:06<20:25,  6.59s/it]
Training Epoch: 12/100, batch 1/375 completed (loss: 0.1248:   1%|          | 1/187 [00:06<20:27,  6.60s/it]
Training Epoch: 12/100, batch 2/375 completed (loss: 0.0414:   1%|          | 1/187 [00:09<20:25,  6.59s/it]
Training Epoch: 12/100, batch 2/375 completed (loss: 0.0198:   1%|          | 1/187 [00:09<20:25,  6.59s/it]
Training Epoch: 12/100, batch 2/375 completed (loss: 0.2635:   1%|          | 1/187 [00:09<20:25,  6.59s/it]
Training Epoch: 12/100, batch 2/375 completed (loss: 0.0185:   1%|          | 1/187 [00:09<20:27,  6.60s/it]
Training Epoch: 12/100, batch 2/375 completed (loss: 0.0185:   1%|          | 2/187 [00:12<18:32,  6.01s/it]
Training Epoch: 12/100, batch 2/375 completed (loss: 0.0414:   1%|          | 2/187 [00:12<18:29,  6.00s/it]
Training Epoch: 12/100, batch 2/375 completed (loss: 0.0198:   1%|          | 2/187 [00:12<18:30,  6.00s/it]
Training Epoch: 12/100, batch 2/375 completed (loss: 0.2635:   1%|          | 2/187 [00:12<18:31,  6.01s/it]
Training Epoch: 12/100, batch 3/375 completed (loss: 0.0557:   1%|          | 2/187 [00:12<18:31,  6.01s/it]
Training Epoch: 12/100, batch 3/375 completed (loss: 0.1143:   1%|          | 2/187 [00:12<18:30,  6.00s/it]
Training Epoch: 12/100, batch 3/375 completed (loss: 0.032:   1%|          | 2/187 [00:12<18:29,  6.00s/it] 
Training Epoch: 12/100, batch 3/375 completed (loss: 0.029:   1%|          | 2/187 [00:12<18:32,  6.01s/it] 
Training Epoch: 12/100, batch 4/375 completed (loss: 0.0331:   1%|          | 2/187 [00:15<18:30,  6.00s/it]
Training Epoch: 12/100, batch 4/375 completed (loss: 0.0234:   1%|          | 2/187 [00:15<18:29,  6.00s/it]
Training Epoch: 12/100, batch 4/375 completed (loss: 0.1124:   1%|          | 2/187 [00:15<18:31,  6.01s/it]
Training Epoch: 12/100, batch 4/375 completed (loss: 0.0848:   1%|          | 2/187 [00:15<18:32,  6.01s/it]
Training Epoch: 12/100, batch 4/375 completed (loss: 0.0234:   2%|▏         | 3/187 [00:17<18:06,  5.90s/it]
Training Epoch: 12/100, batch 4/375 completed (loss: 0.0848:   2%|▏         | 3/187 [00:17<18:08,  5.91s/it]
Training Epoch: 12/100, batch 4/375 completed (loss: 0.0331:   2%|▏         | 3/187 [00:17<18:06,  5.90s/it]
Training Epoch: 12/100, batch 4/375 completed (loss: 0.1124:   2%|▏         | 3/187 [00:17<18:06,  5.91s/it]
Training Epoch: 12/100, batch 5/375 completed (loss: 0.1164:   2%|▏         | 3/187 [00:18<18:06,  5.90s/it]
Training Epoch: 12/100, batch 5/375 completed (loss: 0.0954:   2%|▏         | 3/187 [00:18<18:06,  5.90s/it]
Training Epoch: 12/100, batch 5/375 completed (loss: 0.1069:   2%|▏         | 3/187 [00:18<18:06,  5.91s/it]
Training Epoch: 12/100, batch 5/375 completed (loss: 0.073:   2%|▏         | 3/187 [00:18<18:08,  5.91s/it] 
Training Epoch: 12/100, batch 6/375 completed (loss: 0.1094:   2%|▏         | 3/187 [00:20<18:06,  5.91s/it]
Training Epoch: 12/100, batch 6/375 completed (loss: 0.0118:   2%|▏         | 3/187 [00:20<18:06,  5.90s/it]
Training Epoch: 12/100, batch 6/375 completed (loss: 0.0345:   2%|▏         | 3/187 [00:20<18:06,  5.90s/it]
Training Epoch: 12/100, batch 6/375 completed (loss: 0.1025:   2%|▏         | 3/187 [00:21<18:08,  5.91s/it]
Training Epoch: 12/100, batch 6/375 completed (loss: 0.1025:   2%|▏         | 4/187 [00:23<18:08,  5.95s/it]
Training Epoch: 12/100, batch 6/375 completed (loss: 0.1094:   2%|▏         | 4/187 [00:23<18:07,  5.94s/it]
Training Epoch: 12/100, batch 6/375 completed (loss: 0.0345:   2%|▏         | 4/187 [00:23<18:07,  5.94s/it]
Training Epoch: 12/100, batch 6/375 completed (loss: 0.0118:   2%|▏         | 4/187 [00:23<18:08,  5.95s/it]
Training Epoch: 12/100, batch 7/375 completed (loss: 0.2247:   2%|▏         | 4/187 [00:24<18:08,  5.95s/it]
Training Epoch: 12/100, batch 7/375 completed (loss: 0.0824:   2%|▏         | 4/187 [00:24<18:07,  5.94s/it]
Training Epoch: 12/100, batch 7/375 completed (loss: 0.0538:   2%|▏         | 4/187 [00:24<18:07,  5.94s/it]
Training Epoch: 12/100, batch 7/375 completed (loss: 0.1198:   2%|▏         | 4/187 [00:24<18:08,  5.95s/it]
Training Epoch: 12/100, batch 8/375 completed (loss: 0.0823:   2%|▏         | 4/187 [00:26<18:08,  5.95s/it]
Training Epoch: 12/100, batch 8/375 completed (loss: 0.228:   2%|▏         | 4/187 [00:26<18:07,  5.94s/it] 
Training Epoch: 12/100, batch 8/375 completed (loss: 0.0512:   2%|▏         | 4/187 [00:26<18:07,  5.94s/it]
Training Epoch: 12/100, batch 8/375 completed (loss: 0.0275:   2%|▏         | 4/187 [00:26<18:08,  5.95s/it]
Training Epoch: 12/100, batch 8/375 completed (loss: 0.228:   3%|▎         | 5/187 [00:29<17:41,  5.83s/it]
Training Epoch: 12/100, batch 8/375 completed (loss: 0.0512:   3%|▎         | 5/187 [00:29<17:41,  5.83s/it]
Training Epoch: 12/100, batch 8/375 completed (loss: 0.0823:   3%|▎         | 5/187 [00:29<17:41,  5.83s/it]
Training Epoch: 12/100, batch 8/375 completed (loss: 0.0275:   3%|▎         | 5/187 [00:29<17:42,  5.84s/it]
Training Epoch: 12/100, batch 9/375 completed (loss: 0.0493:   3%|▎         | 5/187 [00:29<17:41,  5.83s/it]
Training Epoch: 12/100, batch 9/375 completed (loss: 0.061:   3%|▎         | 5/187 [00:29<17:41,  5.83s/it] 
Training Epoch: 12/100, batch 9/375 completed (loss: 0.0773:   3%|▎         | 5/187 [00:29<17:41,  5.83s/it]
Training Epoch: 12/100, batch 9/375 completed (loss: 0.0652:   3%|▎         | 5/187 [00:29<17:42,  5.84s/it]
Training Epoch: 12/100, batch 10/375 completed (loss: 0.05:   3%|▎         | 5/187 [00:32<17:41,  5.83s/it] 
Training Epoch: 12/100, batch 10/375 completed (loss: 0.021:   3%|▎         | 5/187 [00:32<17:41,  5.83s/it]
Training Epoch: 12/100, batch 10/375 completed (loss: 0.0361:   3%|▎         | 5/187 [00:32<17:41,  5.83s/it]
Training Epoch: 12/100, batch 10/375 completed (loss: 0.189:   3%|▎         | 5/187 [00:32<17:42,  5.84s/it]
Training Epoch: 12/100, batch 10/375 completed (loss: 0.189:   3%|▎         | 6/187 [00:35<17:48,  5.90s/it]
Training Epoch: 12/100, batch 10/375 completed (loss: 0.021:   3%|▎         | 6/187 [00:35<17:48,  5.90s/it]
Training Epoch: 12/100, batch 10/375 completed (loss: 0.05:   3%|▎         | 6/187 [00:35<17:48,  5.91s/it]
Training Epoch: 12/100, batch 10/375 completed (loss: 0.0361:   3%|▎         | 6/187 [00:35<17:48,  5.91s/it]
Training Epoch: 12/100, batch 11/375 completed (loss: 0.0617:   3%|▎         | 6/187 [00:35<17:48,  5.91s/it]
Training Epoch: 12/100, batch 11/375 completed (loss: 0.0866:   3%|▎         | 6/187 [00:35<17:48,  5.91s/it]
Training Epoch: 12/100, batch 11/375 completed (loss: 0.0117:   3%|▎         | 6/187 [00:35<17:48,  5.90s/it]
Training Epoch: 12/100, batch 11/375 completed (loss: 0.3382:   3%|▎         | 6/187 [00:35<17:48,  5.90s/it]
Training Epoch: 12/100, batch 12/375 completed (loss: 0.0988:   3%|▎         | 6/187 [00:38<17:48,  5.90s/it]
Training Epoch: 12/100, batch 12/375 completed (loss: 0.0458:   3%|▎         | 6/187 [00:38<17:48,  5.91s/it]
Training Epoch: 12/100, batch 12/375 completed (loss: 0.0345:   3%|▎         | 6/187 [00:38<17:48,  5.91s/it]
Training Epoch: 12/100, batch 12/375 completed (loss: 0.0717:   3%|▎         | 6/187 [00:38<17:48,  5.90s/it]
Training Epoch: 12/100, batch 12/375 completed (loss: 0.0458:   4%|▎         | 7/187 [00:41<17:41,  5.90s/it]
Training Epoch: 12/100, batch 12/375 completed (loss: 0.0345:   4%|▎         | 7/187 [00:41<17:41,  5.90s/it]
Training Epoch: 12/100, batch 12/375 completed (loss: 0.0717:   4%|▎         | 7/187 [00:41<17:42,  5.90s/it]
Training Epoch: 12/100, batch 12/375 completed (loss: 0.0988:   4%|▎         | 7/187 [00:41<17:42,  5.90s/it]
Training Epoch: 12/100, batch 13/375 completed (loss: 0.1061:   4%|▎         | 7/187 [00:41<17:41,  5.90s/it]
Training Epoch: 12/100, batch 13/375 completed (loss: 0.2867:   4%|▎         | 7/187 [00:41<17:42,  5.90s/it]
Training Epoch: 12/100, batch 13/375 completed (loss: 0.0813:   4%|▎         | 7/187 [00:41<17:41,  5.90s/it]
Training Epoch: 12/100, batch 13/375 completed (loss: 0.0256:   4%|▎         | 7/187 [00:41<17:42,  5.90s/it]
Training Epoch: 12/100, batch 14/375 completed (loss: 0.0172:   4%|▎         | 7/187 [00:44<17:42,  5.90s/it]
Training Epoch: 12/100, batch 14/375 completed (loss: 0.2086:   4%|▎         | 7/187 [00:44<17:41,  5.90s/it]
Training Epoch: 12/100, batch 14/375 completed (loss: 0.1411:   4%|▎         | 7/187 [00:44<17:41,  5.90s/it]
Training Epoch: 12/100, batch 14/375 completed (loss: 0.189:   4%|▎         | 7/187 [00:44<17:42,  5.90s/it] 
Training Epoch: 12/100, batch 14/375 completed (loss: 0.189:   4%|▍         | 8/187 [00:47<17:34,  5.89s/it]
Training Epoch: 12/100, batch 14/375 completed (loss: 0.1411:   4%|▍         | 8/187 [00:47<17:34,  5.89s/it]
Training Epoch: 12/100, batch 14/375 completed (loss: 0.2086:   4%|▍         | 8/187 [00:47<17:34,  5.89s/it]
Training Epoch: 12/100, batch 14/375 completed (loss: 0.0172:   4%|▍         | 8/187 [00:47<17:34,  5.89s/it]
Training Epoch: 12/100, batch 15/375 completed (loss: 0.0566:   4%|▍         | 8/187 [00:47<17:34,  5.89s/it]
Training Epoch: 12/100, batch 15/375 completed (loss: 0.0424:   4%|▍         | 8/187 [00:47<17:34,  5.89s/it]
Training Epoch: 12/100, batch 15/375 completed (loss: 0.1725:   4%|▍         | 8/187 [00:47<17:34,  5.89s/it]
Training Epoch: 12/100, batch 15/375 completed (loss: 0.0935:   4%|▍         | 8/187 [00:47<17:34,  5.89s/it]
Training Epoch: 12/100, batch 16/375 completed (loss: 0.0216:   4%|▍         | 8/187 [00:50<17:34,  5.89s/it]
Training Epoch: 12/100, batch 16/375 completed (loss: 0.0163:   4%|▍         | 8/187 [00:50<17:34,  5.89s/it]
Training Epoch: 12/100, batch 16/375 completed (loss: 0.1054:   4%|▍         | 8/187 [00:50<17:34,  5.89s/it]
Training Epoch: 12/100, batch 16/375 completed (loss: 0.0371:   4%|▍         | 8/187 [00:50<17:34,  5.89s/it]
Training Epoch: 12/100, batch 16/375 completed (loss: 0.0371:   5%|▍         | 9/187 [00:53<17:43,  5.97s/it]
Training Epoch: 12/100, batch 16/375 completed (loss: 0.1054:   5%|▍         | 9/187 [00:53<17:43,  5.97s/it]
Training Epoch: 12/100, batch 16/375 completed (loss: 0.0216:   5%|▍         | 9/187 [00:53<17:44,  5.98s/it]
Training Epoch: 12/100, batch 16/375 completed (loss: 0.0163:   5%|▍         | 9/187 [00:53<17:43,  5.98s/it]
Training Epoch: 12/100, batch 17/375 completed (loss: 0.2373:   5%|▍         | 9/187 [00:53<17:43,  5.97s/it]
Training Epoch: 12/100, batch 17/375 completed (loss: 0.1177:   5%|▍         | 9/187 [00:53<17:44,  5.98s/it]
Training Epoch: 12/100, batch 17/375 completed (loss: 0.0499:   5%|▍         | 9/187 [00:53<17:43,  5.98s/it]
Training Epoch: 12/100, batch 17/375 completed (loss: 0.0211:   5%|▍         | 9/187 [00:53<17:43,  5.97s/it]
Training Epoch: 12/100, batch 18/375 completed (loss: 0.0455:   5%|▍         | 9/187 [00:56<17:44,  5.98s/it]
Training Epoch: 12/100, batch 18/375 completed (loss: 0.1858:   5%|▍         | 9/187 [00:56<17:43,  5.98s/it]
Training Epoch: 12/100, batch 18/375 completed (loss: 0.0901:   5%|▍         | 9/187 [00:56<17:43,  5.97s/it]
Training Epoch: 12/100, batch 18/375 completed (loss: 0.019:   5%|▍         | 9/187 [00:56<17:43,  5.97s/it] 
Training Epoch: 12/100, batch 18/375 completed (loss: 0.019:   5%|▌         | 10/187 [00:59<17:23,  5.89s/it]
Training Epoch: 12/100, batch 18/375 completed (loss: 0.0901:   5%|▌         | 10/187 [00:59<17:22,  5.89s/it]
Training Epoch: 12/100, batch 18/375 completed (loss: 0.0455:   5%|▌         | 10/187 [00:59<17:22,  5.89s/it]
Training Epoch: 12/100, batch 18/375 completed (loss: 0.1858:   5%|▌         | 10/187 [00:59<17:22,  5.89s/it]
Training Epoch: 12/100, batch 19/375 completed (loss: 0.1286:   5%|▌         | 10/187 [00:59<17:22,  5.89s/it]
Training Epoch: 12/100, batch 19/375 completed (loss: 0.0718:   5%|▌         | 10/187 [00:59<17:22,  5.89s/it]
Training Epoch: 12/100, batch 19/375 completed (loss: 0.0321:   5%|▌         | 10/187 [00:59<17:22,  5.89s/it]
Training Epoch: 12/100, batch 19/375 completed (loss: 0.0418:   5%|▌         | 10/187 [00:59<17:23,  5.89s/it]
Training Epoch: 12/100, batch 20/375 completed (loss: 0.2985:   5%|▌         | 10/187 [01:02<17:22,  5.89s/it]
Training Epoch: 12/100, batch 20/375 completed (loss: 0.3886:   5%|▌         | 10/187 [01:02<17:22,  5.89s/it]
Training Epoch: 12/100, batch 20/375 completed (loss: 0.1331:   5%|▌         | 10/187 [01:02<17:22,  5.89s/it]
Training Epoch: 12/100, batch 20/375 completed (loss: 0.2566:   5%|▌         | 10/187 [01:02<17:23,  5.89s/it]
Training Epoch: 12/100, batch 20/375 completed (loss: 0.2566:   6%|▌         | 11/187 [01:05<17:15,  5.89s/it]
Training Epoch: 12/100, batch 20/375 completed (loss: 0.1331:   6%|▌         | 11/187 [01:05<17:15,  5.88s/it]
Training Epoch: 12/100, batch 20/375 completed (loss: 0.2985:   6%|▌         | 11/187 [01:05<17:15,  5.88s/it]
Training Epoch: 12/100, batch 20/375 completed (loss: 0.3886:   6%|▌         | 11/187 [01:05<17:15,  5.89s/it]
Training Epoch: 12/100, batch 21/375 completed (loss: 0.162:   6%|▌         | 11/187 [01:05<17:15,  5.88s/it] 
Training Epoch: 12/100, batch 21/375 completed (loss: 0.0874:   6%|▌         | 11/187 [01:05<17:15,  5.89s/it]
Training Epoch: 12/100, batch 21/375 completed (loss: 0.1883:   6%|▌         | 11/187 [01:05<17:15,  5.88s/it]
Training Epoch: 12/100, batch 21/375 completed (loss: 0.4687:   6%|▌         | 11/187 [01:05<17:15,  5.89s/it]
Training Epoch: 12/100, batch 22/375 completed (loss: 0.0224:   6%|▌         | 11/187 [01:08<17:15,  5.88s/it]
Training Epoch: 12/100, batch 22/375 completed (loss: 0.2286:   6%|▌         | 11/187 [01:08<17:15,  5.88s/it]
Training Epoch: 12/100, batch 22/375 completed (loss: 0.2576:   6%|▌         | 11/187 [01:08<17:15,  5.89s/it]
Training Epoch: 12/100, batch 22/375 completed (loss: 0.5835:   6%|▌         | 11/187 [01:08<17:15,  5.89s/it]
Training Epoch: 12/100, batch 22/375 completed (loss: 0.5835:   6%|▋         | 12/187 [01:11<17:24,  5.97s/it]
Training Epoch: 12/100, batch 22/375 completed (loss: 0.2576:   6%|▋         | 12/187 [01:11<17:24,  5.97s/it]
Training Epoch: 12/100, batch 22/375 completed (loss: 0.0224:   6%|▋         | 12/187 [01:11<17:24,  5.97s/it]
Training Epoch: 12/100, batch 22/375 completed (loss: 0.2286:   6%|▋         | 12/187 [01:11<17:24,  5.97s/it]
Training Epoch: 12/100, batch 23/375 completed (loss: 0.0467:   6%|▋         | 12/187 [01:11<17:24,  5.97s/it]
Training Epoch: 12/100, batch 23/375 completed (loss: 0.3235:   6%|▋         | 12/187 [01:11<17:24,  5.97s/it]
Training Epoch: 12/100, batch 23/375 completed (loss: 0.1496:   6%|▋         | 12/187 [01:11<17:24,  5.97s/it]
Training Epoch: 12/100, batch 23/375 completed (loss: 0.2473:   6%|▋         | 12/187 [01:11<17:24,  5.97s/it]
Training Epoch: 12/100, batch 24/375 completed (loss: 0.2269:   6%|▋         | 12/187 [01:14<17:24,  5.97s/it]
Training Epoch: 12/100, batch 24/375 completed (loss: 0.0944:   6%|▋         | 12/187 [01:14<17:24,  5.97s/it]
Training Epoch: 12/100, batch 24/375 completed (loss: 0.0354:   6%|▋         | 12/187 [01:14<17:24,  5.97s/it]
Training Epoch: 12/100, batch 24/375 completed (loss: 0.0751:   6%|▋         | 12/187 [01:14<17:24,  5.97s/it]
Training Epoch: 12/100, batch 24/375 completed (loss: 0.2269:   7%|▋         | 13/187 [01:17<17:05,  5.89s/it]
Training Epoch: 12/100, batch 24/375 completed (loss: 0.0944:   7%|▋         | 13/187 [01:17<17:05,  5.90s/it]
Training Epoch: 12/100, batch 24/375 completed (loss: 0.0354:   7%|▋         | 13/187 [01:17<17:05,  5.89s/it]
Training Epoch: 12/100, batch 24/375 completed (loss: 0.0751:   7%|▋         | 13/187 [01:17<17:06,  5.90s/it]
Training Epoch: 12/100, batch 25/375 completed (loss: 0.1596:   7%|▋         | 13/187 [01:17<17:05,  5.89s/it]
Training Epoch: 12/100, batch 25/375 completed (loss: 0.3362:   7%|▋         | 13/187 [01:17<17:05,  5.90s/it]
Training Epoch: 12/100, batch 25/375 completed (loss: 0.0695:   7%|▋         | 13/187 [01:17<17:05,  5.89s/it]
Training Epoch: 12/100, batch 25/375 completed (loss: 0.049:   7%|▋         | 13/187 [01:17<17:06,  5.90s/it] 
Training Epoch: 12/100, batch 26/375 completed (loss: 0.065:   7%|▋         | 13/187 [01:20<17:05,  5.90s/it] 
Training Epoch: 12/100, batch 26/375 completed (loss: 0.0386:   7%|▋         | 13/187 [01:20<17:05,  5.89s/it]
Training Epoch: 12/100, batch 26/375 completed (loss: 0.1037:   7%|▋         | 13/187 [01:20<17:05,  5.89s/it]
Training Epoch: 12/100, batch 26/375 completed (loss: 0.0845:   7%|▋         | 13/187 [01:20<17:06,  5.90s/it]
Training Epoch: 12/100, batch 26/375 completed (loss: 0.065:   7%|▋         | 14/187 [01:23<17:08,  5.95s/it]
Training Epoch: 12/100, batch 26/375 completed (loss: 0.0845:   7%|▋         | 14/187 [01:23<17:09,  5.95s/it]
Training Epoch: 12/100, batch 26/375 completed (loss: 0.1037:   7%|▋         | 14/187 [01:23<17:08,  5.95s/it]
Training Epoch: 12/100, batch 26/375 completed (loss: 0.0386:   7%|▋         | 14/187 [01:23<17:09,  5.95s/it]
Training Epoch: 12/100, batch 27/375 completed (loss: 0.0525:   7%|▋         | 14/187 [01:23<17:08,  5.95s/it]
Training Epoch: 12/100, batch 27/375 completed (loss: 0.2328:   7%|▋         | 14/187 [01:23<17:08,  5.95s/it]
Training Epoch: 12/100, batch 27/375 completed (loss: 0.1837:   7%|▋         | 14/187 [01:23<17:09,  5.95s/it]
Training Epoch: 12/100, batch 27/375 completed (loss: 0.052:   7%|▋         | 14/187 [01:23<17:09,  5.95s/it] 
Training Epoch: 12/100, batch 28/375 completed (loss: 0.0641:   7%|▋         | 14/187 [01:26<17:08,  5.95s/it]
Training Epoch: 12/100, batch 28/375 completed (loss: 0.4459:   7%|▋         | 14/187 [01:26<17:08,  5.95s/it]
Training Epoch: 12/100, batch 28/375 completed (loss: 0.0553:   7%|▋         | 14/187 [01:26<17:09,  5.95s/it]
Training Epoch: 12/100, batch 28/375 completed (loss: 0.0464:   7%|▋         | 14/187 [01:26<17:09,  5.95s/it]
Training Epoch: 12/100, batch 28/375 completed (loss: 0.0641:   8%|▊         | 15/187 [01:29<17:02,  5.94s/it]
Training Epoch: 12/100, batch 28/375 completed (loss: 0.0553:   8%|▊         | 15/187 [01:29<17:01,  5.94s/it]
Training Epoch: 12/100, batch 28/375 completed (loss: 0.4459:   8%|▊         | 15/187 [01:29<17:02,  5.94s/it]
Training Epoch: 12/100, batch 28/375 completed (loss: 0.0464:   8%|▊         | 15/187 [01:29<17:02,  5.94s/it]
Training Epoch: 12/100, batch 29/375 completed (loss: 0.2758:   8%|▊         | 15/187 [01:29<17:02,  5.94s/it]
Training Epoch: 12/100, batch 29/375 completed (loss: 0.2604:   8%|▊         | 15/187 [01:29<17:02,  5.94s/it]
Training Epoch: 12/100, batch 29/375 completed (loss: 0.1043:   8%|▊         | 15/187 [01:29<17:01,  5.94s/it]
Training Epoch: 12/100, batch 29/375 completed (loss: 0.0632:   8%|▊         | 15/187 [01:29<17:02,  5.94s/it]
Training Epoch: 12/100, batch 30/375 completed (loss: 0.1001:   8%|▊         | 15/187 [01:31<17:01,  5.94s/it]
Training Epoch: 12/100, batch 30/375 completed (loss: 0.0612:   8%|▊         | 15/187 [01:31<17:02,  5.94s/it]
Training Epoch: 12/100, batch 30/375 completed (loss: 0.0223:   8%|▊         | 15/187 [01:31<17:02,  5.94s/it]
Training Epoch: 12/100, batch 30/375 completed (loss: 0.0209:   8%|▊         | 15/187 [01:31<17:02,  5.94s/it]
Training Epoch: 12/100, batch 30/375 completed (loss: 0.0209:   9%|▊         | 16/187 [01:34<16:46,  5.89s/it]
Training Epoch: 12/100, batch 30/375 completed (loss: 0.1001:   9%|▊         | 16/187 [01:34<16:46,  5.89s/it]
Training Epoch: 12/100, batch 30/375 completed (loss: 0.0612:   9%|▊         | 16/187 [01:34<16:47,  5.89s/it]
Training Epoch: 12/100, batch 30/375 completed (loss: 0.0223:   9%|▊         | 16/187 [01:34<16:47,  5.89s/it]
Training Epoch: 12/100, batch 31/375 completed (loss: 0.0835:   9%|▊         | 16/187 [01:34<16:47,  5.89s/it]
Training Epoch: 12/100, batch 31/375 completed (loss: 0.0479:   9%|▊         | 16/187 [01:34<16:47,  5.89s/it]
Training Epoch: 12/100, batch 31/375 completed (loss: 0.1467:   9%|▊         | 16/187 [01:34<16:46,  5.89s/it]
Training Epoch: 12/100, batch 31/375 completed (loss: 0.0146:   9%|▊         | 16/187 [01:34<16:46,  5.89s/it]
Training Epoch: 12/100, batch 32/375 completed (loss: 0.0237:   9%|▊         | 16/187 [01:38<16:46,  5.89s/it]
Training Epoch: 12/100, batch 32/375 completed (loss: 0.4573:   9%|▊         | 16/187 [01:38<16:47,  5.89s/it]
Training Epoch: 12/100, batch 32/375 completed (loss: 0.0133:   9%|▊         | 16/187 [01:38<16:47,  5.89s/it]
Training Epoch: 12/100, batch 32/375 completed (loss: 0.3471:   9%|▊         | 16/187 [01:38<16:46,  5.89s/it]
Training Epoch: 12/100, batch 32/375 completed (loss: 0.4573:   9%|▉         | 17/187 [01:41<17:02,  6.02s/it]
Training Epoch: 12/100, batch 32/375 completed (loss: 0.0237:   9%|▉         | 17/187 [01:41<17:02,  6.02s/it]
Training Epoch: 12/100, batch 32/375 completed (loss: 0.0133:   9%|▉         | 17/187 [01:41<17:02,  6.02s/it]
Training Epoch: 12/100, batch 32/375 completed (loss: 0.3471:   9%|▉         | 17/187 [01:41<17:03,  6.02s/it]
Training Epoch: 12/100, batch 33/375 completed (loss: 0.0544:   9%|▉         | 17/187 [01:41<17:02,  6.02s/it]
Training Epoch: 12/100, batch 33/375 completed (loss: 0.0945:   9%|▉         | 17/187 [01:41<17:02,  6.02s/it]
Training Epoch: 12/100, batch 33/375 completed (loss: 0.0843:   9%|▉         | 17/187 [01:41<17:02,  6.02s/it]
Training Epoch: 12/100, batch 33/375 completed (loss: 0.1086:   9%|▉         | 17/187 [01:41<17:03,  6.02s/it]
Training Epoch: 12/100, batch 34/375 completed (loss: 0.0182:   9%|▉         | 17/187 [01:44<17:02,  6.02s/it]
Training Epoch: 12/100, batch 34/375 completed (loss: 0.1399:   9%|▉         | 17/187 [01:44<17:02,  6.02s/it]
Training Epoch: 12/100, batch 34/375 completed (loss: 0.0167:   9%|▉         | 17/187 [01:44<17:02,  6.02s/it]
Training Epoch: 12/100, batch 34/375 completed (loss: 0.065:   9%|▉         | 17/187 [01:44<17:03,  6.02s/it] 
Training Epoch: 12/100, batch 34/375 completed (loss: 0.065:  10%|▉         | 18/187 [01:46<16:43,  5.94s/it]
Training Epoch: 12/100, batch 34/375 completed (loss: 0.1399:  10%|▉         | 18/187 [01:46<16:43,  5.94s/it]
Training Epoch: 12/100, batch 34/375 completed (loss: 0.0167:  10%|▉         | 18/187 [01:46<16:43,  5.94s/it]
Training Epoch: 12/100, batch 34/375 completed (loss: 0.0182:  10%|▉         | 18/187 [01:46<16:44,  5.94s/it]
Training Epoch: 12/100, batch 35/375 completed (loss: 0.0949:  10%|▉         | 18/187 [01:46<16:43,  5.94s/it]
Training Epoch: 12/100, batch 35/375 completed (loss: 0.0173:  10%|▉         | 18/187 [01:46<16:43,  5.94s/it]
Training Epoch: 12/100, batch 35/375 completed (loss: 0.0578:  10%|▉         | 18/187 [01:46<16:44,  5.94s/it]
Training Epoch: 12/100, batch 35/375 completed (loss: 0.1153:  10%|▉         | 18/187 [01:46<16:43,  5.94s/it]
Training Epoch: 12/100, batch 36/375 completed (loss: 0.1895:  10%|▉         | 18/187 [01:49<16:43,  5.94s/it]
Training Epoch: 12/100, batch 36/375 completed (loss: 0.3134:  10%|▉         | 18/187 [01:49<16:43,  5.94s/it]
Training Epoch: 12/100, batch 36/375 completed (loss: 0.0688:  10%|▉         | 18/187 [01:49<16:44,  5.94s/it]
Training Epoch: 12/100, batch 36/375 completed (loss: 0.0487:  10%|▉         | 18/187 [01:49<16:43,  5.94s/it]
Training Epoch: 12/100, batch 36/375 completed (loss: 0.0487:  10%|█         | 19/187 [01:53<16:47,  5.99s/it]
Training Epoch: 12/100, batch 36/375 completed (loss: 0.1895:  10%|█         | 19/187 [01:52<16:47,  5.99s/it]
Training Epoch: 12/100, batch 36/375 completed (loss: 0.0688:  10%|█         | 19/187 [01:52<16:47,  6.00s/it]
Training Epoch: 12/100, batch 36/375 completed (loss: 0.3134:  10%|█         | 19/187 [01:53<16:47,  6.00s/it]
Training Epoch: 12/100, batch 37/375 completed (loss: 0.1343:  10%|█         | 19/187 [01:53<16:47,  5.99s/it]
Training Epoch: 12/100, batch 37/375 completed (loss: 0.0076:  10%|█         | 19/187 [01:53<16:47,  6.00s/it]
Training Epoch: 12/100, batch 37/375 completed (loss: 0.1547:  10%|█         | 19/187 [01:53<16:47,  6.00s/it]
Training Epoch: 12/100, batch 37/375 completed (loss: 0.2581:  10%|█         | 19/187 [01:53<16:47,  5.99s/it]
Training Epoch: 12/100, batch 38/375 completed (loss: 0.0158:  10%|█         | 19/187 [01:56<16:47,  6.00s/it]
Training Epoch: 12/100, batch 38/375 completed (loss: 0.0532:  10%|█         | 19/187 [01:56<16:47,  6.00s/it]
Training Epoch: 12/100, batch 38/375 completed (loss: 0.3675:  10%|█         | 19/187 [01:56<16:47,  5.99s/it]
Training Epoch: 12/100, batch 38/375 completed (loss: 0.1962:  10%|█         | 19/187 [01:56<16:47,  5.99s/it]
Training Epoch: 12/100, batch 38/375 completed (loss: 0.3675:  11%|█         | 20/187 [01:58<16:37,  5.97s/it]
Training Epoch: 12/100, batch 38/375 completed (loss: 0.1962:  11%|█         | 20/187 [01:58<16:37,  5.97s/it]
Training Epoch: 12/100, batch 38/375 completed (loss: 0.0532:  11%|█         | 20/187 [01:58<16:37,  5.97s/it]
Training Epoch: 12/100, batch 38/375 completed (loss: 0.0158:  11%|█         | 20/187 [01:58<16:38,  5.98s/it]
Training Epoch: 12/100, batch 39/375 completed (loss: 0.0349:  11%|█         | 20/187 [01:58<16:38,  5.98s/it]
Training Epoch: 12/100, batch 39/375 completed (loss: 0.3325:  11%|█         | 20/187 [01:58<16:37,  5.97s/it]
Training Epoch: 12/100, batch 39/375 completed (loss: 0.0243:  11%|█         | 20/187 [01:58<16:37,  5.97s/it]
Training Epoch: 12/100, batch 39/375 completed (loss: 0.0414:  11%|█         | 20/187 [01:59<16:37,  5.97s/it]
Training Epoch: 12/100, batch 40/375 completed (loss: 0.037:  11%|█         | 20/187 [02:01<16:38,  5.98s/it] 
Training Epoch: 12/100, batch 40/375 completed (loss: 0.0541:  11%|█         | 20/187 [02:01<16:37,  5.97s/it]
Training Epoch: 12/100, batch 40/375 completed (loss: 0.1093:  11%|█         | 20/187 [02:01<16:37,  5.97s/it]
Training Epoch: 12/100, batch 40/375 completed (loss: 0.0834:  11%|█         | 20/187 [02:01<16:37,  5.97s/it]
Training Epoch: 12/100, batch 40/375 completed (loss: 0.0834:  11%|█         | 21/187 [02:04<16:23,  5.92s/it]
Training Epoch: 12/100, batch 40/375 completed (loss: 0.0541:  11%|█         | 21/187 [02:04<16:23,  5.93s/it]
Training Epoch: 12/100, batch 40/375 completed (loss: 0.1093:  11%|█         | 21/187 [02:04<16:24,  5.93s/it]
Training Epoch: 12/100, batch 40/375 completed (loss: 0.037:  11%|█         | 21/187 [02:04<16:24,  5.93s/it]
Training Epoch: 12/100, batch 41/375 completed (loss: 0.3894:  11%|█         | 21/187 [02:04<16:24,  5.93s/it]
Training Epoch: 12/100, batch 41/375 completed (loss: 0.0119:  11%|█         | 21/187 [02:04<16:23,  5.93s/it]
Training Epoch: 12/100, batch 41/375 completed (loss: 0.0188:  11%|█         | 21/187 [02:04<16:24,  5.93s/it]
Training Epoch: 12/100, batch 41/375 completed (loss: 0.0161:  11%|█         | 21/187 [02:04<16:23,  5.92s/it]
Training Epoch: 12/100, batch 42/375 completed (loss: 0.013:  11%|█         | 21/187 [02:07<16:24,  5.93s/it] 
Training Epoch: 12/100, batch 42/375 completed (loss: 0.1525:  11%|█         | 21/187 [02:07<16:23,  5.93s/it]
Training Epoch: 12/100, batch 42/375 completed (loss: 0.1298:  11%|█         | 21/187 [02:07<16:24,  5.93s/it]
Training Epoch: 12/100, batch 42/375 completed (loss: 0.0146:  11%|█         | 21/187 [02:07<16:23,  5.92s/it]
Training Epoch: 12/100, batch 42/375 completed (loss: 0.0146:  12%|█▏        | 22/187 [02:11<16:42,  6.07s/it]
Training Epoch: 12/100, batch 42/375 completed (loss: 0.1525:  12%|█▏        | 22/187 [02:11<16:42,  6.08s/it]
Training Epoch: 12/100, batch 42/375 completed (loss: 0.1298:  12%|█▏        | 22/187 [02:11<16:42,  6.08s/it]
Training Epoch: 12/100, batch 42/375 completed (loss: 0.013:  12%|█▏        | 22/187 [02:11<16:42,  6.08s/it]
Training Epoch: 12/100, batch 43/375 completed (loss: 0.0321:  12%|█▏        | 22/187 [02:11<16:42,  6.08s/it]
Training Epoch: 12/100, batch 43/375 completed (loss: 0.1233:  12%|█▏        | 22/187 [02:11<16:42,  6.08s/it]
Training Epoch: 12/100, batch 43/375 completed (loss: 0.0314:  12%|█▏        | 22/187 [02:11<16:42,  6.08s/it]
Training Epoch: 12/100, batch 43/375 completed (loss: 0.0495:  12%|█▏        | 22/187 [02:11<16:42,  6.07s/it]
Training Epoch: 12/100, batch 44/375 completed (loss: 0.6045:  12%|█▏        | 22/187 [02:14<16:42,  6.08s/it]
Training Epoch: 12/100, batch 44/375 completed (loss: 0.0096:  12%|█▏        | 22/187 [02:14<16:42,  6.08s/it]
Training Epoch: 12/100, batch 44/375 completed (loss: 0.2266:  12%|█▏        | 22/187 [02:14<16:42,  6.08s/it]
Training Epoch: 12/100, batch 44/375 completed (loss: 0.1479:  12%|█▏        | 22/187 [02:14<16:42,  6.07s/it]
Training Epoch: 12/100, batch 44/375 completed (loss: 0.0096:  12%|█▏        | 23/187 [02:16<16:23,  6.00s/it]
Training Epoch: 12/100, batch 44/375 completed (loss: 0.1479:  12%|█▏        | 23/187 [02:17<16:24,  6.00s/it]
Training Epoch: 12/100, batch 44/375 completed (loss: 0.6045:  12%|█▏        | 23/187 [02:16<16:23,  6.00s/it]
Training Epoch: 12/100, batch 44/375 completed (loss: 0.2266:  12%|█▏        | 23/187 [02:16<16:23,  6.00s/it]
Training Epoch: 12/100, batch 45/375 completed (loss: 0.0389:  12%|█▏        | 23/187 [02:17<16:23,  6.00s/it]
Training Epoch: 12/100, batch 45/375 completed (loss: 0.0732:  12%|█▏        | 23/187 [02:17<16:23,  6.00s/it]
Training Epoch: 12/100, batch 45/375 completed (loss: 0.0831:  12%|█▏        | 23/187 [02:17<16:23,  6.00s/it]
Training Epoch: 12/100, batch 45/375 completed (loss: 0.185:  12%|█▏        | 23/187 [02:17<16:24,  6.00s/it] 
Training Epoch: 12/100, batch 46/375 completed (loss: 0.0638:  12%|█▏        | 23/187 [02:19<16:23,  6.00s/it]
Training Epoch: 12/100, batch 46/375 completed (loss: 0.0803:  12%|█▏        | 23/187 [02:19<16:23,  6.00s/it]
Training Epoch: 12/100, batch 46/375 completed (loss: 0.0973:  12%|█▏        | 23/187 [02:19<16:23,  6.00s/it]
Training Epoch: 12/100, batch 46/375 completed (loss: 0.3869:  12%|█▏        | 23/187 [02:19<16:24,  6.00s/it]
Training Epoch: 12/100, batch 46/375 completed (loss: 0.3869:  13%|█▎        | 24/187 [02:22<16:08,  5.94s/it]
Training Epoch: 12/100, batch 46/375 completed (loss: 0.0973:  13%|█▎        | 24/187 [02:22<16:08,  5.94s/it]
Training Epoch: 12/100, batch 46/375 completed (loss: 0.0803:  13%|█▎        | 24/187 [02:22<16:08,  5.94s/it]
Training Epoch: 12/100, batch 46/375 completed (loss: 0.0638:  13%|█▎        | 24/187 [02:22<16:08,  5.94s/it]
Training Epoch: 12/100, batch 47/375 completed (loss: 0.1137:  13%|█▎        | 24/187 [02:22<16:08,  5.94s/it]
Training Epoch: 12/100, batch 47/375 completed (loss: 0.0446:  13%|█▎        | 24/187 [02:22<16:08,  5.94s/it]
Training Epoch: 12/100, batch 47/375 completed (loss: 0.0327:  13%|█▎        | 24/187 [02:22<16:08,  5.94s/it]
Training Epoch: 12/100, batch 47/375 completed (loss: 0.2035:  13%|█▎        | 24/187 [02:22<16:08,  5.94s/it]
Training Epoch: 12/100, batch 48/375 completed (loss: 0.059:  13%|█▎        | 24/187 [02:26<16:08,  5.94s/it] 
Training Epoch: 12/100, batch 48/375 completed (loss: 0.0417:  13%|█▎        | 24/187 [02:26<16:08,  5.94s/it]
Training Epoch: 12/100, batch 48/375 completed (loss: 0.2258:  13%|█▎        | 24/187 [02:26<16:08,  5.94s/it]
Training Epoch: 12/100, batch 48/375 completed (loss: 0.139:  13%|█▎        | 24/187 [02:26<16:08,  5.94s/it] 
Training Epoch: 12/100, batch 48/375 completed (loss: 0.0417:  13%|█▎        | 25/187 [02:29<16:25,  6.08s/it]
Training Epoch: 12/100, batch 48/375 completed (loss: 0.059:  13%|█▎        | 25/187 [02:29<16:25,  6.08s/it]
Training Epoch: 12/100, batch 48/375 completed (loss: 0.139:  13%|█▎        | 25/187 [02:29<16:25,  6.08s/it]
Training Epoch: 12/100, batch 48/375 completed (loss: 0.2258:  13%|█▎        | 25/187 [02:29<16:24,  6.08s/it]
Training Epoch: 12/100, batch 49/375 completed (loss: 0.2031:  13%|█▎        | 25/187 [02:29<16:25,  6.08s/it]
Training Epoch: 12/100, batch 49/375 completed (loss: 0.0306:  13%|█▎        | 25/187 [02:29<16:25,  6.08s/it]
Training Epoch: 12/100, batch 49/375 completed (loss: 0.14:  13%|█▎        | 25/187 [02:29<16:24,  6.08s/it]  
Training Epoch: 12/100, batch 49/375 completed (loss: 0.0819:  13%|█▎        | 25/187 [02:29<16:25,  6.08s/it]
Training Epoch: 12/100, batch 50/375 completed (loss: 0.0287:  13%|█▎        | 25/187 [02:32<16:25,  6.08s/it]
Training Epoch: 12/100, batch 50/375 completed (loss: 0.0303:  13%|█▎        | 25/187 [02:32<16:24,  6.08s/it]
Training Epoch: 12/100, batch 50/375 completed (loss: 0.1289:  13%|█▎        | 25/187 [02:32<16:25,  6.08s/it]
Training Epoch: 12/100, batch 50/375 completed (loss: 0.026:  13%|█▎        | 25/187 [02:32<16:25,  6.08s/it] 
Training Epoch: 12/100, batch 50/375 completed (loss: 0.1289:  14%|█▍        | 26/187 [02:34<16:04,  5.99s/it]
Training Epoch: 12/100, batch 50/375 completed (loss: 0.0287:  14%|█▍        | 26/187 [02:34<16:04,  5.99s/it]
Training Epoch: 12/100, batch 50/375 completed (loss: 0.026:  14%|█▍        | 26/187 [02:35<16:05,  5.99s/it]
Training Epoch: 12/100, batch 50/375 completed (loss: 0.0303:  14%|█▍        | 26/187 [02:34<16:04,  5.99s/it]
Training Epoch: 12/100, batch 51/375 completed (loss: 0.1674:  14%|█▍        | 26/187 [02:35<16:04,  5.99s/it]
Training Epoch: 12/100, batch 51/375 completed (loss: 0.1334:  14%|█▍        | 26/187 [02:35<16:04,  5.99s/it]
Training Epoch: 12/100, batch 51/375 completed (loss: 0.026:  14%|█▍        | 26/187 [02:35<16:04,  5.99s/it] 
Training Epoch: 12/100, batch 51/375 completed (loss: 0.048:  14%|█▍        | 26/187 [02:35<16:05,  5.99s/it]
Training Epoch: 12/100, batch 52/375 completed (loss: 0.364:  14%|█▍        | 26/187 [02:38<16:04,  5.99s/it]
Training Epoch: 12/100, batch 52/375 completed (loss: 0.1155:  14%|█▍        | 26/187 [02:38<16:04,  5.99s/it]
Training Epoch: 12/100, batch 52/375 completed (loss: 0.0376:  14%|█▍        | 26/187 [02:38<16:04,  5.99s/it]
Training Epoch: 12/100, batch 52/375 completed (loss: 0.0256:  14%|█▍        | 26/187 [02:38<16:05,  5.99s/it]
Training Epoch: 12/100, batch 52/375 completed (loss: 0.0376:  14%|█▍        | 27/187 [02:41<16:11,  6.07s/it]
Training Epoch: 12/100, batch 52/375 completed (loss: 0.364:  14%|█▍        | 27/187 [02:41<16:11,  6.07s/it]
Training Epoch: 12/100, batch 52/375 completed (loss: 0.0256:  14%|█▍        | 27/187 [02:41<16:12,  6.08s/it]
Training Epoch: 12/100, batch 52/375 completed (loss: 0.1155:  14%|█▍        | 27/187 [02:41<16:11,  6.07s/it]
Training Epoch: 12/100, batch 53/375 completed (loss: 0.0651:  14%|█▍        | 27/187 [02:41<16:11,  6.07s/it]
Training Epoch: 12/100, batch 53/375 completed (loss: 0.1131:  14%|█▍        | 27/187 [02:41<16:11,  6.07s/it]
Training Epoch: 12/100, batch 53/375 completed (loss: 0.023:  14%|█▍        | 27/187 [02:41<16:11,  6.07s/it] 
Training Epoch: 12/100, batch 53/375 completed (loss: 0.0329:  14%|█▍        | 27/187 [02:41<16:12,  6.08s/it]
Training Epoch: 12/100, batch 54/375 completed (loss: 0.0662:  14%|█▍        | 27/187 [02:44<16:11,  6.07s/it]
Training Epoch: 12/100, batch 54/375 completed (loss: 0.0682:  14%|█▍        | 27/187 [02:44<16:11,  6.07s/it]
Training Epoch: 12/100, batch 54/375 completed (loss: 0.2223:  14%|█▍        | 27/187 [02:44<16:11,  6.07s/it]
Training Epoch: 12/100, batch 54/375 completed (loss: 0.0278:  14%|█▍        | 27/187 [02:44<16:12,  6.08s/it]
Training Epoch: 12/100, batch 54/375 completed (loss: 0.0278:  15%|█▍        | 28/187 [02:47<16:02,  6.05s/it]
Training Epoch: 12/100, batch 54/375 completed (loss: 0.0682:  15%|█▍        | 28/187 [02:47<16:02,  6.05s/it]
Training Epoch: 12/100, batch 54/375 completed (loss: 0.0662:  15%|█▍        | 28/187 [02:47<16:02,  6.05s/it]
Training Epoch: 12/100, batch 54/375 completed (loss: 0.2223:  15%|█▍        | 28/187 [02:47<16:02,  6.05s/it]
Training Epoch: 12/100, batch 55/375 completed (loss: 0.0696:  15%|█▍        | 28/187 [02:47<16:02,  6.05s/it]
Training Epoch: 12/100, batch 55/375 completed (loss: 0.1588:  15%|█▍        | 28/187 [02:47<16:02,  6.05s/it]
Training Epoch: 12/100, batch 55/375 completed (loss: 0.1157:  15%|█▍        | 28/187 [02:47<16:02,  6.05s/it]
Training Epoch: 12/100, batch 55/375 completed (loss: 0.0404:  15%|█▍        | 28/187 [02:47<16:02,  6.05s/it]
Training Epoch: 12/100, batch 56/375 completed (loss: 0.046:  15%|█▍        | 28/187 [02:50<16:02,  6.05s/it] 
Training Epoch: 12/100, batch 56/375 completed (loss: 0.1587:  15%|█▍        | 28/187 [02:50<16:02,  6.05s/it]
Training Epoch: 12/100, batch 56/375 completed (loss: 0.2056:  15%|█▍        | 28/187 [02:50<16:02,  6.05s/it]
Training Epoch: 12/100, batch 56/375 completed (loss: 0.4418:  15%|█▍        | 28/187 [02:50<16:02,  6.05s/it]
Training Epoch: 12/100, batch 56/375 completed (loss: 0.2056:  16%|█▌        | 29/187 [02:53<15:45,  5.98s/it]
Training Epoch: 12/100, batch 56/375 completed (loss: 0.1587:  16%|█▌        | 29/187 [02:53<15:45,  5.98s/it]
Training Epoch: 12/100, batch 56/375 completed (loss: 0.046:  16%|█▌        | 29/187 [02:53<15:45,  5.99s/it]
Training Epoch: 12/100, batch 56/375 completed (loss: 0.4418:  16%|█▌        | 29/187 [02:53<15:45,  5.99s/it]
Training Epoch: 12/100, batch 57/375 completed (loss: 0.0561:  16%|█▌        | 29/187 [02:53<15:45,  5.98s/it]
Training Epoch: 12/100, batch 57/375 completed (loss: 0.0314:  16%|█▌        | 29/187 [02:53<15:45,  5.99s/it]
Training Epoch: 12/100, batch 57/375 completed (loss: 0.0916:  16%|█▌        | 29/187 [02:53<15:45,  5.98s/it]
Training Epoch: 12/100, batch 57/375 completed (loss: 0.1434:  16%|█▌        | 29/187 [02:53<15:45,  5.99s/it]
Training Epoch: 12/100, batch 58/375 completed (loss: 0.0135:  16%|█▌        | 29/187 [02:56<15:45,  5.99s/it]
Training Epoch: 12/100, batch 58/375 completed (loss: 0.2112:  16%|█▌        | 29/187 [02:56<15:45,  5.98s/it]
Training Epoch: 12/100, batch 58/375 completed (loss: 0.0407:  16%|█▌        | 29/187 [02:56<15:45,  5.98s/it]
Training Epoch: 12/100, batch 58/375 completed (loss: 0.0425:  16%|█▌        | 29/187 [02:56<15:45,  5.99s/it]
Training Epoch: 12/100, batch 58/375 completed (loss: 0.0135:  16%|█▌        | 30/187 [02:59<16:00,  6.12s/it]
Training Epoch: 12/100, batch 58/375 completed (loss: 0.2112:  16%|█▌        | 30/187 [02:59<16:00,  6.12s/it]
Training Epoch: 12/100, batch 58/375 completed (loss: 0.0407:  16%|█▌        | 30/187 [02:59<16:00,  6.12s/it]
Training Epoch: 12/100, batch 58/375 completed (loss: 0.0425:  16%|█▌        | 30/187 [02:59<16:00,  6.12s/it]
Training Epoch: 12/100, batch 59/375 completed (loss: 0.0401:  16%|█▌        | 30/187 [02:59<16:00,  6.12s/it]
Training Epoch: 12/100, batch 59/375 completed (loss: 0.2672:  16%|█▌        | 30/187 [02:59<16:00,  6.12s/it]
Training Epoch: 12/100, batch 59/375 completed (loss: 0.0895:  16%|█▌        | 30/187 [02:59<16:00,  6.12s/it]
Training Epoch: 12/100, batch 59/375 completed (loss: 0.025:  16%|█▌        | 30/187 [02:59<16:00,  6.12s/it] 
Training Epoch: 12/100, batch 60/375 completed (loss: 0.0559:  16%|█▌        | 30/187 [03:02<16:00,  6.12s/it]
Training Epoch: 12/100, batch 60/375 completed (loss: 0.238:  16%|█▌        | 30/187 [03:02<16:00,  6.12s/it] 
Training Epoch: 12/100, batch 60/375 completed (loss: 0.1786:  16%|█▌        | 30/187 [03:02<16:00,  6.12s/it]
Training Epoch: 12/100, batch 60/375 completed (loss: 0.242:  16%|█▌        | 30/187 [03:02<16:00,  6.12s/it]
Training Epoch: 12/100, batch 60/375 completed (loss: 0.238:  17%|█▋        | 31/187 [03:05<15:38,  6.02s/it]
Training Epoch: 12/100, batch 60/375 completed (loss: 0.0559:  17%|█▋        | 31/187 [03:05<15:38,  6.02s/it]
Training Epoch: 12/100, batch 60/375 completed (loss: 0.1786:  17%|█▋        | 31/187 [03:05<15:38,  6.02s/it]
Training Epoch: 12/100, batch 60/375 completed (loss: 0.242:  17%|█▋        | 31/187 [03:05<15:38,  6.02s/it]
Training Epoch: 12/100, batch 61/375 completed (loss: 0.1121:  17%|█▋        | 31/187 [03:05<15:38,  6.02s/it]
Training Epoch: 12/100, batch 61/375 completed (loss: 0.0569:  17%|█▋        | 31/187 [03:05<15:38,  6.02s/it]
Training Epoch: 12/100, batch 61/375 completed (loss: 0.0437:  17%|█▋        | 31/187 [03:05<15:38,  6.02s/it]
Training Epoch: 12/100, batch 61/375 completed (loss: 0.0289:  17%|█▋        | 31/187 [03:05<15:38,  6.02s/it]
Training Epoch: 12/100, batch 62/375 completed (loss: 0.0434:  17%|█▋        | 31/187 [03:08<15:38,  6.02s/it]
Training Epoch: 12/100, batch 62/375 completed (loss: 0.1484:  17%|█▋        | 31/187 [03:08<15:38,  6.02s/it]
Training Epoch: 12/100, batch 62/375 completed (loss: 0.1476:  17%|█▋        | 31/187 [03:08<15:38,  6.02s/it]
Training Epoch: 12/100, batch 62/375 completed (loss: 0.0149:  17%|█▋        | 31/187 [03:08<15:38,  6.02s/it]
Training Epoch: 12/100, batch 62/375 completed (loss: 0.0149:  17%|█▋        | 32/187 [03:11<15:30,  6.00s/it]
Training Epoch: 12/100, batch 62/375 completed (loss: 0.1484:  17%|█▋        | 32/187 [03:11<15:31,  6.01s/it]
Training Epoch: 12/100, batch 62/375 completed (loss: 0.1476:  17%|█▋        | 32/187 [03:11<15:31,  6.01s/it]
Training Epoch: 12/100, batch 62/375 completed (loss: 0.0434:  17%|█▋        | 32/187 [03:11<15:31,  6.01s/it]
Training Epoch: 12/100, batch 63/375 completed (loss: 0.1237:  17%|█▋        | 32/187 [03:11<15:31,  6.01s/it]
Training Epoch: 12/100, batch 63/375 completed (loss: 0.0834:  17%|█▋        | 32/187 [03:11<15:31,  6.01s/it]
Training Epoch: 12/100, batch 63/375 completed (loss: 0.0748:  17%|█▋        | 32/187 [03:11<15:31,  6.01s/it]
Training Epoch: 12/100, batch 63/375 completed (loss: 0.012:  17%|█▋        | 32/187 [03:11<15:30,  6.00s/it] 
Training Epoch: 12/100, batch 64/375 completed (loss: 0.0239:  17%|█▋        | 32/187 [03:14<15:31,  6.01s/it]
Training Epoch: 12/100, batch 64/375 completed (loss: 0.0304:  17%|█▋        | 32/187 [03:14<15:31,  6.01s/it]
Training Epoch: 12/100, batch 64/375 completed (loss: 0.0907:  17%|█▋        | 32/187 [03:14<15:31,  6.01s/it]
Training Epoch: 12/100, batch 64/375 completed (loss: 0.1686:  17%|█▋        | 32/187 [03:14<15:30,  6.00s/it]
Training Epoch: 12/100, batch 64/375 completed (loss: 0.1686:  18%|█▊        | 33/187 [03:17<15:39,  6.10s/it]
Training Epoch: 12/100, batch 64/375 completed (loss: 0.0304:  18%|█▊        | 33/187 [03:17<15:39,  6.10s/it]
Training Epoch: 12/100, batch 64/375 completed (loss: 0.0907:  18%|█▊        | 33/187 [03:17<15:39,  6.10s/it]
Training Epoch: 12/100, batch 64/375 completed (loss: 0.0239:  18%|█▊        | 33/187 [03:17<15:39,  6.10s/it]
Training Epoch: 12/100, batch 65/375 completed (loss: 0.4471:  18%|█▊        | 33/187 [03:17<15:39,  6.10s/it]
Training Epoch: 12/100, batch 65/375 completed (loss: 0.0357:  18%|█▊        | 33/187 [03:17<15:39,  6.10s/it]
Training Epoch: 12/100, batch 65/375 completed (loss: 0.0421:  18%|█▊        | 33/187 [03:17<15:39,  6.10s/it]
Training Epoch: 12/100, batch 65/375 completed (loss: 0.1261:  18%|█▊        | 33/187 [03:17<15:39,  6.10s/it]
Training Epoch: 12/100, batch 66/375 completed (loss: 0.2598:  18%|█▊        | 33/187 [03:20<15:39,  6.10s/it]
Training Epoch: 12/100, batch 66/375 completed (loss: 0.2523:  18%|█▊        | 33/187 [03:20<15:39,  6.10s/it]
Training Epoch: 12/100, batch 66/375 completed (loss: 0.0408:  18%|█▊        | 33/187 [03:20<15:39,  6.10s/it]
Training Epoch: 12/100, batch 66/375 completed (loss: 0.2564:  18%|█▊        | 33/187 [03:20<15:39,  6.10s/it]
Training Epoch: 12/100, batch 66/375 completed (loss: 0.0408:  18%|█▊        | 34/187 [03:23<15:21,  6.02s/it]
Training Epoch: 12/100, batch 66/375 completed (loss: 0.2598:  18%|█▊        | 34/187 [03:23<15:21,  6.02s/it]
Training Epoch: 12/100, batch 66/375 completed (loss: 0.2523:  18%|█▊        | 34/187 [03:23<15:21,  6.02s/it]
Training Epoch: 12/100, batch 66/375 completed (loss: 0.2564:  18%|█▊        | 34/187 [03:23<15:21,  6.03s/it]
Training Epoch: 12/100, batch 67/375 completed (loss: 0.1417:  18%|█▊        | 34/187 [03:23<15:21,  6.02s/it]
Training Epoch: 12/100, batch 67/375 completed (loss: 0.1175:  18%|█▊        | 34/187 [03:23<15:21,  6.02s/it]
Training Epoch: 12/100, batch 67/375 completed (loss: 0.0328:  18%|█▊        | 34/187 [03:23<15:21,  6.02s/it]
Training Epoch: 12/100, batch 67/375 completed (loss: 0.0979:  18%|█▊        | 34/187 [03:23<15:21,  6.03s/it]
Training Epoch: 12/100, batch 68/375 completed (loss: 0.0704:  18%|█▊        | 34/187 [03:26<15:21,  6.02s/it]
Training Epoch: 12/100, batch 68/375 completed (loss: 0.0611:  18%|█▊        | 34/187 [03:26<15:21,  6.02s/it]
Training Epoch: 12/100, batch 68/375 completed (loss: 0.178:  18%|█▊        | 34/187 [03:26<15:21,  6.02s/it] 
Training Epoch: 12/100, batch 68/375 completed (loss: 0.0761:  18%|█▊        | 34/187 [03:26<15:21,  6.03s/it]
Training Epoch: 12/100, batch 68/375 completed (loss: 0.0761:  19%|█▊        | 35/187 [03:29<15:27,  6.10s/it]
Training Epoch: 12/100, batch 68/375 completed (loss: 0.178:  19%|█▊        | 35/187 [03:29<15:27,  6.10s/it]
Training Epoch: 12/100, batch 68/375 completed (loss: 0.0704:  19%|█▊        | 35/187 [03:29<15:27,  6.10s/it]
Training Epoch: 12/100, batch 68/375 completed (loss: 0.0611:  19%|█▊        | 35/187 [03:29<15:27,  6.10s/it]
Training Epoch: 12/100, batch 69/375 completed (loss: 0.1224:  19%|█▊        | 35/187 [03:29<15:27,  6.10s/it]
Training Epoch: 12/100, batch 69/375 completed (loss: 0.1362:  19%|█▊        | 35/187 [03:29<15:27,  6.10s/it]
Training Epoch: 12/100, batch 69/375 completed (loss: 0.0319:  19%|█▊        | 35/187 [03:29<15:27,  6.10s/it]
Training Epoch: 12/100, batch 69/375 completed (loss: 0.0455:  19%|█▊        | 35/187 [03:29<15:27,  6.10s/it]
Training Epoch: 12/100, batch 70/375 completed (loss: 0.1175:  19%|█▊        | 35/187 [03:32<15:27,  6.10s/it]
Training Epoch: 12/100, batch 70/375 completed (loss: 0.0346:  19%|█▊        | 35/187 [03:32<15:27,  6.10s/it]
Training Epoch: 12/100, batch 70/375 completed (loss: 0.0175:  19%|█▊        | 35/187 [03:32<15:27,  6.10s/it]
Training Epoch: 12/100, batch 70/375 completed (loss: 0.1307:  19%|█▊        | 35/187 [03:32<15:27,  6.10s/it]
Training Epoch: 12/100, batch 70/375 completed (loss: 0.0346:  19%|█▉        | 36/187 [03:35<15:15,  6.06s/it]
Training Epoch: 12/100, batch 70/375 completed (loss: 0.1307:  19%|█▉        | 36/187 [03:35<15:15,  6.06s/it]
Training Epoch: 12/100, batch 70/375 completed (loss: 0.0175:  19%|█▉        | 36/187 [03:35<15:15,  6.06s/it]
Training Epoch: 12/100, batch 70/375 completed (loss: 0.1175:  19%|█▉        | 36/187 [03:35<15:15,  6.06s/it]
Training Epoch: 12/100, batch 71/375 completed (loss: 0.0964:  19%|█▉        | 36/187 [03:35<15:15,  6.06s/it]
Training Epoch: 12/100, batch 71/375 completed (loss: 0.0838:  19%|█▉        | 36/187 [03:35<15:15,  6.06s/it]
Training Epoch: 12/100, batch 71/375 completed (loss: 0.0782:  19%|█▉        | 36/187 [03:35<15:15,  6.06s/it]
Training Epoch: 12/100, batch 71/375 completed (loss: 0.0216:  19%|█▉        | 36/187 [03:35<15:15,  6.06s/it]
Training Epoch: 12/100, batch 72/375 completed (loss: 0.0649:  19%|█▉        | 36/187 [03:38<15:15,  6.06s/it]
Training Epoch: 12/100, batch 72/375 completed (loss: 0.3523:  19%|█▉        | 36/187 [03:38<15:15,  6.06s/it]
Training Epoch: 12/100, batch 72/375 completed (loss: 0.0252:  19%|█▉        | 36/187 [03:38<15:15,  6.06s/it]
Training Epoch: 12/100, batch 72/375 completed (loss: 0.0721:  19%|█▉        | 36/187 [03:38<15:15,  6.06s/it]
Training Epoch: 12/100, batch 72/375 completed (loss: 0.3523:  20%|█▉        | 37/187 [03:41<15:05,  6.03s/it]
Training Epoch: 12/100, batch 72/375 completed (loss: 0.0721:  20%|█▉        | 37/187 [03:41<15:05,  6.04s/it]
Training Epoch: 12/100, batch 72/375 completed (loss: 0.0649:  20%|█▉        | 37/187 [03:41<15:05,  6.04s/it]
Training Epoch: 12/100, batch 72/375 completed (loss: 0.0252:  20%|█▉        | 37/187 [03:41<15:05,  6.04s/it]
Training Epoch: 12/100, batch 73/375 completed (loss: 0.0398:  20%|█▉        | 37/187 [03:41<15:05,  6.04s/it]
Training Epoch: 12/100, batch 73/375 completed (loss: 0.0928:  20%|█▉        | 37/187 [03:41<15:05,  6.03s/it]
Training Epoch: 12/100, batch 73/375 completed (loss: 0.0777:  20%|█▉        | 37/187 [03:41<15:05,  6.04s/it]
Training Epoch: 12/100, batch 73/375 completed (loss: 0.2294:  20%|█▉        | 37/187 [03:41<15:05,  6.04s/it]
Training Epoch: 12/100, batch 74/375 completed (loss: 0.1111:  20%|█▉        | 37/187 [03:44<15:05,  6.04s/it]
Training Epoch: 12/100, batch 74/375 completed (loss: 0.1199:  20%|█▉        | 37/187 [03:44<15:05,  6.03s/it]
Training Epoch: 12/100, batch 74/375 completed (loss: 0.0401:  20%|█▉        | 37/187 [03:44<15:05,  6.04s/it]
Training Epoch: 12/100, batch 74/375 completed (loss: 0.0966:  20%|█▉        | 37/187 [03:44<15:05,  6.04s/it]
Training Epoch: 12/100, batch 74/375 completed (loss: 0.1199:  20%|██        | 38/187 [03:47<15:02,  6.06s/it]
Training Epoch: 12/100, batch 74/375 completed (loss: 0.1111:  20%|██        | 38/187 [03:47<15:02,  6.06s/it]
Training Epoch: 12/100, batch 74/375 completed (loss: 0.0401:  20%|██        | 38/187 [03:47<15:02,  6.06s/it]
Training Epoch: 12/100, batch 74/375 completed (loss: 0.0966:  20%|██        | 38/187 [03:47<15:03,  6.06s/it]
Training Epoch: 12/100, batch 75/375 completed (loss: 0.069:  20%|██        | 38/187 [03:47<15:02,  6.06s/it] 
Training Epoch: 12/100, batch 75/375 completed (loss: 0.0116:  20%|██        | 38/187 [03:47<15:02,  6.06s/it]
Training Epoch: 12/100, batch 75/375 completed (loss: 0.14:  20%|██        | 38/187 [03:47<15:02,  6.06s/it]  
Training Epoch: 12/100, batch 75/375 completed (loss: 0.0368:  20%|██        | 38/187 [03:47<15:03,  6.06s/it]
Training Epoch: 12/100, batch 76/375 completed (loss: 0.1077:  20%|██        | 38/187 [03:50<15:02,  6.06s/it]
Training Epoch: 12/100, batch 76/375 completed (loss: 0.0305:  20%|██        | 38/187 [03:50<15:02,  6.06s/it]
Training Epoch: 12/100, batch 76/375 completed (loss: 0.2006:  20%|██        | 38/187 [03:50<15:02,  6.06s/it]
Training Epoch: 12/100, batch 76/375 completed (loss: 0.2796:  20%|██        | 38/187 [03:50<15:03,  6.06s/it]
Training Epoch: 12/100, batch 76/375 completed (loss: 0.0305:  21%|██        | 39/187 [03:53<14:53,  6.03s/it]
Training Epoch: 12/100, batch 76/375 completed (loss: 0.2006:  21%|██        | 39/187 [03:53<14:52,  6.03s/it]
Training Epoch: 12/100, batch 76/375 completed (loss: 0.2796:  21%|██        | 39/187 [03:53<14:53,  6.03s/it]
Training Epoch: 12/100, batch 76/375 completed (loss: 0.1077:  21%|██        | 39/187 [03:53<14:53,  6.04s/it]
Training Epoch: 12/100, batch 77/375 completed (loss: 0.0317:  21%|██        | 39/187 [03:53<14:52,  6.03s/it]
Training Epoch: 12/100, batch 77/375 completed (loss: 0.0802:  21%|██        | 39/187 [03:53<14:53,  6.04s/it]
Training Epoch: 12/100, batch 77/375 completed (loss: 0.0538:  21%|██        | 39/187 [03:53<14:53,  6.03s/it]
Training Epoch: 12/100, batch 77/375 completed (loss: 0.0095:  21%|██        | 39/187 [03:53<14:53,  6.03s/it]
Training Epoch: 12/100, batch 78/375 completed (loss: 0.0548:  21%|██        | 39/187 [03:56<14:52,  6.03s/it]
Training Epoch: 12/100, batch 78/375 completed (loss: 0.1799:  21%|██        | 39/187 [03:56<14:53,  6.04s/it]
Training Epoch: 12/100, batch 78/375 completed (loss: 0.1263:  21%|██        | 39/187 [03:56<14:53,  6.03s/it]
Training Epoch: 12/100, batch 78/375 completed (loss: 0.2179:  21%|██        | 39/187 [03:56<14:53,  6.03s/it]
Training Epoch: 12/100, batch 78/375 completed (loss: 0.1263:  21%|██▏       | 40/187 [03:59<14:52,  6.07s/it]
Training Epoch: 12/100, batch 78/375 completed (loss: 0.1799:  21%|██▏       | 40/187 [03:59<14:51,  6.07s/it]
Training Epoch: 12/100, batch 78/375 completed (loss: 0.0548:  21%|██▏       | 40/187 [03:59<14:51,  6.07s/it]
Training Epoch: 12/100, batch 78/375 completed (loss: 0.2179:  21%|██▏       | 40/187 [03:59<14:52,  6.07s/it]
Training Epoch: 12/100, batch 79/375 completed (loss: 0.0114:  21%|██▏       | 40/187 [03:59<14:51,  6.07s/it]
Training Epoch: 12/100, batch 79/375 completed (loss: 0.1215:  21%|██▏       | 40/187 [03:59<14:52,  6.07s/it]
Training Epoch: 12/100, batch 79/375 completed (loss: 0.2488:  21%|██▏       | 40/187 [03:59<14:51,  6.07s/it]
Training Epoch: 12/100, batch 79/375 completed (loss: 0.1589:  21%|██▏       | 40/187 [03:59<14:52,  6.07s/it]
Training Epoch: 12/100, batch 80/375 completed (loss: 0.0268:  21%|██▏       | 40/187 [04:03<14:51,  6.07s/it]
Training Epoch: 12/100, batch 80/375 completed (loss: 0.0867:  21%|██▏       | 40/187 [04:03<14:51,  6.07s/it]
Training Epoch: 12/100, batch 80/375 completed (loss: 0.0745:  21%|██▏       | 40/187 [04:03<14:52,  6.07s/it]
Training Epoch: 12/100, batch 80/375 completed (loss: 0.0469:  21%|██▏       | 40/187 [04:03<14:52,  6.07s/it]
Training Epoch: 12/100, batch 80/375 completed (loss: 0.0268:  22%|██▏       | 41/187 [04:06<14:48,  6.08s/it]
Training Epoch: 12/100, batch 80/375 completed (loss: 0.0867:  22%|██▏       | 41/187 [04:06<14:48,  6.08s/it]
Training Epoch: 12/100, batch 80/375 completed (loss: 0.0469:  22%|██▏       | 41/187 [04:06<14:48,  6.08s/it]
Training Epoch: 12/100, batch 80/375 completed (loss: 0.0745:  22%|██▏       | 41/187 [04:06<14:48,  6.09s/it]
Training Epoch: 12/100, batch 81/375 completed (loss: 0.0258:  22%|██▏       | 41/187 [04:06<14:48,  6.08s/it]
Training Epoch: 12/100, batch 81/375 completed (loss: 0.1468:  22%|██▏       | 41/187 [04:06<14:48,  6.08s/it]
Training Epoch: 12/100, batch 81/375 completed (loss: 0.1319:  22%|██▏       | 41/187 [04:06<14:48,  6.09s/it]
Training Epoch: 12/100, batch 81/375 completed (loss: 0.0325:  22%|██▏       | 41/187 [04:06<14:48,  6.08s/it]
Training Epoch: 12/100, batch 82/375 completed (loss: 0.1086:  22%|██▏       | 41/187 [04:08<14:48,  6.08s/it]
Training Epoch: 12/100, batch 82/375 completed (loss: 0.1063:  22%|██▏       | 41/187 [04:08<14:48,  6.09s/it]
Training Epoch: 12/100, batch 82/375 completed (loss: 0.0234:  22%|██▏       | 41/187 [04:08<14:48,  6.08s/it]
Training Epoch: 12/100, batch 82/375 completed (loss: 0.0844:  22%|██▏       | 41/187 [04:09<14:48,  6.08s/it]
Training Epoch: 12/100, batch 82/375 completed (loss: 0.1086:  22%|██▏       | 42/187 [04:11<14:29,  6.00s/it]
Training Epoch: 12/100, batch 82/375 completed (loss: 0.0844:  22%|██▏       | 42/187 [04:11<14:29,  6.00s/it]
Training Epoch: 12/100, batch 82/375 completed (loss: 0.0234:  22%|██▏       | 42/187 [04:11<14:29,  6.00s/it]
Training Epoch: 12/100, batch 82/375 completed (loss: 0.1063:  22%|██▏       | 42/187 [04:11<14:30,  6.00s/it]
Training Epoch: 12/100, batch 83/375 completed (loss: 0.0627:  22%|██▏       | 42/187 [04:11<14:29,  6.00s/it]
Training Epoch: 12/100, batch 83/375 completed (loss: 0.0534:  22%|██▏       | 42/187 [04:11<14:29,  6.00s/it]
Training Epoch: 12/100, batch 83/375 completed (loss: 0.0305:  22%|██▏       | 42/187 [04:11<14:30,  6.00s/it]
Training Epoch: 12/100, batch 83/375 completed (loss: 0.1234:  22%|██▏       | 42/187 [04:11<14:29,  6.00s/it]
Training Epoch: 12/100, batch 84/375 completed (loss: 0.2295:  22%|██▏       | 42/187 [04:15<14:29,  6.00s/it]
Training Epoch: 12/100, batch 84/375 completed (loss: 0.0289:  22%|██▏       | 42/187 [04:15<14:30,  6.00s/it]
Training Epoch: 12/100, batch 84/375 completed (loss: 0.0192:  22%|██▏       | 42/187 [04:15<14:29,  6.00s/it]
Training Epoch: 12/100, batch 84/375 completed (loss: 0.09:  22%|██▏       | 42/187 [04:15<14:29,  6.00s/it]  
Training Epoch: 12/100, batch 84/375 completed (loss: 0.2295:  23%|██▎       | 43/187 [04:18<14:36,  6.08s/it]
Training Epoch: 12/100, batch 84/375 completed (loss: 0.0289:  23%|██▎       | 43/187 [04:18<14:36,  6.08s/it]
Training Epoch: 12/100, batch 84/375 completed (loss: 0.0192:  23%|██▎       | 43/187 [04:18<14:36,  6.08s/it]
Training Epoch: 12/100, batch 84/375 completed (loss: 0.09:  23%|██▎       | 43/187 [04:18<14:36,  6.09s/it]
Training Epoch: 12/100, batch 85/375 completed (loss: 0.2007:  23%|██▎       | 43/187 [04:18<14:36,  6.08s/it]
Training Epoch: 12/100, batch 85/375 completed (loss: 0.0397:  23%|██▎       | 43/187 [04:18<14:36,  6.08s/it]
Training Epoch: 12/100, batch 85/375 completed (loss: 0.2444:  23%|██▎       | 43/187 [04:18<14:36,  6.08s/it]
Training Epoch: 12/100, batch 85/375 completed (loss: 0.0628:  23%|██▎       | 43/187 [04:18<14:36,  6.09s/it]
Training Epoch: 12/100, batch 86/375 completed (loss: 0.1605:  23%|██▎       | 43/187 [04:21<14:36,  6.08s/it]
Training Epoch: 12/100, batch 86/375 completed (loss: 0.0163:  23%|██▎       | 43/187 [04:21<14:36,  6.08s/it]
Training Epoch: 12/100, batch 86/375 completed (loss: 0.0787:  23%|██▎       | 43/187 [04:21<14:36,  6.08s/it]
Training Epoch: 12/100, batch 86/375 completed (loss: 0.02:  23%|██▎       | 43/187 [04:21<14:36,  6.09s/it]  
Training Epoch: 12/100, batch 86/375 completed (loss: 0.0163:  24%|██▎       | 44/187 [04:24<14:25,  6.05s/it]
Training Epoch: 12/100, batch 86/375 completed (loss: 0.1605:  24%|██▎       | 44/187 [04:24<14:25,  6.05s/it]
Training Epoch: 12/100, batch 86/375 completed (loss: 0.02:  24%|██▎       | 44/187 [04:24<14:25,  6.05s/it]
Training Epoch: 12/100, batch 86/375 completed (loss: 0.0787:  24%|██▎       | 44/187 [04:24<14:25,  6.05s/it]
Training Epoch: 12/100, batch 87/375 completed (loss: 0.2403:  24%|██▎       | 44/187 [04:24<14:25,  6.05s/it]
Training Epoch: 12/100, batch 87/375 completed (loss: 0.0655:  24%|██▎       | 44/187 [04:24<14:25,  6.05s/it]
Training Epoch: 12/100, batch 87/375 completed (loss: 0.5346:  24%|██▎       | 44/187 [04:24<14:25,  6.05s/it]
Training Epoch: 12/100, batch 87/375 completed (loss: 0.1472:  24%|██▎       | 44/187 [04:24<14:25,  6.05s/it]
Training Epoch: 12/100, batch 88/375 completed (loss: 0.0261:  24%|██▎       | 44/187 [04:27<14:25,  6.05s/it]
Training Epoch: 12/100, batch 88/375 completed (loss: 0.2749:  24%|██▎       | 44/187 [04:27<14:25,  6.05s/it]
Training Epoch: 12/100, batch 88/375 completed (loss: 0.0365:  24%|██▎       | 44/187 [04:27<14:25,  6.05s/it]
Training Epoch: 12/100, batch 88/375 completed (loss: 0.0699:  24%|██▎       | 44/187 [04:27<14:25,  6.05s/it]
Training Epoch: 12/100, batch 88/375 completed (loss: 0.0699:  24%|██▍       | 45/187 [04:30<14:23,  6.08s/it]
Training Epoch: 12/100, batch 88/375 completed (loss: 0.0261:  24%|██▍       | 45/187 [04:30<14:24,  6.09s/it]
Training Epoch: 12/100, batch 88/375 completed (loss: 0.0365:  24%|██▍       | 45/187 [04:30<14:24,  6.09s/it]
Training Epoch: 12/100, batch 88/375 completed (loss: 0.2749:  24%|██▍       | 45/187 [04:30<14:24,  6.09s/it]
Training Epoch: 12/100, batch 89/375 completed (loss: 0.0913:  24%|██▍       | 45/187 [04:30<14:24,  6.09s/it]
Training Epoch: 12/100, batch 89/375 completed (loss: 0.1118:  24%|██▍       | 45/187 [04:30<14:24,  6.09s/it]
Training Epoch: 12/100, batch 89/375 completed (loss: 0.0125:  24%|██▍       | 45/187 [04:30<14:24,  6.09s/it]
Training Epoch: 12/100, batch 89/375 completed (loss: 0.018:  24%|██▍       | 45/187 [04:30<14:23,  6.08s/it] 
Training Epoch: 12/100, batch 90/375 completed (loss: 0.0682:  24%|██▍       | 45/187 [04:33<14:24,  6.09s/it]
Training Epoch: 12/100, batch 90/375 completed (loss: 0.0961:  24%|██▍       | 45/187 [04:33<14:24,  6.09s/it]
Training Epoch: 12/100, batch 90/375 completed (loss: 0.0377:  24%|██▍       | 45/187 [04:33<14:24,  6.09s/it]
Training Epoch: 12/100, batch 90/375 completed (loss: 0.5168:  24%|██▍       | 45/187 [04:33<14:23,  6.08s/it]
Training Epoch: 12/100, batch 90/375 completed (loss: 0.5168:  25%|██▍       | 46/187 [04:36<14:15,  6.07s/it]
Training Epoch: 12/100, batch 90/375 completed (loss: 0.0377:  25%|██▍       | 46/187 [04:36<14:15,  6.07s/it]
Training Epoch: 12/100, batch 90/375 completed (loss: 0.0961:  25%|██▍       | 46/187 [04:36<14:15,  6.07s/it]
Training Epoch: 12/100, batch 90/375 completed (loss: 0.0682:  25%|██▍       | 46/187 [04:36<14:15,  6.07s/it]
Training Epoch: 12/100, batch 91/375 completed (loss: 0.3691:  25%|██▍       | 46/187 [04:36<14:15,  6.07s/it]
Training Epoch: 12/100, batch 91/375 completed (loss: 0.011:  25%|██▍       | 46/187 [04:36<14:15,  6.07s/it] 
Training Epoch: 12/100, batch 91/375 completed (loss: 0.0382:  25%|██▍       | 46/187 [04:36<14:15,  6.07s/it]
Training Epoch: 12/100, batch 91/375 completed (loss: 0.1111:  25%|██▍       | 46/187 [04:36<14:15,  6.07s/it]
Training Epoch: 12/100, batch 92/375 completed (loss: 0.0186:  25%|██▍       | 46/187 [04:39<14:15,  6.07s/it]
Training Epoch: 12/100, batch 92/375 completed (loss: 0.0126:  25%|██▍       | 46/187 [04:39<14:15,  6.07s/it]
Training Epoch: 12/100, batch 92/375 completed (loss: 0.1665:  25%|██▍       | 46/187 [04:39<14:15,  6.07s/it]
Training Epoch: 12/100, batch 92/375 completed (loss: 0.0237:  25%|██▍       | 46/187 [04:39<14:15,  6.07s/it]
Training Epoch: 12/100, batch 92/375 completed (loss: 0.0237:  25%|██▌       | 47/187 [04:42<14:06,  6.05s/it]
Training Epoch: 12/100, batch 92/375 completed (loss: 0.1665:  25%|██▌       | 47/187 [04:42<14:06,  6.05s/it]
Training Epoch: 12/100, batch 92/375 completed (loss: 0.0186:  25%|██▌       | 47/187 [04:42<14:06,  6.05s/it]
Training Epoch: 12/100, batch 92/375 completed (loss: 0.0126:  25%|██▌       | 47/187 [04:42<14:07,  6.05s/it]
Training Epoch: 12/100, batch 93/375 completed (loss: 0.027:  25%|██▌       | 47/187 [04:42<14:06,  6.05s/it] 
Training Epoch: 12/100, batch 93/375 completed (loss: 0.0812:  25%|██▌       | 47/187 [04:42<14:06,  6.05s/it]
Training Epoch: 12/100, batch 93/375 completed (loss: 0.0567:  25%|██▌       | 47/187 [04:42<14:07,  6.05s/it]
Training Epoch: 12/100, batch 93/375 completed (loss: 0.1441:  25%|██▌       | 47/187 [04:42<14:06,  6.05s/it]
Training Epoch: 12/100, batch 94/375 completed (loss: 0.0193:  25%|██▌       | 47/187 [04:45<14:06,  6.05s/it]
Training Epoch: 12/100, batch 94/375 completed (loss: 0.1006:  25%|██▌       | 47/187 [04:45<14:07,  6.05s/it]
Training Epoch: 12/100, batch 94/375 completed (loss: 0.0809:  25%|██▌       | 47/187 [04:45<14:06,  6.05s/it]
Training Epoch: 12/100, batch 94/375 completed (loss: 0.1371:  25%|██▌       | 47/187 [04:45<14:06,  6.05s/it]
Training Epoch: 12/100, batch 94/375 completed (loss: 0.0809:  26%|██▌       | 48/187 [04:48<14:05,  6.09s/it]
Training Epoch: 12/100, batch 94/375 completed (loss: 0.1006:  26%|██▌       | 48/187 [04:48<14:05,  6.09s/it]
Training Epoch: 12/100, batch 94/375 completed (loss: 0.0193:  26%|██▌       | 48/187 [04:48<14:05,  6.09s/it]
Training Epoch: 12/100, batch 94/375 completed (loss: 0.1371:  26%|██▌       | 48/187 [04:48<14:06,  6.09s/it]
Training Epoch: 12/100, batch 95/375 completed (loss: 0.3612:  26%|██▌       | 48/187 [04:48<14:05,  6.09s/it]
Training Epoch: 12/100, batch 95/375 completed (loss: 0.0182:  26%|██▌       | 48/187 [04:48<14:05,  6.09s/it]
Training Epoch: 12/100, batch 95/375 completed (loss: 0.3355:  26%|██▌       | 48/187 [04:48<14:05,  6.09s/it]
Training Epoch: 12/100, batch 95/375 completed (loss: 0.0103:  26%|██▌       | 48/187 [04:48<14:06,  6.09s/it]
Training Epoch: 12/100, batch 96/375 completed (loss: 0.1833:  26%|██▌       | 48/187 [04:51<14:05,  6.09s/it]
Training Epoch: 12/100, batch 96/375 completed (loss: 0.1686:  26%|██▌       | 48/187 [04:51<14:05,  6.09s/it]
Training Epoch: 12/100, batch 96/375 completed (loss: 0.0832:  26%|██▌       | 48/187 [04:51<14:05,  6.09s/it]
Training Epoch: 12/100, batch 96/375 completed (loss: 0.0739:  26%|██▌       | 48/187 [04:51<14:06,  6.09s/it]
Training Epoch: 12/100, batch 96/375 completed (loss: 0.1833:  26%|██▌       | 49/187 [04:54<13:57,  6.07s/it]
Training Epoch: 12/100, batch 96/375 completed (loss: 0.0739:  26%|██▌       | 49/187 [04:54<13:57,  6.07s/it]
Training Epoch: 12/100, batch 96/375 completed (loss: 0.1686:  26%|██▌       | 49/187 [04:54<13:57,  6.07s/it]
Training Epoch: 12/100, batch 96/375 completed (loss: 0.0832:  26%|██▌       | 49/187 [04:54<13:57,  6.07s/it]
Training Epoch: 12/100, batch 97/375 completed (loss: 0.042:  26%|██▌       | 49/187 [04:54<13:57,  6.07s/it] 
Training Epoch: 12/100, batch 97/375 completed (loss: 0.0155:  26%|██▌       | 49/187 [04:54<13:57,  6.07s/it]
Training Epoch: 12/100, batch 97/375 completed (loss: 0.0253:  26%|██▌       | 49/187 [04:54<13:57,  6.07s/it]
Training Epoch: 12/100, batch 97/375 completed (loss: 0.0791:  26%|██▌       | 49/187 [04:54<13:57,  6.07s/it]
Training Epoch: 12/100, batch 98/375 completed (loss: 0.1786:  26%|██▌       | 49/187 [04:57<13:57,  6.07s/it]
Training Epoch: 12/100, batch 98/375 completed (loss: 0.133:  26%|██▌       | 49/187 [04:57<13:57,  6.07s/it] 
Training Epoch: 12/100, batch 98/375 completed (loss: 0.1878:  26%|██▌       | 49/187 [04:57<13:57,  6.07s/it]
Training Epoch: 12/100, batch 98/375 completed (loss: 0.1125:  26%|██▌       | 49/187 [04:57<13:57,  6.07s/it]
Training Epoch: 12/100, batch 98/375 completed (loss: 0.1125:  27%|██▋       | 50/187 [05:00<13:48,  6.05s/it]
Training Epoch: 12/100, batch 98/375 completed (loss: 0.1878:  27%|██▋       | 50/187 [05:00<13:48,  6.05s/it]
Training Epoch: 12/100, batch 98/375 completed (loss: 0.1786:  27%|██▋       | 50/187 [05:00<13:48,  6.05s/it]
Training Epoch: 12/100, batch 98/375 completed (loss: 0.133:  27%|██▋       | 50/187 [05:00<13:48,  6.05s/it]
Training Epoch: 12/100, batch 99/375 completed (loss: 0.0423:  27%|██▋       | 50/187 [05:00<13:48,  6.05s/it]
Training Epoch: 12/100, batch 99/375 completed (loss: 0.0322:  27%|██▋       | 50/187 [05:00<13:48,  6.05s/it]
Training Epoch: 12/100, batch 99/375 completed (loss: 0.0588:  27%|██▋       | 50/187 [05:00<13:48,  6.05s/it]
Training Epoch: 12/100, batch 99/375 completed (loss: 0.0318:  27%|██▋       | 50/187 [05:00<13:48,  6.05s/it]
Training Epoch: 12/100, batch 100/375 completed (loss: 0.0178:  27%|██▋       | 50/187 [05:03<13:48,  6.05s/it]
Training Epoch: 12/100, batch 100/375 completed (loss: 0.0516:  27%|██▋       | 50/187 [05:03<13:48,  6.05s/it]
Training Epoch: 12/100, batch 100/375 completed (loss: 0.066:  27%|██▋       | 50/187 [05:03<13:48,  6.05s/it]
Training Epoch: 12/100, batch 100/375 completed (loss: 0.1474:  27%|██▋       | 50/187 [05:03<13:48,  6.05s/it]
Training Epoch: 12/100, batch 100/375 completed (loss: 0.0516:  27%|██▋       | 51/187 [05:06<13:40,  6.03s/it]
Training Epoch: 12/100, batch 100/375 completed (loss: 0.0178:  27%|██▋       | 51/187 [05:06<13:40,  6.03s/it]
Training Epoch: 12/100, batch 100/375 completed (loss: 0.1474:  27%|██▋       | 51/187 [05:06<13:40,  6.03s/it]
Training Epoch: 12/100, batch 100/375 completed (loss: 0.066:  27%|██▋       | 51/187 [05:06<13:40,  6.03s/it]
Training Epoch: 12/100, batch 101/375 completed (loss: 0.2705:  27%|██▋       | 51/187 [05:06<13:40,  6.03s/it]
Training Epoch: 12/100, batch 101/375 completed (loss: 0.0194:  27%|██▋       | 51/187 [05:06<13:40,  6.03s/it]
Training Epoch: 12/100, batch 101/375 completed (loss: 0.0159:  27%|██▋       | 51/187 [05:06<13:40,  6.03s/it]
Training Epoch: 12/100, batch 101/375 completed (loss: 0.042:  27%|██▋       | 51/187 [05:06<13:40,  6.03s/it] 
Training Epoch: 12/100, batch 102/375 completed (loss: 0.0344:  27%|██▋       | 51/187 [05:09<13:40,  6.03s/it]
Training Epoch: 12/100, batch 102/375 completed (loss: 0.0299:  27%|██▋       | 51/187 [05:09<13:40,  6.03s/it]
Training Epoch: 12/100, batch 102/375 completed (loss: 0.0272:  27%|██▋       | 51/187 [05:09<13:40,  6.03s/it]
Training Epoch: 12/100, batch 102/375 completed (loss: 0.1213:  27%|██▋       | 51/187 [05:09<13:40,  6.03s/it]
Training Epoch: 12/100, batch 102/375 completed (loss: 0.1213:  28%|██▊       | 52/187 [05:12<13:34,  6.03s/it]
Training Epoch: 12/100, batch 102/375 completed (loss: 0.0272:  28%|██▊       | 52/187 [05:12<13:34,  6.03s/it]
Training Epoch: 12/100, batch 102/375 completed (loss: 0.0344:  28%|██▊       | 52/187 [05:12<13:34,  6.03s/it]
Training Epoch: 12/100, batch 102/375 completed (loss: 0.0299:  28%|██▊       | 52/187 [05:12<13:34,  6.03s/it]
Training Epoch: 12/100, batch 103/375 completed (loss: 0.0281:  28%|██▊       | 52/187 [05:12<13:34,  6.03s/it]
Training Epoch: 12/100, batch 103/375 completed (loss: 0.126:  28%|██▊       | 52/187 [05:12<13:34,  6.03s/it] 
Training Epoch: 12/100, batch 103/375 completed (loss: 0.0958:  28%|██▊       | 52/187 [05:12<13:34,  6.03s/it]
Training Epoch: 12/100, batch 103/375 completed (loss: 0.032:  28%|██▊       | 52/187 [05:12<13:34,  6.03s/it] 
Training Epoch: 12/100, batch 104/375 completed (loss: 0.1656:  28%|██▊       | 52/187 [05:15<13:34,  6.03s/it]
Training Epoch: 12/100, batch 104/375 completed (loss: 0.2955:  28%|██▊       | 52/187 [05:15<13:34,  6.03s/it]
Training Epoch: 12/100, batch 104/375 completed (loss: 0.1996:  28%|██▊       | 52/187 [05:15<13:34,  6.03s/it]
Training Epoch: 12/100, batch 104/375 completed (loss: 0.0788:  28%|██▊       | 52/187 [05:15<13:34,  6.03s/it]
Training Epoch: 12/100, batch 104/375 completed (loss: 0.0788:  28%|██▊       | 53/187 [05:18<13:33,  6.07s/it]
Training Epoch: 12/100, batch 104/375 completed (loss: 0.2955:  28%|██▊       | 53/187 [05:18<13:34,  6.07s/it]
Training Epoch: 12/100, batch 104/375 completed (loss: 0.1656:  28%|██▊       | 53/187 [05:18<13:34,  6.08s/it]
Training Epoch: 12/100, batch 104/375 completed (loss: 0.1996:  28%|██▊       | 53/187 [05:18<13:34,  6.08s/it]
Training Epoch: 12/100, batch 105/375 completed (loss: 0.466:  28%|██▊       | 53/187 [05:18<13:34,  6.08s/it] 
Training Epoch: 12/100, batch 105/375 completed (loss: 0.0942:  28%|██▊       | 53/187 [05:18<13:34,  6.08s/it]
Training Epoch: 12/100, batch 105/375 completed (loss: 0.1208:  28%|██▊       | 53/187 [05:18<13:34,  6.07s/it]
Training Epoch: 12/100, batch 105/375 completed (loss: 0.2008:  28%|██▊       | 53/187 [05:18<13:33,  6.07s/it]
Training Epoch: 12/100, batch 106/375 completed (loss: 0.5057:  28%|██▊       | 53/187 [05:21<13:34,  6.07s/it]
Training Epoch: 12/100, batch 106/375 completed (loss: 0.0509:  28%|██▊       | 53/187 [05:21<13:34,  6.08s/it]
Training Epoch: 12/100, batch 106/375 completed (loss: 0.224:  28%|██▊       | 53/187 [05:21<13:34,  6.08s/it] 
Training Epoch: 12/100, batch 106/375 completed (loss: 0.1244:  28%|██▊       | 53/187 [05:21<13:33,  6.07s/it]
Training Epoch: 12/100, batch 106/375 completed (loss: 0.5057:  29%|██▉       | 54/187 [05:24<13:23,  6.04s/it]
Training Epoch: 12/100, batch 106/375 completed (loss: 0.224:  29%|██▉       | 54/187 [05:24<13:23,  6.04s/it]
Training Epoch: 12/100, batch 106/375 completed (loss: 0.1244:  29%|██▉       | 54/187 [05:24<13:23,  6.04s/it]
Training Epoch: 12/100, batch 106/375 completed (loss: 0.0509:  29%|██▉       | 54/187 [05:24<13:24,  6.05s/it]
Training Epoch: 12/100, batch 107/375 completed (loss: 0.0426:  29%|██▉       | 54/187 [05:24<13:24,  6.05s/it]
Training Epoch: 12/100, batch 107/375 completed (loss: 0.1883:  29%|██▉       | 54/187 [05:24<13:23,  6.04s/it]
Training Epoch: 12/100, batch 107/375 completed (loss: 0.041:  29%|██▉       | 54/187 [05:24<13:23,  6.04s/it]
Training Epoch: 12/100, batch 107/375 completed (loss: 0.1188:  29%|██▉       | 54/187 [05:24<13:23,  6.04s/it]
Training Epoch: 12/100, batch 108/375 completed (loss: 0.095:  29%|██▉       | 54/187 [05:27<13:23,  6.04s/it] 
Training Epoch: 12/100, batch 108/375 completed (loss: 0.0275:  29%|██▉       | 54/187 [05:27<13:24,  6.05s/it]
Training Epoch: 12/100, batch 108/375 completed (loss: 0.0279:  29%|██▉       | 54/187 [05:27<13:23,  6.04s/it]
Training Epoch: 12/100, batch 108/375 completed (loss: 0.0451:  29%|██▉       | 54/187 [05:27<13:23,  6.04s/it]
Training Epoch: 12/100, batch 108/375 completed (loss: 0.0451:  29%|██▉       | 55/187 [05:30<13:15,  6.02s/it]
Training Epoch: 12/100, batch 108/375 completed (loss: 0.095:  29%|██▉       | 55/187 [05:30<13:15,  6.02s/it]
Training Epoch: 12/100, batch 108/375 completed (loss: 0.0279:  29%|██▉       | 55/187 [05:30<13:14,  6.02s/it]
Training Epoch: 12/100, batch 108/375 completed (loss: 0.0275:  29%|██▉       | 55/187 [05:30<13:15,  6.02s/it]
Training Epoch: 12/100, batch 109/375 completed (loss: 0.1659:  29%|██▉       | 55/187 [05:30<13:15,  6.02s/it]
Training Epoch: 12/100, batch 109/375 completed (loss: 0.1107:  29%|██▉       | 55/187 [05:30<13:14,  6.02s/it]
Training Epoch: 12/100, batch 109/375 completed (loss: 0.0425:  29%|██▉       | 55/187 [05:30<13:15,  6.02s/it]
Training Epoch: 12/100, batch 109/375 completed (loss: 0.0534:  29%|██▉       | 55/187 [05:30<13:15,  6.02s/it]
Training Epoch: 12/100, batch 110/375 completed (loss: 0.0953:  29%|██▉       | 55/187 [05:33<13:15,  6.02s/it]
Training Epoch: 12/100, batch 110/375 completed (loss: 0.058:  29%|██▉       | 55/187 [05:33<13:15,  6.02s/it] 
Training Epoch: 12/100, batch 110/375 completed (loss: 0.1221:  29%|██▉       | 55/187 [05:33<13:14,  6.02s/it]
Training Epoch: 12/100, batch 110/375 completed (loss: 0.0128:  29%|██▉       | 55/187 [05:33<13:15,  6.02s/it]
Training Epoch: 12/100, batch 110/375 completed (loss: 0.1221:  30%|██▉       | 56/187 [05:36<13:17,  6.09s/it]
Training Epoch: 12/100, batch 110/375 completed (loss: 0.0128:  30%|██▉       | 56/187 [05:36<13:17,  6.09s/it]
Training Epoch: 12/100, batch 110/375 completed (loss: 0.0953:  30%|██▉       | 56/187 [05:36<13:17,  6.09s/it]
Training Epoch: 12/100, batch 110/375 completed (loss: 0.058:  30%|██▉       | 56/187 [05:36<13:17,  6.09s/it]
Training Epoch: 12/100, batch 111/375 completed (loss: 0.048:  30%|██▉       | 56/187 [05:36<13:17,  6.09s/it]
Training Epoch: 12/100, batch 111/375 completed (loss: 0.0713:  30%|██▉       | 56/187 [05:36<13:17,  6.09s/it]
Training Epoch: 12/100, batch 111/375 completed (loss: 0.0327:  30%|██▉       | 56/187 [05:36<13:17,  6.09s/it]
Training Epoch: 12/100, batch 111/375 completed (loss: 0.1022:  30%|██▉       | 56/187 [05:36<13:17,  6.09s/it]
Training Epoch: 12/100, batch 112/375 completed (loss: 0.0602:  30%|██▉       | 56/187 [05:39<13:17,  6.09s/it]
Training Epoch: 12/100, batch 112/375 completed (loss: 0.0341:  30%|██▉       | 56/187 [05:39<13:17,  6.09s/it]
Training Epoch: 12/100, batch 112/375 completed (loss: 0.4055:  30%|██▉       | 56/187 [05:39<13:17,  6.09s/it]
Training Epoch: 12/100, batch 112/375 completed (loss: 0.2746:  30%|██▉       | 56/187 [05:39<13:17,  6.09s/it]
Training Epoch: 12/100, batch 112/375 completed (loss: 0.0341:  30%|███       | 57/187 [05:42<13:02,  6.02s/it]
Training Epoch: 12/100, batch 112/375 completed (loss: 0.4055:  30%|███       | 57/187 [05:42<13:02,  6.02s/it]
Training Epoch: 12/100, batch 112/375 completed (loss: 0.2746:  30%|███       | 57/187 [05:42<13:02,  6.02s/it]
Training Epoch: 12/100, batch 112/375 completed (loss: 0.0602:  30%|███       | 57/187 [05:42<13:02,  6.02s/it]
Training Epoch: 12/100, batch 113/375 completed (loss: 0.0161:  30%|███       | 57/187 [05:42<13:02,  6.02s/it]
Training Epoch: 12/100, batch 113/375 completed (loss: 0.0757:  30%|███       | 57/187 [05:42<13:02,  6.02s/it]
Training Epoch: 12/100, batch 113/375 completed (loss: 0.0113:  30%|███       | 57/187 [05:42<13:02,  6.02s/it]
Training Epoch: 12/100, batch 113/375 completed (loss: 0.015:  30%|███       | 57/187 [05:42<13:02,  6.02s/it] 
Training Epoch: 12/100, batch 114/375 completed (loss: 0.1308:  30%|███       | 57/187 [05:45<13:02,  6.02s/it]
Training Epoch: 12/100, batch 114/375 completed (loss: 0.0617:  30%|███       | 57/187 [05:45<13:02,  6.02s/it]
Training Epoch: 12/100, batch 114/375 completed (loss: 0.1164:  30%|███       | 57/187 [05:45<13:02,  6.02s/it]
Training Epoch: 12/100, batch 114/375 completed (loss: 0.0086:  30%|███       | 57/187 [05:45<13:02,  6.02s/it]
Training Epoch: 12/100, batch 114/375 completed (loss: 0.0086:  31%|███       | 58/187 [05:49<13:05,  6.09s/it]
Training Epoch: 12/100, batch 114/375 completed (loss: 0.0617:  31%|███       | 58/187 [05:48<13:05,  6.09s/it]
Training Epoch: 12/100, batch 114/375 completed (loss: 0.1164:  31%|███       | 58/187 [05:48<13:05,  6.09s/it]
Training Epoch: 12/100, batch 114/375 completed (loss: 0.1308:  31%|███       | 58/187 [05:48<13:05,  6.09s/it]
Training Epoch: 12/100, batch 115/375 completed (loss: 0.0327:  31%|███       | 58/187 [05:49<13:05,  6.09s/it]
Training Epoch: 12/100, batch 115/375 completed (loss: 0.0379:  31%|███       | 58/187 [05:49<13:05,  6.09s/it]
Training Epoch: 12/100, batch 115/375 completed (loss: 0.1203:  31%|███       | 58/187 [05:49<13:05,  6.09s/it]
Training Epoch: 12/100, batch 115/375 completed (loss: 0.0156:  31%|███       | 58/187 [05:49<13:05,  6.09s/it]
Training Epoch: 12/100, batch 116/375 completed (loss: 0.0434:  31%|███       | 58/187 [05:52<13:05,  6.09s/it]
Training Epoch: 12/100, batch 116/375 completed (loss: 0.0077:  31%|███       | 58/187 [05:52<13:05,  6.09s/it]
Training Epoch: 12/100, batch 116/375 completed (loss: 0.1462:  31%|███       | 58/187 [05:52<13:05,  6.09s/it]
Training Epoch: 12/100, batch 116/375 completed (loss: 0.0934:  31%|███       | 58/187 [05:52<13:05,  6.09s/it]
Training Epoch: 12/100, batch 116/375 completed (loss: 0.0934:  32%|███▏      | 59/187 [05:55<12:55,  6.06s/it]
Training Epoch: 12/100, batch 116/375 completed (loss: 0.0077:  32%|███▏      | 59/187 [05:54<12:55,  6.06s/it]
Training Epoch: 12/100, batch 116/375 completed (loss: 0.0434:  32%|███▏      | 59/187 [05:54<12:55,  6.06s/it]
Training Epoch: 12/100, batch 116/375 completed (loss: 0.1462:  32%|███▏      | 59/187 [05:54<12:55,  6.06s/it]
Training Epoch: 12/100, batch 117/375 completed (loss: 0.0244:  32%|███▏      | 59/187 [05:55<12:55,  6.06s/it]
Training Epoch: 12/100, batch 117/375 completed (loss: 0.1104:  32%|███▏      | 59/187 [05:55<12:55,  6.06s/it]
Training Epoch: 12/100, batch 117/375 completed (loss: 0.1382:  32%|███▏      | 59/187 [05:55<12:55,  6.06s/it]
Training Epoch: 12/100, batch 117/375 completed (loss: 0.0703:  32%|███▏      | 59/187 [05:55<12:55,  6.06s/it]
Training Epoch: 12/100, batch 118/375 completed (loss: 0.0811:  32%|███▏      | 59/187 [05:57<12:55,  6.06s/it]
Training Epoch: 12/100, batch 118/375 completed (loss: 0.0322:  32%|███▏      | 59/187 [05:57<12:55,  6.06s/it]
Training Epoch: 12/100, batch 118/375 completed (loss: 0.0473:  32%|███▏      | 59/187 [05:57<12:55,  6.06s/it]
Training Epoch: 12/100, batch 118/375 completed (loss: 0.0524:  32%|███▏      | 59/187 [05:57<12:55,  6.06s/it]
Training Epoch: 12/100, batch 118/375 completed (loss: 0.0322:  32%|███▏      | 60/187 [06:01<12:48,  6.05s/it]
Training Epoch: 12/100, batch 118/375 completed (loss: 0.0524:  32%|███▏      | 60/187 [06:01<12:48,  6.05s/it]
Training Epoch: 12/100, batch 118/375 completed (loss: 0.0811:  32%|███▏      | 60/187 [06:01<12:48,  6.05s/it]
Training Epoch: 12/100, batch 118/375 completed (loss: 0.0473:  32%|███▏      | 60/187 [06:01<12:48,  6.05s/it]
Training Epoch: 12/100, batch 119/375 completed (loss: 0.0267:  32%|███▏      | 60/187 [06:01<12:48,  6.05s/it]
Training Epoch: 12/100, batch 119/375 completed (loss: 0.1242:  32%|███▏      | 60/187 [06:01<12:48,  6.05s/it]
Training Epoch: 12/100, batch 119/375 completed (loss: 0.0462:  32%|███▏      | 60/187 [06:01<12:48,  6.05s/it]
Training Epoch: 12/100, batch 119/375 completed (loss: 0.042:  32%|███▏      | 60/187 [06:01<12:48,  6.05s/it] 
Training Epoch: 12/100, batch 120/375 completed (loss: 0.0692:  32%|███▏      | 60/187 [06:04<12:48,  6.05s/it]
Training Epoch: 12/100, batch 120/375 completed (loss: 0.1139:  32%|███▏      | 60/187 [06:04<12:48,  6.05s/it]
Training Epoch: 12/100, batch 120/375 completed (loss: 0.0513:  32%|███▏      | 60/187 [06:04<12:48,  6.05s/it]
Training Epoch: 12/100, batch 120/375 completed (loss: 0.2972:  32%|███▏      | 60/187 [06:04<12:48,  6.05s/it]
Training Epoch: 12/100, batch 120/375 completed (loss: 0.1139:  33%|███▎      | 61/187 [06:07<12:51,  6.12s/it]
Training Epoch: 12/100, batch 120/375 completed (loss: 0.2972:  33%|███▎      | 61/187 [06:07<12:51,  6.12s/it]
Training Epoch: 12/100, batch 120/375 completed (loss: 0.0692:  33%|███▎      | 61/187 [06:07<12:51,  6.12s/it]
Training Epoch: 12/100, batch 120/375 completed (loss: 0.0513:  33%|███▎      | 61/187 [06:07<12:51,  6.12s/it]
Training Epoch: 12/100, batch 121/375 completed (loss: 0.0664:  33%|███▎      | 61/187 [06:07<12:51,  6.12s/it]
Training Epoch: 12/100, batch 121/375 completed (loss: 0.092:  33%|███▎      | 61/187 [06:07<12:51,  6.12s/it] 
Training Epoch: 12/100, batch 121/375 completed (loss: 0.0806:  33%|███▎      | 61/187 [06:07<12:51,  6.12s/it]
Training Epoch: 12/100, batch 121/375 completed (loss: 0.1123:  33%|███▎      | 61/187 [06:07<12:51,  6.12s/it]
Training Epoch: 12/100, batch 122/375 completed (loss: 0.0496:  33%|███▎      | 61/187 [06:10<12:51,  6.12s/it]
Training Epoch: 12/100, batch 122/375 completed (loss: 0.1242:  33%|███▎      | 61/187 [06:10<12:51,  6.12s/it]
Training Epoch: 12/100, batch 122/375 completed (loss: 0.0877:  33%|███▎      | 61/187 [06:10<12:51,  6.12s/it]
Training Epoch: 12/100, batch 122/375 completed (loss: 0.1304:  33%|███▎      | 61/187 [06:10<12:51,  6.12s/it]
Training Epoch: 12/100, batch 122/375 completed (loss: 0.0496:  33%|███▎      | 62/187 [06:13<12:34,  6.04s/it]
Training Epoch: 12/100, batch 122/375 completed (loss: 0.0877:  33%|███▎      | 62/187 [06:13<12:34,  6.04s/it]
Training Epoch: 12/100, batch 122/375 completed (loss: 0.1242:  33%|███▎      | 62/187 [06:13<12:34,  6.04s/it]
Training Epoch: 12/100, batch 122/375 completed (loss: 0.1304:  33%|███▎      | 62/187 [06:13<12:34,  6.04s/it]
Training Epoch: 12/100, batch 123/375 completed (loss: 0.1361:  33%|███▎      | 62/187 [06:13<12:34,  6.04s/it]
Training Epoch: 12/100, batch 123/375 completed (loss: 0.0289:  33%|███▎      | 62/187 [06:13<12:34,  6.04s/it]
Training Epoch: 12/100, batch 123/375 completed (loss: 0.1528:  33%|███▎      | 62/187 [06:13<12:34,  6.04s/it]
Training Epoch: 12/100, batch 123/375 completed (loss: 0.042:  33%|███▎      | 62/187 [06:13<12:34,  6.04s/it] 
Training Epoch: 12/100, batch 124/375 completed (loss: 0.046:  33%|███▎      | 62/187 [06:16<12:34,  6.04s/it] 
Training Epoch: 12/100, batch 124/375 completed (loss: 0.0248:  33%|███▎      | 62/187 [06:16<12:34,  6.04s/it]
Training Epoch: 12/100, batch 124/375 completed (loss: 0.0239:  33%|███▎      | 62/187 [06:16<12:34,  6.04s/it]
Training Epoch: 12/100, batch 124/375 completed (loss: 0.2968:  33%|███▎      | 62/187 [06:16<12:34,  6.04s/it]
Training Epoch: 12/100, batch 124/375 completed (loss: 0.2968:  34%|███▎      | 63/187 [06:19<12:37,  6.11s/it]
Training Epoch: 12/100, batch 124/375 completed (loss: 0.0248:  34%|███▎      | 63/187 [06:19<12:38,  6.12s/it]
Training Epoch: 12/100, batch 124/375 completed (loss: 0.046:  34%|███▎      | 63/187 [06:19<12:38,  6.12s/it]
Training Epoch: 12/100, batch 124/375 completed (loss: 0.0239:  34%|███▎      | 63/187 [06:19<12:39,  6.12s/it]
Training Epoch: 12/100, batch 125/375 completed (loss: 0.0417:  34%|███▎      | 63/187 [06:19<12:38,  6.12s/it]
Training Epoch: 12/100, batch 125/375 completed (loss: 0.0914:  34%|███▎      | 63/187 [06:19<12:39,  6.12s/it]
Training Epoch: 12/100, batch 125/375 completed (loss: 0.0456:  34%|███▎      | 63/187 [06:19<12:38,  6.12s/it]
Training Epoch: 12/100, batch 125/375 completed (loss: 0.138:  34%|███▎      | 63/187 [06:19<12:37,  6.11s/it] 
Training Epoch: 12/100, batch 126/375 completed (loss: 0.1591:  34%|███▎      | 63/187 [06:22<12:39,  6.12s/it]
Training Epoch: 12/100, batch 126/375 completed (loss: 0.0343:  34%|███▎      | 63/187 [06:22<12:38,  6.12s/it]
Training Epoch: 12/100, batch 126/375 completed (loss: 0.0587:  34%|███▎      | 63/187 [06:22<12:38,  6.12s/it]
Training Epoch: 12/100, batch 126/375 completed (loss: 0.0764:  34%|███▎      | 63/187 [06:22<12:37,  6.11s/it]
Training Epoch: 12/100, batch 126/375 completed (loss: 0.0764:  34%|███▍      | 64/187 [06:25<12:28,  6.09s/it]
Training Epoch: 12/100, batch 126/375 completed (loss: 0.0587:  34%|███▍      | 64/187 [06:25<12:28,  6.09s/it]
Training Epoch: 12/100, batch 126/375 completed (loss: 0.1591:  34%|███▍      | 64/187 [06:25<12:28,  6.09s/it]
Training Epoch: 12/100, batch 126/375 completed (loss: 0.0343:  34%|███▍      | 64/187 [06:25<12:28,  6.09s/it]
Training Epoch: 12/100, batch 127/375 completed (loss: 0.2607:  34%|███▍      | 64/187 [06:25<12:28,  6.09s/it]
Training Epoch: 12/100, batch 127/375 completed (loss: 0.0151:  34%|███▍      | 64/187 [06:25<12:28,  6.09s/it]
Training Epoch: 12/100, batch 127/375 completed (loss: 0.1495:  34%|███▍      | 64/187 [06:25<12:28,  6.09s/it]
Training Epoch: 12/100, batch 127/375 completed (loss: 0.0141:  34%|███▍      | 64/187 [06:25<12:28,  6.09s/it]
Training Epoch: 12/100, batch 128/375 completed (loss: 0.0518:  34%|███▍      | 64/187 [06:28<12:28,  6.09s/it]
Training Epoch: 12/100, batch 128/375 completed (loss: 0.0214:  34%|███▍      | 64/187 [06:28<12:28,  6.09s/it]
Training Epoch: 12/100, batch 128/375 completed (loss: 1.0418:  34%|███▍      | 64/187 [06:28<12:28,  6.09s/it]
Training Epoch: 12/100, batch 128/375 completed (loss: 0.1467:  34%|███▍      | 64/187 [06:28<12:28,  6.09s/it]
Training Epoch: 12/100, batch 128/375 completed (loss: 0.0214:  35%|███▍      | 65/187 [06:31<12:19,  6.06s/it]
Training Epoch: 12/100, batch 128/375 completed (loss: 0.1467:  35%|███▍      | 65/187 [06:31<12:20,  6.07s/it]
Training Epoch: 12/100, batch 128/375 completed (loss: 1.0418:  35%|███▍      | 65/187 [06:31<12:19,  6.06s/it]
Training Epoch: 12/100, batch 128/375 completed (loss: 0.0518:  35%|███▍      | 65/187 [06:31<12:19,  6.06s/it]
Training Epoch: 12/100, batch 129/375 completed (loss: 0.0673:  35%|███▍      | 65/187 [06:31<12:19,  6.06s/it]
Training Epoch: 12/100, batch 129/375 completed (loss: 0.0506:  35%|███▍      | 65/187 [06:31<12:19,  6.06s/it]
Training Epoch: 12/100, batch 129/375 completed (loss: 0.3049:  35%|███▍      | 65/187 [06:31<12:19,  6.06s/it]
Training Epoch: 12/100, batch 129/375 completed (loss: 0.2693:  35%|███▍      | 65/187 [06:31<12:20,  6.07s/it]
Training Epoch: 12/100, batch 130/375 completed (loss: 0.0067:  35%|███▍      | 65/187 [06:34<12:19,  6.06s/it]
Training Epoch: 12/100, batch 130/375 completed (loss: 0.0408:  35%|███▍      | 65/187 [06:34<12:19,  6.06s/it]
Training Epoch: 12/100, batch 130/375 completed (loss: 0.1241:  35%|███▍      | 65/187 [06:34<12:19,  6.06s/it]
Training Epoch: 12/100, batch 130/375 completed (loss: 0.0448:  35%|███▍      | 65/187 [06:34<12:20,  6.07s/it]
Training Epoch: 12/100, batch 130/375 completed (loss: 0.0448:  35%|███▌      | 66/187 [06:37<12:12,  6.05s/it]
Training Epoch: 12/100, batch 130/375 completed (loss: 0.0067:  35%|███▌      | 66/187 [06:37<12:12,  6.05s/it]
Training Epoch: 12/100, batch 130/375 completed (loss: 0.1241:  35%|███▌      | 66/187 [06:37<12:12,  6.05s/it]
Training Epoch: 12/100, batch 130/375 completed (loss: 0.0408:  35%|███▌      | 66/187 [06:37<12:12,  6.05s/it]
Training Epoch: 12/100, batch 131/375 completed (loss: 0.065:  35%|███▌      | 66/187 [06:37<12:12,  6.05s/it] 
Training Epoch: 12/100, batch 131/375 completed (loss: 0.012:  35%|███▌      | 66/187 [06:37<12:12,  6.05s/it] 
Training Epoch: 12/100, batch 131/375 completed (loss: 0.096:  35%|███▌      | 66/187 [06:37<12:12,  6.05s/it] 
Training Epoch: 12/100, batch 131/375 completed (loss: 0.0307:  35%|███▌      | 66/187 [06:37<12:12,  6.05s/it]
Training Epoch: 12/100, batch 132/375 completed (loss: 0.0204:  35%|███▌      | 66/187 [06:40<12:12,  6.05s/it]
Training Epoch: 12/100, batch 132/375 completed (loss: 0.1322:  35%|███▌      | 66/187 [06:40<12:12,  6.05s/it]
Training Epoch: 12/100, batch 132/375 completed (loss: 0.0318:  35%|███▌      | 66/187 [06:40<12:12,  6.05s/it]
Training Epoch: 12/100, batch 132/375 completed (loss: 0.0203:  35%|███▌      | 66/187 [06:40<12:12,  6.05s/it]
Training Epoch: 12/100, batch 132/375 completed (loss: 0.0204:  36%|███▌      | 67/187 [06:43<12:03,  6.03s/it]
Training Epoch: 12/100, batch 132/375 completed (loss: 0.0203:  36%|███▌      | 67/187 [06:43<12:04,  6.04s/it]
Training Epoch: 12/100, batch 132/375 completed (loss: 0.0318:  36%|███▌      | 67/187 [06:43<12:03,  6.03s/it]
Training Epoch: 12/100, batch 132/375 completed (loss: 0.1322:  36%|███▌      | 67/187 [06:43<12:04,  6.03s/it]
Training Epoch: 12/100, batch 133/375 completed (loss: 0.0947:  36%|███▌      | 67/187 [06:43<12:04,  6.03s/it]
Training Epoch: 12/100, batch 133/375 completed (loss: 0.1136:  36%|███▌      | 67/187 [06:43<12:03,  6.03s/it]
Training Epoch: 12/100, batch 133/375 completed (loss: 0.0851:  36%|███▌      | 67/187 [06:43<12:03,  6.03s/it]
Training Epoch: 12/100, batch 133/375 completed (loss: 0.033:  36%|███▌      | 67/187 [06:43<12:04,  6.04s/it] 
Training Epoch: 12/100, batch 134/375 completed (loss: 0.0688:  36%|███▌      | 67/187 [06:46<12:03,  6.03s/it]
Training Epoch: 12/100, batch 134/375 completed (loss: 0.3184:  36%|███▌      | 67/187 [06:46<12:04,  6.03s/it]
Training Epoch: 12/100, batch 134/375 completed (loss: 0.2225:  36%|███▌      | 67/187 [06:46<12:03,  6.03s/it]
Training Epoch: 12/100, batch 134/375 completed (loss: 0.0164:  36%|███▌      | 67/187 [06:46<12:04,  6.04s/it]
Training Epoch: 12/100, batch 134/375 completed (loss: 0.2225:  36%|███▋      | 68/187 [06:49<11:57,  6.03s/it]
Training Epoch: 12/100, batch 134/375 completed (loss: 0.3184:  36%|███▋      | 68/187 [06:49<11:57,  6.03s/it]
Training Epoch: 12/100, batch 134/375 completed (loss: 0.0164:  36%|███▋      | 68/187 [06:49<11:57,  6.03s/it]
Training Epoch: 12/100, batch 134/375 completed (loss: 0.0688:  36%|███▋      | 68/187 [06:49<11:57,  6.03s/it]
Training Epoch: 12/100, batch 135/375 completed (loss: 0.3162:  36%|███▋      | 68/187 [06:49<11:57,  6.03s/it]
Training Epoch: 12/100, batch 135/375 completed (loss: 0.093:  36%|███▋      | 68/187 [06:49<11:57,  6.03s/it] 
Training Epoch: 12/100, batch 135/375 completed (loss: 0.0784:  36%|███▋      | 68/187 [06:49<11:57,  6.03s/it]
Training Epoch: 12/100, batch 135/375 completed (loss: 0.2409:  36%|███▋      | 68/187 [06:49<11:57,  6.03s/it]
Training Epoch: 12/100, batch 136/375 completed (loss: 0.1543:  36%|███▋      | 68/187 [06:52<11:57,  6.03s/it]
Training Epoch: 12/100, batch 136/375 completed (loss: 0.141:  36%|███▋      | 68/187 [06:52<11:57,  6.03s/it] 
Training Epoch: 12/100, batch 136/375 completed (loss: 0.035:  36%|███▋      | 68/187 [06:52<11:57,  6.03s/it]
Training Epoch: 12/100, batch 136/375 completed (loss: 0.07:  36%|███▋      | 68/187 [06:52<11:57,  6.03s/it]  
Training Epoch: 12/100, batch 136/375 completed (loss: 0.035:  37%|███▋      | 69/187 [06:55<12:00,  6.10s/it]
Training Epoch: 12/100, batch 136/375 completed (loss: 0.141:  37%|███▋      | 69/187 [06:55<12:00,  6.10s/it]
Training Epoch: 12/100, batch 136/375 completed (loss: 0.07:  37%|███▋      | 69/187 [06:55<12:00,  6.10s/it]
Training Epoch: 12/100, batch 136/375 completed (loss: 0.1543:  37%|███▋      | 69/187 [06:55<12:00,  6.10s/it]
Training Epoch: 12/100, batch 137/375 completed (loss: 0.0921:  37%|███▋      | 69/187 [06:55<12:00,  6.10s/it]
Training Epoch: 12/100, batch 137/375 completed (loss: 0.0418:  37%|███▋      | 69/187 [06:55<12:00,  6.10s/it]
Training Epoch: 12/100, batch 137/375 completed (loss: 0.0041:  37%|███▋      | 69/187 [06:55<12:00,  6.10s/it]
Training Epoch: 12/100, batch 137/375 completed (loss: 0.2145:  37%|███▋      | 69/187 [06:55<12:00,  6.10s/it]
Training Epoch: 12/100, batch 138/375 completed (loss: 0.0363:  37%|███▋      | 69/187 [06:58<12:00,  6.10s/it]
Training Epoch: 12/100, batch 138/375 completed (loss: 0.0546:  37%|███▋      | 69/187 [06:58<12:00,  6.10s/it]
Training Epoch: 12/100, batch 138/375 completed (loss: 0.3665:  37%|███▋      | 69/187 [06:58<12:00,  6.10s/it]
Training Epoch: 12/100, batch 138/375 completed (loss: 0.0146:  37%|███▋      | 69/187 [06:58<12:00,  6.10s/it]
Training Epoch: 12/100, batch 138/375 completed (loss: 0.0146:  37%|███▋      | 70/187 [07:01<11:45,  6.03s/it]
Training Epoch: 12/100, batch 138/375 completed (loss: 0.3665:  37%|███▋      | 70/187 [07:01<11:45,  6.03s/it]
Training Epoch: 12/100, batch 138/375 completed (loss: 0.0546:  37%|███▋      | 70/187 [07:01<11:45,  6.03s/it]
Training Epoch: 12/100, batch 138/375 completed (loss: 0.0363:  37%|███▋      | 70/187 [07:01<11:45,  6.03s/it]
Training Epoch: 12/100, batch 139/375 completed (loss: 0.1192:  37%|███▋      | 70/187 [07:01<11:45,  6.03s/it]
Training Epoch: 12/100, batch 139/375 completed (loss: 0.1338:  37%|███▋      | 70/187 [07:01<11:45,  6.03s/it]
Training Epoch: 12/100, batch 139/375 completed (loss: 0.0853:  37%|███▋      | 70/187 [07:01<11:45,  6.03s/it]
Training Epoch: 12/100, batch 139/375 completed (loss: 0.0237:  37%|███▋      | 70/187 [07:01<11:45,  6.03s/it]
Training Epoch: 12/100, batch 140/375 completed (loss: 0.0385:  37%|███▋      | 70/187 [07:04<11:45,  6.03s/it]
Training Epoch: 12/100, batch 140/375 completed (loss: 0.0869:  37%|███▋      | 70/187 [07:04<11:45,  6.03s/it]
Training Epoch: 12/100, batch 140/375 completed (loss: 0.1478:  37%|███▋      | 70/187 [07:04<11:45,  6.03s/it]
Training Epoch: 12/100, batch 140/375 completed (loss: 0.1054:  37%|███▋      | 70/187 [07:04<11:45,  6.03s/it]
Training Epoch: 12/100, batch 140/375 completed (loss: 0.1054:  38%|███▊      | 71/187 [07:07<11:49,  6.12s/it]
Training Epoch: 12/100, batch 140/375 completed (loss: 0.0869:  38%|███▊      | 71/187 [07:07<11:49,  6.11s/it]
Training Epoch: 12/100, batch 140/375 completed (loss: 0.0385:  38%|███▊      | 71/187 [07:07<11:49,  6.12s/it]
Training Epoch: 12/100, batch 140/375 completed (loss: 0.1478:  38%|███▊      | 71/187 [07:07<11:49,  6.12s/it]
Training Epoch: 12/100, batch 141/375 completed (loss: 0.1513:  38%|███▊      | 71/187 [07:08<11:49,  6.12s/it]
Training Epoch: 12/100, batch 141/375 completed (loss: 0.1013:  38%|███▊      | 71/187 [07:08<11:49,  6.12s/it]
Training Epoch: 12/100, batch 141/375 completed (loss: 0.2485:  38%|███▊      | 71/187 [07:08<11:49,  6.11s/it]
Training Epoch: 12/100, batch 141/375 completed (loss: 0.0105:  38%|███▊      | 71/187 [07:08<11:49,  6.12s/it]
Training Epoch: 12/100, batch 142/375 completed (loss: 0.1234:  38%|███▊      | 71/187 [07:11<11:49,  6.11s/it]
Training Epoch: 12/100, batch 142/375 completed (loss: 0.0134:  38%|███▊      | 71/187 [07:11<11:49,  6.12s/it]
Training Epoch: 12/100, batch 142/375 completed (loss: 0.0662:  38%|███▊      | 71/187 [07:11<11:49,  6.12s/it]
Training Epoch: 12/100, batch 142/375 completed (loss: 0.0681:  38%|███▊      | 71/187 [07:11<11:49,  6.12s/it]
Training Epoch: 12/100, batch 142/375 completed (loss: 0.0681:  39%|███▊      | 72/187 [07:13<11:39,  6.08s/it]
Training Epoch: 12/100, batch 142/375 completed (loss: 0.1234:  39%|███▊      | 72/187 [07:13<11:39,  6.08s/it]
Training Epoch: 12/100, batch 142/375 completed (loss: 0.0134:  39%|███▊      | 72/187 [07:13<11:39,  6.08s/it]
Training Epoch: 12/100, batch 142/375 completed (loss: 0.0662:  39%|███▊      | 72/187 [07:13<11:39,  6.08s/it]
Training Epoch: 12/100, batch 143/375 completed (loss: 0.1202:  39%|███▊      | 72/187 [07:14<11:39,  6.08s/it]
Training Epoch: 12/100, batch 143/375 completed (loss: 0.0405:  39%|███▊      | 72/187 [07:14<11:39,  6.08s/it]
Training Epoch: 12/100, batch 143/375 completed (loss: 0.0409:  39%|███▊      | 72/187 [07:14<11:39,  6.08s/it]
Training Epoch: 12/100, batch 143/375 completed (loss: 0.0964:  39%|███▊      | 72/187 [07:14<11:39,  6.08s/it]
Training Epoch: 12/100, batch 144/375 completed (loss: 0.1641:  39%|███▊      | 72/187 [07:16<11:39,  6.08s/it]
Training Epoch: 12/100, batch 144/375 completed (loss: 0.0566:  39%|███▊      | 72/187 [07:16<11:39,  6.08s/it]
Training Epoch: 12/100, batch 144/375 completed (loss: 0.0599:  39%|███▊      | 72/187 [07:16<11:39,  6.08s/it]
Training Epoch: 12/100, batch 144/375 completed (loss: 0.0458:  39%|███▊      | 72/187 [07:16<11:39,  6.08s/it]
Training Epoch: 12/100, batch 144/375 completed (loss: 0.0458:  39%|███▉      | 73/187 [07:19<11:26,  6.02s/it]
Training Epoch: 12/100, batch 144/375 completed (loss: 0.0599:  39%|███▉      | 73/187 [07:19<11:26,  6.02s/it]
Training Epoch: 12/100, batch 144/375 completed (loss: 0.1641:  39%|███▉      | 73/187 [07:19<11:26,  6.02s/it]
Training Epoch: 12/100, batch 144/375 completed (loss: 0.0566:  39%|███▉      | 73/187 [07:19<11:26,  6.02s/it]
Training Epoch: 12/100, batch 145/375 completed (loss: 0.0579:  39%|███▉      | 73/187 [07:19<11:26,  6.02s/it]
Training Epoch: 12/100, batch 145/375 completed (loss: 0.0668:  39%|███▉      | 73/187 [07:19<11:26,  6.02s/it]
Training Epoch: 12/100, batch 145/375 completed (loss: 0.0305:  39%|███▉      | 73/187 [07:19<11:26,  6.02s/it]
Training Epoch: 12/100, batch 145/375 completed (loss: 0.0195:  39%|███▉      | 73/187 [07:19<11:26,  6.02s/it]
Training Epoch: 12/100, batch 146/375 completed (loss: 0.0355:  39%|███▉      | 73/187 [07:22<11:26,  6.02s/it]
Training Epoch: 12/100, batch 146/375 completed (loss: 0.0913:  39%|███▉      | 73/187 [07:22<11:26,  6.02s/it]
Training Epoch: 12/100, batch 146/375 completed (loss: 0.0205:  39%|███▉      | 73/187 [07:22<11:26,  6.02s/it]
Training Epoch: 12/100, batch 146/375 completed (loss: 0.0129:  39%|███▉      | 73/187 [07:23<11:26,  6.02s/it]
Training Epoch: 12/100, batch 146/375 completed (loss: 0.0205:  40%|███▉      | 74/187 [07:26<11:34,  6.14s/it]
Training Epoch: 12/100, batch 146/375 completed (loss: 0.0355:  40%|███▉      | 74/187 [07:26<11:34,  6.14s/it]
Training Epoch: 12/100, batch 146/375 completed (loss: 0.0129:  40%|███▉      | 74/187 [07:26<11:34,  6.15s/it]
Training Epoch: 12/100, batch 146/375 completed (loss: 0.0913:  40%|███▉      | 74/187 [07:26<11:34,  6.15s/it]
Training Epoch: 12/100, batch 147/375 completed (loss: 0.096:  40%|███▉      | 74/187 [07:26<11:34,  6.15s/it] 
Training Epoch: 12/100, batch 147/375 completed (loss: 0.0389:  40%|███▉      | 74/187 [07:26<11:34,  6.14s/it]
Training Epoch: 12/100, batch 147/375 completed (loss: 0.086:  40%|███▉      | 74/187 [07:26<11:34,  6.14s/it] 
Training Epoch: 12/100, batch 147/375 completed (loss: 0.0964:  40%|███▉      | 74/187 [07:26<11:34,  6.15s/it]
Training Epoch: 12/100, batch 148/375 completed (loss: 0.0226:  40%|███▉      | 74/187 [07:29<11:34,  6.15s/it]
Training Epoch: 12/100, batch 148/375 completed (loss: 0.1957:  40%|███▉      | 74/187 [07:29<11:34,  6.14s/it]
Training Epoch: 12/100, batch 148/375 completed (loss: 0.0995:  40%|███▉      | 74/187 [07:29<11:34,  6.14s/it]
Training Epoch: 12/100, batch 148/375 completed (loss: 0.1346:  40%|███▉      | 74/187 [07:29<11:34,  6.15s/it]
Training Epoch: 12/100, batch 148/375 completed (loss: 0.1346:  40%|████      | 75/187 [07:32<11:18,  6.05s/it]
Training Epoch: 12/100, batch 148/375 completed (loss: 0.0995:  40%|████      | 75/187 [07:32<11:18,  6.06s/it]
Training Epoch: 12/100, batch 148/375 completed (loss: 0.0226:  40%|████      | 75/187 [07:32<11:18,  6.06s/it]
Training Epoch: 12/100, batch 148/375 completed (loss: 0.1957:  40%|████      | 75/187 [07:32<11:18,  6.06s/it]
Training Epoch: 12/100, batch 149/375 completed (loss: 0.005:  40%|████      | 75/187 [07:32<11:18,  6.06s/it] 
Training Epoch: 12/100, batch 149/375 completed (loss: 0.16:  40%|████      | 75/187 [07:32<11:18,  6.06s/it]  
Training Epoch: 12/100, batch 149/375 completed (loss: 0.4407:  40%|████      | 75/187 [07:32<11:18,  6.06s/it]
Training Epoch: 12/100, batch 149/375 completed (loss: 0.0821:  40%|████      | 75/187 [07:32<11:18,  6.05s/it]
Training Epoch: 12/100, batch 150/375 completed (loss: 0.1168:  40%|████      | 75/187 [07:35<11:18,  6.06s/it]
Training Epoch: 12/100, batch 150/375 completed (loss: 0.2088:  40%|████      | 75/187 [07:35<11:18,  6.06s/it]
Training Epoch: 12/100, batch 150/375 completed (loss: 0.0418:  40%|████      | 75/187 [07:35<11:18,  6.06s/it]
Training Epoch: 12/100, batch 150/375 completed (loss: 0.0922:  40%|████      | 75/187 [07:35<11:18,  6.05s/it]
Training Epoch: 12/100, batch 150/375 completed (loss: 0.0418:  41%|████      | 76/187 [07:38<11:10,  6.04s/it]
Training Epoch: 12/100, batch 150/375 completed (loss: 0.0922:  41%|████      | 76/187 [07:38<11:11,  6.05s/it]
Training Epoch: 12/100, batch 150/375 completed (loss: 0.1168:  41%|████      | 76/187 [07:38<11:11,  6.05s/it]
Training Epoch: 12/100, batch 150/375 completed (loss: 0.2088:  41%|████      | 76/187 [07:38<11:11,  6.05s/it]
Training Epoch: 12/100, batch 151/375 completed (loss: 0.4118:  41%|████      | 76/187 [07:38<11:11,  6.05s/it]
Training Epoch: 12/100, batch 151/375 completed (loss: 0.2197:  41%|████      | 76/187 [07:38<11:11,  6.05s/it]
Training Epoch: 12/100, batch 151/375 completed (loss: 0.0382:  41%|████      | 76/187 [07:38<11:10,  6.04s/it]
Training Epoch: 12/100, batch 151/375 completed (loss: 0.0376:  41%|████      | 76/187 [07:38<11:11,  6.05s/it]
Training Epoch: 12/100, batch 152/375 completed (loss: 0.0191:  41%|████      | 76/187 [07:41<11:10,  6.04s/it]
Training Epoch: 12/100, batch 152/375 completed (loss: 0.0758:  41%|████      | 76/187 [07:41<11:11,  6.05s/it]
Training Epoch: 12/100, batch 152/375 completed (loss: 0.0259:  41%|████      | 76/187 [07:41<11:11,  6.05s/it]
Training Epoch: 12/100, batch 152/375 completed (loss: 0.0949:  41%|████      | 76/187 [07:41<11:11,  6.05s/it]
Training Epoch: 12/100, batch 152/375 completed (loss: 0.0191:  41%|████      | 77/187 [07:44<11:13,  6.12s/it]
Training Epoch: 12/100, batch 152/375 completed (loss: 0.0949:  41%|████      | 77/187 [07:44<11:13,  6.12s/it]
Training Epoch: 12/100, batch 152/375 completed (loss: 0.0259:  41%|████      | 77/187 [07:44<11:13,  6.12s/it]
Training Epoch: 12/100, batch 152/375 completed (loss: 0.0758:  41%|████      | 77/187 [07:44<11:13,  6.12s/it]
Training Epoch: 12/100, batch 153/375 completed (loss: 0.072:  41%|████      | 77/187 [07:44<11:13,  6.12s/it] 
Training Epoch: 12/100, batch 153/375 completed (loss: 0.0951:  41%|████      | 77/187 [07:44<11:13,  6.12s/it]
Training Epoch: 12/100, batch 153/375 completed (loss: 0.0666:  41%|████      | 77/187 [07:44<11:13,  6.12s/it]
Training Epoch: 12/100, batch 153/375 completed (loss: 0.0551:  41%|████      | 77/187 [07:44<11:13,  6.12s/it]
Training Epoch: 12/100, batch 154/375 completed (loss: 0.0269:  41%|████      | 77/187 [07:47<11:13,  6.12s/it]
Training Epoch: 12/100, batch 154/375 completed (loss: 0.0165:  41%|████      | 77/187 [07:47<11:13,  6.12s/it]
Training Epoch: 12/100, batch 154/375 completed (loss: 0.1176:  41%|████      | 77/187 [07:47<11:13,  6.12s/it]
Training Epoch: 12/100, batch 154/375 completed (loss: 0.3061:  41%|████      | 77/187 [07:47<11:13,  6.12s/it]
Training Epoch: 12/100, batch 154/375 completed (loss: 0.3061:  42%|████▏     | 78/187 [07:50<10:57,  6.04s/it]
Training Epoch: 12/100, batch 154/375 completed (loss: 0.0165:  42%|████▏     | 78/187 [07:50<10:58,  6.04s/it]
Training Epoch: 12/100, batch 154/375 completed (loss: 0.1176:  42%|████▏     | 78/187 [07:50<10:58,  6.04s/it]
Training Epoch: 12/100, batch 154/375 completed (loss: 0.0269:  42%|████▏     | 78/187 [07:50<10:58,  6.04s/it]
Training Epoch: 12/100, batch 155/375 completed (loss: 0.1128:  42%|████▏     | 78/187 [07:50<10:58,  6.04s/it]
Training Epoch: 12/100, batch 155/375 completed (loss: 0.5017:  42%|████▏     | 78/187 [07:50<10:58,  6.04s/it]
Training Epoch: 12/100, batch 155/375 completed (loss: 0.043:  42%|████▏     | 78/187 [07:50<10:58,  6.04s/it] 
Training Epoch: 12/100, batch 155/375 completed (loss: 0.0201:  42%|████▏     | 78/187 [07:50<10:57,  6.04s/it]
Training Epoch: 12/100, batch 156/375 completed (loss: 0.0186:  42%|████▏     | 78/187 [07:53<10:58,  6.04s/it]
Training Epoch: 12/100, batch 156/375 completed (loss: 0.1084:  42%|████▏     | 78/187 [07:53<10:58,  6.04s/it]
Training Epoch: 12/100, batch 156/375 completed (loss: 0.0115:  42%|████▏     | 78/187 [07:53<10:58,  6.04s/it]
Training Epoch: 12/100, batch 156/375 completed (loss: 0.3602:  42%|████▏     | 78/187 [07:53<10:57,  6.04s/it]
Training Epoch: 12/100, batch 156/375 completed (loss: 0.0115:  42%|████▏     | 79/187 [07:56<10:55,  6.07s/it]
Training Epoch: 12/100, batch 156/375 completed (loss: 0.1084:  42%|████▏     | 79/187 [07:56<10:56,  6.07s/it]
Training Epoch: 12/100, batch 156/375 completed (loss: 0.3602:  42%|████▏     | 79/187 [07:56<10:56,  6.08s/it]
Training Epoch: 12/100, batch 156/375 completed (loss: 0.0186:  42%|████▏     | 79/187 [07:56<10:55,  6.07s/it]
Training Epoch: 12/100, batch 157/375 completed (loss: 0.0976:  42%|████▏     | 79/187 [07:56<10:55,  6.07s/it]
Training Epoch: 12/100, batch 157/375 completed (loss: 0.0271:  42%|████▏     | 79/187 [07:56<10:56,  6.07s/it]
Training Epoch: 12/100, batch 157/375 completed (loss: 0.0079:  42%|████▏     | 79/187 [07:56<10:55,  6.07s/it]
Training Epoch: 12/100, batch 157/375 completed (loss: 0.2238:  42%|████▏     | 79/187 [07:56<10:56,  6.08s/it]
Training Epoch: 12/100, batch 158/375 completed (loss: 0.3282:  42%|████▏     | 79/187 [07:59<10:56,  6.07s/it]
Training Epoch: 12/100, batch 158/375 completed (loss: 0.5263:  42%|████▏     | 79/187 [07:59<10:55,  6.07s/it]
Training Epoch: 12/100, batch 158/375 completed (loss: 0.0227:  42%|████▏     | 79/187 [07:59<10:55,  6.07s/it]
Training Epoch: 12/100, batch 158/375 completed (loss: 0.06:  42%|████▏     | 79/187 [07:59<10:56,  6.08s/it]  
Training Epoch: 12/100, batch 158/375 completed (loss: 0.06:  43%|████▎     | 80/187 [08:02<10:47,  6.05s/it]
Training Epoch: 12/100, batch 158/375 completed (loss: 0.3282:  43%|████▎     | 80/187 [08:02<10:47,  6.05s/it]
Training Epoch: 12/100, batch 158/375 completed (loss: 0.5263:  43%|████▎     | 80/187 [08:02<10:47,  6.05s/it]
Training Epoch: 12/100, batch 158/375 completed (loss: 0.0227:  43%|████▎     | 80/187 [08:02<10:47,  6.05s/it]
Training Epoch: 12/100, batch 159/375 completed (loss: 0.071:  43%|████▎     | 80/187 [08:02<10:47,  6.05s/it] 
Training Epoch: 12/100, batch 159/375 completed (loss: 0.1746:  43%|████▎     | 80/187 [08:02<10:47,  6.05s/it]
Training Epoch: 12/100, batch 159/375 completed (loss: 0.0604:  43%|████▎     | 80/187 [08:02<10:47,  6.05s/it]
Training Epoch: 12/100, batch 159/375 completed (loss: 0.0248:  43%|████▎     | 80/187 [08:02<10:47,  6.05s/it]
Training Epoch: 12/100, batch 160/375 completed (loss: 0.4095:  43%|████▎     | 80/187 [08:05<10:47,  6.05s/it]
Training Epoch: 12/100, batch 160/375 completed (loss: 0.0405:  43%|████▎     | 80/187 [08:05<10:47,  6.05s/it]
Training Epoch: 12/100, batch 160/375 completed (loss: 0.0707:  43%|████▎     | 80/187 [08:05<10:47,  6.05s/it]
Training Epoch: 12/100, batch 160/375 completed (loss: 0.0809:  43%|████▎     | 80/187 [08:05<10:47,  6.05s/it]
Training Epoch: 12/100, batch 160/375 completed (loss: 0.0809:  43%|████▎     | 81/187 [08:08<10:34,  5.98s/it]
Training Epoch: 12/100, batch 160/375 completed (loss: 0.4095:  43%|████▎     | 81/187 [08:08<10:34,  5.98s/it]
Training Epoch: 12/100, batch 160/375 completed (loss: 0.0405:  43%|████▎     | 81/187 [08:08<10:34,  5.99s/it]
Training Epoch: 12/100, batch 160/375 completed (loss: 0.0707:  43%|████▎     | 81/187 [08:08<10:34,  5.99s/it]
Training Epoch: 12/100, batch 161/375 completed (loss: 0.0642:  43%|████▎     | 81/187 [08:08<10:34,  5.99s/it]
Training Epoch: 12/100, batch 161/375 completed (loss: 0.2237:  43%|████▎     | 81/187 [08:08<10:34,  5.98s/it]
Training Epoch: 12/100, batch 161/375 completed (loss: 0.0838:  43%|████▎     | 81/187 [08:08<10:34,  5.99s/it]
Training Epoch: 12/100, batch 161/375 completed (loss: 0.0142:  43%|████▎     | 81/187 [08:08<10:34,  5.98s/it]
Training Epoch: 12/100, batch 162/375 completed (loss: 0.1248:  43%|████▎     | 81/187 [08:11<10:34,  5.99s/it]
Training Epoch: 12/100, batch 162/375 completed (loss: 0.0934:  43%|████▎     | 81/187 [08:11<10:34,  5.98s/it]
Training Epoch: 12/100, batch 162/375 completed (loss: 0.2648:  43%|████▎     | 81/187 [08:11<10:34,  5.99s/it]
Training Epoch: 12/100, batch 162/375 completed (loss: 0.2057:  43%|████▎     | 81/187 [08:11<10:34,  5.98s/it]
Training Epoch: 12/100, batch 162/375 completed (loss: 0.1248:  44%|████▍     | 82/187 [08:14<10:39,  6.09s/it]
Training Epoch: 12/100, batch 162/375 completed (loss: 0.2648:  44%|████▍     | 82/187 [08:14<10:39,  6.09s/it]
Training Epoch: 12/100, batch 162/375 completed (loss: 0.2057:  44%|████▍     | 82/187 [08:14<10:39,  6.09s/it]
Training Epoch: 12/100, batch 162/375 completed (loss: 0.0934:  44%|████▍     | 82/187 [08:14<10:39,  6.09s/it]
Training Epoch: 12/100, batch 163/375 completed (loss: 0.0624:  44%|████▍     | 82/187 [08:14<10:39,  6.09s/it]
Training Epoch: 12/100, batch 163/375 completed (loss: 0.0485:  44%|████▍     | 82/187 [08:14<10:39,  6.09s/it]
Training Epoch: 12/100, batch 163/375 completed (loss: 0.0338:  44%|████▍     | 82/187 [08:14<10:39,  6.09s/it]
Training Epoch: 12/100, batch 163/375 completed (loss: 0.4179:  44%|████▍     | 82/187 [08:14<10:39,  6.09s/it]
Training Epoch: 12/100, batch 164/375 completed (loss: 0.0581:  44%|████▍     | 82/187 [08:17<10:39,  6.09s/it]
Training Epoch: 12/100, batch 164/375 completed (loss: 0.023:  44%|████▍     | 82/187 [08:17<10:39,  6.09s/it] 
Training Epoch: 12/100, batch 164/375 completed (loss: 0.0635:  44%|████▍     | 82/187 [08:17<10:39,  6.09s/it]
Training Epoch: 12/100, batch 164/375 completed (loss: 0.1578:  44%|████▍     | 82/187 [08:17<10:39,  6.09s/it]
Training Epoch: 12/100, batch 164/375 completed (loss: 0.0581:  44%|████▍     | 83/187 [08:20<10:25,  6.02s/it]
Training Epoch: 12/100, batch 164/375 completed (loss: 0.0635:  44%|████▍     | 83/187 [08:20<10:25,  6.02s/it]
Training Epoch: 12/100, batch 164/375 completed (loss: 0.1578:  44%|████▍     | 83/187 [08:20<10:25,  6.02s/it]
Training Epoch: 12/100, batch 164/375 completed (loss: 0.023:  44%|████▍     | 83/187 [08:20<10:25,  6.02s/it]
Training Epoch: 12/100, batch 165/375 completed (loss: 0.0609:  44%|████▍     | 83/187 [08:20<10:25,  6.02s/it]
Training Epoch: 12/100, batch 165/375 completed (loss: 0.0732:  44%|████▍     | 83/187 [08:20<10:25,  6.02s/it]
Training Epoch: 12/100, batch 165/375 completed (loss: 0.1001:  44%|████▍     | 83/187 [08:20<10:25,  6.02s/it]
Training Epoch: 12/100, batch 165/375 completed (loss: 0.0612:  44%|████▍     | 83/187 [08:20<10:25,  6.02s/it]
Training Epoch: 12/100, batch 166/375 completed (loss: 0.0285:  44%|████▍     | 83/187 [08:23<10:25,  6.02s/it]
Training Epoch: 12/100, batch 166/375 completed (loss: 0.1315:  44%|████▍     | 83/187 [08:23<10:25,  6.02s/it]
Training Epoch: 12/100, batch 166/375 completed (loss: 0.0553:  44%|████▍     | 83/187 [08:23<10:25,  6.02s/it]
Training Epoch: 12/100, batch 166/375 completed (loss: 0.0265:  44%|████▍     | 83/187 [08:23<10:25,  6.02s/it]
Training Epoch: 12/100, batch 166/375 completed (loss: 0.0553:  45%|████▍     | 84/187 [08:26<10:18,  6.01s/it]
Training Epoch: 12/100, batch 166/375 completed (loss: 0.1315:  45%|████▍     | 84/187 [08:26<10:18,  6.01s/it]
Training Epoch: 12/100, batch 166/375 completed (loss: 0.0285:  45%|████▍     | 84/187 [08:26<10:18,  6.01s/it]
Training Epoch: 12/100, batch 166/375 completed (loss: 0.0265:  45%|████▍     | 84/187 [08:26<10:19,  6.01s/it]
Training Epoch: 12/100, batch 167/375 completed (loss: 0.3178:  45%|████▍     | 84/187 [08:26<10:18,  6.01s/it]
Training Epoch: 12/100, batch 167/375 completed (loss: 0.0783:  45%|████▍     | 84/187 [08:26<10:18,  6.01s/it]
Training Epoch: 12/100, batch 167/375 completed (loss: 0.17:  45%|████▍     | 84/187 [08:26<10:18,  6.01s/it]  
Training Epoch: 12/100, batch 167/375 completed (loss: 0.0925:  45%|████▍     | 84/187 [08:26<10:19,  6.01s/it]
Training Epoch: 12/100, batch 168/375 completed (loss: 0.0415:  45%|████▍     | 84/187 [08:29<10:18,  6.01s/it]
Training Epoch: 12/100, batch 168/375 completed (loss: 0.0053:  45%|████▍     | 84/187 [08:29<10:18,  6.01s/it]
Training Epoch: 12/100, batch 168/375 completed (loss: 0.0781:  45%|████▍     | 84/187 [08:29<10:18,  6.01s/it]
Training Epoch: 12/100, batch 168/375 completed (loss: 0.0711:  45%|████▍     | 84/187 [08:29<10:19,  6.01s/it]
Training Epoch: 12/100, batch 168/375 completed (loss: 0.0711:  45%|████▌     | 85/187 [08:32<10:21,  6.10s/it]
Training Epoch: 12/100, batch 168/375 completed (loss: 0.0781:  45%|████▌     | 85/187 [08:32<10:22,  6.10s/it]
Training Epoch: 12/100, batch 168/375 completed (loss: 0.0415:  45%|████▌     | 85/187 [08:32<10:22,  6.10s/it]
Training Epoch: 12/100, batch 168/375 completed (loss: 0.0053:  45%|████▌     | 85/187 [08:32<10:22,  6.10s/it]
Training Epoch: 12/100, batch 169/375 completed (loss: 0.0956:  45%|████▌     | 85/187 [08:32<10:22,  6.10s/it]
Training Epoch: 12/100, batch 169/375 completed (loss: 0.0732:  45%|████▌     | 85/187 [08:32<10:22,  6.10s/it]
Training Epoch: 12/100, batch 169/375 completed (loss: 0.0343:  45%|████▌     | 85/187 [08:32<10:22,  6.10s/it]
Training Epoch: 12/100, batch 169/375 completed (loss: 0.1163:  45%|████▌     | 85/187 [08:32<10:21,  6.10s/it]
Training Epoch: 12/100, batch 170/375 completed (loss: 0.2698:  45%|████▌     | 85/187 [08:35<10:22,  6.10s/it]
Training Epoch: 12/100, batch 170/375 completed (loss: 0.1201:  45%|████▌     | 85/187 [08:35<10:22,  6.10s/it]
Training Epoch: 12/100, batch 170/375 completed (loss: 0.1081:  45%|████▌     | 85/187 [08:35<10:22,  6.10s/it]
Training Epoch: 12/100, batch 170/375 completed (loss: 0.2275:  45%|████▌     | 85/187 [08:35<10:21,  6.10s/it]
Training Epoch: 12/100, batch 170/375 completed (loss: 0.2275:  46%|████▌     | 86/187 [08:38<10:07,  6.01s/it]
Training Epoch: 12/100, batch 170/375 completed (loss: 0.1201:  46%|████▌     | 86/187 [08:38<10:07,  6.01s/it]
Training Epoch: 12/100, batch 170/375 completed (loss: 0.2698:  46%|████▌     | 86/187 [08:38<10:07,  6.01s/it]
Training Epoch: 12/100, batch 170/375 completed (loss: 0.1081:  46%|████▌     | 86/187 [08:38<10:07,  6.02s/it]
Training Epoch: 12/100, batch 171/375 completed (loss: 0.0107:  46%|████▌     | 86/187 [08:38<10:07,  6.01s/it]
Training Epoch: 12/100, batch 171/375 completed (loss: 0.0322:  46%|████▌     | 86/187 [08:38<10:07,  6.01s/it]
Training Epoch: 12/100, batch 171/375 completed (loss: 0.0167:  46%|████▌     | 86/187 [08:38<10:07,  6.02s/it]
Training Epoch: 12/100, batch 171/375 completed (loss: 0.0551:  46%|████▌     | 86/187 [08:38<10:07,  6.01s/it]
Training Epoch: 12/100, batch 172/375 completed (loss: 0.0296:  46%|████▌     | 86/187 [08:41<10:07,  6.01s/it]
Training Epoch: 12/100, batch 172/375 completed (loss: 0.158:  46%|████▌     | 86/187 [08:41<10:07,  6.02s/it] 
Training Epoch: 12/100, batch 172/375 completed (loss: 0.048:  46%|████▌     | 86/187 [08:41<10:07,  6.01s/it] 
Training Epoch: 12/100, batch 172/375 completed (loss: 0.0602:  46%|████▌     | 86/187 [08:41<10:07,  6.01s/it]
Training Epoch: 12/100, batch 172/375 completed (loss: 0.0602:  47%|████▋     | 87/187 [08:45<10:15,  6.15s/it]
Training Epoch: 12/100, batch 172/375 completed (loss: 0.048:  47%|████▋     | 87/187 [08:45<10:15,  6.15s/it]
Training Epoch: 12/100, batch 172/375 completed (loss: 0.0296:  47%|████▋     | 87/187 [08:45<10:15,  6.15s/it]
Training Epoch: 12/100, batch 172/375 completed (loss: 0.158:  47%|████▋     | 87/187 [08:45<10:15,  6.16s/it]
Training Epoch: 12/100, batch 173/375 completed (loss: 0.0184:  47%|████▋     | 87/187 [08:45<10:15,  6.15s/it]
Training Epoch: 12/100, batch 173/375 completed (loss: 0.046:  47%|████▋     | 87/187 [08:45<10:15,  6.15s/it]
Training Epoch: 12/100, batch 173/375 completed (loss: 0.0381:  47%|████▋     | 87/187 [08:45<10:15,  6.16s/it]
Training Epoch: 12/100, batch 173/375 completed (loss: 0.0472:  47%|████▋     | 87/187 [08:45<10:15,  6.15s/it]
Training Epoch: 12/100, batch 174/375 completed (loss: 0.0091:  47%|████▋     | 87/187 [08:48<10:15,  6.15s/it]
Training Epoch: 12/100, batch 174/375 completed (loss: 0.0275:  47%|████▋     | 87/187 [08:48<10:15,  6.16s/it]
Training Epoch: 12/100, batch 174/375 completed (loss: 0.3362:  47%|████▋     | 87/187 [08:48<10:15,  6.15s/it]
Training Epoch: 12/100, batch 174/375 completed (loss: 0.012:  47%|████▋     | 87/187 [08:48<10:15,  6.15s/it] 
Training Epoch: 12/100, batch 174/375 completed (loss: 0.012:  47%|████▋     | 88/187 [08:50<10:00,  6.07s/it]
Training Epoch: 12/100, batch 174/375 completed (loss: 0.0275:  47%|████▋     | 88/187 [08:50<10:00,  6.07s/it]
Training Epoch: 12/100, batch 174/375 completed (loss: 0.3362:  47%|████▋     | 88/187 [08:50<10:00,  6.07s/it]
Training Epoch: 12/100, batch 174/375 completed (loss: 0.0091:  47%|████▋     | 88/187 [08:50<10:00,  6.07s/it]
Training Epoch: 12/100, batch 175/375 completed (loss: 0.0964:  47%|████▋     | 88/187 [08:50<10:00,  6.07s/it]
Training Epoch: 12/100, batch 175/375 completed (loss: 0.0745:  47%|████▋     | 88/187 [08:50<10:00,  6.07s/it]
Training Epoch: 12/100, batch 175/375 completed (loss: 0.4772:  47%|████▋     | 88/187 [08:50<10:00,  6.07s/it]
Training Epoch: 12/100, batch 175/375 completed (loss: 0.039:  47%|████▋     | 88/187 [08:50<10:00,  6.07s/it]
Training Epoch: 12/100, batch 176/375 completed (loss: 0.1834:  47%|████▋     | 88/187 [08:53<10:00,  6.07s/it]
Training Epoch: 12/100, batch 176/375 completed (loss: 0.0337:  47%|████▋     | 88/187 [08:53<10:00,  6.07s/it]
Training Epoch: 12/100, batch 176/375 completed (loss: 0.071:  47%|████▋     | 88/187 [08:53<10:00,  6.07s/it] 
Training Epoch: 12/100, batch 176/375 completed (loss: 0.2518:  47%|████▋     | 88/187 [08:53<10:00,  6.07s/it]
Training Epoch: 12/100, batch 176/375 completed (loss: 0.2518:  48%|████▊     | 89/187 [08:56<09:48,  6.00s/it]
Training Epoch: 12/100, batch 176/375 completed (loss: 0.0337:  48%|████▊     | 89/187 [08:56<09:48,  6.00s/it]
Training Epoch: 12/100, batch 176/375 completed (loss: 0.1834:  48%|████▊     | 89/187 [08:56<09:48,  6.00s/it]
Training Epoch: 12/100, batch 176/375 completed (loss: 0.071:  48%|████▊     | 89/187 [08:56<09:48,  6.00s/it]
Training Epoch: 12/100, batch 177/375 completed (loss: 0.0166:  48%|████▊     | 89/187 [08:56<09:48,  6.00s/it]
Training Epoch: 12/100, batch 177/375 completed (loss: 0.3797:  48%|████▊     | 89/187 [08:56<09:48,  6.00s/it]
Training Epoch: 12/100, batch 177/375 completed (loss: 0.14:  48%|████▊     | 89/187 [08:56<09:48,  6.00s/it]  
Training Epoch: 12/100, batch 177/375 completed (loss: 0.3795:  48%|████▊     | 89/187 [08:56<09:48,  6.00s/it]
Training Epoch: 12/100, batch 178/375 completed (loss: 0.0882:  48%|████▊     | 89/187 [09:00<09:48,  6.00s/it]
Training Epoch: 12/100, batch 178/375 completed (loss: 0.0897:  48%|████▊     | 89/187 [09:00<09:48,  6.00s/it]
Training Epoch: 12/100, batch 178/375 completed (loss: 0.2001:  48%|████▊     | 89/187 [09:00<09:48,  6.00s/it]
Training Epoch: 12/100, batch 178/375 completed (loss: 0.0704:  48%|████▊     | 89/187 [09:00<09:48,  6.00s/it]
Training Epoch: 12/100, batch 178/375 completed (loss: 0.0897:  48%|████▊     | 90/187 [09:03<09:52,  6.11s/it]
Training Epoch: 12/100, batch 178/375 completed (loss: 0.2001:  48%|████▊     | 90/187 [09:03<09:52,  6.11s/it]
Training Epoch: 12/100, batch 178/375 completed (loss: 0.0882:  48%|████▊     | 90/187 [09:03<09:52,  6.11s/it]
Training Epoch: 12/100, batch 178/375 completed (loss: 0.0704:  48%|████▊     | 90/187 [09:03<09:52,  6.11s/it]
Training Epoch: 12/100, batch 179/375 completed (loss: 0.3013:  48%|████▊     | 90/187 [09:03<09:52,  6.11s/it]
Training Epoch: 12/100, batch 179/375 completed (loss: 0.1019:  48%|████▊     | 90/187 [09:03<09:52,  6.11s/it]
Training Epoch: 12/100, batch 179/375 completed (loss: 0.1899:  48%|████▊     | 90/187 [09:03<09:52,  6.11s/it]
Training Epoch: 12/100, batch 179/375 completed (loss: 0.1342:  48%|████▊     | 90/187 [09:03<09:52,  6.11s/it]
Training Epoch: 12/100, batch 180/375 completed (loss: 0.0551:  48%|████▊     | 90/187 [09:06<09:52,  6.11s/it]
Training Epoch: 12/100, batch 180/375 completed (loss: 0.1736:  48%|████▊     | 90/187 [09:06<09:52,  6.11s/it]
Training Epoch: 12/100, batch 180/375 completed (loss: 0.0393:  48%|████▊     | 90/187 [09:06<09:52,  6.11s/it]
Training Epoch: 12/100, batch 180/375 completed (loss: 0.3205:  48%|████▊     | 90/187 [09:06<09:52,  6.11s/it]
Training Epoch: 12/100, batch 180/375 completed (loss: 0.3205:  49%|████▊     | 91/187 [09:08<09:39,  6.03s/it]
Training Epoch: 12/100, batch 180/375 completed (loss: 0.0551:  49%|████▊     | 91/187 [09:08<09:39,  6.03s/it]
Training Epoch: 12/100, batch 180/375 completed (loss: 0.1736:  49%|████▊     | 91/187 [09:08<09:39,  6.03s/it]
Training Epoch: 12/100, batch 180/375 completed (loss: 0.0393:  49%|████▊     | 91/187 [09:08<09:39,  6.03s/it]
Training Epoch: 12/100, batch 181/375 completed (loss: 0.1057:  49%|████▊     | 91/187 [09:09<09:39,  6.03s/it]
Training Epoch: 12/100, batch 181/375 completed (loss: 0.0886:  49%|████▊     | 91/187 [09:09<09:39,  6.03s/it]
Training Epoch: 12/100, batch 181/375 completed (loss: 0.071:  49%|████▊     | 91/187 [09:09<09:39,  6.03s/it] 
Training Epoch: 12/100, batch 181/375 completed (loss: 0.0555:  49%|████▊     | 91/187 [09:09<09:39,  6.03s/it]
Training Epoch: 12/100, batch 182/375 completed (loss: 0.0118:  49%|████▊     | 91/187 [09:11<09:39,  6.03s/it]
Training Epoch: 12/100, batch 182/375 completed (loss: 0.0287:  49%|████▊     | 91/187 [09:11<09:39,  6.03s/it]
Training Epoch: 12/100, batch 182/375 completed (loss: 0.0977:  49%|████▊     | 91/187 [09:11<09:39,  6.03s/it]
Training Epoch: 12/100, batch 182/375 completed (loss: 0.1867:  49%|████▊     | 91/187 [09:11<09:39,  6.03s/it]
Training Epoch: 12/100, batch 182/375 completed (loss: 0.1867:  49%|████▉     | 92/187 [09:15<09:35,  6.06s/it]
Training Epoch: 12/100, batch 182/375 completed (loss: 0.0977:  49%|████▉     | 92/187 [09:15<09:35,  6.06s/it]
Training Epoch: 12/100, batch 182/375 completed (loss: 0.0287:  49%|████▉     | 92/187 [09:15<09:35,  6.06s/it]
Training Epoch: 12/100, batch 182/375 completed (loss: 0.0118:  49%|████▉     | 92/187 [09:15<09:35,  6.06s/it]
Training Epoch: 12/100, batch 183/375 completed (loss: 0.0862:  49%|████▉     | 92/187 [09:15<09:35,  6.06s/it]
Training Epoch: 12/100, batch 183/375 completed (loss: 0.0176:  49%|████▉     | 92/187 [09:15<09:35,  6.06s/it]
Training Epoch: 12/100, batch 183/375 completed (loss: 0.015:  49%|████▉     | 92/187 [09:15<09:35,  6.06s/it] 
Training Epoch: 12/100, batch 183/375 completed (loss: 0.3432:  49%|████▉     | 92/187 [09:15<09:35,  6.06s/it]
Training Epoch: 12/100, batch 184/375 completed (loss: 0.0255:  49%|████▉     | 92/187 [09:18<09:35,  6.06s/it]
Training Epoch: 12/100, batch 184/375 completed (loss: 0.0159:  49%|████▉     | 92/187 [09:18<09:35,  6.06s/it]
Training Epoch: 12/100, batch 184/375 completed (loss: 0.3119:  49%|████▉     | 92/187 [09:18<09:35,  6.06s/it]
Training Epoch: 12/100, batch 184/375 completed (loss: 0.0158:  49%|████▉     | 92/187 [09:18<09:35,  6.06s/it]
Training Epoch: 12/100, batch 184/375 completed (loss: 0.0255:  50%|████▉     | 93/187 [09:21<09:32,  6.09s/it]
Training Epoch: 12/100, batch 184/375 completed (loss: 0.0159:  50%|████▉     | 93/187 [09:21<09:32,  6.09s/it]
Training Epoch: 12/100, batch 184/375 completed (loss: 0.3119:  50%|████▉     | 93/187 [09:21<09:32,  6.09s/it]
Training Epoch: 12/100, batch 184/375 completed (loss: 0.0158:  50%|████▉     | 93/187 [09:21<09:32,  6.09s/it]
Training Epoch: 12/100, batch 185/375 completed (loss: 0.0438:  50%|████▉     | 93/187 [09:21<09:32,  6.09s/it]
Training Epoch: 12/100, batch 185/375 completed (loss: 0.0119:  50%|████▉     | 93/187 [09:21<09:32,  6.09s/it]
Training Epoch: 12/100, batch 185/375 completed (loss: 0.1244:  50%|████▉     | 93/187 [09:21<09:32,  6.09s/it]
Training Epoch: 12/100, batch 185/375 completed (loss: 0.0068:  50%|████▉     | 93/187 [09:21<09:32,  6.09s/it]
Training Epoch: 12/100, batch 186/375 completed (loss: 0.0764:  50%|████▉     | 93/187 [09:24<09:32,  6.09s/it]
Training Epoch: 12/100, batch 186/375 completed (loss: 0.0083:  50%|████▉     | 93/187 [09:24<09:32,  6.09s/it]
Training Epoch: 12/100, batch 186/375 completed (loss: 0.0316:  50%|████▉     | 93/187 [09:24<09:32,  6.09s/it]
Training Epoch: 12/100, batch 186/375 completed (loss: 0.0243:  50%|████▉     | 93/187 [09:24<09:32,  6.09s/it]
Training Epoch: 12/100, batch 186/375 completed (loss: 0.0764:  50%|█████     | 94/187 [09:27<09:19,  6.02s/it]
Training Epoch: 12/100, batch 186/375 completed (loss: 0.0083:  50%|█████     | 94/187 [09:27<09:19,  6.02s/it]
Training Epoch: 12/100, batch 186/375 completed (loss: 0.0243:  50%|█████     | 94/187 [09:27<09:20,  6.02s/it]
Training Epoch: 12/100, batch 186/375 completed (loss: 0.0316:  50%|█████     | 94/187 [09:27<09:19,  6.02s/it]
Training Epoch: 12/100, batch 187/375 completed (loss: 0.4412:  50%|█████     | 94/187 [09:27<09:19,  6.02s/it]
Training Epoch: 12/100, batch 187/375 completed (loss: 0.0267:  50%|█████     | 94/187 [09:27<09:19,  6.02s/it]
Training Epoch: 12/100, batch 187/375 completed (loss: 0.0065:  50%|█████     | 94/187 [09:27<09:19,  6.02s/it]
Training Epoch: 12/100, batch 187/375 completed (loss: 0.0423:  50%|█████     | 94/187 [09:27<09:20,  6.02s/it]
Training Epoch: 12/100, batch 188/375 completed (loss: 0.1308:  50%|█████     | 94/187 [09:30<09:19,  6.02s/it]
Training Epoch: 12/100, batch 188/375 completed (loss: 0.023:  50%|█████     | 94/187 [09:30<09:19,  6.02s/it] 
Training Epoch: 12/100, batch 188/375 completed (loss: 0.0173:  50%|█████     | 94/187 [09:30<09:19,  6.02s/it]
Training Epoch: 12/100, batch 188/375 completed (loss: 0.0904:  50%|█████     | 94/187 [09:30<09:20,  6.02s/it]
Training Epoch: 12/100, batch 188/375 completed (loss: 0.0904:  51%|█████     | 95/187 [09:33<09:25,  6.15s/it]
Training Epoch: 12/100, batch 188/375 completed (loss: 0.1308:  51%|█████     | 95/187 [09:33<09:26,  6.16s/it]
Training Epoch: 12/100, batch 188/375 completed (loss: 0.023:  51%|█████     | 95/187 [09:33<09:26,  6.16s/it]
Training Epoch: 12/100, batch 188/375 completed (loss: 0.0173:  51%|█████     | 95/187 [09:33<09:26,  6.16s/it]
Training Epoch: 12/100, batch 189/375 completed (loss: 0.0352:  51%|█████     | 95/187 [09:33<09:26,  6.16s/it]
Training Epoch: 12/100, batch 189/375 completed (loss: 0.0365:  51%|█████     | 95/187 [09:33<09:26,  6.16s/it]
Training Epoch: 12/100, batch 189/375 completed (loss: 0.0889:  51%|█████     | 95/187 [09:33<09:26,  6.16s/it]
Training Epoch: 12/100, batch 189/375 completed (loss: 0.0222:  51%|█████     | 95/187 [09:33<09:25,  6.15s/it]
Training Epoch: 12/100, batch 190/375 completed (loss: 0.0355:  51%|█████     | 95/187 [09:36<09:26,  6.16s/it]
Training Epoch: 12/100, batch 190/375 completed (loss: 0.147:  51%|█████     | 95/187 [09:36<09:26,  6.16s/it] 
Training Epoch: 12/100, batch 190/375 completed (loss: 0.0198:  51%|█████     | 95/187 [09:36<09:26,  6.16s/it]
Training Epoch: 12/100, batch 190/375 completed (loss: 0.0312:  51%|█████     | 95/187 [09:36<09:25,  6.15s/it]
Training Epoch: 12/100, batch 190/375 completed (loss: 0.0312:  51%|█████▏    | 96/187 [09:39<09:12,  6.07s/it]
Training Epoch: 12/100, batch 190/375 completed (loss: 0.0198:  51%|█████▏    | 96/187 [09:39<09:11,  6.06s/it]
Training Epoch: 12/100, batch 190/375 completed (loss: 0.0355:  51%|█████▏    | 96/187 [09:39<09:11,  6.06s/it]
Training Epoch: 12/100, batch 190/375 completed (loss: 0.147:  51%|█████▏    | 96/187 [09:39<09:11,  6.06s/it]
Training Epoch: 12/100, batch 191/375 completed (loss: 0.1654:  51%|█████▏    | 96/187 [09:39<09:11,  6.06s/it]
Training Epoch: 12/100, batch 191/375 completed (loss: 0.3282:  51%|█████▏    | 96/187 [09:39<09:11,  6.06s/it]
Training Epoch: 12/100, batch 191/375 completed (loss: 0.0917:  51%|█████▏    | 96/187 [09:39<09:11,  6.06s/it]
Training Epoch: 12/100, batch 191/375 completed (loss: 0.0802:  51%|█████▏    | 96/187 [09:39<09:12,  6.07s/it]
Training Epoch: 12/100, batch 192/375 completed (loss: 0.0298:  51%|█████▏    | 96/187 [09:42<09:11,  6.06s/it]
Training Epoch: 12/100, batch 192/375 completed (loss: 0.0823:  51%|█████▏    | 96/187 [09:42<09:11,  6.06s/it]
Training Epoch: 12/100, batch 192/375 completed (loss: 0.0946:  51%|█████▏    | 96/187 [09:42<09:11,  6.06s/it]
Training Epoch: 12/100, batch 192/375 completed (loss: 0.0278:  51%|█████▏    | 96/187 [09:42<09:12,  6.07s/it]
Training Epoch: 12/100, batch 192/375 completed (loss: 0.0278:  52%|█████▏    | 97/187 [09:45<09:04,  6.05s/it]
Training Epoch: 12/100, batch 192/375 completed (loss: 0.0823:  52%|█████▏    | 97/187 [09:45<09:04,  6.05s/it]
Training Epoch: 12/100, batch 192/375 completed (loss: 0.0946:  52%|█████▏    | 97/187 [09:45<09:05,  6.06s/it]
Training Epoch: 12/100, batch 192/375 completed (loss: 0.0298:  52%|█████▏    | 97/187 [09:45<09:05,  6.06s/it]
Training Epoch: 12/100, batch 193/375 completed (loss: 0.07:  52%|█████▏    | 97/187 [09:45<09:05,  6.06s/it]  
Training Epoch: 12/100, batch 193/375 completed (loss: 0.2701:  52%|█████▏    | 97/187 [09:45<09:04,  6.05s/it]
Training Epoch: 12/100, batch 193/375 completed (loss: 0.1834:  52%|█████▏    | 97/187 [09:45<09:05,  6.06s/it]
Training Epoch: 12/100, batch 193/375 completed (loss: 0.1174:  52%|█████▏    | 97/187 [09:45<09:04,  6.05s/it]
Training Epoch: 12/100, batch 194/375 completed (loss: 0.04:  52%|█████▏    | 97/187 [09:48<09:05,  6.06s/it]  
Training Epoch: 12/100, batch 194/375 completed (loss: 0.0226:  52%|█████▏    | 97/187 [09:48<09:05,  6.06s/it]
Training Epoch: 12/100, batch 194/375 completed (loss: 0.0245:  52%|█████▏    | 97/187 [09:48<09:04,  6.05s/it]
Training Epoch: 12/100, batch 194/375 completed (loss: 0.0248:  52%|█████▏    | 97/187 [09:48<09:04,  6.05s/it]
Training Epoch: 12/100, batch 194/375 completed (loss: 0.0245:  52%|█████▏    | 98/187 [09:51<09:05,  6.13s/it]
Training Epoch: 12/100, batch 194/375 completed (loss: 0.0226:  52%|█████▏    | 98/187 [09:51<09:05,  6.13s/it]
Training Epoch: 12/100, batch 194/375 completed (loss: 0.04:  52%|█████▏    | 98/187 [09:51<09:05,  6.13s/it]
Training Epoch: 12/100, batch 194/375 completed (loss: 0.0248:  52%|█████▏    | 98/187 [09:51<09:06,  6.14s/it]
Training Epoch: 12/100, batch 195/375 completed (loss: 0.1147:  52%|█████▏    | 98/187 [09:51<09:05,  6.13s/it]
Training Epoch: 12/100, batch 195/375 completed (loss: 0.0743:  52%|█████▏    | 98/187 [09:51<09:05,  6.13s/it]
Training Epoch: 12/100, batch 195/375 completed (loss: 0.0752:  52%|█████▏    | 98/187 [09:51<09:05,  6.13s/it]
Training Epoch: 12/100, batch 195/375 completed (loss: 0.4826:  52%|█████▏    | 98/187 [09:51<09:06,  6.14s/it]
Training Epoch: 12/100, batch 196/375 completed (loss: 0.0102:  52%|█████▏    | 98/187 [09:54<09:05,  6.13s/it]
Training Epoch: 12/100, batch 196/375 completed (loss: 0.0721:  52%|█████▏    | 98/187 [09:54<09:05,  6.13s/it]
Training Epoch: 12/100, batch 196/375 completed (loss: 0.1319:  52%|█████▏    | 98/187 [09:54<09:05,  6.13s/it]
Training Epoch: 12/100, batch 196/375 completed (loss: 0.1594:  52%|█████▏    | 98/187 [09:54<09:06,  6.14s/it]
Training Epoch: 12/100, batch 196/375 completed (loss: 0.1594:  53%|█████▎    | 99/187 [09:57<08:51,  6.04s/it]
Training Epoch: 12/100, batch 196/375 completed (loss: 0.1319:  53%|█████▎    | 99/187 [09:57<08:51,  6.04s/it]
Training Epoch: 12/100, batch 196/375 completed (loss: 0.0102:  53%|█████▎    | 99/187 [09:57<08:52,  6.05s/it]
Training Epoch: 12/100, batch 196/375 completed (loss: 0.0721:  53%|█████▎    | 99/187 [09:57<08:51,  6.05s/it]
Training Epoch: 12/100, batch 197/375 completed (loss: 0.2424:  53%|█████▎    | 99/187 [09:57<08:51,  6.05s/it]
Training Epoch: 12/100, batch 197/375 completed (loss: 0.0158:  53%|█████▎    | 99/187 [09:57<08:52,  6.05s/it]
Training Epoch: 12/100, batch 197/375 completed (loss: 0.1788:  53%|█████▎    | 99/187 [09:57<08:51,  6.04s/it]
Training Epoch: 12/100, batch 197/375 completed (loss: 0.2694:  53%|█████▎    | 99/187 [09:57<08:51,  6.04s/it]
Training Epoch: 12/100, batch 198/375 completed (loss: 0.1183:  53%|█████▎    | 99/187 [10:00<08:52,  6.05s/it]
Training Epoch: 12/100, batch 198/375 completed (loss: 0.247:  53%|█████▎    | 99/187 [10:00<08:51,  6.05s/it] 
Training Epoch: 12/100, batch 198/375 completed (loss: 0.2755:  53%|█████▎    | 99/187 [10:00<08:51,  6.04s/it]
Training Epoch: 12/100, batch 198/375 completed (loss: 0.0214:  53%|█████▎    | 99/187 [10:00<08:51,  6.04s/it]
Training Epoch: 12/100, batch 198/375 completed (loss: 0.2755:  53%|█████▎    | 100/187 [10:03<08:52,  6.12s/it]
Training Epoch: 12/100, batch 198/375 completed (loss: 0.247:  53%|█████▎    | 100/187 [10:03<08:52,  6.12s/it]
Training Epoch: 12/100, batch 198/375 completed (loss: 0.0214:  53%|█████▎    | 100/187 [10:03<08:52,  6.12s/it]
Training Epoch: 12/100, batch 198/375 completed (loss: 0.1183:  53%|█████▎    | 100/187 [10:03<08:52,  6.12s/it]
Training Epoch: 12/100, batch 199/375 completed (loss: 0.1236:  53%|█████▎    | 100/187 [10:03<08:52,  6.12s/it]
Training Epoch: 12/100, batch 199/375 completed (loss: 0.0292:  53%|█████▎    | 100/187 [10:03<08:52,  6.12s/it]
Training Epoch: 12/100, batch 199/375 completed (loss: 0.0628:  53%|█████▎    | 100/187 [10:03<08:52,  6.12s/it]
Training Epoch: 12/100, batch 199/375 completed (loss: 0.5091:  53%|█████▎    | 100/187 [10:03<08:52,  6.12s/it]
Training Epoch: 12/100, batch 200/375 completed (loss: 0.1387:  53%|█████▎    | 100/187 [10:07<08:52,  6.12s/it]
Training Epoch: 12/100, batch 200/375 completed (loss: 0.0093:  53%|█████▎    | 100/187 [10:07<08:52,  6.12s/it]
Training Epoch: 12/100, batch 200/375 completed (loss: 0.0649:  53%|█████▎    | 100/187 [10:07<08:52,  6.12s/it]
Training Epoch: 12/100, batch 200/375 completed (loss: 0.1528:  53%|█████▎    | 100/187 [10:07<08:52,  6.12s/it]
Training Epoch: 12/100, batch 200/375 completed (loss: 0.0649:  54%|█████▍    | 101/187 [10:09<08:43,  6.09s/it]
Training Epoch: 12/100, batch 200/375 completed (loss: 0.1528:  54%|█████▍    | 101/187 [10:09<08:44,  6.10s/it]
Training Epoch: 12/100, batch 200/375 completed (loss: 0.1387:  54%|█████▍    | 101/187 [10:09<08:44,  6.09s/it]
Training Epoch: 12/100, batch 200/375 completed (loss: 0.0093:  54%|█████▍    | 101/187 [10:09<08:44,  6.10s/it]
Training Epoch: 12/100, batch 201/375 completed (loss: 0.2113:  54%|█████▍    | 101/187 [10:09<08:44,  6.10s/it]
Training Epoch: 12/100, batch 201/375 completed (loss: 0.0363:  54%|█████▍    | 101/187 [10:09<08:44,  6.09s/it]
Training Epoch: 12/100, batch 201/375 completed (loss: 0.1708:  54%|█████▍    | 101/187 [10:09<08:43,  6.09s/it]
Training Epoch: 12/100, batch 201/375 completed (loss: 0.0359:  54%|█████▍    | 101/187 [10:10<08:44,  6.10s/it]
Training Epoch: 12/100, batch 202/375 completed (loss: 0.0305:  54%|█████▍    | 101/187 [10:12<08:44,  6.10s/it]
Training Epoch: 12/100, batch 202/375 completed (loss: 0.4058:  54%|█████▍    | 101/187 [10:12<08:43,  6.09s/it]
Training Epoch: 12/100, batch 202/375 completed (loss: 0.0225:  54%|█████▍    | 101/187 [10:12<08:44,  6.09s/it]
Training Epoch: 12/100, batch 202/375 completed (loss: 0.0359:  54%|█████▍    | 101/187 [10:12<08:44,  6.10s/it]
Training Epoch: 12/100, batch 202/375 completed (loss: 0.0359:  55%|█████▍    | 102/187 [10:15<08:35,  6.06s/it]
Training Epoch: 12/100, batch 202/375 completed (loss: 0.4058:  55%|█████▍    | 102/187 [10:15<08:35,  6.06s/it]
Training Epoch: 12/100, batch 202/375 completed (loss: 0.0305:  55%|█████▍    | 102/187 [10:15<08:35,  6.06s/it]
Training Epoch: 12/100, batch 202/375 completed (loss: 0.0225:  55%|█████▍    | 102/187 [10:15<08:35,  6.06s/it]
Training Epoch: 12/100, batch 203/375 completed (loss: 0.2995:  55%|█████▍    | 102/187 [10:15<08:35,  6.06s/it]
Training Epoch: 12/100, batch 203/375 completed (loss: 0.4493:  55%|█████▍    | 102/187 [10:15<08:35,  6.06s/it]
Training Epoch: 12/100, batch 203/375 completed (loss: 0.1107:  55%|█████▍    | 102/187 [10:15<08:35,  6.06s/it]
Training Epoch: 12/100, batch 203/375 completed (loss: 0.1723:  55%|█████▍    | 102/187 [10:15<08:35,  6.06s/it]
Training Epoch: 12/100, batch 204/375 completed (loss: 0.1604:  55%|█████▍    | 102/187 [10:19<08:35,  6.06s/it]
Training Epoch: 12/100, batch 204/375 completed (loss: 0.1884:  55%|█████▍    | 102/187 [10:19<08:35,  6.06s/it]
Training Epoch: 12/100, batch 204/375 completed (loss: 0.0448:  55%|█████▍    | 102/187 [10:19<08:35,  6.06s/it]
Training Epoch: 12/100, batch 204/375 completed (loss: 0.1779:  55%|█████▍    | 102/187 [10:19<08:35,  6.06s/it]
Training Epoch: 12/100, batch 204/375 completed (loss: 0.0448:  55%|█████▌    | 103/187 [10:22<08:35,  6.14s/it]
Training Epoch: 12/100, batch 204/375 completed (loss: 0.1884:  55%|█████▌    | 103/187 [10:22<08:35,  6.14s/it]
Training Epoch: 12/100, batch 204/375 completed (loss: 0.1604:  55%|█████▌    | 103/187 [10:22<08:35,  6.14s/it]
Training Epoch: 12/100, batch 204/375 completed (loss: 0.1779:  55%|█████▌    | 103/187 [10:22<08:35,  6.14s/it]
Training Epoch: 12/100, batch 205/375 completed (loss: 0.0757:  55%|█████▌    | 103/187 [10:22<08:35,  6.14s/it]
Training Epoch: 12/100, batch 205/375 completed (loss: 0.0675:  55%|█████▌    | 103/187 [10:22<08:35,  6.14s/it]
Training Epoch: 12/100, batch 205/375 completed (loss: 0.0139:  55%|█████▌    | 103/187 [10:22<08:35,  6.14s/it]
Training Epoch: 12/100, batch 205/375 completed (loss: 0.0323:  55%|█████▌    | 103/187 [10:22<08:35,  6.14s/it]
Training Epoch: 12/100, batch 206/375 completed (loss: 0.0287:  55%|█████▌    | 103/187 [10:25<08:35,  6.14s/it]
Training Epoch: 12/100, batch 206/375 completed (loss: 0.0407:  55%|█████▌    | 103/187 [10:25<08:35,  6.14s/it]
Training Epoch: 12/100, batch 206/375 completed (loss: 0.139:  55%|█████▌    | 103/187 [10:25<08:35,  6.14s/it] 
Training Epoch: 12/100, batch 206/375 completed (loss: 0.0758:  55%|█████▌    | 103/187 [10:25<08:35,  6.14s/it]
Training Epoch: 12/100, batch 206/375 completed (loss: 0.0287:  56%|█████▌    | 104/187 [10:28<08:21,  6.04s/it]
Training Epoch: 12/100, batch 206/375 completed (loss: 0.139:  56%|█████▌    | 104/187 [10:28<08:21,  6.04s/it]
Training Epoch: 12/100, batch 206/375 completed (loss: 0.0758:  56%|█████▌    | 104/187 [10:28<08:21,  6.04s/it]
Training Epoch: 12/100, batch 206/375 completed (loss: 0.0407:  56%|█████▌    | 104/187 [10:28<08:21,  6.04s/it]
Training Epoch: 12/100, batch 207/375 completed (loss: 0.0788:  56%|█████▌    | 104/187 [10:28<08:21,  6.04s/it]
Training Epoch: 12/100, batch 207/375 completed (loss: 0.2359:  56%|█████▌    | 104/187 [10:28<08:21,  6.04s/it]
Training Epoch: 12/100, batch 207/375 completed (loss: 0.009:  56%|█████▌    | 104/187 [10:28<08:21,  6.04s/it] 
Training Epoch: 12/100, batch 207/375 completed (loss: 0.0543:  56%|█████▌    | 104/187 [10:28<08:21,  6.04s/it]
Training Epoch: 12/100, batch 208/375 completed (loss: 0.0739:  56%|█████▌    | 104/187 [10:31<08:21,  6.04s/it]
Training Epoch: 12/100, batch 208/375 completed (loss: 0.0244:  56%|█████▌    | 104/187 [10:31<08:21,  6.04s/it]
Training Epoch: 12/100, batch 208/375 completed (loss: 0.0265:  56%|█████▌    | 104/187 [10:31<08:21,  6.04s/it]
Training Epoch: 12/100, batch 208/375 completed (loss: 0.3022:  56%|█████▌    | 104/187 [10:31<08:21,  6.04s/it]
Training Epoch: 12/100, batch 208/375 completed (loss: 0.3022:  56%|█████▌    | 105/187 [10:34<08:17,  6.07s/it]
Training Epoch: 12/100, batch 208/375 completed (loss: 0.0265:  56%|█████▌    | 105/187 [10:34<08:17,  6.07s/it]
Training Epoch: 12/100, batch 208/375 completed (loss: 0.0739:  56%|█████▌    | 105/187 [10:34<08:17,  6.07s/it]
Training Epoch: 12/100, batch 208/375 completed (loss: 0.0244:  56%|█████▌    | 105/187 [10:34<08:18,  6.08s/it]
Training Epoch: 12/100, batch 209/375 completed (loss: 0.0482:  56%|█████▌    | 105/187 [10:34<08:18,  6.08s/it]
Training Epoch: 12/100, batch 209/375 completed (loss: 0.1625:  56%|█████▌    | 105/187 [10:34<08:17,  6.07s/it]
Training Epoch: 12/100, batch 209/375 completed (loss: 0.0698:  56%|█████▌    | 105/187 [10:34<08:17,  6.07s/it]
Training Epoch: 12/100, batch 209/375 completed (loss: 0.1038:  56%|█████▌    | 105/187 [10:34<08:17,  6.07s/it]
Training Epoch: 12/100, batch 210/375 completed (loss: 0.0154:  56%|█████▌    | 105/187 [10:37<08:17,  6.07s/it]
Training Epoch: 12/100, batch 210/375 completed (loss: 0.0746:  56%|█████▌    | 105/187 [10:37<08:18,  6.08s/it]
Training Epoch: 12/100, batch 210/375 completed (loss: 0.3302:  56%|█████▌    | 105/187 [10:37<08:17,  6.07s/it]
Training Epoch: 12/100, batch 210/375 completed (loss: 0.073:  56%|█████▌    | 105/187 [10:37<08:17,  6.07s/it] 
Training Epoch: 12/100, batch 210/375 completed (loss: 0.073:  57%|█████▋    | 106/187 [10:40<08:13,  6.09s/it]
Training Epoch: 12/100, batch 210/375 completed (loss: 0.0746:  57%|█████▋    | 106/187 [10:40<08:13,  6.09s/it]
Training Epoch: 12/100, batch 210/375 completed (loss: 0.3302:  57%|█████▋    | 106/187 [10:40<08:13,  6.09s/it]
Training Epoch: 12/100, batch 210/375 completed (loss: 0.0154:  57%|█████▋    | 106/187 [10:40<08:13,  6.09s/it]
Training Epoch: 12/100, batch 211/375 completed (loss: 0.0219:  57%|█████▋    | 106/187 [10:40<08:13,  6.09s/it]
Training Epoch: 12/100, batch 211/375 completed (loss: 0.6378:  57%|█████▋    | 106/187 [10:40<08:13,  6.09s/it]
Training Epoch: 12/100, batch 211/375 completed (loss: 0.0654:  57%|█████▋    | 106/187 [10:40<08:13,  6.09s/it]
Training Epoch: 12/100, batch 211/375 completed (loss: 0.5847:  57%|█████▋    | 106/187 [10:40<08:13,  6.09s/it]
Training Epoch: 12/100, batch 212/375 completed (loss: 0.0452:  57%|█████▋    | 106/187 [10:43<08:13,  6.09s/it]
Training Epoch: 12/100, batch 212/375 completed (loss: 0.2182:  57%|█████▋    | 106/187 [10:43<08:13,  6.09s/it]
Training Epoch: 12/100, batch 212/375 completed (loss: 0.0597:  57%|█████▋    | 106/187 [10:43<08:13,  6.09s/it]
Training Epoch: 12/100, batch 212/375 completed (loss: 0.0934:  57%|█████▋    | 106/187 [10:43<08:13,  6.09s/it]
Training Epoch: 12/100, batch 212/375 completed (loss: 0.2182:  57%|█████▋    | 107/187 [10:46<08:00,  6.01s/it]
Training Epoch: 12/100, batch 212/375 completed (loss: 0.0452:  57%|█████▋    | 107/187 [10:46<08:00,  6.01s/it]
Training Epoch: 12/100, batch 212/375 completed (loss: 0.0597:  57%|█████▋    | 107/187 [10:46<08:00,  6.01s/it]
Training Epoch: 12/100, batch 212/375 completed (loss: 0.0934:  57%|█████▋    | 107/187 [10:46<08:00,  6.01s/it]
Training Epoch: 12/100, batch 213/375 completed (loss: 0.0197:  57%|█████▋    | 107/187 [10:46<08:00,  6.01s/it]
Training Epoch: 12/100, batch 213/375 completed (loss: 0.0237:  57%|█████▋    | 107/187 [10:46<08:00,  6.01s/it]
Training Epoch: 12/100, batch 213/375 completed (loss: 0.0157:  57%|█████▋    | 107/187 [10:46<08:00,  6.01s/it]
Training Epoch: 12/100, batch 213/375 completed (loss: 0.2453:  57%|█████▋    | 107/187 [10:46<08:00,  6.01s/it]
Training Epoch: 12/100, batch 214/375 completed (loss: 0.028:  57%|█████▋    | 107/187 [10:49<08:00,  6.01s/it] 
Training Epoch: 12/100, batch 214/375 completed (loss: 0.2801:  57%|█████▋    | 107/187 [10:49<08:00,  6.01s/it]
Training Epoch: 12/100, batch 214/375 completed (loss: 0.1804:  57%|█████▋    | 107/187 [10:49<08:00,  6.01s/it]
Training Epoch: 12/100, batch 214/375 completed (loss: 0.0403:  57%|█████▋    | 107/187 [10:49<08:00,  6.01s/it]
Training Epoch: 12/100, batch 214/375 completed (loss: 0.0403:  58%|█████▊    | 108/187 [10:52<08:01,  6.09s/it]
Training Epoch: 12/100, batch 214/375 completed (loss: 0.2801:  58%|█████▊    | 108/187 [10:52<08:01,  6.09s/it]
Training Epoch: 12/100, batch 214/375 completed (loss: 0.1804:  58%|█████▊    | 108/187 [10:52<08:01,  6.09s/it]
Training Epoch: 12/100, batch 214/375 completed (loss: 0.028:  58%|█████▊    | 108/187 [10:52<08:01,  6.09s/it]
Training Epoch: 12/100, batch 215/375 completed (loss: 0.1512:  58%|█████▊    | 108/187 [10:52<08:01,  6.09s/it]
Training Epoch: 12/100, batch 215/375 completed (loss: 0.1145:  58%|█████▊    | 108/187 [10:52<08:01,  6.09s/it]
Training Epoch: 12/100, batch 215/375 completed (loss: 0.0305:  58%|█████▊    | 108/187 [10:52<08:01,  6.09s/it]
Training Epoch: 12/100, batch 215/375 completed (loss: 0.0577:  58%|█████▊    | 108/187 [10:52<08:01,  6.09s/it]
Training Epoch: 12/100, batch 216/375 completed (loss: 0.0692:  58%|█████▊    | 108/187 [10:55<08:01,  6.09s/it]
Training Epoch: 12/100, batch 216/375 completed (loss: 0.0831:  58%|█████▊    | 108/187 [10:55<08:01,  6.09s/it]
Training Epoch: 12/100, batch 216/375 completed (loss: 0.1644:  58%|█████▊    | 108/187 [10:55<08:01,  6.09s/it]
Training Epoch: 12/100, batch 216/375 completed (loss: 0.0215:  58%|█████▊    | 108/187 [10:55<08:01,  6.09s/it]
Training Epoch: 12/100, batch 216/375 completed (loss: 0.1644:  58%|█████▊    | 109/187 [10:58<07:53,  6.07s/it]
Training Epoch: 12/100, batch 216/375 completed (loss: 0.0692:  58%|█████▊    | 109/187 [10:58<07:53,  6.07s/it]
Training Epoch: 12/100, batch 216/375 completed (loss: 0.0831:  58%|█████▊    | 109/187 [10:58<07:53,  6.07s/it]
Training Epoch: 12/100, batch 216/375 completed (loss: 0.0215:  58%|█████▊    | 109/187 [10:58<07:54,  6.08s/it]
Training Epoch: 12/100, batch 217/375 completed (loss: 0.1748:  58%|█████▊    | 109/187 [10:58<07:53,  6.07s/it]
Training Epoch: 12/100, batch 217/375 completed (loss: 0.0806:  58%|█████▊    | 109/187 [10:58<07:53,  6.07s/it]
Training Epoch: 12/100, batch 217/375 completed (loss: 0.392:  58%|█████▊    | 109/187 [10:58<07:53,  6.07s/it] 
Training Epoch: 12/100, batch 217/375 completed (loss: 0.4079:  58%|█████▊    | 109/187 [10:58<07:54,  6.08s/it]
Training Epoch: 12/100, batch 218/375 completed (loss: 0.0098:  58%|█████▊    | 109/187 [11:01<07:53,  6.07s/it]
Training Epoch: 12/100, batch 218/375 completed (loss: 0.0262:  58%|█████▊    | 109/187 [11:01<07:53,  6.07s/it]
Training Epoch: 12/100, batch 218/375 completed (loss: 0.0326:  58%|█████▊    | 109/187 [11:01<07:53,  6.07s/it]
Training Epoch: 12/100, batch 218/375 completed (loss: 0.1981:  58%|█████▊    | 109/187 [11:01<07:54,  6.08s/it]
Training Epoch: 12/100, batch 218/375 completed (loss: 0.1981:  59%|█████▉    | 110/187 [11:04<07:45,  6.04s/it]
Training Epoch: 12/100, batch 218/375 completed (loss: 0.0262:  59%|█████▉    | 110/187 [11:04<07:45,  6.04s/it]
Training Epoch: 12/100, batch 218/375 completed (loss: 0.0326:  59%|█████▉    | 110/187 [11:04<07:45,  6.05s/it]
Training Epoch: 12/100, batch 218/375 completed (loss: 0.0098:  59%|█████▉    | 110/187 [11:04<07:45,  6.04s/it]
Training Epoch: 12/100, batch 219/375 completed (loss: 0.045:  59%|█████▉    | 110/187 [11:04<07:45,  6.04s/it] 
Training Epoch: 12/100, batch 219/375 completed (loss: 0.079:  59%|█████▉    | 110/187 [11:04<07:45,  6.05s/it] 
Training Epoch: 12/100, batch 219/375 completed (loss: 0.0541:  59%|█████▉    | 110/187 [11:04<07:45,  6.04s/it]
Training Epoch: 12/100, batch 219/375 completed (loss: 0.1408:  59%|█████▉    | 110/187 [11:04<07:45,  6.04s/it]
Training Epoch: 12/100, batch 220/375 completed (loss: 0.0673:  59%|█████▉    | 110/187 [11:07<07:45,  6.04s/it]
Training Epoch: 12/100, batch 220/375 completed (loss: 0.0569:  59%|█████▉    | 110/187 [11:07<07:45,  6.05s/it]
Training Epoch: 12/100, batch 220/375 completed (loss: 0.0264:  59%|█████▉    | 110/187 [11:07<07:45,  6.04s/it]
Training Epoch: 12/100, batch 220/375 completed (loss: 0.5034:  59%|█████▉    | 110/187 [11:07<07:45,  6.04s/it]
Training Epoch: 12/100, batch 220/375 completed (loss: 0.0673:  59%|█████▉    | 111/187 [11:10<07:45,  6.13s/it]
Training Epoch: 12/100, batch 220/375 completed (loss: 0.0264:  59%|█████▉    | 111/187 [11:10<07:45,  6.13s/it]
Training Epoch: 12/100, batch 220/375 completed (loss: 0.0569:  59%|█████▉    | 111/187 [11:10<07:45,  6.13s/it]
Training Epoch: 12/100, batch 220/375 completed (loss: 0.5034:  59%|█████▉    | 111/187 [11:10<07:45,  6.13s/it]
Training Epoch: 12/100, batch 221/375 completed (loss: 0.5257:  59%|█████▉    | 111/187 [11:10<07:45,  6.13s/it]
Training Epoch: 12/100, batch 221/375 completed (loss: 0.1663:  59%|█████▉    | 111/187 [11:10<07:45,  6.13s/it]
Training Epoch: 12/100, batch 221/375 completed (loss: 0.0576:  59%|█████▉    | 111/187 [11:10<07:45,  6.13s/it]
Training Epoch: 12/100, batch 221/375 completed (loss: 0.0137:  59%|█████▉    | 111/187 [11:10<07:45,  6.13s/it]
Training Epoch: 12/100, batch 222/375 completed (loss: 0.3132:  59%|█████▉    | 111/187 [11:13<07:45,  6.13s/it]
Training Epoch: 12/100, batch 222/375 completed (loss: 0.6723:  59%|█████▉    | 111/187 [11:13<07:45,  6.13s/it]
Training Epoch: 12/100, batch 222/375 completed (loss: 0.0363:  59%|█████▉    | 111/187 [11:13<07:45,  6.13s/it]
Training Epoch: 12/100, batch 222/375 completed (loss: 0.2557:  59%|█████▉    | 111/187 [11:13<07:45,  6.13s/it]
Training Epoch: 12/100, batch 222/375 completed (loss: 0.2557:  60%|█████▉    | 112/187 [11:16<07:33,  6.04s/it]
Training Epoch: 12/100, batch 222/375 completed (loss: 0.6723:  60%|█████▉    | 112/187 [11:16<07:33,  6.04s/it]
Training Epoch: 12/100, batch 222/375 completed (loss: 0.3132:  60%|█████▉    | 112/187 [11:16<07:33,  6.04s/it]
Training Epoch: 12/100, batch 222/375 completed (loss: 0.0363:  60%|█████▉    | 112/187 [11:16<07:33,  6.04s/it]
Training Epoch: 12/100, batch 223/375 completed (loss: 0.0317:  60%|█████▉    | 112/187 [11:16<07:33,  6.04s/it]
Training Epoch: 12/100, batch 223/375 completed (loss: 0.2223:  60%|█████▉    | 112/187 [11:16<07:33,  6.04s/it]
Training Epoch: 12/100, batch 223/375 completed (loss: 0.0858:  60%|█████▉    | 112/187 [11:16<07:33,  6.04s/it]
Training Epoch: 12/100, batch 223/375 completed (loss: 0.0334:  60%|█████▉    | 112/187 [11:16<07:33,  6.04s/it]
Training Epoch: 12/100, batch 224/375 completed (loss: 0.1473:  60%|█████▉    | 112/187 [11:19<07:33,  6.04s/it]
Training Epoch: 12/100, batch 224/375 completed (loss: 0.0503:  60%|█████▉    | 112/187 [11:19<07:33,  6.04s/it]
Training Epoch: 12/100, batch 224/375 completed (loss: 0.0738:  60%|█████▉    | 112/187 [11:19<07:33,  6.04s/it]
Training Epoch: 12/100, batch 224/375 completed (loss: 0.0611:  60%|█████▉    | 112/187 [11:19<07:33,  6.04s/it]
Training Epoch: 12/100, batch 224/375 completed (loss: 0.0611:  60%|██████    | 113/187 [11:22<07:32,  6.11s/it]
Training Epoch: 12/100, batch 224/375 completed (loss: 0.0738:  60%|██████    | 113/187 [11:22<07:32,  6.11s/it]
Training Epoch: 12/100, batch 224/375 completed (loss: 0.1473:  60%|██████    | 113/187 [11:22<07:32,  6.11s/it]
Training Epoch: 12/100, batch 224/375 completed (loss: 0.0503:  60%|██████    | 113/187 [11:22<07:32,  6.11s/it]
Training Epoch: 12/100, batch 225/375 completed (loss: 0.1404:  60%|██████    | 113/187 [11:22<07:32,  6.11s/it]
Training Epoch: 12/100, batch 225/375 completed (loss: 0.4918:  60%|██████    | 113/187 [11:22<07:32,  6.11s/it]
Training Epoch: 12/100, batch 225/375 completed (loss: 0.1765:  60%|██████    | 113/187 [11:22<07:32,  6.11s/it]
Training Epoch: 12/100, batch 225/375 completed (loss: 0.0316:  60%|██████    | 113/187 [11:22<07:32,  6.11s/it]
Training Epoch: 12/100, batch 226/375 completed (loss: 0.1584:  60%|██████    | 113/187 [11:25<07:32,  6.11s/it]
Training Epoch: 12/100, batch 226/375 completed (loss: 0.0769:  60%|██████    | 113/187 [11:25<07:32,  6.11s/it]
Training Epoch: 12/100, batch 226/375 completed (loss: 0.0545:  60%|██████    | 113/187 [11:25<07:32,  6.11s/it]
Training Epoch: 12/100, batch 226/375 completed (loss: 0.02:  60%|██████    | 113/187 [11:25<07:32,  6.11s/it]  
Training Epoch: 12/100, batch 226/375 completed (loss: 0.1584:  61%|██████    | 114/187 [11:28<07:23,  6.08s/it]
Training Epoch: 12/100, batch 226/375 completed (loss: 0.02:  61%|██████    | 114/187 [11:28<07:23,  6.08s/it]
Training Epoch: 12/100, batch 226/375 completed (loss: 0.0545:  61%|██████    | 114/187 [11:28<07:23,  6.08s/it]
Training Epoch: 12/100, batch 226/375 completed (loss: 0.0769:  61%|██████    | 114/187 [11:28<07:24,  6.08s/it]
Training Epoch: 12/100, batch 227/375 completed (loss: 0.0441:  61%|██████    | 114/187 [11:28<07:23,  6.08s/it]
Training Epoch: 12/100, batch 227/375 completed (loss: 0.111:  61%|██████    | 114/187 [11:28<07:24,  6.08s/it] 
Training Epoch: 12/100, batch 227/375 completed (loss: 0.095:  61%|██████    | 114/187 [11:28<07:23,  6.08s/it] 
Training Epoch: 12/100, batch 227/375 completed (loss: 0.0221:  61%|██████    | 114/187 [11:28<07:23,  6.08s/it]
Training Epoch: 12/100, batch 228/375 completed (loss: 0.079:  61%|██████    | 114/187 [11:31<07:23,  6.08s/it] 
Training Epoch: 12/100, batch 228/375 completed (loss: 0.0592:  61%|██████    | 114/187 [11:31<07:24,  6.08s/it]
Training Epoch: 12/100, batch 228/375 completed (loss: 0.0851:  61%|██████    | 114/187 [11:31<07:23,  6.08s/it]
Training Epoch: 12/100, batch 228/375 completed (loss: 0.1463:  61%|██████    | 114/187 [11:31<07:23,  6.08s/it]
Training Epoch: 12/100, batch 228/375 completed (loss: 0.1463:  61%|██████▏   | 115/187 [11:34<07:15,  6.05s/it]
Training Epoch: 12/100, batch 228/375 completed (loss: 0.079:  61%|██████▏   | 115/187 [11:34<07:15,  6.05s/it]
Training Epoch: 12/100, batch 228/375 completed (loss: 0.0851:  61%|██████▏   | 115/187 [11:34<07:16,  6.06s/it]
Training Epoch: 12/100, batch 228/375 completed (loss: 0.0592:  61%|██████▏   | 115/187 [11:34<07:16,  6.06s/it]
Training Epoch: 12/100, batch 229/375 completed (loss: 0.24:  61%|██████▏   | 115/187 [11:34<07:15,  6.05s/it] 
Training Epoch: 12/100, batch 229/375 completed (loss: 0.4048:  61%|██████▏   | 115/187 [11:34<07:16,  6.06s/it]
Training Epoch: 12/100, batch 229/375 completed (loss: 0.0155:  61%|██████▏   | 115/187 [11:34<07:16,  6.06s/it]
Training Epoch: 12/100, batch 229/375 completed (loss: 0.0521:  61%|██████▏   | 115/187 [11:34<07:15,  6.05s/it]
Training Epoch: 12/100, batch 230/375 completed (loss: 0.082:  61%|██████▏   | 115/187 [11:38<07:16,  6.06s/it] 
Training Epoch: 12/100, batch 230/375 completed (loss: 0.4575:  61%|██████▏   | 115/187 [11:38<07:15,  6.05s/it]
Training Epoch: 12/100, batch 230/375 completed (loss: 0.3865:  61%|██████▏   | 115/187 [11:38<07:16,  6.06s/it]
Training Epoch: 12/100, batch 230/375 completed (loss: 0.0361:  61%|██████▏   | 115/187 [11:38<07:15,  6.05s/it]
Training Epoch: 12/100, batch 230/375 completed (loss: 0.3865:  62%|██████▏   | 116/187 [11:41<07:11,  6.08s/it]
Training Epoch: 12/100, batch 230/375 completed (loss: 0.082:  62%|██████▏   | 116/187 [11:41<07:12,  6.09s/it]
Training Epoch: 12/100, batch 230/375 completed (loss: 0.0361:  62%|██████▏   | 116/187 [11:41<07:12,  6.09s/it]
Training Epoch: 12/100, batch 230/375 completed (loss: 0.4575:  62%|██████▏   | 116/187 [11:41<07:12,  6.09s/it]
Training Epoch: 12/100, batch 231/375 completed (loss: 0.0444:  62%|██████▏   | 116/187 [11:41<07:11,  6.08s/it]
Training Epoch: 12/100, batch 231/375 completed (loss: 0.1467:  62%|██████▏   | 116/187 [11:41<07:12,  6.09s/it]
Training Epoch: 12/100, batch 231/375 completed (loss: 0.0173:  62%|██████▏   | 116/187 [11:41<07:12,  6.09s/it]
Training Epoch: 12/100, batch 231/375 completed (loss: 0.1545:  62%|██████▏   | 116/187 [11:41<07:12,  6.09s/it]
Training Epoch: 12/100, batch 232/375 completed (loss: 0.2258:  62%|██████▏   | 116/187 [11:44<07:12,  6.09s/it]
Training Epoch: 12/100, batch 232/375 completed (loss: 0.4671:  62%|██████▏   | 116/187 [11:44<07:11,  6.08s/it]
Training Epoch: 12/100, batch 232/375 completed (loss: 0.1337:  62%|██████▏   | 116/187 [11:44<07:12,  6.09s/it]
Training Epoch: 12/100, batch 232/375 completed (loss: 0.0166:  62%|██████▏   | 116/187 [11:44<07:12,  6.09s/it]
Training Epoch: 12/100, batch 232/375 completed (loss: 0.0166:  63%|██████▎   | 117/187 [11:47<07:04,  6.07s/it]
Training Epoch: 12/100, batch 232/375 completed (loss: 0.1337:  63%|██████▎   | 117/187 [11:47<07:04,  6.07s/it]
Training Epoch: 12/100, batch 232/375 completed (loss: 0.4671:  63%|██████▎   | 117/187 [11:47<07:04,  6.07s/it]
Training Epoch: 12/100, batch 232/375 completed (loss: 0.2258:  63%|██████▎   | 117/187 [11:47<07:04,  6.07s/it]
Training Epoch: 12/100, batch 233/375 completed (loss: 0.1328:  63%|██████▎   | 117/187 [11:47<07:04,  6.07s/it]
Training Epoch: 12/100, batch 233/375 completed (loss: 0.0934:  63%|██████▎   | 117/187 [11:47<07:04,  6.07s/it]
Training Epoch: 12/100, batch 233/375 completed (loss: 0.0841:  63%|██████▎   | 117/187 [11:47<07:04,  6.07s/it]
Training Epoch: 12/100, batch 233/375 completed (loss: 0.1171:  63%|██████▎   | 117/187 [11:47<07:04,  6.07s/it]
Training Epoch: 12/100, batch 234/375 completed (loss: 0.0174:  63%|██████▎   | 117/187 [11:50<07:04,  6.07s/it]
Training Epoch: 12/100, batch 234/375 completed (loss: 0.0774:  63%|██████▎   | 117/187 [11:50<07:04,  6.07s/it]
Training Epoch: 12/100, batch 234/375 completed (loss: 0.031:  63%|██████▎   | 117/187 [11:50<07:04,  6.07s/it] 
Training Epoch: 12/100, batch 234/375 completed (loss: 0.0605:  63%|██████▎   | 117/187 [11:50<07:04,  6.07s/it]
Training Epoch: 12/100, batch 234/375 completed (loss: 0.0605:  63%|██████▎   | 118/187 [11:53<07:00,  6.09s/it]
Training Epoch: 12/100, batch 234/375 completed (loss: 0.031:  63%|██████▎   | 118/187 [11:53<06:59,  6.08s/it]
Training Epoch: 12/100, batch 234/375 completed (loss: 0.0774:  63%|██████▎   | 118/187 [11:53<06:59,  6.09s/it]
Training Epoch: 12/100, batch 234/375 completed (loss: 0.0174:  63%|██████▎   | 118/187 [11:53<07:00,  6.09s/it]
Training Epoch: 12/100, batch 235/375 completed (loss: 0.3283:  63%|██████▎   | 118/187 [11:53<07:00,  6.09s/it]
Training Epoch: 12/100, batch 235/375 completed (loss: 0.0434:  63%|██████▎   | 118/187 [11:53<06:59,  6.08s/it]
Training Epoch: 12/100, batch 235/375 completed (loss: 0.0865:  63%|██████▎   | 118/187 [11:53<06:59,  6.09s/it]
Training Epoch: 12/100, batch 235/375 completed (loss: 0.0496:  63%|██████▎   | 118/187 [11:53<07:00,  6.09s/it]
Training Epoch: 12/100, batch 236/375 completed (loss: 0.1958:  63%|██████▎   | 118/187 [11:56<06:59,  6.09s/it]
Training Epoch: 12/100, batch 236/375 completed (loss: 0.0173:  63%|██████▎   | 118/187 [11:56<06:59,  6.08s/it]
Training Epoch: 12/100, batch 236/375 completed (loss: 0.4148:  63%|██████▎   | 118/187 [11:56<07:00,  6.09s/it]
Training Epoch: 12/100, batch 236/375 completed (loss: 0.051:  63%|██████▎   | 118/187 [11:56<07:00,  6.09s/it] 
Training Epoch: 12/100, batch 236/375 completed (loss: 0.0173:  64%|██████▎   | 119/187 [11:59<06:55,  6.11s/it]
Training Epoch: 12/100, batch 236/375 completed (loss: 0.051:  64%|██████▎   | 119/187 [11:59<06:55,  6.11s/it]
Training Epoch: 12/100, batch 236/375 completed (loss: 0.1958:  64%|██████▎   | 119/187 [11:59<06:55,  6.11s/it]
Training Epoch: 12/100, batch 236/375 completed (loss: 0.4148:  64%|██████▎   | 119/187 [11:59<06:55,  6.11s/it]
Training Epoch: 12/100, batch 237/375 completed (loss: 0.5151:  64%|██████▎   | 119/187 [11:59<06:55,  6.11s/it]
Training Epoch: 12/100, batch 237/375 completed (loss: 0.05:  64%|██████▎   | 119/187 [11:59<06:55,  6.11s/it]  
Training Epoch: 12/100, batch 237/375 completed (loss: 0.0484:  64%|██████▎   | 119/187 [11:59<06:55,  6.11s/it]
Training Epoch: 12/100, batch 237/375 completed (loss: 0.0339:  64%|██████▎   | 119/187 [11:59<06:55,  6.11s/it]
Training Epoch: 12/100, batch 238/375 completed (loss: 0.1219:  64%|██████▎   | 119/187 [12:02<06:55,  6.11s/it]
Training Epoch: 12/100, batch 238/375 completed (loss: 0.1056:  64%|██████▎   | 119/187 [12:02<06:55,  6.11s/it]
Training Epoch: 12/100, batch 238/375 completed (loss: 0.1015:  64%|██████▎   | 119/187 [12:02<06:55,  6.11s/it]
Training Epoch: 12/100, batch 238/375 completed (loss: 0.2285:  64%|██████▎   | 119/187 [12:02<06:55,  6.11s/it]
Training Epoch: 12/100, batch 238/375 completed (loss: 0.1015:  64%|██████▍   | 120/187 [12:05<06:46,  6.07s/it]
Training Epoch: 12/100, batch 238/375 completed (loss: 0.1219:  64%|██████▍   | 120/187 [12:05<06:46,  6.07s/it]
Training Epoch: 12/100, batch 238/375 completed (loss: 0.1056:  64%|██████▍   | 120/187 [12:05<06:46,  6.07s/it]
Training Epoch: 12/100, batch 238/375 completed (loss: 0.2285:  64%|██████▍   | 120/187 [12:05<06:46,  6.07s/it]
Training Epoch: 12/100, batch 239/375 completed (loss: 0.0387:  64%|██████▍   | 120/187 [12:05<06:46,  6.07s/it]
Training Epoch: 12/100, batch 239/375 completed (loss: 0.0529:  64%|██████▍   | 120/187 [12:05<06:46,  6.07s/it]
Training Epoch: 12/100, batch 239/375 completed (loss: 0.0782:  64%|██████▍   | 120/187 [12:05<06:46,  6.07s/it]
Training Epoch: 12/100, batch 239/375 completed (loss: 0.0652:  64%|██████▍   | 120/187 [12:05<06:46,  6.07s/it]
Training Epoch: 12/100, batch 240/375 completed (loss: 0.1997:  64%|██████▍   | 120/187 [12:08<06:46,  6.07s/it]
Training Epoch: 12/100, batch 240/375 completed (loss: 0.0428:  64%|██████▍   | 120/187 [12:08<06:46,  6.07s/it]
Training Epoch: 12/100, batch 240/375 completed (loss: 0.0459:  64%|██████▍   | 120/187 [12:08<06:46,  6.07s/it]
Training Epoch: 12/100, batch 240/375 completed (loss: 0.028:  64%|██████▍   | 120/187 [12:08<06:46,  6.07s/it] 
Training Epoch: 12/100, batch 240/375 completed (loss: 0.0459:  65%|██████▍   | 121/187 [12:11<06:41,  6.08s/it]
Training Epoch: 12/100, batch 240/375 completed (loss: 0.1997:  65%|██████▍   | 121/187 [12:11<06:41,  6.08s/it]
Training Epoch: 12/100, batch 240/375 completed (loss: 0.0428:  65%|██████▍   | 121/187 [12:11<06:41,  6.08s/it]
Training Epoch: 12/100, batch 240/375 completed (loss: 0.028:  65%|██████▍   | 121/187 [12:11<06:41,  6.09s/it]
Training Epoch: 12/100, batch 241/375 completed (loss: 0.2398:  65%|██████▍   | 121/187 [12:11<06:41,  6.08s/it]
Training Epoch: 12/100, batch 241/375 completed (loss: 0.1299:  65%|██████▍   | 121/187 [12:11<06:41,  6.08s/it]
Training Epoch: 12/100, batch 241/375 completed (loss: 0.2944:  65%|██████▍   | 121/187 [12:11<06:41,  6.08s/it]
Training Epoch: 12/100, batch 241/375 completed (loss: 0.255:  65%|██████▍   | 121/187 [12:11<06:41,  6.09s/it]
Training Epoch: 12/100, batch 242/375 completed (loss: 0.0114:  65%|██████▍   | 121/187 [12:14<06:41,  6.08s/it]
Training Epoch: 12/100, batch 242/375 completed (loss: 0.0596:  65%|██████▍   | 121/187 [12:14<06:41,  6.08s/it]
Training Epoch: 12/100, batch 242/375 completed (loss: 0.3511:  65%|██████▍   | 121/187 [12:14<06:41,  6.08s/it]
Training Epoch: 12/100, batch 242/375 completed (loss: 0.3876:  65%|██████▍   | 121/187 [12:14<06:41,  6.09s/it]
Training Epoch: 12/100, batch 242/375 completed (loss: 0.3876:  65%|██████▌   | 122/187 [12:17<06:34,  6.06s/it]
Training Epoch: 12/100, batch 242/375 completed (loss: 0.3511:  65%|██████▌   | 122/187 [12:17<06:34,  6.06s/it]
Training Epoch: 12/100, batch 242/375 completed (loss: 0.0114:  65%|██████▌   | 122/187 [12:17<06:34,  6.06s/it]
Training Epoch: 12/100, batch 242/375 completed (loss: 0.0596:  65%|██████▌   | 122/187 [12:17<06:34,  6.06s/it]
Training Epoch: 12/100, batch 243/375 completed (loss: 0.0389:  65%|██████▌   | 122/187 [12:17<06:34,  6.06s/it]
Training Epoch: 12/100, batch 243/375 completed (loss: 0.1486:  65%|██████▌   | 122/187 [12:17<06:34,  6.06s/it]
Training Epoch: 12/100, batch 243/375 completed (loss: 0.0312:  65%|██████▌   | 122/187 [12:17<06:34,  6.06s/it]
Training Epoch: 12/100, batch 243/375 completed (loss: 0.0142:  65%|██████▌   | 122/187 [12:17<06:34,  6.06s/it]
Training Epoch: 12/100, batch 244/375 completed (loss: 0.0417:  65%|██████▌   | 122/187 [12:20<06:34,  6.06s/it]
Training Epoch: 12/100, batch 244/375 completed (loss: 0.0521:  65%|██████▌   | 122/187 [12:20<06:34,  6.06s/it]
Training Epoch: 12/100, batch 244/375 completed (loss: 0.4237:  65%|██████▌   | 122/187 [12:20<06:34,  6.06s/it]
Training Epoch: 12/100, batch 244/375 completed (loss: 0.1606:  65%|██████▌   | 122/187 [12:20<06:34,  6.06s/it]
Training Epoch: 12/100, batch 244/375 completed (loss: 0.4237:  66%|██████▌   | 123/187 [12:23<06:28,  6.08s/it]
Training Epoch: 12/100, batch 244/375 completed (loss: 0.1606:  66%|██████▌   | 123/187 [12:23<06:29,  6.08s/it]
Training Epoch: 12/100, batch 244/375 completed (loss: 0.0521:  66%|██████▌   | 123/187 [12:23<06:29,  6.08s/it]
Training Epoch: 12/100, batch 244/375 completed (loss: 0.0417:  66%|██████▌   | 123/187 [12:23<06:29,  6.08s/it]
Training Epoch: 12/100, batch 245/375 completed (loss: 0.0686:  66%|██████▌   | 123/187 [12:23<06:28,  6.08s/it]
Training Epoch: 12/100, batch 245/375 completed (loss: 0.0772:  66%|██████▌   | 123/187 [12:23<06:29,  6.08s/it]
Training Epoch: 12/100, batch 245/375 completed (loss: 0.0298:  66%|██████▌   | 123/187 [12:23<06:29,  6.08s/it]
Training Epoch: 12/100, batch 245/375 completed (loss: 0.0572:  66%|██████▌   | 123/187 [12:23<06:29,  6.08s/it]
Training Epoch: 12/100, batch 246/375 completed (loss: 0.0499:  66%|██████▌   | 123/187 [12:26<06:29,  6.08s/it]
Training Epoch: 12/100, batch 246/375 completed (loss: 0.1299:  66%|██████▌   | 123/187 [12:26<06:29,  6.08s/it]
Training Epoch: 12/100, batch 246/375 completed (loss: 0.0192:  66%|██████▌   | 123/187 [12:26<06:28,  6.08s/it]
Training Epoch: 12/100, batch 246/375 completed (loss: 0.0619:  66%|██████▌   | 123/187 [12:26<06:29,  6.08s/it]
Training Epoch: 12/100, batch 246/375 completed (loss: 0.0619:  66%|██████▋   | 124/187 [12:29<06:21,  6.05s/it]
Training Epoch: 12/100, batch 246/375 completed (loss: 0.1299:  66%|██████▋   | 124/187 [12:29<06:21,  6.05s/it]
Training Epoch: 12/100, batch 246/375 completed (loss: 0.0499:  66%|██████▋   | 124/187 [12:29<06:21,  6.05s/it]
Training Epoch: 12/100, batch 246/375 completed (loss: 0.0192:  66%|██████▋   | 124/187 [12:29<06:21,  6.06s/it]
Training Epoch: 12/100, batch 247/375 completed (loss: 0.0108:  66%|██████▋   | 124/187 [12:29<06:21,  6.06s/it]
Training Epoch: 12/100, batch 247/375 completed (loss: 0.0835:  66%|██████▋   | 124/187 [12:29<06:21,  6.05s/it]
Training Epoch: 12/100, batch 247/375 completed (loss: 0.1174:  66%|██████▋   | 124/187 [12:29<06:21,  6.05s/it]
Training Epoch: 12/100, batch 247/375 completed (loss: 0.0473:  66%|██████▋   | 124/187 [12:29<06:21,  6.05s/it]
Training Epoch: 12/100, batch 248/375 completed (loss: 0.2599:  66%|██████▋   | 124/187 [12:32<06:21,  6.05s/it]
Training Epoch: 12/100, batch 248/375 completed (loss: 0.0496:  66%|██████▋   | 124/187 [12:32<06:21,  6.05s/it]
Training Epoch: 12/100, batch 248/375 completed (loss: 0.2491:  66%|██████▋   | 124/187 [12:32<06:21,  6.06s/it]
Training Epoch: 12/100, batch 248/375 completed (loss: 0.1835:  66%|██████▋   | 124/187 [12:32<06:21,  6.05s/it]
Training Epoch: 12/100, batch 248/375 completed (loss: 0.1835:  67%|██████▋   | 125/187 [12:35<06:14,  6.04s/it]
Training Epoch: 12/100, batch 248/375 completed (loss: 0.0496:  67%|██████▋   | 125/187 [12:35<06:14,  6.04s/it]
Training Epoch: 12/100, batch 248/375 completed (loss: 0.2491:  67%|██████▋   | 125/187 [12:35<06:14,  6.04s/it]
Training Epoch: 12/100, batch 248/375 completed (loss: 0.2599:  67%|██████▋   | 125/187 [12:35<06:14,  6.04s/it]
Training Epoch: 12/100, batch 249/375 completed (loss: 0.0416:  67%|██████▋   | 125/187 [12:35<06:14,  6.04s/it]
Training Epoch: 12/100, batch 249/375 completed (loss: 0.155:  67%|██████▋   | 125/187 [12:35<06:14,  6.04s/it] 
Training Epoch: 12/100, batch 249/375 completed (loss: 0.0285:  67%|██████▋   | 125/187 [12:35<06:14,  6.04s/it]
Training Epoch: 12/100, batch 249/375 completed (loss: 0.1753:  67%|██████▋   | 125/187 [12:35<06:14,  6.04s/it]
Training Epoch: 12/100, batch 250/375 completed (loss: 0.0168:  67%|██████▋   | 125/187 [12:38<06:14,  6.04s/it]
Training Epoch: 12/100, batch 250/375 completed (loss: 0.069:  67%|██████▋   | 125/187 [12:38<06:14,  6.04s/it] 
Training Epoch: 12/100, batch 250/375 completed (loss: 0.031:  67%|██████▋   | 125/187 [12:38<06:14,  6.04s/it] 
Training Epoch: 12/100, batch 250/375 completed (loss: 0.2151:  67%|██████▋   | 125/187 [12:38<06:14,  6.04s/it]
Training Epoch: 12/100, batch 250/375 completed (loss: 0.2151:  67%|██████▋   | 126/187 [12:41<06:12,  6.11s/it]
Training Epoch: 12/100, batch 250/375 completed (loss: 0.0168:  67%|██████▋   | 126/187 [12:41<06:12,  6.11s/it]
Training Epoch: 12/100, batch 250/375 completed (loss: 0.031:  67%|██████▋   | 126/187 [12:41<06:12,  6.11s/it]
Training Epoch: 12/100, batch 250/375 completed (loss: 0.069:  67%|██████▋   | 126/187 [12:41<06:12,  6.11s/it]
Training Epoch: 12/100, batch 251/375 completed (loss: 0.0544:  67%|██████▋   | 126/187 [12:41<06:12,  6.11s/it]
Training Epoch: 12/100, batch 251/375 completed (loss: 0.0779:  67%|██████▋   | 126/187 [12:41<06:12,  6.11s/it]
Training Epoch: 12/100, batch 251/375 completed (loss: 0.2944:  67%|██████▋   | 126/187 [12:41<06:12,  6.11s/it]
Training Epoch: 12/100, batch 251/375 completed (loss: 0.1899:  67%|██████▋   | 126/187 [12:41<06:12,  6.11s/it]
Training Epoch: 12/100, batch 252/375 completed (loss: 0.0471:  67%|██████▋   | 126/187 [12:44<06:12,  6.11s/it]
Training Epoch: 12/100, batch 252/375 completed (loss: 0.0414:  67%|██████▋   | 126/187 [12:44<06:12,  6.11s/it]
Training Epoch: 12/100, batch 252/375 completed (loss: 0.0467:  67%|██████▋   | 126/187 [12:44<06:12,  6.11s/it]
Training Epoch: 12/100, batch 252/375 completed (loss: 0.0251:  67%|██████▋   | 126/187 [12:44<06:12,  6.11s/it]
Training Epoch: 12/100, batch 252/375 completed (loss: 0.0467:  68%|██████▊   | 127/187 [12:47<06:02,  6.04s/it]
Training Epoch: 12/100, batch 252/375 completed (loss: 0.0414:  68%|██████▊   | 127/187 [12:47<06:02,  6.04s/it]
Training Epoch: 12/100, batch 252/375 completed (loss: 0.0471:  68%|██████▊   | 127/187 [12:47<06:02,  6.04s/it]
Training Epoch: 12/100, batch 252/375 completed (loss: 0.0251:  68%|██████▊   | 127/187 [12:47<06:02,  6.04s/it]
Training Epoch: 12/100, batch 253/375 completed (loss: 0.1132:  68%|██████▊   | 127/187 [12:47<06:02,  6.04s/it]
Training Epoch: 12/100, batch 253/375 completed (loss: 0.1082:  68%|██████▊   | 127/187 [12:47<06:02,  6.04s/it]
Training Epoch: 12/100, batch 253/375 completed (loss: 0.0299:  68%|██████▊   | 127/187 [12:47<06:02,  6.04s/it]
Training Epoch: 12/100, batch 253/375 completed (loss: 0.187:  68%|██████▊   | 127/187 [12:47<06:02,  6.04s/it] 
Training Epoch: 12/100, batch 254/375 completed (loss: 0.1366:  68%|██████▊   | 127/187 [12:50<06:02,  6.04s/it]
Training Epoch: 12/100, batch 254/375 completed (loss: 0.1539:  68%|██████▊   | 127/187 [12:50<06:02,  6.04s/it]
Training Epoch: 12/100, batch 254/375 completed (loss: 0.0554:  68%|██████▊   | 127/187 [12:50<06:02,  6.04s/it]
Training Epoch: 12/100, batch 254/375 completed (loss: 0.1253:  68%|██████▊   | 127/187 [12:50<06:02,  6.04s/it]
Training Epoch: 12/100, batch 254/375 completed (loss: 0.1253:  68%|██████▊   | 128/187 [12:53<05:59,  6.09s/it]
Training Epoch: 12/100, batch 254/375 completed (loss: 0.1366:  68%|██████▊   | 128/187 [12:53<05:59,  6.09s/it]
Training Epoch: 12/100, batch 254/375 completed (loss: 0.0554:  68%|██████▊   | 128/187 [12:53<05:59,  6.09s/it]
Training Epoch: 12/100, batch 254/375 completed (loss: 0.1539:  68%|██████▊   | 128/187 [12:53<05:59,  6.09s/it]
Training Epoch: 12/100, batch 255/375 completed (loss: 0.0132:  68%|██████▊   | 128/187 [12:53<05:59,  6.09s/it]
Training Epoch: 12/100, batch 255/375 completed (loss: 0.0579:  68%|██████▊   | 128/187 [12:53<05:59,  6.09s/it]
Training Epoch: 12/100, batch 255/375 completed (loss: 0.1745:  68%|██████▊   | 128/187 [12:53<05:59,  6.09s/it]
Training Epoch: 12/100, batch 255/375 completed (loss: 0.0321:  68%|██████▊   | 128/187 [12:54<05:59,  6.09s/it]
Training Epoch: 12/100, batch 256/375 completed (loss: 0.0279:  68%|██████▊   | 128/187 [12:57<05:59,  6.09s/it]
Training Epoch: 12/100, batch 256/375 completed (loss: 0.0265:  68%|██████▊   | 128/187 [12:57<05:59,  6.09s/it]
Training Epoch: 12/100, batch 256/375 completed (loss: 0.0336:  68%|██████▊   | 128/187 [12:57<05:59,  6.09s/it]
Training Epoch: 12/100, batch 256/375 completed (loss: 0.0836:  68%|██████▊   | 128/187 [12:57<05:59,  6.09s/it]
Training Epoch: 12/100, batch 256/375 completed (loss: 0.0265:  69%|██████▉   | 129/187 [12:59<05:52,  6.07s/it]
Training Epoch: 12/100, batch 256/375 completed (loss: 0.0336:  69%|██████▉   | 129/187 [12:59<05:52,  6.08s/it]
Training Epoch: 12/100, batch 256/375 completed (loss: 0.0279:  69%|██████▉   | 129/187 [12:59<05:52,  6.08s/it]
Training Epoch: 12/100, batch 256/375 completed (loss: 0.0836:  69%|██████▉   | 129/187 [13:00<05:52,  6.08s/it]
Training Epoch: 12/100, batch 257/375 completed (loss: 0.4074:  69%|██████▉   | 129/187 [13:00<05:52,  6.07s/it]
Training Epoch: 12/100, batch 257/375 completed (loss: 0.0233:  69%|██████▉   | 129/187 [13:00<05:52,  6.08s/it]
Training Epoch: 12/100, batch 257/375 completed (loss: 0.0676:  69%|██████▉   | 129/187 [13:00<05:52,  6.08s/it]
Training Epoch: 12/100, batch 257/375 completed (loss: 0.2516:  69%|██████▉   | 129/187 [13:00<05:52,  6.08s/it]
Training Epoch: 12/100, batch 258/375 completed (loss: 0.1574:  69%|██████▉   | 129/187 [13:02<05:52,  6.08s/it]
Training Epoch: 12/100, batch 258/375 completed (loss: 0.103:  69%|██████▉   | 129/187 [13:02<05:52,  6.07s/it] 
Training Epoch: 12/100, batch 258/375 completed (loss: 0.0574:  69%|██████▉   | 129/187 [13:02<05:52,  6.08s/it]
Training Epoch: 12/100, batch 258/375 completed (loss: 0.1331:  69%|██████▉   | 129/187 [13:02<05:52,  6.08s/it]
Training Epoch: 12/100, batch 258/375 completed (loss: 0.1331:  70%|██████▉   | 130/187 [13:06<05:45,  6.06s/it]
Training Epoch: 12/100, batch 258/375 completed (loss: 0.1574:  70%|██████▉   | 130/187 [13:06<05:45,  6.06s/it]
Training Epoch: 12/100, batch 258/375 completed (loss: 0.0574:  70%|██████▉   | 130/187 [13:06<05:45,  6.06s/it]
Training Epoch: 12/100, batch 258/375 completed (loss: 0.103:  70%|██████▉   | 130/187 [13:06<05:45,  6.06s/it]
Training Epoch: 12/100, batch 259/375 completed (loss: 0.0299:  70%|██████▉   | 130/187 [13:06<05:45,  6.06s/it]
Training Epoch: 12/100, batch 259/375 completed (loss: 0.0329:  70%|██████▉   | 130/187 [13:06<05:45,  6.06s/it]
Training Epoch: 12/100, batch 259/375 completed (loss: 0.5436:  70%|██████▉   | 130/187 [13:06<05:45,  6.06s/it]
Training Epoch: 12/100, batch 259/375 completed (loss: 0.0513:  70%|██████▉   | 130/187 [13:06<05:45,  6.06s/it]
Training Epoch: 12/100, batch 260/375 completed (loss: 0.4121:  70%|██████▉   | 130/187 [13:09<05:45,  6.06s/it]
Training Epoch: 12/100, batch 260/375 completed (loss: 0.1011:  70%|██████▉   | 130/187 [13:09<05:45,  6.06s/it]
Training Epoch: 12/100, batch 260/375 completed (loss: 0.0226:  70%|██████▉   | 130/187 [13:09<05:45,  6.06s/it]
Training Epoch: 12/100, batch 260/375 completed (loss: 0.1604:  70%|██████▉   | 130/187 [13:09<05:45,  6.06s/it]
Training Epoch: 12/100, batch 260/375 completed (loss: 0.1011:  70%|███████   | 131/187 [13:12<05:40,  6.09s/it]
Training Epoch: 12/100, batch 260/375 completed (loss: 0.1604:  70%|███████   | 131/187 [13:12<05:40,  6.09s/it]
Training Epoch: 12/100, batch 260/375 completed (loss: 0.4121:  70%|███████   | 131/187 [13:12<05:40,  6.09s/it]
Training Epoch: 12/100, batch 260/375 completed (loss: 0.0226:  70%|███████   | 131/187 [13:12<05:41,  6.09s/it]
Training Epoch: 12/100, batch 261/375 completed (loss: 0.0725:  70%|███████   | 131/187 [13:12<05:40,  6.09s/it]
Training Epoch: 12/100, batch 261/375 completed (loss: 0.1198:  70%|███████   | 131/187 [13:12<05:40,  6.09s/it]
Training Epoch: 12/100, batch 261/375 completed (loss: 0.0598:  70%|███████   | 131/187 [13:12<05:41,  6.09s/it]
Training Epoch: 12/100, batch 261/375 completed (loss: 0.2028:  70%|███████   | 131/187 [13:12<05:40,  6.09s/it]
Training Epoch: 12/100, batch 262/375 completed (loss: 0.0798:  70%|███████   | 131/187 [13:15<05:40,  6.09s/it]
Training Epoch: 12/100, batch 262/375 completed (loss: 0.0593:  70%|███████   | 131/187 [13:15<05:40,  6.09s/it]
Training Epoch: 12/100, batch 262/375 completed (loss: 0.0983:  70%|███████   | 131/187 [13:15<05:41,  6.09s/it]
Training Epoch: 12/100, batch 262/375 completed (loss: 0.1922:  70%|███████   | 131/187 [13:15<05:40,  6.09s/it]
Training Epoch: 12/100, batch 262/375 completed (loss: 0.1922:  71%|███████   | 132/187 [13:18<05:33,  6.07s/it]
Training Epoch: 12/100, batch 262/375 completed (loss: 0.0593:  71%|███████   | 132/187 [13:18<05:33,  6.07s/it]
Training Epoch: 12/100, batch 262/375 completed (loss: 0.0798:  71%|███████   | 132/187 [13:18<05:34,  6.07s/it]
Training Epoch: 12/100, batch 262/375 completed (loss: 0.0983:  71%|███████   | 132/187 [13:18<05:34,  6.08s/it]
Training Epoch: 12/100, batch 263/375 completed (loss: 0.0713:  71%|███████   | 132/187 [13:18<05:34,  6.08s/it]
Training Epoch: 12/100, batch 263/375 completed (loss: 0.1288:  71%|███████   | 132/187 [13:18<05:34,  6.07s/it]
Training Epoch: 12/100, batch 263/375 completed (loss: 0.0451:  71%|███████   | 132/187 [13:18<05:33,  6.07s/it]
Training Epoch: 12/100, batch 263/375 completed (loss: 0.0133:  71%|███████   | 132/187 [13:18<05:33,  6.07s/it]
Training Epoch: 12/100, batch 264/375 completed (loss: 0.0526:  71%|███████   | 132/187 [13:21<05:33,  6.07s/it]
Training Epoch: 12/100, batch 264/375 completed (loss: 0.3498:  71%|███████   | 132/187 [13:21<05:34,  6.07s/it]
Training Epoch: 12/100, batch 264/375 completed (loss: 0.242:  71%|███████   | 132/187 [13:21<05:34,  6.08s/it] 
Training Epoch: 12/100, batch 264/375 completed (loss: 0.2007:  71%|███████   | 132/187 [13:21<05:33,  6.07s/it]
Training Epoch: 12/100, batch 264/375 completed (loss: 0.3498:  71%|███████   | 133/187 [13:24<05:28,  6.09s/it]
Training Epoch: 12/100, batch 264/375 completed (loss: 0.2007:  71%|███████   | 133/187 [13:24<05:28,  6.09s/it]
Training Epoch: 12/100, batch 264/375 completed (loss: 0.0526:  71%|███████   | 133/187 [13:24<05:28,  6.09s/it]
Training Epoch: 12/100, batch 264/375 completed (loss: 0.242:  71%|███████   | 133/187 [13:24<05:28,  6.09s/it]
Training Epoch: 12/100, batch 265/375 completed (loss: 0.1392:  71%|███████   | 133/187 [13:24<05:28,  6.09s/it]
Training Epoch: 12/100, batch 265/375 completed (loss: 0.104:  71%|███████   | 133/187 [13:24<05:28,  6.09s/it] 
Training Epoch: 12/100, batch 265/375 completed (loss: 0.1112:  71%|███████   | 133/187 [13:24<05:28,  6.09s/it]
Training Epoch: 12/100, batch 265/375 completed (loss: 0.0175:  71%|███████   | 133/187 [13:24<05:28,  6.09s/it]
Training Epoch: 12/100, batch 266/375 completed (loss: 0.0945:  71%|███████   | 133/187 [13:27<05:28,  6.09s/it]
Training Epoch: 12/100, batch 266/375 completed (loss: 0.0168:  71%|███████   | 133/187 [13:27<05:28,  6.09s/it]
Training Epoch: 12/100, batch 266/375 completed (loss: 0.0471:  71%|███████   | 133/187 [13:27<05:28,  6.09s/it]
Training Epoch: 12/100, batch 266/375 completed (loss: 0.0255:  71%|███████   | 133/187 [13:27<05:28,  6.09s/it]
Training Epoch: 12/100, batch 266/375 completed (loss: 0.0945:  72%|███████▏  | 134/187 [13:30<05:23,  6.11s/it]
Training Epoch: 12/100, batch 266/375 completed (loss: 0.0255:  72%|███████▏  | 134/187 [13:30<05:23,  6.11s/it]
Training Epoch: 12/100, batch 266/375 completed (loss: 0.0168:  72%|███████▏  | 134/187 [13:30<05:23,  6.11s/it]
Training Epoch: 12/100, batch 266/375 completed (loss: 0.0471:  72%|███████▏  | 134/187 [13:30<05:23,  6.11s/it]
Training Epoch: 12/100, batch 267/375 completed (loss: 0.1767:  72%|███████▏  | 134/187 [13:30<05:23,  6.11s/it]
Training Epoch: 12/100, batch 267/375 completed (loss: 0.0308:  72%|███████▏  | 134/187 [13:30<05:23,  6.11s/it]
Training Epoch: 12/100, batch 267/375 completed (loss: 0.0573:  72%|███████▏  | 134/187 [13:30<05:23,  6.11s/it]
Training Epoch: 12/100, batch 267/375 completed (loss: 0.1863:  72%|███████▏  | 134/187 [13:30<05:23,  6.11s/it]
Training Epoch: 12/100, batch 268/375 completed (loss: 0.0569:  72%|███████▏  | 134/187 [13:33<05:23,  6.11s/it]
Training Epoch: 12/100, batch 268/375 completed (loss: 0.1526:  72%|███████▏  | 134/187 [13:33<05:23,  6.11s/it]
Training Epoch: 12/100, batch 268/375 completed (loss: 0.1228:  72%|███████▏  | 134/187 [13:33<05:23,  6.11s/it]
Training Epoch: 12/100, batch 268/375 completed (loss: 0.1352:  72%|███████▏  | 134/187 [13:33<05:23,  6.11s/it]
Training Epoch: 12/100, batch 268/375 completed (loss: 0.1352:  72%|███████▏  | 135/187 [13:36<05:13,  6.04s/it]
Training Epoch: 12/100, batch 268/375 completed (loss: 0.1228:  72%|███████▏  | 135/187 [13:36<05:14,  6.04s/it]
Training Epoch: 12/100, batch 268/375 completed (loss: 0.1526:  72%|███████▏  | 135/187 [13:36<05:14,  6.04s/it]
Training Epoch: 12/100, batch 268/375 completed (loss: 0.0569:  72%|███████▏  | 135/187 [13:36<05:14,  6.04s/it]
Training Epoch: 12/100, batch 269/375 completed (loss: 0.1069:  72%|███████▏  | 135/187 [13:36<05:14,  6.04s/it]
Training Epoch: 12/100, batch 269/375 completed (loss: 0.0591:  72%|███████▏  | 135/187 [13:36<05:14,  6.04s/it]
Training Epoch: 12/100, batch 269/375 completed (loss: 0.0931:  72%|███████▏  | 135/187 [13:36<05:14,  6.04s/it]
Training Epoch: 12/100, batch 269/375 completed (loss: 0.2692:  72%|███████▏  | 135/187 [13:36<05:13,  6.04s/it]
Training Epoch: 12/100, batch 270/375 completed (loss: 0.042:  72%|███████▏  | 135/187 [13:39<05:14,  6.04s/it] 
Training Epoch: 12/100, batch 270/375 completed (loss: 0.0249:  72%|███████▏  | 135/187 [13:39<05:14,  6.04s/it]
Training Epoch: 12/100, batch 270/375 completed (loss: 0.109:  72%|███████▏  | 135/187 [13:39<05:14,  6.04s/it] 
Training Epoch: 12/100, batch 270/375 completed (loss: 0.0248:  72%|███████▏  | 135/187 [13:39<05:13,  6.04s/it]
Training Epoch: 12/100, batch 270/375 completed (loss: 0.0249:  73%|███████▎  | 136/187 [13:42<05:12,  6.12s/it]
Training Epoch: 12/100, batch 270/375 completed (loss: 0.042:  73%|███████▎  | 136/187 [13:42<05:12,  6.12s/it]
Training Epoch: 12/100, batch 270/375 completed (loss: 0.109:  73%|███████▎  | 136/187 [13:42<05:12,  6.12s/it]
Training Epoch: 12/100, batch 270/375 completed (loss: 0.0248:  73%|███████▎  | 136/187 [13:42<05:12,  6.13s/it]
Training Epoch: 12/100, batch 271/375 completed (loss: 0.1399:  73%|███████▎  | 136/187 [13:42<05:12,  6.12s/it]
Training Epoch: 12/100, batch 271/375 completed (loss: 0.354:  73%|███████▎  | 136/187 [13:42<05:12,  6.12s/it]
Training Epoch: 12/100, batch 271/375 completed (loss: 0.0738:  73%|███████▎  | 136/187 [13:42<05:12,  6.12s/it]
Training Epoch: 12/100, batch 271/375 completed (loss: 0.1675:  73%|███████▎  | 136/187 [13:42<05:12,  6.13s/it]
Training Epoch: 12/100, batch 272/375 completed (loss: 0.054:  73%|███████▎  | 136/187 [13:45<05:12,  6.12s/it] 
Training Epoch: 12/100, batch 272/375 completed (loss: 0.2266:  73%|███████▎  | 136/187 [13:45<05:12,  6.12s/it]
Training Epoch: 12/100, batch 272/375 completed (loss: 0.0629:  73%|███████▎  | 136/187 [13:45<05:12,  6.12s/it]
Training Epoch: 12/100, batch 272/375 completed (loss: 0.1866:  73%|███████▎  | 136/187 [13:45<05:12,  6.13s/it]
Training Epoch: 12/100, batch 272/375 completed (loss: 0.0629:  73%|███████▎  | 137/187 [13:48<05:03,  6.08s/it]
Training Epoch: 12/100, batch 272/375 completed (loss: 0.2266:  73%|███████▎  | 137/187 [13:48<05:03,  6.08s/it]
Training Epoch: 12/100, batch 272/375 completed (loss: 0.054:  73%|███████▎  | 137/187 [13:48<05:03,  6.08s/it]
Training Epoch: 12/100, batch 272/375 completed (loss: 0.1866:  73%|███████▎  | 137/187 [13:48<05:04,  6.08s/it]
Training Epoch: 12/100, batch 273/375 completed (loss: 0.0775:  73%|███████▎  | 137/187 [13:48<05:03,  6.08s/it]
Training Epoch: 12/100, batch 273/375 completed (loss: 0.0145:  73%|███████▎  | 137/187 [13:48<05:03,  6.08s/it]
Training Epoch: 12/100, batch 273/375 completed (loss: 0.0193:  73%|███████▎  | 137/187 [13:48<05:03,  6.08s/it]
Training Epoch: 12/100, batch 273/375 completed (loss: 0.1702:  73%|███████▎  | 137/187 [13:48<05:04,  6.08s/it]
Training Epoch: 12/100, batch 274/375 completed (loss: 0.2394:  73%|███████▎  | 137/187 [13:51<05:03,  6.08s/it]
Training Epoch: 12/100, batch 274/375 completed (loss: 0.0125:  73%|███████▎  | 137/187 [13:51<05:03,  6.08s/it]
Training Epoch: 12/100, batch 274/375 completed (loss: 0.05:  73%|███████▎  | 137/187 [13:51<05:03,  6.08s/it]  
Training Epoch: 12/100, batch 274/375 completed (loss: 0.0203:  73%|███████▎  | 137/187 [13:51<05:04,  6.08s/it]
Training Epoch: 12/100, batch 274/375 completed (loss: 0.05:  74%|███████▍  | 138/187 [13:54<04:56,  6.05s/it]
Training Epoch: 12/100, batch 274/375 completed (loss: 0.2394:  74%|███████▍  | 138/187 [13:54<04:56,  6.05s/it]
Training Epoch: 12/100, batch 274/375 completed (loss: 0.0125:  74%|███████▍  | 138/187 [13:54<04:56,  6.05s/it]
Training Epoch: 12/100, batch 274/375 completed (loss: 0.0203:  74%|███████▍  | 138/187 [13:54<04:56,  6.05s/it]
Training Epoch: 12/100, batch 275/375 completed (loss: 0.0461:  74%|███████▍  | 138/187 [13:54<04:56,  6.05s/it]
Training Epoch: 12/100, batch 275/375 completed (loss: 0.1043:  74%|███████▍  | 138/187 [13:54<04:56,  6.05s/it]
Training Epoch: 12/100, batch 275/375 completed (loss: 0.0281:  74%|███████▍  | 138/187 [13:54<04:56,  6.05s/it]
Training Epoch: 12/100, batch 275/375 completed (loss: 0.0537:  74%|███████▍  | 138/187 [13:54<04:56,  6.05s/it]
Training Epoch: 12/100, batch 276/375 completed (loss: 0.0244:  74%|███████▍  | 138/187 [13:57<04:56,  6.05s/it]
Training Epoch: 12/100, batch 276/375 completed (loss: 0.1085:  74%|███████▍  | 138/187 [13:57<04:56,  6.05s/it]
Training Epoch: 12/100, batch 276/375 completed (loss: 0.0847:  74%|███████▍  | 138/187 [13:57<04:56,  6.05s/it]
Training Epoch: 12/100, batch 276/375 completed (loss: 0.111:  74%|███████▍  | 138/187 [13:57<04:56,  6.05s/it] 
Training Epoch: 12/100, batch 276/375 completed (loss: 0.1085:  74%|███████▍  | 139/187 [14:00<04:51,  6.07s/it]
Training Epoch: 12/100, batch 276/375 completed (loss: 0.0244:  74%|███████▍  | 139/187 [14:00<04:51,  6.07s/it]
Training Epoch: 12/100, batch 276/375 completed (loss: 0.0847:  74%|███████▍  | 139/187 [14:00<04:51,  6.07s/it]
Training Epoch: 12/100, batch 276/375 completed (loss: 0.111:  74%|███████▍  | 139/187 [14:00<04:51,  6.07s/it]
Training Epoch: 12/100, batch 277/375 completed (loss: 0.0195:  74%|███████▍  | 139/187 [14:00<04:51,  6.07s/it]
Training Epoch: 12/100, batch 277/375 completed (loss: 0.3528:  74%|███████▍  | 139/187 [14:00<04:51,  6.07s/it]
Training Epoch: 12/100, batch 277/375 completed (loss: 0.0173:  74%|███████▍  | 139/187 [14:00<04:51,  6.07s/it]
Training Epoch: 12/100, batch 277/375 completed (loss: 0.1119:  74%|███████▍  | 139/187 [14:00<04:51,  6.07s/it]
Training Epoch: 12/100, batch 278/375 completed (loss: 0.0585:  74%|███████▍  | 139/187 [14:03<04:51,  6.07s/it]
Training Epoch: 12/100, batch 278/375 completed (loss: 0.2145:  74%|███████▍  | 139/187 [14:03<04:51,  6.07s/it]
Training Epoch: 12/100, batch 278/375 completed (loss: 0.2154:  74%|███████▍  | 139/187 [14:03<04:51,  6.07s/it]
Training Epoch: 12/100, batch 278/375 completed (loss: 0.0284:  74%|███████▍  | 139/187 [14:03<04:51,  6.07s/it]
Training Epoch: 12/100, batch 278/375 completed (loss: 0.2145:  75%|███████▍  | 140/187 [14:06<04:44,  6.05s/it]
Training Epoch: 12/100, batch 278/375 completed (loss: 0.0585:  75%|███████▍  | 140/187 [14:06<04:44,  6.05s/it]
Training Epoch: 12/100, batch 278/375 completed (loss: 0.2154:  75%|███████▍  | 140/187 [14:06<04:44,  6.05s/it]
Training Epoch: 12/100, batch 278/375 completed (loss: 0.0284:  75%|███████▍  | 140/187 [14:06<04:44,  6.05s/it]
Training Epoch: 12/100, batch 279/375 completed (loss: 0.2689:  75%|███████▍  | 140/187 [14:06<04:44,  6.05s/it]
Training Epoch: 12/100, batch 279/375 completed (loss: 0.2095:  75%|███████▍  | 140/187 [14:06<04:44,  6.05s/it]
Training Epoch: 12/100, batch 279/375 completed (loss: 0.0339:  75%|███████▍  | 140/187 [14:06<04:44,  6.05s/it]
Training Epoch: 12/100, batch 279/375 completed (loss: 0.1981:  75%|███████▍  | 140/187 [14:06<04:44,  6.05s/it]
Training Epoch: 12/100, batch 280/375 completed (loss: 0.3544:  75%|███████▍  | 140/187 [14:09<04:44,  6.05s/it]
Training Epoch: 12/100, batch 280/375 completed (loss: 0.0696:  75%|███████▍  | 140/187 [14:09<04:44,  6.05s/it]
Training Epoch: 12/100, batch 280/375 completed (loss: 0.0358:  75%|███████▍  | 140/187 [14:09<04:44,  6.05s/it]
Training Epoch: 12/100, batch 280/375 completed (loss: 0.024:  75%|███████▍  | 140/187 [14:09<04:44,  6.05s/it] 
Training Epoch: 12/100, batch 280/375 completed (loss: 0.0358:  75%|███████▌  | 141/187 [14:12<04:39,  6.08s/it]
Training Epoch: 12/100, batch 280/375 completed (loss: 0.3544:  75%|███████▌  | 141/187 [14:12<04:39,  6.08s/it]
Training Epoch: 12/100, batch 280/375 completed (loss: 0.0696:  75%|███████▌  | 141/187 [14:12<04:39,  6.08s/it]
Training Epoch: 12/100, batch 280/375 completed (loss: 0.024:  75%|███████▌  | 141/187 [14:12<04:39,  6.08s/it]
Training Epoch: 12/100, batch 281/375 completed (loss: 0.0555:  75%|███████▌  | 141/187 [14:12<04:39,  6.08s/it]
Training Epoch: 12/100, batch 281/375 completed (loss: 0.0649:  75%|███████▌  | 141/187 [14:12<04:39,  6.08s/it]
Training Epoch: 12/100, batch 281/375 completed (loss: 0.019:  75%|███████▌  | 141/187 [14:12<04:39,  6.08s/it] 
Training Epoch: 12/100, batch 281/375 completed (loss: 0.0939:  75%|███████▌  | 141/187 [14:13<04:39,  6.08s/it]
Training Epoch: 12/100, batch 282/375 completed (loss: 0.1086:  75%|███████▌  | 141/187 [14:16<04:39,  6.08s/it]
Training Epoch: 12/100, batch 282/375 completed (loss: 0.1415:  75%|███████▌  | 141/187 [14:16<04:39,  6.08s/it]
Training Epoch: 12/100, batch 282/375 completed (loss: 0.086:  75%|███████▌  | 141/187 [14:16<04:39,  6.08s/it]
Training Epoch: 12/100, batch 282/375 completed (loss: 0.2217:  75%|███████▌  | 141/187 [14:16<04:39,  6.08s/it]
Training Epoch: 12/100, batch 282/375 completed (loss: 0.1086:  76%|███████▌  | 142/187 [14:19<04:34,  6.10s/it]
Training Epoch: 12/100, batch 282/375 completed (loss: 0.1415:  76%|███████▌  | 142/187 [14:19<04:34,  6.10s/it]
Training Epoch: 12/100, batch 282/375 completed (loss: 0.086:  76%|███████▌  | 142/187 [14:19<04:34,  6.10s/it]
Training Epoch: 12/100, batch 282/375 completed (loss: 0.2217:  76%|███████▌  | 142/187 [14:19<04:34,  6.11s/it]
Training Epoch: 12/100, batch 283/375 completed (loss: 0.0152:  76%|███████▌  | 142/187 [14:19<04:34,  6.10s/it]
Training Epoch: 12/100, batch 283/375 completed (loss: 0.0549:  76%|███████▌  | 142/187 [14:19<04:34,  6.10s/it]
Training Epoch: 12/100, batch 283/375 completed (loss: 0.0632:  76%|███████▌  | 142/187 [14:19<04:34,  6.10s/it]
Training Epoch: 12/100, batch 283/375 completed (loss: 0.0555:  76%|███████▌  | 142/187 [14:19<04:34,  6.11s/it]
Training Epoch: 12/100, batch 284/375 completed (loss: 0.1192:  76%|███████▌  | 142/187 [14:22<04:34,  6.10s/it]
Training Epoch: 12/100, batch 284/375 completed (loss: 0.2089:  76%|███████▌  | 142/187 [14:22<04:34,  6.10s/it]
Training Epoch: 12/100, batch 284/375 completed (loss: 0.0273:  76%|███████▌  | 142/187 [14:22<04:34,  6.10s/it]
Training Epoch: 12/100, batch 284/375 completed (loss: 0.0336:  76%|███████▌  | 142/187 [14:22<04:34,  6.11s/it]
Training Epoch: 12/100, batch 284/375 completed (loss: 0.0273:  76%|███████▋  | 143/187 [14:24<04:25,  6.02s/it]
Training Epoch: 12/100, batch 284/375 completed (loss: 0.1192:  76%|███████▋  | 143/187 [14:24<04:25,  6.02s/it]
Training Epoch: 12/100, batch 284/375 completed (loss: 0.2089:  76%|███████▋  | 143/187 [14:24<04:25,  6.02s/it]
Training Epoch: 12/100, batch 284/375 completed (loss: 0.0336:  76%|███████▋  | 143/187 [14:24<04:25,  6.02s/it]
Training Epoch: 12/100, batch 285/375 completed (loss: 0.1167:  76%|███████▋  | 143/187 [14:24<04:25,  6.02s/it]
Training Epoch: 12/100, batch 285/375 completed (loss: 0.1218:  76%|███████▋  | 143/187 [14:24<04:25,  6.02s/it]
Training Epoch: 12/100, batch 285/375 completed (loss: 0.2614:  76%|███████▋  | 143/187 [14:24<04:25,  6.02s/it]
Training Epoch: 12/100, batch 285/375 completed (loss: 0.0503:  76%|███████▋  | 143/187 [14:24<04:25,  6.02s/it]
Training Epoch: 12/100, batch 286/375 completed (loss: 0.0745:  76%|███████▋  | 143/187 [14:28<04:25,  6.02s/it]
Training Epoch: 12/100, batch 286/375 completed (loss: 0.0318:  76%|███████▋  | 143/187 [14:28<04:25,  6.02s/it]
Training Epoch: 12/100, batch 286/375 completed (loss: 0.0243:  76%|███████▋  | 143/187 [14:28<04:25,  6.02s/it]
Training Epoch: 12/100, batch 286/375 completed (loss: 0.0183:  76%|███████▋  | 143/187 [14:28<04:25,  6.02s/it]
Training Epoch: 12/100, batch 286/375 completed (loss: 0.0243:  77%|███████▋  | 144/187 [14:31<04:22,  6.10s/it]
Training Epoch: 12/100, batch 286/375 completed (loss: 0.0745:  77%|███████▋  | 144/187 [14:31<04:22,  6.11s/it]
Training Epoch: 12/100, batch 286/375 completed (loss: 0.0318:  77%|███████▋  | 144/187 [14:31<04:22,  6.11s/it]
Training Epoch: 12/100, batch 286/375 completed (loss: 0.0183:  77%|███████▋  | 144/187 [14:31<04:22,  6.11s/it]
Training Epoch: 12/100, batch 287/375 completed (loss: 0.045:  77%|███████▋  | 144/187 [14:31<04:22,  6.11s/it] 
Training Epoch: 12/100, batch 287/375 completed (loss: 0.0232:  77%|███████▋  | 144/187 [14:31<04:22,  6.10s/it]
Training Epoch: 12/100, batch 287/375 completed (loss: 0.1341:  77%|███████▋  | 144/187 [14:31<04:22,  6.11s/it]
Training Epoch: 12/100, batch 287/375 completed (loss: 0.0605:  77%|███████▋  | 144/187 [14:31<04:22,  6.11s/it]
Training Epoch: 12/100, batch 288/375 completed (loss: 0.0734:  77%|███████▋  | 144/187 [14:34<04:22,  6.11s/it]
Training Epoch: 12/100, batch 288/375 completed (loss: 0.1178:  77%|███████▋  | 144/187 [14:34<04:22,  6.11s/it]
Training Epoch: 12/100, batch 288/375 completed (loss: 0.0912:  77%|███████▋  | 144/187 [14:34<04:22,  6.10s/it]
Training Epoch: 12/100, batch 288/375 completed (loss: 0.3959:  77%|███████▋  | 144/187 [14:34<04:22,  6.11s/it]
Training Epoch: 12/100, batch 288/375 completed (loss: 0.3959:  78%|███████▊  | 145/187 [14:37<04:14,  6.07s/it]
Training Epoch: 12/100, batch 288/375 completed (loss: 0.0912:  78%|███████▊  | 145/187 [14:37<04:15,  6.07s/it]
Training Epoch: 12/100, batch 288/375 completed (loss: 0.1178:  78%|███████▊  | 145/187 [14:37<04:15,  6.08s/it]
Training Epoch: 12/100, batch 288/375 completed (loss: 0.0734:  78%|███████▊  | 145/187 [14:37<04:15,  6.08s/it]
Training Epoch: 12/100, batch 289/375 completed (loss: 0.2123:  78%|███████▊  | 145/187 [14:37<04:15,  6.08s/it]
Training Epoch: 12/100, batch 289/375 completed (loss: 0.0987:  78%|███████▊  | 145/187 [14:37<04:15,  6.07s/it]
Training Epoch: 12/100, batch 289/375 completed (loss: 0.0548:  78%|███████▊  | 145/187 [14:37<04:15,  6.08s/it]
Training Epoch: 12/100, batch 289/375 completed (loss: 0.0577:  78%|███████▊  | 145/187 [14:37<04:14,  6.07s/it]
Training Epoch: 12/100, batch 290/375 completed (loss: 0.0673:  78%|███████▊  | 145/187 [14:40<04:15,  6.08s/it]
Training Epoch: 12/100, batch 290/375 completed (loss: 0.0509:  78%|███████▊  | 145/187 [14:40<04:15,  6.08s/it]
Training Epoch: 12/100, batch 290/375 completed (loss: 0.2493:  78%|███████▊  | 145/187 [14:40<04:15,  6.07s/it]
Training Epoch: 12/100, batch 290/375 completed (loss: 0.0849:  78%|███████▊  | 145/187 [14:40<04:14,  6.07s/it]
Training Epoch: 12/100, batch 290/375 completed (loss: 0.0509:  78%|███████▊  | 146/187 [14:43<04:07,  6.05s/it]
Training Epoch: 12/100, batch 290/375 completed (loss: 0.0673:  78%|███████▊  | 146/187 [14:43<04:07,  6.05s/it]
Training Epoch: 12/100, batch 290/375 completed (loss: 0.2493:  78%|███████▊  | 146/187 [14:43<04:08,  6.05s/it]
Training Epoch: 12/100, batch 290/375 completed (loss: 0.0849:  78%|███████▊  | 146/187 [14:43<04:08,  6.05s/it]
Training Epoch: 12/100, batch 291/375 completed (loss: 0.0795:  78%|███████▊  | 146/187 [14:43<04:08,  6.05s/it]
Training Epoch: 12/100, batch 291/375 completed (loss: 0.3075:  78%|███████▊  | 146/187 [14:43<04:07,  6.05s/it]
Training Epoch: 12/100, batch 291/375 completed (loss: 0.0546:  78%|███████▊  | 146/187 [14:43<04:07,  6.05s/it]
Training Epoch: 12/100, batch 291/375 completed (loss: 0.1402:  78%|███████▊  | 146/187 [14:43<04:08,  6.05s/it]
Training Epoch: 12/100, batch 292/375 completed (loss: 0.1803:  78%|███████▊  | 146/187 [14:46<04:07,  6.05s/it]
Training Epoch: 12/100, batch 292/375 completed (loss: 0.1342:  78%|███████▊  | 146/187 [14:46<04:07,  6.05s/it]
Training Epoch: 12/100, batch 292/375 completed (loss: 0.0401:  78%|███████▊  | 146/187 [14:46<04:08,  6.05s/it]
Training Epoch: 12/100, batch 292/375 completed (loss: 0.2236:  78%|███████▊  | 146/187 [14:46<04:08,  6.05s/it]
Training Epoch: 12/100, batch 292/375 completed (loss: 0.1803:  79%|███████▊  | 147/187 [14:49<04:05,  6.14s/it]
Training Epoch: 12/100, batch 292/375 completed (loss: 0.1342:  79%|███████▊  | 147/187 [14:49<04:05,  6.14s/it]
Training Epoch: 12/100, batch 292/375 completed (loss: 0.2236:  79%|███████▊  | 147/187 [14:49<04:05,  6.13s/it]
Training Epoch: 12/100, batch 292/375 completed (loss: 0.0401:  79%|███████▊  | 147/187 [14:49<04:05,  6.13s/it]
Training Epoch: 12/100, batch 293/375 completed (loss: 0.108:  79%|███████▊  | 147/187 [14:49<04:05,  6.14s/it] 
Training Epoch: 12/100, batch 293/375 completed (loss: 0.0176:  79%|███████▊  | 147/187 [14:49<04:05,  6.13s/it]
Training Epoch: 12/100, batch 293/375 completed (loss: 0.0954:  79%|███████▊  | 147/187 [14:49<04:05,  6.14s/it]
Training Epoch: 12/100, batch 293/375 completed (loss: 0.0966:  79%|███████▊  | 147/187 [14:49<04:05,  6.13s/it]
Training Epoch: 12/100, batch 294/375 completed (loss: 0.1505:  79%|███████▊  | 147/187 [14:52<04:05,  6.14s/it]
Training Epoch: 12/100, batch 294/375 completed (loss: 0.0849:  79%|███████▊  | 147/187 [14:52<04:05,  6.14s/it]
Training Epoch: 12/100, batch 294/375 completed (loss: 0.0722:  79%|███████▊  | 147/187 [14:52<04:05,  6.13s/it]
Training Epoch: 12/100, batch 294/375 completed (loss: 0.2841:  79%|███████▊  | 147/187 [14:52<04:05,  6.13s/it]
Training Epoch: 12/100, batch 294/375 completed (loss: 0.0849:  79%|███████▉  | 148/187 [14:55<03:56,  6.05s/it]
Training Epoch: 12/100, batch 294/375 completed (loss: 0.2841:  79%|███████▉  | 148/187 [14:55<03:55,  6.05s/it]
Training Epoch: 12/100, batch 294/375 completed (loss: 0.1505:  79%|███████▉  | 148/187 [14:55<03:56,  6.05s/it]
Training Epoch: 12/100, batch 294/375 completed (loss: 0.0722:  79%|███████▉  | 148/187 [14:55<03:56,  6.05s/it]
Training Epoch: 12/100, batch 295/375 completed (loss: 0.2815:  79%|███████▉  | 148/187 [14:55<03:56,  6.05s/it]
Training Epoch: 12/100, batch 295/375 completed (loss: 0.0972:  79%|███████▉  | 148/187 [14:55<03:56,  6.05s/it]
Training Epoch: 12/100, batch 295/375 completed (loss: 0.1411:  79%|███████▉  | 148/187 [14:55<03:56,  6.05s/it]
Training Epoch: 12/100, batch 295/375 completed (loss: 0.0214:  79%|███████▉  | 148/187 [14:55<03:55,  6.05s/it]
Training Epoch: 12/100, batch 296/375 completed (loss: 0.0374:  79%|███████▉  | 148/187 [14:58<03:56,  6.05s/it]
Training Epoch: 12/100, batch 296/375 completed (loss: 0.0316:  79%|███████▉  | 148/187 [14:58<03:56,  6.05s/it]
Training Epoch: 12/100, batch 296/375 completed (loss: 0.1084:  79%|███████▉  | 148/187 [14:58<03:56,  6.05s/it]
Training Epoch: 12/100, batch 296/375 completed (loss: 0.031:  79%|███████▉  | 148/187 [14:58<03:55,  6.05s/it] 
Training Epoch: 12/100, batch 296/375 completed (loss: 0.1084:  80%|███████▉  | 149/187 [15:01<03:51,  6.09s/it]
Training Epoch: 12/100, batch 296/375 completed (loss: 0.0374:  80%|███████▉  | 149/187 [15:01<03:51,  6.09s/it]
Training Epoch: 12/100, batch 296/375 completed (loss: 0.0316:  80%|███████▉  | 149/187 [15:01<03:51,  6.09s/it]
Training Epoch: 12/100, batch 296/375 completed (loss: 0.031:  80%|███████▉  | 149/187 [15:01<03:51,  6.09s/it]
Training Epoch: 12/100, batch 297/375 completed (loss: 0.2506:  80%|███████▉  | 149/187 [15:01<03:51,  6.09s/it]
Training Epoch: 12/100, batch 297/375 completed (loss: 0.2787:  80%|███████▉  | 149/187 [15:01<03:51,  6.09s/it]
Training Epoch: 12/100, batch 297/375 completed (loss: 0.2706:  80%|███████▉  | 149/187 [15:01<03:51,  6.09s/it]
Training Epoch: 12/100, batch 297/375 completed (loss: 0.0184:  80%|███████▉  | 149/187 [15:01<03:51,  6.09s/it]
Training Epoch: 12/100, batch 298/375 completed (loss: 0.181:  80%|███████▉  | 149/187 [15:04<03:51,  6.09s/it] 
Training Epoch: 12/100, batch 298/375 completed (loss: 0.0973:  80%|███████▉  | 149/187 [15:04<03:51,  6.09s/it]
Training Epoch: 12/100, batch 298/375 completed (loss: 0.0522:  80%|███████▉  | 149/187 [15:04<03:51,  6.09s/it]
Training Epoch: 12/100, batch 298/375 completed (loss: 0.06:  80%|███████▉  | 149/187 [15:04<03:51,  6.09s/it]  
Training Epoch: 12/100, batch 298/375 completed (loss: 0.0973:  80%|████████  | 150/187 [15:07<03:46,  6.11s/it]
Training Epoch: 12/100, batch 298/375 completed (loss: 0.0522:  80%|████████  | 150/187 [15:07<03:46,  6.11s/it]
Training Epoch: 12/100, batch 298/375 completed (loss: 0.06:  80%|████████  | 150/187 [15:07<03:46,  6.11s/it]
Training Epoch: 12/100, batch 298/375 completed (loss: 0.181:  80%|████████  | 150/187 [15:07<03:46,  6.11s/it]
Training Epoch: 12/100, batch 299/375 completed (loss: 0.1731:  80%|████████  | 150/187 [15:07<03:46,  6.11s/it]
Training Epoch: 12/100, batch 299/375 completed (loss: 0.1781:  80%|████████  | 150/187 [15:07<03:46,  6.11s/it]
Training Epoch: 12/100, batch 299/375 completed (loss: 0.2825:  80%|████████  | 150/187 [15:07<03:46,  6.11s/it]
Training Epoch: 12/100, batch 299/375 completed (loss: 0.1017:  80%|████████  | 150/187 [15:07<03:46,  6.11s/it]
Training Epoch: 12/100, batch 300/375 completed (loss: 0.0615:  80%|████████  | 150/187 [15:10<03:46,  6.11s/it]
Training Epoch: 12/100, batch 300/375 completed (loss: 0.0296:  80%|████████  | 150/187 [15:10<03:46,  6.11s/it]
Training Epoch: 12/100, batch 300/375 completed (loss: 0.254:  80%|████████  | 150/187 [15:10<03:46,  6.11s/it] 
Training Epoch: 12/100, batch 300/375 completed (loss: 0.3912:  80%|████████  | 150/187 [15:10<03:46,  6.11s/it]
Training Epoch: 12/100, batch 300/375 completed (loss: 0.254:  81%|████████  | 151/187 [15:13<03:38,  6.08s/it]
Training Epoch: 12/100, batch 300/375 completed (loss: 0.0296:  81%|████████  | 151/187 [15:13<03:38,  6.08s/it]
Training Epoch: 12/100, batch 300/375 completed (loss: 0.3912:  81%|████████  | 151/187 [15:13<03:38,  6.08s/it]
Training Epoch: 12/100, batch 300/375 completed (loss: 0.0615:  81%|████████  | 151/187 [15:13<03:39,  6.08s/it]
Training Epoch: 12/100, batch 301/375 completed (loss: 0.2149:  81%|████████  | 151/187 [15:13<03:38,  6.08s/it]
Training Epoch: 12/100, batch 301/375 completed (loss: 0.0109:  81%|████████  | 151/187 [15:13<03:39,  6.08s/it]
Training Epoch: 12/100, batch 301/375 completed (loss: 0.0286:  81%|████████  | 151/187 [15:13<03:38,  6.08s/it]
Training Epoch: 12/100, batch 301/375 completed (loss: 0.2363:  81%|████████  | 151/187 [15:13<03:38,  6.08s/it]
Training Epoch: 12/100, batch 302/375 completed (loss: 0.1104:  81%|████████  | 151/187 [15:16<03:38,  6.08s/it]
Training Epoch: 12/100, batch 302/375 completed (loss: 0.0365:  81%|████████  | 151/187 [15:16<03:39,  6.08s/it]
Training Epoch: 12/100, batch 302/375 completed (loss: 0.1041:  81%|████████  | 151/187 [15:16<03:38,  6.08s/it]
Training Epoch: 12/100, batch 302/375 completed (loss: 0.1007:  81%|████████  | 151/187 [15:16<03:38,  6.08s/it]
Training Epoch: 12/100, batch 302/375 completed (loss: 0.1007:  81%|████████▏ | 152/187 [15:20<03:35,  6.15s/it]
Training Epoch: 12/100, batch 302/375 completed (loss: 0.0365:  81%|████████▏ | 152/187 [15:20<03:35,  6.15s/it]
Training Epoch: 12/100, batch 302/375 completed (loss: 0.1041:  81%|████████▏ | 152/187 [15:20<03:35,  6.15s/it]
Training Epoch: 12/100, batch 302/375 completed (loss: 0.1104:  81%|████████▏ | 152/187 [15:20<03:35,  6.15s/it]
Training Epoch: 12/100, batch 303/375 completed (loss: 0.0298:  81%|████████▏ | 152/187 [15:20<03:35,  6.15s/it]
Training Epoch: 12/100, batch 303/375 completed (loss: 0.0299:  81%|████████▏ | 152/187 [15:20<03:35,  6.15s/it]
Training Epoch: 12/100, batch 303/375 completed (loss: 0.1785:  81%|████████▏ | 152/187 [15:20<03:35,  6.15s/it]
Training Epoch: 12/100, batch 303/375 completed (loss: 0.0514:  81%|████████▏ | 152/187 [15:20<03:35,  6.15s/it]
Training Epoch: 12/100, batch 304/375 completed (loss: 0.3123:  81%|████████▏ | 152/187 [15:23<03:35,  6.15s/it]
Training Epoch: 12/100, batch 304/375 completed (loss: 0.1207:  81%|████████▏ | 152/187 [15:23<03:35,  6.15s/it]
Training Epoch: 12/100, batch 304/375 completed (loss: 0.0664:  81%|████████▏ | 152/187 [15:23<03:35,  6.15s/it]
Training Epoch: 12/100, batch 304/375 completed (loss: 0.0402:  81%|████████▏ | 152/187 [15:23<03:35,  6.15s/it]
Training Epoch: 12/100, batch 304/375 completed (loss: 0.0664:  82%|████████▏ | 153/187 [15:25<03:25,  6.05s/it]
Training Epoch: 12/100, batch 304/375 completed (loss: 0.3123:  82%|████████▏ | 153/187 [15:25<03:25,  6.05s/it]
Training Epoch: 12/100, batch 304/375 completed (loss: 0.0402:  82%|████████▏ | 153/187 [15:25<03:25,  6.06s/it]
Training Epoch: 12/100, batch 304/375 completed (loss: 0.1207:  82%|████████▏ | 153/187 [15:25<03:25,  6.06s/it]
Training Epoch: 12/100, batch 305/375 completed (loss: 0.2158:  82%|████████▏ | 153/187 [15:25<03:25,  6.06s/it]
Training Epoch: 12/100, batch 305/375 completed (loss: 0.1168:  82%|████████▏ | 153/187 [15:25<03:25,  6.05s/it]
Training Epoch: 12/100, batch 305/375 completed (loss: 0.4905:  82%|████████▏ | 153/187 [15:25<03:25,  6.05s/it]
Training Epoch: 12/100, batch 305/375 completed (loss: 0.0648:  82%|████████▏ | 153/187 [15:25<03:25,  6.06s/it]
Training Epoch: 12/100, batch 306/375 completed (loss: 0.0301:  82%|████████▏ | 153/187 [15:29<03:25,  6.05s/it]
Training Epoch: 12/100, batch 306/375 completed (loss: 0.1289:  82%|████████▏ | 153/187 [15:29<03:25,  6.05s/it]
Training Epoch: 12/100, batch 306/375 completed (loss: 0.0104:  82%|████████▏ | 153/187 [15:29<03:25,  6.06s/it]
Training Epoch: 12/100, batch 306/375 completed (loss: 0.0537:  82%|████████▏ | 153/187 [15:29<03:25,  6.06s/it]
Training Epoch: 12/100, batch 306/375 completed (loss: 0.0301:  82%|████████▏ | 154/187 [15:31<03:19,  6.05s/it]
Training Epoch: 12/100, batch 306/375 completed (loss: 0.1289:  82%|████████▏ | 154/187 [15:31<03:19,  6.05s/it]
Training Epoch: 12/100, batch 306/375 completed (loss: 0.0537:  82%|████████▏ | 154/187 [15:31<03:19,  6.05s/it]
Training Epoch: 12/100, batch 306/375 completed (loss: 0.0104:  82%|████████▏ | 154/187 [15:31<03:19,  6.05s/it]
Training Epoch: 12/100, batch 307/375 completed (loss: 0.1123:  82%|████████▏ | 154/187 [15:31<03:19,  6.05s/it]
Training Epoch: 12/100, batch 307/375 completed (loss: 0.3458:  82%|████████▏ | 154/187 [15:31<03:19,  6.05s/it]
Training Epoch: 12/100, batch 307/375 completed (loss: 0.0312:  82%|████████▏ | 154/187 [15:31<03:19,  6.05s/it]
Training Epoch: 12/100, batch 307/375 completed (loss: 0.0247:  82%|████████▏ | 154/187 [15:32<03:19,  6.05s/it]
Training Epoch: 12/100, batch 308/375 completed (loss: 0.1167:  82%|████████▏ | 154/187 [15:35<03:19,  6.05s/it]
Training Epoch: 12/100, batch 308/375 completed (loss: 0.1617:  82%|████████▏ | 154/187 [15:35<03:19,  6.05s/it]
Training Epoch: 12/100, batch 308/375 completed (loss: 0.1442:  82%|████████▏ | 154/187 [15:35<03:19,  6.05s/it]
Training Epoch: 12/100, batch 308/375 completed (loss: 0.0736:  82%|████████▏ | 154/187 [15:35<03:19,  6.05s/it]
Training Epoch: 12/100, batch 308/375 completed (loss: 0.0736:  83%|████████▎ | 155/187 [15:38<03:16,  6.13s/it]
Training Epoch: 12/100, batch 308/375 completed (loss: 0.1167:  83%|████████▎ | 155/187 [15:38<03:16,  6.13s/it]
Training Epoch: 12/100, batch 308/375 completed (loss: 0.1617:  83%|████████▎ | 155/187 [15:38<03:16,  6.13s/it]
Training Epoch: 12/100, batch 308/375 completed (loss: 0.1442:  83%|████████▎ | 155/187 [15:38<03:16,  6.13s/it]
Training Epoch: 12/100, batch 309/375 completed (loss: 0.0726:  83%|████████▎ | 155/187 [15:38<03:16,  6.13s/it]
Training Epoch: 12/100, batch 309/375 completed (loss: 0.115:  83%|████████▎ | 155/187 [15:38<03:16,  6.13s/it] 
Training Epoch: 12/100, batch 309/375 completed (loss: 0.3873:  83%|████████▎ | 155/187 [15:38<03:16,  6.13s/it]
Training Epoch: 12/100, batch 309/375 completed (loss: 0.0777:  83%|████████▎ | 155/187 [15:38<03:16,  6.13s/it]
Training Epoch: 12/100, batch 310/375 completed (loss: 0.0504:  83%|████████▎ | 155/187 [15:41<03:16,  6.13s/it]
Training Epoch: 12/100, batch 310/375 completed (loss: 0.1722:  83%|████████▎ | 155/187 [15:41<03:16,  6.13s/it]
Training Epoch: 12/100, batch 310/375 completed (loss: 0.0289:  83%|████████▎ | 155/187 [15:41<03:16,  6.13s/it]
Training Epoch: 12/100, batch 310/375 completed (loss: 0.0028:  83%|████████▎ | 155/187 [15:41<03:16,  6.13s/it]
Training Epoch: 12/100, batch 310/375 completed (loss: 0.0028:  83%|████████▎ | 156/187 [15:44<03:09,  6.11s/it]
Training Epoch: 12/100, batch 310/375 completed (loss: 0.1722:  83%|████████▎ | 156/187 [15:44<03:09,  6.11s/it]
Training Epoch: 12/100, batch 310/375 completed (loss: 0.0504:  83%|████████▎ | 156/187 [15:44<03:09,  6.11s/it]
Training Epoch: 12/100, batch 310/375 completed (loss: 0.0289:  83%|████████▎ | 156/187 [15:44<03:09,  6.11s/it]
Training Epoch: 12/100, batch 311/375 completed (loss: 0.2107:  83%|████████▎ | 156/187 [15:44<03:09,  6.11s/it]
Training Epoch: 12/100, batch 311/375 completed (loss: 0.0721:  83%|████████▎ | 156/187 [15:44<03:09,  6.11s/it]
Training Epoch: 12/100, batch 311/375 completed (loss: 0.3559:  83%|████████▎ | 156/187 [15:44<03:09,  6.11s/it]
Training Epoch: 12/100, batch 311/375 completed (loss: 0.0184:  83%|████████▎ | 156/187 [15:44<03:09,  6.11s/it]
Training Epoch: 12/100, batch 312/375 completed (loss: 0.159:  83%|████████▎ | 156/187 [15:47<03:09,  6.11s/it] 
Training Epoch: 12/100, batch 312/375 completed (loss: 0.1902:  83%|████████▎ | 156/187 [15:47<03:09,  6.11s/it]
Training Epoch: 12/100, batch 312/375 completed (loss: 0.0535:  83%|████████▎ | 156/187 [15:47<03:09,  6.11s/it]
Training Epoch: 12/100, batch 312/375 completed (loss: 0.046:  83%|████████▎ | 156/187 [15:47<03:09,  6.11s/it] 
Training Epoch: 12/100, batch 312/375 completed (loss: 0.046:  84%|████████▍ | 157/187 [15:50<03:03,  6.13s/it]
Training Epoch: 12/100, batch 312/375 completed (loss: 0.159:  84%|████████▍ | 157/187 [15:50<03:03,  6.13s/it]
Training Epoch: 12/100, batch 312/375 completed (loss: 0.1902:  84%|████████▍ | 157/187 [15:50<03:03,  6.13s/it]
Training Epoch: 12/100, batch 312/375 completed (loss: 0.0535:  84%|████████▍ | 157/187 [15:50<03:03,  6.13s/it]
Training Epoch: 12/100, batch 313/375 completed (loss: 0.033:  84%|████████▍ | 157/187 [15:50<03:03,  6.13s/it] 
Training Epoch: 12/100, batch 313/375 completed (loss: 0.0749:  84%|████████▍ | 157/187 [15:50<03:03,  6.13s/it]
Training Epoch: 12/100, batch 313/375 completed (loss: 0.0145:  84%|████████▍ | 157/187 [15:50<03:03,  6.13s/it]
Training Epoch: 12/100, batch 313/375 completed (loss: 0.0936:  84%|████████▍ | 157/187 [15:50<03:03,  6.13s/it]
Training Epoch: 12/100, batch 314/375 completed (loss: 0.2876:  84%|████████▍ | 157/187 [15:53<03:03,  6.13s/it]
Training Epoch: 12/100, batch 314/375 completed (loss: 0.046:  84%|████████▍ | 157/187 [15:53<03:03,  6.13s/it]
Training Epoch: 12/100, batch 314/375 completed (loss: 0.0231:  84%|████████▍ | 157/187 [15:53<03:03,  6.13s/it]
Training Epoch: 12/100, batch 314/375 completed (loss: 0.2654:  84%|████████▍ | 157/187 [15:53<03:03,  6.13s/it]
Training Epoch: 12/100, batch 314/375 completed (loss: 0.2654:  84%|████████▍ | 158/187 [15:56<02:56,  6.10s/it]
Training Epoch: 12/100, batch 314/375 completed (loss: 0.2876:  84%|████████▍ | 158/187 [15:56<02:56,  6.10s/it]
Training Epoch: 12/100, batch 314/375 completed (loss: 0.0231:  84%|████████▍ | 158/187 [15:56<02:56,  6.10s/it]
Training Epoch: 12/100, batch 314/375 completed (loss: 0.046:  84%|████████▍ | 158/187 [15:56<02:57,  6.10s/it]
Training Epoch: 12/100, batch 315/375 completed (loss: 0.0392:  84%|████████▍ | 158/187 [15:56<02:56,  6.10s/it]
Training Epoch: 12/100, batch 315/375 completed (loss: 0.0148:  84%|████████▍ | 158/187 [15:56<02:57,  6.10s/it]
Training Epoch: 12/100, batch 315/375 completed (loss: 0.0062:  84%|████████▍ | 158/187 [15:56<02:56,  6.10s/it]
Training Epoch: 12/100, batch 315/375 completed (loss: 0.2766:  84%|████████▍ | 158/187 [15:56<02:56,  6.10s/it]
Training Epoch: 12/100, batch 316/375 completed (loss: 0.152:  84%|████████▍ | 158/187 [15:59<02:56,  6.10s/it] 
Training Epoch: 12/100, batch 316/375 completed (loss: 0.0465:  84%|████████▍ | 158/187 [15:59<02:57,  6.10s/it]
Training Epoch: 12/100, batch 316/375 completed (loss: 0.384:  84%|████████▍ | 158/187 [15:59<02:56,  6.10s/it] 
Training Epoch: 12/100, batch 316/375 completed (loss: 0.1056:  84%|████████▍ | 158/187 [15:59<02:56,  6.10s/it]
Training Epoch: 12/100, batch 316/375 completed (loss: 0.1056:  85%|████████▌ | 159/187 [16:02<02:50,  6.08s/it]
Training Epoch: 12/100, batch 316/375 completed (loss: 0.0465:  85%|████████▌ | 159/187 [16:02<02:50,  6.08s/it]
Training Epoch: 12/100, batch 316/375 completed (loss: 0.152:  85%|████████▌ | 159/187 [16:02<02:50,  6.08s/it]
Training Epoch: 12/100, batch 316/375 completed (loss: 0.384:  85%|████████▌ | 159/187 [16:02<02:50,  6.08s/it]
Training Epoch: 12/100, batch 317/375 completed (loss: 0.1829:  85%|████████▌ | 159/187 [16:02<02:50,  6.08s/it]
Training Epoch: 12/100, batch 317/375 completed (loss: 0.0473:  85%|████████▌ | 159/187 [16:02<02:50,  6.08s/it]
Training Epoch: 12/100, batch 317/375 completed (loss: 0.0604:  85%|████████▌ | 159/187 [16:02<02:50,  6.08s/it]
Training Epoch: 12/100, batch 317/375 completed (loss: 0.0199:  85%|████████▌ | 159/187 [16:02<02:50,  6.08s/it]
Training Epoch: 12/100, batch 318/375 completed (loss: 0.1862:  85%|████████▌ | 159/187 [16:05<02:50,  6.08s/it]
Training Epoch: 12/100, batch 318/375 completed (loss: 0.0614:  85%|████████▌ | 159/187 [16:05<02:50,  6.08s/it]
Training Epoch: 12/100, batch 318/375 completed (loss: 0.1563:  85%|████████▌ | 159/187 [16:05<02:50,  6.08s/it]
Training Epoch: 12/100, batch 318/375 completed (loss: 0.0308:  85%|████████▌ | 159/187 [16:05<02:50,  6.08s/it]
Training Epoch: 12/100, batch 318/375 completed (loss: 0.0308:  86%|████████▌ | 160/187 [16:08<02:46,  6.16s/it]
Training Epoch: 12/100, batch 318/375 completed (loss: 0.0614:  86%|████████▌ | 160/187 [16:08<02:46,  6.16s/it]
Training Epoch: 12/100, batch 318/375 completed (loss: 0.1563:  86%|████████▌ | 160/187 [16:08<02:46,  6.17s/it]
Training Epoch: 12/100, batch 318/375 completed (loss: 0.1862:  86%|████████▌ | 160/187 [16:08<02:46,  6.17s/it]
Training Epoch: 12/100, batch 319/375 completed (loss: 0.0189:  86%|████████▌ | 160/187 [16:08<02:46,  6.16s/it]
Training Epoch: 12/100, batch 319/375 completed (loss: 0.0204:  86%|████████▌ | 160/187 [16:08<02:46,  6.17s/it]
Training Epoch: 12/100, batch 319/375 completed (loss: 0.3423:  86%|████████▌ | 160/187 [16:08<02:46,  6.17s/it]
Training Epoch: 12/100, batch 319/375 completed (loss: 0.0583:  86%|████████▌ | 160/187 [16:08<02:46,  6.16s/it]
Training Epoch: 12/100, batch 320/375 completed (loss: 0.1453:  86%|████████▌ | 160/187 [16:11<02:46,  6.17s/it]
Training Epoch: 12/100, batch 320/375 completed (loss: 0.5724:  86%|████████▌ | 160/187 [16:11<02:46,  6.17s/it]
Training Epoch: 12/100, batch 320/375 completed (loss: 0.2475:  86%|████████▌ | 160/187 [16:11<02:46,  6.16s/it]
Training Epoch: 12/100, batch 320/375 completed (loss: 0.0476:  86%|████████▌ | 160/187 [16:11<02:46,  6.16s/it]
Training Epoch: 12/100, batch 320/375 completed (loss: 0.0476:  86%|████████▌ | 161/187 [16:14<02:37,  6.07s/it]
Training Epoch: 12/100, batch 320/375 completed (loss: 0.1453:  86%|████████▌ | 161/187 [16:14<02:37,  6.07s/it]
Training Epoch: 12/100, batch 320/375 completed (loss: 0.2475:  86%|████████▌ | 161/187 [16:14<02:37,  6.07s/it]
Training Epoch: 12/100, batch 320/375 completed (loss: 0.5724:  86%|████████▌ | 161/187 [16:14<02:37,  6.07s/it]
Training Epoch: 12/100, batch 321/375 completed (loss: 0.0262:  86%|████████▌ | 161/187 [16:14<02:37,  6.07s/it]
Training Epoch: 12/100, batch 321/375 completed (loss: 0.1256:  86%|████████▌ | 161/187 [16:14<02:37,  6.07s/it]
Training Epoch: 12/100, batch 321/375 completed (loss: 0.0972:  86%|████████▌ | 161/187 [16:14<02:37,  6.07s/it]
Training Epoch: 12/100, batch 321/375 completed (loss: 0.1042:  86%|████████▌ | 161/187 [16:14<02:37,  6.07s/it]
Training Epoch: 12/100, batch 322/375 completed (loss: 0.0602:  86%|████████▌ | 161/187 [16:17<02:37,  6.07s/it]
Training Epoch: 12/100, batch 322/375 completed (loss: 0.1252:  86%|████████▌ | 161/187 [16:17<02:37,  6.07s/it]
Training Epoch: 12/100, batch 322/375 completed (loss: 0.3232:  86%|████████▌ | 161/187 [16:17<02:37,  6.07s/it]
Training Epoch: 12/100, batch 322/375 completed (loss: 0.1091:  86%|████████▌ | 161/187 [16:17<02:37,  6.07s/it]
Training Epoch: 12/100, batch 322/375 completed (loss: 0.1252:  87%|████████▋ | 162/187 [16:20<02:30,  6.04s/it]
Training Epoch: 12/100, batch 322/375 completed (loss: 0.3232:  87%|████████▋ | 162/187 [16:20<02:30,  6.04s/it]
Training Epoch: 12/100, batch 322/375 completed (loss: 0.0602:  87%|████████▋ | 162/187 [16:20<02:30,  6.04s/it]
Training Epoch: 12/100, batch 322/375 completed (loss: 0.1091:  87%|████████▋ | 162/187 [16:20<02:31,  6.04s/it]
Training Epoch: 12/100, batch 323/375 completed (loss: 0.0439:  87%|████████▋ | 162/187 [16:20<02:30,  6.04s/it]
Training Epoch: 12/100, batch 323/375 completed (loss: 0.069:  87%|████████▋ | 162/187 [16:20<02:30,  6.04s/it] 
Training Epoch: 12/100, batch 323/375 completed (loss: 0.2172:  87%|████████▋ | 162/187 [16:20<02:30,  6.04s/it]
Training Epoch: 12/100, batch 323/375 completed (loss: 0.0667:  87%|████████▋ | 162/187 [16:20<02:31,  6.04s/it]
Training Epoch: 12/100, batch 324/375 completed (loss: 0.1979:  87%|████████▋ | 162/187 [16:24<02:30,  6.04s/it]
Training Epoch: 12/100, batch 324/375 completed (loss: 0.0282:  87%|████████▋ | 162/187 [16:24<02:30,  6.04s/it]
Training Epoch: 12/100, batch 324/375 completed (loss: 0.0109:  87%|████████▋ | 162/187 [16:24<02:30,  6.04s/it]
Training Epoch: 12/100, batch 324/375 completed (loss: 0.1453:  87%|████████▋ | 162/187 [16:24<02:31,  6.04s/it]
Training Epoch: 12/100, batch 324/375 completed (loss: 0.1453:  87%|████████▋ | 163/187 [16:27<02:27,  6.13s/it]
Training Epoch: 12/100, batch 324/375 completed (loss: 0.0109:  87%|████████▋ | 163/187 [16:27<02:27,  6.13s/it]
Training Epoch: 12/100, batch 324/375 completed (loss: 0.1979:  87%|████████▋ | 163/187 [16:27<02:27,  6.13s/it]
Training Epoch: 12/100, batch 324/375 completed (loss: 0.0282:  87%|████████▋ | 163/187 [16:27<02:27,  6.13s/it]
Training Epoch: 12/100, batch 325/375 completed (loss: 0.3275:  87%|████████▋ | 163/187 [16:27<02:27,  6.13s/it]
Training Epoch: 12/100, batch 325/375 completed (loss: 0.1996:  87%|████████▋ | 163/187 [16:27<02:27,  6.13s/it]
Training Epoch: 12/100, batch 325/375 completed (loss: 0.089:  87%|████████▋ | 163/187 [16:27<02:27,  6.13s/it] 
Training Epoch: 12/100, batch 325/375 completed (loss: 0.0167:  87%|████████▋ | 163/187 [16:27<02:27,  6.13s/it]
Training Epoch: 12/100, batch 326/375 completed (loss: 0.1208:  87%|████████▋ | 163/187 [16:30<02:27,  6.13s/it]
Training Epoch: 12/100, batch 326/375 completed (loss: 0.1481:  87%|████████▋ | 163/187 [16:30<02:27,  6.13s/it]
Training Epoch: 12/100, batch 326/375 completed (loss: 0.243:  87%|████████▋ | 163/187 [16:30<02:27,  6.13s/it] 
Training Epoch: 12/100, batch 326/375 completed (loss: 0.071:  87%|████████▋ | 163/187 [16:30<02:27,  6.13s/it] 
Training Epoch: 12/100, batch 326/375 completed (loss: 0.1208:  88%|████████▊ | 164/187 [16:33<02:20,  6.09s/it]
Training Epoch: 12/100, batch 326/375 completed (loss: 0.243:  88%|████████▊ | 164/187 [16:33<02:20,  6.09s/it]
Training Epoch: 12/100, batch 326/375 completed (loss: 0.1481:  88%|████████▊ | 164/187 [16:33<02:20,  6.09s/it]
Training Epoch: 12/100, batch 326/375 completed (loss: 0.071:  88%|████████▊ | 164/187 [16:33<02:20,  6.09s/it]
Training Epoch: 12/100, batch 327/375 completed (loss: 0.0466:  88%|████████▊ | 164/187 [16:33<02:20,  6.09s/it]
Training Epoch: 12/100, batch 327/375 completed (loss: 0.0292:  88%|████████▊ | 164/187 [16:33<02:20,  6.09s/it]
Training Epoch: 12/100, batch 327/375 completed (loss: 0.0699:  88%|████████▊ | 164/187 [16:33<02:20,  6.09s/it]
Training Epoch: 12/100, batch 327/375 completed (loss: 0.0866:  88%|████████▊ | 164/187 [16:33<02:20,  6.09s/it]
Training Epoch: 12/100, batch 328/375 completed (loss: 0.1468:  88%|████████▊ | 164/187 [16:36<02:20,  6.09s/it]
Training Epoch: 12/100, batch 328/375 completed (loss: 0.1326:  88%|████████▊ | 164/187 [16:36<02:20,  6.09s/it]
Training Epoch: 12/100, batch 328/375 completed (loss: 0.0096:  88%|████████▊ | 164/187 [16:36<02:20,  6.09s/it]
Training Epoch: 12/100, batch 328/375 completed (loss: 0.0921:  88%|████████▊ | 164/187 [16:36<02:20,  6.09s/it]
Training Epoch: 12/100, batch 328/375 completed (loss: 0.1326:  88%|████████▊ | 165/187 [16:39<02:14,  6.11s/it]
Training Epoch: 12/100, batch 328/375 completed (loss: 0.0096:  88%|████████▊ | 165/187 [16:39<02:14,  6.12s/it]
Training Epoch: 12/100, batch 328/375 completed (loss: 0.0921:  88%|████████▊ | 165/187 [16:39<02:14,  6.12s/it]
Training Epoch: 12/100, batch 328/375 completed (loss: 0.1468:  88%|████████▊ | 165/187 [16:39<02:14,  6.11s/it]
Training Epoch: 12/100, batch 329/375 completed (loss: 0.0158:  88%|████████▊ | 165/187 [16:39<02:14,  6.11s/it]
Training Epoch: 12/100, batch 329/375 completed (loss: 0.0295:  88%|████████▊ | 165/187 [16:39<02:14,  6.11s/it]
Training Epoch: 12/100, batch 329/375 completed (loss: 0.2183:  88%|████████▊ | 165/187 [16:39<02:14,  6.12s/it]
Training Epoch: 12/100, batch 329/375 completed (loss: 0.0198:  88%|████████▊ | 165/187 [16:39<02:14,  6.12s/it]
Training Epoch: 12/100, batch 330/375 completed (loss: 0.0651:  88%|████████▊ | 165/187 [16:42<02:14,  6.11s/it]
Training Epoch: 12/100, batch 330/375 completed (loss: 0.2861:  88%|████████▊ | 165/187 [16:42<02:14,  6.11s/it]
Training Epoch: 12/100, batch 330/375 completed (loss: 0.0332:  88%|████████▊ | 165/187 [16:42<02:14,  6.12s/it]
Training Epoch: 12/100, batch 330/375 completed (loss: 0.02:  88%|████████▊ | 165/187 [16:42<02:14,  6.12s/it]  
Training Epoch: 12/100, batch 330/375 completed (loss: 0.0651:  89%|████████▉ | 166/187 [16:45<02:07,  6.08s/it]
Training Epoch: 12/100, batch 330/375 completed (loss: 0.0332:  89%|████████▉ | 166/187 [16:45<02:07,  6.08s/it]
Training Epoch: 12/100, batch 330/375 completed (loss: 0.2861:  89%|████████▉ | 166/187 [16:45<02:07,  6.08s/it]
Training Epoch: 12/100, batch 330/375 completed (loss: 0.02:  89%|████████▉ | 166/187 [16:45<02:07,  6.09s/it]
Training Epoch: 12/100, batch 331/375 completed (loss: 0.1688:  89%|████████▉ | 166/187 [16:45<02:07,  6.08s/it]
Training Epoch: 12/100, batch 331/375 completed (loss: 0.0971:  89%|████████▉ | 166/187 [16:45<02:07,  6.08s/it]
Training Epoch: 12/100, batch 331/375 completed (loss: 0.1518:  89%|████████▉ | 166/187 [16:45<02:07,  6.08s/it]
Training Epoch: 12/100, batch 331/375 completed (loss: 0.0964:  89%|████████▉ | 166/187 [16:45<02:07,  6.09s/it]
Training Epoch: 12/100, batch 332/375 completed (loss: 0.1301:  89%|████████▉ | 166/187 [16:48<02:07,  6.08s/it]
Training Epoch: 12/100, batch 332/375 completed (loss: 0.1703:  89%|████████▉ | 166/187 [16:48<02:07,  6.08s/it]
Training Epoch: 12/100, batch 332/375 completed (loss: 0.2815:  89%|████████▉ | 166/187 [16:48<02:07,  6.08s/it]
Training Epoch: 12/100, batch 332/375 completed (loss: 0.2322:  89%|████████▉ | 166/187 [16:48<02:07,  6.09s/it]
Training Epoch: 12/100, batch 332/375 completed (loss: 0.1703:  89%|████████▉ | 167/187 [16:51<02:01,  6.05s/it]
Training Epoch: 12/100, batch 332/375 completed (loss: 0.1301:  89%|████████▉ | 167/187 [16:51<02:01,  6.06s/it]
Training Epoch: 12/100, batch 332/375 completed (loss: 0.2815:  89%|████████▉ | 167/187 [16:51<02:01,  6.06s/it]
Training Epoch: 12/100, batch 332/375 completed (loss: 0.2322:  89%|████████▉ | 167/187 [16:51<02:01,  6.06s/it]
Training Epoch: 12/100, batch 333/375 completed (loss: 0.0674:  89%|████████▉ | 167/187 [16:51<02:01,  6.06s/it]
Training Epoch: 12/100, batch 333/375 completed (loss: 0.184:  89%|████████▉ | 167/187 [16:51<02:01,  6.06s/it] 
Training Epoch: 12/100, batch 333/375 completed (loss: 0.2913:  89%|████████▉ | 167/187 [16:51<02:01,  6.05s/it]
Training Epoch: 12/100, batch 333/375 completed (loss: 0.0205:  89%|████████▉ | 167/187 [16:51<02:01,  6.06s/it]
Training Epoch: 12/100, batch 334/375 completed (loss: 0.5503:  89%|████████▉ | 167/187 [16:54<02:01,  6.06s/it]
Training Epoch: 12/100, batch 334/375 completed (loss: 0.0449:  89%|████████▉ | 167/187 [16:54<02:01,  6.06s/it]
Training Epoch: 12/100, batch 334/375 completed (loss: 0.4922:  89%|████████▉ | 167/187 [16:54<02:01,  6.05s/it]
Training Epoch: 12/100, batch 334/375 completed (loss: 0.0278:  89%|████████▉ | 167/187 [16:54<02:01,  6.06s/it]
Training Epoch: 12/100, batch 334/375 completed (loss: 0.0449:  90%|████████▉ | 168/187 [16:57<01:55,  6.10s/it]
Training Epoch: 12/100, batch 334/375 completed (loss: 0.5503:  90%|████████▉ | 168/187 [16:57<01:55,  6.10s/it]
Training Epoch: 12/100, batch 334/375 completed (loss: 0.4922:  90%|████████▉ | 168/187 [16:57<01:55,  6.10s/it]
Training Epoch: 12/100, batch 334/375 completed (loss: 0.0278:  90%|████████▉ | 168/187 [16:57<01:55,  6.10s/it]
Training Epoch: 12/100, batch 335/375 completed (loss: 0.0359:  90%|████████▉ | 168/187 [16:57<01:55,  6.10s/it]
Training Epoch: 12/100, batch 335/375 completed (loss: 0.0886:  90%|████████▉ | 168/187 [16:57<01:55,  6.10s/it]
Training Epoch: 12/100, batch 335/375 completed (loss: 0.0356:  90%|████████▉ | 168/187 [16:57<01:55,  6.10s/it]
Training Epoch: 12/100, batch 335/375 completed (loss: 0.2765:  90%|████████▉ | 168/187 [16:57<01:55,  6.10s/it]
Training Epoch: 12/100, batch 336/375 completed (loss: 0.0917:  90%|████████▉ | 168/187 [17:00<01:55,  6.10s/it]
Training Epoch: 12/100, batch 336/375 completed (loss: 0.0655:  90%|████████▉ | 168/187 [17:00<01:55,  6.10s/it]
Training Epoch: 12/100, batch 336/375 completed (loss: 0.1286:  90%|████████▉ | 168/187 [17:00<01:55,  6.10s/it]
Training Epoch: 12/100, batch 336/375 completed (loss: 0.0713:  90%|████████▉ | 168/187 [17:00<01:55,  6.10s/it]
Training Epoch: 12/100, batch 336/375 completed (loss: 0.1286:  90%|█████████ | 169/187 [17:03<01:49,  6.08s/it]
Training Epoch: 12/100, batch 336/375 completed (loss: 0.0655:  90%|█████████ | 169/187 [17:03<01:49,  6.07s/it]
Training Epoch: 12/100, batch 336/375 completed (loss: 0.0917:  90%|█████████ | 169/187 [17:03<01:49,  6.08s/it]
Training Epoch: 12/100, batch 336/375 completed (loss: 0.0713:  90%|█████████ | 169/187 [17:03<01:49,  6.07s/it]
Training Epoch: 12/100, batch 337/375 completed (loss: 0.148:  90%|█████████ | 169/187 [17:03<01:49,  6.07s/it] 
Training Epoch: 12/100, batch 337/375 completed (loss: 0.1736:  90%|█████████ | 169/187 [17:03<01:49,  6.08s/it]
Training Epoch: 12/100, batch 337/375 completed (loss: 0.2805:  90%|█████████ | 169/187 [17:03<01:49,  6.08s/it]
Training Epoch: 12/100, batch 337/375 completed (loss: 0.1565:  90%|█████████ | 169/187 [17:03<01:49,  6.07s/it]
Training Epoch: 12/100, batch 338/375 completed (loss: 0.0612:  90%|█████████ | 169/187 [17:06<01:49,  6.08s/it]
Training Epoch: 12/100, batch 338/375 completed (loss: 0.0756:  90%|█████████ | 169/187 [17:06<01:49,  6.07s/it]
Training Epoch: 12/100, batch 338/375 completed (loss: 0.0508:  90%|█████████ | 169/187 [17:06<01:49,  6.08s/it]
Training Epoch: 12/100, batch 338/375 completed (loss: 0.0497:  90%|█████████ | 169/187 [17:06<01:49,  6.07s/it]
Training Epoch: 12/100, batch 338/375 completed (loss: 0.0508:  91%|█████████ | 170/187 [17:09<01:42,  6.01s/it]
Training Epoch: 12/100, batch 338/375 completed (loss: 0.0756:  91%|█████████ | 170/187 [17:09<01:42,  6.01s/it]
Training Epoch: 12/100, batch 338/375 completed (loss: 0.0612:  91%|█████████ | 170/187 [17:09<01:42,  6.01s/it]
Training Epoch: 12/100, batch 338/375 completed (loss: 0.0497:  91%|█████████ | 170/187 [17:09<01:42,  6.01s/it]
Training Epoch: 12/100, batch 339/375 completed (loss: 0.0816:  91%|█████████ | 170/187 [17:09<01:42,  6.01s/it]
Training Epoch: 12/100, batch 339/375 completed (loss: 0.1554:  91%|█████████ | 170/187 [17:09<01:42,  6.01s/it]
Training Epoch: 12/100, batch 339/375 completed (loss: 0.0923:  91%|█████████ | 170/187 [17:09<01:42,  6.01s/it]
Training Epoch: 12/100, batch 339/375 completed (loss: 0.3265:  91%|█████████ | 170/187 [17:09<01:42,  6.01s/it]
Training Epoch: 12/100, batch 340/375 completed (loss: 0.1575:  91%|█████████ | 170/187 [17:12<01:42,  6.01s/it]
Training Epoch: 12/100, batch 340/375 completed (loss: 0.4902:  91%|█████████ | 170/187 [17:12<01:42,  6.01s/it]
Training Epoch: 12/100, batch 340/375 completed (loss: 0.0584:  91%|█████████ | 170/187 [17:12<01:42,  6.01s/it]
Training Epoch: 12/100, batch 340/375 completed (loss: 0.0901:  91%|█████████ | 170/187 [17:12<01:42,  6.01s/it]
Training Epoch: 12/100, batch 340/375 completed (loss: 0.1575:  91%|█████████▏| 171/187 [17:15<01:37,  6.10s/it]
Training Epoch: 12/100, batch 340/375 completed (loss: 0.4902:  91%|█████████▏| 171/187 [17:15<01:37,  6.09s/it]
Training Epoch: 12/100, batch 340/375 completed (loss: 0.0584:  91%|█████████▏| 171/187 [17:15<01:37,  6.10s/it]
Training Epoch: 12/100, batch 340/375 completed (loss: 0.0901:  91%|█████████▏| 171/187 [17:15<01:37,  6.10s/it]
Training Epoch: 12/100, batch 341/375 completed (loss: 0.3151:  91%|█████████▏| 171/187 [17:15<01:37,  6.09s/it]
Training Epoch: 12/100, batch 341/375 completed (loss: 0.1295:  91%|█████████▏| 171/187 [17:15<01:37,  6.10s/it]
Training Epoch: 12/100, batch 341/375 completed (loss: 0.0306:  91%|█████████▏| 171/187 [17:15<01:37,  6.10s/it]
Training Epoch: 12/100, batch 341/375 completed (loss: 0.1284:  91%|█████████▏| 171/187 [17:15<01:37,  6.10s/it]
Training Epoch: 12/100, batch 342/375 completed (loss: 0.1808:  91%|█████████▏| 171/187 [17:18<01:37,  6.10s/it]
Training Epoch: 12/100, batch 342/375 completed (loss: 0.076:  91%|█████████▏| 171/187 [17:18<01:37,  6.09s/it] 
Training Epoch: 12/100, batch 342/375 completed (loss: 0.1812:  91%|█████████▏| 171/187 [17:18<01:37,  6.10s/it]
Training Epoch: 12/100, batch 342/375 completed (loss: 0.0812:  91%|█████████▏| 171/187 [17:18<01:37,  6.10s/it]
Training Epoch: 12/100, batch 342/375 completed (loss: 0.076:  92%|█████████▏| 172/187 [17:21<01:30,  6.06s/it]
Training Epoch: 12/100, batch 342/375 completed (loss: 0.1808:  92%|█████████▏| 172/187 [17:21<01:30,  6.06s/it]
Training Epoch: 12/100, batch 342/375 completed (loss: 0.1812:  92%|█████████▏| 172/187 [17:21<01:30,  6.07s/it]
Training Epoch: 12/100, batch 342/375 completed (loss: 0.0812:  92%|█████████▏| 172/187 [17:21<01:30,  6.07s/it]
Training Epoch: 12/100, batch 343/375 completed (loss: 0.1009:  92%|█████████▏| 172/187 [17:21<01:30,  6.07s/it]
Training Epoch: 12/100, batch 343/375 completed (loss: 0.0347:  92%|█████████▏| 172/187 [17:21<01:30,  6.06s/it]
Training Epoch: 12/100, batch 343/375 completed (loss: 0.4022:  92%|█████████▏| 172/187 [17:21<01:30,  6.06s/it]
Training Epoch: 12/100, batch 343/375 completed (loss: 0.031:  92%|█████████▏| 172/187 [17:21<01:30,  6.07s/it] 
Training Epoch: 12/100, batch 344/375 completed (loss: 0.1306:  92%|█████████▏| 172/187 [17:24<01:30,  6.06s/it]
Training Epoch: 12/100, batch 344/375 completed (loss: 0.3552:  92%|█████████▏| 172/187 [17:24<01:30,  6.07s/it]
Training Epoch: 12/100, batch 344/375 completed (loss: 0.0693:  92%|█████████▏| 172/187 [17:24<01:30,  6.06s/it]
Training Epoch: 12/100, batch 344/375 completed (loss: 0.086:  92%|█████████▏| 172/187 [17:24<01:30,  6.07s/it]
Training Epoch: 12/100, batch 344/375 completed (loss: 0.3552:  93%|█████████▎| 173/187 [17:27<01:25,  6.09s/it]
Training Epoch: 12/100, batch 344/375 completed (loss: 0.086:  93%|█████████▎| 173/187 [17:27<01:25,  6.09s/it]
Training Epoch: 12/100, batch 344/375 completed (loss: 0.1306:  93%|█████████▎| 173/187 [17:27<01:25,  6.09s/it]
Training Epoch: 12/100, batch 344/375 completed (loss: 0.0693:  93%|█████████▎| 173/187 [17:27<01:25,  6.09s/it]
Training Epoch: 12/100, batch 345/375 completed (loss: 0.0287:  93%|█████████▎| 173/187 [17:27<01:25,  6.09s/it]
Training Epoch: 12/100, batch 345/375 completed (loss: 0.051:  93%|█████████▎| 173/187 [17:27<01:25,  6.09s/it] 
Training Epoch: 12/100, batch 345/375 completed (loss: 0.0513:  93%|█████████▎| 173/187 [17:27<01:25,  6.09s/it]
Training Epoch: 12/100, batch 345/375 completed (loss: 0.0142:  93%|█████████▎| 173/187 [17:27<01:25,  6.09s/it]
Training Epoch: 12/100, batch 346/375 completed (loss: 0.0159:  93%|█████████▎| 173/187 [17:30<01:25,  6.09s/it]
Training Epoch: 12/100, batch 346/375 completed (loss: 0.0795:  93%|█████████▎| 173/187 [17:30<01:25,  6.09s/it]
Training Epoch: 12/100, batch 346/375 completed (loss: 0.3643:  93%|█████████▎| 173/187 [17:30<01:25,  6.09s/it]
Training Epoch: 12/100, batch 346/375 completed (loss: 0.0802:  93%|█████████▎| 173/187 [17:30<01:25,  6.09s/it]
Training Epoch: 12/100, batch 346/375 completed (loss: 0.3643:  93%|█████████▎| 174/187 [17:33<01:19,  6.12s/it]
Training Epoch: 12/100, batch 346/375 completed (loss: 0.0802:  93%|█████████▎| 174/187 [17:33<01:19,  6.12s/it]
Training Epoch: 12/100, batch 346/375 completed (loss: 0.0795:  93%|█████████▎| 174/187 [17:33<01:19,  6.12s/it]
Training Epoch: 12/100, batch 346/375 completed (loss: 0.0159:  93%|█████████▎| 174/187 [17:33<01:19,  6.12s/it]
Training Epoch: 12/100, batch 347/375 completed (loss: 0.1587:  93%|█████████▎| 174/187 [17:34<01:19,  6.12s/it]
Training Epoch: 12/100, batch 347/375 completed (loss: 0.1227:  93%|█████████▎| 174/187 [17:33<01:19,  6.12s/it]
Training Epoch: 12/100, batch 347/375 completed (loss: 0.0892:  93%|█████████▎| 174/187 [17:34<01:19,  6.12s/it]
Training Epoch: 12/100, batch 347/375 completed (loss: 0.2981:  93%|█████████▎| 174/187 [17:34<01:19,  6.12s/it]
Training Epoch: 12/100, batch 348/375 completed (loss: 0.0938:  93%|█████████▎| 174/187 [17:36<01:19,  6.12s/it]
Training Epoch: 12/100, batch 348/375 completed (loss: 0.6745:  93%|█████████▎| 174/187 [17:36<01:19,  6.12s/it]
Training Epoch: 12/100, batch 348/375 completed (loss: 0.018:  93%|█████████▎| 174/187 [17:36<01:19,  6.12s/it] 
Training Epoch: 12/100, batch 348/375 completed (loss: 0.3209:  93%|█████████▎| 174/187 [17:36<01:19,  6.12s/it]
Training Epoch: 12/100, batch 348/375 completed (loss: 0.0938:  94%|█████████▎| 175/187 [17:39<01:12,  6.04s/it]
Training Epoch: 12/100, batch 348/375 completed (loss: 0.6745:  94%|█████████▎| 175/187 [17:39<01:12,  6.04s/it]
Training Epoch: 12/100, batch 348/375 completed (loss: 0.018:  94%|█████████▎| 175/187 [17:39<01:12,  6.05s/it]
Training Epoch: 12/100, batch 348/375 completed (loss: 0.3209:  94%|█████████▎| 175/187 [17:39<01:12,  6.05s/it]
Training Epoch: 12/100, batch 349/375 completed (loss: 0.0289:  94%|█████████▎| 175/187 [17:39<01:12,  6.05s/it]
Training Epoch: 12/100, batch 349/375 completed (loss: 0.4118:  94%|█████████▎| 175/187 [17:39<01:12,  6.04s/it]
Training Epoch: 12/100, batch 349/375 completed (loss: 0.0337:  94%|█████████▎| 175/187 [17:39<01:12,  6.04s/it]
Training Epoch: 12/100, batch 349/375 completed (loss: 0.1793:  94%|█████████▎| 175/187 [17:39<01:12,  6.05s/it]
Training Epoch: 12/100, batch 350/375 completed (loss: 0.0364:  94%|█████████▎| 175/187 [17:43<01:12,  6.04s/it]
Training Epoch: 12/100, batch 350/375 completed (loss: 0.2024:  94%|█████████▎| 175/187 [17:43<01:12,  6.05s/it]
Training Epoch: 12/100, batch 350/375 completed (loss: 0.1914:  94%|█████████▎| 175/187 [17:43<01:12,  6.04s/it]
Training Epoch: 12/100, batch 350/375 completed (loss: 0.439:  94%|█████████▎| 175/187 [17:43<01:12,  6.05s/it] 
Training Epoch: 12/100, batch 350/375 completed (loss: 0.0364:  94%|█████████▍| 176/187 [17:46<01:07,  6.12s/it]
Training Epoch: 12/100, batch 350/375 completed (loss: 0.1914:  94%|█████████▍| 176/187 [17:46<01:07,  6.12s/it]
Training Epoch: 12/100, batch 350/375 completed (loss: 0.2024:  94%|█████████▍| 176/187 [17:46<01:07,  6.12s/it]
Training Epoch: 12/100, batch 350/375 completed (loss: 0.439:  94%|█████████▍| 176/187 [17:46<01:07,  6.12s/it]
Training Epoch: 12/100, batch 351/375 completed (loss: 0.0868:  94%|█████████▍| 176/187 [17:46<01:07,  6.12s/it]
Training Epoch: 12/100, batch 351/375 completed (loss: 0.0545:  94%|█████████▍| 176/187 [17:46<01:07,  6.12s/it]
Training Epoch: 12/100, batch 351/375 completed (loss: 0.0289:  94%|█████████▍| 176/187 [17:46<01:07,  6.12s/it]
Training Epoch: 12/100, batch 351/375 completed (loss: 0.2347:  94%|█████████▍| 176/187 [17:46<01:07,  6.12s/it]
Training Epoch: 12/100, batch 352/375 completed (loss: 0.1047:  94%|█████████▍| 176/187 [17:49<01:07,  6.12s/it]
Training Epoch: 12/100, batch 352/375 completed (loss: 0.0318:  94%|█████████▍| 176/187 [17:49<01:07,  6.12s/it]
Training Epoch: 12/100, batch 352/375 completed (loss: 0.0461:  94%|█████████▍| 176/187 [17:49<01:07,  6.12s/it]
Training Epoch: 12/100, batch 352/375 completed (loss: 0.2543:  94%|█████████▍| 176/187 [17:49<01:07,  6.12s/it]
Training Epoch: 12/100, batch 352/375 completed (loss: 0.1047:  95%|█████████▍| 177/187 [17:52<01:00,  6.08s/it]
Training Epoch: 12/100, batch 352/375 completed (loss: 0.0461:  95%|█████████▍| 177/187 [17:52<01:00,  6.08s/it]
Training Epoch: 12/100, batch 352/375 completed (loss: 0.2543:  95%|█████████▍| 177/187 [17:52<01:00,  6.08s/it]
Training Epoch: 12/100, batch 352/375 completed (loss: 0.0318:  95%|█████████▍| 177/187 [17:52<01:00,  6.08s/it]
Training Epoch: 12/100, batch 353/375 completed (loss: 0.036:  95%|█████████▍| 177/187 [17:52<01:00,  6.08s/it] 
Training Epoch: 12/100, batch 353/375 completed (loss: 0.1041:  95%|█████████▍| 177/187 [17:52<01:00,  6.08s/it]
Training Epoch: 12/100, batch 353/375 completed (loss: 0.0231:  95%|█████████▍| 177/187 [17:52<01:00,  6.08s/it]
Training Epoch: 12/100, batch 353/375 completed (loss: 0.1091:  95%|█████████▍| 177/187 [17:52<01:00,  6.08s/it]
Training Epoch: 12/100, batch 354/375 completed (loss: 0.2095:  95%|█████████▍| 177/187 [17:55<01:00,  6.08s/it]
Training Epoch: 12/100, batch 354/375 completed (loss: 0.2797:  95%|█████████▍| 177/187 [17:55<01:00,  6.08s/it]
Training Epoch: 12/100, batch 354/375 completed (loss: 0.0313:  95%|█████████▍| 177/187 [17:55<01:00,  6.08s/it]
Training Epoch: 12/100, batch 354/375 completed (loss: 0.0625:  95%|█████████▍| 177/187 [17:55<01:00,  6.08s/it]
Training Epoch: 12/100, batch 354/375 completed (loss: 0.0313:  95%|█████████▌| 178/187 [17:58<00:54,  6.05s/it]
Training Epoch: 12/100, batch 354/375 completed (loss: 0.2797:  95%|█████████▌| 178/187 [17:58<00:54,  6.05s/it]
Training Epoch: 12/100, batch 354/375 completed (loss: 0.2095:  95%|█████████▌| 178/187 [17:58<00:54,  6.05s/it]
Training Epoch: 12/100, batch 354/375 completed (loss: 0.0625:  95%|█████████▌| 178/187 [17:58<00:54,  6.05s/it]
Training Epoch: 12/100, batch 355/375 completed (loss: 0.1127:  95%|█████████▌| 178/187 [17:58<00:54,  6.05s/it]
Training Epoch: 12/100, batch 355/375 completed (loss: 0.009:  95%|█████████▌| 178/187 [17:58<00:54,  6.05s/it] 
Training Epoch: 12/100, batch 355/375 completed (loss: 0.0817:  95%|█████████▌| 178/187 [17:58<00:54,  6.05s/it]
Training Epoch: 12/100, batch 355/375 completed (loss: 0.0602:  95%|█████████▌| 178/187 [17:58<00:54,  6.05s/it]
Training Epoch: 12/100, batch 356/375 completed (loss: 0.2135:  95%|█████████▌| 178/187 [18:01<00:54,  6.05s/it]
Training Epoch: 12/100, batch 356/375 completed (loss: 0.2781:  95%|█████████▌| 178/187 [18:01<00:54,  6.05s/it]
Training Epoch: 12/100, batch 356/375 completed (loss: 0.3878:  95%|█████████▌| 178/187 [18:01<00:54,  6.05s/it]
Training Epoch: 12/100, batch 356/375 completed (loss: 0.1845:  95%|█████████▌| 178/187 [18:01<00:54,  6.05s/it]
Training Epoch: 12/100, batch 356/375 completed (loss: 0.1845:  96%|█████████▌| 179/187 [18:04<00:48,  6.09s/it]
Training Epoch: 12/100, batch 356/375 completed (loss: 0.2781:  96%|█████████▌| 179/187 [18:04<00:48,  6.09s/it]
Training Epoch: 12/100, batch 356/375 completed (loss: 0.2135:  96%|█████████▌| 179/187 [18:04<00:48,  6.09s/it]
Training Epoch: 12/100, batch 356/375 completed (loss: 0.3878:  96%|█████████▌| 179/187 [18:04<00:48,  6.09s/it]
Training Epoch: 12/100, batch 357/375 completed (loss: 0.0612:  96%|█████████▌| 179/187 [18:04<00:48,  6.09s/it]
Training Epoch: 12/100, batch 357/375 completed (loss: 0.0127:  96%|█████████▌| 179/187 [18:04<00:48,  6.09s/it]
Training Epoch: 12/100, batch 357/375 completed (loss: 0.1151:  96%|█████████▌| 179/187 [18:04<00:48,  6.09s/it]
Training Epoch: 12/100, batch 357/375 completed (loss: 0.2476:  96%|█████████▌| 179/187 [18:04<00:48,  6.09s/it]
Training Epoch: 12/100, batch 358/375 completed (loss: 0.0803:  96%|█████████▌| 179/187 [18:07<00:48,  6.09s/it]
Training Epoch: 12/100, batch 358/375 completed (loss: 0.0896:  96%|█████████▌| 179/187 [18:07<00:48,  6.09s/it]
Training Epoch: 12/100, batch 358/375 completed (loss: 0.0996:  96%|█████████▌| 179/187 [18:07<00:48,  6.09s/it]
Training Epoch: 12/100, batch 358/375 completed (loss: 0.2733:  96%|█████████▌| 179/187 [18:07<00:48,  6.09s/it]
Training Epoch: 12/100, batch 358/375 completed (loss: 0.2733:  96%|█████████▋| 180/187 [18:10<00:42,  6.07s/it]
Training Epoch: 12/100, batch 358/375 completed (loss: 0.0996:  96%|█████████▋| 180/187 [18:10<00:42,  6.07s/it]
Training Epoch: 12/100, batch 358/375 completed (loss: 0.0896:  96%|█████████▋| 180/187 [18:10<00:42,  6.07s/it]
Training Epoch: 12/100, batch 358/375 completed (loss: 0.0803:  96%|█████████▋| 180/187 [18:10<00:42,  6.07s/it]
Training Epoch: 12/100, batch 359/375 completed (loss: 0.0404:  96%|█████████▋| 180/187 [18:10<00:42,  6.07s/it]
Training Epoch: 12/100, batch 359/375 completed (loss: 0.0245:  96%|█████████▋| 180/187 [18:10<00:42,  6.07s/it]
Training Epoch: 12/100, batch 359/375 completed (loss: 0.238:  96%|█████████▋| 180/187 [18:10<00:42,  6.07s/it] 
Training Epoch: 12/100, batch 359/375 completed (loss: 0.0901:  96%|█████████▋| 180/187 [18:10<00:42,  6.07s/it]
Training Epoch: 12/100, batch 360/375 completed (loss: 0.0202:  96%|█████████▋| 180/187 [18:13<00:42,  6.07s/it]
Training Epoch: 12/100, batch 360/375 completed (loss: 0.0238:  96%|█████████▋| 180/187 [18:13<00:42,  6.07s/it]
Training Epoch: 12/100, batch 360/375 completed (loss: 0.0454:  96%|█████████▋| 180/187 [18:13<00:42,  6.07s/it]
Training Epoch: 12/100, batch 360/375 completed (loss: 0.3935:  96%|█████████▋| 180/187 [18:13<00:42,  6.07s/it]
Training Epoch: 12/100, batch 360/375 completed (loss: 0.0202:  97%|█████████▋| 181/187 [18:16<00:36,  6.10s/it]
Training Epoch: 12/100, batch 360/375 completed (loss: 0.0238:  97%|█████████▋| 181/187 [18:16<00:36,  6.10s/it]
Training Epoch: 12/100, batch 360/375 completed (loss: 0.3935:  97%|█████████▋| 181/187 [18:16<00:36,  6.10s/it]
Training Epoch: 12/100, batch 360/375 completed (loss: 0.0454:  97%|█████████▋| 181/187 [18:16<00:36,  6.10s/it]
Training Epoch: 12/100, batch 361/375 completed (loss: 0.0197:  97%|█████████▋| 181/187 [18:16<00:36,  6.10s/it]
Training Epoch: 12/100, batch 361/375 completed (loss: 0.0628:  97%|█████████▋| 181/187 [18:16<00:36,  6.10s/it]
Training Epoch: 12/100, batch 361/375 completed (loss: 0.1917:  97%|█████████▋| 181/187 [18:16<00:36,  6.10s/it]
Training Epoch: 12/100, batch 361/375 completed (loss: 0.0582:  97%|█████████▋| 181/187 [18:16<00:36,  6.10s/it]
Training Epoch: 12/100, batch 362/375 completed (loss: 0.0309:  97%|█████████▋| 181/187 [18:19<00:36,  6.10s/it]
Training Epoch: 12/100, batch 362/375 completed (loss: 0.2635:  97%|█████████▋| 181/187 [18:19<00:36,  6.10s/it]
Training Epoch: 12/100, batch 362/375 completed (loss: 0.1701:  97%|█████████▋| 181/187 [18:19<00:36,  6.10s/it]
Training Epoch: 12/100, batch 362/375 completed (loss: 0.2064:  97%|█████████▋| 181/187 [18:19<00:36,  6.10s/it]
Training Epoch: 12/100, batch 362/375 completed (loss: 0.0309:  97%|█████████▋| 182/187 [18:22<00:30,  6.14s/it]
Training Epoch: 12/100, batch 362/375 completed (loss: 0.2635:  97%|█████████▋| 182/187 [18:22<00:30,  6.14s/it]
Training Epoch: 12/100, batch 362/375 completed (loss: 0.1701:  97%|█████████▋| 182/187 [18:22<00:30,  6.14s/it]
Training Epoch: 12/100, batch 362/375 completed (loss: 0.2064:  97%|█████████▋| 182/187 [18:22<00:30,  6.14s/it]
Training Epoch: 12/100, batch 363/375 completed (loss: 0.0212:  97%|█████████▋| 182/187 [18:22<00:30,  6.14s/it]
Training Epoch: 12/100, batch 363/375 completed (loss: 0.0741:  97%|█████████▋| 182/187 [18:22<00:30,  6.14s/it]
Training Epoch: 12/100, batch 363/375 completed (loss: 0.1869:  97%|█████████▋| 182/187 [18:22<00:30,  6.14s/it]
Training Epoch: 12/100, batch 363/375 completed (loss: 0.1495:  97%|█████████▋| 182/187 [18:22<00:30,  6.14s/it]
Training Epoch: 12/100, batch 364/375 completed (loss: 0.1556:  97%|█████████▋| 182/187 [18:25<00:30,  6.14s/it]
Training Epoch: 12/100, batch 364/375 completed (loss: 0.1667:  97%|█████████▋| 182/187 [18:25<00:30,  6.14s/it]
Training Epoch: 12/100, batch 364/375 completed (loss: 0.2733:  97%|█████████▋| 182/187 [18:25<00:30,  6.14s/it]
Training Epoch: 12/100, batch 364/375 completed (loss: 0.2507:  97%|█████████▋| 182/187 [18:25<00:30,  6.14s/it]
Training Epoch: 12/100, batch 364/375 completed (loss: 0.2733:  98%|█████████▊| 183/187 [18:28<00:24,  6.05s/it]
Training Epoch: 12/100, batch 364/375 completed (loss: 0.1556:  98%|█████████▊| 183/187 [18:28<00:24,  6.06s/it]
Training Epoch: 12/100, batch 364/375 completed (loss: 0.2507:  98%|█████████▊| 183/187 [18:28<00:24,  6.06s/it]
Training Epoch: 12/100, batch 364/375 completed (loss: 0.1667:  98%|█████████▊| 183/187 [18:28<00:24,  6.06s/it]
Training Epoch: 12/100, batch 365/375 completed (loss: 0.0318:  98%|█████████▊| 183/187 [18:28<00:24,  6.06s/it]
Training Epoch: 12/100, batch 365/375 completed (loss: 0.2768:  98%|█████████▊| 183/187 [18:28<00:24,  6.05s/it]
Training Epoch: 12/100, batch 365/375 completed (loss: 0.0651:  98%|█████████▊| 183/187 [18:28<00:24,  6.06s/it]
Training Epoch: 12/100, batch 365/375 completed (loss: 0.0703:  98%|█████████▊| 183/187 [18:28<00:24,  6.06s/it]
Training Epoch: 12/100, batch 366/375 completed (loss: 0.0159:  98%|█████████▊| 183/187 [18:31<00:24,  6.06s/it]
Training Epoch: 12/100, batch 366/375 completed (loss: 0.0318:  98%|█████████▊| 183/187 [18:31<00:24,  6.06s/it]
Training Epoch: 12/100, batch 366/375 completed (loss: 0.2741:  98%|█████████▊| 183/187 [18:31<00:24,  6.05s/it]
Training Epoch: 12/100, batch 366/375 completed (loss: 0.1853:  98%|█████████▊| 183/187 [18:31<00:24,  6.06s/it]
Training Epoch: 12/100, batch 366/375 completed (loss: 0.0318:  98%|█████████▊| 184/187 [18:35<00:18,  6.18s/it]
Training Epoch: 12/100, batch 366/375 completed (loss: 0.2741:  98%|█████████▊| 184/187 [18:35<00:18,  6.18s/it]
Training Epoch: 12/100, batch 366/375 completed (loss: 0.0159:  98%|█████████▊| 184/187 [18:35<00:18,  6.18s/it]
Training Epoch: 12/100, batch 366/375 completed (loss: 0.1853:  98%|█████████▊| 184/187 [18:35<00:18,  6.18s/it]
Training Epoch: 12/100, batch 367/375 completed (loss: 0.0581:  98%|█████████▊| 184/187 [18:35<00:18,  6.18s/it]
Training Epoch: 12/100, batch 367/375 completed (loss: 0.1098:  98%|█████████▊| 184/187 [18:35<00:18,  6.18s/it]
Training Epoch: 12/100, batch 367/375 completed (loss: 0.2056:  98%|█████████▊| 184/187 [18:35<00:18,  6.18s/it]
Training Epoch: 12/100, batch 367/375 completed (loss: 0.1127:  98%|█████████▊| 184/187 [18:35<00:18,  6.18s/it]
Training Epoch: 12/100, batch 368/375 completed (loss: 0.1481:  98%|█████████▊| 184/187 [18:37<00:18,  6.18s/it]
Training Epoch: 12/100, batch 368/375 completed (loss: 0.0944:  98%|█████████▊| 184/187 [18:37<00:18,  6.18s/it]
Training Epoch: 12/100, batch 368/375 completed (loss: 0.113:  98%|█████████▊| 184/187 [18:37<00:18,  6.18s/it] 
Training Epoch: 12/100, batch 368/375 completed (loss: 0.1689:  98%|█████████▊| 184/187 [18:38<00:18,  6.18s/it]
Training Epoch: 12/100, batch 368/375 completed (loss: 0.0944:  99%|█████████▉| 185/187 [18:40<00:12,  6.07s/it]
Training Epoch: 12/100, batch 368/375 completed (loss: 0.113:  99%|█████████▉| 185/187 [18:40<00:12,  6.07s/it]
Training Epoch: 12/100, batch 368/375 completed (loss: 0.1481:  99%|█████████▉| 185/187 [18:40<00:12,  6.08s/it]
Training Epoch: 12/100, batch 368/375 completed (loss: 0.1689:  99%|█████████▉| 185/187 [18:40<00:12,  6.08s/it]
Training Epoch: 12/100, batch 369/375 completed (loss: 0.3931:  99%|█████████▉| 185/187 [18:40<00:12,  6.08s/it]
Training Epoch: 12/100, batch 369/375 completed (loss: 0.0959:  99%|█████████▉| 185/187 [18:40<00:12,  6.07s/it]
Training Epoch: 12/100, batch 369/375 completed (loss: 0.1748:  99%|█████████▉| 185/187 [18:40<00:12,  6.07s/it]
Training Epoch: 12/100, batch 369/375 completed (loss: 0.0444:  99%|█████████▉| 185/187 [18:40<00:12,  6.08s/it]
Training Epoch: 12/100, batch 370/375 completed (loss: 0.1138:  99%|█████████▉| 185/187 [18:43<00:12,  6.08s/it]
Training Epoch: 12/100, batch 370/375 completed (loss: 0.115:  99%|█████████▉| 185/187 [18:43<00:12,  6.07s/it] 
Training Epoch: 12/100, batch 370/375 completed (loss: 0.094:  99%|█████████▉| 185/187 [18:43<00:12,  6.07s/it] 
Training Epoch: 12/100, batch 370/375 completed (loss: 0.2004:  99%|█████████▉| 185/187 [18:43<00:12,  6.08s/it]
Training Epoch: 12/100, batch 370/375 completed (loss: 0.2004:  99%|█████████▉| 186/187 [18:47<00:06,  6.10s/it]
Training Epoch: 12/100, batch 370/375 completed (loss: 0.094:  99%|█████████▉| 186/187 [18:47<00:06,  6.10s/it]
Training Epoch: 12/100, batch 370/375 completed (loss: 0.1138:  99%|█████████▉| 186/187 [18:47<00:06,  6.10s/it]
Training Epoch: 12/100, batch 370/375 completed (loss: 0.115:  99%|█████████▉| 186/187 [18:47<00:06,  6.11s/it]
Training Epoch: 12/100, batch 371/375 completed (loss: 0.2075:  99%|█████████▉| 186/187 [18:47<00:06,  6.10s/it]
Training Epoch: 12/100, batch 371/375 completed (loss: 0.2657:  99%|█████████▉| 186/187 [18:47<00:06,  6.11s/it]
Training Epoch: 12/100, batch 371/375 completed (loss: 0.2865:  99%|█████████▉| 186/187 [18:47<00:06,  6.10s/it]
Training Epoch: 12/100, batch 371/375 completed (loss: 0.3133:  99%|█████████▉| 186/187 [18:47<00:06,  6.10s/it]
Training Epoch: 12/100, batch 372/375 completed (loss: 0.0525:  99%|█████████▉| 186/187 [18:49<00:06,  6.10s/it]
Training Epoch: 12/100, batch 372/375 completed (loss: 0.0266:  99%|█████████▉| 186/187 [18:49<00:06,  6.11s/it]
Training Epoch: 12/100, batch 372/375 completed (loss: 0.3497:  99%|█████████▉| 186/187 [18:49<00:06,  6.10s/it]
Training Epoch: 12/100, batch 372/375 completed (loss: 0.145:  99%|█████████▉| 186/187 [18:50<00:06,  6.10s/it] 
Training Epoch: 12/100, batch 372/375 completed (loss: 0.0525: 100%|██████████| 187/187 [18:53<00:00,  6.11s/it]
Training Epoch: 12/100, batch 372/375 completed (loss: 0.3497: 100%|██████████| 187/187 [18:53<00:00,  6.11s/it]
Training Epoch: 12/100, batch 372/375 completed (loss: 0.0266: 100%|██████████| 187/187 [18:53<00:00,  6.11s/it]
Training Epoch: 12/100, batch 372/375 completed (loss: 0.145: 100%|██████████| 187/187 [18:53<00:00,  6.11s/it]
Training Epoch: 12/100, batch 373/375 completed (loss: 0.3383: 100%|██████████| 187/187 [18:53<00:00,  6.11s/it]
Training Epoch: 12/100, batch 373/375 completed (loss: 0.1518: 100%|██████████| 187/187 [18:53<00:00,  6.11s/it]
Training Epoch: 12/100, batch 373/375 completed (loss: 0.0298: 100%|██████████| 187/187 [18:53<00:00,  6.11s/it]
Training Epoch: 12/100, batch 373/375 completed (loss: 0.0565: 100%|██████████| 187/187 [18:53<00:00,  6.11s/it]
Training Epoch: 12/100, batch 373/375 completed (loss: 0.3383: : 188it [18:56,  5.15s/it]                       
Training Epoch: 12/100, batch 373/375 completed (loss: 0.0298: : 188it [18:56,  5.15s/it]                       
Training Epoch: 12/100, batch 373/375 completed (loss: 0.1518: : 188it [18:56,  5.15s/it]                       
Training Epoch: 12/100, batch 373/375 completed (loss: 0.0565: : 188it [18:56,  5.15s/it]                       
Training Epoch: 12/100, batch 374/375 completed (loss: 0.0419: : 188it [18:56,  5.15s/it]
Training Epoch: 12/100, batch 374/375 completed (loss: 0.3173: : 188it [18:56,  5.15s/it]
Training Epoch: 12/100, batch 374/375 completed (loss: 0.2184: : 188it [18:56,  5.15s/it]
Training Epoch: 12/100, batch 374/375 completed (loss: 0.428: : 188it [18:56,  5.15s/it] 
Training Epoch: 12/100, batch 374/375 completed (loss: 0.2184: : 188it [18:56,  6.04s/it]

Training Epoch: 12/100, batch 374/375 completed (loss: 0.0419: : 188it [18:56,  6.04s/it]

Training Epoch: 12/100, batch 374/375 completed (loss: 0.3173: : 188it [18:56,  6.04s/it]

Training Epoch: 12/100, batch 374/375 completed (loss: 0.428: : 188it [18:56,  6.04s/it]
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/c10d_logger.py:83: UserWarning: barrier(): using the device under current context. You can specify `device_id` in `init_process_group` to mute this warning.
  return func(*args, **kwargs)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
saving model + train state.
Warning: failed to upload checkpoint_12 to HF: (Request ID: Root=1-6a1b8555-4f95d511137ef8d53a0041ae;de1872d7-5f7e-46e1-a25b-7ccebb74d5a1)

403 Forbidden: You need to setup automatic credit recharge in order to upload more data. You can do so at /settings/billing..
Cannot access content at: https://huggingface.co/seyedparsa/thoughtformer-checkpoints.git/info/lfs/objects/batch.
Make sure your token has the correct permissions.
Warning: failed to upload train_state_12 to HF: (Request ID: Root=1-6a1b8559-7c96ac387d402d03254b7405;ca9485e9-aee4-4d7e-96a5-c71f80c20c44)

403 Forbidden: You need to setup automatic credit recharge in order to upload more data. You can do so at /settings/billing..
Cannot access content at: https://huggingface.co/seyedparsa/thoughtformer-checkpoints.git/info/lfs/objects/batch.
Make sure your token has the correct permissions.
Deleted old checkpoint: train_state_11
Deleted old checkpoint: checkpoint_11
[warn] HF upload failed for ckpt_12; kept locally, earlier ckpts deleted.

Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]
Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]
Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]eval loss 0.05611330805052268

Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]
[CORRECT] Example 0 (diff=3)
  Expected : 'Claude'
  Extracted: 'Claude'
  Generated: '### Claude'

[CORRECT] Example 1 (diff=8)
  Expected : 'Guy'
  Extracted: 'Guy'
Test Accuracy:   3%|▎         | 1/32 [00:07<03:38,  7.05s/it]

Test Accuracy:   3%|▎         | 1/32 [00:07<03:38,  7.06s/it]
Test Accuracy:   3%|▎         | 1/32 [00:07<03:38,  7.06s/it]  Generated: '### Guy'

Test accuracy: 1.0:   3%|▎         | 1/32 [00:07<03:38,  7.05s/it]
Test accuracy: 0.62:   3%|▎         | 1/32 [00:07<03:38,  7.06s/it]
Test accuracy: 0.88:   3%|▎         | 1/32 [00:07<03:38,  7.06s/it]
[CORRECT] Example 2 (diff=10)
  Expected : 'Elvis'
  Extracted: 'Elvis'
  Generated: '### Elvis'

[CORRECT] Example 3 (diff=8)
  Expected : 'Coral'
  Extracted: 'Coral'
  Generated: '### Coral'

[CORRECT] Example 4 (diff=1)
  Expected : 'Patricia'
  Extracted: 'Patricia'
  Generated: '### Patricia'

Test Accuracy:   3%|▎         | 1/32 [00:06<03:34,  6.91s/it]
Test accuracy: 1.0:   3%|▎         | 1/32 [00:06<03:34,  6.91s/it]
Test accuracy: 0.88:   6%|▋         | 2/32 [00:13<03:15,  6.53s/it]
Test accuracy: 1.0:   6%|▋         | 2/32 [00:13<03:15,  6.52s/it]
Test accuracy: 0.62:   6%|▋         | 2/32 [00:13<03:15,  6.52s/it]
Test accuracy: 1.0:   6%|▋         | 2/32 [00:13<03:15,  6.52s/it]
Test accuracy: 0.88:   6%|▋         | 2/32 [00:13<03:15,  6.53s/it]
Test accuracy: 0.56:   6%|▋         | 2/32 [00:13<03:15,  6.52s/it]
Test accuracy: 1.0:   6%|▋         | 2/32 [00:13<03:13,  6.46s/it]
Test accuracy: 0.88:   6%|▋         | 2/32 [00:13<03:13,  6.46s/it]
Test accuracy: 0.56:   9%|▉         | 3/32 [00:19<03:12,  6.64s/it]
Test accuracy: 1.0:   9%|▉         | 3/32 [00:19<03:12,  6.64s/it]
Test accuracy: 0.67:   9%|▉         | 3/32 [00:19<03:12,  6.64s/it]
Test accuracy: 0.96:   9%|▉         | 3/32 [00:19<03:12,  6.64s/it]
Test accuracy: 0.88:   9%|▉         | 3/32 [00:19<03:11,  6.61s/it]
Test accuracy: 0.79:   9%|▉         | 3/32 [00:19<03:11,  6.61s/it]
Test accuracy: 0.88:   9%|▉         | 3/32 [00:20<03:12,  6.65s/it]
Test accuracy: 0.83:   9%|▉         | 3/32 [00:20<03:12,  6.65s/it]
Test accuracy: 0.96:  12%|█▎        | 4/32 [00:26<03:04,  6.61s/it]
Test accuracy: 0.83:  12%|█▎        | 4/32 [00:26<03:04,  6.61s/it]
Test accuracy: 0.67:  12%|█▎        | 4/32 [00:26<03:05,  6.61s/it]
Test accuracy: 0.79:  12%|█▎        | 4/32 [00:26<03:04,  6.59s/it]
Test accuracy: 0.88:  12%|█▎        | 4/32 [00:26<03:04,  6.61s/it]
Test accuracy: 0.78:  12%|█▎        | 4/32 [00:26<03:04,  6.61s/it]
Test accuracy: 0.66:  12%|█▎        | 4/32 [00:26<03:05,  6.61s/it]
Test accuracy: 0.81:  12%|█▎        | 4/32 [00:26<03:04,  6.59s/it]
Test accuracy: 0.78:  16%|█▌        | 5/32 [00:32<02:53,  6.44s/it]
Test accuracy: 0.88:  16%|█▌        | 5/32 [00:32<02:53,  6.44s/it]
Test accuracy: 0.66:  16%|█▌        | 5/32 [00:32<02:53,  6.44s/it]
Test accuracy: 0.8:  16%|█▌        | 5/32 [00:32<02:53,  6.44s/it] 
Test accuracy: 0.77:  16%|█▌        | 5/32 [00:32<02:53,  6.44s/it]
Test accuracy: 0.68:  16%|█▌        | 5/32 [00:32<02:53,  6.44s/it]
Test accuracy: 0.81:  16%|█▌        | 5/32 [00:32<02:53,  6.42s/it]
Test accuracy: 0.8:  16%|█▌        | 5/32 [00:32<02:53,  6.42s/it] 
Test accuracy: 0.8:  19%|█▉        | 6/32 [00:39<02:46,  6.40s/it]
Test accuracy: 0.68:  19%|█▉        | 6/32 [00:39<02:46,  6.40s/it]
Test accuracy: 0.77:  19%|█▉        | 6/32 [00:39<02:46,  6.40s/it]
Test accuracy: 0.79:  19%|█▉        | 6/32 [00:39<02:46,  6.40s/it]
Test accuracy: 0.71:  19%|█▉        | 6/32 [00:39<02:46,  6.40s/it]
Test accuracy: 0.79:  19%|█▉        | 6/32 [00:39<02:46,  6.40s/it]
Test accuracy: 0.8:  19%|█▉        | 6/32 [00:38<02:46,  6.40s/it]
Test accuracy: 0.81:  19%|█▉        | 6/32 [00:38<02:46,  6.40s/it]
Test accuracy: 0.71:  22%|██▏       | 7/32 [00:45<02:39,  6.37s/it]
Test accuracy: 0.79:  22%|██▏       | 7/32 [00:45<02:39,  6.37s/it]
Test accuracy: 0.79:  22%|██▏       | 7/32 [00:45<02:39,  6.37s/it]
Test accuracy: 0.73:  22%|██▏       | 7/32 [00:45<02:39,  6.37s/it]
Test accuracy: 0.77:  22%|██▏       | 7/32 [00:45<02:39,  6.37s/it]
Test accuracy: 0.77:  22%|██▏       | 7/32 [00:45<02:39,  6.37s/it]
Test accuracy: 0.81:  22%|██▏       | 7/32 [00:45<02:39,  6.37s/it]
Test accuracy: 0.8:  22%|██▏       | 7/32 [00:45<02:39,  6.37s/it] 
Test accuracy: 0.73:  25%|██▌       | 8/32 [00:51<02:31,  6.29s/it]
Test accuracy: 0.77:  25%|██▌       | 8/32 [00:51<02:31,  6.29s/it]
Test accuracy: 0.77:  25%|██▌       | 8/32 [00:51<02:31,  6.29s/it]
Test accuracy: 0.73:  25%|██▌       | 8/32 [00:51<02:31,  6.29s/it]
Test accuracy: 0.8:  25%|██▌       | 8/32 [00:51<02:31,  6.29s/it] 
Test accuracy: 0.8:  25%|██▌       | 8/32 [00:51<02:31,  6.29s/it] 
Test accuracy: 0.8:  25%|██▌       | 8/32 [00:51<02:30,  6.29s/it]
Test accuracy: 0.83:  25%|██▌       | 8/32 [00:51<02:30,  6.29s/it]
Test accuracy: 0.8:  28%|██▊       | 9/32 [00:58<02:29,  6.51s/it]
Test accuracy: 0.8:  28%|██▊       | 9/32 [00:58<02:29,  6.51s/it]
Test accuracy: 0.78:  28%|██▊       | 9/32 [00:58<02:29,  6.51s/it]
Test accuracy: 0.79:  28%|██▊       | 9/32 [00:58<02:29,  6.51s/it]
Test accuracy: 0.83:  28%|██▊       | 9/32 [00:58<02:29,  6.50s/it]
Test accuracy: 0.83:  28%|██▊       | 9/32 [00:58<02:29,  6.50s/it]
Test accuracy: 0.73:  28%|██▊       | 9/32 [00:58<02:29,  6.51s/it]
Test accuracy: 0.75:  28%|██▊       | 9/32 [00:58<02:29,  6.51s/it]
Test accuracy: 0.79:  31%|███▏      | 10/32 [01:04<02:21,  6.45s/it]
Test accuracy: 0.75:  31%|███▏      | 10/32 [01:04<02:21,  6.45s/it]
Test accuracy: 0.8:  31%|███▏      | 10/32 [01:04<02:21,  6.45s/it] 
Test accuracy: 0.78:  31%|███▏      | 10/32 [01:04<02:21,  6.45s/it]
Test accuracy: 0.75:  31%|███▏      | 10/32 [01:04<02:21,  6.45s/it]
Test accuracy: 0.76:  31%|███▏      | 10/32 [01:04<02:21,  6.45s/it]
Test accuracy: 0.83:  31%|███▏      | 10/32 [01:04<02:21,  6.45s/it]
Test accuracy: 0.82:  31%|███▏      | 10/32 [01:04<02:21,  6.45s/it]
Test accuracy: 0.8:  34%|███▍      | 11/32 [01:10<02:13,  6.36s/it]
Test accuracy: 0.75:  34%|███▍      | 11/32 [01:10<02:13,  6.36s/it]
Test accuracy: 0.77:  34%|███▍      | 11/32 [01:10<02:13,  6.36s/it]
Test accuracy: 0.77:  34%|███▍      | 11/32 [01:10<02:13,  6.36s/it]
Test accuracy: 0.82:  34%|███▍      | 11/32 [01:10<02:13,  6.36s/it]
Test accuracy: 0.8:  34%|███▍      | 11/32 [01:10<02:13,  6.36s/it] 
Test accuracy: 0.76:  34%|███▍      | 11/32 [01:10<02:13,  6.36s/it]
Test accuracy: 0.77:  34%|███▍      | 11/32 [01:10<02:13,  6.36s/it]
Test accuracy: 0.77:  38%|███▊      | 12/32 [01:17<02:07,  6.35s/it]
Test accuracy: 0.8:  38%|███▊      | 12/32 [01:17<02:07,  6.35s/it]
Test accuracy: 0.77:  38%|███▊      | 12/32 [01:17<02:07,  6.35s/it]
Test accuracy: 0.75:  38%|███▊      | 12/32 [01:17<02:07,  6.35s/it]
Test accuracy: 0.77:  38%|███▊      | 12/32 [01:17<02:07,  6.35s/it]
Test accuracy: 0.8:  38%|███▊      | 12/32 [01:17<02:07,  6.35s/it]
Test accuracy: 0.77:  38%|███▊      | 12/32 [01:17<02:07,  6.36s/it]
Test accuracy: 0.76:  38%|███▊      | 12/32 [01:17<02:07,  6.36s/it]
Test accuracy: 0.76:  41%|████      | 13/32 [01:23<02:00,  6.35s/it]
Test accuracy: 0.75:  41%|████      | 13/32 [01:23<02:00,  6.35s/it]
Test accuracy: 0.77:  41%|████      | 13/32 [01:23<02:00,  6.35s/it]
Test accuracy: 0.8:  41%|████      | 13/32 [01:23<02:00,  6.35s/it]
Test accuracy: 0.77:  41%|████      | 13/32 [01:23<02:00,  6.35s/it]
Test accuracy: 0.77:  41%|████      | 13/32 [01:23<02:00,  6.35s/it]
Test accuracy: 0.76:  41%|████      | 13/32 [01:23<02:00,  6.35s/it]
Test accuracy: 0.78:  41%|████      | 13/32 [01:23<02:00,  6.35s/it]
Test accuracy: 0.77:  44%|████▍     | 14/32 [01:29<01:53,  6.29s/it]
Test accuracy: 0.76:  44%|████▍     | 14/32 [01:29<01:53,  6.29s/it]
Test accuracy: 0.79:  44%|████▍     | 14/32 [01:29<01:53,  6.29s/it]
Test accuracy: 0.78:  44%|████▍     | 14/32 [01:29<01:53,  6.29s/it]
Test accuracy: 0.78:  44%|████▍     | 14/32 [01:29<01:53,  6.29s/it]
Test accuracy: 0.77:  44%|████▍     | 14/32 [01:29<01:53,  6.29s/it]
Test accuracy: 0.79:  44%|████▍     | 14/32 [01:29<01:53,  6.29s/it]
Test accuracy: 0.76:  44%|████▍     | 14/32 [01:29<01:53,  6.29s/it]
Test accuracy: 0.78:  47%|████▋     | 15/32 [01:36<01:47,  6.31s/it]
Test accuracy: 0.79:  47%|████▋     | 15/32 [01:36<01:47,  6.31s/it]
Test accuracy: 0.79:  47%|████▋     | 15/32 [01:36<01:47,  6.31s/it]
Test accuracy: 0.77:  47%|████▋     | 15/32 [01:36<01:47,  6.31s/it]
Test accuracy: 0.79:  47%|████▋     | 15/32 [01:36<01:47,  6.31s/it]
Test accuracy: 0.77:  47%|████▋     | 15/32 [01:36<01:47,  6.31s/it]
Test accuracy: 0.76:  47%|████▋     | 15/32 [01:35<01:47,  6.31s/it]
Test accuracy: 0.76:  47%|████▋     | 15/32 [01:35<01:47,  6.31s/it]
Test accuracy: 0.77:  50%|█████     | 16/32 [01:42<01:40,  6.30s/it]
Test accuracy: 0.79:  50%|█████     | 16/32 [01:42<01:40,  6.30s/it]
Test accuracy: 0.77:  50%|█████     | 16/32 [01:42<01:40,  6.30s/it]
Test accuracy: 0.77:  50%|█████     | 16/32 [01:42<01:40,  6.30s/it]
Test accuracy: 0.8:  50%|█████     | 16/32 [01:42<01:40,  6.30s/it] 
Test accuracy: 0.77:  50%|█████     | 16/32 [01:42<01:40,  6.30s/it]
Test accuracy: 0.76:  50%|█████     | 16/32 [01:42<01:40,  6.30s/it]
Test accuracy: 0.76:  50%|█████     | 16/32 [01:42<01:40,  6.30s/it]
Test accuracy: 0.77:  53%|█████▎    | 17/32 [01:48<01:34,  6.32s/it]
Test accuracy: 0.8:  53%|█████▎    | 17/32 [01:48<01:34,  6.32s/it]
Test accuracy: 0.77:  53%|█████▎    | 17/32 [01:48<01:34,  6.32s/it]
Test accuracy: 0.8:  53%|█████▎    | 17/32 [01:48<01:34,  6.32s/it]
Test accuracy: 0.76:  53%|█████▎    | 17/32 [01:48<01:34,  6.32s/it]
Test accuracy: 0.76:  53%|█████▎    | 17/32 [01:48<01:34,  6.32s/it]
Test accuracy: 0.77:  53%|█████▎    | 17/32 [01:48<01:34,  6.32s/it]
Test accuracy: 0.78:  53%|█████▎    | 17/32 [01:48<01:34,  6.32s/it]
Test accuracy: 0.8:  56%|█████▋    | 18/32 [01:55<01:28,  6.33s/it]
Test accuracy: 0.77:  56%|█████▋    | 18/32 [01:55<01:28,  6.33s/it]
Test accuracy: 0.78:  56%|█████▋    | 18/32 [01:55<01:28,  6.33s/it]
Test accuracy: 0.81:  56%|█████▋    | 18/32 [01:55<01:28,  6.33s/it]
Test accuracy: 0.78:  56%|█████▋    | 18/32 [01:55<01:28,  6.33s/it]
Test accuracy: 0.76:  56%|█████▋    | 18/32 [01:54<01:28,  6.33s/it]
Test accuracy: 0.78:  56%|█████▋    | 18/32 [01:55<01:28,  6.33s/it]
Test accuracy: 0.77:  56%|█████▋    | 18/32 [01:54<01:28,  6.33s/it]
Test accuracy: 0.78:  59%|█████▉    | 19/32 [02:01<01:21,  6.28s/it]
Test accuracy: 0.81:  59%|█████▉    | 19/32 [02:01<01:21,  6.28s/it]
Test accuracy: 0.78:  59%|█████▉    | 19/32 [02:01<01:21,  6.28s/it]
Test accuracy: 0.81:  59%|█████▉    | 19/32 [02:01<01:21,  6.28s/it]
Test accuracy: 0.78:  59%|█████▉    | 19/32 [02:01<01:21,  6.28s/it]
Test accuracy: 0.78:  59%|█████▉    | 19/32 [02:01<01:21,  6.28s/it]
Test accuracy: 0.77:  59%|█████▉    | 19/32 [02:01<01:21,  6.28s/it]
Test accuracy: 0.76:  59%|█████▉    | 19/32 [02:01<01:21,  6.28s/it]
Test accuracy: 0.81:  62%|██████▎   | 20/32 [02:07<01:15,  6.30s/it]
Test accuracy: 0.78:  62%|██████▎   | 20/32 [02:07<01:15,  6.30s/it]
Test accuracy: 0.79:  62%|██████▎   | 20/32 [02:07<01:15,  6.30s/it]
Test accuracy: 0.77:  62%|██████▎   | 20/32 [02:07<01:15,  6.30s/it]
Test accuracy: 0.78:  62%|██████▎   | 20/32 [02:07<01:15,  6.30s/it]
Test accuracy: 0.78:  62%|██████▎   | 20/32 [02:07<01:15,  6.30s/it]
Test accuracy: 0.76:  62%|██████▎   | 20/32 [02:07<01:15,  6.30s/it]
Test accuracy: 0.77:  62%|██████▎   | 20/32 [02:07<01:15,  6.30s/it]
Test accuracy: 0.78:  66%|██████▌   | 21/32 [02:14<01:11,  6.52s/it]
Test accuracy: 0.79:  66%|██████▌   | 21/32 [02:14<01:11,  6.52s/it]
Test accuracy: 0.79:  66%|██████▌   | 21/32 [02:14<01:11,  6.52s/it]
Test accuracy: 0.78:  66%|██████▌   | 21/32 [02:14<01:11,  6.52s/it]
Test accuracy: 0.77:  66%|██████▌   | 21/32 [02:14<01:11,  6.52s/it]
Test accuracy: 0.76:  66%|██████▌   | 21/32 [02:14<01:11,  6.52s/it]
Test accuracy: 0.77:  66%|██████▌   | 21/32 [02:14<01:11,  6.52s/it]
Test accuracy: 0.78:  66%|██████▌   | 21/32 [02:14<01:11,  6.52s/it]
Test accuracy: 0.79:  69%|██████▉   | 22/32 [02:20<01:04,  6.41s/it]
Test accuracy: 0.78:  69%|██████▉   | 22/32 [02:20<01:04,  6.41s/it]
Test accuracy: 0.78:  69%|██████▉   | 22/32 [02:20<01:04,  6.41s/it]
Test accuracy: 0.8:  69%|██████▉   | 22/32 [02:20<01:04,  6.41s/it] 
Test accuracy: 0.78:  69%|██████▉   | 22/32 [02:20<01:04,  6.41s/it]
Test accuracy: 0.77:  69%|██████▉   | 22/32 [02:20<01:04,  6.41s/it]
Test accuracy: 0.76:  69%|██████▉   | 22/32 [02:20<01:04,  6.41s/it]
Test accuracy: 0.76:  69%|██████▉   | 22/32 [02:20<01:04,  6.41s/it]
Test accuracy: 0.77:  72%|███████▏  | 23/32 [02:27<00:57,  6.40s/it]
Test accuracy: 0.8:  72%|███████▏  | 23/32 [02:27<00:57,  6.40s/it]
Test accuracy: 0.77:  72%|███████▏  | 23/32 [02:27<00:57,  6.40s/it]
Test accuracy: 0.76:  72%|███████▏  | 23/32 [02:27<00:57,  6.40s/it]
Test accuracy: 0.79:  72%|███████▏  | 23/32 [02:27<00:57,  6.40s/it]
Test accuracy: 0.76:  72%|███████▏  | 23/32 [02:27<00:57,  6.40s/it]
Test accuracy: 0.78:  72%|███████▏  | 23/32 [02:27<00:57,  6.40s/it]
Test accuracy: 0.78:  72%|███████▏  | 23/32 [02:27<00:57,  6.40s/it]
Test accuracy: 0.78:  75%|███████▌  | 24/32 [02:34<00:52,  6.59s/it]
Test accuracy: 0.79:  75%|███████▌  | 24/32 [02:34<00:52,  6.59s/it]
Test accuracy: 0.78:  75%|███████▌  | 24/32 [02:34<00:52,  6.59s/it]
Test accuracy: 0.77:  75%|███████▌  | 24/32 [02:34<00:52,  6.59s/it]
Test accuracy: 0.8:  75%|███████▌  | 24/32 [02:34<00:52,  6.59s/it] 
Test accuracy: 0.78:  75%|███████▌  | 24/32 [02:34<00:52,  6.59s/it]
Test accuracy: 0.76:  75%|███████▌  | 24/32 [02:34<00:52,  6.60s/it]
Test accuracy: 0.76:  75%|███████▌  | 24/32 [02:34<00:52,  6.60s/it]
Test accuracy: 0.78:  78%|███████▊  | 25/32 [02:40<00:45,  6.46s/it]
Test accuracy: 0.8:  78%|███████▊  | 25/32 [02:40<00:45,  6.46s/it]
Test accuracy: 0.78:  78%|███████▊  | 25/32 [02:40<00:45,  6.46s/it]
Test accuracy: 0.79:  78%|███████▊  | 25/32 [02:40<00:45,  6.46s/it]
Test accuracy: 0.79:  78%|███████▊  | 25/32 [02:40<00:45,  6.46s/it]
Test accuracy: 0.77:  78%|███████▊  | 25/32 [02:40<00:45,  6.46s/it]
Test accuracy: 0.76:  78%|███████▊  | 25/32 [02:40<00:45,  6.46s/it]
Test accuracy: 0.75:  78%|███████▊  | 25/32 [02:40<00:45,  6.46s/it]
Test accuracy: 0.79:  81%|████████▏ | 26/32 [02:46<00:38,  6.43s/it]
Test accuracy: 0.79:  81%|████████▏ | 26/32 [02:46<00:38,  6.43s/it]
Test accuracy: 0.77:  81%|████████▏ | 26/32 [02:46<00:38,  6.43s/it]
Test accuracy: 0.75:  81%|████████▏ | 26/32 [02:46<00:38,  6.43s/it]
Test accuracy: 0.78:  81%|████████▏ | 26/32 [02:46<00:38,  6.43s/it]
Test accuracy: 0.79:  81%|████████▏ | 26/32 [02:46<00:38,  6.43s/it]
Test accuracy: 0.77:  81%|████████▏ | 26/32 [02:46<00:38,  6.43s/it]
Test accuracy: 0.75:  81%|████████▏ | 26/32 [02:46<00:38,  6.43s/it]
Test accuracy: 0.77:  84%|████████▍ | 27/32 [02:53<00:32,  6.40s/it]
Test accuracy: 0.79:  84%|████████▍ | 27/32 [02:53<00:32,  6.40s/it]
Test accuracy: 0.78:  84%|████████▍ | 27/32 [02:53<00:32,  6.40s/it]
Test accuracy: 0.8:  84%|████████▍ | 27/32 [02:53<00:32,  6.40s/it] 
Test accuracy: 0.75:  84%|████████▍ | 27/32 [02:52<00:32,  6.40s/it]
Test accuracy: 0.74:  84%|████████▍ | 27/32 [02:52<00:32,  6.40s/it]
Test accuracy: 0.78:  84%|████████▍ | 27/32 [02:53<00:32,  6.40s/it]
Test accuracy: 0.78:  84%|████████▍ | 27/32 [02:53<00:32,  6.40s/it]
Test accuracy: 0.8:  88%|████████▊ | 28/32 [03:00<00:26,  6.60s/it]
Test accuracy: 0.74:  88%|████████▊ | 28/32 [02:59<00:26,  6.60s/it]
Test accuracy: 0.78:  88%|████████▊ | 28/32 [03:00<00:26,  6.60s/it]
Test accuracy: 0.79:  88%|████████▊ | 28/32 [03:00<00:26,  6.60s/it]
Test accuracy: 0.78:  88%|████████▊ | 28/32 [03:00<00:26,  6.60s/it]
Test accuracy: 0.75:  88%|████████▊ | 28/32 [02:59<00:26,  6.60s/it]
Test accuracy: 0.78:  88%|████████▊ | 28/32 [03:00<00:26,  6.60s/it]
Test accuracy: 0.77:  88%|████████▊ | 28/32 [03:00<00:26,  6.60s/it]
Test accuracy: 0.78:  91%|█████████ | 29/32 [03:06<00:19,  6.47s/it]
Test accuracy: 0.79:  91%|█████████ | 29/32 [03:06<00:19,  6.47s/it]
Test accuracy: 0.77:  91%|█████████ | 29/32 [03:06<00:19,  6.46s/it]
Test accuracy: 0.78:  91%|█████████ | 29/32 [03:06<00:19,  6.47s/it]
Test accuracy: 0.75:  91%|█████████ | 29/32 [03:06<00:19,  6.47s/it]
Test accuracy: 0.77:  91%|█████████ | 29/32 [03:06<00:19,  6.47s/it]
Test accuracy: 0.77:  91%|█████████ | 29/32 [03:06<00:19,  6.46s/it]
Test accuracy: 0.75:  91%|█████████ | 29/32 [03:06<00:19,  6.47s/it]
Test accuracy: 0.77:  94%|█████████▍| 30/32 [03:12<00:12,  6.42s/it]
Test accuracy: 0.77:  94%|█████████▍| 30/32 [03:12<00:12,  6.42s/it]
Test accuracy: 0.78:  94%|█████████▍| 30/32 [03:12<00:12,  6.42s/it]
Test accuracy: 0.77:  94%|█████████▍| 30/32 [03:12<00:12,  6.42s/it]
Test accuracy: 0.77:  94%|█████████▍| 30/32 [03:12<00:12,  6.42s/it]
Test accuracy: 0.79:  94%|█████████▍| 30/32 [03:12<00:12,  6.42s/it]
Test accuracy: 0.75:  94%|█████████▍| 30/32 [03:12<00:12,  6.42s/it]
Test accuracy: 0.75:  94%|█████████▍| 30/32 [03:12<00:12,  6.42s/it]
Test accuracy: 0.77:  97%|█████████▋| 31/32 [03:19<00:06,  6.62s/it]
Test accuracy: 0.79:  97%|█████████▋| 31/32 [03:19<00:06,  6.62s/it]
Test accuracy: 0.78:  97%|█████████▋| 31/32 [03:19<00:06,  6.62s/it]
Test accuracy: 0.79:  97%|█████████▋| 31/32 [03:19<00:06,  6.62s/it]
Test accuracy: 0.75:  97%|█████████▋| 31/32 [03:19<00:06,  6.62s/it]
Test accuracy: 0.75:  97%|█████████▋| 31/32 [03:19<00:06,  6.62s/it]
Test accuracy: 0.77:  97%|█████████▋| 31/32 [03:19<00:06,  6.62s/it]
Test accuracy: 0.77:  97%|█████████▋| 31/32 [03:19<00:06,  6.62s/it]
Test accuracy: 0.75: 100%|██████████| 32/32 [03:21<00:00,  5.30s/it]
Test accuracy: 0.75: 100%|██████████| 32/32 [03:21<00:00,  5.30s/it]
Test accuracy: 0.77: 100%|██████████| 32/32 [03:21<00:00,  5.30s/it]
Test accuracy: 0.79: 100%|██████████| 32/32 [03:21<00:00,  5.31s/it]
Test accuracy: 0.76: 100%|██████████| 32/32 [03:21<00:00,  5.30s/it]
Test accuracy: 0.78: 100%|██████████| 32/32 [03:21<00:00,  5.31s/it]
Test accuracy: 0.79: 100%|██████████| 32/32 [03:21<00:00,  5.31s/it]
Test accuracy: 0.78: 100%|██████████| 32/32 [03:21<00:00,  5.31s/it]
Test accuracy: 0.76: 100%|██████████| 32/32 [03:22<00:00,  6.31s/it]
Device 2: Cor=191, CoT=0, Total=250

Test accuracy: 0.79: 100%|██████████| 32/32 [03:22<00:00,  6.31s/it]
Device 3: Cor=197, CoT=0, Total=250

Test accuracy: 0.78: 100%|██████████| 32/32 [03:22<00:00,  6.31s/it]
Device 1: Cor=195, CoT=0, Total=250

Test accuracy: 0.75: 100%|██████████| 32/32 [03:22<00:00,  6.31s/it]
Device 0: Cor=188, CoT=0, Total=250

============================================================
EVALUATION RESULTS (Epoch 11)
============================================================
Config: data_curriculum=ON, thought_curriculum=OFF
thought_stage=10, data_stage=6, n_latent_tokens=10
Training on difficulties: 0-6

Overall accuracy: 771 / 1000 = 0.7710

Per-difficulty accuracy:
  diff=1: 94/100 = 0.9400 (trained)
  diff=2: 98/100 = 0.9800 (trained)
  diff=3: 98/100 = 0.9800 (trained)
  diff=4: 98/100 = 0.9800 (trained)
  diff=5: 91/100 = 0.9100 (trained)
  diff=6: 88/100 = 0.8800 (trained)
  diff=7: 51/100 = 0.5100 (untrained)
  diff=8: 48/100 = 0.4800 (untrained)
  diff=9: 56/100 = 0.5600 (untrained)
  diff=10: 49/100 = 0.4900 (untrained)

Trained (diff 0-6): 567/600 = 0.9450
============================================================

  Staging check: acc_trained=0.9450, acc_diff6=0.8800 (threshold=0.9)
Warning: failed to upload checkpoint_12 to HF: (Request ID: Root=1-6a1b8641-7f57cd053d43b8a82d284d57;a61bedfb-34c8-42c8-84d6-a1ef23569a6b)

403 Forbidden: You need to setup automatic credit recharge in order to upload more data. You can do so at /settings/billing..
Cannot access content at: https://huggingface.co/seyedparsa/thoughtformer-checkpoints.git/info/lfs/objects/batch.
Make sure your token has the correct permissions.

============================================================
EPOCH 12/100 (thought_stage=10, data_stage=6)
============================================================
  thought_stage=10, c_thought=1, max_difficulty=6

Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):  16%|█▌        | 156/1000 [00:00<00:00, 982.76 examples/s]
Map (num_proc=4):  16%|█▌        | 157/1000 [00:00<00:00, 937.92 examples/s]
Map (num_proc=4):  16%|█▌        | 157/1000 [00:00<00:00, 928.83 examples/s]
Map (num_proc=4):  81%|████████▏ | 813/1000 [00:00<00:00, 3510.41 examples/s]
Map (num_proc=4):  82%|████████▏ | 816/1000 [00:00<00:00, 3420.63 examples/s]
Map (num_proc=4):  82%|████████▏ | 816/1000 [00:00<00:00, 3080.06 examples/s]
Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2576.01 examples/s]

Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2486.73 examples/s]

Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2441.77 examples/s]

Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Map (num_proc=4):  15%|█▌        | 154/1000 [00:00<00:01, 510.05 examples/s]
Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Map (num_proc=4):  81%|████████  | 809/1000 [00:00<00:00, 2349.80 examples/s]
Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 1257.53 examples/s]

Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:04, 1943.27 examples/s]
Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:05, 1694.73 examples/s]
Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:04, 1855.95 examples/s]
Filter (num_proc=4):  20%|██        | 2000/10000 [00:00<00:02, 3604.31 examples/s]
Filter (num_proc=4):  20%|██        | 2000/10000 [00:00<00:02, 3224.19 examples/s]
Filter (num_proc=4):  20%|██        | 2000/10000 [00:00<00:02, 3558.02 examples/s]
Filter (num_proc=4):  30%|███       | 3000/10000 [00:00<00:01, 4856.52 examples/s]
Filter (num_proc=4):  30%|███       | 3000/10000 [00:00<00:01, 5014.03 examples/s]
Filter (num_proc=4):  30%|███       | 3000/10000 [00:00<00:01, 4526.12 examples/s]
Filter (num_proc=4):  45%|████▌     | 4500/10000 [00:00<00:00, 7016.48 examples/s]
Filter (num_proc=4):  45%|████▌     | 4500/10000 [00:00<00:00, 6648.82 examples/s]
Filter (num_proc=4):  60%|██████    | 6000/10000 [00:00<00:00, 10184.13 examples/s]
Filter (num_proc=4):  70%|███████   | 7000/10000 [00:00<00:00, 10877.91 examples/s]
Filter (num_proc=4):  75%|███████▌  | 7500/10000 [00:00<00:00, 11265.42 examples/s]
Filter (num_proc=4):  75%|███████▌  | 7500/10000 [00:01<00:00, 11338.15 examples/s]
Filter (num_proc=4):  90%|█████████ | 9000/10000 [00:01<00:00, 12392.19 examples/s]
Filter (num_proc=4):  95%|█████████▌| 9500/10000 [00:01<00:00, 12842.34 examples/s]
Filter (num_proc=4):  95%|█████████▌| 9500/10000 [00:01<00:00, 11987.80 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 7431.13 examples/s]

Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 7264.09 examples/s]

Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 6763.04 examples/s]

Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:05, 1531.30 examples/s]
Filter (num_proc=4):  20%|██        | 2000/10000 [00:00<00:02, 2916.85 examples/s]
Filter (num_proc=4):  30%|███       | 3000/10000 [00:00<00:01, 4261.18 examples/s]
Filter (num_proc=4):  45%|████▌     | 4500/10000 [00:00<00:00, 6386.89 examples/s]
Filter (num_proc=4):  70%|███████   | 7000/10000 [00:01<00:00, 10183.60 examples/s]
Filter (num_proc=4):  90%|█████████ | 9000/10000 [00:01<00:00, 11675.65 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 6025.58 examples/s]

Map (num_proc=4):   0%|          | 0/6000 [00:00<?, ? examples/s]
Map (num_proc=4):   2%|▏         | 137/6000 [00:01<00:51, 114.37 examples/s]
Map (num_proc=4):   9%|▉         | 546/6000 [00:01<00:10, 510.89 examples/s]
Map (num_proc=4):  14%|█▎        | 815/6000 [00:01<00:06, 774.29 examples/s]
Map (num_proc=4):  18%|█▊        | 1091/6000 [00:01<00:06, 783.39 examples/s]
Map (num_proc=4):  29%|██▊       | 1720/6000 [00:01<00:02, 1516.36 examples/s]
Map (num_proc=4):  38%|███▊      | 2287/6000 [00:02<00:01, 2001.51 examples/s]
Map (num_proc=4):  44%|████▍     | 2636/6000 [00:02<00:02, 1441.43 examples/s]
Map (num_proc=4):  52%|█████▏    | 3147/6000 [00:02<00:01, 1940.88 examples/s]
Map (num_proc=4):  60%|██████    | 3600/6000 [00:02<00:01, 2363.58 examples/s]
Map (num_proc=4):  67%|██████▋   | 4000/6000 [00:03<00:01, 1461.15 examples/s]
Map (num_proc=4):  86%|████████▌ | 5133/6000 [00:03<00:00, 2726.43 examples/s]
Map (num_proc=4):  97%|█████████▋| 5818/6000 [00:03<00:00, 2064.62 examples/s]
Map (num_proc=4): 100%|██████████| 6000/6000 [00:04<00:00, 1278.39 examples/s]

Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 1424.85 examples/s]
Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 1348.37 examples/s]
Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 1354.46 examples/s]
Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2737.14 examples/s]

Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2649.26 examples/s]

Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2580.40 examples/s]

Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 761.00 examples/s]
Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 1436.44 examples/s]

Map (num_proc=4):   0%|          | 0/600 [00:00<?, ? examples/s]
Map (num_proc=4):  22%|██▎       | 135/600 [00:00<00:01, 382.40 examples/s]
Map (num_proc=4):  50%|█████     | 300/600 [00:00<00:00, 681.65 examples/s]
Map (num_proc=4): 100%|██████████| 600/600 [00:00<00:00, 1196.08 examples/s]
Map (num_proc=4): 100%|██████████| 600/600 [00:00<00:00, 627.34 examples/s] 

Training Epoch: 13:   0%|          | 0/187 [00:00<?, ?it/s]
Training Epoch: 13:   0%|          | 0/187 [00:00<?, ?it/s]
Training Epoch: 13:   0%|          | 0/187 [00:00<?, ?it/s]
Training Epoch: 13:   0%|          | 0/187 [00:00<?, ?it/s]logging training data
rank 0 | gpu 0 | alloc=4.20 GB | reserved=47.73 GB | max=41.99 GB
rank 1 | gpu 1 | alloc=4.20 GB | reserved=47.74 GB | max=41.98 GB
rank 3 | gpu 3 | alloc=4.20 GB | reserved=47.74 GB | max=41.99 GB
rank 2 | gpu 2 | alloc=4.20 GB | reserved=53.36 GB | max=41.98 GB

Training Epoch: 13/100, batch 0/375 completed (loss: 0.0585:   0%|          | 0/187 [00:03<?, ?it/s]
Training Epoch: 13/100, batch 0/375 completed (loss: 0.0284:   0%|          | 0/187 [00:03<?, ?it/s]
Training Epoch: 13/100, batch 0/375 completed (loss: 0.0199:   0%|          | 0/187 [00:03<?, ?it/s]
Training Epoch: 13/100, batch 0/375 completed (loss: 0.057:   0%|          | 0/187 [00:03<?, ?it/s]
Training Epoch: 13/100, batch 0/375 completed (loss: 0.057:   1%|          | 1/187 [00:06<18:47,  6.06s/it]
Training Epoch: 13/100, batch 0/375 completed (loss: 0.0199:   1%|          | 1/187 [00:06<18:45,  6.05s/it]
Training Epoch: 13/100, batch 0/375 completed (loss: 0.0284:   1%|          | 1/187 [00:06<18:44,  6.05s/it]
Training Epoch: 13/100, batch 0/375 completed (loss: 0.0585:   1%|          | 1/187 [00:06<18:43,  6.04s/it]
Training Epoch: 13/100, batch 1/375 completed (loss: 0.0409:   1%|          | 1/187 [00:06<18:44,  6.05s/it]
Training Epoch: 13/100, batch 1/375 completed (loss: 0.0495:   1%|          | 1/187 [00:06<18:45,  6.05s/it]
Training Epoch: 13/100, batch 1/375 completed (loss: 0.0247:   1%|          | 1/187 [00:06<18:43,  6.04s/it]
Training Epoch: 13/100, batch 1/375 completed (loss: 0.08:   1%|          | 1/187 [00:06<18:47,  6.06s/it] 
Training Epoch: 13/100, batch 2/375 completed (loss: 0.0329:   1%|          | 1/187 [00:09<18:43,  6.04s/it]
Training Epoch: 13/100, batch 2/375 completed (loss: 0.0217:   1%|          | 1/187 [00:09<18:45,  6.05s/it]
Training Epoch: 13/100, batch 2/375 completed (loss: 0.0353:   1%|          | 1/187 [00:09<18:44,  6.05s/it]
Training Epoch: 13/100, batch 2/375 completed (loss: 0.063:   1%|          | 1/187 [00:09<18:47,  6.06s/it]
Training Epoch: 13/100, batch 2/375 completed (loss: 0.063:   1%|          | 2/187 [00:11<18:20,  5.95s/it]
Training Epoch: 13/100, batch 2/375 completed (loss: 0.0353:   1%|          | 2/187 [00:11<18:18,  5.94s/it]
Training Epoch: 13/100, batch 2/375 completed (loss: 0.0217:   1%|          | 2/187 [00:11<18:18,  5.94s/it]
Training Epoch: 13/100, batch 2/375 completed (loss: 0.0329:   1%|          | 2/187 [00:11<18:18,  5.94s/it]
Training Epoch: 13/100, batch 3/375 completed (loss: 0.0212:   1%|          | 2/187 [00:11<18:18,  5.94s/it]
Training Epoch: 13/100, batch 3/375 completed (loss: 0.0361:   1%|          | 2/187 [00:11<18:18,  5.94s/it]
Training Epoch: 13/100, batch 3/375 completed (loss: 0.079:   1%|          | 2/187 [00:11<18:18,  5.94s/it] 
Training Epoch: 13/100, batch 3/375 completed (loss: 0.0535:   1%|          | 2/187 [00:11<18:20,  5.95s/it]
Training Epoch: 13/100, batch 4/375 completed (loss: 0.0353:   1%|          | 2/187 [00:15<18:18,  5.94s/it]
Training Epoch: 13/100, batch 4/375 completed (loss: 0.1181:   1%|          | 2/187 [00:15<18:18,  5.94s/it]
Training Epoch: 13/100, batch 4/375 completed (loss: 0.0936:   1%|          | 2/187 [00:15<18:18,  5.94s/it]
Training Epoch: 13/100, batch 4/375 completed (loss: 0.0231:   1%|          | 2/187 [00:15<18:20,  5.95s/it]
Training Epoch: 13/100, batch 4/375 completed (loss: 0.0231:   2%|▏         | 3/187 [00:17<18:23,  6.00s/it]
Training Epoch: 13/100, batch 4/375 completed (loss: 0.1181:   2%|▏         | 3/187 [00:17<18:22,  5.99s/it]
Training Epoch: 13/100, batch 4/375 completed (loss: 0.0353:   2%|▏         | 3/187 [00:17<18:22,  5.99s/it]
Training Epoch: 13/100, batch 4/375 completed (loss: 0.0936:   2%|▏         | 3/187 [00:17<18:23,  6.00s/it]
Training Epoch: 13/100, batch 5/375 completed (loss: 0.1418:   2%|▏         | 3/187 [00:18<18:22,  5.99s/it]
Training Epoch: 13/100, batch 5/375 completed (loss: 0.032:   2%|▏         | 3/187 [00:18<18:23,  6.00s/it] 
Training Epoch: 13/100, batch 5/375 completed (loss: 0.0224:   2%|▏         | 3/187 [00:18<18:22,  5.99s/it]
Training Epoch: 13/100, batch 5/375 completed (loss: 0.0368:   2%|▏         | 3/187 [00:18<18:23,  6.00s/it]
Training Epoch: 13/100, batch 6/375 completed (loss: 0.0176:   2%|▏         | 3/187 [00:20<18:22,  5.99s/it]
Training Epoch: 13/100, batch 6/375 completed (loss: 0.015:   2%|▏         | 3/187 [00:20<18:22,  5.99s/it] 
Training Epoch: 13/100, batch 6/375 completed (loss: 0.0731:   2%|▏         | 3/187 [00:20<18:23,  6.00s/it]
Training Epoch: 13/100, batch 6/375 completed (loss: 0.0598:   2%|▏         | 3/187 [00:20<18:23,  6.00s/it]
Training Epoch: 13/100, batch 6/375 completed (loss: 0.0598:   2%|▏         | 4/187 [00:23<18:05,  5.93s/it]
Training Epoch: 13/100, batch 6/375 completed (loss: 0.015:   2%|▏         | 4/187 [00:23<18:04,  5.93s/it]
Training Epoch: 13/100, batch 6/375 completed (loss: 0.0731:   2%|▏         | 4/187 [00:23<18:04,  5.93s/it]
Training Epoch: 13/100, batch 6/375 completed (loss: 0.0176:   2%|▏         | 4/187 [00:23<18:04,  5.93s/it]
Training Epoch: 13/100, batch 7/375 completed (loss: 0.0073:   2%|▏         | 4/187 [00:23<18:04,  5.93s/it]
Training Epoch: 13/100, batch 7/375 completed (loss: 0.0265:   2%|▏         | 4/187 [00:23<18:04,  5.93s/it]
Training Epoch: 13/100, batch 7/375 completed (loss: 0.0878:   2%|▏         | 4/187 [00:23<18:04,  5.93s/it]
Training Epoch: 13/100, batch 7/375 completed (loss: 0.1947:   2%|▏         | 4/187 [00:23<18:05,  5.93s/it]
Training Epoch: 13/100, batch 8/375 completed (loss: 0.0489:   2%|▏         | 4/187 [00:26<18:04,  5.93s/it]
Training Epoch: 13/100, batch 8/375 completed (loss: 0.0541:   2%|▏         | 4/187 [00:26<18:04,  5.93s/it]
Training Epoch: 13/100, batch 8/375 completed (loss: 0.0222:   2%|▏         | 4/187 [00:26<18:04,  5.93s/it]
Training Epoch: 13/100, batch 8/375 completed (loss: 0.0322:   2%|▏         | 4/187 [00:26<18:05,  5.93s/it]
Training Epoch: 13/100, batch 8/375 completed (loss: 0.0322:   3%|▎         | 5/187 [00:29<18:05,  5.96s/it]
Training Epoch: 13/100, batch 8/375 completed (loss: 0.0222:   3%|▎         | 5/187 [00:29<18:05,  5.96s/it]
Training Epoch: 13/100, batch 8/375 completed (loss: 0.0489:   3%|▎         | 5/187 [00:29<18:05,  5.97s/it]
Training Epoch: 13/100, batch 8/375 completed (loss: 0.0541:   3%|▎         | 5/187 [00:29<18:05,  5.97s/it]
Training Epoch: 13/100, batch 9/375 completed (loss: 0.0099:   3%|▎         | 5/187 [00:29<18:05,  5.97s/it]
Training Epoch: 13/100, batch 9/375 completed (loss: 0.0877:   3%|▎         | 5/187 [00:29<18:05,  5.97s/it]
Training Epoch: 13/100, batch 9/375 completed (loss: 0.0214:   3%|▎         | 5/187 [00:29<18:05,  5.96s/it]
Training Epoch: 13/100, batch 9/375 completed (loss: 0.0274:   3%|▎         | 5/187 [00:29<18:05,  5.96s/it]
Training Epoch: 13/100, batch 10/375 completed (loss: 0.0108:   3%|▎         | 5/187 [00:32<18:05,  5.97s/it]
Training Epoch: 13/100, batch 10/375 completed (loss: 0.0796:   3%|▎         | 5/187 [00:32<18:05,  5.97s/it]
Training Epoch: 13/100, batch 10/375 completed (loss: 0.0206:   3%|▎         | 5/187 [00:32<18:05,  5.96s/it]
Training Epoch: 13/100, batch 10/375 completed (loss: 0.0168:   3%|▎         | 5/187 [00:32<18:05,  5.96s/it]
Training Epoch: 13/100, batch 10/375 completed (loss: 0.0796:   3%|▎         | 6/187 [00:35<17:40,  5.86s/it]
Training Epoch: 13/100, batch 10/375 completed (loss: 0.0206:   3%|▎         | 6/187 [00:35<17:41,  5.86s/it]
Training Epoch: 13/100, batch 10/375 completed (loss: 0.0168:   3%|▎         | 6/187 [00:35<17:42,  5.87s/it]
Training Epoch: 13/100, batch 10/375 completed (loss: 0.0108:   3%|▎         | 6/187 [00:35<17:41,  5.87s/it]
Training Epoch: 13/100, batch 11/375 completed (loss: 0.034:   3%|▎         | 6/187 [00:35<17:41,  5.86s/it] 
Training Epoch: 13/100, batch 11/375 completed (loss: 0.0387:   3%|▎         | 6/187 [00:35<17:41,  5.87s/it]
Training Epoch: 13/100, batch 11/375 completed (loss: 0.081:   3%|▎         | 6/187 [00:35<17:40,  5.86s/it] 
Training Epoch: 13/100, batch 11/375 completed (loss: 0.0686:   3%|▎         | 6/187 [00:35<17:42,  5.87s/it]
Training Epoch: 13/100, batch 12/375 completed (loss: 0.0171:   3%|▎         | 6/187 [00:38<17:40,  5.86s/it]
Training Epoch: 13/100, batch 12/375 completed (loss: 0.0292:   3%|▎         | 6/187 [00:38<17:41,  5.86s/it]
Training Epoch: 13/100, batch 12/375 completed (loss: 0.0783:   3%|▎         | 6/187 [00:38<17:41,  5.87s/it]
Training Epoch: 13/100, batch 12/375 completed (loss: 0.0076:   3%|▎         | 6/187 [00:38<17:42,  5.87s/it]
Training Epoch: 13/100, batch 12/375 completed (loss: 0.0076:   4%|▎         | 7/187 [00:41<17:34,  5.86s/it]
Training Epoch: 13/100, batch 12/375 completed (loss: 0.0783:   4%|▎         | 7/187 [00:41<17:33,  5.85s/it]
Training Epoch: 13/100, batch 12/375 completed (loss: 0.0171:   4%|▎         | 7/187 [00:41<17:33,  5.85s/it]
Training Epoch: 13/100, batch 12/375 completed (loss: 0.0292:   4%|▎         | 7/187 [00:41<17:33,  5.86s/it]
Training Epoch: 13/100, batch 13/375 completed (loss: 0.0554:   4%|▎         | 7/187 [00:41<17:33,  5.86s/it]
Training Epoch: 13/100, batch 13/375 completed (loss: 0.0228:   4%|▎         | 7/187 [00:41<17:33,  5.85s/it]
Training Epoch: 13/100, batch 13/375 completed (loss: 0.1236:   4%|▎         | 7/187 [00:41<17:33,  5.85s/it]
Training Epoch: 13/100, batch 13/375 completed (loss: 0.0163:   4%|▎         | 7/187 [00:41<17:34,  5.86s/it]
Training Epoch: 13/100, batch 14/375 completed (loss: 0.064:   4%|▎         | 7/187 [00:44<17:33,  5.85s/it] 
Training Epoch: 13/100, batch 14/375 completed (loss: 0.1421:   4%|▎         | 7/187 [00:44<17:33,  5.85s/it]
Training Epoch: 13/100, batch 14/375 completed (loss: 0.0438:   4%|▎         | 7/187 [00:44<17:33,  5.86s/it]
Training Epoch: 13/100, batch 14/375 completed (loss: 0.0478:   4%|▎         | 7/187 [00:44<17:34,  5.86s/it]
Training Epoch: 13/100, batch 14/375 completed (loss: 0.0478:   4%|▍         | 8/187 [00:47<17:42,  5.94s/it]
Training Epoch: 13/100, batch 14/375 completed (loss: 0.1421:   4%|▍         | 8/187 [00:47<17:42,  5.94s/it]
Training Epoch: 13/100, batch 14/375 completed (loss: 0.0438:   4%|▍         | 8/187 [00:47<17:42,  5.94s/it]
Training Epoch: 13/100, batch 14/375 completed (loss: 0.064:   4%|▍         | 8/187 [00:47<17:42,  5.94s/it]
Training Epoch: 13/100, batch 15/375 completed (loss: 0.0266:   4%|▍         | 8/187 [00:47<17:42,  5.94s/it]
Training Epoch: 13/100, batch 15/375 completed (loss: 0.0542:   4%|▍         | 8/187 [00:47<17:42,  5.94s/it]
Training Epoch: 13/100, batch 15/375 completed (loss: 0.0322:   4%|▍         | 8/187 [00:47<17:42,  5.94s/it]
Training Epoch: 13/100, batch 15/375 completed (loss: 0.0496:   4%|▍         | 8/187 [00:47<17:42,  5.94s/it]
Training Epoch: 13/100, batch 16/375 completed (loss: 0.0891:   4%|▍         | 8/187 [00:50<17:42,  5.94s/it]
Training Epoch: 13/100, batch 16/375 completed (loss: 0.0603:   4%|▍         | 8/187 [00:50<17:42,  5.94s/it]
Training Epoch: 13/100, batch 16/375 completed (loss: 0.1194:   4%|▍         | 8/187 [00:50<17:42,  5.94s/it]
Training Epoch: 13/100, batch 16/375 completed (loss: 0.0306:   4%|▍         | 8/187 [00:50<17:42,  5.94s/it]
Training Epoch: 13/100, batch 16/375 completed (loss: 0.0306:   5%|▍         | 9/187 [00:53<17:23,  5.86s/it]
Training Epoch: 13/100, batch 16/375 completed (loss: 0.1194:   5%|▍         | 9/187 [00:53<17:24,  5.87s/it]
Training Epoch: 13/100, batch 16/375 completed (loss: 0.0891:   5%|▍         | 9/187 [00:53<17:24,  5.87s/it]
Training Epoch: 13/100, batch 16/375 completed (loss: 0.0603:   5%|▍         | 9/187 [00:53<17:25,  5.87s/it]
Training Epoch: 13/100, batch 17/375 completed (loss: 0.0909:   5%|▍         | 9/187 [00:53<17:24,  5.87s/it]
Training Epoch: 13/100, batch 17/375 completed (loss: 0.0588:   5%|▍         | 9/187 [00:53<17:24,  5.87s/it]
Training Epoch: 13/100, batch 17/375 completed (loss: 0.0454:   5%|▍         | 9/187 [00:53<17:25,  5.87s/it]
Training Epoch: 13/100, batch 17/375 completed (loss: 0.1032:   5%|▍         | 9/187 [00:53<17:23,  5.86s/it]
Training Epoch: 13/100, batch 18/375 completed (loss: 0.0328:   5%|▍         | 9/187 [00:56<17:24,  5.87s/it]
Training Epoch: 13/100, batch 18/375 completed (loss: 0.0428:   5%|▍         | 9/187 [00:56<17:24,  5.87s/it]
Training Epoch: 13/100, batch 18/375 completed (loss: 0.0825:   5%|▍         | 9/187 [00:56<17:25,  5.87s/it]
Training Epoch: 13/100, batch 18/375 completed (loss: 0.0217:   5%|▍         | 9/187 [00:56<17:23,  5.86s/it]
Training Epoch: 13/100, batch 18/375 completed (loss: 0.0825:   5%|▌         | 10/187 [00:59<17:43,  6.01s/it]
Training Epoch: 13/100, batch 18/375 completed (loss: 0.0428:   5%|▌         | 10/187 [00:59<17:43,  6.01s/it]
Training Epoch: 13/100, batch 18/375 completed (loss: 0.0217:   5%|▌         | 10/187 [00:59<17:43,  6.01s/it]
Training Epoch: 13/100, batch 18/375 completed (loss: 0.0328:   5%|▌         | 10/187 [00:59<17:43,  6.01s/it]
Training Epoch: 13/100, batch 19/375 completed (loss: 0.1065:   5%|▌         | 10/187 [00:59<17:43,  6.01s/it]
Training Epoch: 13/100, batch 19/375 completed (loss: 0.0303:   5%|▌         | 10/187 [00:59<17:43,  6.01s/it]
Training Epoch: 13/100, batch 19/375 completed (loss: 0.0215:   5%|▌         | 10/187 [00:59<17:43,  6.01s/it]
Training Epoch: 13/100, batch 19/375 completed (loss: 0.0125:   5%|▌         | 10/187 [00:59<17:43,  6.01s/it]
Training Epoch: 13/100, batch 20/375 completed (loss: 0.0369:   5%|▌         | 10/187 [01:02<17:43,  6.01s/it]
Training Epoch: 13/100, batch 20/375 completed (loss: 0.0611:   5%|▌         | 10/187 [01:02<17:43,  6.01s/it]
Training Epoch: 13/100, batch 20/375 completed (loss: 0.1581:   5%|▌         | 10/187 [01:02<17:43,  6.01s/it]
Training Epoch: 13/100, batch 20/375 completed (loss: 0.062:   5%|▌         | 10/187 [01:02<17:43,  6.01s/it] 
Training Epoch: 13/100, batch 20/375 completed (loss: 0.0611:   6%|▌         | 11/187 [01:05<17:20,  5.91s/it]
Training Epoch: 13/100, batch 20/375 completed (loss: 0.0369:   6%|▌         | 11/187 [01:05<17:20,  5.91s/it]
Training Epoch: 13/100, batch 20/375 completed (loss: 0.1581:   6%|▌         | 11/187 [01:05<17:20,  5.91s/it]
Training Epoch: 13/100, batch 20/375 completed (loss: 0.062:   6%|▌         | 11/187 [01:05<17:21,  5.92s/it]
Training Epoch: 13/100, batch 21/375 completed (loss: 0.0064:   6%|▌         | 11/187 [01:05<17:20,  5.91s/it]
Training Epoch: 13/100, batch 21/375 completed (loss: 0.1069:   6%|▌         | 11/187 [01:05<17:20,  5.91s/it]
Training Epoch: 13/100, batch 21/375 completed (loss: 0.0234:   6%|▌         | 11/187 [01:05<17:20,  5.91s/it]
Training Epoch: 13/100, batch 21/375 completed (loss: 0.0095:   6%|▌         | 11/187 [01:05<17:21,  5.92s/it]
Training Epoch: 13/100, batch 22/375 completed (loss: 0.0301:   6%|▌         | 11/187 [01:08<17:20,  5.91s/it]
Training Epoch: 13/100, batch 22/375 completed (loss: 0.0603:   6%|▌         | 11/187 [01:08<17:20,  5.91s/it]
Training Epoch: 13/100, batch 22/375 completed (loss: 0.0118:   6%|▌         | 11/187 [01:08<17:20,  5.91s/it]
Training Epoch: 13/100, batch 22/375 completed (loss: 0.0632:   6%|▌         | 11/187 [01:08<17:21,  5.92s/it]
Training Epoch: 13/100, batch 22/375 completed (loss: 0.0632:   6%|▋         | 12/187 [01:11<17:11,  5.90s/it]
Training Epoch: 13/100, batch 22/375 completed (loss: 0.0301:   6%|▋         | 12/187 [01:11<17:12,  5.90s/it]
Training Epoch: 13/100, batch 22/375 completed (loss: 0.0603:   6%|▋         | 12/187 [01:11<17:12,  5.90s/it]
Training Epoch: 13/100, batch 22/375 completed (loss: 0.0118:   6%|▋         | 12/187 [01:11<17:12,  5.90s/it]
Training Epoch: 13/100, batch 23/375 completed (loss: 0.075:   6%|▋         | 12/187 [01:11<17:12,  5.90s/it] 
Training Epoch: 13/100, batch 23/375 completed (loss: 0.0419:   6%|▋         | 12/187 [01:11<17:12,  5.90s/it]
Training Epoch: 13/100, batch 23/375 completed (loss: 0.0758:   6%|▋         | 12/187 [01:11<17:12,  5.90s/it]
Training Epoch: 13/100, batch 23/375 completed (loss: 0.0836:   6%|▋         | 12/187 [01:11<17:11,  5.90s/it]
Training Epoch: 13/100, batch 24/375 completed (loss: 0.033:   6%|▋         | 12/187 [01:14<17:12,  5.90s/it]
Training Epoch: 13/100, batch 24/375 completed (loss: 0.0985:   6%|▋         | 12/187 [01:14<17:12,  5.90s/it]
Training Epoch: 13/100, batch 24/375 completed (loss: 0.0703:   6%|▋         | 12/187 [01:14<17:12,  5.90s/it]
Training Epoch: 13/100, batch 24/375 completed (loss: 0.0381:   6%|▋         | 12/187 [01:14<17:11,  5.90s/it]
Training Epoch: 13/100, batch 24/375 completed (loss: 0.033:   7%|▋         | 13/187 [01:17<17:19,  5.98s/it]
Training Epoch: 13/100, batch 24/375 completed (loss: 0.0381:   7%|▋         | 13/187 [01:17<17:20,  5.98s/it]
Training Epoch: 13/100, batch 24/375 completed (loss: 0.0703:   7%|▋         | 13/187 [01:17<17:20,  5.98s/it]
Training Epoch: 13/100, batch 24/375 completed (loss: 0.0985:   7%|▋         | 13/187 [01:17<17:20,  5.98s/it]
Training Epoch: 13/100, batch 25/375 completed (loss: 0.0621:   7%|▋         | 13/187 [01:17<17:19,  5.98s/it]
Training Epoch: 13/100, batch 25/375 completed (loss: 0.0639:   7%|▋         | 13/187 [01:17<17:20,  5.98s/it]
Training Epoch: 13/100, batch 25/375 completed (loss: 0.0156:   7%|▋         | 13/187 [01:17<17:20,  5.98s/it]
Training Epoch: 13/100, batch 25/375 completed (loss: 0.0214:   7%|▋         | 13/187 [01:17<17:20,  5.98s/it]
Training Epoch: 13/100, batch 26/375 completed (loss: 0.057:   7%|▋         | 13/187 [01:20<17:19,  5.98s/it] 
Training Epoch: 13/100, batch 26/375 completed (loss: 0.0167:   7%|▋         | 13/187 [01:20<17:20,  5.98s/it]
Training Epoch: 13/100, batch 26/375 completed (loss: 0.0331:   7%|▋         | 13/187 [01:20<17:20,  5.98s/it]
Training Epoch: 13/100, batch 26/375 completed (loss: 0.012:   7%|▋         | 13/187 [01:20<17:20,  5.98s/it] 
Training Epoch: 13/100, batch 26/375 completed (loss: 0.012:   7%|▋         | 14/187 [01:22<17:03,  5.91s/it]
Training Epoch: 13/100, batch 26/375 completed (loss: 0.0331:   7%|▋         | 14/187 [01:22<17:03,  5.92s/it]
Training Epoch: 13/100, batch 26/375 completed (loss: 0.057:   7%|▋         | 14/187 [01:22<17:03,  5.92s/it]
Training Epoch: 13/100, batch 26/375 completed (loss: 0.0167:   7%|▋         | 14/187 [01:22<17:03,  5.92s/it]
Training Epoch: 13/100, batch 27/375 completed (loss: 0.0365:   7%|▋         | 14/187 [01:23<17:03,  5.92s/it]
Training Epoch: 13/100, batch 27/375 completed (loss: 0.0045:   7%|▋         | 14/187 [01:23<17:03,  5.92s/it]
Training Epoch: 13/100, batch 27/375 completed (loss: 0.0222:   7%|▋         | 14/187 [01:23<17:03,  5.92s/it]
Training Epoch: 13/100, batch 27/375 completed (loss: 0.0138:   7%|▋         | 14/187 [01:23<17:03,  5.91s/it]
Training Epoch: 13/100, batch 28/375 completed (loss: 0.0179:   7%|▋         | 14/187 [01:25<17:03,  5.92s/it]
Training Epoch: 13/100, batch 28/375 completed (loss: 0.0162:   7%|▋         | 14/187 [01:25<17:03,  5.92s/it]
Training Epoch: 13/100, batch 28/375 completed (loss: 0.0512:   7%|▋         | 14/187 [01:25<17:03,  5.92s/it]
Training Epoch: 13/100, batch 28/375 completed (loss: 0.0354:   7%|▋         | 14/187 [01:25<17:03,  5.91s/it]
Training Epoch: 13/100, batch 28/375 completed (loss: 0.0162:   8%|▊         | 15/187 [01:29<17:04,  5.96s/it]
Training Epoch: 13/100, batch 28/375 completed (loss: 0.0354:   8%|▊         | 15/187 [01:29<17:04,  5.96s/it]
Training Epoch: 13/100, batch 28/375 completed (loss: 0.0179:   8%|▊         | 15/187 [01:29<17:05,  5.96s/it]
Training Epoch: 13/100, batch 28/375 completed (loss: 0.0512:   8%|▊         | 15/187 [01:29<17:04,  5.96s/it]
Training Epoch: 13/100, batch 29/375 completed (loss: 0.0551:   8%|▊         | 15/187 [01:29<17:04,  5.96s/it]
Training Epoch: 13/100, batch 29/375 completed (loss: 0.015:   8%|▊         | 15/187 [01:29<17:05,  5.96s/it] 
Training Epoch: 13/100, batch 29/375 completed (loss: 0.2037:   8%|▊         | 15/187 [01:29<17:04,  5.96s/it]
Training Epoch: 13/100, batch 29/375 completed (loss: 0.0035:   8%|▊         | 15/187 [01:29<17:04,  5.96s/it]
Training Epoch: 13/100, batch 30/375 completed (loss: 0.02:   8%|▊         | 15/187 [01:32<17:05,  5.96s/it] 
Training Epoch: 13/100, batch 30/375 completed (loss: 0.0109:   8%|▊         | 15/187 [01:32<17:04,  5.96s/it]
Training Epoch: 13/100, batch 30/375 completed (loss: 0.0142:   8%|▊         | 15/187 [01:32<17:04,  5.96s/it]
Training Epoch: 13/100, batch 30/375 completed (loss: 0.0343:   8%|▊         | 15/187 [01:32<17:04,  5.96s/it]
Training Epoch: 13/100, batch 30/375 completed (loss: 0.0109:   9%|▊         | 16/187 [01:35<17:03,  5.99s/it]
Training Epoch: 13/100, batch 30/375 completed (loss: 0.0142:   9%|▊         | 16/187 [01:35<17:03,  5.99s/it]
Training Epoch: 13/100, batch 30/375 completed (loss: 0.02:   9%|▊         | 16/187 [01:35<17:04,  5.99s/it]
Training Epoch: 13/100, batch 30/375 completed (loss: 0.0343:   9%|▊         | 16/187 [01:35<17:04,  5.99s/it]
Training Epoch: 13/100, batch 31/375 completed (loss: 0.0062:   9%|▊         | 16/187 [01:35<17:04,  5.99s/it]
Training Epoch: 13/100, batch 31/375 completed (loss: 0.0288:   9%|▊         | 16/187 [01:35<17:03,  5.99s/it]
Training Epoch: 13/100, batch 31/375 completed (loss: 0.0115:   9%|▊         | 16/187 [01:35<17:03,  5.99s/it]
Training Epoch: 13/100, batch 31/375 completed (loss: 0.0962:   9%|▊         | 16/187 [01:35<17:04,  5.99s/it]
Training Epoch: 13/100, batch 32/375 completed (loss: 0.0026:   9%|▊         | 16/187 [01:38<17:03,  5.99s/it]
Training Epoch: 13/100, batch 32/375 completed (loss: 0.0883:   9%|▊         | 16/187 [01:38<17:04,  5.99s/it]
Training Epoch: 13/100, batch 32/375 completed (loss: 0.0548:   9%|▊         | 16/187 [01:38<17:03,  5.99s/it]
Training Epoch: 13/100, batch 32/375 completed (loss: 0.041:   9%|▊         | 16/187 [01:38<17:04,  5.99s/it] 
Training Epoch: 13/100, batch 32/375 completed (loss: 0.041:   9%|▉         | 17/187 [01:40<16:46,  5.92s/it]
Training Epoch: 13/100, batch 32/375 completed (loss: 0.0026:   9%|▉         | 17/187 [01:40<16:47,  5.92s/it]
Training Epoch: 13/100, batch 32/375 completed (loss: 0.0883:   9%|▉         | 17/187 [01:40<16:46,  5.92s/it]
Training Epoch: 13/100, batch 32/375 completed (loss: 0.0548:   9%|▉         | 17/187 [01:40<16:47,  5.92s/it]
Training Epoch: 13/100, batch 33/375 completed (loss: 0.0888:   9%|▉         | 17/187 [01:40<16:46,  5.92s/it]
Training Epoch: 13/100, batch 33/375 completed (loss: 0.0443:   9%|▉         | 17/187 [01:40<16:47,  5.92s/it]
Training Epoch: 13/100, batch 33/375 completed (loss: 0.0147:   9%|▉         | 17/187 [01:40<16:47,  5.92s/it]
Training Epoch: 13/100, batch 33/375 completed (loss: 0.0176:   9%|▉         | 17/187 [01:40<16:46,  5.92s/it]
Training Epoch: 13/100, batch 34/375 completed (loss: 0.0044:   9%|▉         | 17/187 [01:43<16:47,  5.92s/it]
Training Epoch: 13/100, batch 34/375 completed (loss: 0.0133:   9%|▉         | 17/187 [01:43<16:47,  5.92s/it]
Training Epoch: 13/100, batch 34/375 completed (loss: 0.0314:   9%|▉         | 17/187 [01:43<16:46,  5.92s/it]
Training Epoch: 13/100, batch 34/375 completed (loss: 0.0298:   9%|▉         | 17/187 [01:43<16:46,  5.92s/it]
Training Epoch: 13/100, batch 34/375 completed (loss: 0.0044:  10%|▉         | 18/187 [01:47<17:04,  6.06s/it]
Training Epoch: 13/100, batch 34/375 completed (loss: 0.0314:  10%|▉         | 18/187 [01:47<17:04,  6.06s/it]
Training Epoch: 13/100, batch 34/375 completed (loss: 0.0298:  10%|▉         | 18/187 [01:47<17:04,  6.06s/it]
Training Epoch: 13/100, batch 34/375 completed (loss: 0.0133:  10%|▉         | 18/187 [01:47<17:04,  6.06s/it]
Training Epoch: 13/100, batch 35/375 completed (loss: 0.0587:  10%|▉         | 18/187 [01:47<17:04,  6.06s/it]
Training Epoch: 13/100, batch 35/375 completed (loss: 0.0585:  10%|▉         | 18/187 [01:47<17:04,  6.06s/it]
Training Epoch: 13/100, batch 35/375 completed (loss: 0.0174:  10%|▉         | 18/187 [01:47<17:04,  6.06s/it]
Training Epoch: 13/100, batch 35/375 completed (loss: 0.0049:  10%|▉         | 18/187 [01:47<17:04,  6.06s/it]
Training Epoch: 13/100, batch 36/375 completed (loss: 0.1014:  10%|▉         | 18/187 [01:50<17:04,  6.06s/it]
Training Epoch: 13/100, batch 36/375 completed (loss: 0.0186:  10%|▉         | 18/187 [01:50<17:04,  6.06s/it]
Training Epoch: 13/100, batch 36/375 completed (loss: 0.0032:  10%|▉         | 18/187 [01:50<17:04,  6.06s/it]
Training Epoch: 13/100, batch 36/375 completed (loss: 0.0025:  10%|▉         | 18/187 [01:50<17:04,  6.06s/it]
Training Epoch: 13/100, batch 36/375 completed (loss: 0.0025:  10%|█         | 19/187 [01:53<16:43,  5.97s/it]
Training Epoch: 13/100, batch 36/375 completed (loss: 0.0186:  10%|█         | 19/187 [01:53<16:43,  5.97s/it]
Training Epoch: 13/100, batch 36/375 completed (loss: 0.0032:  10%|█         | 19/187 [01:53<16:43,  5.97s/it]
Training Epoch: 13/100, batch 36/375 completed (loss: 0.1014:  10%|█         | 19/187 [01:53<16:44,  5.98s/it]
Training Epoch: 13/100, batch 37/375 completed (loss: 0.0098:  10%|█         | 19/187 [01:53<16:44,  5.98s/it]
Training Epoch: 13/100, batch 37/375 completed (loss: 0.0064:  10%|█         | 19/187 [01:53<16:43,  5.97s/it]
Training Epoch: 13/100, batch 37/375 completed (loss: 0.0266:  10%|█         | 19/187 [01:53<16:43,  5.97s/it]
Training Epoch: 13/100, batch 37/375 completed (loss: 0.0245:  10%|█         | 19/187 [01:53<16:43,  5.97s/it]
Training Epoch: 13/100, batch 38/375 completed (loss: 0.0047:  10%|█         | 19/187 [01:55<16:43,  5.97s/it]
Training Epoch: 13/100, batch 38/375 completed (loss: 0.0142:  10%|█         | 19/187 [01:55<16:44,  5.98s/it]
Training Epoch: 13/100, batch 38/375 completed (loss: 0.0196:  10%|█         | 19/187 [01:55<16:43,  5.97s/it]
Training Epoch: 13/100, batch 38/375 completed (loss: 0.0716:  10%|█         | 19/187 [01:56<16:43,  5.97s/it]
Training Epoch: 13/100, batch 38/375 completed (loss: 0.0142:  11%|█         | 20/187 [01:58<16:29,  5.93s/it]
Training Epoch: 13/100, batch 38/375 completed (loss: 0.0716:  11%|█         | 20/187 [01:58<16:30,  5.93s/it]
Training Epoch: 13/100, batch 38/375 completed (loss: 0.0196:  11%|█         | 20/187 [01:58<16:30,  5.93s/it]
Training Epoch: 13/100, batch 38/375 completed (loss: 0.0047:  11%|█         | 20/187 [01:58<16:30,  5.93s/it]
Training Epoch: 13/100, batch 39/375 completed (loss: 0.1012:  11%|█         | 20/187 [01:58<16:30,  5.93s/it]
Training Epoch: 13/100, batch 39/375 completed (loss: 0.0161:  11%|█         | 20/187 [01:58<16:30,  5.93s/it]
Training Epoch: 13/100, batch 39/375 completed (loss: 0.0606:  11%|█         | 20/187 [01:58<16:29,  5.93s/it]
Training Epoch: 13/100, batch 39/375 completed (loss: 0.012:  11%|█         | 20/187 [01:58<16:30,  5.93s/it] 
Training Epoch: 13/100, batch 40/375 completed (loss: 0.1118:  11%|█         | 20/187 [02:02<16:30,  5.93s/it]
Training Epoch: 13/100, batch 40/375 completed (loss: 0.0771:  11%|█         | 20/187 [02:02<16:29,  5.93s/it]
Training Epoch: 13/100, batch 40/375 completed (loss: 0.0307:  11%|█         | 20/187 [02:02<16:30,  5.93s/it]
Training Epoch: 13/100, batch 40/375 completed (loss: 0.051:  11%|█         | 20/187 [02:02<16:30,  5.93s/it]
Training Epoch: 13/100, batch 40/375 completed (loss: 0.0771:  11%|█         | 21/187 [02:05<16:40,  6.03s/it]
Training Epoch: 13/100, batch 40/375 completed (loss: 0.051:  11%|█         | 21/187 [02:05<16:40,  6.03s/it]
Training Epoch: 13/100, batch 40/375 completed (loss: 0.1118:  11%|█         | 21/187 [02:05<16:41,  6.03s/it]
Training Epoch: 13/100, batch 40/375 completed (loss: 0.0307:  11%|█         | 21/187 [02:05<16:40,  6.03s/it]
Training Epoch: 13/100, batch 41/375 completed (loss: 0.0231:  11%|█         | 21/187 [02:05<16:40,  6.03s/it]
Training Epoch: 13/100, batch 41/375 completed (loss: 0.0149:  11%|█         | 21/187 [02:05<16:40,  6.03s/it]
Training Epoch: 13/100, batch 41/375 completed (loss: 0.026:  11%|█         | 21/187 [02:05<16:41,  6.03s/it] 
Training Epoch: 13/100, batch 41/375 completed (loss: 0.0096:  11%|█         | 21/187 [02:05<16:40,  6.03s/it]
Training Epoch: 13/100, batch 42/375 completed (loss: 0.0191:  11%|█         | 21/187 [02:08<16:40,  6.03s/it]
Training Epoch: 13/100, batch 42/375 completed (loss: 0.0165:  11%|█         | 21/187 [02:08<16:41,  6.03s/it]
Training Epoch: 13/100, batch 42/375 completed (loss: 0.0524:  11%|█         | 21/187 [02:08<16:40,  6.03s/it]
Training Epoch: 13/100, batch 42/375 completed (loss: 0.0586:  11%|█         | 21/187 [02:08<16:40,  6.03s/it]
Training Epoch: 13/100, batch 42/375 completed (loss: 0.0191:  12%|█▏        | 22/187 [02:10<16:23,  5.96s/it]
Training Epoch: 13/100, batch 42/375 completed (loss: 0.0586:  12%|█▏        | 22/187 [02:10<16:23,  5.96s/it]
Training Epoch: 13/100, batch 42/375 completed (loss: 0.0524:  12%|█▏        | 22/187 [02:10<16:23,  5.96s/it]
Training Epoch: 13/100, batch 42/375 completed (loss: 0.0165:  12%|█▏        | 22/187 [02:10<16:24,  5.96s/it]
Training Epoch: 13/100, batch 43/375 completed (loss: 0.0613:  12%|█▏        | 22/187 [02:10<16:23,  5.96s/it]
Training Epoch: 13/100, batch 43/375 completed (loss: 0.0502:  12%|█▏        | 22/187 [02:10<16:23,  5.96s/it]
Training Epoch: 13/100, batch 43/375 completed (loss: 0.0103:  12%|█▏        | 22/187 [02:10<16:24,  5.96s/it]
Training Epoch: 13/100, batch 43/375 completed (loss: 0.0264:  12%|█▏        | 22/187 [02:10<16:23,  5.96s/it]
Training Epoch: 13/100, batch 44/375 completed (loss: 0.0124:  12%|█▏        | 22/187 [02:13<16:24,  5.96s/it]
Training Epoch: 13/100, batch 44/375 completed (loss: 0.0135:  12%|█▏        | 22/187 [02:13<16:23,  5.96s/it]
Training Epoch: 13/100, batch 44/375 completed (loss: 0.0089:  12%|█▏        | 22/187 [02:13<16:23,  5.96s/it]
Training Epoch: 13/100, batch 44/375 completed (loss: 0.0384:  12%|█▏        | 22/187 [02:13<16:23,  5.96s/it]
Training Epoch: 13/100, batch 44/375 completed (loss: 0.0384:  12%|█▏        | 23/187 [02:17<16:32,  6.05s/it]
Training Epoch: 13/100, batch 44/375 completed (loss: 0.0135:  12%|█▏        | 23/187 [02:17<16:32,  6.05s/it]
Training Epoch: 13/100, batch 44/375 completed (loss: 0.0089:  12%|█▏        | 23/187 [02:17<16:32,  6.05s/it]
Training Epoch: 13/100, batch 44/375 completed (loss: 0.0124:  12%|█▏        | 23/187 [02:17<16:32,  6.05s/it]
Training Epoch: 13/100, batch 45/375 completed (loss: 0.0076:  12%|█▏        | 23/187 [02:17<16:32,  6.05s/it]
Training Epoch: 13/100, batch 45/375 completed (loss: 0.0148:  12%|█▏        | 23/187 [02:17<16:32,  6.05s/it]
Training Epoch: 13/100, batch 45/375 completed (loss: 0.0496:  12%|█▏        | 23/187 [02:17<16:32,  6.05s/it]
Training Epoch: 13/100, batch 45/375 completed (loss: 0.2148:  12%|█▏        | 23/187 [02:17<16:32,  6.05s/it]
Training Epoch: 13/100, batch 46/375 completed (loss: 0.0513:  12%|█▏        | 23/187 [02:20<16:32,  6.05s/it]
Training Epoch: 13/100, batch 46/375 completed (loss: 0.0106:  12%|█▏        | 23/187 [02:20<16:32,  6.05s/it]
Training Epoch: 13/100, batch 46/375 completed (loss: 0.0087:  12%|█▏        | 23/187 [02:20<16:32,  6.05s/it]
Training Epoch: 13/100, batch 46/375 completed (loss: 0.0103:  12%|█▏        | 23/187 [02:20<16:32,  6.05s/it]
Training Epoch: 13/100, batch 46/375 completed (loss: 0.0103:  13%|█▎        | 24/187 [02:23<16:15,  5.99s/it]
Training Epoch: 13/100, batch 46/375 completed (loss: 0.0087:  13%|█▎        | 24/187 [02:23<16:16,  5.99s/it]
Training Epoch: 13/100, batch 46/375 completed (loss: 0.0106:  13%|█▎        | 24/187 [02:23<16:16,  5.99s/it]
Training Epoch: 13/100, batch 46/375 completed (loss: 0.0513:  13%|█▎        | 24/187 [02:23<16:16,  5.99s/it]
Training Epoch: 13/100, batch 47/375 completed (loss: 0.2315:  13%|█▎        | 24/187 [02:23<16:16,  5.99s/it]
Training Epoch: 13/100, batch 47/375 completed (loss: 0.0108:  13%|█▎        | 24/187 [02:23<16:16,  5.99s/it]
Training Epoch: 13/100, batch 47/375 completed (loss: 0.0026:  13%|█▎        | 24/187 [02:23<16:16,  5.99s/it]
Training Epoch: 13/100, batch 47/375 completed (loss: 0.0564:  13%|█▎        | 24/187 [02:23<16:15,  5.99s/it]
Training Epoch: 13/100, batch 48/375 completed (loss: 0.0313:  13%|█▎        | 24/187 [02:25<16:16,  5.99s/it]
Training Epoch: 13/100, batch 48/375 completed (loss: 0.0627:  13%|█▎        | 24/187 [02:25<16:16,  5.99s/it]
Training Epoch: 13/100, batch 48/375 completed (loss: 0.0872:  13%|█▎        | 24/187 [02:25<16:16,  5.99s/it]
Training Epoch: 13/100, batch 48/375 completed (loss: 0.0311:  13%|█▎        | 24/187 [02:26<16:15,  5.99s/it]
Training Epoch: 13/100, batch 48/375 completed (loss: 0.0313:  13%|█▎        | 25/187 [02:28<16:01,  5.94s/it]
Training Epoch: 13/100, batch 48/375 completed (loss: 0.0872:  13%|█▎        | 25/187 [02:28<16:01,  5.94s/it]
Training Epoch: 13/100, batch 48/375 completed (loss: 0.0311:  13%|█▎        | 25/187 [02:28<16:02,  5.94s/it]
Training Epoch: 13/100, batch 48/375 completed (loss: 0.0627:  13%|█▎        | 25/187 [02:28<16:01,  5.94s/it]
Training Epoch: 13/100, batch 49/375 completed (loss: 0.0144:  13%|█▎        | 25/187 [02:28<16:01,  5.94s/it]
Training Epoch: 13/100, batch 49/375 completed (loss: 0.0208:  13%|█▎        | 25/187 [02:28<16:01,  5.94s/it]
Training Epoch: 13/100, batch 49/375 completed (loss: 0.0041:  13%|█▎        | 25/187 [02:28<16:01,  5.94s/it]
Training Epoch: 13/100, batch 49/375 completed (loss: 0.0125:  13%|█▎        | 25/187 [02:28<16:02,  5.94s/it]
Training Epoch: 13/100, batch 50/375 completed (loss: 0.0473:  13%|█▎        | 25/187 [02:32<16:01,  5.94s/it]
Training Epoch: 13/100, batch 50/375 completed (loss: 0.0068:  13%|█▎        | 25/187 [02:32<16:01,  5.94s/it]
Training Epoch: 13/100, batch 50/375 completed (loss: 0.0156:  13%|█▎        | 25/187 [02:32<16:01,  5.94s/it]
Training Epoch: 13/100, batch 50/375 completed (loss: 0.0054:  13%|█▎        | 25/187 [02:32<16:02,  5.94s/it]
Training Epoch: 13/100, batch 50/375 completed (loss: 0.0054:  14%|█▍        | 26/187 [02:35<16:19,  6.08s/it]
Training Epoch: 13/100, batch 50/375 completed (loss: 0.0473:  14%|█▍        | 26/187 [02:35<16:18,  6.08s/it]
Training Epoch: 13/100, batch 50/375 completed (loss: 0.0156:  14%|█▍        | 26/187 [02:35<16:19,  6.08s/it]
Training Epoch: 13/100, batch 50/375 completed (loss: 0.0068:  14%|█▍        | 26/187 [02:35<16:19,  6.08s/it]
Training Epoch: 13/100, batch 51/375 completed (loss: 0.0178:  14%|█▍        | 26/187 [02:35<16:19,  6.08s/it]
Training Epoch: 13/100, batch 51/375 completed (loss: 0.0123:  14%|█▍        | 26/187 [02:35<16:18,  6.08s/it]
Training Epoch: 13/100, batch 51/375 completed (loss: 0.1841:  14%|█▍        | 26/187 [02:35<16:19,  6.08s/it]
Training Epoch: 13/100, batch 51/375 completed (loss: 0.0423:  14%|█▍        | 26/187 [02:35<16:19,  6.08s/it]
Training Epoch: 13/100, batch 52/375 completed (loss: 0.0311:  14%|█▍        | 26/187 [02:38<16:19,  6.08s/it]
Training Epoch: 13/100, batch 52/375 completed (loss: 0.007:  14%|█▍        | 26/187 [02:38<16:18,  6.08s/it] 
Training Epoch: 13/100, batch 52/375 completed (loss: 0.0312:  14%|█▍        | 26/187 [02:38<16:19,  6.08s/it]
Training Epoch: 13/100, batch 52/375 completed (loss: 0.0189:  14%|█▍        | 26/187 [02:38<16:19,  6.08s/it]
Training Epoch: 13/100, batch 52/375 completed (loss: 0.0189:  14%|█▍        | 27/187 [02:41<15:58,  5.99s/it]
Training Epoch: 13/100, batch 52/375 completed (loss: 0.007:  14%|█▍        | 27/187 [02:41<15:58,  5.99s/it]
Training Epoch: 13/100, batch 52/375 completed (loss: 0.0311:  14%|█▍        | 27/187 [02:41<15:58,  5.99s/it]
Training Epoch: 13/100, batch 52/375 completed (loss: 0.0312:  14%|█▍        | 27/187 [02:41<15:58,  5.99s/it]
Training Epoch: 13/100, batch 53/375 completed (loss: 0.1333:  14%|█▍        | 27/187 [02:41<15:58,  5.99s/it]
Training Epoch: 13/100, batch 53/375 completed (loss: 0.0145:  14%|█▍        | 27/187 [02:41<15:58,  5.99s/it]
Training Epoch: 13/100, batch 53/375 completed (loss: 0.0232:  14%|█▍        | 27/187 [02:41<15:58,  5.99s/it]
Training Epoch: 13/100, batch 53/375 completed (loss: 0.1468:  14%|█▍        | 27/187 [02:41<15:58,  5.99s/it]
Training Epoch: 13/100, batch 54/375 completed (loss: 0.0321:  14%|█▍        | 27/187 [02:43<15:58,  5.99s/it]
Training Epoch: 13/100, batch 54/375 completed (loss: 0.0225:  14%|█▍        | 27/187 [02:44<15:58,  5.99s/it]
Training Epoch: 13/100, batch 54/375 completed (loss: 0.0668:  14%|█▍        | 27/187 [02:43<15:58,  5.99s/it]
Training Epoch: 13/100, batch 54/375 completed (loss: 0.0108:  14%|█▍        | 27/187 [02:44<15:58,  5.99s/it]
Training Epoch: 13/100, batch 54/375 completed (loss: 0.0668:  15%|█▍        | 28/187 [02:47<16:00,  6.04s/it]
Training Epoch: 13/100, batch 54/375 completed (loss: 0.0225:  15%|█▍        | 28/187 [02:47<16:00,  6.04s/it]
Training Epoch: 13/100, batch 54/375 completed (loss: 0.0108:  15%|█▍        | 28/187 [02:47<16:00,  6.04s/it]
Training Epoch: 13/100, batch 54/375 completed (loss: 0.0321:  15%|█▍        | 28/187 [02:47<16:00,  6.04s/it]
Training Epoch: 13/100, batch 55/375 completed (loss: 0.0156:  15%|█▍        | 28/187 [02:47<16:00,  6.04s/it]
Training Epoch: 13/100, batch 55/375 completed (loss: 0.0227:  15%|█▍        | 28/187 [02:47<16:00,  6.04s/it]
Training Epoch: 13/100, batch 55/375 completed (loss: 0.0305:  15%|█▍        | 28/187 [02:47<16:00,  6.04s/it]
Training Epoch: 13/100, batch 55/375 completed (loss: 0.003:  15%|█▍        | 28/187 [02:47<16:00,  6.04s/it] 
Training Epoch: 13/100, batch 56/375 completed (loss: 0.0012:  15%|█▍        | 28/187 [02:50<16:00,  6.04s/it]
Training Epoch: 13/100, batch 56/375 completed (loss: 0.0074:  15%|█▍        | 28/187 [02:50<16:00,  6.04s/it]
Training Epoch: 13/100, batch 56/375 completed (loss: 0.0123:  15%|█▍        | 28/187 [02:50<16:00,  6.04s/it]
Training Epoch: 13/100, batch 56/375 completed (loss: 0.0051:  15%|█▍        | 28/187 [02:50<16:00,  6.04s/it]
Training Epoch: 13/100, batch 56/375 completed (loss: 0.0051:  16%|█▌        | 29/187 [02:53<15:51,  6.02s/it]
Training Epoch: 13/100, batch 56/375 completed (loss: 0.0012:  16%|█▌        | 29/187 [02:53<15:51,  6.02s/it]
Training Epoch: 13/100, batch 56/375 completed (loss: 0.0074:  16%|█▌        | 29/187 [02:53<15:51,  6.02s/it]
Training Epoch: 13/100, batch 56/375 completed (loss: 0.0123:  16%|█▌        | 29/187 [02:53<15:52,  6.03s/it]
Training Epoch: 13/100, batch 57/375 completed (loss: 0.3027:  16%|█▌        | 29/187 [02:53<15:51,  6.02s/it]
Training Epoch: 13/100, batch 57/375 completed (loss: 0.0934:  16%|█▌        | 29/187 [02:53<15:52,  6.03s/it]
Training Epoch: 13/100, batch 57/375 completed (loss: 0.0233:  16%|█▌        | 29/187 [02:53<15:51,  6.02s/it]
Training Epoch: 13/100, batch 57/375 completed (loss: 0.0061:  16%|█▌        | 29/187 [02:53<15:51,  6.02s/it]
Training Epoch: 13/100, batch 58/375 completed (loss: 0.018:  16%|█▌        | 29/187 [02:56<15:51,  6.02s/it] 
Training Epoch: 13/100, batch 58/375 completed (loss: 0.0362:  16%|█▌        | 29/187 [02:56<15:51,  6.02s/it]
Training Epoch: 13/100, batch 58/375 completed (loss: 0.031:  16%|█▌        | 29/187 [02:56<15:52,  6.03s/it] 
Training Epoch: 13/100, batch 58/375 completed (loss: 0.0221:  16%|█▌        | 29/187 [02:56<15:51,  6.02s/it]
Training Epoch: 13/100, batch 58/375 completed (loss: 0.0221:  16%|█▌        | 30/187 [02:59<15:37,  5.97s/it]
Training Epoch: 13/100, batch 58/375 completed (loss: 0.031:  16%|█▌        | 30/187 [02:59<15:37,  5.97s/it]
Training Epoch: 13/100, batch 58/375 completed (loss: 0.0362:  16%|█▌        | 30/187 [02:59<15:37,  5.97s/it]
Training Epoch: 13/100, batch 58/375 completed (loss: 0.018:  16%|█▌        | 30/187 [02:59<15:37,  5.97s/it]
Training Epoch: 13/100, batch 59/375 completed (loss: 0.0576:  16%|█▌        | 30/187 [02:59<15:37,  5.97s/it]
Training Epoch: 13/100, batch 59/375 completed (loss: 0.0074:  16%|█▌        | 30/187 [02:59<15:37,  5.97s/it]
Training Epoch: 13/100, batch 59/375 completed (loss: 0.022:  16%|█▌        | 30/187 [02:59<15:37,  5.97s/it]
Training Epoch: 13/100, batch 59/375 completed (loss: 0.0344:  16%|█▌        | 30/187 [02:59<15:37,  5.97s/it]
Training Epoch: 13/100, batch 60/375 completed (loss: 0.0553:  16%|█▌        | 30/187 [03:02<15:37,  5.97s/it]
Training Epoch: 13/100, batch 60/375 completed (loss: 0.1261:  16%|█▌        | 30/187 [03:02<15:37,  5.97s/it]
Training Epoch: 13/100, batch 60/375 completed (loss: 0.0046:  16%|█▌        | 30/187 [03:02<15:37,  5.97s/it]
Training Epoch: 13/100, batch 60/375 completed (loss: 0.0347:  16%|█▌        | 30/187 [03:02<15:37,  5.97s/it]
Training Epoch: 13/100, batch 60/375 completed (loss: 0.1261:  17%|█▋        | 31/187 [03:05<15:52,  6.11s/it]
Training Epoch: 13/100, batch 60/375 completed (loss: 0.0553:  17%|█▋        | 31/187 [03:05<15:52,  6.11s/it]
Training Epoch: 13/100, batch 60/375 completed (loss: 0.0046:  17%|█▋        | 31/187 [03:05<15:52,  6.11s/it]
Training Epoch: 13/100, batch 60/375 completed (loss: 0.0347:  17%|█▋        | 31/187 [03:05<15:53,  6.11s/it]
Training Epoch: 13/100, batch 61/375 completed (loss: 0.0272:  17%|█▋        | 31/187 [03:05<15:52,  6.11s/it]
Training Epoch: 13/100, batch 61/375 completed (loss: 0.0064:  17%|█▋        | 31/187 [03:05<15:52,  6.11s/it]
Training Epoch: 13/100, batch 61/375 completed (loss: 0.1044:  17%|█▋        | 31/187 [03:05<15:52,  6.11s/it]
Training Epoch: 13/100, batch 61/375 completed (loss: 0.0258:  17%|█▋        | 31/187 [03:05<15:53,  6.11s/it]
Training Epoch: 13/100, batch 62/375 completed (loss: 0.1658:  17%|█▋        | 31/187 [03:08<15:52,  6.11s/it]
Training Epoch: 13/100, batch 62/375 completed (loss: 0.0191:  17%|█▋        | 31/187 [03:08<15:52,  6.11s/it]
Training Epoch: 13/100, batch 62/375 completed (loss: 0.0017:  17%|█▋        | 31/187 [03:08<15:52,  6.11s/it]
Training Epoch: 13/100, batch 62/375 completed (loss: 0.0242:  17%|█▋        | 31/187 [03:08<15:53,  6.11s/it]
Training Epoch: 13/100, batch 62/375 completed (loss: 0.0017:  17%|█▋        | 32/187 [03:11<15:33,  6.02s/it]
Training Epoch: 13/100, batch 62/375 completed (loss: 0.0242:  17%|█▋        | 32/187 [03:11<15:33,  6.02s/it]
Training Epoch: 13/100, batch 62/375 completed (loss: 0.1658:  17%|█▋        | 32/187 [03:11<15:33,  6.02s/it]
Training Epoch: 13/100, batch 62/375 completed (loss: 0.0191:  17%|█▋        | 32/187 [03:11<15:33,  6.02s/it]
Training Epoch: 13/100, batch 63/375 completed (loss: 0.0089:  17%|█▋        | 32/187 [03:11<15:33,  6.02s/it]
Training Epoch: 13/100, batch 63/375 completed (loss: 0.0135:  17%|█▋        | 32/187 [03:11<15:33,  6.02s/it]
Training Epoch: 13/100, batch 63/375 completed (loss: 0.0201:  17%|█▋        | 32/187 [03:11<15:33,  6.02s/it]
Training Epoch: 13/100, batch 63/375 completed (loss: 0.0213:  17%|█▋        | 32/187 [03:11<15:33,  6.02s/it]
Training Epoch: 13/100, batch 64/375 completed (loss: 0.4369:  17%|█▋        | 32/187 [03:14<15:33,  6.02s/it]
Training Epoch: 13/100, batch 64/375 completed (loss: 0.0116:  17%|█▋        | 32/187 [03:14<15:33,  6.02s/it]
Training Epoch: 13/100, batch 64/375 completed (loss: 0.0197:  17%|█▋        | 32/187 [03:14<15:33,  6.02s/it]
Training Epoch: 13/100, batch 64/375 completed (loss: 0.0591:  17%|█▋        | 32/187 [03:14<15:33,  6.02s/it]
Training Epoch: 13/100, batch 64/375 completed (loss: 0.0591:  18%|█▊        | 33/187 [03:17<15:25,  6.01s/it]
Training Epoch: 13/100, batch 64/375 completed (loss: 0.4369:  18%|█▊        | 33/187 [03:17<15:25,  6.01s/it]
Training Epoch: 13/100, batch 64/375 completed (loss: 0.0116:  18%|█▊        | 33/187 [03:17<15:25,  6.01s/it]
Training Epoch: 13/100, batch 64/375 completed (loss: 0.0197:  18%|█▊        | 33/187 [03:17<15:25,  6.01s/it]
Training Epoch: 13/100, batch 65/375 completed (loss: 0.0105:  18%|█▊        | 33/187 [03:17<15:25,  6.01s/it]
Training Epoch: 13/100, batch 65/375 completed (loss: 0.034:  18%|█▊        | 33/187 [03:17<15:25,  6.01s/it] 
Training Epoch: 13/100, batch 65/375 completed (loss: 0.0043:  18%|█▊        | 33/187 [03:17<15:25,  6.01s/it]
Training Epoch: 13/100, batch 65/375 completed (loss: 0.0052:  18%|█▊        | 33/187 [03:17<15:25,  6.01s/it]
Training Epoch: 13/100, batch 66/375 completed (loss: 0.0542:  18%|█▊        | 33/187 [03:20<15:25,  6.01s/it]
Training Epoch: 13/100, batch 66/375 completed (loss: 0.0046:  18%|█▊        | 33/187 [03:20<15:25,  6.01s/it]
Training Epoch: 13/100, batch 66/375 completed (loss: 0.0084:  18%|█▊        | 33/187 [03:20<15:25,  6.01s/it]
Training Epoch: 13/100, batch 66/375 completed (loss: 0.0109:  18%|█▊        | 33/187 [03:20<15:25,  6.01s/it]
Training Epoch: 13/100, batch 66/375 completed (loss: 0.0109:  18%|█▊        | 34/187 [03:23<15:31,  6.09s/it]
Training Epoch: 13/100, batch 66/375 completed (loss: 0.0046:  18%|█▊        | 34/187 [03:23<15:31,  6.09s/it]
Training Epoch: 13/100, batch 66/375 completed (loss: 0.0084:  18%|█▊        | 34/187 [03:23<15:31,  6.09s/it]
Training Epoch: 13/100, batch 66/375 completed (loss: 0.0542:  18%|█▊        | 34/187 [03:23<15:31,  6.09s/it]
Training Epoch: 13/100, batch 67/375 completed (loss: 0.012:  18%|█▊        | 34/187 [03:23<15:31,  6.09s/it] 
Training Epoch: 13/100, batch 67/375 completed (loss: 0.0839:  18%|█▊        | 34/187 [03:23<15:31,  6.09s/it]
Training Epoch: 13/100, batch 67/375 completed (loss: 0.0149:  18%|█▊        | 34/187 [03:23<15:31,  6.09s/it]
Training Epoch: 13/100, batch 67/375 completed (loss: 0.031:  18%|█▊        | 34/187 [03:23<15:31,  6.09s/it] 
Training Epoch: 13/100, batch 68/375 completed (loss: 0.0317:  18%|█▊        | 34/187 [03:26<15:31,  6.09s/it]
Training Epoch: 13/100, batch 68/375 completed (loss: 0.0135:  18%|█▊        | 34/187 [03:26<15:31,  6.09s/it]
Training Epoch: 13/100, batch 68/375 completed (loss: 0.0339:  18%|█▊        | 34/187 [03:26<15:31,  6.09s/it]
Training Epoch: 13/100, batch 68/375 completed (loss: 0.0113:  18%|█▊        | 34/187 [03:26<15:31,  6.09s/it]
Training Epoch: 13/100, batch 68/375 completed (loss: 0.0339:  19%|█▊        | 35/187 [03:29<15:11,  6.00s/it]
Training Epoch: 13/100, batch 68/375 completed (loss: 0.0135:  19%|█▊        | 35/187 [03:29<15:11,  6.00s/it]
Training Epoch: 13/100, batch 68/375 completed (loss: 0.0113:  19%|█▊        | 35/187 [03:29<15:12,  6.00s/it]
Training Epoch: 13/100, batch 68/375 completed (loss: 0.0317:  19%|█▊        | 35/187 [03:29<15:12,  6.00s/it]
Training Epoch: 13/100, batch 69/375 completed (loss: 0.0051:  19%|█▊        | 35/187 [03:29<15:11,  6.00s/it]
Training Epoch: 13/100, batch 69/375 completed (loss: 0.0904:  19%|█▊        | 35/187 [03:29<15:12,  6.00s/it]
Training Epoch: 13/100, batch 69/375 completed (loss: 0.1039:  19%|█▊        | 35/187 [03:29<15:11,  6.00s/it]
Training Epoch: 13/100, batch 69/375 completed (loss: 0.0095:  19%|█▊        | 35/187 [03:29<15:12,  6.00s/it]
Training Epoch: 13/100, batch 70/375 completed (loss: 0.073:  19%|█▊        | 35/187 [03:32<15:11,  6.00s/it] 
Training Epoch: 13/100, batch 70/375 completed (loss: 0.0194:  19%|█▊        | 35/187 [03:32<15:12,  6.00s/it]
Training Epoch: 13/100, batch 70/375 completed (loss: 0.0071:  19%|█▊        | 35/187 [03:32<15:11,  6.00s/it]
Training Epoch: 13/100, batch 70/375 completed (loss: 0.0226:  19%|█▊        | 35/187 [03:32<15:12,  6.00s/it]
Training Epoch: 13/100, batch 70/375 completed (loss: 0.0071:  19%|█▉        | 36/187 [03:35<15:24,  6.12s/it]
Training Epoch: 13/100, batch 70/375 completed (loss: 0.073:  19%|█▉        | 36/187 [03:35<15:24,  6.12s/it]
Training Epoch: 13/100, batch 70/375 completed (loss: 0.0226:  19%|█▉        | 36/187 [03:35<15:24,  6.13s/it]
Training Epoch: 13/100, batch 70/375 completed (loss: 0.0194:  19%|█▉        | 36/187 [03:35<15:24,  6.12s/it]
Training Epoch: 13/100, batch 71/375 completed (loss: 0.0165:  19%|█▉        | 36/187 [03:35<15:24,  6.12s/it]
Training Epoch: 13/100, batch 71/375 completed (loss: 0.0005:  19%|█▉        | 36/187 [03:35<15:24,  6.12s/it]
Training Epoch: 13/100, batch 71/375 completed (loss: 0.0515:  19%|█▉        | 36/187 [03:35<15:24,  6.12s/it]
Training Epoch: 13/100, batch 71/375 completed (loss: 0.0814:  19%|█▉        | 36/187 [03:35<15:24,  6.13s/it]
Training Epoch: 13/100, batch 72/375 completed (loss: 0.015:  19%|█▉        | 36/187 [03:38<15:24,  6.12s/it] 
Training Epoch: 13/100, batch 72/375 completed (loss: 0.0046:  19%|█▉        | 36/187 [03:38<15:24,  6.12s/it]
Training Epoch: 13/100, batch 72/375 completed (loss: 0.0352:  19%|█▉        | 36/187 [03:38<15:24,  6.12s/it]
Training Epoch: 13/100, batch 72/375 completed (loss: 0.0068:  19%|█▉        | 36/187 [03:38<15:24,  6.13s/it]
Training Epoch: 13/100, batch 72/375 completed (loss: 0.0068:  20%|█▉        | 37/187 [03:41<15:04,  6.03s/it]
Training Epoch: 13/100, batch 72/375 completed (loss: 0.015:  20%|█▉        | 37/187 [03:41<15:04,  6.03s/it]
Training Epoch: 13/100, batch 72/375 completed (loss: 0.0046:  20%|█▉        | 37/187 [03:41<15:04,  6.03s/it]
Training Epoch: 13/100, batch 72/375 completed (loss: 0.0352:  20%|█▉        | 37/187 [03:41<15:04,  6.03s/it]
Training Epoch: 13/100, batch 73/375 completed (loss: 0.031:  20%|█▉        | 37/187 [03:41<15:04,  6.03s/it] 
Training Epoch: 13/100, batch 73/375 completed (loss: 0.0052:  20%|█▉        | 37/187 [03:41<15:04,  6.03s/it]
Training Epoch: 13/100, batch 73/375 completed (loss: 0.0708:  20%|█▉        | 37/187 [03:41<15:04,  6.03s/it]
Training Epoch: 13/100, batch 73/375 completed (loss: 0.0065:  20%|█▉        | 37/187 [03:41<15:04,  6.03s/it]
Training Epoch: 13/100, batch 74/375 completed (loss: 0.0623:  20%|█▉        | 37/187 [03:44<15:04,  6.03s/it]
Training Epoch: 13/100, batch 74/375 completed (loss: 0.008:  20%|█▉        | 37/187 [03:44<15:04,  6.03s/it] 
Training Epoch: 13/100, batch 74/375 completed (loss: 0.0865:  20%|█▉        | 37/187 [03:44<15:04,  6.03s/it]
Training Epoch: 13/100, batch 74/375 completed (loss: 0.042:  20%|█▉        | 37/187 [03:44<15:04,  6.03s/it] 
Training Epoch: 13/100, batch 74/375 completed (loss: 0.042:  20%|██        | 38/187 [03:47<14:56,  6.01s/it]
Training Epoch: 13/100, batch 74/375 completed (loss: 0.0623:  20%|██        | 38/187 [03:47<14:56,  6.01s/it]
Training Epoch: 13/100, batch 74/375 completed (loss: 0.008:  20%|██        | 38/187 [03:47<14:56,  6.01s/it]
Training Epoch: 13/100, batch 74/375 completed (loss: 0.0865:  20%|██        | 38/187 [03:47<14:56,  6.01s/it]
Training Epoch: 13/100, batch 75/375 completed (loss: 0.1585:  20%|██        | 38/187 [03:47<14:56,  6.01s/it]
Training Epoch: 13/100, batch 75/375 completed (loss: 0.0231:  20%|██        | 38/187 [03:47<14:56,  6.01s/it]
Training Epoch: 13/100, batch 75/375 completed (loss: 0.0033:  20%|██        | 38/187 [03:47<14:56,  6.01s/it]
Training Epoch: 13/100, batch 75/375 completed (loss: 0.0203:  20%|██        | 38/187 [03:47<14:56,  6.01s/it]
Training Epoch: 13/100, batch 76/375 completed (loss: 0.0237:  20%|██        | 38/187 [03:50<14:56,  6.01s/it]
Training Epoch: 13/100, batch 76/375 completed (loss: 0.0226:  20%|██        | 38/187 [03:50<14:56,  6.01s/it]
Training Epoch: 13/100, batch 76/375 completed (loss: 0.0075:  20%|██        | 38/187 [03:50<14:56,  6.01s/it]
Training Epoch: 13/100, batch 76/375 completed (loss: 0.0402:  20%|██        | 38/187 [03:50<14:56,  6.01s/it]
Training Epoch: 13/100, batch 76/375 completed (loss: 0.0237:  21%|██        | 39/187 [03:53<15:02,  6.10s/it]
Training Epoch: 13/100, batch 76/375 completed (loss: 0.0075:  21%|██        | 39/187 [03:53<15:02,  6.10s/it]
Training Epoch: 13/100, batch 76/375 completed (loss: 0.0226:  21%|██        | 39/187 [03:53<15:02,  6.10s/it]
Training Epoch: 13/100, batch 76/375 completed (loss: 0.0402:  21%|██        | 39/187 [03:53<15:03,  6.10s/it]
Training Epoch: 13/100, batch 77/375 completed (loss: 0.0148:  21%|██        | 39/187 [03:53<15:02,  6.10s/it]
Training Epoch: 13/100, batch 77/375 completed (loss: 0.0095:  21%|██        | 39/187 [03:53<15:02,  6.10s/it]
Training Epoch: 13/100, batch 77/375 completed (loss: 0.0419:  21%|██        | 39/187 [03:53<15:02,  6.10s/it]
Training Epoch: 13/100, batch 77/375 completed (loss: 0.0497:  21%|██        | 39/187 [03:53<15:03,  6.10s/it]
Training Epoch: 13/100, batch 78/375 completed (loss: 0.0135:  21%|██        | 39/187 [03:56<15:02,  6.10s/it]
Training Epoch: 13/100, batch 78/375 completed (loss: 0.0349:  21%|██        | 39/187 [03:56<15:02,  6.10s/it]
Training Epoch: 13/100, batch 78/375 completed (loss: 0.0114:  21%|██        | 39/187 [03:56<15:02,  6.10s/it]
Training Epoch: 13/100, batch 78/375 completed (loss: 0.0261:  21%|██        | 39/187 [03:56<15:03,  6.10s/it]
Training Epoch: 13/100, batch 78/375 completed (loss: 0.0349:  21%|██▏       | 40/187 [03:59<14:45,  6.02s/it]
Training Epoch: 13/100, batch 78/375 completed (loss: 0.0135:  21%|██▏       | 40/187 [03:59<14:45,  6.02s/it]
Training Epoch: 13/100, batch 78/375 completed (loss: 0.0114:  21%|██▏       | 40/187 [03:59<14:45,  6.02s/it]
Training Epoch: 13/100, batch 78/375 completed (loss: 0.0261:  21%|██▏       | 40/187 [03:59<14:45,  6.03s/it]
Training Epoch: 13/100, batch 79/375 completed (loss: 0.0168:  21%|██▏       | 40/187 [03:59<14:45,  6.02s/it]
Training Epoch: 13/100, batch 79/375 completed (loss: 0.0248:  21%|██▏       | 40/187 [03:59<14:45,  6.02s/it]
Training Epoch: 13/100, batch 79/375 completed (loss: 0.0321:  21%|██▏       | 40/187 [03:59<14:45,  6.02s/it]
Training Epoch: 13/100, batch 79/375 completed (loss: 0.0057:  21%|██▏       | 40/187 [03:59<14:45,  6.03s/it]
Training Epoch: 13/100, batch 80/375 completed (loss: 0.0194:  21%|██▏       | 40/187 [04:02<14:45,  6.02s/it]
Training Epoch: 13/100, batch 80/375 completed (loss: 0.0059:  21%|██▏       | 40/187 [04:02<14:45,  6.02s/it]
Training Epoch: 13/100, batch 80/375 completed (loss: 0.0254:  21%|██▏       | 40/187 [04:02<14:45,  6.02s/it]
Training Epoch: 13/100, batch 80/375 completed (loss: 0.0057:  21%|██▏       | 40/187 [04:02<14:45,  6.03s/it]
Training Epoch: 13/100, batch 80/375 completed (loss: 0.0194:  22%|██▏       | 41/187 [04:05<14:44,  6.06s/it]
Training Epoch: 13/100, batch 80/375 completed (loss: 0.0059:  22%|██▏       | 41/187 [04:05<14:44,  6.06s/it]
Training Epoch: 13/100, batch 80/375 completed (loss: 0.0057:  22%|██▏       | 41/187 [04:05<14:44,  6.06s/it]
Training Epoch: 13/100, batch 80/375 completed (loss: 0.0254:  22%|██▏       | 41/187 [04:05<14:44,  6.06s/it]
Training Epoch: 13/100, batch 81/375 completed (loss: 0.0309:  22%|██▏       | 41/187 [04:05<14:44,  6.06s/it]
Training Epoch: 13/100, batch 81/375 completed (loss: 0.0567:  22%|██▏       | 41/187 [04:05<14:44,  6.06s/it]
Training Epoch: 13/100, batch 81/375 completed (loss: 0.0107:  22%|██▏       | 41/187 [04:05<14:44,  6.06s/it]
Training Epoch: 13/100, batch 81/375 completed (loss: 0.0394:  22%|██▏       | 41/187 [04:05<14:44,  6.06s/it]
Training Epoch: 13/100, batch 82/375 completed (loss: 0.0094:  22%|██▏       | 41/187 [04:08<14:44,  6.06s/it]
Training Epoch: 13/100, batch 82/375 completed (loss: 0.0029:  22%|██▏       | 41/187 [04:08<14:44,  6.06s/it]
Training Epoch: 13/100, batch 82/375 completed (loss: 0.0204:  22%|██▏       | 41/187 [04:08<14:44,  6.06s/it]
Training Epoch: 13/100, batch 82/375 completed (loss: 0.0237:  22%|██▏       | 41/187 [04:08<14:44,  6.06s/it]
Training Epoch: 13/100, batch 82/375 completed (loss: 0.0094:  22%|██▏       | 42/187 [04:11<14:41,  6.08s/it]
Training Epoch: 13/100, batch 82/375 completed (loss: 0.0204:  22%|██▏       | 42/187 [04:11<14:41,  6.08s/it]
Training Epoch: 13/100, batch 82/375 completed (loss: 0.0237:  22%|██▏       | 42/187 [04:11<14:41,  6.08s/it]
Training Epoch: 13/100, batch 82/375 completed (loss: 0.0029:  22%|██▏       | 42/187 [04:11<14:41,  6.08s/it]
Training Epoch: 13/100, batch 83/375 completed (loss: 0.0044:  22%|██▏       | 42/187 [04:11<14:41,  6.08s/it]
Training Epoch: 13/100, batch 83/375 completed (loss: 0.0204:  22%|██▏       | 42/187 [04:11<14:41,  6.08s/it]
Training Epoch: 13/100, batch 83/375 completed (loss: 0.0108:  22%|██▏       | 42/187 [04:11<14:41,  6.08s/it]
Training Epoch: 13/100, batch 83/375 completed (loss: 0.0117:  22%|██▏       | 42/187 [04:12<14:41,  6.08s/it]
Training Epoch: 13/100, batch 84/375 completed (loss: 0.0255:  22%|██▏       | 42/187 [04:14<14:41,  6.08s/it]
Training Epoch: 13/100, batch 84/375 completed (loss: 0.0043:  22%|██▏       | 42/187 [04:14<14:41,  6.08s/it]
Training Epoch: 13/100, batch 84/375 completed (loss: 0.002:  22%|██▏       | 42/187 [04:14<14:41,  6.08s/it] 
Training Epoch: 13/100, batch 84/375 completed (loss: 0.0023:  22%|██▏       | 42/187 [04:14<14:41,  6.08s/it]
Training Epoch: 13/100, batch 84/375 completed (loss: 0.0023:  23%|██▎       | 43/187 [04:17<14:25,  6.01s/it]
Training Epoch: 13/100, batch 84/375 completed (loss: 0.0043:  23%|██▎       | 43/187 [04:17<14:25,  6.01s/it]
Training Epoch: 13/100, batch 84/375 completed (loss: 0.002:  23%|██▎       | 43/187 [04:17<14:25,  6.01s/it]
Training Epoch: 13/100, batch 84/375 completed (loss: 0.0255:  23%|██▎       | 43/187 [04:17<14:25,  6.01s/it]
Training Epoch: 13/100, batch 85/375 completed (loss: 0.0055:  23%|██▎       | 43/187 [04:17<14:25,  6.01s/it]
Training Epoch: 13/100, batch 85/375 completed (loss: 0.0442:  23%|██▎       | 43/187 [04:17<14:25,  6.01s/it]
Training Epoch: 13/100, batch 85/375 completed (loss: 0.0013:  23%|██▎       | 43/187 [04:17<14:25,  6.01s/it]
Training Epoch: 13/100, batch 85/375 completed (loss: 0.0232:  23%|██▎       | 43/187 [04:17<14:25,  6.01s/it]
Training Epoch: 13/100, batch 86/375 completed (loss: 0.0087:  23%|██▎       | 43/187 [04:20<14:25,  6.01s/it]
Training Epoch: 13/100, batch 86/375 completed (loss: 0.0815:  23%|██▎       | 43/187 [04:20<14:25,  6.01s/it]
Training Epoch: 13/100, batch 86/375 completed (loss: 0.0072:  23%|██▎       | 43/187 [04:20<14:25,  6.01s/it]
Training Epoch: 13/100, batch 86/375 completed (loss: 0.0047:  23%|██▎       | 43/187 [04:20<14:25,  6.01s/it]
Training Epoch: 13/100, batch 86/375 completed (loss: 0.0072:  24%|██▎       | 44/187 [04:24<14:37,  6.14s/it]
Training Epoch: 13/100, batch 86/375 completed (loss: 0.0087:  24%|██▎       | 44/187 [04:24<14:37,  6.14s/it]
Training Epoch: 13/100, batch 86/375 completed (loss: 0.0815:  24%|██▎       | 44/187 [04:24<14:37,  6.14s/it]
Training Epoch: 13/100, batch 86/375 completed (loss: 0.0047:  24%|██▎       | 44/187 [04:24<14:37,  6.14s/it]
Training Epoch: 13/100, batch 87/375 completed (loss: 0.0047:  24%|██▎       | 44/187 [04:24<14:37,  6.14s/it]
Training Epoch: 13/100, batch 87/375 completed (loss: 0.0295:  24%|██▎       | 44/187 [04:24<14:37,  6.14s/it]
Training Epoch: 13/100, batch 87/375 completed (loss: 0.0589:  24%|██▎       | 44/187 [04:24<14:37,  6.14s/it]
Training Epoch: 13/100, batch 87/375 completed (loss: 0.0652:  24%|██▎       | 44/187 [04:24<14:37,  6.14s/it]
Training Epoch: 13/100, batch 88/375 completed (loss: 0.0676:  24%|██▎       | 44/187 [04:27<14:37,  6.14s/it]
Training Epoch: 13/100, batch 88/375 completed (loss: 0.016:  24%|██▎       | 44/187 [04:27<14:37,  6.14s/it] 
Training Epoch: 13/100, batch 88/375 completed (loss: 0.0372:  24%|██▎       | 44/187 [04:27<14:37,  6.14s/it]
Training Epoch: 13/100, batch 88/375 completed (loss: 0.0337:  24%|██▎       | 44/187 [04:27<14:37,  6.14s/it]
Training Epoch: 13/100, batch 88/375 completed (loss: 0.0337:  24%|██▍       | 45/187 [04:30<14:18,  6.05s/it]
Training Epoch: 13/100, batch 88/375 completed (loss: 0.0676:  24%|██▍       | 45/187 [04:30<14:18,  6.05s/it]
Training Epoch: 13/100, batch 88/375 completed (loss: 0.0372:  24%|██▍       | 45/187 [04:30<14:19,  6.05s/it]
Training Epoch: 13/100, batch 88/375 completed (loss: 0.016:  24%|██▍       | 45/187 [04:30<14:19,  6.05s/it]
Training Epoch: 13/100, batch 89/375 completed (loss: 0.0126:  24%|██▍       | 45/187 [04:30<14:19,  6.05s/it]
Training Epoch: 13/100, batch 89/375 completed (loss: 0.072:  24%|██▍       | 45/187 [04:30<14:19,  6.05s/it]
Training Epoch: 13/100, batch 89/375 completed (loss: 0.0033:  24%|██▍       | 45/187 [04:30<14:18,  6.05s/it]
Training Epoch: 13/100, batch 89/375 completed (loss: 0.0154:  24%|██▍       | 45/187 [04:30<14:18,  6.05s/it]
Training Epoch: 13/100, batch 90/375 completed (loss: 0.0038:  24%|██▍       | 45/187 [04:33<14:19,  6.05s/it]
Training Epoch: 13/100, batch 90/375 completed (loss: 0.0094:  24%|██▍       | 45/187 [04:33<14:18,  6.05s/it]
Training Epoch: 13/100, batch 90/375 completed (loss: 0.0045:  24%|██▍       | 45/187 [04:33<14:19,  6.05s/it]
Training Epoch: 13/100, batch 90/375 completed (loss: 0.1818:  24%|██▍       | 45/187 [04:33<14:18,  6.05s/it]
Training Epoch: 13/100, batch 90/375 completed (loss: 0.0045:  25%|██▍       | 46/187 [04:36<14:10,  6.03s/it]
Training Epoch: 13/100, batch 90/375 completed (loss: 0.1818:  25%|██▍       | 46/187 [04:36<14:10,  6.03s/it]
Training Epoch: 13/100, batch 90/375 completed (loss: 0.0038:  25%|██▍       | 46/187 [04:36<14:10,  6.03s/it]
Training Epoch: 13/100, batch 90/375 completed (loss: 0.0094:  25%|██▍       | 46/187 [04:36<14:10,  6.03s/it]
Training Epoch: 13/100, batch 91/375 completed (loss: 0.0382:  25%|██▍       | 46/187 [04:36<14:10,  6.03s/it]
Training Epoch: 13/100, batch 91/375 completed (loss: 0.0037:  25%|██▍       | 46/187 [04:36<14:10,  6.03s/it]
Training Epoch: 13/100, batch 91/375 completed (loss: 0.1948:  25%|██▍       | 46/187 [04:36<14:10,  6.03s/it]
Training Epoch: 13/100, batch 91/375 completed (loss: 0.0052:  25%|██▍       | 46/187 [04:36<14:10,  6.03s/it]
Training Epoch: 13/100, batch 92/375 completed (loss: 0.0031:  25%|██▍       | 46/187 [04:39<14:10,  6.03s/it]
Training Epoch: 13/100, batch 92/375 completed (loss: 0.0047:  25%|██▍       | 46/187 [04:39<14:10,  6.03s/it]
Training Epoch: 13/100, batch 92/375 completed (loss: 0.1983:  25%|██▍       | 46/187 [04:39<14:10,  6.03s/it]
Training Epoch: 13/100, batch 92/375 completed (loss: 0.0124:  25%|██▍       | 46/187 [04:39<14:10,  6.03s/it]
Training Epoch: 13/100, batch 92/375 completed (loss: 0.0124:  25%|██▌       | 47/187 [04:42<14:15,  6.11s/it]
Training Epoch: 13/100, batch 92/375 completed (loss: 0.0031:  25%|██▌       | 47/187 [04:42<14:15,  6.11s/it]
Training Epoch: 13/100, batch 92/375 completed (loss: 0.1983:  25%|██▌       | 47/187 [04:42<14:15,  6.11s/it]
Training Epoch: 13/100, batch 92/375 completed (loss: 0.0047:  25%|██▌       | 47/187 [04:42<14:15,  6.11s/it]
Training Epoch: 13/100, batch 93/375 completed (loss: 0.0065:  25%|██▌       | 47/187 [04:42<14:15,  6.11s/it]
Training Epoch: 13/100, batch 93/375 completed (loss: 0.0024:  25%|██▌       | 47/187 [04:42<14:15,  6.11s/it]
Training Epoch: 13/100, batch 93/375 completed (loss: 0.0016:  25%|██▌       | 47/187 [04:42<14:15,  6.11s/it]
Training Epoch: 13/100, batch 93/375 completed (loss: 0.0041:  25%|██▌       | 47/187 [04:42<14:15,  6.11s/it]
Training Epoch: 13/100, batch 94/375 completed (loss: 0.0016:  25%|██▌       | 47/187 [04:45<14:15,  6.11s/it]
Training Epoch: 13/100, batch 94/375 completed (loss: 0.0038:  25%|██▌       | 47/187 [04:45<14:15,  6.11s/it]
Training Epoch: 13/100, batch 94/375 completed (loss: 0.0146:  25%|██▌       | 47/187 [04:45<14:15,  6.11s/it]
Training Epoch: 13/100, batch 94/375 completed (loss: 0.0956:  25%|██▌       | 47/187 [04:45<14:15,  6.11s/it]
Training Epoch: 13/100, batch 94/375 completed (loss: 0.0038:  26%|██▌       | 48/187 [04:48<13:58,  6.04s/it]
Training Epoch: 13/100, batch 94/375 completed (loss: 0.0016:  26%|██▌       | 48/187 [04:48<13:59,  6.04s/it]
Training Epoch: 13/100, batch 94/375 completed (loss: 0.0146:  26%|██▌       | 48/187 [04:48<13:59,  6.04s/it]
Training Epoch: 13/100, batch 94/375 completed (loss: 0.0956:  26%|██▌       | 48/187 [04:48<13:59,  6.04s/it]
Training Epoch: 13/100, batch 95/375 completed (loss: 0.0413:  26%|██▌       | 48/187 [04:48<13:58,  6.04s/it]
Training Epoch: 13/100, batch 95/375 completed (loss: 0.0245:  26%|██▌       | 48/187 [04:48<13:59,  6.04s/it]
Training Epoch: 13/100, batch 95/375 completed (loss: 0.0074:  26%|██▌       | 48/187 [04:48<13:59,  6.04s/it]
Training Epoch: 13/100, batch 95/375 completed (loss: 0.0927:  26%|██▌       | 48/187 [04:48<13:59,  6.04s/it]
Training Epoch: 13/100, batch 96/375 completed (loss: 0.0748:  26%|██▌       | 48/187 [04:51<13:58,  6.04s/it]
Training Epoch: 13/100, batch 96/375 completed (loss: 0.0124:  26%|██▌       | 48/187 [04:51<13:59,  6.04s/it]
Training Epoch: 13/100, batch 96/375 completed (loss: 0.034:  26%|██▌       | 48/187 [04:51<13:59,  6.04s/it] 
Training Epoch: 13/100, batch 96/375 completed (loss: 0.0164:  26%|██▌       | 48/187 [04:51<13:59,  6.04s/it]
Training Epoch: 13/100, batch 96/375 completed (loss: 0.0124:  26%|██▌       | 49/187 [04:54<14:03,  6.11s/it]
Training Epoch: 13/100, batch 96/375 completed (loss: 0.0164:  26%|██▌       | 49/187 [04:54<14:03,  6.11s/it]
Training Epoch: 13/100, batch 96/375 completed (loss: 0.0748:  26%|██▌       | 49/187 [04:54<14:03,  6.11s/it]
Training Epoch: 13/100, batch 96/375 completed (loss: 0.034:  26%|██▌       | 49/187 [04:54<14:04,  6.12s/it]
Training Epoch: 13/100, batch 97/375 completed (loss: 0.0039:  26%|██▌       | 49/187 [04:54<14:03,  6.11s/it]
Training Epoch: 13/100, batch 97/375 completed (loss: 0.0486:  26%|██▌       | 49/187 [04:54<14:03,  6.11s/it]
Training Epoch: 13/100, batch 97/375 completed (loss: 0.1099:  26%|██▌       | 49/187 [04:54<14:04,  6.12s/it]
Training Epoch: 13/100, batch 97/375 completed (loss: 0.007:  26%|██▌       | 49/187 [04:54<14:03,  6.11s/it] 
Training Epoch: 13/100, batch 98/375 completed (loss: 0.0087:  26%|██▌       | 49/187 [04:57<14:03,  6.11s/it]
Training Epoch: 13/100, batch 98/375 completed (loss: 0.0142:  26%|██▌       | 49/187 [04:57<14:04,  6.12s/it]
Training Epoch: 13/100, batch 98/375 completed (loss: 0.0033:  26%|██▌       | 49/187 [04:57<14:03,  6.11s/it]
Training Epoch: 13/100, batch 98/375 completed (loss: 0.0305:  26%|██▌       | 49/187 [04:57<14:03,  6.11s/it]
Training Epoch: 13/100, batch 98/375 completed (loss: 0.0142:  27%|██▋       | 50/187 [05:00<13:54,  6.09s/it]
Training Epoch: 13/100, batch 98/375 completed (loss: 0.0033:  27%|██▋       | 50/187 [05:00<13:55,  6.10s/it]
Training Epoch: 13/100, batch 98/375 completed (loss: 0.0087:  27%|██▋       | 50/187 [05:00<13:55,  6.10s/it]
Training Epoch: 13/100, batch 98/375 completed (loss: 0.0305:  27%|██▋       | 50/187 [05:00<13:55,  6.10s/it]
Training Epoch: 13/100, batch 99/375 completed (loss: 0.0105:  27%|██▋       | 50/187 [05:00<13:55,  6.10s/it]
Training Epoch: 13/100, batch 99/375 completed (loss: 0.0056:  27%|██▋       | 50/187 [05:00<13:55,  6.10s/it]
Training Epoch: 13/100, batch 99/375 completed (loss: 0.0044:  27%|██▋       | 50/187 [05:00<13:54,  6.09s/it]
Training Epoch: 13/100, batch 99/375 completed (loss: 0.0152:  27%|██▋       | 50/187 [05:00<13:55,  6.10s/it]
Training Epoch: 13/100, batch 100/375 completed (loss: 0.1252:  27%|██▋       | 50/187 [05:03<13:55,  6.10s/it]
Training Epoch: 13/100, batch 100/375 completed (loss: 0.0056:  27%|██▋       | 50/187 [05:03<13:54,  6.09s/it]
Training Epoch: 13/100, batch 100/375 completed (loss: 0.0967:  27%|██▋       | 50/187 [05:03<13:55,  6.10s/it]
Training Epoch: 13/100, batch 100/375 completed (loss: 0.0376:  27%|██▋       | 50/187 [05:03<13:55,  6.10s/it]
Training Epoch: 13/100, batch 100/375 completed (loss: 0.1252:  27%|██▋       | 51/187 [05:06<13:44,  6.06s/it]
Training Epoch: 13/100, batch 100/375 completed (loss: 0.0056:  27%|██▋       | 51/187 [05:06<13:44,  6.06s/it]
Training Epoch: 13/100, batch 100/375 completed (loss: 0.0967:  27%|██▋       | 51/187 [05:06<13:44,  6.06s/it]
Training Epoch: 13/100, batch 100/375 completed (loss: 0.0376:  27%|██▋       | 51/187 [05:06<13:44,  6.06s/it]
Training Epoch: 13/100, batch 101/375 completed (loss: 0.0073:  27%|██▋       | 51/187 [05:06<13:44,  6.06s/it]
Training Epoch: 13/100, batch 101/375 completed (loss: 0.0611:  27%|██▋       | 51/187 [05:06<13:44,  6.06s/it]
Training Epoch: 13/100, batch 101/375 completed (loss: 0.0054:  27%|██▋       | 51/187 [05:06<13:44,  6.06s/it]
Training Epoch: 13/100, batch 101/375 completed (loss: 0.0118:  27%|██▋       | 51/187 [05:06<13:44,  6.06s/it]
Training Epoch: 13/100, batch 102/375 completed (loss: 0.0049:  27%|██▋       | 51/187 [05:09<13:44,  6.06s/it]
Training Epoch: 13/100, batch 102/375 completed (loss: 0.0208:  27%|██▋       | 51/187 [05:09<13:44,  6.06s/it]
Training Epoch: 13/100, batch 102/375 completed (loss: 0.0042:  27%|██▋       | 51/187 [05:09<13:44,  6.06s/it]
Training Epoch: 13/100, batch 102/375 completed (loss: 0.0015:  27%|██▋       | 51/187 [05:09<13:44,  6.06s/it]
Training Epoch: 13/100, batch 102/375 completed (loss: 0.0042:  28%|██▊       | 52/187 [05:12<13:48,  6.13s/it]
Training Epoch: 13/100, batch 102/375 completed (loss: 0.0049:  28%|██▊       | 52/187 [05:12<13:48,  6.14s/it]
Training Epoch: 13/100, batch 102/375 completed (loss: 0.0015:  28%|██▊       | 52/187 [05:12<13:48,  6.14s/it]
Training Epoch: 13/100, batch 102/375 completed (loss: 0.0208:  28%|██▊       | 52/187 [05:12<13:48,  6.14s/it]
Training Epoch: 13/100, batch 103/375 completed (loss: 0.0371:  28%|██▊       | 52/187 [05:12<13:48,  6.14s/it]
Training Epoch: 13/100, batch 103/375 completed (loss: 0.0105:  28%|██▊       | 52/187 [05:12<13:48,  6.13s/it]
Training Epoch: 13/100, batch 103/375 completed (loss: 0.0548:  28%|██▊       | 52/187 [05:12<13:48,  6.14s/it]
Training Epoch: 13/100, batch 103/375 completed (loss: 0.0142:  28%|██▊       | 52/187 [05:12<13:48,  6.14s/it]
Training Epoch: 13/100, batch 104/375 completed (loss: 0.0033:  28%|██▊       | 52/187 [05:15<13:48,  6.13s/it]
Training Epoch: 13/100, batch 104/375 completed (loss: 0.1161:  28%|██▊       | 52/187 [05:15<13:48,  6.14s/it]
Training Epoch: 13/100, batch 104/375 completed (loss: 0.125:  28%|██▊       | 52/187 [05:15<13:48,  6.14s/it] 
Training Epoch: 13/100, batch 104/375 completed (loss: 0.0891:  28%|██▊       | 52/187 [05:15<13:48,  6.14s/it]
Training Epoch: 13/100, batch 104/375 completed (loss: 0.1161:  28%|██▊       | 53/187 [05:18<13:29,  6.04s/it]
Training Epoch: 13/100, batch 104/375 completed (loss: 0.125:  28%|██▊       | 53/187 [05:18<13:30,  6.04s/it]
Training Epoch: 13/100, batch 104/375 completed (loss: 0.0033:  28%|██▊       | 53/187 [05:18<13:30,  6.04s/it]
Training Epoch: 13/100, batch 104/375 completed (loss: 0.0891:  28%|██▊       | 53/187 [05:18<13:30,  6.05s/it]
Training Epoch: 13/100, batch 105/375 completed (loss: 0.0263:  28%|██▊       | 53/187 [05:18<13:29,  6.04s/it]
Training Epoch: 13/100, batch 105/375 completed (loss: 0.0044:  28%|██▊       | 53/187 [05:18<13:30,  6.04s/it]
Training Epoch: 13/100, batch 105/375 completed (loss: 0.0669:  28%|██▊       | 53/187 [05:18<13:30,  6.04s/it]
Training Epoch: 13/100, batch 105/375 completed (loss: 0.0102:  28%|██▊       | 53/187 [05:18<13:30,  6.05s/it]
Training Epoch: 13/100, batch 106/375 completed (loss: 0.1483:  28%|██▊       | 53/187 [05:21<13:29,  6.04s/it]
Training Epoch: 13/100, batch 106/375 completed (loss: 0.0237:  28%|██▊       | 53/187 [05:21<13:30,  6.04s/it]
Training Epoch: 13/100, batch 106/375 completed (loss: 0.0013:  28%|██▊       | 53/187 [05:21<13:30,  6.04s/it]
Training Epoch: 13/100, batch 106/375 completed (loss: 0.0252:  28%|██▊       | 53/187 [05:21<13:30,  6.05s/it]
Training Epoch: 13/100, batch 106/375 completed (loss: 0.0252:  29%|██▉       | 54/187 [05:24<13:28,  6.08s/it]
Training Epoch: 13/100, batch 106/375 completed (loss: 0.0013:  29%|██▉       | 54/187 [05:24<13:29,  6.08s/it]
Training Epoch: 13/100, batch 106/375 completed (loss: 0.0237:  29%|██▉       | 54/187 [05:24<13:29,  6.08s/it]
Training Epoch: 13/100, batch 106/375 completed (loss: 0.1483:  29%|██▉       | 54/187 [05:24<13:29,  6.09s/it]
Training Epoch: 13/100, batch 107/375 completed (loss: 0.0202:  29%|██▉       | 54/187 [05:24<13:29,  6.08s/it]
Training Epoch: 13/100, batch 107/375 completed (loss: 0.0009:  29%|██▉       | 54/187 [05:24<13:29,  6.09s/it]
Training Epoch: 13/100, batch 107/375 completed (loss: 0.0385:  29%|██▉       | 54/187 [05:24<13:29,  6.08s/it]
Training Epoch: 13/100, batch 107/375 completed (loss: 0.0199:  29%|██▉       | 54/187 [05:24<13:28,  6.08s/it]
Training Epoch: 13/100, batch 108/375 completed (loss: 0.0144:  29%|██▉       | 54/187 [05:27<13:29,  6.08s/it]
Training Epoch: 13/100, batch 108/375 completed (loss: 0.0021:  29%|██▉       | 54/187 [05:27<13:29,  6.09s/it]
Training Epoch: 13/100, batch 108/375 completed (loss: 0.0018:  29%|██▉       | 54/187 [05:27<13:29,  6.08s/it]
Training Epoch: 13/100, batch 108/375 completed (loss: 0.0072:  29%|██▉       | 54/187 [05:27<13:28,  6.08s/it]
Training Epoch: 13/100, batch 108/375 completed (loss: 0.0144:  29%|██▉       | 55/187 [05:30<13:19,  6.06s/it]
Training Epoch: 13/100, batch 108/375 completed (loss: 0.0072:  29%|██▉       | 55/187 [05:30<13:19,  6.06s/it]
Training Epoch: 13/100, batch 108/375 completed (loss: 0.0021:  29%|██▉       | 55/187 [05:30<13:19,  6.06s/it]
Training Epoch: 13/100, batch 108/375 completed (loss: 0.0018:  29%|██▉       | 55/187 [05:30<13:19,  6.06s/it]
Training Epoch: 13/100, batch 109/375 completed (loss: 0.0626:  29%|██▉       | 55/187 [05:30<13:19,  6.06s/it]
Training Epoch: 13/100, batch 109/375 completed (loss: 0.0045:  29%|██▉       | 55/187 [05:30<13:19,  6.06s/it]
Training Epoch: 13/100, batch 109/375 completed (loss: 0.0042:  29%|██▉       | 55/187 [05:30<13:19,  6.06s/it]
Training Epoch: 13/100, batch 109/375 completed (loss: 0.0089:  29%|██▉       | 55/187 [05:30<13:19,  6.06s/it]
Training Epoch: 13/100, batch 110/375 completed (loss: 0.0031:  29%|██▉       | 55/187 [05:33<13:19,  6.06s/it]
Training Epoch: 13/100, batch 110/375 completed (loss: 0.0022:  29%|██▉       | 55/187 [05:33<13:19,  6.06s/it]
Training Epoch: 13/100, batch 110/375 completed (loss: 0.0127:  29%|██▉       | 55/187 [05:33<13:19,  6.06s/it]
Training Epoch: 13/100, batch 110/375 completed (loss: 0.2999:  29%|██▉       | 55/187 [05:33<13:19,  6.06s/it]
Training Epoch: 13/100, batch 110/375 completed (loss: 0.0022:  30%|██▉       | 56/187 [05:36<13:09,  6.03s/it]
Training Epoch: 13/100, batch 110/375 completed (loss: 0.0127:  30%|██▉       | 56/187 [05:36<13:09,  6.03s/it]
Training Epoch: 13/100, batch 110/375 completed (loss: 0.2999:  30%|██▉       | 56/187 [05:36<13:10,  6.03s/it]
Training Epoch: 13/100, batch 110/375 completed (loss: 0.0031:  30%|██▉       | 56/187 [05:36<13:10,  6.03s/it]
Training Epoch: 13/100, batch 111/375 completed (loss: 0.0412:  30%|██▉       | 56/187 [05:36<13:10,  6.03s/it]
Training Epoch: 13/100, batch 111/375 completed (loss: 0.1182:  30%|██▉       | 56/187 [05:36<13:09,  6.03s/it]
Training Epoch: 13/100, batch 111/375 completed (loss: 0.0063:  30%|██▉       | 56/187 [05:36<13:09,  6.03s/it]
Training Epoch: 13/100, batch 111/375 completed (loss: 0.034:  30%|██▉       | 56/187 [05:36<13:10,  6.03s/it] 
Training Epoch: 13/100, batch 112/375 completed (loss: 0.2487:  30%|██▉       | 56/187 [05:39<13:09,  6.03s/it]
Training Epoch: 13/100, batch 112/375 completed (loss: 0.0026:  30%|██▉       | 56/187 [05:39<13:09,  6.03s/it]
Training Epoch: 13/100, batch 112/375 completed (loss: 0.0031:  30%|██▉       | 56/187 [05:39<13:10,  6.03s/it]
Training Epoch: 13/100, batch 112/375 completed (loss: 0.0192:  30%|██▉       | 56/187 [05:39<13:10,  6.03s/it]
Training Epoch: 13/100, batch 112/375 completed (loss: 0.0026:  30%|███       | 57/187 [05:42<13:07,  6.06s/it]
Training Epoch: 13/100, batch 112/375 completed (loss: 0.2487:  30%|███       | 57/187 [05:42<13:07,  6.06s/it]
Training Epoch: 13/100, batch 112/375 completed (loss: 0.0031:  30%|███       | 57/187 [05:42<13:07,  6.06s/it]
Training Epoch: 13/100, batch 112/375 completed (loss: 0.0192:  30%|███       | 57/187 [05:42<13:07,  6.06s/it]
Training Epoch: 13/100, batch 113/375 completed (loss: 0.0121:  30%|███       | 57/187 [05:42<13:07,  6.06s/it]
Training Epoch: 13/100, batch 113/375 completed (loss: 0.0072:  30%|███       | 57/187 [05:42<13:07,  6.06s/it]
Training Epoch: 13/100, batch 113/375 completed (loss: 0.1679:  30%|███       | 57/187 [05:42<13:07,  6.06s/it]
Training Epoch: 13/100, batch 113/375 completed (loss: 0.0359:  30%|███       | 57/187 [05:43<13:07,  6.06s/it]
Training Epoch: 13/100, batch 114/375 completed (loss: 0.0582:  30%|███       | 57/187 [05:46<13:07,  6.06s/it]
Training Epoch: 13/100, batch 114/375 completed (loss: 0.0047:  30%|███       | 57/187 [05:46<13:07,  6.06s/it]
Training Epoch: 13/100, batch 114/375 completed (loss: 0.0023:  30%|███       | 57/187 [05:46<13:07,  6.06s/it]
Training Epoch: 13/100, batch 114/375 completed (loss: 0.0298:  30%|███       | 57/187 [05:46<13:07,  6.06s/it]
Training Epoch: 13/100, batch 114/375 completed (loss: 0.0047:  31%|███       | 58/187 [05:48<12:59,  6.04s/it]
Training Epoch: 13/100, batch 114/375 completed (loss: 0.0582:  31%|███       | 58/187 [05:48<12:59,  6.04s/it]
Training Epoch: 13/100, batch 114/375 completed (loss: 0.0023:  31%|███       | 58/187 [05:48<12:59,  6.04s/it]
Training Epoch: 13/100, batch 114/375 completed (loss: 0.0298:  31%|███       | 58/187 [05:48<12:59,  6.05s/it]
Training Epoch: 13/100, batch 115/375 completed (loss: 0.0006:  31%|███       | 58/187 [05:49<12:59,  6.04s/it]
Training Epoch: 13/100, batch 115/375 completed (loss: 0.0087:  31%|███       | 58/187 [05:49<12:59,  6.04s/it]
Training Epoch: 13/100, batch 115/375 completed (loss: 0.0127:  31%|███       | 58/187 [05:48<12:59,  6.04s/it]
Training Epoch: 13/100, batch 115/375 completed (loss: 0.0004:  31%|███       | 58/187 [05:49<12:59,  6.05s/it]
Training Epoch: 13/100, batch 116/375 completed (loss: 0.0028:  31%|███       | 58/187 [05:52<12:59,  6.04s/it]
Training Epoch: 13/100, batch 116/375 completed (loss: 0.0032:  31%|███       | 58/187 [05:52<12:59,  6.04s/it]
Training Epoch: 13/100, batch 116/375 completed (loss: 0.0568:  31%|███       | 58/187 [05:52<12:59,  6.04s/it]
Training Epoch: 13/100, batch 116/375 completed (loss: 0.032:  31%|███       | 58/187 [05:52<12:59,  6.05s/it] 
Training Epoch: 13/100, batch 116/375 completed (loss: 0.0568:  32%|███▏      | 59/187 [05:55<12:57,  6.07s/it]
Training Epoch: 13/100, batch 116/375 completed (loss: 0.0032:  32%|███▏      | 59/187 [05:55<12:57,  6.07s/it]
Training Epoch: 13/100, batch 116/375 completed (loss: 0.0028:  32%|███▏      | 59/187 [05:55<12:57,  6.07s/it]
Training Epoch: 13/100, batch 116/375 completed (loss: 0.032:  32%|███▏      | 59/187 [05:55<12:57,  6.08s/it]
Training Epoch: 13/100, batch 117/375 completed (loss: 0.0052:  32%|███▏      | 59/187 [05:55<12:57,  6.07s/it]
Training Epoch: 13/100, batch 117/375 completed (loss: 0.0036:  32%|███▏      | 59/187 [05:55<12:57,  6.07s/it]
Training Epoch: 13/100, batch 117/375 completed (loss: 0.0013:  32%|███▏      | 59/187 [05:55<12:57,  6.07s/it]
Training Epoch: 13/100, batch 117/375 completed (loss: 0.0453:  32%|███▏      | 59/187 [05:55<12:57,  6.08s/it]
Training Epoch: 13/100, batch 118/375 completed (loss: 0.0089:  32%|███▏      | 59/187 [05:58<12:57,  6.07s/it]
Training Epoch: 13/100, batch 118/375 completed (loss: 0.0304:  32%|███▏      | 59/187 [05:58<12:57,  6.07s/it]
Training Epoch: 13/100, batch 118/375 completed (loss: 0.045:  32%|███▏      | 59/187 [05:58<12:57,  6.07s/it] 
Training Epoch: 13/100, batch 118/375 completed (loss: 0.043:  32%|███▏      | 59/187 [05:58<12:57,  6.08s/it] 
Training Epoch: 13/100, batch 118/375 completed (loss: 0.045:  32%|███▏      | 60/187 [06:01<12:47,  6.04s/it]
Training Epoch: 13/100, batch 118/375 completed (loss: 0.0304:  32%|███▏      | 60/187 [06:01<12:47,  6.05s/it]
Training Epoch: 13/100, batch 118/375 completed (loss: 0.043:  32%|███▏      | 60/187 [06:01<12:47,  6.05s/it]
Training Epoch: 13/100, batch 118/375 completed (loss: 0.0089:  32%|███▏      | 60/187 [06:01<12:48,  6.05s/it]
Training Epoch: 13/100, batch 119/375 completed (loss: 0.0819:  32%|███▏      | 60/187 [06:01<12:48,  6.05s/it]
Training Epoch: 13/100, batch 119/375 completed (loss: 0.033:  32%|███▏      | 60/187 [06:01<12:47,  6.05s/it] 
Training Epoch: 13/100, batch 119/375 completed (loss: 0.0115:  32%|███▏      | 60/187 [06:01<12:47,  6.04s/it]
Training Epoch: 13/100, batch 119/375 completed (loss: 0.0429:  32%|███▏      | 60/187 [06:01<12:47,  6.05s/it]
Training Epoch: 13/100, batch 120/375 completed (loss: 0.0192:  32%|███▏      | 60/187 [06:04<12:48,  6.05s/it]
Training Epoch: 13/100, batch 120/375 completed (loss: 0.0005:  32%|███▏      | 60/187 [06:04<12:47,  6.05s/it]
Training Epoch: 13/100, batch 120/375 completed (loss: 0.1617:  32%|███▏      | 60/187 [06:04<12:47,  6.04s/it]
Training Epoch: 13/100, batch 120/375 completed (loss: 0.0029:  32%|███▏      | 60/187 [06:04<12:47,  6.05s/it]
Training Epoch: 13/100, batch 120/375 completed (loss: 0.0029:  33%|███▎      | 61/187 [06:07<12:39,  6.03s/it]
Training Epoch: 13/100, batch 120/375 completed (loss: 0.1617:  33%|███▎      | 61/187 [06:07<12:40,  6.03s/it]
Training Epoch: 13/100, batch 120/375 completed (loss: 0.0005:  33%|███▎      | 61/187 [06:07<12:40,  6.04s/it]
Training Epoch: 13/100, batch 120/375 completed (loss: 0.0192:  33%|███▎      | 61/187 [06:07<12:40,  6.04s/it]
Training Epoch: 13/100, batch 121/375 completed (loss: 0.0009:  33%|███▎      | 61/187 [06:07<12:40,  6.04s/it]
Training Epoch: 13/100, batch 121/375 completed (loss: 0.0032:  33%|███▎      | 61/187 [06:07<12:40,  6.04s/it]
Training Epoch: 13/100, batch 121/375 completed (loss: 0.0076:  33%|███▎      | 61/187 [06:07<12:40,  6.03s/it]
Training Epoch: 13/100, batch 121/375 completed (loss: 0.0127:  33%|███▎      | 61/187 [06:07<12:39,  6.03s/it]
Training Epoch: 13/100, batch 122/375 completed (loss: 0.0067:  33%|███▎      | 61/187 [06:10<12:40,  6.04s/it]
Training Epoch: 13/100, batch 122/375 completed (loss: 0.0033:  33%|███▎      | 61/187 [06:10<12:40,  6.04s/it]
Training Epoch: 13/100, batch 122/375 completed (loss: 0.0134:  33%|███▎      | 61/187 [06:10<12:40,  6.03s/it]
Training Epoch: 13/100, batch 122/375 completed (loss: 0.0279:  33%|███▎      | 61/187 [06:10<12:39,  6.03s/it]
Training Epoch: 13/100, batch 122/375 completed (loss: 0.0067:  33%|███▎      | 62/187 [06:13<12:43,  6.11s/it]
Training Epoch: 13/100, batch 122/375 completed (loss: 0.0279:  33%|███▎      | 62/187 [06:13<12:43,  6.11s/it]
Training Epoch: 13/100, batch 122/375 completed (loss: 0.0134:  33%|███▎      | 62/187 [06:13<12:44,  6.11s/it]
Training Epoch: 13/100, batch 122/375 completed (loss: 0.0033:  33%|███▎      | 62/187 [06:13<12:44,  6.12s/it]
Training Epoch: 13/100, batch 123/375 completed (loss: 0.013:  33%|███▎      | 62/187 [06:13<12:44,  6.12s/it] 
Training Epoch: 13/100, batch 123/375 completed (loss: 0.0104:  33%|███▎      | 62/187 [06:13<12:44,  6.11s/it]
Training Epoch: 13/100, batch 123/375 completed (loss: 0.0133:  33%|███▎      | 62/187 [06:13<12:43,  6.11s/it]
Training Epoch: 13/100, batch 123/375 completed (loss: 0.0331:  33%|███▎      | 62/187 [06:13<12:43,  6.11s/it]
Training Epoch: 13/100, batch 124/375 completed (loss: 0.0045:  33%|███▎      | 62/187 [06:16<12:43,  6.11s/it]
Training Epoch: 13/100, batch 124/375 completed (loss: 0.001:  33%|███▎      | 62/187 [06:16<12:44,  6.12s/it]
Training Epoch: 13/100, batch 124/375 completed (loss: 0.126:  33%|███▎      | 62/187 [06:16<12:44,  6.11s/it] 
Training Epoch: 13/100, batch 124/375 completed (loss: 0.0023:  33%|███▎      | 62/187 [06:16<12:43,  6.11s/it]
Training Epoch: 13/100, batch 124/375 completed (loss: 0.001:  34%|███▎      | 63/187 [06:19<12:33,  6.07s/it]
Training Epoch: 13/100, batch 124/375 completed (loss: 0.126:  34%|███▎      | 63/187 [06:19<12:33,  6.07s/it]
Training Epoch: 13/100, batch 124/375 completed (loss: 0.0023:  34%|███▎      | 63/187 [06:19<12:33,  6.07s/it]
Training Epoch: 13/100, batch 124/375 completed (loss: 0.0045:  34%|███▎      | 63/187 [06:19<12:33,  6.07s/it]
Training Epoch: 13/100, batch 125/375 completed (loss: 0.0002:  34%|███▎      | 63/187 [06:19<12:33,  6.07s/it]
Training Epoch: 13/100, batch 125/375 completed (loss: 0.0248:  34%|███▎      | 63/187 [06:19<12:33,  6.07s/it]
Training Epoch: 13/100, batch 125/375 completed (loss: 0.0665:  34%|███▎      | 63/187 [06:19<12:33,  6.07s/it]
Training Epoch: 13/100, batch 125/375 completed (loss: 0.0053:  34%|███▎      | 63/187 [06:19<12:33,  6.07s/it]
Training Epoch: 13/100, batch 126/375 completed (loss: 0.0102:  34%|███▎      | 63/187 [06:22<12:33,  6.07s/it]
Training Epoch: 13/100, batch 126/375 completed (loss: 0.0531:  34%|███▎      | 63/187 [06:22<12:33,  6.07s/it]
Training Epoch: 13/100, batch 126/375 completed (loss: 0.0051:  34%|███▎      | 63/187 [06:22<12:33,  6.07s/it]
Training Epoch: 13/100, batch 126/375 completed (loss: 0.008:  34%|███▎      | 63/187 [06:22<12:33,  6.07s/it] 
Training Epoch: 13/100, batch 126/375 completed (loss: 0.0531:  34%|███▍      | 64/187 [06:25<12:23,  6.05s/it]
Training Epoch: 13/100, batch 126/375 completed (loss: 0.0051:  34%|███▍      | 64/187 [06:25<12:23,  6.05s/it]
Training Epoch: 13/100, batch 126/375 completed (loss: 0.0102:  34%|███▍      | 64/187 [06:25<12:23,  6.05s/it]
Training Epoch: 13/100, batch 126/375 completed (loss: 0.008:  34%|███▍      | 64/187 [06:25<12:24,  6.05s/it]
Training Epoch: 13/100, batch 127/375 completed (loss: 0.0853:  34%|███▍      | 64/187 [06:25<12:23,  6.05s/it]
Training Epoch: 13/100, batch 127/375 completed (loss: 0.0054:  34%|███▍      | 64/187 [06:25<12:23,  6.05s/it]
Training Epoch: 13/100, batch 127/375 completed (loss: 0.0035:  34%|███▍      | 64/187 [06:25<12:23,  6.05s/it]
Training Epoch: 13/100, batch 127/375 completed (loss: 0.0012:  34%|███▍      | 64/187 [06:25<12:24,  6.05s/it]
Training Epoch: 13/100, batch 128/375 completed (loss: 0.0022:  34%|███▍      | 64/187 [06:28<12:23,  6.05s/it]
Training Epoch: 13/100, batch 128/375 completed (loss: 0.0402:  34%|███▍      | 64/187 [06:28<12:23,  6.05s/it]
Training Epoch: 13/100, batch 128/375 completed (loss: 0.0115:  34%|███▍      | 64/187 [06:28<12:23,  6.05s/it]
Training Epoch: 13/100, batch 128/375 completed (loss: 0.0216:  34%|███▍      | 64/187 [06:28<12:24,  6.05s/it]
Training Epoch: 13/100, batch 128/375 completed (loss: 0.0022:  35%|███▍      | 65/187 [06:31<12:18,  6.05s/it]
Training Epoch: 13/100, batch 128/375 completed (loss: 0.0216:  35%|███▍      | 65/187 [06:31<12:18,  6.05s/it]
Training Epoch: 13/100, batch 128/375 completed (loss: 0.0115:  35%|███▍      | 65/187 [06:31<12:18,  6.05s/it]
Training Epoch: 13/100, batch 128/375 completed (loss: 0.0402:  35%|███▍      | 65/187 [06:31<12:18,  6.05s/it]
Training Epoch: 13/100, batch 129/375 completed (loss: 0.004:  35%|███▍      | 65/187 [06:31<12:18,  6.05s/it] 
Training Epoch: 13/100, batch 129/375 completed (loss: 0.018:  35%|███▍      | 65/187 [06:31<12:18,  6.05s/it] 
Training Epoch: 13/100, batch 129/375 completed (loss: 0.065:  35%|███▍      | 65/187 [06:31<12:18,  6.05s/it] 
Training Epoch: 13/100, batch 129/375 completed (loss: 0.2826:  35%|███▍      | 65/187 [06:31<12:18,  6.05s/it]
Training Epoch: 13/100, batch 130/375 completed (loss: 0.0168:  35%|███▍      | 65/187 [06:34<12:18,  6.05s/it]
Training Epoch: 13/100, batch 130/375 completed (loss: 0.0045:  35%|███▍      | 65/187 [06:34<12:18,  6.05s/it]
Training Epoch: 13/100, batch 130/375 completed (loss: 0.0072:  35%|███▍      | 65/187 [06:34<12:18,  6.05s/it]
Training Epoch: 13/100, batch 130/375 completed (loss: 0.0802:  35%|███▍      | 65/187 [06:34<12:18,  6.05s/it]
Training Epoch: 13/100, batch 130/375 completed (loss: 0.0072:  35%|███▌      | 66/187 [06:37<12:10,  6.03s/it]
Training Epoch: 13/100, batch 130/375 completed (loss: 0.0045:  35%|███▌      | 66/187 [06:37<12:10,  6.03s/it]
Training Epoch: 13/100, batch 130/375 completed (loss: 0.0802:  35%|███▌      | 66/187 [06:37<12:10,  6.03s/it]
Training Epoch: 13/100, batch 130/375 completed (loss: 0.0168:  35%|███▌      | 66/187 [06:37<12:10,  6.04s/it]
Training Epoch: 13/100, batch 131/375 completed (loss: 0.0394:  35%|███▌      | 66/187 [06:37<12:10,  6.04s/it]
Training Epoch: 13/100, batch 131/375 completed (loss: 0.3096:  35%|███▌      | 66/187 [06:37<12:10,  6.03s/it]
Training Epoch: 13/100, batch 131/375 completed (loss: 0.0098:  35%|███▌      | 66/187 [06:37<12:10,  6.03s/it]
Training Epoch: 13/100, batch 131/375 completed (loss: 0.0991:  35%|███▌      | 66/187 [06:37<12:10,  6.03s/it]
Training Epoch: 13/100, batch 132/375 completed (loss: 0.0014:  35%|███▌      | 66/187 [06:40<12:10,  6.03s/it]
Training Epoch: 13/100, batch 132/375 completed (loss: 0.0651:  35%|███▌      | 66/187 [06:40<12:10,  6.04s/it]
Training Epoch: 13/100, batch 132/375 completed (loss: 0.0609:  35%|███▌      | 66/187 [06:40<12:10,  6.03s/it]
Training Epoch: 13/100, batch 132/375 completed (loss: 0.0222:  35%|███▌      | 66/187 [06:40<12:10,  6.03s/it]
Training Epoch: 13/100, batch 132/375 completed (loss: 0.0651:  36%|███▌      | 67/187 [06:43<12:13,  6.11s/it]
Training Epoch: 13/100, batch 132/375 completed (loss: 0.0222:  36%|███▌      | 67/187 [06:43<12:13,  6.11s/it]
Training Epoch: 13/100, batch 132/375 completed (loss: 0.0609:  36%|███▌      | 67/187 [06:43<12:14,  6.12s/it]
Training Epoch: 13/100, batch 132/375 completed (loss: 0.0014:  36%|███▌      | 67/187 [06:43<12:14,  6.12s/it]
Training Epoch: 13/100, batch 133/375 completed (loss: 0.0103:  36%|███▌      | 67/187 [06:43<12:14,  6.12s/it]
Training Epoch: 13/100, batch 133/375 completed (loss: 0.0017:  36%|███▌      | 67/187 [06:43<12:14,  6.12s/it]
Training Epoch: 13/100, batch 133/375 completed (loss: 0.0209:  36%|███▌      | 67/187 [06:43<12:13,  6.11s/it]
Training Epoch: 13/100, batch 133/375 completed (loss: 0.0542:  36%|███▌      | 67/187 [06:43<12:13,  6.11s/it]
Training Epoch: 13/100, batch 134/375 completed (loss: 0.013:  36%|███▌      | 67/187 [06:46<12:14,  6.12s/it] 
Training Epoch: 13/100, batch 134/375 completed (loss: 0.0113:  36%|███▌      | 67/187 [06:46<12:13,  6.11s/it]
Training Epoch: 13/100, batch 134/375 completed (loss: 0.0025:  36%|███▌      | 67/187 [06:46<12:14,  6.12s/it]
Training Epoch: 13/100, batch 134/375 completed (loss: 0.0042:  36%|███▌      | 67/187 [06:46<12:13,  6.11s/it]
Training Epoch: 13/100, batch 134/375 completed (loss: 0.0025:  36%|███▋      | 68/187 [06:49<11:56,  6.02s/it]
Training Epoch: 13/100, batch 134/375 completed (loss: 0.013:  36%|███▋      | 68/187 [06:49<11:56,  6.02s/it]
Training Epoch: 13/100, batch 134/375 completed (loss: 0.0042:  36%|███▋      | 68/187 [06:49<11:56,  6.02s/it]
Training Epoch: 13/100, batch 134/375 completed (loss: 0.0113:  36%|███▋      | 68/187 [06:49<11:56,  6.02s/it]
Training Epoch: 13/100, batch 135/375 completed (loss: 0.0078:  36%|███▋      | 68/187 [06:49<11:56,  6.02s/it]
Training Epoch: 13/100, batch 135/375 completed (loss: 0.0087:  36%|███▋      | 68/187 [06:49<11:56,  6.02s/it]
Training Epoch: 13/100, batch 135/375 completed (loss: 0.0353:  36%|███▋      | 68/187 [06:49<11:56,  6.02s/it]
Training Epoch: 13/100, batch 135/375 completed (loss: 0.0137:  36%|███▋      | 68/187 [06:49<11:56,  6.02s/it]
Training Epoch: 13/100, batch 136/375 completed (loss: 0.0584:  36%|███▋      | 68/187 [06:52<11:56,  6.02s/it]
Training Epoch: 13/100, batch 136/375 completed (loss: 0.0012:  36%|███▋      | 68/187 [06:52<11:56,  6.02s/it]
Training Epoch: 13/100, batch 136/375 completed (loss: 0.0014:  36%|███▋      | 68/187 [06:52<11:56,  6.02s/it]
Training Epoch: 13/100, batch 136/375 completed (loss: 0.026:  36%|███▋      | 68/187 [06:52<11:56,  6.02s/it] 
Training Epoch: 13/100, batch 136/375 completed (loss: 0.0012:  37%|███▋      | 69/187 [06:55<11:57,  6.08s/it]
Training Epoch: 13/100, batch 136/375 completed (loss: 0.026:  37%|███▋      | 69/187 [06:55<11:57,  6.08s/it]
Training Epoch: 13/100, batch 136/375 completed (loss: 0.0014:  37%|███▋      | 69/187 [06:55<11:57,  6.08s/it]
Training Epoch: 13/100, batch 136/375 completed (loss: 0.0584:  37%|███▋      | 69/187 [06:55<11:57,  6.08s/it]
Training Epoch: 13/100, batch 137/375 completed (loss: 0.0856:  37%|███▋      | 69/187 [06:55<11:57,  6.08s/it]
Training Epoch: 13/100, batch 137/375 completed (loss: 0.1272:  37%|███▋      | 69/187 [06:55<11:57,  6.08s/it]
Training Epoch: 13/100, batch 137/375 completed (loss: 0.0988:  37%|███▋      | 69/187 [06:55<11:57,  6.08s/it]
Training Epoch: 13/100, batch 137/375 completed (loss: 0.0276:  37%|███▋      | 69/187 [06:55<11:57,  6.08s/it]
Training Epoch: 13/100, batch 138/375 completed (loss: 0.0161:  37%|███▋      | 69/187 [06:58<11:57,  6.08s/it]
Training Epoch: 13/100, batch 138/375 completed (loss: 0.0302:  37%|███▋      | 69/187 [06:58<11:57,  6.08s/it]
Training Epoch: 13/100, batch 138/375 completed (loss: 0.0019:  37%|███▋      | 69/187 [06:58<11:57,  6.08s/it]
Training Epoch: 13/100, batch 138/375 completed (loss: 0.0742:  37%|███▋      | 69/187 [06:58<11:57,  6.08s/it]
Training Epoch: 13/100, batch 138/375 completed (loss: 0.0161:  37%|███▋      | 70/187 [07:01<11:48,  6.05s/it]
Training Epoch: 13/100, batch 138/375 completed (loss: 0.0019:  37%|███▋      | 70/187 [07:01<11:48,  6.05s/it]
Training Epoch: 13/100, batch 138/375 completed (loss: 0.0302:  37%|███▋      | 70/187 [07:01<11:48,  6.05s/it]
Training Epoch: 13/100, batch 138/375 completed (loss: 0.0742:  37%|███▋      | 70/187 [07:01<11:48,  6.06s/it]
Training Epoch: 13/100, batch 139/375 completed (loss: 0.0102:  37%|███▋      | 70/187 [07:01<11:48,  6.05s/it]
Training Epoch: 13/100, batch 139/375 completed (loss: 0.0341:  37%|███▋      | 70/187 [07:01<11:48,  6.05s/it]
Training Epoch: 13/100, batch 139/375 completed (loss: 0.0425:  37%|███▋      | 70/187 [07:01<11:48,  6.05s/it]
Training Epoch: 13/100, batch 139/375 completed (loss: 0.0442:  37%|███▋      | 70/187 [07:01<11:48,  6.06s/it]
Training Epoch: 13/100, batch 140/375 completed (loss: 0.025:  37%|███▋      | 70/187 [07:04<11:48,  6.05s/it] 
Training Epoch: 13/100, batch 140/375 completed (loss: 0.0015:  37%|███▋      | 70/187 [07:04<11:48,  6.05s/it]
Training Epoch: 13/100, batch 140/375 completed (loss: 0.1369:  37%|███▋      | 70/187 [07:04<11:48,  6.05s/it]
Training Epoch: 13/100, batch 140/375 completed (loss: 0.0309:  37%|███▋      | 70/187 [07:04<11:48,  6.06s/it]
Training Epoch: 13/100, batch 140/375 completed (loss: 0.025:  38%|███▊      | 71/187 [07:07<11:39,  6.03s/it]
Training Epoch: 13/100, batch 140/375 completed (loss: 0.0015:  38%|███▊      | 71/187 [07:07<11:39,  6.03s/it]
Training Epoch: 13/100, batch 140/375 completed (loss: 0.0309:  38%|███▊      | 71/187 [07:07<11:39,  6.03s/it]
Training Epoch: 13/100, batch 140/375 completed (loss: 0.1369:  38%|███▊      | 71/187 [07:07<11:39,  6.03s/it]
Training Epoch: 13/100, batch 141/375 completed (loss: 0.0044:  38%|███▊      | 71/187 [07:07<11:39,  6.03s/it]
Training Epoch: 13/100, batch 141/375 completed (loss: 0.0123:  38%|███▊      | 71/187 [07:07<11:39,  6.03s/it]
Training Epoch: 13/100, batch 141/375 completed (loss: 0.0031:  38%|███▊      | 71/187 [07:07<11:39,  6.03s/it]
Training Epoch: 13/100, batch 141/375 completed (loss: 0.0142:  38%|███▊      | 71/187 [07:07<11:39,  6.03s/it]
Training Epoch: 13/100, batch 142/375 completed (loss: 0.0041:  38%|███▊      | 71/187 [07:10<11:39,  6.03s/it]
Training Epoch: 13/100, batch 142/375 completed (loss: 0.0064:  38%|███▊      | 71/187 [07:10<11:39,  6.03s/it]
Training Epoch: 13/100, batch 142/375 completed (loss: 0.0439:  38%|███▊      | 71/187 [07:10<11:39,  6.03s/it]
Training Epoch: 13/100, batch 142/375 completed (loss: 0.025:  38%|███▊      | 71/187 [07:10<11:39,  6.03s/it] 
Training Epoch: 13/100, batch 142/375 completed (loss: 0.025:  39%|███▊      | 72/187 [07:13<11:32,  6.02s/it]
Training Epoch: 13/100, batch 142/375 completed (loss: 0.0064:  39%|███▊      | 72/187 [07:13<11:32,  6.02s/it]
Training Epoch: 13/100, batch 142/375 completed (loss: 0.0439:  39%|███▊      | 72/187 [07:13<11:32,  6.02s/it]
Training Epoch: 13/100, batch 142/375 completed (loss: 0.0041:  39%|███▊      | 72/187 [07:13<11:33,  6.03s/it]
Training Epoch: 13/100, batch 143/375 completed (loss: 0.0038:  39%|███▊      | 72/187 [07:13<11:32,  6.02s/it]
Training Epoch: 13/100, batch 143/375 completed (loss: 0.0116:  39%|███▊      | 72/187 [07:13<11:32,  6.02s/it]
Training Epoch: 13/100, batch 143/375 completed (loss: 0.01:  39%|███▊      | 72/187 [07:13<11:33,  6.03s/it]  
Training Epoch: 13/100, batch 143/375 completed (loss: 0.0404:  39%|███▊      | 72/187 [07:13<11:32,  6.02s/it]
Training Epoch: 13/100, batch 144/375 completed (loss: 0.0118:  39%|███▊      | 72/187 [07:16<11:32,  6.02s/it]
Training Epoch: 13/100, batch 144/375 completed (loss: 0.007:  39%|███▊      | 72/187 [07:16<11:33,  6.03s/it]
Training Epoch: 13/100, batch 144/375 completed (loss: 0.0041:  39%|███▊      | 72/187 [07:16<11:32,  6.02s/it]
Training Epoch: 13/100, batch 144/375 completed (loss: 0.0055:  39%|███▊      | 72/187 [07:16<11:32,  6.02s/it]
Training Epoch: 13/100, batch 144/375 completed (loss: 0.0041:  39%|███▉      | 73/187 [07:19<11:25,  6.02s/it]
Training Epoch: 13/100, batch 144/375 completed (loss: 0.007:  39%|███▉      | 73/187 [07:19<11:25,  6.02s/it]
Training Epoch: 13/100, batch 144/375 completed (loss: 0.0055:  39%|███▉      | 73/187 [07:19<11:25,  6.02s/it]
Training Epoch: 13/100, batch 144/375 completed (loss: 0.0118:  39%|███▉      | 73/187 [07:19<11:26,  6.02s/it]
Training Epoch: 13/100, batch 145/375 completed (loss: 0.0321:  39%|███▉      | 73/187 [07:19<11:26,  6.02s/it]
Training Epoch: 13/100, batch 145/375 completed (loss: 0.0058:  39%|███▉      | 73/187 [07:19<11:25,  6.02s/it]
Training Epoch: 13/100, batch 145/375 completed (loss: 0.0071:  39%|███▉      | 73/187 [07:19<11:25,  6.02s/it]
Training Epoch: 13/100, batch 145/375 completed (loss: 0.0164:  39%|███▉      | 73/187 [07:19<11:25,  6.02s/it]
Training Epoch: 13/100, batch 146/375 completed (loss: 0.0743:  39%|███▉      | 73/187 [07:22<11:25,  6.02s/it]
Training Epoch: 13/100, batch 146/375 completed (loss: 0.0061:  39%|███▉      | 73/187 [07:22<11:25,  6.02s/it]
Training Epoch: 13/100, batch 146/375 completed (loss: 0.0076:  39%|███▉      | 73/187 [07:22<11:26,  6.02s/it]
Training Epoch: 13/100, batch 146/375 completed (loss: 0.0007:  39%|███▉      | 73/187 [07:22<11:25,  6.02s/it]
Training Epoch: 13/100, batch 146/375 completed (loss: 0.0061:  40%|███▉      | 74/187 [07:25<11:18,  6.01s/it]
Training Epoch: 13/100, batch 146/375 completed (loss: 0.0076:  40%|███▉      | 74/187 [07:25<11:18,  6.00s/it]
Training Epoch: 13/100, batch 146/375 completed (loss: 0.0007:  40%|███▉      | 74/187 [07:25<11:18,  6.01s/it]
Training Epoch: 13/100, batch 146/375 completed (loss: 0.0743:  40%|███▉      | 74/187 [07:25<11:18,  6.01s/it]
Training Epoch: 13/100, batch 147/375 completed (loss: 0.0011:  40%|███▉      | 74/187 [07:25<11:18,  6.00s/it]
Training Epoch: 13/100, batch 147/375 completed (loss: 0.0083:  40%|███▉      | 74/187 [07:25<11:18,  6.01s/it]
Training Epoch: 13/100, batch 147/375 completed (loss: 0.0789:  40%|███▉      | 74/187 [07:25<11:18,  6.01s/it]
Training Epoch: 13/100, batch 147/375 completed (loss: 0.0042:  40%|███▉      | 74/187 [07:25<11:18,  6.01s/it]
Training Epoch: 13/100, batch 148/375 completed (loss: 0.004:  40%|███▉      | 74/187 [07:28<11:18,  6.01s/it] 
Training Epoch: 13/100, batch 148/375 completed (loss: 0.0118:  40%|███▉      | 74/187 [07:28<11:18,  6.01s/it]
Training Epoch: 13/100, batch 148/375 completed (loss: 0.004:  40%|███▉      | 74/187 [07:28<11:18,  6.00s/it] 
Training Epoch: 13/100, batch 148/375 completed (loss: 0.0951:  40%|███▉      | 74/187 [07:28<11:18,  6.01s/it]
Training Epoch: 13/100, batch 148/375 completed (loss: 0.004:  40%|████      | 75/187 [07:31<11:22,  6.09s/it]
Training Epoch: 13/100, batch 148/375 completed (loss: 0.0951:  40%|████      | 75/187 [07:32<11:22,  6.09s/it]
Training Epoch: 13/100, batch 148/375 completed (loss: 0.0118:  40%|████      | 75/187 [07:31<11:22,  6.09s/it]
Training Epoch: 13/100, batch 148/375 completed (loss: 0.004:  40%|████      | 75/187 [07:31<11:22,  6.10s/it]
Training Epoch: 13/100, batch 149/375 completed (loss: 0.0119:  40%|████      | 75/187 [07:32<11:22,  6.10s/it]
Training Epoch: 13/100, batch 149/375 completed (loss: 0.0169:  40%|████      | 75/187 [07:32<11:22,  6.09s/it]
Training Epoch: 13/100, batch 149/375 completed (loss: 0.0094:  40%|████      | 75/187 [07:32<11:22,  6.09s/it]
Training Epoch: 13/100, batch 149/375 completed (loss: 0.024:  40%|████      | 75/187 [07:32<11:22,  6.09s/it] 
Training Epoch: 13/100, batch 150/375 completed (loss: 0.0031:  40%|████      | 75/187 [07:34<11:22,  6.09s/it]
Training Epoch: 13/100, batch 150/375 completed (loss: 0.0465:  40%|████      | 75/187 [07:34<11:22,  6.09s/it]
Training Epoch: 13/100, batch 150/375 completed (loss: 0.0322:  40%|████      | 75/187 [07:34<11:22,  6.10s/it]
Training Epoch: 13/100, batch 150/375 completed (loss: 0.0056:  40%|████      | 75/187 [07:34<11:22,  6.09s/it]
Training Epoch: 13/100, batch 150/375 completed (loss: 0.0056:  41%|████      | 76/187 [07:37<11:07,  6.02s/it]
Training Epoch: 13/100, batch 150/375 completed (loss: 0.0322:  41%|████      | 76/187 [07:37<11:07,  6.02s/it]
Training Epoch: 13/100, batch 150/375 completed (loss: 0.0031:  41%|████      | 76/187 [07:37<11:07,  6.02s/it]
Training Epoch: 13/100, batch 150/375 completed (loss: 0.0465:  41%|████      | 76/187 [07:37<11:07,  6.02s/it]
Training Epoch: 13/100, batch 151/375 completed (loss: 0.0083:  41%|████      | 76/187 [07:37<11:07,  6.02s/it]
Training Epoch: 13/100, batch 151/375 completed (loss: 0.0316:  41%|████      | 76/187 [07:37<11:07,  6.02s/it]
Training Epoch: 13/100, batch 151/375 completed (loss: 0.0251:  41%|████      | 76/187 [07:37<11:07,  6.02s/it]
Training Epoch: 13/100, batch 151/375 completed (loss: 0.0143:  41%|████      | 76/187 [07:37<11:07,  6.02s/it]
Training Epoch: 13/100, batch 152/375 completed (loss: 0.0004:  41%|████      | 76/187 [07:40<11:07,  6.02s/it]
Training Epoch: 13/100, batch 152/375 completed (loss: 0.0165:  41%|████      | 76/187 [07:40<11:07,  6.02s/it]
Training Epoch: 13/100, batch 152/375 completed (loss: 0.0093:  41%|████      | 76/187 [07:40<11:07,  6.02s/it]
Training Epoch: 13/100, batch 152/375 completed (loss: 0.0107:  41%|████      | 76/187 [07:40<11:07,  6.02s/it]
Training Epoch: 13/100, batch 152/375 completed (loss: 0.0093:  41%|████      | 77/187 [07:44<11:11,  6.10s/it]
Training Epoch: 13/100, batch 152/375 completed (loss: 0.0165:  41%|████      | 77/187 [07:44<11:11,  6.10s/it]
Training Epoch: 13/100, batch 152/375 completed (loss: 0.0107:  41%|████      | 77/187 [07:44<11:11,  6.10s/it]
Training Epoch: 13/100, batch 152/375 completed (loss: 0.0004:  41%|████      | 77/187 [07:44<11:11,  6.11s/it]
Training Epoch: 13/100, batch 153/375 completed (loss: 0.0502:  41%|████      | 77/187 [07:44<11:11,  6.10s/it]
Training Epoch: 13/100, batch 153/375 completed (loss: 0.0017:  41%|████      | 77/187 [07:44<11:11,  6.11s/it]
Training Epoch: 13/100, batch 153/375 completed (loss: 0.068:  41%|████      | 77/187 [07:44<11:11,  6.10s/it] 
Training Epoch: 13/100, batch 153/375 completed (loss: 0.0083:  41%|████      | 77/187 [07:44<11:11,  6.10s/it]
Training Epoch: 13/100, batch 154/375 completed (loss: 0.0012:  41%|████      | 77/187 [07:47<11:11,  6.10s/it]
Training Epoch: 13/100, batch 154/375 completed (loss: 0.006:  41%|████      | 77/187 [07:47<11:11,  6.10s/it] 
Training Epoch: 13/100, batch 154/375 completed (loss: 0.0191:  41%|████      | 77/187 [07:47<11:11,  6.11s/it]
Training Epoch: 13/100, batch 154/375 completed (loss: 0.0028:  41%|████      | 77/187 [07:47<11:11,  6.10s/it]
Training Epoch: 13/100, batch 154/375 completed (loss: 0.006:  42%|████▏     | 78/187 [07:50<11:00,  6.06s/it]
Training Epoch: 13/100, batch 154/375 completed (loss: 0.0012:  42%|████▏     | 78/187 [07:50<11:00,  6.06s/it]
Training Epoch: 13/100, batch 154/375 completed (loss: 0.0191:  42%|████▏     | 78/187 [07:50<11:00,  6.06s/it]
Training Epoch: 13/100, batch 154/375 completed (loss: 0.0028:  42%|████▏     | 78/187 [07:50<11:01,  6.07s/it]
Training Epoch: 13/100, batch 155/375 completed (loss: 0.0573:  42%|████▏     | 78/187 [07:50<11:00,  6.06s/it]
Training Epoch: 13/100, batch 155/375 completed (loss: 0.0089:  42%|████▏     | 78/187 [07:50<11:00,  6.06s/it]
Training Epoch: 13/100, batch 155/375 completed (loss: 0.0077:  42%|████▏     | 78/187 [07:50<11:00,  6.06s/it]
Training Epoch: 13/100, batch 155/375 completed (loss: 0.0044:  42%|████▏     | 78/187 [07:50<11:01,  6.07s/it]
Training Epoch: 13/100, batch 156/375 completed (loss: 0.0496:  42%|████▏     | 78/187 [07:53<11:00,  6.06s/it]
Training Epoch: 13/100, batch 156/375 completed (loss: 0.004:  42%|████▏     | 78/187 [07:53<11:00,  6.06s/it] 
Training Epoch: 13/100, batch 156/375 completed (loss: 0.0021:  42%|████▏     | 78/187 [07:53<11:00,  6.06s/it]
Training Epoch: 13/100, batch 156/375 completed (loss: 0.05:  42%|████▏     | 78/187 [07:53<11:01,  6.07s/it]  
Training Epoch: 13/100, batch 156/375 completed (loss: 0.0021:  42%|████▏     | 79/187 [07:55<10:47,  6.00s/it]
Training Epoch: 13/100, batch 156/375 completed (loss: 0.05:  42%|████▏     | 79/187 [07:55<10:48,  6.00s/it]
Training Epoch: 13/100, batch 156/375 completed (loss: 0.0496:  42%|████▏     | 79/187 [07:55<10:48,  6.00s/it]
Training Epoch: 13/100, batch 156/375 completed (loss: 0.004:  42%|████▏     | 79/187 [07:55<10:48,  6.01s/it]
Training Epoch: 13/100, batch 157/375 completed (loss: 0.042:  42%|████▏     | 79/187 [07:56<10:48,  6.00s/it] 
Training Epoch: 13/100, batch 157/375 completed (loss: 0.0066:  42%|████▏     | 79/187 [07:56<10:47,  6.00s/it]
Training Epoch: 13/100, batch 157/375 completed (loss: 0.0074:  42%|████▏     | 79/187 [07:56<10:48,  6.01s/it]
Training Epoch: 13/100, batch 157/375 completed (loss: 0.0209:  42%|████▏     | 79/187 [07:56<10:48,  6.00s/it]
Training Epoch: 13/100, batch 158/375 completed (loss: 0.0132:  42%|████▏     | 79/187 [07:59<10:47,  6.00s/it]
Training Epoch: 13/100, batch 158/375 completed (loss: 0.0657:  42%|████▏     | 79/187 [07:59<10:48,  6.01s/it]
Training Epoch: 13/100, batch 158/375 completed (loss: 0.0025:  42%|████▏     | 79/187 [07:59<10:48,  6.00s/it]
Training Epoch: 13/100, batch 158/375 completed (loss: 0.058:  42%|████▏     | 79/187 [07:59<10:48,  6.00s/it] 
Training Epoch: 13/100, batch 158/375 completed (loss: 0.0132:  43%|████▎     | 80/187 [08:02<10:53,  6.11s/it]
Training Epoch: 13/100, batch 158/375 completed (loss: 0.0657:  43%|████▎     | 80/187 [08:02<10:53,  6.11s/it]
Training Epoch: 13/100, batch 158/375 completed (loss: 0.0025:  43%|████▎     | 80/187 [08:02<10:53,  6.11s/it]
Training Epoch: 13/100, batch 158/375 completed (loss: 0.058:  43%|████▎     | 80/187 [08:02<10:53,  6.11s/it]
Training Epoch: 13/100, batch 159/375 completed (loss: 0.0151:  43%|████▎     | 80/187 [08:02<10:53,  6.11s/it]
Training Epoch: 13/100, batch 159/375 completed (loss: 0.0049:  43%|████▎     | 80/187 [08:02<10:53,  6.11s/it]
Training Epoch: 13/100, batch 159/375 completed (loss: 0.1263:  43%|████▎     | 80/187 [08:02<10:53,  6.11s/it]
Training Epoch: 13/100, batch 159/375 completed (loss: 0.0188:  43%|████▎     | 80/187 [08:02<10:53,  6.11s/it]
Training Epoch: 13/100, batch 160/375 completed (loss: 0.0223:  43%|████▎     | 80/187 [08:05<10:53,  6.11s/it]
Training Epoch: 13/100, batch 160/375 completed (loss: 0.0015:  43%|████▎     | 80/187 [08:05<10:53,  6.11s/it]
Training Epoch: 13/100, batch 160/375 completed (loss: 0.0067:  43%|████▎     | 80/187 [08:05<10:53,  6.11s/it]
Training Epoch: 13/100, batch 160/375 completed (loss: 0.081:  43%|████▎     | 80/187 [08:05<10:53,  6.11s/it] 
Training Epoch: 13/100, batch 160/375 completed (loss: 0.081:  43%|████▎     | 81/187 [08:08<10:39,  6.03s/it]
Training Epoch: 13/100, batch 160/375 completed (loss: 0.0015:  43%|████▎     | 81/187 [08:08<10:39,  6.03s/it]
Training Epoch: 13/100, batch 160/375 completed (loss: 0.0223:  43%|████▎     | 81/187 [08:08<10:39,  6.03s/it]
Training Epoch: 13/100, batch 160/375 completed (loss: 0.0067:  43%|████▎     | 81/187 [08:08<10:39,  6.03s/it]
Training Epoch: 13/100, batch 161/375 completed (loss: 0.0868:  43%|████▎     | 81/187 [08:08<10:39,  6.03s/it]
Training Epoch: 13/100, batch 161/375 completed (loss: 0.0052:  43%|████▎     | 81/187 [08:08<10:39,  6.03s/it]
Training Epoch: 13/100, batch 161/375 completed (loss: 0.0349:  43%|████▎     | 81/187 [08:08<10:39,  6.03s/it]
Training Epoch: 13/100, batch 161/375 completed (loss: 0.0012:  43%|████▎     | 81/187 [08:08<10:39,  6.03s/it]
Training Epoch: 13/100, batch 162/375 completed (loss: 0.0441:  43%|████▎     | 81/187 [08:11<10:39,  6.03s/it]
Training Epoch: 13/100, batch 162/375 completed (loss: 0.013:  43%|████▎     | 81/187 [08:11<10:39,  6.03s/it] 
Training Epoch: 13/100, batch 162/375 completed (loss: 0.0685:  43%|████▎     | 81/187 [08:11<10:39,  6.03s/it]
Training Epoch: 13/100, batch 162/375 completed (loss: 0.1181:  43%|████▎     | 81/187 [08:11<10:39,  6.03s/it]
Training Epoch: 13/100, batch 162/375 completed (loss: 0.1181:  44%|████▍     | 82/187 [08:14<10:42,  6.11s/it]
Training Epoch: 13/100, batch 162/375 completed (loss: 0.0685:  44%|████▍     | 82/187 [08:14<10:42,  6.12s/it]
Training Epoch: 13/100, batch 162/375 completed (loss: 0.013:  44%|████▍     | 82/187 [08:14<10:42,  6.12s/it]
Training Epoch: 13/100, batch 162/375 completed (loss: 0.0441:  44%|████▍     | 82/187 [08:14<10:42,  6.12s/it]
Training Epoch: 13/100, batch 163/375 completed (loss: 0.0119:  44%|████▍     | 82/187 [08:14<10:42,  6.12s/it]
Training Epoch: 13/100, batch 163/375 completed (loss: 0.0005:  44%|████▍     | 82/187 [08:14<10:42,  6.12s/it]
Training Epoch: 13/100, batch 163/375 completed (loss: 0.0015:  44%|████▍     | 82/187 [08:14<10:42,  6.12s/it]
Training Epoch: 13/100, batch 163/375 completed (loss: 0.0017:  44%|████▍     | 82/187 [08:14<10:42,  6.11s/it]
Training Epoch: 13/100, batch 164/375 completed (loss: 0.0077:  44%|████▍     | 82/187 [08:17<10:42,  6.12s/it]
Training Epoch: 13/100, batch 164/375 completed (loss: 0.0006:  44%|████▍     | 82/187 [08:17<10:42,  6.12s/it]
Training Epoch: 13/100, batch 164/375 completed (loss: 0.0109:  44%|████▍     | 82/187 [08:17<10:42,  6.12s/it]
Training Epoch: 13/100, batch 164/375 completed (loss: 0.0065:  44%|████▍     | 82/187 [08:17<10:42,  6.11s/it]
Training Epoch: 13/100, batch 164/375 completed (loss: 0.0006:  44%|████▍     | 83/187 [08:20<10:29,  6.06s/it]
Training Epoch: 13/100, batch 164/375 completed (loss: 0.0109:  44%|████▍     | 83/187 [08:20<10:29,  6.06s/it]
Training Epoch: 13/100, batch 164/375 completed (loss: 0.0077:  44%|████▍     | 83/187 [08:20<10:30,  6.06s/it]
Training Epoch: 13/100, batch 164/375 completed (loss: 0.0065:  44%|████▍     | 83/187 [08:20<10:30,  6.06s/it]
Training Epoch: 13/100, batch 165/375 completed (loss: 0.0067:  44%|████▍     | 83/187 [08:20<10:30,  6.06s/it]
Training Epoch: 13/100, batch 165/375 completed (loss: 0.0015:  44%|████▍     | 83/187 [08:20<10:29,  6.06s/it]
Training Epoch: 13/100, batch 165/375 completed (loss: 0.0166:  44%|████▍     | 83/187 [08:20<10:29,  6.06s/it]
Training Epoch: 13/100, batch 165/375 completed (loss: 0.1226:  44%|████▍     | 83/187 [08:20<10:30,  6.06s/it]
Training Epoch: 13/100, batch 166/375 completed (loss: 0.0152:  44%|████▍     | 83/187 [08:23<10:30,  6.06s/it]
Training Epoch: 13/100, batch 166/375 completed (loss: 0.0035:  44%|████▍     | 83/187 [08:23<10:29,  6.06s/it]
Training Epoch: 13/100, batch 166/375 completed (loss: 0.0997:  44%|████▍     | 83/187 [08:23<10:29,  6.06s/it]
Training Epoch: 13/100, batch 166/375 completed (loss: 0.0045:  44%|████▍     | 83/187 [08:23<10:30,  6.06s/it]
Training Epoch: 13/100, batch 166/375 completed (loss: 0.0152:  45%|████▍     | 84/187 [08:26<10:16,  5.99s/it]
Training Epoch: 13/100, batch 166/375 completed (loss: 0.0035:  45%|████▍     | 84/187 [08:26<10:17,  5.99s/it]
Training Epoch: 13/100, batch 166/375 completed (loss: 0.0045:  45%|████▍     | 84/187 [08:26<10:17,  5.99s/it]
Training Epoch: 13/100, batch 166/375 completed (loss: 0.0997:  45%|████▍     | 84/187 [08:26<10:17,  5.99s/it]
Training Epoch: 13/100, batch 167/375 completed (loss: 0.0283:  45%|████▍     | 84/187 [08:26<10:17,  5.99s/it]
Training Epoch: 13/100, batch 167/375 completed (loss: 0.007:  45%|████▍     | 84/187 [08:26<10:16,  5.99s/it] 
Training Epoch: 13/100, batch 167/375 completed (loss: 0.003:  45%|████▍     | 84/187 [08:26<10:17,  5.99s/it] 
Training Epoch: 13/100, batch 167/375 completed (loss: 0.0011:  45%|████▍     | 84/187 [08:26<10:17,  5.99s/it]
Training Epoch: 13/100, batch 168/375 completed (loss: 0.0206:  45%|████▍     | 84/187 [08:29<10:16,  5.99s/it]
Training Epoch: 13/100, batch 168/375 completed (loss: 0.0011:  45%|████▍     | 84/187 [08:29<10:17,  5.99s/it]
Training Epoch: 13/100, batch 168/375 completed (loss: 0.0003:  45%|████▍     | 84/187 [08:29<10:17,  5.99s/it]
Training Epoch: 13/100, batch 168/375 completed (loss: 0.0012:  45%|████▍     | 84/187 [08:29<10:17,  5.99s/it]
Training Epoch: 13/100, batch 168/375 completed (loss: 0.0011:  45%|████▌     | 85/187 [08:32<10:21,  6.10s/it]
Training Epoch: 13/100, batch 168/375 completed (loss: 0.0206:  45%|████▌     | 85/187 [08:32<10:22,  6.10s/it]
Training Epoch: 13/100, batch 168/375 completed (loss: 0.0012:  45%|████▌     | 85/187 [08:32<10:22,  6.10s/it]
Training Epoch: 13/100, batch 168/375 completed (loss: 0.0003:  45%|████▌     | 85/187 [08:32<10:22,  6.10s/it]
Training Epoch: 13/100, batch 169/375 completed (loss: 0.1315:  45%|████▌     | 85/187 [08:32<10:21,  6.10s/it]
Training Epoch: 13/100, batch 169/375 completed (loss: 0.0223:  45%|████▌     | 85/187 [08:32<10:22,  6.10s/it]
Training Epoch: 13/100, batch 169/375 completed (loss: 0.0052:  45%|████▌     | 85/187 [08:32<10:22,  6.10s/it]
Training Epoch: 13/100, batch 169/375 completed (loss: 0.0046:  45%|████▌     | 85/187 [08:32<10:22,  6.10s/it]
Training Epoch: 13/100, batch 170/375 completed (loss: 0.0281:  45%|████▌     | 85/187 [08:35<10:21,  6.10s/it]
Training Epoch: 13/100, batch 170/375 completed (loss: 0.0122:  45%|████▌     | 85/187 [08:35<10:22,  6.10s/it]
Training Epoch: 13/100, batch 170/375 completed (loss: 0.003:  45%|████▌     | 85/187 [08:35<10:22,  6.10s/it] 
Training Epoch: 13/100, batch 170/375 completed (loss: 0.0052:  45%|████▌     | 85/187 [08:35<10:22,  6.10s/it]
Training Epoch: 13/100, batch 170/375 completed (loss: 0.0281:  46%|████▌     | 86/187 [08:38<10:12,  6.06s/it]
Training Epoch: 13/100, batch 170/375 completed (loss: 0.003:  46%|████▌     | 86/187 [08:38<10:12,  6.06s/it]
Training Epoch: 13/100, batch 170/375 completed (loss: 0.0122:  46%|████▌     | 86/187 [08:38<10:12,  6.06s/it]
Training Epoch: 13/100, batch 170/375 completed (loss: 0.0052:  46%|████▌     | 86/187 [08:38<10:12,  6.06s/it]
Training Epoch: 13/100, batch 171/375 completed (loss: 0.0042:  46%|████▌     | 86/187 [08:38<10:12,  6.06s/it]
Training Epoch: 13/100, batch 171/375 completed (loss: 0.0035:  46%|████▌     | 86/187 [08:38<10:12,  6.06s/it]
Training Epoch: 13/100, batch 171/375 completed (loss: 0.0014:  46%|████▌     | 86/187 [08:38<10:12,  6.06s/it]
Training Epoch: 13/100, batch 171/375 completed (loss: 0.0254:  46%|████▌     | 86/187 [08:38<10:12,  6.06s/it]
Training Epoch: 13/100, batch 172/375 completed (loss: 0.004:  46%|████▌     | 86/187 [08:41<10:12,  6.06s/it] 
Training Epoch: 13/100, batch 172/375 completed (loss: 0.1841:  46%|████▌     | 86/187 [08:41<10:12,  6.06s/it]
Training Epoch: 13/100, batch 172/375 completed (loss: 0.0039:  46%|████▌     | 86/187 [08:41<10:12,  6.06s/it]
Training Epoch: 13/100, batch 172/375 completed (loss: 0.0014:  46%|████▌     | 86/187 [08:41<10:12,  6.06s/it]
Training Epoch: 13/100, batch 172/375 completed (loss: 0.004:  47%|████▋     | 87/187 [08:44<10:03,  6.04s/it]
Training Epoch: 13/100, batch 172/375 completed (loss: 0.1841:  47%|████▋     | 87/187 [08:44<10:03,  6.04s/it]
Training Epoch: 13/100, batch 172/375 completed (loss: 0.0039:  47%|████▋     | 87/187 [08:44<10:03,  6.04s/it]
Training Epoch: 13/100, batch 172/375 completed (loss: 0.0014:  47%|████▋     | 87/187 [08:44<10:03,  6.04s/it]
Training Epoch: 13/100, batch 173/375 completed (loss: 0.0035:  47%|████▋     | 87/187 [08:44<10:03,  6.04s/it]
Training Epoch: 13/100, batch 173/375 completed (loss: 0.0153:  47%|████▋     | 87/187 [08:44<10:03,  6.04s/it]
Training Epoch: 13/100, batch 173/375 completed (loss: 0.0206:  47%|████▋     | 87/187 [08:44<10:03,  6.04s/it]
Training Epoch: 13/100, batch 173/375 completed (loss: 0.0502:  47%|████▋     | 87/187 [08:44<10:03,  6.04s/it]
Training Epoch: 13/100, batch 174/375 completed (loss: 0.0112:  47%|████▋     | 87/187 [08:47<10:03,  6.04s/it]
Training Epoch: 13/100, batch 174/375 completed (loss: 0.005:  47%|████▋     | 87/187 [08:47<10:03,  6.04s/it] 
Training Epoch: 13/100, batch 174/375 completed (loss: 0.0358:  47%|████▋     | 87/187 [08:47<10:03,  6.04s/it]
Training Epoch: 13/100, batch 174/375 completed (loss: 0.0062:  47%|████▋     | 87/187 [08:47<10:03,  6.04s/it]
Training Epoch: 13/100, batch 174/375 completed (loss: 0.005:  47%|████▋     | 88/187 [08:50<10:06,  6.13s/it]
Training Epoch: 13/100, batch 174/375 completed (loss: 0.0358:  47%|████▋     | 88/187 [08:50<10:07,  6.13s/it]
Training Epoch: 13/100, batch 174/375 completed (loss: 0.0062:  47%|████▋     | 88/187 [08:50<10:07,  6.13s/it]
Training Epoch: 13/100, batch 174/375 completed (loss: 0.0112:  47%|████▋     | 88/187 [08:50<10:07,  6.13s/it]
Training Epoch: 13/100, batch 175/375 completed (loss: 0.0012:  47%|████▋     | 88/187 [08:50<10:07,  6.13s/it]
Training Epoch: 13/100, batch 175/375 completed (loss: 0.0442:  47%|████▋     | 88/187 [08:50<10:07,  6.13s/it]
Training Epoch: 13/100, batch 175/375 completed (loss: 0.0328:  47%|████▋     | 88/187 [08:50<10:06,  6.13s/it]
Training Epoch: 13/100, batch 175/375 completed (loss: 0.0032:  47%|████▋     | 88/187 [08:50<10:07,  6.13s/it]
Training Epoch: 13/100, batch 176/375 completed (loss: 0.013:  47%|████▋     | 88/187 [08:53<10:07,  6.13s/it] 
Training Epoch: 13/100, batch 176/375 completed (loss: 0.0988:  47%|████▋     | 88/187 [08:53<10:07,  6.13s/it]
Training Epoch: 13/100, batch 176/375 completed (loss: 0.003:  47%|████▋     | 88/187 [08:53<10:06,  6.13s/it] 
Training Epoch: 13/100, batch 176/375 completed (loss: 0.0342:  47%|████▋     | 88/187 [08:53<10:07,  6.13s/it]
Training Epoch: 13/100, batch 176/375 completed (loss: 0.003:  48%|████▊     | 89/187 [08:56<09:53,  6.05s/it]
Training Epoch: 13/100, batch 176/375 completed (loss: 0.013:  48%|████▊     | 89/187 [08:56<09:53,  6.06s/it]
Training Epoch: 13/100, batch 176/375 completed (loss: 0.0342:  48%|████▊     | 89/187 [08:56<09:53,  6.06s/it]
Training Epoch: 13/100, batch 176/375 completed (loss: 0.0988:  48%|████▊     | 89/187 [08:56<09:54,  6.06s/it]
Training Epoch: 13/100, batch 177/375 completed (loss: 0.0132:  48%|████▊     | 89/187 [08:56<09:54,  6.06s/it]
Training Epoch: 13/100, batch 177/375 completed (loss: 0.0011:  48%|████▊     | 89/187 [08:56<09:53,  6.06s/it]
Training Epoch: 13/100, batch 177/375 completed (loss: 0.0867:  48%|████▊     | 89/187 [08:56<09:53,  6.05s/it]
Training Epoch: 13/100, batch 177/375 completed (loss: 0.0005:  48%|████▊     | 89/187 [08:56<09:53,  6.06s/it]
Training Epoch: 13/100, batch 178/375 completed (loss: 0.0067:  48%|████▊     | 89/187 [08:59<09:53,  6.06s/it]
Training Epoch: 13/100, batch 178/375 completed (loss: 0.0026:  48%|████▊     | 89/187 [08:59<09:54,  6.06s/it]
Training Epoch: 13/100, batch 178/375 completed (loss: 0.0035:  48%|████▊     | 89/187 [08:59<09:53,  6.05s/it]
Training Epoch: 13/100, batch 178/375 completed (loss: 0.0476:  48%|████▊     | 89/187 [08:59<09:53,  6.06s/it]
Training Epoch: 13/100, batch 178/375 completed (loss: 0.0026:  48%|████▊     | 90/187 [09:02<09:49,  6.07s/it]
Training Epoch: 13/100, batch 178/375 completed (loss: 0.0067:  48%|████▊     | 90/187 [09:02<09:49,  6.08s/it]
Training Epoch: 13/100, batch 178/375 completed (loss: 0.0035:  48%|████▊     | 90/187 [09:02<09:49,  6.08s/it]
Training Epoch: 13/100, batch 178/375 completed (loss: 0.0476:  48%|████▊     | 90/187 [09:02<09:49,  6.08s/it]
Training Epoch: 13/100, batch 179/375 completed (loss: 0.0336:  48%|████▊     | 90/187 [09:02<09:49,  6.07s/it]
Training Epoch: 13/100, batch 179/375 completed (loss: 0.0238:  48%|████▊     | 90/187 [09:02<09:49,  6.08s/it]
Training Epoch: 13/100, batch 179/375 completed (loss: 0.0369:  48%|████▊     | 90/187 [09:02<09:49,  6.08s/it]
Training Epoch: 13/100, batch 179/375 completed (loss: 0.1366:  48%|████▊     | 90/187 [09:02<09:49,  6.08s/it]
Training Epoch: 13/100, batch 180/375 completed (loss: 0.0092:  48%|████▊     | 90/187 [09:06<09:49,  6.07s/it]
Training Epoch: 13/100, batch 180/375 completed (loss: 0.016:  48%|████▊     | 90/187 [09:05<09:49,  6.08s/it] 
Training Epoch: 13/100, batch 180/375 completed (loss: 0.0325:  48%|████▊     | 90/187 [09:06<09:49,  6.08s/it]
Training Epoch: 13/100, batch 180/375 completed (loss: 0.0042:  48%|████▊     | 90/187 [09:06<09:49,  6.08s/it]
Training Epoch: 13/100, batch 180/375 completed (loss: 0.016:  49%|████▊     | 91/187 [09:09<09:45,  6.10s/it]
Training Epoch: 13/100, batch 180/375 completed (loss: 0.0092:  49%|████▊     | 91/187 [09:09<09:45,  6.10s/it]
Training Epoch: 13/100, batch 180/375 completed (loss: 0.0042:  49%|████▊     | 91/187 [09:09<09:45,  6.10s/it]
Training Epoch: 13/100, batch 180/375 completed (loss: 0.0325:  49%|████▊     | 91/187 [09:09<09:45,  6.10s/it]
Training Epoch: 13/100, batch 181/375 completed (loss: 0.0029:  49%|████▊     | 91/187 [09:09<09:45,  6.10s/it]
Training Epoch: 13/100, batch 181/375 completed (loss: 0.0029:  49%|████▊     | 91/187 [09:09<09:45,  6.10s/it]
Training Epoch: 13/100, batch 181/375 completed (loss: 0.1246:  49%|████▊     | 91/187 [09:09<09:45,  6.10s/it]
Training Epoch: 13/100, batch 181/375 completed (loss: 0.0018:  49%|████▊     | 91/187 [09:09<09:45,  6.10s/it]
Training Epoch: 13/100, batch 182/375 completed (loss: 0.0135:  49%|████▊     | 91/187 [09:11<09:45,  6.10s/it]
Training Epoch: 13/100, batch 182/375 completed (loss: 0.0076:  49%|████▊     | 91/187 [09:12<09:45,  6.10s/it]
Training Epoch: 13/100, batch 182/375 completed (loss: 0.0085:  49%|████▊     | 91/187 [09:11<09:45,  6.10s/it]
Training Epoch: 13/100, batch 182/375 completed (loss: 0.0028:  49%|████▊     | 91/187 [09:12<09:45,  6.10s/it]
Training Epoch: 13/100, batch 182/375 completed (loss: 0.0135:  49%|████▉     | 92/187 [09:14<09:31,  6.01s/it]
Training Epoch: 13/100, batch 182/375 completed (loss: 0.0085:  49%|████▉     | 92/187 [09:14<09:31,  6.01s/it]
Training Epoch: 13/100, batch 182/375 completed (loss: 0.0076:  49%|████▉     | 92/187 [09:14<09:31,  6.01s/it]
Training Epoch: 13/100, batch 182/375 completed (loss: 0.0028:  49%|████▉     | 92/187 [09:14<09:31,  6.02s/it]
Training Epoch: 13/100, batch 183/375 completed (loss: 0.0101:  49%|████▉     | 92/187 [09:14<09:31,  6.01s/it]
Training Epoch: 13/100, batch 183/375 completed (loss: 0.0564:  49%|████▉     | 92/187 [09:14<09:31,  6.01s/it]
Training Epoch: 13/100, batch 183/375 completed (loss: 0.0056:  49%|████▉     | 92/187 [09:14<09:31,  6.01s/it]
Training Epoch: 13/100, batch 183/375 completed (loss: 0.0028:  49%|████▉     | 92/187 [09:14<09:31,  6.02s/it]
Training Epoch: 13/100, batch 184/375 completed (loss: 0.0417:  49%|████▉     | 92/187 [09:17<09:31,  6.01s/it]
Training Epoch: 13/100, batch 184/375 completed (loss: 0.0067:  49%|████▉     | 92/187 [09:17<09:31,  6.01s/it]
Training Epoch: 13/100, batch 184/375 completed (loss: 0.0106:  49%|████▉     | 92/187 [09:17<09:31,  6.01s/it]
Training Epoch: 13/100, batch 184/375 completed (loss: 0.0071:  49%|████▉     | 92/187 [09:18<09:31,  6.02s/it]
Training Epoch: 13/100, batch 184/375 completed (loss: 0.0071:  50%|████▉     | 93/187 [09:21<09:37,  6.14s/it]
Training Epoch: 13/100, batch 184/375 completed (loss: 0.0106:  50%|████▉     | 93/187 [09:21<09:37,  6.14s/it]
Training Epoch: 13/100, batch 184/375 completed (loss: 0.0067:  50%|████▉     | 93/187 [09:21<09:37,  6.15s/it]
Training Epoch: 13/100, batch 184/375 completed (loss: 0.0417:  50%|████▉     | 93/187 [09:21<09:37,  6.14s/it]
Training Epoch: 13/100, batch 185/375 completed (loss: 0.0045:  50%|████▉     | 93/187 [09:21<09:37,  6.14s/it]
Training Epoch: 13/100, batch 185/375 completed (loss: 0.0005:  50%|████▉     | 93/187 [09:21<09:37,  6.15s/it]
Training Epoch: 13/100, batch 185/375 completed (loss: 0.0086:  50%|████▉     | 93/187 [09:21<09:37,  6.14s/it]
Training Epoch: 13/100, batch 185/375 completed (loss: 0.004:  50%|████▉     | 93/187 [09:21<09:37,  6.14s/it] 
Training Epoch: 13/100, batch 186/375 completed (loss: 0.0024:  50%|████▉     | 93/187 [09:24<09:37,  6.14s/it]
Training Epoch: 13/100, batch 186/375 completed (loss: 0.0068:  50%|████▉     | 93/187 [09:24<09:37,  6.14s/it]
Training Epoch: 13/100, batch 186/375 completed (loss: 0.1937:  50%|████▉     | 93/187 [09:24<09:37,  6.15s/it]
Training Epoch: 13/100, batch 186/375 completed (loss: 0.0437:  50%|████▉     | 93/187 [09:24<09:37,  6.14s/it]
Training Epoch: 13/100, batch 186/375 completed (loss: 0.0024:  50%|█████     | 94/187 [09:27<09:21,  6.04s/it]
Training Epoch: 13/100, batch 186/375 completed (loss: 0.1937:  50%|█████     | 94/187 [09:27<09:22,  6.04s/it]
Training Epoch: 13/100, batch 186/375 completed (loss: 0.0068:  50%|█████     | 94/187 [09:27<09:22,  6.04s/it]
Training Epoch: 13/100, batch 186/375 completed (loss: 0.0437:  50%|█████     | 94/187 [09:27<09:22,  6.04s/it]
Training Epoch: 13/100, batch 187/375 completed (loss: 0.0609:  50%|█████     | 94/187 [09:27<09:22,  6.04s/it]
Training Epoch: 13/100, batch 187/375 completed (loss: 0.0091:  50%|█████     | 94/187 [09:27<09:22,  6.04s/it]
Training Epoch: 13/100, batch 187/375 completed (loss: 0.0048:  50%|█████     | 94/187 [09:27<09:21,  6.04s/it]
Training Epoch: 13/100, batch 187/375 completed (loss: 0.0359:  50%|█████     | 94/187 [09:27<09:22,  6.04s/it]
Training Epoch: 13/100, batch 188/375 completed (loss: 0.0078:  50%|█████     | 94/187 [09:30<09:22,  6.04s/it]
Training Epoch: 13/100, batch 188/375 completed (loss: 0.0189:  50%|█████     | 94/187 [09:30<09:21,  6.04s/it]
Training Epoch: 13/100, batch 188/375 completed (loss: 0.0115:  50%|█████     | 94/187 [09:30<09:22,  6.04s/it]
Training Epoch: 13/100, batch 188/375 completed (loss: 0.0384:  50%|█████     | 94/187 [09:30<09:22,  6.04s/it]
Training Epoch: 13/100, batch 188/375 completed (loss: 0.0189:  51%|█████     | 95/187 [09:33<09:17,  6.06s/it]
Training Epoch: 13/100, batch 188/375 completed (loss: 0.0384:  51%|█████     | 95/187 [09:33<09:17,  6.06s/it]
Training Epoch: 13/100, batch 188/375 completed (loss: 0.0078:  51%|█████     | 95/187 [09:33<09:17,  6.06s/it]
Training Epoch: 13/100, batch 188/375 completed (loss: 0.0115:  51%|█████     | 95/187 [09:33<09:17,  6.06s/it]
Training Epoch: 13/100, batch 189/375 completed (loss: 0.0418:  51%|█████     | 95/187 [09:33<09:17,  6.06s/it]
Training Epoch: 13/100, batch 189/375 completed (loss: 0.012:  51%|█████     | 95/187 [09:33<09:17,  6.06s/it] 
Training Epoch: 13/100, batch 189/375 completed (loss: 0.0599:  51%|█████     | 95/187 [09:33<09:17,  6.06s/it]
Training Epoch: 13/100, batch 189/375 completed (loss: 0.0359:  51%|█████     | 95/187 [09:33<09:17,  6.06s/it]
Training Epoch: 13/100, batch 190/375 completed (loss: 0.0187:  51%|█████     | 95/187 [09:36<09:17,  6.06s/it]
Training Epoch: 13/100, batch 190/375 completed (loss: 0.0054:  51%|█████     | 95/187 [09:36<09:17,  6.06s/it]
Training Epoch: 13/100, batch 190/375 completed (loss: 0.203:  51%|█████     | 95/187 [09:36<09:17,  6.06s/it] 
Training Epoch: 13/100, batch 190/375 completed (loss: 0.0131:  51%|█████     | 95/187 [09:36<09:17,  6.06s/it]
Training Epoch: 13/100, batch 190/375 completed (loss: 0.0187:  51%|█████▏    | 96/187 [09:39<09:13,  6.08s/it]
Training Epoch: 13/100, batch 190/375 completed (loss: 0.0131:  51%|█████▏    | 96/187 [09:39<09:13,  6.08s/it]
Training Epoch: 13/100, batch 190/375 completed (loss: 0.0054:  51%|█████▏    | 96/187 [09:39<09:13,  6.08s/it]
Training Epoch: 13/100, batch 190/375 completed (loss: 0.203:  51%|█████▏    | 96/187 [09:39<09:14,  6.09s/it]
Training Epoch: 13/100, batch 191/375 completed (loss: 0.0039:  51%|█████▏    | 96/187 [09:39<09:14,  6.09s/it]
Training Epoch: 13/100, batch 191/375 completed (loss: 0.0463:  51%|█████▏    | 96/187 [09:39<09:13,  6.08s/it]
Training Epoch: 13/100, batch 191/375 completed (loss: 0.0127:  51%|█████▏    | 96/187 [09:39<09:13,  6.08s/it]
Training Epoch: 13/100, batch 191/375 completed (loss: 0.0144:  51%|█████▏    | 96/187 [09:39<09:13,  6.08s/it]
Training Epoch: 13/100, batch 192/375 completed (loss: 0.0339:  51%|█████▏    | 96/187 [09:42<09:14,  6.09s/it]
Training Epoch: 13/100, batch 192/375 completed (loss: 0.0829:  51%|█████▏    | 96/187 [09:42<09:13,  6.08s/it]
Training Epoch: 13/100, batch 192/375 completed (loss: 0.0246:  51%|█████▏    | 96/187 [09:42<09:13,  6.08s/it]
Training Epoch: 13/100, batch 192/375 completed (loss: 0.0017:  51%|█████▏    | 96/187 [09:42<09:13,  6.08s/it]
Training Epoch: 13/100, batch 192/375 completed (loss: 0.0339:  52%|█████▏    | 97/187 [09:45<09:01,  6.02s/it]
Training Epoch: 13/100, batch 192/375 completed (loss: 0.0829:  52%|█████▏    | 97/187 [09:45<09:01,  6.02s/it]
Training Epoch: 13/100, batch 192/375 completed (loss: 0.0246:  52%|█████▏    | 97/187 [09:45<09:02,  6.02s/it]
Training Epoch: 13/100, batch 192/375 completed (loss: 0.0017:  52%|█████▏    | 97/187 [09:45<09:02,  6.03s/it]
Training Epoch: 13/100, batch 193/375 completed (loss: 0.0164:  52%|█████▏    | 97/187 [09:45<09:01,  6.02s/it]
Training Epoch: 13/100, batch 193/375 completed (loss: 0.0121:  52%|█████▏    | 97/187 [09:45<09:02,  6.02s/it]
Training Epoch: 13/100, batch 193/375 completed (loss: 0.0601:  52%|█████▏    | 97/187 [09:45<09:01,  6.02s/it]
Training Epoch: 13/100, batch 193/375 completed (loss: 0.0052:  52%|█████▏    | 97/187 [09:45<09:02,  6.03s/it]
Training Epoch: 13/100, batch 194/375 completed (loss: 0.0144:  52%|█████▏    | 97/187 [09:48<09:01,  6.02s/it]
Training Epoch: 13/100, batch 194/375 completed (loss: 0.0739:  52%|█████▏    | 97/187 [09:48<09:01,  6.02s/it]
Training Epoch: 13/100, batch 194/375 completed (loss: 0.0113:  52%|█████▏    | 97/187 [09:48<09:02,  6.02s/it]
Training Epoch: 13/100, batch 194/375 completed (loss: 0.0342:  52%|█████▏    | 97/187 [09:48<09:02,  6.03s/it]
Training Epoch: 13/100, batch 194/375 completed (loss: 0.0144:  52%|█████▏    | 98/187 [09:51<09:02,  6.10s/it]
Training Epoch: 13/100, batch 194/375 completed (loss: 0.0739:  52%|█████▏    | 98/187 [09:51<09:02,  6.10s/it]
Training Epoch: 13/100, batch 194/375 completed (loss: 0.0342:  52%|█████▏    | 98/187 [09:51<09:03,  6.10s/it]
Training Epoch: 13/100, batch 194/375 completed (loss: 0.0113:  52%|█████▏    | 98/187 [09:51<09:03,  6.10s/it]
Training Epoch: 13/100, batch 195/375 completed (loss: 0.0026:  52%|█████▏    | 98/187 [09:51<09:03,  6.10s/it]
Training Epoch: 13/100, batch 195/375 completed (loss: 0.0034:  52%|█████▏    | 98/187 [09:51<09:02,  6.10s/it]
Training Epoch: 13/100, batch 195/375 completed (loss: 0.0036:  52%|█████▏    | 98/187 [09:51<09:02,  6.10s/it]
Training Epoch: 13/100, batch 195/375 completed (loss: 0.0051:  52%|█████▏    | 98/187 [09:51<09:03,  6.10s/it]
Training Epoch: 13/100, batch 196/375 completed (loss: 0.0423:  52%|█████▏    | 98/187 [09:54<09:02,  6.10s/it]
Training Epoch: 13/100, batch 196/375 completed (loss: 0.0107:  52%|█████▏    | 98/187 [09:54<09:03,  6.10s/it]
Training Epoch: 13/100, batch 196/375 completed (loss: 0.0099:  52%|█████▏    | 98/187 [09:54<09:02,  6.10s/it]
Training Epoch: 13/100, batch 196/375 completed (loss: 0.0082:  52%|█████▏    | 98/187 [09:54<09:03,  6.10s/it]
Training Epoch: 13/100, batch 196/375 completed (loss: 0.0423:  53%|█████▎    | 99/187 [09:57<08:55,  6.09s/it]
Training Epoch: 13/100, batch 196/375 completed (loss: 0.0082:  53%|█████▎    | 99/187 [09:57<08:55,  6.09s/it]
Training Epoch: 13/100, batch 196/375 completed (loss: 0.0099:  53%|█████▎    | 99/187 [09:57<08:56,  6.09s/it]
Training Epoch: 13/100, batch 196/375 completed (loss: 0.0107:  53%|█████▎    | 99/187 [09:57<08:56,  6.09s/it]
Training Epoch: 13/100, batch 197/375 completed (loss: 0.0412:  53%|█████▎    | 99/187 [09:57<08:56,  6.09s/it]
Training Epoch: 13/100, batch 197/375 completed (loss: 0.0252:  53%|█████▎    | 99/187 [09:57<08:56,  6.09s/it]
Training Epoch: 13/100, batch 197/375 completed (loss: 0.0199:  53%|█████▎    | 99/187 [09:57<08:55,  6.09s/it]
Training Epoch: 13/100, batch 197/375 completed (loss: 0.0013:  53%|█████▎    | 99/187 [09:57<08:55,  6.09s/it]
Training Epoch: 13/100, batch 198/375 completed (loss: 0.0067:  53%|█████▎    | 99/187 [10:00<08:56,  6.09s/it]
Training Epoch: 13/100, batch 198/375 completed (loss: 0.0296:  53%|█████▎    | 99/187 [10:00<08:55,  6.09s/it]
Training Epoch: 13/100, batch 198/375 completed (loss: 0.0397:  53%|█████▎    | 99/187 [10:00<08:56,  6.09s/it]
Training Epoch: 13/100, batch 198/375 completed (loss: 0.0348:  53%|█████▎    | 99/187 [10:00<08:55,  6.09s/it]
Training Epoch: 13/100, batch 198/375 completed (loss: 0.0067:  53%|█████▎    | 100/187 [10:03<08:43,  6.01s/it]
Training Epoch: 13/100, batch 198/375 completed (loss: 0.0397:  53%|█████▎    | 100/187 [10:03<08:43,  6.01s/it]
Training Epoch: 13/100, batch 198/375 completed (loss: 0.0348:  53%|█████▎    | 100/187 [10:03<08:43,  6.01s/it]
Training Epoch: 13/100, batch 198/375 completed (loss: 0.0296:  53%|█████▎    | 100/187 [10:03<08:43,  6.02s/it]
Training Epoch: 13/100, batch 199/375 completed (loss: 0.0078:  53%|█████▎    | 100/187 [10:03<08:43,  6.02s/it]
Training Epoch: 13/100, batch 199/375 completed (loss: 0.0148:  53%|█████▎    | 100/187 [10:03<08:43,  6.01s/it]
Training Epoch: 13/100, batch 199/375 completed (loss: 0.0001:  53%|█████▎    | 100/187 [10:03<08:43,  6.01s/it]
Training Epoch: 13/100, batch 199/375 completed (loss: 0.0046:  53%|█████▎    | 100/187 [10:03<08:43,  6.01s/it]
Training Epoch: 13/100, batch 200/375 completed (loss: 0.0056:  53%|█████▎    | 100/187 [10:06<08:43,  6.01s/it]
Training Epoch: 13/100, batch 200/375 completed (loss: 0.0639:  53%|█████▎    | 100/187 [10:06<08:43,  6.02s/it]
Training Epoch: 13/100, batch 200/375 completed (loss: 0.1346:  53%|█████▎    | 100/187 [10:06<08:43,  6.01s/it]
Training Epoch: 13/100, batch 200/375 completed (loss: 0.0041:  53%|█████▎    | 100/187 [10:06<08:43,  6.01s/it]
Training Epoch: 13/100, batch 200/375 completed (loss: 0.0639:  54%|█████▍    | 101/187 [10:09<08:44,  6.09s/it]
Training Epoch: 13/100, batch 200/375 completed (loss: 0.0056:  54%|█████▍    | 101/187 [10:09<08:43,  6.09s/it]
Training Epoch: 13/100, batch 200/375 completed (loss: 0.1346:  54%|█████▍    | 101/187 [10:09<08:44,  6.09s/it]
Training Epoch: 13/100, batch 200/375 completed (loss: 0.0041:  54%|█████▍    | 101/187 [10:09<08:44,  6.09s/it]
Training Epoch: 13/100, batch 201/375 completed (loss: 0.0021:  54%|█████▍    | 101/187 [10:09<08:43,  6.09s/it]
Training Epoch: 13/100, batch 201/375 completed (loss: 0.058:  54%|█████▍    | 101/187 [10:09<08:44,  6.09s/it] 
Training Epoch: 13/100, batch 201/375 completed (loss: 0.2121:  54%|█████▍    | 101/187 [10:09<08:44,  6.09s/it]
Training Epoch: 13/100, batch 201/375 completed (loss: 0.0013:  54%|█████▍    | 101/187 [10:09<08:44,  6.09s/it]
Training Epoch: 13/100, batch 202/375 completed (loss: 0.0071:  54%|█████▍    | 101/187 [10:12<08:44,  6.09s/it]
Training Epoch: 13/100, batch 202/375 completed (loss: 0.0102:  54%|█████▍    | 101/187 [10:12<08:44,  6.09s/it]
Training Epoch: 13/100, batch 202/375 completed (loss: 0.0012:  54%|█████▍    | 101/187 [10:12<08:43,  6.09s/it]
Training Epoch: 13/100, batch 202/375 completed (loss: 0.0024:  54%|█████▍    | 101/187 [10:12<08:44,  6.09s/it]
Training Epoch: 13/100, batch 202/375 completed (loss: 0.0102:  55%|█████▍    | 102/187 [10:15<08:35,  6.07s/it]
Training Epoch: 13/100, batch 202/375 completed (loss: 0.0071:  55%|█████▍    | 102/187 [10:15<08:36,  6.07s/it]
Training Epoch: 13/100, batch 202/375 completed (loss: 0.0024:  55%|█████▍    | 102/187 [10:15<08:36,  6.07s/it]
Training Epoch: 13/100, batch 202/375 completed (loss: 0.0012:  55%|█████▍    | 102/187 [10:15<08:36,  6.07s/it]
Training Epoch: 13/100, batch 203/375 completed (loss: 0.032:  55%|█████▍    | 102/187 [10:15<08:35,  6.07s/it] 
Training Epoch: 13/100, batch 203/375 completed (loss: 0.0076:  55%|█████▍    | 102/187 [10:15<08:36,  6.07s/it]
Training Epoch: 13/100, batch 203/375 completed (loss: 0.0131:  55%|█████▍    | 102/187 [10:15<08:36,  6.07s/it]
Training Epoch: 13/100, batch 203/375 completed (loss: 0.0283:  55%|█████▍    | 102/187 [10:15<08:36,  6.07s/it]
Training Epoch: 13/100, batch 204/375 completed (loss: 0.1134:  55%|█████▍    | 102/187 [10:18<08:36,  6.07s/it]
Training Epoch: 13/100, batch 204/375 completed (loss: 0.0049:  55%|█████▍    | 102/187 [10:18<08:36,  6.07s/it]
Training Epoch: 13/100, batch 204/375 completed (loss: 0.0024:  55%|█████▍    | 102/187 [10:18<08:35,  6.07s/it]
Training Epoch: 13/100, batch 204/375 completed (loss: 0.0137:  55%|█████▍    | 102/187 [10:18<08:36,  6.07s/it]
Training Epoch: 13/100, batch 204/375 completed (loss: 0.0049:  55%|█████▌    | 103/187 [10:21<08:31,  6.09s/it]
Training Epoch: 13/100, batch 204/375 completed (loss: 0.0024:  55%|█████▌    | 103/187 [10:21<08:31,  6.09s/it]
Training Epoch: 13/100, batch 204/375 completed (loss: 0.1134:  55%|█████▌    | 103/187 [10:21<08:31,  6.09s/it]
Training Epoch: 13/100, batch 204/375 completed (loss: 0.0137:  55%|█████▌    | 103/187 [10:21<08:31,  6.09s/it]
Training Epoch: 13/100, batch 205/375 completed (loss: 0.0075:  55%|█████▌    | 103/187 [10:21<08:31,  6.09s/it]
Training Epoch: 13/100, batch 205/375 completed (loss: 0.1559:  55%|█████▌    | 103/187 [10:21<08:31,  6.09s/it]
Training Epoch: 13/100, batch 205/375 completed (loss: 0.0155:  55%|█████▌    | 103/187 [10:21<08:31,  6.09s/it]
Training Epoch: 13/100, batch 205/375 completed (loss: 0.0497:  55%|█████▌    | 103/187 [10:21<08:31,  6.09s/it]
Training Epoch: 13/100, batch 206/375 completed (loss: 0.0071:  55%|█████▌    | 103/187 [10:24<08:31,  6.09s/it]
Training Epoch: 13/100, batch 206/375 completed (loss: 0.1684:  55%|█████▌    | 103/187 [10:24<08:31,  6.09s/it]
Training Epoch: 13/100, batch 206/375 completed (loss: 0.0604:  55%|█████▌    | 103/187 [10:24<08:31,  6.09s/it]
Training Epoch: 13/100, batch 206/375 completed (loss: 0.0226:  55%|█████▌    | 103/187 [10:24<08:31,  6.09s/it]
Training Epoch: 13/100, batch 206/375 completed (loss: 0.0604:  56%|█████▌    | 104/187 [10:28<08:26,  6.11s/it]
Training Epoch: 13/100, batch 206/375 completed (loss: 0.1684:  56%|█████▌    | 104/187 [10:28<08:27,  6.11s/it]
Training Epoch: 13/100, batch 206/375 completed (loss: 0.0071:  56%|█████▌    | 104/187 [10:28<08:27,  6.11s/it]
Training Epoch: 13/100, batch 206/375 completed (loss: 0.0226:  56%|█████▌    | 104/187 [10:28<08:27,  6.11s/it]
Training Epoch: 13/100, batch 207/375 completed (loss: 0.0302:  56%|█████▌    | 104/187 [10:28<08:27,  6.11s/it]
Training Epoch: 13/100, batch 207/375 completed (loss: 0.0011:  56%|█████▌    | 104/187 [10:28<08:26,  6.11s/it]
Training Epoch: 13/100, batch 207/375 completed (loss: 0.0518:  56%|█████▌    | 104/187 [10:28<08:27,  6.11s/it]
Training Epoch: 13/100, batch 207/375 completed (loss: 0.0096:  56%|█████▌    | 104/187 [10:28<08:27,  6.11s/it]
Training Epoch: 13/100, batch 208/375 completed (loss: 0.0036:  56%|█████▌    | 104/187 [10:30<08:26,  6.11s/it]
Training Epoch: 13/100, batch 208/375 completed (loss: 0.0387:  56%|█████▌    | 104/187 [10:30<08:27,  6.11s/it]
Training Epoch: 13/100, batch 208/375 completed (loss: 0.0382:  56%|█████▌    | 104/187 [10:30<08:27,  6.11s/it]
Training Epoch: 13/100, batch 208/375 completed (loss: 0.0018:  56%|█████▌    | 104/187 [10:31<08:27,  6.11s/it]
Training Epoch: 13/100, batch 208/375 completed (loss: 0.0387:  56%|█████▌    | 105/187 [10:33<08:14,  6.03s/it]
Training Epoch: 13/100, batch 208/375 completed (loss: 0.0036:  56%|█████▌    | 105/187 [10:33<08:14,  6.03s/it]
Training Epoch: 13/100, batch 208/375 completed (loss: 0.0382:  56%|█████▌    | 105/187 [10:33<08:14,  6.03s/it]
Training Epoch: 13/100, batch 208/375 completed (loss: 0.0018:  56%|█████▌    | 105/187 [10:33<08:14,  6.03s/it]
Training Epoch: 13/100, batch 209/375 completed (loss: 0.0031:  56%|█████▌    | 105/187 [10:33<08:14,  6.03s/it]
Training Epoch: 13/100, batch 209/375 completed (loss: 0.0075:  56%|█████▌    | 105/187 [10:33<08:14,  6.03s/it]
Training Epoch: 13/100, batch 209/375 completed (loss: 0.0412:  56%|█████▌    | 105/187 [10:33<08:14,  6.03s/it]
Training Epoch: 13/100, batch 209/375 completed (loss: 0.0023:  56%|█████▌    | 105/187 [10:33<08:14,  6.03s/it]
Training Epoch: 13/100, batch 210/375 completed (loss: 0.0005:  56%|█████▌    | 105/187 [10:37<08:14,  6.03s/it]
Training Epoch: 13/100, batch 210/375 completed (loss: 0.0064:  56%|█████▌    | 105/187 [10:37<08:14,  6.03s/it]
Training Epoch: 13/100, batch 210/375 completed (loss: 0.0093:  56%|█████▌    | 105/187 [10:37<08:14,  6.03s/it]
Training Epoch: 13/100, batch 210/375 completed (loss: 0.0134:  56%|█████▌    | 105/187 [10:37<08:14,  6.03s/it]
Training Epoch: 13/100, batch 210/375 completed (loss: 0.0064:  57%|█████▋    | 106/187 [10:40<08:14,  6.10s/it]
Training Epoch: 13/100, batch 210/375 completed (loss: 0.0093:  57%|█████▋    | 106/187 [10:40<08:14,  6.10s/it]
Training Epoch: 13/100, batch 210/375 completed (loss: 0.0134:  57%|█████▋    | 106/187 [10:40<08:14,  6.10s/it]
Training Epoch: 13/100, batch 210/375 completed (loss: 0.0005:  57%|█████▋    | 106/187 [10:40<08:14,  6.11s/it]
Training Epoch: 13/100, batch 211/375 completed (loss: 0.0538:  57%|█████▋    | 106/187 [10:40<08:14,  6.10s/it]
Training Epoch: 13/100, batch 211/375 completed (loss: 0.0684:  57%|█████▋    | 106/187 [10:40<08:14,  6.11s/it]
Training Epoch: 13/100, batch 211/375 completed (loss: 0.0025:  57%|█████▋    | 106/187 [10:40<08:14,  6.10s/it]
Training Epoch: 13/100, batch 211/375 completed (loss: 0.0604:  57%|█████▋    | 106/187 [10:40<08:14,  6.10s/it]
Training Epoch: 13/100, batch 212/375 completed (loss: 0.001:  57%|█████▋    | 106/187 [10:43<08:14,  6.10s/it] 
Training Epoch: 13/100, batch 212/375 completed (loss: 0.0078:  57%|█████▋    | 106/187 [10:43<08:14,  6.11s/it]
Training Epoch: 13/100, batch 212/375 completed (loss: 0.1289:  57%|█████▋    | 106/187 [10:43<08:14,  6.10s/it]
Training Epoch: 13/100, batch 212/375 completed (loss: 0.027:  57%|█████▋    | 106/187 [10:43<08:14,  6.10s/it] 
Training Epoch: 13/100, batch 212/375 completed (loss: 0.0078:  57%|█████▋    | 107/187 [10:46<08:06,  6.08s/it]
Training Epoch: 13/100, batch 212/375 completed (loss: 0.1289:  57%|█████▋    | 107/187 [10:46<08:06,  6.08s/it]
Training Epoch: 13/100, batch 212/375 completed (loss: 0.027:  57%|█████▋    | 107/187 [10:46<08:06,  6.08s/it]
Training Epoch: 13/100, batch 212/375 completed (loss: 0.001:  57%|█████▋    | 107/187 [10:46<08:06,  6.08s/it]
Training Epoch: 13/100, batch 213/375 completed (loss: 0.0135:  57%|█████▋    | 107/187 [10:46<08:06,  6.08s/it]
Training Epoch: 13/100, batch 213/375 completed (loss: 0.0016:  57%|█████▋    | 107/187 [10:46<08:06,  6.08s/it]
Training Epoch: 13/100, batch 213/375 completed (loss: 0.002:  57%|█████▋    | 107/187 [10:46<08:06,  6.08s/it]
Training Epoch: 13/100, batch 213/375 completed (loss: 0.0701:  57%|█████▋    | 107/187 [10:46<08:06,  6.08s/it]
Training Epoch: 13/100, batch 214/375 completed (loss: 0.0121:  57%|█████▋    | 107/187 [10:49<08:06,  6.08s/it]
Training Epoch: 13/100, batch 214/375 completed (loss: 0.0046:  57%|█████▋    | 107/187 [10:49<08:06,  6.08s/it]
Training Epoch: 13/100, batch 214/375 completed (loss: 0.005:  57%|█████▋    | 107/187 [10:49<08:06,  6.08s/it]
Training Epoch: 13/100, batch 214/375 completed (loss: 0.0358:  57%|█████▋    | 107/187 [10:49<08:06,  6.08s/it]
Training Epoch: 13/100, batch 214/375 completed (loss: 0.0121:  58%|█████▊    | 108/187 [10:52<07:58,  6.05s/it]
Training Epoch: 13/100, batch 214/375 completed (loss: 0.0046:  58%|█████▊    | 108/187 [10:52<07:58,  6.05s/it]
Training Epoch: 13/100, batch 214/375 completed (loss: 0.005:  58%|█████▊    | 108/187 [10:52<07:58,  6.06s/it]
Training Epoch: 13/100, batch 214/375 completed (loss: 0.0358:  58%|█████▊    | 108/187 [10:52<08:00,  6.09s/it]
Training Epoch: 13/100, batch 215/375 completed (loss: 0.0117:  58%|█████▊    | 108/187 [10:52<07:58,  6.05s/it]
Training Epoch: 13/100, batch 215/375 completed (loss: 0.1364:  58%|█████▊    | 108/187 [10:52<07:58,  6.06s/it]
Training Epoch: 13/100, batch 215/375 completed (loss: 0.003:  58%|█████▊    | 108/187 [10:52<07:58,  6.05s/it] 
Training Epoch: 13/100, batch 215/375 completed (loss: 0.013:  58%|█████▊    | 108/187 [10:52<08:00,  6.09s/it] 
Training Epoch: 13/100, batch 216/375 completed (loss: 0.0129:  58%|█████▊    | 108/187 [10:55<07:58,  6.05s/it]
Training Epoch: 13/100, batch 216/375 completed (loss: 0.0432:  58%|█████▊    | 108/187 [10:55<07:58,  6.06s/it]
Training Epoch: 13/100, batch 216/375 completed (loss: 0.2791:  58%|█████▊    | 108/187 [10:55<07:58,  6.05s/it]
Training Epoch: 13/100, batch 216/375 completed (loss: 0.057:  58%|█████▊    | 108/187 [10:55<08:00,  6.09s/it]
Training Epoch: 13/100, batch 216/375 completed (loss: 0.2791:  58%|█████▊    | 109/187 [10:58<07:57,  6.12s/it]
Training Epoch: 13/100, batch 216/375 completed (loss: 0.057:  58%|█████▊    | 109/187 [10:58<07:56,  6.11s/it]
Training Epoch: 13/100, batch 216/375 completed (loss: 0.0129:  58%|█████▊    | 109/187 [10:58<07:57,  6.12s/it]
Training Epoch: 13/100, batch 216/375 completed (loss: 0.0432:  58%|█████▊    | 109/187 [10:58<07:57,  6.12s/it]
Training Epoch: 13/100, batch 217/375 completed (loss: 0.0015:  58%|█████▊    | 109/187 [10:58<07:57,  6.12s/it]
Training Epoch: 13/100, batch 217/375 completed (loss: 0.019:  58%|█████▊    | 109/187 [10:58<07:57,  6.12s/it] 
Training Epoch: 13/100, batch 217/375 completed (loss: 0.0037:  58%|█████▊    | 109/187 [10:58<07:57,  6.12s/it]
Training Epoch: 13/100, batch 217/375 completed (loss: 0.0057:  58%|█████▊    | 109/187 [10:58<07:56,  6.11s/it]
Training Epoch: 13/100, batch 218/375 completed (loss: 0.0226:  58%|█████▊    | 109/187 [11:01<07:57,  6.12s/it]
Training Epoch: 13/100, batch 218/375 completed (loss: 0.1297:  58%|█████▊    | 109/187 [11:01<07:57,  6.12s/it]
Training Epoch: 13/100, batch 218/375 completed (loss: 0.0308:  58%|█████▊    | 109/187 [11:01<07:57,  6.12s/it]
Training Epoch: 13/100, batch 218/375 completed (loss: 0.0044:  58%|█████▊    | 109/187 [11:01<07:56,  6.11s/it]
Training Epoch: 13/100, batch 218/375 completed (loss: 0.0308:  59%|█████▉    | 110/187 [11:04<07:48,  6.08s/it]
Training Epoch: 13/100, batch 218/375 completed (loss: 0.0044:  59%|█████▉    | 110/187 [11:04<07:47,  6.07s/it]
Training Epoch: 13/100, batch 218/375 completed (loss: 0.0226:  59%|█████▉    | 110/187 [11:04<07:48,  6.08s/it]
Training Epoch: 13/100, batch 218/375 completed (loss: 0.1297:  59%|█████▉    | 110/187 [11:04<07:48,  6.08s/it]
Training Epoch: 13/100, batch 219/375 completed (loss: 0.03:  59%|█████▉    | 110/187 [11:04<07:48,  6.08s/it]  
Training Epoch: 13/100, batch 219/375 completed (loss: 0.012:  59%|█████▉    | 110/187 [11:04<07:48,  6.08s/it] 
Training Epoch: 13/100, batch 219/375 completed (loss: 0.0319:  59%|█████▉    | 110/187 [11:04<07:48,  6.08s/it]
Training Epoch: 13/100, batch 219/375 completed (loss: 0.1168:  59%|█████▉    | 110/187 [11:04<07:47,  6.07s/it]
Training Epoch: 13/100, batch 220/375 completed (loss: 0.0148:  59%|█████▉    | 110/187 [11:07<07:48,  6.08s/it]
Training Epoch: 13/100, batch 220/375 completed (loss: 0.0207:  59%|█████▉    | 110/187 [11:07<07:48,  6.08s/it]
Training Epoch: 13/100, batch 220/375 completed (loss: 0.0204:  59%|█████▉    | 110/187 [11:07<07:48,  6.08s/it]
Training Epoch: 13/100, batch 220/375 completed (loss: 0.0088:  59%|█████▉    | 110/187 [11:07<07:47,  6.07s/it]
Training Epoch: 13/100, batch 220/375 completed (loss: 0.0207:  59%|█████▉    | 111/187 [11:10<07:46,  6.14s/it]
Training Epoch: 13/100, batch 220/375 completed (loss: 0.0148:  59%|█████▉    | 111/187 [11:10<07:46,  6.14s/it]
Training Epoch: 13/100, batch 220/375 completed (loss: 0.0204:  59%|█████▉    | 111/187 [11:10<07:46,  6.14s/it]
Training Epoch: 13/100, batch 220/375 completed (loss: 0.0088:  59%|█████▉    | 111/187 [11:10<07:46,  6.14s/it]
Training Epoch: 13/100, batch 221/375 completed (loss: 0.0031:  59%|█████▉    | 111/187 [11:10<07:46,  6.14s/it]
Training Epoch: 13/100, batch 221/375 completed (loss: 0.0047:  59%|█████▉    | 111/187 [11:10<07:46,  6.14s/it]
Training Epoch: 13/100, batch 221/375 completed (loss: 0.0019:  59%|█████▉    | 111/187 [11:10<07:46,  6.14s/it]
Training Epoch: 13/100, batch 221/375 completed (loss: 0.0356:  59%|█████▉    | 111/187 [11:10<07:46,  6.14s/it]
Training Epoch: 13/100, batch 222/375 completed (loss: 0.032:  59%|█████▉    | 111/187 [11:13<07:46,  6.14s/it] 
Training Epoch: 13/100, batch 222/375 completed (loss: 0.0133:  59%|█████▉    | 111/187 [11:13<07:46,  6.14s/it]
Training Epoch: 13/100, batch 222/375 completed (loss: 0.0131:  59%|█████▉    | 111/187 [11:13<07:46,  6.14s/it]
Training Epoch: 13/100, batch 222/375 completed (loss: 0.0036:  59%|█████▉    | 111/187 [11:13<07:46,  6.14s/it]
Training Epoch: 13/100, batch 222/375 completed (loss: 0.0133:  60%|█████▉    | 112/187 [11:16<07:34,  6.06s/it]
Training Epoch: 13/100, batch 222/375 completed (loss: 0.0036:  60%|█████▉    | 112/187 [11:16<07:34,  6.05s/it]
Training Epoch: 13/100, batch 222/375 completed (loss: 0.032:  60%|█████▉    | 112/187 [11:16<07:34,  6.06s/it]
Training Epoch: 13/100, batch 222/375 completed (loss: 0.0131:  60%|█████▉    | 112/187 [11:16<07:34,  6.06s/it]
Training Epoch: 13/100, batch 223/375 completed (loss: 0.0138:  60%|█████▉    | 112/187 [11:16<07:34,  6.06s/it]
Training Epoch: 13/100, batch 223/375 completed (loss: 0.0209:  60%|█████▉    | 112/187 [11:16<07:34,  6.06s/it]
Training Epoch: 13/100, batch 223/375 completed (loss: 0.0152:  60%|█████▉    | 112/187 [11:16<07:34,  6.06s/it]
Training Epoch: 13/100, batch 223/375 completed (loss: 0.2295:  60%|█████▉    | 112/187 [11:16<07:34,  6.05s/it]
Training Epoch: 13/100, batch 224/375 completed (loss: 0.4591:  60%|█████▉    | 112/187 [11:19<07:34,  6.06s/it]
Training Epoch: 13/100, batch 224/375 completed (loss: 0.2967:  60%|█████▉    | 112/187 [11:19<07:34,  6.06s/it]
Training Epoch: 13/100, batch 224/375 completed (loss: 0.0503:  60%|█████▉    | 112/187 [11:19<07:34,  6.06s/it]
Training Epoch: 13/100, batch 224/375 completed (loss: 0.0015:  60%|█████▉    | 112/187 [11:19<07:34,  6.05s/it]
Training Epoch: 13/100, batch 224/375 completed (loss: 0.0015:  60%|██████    | 113/187 [11:22<07:27,  6.05s/it]
Training Epoch: 13/100, batch 224/375 completed (loss: 0.4591:  60%|██████    | 113/187 [11:22<07:27,  6.05s/it]
Training Epoch: 13/100, batch 224/375 completed (loss: 0.2967:  60%|██████    | 113/187 [11:22<07:27,  6.05s/it]
Training Epoch: 13/100, batch 224/375 completed (loss: 0.0503:  60%|██████    | 113/187 [11:22<07:27,  6.05s/it]
Training Epoch: 13/100, batch 225/375 completed (loss: 0.026:  60%|██████    | 113/187 [11:22<07:27,  6.05s/it] 
Training Epoch: 13/100, batch 225/375 completed (loss: 0.0201:  60%|██████    | 113/187 [11:22<07:27,  6.05s/it]
Training Epoch: 13/100, batch 225/375 completed (loss: 0.0037:  60%|██████    | 113/187 [11:22<07:27,  6.05s/it]
Training Epoch: 13/100, batch 225/375 completed (loss: 0.0014:  60%|██████    | 113/187 [11:22<07:27,  6.05s/it]
Training Epoch: 13/100, batch 226/375 completed (loss: 0.0812:  60%|██████    | 113/187 [11:25<07:27,  6.05s/it]
Training Epoch: 13/100, batch 226/375 completed (loss: 0.0066:  60%|██████    | 113/187 [11:25<07:27,  6.05s/it]
Training Epoch: 13/100, batch 226/375 completed (loss: 0.0045:  60%|██████    | 113/187 [11:25<07:27,  6.05s/it]
Training Epoch: 13/100, batch 226/375 completed (loss: 0.0135:  60%|██████    | 113/187 [11:26<07:27,  6.05s/it]
Training Epoch: 13/100, batch 226/375 completed (loss: 0.0135:  61%|██████    | 114/187 [11:28<07:25,  6.11s/it]
Training Epoch: 13/100, batch 226/375 completed (loss: 0.0812:  61%|██████    | 114/187 [11:28<07:25,  6.11s/it]
Training Epoch: 13/100, batch 226/375 completed (loss: 0.0045:  61%|██████    | 114/187 [11:28<07:25,  6.11s/it]
Training Epoch: 13/100, batch 226/375 completed (loss: 0.0066:  61%|██████    | 114/187 [11:28<07:26,  6.11s/it]
Training Epoch: 13/100, batch 227/375 completed (loss: 0.0105:  61%|██████    | 114/187 [11:28<07:26,  6.11s/it]
Training Epoch: 13/100, batch 227/375 completed (loss: 0.0193:  61%|██████    | 114/187 [11:28<07:25,  6.11s/it]
Training Epoch: 13/100, batch 227/375 completed (loss: 0.0232:  61%|██████    | 114/187 [11:28<07:25,  6.11s/it]
Training Epoch: 13/100, batch 227/375 completed (loss: 0.0054:  61%|██████    | 114/187 [11:28<07:25,  6.11s/it]
Training Epoch: 13/100, batch 228/375 completed (loss: 0.0499:  61%|██████    | 114/187 [11:31<07:26,  6.11s/it]
Training Epoch: 13/100, batch 228/375 completed (loss: 0.1383:  61%|██████    | 114/187 [11:31<07:25,  6.11s/it]
Training Epoch: 13/100, batch 228/375 completed (loss: 0.0189:  61%|██████    | 114/187 [11:31<07:25,  6.11s/it]
Training Epoch: 13/100, batch 228/375 completed (loss: 0.0093:  61%|██████    | 114/187 [11:31<07:25,  6.11s/it]
Training Epoch: 13/100, batch 228/375 completed (loss: 0.0189:  61%|██████▏   | 115/187 [11:34<07:17,  6.08s/it]
Training Epoch: 13/100, batch 228/375 completed (loss: 0.1383:  61%|██████▏   | 115/187 [11:34<07:17,  6.08s/it]
Training Epoch: 13/100, batch 228/375 completed (loss: 0.0093:  61%|██████▏   | 115/187 [11:34<07:17,  6.08s/it]
Training Epoch: 13/100, batch 228/375 completed (loss: 0.0499:  61%|██████▏   | 115/187 [11:34<07:17,  6.08s/it]
Training Epoch: 13/100, batch 229/375 completed (loss: 0.0059:  61%|██████▏   | 115/187 [11:34<07:17,  6.08s/it]
Training Epoch: 13/100, batch 229/375 completed (loss: 0.0005:  61%|██████▏   | 115/187 [11:34<07:17,  6.08s/it]
Training Epoch: 13/100, batch 229/375 completed (loss: 0.0066:  61%|██████▏   | 115/187 [11:34<07:17,  6.08s/it]
Training Epoch: 13/100, batch 229/375 completed (loss: 0.0253:  61%|██████▏   | 115/187 [11:34<07:17,  6.08s/it]
Training Epoch: 13/100, batch 230/375 completed (loss: 0.012:  61%|██████▏   | 115/187 [11:37<07:17,  6.08s/it] 
Training Epoch: 13/100, batch 230/375 completed (loss: 0.0081:  61%|██████▏   | 115/187 [11:37<07:17,  6.08s/it]
Training Epoch: 13/100, batch 230/375 completed (loss: 0.2113:  61%|██████▏   | 115/187 [11:37<07:17,  6.08s/it]
Training Epoch: 13/100, batch 230/375 completed (loss: 0.0374:  61%|██████▏   | 115/187 [11:37<07:17,  6.08s/it]
Training Epoch: 13/100, batch 230/375 completed (loss: 0.012:  62%|██████▏   | 116/187 [11:40<07:13,  6.10s/it]
Training Epoch: 13/100, batch 230/375 completed (loss: 0.2113:  62%|██████▏   | 116/187 [11:40<07:13,  6.10s/it]
Training Epoch: 13/100, batch 230/375 completed (loss: 0.0081:  62%|██████▏   | 116/187 [11:40<07:13,  6.10s/it]
Training Epoch: 13/100, batch 230/375 completed (loss: 0.0374:  62%|██████▏   | 116/187 [11:41<07:13,  6.10s/it]
Training Epoch: 13/100, batch 231/375 completed (loss: 0.002:  62%|██████▏   | 116/187 [11:41<07:13,  6.10s/it]
Training Epoch: 13/100, batch 231/375 completed (loss: 0.0263:  62%|██████▏   | 116/187 [11:41<07:13,  6.10s/it]
Training Epoch: 13/100, batch 231/375 completed (loss: 0.0763:  62%|██████▏   | 116/187 [11:41<07:13,  6.10s/it]
Training Epoch: 13/100, batch 231/375 completed (loss: 0.0742:  62%|██████▏   | 116/187 [11:41<07:13,  6.10s/it]
Training Epoch: 13/100, batch 232/375 completed (loss: 0.0074:  62%|██████▏   | 116/187 [11:44<07:13,  6.10s/it]
Training Epoch: 13/100, batch 232/375 completed (loss: 0.0261:  62%|██████▏   | 116/187 [11:44<07:13,  6.10s/it]
Training Epoch: 13/100, batch 232/375 completed (loss: 0.2085:  62%|██████▏   | 116/187 [11:44<07:13,  6.10s/it]
Training Epoch: 13/100, batch 232/375 completed (loss: 0.0104:  62%|██████▏   | 116/187 [11:44<07:13,  6.10s/it]
Training Epoch: 13/100, batch 232/375 completed (loss: 0.0261:  63%|██████▎   | 117/187 [11:46<07:04,  6.06s/it]
Training Epoch: 13/100, batch 232/375 completed (loss: 0.2085:  63%|██████▎   | 117/187 [11:46<07:04,  6.06s/it]
Training Epoch: 13/100, batch 232/375 completed (loss: 0.0104:  63%|██████▎   | 117/187 [11:47<07:04,  6.06s/it]
Training Epoch: 13/100, batch 232/375 completed (loss: 0.0074:  63%|██████▎   | 117/187 [11:46<07:04,  6.06s/it]
Training Epoch: 13/100, batch 233/375 completed (loss: 0.0014:  63%|██████▎   | 117/187 [11:47<07:04,  6.06s/it]
Training Epoch: 13/100, batch 233/375 completed (loss: 0.005:  63%|██████▎   | 117/187 [11:47<07:04,  6.06s/it] 
Training Epoch: 13/100, batch 233/375 completed (loss: 0.0081:  63%|██████▎   | 117/187 [11:47<07:04,  6.06s/it]
Training Epoch: 13/100, batch 233/375 completed (loss: 0.0529:  63%|██████▎   | 117/187 [11:47<07:04,  6.06s/it]
Training Epoch: 13/100, batch 234/375 completed (loss: 0.0337:  63%|██████▎   | 117/187 [11:49<07:04,  6.06s/it]
Training Epoch: 13/100, batch 234/375 completed (loss: 0.069:  63%|██████▎   | 117/187 [11:49<07:04,  6.06s/it]
Training Epoch: 13/100, batch 234/375 completed (loss: 0.0197:  63%|██████▎   | 117/187 [11:49<07:04,  6.06s/it]
Training Epoch: 13/100, batch 234/375 completed (loss: 0.217:  63%|██████▎   | 117/187 [11:49<07:04,  6.06s/it] 
Training Epoch: 13/100, batch 234/375 completed (loss: 0.0197:  63%|██████▎   | 118/187 [11:52<06:56,  6.04s/it]
Training Epoch: 13/100, batch 234/375 completed (loss: 0.217:  63%|██████▎   | 118/187 [11:52<06:56,  6.04s/it]
Training Epoch: 13/100, batch 234/375 completed (loss: 0.0337:  63%|██████▎   | 118/187 [11:52<06:56,  6.04s/it]
Training Epoch: 13/100, batch 234/375 completed (loss: 0.069:  63%|██████▎   | 118/187 [11:52<06:56,  6.04s/it]
Training Epoch: 13/100, batch 235/375 completed (loss: 0.0027:  63%|██████▎   | 118/187 [11:53<06:56,  6.04s/it]
Training Epoch: 13/100, batch 235/375 completed (loss: 0.0234:  63%|██████▎   | 118/187 [11:52<06:56,  6.04s/it]
Training Epoch: 13/100, batch 235/375 completed (loss: 0.0067:  63%|██████▎   | 118/187 [11:52<06:56,  6.04s/it]
Training Epoch: 13/100, batch 235/375 completed (loss: 0.016:  63%|██████▎   | 118/187 [11:53<06:56,  6.04s/it]
Training Epoch: 13/100, batch 236/375 completed (loss: 0.039:  63%|██████▎   | 118/187 [11:56<06:56,  6.04s/it] 
Training Epoch: 13/100, batch 236/375 completed (loss: 0.0088:  63%|██████▎   | 118/187 [11:56<06:56,  6.04s/it]
Training Epoch: 13/100, batch 236/375 completed (loss: 0.0829:  63%|██████▎   | 118/187 [11:56<06:56,  6.04s/it]
Training Epoch: 13/100, batch 236/375 completed (loss: 0.0067:  63%|██████▎   | 118/187 [11:56<06:56,  6.04s/it]
Training Epoch: 13/100, batch 236/375 completed (loss: 0.039:  64%|██████▎   | 119/187 [11:59<06:55,  6.11s/it]
Training Epoch: 13/100, batch 236/375 completed (loss: 0.0088:  64%|██████▎   | 119/187 [11:59<06:55,  6.11s/it]
Training Epoch: 13/100, batch 236/375 completed (loss: 0.0829:  64%|██████▎   | 119/187 [11:59<06:55,  6.11s/it]
Training Epoch: 13/100, batch 236/375 completed (loss: 0.0067:  64%|██████▎   | 119/187 [11:59<06:55,  6.11s/it]
Training Epoch: 13/100, batch 237/375 completed (loss: 0.0119:  64%|██████▎   | 119/187 [11:59<06:55,  6.11s/it]
Training Epoch: 13/100, batch 237/375 completed (loss: 0.0092:  64%|██████▎   | 119/187 [11:59<06:55,  6.11s/it]
Training Epoch: 13/100, batch 237/375 completed (loss: 0.0414:  64%|██████▎   | 119/187 [11:59<06:55,  6.11s/it]
Training Epoch: 13/100, batch 237/375 completed (loss: 0.002:  64%|██████▎   | 119/187 [11:59<06:55,  6.11s/it] 
Training Epoch: 13/100, batch 238/375 completed (loss: 0.0083:  64%|██████▎   | 119/187 [12:02<06:55,  6.11s/it]
Training Epoch: 13/100, batch 238/375 completed (loss: 0.0535:  64%|██████▎   | 119/187 [12:02<06:55,  6.11s/it]
Training Epoch: 13/100, batch 238/375 completed (loss: 0.0281:  64%|██████▎   | 119/187 [12:02<06:55,  6.11s/it]
Training Epoch: 13/100, batch 238/375 completed (loss: 0.0103:  64%|██████▎   | 119/187 [12:02<06:55,  6.11s/it]
Training Epoch: 13/100, batch 238/375 completed (loss: 0.0281:  64%|██████▍   | 120/187 [12:05<06:43,  6.02s/it]
Training Epoch: 13/100, batch 238/375 completed (loss: 0.0535:  64%|██████▍   | 120/187 [12:05<06:43,  6.02s/it]
Training Epoch: 13/100, batch 238/375 completed (loss: 0.0083:  64%|██████▍   | 120/187 [12:05<06:43,  6.03s/it]
Training Epoch: 13/100, batch 238/375 completed (loss: 0.0103:  64%|██████▍   | 120/187 [12:05<06:43,  6.03s/it]
Training Epoch: 13/100, batch 239/375 completed (loss: 0.0087:  64%|██████▍   | 120/187 [12:05<06:43,  6.02s/it]
Training Epoch: 13/100, batch 239/375 completed (loss: 0.0108:  64%|██████▍   | 120/187 [12:05<06:43,  6.03s/it]
Training Epoch: 13/100, batch 239/375 completed (loss: 0.0049:  64%|██████▍   | 120/187 [12:05<06:43,  6.02s/it]
Training Epoch: 13/100, batch 239/375 completed (loss: 0.0094:  64%|██████▍   | 120/187 [12:05<06:43,  6.03s/it]
Training Epoch: 13/100, batch 240/375 completed (loss: 0.0729:  64%|██████▍   | 120/187 [12:08<06:43,  6.02s/it]
Training Epoch: 13/100, batch 240/375 completed (loss: 0.0147:  64%|██████▍   | 120/187 [12:08<06:43,  6.03s/it]
Training Epoch: 13/100, batch 240/375 completed (loss: 0.0113:  64%|██████▍   | 120/187 [12:08<06:43,  6.02s/it]
Training Epoch: 13/100, batch 240/375 completed (loss: 0.0068:  64%|██████▍   | 120/187 [12:08<06:43,  6.03s/it]
Training Epoch: 13/100, batch 240/375 completed (loss: 0.0729:  65%|██████▍   | 121/187 [12:11<06:40,  6.07s/it]
Training Epoch: 13/100, batch 240/375 completed (loss: 0.0113:  65%|██████▍   | 121/187 [12:11<06:40,  6.07s/it]
Training Epoch: 13/100, batch 240/375 completed (loss: 0.0068:  65%|██████▍   | 121/187 [12:11<06:40,  6.07s/it]
Training Epoch: 13/100, batch 240/375 completed (loss: 0.0147:  65%|██████▍   | 121/187 [12:11<06:40,  6.07s/it]
Training Epoch: 13/100, batch 241/375 completed (loss: 0.1116:  65%|██████▍   | 121/187 [12:11<06:40,  6.07s/it]
Training Epoch: 13/100, batch 241/375 completed (loss: 0.0127:  65%|██████▍   | 121/187 [12:11<06:40,  6.07s/it]
Training Epoch: 13/100, batch 241/375 completed (loss: 0.0237:  65%|██████▍   | 121/187 [12:11<06:40,  6.07s/it]
Training Epoch: 13/100, batch 241/375 completed (loss: 0.1288:  65%|██████▍   | 121/187 [12:11<06:40,  6.07s/it]
Training Epoch: 13/100, batch 242/375 completed (loss: 0.1098:  65%|██████▍   | 121/187 [12:14<06:40,  6.07s/it]
Training Epoch: 13/100, batch 242/375 completed (loss: 0.0035:  65%|██████▍   | 121/187 [12:14<06:40,  6.07s/it]
Training Epoch: 13/100, batch 242/375 completed (loss: 0.0126:  65%|██████▍   | 121/187 [12:14<06:40,  6.07s/it]
Training Epoch: 13/100, batch 242/375 completed (loss: 0.0112:  65%|██████▍   | 121/187 [12:14<06:40,  6.07s/it]
Training Epoch: 13/100, batch 242/375 completed (loss: 0.1098:  65%|██████▌   | 122/187 [12:17<06:33,  6.05s/it]
Training Epoch: 13/100, batch 242/375 completed (loss: 0.0035:  65%|██████▌   | 122/187 [12:17<06:33,  6.05s/it]
Training Epoch: 13/100, batch 242/375 completed (loss: 0.0126:  65%|██████▌   | 122/187 [12:17<06:32,  6.05s/it]
Training Epoch: 13/100, batch 242/375 completed (loss: 0.0112:  65%|██████▌   | 122/187 [12:17<06:33,  6.05s/it]
Training Epoch: 13/100, batch 243/375 completed (loss: 0.11:  65%|██████▌   | 122/187 [12:17<06:32,  6.05s/it]  
Training Epoch: 13/100, batch 243/375 completed (loss: 0.0873:  65%|██████▌   | 122/187 [12:17<06:33,  6.05s/it]
Training Epoch: 13/100, batch 243/375 completed (loss: 0.0028:  65%|██████▌   | 122/187 [12:17<06:33,  6.05s/it]
Training Epoch: 13/100, batch 243/375 completed (loss: 0.0367:  65%|██████▌   | 122/187 [12:17<06:33,  6.05s/it]
Training Epoch: 13/100, batch 244/375 completed (loss: 0.1288:  65%|██████▌   | 122/187 [12:20<06:33,  6.05s/it]
Training Epoch: 13/100, batch 244/375 completed (loss: 0.0538:  65%|██████▌   | 122/187 [12:20<06:32,  6.05s/it]
Training Epoch: 13/100, batch 244/375 completed (loss: 0.0056:  65%|██████▌   | 122/187 [12:20<06:33,  6.05s/it]
Training Epoch: 13/100, batch 244/375 completed (loss: 0.0304:  65%|██████▌   | 122/187 [12:20<06:33,  6.05s/it]
Training Epoch: 13/100, batch 244/375 completed (loss: 0.1288:  66%|██████▌   | 123/187 [12:23<06:25,  6.03s/it]
Training Epoch: 13/100, batch 244/375 completed (loss: 0.0056:  66%|██████▌   | 123/187 [12:23<06:26,  6.03s/it]
Training Epoch: 13/100, batch 244/375 completed (loss: 0.0538:  66%|██████▌   | 123/187 [12:23<06:25,  6.03s/it]
Training Epoch: 13/100, batch 244/375 completed (loss: 0.0304:  66%|██████▌   | 123/187 [12:23<06:26,  6.03s/it]
Training Epoch: 13/100, batch 245/375 completed (loss: 0.0312:  66%|██████▌   | 123/187 [12:23<06:25,  6.03s/it]
Training Epoch: 13/100, batch 245/375 completed (loss: 0.0283:  66%|██████▌   | 123/187 [12:23<06:25,  6.03s/it]
Training Epoch: 13/100, batch 245/375 completed (loss: 0.0477:  66%|██████▌   | 123/187 [12:23<06:26,  6.03s/it]
Training Epoch: 13/100, batch 245/375 completed (loss: 0.0719:  66%|██████▌   | 123/187 [12:23<06:26,  6.03s/it]
Training Epoch: 13/100, batch 246/375 completed (loss: 0.0275:  66%|██████▌   | 123/187 [12:26<06:25,  6.03s/it]
Training Epoch: 13/100, batch 246/375 completed (loss: 0.0098:  66%|██████▌   | 123/187 [12:26<06:25,  6.03s/it]
Training Epoch: 13/100, batch 246/375 completed (loss: 0.0012:  66%|██████▌   | 123/187 [12:26<06:26,  6.03s/it]
Training Epoch: 13/100, batch 246/375 completed (loss: 0.0387:  66%|██████▌   | 123/187 [12:26<06:26,  6.03s/it]
Training Epoch: 13/100, batch 246/375 completed (loss: 0.0275:  66%|██████▋   | 124/187 [12:29<06:24,  6.11s/it]
Training Epoch: 13/100, batch 246/375 completed (loss: 0.0098:  66%|██████▋   | 124/187 [12:29<06:24,  6.11s/it]
Training Epoch: 13/100, batch 246/375 completed (loss: 0.0387:  66%|██████▋   | 124/187 [12:29<06:24,  6.11s/it]
Training Epoch: 13/100, batch 246/375 completed (loss: 0.0012:  66%|██████▋   | 124/187 [12:29<06:25,  6.11s/it]
Training Epoch: 13/100, batch 247/375 completed (loss: 0.0048:  66%|██████▋   | 124/187 [12:29<06:25,  6.11s/it]
Training Epoch: 13/100, batch 247/375 completed (loss: 0.0045:  66%|██████▋   | 124/187 [12:29<06:24,  6.11s/it]
Training Epoch: 13/100, batch 247/375 completed (loss: 0.0066:  66%|██████▋   | 124/187 [12:29<06:24,  6.11s/it]
Training Epoch: 13/100, batch 247/375 completed (loss: 0.0352:  66%|██████▋   | 124/187 [12:29<06:24,  6.11s/it]
Training Epoch: 13/100, batch 248/375 completed (loss: 0.0044:  66%|██████▋   | 124/187 [12:32<06:24,  6.11s/it]
Training Epoch: 13/100, batch 248/375 completed (loss: 0.034:  66%|██████▋   | 124/187 [12:32<06:24,  6.11s/it] 
Training Epoch: 13/100, batch 248/375 completed (loss: 0.0522:  66%|██████▋   | 124/187 [12:32<06:25,  6.11s/it]
Training Epoch: 13/100, batch 248/375 completed (loss: 0.0155:  66%|██████▋   | 124/187 [12:32<06:24,  6.11s/it]
Training Epoch: 13/100, batch 248/375 completed (loss: 0.0044:  67%|██████▋   | 125/187 [12:35<06:14,  6.03s/it]
Training Epoch: 13/100, batch 248/375 completed (loss: 0.034:  67%|██████▋   | 125/187 [12:35<06:14,  6.03s/it]
Training Epoch: 13/100, batch 248/375 completed (loss: 0.0155:  67%|██████▋   | 125/187 [12:35<06:14,  6.03s/it]
Training Epoch: 13/100, batch 248/375 completed (loss: 0.0522:  67%|██████▋   | 125/187 [12:35<06:14,  6.04s/it]
Training Epoch: 13/100, batch 249/375 completed (loss: 0.0156:  67%|██████▋   | 125/187 [12:35<06:14,  6.04s/it]
Training Epoch: 13/100, batch 249/375 completed (loss: 0.012:  67%|██████▋   | 125/187 [12:35<06:14,  6.03s/it]
Training Epoch: 13/100, batch 249/375 completed (loss: 0.0115:  67%|██████▋   | 125/187 [12:35<06:14,  6.03s/it]
Training Epoch: 13/100, batch 249/375 completed (loss: 0.128:  67%|██████▋   | 125/187 [12:35<06:14,  6.03s/it] 
Training Epoch: 13/100, batch 250/375 completed (loss: 0.0196:  67%|██████▋   | 125/187 [12:38<06:14,  6.03s/it]
Training Epoch: 13/100, batch 250/375 completed (loss: 0.0143:  67%|██████▋   | 125/187 [12:38<06:14,  6.04s/it]
Training Epoch: 13/100, batch 250/375 completed (loss: 0.0036:  67%|██████▋   | 125/187 [12:38<06:14,  6.03s/it]
Training Epoch: 13/100, batch 250/375 completed (loss: 0.0015:  67%|██████▋   | 125/187 [12:38<06:14,  6.03s/it]
Training Epoch: 13/100, batch 250/375 completed (loss: 0.0036:  67%|██████▋   | 126/187 [12:41<06:11,  6.08s/it]
Training Epoch: 13/100, batch 250/375 completed (loss: 0.0196:  67%|██████▋   | 126/187 [12:41<06:11,  6.08s/it]
Training Epoch: 13/100, batch 250/375 completed (loss: 0.0143:  67%|██████▋   | 126/187 [12:41<06:11,  6.09s/it]
Training Epoch: 13/100, batch 250/375 completed (loss: 0.0015:  67%|██████▋   | 126/187 [12:41<06:11,  6.08s/it]
Training Epoch: 13/100, batch 251/375 completed (loss: 0.0088:  67%|██████▋   | 126/187 [12:41<06:11,  6.09s/it]
Training Epoch: 13/100, batch 251/375 completed (loss: 0.0215:  67%|██████▋   | 126/187 [12:41<06:11,  6.08s/it]
Training Epoch: 13/100, batch 251/375 completed (loss: 0.0784:  67%|██████▋   | 126/187 [12:41<06:11,  6.08s/it]
Training Epoch: 13/100, batch 251/375 completed (loss: 0.0484:  67%|██████▋   | 126/187 [12:41<06:11,  6.08s/it]
Training Epoch: 13/100, batch 252/375 completed (loss: 0.0079:  67%|██████▋   | 126/187 [12:44<06:11,  6.08s/it]
Training Epoch: 13/100, batch 252/375 completed (loss: 0.1053:  67%|██████▋   | 126/187 [12:44<06:11,  6.09s/it]
Training Epoch: 13/100, batch 252/375 completed (loss: 0.0046:  67%|██████▋   | 126/187 [12:44<06:11,  6.08s/it]
Training Epoch: 13/100, batch 252/375 completed (loss: 0.0284:  67%|██████▋   | 126/187 [12:44<06:11,  6.08s/it]
Training Epoch: 13/100, batch 252/375 completed (loss: 0.1053:  68%|██████▊   | 127/187 [12:47<06:04,  6.07s/it]
Training Epoch: 13/100, batch 252/375 completed (loss: 0.0079:  68%|██████▊   | 127/187 [12:47<06:04,  6.07s/it]
Training Epoch: 13/100, batch 252/375 completed (loss: 0.0284:  68%|██████▊   | 127/187 [12:47<06:04,  6.07s/it]
Training Epoch: 13/100, batch 252/375 completed (loss: 0.0046:  68%|██████▊   | 127/187 [12:47<06:04,  6.08s/it]
Training Epoch: 13/100, batch 253/375 completed (loss: 0.0231:  68%|██████▊   | 127/187 [12:47<06:04,  6.07s/it]
Training Epoch: 13/100, batch 253/375 completed (loss: 0.0104:  68%|██████▊   | 127/187 [12:47<06:04,  6.08s/it]
Training Epoch: 13/100, batch 253/375 completed (loss: 0.0182:  68%|██████▊   | 127/187 [12:47<06:04,  6.07s/it]
Training Epoch: 13/100, batch 253/375 completed (loss: 0.0076:  68%|██████▊   | 127/187 [12:47<06:04,  6.07s/it]
Training Epoch: 13/100, batch 254/375 completed (loss: 0.0094:  68%|██████▊   | 127/187 [12:50<06:04,  6.07s/it]
Training Epoch: 13/100, batch 254/375 completed (loss: 0.0313:  68%|██████▊   | 127/187 [12:50<06:04,  6.07s/it]
Training Epoch: 13/100, batch 254/375 completed (loss: 0.0094:  68%|██████▊   | 127/187 [12:50<06:04,  6.08s/it]
Training Epoch: 13/100, batch 254/375 completed (loss: 0.0068:  68%|██████▊   | 127/187 [12:50<06:04,  6.07s/it]
Training Epoch: 13/100, batch 254/375 completed (loss: 0.0094:  68%|██████▊   | 128/187 [12:53<05:54,  6.01s/it]
Training Epoch: 13/100, batch 254/375 completed (loss: 0.0068:  68%|██████▊   | 128/187 [12:53<05:54,  6.01s/it]
Training Epoch: 13/100, batch 254/375 completed (loss: 0.0313:  68%|██████▊   | 128/187 [12:53<05:54,  6.01s/it]
Training Epoch: 13/100, batch 254/375 completed (loss: 0.0094:  68%|██████▊   | 128/187 [12:53<05:54,  6.01s/it]
Training Epoch: 13/100, batch 255/375 completed (loss: 0.0194:  68%|██████▊   | 128/187 [12:53<05:54,  6.01s/it]
Training Epoch: 13/100, batch 255/375 completed (loss: 0.0095:  68%|██████▊   | 128/187 [12:53<05:54,  6.01s/it]
Training Epoch: 13/100, batch 255/375 completed (loss: 0.0692:  68%|██████▊   | 128/187 [12:53<05:54,  6.01s/it]
Training Epoch: 13/100, batch 255/375 completed (loss: 0.0762:  68%|██████▊   | 128/187 [12:53<05:54,  6.01s/it]
Training Epoch: 13/100, batch 256/375 completed (loss: 0.0088:  68%|██████▊   | 128/187 [12:56<05:54,  6.01s/it]
Training Epoch: 13/100, batch 256/375 completed (loss: 0.0077:  68%|██████▊   | 128/187 [12:56<05:54,  6.01s/it]
Training Epoch: 13/100, batch 256/375 completed (loss: 0.009:  68%|██████▊   | 128/187 [12:56<05:54,  6.01s/it] 
Training Epoch: 13/100, batch 256/375 completed (loss: 0.0055:  68%|██████▊   | 128/187 [12:56<05:54,  6.01s/it]
Training Epoch: 13/100, batch 256/375 completed (loss: 0.0055:  69%|██████▉   | 129/187 [12:59<05:54,  6.11s/it]
Training Epoch: 13/100, batch 256/375 completed (loss: 0.0077:  69%|██████▉   | 129/187 [12:59<05:54,  6.11s/it]
Training Epoch: 13/100, batch 256/375 completed (loss: 0.0088:  69%|██████▉   | 129/187 [12:59<05:54,  6.11s/it]
Training Epoch: 13/100, batch 256/375 completed (loss: 0.009:  69%|██████▉   | 129/187 [12:59<05:54,  6.11s/it]
Training Epoch: 13/100, batch 257/375 completed (loss: 0.0083:  69%|██████▉   | 129/187 [12:59<05:54,  6.11s/it]
Training Epoch: 13/100, batch 257/375 completed (loss: 0.0011:  69%|██████▉   | 129/187 [12:59<05:54,  6.11s/it]
Training Epoch: 13/100, batch 257/375 completed (loss: 0.0395:  69%|██████▉   | 129/187 [12:59<05:54,  6.11s/it]
Training Epoch: 13/100, batch 257/375 completed (loss: 0.0003:  69%|██████▉   | 129/187 [12:59<05:54,  6.11s/it]
Training Epoch: 13/100, batch 258/375 completed (loss: 0.0011:  69%|██████▉   | 129/187 [13:02<05:54,  6.11s/it]
Training Epoch: 13/100, batch 258/375 completed (loss: 0.0247:  69%|██████▉   | 129/187 [13:02<05:54,  6.11s/it]
Training Epoch: 13/100, batch 258/375 completed (loss: 0.0171:  69%|██████▉   | 129/187 [13:02<05:54,  6.11s/it]
Training Epoch: 13/100, batch 258/375 completed (loss: 0.0119:  69%|██████▉   | 129/187 [13:02<05:54,  6.11s/it]
Training Epoch: 13/100, batch 258/375 completed (loss: 0.0247:  70%|██████▉   | 130/187 [13:05<05:44,  6.04s/it]
Training Epoch: 13/100, batch 258/375 completed (loss: 0.0011:  70%|██████▉   | 130/187 [13:05<05:44,  6.04s/it]
Training Epoch: 13/100, batch 258/375 completed (loss: 0.0171:  70%|██████▉   | 130/187 [13:05<05:44,  6.04s/it]
Training Epoch: 13/100, batch 258/375 completed (loss: 0.0119:  70%|██████▉   | 130/187 [13:05<05:44,  6.04s/it]
Training Epoch: 13/100, batch 259/375 completed (loss: 0.06:  70%|██████▉   | 130/187 [13:05<05:44,  6.04s/it]  
Training Epoch: 13/100, batch 259/375 completed (loss: 0.0035:  70%|██████▉   | 130/187 [13:05<05:44,  6.04s/it]
Training Epoch: 13/100, batch 259/375 completed (loss: 0.0683:  70%|██████▉   | 130/187 [13:05<05:44,  6.04s/it]
Training Epoch: 13/100, batch 259/375 completed (loss: 0.2942:  70%|██████▉   | 130/187 [13:05<05:44,  6.04s/it]
Training Epoch: 13/100, batch 260/375 completed (loss: 0.017:  70%|██████▉   | 130/187 [13:08<05:44,  6.04s/it] 
Training Epoch: 13/100, batch 260/375 completed (loss: 0.0451:  70%|██████▉   | 130/187 [13:08<05:44,  6.04s/it]
Training Epoch: 13/100, batch 260/375 completed (loss: 0.0124:  70%|██████▉   | 130/187 [13:08<05:44,  6.04s/it]
Training Epoch: 13/100, batch 260/375 completed (loss: 0.0592:  70%|██████▉   | 130/187 [13:08<05:44,  6.04s/it]
Training Epoch: 13/100, batch 260/375 completed (loss: 0.0451:  70%|███████   | 131/187 [13:11<05:34,  5.97s/it]
Training Epoch: 13/100, batch 260/375 completed (loss: 0.0592:  70%|███████   | 131/187 [13:11<05:34,  5.97s/it]
Training Epoch: 13/100, batch 260/375 completed (loss: 0.0124:  70%|███████   | 131/187 [13:11<05:34,  5.97s/it]
Training Epoch: 13/100, batch 260/375 completed (loss: 0.017:  70%|███████   | 131/187 [13:11<05:34,  5.98s/it]
Training Epoch: 13/100, batch 261/375 completed (loss: 0.0074:  70%|███████   | 131/187 [13:11<05:34,  5.98s/it]
Training Epoch: 13/100, batch 261/375 completed (loss: 0.0051:  70%|███████   | 131/187 [13:11<05:34,  5.97s/it]
Training Epoch: 13/100, batch 261/375 completed (loss: 0.0142:  70%|███████   | 131/187 [13:11<05:34,  5.97s/it]
Training Epoch: 13/100, batch 261/375 completed (loss: 0.0205:  70%|███████   | 131/187 [13:11<05:34,  5.97s/it]
Training Epoch: 13/100, batch 262/375 completed (loss: 0.0717:  70%|███████   | 131/187 [13:14<05:34,  5.97s/it]
Training Epoch: 13/100, batch 262/375 completed (loss: 0.081:  70%|███████   | 131/187 [13:14<05:34,  5.97s/it] 
Training Epoch: 13/100, batch 262/375 completed (loss: 0.064:  70%|███████   | 131/187 [13:14<05:34,  5.98s/it] 
Training Epoch: 13/100, batch 262/375 completed (loss: 0.0405:  70%|███████   | 131/187 [13:15<05:34,  5.97s/it]
Training Epoch: 13/100, batch 262/375 completed (loss: 0.081:  71%|███████   | 132/187 [13:18<05:37,  6.13s/it]
Training Epoch: 13/100, batch 262/375 completed (loss: 0.0717:  71%|███████   | 132/187 [13:18<05:37,  6.13s/it]
Training Epoch: 13/100, batch 262/375 completed (loss: 0.0405:  71%|███████   | 132/187 [13:18<05:37,  6.13s/it]
Training Epoch: 13/100, batch 262/375 completed (loss: 0.064:  71%|███████   | 132/187 [13:18<05:37,  6.13s/it]
Training Epoch: 13/100, batch 263/375 completed (loss: 0.0336:  71%|███████   | 132/187 [13:18<05:37,  6.13s/it]
Training Epoch: 13/100, batch 263/375 completed (loss: 0.0079:  71%|███████   | 132/187 [13:18<05:37,  6.13s/it]
Training Epoch: 13/100, batch 263/375 completed (loss: 0.0089:  71%|███████   | 132/187 [13:18<05:37,  6.13s/it]
Training Epoch: 13/100, batch 263/375 completed (loss: 0.0319:  71%|███████   | 132/187 [13:18<05:37,  6.13s/it]
Training Epoch: 13/100, batch 264/375 completed (loss: 0.0664:  71%|███████   | 132/187 [13:20<05:37,  6.13s/it]
Training Epoch: 13/100, batch 264/375 completed (loss: 0.0113:  71%|███████   | 132/187 [13:20<05:37,  6.13s/it]
Training Epoch: 13/100, batch 264/375 completed (loss: 0.0303:  71%|███████   | 132/187 [13:20<05:37,  6.13s/it]
Training Epoch: 13/100, batch 264/375 completed (loss: 0.0263:  71%|███████   | 132/187 [13:21<05:37,  6.13s/it]
Training Epoch: 13/100, batch 264/375 completed (loss: 0.0664:  71%|███████   | 133/187 [13:23<05:26,  6.05s/it]
Training Epoch: 13/100, batch 264/375 completed (loss: 0.0113:  71%|███████   | 133/187 [13:23<05:26,  6.05s/it]
Training Epoch: 13/100, batch 264/375 completed (loss: 0.0263:  71%|███████   | 133/187 [13:23<05:26,  6.05s/it]
Training Epoch: 13/100, batch 264/375 completed (loss: 0.0303:  71%|███████   | 133/187 [13:23<05:26,  6.05s/it]
Training Epoch: 13/100, batch 265/375 completed (loss: 0.0068:  71%|███████   | 133/187 [13:23<05:26,  6.05s/it]
Training Epoch: 13/100, batch 265/375 completed (loss: 0.016:  71%|███████   | 133/187 [13:23<05:26,  6.05s/it] 
Training Epoch: 13/100, batch 265/375 completed (loss: 0.0045:  71%|███████   | 133/187 [13:23<05:26,  6.05s/it]
Training Epoch: 13/100, batch 265/375 completed (loss: 0.0045:  71%|███████   | 133/187 [13:23<05:26,  6.05s/it]
Training Epoch: 13/100, batch 266/375 completed (loss: 0.0043:  71%|███████   | 133/187 [13:26<05:26,  6.05s/it]
Training Epoch: 13/100, batch 266/375 completed (loss: 0.1246:  71%|███████   | 133/187 [13:26<05:26,  6.05s/it]
Training Epoch: 13/100, batch 266/375 completed (loss: 0.0131:  71%|███████   | 133/187 [13:26<05:26,  6.05s/it]
Training Epoch: 13/100, batch 266/375 completed (loss: 0.0007:  71%|███████   | 133/187 [13:26<05:26,  6.05s/it]
Training Epoch: 13/100, batch 266/375 completed (loss: 0.0131:  72%|███████▏  | 134/187 [13:30<05:22,  6.08s/it]
Training Epoch: 13/100, batch 266/375 completed (loss: 0.0043:  72%|███████▏  | 134/187 [13:30<05:22,  6.09s/it]
Training Epoch: 13/100, batch 266/375 completed (loss: 0.1246:  72%|███████▏  | 134/187 [13:30<05:22,  6.09s/it]
Training Epoch: 13/100, batch 266/375 completed (loss: 0.0007:  72%|███████▏  | 134/187 [13:30<05:22,  6.09s/it]
Training Epoch: 13/100, batch 267/375 completed (loss: 0.0513:  72%|███████▏  | 134/187 [13:30<05:22,  6.09s/it]
Training Epoch: 13/100, batch 267/375 completed (loss: 0.0089:  72%|███████▏  | 134/187 [13:30<05:22,  6.08s/it]
Training Epoch: 13/100, batch 267/375 completed (loss: 0.0066:  72%|███████▏  | 134/187 [13:30<05:22,  6.09s/it]
Training Epoch: 13/100, batch 267/375 completed (loss: 0.0025:  72%|███████▏  | 134/187 [13:30<05:22,  6.09s/it]
Training Epoch: 13/100, batch 268/375 completed (loss: 0.1378:  72%|███████▏  | 134/187 [13:33<05:22,  6.09s/it]
Training Epoch: 13/100, batch 268/375 completed (loss: 0.0438:  72%|███████▏  | 134/187 [13:33<05:22,  6.08s/it]
Training Epoch: 13/100, batch 268/375 completed (loss: 0.0001:  72%|███████▏  | 134/187 [13:33<05:22,  6.09s/it]
Training Epoch: 13/100, batch 268/375 completed (loss: 0.0089:  72%|███████▏  | 134/187 [13:33<05:22,  6.09s/it]
Training Epoch: 13/100, batch 268/375 completed (loss: 0.0001:  72%|███████▏  | 135/187 [13:36<05:16,  6.08s/it]
Training Epoch: 13/100, batch 268/375 completed (loss: 0.1378:  72%|███████▏  | 135/187 [13:36<05:16,  6.08s/it]
Training Epoch: 13/100, batch 268/375 completed (loss: 0.0438:  72%|███████▏  | 135/187 [13:36<05:16,  6.08s/it]
Training Epoch: 13/100, batch 268/375 completed (loss: 0.0089:  72%|███████▏  | 135/187 [13:36<05:16,  6.08s/it]
Training Epoch: 13/100, batch 269/375 completed (loss: 0.0028:  72%|███████▏  | 135/187 [13:36<05:16,  6.08s/it]
Training Epoch: 13/100, batch 269/375 completed (loss: 0.0009:  72%|███████▏  | 135/187 [13:36<05:16,  6.08s/it]
Training Epoch: 13/100, batch 269/375 completed (loss: 0.0015:  72%|███████▏  | 135/187 [13:36<05:16,  6.08s/it]
Training Epoch: 13/100, batch 269/375 completed (loss: 0.0543:  72%|███████▏  | 135/187 [13:36<05:16,  6.08s/it]
Training Epoch: 13/100, batch 270/375 completed (loss: 0.5128:  72%|███████▏  | 135/187 [13:39<05:16,  6.08s/it]
Training Epoch: 13/100, batch 270/375 completed (loss: 0.003:  72%|███████▏  | 135/187 [13:39<05:16,  6.08s/it] 
Training Epoch: 13/100, batch 270/375 completed (loss: 0.0002:  72%|███████▏  | 135/187 [13:39<05:16,  6.08s/it]
Training Epoch: 13/100, batch 270/375 completed (loss: 0.0787:  72%|███████▏  | 135/187 [13:39<05:16,  6.08s/it]
Training Epoch: 13/100, batch 270/375 completed (loss: 0.0002:  73%|███████▎  | 136/187 [13:41<05:06,  6.01s/it]
Training Epoch: 13/100, batch 270/375 completed (loss: 0.5128:  73%|███████▎  | 136/187 [13:41<05:06,  6.02s/it]
Training Epoch: 13/100, batch 270/375 completed (loss: 0.003:  73%|███████▎  | 136/187 [13:41<05:06,  6.02s/it]
Training Epoch: 13/100, batch 270/375 completed (loss: 0.0787:  73%|███████▎  | 136/187 [13:42<05:06,  6.02s/it]
Training Epoch: 13/100, batch 271/375 completed (loss: 0.0148:  73%|███████▎  | 136/187 [13:42<05:06,  6.01s/it]
Training Epoch: 13/100, batch 271/375 completed (loss: 0.0821:  73%|███████▎  | 136/187 [13:42<05:06,  6.02s/it]
Training Epoch: 13/100, batch 271/375 completed (loss: 0.0099:  73%|███████▎  | 136/187 [13:42<05:06,  6.02s/it]
Training Epoch: 13/100, batch 271/375 completed (loss: 0.1103:  73%|███████▎  | 136/187 [13:42<05:06,  6.02s/it]
Training Epoch: 13/100, batch 272/375 completed (loss: 0.0017:  73%|███████▎  | 136/187 [13:45<05:06,  6.02s/it]
Training Epoch: 13/100, batch 272/375 completed (loss: 0.0985:  73%|███████▎  | 136/187 [13:45<05:06,  6.01s/it]
Training Epoch: 13/100, batch 272/375 completed (loss: 0.0031:  73%|███████▎  | 136/187 [13:45<05:06,  6.02s/it]
Training Epoch: 13/100, batch 272/375 completed (loss: 0.0369:  73%|███████▎  | 136/187 [13:45<05:06,  6.02s/it]
Training Epoch: 13/100, batch 272/375 completed (loss: 0.0017:  73%|███████▎  | 137/187 [13:48<05:05,  6.11s/it]
Training Epoch: 13/100, batch 272/375 completed (loss: 0.0985:  73%|███████▎  | 137/187 [13:48<05:05,  6.11s/it]
Training Epoch: 13/100, batch 272/375 completed (loss: 0.0031:  73%|███████▎  | 137/187 [13:48<05:05,  6.11s/it]
Training Epoch: 13/100, batch 272/375 completed (loss: 0.0369:  73%|███████▎  | 137/187 [13:48<05:05,  6.11s/it]
Training Epoch: 13/100, batch 273/375 completed (loss: 0.0041:  73%|███████▎  | 137/187 [13:48<05:05,  6.11s/it]
Training Epoch: 13/100, batch 273/375 completed (loss: 0.0209:  73%|███████▎  | 137/187 [13:48<05:05,  6.11s/it]
Training Epoch: 13/100, batch 273/375 completed (loss: 0.0231:  73%|███████▎  | 137/187 [13:48<05:05,  6.11s/it]
Training Epoch: 13/100, batch 273/375 completed (loss: 0.0259:  73%|███████▎  | 137/187 [13:48<05:05,  6.11s/it]
Training Epoch: 13/100, batch 274/375 completed (loss: 0.0028:  73%|███████▎  | 137/187 [13:51<05:05,  6.11s/it]
Training Epoch: 13/100, batch 274/375 completed (loss: 0.1338:  73%|███████▎  | 137/187 [13:51<05:05,  6.11s/it]
Training Epoch: 13/100, batch 274/375 completed (loss: 0.3434:  73%|███████▎  | 137/187 [13:51<05:05,  6.11s/it]
Training Epoch: 13/100, batch 274/375 completed (loss: 0.0267:  73%|███████▎  | 137/187 [13:51<05:05,  6.11s/it]
Training Epoch: 13/100, batch 274/375 completed (loss: 0.0028:  74%|███████▍  | 138/187 [13:54<04:54,  6.02s/it]
Training Epoch: 13/100, batch 274/375 completed (loss: 0.3434:  74%|███████▍  | 138/187 [13:54<04:54,  6.02s/it]
Training Epoch: 13/100, batch 274/375 completed (loss: 0.0267:  74%|███████▍  | 138/187 [13:54<04:54,  6.02s/it]
Training Epoch: 13/100, batch 274/375 completed (loss: 0.1338:  74%|███████▍  | 138/187 [13:54<04:55,  6.02s/it]
Training Epoch: 13/100, batch 275/375 completed (loss: 0.0086:  74%|███████▍  | 138/187 [13:54<04:54,  6.02s/it]
Training Epoch: 13/100, batch 275/375 completed (loss: 0.0016:  74%|███████▍  | 138/187 [13:54<04:54,  6.02s/it]
Training Epoch: 13/100, batch 275/375 completed (loss: 0.0042:  74%|███████▍  | 138/187 [13:54<04:55,  6.02s/it]
Training Epoch: 13/100, batch 275/375 completed (loss: 0.1582:  74%|███████▍  | 138/187 [13:54<04:54,  6.02s/it]
Training Epoch: 13/100, batch 276/375 completed (loss: 0.0066:  74%|███████▍  | 138/187 [13:57<04:55,  6.02s/it]
Training Epoch: 13/100, batch 276/375 completed (loss: 0.0244:  74%|███████▍  | 138/187 [13:57<04:54,  6.02s/it]
Training Epoch: 13/100, batch 276/375 completed (loss: 0.0224:  74%|███████▍  | 138/187 [13:57<04:54,  6.02s/it]
Training Epoch: 13/100, batch 276/375 completed (loss: 0.0273:  74%|███████▍  | 138/187 [13:57<04:54,  6.02s/it]
Training Epoch: 13/100, batch 276/375 completed (loss: 0.0224:  74%|███████▍  | 139/187 [14:00<04:49,  6.02s/it]
Training Epoch: 13/100, batch 276/375 completed (loss: 0.0066:  74%|███████▍  | 139/187 [14:00<04:48,  6.02s/it]
Training Epoch: 13/100, batch 276/375 completed (loss: 0.0244:  74%|███████▍  | 139/187 [14:00<04:49,  6.02s/it]
Training Epoch: 13/100, batch 276/375 completed (loss: 0.0273:  74%|███████▍  | 139/187 [14:00<04:48,  6.02s/it]
Training Epoch: 13/100, batch 277/375 completed (loss: 0.0613:  74%|███████▍  | 139/187 [14:00<04:48,  6.02s/it]
Training Epoch: 13/100, batch 277/375 completed (loss: 0.0152:  74%|███████▍  | 139/187 [14:00<04:49,  6.02s/it]
Training Epoch: 13/100, batch 277/375 completed (loss: 0.0076:  74%|███████▍  | 139/187 [14:00<04:49,  6.02s/it]
Training Epoch: 13/100, batch 277/375 completed (loss: 0.0111:  74%|███████▍  | 139/187 [14:00<04:48,  6.02s/it]
Training Epoch: 13/100, batch 278/375 completed (loss: 0.0139:  74%|███████▍  | 139/187 [14:03<04:49,  6.02s/it]
Training Epoch: 13/100, batch 278/375 completed (loss: 0.004:  74%|███████▍  | 139/187 [14:03<04:49,  6.02s/it] 
Training Epoch: 13/100, batch 278/375 completed (loss: 0.0005:  74%|███████▍  | 139/187 [14:03<04:48,  6.02s/it]
Training Epoch: 13/100, batch 278/375 completed (loss: 0.032:  74%|███████▍  | 139/187 [14:03<04:48,  6.02s/it] 
Training Epoch: 13/100, batch 278/375 completed (loss: 0.032:  75%|███████▍  | 140/187 [14:06<04:46,  6.10s/it]
Training Epoch: 13/100, batch 278/375 completed (loss: 0.004:  75%|███████▍  | 140/187 [14:06<04:46,  6.10s/it]
Training Epoch: 13/100, batch 278/375 completed (loss: 0.0139:  75%|███████▍  | 140/187 [14:06<04:46,  6.10s/it]
Training Epoch: 13/100, batch 278/375 completed (loss: 0.0005:  75%|███████▍  | 140/187 [14:06<04:46,  6.10s/it]
Training Epoch: 13/100, batch 279/375 completed (loss: 0.023:  75%|███████▍  | 140/187 [14:06<04:46,  6.10s/it]
Training Epoch: 13/100, batch 279/375 completed (loss: 0.1069:  75%|███████▍  | 140/187 [14:06<04:46,  6.10s/it]
Training Epoch: 13/100, batch 279/375 completed (loss: 0.0041:  75%|███████▍  | 140/187 [14:06<04:46,  6.10s/it]
Training Epoch: 13/100, batch 279/375 completed (loss: 0.0287:  75%|███████▍  | 140/187 [14:06<04:46,  6.10s/it]
Training Epoch: 13/100, batch 280/375 completed (loss: 0.0395:  75%|███████▍  | 140/187 [14:09<04:46,  6.10s/it]
Training Epoch: 13/100, batch 280/375 completed (loss: 0.0253:  75%|███████▍  | 140/187 [14:09<04:46,  6.10s/it]
Training Epoch: 13/100, batch 280/375 completed (loss: 0.004:  75%|███████▍  | 140/187 [14:09<04:46,  6.10s/it] 
Training Epoch: 13/100, batch 280/375 completed (loss: 0.0165:  75%|███████▍  | 140/187 [14:09<04:46,  6.10s/it]
Training Epoch: 13/100, batch 280/375 completed (loss: 0.004:  75%|███████▌  | 141/187 [14:12<04:36,  6.02s/it]
Training Epoch: 13/100, batch 280/375 completed (loss: 0.0395:  75%|███████▌  | 141/187 [14:12<04:36,  6.02s/it]
Training Epoch: 13/100, batch 280/375 completed (loss: 0.0165:  75%|███████▌  | 141/187 [14:12<04:36,  6.02s/it]
Training Epoch: 13/100, batch 280/375 completed (loss: 0.0253:  75%|███████▌  | 141/187 [14:12<04:37,  6.02s/it]
Training Epoch: 13/100, batch 281/375 completed (loss: 0.0009:  75%|███████▌  | 141/187 [14:12<04:37,  6.02s/it]
Training Epoch: 13/100, batch 281/375 completed (loss: 0.07:  75%|███████▌  | 141/187 [14:12<04:36,  6.02s/it]  
Training Epoch: 13/100, batch 281/375 completed (loss: 0.0857:  75%|███████▌  | 141/187 [14:12<04:36,  6.02s/it]
Training Epoch: 13/100, batch 281/375 completed (loss: 0.3027:  75%|███████▌  | 141/187 [14:12<04:36,  6.02s/it]
Training Epoch: 13/100, batch 282/375 completed (loss: 0.0272:  75%|███████▌  | 141/187 [14:15<04:36,  6.02s/it]
Training Epoch: 13/100, batch 282/375 completed (loss: 0.0116:  75%|███████▌  | 141/187 [14:15<04:37,  6.02s/it]
Training Epoch: 13/100, batch 282/375 completed (loss: 0.0727:  75%|███████▌  | 141/187 [14:15<04:36,  6.02s/it]
Training Epoch: 13/100, batch 282/375 completed (loss: 0.1463:  75%|███████▌  | 141/187 [14:15<04:36,  6.02s/it]
Training Epoch: 13/100, batch 282/375 completed (loss: 0.0727:  76%|███████▌  | 142/187 [14:18<04:34,  6.10s/it]
Training Epoch: 13/100, batch 282/375 completed (loss: 0.0272:  76%|███████▌  | 142/187 [14:18<04:34,  6.10s/it]
Training Epoch: 13/100, batch 282/375 completed (loss: 0.1463:  76%|███████▌  | 142/187 [14:18<04:34,  6.10s/it]
Training Epoch: 13/100, batch 282/375 completed (loss: 0.0116:  76%|███████▌  | 142/187 [14:18<04:34,  6.10s/it]
Training Epoch: 13/100, batch 283/375 completed (loss: 0.0013:  76%|███████▌  | 142/187 [14:18<04:34,  6.10s/it]
Training Epoch: 13/100, batch 283/375 completed (loss: 0.2701:  76%|███████▌  | 142/187 [14:18<04:34,  6.10s/it]
Training Epoch: 13/100, batch 283/375 completed (loss: 0.0512:  76%|███████▌  | 142/187 [14:18<04:34,  6.10s/it]
Training Epoch: 13/100, batch 283/375 completed (loss: 0.0586:  76%|███████▌  | 142/187 [14:18<04:34,  6.10s/it]
Training Epoch: 13/100, batch 284/375 completed (loss: 0.0022:  76%|███████▌  | 142/187 [14:21<04:34,  6.10s/it]
Training Epoch: 13/100, batch 284/375 completed (loss: 0.0164:  76%|███████▌  | 142/187 [14:21<04:34,  6.10s/it]
Training Epoch: 13/100, batch 284/375 completed (loss: 0.0079:  76%|███████▌  | 142/187 [14:21<04:34,  6.10s/it]
Training Epoch: 13/100, batch 284/375 completed (loss: 0.0016:  76%|███████▌  | 142/187 [14:21<04:34,  6.10s/it]
Training Epoch: 13/100, batch 284/375 completed (loss: 0.0079:  76%|███████▋  | 143/187 [14:24<04:26,  6.06s/it]
Training Epoch: 13/100, batch 284/375 completed (loss: 0.0022:  76%|███████▋  | 143/187 [14:24<04:26,  6.06s/it]
Training Epoch: 13/100, batch 284/375 completed (loss: 0.0164:  76%|███████▋  | 143/187 [14:24<04:26,  6.06s/it]
Training Epoch: 13/100, batch 284/375 completed (loss: 0.0016:  76%|███████▋  | 143/187 [14:24<04:26,  6.06s/it]
Training Epoch: 13/100, batch 285/375 completed (loss: 0.0177:  76%|███████▋  | 143/187 [14:24<04:26,  6.06s/it]
Training Epoch: 13/100, batch 285/375 completed (loss: 0.045:  76%|███████▋  | 143/187 [14:24<04:26,  6.06s/it] 
Training Epoch: 13/100, batch 285/375 completed (loss: 0.1671:  76%|███████▋  | 143/187 [14:24<04:26,  6.06s/it]
Training Epoch: 13/100, batch 285/375 completed (loss: 0.033:  76%|███████▋  | 143/187 [14:24<04:26,  6.06s/it] 
Training Epoch: 13/100, batch 286/375 completed (loss: 0.0101:  76%|███████▋  | 143/187 [14:27<04:26,  6.06s/it]
Training Epoch: 13/100, batch 286/375 completed (loss: 0.0126:  76%|███████▋  | 143/187 [14:27<04:26,  6.06s/it]
Training Epoch: 13/100, batch 286/375 completed (loss: 0.0907:  76%|███████▋  | 143/187 [14:27<04:26,  6.06s/it]
Training Epoch: 13/100, batch 286/375 completed (loss: 0.0232:  76%|███████▋  | 143/187 [14:27<04:26,  6.06s/it]
Training Epoch: 13/100, batch 286/375 completed (loss: 0.0101:  77%|███████▋  | 144/187 [14:30<04:17,  6.00s/it]
Training Epoch: 13/100, batch 286/375 completed (loss: 0.0126:  77%|███████▋  | 144/187 [14:30<04:18,  6.00s/it]
Training Epoch: 13/100, batch 286/375 completed (loss: 0.0232:  77%|███████▋  | 144/187 [14:30<04:18,  6.00s/it]
Training Epoch: 13/100, batch 286/375 completed (loss: 0.0907:  77%|███████▋  | 144/187 [14:30<04:18,  6.00s/it]
Training Epoch: 13/100, batch 287/375 completed (loss: 0.0793:  77%|███████▋  | 144/187 [14:30<04:17,  6.00s/it]
Training Epoch: 13/100, batch 287/375 completed (loss: 0.0008:  77%|███████▋  | 144/187 [14:30<04:18,  6.00s/it]
Training Epoch: 13/100, batch 287/375 completed (loss: 0.0012:  77%|███████▋  | 144/187 [14:30<04:18,  6.00s/it]
Training Epoch: 13/100, batch 287/375 completed (loss: 0.0319:  77%|███████▋  | 144/187 [14:30<04:18,  6.00s/it]
Training Epoch: 13/100, batch 288/375 completed (loss: 0.012:  77%|███████▋  | 144/187 [14:33<04:18,  6.00s/it] 
Training Epoch: 13/100, batch 288/375 completed (loss: 0.14:  77%|███████▋  | 144/187 [14:33<04:18,  6.00s/it]  
Training Epoch: 13/100, batch 288/375 completed (loss: 0.0349:  77%|███████▋  | 144/187 [14:33<04:17,  6.00s/it]
Training Epoch: 13/100, batch 288/375 completed (loss: 0.0169:  77%|███████▋  | 144/187 [14:33<04:18,  6.00s/it]
Training Epoch: 13/100, batch 288/375 completed (loss: 0.0169:  78%|███████▊  | 145/187 [14:36<04:17,  6.14s/it]
Training Epoch: 13/100, batch 288/375 completed (loss: 0.14:  78%|███████▊  | 145/187 [14:36<04:18,  6.14s/it]
Training Epoch: 13/100, batch 288/375 completed (loss: 0.0349:  78%|███████▊  | 145/187 [14:36<04:18,  6.14s/it]
Training Epoch: 13/100, batch 288/375 completed (loss: 0.012:  78%|███████▊  | 145/187 [14:36<04:18,  6.14s/it]
Training Epoch: 13/100, batch 289/375 completed (loss: 0.0624:  78%|███████▊  | 145/187 [14:36<04:18,  6.14s/it]
Training Epoch: 13/100, batch 289/375 completed (loss: 0.0037:  78%|███████▊  | 145/187 [14:36<04:18,  6.14s/it]
Training Epoch: 13/100, batch 289/375 completed (loss: 0.0373:  78%|███████▊  | 145/187 [14:36<04:18,  6.14s/it]
Training Epoch: 13/100, batch 289/375 completed (loss: 0.0081:  78%|███████▊  | 145/187 [14:36<04:17,  6.14s/it]
Training Epoch: 13/100, batch 290/375 completed (loss: 0.0196:  78%|███████▊  | 145/187 [14:39<04:18,  6.14s/it]
Training Epoch: 13/100, batch 290/375 completed (loss: 0.0108:  78%|███████▊  | 145/187 [14:39<04:18,  6.14s/it]
Training Epoch: 13/100, batch 290/375 completed (loss: 0.0248:  78%|███████▊  | 145/187 [14:39<04:18,  6.14s/it]
Training Epoch: 13/100, batch 290/375 completed (loss: 0.0111:  78%|███████▊  | 145/187 [14:39<04:17,  6.14s/it]
Training Epoch: 13/100, batch 290/375 completed (loss: 0.0108:  78%|███████▊  | 146/187 [14:42<04:08,  6.06s/it]
Training Epoch: 13/100, batch 290/375 completed (loss: 0.0111:  78%|███████▊  | 146/187 [14:42<04:08,  6.06s/it]
Training Epoch: 13/100, batch 290/375 completed (loss: 0.0248:  78%|███████▊  | 146/187 [14:42<04:08,  6.06s/it]
Training Epoch: 13/100, batch 290/375 completed (loss: 0.0196:  78%|███████▊  | 146/187 [14:42<04:08,  6.06s/it]
Training Epoch: 13/100, batch 291/375 completed (loss: 0.0761:  78%|███████▊  | 146/187 [14:42<04:08,  6.06s/it]
Training Epoch: 13/100, batch 291/375 completed (loss: 0.2091:  78%|███████▊  | 146/187 [14:42<04:08,  6.06s/it]
Training Epoch: 13/100, batch 291/375 completed (loss: 0.0047:  78%|███████▊  | 146/187 [14:42<04:08,  6.06s/it]
Training Epoch: 13/100, batch 291/375 completed (loss: 0.1867:  78%|███████▊  | 146/187 [14:42<04:08,  6.06s/it]
Training Epoch: 13/100, batch 292/375 completed (loss: 0.0643:  78%|███████▊  | 146/187 [14:45<04:08,  6.06s/it]
Training Epoch: 13/100, batch 292/375 completed (loss: 0.0316:  78%|███████▊  | 146/187 [14:45<04:08,  6.06s/it]
Training Epoch: 13/100, batch 292/375 completed (loss: 0.0014:  78%|███████▊  | 146/187 [14:45<04:08,  6.06s/it]
Training Epoch: 13/100, batch 292/375 completed (loss: 0.0032:  78%|███████▊  | 146/187 [14:45<04:08,  6.06s/it]
Training Epoch: 13/100, batch 292/375 completed (loss: 0.0032:  79%|███████▊  | 147/187 [14:49<04:05,  6.14s/it]
Training Epoch: 13/100, batch 292/375 completed (loss: 0.0643:  79%|███████▊  | 147/187 [14:49<04:05,  6.14s/it]
Training Epoch: 13/100, batch 292/375 completed (loss: 0.0014:  79%|███████▊  | 147/187 [14:49<04:05,  6.14s/it]
Training Epoch: 13/100, batch 292/375 completed (loss: 0.0316:  79%|███████▊  | 147/187 [14:49<04:05,  6.14s/it]
Training Epoch: 13/100, batch 293/375 completed (loss: 0.0055:  79%|███████▊  | 147/187 [14:49<04:05,  6.14s/it]
Training Epoch: 13/100, batch 293/375 completed (loss: 0.0787:  79%|███████▊  | 147/187 [14:49<04:05,  6.14s/it]
Training Epoch: 13/100, batch 293/375 completed (loss: 0.017:  79%|███████▊  | 147/187 [14:49<04:05,  6.14s/it] 
Training Epoch: 13/100, batch 293/375 completed (loss: 0.0043:  79%|███████▊  | 147/187 [14:49<04:05,  6.14s/it]
Training Epoch: 13/100, batch 294/375 completed (loss: 0.0231:  79%|███████▊  | 147/187 [14:52<04:05,  6.14s/it]
Training Epoch: 13/100, batch 294/375 completed (loss: 0.0049:  79%|███████▊  | 147/187 [14:52<04:05,  6.14s/it]
Training Epoch: 13/100, batch 294/375 completed (loss: 0.0012:  79%|███████▊  | 147/187 [14:52<04:05,  6.14s/it]
Training Epoch: 13/100, batch 294/375 completed (loss: 0.0387:  79%|███████▊  | 147/187 [14:52<04:05,  6.14s/it]
Training Epoch: 13/100, batch 294/375 completed (loss: 0.0231:  79%|███████▉  | 148/187 [14:55<03:58,  6.10s/it]
Training Epoch: 13/100, batch 294/375 completed (loss: 0.0387:  79%|███████▉  | 148/187 [14:55<03:58,  6.10s/it]
Training Epoch: 13/100, batch 294/375 completed (loss: 0.0049:  79%|███████▉  | 148/187 [14:55<03:58,  6.10s/it]
Training Epoch: 13/100, batch 294/375 completed (loss: 0.0012:  79%|███████▉  | 148/187 [14:55<03:58,  6.10s/it]
Training Epoch: 13/100, batch 295/375 completed (loss: 0.0052:  79%|███████▉  | 148/187 [14:55<03:58,  6.10s/it]
Training Epoch: 13/100, batch 295/375 completed (loss: 0.0118:  79%|███████▉  | 148/187 [14:55<03:58,  6.10s/it]
Training Epoch: 13/100, batch 295/375 completed (loss: 0.0244:  79%|███████▉  | 148/187 [14:55<03:58,  6.10s/it]
Training Epoch: 13/100, batch 295/375 completed (loss: 0.0279:  79%|███████▉  | 148/187 [14:55<03:58,  6.10s/it]
Training Epoch: 13/100, batch 296/375 completed (loss: 0.0574:  79%|███████▉  | 148/187 [14:58<03:58,  6.10s/it]
Training Epoch: 13/100, batch 296/375 completed (loss: 0.0324:  79%|███████▉  | 148/187 [14:58<03:58,  6.10s/it]
Training Epoch: 13/100, batch 296/375 completed (loss: 0.0002:  79%|███████▉  | 148/187 [14:58<03:58,  6.10s/it]
Training Epoch: 13/100, batch 296/375 completed (loss: 0.0336:  79%|███████▉  | 148/187 [14:58<03:58,  6.10s/it]
Training Epoch: 13/100, batch 296/375 completed (loss: 0.0002:  80%|███████▉  | 149/187 [15:00<03:48,  6.03s/it]
Training Epoch: 13/100, batch 296/375 completed (loss: 0.0336:  80%|███████▉  | 149/187 [15:00<03:49,  6.03s/it]
Training Epoch: 13/100, batch 296/375 completed (loss: 0.0574:  80%|███████▉  | 149/187 [15:00<03:49,  6.03s/it]
Training Epoch: 13/100, batch 296/375 completed (loss: 0.0324:  80%|███████▉  | 149/187 [15:00<03:49,  6.03s/it]
Training Epoch: 13/100, batch 297/375 completed (loss: 0.0294:  80%|███████▉  | 149/187 [15:00<03:49,  6.03s/it]
Training Epoch: 13/100, batch 297/375 completed (loss: 0.0008:  80%|███████▉  | 149/187 [15:00<03:49,  6.03s/it]
Training Epoch: 13/100, batch 297/375 completed (loss: 0.0525:  80%|███████▉  | 149/187 [15:00<03:48,  6.03s/it]
Training Epoch: 13/100, batch 297/375 completed (loss: 0.0093:  80%|███████▉  | 149/187 [15:00<03:49,  6.03s/it]
Training Epoch: 13/100, batch 298/375 completed (loss: 0.0258:  80%|███████▉  | 149/187 [15:04<03:49,  6.03s/it]
Training Epoch: 13/100, batch 298/375 completed (loss: 0.0105:  80%|███████▉  | 149/187 [15:04<03:49,  6.03s/it]
Training Epoch: 13/100, batch 298/375 completed (loss: 0.0099:  80%|███████▉  | 149/187 [15:04<03:48,  6.03s/it]
Training Epoch: 13/100, batch 298/375 completed (loss: 0.0182:  80%|███████▉  | 149/187 [15:04<03:49,  6.03s/it]
Training Epoch: 13/100, batch 298/375 completed (loss: 0.0182:  80%|████████  | 150/187 [15:07<03:47,  6.15s/it]
Training Epoch: 13/100, batch 298/375 completed (loss: 0.0099:  80%|████████  | 150/187 [15:07<03:47,  6.15s/it]
Training Epoch: 13/100, batch 298/375 completed (loss: 0.0105:  80%|████████  | 150/187 [15:07<03:47,  6.15s/it]
Training Epoch: 13/100, batch 298/375 completed (loss: 0.0258:  80%|████████  | 150/187 [15:07<03:47,  6.15s/it]
Training Epoch: 13/100, batch 299/375 completed (loss: 0.0087:  80%|████████  | 150/187 [15:07<03:47,  6.15s/it]
Training Epoch: 13/100, batch 299/375 completed (loss: 0.0005:  80%|████████  | 150/187 [15:07<03:47,  6.15s/it]
Training Epoch: 13/100, batch 299/375 completed (loss: 0.0008:  80%|████████  | 150/187 [15:07<03:47,  6.15s/it]
Training Epoch: 13/100, batch 299/375 completed (loss: 0.0222:  80%|████████  | 150/187 [15:07<03:47,  6.15s/it]
Training Epoch: 13/100, batch 300/375 completed (loss: 0.0908:  80%|████████  | 150/187 [15:10<03:47,  6.15s/it]
Training Epoch: 13/100, batch 300/375 completed (loss: 0.0129:  80%|████████  | 150/187 [15:10<03:47,  6.15s/it]
Training Epoch: 13/100, batch 300/375 completed (loss: 0.0278:  80%|████████  | 150/187 [15:10<03:47,  6.15s/it]
Training Epoch: 13/100, batch 300/375 completed (loss: 0.0101:  80%|████████  | 150/187 [15:10<03:47,  6.15s/it]
Training Epoch: 13/100, batch 300/375 completed (loss: 0.0908:  81%|████████  | 151/187 [15:13<03:38,  6.06s/it]
Training Epoch: 13/100, batch 300/375 completed (loss: 0.0129:  81%|████████  | 151/187 [15:13<03:38,  6.07s/it]
Training Epoch: 13/100, batch 300/375 completed (loss: 0.0278:  81%|████████  | 151/187 [15:13<03:38,  6.07s/it]
Training Epoch: 13/100, batch 300/375 completed (loss: 0.0101:  81%|████████  | 151/187 [15:13<03:38,  6.07s/it]
Training Epoch: 13/100, batch 301/375 completed (loss: 0.0583:  81%|████████  | 151/187 [15:13<03:38,  6.06s/it]
Training Epoch: 13/100, batch 301/375 completed (loss: 0.0014:  81%|████████  | 151/187 [15:13<03:38,  6.07s/it]
Training Epoch: 13/100, batch 301/375 completed (loss: 0.0709:  81%|████████  | 151/187 [15:13<03:38,  6.07s/it]
Training Epoch: 13/100, batch 301/375 completed (loss: 0.0824:  81%|████████  | 151/187 [15:13<03:38,  6.07s/it]
Training Epoch: 13/100, batch 302/375 completed (loss: 0.0062:  81%|████████  | 151/187 [15:16<03:38,  6.06s/it]
Training Epoch: 13/100, batch 302/375 completed (loss: 0.4817:  81%|████████  | 151/187 [15:16<03:38,  6.07s/it]
Training Epoch: 13/100, batch 302/375 completed (loss: 0.0047:  81%|████████  | 151/187 [15:16<03:38,  6.07s/it]
Training Epoch: 13/100, batch 302/375 completed (loss: 0.0436:  81%|████████  | 151/187 [15:16<03:38,  6.07s/it]
Training Epoch: 13/100, batch 302/375 completed (loss: 0.0436:  81%|████████▏ | 152/187 [15:19<03:31,  6.06s/it]
Training Epoch: 13/100, batch 302/375 completed (loss: 0.0062:  81%|████████▏ | 152/187 [15:19<03:31,  6.06s/it]
Training Epoch: 13/100, batch 302/375 completed (loss: 0.4817:  81%|████████▏ | 152/187 [15:19<03:32,  6.06s/it]
Training Epoch: 13/100, batch 302/375 completed (loss: 0.0047:  81%|████████▏ | 152/187 [15:19<03:31,  6.06s/it]
Training Epoch: 13/100, batch 303/375 completed (loss: 0.0015:  81%|████████▏ | 152/187 [15:19<03:31,  6.06s/it]
Training Epoch: 13/100, batch 303/375 completed (loss: 0.0036:  81%|████████▏ | 152/187 [15:19<03:31,  6.06s/it]
Training Epoch: 13/100, batch 303/375 completed (loss: 0.0038:  81%|████████▏ | 152/187 [15:19<03:32,  6.06s/it]
Training Epoch: 13/100, batch 303/375 completed (loss: 0.008:  81%|████████▏ | 152/187 [15:19<03:31,  6.06s/it] 
Training Epoch: 13/100, batch 304/375 completed (loss: 0.1151:  81%|████████▏ | 152/187 [15:22<03:31,  6.06s/it]
Training Epoch: 13/100, batch 304/375 completed (loss: 0.0113:  81%|████████▏ | 152/187 [15:22<03:32,  6.06s/it]
Training Epoch: 13/100, batch 304/375 completed (loss: 0.0077:  81%|████████▏ | 152/187 [15:22<03:31,  6.06s/it]
Training Epoch: 13/100, batch 304/375 completed (loss: 0.011:  81%|████████▏ | 152/187 [15:22<03:31,  6.06s/it]
Training Epoch: 13/100, batch 304/375 completed (loss: 0.0113:  82%|████████▏ | 153/187 [15:25<03:28,  6.12s/it]
Training Epoch: 13/100, batch 304/375 completed (loss: 0.0077:  82%|████████▏ | 153/187 [15:25<03:28,  6.12s/it]
Training Epoch: 13/100, batch 304/375 completed (loss: 0.011:  82%|████████▏ | 153/187 [15:25<03:28,  6.13s/it]
Training Epoch: 13/100, batch 304/375 completed (loss: 0.1151:  82%|████████▏ | 153/187 [15:25<03:28,  6.13s/it]
Training Epoch: 13/100, batch 305/375 completed (loss: 0.0711:  82%|████████▏ | 153/187 [15:25<03:28,  6.13s/it]
Training Epoch: 13/100, batch 305/375 completed (loss: 0.1086:  82%|████████▏ | 153/187 [15:25<03:28,  6.12s/it]
Training Epoch: 13/100, batch 305/375 completed (loss: 0.0131:  82%|████████▏ | 153/187 [15:25<03:28,  6.12s/it]
Training Epoch: 13/100, batch 305/375 completed (loss: 0.2063:  82%|████████▏ | 153/187 [15:25<03:28,  6.13s/it]
Training Epoch: 13/100, batch 306/375 completed (loss: 0.0039:  82%|████████▏ | 153/187 [15:28<03:28,  6.12s/it]
Training Epoch: 13/100, batch 306/375 completed (loss: 0.073:  82%|████████▏ | 153/187 [15:28<03:28,  6.12s/it] 
Training Epoch: 13/100, batch 306/375 completed (loss: 0.0099:  82%|████████▏ | 153/187 [15:28<03:28,  6.13s/it]
Training Epoch: 13/100, batch 306/375 completed (loss: 0.0452:  82%|████████▏ | 153/187 [15:28<03:28,  6.13s/it]
Training Epoch: 13/100, batch 306/375 completed (loss: 0.0452:  82%|████████▏ | 154/187 [15:31<03:19,  6.04s/it]
Training Epoch: 13/100, batch 306/375 completed (loss: 0.073:  82%|████████▏ | 154/187 [15:31<03:19,  6.04s/it]
Training Epoch: 13/100, batch 306/375 completed (loss: 0.0039:  82%|████████▏ | 154/187 [15:31<03:19,  6.04s/it]
Training Epoch: 13/100, batch 306/375 completed (loss: 0.0099:  82%|████████▏ | 154/187 [15:31<03:19,  6.04s/it]
Training Epoch: 13/100, batch 307/375 completed (loss: 0.023:  82%|████████▏ | 154/187 [15:31<03:19,  6.04s/it] 
Training Epoch: 13/100, batch 307/375 completed (loss: 0.007:  82%|████████▏ | 154/187 [15:31<03:19,  6.04s/it]
Training Epoch: 13/100, batch 307/375 completed (loss: 0.0007:  82%|████████▏ | 154/187 [15:31<03:19,  6.04s/it]
Training Epoch: 13/100, batch 307/375 completed (loss: 0.0543:  82%|████████▏ | 154/187 [15:31<03:19,  6.04s/it]
Training Epoch: 13/100, batch 308/375 completed (loss: 0.0134:  82%|████████▏ | 154/187 [15:34<03:19,  6.04s/it]
Training Epoch: 13/100, batch 308/375 completed (loss: 0.035:  82%|████████▏ | 154/187 [15:34<03:19,  6.04s/it] 
Training Epoch: 13/100, batch 308/375 completed (loss: 0.0456:  82%|████████▏ | 154/187 [15:34<03:19,  6.04s/it]
Training Epoch: 13/100, batch 308/375 completed (loss: 0.0208:  82%|████████▏ | 154/187 [15:34<03:19,  6.04s/it]
Training Epoch: 13/100, batch 308/375 completed (loss: 0.0134:  83%|████████▎ | 155/187 [15:37<03:15,  6.12s/it]
Training Epoch: 13/100, batch 308/375 completed (loss: 0.0456:  83%|████████▎ | 155/187 [15:37<03:15,  6.12s/it]
Training Epoch: 13/100, batch 308/375 completed (loss: 0.035:  83%|████████▎ | 155/187 [15:37<03:15,  6.12s/it]
Training Epoch: 13/100, batch 308/375 completed (loss: 0.0208:  83%|████████▎ | 155/187 [15:37<03:15,  6.12s/it]
Training Epoch: 13/100, batch 309/375 completed (loss: 0.0379:  83%|████████▎ | 155/187 [15:37<03:15,  6.12s/it]
Training Epoch: 13/100, batch 309/375 completed (loss: 0.049:  83%|████████▎ | 155/187 [15:37<03:15,  6.12s/it] 
Training Epoch: 13/100, batch 309/375 completed (loss: 0.0033:  83%|████████▎ | 155/187 [15:37<03:15,  6.12s/it]
Training Epoch: 13/100, batch 309/375 completed (loss: 0.0013:  83%|████████▎ | 155/187 [15:37<03:15,  6.12s/it]
Training Epoch: 13/100, batch 310/375 completed (loss: 0.0099:  83%|████████▎ | 155/187 [15:40<03:15,  6.12s/it]
Training Epoch: 13/100, batch 310/375 completed (loss: 0.0032:  83%|████████▎ | 155/187 [15:40<03:15,  6.12s/it]
Training Epoch: 13/100, batch 310/375 completed (loss: 0.022:  83%|████████▎ | 155/187 [15:40<03:15,  6.12s/it] 
Training Epoch: 13/100, batch 310/375 completed (loss: 0.0273:  83%|████████▎ | 155/187 [15:40<03:15,  6.12s/it]
Training Epoch: 13/100, batch 310/375 completed (loss: 0.0273:  83%|████████▎ | 156/187 [15:43<03:08,  6.09s/it]
Training Epoch: 13/100, batch 310/375 completed (loss: 0.0032:  83%|████████▎ | 156/187 [15:43<03:08,  6.10s/it]
Training Epoch: 13/100, batch 310/375 completed (loss: 0.022:  83%|████████▎ | 156/187 [15:43<03:08,  6.10s/it]
Training Epoch: 13/100, batch 310/375 completed (loss: 0.0099:  83%|████████▎ | 156/187 [15:43<03:09,  6.10s/it]
Training Epoch: 13/100, batch 311/375 completed (loss: 0.0034:  83%|████████▎ | 156/187 [15:43<03:08,  6.10s/it]
Training Epoch: 13/100, batch 311/375 completed (loss: 0.001:  83%|████████▎ | 156/187 [15:43<03:08,  6.10s/it] 
Training Epoch: 13/100, batch 311/375 completed (loss: 0.01:  83%|████████▎ | 156/187 [15:43<03:09,  6.10s/it]  
Training Epoch: 13/100, batch 311/375 completed (loss: 0.0111:  83%|████████▎ | 156/187 [15:43<03:08,  6.09s/it]
Training Epoch: 13/100, batch 312/375 completed (loss: 0.0539:  83%|████████▎ | 156/187 [15:46<03:08,  6.10s/it]
Training Epoch: 13/100, batch 312/375 completed (loss: 0.0916:  83%|████████▎ | 156/187 [15:46<03:09,  6.10s/it]
Training Epoch: 13/100, batch 312/375 completed (loss: 0.0027:  83%|████████▎ | 156/187 [15:46<03:08,  6.10s/it]
Training Epoch: 13/100, batch 312/375 completed (loss: 0.0253:  83%|████████▎ | 156/187 [15:46<03:08,  6.09s/it]
Training Epoch: 13/100, batch 312/375 completed (loss: 0.0027:  84%|████████▍ | 157/187 [15:49<03:00,  6.02s/it]
Training Epoch: 13/100, batch 312/375 completed (loss: 0.0916:  84%|████████▍ | 157/187 [15:49<03:00,  6.02s/it]
Training Epoch: 13/100, batch 312/375 completed (loss: 0.0539:  84%|████████▍ | 157/187 [15:49<03:00,  6.02s/it]
Training Epoch: 13/100, batch 312/375 completed (loss: 0.0253:  84%|████████▍ | 157/187 [15:49<03:00,  6.03s/it]
Training Epoch: 13/100, batch 313/375 completed (loss: 0.0573:  84%|████████▍ | 157/187 [15:49<03:00,  6.02s/it]
Training Epoch: 13/100, batch 313/375 completed (loss: 0.0098:  84%|████████▍ | 157/187 [15:49<03:00,  6.02s/it]
Training Epoch: 13/100, batch 313/375 completed (loss: 0.0055:  84%|████████▍ | 157/187 [15:49<03:00,  6.02s/it]
Training Epoch: 13/100, batch 313/375 completed (loss: 0.0084:  84%|████████▍ | 157/187 [15:49<03:00,  6.03s/it]
Training Epoch: 13/100, batch 314/375 completed (loss: 0.0027:  84%|████████▍ | 157/187 [15:52<03:00,  6.02s/it]
Training Epoch: 13/100, batch 314/375 completed (loss: 0.008:  84%|████████▍ | 157/187 [15:52<03:00,  6.02s/it] 
Training Epoch: 13/100, batch 314/375 completed (loss: 0.0082:  84%|████████▍ | 157/187 [15:52<03:00,  6.02s/it]
Training Epoch: 13/100, batch 314/375 completed (loss: 0.0702:  84%|████████▍ | 157/187 [15:52<03:00,  6.03s/it]
Training Epoch: 13/100, batch 314/375 completed (loss: 0.0082:  84%|████████▍ | 158/187 [15:56<02:58,  6.15s/it]
Training Epoch: 13/100, batch 314/375 completed (loss: 0.0027:  84%|████████▍ | 158/187 [15:56<02:58,  6.16s/it]
Training Epoch: 13/100, batch 314/375 completed (loss: 0.008:  84%|████████▍ | 158/187 [15:56<02:58,  6.16s/it]
Training Epoch: 13/100, batch 314/375 completed (loss: 0.0702:  84%|████████▍ | 158/187 [15:56<02:58,  6.16s/it]
Training Epoch: 13/100, batch 315/375 completed (loss: 0.0097:  84%|████████▍ | 158/187 [15:56<02:58,  6.16s/it]
Training Epoch: 13/100, batch 315/375 completed (loss: 0.0153:  84%|████████▍ | 158/187 [15:56<02:58,  6.15s/it]
Training Epoch: 13/100, batch 315/375 completed (loss: 0.2018:  84%|████████▍ | 158/187 [15:56<02:58,  6.16s/it]
Training Epoch: 13/100, batch 315/375 completed (loss: 0.033:  84%|████████▍ | 158/187 [15:56<02:58,  6.16s/it] 
Training Epoch: 13/100, batch 316/375 completed (loss: 0.0338:  84%|████████▍ | 158/187 [15:59<02:58,  6.15s/it]
Training Epoch: 13/100, batch 316/375 completed (loss: 0.0261:  84%|████████▍ | 158/187 [15:59<02:58,  6.16s/it]
Training Epoch: 13/100, batch 316/375 completed (loss: 0.0076:  84%|████████▍ | 158/187 [15:59<02:58,  6.16s/it]
Training Epoch: 13/100, batch 316/375 completed (loss: 0.0051:  84%|████████▍ | 158/187 [15:59<02:58,  6.16s/it]
Training Epoch: 13/100, batch 316/375 completed (loss: 0.0338:  85%|████████▌ | 159/187 [16:01<02:49,  6.06s/it]
Training Epoch: 13/100, batch 316/375 completed (loss: 0.0076:  85%|████████▌ | 159/187 [16:01<02:49,  6.07s/it]
Training Epoch: 13/100, batch 316/375 completed (loss: 0.0261:  85%|████████▌ | 159/187 [16:01<02:49,  6.07s/it]
Training Epoch: 13/100, batch 316/375 completed (loss: 0.0051:  85%|████████▌ | 159/187 [16:01<02:49,  6.06s/it]
Training Epoch: 13/100, batch 317/375 completed (loss: 0.0369:  85%|████████▌ | 159/187 [16:01<02:49,  6.07s/it]
Training Epoch: 13/100, batch 317/375 completed (loss: 0.0216:  85%|████████▌ | 159/187 [16:01<02:49,  6.07s/it]
Training Epoch: 13/100, batch 317/375 completed (loss: 0.005:  85%|████████▌ | 159/187 [16:01<02:49,  6.06s/it] 
Training Epoch: 13/100, batch 317/375 completed (loss: 0.079:  85%|████████▌ | 159/187 [16:01<02:49,  6.06s/it] 
Training Epoch: 13/100, batch 318/375 completed (loss: 0.189:  85%|████████▌ | 159/187 [16:04<02:49,  6.07s/it] 
Training Epoch: 13/100, batch 318/375 completed (loss: 0.0478:  85%|████████▌ | 159/187 [16:04<02:49,  6.07s/it]
Training Epoch: 13/100, batch 318/375 completed (loss: 0.0224:  85%|████████▌ | 159/187 [16:04<02:49,  6.06s/it]
Training Epoch: 13/100, batch 318/375 completed (loss: 0.0478:  85%|████████▌ | 159/187 [16:04<02:49,  6.06s/it]
Training Epoch: 13/100, batch 318/375 completed (loss: 0.189:  86%|████████▌ | 160/187 [16:08<02:44,  6.10s/it]
Training Epoch: 13/100, batch 318/375 completed (loss: 0.0478:  86%|████████▌ | 160/187 [16:08<02:44,  6.10s/it]
Training Epoch: 13/100, batch 318/375 completed (loss: 0.0224:  86%|████████▌ | 160/187 [16:08<02:44,  6.10s/it]
Training Epoch: 13/100, batch 318/375 completed (loss: 0.0478:  86%|████████▌ | 160/187 [16:08<02:44,  6.10s/it]
Training Epoch: 13/100, batch 319/375 completed (loss: 0.0038:  86%|████████▌ | 160/187 [16:08<02:44,  6.10s/it]
Training Epoch: 13/100, batch 319/375 completed (loss: 0.0398:  86%|████████▌ | 160/187 [16:08<02:44,  6.10s/it]
Training Epoch: 13/100, batch 319/375 completed (loss: 0.0025:  86%|████████▌ | 160/187 [16:08<02:44,  6.10s/it]
Training Epoch: 13/100, batch 319/375 completed (loss: 0.0069:  86%|████████▌ | 160/187 [16:08<02:44,  6.10s/it]
Training Epoch: 13/100, batch 320/375 completed (loss: 0.0182:  86%|████████▌ | 160/187 [16:11<02:44,  6.10s/it]
Training Epoch: 13/100, batch 320/375 completed (loss: 0.0008:  86%|████████▌ | 160/187 [16:11<02:44,  6.10s/it]
Training Epoch: 13/100, batch 320/375 completed (loss: 0.0264:  86%|████████▌ | 160/187 [16:11<02:44,  6.10s/it]
Training Epoch: 13/100, batch 320/375 completed (loss: 0.06:  86%|████████▌ | 160/187 [16:11<02:44,  6.10s/it]  
Training Epoch: 13/100, batch 320/375 completed (loss: 0.0182:  86%|████████▌ | 161/187 [16:14<02:39,  6.12s/it]
Training Epoch: 13/100, batch 320/375 completed (loss: 0.0264:  86%|████████▌ | 161/187 [16:14<02:39,  6.12s/it]
Training Epoch: 13/100, batch 320/375 completed (loss: 0.0008:  86%|████████▌ | 161/187 [16:14<02:39,  6.12s/it]
Training Epoch: 13/100, batch 320/375 completed (loss: 0.06:  86%|████████▌ | 161/187 [16:14<02:39,  6.12s/it]
Training Epoch: 13/100, batch 321/375 completed (loss: 0.0257:  86%|████████▌ | 161/187 [16:14<02:39,  6.12s/it]
Training Epoch: 13/100, batch 321/375 completed (loss: 0.545:  86%|████████▌ | 161/187 [16:14<02:39,  6.12s/it] 
Training Epoch: 13/100, batch 321/375 completed (loss: 0.0059:  86%|████████▌ | 161/187 [16:14<02:39,  6.12s/it]
Training Epoch: 13/100, batch 321/375 completed (loss: 0.121:  86%|████████▌ | 161/187 [16:14<02:39,  6.12s/it]
Training Epoch: 13/100, batch 322/375 completed (loss: 0.2847:  86%|████████▌ | 161/187 [16:17<02:39,  6.12s/it]
Training Epoch: 13/100, batch 322/375 completed (loss: 0.197:  86%|████████▌ | 161/187 [16:17<02:39,  6.12s/it] 
Training Epoch: 13/100, batch 322/375 completed (loss: 0.0065:  86%|████████▌ | 161/187 [16:17<02:39,  6.12s/it]
Training Epoch: 13/100, batch 322/375 completed (loss: 0.0031:  86%|████████▌ | 161/187 [16:17<02:39,  6.12s/it]
Training Epoch: 13/100, batch 322/375 completed (loss: 0.0031:  87%|████████▋ | 162/187 [16:20<02:30,  6.04s/it]
Training Epoch: 13/100, batch 322/375 completed (loss: 0.197:  87%|████████▋ | 162/187 [16:20<02:30,  6.04s/it]
Training Epoch: 13/100, batch 322/375 completed (loss: 0.0065:  87%|████████▋ | 162/187 [16:20<02:31,  6.04s/it]
Training Epoch: 13/100, batch 322/375 completed (loss: 0.2847:  87%|████████▋ | 162/187 [16:20<02:31,  6.04s/it]
Training Epoch: 13/100, batch 323/375 completed (loss: 0.0064:  87%|████████▋ | 162/187 [16:20<02:31,  6.04s/it]
Training Epoch: 13/100, batch 323/375 completed (loss: 0.0046:  87%|████████▋ | 162/187 [16:20<02:31,  6.04s/it]
Training Epoch: 13/100, batch 323/375 completed (loss: 0.0351:  87%|████████▋ | 162/187 [16:20<02:30,  6.04s/it]
Training Epoch: 13/100, batch 323/375 completed (loss: 0.001:  87%|████████▋ | 162/187 [16:20<02:30,  6.04s/it] 
Training Epoch: 13/100, batch 324/375 completed (loss: 0.0033:  87%|████████▋ | 162/187 [16:23<02:31,  6.04s/it]
Training Epoch: 13/100, batch 324/375 completed (loss: 0.0152:  87%|████████▋ | 162/187 [16:23<02:30,  6.04s/it]
Training Epoch: 13/100, batch 324/375 completed (loss: 0.0176:  87%|████████▋ | 162/187 [16:23<02:31,  6.04s/it]
Training Epoch: 13/100, batch 324/375 completed (loss: 0.0441:  87%|████████▋ | 162/187 [16:23<02:30,  6.04s/it]
Training Epoch: 13/100, batch 324/375 completed (loss: 0.0176:  87%|████████▋ | 163/187 [16:26<02:26,  6.12s/it]
Training Epoch: 13/100, batch 324/375 completed (loss: 0.0152:  87%|████████▋ | 163/187 [16:26<02:26,  6.12s/it]
Training Epoch: 13/100, batch 324/375 completed (loss: 0.0033:  87%|████████▋ | 163/187 [16:26<02:26,  6.12s/it]
Training Epoch: 13/100, batch 324/375 completed (loss: 0.0441:  87%|████████▋ | 163/187 [16:26<02:26,  6.12s/it]
Training Epoch: 13/100, batch 325/375 completed (loss: 0.0885:  87%|████████▋ | 163/187 [16:26<02:26,  6.12s/it]
Training Epoch: 13/100, batch 325/375 completed (loss: 0.0025:  87%|████████▋ | 163/187 [16:26<02:26,  6.12s/it]
Training Epoch: 13/100, batch 325/375 completed (loss: 0.0088:  87%|████████▋ | 163/187 [16:26<02:26,  6.12s/it]
Training Epoch: 13/100, batch 325/375 completed (loss: 0.0069:  87%|████████▋ | 163/187 [16:26<02:26,  6.12s/it]
Training Epoch: 13/100, batch 326/375 completed (loss: 0.0799:  87%|████████▋ | 163/187 [16:29<02:26,  6.12s/it]
Training Epoch: 13/100, batch 326/375 completed (loss: 0.0281:  87%|████████▋ | 163/187 [16:29<02:26,  6.12s/it]
Training Epoch: 13/100, batch 326/375 completed (loss: 0.0334:  87%|████████▋ | 163/187 [16:29<02:26,  6.12s/it]
Training Epoch: 13/100, batch 326/375 completed (loss: 0.0202:  87%|████████▋ | 163/187 [16:29<02:26,  6.12s/it]
Training Epoch: 13/100, batch 326/375 completed (loss: 0.0334:  88%|████████▊ | 164/187 [16:32<02:19,  6.08s/it]
Training Epoch: 13/100, batch 326/375 completed (loss: 0.0281:  88%|████████▊ | 164/187 [16:32<02:19,  6.08s/it]
Training Epoch: 13/100, batch 326/375 completed (loss: 0.0799:  88%|████████▊ | 164/187 [16:32<02:19,  6.08s/it]
Training Epoch: 13/100, batch 326/375 completed (loss: 0.0202:  88%|████████▊ | 164/187 [16:32<02:19,  6.08s/it]
Training Epoch: 13/100, batch 327/375 completed (loss: 0.0034:  88%|████████▊ | 164/187 [16:32<02:19,  6.08s/it]
Training Epoch: 13/100, batch 327/375 completed (loss: 0.0064:  88%|████████▊ | 164/187 [16:32<02:19,  6.08s/it]
Training Epoch: 13/100, batch 327/375 completed (loss: 0.1115:  88%|████████▊ | 164/187 [16:32<02:19,  6.08s/it]
Training Epoch: 13/100, batch 327/375 completed (loss: 0.0261:  88%|████████▊ | 164/187 [16:32<02:19,  6.08s/it]
Training Epoch: 13/100, batch 328/375 completed (loss: 0.0067:  88%|████████▊ | 164/187 [16:35<02:19,  6.08s/it]
Training Epoch: 13/100, batch 328/375 completed (loss: 0.4042:  88%|████████▊ | 164/187 [16:35<02:19,  6.08s/it]
Training Epoch: 13/100, batch 328/375 completed (loss: 0.0022:  88%|████████▊ | 164/187 [16:35<02:19,  6.08s/it]
Training Epoch: 13/100, batch 328/375 completed (loss: 0.1064:  88%|████████▊ | 164/187 [16:35<02:19,  6.08s/it]
Training Epoch: 13/100, batch 328/375 completed (loss: 0.0022:  88%|████████▊ | 165/187 [16:38<02:13,  6.05s/it]
Training Epoch: 13/100, batch 328/375 completed (loss: 0.0067:  88%|████████▊ | 165/187 [16:38<02:13,  6.06s/it]
Training Epoch: 13/100, batch 328/375 completed (loss: 0.4042:  88%|████████▊ | 165/187 [16:38<02:13,  6.06s/it]
Training Epoch: 13/100, batch 328/375 completed (loss: 0.1064:  88%|████████▊ | 165/187 [16:38<02:13,  6.05s/it]
Training Epoch: 13/100, batch 329/375 completed (loss: 0.0155:  88%|████████▊ | 165/187 [16:38<02:13,  6.06s/it]
Training Epoch: 13/100, batch 329/375 completed (loss: 0.0071:  88%|████████▊ | 165/187 [16:38<02:13,  6.06s/it]
Training Epoch: 13/100, batch 329/375 completed (loss: 0.0014:  88%|████████▊ | 165/187 [16:38<02:13,  6.05s/it]
Training Epoch: 13/100, batch 329/375 completed (loss: 0.2438:  88%|████████▊ | 165/187 [16:38<02:13,  6.05s/it]
Training Epoch: 13/100, batch 330/375 completed (loss: 0.0014:  88%|████████▊ | 165/187 [16:41<02:13,  6.05s/it]
Training Epoch: 13/100, batch 330/375 completed (loss: 0.0523:  88%|████████▊ | 165/187 [16:41<02:13,  6.06s/it]
Training Epoch: 13/100, batch 330/375 completed (loss: 0.0026:  88%|████████▊ | 165/187 [16:41<02:13,  6.06s/it]
Training Epoch: 13/100, batch 330/375 completed (loss: 0.0119:  88%|████████▊ | 165/187 [16:41<02:13,  6.05s/it]
Training Epoch: 13/100, batch 330/375 completed (loss: 0.0026:  89%|████████▉ | 166/187 [16:44<02:08,  6.13s/it]
Training Epoch: 13/100, batch 330/375 completed (loss: 0.0523:  89%|████████▉ | 166/187 [16:44<02:08,  6.13s/it]
Training Epoch: 13/100, batch 330/375 completed (loss: 0.0014:  89%|████████▉ | 166/187 [16:44<02:08,  6.13s/it]
Training Epoch: 13/100, batch 330/375 completed (loss: 0.0119:  89%|████████▉ | 166/187 [16:44<02:08,  6.13s/it]
Training Epoch: 13/100, batch 331/375 completed (loss: 0.2166:  89%|████████▉ | 166/187 [16:44<02:08,  6.13s/it]
Training Epoch: 13/100, batch 331/375 completed (loss: 0.008:  89%|████████▉ | 166/187 [16:44<02:08,  6.13s/it] 
Training Epoch: 13/100, batch 331/375 completed (loss: 0.0115:  89%|████████▉ | 166/187 [16:44<02:08,  6.13s/it]
Training Epoch: 13/100, batch 331/375 completed (loss: 0.0035:  89%|████████▉ | 166/187 [16:44<02:08,  6.13s/it]
Training Epoch: 13/100, batch 332/375 completed (loss: 0.0014:  89%|████████▉ | 166/187 [16:47<02:08,  6.13s/it]
Training Epoch: 13/100, batch 332/375 completed (loss: 0.0006:  89%|████████▉ | 166/187 [16:47<02:08,  6.13s/it]
Training Epoch: 13/100, batch 332/375 completed (loss: 0.0014:  89%|████████▉ | 166/187 [16:47<02:08,  6.13s/it]
Training Epoch: 13/100, batch 332/375 completed (loss: 0.0532:  89%|████████▉ | 166/187 [16:47<02:08,  6.13s/it]
Training Epoch: 13/100, batch 332/375 completed (loss: 0.0014:  89%|████████▉ | 167/187 [16:50<02:00,  6.04s/it]
Training Epoch: 13/100, batch 332/375 completed (loss: 0.0006:  89%|████████▉ | 167/187 [16:50<02:00,  6.04s/it]
Training Epoch: 13/100, batch 332/375 completed (loss: 0.0014:  89%|████████▉ | 167/187 [16:50<02:00,  6.04s/it]
Training Epoch: 13/100, batch 332/375 completed (loss: 0.0532:  89%|████████▉ | 167/187 [16:50<02:00,  6.03s/it]
Training Epoch: 13/100, batch 333/375 completed (loss: 0.011:  89%|████████▉ | 167/187 [16:50<02:00,  6.04s/it] 
Training Epoch: 13/100, batch 333/375 completed (loss: 0.0207:  89%|████████▉ | 167/187 [16:50<02:00,  6.04s/it]
Training Epoch: 13/100, batch 333/375 completed (loss: 0.0179:  89%|████████▉ | 167/187 [16:50<02:00,  6.04s/it]
Training Epoch: 13/100, batch 333/375 completed (loss: 0.0026:  89%|████████▉ | 167/187 [16:50<02:00,  6.03s/it]
Training Epoch: 13/100, batch 334/375 completed (loss: 0.1634:  89%|████████▉ | 167/187 [16:53<02:00,  6.04s/it]
Training Epoch: 13/100, batch 334/375 completed (loss: 0.0074:  89%|████████▉ | 167/187 [16:53<02:00,  6.04s/it]
Training Epoch: 13/100, batch 334/375 completed (loss: 0.1817:  89%|████████▉ | 167/187 [16:53<02:00,  6.04s/it]
Training Epoch: 13/100, batch 334/375 completed (loss: 0.0079:  89%|████████▉ | 167/187 [16:53<02:00,  6.03s/it]
Training Epoch: 13/100, batch 334/375 completed (loss: 0.0074:  90%|████████▉ | 168/187 [16:56<01:56,  6.11s/it]
Training Epoch: 13/100, batch 334/375 completed (loss: 0.1817:  90%|████████▉ | 168/187 [16:56<01:56,  6.11s/it]
Training Epoch: 13/100, batch 334/375 completed (loss: 0.1634:  90%|████████▉ | 168/187 [16:56<01:56,  6.11s/it]
Training Epoch: 13/100, batch 334/375 completed (loss: 0.0079:  90%|████████▉ | 168/187 [16:56<01:56,  6.11s/it]
Training Epoch: 13/100, batch 335/375 completed (loss: 0.3191:  90%|████████▉ | 168/187 [16:56<01:56,  6.11s/it]
Training Epoch: 13/100, batch 335/375 completed (loss: 0.0091:  90%|████████▉ | 168/187 [16:56<01:56,  6.11s/it]
Training Epoch: 13/100, batch 335/375 completed (loss: 0.2982:  90%|████████▉ | 168/187 [16:56<01:56,  6.11s/it]
Training Epoch: 13/100, batch 335/375 completed (loss: 0.006:  90%|████████▉ | 168/187 [16:56<01:56,  6.11s/it] 
Training Epoch: 13/100, batch 336/375 completed (loss: 0.0183:  90%|████████▉ | 168/187 [16:59<01:56,  6.11s/it]
Training Epoch: 13/100, batch 336/375 completed (loss: 0.0616:  90%|████████▉ | 168/187 [16:59<01:56,  6.11s/it]
Training Epoch: 13/100, batch 336/375 completed (loss: 0.0039:  90%|████████▉ | 168/187 [16:59<01:56,  6.11s/it]
Training Epoch: 13/100, batch 336/375 completed (loss: 0.0665:  90%|████████▉ | 168/187 [16:59<01:56,  6.11s/it]
Training Epoch: 13/100, batch 336/375 completed (loss: 0.0039:  90%|█████████ | 169/187 [17:02<01:49,  6.09s/it]
Training Epoch: 13/100, batch 336/375 completed (loss: 0.0665:  90%|█████████ | 169/187 [17:02<01:49,  6.09s/it]
Training Epoch: 13/100, batch 336/375 completed (loss: 0.0616:  90%|█████████ | 169/187 [17:02<01:49,  6.09s/it]
Training Epoch: 13/100, batch 336/375 completed (loss: 0.0183:  90%|█████████ | 169/187 [17:02<01:49,  6.09s/it]
Training Epoch: 13/100, batch 337/375 completed (loss: 0.3879:  90%|█████████ | 169/187 [17:02<01:49,  6.09s/it]
Training Epoch: 13/100, batch 337/375 completed (loss: 0.0226:  90%|█████████ | 169/187 [17:02<01:49,  6.09s/it]
Training Epoch: 13/100, batch 337/375 completed (loss: 0.0737:  90%|█████████ | 169/187 [17:02<01:49,  6.09s/it]
Training Epoch: 13/100, batch 337/375 completed (loss: 0.1343:  90%|█████████ | 169/187 [17:02<01:49,  6.09s/it]
Training Epoch: 13/100, batch 338/375 completed (loss: 0.0196:  90%|█████████ | 169/187 [17:05<01:49,  6.09s/it]
Training Epoch: 13/100, batch 338/375 completed (loss: 0.0171:  90%|█████████ | 169/187 [17:05<01:49,  6.09s/it]
Training Epoch: 13/100, batch 338/375 completed (loss: 0.0195:  90%|█████████ | 169/187 [17:05<01:49,  6.09s/it]
Training Epoch: 13/100, batch 338/375 completed (loss: 0.096:  90%|█████████ | 169/187 [17:05<01:49,  6.09s/it] 
Training Epoch: 13/100, batch 338/375 completed (loss: 0.0195:  91%|█████████ | 170/187 [17:08<01:42,  6.01s/it]
Training Epoch: 13/100, batch 338/375 completed (loss: 0.0196:  91%|█████████ | 170/187 [17:08<01:42,  6.01s/it]
Training Epoch: 13/100, batch 338/375 completed (loss: 0.096:  91%|█████████ | 170/187 [17:08<01:42,  6.01s/it]
Training Epoch: 13/100, batch 338/375 completed (loss: 0.0171:  91%|█████████ | 170/187 [17:08<01:42,  6.01s/it]
Training Epoch: 13/100, batch 339/375 completed (loss: 0.0804:  91%|█████████ | 170/187 [17:08<01:42,  6.01s/it]
Training Epoch: 13/100, batch 339/375 completed (loss: 0.0025:  91%|█████████ | 170/187 [17:08<01:42,  6.01s/it]
Training Epoch: 13/100, batch 339/375 completed (loss: 0.0378:  91%|█████████ | 170/187 [17:08<01:42,  6.01s/it]
Training Epoch: 13/100, batch 339/375 completed (loss: 0.1146:  91%|█████████ | 170/187 [17:08<01:42,  6.01s/it]
Training Epoch: 13/100, batch 340/375 completed (loss: 0.0116:  91%|█████████ | 170/187 [17:12<01:42,  6.01s/it]
Training Epoch: 13/100, batch 340/375 completed (loss: 0.0175:  91%|█████████ | 170/187 [17:12<01:42,  6.01s/it]
Training Epoch: 13/100, batch 340/375 completed (loss: 0.0038:  91%|█████████ | 170/187 [17:12<01:42,  6.01s/it]
Training Epoch: 13/100, batch 340/375 completed (loss: 0.0243:  91%|█████████ | 170/187 [17:12<01:42,  6.01s/it]
Training Epoch: 13/100, batch 340/375 completed (loss: 0.0038:  91%|█████████▏| 171/187 [17:15<01:38,  6.14s/it]
Training Epoch: 13/100, batch 340/375 completed (loss: 0.0116:  91%|█████████▏| 171/187 [17:15<01:38,  6.14s/it]
Training Epoch: 13/100, batch 340/375 completed (loss: 0.0243:  91%|█████████▏| 171/187 [17:15<01:38,  6.14s/it]
Training Epoch: 13/100, batch 340/375 completed (loss: 0.0175:  91%|█████████▏| 171/187 [17:15<01:38,  6.14s/it]
Training Epoch: 13/100, batch 341/375 completed (loss: 0.0086:  91%|█████████▏| 171/187 [17:15<01:38,  6.14s/it]
Training Epoch: 13/100, batch 341/375 completed (loss: 0.008:  91%|█████████▏| 171/187 [17:15<01:38,  6.14s/it] 
Training Epoch: 13/100, batch 341/375 completed (loss: 0.0011:  91%|█████████▏| 171/187 [17:15<01:38,  6.14s/it]
Training Epoch: 13/100, batch 341/375 completed (loss: 0.01:  91%|█████████▏| 171/187 [17:15<01:38,  6.14s/it]  
Training Epoch: 13/100, batch 342/375 completed (loss: 0.0306:  91%|█████████▏| 171/187 [17:18<01:38,  6.14s/it]
Training Epoch: 13/100, batch 342/375 completed (loss: 0.1323:  91%|█████████▏| 171/187 [17:18<01:38,  6.14s/it]
Training Epoch: 13/100, batch 342/375 completed (loss: 0.0279:  91%|█████████▏| 171/187 [17:18<01:38,  6.14s/it]
Training Epoch: 13/100, batch 342/375 completed (loss: 0.0626:  91%|█████████▏| 171/187 [17:18<01:38,  6.14s/it]
Training Epoch: 13/100, batch 342/375 completed (loss: 0.1323:  92%|█████████▏| 172/187 [17:20<01:30,  6.05s/it]
Training Epoch: 13/100, batch 342/375 completed (loss: 0.0279:  92%|█████████▏| 172/187 [17:20<01:30,  6.05s/it]
Training Epoch: 13/100, batch 342/375 completed (loss: 0.0306:  92%|█████████▏| 172/187 [17:20<01:30,  6.05s/it]
Training Epoch: 13/100, batch 342/375 completed (loss: 0.0626:  92%|█████████▏| 172/187 [17:20<01:30,  6.05s/it]
Training Epoch: 13/100, batch 343/375 completed (loss: 0.0214:  92%|█████████▏| 172/187 [17:20<01:30,  6.05s/it]
Training Epoch: 13/100, batch 343/375 completed (loss: 0.0535:  92%|█████████▏| 172/187 [17:20<01:30,  6.05s/it]
Training Epoch: 13/100, batch 343/375 completed (loss: 0.012:  92%|█████████▏| 172/187 [17:20<01:30,  6.05s/it] 
Training Epoch: 13/100, batch 343/375 completed (loss: 0.013:  92%|█████████▏| 172/187 [17:21<01:30,  6.05s/it] 
Training Epoch: 13/100, batch 344/375 completed (loss: 0.0591:  92%|█████████▏| 172/187 [17:23<01:30,  6.05s/it]
Training Epoch: 13/100, batch 344/375 completed (loss: 0.0233:  92%|█████████▏| 172/187 [17:23<01:30,  6.05s/it]
Training Epoch: 13/100, batch 344/375 completed (loss: 0.2341:  92%|█████████▏| 172/187 [17:23<01:30,  6.05s/it]
Training Epoch: 13/100, batch 344/375 completed (loss: 0.0569:  92%|█████████▏| 172/187 [17:23<01:30,  6.05s/it]
Training Epoch: 13/100, batch 344/375 completed (loss: 0.2341:  93%|█████████▎| 173/187 [17:26<01:24,  6.04s/it]
Training Epoch: 13/100, batch 344/375 completed (loss: 0.0569:  93%|█████████▎| 173/187 [17:26<01:24,  6.03s/it]
Training Epoch: 13/100, batch 344/375 completed (loss: 0.0591:  93%|█████████▎| 173/187 [17:26<01:24,  6.04s/it]
Training Epoch: 13/100, batch 344/375 completed (loss: 0.0233:  93%|█████████▎| 173/187 [17:26<01:24,  6.04s/it]
Training Epoch: 13/100, batch 345/375 completed (loss: 0.1655:  93%|█████████▎| 173/187 [17:26<01:24,  6.04s/it]
Training Epoch: 13/100, batch 345/375 completed (loss: 0.0209:  93%|█████████▎| 173/187 [17:26<01:24,  6.04s/it]
Training Epoch: 13/100, batch 345/375 completed (loss: 0.0271:  93%|█████████▎| 173/187 [17:26<01:24,  6.04s/it]
Training Epoch: 13/100, batch 345/375 completed (loss: 0.0147:  93%|█████████▎| 173/187 [17:27<01:24,  6.03s/it]
Training Epoch: 13/100, batch 346/375 completed (loss: 0.0469:  93%|█████████▎| 173/187 [17:30<01:24,  6.04s/it]
Training Epoch: 13/100, batch 346/375 completed (loss: 0.0073:  93%|█████████▎| 173/187 [17:30<01:24,  6.04s/it]
Training Epoch: 13/100, batch 346/375 completed (loss: 0.0003:  93%|█████████▎| 173/187 [17:30<01:24,  6.04s/it]
Training Epoch: 13/100, batch 346/375 completed (loss: 0.0181:  93%|█████████▎| 173/187 [17:30<01:24,  6.03s/it]
Training Epoch: 13/100, batch 346/375 completed (loss: 0.0073:  93%|█████████▎| 174/187 [17:33<01:19,  6.12s/it]
Training Epoch: 13/100, batch 346/375 completed (loss: 0.0003:  93%|█████████▎| 174/187 [17:33<01:19,  6.12s/it]
Training Epoch: 13/100, batch 346/375 completed (loss: 0.0469:  93%|█████████▎| 174/187 [17:33<01:19,  6.12s/it]
Training Epoch: 13/100, batch 346/375 completed (loss: 0.0181:  93%|█████████▎| 174/187 [17:33<01:19,  6.12s/it]
Training Epoch: 13/100, batch 347/375 completed (loss: 0.1584:  93%|█████████▎| 174/187 [17:33<01:19,  6.12s/it]
Training Epoch: 13/100, batch 347/375 completed (loss: 0.0233:  93%|█████████▎| 174/187 [17:33<01:19,  6.12s/it]
Training Epoch: 13/100, batch 347/375 completed (loss: 0.0681:  93%|█████████▎| 174/187 [17:33<01:19,  6.12s/it]
Training Epoch: 13/100, batch 347/375 completed (loss: 0.014:  93%|█████████▎| 174/187 [17:33<01:19,  6.12s/it] 
Training Epoch: 13/100, batch 348/375 completed (loss: 0.0567:  93%|█████████▎| 174/187 [17:36<01:19,  6.12s/it]
Training Epoch: 13/100, batch 348/375 completed (loss: 0.0015:  93%|█████████▎| 174/187 [17:36<01:19,  6.12s/it]
Training Epoch: 13/100, batch 348/375 completed (loss: 0.0036:  93%|█████████▎| 174/187 [17:36<01:19,  6.12s/it]
Training Epoch: 13/100, batch 348/375 completed (loss: 0.0883:  93%|█████████▎| 174/187 [17:36<01:19,  6.12s/it]
Training Epoch: 13/100, batch 348/375 completed (loss: 0.0036:  94%|█████████▎| 175/187 [17:39<01:12,  6.04s/it]
Training Epoch: 13/100, batch 348/375 completed (loss: 0.0015:  94%|█████████▎| 175/187 [17:39<01:12,  6.04s/it]
Training Epoch: 13/100, batch 348/375 completed (loss: 0.0567:  94%|█████████▎| 175/187 [17:39<01:12,  6.04s/it]
Training Epoch: 13/100, batch 348/375 completed (loss: 0.0883:  94%|█████████▎| 175/187 [17:39<01:12,  6.04s/it]
Training Epoch: 13/100, batch 349/375 completed (loss: 0.0469:  94%|█████████▎| 175/187 [17:39<01:12,  6.04s/it]
Training Epoch: 13/100, batch 349/375 completed (loss: 0.1459:  94%|█████████▎| 175/187 [17:39<01:12,  6.04s/it]
Training Epoch: 13/100, batch 349/375 completed (loss: 0.0083:  94%|█████████▎| 175/187 [17:39<01:12,  6.04s/it]
Training Epoch: 13/100, batch 349/375 completed (loss: 0.1765:  94%|█████████▎| 175/187 [17:39<01:12,  6.04s/it]
Training Epoch: 13/100, batch 350/375 completed (loss: 0.0153:  94%|█████████▎| 175/187 [17:42<01:12,  6.04s/it]
Training Epoch: 13/100, batch 350/375 completed (loss: 0.0119:  94%|█████████▎| 175/187 [17:42<01:12,  6.04s/it]
Training Epoch: 13/100, batch 350/375 completed (loss: 0.026:  94%|█████████▎| 175/187 [17:42<01:12,  6.04s/it] 
Training Epoch: 13/100, batch 350/375 completed (loss: 0.0524:  94%|█████████▎| 175/187 [17:42<01:12,  6.04s/it]
Training Epoch: 13/100, batch 350/375 completed (loss: 0.0119:  94%|█████████▍| 176/187 [17:45<01:07,  6.11s/it]
Training Epoch: 13/100, batch 350/375 completed (loss: 0.026:  94%|█████████▍| 176/187 [17:45<01:07,  6.11s/it]
Training Epoch: 13/100, batch 350/375 completed (loss: 0.0153:  94%|█████████▍| 176/187 [17:45<01:07,  6.11s/it]
Training Epoch: 13/100, batch 350/375 completed (loss: 0.0524:  94%|█████████▍| 176/187 [17:45<01:07,  6.11s/it]
Training Epoch: 13/100, batch 351/375 completed (loss: 0.0228:  94%|█████████▍| 176/187 [17:45<01:07,  6.11s/it]
Training Epoch: 13/100, batch 351/375 completed (loss: 0.0256:  94%|█████████▍| 176/187 [17:45<01:07,  6.11s/it]
Training Epoch: 13/100, batch 351/375 completed (loss: 0.0302:  94%|█████████▍| 176/187 [17:45<01:07,  6.11s/it]
Training Epoch: 13/100, batch 351/375 completed (loss: 0.0161:  94%|█████████▍| 176/187 [17:45<01:07,  6.11s/it]
Training Epoch: 13/100, batch 352/375 completed (loss: 0.1457:  94%|█████████▍| 176/187 [17:48<01:07,  6.11s/it]
Training Epoch: 13/100, batch 352/375 completed (loss: 0.0114:  94%|█████████▍| 176/187 [17:48<01:07,  6.11s/it]
Training Epoch: 13/100, batch 352/375 completed (loss: 0.0028:  94%|█████████▍| 176/187 [17:48<01:07,  6.11s/it]
Training Epoch: 13/100, batch 352/375 completed (loss: 0.1104:  94%|█████████▍| 176/187 [17:48<01:07,  6.11s/it]
Training Epoch: 13/100, batch 352/375 completed (loss: 0.1104:  95%|█████████▍| 177/187 [17:51<01:00,  6.07s/it]
Training Epoch: 13/100, batch 352/375 completed (loss: 0.0114:  95%|█████████▍| 177/187 [17:51<01:00,  6.07s/it]
Training Epoch: 13/100, batch 352/375 completed (loss: 0.1457:  95%|█████████▍| 177/187 [17:51<01:00,  6.07s/it]
Training Epoch: 13/100, batch 352/375 completed (loss: 0.0028:  95%|█████████▍| 177/187 [17:51<01:00,  6.07s/it]
Training Epoch: 13/100, batch 353/375 completed (loss: 0.074:  95%|█████████▍| 177/187 [17:51<01:00,  6.07s/it] 
Training Epoch: 13/100, batch 353/375 completed (loss: 0.2124:  95%|█████████▍| 177/187 [17:51<01:00,  6.07s/it]
Training Epoch: 13/100, batch 353/375 completed (loss: 0.0435:  95%|█████████▍| 177/187 [17:51<01:00,  6.07s/it]
Training Epoch: 13/100, batch 353/375 completed (loss: 0.0583:  95%|█████████▍| 177/187 [17:51<01:00,  6.07s/it]
Training Epoch: 13/100, batch 354/375 completed (loss: 0.0332:  95%|█████████▍| 177/187 [17:54<01:00,  6.07s/it]
Training Epoch: 13/100, batch 354/375 completed (loss: 0.027:  95%|█████████▍| 177/187 [17:54<01:00,  6.07s/it] 
Training Epoch: 13/100, batch 354/375 completed (loss: 0.0025:  95%|█████████▍| 177/187 [17:54<01:00,  6.07s/it]
Training Epoch: 13/100, batch 354/375 completed (loss: 0.0008:  95%|█████████▍| 177/187 [17:54<01:00,  6.07s/it]
Training Epoch: 13/100, batch 354/375 completed (loss: 0.0025:  95%|█████████▌| 178/187 [17:57<00:53,  6.00s/it]
Training Epoch: 13/100, batch 354/375 completed (loss: 0.027:  95%|█████████▌| 178/187 [17:57<00:53,  6.00s/it]
Training Epoch: 13/100, batch 354/375 completed (loss: 0.0332:  95%|█████████▌| 178/187 [17:57<00:53,  6.00s/it]
Training Epoch: 13/100, batch 354/375 completed (loss: 0.0008:  95%|█████████▌| 178/187 [17:57<00:53,  6.00s/it]
Training Epoch: 13/100, batch 355/375 completed (loss: 0.025:  95%|█████████▌| 178/187 [17:57<00:53,  6.00s/it] 
Training Epoch: 13/100, batch 355/375 completed (loss: 0.0175:  95%|█████████▌| 178/187 [17:57<00:53,  6.00s/it]
Training Epoch: 13/100, batch 355/375 completed (loss: 0.1851:  95%|█████████▌| 178/187 [17:57<00:53,  6.00s/it]
Training Epoch: 13/100, batch 355/375 completed (loss: 0.0224:  95%|█████████▌| 178/187 [17:57<00:53,  6.00s/it]
Training Epoch: 13/100, batch 356/375 completed (loss: 0.0173:  95%|█████████▌| 178/187 [18:00<00:53,  6.00s/it]
Training Epoch: 13/100, batch 356/375 completed (loss: 0.028:  95%|█████████▌| 178/187 [18:00<00:53,  6.00s/it]
Training Epoch: 13/100, batch 356/375 completed (loss: 0.0686:  95%|█████████▌| 178/187 [18:00<00:53,  6.00s/it]
Training Epoch: 13/100, batch 356/375 completed (loss: 0.0105:  95%|█████████▌| 178/187 [18:00<00:53,  6.00s/it]
Training Epoch: 13/100, batch 356/375 completed (loss: 0.0173:  96%|█████████▌| 179/187 [18:03<00:48,  6.08s/it]
Training Epoch: 13/100, batch 356/375 completed (loss: 0.028:  96%|█████████▌| 179/187 [18:03<00:48,  6.08s/it]
Training Epoch: 13/100, batch 356/375 completed (loss: 0.0686:  96%|█████████▌| 179/187 [18:03<00:48,  6.08s/it]
Training Epoch: 13/100, batch 356/375 completed (loss: 0.0105:  96%|█████████▌| 179/187 [18:03<00:48,  6.08s/it]
Training Epoch: 13/100, batch 357/375 completed (loss: 0.0313:  96%|█████████▌| 179/187 [18:03<00:48,  6.08s/it]
Training Epoch: 13/100, batch 357/375 completed (loss: 0.0354:  96%|█████████▌| 179/187 [18:03<00:48,  6.08s/it]
Training Epoch: 13/100, batch 357/375 completed (loss: 0.0435:  96%|█████████▌| 179/187 [18:03<00:48,  6.08s/it]
Training Epoch: 13/100, batch 357/375 completed (loss: 0.0036:  96%|█████████▌| 179/187 [18:03<00:48,  6.08s/it]
Training Epoch: 13/100, batch 358/375 completed (loss: 0.0086:  96%|█████████▌| 179/187 [18:06<00:48,  6.08s/it]
Training Epoch: 13/100, batch 358/375 completed (loss: 0.0067:  96%|█████████▌| 179/187 [18:06<00:48,  6.08s/it]
Training Epoch: 13/100, batch 358/375 completed (loss: 0.0174:  96%|█████████▌| 179/187 [18:06<00:48,  6.08s/it]
Training Epoch: 13/100, batch 358/375 completed (loss: 0.0093:  96%|█████████▌| 179/187 [18:06<00:48,  6.08s/it]
Training Epoch: 13/100, batch 358/375 completed (loss: 0.0174:  96%|█████████▋| 180/187 [18:09<00:42,  6.06s/it]
Training Epoch: 13/100, batch 358/375 completed (loss: 0.0093:  96%|█████████▋| 180/187 [18:09<00:42,  6.06s/it]
Training Epoch: 13/100, batch 358/375 completed (loss: 0.0067:  96%|█████████▋| 180/187 [18:09<00:42,  6.06s/it]
Training Epoch: 13/100, batch 358/375 completed (loss: 0.0086:  96%|█████████▋| 180/187 [18:09<00:42,  6.07s/it]
Training Epoch: 13/100, batch 359/375 completed (loss: 0.0029:  96%|█████████▋| 180/187 [18:09<00:42,  6.06s/it]
Training Epoch: 13/100, batch 359/375 completed (loss: 0.0043:  96%|█████████▋| 180/187 [18:09<00:42,  6.07s/it]
Training Epoch: 13/100, batch 359/375 completed (loss: 0.0363:  96%|█████████▋| 180/187 [18:09<00:42,  6.06s/it]
Training Epoch: 13/100, batch 359/375 completed (loss: 0.0092:  96%|█████████▋| 180/187 [18:09<00:42,  6.06s/it]
Training Epoch: 13/100, batch 360/375 completed (loss: 0.0095:  96%|█████████▋| 180/187 [18:12<00:42,  6.06s/it]
Training Epoch: 13/100, batch 360/375 completed (loss: 0.0212:  96%|█████████▋| 180/187 [18:12<00:42,  6.06s/it]
Training Epoch: 13/100, batch 360/375 completed (loss: 0.0268:  96%|█████████▋| 180/187 [18:12<00:42,  6.07s/it]
Training Epoch: 13/100, batch 360/375 completed (loss: 0.0121:  96%|█████████▋| 180/187 [18:12<00:42,  6.06s/it]
Training Epoch: 13/100, batch 360/375 completed (loss: 0.0121:  97%|█████████▋| 181/187 [18:15<00:36,  6.13s/it]
Training Epoch: 13/100, batch 360/375 completed (loss: 0.0212:  97%|█████████▋| 181/187 [18:15<00:36,  6.13s/it]
Training Epoch: 13/100, batch 360/375 completed (loss: 0.0095:  97%|█████████▋| 181/187 [18:15<00:36,  6.14s/it]
Training Epoch: 13/100, batch 360/375 completed (loss: 0.0268:  97%|█████████▋| 181/187 [18:15<00:36,  6.14s/it]
Training Epoch: 13/100, batch 361/375 completed (loss: 0.0688:  97%|█████████▋| 181/187 [18:15<00:36,  6.14s/it]
Training Epoch: 13/100, batch 361/375 completed (loss: 0.101:  97%|█████████▋| 181/187 [18:15<00:36,  6.13s/it] 
Training Epoch: 13/100, batch 361/375 completed (loss: 0.0569:  97%|█████████▋| 181/187 [18:15<00:36,  6.14s/it]
Training Epoch: 13/100, batch 361/375 completed (loss: 0.059:  97%|█████████▋| 181/187 [18:15<00:36,  6.13s/it] 
Training Epoch: 13/100, batch 362/375 completed (loss: 0.056:  97%|█████████▋| 181/187 [18:18<00:36,  6.13s/it]
Training Epoch: 13/100, batch 362/375 completed (loss: 0.001:  97%|█████████▋| 181/187 [18:18<00:36,  6.14s/it] 
Training Epoch: 13/100, batch 362/375 completed (loss: 0.0059:  97%|█████████▋| 181/187 [18:18<00:36,  6.14s/it]
Training Epoch: 13/100, batch 362/375 completed (loss: 0.0084:  97%|█████████▋| 181/187 [18:18<00:36,  6.13s/it]
Training Epoch: 13/100, batch 362/375 completed (loss: 0.056:  97%|█████████▋| 182/187 [18:21<00:30,  6.10s/it]
Training Epoch: 13/100, batch 362/375 completed (loss: 0.0059:  97%|█████████▋| 182/187 [18:21<00:30,  6.10s/it]
Training Epoch: 13/100, batch 362/375 completed (loss: 0.001:  97%|█████████▋| 182/187 [18:21<00:30,  6.10s/it]
Training Epoch: 13/100, batch 362/375 completed (loss: 0.0084:  97%|█████████▋| 182/187 [18:21<00:30,  6.10s/it]
Training Epoch: 13/100, batch 363/375 completed (loss: 0.0173:  97%|█████████▋| 182/187 [18:21<00:30,  6.10s/it]
Training Epoch: 13/100, batch 363/375 completed (loss: 0.0069:  97%|█████████▋| 182/187 [18:21<00:30,  6.10s/it]
Training Epoch: 13/100, batch 363/375 completed (loss: 0.013:  97%|█████████▋| 182/187 [18:21<00:30,  6.10s/it]
Training Epoch: 13/100, batch 363/375 completed (loss: 0.0191:  97%|█████████▋| 182/187 [18:21<00:30,  6.10s/it]
Training Epoch: 13/100, batch 364/375 completed (loss: 0.2681:  97%|█████████▋| 182/187 [18:24<00:30,  6.10s/it]
Training Epoch: 13/100, batch 364/375 completed (loss: 0.0201:  97%|█████████▋| 182/187 [18:24<00:30,  6.10s/it]
Training Epoch: 13/100, batch 364/375 completed (loss: 0.0002:  97%|█████████▋| 182/187 [18:24<00:30,  6.10s/it]
Training Epoch: 13/100, batch 364/375 completed (loss: 0.0076:  97%|█████████▋| 182/187 [18:24<00:30,  6.10s/it]
Training Epoch: 13/100, batch 364/375 completed (loss: 0.0201:  98%|█████████▊| 183/187 [18:27<00:24,  6.02s/it]
Training Epoch: 13/100, batch 364/375 completed (loss: 0.0076:  98%|█████████▊| 183/187 [18:27<00:24,  6.03s/it]
Training Epoch: 13/100, batch 364/375 completed (loss: 0.2681:  98%|█████████▊| 183/187 [18:27<00:24,  6.03s/it]
Training Epoch: 13/100, batch 364/375 completed (loss: 0.0002:  98%|█████████▊| 183/187 [18:27<00:24,  6.03s/it]
Training Epoch: 13/100, batch 365/375 completed (loss: 0.067:  98%|█████████▊| 183/187 [18:27<00:24,  6.02s/it] 
Training Epoch: 13/100, batch 365/375 completed (loss: 0.0135:  98%|█████████▊| 183/187 [18:27<00:24,  6.03s/it]
Training Epoch: 13/100, batch 365/375 completed (loss: 0.0559:  98%|█████████▊| 183/187 [18:27<00:24,  6.03s/it]
Training Epoch: 13/100, batch 365/375 completed (loss: 0.017:  98%|█████████▊| 183/187 [18:27<00:24,  6.03s/it] 
Training Epoch: 13/100, batch 366/375 completed (loss: 0.0092:  98%|█████████▊| 183/187 [18:30<00:24,  6.02s/it]
Training Epoch: 13/100, batch 366/375 completed (loss: 0.055:  98%|█████████▊| 183/187 [18:30<00:24,  6.03s/it] 
Training Epoch: 13/100, batch 366/375 completed (loss: 0.1493:  98%|█████████▊| 183/187 [18:30<00:24,  6.03s/it]
Training Epoch: 13/100, batch 366/375 completed (loss: 0.0054:  98%|█████████▊| 183/187 [18:31<00:24,  6.03s/it]
Training Epoch: 13/100, batch 366/375 completed (loss: 0.0054:  98%|█████████▊| 184/187 [18:34<00:18,  6.12s/it]
Training Epoch: 13/100, batch 366/375 completed (loss: 0.1493:  98%|█████████▊| 184/187 [18:34<00:18,  6.13s/it]
Training Epoch: 13/100, batch 366/375 completed (loss: 0.0092:  98%|█████████▊| 184/187 [18:34<00:18,  6.13s/it]
Training Epoch: 13/100, batch 366/375 completed (loss: 0.055:  98%|█████████▊| 184/187 [18:34<00:18,  6.13s/it]
Training Epoch: 13/100, batch 367/375 completed (loss: 0.0374:  98%|█████████▊| 184/187 [18:34<00:18,  6.13s/it]
Training Epoch: 13/100, batch 367/375 completed (loss: 0.06:  98%|█████████▊| 184/187 [18:34<00:18,  6.13s/it]  
Training Epoch: 13/100, batch 367/375 completed (loss: 0.0071:  98%|█████████▊| 184/187 [18:34<00:18,  6.13s/it]
Training Epoch: 13/100, batch 367/375 completed (loss: 0.0033:  98%|█████████▊| 184/187 [18:34<00:18,  6.12s/it]
Training Epoch: 13/100, batch 368/375 completed (loss: 0.0321:  98%|█████████▊| 184/187 [18:37<00:18,  6.13s/it]
Training Epoch: 13/100, batch 368/375 completed (loss: 0.0029:  98%|█████████▊| 184/187 [18:37<00:18,  6.13s/it]
Training Epoch: 13/100, batch 368/375 completed (loss: 0.0169:  98%|█████████▊| 184/187 [18:37<00:18,  6.13s/it]
Training Epoch: 13/100, batch 368/375 completed (loss: 0.0034:  98%|█████████▊| 184/187 [18:37<00:18,  6.12s/it]
Training Epoch: 13/100, batch 368/375 completed (loss: 0.0169:  99%|█████████▉| 185/187 [18:40<00:12,  6.09s/it]
Training Epoch: 13/100, batch 368/375 completed (loss: 0.0034:  99%|█████████▉| 185/187 [18:40<00:12,  6.09s/it]
Training Epoch: 13/100, batch 368/375 completed (loss: 0.0029:  99%|█████████▉| 185/187 [18:40<00:12,  6.10s/it]
Training Epoch: 13/100, batch 368/375 completed (loss: 0.0321:  99%|█████████▉| 185/187 [18:40<00:12,  6.10s/it]
Training Epoch: 13/100, batch 369/375 completed (loss: 0.2615:  99%|█████████▉| 185/187 [18:40<00:12,  6.09s/it]
Training Epoch: 13/100, batch 369/375 completed (loss: 0.0602:  99%|█████████▉| 185/187 [18:40<00:12,  6.10s/it]
Training Epoch: 13/100, batch 369/375 completed (loss: 0.0032:  99%|█████████▉| 185/187 [18:40<00:12,  6.10s/it]
Training Epoch: 13/100, batch 369/375 completed (loss: 0.004:  99%|█████████▉| 185/187 [18:40<00:12,  6.09s/it] 
Training Epoch: 13/100, batch 370/375 completed (loss: 0.0442:  99%|█████████▉| 185/187 [18:42<00:12,  6.09s/it]
Training Epoch: 13/100, batch 370/375 completed (loss: 0.0852:  99%|█████████▉| 185/187 [18:42<00:12,  6.10s/it]
Training Epoch: 13/100, batch 370/375 completed (loss: 0.165:  99%|█████████▉| 185/187 [18:43<00:12,  6.10s/it] 
Training Epoch: 13/100, batch 370/375 completed (loss: 0.0268:  99%|█████████▉| 185/187 [18:43<00:12,  6.09s/it]
Training Epoch: 13/100, batch 370/375 completed (loss: 0.165:  99%|█████████▉| 186/187 [18:46<00:06,  6.10s/it]
Training Epoch: 13/100, batch 370/375 completed (loss: 0.0442:  99%|█████████▉| 186/187 [18:46<00:06,  6.11s/it]
Training Epoch: 13/100, batch 370/375 completed (loss: 0.0268:  99%|█████████▉| 186/187 [18:46<00:06,  6.11s/it]
Training Epoch: 13/100, batch 370/375 completed (loss: 0.0852:  99%|█████████▉| 186/187 [18:46<00:06,  6.10s/it]
Training Epoch: 13/100, batch 371/375 completed (loss: 0.1469:  99%|█████████▉| 186/187 [18:46<00:06,  6.11s/it]
Training Epoch: 13/100, batch 371/375 completed (loss: 0.0101:  99%|█████████▉| 186/187 [18:46<00:06,  6.10s/it]
Training Epoch: 13/100, batch 371/375 completed (loss: 0.065:  99%|█████████▉| 186/187 [18:46<00:06,  6.10s/it]
Training Epoch: 13/100, batch 371/375 completed (loss: 0.0705:  99%|█████████▉| 186/187 [18:46<00:06,  6.11s/it]
Training Epoch: 13/100, batch 372/375 completed (loss: 0.0058:  99%|█████████▉| 186/187 [18:49<00:06,  6.11s/it]
Training Epoch: 13/100, batch 372/375 completed (loss: 0.0043:  99%|█████████▉| 186/187 [18:49<00:06,  6.10s/it]
Training Epoch: 13/100, batch 372/375 completed (loss: 0.0767:  99%|█████████▉| 186/187 [18:49<00:06,  6.10s/it]
Training Epoch: 13/100, batch 372/375 completed (loss: 0.2398:  99%|█████████▉| 186/187 [18:49<00:06,  6.11s/it]
Training Epoch: 13/100, batch 372/375 completed (loss: 0.2398: 100%|██████████| 187/187 [18:52<00:00,  6.08s/it]
Training Epoch: 13/100, batch 372/375 completed (loss: 0.0767: 100%|██████████| 187/187 [18:52<00:00,  6.08s/it]
Training Epoch: 13/100, batch 372/375 completed (loss: 0.0043: 100%|██████████| 187/187 [18:52<00:00,  6.08s/it]
Training Epoch: 13/100, batch 372/375 completed (loss: 0.0058: 100%|██████████| 187/187 [18:52<00:00,  6.08s/it]
Training Epoch: 13/100, batch 373/375 completed (loss: 0.0799: 100%|██████████| 187/187 [18:52<00:00,  6.08s/it]
Training Epoch: 13/100, batch 373/375 completed (loss: 0.0061: 100%|██████████| 187/187 [18:52<00:00,  6.08s/it]
Training Epoch: 13/100, batch 373/375 completed (loss: 0.0551: 100%|██████████| 187/187 [18:52<00:00,  6.08s/it]
Training Epoch: 13/100, batch 373/375 completed (loss: 0.0279: 100%|██████████| 187/187 [18:52<00:00,  6.08s/it]
Training Epoch: 13/100, batch 373/375 completed (loss: 0.0279: : 188it [18:55,  5.18s/it]                       
Training Epoch: 13/100, batch 373/375 completed (loss: 0.0061: : 188it [18:55,  5.18s/it]                       
Training Epoch: 13/100, batch 373/375 completed (loss: 0.0799: : 188it [18:55,  5.18s/it]                       
Training Epoch: 13/100, batch 373/375 completed (loss: 0.0551: : 188it [18:55,  5.18s/it]                       
Training Epoch: 13/100, batch 374/375 completed (loss: 0.0082: : 188it [18:55,  5.18s/it]
Training Epoch: 13/100, batch 374/375 completed (loss: 0.0054: : 188it [18:55,  5.18s/it]
Training Epoch: 13/100, batch 374/375 completed (loss: 0.0244: : 188it [18:55,  5.18s/it]
Training Epoch: 13/100, batch 374/375 completed (loss: 0.091: : 188it [18:55,  5.18s/it] 
Training Epoch: 13/100, batch 374/375 completed (loss: 0.0244: : 188it [18:55,  6.04s/it]

Training Epoch: 13/100, batch 374/375 completed (loss: 0.0054: : 188it [18:55,  6.04s/it]

Training Epoch: 13/100, batch 374/375 completed (loss: 0.0082: : 188it [18:55,  6.04s/it]

Training Epoch: 13/100, batch 374/375 completed (loss: 0.091: : 188it [18:55,  6.04s/it]
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/c10d_logger.py:83: UserWarning: barrier(): using the device under current context. You can specify `device_id` in `init_process_group` to mute this warning.
  return func(*args, **kwargs)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
saving model + train state.
Warning: failed to upload checkpoint_13 to HF: (Request ID: Root=1-6a1b8acd-353c2fea2a8f9d0245443430;4fb86161-30c3-4f2d-bb86-ae2614dda988)

403 Forbidden: You need to setup automatic credit recharge in order to upload more data. You can do so at /settings/billing..
Cannot access content at: https://huggingface.co/seyedparsa/thoughtformer-checkpoints.git/info/lfs/objects/batch.
Make sure your token has the correct permissions.
Warning: failed to upload train_state_13 to HF: (Request ID: Root=1-6a1b8ad1-1721900a2d4a795d4438e88c;db9cbb82-abaa-4413-a63f-864a6870de2c)

403 Forbidden: You need to setup automatic credit recharge in order to upload more data. You can do so at /settings/billing..
Cannot access content at: https://huggingface.co/seyedparsa/thoughtformer-checkpoints.git/info/lfs/objects/batch.
Make sure your token has the correct permissions.
Deleted old checkpoint: train_state_12
Deleted old checkpoint: checkpoint_12
[warn] HF upload failed for ckpt_13; kept locally, earlier ckpts deleted.

Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]
Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]
Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]eval loss 0.021239456579726385

Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]
[CORRECT] Example 0 (diff=3)
  Expected : 'Claude'
  Extracted: 'Claude'
  Generated: '### Claude'

[WRONG] Example 1 (diff=8)
  Expected : 'Guy'
  Extracted: 'Irma'
  Generated: '### Irma'

Test Accuracy:   3%|▎         | 1/32 [00:07<03:39,  7.06s/it]
Test Accuracy:   3%|▎         | 1/32 [00:07<03:39,  7.07s/it]
Test accuracy: 1.0:   3%|▎         | 1/32 [00:07<03:39,  7.06s/it]
Test accuracy: 0.75:   3%|▎         | 1/32 [00:07<03:39,  7.07s/it]
[WRONG] Example 2 (diff=10)
Test Accuracy:   3%|▎         | 1/32 [00:07<03:39,  7.07s/it]
  Expected : 'Elvis'
  Extracted: 'Shannon'
  Generated: '### Shannon'

Test accuracy: 0.75:   3%|▎         | 1/32 [00:07<03:39,  7.07s/it]
[CORRECT] Example 3 (diff=8)
  Expected : 'Coral'
  Extracted: 'Coral'
  Generated: '### Coral'

[CORRECT] Example 4 (diff=1)
  Expected : 'Patricia'
  Extracted: 'Patricia'
  Generated: '### Patricia'

Test Accuracy:   3%|▎         | 1/32 [00:06<03:34,  6.90s/it]
Test accuracy: 0.75:   3%|▎         | 1/32 [00:06<03:34,  6.90s/it]
Test accuracy: 1.0:   6%|▋         | 2/32 [00:13<03:15,  6.52s/it]
Test accuracy: 0.75:   6%|▋         | 2/32 [00:13<03:15,  6.52s/it]
Test accuracy: 0.75:   6%|▋         | 2/32 [00:13<03:15,  6.52s/it]
Test accuracy: 1.0:   6%|▋         | 2/32 [00:13<03:15,  6.52s/it]
Test accuracy: 0.81:   6%|▋         | 2/32 [00:13<03:15,  6.52s/it]
Test accuracy: 0.56:   6%|▋         | 2/32 [00:13<03:15,  6.52s/it]
Test accuracy: 0.75:   6%|▋         | 2/32 [00:13<03:13,  6.45s/it]
Test accuracy: 0.75:   6%|▋         | 2/32 [00:13<03:13,  6.45s/it]
Test accuracy: 0.56:   9%|▉         | 3/32 [00:19<03:12,  6.64s/it]
Test accuracy: 1.0:   9%|▉         | 3/32 [00:19<03:12,  6.64s/it]
Test accuracy: 0.67:   9%|▉         | 3/32 [00:19<03:12,  6.64s/it]
Test accuracy: 0.96:   9%|▉         | 3/32 [00:19<03:12,  6.64s/it]
Test accuracy: 0.75:   9%|▉         | 3/32 [00:19<03:11,  6.60s/it]
Test accuracy: 0.75:   9%|▉         | 3/32 [00:19<03:11,  6.60s/it]
Test accuracy: 0.81:   9%|▉         | 3/32 [00:19<03:12,  6.64s/it]
Test accuracy: 0.79:   9%|▉         | 3/32 [00:19<03:12,  6.64s/it]
Test accuracy: 0.79:  12%|█▎        | 4/32 [00:26<03:04,  6.60s/it]
Test accuracy: 0.67:  12%|█▎        | 4/32 [00:26<03:04,  6.61s/it]
Test accuracy: 0.96:  12%|█▎        | 4/32 [00:26<03:04,  6.61s/it]
Test accuracy: 0.66:  12%|█▎        | 4/32 [00:26<03:04,  6.61s/it]
Test accuracy: 0.88:  12%|█▎        | 4/32 [00:26<03:04,  6.61s/it]
Test accuracy: 0.75:  12%|█▎        | 4/32 [00:26<03:04,  6.60s/it]
Test accuracy: 0.75:  12%|█▎        | 4/32 [00:26<03:04,  6.58s/it]
Test accuracy: 0.81:  12%|█▎        | 4/32 [00:26<03:04,  6.58s/it]
Test accuracy: 0.75:  16%|█▌        | 5/32 [00:32<02:53,  6.44s/it]
Test accuracy: 0.66:  16%|█▌        | 5/32 [00:32<02:53,  6.44s/it]
Test accuracy: 0.88:  16%|█▌        | 5/32 [00:32<02:53,  6.44s/it]
Test accuracy: 0.81:  16%|█▌        | 5/32 [00:32<02:53,  6.42s/it]
Test accuracy: 0.7:  16%|█▌        | 5/32 [00:32<02:53,  6.44s/it] 
Test accuracy: 0.65:  16%|█▌        | 5/32 [00:32<02:53,  6.44s/it]
Test accuracy: 0.82:  16%|█▌        | 5/32 [00:32<02:53,  6.44s/it]
Test accuracy: 0.82:  16%|█▌        | 5/32 [00:32<02:53,  6.42s/it]
Test accuracy: 0.82:  19%|█▉        | 6/32 [00:39<02:46,  6.41s/it]
Test accuracy: 0.65:  19%|█▉        | 6/32 [00:39<02:46,  6.41s/it]
Test accuracy: 0.7:  19%|█▉        | 6/32 [00:39<02:46,  6.41s/it]
Test accuracy: 0.83:  19%|█▉        | 6/32 [00:39<02:46,  6.41s/it]
Test accuracy: 0.69:  19%|█▉        | 6/32 [00:39<02:46,  6.41s/it]
Test accuracy: 0.82:  19%|█▉        | 6/32 [00:38<02:46,  6.40s/it]
Test accuracy: 0.73:  19%|█▉        | 6/32 [00:39<02:46,  6.41s/it]
Test accuracy: 0.83:  19%|█▉        | 6/32 [00:38<02:46,  6.40s/it]
Test accuracy: 0.83:  22%|██▏       | 7/32 [00:45<02:39,  6.37s/it]
Test accuracy: 0.69:  22%|██▏       | 7/32 [00:45<02:39,  6.37s/it]
Test accuracy: 0.83:  22%|██▏       | 7/32 [00:45<02:39,  6.37s/it]
Test accuracy: 0.73:  22%|██▏       | 7/32 [00:45<02:39,  6.37s/it]
Test accuracy: 0.8:  22%|██▏       | 7/32 [00:45<02:39,  6.37s/it] 
Test accuracy: 0.68:  22%|██▏       | 7/32 [00:45<02:39,  6.37s/it]
Test accuracy: 0.75:  22%|██▏       | 7/32 [00:45<02:39,  6.37s/it]
Test accuracy: 0.82:  22%|██▏       | 7/32 [00:45<02:39,  6.37s/it]
Test accuracy: 0.75:  25%|██▌       | 8/32 [00:51<02:31,  6.29s/it]
Test accuracy: 0.8:  25%|██▌       | 8/32 [00:51<02:31,  6.29s/it]
Test accuracy: 0.68:  25%|██▌       | 8/32 [00:51<02:31,  6.29s/it]
Test accuracy: 0.83:  25%|██▌       | 8/32 [00:51<02:31,  6.29s/it]
Test accuracy: 0.77:  25%|██▌       | 8/32 [00:51<02:31,  6.29s/it]
Test accuracy: 0.67:  25%|██▌       | 8/32 [00:51<02:31,  6.29s/it]
Test accuracy: 0.82:  25%|██▌       | 8/32 [00:51<02:30,  6.29s/it]
Test accuracy: 0.84:  25%|██▌       | 8/32 [00:51<02:30,  6.29s/it]
Test accuracy: 0.77:  28%|██▊       | 9/32 [00:58<02:29,  6.51s/it]
Test accuracy: 0.83:  28%|██▊       | 9/32 [00:58<02:29,  6.51s/it]
Test accuracy: 0.78:  28%|██▊       | 9/32 [00:58<02:29,  6.51s/it]
Test accuracy: 0.85:  28%|██▊       | 9/32 [00:58<02:29,  6.51s/it]
Test accuracy: 0.84:  28%|██▊       | 9/32 [00:58<02:29,  6.50s/it]
Test accuracy: 0.86:  28%|██▊       | 9/32 [00:58<02:29,  6.50s/it]
Test accuracy: 0.67:  28%|██▊       | 9/32 [00:58<02:29,  6.51s/it]
Test accuracy: 0.68:  28%|██▊       | 9/32 [00:58<02:29,  6.51s/it]
Test accuracy: 0.68:  31%|███▏      | 10/32 [01:04<02:21,  6.45s/it]
Test accuracy: 0.78:  31%|███▏      | 10/32 [01:04<02:21,  6.45s/it]
Test accuracy: 0.85:  31%|███▏      | 10/32 [01:04<02:21,  6.45s/it]
Test accuracy: 0.68:  31%|███▏      | 10/32 [01:04<02:21,  6.45s/it]
Test accuracy: 0.86:  31%|███▏      | 10/32 [01:04<02:21,  6.45s/it]
Test accuracy: 0.77:  31%|███▏      | 10/32 [01:04<02:21,  6.45s/it]
Test accuracy: 0.86:  31%|███▏      | 10/32 [01:04<02:21,  6.45s/it]
Test accuracy: 0.86:  31%|███▏      | 10/32 [01:04<02:21,  6.45s/it]
Test accuracy: 0.68:  34%|███▍      | 11/32 [01:10<02:13,  6.35s/it]
Test accuracy: 0.77:  34%|███▍      | 11/32 [01:10<02:13,  6.35s/it]
Test accuracy: 0.86:  34%|███▍      | 11/32 [01:10<02:13,  6.35s/it]
Test accuracy: 0.78:  34%|███▍      | 11/32 [01:10<02:13,  6.35s/it]
Test accuracy: 0.69:  34%|███▍      | 11/32 [01:10<02:13,  6.35s/it]
Test accuracy: 0.83:  34%|███▍      | 11/32 [01:10<02:13,  6.35s/it]
Test accuracy: 0.86:  34%|███▍      | 11/32 [01:10<02:13,  6.35s/it]
Test accuracy: 0.84:  34%|███▍      | 11/32 [01:10<02:13,  6.35s/it]
Test accuracy: 0.78:  38%|███▊      | 12/32 [01:17<02:07,  6.35s/it]
Test accuracy: 0.83:  38%|███▊      | 12/32 [01:17<02:07,  6.35s/it]
Test accuracy: 0.69:  38%|███▊      | 12/32 [01:17<02:07,  6.35s/it]
Test accuracy: 0.84:  38%|███▊      | 12/32 [01:17<02:06,  6.35s/it]
Test accuracy: 0.77:  38%|███▊      | 12/32 [01:17<02:07,  6.35s/it]
Test accuracy: 0.8:  38%|███▊      | 12/32 [01:17<02:07,  6.35s/it] 
Test accuracy: 0.71:  38%|███▊      | 12/32 [01:17<02:07,  6.35s/it]
Test accuracy: 0.84:  38%|███▊      | 12/32 [01:17<02:06,  6.35s/it]
Test accuracy: 0.71:  41%|████      | 13/32 [01:23<02:00,  6.34s/it]
Test accuracy: 0.8:  41%|████      | 13/32 [01:23<02:00,  6.34s/it]
Test accuracy: 0.77:  41%|████      | 13/32 [01:23<02:00,  6.34s/it]
Test accuracy: 0.7:  41%|████      | 13/32 [01:23<02:00,  6.34s/it] 
Test accuracy: 0.82:  41%|████      | 13/32 [01:23<02:00,  6.34s/it]
Test accuracy: 0.78:  41%|████      | 13/32 [01:23<02:00,  6.34s/it]
Test accuracy: 0.84:  41%|████      | 13/32 [01:23<02:00,  6.34s/it]
Test accuracy: 0.83:  41%|████      | 13/32 [01:23<02:00,  6.34s/it]
Test accuracy: 0.82:  44%|████▍     | 14/32 [01:29<01:53,  6.28s/it]
Test accuracy: 0.83:  44%|████▍     | 14/32 [01:29<01:53,  6.28s/it]
Test accuracy: 0.7:  44%|████▍     | 14/32 [01:29<01:53,  6.28s/it]
Test accuracy: 0.83:  44%|████▍     | 14/32 [01:29<01:53,  6.28s/it]
Test accuracy: 0.71:  44%|████▍     | 14/32 [01:29<01:53,  6.28s/it]
Test accuracy: 0.8:  44%|████▍     | 14/32 [01:29<01:53,  6.28s/it] 
Test accuracy: 0.78:  44%|████▍     | 14/32 [01:29<01:53,  6.28s/it]
Test accuracy: 0.79:  44%|████▍     | 14/32 [01:29<01:53,  6.28s/it]
Test accuracy: 0.71:  47%|████▋     | 15/32 [01:36<01:47,  6.31s/it]
Test accuracy: 0.79:  47%|████▋     | 15/32 [01:36<01:47,  6.31s/it]
Test accuracy: 0.83:  47%|████▋     | 15/32 [01:36<01:47,  6.31s/it]
Test accuracy: 0.71:  47%|████▋     | 15/32 [01:36<01:47,  6.31s/it]
Test accuracy: 0.8:  47%|████▋     | 15/32 [01:35<01:47,  6.30s/it]
Test accuracy: 0.78:  47%|████▋     | 15/32 [01:36<01:47,  6.31s/it]
Test accuracy: 0.83:  47%|████▋     | 15/32 [01:36<01:47,  6.31s/it]
Test accuracy: 0.8:  47%|████▋     | 15/32 [01:35<01:47,  6.30s/it]
Test accuracy: 0.71:  50%|█████     | 16/32 [01:42<01:40,  6.30s/it]
Test accuracy: 0.78:  50%|█████     | 16/32 [01:42<01:40,  6.30s/it]
Test accuracy: 0.83:  50%|█████     | 16/32 [01:42<01:40,  6.30s/it]
Test accuracy: 0.71:  50%|█████     | 16/32 [01:42<01:40,  6.30s/it]
Test accuracy: 0.78:  50%|█████     | 16/32 [01:42<01:40,  6.30s/it]
Test accuracy: 0.84:  50%|█████     | 16/32 [01:42<01:40,  6.30s/it]
Test accuracy: 0.8:  50%|█████     | 16/32 [01:42<01:40,  6.30s/it]
Test accuracy: 0.8:  50%|█████     | 16/32 [01:42<01:40,  6.30s/it]
Test accuracy: 0.84:  53%|█████▎    | 17/32 [01:48<01:34,  6.31s/it]
Test accuracy: 0.71:  53%|█████▎    | 17/32 [01:48<01:34,  6.31s/it]
Test accuracy: 0.78:  53%|█████▎    | 17/32 [01:48<01:34,  6.31s/it]
Test accuracy: 0.83:  53%|█████▎    | 17/32 [01:48<01:34,  6.31s/it]
Test accuracy: 0.72:  53%|█████▎    | 17/32 [01:48<01:34,  6.31s/it]
Test accuracy: 0.79:  53%|█████▎    | 17/32 [01:48<01:34,  6.31s/it]
Test accuracy: 0.8:  53%|█████▎    | 17/32 [01:48<01:34,  6.31s/it]
Test accuracy: 0.81:  53%|█████▎    | 17/32 [01:48<01:34,  6.31s/it]
Test accuracy: 0.79:  56%|█████▋    | 18/32 [01:55<01:28,  6.32s/it]
Test accuracy: 0.72:  56%|█████▋    | 18/32 [01:55<01:28,  6.32s/it]
Test accuracy: 0.83:  56%|█████▋    | 18/32 [01:55<01:28,  6.32s/it]
Test accuracy: 0.73:  56%|█████▋    | 18/32 [01:55<01:28,  6.32s/it]
Test accuracy: 0.79:  56%|█████▋    | 18/32 [01:55<01:28,  6.32s/it]
Test accuracy: 0.83:  56%|█████▋    | 18/32 [01:55<01:28,  6.32s/it]
Test accuracy: 0.81:  56%|█████▋    | 18/32 [01:54<01:28,  6.32s/it]
Test accuracy: 0.81:  56%|█████▋    | 18/32 [01:54<01:28,  6.32s/it]
Test accuracy: 0.83:  59%|█████▉    | 19/32 [02:01<01:21,  6.27s/it]
Test accuracy: 0.73:  59%|█████▉    | 19/32 [02:01<01:21,  6.27s/it]
Test accuracy: 0.83:  59%|█████▉    | 19/32 [02:01<01:21,  6.27s/it]
Test accuracy: 0.72:  59%|█████▉    | 19/32 [02:01<01:21,  6.27s/it]
Test accuracy: 0.79:  59%|█████▉    | 19/32 [02:01<01:21,  6.27s/it]
Test accuracy: 0.81:  59%|█████▉    | 19/32 [02:01<01:21,  6.27s/it]
Test accuracy: 0.8:  59%|█████▉    | 19/32 [02:01<01:21,  6.27s/it] 
Test accuracy: 0.82:  59%|█████▉    | 19/32 [02:01<01:21,  6.27s/it]
Test accuracy: 0.72:  62%|██████▎   | 20/32 [02:07<01:15,  6.30s/it]
Test accuracy: 0.83:  62%|██████▎   | 20/32 [02:07<01:15,  6.30s/it]
Test accuracy: 0.74:  62%|██████▎   | 20/32 [02:07<01:15,  6.30s/it]
Test accuracy: 0.82:  62%|██████▎   | 20/32 [02:07<01:15,  6.30s/it]
Test accuracy: 0.8:  62%|██████▎   | 20/32 [02:07<01:15,  6.30s/it]
Test accuracy: 0.82:  62%|██████▎   | 20/32 [02:07<01:15,  6.30s/it]
Test accuracy: 0.8:  62%|██████▎   | 20/32 [02:07<01:15,  6.30s/it]
Test accuracy: 0.82:  62%|██████▎   | 20/32 [02:07<01:15,  6.30s/it]
Test accuracy: 0.74:  66%|██████▌   | 21/32 [02:14<01:11,  6.52s/it]
Test accuracy: 0.82:  66%|██████▌   | 21/32 [02:14<01:11,  6.52s/it]
Test accuracy: 0.74:  66%|██████▌   | 21/32 [02:14<01:11,  6.52s/it]
Test accuracy: 0.82:  66%|██████▌   | 21/32 [02:14<01:11,  6.52s/it]
Test accuracy: 0.82:  66%|██████▌   | 21/32 [02:14<01:11,  6.52s/it]
Test accuracy: 0.81:  66%|██████▌   | 21/32 [02:14<01:11,  6.52s/it]
Test accuracy: 0.8:  66%|██████▌   | 21/32 [02:14<01:11,  6.52s/it]
Test accuracy: 0.8:  66%|██████▌   | 21/32 [02:14<01:11,  6.52s/it]
Test accuracy: 0.74:  69%|██████▉   | 22/32 [02:20<01:04,  6.41s/it]
Test accuracy: 0.8:  69%|██████▉   | 22/32 [02:20<01:04,  6.41s/it]
Test accuracy: 0.82:  69%|██████▉   | 22/32 [02:20<01:04,  6.41s/it]
Test accuracy: 0.8:  69%|██████▉   | 22/32 [02:20<01:04,  6.41s/it]
Test accuracy: 0.73:  69%|██████▉   | 22/32 [02:20<01:04,  6.41s/it]
Test accuracy: 0.82:  69%|██████▉   | 22/32 [02:20<01:04,  6.41s/it]
Test accuracy: 0.81:  69%|██████▉   | 22/32 [02:20<01:04,  6.41s/it]
Test accuracy: 0.82:  69%|██████▉   | 22/32 [02:20<01:04,  6.41s/it]
Test accuracy: 0.73:  72%|███████▏  | 23/32 [02:27<00:57,  6.39s/it]
Test accuracy: 0.82:  72%|███████▏  | 23/32 [02:27<00:57,  6.39s/it]
Test accuracy: 0.74:  72%|███████▏  | 23/32 [02:27<00:57,  6.39s/it]
Test accuracy: 0.83:  72%|███████▏  | 23/32 [02:27<00:57,  6.39s/it]
Test accuracy: 0.8:  72%|███████▏  | 23/32 [02:27<00:57,  6.39s/it]
Test accuracy: 0.8:  72%|███████▏  | 23/32 [02:27<00:57,  6.39s/it]
Test accuracy: 0.82:  72%|███████▏  | 23/32 [02:26<00:57,  6.39s/it]
Test accuracy: 0.82:  72%|███████▏  | 23/32 [02:26<00:57,  6.39s/it]
Test accuracy: 0.8:  75%|███████▌  | 24/32 [02:34<00:52,  6.59s/it]
Test accuracy: 0.83:  75%|███████▌  | 24/32 [02:34<00:52,  6.59s/it]
Test accuracy: 0.74:  75%|███████▌  | 24/32 [02:34<00:52,  6.59s/it]
Test accuracy: 0.8:  75%|███████▌  | 24/32 [02:34<00:52,  6.59s/it]
Test accuracy: 0.83:  75%|███████▌  | 24/32 [02:34<00:52,  6.59s/it]
Test accuracy: 0.74:  75%|███████▌  | 24/32 [02:34<00:52,  6.59s/it]
Test accuracy: 0.82:  75%|███████▌  | 24/32 [02:33<00:52,  6.60s/it]
Test accuracy: 0.81:  75%|███████▌  | 24/32 [02:33<00:52,  6.60s/it]
Test accuracy: 0.74:  78%|███████▊  | 25/32 [02:40<00:45,  6.46s/it]
Test accuracy: 0.8:  78%|███████▊  | 25/32 [02:40<00:45,  6.46s/it]
Test accuracy: 0.81:  78%|███████▊  | 25/32 [02:40<00:45,  6.46s/it]
Test accuracy: 0.81:  78%|███████▊  | 25/32 [02:40<00:45,  6.46s/it]
Test accuracy: 0.75:  78%|███████▊  | 25/32 [02:40<00:45,  6.46s/it]
Test accuracy: 0.83:  78%|███████▊  | 25/32 [02:40<00:45,  6.46s/it]
Test accuracy: 0.82:  78%|███████▊  | 25/32 [02:40<00:45,  6.46s/it]
Test accuracy: 0.81:  78%|███████▊  | 25/32 [02:40<00:45,  6.46s/it]
Test accuracy: 0.82:  81%|████████▏ | 26/32 [02:46<00:38,  6.43s/it]
Test accuracy: 0.75:  81%|████████▏ | 26/32 [02:46<00:38,  6.43s/it]
Test accuracy: 0.83:  81%|████████▏ | 26/32 [02:46<00:38,  6.43s/it]
Test accuracy: 0.74:  81%|████████▏ | 26/32 [02:46<00:38,  6.43s/it]
Test accuracy: 0.81:  81%|████████▏ | 26/32 [02:46<00:38,  6.43s/it]
Test accuracy: 0.81:  81%|████████▏ | 26/32 [02:46<00:38,  6.43s/it]
Test accuracy: 0.81:  81%|████████▏ | 26/32 [02:46<00:38,  6.43s/it]
Test accuracy: 0.8:  81%|████████▏ | 26/32 [02:46<00:38,  6.43s/it] 
Test accuracy: 0.81:  84%|████████▍ | 27/32 [02:52<00:31,  6.40s/it]
Test accuracy: 0.83:  84%|████████▍ | 27/32 [02:52<00:31,  6.40s/it]
Test accuracy: 0.74:  84%|████████▍ | 27/32 [02:52<00:31,  6.40s/it]
Test accuracy: 0.81:  84%|████████▍ | 27/32 [02:52<00:31,  6.40s/it]
Test accuracy: 0.83:  84%|████████▍ | 27/32 [02:52<00:31,  6.40s/it]
Test accuracy: 0.75:  84%|████████▍ | 27/32 [02:52<00:31,  6.40s/it]
Test accuracy: 0.8:  84%|████████▍ | 27/32 [02:52<00:31,  6.40s/it]
Test accuracy: 0.8:  84%|████████▍ | 27/32 [02:52<00:31,  6.40s/it]
Test accuracy: 0.83:  88%|████████▊ | 28/32 [03:00<00:26,  6.60s/it]
Test accuracy: 0.75:  88%|████████▊ | 28/32 [03:00<00:26,  6.60s/it]
Test accuracy: 0.75:  88%|████████▊ | 28/32 [03:00<00:26,  6.60s/it]
Test accuracy: 0.82:  88%|████████▊ | 28/32 [03:00<00:26,  6.60s/it]
Test accuracy: 0.8:  88%|████████▊ | 28/32 [02:59<00:26,  6.60s/it]
Test accuracy: 0.8:  88%|████████▊ | 28/32 [02:59<00:26,  6.60s/it]
Test accuracy: 0.81:  88%|████████▊ | 28/32 [03:00<00:26,  6.60s/it]
Test accuracy: 0.8:  88%|████████▊ | 28/32 [03:00<00:26,  6.60s/it] 
Test accuracy: 0.8:  91%|█████████ | 29/32 [03:06<00:19,  6.46s/it]
Test accuracy: 0.82:  91%|█████████ | 29/32 [03:06<00:19,  6.46s/it]
Test accuracy: 0.75:  91%|█████████ | 29/32 [03:06<00:19,  6.46s/it]
Test accuracy: 0.8:  91%|█████████ | 29/32 [03:06<00:19,  6.46s/it]
Test accuracy: 0.8:  91%|█████████ | 29/32 [03:06<00:19,  6.46s/it]
Test accuracy: 0.82:  91%|█████████ | 29/32 [03:06<00:19,  6.46s/it]
Test accuracy: 0.76:  91%|█████████ | 29/32 [03:06<00:19,  6.46s/it]
Test accuracy: 0.8:  91%|█████████ | 29/32 [03:06<00:19,  6.46s/it]
Test accuracy: 0.8:  94%|█████████▍| 30/32 [03:12<00:12,  6.42s/it]
Test accuracy: 0.76:  94%|█████████▍| 30/32 [03:12<00:12,  6.42s/it]
Test accuracy: 0.82:  94%|█████████▍| 30/32 [03:12<00:12,  6.42s/it]
Test accuracy: 0.8:  94%|█████████▍| 30/32 [03:12<00:12,  6.42s/it]
Test accuracy: 0.76:  94%|█████████▍| 30/32 [03:12<00:12,  6.42s/it]
Test accuracy: 0.8:  94%|█████████▍| 30/32 [03:12<00:12,  6.42s/it]
Test accuracy: 0.82:  94%|█████████▍| 30/32 [03:12<00:12,  6.42s/it]
Test accuracy: 0.8:  94%|█████████▍| 30/32 [03:12<00:12,  6.42s/it]
Test accuracy: 0.76:  97%|█████████▋| 31/32 [03:19<00:06,  6.61s/it]
Test accuracy: 0.82:  97%|█████████▋| 31/32 [03:19<00:06,  6.61s/it]
Test accuracy: 0.76:  97%|█████████▋| 31/32 [03:19<00:06,  6.61s/it]
Test accuracy: 0.82:  97%|█████████▋| 31/32 [03:19<00:06,  6.61s/it]
Test accuracy: 0.8:  97%|█████████▋| 31/32 [03:19<00:06,  6.61s/it]
Test accuracy: 0.8:  97%|█████████▋| 31/32 [03:19<00:06,  6.61s/it]
Test accuracy: 0.8:  97%|█████████▋| 31/32 [03:19<00:06,  6.62s/it]
Test accuracy: 0.8:  97%|█████████▋| 31/32 [03:19<00:06,  6.62s/it]
Test accuracy: 0.8: 100%|██████████| 32/32 [03:21<00:00,  5.30s/it]
Test accuracy: 0.8: 100%|██████████| 32/32 [03:21<00:00,  5.30s/it]
Test accuracy: 0.76: 100%|██████████| 32/32 [03:21<00:00,  5.30s/it]
Test accuracy: 0.8: 100%|██████████| 32/32 [03:21<00:00,  5.30s/it]
Test accuracy: 0.82: 100%|██████████| 32/32 [03:21<00:00,  5.30s/it]
Test accuracy: 0.76: 100%|██████████| 32/32 [03:21<00:00,  5.30s/it]
Test accuracy: 0.79: 100%|██████████| 32/32 [03:21<00:00,  5.30s/it]
Test accuracy: 0.82: 100%|██████████| 32/32 [03:21<00:00,  5.30s/it]
Test accuracy: 0.79: 100%|██████████| 32/32 [03:21<00:00,  6.31s/it]
Device 2: Cor=198, CoT=0, Total=250

Test accuracy: 0.76: 100%|██████████| 32/32 [03:21<00:00,  6.31s/it]

Test accuracy: 0.82: 100%|██████████| 32/32 [03:21<00:00,  6.31s/it]Device 3: Cor=191, CoT=0, Total=250

Device 1: Cor=206, CoT=0, Total=250

Test accuracy: 0.8: 100%|██████████| 32/32 [03:21<00:00,  6.31s/it]
Device 0: Cor=199, CoT=0, Total=250

============================================================
EVALUATION RESULTS (Epoch 12)
============================================================
Config: data_curriculum=ON, thought_curriculum=OFF
thought_stage=10, data_stage=6, n_latent_tokens=10
Training on difficulties: 0-6

Overall accuracy: 794 / 1000 = 0.7940

Per-difficulty accuracy:
  diff=1: 98/100 = 0.9800 (trained)
  diff=2: 100/100 = 1.0000 (trained)
  diff=3: 100/100 = 1.0000 (trained)
  diff=4: 96/100 = 0.9600 (trained)
  diff=5: 95/100 = 0.9500 (trained)
  diff=6: 96/100 = 0.9600 (trained)
  diff=7: 52/100 = 0.5200 (untrained)
  diff=8: 44/100 = 0.4400 (untrained)
  diff=9: 63/100 = 0.6300 (untrained)
  diff=10: 50/100 = 0.5000 (untrained)

Trained (diff 0-6): 585/600 = 0.9750
============================================================


*** ADAPTIVE STAGING: acc_trained=0.9750, acc_diff6=0.9600 >= 0.9 ***
*** Advancing to stage 7 ***

Warning: failed to upload checkpoint_13 to HF: (Request ID: Root=1-6a1b8bba-60022d5014cd386e705139ce;739ae0e8-f85d-41b4-a9e0-de435c8b8ca8)

403 Forbidden: You need to setup automatic credit recharge in order to upload more data. You can do so at /settings/billing..
Cannot access content at: https://huggingface.co/seyedparsa/thoughtformer-checkpoints.git/info/lfs/objects/batch.
Make sure your token has the correct permissions.

============================================================
EPOCH 13/100 (thought_stage=10, data_stage=7)
============================================================
  thought_stage=10, c_thought=1, max_difficulty=7

Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):  16%|█▌        | 156/1000 [00:00<00:00, 946.64 examples/s]
Map (num_proc=4):  16%|█▌        | 156/1000 [00:00<00:00, 921.56 examples/s]
Map (num_proc=4):  16%|█▌        | 156/1000 [00:00<00:00, 912.82 examples/s]
Map (num_proc=4):  81%|████████  | 810/1000 [00:00<00:00, 3311.53 examples/s]
Map (num_proc=4):  81%|████████  | 812/1000 [00:00<00:00, 3011.50 examples/s]
Map (num_proc=4):  81%|████████  | 810/1000 [00:00<00:00, 2986.91 examples/s]
Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2407.53 examples/s]

Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2440.54 examples/s]

Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2371.60 examples/s]

Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Map (num_proc=4):  16%|█▌        | 157/1000 [00:00<00:01, 509.05 examples/s]
Map (num_proc=4):  82%|████████▏ | 815/1000 [00:00<00:00, 2337.74 examples/s]
Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:04, 1949.37 examples/s]
Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:04, 1930.77 examples/s]
Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:04, 1907.83 examples/s]
Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 1237.14 examples/s]

Filter (num_proc=4):  20%|██        | 2000/10000 [00:00<00:02, 3511.00 examples/s]
Filter (num_proc=4):  20%|██        | 2000/10000 [00:00<00:02, 3662.74 examples/s]
Filter (num_proc=4):  20%|██        | 2000/10000 [00:00<00:02, 3555.93 examples/s]
Filter (num_proc=4):  30%|███       | 3000/10000 [00:00<00:01, 5124.40 examples/s]
Filter (num_proc=4):  30%|███       | 3000/10000 [00:00<00:01, 4900.14 examples/s]
Filter (num_proc=4):  30%|███       | 3000/10000 [00:00<00:01, 4882.26 examples/s]
Filter (num_proc=4):  50%|█████     | 5000/10000 [00:00<00:00, 8484.91 examples/s]
Filter (num_proc=4):  50%|█████     | 5000/10000 [00:00<00:00, 8678.79 examples/s]
Filter (num_proc=4):  50%|█████     | 5000/10000 [00:00<00:00, 8306.05 examples/s]
Filter (num_proc=4):  75%|███████▌  | 7500/10000 [00:00<00:00, 12100.38 examples/s]
Filter (num_proc=4):  75%|███████▌  | 7500/10000 [00:00<00:00, 11783.70 examples/s]
Filter (num_proc=4):  75%|███████▌  | 7500/10000 [00:00<00:00, 11600.84 examples/s]
Filter (num_proc=4):  90%|█████████ | 9000/10000 [00:01<00:00, 12450.72 examples/s]
Filter (num_proc=4):  90%|█████████ | 9000/10000 [00:01<00:00, 12592.00 examples/s]
Filter (num_proc=4):  90%|█████████ | 9000/10000 [00:01<00:00, 12051.39 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 7441.22 examples/s]

Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 7545.19 examples/s]

Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 7306.93 examples/s]

Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:07, 1263.54 examples/s]
Filter (num_proc=4):  20%|██        | 2000/10000 [00:00<00:03, 2267.89 examples/s]
Filter (num_proc=4):  35%|███▌      | 3500/10000 [00:01<00:01, 3921.10 examples/s]
Filter (num_proc=4):  60%|██████    | 6000/10000 [00:01<00:00, 6563.33 examples/s]
Filter (num_proc=4):  85%|████████▌ | 8500/10000 [00:01<00:00, 8453.77 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 8388.91 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:02<00:00, 4938.91 examples/s]

Map (num_proc=4):   0%|          | 0/7000 [00:00<?, ? examples/s]
Map (num_proc=4):   2%|▏         | 137/7000 [00:01<00:55, 123.83 examples/s]
Map (num_proc=4):   8%|▊         | 547/7000 [00:01<00:11, 547.31 examples/s]
Map (num_proc=4):  12%|█▏        | 816/7000 [00:01<00:07, 821.88 examples/s]
Map (num_proc=4):  16%|█▌        | 1088/7000 [00:01<00:06, 858.11 examples/s]
Map (num_proc=4):  22%|██▏       | 1570/7000 [00:01<00:03, 1422.95 examples/s]
Map (num_proc=4):  29%|██▊       | 2011/7000 [00:01<00:02, 1936.01 examples/s]
Map (num_proc=4):  37%|███▋      | 2585/7000 [00:01<00:01, 2657.70 examples/s]
Map (num_proc=4):  43%|████▎     | 3021/7000 [00:02<00:01, 2092.52 examples/s]
Map (num_proc=4):  49%|████▉     | 3459/7000 [00:02<00:01, 2343.34 examples/s]
Map (num_proc=4):  55%|█████▍    | 3841/7000 [00:02<00:01, 2158.80 examples/s]
Map (num_proc=4):  64%|██████▍   | 4493/7000 [00:02<00:00, 2719.16 examples/s]
Map (num_proc=4):  70%|███████   | 4906/7000 [00:02<00:00, 2878.22 examples/s]
Map (num_proc=4):  77%|███████▋  | 5360/7000 [00:03<00:00, 3019.91 examples/s]
Map (num_proc=4):  82%|████████▏ | 5741/7000 [00:03<00:00, 2622.68 examples/s]
Map (num_proc=4):  89%|████████▉ | 6231/7000 [00:03<00:00, 2809.81 examples/s]
Map (num_proc=4):  94%|█████████▍| 6596/7000 [00:03<00:00, 1617.69 examples/s]
Map (num_proc=4): 100%|██████████| 7000/7000 [00:03<00:00, 1946.37 examples/s]
Map (num_proc=4): 100%|██████████| 7000/7000 [00:04<00:00, 1473.21 examples/s]

Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 1400.91 examples/s]
Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 1380.12 examples/s]
Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 1350.35 examples/s]
Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2684.98 examples/s]

Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2668.99 examples/s]

Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2584.76 examples/s]

Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:01, 733.96 examples/s]
Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 1362.87 examples/s]

Map (num_proc=4):   0%|          | 0/700 [00:00<?, ? examples/s]
Map (num_proc=4):  19%|█▉        | 136/700 [00:00<00:01, 359.00 examples/s]
Map (num_proc=4):  44%|████▍     | 308/700 [00:00<00:00, 659.34 examples/s]
Map (num_proc=4):  75%|███████▌  | 525/700 [00:00<00:00, 1037.16 examples/s]
Map (num_proc=4): 100%|██████████| 700/700 [00:00<00:00, 1226.62 examples/s]
Map (num_proc=4): 100%|██████████| 700/700 [00:01<00:00, 650.86 examples/s] 

Training Epoch: 14:   0%|          | 0/219 [00:00<?, ?it/s]
Training Epoch: 14:   0%|          | 0/219 [00:00<?, ?it/s]
Training Epoch: 14:   0%|          | 0/219 [00:00<?, ?it/s]
Training Epoch: 14:   0%|          | 0/219 [00:00<?, ?it/s]logging training data
rank 1 | gpu 1 | alloc=4.20 GB | reserved=47.74 GB | max=41.98 GB
rank 3 | gpu 3 | alloc=4.20 GB | reserved=51.43 GB | max=41.99 GB
rank 2 | gpu 2 | alloc=4.20 GB | reserved=47.73 GB | max=41.98 GB
rank 0 | gpu 0 | alloc=4.20 GB | reserved=53.37 GB | max=41.99 GB

Training Epoch: 14/100, batch 0/438 completed (loss: 0.0114:   0%|          | 0/219 [00:03<?, ?it/s]
Training Epoch: 14/100, batch 0/438 completed (loss: 0.1824:   0%|          | 0/219 [00:03<?, ?it/s]
Training Epoch: 14/100, batch 0/438 completed (loss: 0.2132:   0%|          | 0/219 [00:03<?, ?it/s]
Training Epoch: 14/100, batch 0/438 completed (loss: 0.1593:   0%|          | 0/219 [00:03<?, ?it/s]
Training Epoch: 14/100, batch 0/438 completed (loss: 0.1593:   0%|          | 1/219 [00:06<23:32,  6.48s/it]
Training Epoch: 14/100, batch 0/438 completed (loss: 0.1824:   0%|          | 1/219 [00:06<23:26,  6.45s/it]
Training Epoch: 14/100, batch 0/438 completed (loss: 0.2132:   0%|          | 1/219 [00:06<23:26,  6.45s/it]
Training Epoch: 14/100, batch 0/438 completed (loss: 0.0114:   0%|          | 1/219 [00:06<23:26,  6.45s/it]
Training Epoch: 14/100, batch 1/438 completed (loss: 0.006:   0%|          | 1/219 [00:06<23:26,  6.45s/it] 
Training Epoch: 14/100, batch 1/438 completed (loss: 0.017:   0%|          | 1/219 [00:06<23:26,  6.45s/it] 
Training Epoch: 14/100, batch 1/438 completed (loss: 0.1443:   0%|          | 1/219 [00:06<23:26,  6.45s/it]
Training Epoch: 14/100, batch 1/438 completed (loss: 0.0894:   0%|          | 1/219 [00:06<23:32,  6.48s/it]
Training Epoch: 14/100, batch 2/438 completed (loss: 0.0033:   0%|          | 1/219 [00:09<23:26,  6.45s/it]
Training Epoch: 14/100, batch 2/438 completed (loss: 0.0047:   0%|          | 1/219 [00:09<23:26,  6.45s/it]
Training Epoch: 14/100, batch 2/438 completed (loss: 0.0172:   0%|          | 1/219 [00:09<23:26,  6.45s/it]
Training Epoch: 14/100, batch 2/438 completed (loss: 0.0084:   0%|          | 1/219 [00:09<23:32,  6.48s/it]
Training Epoch: 14/100, batch 2/438 completed (loss: 0.0033:   1%|          | 2/219 [00:12<22:18,  6.17s/it]
Training Epoch: 14/100, batch 2/438 completed (loss: 0.0172:   1%|          | 2/219 [00:12<22:18,  6.17s/it]
Training Epoch: 14/100, batch 2/438 completed (loss: 0.0047:   1%|          | 2/219 [00:12<22:18,  6.17s/it]
Training Epoch: 14/100, batch 2/438 completed (loss: 0.0084:   1%|          | 2/219 [00:12<22:21,  6.18s/it]
Training Epoch: 14/100, batch 3/438 completed (loss: 0.4639:   1%|          | 2/219 [00:12<22:18,  6.17s/it]
Training Epoch: 14/100, batch 3/438 completed (loss: 0.3487:   1%|          | 2/219 [00:12<22:18,  6.17s/it]
Training Epoch: 14/100, batch 3/438 completed (loss: 0.0026:   1%|          | 2/219 [00:12<22:18,  6.17s/it]
Training Epoch: 14/100, batch 3/438 completed (loss: 0.0682:   1%|          | 2/219 [00:12<22:21,  6.18s/it]
Training Epoch: 14/100, batch 4/438 completed (loss: 0.0035:   1%|          | 2/219 [00:15<22:18,  6.17s/it]
Training Epoch: 14/100, batch 4/438 completed (loss: 0.3589:   1%|          | 2/219 [00:15<22:18,  6.17s/it]
Training Epoch: 14/100, batch 4/438 completed (loss: 0.0107:   1%|          | 2/219 [00:15<22:18,  6.17s/it]
Training Epoch: 14/100, batch 4/438 completed (loss: 0.0022:   1%|          | 2/219 [00:15<22:21,  6.18s/it]
Training Epoch: 14/100, batch 4/438 completed (loss: 0.0022:   1%|▏         | 3/219 [00:18<21:43,  6.03s/it]
Training Epoch: 14/100, batch 4/438 completed (loss: 0.0107:   1%|▏         | 3/219 [00:18<21:43,  6.03s/it]
Training Epoch: 14/100, batch 4/438 completed (loss: 0.3589:   1%|▏         | 3/219 [00:18<21:43,  6.03s/it]
Training Epoch: 14/100, batch 4/438 completed (loss: 0.0035:   1%|▏         | 3/219 [00:18<21:43,  6.04s/it]
Training Epoch: 14/100, batch 5/438 completed (loss: 0.0107:   1%|▏         | 3/219 [00:18<21:43,  6.04s/it]
Training Epoch: 14/100, batch 5/438 completed (loss: 0.0077:   1%|▏         | 3/219 [00:18<21:43,  6.03s/it]
Training Epoch: 14/100, batch 5/438 completed (loss: 0.0038:   1%|▏         | 3/219 [00:18<21:43,  6.03s/it]
Training Epoch: 14/100, batch 5/438 completed (loss: 0.1226:   1%|▏         | 3/219 [00:18<21:43,  6.03s/it]
Training Epoch: 14/100, batch 6/438 completed (loss: 0.0116:   1%|▏         | 3/219 [00:21<21:43,  6.03s/it]
Training Epoch: 14/100, batch 6/438 completed (loss: 0.081:   1%|▏         | 3/219 [00:21<21:43,  6.03s/it] 
Training Epoch: 14/100, batch 6/438 completed (loss: 0.0084:   1%|▏         | 3/219 [00:21<21:43,  6.04s/it]
Training Epoch: 14/100, batch 6/438 completed (loss: 0.0024:   1%|▏         | 3/219 [00:21<21:43,  6.03s/it]
Training Epoch: 14/100, batch 6/438 completed (loss: 0.0084:   2%|▏         | 4/219 [00:24<21:16,  5.94s/it]
Training Epoch: 14/100, batch 6/438 completed (loss: 0.0116:   2%|▏         | 4/219 [00:24<21:17,  5.94s/it]
Training Epoch: 14/100, batch 6/438 completed (loss: 0.081:   2%|▏         | 4/219 [00:24<21:17,  5.94s/it]
Training Epoch: 14/100, batch 6/438 completed (loss: 0.0024:   2%|▏         | 4/219 [00:24<21:18,  5.95s/it]
Training Epoch: 14/100, batch 7/438 completed (loss: 0.0027:   2%|▏         | 4/219 [00:24<21:16,  5.94s/it]
Training Epoch: 14/100, batch 7/438 completed (loss: 0.0007:   2%|▏         | 4/219 [00:24<21:17,  5.94s/it]
Training Epoch: 14/100, batch 7/438 completed (loss: 0.6978:   2%|▏         | 4/219 [00:24<21:17,  5.94s/it]
Training Epoch: 14/100, batch 7/438 completed (loss: 0.0012:   2%|▏         | 4/219 [00:24<21:18,  5.95s/it]
Training Epoch: 14/100, batch 8/438 completed (loss: 0.0043:   2%|▏         | 4/219 [00:27<21:16,  5.94s/it]
Training Epoch: 14/100, batch 8/438 completed (loss: 0.0718:   2%|▏         | 4/219 [00:27<21:17,  5.94s/it]
Training Epoch: 14/100, batch 8/438 completed (loss: 0.0031:   2%|▏         | 4/219 [00:27<21:17,  5.94s/it]
Training Epoch: 14/100, batch 8/438 completed (loss: 0.0007:   2%|▏         | 4/219 [00:27<21:18,  5.95s/it]
Training Epoch: 14/100, batch 8/438 completed (loss: 0.0007:   2%|▏         | 5/219 [00:30<21:15,  5.96s/it]
Training Epoch: 14/100, batch 8/438 completed (loss: 0.0031:   2%|▏         | 5/219 [00:30<21:15,  5.96s/it]
Training Epoch: 14/100, batch 8/438 completed (loss: 0.0718:   2%|▏         | 5/219 [00:30<21:15,  5.96s/it]
Training Epoch: 14/100, batch 8/438 completed (loss: 0.0043:   2%|▏         | 5/219 [00:30<21:15,  5.96s/it]
Training Epoch: 14/100, batch 9/438 completed (loss: 0.2012:   2%|▏         | 5/219 [00:30<21:15,  5.96s/it]
Training Epoch: 14/100, batch 9/438 completed (loss: 0.2355:   2%|▏         | 5/219 [00:30<21:15,  5.96s/it]
Training Epoch: 14/100, batch 9/438 completed (loss: 0.7923:   2%|▏         | 5/219 [00:30<21:15,  5.96s/it]
Training Epoch: 14/100, batch 9/438 completed (loss: 0.5335:   2%|▏         | 5/219 [00:30<21:15,  5.96s/it]
Training Epoch: 14/100, batch 10/438 completed (loss: 0.0062:   2%|▏         | 5/219 [00:32<21:15,  5.96s/it]
Training Epoch: 14/100, batch 10/438 completed (loss: 0.217:   2%|▏         | 5/219 [00:32<21:15,  5.96s/it]
Training Epoch: 14/100, batch 10/438 completed (loss: 0.0322:   2%|▏         | 5/219 [00:32<21:15,  5.96s/it]
Training Epoch: 14/100, batch 10/438 completed (loss: 0.0921:   2%|▏         | 5/219 [00:32<21:15,  5.96s/it]
Training Epoch: 14/100, batch 10/438 completed (loss: 0.0921:   3%|▎         | 6/219 [00:35<20:49,  5.86s/it]
Training Epoch: 14/100, batch 10/438 completed (loss: 0.0062:   3%|▎         | 6/219 [00:35<20:48,  5.86s/it]
Training Epoch: 14/100, batch 10/438 completed (loss: 0.0322:   3%|▎         | 6/219 [00:35<20:48,  5.86s/it]
Training Epoch: 14/100, batch 10/438 completed (loss: 0.217:   3%|▎         | 6/219 [00:35<20:48,  5.86s/it]
Training Epoch: 14/100, batch 11/438 completed (loss: 0.02:   3%|▎         | 6/219 [00:35<20:48,  5.86s/it]  
Training Epoch: 14/100, batch 11/438 completed (loss: 0.0173:   3%|▎         | 6/219 [00:35<20:48,  5.86s/it]
Training Epoch: 14/100, batch 11/438 completed (loss: 0.0122:   3%|▎         | 6/219 [00:35<20:48,  5.86s/it]
Training Epoch: 14/100, batch 11/438 completed (loss: 0.1943:   3%|▎         | 6/219 [00:35<20:49,  5.86s/it]
Training Epoch: 14/100, batch 12/438 completed (loss: 0.0077:   3%|▎         | 6/219 [00:38<20:48,  5.86s/it]
Training Epoch: 14/100, batch 12/438 completed (loss: 0.0717:   3%|▎         | 6/219 [00:38<20:48,  5.86s/it]
Training Epoch: 14/100, batch 12/438 completed (loss: 0.0457:   3%|▎         | 6/219 [00:38<20:48,  5.86s/it]
Training Epoch: 14/100, batch 12/438 completed (loss: 0.0093:   3%|▎         | 6/219 [00:38<20:49,  5.86s/it]
Training Epoch: 14/100, batch 12/438 completed (loss: 0.0093:   3%|▎         | 7/219 [00:41<20:40,  5.85s/it]
Training Epoch: 14/100, batch 12/438 completed (loss: 0.0457:   3%|▎         | 7/219 [00:41<20:39,  5.85s/it]
Training Epoch: 14/100, batch 12/438 completed (loss: 0.0077:   3%|▎         | 7/219 [00:41<20:39,  5.85s/it]
Training Epoch: 14/100, batch 12/438 completed (loss: 0.0717:   3%|▎         | 7/219 [00:41<20:40,  5.85s/it]
Training Epoch: 14/100, batch 13/438 completed (loss: 0.5482:   3%|▎         | 7/219 [00:41<20:39,  5.85s/it]
Training Epoch: 14/100, batch 13/438 completed (loss: 0.306:   3%|▎         | 7/219 [00:41<20:40,  5.85s/it] 
Training Epoch: 14/100, batch 13/438 completed (loss: 0.0854:   3%|▎         | 7/219 [00:41<20:39,  5.85s/it]
Training Epoch: 14/100, batch 13/438 completed (loss: 0.0179:   3%|▎         | 7/219 [00:41<20:40,  5.85s/it]
Training Epoch: 14/100, batch 14/438 completed (loss: 0.1817:   3%|▎         | 7/219 [00:44<20:39,  5.85s/it]
Training Epoch: 14/100, batch 14/438 completed (loss: 0.0105:   3%|▎         | 7/219 [00:44<20:39,  5.85s/it]
Training Epoch: 14/100, batch 14/438 completed (loss: 0.0342:   3%|▎         | 7/219 [00:44<20:40,  5.85s/it]
Training Epoch: 14/100, batch 14/438 completed (loss: 0.354:   3%|▎         | 7/219 [00:44<20:40,  5.85s/it] 
Training Epoch: 14/100, batch 14/438 completed (loss: 0.354:   4%|▎         | 8/219 [00:47<20:54,  5.94s/it]
Training Epoch: 14/100, batch 14/438 completed (loss: 0.0342:   4%|▎         | 8/219 [00:47<20:53,  5.94s/it]
Training Epoch: 14/100, batch 14/438 completed (loss: 0.1817:   4%|▎         | 8/219 [00:47<20:53,  5.94s/it]
Training Epoch: 14/100, batch 14/438 completed (loss: 0.0105:   4%|▎         | 8/219 [00:47<20:53,  5.94s/it]
Training Epoch: 14/100, batch 15/438 completed (loss: 0.0084:   4%|▎         | 8/219 [00:47<20:53,  5.94s/it]
Training Epoch: 14/100, batch 15/438 completed (loss: 0.0049:   4%|▎         | 8/219 [00:47<20:53,  5.94s/it]
Training Epoch: 14/100, batch 15/438 completed (loss: 0.271:   4%|▎         | 8/219 [00:47<20:53,  5.94s/it] 
Training Epoch: 14/100, batch 15/438 completed (loss: 0.2596:   4%|▎         | 8/219 [00:47<20:54,  5.94s/it]
Training Epoch: 14/100, batch 16/438 completed (loss: 0.0173:   4%|▎         | 8/219 [00:50<20:53,  5.94s/it]
Training Epoch: 14/100, batch 16/438 completed (loss: 0.0237:   4%|▎         | 8/219 [00:50<20:53,  5.94s/it]
Training Epoch: 14/100, batch 16/438 completed (loss: 0.2683:   4%|▎         | 8/219 [00:50<20:53,  5.94s/it]
Training Epoch: 14/100, batch 16/438 completed (loss: 0.0124:   4%|▎         | 8/219 [00:50<20:54,  5.94s/it]
Training Epoch: 14/100, batch 16/438 completed (loss: 0.0173:   4%|▍         | 9/219 [00:53<20:30,  5.86s/it]
Training Epoch: 14/100, batch 16/438 completed (loss: 0.0124:   4%|▍         | 9/219 [00:53<20:30,  5.86s/it]
Training Epoch: 14/100, batch 16/438 completed (loss: 0.0237:   4%|▍         | 9/219 [00:53<20:30,  5.86s/it]
Training Epoch: 14/100, batch 16/438 completed (loss: 0.2683:   4%|▍         | 9/219 [00:53<20:30,  5.86s/it]
Training Epoch: 14/100, batch 17/438 completed (loss: 0.3977:   4%|▍         | 9/219 [00:53<20:30,  5.86s/it]
Training Epoch: 14/100, batch 17/438 completed (loss: 0.0064:   4%|▍         | 9/219 [00:53<20:30,  5.86s/it]
Training Epoch: 14/100, batch 17/438 completed (loss: 0.2348:   4%|▍         | 9/219 [00:53<20:30,  5.86s/it]
Training Epoch: 14/100, batch 17/438 completed (loss: 0.0105:   4%|▍         | 9/219 [00:53<20:30,  5.86s/it]
Training Epoch: 14/100, batch 18/438 completed (loss: 0.1217:   4%|▍         | 9/219 [00:56<20:30,  5.86s/it]
Training Epoch: 14/100, batch 18/438 completed (loss: 0.045:   4%|▍         | 9/219 [00:56<20:30,  5.86s/it] 
Training Epoch: 14/100, batch 18/438 completed (loss: 0.0762:   4%|▍         | 9/219 [00:56<20:30,  5.86s/it]
Training Epoch: 14/100, batch 18/438 completed (loss: 0.0165:   4%|▍         | 9/219 [00:56<20:30,  5.86s/it]
Training Epoch: 14/100, batch 18/438 completed (loss: 0.0165:   5%|▍         | 10/219 [00:59<20:54,  6.00s/it]
Training Epoch: 14/100, batch 18/438 completed (loss: 0.0762:   5%|▍         | 10/219 [00:59<20:54,  6.00s/it]
Training Epoch: 14/100, batch 18/438 completed (loss: 0.1217:   5%|▍         | 10/219 [00:59<20:54,  6.00s/it]
Training Epoch: 14/100, batch 18/438 completed (loss: 0.045:   5%|▍         | 10/219 [00:59<20:54,  6.00s/it]
Training Epoch: 14/100, batch 19/438 completed (loss: 0.011:   5%|▍         | 10/219 [00:59<20:54,  6.00s/it]
Training Epoch: 14/100, batch 19/438 completed (loss: 0.0318:   5%|▍         | 10/219 [00:59<20:54,  6.00s/it]
Training Epoch: 14/100, batch 19/438 completed (loss: 0.1412:   5%|▍         | 10/219 [00:59<20:54,  6.00s/it]
Training Epoch: 14/100, batch 19/438 completed (loss: 0.1157:   5%|▍         | 10/219 [00:59<20:54,  6.00s/it]
Training Epoch: 14/100, batch 20/438 completed (loss: 0.0054:   5%|▍         | 10/219 [01:02<20:54,  6.00s/it]
Training Epoch: 14/100, batch 20/438 completed (loss: 0.0039:   5%|▍         | 10/219 [01:02<20:54,  6.00s/it]
Training Epoch: 14/100, batch 20/438 completed (loss: 0.4545:   5%|▍         | 10/219 [01:02<20:54,  6.00s/it]
Training Epoch: 14/100, batch 20/438 completed (loss: 0.0062:   5%|▍         | 10/219 [01:02<20:54,  6.00s/it]
Training Epoch: 14/100, batch 20/438 completed (loss: 0.0054:   5%|▌         | 11/219 [01:05<20:31,  5.92s/it]
Training Epoch: 14/100, batch 20/438 completed (loss: 0.0039:   5%|▌         | 11/219 [01:05<20:32,  5.92s/it]
Training Epoch: 14/100, batch 20/438 completed (loss: 0.0062:   5%|▌         | 11/219 [01:05<20:32,  5.93s/it]
Training Epoch: 14/100, batch 20/438 completed (loss: 0.4545:   5%|▌         | 11/219 [01:05<20:32,  5.93s/it]
Training Epoch: 14/100, batch 21/438 completed (loss: 0.0788:   5%|▌         | 11/219 [01:05<20:32,  5.92s/it]
Training Epoch: 14/100, batch 21/438 completed (loss: 0.0387:   5%|▌         | 11/219 [01:05<20:31,  5.92s/it]
Training Epoch: 14/100, batch 21/438 completed (loss: 0.1799:   5%|▌         | 11/219 [01:05<20:32,  5.93s/it]
Training Epoch: 14/100, batch 21/438 completed (loss: 0.2843:   5%|▌         | 11/219 [01:05<20:32,  5.93s/it]
Training Epoch: 14/100, batch 22/438 completed (loss: 0.0148:   5%|▌         | 11/219 [01:08<20:32,  5.92s/it]
Training Epoch: 14/100, batch 22/438 completed (loss: 0.0838:   5%|▌         | 11/219 [01:08<20:31,  5.92s/it]
Training Epoch: 14/100, batch 22/438 completed (loss: 0.0118:   5%|▌         | 11/219 [01:08<20:32,  5.93s/it]
Training Epoch: 14/100, batch 22/438 completed (loss: 0.0286:   5%|▌         | 11/219 [01:08<20:32,  5.93s/it]
Training Epoch: 14/100, batch 22/438 completed (loss: 0.0838:   5%|▌         | 12/219 [01:11<20:27,  5.93s/it]
Training Epoch: 14/100, batch 22/438 completed (loss: 0.0118:   5%|▌         | 12/219 [01:11<20:27,  5.93s/it]
Training Epoch: 14/100, batch 22/438 completed (loss: 0.0286:   5%|▌         | 12/219 [01:11<20:27,  5.93s/it]
Training Epoch: 14/100, batch 22/438 completed (loss: 0.0148:   5%|▌         | 12/219 [01:11<20:27,  5.93s/it]
Training Epoch: 14/100, batch 23/438 completed (loss: 0.1038:   5%|▌         | 12/219 [01:11<20:27,  5.93s/it]
Training Epoch: 14/100, batch 23/438 completed (loss: 0.0097:   5%|▌         | 12/219 [01:11<20:27,  5.93s/it]
Training Epoch: 14/100, batch 23/438 completed (loss: 0.0614:   5%|▌         | 12/219 [01:11<20:27,  5.93s/it]
Training Epoch: 14/100, batch 23/438 completed (loss: 0.0069:   5%|▌         | 12/219 [01:11<20:27,  5.93s/it]
Training Epoch: 14/100, batch 24/438 completed (loss: 0.4757:   5%|▌         | 12/219 [01:14<20:27,  5.93s/it]
Training Epoch: 14/100, batch 24/438 completed (loss: 0.0502:   5%|▌         | 12/219 [01:14<20:27,  5.93s/it]
Training Epoch: 14/100, batch 24/438 completed (loss: 0.2213:   5%|▌         | 12/219 [01:14<20:27,  5.93s/it]
Training Epoch: 14/100, batch 24/438 completed (loss: 0.0177:   5%|▌         | 12/219 [01:14<20:27,  5.93s/it]
Training Epoch: 14/100, batch 24/438 completed (loss: 0.0177:   6%|▌         | 13/219 [01:17<20:37,  6.01s/it]
Training Epoch: 14/100, batch 24/438 completed (loss: 0.0502:   6%|▌         | 13/219 [01:17<20:37,  6.01s/it]
Training Epoch: 14/100, batch 24/438 completed (loss: 0.4757:   6%|▌         | 13/219 [01:17<20:37,  6.01s/it]
Training Epoch: 14/100, batch 24/438 completed (loss: 0.2213:   6%|▌         | 13/219 [01:17<20:37,  6.01s/it]
Training Epoch: 14/100, batch 25/438 completed (loss: 0.0159:   6%|▌         | 13/219 [01:17<20:37,  6.01s/it]
Training Epoch: 14/100, batch 25/438 completed (loss: 0.006:   6%|▌         | 13/219 [01:17<20:37,  6.01s/it] 
Training Epoch: 14/100, batch 25/438 completed (loss: 0.0756:   6%|▌         | 13/219 [01:17<20:37,  6.01s/it]
Training Epoch: 14/100, batch 25/438 completed (loss: 0.2362:   6%|▌         | 13/219 [01:17<20:37,  6.01s/it]
Training Epoch: 14/100, batch 26/438 completed (loss: 0.0272:   6%|▌         | 13/219 [01:20<20:37,  6.01s/it]
Training Epoch: 14/100, batch 26/438 completed (loss: 0.3479:   6%|▌         | 13/219 [01:20<20:37,  6.01s/it]
Training Epoch: 14/100, batch 26/438 completed (loss: 0.0292:   6%|▌         | 13/219 [01:20<20:37,  6.01s/it]
Training Epoch: 14/100, batch 26/438 completed (loss: 0.0371:   6%|▌         | 13/219 [01:20<20:37,  6.01s/it]
Training Epoch: 14/100, batch 26/438 completed (loss: 0.0272:   6%|▋         | 14/219 [01:23<20:17,  5.94s/it]
Training Epoch: 14/100, batch 26/438 completed (loss: 0.0292:   6%|▋         | 14/219 [01:23<20:18,  5.94s/it]
Training Epoch: 14/100, batch 26/438 completed (loss: 0.0371:   6%|▋         | 14/219 [01:23<20:18,  5.94s/it]
Training Epoch: 14/100, batch 26/438 completed (loss: 0.3479:   6%|▋         | 14/219 [01:23<20:18,  5.95s/it]
Training Epoch: 14/100, batch 27/438 completed (loss: 0.0116:   6%|▋         | 14/219 [01:23<20:18,  5.95s/it]
Training Epoch: 14/100, batch 27/438 completed (loss: 0.042:   6%|▋         | 14/219 [01:23<20:17,  5.94s/it] 
Training Epoch: 14/100, batch 27/438 completed (loss: 0.0138:   6%|▋         | 14/219 [01:23<20:18,  5.94s/it]
Training Epoch: 14/100, batch 27/438 completed (loss: 0.0698:   6%|▋         | 14/219 [01:23<20:18,  5.94s/it]
Training Epoch: 14/100, batch 28/438 completed (loss: 0.0473:   6%|▋         | 14/219 [01:26<20:17,  5.94s/it]
Training Epoch: 14/100, batch 28/438 completed (loss: 0.1188:   6%|▋         | 14/219 [01:26<20:18,  5.94s/it]
Training Epoch: 14/100, batch 28/438 completed (loss: 0.0263:   6%|▋         | 14/219 [01:26<20:18,  5.95s/it]
Training Epoch: 14/100, batch 28/438 completed (loss: 0.0142:   6%|▋         | 14/219 [01:26<20:18,  5.94s/it]
Training Epoch: 14/100, batch 28/438 completed (loss: 0.0263:   7%|▋         | 15/219 [01:29<20:19,  5.98s/it]
Training Epoch: 14/100, batch 28/438 completed (loss: 0.0142:   7%|▋         | 15/219 [01:29<20:19,  5.98s/it]
Training Epoch: 14/100, batch 28/438 completed (loss: 0.1188:   7%|▋         | 15/219 [01:29<20:19,  5.98s/it]
Training Epoch: 14/100, batch 28/438 completed (loss: 0.0473:   7%|▋         | 15/219 [01:29<20:20,  5.98s/it]
Training Epoch: 14/100, batch 29/438 completed (loss: 0.1231:   7%|▋         | 15/219 [01:29<20:19,  5.98s/it]
Training Epoch: 14/100, batch 29/438 completed (loss: 0.0647:   7%|▋         | 15/219 [01:29<20:19,  5.98s/it]
Training Epoch: 14/100, batch 29/438 completed (loss: 0.026:   7%|▋         | 15/219 [01:29<20:20,  5.98s/it] 
Training Epoch: 14/100, batch 29/438 completed (loss: 0.0467:   7%|▋         | 15/219 [01:29<20:19,  5.98s/it]
Training Epoch: 14/100, batch 30/438 completed (loss: 0.0143:   7%|▋         | 15/219 [01:32<20:20,  5.98s/it]
Training Epoch: 14/100, batch 30/438 completed (loss: 0.2694:   7%|▋         | 15/219 [01:32<20:19,  5.98s/it]
Training Epoch: 14/100, batch 30/438 completed (loss: 0.0184:   7%|▋         | 15/219 [01:32<20:19,  5.98s/it]
Training Epoch: 14/100, batch 30/438 completed (loss: 0.5474:   7%|▋         | 15/219 [01:32<20:19,  5.98s/it]
Training Epoch: 14/100, batch 30/438 completed (loss: 0.0143:   7%|▋         | 16/219 [01:35<20:17,  6.00s/it]
Training Epoch: 14/100, batch 30/438 completed (loss: 0.5474:   7%|▋         | 16/219 [01:35<20:17,  6.00s/it]
Training Epoch: 14/100, batch 30/438 completed (loss: 0.0184:   7%|▋         | 16/219 [01:35<20:17,  6.00s/it]
Training Epoch: 14/100, batch 30/438 completed (loss: 0.2694:   7%|▋         | 16/219 [01:35<20:18,  6.00s/it]
Training Epoch: 14/100, batch 31/438 completed (loss: 0.0103:   7%|▋         | 16/219 [01:35<20:17,  6.00s/it]
Training Epoch: 14/100, batch 31/438 completed (loss: 0.0459:   7%|▋         | 16/219 [01:35<20:17,  6.00s/it]
Training Epoch: 14/100, batch 31/438 completed (loss: 0.0178:   7%|▋         | 16/219 [01:35<20:18,  6.00s/it]
Training Epoch: 14/100, batch 31/438 completed (loss: 0.0304:   7%|▋         | 16/219 [01:35<20:17,  6.00s/it]
Training Epoch: 14/100, batch 32/438 completed (loss: 0.0768:   7%|▋         | 16/219 [01:38<20:17,  6.00s/it]
Training Epoch: 14/100, batch 32/438 completed (loss: 0.022:   7%|▋         | 16/219 [01:38<20:18,  6.00s/it] 
Training Epoch: 14/100, batch 32/438 completed (loss: 0.0089:   7%|▋         | 16/219 [01:38<20:17,  6.00s/it]
Training Epoch: 14/100, batch 32/438 completed (loss: 0.0612:   7%|▋         | 16/219 [01:38<20:17,  6.00s/it]
Training Epoch: 14/100, batch 32/438 completed (loss: 0.0768:   8%|▊         | 17/219 [01:41<19:57,  5.93s/it]
Training Epoch: 14/100, batch 32/438 completed (loss: 0.022:   8%|▊         | 17/219 [01:41<19:57,  5.93s/it]
Training Epoch: 14/100, batch 32/438 completed (loss: 0.0089:   8%|▊         | 17/219 [01:41<19:57,  5.93s/it]
Training Epoch: 14/100, batch 32/438 completed (loss: 0.0612:   8%|▊         | 17/219 [01:41<19:57,  5.93s/it]
Training Epoch: 14/100, batch 33/438 completed (loss: 0.0379:   8%|▊         | 17/219 [01:41<19:57,  5.93s/it]
Training Epoch: 14/100, batch 33/438 completed (loss: 0.0069:   8%|▊         | 17/219 [01:41<19:57,  5.93s/it]
Training Epoch: 14/100, batch 33/438 completed (loss: 0.0096:   8%|▊         | 17/219 [01:41<19:57,  5.93s/it]
Training Epoch: 14/100, batch 33/438 completed (loss: 0.2255:   8%|▊         | 17/219 [01:41<19:57,  5.93s/it]
Training Epoch: 14/100, batch 34/438 completed (loss: 0.0857:   8%|▊         | 17/219 [01:44<19:57,  5.93s/it]
Training Epoch: 14/100, batch 34/438 completed (loss: 0.2509:   8%|▊         | 17/219 [01:44<19:57,  5.93s/it]
Training Epoch: 14/100, batch 34/438 completed (loss: 0.102:   8%|▊         | 17/219 [01:44<19:57,  5.93s/it] 
Training Epoch: 14/100, batch 34/438 completed (loss: 0.0166:   8%|▊         | 17/219 [01:44<19:57,  5.93s/it]
Training Epoch: 14/100, batch 34/438 completed (loss: 0.102:   8%|▊         | 18/219 [01:47<20:16,  6.05s/it]
Training Epoch: 14/100, batch 34/438 completed (loss: 0.0166:   8%|▊         | 18/219 [01:47<20:17,  6.06s/it]
Training Epoch: 14/100, batch 34/438 completed (loss: 0.0857:   8%|▊         | 18/219 [01:47<20:16,  6.05s/it]
Training Epoch: 14/100, batch 34/438 completed (loss: 0.2509:   8%|▊         | 18/219 [01:47<20:16,  6.05s/it]
Training Epoch: 14/100, batch 35/438 completed (loss: 0.0298:   8%|▊         | 18/219 [01:47<20:16,  6.05s/it]
Training Epoch: 14/100, batch 35/438 completed (loss: 0.0068:   8%|▊         | 18/219 [01:47<20:16,  6.05s/it]
Training Epoch: 14/100, batch 35/438 completed (loss: 0.0115:   8%|▊         | 18/219 [01:47<20:16,  6.05s/it]
Training Epoch: 14/100, batch 35/438 completed (loss: 0.0386:   8%|▊         | 18/219 [01:47<20:17,  6.06s/it]
Training Epoch: 14/100, batch 36/438 completed (loss: 0.0172:   8%|▊         | 18/219 [01:50<20:16,  6.05s/it]
Training Epoch: 14/100, batch 36/438 completed (loss: 0.1422:   8%|▊         | 18/219 [01:50<20:16,  6.05s/it]
Training Epoch: 14/100, batch 36/438 completed (loss: 0.063:   8%|▊         | 18/219 [01:50<20:16,  6.05s/it] 
Training Epoch: 14/100, batch 36/438 completed (loss: 0.0921:   8%|▊         | 18/219 [01:50<20:17,  6.06s/it]
Training Epoch: 14/100, batch 36/438 completed (loss: 0.0172:   9%|▊         | 19/219 [01:53<19:56,  5.98s/it]
Training Epoch: 14/100, batch 36/438 completed (loss: 0.0921:   9%|▊         | 19/219 [01:53<19:57,  5.99s/it]
Training Epoch: 14/100, batch 36/438 completed (loss: 0.063:   9%|▊         | 19/219 [01:53<19:57,  5.99s/it]
Training Epoch: 14/100, batch 36/438 completed (loss: 0.1422:   9%|▊         | 19/219 [01:53<19:57,  5.99s/it]
Training Epoch: 14/100, batch 37/438 completed (loss: 0.0515:   9%|▊         | 19/219 [01:53<19:57,  5.99s/it]
Training Epoch: 14/100, batch 37/438 completed (loss: 0.1301:   9%|▊         | 19/219 [01:53<19:57,  5.99s/it]
Training Epoch: 14/100, batch 37/438 completed (loss: 0.007:   9%|▊         | 19/219 [01:53<19:56,  5.98s/it] 
Training Epoch: 14/100, batch 37/438 completed (loss: 0.0521:   9%|▊         | 19/219 [01:53<19:57,  5.99s/it]
Training Epoch: 14/100, batch 38/438 completed (loss: 0.0025:   9%|▊         | 19/219 [01:56<19:57,  5.99s/it]
Training Epoch: 14/100, batch 38/438 completed (loss: 0.0205:   9%|▊         | 19/219 [01:56<19:57,  5.99s/it]
Training Epoch: 14/100, batch 38/438 completed (loss: 0.0199:   9%|▊         | 19/219 [01:56<19:56,  5.98s/it]
Training Epoch: 14/100, batch 38/438 completed (loss: 0.1355:   9%|▊         | 19/219 [01:56<19:57,  5.99s/it]
Training Epoch: 14/100, batch 38/438 completed (loss: 0.0205:   9%|▉         | 20/219 [01:59<19:48,  5.97s/it]
Training Epoch: 14/100, batch 38/438 completed (loss: 0.1355:   9%|▉         | 20/219 [01:59<19:48,  5.97s/it]
Training Epoch: 14/100, batch 38/438 completed (loss: 0.0199:   9%|▉         | 20/219 [01:59<19:48,  5.97s/it]
Training Epoch: 14/100, batch 38/438 completed (loss: 0.0025:   9%|▉         | 20/219 [01:59<19:48,  5.97s/it]
Training Epoch: 14/100, batch 39/438 completed (loss: 0.021:   9%|▉         | 20/219 [01:59<19:48,  5.97s/it] 
Training Epoch: 14/100, batch 39/438 completed (loss: 0.0568:   9%|▉         | 20/219 [01:59<19:48,  5.97s/it]
Training Epoch: 14/100, batch 39/438 completed (loss: 0.2432:   9%|▉         | 20/219 [01:59<19:48,  5.97s/it]
Training Epoch: 14/100, batch 39/438 completed (loss: 0.0428:   9%|▉         | 20/219 [01:59<19:48,  5.97s/it]
Training Epoch: 14/100, batch 40/438 completed (loss: 0.0376:   9%|▉         | 20/219 [02:02<19:48,  5.97s/it]
Training Epoch: 14/100, batch 40/438 completed (loss: 0.0156:   9%|▉         | 20/219 [02:02<19:48,  5.97s/it]
Training Epoch: 14/100, batch 40/438 completed (loss: 0.0334:   9%|▉         | 20/219 [02:02<19:48,  5.97s/it]
Training Epoch: 14/100, batch 40/438 completed (loss: 0.0396:   9%|▉         | 20/219 [02:02<19:48,  5.97s/it]
Training Epoch: 14/100, batch 40/438 completed (loss: 0.0334:  10%|▉         | 21/219 [02:05<19:43,  5.97s/it]
Training Epoch: 14/100, batch 40/438 completed (loss: 0.0376:  10%|▉         | 21/219 [02:05<19:43,  5.98s/it]
Training Epoch: 14/100, batch 40/438 completed (loss: 0.0156:  10%|▉         | 21/219 [02:05<19:43,  5.98s/it]
Training Epoch: 14/100, batch 40/438 completed (loss: 0.0396:  10%|▉         | 21/219 [02:05<19:43,  5.98s/it]
Training Epoch: 14/100, batch 41/438 completed (loss: 0.2485:  10%|▉         | 21/219 [02:05<19:43,  5.97s/it]
Training Epoch: 14/100, batch 41/438 completed (loss: 0.0566:  10%|▉         | 21/219 [02:05<19:43,  5.98s/it]
Training Epoch: 14/100, batch 41/438 completed (loss: 0.0279:  10%|▉         | 21/219 [02:05<19:43,  5.98s/it]
Training Epoch: 14/100, batch 41/438 completed (loss: 0.0347:  10%|▉         | 21/219 [02:05<19:43,  5.98s/it]
Training Epoch: 14/100, batch 42/438 completed (loss: 0.037:  10%|▉         | 21/219 [02:08<19:43,  5.98s/it] 
Training Epoch: 14/100, batch 42/438 completed (loss: 0.0564:  10%|▉         | 21/219 [02:08<19:43,  5.98s/it]
Training Epoch: 14/100, batch 42/438 completed (loss: 0.0174:  10%|▉         | 21/219 [02:08<19:43,  5.97s/it]
Training Epoch: 14/100, batch 42/438 completed (loss: 0.1965:  10%|▉         | 21/219 [02:08<19:43,  5.98s/it]
Training Epoch: 14/100, batch 42/438 completed (loss: 0.0564:  10%|█         | 22/219 [02:11<19:27,  5.92s/it]
Training Epoch: 14/100, batch 42/438 completed (loss: 0.0174:  10%|█         | 22/219 [02:11<19:27,  5.92s/it]
Training Epoch: 14/100, batch 42/438 completed (loss: 0.1965:  10%|█         | 22/219 [02:11<19:27,  5.93s/it]
Training Epoch: 14/100, batch 42/438 completed (loss: 0.037:  10%|█         | 22/219 [02:11<19:27,  5.92s/it]
Training Epoch: 14/100, batch 43/438 completed (loss: 0.404:  10%|█         | 22/219 [02:11<19:27,  5.92s/it] 
Training Epoch: 14/100, batch 43/438 completed (loss: 0.0242:  10%|█         | 22/219 [02:11<19:27,  5.92s/it]
Training Epoch: 14/100, batch 43/438 completed (loss: 0.0938:  10%|█         | 22/219 [02:11<19:27,  5.92s/it]
Training Epoch: 14/100, batch 43/438 completed (loss: 0.0169:  10%|█         | 22/219 [02:11<19:27,  5.93s/it]
Training Epoch: 14/100, batch 44/438 completed (loss: 0.2492:  10%|█         | 22/219 [02:14<19:27,  5.92s/it]
Training Epoch: 14/100, batch 44/438 completed (loss: 0.0352:  10%|█         | 22/219 [02:14<19:27,  5.92s/it]
Training Epoch: 14/100, batch 44/438 completed (loss: 0.0104:  10%|█         | 22/219 [02:14<19:27,  5.92s/it]
Training Epoch: 14/100, batch 44/438 completed (loss: 0.1019:  10%|█         | 22/219 [02:14<19:27,  5.93s/it]
Training Epoch: 14/100, batch 44/438 completed (loss: 0.1019:  11%|█         | 23/219 [02:17<19:43,  6.04s/it]
Training Epoch: 14/100, batch 44/438 completed (loss: 0.2492:  11%|█         | 23/219 [02:17<19:43,  6.04s/it]
Training Epoch: 14/100, batch 44/438 completed (loss: 0.0352:  11%|█         | 23/219 [02:17<19:43,  6.04s/it]
Training Epoch: 14/100, batch 44/438 completed (loss: 0.0104:  11%|█         | 23/219 [02:17<19:43,  6.04s/it]
Training Epoch: 14/100, batch 45/438 completed (loss: 0.0135:  11%|█         | 23/219 [02:17<19:43,  6.04s/it]
Training Epoch: 14/100, batch 45/438 completed (loss: 0.0089:  11%|█         | 23/219 [02:17<19:43,  6.04s/it]
Training Epoch: 14/100, batch 45/438 completed (loss: 0.0279:  11%|█         | 23/219 [02:17<19:43,  6.04s/it]
Training Epoch: 14/100, batch 45/438 completed (loss: 0.0429:  11%|█         | 23/219 [02:17<19:43,  6.04s/it]
Training Epoch: 14/100, batch 46/438 completed (loss: 0.1325:  11%|█         | 23/219 [02:20<19:43,  6.04s/it]
Training Epoch: 14/100, batch 46/438 completed (loss: 0.0106:  11%|█         | 23/219 [02:20<19:43,  6.04s/it]
Training Epoch: 14/100, batch 46/438 completed (loss: 0.099:  11%|█         | 23/219 [02:20<19:43,  6.04s/it] 
Training Epoch: 14/100, batch 46/438 completed (loss: 0.021:  11%|█         | 23/219 [02:20<19:43,  6.04s/it] 
Training Epoch: 14/100, batch 46/438 completed (loss: 0.021:  11%|█         | 24/219 [02:23<19:22,  5.96s/it]
Training Epoch: 14/100, batch 46/438 completed (loss: 0.099:  11%|█         | 24/219 [02:23<19:22,  5.96s/it]
Training Epoch: 14/100, batch 46/438 completed (loss: 0.1325:  11%|█         | 24/219 [02:23<19:22,  5.96s/it]
Training Epoch: 14/100, batch 46/438 completed (loss: 0.0106:  11%|█         | 24/219 [02:23<19:22,  5.96s/it]
Training Epoch: 14/100, batch 47/438 completed (loss: 0.0203:  11%|█         | 24/219 [02:23<19:22,  5.96s/it]
Training Epoch: 14/100, batch 47/438 completed (loss: 0.0451:  11%|█         | 24/219 [02:23<19:22,  5.96s/it]
Training Epoch: 14/100, batch 47/438 completed (loss: 0.0277:  11%|█         | 24/219 [02:23<19:22,  5.96s/it]
Training Epoch: 14/100, batch 47/438 completed (loss: 0.0707:  11%|█         | 24/219 [02:23<19:22,  5.96s/it]
Training Epoch: 14/100, batch 48/438 completed (loss: 0.0495:  11%|█         | 24/219 [02:26<19:22,  5.96s/it]
Training Epoch: 14/100, batch 48/438 completed (loss: 0.239:  11%|█         | 24/219 [02:26<19:22,  5.96s/it] 
Training Epoch: 14/100, batch 48/438 completed (loss: 0.0227:  11%|█         | 24/219 [02:26<19:22,  5.96s/it]
Training Epoch: 14/100, batch 48/438 completed (loss: 0.0667:  11%|█         | 24/219 [02:26<19:22,  5.96s/it]
Training Epoch: 14/100, batch 48/438 completed (loss: 0.0667:  11%|█▏        | 25/219 [02:29<19:06,  5.91s/it]
Training Epoch: 14/100, batch 48/438 completed (loss: 0.0227:  11%|█▏        | 25/219 [02:29<19:07,  5.91s/it]
Training Epoch: 14/100, batch 48/438 completed (loss: 0.239:  11%|█▏        | 25/219 [02:29<19:07,  5.91s/it]
Training Epoch: 14/100, batch 48/438 completed (loss: 0.0495:  11%|█▏        | 25/219 [02:29<19:07,  5.92s/it]
Training Epoch: 14/100, batch 49/438 completed (loss: 0.2222:  11%|█▏        | 25/219 [02:29<19:07,  5.91s/it]
Training Epoch: 14/100, batch 49/438 completed (loss: 0.0379:  11%|█▏        | 25/219 [02:29<19:07,  5.92s/it]
Training Epoch: 14/100, batch 49/438 completed (loss: 0.053:  11%|█▏        | 25/219 [02:29<19:07,  5.91s/it] 
Training Epoch: 14/100, batch 49/438 completed (loss: 0.0496:  11%|█▏        | 25/219 [02:29<19:06,  5.91s/it]
Training Epoch: 14/100, batch 50/438 completed (loss: 0.0463:  11%|█▏        | 25/219 [02:32<19:07,  5.91s/it]
Training Epoch: 14/100, batch 50/438 completed (loss: 0.247:  11%|█▏        | 25/219 [02:32<19:07,  5.92s/it] 
Training Epoch: 14/100, batch 50/438 completed (loss: 0.0144:  11%|█▏        | 25/219 [02:32<19:07,  5.91s/it]
Training Epoch: 14/100, batch 50/438 completed (loss: 0.0106:  11%|█▏        | 25/219 [02:32<19:06,  5.91s/it]
Training Epoch: 14/100, batch 50/438 completed (loss: 0.0106:  12%|█▏        | 26/219 [02:35<19:30,  6.06s/it]
Training Epoch: 14/100, batch 50/438 completed (loss: 0.0144:  12%|█▏        | 26/219 [02:35<19:30,  6.06s/it]
Training Epoch: 14/100, batch 50/438 completed (loss: 0.247:  12%|█▏        | 26/219 [02:35<19:30,  6.06s/it]
Training Epoch: 14/100, batch 50/438 completed (loss: 0.0463:  12%|█▏        | 26/219 [02:35<19:30,  6.06s/it]
Training Epoch: 14/100, batch 51/438 completed (loss: 0.1631:  12%|█▏        | 26/219 [02:35<19:30,  6.06s/it]
Training Epoch: 14/100, batch 51/438 completed (loss: 0.2515:  12%|█▏        | 26/219 [02:35<19:30,  6.06s/it]
Training Epoch: 14/100, batch 51/438 completed (loss: 0.0514:  12%|█▏        | 26/219 [02:35<19:30,  6.06s/it]
Training Epoch: 14/100, batch 51/438 completed (loss: 0.0162:  12%|█▏        | 26/219 [02:35<19:30,  6.06s/it]
Training Epoch: 14/100, batch 52/438 completed (loss: 0.2537:  12%|█▏        | 26/219 [02:38<19:30,  6.06s/it]
Training Epoch: 14/100, batch 52/438 completed (loss: 0.0438:  12%|█▏        | 26/219 [02:38<19:30,  6.06s/it]
Training Epoch: 14/100, batch 52/438 completed (loss: 0.0861:  12%|█▏        | 26/219 [02:38<19:30,  6.06s/it]
Training Epoch: 14/100, batch 52/438 completed (loss: 0.0315:  12%|█▏        | 26/219 [02:38<19:30,  6.06s/it]
Training Epoch: 14/100, batch 52/438 completed (loss: 0.0438:  12%|█▏        | 27/219 [02:41<19:08,  5.98s/it]
Training Epoch: 14/100, batch 52/438 completed (loss: 0.2537:  12%|█▏        | 27/219 [02:41<19:08,  5.98s/it]
Training Epoch: 14/100, batch 52/438 completed (loss: 0.0315:  12%|█▏        | 27/219 [02:41<19:08,  5.98s/it]
Training Epoch: 14/100, batch 52/438 completed (loss: 0.0861:  12%|█▏        | 27/219 [02:41<19:08,  5.98s/it]
Training Epoch: 14/100, batch 53/438 completed (loss: 0.0332:  12%|█▏        | 27/219 [02:41<19:08,  5.98s/it]
Training Epoch: 14/100, batch 53/438 completed (loss: 0.0258:  12%|█▏        | 27/219 [02:41<19:08,  5.98s/it]
Training Epoch: 14/100, batch 53/438 completed (loss: 0.0341:  12%|█▏        | 27/219 [02:41<19:08,  5.98s/it]
Training Epoch: 14/100, batch 53/438 completed (loss: 0.008:  12%|█▏        | 27/219 [02:41<19:08,  5.98s/it] 
Training Epoch: 14/100, batch 54/438 completed (loss: 0.4379:  12%|█▏        | 27/219 [02:44<19:08,  5.98s/it]
Training Epoch: 14/100, batch 54/438 completed (loss: 0.0322:  12%|█▏        | 27/219 [02:44<19:08,  5.98s/it]
Training Epoch: 14/100, batch 54/438 completed (loss: 0.0034:  12%|█▏        | 27/219 [02:44<19:08,  5.98s/it]
Training Epoch: 14/100, batch 54/438 completed (loss: 0.0767:  12%|█▏        | 27/219 [02:44<19:08,  5.98s/it]
Training Epoch: 14/100, batch 54/438 completed (loss: 0.0767:  13%|█▎        | 28/219 [02:47<19:19,  6.07s/it]
Training Epoch: 14/100, batch 54/438 completed (loss: 0.0322:  13%|█▎        | 28/219 [02:47<19:19,  6.07s/it]
Training Epoch: 14/100, batch 54/438 completed (loss: 0.0034:  13%|█▎        | 28/219 [02:47<19:19,  6.07s/it]
Training Epoch: 14/100, batch 54/438 completed (loss: 0.4379:  13%|█▎        | 28/219 [02:47<19:19,  6.07s/it]
Training Epoch: 14/100, batch 55/438 completed (loss: 0.0053:  13%|█▎        | 28/219 [02:47<19:19,  6.07s/it]
Training Epoch: 14/100, batch 55/438 completed (loss: 0.0281:  13%|█▎        | 28/219 [02:47<19:19,  6.07s/it]
Training Epoch: 14/100, batch 55/438 completed (loss: 0.0223:  13%|█▎        | 28/219 [02:47<19:19,  6.07s/it]
Training Epoch: 14/100, batch 55/438 completed (loss: 0.1382:  13%|█▎        | 28/219 [02:47<19:19,  6.07s/it]
Training Epoch: 14/100, batch 56/438 completed (loss: 0.1951:  13%|█▎        | 28/219 [02:50<19:19,  6.07s/it]
Training Epoch: 14/100, batch 56/438 completed (loss: 0.0315:  13%|█▎        | 28/219 [02:50<19:19,  6.07s/it]
Training Epoch: 14/100, batch 56/438 completed (loss: 0.0207:  13%|█▎        | 28/219 [02:50<19:19,  6.07s/it]
Training Epoch: 14/100, batch 56/438 completed (loss: 0.0065:  13%|█▎        | 28/219 [02:50<19:19,  6.07s/it]
Training Epoch: 14/100, batch 56/438 completed (loss: 0.0315:  13%|█▎        | 29/219 [02:53<19:07,  6.04s/it]
Training Epoch: 14/100, batch 56/438 completed (loss: 0.0207:  13%|█▎        | 29/219 [02:53<19:07,  6.04s/it]
Training Epoch: 14/100, batch 56/438 completed (loss: 0.0065:  13%|█▎        | 29/219 [02:53<19:07,  6.04s/it]
Training Epoch: 14/100, batch 56/438 completed (loss: 0.1951:  13%|█▎        | 29/219 [02:53<19:07,  6.04s/it]
Training Epoch: 14/100, batch 57/438 completed (loss: 0.028:  13%|█▎        | 29/219 [02:53<19:07,  6.04s/it] 
Training Epoch: 14/100, batch 57/438 completed (loss: 0.0124:  13%|█▎        | 29/219 [02:53<19:07,  6.04s/it]
Training Epoch: 14/100, batch 57/438 completed (loss: 0.4034:  13%|█▎        | 29/219 [02:53<19:07,  6.04s/it]
Training Epoch: 14/100, batch 57/438 completed (loss: 0.0283:  13%|█▎        | 29/219 [02:53<19:07,  6.04s/it]
Training Epoch: 14/100, batch 58/438 completed (loss: 0.0261:  13%|█▎        | 29/219 [02:56<19:07,  6.04s/it]
Training Epoch: 14/100, batch 58/438 completed (loss: 0.008:  13%|█▎        | 29/219 [02:56<19:07,  6.04s/it] 
Training Epoch: 14/100, batch 58/438 completed (loss: 0.2906:  13%|█▎        | 29/219 [02:56<19:07,  6.04s/it]
Training Epoch: 14/100, batch 58/438 completed (loss: 0.0867:  13%|█▎        | 29/219 [02:56<19:07,  6.04s/it]
Training Epoch: 14/100, batch 58/438 completed (loss: 0.0261:  14%|█▎        | 30/219 [02:59<18:46,  5.96s/it]
Training Epoch: 14/100, batch 58/438 completed (loss: 0.2906:  14%|█▎        | 30/219 [02:59<18:46,  5.96s/it]
Training Epoch: 14/100, batch 58/438 completed (loss: 0.008:  14%|█▎        | 30/219 [02:59<18:47,  5.96s/it]
Training Epoch: 14/100, batch 58/438 completed (loss: 0.0867:  14%|█▎        | 30/219 [02:59<18:47,  5.96s/it]
Training Epoch: 14/100, batch 59/438 completed (loss: 0.1232:  14%|█▎        | 30/219 [02:59<18:46,  5.96s/it]
Training Epoch: 14/100, batch 59/438 completed (loss: 0.2263:  14%|█▎        | 30/219 [02:59<18:47,  5.96s/it]
Training Epoch: 14/100, batch 59/438 completed (loss: 0.3167:  14%|█▎        | 30/219 [02:59<18:46,  5.96s/it]
Training Epoch: 14/100, batch 59/438 completed (loss: 0.0184:  14%|█▎        | 30/219 [02:59<18:47,  5.96s/it]
Training Epoch: 14/100, batch 60/438 completed (loss: 0.0744:  14%|█▎        | 30/219 [03:02<18:46,  5.96s/it]
Training Epoch: 14/100, batch 60/438 completed (loss: 0.0196:  14%|█▎        | 30/219 [03:02<18:46,  5.96s/it]
Training Epoch: 14/100, batch 60/438 completed (loss: 0.0823:  14%|█▎        | 30/219 [03:02<18:47,  5.96s/it]
Training Epoch: 14/100, batch 60/438 completed (loss: 0.0683:  14%|█▎        | 30/219 [03:02<18:47,  5.96s/it]
Training Epoch: 14/100, batch 60/438 completed (loss: 0.0744:  14%|█▍        | 31/219 [03:05<19:07,  6.10s/it]
Training Epoch: 14/100, batch 60/438 completed (loss: 0.0196:  14%|█▍        | 31/219 [03:05<19:07,  6.10s/it]
Training Epoch: 14/100, batch 60/438 completed (loss: 0.0823:  14%|█▍        | 31/219 [03:05<19:07,  6.10s/it]
Training Epoch: 14/100, batch 60/438 completed (loss: 0.0683:  14%|█▍        | 31/219 [03:05<19:07,  6.11s/it]
Training Epoch: 14/100, batch 61/438 completed (loss: 0.0636:  14%|█▍        | 31/219 [03:05<19:07,  6.10s/it]
Training Epoch: 14/100, batch 61/438 completed (loss: 0.044:  14%|█▍        | 31/219 [03:05<19:07,  6.10s/it] 
Training Epoch: 14/100, batch 61/438 completed (loss: 0.121:  14%|█▍        | 31/219 [03:05<19:07,  6.10s/it] 
Training Epoch: 14/100, batch 61/438 completed (loss: 0.0046:  14%|█▍        | 31/219 [03:05<19:07,  6.11s/it]
Training Epoch: 14/100, batch 62/438 completed (loss: 0.0108:  14%|█▍        | 31/219 [03:08<19:07,  6.10s/it]
Training Epoch: 14/100, batch 62/438 completed (loss: 0.0205:  14%|█▍        | 31/219 [03:08<19:07,  6.10s/it]
Training Epoch: 14/100, batch 62/438 completed (loss: 0.0541:  14%|█▍        | 31/219 [03:08<19:07,  6.10s/it]
Training Epoch: 14/100, batch 62/438 completed (loss: 0.0405:  14%|█▍        | 31/219 [03:08<19:07,  6.11s/it]
Training Epoch: 14/100, batch 62/438 completed (loss: 0.0108:  15%|█▍        | 32/219 [03:11<18:44,  6.02s/it]
Training Epoch: 14/100, batch 62/438 completed (loss: 0.0205:  15%|█▍        | 32/219 [03:11<18:45,  6.02s/it]
Training Epoch: 14/100, batch 62/438 completed (loss: 0.0541:  15%|█▍        | 32/219 [03:11<18:44,  6.02s/it]
Training Epoch: 14/100, batch 62/438 completed (loss: 0.0405:  15%|█▍        | 32/219 [03:11<18:45,  6.02s/it]
Training Epoch: 14/100, batch 63/438 completed (loss: 0.035:  15%|█▍        | 32/219 [03:11<18:44,  6.02s/it] 
Training Epoch: 14/100, batch 63/438 completed (loss: 0.0482:  15%|█▍        | 32/219 [03:11<18:44,  6.02s/it]
Training Epoch: 14/100, batch 63/438 completed (loss: 0.2873:  15%|█▍        | 32/219 [03:11<18:45,  6.02s/it]
Training Epoch: 14/100, batch 63/438 completed (loss: 0.0073:  15%|█▍        | 32/219 [03:11<18:45,  6.02s/it]
Training Epoch: 14/100, batch 64/438 completed (loss: 0.099:  15%|█▍        | 32/219 [03:14<18:45,  6.02s/it] 
Training Epoch: 14/100, batch 64/438 completed (loss: 0.0787:  15%|█▍        | 32/219 [03:14<18:44,  6.02s/it]
Training Epoch: 14/100, batch 64/438 completed (loss: 0.0062:  15%|█▍        | 32/219 [03:14<18:44,  6.02s/it]
Training Epoch: 14/100, batch 64/438 completed (loss: 0.1628:  15%|█▍        | 32/219 [03:14<18:45,  6.02s/it]
Training Epoch: 14/100, batch 64/438 completed (loss: 0.0062:  15%|█▌        | 33/219 [03:17<18:36,  6.00s/it]
Training Epoch: 14/100, batch 64/438 completed (loss: 0.0787:  15%|█▌        | 33/219 [03:17<18:36,  6.00s/it]
Training Epoch: 14/100, batch 64/438 completed (loss: 0.099:  15%|█▌        | 33/219 [03:17<18:36,  6.00s/it]
Training Epoch: 14/100, batch 64/438 completed (loss: 0.1628:  15%|█▌        | 33/219 [03:17<18:36,  6.00s/it]
Training Epoch: 14/100, batch 65/438 completed (loss: 0.1057:  15%|█▌        | 33/219 [03:17<18:36,  6.00s/it]
Training Epoch: 14/100, batch 65/438 completed (loss: 0.0141:  15%|█▌        | 33/219 [03:17<18:36,  6.00s/it]
Training Epoch: 14/100, batch 65/438 completed (loss: 0.0219:  15%|█▌        | 33/219 [03:17<18:36,  6.00s/it]
Training Epoch: 14/100, batch 65/438 completed (loss: 0.0054:  15%|█▌        | 33/219 [03:17<18:36,  6.00s/it]
Training Epoch: 14/100, batch 66/438 completed (loss: 0.066:  15%|█▌        | 33/219 [03:20<18:36,  6.00s/it] 
Training Epoch: 14/100, batch 66/438 completed (loss: 0.0231:  15%|█▌        | 33/219 [03:20<18:36,  6.00s/it]
Training Epoch: 14/100, batch 66/438 completed (loss: 0.0393:  15%|█▌        | 33/219 [03:20<18:36,  6.00s/it]
Training Epoch: 14/100, batch 66/438 completed (loss: 0.1038:  15%|█▌        | 33/219 [03:20<18:36,  6.00s/it]
Training Epoch: 14/100, batch 66/438 completed (loss: 0.0393:  16%|█▌        | 34/219 [03:23<18:44,  6.08s/it]
Training Epoch: 14/100, batch 66/438 completed (loss: 0.066:  16%|█▌        | 34/219 [03:23<18:44,  6.08s/it]
Training Epoch: 14/100, batch 66/438 completed (loss: 0.0231:  16%|█▌        | 34/219 [03:23<18:44,  6.08s/it]
Training Epoch: 14/100, batch 66/438 completed (loss: 0.1038:  16%|█▌        | 34/219 [03:23<18:44,  6.08s/it]
Training Epoch: 14/100, batch 67/438 completed (loss: 0.0447:  16%|█▌        | 34/219 [03:23<18:44,  6.08s/it]
Training Epoch: 14/100, batch 67/438 completed (loss: 0.0044:  16%|█▌        | 34/219 [03:23<18:44,  6.08s/it]
Training Epoch: 14/100, batch 67/438 completed (loss: 0.7706:  16%|█▌        | 34/219 [03:23<18:44,  6.08s/it]
Training Epoch: 14/100, batch 67/438 completed (loss: 0.3378:  16%|█▌        | 34/219 [03:23<18:44,  6.08s/it]
Training Epoch: 14/100, batch 68/438 completed (loss: 0.0602:  16%|█▌        | 34/219 [03:26<18:44,  6.08s/it]
Training Epoch: 14/100, batch 68/438 completed (loss: 0.0219:  16%|█▌        | 34/219 [03:26<18:44,  6.08s/it]
Training Epoch: 14/100, batch 68/438 completed (loss: 0.1189:  16%|█▌        | 34/219 [03:26<18:44,  6.08s/it]
Training Epoch: 14/100, batch 68/438 completed (loss: 0.2428:  16%|█▌        | 34/219 [03:26<18:44,  6.08s/it]
Training Epoch: 14/100, batch 68/438 completed (loss: 0.0219:  16%|█▌        | 35/219 [03:29<18:23,  6.00s/it]
Training Epoch: 14/100, batch 68/438 completed (loss: 0.2428:  16%|█▌        | 35/219 [03:29<18:23,  6.00s/it]
Training Epoch: 14/100, batch 68/438 completed (loss: 0.1189:  16%|█▌        | 35/219 [03:29<18:23,  6.00s/it]
Training Epoch: 14/100, batch 68/438 completed (loss: 0.0602:  16%|█▌        | 35/219 [03:29<18:23,  6.00s/it]
Training Epoch: 14/100, batch 69/438 completed (loss: 0.0273:  16%|█▌        | 35/219 [03:29<18:23,  6.00s/it]
Training Epoch: 14/100, batch 69/438 completed (loss: 0.3914:  16%|█▌        | 35/219 [03:29<18:23,  6.00s/it]
Training Epoch: 14/100, batch 69/438 completed (loss: 0.001:  16%|█▌        | 35/219 [03:29<18:23,  6.00s/it] 
Training Epoch: 14/100, batch 69/438 completed (loss: 0.1031:  16%|█▌        | 35/219 [03:29<18:23,  6.00s/it]
Training Epoch: 14/100, batch 70/438 completed (loss: 0.0484:  16%|█▌        | 35/219 [03:32<18:23,  6.00s/it]
Training Epoch: 14/100, batch 70/438 completed (loss: 0.0849:  16%|█▌        | 35/219 [03:32<18:23,  6.00s/it]
Training Epoch: 14/100, batch 70/438 completed (loss: 0.7348:  16%|█▌        | 35/219 [03:32<18:23,  6.00s/it]
Training Epoch: 14/100, batch 70/438 completed (loss: 0.0085:  16%|█▌        | 35/219 [03:32<18:23,  6.00s/it]
Training Epoch: 14/100, batch 70/438 completed (loss: 0.0484:  16%|█▋        | 36/219 [03:35<18:31,  6.08s/it]
Training Epoch: 14/100, batch 70/438 completed (loss: 0.0849:  16%|█▋        | 36/219 [03:35<18:31,  6.08s/it]
Training Epoch: 14/100, batch 70/438 completed (loss: 0.7348:  16%|█▋        | 36/219 [03:35<18:31,  6.08s/it]
Training Epoch: 14/100, batch 70/438 completed (loss: 0.0085:  16%|█▋        | 36/219 [03:35<18:31,  6.08s/it]
Training Epoch: 14/100, batch 71/438 completed (loss: 0.0367:  16%|█▋        | 36/219 [03:35<18:31,  6.08s/it]
Training Epoch: 14/100, batch 71/438 completed (loss: 0.1003:  16%|█▋        | 36/219 [03:35<18:31,  6.08s/it]
Training Epoch: 14/100, batch 71/438 completed (loss: 0.5062:  16%|█▋        | 36/219 [03:35<18:31,  6.08s/it]
Training Epoch: 14/100, batch 71/438 completed (loss: 0.0342:  16%|█▋        | 36/219 [03:35<18:31,  6.08s/it]
Training Epoch: 14/100, batch 72/438 completed (loss: 0.4436:  16%|█▋        | 36/219 [03:38<18:31,  6.08s/it]
Training Epoch: 14/100, batch 72/438 completed (loss: 0.0517:  16%|█▋        | 36/219 [03:38<18:31,  6.08s/it]
Training Epoch: 14/100, batch 72/438 completed (loss: 0.0455:  16%|█▋        | 36/219 [03:38<18:31,  6.08s/it]
Training Epoch: 14/100, batch 72/438 completed (loss: 0.1314:  16%|█▋        | 36/219 [03:38<18:31,  6.08s/it]
Training Epoch: 14/100, batch 72/438 completed (loss: 0.0455:  17%|█▋        | 37/219 [03:41<18:20,  6.05s/it]
Training Epoch: 14/100, batch 72/438 completed (loss: 0.1314:  17%|█▋        | 37/219 [03:41<18:20,  6.05s/it]
Training Epoch: 14/100, batch 72/438 completed (loss: 0.0517:  17%|█▋        | 37/219 [03:41<18:20,  6.05s/it]
Training Epoch: 14/100, batch 72/438 completed (loss: 0.4436:  17%|█▋        | 37/219 [03:41<18:20,  6.05s/it]
Training Epoch: 14/100, batch 73/438 completed (loss: 0.0427:  17%|█▋        | 37/219 [03:41<18:20,  6.05s/it]
Training Epoch: 14/100, batch 73/438 completed (loss: 0.0076:  17%|█▋        | 37/219 [03:41<18:20,  6.05s/it]
Training Epoch: 14/100, batch 73/438 completed (loss: 0.0245:  17%|█▋        | 37/219 [03:41<18:20,  6.05s/it]
Training Epoch: 14/100, batch 73/438 completed (loss: 0.0901:  17%|█▋        | 37/219 [03:41<18:20,  6.05s/it]
Training Epoch: 14/100, batch 74/438 completed (loss: 0.1061:  17%|█▋        | 37/219 [03:44<18:20,  6.05s/it]
Training Epoch: 14/100, batch 74/438 completed (loss: 0.0112:  17%|█▋        | 37/219 [03:44<18:20,  6.05s/it]
Training Epoch: 14/100, batch 74/438 completed (loss: 0.0405:  17%|█▋        | 37/219 [03:44<18:20,  6.05s/it]
Training Epoch: 14/100, batch 74/438 completed (loss: 0.0265:  17%|█▋        | 37/219 [03:44<18:20,  6.05s/it]
Training Epoch: 14/100, batch 74/438 completed (loss: 0.0112:  17%|█▋        | 38/219 [03:47<18:09,  6.02s/it]
Training Epoch: 14/100, batch 74/438 completed (loss: 0.0265:  17%|█▋        | 38/219 [03:47<18:09,  6.02s/it]
Training Epoch: 14/100, batch 74/438 completed (loss: 0.1061:  17%|█▋        | 38/219 [03:47<18:09,  6.02s/it]
Training Epoch: 14/100, batch 74/438 completed (loss: 0.0405:  17%|█▋        | 38/219 [03:47<18:09,  6.02s/it]
Training Epoch: 14/100, batch 75/438 completed (loss: 0.071:  17%|█▋        | 38/219 [03:47<18:09,  6.02s/it] 
Training Epoch: 14/100, batch 75/438 completed (loss: 0.0034:  17%|█▋        | 38/219 [03:47<18:09,  6.02s/it]
Training Epoch: 14/100, batch 75/438 completed (loss: 0.0236:  17%|█▋        | 38/219 [03:47<18:09,  6.02s/it]
Training Epoch: 14/100, batch 75/438 completed (loss: 0.0818:  17%|█▋        | 38/219 [03:47<18:09,  6.02s/it]
Training Epoch: 14/100, batch 76/438 completed (loss: 0.0447:  17%|█▋        | 38/219 [03:50<18:09,  6.02s/it]
Training Epoch: 14/100, batch 76/438 completed (loss: 0.0051:  17%|█▋        | 38/219 [03:50<18:09,  6.02s/it]
Training Epoch: 14/100, batch 76/438 completed (loss: 0.1022:  17%|█▋        | 38/219 [03:50<18:09,  6.02s/it]
Training Epoch: 14/100, batch 76/438 completed (loss: 0.0983:  17%|█▋        | 38/219 [03:50<18:09,  6.02s/it]
Training Epoch: 14/100, batch 76/438 completed (loss: 0.1022:  18%|█▊        | 39/219 [03:54<18:16,  6.09s/it]
Training Epoch: 14/100, batch 76/438 completed (loss: 0.0051:  18%|█▊        | 39/219 [03:54<18:16,  6.09s/it]
Training Epoch: 14/100, batch 76/438 completed (loss: 0.0447:  18%|█▊        | 39/219 [03:54<18:16,  6.09s/it]
Training Epoch: 14/100, batch 76/438 completed (loss: 0.0983:  18%|█▊        | 39/219 [03:54<18:16,  6.09s/it]
Training Epoch: 14/100, batch 77/438 completed (loss: 0.0511:  18%|█▊        | 39/219 [03:54<18:16,  6.09s/it]
Training Epoch: 14/100, batch 77/438 completed (loss: 0.027:  18%|█▊        | 39/219 [03:54<18:16,  6.09s/it] 
Training Epoch: 14/100, batch 77/438 completed (loss: 0.0385:  18%|█▊        | 39/219 [03:54<18:16,  6.09s/it]
Training Epoch: 14/100, batch 77/438 completed (loss: 0.0726:  18%|█▊        | 39/219 [03:54<18:16,  6.09s/it]
Training Epoch: 14/100, batch 78/438 completed (loss: 0.0349:  18%|█▊        | 39/219 [03:56<18:16,  6.09s/it]
Training Epoch: 14/100, batch 78/438 completed (loss: 0.0181:  18%|█▊        | 39/219 [03:56<18:16,  6.09s/it]
Training Epoch: 14/100, batch 78/438 completed (loss: 0.0702:  18%|█▊        | 39/219 [03:56<18:16,  6.09s/it]
Training Epoch: 14/100, batch 78/438 completed (loss: 0.0113:  18%|█▊        | 39/219 [03:57<18:16,  6.09s/it]
Training Epoch: 14/100, batch 78/438 completed (loss: 0.0181:  18%|█▊        | 40/219 [03:59<17:56,  6.02s/it]
Training Epoch: 14/100, batch 78/438 completed (loss: 0.0702:  18%|█▊        | 40/219 [03:59<17:56,  6.02s/it]
Training Epoch: 14/100, batch 78/438 completed (loss: 0.0113:  18%|█▊        | 40/219 [03:59<17:56,  6.02s/it]
Training Epoch: 14/100, batch 78/438 completed (loss: 0.0349:  18%|█▊        | 40/219 [03:59<17:57,  6.02s/it]
Training Epoch: 14/100, batch 79/438 completed (loss: 0.1884:  18%|█▊        | 40/219 [03:59<17:56,  6.02s/it]
Training Epoch: 14/100, batch 79/438 completed (loss: 0.1086:  18%|█▊        | 40/219 [03:59<17:57,  6.02s/it]
Training Epoch: 14/100, batch 79/438 completed (loss: 0.0221:  18%|█▊        | 40/219 [03:59<17:56,  6.02s/it]
Training Epoch: 14/100, batch 79/438 completed (loss: 0.0349:  18%|█▊        | 40/219 [03:59<17:56,  6.02s/it]
Training Epoch: 14/100, batch 80/438 completed (loss: 0.0147:  18%|█▊        | 40/219 [04:02<17:57,  6.02s/it]
Training Epoch: 14/100, batch 80/438 completed (loss: 0.2613:  18%|█▊        | 40/219 [04:02<17:56,  6.02s/it]
Training Epoch: 14/100, batch 80/438 completed (loss: 0.0169:  18%|█▊        | 40/219 [04:02<17:56,  6.02s/it]
Training Epoch: 14/100, batch 80/438 completed (loss: 0.4087:  18%|█▊        | 40/219 [04:02<17:56,  6.02s/it]
Training Epoch: 14/100, batch 80/438 completed (loss: 0.4087:  19%|█▊        | 41/219 [04:06<18:04,  6.09s/it]
Training Epoch: 14/100, batch 80/438 completed (loss: 0.0147:  19%|█▊        | 41/219 [04:06<18:04,  6.10s/it]
Training Epoch: 14/100, batch 80/438 completed (loss: 0.0169:  19%|█▊        | 41/219 [04:06<18:05,  6.10s/it]
Training Epoch: 14/100, batch 80/438 completed (loss: 0.2613:  19%|█▊        | 41/219 [04:06<18:05,  6.10s/it]
Training Epoch: 14/100, batch 81/438 completed (loss: 0.1849:  19%|█▊        | 41/219 [04:06<18:05,  6.10s/it]
Training Epoch: 14/100, batch 81/438 completed (loss: 0.0078:  19%|█▊        | 41/219 [04:06<18:04,  6.10s/it]
Training Epoch: 14/100, batch 81/438 completed (loss: 0.0744:  19%|█▊        | 41/219 [04:06<18:05,  6.10s/it]
Training Epoch: 14/100, batch 81/438 completed (loss: 0.3846:  19%|█▊        | 41/219 [04:06<18:04,  6.09s/it]
Training Epoch: 14/100, batch 82/438 completed (loss: 0.0277:  19%|█▊        | 41/219 [04:09<18:05,  6.10s/it]
Training Epoch: 14/100, batch 82/438 completed (loss: 0.0178:  19%|█▊        | 41/219 [04:09<18:05,  6.10s/it]
Training Epoch: 14/100, batch 82/438 completed (loss: 0.0306:  19%|█▊        | 41/219 [04:09<18:04,  6.10s/it]
Training Epoch: 14/100, batch 82/438 completed (loss: 0.1977:  19%|█▊        | 41/219 [04:09<18:04,  6.09s/it]
Training Epoch: 14/100, batch 82/438 completed (loss: 0.0306:  19%|█▉        | 42/219 [04:12<17:53,  6.07s/it]
Training Epoch: 14/100, batch 82/438 completed (loss: 0.0178:  19%|█▉        | 42/219 [04:12<17:53,  6.07s/it]
Training Epoch: 14/100, batch 82/438 completed (loss: 0.0277:  19%|█▉        | 42/219 [04:12<17:53,  6.07s/it]
Training Epoch: 14/100, batch 82/438 completed (loss: 0.1977:  19%|█▉        | 42/219 [04:12<17:54,  6.07s/it]
Training Epoch: 14/100, batch 83/438 completed (loss: 0.2078:  19%|█▉        | 42/219 [04:12<17:53,  6.07s/it]
Training Epoch: 14/100, batch 83/438 completed (loss: 0.0501:  19%|█▉        | 42/219 [04:12<17:53,  6.07s/it]
Training Epoch: 14/100, batch 83/438 completed (loss: 0.0944:  19%|█▉        | 42/219 [04:12<17:53,  6.07s/it]
Training Epoch: 14/100, batch 83/438 completed (loss: 0.057:  19%|█▉        | 42/219 [04:12<17:54,  6.07s/it] 
Training Epoch: 14/100, batch 84/438 completed (loss: 0.1469:  19%|█▉        | 42/219 [04:15<17:53,  6.07s/it]
Training Epoch: 14/100, batch 84/438 completed (loss: 0.032:  19%|█▉        | 42/219 [04:15<17:53,  6.07s/it] 
Training Epoch: 14/100, batch 84/438 completed (loss: 0.009:  19%|█▉        | 42/219 [04:15<17:53,  6.07s/it] 
Training Epoch: 14/100, batch 84/438 completed (loss: 0.1309:  19%|█▉        | 42/219 [04:15<17:54,  6.07s/it]
Training Epoch: 14/100, batch 84/438 completed (loss: 0.1309:  20%|█▉        | 43/219 [04:18<17:43,  6.04s/it]
Training Epoch: 14/100, batch 84/438 completed (loss: 0.1469:  20%|█▉        | 43/219 [04:18<17:43,  6.04s/it]
Training Epoch: 14/100, batch 84/438 completed (loss: 0.009:  20%|█▉        | 43/219 [04:18<17:43,  6.04s/it]
Training Epoch: 14/100, batch 84/438 completed (loss: 0.032:  20%|█▉        | 43/219 [04:18<17:43,  6.04s/it]
Training Epoch: 14/100, batch 85/438 completed (loss: 0.221:  20%|█▉        | 43/219 [04:18<17:43,  6.04s/it]
Training Epoch: 14/100, batch 85/438 completed (loss: 0.018:  20%|█▉        | 43/219 [04:18<17:43,  6.04s/it] 
Training Epoch: 14/100, batch 85/438 completed (loss: 0.0237:  20%|█▉        | 43/219 [04:18<17:43,  6.04s/it]
Training Epoch: 14/100, batch 85/438 completed (loss: 0.0172:  20%|█▉        | 43/219 [04:18<17:43,  6.04s/it]
Training Epoch: 14/100, batch 86/438 completed (loss: 0.0099:  20%|█▉        | 43/219 [04:21<17:43,  6.04s/it]
Training Epoch: 14/100, batch 86/438 completed (loss: 0.0214:  20%|█▉        | 43/219 [04:21<17:43,  6.04s/it]
Training Epoch: 14/100, batch 86/438 completed (loss: 0.1197:  20%|█▉        | 43/219 [04:21<17:43,  6.04s/it]
Training Epoch: 14/100, batch 86/438 completed (loss: 0.0041:  20%|█▉        | 43/219 [04:21<17:43,  6.04s/it]
Training Epoch: 14/100, batch 86/438 completed (loss: 0.0099:  20%|██        | 44/219 [04:24<17:40,  6.06s/it]
Training Epoch: 14/100, batch 86/438 completed (loss: 0.0214:  20%|██        | 44/219 [04:24<17:40,  6.06s/it]
Training Epoch: 14/100, batch 86/438 completed (loss: 0.1197:  20%|██        | 44/219 [04:24<17:40,  6.06s/it]
Training Epoch: 14/100, batch 86/438 completed (loss: 0.0041:  20%|██        | 44/219 [04:24<17:41,  6.06s/it]
Training Epoch: 14/100, batch 87/438 completed (loss: 0.5234:  20%|██        | 44/219 [04:24<17:40,  6.06s/it]
Training Epoch: 14/100, batch 87/438 completed (loss: 0.0048:  20%|██        | 44/219 [04:24<17:40,  6.06s/it]
Training Epoch: 14/100, batch 87/438 completed (loss: 0.0384:  20%|██        | 44/219 [04:24<17:40,  6.06s/it]
Training Epoch: 14/100, batch 87/438 completed (loss: 0.0276:  20%|██        | 44/219 [04:24<17:41,  6.06s/it]
Training Epoch: 14/100, batch 88/438 completed (loss: 0.9274:  20%|██        | 44/219 [04:27<17:40,  6.06s/it]
Training Epoch: 14/100, batch 88/438 completed (loss: 0.0123:  20%|██        | 44/219 [04:27<17:40,  6.06s/it]
Training Epoch: 14/100, batch 88/438 completed (loss: 0.0198:  20%|██        | 44/219 [04:27<17:40,  6.06s/it]
Training Epoch: 14/100, batch 88/438 completed (loss: 0.3006:  20%|██        | 44/219 [04:27<17:41,  6.06s/it]
Training Epoch: 14/100, batch 88/438 completed (loss: 0.0198:  21%|██        | 45/219 [04:30<17:31,  6.04s/it]
Training Epoch: 14/100, batch 88/438 completed (loss: 0.0123:  21%|██        | 45/219 [04:30<17:31,  6.04s/it]
Training Epoch: 14/100, batch 88/438 completed (loss: 0.9274:  21%|██        | 45/219 [04:30<17:31,  6.04s/it]
Training Epoch: 14/100, batch 88/438 completed (loss: 0.3006:  21%|██        | 45/219 [04:30<17:31,  6.04s/it]
Training Epoch: 14/100, batch 89/438 completed (loss: 0.0657:  21%|██        | 45/219 [04:30<17:31,  6.04s/it]
Training Epoch: 14/100, batch 89/438 completed (loss: 0.004:  21%|██        | 45/219 [04:30<17:31,  6.04s/it] 
Training Epoch: 14/100, batch 89/438 completed (loss: 0.0289:  21%|██        | 45/219 [04:30<17:31,  6.04s/it]
Training Epoch: 14/100, batch 89/438 completed (loss: 0.4428:  21%|██        | 45/219 [04:30<17:31,  6.04s/it]
Training Epoch: 14/100, batch 90/438 completed (loss: 0.0995:  21%|██        | 45/219 [04:33<17:31,  6.04s/it]
Training Epoch: 14/100, batch 90/438 completed (loss: 0.1035:  21%|██        | 45/219 [04:33<17:31,  6.04s/it]
Training Epoch: 14/100, batch 90/438 completed (loss: 0.1165:  21%|██        | 45/219 [04:33<17:31,  6.04s/it]
Training Epoch: 14/100, batch 90/438 completed (loss: 0.0869:  21%|██        | 45/219 [04:33<17:31,  6.04s/it]
Training Epoch: 14/100, batch 90/438 completed (loss: 0.0869:  21%|██        | 46/219 [04:36<17:29,  6.07s/it]
Training Epoch: 14/100, batch 90/438 completed (loss: 0.1035:  21%|██        | 46/219 [04:36<17:29,  6.07s/it]
Training Epoch: 14/100, batch 90/438 completed (loss: 0.0995:  21%|██        | 46/219 [04:36<17:30,  6.07s/it]
Training Epoch: 14/100, batch 90/438 completed (loss: 0.1165:  21%|██        | 46/219 [04:36<17:30,  6.07s/it]
Training Epoch: 14/100, batch 91/438 completed (loss: 0.268:  21%|██        | 46/219 [04:36<17:30,  6.07s/it] 
Training Epoch: 14/100, batch 91/438 completed (loss: 0.0563:  21%|██        | 46/219 [04:36<17:30,  6.07s/it]
Training Epoch: 14/100, batch 91/438 completed (loss: 0.0048:  21%|██        | 46/219 [04:36<17:29,  6.07s/it]
Training Epoch: 14/100, batch 91/438 completed (loss: 0.0065:  21%|██        | 46/219 [04:36<17:29,  6.07s/it]
Training Epoch: 14/100, batch 92/438 completed (loss: 0.0029:  21%|██        | 46/219 [04:39<17:29,  6.07s/it]
Training Epoch: 14/100, batch 92/438 completed (loss: 0.0556:  21%|██        | 46/219 [04:39<17:30,  6.07s/it]
Training Epoch: 14/100, batch 92/438 completed (loss: 0.1281:  21%|██        | 46/219 [04:39<17:30,  6.07s/it]
Training Epoch: 14/100, batch 92/438 completed (loss: 0.0201:  21%|██        | 46/219 [04:39<17:29,  6.07s/it]
Training Epoch: 14/100, batch 92/438 completed (loss: 0.1281:  21%|██▏       | 47/219 [04:42<17:27,  6.09s/it]
Training Epoch: 14/100, batch 92/438 completed (loss: 0.0556:  21%|██▏       | 47/219 [04:42<17:27,  6.09s/it]
Training Epoch: 14/100, batch 92/438 completed (loss: 0.0029:  21%|██▏       | 47/219 [04:42<17:28,  6.09s/it]
Training Epoch: 14/100, batch 92/438 completed (loss: 0.0201:  21%|██▏       | 47/219 [04:42<17:28,  6.09s/it]
Training Epoch: 14/100, batch 93/438 completed (loss: 0.0154:  21%|██▏       | 47/219 [04:42<17:28,  6.09s/it]
Training Epoch: 14/100, batch 93/438 completed (loss: 0.0134:  21%|██▏       | 47/219 [04:42<17:27,  6.09s/it]
Training Epoch: 14/100, batch 93/438 completed (loss: 0.386:  21%|██▏       | 47/219 [04:42<17:27,  6.09s/it] 
Training Epoch: 14/100, batch 93/438 completed (loss: 0.0054:  21%|██▏       | 47/219 [04:42<17:28,  6.09s/it]
Training Epoch: 14/100, batch 94/438 completed (loss: 0.1016:  21%|██▏       | 47/219 [04:45<17:28,  6.09s/it]
Training Epoch: 14/100, batch 94/438 completed (loss: 0.0474:  21%|██▏       | 47/219 [04:45<17:27,  6.09s/it]
Training Epoch: 14/100, batch 94/438 completed (loss: 0.4723:  21%|██▏       | 47/219 [04:45<17:27,  6.09s/it]
Training Epoch: 14/100, batch 94/438 completed (loss: 0.0039:  21%|██▏       | 47/219 [04:45<17:28,  6.09s/it]
Training Epoch: 14/100, batch 94/438 completed (loss: 0.0039:  22%|██▏       | 48/219 [04:48<17:09,  6.02s/it]
Training Epoch: 14/100, batch 94/438 completed (loss: 0.4723:  22%|██▏       | 48/219 [04:48<17:10,  6.02s/it]
Training Epoch: 14/100, batch 94/438 completed (loss: 0.1016:  22%|██▏       | 48/219 [04:48<17:10,  6.02s/it]
Training Epoch: 14/100, batch 94/438 completed (loss: 0.0474:  22%|██▏       | 48/219 [04:48<17:10,  6.03s/it]
Training Epoch: 14/100, batch 95/438 completed (loss: 0.0119:  22%|██▏       | 48/219 [04:48<17:10,  6.02s/it]
Training Epoch: 14/100, batch 95/438 completed (loss: 0.0318:  22%|██▏       | 48/219 [04:48<17:10,  6.03s/it]
Training Epoch: 14/100, batch 95/438 completed (loss: 0.0178:  22%|██▏       | 48/219 [04:48<17:10,  6.02s/it]
Training Epoch: 14/100, batch 95/438 completed (loss: 0.1781:  22%|██▏       | 48/219 [04:48<17:09,  6.02s/it]
Training Epoch: 14/100, batch 96/438 completed (loss: 0.3734:  22%|██▏       | 48/219 [04:51<17:10,  6.02s/it]
Training Epoch: 14/100, batch 96/438 completed (loss: 0.0071:  22%|██▏       | 48/219 [04:51<17:10,  6.03s/it]
Training Epoch: 14/100, batch 96/438 completed (loss: 0.0246:  22%|██▏       | 48/219 [04:51<17:10,  6.02s/it]
Training Epoch: 14/100, batch 96/438 completed (loss: 0.2617:  22%|██▏       | 48/219 [04:51<17:09,  6.02s/it]
Training Epoch: 14/100, batch 96/438 completed (loss: 0.0246:  22%|██▏       | 49/219 [04:54<17:15,  6.09s/it]
Training Epoch: 14/100, batch 96/438 completed (loss: 0.0071:  22%|██▏       | 49/219 [04:54<17:15,  6.09s/it]
Training Epoch: 14/100, batch 96/438 completed (loss: 0.3734:  22%|██▏       | 49/219 [04:54<17:15,  6.09s/it]
Training Epoch: 14/100, batch 96/438 completed (loss: 0.2617:  22%|██▏       | 49/219 [04:54<17:16,  6.10s/it]
Training Epoch: 14/100, batch 97/438 completed (loss: 0.0345:  22%|██▏       | 49/219 [04:54<17:15,  6.09s/it]
Training Epoch: 14/100, batch 97/438 completed (loss: 0.029:  22%|██▏       | 49/219 [04:54<17:15,  6.09s/it] 
Training Epoch: 14/100, batch 97/438 completed (loss: 0.0226:  22%|██▏       | 49/219 [04:54<17:15,  6.09s/it]
Training Epoch: 14/100, batch 97/438 completed (loss: 0.006:  22%|██▏       | 49/219 [04:54<17:16,  6.10s/it] 
Training Epoch: 14/100, batch 98/438 completed (loss: 0.0064:  22%|██▏       | 49/219 [04:57<17:15,  6.09s/it]
Training Epoch: 14/100, batch 98/438 completed (loss: 0.0515:  22%|██▏       | 49/219 [04:57<17:15,  6.09s/it]
Training Epoch: 14/100, batch 98/438 completed (loss: 0.1015:  22%|██▏       | 49/219 [04:57<17:15,  6.09s/it]
Training Epoch: 14/100, batch 98/438 completed (loss: 0.0184:  22%|██▏       | 49/219 [04:57<17:16,  6.10s/it]
Training Epoch: 14/100, batch 98/438 completed (loss: 0.0064:  23%|██▎       | 50/219 [05:00<17:04,  6.06s/it]
Training Epoch: 14/100, batch 98/438 completed (loss: 0.0515:  23%|██▎       | 50/219 [05:00<17:04,  6.06s/it]
Training Epoch: 14/100, batch 98/438 completed (loss: 0.0184:  23%|██▎       | 50/219 [05:00<17:04,  6.06s/it]
Training Epoch: 14/100, batch 98/438 completed (loss: 0.1015:  23%|██▎       | 50/219 [05:00<17:04,  6.06s/it]
Training Epoch: 14/100, batch 99/438 completed (loss: 0.257:  23%|██▎       | 50/219 [05:00<17:04,  6.06s/it] 
Training Epoch: 14/100, batch 99/438 completed (loss: 0.0469:  23%|██▎       | 50/219 [05:00<17:04,  6.06s/it]
Training Epoch: 14/100, batch 99/438 completed (loss: 0.1305:  23%|██▎       | 50/219 [05:00<17:04,  6.06s/it]
Training Epoch: 14/100, batch 99/438 completed (loss: 0.0989:  23%|██▎       | 50/219 [05:00<17:04,  6.06s/it]
Training Epoch: 14/100, batch 100/438 completed (loss: 0.0458:  23%|██▎       | 50/219 [05:03<17:04,  6.06s/it]
Training Epoch: 14/100, batch 100/438 completed (loss: 0.1162:  23%|██▎       | 50/219 [05:03<17:04,  6.06s/it]
Training Epoch: 14/100, batch 100/438 completed (loss: 0.051:  23%|██▎       | 50/219 [05:03<17:04,  6.06s/it]
Training Epoch: 14/100, batch 100/438 completed (loss: 0.0046:  23%|██▎       | 50/219 [05:03<17:04,  6.06s/it]
Training Epoch: 14/100, batch 100/438 completed (loss: 0.051:  23%|██▎       | 51/219 [05:06<16:54,  6.04s/it]
Training Epoch: 14/100, batch 100/438 completed (loss: 0.0458:  23%|██▎       | 51/219 [05:06<16:54,  6.04s/it]
Training Epoch: 14/100, batch 100/438 completed (loss: 0.0046:  23%|██▎       | 51/219 [05:06<16:54,  6.04s/it]
Training Epoch: 14/100, batch 100/438 completed (loss: 0.1162:  23%|██▎       | 51/219 [05:06<16:54,  6.04s/it]
Training Epoch: 14/100, batch 101/438 completed (loss: 0.3877:  23%|██▎       | 51/219 [05:06<16:54,  6.04s/it]
Training Epoch: 14/100, batch 101/438 completed (loss: 0.0092:  23%|██▎       | 51/219 [05:06<16:54,  6.04s/it]
Training Epoch: 14/100, batch 101/438 completed (loss: 0.044:  23%|██▎       | 51/219 [05:06<16:54,  6.04s/it]
Training Epoch: 14/100, batch 101/438 completed (loss: 0.0379:  23%|██▎       | 51/219 [05:06<16:54,  6.04s/it]
Training Epoch: 14/100, batch 102/438 completed (loss: 0.1874:  23%|██▎       | 51/219 [05:09<16:54,  6.04s/it]
Training Epoch: 14/100, batch 102/438 completed (loss: 0.22:  23%|██▎       | 51/219 [05:09<16:54,  6.04s/it]  
Training Epoch: 14/100, batch 102/438 completed (loss: 0.148:  23%|██▎       | 51/219 [05:09<16:54,  6.04s/it] 
Training Epoch: 14/100, batch 102/438 completed (loss: 0.3675:  23%|██▎       | 51/219 [05:09<16:54,  6.04s/it]
Training Epoch: 14/100, batch 102/438 completed (loss: 0.1874:  24%|██▎       | 52/219 [05:12<16:51,  6.06s/it]
Training Epoch: 14/100, batch 102/438 completed (loss: 0.22:  24%|██▎       | 52/219 [05:12<16:51,  6.06s/it]
Training Epoch: 14/100, batch 102/438 completed (loss: 0.148:  24%|██▎       | 52/219 [05:12<16:51,  6.06s/it]
Training Epoch: 14/100, batch 102/438 completed (loss: 0.3675:  24%|██▎       | 52/219 [05:12<16:51,  6.06s/it]
Training Epoch: 14/100, batch 103/438 completed (loss: 0.4705:  24%|██▎       | 52/219 [05:12<16:51,  6.06s/it]
Training Epoch: 14/100, batch 103/438 completed (loss: 0.0679:  24%|██▎       | 52/219 [05:12<16:51,  6.06s/it]
Training Epoch: 14/100, batch 103/438 completed (loss: 0.3216:  24%|██▎       | 52/219 [05:12<16:51,  6.06s/it]
Training Epoch: 14/100, batch 103/438 completed (loss: 0.1578:  24%|██▎       | 52/219 [05:12<16:51,  6.06s/it]
Training Epoch: 14/100, batch 104/438 completed (loss: 0.0071:  24%|██▎       | 52/219 [05:15<16:51,  6.06s/it]
Training Epoch: 14/100, batch 104/438 completed (loss: 0.1836:  24%|██▎       | 52/219 [05:15<16:51,  6.06s/it]
Training Epoch: 14/100, batch 104/438 completed (loss: 0.17:  24%|██▎       | 52/219 [05:15<16:51,  6.06s/it]  
Training Epoch: 14/100, batch 104/438 completed (loss: 0.034:  24%|██▎       | 52/219 [05:15<16:51,  6.06s/it] 
Training Epoch: 14/100, batch 104/438 completed (loss: 0.0071:  24%|██▍       | 53/219 [05:18<16:41,  6.03s/it]
Training Epoch: 14/100, batch 104/438 completed (loss: 0.17:  24%|██▍       | 53/219 [05:18<16:41,  6.03s/it]
Training Epoch: 14/100, batch 104/438 completed (loss: 0.034:  24%|██▍       | 53/219 [05:18<16:41,  6.03s/it]
Training Epoch: 14/100, batch 104/438 completed (loss: 0.1836:  24%|██▍       | 53/219 [05:18<16:41,  6.03s/it]
Training Epoch: 14/100, batch 105/438 completed (loss: 0.4663:  24%|██▍       | 53/219 [05:18<16:41,  6.03s/it]
Training Epoch: 14/100, batch 105/438 completed (loss: 0.4091:  24%|██▍       | 53/219 [05:18<16:41,  6.03s/it]
Training Epoch: 14/100, batch 105/438 completed (loss: 0.3927:  24%|██▍       | 53/219 [05:18<16:41,  6.03s/it]
Training Epoch: 14/100, batch 105/438 completed (loss: 0.0628:  24%|██▍       | 53/219 [05:18<16:41,  6.03s/it]
Training Epoch: 14/100, batch 106/438 completed (loss: 0.1511:  24%|██▍       | 53/219 [05:21<16:41,  6.03s/it]
Training Epoch: 14/100, batch 106/438 completed (loss: 0.0161:  24%|██▍       | 53/219 [05:21<16:41,  6.03s/it]
Training Epoch: 14/100, batch 106/438 completed (loss: 0.0683:  24%|██▍       | 53/219 [05:21<16:41,  6.03s/it]
Training Epoch: 14/100, batch 106/438 completed (loss: 0.0687:  24%|██▍       | 53/219 [05:21<16:41,  6.03s/it]
Training Epoch: 14/100, batch 106/438 completed (loss: 0.1511:  25%|██▍       | 54/219 [05:24<16:42,  6.08s/it]
Training Epoch: 14/100, batch 106/438 completed (loss: 0.0683:  25%|██▍       | 54/219 [05:24<16:42,  6.08s/it]
Training Epoch: 14/100, batch 106/438 completed (loss: 0.0687:  25%|██▍       | 54/219 [05:24<16:42,  6.08s/it]
Training Epoch: 14/100, batch 106/438 completed (loss: 0.0161:  25%|██▍       | 54/219 [05:24<16:43,  6.08s/it]
Training Epoch: 14/100, batch 107/438 completed (loss: 0.1381:  25%|██▍       | 54/219 [05:24<16:42,  6.08s/it]
Training Epoch: 14/100, batch 107/438 completed (loss: 0.0396:  25%|██▍       | 54/219 [05:24<16:43,  6.08s/it]
Training Epoch: 14/100, batch 107/438 completed (loss: 0.0739:  25%|██▍       | 54/219 [05:24<16:42,  6.08s/it]
Training Epoch: 14/100, batch 107/438 completed (loss: 0.0192:  25%|██▍       | 54/219 [05:24<16:42,  6.08s/it]
Training Epoch: 14/100, batch 108/438 completed (loss: 0.1773:  25%|██▍       | 54/219 [05:27<16:42,  6.08s/it]
Training Epoch: 14/100, batch 108/438 completed (loss: 0.5285:  25%|██▍       | 54/219 [05:27<16:43,  6.08s/it]
Training Epoch: 14/100, batch 108/438 completed (loss: 0.0769:  25%|██▍       | 54/219 [05:27<16:42,  6.08s/it]
Training Epoch: 14/100, batch 108/438 completed (loss: 0.4396:  25%|██▍       | 54/219 [05:27<16:42,  6.08s/it]
Training Epoch: 14/100, batch 108/438 completed (loss: 0.1773:  25%|██▌       | 55/219 [05:30<16:26,  6.01s/it]
Training Epoch: 14/100, batch 108/438 completed (loss: 0.5285:  25%|██▌       | 55/219 [05:30<16:26,  6.01s/it]
Training Epoch: 14/100, batch 108/438 completed (loss: 0.0769:  25%|██▌       | 55/219 [05:30<16:26,  6.02s/it]
Training Epoch: 14/100, batch 108/438 completed (loss: 0.4396:  25%|██▌       | 55/219 [05:30<16:26,  6.02s/it]
Training Epoch: 14/100, batch 109/438 completed (loss: 0.2021:  25%|██▌       | 55/219 [05:30<16:26,  6.01s/it]
Training Epoch: 14/100, batch 109/438 completed (loss: 0.0158:  25%|██▌       | 55/219 [05:30<16:26,  6.02s/it]
Training Epoch: 14/100, batch 109/438 completed (loss: 0.0184:  25%|██▌       | 55/219 [05:30<16:26,  6.01s/it]
Training Epoch: 14/100, batch 109/438 completed (loss: 0.0512:  25%|██▌       | 55/219 [05:30<16:26,  6.02s/it]
Training Epoch: 14/100, batch 110/438 completed (loss: 0.0077:  25%|██▌       | 55/219 [05:33<16:26,  6.02s/it]
Training Epoch: 14/100, batch 110/438 completed (loss: 0.0491:  25%|██▌       | 55/219 [05:33<16:26,  6.01s/it]
Training Epoch: 14/100, batch 110/438 completed (loss: 0.1845:  25%|██▌       | 55/219 [05:33<16:26,  6.01s/it]
Training Epoch: 14/100, batch 110/438 completed (loss: 0.2471:  25%|██▌       | 55/219 [05:33<16:26,  6.02s/it]
Training Epoch: 14/100, batch 110/438 completed (loss: 0.0491:  26%|██▌       | 56/219 [05:36<16:20,  6.02s/it]
Training Epoch: 14/100, batch 110/438 completed (loss: 0.2471:  26%|██▌       | 56/219 [05:36<16:20,  6.02s/it]
Training Epoch: 14/100, batch 110/438 completed (loss: 0.0077:  26%|██▌       | 56/219 [05:36<16:20,  6.02s/it]
Training Epoch: 14/100, batch 110/438 completed (loss: 0.1845:  26%|██▌       | 56/219 [05:36<16:21,  6.02s/it]
Training Epoch: 14/100, batch 111/438 completed (loss: 0.1327:  26%|██▌       | 56/219 [05:36<16:21,  6.02s/it]
Training Epoch: 14/100, batch 111/438 completed (loss: 0.03:  26%|██▌       | 56/219 [05:36<16:20,  6.02s/it]  
Training Epoch: 14/100, batch 111/438 completed (loss: 0.0137:  26%|██▌       | 56/219 [05:36<16:20,  6.02s/it]
Training Epoch: 14/100, batch 111/438 completed (loss: 0.1432:  26%|██▌       | 56/219 [05:36<16:20,  6.02s/it]
Training Epoch: 14/100, batch 112/438 completed (loss: 0.0171:  26%|██▌       | 56/219 [05:40<16:21,  6.02s/it]
Training Epoch: 14/100, batch 112/438 completed (loss: 0.024:  26%|██▌       | 56/219 [05:40<16:20,  6.02s/it] 
Training Epoch: 14/100, batch 112/438 completed (loss: 0.1019:  26%|██▌       | 56/219 [05:40<16:20,  6.02s/it]
Training Epoch: 14/100, batch 112/438 completed (loss: 0.2513:  26%|██▌       | 56/219 [05:40<16:20,  6.02s/it]
Training Epoch: 14/100, batch 112/438 completed (loss: 0.0171:  26%|██▌       | 57/219 [05:42<16:20,  6.05s/it]
Training Epoch: 14/100, batch 112/438 completed (loss: 0.1019:  26%|██▌       | 57/219 [05:42<16:20,  6.05s/it]
Training Epoch: 14/100, batch 112/438 completed (loss: 0.024:  26%|██▌       | 57/219 [05:42<16:20,  6.06s/it]
Training Epoch: 14/100, batch 112/438 completed (loss: 0.2513:  26%|██▌       | 57/219 [05:42<16:21,  6.06s/it]
Training Epoch: 14/100, batch 113/438 completed (loss: 0.118:  26%|██▌       | 57/219 [05:42<16:20,  6.06s/it]
Training Epoch: 14/100, batch 113/438 completed (loss: 0.0073:  26%|██▌       | 57/219 [05:42<16:20,  6.05s/it]
Training Epoch: 14/100, batch 113/438 completed (loss: 0.1546:  26%|██▌       | 57/219 [05:42<16:20,  6.05s/it]
Training Epoch: 14/100, batch 113/438 completed (loss: 0.1347:  26%|██▌       | 57/219 [05:42<16:21,  6.06s/it]
Training Epoch: 14/100, batch 114/438 completed (loss: 0.0622:  26%|██▌       | 57/219 [05:45<16:20,  6.05s/it]
Training Epoch: 14/100, batch 114/438 completed (loss: 0.0737:  26%|██▌       | 57/219 [05:45<16:20,  6.05s/it]
Training Epoch: 14/100, batch 114/438 completed (loss: 0.0265:  26%|██▌       | 57/219 [05:45<16:20,  6.06s/it]
Training Epoch: 14/100, batch 114/438 completed (loss: 0.0452:  26%|██▌       | 57/219 [05:45<16:21,  6.06s/it]
Training Epoch: 14/100, batch 114/438 completed (loss: 0.0452:  26%|██▋       | 58/219 [05:48<16:12,  6.04s/it]
Training Epoch: 14/100, batch 114/438 completed (loss: 0.0622:  26%|██▋       | 58/219 [05:48<16:12,  6.04s/it]
Training Epoch: 14/100, batch 114/438 completed (loss: 0.0737:  26%|██▋       | 58/219 [05:48<16:12,  6.04s/it]
Training Epoch: 14/100, batch 114/438 completed (loss: 0.0265:  26%|██▋       | 58/219 [05:48<16:12,  6.04s/it]
Training Epoch: 14/100, batch 115/438 completed (loss: 0.0501:  26%|██▋       | 58/219 [05:48<16:12,  6.04s/it]
Training Epoch: 14/100, batch 115/438 completed (loss: 0.0112:  26%|██▋       | 58/219 [05:48<16:12,  6.04s/it]
Training Epoch: 14/100, batch 115/438 completed (loss: 0.4175:  26%|██▋       | 58/219 [05:48<16:12,  6.04s/it]
Training Epoch: 14/100, batch 115/438 completed (loss: 0.0122:  26%|██▋       | 58/219 [05:48<16:12,  6.04s/it]
Training Epoch: 14/100, batch 116/438 completed (loss: 0.0526:  26%|██▋       | 58/219 [05:51<16:12,  6.04s/it]
Training Epoch: 14/100, batch 116/438 completed (loss: 0.0454:  26%|██▋       | 58/219 [05:51<16:12,  6.04s/it]
Training Epoch: 14/100, batch 116/438 completed (loss: 0.2958:  26%|██▋       | 58/219 [05:51<16:12,  6.04s/it]
Training Epoch: 14/100, batch 116/438 completed (loss: 0.145:  26%|██▋       | 58/219 [05:51<16:12,  6.04s/it] 
Training Epoch: 14/100, batch 116/438 completed (loss: 0.0454:  27%|██▋       | 59/219 [05:55<16:13,  6.08s/it]
Training Epoch: 14/100, batch 116/438 completed (loss: 0.145:  27%|██▋       | 59/219 [05:55<16:12,  6.08s/it]
Training Epoch: 14/100, batch 116/438 completed (loss: 0.2958:  27%|██▋       | 59/219 [05:55<16:13,  6.08s/it]
Training Epoch: 14/100, batch 116/438 completed (loss: 0.0526:  27%|██▋       | 59/219 [05:55<16:13,  6.08s/it]
Training Epoch: 14/100, batch 117/438 completed (loss: 0.0274:  27%|██▋       | 59/219 [05:55<16:13,  6.08s/it]
Training Epoch: 14/100, batch 117/438 completed (loss: 0.0279:  27%|██▋       | 59/219 [05:55<16:13,  6.08s/it]
Training Epoch: 14/100, batch 117/438 completed (loss: 0.0275:  27%|██▋       | 59/219 [05:55<16:13,  6.08s/it]
Training Epoch: 14/100, batch 117/438 completed (loss: 0.0279:  27%|██▋       | 59/219 [05:55<16:12,  6.08s/it]
Training Epoch: 14/100, batch 118/438 completed (loss: 0.1633:  27%|██▋       | 59/219 [05:58<16:13,  6.08s/it]
Training Epoch: 14/100, batch 118/438 completed (loss: 0.0448:  27%|██▋       | 59/219 [05:58<16:13,  6.08s/it]
Training Epoch: 14/100, batch 118/438 completed (loss: 0.0154:  27%|██▋       | 59/219 [05:58<16:13,  6.08s/it]
Training Epoch: 14/100, batch 118/438 completed (loss: 0.02:  27%|██▋       | 59/219 [05:58<16:12,  6.08s/it]  
Training Epoch: 14/100, batch 118/438 completed (loss: 0.0154:  27%|██▋       | 60/219 [06:01<16:01,  6.05s/it]
Training Epoch: 14/100, batch 118/438 completed (loss: 0.1633:  27%|██▋       | 60/219 [06:01<16:02,  6.05s/it]
Training Epoch: 14/100, batch 118/438 completed (loss: 0.02:  27%|██▋       | 60/219 [06:01<16:01,  6.05s/it]
Training Epoch: 14/100, batch 118/438 completed (loss: 0.0448:  27%|██▋       | 60/219 [06:01<16:02,  6.05s/it]
Training Epoch: 14/100, batch 119/438 completed (loss: 0.0798:  27%|██▋       | 60/219 [06:01<16:02,  6.05s/it]
Training Epoch: 14/100, batch 119/438 completed (loss: 0.0384:  27%|██▋       | 60/219 [06:01<16:02,  6.05s/it]
Training Epoch: 14/100, batch 119/438 completed (loss: 0.1294:  27%|██▋       | 60/219 [06:01<16:01,  6.05s/it]
Training Epoch: 14/100, batch 119/438 completed (loss: 0.0098:  27%|██▋       | 60/219 [06:01<16:01,  6.05s/it]
Training Epoch: 14/100, batch 120/438 completed (loss: 0.0088:  27%|██▋       | 60/219 [06:04<16:01,  6.05s/it]
Training Epoch: 14/100, batch 120/438 completed (loss: 0.2513:  27%|██▋       | 60/219 [06:04<16:02,  6.05s/it]
Training Epoch: 14/100, batch 120/438 completed (loss: 0.2361:  27%|██▋       | 60/219 [06:04<16:02,  6.05s/it]
Training Epoch: 14/100, batch 120/438 completed (loss: 0.0343:  27%|██▋       | 60/219 [06:04<16:01,  6.05s/it]
Training Epoch: 14/100, batch 120/438 completed (loss: 0.0343:  28%|██▊       | 61/219 [06:07<15:53,  6.03s/it]
Training Epoch: 14/100, batch 120/438 completed (loss: 0.0088:  28%|██▊       | 61/219 [06:07<15:53,  6.03s/it]
Training Epoch: 14/100, batch 120/438 completed (loss: 0.2361:  28%|██▊       | 61/219 [06:07<15:53,  6.03s/it]
Training Epoch: 14/100, batch 120/438 completed (loss: 0.2513:  28%|██▊       | 61/219 [06:07<15:53,  6.03s/it]
Training Epoch: 14/100, batch 121/438 completed (loss: 0.1038:  28%|██▊       | 61/219 [06:07<15:53,  6.03s/it]
Training Epoch: 14/100, batch 121/438 completed (loss: 0.0077:  28%|██▊       | 61/219 [06:07<15:53,  6.03s/it]
Training Epoch: 14/100, batch 121/438 completed (loss: 0.1489:  28%|██▊       | 61/219 [06:07<15:53,  6.03s/it]
Training Epoch: 14/100, batch 121/438 completed (loss: 0.0387:  28%|██▊       | 61/219 [06:07<15:53,  6.03s/it]
Training Epoch: 14/100, batch 122/438 completed (loss: 0.4285:  28%|██▊       | 61/219 [06:10<15:53,  6.03s/it]
Training Epoch: 14/100, batch 122/438 completed (loss: 0.0279:  28%|██▊       | 61/219 [06:10<15:53,  6.03s/it]
Training Epoch: 14/100, batch 122/438 completed (loss: 0.0312:  28%|██▊       | 61/219 [06:10<15:53,  6.03s/it]
Training Epoch: 14/100, batch 122/438 completed (loss: 0.0121:  28%|██▊       | 61/219 [06:10<15:53,  6.03s/it]
Training Epoch: 14/100, batch 122/438 completed (loss: 0.0279:  28%|██▊       | 62/219 [06:13<15:58,  6.11s/it]
Training Epoch: 14/100, batch 122/438 completed (loss: 0.4285:  28%|██▊       | 62/219 [06:13<15:58,  6.11s/it]
Training Epoch: 14/100, batch 122/438 completed (loss: 0.0121:  28%|██▊       | 62/219 [06:13<15:58,  6.11s/it]
Training Epoch: 14/100, batch 122/438 completed (loss: 0.0312:  28%|██▊       | 62/219 [06:13<15:58,  6.11s/it]
Training Epoch: 14/100, batch 123/438 completed (loss: 0.231:  28%|██▊       | 62/219 [06:13<15:58,  6.11s/it] 
Training Epoch: 14/100, batch 123/438 completed (loss: 0.0405:  28%|██▊       | 62/219 [06:13<15:58,  6.11s/it]
Training Epoch: 14/100, batch 123/438 completed (loss: 0.1187:  28%|██▊       | 62/219 [06:13<15:58,  6.11s/it]
Training Epoch: 14/100, batch 123/438 completed (loss: 0.0509:  28%|██▊       | 62/219 [06:13<15:58,  6.11s/it]
Training Epoch: 14/100, batch 124/438 completed (loss: 0.2817:  28%|██▊       | 62/219 [06:16<15:58,  6.11s/it]
Training Epoch: 14/100, batch 124/438 completed (loss: 0.0422:  28%|██▊       | 62/219 [06:16<15:58,  6.11s/it]
Training Epoch: 14/100, batch 124/438 completed (loss: 0.0791:  28%|██▊       | 62/219 [06:16<15:58,  6.11s/it]
Training Epoch: 14/100, batch 124/438 completed (loss: 0.4183:  28%|██▊       | 62/219 [06:16<15:58,  6.11s/it]
Training Epoch: 14/100, batch 124/438 completed (loss: 0.4183:  29%|██▉       | 63/219 [06:19<15:39,  6.02s/it]
Training Epoch: 14/100, batch 124/438 completed (loss: 0.2817:  29%|██▉       | 63/219 [06:19<15:39,  6.02s/it]
Training Epoch: 14/100, batch 124/438 completed (loss: 0.0791:  29%|██▉       | 63/219 [06:19<15:39,  6.02s/it]
Training Epoch: 14/100, batch 124/438 completed (loss: 0.0422:  29%|██▉       | 63/219 [06:19<15:39,  6.02s/it]
Training Epoch: 14/100, batch 125/438 completed (loss: 0.1513:  29%|██▉       | 63/219 [06:19<15:39,  6.02s/it]
Training Epoch: 14/100, batch 125/438 completed (loss: 0.3687:  29%|██▉       | 63/219 [06:19<15:39,  6.02s/it]
Training Epoch: 14/100, batch 125/438 completed (loss: 0.0119:  29%|██▉       | 63/219 [06:19<15:39,  6.02s/it]
Training Epoch: 14/100, batch 125/438 completed (loss: 0.0345:  29%|██▉       | 63/219 [06:19<15:39,  6.02s/it]
Training Epoch: 14/100, batch 126/438 completed (loss: 0.0886:  29%|██▉       | 63/219 [06:22<15:39,  6.02s/it]
Training Epoch: 14/100, batch 126/438 completed (loss: 0.4717:  29%|██▉       | 63/219 [06:22<15:39,  6.02s/it]
Training Epoch: 14/100, batch 126/438 completed (loss: 0.0093:  29%|██▉       | 63/219 [06:22<15:39,  6.02s/it]
Training Epoch: 14/100, batch 126/438 completed (loss: 0.1181:  29%|██▉       | 63/219 [06:22<15:39,  6.02s/it]
Training Epoch: 14/100, batch 126/438 completed (loss: 0.0093:  29%|██▉       | 64/219 [06:25<15:38,  6.06s/it]
Training Epoch: 14/100, batch 126/438 completed (loss: 0.1181:  29%|██▉       | 64/219 [06:25<15:39,  6.06s/it]
Training Epoch: 14/100, batch 126/438 completed (loss: 0.0886:  29%|██▉       | 64/219 [06:25<15:39,  6.06s/it]
Training Epoch: 14/100, batch 126/438 completed (loss: 0.4717:  29%|██▉       | 64/219 [06:25<15:39,  6.06s/it]
Training Epoch: 14/100, batch 127/438 completed (loss: 0.0491:  29%|██▉       | 64/219 [06:25<15:39,  6.06s/it]
Training Epoch: 14/100, batch 127/438 completed (loss: 0.0399:  29%|██▉       | 64/219 [06:25<15:38,  6.06s/it]
Training Epoch: 14/100, batch 127/438 completed (loss: 0.0111:  29%|██▉       | 64/219 [06:25<15:39,  6.06s/it]
Training Epoch: 14/100, batch 127/438 completed (loss: 0.0844:  29%|██▉       | 64/219 [06:25<15:39,  6.06s/it]
Training Epoch: 14/100, batch 128/438 completed (loss: 0.0199:  29%|██▉       | 64/219 [06:28<15:38,  6.06s/it]
Training Epoch: 14/100, batch 128/438 completed (loss: 0.0879:  29%|██▉       | 64/219 [06:28<15:39,  6.06s/it]
Training Epoch: 14/100, batch 128/438 completed (loss: 0.1352:  29%|██▉       | 64/219 [06:28<15:39,  6.06s/it]
Training Epoch: 14/100, batch 128/438 completed (loss: 0.1242:  29%|██▉       | 64/219 [06:28<15:39,  6.06s/it]
Training Epoch: 14/100, batch 128/438 completed (loss: 0.0199:  30%|██▉       | 65/219 [06:31<15:30,  6.04s/it]
Training Epoch: 14/100, batch 128/438 completed (loss: 0.1352:  30%|██▉       | 65/219 [06:31<15:30,  6.04s/it]
Training Epoch: 14/100, batch 128/438 completed (loss: 0.0879:  30%|██▉       | 65/219 [06:31<15:30,  6.04s/it]
Training Epoch: 14/100, batch 128/438 completed (loss: 0.1242:  30%|██▉       | 65/219 [06:31<15:30,  6.04s/it]
Training Epoch: 14/100, batch 129/438 completed (loss: 0.3763:  30%|██▉       | 65/219 [06:31<15:30,  6.04s/it]
Training Epoch: 14/100, batch 129/438 completed (loss: 0.0147:  30%|██▉       | 65/219 [06:31<15:30,  6.04s/it]
Training Epoch: 14/100, batch 129/438 completed (loss: 0.0664:  30%|██▉       | 65/219 [06:31<15:30,  6.04s/it]
Training Epoch: 14/100, batch 129/438 completed (loss: 0.2558:  30%|██▉       | 65/219 [06:31<15:30,  6.04s/it]
Training Epoch: 14/100, batch 130/438 completed (loss: 0.151:  30%|██▉       | 65/219 [06:34<15:30,  6.04s/it] 
Training Epoch: 14/100, batch 130/438 completed (loss: 0.0793:  30%|██▉       | 65/219 [06:34<15:30,  6.04s/it]
Training Epoch: 14/100, batch 130/438 completed (loss: 0.0114:  30%|██▉       | 65/219 [06:34<15:30,  6.04s/it]
Training Epoch: 14/100, batch 130/438 completed (loss: 0.012:  30%|██▉       | 65/219 [06:34<15:30,  6.04s/it] 
Training Epoch: 14/100, batch 130/438 completed (loss: 0.012:  30%|███       | 66/219 [06:37<15:15,  5.98s/it]
Training Epoch: 14/100, batch 130/438 completed (loss: 0.151:  30%|███       | 66/219 [06:37<15:15,  5.98s/it]
Training Epoch: 14/100, batch 130/438 completed (loss: 0.0793:  30%|███       | 66/219 [06:37<15:15,  5.98s/it]
Training Epoch: 14/100, batch 130/438 completed (loss: 0.0114:  30%|███       | 66/219 [06:37<15:15,  5.98s/it]
Training Epoch: 14/100, batch 131/438 completed (loss: 0.0131:  30%|███       | 66/219 [06:37<15:15,  5.98s/it]
Training Epoch: 14/100, batch 131/438 completed (loss: 0.0117:  30%|███       | 66/219 [06:37<15:15,  5.98s/it]
Training Epoch: 14/100, batch 131/438 completed (loss: 0.1052:  30%|███       | 66/219 [06:37<15:15,  5.98s/it]
Training Epoch: 14/100, batch 131/438 completed (loss: 0.2949:  30%|███       | 66/219 [06:37<15:15,  5.98s/it]
Training Epoch: 14/100, batch 132/438 completed (loss: 0.0584:  30%|███       | 66/219 [06:40<15:15,  5.98s/it]
Training Epoch: 14/100, batch 132/438 completed (loss: 0.022:  30%|███       | 66/219 [06:40<15:15,  5.98s/it] 
Training Epoch: 14/100, batch 132/438 completed (loss: 0.2289:  30%|███       | 66/219 [06:40<15:15,  5.98s/it]
Training Epoch: 14/100, batch 132/438 completed (loss: 0.0148:  30%|███       | 66/219 [06:40<15:15,  5.98s/it]
Training Epoch: 14/100, batch 132/438 completed (loss: 0.0148:  31%|███       | 67/219 [06:43<15:24,  6.08s/it]
Training Epoch: 14/100, batch 132/438 completed (loss: 0.2289:  31%|███       | 67/219 [06:43<15:24,  6.08s/it]
Training Epoch: 14/100, batch 132/438 completed (loss: 0.022:  31%|███       | 67/219 [06:43<15:24,  6.08s/it]
Training Epoch: 14/100, batch 132/438 completed (loss: 0.0584:  31%|███       | 67/219 [06:43<15:24,  6.09s/it]
Training Epoch: 14/100, batch 133/438 completed (loss: 0.1963:  31%|███       | 67/219 [06:43<15:24,  6.08s/it]
Training Epoch: 14/100, batch 133/438 completed (loss: 0.0046:  31%|███       | 67/219 [06:43<15:24,  6.09s/it]
Training Epoch: 14/100, batch 133/438 completed (loss: 0.0606:  31%|███       | 67/219 [06:43<15:24,  6.08s/it]
Training Epoch: 14/100, batch 133/438 completed (loss: 0.0775:  31%|███       | 67/219 [06:43<15:24,  6.08s/it]
Training Epoch: 14/100, batch 134/438 completed (loss: 0.0245:  31%|███       | 67/219 [06:46<15:24,  6.09s/it]
Training Epoch: 14/100, batch 134/438 completed (loss: 0.0233:  31%|███       | 67/219 [06:46<15:24,  6.08s/it]
Training Epoch: 14/100, batch 134/438 completed (loss: 0.0341:  31%|███       | 67/219 [06:46<15:24,  6.08s/it]
Training Epoch: 14/100, batch 134/438 completed (loss: 0.0247:  31%|███       | 67/219 [06:46<15:24,  6.08s/it]
Training Epoch: 14/100, batch 134/438 completed (loss: 0.0341:  31%|███       | 68/219 [06:49<15:13,  6.05s/it]
Training Epoch: 14/100, batch 134/438 completed (loss: 0.0233:  31%|███       | 68/219 [06:49<15:13,  6.05s/it]
Training Epoch: 14/100, batch 134/438 completed (loss: 0.0247:  31%|███       | 68/219 [06:49<15:13,  6.05s/it]
Training Epoch: 14/100, batch 134/438 completed (loss: 0.0245:  31%|███       | 68/219 [06:49<15:13,  6.05s/it]
Training Epoch: 14/100, batch 135/438 completed (loss: 0.028:  31%|███       | 68/219 [06:49<15:13,  6.05s/it] 
Training Epoch: 14/100, batch 135/438 completed (loss: 0.0045:  31%|███       | 68/219 [06:49<15:13,  6.05s/it]
Training Epoch: 14/100, batch 135/438 completed (loss: 0.0574:  31%|███       | 68/219 [06:49<15:13,  6.05s/it]
Training Epoch: 14/100, batch 135/438 completed (loss: 0.0722:  31%|███       | 68/219 [06:49<15:13,  6.05s/it]
Training Epoch: 14/100, batch 136/438 completed (loss: 0.2626:  31%|███       | 68/219 [06:52<15:13,  6.05s/it]
Training Epoch: 14/100, batch 136/438 completed (loss: 0.0504:  31%|███       | 68/219 [06:52<15:13,  6.05s/it]
Training Epoch: 14/100, batch 136/438 completed (loss: 0.1764:  31%|███       | 68/219 [06:52<15:13,  6.05s/it]
Training Epoch: 14/100, batch 136/438 completed (loss: 0.006:  31%|███       | 68/219 [06:52<15:13,  6.05s/it] 
Training Epoch: 14/100, batch 136/438 completed (loss: 0.1764:  32%|███▏      | 69/219 [06:55<15:05,  6.04s/it]
Training Epoch: 14/100, batch 136/438 completed (loss: 0.006:  32%|███▏      | 69/219 [06:55<15:05,  6.04s/it]
Training Epoch: 14/100, batch 136/438 completed (loss: 0.2626:  32%|███▏      | 69/219 [06:55<15:05,  6.04s/it]
Training Epoch: 14/100, batch 136/438 completed (loss: 0.0504:  32%|███▏      | 69/219 [06:55<15:05,  6.04s/it]
Training Epoch: 14/100, batch 137/438 completed (loss: 0.011:  32%|███▏      | 69/219 [06:55<15:05,  6.04s/it] 
Training Epoch: 14/100, batch 137/438 completed (loss: 0.1262:  32%|███▏      | 69/219 [06:55<15:05,  6.04s/it]
Training Epoch: 14/100, batch 137/438 completed (loss: 0.018:  32%|███▏      | 69/219 [06:55<15:05,  6.04s/it] 
Training Epoch: 14/100, batch 137/438 completed (loss: 0.1089:  32%|███▏      | 69/219 [06:55<15:05,  6.04s/it]
Training Epoch: 14/100, batch 138/438 completed (loss: 0.1696:  32%|███▏      | 69/219 [06:58<15:05,  6.04s/it]
Training Epoch: 14/100, batch 138/438 completed (loss: 0.5749:  32%|███▏      | 69/219 [06:58<15:05,  6.04s/it]
Training Epoch: 14/100, batch 138/438 completed (loss: 0.0108:  32%|███▏      | 69/219 [06:58<15:05,  6.04s/it]
Training Epoch: 14/100, batch 138/438 completed (loss: 0.0652:  32%|███▏      | 69/219 [06:58<15:05,  6.04s/it]
Training Epoch: 14/100, batch 138/438 completed (loss: 0.0108:  32%|███▏      | 70/219 [07:01<15:09,  6.10s/it]
Training Epoch: 14/100, batch 138/438 completed (loss: 0.0652:  32%|███▏      | 70/219 [07:01<15:09,  6.11s/it]
Training Epoch: 14/100, batch 138/438 completed (loss: 0.1696:  32%|███▏      | 70/219 [07:01<15:09,  6.10s/it]
Training Epoch: 14/100, batch 138/438 completed (loss: 0.5749:  32%|███▏      | 70/219 [07:01<15:09,  6.11s/it]
Training Epoch: 14/100, batch 139/438 completed (loss: 0.3267:  32%|███▏      | 70/219 [07:01<15:09,  6.10s/it]
Training Epoch: 14/100, batch 139/438 completed (loss: 0.0774:  32%|███▏      | 70/219 [07:01<15:09,  6.10s/it]
Training Epoch: 14/100, batch 139/438 completed (loss: 0.1052:  32%|███▏      | 70/219 [07:01<15:09,  6.11s/it]
Training Epoch: 14/100, batch 139/438 completed (loss: 0.0738:  32%|███▏      | 70/219 [07:01<15:09,  6.11s/it]
Training Epoch: 14/100, batch 140/438 completed (loss: 0.0401:  32%|███▏      | 70/219 [07:04<15:09,  6.10s/it]
Training Epoch: 14/100, batch 140/438 completed (loss: 0.0371:  32%|███▏      | 70/219 [07:04<15:09,  6.10s/it]
Training Epoch: 14/100, batch 140/438 completed (loss: 0.0247:  32%|███▏      | 70/219 [07:04<15:09,  6.11s/it]
Training Epoch: 14/100, batch 140/438 completed (loss: 0.0463:  32%|███▏      | 70/219 [07:04<15:09,  6.11s/it]
Training Epoch: 14/100, batch 140/438 completed (loss: 0.0247:  32%|███▏      | 71/219 [07:07<14:50,  6.02s/it]
Training Epoch: 14/100, batch 140/438 completed (loss: 0.0401:  32%|███▏      | 71/219 [07:07<14:51,  6.02s/it]
Training Epoch: 14/100, batch 140/438 completed (loss: 0.0371:  32%|███▏      | 71/219 [07:07<14:50,  6.02s/it]
Training Epoch: 14/100, batch 140/438 completed (loss: 0.0463:  32%|███▏      | 71/219 [07:07<14:51,  6.02s/it]
Training Epoch: 14/100, batch 141/438 completed (loss: 0.0139:  32%|███▏      | 71/219 [07:07<14:50,  6.02s/it]
Training Epoch: 14/100, batch 141/438 completed (loss: 0.1353:  32%|███▏      | 71/219 [07:07<14:51,  6.02s/it]
Training Epoch: 14/100, batch 141/438 completed (loss: 0.015:  32%|███▏      | 71/219 [07:07<14:50,  6.02s/it] 
Training Epoch: 14/100, batch 141/438 completed (loss: 0.0332:  32%|███▏      | 71/219 [07:07<14:51,  6.02s/it]
Training Epoch: 14/100, batch 142/438 completed (loss: 0.1005:  32%|███▏      | 71/219 [07:10<14:50,  6.02s/it]
Training Epoch: 14/100, batch 142/438 completed (loss: 0.0468:  32%|███▏      | 71/219 [07:10<14:51,  6.02s/it]
Training Epoch: 14/100, batch 142/438 completed (loss: 0.1439:  32%|███▏      | 71/219 [07:10<14:50,  6.02s/it]
Training Epoch: 14/100, batch 142/438 completed (loss: 0.0239:  32%|███▏      | 71/219 [07:10<14:51,  6.02s/it]
Training Epoch: 14/100, batch 142/438 completed (loss: 0.0468:  33%|███▎      | 72/219 [07:13<14:55,  6.10s/it]
Training Epoch: 14/100, batch 142/438 completed (loss: 0.1439:  33%|███▎      | 72/219 [07:13<14:55,  6.09s/it]
Training Epoch: 14/100, batch 142/438 completed (loss: 0.0239:  33%|███▎      | 72/219 [07:13<14:55,  6.10s/it]
Training Epoch: 14/100, batch 142/438 completed (loss: 0.1005:  33%|███▎      | 72/219 [07:13<14:56,  6.10s/it]
Training Epoch: 14/100, batch 143/438 completed (loss: 0.191:  33%|███▎      | 72/219 [07:13<14:56,  6.10s/it] 
Training Epoch: 14/100, batch 143/438 completed (loss: 0.3851:  33%|███▎      | 72/219 [07:13<14:55,  6.09s/it]
Training Epoch: 14/100, batch 143/438 completed (loss: 0.0468:  33%|███▎      | 72/219 [07:13<14:55,  6.10s/it]
Training Epoch: 14/100, batch 143/438 completed (loss: 0.0182:  33%|███▎      | 72/219 [07:13<14:55,  6.10s/it]
Training Epoch: 14/100, batch 144/438 completed (loss: 0.1198:  33%|███▎      | 72/219 [07:16<14:56,  6.10s/it]
Training Epoch: 14/100, batch 144/438 completed (loss: 0.1232:  33%|███▎      | 72/219 [07:16<14:55,  6.10s/it]
Training Epoch: 14/100, batch 144/438 completed (loss: 0.1491:  33%|███▎      | 72/219 [07:16<14:55,  6.09s/it]
Training Epoch: 14/100, batch 144/438 completed (loss: 0.0117:  33%|███▎      | 72/219 [07:16<14:55,  6.10s/it]
Training Epoch: 14/100, batch 144/438 completed (loss: 0.0117:  33%|███▎      | 73/219 [07:19<14:45,  6.06s/it]
Training Epoch: 14/100, batch 144/438 completed (loss: 0.1198:  33%|███▎      | 73/219 [07:19<14:45,  6.07s/it]
Training Epoch: 14/100, batch 144/438 completed (loss: 0.1491:  33%|███▎      | 73/219 [07:19<14:45,  6.07s/it]
Training Epoch: 14/100, batch 144/438 completed (loss: 0.1232:  33%|███▎      | 73/219 [07:19<14:45,  6.07s/it]
Training Epoch: 14/100, batch 145/438 completed (loss: 0.0905:  33%|███▎      | 73/219 [07:19<14:45,  6.07s/it]
Training Epoch: 14/100, batch 145/438 completed (loss: 0.0208:  33%|███▎      | 73/219 [07:19<14:45,  6.07s/it]
Training Epoch: 14/100, batch 145/438 completed (loss: 0.094:  33%|███▎      | 73/219 [07:19<14:45,  6.07s/it] 
Training Epoch: 14/100, batch 145/438 completed (loss: 0.0877:  33%|███▎      | 73/219 [07:19<14:45,  6.06s/it]
Training Epoch: 14/100, batch 146/438 completed (loss: 0.0358:  33%|███▎      | 73/219 [07:22<14:45,  6.07s/it]
Training Epoch: 14/100, batch 146/438 completed (loss: 0.0075:  33%|███▎      | 73/219 [07:22<14:45,  6.07s/it]
Training Epoch: 14/100, batch 146/438 completed (loss: 0.0395:  33%|███▎      | 73/219 [07:22<14:45,  6.07s/it]
Training Epoch: 14/100, batch 146/438 completed (loss: 0.0224:  33%|███▎      | 73/219 [07:22<14:45,  6.06s/it]
Training Epoch: 14/100, batch 146/438 completed (loss: 0.0224:  34%|███▍      | 74/219 [07:25<14:29,  6.00s/it]
Training Epoch: 14/100, batch 146/438 completed (loss: 0.0395:  34%|███▍      | 74/219 [07:25<14:30,  6.00s/it]
Training Epoch: 14/100, batch 146/438 completed (loss: 0.0358:  34%|███▍      | 74/219 [07:25<14:30,  6.00s/it]
Training Epoch: 14/100, batch 146/438 completed (loss: 0.0075:  34%|███▍      | 74/219 [07:25<14:30,  6.00s/it]
Training Epoch: 14/100, batch 147/438 completed (loss: 0.0827:  34%|███▍      | 74/219 [07:25<14:30,  6.00s/it]
Training Epoch: 14/100, batch 147/438 completed (loss: 0.4459:  34%|███▍      | 74/219 [07:25<14:30,  6.00s/it]
Training Epoch: 14/100, batch 147/438 completed (loss: 0.0606:  34%|███▍      | 74/219 [07:25<14:30,  6.00s/it]
Training Epoch: 14/100, batch 147/438 completed (loss: 0.0307:  34%|███▍      | 74/219 [07:25<14:29,  6.00s/it]
Training Epoch: 14/100, batch 148/438 completed (loss: 0.1272:  34%|███▍      | 74/219 [07:29<14:30,  6.00s/it]
Training Epoch: 14/100, batch 148/438 completed (loss: 0.0994:  34%|███▍      | 74/219 [07:29<14:30,  6.00s/it]
Training Epoch: 14/100, batch 148/438 completed (loss: 0.1445:  34%|███▍      | 74/219 [07:29<14:30,  6.00s/it]
Training Epoch: 14/100, batch 148/438 completed (loss: 0.0977:  34%|███▍      | 74/219 [07:29<14:29,  6.00s/it]
Training Epoch: 14/100, batch 148/438 completed (loss: 0.0977:  34%|███▍      | 75/219 [07:32<14:44,  6.14s/it]
Training Epoch: 14/100, batch 148/438 completed (loss: 0.1272:  34%|███▍      | 75/219 [07:32<14:44,  6.14s/it]
Training Epoch: 14/100, batch 148/438 completed (loss: 0.0994:  34%|███▍      | 75/219 [07:32<14:44,  6.14s/it]
Training Epoch: 14/100, batch 148/438 completed (loss: 0.1445:  34%|███▍      | 75/219 [07:32<14:44,  6.14s/it]
Training Epoch: 14/100, batch 149/438 completed (loss: 0.1936:  34%|███▍      | 75/219 [07:32<14:44,  6.14s/it]
Training Epoch: 14/100, batch 149/438 completed (loss: 0.1649:  34%|███▍      | 75/219 [07:32<14:44,  6.14s/it]
Training Epoch: 14/100, batch 149/438 completed (loss: 0.0401:  34%|███▍      | 75/219 [07:32<14:44,  6.14s/it]
Training Epoch: 14/100, batch 149/438 completed (loss: 0.1316:  34%|███▍      | 75/219 [07:32<14:44,  6.14s/it]
Training Epoch: 14/100, batch 150/438 completed (loss: 0.0186:  34%|███▍      | 75/219 [07:35<14:44,  6.14s/it]
Training Epoch: 14/100, batch 150/438 completed (loss: 0.0629:  34%|███▍      | 75/219 [07:35<14:44,  6.14s/it]
Training Epoch: 14/100, batch 150/438 completed (loss: 0.0197:  34%|███▍      | 75/219 [07:35<14:44,  6.14s/it]
Training Epoch: 14/100, batch 150/438 completed (loss: 0.0274:  34%|███▍      | 75/219 [07:35<14:44,  6.14s/it]
Training Epoch: 14/100, batch 150/438 completed (loss: 0.0274:  35%|███▍      | 76/219 [07:37<14:24,  6.05s/it]
Training Epoch: 14/100, batch 150/438 completed (loss: 0.0186:  35%|███▍      | 76/219 [07:37<14:24,  6.05s/it]
Training Epoch: 14/100, batch 150/438 completed (loss: 0.0629:  35%|███▍      | 76/219 [07:37<14:24,  6.05s/it]
Training Epoch: 14/100, batch 150/438 completed (loss: 0.0197:  35%|███▍      | 76/219 [07:37<14:24,  6.05s/it]
Training Epoch: 14/100, batch 151/438 completed (loss: 0.0319:  35%|███▍      | 76/219 [07:37<14:24,  6.05s/it]
Training Epoch: 14/100, batch 151/438 completed (loss: 0.027:  35%|███▍      | 76/219 [07:37<14:24,  6.05s/it] 
Training Epoch: 14/100, batch 151/438 completed (loss: 0.0543:  35%|███▍      | 76/219 [07:37<14:24,  6.05s/it]
Training Epoch: 14/100, batch 151/438 completed (loss: 0.0178:  35%|███▍      | 76/219 [07:38<14:24,  6.05s/it]
Training Epoch: 14/100, batch 152/438 completed (loss: 0.0229:  35%|███▍      | 76/219 [07:40<14:24,  6.05s/it]
Training Epoch: 14/100, batch 152/438 completed (loss: 0.0181:  35%|███▍      | 76/219 [07:40<14:24,  6.05s/it]
Training Epoch: 14/100, batch 152/438 completed (loss: 0.1016:  35%|███▍      | 76/219 [07:40<14:24,  6.05s/it]
Training Epoch: 14/100, batch 152/438 completed (loss: 0.0991:  35%|███▍      | 76/219 [07:40<14:24,  6.05s/it]
Training Epoch: 14/100, batch 152/438 completed (loss: 0.0991:  35%|███▌      | 77/219 [07:43<14:16,  6.03s/it]
Training Epoch: 14/100, batch 152/438 completed (loss: 0.0229:  35%|███▌      | 77/219 [07:43<14:16,  6.03s/it]
Training Epoch: 14/100, batch 152/438 completed (loss: 0.1016:  35%|███▌      | 77/219 [07:43<14:16,  6.03s/it]
Training Epoch: 14/100, batch 152/438 completed (loss: 0.0181:  35%|███▌      | 77/219 [07:43<14:16,  6.03s/it]
Training Epoch: 14/100, batch 153/438 completed (loss: 0.2905:  35%|███▌      | 77/219 [07:43<14:16,  6.03s/it]
Training Epoch: 14/100, batch 153/438 completed (loss: 0.0368:  35%|███▌      | 77/219 [07:43<14:16,  6.03s/it]
Training Epoch: 14/100, batch 153/438 completed (loss: 0.0619:  35%|███▌      | 77/219 [07:43<14:16,  6.03s/it]
Training Epoch: 14/100, batch 153/438 completed (loss: 0.1681:  35%|███▌      | 77/219 [07:44<14:16,  6.03s/it]
Training Epoch: 14/100, batch 154/438 completed (loss: 0.1033:  35%|███▌      | 77/219 [07:47<14:16,  6.03s/it]
Training Epoch: 14/100, batch 154/438 completed (loss: 0.143:  35%|███▌      | 77/219 [07:47<14:16,  6.03s/it] 
Training Epoch: 14/100, batch 154/438 completed (loss: 0.053:  35%|███▌      | 77/219 [07:47<14:16,  6.03s/it] 
Training Epoch: 14/100, batch 154/438 completed (loss: 0.0072:  35%|███▌      | 77/219 [07:47<14:16,  6.03s/it]
Training Epoch: 14/100, batch 154/438 completed (loss: 0.143:  36%|███▌      | 78/219 [07:50<14:21,  6.11s/it]
Training Epoch: 14/100, batch 154/438 completed (loss: 0.053:  36%|███▌      | 78/219 [07:50<14:21,  6.11s/it]
Training Epoch: 14/100, batch 154/438 completed (loss: 0.0072:  36%|███▌      | 78/219 [07:50<14:21,  6.11s/it]
Training Epoch: 14/100, batch 154/438 completed (loss: 0.1033:  36%|███▌      | 78/219 [07:50<14:21,  6.11s/it]
Training Epoch: 14/100, batch 155/438 completed (loss: 0.0393:  36%|███▌      | 78/219 [07:50<14:21,  6.11s/it]
Training Epoch: 14/100, batch 155/438 completed (loss: 0.0639:  36%|███▌      | 78/219 [07:50<14:21,  6.11s/it]
Training Epoch: 14/100, batch 155/438 completed (loss: 0.0085:  36%|███▌      | 78/219 [07:50<14:21,  6.11s/it]
Training Epoch: 14/100, batch 155/438 completed (loss: 0.0622:  36%|███▌      | 78/219 [07:50<14:21,  6.11s/it]
Training Epoch: 14/100, batch 156/438 completed (loss: 0.0297:  36%|███▌      | 78/219 [07:53<14:21,  6.11s/it]
Training Epoch: 14/100, batch 156/438 completed (loss: 0.046:  36%|███▌      | 78/219 [07:53<14:21,  6.11s/it] 
Training Epoch: 14/100, batch 156/438 completed (loss: 0.4692:  36%|███▌      | 78/219 [07:53<14:21,  6.11s/it]
Training Epoch: 14/100, batch 156/438 completed (loss: 0.0709:  36%|███▌      | 78/219 [07:53<14:21,  6.11s/it]
Training Epoch: 14/100, batch 156/438 completed (loss: 0.4692:  36%|███▌      | 79/219 [07:56<14:03,  6.02s/it]
Training Epoch: 14/100, batch 156/438 completed (loss: 0.046:  36%|███▌      | 79/219 [07:56<14:03,  6.02s/it]
Training Epoch: 14/100, batch 156/438 completed (loss: 0.0297:  36%|███▌      | 79/219 [07:56<14:03,  6.02s/it]
Training Epoch: 14/100, batch 156/438 completed (loss: 0.0709:  36%|███▌      | 79/219 [07:56<14:03,  6.03s/it]
Training Epoch: 14/100, batch 157/438 completed (loss: 0.1063:  36%|███▌      | 79/219 [07:56<14:03,  6.02s/it]
Training Epoch: 14/100, batch 157/438 completed (loss: 0.2048:  36%|███▌      | 79/219 [07:56<14:03,  6.02s/it]
Training Epoch: 14/100, batch 157/438 completed (loss: 0.0994:  36%|███▌      | 79/219 [07:56<14:03,  6.02s/it]
Training Epoch: 14/100, batch 157/438 completed (loss: 0.0697:  36%|███▌      | 79/219 [07:56<14:03,  6.03s/it]
Training Epoch: 14/100, batch 158/438 completed (loss: 0.2721:  36%|███▌      | 79/219 [07:59<14:03,  6.02s/it]
Training Epoch: 14/100, batch 158/438 completed (loss: 0.0214:  36%|███▌      | 79/219 [07:59<14:03,  6.02s/it]
Training Epoch: 14/100, batch 158/438 completed (loss: 0.0359:  36%|███▌      | 79/219 [07:59<14:03,  6.02s/it]
Training Epoch: 14/100, batch 158/438 completed (loss: 0.0489:  36%|███▌      | 79/219 [07:59<14:03,  6.03s/it]
Training Epoch: 14/100, batch 158/438 completed (loss: 0.2721:  37%|███▋      | 80/219 [08:02<14:15,  6.16s/it]
Training Epoch: 14/100, batch 158/438 completed (loss: 0.0489:  37%|███▋      | 80/219 [08:02<14:16,  6.16s/it]
Training Epoch: 14/100, batch 158/438 completed (loss: 0.0359:  37%|███▋      | 80/219 [08:02<14:16,  6.16s/it]
Training Epoch: 14/100, batch 158/438 completed (loss: 0.0214:  37%|███▋      | 80/219 [08:02<14:16,  6.16s/it]
Training Epoch: 14/100, batch 159/438 completed (loss: 0.017:  37%|███▋      | 80/219 [08:02<14:15,  6.16s/it] 
Training Epoch: 14/100, batch 159/438 completed (loss: 0.0039:  37%|███▋      | 80/219 [08:02<14:16,  6.16s/it]
Training Epoch: 14/100, batch 159/438 completed (loss: 0.0235:  37%|███▋      | 80/219 [08:02<14:16,  6.16s/it]
Training Epoch: 14/100, batch 159/438 completed (loss: 0.0273:  37%|███▋      | 80/219 [08:02<14:16,  6.16s/it]
Training Epoch: 14/100, batch 160/438 completed (loss: 0.0374:  37%|███▋      | 80/219 [08:05<14:15,  6.16s/it]
Training Epoch: 14/100, batch 160/438 completed (loss: 0.0193:  37%|███▋      | 80/219 [08:05<14:16,  6.16s/it]
Training Epoch: 14/100, batch 160/438 completed (loss: 0.1695:  37%|███▋      | 80/219 [08:05<14:16,  6.16s/it]
Training Epoch: 14/100, batch 160/438 completed (loss: 0.0838:  37%|███▋      | 80/219 [08:05<14:16,  6.16s/it]
Training Epoch: 14/100, batch 160/438 completed (loss: 0.0838:  37%|███▋      | 81/219 [08:08<13:56,  6.06s/it]
Training Epoch: 14/100, batch 160/438 completed (loss: 0.0374:  37%|███▋      | 81/219 [08:08<13:56,  6.06s/it]
Training Epoch: 14/100, batch 160/438 completed (loss: 0.1695:  37%|███▋      | 81/219 [08:08<13:56,  6.06s/it]
Training Epoch: 14/100, batch 160/438 completed (loss: 0.0193:  37%|███▋      | 81/219 [08:08<13:56,  6.06s/it]
Training Epoch: 14/100, batch 161/438 completed (loss: 0.0154:  37%|███▋      | 81/219 [08:08<13:56,  6.06s/it]
Training Epoch: 14/100, batch 161/438 completed (loss: 0.2213:  37%|███▋      | 81/219 [08:08<13:56,  6.06s/it]
Training Epoch: 14/100, batch 161/438 completed (loss: 0.0541:  37%|███▋      | 81/219 [08:08<13:56,  6.06s/it]
Training Epoch: 14/100, batch 161/438 completed (loss: 0.2603:  37%|███▋      | 81/219 [08:08<13:56,  6.06s/it]
Training Epoch: 14/100, batch 162/438 completed (loss: 0.0452:  37%|███▋      | 81/219 [08:11<13:56,  6.06s/it]
Training Epoch: 14/100, batch 162/438 completed (loss: 0.0411:  37%|███▋      | 81/219 [08:11<13:56,  6.06s/it]
Training Epoch: 14/100, batch 162/438 completed (loss: 0.0539:  37%|███▋      | 81/219 [08:11<13:56,  6.06s/it]
Training Epoch: 14/100, batch 162/438 completed (loss: 0.0413:  37%|███▋      | 81/219 [08:11<13:56,  6.06s/it]
Training Epoch: 14/100, batch 162/438 completed (loss: 0.0452:  37%|███▋      | 82/219 [08:14<13:48,  6.05s/it]
Training Epoch: 14/100, batch 162/438 completed (loss: 0.0539:  37%|███▋      | 82/219 [08:14<13:48,  6.05s/it]
Training Epoch: 14/100, batch 162/438 completed (loss: 0.0413:  37%|███▋      | 82/219 [08:14<13:48,  6.05s/it]
Training Epoch: 14/100, batch 162/438 completed (loss: 0.0411:  37%|███▋      | 82/219 [08:14<13:48,  6.05s/it]
Training Epoch: 14/100, batch 163/438 completed (loss: 0.1007:  37%|███▋      | 82/219 [08:14<13:48,  6.05s/it]
Training Epoch: 14/100, batch 163/438 completed (loss: 0.1768:  37%|███▋      | 82/219 [08:14<13:48,  6.05s/it]
Training Epoch: 14/100, batch 163/438 completed (loss: 0.0566:  37%|███▋      | 82/219 [08:14<13:48,  6.05s/it]
Training Epoch: 14/100, batch 163/438 completed (loss: 0.05:  37%|███▋      | 82/219 [08:14<13:48,  6.05s/it]  
Training Epoch: 14/100, batch 164/438 completed (loss: 0.0172:  37%|███▋      | 82/219 [08:17<13:48,  6.05s/it]
Training Epoch: 14/100, batch 164/438 completed (loss: 0.0516:  37%|███▋      | 82/219 [08:17<13:48,  6.05s/it]
Training Epoch: 14/100, batch 164/438 completed (loss: 0.0585:  37%|███▋      | 82/219 [08:17<13:48,  6.05s/it]
Training Epoch: 14/100, batch 164/438 completed (loss: 0.0401:  37%|███▋      | 82/219 [08:17<13:48,  6.05s/it]
Training Epoch: 14/100, batch 164/438 completed (loss: 0.0172:  38%|███▊      | 83/219 [08:20<13:51,  6.12s/it]
Training Epoch: 14/100, batch 164/438 completed (loss: 0.0585:  38%|███▊      | 83/219 [08:20<13:51,  6.12s/it]
Training Epoch: 14/100, batch 164/438 completed (loss: 0.0516:  38%|███▊      | 83/219 [08:20<13:51,  6.12s/it]
Training Epoch: 14/100, batch 164/438 completed (loss: 0.0401:  38%|███▊      | 83/219 [08:20<13:52,  6.12s/it]
Training Epoch: 14/100, batch 165/438 completed (loss: 0.055:  38%|███▊      | 83/219 [08:20<13:51,  6.12s/it] 
Training Epoch: 14/100, batch 165/438 completed (loss: 0.0528:  38%|███▊      | 83/219 [08:20<13:51,  6.12s/it]
Training Epoch: 14/100, batch 165/438 completed (loss: 0.0766:  38%|███▊      | 83/219 [08:20<13:51,  6.12s/it]
Training Epoch: 14/100, batch 165/438 completed (loss: 0.0133:  38%|███▊      | 83/219 [08:20<13:52,  6.12s/it]
Training Epoch: 14/100, batch 166/438 completed (loss: 0.0087:  38%|███▊      | 83/219 [08:23<13:51,  6.12s/it]
Training Epoch: 14/100, batch 166/438 completed (loss: 0.0373:  38%|███▊      | 83/219 [08:23<13:51,  6.12s/it]
Training Epoch: 14/100, batch 166/438 completed (loss: 0.0178:  38%|███▊      | 83/219 [08:23<13:51,  6.12s/it]
Training Epoch: 14/100, batch 166/438 completed (loss: 0.0722:  38%|███▊      | 83/219 [08:23<13:52,  6.12s/it]
Training Epoch: 14/100, batch 166/438 completed (loss: 0.0722:  38%|███▊      | 84/219 [08:26<13:35,  6.04s/it]
Training Epoch: 14/100, batch 166/438 completed (loss: 0.0087:  38%|███▊      | 84/219 [08:26<13:35,  6.04s/it]
Training Epoch: 14/100, batch 166/438 completed (loss: 0.0373:  38%|███▊      | 84/219 [08:26<13:35,  6.04s/it]
Training Epoch: 14/100, batch 166/438 completed (loss: 0.0178:  38%|███▊      | 84/219 [08:26<13:35,  6.04s/it]
Training Epoch: 14/100, batch 167/438 completed (loss: 0.0083:  38%|███▊      | 84/219 [08:26<13:35,  6.04s/it]
Training Epoch: 14/100, batch 167/438 completed (loss: 0.0034:  38%|███▊      | 84/219 [08:26<13:35,  6.04s/it]
Training Epoch: 14/100, batch 167/438 completed (loss: 0.0774:  38%|███▊      | 84/219 [08:26<13:35,  6.04s/it]
Training Epoch: 14/100, batch 167/438 completed (loss: 0.1488:  38%|███▊      | 84/219 [08:26<13:35,  6.04s/it]
Training Epoch: 14/100, batch 168/438 completed (loss: 0.4458:  38%|███▊      | 84/219 [08:29<13:35,  6.04s/it]
Training Epoch: 14/100, batch 168/438 completed (loss: 0.0287:  38%|███▊      | 84/219 [08:29<13:35,  6.04s/it]
Training Epoch: 14/100, batch 168/438 completed (loss: 0.0391:  38%|███▊      | 84/219 [08:29<13:35,  6.04s/it]
Training Epoch: 14/100, batch 168/438 completed (loss: 0.0025:  38%|███▊      | 84/219 [08:29<13:35,  6.04s/it]
Training Epoch: 14/100, batch 168/438 completed (loss: 0.0287:  39%|███▉      | 85/219 [08:32<13:33,  6.07s/it]
Training Epoch: 14/100, batch 168/438 completed (loss: 0.0025:  39%|███▉      | 85/219 [08:32<13:33,  6.07s/it]
Training Epoch: 14/100, batch 168/438 completed (loss: 0.4458:  39%|███▉      | 85/219 [08:32<13:33,  6.07s/it]
Training Epoch: 14/100, batch 168/438 completed (loss: 0.0391:  39%|███▉      | 85/219 [08:32<13:33,  6.07s/it]
Training Epoch: 14/100, batch 169/438 completed (loss: 0.0752:  39%|███▉      | 85/219 [08:32<13:33,  6.07s/it]
Training Epoch: 14/100, batch 169/438 completed (loss: 0.0152:  39%|███▉      | 85/219 [08:32<13:33,  6.07s/it]
Training Epoch: 14/100, batch 169/438 completed (loss: 0.0377:  39%|███▉      | 85/219 [08:32<13:33,  6.07s/it]
Training Epoch: 14/100, batch 169/438 completed (loss: 0.1291:  39%|███▉      | 85/219 [08:32<13:33,  6.07s/it]
Training Epoch: 14/100, batch 170/438 completed (loss: 0.0923:  39%|███▉      | 85/219 [08:35<13:33,  6.07s/it]
Training Epoch: 14/100, batch 170/438 completed (loss: 0.7465:  39%|███▉      | 85/219 [08:35<13:33,  6.07s/it]
Training Epoch: 14/100, batch 170/438 completed (loss: 0.0306:  39%|███▉      | 85/219 [08:35<13:33,  6.07s/it]
Training Epoch: 14/100, batch 170/438 completed (loss: 0.0169:  39%|███▉      | 85/219 [08:35<13:33,  6.07s/it]
Training Epoch: 14/100, batch 170/438 completed (loss: 0.0169:  39%|███▉      | 86/219 [08:38<13:29,  6.08s/it]
Training Epoch: 14/100, batch 170/438 completed (loss: 0.0923:  39%|███▉      | 86/219 [08:38<13:29,  6.08s/it]
Training Epoch: 14/100, batch 170/438 completed (loss: 0.7465:  39%|███▉      | 86/219 [08:38<13:29,  6.08s/it]
Training Epoch: 14/100, batch 170/438 completed (loss: 0.0306:  39%|███▉      | 86/219 [08:38<13:29,  6.09s/it]
Training Epoch: 14/100, batch 171/438 completed (loss: 0.1669:  39%|███▉      | 86/219 [08:38<13:29,  6.08s/it]
Training Epoch: 14/100, batch 171/438 completed (loss: 0.3088:  39%|███▉      | 86/219 [08:38<13:29,  6.08s/it]
Training Epoch: 14/100, batch 171/438 completed (loss: 0.0112:  39%|███▉      | 86/219 [08:38<13:29,  6.09s/it]
Training Epoch: 14/100, batch 171/438 completed (loss: 0.0085:  39%|███▉      | 86/219 [08:38<13:29,  6.08s/it]
Training Epoch: 14/100, batch 172/438 completed (loss: 0.0333:  39%|███▉      | 86/219 [08:41<13:29,  6.08s/it]
Training Epoch: 14/100, batch 172/438 completed (loss: 0.0374:  39%|███▉      | 86/219 [08:41<13:29,  6.08s/it]
Training Epoch: 14/100, batch 172/438 completed (loss: 0.1058:  39%|███▉      | 86/219 [08:41<13:29,  6.09s/it]
Training Epoch: 14/100, batch 172/438 completed (loss: 0.0152:  39%|███▉      | 86/219 [08:41<13:29,  6.08s/it]
Training Epoch: 14/100, batch 172/438 completed (loss: 0.0374:  40%|███▉      | 87/219 [08:44<13:14,  6.02s/it]
Training Epoch: 14/100, batch 172/438 completed (loss: 0.0333:  40%|███▉      | 87/219 [08:44<13:13,  6.02s/it]
Training Epoch: 14/100, batch 172/438 completed (loss: 0.1058:  40%|███▉      | 87/219 [08:44<13:14,  6.02s/it]
Training Epoch: 14/100, batch 172/438 completed (loss: 0.0152:  40%|███▉      | 87/219 [08:44<13:14,  6.02s/it]
Training Epoch: 14/100, batch 173/438 completed (loss: 0.082:  40%|███▉      | 87/219 [08:44<13:13,  6.02s/it] 
Training Epoch: 14/100, batch 173/438 completed (loss: 0.009:  40%|███▉      | 87/219 [08:44<13:14,  6.02s/it] 
Training Epoch: 14/100, batch 173/438 completed (loss: 0.1044:  40%|███▉      | 87/219 [08:44<13:14,  6.02s/it]
Training Epoch: 14/100, batch 173/438 completed (loss: 0.0366:  40%|███▉      | 87/219 [08:44<13:14,  6.02s/it]
Training Epoch: 14/100, batch 174/438 completed (loss: 0.0189:  40%|███▉      | 87/219 [08:47<13:14,  6.02s/it]
Training Epoch: 14/100, batch 174/438 completed (loss: 0.0277:  40%|███▉      | 87/219 [08:47<13:14,  6.02s/it]
Training Epoch: 14/100, batch 174/438 completed (loss: 0.0945:  40%|███▉      | 87/219 [08:47<13:13,  6.02s/it]
Training Epoch: 14/100, batch 174/438 completed (loss: 0.0717:  40%|███▉      | 87/219 [08:47<13:14,  6.02s/it]
Training Epoch: 14/100, batch 174/438 completed (loss: 0.0717:  40%|████      | 88/219 [08:50<13:12,  6.05s/it]
Training Epoch: 14/100, batch 174/438 completed (loss: 0.0189:  40%|████      | 88/219 [08:50<13:12,  6.05s/it]
Training Epoch: 14/100, batch 174/438 completed (loss: 0.0945:  40%|████      | 88/219 [08:50<13:12,  6.05s/it]
Training Epoch: 14/100, batch 174/438 completed (loss: 0.0277:  40%|████      | 88/219 [08:50<13:13,  6.05s/it]
Training Epoch: 14/100, batch 175/438 completed (loss: 0.0253:  40%|████      | 88/219 [08:50<13:12,  6.05s/it]
Training Epoch: 14/100, batch 175/438 completed (loss: 0.1252:  40%|████      | 88/219 [08:50<13:12,  6.05s/it]
Training Epoch: 14/100, batch 175/438 completed (loss: 0.0279:  40%|████      | 88/219 [08:50<13:13,  6.05s/it]
Training Epoch: 14/100, batch 175/438 completed (loss: 0.0581:  40%|████      | 88/219 [08:50<13:12,  6.05s/it]
Training Epoch: 14/100, batch 176/438 completed (loss: 0.1109:  40%|████      | 88/219 [08:53<13:12,  6.05s/it]
Training Epoch: 14/100, batch 176/438 completed (loss: 0.0538:  40%|████      | 88/219 [08:53<13:13,  6.05s/it]
Training Epoch: 14/100, batch 176/438 completed (loss: 0.0822:  40%|████      | 88/219 [08:53<13:12,  6.05s/it]
Training Epoch: 14/100, batch 176/438 completed (loss: 0.0066:  40%|████      | 88/219 [08:53<13:12,  6.05s/it]
Training Epoch: 14/100, batch 176/438 completed (loss: 0.0066:  41%|████      | 89/219 [08:56<13:05,  6.04s/it]
Training Epoch: 14/100, batch 176/438 completed (loss: 0.0538:  41%|████      | 89/219 [08:56<13:05,  6.04s/it]
Training Epoch: 14/100, batch 176/438 completed (loss: 0.1109:  41%|████      | 89/219 [08:56<13:05,  6.04s/it]
Training Epoch: 14/100, batch 176/438 completed (loss: 0.0822:  41%|████      | 89/219 [08:56<13:05,  6.04s/it]
Training Epoch: 14/100, batch 177/438 completed (loss: 0.12:  41%|████      | 89/219 [08:56<13:05,  6.04s/it]  
Training Epoch: 14/100, batch 177/438 completed (loss: 0.0058:  41%|████      | 89/219 [08:56<13:05,  6.04s/it]
Training Epoch: 14/100, batch 177/438 completed (loss: 0.0056:  41%|████      | 89/219 [08:56<13:05,  6.04s/it]
Training Epoch: 14/100, batch 177/438 completed (loss: 0.0282:  41%|████      | 89/219 [08:56<13:05,  6.04s/it]
Training Epoch: 14/100, batch 178/438 completed (loss: 0.0091:  41%|████      | 89/219 [08:59<13:05,  6.04s/it]
Training Epoch: 14/100, batch 178/438 completed (loss: 0.0055:  41%|████      | 89/219 [08:59<13:05,  6.04s/it]
Training Epoch: 14/100, batch 178/438 completed (loss: 0.0076:  41%|████      | 89/219 [08:59<13:05,  6.04s/it]
Training Epoch: 14/100, batch 178/438 completed (loss: 0.3257:  41%|████      | 89/219 [08:59<13:05,  6.04s/it]
Training Epoch: 14/100, batch 178/438 completed (loss: 0.0055:  41%|████      | 90/219 [09:02<12:58,  6.03s/it]
Training Epoch: 14/100, batch 178/438 completed (loss: 0.3257:  41%|████      | 90/219 [09:02<12:58,  6.03s/it]
Training Epoch: 14/100, batch 178/438 completed (loss: 0.0091:  41%|████      | 90/219 [09:02<12:58,  6.03s/it]
Training Epoch: 14/100, batch 178/438 completed (loss: 0.0076:  41%|████      | 90/219 [09:02<12:58,  6.04s/it]
Training Epoch: 14/100, batch 179/438 completed (loss: 0.2769:  41%|████      | 90/219 [09:02<12:58,  6.03s/it]
Training Epoch: 14/100, batch 179/438 completed (loss: 0.0882:  41%|████      | 90/219 [09:02<12:58,  6.04s/it]
Training Epoch: 14/100, batch 179/438 completed (loss: 0.0506:  41%|████      | 90/219 [09:02<12:58,  6.03s/it]
Training Epoch: 14/100, batch 179/438 completed (loss: 0.2231:  41%|████      | 90/219 [09:02<12:58,  6.03s/it]
Training Epoch: 14/100, batch 180/438 completed (loss: 0.0185:  41%|████      | 90/219 [09:05<12:58,  6.04s/it]
Training Epoch: 14/100, batch 180/438 completed (loss: 0.1748:  41%|████      | 90/219 [09:05<12:58,  6.03s/it]
Training Epoch: 14/100, batch 180/438 completed (loss: 0.111:  41%|████      | 90/219 [09:05<12:58,  6.03s/it] 
Training Epoch: 14/100, batch 180/438 completed (loss: 0.0762:  41%|████      | 90/219 [09:05<12:58,  6.03s/it]
Training Epoch: 14/100, batch 180/438 completed (loss: 0.111:  42%|████▏     | 91/219 [09:08<12:58,  6.08s/it]
Training Epoch: 14/100, batch 180/438 completed (loss: 0.1748:  42%|████▏     | 91/219 [09:08<12:58,  6.08s/it]
Training Epoch: 14/100, batch 180/438 completed (loss: 0.0762:  42%|████▏     | 91/219 [09:09<12:58,  6.08s/it]
Training Epoch: 14/100, batch 180/438 completed (loss: 0.0185:  42%|████▏     | 91/219 [09:08<12:58,  6.08s/it]
Training Epoch: 14/100, batch 181/438 completed (loss: 0.3327:  42%|████▏     | 91/219 [09:09<12:58,  6.08s/it]
Training Epoch: 14/100, batch 181/438 completed (loss: 0.1078:  42%|████▏     | 91/219 [09:09<12:58,  6.08s/it]
Training Epoch: 14/100, batch 181/438 completed (loss: 0.2668:  42%|████▏     | 91/219 [09:09<12:58,  6.08s/it]
Training Epoch: 14/100, batch 181/438 completed (loss: 0.1287:  42%|████▏     | 91/219 [09:09<12:58,  6.08s/it]
Training Epoch: 14/100, batch 182/438 completed (loss: 0.048:  42%|████▏     | 91/219 [09:11<12:58,  6.08s/it] 
Training Epoch: 14/100, batch 182/438 completed (loss: 0.1605:  42%|████▏     | 91/219 [09:11<12:58,  6.08s/it]
Training Epoch: 14/100, batch 182/438 completed (loss: 0.0912:  42%|████▏     | 91/219 [09:11<12:58,  6.08s/it]
Training Epoch: 14/100, batch 182/438 completed (loss: 0.0283:  42%|████▏     | 91/219 [09:12<12:58,  6.08s/it]
Training Epoch: 14/100, batch 182/438 completed (loss: 0.048:  42%|████▏     | 92/219 [09:14<12:43,  6.01s/it]
Training Epoch: 14/100, batch 182/438 completed (loss: 0.0283:  42%|████▏     | 92/219 [09:14<12:43,  6.01s/it]
Training Epoch: 14/100, batch 182/438 completed (loss: 0.0912:  42%|████▏     | 92/219 [09:14<12:43,  6.01s/it]
Training Epoch: 14/100, batch 182/438 completed (loss: 0.1605:  42%|████▏     | 92/219 [09:14<12:43,  6.01s/it]
Training Epoch: 14/100, batch 183/438 completed (loss: 0.2262:  42%|████▏     | 92/219 [09:14<12:43,  6.01s/it]
Training Epoch: 14/100, batch 183/438 completed (loss: 0.2533:  42%|████▏     | 92/219 [09:14<12:43,  6.01s/it]
Training Epoch: 14/100, batch 183/438 completed (loss: 0.1511:  42%|████▏     | 92/219 [09:14<12:43,  6.01s/it]
Training Epoch: 14/100, batch 183/438 completed (loss: 0.0593:  42%|████▏     | 92/219 [09:14<12:43,  6.01s/it]
Training Epoch: 14/100, batch 184/438 completed (loss: 0.857:  42%|████▏     | 92/219 [09:17<12:43,  6.01s/it] 
Training Epoch: 14/100, batch 184/438 completed (loss: 0.0402:  42%|████▏     | 92/219 [09:17<12:43,  6.01s/it]
Training Epoch: 14/100, batch 184/438 completed (loss: 0.0175:  42%|████▏     | 92/219 [09:17<12:43,  6.01s/it]
Training Epoch: 14/100, batch 184/438 completed (loss: 0.0497:  42%|████▏     | 92/219 [09:18<12:43,  6.01s/it]
Training Epoch: 14/100, batch 184/438 completed (loss: 0.0175:  42%|████▏     | 93/219 [09:21<12:48,  6.10s/it]
Training Epoch: 14/100, batch 184/438 completed (loss: 0.857:  42%|████▏     | 93/219 [09:21<12:48,  6.10s/it]
Training Epoch: 14/100, batch 184/438 completed (loss: 0.0402:  42%|████▏     | 93/219 [09:21<12:48,  6.10s/it]
Training Epoch: 14/100, batch 184/438 completed (loss: 0.0497:  42%|████▏     | 93/219 [09:21<12:48,  6.10s/it]
Training Epoch: 14/100, batch 185/438 completed (loss: 0.0795:  42%|████▏     | 93/219 [09:21<12:48,  6.10s/it]
Training Epoch: 14/100, batch 185/438 completed (loss: 0.0469:  42%|████▏     | 93/219 [09:21<12:48,  6.10s/it]
Training Epoch: 14/100, batch 185/438 completed (loss: 0.0408:  42%|████▏     | 93/219 [09:21<12:48,  6.10s/it]
Training Epoch: 14/100, batch 185/438 completed (loss: 0.0028:  42%|████▏     | 93/219 [09:21<12:48,  6.10s/it]
Training Epoch: 14/100, batch 186/438 completed (loss: 0.0722:  42%|████▏     | 93/219 [09:24<12:48,  6.10s/it]
Training Epoch: 14/100, batch 186/438 completed (loss: 0.0359:  42%|████▏     | 93/219 [09:24<12:48,  6.10s/it]
Training Epoch: 14/100, batch 186/438 completed (loss: 0.1076:  42%|████▏     | 93/219 [09:24<12:48,  6.10s/it]
Training Epoch: 14/100, batch 186/438 completed (loss: 0.2677:  42%|████▏     | 93/219 [09:24<12:48,  6.10s/it]
Training Epoch: 14/100, batch 186/438 completed (loss: 0.2677:  43%|████▎     | 94/219 [09:27<12:39,  6.07s/it]
Training Epoch: 14/100, batch 186/438 completed (loss: 0.1076:  43%|████▎     | 94/219 [09:27<12:39,  6.07s/it]
Training Epoch: 14/100, batch 186/438 completed (loss: 0.0359:  43%|████▎     | 94/219 [09:27<12:39,  6.08s/it]
Training Epoch: 14/100, batch 186/438 completed (loss: 0.0722:  43%|████▎     | 94/219 [09:27<12:39,  6.08s/it]
Training Epoch: 14/100, batch 187/438 completed (loss: 0.046:  43%|████▎     | 94/219 [09:27<12:39,  6.07s/it] 
Training Epoch: 14/100, batch 187/438 completed (loss: 0.0116:  43%|████▎     | 94/219 [09:27<12:39,  6.08s/it]
Training Epoch: 14/100, batch 187/438 completed (loss: 0.0304:  43%|████▎     | 94/219 [09:27<12:39,  6.08s/it]
Training Epoch: 14/100, batch 187/438 completed (loss: 0.0607:  43%|████▎     | 94/219 [09:27<12:39,  6.07s/it]
Training Epoch: 14/100, batch 188/438 completed (loss: 0.0484:  43%|████▎     | 94/219 [09:30<12:39,  6.08s/it]
Training Epoch: 14/100, batch 188/438 completed (loss: 0.3854:  43%|████▎     | 94/219 [09:30<12:39,  6.07s/it]
Training Epoch: 14/100, batch 188/438 completed (loss: 0.1094:  43%|████▎     | 94/219 [09:30<12:39,  6.08s/it]
Training Epoch: 14/100, batch 188/438 completed (loss: 0.0974:  43%|████▎     | 94/219 [09:30<12:39,  6.07s/it]
Training Epoch: 14/100, batch 188/438 completed (loss: 0.0974:  43%|████▎     | 95/219 [09:33<12:24,  6.00s/it]
Training Epoch: 14/100, batch 188/438 completed (loss: 0.1094:  43%|████▎     | 95/219 [09:33<12:25,  6.01s/it]
Training Epoch: 14/100, batch 188/438 completed (loss: 0.3854:  43%|████▎     | 95/219 [09:33<12:25,  6.01s/it]
Training Epoch: 14/100, batch 188/438 completed (loss: 0.0484:  43%|████▎     | 95/219 [09:33<12:25,  6.01s/it]
Training Epoch: 14/100, batch 189/438 completed (loss: 0.0972:  43%|████▎     | 95/219 [09:33<12:25,  6.01s/it]
Training Epoch: 14/100, batch 189/438 completed (loss: 0.0117:  43%|████▎     | 95/219 [09:33<12:25,  6.01s/it]
Training Epoch: 14/100, batch 189/438 completed (loss: 0.0147:  43%|████▎     | 95/219 [09:33<12:25,  6.01s/it]
Training Epoch: 14/100, batch 189/438 completed (loss: 0.017:  43%|████▎     | 95/219 [09:33<12:24,  6.00s/it] 
Training Epoch: 14/100, batch 190/438 completed (loss: 0.3209:  43%|████▎     | 95/219 [09:36<12:25,  6.01s/it]
Training Epoch: 14/100, batch 190/438 completed (loss: 0.0997:  43%|████▎     | 95/219 [09:36<12:25,  6.01s/it]
Training Epoch: 14/100, batch 190/438 completed (loss: 0.235:  43%|████▎     | 95/219 [09:36<12:25,  6.01s/it] 
Training Epoch: 14/100, batch 190/438 completed (loss: 0.0924:  43%|████▎     | 95/219 [09:36<12:24,  6.00s/it]
Training Epoch: 14/100, batch 190/438 completed (loss: 0.235:  44%|████▍     | 96/219 [09:39<12:32,  6.12s/it]
Training Epoch: 14/100, batch 190/438 completed (loss: 0.0997:  44%|████▍     | 96/219 [09:39<12:32,  6.12s/it]
Training Epoch: 14/100, batch 190/438 completed (loss: 0.0924:  44%|████▍     | 96/219 [09:39<12:33,  6.12s/it]
Training Epoch: 14/100, batch 190/438 completed (loss: 0.3209:  44%|████▍     | 96/219 [09:39<12:32,  6.12s/it]
Training Epoch: 14/100, batch 191/438 completed (loss: 0.0233:  44%|████▍     | 96/219 [09:39<12:32,  6.12s/it]
Training Epoch: 14/100, batch 191/438 completed (loss: 0.0185:  44%|████▍     | 96/219 [09:39<12:32,  6.12s/it]
Training Epoch: 14/100, batch 191/438 completed (loss: 0.0399:  44%|████▍     | 96/219 [09:39<12:32,  6.12s/it]
Training Epoch: 14/100, batch 191/438 completed (loss: 0.0685:  44%|████▍     | 96/219 [09:39<12:33,  6.12s/it]
Training Epoch: 14/100, batch 192/438 completed (loss: 0.0253:  44%|████▍     | 96/219 [09:42<12:32,  6.12s/it]
Training Epoch: 14/100, batch 192/438 completed (loss: 0.019:  44%|████▍     | 96/219 [09:42<12:32,  6.12s/it] 
Training Epoch: 14/100, batch 192/438 completed (loss: 0.0275:  44%|████▍     | 96/219 [09:42<12:32,  6.12s/it]
Training Epoch: 14/100, batch 192/438 completed (loss: 0.1091:  44%|████▍     | 96/219 [09:42<12:33,  6.12s/it]
Training Epoch: 14/100, batch 192/438 completed (loss: 0.1091:  44%|████▍     | 97/219 [09:45<12:17,  6.04s/it]
Training Epoch: 14/100, batch 192/438 completed (loss: 0.0275:  44%|████▍     | 97/219 [09:45<12:16,  6.04s/it]
Training Epoch: 14/100, batch 192/438 completed (loss: 0.0253:  44%|████▍     | 97/219 [09:45<12:17,  6.04s/it]
Training Epoch: 14/100, batch 192/438 completed (loss: 0.019:  44%|████▍     | 97/219 [09:45<12:17,  6.04s/it]
Training Epoch: 14/100, batch 193/438 completed (loss: 0.2345:  44%|████▍     | 97/219 [09:45<12:17,  6.04s/it]
Training Epoch: 14/100, batch 193/438 completed (loss: 0.0289:  44%|████▍     | 97/219 [09:45<12:17,  6.04s/it]
Training Epoch: 14/100, batch 193/438 completed (loss: 0.0278:  44%|████▍     | 97/219 [09:45<12:16,  6.04s/it]
Training Epoch: 14/100, batch 193/438 completed (loss: 0.0274:  44%|████▍     | 97/219 [09:45<12:17,  6.04s/it]
Training Epoch: 14/100, batch 194/438 completed (loss: 0.319:  44%|████▍     | 97/219 [09:48<12:16,  6.04s/it] 
Training Epoch: 14/100, batch 194/438 completed (loss: 0.1415:  44%|████▍     | 97/219 [09:48<12:17,  6.04s/it]
Training Epoch: 14/100, batch 194/438 completed (loss: 0.0732:  44%|████▍     | 97/219 [09:48<12:17,  6.04s/it]
Training Epoch: 14/100, batch 194/438 completed (loss: 0.1563:  44%|████▍     | 97/219 [09:48<12:17,  6.04s/it]
Training Epoch: 14/100, batch 194/438 completed (loss: 0.1563:  45%|████▍     | 98/219 [09:51<12:21,  6.12s/it]
Training Epoch: 14/100, batch 194/438 completed (loss: 0.319:  45%|████▍     | 98/219 [09:51<12:21,  6.13s/it]
Training Epoch: 14/100, batch 194/438 completed (loss: 0.1415:  45%|████▍     | 98/219 [09:51<12:21,  6.13s/it]
Training Epoch: 14/100, batch 194/438 completed (loss: 0.0732:  45%|████▍     | 98/219 [09:51<12:21,  6.13s/it]
Training Epoch: 14/100, batch 195/438 completed (loss: 0.3731:  45%|████▍     | 98/219 [09:51<12:21,  6.13s/it]
Training Epoch: 14/100, batch 195/438 completed (loss: 0.0325:  45%|████▍     | 98/219 [09:51<12:21,  6.13s/it]
Training Epoch: 14/100, batch 195/438 completed (loss: 0.0166:  45%|████▍     | 98/219 [09:51<12:21,  6.13s/it]
Training Epoch: 14/100, batch 195/438 completed (loss: 0.0078:  45%|████▍     | 98/219 [09:51<12:21,  6.12s/it]
Training Epoch: 14/100, batch 196/438 completed (loss: 0.0254:  45%|████▍     | 98/219 [09:54<12:21,  6.13s/it]
Training Epoch: 14/100, batch 196/438 completed (loss: 0.0116:  45%|████▍     | 98/219 [09:54<12:21,  6.13s/it]
Training Epoch: 14/100, batch 196/438 completed (loss: 0.0134:  45%|████▍     | 98/219 [09:54<12:21,  6.13s/it]
Training Epoch: 14/100, batch 196/438 completed (loss: 0.5703:  45%|████▍     | 98/219 [09:54<12:21,  6.12s/it]
Training Epoch: 14/100, batch 196/438 completed (loss: 0.0116:  45%|████▌     | 99/219 [09:57<12:11,  6.09s/it]
Training Epoch: 14/100, batch 196/438 completed (loss: 0.5703:  45%|████▌     | 99/219 [09:57<12:11,  6.09s/it]
Training Epoch: 14/100, batch 196/438 completed (loss: 0.0134:  45%|████▌     | 99/219 [09:57<12:11,  6.09s/it]
Training Epoch: 14/100, batch 196/438 completed (loss: 0.0254:  45%|████▌     | 99/219 [09:57<12:11,  6.10s/it]
Training Epoch: 14/100, batch 197/438 completed (loss: 0.0057:  45%|████▌     | 99/219 [09:57<12:11,  6.10s/it]
Training Epoch: 14/100, batch 197/438 completed (loss: 0.3765:  45%|████▌     | 99/219 [09:57<12:11,  6.09s/it]
Training Epoch: 14/100, batch 197/438 completed (loss: 0.1251:  45%|████▌     | 99/219 [09:57<12:11,  6.09s/it]
Training Epoch: 14/100, batch 197/438 completed (loss: 0.0435:  45%|████▌     | 99/219 [09:57<12:11,  6.09s/it]
Training Epoch: 14/100, batch 198/438 completed (loss: 0.0975:  45%|████▌     | 99/219 [10:00<12:11,  6.09s/it]
Training Epoch: 14/100, batch 198/438 completed (loss: 0.011:  45%|████▌     | 99/219 [10:00<12:11,  6.09s/it] 
Training Epoch: 14/100, batch 198/438 completed (loss: 0.0126:  45%|████▌     | 99/219 [10:00<12:11,  6.10s/it]
Training Epoch: 14/100, batch 198/438 completed (loss: 0.3819:  45%|████▌     | 99/219 [10:00<12:11,  6.09s/it]
Training Epoch: 14/100, batch 198/438 completed (loss: 0.0126:  46%|████▌     | 100/219 [10:03<11:55,  6.02s/it]
Training Epoch: 14/100, batch 198/438 completed (loss: 0.3819:  46%|████▌     | 100/219 [10:03<11:56,  6.02s/it]
Training Epoch: 14/100, batch 198/438 completed (loss: 0.0975:  46%|████▌     | 100/219 [10:03<11:55,  6.02s/it]
Training Epoch: 14/100, batch 198/438 completed (loss: 0.011:  46%|████▌     | 100/219 [10:03<11:56,  6.02s/it]
Training Epoch: 14/100, batch 199/438 completed (loss: 0.1018:  46%|████▌     | 100/219 [10:03<11:55,  6.02s/it]
Training Epoch: 14/100, batch 199/438 completed (loss: 0.1795:  46%|████▌     | 100/219 [10:03<11:55,  6.02s/it]
Training Epoch: 14/100, batch 199/438 completed (loss: 0.0474:  46%|████▌     | 100/219 [10:03<11:56,  6.02s/it]
Training Epoch: 14/100, batch 199/438 completed (loss: 0.1348:  46%|████▌     | 100/219 [10:03<11:56,  6.02s/it]
Training Epoch: 14/100, batch 200/438 completed (loss: 0.0551:  46%|████▌     | 100/219 [10:06<11:56,  6.02s/it]
Training Epoch: 14/100, batch 200/438 completed (loss: 0.0621:  46%|████▌     | 100/219 [10:06<11:55,  6.02s/it]
Training Epoch: 14/100, batch 200/438 completed (loss: 0.0596:  46%|████▌     | 100/219 [10:06<11:55,  6.02s/it]
Training Epoch: 14/100, batch 200/438 completed (loss: 0.2817:  46%|████▌     | 100/219 [10:06<11:56,  6.02s/it]
Training Epoch: 14/100, batch 200/438 completed (loss: 0.0596:  46%|████▌     | 101/219 [10:09<11:54,  6.06s/it]
Training Epoch: 14/100, batch 200/438 completed (loss: 0.0621:  46%|████▌     | 101/219 [10:09<11:54,  6.06s/it]
Training Epoch: 14/100, batch 200/438 completed (loss: 0.2817:  46%|████▌     | 101/219 [10:09<11:55,  6.06s/it]
Training Epoch: 14/100, batch 200/438 completed (loss: 0.0551:  46%|████▌     | 101/219 [10:09<11:55,  6.06s/it]
Training Epoch: 14/100, batch 201/438 completed (loss: 0.385:  46%|████▌     | 101/219 [10:09<11:54,  6.06s/it] 
Training Epoch: 14/100, batch 201/438 completed (loss: 0.0255:  46%|████▌     | 101/219 [10:09<11:55,  6.06s/it]
Training Epoch: 14/100, batch 201/438 completed (loss: 0.0375:  46%|████▌     | 101/219 [10:09<11:54,  6.06s/it]
Training Epoch: 14/100, batch 201/438 completed (loss: 0.0189:  46%|████▌     | 101/219 [10:09<11:55,  6.06s/it]
Training Epoch: 14/100, batch 202/438 completed (loss: 0.6099:  46%|████▌     | 101/219 [10:12<11:54,  6.06s/it]
Training Epoch: 14/100, batch 202/438 completed (loss: 0.1911:  46%|████▌     | 101/219 [10:12<11:54,  6.06s/it]
Training Epoch: 14/100, batch 202/438 completed (loss: 0.0484:  46%|████▌     | 101/219 [10:12<11:55,  6.06s/it]
Training Epoch: 14/100, batch 202/438 completed (loss: 0.5669:  46%|████▌     | 101/219 [10:12<11:55,  6.06s/it]
Training Epoch: 14/100, batch 202/438 completed (loss: 0.0484:  47%|████▋     | 102/219 [10:15<11:46,  6.04s/it]
Training Epoch: 14/100, batch 202/438 completed (loss: 0.5669:  47%|████▋     | 102/219 [10:15<11:47,  6.04s/it]
Training Epoch: 14/100, batch 202/438 completed (loss: 0.6099:  47%|████▋     | 102/219 [10:15<11:46,  6.04s/it]
Training Epoch: 14/100, batch 202/438 completed (loss: 0.1911:  47%|████▋     | 102/219 [10:15<11:46,  6.04s/it]
Training Epoch: 14/100, batch 203/438 completed (loss: 0.1327:  47%|████▋     | 102/219 [10:15<11:46,  6.04s/it]
Training Epoch: 14/100, batch 203/438 completed (loss: 0.019:  47%|████▋     | 102/219 [10:15<11:46,  6.04s/it] 
Training Epoch: 14/100, batch 203/438 completed (loss: 0.022:  47%|████▋     | 102/219 [10:15<11:46,  6.04s/it] 
Training Epoch: 14/100, batch 203/438 completed (loss: 0.2232:  47%|████▋     | 102/219 [10:15<11:47,  6.04s/it]
Training Epoch: 14/100, batch 204/438 completed (loss: 0.0028:  47%|████▋     | 102/219 [10:18<11:46,  6.04s/it]
Training Epoch: 14/100, batch 204/438 completed (loss: 0.0134:  47%|████▋     | 102/219 [10:18<11:46,  6.04s/it]
Training Epoch: 14/100, batch 204/438 completed (loss: 0.018:  47%|████▋     | 102/219 [10:18<11:46,  6.04s/it] 
Training Epoch: 14/100, batch 204/438 completed (loss: 0.0284:  47%|████▋     | 102/219 [10:18<11:47,  6.04s/it]
Training Epoch: 14/100, batch 204/438 completed (loss: 0.018:  47%|████▋     | 103/219 [10:21<11:34,  5.99s/it]
Training Epoch: 14/100, batch 204/438 completed (loss: 0.0134:  47%|████▋     | 103/219 [10:21<11:34,  5.99s/it]
Training Epoch: 14/100, batch 204/438 completed (loss: 0.0028:  47%|████▋     | 103/219 [10:21<11:34,  5.99s/it]
Training Epoch: 14/100, batch 204/438 completed (loss: 0.0284:  47%|████▋     | 103/219 [10:21<11:35,  5.99s/it]
Training Epoch: 14/100, batch 205/438 completed (loss: 0.498:  47%|████▋     | 103/219 [10:21<11:34,  5.99s/it] 
Training Epoch: 14/100, batch 205/438 completed (loss: 0.1431:  47%|████▋     | 103/219 [10:21<11:34,  5.99s/it]
Training Epoch: 14/100, batch 205/438 completed (loss: 0.5902:  47%|████▋     | 103/219 [10:21<11:34,  5.99s/it]
Training Epoch: 14/100, batch 205/438 completed (loss: 0.014:  47%|████▋     | 103/219 [10:21<11:35,  5.99s/it] 
Training Epoch: 14/100, batch 206/438 completed (loss: 0.1098:  47%|████▋     | 103/219 [10:24<11:34,  5.99s/it]
Training Epoch: 14/100, batch 206/438 completed (loss: 0.0176:  47%|████▋     | 103/219 [10:24<11:34,  5.99s/it]
Training Epoch: 14/100, batch 206/438 completed (loss: 0.0176:  47%|████▋     | 103/219 [10:24<11:34,  5.99s/it]
Training Epoch: 14/100, batch 206/438 completed (loss: 0.2597:  47%|████▋     | 103/219 [10:24<11:35,  5.99s/it]
Training Epoch: 14/100, batch 206/438 completed (loss: 0.2597:  47%|████▋     | 104/219 [10:27<11:38,  6.07s/it]
Training Epoch: 14/100, batch 206/438 completed (loss: 0.1098:  47%|████▋     | 104/219 [10:27<11:38,  6.07s/it]
Training Epoch: 14/100, batch 206/438 completed (loss: 0.0176:  47%|████▋     | 104/219 [10:27<11:38,  6.07s/it]
Training Epoch: 14/100, batch 206/438 completed (loss: 0.0176:  47%|████▋     | 104/219 [10:27<11:38,  6.07s/it]
Training Epoch: 14/100, batch 207/438 completed (loss: 0.0381:  47%|████▋     | 104/219 [10:27<11:38,  6.07s/it]
Training Epoch: 14/100, batch 207/438 completed (loss: 0.1007:  47%|████▋     | 104/219 [10:27<11:38,  6.07s/it]
Training Epoch: 14/100, batch 207/438 completed (loss: 0.0222:  47%|████▋     | 104/219 [10:27<11:38,  6.07s/it]
Training Epoch: 14/100, batch 207/438 completed (loss: 0.0472:  47%|████▋     | 104/219 [10:27<11:38,  6.07s/it]
Training Epoch: 14/100, batch 208/438 completed (loss: 0.0663:  47%|████▋     | 104/219 [10:30<11:38,  6.07s/it]
Training Epoch: 14/100, batch 208/438 completed (loss: 0.0303:  47%|████▋     | 104/219 [10:30<11:38,  6.07s/it]
Training Epoch: 14/100, batch 208/438 completed (loss: 0.1199:  47%|████▋     | 104/219 [10:30<11:38,  6.07s/it]
Training Epoch: 14/100, batch 208/438 completed (loss: 0.012:  47%|████▋     | 104/219 [10:30<11:38,  6.07s/it] 
Training Epoch: 14/100, batch 208/438 completed (loss: 0.0303:  48%|████▊     | 105/219 [10:33<11:29,  6.05s/it]
Training Epoch: 14/100, batch 208/438 completed (loss: 0.1199:  48%|████▊     | 105/219 [10:33<11:29,  6.05s/it]
Training Epoch: 14/100, batch 208/438 completed (loss: 0.0663:  48%|████▊     | 105/219 [10:33<11:29,  6.05s/it]
Training Epoch: 14/100, batch 208/438 completed (loss: 0.012:  48%|████▊     | 105/219 [10:33<11:30,  6.05s/it]
Training Epoch: 14/100, batch 209/438 completed (loss: 0.0841:  48%|████▊     | 105/219 [10:33<11:29,  6.05s/it]
Training Epoch: 14/100, batch 209/438 completed (loss: 0.0055:  48%|████▊     | 105/219 [10:33<11:29,  6.05s/it]
Training Epoch: 14/100, batch 209/438 completed (loss: 0.0222:  48%|████▊     | 105/219 [10:33<11:29,  6.05s/it]
Training Epoch: 14/100, batch 209/438 completed (loss: 0.0485:  48%|████▊     | 105/219 [10:33<11:30,  6.05s/it]
Training Epoch: 14/100, batch 210/438 completed (loss: 0.0556:  48%|████▊     | 105/219 [10:36<11:29,  6.05s/it]
Training Epoch: 14/100, batch 210/438 completed (loss: 0.064:  48%|████▊     | 105/219 [10:36<11:29,  6.05s/it] 
Training Epoch: 14/100, batch 210/438 completed (loss: 0.0231:  48%|████▊     | 105/219 [10:36<11:29,  6.05s/it]
Training Epoch: 14/100, batch 210/438 completed (loss: 0.1057:  48%|████▊     | 105/219 [10:36<11:30,  6.05s/it]
Training Epoch: 14/100, batch 210/438 completed (loss: 0.064:  48%|████▊     | 106/219 [10:39<11:27,  6.08s/it]
Training Epoch: 14/100, batch 210/438 completed (loss: 0.0556:  48%|████▊     | 106/219 [10:39<11:27,  6.08s/it]
Training Epoch: 14/100, batch 210/438 completed (loss: 0.0231:  48%|████▊     | 106/219 [10:39<11:27,  6.09s/it]
Training Epoch: 14/100, batch 210/438 completed (loss: 0.1057:  48%|████▊     | 106/219 [10:39<11:28,  6.09s/it]
Training Epoch: 14/100, batch 211/438 completed (loss: 0.0306:  48%|████▊     | 106/219 [10:39<11:27,  6.08s/it]
Training Epoch: 14/100, batch 211/438 completed (loss: 0.029:  48%|████▊     | 106/219 [10:39<11:27,  6.09s/it] 
Training Epoch: 14/100, batch 211/438 completed (loss: 0.3038:  48%|████▊     | 106/219 [10:39<11:27,  6.08s/it]
Training Epoch: 14/100, batch 211/438 completed (loss: 0.0652:  48%|████▊     | 106/219 [10:39<11:28,  6.09s/it]
Training Epoch: 14/100, batch 212/438 completed (loss: 0.0274:  48%|████▊     | 106/219 [10:42<11:27,  6.08s/it]
Training Epoch: 14/100, batch 212/438 completed (loss: 0.0211:  48%|████▊     | 106/219 [10:42<11:27,  6.09s/it]
Training Epoch: 14/100, batch 212/438 completed (loss: 0.0735:  48%|████▊     | 106/219 [10:42<11:27,  6.08s/it]
Training Epoch: 14/100, batch 212/438 completed (loss: 0.2545:  48%|████▊     | 106/219 [10:42<11:28,  6.09s/it]
Training Epoch: 14/100, batch 212/438 completed (loss: 0.0211:  49%|████▉     | 107/219 [10:45<11:18,  6.05s/it]
Training Epoch: 14/100, batch 212/438 completed (loss: 0.0735:  49%|████▉     | 107/219 [10:45<11:18,  6.05s/it]
Training Epoch: 14/100, batch 212/438 completed (loss: 0.0274:  49%|████▉     | 107/219 [10:45<11:18,  6.06s/it]
Training Epoch: 14/100, batch 212/438 completed (loss: 0.2545:  49%|████▉     | 107/219 [10:45<11:17,  6.05s/it]
Training Epoch: 14/100, batch 213/438 completed (loss: 0.0873:  49%|████▉     | 107/219 [10:45<11:18,  6.05s/it]
Training Epoch: 14/100, batch 213/438 completed (loss: 0.0345:  49%|████▉     | 107/219 [10:45<11:18,  6.06s/it]
Training Epoch: 14/100, batch 213/438 completed (loss: 0.0235:  49%|████▉     | 107/219 [10:45<11:18,  6.05s/it]
Training Epoch: 14/100, batch 213/438 completed (loss: 0.0144:  49%|████▉     | 107/219 [10:45<11:17,  6.05s/it]
Training Epoch: 14/100, batch 214/438 completed (loss: 0.0068:  49%|████▉     | 107/219 [10:48<11:18,  6.05s/it]
Training Epoch: 14/100, batch 214/438 completed (loss: 0.1845:  49%|████▉     | 107/219 [10:48<11:18,  6.06s/it]
Training Epoch: 14/100, batch 214/438 completed (loss: 0.0863:  49%|████▉     | 107/219 [10:48<11:18,  6.05s/it]
Training Epoch: 14/100, batch 214/438 completed (loss: 0.3168:  49%|████▉     | 107/219 [10:48<11:17,  6.05s/it]
Training Epoch: 14/100, batch 214/438 completed (loss: 0.1845:  49%|████▉     | 108/219 [10:51<11:05,  5.99s/it]
Training Epoch: 14/100, batch 214/438 completed (loss: 0.3168:  49%|████▉     | 108/219 [10:51<11:05,  5.99s/it]
Training Epoch: 14/100, batch 214/438 completed (loss: 0.0863:  49%|████▉     | 108/219 [10:51<11:05,  5.99s/it]
Training Epoch: 14/100, batch 214/438 completed (loss: 0.0068:  49%|████▉     | 108/219 [10:51<11:05,  6.00s/it]
Training Epoch: 14/100, batch 215/438 completed (loss: 0.5307:  49%|████▉     | 108/219 [10:51<11:05,  5.99s/it]
Training Epoch: 14/100, batch 215/438 completed (loss: 0.0416:  49%|████▉     | 108/219 [10:51<11:05,  5.99s/it]
Training Epoch: 14/100, batch 215/438 completed (loss: 0.2035:  49%|████▉     | 108/219 [10:51<11:05,  6.00s/it]
Training Epoch: 14/100, batch 215/438 completed (loss: 0.0252:  49%|████▉     | 108/219 [10:51<11:05,  5.99s/it]
Training Epoch: 14/100, batch 216/438 completed (loss: 0.0497:  49%|████▉     | 108/219 [10:55<11:05,  5.99s/it]
Training Epoch: 14/100, batch 216/438 completed (loss: 0.0176:  49%|████▉     | 108/219 [10:55<11:05,  6.00s/it]
Training Epoch: 14/100, batch 216/438 completed (loss: 0.0582:  49%|████▉     | 108/219 [10:55<11:05,  5.99s/it]
Training Epoch: 14/100, batch 216/438 completed (loss: 0.0088:  49%|████▉     | 108/219 [10:55<11:05,  5.99s/it]
Training Epoch: 14/100, batch 216/438 completed (loss: 0.0497:  50%|████▉     | 109/219 [10:58<11:09,  6.08s/it]
Training Epoch: 14/100, batch 216/438 completed (loss: 0.0176:  50%|████▉     | 109/219 [10:58<11:09,  6.08s/it]
Training Epoch: 14/100, batch 216/438 completed (loss: 0.0582:  50%|████▉     | 109/219 [10:58<11:09,  6.08s/it]
Training Epoch: 14/100, batch 216/438 completed (loss: 0.0088:  50%|████▉     | 109/219 [10:58<11:09,  6.08s/it]
Training Epoch: 14/100, batch 217/438 completed (loss: 0.0121:  50%|████▉     | 109/219 [10:58<11:09,  6.08s/it]
Training Epoch: 14/100, batch 217/438 completed (loss: 0.2815:  50%|████▉     | 109/219 [10:58<11:09,  6.08s/it]
Training Epoch: 14/100, batch 217/438 completed (loss: 0.025:  50%|████▉     | 109/219 [10:58<11:09,  6.08s/it] 
Training Epoch: 14/100, batch 217/438 completed (loss: 0.0756:  50%|████▉     | 109/219 [10:58<11:09,  6.08s/it]
Training Epoch: 14/100, batch 218/438 completed (loss: 0.0275:  50%|████▉     | 109/219 [11:01<11:09,  6.08s/it]
Training Epoch: 14/100, batch 218/438 completed (loss: 0.3678:  50%|████▉     | 109/219 [11:01<11:09,  6.08s/it]
Training Epoch: 14/100, batch 218/438 completed (loss: 0.3398:  50%|████▉     | 109/219 [11:01<11:09,  6.08s/it]
Training Epoch: 14/100, batch 218/438 completed (loss: 0.0214:  50%|████▉     | 109/219 [11:01<11:09,  6.08s/it]
Training Epoch: 14/100, batch 218/438 completed (loss: 0.3398:  50%|█████     | 110/219 [11:04<11:00,  6.06s/it]
Training Epoch: 14/100, batch 218/438 completed (loss: 0.0214:  50%|█████     | 110/219 [11:04<11:00,  6.06s/it]
Training Epoch: 14/100, batch 218/438 completed (loss: 0.3678:  50%|█████     | 110/219 [11:04<11:00,  6.06s/it]
Training Epoch: 14/100, batch 218/438 completed (loss: 0.0275:  50%|█████     | 110/219 [11:04<11:00,  6.06s/it]
Training Epoch: 14/100, batch 219/438 completed (loss: 0.019:  50%|█████     | 110/219 [11:04<11:00,  6.06s/it] 
Training Epoch: 14/100, batch 219/438 completed (loss: 0.0046:  50%|█████     | 110/219 [11:04<11:00,  6.06s/it]
Training Epoch: 14/100, batch 219/438 completed (loss: 0.032:  50%|█████     | 110/219 [11:04<11:00,  6.06s/it] 
Training Epoch: 14/100, batch 219/438 completed (loss: 0.1168:  50%|█████     | 110/219 [11:04<11:00,  6.06s/it]
Training Epoch: 14/100, batch 220/438 completed (loss: 0.1011:  50%|█████     | 110/219 [11:07<11:00,  6.06s/it]
Training Epoch: 14/100, batch 220/438 completed (loss: 0.0733:  50%|█████     | 110/219 [11:07<11:00,  6.06s/it]
Training Epoch: 14/100, batch 220/438 completed (loss: 0.0147:  50%|█████     | 110/219 [11:07<11:00,  6.06s/it]
Training Epoch: 14/100, batch 220/438 completed (loss: 0.0193:  50%|█████     | 110/219 [11:07<11:00,  6.06s/it]
Training Epoch: 14/100, batch 220/438 completed (loss: 0.0193:  51%|█████     | 111/219 [11:10<10:58,  6.10s/it]
Training Epoch: 14/100, batch 220/438 completed (loss: 0.1011:  51%|█████     | 111/219 [11:10<10:58,  6.10s/it]
Training Epoch: 14/100, batch 220/438 completed (loss: 0.0147:  51%|█████     | 111/219 [11:10<10:58,  6.10s/it]
Training Epoch: 14/100, batch 220/438 completed (loss: 0.0733:  51%|█████     | 111/219 [11:10<10:58,  6.10s/it]
Training Epoch: 14/100, batch 221/438 completed (loss: 0.0338:  51%|█████     | 111/219 [11:10<10:58,  6.10s/it]
Training Epoch: 14/100, batch 221/438 completed (loss: 0.0376:  51%|█████     | 111/219 [11:10<10:58,  6.10s/it]
Training Epoch: 14/100, batch 221/438 completed (loss: 0.0281:  51%|█████     | 111/219 [11:10<10:58,  6.10s/it]
Training Epoch: 14/100, batch 221/438 completed (loss: 0.0755:  51%|█████     | 111/219 [11:10<10:58,  6.10s/it]
Training Epoch: 14/100, batch 222/438 completed (loss: 0.0716:  51%|█████     | 111/219 [11:13<10:58,  6.10s/it]
Training Epoch: 14/100, batch 222/438 completed (loss: 0.0193:  51%|█████     | 111/219 [11:13<10:58,  6.10s/it]
Training Epoch: 14/100, batch 222/438 completed (loss: 0.0466:  51%|█████     | 111/219 [11:13<10:58,  6.10s/it]
Training Epoch: 14/100, batch 222/438 completed (loss: 0.2215:  51%|█████     | 111/219 [11:13<10:58,  6.10s/it]
Training Epoch: 14/100, batch 222/438 completed (loss: 0.0716:  51%|█████     | 112/219 [11:16<10:48,  6.06s/it]
Training Epoch: 14/100, batch 222/438 completed (loss: 0.0466:  51%|█████     | 112/219 [11:16<10:48,  6.06s/it]
Training Epoch: 14/100, batch 222/438 completed (loss: 0.0193:  51%|█████     | 112/219 [11:16<10:48,  6.06s/it]
Training Epoch: 14/100, batch 222/438 completed (loss: 0.2215:  51%|█████     | 112/219 [11:16<10:48,  6.06s/it]
Training Epoch: 14/100, batch 223/438 completed (loss: 0.025:  51%|█████     | 112/219 [11:16<10:48,  6.06s/it] 
Training Epoch: 14/100, batch 223/438 completed (loss: 0.3642:  51%|█████     | 112/219 [11:16<10:48,  6.06s/it]
Training Epoch: 14/100, batch 223/438 completed (loss: 0.1547:  51%|█████     | 112/219 [11:16<10:48,  6.06s/it]
Training Epoch: 14/100, batch 223/438 completed (loss: 0.0875:  51%|█████     | 112/219 [11:16<10:48,  6.06s/it]
Training Epoch: 14/100, batch 224/438 completed (loss: 0.1533:  51%|█████     | 112/219 [11:19<10:48,  6.06s/it]
Training Epoch: 14/100, batch 224/438 completed (loss: 0.1507:  51%|█████     | 112/219 [11:19<10:48,  6.06s/it]
Training Epoch: 14/100, batch 224/438 completed (loss: 0.3222:  51%|█████     | 112/219 [11:19<10:48,  6.06s/it]
Training Epoch: 14/100, batch 224/438 completed (loss: 0.0796:  51%|█████     | 112/219 [11:19<10:48,  6.06s/it]
Training Epoch: 14/100, batch 224/438 completed (loss: 0.3222:  52%|█████▏    | 113/219 [11:22<10:40,  6.05s/it]
Training Epoch: 14/100, batch 224/438 completed (loss: 0.1507:  52%|█████▏    | 113/219 [11:22<10:40,  6.05s/it]
Training Epoch: 14/100, batch 224/438 completed (loss: 0.1533:  52%|█████▏    | 113/219 [11:22<10:41,  6.05s/it]
Training Epoch: 14/100, batch 224/438 completed (loss: 0.0796:  52%|█████▏    | 113/219 [11:22<10:41,  6.05s/it]
Training Epoch: 14/100, batch 225/438 completed (loss: 0.1349:  52%|█████▏    | 113/219 [11:22<10:41,  6.05s/it]
Training Epoch: 14/100, batch 225/438 completed (loss: 0.0517:  52%|█████▏    | 113/219 [11:22<10:40,  6.05s/it]
Training Epoch: 14/100, batch 225/438 completed (loss: 0.0477:  52%|█████▏    | 113/219 [11:22<10:40,  6.05s/it]
Training Epoch: 14/100, batch 225/438 completed (loss: 0.0497:  52%|█████▏    | 113/219 [11:22<10:41,  6.05s/it]
Training Epoch: 14/100, batch 226/438 completed (loss: 0.0065:  52%|█████▏    | 113/219 [11:25<10:40,  6.05s/it]
Training Epoch: 14/100, batch 226/438 completed (loss: 0.016:  52%|█████▏    | 113/219 [11:25<10:40,  6.05s/it] 
Training Epoch: 14/100, batch 226/438 completed (loss: 0.0449:  52%|█████▏    | 113/219 [11:25<10:41,  6.05s/it]
Training Epoch: 14/100, batch 226/438 completed (loss: 0.0747:  52%|█████▏    | 113/219 [11:25<10:41,  6.05s/it]
Training Epoch: 14/100, batch 226/438 completed (loss: 0.016:  52%|█████▏    | 114/219 [11:28<10:41,  6.11s/it]
Training Epoch: 14/100, batch 226/438 completed (loss: 0.0065:  52%|█████▏    | 114/219 [11:28<10:41,  6.11s/it]
Training Epoch: 14/100, batch 226/438 completed (loss: 0.0747:  52%|█████▏    | 114/219 [11:28<10:41,  6.11s/it]
Training Epoch: 14/100, batch 226/438 completed (loss: 0.0449:  52%|█████▏    | 114/219 [11:28<10:42,  6.11s/it]
Training Epoch: 14/100, batch 227/438 completed (loss: 0.2435:  52%|█████▏    | 114/219 [11:28<10:41,  6.11s/it]
Training Epoch: 14/100, batch 227/438 completed (loss: 0.2659:  52%|█████▏    | 114/219 [11:28<10:42,  6.11s/it]
Training Epoch: 14/100, batch 227/438 completed (loss: 0.163:  52%|█████▏    | 114/219 [11:28<10:41,  6.11s/it]
Training Epoch: 14/100, batch 227/438 completed (loss: 0.0387:  52%|█████▏    | 114/219 [11:28<10:41,  6.11s/it]
Training Epoch: 14/100, batch 228/438 completed (loss: 0.0181:  52%|█████▏    | 114/219 [11:31<10:41,  6.11s/it]
Training Epoch: 14/100, batch 228/438 completed (loss: 0.2993:  52%|█████▏    | 114/219 [11:31<10:42,  6.11s/it]
Training Epoch: 14/100, batch 228/438 completed (loss: 0.3021:  52%|█████▏    | 114/219 [11:31<10:41,  6.11s/it]
Training Epoch: 14/100, batch 228/438 completed (loss: 0.1004:  52%|█████▏    | 114/219 [11:31<10:41,  6.11s/it]
Training Epoch: 14/100, batch 228/438 completed (loss: 0.1004:  53%|█████▎    | 115/219 [11:34<10:32,  6.08s/it]
Training Epoch: 14/100, batch 228/438 completed (loss: 0.2993:  53%|█████▎    | 115/219 [11:34<10:32,  6.08s/it]
Training Epoch: 14/100, batch 228/438 completed (loss: 0.0181:  53%|█████▎    | 115/219 [11:34<10:32,  6.08s/it]
Training Epoch: 14/100, batch 228/438 completed (loss: 0.3021:  53%|█████▎    | 115/219 [11:34<10:32,  6.08s/it]
Training Epoch: 14/100, batch 229/438 completed (loss: 0.0907:  53%|█████▎    | 115/219 [11:34<10:32,  6.08s/it]
Training Epoch: 14/100, batch 229/438 completed (loss: 0.2297:  53%|█████▎    | 115/219 [11:34<10:32,  6.08s/it]
Training Epoch: 14/100, batch 229/438 completed (loss: 0.0943:  53%|█████▎    | 115/219 [11:34<10:32,  6.08s/it]
Training Epoch: 14/100, batch 229/438 completed (loss: 0.2519:  53%|█████▎    | 115/219 [11:34<10:32,  6.08s/it]
Training Epoch: 14/100, batch 230/438 completed (loss: 0.1923:  53%|█████▎    | 115/219 [11:37<10:32,  6.08s/it]
Training Epoch: 14/100, batch 230/438 completed (loss: 0.0632:  53%|█████▎    | 115/219 [11:37<10:32,  6.08s/it]
Training Epoch: 14/100, batch 230/438 completed (loss: 0.0466:  53%|█████▎    | 115/219 [11:37<10:32,  6.08s/it]
Training Epoch: 14/100, batch 230/438 completed (loss: 0.1036:  53%|█████▎    | 115/219 [11:37<10:32,  6.08s/it]
Training Epoch: 14/100, batch 230/438 completed (loss: 0.1923:  53%|█████▎    | 116/219 [11:40<10:27,  6.09s/it]
Training Epoch: 14/100, batch 230/438 completed (loss: 0.0632:  53%|█████▎    | 116/219 [11:40<10:27,  6.09s/it]
Training Epoch: 14/100, batch 230/438 completed (loss: 0.0466:  53%|█████▎    | 116/219 [11:40<10:27,  6.09s/it]
Training Epoch: 14/100, batch 230/438 completed (loss: 0.1036:  53%|█████▎    | 116/219 [11:40<10:27,  6.09s/it]
Training Epoch: 14/100, batch 231/438 completed (loss: 0.1035:  53%|█████▎    | 116/219 [11:40<10:27,  6.09s/it]
Training Epoch: 14/100, batch 231/438 completed (loss: 0.0368:  53%|█████▎    | 116/219 [11:40<10:27,  6.09s/it]
Training Epoch: 14/100, batch 231/438 completed (loss: 0.0963:  53%|█████▎    | 116/219 [11:40<10:27,  6.09s/it]
Training Epoch: 14/100, batch 231/438 completed (loss: 0.0732:  53%|█████▎    | 116/219 [11:40<10:27,  6.09s/it]
Training Epoch: 14/100, batch 232/438 completed (loss: 0.0328:  53%|█████▎    | 116/219 [11:43<10:27,  6.09s/it]
Training Epoch: 14/100, batch 232/438 completed (loss: 0.0796:  53%|█████▎    | 116/219 [11:43<10:27,  6.09s/it]
Training Epoch: 14/100, batch 232/438 completed (loss: 0.0839:  53%|█████▎    | 116/219 [11:43<10:27,  6.09s/it]
Training Epoch: 14/100, batch 232/438 completed (loss: 0.0635:  53%|█████▎    | 116/219 [11:43<10:27,  6.09s/it]
Training Epoch: 14/100, batch 232/438 completed (loss: 0.0635:  53%|█████▎    | 117/219 [11:46<10:17,  6.05s/it]
Training Epoch: 14/100, batch 232/438 completed (loss: 0.0839:  53%|█████▎    | 117/219 [11:46<10:17,  6.05s/it]
Training Epoch: 14/100, batch 232/438 completed (loss: 0.0328:  53%|█████▎    | 117/219 [11:46<10:17,  6.05s/it]
Training Epoch: 14/100, batch 232/438 completed (loss: 0.0796:  53%|█████▎    | 117/219 [11:46<10:17,  6.05s/it]
Training Epoch: 14/100, batch 233/438 completed (loss: 0.0888:  53%|█████▎    | 117/219 [11:46<10:17,  6.05s/it]
Training Epoch: 14/100, batch 233/438 completed (loss: 0.1717:  53%|█████▎    | 117/219 [11:46<10:17,  6.05s/it]
Training Epoch: 14/100, batch 233/438 completed (loss: 0.2259:  53%|█████▎    | 117/219 [11:46<10:17,  6.05s/it]
Training Epoch: 14/100, batch 233/438 completed (loss: 0.0382:  53%|█████▎    | 117/219 [11:46<10:17,  6.05s/it]
Training Epoch: 14/100, batch 234/438 completed (loss: 0.753:  53%|█████▎    | 117/219 [11:49<10:17,  6.05s/it] 
Training Epoch: 14/100, batch 234/438 completed (loss: 0.024:  53%|█████▎    | 117/219 [11:49<10:17,  6.05s/it] 
Training Epoch: 14/100, batch 234/438 completed (loss: 0.049:  53%|█████▎    | 117/219 [11:49<10:17,  6.05s/it] 
Training Epoch: 14/100, batch 234/438 completed (loss: 0.0084:  53%|█████▎    | 117/219 [11:49<10:17,  6.05s/it]
Training Epoch: 14/100, batch 234/438 completed (loss: 0.753:  54%|█████▍    | 118/219 [11:52<10:08,  6.03s/it]
Training Epoch: 14/100, batch 234/438 completed (loss: 0.024:  54%|█████▍    | 118/219 [11:52<10:08,  6.03s/it]
Training Epoch: 14/100, batch 234/438 completed (loss: 0.049:  54%|█████▍    | 118/219 [11:52<10:08,  6.03s/it]
Training Epoch: 14/100, batch 234/438 completed (loss: 0.0084:  54%|█████▍    | 118/219 [11:52<10:09,  6.03s/it]
Training Epoch: 14/100, batch 235/438 completed (loss: 0.0355:  54%|█████▍    | 118/219 [11:52<10:08,  6.03s/it]
Training Epoch: 14/100, batch 235/438 completed (loss: 0.1019:  54%|█████▍    | 118/219 [11:52<10:08,  6.03s/it]
Training Epoch: 14/100, batch 235/438 completed (loss: 0.1661:  54%|█████▍    | 118/219 [11:52<10:08,  6.03s/it]
Training Epoch: 14/100, batch 235/438 completed (loss: 0.0054:  54%|█████▍    | 118/219 [11:52<10:09,  6.03s/it]
Training Epoch: 14/100, batch 236/438 completed (loss: 0.1651:  54%|█████▍    | 118/219 [11:55<10:08,  6.03s/it]
Training Epoch: 14/100, batch 236/438 completed (loss: 0.0145:  54%|█████▍    | 118/219 [11:55<10:08,  6.03s/it]
Training Epoch: 14/100, batch 236/438 completed (loss: 0.0176:  54%|█████▍    | 118/219 [11:55<10:08,  6.03s/it]
Training Epoch: 14/100, batch 236/438 completed (loss: 0.0764:  54%|█████▍    | 118/219 [11:55<10:09,  6.03s/it]
Training Epoch: 14/100, batch 236/438 completed (loss: 0.0145:  54%|█████▍    | 119/219 [11:58<10:10,  6.11s/it]
Training Epoch: 14/100, batch 236/438 completed (loss: 0.0176:  54%|█████▍    | 119/219 [11:58<10:10,  6.11s/it]
Training Epoch: 14/100, batch 236/438 completed (loss: 0.1651:  54%|█████▍    | 119/219 [11:58<10:10,  6.11s/it]
Training Epoch: 14/100, batch 236/438 completed (loss: 0.0764:  54%|█████▍    | 119/219 [11:58<10:10,  6.11s/it]
Training Epoch: 14/100, batch 237/438 completed (loss: 0.0353:  54%|█████▍    | 119/219 [11:58<10:10,  6.11s/it]
Training Epoch: 14/100, batch 237/438 completed (loss: 0.0503:  54%|█████▍    | 119/219 [11:58<10:10,  6.11s/it]
Training Epoch: 14/100, batch 237/438 completed (loss: 0.0559:  54%|█████▍    | 119/219 [11:58<10:10,  6.11s/it]
Training Epoch: 14/100, batch 237/438 completed (loss: 0.3188:  54%|█████▍    | 119/219 [11:58<10:10,  6.11s/it]
Training Epoch: 14/100, batch 238/438 completed (loss: 0.0222:  54%|█████▍    | 119/219 [12:01<10:10,  6.11s/it]
Training Epoch: 14/100, batch 238/438 completed (loss: 0.1039:  54%|█████▍    | 119/219 [12:01<10:10,  6.11s/it]
Training Epoch: 14/100, batch 238/438 completed (loss: 0.357:  54%|█████▍    | 119/219 [12:01<10:10,  6.11s/it] 
Training Epoch: 14/100, batch 238/438 completed (loss: 0.0646:  54%|█████▍    | 119/219 [12:01<10:10,  6.11s/it]
Training Epoch: 14/100, batch 238/438 completed (loss: 0.1039:  55%|█████▍    | 120/219 [12:04<09:56,  6.03s/it]
Training Epoch: 14/100, batch 238/438 completed (loss: 0.0222:  55%|█████▍    | 120/219 [12:04<09:56,  6.03s/it]
Training Epoch: 14/100, batch 238/438 completed (loss: 0.357:  55%|█████▍    | 120/219 [12:04<09:56,  6.03s/it]
Training Epoch: 14/100, batch 238/438 completed (loss: 0.0646:  55%|█████▍    | 120/219 [12:04<09:57,  6.03s/it]
Training Epoch: 14/100, batch 239/438 completed (loss: 0.3533:  55%|█████▍    | 120/219 [12:04<09:56,  6.03s/it]
Training Epoch: 14/100, batch 239/438 completed (loss: 0.0228:  55%|█████▍    | 120/219 [12:04<09:56,  6.03s/it]
Training Epoch: 14/100, batch 239/438 completed (loss: 0.0484:  55%|█████▍    | 120/219 [12:04<09:56,  6.03s/it]
Training Epoch: 14/100, batch 239/438 completed (loss: 0.0438:  55%|█████▍    | 120/219 [12:04<09:57,  6.03s/it]
Training Epoch: 14/100, batch 240/438 completed (loss: 0.0467:  55%|█████▍    | 120/219 [12:07<09:56,  6.03s/it]
Training Epoch: 14/100, batch 240/438 completed (loss: 0.011:  55%|█████▍    | 120/219 [12:07<09:56,  6.03s/it] 
Training Epoch: 14/100, batch 240/438 completed (loss: 0.0223:  55%|█████▍    | 120/219 [12:07<09:56,  6.03s/it]
Training Epoch: 14/100, batch 240/438 completed (loss: 0.0172:  55%|█████▍    | 120/219 [12:07<09:57,  6.03s/it]
Training Epoch: 14/100, batch 240/438 completed (loss: 0.011:  55%|█████▌    | 121/219 [12:10<09:54,  6.06s/it]
Training Epoch: 14/100, batch 240/438 completed (loss: 0.0467:  55%|█████▌    | 121/219 [12:10<09:54,  6.06s/it]
Training Epoch: 14/100, batch 240/438 completed (loss: 0.0172:  55%|█████▌    | 121/219 [12:10<09:54,  6.06s/it]
Training Epoch: 14/100, batch 240/438 completed (loss: 0.0223:  55%|█████▌    | 121/219 [12:10<09:54,  6.07s/it]
Training Epoch: 14/100, batch 241/438 completed (loss: 0.2364:  55%|█████▌    | 121/219 [12:10<09:54,  6.06s/it]
Training Epoch: 14/100, batch 241/438 completed (loss: 0.0422:  55%|█████▌    | 121/219 [12:10<09:54,  6.07s/it]
Training Epoch: 14/100, batch 241/438 completed (loss: 0.1175:  55%|█████▌    | 121/219 [12:10<09:54,  6.06s/it]
Training Epoch: 14/100, batch 241/438 completed (loss: 0.0571:  55%|█████▌    | 121/219 [12:10<09:54,  6.06s/it]
Training Epoch: 14/100, batch 242/438 completed (loss: 0.0582:  55%|█████▌    | 121/219 [12:13<09:54,  6.06s/it]
Training Epoch: 14/100, batch 242/438 completed (loss: 0.0131:  55%|█████▌    | 121/219 [12:13<09:54,  6.06s/it]
Training Epoch: 14/100, batch 242/438 completed (loss: 0.2237:  55%|█████▌    | 121/219 [12:13<09:54,  6.07s/it]
Training Epoch: 14/100, batch 242/438 completed (loss: 0.0895:  55%|█████▌    | 121/219 [12:13<09:54,  6.06s/it]
Training Epoch: 14/100, batch 242/438 completed (loss: 0.0131:  56%|█████▌    | 122/219 [12:16<09:46,  6.05s/it]
Training Epoch: 14/100, batch 242/438 completed (loss: 0.0582:  56%|█████▌    | 122/219 [12:16<09:46,  6.05s/it]
Training Epoch: 14/100, batch 242/438 completed (loss: 0.2237:  56%|█████▌    | 122/219 [12:16<09:46,  6.05s/it]
Training Epoch: 14/100, batch 242/438 completed (loss: 0.0895:  56%|█████▌    | 122/219 [12:16<09:47,  6.05s/it]
Training Epoch: 14/100, batch 243/438 completed (loss: 0.0506:  56%|█████▌    | 122/219 [12:16<09:46,  6.05s/it]
Training Epoch: 14/100, batch 243/438 completed (loss: 0.0365:  56%|█████▌    | 122/219 [12:16<09:46,  6.05s/it]
Training Epoch: 14/100, batch 243/438 completed (loss: 0.2633:  56%|█████▌    | 122/219 [12:16<09:46,  6.05s/it]
Training Epoch: 14/100, batch 243/438 completed (loss: 0.0367:  56%|█████▌    | 122/219 [12:16<09:47,  6.05s/it]
Training Epoch: 14/100, batch 244/438 completed (loss: 0.0588:  56%|█████▌    | 122/219 [12:19<09:46,  6.05s/it]
Training Epoch: 14/100, batch 244/438 completed (loss: 0.0169:  56%|█████▌    | 122/219 [12:19<09:46,  6.05s/it]
Training Epoch: 14/100, batch 244/438 completed (loss: 0.0654:  56%|█████▌    | 122/219 [12:19<09:46,  6.05s/it]
Training Epoch: 14/100, batch 244/438 completed (loss: 0.0218:  56%|█████▌    | 122/219 [12:19<09:47,  6.05s/it]
Training Epoch: 14/100, batch 244/438 completed (loss: 0.0588:  56%|█████▌    | 123/219 [12:22<09:38,  6.02s/it]
Training Epoch: 14/100, batch 244/438 completed (loss: 0.0654:  56%|█████▌    | 123/219 [12:22<09:38,  6.02s/it]
Training Epoch: 14/100, batch 244/438 completed (loss: 0.0169:  56%|█████▌    | 123/219 [12:22<09:38,  6.02s/it]
Training Epoch: 14/100, batch 244/438 completed (loss: 0.0218:  56%|█████▌    | 123/219 [12:22<09:38,  6.02s/it]
Training Epoch: 14/100, batch 245/438 completed (loss: 0.0094:  56%|█████▌    | 123/219 [12:22<09:38,  6.02s/it]
Training Epoch: 14/100, batch 245/438 completed (loss: 0.2221:  56%|█████▌    | 123/219 [12:22<09:38,  6.02s/it]
Training Epoch: 14/100, batch 245/438 completed (loss: 0.0911:  56%|█████▌    | 123/219 [12:22<09:38,  6.02s/it]
Training Epoch: 14/100, batch 245/438 completed (loss: 0.1151:  56%|█████▌    | 123/219 [12:22<09:38,  6.02s/it]
Training Epoch: 14/100, batch 246/438 completed (loss: 0.0273:  56%|█████▌    | 123/219 [12:25<09:38,  6.02s/it]
Training Epoch: 14/100, batch 246/438 completed (loss: 0.0338:  56%|█████▌    | 123/219 [12:25<09:38,  6.02s/it]
Training Epoch: 14/100, batch 246/438 completed (loss: 0.0254:  56%|█████▌    | 123/219 [12:25<09:38,  6.02s/it]
Training Epoch: 14/100, batch 246/438 completed (loss: 0.0336:  56%|█████▌    | 123/219 [12:25<09:38,  6.02s/it]
Training Epoch: 14/100, batch 246/438 completed (loss: 0.0338:  57%|█████▋    | 124/219 [12:28<09:35,  6.05s/it]
Training Epoch: 14/100, batch 246/438 completed (loss: 0.0273:  57%|█████▋    | 124/219 [12:28<09:35,  6.06s/it]
Training Epoch: 14/100, batch 246/438 completed (loss: 0.0254:  57%|█████▋    | 124/219 [12:28<09:35,  6.06s/it]
Training Epoch: 14/100, batch 246/438 completed (loss: 0.0336:  57%|█████▋    | 124/219 [12:28<09:35,  6.05s/it]
Training Epoch: 14/100, batch 247/438 completed (loss: 0.0386:  57%|█████▋    | 124/219 [12:28<09:35,  6.05s/it]
Training Epoch: 14/100, batch 247/438 completed (loss: 0.0153:  57%|█████▋    | 124/219 [12:28<09:35,  6.06s/it]
Training Epoch: 14/100, batch 247/438 completed (loss: 0.087:  57%|█████▋    | 124/219 [12:28<09:35,  6.06s/it] 
Training Epoch: 14/100, batch 247/438 completed (loss: 0.0718:  57%|█████▋    | 124/219 [12:29<09:35,  6.05s/it]
Training Epoch: 14/100, batch 248/438 completed (loss: 0.0349:  57%|█████▋    | 124/219 [12:32<09:35,  6.06s/it]
Training Epoch: 14/100, batch 248/438 completed (loss: 0.0676:  57%|█████▋    | 124/219 [12:32<09:35,  6.05s/it]
Training Epoch: 14/100, batch 248/438 completed (loss: 0.3825:  57%|█████▋    | 124/219 [12:32<09:35,  6.06s/it]
Training Epoch: 14/100, batch 248/438 completed (loss: 0.2024:  57%|█████▋    | 124/219 [12:32<09:35,  6.05s/it]
Training Epoch: 14/100, batch 248/438 completed (loss: 0.3825:  57%|█████▋    | 125/219 [12:34<09:26,  6.03s/it]
Training Epoch: 14/100, batch 248/438 completed (loss: 0.0349:  57%|█████▋    | 125/219 [12:34<09:26,  6.03s/it]
Training Epoch: 14/100, batch 248/438 completed (loss: 0.0676:  57%|█████▋    | 125/219 [12:34<09:26,  6.03s/it]
Training Epoch: 14/100, batch 248/438 completed (loss: 0.2024:  57%|█████▋    | 125/219 [12:34<09:26,  6.03s/it]
Training Epoch: 14/100, batch 249/438 completed (loss: 0.0337:  57%|█████▋    | 125/219 [12:34<09:26,  6.03s/it]
Training Epoch: 14/100, batch 249/438 completed (loss: 0.0617:  57%|█████▋    | 125/219 [12:34<09:26,  6.03s/it]
Training Epoch: 14/100, batch 249/438 completed (loss: 0.1301:  57%|█████▋    | 125/219 [12:34<09:26,  6.03s/it]
Training Epoch: 14/100, batch 249/438 completed (loss: 0.0619:  57%|█████▋    | 125/219 [12:34<09:26,  6.03s/it]
Training Epoch: 14/100, batch 250/438 completed (loss: 0.0318:  57%|█████▋    | 125/219 [12:37<09:26,  6.03s/it]
Training Epoch: 14/100, batch 250/438 completed (loss: 0.0619:  57%|█████▋    | 125/219 [12:37<09:26,  6.03s/it]
Training Epoch: 14/100, batch 250/438 completed (loss: 0.4663:  57%|█████▋    | 125/219 [12:37<09:26,  6.03s/it]
Training Epoch: 14/100, batch 250/438 completed (loss: 0.1912:  57%|█████▋    | 125/219 [12:37<09:26,  6.03s/it]
Training Epoch: 14/100, batch 250/438 completed (loss: 0.0318:  58%|█████▊    | 126/219 [12:41<09:24,  6.07s/it]
Training Epoch: 14/100, batch 250/438 completed (loss: 0.0619:  58%|█████▊    | 126/219 [12:41<09:24,  6.07s/it]
Training Epoch: 14/100, batch 250/438 completed (loss: 0.4663:  58%|█████▊    | 126/219 [12:41<09:24,  6.07s/it]
Training Epoch: 14/100, batch 250/438 completed (loss: 0.1912:  58%|█████▊    | 126/219 [12:41<09:24,  6.07s/it]
Training Epoch: 14/100, batch 251/438 completed (loss: 0.0123:  58%|█████▊    | 126/219 [12:41<09:24,  6.07s/it]
Training Epoch: 14/100, batch 251/438 completed (loss: 0.1556:  58%|█████▊    | 126/219 [12:41<09:24,  6.07s/it]
Training Epoch: 14/100, batch 251/438 completed (loss: 0.1235:  58%|█████▊    | 126/219 [12:41<09:24,  6.07s/it]
Training Epoch: 14/100, batch 251/438 completed (loss: 0.1329:  58%|█████▊    | 126/219 [12:41<09:24,  6.07s/it]
Training Epoch: 14/100, batch 252/438 completed (loss: 0.0261:  58%|█████▊    | 126/219 [12:44<09:24,  6.07s/it]
Training Epoch: 14/100, batch 252/438 completed (loss: 0.0409:  58%|█████▊    | 126/219 [12:44<09:24,  6.07s/it]
Training Epoch: 14/100, batch 252/438 completed (loss: 0.0448:  58%|█████▊    | 126/219 [12:44<09:24,  6.07s/it]
Training Epoch: 14/100, batch 252/438 completed (loss: 0.0522:  58%|█████▊    | 126/219 [12:44<09:24,  6.07s/it]
Training Epoch: 14/100, batch 252/438 completed (loss: 0.0261:  58%|█████▊    | 127/219 [12:47<09:20,  6.09s/it]
Training Epoch: 14/100, batch 252/438 completed (loss: 0.0409:  58%|█████▊    | 127/219 [12:47<09:20,  6.09s/it]
Training Epoch: 14/100, batch 252/438 completed (loss: 0.0448:  58%|█████▊    | 127/219 [12:47<09:20,  6.09s/it]
Training Epoch: 14/100, batch 252/438 completed (loss: 0.0522:  58%|█████▊    | 127/219 [12:47<09:20,  6.09s/it]
Training Epoch: 14/100, batch 253/438 completed (loss: 0.0542:  58%|█████▊    | 127/219 [12:47<09:20,  6.09s/it]
Training Epoch: 14/100, batch 253/438 completed (loss: 0.0246:  58%|█████▊    | 127/219 [12:47<09:20,  6.09s/it]
Training Epoch: 14/100, batch 253/438 completed (loss: 0.0736:  58%|█████▊    | 127/219 [12:47<09:20,  6.09s/it]
Training Epoch: 14/100, batch 253/438 completed (loss: 0.2777:  58%|█████▊    | 127/219 [12:47<09:20,  6.09s/it]
Training Epoch: 14/100, batch 254/438 completed (loss: 0.188:  58%|█████▊    | 127/219 [12:50<09:20,  6.09s/it] 
Training Epoch: 14/100, batch 254/438 completed (loss: 0.0162:  58%|█████▊    | 127/219 [12:50<09:20,  6.09s/it]
Training Epoch: 14/100, batch 254/438 completed (loss: 0.0303:  58%|█████▊    | 127/219 [12:50<09:20,  6.09s/it]
Training Epoch: 14/100, batch 254/438 completed (loss: 0.021:  58%|█████▊    | 127/219 [12:50<09:20,  6.09s/it] 
Training Epoch: 14/100, batch 254/438 completed (loss: 0.0162:  58%|█████▊    | 128/219 [12:53<09:08,  6.02s/it]
Training Epoch: 14/100, batch 254/438 completed (loss: 0.188:  58%|█████▊    | 128/219 [12:53<09:08,  6.02s/it]
Training Epoch: 14/100, batch 254/438 completed (loss: 0.021:  58%|█████▊    | 128/219 [12:53<09:08,  6.02s/it]
Training Epoch: 14/100, batch 254/438 completed (loss: 0.0303:  58%|█████▊    | 128/219 [12:53<09:08,  6.03s/it]
Training Epoch: 14/100, batch 255/438 completed (loss: 0.0544:  58%|█████▊    | 128/219 [12:53<09:08,  6.02s/it]
Training Epoch: 14/100, batch 255/438 completed (loss: 0.0121:  58%|█████▊    | 128/219 [12:53<09:08,  6.02s/it]
Training Epoch: 14/100, batch 255/438 completed (loss: 0.1011:  58%|█████▊    | 128/219 [12:53<09:08,  6.03s/it]
Training Epoch: 14/100, batch 255/438 completed (loss: 0.1573:  58%|█████▊    | 128/219 [12:53<09:08,  6.02s/it]
Training Epoch: 14/100, batch 256/438 completed (loss: 0.032:  58%|█████▊    | 128/219 [12:56<09:08,  6.02s/it] 
Training Epoch: 14/100, batch 256/438 completed (loss: 0.4377:  58%|█████▊    | 128/219 [12:56<09:08,  6.02s/it]
Training Epoch: 14/100, batch 256/438 completed (loss: 0.7032:  58%|█████▊    | 128/219 [12:56<09:08,  6.03s/it]
Training Epoch: 14/100, batch 256/438 completed (loss: 0.0032:  58%|█████▊    | 128/219 [12:56<09:08,  6.02s/it]
Training Epoch: 14/100, batch 256/438 completed (loss: 0.0032:  59%|█████▉    | 129/219 [12:59<09:03,  6.03s/it]
Training Epoch: 14/100, batch 256/438 completed (loss: 0.7032:  59%|█████▉    | 129/219 [12:59<09:03,  6.04s/it]
Training Epoch: 14/100, batch 256/438 completed (loss: 0.032:  59%|█████▉    | 129/219 [12:59<09:03,  6.04s/it]
Training Epoch: 14/100, batch 256/438 completed (loss: 0.4377:  59%|█████▉    | 129/219 [12:59<09:03,  6.04s/it]
Training Epoch: 14/100, batch 257/438 completed (loss: 0.0223:  59%|█████▉    | 129/219 [12:59<09:03,  6.04s/it]
Training Epoch: 14/100, batch 257/438 completed (loss: 0.0574:  59%|█████▉    | 129/219 [12:59<09:03,  6.04s/it]
Training Epoch: 14/100, batch 257/438 completed (loss: 0.1302:  59%|█████▉    | 129/219 [12:59<09:03,  6.04s/it]
Training Epoch: 14/100, batch 257/438 completed (loss: 0.4899:  59%|█████▉    | 129/219 [12:59<09:03,  6.03s/it]
Training Epoch: 14/100, batch 258/438 completed (loss: 0.2186:  59%|█████▉    | 129/219 [13:02<09:03,  6.04s/it]
Training Epoch: 14/100, batch 258/438 completed (loss: 0.0162:  59%|█████▉    | 129/219 [13:02<09:03,  6.04s/it]
Training Epoch: 14/100, batch 258/438 completed (loss: 0.0086:  59%|█████▉    | 129/219 [13:02<09:03,  6.04s/it]
Training Epoch: 14/100, batch 258/438 completed (loss: 0.0167:  59%|█████▉    | 129/219 [13:02<09:03,  6.03s/it]
Training Epoch: 14/100, batch 258/438 completed (loss: 0.0162:  59%|█████▉    | 130/219 [13:05<08:57,  6.04s/it]
Training Epoch: 14/100, batch 258/438 completed (loss: 0.0167:  59%|█████▉    | 130/219 [13:05<08:57,  6.04s/it]
Training Epoch: 14/100, batch 258/438 completed (loss: 0.2186:  59%|█████▉    | 130/219 [13:05<08:57,  6.04s/it]
Training Epoch: 14/100, batch 258/438 completed (loss: 0.0086:  59%|█████▉    | 130/219 [13:05<08:57,  6.04s/it]
Training Epoch: 14/100, batch 259/438 completed (loss: 0.0324:  59%|█████▉    | 130/219 [13:05<08:57,  6.04s/it]
Training Epoch: 14/100, batch 259/438 completed (loss: 0.0068:  59%|█████▉    | 130/219 [13:05<08:57,  6.04s/it]
Training Epoch: 14/100, batch 259/438 completed (loss: 0.1392:  59%|█████▉    | 130/219 [13:05<08:57,  6.04s/it]
Training Epoch: 14/100, batch 259/438 completed (loss: 0.007:  59%|█████▉    | 130/219 [13:05<08:57,  6.04s/it] 
Training Epoch: 14/100, batch 260/438 completed (loss: 0.0333:  59%|█████▉    | 130/219 [13:08<08:57,  6.04s/it]
Training Epoch: 14/100, batch 260/438 completed (loss: 0.4666:  59%|█████▉    | 130/219 [13:08<08:57,  6.04s/it]
Training Epoch: 14/100, batch 260/438 completed (loss: 0.06:  59%|█████▉    | 130/219 [13:08<08:57,  6.04s/it]  
Training Epoch: 14/100, batch 260/438 completed (loss: 0.2259:  59%|█████▉    | 130/219 [13:08<08:57,  6.04s/it]
Training Epoch: 14/100, batch 260/438 completed (loss: 0.2259:  60%|█████▉    | 131/219 [13:11<08:49,  6.02s/it]
Training Epoch: 14/100, batch 260/438 completed (loss: 0.06:  60%|█████▉    | 131/219 [13:11<08:49,  6.02s/it]
Training Epoch: 14/100, batch 260/438 completed (loss: 0.4666:  60%|█████▉    | 131/219 [13:11<08:49,  6.02s/it]
Training Epoch: 14/100, batch 260/438 completed (loss: 0.0333:  60%|█████▉    | 131/219 [13:11<08:49,  6.02s/it]
Training Epoch: 14/100, batch 261/438 completed (loss: 0.025:  60%|█████▉    | 131/219 [13:11<08:49,  6.02s/it] 
Training Epoch: 14/100, batch 261/438 completed (loss: 0.0491:  60%|█████▉    | 131/219 [13:11<08:49,  6.02s/it]
Training Epoch: 14/100, batch 261/438 completed (loss: 0.1938:  60%|█████▉    | 131/219 [13:11<08:49,  6.02s/it]
Training Epoch: 14/100, batch 261/438 completed (loss: 0.0476:  60%|█████▉    | 131/219 [13:11<08:49,  6.02s/it]
Training Epoch: 14/100, batch 262/438 completed (loss: 0.0496:  60%|█████▉    | 131/219 [13:14<08:49,  6.02s/it]
Training Epoch: 14/100, batch 262/438 completed (loss: 0.1997:  60%|█████▉    | 131/219 [13:14<08:49,  6.02s/it]
Training Epoch: 14/100, batch 262/438 completed (loss: 0.1004:  60%|█████▉    | 131/219 [13:14<08:49,  6.02s/it]
Training Epoch: 14/100, batch 262/438 completed (loss: 0.129:  60%|█████▉    | 131/219 [13:14<08:49,  6.02s/it] 
Training Epoch: 14/100, batch 262/438 completed (loss: 0.1004:  60%|██████    | 132/219 [13:17<08:51,  6.11s/it]
Training Epoch: 14/100, batch 262/438 completed (loss: 0.1997:  60%|██████    | 132/219 [13:17<08:51,  6.11s/it]
Training Epoch: 14/100, batch 262/438 completed (loss: 0.0496:  60%|██████    | 132/219 [13:17<08:51,  6.11s/it]
Training Epoch: 14/100, batch 262/438 completed (loss: 0.129:  60%|██████    | 132/219 [13:17<08:51,  6.11s/it]
Training Epoch: 14/100, batch 263/438 completed (loss: 0.0635:  60%|██████    | 132/219 [13:17<08:51,  6.11s/it]
Training Epoch: 14/100, batch 263/438 completed (loss: 0.0213:  60%|██████    | 132/219 [13:17<08:51,  6.11s/it]
Training Epoch: 14/100, batch 263/438 completed (loss: 0.0655:  60%|██████    | 132/219 [13:17<08:51,  6.11s/it]
Training Epoch: 14/100, batch 263/438 completed (loss: 0.1883:  60%|██████    | 132/219 [13:17<08:51,  6.11s/it]
Training Epoch: 14/100, batch 264/438 completed (loss: 0.2967:  60%|██████    | 132/219 [13:20<08:51,  6.11s/it]
Training Epoch: 14/100, batch 264/438 completed (loss: 0.0207:  60%|██████    | 132/219 [13:20<08:51,  6.11s/it]
Training Epoch: 14/100, batch 264/438 completed (loss: 0.1513:  60%|██████    | 132/219 [13:20<08:51,  6.11s/it]
Training Epoch: 14/100, batch 264/438 completed (loss: 0.04:  60%|██████    | 132/219 [13:20<08:51,  6.11s/it]  
Training Epoch: 14/100, batch 264/438 completed (loss: 0.04:  61%|██████    | 133/219 [13:23<08:38,  6.03s/it]
Training Epoch: 14/100, batch 264/438 completed (loss: 0.0207:  61%|██████    | 133/219 [13:23<08:38,  6.03s/it]
Training Epoch: 14/100, batch 264/438 completed (loss: 0.1513:  61%|██████    | 133/219 [13:23<08:38,  6.03s/it]
Training Epoch: 14/100, batch 264/438 completed (loss: 0.2967:  61%|██████    | 133/219 [13:23<08:38,  6.03s/it]
Training Epoch: 14/100, batch 265/438 completed (loss: 0.138:  61%|██████    | 133/219 [13:23<08:38,  6.03s/it] 
Training Epoch: 14/100, batch 265/438 completed (loss: 0.0142:  61%|██████    | 133/219 [13:23<08:38,  6.03s/it]
Training Epoch: 14/100, batch 265/438 completed (loss: 0.0166:  61%|██████    | 133/219 [13:23<08:38,  6.03s/it]
Training Epoch: 14/100, batch 265/438 completed (loss: 0.3109:  61%|██████    | 133/219 [13:23<08:38,  6.03s/it]
Training Epoch: 14/100, batch 266/438 completed (loss: 0.6479:  61%|██████    | 133/219 [13:26<08:38,  6.03s/it]
Training Epoch: 14/100, batch 266/438 completed (loss: 0.0088:  61%|██████    | 133/219 [13:26<08:38,  6.03s/it]
Training Epoch: 14/100, batch 266/438 completed (loss: 0.0611:  61%|██████    | 133/219 [13:26<08:38,  6.03s/it]
Training Epoch: 14/100, batch 266/438 completed (loss: 0.0208:  61%|██████    | 133/219 [13:26<08:38,  6.03s/it]
Training Epoch: 14/100, batch 266/438 completed (loss: 0.0208:  61%|██████    | 134/219 [13:29<08:35,  6.07s/it]
Training Epoch: 14/100, batch 266/438 completed (loss: 0.0088:  61%|██████    | 134/219 [13:29<08:36,  6.07s/it]
Training Epoch: 14/100, batch 266/438 completed (loss: 0.6479:  61%|██████    | 134/219 [13:29<08:36,  6.07s/it]
Training Epoch: 14/100, batch 266/438 completed (loss: 0.0611:  61%|██████    | 134/219 [13:29<08:36,  6.08s/it]
Training Epoch: 14/100, batch 267/438 completed (loss: 0.3656:  61%|██████    | 134/219 [13:29<08:36,  6.07s/it]
Training Epoch: 14/100, batch 267/438 completed (loss: 0.0396:  61%|██████    | 134/219 [13:29<08:36,  6.07s/it]
Training Epoch: 14/100, batch 267/438 completed (loss: 0.0603:  61%|██████    | 134/219 [13:29<08:36,  6.08s/it]
Training Epoch: 14/100, batch 267/438 completed (loss: 0.0578:  61%|██████    | 134/219 [13:29<08:35,  6.07s/it]
Training Epoch: 14/100, batch 268/438 completed (loss: 0.0242:  61%|██████    | 134/219 [13:32<08:36,  6.07s/it]
Training Epoch: 14/100, batch 268/438 completed (loss: 0.0681:  61%|██████    | 134/219 [13:32<08:36,  6.08s/it]
Training Epoch: 14/100, batch 268/438 completed (loss: 0.0235:  61%|██████    | 134/219 [13:32<08:36,  6.07s/it]
Training Epoch: 14/100, batch 268/438 completed (loss: 0.6018:  61%|██████    | 134/219 [13:32<08:35,  6.07s/it]
Training Epoch: 14/100, batch 268/438 completed (loss: 0.0235:  62%|██████▏   | 135/219 [13:35<08:31,  6.09s/it]
Training Epoch: 14/100, batch 268/438 completed (loss: 0.0242:  62%|██████▏   | 135/219 [13:35<08:31,  6.09s/it]
Training Epoch: 14/100, batch 268/438 completed (loss: 0.0681:  62%|██████▏   | 135/219 [13:35<08:31,  6.09s/it]
Training Epoch: 14/100, batch 268/438 completed (loss: 0.6018:  62%|██████▏   | 135/219 [13:35<08:31,  6.09s/it]
Training Epoch: 14/100, batch 269/438 completed (loss: 0.0488:  62%|██████▏   | 135/219 [13:35<08:31,  6.09s/it]
Training Epoch: 14/100, batch 269/438 completed (loss: 0.2322:  62%|██████▏   | 135/219 [13:35<08:31,  6.09s/it]
Training Epoch: 14/100, batch 269/438 completed (loss: 0.051:  62%|██████▏   | 135/219 [13:35<08:31,  6.09s/it] 
Training Epoch: 14/100, batch 269/438 completed (loss: 0.2174:  62%|██████▏   | 135/219 [13:35<08:31,  6.09s/it]
Training Epoch: 14/100, batch 270/438 completed (loss: 0.0887:  62%|██████▏   | 135/219 [13:38<08:31,  6.09s/it]
Training Epoch: 14/100, batch 270/438 completed (loss: 0.0047:  62%|██████▏   | 135/219 [13:38<08:31,  6.09s/it]
Training Epoch: 14/100, batch 270/438 completed (loss: 0.0126:  62%|██████▏   | 135/219 [13:38<08:31,  6.09s/it]
Training Epoch: 14/100, batch 270/438 completed (loss: 0.0099:  62%|██████▏   | 135/219 [13:38<08:31,  6.09s/it]
Training Epoch: 14/100, batch 270/438 completed (loss: 0.0047:  62%|██████▏   | 136/219 [13:41<08:19,  6.02s/it]
Training Epoch: 14/100, batch 270/438 completed (loss: 0.0887:  62%|██████▏   | 136/219 [13:41<08:19,  6.02s/it]
Training Epoch: 14/100, batch 270/438 completed (loss: 0.0126:  62%|██████▏   | 136/219 [13:41<08:19,  6.02s/it]
Training Epoch: 14/100, batch 270/438 completed (loss: 0.0099:  62%|██████▏   | 136/219 [13:41<08:20,  6.03s/it]
Training Epoch: 14/100, batch 271/438 completed (loss: 0.3695:  62%|██████▏   | 136/219 [13:41<08:19,  6.02s/it]
Training Epoch: 14/100, batch 271/438 completed (loss: 0.0373:  62%|██████▏   | 136/219 [13:41<08:19,  6.02s/it]
Training Epoch: 14/100, batch 271/438 completed (loss: 0.0099:  62%|██████▏   | 136/219 [13:41<08:19,  6.02s/it]
Training Epoch: 14/100, batch 271/438 completed (loss: 0.2167:  62%|██████▏   | 136/219 [13:41<08:20,  6.03s/it]
Training Epoch: 14/100, batch 272/438 completed (loss: 0.0683:  62%|██████▏   | 136/219 [13:44<08:19,  6.02s/it]
Training Epoch: 14/100, batch 272/438 completed (loss: 0.0101:  62%|██████▏   | 136/219 [13:44<08:19,  6.02s/it]
Training Epoch: 14/100, batch 272/438 completed (loss: 0.2964:  62%|██████▏   | 136/219 [13:44<08:19,  6.02s/it]
Training Epoch: 14/100, batch 272/438 completed (loss: 0.0144:  62%|██████▏   | 136/219 [13:44<08:20,  6.03s/it]
Training Epoch: 14/100, batch 272/438 completed (loss: 0.0144:  63%|██████▎   | 137/219 [13:47<08:16,  6.06s/it]
Training Epoch: 14/100, batch 272/438 completed (loss: 0.0101:  63%|██████▎   | 137/219 [13:47<08:17,  6.06s/it]
Training Epoch: 14/100, batch 272/438 completed (loss: 0.0683:  63%|██████▎   | 137/219 [13:47<08:17,  6.06s/it]
Training Epoch: 14/100, batch 272/438 completed (loss: 0.2964:  63%|██████▎   | 137/219 [13:47<08:17,  6.06s/it]
Training Epoch: 14/100, batch 273/438 completed (loss: 0.2809:  63%|██████▎   | 137/219 [13:47<08:17,  6.06s/it]
Training Epoch: 14/100, batch 273/438 completed (loss: 0.0155:  63%|██████▎   | 137/219 [13:47<08:17,  6.06s/it]
Training Epoch: 14/100, batch 273/438 completed (loss: 0.3587:  63%|██████▎   | 137/219 [13:47<08:17,  6.06s/it]
Training Epoch: 14/100, batch 273/438 completed (loss: 0.0322:  63%|██████▎   | 137/219 [13:47<08:16,  6.06s/it]
Training Epoch: 14/100, batch 274/438 completed (loss: 0.029:  63%|██████▎   | 137/219 [13:50<08:17,  6.06s/it] 
Training Epoch: 14/100, batch 274/438 completed (loss: 0.1447:  63%|██████▎   | 137/219 [13:50<08:17,  6.06s/it]
Training Epoch: 14/100, batch 274/438 completed (loss: 0.0493:  63%|██████▎   | 137/219 [13:50<08:17,  6.06s/it]
Training Epoch: 14/100, batch 274/438 completed (loss: 0.0287:  63%|██████▎   | 137/219 [13:50<08:16,  6.06s/it]
Training Epoch: 14/100, batch 274/438 completed (loss: 0.1447:  63%|██████▎   | 138/219 [13:53<08:08,  6.03s/it]
Training Epoch: 14/100, batch 274/438 completed (loss: 0.0493:  63%|██████▎   | 138/219 [13:53<08:08,  6.03s/it]
Training Epoch: 14/100, batch 274/438 completed (loss: 0.029:  63%|██████▎   | 138/219 [13:53<08:08,  6.03s/it]
Training Epoch: 14/100, batch 274/438 completed (loss: 0.0287:  63%|██████▎   | 138/219 [13:53<08:08,  6.04s/it]
Training Epoch: 14/100, batch 275/438 completed (loss: 0.0473:  63%|██████▎   | 138/219 [13:53<08:08,  6.03s/it]
Training Epoch: 14/100, batch 275/438 completed (loss: 0.05:  63%|██████▎   | 138/219 [13:53<08:08,  6.03s/it]  
Training Epoch: 14/100, batch 275/438 completed (loss: 0.0771:  63%|██████▎   | 138/219 [13:53<08:08,  6.03s/it]
Training Epoch: 14/100, batch 275/438 completed (loss: 0.0252:  63%|██████▎   | 138/219 [13:53<08:08,  6.04s/it]
Training Epoch: 14/100, batch 276/438 completed (loss: 0.0258:  63%|██████▎   | 138/219 [13:56<08:08,  6.03s/it]
Training Epoch: 14/100, batch 276/438 completed (loss: 0.0623:  63%|██████▎   | 138/219 [13:56<08:08,  6.03s/it]
Training Epoch: 14/100, batch 276/438 completed (loss: 0.0694:  63%|██████▎   | 138/219 [13:56<08:08,  6.03s/it]
Training Epoch: 14/100, batch 276/438 completed (loss: 0.5081:  63%|██████▎   | 138/219 [13:56<08:08,  6.04s/it]
Training Epoch: 14/100, batch 276/438 completed (loss: 0.0694:  63%|██████▎   | 139/219 [13:59<08:05,  6.07s/it]
Training Epoch: 14/100, batch 276/438 completed (loss: 0.0623:  63%|██████▎   | 139/219 [13:59<08:05,  6.07s/it]
Training Epoch: 14/100, batch 276/438 completed (loss: 0.0258:  63%|██████▎   | 139/219 [13:59<08:05,  6.07s/it]
Training Epoch: 14/100, batch 276/438 completed (loss: 0.5081:  63%|██████▎   | 139/219 [13:59<08:05,  6.07s/it]
Training Epoch: 14/100, batch 277/438 completed (loss: 0.035:  63%|██████▎   | 139/219 [13:59<08:05,  6.07s/it] 
Training Epoch: 14/100, batch 277/438 completed (loss: 0.2136:  63%|██████▎   | 139/219 [13:59<08:05,  6.07s/it]
Training Epoch: 14/100, batch 277/438 completed (loss: 0.0147:  63%|██████▎   | 139/219 [13:59<08:05,  6.07s/it]
Training Epoch: 14/100, batch 277/438 completed (loss: 0.0719:  63%|██████▎   | 139/219 [13:59<08:05,  6.07s/it]
Training Epoch: 14/100, batch 278/438 completed (loss: 0.1289:  63%|██████▎   | 139/219 [14:02<08:05,  6.07s/it]
Training Epoch: 14/100, batch 278/438 completed (loss: 0.0195:  63%|██████▎   | 139/219 [14:02<08:05,  6.07s/it]
Training Epoch: 14/100, batch 278/438 completed (loss: 0.0283:  63%|██████▎   | 139/219 [14:02<08:05,  6.07s/it]
Training Epoch: 14/100, batch 278/438 completed (loss: 0.052:  63%|██████▎   | 139/219 [14:02<08:05,  6.07s/it] 
Training Epoch: 14/100, batch 278/438 completed (loss: 0.0283:  64%|██████▍   | 140/219 [14:05<08:00,  6.08s/it]
Training Epoch: 14/100, batch 278/438 completed (loss: 0.0195:  64%|██████▍   | 140/219 [14:05<08:00,  6.08s/it]
Training Epoch: 14/100, batch 278/438 completed (loss: 0.1289:  64%|██████▍   | 140/219 [14:05<08:00,  6.08s/it]
Training Epoch: 14/100, batch 278/438 completed (loss: 0.052:  64%|██████▍   | 140/219 [14:05<08:00,  6.08s/it]
Training Epoch: 14/100, batch 279/438 completed (loss: 0.1615:  64%|██████▍   | 140/219 [14:05<08:00,  6.08s/it]
Training Epoch: 14/100, batch 279/438 completed (loss: 0.0733:  64%|██████▍   | 140/219 [14:05<08:00,  6.08s/it]
Training Epoch: 14/100, batch 279/438 completed (loss: 0.1395:  64%|██████▍   | 140/219 [14:05<08:00,  6.08s/it]
Training Epoch: 14/100, batch 279/438 completed (loss: 0.016:  64%|██████▍   | 140/219 [14:05<08:00,  6.08s/it]
Training Epoch: 14/100, batch 280/438 completed (loss: 0.3592:  64%|██████▍   | 140/219 [14:08<08:00,  6.08s/it]
Training Epoch: 14/100, batch 280/438 completed (loss: 0.4383:  64%|██████▍   | 140/219 [14:08<08:00,  6.08s/it]
Training Epoch: 14/100, batch 280/438 completed (loss: 0.0634:  64%|██████▍   | 140/219 [14:08<08:00,  6.08s/it]
Training Epoch: 14/100, batch 280/438 completed (loss: 0.1572:  64%|██████▍   | 140/219 [14:08<08:00,  6.08s/it]
Training Epoch: 14/100, batch 280/438 completed (loss: 0.0634:  64%|██████▍   | 141/219 [14:11<07:49,  6.01s/it]
Training Epoch: 14/100, batch 280/438 completed (loss: 0.4383:  64%|██████▍   | 141/219 [14:11<07:49,  6.01s/it]
Training Epoch: 14/100, batch 280/438 completed (loss: 0.3592:  64%|██████▍   | 141/219 [14:11<07:48,  6.01s/it]
Training Epoch: 14/100, batch 280/438 completed (loss: 0.1572:  64%|██████▍   | 141/219 [14:11<07:48,  6.01s/it]
Training Epoch: 14/100, batch 281/438 completed (loss: 0.1076:  64%|██████▍   | 141/219 [14:11<07:49,  6.01s/it]
Training Epoch: 14/100, batch 281/438 completed (loss: 0.01:  64%|██████▍   | 141/219 [14:11<07:49,  6.01s/it]  
Training Epoch: 14/100, batch 281/438 completed (loss: 0.0342:  64%|██████▍   | 141/219 [14:11<07:48,  6.01s/it]
Training Epoch: 14/100, batch 281/438 completed (loss: 0.2097:  64%|██████▍   | 141/219 [14:11<07:48,  6.01s/it]
Training Epoch: 14/100, batch 282/438 completed (loss: 0.015:  64%|██████▍   | 141/219 [14:14<07:48,  6.01s/it] 
Training Epoch: 14/100, batch 282/438 completed (loss: 0.0718:  64%|██████▍   | 141/219 [14:14<07:49,  6.01s/it]
Training Epoch: 14/100, batch 282/438 completed (loss: 0.2505:  64%|██████▍   | 141/219 [14:14<07:49,  6.01s/it]
Training Epoch: 14/100, batch 282/438 completed (loss: 0.0506:  64%|██████▍   | 141/219 [14:14<07:48,  6.01s/it]
Training Epoch: 14/100, batch 282/438 completed (loss: 0.2505:  65%|██████▍   | 142/219 [14:18<07:50,  6.10s/it]
Training Epoch: 14/100, batch 282/438 completed (loss: 0.0718:  65%|██████▍   | 142/219 [14:18<07:50,  6.11s/it]
Training Epoch: 14/100, batch 282/438 completed (loss: 0.015:  65%|██████▍   | 142/219 [14:18<07:50,  6.11s/it]
Training Epoch: 14/100, batch 282/438 completed (loss: 0.0506:  65%|██████▍   | 142/219 [14:18<07:50,  6.11s/it]
Training Epoch: 14/100, batch 283/438 completed (loss: 0.1859:  65%|██████▍   | 142/219 [14:18<07:50,  6.10s/it]
Training Epoch: 14/100, batch 283/438 completed (loss: 0.0896:  65%|██████▍   | 142/219 [14:18<07:50,  6.11s/it]
Training Epoch: 14/100, batch 283/438 completed (loss: 0.0288:  65%|██████▍   | 142/219 [14:18<07:50,  6.11s/it]
Training Epoch: 14/100, batch 283/438 completed (loss: 0.2736:  65%|██████▍   | 142/219 [14:18<07:50,  6.11s/it]
Training Epoch: 14/100, batch 284/438 completed (loss: 0.0627:  65%|██████▍   | 142/219 [14:21<07:50,  6.11s/it]
Training Epoch: 14/100, batch 284/438 completed (loss: 0.0409:  65%|██████▍   | 142/219 [14:21<07:50,  6.10s/it]
Training Epoch: 14/100, batch 284/438 completed (loss: 0.0583:  65%|██████▍   | 142/219 [14:21<07:50,  6.11s/it]
Training Epoch: 14/100, batch 284/438 completed (loss: 0.1801:  65%|██████▍   | 142/219 [14:21<07:50,  6.11s/it]
Training Epoch: 14/100, batch 284/438 completed (loss: 0.0409:  65%|██████▌   | 143/219 [14:24<07:41,  6.08s/it]
Training Epoch: 14/100, batch 284/438 completed (loss: 0.0583:  65%|██████▌   | 143/219 [14:24<07:41,  6.08s/it]
Training Epoch: 14/100, batch 284/438 completed (loss: 0.0627:  65%|██████▌   | 143/219 [14:24<07:41,  6.08s/it]
Training Epoch: 14/100, batch 284/438 completed (loss: 0.1801:  65%|██████▌   | 143/219 [14:24<07:41,  6.08s/it]
Training Epoch: 14/100, batch 285/438 completed (loss: 0.0786:  65%|██████▌   | 143/219 [14:24<07:41,  6.08s/it]
Training Epoch: 14/100, batch 285/438 completed (loss: 0.1892:  65%|██████▌   | 143/219 [14:24<07:41,  6.08s/it]
Training Epoch: 14/100, batch 285/438 completed (loss: 0.0154:  65%|██████▌   | 143/219 [14:24<07:41,  6.08s/it]
Training Epoch: 14/100, batch 285/438 completed (loss: 0.119:  65%|██████▌   | 143/219 [14:24<07:41,  6.08s/it] 
Training Epoch: 14/100, batch 286/438 completed (loss: 0.0304:  65%|██████▌   | 143/219 [14:27<07:41,  6.08s/it]
Training Epoch: 14/100, batch 286/438 completed (loss: 0.0211:  65%|██████▌   | 143/219 [14:27<07:41,  6.08s/it]
Training Epoch: 14/100, batch 286/438 completed (loss: 0.0529:  65%|██████▌   | 143/219 [14:27<07:41,  6.08s/it]
Training Epoch: 14/100, batch 286/438 completed (loss: 0.0654:  65%|██████▌   | 143/219 [14:27<07:41,  6.08s/it]
Training Epoch: 14/100, batch 286/438 completed (loss: 0.0654:  66%|██████▌   | 144/219 [14:29<07:31,  6.02s/it]
Training Epoch: 14/100, batch 286/438 completed (loss: 0.0529:  66%|██████▌   | 144/219 [14:29<07:31,  6.02s/it]
Training Epoch: 14/100, batch 286/438 completed (loss: 0.0211:  66%|██████▌   | 144/219 [14:29<07:31,  6.02s/it]
Training Epoch: 14/100, batch 286/438 completed (loss: 0.0304:  66%|██████▌   | 144/219 [14:29<07:31,  6.02s/it]
Training Epoch: 14/100, batch 287/438 completed (loss: 0.0441:  66%|██████▌   | 144/219 [14:29<07:31,  6.02s/it]
Training Epoch: 14/100, batch 287/438 completed (loss: 0.0987:  66%|██████▌   | 144/219 [14:29<07:31,  6.02s/it]
Training Epoch: 14/100, batch 287/438 completed (loss: 0.0879:  66%|██████▌   | 144/219 [14:29<07:31,  6.02s/it]
Training Epoch: 14/100, batch 287/438 completed (loss: 0.0601:  66%|██████▌   | 144/219 [14:30<07:31,  6.02s/it]
Training Epoch: 14/100, batch 288/438 completed (loss: 0.0177:  66%|██████▌   | 144/219 [14:33<07:31,  6.02s/it]
Training Epoch: 14/100, batch 288/438 completed (loss: 0.1212:  66%|██████▌   | 144/219 [14:33<07:31,  6.02s/it]
Training Epoch: 14/100, batch 288/438 completed (loss: 0.0776:  66%|██████▌   | 144/219 [14:33<07:31,  6.02s/it]
Training Epoch: 14/100, batch 288/438 completed (loss: 0.1689:  66%|██████▌   | 144/219 [14:33<07:31,  6.02s/it]
Training Epoch: 14/100, batch 288/438 completed (loss: 0.1689:  66%|██████▌   | 145/219 [14:36<07:35,  6.15s/it]
Training Epoch: 14/100, batch 288/438 completed (loss: 0.1212:  66%|██████▌   | 145/219 [14:36<07:35,  6.16s/it]
Training Epoch: 14/100, batch 288/438 completed (loss: 0.0177:  66%|██████▌   | 145/219 [14:36<07:35,  6.16s/it]
Training Epoch: 14/100, batch 288/438 completed (loss: 0.0776:  66%|██████▌   | 145/219 [14:36<07:35,  6.16s/it]
Training Epoch: 14/100, batch 289/438 completed (loss: 0.2383:  66%|██████▌   | 145/219 [14:36<07:35,  6.16s/it]
Training Epoch: 14/100, batch 289/438 completed (loss: 0.0392:  66%|██████▌   | 145/219 [14:36<07:35,  6.16s/it]
Training Epoch: 14/100, batch 289/438 completed (loss: 0.0392:  66%|██████▌   | 145/219 [14:36<07:35,  6.16s/it]
Training Epoch: 14/100, batch 289/438 completed (loss: 0.2541:  66%|██████▌   | 145/219 [14:36<07:35,  6.15s/it]
Training Epoch: 14/100, batch 290/438 completed (loss: 0.0949:  66%|██████▌   | 145/219 [14:39<07:35,  6.16s/it]
Training Epoch: 14/100, batch 290/438 completed (loss: 0.1052:  66%|██████▌   | 145/219 [14:39<07:35,  6.16s/it]
Training Epoch: 14/100, batch 290/438 completed (loss: 0.0653:  66%|██████▌   | 145/219 [14:39<07:35,  6.16s/it]
Training Epoch: 14/100, batch 290/438 completed (loss: 0.2937:  66%|██████▌   | 145/219 [14:39<07:35,  6.15s/it]
Training Epoch: 14/100, batch 290/438 completed (loss: 0.2937:  67%|██████▋   | 146/219 [14:42<07:22,  6.07s/it]
Training Epoch: 14/100, batch 290/438 completed (loss: 0.1052:  67%|██████▋   | 146/219 [14:42<07:22,  6.07s/it]
Training Epoch: 14/100, batch 290/438 completed (loss: 0.0949:  67%|██████▋   | 146/219 [14:42<07:22,  6.07s/it]
Training Epoch: 14/100, batch 290/438 completed (loss: 0.0653:  67%|██████▋   | 146/219 [14:42<07:23,  6.07s/it]
Training Epoch: 14/100, batch 291/438 completed (loss: 0.5929:  67%|██████▋   | 146/219 [14:42<07:23,  6.07s/it]
Training Epoch: 14/100, batch 291/438 completed (loss: 0.051:  67%|██████▋   | 146/219 [14:42<07:22,  6.07s/it] 
Training Epoch: 14/100, batch 291/438 completed (loss: 0.1241:  67%|██████▋   | 146/219 [14:42<07:22,  6.07s/it]
Training Epoch: 14/100, batch 291/438 completed (loss: 0.0928:  67%|██████▋   | 146/219 [14:42<07:22,  6.07s/it]
Training Epoch: 14/100, batch 292/438 completed (loss: 0.0323:  67%|██████▋   | 146/219 [14:45<07:22,  6.07s/it]
Training Epoch: 14/100, batch 292/438 completed (loss: 0.051:  67%|██████▋   | 146/219 [14:45<07:23,  6.07s/it] 
Training Epoch: 14/100, batch 292/438 completed (loss: 0.053:  67%|██████▋   | 146/219 [14:45<07:22,  6.07s/it] 
Training Epoch: 14/100, batch 292/438 completed (loss: 0.0888:  67%|██████▋   | 146/219 [14:45<07:22,  6.07s/it]
Training Epoch: 14/100, batch 292/438 completed (loss: 0.0888:  67%|██████▋   | 147/219 [14:48<07:15,  6.05s/it]
Training Epoch: 14/100, batch 292/438 completed (loss: 0.0323:  67%|██████▋   | 147/219 [14:48<07:15,  6.05s/it]
Training Epoch: 14/100, batch 292/438 completed (loss: 0.053:  67%|██████▋   | 147/219 [14:48<07:15,  6.05s/it]
Training Epoch: 14/100, batch 292/438 completed (loss: 0.051:  67%|██████▋   | 147/219 [14:48<07:15,  6.05s/it]
Training Epoch: 14/100, batch 293/438 completed (loss: 0.1393:  67%|██████▋   | 147/219 [14:48<07:15,  6.05s/it]
Training Epoch: 14/100, batch 293/438 completed (loss: 0.1481:  67%|██████▋   | 147/219 [14:48<07:15,  6.05s/it]
Training Epoch: 14/100, batch 293/438 completed (loss: 0.3412:  67%|██████▋   | 147/219 [14:48<07:15,  6.05s/it]
Training Epoch: 14/100, batch 293/438 completed (loss: 0.0521:  67%|██████▋   | 147/219 [14:48<07:15,  6.05s/it]
Training Epoch: 14/100, batch 294/438 completed (loss: 0.0871:  67%|██████▋   | 147/219 [14:51<07:15,  6.05s/it]
Training Epoch: 14/100, batch 294/438 completed (loss: 0.0582:  67%|██████▋   | 147/219 [14:51<07:15,  6.05s/it]
Training Epoch: 14/100, batch 294/438 completed (loss: 0.0571:  67%|██████▋   | 147/219 [14:51<07:15,  6.05s/it]
Training Epoch: 14/100, batch 294/438 completed (loss: 0.233:  67%|██████▋   | 147/219 [14:51<07:15,  6.05s/it] 
Training Epoch: 14/100, batch 294/438 completed (loss: 0.0571:  68%|██████▊   | 148/219 [14:54<07:14,  6.12s/it]
Training Epoch: 14/100, batch 294/438 completed (loss: 0.233:  68%|██████▊   | 148/219 [14:54<07:14,  6.12s/it]
Training Epoch: 14/100, batch 294/438 completed (loss: 0.0871:  68%|██████▊   | 148/219 [14:54<07:14,  6.12s/it]
Training Epoch: 14/100, batch 294/438 completed (loss: 0.0582:  68%|██████▊   | 148/219 [14:54<07:14,  6.13s/it]
Training Epoch: 14/100, batch 295/438 completed (loss: 0.0739:  68%|██████▊   | 148/219 [14:54<07:14,  6.13s/it]
Training Epoch: 14/100, batch 295/438 completed (loss: 0.4388:  68%|██████▊   | 148/219 [14:54<07:14,  6.12s/it]
Training Epoch: 14/100, batch 295/438 completed (loss: 0.0046:  68%|██████▊   | 148/219 [14:54<07:14,  6.12s/it]
Training Epoch: 14/100, batch 295/438 completed (loss: 0.1116:  68%|██████▊   | 148/219 [14:54<07:14,  6.12s/it]
Training Epoch: 14/100, batch 296/438 completed (loss: 0.2334:  68%|██████▊   | 148/219 [14:57<07:14,  6.13s/it]
Training Epoch: 14/100, batch 296/438 completed (loss: 0.0302:  68%|██████▊   | 148/219 [14:57<07:14,  6.12s/it]
Training Epoch: 14/100, batch 296/438 completed (loss: 0.0212:  68%|██████▊   | 148/219 [14:57<07:14,  6.12s/it]
Training Epoch: 14/100, batch 296/438 completed (loss: 0.0363:  68%|██████▊   | 148/219 [14:57<07:14,  6.12s/it]
Training Epoch: 14/100, batch 296/438 completed (loss: 0.0212:  68%|██████▊   | 149/219 [15:00<07:05,  6.08s/it]
Training Epoch: 14/100, batch 296/438 completed (loss: 0.0302:  68%|██████▊   | 149/219 [15:00<07:05,  6.08s/it]
Training Epoch: 14/100, batch 296/438 completed (loss: 0.0363:  68%|██████▊   | 149/219 [15:00<07:05,  6.08s/it]
Training Epoch: 14/100, batch 296/438 completed (loss: 0.2334:  68%|██████▊   | 149/219 [15:00<07:05,  6.08s/it]
Training Epoch: 14/100, batch 297/438 completed (loss: 0.0893:  68%|██████▊   | 149/219 [15:00<07:05,  6.08s/it]
Training Epoch: 14/100, batch 297/438 completed (loss: 0.1272:  68%|██████▊   | 149/219 [15:00<07:05,  6.08s/it]
Training Epoch: 14/100, batch 297/438 completed (loss: 0.0432:  68%|██████▊   | 149/219 [15:00<07:05,  6.08s/it]
Training Epoch: 14/100, batch 297/438 completed (loss: 0.0684:  68%|██████▊   | 149/219 [15:00<07:05,  6.08s/it]
Training Epoch: 14/100, batch 298/438 completed (loss: 0.0205:  68%|██████▊   | 149/219 [15:03<07:05,  6.08s/it]
Training Epoch: 14/100, batch 298/438 completed (loss: 0.0164:  68%|██████▊   | 149/219 [15:03<07:05,  6.08s/it]
Training Epoch: 14/100, batch 298/438 completed (loss: 0.0327:  68%|██████▊   | 149/219 [15:03<07:05,  6.08s/it]
Training Epoch: 14/100, batch 298/438 completed (loss: 0.0202:  68%|██████▊   | 149/219 [15:03<07:05,  6.08s/it]
Training Epoch: 14/100, batch 298/438 completed (loss: 0.0202:  68%|██████▊   | 150/219 [15:06<07:01,  6.11s/it]
Training Epoch: 14/100, batch 298/438 completed (loss: 0.0164:  68%|██████▊   | 150/219 [15:06<07:01,  6.11s/it]
Training Epoch: 14/100, batch 298/438 completed (loss: 0.0327:  68%|██████▊   | 150/219 [15:06<07:01,  6.11s/it]
Training Epoch: 14/100, batch 298/438 completed (loss: 0.0205:  68%|██████▊   | 150/219 [15:06<07:01,  6.11s/it]
Training Epoch: 14/100, batch 299/438 completed (loss: 0.1411:  68%|██████▊   | 150/219 [15:06<07:01,  6.11s/it]
Training Epoch: 14/100, batch 299/438 completed (loss: 0.1897:  68%|██████▊   | 150/219 [15:06<07:01,  6.11s/it]
Training Epoch: 14/100, batch 299/438 completed (loss: 0.0166:  68%|██████▊   | 150/219 [15:06<07:01,  6.11s/it]
Training Epoch: 14/100, batch 299/438 completed (loss: 0.1289:  68%|██████▊   | 150/219 [15:06<07:01,  6.11s/it]
Training Epoch: 14/100, batch 300/438 completed (loss: 0.0236:  68%|██████▊   | 150/219 [15:09<07:01,  6.11s/it]
Training Epoch: 14/100, batch 300/438 completed (loss: 0.0043:  68%|██████▊   | 150/219 [15:09<07:01,  6.11s/it]
Training Epoch: 14/100, batch 300/438 completed (loss: 0.1616:  68%|██████▊   | 150/219 [15:09<07:01,  6.11s/it]
Training Epoch: 14/100, batch 300/438 completed (loss: 0.0364:  68%|██████▊   | 150/219 [15:09<07:01,  6.11s/it]
Training Epoch: 14/100, batch 300/438 completed (loss: 0.1616:  69%|██████▉   | 151/219 [15:12<06:52,  6.07s/it]
Training Epoch: 14/100, batch 300/438 completed (loss: 0.0043:  69%|██████▉   | 151/219 [15:12<06:52,  6.07s/it]
Training Epoch: 14/100, batch 300/438 completed (loss: 0.0364:  69%|██████▉   | 151/219 [15:12<06:52,  6.07s/it]
Training Epoch: 14/100, batch 300/438 completed (loss: 0.0236:  69%|██████▉   | 151/219 [15:12<06:52,  6.07s/it]
Training Epoch: 14/100, batch 301/438 completed (loss: 0.0662:  69%|██████▉   | 151/219 [15:12<06:52,  6.07s/it]
Training Epoch: 14/100, batch 301/438 completed (loss: 0.0221:  69%|██████▉   | 151/219 [15:12<06:52,  6.07s/it]
Training Epoch: 14/100, batch 301/438 completed (loss: 0.417:  69%|██████▉   | 151/219 [15:12<06:52,  6.07s/it] 
Training Epoch: 14/100, batch 301/438 completed (loss: 0.2452:  69%|██████▉   | 151/219 [15:12<06:52,  6.07s/it]
Training Epoch: 14/100, batch 302/438 completed (loss: 0.0906:  69%|██████▉   | 151/219 [15:15<06:52,  6.07s/it]
Training Epoch: 14/100, batch 302/438 completed (loss: 0.0251:  69%|██████▉   | 151/219 [15:15<06:52,  6.07s/it]
Training Epoch: 14/100, batch 302/438 completed (loss: 0.0371:  69%|██████▉   | 151/219 [15:15<06:52,  6.07s/it]
Training Epoch: 14/100, batch 302/438 completed (loss: 0.0608:  69%|██████▉   | 151/219 [15:15<06:52,  6.07s/it]
Training Epoch: 14/100, batch 302/438 completed (loss: 0.0608:  69%|██████▉   | 152/219 [15:18<06:45,  6.05s/it]
Training Epoch: 14/100, batch 302/438 completed (loss: 0.0251:  69%|██████▉   | 152/219 [15:18<06:45,  6.05s/it]
Training Epoch: 14/100, batch 302/438 completed (loss: 0.0906:  69%|██████▉   | 152/219 [15:18<06:45,  6.06s/it]
Training Epoch: 14/100, batch 302/438 completed (loss: 0.0371:  69%|██████▉   | 152/219 [15:18<06:45,  6.06s/it]
Training Epoch: 14/100, batch 303/438 completed (loss: 0.0792:  69%|██████▉   | 152/219 [15:18<06:45,  6.05s/it]
Training Epoch: 14/100, batch 303/438 completed (loss: 0.2175:  69%|██████▉   | 152/219 [15:18<06:45,  6.06s/it]
Training Epoch: 14/100, batch 303/438 completed (loss: 0.0868:  69%|██████▉   | 152/219 [15:18<06:45,  6.06s/it]
Training Epoch: 14/100, batch 303/438 completed (loss: 0.0654:  69%|██████▉   | 152/219 [15:18<06:45,  6.05s/it]
Training Epoch: 14/100, batch 304/438 completed (loss: 0.0855:  69%|██████▉   | 152/219 [15:21<06:45,  6.05s/it]
Training Epoch: 14/100, batch 304/438 completed (loss: 0.1052:  69%|██████▉   | 152/219 [15:21<06:45,  6.06s/it]
Training Epoch: 14/100, batch 304/438 completed (loss: 0.1839:  69%|██████▉   | 152/219 [15:21<06:45,  6.06s/it]
Training Epoch: 14/100, batch 304/438 completed (loss: 0.0948:  69%|██████▉   | 152/219 [15:21<06:45,  6.05s/it]
Training Epoch: 14/100, batch 304/438 completed (loss: 0.0855:  70%|██████▉   | 153/219 [15:25<06:43,  6.12s/it]
Training Epoch: 14/100, batch 304/438 completed (loss: 0.1839:  70%|██████▉   | 153/219 [15:25<06:43,  6.12s/it]
Training Epoch: 14/100, batch 304/438 completed (loss: 0.1052:  70%|██████▉   | 153/219 [15:25<06:43,  6.12s/it]
Training Epoch: 14/100, batch 304/438 completed (loss: 0.0948:  70%|██████▉   | 153/219 [15:25<06:44,  6.12s/it]
Training Epoch: 14/100, batch 305/438 completed (loss: 0.092:  70%|██████▉   | 153/219 [15:25<06:43,  6.12s/it] 
Training Epoch: 14/100, batch 305/438 completed (loss: 0.0613:  70%|██████▉   | 153/219 [15:25<06:43,  6.12s/it]
Training Epoch: 14/100, batch 305/438 completed (loss: 0.1293:  70%|██████▉   | 153/219 [15:25<06:43,  6.12s/it]
Training Epoch: 14/100, batch 305/438 completed (loss: 0.193:  70%|██████▉   | 153/219 [15:25<06:44,  6.12s/it] 
Training Epoch: 14/100, batch 306/438 completed (loss: 0.0088:  70%|██████▉   | 153/219 [15:27<06:43,  6.12s/it]
Training Epoch: 14/100, batch 306/438 completed (loss: 0.0107:  70%|██████▉   | 153/219 [15:27<06:43,  6.12s/it]
Training Epoch: 14/100, batch 306/438 completed (loss: 0.0819:  70%|██████▉   | 153/219 [15:27<06:43,  6.12s/it]
Training Epoch: 14/100, batch 306/438 completed (loss: 0.0723:  70%|██████▉   | 153/219 [15:27<06:44,  6.12s/it]
Training Epoch: 14/100, batch 306/438 completed (loss: 0.0819:  70%|███████   | 154/219 [15:30<06:32,  6.03s/it]
Training Epoch: 14/100, batch 306/438 completed (loss: 0.0723:  70%|███████   | 154/219 [15:30<06:32,  6.03s/it]
Training Epoch: 14/100, batch 306/438 completed (loss: 0.0088:  70%|███████   | 154/219 [15:30<06:32,  6.03s/it]
Training Epoch: 14/100, batch 306/438 completed (loss: 0.0107:  70%|███████   | 154/219 [15:30<06:32,  6.04s/it]
Training Epoch: 14/100, batch 307/438 completed (loss: 0.1002:  70%|███████   | 154/219 [15:30<06:32,  6.03s/it]
Training Epoch: 14/100, batch 307/438 completed (loss: 0.0218:  70%|███████   | 154/219 [15:30<06:32,  6.04s/it]
Training Epoch: 14/100, batch 307/438 completed (loss: 0.0239:  70%|███████   | 154/219 [15:30<06:32,  6.03s/it]
Training Epoch: 14/100, batch 307/438 completed (loss: 0.0187:  70%|███████   | 154/219 [15:30<06:32,  6.03s/it]
Training Epoch: 14/100, batch 308/438 completed (loss: 0.0274:  70%|███████   | 154/219 [15:33<06:32,  6.03s/it]
Training Epoch: 14/100, batch 308/438 completed (loss: 0.0328:  70%|███████   | 154/219 [15:33<06:32,  6.03s/it]
Training Epoch: 14/100, batch 308/438 completed (loss: 0.4074:  70%|███████   | 154/219 [15:33<06:32,  6.04s/it]
Training Epoch: 14/100, batch 308/438 completed (loss: 0.0764:  70%|███████   | 154/219 [15:33<06:32,  6.03s/it]
Training Epoch: 14/100, batch 308/438 completed (loss: 0.0764:  71%|███████   | 155/219 [15:37<06:31,  6.11s/it]
Training Epoch: 14/100, batch 308/438 completed (loss: 0.4074:  71%|███████   | 155/219 [15:37<06:31,  6.11s/it]
Training Epoch: 14/100, batch 308/438 completed (loss: 0.0328:  71%|███████   | 155/219 [15:37<06:31,  6.11s/it]
Training Epoch: 14/100, batch 308/438 completed (loss: 0.0274:  71%|███████   | 155/219 [15:37<06:31,  6.12s/it]
Training Epoch: 14/100, batch 309/438 completed (loss: 0.0133:  71%|███████   | 155/219 [15:37<06:31,  6.11s/it]
Training Epoch: 14/100, batch 309/438 completed (loss: 0.0423:  71%|███████   | 155/219 [15:37<06:31,  6.12s/it]
Training Epoch: 14/100, batch 309/438 completed (loss: 0.0754:  71%|███████   | 155/219 [15:37<06:31,  6.11s/it]
Training Epoch: 14/100, batch 309/438 completed (loss: 0.0727:  71%|███████   | 155/219 [15:37<06:31,  6.11s/it]
Training Epoch: 14/100, batch 310/438 completed (loss: 0.0403:  71%|███████   | 155/219 [15:40<06:31,  6.12s/it]
Training Epoch: 14/100, batch 310/438 completed (loss: 0.0337:  71%|███████   | 155/219 [15:40<06:31,  6.11s/it]
Training Epoch: 14/100, batch 310/438 completed (loss: 0.0128:  71%|███████   | 155/219 [15:40<06:31,  6.11s/it]
Training Epoch: 14/100, batch 310/438 completed (loss: 0.0741:  71%|███████   | 155/219 [15:40<06:31,  6.11s/it]
Training Epoch: 14/100, batch 310/438 completed (loss: 0.0128:  71%|███████   | 156/219 [15:43<06:22,  6.07s/it]
Training Epoch: 14/100, batch 310/438 completed (loss: 0.0403:  71%|███████   | 156/219 [15:43<06:22,  6.07s/it]
Training Epoch: 14/100, batch 310/438 completed (loss: 0.0337:  71%|███████   | 156/219 [15:43<06:22,  6.07s/it]
Training Epoch: 14/100, batch 310/438 completed (loss: 0.0741:  71%|███████   | 156/219 [15:43<06:22,  6.08s/it]
Training Epoch: 14/100, batch 311/438 completed (loss: 0.038:  71%|███████   | 156/219 [15:43<06:22,  6.07s/it] 
Training Epoch: 14/100, batch 311/438 completed (loss: 0.086:  71%|███████   | 156/219 [15:43<06:22,  6.07s/it] 
Training Epoch: 14/100, batch 311/438 completed (loss: 0.2122:  71%|███████   | 156/219 [15:43<06:22,  6.07s/it]
Training Epoch: 14/100, batch 311/438 completed (loss: 0.1009:  71%|███████   | 156/219 [15:43<06:22,  6.08s/it]
Training Epoch: 14/100, batch 312/438 completed (loss: 0.3113:  71%|███████   | 156/219 [15:46<06:22,  6.07s/it]
Training Epoch: 14/100, batch 312/438 completed (loss: 0.2222:  71%|███████   | 156/219 [15:46<06:22,  6.07s/it]
Training Epoch: 14/100, batch 312/438 completed (loss: 0.2621:  71%|███████   | 156/219 [15:46<06:22,  6.07s/it]
Training Epoch: 14/100, batch 312/438 completed (loss: 0.0428:  71%|███████   | 156/219 [15:46<06:22,  6.08s/it]
Training Epoch: 14/100, batch 312/438 completed (loss: 0.2621:  72%|███████▏  | 157/219 [15:49<06:14,  6.04s/it]
Training Epoch: 14/100, batch 312/438 completed (loss: 0.3113:  72%|███████▏  | 157/219 [15:49<06:14,  6.04s/it]
Training Epoch: 14/100, batch 312/438 completed (loss: 0.2222:  72%|███████▏  | 157/219 [15:49<06:14,  6.04s/it]
Training Epoch: 14/100, batch 312/438 completed (loss: 0.0428:  72%|███████▏  | 157/219 [15:49<06:14,  6.04s/it]
Training Epoch: 14/100, batch 313/438 completed (loss: 0.3398:  72%|███████▏  | 157/219 [15:49<06:14,  6.04s/it]
Training Epoch: 14/100, batch 313/438 completed (loss: 0.0177:  72%|███████▏  | 157/219 [15:49<06:14,  6.04s/it]
Training Epoch: 14/100, batch 313/438 completed (loss: 0.0472:  72%|███████▏  | 157/219 [15:49<06:14,  6.04s/it]
Training Epoch: 14/100, batch 313/438 completed (loss: 0.0226:  72%|███████▏  | 157/219 [15:49<06:14,  6.04s/it]
Training Epoch: 14/100, batch 314/438 completed (loss: 0.0679:  72%|███████▏  | 157/219 [15:52<06:14,  6.04s/it]
Training Epoch: 14/100, batch 314/438 completed (loss: 0.0603:  72%|███████▏  | 157/219 [15:52<06:14,  6.04s/it]
Training Epoch: 14/100, batch 314/438 completed (loss: 0.0071:  72%|███████▏  | 157/219 [15:52<06:14,  6.04s/it]
Training Epoch: 14/100, batch 314/438 completed (loss: 0.0018:  72%|███████▏  | 157/219 [15:52<06:14,  6.04s/it]
Training Epoch: 14/100, batch 314/438 completed (loss: 0.0603:  72%|███████▏  | 158/219 [15:55<06:08,  6.04s/it]
Training Epoch: 14/100, batch 314/438 completed (loss: 0.0679:  72%|███████▏  | 158/219 [15:55<06:08,  6.04s/it]
Training Epoch: 14/100, batch 314/438 completed (loss: 0.0071:  72%|███████▏  | 158/219 [15:55<06:08,  6.04s/it]
Training Epoch: 14/100, batch 314/438 completed (loss: 0.0018:  72%|███████▏  | 158/219 [15:55<06:08,  6.04s/it]
Training Epoch: 14/100, batch 315/438 completed (loss: 0.1122:  72%|███████▏  | 158/219 [15:55<06:08,  6.04s/it]
Training Epoch: 14/100, batch 315/438 completed (loss: 0.0966:  72%|███████▏  | 158/219 [15:55<06:08,  6.04s/it]
Training Epoch: 14/100, batch 315/438 completed (loss: 0.1445:  72%|███████▏  | 158/219 [15:55<06:08,  6.04s/it]
Training Epoch: 14/100, batch 315/438 completed (loss: 0.0531:  72%|███████▏  | 158/219 [15:55<06:08,  6.04s/it]
Training Epoch: 14/100, batch 316/438 completed (loss: 0.0448:  72%|███████▏  | 158/219 [15:58<06:08,  6.04s/it]
Training Epoch: 14/100, batch 316/438 completed (loss: 0.2236:  72%|███████▏  | 158/219 [15:58<06:08,  6.04s/it]
Training Epoch: 14/100, batch 316/438 completed (loss: 0.0682:  72%|███████▏  | 158/219 [15:58<06:08,  6.04s/it]
Training Epoch: 14/100, batch 316/438 completed (loss: 0.2656:  72%|███████▏  | 158/219 [15:58<06:08,  6.04s/it]
Training Epoch: 14/100, batch 316/438 completed (loss: 0.0448:  73%|███████▎  | 159/219 [16:01<06:01,  6.02s/it]
Training Epoch: 14/100, batch 316/438 completed (loss: 0.0682:  73%|███████▎  | 159/219 [16:01<06:01,  6.02s/it]
Training Epoch: 14/100, batch 316/438 completed (loss: 0.2236:  73%|███████▎  | 159/219 [16:01<06:01,  6.02s/it]
Training Epoch: 14/100, batch 316/438 completed (loss: 0.2656:  73%|███████▎  | 159/219 [16:01<06:01,  6.02s/it]
Training Epoch: 14/100, batch 317/438 completed (loss: 0.076:  73%|███████▎  | 159/219 [16:01<06:01,  6.02s/it] 
Training Epoch: 14/100, batch 317/438 completed (loss: 0.0732:  73%|███████▎  | 159/219 [16:01<06:01,  6.02s/it]
Training Epoch: 14/100, batch 317/438 completed (loss: 0.2843:  73%|███████▎  | 159/219 [16:01<06:01,  6.02s/it]
Training Epoch: 14/100, batch 317/438 completed (loss: 0.054:  73%|███████▎  | 159/219 [16:01<06:01,  6.02s/it] 
Training Epoch: 14/100, batch 318/438 completed (loss: 0.0116:  73%|███████▎  | 159/219 [16:04<06:01,  6.02s/it]
Training Epoch: 14/100, batch 318/438 completed (loss: 0.3025:  73%|███████▎  | 159/219 [16:04<06:01,  6.02s/it]
Training Epoch: 14/100, batch 318/438 completed (loss: 0.1908:  73%|███████▎  | 159/219 [16:04<06:01,  6.02s/it]
Training Epoch: 14/100, batch 318/438 completed (loss: 0.0572:  73%|███████▎  | 159/219 [16:04<06:01,  6.02s/it]
Training Epoch: 14/100, batch 318/438 completed (loss: 0.3025:  73%|███████▎  | 160/219 [16:07<05:54,  6.00s/it]
Training Epoch: 14/100, batch 318/438 completed (loss: 0.1908:  73%|███████▎  | 160/219 [16:07<05:54,  6.00s/it]
Training Epoch: 14/100, batch 318/438 completed (loss: 0.0116:  73%|███████▎  | 160/219 [16:07<05:54,  6.01s/it]
Training Epoch: 14/100, batch 318/438 completed (loss: 0.0572:  73%|███████▎  | 160/219 [16:07<05:54,  6.01s/it]
Training Epoch: 14/100, batch 319/438 completed (loss: 0.0454:  73%|███████▎  | 160/219 [16:07<05:54,  6.00s/it]
Training Epoch: 14/100, batch 319/438 completed (loss: 0.0071:  73%|███████▎  | 160/219 [16:07<05:54,  6.01s/it]
Training Epoch: 14/100, batch 319/438 completed (loss: 0.0529:  73%|███████▎  | 160/219 [16:07<05:54,  6.00s/it]
Training Epoch: 14/100, batch 319/438 completed (loss: 0.0515:  73%|███████▎  | 160/219 [16:07<05:54,  6.01s/it]
Training Epoch: 14/100, batch 320/438 completed (loss: 0.0236:  73%|███████▎  | 160/219 [16:10<05:54,  6.01s/it]
Training Epoch: 14/100, batch 320/438 completed (loss: 0.1258:  73%|███████▎  | 160/219 [16:10<05:54,  6.00s/it]
Training Epoch: 14/100, batch 320/438 completed (loss: 0.0041:  73%|███████▎  | 160/219 [16:10<05:54,  6.00s/it]
Training Epoch: 14/100, batch 320/438 completed (loss: 0.4721:  73%|███████▎  | 160/219 [16:10<05:54,  6.01s/it]
Training Epoch: 14/100, batch 320/438 completed (loss: 0.0041:  74%|███████▎  | 161/219 [16:13<05:51,  6.05s/it]
Training Epoch: 14/100, batch 320/438 completed (loss: 0.4721:  74%|███████▎  | 161/219 [16:13<05:51,  6.05s/it]
Training Epoch: 14/100, batch 320/438 completed (loss: 0.1258:  74%|███████▎  | 161/219 [16:13<05:51,  6.06s/it]
Training Epoch: 14/100, batch 320/438 completed (loss: 0.0236:  74%|███████▎  | 161/219 [16:13<05:51,  6.06s/it]
Training Epoch: 14/100, batch 321/438 completed (loss: 0.141:  74%|███████▎  | 161/219 [16:13<05:51,  6.05s/it] 
Training Epoch: 14/100, batch 321/438 completed (loss: 0.1096:  74%|███████▎  | 161/219 [16:13<05:51,  6.06s/it]
Training Epoch: 14/100, batch 321/438 completed (loss: 0.0404:  74%|███████▎  | 161/219 [16:13<05:51,  6.06s/it]
Training Epoch: 14/100, batch 321/438 completed (loss: 0.0115:  74%|███████▎  | 161/219 [16:13<05:51,  6.05s/it]
Training Epoch: 14/100, batch 322/438 completed (loss: 0.0909:  74%|███████▎  | 161/219 [16:16<05:51,  6.05s/it]
Training Epoch: 14/100, batch 322/438 completed (loss: 0.3212:  74%|███████▎  | 161/219 [16:16<05:51,  6.06s/it]
Training Epoch: 14/100, batch 322/438 completed (loss: 0.0311:  74%|███████▎  | 161/219 [16:16<05:51,  6.06s/it]
Training Epoch: 14/100, batch 322/438 completed (loss: 0.0509:  74%|███████▎  | 161/219 [16:16<05:51,  6.05s/it]
Training Epoch: 14/100, batch 322/438 completed (loss: 0.0311:  74%|███████▍  | 162/219 [16:19<05:44,  6.04s/it]
Training Epoch: 14/100, batch 322/438 completed (loss: 0.0909:  74%|███████▍  | 162/219 [16:19<05:44,  6.04s/it]
Training Epoch: 14/100, batch 322/438 completed (loss: 0.0509:  74%|███████▍  | 162/219 [16:19<05:44,  6.04s/it]
Training Epoch: 14/100, batch 322/438 completed (loss: 0.3212:  74%|███████▍  | 162/219 [16:19<05:44,  6.05s/it]
Training Epoch: 14/100, batch 323/438 completed (loss: 0.0486:  74%|███████▍  | 162/219 [16:19<05:44,  6.04s/it]
Training Epoch: 14/100, batch 323/438 completed (loss: 0.1548:  74%|███████▍  | 162/219 [16:19<05:44,  6.04s/it]
Training Epoch: 14/100, batch 323/438 completed (loss: 0.0657:  74%|███████▍  | 162/219 [16:19<05:44,  6.05s/it]
Training Epoch: 14/100, batch 323/438 completed (loss: 0.0589:  74%|███████▍  | 162/219 [16:19<05:44,  6.04s/it]
Training Epoch: 14/100, batch 324/438 completed (loss: 0.0558:  74%|███████▍  | 162/219 [16:22<05:44,  6.05s/it]
Training Epoch: 14/100, batch 324/438 completed (loss: 0.1178:  74%|███████▍  | 162/219 [16:22<05:44,  6.04s/it]
Training Epoch: 14/100, batch 324/438 completed (loss: 0.0953:  74%|███████▍  | 162/219 [16:22<05:44,  6.04s/it]
Training Epoch: 14/100, batch 324/438 completed (loss: 0.0092:  74%|███████▍  | 162/219 [16:22<05:44,  6.04s/it]
Training Epoch: 14/100, batch 324/438 completed (loss: 0.0092:  74%|███████▍  | 163/219 [16:25<05:37,  6.03s/it]
Training Epoch: 14/100, batch 324/438 completed (loss: 0.0558:  74%|███████▍  | 163/219 [16:25<05:38,  6.04s/it]
Training Epoch: 14/100, batch 324/438 completed (loss: 0.0953:  74%|███████▍  | 163/219 [16:25<05:38,  6.04s/it]
Training Epoch: 14/100, batch 324/438 completed (loss: 0.1178:  74%|███████▍  | 163/219 [16:25<05:38,  6.04s/it]
Training Epoch: 14/100, batch 325/438 completed (loss: 0.4414:  74%|███████▍  | 163/219 [16:25<05:38,  6.04s/it]
Training Epoch: 14/100, batch 325/438 completed (loss: 0.3601:  74%|███████▍  | 163/219 [16:25<05:38,  6.04s/it]
Training Epoch: 14/100, batch 325/438 completed (loss: 0.0432:  74%|███████▍  | 163/219 [16:25<05:38,  6.04s/it]
Training Epoch: 14/100, batch 325/438 completed (loss: 0.0237:  74%|███████▍  | 163/219 [16:25<05:37,  6.03s/it]
Training Epoch: 14/100, batch 326/438 completed (loss: 0.3331:  74%|███████▍  | 163/219 [16:28<05:38,  6.04s/it]
Training Epoch: 14/100, batch 326/438 completed (loss: 0.0164:  74%|███████▍  | 163/219 [16:28<05:38,  6.04s/it]
Training Epoch: 14/100, batch 326/438 completed (loss: 0.0541:  74%|███████▍  | 163/219 [16:28<05:38,  6.04s/it]
Training Epoch: 14/100, batch 326/438 completed (loss: 0.3731:  74%|███████▍  | 163/219 [16:28<05:37,  6.03s/it]
Training Epoch: 14/100, batch 326/438 completed (loss: 0.3731:  75%|███████▍  | 164/219 [16:31<05:34,  6.08s/it]
Training Epoch: 14/100, batch 326/438 completed (loss: 0.3331:  75%|███████▍  | 164/219 [16:31<05:34,  6.08s/it]
Training Epoch: 14/100, batch 326/438 completed (loss: 0.0541:  75%|███████▍  | 164/219 [16:31<05:34,  6.08s/it]
Training Epoch: 14/100, batch 326/438 completed (loss: 0.0164:  75%|███████▍  | 164/219 [16:31<05:34,  6.08s/it]
Training Epoch: 14/100, batch 327/438 completed (loss: 0.0441:  75%|███████▍  | 164/219 [16:31<05:34,  6.08s/it]
Training Epoch: 14/100, batch 327/438 completed (loss: 0.002:  75%|███████▍  | 164/219 [16:31<05:34,  6.08s/it] 
Training Epoch: 14/100, batch 327/438 completed (loss: 0.0091:  75%|███████▍  | 164/219 [16:31<05:34,  6.08s/it]
Training Epoch: 14/100, batch 327/438 completed (loss: 0.0269:  75%|███████▍  | 164/219 [16:31<05:34,  6.08s/it]
Training Epoch: 14/100, batch 328/438 completed (loss: 0.0173:  75%|███████▍  | 164/219 [16:34<05:34,  6.08s/it]
Training Epoch: 14/100, batch 328/438 completed (loss: 0.0223:  75%|███████▍  | 164/219 [16:34<05:34,  6.08s/it]
Training Epoch: 14/100, batch 328/438 completed (loss: 0.0091:  75%|███████▍  | 164/219 [16:34<05:34,  6.08s/it]
Training Epoch: 14/100, batch 328/438 completed (loss: 0.0394:  75%|███████▍  | 164/219 [16:34<05:34,  6.08s/it]
Training Epoch: 14/100, batch 328/438 completed (loss: 0.0091:  75%|███████▌  | 165/219 [16:37<05:27,  6.07s/it]
Training Epoch: 14/100, batch 328/438 completed (loss: 0.0223:  75%|███████▌  | 165/219 [16:37<05:27,  6.07s/it]
Training Epoch: 14/100, batch 328/438 completed (loss: 0.0394:  75%|███████▌  | 165/219 [16:37<05:27,  6.07s/it]
Training Epoch: 14/100, batch 328/438 completed (loss: 0.0173:  75%|███████▌  | 165/219 [16:37<05:27,  6.07s/it]
Training Epoch: 14/100, batch 329/438 completed (loss: 0.3494:  75%|███████▌  | 165/219 [16:37<05:27,  6.07s/it]
Training Epoch: 14/100, batch 329/438 completed (loss: 0.0327:  75%|███████▌  | 165/219 [16:37<05:27,  6.07s/it]
Training Epoch: 14/100, batch 329/438 completed (loss: 0.376:  75%|███████▌  | 165/219 [16:37<05:27,  6.07s/it] 
Training Epoch: 14/100, batch 329/438 completed (loss: 0.0299:  75%|███████▌  | 165/219 [16:37<05:27,  6.07s/it]
Training Epoch: 14/100, batch 330/438 completed (loss: 0.0294:  75%|███████▌  | 165/219 [16:40<05:27,  6.07s/it]
Training Epoch: 14/100, batch 330/438 completed (loss: 0.0275:  75%|███████▌  | 165/219 [16:40<05:27,  6.07s/it]
Training Epoch: 14/100, batch 330/438 completed (loss: 0.2861:  75%|███████▌  | 165/219 [16:40<05:27,  6.07s/it]
Training Epoch: 14/100, batch 330/438 completed (loss: 0.0393:  75%|███████▌  | 165/219 [16:40<05:27,  6.07s/it]
Training Epoch: 14/100, batch 330/438 completed (loss: 0.0275:  76%|███████▌  | 166/219 [16:43<05:25,  6.14s/it]
Training Epoch: 14/100, batch 330/438 completed (loss: 0.0294:  76%|███████▌  | 166/219 [16:43<05:25,  6.13s/it]
Training Epoch: 14/100, batch 330/438 completed (loss: 0.2861:  76%|███████▌  | 166/219 [16:43<05:25,  6.14s/it]
Training Epoch: 14/100, batch 330/438 completed (loss: 0.0393:  76%|███████▌  | 166/219 [16:43<05:25,  6.14s/it]
Training Epoch: 14/100, batch 331/438 completed (loss: 0.3867:  76%|███████▌  | 166/219 [16:43<05:25,  6.14s/it]
Training Epoch: 14/100, batch 331/438 completed (loss: 0.1085:  76%|███████▌  | 166/219 [16:43<05:25,  6.13s/it]
Training Epoch: 14/100, batch 331/438 completed (loss: 0.0153:  76%|███████▌  | 166/219 [16:43<05:25,  6.14s/it]
Training Epoch: 14/100, batch 331/438 completed (loss: 0.0845:  76%|███████▌  | 166/219 [16:43<05:25,  6.14s/it]
Training Epoch: 14/100, batch 332/438 completed (loss: 0.0312:  76%|███████▌  | 166/219 [16:46<05:25,  6.14s/it]
Training Epoch: 14/100, batch 332/438 completed (loss: 0.7075:  76%|███████▌  | 166/219 [16:46<05:25,  6.13s/it]
Training Epoch: 14/100, batch 332/438 completed (loss: 0.1141:  76%|███████▌  | 166/219 [16:46<05:25,  6.14s/it]
Training Epoch: 14/100, batch 332/438 completed (loss: 0.0797:  76%|███████▌  | 166/219 [16:46<05:25,  6.14s/it]
Training Epoch: 14/100, batch 332/438 completed (loss: 0.0312:  76%|███████▋  | 167/219 [16:49<05:16,  6.09s/it]
Training Epoch: 14/100, batch 332/438 completed (loss: 0.1141:  76%|███████▋  | 167/219 [16:49<05:16,  6.09s/it]
Training Epoch: 14/100, batch 332/438 completed (loss: 0.7075:  76%|███████▋  | 167/219 [16:49<05:16,  6.09s/it]
Training Epoch: 14/100, batch 332/438 completed (loss: 0.0797:  76%|███████▋  | 167/219 [16:49<05:16,  6.10s/it]
Training Epoch: 14/100, batch 333/438 completed (loss: 0.1464:  76%|███████▋  | 167/219 [16:49<05:16,  6.09s/it]
Training Epoch: 14/100, batch 333/438 completed (loss: 0.059:  76%|███████▋  | 167/219 [16:49<05:16,  6.09s/it] 
Training Epoch: 14/100, batch 333/438 completed (loss: 0.0886:  76%|███████▋  | 167/219 [16:49<05:16,  6.09s/it]
Training Epoch: 14/100, batch 333/438 completed (loss: 0.4159:  76%|███████▋  | 167/219 [16:49<05:16,  6.10s/it]
Training Epoch: 14/100, batch 334/438 completed (loss: 0.192:  76%|███████▋  | 167/219 [16:52<05:16,  6.09s/it] 
Training Epoch: 14/100, batch 334/438 completed (loss: 0.3726:  76%|███████▋  | 167/219 [16:52<05:16,  6.09s/it]
Training Epoch: 14/100, batch 334/438 completed (loss: 0.0341:  76%|███████▋  | 167/219 [16:52<05:16,  6.09s/it]
Training Epoch: 14/100, batch 334/438 completed (loss: 0.0238:  76%|███████▋  | 167/219 [16:52<05:16,  6.10s/it]
Training Epoch: 14/100, batch 334/438 completed (loss: 0.0341:  77%|███████▋  | 168/219 [16:55<05:06,  6.02s/it]
Training Epoch: 14/100, batch 334/438 completed (loss: 0.192:  77%|███████▋  | 168/219 [16:55<05:06,  6.02s/it]
Training Epoch: 14/100, batch 334/438 completed (loss: 0.0238:  77%|███████▋  | 168/219 [16:55<05:06,  6.02s/it]
Training Epoch: 14/100, batch 334/438 completed (loss: 0.3726:  77%|███████▋  | 168/219 [16:55<05:06,  6.02s/it]
Training Epoch: 14/100, batch 335/438 completed (loss: 0.0963:  77%|███████▋  | 168/219 [16:55<05:06,  6.02s/it]
Training Epoch: 14/100, batch 335/438 completed (loss: 0.007:  77%|███████▋  | 168/219 [16:55<05:06,  6.02s/it] 
Training Epoch: 14/100, batch 335/438 completed (loss: 0.3226:  77%|███████▋  | 168/219 [16:55<05:06,  6.02s/it]
Training Epoch: 14/100, batch 335/438 completed (loss: 0.0171:  77%|███████▋  | 168/219 [16:55<05:06,  6.02s/it]
Training Epoch: 14/100, batch 336/438 completed (loss: 0.032:  77%|███████▋  | 168/219 [16:59<05:06,  6.02s/it] 
Training Epoch: 14/100, batch 336/438 completed (loss: 0.3561:  77%|███████▋  | 168/219 [16:59<05:06,  6.02s/it]
Training Epoch: 14/100, batch 336/438 completed (loss: 0.075:  77%|███████▋  | 168/219 [16:59<05:06,  6.02s/it] 
Training Epoch: 14/100, batch 336/438 completed (loss: 0.2127:  77%|███████▋  | 168/219 [16:59<05:06,  6.02s/it]
Training Epoch: 14/100, batch 336/438 completed (loss: 0.2127:  77%|███████▋  | 169/219 [17:01<05:05,  6.10s/it]
Training Epoch: 14/100, batch 336/438 completed (loss: 0.032:  77%|███████▋  | 169/219 [17:01<05:05,  6.10s/it]
Training Epoch: 14/100, batch 336/438 completed (loss: 0.3561:  77%|███████▋  | 169/219 [17:01<05:05,  6.10s/it]
Training Epoch: 14/100, batch 336/438 completed (loss: 0.075:  77%|███████▋  | 169/219 [17:01<05:05,  6.11s/it]
Training Epoch: 14/100, batch 337/438 completed (loss: 0.0274:  77%|███████▋  | 169/219 [17:01<05:05,  6.10s/it]
Training Epoch: 14/100, batch 337/438 completed (loss: 0.0128:  77%|███████▋  | 169/219 [17:01<05:05,  6.11s/it]
Training Epoch: 14/100, batch 337/438 completed (loss: 0.0569:  77%|███████▋  | 169/219 [17:01<05:05,  6.10s/it]
Training Epoch: 14/100, batch 337/438 completed (loss: 0.0219:  77%|███████▋  | 169/219 [17:02<05:05,  6.10s/it]
Training Epoch: 14/100, batch 338/438 completed (loss: 0.0137:  77%|███████▋  | 169/219 [17:05<05:05,  6.10s/it]
Training Epoch: 14/100, batch 338/438 completed (loss: 0.0213:  77%|███████▋  | 169/219 [17:05<05:05,  6.10s/it]
Training Epoch: 14/100, batch 338/438 completed (loss: 0.0451:  77%|███████▋  | 169/219 [17:05<05:05,  6.11s/it]
Training Epoch: 14/100, batch 338/438 completed (loss: 0.149:  77%|███████▋  | 169/219 [17:05<05:05,  6.10s/it] 
Training Epoch: 14/100, batch 338/438 completed (loss: 0.0213:  78%|███████▊  | 170/219 [17:07<04:57,  6.08s/it]
Training Epoch: 14/100, batch 338/438 completed (loss: 0.149:  78%|███████▊  | 170/219 [17:07<04:57,  6.08s/it]
Training Epoch: 14/100, batch 338/438 completed (loss: 0.0137:  78%|███████▊  | 170/219 [17:07<04:57,  6.08s/it]
Training Epoch: 14/100, batch 338/438 completed (loss: 0.0451:  78%|███████▊  | 170/219 [17:07<04:58,  6.08s/it]
Training Epoch: 14/100, batch 339/438 completed (loss: 0.0093:  78%|███████▊  | 170/219 [17:07<04:57,  6.08s/it]
Training Epoch: 14/100, batch 339/438 completed (loss: 0.0115:  78%|███████▊  | 170/219 [17:07<04:58,  6.08s/it]
Training Epoch: 14/100, batch 339/438 completed (loss: 0.0535:  78%|███████▊  | 170/219 [17:07<04:57,  6.08s/it]
Training Epoch: 14/100, batch 339/438 completed (loss: 0.0445:  78%|███████▊  | 170/219 [17:08<04:57,  6.08s/it]
Training Epoch: 14/100, batch 340/438 completed (loss: 0.0873:  78%|███████▊  | 170/219 [17:10<04:57,  6.08s/it]
Training Epoch: 14/100, batch 340/438 completed (loss: 0.0078:  78%|███████▊  | 170/219 [17:10<04:58,  6.08s/it]
Training Epoch: 14/100, batch 340/438 completed (loss: 0.026:  78%|███████▊  | 170/219 [17:10<04:57,  6.08s/it] 
Training Epoch: 14/100, batch 340/438 completed (loss: 0.0052:  78%|███████▊  | 170/219 [17:10<04:57,  6.08s/it]
Training Epoch: 14/100, batch 340/438 completed (loss: 0.026:  78%|███████▊  | 171/219 [17:14<04:51,  6.08s/it]
Training Epoch: 14/100, batch 340/438 completed (loss: 0.0052:  78%|███████▊  | 171/219 [17:14<04:51,  6.08s/it]
Training Epoch: 14/100, batch 340/438 completed (loss: 0.0078:  78%|███████▊  | 171/219 [17:14<04:52,  6.08s/it]
Training Epoch: 14/100, batch 340/438 completed (loss: 0.0873:  78%|███████▊  | 171/219 [17:14<04:52,  6.08s/it]
Training Epoch: 14/100, batch 341/438 completed (loss: 0.0343:  78%|███████▊  | 171/219 [17:14<04:52,  6.08s/it]
Training Epoch: 14/100, batch 341/438 completed (loss: 0.0305:  78%|███████▊  | 171/219 [17:14<04:52,  6.08s/it]
Training Epoch: 14/100, batch 341/438 completed (loss: 0.1392:  78%|███████▊  | 171/219 [17:14<04:51,  6.08s/it]
Training Epoch: 14/100, batch 341/438 completed (loss: 0.0606:  78%|███████▊  | 171/219 [17:14<04:51,  6.08s/it]
Training Epoch: 14/100, batch 342/438 completed (loss: 0.0303:  78%|███████▊  | 171/219 [17:17<04:51,  6.08s/it]
Training Epoch: 14/100, batch 342/438 completed (loss: 0.3554:  78%|███████▊  | 171/219 [17:17<04:52,  6.08s/it]
Training Epoch: 14/100, batch 342/438 completed (loss: 0.0143:  78%|███████▊  | 171/219 [17:17<04:52,  6.08s/it]
Training Epoch: 14/100, batch 342/438 completed (loss: 0.0195:  78%|███████▊  | 171/219 [17:17<04:51,  6.08s/it]
Training Epoch: 14/100, batch 342/438 completed (loss: 0.0143:  79%|███████▊  | 172/219 [17:20<04:47,  6.11s/it]
Training Epoch: 14/100, batch 342/438 completed (loss: 0.3554:  79%|███████▊  | 172/219 [17:20<04:47,  6.11s/it]
Training Epoch: 14/100, batch 342/438 completed (loss: 0.0195:  79%|███████▊  | 172/219 [17:20<04:47,  6.11s/it]
Training Epoch: 14/100, batch 342/438 completed (loss: 0.0303:  79%|███████▊  | 172/219 [17:20<04:47,  6.12s/it]
Training Epoch: 14/100, batch 343/438 completed (loss: 0.0231:  79%|███████▊  | 172/219 [17:20<04:47,  6.12s/it]
Training Epoch: 14/100, batch 343/438 completed (loss: 0.0191:  79%|███████▊  | 172/219 [17:20<04:47,  6.11s/it]
Training Epoch: 14/100, batch 343/438 completed (loss: 0.1221:  79%|███████▊  | 172/219 [17:20<04:47,  6.11s/it]
Training Epoch: 14/100, batch 343/438 completed (loss: 0.0984:  79%|███████▊  | 172/219 [17:20<04:47,  6.11s/it]
Training Epoch: 14/100, batch 344/438 completed (loss: 0.0157:  79%|███████▊  | 172/219 [17:23<04:47,  6.11s/it]
Training Epoch: 14/100, batch 344/438 completed (loss: 0.1759:  79%|███████▊  | 172/219 [17:23<04:47,  6.12s/it]
Training Epoch: 14/100, batch 344/438 completed (loss: 0.1046:  79%|███████▊  | 172/219 [17:23<04:47,  6.11s/it]
Training Epoch: 14/100, batch 344/438 completed (loss: 0.0421:  79%|███████▊  | 172/219 [17:23<04:47,  6.11s/it]
Training Epoch: 14/100, batch 344/438 completed (loss: 0.1759:  79%|███████▉  | 173/219 [17:26<04:37,  6.03s/it]
Training Epoch: 14/100, batch 344/438 completed (loss: 0.0157:  79%|███████▉  | 173/219 [17:26<04:37,  6.03s/it]
Training Epoch: 14/100, batch 344/438 completed (loss: 0.1046:  79%|███████▉  | 173/219 [17:26<04:37,  6.04s/it]
Training Epoch: 14/100, batch 344/438 completed (loss: 0.0421:  79%|███████▉  | 173/219 [17:26<04:37,  6.04s/it]
Training Epoch: 14/100, batch 345/438 completed (loss: 0.3491:  79%|███████▉  | 173/219 [17:26<04:37,  6.03s/it]
Training Epoch: 14/100, batch 345/438 completed (loss: 0.0327:  79%|███████▉  | 173/219 [17:26<04:37,  6.04s/it]
Training Epoch: 14/100, batch 345/438 completed (loss: 0.0725:  79%|███████▉  | 173/219 [17:26<04:37,  6.03s/it]
Training Epoch: 14/100, batch 345/438 completed (loss: 0.0516:  79%|███████▉  | 173/219 [17:26<04:37,  6.04s/it]
Training Epoch: 14/100, batch 346/438 completed (loss: 0.0429:  79%|███████▉  | 173/219 [17:29<04:37,  6.03s/it]
Training Epoch: 14/100, batch 346/438 completed (loss: 0.0078:  79%|███████▉  | 173/219 [17:29<04:37,  6.03s/it]
Training Epoch: 14/100, batch 346/438 completed (loss: 0.018:  79%|███████▉  | 173/219 [17:29<04:37,  6.04s/it] 
Training Epoch: 14/100, batch 346/438 completed (loss: 0.1157:  79%|███████▉  | 173/219 [17:29<04:37,  6.04s/it]
Training Epoch: 14/100, batch 346/438 completed (loss: 0.0429:  79%|███████▉  | 174/219 [17:32<04:35,  6.12s/it]
Training Epoch: 14/100, batch 346/438 completed (loss: 0.0078:  79%|███████▉  | 174/219 [17:32<04:35,  6.13s/it]
Training Epoch: 14/100, batch 346/438 completed (loss: 0.1157:  79%|███████▉  | 174/219 [17:32<04:35,  6.13s/it]
Training Epoch: 14/100, batch 346/438 completed (loss: 0.018:  79%|███████▉  | 174/219 [17:32<04:35,  6.13s/it]
Training Epoch: 14/100, batch 347/438 completed (loss: 0.0659:  79%|███████▉  | 174/219 [17:32<04:35,  6.13s/it]
Training Epoch: 14/100, batch 347/438 completed (loss: 0.1685:  79%|███████▉  | 174/219 [17:32<04:35,  6.12s/it]
Training Epoch: 14/100, batch 347/438 completed (loss: 0.1294:  79%|███████▉  | 174/219 [17:32<04:35,  6.13s/it]
Training Epoch: 14/100, batch 347/438 completed (loss: 0.1246:  79%|███████▉  | 174/219 [17:32<04:35,  6.13s/it]
Training Epoch: 14/100, batch 348/438 completed (loss: 0.0168:  79%|███████▉  | 174/219 [17:35<04:35,  6.13s/it]
Training Epoch: 14/100, batch 348/438 completed (loss: 0.0181:  79%|███████▉  | 174/219 [17:35<04:35,  6.13s/it]
Training Epoch: 14/100, batch 348/438 completed (loss: 0.1994:  79%|███████▉  | 174/219 [17:35<04:35,  6.12s/it]
Training Epoch: 14/100, batch 348/438 completed (loss: 0.0262:  79%|███████▉  | 174/219 [17:35<04:35,  6.13s/it]
Training Epoch: 14/100, batch 348/438 completed (loss: 0.1994:  80%|███████▉  | 175/219 [17:38<04:27,  6.09s/it]
Training Epoch: 14/100, batch 348/438 completed (loss: 0.0181:  80%|███████▉  | 175/219 [17:38<04:27,  6.09s/it]
Training Epoch: 14/100, batch 348/438 completed (loss: 0.0168:  80%|███████▉  | 175/219 [17:38<04:28,  6.09s/it]
Training Epoch: 14/100, batch 348/438 completed (loss: 0.0262:  80%|███████▉  | 175/219 [17:38<04:28,  6.09s/it]
Training Epoch: 14/100, batch 349/438 completed (loss: 0.0437:  80%|███████▉  | 175/219 [17:38<04:28,  6.09s/it]
Training Epoch: 14/100, batch 349/438 completed (loss: 0.0097:  80%|███████▉  | 175/219 [17:38<04:27,  6.09s/it]
Training Epoch: 14/100, batch 349/438 completed (loss: 0.0354:  80%|███████▉  | 175/219 [17:38<04:27,  6.09s/it]
Training Epoch: 14/100, batch 349/438 completed (loss: 0.1074:  80%|███████▉  | 175/219 [17:38<04:28,  6.09s/it]
Training Epoch: 14/100, batch 350/438 completed (loss: 0.0645:  80%|███████▉  | 175/219 [17:41<04:27,  6.09s/it]
Training Epoch: 14/100, batch 350/438 completed (loss: 0.1255:  80%|███████▉  | 175/219 [17:41<04:27,  6.09s/it]
Training Epoch: 14/100, batch 350/438 completed (loss: 0.5756:  80%|███████▉  | 175/219 [17:41<04:28,  6.09s/it]
Training Epoch: 14/100, batch 350/438 completed (loss: 0.0178:  80%|███████▉  | 175/219 [17:41<04:28,  6.09s/it]
Training Epoch: 14/100, batch 350/438 completed (loss: 0.0645:  80%|████████  | 176/219 [17:44<04:20,  6.06s/it]
Training Epoch: 14/100, batch 350/438 completed (loss: 0.1255:  80%|████████  | 176/219 [17:44<04:20,  6.07s/it]
Training Epoch: 14/100, batch 350/438 completed (loss: 0.5756:  80%|████████  | 176/219 [17:44<04:20,  6.07s/it]
Training Epoch: 14/100, batch 350/438 completed (loss: 0.0178:  80%|████████  | 176/219 [17:44<04:20,  6.07s/it]
Training Epoch: 14/100, batch 351/438 completed (loss: 0.0342:  80%|████████  | 176/219 [17:44<04:20,  6.07s/it]
Training Epoch: 14/100, batch 351/438 completed (loss: 0.0604:  80%|████████  | 176/219 [17:44<04:20,  6.07s/it]
Training Epoch: 14/100, batch 351/438 completed (loss: 0.0499:  80%|████████  | 176/219 [17:44<04:20,  6.06s/it]
Training Epoch: 14/100, batch 351/438 completed (loss: 0.0505:  80%|████████  | 176/219 [17:44<04:20,  6.07s/it]
Training Epoch: 14/100, batch 352/438 completed (loss: 0.0377:  80%|████████  | 176/219 [17:47<04:20,  6.06s/it]
Training Epoch: 14/100, batch 352/438 completed (loss: 0.1836:  80%|████████  | 176/219 [17:47<04:20,  6.07s/it]
Training Epoch: 14/100, batch 352/438 completed (loss: 0.3777:  80%|████████  | 176/219 [17:47<04:20,  6.07s/it]
Training Epoch: 14/100, batch 352/438 completed (loss: 0.0376:  80%|████████  | 176/219 [17:47<04:20,  6.07s/it]
Training Epoch: 14/100, batch 352/438 completed (loss: 0.3777:  81%|████████  | 177/219 [17:50<04:13,  6.05s/it]
Training Epoch: 14/100, batch 352/438 completed (loss: 0.0377:  81%|████████  | 177/219 [17:50<04:13,  6.05s/it]
Training Epoch: 14/100, batch 352/438 completed (loss: 0.0376:  81%|████████  | 177/219 [17:50<04:14,  6.05s/it]
Training Epoch: 14/100, batch 352/438 completed (loss: 0.1836:  81%|████████  | 177/219 [17:50<04:14,  6.05s/it]
Training Epoch: 14/100, batch 353/438 completed (loss: 0.0607:  81%|████████  | 177/219 [17:50<04:13,  6.05s/it]
Training Epoch: 14/100, batch 353/438 completed (loss: 0.0091:  81%|████████  | 177/219 [17:50<04:14,  6.05s/it]
Training Epoch: 14/100, batch 353/438 completed (loss: 0.0802:  81%|████████  | 177/219 [17:50<04:13,  6.05s/it]
Training Epoch: 14/100, batch 353/438 completed (loss: 0.2903:  81%|████████  | 177/219 [17:50<04:14,  6.05s/it]
Training Epoch: 14/100, batch 354/438 completed (loss: 0.0172:  81%|████████  | 177/219 [17:53<04:14,  6.05s/it]
Training Epoch: 14/100, batch 354/438 completed (loss: 0.1408:  81%|████████  | 177/219 [17:53<04:13,  6.05s/it]
Training Epoch: 14/100, batch 354/438 completed (loss: 0.1072:  81%|████████  | 177/219 [17:53<04:13,  6.05s/it]
Training Epoch: 14/100, batch 354/438 completed (loss: 0.0352:  81%|████████  | 177/219 [17:53<04:14,  6.05s/it]
Training Epoch: 14/100, batch 354/438 completed (loss: 0.1072:  81%|████████▏ | 178/219 [17:56<04:07,  6.03s/it]
Training Epoch: 14/100, batch 354/438 completed (loss: 0.0172:  81%|████████▏ | 178/219 [17:56<04:07,  6.03s/it]
Training Epoch: 14/100, batch 354/438 completed (loss: 0.1408:  81%|████████▏ | 178/219 [17:56<04:07,  6.04s/it]
Training Epoch: 14/100, batch 354/438 completed (loss: 0.0352:  81%|████████▏ | 178/219 [17:56<04:07,  6.03s/it]
Training Epoch: 14/100, batch 355/438 completed (loss: 0.0171:  81%|████████▏ | 178/219 [17:56<04:07,  6.03s/it]
Training Epoch: 14/100, batch 355/438 completed (loss: 0.0157:  81%|████████▏ | 178/219 [17:56<04:07,  6.04s/it]
Training Epoch: 14/100, batch 355/438 completed (loss: 0.0376:  81%|████████▏ | 178/219 [17:56<04:07,  6.03s/it]
Training Epoch: 14/100, batch 355/438 completed (loss: 0.0383:  81%|████████▏ | 178/219 [17:56<04:07,  6.03s/it]
Training Epoch: 14/100, batch 356/438 completed (loss: 0.4717:  81%|████████▏ | 178/219 [17:59<04:07,  6.03s/it]
Training Epoch: 14/100, batch 356/438 completed (loss: 0.0629:  81%|████████▏ | 178/219 [17:59<04:07,  6.03s/it]
Training Epoch: 14/100, batch 356/438 completed (loss: 0.0876:  81%|████████▏ | 178/219 [17:59<04:07,  6.04s/it]
Training Epoch: 14/100, batch 356/438 completed (loss: 0.0361:  81%|████████▏ | 178/219 [17:59<04:07,  6.03s/it]
Training Epoch: 14/100, batch 356/438 completed (loss: 0.0876:  82%|████████▏ | 179/219 [18:02<04:04,  6.11s/it]
Training Epoch: 14/100, batch 356/438 completed (loss: 0.0629:  82%|████████▏ | 179/219 [18:02<04:04,  6.11s/it]
Training Epoch: 14/100, batch 356/438 completed (loss: 0.4717:  82%|████████▏ | 179/219 [18:02<04:04,  6.11s/it]
Training Epoch: 14/100, batch 356/438 completed (loss: 0.0361:  82%|████████▏ | 179/219 [18:02<04:04,  6.11s/it]
Training Epoch: 14/100, batch 357/438 completed (loss: 0.047:  82%|████████▏ | 179/219 [18:02<04:04,  6.11s/it] 
Training Epoch: 14/100, batch 357/438 completed (loss: 0.0044:  82%|████████▏ | 179/219 [18:02<04:04,  6.11s/it]
Training Epoch: 14/100, batch 357/438 completed (loss: 0.0136:  82%|████████▏ | 179/219 [18:02<04:04,  6.11s/it]
Training Epoch: 14/100, batch 357/438 completed (loss: 0.0373:  82%|████████▏ | 179/219 [18:02<04:04,  6.11s/it]
Training Epoch: 14/100, batch 358/438 completed (loss: 0.0046:  82%|████████▏ | 179/219 [18:05<04:04,  6.11s/it]
Training Epoch: 14/100, batch 358/438 completed (loss: 0.0192:  82%|████████▏ | 179/219 [18:05<04:04,  6.11s/it]
Training Epoch: 14/100, batch 358/438 completed (loss: 0.0806:  82%|████████▏ | 179/219 [18:05<04:04,  6.11s/it]
Training Epoch: 14/100, batch 358/438 completed (loss: 0.2058:  82%|████████▏ | 179/219 [18:05<04:04,  6.11s/it]
Training Epoch: 14/100, batch 358/438 completed (loss: 0.0192:  82%|████████▏ | 180/219 [18:08<03:57,  6.08s/it]
Training Epoch: 14/100, batch 358/438 completed (loss: 0.0046:  82%|████████▏ | 180/219 [18:08<03:57,  6.08s/it]
Training Epoch: 14/100, batch 358/438 completed (loss: 0.0806:  82%|████████▏ | 180/219 [18:08<03:57,  6.09s/it]
Training Epoch: 14/100, batch 358/438 completed (loss: 0.2058:  82%|████████▏ | 180/219 [18:08<03:57,  6.08s/it]
Training Epoch: 14/100, batch 359/438 completed (loss: 0.0518:  82%|████████▏ | 180/219 [18:08<03:57,  6.09s/it]
Training Epoch: 14/100, batch 359/438 completed (loss: 0.0409:  82%|████████▏ | 180/219 [18:08<03:57,  6.08s/it]
Training Epoch: 14/100, batch 359/438 completed (loss: 0.196:  82%|████████▏ | 180/219 [18:08<03:57,  6.08s/it] 
Training Epoch: 14/100, batch 359/438 completed (loss: 0.392:  82%|████████▏ | 180/219 [18:08<03:57,  6.08s/it] 
Training Epoch: 14/100, batch 360/438 completed (loss: 0.0843:  82%|████████▏ | 180/219 [18:11<03:57,  6.08s/it]
Training Epoch: 14/100, batch 360/438 completed (loss: 0.1169:  82%|████████▏ | 180/219 [18:11<03:57,  6.08s/it]
Training Epoch: 14/100, batch 360/438 completed (loss: 0.1198:  82%|████████▏ | 180/219 [18:11<03:57,  6.09s/it]
Training Epoch: 14/100, batch 360/438 completed (loss: 0.0948:  82%|████████▏ | 180/219 [18:11<03:57,  6.08s/it]
Training Epoch: 14/100, batch 360/438 completed (loss: 0.0843:  83%|████████▎ | 181/219 [18:14<03:51,  6.10s/it]
Training Epoch: 14/100, batch 360/438 completed (loss: 0.1198:  83%|████████▎ | 181/219 [18:14<03:51,  6.10s/it]
Training Epoch: 14/100, batch 360/438 completed (loss: 0.1169:  83%|████████▎ | 181/219 [18:14<03:51,  6.10s/it]
Training Epoch: 14/100, batch 360/438 completed (loss: 0.0948:  83%|████████▎ | 181/219 [18:14<03:51,  6.10s/it]
Training Epoch: 14/100, batch 361/438 completed (loss: 0.1656:  83%|████████▎ | 181/219 [18:14<03:51,  6.10s/it]
Training Epoch: 14/100, batch 361/438 completed (loss: 0.1264:  83%|████████▎ | 181/219 [18:14<03:51,  6.10s/it]
Training Epoch: 14/100, batch 361/438 completed (loss: 0.0296:  83%|████████▎ | 181/219 [18:14<03:51,  6.10s/it]
Training Epoch: 14/100, batch 361/438 completed (loss: 0.2975:  83%|████████▎ | 181/219 [18:14<03:51,  6.10s/it]
Training Epoch: 14/100, batch 362/438 completed (loss: 0.0674:  83%|████████▎ | 181/219 [18:17<03:51,  6.10s/it]
Training Epoch: 14/100, batch 362/438 completed (loss: 0.2145:  83%|████████▎ | 181/219 [18:17<03:51,  6.10s/it]
Training Epoch: 14/100, batch 362/438 completed (loss: 0.1856:  83%|████████▎ | 181/219 [18:17<03:51,  6.10s/it]
Training Epoch: 14/100, batch 362/438 completed (loss: 0.2772:  83%|████████▎ | 181/219 [18:17<03:51,  6.10s/it]
Training Epoch: 14/100, batch 362/438 completed (loss: 0.0674:  83%|████████▎ | 182/219 [18:20<03:42,  6.02s/it]
Training Epoch: 14/100, batch 362/438 completed (loss: 0.1856:  83%|████████▎ | 182/219 [18:20<03:42,  6.02s/it]
Training Epoch: 14/100, batch 362/438 completed (loss: 0.2145:  83%|████████▎ | 182/219 [18:20<03:42,  6.03s/it]
Training Epoch: 14/100, batch 362/438 completed (loss: 0.2772:  83%|████████▎ | 182/219 [18:20<03:42,  6.03s/it]
Training Epoch: 14/100, batch 363/438 completed (loss: 0.0045:  83%|████████▎ | 182/219 [18:20<03:42,  6.02s/it]
Training Epoch: 14/100, batch 363/438 completed (loss: 0.0171:  83%|████████▎ | 182/219 [18:20<03:42,  6.03s/it]
Training Epoch: 14/100, batch 363/438 completed (loss: 0.0293:  83%|████████▎ | 182/219 [18:20<03:42,  6.02s/it]
Training Epoch: 14/100, batch 363/438 completed (loss: 0.232:  83%|████████▎ | 182/219 [18:20<03:42,  6.03s/it] 
Training Epoch: 14/100, batch 364/438 completed (loss: 0.053:  83%|████████▎ | 182/219 [18:24<03:42,  6.02s/it] 
Training Epoch: 14/100, batch 364/438 completed (loss: 0.0425:  83%|████████▎ | 182/219 [18:24<03:42,  6.02s/it]
Training Epoch: 14/100, batch 364/438 completed (loss: 0.023:  83%|████████▎ | 182/219 [18:24<03:42,  6.03s/it] 
Training Epoch: 14/100, batch 364/438 completed (loss: 0.0193:  83%|████████▎ | 182/219 [18:24<03:42,  6.03s/it]
Training Epoch: 14/100, batch 364/438 completed (loss: 0.023:  84%|████████▎ | 183/219 [18:26<03:38,  6.07s/it]
Training Epoch: 14/100, batch 364/438 completed (loss: 0.0425:  84%|████████▎ | 183/219 [18:26<03:38,  6.07s/it]
Training Epoch: 14/100, batch 364/438 completed (loss: 0.053:  84%|████████▎ | 183/219 [18:26<03:38,  6.07s/it]
Training Epoch: 14/100, batch 364/438 completed (loss: 0.0193:  84%|████████▎ | 183/219 [18:26<03:38,  6.07s/it]
Training Epoch: 14/100, batch 365/438 completed (loss: 0.1174:  84%|████████▎ | 183/219 [18:26<03:38,  6.07s/it]
Training Epoch: 14/100, batch 365/438 completed (loss: 0.1392:  84%|████████▎ | 183/219 [18:26<03:38,  6.07s/it]
Training Epoch: 14/100, batch 365/438 completed (loss: 0.1533:  84%|████████▎ | 183/219 [18:26<03:38,  6.07s/it]
Training Epoch: 14/100, batch 365/438 completed (loss: 0.0201:  84%|████████▎ | 183/219 [18:26<03:38,  6.07s/it]
Training Epoch: 14/100, batch 366/438 completed (loss: 0.148:  84%|████████▎ | 183/219 [18:30<03:38,  6.07s/it] 
Training Epoch: 14/100, batch 366/438 completed (loss: 0.171:  84%|████████▎ | 183/219 [18:30<03:38,  6.07s/it] 
Training Epoch: 14/100, batch 366/438 completed (loss: 0.1354:  84%|████████▎ | 183/219 [18:30<03:38,  6.07s/it]
Training Epoch: 14/100, batch 366/438 completed (loss: 0.045:  84%|████████▎ | 183/219 [18:30<03:38,  6.07s/it] 
Training Epoch: 14/100, batch 366/438 completed (loss: 0.148:  84%|████████▍ | 184/219 [18:33<03:33,  6.09s/it]
Training Epoch: 14/100, batch 366/438 completed (loss: 0.171:  84%|████████▍ | 184/219 [18:33<03:33,  6.09s/it]
Training Epoch: 14/100, batch 366/438 completed (loss: 0.1354:  84%|████████▍ | 184/219 [18:33<03:33,  6.09s/it]
Training Epoch: 14/100, batch 366/438 completed (loss: 0.045:  84%|████████▍ | 184/219 [18:33<03:33,  6.09s/it]
Training Epoch: 14/100, batch 367/438 completed (loss: 0.0763:  84%|████████▍ | 184/219 [18:33<03:33,  6.09s/it]
Training Epoch: 14/100, batch 367/438 completed (loss: 0.0298:  84%|████████▍ | 184/219 [18:33<03:33,  6.09s/it]
Training Epoch: 14/100, batch 367/438 completed (loss: 0.04:  84%|████████▍ | 184/219 [18:33<03:33,  6.09s/it] 
Training Epoch: 14/100, batch 367/438 completed (loss: 0.0226:  84%|████████▍ | 184/219 [18:33<03:33,  6.09s/it]
Training Epoch: 14/100, batch 368/438 completed (loss: 0.0643:  84%|████████▍ | 184/219 [18:36<03:33,  6.09s/it]
Training Epoch: 14/100, batch 368/438 completed (loss: 0.0507:  84%|████████▍ | 184/219 [18:36<03:33,  6.09s/it]
Training Epoch: 14/100, batch 368/438 completed (loss: 0.033:  84%|████████▍ | 184/219 [18:36<03:33,  6.09s/it]
Training Epoch: 14/100, batch 368/438 completed (loss: 0.3364:  84%|████████▍ | 184/219 [18:36<03:33,  6.09s/it]
Training Epoch: 14/100, batch 368/438 completed (loss: 0.0507:  84%|████████▍ | 185/219 [18:39<03:27,  6.11s/it]
Training Epoch: 14/100, batch 368/438 completed (loss: 0.3364:  84%|████████▍ | 185/219 [18:39<03:27,  6.11s/it]
Training Epoch: 14/100, batch 368/438 completed (loss: 0.0643:  84%|████████▍ | 185/219 [18:39<03:28,  6.12s/it]
Training Epoch: 14/100, batch 368/438 completed (loss: 0.033:  84%|████████▍ | 185/219 [18:39<03:28,  6.12s/it]
Training Epoch: 14/100, batch 369/438 completed (loss: 0.2471:  84%|████████▍ | 185/219 [18:39<03:28,  6.12s/it]
Training Epoch: 14/100, batch 369/438 completed (loss: 0.0822:  84%|████████▍ | 185/219 [18:39<03:28,  6.12s/it]
Training Epoch: 14/100, batch 369/438 completed (loss: 0.0664:  84%|████████▍ | 185/219 [18:39<03:27,  6.11s/it]
Training Epoch: 14/100, batch 369/438 completed (loss: 0.1523:  84%|████████▍ | 185/219 [18:39<03:27,  6.11s/it]
Training Epoch: 14/100, batch 370/438 completed (loss: 0.0125:  84%|████████▍ | 185/219 [18:42<03:27,  6.11s/it]
Training Epoch: 14/100, batch 370/438 completed (loss: 0.0886:  84%|████████▍ | 185/219 [18:42<03:28,  6.12s/it]
Training Epoch: 14/100, batch 370/438 completed (loss: 0.1988:  84%|████████▍ | 185/219 [18:42<03:28,  6.12s/it]
Training Epoch: 14/100, batch 370/438 completed (loss: 0.0899:  84%|████████▍ | 185/219 [18:42<03:27,  6.11s/it]
Training Epoch: 14/100, batch 370/438 completed (loss: 0.0886:  85%|████████▍ | 186/219 [18:45<03:20,  6.08s/it]
Training Epoch: 14/100, batch 370/438 completed (loss: 0.0899:  85%|████████▍ | 186/219 [18:45<03:20,  6.08s/it]
Training Epoch: 14/100, batch 370/438 completed (loss: 0.0125:  85%|████████▍ | 186/219 [18:45<03:20,  6.09s/it]
Training Epoch: 14/100, batch 370/438 completed (loss: 0.1988:  85%|████████▍ | 186/219 [18:45<03:20,  6.09s/it]
Training Epoch: 14/100, batch 371/438 completed (loss: 0.0775:  85%|████████▍ | 186/219 [18:45<03:20,  6.08s/it]
Training Epoch: 14/100, batch 371/438 completed (loss: 0.0325:  85%|████████▍ | 186/219 [18:45<03:20,  6.09s/it]
Training Epoch: 14/100, batch 371/438 completed (loss: 0.0478:  85%|████████▍ | 186/219 [18:45<03:20,  6.09s/it]
Training Epoch: 14/100, batch 371/438 completed (loss: 0.0375:  85%|████████▍ | 186/219 [18:45<03:20,  6.08s/it]
Training Epoch: 14/100, batch 372/438 completed (loss: 0.081:  85%|████████▍ | 186/219 [18:48<03:20,  6.09s/it] 
Training Epoch: 14/100, batch 372/438 completed (loss: 0.1641:  85%|████████▍ | 186/219 [18:48<03:20,  6.09s/it]
Training Epoch: 14/100, batch 372/438 completed (loss: 0.1203:  85%|████████▍ | 186/219 [18:48<03:20,  6.08s/it]
Training Epoch: 14/100, batch 372/438 completed (loss: 0.0247:  85%|████████▍ | 186/219 [18:48<03:20,  6.08s/it]
Training Epoch: 14/100, batch 372/438 completed (loss: 0.0247:  85%|████████▌ | 187/219 [18:51<03:13,  6.06s/it]
Training Epoch: 14/100, batch 372/438 completed (loss: 0.1641:  85%|████████▌ | 187/219 [18:51<03:14,  6.06s/it]
Training Epoch: 14/100, batch 372/438 completed (loss: 0.1203:  85%|████████▌ | 187/219 [18:51<03:14,  6.07s/it]
Training Epoch: 14/100, batch 372/438 completed (loss: 0.081:  85%|████████▌ | 187/219 [18:51<03:14,  6.07s/it]
Training Epoch: 14/100, batch 373/438 completed (loss: 0.0393:  85%|████████▌ | 187/219 [18:51<03:14,  6.07s/it]
Training Epoch: 14/100, batch 373/438 completed (loss: 0.3026:  85%|████████▌ | 187/219 [18:51<03:14,  6.07s/it]
Training Epoch: 14/100, batch 373/438 completed (loss: 0.3866:  85%|████████▌ | 187/219 [18:51<03:14,  6.06s/it]
Training Epoch: 14/100, batch 373/438 completed (loss: 0.2008:  85%|████████▌ | 187/219 [18:51<03:13,  6.06s/it]
Training Epoch: 14/100, batch 374/438 completed (loss: 0.165:  85%|████████▌ | 187/219 [18:54<03:14,  6.07s/it] 
Training Epoch: 14/100, batch 374/438 completed (loss: 0.1347:  85%|████████▌ | 187/219 [18:54<03:14,  6.07s/it]
Training Epoch: 14/100, batch 374/438 completed (loss: 0.1053:  85%|████████▌ | 187/219 [18:54<03:14,  6.06s/it]
Training Epoch: 14/100, batch 374/438 completed (loss: 0.006:  85%|████████▌ | 187/219 [18:54<03:13,  6.06s/it] 
Training Epoch: 14/100, batch 374/438 completed (loss: 0.1347:  86%|████████▌ | 188/219 [18:57<03:08,  6.10s/it]
Training Epoch: 14/100, batch 374/438 completed (loss: 0.165:  86%|████████▌ | 188/219 [18:57<03:08,  6.09s/it]
Training Epoch: 14/100, batch 374/438 completed (loss: 0.1053:  86%|████████▌ | 188/219 [18:57<03:09,  6.10s/it]
Training Epoch: 14/100, batch 374/438 completed (loss: 0.006:  86%|████████▌ | 188/219 [18:57<03:09,  6.10s/it]
Training Epoch: 14/100, batch 375/438 completed (loss: 0.0771:  86%|████████▌ | 188/219 [18:57<03:09,  6.10s/it]
Training Epoch: 14/100, batch 375/438 completed (loss: 0.036:  86%|████████▌ | 188/219 [18:57<03:08,  6.09s/it]
Training Epoch: 14/100, batch 375/438 completed (loss: 0.0245:  86%|████████▌ | 188/219 [18:57<03:08,  6.10s/it]
Training Epoch: 14/100, batch 375/438 completed (loss: 0.1464:  86%|████████▌ | 188/219 [18:57<03:09,  6.10s/it]
Training Epoch: 14/100, batch 376/438 completed (loss: 0.0179:  86%|████████▌ | 188/219 [19:00<03:08,  6.09s/it]
Training Epoch: 14/100, batch 376/438 completed (loss: 0.2184:  86%|████████▌ | 188/219 [19:00<03:08,  6.10s/it]
Training Epoch: 14/100, batch 376/438 completed (loss: 0.6186:  86%|████████▌ | 188/219 [19:00<03:09,  6.10s/it]
Training Epoch: 14/100, batch 376/438 completed (loss: 0.1174:  86%|████████▌ | 188/219 [19:00<03:09,  6.10s/it]
Training Epoch: 14/100, batch 376/438 completed (loss: 0.6186:  86%|████████▋ | 189/219 [19:03<03:03,  6.10s/it]
Training Epoch: 14/100, batch 376/438 completed (loss: 0.2184:  86%|████████▋ | 189/219 [19:03<03:03,  6.11s/it]
Training Epoch: 14/100, batch 376/438 completed (loss: 0.1174:  86%|████████▋ | 189/219 [19:03<03:03,  6.10s/it]
Training Epoch: 14/100, batch 376/438 completed (loss: 0.0179:  86%|████████▋ | 189/219 [19:03<03:03,  6.11s/it]
Training Epoch: 14/100, batch 377/438 completed (loss: 0.0211:  86%|████████▋ | 189/219 [19:03<03:03,  6.11s/it]
Training Epoch: 14/100, batch 377/438 completed (loss: 0.042:  86%|████████▋ | 189/219 [19:03<03:03,  6.11s/it] 
Training Epoch: 14/100, batch 377/438 completed (loss: 0.1014:  86%|████████▋ | 189/219 [19:03<03:03,  6.10s/it]
Training Epoch: 14/100, batch 377/438 completed (loss: 0.3445:  86%|████████▋ | 189/219 [19:03<03:03,  6.10s/it]
Training Epoch: 14/100, batch 378/438 completed (loss: 0.0751:  86%|████████▋ | 189/219 [19:06<03:03,  6.11s/it]
Training Epoch: 14/100, batch 378/438 completed (loss: 0.0254:  86%|████████▋ | 189/219 [19:06<03:03,  6.10s/it]
Training Epoch: 14/100, batch 378/438 completed (loss: 0.3069:  86%|████████▋ | 189/219 [19:06<03:03,  6.11s/it]
Training Epoch: 14/100, batch 378/438 completed (loss: 0.0344:  86%|████████▋ | 189/219 [19:06<03:03,  6.10s/it]
Training Epoch: 14/100, batch 378/438 completed (loss: 0.0344:  87%|████████▋ | 190/219 [19:09<02:57,  6.13s/it]
Training Epoch: 14/100, batch 378/438 completed (loss: 0.0254:  87%|████████▋ | 190/219 [19:09<02:57,  6.13s/it]
Training Epoch: 14/100, batch 378/438 completed (loss: 0.0751:  87%|████████▋ | 190/219 [19:09<02:57,  6.13s/it]
Training Epoch: 14/100, batch 378/438 completed (loss: 0.3069:  87%|████████▋ | 190/219 [19:09<02:57,  6.13s/it]
Training Epoch: 14/100, batch 379/438 completed (loss: 0.0917:  87%|████████▋ | 190/219 [19:09<02:57,  6.13s/it]
Training Epoch: 14/100, batch 379/438 completed (loss: 0.0434:  87%|████████▋ | 190/219 [19:09<02:57,  6.13s/it]
Training Epoch: 14/100, batch 379/438 completed (loss: 0.0498:  87%|████████▋ | 190/219 [19:09<02:57,  6.13s/it]
Training Epoch: 14/100, batch 379/438 completed (loss: 0.0293:  87%|████████▋ | 190/219 [19:09<02:57,  6.13s/it]
Training Epoch: 14/100, batch 380/438 completed (loss: 0.0175:  87%|████████▋ | 190/219 [19:12<02:57,  6.13s/it]
Training Epoch: 14/100, batch 380/438 completed (loss: 0.0807:  87%|████████▋ | 190/219 [19:12<02:57,  6.13s/it]
Training Epoch: 14/100, batch 380/438 completed (loss: 0.0407:  87%|████████▋ | 190/219 [19:12<02:57,  6.13s/it]
Training Epoch: 14/100, batch 380/438 completed (loss: 0.0872:  87%|████████▋ | 190/219 [19:12<02:57,  6.13s/it]
Training Epoch: 14/100, batch 380/438 completed (loss: 0.0175:  87%|████████▋ | 191/219 [19:15<02:48,  6.03s/it]
Training Epoch: 14/100, batch 380/438 completed (loss: 0.0407:  87%|████████▋ | 191/219 [19:15<02:48,  6.03s/it]
Training Epoch: 14/100, batch 380/438 completed (loss: 0.0807:  87%|████████▋ | 191/219 [19:15<02:48,  6.03s/it]
Training Epoch: 14/100, batch 380/438 completed (loss: 0.0872:  87%|████████▋ | 191/219 [19:15<02:49,  6.04s/it]
Training Epoch: 14/100, batch 381/438 completed (loss: 0.1173:  87%|████████▋ | 191/219 [19:15<02:48,  6.03s/it]
Training Epoch: 14/100, batch 381/438 completed (loss: 0.185:  87%|████████▋ | 191/219 [19:15<02:48,  6.03s/it] 
Training Epoch: 14/100, batch 381/438 completed (loss: 0.0825:  87%|████████▋ | 191/219 [19:15<02:48,  6.03s/it]
Training Epoch: 14/100, batch 381/438 completed (loss: 0.659:  87%|████████▋ | 191/219 [19:15<02:49,  6.04s/it] 
Training Epoch: 14/100, batch 382/438 completed (loss: 0.0819:  87%|████████▋ | 191/219 [19:18<02:48,  6.03s/it]
Training Epoch: 14/100, batch 382/438 completed (loss: 0.0294:  87%|████████▋ | 191/219 [19:18<02:48,  6.03s/it]
Training Epoch: 14/100, batch 382/438 completed (loss: 0.0301:  87%|████████▋ | 191/219 [19:18<02:48,  6.03s/it]
Training Epoch: 14/100, batch 382/438 completed (loss: 0.0111:  87%|████████▋ | 191/219 [19:18<02:49,  6.04s/it]
Training Epoch: 14/100, batch 382/438 completed (loss: 0.0301:  88%|████████▊ | 192/219 [19:21<02:43,  6.07s/it]
Training Epoch: 14/100, batch 382/438 completed (loss: 0.0819:  88%|████████▊ | 192/219 [19:21<02:43,  6.07s/it]
Training Epoch: 14/100, batch 382/438 completed (loss: 0.0111:  88%|████████▊ | 192/219 [19:21<02:43,  6.07s/it]
Training Epoch: 14/100, batch 382/438 completed (loss: 0.0294:  88%|████████▊ | 192/219 [19:21<02:43,  6.07s/it]
Training Epoch: 14/100, batch 383/438 completed (loss: 0.3551:  88%|████████▊ | 192/219 [19:21<02:43,  6.07s/it]
Training Epoch: 14/100, batch 383/438 completed (loss: 0.0206:  88%|████████▊ | 192/219 [19:21<02:43,  6.07s/it]
Training Epoch: 14/100, batch 383/438 completed (loss: 0.0822:  88%|████████▊ | 192/219 [19:21<02:43,  6.07s/it]
Training Epoch: 14/100, batch 383/438 completed (loss: 0.0054:  88%|████████▊ | 192/219 [19:21<02:43,  6.07s/it]
Training Epoch: 14/100, batch 384/438 completed (loss: 0.2933:  88%|████████▊ | 192/219 [19:24<02:43,  6.07s/it]
Training Epoch: 14/100, batch 384/438 completed (loss: 0.1737:  88%|████████▊ | 192/219 [19:24<02:43,  6.07s/it]
Training Epoch: 14/100, batch 384/438 completed (loss: 0.0235:  88%|████████▊ | 192/219 [19:24<02:43,  6.07s/it]
Training Epoch: 14/100, batch 384/438 completed (loss: 0.1152:  88%|████████▊ | 192/219 [19:24<02:43,  6.07s/it]
Training Epoch: 14/100, batch 384/438 completed (loss: 0.0235:  88%|████████▊ | 193/219 [19:27<02:38,  6.11s/it]
Training Epoch: 14/100, batch 384/438 completed (loss: 0.1737:  88%|████████▊ | 193/219 [19:27<02:39,  6.12s/it]
Training Epoch: 14/100, batch 384/438 completed (loss: 0.2933:  88%|████████▊ | 193/219 [19:27<02:38,  6.11s/it]
Training Epoch: 14/100, batch 384/438 completed (loss: 0.1152:  88%|████████▊ | 193/219 [19:27<02:38,  6.11s/it]
Training Epoch: 14/100, batch 385/438 completed (loss: 0.0593:  88%|████████▊ | 193/219 [19:27<02:38,  6.11s/it]
Training Epoch: 14/100, batch 385/438 completed (loss: 0.0859:  88%|████████▊ | 193/219 [19:27<02:38,  6.11s/it]
Training Epoch: 14/100, batch 385/438 completed (loss: 0.1488:  88%|████████▊ | 193/219 [19:27<02:39,  6.12s/it]
Training Epoch: 14/100, batch 385/438 completed (loss: 0.271:  88%|████████▊ | 193/219 [19:28<02:38,  6.11s/it] 
Training Epoch: 14/100, batch 386/438 completed (loss: 0.1793:  88%|████████▊ | 193/219 [19:30<02:39,  6.12s/it]
Training Epoch: 14/100, batch 386/438 completed (loss: 0.103:  88%|████████▊ | 193/219 [19:30<02:38,  6.11s/it] 
Training Epoch: 14/100, batch 386/438 completed (loss: 0.1296:  88%|████████▊ | 193/219 [19:30<02:38,  6.11s/it]
Training Epoch: 14/100, batch 386/438 completed (loss: 0.144:  88%|████████▊ | 193/219 [19:30<02:38,  6.11s/it]
Training Epoch: 14/100, batch 386/438 completed (loss: 0.1793:  89%|████████▊ | 194/219 [19:33<02:31,  6.07s/it]
Training Epoch: 14/100, batch 386/438 completed (loss: 0.103:  89%|████████▊ | 194/219 [19:33<02:31,  6.07s/it]
Training Epoch: 14/100, batch 386/438 completed (loss: 0.1296:  89%|████████▊ | 194/219 [19:33<02:31,  6.07s/it]
Training Epoch: 14/100, batch 386/438 completed (loss: 0.144:  89%|████████▊ | 194/219 [19:33<02:31,  6.07s/it]
Training Epoch: 14/100, batch 387/438 completed (loss: 0.522:  89%|████████▊ | 194/219 [19:33<02:31,  6.07s/it]
Training Epoch: 14/100, batch 387/438 completed (loss: 0.0248:  89%|████████▊ | 194/219 [19:33<02:31,  6.07s/it]
Training Epoch: 14/100, batch 387/438 completed (loss: 0.0385:  89%|████████▊ | 194/219 [19:33<02:31,  6.07s/it]
Training Epoch: 14/100, batch 387/438 completed (loss: 0.026:  89%|████████▊ | 194/219 [19:33<02:31,  6.07s/it]
Training Epoch: 14/100, batch 388/438 completed (loss: 0.1414:  89%|████████▊ | 194/219 [19:36<02:31,  6.07s/it]
Training Epoch: 14/100, batch 388/438 completed (loss: 0.099:  89%|████████▊ | 194/219 [19:36<02:31,  6.07s/it] 
Training Epoch: 14/100, batch 388/438 completed (loss: 0.0228:  89%|████████▊ | 194/219 [19:36<02:31,  6.07s/it]
Training Epoch: 14/100, batch 388/438 completed (loss: 0.0197:  89%|████████▊ | 194/219 [19:36<02:31,  6.07s/it]
Training Epoch: 14/100, batch 388/438 completed (loss: 0.0197:  89%|████████▉ | 195/219 [19:40<02:25,  6.08s/it]
Training Epoch: 14/100, batch 388/438 completed (loss: 0.0228:  89%|████████▉ | 195/219 [19:40<02:26,  6.09s/it]
Training Epoch: 14/100, batch 388/438 completed (loss: 0.1414:  89%|████████▉ | 195/219 [19:40<02:26,  6.09s/it]
Training Epoch: 14/100, batch 388/438 completed (loss: 0.099:  89%|████████▉ | 195/219 [19:40<02:26,  6.09s/it]
Training Epoch: 14/100, batch 389/438 completed (loss: 0.0753:  89%|████████▉ | 195/219 [19:40<02:26,  6.09s/it]
Training Epoch: 14/100, batch 389/438 completed (loss: 0.0197:  89%|████████▉ | 195/219 [19:40<02:26,  6.09s/it]
Training Epoch: 14/100, batch 389/438 completed (loss: 0.0412:  89%|████████▉ | 195/219 [19:40<02:26,  6.09s/it]
Training Epoch: 14/100, batch 389/438 completed (loss: 0.405:  89%|████████▉ | 195/219 [19:40<02:25,  6.08s/it] 
Training Epoch: 14/100, batch 390/438 completed (loss: 0.0651:  89%|████████▉ | 195/219 [19:43<02:26,  6.09s/it]
Training Epoch: 14/100, batch 390/438 completed (loss: 0.0837:  89%|████████▉ | 195/219 [19:43<02:26,  6.09s/it]
Training Epoch: 14/100, batch 390/438 completed (loss: 0.0145:  89%|████████▉ | 195/219 [19:43<02:26,  6.09s/it]
Training Epoch: 14/100, batch 390/438 completed (loss: 0.0534:  89%|████████▉ | 195/219 [19:43<02:25,  6.08s/it]
Training Epoch: 14/100, batch 390/438 completed (loss: 0.0534:  89%|████████▉ | 196/219 [19:46<02:19,  6.07s/it]
Training Epoch: 14/100, batch 390/438 completed (loss: 0.0837:  89%|████████▉ | 196/219 [19:46<02:19,  6.07s/it]
Training Epoch: 14/100, batch 390/438 completed (loss: 0.0651:  89%|████████▉ | 196/219 [19:46<02:19,  6.07s/it]
Training Epoch: 14/100, batch 390/438 completed (loss: 0.0145:  89%|████████▉ | 196/219 [19:46<02:19,  6.07s/it]
Training Epoch: 14/100, batch 391/438 completed (loss: 0.0482:  89%|████████▉ | 196/219 [19:46<02:19,  6.07s/it]
Training Epoch: 14/100, batch 391/438 completed (loss: 0.1847:  89%|████████▉ | 196/219 [19:46<02:19,  6.07s/it]
Training Epoch: 14/100, batch 391/438 completed (loss: 0.13:  89%|████████▉ | 196/219 [19:46<02:19,  6.07s/it]  
Training Epoch: 14/100, batch 391/438 completed (loss: 0.0347:  89%|████████▉ | 196/219 [19:46<02:19,  6.07s/it]
Training Epoch: 14/100, batch 392/438 completed (loss: 0.0636:  89%|████████▉ | 196/219 [19:49<02:19,  6.07s/it]
Training Epoch: 14/100, batch 392/438 completed (loss: 0.0259:  89%|████████▉ | 196/219 [19:49<02:19,  6.07s/it]
Training Epoch: 14/100, batch 392/438 completed (loss: 0.0765:  89%|████████▉ | 196/219 [19:49<02:19,  6.07s/it]
Training Epoch: 14/100, batch 392/438 completed (loss: 0.1892:  89%|████████▉ | 196/219 [19:49<02:19,  6.07s/it]
Training Epoch: 14/100, batch 392/438 completed (loss: 0.0765:  90%|████████▉ | 197/219 [19:52<02:14,  6.10s/it]
Training Epoch: 14/100, batch 392/438 completed (loss: 0.0636:  90%|████████▉ | 197/219 [19:52<02:14,  6.10s/it]
Training Epoch: 14/100, batch 392/438 completed (loss: 0.0259:  90%|████████▉ | 197/219 [19:52<02:14,  6.10s/it]
Training Epoch: 14/100, batch 392/438 completed (loss: 0.1892:  90%|████████▉ | 197/219 [19:52<02:14,  6.10s/it]
Training Epoch: 14/100, batch 393/438 completed (loss: 0.116:  90%|████████▉ | 197/219 [19:52<02:14,  6.10s/it] 
Training Epoch: 14/100, batch 393/438 completed (loss: 0.0036:  90%|████████▉ | 197/219 [19:52<02:14,  6.10s/it]
Training Epoch: 14/100, batch 393/438 completed (loss: 0.0377:  90%|████████▉ | 197/219 [19:52<02:14,  6.10s/it]
Training Epoch: 14/100, batch 393/438 completed (loss: 0.1134:  90%|████████▉ | 197/219 [19:52<02:14,  6.10s/it]
Training Epoch: 14/100, batch 394/438 completed (loss: 0.2223:  90%|████████▉ | 197/219 [19:55<02:14,  6.10s/it]
Training Epoch: 14/100, batch 394/438 completed (loss: 0.062:  90%|████████▉ | 197/219 [19:55<02:14,  6.10s/it] 
Training Epoch: 14/100, batch 394/438 completed (loss: 0.0209:  90%|████████▉ | 197/219 [19:55<02:14,  6.10s/it]
Training Epoch: 14/100, batch 394/438 completed (loss: 0.1735:  90%|████████▉ | 197/219 [19:55<02:14,  6.10s/it]
Training Epoch: 14/100, batch 394/438 completed (loss: 0.1735:  90%|█████████ | 198/219 [19:58<02:07,  6.07s/it]
Training Epoch: 14/100, batch 394/438 completed (loss: 0.0209:  90%|█████████ | 198/219 [19:58<02:07,  6.07s/it]
Training Epoch: 14/100, batch 394/438 completed (loss: 0.062:  90%|█████████ | 198/219 [19:58<02:07,  6.07s/it]
Training Epoch: 14/100, batch 394/438 completed (loss: 0.2223:  90%|█████████ | 198/219 [19:58<02:07,  6.07s/it]
Training Epoch: 14/100, batch 395/438 completed (loss: 0.3572:  90%|█████████ | 198/219 [19:58<02:07,  6.07s/it]
Training Epoch: 14/100, batch 395/438 completed (loss: 0.0309:  90%|█████████ | 198/219 [19:58<02:07,  6.07s/it]
Training Epoch: 14/100, batch 395/438 completed (loss: 0.2372:  90%|█████████ | 198/219 [19:58<02:07,  6.07s/it]
Training Epoch: 14/100, batch 395/438 completed (loss: 0.2055:  90%|█████████ | 198/219 [19:58<02:07,  6.07s/it]
Training Epoch: 14/100, batch 396/438 completed (loss: 0.314:  90%|█████████ | 198/219 [20:01<02:07,  6.07s/it] 
Training Epoch: 14/100, batch 396/438 completed (loss: 0.3622:  90%|█████████ | 198/219 [20:01<02:07,  6.07s/it]
Training Epoch: 14/100, batch 396/438 completed (loss: 0.2704:  90%|█████████ | 198/219 [20:01<02:07,  6.07s/it]
Training Epoch: 14/100, batch 396/438 completed (loss: 0.057:  90%|█████████ | 198/219 [20:01<02:07,  6.07s/it] 
Training Epoch: 14/100, batch 396/438 completed (loss: 0.2704:  91%|█████████ | 199/219 [20:04<02:01,  6.09s/it]
Training Epoch: 14/100, batch 396/438 completed (loss: 0.3622:  91%|█████████ | 199/219 [20:04<02:01,  6.09s/it]
Training Epoch: 14/100, batch 396/438 completed (loss: 0.314:  91%|█████████ | 199/219 [20:04<02:01,  6.09s/it]
Training Epoch: 14/100, batch 396/438 completed (loss: 0.057:  91%|█████████ | 199/219 [20:04<02:01,  6.09s/it]
Training Epoch: 14/100, batch 397/438 completed (loss: 0.1018:  91%|█████████ | 199/219 [20:04<02:01,  6.09s/it]
Training Epoch: 14/100, batch 397/438 completed (loss: 0.0786:  91%|█████████ | 199/219 [20:04<02:01,  6.09s/it]
Training Epoch: 14/100, batch 397/438 completed (loss: 0.0117:  91%|█████████ | 199/219 [20:04<02:01,  6.09s/it]
Training Epoch: 14/100, batch 397/438 completed (loss: 0.0592:  91%|█████████ | 199/219 [20:04<02:01,  6.09s/it]
Training Epoch: 14/100, batch 398/438 completed (loss: 0.0737:  91%|█████████ | 199/219 [20:07<02:01,  6.09s/it]
Training Epoch: 14/100, batch 398/438 completed (loss: 0.1029:  91%|█████████ | 199/219 [20:07<02:01,  6.09s/it]
Training Epoch: 14/100, batch 398/438 completed (loss: 0.273:  91%|█████████ | 199/219 [20:07<02:01,  6.09s/it] 
Training Epoch: 14/100, batch 398/438 completed (loss: 0.0989:  91%|█████████ | 199/219 [20:07<02:01,  6.09s/it]
Training Epoch: 14/100, batch 398/438 completed (loss: 0.0737:  91%|█████████▏| 200/219 [20:10<01:55,  6.06s/it]
Training Epoch: 14/100, batch 398/438 completed (loss: 0.273:  91%|█████████▏| 200/219 [20:10<01:55,  6.06s/it]
Training Epoch: 14/100, batch 398/438 completed (loss: 0.1029:  91%|█████████▏| 200/219 [20:10<01:55,  6.06s/it]
Training Epoch: 14/100, batch 398/438 completed (loss: 0.0989:  91%|█████████▏| 200/219 [20:10<01:55,  6.07s/it]
Training Epoch: 14/100, batch 399/438 completed (loss: 0.2445:  91%|█████████▏| 200/219 [20:10<01:55,  6.06s/it]
Training Epoch: 14/100, batch 399/438 completed (loss: 0.0082:  91%|█████████▏| 200/219 [20:10<01:55,  6.06s/it]
Training Epoch: 14/100, batch 399/438 completed (loss: 0.0485:  91%|█████████▏| 200/219 [20:10<01:55,  6.06s/it]
Training Epoch: 14/100, batch 399/438 completed (loss: 0.1393:  91%|█████████▏| 200/219 [20:10<01:55,  6.07s/it]
Training Epoch: 14/100, batch 400/438 completed (loss: 0.1018:  91%|█████████▏| 200/219 [20:13<01:55,  6.06s/it]
Training Epoch: 14/100, batch 400/438 completed (loss: 0.1353:  91%|█████████▏| 200/219 [20:13<01:55,  6.06s/it]
Training Epoch: 14/100, batch 400/438 completed (loss: 0.012:  91%|█████████▏| 200/219 [20:13<01:55,  6.06s/it] 
Training Epoch: 14/100, batch 400/438 completed (loss: 0.0389:  91%|█████████▏| 200/219 [20:13<01:55,  6.07s/it]
Training Epoch: 14/100, batch 400/438 completed (loss: 0.0389:  92%|█████████▏| 201/219 [20:16<01:49,  6.11s/it]
Training Epoch: 14/100, batch 400/438 completed (loss: 0.012:  92%|█████████▏| 201/219 [20:16<01:49,  6.11s/it]
Training Epoch: 14/100, batch 400/438 completed (loss: 0.1353:  92%|█████████▏| 201/219 [20:16<01:49,  6.11s/it]
Training Epoch: 14/100, batch 400/438 completed (loss: 0.1018:  92%|█████████▏| 201/219 [20:16<01:49,  6.11s/it]
Training Epoch: 14/100, batch 401/438 completed (loss: 0.0749:  92%|█████████▏| 201/219 [20:16<01:49,  6.11s/it]
Training Epoch: 14/100, batch 401/438 completed (loss: 0.0546:  92%|█████████▏| 201/219 [20:16<01:49,  6.11s/it]
Training Epoch: 14/100, batch 401/438 completed (loss: 0.4828:  92%|█████████▏| 201/219 [20:16<01:49,  6.11s/it]
Training Epoch: 14/100, batch 401/438 completed (loss: 0.5448:  92%|█████████▏| 201/219 [20:16<01:49,  6.11s/it]
Training Epoch: 14/100, batch 402/438 completed (loss: 0.2236:  92%|█████████▏| 201/219 [20:19<01:49,  6.11s/it]
Training Epoch: 14/100, batch 402/438 completed (loss: 0.0321:  92%|█████████▏| 201/219 [20:19<01:49,  6.11s/it]
Training Epoch: 14/100, batch 402/438 completed (loss: 0.0154:  92%|█████████▏| 201/219 [20:19<01:49,  6.11s/it]
Training Epoch: 14/100, batch 402/438 completed (loss: 0.0445:  92%|█████████▏| 201/219 [20:19<01:49,  6.11s/it]
Training Epoch: 14/100, batch 402/438 completed (loss: 0.0445:  92%|█████████▏| 202/219 [20:22<01:44,  6.12s/it]
Training Epoch: 14/100, batch 402/438 completed (loss: 0.0321:  92%|█████████▏| 202/219 [20:22<01:44,  6.12s/it]
Training Epoch: 14/100, batch 402/438 completed (loss: 0.0154:  92%|█████████▏| 202/219 [20:22<01:44,  6.13s/it]
Training Epoch: 14/100, batch 402/438 completed (loss: 0.2236:  92%|█████████▏| 202/219 [20:22<01:44,  6.13s/it]
Training Epoch: 14/100, batch 403/438 completed (loss: 0.012:  92%|█████████▏| 202/219 [20:22<01:44,  6.13s/it] 
Training Epoch: 14/100, batch 403/438 completed (loss: 0.0438:  92%|█████████▏| 202/219 [20:22<01:44,  6.13s/it]
Training Epoch: 14/100, batch 403/438 completed (loss: 0.0204:  92%|█████████▏| 202/219 [20:22<01:44,  6.12s/it]
Training Epoch: 14/100, batch 403/438 completed (loss: 0.0615:  92%|█████████▏| 202/219 [20:22<01:44,  6.12s/it]
Training Epoch: 14/100, batch 404/438 completed (loss: 0.0502:  92%|█████████▏| 202/219 [20:25<01:44,  6.13s/it]
Training Epoch: 14/100, batch 404/438 completed (loss: 0.0778:  92%|█████████▏| 202/219 [20:25<01:44,  6.12s/it]
Training Epoch: 14/100, batch 404/438 completed (loss: 0.0359:  92%|█████████▏| 202/219 [20:25<01:44,  6.13s/it]
Training Epoch: 14/100, batch 404/438 completed (loss: 0.1296:  92%|█████████▏| 202/219 [20:25<01:44,  6.12s/it]
Training Epoch: 14/100, batch 404/438 completed (loss: 0.1296:  93%|█████████▎| 203/219 [20:28<01:37,  6.10s/it]
Training Epoch: 14/100, batch 404/438 completed (loss: 0.0778:  93%|█████████▎| 203/219 [20:28<01:37,  6.11s/it]
Training Epoch: 14/100, batch 404/438 completed (loss: 0.0359:  93%|█████████▎| 203/219 [20:28<01:37,  6.11s/it]
Training Epoch: 14/100, batch 404/438 completed (loss: 0.0502:  93%|█████████▎| 203/219 [20:28<01:37,  6.11s/it]
Training Epoch: 14/100, batch 405/438 completed (loss: 0.0372:  93%|█████████▎| 203/219 [20:28<01:37,  6.11s/it]
Training Epoch: 14/100, batch 405/438 completed (loss: 0.4616:  93%|█████████▎| 203/219 [20:28<01:37,  6.11s/it]
Training Epoch: 14/100, batch 405/438 completed (loss: 0.0087:  93%|█████████▎| 203/219 [20:28<01:37,  6.11s/it]
Training Epoch: 14/100, batch 405/438 completed (loss: 0.0476:  93%|█████████▎| 203/219 [20:28<01:37,  6.10s/it]
Training Epoch: 14/100, batch 406/438 completed (loss: 0.0293:  93%|█████████▎| 203/219 [20:31<01:37,  6.11s/it]
Training Epoch: 14/100, batch 406/438 completed (loss: 0.1196:  93%|█████████▎| 203/219 [20:31<01:37,  6.11s/it]
Training Epoch: 14/100, batch 406/438 completed (loss: 0.0491:  93%|█████████▎| 203/219 [20:31<01:37,  6.11s/it]
Training Epoch: 14/100, batch 406/438 completed (loss: 0.1141:  93%|█████████▎| 203/219 [20:31<01:37,  6.10s/it]
Training Epoch: 14/100, batch 406/438 completed (loss: 0.0491:  93%|█████████▎| 204/219 [20:34<01:31,  6.09s/it]
Training Epoch: 14/100, batch 406/438 completed (loss: 0.1141:  93%|█████████▎| 204/219 [20:34<01:31,  6.09s/it]
Training Epoch: 14/100, batch 406/438 completed (loss: 0.1196:  93%|█████████▎| 204/219 [20:34<01:31,  6.09s/it]
Training Epoch: 14/100, batch 406/438 completed (loss: 0.0293:  93%|█████████▎| 204/219 [20:34<01:31,  6.09s/it]
Training Epoch: 14/100, batch 407/438 completed (loss: 0.1138:  93%|█████████▎| 204/219 [20:34<01:31,  6.09s/it]
Training Epoch: 14/100, batch 407/438 completed (loss: 0.0269:  93%|█████████▎| 204/219 [20:34<01:31,  6.09s/it]
Training Epoch: 14/100, batch 407/438 completed (loss: 0.1174:  93%|█████████▎| 204/219 [20:34<01:31,  6.09s/it]
Training Epoch: 14/100, batch 407/438 completed (loss: 0.1923:  93%|█████████▎| 204/219 [20:34<01:31,  6.09s/it]
Training Epoch: 14/100, batch 408/438 completed (loss: 0.1826:  93%|█████████▎| 204/219 [20:37<01:31,  6.09s/it]
Training Epoch: 14/100, batch 408/438 completed (loss: 0.0311:  93%|█████████▎| 204/219 [20:37<01:31,  6.09s/it]
Training Epoch: 14/100, batch 408/438 completed (loss: 0.0151:  93%|█████████▎| 204/219 [20:37<01:31,  6.09s/it]
Training Epoch: 14/100, batch 408/438 completed (loss: 0.0339:  93%|█████████▎| 204/219 [20:37<01:31,  6.09s/it]
Training Epoch: 14/100, batch 408/438 completed (loss: 0.1826:  94%|█████████▎| 205/219 [20:40<01:25,  6.10s/it]
Training Epoch: 14/100, batch 408/438 completed (loss: 0.0151:  94%|█████████▎| 205/219 [20:40<01:25,  6.10s/it]
Training Epoch: 14/100, batch 408/438 completed (loss: 0.0339:  94%|█████████▎| 205/219 [20:41<01:25,  6.10s/it]
Training Epoch: 14/100, batch 408/438 completed (loss: 0.0311:  94%|█████████▎| 205/219 [20:40<01:25,  6.10s/it]
Training Epoch: 14/100, batch 409/438 completed (loss: 0.1163:  94%|█████████▎| 205/219 [20:41<01:25,  6.10s/it]
Training Epoch: 14/100, batch 409/438 completed (loss: 0.0656:  94%|█████████▎| 205/219 [20:41<01:25,  6.10s/it]
Training Epoch: 14/100, batch 409/438 completed (loss: 0.2178:  94%|█████████▎| 205/219 [20:41<01:25,  6.10s/it]
Training Epoch: 14/100, batch 409/438 completed (loss: 0.0189:  94%|█████████▎| 205/219 [20:41<01:25,  6.10s/it]
Training Epoch: 14/100, batch 410/438 completed (loss: 0.0353:  94%|█████████▎| 205/219 [20:44<01:25,  6.10s/it]
Training Epoch: 14/100, batch 410/438 completed (loss: 0.2796:  94%|█████████▎| 205/219 [20:44<01:25,  6.10s/it]
Training Epoch: 14/100, batch 410/438 completed (loss: 0.0109:  94%|█████████▎| 205/219 [20:44<01:25,  6.10s/it]
Training Epoch: 14/100, batch 410/438 completed (loss: 0.1279:  94%|█████████▎| 205/219 [20:44<01:25,  6.10s/it]
Training Epoch: 14/100, batch 410/438 completed (loss: 0.0353:  94%|█████████▍| 206/219 [20:46<01:18,  6.07s/it]
Training Epoch: 14/100, batch 410/438 completed (loss: 0.0109:  94%|█████████▍| 206/219 [20:46<01:18,  6.08s/it]
Training Epoch: 14/100, batch 410/438 completed (loss: 0.2796:  94%|█████████▍| 206/219 [20:46<01:18,  6.08s/it]
Training Epoch: 14/100, batch 410/438 completed (loss: 0.1279:  94%|█████████▍| 206/219 [20:47<01:19,  6.08s/it]
Training Epoch: 14/100, batch 411/438 completed (loss: 0.314:  94%|█████████▍| 206/219 [20:47<01:18,  6.08s/it] 
Training Epoch: 14/100, batch 411/438 completed (loss: 0.0163:  94%|█████████▍| 206/219 [20:47<01:18,  6.07s/it]
Training Epoch: 14/100, batch 411/438 completed (loss: 0.0316:  94%|█████████▍| 206/219 [20:47<01:18,  6.08s/it]
Training Epoch: 14/100, batch 411/438 completed (loss: 0.033:  94%|█████████▍| 206/219 [20:47<01:19,  6.08s/it] 
Training Epoch: 14/100, batch 412/438 completed (loss: 0.03:  94%|█████████▍| 206/219 [20:50<01:18,  6.08s/it]  
Training Epoch: 14/100, batch 412/438 completed (loss: 0.0243:  94%|█████████▍| 206/219 [20:50<01:18,  6.07s/it]
Training Epoch: 14/100, batch 412/438 completed (loss: 0.0121:  94%|█████████▍| 206/219 [20:50<01:18,  6.08s/it]
Training Epoch: 14/100, batch 412/438 completed (loss: 0.0382:  94%|█████████▍| 206/219 [20:50<01:19,  6.08s/it]
Training Epoch: 14/100, batch 412/438 completed (loss: 0.0121:  95%|█████████▍| 207/219 [20:53<01:13,  6.09s/it]
Training Epoch: 14/100, batch 412/438 completed (loss: 0.0243:  95%|█████████▍| 207/219 [20:53<01:13,  6.10s/it]
Training Epoch: 14/100, batch 412/438 completed (loss: 0.03:  95%|█████████▍| 207/219 [20:53<01:13,  6.10s/it]
Training Epoch: 14/100, batch 412/438 completed (loss: 0.0382:  95%|█████████▍| 207/219 [20:53<01:13,  6.10s/it]
Training Epoch: 14/100, batch 413/438 completed (loss: 0.1546:  95%|█████████▍| 207/219 [20:53<01:13,  6.10s/it]
Training Epoch: 14/100, batch 413/438 completed (loss: 0.0108:  95%|█████████▍| 207/219 [20:53<01:13,  6.09s/it]
Training Epoch: 14/100, batch 413/438 completed (loss: 0.073:  95%|█████████▍| 207/219 [20:53<01:13,  6.10s/it] 
Training Epoch: 14/100, batch 413/438 completed (loss: 0.0297:  95%|█████████▍| 207/219 [20:53<01:13,  6.10s/it]
Training Epoch: 14/100, batch 414/438 completed (loss: 0.0608:  95%|█████████▍| 207/219 [20:56<01:13,  6.10s/it]
Training Epoch: 14/100, batch 414/438 completed (loss: 0.1378:  95%|█████████▍| 207/219 [20:56<01:13,  6.09s/it]
Training Epoch: 14/100, batch 414/438 completed (loss: 0.061:  95%|█████████▍| 207/219 [20:56<01:13,  6.10s/it] 
Training Epoch: 14/100, batch 414/438 completed (loss: 0.0325:  95%|█████████▍| 207/219 [20:56<01:13,  6.10s/it]
Training Epoch: 14/100, batch 414/438 completed (loss: 0.1378:  95%|█████████▍| 208/219 [20:59<01:06,  6.07s/it]
Training Epoch: 14/100, batch 414/438 completed (loss: 0.0608:  95%|█████████▍| 208/219 [20:59<01:06,  6.07s/it]
Training Epoch: 14/100, batch 414/438 completed (loss: 0.061:  95%|█████████▍| 208/219 [20:59<01:06,  6.06s/it]
Training Epoch: 14/100, batch 414/438 completed (loss: 0.0325:  95%|█████████▍| 208/219 [20:59<01:06,  6.07s/it]
Training Epoch: 14/100, batch 415/438 completed (loss: 0.0461:  95%|█████████▍| 208/219 [20:59<01:06,  6.07s/it]
Training Epoch: 14/100, batch 415/438 completed (loss: 0.1971:  95%|█████████▍| 208/219 [20:59<01:06,  6.07s/it]
Training Epoch: 14/100, batch 415/438 completed (loss: 0.0327:  95%|█████████▍| 208/219 [20:59<01:06,  6.06s/it]
Training Epoch: 14/100, batch 415/438 completed (loss: 0.1706:  95%|█████████▍| 208/219 [20:59<01:06,  6.07s/it]
Training Epoch: 14/100, batch 416/438 completed (loss: 0.245:  95%|█████████▍| 208/219 [21:02<01:06,  6.07s/it] 
Training Epoch: 14/100, batch 416/438 completed (loss: 0.0279:  95%|█████████▍| 208/219 [21:02<01:06,  6.07s/it]
Training Epoch: 14/100, batch 416/438 completed (loss: 0.1078:  95%|█████████▍| 208/219 [21:02<01:06,  6.06s/it]
Training Epoch: 14/100, batch 416/438 completed (loss: 0.5182:  95%|█████████▍| 208/219 [21:02<01:06,  6.07s/it]
Training Epoch: 14/100, batch 416/438 completed (loss: 0.1078:  95%|█████████▌| 209/219 [21:05<01:00,  6.09s/it]
Training Epoch: 14/100, batch 416/438 completed (loss: 0.0279:  95%|█████████▌| 209/219 [21:05<01:00,  6.10s/it]
Training Epoch: 14/100, batch 416/438 completed (loss: 0.245:  95%|█████████▌| 209/219 [21:05<01:00,  6.10s/it]
Training Epoch: 14/100, batch 416/438 completed (loss: 0.5182:  95%|█████████▌| 209/219 [21:05<01:00,  6.10s/it]
Training Epoch: 14/100, batch 417/438 completed (loss: 0.1286:  95%|█████████▌| 209/219 [21:05<01:00,  6.10s/it]
Training Epoch: 14/100, batch 417/438 completed (loss: 0.0481:  95%|█████████▌| 209/219 [21:05<01:00,  6.09s/it]
Training Epoch: 14/100, batch 417/438 completed (loss: 0.2998:  95%|█████████▌| 209/219 [21:05<01:00,  6.10s/it]
Training Epoch: 14/100, batch 417/438 completed (loss: 0.1228:  95%|█████████▌| 209/219 [21:05<01:00,  6.10s/it]
Training Epoch: 14/100, batch 418/438 completed (loss: 0.0967:  95%|█████████▌| 209/219 [21:08<01:00,  6.09s/it]
Training Epoch: 14/100, batch 418/438 completed (loss: 0.0036:  95%|█████████▌| 209/219 [21:08<01:00,  6.10s/it]
Training Epoch: 14/100, batch 418/438 completed (loss: 0.0276:  95%|█████████▌| 209/219 [21:08<01:00,  6.10s/it]
Training Epoch: 14/100, batch 418/438 completed (loss: 0.0672:  95%|█████████▌| 209/219 [21:08<01:00,  6.10s/it]
Training Epoch: 14/100, batch 418/438 completed (loss: 0.0036:  96%|█████████▌| 210/219 [21:11<00:54,  6.11s/it]
Training Epoch: 14/100, batch 418/438 completed (loss: 0.0967:  96%|█████████▌| 210/219 [21:11<00:54,  6.11s/it]
Training Epoch: 14/100, batch 418/438 completed (loss: 0.0276:  96%|█████████▌| 210/219 [21:11<00:54,  6.11s/it]
Training Epoch: 14/100, batch 418/438 completed (loss: 0.0672:  96%|█████████▌| 210/219 [21:11<00:54,  6.11s/it]
Training Epoch: 14/100, batch 419/438 completed (loss: 0.0582:  96%|█████████▌| 210/219 [21:11<00:54,  6.11s/it]
Training Epoch: 14/100, batch 419/438 completed (loss: 0.0038:  96%|█████████▌| 210/219 [21:11<00:54,  6.11s/it]
Training Epoch: 14/100, batch 419/438 completed (loss: 0.0828:  96%|█████████▌| 210/219 [21:11<00:54,  6.11s/it]
Training Epoch: 14/100, batch 419/438 completed (loss: 0.055:  96%|█████████▌| 210/219 [21:11<00:54,  6.11s/it] 
Training Epoch: 14/100, batch 420/438 completed (loss: 0.0705:  96%|█████████▌| 210/219 [21:14<00:54,  6.11s/it]
Training Epoch: 14/100, batch 420/438 completed (loss: 0.0724:  96%|█████████▌| 210/219 [21:14<00:54,  6.11s/it]
Training Epoch: 14/100, batch 420/438 completed (loss: 0.0407:  96%|█████████▌| 210/219 [21:14<00:54,  6.11s/it]
Training Epoch: 14/100, batch 420/438 completed (loss: 0.1028:  96%|█████████▌| 210/219 [21:14<00:54,  6.11s/it]
Training Epoch: 14/100, batch 420/438 completed (loss: 0.0724:  96%|█████████▋| 211/219 [21:17<00:48,  6.07s/it]
Training Epoch: 14/100, batch 420/438 completed (loss: 0.0407:  96%|█████████▋| 211/219 [21:17<00:48,  6.07s/it]
Training Epoch: 14/100, batch 420/438 completed (loss: 0.1028:  96%|█████████▋| 211/219 [21:17<00:48,  6.07s/it]
Training Epoch: 14/100, batch 420/438 completed (loss: 0.0705:  96%|█████████▋| 211/219 [21:17<00:48,  6.07s/it]
Training Epoch: 14/100, batch 421/438 completed (loss: 0.059:  96%|█████████▋| 211/219 [21:17<00:48,  6.07s/it] 
Training Epoch: 14/100, batch 421/438 completed (loss: 0.0124:  96%|█████████▋| 211/219 [21:17<00:48,  6.07s/it]
Training Epoch: 14/100, batch 421/438 completed (loss: 0.2019:  96%|█████████▋| 211/219 [21:17<00:48,  6.07s/it]
Training Epoch: 14/100, batch 421/438 completed (loss: 0.047:  96%|█████████▋| 211/219 [21:17<00:48,  6.07s/it] 
Training Epoch: 14/100, batch 422/438 completed (loss: 0.0469:  96%|█████████▋| 211/219 [21:20<00:48,  6.07s/it]
Training Epoch: 14/100, batch 422/438 completed (loss: 0.0666:  96%|█████████▋| 211/219 [21:20<00:48,  6.07s/it]
Training Epoch: 14/100, batch 422/438 completed (loss: 0.0343:  96%|█████████▋| 211/219 [21:20<00:48,  6.07s/it]
Training Epoch: 14/100, batch 422/438 completed (loss: 0.1181:  96%|█████████▋| 211/219 [21:20<00:48,  6.07s/it]
Training Epoch: 14/100, batch 422/438 completed (loss: 0.0666:  97%|█████████▋| 212/219 [21:23<00:42,  6.10s/it]
Training Epoch: 14/100, batch 422/438 completed (loss: 0.0469:  97%|█████████▋| 212/219 [21:23<00:42,  6.10s/it]
Training Epoch: 14/100, batch 422/438 completed (loss: 0.0343:  97%|█████████▋| 212/219 [21:23<00:42,  6.10s/it]
Training Epoch: 14/100, batch 422/438 completed (loss: 0.1181:  97%|█████████▋| 212/219 [21:23<00:42,  6.10s/it]
Training Epoch: 14/100, batch 423/438 completed (loss: 0.0397:  97%|█████████▋| 212/219 [21:23<00:42,  6.10s/it]
Training Epoch: 14/100, batch 423/438 completed (loss: 0.0883:  97%|█████████▋| 212/219 [21:23<00:42,  6.10s/it]
Training Epoch: 14/100, batch 423/438 completed (loss: 0.0259:  97%|█████████▋| 212/219 [21:23<00:42,  6.10s/it]
Training Epoch: 14/100, batch 423/438 completed (loss: 0.1465:  97%|█████████▋| 212/219 [21:23<00:42,  6.10s/it]
Training Epoch: 14/100, batch 424/438 completed (loss: 0.0021:  97%|█████████▋| 212/219 [21:26<00:42,  6.10s/it]
Training Epoch: 14/100, batch 424/438 completed (loss: 0.0429:  97%|█████████▋| 212/219 [21:26<00:42,  6.10s/it]
Training Epoch: 14/100, batch 424/438 completed (loss: 0.2571:  97%|█████████▋| 212/219 [21:26<00:42,  6.10s/it]
Training Epoch: 14/100, batch 424/438 completed (loss: 0.0616:  97%|█████████▋| 212/219 [21:26<00:42,  6.10s/it]
Training Epoch: 14/100, batch 424/438 completed (loss: 0.0429:  97%|█████████▋| 213/219 [21:29<00:36,  6.07s/it]
Training Epoch: 14/100, batch 424/438 completed (loss: 0.2571:  97%|█████████▋| 213/219 [21:29<00:36,  6.07s/it]
Training Epoch: 14/100, batch 424/438 completed (loss: 0.0021:  97%|█████████▋| 213/219 [21:29<00:36,  6.07s/it]
Training Epoch: 14/100, batch 424/438 completed (loss: 0.0616:  97%|█████████▋| 213/219 [21:29<00:36,  6.07s/it]
Training Epoch: 14/100, batch 425/438 completed (loss: 0.2151:  97%|█████████▋| 213/219 [21:29<00:36,  6.07s/it]
Training Epoch: 14/100, batch 425/438 completed (loss: 0.0304:  97%|█████████▋| 213/219 [21:29<00:36,  6.07s/it]
Training Epoch: 14/100, batch 425/438 completed (loss: 0.0807:  97%|█████████▋| 213/219 [21:29<00:36,  6.07s/it]
Training Epoch: 14/100, batch 425/438 completed (loss: 0.0859:  97%|█████████▋| 213/219 [21:29<00:36,  6.07s/it]
Training Epoch: 14/100, batch 426/438 completed (loss: 0.1111:  97%|█████████▋| 213/219 [21:32<00:36,  6.07s/it]
Training Epoch: 14/100, batch 426/438 completed (loss: 0.1422:  97%|█████████▋| 213/219 [21:32<00:36,  6.07s/it]
Training Epoch: 14/100, batch 426/438 completed (loss: 0.0908:  97%|█████████▋| 213/219 [21:32<00:36,  6.07s/it]
Training Epoch: 14/100, batch 426/438 completed (loss: 0.3582:  97%|█████████▋| 213/219 [21:32<00:36,  6.07s/it]
Training Epoch: 14/100, batch 426/438 completed (loss: 0.0908:  98%|█████████▊| 214/219 [21:35<00:30,  6.04s/it]
Training Epoch: 14/100, batch 426/438 completed (loss: 0.1111:  98%|█████████▊| 214/219 [21:35<00:30,  6.04s/it]
Training Epoch: 14/100, batch 426/438 completed (loss: 0.3582:  98%|█████████▊| 214/219 [21:35<00:30,  6.04s/it]
Training Epoch: 14/100, batch 426/438 completed (loss: 0.1422:  98%|█████████▊| 214/219 [21:35<00:30,  6.04s/it]
Training Epoch: 14/100, batch 427/438 completed (loss: 0.1378:  98%|█████████▊| 214/219 [21:35<00:30,  6.04s/it]
Training Epoch: 14/100, batch 427/438 completed (loss: 0.0524:  98%|█████████▊| 214/219 [21:35<00:30,  6.04s/it]
Training Epoch: 14/100, batch 427/438 completed (loss: 0.2158:  98%|█████████▊| 214/219 [21:35<00:30,  6.04s/it]
Training Epoch: 14/100, batch 427/438 completed (loss: 0.2779:  98%|█████████▊| 214/219 [21:35<00:30,  6.04s/it]
Training Epoch: 14/100, batch 428/438 completed (loss: 0.0103:  98%|█████████▊| 214/219 [21:38<00:30,  6.04s/it]
Training Epoch: 14/100, batch 428/438 completed (loss: 0.0515:  98%|█████████▊| 214/219 [21:38<00:30,  6.04s/it]
Training Epoch: 14/100, batch 428/438 completed (loss: 0.0743:  98%|█████████▊| 214/219 [21:38<00:30,  6.04s/it]
Training Epoch: 14/100, batch 428/438 completed (loss: 0.2435:  98%|█████████▊| 214/219 [21:38<00:30,  6.04s/it]
Training Epoch: 14/100, batch 428/438 completed (loss: 0.2435:  98%|█████████▊| 215/219 [21:41<00:24,  6.12s/it]
Training Epoch: 14/100, batch 428/438 completed (loss: 0.0103:  98%|█████████▊| 215/219 [21:41<00:24,  6.12s/it]
Training Epoch: 14/100, batch 428/438 completed (loss: 0.0515:  98%|█████████▊| 215/219 [21:41<00:24,  6.12s/it]
Training Epoch: 14/100, batch 428/438 completed (loss: 0.0743:  98%|█████████▊| 215/219 [21:41<00:24,  6.13s/it]
Training Epoch: 14/100, batch 429/438 completed (loss: 0.0714:  98%|█████████▊| 215/219 [21:41<00:24,  6.12s/it]
Training Epoch: 14/100, batch 429/438 completed (loss: 0.0821:  98%|█████████▊| 215/219 [21:41<00:24,  6.13s/it]
Training Epoch: 14/100, batch 429/438 completed (loss: 0.038:  98%|█████████▊| 215/219 [21:41<00:24,  6.12s/it] 
Training Epoch: 14/100, batch 429/438 completed (loss: 0.1591:  98%|█████████▊| 215/219 [21:41<00:24,  6.12s/it]
Training Epoch: 14/100, batch 430/438 completed (loss: 0.1866:  98%|█████████▊| 215/219 [21:44<00:24,  6.12s/it]
Training Epoch: 14/100, batch 430/438 completed (loss: 0.0718:  98%|█████████▊| 215/219 [21:44<00:24,  6.12s/it]
Training Epoch: 14/100, batch 430/438 completed (loss: 0.1101:  98%|█████████▊| 215/219 [21:44<00:24,  6.13s/it]
Training Epoch: 14/100, batch 430/438 completed (loss: 0.0317:  98%|█████████▊| 215/219 [21:44<00:24,  6.12s/it]
Training Epoch: 14/100, batch 430/438 completed (loss: 0.1866:  99%|█████████▊| 216/219 [21:47<00:18,  6.09s/it]
Training Epoch: 14/100, batch 430/438 completed (loss: 0.0317:  99%|█████████▊| 216/219 [21:47<00:18,  6.09s/it]
Training Epoch: 14/100, batch 430/438 completed (loss: 0.0718:  99%|█████████▊| 216/219 [21:47<00:18,  6.09s/it]
Training Epoch: 14/100, batch 430/438 completed (loss: 0.1101:  99%|█████████▊| 216/219 [21:47<00:18,  6.09s/it]
Training Epoch: 14/100, batch 431/438 completed (loss: 0.1631:  99%|█████████▊| 216/219 [21:47<00:18,  6.09s/it]
Training Epoch: 14/100, batch 431/438 completed (loss: 0.0621:  99%|█████████▊| 216/219 [21:47<00:18,  6.09s/it]
Training Epoch: 14/100, batch 431/438 completed (loss: 0.0113:  99%|█████████▊| 216/219 [21:47<00:18,  6.09s/it]
Training Epoch: 14/100, batch 431/438 completed (loss: 0.056:  99%|█████████▊| 216/219 [21:47<00:18,  6.09s/it] 
Training Epoch: 14/100, batch 432/438 completed (loss: 0.1359:  99%|█████████▊| 216/219 [21:50<00:18,  6.09s/it]
Training Epoch: 14/100, batch 432/438 completed (loss: 0.0505:  99%|█████████▊| 216/219 [21:50<00:18,  6.09s/it]
Training Epoch: 14/100, batch 432/438 completed (loss: 0.0403:  99%|█████████▊| 216/219 [21:50<00:18,  6.09s/it]
Training Epoch: 14/100, batch 432/438 completed (loss: 0.0809:  99%|█████████▊| 216/219 [21:50<00:18,  6.09s/it]
Training Epoch: 14/100, batch 432/438 completed (loss: 0.0505:  99%|█████████▉| 217/219 [21:54<00:12,  6.11s/it]
Training Epoch: 14/100, batch 432/438 completed (loss: 0.0809:  99%|█████████▉| 217/219 [21:54<00:12,  6.11s/it]
Training Epoch: 14/100, batch 432/438 completed (loss: 0.1359:  99%|█████████▉| 217/219 [21:54<00:12,  6.11s/it]
Training Epoch: 14/100, batch 432/438 completed (loss: 0.0403:  99%|█████████▉| 217/219 [21:54<00:12,  6.11s/it]
Training Epoch: 14/100, batch 433/438 completed (loss: 0.0655:  99%|█████████▉| 217/219 [21:54<00:12,  6.11s/it]
Training Epoch: 14/100, batch 433/438 completed (loss: 0.1449:  99%|█████████▉| 217/219 [21:54<00:12,  6.11s/it]
Training Epoch: 14/100, batch 433/438 completed (loss: 0.043:  99%|█████████▉| 217/219 [21:54<00:12,  6.11s/it] 
Training Epoch: 14/100, batch 433/438 completed (loss: 0.0126:  99%|█████████▉| 217/219 [21:54<00:12,  6.11s/it]
Training Epoch: 14/100, batch 434/438 completed (loss: 0.0376:  99%|█████████▉| 217/219 [21:56<00:12,  6.11s/it]
Training Epoch: 14/100, batch 434/438 completed (loss: 0.1132:  99%|█████████▉| 217/219 [21:56<00:12,  6.11s/it]
Training Epoch: 14/100, batch 434/438 completed (loss: 0.015:  99%|█████████▉| 217/219 [21:56<00:12,  6.11s/it]
Training Epoch: 14/100, batch 434/438 completed (loss: 0.0253:  99%|█████████▉| 217/219 [21:57<00:12,  6.11s/it]
Training Epoch: 14/100, batch 434/438 completed (loss: 0.015: 100%|█████████▉| 218/219 [21:59<00:06,  6.06s/it]
Training Epoch: 14/100, batch 434/438 completed (loss: 0.0253: 100%|█████████▉| 218/219 [22:00<00:06,  6.06s/it]
Training Epoch: 14/100, batch 434/438 completed (loss: 0.1132: 100%|█████████▉| 218/219 [21:59<00:06,  6.06s/it]
Training Epoch: 14/100, batch 434/438 completed (loss: 0.0376: 100%|█████████▉| 218/219 [21:59<00:06,  6.07s/it]
Training Epoch: 14/100, batch 435/438 completed (loss: 0.0752: 100%|█████████▉| 218/219 [22:00<00:06,  6.06s/it]
Training Epoch: 14/100, batch 435/438 completed (loss: 0.0433: 100%|█████████▉| 218/219 [22:00<00:06,  6.06s/it]
Training Epoch: 14/100, batch 435/438 completed (loss: 0.3378: 100%|█████████▉| 218/219 [22:00<00:06,  6.07s/it]
Training Epoch: 14/100, batch 435/438 completed (loss: 0.0096: 100%|█████████▉| 218/219 [22:00<00:06,  6.06s/it]
Training Epoch: 14/100, batch 436/438 completed (loss: 0.2844: 100%|█████████▉| 218/219 [22:03<00:06,  6.06s/it]
Training Epoch: 14/100, batch 436/438 completed (loss: 0.0152: 100%|█████████▉| 218/219 [22:03<00:06,  6.06s/it]
Training Epoch: 14/100, batch 436/438 completed (loss: 0.0227: 100%|█████████▉| 218/219 [22:03<00:06,  6.07s/it]
Training Epoch: 14/100, batch 436/438 completed (loss: 0.0153: 100%|█████████▉| 218/219 [22:03<00:06,  6.06s/it]
Training Epoch: 14/100, batch 436/438 completed (loss: 0.0227: 100%|██████████| 219/219 [22:05<00:00,  5.79s/it]
Training Epoch: 14/100, batch 436/438 completed (loss: 0.2844: 100%|██████████| 219/219 [22:05<00:00,  5.79s/it]
Training Epoch: 14/100, batch 436/438 completed (loss: 0.0152: 100%|██████████| 219/219 [22:05<00:00,  5.79s/it]
Training Epoch: 14/100, batch 436/438 completed (loss: 0.0153: 100%|██████████| 219/219 [22:05<00:00,  5.79s/it]
Training Epoch: 14/100, batch 437/438 completed (loss: 0.5691: 100%|██████████| 219/219 [22:05<00:00,  5.79s/it]
Training Epoch: 14/100, batch 437/438 completed (loss: 0.0369: 100%|██████████| 219/219 [22:05<00:00,  5.79s/it]
Training Epoch: 14/100, batch 437/438 completed (loss: 0.5025: 100%|██████████| 219/219 [22:05<00:00,  5.79s/it]
Training Epoch: 14/100, batch 437/438 completed (loss: 0.0127: 100%|██████████| 219/219 [22:05<00:00,  5.79s/it]
Training Epoch: 14/100, batch 437/438 completed (loss: 0.5691: 100%|██████████| 219/219 [22:05<00:00,  6.05s/it]

Training Epoch: 14/100, batch 437/438 completed (loss: 0.0369: 100%|██████████| 219/219 [22:05<00:00,  6.05s/it]

Training Epoch: 14/100, batch 437/438 completed (loss: 0.5025: 100%|██████████| 219/219 [22:05<00:00,  6.05s/it]

Training Epoch: 14/100, batch 437/438 completed (loss: 0.0127: 100%|██████████| 219/219 [22:05<00:00,  6.05s/it]
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/c10d_logger.py:83: UserWarning: barrier(): using the device under current context. You can specify `device_id` in `init_process_group` to mute this warning.
  return func(*args, **kwargs)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
saving model + train state.

Processing Files (0 / 0)      : |          |  0.00B /  0.00B            

New Data Upload               : |          |  0.00B /  0.00B            


  ...ve-03b89ee9/checkpoint_14:   0%|          | 82.1kB / 3.63GB            


  ...ve-03b89ee9/checkpoint_14:   0%|          | 82.1kB / 3.63GB            
Processing Files (0 / 1)      :   0%|          | 82.1kB / 3.63GB,   ???B/s  


  ...ve-03b89ee9/checkpoint_14:   1%|          | 18.9MB / 3.63GB            
Processing Files (0 / 1)      :   1%|          | 18.9MB / 3.63GB, 1.84MB/s  

New Data Upload               :  14%|█▍        | 18.8MB /  134MB, 1.84MB/s  


  ...ve-03b89ee9/checkpoint_14:   3%|▎         |  103MB / 3.63GB            
Processing Files (0 / 1)      :   3%|▎         |  103MB / 3.63GB, 9.94MB/s  

New Data Upload               :  51%|█████     |  103MB /  201MB, 9.93MB/s  


  ...ve-03b89ee9/checkpoint_14:   4%|▎         |  134MB / 3.63GB            
Processing Files (0 / 1)      :   4%|▎         |  134MB / 3.63GB, 12.7MB/s  

New Data Upload               :  66%|██████▋   |  133MB /  201MB, 12.7MB/s  


  ...ve-03b89ee9/checkpoint_14:   5%|▍         |  174MB / 3.63GB            
Processing Files (0 / 1)      :   5%|▍         |  174MB / 3.63GB, 15.3MB/s  

New Data Upload               :  65%|██████▍   |  174MB /  268MB, 15.3MB/s  


  ...ve-03b89ee9/checkpoint_14:   7%|▋         |  253MB / 3.63GB            
Processing Files (0 / 1)      :   7%|▋         |  253MB / 3.63GB, 22.3MB/s  

New Data Upload               :  63%|██████▎   |  253MB /  402MB, 22.3MB/s  


  ...ve-03b89ee9/checkpoint_14:   9%|▊         |  314MB / 3.63GB            
Processing Files (0 / 1)      :   9%|▊         |  314MB / 3.63GB, 27.5MB/s  

New Data Upload               :  67%|██████▋   |  314MB /  469MB, 27.5MB/s  


  ...ve-03b89ee9/checkpoint_14:   9%|▉         |  335MB / 3.63GB            
Processing Files (0 / 1)      :   9%|▉         |  335MB / 3.63GB, 28.7MB/s  

New Data Upload               :  71%|███████▏  |  335MB /  469MB, 28.7MB/s  


  ...ve-03b89ee9/checkpoint_14:  10%|█         |  373MB / 3.63GB            
Processing Files (0 / 1)      :  10%|█         |  373MB / 3.63GB, 31.5MB/s  

New Data Upload               :  69%|██████▉   |  373MB /  536MB, 31.5MB/s  


  ...ve-03b89ee9/checkpoint_14:  12%|█▏        |  435MB / 3.63GB            
Processing Files (0 / 1)      :  12%|█▏        |  435MB / 3.63GB, 36.5MB/s  

New Data Upload               :  72%|███████▏  |  434MB /  604MB, 36.5MB/s  


  ...ve-03b89ee9/checkpoint_14:  14%|█▍        |  503MB / 3.63GB            
Processing Files (0 / 1)      :  14%|█▍        |  503MB / 3.63GB, 41.9MB/s  

New Data Upload               :  68%|██████▊   |  503MB /  738MB, 41.8MB/s  


  ...ve-03b89ee9/checkpoint_14:  17%|█▋        |  609MB / 3.63GB            
Processing Files (0 / 1)      :  17%|█▋        |  609MB / 3.63GB, 50.4MB/s  

New Data Upload               :  76%|███████▌  |  609MB /  805MB, 50.4MB/s  


  ...ve-03b89ee9/checkpoint_14:  19%|█▉        |  703MB / 3.63GB            
Processing Files (0 / 1)      :  19%|█▉        |  703MB / 3.63GB, 57.7MB/s  

New Data Upload               :  81%|████████  |  703MB /  872MB, 57.7MB/s  


  ...ve-03b89ee9/checkpoint_14:  21%|██        |  767MB / 3.63GB            
Processing Files (0 / 1)      :  21%|██        |  767MB / 3.63GB, 62.1MB/s  

New Data Upload               :  82%|████████▏ |  766MB /  939MB, 62.1MB/s  


  ...ve-03b89ee9/checkpoint_14:  23%|██▎       |  831MB / 3.63GB            
Processing Files (0 / 1)      :  23%|██▎       |  831MB / 3.63GB, 66.5MB/s  

New Data Upload               :  83%|████████▎ |  831MB / 1.01GB, 66.5MB/s  


  ...ve-03b89ee9/checkpoint_14:  25%|██▌       |  922MB / 3.63GB            
Processing Files (0 / 1)      :  25%|██▌       |  922MB / 3.63GB, 73.0MB/s  

New Data Upload               :  81%|████████  |  922MB / 1.14GB, 73.0MB/s  


  ...ve-03b89ee9/checkpoint_14:  29%|██▉       | 1.04GB / 3.63GB            
Processing Files (0 / 1)      :  29%|██▉       | 1.04GB / 3.63GB, 82.1MB/s  

New Data Upload               :  86%|████████▋ | 1.04GB / 1.21GB, 82.1MB/s  


  ...ve-03b89ee9/checkpoint_14:  31%|███       | 1.13GB / 3.63GB            
Processing Files (0 / 1)      :  31%|███       | 1.13GB / 3.63GB, 88.0MB/s  

New Data Upload               :  89%|████████▉ | 1.13GB / 1.27GB, 88.0MB/s  


  ...ve-03b89ee9/checkpoint_14:  33%|███▎      | 1.20GB / 3.63GB            
Processing Files (0 / 1)      :  33%|███▎      | 1.20GB / 3.63GB, 92.1MB/s  

New Data Upload               :  85%|████████▌ | 1.20GB / 1.41GB, 92.1MB/s  


  ...ve-03b89ee9/checkpoint_14:  36%|███▌      | 1.31GB / 3.63GB            
Processing Files (0 / 1)      :  36%|███▌      | 1.31GB / 3.63GB, 99.3MB/s  

New Data Upload               :  89%|████████▊ | 1.31GB / 1.48GB, 99.3MB/s  


  ...ve-03b89ee9/checkpoint_14:  38%|███▊      | 1.38GB / 3.63GB            
Processing Files (0 / 1)      :  38%|███▊      | 1.38GB / 3.63GB,  103MB/s  

New Data Upload               :  89%|████████▉ | 1.38GB / 1.54GB,  103MB/s  


  ...ve-03b89ee9/checkpoint_14:  41%|████      | 1.47GB / 3.63GB            
Processing Files (0 / 1)      :  41%|████      | 1.47GB / 3.63GB,  109MB/s  

New Data Upload               :  91%|█████████▏| 1.47GB / 1.61GB,  109MB/s  


  ...ve-03b89ee9/checkpoint_14:  43%|████▎     | 1.57GB / 3.63GB            
Processing Files (0 / 1)      :  43%|████▎     | 1.57GB / 3.63GB,  115MB/s  

New Data Upload               :  90%|████████▉ | 1.56GB / 1.74GB,  115MB/s  


  ...ve-03b89ee9/checkpoint_14:  47%|████▋     | 1.69GB / 3.63GB            
Processing Files (0 / 1)      :  47%|████▋     | 1.69GB / 3.63GB,  123MB/s  

New Data Upload               :  93%|█████████▎| 1.69GB / 1.81GB,  123MB/s  


  ...ve-03b89ee9/checkpoint_14:  49%|████▉     | 1.77GB / 3.63GB            
Processing Files (0 / 1)      :  49%|████▉     | 1.77GB / 3.63GB,  127MB/s  

New Data Upload               :  94%|█████████▍| 1.77GB / 1.88GB,  127MB/s  


  ...ve-03b89ee9/checkpoint_14:  52%|█████▏    | 1.87GB / 3.63GB            
Processing Files (0 / 1)      :  52%|█████▏    | 1.87GB / 3.63GB,  133MB/s  

New Data Upload               :  93%|█████████▎| 1.87GB / 2.01GB,  133MB/s  


  ...ve-03b89ee9/checkpoint_14:  54%|█████▍    | 1.95GB / 3.63GB            
Processing Files (0 / 1)      :  54%|█████▍    | 1.95GB / 3.63GB,  137MB/s  

New Data Upload               :  94%|█████████▍| 1.95GB / 2.08GB,  137MB/s  


  ...ve-03b89ee9/checkpoint_14:  56%|█████▌    | 2.02GB / 3.63GB            
Processing Files (0 / 1)      :  56%|█████▌    | 2.02GB / 3.63GB,  140MB/s  

New Data Upload               :  94%|█████████▍| 2.02GB / 2.15GB,  140MB/s  


  ...ve-03b89ee9/checkpoint_14:  59%|█████▉    | 2.13GB / 3.63GB            
Processing Files (0 / 1)      :  59%|█████▉    | 2.13GB / 3.63GB,  146MB/s  

New Data Upload               :  94%|█████████▎| 2.13GB / 2.28GB,  146MB/s  


  ...ve-03b89ee9/checkpoint_14:  60%|█████▉    | 2.17GB / 3.63GB            
Processing Files (0 / 1)      :  60%|█████▉    | 2.17GB / 3.63GB,  147MB/s  

New Data Upload               :  93%|█████████▎| 2.17GB / 2.35GB,  147MB/s  


  ...ve-03b89ee9/checkpoint_14:  71%|███████   | 2.58GB / 3.63GB            
Processing Files (0 / 1)      :  71%|███████   | 2.58GB / 3.63GB,  177MB/s  

New Data Upload               :  97%|█████████▋| 2.27GB / 2.35GB,  151MB/s  


  ...ve-03b89ee9/checkpoint_14:  82%|████████▏ | 2.99GB / 3.63GB            
Processing Files (0 / 1)      :  82%|████████▏ | 2.99GB / 3.63GB,  208MB/s  

New Data Upload               :  98%|█████████▊| 2.30GB / 2.35GB,  151MB/s  


  ...ve-03b89ee9/checkpoint_14:  93%|█████████▎| 3.37GB / 3.63GB            
Processing Files (0 / 1)      :  93%|█████████▎| 3.37GB / 3.63GB,  234MB/s  

New Data Upload               :  98%|█████████▊| 2.31GB / 2.35GB,  149MB/s  


  ...ve-03b89ee9/checkpoint_14:  98%|█████████▊| 3.57GB / 3.63GB            
Processing Files (0 / 1)      :  98%|█████████▊| 3.57GB / 3.63GB,  246MB/s  

New Data Upload               :  97%|█████████▋| 2.32GB / 2.38GB,  147MB/s  


  ...ve-03b89ee9/checkpoint_14:  99%|█████████▉| 3.61GB / 3.63GB            
Processing Files (0 / 1)      :  99%|█████████▉| 3.61GB / 3.63GB,  244MB/s  

New Data Upload               :  99%|█████████▉| 2.36GB / 2.38GB,  147MB/s  


  ...ve-03b89ee9/checkpoint_14: 100%|█████████▉| 3.62GB / 3.63GB            
Processing Files (0 / 1)      : 100%|█████████▉| 3.62GB / 3.63GB,  240MB/s  

New Data Upload               : 100%|█████████▉| 2.37GB / 2.38GB,  145MB/s  


  ...ve-03b89ee9/checkpoint_14: 100%|█████████▉| 3.62GB / 3.63GB            
Processing Files (0 / 1)      : 100%|█████████▉| 3.62GB / 3.63GB,  236MB/s  

New Data Upload               : 100%|█████████▉| 2.38GB / 2.38GB,  143MB/s  


  ...ve-03b89ee9/checkpoint_14: 100%|█████████▉| 3.63GB / 3.63GB            
Processing Files (0 / 1)      : 100%|█████████▉| 3.63GB / 3.63GB,  231MB/s  

New Data Upload               : 100%|█████████▉| 2.38GB / 2.38GB,  140MB/s  


  ...ve-03b89ee9/checkpoint_14: 100%|█████████▉| 3.63GB / 3.63GB            
Processing Files (0 / 1)      : 100%|█████████▉| 3.63GB / 3.63GB,  217MB/s  

New Data Upload               : 100%|██████████| 2.38GB / 2.38GB,  132MB/s  
Processing Files (0 / 1)      : 100%|█████████▉| 3.63GB / 3.63GB,  217MB/s  

New Data Upload               : 100%|██████████| 2.38GB / 2.38GB,  132MB/s  

  ...ve-03b89ee9/checkpoint_14: 100%|█████████▉| 3.63GB / 3.63GB            
Uploaded checkpoint_14 to seyedparsa/thoughtformer-checkpoints/checkpoints/star-datacurr-only-4gpu-bs32-lr1e-5_n1000_qwen3-0.6b-base_thoughtformer_lr1e-5_no-reset_dc_no-tc_adaptive-03b89ee9/checkpoint_14

Processing Files (0 / 0)      : |          |  0.00B /  0.00B            

New Data Upload               : |          |  0.00B /  0.00B            


  ...e-03b89ee9/train_state_14:   0%|          | 54.6kB / 4.77GB            


  ...e-03b89ee9/train_state_14:   0%|          | 54.6kB / 4.77GB            
Processing Files (0 / 1)      :   0%|          | 54.6kB / 4.77GB,   ???B/s  


  ...e-03b89ee9/train_state_14:   1%|          | 32.2MB / 4.77GB            
Processing Files (0 / 1)      :   1%|          | 32.2MB / 4.77GB, 3.14MB/s  

New Data Upload               :  16%|█▌        | 32.1MB /  201MB, 3.14MB/s  


  ...e-03b89ee9/train_state_14:   2%|▏         | 75.3MB / 4.77GB            
Processing Files (0 / 1)      :   2%|▏         | 75.3MB / 4.77GB, 7.25MB/s  

New Data Upload               :  37%|███▋      | 75.2MB /  201MB, 7.25MB/s  


  ...e-03b89ee9/train_state_14:   2%|▏         | 83.2MB / 4.77GB            
Processing Files (0 / 1)      :   2%|▏         | 83.2MB / 4.77GB, 7.84MB/s  

New Data Upload               :  41%|████▏     | 83.1MB /  201MB, 7.83MB/s  


  ...e-03b89ee9/train_state_14:   2%|▏         | 90.5MB / 4.77GB            
Processing Files (0 / 1)      :   2%|▏         | 90.5MB / 4.77GB, 8.35MB/s  

New Data Upload               :  45%|████▍     | 90.5MB /  201MB, 8.35MB/s  


  ...e-03b89ee9/train_state_14:   2%|▏         | 97.9MB / 4.77GB            
Processing Files (0 / 1)      :   2%|▏         | 97.9MB / 4.77GB, 8.85MB/s  

New Data Upload               :  49%|████▊     | 97.8MB /  201MB, 8.85MB/s  


  ...e-03b89ee9/train_state_14:   2%|▏         |  107MB / 4.77GB            
Processing Files (0 / 1)      :   2%|▏         |  107MB / 4.77GB, 9.46MB/s  

New Data Upload               :  53%|█████▎    |  106MB /  201MB, 9.46MB/s  


  ...e-03b89ee9/train_state_14:   3%|▎         |  156MB / 4.77GB            
Processing Files (0 / 1)      :   3%|▎         |  156MB / 4.77GB, 13.9MB/s  

New Data Upload               :  46%|████▋     |  156MB /  335MB, 13.9MB/s  


  ...e-03b89ee9/train_state_14:   5%|▌         |  257MB / 4.77GB            
Processing Files (0 / 1)      :   5%|▌         |  257MB / 4.77GB, 23.0MB/s  

New Data Upload               :  77%|███████▋  |  257MB /  335MB, 23.0MB/s  


  ...e-03b89ee9/train_state_14:   6%|▌         |  269MB / 4.77GB            
Processing Files (0 / 1)      :   6%|▌         |  269MB / 4.77GB, 23.6MB/s  

New Data Upload               :  67%|██████▋   |  269MB /  402MB, 23.5MB/s  


  ...e-03b89ee9/train_state_14:   7%|▋         |  310MB / 4.77GB            
Processing Files (0 / 1)      :   7%|▋         |  310MB / 4.77GB, 26.8MB/s  

New Data Upload               :  77%|███████▋  |  310MB /  402MB, 26.8MB/s  


  ...e-03b89ee9/train_state_14:   8%|▊         |  358MB / 4.77GB            
Processing Files (0 / 1)      :   8%|▊         |  358MB / 4.77GB, 30.5MB/s  

New Data Upload               :  67%|██████▋   |  357MB /  537MB, 30.5MB/s  


  ...e-03b89ee9/train_state_14:  10%|▉         |  467MB / 4.77GB            
Processing Files (0 / 1)      :  10%|▉         |  467MB / 4.77GB, 39.8MB/s  

New Data Upload               :  77%|███████▋  |  467MB /  604MB, 39.8MB/s  


  ...e-03b89ee9/train_state_14:  11%|█         |  509MB / 4.77GB            
Processing Files (0 / 1)      :  11%|█         |  509MB / 4.77GB, 42.7MB/s  

New Data Upload               :  84%|████████▍ |  509MB /  604MB, 42.7MB/s  


  ...e-03b89ee9/train_state_14:  11%|█▏        |  542MB / 4.77GB            
Processing Files (0 / 1)      :  11%|█▏        |  542MB / 4.77GB, 44.7MB/s  

New Data Upload               :  81%|████████  |  541MB /  671MB, 44.7MB/s  


  ...e-03b89ee9/train_state_14:  13%|█▎        |  605MB / 4.77GB            
Processing Files (0 / 1)      :  13%|█▎        |  605MB / 4.77GB, 49.4MB/s  

New Data Upload               :  75%|███████▌  |  605MB /  805MB, 49.4MB/s  


  ...e-03b89ee9/train_state_14:  14%|█▍        |  689MB / 4.77GB            
Processing Files (0 / 1)      :  14%|█▍        |  689MB / 4.77GB, 55.8MB/s  

New Data Upload               :  79%|███████▉  |  689MB /  872MB, 55.7MB/s  


  ...e-03b89ee9/train_state_14:  17%|█▋        |  792MB / 4.77GB            
Processing Files (0 / 1)      :  17%|█▋        |  792MB / 4.77GB, 63.6MB/s  

New Data Upload               :  84%|████████▍ |  791MB /  939MB, 63.6MB/s  


  ...e-03b89ee9/train_state_14:  18%|█▊        |  873MB / 4.77GB            
Processing Files (0 / 1)      :  18%|█▊        |  873MB / 4.77GB, 69.4MB/s  

New Data Upload               :  81%|████████▏ |  873MB / 1.07GB, 69.4MB/s  


  ...e-03b89ee9/train_state_14:  20%|██        |  967MB / 4.77GB            
Processing Files (0 / 1)      :  20%|██        |  967MB / 4.77GB, 76.1MB/s  

New Data Upload               :  85%|████████▍ |  967MB / 1.14GB, 76.1MB/s  


  ...e-03b89ee9/train_state_14:  23%|██▎       | 1.11GB / 4.77GB            
Processing Files (0 / 1)      :  23%|██▎       | 1.11GB / 4.77GB, 87.1MB/s  

New Data Upload               :  87%|████████▋ | 1.11GB / 1.27GB, 87.1MB/s  


  ...e-03b89ee9/train_state_14:  26%|██▌       | 1.22GB / 4.77GB            
Processing Files (0 / 1)      :  26%|██▌       | 1.22GB / 4.77GB, 95.1MB/s  

New Data Upload               :  87%|████████▋ | 1.22GB / 1.41GB, 95.1MB/s  


  ...e-03b89ee9/train_state_14:  28%|██▊       | 1.34GB / 4.77GB            
Processing Files (0 / 1)      :  28%|██▊       | 1.34GB / 4.77GB,  103MB/s  

New Data Upload               :  91%|█████████ | 1.34GB / 1.48GB,  103MB/s  


  ...e-03b89ee9/train_state_14:  30%|███       | 1.44GB / 4.77GB            
Processing Files (0 / 1)      :  30%|███       | 1.44GB / 4.77GB,  109MB/s  

New Data Upload               :  89%|████████▉ | 1.44GB / 1.61GB,  109MB/s  


  ...e-03b89ee9/train_state_14:  32%|███▏      | 1.53GB / 4.77GB            
Processing Files (0 / 1)      :  32%|███▏      | 1.53GB / 4.77GB,  115MB/s  

New Data Upload               :  92%|█████████▏| 1.53GB / 1.68GB,  115MB/s  


  ...e-03b89ee9/train_state_14:  35%|███▍      | 1.66GB / 4.77GB            
Processing Files (0 / 1)      :  35%|███▍      | 1.66GB / 4.77GB,  123MB/s  

New Data Upload               :  92%|█████████▏| 1.66GB / 1.81GB,  123MB/s  


  ...e-03b89ee9/train_state_14:  37%|███▋      | 1.75GB / 4.77GB            
Processing Files (0 / 1)      :  37%|███▋      | 1.75GB / 4.77GB,  128MB/s  

New Data Upload               :  93%|█████████▎| 1.75GB / 1.88GB,  128MB/s  


  ...e-03b89ee9/train_state_14:  38%|███▊      | 1.80GB / 4.77GB            
Processing Files (0 / 1)      :  38%|███▊      | 1.80GB / 4.77GB,  130MB/s  

New Data Upload               :  90%|████████▉ | 1.80GB / 2.01GB,  130MB/s  


  ...e-03b89ee9/train_state_14:  39%|███▉      | 1.88GB / 4.77GB            
Processing Files (0 / 1)      :  39%|███▉      | 1.88GB / 4.77GB,  134MB/s  

New Data Upload               :  90%|█████████ | 1.88GB / 2.08GB,  134MB/s  


  ...e-03b89ee9/train_state_14:  42%|████▏     | 1.99GB / 4.77GB            
Processing Files (0 / 1)      :  42%|████▏     | 1.99GB / 4.77GB,  140MB/s  

New Data Upload               :  90%|████████▉ | 1.99GB / 2.21GB,  140MB/s  


  ...e-03b89ee9/train_state_14:  44%|████▍     | 2.11GB / 4.77GB            
Processing Files (0 / 1)      :  44%|████▍     | 2.11GB / 4.77GB,  148MB/s  

New Data Upload               :  93%|█████████▎| 2.11GB / 2.28GB,  148MB/s  


  ...e-03b89ee9/train_state_14:  46%|████▋     | 2.21GB / 4.77GB            
Processing Files (0 / 1)      :  46%|████▋     | 2.21GB / 4.77GB,  153MB/s  

New Data Upload               :  92%|█████████▏| 2.21GB / 2.41GB,  153MB/s  


  ...e-03b89ee9/train_state_14:  48%|████▊     | 2.31GB / 4.77GB            
Processing Files (0 / 1)      :  48%|████▊     | 2.31GB / 4.77GB,  158MB/s  

New Data Upload               :  93%|█████████▎| 2.31GB / 2.48GB,  158MB/s  


  ...e-03b89ee9/train_state_14:  51%|█████     | 2.42GB / 4.77GB            
Processing Files (0 / 1)      :  51%|█████     | 2.42GB / 4.77GB,  164MB/s  

New Data Upload               :  93%|█████████▎| 2.42GB / 2.62GB,  164MB/s  


  ...e-03b89ee9/train_state_14:  52%|█████▏    | 2.49GB / 4.77GB            
Processing Files (0 / 1)      :  52%|█████▏    | 2.49GB / 4.77GB,  166MB/s  

New Data Upload               :  93%|█████████▎| 2.49GB / 2.68GB,  166MB/s  


  ...e-03b89ee9/train_state_14:  55%|█████▍    | 2.62GB / 4.77GB            
Processing Files (0 / 1)      :  55%|█████▍    | 2.62GB / 4.77GB,  173MB/s  

New Data Upload               :  95%|█████████▌| 2.62GB / 2.75GB,  173MB/s  


  ...e-03b89ee9/train_state_14:  57%|█████▋    | 2.70GB / 4.77GB            
Processing Files (0 / 1)      :  57%|█████▋    | 2.70GB / 4.77GB,  177MB/s  

New Data Upload               :  94%|█████████▎| 2.70GB / 2.88GB,  177MB/s  


  ...e-03b89ee9/train_state_14:  59%|█████▊    | 2.80GB / 4.77GB            
Processing Files (0 / 1)      :  59%|█████▊    | 2.80GB / 4.77GB,  181MB/s  

New Data Upload               :  95%|█████████▍| 2.80GB / 2.95GB,  181MB/s  


  ...e-03b89ee9/train_state_14:  61%|██████    | 2.90GB / 4.77GB            
Processing Files (0 / 1)      :  61%|██████    | 2.90GB / 4.77GB,  186MB/s  

New Data Upload               :  96%|█████████▌| 2.90GB / 3.02GB,  186MB/s  


  ...e-03b89ee9/train_state_14:  62%|██████▏   | 2.97GB / 4.77GB            
Processing Files (0 / 1)      :  62%|██████▏   | 2.97GB / 4.77GB,  188MB/s  

New Data Upload               :  94%|█████████▍| 2.97GB / 3.15GB,  188MB/s  


  ...e-03b89ee9/train_state_14:  64%|██████▍   | 3.06GB / 4.77GB            
Processing Files (0 / 1)      :  64%|██████▍   | 3.06GB / 4.77GB,  191MB/s  

New Data Upload               :  95%|█████████▌| 3.06GB / 3.22GB,  191MB/s  


  ...e-03b89ee9/train_state_14:  66%|██████▌   | 3.15GB / 4.77GB            
Processing Files (0 / 1)      :  66%|██████▌   | 3.15GB / 4.77GB,  194MB/s  

New Data Upload               :  94%|█████████▍| 3.15GB / 3.35GB,  194MB/s  


  ...e-03b89ee9/train_state_14:  67%|██████▋   | 3.21GB / 4.77GB            
Processing Files (0 / 1)      :  67%|██████▋   | 3.21GB / 4.77GB,  195MB/s  

New Data Upload               :  94%|█████████▍| 3.21GB / 3.42GB,  195MB/s  


  ...e-03b89ee9/train_state_14:  70%|███████   | 3.34GB / 4.77GB            
Processing Files (0 / 1)      :  70%|███████   | 3.34GB / 4.77GB,  202MB/s  

New Data Upload               :  94%|█████████▍| 3.34GB / 3.55GB,  202MB/s  


  ...e-03b89ee9/train_state_14:  72%|███████▏  | 3.45GB / 4.77GB            
Processing Files (0 / 1)      :  72%|███████▏  | 3.45GB / 4.77GB,  207MB/s  

New Data Upload               :  95%|█████████▌| 3.45GB / 3.62GB,  207MB/s  


  ...e-03b89ee9/train_state_14:  75%|███████▍  | 3.56GB / 4.77GB            
Processing Files (0 / 1)      :  75%|███████▍  | 3.56GB / 4.77GB,  211MB/s  

New Data Upload               :  95%|█████████▍| 3.56GB / 3.76GB,  211MB/s  


  ...e-03b89ee9/train_state_14:  76%|███████▋  | 3.64GB / 4.77GB            
Processing Files (0 / 1)      :  76%|███████▋  | 3.64GB / 4.77GB,  213MB/s  

New Data Upload               :  95%|█████████▌| 3.64GB / 3.82GB,  213MB/s  


  ...e-03b89ee9/train_state_14:  78%|███████▊  | 3.73GB / 4.77GB            
Processing Files (0 / 1)      :  78%|███████▊  | 3.73GB / 4.77GB,  216MB/s  

New Data Upload               :  94%|█████████▍| 3.73GB / 3.96GB,  216MB/s  


  ...e-03b89ee9/train_state_14:  81%|████████  | 3.87GB / 4.77GB            
Processing Files (0 / 1)      :  81%|████████  | 3.87GB / 4.77GB,  223MB/s  

New Data Upload               :  96%|█████████▌| 3.87GB / 4.02GB,  223MB/s  


  ...e-03b89ee9/train_state_14:  83%|████████▎ | 3.96GB / 4.77GB            
Processing Files (0 / 1)      :  83%|████████▎ | 3.96GB / 4.77GB,  226MB/s  

New Data Upload               :  97%|█████████▋| 3.96GB / 4.09GB,  226MB/s  


  ...e-03b89ee9/train_state_14:  86%|████████▌ | 4.09GB / 4.77GB            
Processing Files (0 / 1)      :  86%|████████▌ | 4.09GB / 4.77GB,  231MB/s  

New Data Upload               :  97%|█████████▋| 4.09GB / 4.23GB,  231MB/s  


  ...e-03b89ee9/train_state_14:  87%|████████▋ | 4.17GB / 4.77GB            
Processing Files (0 / 1)      :  87%|████████▋ | 4.17GB / 4.77GB,  233MB/s  

New Data Upload               :  97%|█████████▋| 4.17GB / 4.29GB,  233MB/s  


  ...e-03b89ee9/train_state_14:  89%|████████▉ | 4.25GB / 4.77GB            
Processing Files (0 / 1)      :  89%|████████▉ | 4.25GB / 4.77GB,  235MB/s  

New Data Upload               :  96%|█████████▌| 4.25GB / 4.43GB,  235MB/s  


  ...e-03b89ee9/train_state_14:  91%|█████████ | 4.33GB / 4.77GB            
Processing Files (0 / 1)      :  91%|█████████ | 4.33GB / 4.77GB,  236MB/s  

New Data Upload               :  96%|█████████▋| 4.33GB / 4.49GB,  236MB/s  


  ...e-03b89ee9/train_state_14:  92%|█████████▏| 4.40GB / 4.77GB            
Processing Files (0 / 1)      :  92%|█████████▏| 4.40GB / 4.77GB,  237MB/s  

New Data Upload               :  95%|█████████▌| 4.40GB / 4.63GB,  237MB/s  


  ...e-03b89ee9/train_state_14:  94%|█████████▍| 4.50GB / 4.77GB            
Processing Files (0 / 1)      :  94%|█████████▍| 4.50GB / 4.77GB,  241MB/s  

New Data Upload               :  96%|█████████▌| 4.50GB / 4.69GB,  241MB/s  


  ...e-03b89ee9/train_state_14:  97%|█████████▋| 4.63GB / 4.77GB            
Processing Files (0 / 1)      :  97%|█████████▋| 4.63GB / 4.77GB,  246MB/s  

New Data Upload               :  97%|█████████▋| 4.63GB / 4.77GB,  246MB/s  


  ...e-03b89ee9/train_state_14:  99%|█████████▉| 4.72GB / 4.77GB            
Processing Files (0 / 1)      :  99%|█████████▉| 4.72GB / 4.77GB,  248MB/s  

New Data Upload               :  99%|█████████▉| 4.72GB / 4.77GB,  248MB/s  


  ...e-03b89ee9/train_state_14: 100%|█████████▉| 4.76GB / 4.77GB            
Processing Files (0 / 1)      : 100%|█████████▉| 4.76GB / 4.77GB,  247MB/s  

New Data Upload               : 100%|█████████▉| 4.76GB / 4.77GB,  247MB/s  


  ...e-03b89ee9/train_state_14: 100%|█████████▉| 4.77GB / 4.77GB            
Processing Files (0 / 1)      : 100%|█████████▉| 4.77GB / 4.77GB,  242MB/s  

New Data Upload               : 100%|█████████▉| 4.77GB / 4.77GB,  242MB/s  


  ...e-03b89ee9/train_state_14: 100%|█████████▉| 4.77GB / 4.77GB            
Processing Files (0 / 1)      : 100%|█████████▉| 4.77GB / 4.77GB,  233MB/s  

New Data Upload               : 100%|█████████▉| 4.77GB / 4.77GB,  233MB/s  


  ...e-03b89ee9/train_state_14: 100%|█████████▉| 4.77GB / 4.77GB            
Processing Files (0 / 1)      : 100%|█████████▉| 4.77GB / 4.77GB,  229MB/s  

New Data Upload               : 100%|██████████| 4.77GB / 4.77GB,  229MB/s  
Processing Files (0 / 1)      : 100%|█████████▉| 4.77GB / 4.77GB,  229MB/s  

New Data Upload               : 100%|██████████| 4.77GB / 4.77GB,  229MB/s  

  ...e-03b89ee9/train_state_14: 100%|█████████▉| 4.77GB / 4.77GB            
Uploaded train_state_14 to seyedparsa/thoughtformer-checkpoints/checkpoints/star-datacurr-only-4gpu-bs32-lr1e-5_n1000_qwen3-0.6b-base_thoughtformer_lr1e-5_no-reset_dc_no-tc_adaptive-03b89ee9/train_state_14
Deleted old checkpoint: train_state_13
Deleted old checkpoint: checkpoint_13

Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]
Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]
Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]eval loss 0.03036698764613406

Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]
[CORRECT] Example 0 (diff=3)
  Expected : 'Claude'
  Extracted: 'Claude'
  Generated: '### Claude'

[WRONG] Example 1 (diff=8)
  Expected : 'Guy'
  Extracted: 'Irma'
  Generated: '### Irma'

Test Accuracy:   3%|▎         | 1/32 [00:06<03:23,  6.55s/it]
Test Accuracy:   3%|▎         | 1/32 [00:06<03:23,  6.55s/it]
Test Accuracy:   3%|▎         | 1/32 [00:06<03:23,  6.55s/it]
[CORRECT] Example 2 (diff=10)
  Expected : 'Elvis'
  Extracted: 'Elvis'
Test accuracy: 1.0:   3%|▎         | 1/32 [00:06<03:23,  6.55s/it]
  Generated: '### Elvis'

Test accuracy: 1.0:   3%|▎         | 1/32 [00:06<03:23,  6.55s/it]
Test accuracy: 0.88:   3%|▎         | 1/32 [00:06<03:23,  6.55s/it]
[WRONG] Example 3 (diff=8)
  Expected : 'Coral'
  Extracted: 'Claude'
  Generated: '### Claude'

[CORRECT] Example 4 (diff=1)
  Expected : 'Patricia'
  Extracted: 'Patricia'
  Generated: '### Patricia'

Test Accuracy:   3%|▎         | 1/32 [00:06<03:18,  6.41s/it]
Test accuracy: 0.75:   3%|▎         | 1/32 [00:06<03:18,  6.41s/it]
Test accuracy: 0.88:   6%|▋         | 2/32 [00:13<03:18,  6.61s/it]
Test accuracy: 1.0:   6%|▋         | 2/32 [00:13<03:18,  6.60s/it]
Test accuracy: 1.0:   6%|▋         | 2/32 [00:13<03:18,  6.60s/it]
Test accuracy: 0.88:   6%|▋         | 2/32 [00:13<03:18,  6.61s/it]
Test accuracy: 1.0:   6%|▋         | 2/32 [00:13<03:18,  6.60s/it]
Test accuracy: 0.81:   6%|▋         | 2/32 [00:13<03:18,  6.60s/it]
Test accuracy: 0.75:   6%|▋         | 2/32 [00:13<03:16,  6.55s/it]
Test accuracy: 0.81:   6%|▋         | 2/32 [00:13<03:16,  6.55s/it]
Test accuracy: 0.81:   9%|▉         | 3/32 [00:19<03:14,  6.69s/it]
Test accuracy: 1.0:   9%|▉         | 3/32 [00:19<03:14,  6.69s/it]
Test accuracy: 0.83:   9%|▉         | 3/32 [00:19<03:14,  6.69s/it]
Test accuracy: 0.92:   9%|▉         | 3/32 [00:19<03:14,  6.69s/it]
Test accuracy: 0.81:   9%|▉         | 3/32 [00:19<03:13,  6.66s/it]
Test accuracy: 0.83:   9%|▉         | 3/32 [00:19<03:13,  6.66s/it]
Test accuracy: 0.88:   9%|▉         | 3/32 [00:20<03:14,  6.70s/it]
Test accuracy: 0.92:   9%|▉         | 3/32 [00:20<03:14,  6.70s/it]
Test accuracy: 0.83:  12%|█▎        | 4/32 [00:26<03:01,  6.49s/it]
Test accuracy: 0.92:  12%|█▎        | 4/32 [00:26<03:01,  6.49s/it]
Test accuracy: 0.75:  12%|█▎        | 4/32 [00:26<03:01,  6.49s/it]
Test accuracy: 0.91:  12%|█▎        | 4/32 [00:26<03:01,  6.49s/it]
Test accuracy: 0.92:  12%|█▎        | 4/32 [00:26<03:01,  6.48s/it]
Test accuracy: 0.83:  12%|█▎        | 4/32 [00:26<03:01,  6.47s/it]
Test accuracy: 0.91:  12%|█▎        | 4/32 [00:26<03:01,  6.48s/it]
Test accuracy: 0.84:  12%|█▎        | 4/32 [00:26<03:01,  6.47s/it]
Test accuracy: 0.91:  16%|█▌        | 5/32 [00:32<02:55,  6.49s/it]
Test accuracy: 0.91:  16%|█▌        | 5/32 [00:32<02:55,  6.49s/it]
Test accuracy: 0.75:  16%|█▌        | 5/32 [00:32<02:55,  6.49s/it]
Test accuracy: 0.85:  16%|█▌        | 5/32 [00:32<02:55,  6.49s/it]
Test accuracy: 0.93:  16%|█▌        | 5/32 [00:32<02:55,  6.49s/it]
Test accuracy: 0.73:  16%|█▌        | 5/32 [00:32<02:55,  6.49s/it]
Test accuracy: 0.84:  16%|█▌        | 5/32 [00:32<02:54,  6.48s/it]
Test accuracy: 0.82:  16%|█▌        | 5/32 [00:32<02:54,  6.48s/it]
Test accuracy: 0.85:  19%|█▉        | 6/32 [00:38<02:45,  6.37s/it]
Test accuracy: 0.93:  19%|█▉        | 6/32 [00:38<02:45,  6.37s/it]
Test accuracy: 0.73:  19%|█▉        | 6/32 [00:38<02:45,  6.37s/it]
Test accuracy: 0.82:  19%|█▉        | 6/32 [00:38<02:45,  6.36s/it]
Test accuracy: 0.83:  19%|█▉        | 6/32 [00:38<02:45,  6.37s/it]
Test accuracy: 0.94:  19%|█▉        | 6/32 [00:38<02:45,  6.37s/it]
Test accuracy: 0.73:  19%|█▉        | 6/32 [00:38<02:45,  6.37s/it]
Test accuracy: 0.85:  19%|█▉        | 6/32 [00:38<02:45,  6.36s/it]
Test accuracy: 0.94:  22%|██▏       | 7/32 [00:45<02:39,  6.36s/it]
Test accuracy: 0.83:  22%|██▏       | 7/32 [00:45<02:39,  6.36s/it]
Test accuracy: 0.73:  22%|██▏       | 7/32 [00:45<02:39,  6.36s/it]
Test accuracy: 0.91:  22%|██▏       | 7/32 [00:45<02:39,  6.36s/it]
Test accuracy: 0.8:  22%|██▏       | 7/32 [00:45<02:39,  6.36s/it] 
Test accuracy: 0.75:  22%|██▏       | 7/32 [00:45<02:39,  6.36s/it]
Test accuracy: 0.85:  22%|██▏       | 7/32 [00:44<02:38,  6.36s/it]
Test accuracy: 0.86:  22%|██▏       | 7/32 [00:44<02:38,  6.36s/it]
Test accuracy: 0.8:  25%|██▌       | 8/32 [00:51<02:32,  6.35s/it]
Test accuracy: 0.91:  25%|██▌       | 8/32 [00:51<02:32,  6.35s/it]
Test accuracy: 0.75:  25%|██▌       | 8/32 [00:51<02:32,  6.35s/it]
Test accuracy: 0.92:  25%|██▌       | 8/32 [00:51<02:32,  6.35s/it]
Test accuracy: 0.83:  25%|██▌       | 8/32 [00:51<02:32,  6.35s/it]
Test accuracy: 0.73:  25%|██▌       | 8/32 [00:51<02:32,  6.35s/it]
Test accuracy: 0.86:  25%|██▌       | 8/32 [00:51<02:32,  6.34s/it]
Test accuracy: 0.88:  25%|██▌       | 8/32 [00:51<02:32,  6.34s/it]
Test accuracy: 0.92:  28%|██▊       | 9/32 [00:58<02:29,  6.48s/it]
Test accuracy: 0.83:  28%|██▊       | 9/32 [00:58<02:29,  6.48s/it]
Test accuracy: 0.88:  28%|██▊       | 9/32 [00:58<02:29,  6.48s/it]
Test accuracy: 0.92:  28%|██▊       | 9/32 [00:58<02:29,  6.48s/it]
Test accuracy: 0.83:  28%|██▊       | 9/32 [00:58<02:29,  6.48s/it]
Test accuracy: 0.86:  28%|██▊       | 9/32 [00:58<02:29,  6.48s/it]
Test accuracy: 0.73:  28%|██▊       | 9/32 [00:58<02:29,  6.48s/it]
Test accuracy: 0.76:  28%|██▊       | 9/32 [00:58<02:29,  6.48s/it]
Test accuracy: 0.83:  31%|███▏      | 10/32 [01:04<02:21,  6.44s/it]
Test accuracy: 0.92:  31%|███▏      | 10/32 [01:04<02:21,  6.44s/it]
Test accuracy: 0.91:  31%|███▏      | 10/32 [01:04<02:21,  6.44s/it]
Test accuracy: 0.84:  31%|███▏      | 10/32 [01:04<02:21,  6.44s/it]
Test accuracy: 0.76:  31%|███▏      | 10/32 [01:04<02:21,  6.44s/it]
Test accuracy: 0.75:  31%|███▏      | 10/32 [01:04<02:21,  6.44s/it]
Test accuracy: 0.86:  31%|███▏      | 10/32 [01:04<02:21,  6.44s/it]
Test accuracy: 0.88:  31%|███▏      | 10/32 [01:04<02:21,  6.44s/it]
Test accuracy: 0.84:  34%|███▍      | 11/32 [01:10<02:14,  6.40s/it]
Test accuracy: 0.91:  34%|███▍      | 11/32 [01:10<02:14,  6.40s/it]
Test accuracy: 0.75:  34%|███▍      | 11/32 [01:10<02:14,  6.40s/it]
Test accuracy: 0.83:  34%|███▍      | 11/32 [01:10<02:14,  6.40s/it]
Test accuracy: 0.76:  34%|███▍      | 11/32 [01:10<02:14,  6.40s/it]
Test accuracy: 0.9:  34%|███▍      | 11/32 [01:10<02:14,  6.40s/it] 
Test accuracy: 0.88:  34%|███▍      | 11/32 [01:10<02:14,  6.40s/it]
Test accuracy: 0.86:  34%|███▍      | 11/32 [01:10<02:14,  6.40s/it]
Test accuracy: 0.9:  38%|███▊      | 12/32 [01:17<02:06,  6.32s/it]
Test accuracy: 0.83:  38%|███▊      | 12/32 [01:17<02:06,  6.32s/it]
Test accuracy: 0.76:  38%|███▊      | 12/32 [01:17<02:06,  6.32s/it]
Test accuracy: 0.86:  38%|███▊      | 12/32 [01:16<02:06,  6.32s/it]
Test accuracy: 0.88:  38%|███▊      | 12/32 [01:17<02:06,  6.32s/it]
Test accuracy: 0.82:  38%|███▊      | 12/32 [01:17<02:06,  6.32s/it]
Test accuracy: 0.78:  38%|███▊      | 12/32 [01:17<02:06,  6.32s/it]
Test accuracy: 0.86:  38%|███▊      | 12/32 [01:16<02:06,  6.32s/it]
Test accuracy: 0.82:  41%|████      | 13/32 [01:23<02:00,  6.33s/it]
Test accuracy: 0.78:  41%|████      | 13/32 [01:23<02:00,  6.33s/it]
Test accuracy: 0.88:  41%|████      | 13/32 [01:23<02:00,  6.33s/it]
Test accuracy: 0.83:  41%|████      | 13/32 [01:23<02:00,  6.33s/it]
Test accuracy: 0.79:  41%|████      | 13/32 [01:23<02:00,  6.33s/it]
Test accuracy: 0.88:  41%|████      | 13/32 [01:23<02:00,  6.33s/it]
Test accuracy: 0.86:  41%|████      | 13/32 [01:23<02:00,  6.33s/it]
Test accuracy: 0.85:  41%|████      | 13/32 [01:23<02:00,  6.33s/it]
Test accuracy: 0.88:  44%|████▍     | 14/32 [01:29<01:53,  6.33s/it]
Test accuracy: 0.79:  44%|████▍     | 14/32 [01:29<01:53,  6.33s/it]
Test accuracy: 0.88:  44%|████▍     | 14/32 [01:29<01:53,  6.33s/it]
Test accuracy: 0.77:  44%|████▍     | 14/32 [01:29<01:53,  6.33s/it]
Test accuracy: 0.83:  44%|████▍     | 14/32 [01:29<01:53,  6.33s/it]
Test accuracy: 0.82:  44%|████▍     | 14/32 [01:29<01:53,  6.33s/it]
Test accuracy: 0.85:  44%|████▍     | 14/32 [01:29<01:53,  6.33s/it]
Test accuracy: 0.84:  44%|████▍     | 14/32 [01:29<01:53,  6.33s/it]
Test accuracy: 0.88:  47%|████▋     | 15/32 [01:36<01:47,  6.33s/it]
Test accuracy: 0.82:  47%|████▋     | 15/32 [01:36<01:47,  6.33s/it]
Test accuracy: 0.84:  47%|████▋     | 15/32 [01:35<01:47,  6.33s/it]
Test accuracy: 0.86:  47%|████▋     | 15/32 [01:36<01:47,  6.33s/it]
Test accuracy: 0.82:  47%|████▋     | 15/32 [01:36<01:47,  6.33s/it]
Test accuracy: 0.77:  47%|████▋     | 15/32 [01:36<01:47,  6.33s/it]
Test accuracy: 0.76:  47%|████▋     | 15/32 [01:36<01:47,  6.33s/it]
Test accuracy: 0.85:  47%|████▋     | 15/32 [01:35<01:47,  6.33s/it]
Test accuracy: 0.76:  50%|█████     | 16/32 [01:42<01:40,  6.28s/it]
Test accuracy: 0.82:  50%|█████     | 16/32 [01:42<01:40,  6.28s/it]
Test accuracy: 0.77:  50%|█████     | 16/32 [01:42<01:40,  6.28s/it]
Test accuracy: 0.84:  50%|█████     | 16/32 [01:42<01:40,  6.28s/it]
Test accuracy: 0.86:  50%|█████     | 16/32 [01:42<01:40,  6.28s/it]
Test accuracy: 0.85:  50%|█████     | 16/32 [01:42<01:40,  6.28s/it]
Test accuracy: 0.85:  50%|█████     | 16/32 [01:42<01:40,  6.28s/it]
Test accuracy: 0.84:  50%|█████     | 16/32 [01:42<01:40,  6.28s/it]
Test accuracy: 0.77:  53%|█████▎    | 17/32 [01:48<01:34,  6.29s/it]
Test accuracy: 0.78:  53%|█████▎    | 17/32 [01:48<01:34,  6.29s/it]
Test accuracy: 0.84:  53%|█████▎    | 17/32 [01:48<01:34,  6.29s/it]
Test accuracy: 0.84:  53%|█████▎    | 17/32 [01:48<01:34,  6.29s/it]
Test accuracy: 0.84:  53%|█████▎    | 17/32 [01:48<01:34,  6.29s/it]
Test accuracy: 0.85:  53%|█████▎    | 17/32 [01:48<01:34,  6.29s/it]
Test accuracy: 0.85:  53%|█████▎    | 17/32 [01:48<01:34,  6.29s/it]
Test accuracy: 0.86:  53%|█████▎    | 17/32 [01:48<01:34,  6.29s/it]
Test accuracy: 0.86:  56%|█████▋    | 18/32 [01:54<01:28,  6.31s/it]
Test accuracy: 0.78:  56%|█████▋    | 18/32 [01:54<01:28,  6.31s/it]
Test accuracy: 0.84:  56%|█████▋    | 18/32 [01:54<01:28,  6.31s/it]
Test accuracy: 0.86:  56%|█████▋    | 18/32 [01:54<01:28,  6.31s/it]
Test accuracy: 0.78:  56%|█████▋    | 18/32 [01:54<01:28,  6.31s/it]
Test accuracy: 0.84:  56%|█████▋    | 18/32 [01:54<01:28,  6.31s/it]
Test accuracy: 0.85:  56%|█████▋    | 18/32 [01:54<01:28,  6.31s/it]
Test accuracy: 0.85:  56%|█████▋    | 18/32 [01:54<01:28,  6.31s/it]
Test accuracy: 0.84:  59%|█████▉    | 19/32 [02:01<01:21,  6.27s/it]
Test accuracy: 0.86:  59%|█████▉    | 19/32 [02:01<01:21,  6.27s/it]
Test accuracy: 0.78:  59%|█████▉    | 19/32 [02:01<01:21,  6.27s/it]
Test accuracy: 0.85:  59%|█████▉    | 19/32 [02:01<01:21,  6.27s/it]
Test accuracy: 0.86:  59%|█████▉    | 19/32 [02:01<01:21,  6.27s/it]
Test accuracy: 0.78:  59%|█████▉    | 19/32 [02:01<01:21,  6.27s/it]
Test accuracy: 0.85:  59%|█████▉    | 19/32 [02:00<01:21,  6.27s/it]
Test accuracy: 0.86:  59%|█████▉    | 19/32 [02:00<01:21,  6.27s/it]
Test accuracy: 0.85:  62%|██████▎   | 20/32 [02:07<01:15,  6.29s/it]
Test accuracy: 0.86:  62%|██████▎   | 20/32 [02:07<01:15,  6.29s/it]
Test accuracy: 0.86:  62%|██████▎   | 20/32 [02:07<01:15,  6.29s/it]
Test accuracy: 0.78:  62%|██████▎   | 20/32 [02:07<01:15,  6.29s/it]
Test accuracy: 0.84:  62%|██████▎   | 20/32 [02:07<01:15,  6.29s/it]
Test accuracy: 0.86:  62%|██████▎   | 20/32 [02:07<01:15,  6.29s/it]
Test accuracy: 0.78:  62%|██████▎   | 20/32 [02:07<01:15,  6.29s/it]
Test accuracy: 0.86:  62%|██████▎   | 20/32 [02:07<01:15,  6.29s/it]
Test accuracy: 0.84:  66%|██████▌   | 21/32 [02:14<01:11,  6.52s/it]
Test accuracy: 0.78:  66%|██████▌   | 21/32 [02:14<01:11,  6.52s/it]
Test accuracy: 0.86:  66%|██████▌   | 21/32 [02:14<01:11,  6.52s/it]
Test accuracy: 0.83:  66%|██████▌   | 21/32 [02:14<01:11,  6.52s/it]
Test accuracy: 0.79:  66%|██████▌   | 21/32 [02:14<01:11,  6.52s/it]
Test accuracy: 0.85:  66%|██████▌   | 21/32 [02:14<01:11,  6.52s/it]
Test accuracy: 0.86:  66%|██████▌   | 21/32 [02:14<01:11,  6.52s/it]
Test accuracy: 0.85:  66%|██████▌   | 21/32 [02:14<01:11,  6.52s/it]
Test accuracy: 0.85:  69%|██████▉   | 22/32 [02:20<01:04,  6.46s/it]
Test accuracy: 0.79:  69%|██████▉   | 22/32 [02:20<01:04,  6.46s/it]
Test accuracy: 0.83:  69%|██████▉   | 22/32 [02:20<01:04,  6.46s/it]
Test accuracy: 0.85:  69%|██████▉   | 22/32 [02:20<01:04,  6.46s/it]
Test accuracy: 0.84:  69%|██████▉   | 22/32 [02:20<01:04,  6.46s/it]
Test accuracy: 0.84:  69%|██████▉   | 22/32 [02:20<01:04,  6.46s/it]
Test accuracy: 0.79:  69%|██████▉   | 22/32 [02:20<01:04,  6.46s/it]
Test accuracy: 0.85:  69%|██████▉   | 22/32 [02:20<01:04,  6.46s/it]
Test accuracy: 0.79:  72%|███████▏  | 23/32 [02:26<00:57,  6.37s/it]
Test accuracy: 0.8:  72%|███████▏  | 23/32 [02:26<00:57,  6.37s/it] 
Test accuracy: 0.84:  72%|███████▏  | 23/32 [02:26<00:57,  6.37s/it]
Test accuracy: 0.85:  72%|███████▏  | 23/32 [02:26<00:57,  6.37s/it]
Test accuracy: 0.84:  72%|███████▏  | 23/32 [02:26<00:57,  6.37s/it]
Test accuracy: 0.85:  72%|███████▏  | 23/32 [02:26<00:57,  6.37s/it]
Test accuracy: 0.84:  72%|███████▏  | 23/32 [02:26<00:57,  6.37s/it]
Test accuracy: 0.85:  72%|███████▏  | 23/32 [02:26<00:57,  6.37s/it]
Test accuracy: 0.8:  75%|███████▌  | 24/32 [02:33<00:52,  6.58s/it]
Test accuracy: 0.84:  75%|███████▌  | 24/32 [02:33<00:52,  6.58s/it]
Test accuracy: 0.85:  75%|███████▌  | 24/32 [02:34<00:52,  6.58s/it]
Test accuracy: 0.84:  75%|███████▌  | 24/32 [02:33<00:52,  6.58s/it]
Test accuracy: 0.8:  75%|███████▌  | 24/32 [02:33<00:52,  6.58s/it]
Test accuracy: 0.85:  75%|███████▌  | 24/32 [02:34<00:52,  6.58s/it]
Test accuracy: 0.85:  75%|███████▌  | 24/32 [02:33<00:52,  6.58s/it]
Test accuracy: 0.85:  75%|███████▌  | 24/32 [02:33<00:52,  6.58s/it]
Test accuracy: 0.84:  78%|███████▊  | 25/32 [02:40<00:45,  6.51s/it]
Test accuracy: 0.8:  78%|███████▊  | 25/32 [02:40<00:45,  6.51s/it]
Test accuracy: 0.85:  78%|███████▊  | 25/32 [02:40<00:45,  6.51s/it]
Test accuracy: 0.8:  78%|███████▊  | 25/32 [02:40<00:45,  6.51s/it]
Test accuracy: 0.83:  78%|███████▊  | 25/32 [02:40<00:45,  6.51s/it]
Test accuracy: 0.85:  78%|███████▊  | 25/32 [02:40<00:45,  6.51s/it]
Test accuracy: 0.85:  78%|███████▊  | 25/32 [02:40<00:45,  6.51s/it]
Test accuracy: 0.86:  78%|███████▊  | 25/32 [02:40<00:45,  6.51s/it]
Test accuracy: 0.83:  81%|████████▏ | 26/32 [02:46<00:38,  6.46s/it]
Test accuracy: 0.8:  81%|████████▏ | 26/32 [02:46<00:38,  6.46s/it]
Test accuracy: 0.83:  81%|████████▏ | 26/32 [02:46<00:38,  6.46s/it]
Test accuracy: 0.79:  81%|████████▏ | 26/32 [02:46<00:38,  6.46s/it]
Test accuracy: 0.86:  81%|████████▏ | 26/32 [02:46<00:38,  6.46s/it]
Test accuracy: 0.85:  81%|████████▏ | 26/32 [02:46<00:38,  6.46s/it]
Test accuracy: 0.85:  81%|████████▏ | 26/32 [02:46<00:38,  6.46s/it]
Test accuracy: 0.84:  81%|████████▏ | 26/32 [02:46<00:38,  6.46s/it]
Test accuracy: 0.85:  84%|████████▍ | 27/32 [02:52<00:31,  6.37s/it]
Test accuracy: 0.79:  84%|████████▍ | 27/32 [02:52<00:31,  6.37s/it]
Test accuracy: 0.83:  84%|████████▍ | 27/32 [02:52<00:31,  6.37s/it]
Test accuracy: 0.84:  84%|████████▍ | 27/32 [02:52<00:31,  6.37s/it]
Test accuracy: 0.78:  84%|████████▍ | 27/32 [02:52<00:31,  6.37s/it]
Test accuracy: 0.85:  84%|████████▍ | 27/32 [02:52<00:31,  6.37s/it]
Test accuracy: 0.83:  84%|████████▍ | 27/32 [02:52<00:31,  6.37s/it]
Test accuracy: 0.84:  84%|████████▍ | 27/32 [02:52<00:31,  6.37s/it]
Test accuracy: 0.83:  88%|████████▊ | 28/32 [02:59<00:26,  6.57s/it]
Test accuracy: 0.78:  88%|████████▊ | 28/32 [02:59<00:26,  6.57s/it]
Test accuracy: 0.82:  88%|████████▊ | 28/32 [02:59<00:26,  6.57s/it]
Test accuracy: 0.79:  88%|████████▊ | 28/32 [02:59<00:26,  6.57s/it]
Test accuracy: 0.84:  88%|████████▊ | 28/32 [02:59<00:26,  6.57s/it]
Test accuracy: 0.84:  88%|████████▊ | 28/32 [02:59<00:26,  6.57s/it]
Test accuracy: 0.85:  88%|████████▊ | 28/32 [02:59<00:26,  6.57s/it]
Test accuracy: 0.85:  88%|████████▊ | 28/32 [02:59<00:26,  6.57s/it]
Test accuracy: 0.82:  91%|█████████ | 29/32 [03:06<00:19,  6.51s/it]
Test accuracy: 0.79:  91%|█████████ | 29/32 [03:06<00:19,  6.51s/it]
Test accuracy: 0.85:  91%|█████████ | 29/32 [03:06<00:19,  6.51s/it]
Test accuracy: 0.81:  91%|█████████ | 29/32 [03:06<00:19,  6.51s/it]
Test accuracy: 0.79:  91%|█████████ | 29/32 [03:06<00:19,  6.51s/it]
Test accuracy: 0.85:  91%|█████████ | 29/32 [03:06<00:19,  6.51s/it]
Test accuracy: 0.84:  91%|█████████ | 29/32 [03:06<00:19,  6.51s/it]
Test accuracy: 0.84:  91%|█████████ | 29/32 [03:06<00:19,  6.51s/it]
Test accuracy: 0.79:  94%|█████████▍| 30/32 [03:12<00:12,  6.40s/it]
Test accuracy: 0.8:  94%|█████████▍| 30/32 [03:12<00:12,  6.40s/it] 
Test accuracy: 0.81:  94%|█████████▍| 30/32 [03:12<00:12,  6.40s/it]
Test accuracy: 0.82:  94%|█████████▍| 30/32 [03:12<00:12,  6.40s/it]
Test accuracy: 0.85:  94%|█████████▍| 30/32 [03:12<00:12,  6.40s/it]
Test accuracy: 0.84:  94%|█████████▍| 30/32 [03:12<00:12,  6.40s/it]
Test accuracy: 0.85:  94%|█████████▍| 30/32 [03:12<00:12,  6.40s/it]
Test accuracy: 0.85:  94%|█████████▍| 30/32 [03:12<00:12,  6.40s/it]
Test accuracy: 0.82:  97%|█████████▋| 31/32 [03:19<00:06,  6.60s/it]
Test accuracy: 0.8:  97%|█████████▋| 31/32 [03:19<00:06,  6.60s/it]
Test accuracy: 0.85:  97%|█████████▋| 31/32 [03:19<00:06,  6.60s/it]
Test accuracy: 0.82:  97%|█████████▋| 31/32 [03:19<00:06,  6.60s/it]
Test accuracy: 0.79:  97%|█████████▋| 31/32 [03:19<00:06,  6.60s/it]
Test accuracy: 0.84:  97%|█████████▋| 31/32 [03:19<00:06,  6.60s/it]
Test accuracy: 0.85:  97%|█████████▋| 31/32 [03:19<00:06,  6.60s/it]
Test accuracy: 0.85:  97%|█████████▋| 31/32 [03:19<00:06,  6.60s/it]
Test accuracy: 0.84: 100%|██████████| 32/32 [03:21<00:00,  5.29s/it]
Test accuracy: 0.84: 100%|██████████| 32/32 [03:21<00:00,  5.29s/it]
Test accuracy: 0.85: 100%|██████████| 32/32 [03:21<00:00,  5.29s/it]
Test accuracy: 0.79: 100%|██████████| 32/32 [03:21<00:00,  5.29s/it]
Test accuracy: 0.82: 100%|██████████| 32/32 [03:21<00:00,  5.29s/it]
Test accuracy: 0.85: 100%|██████████| 32/32 [03:21<00:00,  5.29s/it]
Test accuracy: 0.8: 100%|██████████| 32/32 [03:21<00:00,  5.29s/it] 
Test accuracy: 0.82: 100%|██████████| 32/32 [03:21<00:00,  5.29s/it]
Test accuracy: 0.85: 100%|██████████| 32/32 [03:21<00:00,  6.31s/it]
Device 2: Cor=212, CoT=0, Total=250

Test accuracy: 0.8: 100%|██████████| 32/32 [03:21<00:00,  6.31s/it]
Device 3: Cor=199, CoT=0, Total=250

Test accuracy: 0.82: 100%|██████████| 32/32 [03:21<00:00,  6.31s/it]
Device 1: Cor=206, CoT=0, Total=250

Test accuracy: 0.84: 100%|██████████| 32/32 [03:21<00:00,  6.31s/it]
Device 0: Cor=210, CoT=0, Total=250

============================================================
EVALUATION RESULTS (Epoch 13)
============================================================
Config: data_curriculum=ON, thought_curriculum=OFF
thought_stage=10, data_stage=7, n_latent_tokens=10
Training on difficulties: 0-7

Overall accuracy: 827 / 1000 = 0.8270

Per-difficulty accuracy:
  diff=1: 99/100 = 0.9900 (trained)
  diff=2: 99/100 = 0.9900 (trained)
  diff=3: 97/100 = 0.9700 (trained)
  diff=4: 98/100 = 0.9800 (trained)
  diff=5: 98/100 = 0.9800 (trained)
  diff=6: 96/100 = 0.9600 (trained)
  diff=7: 92/100 = 0.9200 (trained)
  diff=8: 49/100 = 0.4900 (untrained)
  diff=9: 53/100 = 0.5300 (untrained)
  diff=10: 46/100 = 0.4600 (untrained)

Trained (diff 0-7): 679/700 = 0.9700
============================================================


*** ADAPTIVE STAGING: acc_trained=0.9700, acc_diff7=0.9200 >= 0.9 ***
*** Advancing to stage 8 ***


Processing Files (0 / 0)      : |          |  0.00B /  0.00B            

New Data Upload               : |          |  0.00B /  0.00B            


  ...ve-03b89ee9/checkpoint_14:  10%|▉         |  360MB / 3.63GB            


  ...ve-03b89ee9/checkpoint_14:  10%|▉         |  360MB / 3.63GB            
Processing Files (0 / 1)      :  10%|▉         |  360MB / 3.63GB,   ???B/s  


  ...ve-03b89ee9/checkpoint_14:  20%|██        |  736MB / 3.63GB            
Processing Files (0 / 1)      :  20%|██        |  736MB / 3.63GB,   ???B/s  


  ...ve-03b89ee9/checkpoint_14:  31%|███       | 1.11GB / 3.63GB            
Processing Files (0 / 1)      :  31%|███       | 1.11GB / 3.63GB,  107MB/s  


  ...ve-03b89ee9/checkpoint_14:  41%|████      | 1.49GB / 3.63GB            
Processing Files (0 / 1)      :  41%|████      | 1.49GB / 3.63GB,  141MB/s  


  ...ve-03b89ee9/checkpoint_14:  51%|█████     | 1.86GB / 3.63GB            
Processing Files (0 / 1)      :  51%|█████     | 1.86GB / 3.63GB,  173MB/s  


  ...ve-03b89ee9/checkpoint_14:  61%|██████▏   | 2.22GB / 3.63GB            
Processing Files (0 / 1)      :  61%|██████▏   | 2.22GB / 3.63GB,  204MB/s  


  ...ve-03b89ee9/checkpoint_14:  72%|███████▏  | 2.60GB / 3.63GB            
Processing Files (0 / 1)      :  72%|███████▏  | 2.60GB / 3.63GB,  235MB/s  


  ...ve-03b89ee9/checkpoint_14:  82%|████████▏ | 2.97GB / 3.63GB            
Processing Files (0 / 1)      :  82%|████████▏ | 2.97GB / 3.63GB,  264MB/s  


  ...ve-03b89ee9/checkpoint_14:  92%|█████████▏| 3.34GB / 3.63GB            
Processing Files (0 / 1)      :  92%|█████████▏| 3.34GB / 3.63GB,  293MB/s  


  ...ve-03b89ee9/checkpoint_14: 100%|██████████| 3.63GB / 3.63GB            
Processing Files (1 / 1)      : 100%|██████████| 3.63GB / 3.63GB,  313MB/s  
Processing Files (1 / 1)      : 100%|██████████| 3.63GB / 3.63GB,  313MB/s  

New Data Upload               : |          |  0.00B /  0.00B,  0.00B/s  

  ...ve-03b89ee9/checkpoint_14: 100%|██████████| 3.63GB / 3.63GB            
No files have been modified since last commit. Skipping to prevent empty commit.
Uploaded checkpoint_14 to seyedparsa/thoughtformer-checkpoints/checkpoints/star-datacurr-only-4gpu-bs32-lr1e-5_n1000_qwen3-0.6b-base_thoughtformer_lr1e-5_no-reset_dc_no-tc_adaptive-03b89ee9/checkpoint_14

============================================================
EPOCH 14/100 (thought_stage=10, data_stage=8)
============================================================
  thought_stage=10, c_thought=1, max_difficulty=8

Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):  16%|█▌        | 156/1000 [00:00<00:00, 964.40 examples/s]
Map (num_proc=4):  16%|█▌        | 156/1000 [00:00<00:00, 924.22 examples/s]
Map (num_proc=4):  16%|█▌        | 156/1000 [00:00<00:00, 908.74 examples/s]
Map (num_proc=4):  81%|████████▏ | 813/1000 [00:00<00:00, 3446.19 examples/s]
Map (num_proc=4):  81%|████████  | 811/1000 [00:00<00:00, 3362.74 examples/s]
Map (num_proc=4):  81%|████████  | 812/1000 [00:00<00:00, 2932.66 examples/s]
Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2490.39 examples/s]

Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2405.88 examples/s]

Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2404.15 examples/s]

Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Map (num_proc=4):  16%|█▌        | 157/1000 [00:00<00:01, 541.44 examples/s]
Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Map (num_proc=4):  81%|████████  | 811/1000 [00:00<00:00, 2454.07 examples/s]
Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 1311.63 examples/s]

Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:04, 1869.08 examples/s]
Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:04, 1881.91 examples/s]
Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:05, 1785.77 examples/s]
Filter (num_proc=4):  20%|██        | 2000/10000 [00:00<00:02, 3275.76 examples/s]
Filter (num_proc=4):  20%|██        | 2000/10000 [00:00<00:02, 3315.59 examples/s]
Filter (num_proc=4):  20%|██        | 2000/10000 [00:00<00:02, 3125.60 examples/s]
Filter (num_proc=4):  30%|███       | 3000/10000 [00:00<00:01, 4716.49 examples/s]
Filter (num_proc=4):  40%|████      | 4000/10000 [00:00<00:00, 6314.45 examples/s]
Filter (num_proc=4):  40%|████      | 4000/10000 [00:00<00:00, 6382.40 examples/s]
Filter (num_proc=4):  45%|████▌     | 4500/10000 [00:00<00:00, 6841.18 examples/s]
Filter (num_proc=4):  65%|██████▌   | 6500/10000 [00:00<00:00, 9680.77 examples/s]
Filter (num_proc=4):  55%|█████▌    | 5500/10000 [00:00<00:00, 8017.49 examples/s]
Filter (num_proc=4):  80%|████████  | 8000/10000 [00:01<00:00, 10760.83 examples/s]
Filter (num_proc=4):  70%|███████   | 7000/10000 [00:01<00:00, 10229.61 examples/s]
Filter (num_proc=4):  70%|███████   | 7000/10000 [00:01<00:00, 9259.96 examples/s]
Filter (num_proc=4):  95%|█████████▌| 9500/10000 [00:01<00:00, 10791.18 examples/s]
Filter (num_proc=4):  85%|████████▌ | 8500/10000 [00:01<00:00, 10729.32 examples/s]
Filter (num_proc=4):  90%|█████████ | 9000/10000 [00:01<00:00, 11240.14 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 10774.33 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 6867.57 examples/s]

Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 6882.44 examples/s] 

Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 6898.71 examples/s]

Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:06, 1330.17 examples/s]
Filter (num_proc=4):  30%|███       | 3000/10000 [00:00<00:01, 3587.83 examples/s]
Filter (num_proc=4):  55%|█████▌    | 5500/10000 [00:01<00:00, 6642.22 examples/s]
Filter (num_proc=4):  80%|████████  | 8000/10000 [00:01<00:00, 9892.80 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 8980.70 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 5663.90 examples/s]

Map (num_proc=4):   0%|          | 0/8000 [00:00<?, ? examples/s]
Map (num_proc=4):   2%|▏         | 138/8000 [00:00<00:55, 142.20 examples/s]
Map (num_proc=4):   7%|▋         | 547/8000 [00:01<00:12, 609.80 examples/s]
Map (num_proc=4):  10%|█         | 817/8000 [00:01<00:07, 915.76 examples/s]
Map (num_proc=4):  14%|█▎        | 1093/8000 [00:01<00:07, 877.55 examples/s]
Map (num_proc=4):  21%|██▏       | 1719/8000 [00:01<00:03, 1639.89 examples/s]
Map (num_proc=4):  29%|██▉       | 2311/8000 [00:01<00:02, 2377.04 examples/s]
Map (num_proc=4):  36%|███▌      | 2874/8000 [00:01<00:01, 3017.95 examples/s]
Map (num_proc=4):  43%|████▎     | 3434/8000 [00:02<00:01, 2312.62 examples/s]
Map (num_proc=4):  52%|█████▏    | 4155/8000 [00:02<00:01, 2816.47 examples/s]
Map (num_proc=4):  58%|█████▊    | 4625/8000 [00:02<00:01, 2331.13 examples/s]
Map (num_proc=4):  66%|██████▌   | 5265/8000 [00:02<00:00, 2852.32 examples/s]
Map (num_proc=4):  73%|███████▎  | 5847/8000 [00:02<00:00, 3367.06 examples/s]
Map (num_proc=4):  80%|███████▉  | 6381/8000 [00:03<00:00, 3544.22 examples/s]
Map (num_proc=4):  86%|████████▌ | 6872/8000 [00:03<00:00, 3168.21 examples/s]
Map (num_proc=4):  91%|█████████▏| 7300/8000 [00:03<00:00, 2792.13 examples/s]
Map (num_proc=4):  96%|█████████▌| 7656/8000 [00:03<00:00, 2936.71 examples/s]
Map (num_proc=4): 100%|██████████| 8000/8000 [00:03<00:00, 1843.91 examples/s]
Map (num_proc=4): 100%|██████████| 8000/8000 [00:04<00:00, 1726.04 examples/s]

Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 1386.25 examples/s]
Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 1371.74 examples/s]
Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 1318.75 examples/s]
Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2639.51 examples/s]

Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 780.44 examples/s]
Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2638.26 examples/s]

Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2508.67 examples/s]

Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 1471.52 examples/s]

Map (num_proc=4):   0%|          | 0/800 [00:00<?, ? examples/s]
Map (num_proc=4):  17%|█▋        | 136/800 [00:00<00:01, 381.27 examples/s]
Map (num_proc=4):  42%|████▏     | 332/800 [00:00<00:00, 830.38 examples/s]
Map (num_proc=4):  67%|██████▋   | 533/800 [00:00<00:00, 1117.70 examples/s]
Map (num_proc=4): 100%|██████████| 800/800 [00:00<00:00, 1267.67 examples/s]
Map (num_proc=4): 100%|██████████| 800/800 [00:01<00:00, 726.99 examples/s] 

Training Epoch: 15:   0%|          | 0/250 [00:00<?, ?it/s]
Training Epoch: 15:   0%|          | 0/250 [00:00<?, ?it/s]
Training Epoch: 15:   0%|          | 0/250 [00:00<?, ?it/s]
Training Epoch: 15:   0%|          | 0/250 [00:00<?, ?it/s]logging training data
rank 0 | gpu 0 | alloc=4.20 GB | reserved=51.39 GB | max=41.99 GB
rank 2 | gpu 2 | alloc=4.20 GB | reserved=51.39 GB | max=41.98 GB
rank 3 | gpu 3 | alloc=4.20 GB | reserved=49.56 GB | max=41.99 GB
rank 1 | gpu 1 | alloc=4.20 GB | reserved=53.36 GB | max=41.98 GB

Training Epoch: 15/100, batch 0/500 completed (loss: 0.4348:   0%|          | 0/250 [00:03<?, ?it/s]
Training Epoch: 15/100, batch 0/500 completed (loss: 0.0507:   0%|          | 0/250 [00:03<?, ?it/s]
Training Epoch: 15/100, batch 0/500 completed (loss: 0.0565:   0%|          | 0/250 [00:03<?, ?it/s]
Training Epoch: 15/100, batch 0/500 completed (loss: 0.1336:   0%|          | 0/250 [00:03<?, ?it/s]
Training Epoch: 15/100, batch 0/500 completed (loss: 0.0507:   0%|          | 1/250 [00:06<25:50,  6.23s/it]
Training Epoch: 15/100, batch 0/500 completed (loss: 0.0565:   0%|          | 1/250 [00:06<25:50,  6.23s/it]
Training Epoch: 15/100, batch 0/500 completed (loss: 0.1336:   0%|          | 1/250 [00:06<25:59,  6.26s/it]
Training Epoch: 15/100, batch 0/500 completed (loss: 0.4348:   0%|          | 1/250 [00:06<25:47,  6.22s/it]
Training Epoch: 15/100, batch 1/500 completed (loss: 0.0275:   0%|          | 1/250 [00:06<25:50,  6.23s/it]
Training Epoch: 15/100, batch 1/500 completed (loss: 0.0599:   0%|          | 1/250 [00:06<25:50,  6.23s/it]
Training Epoch: 15/100, batch 1/500 completed (loss: 0.025:   0%|          | 1/250 [00:06<25:47,  6.22s/it] 
Training Epoch: 15/100, batch 1/500 completed (loss: 0.0852:   0%|          | 1/250 [00:06<25:59,  6.26s/it]
Training Epoch: 15/100, batch 2/500 completed (loss: 0.0262:   0%|          | 1/250 [00:09<25:50,  6.23s/it]
Training Epoch: 15/100, batch 2/500 completed (loss: 0.1149:   0%|          | 1/250 [00:09<25:50,  6.23s/it]
Training Epoch: 15/100, batch 2/500 completed (loss: 0.0425:   0%|          | 1/250 [00:09<25:47,  6.22s/it]
Training Epoch: 15/100, batch 2/500 completed (loss: 0.0633:   0%|          | 1/250 [00:09<25:59,  6.26s/it]
Training Epoch: 15/100, batch 2/500 completed (loss: 0.0425:   1%|          | 2/250 [00:12<24:49,  6.01s/it]
Training Epoch: 15/100, batch 2/500 completed (loss: 0.0633:   1%|          | 2/250 [00:12<24:55,  6.03s/it]
Training Epoch: 15/100, batch 2/500 completed (loss: 0.1149:   1%|          | 2/250 [00:12<24:52,  6.02s/it]
Training Epoch: 15/100, batch 2/500 completed (loss: 0.0262:   1%|          | 2/250 [00:12<24:52,  6.02s/it]
Training Epoch: 15/100, batch 3/500 completed (loss: 0.0166:   1%|          | 2/250 [00:12<24:52,  6.02s/it]
Training Epoch: 15/100, batch 3/500 completed (loss: 0.8441:   1%|          | 2/250 [00:12<24:52,  6.02s/it]
Training Epoch: 15/100, batch 3/500 completed (loss: 0.2818:   1%|          | 2/250 [00:12<24:49,  6.01s/it]
Training Epoch: 15/100, batch 3/500 completed (loss: 0.1514:   1%|          | 2/250 [00:12<24:55,  6.03s/it]
Training Epoch: 15/100, batch 4/500 completed (loss: 0.0985:   1%|          | 2/250 [00:14<24:49,  6.01s/it]
Training Epoch: 15/100, batch 4/500 completed (loss: 0.0133:   1%|          | 2/250 [00:14<24:52,  6.02s/it]
Training Epoch: 15/100, batch 4/500 completed (loss: 0.0219:   1%|          | 2/250 [00:14<24:52,  6.02s/it]
Training Epoch: 15/100, batch 4/500 completed (loss: 0.0203:   1%|          | 2/250 [00:14<24:55,  6.03s/it]
Training Epoch: 15/100, batch 4/500 completed (loss: 0.0219:   1%|          | 3/250 [00:17<24:20,  5.91s/it]
Training Epoch: 15/100, batch 4/500 completed (loss: 0.0133:   1%|          | 3/250 [00:17<24:20,  5.91s/it]
Training Epoch: 15/100, batch 4/500 completed (loss: 0.0985:   1%|          | 3/250 [00:17<24:20,  5.91s/it]
Training Epoch: 15/100, batch 4/500 completed (loss: 0.0203:   1%|          | 3/250 [00:17<24:23,  5.92s/it]
Training Epoch: 15/100, batch 5/500 completed (loss: 0.0164:   1%|          | 3/250 [00:17<24:20,  5.91s/it]
Training Epoch: 15/100, batch 5/500 completed (loss: 0.0073:   1%|          | 3/250 [00:17<24:20,  5.91s/it]
Training Epoch: 15/100, batch 5/500 completed (loss: 0.0345:   1%|          | 3/250 [00:17<24:20,  5.91s/it]
Training Epoch: 15/100, batch 5/500 completed (loss: 0.1948:   1%|          | 3/250 [00:17<24:23,  5.92s/it]
Training Epoch: 15/100, batch 6/500 completed (loss: 0.0204:   1%|          | 3/250 [00:20<24:20,  5.91s/it]
Training Epoch: 15/100, batch 6/500 completed (loss: 0.0058:   1%|          | 3/250 [00:20<24:20,  5.91s/it]
Training Epoch: 15/100, batch 6/500 completed (loss: 0.007:   1%|          | 3/250 [00:20<24:20,  5.91s/it] 
Training Epoch: 15/100, batch 6/500 completed (loss: 0.0044:   1%|          | 3/250 [00:20<24:23,  5.92s/it]
Training Epoch: 15/100, batch 6/500 completed (loss: 0.0058:   2%|▏         | 4/250 [00:23<23:56,  5.84s/it]
Training Epoch: 15/100, batch 6/500 completed (loss: 0.0044:   2%|▏         | 4/250 [00:23<23:58,  5.85s/it]
Training Epoch: 15/100, batch 6/500 completed (loss: 0.007:   2%|▏         | 4/250 [00:23<23:56,  5.84s/it]
Training Epoch: 15/100, batch 6/500 completed (loss: 0.0204:   2%|▏         | 4/250 [00:23<23:57,  5.84s/it]
Training Epoch: 15/100, batch 7/500 completed (loss: 0.0481:   2%|▏         | 4/250 [00:23<23:56,  5.84s/it]
Training Epoch: 15/100, batch 7/500 completed (loss: 0.0397:   2%|▏         | 4/250 [00:23<23:56,  5.84s/it]
Training Epoch: 15/100, batch 7/500 completed (loss: 0.0289:   2%|▏         | 4/250 [00:23<23:57,  5.84s/it]
Training Epoch: 15/100, batch 7/500 completed (loss: 0.0476:   2%|▏         | 4/250 [00:23<23:58,  5.85s/it]
Training Epoch: 15/100, batch 8/500 completed (loss: 0.0249:   2%|▏         | 4/250 [00:26<23:56,  5.84s/it]
Training Epoch: 15/100, batch 8/500 completed (loss: 0.0395:   2%|▏         | 4/250 [00:26<23:56,  5.84s/it]
Training Epoch: 15/100, batch 8/500 completed (loss: 0.0379:   2%|▏         | 4/250 [00:26<23:57,  5.84s/it]
Training Epoch: 15/100, batch 8/500 completed (loss: 0.1231:   2%|▏         | 4/250 [00:26<23:58,  5.85s/it]
Training Epoch: 15/100, batch 8/500 completed (loss: 0.0249:   2%|▏         | 5/250 [00:29<23:47,  5.83s/it]
Training Epoch: 15/100, batch 8/500 completed (loss: 0.0395:   2%|▏         | 5/250 [00:29<23:48,  5.83s/it]
Training Epoch: 15/100, batch 8/500 completed (loss: 0.1231:   2%|▏         | 5/250 [00:29<23:48,  5.83s/it]
Training Epoch: 15/100, batch 8/500 completed (loss: 0.0379:   2%|▏         | 5/250 [00:29<23:48,  5.83s/it]
Training Epoch: 15/100, batch 9/500 completed (loss: 0.0278:   2%|▏         | 5/250 [00:29<23:48,  5.83s/it]
Training Epoch: 15/100, batch 9/500 completed (loss: 0.1405:   2%|▏         | 5/250 [00:29<23:47,  5.83s/it]
Training Epoch: 15/100, batch 9/500 completed (loss: 0.1933:   2%|▏         | 5/250 [00:29<23:48,  5.83s/it]
Training Epoch: 15/100, batch 9/500 completed (loss: 0.0043:   2%|▏         | 5/250 [00:29<23:48,  5.83s/it]
Training Epoch: 15/100, batch 10/500 completed (loss: 0.0253:   2%|▏         | 5/250 [00:32<23:48,  5.83s/it]
Training Epoch: 15/100, batch 10/500 completed (loss: 0.0041:   2%|▏         | 5/250 [00:32<23:47,  5.83s/it]
Training Epoch: 15/100, batch 10/500 completed (loss: 0.1864:   2%|▏         | 5/250 [00:32<23:48,  5.83s/it]
Training Epoch: 15/100, batch 10/500 completed (loss: 0.0411:   2%|▏         | 5/250 [00:32<23:48,  5.83s/it]
Training Epoch: 15/100, batch 10/500 completed (loss: 0.0253:   2%|▏         | 6/250 [00:35<23:57,  5.89s/it]
Training Epoch: 15/100, batch 10/500 completed (loss: 0.1864:   2%|▏         | 6/250 [00:35<23:57,  5.89s/it]
Training Epoch: 15/100, batch 10/500 completed (loss: 0.0411:   2%|▏         | 6/250 [00:35<23:58,  5.90s/it]
Training Epoch: 15/100, batch 10/500 completed (loss: 0.0041:   2%|▏         | 6/250 [00:35<23:58,  5.89s/it]
Training Epoch: 15/100, batch 11/500 completed (loss: 0.0273:   2%|▏         | 6/250 [00:35<23:57,  5.89s/it]
Training Epoch: 15/100, batch 11/500 completed (loss: 0.0264:   2%|▏         | 6/250 [00:35<23:57,  5.89s/it]
Training Epoch: 15/100, batch 11/500 completed (loss: 0.0093:   2%|▏         | 6/250 [00:35<23:58,  5.89s/it]
Training Epoch: 15/100, batch 11/500 completed (loss: 0.0066:   2%|▏         | 6/250 [00:35<23:58,  5.90s/it]
Training Epoch: 15/100, batch 12/500 completed (loss: 0.0241:   2%|▏         | 6/250 [00:38<23:58,  5.89s/it]
Training Epoch: 15/100, batch 12/500 completed (loss: 0.0619:   2%|▏         | 6/250 [00:38<23:57,  5.89s/it]
Training Epoch: 15/100, batch 12/500 completed (loss: 0.5553:   2%|▏         | 6/250 [00:38<23:57,  5.89s/it]
Training Epoch: 15/100, batch 12/500 completed (loss: 0.349:   2%|▏         | 6/250 [00:38<23:58,  5.90s/it] 
Training Epoch: 15/100, batch 12/500 completed (loss: 0.0241:   3%|▎         | 7/250 [00:41<23:35,  5.82s/it]
Training Epoch: 15/100, batch 12/500 completed (loss: 0.349:   3%|▎         | 7/250 [00:41<23:35,  5.83s/it]
Training Epoch: 15/100, batch 12/500 completed (loss: 0.5553:   3%|▎         | 7/250 [00:41<23:35,  5.83s/it]
Training Epoch: 15/100, batch 12/500 completed (loss: 0.0619:   3%|▎         | 7/250 [00:41<23:36,  5.83s/it]
Training Epoch: 15/100, batch 13/500 completed (loss: 0.0611:   3%|▎         | 7/250 [00:41<23:35,  5.83s/it]
Training Epoch: 15/100, batch 13/500 completed (loss: 0.0463:   3%|▎         | 7/250 [00:41<23:36,  5.83s/it]
Training Epoch: 15/100, batch 13/500 completed (loss: 0.0628:   3%|▎         | 7/250 [00:41<23:35,  5.82s/it]
Training Epoch: 15/100, batch 13/500 completed (loss: 0.5405:   3%|▎         | 7/250 [00:41<23:35,  5.83s/it]
Training Epoch: 15/100, batch 14/500 completed (loss: 0.0032:   3%|▎         | 7/250 [00:44<23:35,  5.82s/it]
Training Epoch: 15/100, batch 14/500 completed (loss: 0.0169:   3%|▎         | 7/250 [00:44<23:35,  5.83s/it]
Training Epoch: 15/100, batch 14/500 completed (loss: 0.0878:   3%|▎         | 7/250 [00:44<23:36,  5.83s/it]
Training Epoch: 15/100, batch 14/500 completed (loss: 0.0125:   3%|▎         | 7/250 [00:44<23:35,  5.83s/it]
Training Epoch: 15/100, batch 14/500 completed (loss: 0.0125:   3%|▎         | 8/250 [00:47<23:32,  5.84s/it]
Training Epoch: 15/100, batch 14/500 completed (loss: 0.0878:   3%|▎         | 8/250 [00:46<23:32,  5.84s/it]
Training Epoch: 15/100, batch 14/500 completed (loss: 0.0032:   3%|▎         | 8/250 [00:46<23:32,  5.84s/it]
Training Epoch: 15/100, batch 14/500 completed (loss: 0.0169:   3%|▎         | 8/250 [00:47<23:33,  5.84s/it]
Training Epoch: 15/100, batch 15/500 completed (loss: 0.1212:   3%|▎         | 8/250 [00:47<23:32,  5.84s/it]
Training Epoch: 15/100, batch 15/500 completed (loss: 0.0156:   3%|▎         | 8/250 [00:47<23:33,  5.84s/it]
Training Epoch: 15/100, batch 15/500 completed (loss: 0.1792:   3%|▎         | 8/250 [00:47<23:32,  5.84s/it]
Training Epoch: 15/100, batch 15/500 completed (loss: 0.1456:   3%|▎         | 8/250 [00:47<23:32,  5.84s/it]
Training Epoch: 15/100, batch 16/500 completed (loss: 0.0852:   3%|▎         | 8/250 [00:50<23:32,  5.84s/it]
Training Epoch: 15/100, batch 16/500 completed (loss: 0.0603:   3%|▎         | 8/250 [00:50<23:32,  5.84s/it]
Training Epoch: 15/100, batch 16/500 completed (loss: 0.0384:   3%|▎         | 8/250 [00:50<23:33,  5.84s/it]
Training Epoch: 15/100, batch 16/500 completed (loss: 0.4056:   3%|▎         | 8/250 [00:50<23:32,  5.84s/it]
Training Epoch: 15/100, batch 16/500 completed (loss: 0.4056:   4%|▎         | 9/250 [00:53<23:49,  5.93s/it]
Training Epoch: 15/100, batch 16/500 completed (loss: 0.0384:   4%|▎         | 9/250 [00:53<23:49,  5.93s/it]
Training Epoch: 15/100, batch 16/500 completed (loss: 0.0603:   4%|▎         | 9/250 [00:53<23:49,  5.93s/it]
Training Epoch: 15/100, batch 16/500 completed (loss: 0.0852:   4%|▎         | 9/250 [00:53<23:50,  5.94s/it]
Training Epoch: 15/100, batch 17/500 completed (loss: 0.3572:   4%|▎         | 9/250 [00:53<23:49,  5.93s/it]
Training Epoch: 15/100, batch 17/500 completed (loss: 0.0173:   4%|▎         | 9/250 [00:53<23:49,  5.93s/it]
Training Epoch: 15/100, batch 17/500 completed (loss: 0.1897:   4%|▎         | 9/250 [00:53<23:50,  5.94s/it]
Training Epoch: 15/100, batch 17/500 completed (loss: 0.0188:   4%|▎         | 9/250 [00:53<23:49,  5.93s/it]
Training Epoch: 15/100, batch 18/500 completed (loss: 0.0373:   4%|▎         | 9/250 [00:56<23:49,  5.93s/it]
Training Epoch: 15/100, batch 18/500 completed (loss: 0.0109:   4%|▎         | 9/250 [00:56<23:50,  5.94s/it]
Training Epoch: 15/100, batch 18/500 completed (loss: 0.0513:   4%|▎         | 9/250 [00:56<23:49,  5.93s/it]
Training Epoch: 15/100, batch 18/500 completed (loss: 0.0251:   4%|▎         | 9/250 [00:56<23:49,  5.93s/it]
Training Epoch: 15/100, batch 18/500 completed (loss: 0.0251:   4%|▍         | 10/250 [00:59<23:39,  5.92s/it]
Training Epoch: 15/100, batch 18/500 completed (loss: 0.0373:   4%|▍         | 10/250 [00:59<23:39,  5.91s/it]
Training Epoch: 15/100, batch 18/500 completed (loss: 0.0109:   4%|▍         | 10/250 [00:58<23:39,  5.91s/it]
Training Epoch: 15/100, batch 18/500 completed (loss: 0.0513:   4%|▍         | 10/250 [00:59<23:39,  5.92s/it]
Training Epoch: 15/100, batch 19/500 completed (loss: 0.1362:   4%|▍         | 10/250 [00:59<23:39,  5.92s/it]
Training Epoch: 15/100, batch 19/500 completed (loss: 0.0347:   4%|▍         | 10/250 [00:59<23:39,  5.91s/it]
Training Epoch: 15/100, batch 19/500 completed (loss: 0.0081:   4%|▍         | 10/250 [00:59<23:39,  5.91s/it]
Training Epoch: 15/100, batch 19/500 completed (loss: 0.0194:   4%|▍         | 10/250 [00:59<23:39,  5.92s/it]
Training Epoch: 15/100, batch 20/500 completed (loss: 0.0418:   4%|▍         | 10/250 [01:01<23:39,  5.91s/it]
Training Epoch: 15/100, batch 20/500 completed (loss: 0.0177:   4%|▍         | 10/250 [01:01<23:39,  5.92s/it]
Training Epoch: 15/100, batch 20/500 completed (loss: 0.0209:   4%|▍         | 10/250 [01:01<23:39,  5.91s/it]
Training Epoch: 15/100, batch 20/500 completed (loss: 0.2269:   4%|▍         | 10/250 [01:01<23:39,  5.92s/it]
Training Epoch: 15/100, batch 20/500 completed (loss: 0.2269:   4%|▍         | 11/250 [01:05<23:48,  5.98s/it]
Training Epoch: 15/100, batch 20/500 completed (loss: 0.0418:   4%|▍         | 11/250 [01:05<23:48,  5.98s/it]
Training Epoch: 15/100, batch 20/500 completed (loss: 0.0177:   4%|▍         | 11/250 [01:05<23:48,  5.98s/it]
Training Epoch: 15/100, batch 20/500 completed (loss: 0.0209:   4%|▍         | 11/250 [01:05<23:48,  5.98s/it]
Training Epoch: 15/100, batch 21/500 completed (loss: 0.0648:   4%|▍         | 11/250 [01:05<23:48,  5.98s/it]
Training Epoch: 15/100, batch 21/500 completed (loss: 0.0165:   4%|▍         | 11/250 [01:05<23:48,  5.98s/it]
Training Epoch: 15/100, batch 21/500 completed (loss: 0.0024:   4%|▍         | 11/250 [01:05<23:48,  5.98s/it]
Training Epoch: 15/100, batch 21/500 completed (loss: 0.0064:   4%|▍         | 11/250 [01:05<23:48,  5.98s/it]
Training Epoch: 15/100, batch 22/500 completed (loss: 0.0693:   4%|▍         | 11/250 [01:08<23:48,  5.98s/it]
Training Epoch: 15/100, batch 22/500 completed (loss: 0.0363:   4%|▍         | 11/250 [01:08<23:48,  5.98s/it]
Training Epoch: 15/100, batch 22/500 completed (loss: 0.0146:   4%|▍         | 11/250 [01:08<23:48,  5.98s/it]
Training Epoch: 15/100, batch 22/500 completed (loss: 0.1037:   4%|▍         | 11/250 [01:08<23:48,  5.98s/it]
Training Epoch: 15/100, batch 22/500 completed (loss: 0.0146:   5%|▍         | 12/250 [01:10<23:22,  5.89s/it]
Training Epoch: 15/100, batch 22/500 completed (loss: 0.1037:   5%|▍         | 12/250 [01:10<23:23,  5.90s/it]
Training Epoch: 15/100, batch 22/500 completed (loss: 0.0363:   5%|▍         | 12/250 [01:10<23:22,  5.89s/it]
Training Epoch: 15/100, batch 22/500 completed (loss: 0.0693:   5%|▍         | 12/250 [01:10<23:23,  5.90s/it]
Training Epoch: 15/100, batch 23/500 completed (loss: 0.0445:   5%|▍         | 12/250 [01:10<23:22,  5.89s/it]
Training Epoch: 15/100, batch 23/500 completed (loss: 0.092:   5%|▍         | 12/250 [01:10<23:23,  5.90s/it] 
Training Epoch: 15/100, batch 23/500 completed (loss: 0.016:   5%|▍         | 12/250 [01:10<23:22,  5.89s/it] 
Training Epoch: 15/100, batch 23/500 completed (loss: 0.01:   5%|▍         | 12/250 [01:10<23:23,  5.90s/it]  
Training Epoch: 15/100, batch 24/500 completed (loss: 0.0118:   5%|▍         | 12/250 [01:13<23:22,  5.89s/it]
Training Epoch: 15/100, batch 24/500 completed (loss: 0.0852:   5%|▍         | 12/250 [01:13<23:22,  5.89s/it]
Training Epoch: 15/100, batch 24/500 completed (loss: 0.0217:   5%|▍         | 12/250 [01:13<23:23,  5.90s/it]
Training Epoch: 15/100, batch 24/500 completed (loss: 0.0072:   5%|▍         | 12/250 [01:13<23:23,  5.90s/it]
Training Epoch: 15/100, batch 24/500 completed (loss: 0.0118:   5%|▌         | 13/250 [01:16<23:17,  5.90s/it]
Training Epoch: 15/100, batch 24/500 completed (loss: 0.0072:   5%|▌         | 13/250 [01:16<23:18,  5.90s/it]
Training Epoch: 15/100, batch 24/500 completed (loss: 0.0217:   5%|▌         | 13/250 [01:16<23:17,  5.90s/it]
Training Epoch: 15/100, batch 24/500 completed (loss: 0.0852:   5%|▌         | 13/250 [01:16<23:18,  5.90s/it]
Training Epoch: 15/100, batch 25/500 completed (loss: 0.1266:   5%|▌         | 13/250 [01:16<23:18,  5.90s/it]
Training Epoch: 15/100, batch 25/500 completed (loss: 0.0175:   5%|▌         | 13/250 [01:16<23:17,  5.90s/it]
Training Epoch: 15/100, batch 25/500 completed (loss: 0.0376:   5%|▌         | 13/250 [01:16<23:17,  5.90s/it]
Training Epoch: 15/100, batch 25/500 completed (loss: 0.0009:   5%|▌         | 13/250 [01:16<23:18,  5.90s/it]
Training Epoch: 15/100, batch 26/500 completed (loss: 0.1327:   5%|▌         | 13/250 [01:20<23:17,  5.90s/it]
Training Epoch: 15/100, batch 26/500 completed (loss: 0.0097:   5%|▌         | 13/250 [01:20<23:18,  5.90s/it]
Training Epoch: 15/100, batch 26/500 completed (loss: 0.2231:   5%|▌         | 13/250 [01:20<23:17,  5.90s/it]
Training Epoch: 15/100, batch 26/500 completed (loss: 0.0076:   5%|▌         | 13/250 [01:20<23:18,  5.90s/it]
Training Epoch: 15/100, batch 26/500 completed (loss: 0.0076:   6%|▌         | 14/250 [01:23<23:35,  6.00s/it]
Training Epoch: 15/100, batch 26/500 completed (loss: 0.1327:   6%|▌         | 14/250 [01:22<23:36,  6.00s/it]
Training Epoch: 15/100, batch 26/500 completed (loss: 0.2231:   6%|▌         | 14/250 [01:22<23:36,  6.00s/it]
Training Epoch: 15/100, batch 26/500 completed (loss: 0.0097:   6%|▌         | 14/250 [01:23<23:37,  6.01s/it]
Training Epoch: 15/100, batch 27/500 completed (loss: 0.0243:   6%|▌         | 14/250 [01:23<23:36,  6.00s/it]
Training Epoch: 15/100, batch 27/500 completed (loss: 0.01:   6%|▌         | 14/250 [01:23<23:36,  6.00s/it]  
Training Epoch: 15/100, batch 27/500 completed (loss: 0.1157:   6%|▌         | 14/250 [01:23<23:37,  6.01s/it]
Training Epoch: 15/100, batch 27/500 completed (loss: 0.0375:   6%|▌         | 14/250 [01:23<23:35,  6.00s/it]
Training Epoch: 15/100, batch 28/500 completed (loss: 0.0178:   6%|▌         | 14/250 [01:25<23:36,  6.00s/it]
Training Epoch: 15/100, batch 28/500 completed (loss: 0.0186:   6%|▌         | 14/250 [01:25<23:37,  6.01s/it]
Training Epoch: 15/100, batch 28/500 completed (loss: 0.3234:   6%|▌         | 14/250 [01:25<23:36,  6.00s/it]
Training Epoch: 15/100, batch 28/500 completed (loss: 0.1801:   6%|▌         | 14/250 [01:25<23:35,  6.00s/it]
Training Epoch: 15/100, batch 28/500 completed (loss: 0.1801:   6%|▌         | 15/250 [01:28<23:13,  5.93s/it]
Training Epoch: 15/100, batch 28/500 completed (loss: 0.0186:   6%|▌         | 15/250 [01:28<23:13,  5.93s/it]
Training Epoch: 15/100, batch 28/500 completed (loss: 0.3234:   6%|▌         | 15/250 [01:28<23:13,  5.93s/it]
Training Epoch: 15/100, batch 28/500 completed (loss: 0.0178:   6%|▌         | 15/250 [01:28<23:13,  5.93s/it]
Training Epoch: 15/100, batch 29/500 completed (loss: 0.1081:   6%|▌         | 15/250 [01:28<23:13,  5.93s/it]
Training Epoch: 15/100, batch 29/500 completed (loss: 0.006:   6%|▌         | 15/250 [01:28<23:13,  5.93s/it] 
Training Epoch: 15/100, batch 29/500 completed (loss: 0.0372:   6%|▌         | 15/250 [01:28<23:13,  5.93s/it]
Training Epoch: 15/100, batch 29/500 completed (loss: 0.0276:   6%|▌         | 15/250 [01:28<23:13,  5.93s/it]
Training Epoch: 15/100, batch 30/500 completed (loss: 0.0099:   6%|▌         | 15/250 [01:31<23:13,  5.93s/it]
Training Epoch: 15/100, batch 30/500 completed (loss: 0.0095:   6%|▌         | 15/250 [01:31<23:13,  5.93s/it]
Training Epoch: 15/100, batch 30/500 completed (loss: 0.025:   6%|▌         | 15/250 [01:31<23:13,  5.93s/it] 
Training Epoch: 15/100, batch 30/500 completed (loss: 0.0047:   6%|▌         | 15/250 [01:31<23:13,  5.93s/it]
Training Epoch: 15/100, batch 30/500 completed (loss: 0.025:   6%|▋         | 16/250 [01:34<23:19,  5.98s/it]
Training Epoch: 15/100, batch 30/500 completed (loss: 0.0099:   6%|▋         | 16/250 [01:34<23:19,  5.98s/it]
Training Epoch: 15/100, batch 30/500 completed (loss: 0.0047:   6%|▋         | 16/250 [01:34<23:19,  5.98s/it]
Training Epoch: 15/100, batch 30/500 completed (loss: 0.0095:   6%|▋         | 16/250 [01:34<23:19,  5.98s/it]
Training Epoch: 15/100, batch 31/500 completed (loss: 0.0093:   6%|▋         | 16/250 [01:34<23:19,  5.98s/it]
Training Epoch: 15/100, batch 31/500 completed (loss: 0.0547:   6%|▋         | 16/250 [01:34<23:19,  5.98s/it]
Training Epoch: 15/100, batch 31/500 completed (loss: 0.0608:   6%|▋         | 16/250 [01:34<23:19,  5.98s/it]
Training Epoch: 15/100, batch 31/500 completed (loss: 0.7555:   6%|▋         | 16/250 [01:34<23:19,  5.98s/it]
Training Epoch: 15/100, batch 32/500 completed (loss: 0.041:   6%|▋         | 16/250 [01:38<23:19,  5.98s/it] 
Training Epoch: 15/100, batch 32/500 completed (loss: 0.0342:   6%|▋         | 16/250 [01:38<23:19,  5.98s/it]
Training Epoch: 15/100, batch 32/500 completed (loss: 0.0204:   6%|▋         | 16/250 [01:38<23:19,  5.98s/it]
Training Epoch: 15/100, batch 32/500 completed (loss: 0.0281:   6%|▋         | 16/250 [01:38<23:19,  5.98s/it]
Training Epoch: 15/100, batch 32/500 completed (loss: 0.0204:   7%|▋         | 17/250 [01:40<23:18,  6.00s/it]
Training Epoch: 15/100, batch 32/500 completed (loss: 0.0342:   7%|▋         | 17/250 [01:40<23:18,  6.00s/it]
Training Epoch: 15/100, batch 32/500 completed (loss: 0.0281:   7%|▋         | 17/250 [01:40<23:19,  6.00s/it]
Training Epoch: 15/100, batch 32/500 completed (loss: 0.041:   7%|▋         | 17/250 [01:40<23:19,  6.01s/it]
Training Epoch: 15/100, batch 33/500 completed (loss: 0.1459:   7%|▋         | 17/250 [01:40<23:18,  6.00s/it]
Training Epoch: 15/100, batch 33/500 completed (loss: 0.016:   7%|▋         | 17/250 [01:40<23:19,  6.01s/it]
Training Epoch: 15/100, batch 33/500 completed (loss: 0.0296:   7%|▋         | 17/250 [01:40<23:18,  6.00s/it]
Training Epoch: 15/100, batch 33/500 completed (loss: 0.0418:   7%|▋         | 17/250 [01:40<23:19,  6.00s/it]
Training Epoch: 15/100, batch 34/500 completed (loss: 0.0984:   7%|▋         | 17/250 [01:43<23:19,  6.01s/it]
Training Epoch: 15/100, batch 34/500 completed (loss: 0.0227:   7%|▋         | 17/250 [01:43<23:18,  6.00s/it]
Training Epoch: 15/100, batch 34/500 completed (loss: 0.0195:   7%|▋         | 17/250 [01:43<23:18,  6.00s/it]
Training Epoch: 15/100, batch 34/500 completed (loss: 0.0049:   7%|▋         | 17/250 [01:43<23:19,  6.00s/it]
Training Epoch: 15/100, batch 34/500 completed (loss: 0.0227:   7%|▋         | 18/250 [01:46<22:55,  5.93s/it]
Training Epoch: 15/100, batch 34/500 completed (loss: 0.0049:   7%|▋         | 18/250 [01:46<22:55,  5.93s/it]
Training Epoch: 15/100, batch 34/500 completed (loss: 0.0984:   7%|▋         | 18/250 [01:46<22:55,  5.93s/it]
Training Epoch: 15/100, batch 34/500 completed (loss: 0.0195:   7%|▋         | 18/250 [01:46<22:55,  5.93s/it]
Training Epoch: 15/100, batch 35/500 completed (loss: 0.0193:   7%|▋         | 18/250 [01:46<22:55,  5.93s/it]
Training Epoch: 15/100, batch 35/500 completed (loss: 0.128:   7%|▋         | 18/250 [01:46<22:55,  5.93s/it] 
Training Epoch: 15/100, batch 35/500 completed (loss: 0.1693:   7%|▋         | 18/250 [01:46<22:55,  5.93s/it]
Training Epoch: 15/100, batch 35/500 completed (loss: 0.0054:   7%|▋         | 18/250 [01:46<22:55,  5.93s/it]
Training Epoch: 15/100, batch 36/500 completed (loss: 0.0049:   7%|▋         | 18/250 [01:49<22:55,  5.93s/it]
Training Epoch: 15/100, batch 36/500 completed (loss: 0.2111:   7%|▋         | 18/250 [01:49<22:55,  5.93s/it]
Training Epoch: 15/100, batch 36/500 completed (loss: 0.0254:   7%|▋         | 18/250 [01:49<22:55,  5.93s/it]
Training Epoch: 15/100, batch 36/500 completed (loss: 0.0202:   7%|▋         | 18/250 [01:49<22:55,  5.93s/it]
Training Epoch: 15/100, batch 36/500 completed (loss: 0.0202:   8%|▊         | 19/250 [01:52<23:15,  6.04s/it]
Training Epoch: 15/100, batch 36/500 completed (loss: 0.2111:   8%|▊         | 19/250 [01:52<23:16,  6.04s/it]
Training Epoch: 15/100, batch 36/500 completed (loss: 0.0254:   8%|▊         | 19/250 [01:52<23:16,  6.05s/it]
Training Epoch: 15/100, batch 36/500 completed (loss: 0.0049:   8%|▊         | 19/250 [01:52<23:17,  6.05s/it]
Training Epoch: 15/100, batch 37/500 completed (loss: 0.0854:   8%|▊         | 19/250 [01:53<23:17,  6.05s/it]
Training Epoch: 15/100, batch 37/500 completed (loss: 0.0264:   8%|▊         | 19/250 [01:53<23:16,  6.05s/it]
Training Epoch: 15/100, batch 37/500 completed (loss: 0.0824:   8%|▊         | 19/250 [01:53<23:16,  6.04s/it]
Training Epoch: 15/100, batch 37/500 completed (loss: 0.0253:   8%|▊         | 19/250 [01:53<23:15,  6.04s/it]
Training Epoch: 15/100, batch 38/500 completed (loss: 0.1662:   8%|▊         | 19/250 [01:55<23:17,  6.05s/it]
Training Epoch: 15/100, batch 38/500 completed (loss: 0.0222:   8%|▊         | 19/250 [01:55<23:16,  6.04s/it]
Training Epoch: 15/100, batch 38/500 completed (loss: 0.0447:   8%|▊         | 19/250 [01:55<23:16,  6.05s/it]
Training Epoch: 15/100, batch 38/500 completed (loss: 0.2595:   8%|▊         | 19/250 [01:55<23:15,  6.04s/it]
Training Epoch: 15/100, batch 38/500 completed (loss: 0.2595:   8%|▊         | 20/250 [01:58<22:55,  5.98s/it]
Training Epoch: 15/100, batch 38/500 completed (loss: 0.0447:   8%|▊         | 20/250 [01:58<22:55,  5.98s/it]
Training Epoch: 15/100, batch 38/500 completed (loss: 0.0222:   8%|▊         | 20/250 [01:58<22:56,  5.98s/it]
Training Epoch: 15/100, batch 38/500 completed (loss: 0.1662:   8%|▊         | 20/250 [01:58<22:56,  5.99s/it]
Training Epoch: 15/100, batch 39/500 completed (loss: 0.0161:   8%|▊         | 20/250 [01:58<22:56,  5.99s/it]
Training Epoch: 15/100, batch 39/500 completed (loss: 0.0222:   8%|▊         | 20/250 [01:58<22:55,  5.98s/it]
Training Epoch: 15/100, batch 39/500 completed (loss: 0.0065:   8%|▊         | 20/250 [01:58<22:56,  5.98s/it]
Training Epoch: 15/100, batch 39/500 completed (loss: 0.0062:   8%|▊         | 20/250 [01:58<22:55,  5.98s/it]
Training Epoch: 15/100, batch 40/500 completed (loss: 0.045:   8%|▊         | 20/250 [02:01<22:55,  5.98s/it] 
Training Epoch: 15/100, batch 40/500 completed (loss: 0.1533:   8%|▊         | 20/250 [02:01<22:56,  5.98s/it]
Training Epoch: 15/100, batch 40/500 completed (loss: 0.0159:   8%|▊         | 20/250 [02:01<22:56,  5.99s/it]
Training Epoch: 15/100, batch 40/500 completed (loss: 0.018:   8%|▊         | 20/250 [02:01<22:55,  5.98s/it] 
Training Epoch: 15/100, batch 40/500 completed (loss: 0.018:   8%|▊         | 21/250 [02:05<23:09,  6.07s/it]
Training Epoch: 15/100, batch 40/500 completed (loss: 0.045:   8%|▊         | 21/250 [02:05<23:09,  6.07s/it]
Training Epoch: 15/100, batch 40/500 completed (loss: 0.1533:   8%|▊         | 21/250 [02:05<23:09,  6.07s/it]
Training Epoch: 15/100, batch 40/500 completed (loss: 0.0159:   8%|▊         | 21/250 [02:05<23:09,  6.07s/it]
Training Epoch: 15/100, batch 41/500 completed (loss: 0.0067:   8%|▊         | 21/250 [02:05<23:09,  6.07s/it]
Training Epoch: 15/100, batch 41/500 completed (loss: 0.294:   8%|▊         | 21/250 [02:05<23:09,  6.07s/it] 
Training Epoch: 15/100, batch 41/500 completed (loss: 0.0182:   8%|▊         | 21/250 [02:05<23:09,  6.07s/it]
Training Epoch: 15/100, batch 41/500 completed (loss: 0.0154:   8%|▊         | 21/250 [02:05<23:09,  6.07s/it]
Training Epoch: 15/100, batch 42/500 completed (loss: 0.321:   8%|▊         | 21/250 [02:08<23:09,  6.07s/it] 
Training Epoch: 15/100, batch 42/500 completed (loss: 0.0272:   8%|▊         | 21/250 [02:08<23:09,  6.07s/it]
Training Epoch: 15/100, batch 42/500 completed (loss: 0.019:   8%|▊         | 21/250 [02:08<23:09,  6.07s/it]
Training Epoch: 15/100, batch 42/500 completed (loss: 0.184:   8%|▊         | 21/250 [02:08<23:09,  6.07s/it] 
Training Epoch: 15/100, batch 42/500 completed (loss: 0.184:   9%|▉         | 22/250 [02:11<22:56,  6.04s/it]
Training Epoch: 15/100, batch 42/500 completed (loss: 0.321:   9%|▉         | 22/250 [02:11<22:56,  6.04s/it]
Training Epoch: 15/100, batch 42/500 completed (loss: 0.019:   9%|▉         | 22/250 [02:11<22:55,  6.03s/it]
Training Epoch: 15/100, batch 42/500 completed (loss: 0.0272:   9%|▉         | 22/250 [02:11<22:56,  6.04s/it]
Training Epoch: 15/100, batch 43/500 completed (loss: 0.0173:   9%|▉         | 22/250 [02:11<22:56,  6.04s/it]
Training Epoch: 15/100, batch 43/500 completed (loss: 0.0521:   9%|▉         | 22/250 [02:11<22:56,  6.04s/it]
Training Epoch: 15/100, batch 43/500 completed (loss: 0.0154:   9%|▉         | 22/250 [02:11<22:55,  6.03s/it]
Training Epoch: 15/100, batch 43/500 completed (loss: 0.023:   9%|▉         | 22/250 [02:11<22:56,  6.04s/it]
Training Epoch: 15/100, batch 44/500 completed (loss: 0.0116:   9%|▉         | 22/250 [02:13<22:56,  6.04s/it]
Training Epoch: 15/100, batch 44/500 completed (loss: 0.0326:   9%|▉         | 22/250 [02:13<22:56,  6.04s/it]
Training Epoch: 15/100, batch 44/500 completed (loss: 0.0581:   9%|▉         | 22/250 [02:13<22:55,  6.03s/it]
Training Epoch: 15/100, batch 44/500 completed (loss: 0.1032:   9%|▉         | 22/250 [02:14<22:56,  6.04s/it]
Training Epoch: 15/100, batch 44/500 completed (loss: 0.1032:   9%|▉         | 23/250 [02:16<22:35,  5.97s/it]
Training Epoch: 15/100, batch 44/500 completed (loss: 0.0326:   9%|▉         | 23/250 [02:16<22:34,  5.97s/it]
Training Epoch: 15/100, batch 44/500 completed (loss: 0.0581:   9%|▉         | 23/250 [02:16<22:34,  5.97s/it]
Training Epoch: 15/100, batch 44/500 completed (loss: 0.0116:   9%|▉         | 23/250 [02:16<22:35,  5.97s/it]
Training Epoch: 15/100, batch 45/500 completed (loss: 0.0242:   9%|▉         | 23/250 [02:16<22:34,  5.97s/it]
Training Epoch: 15/100, batch 45/500 completed (loss: 0.021:   9%|▉         | 23/250 [02:16<22:34,  5.97s/it] 
Training Epoch: 15/100, batch 45/500 completed (loss: 0.0855:   9%|▉         | 23/250 [02:16<22:35,  5.97s/it]
Training Epoch: 15/100, batch 45/500 completed (loss: 0.0215:   9%|▉         | 23/250 [02:16<22:35,  5.97s/it]
Training Epoch: 15/100, batch 46/500 completed (loss: 0.1676:   9%|▉         | 23/250 [02:19<22:34,  5.97s/it]
Training Epoch: 15/100, batch 46/500 completed (loss: 0.0457:   9%|▉         | 23/250 [02:19<22:35,  5.97s/it]
Training Epoch: 15/100, batch 46/500 completed (loss: 0.034:   9%|▉         | 23/250 [02:19<22:34,  5.97s/it] 
Training Epoch: 15/100, batch 46/500 completed (loss: 0.0644:   9%|▉         | 23/250 [02:19<22:35,  5.97s/it]
Training Epoch: 15/100, batch 46/500 completed (loss: 0.0457:  10%|▉         | 24/250 [02:23<22:58,  6.10s/it]
Training Epoch: 15/100, batch 46/500 completed (loss: 0.034:  10%|▉         | 24/250 [02:23<22:58,  6.10s/it]
Training Epoch: 15/100, batch 46/500 completed (loss: 0.0644:  10%|▉         | 24/250 [02:23<22:58,  6.10s/it]
Training Epoch: 15/100, batch 46/500 completed (loss: 0.1676:  10%|▉         | 24/250 [02:23<22:58,  6.10s/it]
Training Epoch: 15/100, batch 47/500 completed (loss: 0.0328:  10%|▉         | 24/250 [02:23<22:58,  6.10s/it]
Training Epoch: 15/100, batch 47/500 completed (loss: 0.0676:  10%|▉         | 24/250 [02:23<22:58,  6.10s/it]
Training Epoch: 15/100, batch 47/500 completed (loss: 0.013:  10%|▉         | 24/250 [02:23<22:58,  6.10s/it] 
Training Epoch: 15/100, batch 47/500 completed (loss: 0.3203:  10%|▉         | 24/250 [02:23<22:58,  6.10s/it]
Training Epoch: 15/100, batch 48/500 completed (loss: 0.0363:  10%|▉         | 24/250 [02:26<22:58,  6.10s/it]
Training Epoch: 15/100, batch 48/500 completed (loss: 0.032:  10%|▉         | 24/250 [02:26<22:58,  6.10s/it]
Training Epoch: 15/100, batch 48/500 completed (loss: 0.0549:  10%|▉         | 24/250 [02:26<22:58,  6.10s/it]
Training Epoch: 15/100, batch 48/500 completed (loss: 0.0202:  10%|▉         | 24/250 [02:26<22:58,  6.10s/it]
Training Epoch: 15/100, batch 48/500 completed (loss: 0.0202:  10%|█         | 25/250 [02:29<22:32,  6.01s/it]
Training Epoch: 15/100, batch 48/500 completed (loss: 0.0363:  10%|█         | 25/250 [02:29<22:31,  6.01s/it]
Training Epoch: 15/100, batch 48/500 completed (loss: 0.0549:  10%|█         | 25/250 [02:29<22:32,  6.01s/it]
Training Epoch: 15/100, batch 48/500 completed (loss: 0.032:  10%|█         | 25/250 [02:29<22:32,  6.01s/it]
Training Epoch: 15/100, batch 49/500 completed (loss: 0.0181:  10%|█         | 25/250 [02:29<22:32,  6.01s/it]
Training Epoch: 15/100, batch 49/500 completed (loss: 0.0092:  10%|█         | 25/250 [02:29<22:31,  6.01s/it]
Training Epoch: 15/100, batch 49/500 completed (loss: 0.1591:  10%|█         | 25/250 [02:29<22:32,  6.01s/it]
Training Epoch: 15/100, batch 49/500 completed (loss: 0.0222:  10%|█         | 25/250 [02:29<22:32,  6.01s/it]
Training Epoch: 15/100, batch 50/500 completed (loss: 0.0141:  10%|█         | 25/250 [02:31<22:32,  6.01s/it]
Training Epoch: 15/100, batch 50/500 completed (loss: 0.0518:  10%|█         | 25/250 [02:31<22:31,  6.01s/it]
Training Epoch: 15/100, batch 50/500 completed (loss: 0.1388:  10%|█         | 25/250 [02:31<22:32,  6.01s/it]
Training Epoch: 15/100, batch 50/500 completed (loss: 0.0351:  10%|█         | 25/250 [02:32<22:32,  6.01s/it]
Training Epoch: 15/100, batch 50/500 completed (loss: 0.0518:  10%|█         | 26/250 [02:34<22:12,  5.95s/it]
Training Epoch: 15/100, batch 50/500 completed (loss: 0.1388:  10%|█         | 26/250 [02:34<22:12,  5.95s/it]
Training Epoch: 15/100, batch 50/500 completed (loss: 0.0141:  10%|█         | 26/250 [02:34<22:12,  5.95s/it]
Training Epoch: 15/100, batch 50/500 completed (loss: 0.0351:  10%|█         | 26/250 [02:34<22:13,  5.95s/it]
Training Epoch: 15/100, batch 51/500 completed (loss: 0.0651:  10%|█         | 26/250 [02:34<22:12,  5.95s/it]
Training Epoch: 15/100, batch 51/500 completed (loss: 0.0112:  10%|█         | 26/250 [02:34<22:12,  5.95s/it]
Training Epoch: 15/100, batch 51/500 completed (loss: 0.2531:  10%|█         | 26/250 [02:34<22:12,  5.95s/it]
Training Epoch: 15/100, batch 51/500 completed (loss: 0.1351:  10%|█         | 26/250 [02:34<22:13,  5.95s/it]
Training Epoch: 15/100, batch 52/500 completed (loss: 0.1281:  10%|█         | 26/250 [02:38<22:12,  5.95s/it]
Training Epoch: 15/100, batch 52/500 completed (loss: 0.0589:  10%|█         | 26/250 [02:38<22:12,  5.95s/it]
Training Epoch: 15/100, batch 52/500 completed (loss: 0.0403:  10%|█         | 26/250 [02:38<22:12,  5.95s/it]
Training Epoch: 15/100, batch 52/500 completed (loss: 0.6115:  10%|█         | 26/250 [02:38<22:13,  5.95s/it]
Training Epoch: 15/100, batch 52/500 completed (loss: 0.6115:  11%|█         | 27/250 [02:41<22:40,  6.10s/it]
Training Epoch: 15/100, batch 52/500 completed (loss: 0.1281:  11%|█         | 27/250 [02:41<22:39,  6.10s/it]
Training Epoch: 15/100, batch 52/500 completed (loss: 0.0403:  11%|█         | 27/250 [02:41<22:40,  6.10s/it]
Training Epoch: 15/100, batch 52/500 completed (loss: 0.0589:  11%|█         | 27/250 [02:41<22:40,  6.10s/it]
Training Epoch: 15/100, batch 53/500 completed (loss: 0.0408:  11%|█         | 27/250 [02:41<22:39,  6.10s/it]
Training Epoch: 15/100, batch 53/500 completed (loss: 0.0122:  11%|█         | 27/250 [02:41<22:40,  6.10s/it]
Training Epoch: 15/100, batch 53/500 completed (loss: 0.0997:  11%|█         | 27/250 [02:41<22:40,  6.10s/it]
Training Epoch: 15/100, batch 53/500 completed (loss: 0.0633:  11%|█         | 27/250 [02:41<22:40,  6.10s/it]
Training Epoch: 15/100, batch 54/500 completed (loss: 0.0691:  11%|█         | 27/250 [02:44<22:40,  6.10s/it]
Training Epoch: 15/100, batch 54/500 completed (loss: 0.1097:  11%|█         | 27/250 [02:44<22:40,  6.10s/it]
Training Epoch: 15/100, batch 54/500 completed (loss: 0.069:  11%|█         | 27/250 [02:44<22:39,  6.10s/it] 
Training Epoch: 15/100, batch 54/500 completed (loss: 0.0201:  11%|█         | 27/250 [02:44<22:40,  6.10s/it]
Training Epoch: 15/100, batch 54/500 completed (loss: 0.069:  11%|█         | 28/250 [02:47<22:14,  6.01s/it]
Training Epoch: 15/100, batch 54/500 completed (loss: 0.1097:  11%|█         | 28/250 [02:47<22:14,  6.01s/it]
Training Epoch: 15/100, batch 54/500 completed (loss: 0.0691:  11%|█         | 28/250 [02:47<22:15,  6.01s/it]
Training Epoch: 15/100, batch 54/500 completed (loss: 0.0201:  11%|█         | 28/250 [02:47<22:15,  6.02s/it]
Training Epoch: 15/100, batch 55/500 completed (loss: 0.2063:  11%|█         | 28/250 [02:47<22:14,  6.01s/it]
Training Epoch: 15/100, batch 55/500 completed (loss: 0.0144:  11%|█         | 28/250 [02:47<22:15,  6.01s/it]
Training Epoch: 15/100, batch 55/500 completed (loss: 0.0093:  11%|█         | 28/250 [02:47<22:14,  6.01s/it]
Training Epoch: 15/100, batch 55/500 completed (loss: 0.025:  11%|█         | 28/250 [02:47<22:15,  6.02s/it] 
Training Epoch: 15/100, batch 56/500 completed (loss: 0.0535:  11%|█         | 28/250 [02:50<22:14,  6.01s/it]
Training Epoch: 15/100, batch 56/500 completed (loss: 0.2005:  11%|█         | 28/250 [02:50<22:15,  6.01s/it]
Training Epoch: 15/100, batch 56/500 completed (loss: 0.0784:  11%|█         | 28/250 [02:50<22:14,  6.01s/it]
Training Epoch: 15/100, batch 56/500 completed (loss: 0.1646:  11%|█         | 28/250 [02:50<22:15,  6.02s/it]
Training Epoch: 15/100, batch 56/500 completed (loss: 0.2005:  12%|█▏        | 29/250 [02:53<22:05,  6.00s/it]
Training Epoch: 15/100, batch 56/500 completed (loss: 0.0784:  12%|█▏        | 29/250 [02:53<22:06,  6.00s/it]
Training Epoch: 15/100, batch 56/500 completed (loss: 0.0535:  12%|█▏        | 29/250 [02:53<22:06,  6.00s/it]
Training Epoch: 15/100, batch 56/500 completed (loss: 0.1646:  12%|█▏        | 29/250 [02:53<22:06,  6.00s/it]
Training Epoch: 15/100, batch 57/500 completed (loss: 0.0113:  12%|█▏        | 29/250 [02:53<22:06,  6.00s/it]
Training Epoch: 15/100, batch 57/500 completed (loss: 0.3673:  12%|█▏        | 29/250 [02:53<22:05,  6.00s/it]
Training Epoch: 15/100, batch 57/500 completed (loss: 0.0375:  12%|█▏        | 29/250 [02:53<22:06,  6.00s/it]
Training Epoch: 15/100, batch 57/500 completed (loss: 0.0342:  12%|█▏        | 29/250 [02:53<22:06,  6.00s/it]
Training Epoch: 15/100, batch 58/500 completed (loss: 0.0426:  12%|█▏        | 29/250 [02:56<22:05,  6.00s/it]
Training Epoch: 15/100, batch 58/500 completed (loss: 0.5407:  12%|█▏        | 29/250 [02:56<22:06,  6.00s/it]
Training Epoch: 15/100, batch 58/500 completed (loss: 0.2612:  12%|█▏        | 29/250 [02:56<22:06,  6.00s/it]
Training Epoch: 15/100, batch 58/500 completed (loss: 0.05:  12%|█▏        | 29/250 [02:56<22:06,  6.00s/it]  
Training Epoch: 15/100, batch 58/500 completed (loss: 0.5407:  12%|█▏        | 30/250 [02:59<22:10,  6.05s/it]
Training Epoch: 15/100, batch 58/500 completed (loss: 0.0426:  12%|█▏        | 30/250 [02:59<22:10,  6.05s/it]
Training Epoch: 15/100, batch 58/500 completed (loss: 0.2612:  12%|█▏        | 30/250 [02:59<22:10,  6.05s/it]
Training Epoch: 15/100, batch 58/500 completed (loss: 0.05:  12%|█▏        | 30/250 [02:59<22:10,  6.05s/it]
Training Epoch: 15/100, batch 59/500 completed (loss: 0.0125:  12%|█▏        | 30/250 [02:59<22:10,  6.05s/it]
Training Epoch: 15/100, batch 59/500 completed (loss: 0.0354:  12%|█▏        | 30/250 [02:59<22:10,  6.05s/it]
Training Epoch: 15/100, batch 59/500 completed (loss: 0.1315:  12%|█▏        | 30/250 [02:59<22:10,  6.05s/it]
Training Epoch: 15/100, batch 59/500 completed (loss: 0.1489:  12%|█▏        | 30/250 [02:59<22:10,  6.05s/it]
Training Epoch: 15/100, batch 60/500 completed (loss: 0.0058:  12%|█▏        | 30/250 [03:02<22:10,  6.05s/it]
Training Epoch: 15/100, batch 60/500 completed (loss: 0.7472:  12%|█▏        | 30/250 [03:02<22:10,  6.05s/it]
Training Epoch: 15/100, batch 60/500 completed (loss: 0.0595:  12%|█▏        | 30/250 [03:02<22:10,  6.05s/it]
Training Epoch: 15/100, batch 60/500 completed (loss: 0.0419:  12%|█▏        | 30/250 [03:02<22:10,  6.05s/it]
Training Epoch: 15/100, batch 60/500 completed (loss: 0.0419:  12%|█▏        | 31/250 [03:05<21:50,  5.98s/it]
Training Epoch: 15/100, batch 60/500 completed (loss: 0.0595:  12%|█▏        | 31/250 [03:05<21:50,  5.98s/it]
Training Epoch: 15/100, batch 60/500 completed (loss: 0.7472:  12%|█▏        | 31/250 [03:05<21:50,  5.99s/it]
Training Epoch: 15/100, batch 60/500 completed (loss: 0.0058:  12%|█▏        | 31/250 [03:05<21:51,  5.99s/it]
Training Epoch: 15/100, batch 61/500 completed (loss: 0.2318:  12%|█▏        | 31/250 [03:05<21:51,  5.99s/it]
Training Epoch: 15/100, batch 61/500 completed (loss: 0.0215:  12%|█▏        | 31/250 [03:05<21:50,  5.99s/it]
Training Epoch: 15/100, batch 61/500 completed (loss: 0.0294:  12%|█▏        | 31/250 [03:05<21:50,  5.98s/it]
Training Epoch: 15/100, batch 61/500 completed (loss: 0.079:  12%|█▏        | 31/250 [03:05<21:50,  5.98s/it] 
Training Epoch: 15/100, batch 62/500 completed (loss: 0.0313:  12%|█▏        | 31/250 [03:08<21:51,  5.99s/it]
Training Epoch: 15/100, batch 62/500 completed (loss: 0.0109:  12%|█▏        | 31/250 [03:08<21:50,  5.98s/it]
Training Epoch: 15/100, batch 62/500 completed (loss: 0.1182:  12%|█▏        | 31/250 [03:08<21:50,  5.99s/it]
Training Epoch: 15/100, batch 62/500 completed (loss: 0.0348:  12%|█▏        | 31/250 [03:08<21:50,  5.98s/it]
Training Epoch: 15/100, batch 62/500 completed (loss: 0.1182:  13%|█▎        | 32/250 [03:11<22:04,  6.08s/it]
Training Epoch: 15/100, batch 62/500 completed (loss: 0.0313:  13%|█▎        | 32/250 [03:11<22:04,  6.08s/it]
Training Epoch: 15/100, batch 62/500 completed (loss: 0.0109:  13%|█▎        | 32/250 [03:11<22:05,  6.08s/it]
Training Epoch: 15/100, batch 62/500 completed (loss: 0.0348:  13%|█▎        | 32/250 [03:11<22:05,  6.08s/it]
Training Epoch: 15/100, batch 63/500 completed (loss: 0.0498:  13%|█▎        | 32/250 [03:11<22:05,  6.08s/it]
Training Epoch: 15/100, batch 63/500 completed (loss: 0.165:  13%|█▎        | 32/250 [03:11<22:04,  6.08s/it] 
Training Epoch: 15/100, batch 63/500 completed (loss: 0.1108:  13%|█▎        | 32/250 [03:11<22:04,  6.08s/it]
Training Epoch: 15/100, batch 63/500 completed (loss: 0.0126:  13%|█▎        | 32/250 [03:11<22:05,  6.08s/it]
Training Epoch: 15/100, batch 64/500 completed (loss: 0.0631:  13%|█▎        | 32/250 [03:14<22:04,  6.08s/it]
Training Epoch: 15/100, batch 64/500 completed (loss: 0.1499:  13%|█▎        | 32/250 [03:14<22:05,  6.08s/it]
Training Epoch: 15/100, batch 64/500 completed (loss: 0.0441:  13%|█▎        | 32/250 [03:14<22:04,  6.08s/it]
Training Epoch: 15/100, batch 64/500 completed (loss: 0.0214:  13%|█▎        | 32/250 [03:14<22:05,  6.08s/it]
Training Epoch: 15/100, batch 64/500 completed (loss: 0.0441:  13%|█▎        | 33/250 [03:17<21:52,  6.05s/it]
Training Epoch: 15/100, batch 64/500 completed (loss: 0.0631:  13%|█▎        | 33/250 [03:17<21:52,  6.05s/it]
Training Epoch: 15/100, batch 64/500 completed (loss: 0.0214:  13%|█▎        | 33/250 [03:17<21:52,  6.05s/it]
Training Epoch: 15/100, batch 64/500 completed (loss: 0.1499:  13%|█▎        | 33/250 [03:17<21:52,  6.05s/it]
Training Epoch: 15/100, batch 65/500 completed (loss: 0.0147:  13%|█▎        | 33/250 [03:17<21:52,  6.05s/it]
Training Epoch: 15/100, batch 65/500 completed (loss: 0.0591:  13%|█▎        | 33/250 [03:17<21:52,  6.05s/it]
Training Epoch: 15/100, batch 65/500 completed (loss: 0.1474:  13%|█▎        | 33/250 [03:17<21:52,  6.05s/it]
Training Epoch: 15/100, batch 65/500 completed (loss: 0.0094:  13%|█▎        | 33/250 [03:17<21:52,  6.05s/it]
Training Epoch: 15/100, batch 66/500 completed (loss: 0.0295:  13%|█▎        | 33/250 [03:20<21:52,  6.05s/it]
Training Epoch: 15/100, batch 66/500 completed (loss: 0.0504:  13%|█▎        | 33/250 [03:20<21:52,  6.05s/it]
Training Epoch: 15/100, batch 66/500 completed (loss: 0.4003:  13%|█▎        | 33/250 [03:20<21:52,  6.05s/it]
Training Epoch: 15/100, batch 66/500 completed (loss: 0.3046:  13%|█▎        | 33/250 [03:20<21:52,  6.05s/it]
Training Epoch: 15/100, batch 66/500 completed (loss: 0.3046:  14%|█▎        | 34/250 [03:23<21:39,  6.02s/it]
Training Epoch: 15/100, batch 66/500 completed (loss: 0.0504:  14%|█▎        | 34/250 [03:23<21:39,  6.02s/it]
Training Epoch: 15/100, batch 66/500 completed (loss: 0.4003:  14%|█▎        | 34/250 [03:23<21:39,  6.02s/it]
Training Epoch: 15/100, batch 66/500 completed (loss: 0.0295:  14%|█▎        | 34/250 [03:23<21:39,  6.02s/it]
Training Epoch: 15/100, batch 67/500 completed (loss: 0.0356:  14%|█▎        | 34/250 [03:23<21:39,  6.02s/it]
Training Epoch: 15/100, batch 67/500 completed (loss: 0.1406:  14%|█▎        | 34/250 [03:23<21:39,  6.02s/it]
Training Epoch: 15/100, batch 67/500 completed (loss: 0.0663:  14%|█▎        | 34/250 [03:23<21:39,  6.02s/it]
Training Epoch: 15/100, batch 67/500 completed (loss: 0.0149:  14%|█▎        | 34/250 [03:23<21:39,  6.02s/it]
Training Epoch: 15/100, batch 68/500 completed (loss: 0.2939:  14%|█▎        | 34/250 [03:26<21:39,  6.02s/it]
Training Epoch: 15/100, batch 68/500 completed (loss: 0.0337:  14%|█▎        | 34/250 [03:26<21:39,  6.02s/it]
Training Epoch: 15/100, batch 68/500 completed (loss: 0.0576:  14%|█▎        | 34/250 [03:26<21:39,  6.02s/it]
Training Epoch: 15/100, batch 68/500 completed (loss: 0.0885:  14%|█▎        | 34/250 [03:26<21:39,  6.02s/it]
Training Epoch: 15/100, batch 68/500 completed (loss: 0.2939:  14%|█▍        | 35/250 [03:29<21:53,  6.11s/it]
Training Epoch: 15/100, batch 68/500 completed (loss: 0.0337:  14%|█▍        | 35/250 [03:29<21:53,  6.11s/it]
Training Epoch: 15/100, batch 68/500 completed (loss: 0.0885:  14%|█▍        | 35/250 [03:29<21:53,  6.11s/it]
Training Epoch: 15/100, batch 68/500 completed (loss: 0.0576:  14%|█▍        | 35/250 [03:29<21:53,  6.11s/it]
Training Epoch: 15/100, batch 69/500 completed (loss: 0.0112:  14%|█▍        | 35/250 [03:29<21:53,  6.11s/it]
Training Epoch: 15/100, batch 69/500 completed (loss: 0.018:  14%|█▍        | 35/250 [03:29<21:53,  6.11s/it] 
Training Epoch: 15/100, batch 69/500 completed (loss: 0.4306:  14%|█▍        | 35/250 [03:29<21:53,  6.11s/it]
Training Epoch: 15/100, batch 69/500 completed (loss: 0.0056:  14%|█▍        | 35/250 [03:29<21:53,  6.11s/it]
Training Epoch: 15/100, batch 70/500 completed (loss: 0.0215:  14%|█▍        | 35/250 [03:32<21:53,  6.11s/it]
Training Epoch: 15/100, batch 70/500 completed (loss: 0.0773:  14%|█▍        | 35/250 [03:32<21:53,  6.11s/it]
Training Epoch: 15/100, batch 70/500 completed (loss: 0.0142:  14%|█▍        | 35/250 [03:32<21:53,  6.11s/it]
Training Epoch: 15/100, batch 70/500 completed (loss: 0.1417:  14%|█▍        | 35/250 [03:32<21:53,  6.11s/it]
Training Epoch: 15/100, batch 70/500 completed (loss: 0.0773:  14%|█▍        | 36/250 [03:35<21:28,  6.02s/it]
Training Epoch: 15/100, batch 70/500 completed (loss: 0.1417:  14%|█▍        | 36/250 [03:35<21:28,  6.02s/it]
Training Epoch: 15/100, batch 70/500 completed (loss: 0.0215:  14%|█▍        | 36/250 [03:35<21:28,  6.02s/it]
Training Epoch: 15/100, batch 70/500 completed (loss: 0.0142:  14%|█▍        | 36/250 [03:35<21:28,  6.02s/it]
Training Epoch: 15/100, batch 71/500 completed (loss: 0.0342:  14%|█▍        | 36/250 [03:35<21:28,  6.02s/it]
Training Epoch: 15/100, batch 71/500 completed (loss: 0.0068:  14%|█▍        | 36/250 [03:35<21:28,  6.02s/it]
Training Epoch: 15/100, batch 71/500 completed (loss: 0.0057:  14%|█▍        | 36/250 [03:35<21:28,  6.02s/it]
Training Epoch: 15/100, batch 71/500 completed (loss: 0.0026:  14%|█▍        | 36/250 [03:35<21:28,  6.02s/it]
Training Epoch: 15/100, batch 72/500 completed (loss: 0.0552:  14%|█▍        | 36/250 [03:38<21:28,  6.02s/it]
Training Epoch: 15/100, batch 72/500 completed (loss: 0.064:  14%|█▍        | 36/250 [03:38<21:28,  6.02s/it] 
Training Epoch: 15/100, batch 72/500 completed (loss: 0.0125:  14%|█▍        | 36/250 [03:38<21:28,  6.02s/it]
Training Epoch: 15/100, batch 72/500 completed (loss: 0.1868:  14%|█▍        | 36/250 [03:38<21:28,  6.02s/it]
Training Epoch: 15/100, batch 72/500 completed (loss: 0.0125:  15%|█▍        | 37/250 [03:41<21:38,  6.10s/it]
Training Epoch: 15/100, batch 72/500 completed (loss: 0.064:  15%|█▍        | 37/250 [03:41<21:38,  6.10s/it]
Training Epoch: 15/100, batch 72/500 completed (loss: 0.1868:  15%|█▍        | 37/250 [03:41<21:38,  6.10s/it]
Training Epoch: 15/100, batch 72/500 completed (loss: 0.0552:  15%|█▍        | 37/250 [03:41<21:39,  6.10s/it]
Training Epoch: 15/100, batch 73/500 completed (loss: 0.1306:  15%|█▍        | 37/250 [03:41<21:38,  6.10s/it]
Training Epoch: 15/100, batch 73/500 completed (loss: 0.009:  15%|█▍        | 37/250 [03:41<21:38,  6.10s/it]
Training Epoch: 15/100, batch 73/500 completed (loss: 0.0895:  15%|█▍        | 37/250 [03:41<21:39,  6.10s/it]
Training Epoch: 15/100, batch 73/500 completed (loss: 0.1425:  15%|█▍        | 37/250 [03:41<21:38,  6.10s/it]
Training Epoch: 15/100, batch 74/500 completed (loss: 0.0055:  15%|█▍        | 37/250 [03:44<21:39,  6.10s/it]
Training Epoch: 15/100, batch 74/500 completed (loss: 0.0387:  15%|█▍        | 37/250 [03:44<21:38,  6.10s/it]
Training Epoch: 15/100, batch 74/500 completed (loss: 0.0059:  15%|█▍        | 37/250 [03:44<21:38,  6.10s/it]
Training Epoch: 15/100, batch 74/500 completed (loss: 0.1104:  15%|█▍        | 37/250 [03:44<21:38,  6.10s/it]
Training Epoch: 15/100, batch 74/500 completed (loss: 0.1104:  15%|█▌        | 38/250 [03:47<21:26,  6.07s/it]
Training Epoch: 15/100, batch 74/500 completed (loss: 0.0059:  15%|█▌        | 38/250 [03:47<21:26,  6.07s/it]
Training Epoch: 15/100, batch 74/500 completed (loss: 0.0055:  15%|█▌        | 38/250 [03:47<21:26,  6.07s/it]
Training Epoch: 15/100, batch 74/500 completed (loss: 0.0387:  15%|█▌        | 38/250 [03:47<21:27,  6.07s/it]
Training Epoch: 15/100, batch 75/500 completed (loss: 0.0708:  15%|█▌        | 38/250 [03:47<21:26,  6.07s/it]
Training Epoch: 15/100, batch 75/500 completed (loss: 0.1585:  15%|█▌        | 38/250 [03:47<21:27,  6.07s/it]
Training Epoch: 15/100, batch 75/500 completed (loss: 0.0703:  15%|█▌        | 38/250 [03:47<21:26,  6.07s/it]
Training Epoch: 15/100, batch 75/500 completed (loss: 0.1108:  15%|█▌        | 38/250 [03:47<21:26,  6.07s/it]
Training Epoch: 15/100, batch 76/500 completed (loss: 0.0527:  15%|█▌        | 38/250 [03:50<21:27,  6.07s/it]
Training Epoch: 15/100, batch 76/500 completed (loss: 0.0431:  15%|█▌        | 38/250 [03:50<21:26,  6.07s/it]
Training Epoch: 15/100, batch 76/500 completed (loss: 0.0718:  15%|█▌        | 38/250 [03:50<21:26,  6.07s/it]
Training Epoch: 15/100, batch 76/500 completed (loss: 0.0514:  15%|█▌        | 38/250 [03:50<21:26,  6.07s/it]
Training Epoch: 15/100, batch 76/500 completed (loss: 0.0431:  16%|█▌        | 39/250 [03:53<21:14,  6.04s/it]
Training Epoch: 15/100, batch 76/500 completed (loss: 0.0527:  16%|█▌        | 39/250 [03:53<21:14,  6.04s/it]
Training Epoch: 15/100, batch 76/500 completed (loss: 0.0718:  16%|█▌        | 39/250 [03:53<21:14,  6.04s/it]
Training Epoch: 15/100, batch 76/500 completed (loss: 0.0514:  16%|█▌        | 39/250 [03:53<21:14,  6.04s/it]
Training Epoch: 15/100, batch 77/500 completed (loss: 0.0115:  16%|█▌        | 39/250 [03:53<21:14,  6.04s/it]
Training Epoch: 15/100, batch 77/500 completed (loss: 0.0029:  16%|█▌        | 39/250 [03:53<21:14,  6.04s/it]
Training Epoch: 15/100, batch 77/500 completed (loss: 0.0255:  16%|█▌        | 39/250 [03:53<21:14,  6.04s/it]
Training Epoch: 15/100, batch 77/500 completed (loss: 0.2953:  16%|█▌        | 39/250 [03:53<21:14,  6.04s/it]
Training Epoch: 15/100, batch 78/500 completed (loss: 0.1332:  16%|█▌        | 39/250 [03:56<21:14,  6.04s/it]
Training Epoch: 15/100, batch 78/500 completed (loss: 0.0038:  16%|█▌        | 39/250 [03:56<21:14,  6.04s/it]
Training Epoch: 15/100, batch 78/500 completed (loss: 0.0016:  16%|█▌        | 39/250 [03:56<21:14,  6.04s/it]
Training Epoch: 15/100, batch 78/500 completed (loss: 0.1276:  16%|█▌        | 39/250 [03:56<21:14,  6.04s/it]
Training Epoch: 15/100, batch 78/500 completed (loss: 0.1332:  16%|█▌        | 40/250 [03:59<21:13,  6.07s/it]
Training Epoch: 15/100, batch 78/500 completed (loss: 0.0038:  16%|█▌        | 40/250 [03:59<21:14,  6.07s/it]
Training Epoch: 15/100, batch 78/500 completed (loss: 0.0016:  16%|█▌        | 40/250 [03:59<21:14,  6.07s/it]
Training Epoch: 15/100, batch 78/500 completed (loss: 0.1276:  16%|█▌        | 40/250 [03:59<21:14,  6.07s/it]
Training Epoch: 15/100, batch 79/500 completed (loss: 0.0874:  16%|█▌        | 40/250 [03:59<21:13,  6.07s/it]
Training Epoch: 15/100, batch 79/500 completed (loss: 0.0304:  16%|█▌        | 40/250 [03:59<21:14,  6.07s/it]
Training Epoch: 15/100, batch 79/500 completed (loss: 0.2564:  16%|█▌        | 40/250 [03:59<21:14,  6.07s/it]
Training Epoch: 15/100, batch 79/500 completed (loss: 0.0552:  16%|█▌        | 40/250 [03:59<21:14,  6.07s/it]
Training Epoch: 15/100, batch 80/500 completed (loss: 0.0338:  16%|█▌        | 40/250 [04:02<21:14,  6.07s/it]
Training Epoch: 15/100, batch 80/500 completed (loss: 0.0046:  16%|█▌        | 40/250 [04:02<21:14,  6.07s/it]
Training Epoch: 15/100, batch 80/500 completed (loss: 0.0466:  16%|█▌        | 40/250 [04:02<21:13,  6.07s/it]
Training Epoch: 15/100, batch 80/500 completed (loss: 0.0211:  16%|█▌        | 40/250 [04:02<21:14,  6.07s/it]
Training Epoch: 15/100, batch 80/500 completed (loss: 0.0046:  16%|█▋        | 41/250 [04:05<21:01,  6.03s/it]
Training Epoch: 15/100, batch 80/500 completed (loss: 0.0338:  16%|█▋        | 41/250 [04:05<21:01,  6.03s/it]
Training Epoch: 15/100, batch 80/500 completed (loss: 0.0211:  16%|█▋        | 41/250 [04:05<21:01,  6.03s/it]
Training Epoch: 15/100, batch 80/500 completed (loss: 0.0466:  16%|█▋        | 41/250 [04:05<21:01,  6.03s/it]
Training Epoch: 15/100, batch 81/500 completed (loss: 0.0237:  16%|█▋        | 41/250 [04:05<21:01,  6.03s/it]
Training Epoch: 15/100, batch 81/500 completed (loss: 0.0029:  16%|█▋        | 41/250 [04:05<21:01,  6.03s/it]
Training Epoch: 15/100, batch 81/500 completed (loss: 0.2463:  16%|█▋        | 41/250 [04:05<21:01,  6.03s/it]
Training Epoch: 15/100, batch 81/500 completed (loss: 0.0245:  16%|█▋        | 41/250 [04:05<21:01,  6.03s/it]
Training Epoch: 15/100, batch 82/500 completed (loss: 0.0198:  16%|█▋        | 41/250 [04:08<21:01,  6.03s/it]
Training Epoch: 15/100, batch 82/500 completed (loss: 0.1361:  16%|█▋        | 41/250 [04:08<21:01,  6.03s/it]
Training Epoch: 15/100, batch 82/500 completed (loss: 0.0537:  16%|█▋        | 41/250 [04:08<21:01,  6.03s/it]
Training Epoch: 15/100, batch 82/500 completed (loss: 0.0417:  16%|█▋        | 41/250 [04:08<21:01,  6.03s/it]
Training Epoch: 15/100, batch 82/500 completed (loss: 0.1361:  17%|█▋        | 42/250 [04:11<20:54,  6.03s/it]
Training Epoch: 15/100, batch 82/500 completed (loss: 0.0537:  17%|█▋        | 42/250 [04:11<20:54,  6.03s/it]
Training Epoch: 15/100, batch 82/500 completed (loss: 0.0198:  17%|█▋        | 42/250 [04:11<20:53,  6.03s/it]
Training Epoch: 15/100, batch 82/500 completed (loss: 0.0417:  17%|█▋        | 42/250 [04:11<20:54,  6.03s/it]
Training Epoch: 15/100, batch 83/500 completed (loss: 0.4897:  17%|█▋        | 42/250 [04:11<20:54,  6.03s/it]
Training Epoch: 15/100, batch 83/500 completed (loss: 0.0977:  17%|█▋        | 42/250 [04:11<20:53,  6.03s/it]
Training Epoch: 15/100, batch 83/500 completed (loss: 0.0306:  17%|█▋        | 42/250 [04:11<20:54,  6.03s/it]
Training Epoch: 15/100, batch 83/500 completed (loss: 0.0133:  17%|█▋        | 42/250 [04:11<20:54,  6.03s/it]
Training Epoch: 15/100, batch 84/500 completed (loss: 0.0169:  17%|█▋        | 42/250 [04:14<20:53,  6.03s/it]
Training Epoch: 15/100, batch 84/500 completed (loss: 0.0301:  17%|█▋        | 42/250 [04:14<20:54,  6.03s/it]
Training Epoch: 15/100, batch 84/500 completed (loss: 0.0348:  17%|█▋        | 42/250 [04:14<20:54,  6.03s/it]
Training Epoch: 15/100, batch 84/500 completed (loss: 0.0119:  17%|█▋        | 42/250 [04:14<20:54,  6.03s/it]
Training Epoch: 15/100, batch 84/500 completed (loss: 0.0348:  17%|█▋        | 43/250 [04:17<20:55,  6.06s/it]
Training Epoch: 15/100, batch 84/500 completed (loss: 0.0301:  17%|█▋        | 43/250 [04:17<20:55,  6.06s/it]
Training Epoch: 15/100, batch 84/500 completed (loss: 0.0119:  17%|█▋        | 43/250 [04:17<20:55,  6.06s/it]
Training Epoch: 15/100, batch 84/500 completed (loss: 0.0169:  17%|█▋        | 43/250 [04:17<20:55,  6.07s/it]
Training Epoch: 15/100, batch 85/500 completed (loss: 0.0797:  17%|█▋        | 43/250 [04:17<20:55,  6.06s/it]
Training Epoch: 15/100, batch 85/500 completed (loss: 0.0253:  17%|█▋        | 43/250 [04:18<20:55,  6.07s/it]
Training Epoch: 15/100, batch 85/500 completed (loss: 0.0221:  17%|█▋        | 43/250 [04:18<20:55,  6.06s/it]
Training Epoch: 15/100, batch 85/500 completed (loss: 0.0502:  17%|█▋        | 43/250 [04:18<20:55,  6.06s/it]
Training Epoch: 15/100, batch 86/500 completed (loss: 0.1743:  17%|█▋        | 43/250 [04:20<20:55,  6.07s/it]
Training Epoch: 15/100, batch 86/500 completed (loss: 0.0178:  17%|█▋        | 43/250 [04:20<20:55,  6.06s/it]
Training Epoch: 15/100, batch 86/500 completed (loss: 0.0874:  17%|█▋        | 43/250 [04:20<20:55,  6.06s/it]
Training Epoch: 15/100, batch 86/500 completed (loss: 0.0915:  17%|█▋        | 43/250 [04:20<20:55,  6.06s/it]
Training Epoch: 15/100, batch 86/500 completed (loss: 0.1743:  18%|█▊        | 44/250 [04:23<20:33,  5.99s/it]
Training Epoch: 15/100, batch 86/500 completed (loss: 0.0178:  18%|█▊        | 44/250 [04:23<20:33,  5.99s/it]
Training Epoch: 15/100, batch 86/500 completed (loss: 0.0915:  18%|█▊        | 44/250 [04:23<20:33,  5.99s/it]
Training Epoch: 15/100, batch 86/500 completed (loss: 0.0874:  18%|█▊        | 44/250 [04:23<20:33,  5.99s/it]
Training Epoch: 15/100, batch 87/500 completed (loss: 0.1904:  18%|█▊        | 44/250 [04:23<20:33,  5.99s/it]
Training Epoch: 15/100, batch 87/500 completed (loss: 0.0078:  18%|█▊        | 44/250 [04:23<20:33,  5.99s/it]
Training Epoch: 15/100, batch 87/500 completed (loss: 0.3108:  18%|█▊        | 44/250 [04:23<20:33,  5.99s/it]
Training Epoch: 15/100, batch 87/500 completed (loss: 0.1462:  18%|█▊        | 44/250 [04:23<20:33,  5.99s/it]
Training Epoch: 15/100, batch 88/500 completed (loss: 0.807:  18%|█▊        | 44/250 [04:27<20:33,  5.99s/it] 
Training Epoch: 15/100, batch 88/500 completed (loss: 0.0177:  18%|█▊        | 44/250 [04:27<20:33,  5.99s/it]
Training Epoch: 15/100, batch 88/500 completed (loss: 0.01:  18%|█▊        | 44/250 [04:26<20:33,  5.99s/it]  
Training Epoch: 15/100, batch 88/500 completed (loss: 0.0293:  18%|█▊        | 44/250 [04:27<20:33,  5.99s/it]
Training Epoch: 15/100, batch 88/500 completed (loss: 0.01:  18%|█▊        | 45/250 [04:30<20:44,  6.07s/it]
Training Epoch: 15/100, batch 88/500 completed (loss: 0.0177:  18%|█▊        | 45/250 [04:30<20:43,  6.07s/it]
Training Epoch: 15/100, batch 88/500 completed (loss: 0.0293:  18%|█▊        | 45/250 [04:30<20:43,  6.07s/it]
Training Epoch: 15/100, batch 88/500 completed (loss: 0.807:  18%|█▊        | 45/250 [04:30<20:44,  6.07s/it]
Training Epoch: 15/100, batch 89/500 completed (loss: 0.3776:  18%|█▊        | 45/250 [04:30<20:43,  6.07s/it]
Training Epoch: 15/100, batch 89/500 completed (loss: 0.0367:  18%|█▊        | 45/250 [04:30<20:44,  6.07s/it]
Training Epoch: 15/100, batch 89/500 completed (loss: 0.0181:  18%|█▊        | 45/250 [04:30<20:44,  6.07s/it]
Training Epoch: 15/100, batch 89/500 completed (loss: 0.0405:  18%|█▊        | 45/250 [04:30<20:43,  6.07s/it]
Training Epoch: 15/100, batch 90/500 completed (loss: 0.0987:  18%|█▊        | 45/250 [04:32<20:44,  6.07s/it]
Training Epoch: 15/100, batch 90/500 completed (loss: 0.0559:  18%|█▊        | 45/250 [04:32<20:43,  6.07s/it]
Training Epoch: 15/100, batch 90/500 completed (loss: 0.0074:  18%|█▊        | 45/250 [04:32<20:44,  6.07s/it]
Training Epoch: 15/100, batch 90/500 completed (loss: 0.0425:  18%|█▊        | 45/250 [04:33<20:43,  6.07s/it]
Training Epoch: 15/100, batch 90/500 completed (loss: 0.0559:  18%|█▊        | 46/250 [04:36<20:33,  6.04s/it]
Training Epoch: 15/100, batch 90/500 completed (loss: 0.0987:  18%|█▊        | 46/250 [04:36<20:33,  6.05s/it]
Training Epoch: 15/100, batch 90/500 completed (loss: 0.0425:  18%|█▊        | 46/250 [04:36<20:33,  6.05s/it]
Training Epoch: 15/100, batch 90/500 completed (loss: 0.0074:  18%|█▊        | 46/250 [04:36<20:33,  6.05s/it]
Training Epoch: 15/100, batch 91/500 completed (loss: 0.0978:  18%|█▊        | 46/250 [04:36<20:33,  6.05s/it]
Training Epoch: 15/100, batch 91/500 completed (loss: 0.0499:  18%|█▊        | 46/250 [04:36<20:33,  6.04s/it]
Training Epoch: 15/100, batch 91/500 completed (loss: 0.0311:  18%|█▊        | 46/250 [04:36<20:33,  6.05s/it]
Training Epoch: 15/100, batch 91/500 completed (loss: 0.0107:  18%|█▊        | 46/250 [04:36<20:33,  6.05s/it]
Training Epoch: 15/100, batch 92/500 completed (loss: 0.013:  18%|█▊        | 46/250 [04:38<20:33,  6.05s/it] 
Training Epoch: 15/100, batch 92/500 completed (loss: 0.0951:  18%|█▊        | 46/250 [04:38<20:33,  6.05s/it]
Training Epoch: 15/100, batch 92/500 completed (loss: 0.0375:  18%|█▊        | 46/250 [04:38<20:33,  6.04s/it]
Training Epoch: 15/100, batch 92/500 completed (loss: 0.0357:  18%|█▊        | 46/250 [04:39<20:33,  6.05s/it]
Training Epoch: 15/100, batch 92/500 completed (loss: 0.0951:  19%|█▉        | 47/250 [04:42<20:32,  6.07s/it]
Training Epoch: 15/100, batch 92/500 completed (loss: 0.0375:  19%|█▉        | 47/250 [04:42<20:32,  6.07s/it]
Training Epoch: 15/100, batch 92/500 completed (loss: 0.013:  19%|█▉        | 47/250 [04:42<20:32,  6.07s/it]
Training Epoch: 15/100, batch 92/500 completed (loss: 0.0357:  19%|█▉        | 47/250 [04:42<20:32,  6.07s/it]
Training Epoch: 15/100, batch 93/500 completed (loss: 0.0185:  19%|█▉        | 47/250 [04:42<20:32,  6.07s/it]
Training Epoch: 15/100, batch 93/500 completed (loss: 0.1851:  19%|█▉        | 47/250 [04:42<20:32,  6.07s/it]
Training Epoch: 15/100, batch 93/500 completed (loss: 0.0143:  19%|█▉        | 47/250 [04:42<20:32,  6.07s/it]
Training Epoch: 15/100, batch 93/500 completed (loss: 0.0171:  19%|█▉        | 47/250 [04:42<20:32,  6.07s/it]
Training Epoch: 15/100, batch 94/500 completed (loss: 0.0226:  19%|█▉        | 47/250 [04:45<20:32,  6.07s/it]
Training Epoch: 15/100, batch 94/500 completed (loss: 0.0497:  19%|█▉        | 47/250 [04:45<20:32,  6.07s/it]
Training Epoch: 15/100, batch 94/500 completed (loss: 0.026:  19%|█▉        | 47/250 [04:45<20:32,  6.07s/it] 
Training Epoch: 15/100, batch 94/500 completed (loss: 0.0523:  19%|█▉        | 47/250 [04:45<20:32,  6.07s/it]
Training Epoch: 15/100, batch 94/500 completed (loss: 0.0497:  19%|█▉        | 48/250 [04:48<20:19,  6.04s/it]
Training Epoch: 15/100, batch 94/500 completed (loss: 0.0523:  19%|█▉        | 48/250 [04:48<20:19,  6.04s/it]
Training Epoch: 15/100, batch 94/500 completed (loss: 0.0226:  19%|█▉        | 48/250 [04:48<20:19,  6.04s/it]
Training Epoch: 15/100, batch 94/500 completed (loss: 0.026:  19%|█▉        | 48/250 [04:48<20:20,  6.04s/it]
Training Epoch: 15/100, batch 95/500 completed (loss: 0.1277:  19%|█▉        | 48/250 [04:48<20:20,  6.04s/it]
Training Epoch: 15/100, batch 95/500 completed (loss: 0.0847:  19%|█▉        | 48/250 [04:48<20:19,  6.04s/it]
Training Epoch: 15/100, batch 95/500 completed (loss: 0.0068:  19%|█▉        | 48/250 [04:48<20:19,  6.04s/it]
Training Epoch: 15/100, batch 95/500 completed (loss: 0.107:  19%|█▉        | 48/250 [04:48<20:19,  6.04s/it] 
Training Epoch: 15/100, batch 96/500 completed (loss: 0.0173:  19%|█▉        | 48/250 [04:51<20:19,  6.04s/it]
Training Epoch: 15/100, batch 96/500 completed (loss: 0.0475:  19%|█▉        | 48/250 [04:51<20:20,  6.04s/it]
Training Epoch: 15/100, batch 96/500 completed (loss: 0.0292:  19%|█▉        | 48/250 [04:51<20:19,  6.04s/it]
Training Epoch: 15/100, batch 96/500 completed (loss: 0.0167:  19%|█▉        | 48/250 [04:51<20:19,  6.04s/it]
Training Epoch: 15/100, batch 96/500 completed (loss: 0.0173:  20%|█▉        | 49/250 [04:54<20:11,  6.03s/it]
Training Epoch: 15/100, batch 96/500 completed (loss: 0.0292:  20%|█▉        | 49/250 [04:54<20:11,  6.03s/it]
Training Epoch: 15/100, batch 96/500 completed (loss: 0.0167:  20%|█▉        | 49/250 [04:54<20:11,  6.03s/it]
Training Epoch: 15/100, batch 96/500 completed (loss: 0.0475:  20%|█▉        | 49/250 [04:54<20:11,  6.03s/it]
Training Epoch: 15/100, batch 97/500 completed (loss: 0.0537:  20%|█▉        | 49/250 [04:54<20:11,  6.03s/it]
Training Epoch: 15/100, batch 97/500 completed (loss: 0.0156:  20%|█▉        | 49/250 [04:54<20:11,  6.03s/it]
Training Epoch: 15/100, batch 97/500 completed (loss: 0.0216:  20%|█▉        | 49/250 [04:54<20:11,  6.03s/it]
Training Epoch: 15/100, batch 97/500 completed (loss: 0.165:  20%|█▉        | 49/250 [04:54<20:11,  6.03s/it] 
Training Epoch: 15/100, batch 98/500 completed (loss: 0.0234:  20%|█▉        | 49/250 [04:57<20:11,  6.03s/it]
Training Epoch: 15/100, batch 98/500 completed (loss: 0.0342:  20%|█▉        | 49/250 [04:57<20:11,  6.03s/it]
Training Epoch: 15/100, batch 98/500 completed (loss: 0.3463:  20%|█▉        | 49/250 [04:57<20:11,  6.03s/it]
Training Epoch: 15/100, batch 98/500 completed (loss: 0.0212:  20%|█▉        | 49/250 [04:57<20:11,  6.03s/it]
Training Epoch: 15/100, batch 98/500 completed (loss: 0.0234:  20%|██        | 50/250 [05:00<20:11,  6.06s/it]
Training Epoch: 15/100, batch 98/500 completed (loss: 0.3463:  20%|██        | 50/250 [05:00<20:11,  6.06s/it]
Training Epoch: 15/100, batch 98/500 completed (loss: 0.0212:  20%|██        | 50/250 [05:00<20:11,  6.06s/it]
Training Epoch: 15/100, batch 98/500 completed (loss: 0.0342:  20%|██        | 50/250 [05:00<20:11,  6.06s/it]
Training Epoch: 15/100, batch 99/500 completed (loss: 0.1805:  20%|██        | 50/250 [05:00<20:11,  6.06s/it]
Training Epoch: 15/100, batch 99/500 completed (loss: 0.0789:  20%|██        | 50/250 [05:00<20:11,  6.06s/it]
Training Epoch: 15/100, batch 99/500 completed (loss: 0.0666:  20%|██        | 50/250 [05:00<20:11,  6.06s/it]
Training Epoch: 15/100, batch 99/500 completed (loss: 0.049:  20%|██        | 50/250 [05:00<20:11,  6.06s/it] 
Training Epoch: 15/100, batch 100/500 completed (loss: 0.2197:  20%|██        | 50/250 [05:03<20:11,  6.06s/it]
Training Epoch: 15/100, batch 100/500 completed (loss: 0.1039:  20%|██        | 50/250 [05:03<20:11,  6.06s/it]
Training Epoch: 15/100, batch 100/500 completed (loss: 0.056:  20%|██        | 50/250 [05:03<20:11,  6.06s/it]
Training Epoch: 15/100, batch 100/500 completed (loss: 0.0233:  20%|██        | 50/250 [05:03<20:11,  6.06s/it]
Training Epoch: 15/100, batch 100/500 completed (loss: 0.056:  20%|██        | 51/250 [05:06<19:59,  6.03s/it]
Training Epoch: 15/100, batch 100/500 completed (loss: 0.2197:  20%|██        | 51/250 [05:06<19:59,  6.03s/it]
Training Epoch: 15/100, batch 100/500 completed (loss: 0.0233:  20%|██        | 51/250 [05:06<19:59,  6.03s/it]
Training Epoch: 15/100, batch 100/500 completed (loss: 0.1039:  20%|██        | 51/250 [05:06<19:59,  6.03s/it]
Training Epoch: 15/100, batch 101/500 completed (loss: 0.0281:  20%|██        | 51/250 [05:06<19:59,  6.03s/it]
Training Epoch: 15/100, batch 101/500 completed (loss: 0.0521:  20%|██        | 51/250 [05:06<19:59,  6.03s/it]
Training Epoch: 15/100, batch 101/500 completed (loss: 0.0876:  20%|██        | 51/250 [05:06<19:59,  6.03s/it]
Training Epoch: 15/100, batch 101/500 completed (loss: 0.0238:  20%|██        | 51/250 [05:06<19:59,  6.03s/it]
Training Epoch: 15/100, batch 102/500 completed (loss: 0.0246:  20%|██        | 51/250 [05:09<19:59,  6.03s/it]
Training Epoch: 15/100, batch 102/500 completed (loss: 0.0477:  20%|██        | 51/250 [05:09<19:59,  6.03s/it]
Training Epoch: 15/100, batch 102/500 completed (loss: 0.0077:  20%|██        | 51/250 [05:09<19:59,  6.03s/it]
Training Epoch: 15/100, batch 102/500 completed (loss: 0.0195:  20%|██        | 51/250 [05:09<19:59,  6.03s/it]
Training Epoch: 15/100, batch 102/500 completed (loss: 0.0077:  21%|██        | 52/250 [05:12<20:00,  6.07s/it]
Training Epoch: 15/100, batch 102/500 completed (loss: 0.0195:  21%|██        | 52/250 [05:12<20:00,  6.07s/it]
Training Epoch: 15/100, batch 102/500 completed (loss: 0.0246:  21%|██        | 52/250 [05:12<20:00,  6.07s/it]
Training Epoch: 15/100, batch 102/500 completed (loss: 0.0477:  21%|██        | 52/250 [05:12<20:01,  6.07s/it]
Training Epoch: 15/100, batch 103/500 completed (loss: 0.0817:  21%|██        | 52/250 [05:12<20:01,  6.07s/it]
Training Epoch: 15/100, batch 103/500 completed (loss: 0.0541:  21%|██        | 52/250 [05:12<20:00,  6.07s/it]
Training Epoch: 15/100, batch 103/500 completed (loss: 0.0282:  21%|██        | 52/250 [05:12<20:00,  6.07s/it]
Training Epoch: 15/100, batch 103/500 completed (loss: 0.0119:  21%|██        | 52/250 [05:12<20:00,  6.07s/it]
Training Epoch: 15/100, batch 104/500 completed (loss: 0.0313:  21%|██        | 52/250 [05:15<20:00,  6.07s/it]
Training Epoch: 15/100, batch 104/500 completed (loss: 0.043:  21%|██        | 52/250 [05:15<20:01,  6.07s/it] 
Training Epoch: 15/100, batch 104/500 completed (loss: 0.1325:  21%|██        | 52/250 [05:15<20:00,  6.07s/it]
Training Epoch: 15/100, batch 104/500 completed (loss: 0.1239:  21%|██        | 52/250 [05:15<20:00,  6.07s/it]
Training Epoch: 15/100, batch 104/500 completed (loss: 0.1239:  21%|██        | 53/250 [05:18<19:59,  6.09s/it]
Training Epoch: 15/100, batch 104/500 completed (loss: 0.1325:  21%|██        | 53/250 [05:18<20:00,  6.09s/it]
Training Epoch: 15/100, batch 104/500 completed (loss: 0.0313:  21%|██        | 53/250 [05:18<20:00,  6.09s/it]
Training Epoch: 15/100, batch 104/500 completed (loss: 0.043:  21%|██        | 53/250 [05:18<20:00,  6.09s/it]
Training Epoch: 15/100, batch 105/500 completed (loss: 0.0856:  21%|██        | 53/250 [05:18<20:00,  6.09s/it]
Training Epoch: 15/100, batch 105/500 completed (loss: 0.0194:  21%|██        | 53/250 [05:18<20:00,  6.09s/it]
Training Epoch: 15/100, batch 105/500 completed (loss: 0.1154:  21%|██        | 53/250 [05:18<20:00,  6.09s/it]
Training Epoch: 15/100, batch 105/500 completed (loss: 0.1441:  21%|██        | 53/250 [05:18<19:59,  6.09s/it]
Training Epoch: 15/100, batch 106/500 completed (loss: 0.0064:  21%|██        | 53/250 [05:21<20:00,  6.09s/it]
Training Epoch: 15/100, batch 106/500 completed (loss: 0.0446:  21%|██        | 53/250 [05:21<20:00,  6.09s/it]
Training Epoch: 15/100, batch 106/500 completed (loss: 0.0446:  21%|██        | 53/250 [05:21<20:00,  6.09s/it]
Training Epoch: 15/100, batch 106/500 completed (loss: 0.4041:  21%|██        | 53/250 [05:21<19:59,  6.09s/it]
Training Epoch: 15/100, batch 106/500 completed (loss: 0.0064:  22%|██▏       | 54/250 [05:24<19:49,  6.07s/it]
Training Epoch: 15/100, batch 106/500 completed (loss: 0.0446:  22%|██▏       | 54/250 [05:24<19:49,  6.07s/it]
Training Epoch: 15/100, batch 106/500 completed (loss: 0.0446:  22%|██▏       | 54/250 [05:24<19:49,  6.07s/it]
Training Epoch: 15/100, batch 106/500 completed (loss: 0.4041:  22%|██▏       | 54/250 [05:24<19:49,  6.07s/it]
Training Epoch: 15/100, batch 107/500 completed (loss: 0.0566:  22%|██▏       | 54/250 [05:24<19:49,  6.07s/it]
Training Epoch: 15/100, batch 107/500 completed (loss: 0.2755:  22%|██▏       | 54/250 [05:24<19:49,  6.07s/it]
Training Epoch: 15/100, batch 107/500 completed (loss: 0.0214:  22%|██▏       | 54/250 [05:24<19:49,  6.07s/it]
Training Epoch: 15/100, batch 107/500 completed (loss: 0.0737:  22%|██▏       | 54/250 [05:24<19:49,  6.07s/it]
Training Epoch: 15/100, batch 108/500 completed (loss: 0.027:  22%|██▏       | 54/250 [05:27<19:49,  6.07s/it] 
Training Epoch: 15/100, batch 108/500 completed (loss: 0.0474:  22%|██▏       | 54/250 [05:27<19:49,  6.07s/it]
Training Epoch: 15/100, batch 108/500 completed (loss: 0.0463:  22%|██▏       | 54/250 [05:27<19:49,  6.07s/it]
Training Epoch: 15/100, batch 108/500 completed (loss: 0.0451:  22%|██▏       | 54/250 [05:27<19:49,  6.07s/it]
Training Epoch: 15/100, batch 108/500 completed (loss: 0.0451:  22%|██▏       | 55/250 [05:30<19:41,  6.06s/it]
Training Epoch: 15/100, batch 108/500 completed (loss: 0.0474:  22%|██▏       | 55/250 [05:30<19:41,  6.06s/it]
Training Epoch: 15/100, batch 108/500 completed (loss: 0.0463:  22%|██▏       | 55/250 [05:30<19:42,  6.06s/it]
Training Epoch: 15/100, batch 108/500 completed (loss: 0.027:  22%|██▏       | 55/250 [05:30<19:42,  6.06s/it]
Training Epoch: 15/100, batch 109/500 completed (loss: 0.0187:  22%|██▏       | 55/250 [05:30<19:42,  6.06s/it]
Training Epoch: 15/100, batch 109/500 completed (loss: 0.0155:  22%|██▏       | 55/250 [05:30<19:42,  6.06s/it]
Training Epoch: 15/100, batch 109/500 completed (loss: 0.0473:  22%|██▏       | 55/250 [05:30<19:41,  6.06s/it]
Training Epoch: 15/100, batch 109/500 completed (loss: 0.006:  22%|██▏       | 55/250 [05:30<19:41,  6.06s/it] 
Training Epoch: 15/100, batch 110/500 completed (loss: 0.0168:  22%|██▏       | 55/250 [05:33<19:42,  6.06s/it]
Training Epoch: 15/100, batch 110/500 completed (loss: 0.1961:  22%|██▏       | 55/250 [05:33<19:42,  6.06s/it]
Training Epoch: 15/100, batch 110/500 completed (loss: 0.2227:  22%|██▏       | 55/250 [05:33<19:41,  6.06s/it]
Training Epoch: 15/100, batch 110/500 completed (loss: 0.1723:  22%|██▏       | 55/250 [05:33<19:41,  6.06s/it]
Training Epoch: 15/100, batch 110/500 completed (loss: 0.1723:  22%|██▏       | 56/250 [05:36<19:31,  6.04s/it]
Training Epoch: 15/100, batch 110/500 completed (loss: 0.0168:  22%|██▏       | 56/250 [05:36<19:31,  6.04s/it]
Training Epoch: 15/100, batch 110/500 completed (loss: 0.2227:  22%|██▏       | 56/250 [05:36<19:31,  6.04s/it]
Training Epoch: 15/100, batch 110/500 completed (loss: 0.1961:  22%|██▏       | 56/250 [05:36<19:31,  6.04s/it]
Training Epoch: 15/100, batch 111/500 completed (loss: 0.0059:  22%|██▏       | 56/250 [05:36<19:31,  6.04s/it]
Training Epoch: 15/100, batch 111/500 completed (loss: 0.2305:  22%|██▏       | 56/250 [05:36<19:31,  6.04s/it]
Training Epoch: 15/100, batch 111/500 completed (loss: 0.0117:  22%|██▏       | 56/250 [05:36<19:31,  6.04s/it]
Training Epoch: 15/100, batch 111/500 completed (loss: 0.0222:  22%|██▏       | 56/250 [05:36<19:31,  6.04s/it]
Training Epoch: 15/100, batch 112/500 completed (loss: 0.153:  22%|██▏       | 56/250 [05:39<19:31,  6.04s/it] 
Training Epoch: 15/100, batch 112/500 completed (loss: 0.0035:  22%|██▏       | 56/250 [05:39<19:31,  6.04s/it]
Training Epoch: 15/100, batch 112/500 completed (loss: 0.0075:  22%|██▏       | 56/250 [05:39<19:31,  6.04s/it]
Training Epoch: 15/100, batch 112/500 completed (loss: 0.094:  22%|██▏       | 56/250 [05:39<19:31,  6.04s/it] 
Training Epoch: 15/100, batch 112/500 completed (loss: 0.153:  23%|██▎       | 57/250 [05:42<19:21,  6.02s/it]
Training Epoch: 15/100, batch 112/500 completed (loss: 0.0035:  23%|██▎       | 57/250 [05:42<19:22,  6.02s/it]
Training Epoch: 15/100, batch 112/500 completed (loss: 0.0075:  23%|██▎       | 57/250 [05:42<19:21,  6.02s/it]
Training Epoch: 15/100, batch 112/500 completed (loss: 0.094:  23%|██▎       | 57/250 [05:42<19:22,  6.02s/it]
Training Epoch: 15/100, batch 113/500 completed (loss: 0.1038:  23%|██▎       | 57/250 [05:42<19:21,  6.02s/it]
Training Epoch: 15/100, batch 113/500 completed (loss: 0.0297:  23%|██▎       | 57/250 [05:42<19:21,  6.02s/it]
Training Epoch: 15/100, batch 113/500 completed (loss: 0.0212:  23%|██▎       | 57/250 [05:42<19:22,  6.02s/it]
Training Epoch: 15/100, batch 113/500 completed (loss: 0.0405:  23%|██▎       | 57/250 [05:42<19:22,  6.02s/it]
Training Epoch: 15/100, batch 114/500 completed (loss: 0.0172:  23%|██▎       | 57/250 [05:45<19:21,  6.02s/it]
Training Epoch: 15/100, batch 114/500 completed (loss: 0.0161:  23%|██▎       | 57/250 [05:45<19:21,  6.02s/it]
Training Epoch: 15/100, batch 114/500 completed (loss: 0.0072:  23%|██▎       | 57/250 [05:45<19:22,  6.02s/it]
Training Epoch: 15/100, batch 114/500 completed (loss: 0.0662:  23%|██▎       | 57/250 [05:45<19:22,  6.02s/it]
Training Epoch: 15/100, batch 114/500 completed (loss: 0.0072:  23%|██▎       | 58/250 [05:48<19:22,  6.06s/it]
Training Epoch: 15/100, batch 114/500 completed (loss: 0.0662:  23%|██▎       | 58/250 [05:48<19:23,  6.06s/it]
Training Epoch: 15/100, batch 114/500 completed (loss: 0.0172:  23%|██▎       | 58/250 [05:48<19:23,  6.06s/it]
Training Epoch: 15/100, batch 114/500 completed (loss: 0.0161:  23%|██▎       | 58/250 [05:48<19:23,  6.06s/it]
Training Epoch: 15/100, batch 115/500 completed (loss: 0.0444:  23%|██▎       | 58/250 [05:48<19:23,  6.06s/it]
Training Epoch: 15/100, batch 115/500 completed (loss: 0.0992:  23%|██▎       | 58/250 [05:48<19:22,  6.06s/it]
Training Epoch: 15/100, batch 115/500 completed (loss: 0.1702:  23%|██▎       | 58/250 [05:48<19:23,  6.06s/it]
Training Epoch: 15/100, batch 115/500 completed (loss: 0.0264:  23%|██▎       | 58/250 [05:48<19:23,  6.06s/it]
Training Epoch: 15/100, batch 116/500 completed (loss: 0.003:  23%|██▎       | 58/250 [05:51<19:23,  6.06s/it] 
Training Epoch: 15/100, batch 116/500 completed (loss: 0.0601:  23%|██▎       | 58/250 [05:51<19:22,  6.06s/it]
Training Epoch: 15/100, batch 116/500 completed (loss: 0.3506:  23%|██▎       | 58/250 [05:51<19:23,  6.06s/it]
Training Epoch: 15/100, batch 116/500 completed (loss: 0.1034:  23%|██▎       | 58/250 [05:51<19:23,  6.06s/it]
Training Epoch: 15/100, batch 116/500 completed (loss: 0.0601:  24%|██▎       | 59/250 [05:54<19:14,  6.05s/it]
Training Epoch: 15/100, batch 116/500 completed (loss: 0.3506:  24%|██▎       | 59/250 [05:54<19:14,  6.05s/it]
Training Epoch: 15/100, batch 116/500 completed (loss: 0.1034:  24%|██▎       | 59/250 [05:54<19:15,  6.05s/it]
Training Epoch: 15/100, batch 116/500 completed (loss: 0.003:  24%|██▎       | 59/250 [05:54<19:15,  6.05s/it]
Training Epoch: 15/100, batch 117/500 completed (loss: 0.0419:  24%|██▎       | 59/250 [05:54<19:14,  6.05s/it]
Training Epoch: 15/100, batch 117/500 completed (loss: 0.087:  24%|██▎       | 59/250 [05:54<19:15,  6.05s/it]
Training Epoch: 15/100, batch 117/500 completed (loss: 0.0935:  24%|██▎       | 59/250 [05:54<19:14,  6.05s/it]
Training Epoch: 15/100, batch 117/500 completed (loss: 0.0382:  24%|██▎       | 59/250 [05:54<19:15,  6.05s/it]
Training Epoch: 15/100, batch 118/500 completed (loss: 0.3797:  24%|██▎       | 59/250 [05:57<19:14,  6.05s/it]
Training Epoch: 15/100, batch 118/500 completed (loss: 0.0929:  24%|██▎       | 59/250 [05:57<19:15,  6.05s/it]
Training Epoch: 15/100, batch 118/500 completed (loss: 0.1415:  24%|██▎       | 59/250 [05:57<19:14,  6.05s/it]
Training Epoch: 15/100, batch 118/500 completed (loss: 0.0651:  24%|██▎       | 59/250 [05:57<19:15,  6.05s/it]
Training Epoch: 15/100, batch 118/500 completed (loss: 0.0651:  24%|██▍       | 60/250 [06:01<19:25,  6.13s/it]
Training Epoch: 15/100, batch 118/500 completed (loss: 0.3797:  24%|██▍       | 60/250 [06:01<19:25,  6.13s/it]
Training Epoch: 15/100, batch 118/500 completed (loss: 0.1415:  24%|██▍       | 60/250 [06:01<19:25,  6.14s/it]
Training Epoch: 15/100, batch 118/500 completed (loss: 0.0929:  24%|██▍       | 60/250 [06:01<19:25,  6.14s/it]
Training Epoch: 15/100, batch 119/500 completed (loss: 0.0989:  24%|██▍       | 60/250 [06:01<19:25,  6.14s/it]
Training Epoch: 15/100, batch 119/500 completed (loss: 0.0514:  24%|██▍       | 60/250 [06:01<19:25,  6.13s/it]
Training Epoch: 15/100, batch 119/500 completed (loss: 0.0133:  24%|██▍       | 60/250 [06:01<19:25,  6.14s/it]
Training Epoch: 15/100, batch 119/500 completed (loss: 0.0465:  24%|██▍       | 60/250 [06:01<19:25,  6.13s/it]
Training Epoch: 15/100, batch 120/500 completed (loss: 0.0521:  24%|██▍       | 60/250 [06:04<19:25,  6.14s/it]
Training Epoch: 15/100, batch 120/500 completed (loss: 0.0254:  24%|██▍       | 60/250 [06:04<19:25,  6.14s/it]
Training Epoch: 15/100, batch 120/500 completed (loss: 0.0456:  24%|██▍       | 60/250 [06:03<19:25,  6.13s/it]
Training Epoch: 15/100, batch 120/500 completed (loss: 0.0681:  24%|██▍       | 60/250 [06:04<19:25,  6.13s/it]
Training Epoch: 15/100, batch 120/500 completed (loss: 0.0456:  24%|██▍       | 61/250 [06:07<19:11,  6.09s/it]
Training Epoch: 15/100, batch 120/500 completed (loss: 0.0521:  24%|██▍       | 61/250 [06:07<19:11,  6.09s/it]
Training Epoch: 15/100, batch 120/500 completed (loss: 0.0681:  24%|██▍       | 61/250 [06:07<19:11,  6.09s/it]
Training Epoch: 15/100, batch 120/500 completed (loss: 0.0254:  24%|██▍       | 61/250 [06:07<19:11,  6.09s/it]
Training Epoch: 15/100, batch 121/500 completed (loss: 0.0868:  24%|██▍       | 61/250 [06:07<19:11,  6.09s/it]
Training Epoch: 15/100, batch 121/500 completed (loss: 0.0212:  24%|██▍       | 61/250 [06:07<19:11,  6.09s/it]
Training Epoch: 15/100, batch 121/500 completed (loss: 0.3019:  24%|██▍       | 61/250 [06:07<19:11,  6.09s/it]
Training Epoch: 15/100, batch 121/500 completed (loss: 0.0939:  24%|██▍       | 61/250 [06:07<19:11,  6.09s/it]
Training Epoch: 15/100, batch 122/500 completed (loss: 0.0692:  24%|██▍       | 61/250 [06:10<19:11,  6.09s/it]
Training Epoch: 15/100, batch 122/500 completed (loss: 0.083:  24%|██▍       | 61/250 [06:10<19:11,  6.09s/it] 
Training Epoch: 15/100, batch 122/500 completed (loss: 0.0421:  24%|██▍       | 61/250 [06:10<19:11,  6.09s/it]
Training Epoch: 15/100, batch 122/500 completed (loss: 0.1303:  24%|██▍       | 61/250 [06:10<19:11,  6.09s/it]
Training Epoch: 15/100, batch 122/500 completed (loss: 0.083:  25%|██▍       | 62/250 [06:12<18:57,  6.05s/it]
Training Epoch: 15/100, batch 122/500 completed (loss: 0.0421:  25%|██▍       | 62/250 [06:12<18:57,  6.05s/it]
Training Epoch: 15/100, batch 122/500 completed (loss: 0.1303:  25%|██▍       | 62/250 [06:13<18:57,  6.05s/it]
Training Epoch: 15/100, batch 122/500 completed (loss: 0.0692:  25%|██▍       | 62/250 [06:12<18:57,  6.05s/it]
Training Epoch: 15/100, batch 123/500 completed (loss: 0.1079:  25%|██▍       | 62/250 [06:13<18:57,  6.05s/it]
Training Epoch: 15/100, batch 123/500 completed (loss: 0.2512:  25%|██▍       | 62/250 [06:13<18:57,  6.05s/it]
Training Epoch: 15/100, batch 123/500 completed (loss: 0.0251:  25%|██▍       | 62/250 [06:13<18:57,  6.05s/it]
Training Epoch: 15/100, batch 123/500 completed (loss: 0.2083:  25%|██▍       | 62/250 [06:13<18:57,  6.05s/it]
Training Epoch: 15/100, batch 124/500 completed (loss: 0.1039:  25%|██▍       | 62/250 [06:16<18:57,  6.05s/it]
Training Epoch: 15/100, batch 124/500 completed (loss: 0.0061:  25%|██▍       | 62/250 [06:16<18:57,  6.05s/it]
Training Epoch: 15/100, batch 124/500 completed (loss: 0.1735:  25%|██▍       | 62/250 [06:16<18:57,  6.05s/it]
Training Epoch: 15/100, batch 124/500 completed (loss: 0.0439:  25%|██▍       | 62/250 [06:16<18:57,  6.05s/it]
Training Epoch: 15/100, batch 124/500 completed (loss: 0.1039:  25%|██▌       | 63/250 [06:19<19:06,  6.13s/it]
Training Epoch: 15/100, batch 124/500 completed (loss: 0.1735:  25%|██▌       | 63/250 [06:19<19:06,  6.13s/it]
Training Epoch: 15/100, batch 124/500 completed (loss: 0.0439:  25%|██▌       | 63/250 [06:19<19:06,  6.13s/it]
Training Epoch: 15/100, batch 124/500 completed (loss: 0.0061:  25%|██▌       | 63/250 [06:19<19:06,  6.13s/it]
Training Epoch: 15/100, batch 125/500 completed (loss: 0.0111:  25%|██▌       | 63/250 [06:19<19:06,  6.13s/it]
Training Epoch: 15/100, batch 125/500 completed (loss: 0.0265:  25%|██▌       | 63/250 [06:19<19:06,  6.13s/it]
Training Epoch: 15/100, batch 125/500 completed (loss: 0.4254:  25%|██▌       | 63/250 [06:19<19:06,  6.13s/it]
Training Epoch: 15/100, batch 125/500 completed (loss: 0.0082:  25%|██▌       | 63/250 [06:19<19:06,  6.13s/it]
Training Epoch: 15/100, batch 126/500 completed (loss: 0.0836:  25%|██▌       | 63/250 [06:22<19:06,  6.13s/it]
Training Epoch: 15/100, batch 126/500 completed (loss: 0.2177:  25%|██▌       | 63/250 [06:22<19:06,  6.13s/it]
Training Epoch: 15/100, batch 126/500 completed (loss: 0.3148:  25%|██▌       | 63/250 [06:22<19:06,  6.13s/it]
Training Epoch: 15/100, batch 126/500 completed (loss: 0.0876:  25%|██▌       | 63/250 [06:22<19:06,  6.13s/it]
Training Epoch: 15/100, batch 126/500 completed (loss: 0.2177:  26%|██▌       | 64/250 [06:25<18:54,  6.10s/it]
Training Epoch: 15/100, batch 126/500 completed (loss: 0.0836:  26%|██▌       | 64/250 [06:25<18:54,  6.10s/it]
Training Epoch: 15/100, batch 126/500 completed (loss: 0.3148:  26%|██▌       | 64/250 [06:25<18:54,  6.10s/it]
Training Epoch: 15/100, batch 126/500 completed (loss: 0.0876:  26%|██▌       | 64/250 [06:25<18:54,  6.10s/it]
Training Epoch: 15/100, batch 127/500 completed (loss: 0.0844:  26%|██▌       | 64/250 [06:25<18:54,  6.10s/it]
Training Epoch: 15/100, batch 127/500 completed (loss: 0.163:  26%|██▌       | 64/250 [06:25<18:54,  6.10s/it] 
Training Epoch: 15/100, batch 127/500 completed (loss: 0.0411:  26%|██▌       | 64/250 [06:25<18:54,  6.10s/it]
Training Epoch: 15/100, batch 127/500 completed (loss: 0.0825:  26%|██▌       | 64/250 [06:25<18:54,  6.10s/it]
Training Epoch: 15/100, batch 128/500 completed (loss: 0.0435:  26%|██▌       | 64/250 [06:28<18:54,  6.10s/it]
Training Epoch: 15/100, batch 128/500 completed (loss: 0.0161:  26%|██▌       | 64/250 [06:28<18:54,  6.10s/it]
Training Epoch: 15/100, batch 128/500 completed (loss: 0.062:  26%|██▌       | 64/250 [06:28<18:54,  6.10s/it]
Training Epoch: 15/100, batch 128/500 completed (loss: 0.0255:  26%|██▌       | 64/250 [06:28<18:54,  6.10s/it]
Training Epoch: 15/100, batch 128/500 completed (loss: 0.062:  26%|██▌       | 65/250 [06:31<18:44,  6.08s/it]
Training Epoch: 15/100, batch 128/500 completed (loss: 0.0435:  26%|██▌       | 65/250 [06:31<18:44,  6.08s/it]
Training Epoch: 15/100, batch 128/500 completed (loss: 0.0161:  26%|██▌       | 65/250 [06:31<18:44,  6.08s/it]
Training Epoch: 15/100, batch 128/500 completed (loss: 0.0255:  26%|██▌       | 65/250 [06:31<18:44,  6.08s/it]
Training Epoch: 15/100, batch 129/500 completed (loss: 0.0375:  26%|██▌       | 65/250 [06:31<18:44,  6.08s/it]
Training Epoch: 15/100, batch 129/500 completed (loss: 0.0136:  26%|██▌       | 65/250 [06:31<18:44,  6.08s/it]
Training Epoch: 15/100, batch 129/500 completed (loss: 0.1817:  26%|██▌       | 65/250 [06:31<18:44,  6.08s/it]
Training Epoch: 15/100, batch 129/500 completed (loss: 0.3534:  26%|██▌       | 65/250 [06:31<18:44,  6.08s/it]
Training Epoch: 15/100, batch 130/500 completed (loss: 0.0191:  26%|██▌       | 65/250 [06:34<18:44,  6.08s/it]
Training Epoch: 15/100, batch 130/500 completed (loss: 0.0048:  26%|██▌       | 65/250 [06:34<18:44,  6.08s/it]
Training Epoch: 15/100, batch 130/500 completed (loss: 0.0851:  26%|██▌       | 65/250 [06:34<18:44,  6.08s/it]
Training Epoch: 15/100, batch 130/500 completed (loss: 0.061:  26%|██▌       | 65/250 [06:34<18:44,  6.08s/it] 
Training Epoch: 15/100, batch 130/500 completed (loss: 0.0851:  26%|██▋       | 66/250 [06:37<18:37,  6.08s/it]
Training Epoch: 15/100, batch 130/500 completed (loss: 0.061:  26%|██▋       | 66/250 [06:37<18:37,  6.07s/it]
Training Epoch: 15/100, batch 130/500 completed (loss: 0.0048:  26%|██▋       | 66/250 [06:37<18:38,  6.08s/it]
Training Epoch: 15/100, batch 130/500 completed (loss: 0.0191:  26%|██▋       | 66/250 [06:37<18:38,  6.08s/it]
Training Epoch: 15/100, batch 131/500 completed (loss: 0.0582:  26%|██▋       | 66/250 [06:37<18:38,  6.08s/it]
Training Epoch: 15/100, batch 131/500 completed (loss: 0.0127:  26%|██▋       | 66/250 [06:37<18:38,  6.08s/it]
Training Epoch: 15/100, batch 131/500 completed (loss: 0.0401:  26%|██▋       | 66/250 [06:37<18:37,  6.08s/it]
Training Epoch: 15/100, batch 131/500 completed (loss: 0.0014:  26%|██▋       | 66/250 [06:37<18:37,  6.07s/it]
Training Epoch: 15/100, batch 132/500 completed (loss: 0.1611:  26%|██▋       | 66/250 [06:40<18:38,  6.08s/it]
Training Epoch: 15/100, batch 132/500 completed (loss: 0.3517:  26%|██▋       | 66/250 [06:40<18:37,  6.08s/it]
Training Epoch: 15/100, batch 132/500 completed (loss: 0.2165:  26%|██▋       | 66/250 [06:40<18:38,  6.08s/it]
Training Epoch: 15/100, batch 132/500 completed (loss: 0.1001:  26%|██▋       | 66/250 [06:40<18:37,  6.07s/it]
Training Epoch: 15/100, batch 132/500 completed (loss: 0.2165:  27%|██▋       | 67/250 [06:43<18:23,  6.03s/it]
Training Epoch: 15/100, batch 132/500 completed (loss: 0.3517:  27%|██▋       | 67/250 [06:43<18:23,  6.03s/it]
Training Epoch: 15/100, batch 132/500 completed (loss: 0.1611:  27%|██▋       | 67/250 [06:43<18:23,  6.03s/it]
Training Epoch: 15/100, batch 132/500 completed (loss: 0.1001:  27%|██▋       | 67/250 [06:43<18:23,  6.03s/it]
Training Epoch: 15/100, batch 133/500 completed (loss: 0.0624:  27%|██▋       | 67/250 [06:43<18:23,  6.03s/it]
Training Epoch: 15/100, batch 133/500 completed (loss: 0.0301:  27%|██▋       | 67/250 [06:43<18:23,  6.03s/it]
Training Epoch: 15/100, batch 133/500 completed (loss: 0.1359:  27%|██▋       | 67/250 [06:43<18:23,  6.03s/it]
Training Epoch: 15/100, batch 133/500 completed (loss: 0.0129:  27%|██▋       | 67/250 [06:43<18:23,  6.03s/it]
Training Epoch: 15/100, batch 134/500 completed (loss: 0.0449:  27%|██▋       | 67/250 [06:46<18:23,  6.03s/it]
Training Epoch: 15/100, batch 134/500 completed (loss: 0.1065:  27%|██▋       | 67/250 [06:46<18:23,  6.03s/it]
Training Epoch: 15/100, batch 134/500 completed (loss: 0.2212:  27%|██▋       | 67/250 [06:46<18:23,  6.03s/it]
Training Epoch: 15/100, batch 134/500 completed (loss: 0.0219:  27%|██▋       | 67/250 [06:46<18:23,  6.03s/it]
Training Epoch: 15/100, batch 134/500 completed (loss: 0.2212:  27%|██▋       | 68/250 [06:49<18:24,  6.07s/it]
Training Epoch: 15/100, batch 134/500 completed (loss: 0.1065:  27%|██▋       | 68/250 [06:49<18:24,  6.07s/it]
Training Epoch: 15/100, batch 134/500 completed (loss: 0.0219:  27%|██▋       | 68/250 [06:49<18:24,  6.07s/it]
Training Epoch: 15/100, batch 134/500 completed (loss: 0.0449:  27%|██▋       | 68/250 [06:49<18:24,  6.07s/it]
Training Epoch: 15/100, batch 135/500 completed (loss: 0.0553:  27%|██▋       | 68/250 [06:49<18:24,  6.07s/it]
Training Epoch: 15/100, batch 135/500 completed (loss: 0.0252:  27%|██▋       | 68/250 [06:49<18:24,  6.07s/it]
Training Epoch: 15/100, batch 135/500 completed (loss: 0.0075:  27%|██▋       | 68/250 [06:49<18:24,  6.07s/it]
Training Epoch: 15/100, batch 135/500 completed (loss: 0.03:  27%|██▋       | 68/250 [06:49<18:24,  6.07s/it]  
Training Epoch: 15/100, batch 136/500 completed (loss: 0.1291:  27%|██▋       | 68/250 [06:52<18:24,  6.07s/it]
Training Epoch: 15/100, batch 136/500 completed (loss: 0.0133:  27%|██▋       | 68/250 [06:52<18:24,  6.07s/it]
Training Epoch: 15/100, batch 136/500 completed (loss: 0.046:  27%|██▋       | 68/250 [06:52<18:24,  6.07s/it] 
Training Epoch: 15/100, batch 136/500 completed (loss: 0.0658:  27%|██▋       | 68/250 [06:52<18:24,  6.07s/it]
Training Epoch: 15/100, batch 136/500 completed (loss: 0.046:  28%|██▊       | 69/250 [06:55<18:12,  6.03s/it]
Training Epoch: 15/100, batch 136/500 completed (loss: 0.0133:  28%|██▊       | 69/250 [06:55<18:12,  6.03s/it]
Training Epoch: 15/100, batch 136/500 completed (loss: 0.1291:  28%|██▊       | 69/250 [06:55<18:12,  6.03s/it]
Training Epoch: 15/100, batch 136/500 completed (loss: 0.0658:  28%|██▊       | 69/250 [06:55<18:12,  6.03s/it]
Training Epoch: 15/100, batch 137/500 completed (loss: 0.4019:  28%|██▊       | 69/250 [06:55<18:12,  6.03s/it]
Training Epoch: 15/100, batch 137/500 completed (loss: 0.087:  28%|██▊       | 69/250 [06:55<18:12,  6.03s/it] 
Training Epoch: 15/100, batch 137/500 completed (loss: 0.044:  28%|██▊       | 69/250 [06:55<18:12,  6.03s/it] 
Training Epoch: 15/100, batch 137/500 completed (loss: 0.0294:  28%|██▊       | 69/250 [06:55<18:12,  6.03s/it]
Training Epoch: 15/100, batch 138/500 completed (loss: 0.0125:  28%|██▊       | 69/250 [06:58<18:12,  6.03s/it]
Training Epoch: 15/100, batch 138/500 completed (loss: 0.0373:  28%|██▊       | 69/250 [06:58<18:12,  6.03s/it]
Training Epoch: 15/100, batch 138/500 completed (loss: 0.1383:  28%|██▊       | 69/250 [06:58<18:12,  6.03s/it]
Training Epoch: 15/100, batch 138/500 completed (loss: 0.0235:  28%|██▊       | 69/250 [06:58<18:12,  6.03s/it]
Training Epoch: 15/100, batch 138/500 completed (loss: 0.0235:  28%|██▊       | 70/250 [07:01<18:02,  6.01s/it]
Training Epoch: 15/100, batch 138/500 completed (loss: 0.0373:  28%|██▊       | 70/250 [07:01<18:02,  6.02s/it]
Training Epoch: 15/100, batch 138/500 completed (loss: 0.0125:  28%|██▊       | 70/250 [07:01<18:02,  6.02s/it]
Training Epoch: 15/100, batch 138/500 completed (loss: 0.1383:  28%|██▊       | 70/250 [07:01<18:02,  6.02s/it]
Training Epoch: 15/100, batch 139/500 completed (loss: 0.0089:  28%|██▊       | 70/250 [07:01<18:02,  6.02s/it]
Training Epoch: 15/100, batch 139/500 completed (loss: 0.0568:  28%|██▊       | 70/250 [07:01<18:02,  6.02s/it]
Training Epoch: 15/100, batch 139/500 completed (loss: 0.0841:  28%|██▊       | 70/250 [07:01<18:02,  6.02s/it]
Training Epoch: 15/100, batch 139/500 completed (loss: 0.2266:  28%|██▊       | 70/250 [07:01<18:02,  6.01s/it]
Training Epoch: 15/100, batch 140/500 completed (loss: 0.0889:  28%|██▊       | 70/250 [07:04<18:02,  6.02s/it]
Training Epoch: 15/100, batch 140/500 completed (loss: 0.025:  28%|██▊       | 70/250 [07:04<18:02,  6.02s/it] 
Training Epoch: 15/100, batch 140/500 completed (loss: 0.0069:  28%|██▊       | 70/250 [07:04<18:02,  6.02s/it]
Training Epoch: 15/100, batch 140/500 completed (loss: 0.246:  28%|██▊       | 70/250 [07:04<18:02,  6.01s/it] 
Training Epoch: 15/100, batch 140/500 completed (loss: 0.246:  28%|██▊       | 71/250 [07:07<18:02,  6.05s/it]
Training Epoch: 15/100, batch 140/500 completed (loss: 0.025:  28%|██▊       | 71/250 [07:07<18:03,  6.05s/it]
Training Epoch: 15/100, batch 140/500 completed (loss: 0.0069:  28%|██▊       | 71/250 [07:07<18:03,  6.05s/it]
Training Epoch: 15/100, batch 140/500 completed (loss: 0.0889:  28%|██▊       | 71/250 [07:07<18:03,  6.05s/it]
Training Epoch: 15/100, batch 141/500 completed (loss: 0.2156:  28%|██▊       | 71/250 [07:07<18:03,  6.05s/it]
Training Epoch: 15/100, batch 141/500 completed (loss: 0.386:  28%|██▊       | 71/250 [07:07<18:03,  6.05s/it] 
Training Epoch: 15/100, batch 141/500 completed (loss: 0.0077:  28%|██▊       | 71/250 [07:07<18:03,  6.05s/it]
Training Epoch: 15/100, batch 141/500 completed (loss: 0.0179:  28%|██▊       | 71/250 [07:07<18:02,  6.05s/it]
Training Epoch: 15/100, batch 142/500 completed (loss: 0.1508:  28%|██▊       | 71/250 [07:10<18:03,  6.05s/it]
Training Epoch: 15/100, batch 142/500 completed (loss: 0.0109:  28%|██▊       | 71/250 [07:10<18:03,  6.05s/it]
Training Epoch: 15/100, batch 142/500 completed (loss: 0.1068:  28%|██▊       | 71/250 [07:10<18:03,  6.05s/it]
Training Epoch: 15/100, batch 142/500 completed (loss: 0.0064:  28%|██▊       | 71/250 [07:10<18:02,  6.05s/it]
Training Epoch: 15/100, batch 142/500 completed (loss: 0.0064:  29%|██▉       | 72/250 [07:13<17:53,  6.03s/it]
Training Epoch: 15/100, batch 142/500 completed (loss: 0.1508:  29%|██▉       | 72/250 [07:13<17:53,  6.03s/it]
Training Epoch: 15/100, batch 142/500 completed (loss: 0.0109:  29%|██▉       | 72/250 [07:13<17:54,  6.03s/it]
Training Epoch: 15/100, batch 142/500 completed (loss: 0.1068:  29%|██▉       | 72/250 [07:13<17:53,  6.03s/it]
Training Epoch: 15/100, batch 143/500 completed (loss: 0.4022:  29%|██▉       | 72/250 [07:13<17:53,  6.03s/it]
Training Epoch: 15/100, batch 143/500 completed (loss: 0.0504:  29%|██▉       | 72/250 [07:13<17:53,  6.03s/it]
Training Epoch: 15/100, batch 143/500 completed (loss: 0.292:  29%|██▉       | 72/250 [07:13<17:54,  6.03s/it] 
Training Epoch: 15/100, batch 143/500 completed (loss: 0.0154:  29%|██▉       | 72/250 [07:13<17:53,  6.03s/it]
Training Epoch: 15/100, batch 144/500 completed (loss: 0.0662:  29%|██▉       | 72/250 [07:16<17:54,  6.03s/it]
Training Epoch: 15/100, batch 144/500 completed (loss: 0.0236:  29%|██▉       | 72/250 [07:16<17:53,  6.03s/it]
Training Epoch: 15/100, batch 144/500 completed (loss: 0.042:  29%|██▉       | 72/250 [07:16<17:53,  6.03s/it] 
Training Epoch: 15/100, batch 144/500 completed (loss: 0.0865:  29%|██▉       | 72/250 [07:16<17:53,  6.03s/it]
Training Epoch: 15/100, batch 144/500 completed (loss: 0.0236:  29%|██▉       | 73/250 [07:19<17:53,  6.07s/it]
Training Epoch: 15/100, batch 144/500 completed (loss: 0.0662:  29%|██▉       | 73/250 [07:19<17:54,  6.07s/it]
Training Epoch: 15/100, batch 144/500 completed (loss: 0.042:  29%|██▉       | 73/250 [07:19<17:54,  6.07s/it]
Training Epoch: 15/100, batch 144/500 completed (loss: 0.0865:  29%|██▉       | 73/250 [07:19<17:54,  6.07s/it]
Training Epoch: 15/100, batch 145/500 completed (loss: 0.0321:  29%|██▉       | 73/250 [07:19<17:54,  6.07s/it]
Training Epoch: 15/100, batch 145/500 completed (loss: 0.2362:  29%|██▉       | 73/250 [07:19<17:54,  6.07s/it]
Training Epoch: 15/100, batch 145/500 completed (loss: 0.0373:  29%|██▉       | 73/250 [07:19<17:53,  6.07s/it]
Training Epoch: 15/100, batch 145/500 completed (loss: 0.0342:  29%|██▉       | 73/250 [07:19<17:54,  6.07s/it]
Training Epoch: 15/100, batch 146/500 completed (loss: 0.0094:  29%|██▉       | 73/250 [07:22<17:53,  6.07s/it]
Training Epoch: 15/100, batch 146/500 completed (loss: 0.0503:  29%|██▉       | 73/250 [07:22<17:54,  6.07s/it]
Training Epoch: 15/100, batch 146/500 completed (loss: 0.2096:  29%|██▉       | 73/250 [07:22<17:54,  6.07s/it]
Training Epoch: 15/100, batch 146/500 completed (loss: 0.0156:  29%|██▉       | 73/250 [07:22<17:54,  6.07s/it]
Training Epoch: 15/100, batch 146/500 completed (loss: 0.0156:  30%|██▉       | 74/250 [07:25<17:42,  6.04s/it]
Training Epoch: 15/100, batch 146/500 completed (loss: 0.2096:  30%|██▉       | 74/250 [07:25<17:42,  6.04s/it]
Training Epoch: 15/100, batch 146/500 completed (loss: 0.0503:  30%|██▉       | 74/250 [07:25<17:42,  6.04s/it]
Training Epoch: 15/100, batch 146/500 completed (loss: 0.0094:  30%|██▉       | 74/250 [07:25<17:43,  6.04s/it]
Training Epoch: 15/100, batch 147/500 completed (loss: 0.2597:  30%|██▉       | 74/250 [07:25<17:42,  6.04s/it]
Training Epoch: 15/100, batch 147/500 completed (loss: 0.2045:  30%|██▉       | 74/250 [07:25<17:43,  6.04s/it]
Training Epoch: 15/100, batch 147/500 completed (loss: 0.002:  30%|██▉       | 74/250 [07:25<17:42,  6.04s/it] 
Training Epoch: 15/100, batch 147/500 completed (loss: 0.022:  30%|██▉       | 74/250 [07:25<17:42,  6.04s/it] 
Training Epoch: 15/100, batch 148/500 completed (loss: 0.0062:  30%|██▉       | 74/250 [07:28<17:43,  6.04s/it]
Training Epoch: 15/100, batch 148/500 completed (loss: 0.1343:  30%|██▉       | 74/250 [07:28<17:42,  6.04s/it]
Training Epoch: 15/100, batch 148/500 completed (loss: 0.0691:  30%|██▉       | 74/250 [07:28<17:42,  6.04s/it]
Training Epoch: 15/100, batch 148/500 completed (loss: 0.0511:  30%|██▉       | 74/250 [07:28<17:42,  6.04s/it]
Training Epoch: 15/100, batch 148/500 completed (loss: 0.0062:  30%|███       | 75/250 [07:31<17:24,  5.97s/it]
Training Epoch: 15/100, batch 148/500 completed (loss: 0.0691:  30%|███       | 75/250 [07:31<17:24,  5.97s/it]
Training Epoch: 15/100, batch 148/500 completed (loss: 0.1343:  30%|███       | 75/250 [07:31<17:24,  5.97s/it]
Training Epoch: 15/100, batch 148/500 completed (loss: 0.0511:  30%|███       | 75/250 [07:31<17:24,  5.97s/it]
Training Epoch: 15/100, batch 149/500 completed (loss: 0.0906:  30%|███       | 75/250 [07:31<17:24,  5.97s/it]
Training Epoch: 15/100, batch 149/500 completed (loss: 0.0849:  30%|███       | 75/250 [07:31<17:24,  5.97s/it]
Training Epoch: 15/100, batch 149/500 completed (loss: 0.0437:  30%|███       | 75/250 [07:31<17:24,  5.97s/it]
Training Epoch: 15/100, batch 149/500 completed (loss: 0.0138:  30%|███       | 75/250 [07:31<17:24,  5.97s/it]
Training Epoch: 15/100, batch 150/500 completed (loss: 0.2858:  30%|███       | 75/250 [07:34<17:24,  5.97s/it]
Training Epoch: 15/100, batch 150/500 completed (loss: 0.0486:  30%|███       | 75/250 [07:34<17:24,  5.97s/it]
Training Epoch: 15/100, batch 150/500 completed (loss: 0.0111:  30%|███       | 75/250 [07:34<17:24,  5.97s/it]
Training Epoch: 15/100, batch 150/500 completed (loss: 0.294:  30%|███       | 75/250 [07:34<17:24,  5.97s/it] 
Training Epoch: 15/100, batch 150/500 completed (loss: 0.2858:  30%|███       | 76/250 [07:37<17:34,  6.06s/it]
Training Epoch: 15/100, batch 150/500 completed (loss: 0.0486:  30%|███       | 76/250 [07:37<17:34,  6.06s/it]
Training Epoch: 15/100, batch 150/500 completed (loss: 0.0111:  30%|███       | 76/250 [07:37<17:34,  6.06s/it]
Training Epoch: 15/100, batch 150/500 completed (loss: 0.294:  30%|███       | 76/250 [07:37<17:34,  6.06s/it]
Training Epoch: 15/100, batch 151/500 completed (loss: 0.2581:  30%|███       | 76/250 [07:37<17:34,  6.06s/it]
Training Epoch: 15/100, batch 151/500 completed (loss: 0.0446:  30%|███       | 76/250 [07:37<17:34,  6.06s/it]
Training Epoch: 15/100, batch 151/500 completed (loss: 0.101:  30%|███       | 76/250 [07:37<17:34,  6.06s/it] 
Training Epoch: 15/100, batch 151/500 completed (loss: 0.0222:  30%|███       | 76/250 [07:37<17:34,  6.06s/it]
Training Epoch: 15/100, batch 152/500 completed (loss: 0.0103:  30%|███       | 76/250 [07:40<17:34,  6.06s/it]
Training Epoch: 15/100, batch 152/500 completed (loss: 0.0957:  30%|███       | 76/250 [07:40<17:34,  6.06s/it]
Training Epoch: 15/100, batch 152/500 completed (loss: 0.0637:  30%|███       | 76/250 [07:40<17:34,  6.06s/it]
Training Epoch: 15/100, batch 152/500 completed (loss: 0.417:  30%|███       | 76/250 [07:40<17:34,  6.06s/it] 
Training Epoch: 15/100, batch 152/500 completed (loss: 0.0957:  31%|███       | 77/250 [07:43<17:23,  6.03s/it]
Training Epoch: 15/100, batch 152/500 completed (loss: 0.0103:  31%|███       | 77/250 [07:43<17:23,  6.03s/it]
Training Epoch: 15/100, batch 152/500 completed (loss: 0.0637:  31%|███       | 77/250 [07:43<17:23,  6.03s/it]
Training Epoch: 15/100, batch 152/500 completed (loss: 0.417:  31%|███       | 77/250 [07:43<17:23,  6.03s/it]
Training Epoch: 15/100, batch 153/500 completed (loss: 0.0107:  31%|███       | 77/250 [07:43<17:23,  6.03s/it]
Training Epoch: 15/100, batch 153/500 completed (loss: 0.0066:  31%|███       | 77/250 [07:43<17:23,  6.03s/it]
Training Epoch: 15/100, batch 153/500 completed (loss: 0.3096:  31%|███       | 77/250 [07:43<17:23,  6.03s/it]
Training Epoch: 15/100, batch 153/500 completed (loss: 0.0221:  31%|███       | 77/250 [07:43<17:23,  6.03s/it]
Training Epoch: 15/100, batch 154/500 completed (loss: 0.0272:  31%|███       | 77/250 [07:46<17:23,  6.03s/it]
Training Epoch: 15/100, batch 154/500 completed (loss: 0.093:  31%|███       | 77/250 [07:46<17:23,  6.03s/it] 
Training Epoch: 15/100, batch 154/500 completed (loss: 0.2064:  31%|███       | 77/250 [07:46<17:23,  6.03s/it]
Training Epoch: 15/100, batch 154/500 completed (loss: 0.1081:  31%|███       | 77/250 [07:46<17:23,  6.03s/it]
Training Epoch: 15/100, batch 154/500 completed (loss: 0.1081:  31%|███       | 78/250 [07:49<17:14,  6.02s/it]
Training Epoch: 15/100, batch 154/500 completed (loss: 0.093:  31%|███       | 78/250 [07:49<17:14,  6.02s/it]
Training Epoch: 15/100, batch 154/500 completed (loss: 0.2064:  31%|███       | 78/250 [07:49<17:14,  6.02s/it]
Training Epoch: 15/100, batch 154/500 completed (loss: 0.0272:  31%|███       | 78/250 [07:49<17:14,  6.02s/it]
Training Epoch: 15/100, batch 155/500 completed (loss: 0.0071:  31%|███       | 78/250 [07:49<17:14,  6.02s/it]
Training Epoch: 15/100, batch 155/500 completed (loss: 0.4894:  31%|███       | 78/250 [07:49<17:14,  6.02s/it]
Training Epoch: 15/100, batch 155/500 completed (loss: 0.0157:  31%|███       | 78/250 [07:49<17:14,  6.02s/it]
Training Epoch: 15/100, batch 155/500 completed (loss: 0.0497:  31%|███       | 78/250 [07:49<17:14,  6.02s/it]
Training Epoch: 15/100, batch 156/500 completed (loss: 0.0116:  31%|███       | 78/250 [07:52<17:14,  6.02s/it]
Training Epoch: 15/100, batch 156/500 completed (loss: 0.3593:  31%|███       | 78/250 [07:52<17:14,  6.02s/it]
Training Epoch: 15/100, batch 156/500 completed (loss: 0.0533:  31%|███       | 78/250 [07:52<17:14,  6.02s/it]
Training Epoch: 15/100, batch 156/500 completed (loss: 0.0188:  31%|███       | 78/250 [07:53<17:14,  6.02s/it]
Training Epoch: 15/100, batch 156/500 completed (loss: 0.0116:  32%|███▏      | 79/250 [07:55<17:14,  6.05s/it]
Training Epoch: 15/100, batch 156/500 completed (loss: 0.3593:  32%|███▏      | 79/250 [07:55<17:14,  6.05s/it]
Training Epoch: 15/100, batch 156/500 completed (loss: 0.0533:  32%|███▏      | 79/250 [07:55<17:14,  6.05s/it]
Training Epoch: 15/100, batch 156/500 completed (loss: 0.0188:  32%|███▏      | 79/250 [07:55<17:14,  6.05s/it]
Training Epoch: 15/100, batch 157/500 completed (loss: 0.0195:  32%|███▏      | 79/250 [07:55<17:14,  6.05s/it]
Training Epoch: 15/100, batch 157/500 completed (loss: 0.0539:  32%|███▏      | 79/250 [07:55<17:14,  6.05s/it]
Training Epoch: 15/100, batch 157/500 completed (loss: 0.0302:  32%|███▏      | 79/250 [07:55<17:14,  6.05s/it]
Training Epoch: 15/100, batch 157/500 completed (loss: 0.0338:  32%|███▏      | 79/250 [07:55<17:14,  6.05s/it]
Training Epoch: 15/100, batch 158/500 completed (loss: 0.0141:  32%|███▏      | 79/250 [07:58<17:14,  6.05s/it]
Training Epoch: 15/100, batch 158/500 completed (loss: 0.5408:  32%|███▏      | 79/250 [07:58<17:14,  6.05s/it]
Training Epoch: 15/100, batch 158/500 completed (loss: 0.02:  32%|███▏      | 79/250 [07:58<17:14,  6.05s/it]  
Training Epoch: 15/100, batch 158/500 completed (loss: 0.0692:  32%|███▏      | 79/250 [07:59<17:14,  6.05s/it]
Training Epoch: 15/100, batch 158/500 completed (loss: 0.02:  32%|███▏      | 80/250 [08:01<17:07,  6.04s/it]
Training Epoch: 15/100, batch 158/500 completed (loss: 0.0692:  32%|███▏      | 80/250 [08:01<17:07,  6.04s/it]
Training Epoch: 15/100, batch 158/500 completed (loss: 0.5408:  32%|███▏      | 80/250 [08:01<17:07,  6.04s/it]
Training Epoch: 15/100, batch 158/500 completed (loss: 0.0141:  32%|███▏      | 80/250 [08:01<17:07,  6.04s/it]
Training Epoch: 15/100, batch 159/500 completed (loss: 0.037:  32%|███▏      | 80/250 [08:01<17:07,  6.04s/it]
Training Epoch: 15/100, batch 159/500 completed (loss: 0.0166:  32%|███▏      | 80/250 [08:01<17:07,  6.04s/it]
Training Epoch: 15/100, batch 159/500 completed (loss: 0.0312:  32%|███▏      | 80/250 [08:01<17:07,  6.04s/it]
Training Epoch: 15/100, batch 159/500 completed (loss: 0.092:  32%|███▏      | 80/250 [08:01<17:07,  6.04s/it] 
Training Epoch: 15/100, batch 160/500 completed (loss: 0.1411:  32%|███▏      | 80/250 [08:05<17:07,  6.04s/it]
Training Epoch: 15/100, batch 160/500 completed (loss: 0.3407:  32%|███▏      | 80/250 [08:05<17:07,  6.04s/it]
Training Epoch: 15/100, batch 160/500 completed (loss: 0.0972:  32%|███▏      | 80/250 [08:05<17:07,  6.04s/it]
Training Epoch: 15/100, batch 160/500 completed (loss: 0.0127:  32%|███▏      | 80/250 [08:05<17:07,  6.04s/it]
Training Epoch: 15/100, batch 160/500 completed (loss: 0.1411:  32%|███▏      | 81/250 [08:08<17:13,  6.11s/it]
Training Epoch: 15/100, batch 160/500 completed (loss: 0.0972:  32%|███▏      | 81/250 [08:08<17:13,  6.11s/it]
Training Epoch: 15/100, batch 160/500 completed (loss: 0.3407:  32%|███▏      | 81/250 [08:08<17:13,  6.11s/it]
Training Epoch: 15/100, batch 160/500 completed (loss: 0.0127:  32%|███▏      | 81/250 [08:08<17:13,  6.11s/it]
Training Epoch: 15/100, batch 161/500 completed (loss: 0.0157:  32%|███▏      | 81/250 [08:08<17:13,  6.11s/it]
Training Epoch: 15/100, batch 161/500 completed (loss: 0.0121:  32%|███▏      | 81/250 [08:08<17:13,  6.11s/it]
Training Epoch: 15/100, batch 161/500 completed (loss: 0.0206:  32%|███▏      | 81/250 [08:08<17:13,  6.11s/it]
Training Epoch: 15/100, batch 161/500 completed (loss: 0.0873:  32%|███▏      | 81/250 [08:08<17:13,  6.11s/it]
Training Epoch: 15/100, batch 162/500 completed (loss: 0.0196:  32%|███▏      | 81/250 [08:11<17:13,  6.11s/it]
Training Epoch: 15/100, batch 162/500 completed (loss: 0.014:  32%|███▏      | 81/250 [08:11<17:13,  6.11s/it] 
Training Epoch: 15/100, batch 162/500 completed (loss: 0.1144:  32%|███▏      | 81/250 [08:11<17:13,  6.11s/it]
Training Epoch: 15/100, batch 162/500 completed (loss: 0.3497:  32%|███▏      | 81/250 [08:11<17:13,  6.11s/it]
Training Epoch: 15/100, batch 162/500 completed (loss: 0.0196:  33%|███▎      | 82/250 [08:14<16:59,  6.07s/it]
Training Epoch: 15/100, batch 162/500 completed (loss: 0.3497:  33%|███▎      | 82/250 [08:14<16:59,  6.07s/it]
Training Epoch: 15/100, batch 162/500 completed (loss: 0.1144:  33%|███▎      | 82/250 [08:14<17:00,  6.07s/it]
Training Epoch: 15/100, batch 162/500 completed (loss: 0.014:  33%|███▎      | 82/250 [08:14<17:00,  6.07s/it]
Training Epoch: 15/100, batch 163/500 completed (loss: 0.1032:  33%|███▎      | 82/250 [08:14<16:59,  6.07s/it]
Training Epoch: 15/100, batch 163/500 completed (loss: 0.0161:  33%|███▎      | 82/250 [08:14<17:00,  6.07s/it]
Training Epoch: 15/100, batch 163/500 completed (loss: 0.0392:  33%|███▎      | 82/250 [08:14<17:00,  6.07s/it]
Training Epoch: 15/100, batch 163/500 completed (loss: 0.1615:  33%|███▎      | 82/250 [08:14<16:59,  6.07s/it]
Training Epoch: 15/100, batch 164/500 completed (loss: 0.0832:  33%|███▎      | 82/250 [08:17<16:59,  6.07s/it]
Training Epoch: 15/100, batch 164/500 completed (loss: 0.1833:  33%|███▎      | 82/250 [08:17<17:00,  6.07s/it]
Training Epoch: 15/100, batch 164/500 completed (loss: 0.0245:  33%|███▎      | 82/250 [08:17<17:00,  6.07s/it]
Training Epoch: 15/100, batch 164/500 completed (loss: 0.1369:  33%|███▎      | 82/250 [08:17<16:59,  6.07s/it]
Training Epoch: 15/100, batch 164/500 completed (loss: 0.1369:  33%|███▎      | 83/250 [08:19<16:41,  6.00s/it]
Training Epoch: 15/100, batch 164/500 completed (loss: 0.1833:  33%|███▎      | 83/250 [08:19<16:42,  6.00s/it]
Training Epoch: 15/100, batch 164/500 completed (loss: 0.0832:  33%|███▎      | 83/250 [08:19<16:42,  6.00s/it]
Training Epoch: 15/100, batch 164/500 completed (loss: 0.0245:  33%|███▎      | 83/250 [08:19<16:42,  6.00s/it]
Training Epoch: 15/100, batch 165/500 completed (loss: 0.161:  33%|███▎      | 83/250 [08:19<16:42,  6.00s/it] 
Training Epoch: 15/100, batch 165/500 completed (loss: 0.2587:  33%|███▎      | 83/250 [08:19<16:42,  6.00s/it]
Training Epoch: 15/100, batch 165/500 completed (loss: 0.095:  33%|███▎      | 83/250 [08:19<16:42,  6.00s/it] 
Training Epoch: 15/100, batch 165/500 completed (loss: 0.0546:  33%|███▎      | 83/250 [08:20<16:41,  6.00s/it]
Training Epoch: 15/100, batch 166/500 completed (loss: 0.03:  33%|███▎      | 83/250 [08:23<16:42,  6.00s/it]  
Training Epoch: 15/100, batch 166/500 completed (loss: 0.0399:  33%|███▎      | 83/250 [08:23<16:42,  6.00s/it]
Training Epoch: 15/100, batch 166/500 completed (loss: 0.2141:  33%|███▎      | 83/250 [08:23<16:42,  6.00s/it]
Training Epoch: 15/100, batch 166/500 completed (loss: 0.3075:  33%|███▎      | 83/250 [08:23<16:41,  6.00s/it]
Training Epoch: 15/100, batch 166/500 completed (loss: 0.2141:  34%|███▎      | 84/250 [08:26<16:45,  6.06s/it]
Training Epoch: 15/100, batch 166/500 completed (loss: 0.0399:  34%|███▎      | 84/250 [08:26<16:45,  6.06s/it]
Training Epoch: 15/100, batch 166/500 completed (loss: 0.03:  34%|███▎      | 84/250 [08:26<16:46,  6.06s/it]
Training Epoch: 15/100, batch 166/500 completed (loss: 0.3075:  34%|███▎      | 84/250 [08:26<16:46,  6.06s/it]
Training Epoch: 15/100, batch 167/500 completed (loss: 0.105:  34%|███▎      | 84/250 [08:26<16:45,  6.06s/it] 
Training Epoch: 15/100, batch 167/500 completed (loss: 0.0418:  34%|███▎      | 84/250 [08:26<16:45,  6.06s/it]
Training Epoch: 15/100, batch 167/500 completed (loss: 0.0523:  34%|███▎      | 84/250 [08:26<16:46,  6.06s/it]
Training Epoch: 15/100, batch 167/500 completed (loss: 0.0398:  34%|███▎      | 84/250 [08:26<16:46,  6.06s/it]
Training Epoch: 15/100, batch 168/500 completed (loss: 0.2628:  34%|███▎      | 84/250 [08:29<16:45,  6.06s/it]
Training Epoch: 15/100, batch 168/500 completed (loss: 0.0174:  34%|███▎      | 84/250 [08:29<16:45,  6.06s/it]
Training Epoch: 15/100, batch 168/500 completed (loss: 0.1059:  34%|███▎      | 84/250 [08:29<16:46,  6.06s/it]
Training Epoch: 15/100, batch 168/500 completed (loss: 0.1181:  34%|███▎      | 84/250 [08:29<16:46,  6.06s/it]
Training Epoch: 15/100, batch 168/500 completed (loss: 0.2628:  34%|███▍      | 85/250 [08:32<16:36,  6.04s/it]
Training Epoch: 15/100, batch 168/500 completed (loss: 0.0174:  34%|███▍      | 85/250 [08:32<16:36,  6.04s/it]
Training Epoch: 15/100, batch 168/500 completed (loss: 0.1181:  34%|███▍      | 85/250 [08:32<16:36,  6.04s/it]
Training Epoch: 15/100, batch 168/500 completed (loss: 0.1059:  34%|███▍      | 85/250 [08:32<16:36,  6.04s/it]
Training Epoch: 15/100, batch 169/500 completed (loss: 0.0368:  34%|███▍      | 85/250 [08:32<16:36,  6.04s/it]
Training Epoch: 15/100, batch 169/500 completed (loss: 0.0218:  34%|███▍      | 85/250 [08:32<16:36,  6.04s/it]
Training Epoch: 15/100, batch 169/500 completed (loss: 0.0076:  34%|███▍      | 85/250 [08:32<16:36,  6.04s/it]
Training Epoch: 15/100, batch 169/500 completed (loss: 0.0214:  34%|███▍      | 85/250 [08:32<16:36,  6.04s/it]
Training Epoch: 15/100, batch 170/500 completed (loss: 0.0236:  34%|███▍      | 85/250 [08:35<16:36,  6.04s/it]
Training Epoch: 15/100, batch 170/500 completed (loss: 0.071:  34%|███▍      | 85/250 [08:35<16:36,  6.04s/it] 
Training Epoch: 15/100, batch 170/500 completed (loss: 0.1587:  34%|███▍      | 85/250 [08:35<16:36,  6.04s/it]
Training Epoch: 15/100, batch 170/500 completed (loss: 0.2238:  34%|███▍      | 85/250 [08:35<16:36,  6.04s/it]
Training Epoch: 15/100, batch 170/500 completed (loss: 0.2238:  34%|███▍      | 86/250 [08:38<16:36,  6.08s/it]
Training Epoch: 15/100, batch 170/500 completed (loss: 0.1587:  34%|███▍      | 86/250 [08:38<16:36,  6.08s/it]
Training Epoch: 15/100, batch 170/500 completed (loss: 0.071:  34%|███▍      | 86/250 [08:38<16:36,  6.08s/it]
Training Epoch: 15/100, batch 170/500 completed (loss: 0.0236:  34%|███▍      | 86/250 [08:38<16:37,  6.08s/it]
Training Epoch: 15/100, batch 171/500 completed (loss: 0.0229:  34%|███▍      | 86/250 [08:38<16:37,  6.08s/it]
Training Epoch: 15/100, batch 171/500 completed (loss: 0.0493:  34%|███▍      | 86/250 [08:38<16:36,  6.08s/it]
Training Epoch: 15/100, batch 171/500 completed (loss: 0.0562:  34%|███▍      | 86/250 [08:38<16:36,  6.08s/it]
Training Epoch: 15/100, batch 171/500 completed (loss: 0.0206:  34%|███▍      | 86/250 [08:38<16:36,  6.08s/it]
Training Epoch: 15/100, batch 172/500 completed (loss: 0.0438:  34%|███▍      | 86/250 [08:41<16:36,  6.08s/it]
Training Epoch: 15/100, batch 172/500 completed (loss: 0.2978:  34%|███▍      | 86/250 [08:41<16:37,  6.08s/it]
Training Epoch: 15/100, batch 172/500 completed (loss: 0.1414:  34%|███▍      | 86/250 [08:41<16:36,  6.08s/it]
Training Epoch: 15/100, batch 172/500 completed (loss: 0.1696:  34%|███▍      | 86/250 [08:41<16:36,  6.08s/it]
Training Epoch: 15/100, batch 172/500 completed (loss: 0.1414:  35%|███▍      | 87/250 [08:44<16:34,  6.10s/it]
Training Epoch: 15/100, batch 172/500 completed (loss: 0.0438:  35%|███▍      | 87/250 [08:44<16:34,  6.10s/it]
Training Epoch: 15/100, batch 172/500 completed (loss: 0.2978:  35%|███▍      | 87/250 [08:44<16:34,  6.10s/it]
Training Epoch: 15/100, batch 172/500 completed (loss: 0.1696:  35%|███▍      | 87/250 [08:44<16:35,  6.11s/it]
Training Epoch: 15/100, batch 173/500 completed (loss: 0.085:  35%|███▍      | 87/250 [08:44<16:34,  6.10s/it] 
Training Epoch: 15/100, batch 173/500 completed (loss: 0.371:  35%|███▍      | 87/250 [08:44<16:34,  6.10s/it] 
Training Epoch: 15/100, batch 173/500 completed (loss: 0.0838:  35%|███▍      | 87/250 [08:44<16:34,  6.10s/it]
Training Epoch: 15/100, batch 173/500 completed (loss: 0.0232:  35%|███▍      | 87/250 [08:44<16:35,  6.11s/it]
Training Epoch: 15/100, batch 174/500 completed (loss: 0.0147:  35%|███▍      | 87/250 [08:47<16:34,  6.10s/it]
Training Epoch: 15/100, batch 174/500 completed (loss: 0.061:  35%|███▍      | 87/250 [08:47<16:34,  6.10s/it]
Training Epoch: 15/100, batch 174/500 completed (loss: 0.0868:  35%|███▍      | 87/250 [08:47<16:34,  6.10s/it]
Training Epoch: 15/100, batch 174/500 completed (loss: 0.4667:  35%|███▍      | 87/250 [08:47<16:35,  6.11s/it]
Training Epoch: 15/100, batch 174/500 completed (loss: 0.4667:  35%|███▌      | 88/250 [08:50<16:17,  6.03s/it]
Training Epoch: 15/100, batch 174/500 completed (loss: 0.061:  35%|███▌      | 88/250 [08:50<16:17,  6.03s/it]
Training Epoch: 15/100, batch 174/500 completed (loss: 0.0147:  35%|███▌      | 88/250 [08:50<16:17,  6.03s/it]
Training Epoch: 15/100, batch 174/500 completed (loss: 0.0868:  35%|███▌      | 88/250 [08:50<16:17,  6.03s/it]
Training Epoch: 15/100, batch 175/500 completed (loss: 0.0611:  35%|███▌      | 88/250 [08:50<16:17,  6.03s/it]
Training Epoch: 15/100, batch 175/500 completed (loss: 0.0655:  35%|███▌      | 88/250 [08:50<16:17,  6.03s/it]
Training Epoch: 15/100, batch 175/500 completed (loss: 0.533:  35%|███▌      | 88/250 [08:50<16:17,  6.03s/it] 
Training Epoch: 15/100, batch 175/500 completed (loss: 0.0188:  35%|███▌      | 88/250 [08:50<16:17,  6.03s/it]
Training Epoch: 15/100, batch 176/500 completed (loss: 0.0033:  35%|███▌      | 88/250 [08:53<16:17,  6.03s/it]
Training Epoch: 15/100, batch 176/500 completed (loss: 0.0082:  35%|███▌      | 88/250 [08:53<16:17,  6.03s/it]
Training Epoch: 15/100, batch 176/500 completed (loss: 0.0122:  35%|███▌      | 88/250 [08:53<16:17,  6.03s/it]
Training Epoch: 15/100, batch 176/500 completed (loss: 0.3199:  35%|███▌      | 88/250 [08:53<16:17,  6.03s/it]
Training Epoch: 15/100, batch 176/500 completed (loss: 0.0082:  36%|███▌      | 89/250 [08:56<16:25,  6.12s/it]
Training Epoch: 15/100, batch 176/500 completed (loss: 0.3199:  36%|███▌      | 89/250 [08:56<16:25,  6.12s/it]
Training Epoch: 15/100, batch 176/500 completed (loss: 0.0033:  36%|███▌      | 89/250 [08:56<16:25,  6.12s/it]
Training Epoch: 15/100, batch 176/500 completed (loss: 0.0122:  36%|███▌      | 89/250 [08:56<16:25,  6.12s/it]
Training Epoch: 15/100, batch 177/500 completed (loss: 0.0188:  36%|███▌      | 89/250 [08:56<16:25,  6.12s/it]
Training Epoch: 15/100, batch 177/500 completed (loss: 0.0032:  36%|███▌      | 89/250 [08:56<16:25,  6.12s/it]
Training Epoch: 15/100, batch 177/500 completed (loss: 0.0462:  36%|███▌      | 89/250 [08:56<16:25,  6.12s/it]
Training Epoch: 15/100, batch 177/500 completed (loss: 0.1014:  36%|███▌      | 89/250 [08:56<16:25,  6.12s/it]
Training Epoch: 15/100, batch 178/500 completed (loss: 0.1873:  36%|███▌      | 89/250 [08:59<16:25,  6.12s/it]
Training Epoch: 15/100, batch 178/500 completed (loss: 0.054:  36%|███▌      | 89/250 [08:59<16:25,  6.12s/it] 
Training Epoch: 15/100, batch 178/500 completed (loss: 0.245:  36%|███▌      | 89/250 [08:59<16:25,  6.12s/it] 
Training Epoch: 15/100, batch 178/500 completed (loss: 0.0124:  36%|███▌      | 89/250 [08:59<16:25,  6.12s/it]
Training Epoch: 15/100, batch 178/500 completed (loss: 0.054:  36%|███▌      | 90/250 [09:02<16:14,  6.09s/it]
Training Epoch: 15/100, batch 178/500 completed (loss: 0.0124:  36%|███▌      | 90/250 [09:02<16:15,  6.09s/it]
Training Epoch: 15/100, batch 178/500 completed (loss: 0.1873:  36%|███▌      | 90/250 [09:02<16:15,  6.09s/it]
Training Epoch: 15/100, batch 178/500 completed (loss: 0.245:  36%|███▌      | 90/250 [09:02<16:15,  6.10s/it]
Training Epoch: 15/100, batch 179/500 completed (loss: 0.0258:  36%|███▌      | 90/250 [09:02<16:15,  6.09s/it]
Training Epoch: 15/100, batch 179/500 completed (loss: 0.2946:  36%|███▌      | 90/250 [09:02<16:15,  6.10s/it]
Training Epoch: 15/100, batch 179/500 completed (loss: 0.3288:  36%|███▌      | 90/250 [09:02<16:14,  6.09s/it]
Training Epoch: 15/100, batch 179/500 completed (loss: 0.4721:  36%|███▌      | 90/250 [09:02<16:15,  6.09s/it]
Training Epoch: 15/100, batch 180/500 completed (loss: 0.1196:  36%|███▌      | 90/250 [09:05<16:14,  6.09s/it]
Training Epoch: 15/100, batch 180/500 completed (loss: 0.0882:  36%|███▌      | 90/250 [09:05<16:15,  6.10s/it]
Training Epoch: 15/100, batch 180/500 completed (loss: 0.2059:  36%|███▌      | 90/250 [09:05<16:15,  6.09s/it]
Training Epoch: 15/100, batch 180/500 completed (loss: 0.0227:  36%|███▌      | 90/250 [09:05<16:15,  6.09s/it]
Training Epoch: 15/100, batch 180/500 completed (loss: 0.2059:  36%|███▋      | 91/250 [09:08<16:05,  6.07s/it]
Training Epoch: 15/100, batch 180/500 completed (loss: 0.1196:  36%|███▋      | 91/250 [09:08<16:05,  6.07s/it]
Training Epoch: 15/100, batch 180/500 completed (loss: 0.0882:  36%|███▋      | 91/250 [09:08<16:05,  6.07s/it]
Training Epoch: 15/100, batch 180/500 completed (loss: 0.0227:  36%|███▋      | 91/250 [09:08<16:05,  6.07s/it]
Training Epoch: 15/100, batch 181/500 completed (loss: 0.0994:  36%|███▋      | 91/250 [09:08<16:05,  6.07s/it]
Training Epoch: 15/100, batch 181/500 completed (loss: 0.2002:  36%|███▋      | 91/250 [09:08<16:05,  6.07s/it]
Training Epoch: 15/100, batch 181/500 completed (loss: 0.0292:  36%|███▋      | 91/250 [09:08<16:05,  6.07s/it]
Training Epoch: 15/100, batch 181/500 completed (loss: 0.1026:  36%|███▋      | 91/250 [09:08<16:05,  6.07s/it]
Training Epoch: 15/100, batch 182/500 completed (loss: 0.1075:  36%|███▋      | 91/250 [09:11<16:05,  6.07s/it]
Training Epoch: 15/100, batch 182/500 completed (loss: 0.0263:  36%|███▋      | 91/250 [09:11<16:05,  6.07s/it]
Training Epoch: 15/100, batch 182/500 completed (loss: 0.0399:  36%|███▋      | 91/250 [09:11<16:05,  6.07s/it]
Training Epoch: 15/100, batch 182/500 completed (loss: 0.1051:  36%|███▋      | 91/250 [09:12<16:05,  6.07s/it]
Training Epoch: 15/100, batch 182/500 completed (loss: 0.1051:  37%|███▋      | 92/250 [09:14<16:02,  6.09s/it]
Training Epoch: 15/100, batch 182/500 completed (loss: 0.0263:  37%|███▋      | 92/250 [09:14<16:03,  6.10s/it]
Training Epoch: 15/100, batch 182/500 completed (loss: 0.0399:  37%|███▋      | 92/250 [09:14<16:03,  6.10s/it]
Training Epoch: 15/100, batch 182/500 completed (loss: 0.1075:  37%|███▋      | 92/250 [09:14<16:03,  6.10s/it]
Training Epoch: 15/100, batch 183/500 completed (loss: 0.0215:  37%|███▋      | 92/250 [09:14<16:03,  6.10s/it]
Training Epoch: 15/100, batch 183/500 completed (loss: 0.0061:  37%|███▋      | 92/250 [09:14<16:03,  6.10s/it]
Training Epoch: 15/100, batch 183/500 completed (loss: 0.0094:  37%|███▋      | 92/250 [09:14<16:03,  6.10s/it]
Training Epoch: 15/100, batch 183/500 completed (loss: 0.2691:  37%|███▋      | 92/250 [09:14<16:02,  6.09s/it]
Training Epoch: 15/100, batch 184/500 completed (loss: 0.0114:  37%|███▋      | 92/250 [09:17<16:03,  6.10s/it]
Training Epoch: 15/100, batch 184/500 completed (loss: 0.032:  37%|███▋      | 92/250 [09:17<16:03,  6.10s/it] 
Training Epoch: 15/100, batch 184/500 completed (loss: 0.0515:  37%|███▋      | 92/250 [09:17<16:03,  6.10s/it]
Training Epoch: 15/100, batch 184/500 completed (loss: 0.0242:  37%|███▋      | 92/250 [09:18<16:02,  6.09s/it]
Training Epoch: 15/100, batch 184/500 completed (loss: 0.0242:  37%|███▋      | 93/250 [09:20<15:52,  6.07s/it]
Training Epoch: 15/100, batch 184/500 completed (loss: 0.0515:  37%|███▋      | 93/250 [09:20<15:52,  6.07s/it]
Training Epoch: 15/100, batch 184/500 completed (loss: 0.032:  37%|███▋      | 93/250 [09:20<15:52,  6.07s/it]
Training Epoch: 15/100, batch 184/500 completed (loss: 0.0114:  37%|███▋      | 93/250 [09:20<15:53,  6.07s/it]
Training Epoch: 15/100, batch 185/500 completed (loss: 0.1341:  37%|███▋      | 93/250 [09:20<15:52,  6.07s/it]
Training Epoch: 15/100, batch 185/500 completed (loss: 0.2957:  37%|███▋      | 93/250 [09:20<15:52,  6.07s/it]
Training Epoch: 15/100, batch 185/500 completed (loss: 0.1368:  37%|███▋      | 93/250 [09:20<15:53,  6.07s/it]
Training Epoch: 15/100, batch 185/500 completed (loss: 0.1491:  37%|███▋      | 93/250 [09:20<15:52,  6.07s/it]
Training Epoch: 15/100, batch 186/500 completed (loss: 0.0242:  37%|███▋      | 93/250 [09:23<15:52,  6.07s/it]
Training Epoch: 15/100, batch 186/500 completed (loss: 0.0221:  37%|███▋      | 93/250 [09:23<15:53,  6.07s/it]
Training Epoch: 15/100, batch 186/500 completed (loss: 0.1462:  37%|███▋      | 93/250 [09:23<15:52,  6.07s/it]
Training Epoch: 15/100, batch 186/500 completed (loss: 0.0323:  37%|███▋      | 93/250 [09:23<15:52,  6.07s/it]
Training Epoch: 15/100, batch 186/500 completed (loss: 0.0323:  38%|███▊      | 94/250 [09:27<15:50,  6.09s/it]
Training Epoch: 15/100, batch 186/500 completed (loss: 0.0242:  38%|███▊      | 94/250 [09:27<15:50,  6.09s/it]
Training Epoch: 15/100, batch 186/500 completed (loss: 0.1462:  38%|███▊      | 94/250 [09:27<15:50,  6.09s/it]
Training Epoch: 15/100, batch 186/500 completed (loss: 0.0221:  38%|███▊      | 94/250 [09:27<15:50,  6.09s/it]
Training Epoch: 15/100, batch 187/500 completed (loss: 0.0795:  38%|███▊      | 94/250 [09:27<15:50,  6.09s/it]
Training Epoch: 15/100, batch 187/500 completed (loss: 0.027:  38%|███▊      | 94/250 [09:27<15:50,  6.09s/it] 
Training Epoch: 15/100, batch 187/500 completed (loss: 0.0236:  38%|███▊      | 94/250 [09:27<15:50,  6.09s/it]
Training Epoch: 15/100, batch 187/500 completed (loss: 0.1302:  38%|███▊      | 94/250 [09:27<15:50,  6.09s/it]
Training Epoch: 15/100, batch 188/500 completed (loss: 0.0111:  38%|███▊      | 94/250 [09:30<15:50,  6.09s/it]
Training Epoch: 15/100, batch 188/500 completed (loss: 0.0088:  38%|███▊      | 94/250 [09:30<15:50,  6.09s/it]
Training Epoch: 15/100, batch 188/500 completed (loss: 0.0192:  38%|███▊      | 94/250 [09:30<15:50,  6.09s/it]
Training Epoch: 15/100, batch 188/500 completed (loss: 0.0154:  38%|███▊      | 94/250 [09:30<15:50,  6.09s/it]
Training Epoch: 15/100, batch 188/500 completed (loss: 0.0154:  38%|███▊      | 95/250 [09:33<15:43,  6.09s/it]
Training Epoch: 15/100, batch 188/500 completed (loss: 0.0088:  38%|███▊      | 95/250 [09:33<15:43,  6.09s/it]
Training Epoch: 15/100, batch 188/500 completed (loss: 0.0192:  38%|███▊      | 95/250 [09:33<15:43,  6.09s/it]
Training Epoch: 15/100, batch 188/500 completed (loss: 0.0111:  38%|███▊      | 95/250 [09:33<15:43,  6.09s/it]
Training Epoch: 15/100, batch 189/500 completed (loss: 0.0552:  38%|███▊      | 95/250 [09:33<15:43,  6.09s/it]
Training Epoch: 15/100, batch 189/500 completed (loss: 0.2104:  38%|███▊      | 95/250 [09:33<15:43,  6.09s/it]
Training Epoch: 15/100, batch 189/500 completed (loss: 0.0751:  38%|███▊      | 95/250 [09:33<15:43,  6.09s/it]
Training Epoch: 15/100, batch 189/500 completed (loss: 0.0576:  38%|███▊      | 95/250 [09:33<15:43,  6.09s/it]
Training Epoch: 15/100, batch 190/500 completed (loss: 0.042:  38%|███▊      | 95/250 [09:36<15:43,  6.09s/it] 
Training Epoch: 15/100, batch 190/500 completed (loss: 0.0104:  38%|███▊      | 95/250 [09:36<15:43,  6.09s/it]
Training Epoch: 15/100, batch 190/500 completed (loss: 0.2996:  38%|███▊      | 95/250 [09:36<15:43,  6.09s/it]
Training Epoch: 15/100, batch 190/500 completed (loss: 0.0798:  38%|███▊      | 95/250 [09:36<15:43,  6.09s/it]
Training Epoch: 15/100, batch 190/500 completed (loss: 0.0798:  38%|███▊      | 96/250 [09:39<15:34,  6.07s/it]
Training Epoch: 15/100, batch 190/500 completed (loss: 0.2996:  38%|███▊      | 96/250 [09:39<15:34,  6.07s/it]
Training Epoch: 15/100, batch 190/500 completed (loss: 0.0104:  38%|███▊      | 96/250 [09:39<15:33,  6.06s/it]
Training Epoch: 15/100, batch 190/500 completed (loss: 0.042:  38%|███▊      | 96/250 [09:39<15:34,  6.07s/it]
Training Epoch: 15/100, batch 191/500 completed (loss: 0.3449:  38%|███▊      | 96/250 [09:39<15:33,  6.06s/it]
Training Epoch: 15/100, batch 191/500 completed (loss: 0.0984:  38%|███▊      | 96/250 [09:39<15:34,  6.07s/it]
Training Epoch: 15/100, batch 191/500 completed (loss: 0.0454:  38%|███▊      | 96/250 [09:39<15:34,  6.07s/it]
Training Epoch: 15/100, batch 191/500 completed (loss: 0.0261:  38%|███▊      | 96/250 [09:39<15:34,  6.07s/it]
Training Epoch: 15/100, batch 192/500 completed (loss: 0.0128:  38%|███▊      | 96/250 [09:42<15:33,  6.06s/it]
Training Epoch: 15/100, batch 192/500 completed (loss: 0.0226:  38%|███▊      | 96/250 [09:42<15:34,  6.07s/it]
Training Epoch: 15/100, batch 192/500 completed (loss: 0.0573:  38%|███▊      | 96/250 [09:42<15:34,  6.07s/it]
Training Epoch: 15/100, batch 192/500 completed (loss: 0.012:  38%|███▊      | 96/250 [09:42<15:34,  6.07s/it] 
Training Epoch: 15/100, batch 192/500 completed (loss: 0.0573:  39%|███▉      | 97/250 [09:45<15:38,  6.13s/it]
Training Epoch: 15/100, batch 192/500 completed (loss: 0.0226:  39%|███▉      | 97/250 [09:45<15:38,  6.13s/it]
Training Epoch: 15/100, batch 192/500 completed (loss: 0.012:  39%|███▉      | 97/250 [09:45<15:38,  6.14s/it]
Training Epoch: 15/100, batch 192/500 completed (loss: 0.0128:  39%|███▉      | 97/250 [09:45<15:38,  6.13s/it]
Training Epoch: 15/100, batch 193/500 completed (loss: 0.0751:  39%|███▉      | 97/250 [09:45<15:38,  6.13s/it]
Training Epoch: 15/100, batch 193/500 completed (loss: 0.0071:  39%|███▉      | 97/250 [09:45<15:38,  6.13s/it]
Training Epoch: 15/100, batch 193/500 completed (loss: 0.0365:  39%|███▉      | 97/250 [09:45<15:38,  6.13s/it]
Training Epoch: 15/100, batch 193/500 completed (loss: 0.0553:  39%|███▉      | 97/250 [09:45<15:38,  6.14s/it]
Training Epoch: 15/100, batch 194/500 completed (loss: 0.0063:  39%|███▉      | 97/250 [09:48<15:38,  6.13s/it]
Training Epoch: 15/100, batch 194/500 completed (loss: 0.03:  39%|███▉      | 97/250 [09:48<15:38,  6.13s/it]  
Training Epoch: 15/100, batch 194/500 completed (loss: 0.0145:  39%|███▉      | 97/250 [09:48<15:38,  6.13s/it]
Training Epoch: 15/100, batch 194/500 completed (loss: 0.1806:  39%|███▉      | 97/250 [09:48<15:38,  6.14s/it]
Training Epoch: 15/100, batch 194/500 completed (loss: 0.1806:  39%|███▉      | 98/250 [09:51<15:21,  6.06s/it]
Training Epoch: 15/100, batch 194/500 completed (loss: 0.0145:  39%|███▉      | 98/250 [09:51<15:22,  6.07s/it]
Training Epoch: 15/100, batch 194/500 completed (loss: 0.03:  39%|███▉      | 98/250 [09:51<15:22,  6.07s/it]
Training Epoch: 15/100, batch 194/500 completed (loss: 0.0063:  39%|███▉      | 98/250 [09:51<15:22,  6.07s/it]
Training Epoch: 15/100, batch 195/500 completed (loss: 0.0652:  39%|███▉      | 98/250 [09:51<15:22,  6.07s/it]
Training Epoch: 15/100, batch 195/500 completed (loss: 0.0202:  39%|███▉      | 98/250 [09:51<15:22,  6.07s/it]
Training Epoch: 15/100, batch 195/500 completed (loss: 0.0283:  39%|███▉      | 98/250 [09:51<15:22,  6.07s/it]
Training Epoch: 15/100, batch 195/500 completed (loss: 0.0554:  39%|███▉      | 98/250 [09:51<15:21,  6.06s/it]
Training Epoch: 15/100, batch 196/500 completed (loss: 0.0225:  39%|███▉      | 98/250 [09:54<15:22,  6.07s/it]
Training Epoch: 15/100, batch 196/500 completed (loss: 0.0146:  39%|███▉      | 98/250 [09:54<15:22,  6.07s/it]
Training Epoch: 15/100, batch 196/500 completed (loss: 0.0309:  39%|███▉      | 98/250 [09:54<15:22,  6.07s/it]
Training Epoch: 15/100, batch 196/500 completed (loss: 0.0233:  39%|███▉      | 98/250 [09:54<15:21,  6.06s/it]
Training Epoch: 15/100, batch 196/500 completed (loss: 0.0233:  40%|███▉      | 99/250 [09:57<15:15,  6.06s/it]
Training Epoch: 15/100, batch 196/500 completed (loss: 0.0225:  40%|███▉      | 99/250 [09:57<15:15,  6.06s/it]
Training Epoch: 15/100, batch 196/500 completed (loss: 0.0146:  40%|███▉      | 99/250 [09:57<15:15,  6.06s/it]
Training Epoch: 15/100, batch 196/500 completed (loss: 0.0309:  40%|███▉      | 99/250 [09:57<15:16,  6.07s/it]
Training Epoch: 15/100, batch 197/500 completed (loss: 0.0536:  40%|███▉      | 99/250 [09:57<15:15,  6.06s/it]
Training Epoch: 15/100, batch 197/500 completed (loss: 0.0091:  40%|███▉      | 99/250 [09:57<15:16,  6.07s/it]
Training Epoch: 15/100, batch 197/500 completed (loss: 0.0601:  40%|███▉      | 99/250 [09:57<15:15,  6.06s/it]
Training Epoch: 15/100, batch 197/500 completed (loss: 0.0606:  40%|███▉      | 99/250 [09:57<15:15,  6.06s/it]
Training Epoch: 15/100, batch 198/500 completed (loss: 0.6092:  40%|███▉      | 99/250 [10:00<15:15,  6.06s/it]
Training Epoch: 15/100, batch 198/500 completed (loss: 0.0622:  40%|███▉      | 99/250 [10:00<15:16,  6.07s/it]
Training Epoch: 15/100, batch 198/500 completed (loss: 0.1375:  40%|███▉      | 99/250 [10:00<15:15,  6.06s/it]
Training Epoch: 15/100, batch 198/500 completed (loss: 0.0279:  40%|███▉      | 99/250 [10:00<15:15,  6.06s/it]
Training Epoch: 15/100, batch 198/500 completed (loss: 0.6092:  40%|████      | 100/250 [10:03<15:14,  6.09s/it]
Training Epoch: 15/100, batch 198/500 completed (loss: 0.0622:  40%|████      | 100/250 [10:03<15:14,  6.10s/it]
Training Epoch: 15/100, batch 198/500 completed (loss: 0.1375:  40%|████      | 100/250 [10:03<15:14,  6.10s/it]
Training Epoch: 15/100, batch 198/500 completed (loss: 0.0279:  40%|████      | 100/250 [10:03<15:15,  6.10s/it]
Training Epoch: 15/100, batch 199/500 completed (loss: 0.0086:  40%|████      | 100/250 [10:03<15:14,  6.10s/it]
Training Epoch: 15/100, batch 199/500 completed (loss: 0.0794:  40%|████      | 100/250 [10:03<15:14,  6.09s/it]
Training Epoch: 15/100, batch 199/500 completed (loss: 0.1315:  40%|████      | 100/250 [10:03<15:14,  6.10s/it]
Training Epoch: 15/100, batch 199/500 completed (loss: 0.0137:  40%|████      | 100/250 [10:03<15:15,  6.10s/it]
Training Epoch: 15/100, batch 200/500 completed (loss: 0.005:  40%|████      | 100/250 [10:06<15:14,  6.09s/it] 
Training Epoch: 15/100, batch 200/500 completed (loss: 0.033:  40%|████      | 100/250 [10:06<15:14,  6.10s/it] 
Training Epoch: 15/100, batch 200/500 completed (loss: 0.0107:  40%|████      | 100/250 [10:06<15:14,  6.10s/it]
Training Epoch: 15/100, batch 200/500 completed (loss: 0.0649:  40%|████      | 100/250 [10:06<15:15,  6.10s/it]
Training Epoch: 15/100, batch 200/500 completed (loss: 0.005:  40%|████      | 101/250 [10:09<14:57,  6.02s/it]
Training Epoch: 15/100, batch 200/500 completed (loss: 0.0107:  40%|████      | 101/250 [10:09<14:57,  6.02s/it]
Training Epoch: 15/100, batch 200/500 completed (loss: 0.0649:  40%|████      | 101/250 [10:09<14:57,  6.02s/it]
Training Epoch: 15/100, batch 200/500 completed (loss: 0.033:  40%|████      | 101/250 [10:09<14:57,  6.02s/it]
Training Epoch: 15/100, batch 201/500 completed (loss: 0.1144:  40%|████      | 101/250 [10:09<14:57,  6.02s/it]
Training Epoch: 15/100, batch 201/500 completed (loss: 0.2321:  40%|████      | 101/250 [10:09<14:57,  6.02s/it]
Training Epoch: 15/100, batch 201/500 completed (loss: 0.0334:  40%|████      | 101/250 [10:09<14:57,  6.02s/it]
Training Epoch: 15/100, batch 201/500 completed (loss: 0.0099:  40%|████      | 101/250 [10:09<14:57,  6.02s/it]
Training Epoch: 15/100, batch 202/500 completed (loss: 0.0469:  40%|████      | 101/250 [10:12<14:57,  6.02s/it]
Training Epoch: 15/100, batch 202/500 completed (loss: 0.0059:  40%|████      | 101/250 [10:12<14:57,  6.02s/it]
Training Epoch: 15/100, batch 202/500 completed (loss: 0.0889:  40%|████      | 101/250 [10:12<14:57,  6.02s/it]
Training Epoch: 15/100, batch 202/500 completed (loss: 0.065:  40%|████      | 101/250 [10:12<14:57,  6.02s/it] 
Training Epoch: 15/100, batch 202/500 completed (loss: 0.065:  41%|████      | 102/250 [10:15<14:56,  6.06s/it]
Training Epoch: 15/100, batch 202/500 completed (loss: 0.0059:  41%|████      | 102/250 [10:15<14:57,  6.06s/it]
Training Epoch: 15/100, batch 202/500 completed (loss: 0.0469:  41%|████      | 102/250 [10:15<14:57,  6.06s/it]
Training Epoch: 15/100, batch 202/500 completed (loss: 0.0889:  41%|████      | 102/250 [10:15<14:57,  6.06s/it]
Training Epoch: 15/100, batch 203/500 completed (loss: 0.0307:  41%|████      | 102/250 [10:15<14:57,  6.06s/it]
Training Epoch: 15/100, batch 203/500 completed (loss: 0.0082:  41%|████      | 102/250 [10:15<14:57,  6.06s/it]
Training Epoch: 15/100, batch 203/500 completed (loss: 0.0655:  41%|████      | 102/250 [10:15<14:57,  6.06s/it]
Training Epoch: 15/100, batch 203/500 completed (loss: 0.1081:  41%|████      | 102/250 [10:15<14:56,  6.06s/it]
Training Epoch: 15/100, batch 204/500 completed (loss: 0.0996:  41%|████      | 102/250 [10:18<14:57,  6.06s/it]
Training Epoch: 15/100, batch 204/500 completed (loss: 0.0679:  41%|████      | 102/250 [10:18<14:57,  6.06s/it]
Training Epoch: 15/100, batch 204/500 completed (loss: 0.0937:  41%|████      | 102/250 [10:18<14:57,  6.06s/it]
Training Epoch: 15/100, batch 204/500 completed (loss: 0.0424:  41%|████      | 102/250 [10:18<14:56,  6.06s/it]
Training Epoch: 15/100, batch 204/500 completed (loss: 0.0679:  41%|████      | 103/250 [10:21<14:55,  6.09s/it]
Training Epoch: 15/100, batch 204/500 completed (loss: 0.0937:  41%|████      | 103/250 [10:21<14:55,  6.09s/it]
Training Epoch: 15/100, batch 204/500 completed (loss: 0.0996:  41%|████      | 103/250 [10:21<14:55,  6.09s/it]
Training Epoch: 15/100, batch 204/500 completed (loss: 0.0424:  41%|████      | 103/250 [10:21<14:56,  6.10s/it]
Training Epoch: 15/100, batch 205/500 completed (loss: 0.0259:  41%|████      | 103/250 [10:21<14:55,  6.09s/it]
Training Epoch: 15/100, batch 205/500 completed (loss: 0.019:  41%|████      | 103/250 [10:21<14:55,  6.09s/it] 
Training Epoch: 15/100, batch 205/500 completed (loss: 0.0679:  41%|████      | 103/250 [10:21<14:55,  6.09s/it]
Training Epoch: 15/100, batch 205/500 completed (loss: 0.01:  41%|████      | 103/250 [10:21<14:56,  6.10s/it]  
Training Epoch: 15/100, batch 206/500 completed (loss: 0.0303:  41%|████      | 103/250 [10:24<14:55,  6.09s/it]
Training Epoch: 15/100, batch 206/500 completed (loss: 0.0115:  41%|████      | 103/250 [10:24<14:55,  6.09s/it]
Training Epoch: 15/100, batch 206/500 completed (loss: 0.0254:  41%|████      | 103/250 [10:24<14:55,  6.09s/it]
Training Epoch: 15/100, batch 206/500 completed (loss: 0.0309:  41%|████      | 103/250 [10:24<14:56,  6.10s/it]
Training Epoch: 15/100, batch 206/500 completed (loss: 0.0254:  42%|████▏     | 104/250 [10:27<14:44,  6.06s/it]
Training Epoch: 15/100, batch 206/500 completed (loss: 0.0303:  42%|████▏     | 104/250 [10:27<14:44,  6.06s/it]
Training Epoch: 15/100, batch 206/500 completed (loss: 0.0115:  42%|████▏     | 104/250 [10:27<14:44,  6.06s/it]
Training Epoch: 15/100, batch 206/500 completed (loss: 0.0309:  42%|████▏     | 104/250 [10:27<14:45,  6.06s/it]
Training Epoch: 15/100, batch 207/500 completed (loss: 0.0566:  42%|████▏     | 104/250 [10:27<14:44,  6.06s/it]
Training Epoch: 15/100, batch 207/500 completed (loss: 0.0702:  42%|████▏     | 104/250 [10:27<14:44,  6.06s/it]
Training Epoch: 15/100, batch 207/500 completed (loss: 0.0659:  42%|████▏     | 104/250 [10:27<14:44,  6.06s/it]
Training Epoch: 15/100, batch 207/500 completed (loss: 0.1216:  42%|████▏     | 104/250 [10:27<14:45,  6.06s/it]
Training Epoch: 15/100, batch 208/500 completed (loss: 0.0948:  42%|████▏     | 104/250 [10:30<14:44,  6.06s/it]
Training Epoch: 15/100, batch 208/500 completed (loss: 0.0448:  42%|████▏     | 104/250 [10:30<14:44,  6.06s/it]
Training Epoch: 15/100, batch 208/500 completed (loss: 0.1429:  42%|████▏     | 104/250 [10:30<14:44,  6.06s/it]
Training Epoch: 15/100, batch 208/500 completed (loss: 0.0676:  42%|████▏     | 104/250 [10:30<14:45,  6.06s/it]
Training Epoch: 15/100, batch 208/500 completed (loss: 0.0676:  42%|████▏     | 105/250 [10:33<14:43,  6.09s/it]
Training Epoch: 15/100, batch 208/500 completed (loss: 0.0948:  42%|████▏     | 105/250 [10:33<14:43,  6.09s/it]
Training Epoch: 15/100, batch 208/500 completed (loss: 0.1429:  42%|████▏     | 105/250 [10:33<14:43,  6.09s/it]
Training Epoch: 15/100, batch 208/500 completed (loss: 0.0448:  42%|████▏     | 105/250 [10:33<14:43,  6.09s/it]
Training Epoch: 15/100, batch 209/500 completed (loss: 0.1585:  42%|████▏     | 105/250 [10:33<14:43,  6.09s/it]
Training Epoch: 15/100, batch 209/500 completed (loss: 0.0108:  42%|████▏     | 105/250 [10:33<14:43,  6.09s/it]
Training Epoch: 15/100, batch 209/500 completed (loss: 0.0132:  42%|████▏     | 105/250 [10:33<14:43,  6.09s/it]
Training Epoch: 15/100, batch 209/500 completed (loss: 0.0084:  42%|████▏     | 105/250 [10:33<14:43,  6.09s/it]
Training Epoch: 15/100, batch 210/500 completed (loss: 0.0227:  42%|████▏     | 105/250 [10:36<14:43,  6.09s/it]
Training Epoch: 15/100, batch 210/500 completed (loss: 0.0644:  42%|████▏     | 105/250 [10:37<14:43,  6.09s/it]
Training Epoch: 15/100, batch 210/500 completed (loss: 0.1618:  42%|████▏     | 105/250 [10:37<14:43,  6.09s/it]
Training Epoch: 15/100, batch 210/500 completed (loss: 0.0386:  42%|████▏     | 105/250 [10:37<14:43,  6.09s/it]
Training Epoch: 15/100, batch 210/500 completed (loss: 0.0386:  42%|████▏     | 106/250 [10:39<14:34,  6.07s/it]
Training Epoch: 15/100, batch 210/500 completed (loss: 0.0644:  42%|████▏     | 106/250 [10:39<14:34,  6.07s/it]
Training Epoch: 15/100, batch 210/500 completed (loss: 0.1618:  42%|████▏     | 106/250 [10:39<14:34,  6.07s/it]
Training Epoch: 15/100, batch 210/500 completed (loss: 0.0227:  42%|████▏     | 106/250 [10:39<14:34,  6.07s/it]
Training Epoch: 15/100, batch 211/500 completed (loss: 0.0223:  42%|████▏     | 106/250 [10:39<14:34,  6.07s/it]
Training Epoch: 15/100, batch 211/500 completed (loss: 0.0068:  42%|████▏     | 106/250 [10:39<14:34,  6.07s/it]
Training Epoch: 15/100, batch 211/500 completed (loss: 0.1044:  42%|████▏     | 106/250 [10:39<14:34,  6.07s/it]
Training Epoch: 15/100, batch 211/500 completed (loss: 0.0245:  42%|████▏     | 106/250 [10:39<14:34,  6.07s/it]
Training Epoch: 15/100, batch 212/500 completed (loss: 0.0425:  42%|████▏     | 106/250 [10:43<14:34,  6.07s/it]
Training Epoch: 15/100, batch 212/500 completed (loss: 0.015:  42%|████▏     | 106/250 [10:43<14:34,  6.07s/it] 
Training Epoch: 15/100, batch 212/500 completed (loss: 0.0098:  42%|████▏     | 106/250 [10:42<14:34,  6.07s/it]
Training Epoch: 15/100, batch 212/500 completed (loss: 0.1805:  42%|████▏     | 106/250 [10:43<14:34,  6.07s/it]
Training Epoch: 15/100, batch 212/500 completed (loss: 0.1805:  43%|████▎     | 107/250 [10:46<14:30,  6.09s/it]
Training Epoch: 15/100, batch 212/500 completed (loss: 0.0425:  43%|████▎     | 107/250 [10:46<14:30,  6.09s/it]
Training Epoch: 15/100, batch 212/500 completed (loss: 0.0098:  43%|████▎     | 107/250 [10:46<14:30,  6.09s/it]
Training Epoch: 15/100, batch 212/500 completed (loss: 0.015:  43%|████▎     | 107/250 [10:46<14:30,  6.09s/it]
Training Epoch: 15/100, batch 213/500 completed (loss: 0.0724:  43%|████▎     | 107/250 [10:46<14:30,  6.09s/it]
Training Epoch: 15/100, batch 213/500 completed (loss: 0.1783:  43%|████▎     | 107/250 [10:46<14:30,  6.09s/it]
Training Epoch: 15/100, batch 213/500 completed (loss: 0.0137:  43%|████▎     | 107/250 [10:46<14:30,  6.09s/it]
Training Epoch: 15/100, batch 213/500 completed (loss: 0.008:  43%|████▎     | 107/250 [10:46<14:30,  6.09s/it] 
Training Epoch: 15/100, batch 214/500 completed (loss: 0.0772:  43%|████▎     | 107/250 [10:49<14:30,  6.09s/it]
Training Epoch: 15/100, batch 214/500 completed (loss: 0.0058:  43%|████▎     | 107/250 [10:49<14:30,  6.09s/it]
Training Epoch: 15/100, batch 214/500 completed (loss: 0.145:  43%|████▎     | 107/250 [10:49<14:30,  6.09s/it] 
Training Epoch: 15/100, batch 214/500 completed (loss: 0.0695:  43%|████▎     | 107/250 [10:49<14:30,  6.09s/it]
Training Epoch: 15/100, batch 214/500 completed (loss: 0.0695:  43%|████▎     | 108/250 [10:52<14:27,  6.11s/it]
Training Epoch: 15/100, batch 214/500 completed (loss: 0.0058:  43%|████▎     | 108/250 [10:52<14:27,  6.11s/it]
Training Epoch: 15/100, batch 214/500 completed (loss: 0.145:  43%|████▎     | 108/250 [10:52<14:27,  6.11s/it]
Training Epoch: 15/100, batch 214/500 completed (loss: 0.0772:  43%|████▎     | 108/250 [10:52<14:27,  6.11s/it]
Training Epoch: 15/100, batch 215/500 completed (loss: 0.0346:  43%|████▎     | 108/250 [10:52<14:27,  6.11s/it]
Training Epoch: 15/100, batch 215/500 completed (loss: 0.0038:  43%|████▎     | 108/250 [10:52<14:27,  6.11s/it]
Training Epoch: 15/100, batch 215/500 completed (loss: 0.1403:  43%|████▎     | 108/250 [10:52<14:27,  6.11s/it]
Training Epoch: 15/100, batch 215/500 completed (loss: 0.0174:  43%|████▎     | 108/250 [10:52<14:27,  6.11s/it]
Training Epoch: 15/100, batch 216/500 completed (loss: 0.0661:  43%|████▎     | 108/250 [10:55<14:27,  6.11s/it]
Training Epoch: 15/100, batch 216/500 completed (loss: 0.2715:  43%|████▎     | 108/250 [10:55<14:27,  6.11s/it]
Training Epoch: 15/100, batch 216/500 completed (loss: 0.0217:  43%|████▎     | 108/250 [10:55<14:27,  6.11s/it]
Training Epoch: 15/100, batch 216/500 completed (loss: 0.2434:  43%|████▎     | 108/250 [10:55<14:27,  6.11s/it]
Training Epoch: 15/100, batch 216/500 completed (loss: 0.2715:  44%|████▎     | 109/250 [10:58<14:09,  6.03s/it]
Training Epoch: 15/100, batch 216/500 completed (loss: 0.0661:  44%|████▎     | 109/250 [10:58<14:09,  6.03s/it]
Training Epoch: 15/100, batch 216/500 completed (loss: 0.0217:  44%|████▎     | 109/250 [10:57<14:09,  6.03s/it]
Training Epoch: 15/100, batch 216/500 completed (loss: 0.2434:  44%|████▎     | 109/250 [10:58<14:10,  6.03s/it]
Training Epoch: 15/100, batch 217/500 completed (loss: 0.054:  44%|████▎     | 109/250 [10:58<14:09,  6.03s/it] 
Training Epoch: 15/100, batch 217/500 completed (loss: 0.0649:  44%|████▎     | 109/250 [10:58<14:09,  6.03s/it]
Training Epoch: 15/100, batch 217/500 completed (loss: 0.1399:  44%|████▎     | 109/250 [10:58<14:09,  6.03s/it]
Training Epoch: 15/100, batch 217/500 completed (loss: 0.1681:  44%|████▎     | 109/250 [10:58<14:10,  6.03s/it]
Training Epoch: 15/100, batch 218/500 completed (loss: 0.0946:  44%|████▎     | 109/250 [11:01<14:09,  6.03s/it]
Training Epoch: 15/100, batch 218/500 completed (loss: 0.0217:  44%|████▎     | 109/250 [11:01<14:09,  6.03s/it]
Training Epoch: 15/100, batch 218/500 completed (loss: 0.016:  44%|████▎     | 109/250 [11:01<14:09,  6.03s/it] 
Training Epoch: 15/100, batch 218/500 completed (loss: 0.0225:  44%|████▎     | 109/250 [11:01<14:10,  6.03s/it]
Training Epoch: 15/100, batch 218/500 completed (loss: 0.0225:  44%|████▍     | 110/250 [11:04<14:04,  6.03s/it]
Training Epoch: 15/100, batch 218/500 completed (loss: 0.016:  44%|████▍     | 110/250 [11:04<14:04,  6.03s/it]
Training Epoch: 15/100, batch 218/500 completed (loss: 0.0946:  44%|████▍     | 110/250 [11:04<14:04,  6.03s/it]
Training Epoch: 15/100, batch 218/500 completed (loss: 0.0217:  44%|████▍     | 110/250 [11:04<14:04,  6.03s/it]
Training Epoch: 15/100, batch 219/500 completed (loss: 0.0233:  44%|████▍     | 110/250 [11:04<14:04,  6.03s/it]
Training Epoch: 15/100, batch 219/500 completed (loss: 0.3804:  44%|████▍     | 110/250 [11:04<14:04,  6.03s/it]
Training Epoch: 15/100, batch 219/500 completed (loss: 0.0281:  44%|████▍     | 110/250 [11:04<14:04,  6.03s/it]
Training Epoch: 15/100, batch 219/500 completed (loss: 0.0187:  44%|████▍     | 110/250 [11:04<14:04,  6.03s/it]
Training Epoch: 15/100, batch 220/500 completed (loss: 0.2822:  44%|████▍     | 110/250 [11:07<14:04,  6.03s/it]
Training Epoch: 15/100, batch 220/500 completed (loss: 0.0679:  44%|████▍     | 110/250 [11:07<14:04,  6.03s/it]
Training Epoch: 15/100, batch 220/500 completed (loss: 0.012:  44%|████▍     | 110/250 [11:07<14:04,  6.03s/it] 
Training Epoch: 15/100, batch 220/500 completed (loss: 0.0373:  44%|████▍     | 110/250 [11:07<14:04,  6.03s/it]
Training Epoch: 15/100, batch 220/500 completed (loss: 0.012:  44%|████▍     | 111/250 [11:10<14:02,  6.06s/it]
Training Epoch: 15/100, batch 220/500 completed (loss: 0.0373:  44%|████▍     | 111/250 [11:10<14:02,  6.06s/it]
Training Epoch: 15/100, batch 220/500 completed (loss: 0.2822:  44%|████▍     | 111/250 [11:10<14:02,  6.06s/it]
Training Epoch: 15/100, batch 220/500 completed (loss: 0.0679:  44%|████▍     | 111/250 [11:10<14:02,  6.06s/it]
Training Epoch: 15/100, batch 221/500 completed (loss: 0.0314:  44%|████▍     | 111/250 [11:10<14:02,  6.06s/it]
Training Epoch: 15/100, batch 221/500 completed (loss: 0.0169:  44%|████▍     | 111/250 [11:10<14:02,  6.06s/it]
Training Epoch: 15/100, batch 221/500 completed (loss: 0.0657:  44%|████▍     | 111/250 [11:10<14:02,  6.06s/it]
Training Epoch: 15/100, batch 221/500 completed (loss: 0.0218:  44%|████▍     | 111/250 [11:10<14:02,  6.06s/it]
Training Epoch: 15/100, batch 222/500 completed (loss: 0.0848:  44%|████▍     | 111/250 [11:13<14:02,  6.06s/it]
Training Epoch: 15/100, batch 222/500 completed (loss: 0.0127:  44%|████▍     | 111/250 [11:13<14:02,  6.06s/it]
Training Epoch: 15/100, batch 222/500 completed (loss: 0.0581:  44%|████▍     | 111/250 [11:13<14:02,  6.06s/it]
Training Epoch: 15/100, batch 222/500 completed (loss: 0.0196:  44%|████▍     | 111/250 [11:13<14:02,  6.06s/it]
Training Epoch: 15/100, batch 222/500 completed (loss: 0.0196:  45%|████▍     | 112/250 [11:16<14:00,  6.09s/it]
Training Epoch: 15/100, batch 222/500 completed (loss: 0.0581:  45%|████▍     | 112/250 [11:16<14:00,  6.09s/it]
Training Epoch: 15/100, batch 222/500 completed (loss: 0.0127:  45%|████▍     | 112/250 [11:16<14:00,  6.09s/it]
Training Epoch: 15/100, batch 222/500 completed (loss: 0.0848:  45%|████▍     | 112/250 [11:16<14:01,  6.09s/it]
Training Epoch: 15/100, batch 223/500 completed (loss: 0.0165:  45%|████▍     | 112/250 [11:16<14:00,  6.09s/it]
Training Epoch: 15/100, batch 223/500 completed (loss: 0.0276:  45%|████▍     | 112/250 [11:16<14:00,  6.09s/it]
Training Epoch: 15/100, batch 223/500 completed (loss: 0.0116:  45%|████▍     | 112/250 [11:16<14:01,  6.09s/it]
Training Epoch: 15/100, batch 223/500 completed (loss: 0.1275:  45%|████▍     | 112/250 [11:16<14:00,  6.09s/it]
Training Epoch: 15/100, batch 224/500 completed (loss: 0.0388:  45%|████▍     | 112/250 [11:19<14:00,  6.09s/it]
Training Epoch: 15/100, batch 224/500 completed (loss: 0.0399:  45%|████▍     | 112/250 [11:19<14:01,  6.09s/it]
Training Epoch: 15/100, batch 224/500 completed (loss: 0.0079:  45%|████▍     | 112/250 [11:19<14:00,  6.09s/it]
Training Epoch: 15/100, batch 224/500 completed (loss: 0.0097:  45%|████▍     | 112/250 [11:19<14:00,  6.09s/it]
Training Epoch: 15/100, batch 224/500 completed (loss: 0.0097:  45%|████▌     | 113/250 [11:22<13:51,  6.07s/it]
Training Epoch: 15/100, batch 224/500 completed (loss: 0.0079:  45%|████▌     | 113/250 [11:22<13:51,  6.07s/it]
Training Epoch: 15/100, batch 224/500 completed (loss: 0.0399:  45%|████▌     | 113/250 [11:22<13:51,  6.07s/it]
Training Epoch: 15/100, batch 224/500 completed (loss: 0.0388:  45%|████▌     | 113/250 [11:22<13:51,  6.07s/it]
Training Epoch: 15/100, batch 225/500 completed (loss: 0.0079:  45%|████▌     | 113/250 [11:22<13:51,  6.07s/it]
Training Epoch: 15/100, batch 225/500 completed (loss: 0.0505:  45%|████▌     | 113/250 [11:22<13:51,  6.07s/it]
Training Epoch: 15/100, batch 225/500 completed (loss: 0.0463:  45%|████▌     | 113/250 [11:22<13:51,  6.07s/it]
Training Epoch: 15/100, batch 225/500 completed (loss: 0.1231:  45%|████▌     | 113/250 [11:22<13:51,  6.07s/it]
Training Epoch: 15/100, batch 226/500 completed (loss: 0.1689:  45%|████▌     | 113/250 [11:25<13:51,  6.07s/it]
Training Epoch: 15/100, batch 226/500 completed (loss: 0.1233:  45%|████▌     | 113/250 [11:25<13:51,  6.07s/it]
Training Epoch: 15/100, batch 226/500 completed (loss: 0.2813:  45%|████▌     | 113/250 [11:25<13:51,  6.07s/it]
Training Epoch: 15/100, batch 226/500 completed (loss: 0.3626:  45%|████▌     | 113/250 [11:25<13:51,  6.07s/it]
Training Epoch: 15/100, batch 226/500 completed (loss: 0.3626:  46%|████▌     | 114/250 [11:28<13:44,  6.06s/it]
Training Epoch: 15/100, batch 226/500 completed (loss: 0.2813:  46%|████▌     | 114/250 [11:28<13:44,  6.06s/it]
Training Epoch: 15/100, batch 226/500 completed (loss: 0.1689:  46%|████▌     | 114/250 [11:28<13:44,  6.07s/it]
Training Epoch: 15/100, batch 226/500 completed (loss: 0.1233:  46%|████▌     | 114/250 [11:28<13:45,  6.07s/it]
Training Epoch: 15/100, batch 227/500 completed (loss: 0.1133:  46%|████▌     | 114/250 [11:28<13:44,  6.06s/it]
Training Epoch: 15/100, batch 227/500 completed (loss: 0.1569:  46%|████▌     | 114/250 [11:28<13:45,  6.07s/it]
Training Epoch: 15/100, batch 227/500 completed (loss: 0.0305:  46%|████▌     | 114/250 [11:28<13:44,  6.07s/it]
Training Epoch: 15/100, batch 227/500 completed (loss: 0.0603:  46%|████▌     | 114/250 [11:28<13:44,  6.06s/it]
Training Epoch: 15/100, batch 228/500 completed (loss: 0.3592:  46%|████▌     | 114/250 [11:31<13:44,  6.06s/it]
Training Epoch: 15/100, batch 228/500 completed (loss: 0.1683:  46%|████▌     | 114/250 [11:31<13:45,  6.07s/it]
Training Epoch: 15/100, batch 228/500 completed (loss: 0.0281:  46%|████▌     | 114/250 [11:31<13:44,  6.07s/it]
Training Epoch: 15/100, batch 228/500 completed (loss: 0.2192:  46%|████▌     | 114/250 [11:31<13:44,  6.06s/it]
Training Epoch: 15/100, batch 228/500 completed (loss: 0.2192:  46%|████▌     | 115/250 [11:34<13:44,  6.11s/it]
Training Epoch: 15/100, batch 228/500 completed (loss: 0.3592:  46%|████▌     | 115/250 [11:34<13:44,  6.11s/it]
Training Epoch: 15/100, batch 228/500 completed (loss: 0.0281:  46%|████▌     | 115/250 [11:34<13:44,  6.11s/it]
Training Epoch: 15/100, batch 228/500 completed (loss: 0.1683:  46%|████▌     | 115/250 [11:34<13:44,  6.11s/it]
Training Epoch: 15/100, batch 229/500 completed (loss: 0.0061:  46%|████▌     | 115/250 [11:34<13:44,  6.11s/it]
Training Epoch: 15/100, batch 229/500 completed (loss: 0.1565:  46%|████▌     | 115/250 [11:34<13:44,  6.11s/it]
Training Epoch: 15/100, batch 229/500 completed (loss: 0.0199:  46%|████▌     | 115/250 [11:34<13:44,  6.11s/it]
Training Epoch: 15/100, batch 229/500 completed (loss: 0.0151:  46%|████▌     | 115/250 [11:34<13:44,  6.11s/it]
Training Epoch: 15/100, batch 230/500 completed (loss: 0.0071:  46%|████▌     | 115/250 [11:37<13:44,  6.11s/it]
Training Epoch: 15/100, batch 230/500 completed (loss: 0.1365:  46%|████▌     | 115/250 [11:37<13:44,  6.11s/it]
Training Epoch: 15/100, batch 230/500 completed (loss: 0.0234:  46%|████▌     | 115/250 [11:37<13:44,  6.11s/it]
Training Epoch: 15/100, batch 230/500 completed (loss: 0.0028:  46%|████▌     | 115/250 [11:37<13:44,  6.11s/it]
Training Epoch: 15/100, batch 230/500 completed (loss: 0.0234:  46%|████▋     | 116/250 [11:40<13:41,  6.13s/it]
Training Epoch: 15/100, batch 230/500 completed (loss: 0.0028:  46%|████▋     | 116/250 [11:40<13:41,  6.13s/it]
Training Epoch: 15/100, batch 230/500 completed (loss: 0.1365:  46%|████▋     | 116/250 [11:40<13:41,  6.13s/it]
Training Epoch: 15/100, batch 230/500 completed (loss: 0.0071:  46%|████▋     | 116/250 [11:40<13:41,  6.13s/it]
Training Epoch: 15/100, batch 231/500 completed (loss: 0.0987:  46%|████▋     | 116/250 [11:40<13:41,  6.13s/it]
Training Epoch: 15/100, batch 231/500 completed (loss: 0.0302:  46%|████▋     | 116/250 [11:40<13:41,  6.13s/it]
Training Epoch: 15/100, batch 231/500 completed (loss: 0.1126:  46%|████▋     | 116/250 [11:40<13:41,  6.13s/it]
Training Epoch: 15/100, batch 231/500 completed (loss: 0.6149:  46%|████▋     | 116/250 [11:40<13:41,  6.13s/it]
Training Epoch: 15/100, batch 232/500 completed (loss: 0.0816:  46%|████▋     | 116/250 [11:43<13:41,  6.13s/it]
Training Epoch: 15/100, batch 232/500 completed (loss: 0.0783:  46%|████▋     | 116/250 [11:43<13:41,  6.13s/it]
Training Epoch: 15/100, batch 232/500 completed (loss: 0.015:  46%|████▋     | 116/250 [11:43<13:41,  6.13s/it] 
Training Epoch: 15/100, batch 232/500 completed (loss: 0.0293:  46%|████▋     | 116/250 [11:43<13:41,  6.13s/it]
Training Epoch: 15/100, batch 232/500 completed (loss: 0.0783:  47%|████▋     | 117/250 [11:46<13:37,  6.14s/it]
Training Epoch: 15/100, batch 232/500 completed (loss: 0.0816:  47%|████▋     | 117/250 [11:46<13:37,  6.14s/it]
Training Epoch: 15/100, batch 232/500 completed (loss: 0.0293:  47%|████▋     | 117/250 [11:47<13:37,  6.15s/it]
Training Epoch: 15/100, batch 232/500 completed (loss: 0.015:  47%|████▋     | 117/250 [11:46<13:37,  6.14s/it]
Training Epoch: 15/100, batch 233/500 completed (loss: 0.0146:  47%|████▋     | 117/250 [11:47<13:37,  6.14s/it]
Training Epoch: 15/100, batch 233/500 completed (loss: 0.0513:  47%|████▋     | 117/250 [11:47<13:37,  6.14s/it]
Training Epoch: 15/100, batch 233/500 completed (loss: 0.0597:  47%|████▋     | 117/250 [11:47<13:37,  6.14s/it]
Training Epoch: 15/100, batch 233/500 completed (loss: 0.1374:  47%|████▋     | 117/250 [11:47<13:37,  6.15s/it]
Training Epoch: 15/100, batch 234/500 completed (loss: 0.0105:  47%|████▋     | 117/250 [11:49<13:37,  6.14s/it]
Training Epoch: 15/100, batch 234/500 completed (loss: 0.0558:  47%|████▋     | 117/250 [11:49<13:37,  6.14s/it]
Training Epoch: 15/100, batch 234/500 completed (loss: 0.0381:  47%|████▋     | 117/250 [11:49<13:37,  6.14s/it]
Training Epoch: 15/100, batch 234/500 completed (loss: 0.0192:  47%|████▋     | 117/250 [11:49<13:37,  6.15s/it]
Training Epoch: 15/100, batch 234/500 completed (loss: 0.0192:  47%|████▋     | 118/250 [11:53<13:25,  6.10s/it]
Training Epoch: 15/100, batch 234/500 completed (loss: 0.0105:  47%|████▋     | 118/250 [11:52<13:25,  6.10s/it]
Training Epoch: 15/100, batch 234/500 completed (loss: 0.0381:  47%|████▋     | 118/250 [11:53<13:25,  6.11s/it]
Training Epoch: 15/100, batch 234/500 completed (loss: 0.0558:  47%|████▋     | 118/250 [11:53<13:25,  6.10s/it]
Training Epoch: 15/100, batch 235/500 completed (loss: 0.0967:  47%|████▋     | 118/250 [11:53<13:25,  6.10s/it]
Training Epoch: 15/100, batch 235/500 completed (loss: 0.0588:  47%|████▋     | 118/250 [11:53<13:25,  6.11s/it]
Training Epoch: 15/100, batch 235/500 completed (loss: 0.0251:  47%|████▋     | 118/250 [11:53<13:25,  6.10s/it]
Training Epoch: 15/100, batch 235/500 completed (loss: 0.078:  47%|████▋     | 118/250 [11:53<13:25,  6.10s/it] 
Training Epoch: 15/100, batch 236/500 completed (loss: 0.1716:  47%|████▋     | 118/250 [11:55<13:25,  6.10s/it]
Training Epoch: 15/100, batch 236/500 completed (loss: 0.2573:  47%|████▋     | 118/250 [11:55<13:25,  6.10s/it]
Training Epoch: 15/100, batch 236/500 completed (loss: 0.0033:  47%|████▋     | 118/250 [11:55<13:25,  6.11s/it]
Training Epoch: 15/100, batch 236/500 completed (loss: 0.0264:  47%|████▋     | 118/250 [11:56<13:25,  6.10s/it]
Training Epoch: 15/100, batch 236/500 completed (loss: 0.0033:  48%|████▊     | 119/250 [11:59<13:16,  6.08s/it]
Training Epoch: 15/100, batch 236/500 completed (loss: 0.1716:  48%|████▊     | 119/250 [11:59<13:15,  6.08s/it]
Training Epoch: 15/100, batch 236/500 completed (loss: 0.2573:  48%|████▊     | 119/250 [11:59<13:16,  6.08s/it]
Training Epoch: 15/100, batch 236/500 completed (loss: 0.0264:  48%|████▊     | 119/250 [11:59<13:16,  6.08s/it]
Training Epoch: 15/100, batch 237/500 completed (loss: 0.0196:  48%|████▊     | 119/250 [11:59<13:16,  6.08s/it]
Training Epoch: 15/100, batch 237/500 completed (loss: 0.0562:  48%|████▊     | 119/250 [11:59<13:16,  6.08s/it]
Training Epoch: 15/100, batch 237/500 completed (loss: 0.0206:  48%|████▊     | 119/250 [11:59<13:15,  6.08s/it]
Training Epoch: 15/100, batch 237/500 completed (loss: 0.3728:  48%|████▊     | 119/250 [11:59<13:16,  6.08s/it]
Training Epoch: 15/100, batch 238/500 completed (loss: 0.0049:  48%|████▊     | 119/250 [12:01<13:15,  6.08s/it]
Training Epoch: 15/100, batch 238/500 completed (loss: 0.0119:  48%|████▊     | 119/250 [12:01<13:16,  6.08s/it]
Training Epoch: 15/100, batch 238/500 completed (loss: 0.0172:  48%|████▊     | 119/250 [12:01<13:16,  6.08s/it]
Training Epoch: 15/100, batch 238/500 completed (loss: 0.1067:  48%|████▊     | 119/250 [12:02<13:16,  6.08s/it]
Training Epoch: 15/100, batch 238/500 completed (loss: 0.1067:  48%|████▊     | 120/250 [12:05<13:10,  6.08s/it]
Training Epoch: 15/100, batch 238/500 completed (loss: 0.0119:  48%|████▊     | 120/250 [12:05<13:10,  6.08s/it]
Training Epoch: 15/100, batch 238/500 completed (loss: 0.0172:  48%|████▊     | 120/250 [12:05<13:10,  6.08s/it]
Training Epoch: 15/100, batch 238/500 completed (loss: 0.0049:  48%|████▊     | 120/250 [12:05<13:10,  6.08s/it]
Training Epoch: 15/100, batch 239/500 completed (loss: 0.0362:  48%|████▊     | 120/250 [12:05<13:10,  6.08s/it]
Training Epoch: 15/100, batch 239/500 completed (loss: 0.0156:  48%|████▊     | 120/250 [12:05<13:10,  6.08s/it]
Training Epoch: 15/100, batch 239/500 completed (loss: 0.0335:  48%|████▊     | 120/250 [12:05<13:10,  6.08s/it]
Training Epoch: 15/100, batch 239/500 completed (loss: 0.0277:  48%|████▊     | 120/250 [12:05<13:10,  6.08s/it]
Training Epoch: 15/100, batch 240/500 completed (loss: 0.061:  48%|████▊     | 120/250 [12:08<13:10,  6.08s/it] 
Training Epoch: 15/100, batch 240/500 completed (loss: 0.0054:  48%|████▊     | 120/250 [12:08<13:10,  6.08s/it]
Training Epoch: 15/100, batch 240/500 completed (loss: 0.0166:  48%|████▊     | 120/250 [12:08<13:10,  6.08s/it]
Training Epoch: 15/100, batch 240/500 completed (loss: 0.3607:  48%|████▊     | 120/250 [12:08<13:10,  6.08s/it]
Training Epoch: 15/100, batch 240/500 completed (loss: 0.3607:  48%|████▊     | 121/250 [12:11<13:01,  6.06s/it]
Training Epoch: 15/100, batch 240/500 completed (loss: 0.0166:  48%|████▊     | 121/250 [12:11<13:01,  6.06s/it]
Training Epoch: 15/100, batch 240/500 completed (loss: 0.061:  48%|████▊     | 121/250 [12:11<13:01,  6.06s/it]
Training Epoch: 15/100, batch 240/500 completed (loss: 0.0054:  48%|████▊     | 121/250 [12:11<13:01,  6.06s/it]
Training Epoch: 15/100, batch 241/500 completed (loss: 0.0186:  48%|████▊     | 121/250 [12:11<13:01,  6.06s/it]
Training Epoch: 15/100, batch 241/500 completed (loss: 0.0827:  48%|████▊     | 121/250 [12:11<13:01,  6.06s/it]
Training Epoch: 15/100, batch 241/500 completed (loss: 0.0108:  48%|████▊     | 121/250 [12:11<13:01,  6.06s/it]
Training Epoch: 15/100, batch 241/500 completed (loss: 0.054:  48%|████▊     | 121/250 [12:11<13:01,  6.06s/it] 
Training Epoch: 15/100, batch 242/500 completed (loss: 0.1084:  48%|████▊     | 121/250 [12:14<13:01,  6.06s/it]
Training Epoch: 15/100, batch 242/500 completed (loss: 0.0119:  48%|████▊     | 121/250 [12:14<13:01,  6.06s/it]
Training Epoch: 15/100, batch 242/500 completed (loss: 0.0433:  48%|████▊     | 121/250 [12:14<13:01,  6.06s/it]
Training Epoch: 15/100, batch 242/500 completed (loss: 0.1707:  48%|████▊     | 121/250 [12:14<13:01,  6.06s/it]
Training Epoch: 15/100, batch 242/500 completed (loss: 0.1707:  49%|████▉     | 122/250 [12:17<12:53,  6.04s/it]
Training Epoch: 15/100, batch 242/500 completed (loss: 0.0119:  49%|████▉     | 122/250 [12:17<12:53,  6.04s/it]
Training Epoch: 15/100, batch 242/500 completed (loss: 0.0433:  49%|████▉     | 122/250 [12:17<12:53,  6.04s/it]
Training Epoch: 15/100, batch 242/500 completed (loss: 0.1084:  49%|████▉     | 122/250 [12:17<12:54,  6.05s/it]
Training Epoch: 15/100, batch 243/500 completed (loss: 0.0177:  49%|████▉     | 122/250 [12:17<12:53,  6.04s/it]
Training Epoch: 15/100, batch 243/500 completed (loss: 0.2738:  49%|████▉     | 122/250 [12:17<12:54,  6.05s/it]
Training Epoch: 15/100, batch 243/500 completed (loss: 0.0159:  49%|████▉     | 122/250 [12:17<12:53,  6.04s/it]
Training Epoch: 15/100, batch 243/500 completed (loss: 0.0198:  49%|████▉     | 122/250 [12:17<12:53,  6.04s/it]
Training Epoch: 15/100, batch 244/500 completed (loss: 0.1669:  49%|████▉     | 122/250 [12:20<12:54,  6.05s/it]
Training Epoch: 15/100, batch 244/500 completed (loss: 0.0361:  49%|████▉     | 122/250 [12:20<12:53,  6.04s/it]
Training Epoch: 15/100, batch 244/500 completed (loss: 0.0095:  49%|████▉     | 122/250 [12:20<12:53,  6.04s/it]
Training Epoch: 15/100, batch 244/500 completed (loss: 0.5569:  49%|████▉     | 122/250 [12:20<12:53,  6.04s/it]
Training Epoch: 15/100, batch 244/500 completed (loss: 0.5569:  49%|████▉     | 123/250 [12:23<12:50,  6.07s/it]
Training Epoch: 15/100, batch 244/500 completed (loss: 0.1669:  49%|████▉     | 123/250 [12:23<12:51,  6.07s/it]
Training Epoch: 15/100, batch 244/500 completed (loss: 0.0095:  49%|████▉     | 123/250 [12:23<12:51,  6.07s/it]
Training Epoch: 15/100, batch 244/500 completed (loss: 0.0361:  49%|████▉     | 123/250 [12:23<12:51,  6.07s/it]
Training Epoch: 15/100, batch 245/500 completed (loss: 0.0586:  49%|████▉     | 123/250 [12:23<12:51,  6.07s/it]
Training Epoch: 15/100, batch 245/500 completed (loss: 0.0789:  49%|████▉     | 123/250 [12:23<12:51,  6.07s/it]
Training Epoch: 15/100, batch 245/500 completed (loss: 0.0087:  49%|████▉     | 123/250 [12:23<12:51,  6.07s/it]
Training Epoch: 15/100, batch 245/500 completed (loss: 0.0241:  49%|████▉     | 123/250 [12:23<12:50,  6.07s/it]
Training Epoch: 15/100, batch 246/500 completed (loss: 0.2836:  49%|████▉     | 123/250 [12:26<12:51,  6.07s/it]
Training Epoch: 15/100, batch 246/500 completed (loss: 0.0133:  49%|████▉     | 123/250 [12:26<12:51,  6.07s/it]
Training Epoch: 15/100, batch 246/500 completed (loss: 0.0296:  49%|████▉     | 123/250 [12:26<12:51,  6.07s/it]
Training Epoch: 15/100, batch 246/500 completed (loss: 0.0121:  49%|████▉     | 123/250 [12:26<12:50,  6.07s/it]
Training Epoch: 15/100, batch 246/500 completed (loss: 0.0121:  50%|████▉     | 124/250 [12:29<12:48,  6.10s/it]
Training Epoch: 15/100, batch 246/500 completed (loss: 0.0133:  50%|████▉     | 124/250 [12:29<12:48,  6.10s/it]
Training Epoch: 15/100, batch 246/500 completed (loss: 0.0296:  50%|████▉     | 124/250 [12:29<12:48,  6.10s/it]
Training Epoch: 15/100, batch 246/500 completed (loss: 0.2836:  50%|████▉     | 124/250 [12:29<12:48,  6.10s/it]
Training Epoch: 15/100, batch 247/500 completed (loss: 0.0341:  50%|████▉     | 124/250 [12:29<12:48,  6.10s/it]
Training Epoch: 15/100, batch 247/500 completed (loss: 0.111:  50%|████▉     | 124/250 [12:29<12:48,  6.10s/it] 
Training Epoch: 15/100, batch 247/500 completed (loss: 0.0345:  50%|████▉     | 124/250 [12:29<12:48,  6.10s/it]
Training Epoch: 15/100, batch 247/500 completed (loss: 0.0246:  50%|████▉     | 124/250 [12:29<12:48,  6.10s/it]
Training Epoch: 15/100, batch 248/500 completed (loss: 0.014:  50%|████▉     | 124/250 [12:32<12:48,  6.10s/it]
Training Epoch: 15/100, batch 248/500 completed (loss: 0.0364:  50%|████▉     | 124/250 [12:32<12:48,  6.10s/it]
Training Epoch: 15/100, batch 248/500 completed (loss: 0.0216:  50%|████▉     | 124/250 [12:32<12:48,  6.10s/it]
Training Epoch: 15/100, batch 248/500 completed (loss: 0.0194:  50%|████▉     | 124/250 [12:32<12:48,  6.10s/it]
Training Epoch: 15/100, batch 248/500 completed (loss: 0.0364:  50%|█████     | 125/250 [12:35<12:44,  6.12s/it]
Training Epoch: 15/100, batch 248/500 completed (loss: 0.0216:  50%|█████     | 125/250 [12:35<12:44,  6.12s/it]
Training Epoch: 15/100, batch 248/500 completed (loss: 0.014:  50%|█████     | 125/250 [12:35<12:44,  6.12s/it]
Training Epoch: 15/100, batch 248/500 completed (loss: 0.0194:  50%|█████     | 125/250 [12:35<12:45,  6.12s/it]
Training Epoch: 15/100, batch 249/500 completed (loss: 0.0903:  50%|█████     | 125/250 [12:35<12:44,  6.12s/it]
Training Epoch: 15/100, batch 249/500 completed (loss: 0.031:  50%|█████     | 125/250 [12:35<12:44,  6.12s/it]
Training Epoch: 15/100, batch 249/500 completed (loss: 0.0224:  50%|█████     | 125/250 [12:35<12:44,  6.12s/it]
Training Epoch: 15/100, batch 249/500 completed (loss: 0.243:  50%|█████     | 125/250 [12:35<12:45,  6.12s/it] 
Training Epoch: 15/100, batch 250/500 completed (loss: 0.1932:  50%|█████     | 125/250 [12:38<12:44,  6.12s/it]
Training Epoch: 15/100, batch 250/500 completed (loss: 0.0517:  50%|█████     | 125/250 [12:38<12:44,  6.12s/it]
Training Epoch: 15/100, batch 250/500 completed (loss: 0.0486:  50%|█████     | 125/250 [12:38<12:44,  6.12s/it]
Training Epoch: 15/100, batch 250/500 completed (loss: 0.0244:  50%|█████     | 125/250 [12:38<12:45,  6.12s/it]
Training Epoch: 15/100, batch 250/500 completed (loss: 0.1932:  50%|█████     | 126/250 [12:41<12:34,  6.08s/it]
Training Epoch: 15/100, batch 250/500 completed (loss: 0.0244:  50%|█████     | 126/250 [12:41<12:34,  6.08s/it]
Training Epoch: 15/100, batch 250/500 completed (loss: 0.0486:  50%|█████     | 126/250 [12:41<12:34,  6.08s/it]
Training Epoch: 15/100, batch 250/500 completed (loss: 0.0517:  50%|█████     | 126/250 [12:41<12:34,  6.08s/it]
Training Epoch: 15/100, batch 251/500 completed (loss: 0.1829:  50%|█████     | 126/250 [12:41<12:34,  6.08s/it]
Training Epoch: 15/100, batch 251/500 completed (loss: 0.0437:  50%|█████     | 126/250 [12:41<12:34,  6.08s/it]
Training Epoch: 15/100, batch 251/500 completed (loss: 0.0991:  50%|█████     | 126/250 [12:41<12:34,  6.08s/it]
Training Epoch: 15/100, batch 251/500 completed (loss: 0.275:  50%|█████     | 126/250 [12:41<12:34,  6.08s/it] 
Training Epoch: 15/100, batch 252/500 completed (loss: 0.1147:  50%|█████     | 126/250 [12:44<12:34,  6.08s/it]
Training Epoch: 15/100, batch 252/500 completed (loss: 0.0267:  50%|█████     | 126/250 [12:44<12:34,  6.08s/it]
Training Epoch: 15/100, batch 252/500 completed (loss: 0.0351:  50%|█████     | 126/250 [12:44<12:34,  6.08s/it]
Training Epoch: 15/100, batch 252/500 completed (loss: 0.059:  50%|█████     | 126/250 [12:44<12:34,  6.08s/it]
Training Epoch: 15/100, batch 252/500 completed (loss: 0.059:  51%|█████     | 127/250 [12:47<12:25,  6.06s/it]
Training Epoch: 15/100, batch 252/500 completed (loss: 0.0267:  51%|█████     | 127/250 [12:47<12:25,  6.06s/it]
Training Epoch: 15/100, batch 252/500 completed (loss: 0.1147:  51%|█████     | 127/250 [12:47<12:25,  6.06s/it]
Training Epoch: 15/100, batch 252/500 completed (loss: 0.0351:  51%|█████     | 127/250 [12:47<12:25,  6.06s/it]
Training Epoch: 15/100, batch 253/500 completed (loss: 0.4971:  51%|█████     | 127/250 [12:47<12:25,  6.06s/it]
Training Epoch: 15/100, batch 253/500 completed (loss: 0.0119:  51%|█████     | 127/250 [12:47<12:25,  6.06s/it]
Training Epoch: 15/100, batch 253/500 completed (loss: 0.012:  51%|█████     | 127/250 [12:47<12:25,  6.06s/it] 
Training Epoch: 15/100, batch 253/500 completed (loss: 0.031:  51%|█████     | 127/250 [12:47<12:25,  6.06s/it]
Training Epoch: 15/100, batch 254/500 completed (loss: 0.1063:  51%|█████     | 127/250 [12:50<12:25,  6.06s/it]
Training Epoch: 15/100, batch 254/500 completed (loss: 0.0206:  51%|█████     | 127/250 [12:50<12:25,  6.06s/it]
Training Epoch: 15/100, batch 254/500 completed (loss: 0.4374:  51%|█████     | 127/250 [12:50<12:25,  6.06s/it]
Training Epoch: 15/100, batch 254/500 completed (loss: 0.1264:  51%|█████     | 127/250 [12:50<12:25,  6.06s/it]
Training Epoch: 15/100, batch 254/500 completed (loss: 0.4374:  51%|█████     | 128/250 [12:53<12:22,  6.09s/it]
Training Epoch: 15/100, batch 254/500 completed (loss: 0.0206:  51%|█████     | 128/250 [12:53<12:22,  6.09s/it]
Training Epoch: 15/100, batch 254/500 completed (loss: 0.1063:  51%|█████     | 128/250 [12:53<12:22,  6.09s/it]
Training Epoch: 15/100, batch 254/500 completed (loss: 0.1264:  51%|█████     | 128/250 [12:53<12:23,  6.09s/it]
Training Epoch: 15/100, batch 255/500 completed (loss: 0.0317:  51%|█████     | 128/250 [12:53<12:22,  6.09s/it]
Training Epoch: 15/100, batch 255/500 completed (loss: 0.5165:  51%|█████     | 128/250 [12:53<12:22,  6.09s/it]
Training Epoch: 15/100, batch 255/500 completed (loss: 0.395:  51%|█████     | 128/250 [12:53<12:22,  6.09s/it] 
Training Epoch: 15/100, batch 255/500 completed (loss: 0.0144:  51%|█████     | 128/250 [12:53<12:23,  6.09s/it]
Training Epoch: 15/100, batch 256/500 completed (loss: 0.0072:  51%|█████     | 128/250 [12:56<12:22,  6.09s/it]
Training Epoch: 15/100, batch 256/500 completed (loss: 0.2485:  51%|█████     | 128/250 [12:56<12:22,  6.09s/it]
Training Epoch: 15/100, batch 256/500 completed (loss: 0.0403:  51%|█████     | 128/250 [12:56<12:22,  6.09s/it]
Training Epoch: 15/100, batch 256/500 completed (loss: 0.386:  51%|█████     | 128/250 [12:56<12:23,  6.09s/it] 
Training Epoch: 15/100, batch 256/500 completed (loss: 0.386:  52%|█████▏    | 129/250 [12:59<12:11,  6.05s/it]
Training Epoch: 15/100, batch 256/500 completed (loss: 0.0072:  52%|█████▏    | 129/250 [12:59<12:12,  6.05s/it]
Training Epoch: 15/100, batch 256/500 completed (loss: 0.2485:  52%|█████▏    | 129/250 [12:59<12:12,  6.05s/it]
Training Epoch: 15/100, batch 256/500 completed (loss: 0.0403:  52%|█████▏    | 129/250 [12:59<12:12,  6.05s/it]
Training Epoch: 15/100, batch 257/500 completed (loss: 0.0412:  52%|█████▏    | 129/250 [12:59<12:12,  6.05s/it]
Training Epoch: 15/100, batch 257/500 completed (loss: 0.0111:  52%|█████▏    | 129/250 [12:59<12:12,  6.05s/it]
Training Epoch: 15/100, batch 257/500 completed (loss: 0.226:  52%|█████▏    | 129/250 [12:59<12:12,  6.05s/it] 
Training Epoch: 15/100, batch 257/500 completed (loss: 0.1069:  52%|█████▏    | 129/250 [12:59<12:11,  6.05s/it]
Training Epoch: 15/100, batch 258/500 completed (loss: 0.1287:  52%|█████▏    | 129/250 [13:02<12:12,  6.05s/it]
Training Epoch: 15/100, batch 258/500 completed (loss: 0.0495:  52%|█████▏    | 129/250 [13:02<12:12,  6.05s/it]
Training Epoch: 15/100, batch 258/500 completed (loss: 0.0926:  52%|█████▏    | 129/250 [13:02<12:12,  6.05s/it]
Training Epoch: 15/100, batch 258/500 completed (loss: 0.053:  52%|█████▏    | 129/250 [13:02<12:11,  6.05s/it] 
Training Epoch: 15/100, batch 258/500 completed (loss: 0.0926:  52%|█████▏    | 130/250 [13:05<12:13,  6.11s/it]
Training Epoch: 15/100, batch 258/500 completed (loss: 0.053:  52%|█████▏    | 130/250 [13:06<12:13,  6.11s/it]
Training Epoch: 15/100, batch 258/500 completed (loss: 0.1287:  52%|█████▏    | 130/250 [13:05<12:14,  6.12s/it]
Training Epoch: 15/100, batch 258/500 completed (loss: 0.0495:  52%|█████▏    | 130/250 [13:05<12:14,  6.12s/it]
Training Epoch: 15/100, batch 259/500 completed (loss: 0.0397:  52%|█████▏    | 130/250 [13:06<12:13,  6.11s/it]
Training Epoch: 15/100, batch 259/500 completed (loss: 0.0258:  52%|█████▏    | 130/250 [13:06<12:14,  6.12s/it]
Training Epoch: 15/100, batch 259/500 completed (loss: 0.1795:  52%|█████▏    | 130/250 [13:06<12:14,  6.12s/it]
Training Epoch: 15/100, batch 259/500 completed (loss: 0.2473:  52%|█████▏    | 130/250 [13:06<12:13,  6.11s/it]
Training Epoch: 15/100, batch 260/500 completed (loss: 0.0714:  52%|█████▏    | 130/250 [13:09<12:14,  6.12s/it]
Training Epoch: 15/100, batch 260/500 completed (loss: 0.3896:  52%|█████▏    | 130/250 [13:09<12:13,  6.11s/it]
Training Epoch: 15/100, batch 260/500 completed (loss: 0.0097:  52%|█████▏    | 130/250 [13:09<12:14,  6.12s/it]
Training Epoch: 15/100, batch 260/500 completed (loss: 0.0138:  52%|█████▏    | 130/250 [13:09<12:13,  6.11s/it]
Training Epoch: 15/100, batch 260/500 completed (loss: 0.0138:  52%|█████▏    | 131/250 [13:12<12:03,  6.08s/it]
Training Epoch: 15/100, batch 260/500 completed (loss: 0.0714:  52%|█████▏    | 131/250 [13:11<12:03,  6.08s/it]
Training Epoch: 15/100, batch 260/500 completed (loss: 0.3896:  52%|█████▏    | 131/250 [13:11<12:03,  6.08s/it]
Training Epoch: 15/100, batch 260/500 completed (loss: 0.0097:  52%|█████▏    | 131/250 [13:11<12:03,  6.08s/it]
Training Epoch: 15/100, batch 261/500 completed (loss: 0.1027:  52%|█████▏    | 131/250 [13:12<12:03,  6.08s/it]
Training Epoch: 15/100, batch 261/500 completed (loss: 0.0289:  52%|█████▏    | 131/250 [13:12<12:03,  6.08s/it]
Training Epoch: 15/100, batch 261/500 completed (loss: 0.0551:  52%|█████▏    | 131/250 [13:12<12:03,  6.08s/it]
Training Epoch: 15/100, batch 261/500 completed (loss: 0.0229:  52%|█████▏    | 131/250 [13:12<12:03,  6.08s/it]
Training Epoch: 15/100, batch 262/500 completed (loss: 0.1629:  52%|█████▏    | 131/250 [13:15<12:03,  6.08s/it]
Training Epoch: 15/100, batch 262/500 completed (loss: 0.2359:  52%|█████▏    | 131/250 [13:15<12:03,  6.08s/it]
Training Epoch: 15/100, batch 262/500 completed (loss: 0.1127:  52%|█████▏    | 131/250 [13:15<12:03,  6.08s/it]
Training Epoch: 15/100, batch 262/500 completed (loss: 0.2277:  52%|█████▏    | 131/250 [13:15<12:03,  6.08s/it]
Training Epoch: 15/100, batch 262/500 completed (loss: 0.2277:  53%|█████▎    | 132/250 [13:18<11:54,  6.06s/it]
Training Epoch: 15/100, batch 262/500 completed (loss: 0.1629:  53%|█████▎    | 132/250 [13:17<11:54,  6.06s/it]
Training Epoch: 15/100, batch 262/500 completed (loss: 0.2359:  53%|█████▎    | 132/250 [13:17<11:55,  6.06s/it]
Training Epoch: 15/100, batch 262/500 completed (loss: 0.1127:  53%|█████▎    | 132/250 [13:17<11:55,  6.06s/it]
Training Epoch: 15/100, batch 263/500 completed (loss: 0.1726:  53%|█████▎    | 132/250 [13:18<11:55,  6.06s/it]
Training Epoch: 15/100, batch 263/500 completed (loss: 0.3572:  53%|█████▎    | 132/250 [13:18<11:55,  6.06s/it]
Training Epoch: 15/100, batch 263/500 completed (loss: 0.0886:  53%|█████▎    | 132/250 [13:18<11:54,  6.06s/it]
Training Epoch: 15/100, batch 263/500 completed (loss: 0.2849:  53%|█████▎    | 132/250 [13:18<11:54,  6.06s/it]
Training Epoch: 15/100, batch 264/500 completed (loss: 0.0157:  53%|█████▎    | 132/250 [13:21<11:55,  6.06s/it]
Training Epoch: 15/100, batch 264/500 completed (loss: 0.0123:  53%|█████▎    | 132/250 [13:21<11:55,  6.06s/it]
Training Epoch: 15/100, batch 264/500 completed (loss: 0.4345:  53%|█████▎    | 132/250 [13:21<11:54,  6.06s/it]
Training Epoch: 15/100, batch 264/500 completed (loss: 0.0801:  53%|█████▎    | 132/250 [13:21<11:54,  6.06s/it]
Training Epoch: 15/100, batch 264/500 completed (loss: 0.0123:  53%|█████▎    | 133/250 [13:24<11:59,  6.15s/it]
Training Epoch: 15/100, batch 264/500 completed (loss: 0.0801:  53%|█████▎    | 133/250 [13:24<11:59,  6.15s/it]
Training Epoch: 15/100, batch 264/500 completed (loss: 0.4345:  53%|█████▎    | 133/250 [13:24<11:59,  6.15s/it]
Training Epoch: 15/100, batch 264/500 completed (loss: 0.0157:  53%|█████▎    | 133/250 [13:24<11:59,  6.15s/it]
Training Epoch: 15/100, batch 265/500 completed (loss: 0.0073:  53%|█████▎    | 133/250 [13:24<11:59,  6.15s/it]
Training Epoch: 15/100, batch 265/500 completed (loss: 0.1536:  53%|█████▎    | 133/250 [13:24<11:59,  6.15s/it]
Training Epoch: 15/100, batch 265/500 completed (loss: 0.2564:  53%|█████▎    | 133/250 [13:24<11:59,  6.15s/it]
Training Epoch: 15/100, batch 265/500 completed (loss: 0.0431:  53%|█████▎    | 133/250 [13:24<11:59,  6.15s/it]
Training Epoch: 15/100, batch 266/500 completed (loss: 0.2417:  53%|█████▎    | 133/250 [13:27<11:59,  6.15s/it]
Training Epoch: 15/100, batch 266/500 completed (loss: 0.1259:  53%|█████▎    | 133/250 [13:27<11:59,  6.15s/it]
Training Epoch: 15/100, batch 266/500 completed (loss: 0.1766:  53%|█████▎    | 133/250 [13:27<11:59,  6.15s/it]
Training Epoch: 15/100, batch 266/500 completed (loss: 0.0685:  53%|█████▎    | 133/250 [13:27<11:59,  6.15s/it]
Training Epoch: 15/100, batch 266/500 completed (loss: 0.0685:  54%|█████▎    | 134/250 [13:30<11:47,  6.10s/it]
Training Epoch: 15/100, batch 266/500 completed (loss: 0.2417:  54%|█████▎    | 134/250 [13:30<11:47,  6.10s/it]
Training Epoch: 15/100, batch 266/500 completed (loss: 0.1766:  54%|█████▎    | 134/250 [13:30<11:48,  6.10s/it]
Training Epoch: 15/100, batch 266/500 completed (loss: 0.1259:  54%|█████▎    | 134/250 [13:30<11:48,  6.11s/it]
Training Epoch: 15/100, batch 267/500 completed (loss: 0.0016:  54%|█████▎    | 134/250 [13:30<11:48,  6.10s/it]
Training Epoch: 15/100, batch 267/500 completed (loss: 0.3603:  54%|█████▎    | 134/250 [13:30<11:48,  6.11s/it]
Training Epoch: 15/100, batch 267/500 completed (loss: 0.0262:  54%|█████▎    | 134/250 [13:30<11:47,  6.10s/it]
Training Epoch: 15/100, batch 267/500 completed (loss: 0.0374:  54%|█████▎    | 134/250 [13:30<11:47,  6.10s/it]
Training Epoch: 15/100, batch 268/500 completed (loss: 0.0326:  54%|█████▎    | 134/250 [13:33<11:47,  6.10s/it]
Training Epoch: 15/100, batch 268/500 completed (loss: 0.0588:  54%|█████▎    | 134/250 [13:33<11:48,  6.10s/it]
Training Epoch: 15/100, batch 268/500 completed (loss: 0.0261:  54%|█████▎    | 134/250 [13:33<11:48,  6.11s/it]
Training Epoch: 15/100, batch 268/500 completed (loss: 0.0145:  54%|█████▎    | 134/250 [13:33<11:47,  6.10s/it]
Training Epoch: 15/100, batch 268/500 completed (loss: 0.0261:  54%|█████▍    | 135/250 [13:36<11:40,  6.09s/it]
Training Epoch: 15/100, batch 268/500 completed (loss: 0.0326:  54%|█████▍    | 135/250 [13:36<11:40,  6.09s/it]
Training Epoch: 15/100, batch 268/500 completed (loss: 0.0145:  54%|█████▍    | 135/250 [13:36<11:40,  6.09s/it]
Training Epoch: 15/100, batch 268/500 completed (loss: 0.0588:  54%|█████▍    | 135/250 [13:36<11:40,  6.09s/it]
Training Epoch: 15/100, batch 269/500 completed (loss: 0.0243:  54%|█████▍    | 135/250 [13:36<11:40,  6.09s/it]
Training Epoch: 15/100, batch 269/500 completed (loss: 0.0147:  54%|█████▍    | 135/250 [13:36<11:40,  6.09s/it]
Training Epoch: 15/100, batch 269/500 completed (loss: 0.0879:  54%|█████▍    | 135/250 [13:36<11:40,  6.09s/it]
Training Epoch: 15/100, batch 269/500 completed (loss: 0.0347:  54%|█████▍    | 135/250 [13:36<11:40,  6.09s/it]
Training Epoch: 15/100, batch 270/500 completed (loss: 0.0089:  54%|█████▍    | 135/250 [13:39<11:40,  6.09s/it]
Training Epoch: 15/100, batch 270/500 completed (loss: 0.0436:  54%|█████▍    | 135/250 [13:39<11:40,  6.09s/it]
Training Epoch: 15/100, batch 270/500 completed (loss: 0.0592:  54%|█████▍    | 135/250 [13:39<11:40,  6.09s/it]
Training Epoch: 15/100, batch 270/500 completed (loss: 0.0627:  54%|█████▍    | 135/250 [13:39<11:40,  6.09s/it]
Training Epoch: 15/100, batch 270/500 completed (loss: 0.0627:  54%|█████▍    | 136/250 [13:42<11:37,  6.12s/it]
Training Epoch: 15/100, batch 270/500 completed (loss: 0.0436:  54%|█████▍    | 136/250 [13:42<11:37,  6.12s/it]
Training Epoch: 15/100, batch 270/500 completed (loss: 0.0089:  54%|█████▍    | 136/250 [13:42<11:37,  6.12s/it]
Training Epoch: 15/100, batch 270/500 completed (loss: 0.0592:  54%|█████▍    | 136/250 [13:42<11:37,  6.12s/it]
Training Epoch: 15/100, batch 271/500 completed (loss: 0.4519:  54%|█████▍    | 136/250 [13:42<11:37,  6.12s/it]
Training Epoch: 15/100, batch 271/500 completed (loss: 0.0233:  54%|█████▍    | 136/250 [13:42<11:37,  6.12s/it]
Training Epoch: 15/100, batch 271/500 completed (loss: 0.3595:  54%|█████▍    | 136/250 [13:42<11:37,  6.12s/it]
Training Epoch: 15/100, batch 271/500 completed (loss: 0.0243:  54%|█████▍    | 136/250 [13:42<11:37,  6.12s/it]
Training Epoch: 15/100, batch 272/500 completed (loss: 0.0083:  54%|█████▍    | 136/250 [13:45<11:37,  6.12s/it]
Training Epoch: 15/100, batch 272/500 completed (loss: 0.0678:  54%|█████▍    | 136/250 [13:45<11:37,  6.12s/it]
Training Epoch: 15/100, batch 272/500 completed (loss: 0.0143:  54%|█████▍    | 136/250 [13:45<11:37,  6.12s/it]
Training Epoch: 15/100, batch 272/500 completed (loss: 0.0642:  54%|█████▍    | 136/250 [13:45<11:37,  6.12s/it]
Training Epoch: 15/100, batch 272/500 completed (loss: 0.0642:  55%|█████▍    | 137/250 [13:48<11:26,  6.08s/it]
Training Epoch: 15/100, batch 272/500 completed (loss: 0.0678:  55%|█████▍    | 137/250 [13:48<11:27,  6.08s/it]
Training Epoch: 15/100, batch 272/500 completed (loss: 0.0083:  55%|█████▍    | 137/250 [13:48<11:27,  6.08s/it]
Training Epoch: 15/100, batch 272/500 completed (loss: 0.0143:  55%|█████▍    | 137/250 [13:48<11:27,  6.08s/it]
Training Epoch: 15/100, batch 273/500 completed (loss: 0.0183:  55%|█████▍    | 137/250 [13:48<11:27,  6.08s/it]
Training Epoch: 15/100, batch 273/500 completed (loss: 0.1598:  55%|█████▍    | 137/250 [13:48<11:27,  6.08s/it]
Training Epoch: 15/100, batch 273/500 completed (loss: 0.0982:  55%|█████▍    | 137/250 [13:48<11:27,  6.08s/it]
Training Epoch: 15/100, batch 273/500 completed (loss: 0.1119:  55%|█████▍    | 137/250 [13:48<11:26,  6.08s/it]
Training Epoch: 15/100, batch 274/500 completed (loss: 0.034:  55%|█████▍    | 137/250 [13:51<11:27,  6.08s/it] 
Training Epoch: 15/100, batch 274/500 completed (loss: 0.1467:  55%|█████▍    | 137/250 [13:51<11:27,  6.08s/it]
Training Epoch: 15/100, batch 274/500 completed (loss: 0.0178:  55%|█████▍    | 137/250 [13:51<11:27,  6.08s/it]
Training Epoch: 15/100, batch 274/500 completed (loss: 0.2183:  55%|█████▍    | 137/250 [13:51<11:26,  6.08s/it]
Training Epoch: 15/100, batch 274/500 completed (loss: 0.1467:  55%|█████▌    | 138/250 [13:54<11:22,  6.09s/it]
Training Epoch: 15/100, batch 274/500 completed (loss: 0.2183:  55%|█████▌    | 138/250 [13:54<11:22,  6.10s/it]
Training Epoch: 15/100, batch 274/500 completed (loss: 0.0178:  55%|█████▌    | 138/250 [13:54<11:22,  6.10s/it]
Training Epoch: 15/100, batch 274/500 completed (loss: 0.034:  55%|█████▌    | 138/250 [13:54<11:22,  6.10s/it]
Training Epoch: 15/100, batch 275/500 completed (loss: 0.04:  55%|█████▌    | 138/250 [13:54<11:22,  6.10s/it] 
Training Epoch: 15/100, batch 275/500 completed (loss: 0.3037:  55%|█████▌    | 138/250 [13:54<11:22,  6.10s/it]
Training Epoch: 15/100, batch 275/500 completed (loss: 0.0042:  55%|█████▌    | 138/250 [13:54<11:22,  6.09s/it]
Training Epoch: 15/100, batch 275/500 completed (loss: 0.0741:  55%|█████▌    | 138/250 [13:54<11:22,  6.10s/it]
Training Epoch: 15/100, batch 276/500 completed (loss: 0.1403:  55%|█████▌    | 138/250 [13:57<11:22,  6.10s/it]
Training Epoch: 15/100, batch 276/500 completed (loss: 0.0109:  55%|█████▌    | 138/250 [13:57<11:22,  6.10s/it]
Training Epoch: 15/100, batch 276/500 completed (loss: 0.0811:  55%|█████▌    | 138/250 [13:57<11:22,  6.09s/it]
Training Epoch: 15/100, batch 276/500 completed (loss: 0.051:  55%|█████▌    | 138/250 [13:57<11:22,  6.10s/it] 
Training Epoch: 15/100, batch 276/500 completed (loss: 0.051:  56%|█████▌    | 139/250 [14:00<11:18,  6.12s/it]
Training Epoch: 15/100, batch 276/500 completed (loss: 0.0811:  56%|█████▌    | 139/250 [14:00<11:18,  6.12s/it]
Training Epoch: 15/100, batch 276/500 completed (loss: 0.1403:  56%|█████▌    | 139/250 [14:00<11:18,  6.12s/it]
Training Epoch: 15/100, batch 276/500 completed (loss: 0.0109:  56%|█████▌    | 139/250 [14:00<11:18,  6.11s/it]
Training Epoch: 15/100, batch 277/500 completed (loss: 0.0765:  56%|█████▌    | 139/250 [14:00<11:18,  6.12s/it]
Training Epoch: 15/100, batch 277/500 completed (loss: 0.0137:  56%|█████▌    | 139/250 [14:00<11:18,  6.12s/it]
Training Epoch: 15/100, batch 277/500 completed (loss: 0.0715:  56%|█████▌    | 139/250 [14:00<11:18,  6.11s/it]
Training Epoch: 15/100, batch 277/500 completed (loss: 0.0189:  56%|█████▌    | 139/250 [14:00<11:18,  6.12s/it]
Training Epoch: 15/100, batch 278/500 completed (loss: 0.0221:  56%|█████▌    | 139/250 [14:03<11:18,  6.12s/it]
Training Epoch: 15/100, batch 278/500 completed (loss: 0.1169:  56%|█████▌    | 139/250 [14:03<11:18,  6.12s/it]
Training Epoch: 15/100, batch 278/500 completed (loss: 0.2801:  56%|█████▌    | 139/250 [14:03<11:18,  6.11s/it]
Training Epoch: 15/100, batch 278/500 completed (loss: 0.0221:  56%|█████▌    | 139/250 [14:03<11:18,  6.12s/it]
Training Epoch: 15/100, batch 278/500 completed (loss: 0.0221:  56%|█████▌    | 140/250 [14:06<11:04,  6.04s/it]
Training Epoch: 15/100, batch 278/500 completed (loss: 0.2801:  56%|█████▌    | 140/250 [14:06<11:04,  6.04s/it]
Training Epoch: 15/100, batch 278/500 completed (loss: 0.1169:  56%|█████▌    | 140/250 [14:06<11:04,  6.04s/it]
Training Epoch: 15/100, batch 278/500 completed (loss: 0.0221:  56%|█████▌    | 140/250 [14:06<11:04,  6.04s/it]
Training Epoch: 15/100, batch 279/500 completed (loss: 0.0768:  56%|█████▌    | 140/250 [14:06<11:04,  6.04s/it]
Training Epoch: 15/100, batch 279/500 completed (loss: 0.0628:  56%|█████▌    | 140/250 [14:06<11:04,  6.04s/it]
Training Epoch: 15/100, batch 279/500 completed (loss: 0.0417:  56%|█████▌    | 140/250 [14:06<11:04,  6.04s/it]
Training Epoch: 15/100, batch 279/500 completed (loss: 0.0668:  56%|█████▌    | 140/250 [14:06<11:04,  6.04s/it]
Training Epoch: 15/100, batch 280/500 completed (loss: 0.14:  56%|█████▌    | 140/250 [14:09<11:04,  6.04s/it]  
Training Epoch: 15/100, batch 280/500 completed (loss: 0.0617:  56%|█████▌    | 140/250 [14:09<11:04,  6.04s/it]
Training Epoch: 15/100, batch 280/500 completed (loss: 0.167:  56%|█████▌    | 140/250 [14:09<11:04,  6.04s/it] 
Training Epoch: 15/100, batch 280/500 completed (loss: 0.0351:  56%|█████▌    | 140/250 [14:09<11:04,  6.04s/it]
Training Epoch: 15/100, batch 280/500 completed (loss: 0.14:  56%|█████▋    | 141/250 [14:12<11:01,  6.07s/it]
Training Epoch: 15/100, batch 280/500 completed (loss: 0.0351:  56%|█████▋    | 141/250 [14:12<11:02,  6.07s/it]
Training Epoch: 15/100, batch 280/500 completed (loss: 0.167:  56%|█████▋    | 141/250 [14:12<11:01,  6.07s/it]
Training Epoch: 15/100, batch 280/500 completed (loss: 0.0617:  56%|█████▋    | 141/250 [14:12<11:01,  6.07s/it]
Training Epoch: 15/100, batch 281/500 completed (loss: 0.0255:  56%|█████▋    | 141/250 [14:12<11:01,  6.07s/it]
Training Epoch: 15/100, batch 281/500 completed (loss: 0.0794:  56%|█████▋    | 141/250 [14:12<11:01,  6.07s/it]
Training Epoch: 15/100, batch 281/500 completed (loss: 0.0691:  56%|█████▋    | 141/250 [14:12<11:01,  6.07s/it]
Training Epoch: 15/100, batch 281/500 completed (loss: 0.0022:  56%|█████▋    | 141/250 [14:12<11:02,  6.07s/it]
Training Epoch: 15/100, batch 282/500 completed (loss: 0.053:  56%|█████▋    | 141/250 [14:15<11:01,  6.07s/it] 
Training Epoch: 15/100, batch 282/500 completed (loss: 0.2846:  56%|█████▋    | 141/250 [14:15<11:01,  6.07s/it]
Training Epoch: 15/100, batch 282/500 completed (loss: 0.0311:  56%|█████▋    | 141/250 [14:15<11:01,  6.07s/it]
Training Epoch: 15/100, batch 282/500 completed (loss: 0.087:  56%|█████▋    | 141/250 [14:16<11:02,  6.07s/it] 
Training Epoch: 15/100, batch 282/500 completed (loss: 0.2846:  57%|█████▋    | 142/250 [14:18<10:52,  6.04s/it]
Training Epoch: 15/100, batch 282/500 completed (loss: 0.087:  57%|█████▋    | 142/250 [14:18<10:53,  6.05s/it]
Training Epoch: 15/100, batch 282/500 completed (loss: 0.053:  57%|█████▋    | 142/250 [14:18<10:53,  6.05s/it]
Training Epoch: 15/100, batch 282/500 completed (loss: 0.0311:  57%|█████▋    | 142/250 [14:18<10:53,  6.05s/it]
Training Epoch: 15/100, batch 283/500 completed (loss: 0.0093:  57%|█████▋    | 142/250 [14:18<10:53,  6.05s/it]
Training Epoch: 15/100, batch 283/500 completed (loss: 0.0753:  57%|█████▋    | 142/250 [14:18<10:52,  6.04s/it]
Training Epoch: 15/100, batch 283/500 completed (loss: 0.113:  57%|█████▋    | 142/250 [14:18<10:53,  6.05s/it] 
Training Epoch: 15/100, batch 283/500 completed (loss: 0.019:  57%|█████▋    | 142/250 [14:18<10:53,  6.05s/it]
Training Epoch: 15/100, batch 284/500 completed (loss: 0.0352:  57%|█████▋    | 142/250 [14:21<10:53,  6.05s/it]
Training Epoch: 15/100, batch 284/500 completed (loss: 0.0831:  57%|█████▋    | 142/250 [14:21<10:52,  6.04s/it]
Training Epoch: 15/100, batch 284/500 completed (loss: 0.0207:  57%|█████▋    | 142/250 [14:21<10:53,  6.05s/it]
Training Epoch: 15/100, batch 284/500 completed (loss: 0.0038:  57%|█████▋    | 142/250 [14:21<10:53,  6.05s/it]
Training Epoch: 15/100, batch 284/500 completed (loss: 0.0352:  57%|█████▋    | 143/250 [14:24<10:45,  6.04s/it]
Training Epoch: 15/100, batch 284/500 completed (loss: 0.0038:  57%|█████▋    | 143/250 [14:24<10:46,  6.04s/it]
Training Epoch: 15/100, batch 284/500 completed (loss: 0.0207:  57%|█████▋    | 143/250 [14:24<10:46,  6.04s/it]
Training Epoch: 15/100, batch 284/500 completed (loss: 0.0831:  57%|█████▋    | 143/250 [14:24<10:46,  6.04s/it]
Training Epoch: 15/100, batch 285/500 completed (loss: 0.2861:  57%|█████▋    | 143/250 [14:24<10:46,  6.04s/it]
Training Epoch: 15/100, batch 285/500 completed (loss: 0.0285:  57%|█████▋    | 143/250 [14:24<10:45,  6.04s/it]
Training Epoch: 15/100, batch 285/500 completed (loss: 0.0377:  57%|█████▋    | 143/250 [14:24<10:46,  6.04s/it]
Training Epoch: 15/100, batch 285/500 completed (loss: 0.0223:  57%|█████▋    | 143/250 [14:24<10:46,  6.04s/it]
Training Epoch: 15/100, batch 286/500 completed (loss: 0.0506:  57%|█████▋    | 143/250 [14:28<10:46,  6.04s/it]
Training Epoch: 15/100, batch 286/500 completed (loss: 0.0356:  57%|█████▋    | 143/250 [14:28<10:46,  6.04s/it]
Training Epoch: 15/100, batch 286/500 completed (loss: 0.0121:  57%|█████▋    | 143/250 [14:28<10:45,  6.04s/it]
Training Epoch: 15/100, batch 286/500 completed (loss: 0.1156:  57%|█████▋    | 143/250 [14:28<10:46,  6.04s/it]
Training Epoch: 15/100, batch 286/500 completed (loss: 0.1156:  58%|█████▊    | 144/250 [14:31<10:48,  6.12s/it]
Training Epoch: 15/100, batch 286/500 completed (loss: 0.0506:  58%|█████▊    | 144/250 [14:31<10:48,  6.12s/it]
Training Epoch: 15/100, batch 286/500 completed (loss: 0.0121:  58%|█████▊    | 144/250 [14:31<10:48,  6.12s/it]
Training Epoch: 15/100, batch 286/500 completed (loss: 0.0356:  58%|█████▊    | 144/250 [14:31<10:49,  6.12s/it]
Training Epoch: 15/100, batch 287/500 completed (loss: 0.3439:  58%|█████▊    | 144/250 [14:31<10:48,  6.12s/it]
Training Epoch: 15/100, batch 287/500 completed (loss: 0.0414:  58%|█████▊    | 144/250 [14:31<10:48,  6.12s/it]
Training Epoch: 15/100, batch 287/500 completed (loss: 0.0121:  58%|█████▊    | 144/250 [14:31<10:49,  6.12s/it]
Training Epoch: 15/100, batch 287/500 completed (loss: 0.0144:  58%|█████▊    | 144/250 [14:31<10:48,  6.12s/it]
Training Epoch: 15/100, batch 288/500 completed (loss: 0.0028:  58%|█████▊    | 144/250 [14:34<10:49,  6.12s/it]
Training Epoch: 15/100, batch 288/500 completed (loss: 0.0651:  58%|█████▊    | 144/250 [14:34<10:48,  6.12s/it]
Training Epoch: 15/100, batch 288/500 completed (loss: 0.1155:  58%|█████▊    | 144/250 [14:34<10:48,  6.12s/it]
Training Epoch: 15/100, batch 288/500 completed (loss: 0.0149:  58%|█████▊    | 144/250 [14:34<10:48,  6.12s/it]
Training Epoch: 15/100, batch 288/500 completed (loss: 0.0149:  58%|█████▊    | 145/250 [14:37<10:34,  6.05s/it]
Training Epoch: 15/100, batch 288/500 completed (loss: 0.0028:  58%|█████▊    | 145/250 [14:37<10:35,  6.05s/it]
Training Epoch: 15/100, batch 288/500 completed (loss: 0.1155:  58%|█████▊    | 145/250 [14:37<10:35,  6.05s/it]
Training Epoch: 15/100, batch 288/500 completed (loss: 0.0651:  58%|█████▊    | 145/250 [14:37<10:35,  6.06s/it]
Training Epoch: 15/100, batch 289/500 completed (loss: 0.0246:  58%|█████▊    | 145/250 [14:37<10:35,  6.05s/it]
Training Epoch: 15/100, batch 289/500 completed (loss: 0.1187:  58%|█████▊    | 145/250 [14:37<10:35,  6.05s/it]
Training Epoch: 15/100, batch 289/500 completed (loss: 0.0794:  58%|█████▊    | 145/250 [14:37<10:35,  6.06s/it]
Training Epoch: 15/100, batch 289/500 completed (loss: 0.0172:  58%|█████▊    | 145/250 [14:37<10:34,  6.05s/it]
Training Epoch: 15/100, batch 290/500 completed (loss: 0.006:  58%|█████▊    | 145/250 [14:40<10:35,  6.05s/it] 
Training Epoch: 15/100, batch 290/500 completed (loss: 0.0166:  58%|█████▊    | 145/250 [14:40<10:35,  6.06s/it]
Training Epoch: 15/100, batch 290/500 completed (loss: 0.018:  58%|█████▊    | 145/250 [14:40<10:35,  6.05s/it] 
Training Epoch: 15/100, batch 290/500 completed (loss: 0.0278:  58%|█████▊    | 145/250 [14:40<10:34,  6.05s/it]
Training Epoch: 15/100, batch 290/500 completed (loss: 0.0278:  58%|█████▊    | 146/250 [14:43<10:38,  6.14s/it]
Training Epoch: 15/100, batch 290/500 completed (loss: 0.0166:  58%|█████▊    | 146/250 [14:43<10:38,  6.13s/it]
Training Epoch: 15/100, batch 290/500 completed (loss: 0.018:  58%|█████▊    | 146/250 [14:43<10:38,  6.14s/it]
Training Epoch: 15/100, batch 290/500 completed (loss: 0.006:  58%|█████▊    | 146/250 [14:43<10:38,  6.14s/it]
Training Epoch: 15/100, batch 291/500 completed (loss: 0.2338:  58%|█████▊    | 146/250 [14:43<10:38,  6.14s/it]
Training Epoch: 15/100, batch 291/500 completed (loss: 0.0142:  58%|█████▊    | 146/250 [14:43<10:38,  6.13s/it]
Training Epoch: 15/100, batch 291/500 completed (loss: 0.0523:  58%|█████▊    | 146/250 [14:43<10:38,  6.14s/it]
Training Epoch: 15/100, batch 291/500 completed (loss: 0.0312:  58%|█████▊    | 146/250 [14:43<10:38,  6.14s/it]
Training Epoch: 15/100, batch 292/500 completed (loss: 0.3253:  58%|█████▊    | 146/250 [14:46<10:38,  6.14s/it]
Training Epoch: 15/100, batch 292/500 completed (loss: 0.2994:  58%|█████▊    | 146/250 [14:46<10:38,  6.14s/it]
Training Epoch: 15/100, batch 292/500 completed (loss: 0.2187:  58%|█████▊    | 146/250 [14:46<10:38,  6.13s/it]
Training Epoch: 15/100, batch 292/500 completed (loss: 0.0263:  58%|█████▊    | 146/250 [14:46<10:38,  6.14s/it]
Training Epoch: 15/100, batch 292/500 completed (loss: 0.0263:  59%|█████▉    | 147/250 [14:49<10:27,  6.09s/it]
Training Epoch: 15/100, batch 292/500 completed (loss: 0.3253:  59%|█████▉    | 147/250 [14:49<10:27,  6.09s/it]
Training Epoch: 15/100, batch 292/500 completed (loss: 0.2187:  59%|█████▉    | 147/250 [14:49<10:27,  6.09s/it]
Training Epoch: 15/100, batch 292/500 completed (loss: 0.2994:  59%|█████▉    | 147/250 [14:49<10:27,  6.09s/it]
Training Epoch: 15/100, batch 293/500 completed (loss: 0.4707:  59%|█████▉    | 147/250 [14:49<10:27,  6.09s/it]
Training Epoch: 15/100, batch 293/500 completed (loss: 0.1584:  59%|█████▉    | 147/250 [14:49<10:27,  6.09s/it]
Training Epoch: 15/100, batch 293/500 completed (loss: 0.028:  59%|█████▉    | 147/250 [14:49<10:27,  6.09s/it] 
Training Epoch: 15/100, batch 293/500 completed (loss: 0.0272:  59%|█████▉    | 147/250 [14:49<10:27,  6.09s/it]
Training Epoch: 15/100, batch 294/500 completed (loss: 0.1288:  59%|█████▉    | 147/250 [14:52<10:27,  6.09s/it]
Training Epoch: 15/100, batch 294/500 completed (loss: 0.076:  59%|█████▉    | 147/250 [14:52<10:27,  6.09s/it] 
Training Epoch: 15/100, batch 294/500 completed (loss: 0.0261:  59%|█████▉    | 147/250 [14:52<10:27,  6.09s/it]
Training Epoch: 15/100, batch 294/500 completed (loss: 0.0131:  59%|█████▉    | 147/250 [14:52<10:27,  6.09s/it]
Training Epoch: 15/100, batch 294/500 completed (loss: 0.0131:  59%|█████▉    | 148/250 [14:55<10:18,  6.06s/it]
Training Epoch: 15/100, batch 294/500 completed (loss: 0.1288:  59%|█████▉    | 148/250 [14:55<10:18,  6.06s/it]
Training Epoch: 15/100, batch 294/500 completed (loss: 0.076:  59%|█████▉    | 148/250 [14:55<10:18,  6.06s/it]
Training Epoch: 15/100, batch 294/500 completed (loss: 0.0261:  59%|█████▉    | 148/250 [14:55<10:18,  6.06s/it]
Training Epoch: 15/100, batch 295/500 completed (loss: 0.131:  59%|█████▉    | 148/250 [14:55<10:18,  6.06s/it] 
Training Epoch: 15/100, batch 295/500 completed (loss: 0.1262:  59%|█████▉    | 148/250 [14:55<10:18,  6.06s/it]
Training Epoch: 15/100, batch 295/500 completed (loss: 0.0249:  59%|█████▉    | 148/250 [14:55<10:18,  6.06s/it]
Training Epoch: 15/100, batch 295/500 completed (loss: 0.1567:  59%|█████▉    | 148/250 [14:55<10:18,  6.06s/it]
Training Epoch: 15/100, batch 296/500 completed (loss: 0.1536:  59%|█████▉    | 148/250 [14:58<10:18,  6.06s/it]
Training Epoch: 15/100, batch 296/500 completed (loss: 0.04:  59%|█████▉    | 148/250 [14:58<10:18,  6.06s/it]  
Training Epoch: 15/100, batch 296/500 completed (loss: 0.0511:  59%|█████▉    | 148/250 [14:58<10:18,  6.06s/it]
Training Epoch: 15/100, batch 296/500 completed (loss: 0.3301:  59%|█████▉    | 148/250 [14:58<10:18,  6.06s/it]
Training Epoch: 15/100, batch 296/500 completed (loss: 0.04:  60%|█████▉    | 149/250 [15:01<10:18,  6.13s/it]
Training Epoch: 15/100, batch 296/500 completed (loss: 0.3301:  60%|█████▉    | 149/250 [15:01<10:19,  6.14s/it]
Training Epoch: 15/100, batch 296/500 completed (loss: 0.0511:  60%|█████▉    | 149/250 [15:01<10:19,  6.13s/it]
Training Epoch: 15/100, batch 296/500 completed (loss: 0.1536:  60%|█████▉    | 149/250 [15:01<10:19,  6.13s/it]
Training Epoch: 15/100, batch 297/500 completed (loss: 0.0462:  60%|█████▉    | 149/250 [15:01<10:19,  6.13s/it]
Training Epoch: 15/100, batch 297/500 completed (loss: 0.4289:  60%|█████▉    | 149/250 [15:01<10:19,  6.13s/it]
Training Epoch: 15/100, batch 297/500 completed (loss: 0.0322:  60%|█████▉    | 149/250 [15:01<10:18,  6.13s/it]
Training Epoch: 15/100, batch 297/500 completed (loss: 0.0187:  60%|█████▉    | 149/250 [15:01<10:19,  6.14s/it]
Training Epoch: 15/100, batch 298/500 completed (loss: 0.2129:  60%|█████▉    | 149/250 [15:04<10:19,  6.13s/it]
Training Epoch: 15/100, batch 298/500 completed (loss: 0.0874:  60%|█████▉    | 149/250 [15:04<10:18,  6.13s/it]
Training Epoch: 15/100, batch 298/500 completed (loss: 0.0486:  60%|█████▉    | 149/250 [15:04<10:19,  6.13s/it]
Training Epoch: 15/100, batch 298/500 completed (loss: 0.0372:  60%|█████▉    | 149/250 [15:04<10:19,  6.14s/it]
Training Epoch: 15/100, batch 298/500 completed (loss: 0.0372:  60%|██████    | 150/250 [15:07<10:05,  6.05s/it]
Training Epoch: 15/100, batch 298/500 completed (loss: 0.0486:  60%|██████    | 150/250 [15:07<10:05,  6.05s/it]
Training Epoch: 15/100, batch 298/500 completed (loss: 0.0874:  60%|██████    | 150/250 [15:07<10:05,  6.06s/it]
Training Epoch: 15/100, batch 298/500 completed (loss: 0.2129:  60%|██████    | 150/250 [15:07<10:05,  6.06s/it]
Training Epoch: 15/100, batch 299/500 completed (loss: 0.0129:  60%|██████    | 150/250 [15:07<10:05,  6.05s/it]
Training Epoch: 15/100, batch 299/500 completed (loss: 0.2029:  60%|██████    | 150/250 [15:07<10:05,  6.06s/it]
Training Epoch: 15/100, batch 299/500 completed (loss: 0.0868:  60%|██████    | 150/250 [15:07<10:05,  6.06s/it]
Training Epoch: 15/100, batch 299/500 completed (loss: 0.0603:  60%|██████    | 150/250 [15:07<10:05,  6.05s/it]
Training Epoch: 15/100, batch 300/500 completed (loss: 0.0292:  60%|██████    | 150/250 [15:10<10:05,  6.06s/it]
Training Epoch: 15/100, batch 300/500 completed (loss: 0.0242:  60%|██████    | 150/250 [15:10<10:05,  6.05s/it]
Training Epoch: 15/100, batch 300/500 completed (loss: 0.1415:  60%|██████    | 150/250 [15:10<10:05,  6.06s/it]
Training Epoch: 15/100, batch 300/500 completed (loss: 0.0221:  60%|██████    | 150/250 [15:10<10:05,  6.05s/it]
Training Epoch: 15/100, batch 300/500 completed (loss: 0.0221:  60%|██████    | 151/250 [15:13<10:02,  6.08s/it]
Training Epoch: 15/100, batch 300/500 completed (loss: 0.0242:  60%|██████    | 151/250 [15:13<10:02,  6.08s/it]
Training Epoch: 15/100, batch 300/500 completed (loss: 0.1415:  60%|██████    | 151/250 [15:13<10:02,  6.08s/it]
Training Epoch: 15/100, batch 300/500 completed (loss: 0.0292:  60%|██████    | 151/250 [15:13<10:02,  6.08s/it]
Training Epoch: 15/100, batch 301/500 completed (loss: 0.0145:  60%|██████    | 151/250 [15:13<10:02,  6.08s/it]
Training Epoch: 15/100, batch 301/500 completed (loss: 0.0184:  60%|██████    | 151/250 [15:13<10:02,  6.08s/it]
Training Epoch: 15/100, batch 301/500 completed (loss: 0.0085:  60%|██████    | 151/250 [15:13<10:02,  6.08s/it]
Training Epoch: 15/100, batch 301/500 completed (loss: 0.0234:  60%|██████    | 151/250 [15:13<10:02,  6.08s/it]
Training Epoch: 15/100, batch 302/500 completed (loss: 0.025:  60%|██████    | 151/250 [15:16<10:02,  6.08s/it] 
Training Epoch: 15/100, batch 302/500 completed (loss: 0.1524:  60%|██████    | 151/250 [15:16<10:02,  6.08s/it]
Training Epoch: 15/100, batch 302/500 completed (loss: 0.0465:  60%|██████    | 151/250 [15:16<10:02,  6.08s/it]
Training Epoch: 15/100, batch 302/500 completed (loss: 0.0854:  60%|██████    | 151/250 [15:16<10:02,  6.08s/it]
Training Epoch: 15/100, batch 302/500 completed (loss: 0.0854:  61%|██████    | 152/250 [15:19<09:59,  6.12s/it]
Training Epoch: 15/100, batch 302/500 completed (loss: 0.1524:  61%|██████    | 152/250 [15:19<09:59,  6.12s/it]
Training Epoch: 15/100, batch 302/500 completed (loss: 0.0465:  61%|██████    | 152/250 [15:19<10:00,  6.12s/it]
Training Epoch: 15/100, batch 302/500 completed (loss: 0.025:  61%|██████    | 152/250 [15:19<10:00,  6.13s/it]
Training Epoch: 15/100, batch 303/500 completed (loss: 0.0797:  61%|██████    | 152/250 [15:19<10:00,  6.13s/it]
Training Epoch: 15/100, batch 303/500 completed (loss: 0.1312:  61%|██████    | 152/250 [15:19<10:00,  6.12s/it]
Training Epoch: 15/100, batch 303/500 completed (loss: 0.0255:  61%|██████    | 152/250 [15:19<09:59,  6.12s/it]
Training Epoch: 15/100, batch 303/500 completed (loss: 0.0112:  61%|██████    | 152/250 [15:20<09:59,  6.12s/it]
Training Epoch: 15/100, batch 304/500 completed (loss: 0.0334:  61%|██████    | 152/250 [15:22<09:59,  6.12s/it]
Training Epoch: 15/100, batch 304/500 completed (loss: 0.0188:  61%|██████    | 152/250 [15:22<10:00,  6.12s/it]
Training Epoch: 15/100, batch 304/500 completed (loss: 0.0158:  61%|██████    | 152/250 [15:22<10:00,  6.13s/it]
Training Epoch: 15/100, batch 304/500 completed (loss: 0.118:  61%|██████    | 152/250 [15:22<09:59,  6.12s/it] 
Training Epoch: 15/100, batch 304/500 completed (loss: 0.0188:  61%|██████    | 153/250 [15:25<09:47,  6.05s/it]
Training Epoch: 15/100, batch 304/500 completed (loss: 0.0158:  61%|██████    | 153/250 [15:25<09:46,  6.05s/it]
Training Epoch: 15/100, batch 304/500 completed (loss: 0.0334:  61%|██████    | 153/250 [15:25<09:47,  6.05s/it]
Training Epoch: 15/100, batch 304/500 completed (loss: 0.118:  61%|██████    | 153/250 [15:25<09:47,  6.06s/it]
Training Epoch: 15/100, batch 305/500 completed (loss: 0.0175:  61%|██████    | 153/250 [15:25<09:47,  6.05s/it]
Training Epoch: 15/100, batch 305/500 completed (loss: 0.025:  61%|██████    | 153/250 [15:25<09:47,  6.05s/it] 
Training Epoch: 15/100, batch 305/500 completed (loss: 0.015:  61%|██████    | 153/250 [15:25<09:46,  6.05s/it] 
Training Epoch: 15/100, batch 305/500 completed (loss: 0.4064:  61%|██████    | 153/250 [15:25<09:47,  6.06s/it]
Training Epoch: 15/100, batch 306/500 completed (loss: 0.0211:  61%|██████    | 153/250 [15:29<09:46,  6.05s/it]
Training Epoch: 15/100, batch 306/500 completed (loss: 0.1149:  61%|██████    | 153/250 [15:29<09:47,  6.05s/it]
Training Epoch: 15/100, batch 306/500 completed (loss: 0.0202:  61%|██████    | 153/250 [15:29<09:47,  6.05s/it]
Training Epoch: 15/100, batch 306/500 completed (loss: 0.456:  61%|██████    | 153/250 [15:29<09:47,  6.06s/it] 
Training Epoch: 15/100, batch 306/500 completed (loss: 0.0202:  62%|██████▏   | 154/250 [15:32<09:50,  6.15s/it]
Training Epoch: 15/100, batch 306/500 completed (loss: 0.0211:  62%|██████▏   | 154/250 [15:32<09:50,  6.15s/it]
Training Epoch: 15/100, batch 306/500 completed (loss: 0.1149:  62%|██████▏   | 154/250 [15:32<09:50,  6.15s/it]
Training Epoch: 15/100, batch 306/500 completed (loss: 0.456:  62%|██████▏   | 154/250 [15:32<09:50,  6.15s/it]
Training Epoch: 15/100, batch 307/500 completed (loss: 0.143:  62%|██████▏   | 154/250 [15:32<09:50,  6.15s/it] 
Training Epoch: 15/100, batch 307/500 completed (loss: 0.0354:  62%|██████▏   | 154/250 [15:32<09:50,  6.15s/it]
Training Epoch: 15/100, batch 307/500 completed (loss: 0.0832:  62%|██████▏   | 154/250 [15:32<09:50,  6.15s/it]
Training Epoch: 15/100, batch 307/500 completed (loss: 0.0642:  62%|██████▏   | 154/250 [15:32<09:50,  6.15s/it]
Training Epoch: 15/100, batch 308/500 completed (loss: 0.0388:  62%|██████▏   | 154/250 [15:35<09:50,  6.15s/it]
Training Epoch: 15/100, batch 308/500 completed (loss: 0.487:  62%|██████▏   | 154/250 [15:35<09:50,  6.15s/it]
Training Epoch: 15/100, batch 308/500 completed (loss: 0.0271:  62%|██████▏   | 154/250 [15:35<09:50,  6.15s/it]
Training Epoch: 15/100, batch 308/500 completed (loss: 0.03:  62%|██████▏   | 154/250 [15:35<09:50,  6.15s/it]  
Training Epoch: 15/100, batch 308/500 completed (loss: 0.0271:  62%|██████▏   | 155/250 [15:38<09:35,  6.06s/it]
Training Epoch: 15/100, batch 308/500 completed (loss: 0.0388:  62%|██████▏   | 155/250 [15:38<09:35,  6.06s/it]
Training Epoch: 15/100, batch 308/500 completed (loss: 0.487:  62%|██████▏   | 155/250 [15:38<09:35,  6.06s/it]
Training Epoch: 15/100, batch 308/500 completed (loss: 0.03:  62%|██████▏   | 155/250 [15:38<09:36,  6.06s/it]
Training Epoch: 15/100, batch 309/500 completed (loss: 0.1123:  62%|██████▏   | 155/250 [15:38<09:35,  6.06s/it]
Training Epoch: 15/100, batch 309/500 completed (loss: 0.0514:  62%|██████▏   | 155/250 [15:38<09:35,  6.06s/it]
Training Epoch: 15/100, batch 309/500 completed (loss: 0.0134:  62%|██████▏   | 155/250 [15:38<09:35,  6.06s/it]
Training Epoch: 15/100, batch 309/500 completed (loss: 0.0097:  62%|██████▏   | 155/250 [15:38<09:36,  6.06s/it]
Training Epoch: 15/100, batch 310/500 completed (loss: 0.0366:  62%|██████▏   | 155/250 [15:41<09:35,  6.06s/it]
Training Epoch: 15/100, batch 310/500 completed (loss: 0.0973:  62%|██████▏   | 155/250 [15:41<09:35,  6.06s/it]
Training Epoch: 15/100, batch 310/500 completed (loss: 0.2164:  62%|██████▏   | 155/250 [15:41<09:35,  6.06s/it]
Training Epoch: 15/100, batch 310/500 completed (loss: 0.0612:  62%|██████▏   | 155/250 [15:41<09:36,  6.06s/it]
Training Epoch: 15/100, batch 310/500 completed (loss: 0.0612:  62%|██████▏   | 156/250 [15:44<09:27,  6.04s/it]
Training Epoch: 15/100, batch 310/500 completed (loss: 0.0973:  62%|██████▏   | 156/250 [15:44<09:28,  6.04s/it]
Training Epoch: 15/100, batch 310/500 completed (loss: 0.0366:  62%|██████▏   | 156/250 [15:44<09:28,  6.04s/it]
Training Epoch: 15/100, batch 310/500 completed (loss: 0.2164:  62%|██████▏   | 156/250 [15:44<09:28,  6.04s/it]
Training Epoch: 15/100, batch 311/500 completed (loss: 0.0116:  62%|██████▏   | 156/250 [15:44<09:28,  6.04s/it]
Training Epoch: 15/100, batch 311/500 completed (loss: 0.0136:  62%|██████▏   | 156/250 [15:44<09:28,  6.04s/it]
Training Epoch: 15/100, batch 311/500 completed (loss: 0.4302:  62%|██████▏   | 156/250 [15:44<09:28,  6.04s/it]
Training Epoch: 15/100, batch 311/500 completed (loss: 0.1923:  62%|██████▏   | 156/250 [15:44<09:27,  6.04s/it]
Training Epoch: 15/100, batch 312/500 completed (loss: 0.061:  62%|██████▏   | 156/250 [15:47<09:28,  6.04s/it] 
Training Epoch: 15/100, batch 312/500 completed (loss: 0.0694:  62%|██████▏   | 156/250 [15:47<09:28,  6.04s/it]
Training Epoch: 15/100, batch 312/500 completed (loss: 0.1399:  62%|██████▏   | 156/250 [15:47<09:28,  6.04s/it]
Training Epoch: 15/100, batch 312/500 completed (loss: 0.0999:  62%|██████▏   | 156/250 [15:47<09:27,  6.04s/it]
Training Epoch: 15/100, batch 312/500 completed (loss: 0.1399:  63%|██████▎   | 157/250 [15:50<09:29,  6.12s/it]
Training Epoch: 15/100, batch 312/500 completed (loss: 0.0694:  63%|██████▎   | 157/250 [15:50<09:29,  6.12s/it]
Training Epoch: 15/100, batch 312/500 completed (loss: 0.061:  63%|██████▎   | 157/250 [15:50<09:29,  6.12s/it]
Training Epoch: 15/100, batch 312/500 completed (loss: 0.0999:  63%|██████▎   | 157/250 [15:50<09:29,  6.12s/it]
Training Epoch: 15/100, batch 313/500 completed (loss: 0.0643:  63%|██████▎   | 157/250 [15:50<09:29,  6.12s/it]
Training Epoch: 15/100, batch 313/500 completed (loss: 0.0147:  63%|██████▎   | 157/250 [15:50<09:29,  6.12s/it]
Training Epoch: 15/100, batch 313/500 completed (loss: 0.0431:  63%|██████▎   | 157/250 [15:50<09:29,  6.12s/it]
Training Epoch: 15/100, batch 313/500 completed (loss: 0.0114:  63%|██████▎   | 157/250 [15:50<09:29,  6.12s/it]
Training Epoch: 15/100, batch 314/500 completed (loss: 0.0112:  63%|██████▎   | 157/250 [15:53<09:29,  6.12s/it]
Training Epoch: 15/100, batch 314/500 completed (loss: 0.0682:  63%|██████▎   | 157/250 [15:53<09:29,  6.12s/it]
Training Epoch: 15/100, batch 314/500 completed (loss: 0.2019:  63%|██████▎   | 157/250 [15:53<09:29,  6.12s/it]
Training Epoch: 15/100, batch 314/500 completed (loss: 0.1872:  63%|██████▎   | 157/250 [15:53<09:29,  6.12s/it]
Training Epoch: 15/100, batch 314/500 completed (loss: 0.1872:  63%|██████▎   | 158/250 [15:56<09:15,  6.04s/it]
Training Epoch: 15/100, batch 314/500 completed (loss: 0.2019:  63%|██████▎   | 158/250 [15:56<09:15,  6.04s/it]
Training Epoch: 15/100, batch 314/500 completed (loss: 0.0682:  63%|██████▎   | 158/250 [15:56<09:15,  6.04s/it]
Training Epoch: 15/100, batch 314/500 completed (loss: 0.0112:  63%|██████▎   | 158/250 [15:56<09:15,  6.04s/it]
Training Epoch: 15/100, batch 315/500 completed (loss: 0.2809:  63%|██████▎   | 158/250 [15:56<09:15,  6.04s/it]
Training Epoch: 15/100, batch 315/500 completed (loss: 0.0213:  63%|██████▎   | 158/250 [15:56<09:15,  6.04s/it]
Training Epoch: 15/100, batch 315/500 completed (loss: 0.0359:  63%|██████▎   | 158/250 [15:56<09:15,  6.04s/it]
Training Epoch: 15/100, batch 315/500 completed (loss: 0.0823:  63%|██████▎   | 158/250 [15:56<09:15,  6.04s/it]
Training Epoch: 15/100, batch 316/500 completed (loss: 0.0039:  63%|██████▎   | 158/250 [15:59<09:15,  6.04s/it]
Training Epoch: 15/100, batch 316/500 completed (loss: 0.037:  63%|██████▎   | 158/250 [15:59<09:15,  6.04s/it] 
Training Epoch: 15/100, batch 316/500 completed (loss: 0.0632:  63%|██████▎   | 158/250 [15:59<09:15,  6.04s/it]
Training Epoch: 15/100, batch 316/500 completed (loss: 0.0467:  63%|██████▎   | 158/250 [15:59<09:15,  6.04s/it]
Training Epoch: 15/100, batch 316/500 completed (loss: 0.037:  64%|██████▎   | 159/250 [16:02<09:13,  6.08s/it]
Training Epoch: 15/100, batch 316/500 completed (loss: 0.0632:  64%|██████▎   | 159/250 [16:02<09:13,  6.08s/it]
Training Epoch: 15/100, batch 316/500 completed (loss: 0.0039:  64%|██████▎   | 159/250 [16:02<09:13,  6.08s/it]
Training Epoch: 15/100, batch 316/500 completed (loss: 0.0467:  64%|██████▎   | 159/250 [16:02<09:14,  6.09s/it]
Training Epoch: 15/100, batch 317/500 completed (loss: 0.0113:  64%|██████▎   | 159/250 [16:02<09:13,  6.08s/it]
Training Epoch: 15/100, batch 317/500 completed (loss: 0.0329:  64%|██████▎   | 159/250 [16:02<09:13,  6.08s/it]
Training Epoch: 15/100, batch 317/500 completed (loss: 0.0213:  64%|██████▎   | 159/250 [16:02<09:13,  6.08s/it]
Training Epoch: 15/100, batch 317/500 completed (loss: 0.0067:  64%|██████▎   | 159/250 [16:02<09:14,  6.09s/it]
Training Epoch: 15/100, batch 318/500 completed (loss: 0.0422:  64%|██████▎   | 159/250 [16:05<09:13,  6.08s/it]
Training Epoch: 15/100, batch 318/500 completed (loss: 0.0567:  64%|██████▎   | 159/250 [16:05<09:13,  6.08s/it]
Training Epoch: 15/100, batch 318/500 completed (loss: 0.1216:  64%|██████▎   | 159/250 [16:05<09:13,  6.08s/it]
Training Epoch: 15/100, batch 318/500 completed (loss: 0.007:  64%|██████▎   | 159/250 [16:05<09:14,  6.09s/it] 
Training Epoch: 15/100, batch 318/500 completed (loss: 0.0567:  64%|██████▍   | 160/250 [16:08<09:09,  6.10s/it]
Training Epoch: 15/100, batch 318/500 completed (loss: 0.1216:  64%|██████▍   | 160/250 [16:08<09:09,  6.11s/it]
Training Epoch: 15/100, batch 318/500 completed (loss: 0.0422:  64%|██████▍   | 160/250 [16:08<09:09,  6.11s/it]
Training Epoch: 15/100, batch 318/500 completed (loss: 0.007:  64%|██████▍   | 160/250 [16:08<09:09,  6.11s/it]
Training Epoch: 15/100, batch 319/500 completed (loss: 0.0292:  64%|██████▍   | 160/250 [16:08<09:09,  6.11s/it]
Training Epoch: 15/100, batch 319/500 completed (loss: 0.0283:  64%|██████▍   | 160/250 [16:08<09:09,  6.11s/it]
Training Epoch: 15/100, batch 319/500 completed (loss: 0.174:  64%|██████▍   | 160/250 [16:08<09:09,  6.10s/it] 
Training Epoch: 15/100, batch 319/500 completed (loss: 0.1465:  64%|██████▍   | 160/250 [16:08<09:09,  6.11s/it]
Training Epoch: 15/100, batch 320/500 completed (loss: 0.0158:  64%|██████▍   | 160/250 [16:11<09:09,  6.11s/it]
Training Epoch: 15/100, batch 320/500 completed (loss: 0.047:  64%|██████▍   | 160/250 [16:11<09:09,  6.10s/it]
Training Epoch: 15/100, batch 320/500 completed (loss: 0.0525:  64%|██████▍   | 160/250 [16:11<09:09,  6.11s/it]
Training Epoch: 15/100, batch 320/500 completed (loss: 0.0403:  64%|██████▍   | 160/250 [16:11<09:09,  6.11s/it]
Training Epoch: 15/100, batch 320/500 completed (loss: 0.0403:  64%|██████▍   | 161/250 [16:14<08:55,  6.02s/it]
Training Epoch: 15/100, batch 320/500 completed (loss: 0.0525:  64%|██████▍   | 161/250 [16:14<08:55,  6.02s/it]
Training Epoch: 15/100, batch 320/500 completed (loss: 0.0158:  64%|██████▍   | 161/250 [16:14<08:55,  6.02s/it]
Training Epoch: 15/100, batch 320/500 completed (loss: 0.047:  64%|██████▍   | 161/250 [16:14<08:56,  6.02s/it]
Training Epoch: 15/100, batch 321/500 completed (loss: 0.0551:  64%|██████▍   | 161/250 [16:14<08:56,  6.02s/it]
Training Epoch: 15/100, batch 321/500 completed (loss: 0.0595:  64%|██████▍   | 161/250 [16:14<08:55,  6.02s/it]
Training Epoch: 15/100, batch 321/500 completed (loss: 0.0553:  64%|██████▍   | 161/250 [16:14<08:55,  6.02s/it]
Training Epoch: 15/100, batch 321/500 completed (loss: 0.021:  64%|██████▍   | 161/250 [16:14<08:55,  6.02s/it] 
Training Epoch: 15/100, batch 322/500 completed (loss: 0.0103:  64%|██████▍   | 161/250 [16:17<08:55,  6.02s/it]
Training Epoch: 15/100, batch 322/500 completed (loss: 0.1955:  64%|██████▍   | 161/250 [16:17<08:55,  6.02s/it]
Training Epoch: 15/100, batch 322/500 completed (loss: 0.3853:  64%|██████▍   | 161/250 [16:17<08:56,  6.02s/it]
Training Epoch: 15/100, batch 322/500 completed (loss: 0.0283:  64%|██████▍   | 161/250 [16:17<08:55,  6.02s/it]
Training Epoch: 15/100, batch 322/500 completed (loss: 0.0283:  65%|██████▍   | 162/250 [16:20<09:01,  6.15s/it]
Training Epoch: 15/100, batch 322/500 completed (loss: 0.1955:  65%|██████▍   | 162/250 [16:20<09:01,  6.15s/it]
Training Epoch: 15/100, batch 322/500 completed (loss: 0.0103:  65%|██████▍   | 162/250 [16:20<09:01,  6.15s/it]
Training Epoch: 15/100, batch 322/500 completed (loss: 0.3853:  65%|██████▍   | 162/250 [16:20<09:01,  6.15s/it]
Training Epoch: 15/100, batch 323/500 completed (loss: 0.0442:  65%|██████▍   | 162/250 [16:20<09:01,  6.15s/it]
Training Epoch: 15/100, batch 323/500 completed (loss: 0.01:  65%|██████▍   | 162/250 [16:20<09:01,  6.15s/it]  
Training Epoch: 15/100, batch 323/500 completed (loss: 0.1049:  65%|██████▍   | 162/250 [16:20<09:01,  6.15s/it]
Training Epoch: 15/100, batch 323/500 completed (loss: 0.1227:  65%|██████▍   | 162/250 [16:20<09:01,  6.15s/it]
Training Epoch: 15/100, batch 324/500 completed (loss: 0.0104:  65%|██████▍   | 162/250 [16:23<09:01,  6.15s/it]
Training Epoch: 15/100, batch 324/500 completed (loss: 0.021:  65%|██████▍   | 162/250 [16:23<09:01,  6.15s/it] 
Training Epoch: 15/100, batch 324/500 completed (loss: 0.0359:  65%|██████▍   | 162/250 [16:23<09:01,  6.15s/it]
Training Epoch: 15/100, batch 324/500 completed (loss: 0.01:  65%|██████▍   | 162/250 [16:23<09:01,  6.15s/it]  
Training Epoch: 15/100, batch 324/500 completed (loss: 0.0359:  65%|██████▌   | 163/250 [16:26<08:47,  6.06s/it]
Training Epoch: 15/100, batch 324/500 completed (loss: 0.01:  65%|██████▌   | 163/250 [16:26<08:47,  6.06s/it]
Training Epoch: 15/100, batch 324/500 completed (loss: 0.0104:  65%|██████▌   | 163/250 [16:26<08:47,  6.06s/it]
Training Epoch: 15/100, batch 324/500 completed (loss: 0.021:  65%|██████▌   | 163/250 [16:26<08:47,  6.06s/it]
Training Epoch: 15/100, batch 325/500 completed (loss: 0.1352:  65%|██████▌   | 163/250 [16:26<08:47,  6.06s/it]
Training Epoch: 15/100, batch 325/500 completed (loss: 0.0717:  65%|██████▌   | 163/250 [16:26<08:47,  6.06s/it]
Training Epoch: 15/100, batch 325/500 completed (loss: 0.0596:  65%|██████▌   | 163/250 [16:26<08:47,  6.06s/it]
Training Epoch: 15/100, batch 325/500 completed (loss: 0.1576:  65%|██████▌   | 163/250 [16:26<08:47,  6.06s/it]
Training Epoch: 15/100, batch 326/500 completed (loss: 0.0258:  65%|██████▌   | 163/250 [16:29<08:47,  6.06s/it]
Training Epoch: 15/100, batch 326/500 completed (loss: 0.007:  65%|██████▌   | 163/250 [16:29<08:47,  6.06s/it] 
Training Epoch: 15/100, batch 326/500 completed (loss: 0.4892:  65%|██████▌   | 163/250 [16:29<08:47,  6.06s/it]
Training Epoch: 15/100, batch 326/500 completed (loss: 0.0412:  65%|██████▌   | 163/250 [16:29<08:47,  6.06s/it]
Training Epoch: 15/100, batch 326/500 completed (loss: 0.0412:  66%|██████▌   | 164/250 [16:32<08:43,  6.09s/it]
Training Epoch: 15/100, batch 326/500 completed (loss: 0.0258:  66%|██████▌   | 164/250 [16:32<08:43,  6.09s/it]
Training Epoch: 15/100, batch 326/500 completed (loss: 0.007:  66%|██████▌   | 164/250 [16:32<08:43,  6.09s/it]
Training Epoch: 15/100, batch 326/500 completed (loss: 0.4892:  66%|██████▌   | 164/250 [16:32<08:43,  6.09s/it]
Training Epoch: 15/100, batch 327/500 completed (loss: 0.0957:  66%|██████▌   | 164/250 [16:32<08:43,  6.09s/it]
Training Epoch: 15/100, batch 327/500 completed (loss: 0.0272:  66%|██████▌   | 164/250 [16:32<08:43,  6.09s/it]
Training Epoch: 15/100, batch 327/500 completed (loss: 0.0425:  66%|██████▌   | 164/250 [16:32<08:43,  6.09s/it]
Training Epoch: 15/100, batch 327/500 completed (loss: 0.603:  66%|██████▌   | 164/250 [16:32<08:43,  6.09s/it] 
Training Epoch: 15/100, batch 328/500 completed (loss: 0.3163:  66%|██████▌   | 164/250 [16:35<08:43,  6.09s/it]
Training Epoch: 15/100, batch 328/500 completed (loss: 0.0146:  66%|██████▌   | 164/250 [16:35<08:43,  6.09s/it]
Training Epoch: 15/100, batch 328/500 completed (loss: 0.0112:  66%|██████▌   | 164/250 [16:35<08:43,  6.09s/it]
Training Epoch: 15/100, batch 328/500 completed (loss: 0.0324:  66%|██████▌   | 164/250 [16:36<08:43,  6.09s/it]
Training Epoch: 15/100, batch 328/500 completed (loss: 0.3163:  66%|██████▌   | 165/250 [16:38<08:38,  6.10s/it]
Training Epoch: 15/100, batch 328/500 completed (loss: 0.0324:  66%|██████▌   | 165/250 [16:38<08:38,  6.10s/it]
Training Epoch: 15/100, batch 328/500 completed (loss: 0.0112:  66%|██████▌   | 165/250 [16:38<08:38,  6.10s/it]
Training Epoch: 15/100, batch 328/500 completed (loss: 0.0146:  66%|██████▌   | 165/250 [16:38<08:38,  6.10s/it]
Training Epoch: 15/100, batch 329/500 completed (loss: 0.1753:  66%|██████▌   | 165/250 [16:38<08:38,  6.10s/it]
Training Epoch: 15/100, batch 329/500 completed (loss: 0.1759:  66%|██████▌   | 165/250 [16:39<08:38,  6.10s/it]
Training Epoch: 15/100, batch 329/500 completed (loss: 0.119:  66%|██████▌   | 165/250 [16:39<08:38,  6.10s/it] 
Training Epoch: 15/100, batch 329/500 completed (loss: 0.2159:  66%|██████▌   | 165/250 [16:39<08:38,  6.10s/it]
Training Epoch: 15/100, batch 330/500 completed (loss: 0.1657:  66%|██████▌   | 165/250 [16:41<08:38,  6.10s/it]
Training Epoch: 15/100, batch 330/500 completed (loss: 0.2539:  66%|██████▌   | 165/250 [16:41<08:38,  6.10s/it]
Training Epoch: 15/100, batch 330/500 completed (loss: 0.1409:  66%|██████▌   | 165/250 [16:41<08:38,  6.10s/it]
Training Epoch: 15/100, batch 330/500 completed (loss: 0.0401:  66%|██████▌   | 165/250 [16:41<08:38,  6.10s/it]
Training Epoch: 15/100, batch 330/500 completed (loss: 0.2539:  66%|██████▋   | 166/250 [16:44<08:25,  6.02s/it]
Training Epoch: 15/100, batch 330/500 completed (loss: 0.1657:  66%|██████▋   | 166/250 [16:44<08:25,  6.02s/it]
Training Epoch: 15/100, batch 330/500 completed (loss: 0.1409:  66%|██████▋   | 166/250 [16:44<08:25,  6.02s/it]
Training Epoch: 15/100, batch 330/500 completed (loss: 0.0401:  66%|██████▋   | 166/250 [16:44<08:25,  6.02s/it]
Training Epoch: 15/100, batch 331/500 completed (loss: 0.0407:  66%|██████▋   | 166/250 [16:44<08:25,  6.02s/it]
Training Epoch: 15/100, batch 331/500 completed (loss: 0.159:  66%|██████▋   | 166/250 [16:44<08:25,  6.02s/it] 
Training Epoch: 15/100, batch 331/500 completed (loss: 0.206:  66%|██████▋   | 166/250 [16:44<08:25,  6.02s/it] 
Training Epoch: 15/100, batch 331/500 completed (loss: 0.0259:  66%|██████▋   | 166/250 [16:44<08:25,  6.02s/it]
Training Epoch: 15/100, batch 332/500 completed (loss: 0.2567:  66%|██████▋   | 166/250 [16:47<08:25,  6.02s/it]
Training Epoch: 15/100, batch 332/500 completed (loss: 0.038:  66%|██████▋   | 166/250 [16:47<08:25,  6.02s/it] 
Training Epoch: 15/100, batch 332/500 completed (loss: 0.1315:  66%|██████▋   | 166/250 [16:47<08:25,  6.02s/it]
Training Epoch: 15/100, batch 332/500 completed (loss: 0.0215:  66%|██████▋   | 166/250 [16:47<08:25,  6.02s/it]
Training Epoch: 15/100, batch 332/500 completed (loss: 0.2567:  67%|██████▋   | 167/250 [16:50<08:22,  6.06s/it]
Training Epoch: 15/100, batch 332/500 completed (loss: 0.038:  67%|██████▋   | 167/250 [16:50<08:22,  6.06s/it]
Training Epoch: 15/100, batch 332/500 completed (loss: 0.1315:  67%|██████▋   | 167/250 [16:50<08:23,  6.06s/it]
Training Epoch: 15/100, batch 332/500 completed (loss: 0.0215:  67%|██████▋   | 167/250 [16:50<08:23,  6.06s/it]
Training Epoch: 15/100, batch 333/500 completed (loss: 0.2135:  67%|██████▋   | 167/250 [16:50<08:23,  6.06s/it]
Training Epoch: 15/100, batch 333/500 completed (loss: 0.014:  67%|██████▋   | 167/250 [16:51<08:22,  6.06s/it] 
Training Epoch: 15/100, batch 333/500 completed (loss: 0.0279:  67%|██████▋   | 167/250 [16:51<08:22,  6.06s/it]
Training Epoch: 15/100, batch 333/500 completed (loss: 0.3865:  67%|██████▋   | 167/250 [16:51<08:23,  6.06s/it]
Training Epoch: 15/100, batch 334/500 completed (loss: 0.0504:  67%|██████▋   | 167/250 [16:54<08:23,  6.06s/it]
Training Epoch: 15/100, batch 334/500 completed (loss: 0.1667:  67%|██████▋   | 167/250 [16:54<08:22,  6.06s/it]
Training Epoch: 15/100, batch 334/500 completed (loss: 0.043:  67%|██████▋   | 167/250 [16:54<08:22,  6.06s/it]
Training Epoch: 15/100, batch 334/500 completed (loss: 0.0105:  67%|██████▋   | 167/250 [16:54<08:23,  6.06s/it]
Training Epoch: 15/100, batch 334/500 completed (loss: 0.1667:  67%|██████▋   | 168/250 [16:57<08:19,  6.10s/it]
Training Epoch: 15/100, batch 334/500 completed (loss: 0.0504:  67%|██████▋   | 168/250 [16:57<08:19,  6.10s/it]
Training Epoch: 15/100, batch 334/500 completed (loss: 0.043:  67%|██████▋   | 168/250 [16:57<08:19,  6.10s/it]
Training Epoch: 15/100, batch 334/500 completed (loss: 0.0105:  67%|██████▋   | 168/250 [16:57<08:19,  6.10s/it]
Training Epoch: 15/100, batch 335/500 completed (loss: 0.0201:  67%|██████▋   | 168/250 [16:57<08:19,  6.10s/it]
Training Epoch: 15/100, batch 335/500 completed (loss: 0.0371:  67%|██████▋   | 168/250 [16:57<08:19,  6.10s/it]
Training Epoch: 15/100, batch 335/500 completed (loss: 0.078:  67%|██████▋   | 168/250 [16:57<08:19,  6.10s/it]
Training Epoch: 15/100, batch 335/500 completed (loss: 0.2834:  67%|██████▋   | 168/250 [16:57<08:19,  6.10s/it]
Training Epoch: 15/100, batch 336/500 completed (loss: 0.017:  67%|██████▋   | 168/250 [17:00<08:19,  6.10s/it] 
Training Epoch: 15/100, batch 336/500 completed (loss: 0.0456:  67%|██████▋   | 168/250 [17:00<08:19,  6.10s/it]
Training Epoch: 15/100, batch 336/500 completed (loss: 0.02:  67%|██████▋   | 168/250 [17:00<08:19,  6.10s/it]  
Training Epoch: 15/100, batch 336/500 completed (loss: 0.0106:  67%|██████▋   | 168/250 [17:00<08:19,  6.10s/it]
Training Epoch: 15/100, batch 336/500 completed (loss: 0.0106:  68%|██████▊   | 169/250 [17:03<08:07,  6.02s/it]
Training Epoch: 15/100, batch 336/500 completed (loss: 0.0456:  68%|██████▊   | 169/250 [17:02<08:07,  6.02s/it]
Training Epoch: 15/100, batch 336/500 completed (loss: 0.02:  68%|██████▊   | 169/250 [17:03<08:08,  6.03s/it]
Training Epoch: 15/100, batch 336/500 completed (loss: 0.017:  68%|██████▊   | 169/250 [17:02<08:08,  6.03s/it]
Training Epoch: 15/100, batch 337/500 completed (loss: 0.0486:  68%|██████▊   | 169/250 [17:03<08:08,  6.03s/it]
Training Epoch: 15/100, batch 337/500 completed (loss: 0.1202:  68%|██████▊   | 169/250 [17:03<08:07,  6.02s/it]
Training Epoch: 15/100, batch 337/500 completed (loss: 0.1364:  68%|██████▊   | 169/250 [17:03<08:08,  6.03s/it]
Training Epoch: 15/100, batch 337/500 completed (loss: 0.0343:  68%|██████▊   | 169/250 [17:03<08:07,  6.02s/it]
Training Epoch: 15/100, batch 338/500 completed (loss: 0.2045:  68%|██████▊   | 169/250 [17:06<08:08,  6.03s/it]
Training Epoch: 15/100, batch 338/500 completed (loss: 0.1991:  68%|██████▊   | 169/250 [17:06<08:08,  6.03s/it]
Training Epoch: 15/100, batch 338/500 completed (loss: 0.2783:  68%|██████▊   | 169/250 [17:06<08:07,  6.02s/it]
Training Epoch: 15/100, batch 338/500 completed (loss: 0.0611:  68%|██████▊   | 169/250 [17:06<08:07,  6.02s/it]
Training Epoch: 15/100, batch 338/500 completed (loss: 0.0611:  68%|██████▊   | 170/250 [17:09<08:12,  6.16s/it]
Training Epoch: 15/100, batch 338/500 completed (loss: 0.2783:  68%|██████▊   | 170/250 [17:09<08:12,  6.16s/it]
Training Epoch: 15/100, batch 338/500 completed (loss: 0.2045:  68%|██████▊   | 170/250 [17:09<08:12,  6.16s/it]
Training Epoch: 15/100, batch 338/500 completed (loss: 0.1991:  68%|██████▊   | 170/250 [17:09<08:12,  6.16s/it]
Training Epoch: 15/100, batch 339/500 completed (loss: 0.1195:  68%|██████▊   | 170/250 [17:09<08:12,  6.16s/it]
Training Epoch: 15/100, batch 339/500 completed (loss: 0.0148:  68%|██████▊   | 170/250 [17:09<08:12,  6.16s/it]
Training Epoch: 15/100, batch 339/500 completed (loss: 0.0926:  68%|██████▊   | 170/250 [17:09<08:12,  6.16s/it]
Training Epoch: 15/100, batch 339/500 completed (loss: 0.0021:  68%|██████▊   | 170/250 [17:09<08:12,  6.16s/it]
Training Epoch: 15/100, batch 340/500 completed (loss: 0.0216:  68%|██████▊   | 170/250 [17:12<08:12,  6.16s/it]
Training Epoch: 15/100, batch 340/500 completed (loss: 0.0045:  68%|██████▊   | 170/250 [17:12<08:12,  6.16s/it]
Training Epoch: 15/100, batch 340/500 completed (loss: 0.0632:  68%|██████▊   | 170/250 [17:12<08:12,  6.16s/it]
Training Epoch: 15/100, batch 340/500 completed (loss: 0.0078:  68%|██████▊   | 170/250 [17:12<08:12,  6.16s/it]
Training Epoch: 15/100, batch 340/500 completed (loss: 0.0045:  68%|██████▊   | 171/250 [17:15<07:59,  6.07s/it]
Training Epoch: 15/100, batch 340/500 completed (loss: 0.0216:  68%|██████▊   | 171/250 [17:15<07:59,  6.07s/it]
Training Epoch: 15/100, batch 340/500 completed (loss: 0.0632:  68%|██████▊   | 171/250 [17:15<07:59,  6.07s/it]
Training Epoch: 15/100, batch 340/500 completed (loss: 0.0078:  68%|██████▊   | 171/250 [17:15<08:00,  6.08s/it]
Training Epoch: 15/100, batch 341/500 completed (loss: 0.008:  68%|██████▊   | 171/250 [17:15<07:59,  6.07s/it] 
Training Epoch: 15/100, batch 341/500 completed (loss: 0.0769:  68%|██████▊   | 171/250 [17:15<07:59,  6.07s/it]
Training Epoch: 15/100, batch 341/500 completed (loss: 0.0088:  68%|██████▊   | 171/250 [17:15<07:59,  6.07s/it]
Training Epoch: 15/100, batch 341/500 completed (loss: 0.163:  68%|██████▊   | 171/250 [17:15<08:00,  6.08s/it] 
Training Epoch: 15/100, batch 342/500 completed (loss: 0.0978:  68%|██████▊   | 171/250 [17:18<07:59,  6.07s/it]
Training Epoch: 15/100, batch 342/500 completed (loss: 0.0417:  68%|██████▊   | 171/250 [17:18<07:59,  6.07s/it]
Training Epoch: 15/100, batch 342/500 completed (loss: 0.2692:  68%|██████▊   | 171/250 [17:18<07:59,  6.07s/it]
Training Epoch: 15/100, batch 342/500 completed (loss: 0.0852:  68%|██████▊   | 171/250 [17:18<08:00,  6.08s/it]
Training Epoch: 15/100, batch 342/500 completed (loss: 0.2692:  69%|██████▉   | 172/250 [17:21<07:55,  6.10s/it]
Training Epoch: 15/100, batch 342/500 completed (loss: 0.0417:  69%|██████▉   | 172/250 [17:21<07:56,  6.11s/it]
Training Epoch: 15/100, batch 342/500 completed (loss: 0.0978:  69%|██████▉   | 172/250 [17:21<07:56,  6.10s/it]
Training Epoch: 15/100, batch 342/500 completed (loss: 0.0852:  69%|██████▉   | 172/250 [17:21<07:56,  6.11s/it]
Training Epoch: 15/100, batch 343/500 completed (loss: 0.2253:  69%|██████▉   | 172/250 [17:21<07:56,  6.10s/it]
Training Epoch: 15/100, batch 343/500 completed (loss: 0.2774:  69%|██████▉   | 172/250 [17:21<07:56,  6.11s/it]
Training Epoch: 15/100, batch 343/500 completed (loss: 0.1677:  69%|██████▉   | 172/250 [17:21<07:55,  6.10s/it]
Training Epoch: 15/100, batch 343/500 completed (loss: 0.055:  69%|██████▉   | 172/250 [17:21<07:56,  6.11s/it] 
Training Epoch: 15/100, batch 344/500 completed (loss: 0.1053:  69%|██████▉   | 172/250 [17:24<07:56,  6.11s/it]
Training Epoch: 15/100, batch 344/500 completed (loss: 0.0203:  69%|██████▉   | 172/250 [17:24<07:56,  6.10s/it]
Training Epoch: 15/100, batch 344/500 completed (loss: 0.1098:  69%|██████▉   | 172/250 [17:24<07:55,  6.10s/it]
Training Epoch: 15/100, batch 344/500 completed (loss: 0.098:  69%|██████▉   | 172/250 [17:24<07:56,  6.11s/it]
Training Epoch: 15/100, batch 344/500 completed (loss: 0.1053:  69%|██████▉   | 173/250 [17:27<07:51,  6.13s/it]
Training Epoch: 15/100, batch 344/500 completed (loss: 0.1098:  69%|██████▉   | 173/250 [17:27<07:51,  6.13s/it]
Training Epoch: 15/100, batch 344/500 completed (loss: 0.0203:  69%|██████▉   | 173/250 [17:27<07:51,  6.13s/it]
Training Epoch: 15/100, batch 344/500 completed (loss: 0.098:  69%|██████▉   | 173/250 [17:27<07:51,  6.13s/it]
Training Epoch: 15/100, batch 345/500 completed (loss: 0.2386:  69%|██████▉   | 173/250 [17:27<07:51,  6.13s/it]
Training Epoch: 15/100, batch 345/500 completed (loss: 0.0773:  69%|██████▉   | 173/250 [17:27<07:51,  6.13s/it]
Training Epoch: 15/100, batch 345/500 completed (loss: 0.2286:  69%|██████▉   | 173/250 [17:27<07:51,  6.13s/it]
Training Epoch: 15/100, batch 345/500 completed (loss: 0.2003:  69%|██████▉   | 173/250 [17:27<07:51,  6.13s/it]
Training Epoch: 15/100, batch 346/500 completed (loss: 0.0979:  69%|██████▉   | 173/250 [17:30<07:51,  6.13s/it]
Training Epoch: 15/100, batch 346/500 completed (loss: 0.1767:  69%|██████▉   | 173/250 [17:30<07:51,  6.13s/it]
Training Epoch: 15/100, batch 346/500 completed (loss: 0.0723:  69%|██████▉   | 173/250 [17:30<07:51,  6.13s/it]
Training Epoch: 15/100, batch 346/500 completed (loss: 0.0221:  69%|██████▉   | 173/250 [17:30<07:51,  6.13s/it]
Training Epoch: 15/100, batch 346/500 completed (loss: 0.0723:  70%|██████▉   | 174/250 [17:33<07:39,  6.04s/it]
Training Epoch: 15/100, batch 346/500 completed (loss: 0.0221:  70%|██████▉   | 174/250 [17:33<07:39,  6.04s/it]
Training Epoch: 15/100, batch 346/500 completed (loss: 0.1767:  70%|██████▉   | 174/250 [17:33<07:39,  6.04s/it]
Training Epoch: 15/100, batch 346/500 completed (loss: 0.0979:  70%|██████▉   | 174/250 [17:33<07:39,  6.05s/it]
Training Epoch: 15/100, batch 347/500 completed (loss: 0.0039:  70%|██████▉   | 174/250 [17:33<07:39,  6.04s/it]
Training Epoch: 15/100, batch 347/500 completed (loss: 0.0056:  70%|██████▉   | 174/250 [17:33<07:39,  6.04s/it]
Training Epoch: 15/100, batch 347/500 completed (loss: 0.0097:  70%|██████▉   | 174/250 [17:33<07:39,  6.05s/it]
Training Epoch: 15/100, batch 347/500 completed (loss: 0.1917:  70%|██████▉   | 174/250 [17:33<07:39,  6.04s/it]
Training Epoch: 15/100, batch 348/500 completed (loss: 0.0146:  70%|██████▉   | 174/250 [17:36<07:39,  6.05s/it]
Training Epoch: 15/100, batch 348/500 completed (loss: 0.0654:  70%|██████▉   | 174/250 [17:36<07:39,  6.04s/it]
Training Epoch: 15/100, batch 348/500 completed (loss: 0.2877:  70%|██████▉   | 174/250 [17:36<07:39,  6.04s/it]
Training Epoch: 15/100, batch 348/500 completed (loss: 0.0052:  70%|██████▉   | 174/250 [17:36<07:39,  6.04s/it]
Training Epoch: 15/100, batch 348/500 completed (loss: 0.0052:  70%|███████   | 175/250 [17:39<07:38,  6.12s/it]
Training Epoch: 15/100, batch 348/500 completed (loss: 0.0654:  70%|███████   | 175/250 [17:39<07:39,  6.12s/it]
Training Epoch: 15/100, batch 348/500 completed (loss: 0.0146:  70%|███████   | 175/250 [17:39<07:39,  6.12s/it]
Training Epoch: 15/100, batch 348/500 completed (loss: 0.2877:  70%|███████   | 175/250 [17:39<07:39,  6.12s/it]
Training Epoch: 15/100, batch 349/500 completed (loss: 0.0801:  70%|███████   | 175/250 [17:39<07:39,  6.12s/it]
Training Epoch: 15/100, batch 349/500 completed (loss: 0.1648:  70%|███████   | 175/250 [17:39<07:39,  6.12s/it]
Training Epoch: 15/100, batch 349/500 completed (loss: 0.2947:  70%|███████   | 175/250 [17:39<07:39,  6.12s/it]
Training Epoch: 15/100, batch 349/500 completed (loss: 0.1234:  70%|███████   | 175/250 [17:39<07:38,  6.12s/it]
Training Epoch: 15/100, batch 350/500 completed (loss: 0.0562:  70%|███████   | 175/250 [17:42<07:39,  6.12s/it]
Training Epoch: 15/100, batch 350/500 completed (loss: 0.0581:  70%|███████   | 175/250 [17:42<07:39,  6.12s/it]
Training Epoch: 15/100, batch 350/500 completed (loss: 0.0511:  70%|███████   | 175/250 [17:42<07:39,  6.12s/it]
Training Epoch: 15/100, batch 350/500 completed (loss: 0.1676:  70%|███████   | 175/250 [17:42<07:38,  6.12s/it]
Training Epoch: 15/100, batch 350/500 completed (loss: 0.0511:  70%|███████   | 176/250 [17:45<07:30,  6.09s/it]
Training Epoch: 15/100, batch 350/500 completed (loss: 0.0562:  70%|███████   | 176/250 [17:45<07:30,  6.09s/it]
Training Epoch: 15/100, batch 350/500 completed (loss: 0.0581:  70%|███████   | 176/250 [17:45<07:30,  6.09s/it]
Training Epoch: 15/100, batch 350/500 completed (loss: 0.1676:  70%|███████   | 176/250 [17:45<07:30,  6.09s/it]
Training Epoch: 15/100, batch 351/500 completed (loss: 0.1063:  70%|███████   | 176/250 [17:45<07:30,  6.09s/it]
Training Epoch: 15/100, batch 351/500 completed (loss: 0.1392:  70%|███████   | 176/250 [17:45<07:30,  6.09s/it]
Training Epoch: 15/100, batch 351/500 completed (loss: 0.0708:  70%|███████   | 176/250 [17:45<07:30,  6.09s/it]
Training Epoch: 15/100, batch 351/500 completed (loss: 0.0884:  70%|███████   | 176/250 [17:45<07:30,  6.09s/it]
Training Epoch: 15/100, batch 352/500 completed (loss: 0.0422:  70%|███████   | 176/250 [17:48<07:30,  6.09s/it]
Training Epoch: 15/100, batch 352/500 completed (loss: 0.0044:  70%|███████   | 176/250 [17:48<07:30,  6.09s/it]
Training Epoch: 15/100, batch 352/500 completed (loss: 0.1483:  70%|███████   | 176/250 [17:48<07:30,  6.09s/it]
Training Epoch: 15/100, batch 352/500 completed (loss: 0.0997:  70%|███████   | 176/250 [17:48<07:30,  6.09s/it]
Training Epoch: 15/100, batch 352/500 completed (loss: 0.1483:  71%|███████   | 177/250 [17:51<07:23,  6.07s/it]
Training Epoch: 15/100, batch 352/500 completed (loss: 0.0044:  71%|███████   | 177/250 [17:51<07:23,  6.07s/it]
Training Epoch: 15/100, batch 352/500 completed (loss: 0.0422:  71%|███████   | 177/250 [17:51<07:23,  6.07s/it]
Training Epoch: 15/100, batch 352/500 completed (loss: 0.0997:  71%|███████   | 177/250 [17:51<07:23,  6.07s/it]
Training Epoch: 15/100, batch 353/500 completed (loss: 0.0995:  71%|███████   | 177/250 [17:51<07:23,  6.07s/it]
Training Epoch: 15/100, batch 353/500 completed (loss: 0.0599:  71%|███████   | 177/250 [17:51<07:23,  6.07s/it]
Training Epoch: 15/100, batch 353/500 completed (loss: 0.0741:  71%|███████   | 177/250 [17:51<07:23,  6.07s/it]
Training Epoch: 15/100, batch 353/500 completed (loss: 0.1099:  71%|███████   | 177/250 [17:51<07:23,  6.07s/it]
Training Epoch: 15/100, batch 354/500 completed (loss: 0.0109:  71%|███████   | 177/250 [17:55<07:23,  6.07s/it]
Training Epoch: 15/100, batch 354/500 completed (loss: 0.1614:  71%|███████   | 177/250 [17:55<07:23,  6.07s/it]
Training Epoch: 15/100, batch 354/500 completed (loss: 0.3363:  71%|███████   | 177/250 [17:55<07:23,  6.07s/it]
Training Epoch: 15/100, batch 354/500 completed (loss: 0.1828:  71%|███████   | 177/250 [17:55<07:23,  6.07s/it]
Training Epoch: 15/100, batch 354/500 completed (loss: 0.1828:  71%|███████   | 178/250 [17:58<07:18,  6.09s/it]
Training Epoch: 15/100, batch 354/500 completed (loss: 0.1614:  71%|███████   | 178/250 [17:58<07:18,  6.10s/it]
Training Epoch: 15/100, batch 354/500 completed (loss: 0.3363:  71%|███████   | 178/250 [17:58<07:19,  6.10s/it]
Training Epoch: 15/100, batch 354/500 completed (loss: 0.0109:  71%|███████   | 178/250 [17:58<07:19,  6.10s/it]
Training Epoch: 15/100, batch 355/500 completed (loss: 0.0212:  71%|███████   | 178/250 [17:58<07:18,  6.10s/it]
Training Epoch: 15/100, batch 355/500 completed (loss: 0.0092:  71%|███████   | 178/250 [17:58<07:19,  6.10s/it]
Training Epoch: 15/100, batch 355/500 completed (loss: 0.1789:  71%|███████   | 178/250 [17:58<07:19,  6.10s/it]
Training Epoch: 15/100, batch 355/500 completed (loss: 0.0998:  71%|███████   | 178/250 [17:58<07:18,  6.09s/it]
Training Epoch: 15/100, batch 356/500 completed (loss: 0.0137:  71%|███████   | 178/250 [18:01<07:18,  6.10s/it]
Training Epoch: 15/100, batch 356/500 completed (loss: 0.1729:  71%|███████   | 178/250 [18:01<07:19,  6.10s/it]
Training Epoch: 15/100, batch 356/500 completed (loss: 0.3285:  71%|███████   | 178/250 [18:01<07:19,  6.10s/it]
Training Epoch: 15/100, batch 356/500 completed (loss: 0.3597:  71%|███████   | 178/250 [18:01<07:18,  6.09s/it]
Training Epoch: 15/100, batch 356/500 completed (loss: 0.3597:  72%|███████▏  | 179/250 [18:04<07:11,  6.07s/it]
Training Epoch: 15/100, batch 356/500 completed (loss: 0.3285:  72%|███████▏  | 179/250 [18:04<07:11,  6.07s/it]
Training Epoch: 15/100, batch 356/500 completed (loss: 0.1729:  72%|███████▏  | 179/250 [18:04<07:11,  6.07s/it]
Training Epoch: 15/100, batch 356/500 completed (loss: 0.0137:  72%|███████▏  | 179/250 [18:04<07:11,  6.08s/it]
Training Epoch: 15/100, batch 357/500 completed (loss: 0.0426:  72%|███████▏  | 179/250 [18:04<07:11,  6.08s/it]
Training Epoch: 15/100, batch 357/500 completed (loss: 0.0198:  72%|███████▏  | 179/250 [18:04<07:11,  6.07s/it]
Training Epoch: 15/100, batch 357/500 completed (loss: 0.0602:  72%|███████▏  | 179/250 [18:04<07:11,  6.07s/it]
Training Epoch: 15/100, batch 357/500 completed (loss: 0.0785:  72%|███████▏  | 179/250 [18:04<07:11,  6.07s/it]
Training Epoch: 15/100, batch 358/500 completed (loss: 0.0192:  72%|███████▏  | 179/250 [18:07<07:11,  6.07s/it]
Training Epoch: 15/100, batch 358/500 completed (loss: 0.2119:  72%|███████▏  | 179/250 [18:07<07:11,  6.07s/it]
Training Epoch: 15/100, batch 358/500 completed (loss: 0.0063:  72%|███████▏  | 179/250 [18:07<07:11,  6.08s/it]
Training Epoch: 15/100, batch 358/500 completed (loss: 0.0043:  72%|███████▏  | 179/250 [18:07<07:11,  6.07s/it]
Training Epoch: 15/100, batch 358/500 completed (loss: 0.0192:  72%|███████▏  | 180/250 [18:10<07:10,  6.15s/it]
Training Epoch: 15/100, batch 358/500 completed (loss: 0.2119:  72%|███████▏  | 180/250 [18:10<07:10,  6.15s/it]
Training Epoch: 15/100, batch 358/500 completed (loss: 0.0063:  72%|███████▏  | 180/250 [18:10<07:10,  6.15s/it]
Training Epoch: 15/100, batch 358/500 completed (loss: 0.0043:  72%|███████▏  | 180/250 [18:10<07:10,  6.15s/it]
Training Epoch: 15/100, batch 359/500 completed (loss: 0.0153:  72%|███████▏  | 180/250 [18:10<07:10,  6.15s/it]
Training Epoch: 15/100, batch 359/500 completed (loss: 0.1661:  72%|███████▏  | 180/250 [18:10<07:10,  6.15s/it]
Training Epoch: 15/100, batch 359/500 completed (loss: 0.0397:  72%|███████▏  | 180/250 [18:10<07:10,  6.15s/it]
Training Epoch: 15/100, batch 359/500 completed (loss: 0.0293:  72%|███████▏  | 180/250 [18:10<07:10,  6.15s/it]
Training Epoch: 15/100, batch 360/500 completed (loss: 0.065:  72%|███████▏  | 180/250 [18:13<07:10,  6.15s/it] 
Training Epoch: 15/100, batch 360/500 completed (loss: 0.0193:  72%|███████▏  | 180/250 [18:13<07:10,  6.15s/it]
Training Epoch: 15/100, batch 360/500 completed (loss: 0.0088:  72%|███████▏  | 180/250 [18:13<07:10,  6.15s/it]
Training Epoch: 15/100, batch 360/500 completed (loss: 0.0576:  72%|███████▏  | 180/250 [18:13<07:10,  6.15s/it]
Training Epoch: 15/100, batch 360/500 completed (loss: 0.0088:  72%|███████▏  | 181/250 [18:16<07:01,  6.11s/it]
Training Epoch: 15/100, batch 360/500 completed (loss: 0.0193:  72%|███████▏  | 181/250 [18:16<07:01,  6.11s/it]
Training Epoch: 15/100, batch 360/500 completed (loss: 0.065:  72%|███████▏  | 181/250 [18:16<07:01,  6.11s/it]
Training Epoch: 15/100, batch 360/500 completed (loss: 0.0576:  72%|███████▏  | 181/250 [18:16<07:01,  6.12s/it]
Training Epoch: 15/100, batch 361/500 completed (loss: 0.0541:  72%|███████▏  | 181/250 [18:16<07:01,  6.11s/it]
Training Epoch: 15/100, batch 361/500 completed (loss: 0.0513:  72%|███████▏  | 181/250 [18:16<07:01,  6.11s/it]
Training Epoch: 15/100, batch 361/500 completed (loss: 0.1356:  72%|███████▏  | 181/250 [18:16<07:01,  6.11s/it]
Training Epoch: 15/100, batch 361/500 completed (loss: 0.0967:  72%|███████▏  | 181/250 [18:16<07:01,  6.12s/it]
Training Epoch: 15/100, batch 362/500 completed (loss: 0.0044:  72%|███████▏  | 181/250 [18:19<07:01,  6.11s/it]
Training Epoch: 15/100, batch 362/500 completed (loss: 0.0141:  72%|███████▏  | 181/250 [18:19<07:01,  6.11s/it]
Training Epoch: 15/100, batch 362/500 completed (loss: 0.0654:  72%|███████▏  | 181/250 [18:19<07:01,  6.11s/it]
Training Epoch: 15/100, batch 362/500 completed (loss: 0.018:  72%|███████▏  | 181/250 [18:19<07:01,  6.12s/it] 
Training Epoch: 15/100, batch 362/500 completed (loss: 0.018:  73%|███████▎  | 182/250 [18:22<06:53,  6.08s/it]
Training Epoch: 15/100, batch 362/500 completed (loss: 0.0141:  73%|███████▎  | 182/250 [18:22<06:53,  6.08s/it]
Training Epoch: 15/100, batch 362/500 completed (loss: 0.0654:  73%|███████▎  | 182/250 [18:22<06:53,  6.08s/it]
Training Epoch: 15/100, batch 362/500 completed (loss: 0.0044:  73%|███████▎  | 182/250 [18:22<06:53,  6.08s/it]
Training Epoch: 15/100, batch 363/500 completed (loss: 0.2629:  73%|███████▎  | 182/250 [18:22<06:53,  6.08s/it]
Training Epoch: 15/100, batch 363/500 completed (loss: 0.0254:  73%|███████▎  | 182/250 [18:22<06:53,  6.08s/it]
Training Epoch: 15/100, batch 363/500 completed (loss: 0.0424:  73%|███████▎  | 182/250 [18:22<06:53,  6.08s/it]
Training Epoch: 15/100, batch 363/500 completed (loss: 0.0363:  73%|███████▎  | 182/250 [18:22<06:53,  6.08s/it]
Training Epoch: 15/100, batch 364/500 completed (loss: 0.0339:  73%|███████▎  | 182/250 [18:25<06:53,  6.08s/it]
Training Epoch: 15/100, batch 364/500 completed (loss: 0.0386:  73%|███████▎  | 182/250 [18:25<06:53,  6.08s/it]
Training Epoch: 15/100, batch 364/500 completed (loss: 0.0177:  73%|███████▎  | 182/250 [18:25<06:53,  6.08s/it]
Training Epoch: 15/100, batch 364/500 completed (loss: 0.0272:  73%|███████▎  | 182/250 [18:25<06:53,  6.08s/it]
Training Epoch: 15/100, batch 364/500 completed (loss: 0.0272:  73%|███████▎  | 183/250 [18:28<06:48,  6.10s/it]
Training Epoch: 15/100, batch 364/500 completed (loss: 0.0339:  73%|███████▎  | 183/250 [18:28<06:48,  6.10s/it]
Training Epoch: 15/100, batch 364/500 completed (loss: 0.0386:  73%|███████▎  | 183/250 [18:28<06:48,  6.10s/it]
Training Epoch: 15/100, batch 364/500 completed (loss: 0.0177:  73%|███████▎  | 183/250 [18:28<06:49,  6.11s/it]
Training Epoch: 15/100, batch 365/500 completed (loss: 0.1183:  73%|███████▎  | 183/250 [18:28<06:48,  6.10s/it]
Training Epoch: 15/100, batch 365/500 completed (loss: 0.0299:  73%|███████▎  | 183/250 [18:28<06:49,  6.11s/it]
Training Epoch: 15/100, batch 365/500 completed (loss: 0.012:  73%|███████▎  | 183/250 [18:28<06:48,  6.10s/it] 
Training Epoch: 15/100, batch 365/500 completed (loss: 0.0299:  73%|███████▎  | 183/250 [18:28<06:48,  6.10s/it]
Training Epoch: 15/100, batch 366/500 completed (loss: 0.0446:  73%|███████▎  | 183/250 [18:31<06:48,  6.10s/it]
Training Epoch: 15/100, batch 366/500 completed (loss: 0.006:  73%|███████▎  | 183/250 [18:31<06:48,  6.10s/it]
Training Epoch: 15/100, batch 366/500 completed (loss: 0.1906:  73%|███████▎  | 183/250 [18:31<06:49,  6.11s/it]
Training Epoch: 15/100, batch 366/500 completed (loss: 0.0805:  73%|███████▎  | 183/250 [18:31<06:48,  6.10s/it]
Training Epoch: 15/100, batch 366/500 completed (loss: 0.0446:  74%|███████▎  | 184/250 [18:34<06:41,  6.09s/it]
Training Epoch: 15/100, batch 366/500 completed (loss: 0.006:  74%|███████▎  | 184/250 [18:34<06:41,  6.09s/it]
Training Epoch: 15/100, batch 366/500 completed (loss: 0.1906:  74%|███████▎  | 184/250 [18:34<06:41,  6.09s/it]
Training Epoch: 15/100, batch 366/500 completed (loss: 0.0805:  74%|███████▎  | 184/250 [18:34<06:41,  6.09s/it]
Training Epoch: 15/100, batch 367/500 completed (loss: 0.074:  74%|███████▎  | 184/250 [18:34<06:41,  6.09s/it] 
Training Epoch: 15/100, batch 367/500 completed (loss: 0.0536:  74%|███████▎  | 184/250 [18:34<06:41,  6.09s/it]
Training Epoch: 15/100, batch 367/500 completed (loss: 0.1528:  74%|███████▎  | 184/250 [18:34<06:41,  6.09s/it]
Training Epoch: 15/100, batch 367/500 completed (loss: 0.0103:  74%|███████▎  | 184/250 [18:34<06:41,  6.09s/it]
Training Epoch: 15/100, batch 368/500 completed (loss: 0.0616:  74%|███████▎  | 184/250 [18:37<06:41,  6.09s/it]
Training Epoch: 15/100, batch 368/500 completed (loss: 0.0196:  74%|███████▎  | 184/250 [18:37<06:41,  6.09s/it]
Training Epoch: 15/100, batch 368/500 completed (loss: 0.0278:  74%|███████▎  | 184/250 [18:37<06:41,  6.09s/it]
Training Epoch: 15/100, batch 368/500 completed (loss: 0.0614:  74%|███████▎  | 184/250 [18:37<06:41,  6.09s/it]
Training Epoch: 15/100, batch 368/500 completed (loss: 0.0278:  74%|███████▍  | 185/250 [18:40<06:34,  6.07s/it]
Training Epoch: 15/100, batch 368/500 completed (loss: 0.0616:  74%|███████▍  | 185/250 [18:40<06:34,  6.07s/it]
Training Epoch: 15/100, batch 368/500 completed (loss: 0.0196:  74%|███████▍  | 185/250 [18:40<06:34,  6.07s/it]
Training Epoch: 15/100, batch 368/500 completed (loss: 0.0614:  74%|███████▍  | 185/250 [18:40<06:34,  6.08s/it]
Training Epoch: 15/100, batch 369/500 completed (loss: 0.1023:  74%|███████▍  | 185/250 [18:40<06:34,  6.07s/it]
Training Epoch: 15/100, batch 369/500 completed (loss: 0.0587:  74%|███████▍  | 185/250 [18:40<06:34,  6.07s/it]
Training Epoch: 15/100, batch 369/500 completed (loss: 0.011:  74%|███████▍  | 185/250 [18:40<06:34,  6.07s/it] 
Training Epoch: 15/100, batch 369/500 completed (loss: 0.1302:  74%|███████▍  | 185/250 [18:40<06:34,  6.08s/it]
Training Epoch: 15/100, batch 370/500 completed (loss: 0.0968:  74%|███████▍  | 185/250 [18:43<06:34,  6.07s/it]
Training Epoch: 15/100, batch 370/500 completed (loss: 0.1342:  74%|███████▍  | 185/250 [18:43<06:34,  6.07s/it]
Training Epoch: 15/100, batch 370/500 completed (loss: 0.0739:  74%|███████▍  | 185/250 [18:43<06:34,  6.07s/it]
Training Epoch: 15/100, batch 370/500 completed (loss: 0.0374:  74%|███████▍  | 185/250 [18:43<06:34,  6.08s/it]
Training Epoch: 15/100, batch 370/500 completed (loss: 0.1342:  74%|███████▍  | 186/250 [18:46<06:27,  6.06s/it]
Training Epoch: 15/100, batch 370/500 completed (loss: 0.0374:  74%|███████▍  | 186/250 [18:46<06:27,  6.05s/it]
Training Epoch: 15/100, batch 370/500 completed (loss: 0.0968:  74%|███████▍  | 186/250 [18:46<06:27,  6.06s/it]
Training Epoch: 15/100, batch 370/500 completed (loss: 0.0739:  74%|███████▍  | 186/250 [18:46<06:27,  6.06s/it]
Training Epoch: 15/100, batch 371/500 completed (loss: 0.0906:  74%|███████▍  | 186/250 [18:46<06:27,  6.06s/it]
Training Epoch: 15/100, batch 371/500 completed (loss: 0.0056:  74%|███████▍  | 186/250 [18:46<06:27,  6.06s/it]
Training Epoch: 15/100, batch 371/500 completed (loss: 0.131:  74%|███████▍  | 186/250 [18:46<06:27,  6.06s/it] 
Training Epoch: 15/100, batch 371/500 completed (loss: 0.1752:  74%|███████▍  | 186/250 [18:46<06:27,  6.05s/it]
Training Epoch: 15/100, batch 372/500 completed (loss: 0.0047:  74%|███████▍  | 186/250 [18:49<06:27,  6.06s/it]
Training Epoch: 15/100, batch 372/500 completed (loss: 0.2118:  74%|███████▍  | 186/250 [18:49<06:27,  6.06s/it]
Training Epoch: 15/100, batch 372/500 completed (loss: 0.1214:  74%|███████▍  | 186/250 [18:49<06:27,  6.06s/it]
Training Epoch: 15/100, batch 372/500 completed (loss: 0.0478:  74%|███████▍  | 186/250 [18:49<06:27,  6.05s/it]
Training Epoch: 15/100, batch 372/500 completed (loss: 0.1214:  75%|███████▍  | 187/250 [18:52<06:24,  6.10s/it]
Training Epoch: 15/100, batch 372/500 completed (loss: 0.0047:  75%|███████▍  | 187/250 [18:52<06:24,  6.10s/it]
Training Epoch: 15/100, batch 372/500 completed (loss: 0.2118:  75%|███████▍  | 187/250 [18:52<06:24,  6.10s/it]
Training Epoch: 15/100, batch 372/500 completed (loss: 0.0478:  75%|███████▍  | 187/250 [18:52<06:24,  6.10s/it]
Training Epoch: 15/100, batch 373/500 completed (loss: 0.0691:  75%|███████▍  | 187/250 [18:52<06:24,  6.10s/it]
Training Epoch: 15/100, batch 373/500 completed (loss: 0.428:  75%|███████▍  | 187/250 [18:52<06:24,  6.10s/it] 
Training Epoch: 15/100, batch 373/500 completed (loss: 0.009:  75%|███████▍  | 187/250 [18:52<06:24,  6.10s/it] 
Training Epoch: 15/100, batch 373/500 completed (loss: 0.029:  75%|███████▍  | 187/250 [18:52<06:24,  6.10s/it] 
Training Epoch: 15/100, batch 374/500 completed (loss: 0.0257:  75%|███████▍  | 187/250 [18:56<06:24,  6.10s/it]
Training Epoch: 15/100, batch 374/500 completed (loss: 0.141:  75%|███████▍  | 187/250 [18:56<06:24,  6.10s/it]
Training Epoch: 15/100, batch 374/500 completed (loss: 0.0062:  75%|███████▍  | 187/250 [18:55<06:24,  6.10s/it]
Training Epoch: 15/100, batch 374/500 completed (loss: 0.0519:  75%|███████▍  | 187/250 [18:56<06:24,  6.10s/it]
Training Epoch: 15/100, batch 374/500 completed (loss: 0.0519:  75%|███████▌  | 188/250 [18:59<06:19,  6.12s/it]
Training Epoch: 15/100, batch 374/500 completed (loss: 0.141:  75%|███████▌  | 188/250 [18:59<06:19,  6.12s/it]
Training Epoch: 15/100, batch 374/500 completed (loss: 0.0062:  75%|███████▌  | 188/250 [18:59<06:19,  6.12s/it]
Training Epoch: 15/100, batch 374/500 completed (loss: 0.0257:  75%|███████▌  | 188/250 [18:59<06:19,  6.12s/it]
Training Epoch: 15/100, batch 375/500 completed (loss: 0.0789:  75%|███████▌  | 188/250 [18:59<06:19,  6.12s/it]
Training Epoch: 15/100, batch 375/500 completed (loss: 0.251:  75%|███████▌  | 188/250 [18:59<06:19,  6.12s/it]
Training Epoch: 15/100, batch 375/500 completed (loss: 0.0036:  75%|███████▌  | 188/250 [18:59<06:19,  6.12s/it]
Training Epoch: 15/100, batch 375/500 completed (loss: 0.2243:  75%|███████▌  | 188/250 [18:59<06:19,  6.12s/it]
Training Epoch: 15/100, batch 376/500 completed (loss: 0.0079:  75%|███████▌  | 188/250 [19:02<06:19,  6.12s/it]
Training Epoch: 15/100, batch 376/500 completed (loss: 0.1668:  75%|███████▌  | 188/250 [19:02<06:19,  6.12s/it]
Training Epoch: 15/100, batch 376/500 completed (loss: 0.1434:  75%|███████▌  | 188/250 [19:02<06:19,  6.12s/it]
Training Epoch: 15/100, batch 376/500 completed (loss: 0.0277:  75%|███████▌  | 188/250 [19:02<06:19,  6.12s/it]
Training Epoch: 15/100, batch 376/500 completed (loss: 0.0277:  76%|███████▌  | 189/250 [19:04<06:08,  6.04s/it]
Training Epoch: 15/100, batch 376/500 completed (loss: 0.1668:  76%|███████▌  | 189/250 [19:04<06:08,  6.04s/it]
Training Epoch: 15/100, batch 376/500 completed (loss: 0.1434:  76%|███████▌  | 189/250 [19:04<06:08,  6.04s/it]
Training Epoch: 15/100, batch 376/500 completed (loss: 0.0079:  76%|███████▌  | 189/250 [19:04<06:08,  6.04s/it]
Training Epoch: 15/100, batch 377/500 completed (loss: 0.2205:  76%|███████▌  | 189/250 [19:04<06:08,  6.04s/it]
Training Epoch: 15/100, batch 377/500 completed (loss: 0.031:  76%|███████▌  | 189/250 [19:04<06:08,  6.04s/it] 
Training Epoch: 15/100, batch 377/500 completed (loss: 0.0364:  76%|███████▌  | 189/250 [19:04<06:08,  6.04s/it]
Training Epoch: 15/100, batch 377/500 completed (loss: 0.0238:  76%|███████▌  | 189/250 [19:04<06:08,  6.04s/it]
Training Epoch: 15/100, batch 378/500 completed (loss: 0.021:  76%|███████▌  | 189/250 [19:08<06:08,  6.04s/it] 
Training Epoch: 15/100, batch 378/500 completed (loss: 0.0345:  76%|███████▌  | 189/250 [19:08<06:08,  6.04s/it]
Training Epoch: 15/100, batch 378/500 completed (loss: 0.0039:  76%|███████▌  | 189/250 [19:08<06:08,  6.04s/it]
Training Epoch: 15/100, batch 378/500 completed (loss: 0.0394:  76%|███████▌  | 189/250 [19:08<06:08,  6.04s/it]
Training Epoch: 15/100, batch 378/500 completed (loss: 0.0394:  76%|███████▌  | 190/250 [19:11<06:07,  6.13s/it]
Training Epoch: 15/100, batch 378/500 completed (loss: 0.0345:  76%|███████▌  | 190/250 [19:11<06:08,  6.13s/it]
Training Epoch: 15/100, batch 378/500 completed (loss: 0.0039:  76%|███████▌  | 190/250 [19:11<06:08,  6.14s/it]
Training Epoch: 15/100, batch 378/500 completed (loss: 0.021:  76%|███████▌  | 190/250 [19:11<06:08,  6.14s/it]
Training Epoch: 15/100, batch 379/500 completed (loss: 0.242:  76%|███████▌  | 190/250 [19:11<06:08,  6.14s/it]
Training Epoch: 15/100, batch 379/500 completed (loss: 0.0204:  76%|███████▌  | 190/250 [19:11<06:08,  6.14s/it]
Training Epoch: 15/100, batch 379/500 completed (loss: 0.1431:  76%|███████▌  | 190/250 [19:11<06:08,  6.13s/it]
Training Epoch: 15/100, batch 379/500 completed (loss: 0.0282:  76%|███████▌  | 190/250 [19:11<06:07,  6.13s/it]
Training Epoch: 15/100, batch 380/500 completed (loss: 0.0194:  76%|███████▌  | 190/250 [19:14<06:08,  6.13s/it]
Training Epoch: 15/100, batch 380/500 completed (loss: 0.0303:  76%|███████▌  | 190/250 [19:14<06:08,  6.14s/it]
Training Epoch: 15/100, batch 380/500 completed (loss: 0.0437:  76%|███████▌  | 190/250 [19:14<06:08,  6.14s/it]
Training Epoch: 15/100, batch 380/500 completed (loss: 0.0986:  76%|███████▌  | 190/250 [19:14<06:07,  6.13s/it]
Training Epoch: 15/100, batch 380/500 completed (loss: 0.0437:  76%|███████▋  | 191/250 [19:17<05:59,  6.10s/it]
Training Epoch: 15/100, batch 380/500 completed (loss: 0.0194:  76%|███████▋  | 191/250 [19:17<05:59,  6.10s/it]
Training Epoch: 15/100, batch 380/500 completed (loss: 0.0303:  76%|███████▋  | 191/250 [19:17<05:59,  6.10s/it]
Training Epoch: 15/100, batch 380/500 completed (loss: 0.0986:  76%|███████▋  | 191/250 [19:17<06:00,  6.11s/it]
Training Epoch: 15/100, batch 381/500 completed (loss: 0.1449:  76%|███████▋  | 191/250 [19:17<05:59,  6.10s/it]
Training Epoch: 15/100, batch 381/500 completed (loss: 0.3177:  76%|███████▋  | 191/250 [19:17<05:59,  6.10s/it]
Training Epoch: 15/100, batch 381/500 completed (loss: 0.0049:  76%|███████▋  | 191/250 [19:17<05:59,  6.10s/it]
Training Epoch: 15/100, batch 381/500 completed (loss: 0.007:  76%|███████▋  | 191/250 [19:17<06:00,  6.11s/it] 
Training Epoch: 15/100, batch 382/500 completed (loss: 0.027:  76%|███████▋  | 191/250 [19:20<05:59,  6.10s/it] 
Training Epoch: 15/100, batch 382/500 completed (loss: 0.0593:  76%|███████▋  | 191/250 [19:20<05:59,  6.10s/it]
Training Epoch: 15/100, batch 382/500 completed (loss: 0.2008:  76%|███████▋  | 191/250 [19:20<05:59,  6.10s/it]
Training Epoch: 15/100, batch 382/500 completed (loss: 0.0477:  76%|███████▋  | 191/250 [19:20<06:00,  6.11s/it]
Training Epoch: 15/100, batch 382/500 completed (loss: 0.0477:  77%|███████▋  | 192/250 [19:23<05:49,  6.03s/it]
Training Epoch: 15/100, batch 382/500 completed (loss: 0.027:  77%|███████▋  | 192/250 [19:23<05:49,  6.03s/it]
Training Epoch: 15/100, batch 382/500 completed (loss: 0.0593:  77%|███████▋  | 192/250 [19:23<05:49,  6.03s/it]
Training Epoch: 15/100, batch 382/500 completed (loss: 0.2008:  77%|███████▋  | 192/250 [19:23<05:49,  6.03s/it]
Training Epoch: 15/100, batch 383/500 completed (loss: 0.0963:  77%|███████▋  | 192/250 [19:23<05:49,  6.03s/it]
Training Epoch: 15/100, batch 383/500 completed (loss: 0.0108:  77%|███████▋  | 192/250 [19:23<05:49,  6.03s/it]
Training Epoch: 15/100, batch 383/500 completed (loss: 0.028:  77%|███████▋  | 192/250 [19:23<05:49,  6.03s/it] 
Training Epoch: 15/100, batch 383/500 completed (loss: 0.0297:  77%|███████▋  | 192/250 [19:23<05:49,  6.03s/it]
Training Epoch: 15/100, batch 384/500 completed (loss: 0.0179:  77%|███████▋  | 192/250 [19:26<05:49,  6.03s/it]
Training Epoch: 15/100, batch 384/500 completed (loss: 0.0355:  77%|███████▋  | 192/250 [19:26<05:49,  6.03s/it]
Training Epoch: 15/100, batch 384/500 completed (loss: 0.0693:  77%|███████▋  | 192/250 [19:26<05:49,  6.03s/it]
Training Epoch: 15/100, batch 384/500 completed (loss: 0.0869:  77%|███████▋  | 192/250 [19:26<05:49,  6.03s/it]
Training Epoch: 15/100, batch 384/500 completed (loss: 0.0355:  77%|███████▋  | 193/250 [19:29<05:49,  6.12s/it]
Training Epoch: 15/100, batch 384/500 completed (loss: 0.0693:  77%|███████▋  | 193/250 [19:29<05:49,  6.12s/it]
Training Epoch: 15/100, batch 384/500 completed (loss: 0.0179:  77%|███████▋  | 193/250 [19:29<05:49,  6.13s/it]
Training Epoch: 15/100, batch 384/500 completed (loss: 0.0869:  77%|███████▋  | 193/250 [19:29<05:49,  6.13s/it]
Training Epoch: 15/100, batch 385/500 completed (loss: 0.1839:  77%|███████▋  | 193/250 [19:29<05:49,  6.12s/it]
Training Epoch: 15/100, batch 385/500 completed (loss: 0.0675:  77%|███████▋  | 193/250 [19:29<05:49,  6.12s/it]
Training Epoch: 15/100, batch 385/500 completed (loss: 0.1557:  77%|███████▋  | 193/250 [19:29<05:49,  6.13s/it]
Training Epoch: 15/100, batch 385/500 completed (loss: 0.0945:  77%|███████▋  | 193/250 [19:29<05:49,  6.13s/it]
Training Epoch: 15/100, batch 386/500 completed (loss: 0.0251:  77%|███████▋  | 193/250 [19:32<05:49,  6.12s/it]
Training Epoch: 15/100, batch 386/500 completed (loss: 0.0611:  77%|███████▋  | 193/250 [19:32<05:49,  6.13s/it]
Training Epoch: 15/100, batch 386/500 completed (loss: 0.0542:  77%|███████▋  | 193/250 [19:32<05:49,  6.12s/it]
Training Epoch: 15/100, batch 386/500 completed (loss: 0.1647:  77%|███████▋  | 193/250 [19:32<05:49,  6.13s/it]
Training Epoch: 15/100, batch 386/500 completed (loss: 0.0611:  78%|███████▊  | 194/250 [19:35<05:38,  6.05s/it]
Training Epoch: 15/100, batch 386/500 completed (loss: 0.0542:  78%|███████▊  | 194/250 [19:35<05:38,  6.05s/it]
Training Epoch: 15/100, batch 386/500 completed (loss: 0.0251:  78%|███████▊  | 194/250 [19:35<05:39,  6.05s/it]
Training Epoch: 15/100, batch 386/500 completed (loss: 0.1647:  78%|███████▊  | 194/250 [19:35<05:39,  6.06s/it]
Training Epoch: 15/100, batch 387/500 completed (loss: 0.0145:  78%|███████▊  | 194/250 [19:35<05:38,  6.05s/it]
Training Epoch: 15/100, batch 387/500 completed (loss: 0.0661:  78%|███████▊  | 194/250 [19:35<05:39,  6.05s/it]
Training Epoch: 15/100, batch 387/500 completed (loss: 0.0145:  78%|███████▊  | 194/250 [19:35<05:38,  6.05s/it]
Training Epoch: 15/100, batch 387/500 completed (loss: 0.075:  78%|███████▊  | 194/250 [19:35<05:39,  6.06s/it] 
Training Epoch: 15/100, batch 388/500 completed (loss: 0.2323:  78%|███████▊  | 194/250 [19:38<05:38,  6.05s/it]
Training Epoch: 15/100, batch 388/500 completed (loss: 0.0163:  78%|███████▊  | 194/250 [19:38<05:39,  6.05s/it]
Training Epoch: 15/100, batch 388/500 completed (loss: 0.4024:  78%|███████▊  | 194/250 [19:38<05:38,  6.05s/it]
Training Epoch: 15/100, batch 388/500 completed (loss: 0.0983:  78%|███████▊  | 194/250 [19:38<05:39,  6.06s/it]
Training Epoch: 15/100, batch 388/500 completed (loss: 0.2323:  78%|███████▊  | 195/250 [19:41<05:33,  6.06s/it]
Training Epoch: 15/100, batch 388/500 completed (loss: 0.0163:  78%|███████▊  | 195/250 [19:41<05:33,  6.06s/it]
Training Epoch: 15/100, batch 388/500 completed (loss: 0.0983:  78%|███████▊  | 195/250 [19:41<05:33,  6.06s/it]
Training Epoch: 15/100, batch 388/500 completed (loss: 0.4024:  78%|███████▊  | 195/250 [19:41<05:33,  6.06s/it]
Training Epoch: 15/100, batch 389/500 completed (loss: 0.0038:  78%|███████▊  | 195/250 [19:41<05:33,  6.06s/it]
Training Epoch: 15/100, batch 389/500 completed (loss: 0.0165:  78%|███████▊  | 195/250 [19:41<05:33,  6.06s/it]
Training Epoch: 15/100, batch 389/500 completed (loss: 0.7303:  78%|███████▊  | 195/250 [19:41<05:33,  6.06s/it]
Training Epoch: 15/100, batch 389/500 completed (loss: 0.0594:  78%|███████▊  | 195/250 [19:41<05:33,  6.06s/it]
Training Epoch: 15/100, batch 390/500 completed (loss: 0.0905:  78%|███████▊  | 195/250 [19:44<05:33,  6.06s/it]
Training Epoch: 15/100, batch 390/500 completed (loss: 0.0786:  78%|███████▊  | 195/250 [19:44<05:33,  6.06s/it]
Training Epoch: 15/100, batch 390/500 completed (loss: 0.019:  78%|███████▊  | 195/250 [19:44<05:33,  6.06s/it] 
Training Epoch: 15/100, batch 390/500 completed (loss: 0.0293:  78%|███████▊  | 195/250 [19:44<05:33,  6.06s/it]
Training Epoch: 15/100, batch 390/500 completed (loss: 0.0905:  78%|███████▊  | 196/250 [19:47<05:29,  6.09s/it]
Training Epoch: 15/100, batch 390/500 completed (loss: 0.0786:  78%|███████▊  | 196/250 [19:47<05:29,  6.09s/it]
Training Epoch: 15/100, batch 390/500 completed (loss: 0.019:  78%|███████▊  | 196/250 [19:47<05:29,  6.09s/it]
Training Epoch: 15/100, batch 390/500 completed (loss: 0.0293:  78%|███████▊  | 196/250 [19:47<05:29,  6.09s/it]
Training Epoch: 15/100, batch 391/500 completed (loss: 0.1985:  78%|███████▊  | 196/250 [19:47<05:29,  6.09s/it]
Training Epoch: 15/100, batch 391/500 completed (loss: 0.007:  78%|███████▊  | 196/250 [19:47<05:29,  6.09s/it]
Training Epoch: 15/100, batch 391/500 completed (loss: 0.0087:  78%|███████▊  | 196/250 [19:47<05:29,  6.09s/it]
Training Epoch: 15/100, batch 391/500 completed (loss: 0.0159:  78%|███████▊  | 196/250 [19:47<05:29,  6.09s/it]
Training Epoch: 15/100, batch 392/500 completed (loss: 0.1734:  78%|███████▊  | 196/250 [19:50<05:29,  6.09s/it]
Training Epoch: 15/100, batch 392/500 completed (loss: 0.0344:  78%|███████▊  | 196/250 [19:50<05:29,  6.09s/it]
Training Epoch: 15/100, batch 392/500 completed (loss: 0.0082:  78%|███████▊  | 196/250 [19:50<05:29,  6.09s/it]
Training Epoch: 15/100, batch 392/500 completed (loss: 0.0745:  78%|███████▊  | 196/250 [19:50<05:29,  6.09s/it]
Training Epoch: 15/100, batch 392/500 completed (loss: 0.0082:  79%|███████▉  | 197/250 [19:53<05:18,  6.02s/it]
Training Epoch: 15/100, batch 392/500 completed (loss: 0.0344:  79%|███████▉  | 197/250 [19:53<05:19,  6.02s/it]
Training Epoch: 15/100, batch 392/500 completed (loss: 0.1734:  79%|███████▉  | 197/250 [19:53<05:19,  6.02s/it]
Training Epoch: 15/100, batch 392/500 completed (loss: 0.0745:  79%|███████▉  | 197/250 [19:53<05:19,  6.02s/it]
Training Epoch: 15/100, batch 393/500 completed (loss: 0.04:  79%|███████▉  | 197/250 [19:53<05:19,  6.02s/it]  
Training Epoch: 15/100, batch 393/500 completed (loss: 0.0823:  79%|███████▉  | 197/250 [19:53<05:18,  6.02s/it]
Training Epoch: 15/100, batch 393/500 completed (loss: 0.2455:  79%|███████▉  | 197/250 [19:53<05:19,  6.02s/it]
Training Epoch: 15/100, batch 393/500 completed (loss: 0.0296:  79%|███████▉  | 197/250 [19:53<05:19,  6.02s/it]
Training Epoch: 15/100, batch 394/500 completed (loss: 0.0205:  79%|███████▉  | 197/250 [19:56<05:19,  6.02s/it]
Training Epoch: 15/100, batch 394/500 completed (loss: 0.1358:  79%|███████▉  | 197/250 [19:56<05:19,  6.02s/it]
Training Epoch: 15/100, batch 394/500 completed (loss: 0.011:  79%|███████▉  | 197/250 [19:56<05:18,  6.02s/it] 
Training Epoch: 15/100, batch 394/500 completed (loss: 0.2015:  79%|███████▉  | 197/250 [19:56<05:19,  6.02s/it]
Training Epoch: 15/100, batch 394/500 completed (loss: 0.0205:  79%|███████▉  | 198/250 [19:59<05:17,  6.10s/it]
Training Epoch: 15/100, batch 394/500 completed (loss: 0.011:  79%|███████▉  | 198/250 [19:59<05:17,  6.10s/it]
Training Epoch: 15/100, batch 394/500 completed (loss: 0.1358:  79%|███████▉  | 198/250 [19:59<05:17,  6.11s/it]
Training Epoch: 15/100, batch 394/500 completed (loss: 0.2015:  79%|███████▉  | 198/250 [19:59<05:17,  6.11s/it]
Training Epoch: 15/100, batch 395/500 completed (loss: 0.0633:  79%|███████▉  | 198/250 [19:59<05:17,  6.10s/it]
Training Epoch: 15/100, batch 395/500 completed (loss: 0.0157:  79%|███████▉  | 198/250 [19:59<05:17,  6.11s/it]
Training Epoch: 15/100, batch 395/500 completed (loss: 0.1105:  79%|███████▉  | 198/250 [19:59<05:17,  6.10s/it]
Training Epoch: 15/100, batch 395/500 completed (loss: 0.0163:  79%|███████▉  | 198/250 [19:59<05:17,  6.11s/it]
Training Epoch: 15/100, batch 396/500 completed (loss: 0.009:  79%|███████▉  | 198/250 [20:02<05:17,  6.10s/it] 
Training Epoch: 15/100, batch 396/500 completed (loss: 0.0068:  79%|███████▉  | 198/250 [20:02<05:17,  6.11s/it]
Training Epoch: 15/100, batch 396/500 completed (loss: 0.0153:  79%|███████▉  | 198/250 [20:02<05:17,  6.10s/it]
Training Epoch: 15/100, batch 396/500 completed (loss: 0.1485:  79%|███████▉  | 198/250 [20:02<05:17,  6.11s/it]
Training Epoch: 15/100, batch 396/500 completed (loss: 0.1485:  80%|███████▉  | 199/250 [20:05<05:09,  6.07s/it]
Training Epoch: 15/100, batch 396/500 completed (loss: 0.009:  80%|███████▉  | 199/250 [20:05<05:09,  6.08s/it]
Training Epoch: 15/100, batch 396/500 completed (loss: 0.0153:  80%|███████▉  | 199/250 [20:05<05:09,  6.08s/it]
Training Epoch: 15/100, batch 396/500 completed (loss: 0.0068:  80%|███████▉  | 199/250 [20:05<05:09,  6.07s/it]
Training Epoch: 15/100, batch 397/500 completed (loss: 0.0991:  80%|███████▉  | 199/250 [20:05<05:09,  6.08s/it]
Training Epoch: 15/100, batch 397/500 completed (loss: 0.0167:  80%|███████▉  | 199/250 [20:05<05:09,  6.07s/it]
Training Epoch: 15/100, batch 397/500 completed (loss: 0.0463:  80%|███████▉  | 199/250 [20:05<05:09,  6.08s/it]
Training Epoch: 15/100, batch 397/500 completed (loss: 0.0419:  80%|███████▉  | 199/250 [20:05<05:09,  6.07s/it]
Training Epoch: 15/100, batch 398/500 completed (loss: 0.0439:  80%|███████▉  | 199/250 [20:08<05:09,  6.08s/it]
Training Epoch: 15/100, batch 398/500 completed (loss: 0.6577:  80%|███████▉  | 199/250 [20:08<05:09,  6.08s/it]
Training Epoch: 15/100, batch 398/500 completed (loss: 0.112:  80%|███████▉  | 199/250 [20:08<05:09,  6.07s/it] 
Training Epoch: 15/100, batch 398/500 completed (loss: 0.0448:  80%|███████▉  | 199/250 [20:08<05:09,  6.07s/it]
Training Epoch: 15/100, batch 398/500 completed (loss: 0.0448:  80%|████████  | 200/250 [20:11<05:00,  6.01s/it]
Training Epoch: 15/100, batch 398/500 completed (loss: 0.6577:  80%|████████  | 200/250 [20:11<05:00,  6.02s/it]
Training Epoch: 15/100, batch 398/500 completed (loss: 0.112:  80%|████████  | 200/250 [20:11<05:01,  6.02s/it]
Training Epoch: 15/100, batch 398/500 completed (loss: 0.0439:  80%|████████  | 200/250 [20:11<05:01,  6.02s/it]
Training Epoch: 15/100, batch 399/500 completed (loss: 0.0114:  80%|████████  | 200/250 [20:11<05:00,  6.02s/it]
Training Epoch: 15/100, batch 399/500 completed (loss: 0.3554:  80%|████████  | 200/250 [20:11<05:01,  6.02s/it]
Training Epoch: 15/100, batch 399/500 completed (loss: 0.0328:  80%|████████  | 200/250 [20:11<05:01,  6.02s/it]
Training Epoch: 15/100, batch 399/500 completed (loss: 0.098:  80%|████████  | 200/250 [20:11<05:00,  6.01s/it] 
Training Epoch: 15/100, batch 400/500 completed (loss: 0.01:  80%|████████  | 200/250 [20:15<05:00,  6.02s/it]  
Training Epoch: 15/100, batch 400/500 completed (loss: 0.0044:  80%|████████  | 200/250 [20:15<05:01,  6.02s/it]
Training Epoch: 15/100, batch 400/500 completed (loss: 0.1246:  80%|████████  | 200/250 [20:15<05:01,  6.02s/it]
Training Epoch: 15/100, batch 400/500 completed (loss: 0.1938:  80%|████████  | 200/250 [20:15<05:00,  6.01s/it]
Training Epoch: 15/100, batch 400/500 completed (loss: 0.01:  80%|████████  | 201/250 [20:18<05:01,  6.16s/it]
Training Epoch: 15/100, batch 400/500 completed (loss: 0.0044:  80%|████████  | 201/250 [20:18<05:01,  6.16s/it]
Training Epoch: 15/100, batch 400/500 completed (loss: 0.1938:  80%|████████  | 201/250 [20:18<05:01,  6.16s/it]
Training Epoch: 15/100, batch 400/500 completed (loss: 0.1246:  80%|████████  | 201/250 [20:18<05:01,  6.16s/it]
Training Epoch: 15/100, batch 401/500 completed (loss: 0.0658:  80%|████████  | 201/250 [20:18<05:01,  6.16s/it]
Training Epoch: 15/100, batch 401/500 completed (loss: 0.0127:  80%|████████  | 201/250 [20:18<05:01,  6.16s/it]
Training Epoch: 15/100, batch 401/500 completed (loss: 0.0124:  80%|████████  | 201/250 [20:18<05:01,  6.16s/it]
Training Epoch: 15/100, batch 401/500 completed (loss: 0.0122:  80%|████████  | 201/250 [20:18<05:01,  6.16s/it]
Training Epoch: 15/100, batch 402/500 completed (loss: 0.0946:  80%|████████  | 201/250 [20:21<05:01,  6.16s/it]
Training Epoch: 15/100, batch 402/500 completed (loss: 0.4131:  80%|████████  | 201/250 [20:21<05:01,  6.16s/it]
Training Epoch: 15/100, batch 402/500 completed (loss: 0.1245:  80%|████████  | 201/250 [20:21<05:01,  6.16s/it]
Training Epoch: 15/100, batch 402/500 completed (loss: 0.0051:  80%|████████  | 201/250 [20:21<05:01,  6.16s/it]
Training Epoch: 15/100, batch 402/500 completed (loss: 0.1245:  81%|████████  | 202/250 [20:23<04:51,  6.07s/it]
Training Epoch: 15/100, batch 402/500 completed (loss: 0.4131:  81%|████████  | 202/250 [20:24<04:51,  6.07s/it]
Training Epoch: 15/100, batch 402/500 completed (loss: 0.0946:  81%|████████  | 202/250 [20:23<04:51,  6.07s/it]
Training Epoch: 15/100, batch 402/500 completed (loss: 0.0051:  81%|████████  | 202/250 [20:24<04:51,  6.07s/it]
Training Epoch: 15/100, batch 403/500 completed (loss: 0.0129:  81%|████████  | 202/250 [20:24<04:51,  6.07s/it]
Training Epoch: 15/100, batch 403/500 completed (loss: 0.027:  81%|████████  | 202/250 [20:24<04:51,  6.07s/it] 
Training Epoch: 15/100, batch 403/500 completed (loss: 0.6258:  81%|████████  | 202/250 [20:24<04:51,  6.07s/it]
Training Epoch: 15/100, batch 403/500 completed (loss: 0.2567:  81%|████████  | 202/250 [20:24<04:51,  6.07s/it]
Training Epoch: 15/100, batch 404/500 completed (loss: 0.1575:  81%|████████  | 202/250 [20:26<04:51,  6.07s/it]
Training Epoch: 15/100, batch 404/500 completed (loss: 0.0771:  81%|████████  | 202/250 [20:26<04:51,  6.07s/it]
Training Epoch: 15/100, batch 404/500 completed (loss: 0.1063:  81%|████████  | 202/250 [20:26<04:51,  6.07s/it]
Training Epoch: 15/100, batch 404/500 completed (loss: 0.0649:  81%|████████  | 202/250 [20:27<04:51,  6.07s/it]
Training Epoch: 15/100, batch 404/500 completed (loss: 0.1063:  81%|████████  | 203/250 [20:30<04:49,  6.15s/it]
Training Epoch: 15/100, batch 404/500 completed (loss: 0.0649:  81%|████████  | 203/250 [20:30<04:49,  6.15s/it]
Training Epoch: 15/100, batch 404/500 completed (loss: 0.0771:  81%|████████  | 203/250 [20:30<04:49,  6.15s/it]
Training Epoch: 15/100, batch 404/500 completed (loss: 0.1575:  81%|████████  | 203/250 [20:30<04:49,  6.15s/it]
Training Epoch: 15/100, batch 405/500 completed (loss: 0.3274:  81%|████████  | 203/250 [20:30<04:49,  6.15s/it]
Training Epoch: 15/100, batch 405/500 completed (loss: 0.0262:  81%|████████  | 203/250 [20:30<04:49,  6.15s/it]
Training Epoch: 15/100, batch 405/500 completed (loss: 0.0107:  81%|████████  | 203/250 [20:30<04:49,  6.15s/it]
Training Epoch: 15/100, batch 405/500 completed (loss: 0.0251:  81%|████████  | 203/250 [20:30<04:49,  6.15s/it]
Training Epoch: 15/100, batch 406/500 completed (loss: 0.0256:  81%|████████  | 203/250 [20:33<04:49,  6.15s/it]
Training Epoch: 15/100, batch 406/500 completed (loss: 0.0212:  81%|████████  | 203/250 [20:33<04:49,  6.15s/it]
Training Epoch: 15/100, batch 406/500 completed (loss: 0.1315:  81%|████████  | 203/250 [20:33<04:49,  6.15s/it]
Training Epoch: 15/100, batch 406/500 completed (loss: 0.002:  81%|████████  | 203/250 [20:33<04:49,  6.15s/it] 
Training Epoch: 15/100, batch 406/500 completed (loss: 0.002:  82%|████████▏ | 204/250 [20:36<04:41,  6.11s/it]
Training Epoch: 15/100, batch 406/500 completed (loss: 0.0212:  82%|████████▏ | 204/250 [20:36<04:41,  6.11s/it]
Training Epoch: 15/100, batch 406/500 completed (loss: 0.0256:  82%|████████▏ | 204/250 [20:36<04:41,  6.11s/it]
Training Epoch: 15/100, batch 406/500 completed (loss: 0.1315:  82%|████████▏ | 204/250 [20:36<04:41,  6.12s/it]
Training Epoch: 15/100, batch 407/500 completed (loss: 0.0693:  82%|████████▏ | 204/250 [20:36<04:41,  6.12s/it]
Training Epoch: 15/100, batch 407/500 completed (loss: 0.0282:  82%|████████▏ | 204/250 [20:36<04:41,  6.11s/it]
Training Epoch: 15/100, batch 407/500 completed (loss: 0.1311:  82%|████████▏ | 204/250 [20:36<04:41,  6.11s/it]
Training Epoch: 15/100, batch 407/500 completed (loss: 0.0489:  82%|████████▏ | 204/250 [20:36<04:41,  6.11s/it]
Training Epoch: 15/100, batch 408/500 completed (loss: 1.1776:  82%|████████▏ | 204/250 [20:39<04:41,  6.11s/it]
Training Epoch: 15/100, batch 408/500 completed (loss: 0.0294:  82%|████████▏ | 204/250 [20:39<04:41,  6.12s/it]
Training Epoch: 15/100, batch 408/500 completed (loss: 0.0112:  82%|████████▏ | 204/250 [20:39<04:41,  6.11s/it]
Training Epoch: 15/100, batch 408/500 completed (loss: 0.0135:  82%|████████▏ | 204/250 [20:39<04:41,  6.11s/it]
Training Epoch: 15/100, batch 408/500 completed (loss: 0.0135:  82%|████████▏ | 205/250 [20:42<04:31,  6.03s/it]
Training Epoch: 15/100, batch 408/500 completed (loss: 1.1776:  82%|████████▏ | 205/250 [20:42<04:31,  6.04s/it]
Training Epoch: 15/100, batch 408/500 completed (loss: 0.0294:  82%|████████▏ | 205/250 [20:42<04:31,  6.04s/it]
Training Epoch: 15/100, batch 408/500 completed (loss: 0.0112:  82%|████████▏ | 205/250 [20:42<04:31,  6.04s/it]
Training Epoch: 15/100, batch 409/500 completed (loss: 0.0645:  82%|████████▏ | 205/250 [20:42<04:31,  6.04s/it]
Training Epoch: 15/100, batch 409/500 completed (loss: 0.0074:  82%|████████▏ | 205/250 [20:42<04:31,  6.04s/it]
Training Epoch: 15/100, batch 409/500 completed (loss: 0.0113:  82%|████████▏ | 205/250 [20:42<04:31,  6.04s/it]
Training Epoch: 15/100, batch 409/500 completed (loss: 0.0581:  82%|████████▏ | 205/250 [20:42<04:31,  6.03s/it]
Training Epoch: 15/100, batch 410/500 completed (loss: 0.0264:  82%|████████▏ | 205/250 [20:45<04:31,  6.04s/it]
Training Epoch: 15/100, batch 410/500 completed (loss: 0.0329:  82%|████████▏ | 205/250 [20:45<04:31,  6.04s/it]
Training Epoch: 15/100, batch 410/500 completed (loss: 0.0193:  82%|████████▏ | 205/250 [20:45<04:31,  6.04s/it]
Training Epoch: 15/100, batch 410/500 completed (loss: 0.007:  82%|████████▏ | 205/250 [20:45<04:31,  6.03s/it] 
Training Epoch: 15/100, batch 410/500 completed (loss: 0.0193:  82%|████████▏ | 206/250 [20:48<04:29,  6.13s/it]
Training Epoch: 15/100, batch 410/500 completed (loss: 0.0264:  82%|████████▏ | 206/250 [20:48<04:29,  6.13s/it]
Training Epoch: 15/100, batch 410/500 completed (loss: 0.007:  82%|████████▏ | 206/250 [20:48<04:29,  6.13s/it]
Training Epoch: 15/100, batch 410/500 completed (loss: 0.0329:  82%|████████▏ | 206/250 [20:48<04:29,  6.13s/it]
Training Epoch: 15/100, batch 411/500 completed (loss: 0.0136:  82%|████████▏ | 206/250 [20:48<04:29,  6.13s/it]
Training Epoch: 15/100, batch 411/500 completed (loss: 0.0089:  82%|████████▏ | 206/250 [20:48<04:29,  6.13s/it]
Training Epoch: 15/100, batch 411/500 completed (loss: 0.1557:  82%|████████▏ | 206/250 [20:48<04:29,  6.13s/it]
Training Epoch: 15/100, batch 411/500 completed (loss: 0.2913:  82%|████████▏ | 206/250 [20:48<04:29,  6.13s/it]
Training Epoch: 15/100, batch 412/500 completed (loss: 0.0333:  82%|████████▏ | 206/250 [20:51<04:29,  6.13s/it]
Training Epoch: 15/100, batch 412/500 completed (loss: 0.1505:  82%|████████▏ | 206/250 [20:51<04:29,  6.13s/it]
Training Epoch: 15/100, batch 412/500 completed (loss: 0.0321:  82%|████████▏ | 206/250 [20:51<04:29,  6.13s/it]
Training Epoch: 15/100, batch 412/500 completed (loss: 0.2954:  82%|████████▏ | 206/250 [20:51<04:29,  6.13s/it]
Training Epoch: 15/100, batch 412/500 completed (loss: 0.2954:  83%|████████▎ | 207/250 [20:54<04:22,  6.10s/it]
Training Epoch: 15/100, batch 412/500 completed (loss: 0.0333:  83%|████████▎ | 207/250 [20:54<04:22,  6.10s/it]
Training Epoch: 15/100, batch 412/500 completed (loss: 0.1505:  83%|████████▎ | 207/250 [20:54<04:22,  6.10s/it]
Training Epoch: 15/100, batch 412/500 completed (loss: 0.0321:  83%|████████▎ | 207/250 [20:54<04:22,  6.10s/it]
Training Epoch: 15/100, batch 413/500 completed (loss: 0.2255:  83%|████████▎ | 207/250 [20:54<04:22,  6.10s/it]
Training Epoch: 15/100, batch 413/500 completed (loss: 0.2897:  83%|████████▎ | 207/250 [20:54<04:22,  6.10s/it]
Training Epoch: 15/100, batch 413/500 completed (loss: 0.0088:  83%|████████▎ | 207/250 [20:54<04:22,  6.10s/it]
Training Epoch: 15/100, batch 413/500 completed (loss: 0.4383:  83%|████████▎ | 207/250 [20:54<04:22,  6.10s/it]
Training Epoch: 15/100, batch 414/500 completed (loss: 0.1572:  83%|████████▎ | 207/250 [20:57<04:22,  6.10s/it]
Training Epoch: 15/100, batch 414/500 completed (loss: 0.0283:  83%|████████▎ | 207/250 [20:57<04:22,  6.10s/it]
Training Epoch: 15/100, batch 414/500 completed (loss: 0.0144:  83%|████████▎ | 207/250 [20:57<04:22,  6.10s/it]
Training Epoch: 15/100, batch 414/500 completed (loss: 0.0517:  83%|████████▎ | 207/250 [20:57<04:22,  6.10s/it]
Training Epoch: 15/100, batch 414/500 completed (loss: 0.0517:  83%|████████▎ | 208/250 [21:00<04:15,  6.08s/it]
Training Epoch: 15/100, batch 414/500 completed (loss: 0.0144:  83%|████████▎ | 208/250 [21:00<04:15,  6.08s/it]
Training Epoch: 15/100, batch 414/500 completed (loss: 0.0283:  83%|████████▎ | 208/250 [21:00<04:15,  6.08s/it]
Training Epoch: 15/100, batch 414/500 completed (loss: 0.1572:  83%|████████▎ | 208/250 [21:00<04:15,  6.08s/it]
Training Epoch: 15/100, batch 415/500 completed (loss: 0.4697:  83%|████████▎ | 208/250 [21:00<04:15,  6.08s/it]
Training Epoch: 15/100, batch 415/500 completed (loss: 0.0137:  83%|████████▎ | 208/250 [21:00<04:15,  6.08s/it]
Training Epoch: 15/100, batch 415/500 completed (loss: 0.2228:  83%|████████▎ | 208/250 [21:00<04:15,  6.08s/it]
Training Epoch: 15/100, batch 415/500 completed (loss: 0.1051:  83%|████████▎ | 208/250 [21:00<04:15,  6.08s/it]
Training Epoch: 15/100, batch 416/500 completed (loss: 0.1787:  83%|████████▎ | 208/250 [21:03<04:15,  6.08s/it]
Training Epoch: 15/100, batch 416/500 completed (loss: 0.4076:  83%|████████▎ | 208/250 [21:03<04:15,  6.08s/it]
Training Epoch: 15/100, batch 416/500 completed (loss: 0.1814:  83%|████████▎ | 208/250 [21:03<04:15,  6.08s/it]
Training Epoch: 15/100, batch 416/500 completed (loss: 0.2273:  83%|████████▎ | 208/250 [21:03<04:15,  6.08s/it]
Training Epoch: 15/100, batch 416/500 completed (loss: 0.2273:  84%|████████▎ | 209/250 [21:06<04:12,  6.15s/it]
Training Epoch: 15/100, batch 416/500 completed (loss: 0.1787:  84%|████████▎ | 209/250 [21:06<04:12,  6.15s/it]
Training Epoch: 15/100, batch 416/500 completed (loss: 0.4076:  84%|████████▎ | 209/250 [21:06<04:12,  6.15s/it]
Training Epoch: 15/100, batch 416/500 completed (loss: 0.1814:  84%|████████▎ | 209/250 [21:06<04:12,  6.16s/it]
Training Epoch: 15/100, batch 417/500 completed (loss: 0.1284:  84%|████████▎ | 209/250 [21:07<04:12,  6.15s/it]
Training Epoch: 15/100, batch 417/500 completed (loss: 0.0039:  84%|████████▎ | 209/250 [21:07<04:12,  6.16s/it]
Training Epoch: 15/100, batch 417/500 completed (loss: 0.0495:  84%|████████▎ | 209/250 [21:07<04:12,  6.15s/it]
Training Epoch: 15/100, batch 417/500 completed (loss: 0.4546:  84%|████████▎ | 209/250 [21:07<04:12,  6.15s/it]
Training Epoch: 15/100, batch 418/500 completed (loss: 0.1898:  84%|████████▎ | 209/250 [21:09<04:12,  6.15s/it]
Training Epoch: 15/100, batch 418/500 completed (loss: 0.0164:  84%|████████▎ | 209/250 [21:09<04:12,  6.15s/it]
Training Epoch: 15/100, batch 418/500 completed (loss: 0.0198:  84%|████████▎ | 209/250 [21:09<04:12,  6.16s/it]
Training Epoch: 15/100, batch 418/500 completed (loss: 0.2127:  84%|████████▎ | 209/250 [21:09<04:12,  6.15s/it]
Training Epoch: 15/100, batch 418/500 completed (loss: 0.2127:  84%|████████▍ | 210/250 [21:12<04:02,  6.07s/it]
Training Epoch: 15/100, batch 418/500 completed (loss: 0.0198:  84%|████████▍ | 210/250 [21:12<04:02,  6.07s/it]
Training Epoch: 15/100, batch 418/500 completed (loss: 0.1898:  84%|████████▍ | 210/250 [21:12<04:02,  6.07s/it]
Training Epoch: 15/100, batch 418/500 completed (loss: 0.0164:  84%|████████▍ | 210/250 [21:12<04:03,  6.08s/it]
Training Epoch: 15/100, batch 419/500 completed (loss: 0.0111:  84%|████████▍ | 210/250 [21:12<04:02,  6.07s/it]
Training Epoch: 15/100, batch 419/500 completed (loss: 0.0435:  84%|████████▍ | 210/250 [21:12<04:02,  6.07s/it]
Training Epoch: 15/100, batch 419/500 completed (loss: 0.0391:  84%|████████▍ | 210/250 [21:12<04:03,  6.08s/it]
Training Epoch: 15/100, batch 419/500 completed (loss: 0.0232:  84%|████████▍ | 210/250 [21:12<04:02,  6.07s/it]
Training Epoch: 15/100, batch 420/500 completed (loss: 0.0553:  84%|████████▍ | 210/250 [21:15<04:02,  6.07s/it]
Training Epoch: 15/100, batch 420/500 completed (loss: 0.3071:  84%|████████▍ | 210/250 [21:15<04:03,  6.08s/it]
Training Epoch: 15/100, batch 420/500 completed (loss: 0.0432:  84%|████████▍ | 210/250 [21:15<04:02,  6.07s/it]
Training Epoch: 15/100, batch 420/500 completed (loss: 0.0114:  84%|████████▍ | 210/250 [21:16<04:02,  6.07s/it]
Training Epoch: 15/100, batch 420/500 completed (loss: 0.3071:  84%|████████▍ | 211/250 [21:19<03:58,  6.11s/it]
Training Epoch: 15/100, batch 420/500 completed (loss: 0.0432:  84%|████████▍ | 211/250 [21:19<03:58,  6.11s/it]
Training Epoch: 15/100, batch 420/500 completed (loss: 0.0553:  84%|████████▍ | 211/250 [21:19<03:58,  6.11s/it]
Training Epoch: 15/100, batch 420/500 completed (loss: 0.0114:  84%|████████▍ | 211/250 [21:19<03:58,  6.12s/it]
Training Epoch: 15/100, batch 421/500 completed (loss: 0.1283:  84%|████████▍ | 211/250 [21:19<03:58,  6.11s/it]
Training Epoch: 15/100, batch 421/500 completed (loss: 0.119:  84%|████████▍ | 211/250 [21:19<03:58,  6.11s/it] 
Training Epoch: 15/100, batch 421/500 completed (loss: 0.0211:  84%|████████▍ | 211/250 [21:19<03:58,  6.11s/it]
Training Epoch: 15/100, batch 421/500 completed (loss: 0.312:  84%|████████▍ | 211/250 [21:19<03:58,  6.12s/it] 
Training Epoch: 15/100, batch 422/500 completed (loss: 0.1637:  84%|████████▍ | 211/250 [21:22<03:58,  6.11s/it]
Training Epoch: 15/100, batch 422/500 completed (loss: 0.0166:  84%|████████▍ | 211/250 [21:22<03:58,  6.11s/it]
Training Epoch: 15/100, batch 422/500 completed (loss: 0.1167:  84%|████████▍ | 211/250 [21:22<03:58,  6.11s/it]
Training Epoch: 15/100, batch 422/500 completed (loss: 0.1509:  84%|████████▍ | 211/250 [21:22<03:58,  6.12s/it]
Training Epoch: 15/100, batch 422/500 completed (loss: 0.1637:  85%|████████▍ | 212/250 [21:25<03:52,  6.13s/it]
Training Epoch: 15/100, batch 422/500 completed (loss: 0.0166:  85%|████████▍ | 212/250 [21:25<03:52,  6.13s/it]
Training Epoch: 15/100, batch 422/500 completed (loss: 0.1509:  85%|████████▍ | 212/250 [21:25<03:52,  6.13s/it]
Training Epoch: 15/100, batch 422/500 completed (loss: 0.1167:  85%|████████▍ | 212/250 [21:25<03:52,  6.13s/it]
Training Epoch: 15/100, batch 423/500 completed (loss: 0.0075:  85%|████████▍ | 212/250 [21:25<03:52,  6.13s/it]
Training Epoch: 15/100, batch 423/500 completed (loss: 0.1959:  85%|████████▍ | 212/250 [21:25<03:52,  6.13s/it]
Training Epoch: 15/100, batch 423/500 completed (loss: 0.1662:  85%|████████▍ | 212/250 [21:25<03:52,  6.13s/it]
Training Epoch: 15/100, batch 423/500 completed (loss: 0.0233:  85%|████████▍ | 212/250 [21:25<03:52,  6.13s/it]
Training Epoch: 15/100, batch 424/500 completed (loss: 0.0896:  85%|████████▍ | 212/250 [21:28<03:52,  6.13s/it]
Training Epoch: 15/100, batch 424/500 completed (loss: 0.0985:  85%|████████▍ | 212/250 [21:28<03:52,  6.13s/it]
Training Epoch: 15/100, batch 424/500 completed (loss: 0.0179:  85%|████████▍ | 212/250 [21:28<03:52,  6.13s/it]
Training Epoch: 15/100, batch 424/500 completed (loss: 0.0243:  85%|████████▍ | 212/250 [21:28<03:52,  6.13s/it]
Training Epoch: 15/100, batch 424/500 completed (loss: 0.0985:  85%|████████▌ | 213/250 [21:31<03:43,  6.05s/it]
Training Epoch: 15/100, batch 424/500 completed (loss: 0.0896:  85%|████████▌ | 213/250 [21:31<03:43,  6.05s/it]
Training Epoch: 15/100, batch 424/500 completed (loss: 0.0179:  85%|████████▌ | 213/250 [21:31<03:43,  6.05s/it]
Training Epoch: 15/100, batch 424/500 completed (loss: 0.0243:  85%|████████▌ | 213/250 [21:31<03:43,  6.05s/it]
Training Epoch: 15/100, batch 425/500 completed (loss: 0.0506:  85%|████████▌ | 213/250 [21:31<03:43,  6.05s/it]
Training Epoch: 15/100, batch 425/500 completed (loss: 0.1772:  85%|████████▌ | 213/250 [21:31<03:43,  6.05s/it]
Training Epoch: 15/100, batch 425/500 completed (loss: 0.1554:  85%|████████▌ | 213/250 [21:31<03:43,  6.05s/it]
Training Epoch: 15/100, batch 425/500 completed (loss: 0.0554:  85%|████████▌ | 213/250 [21:31<03:43,  6.05s/it]
Training Epoch: 15/100, batch 426/500 completed (loss: 0.0602:  85%|████████▌ | 213/250 [21:34<03:43,  6.05s/it]
Training Epoch: 15/100, batch 426/500 completed (loss: 0.0431:  85%|████████▌ | 213/250 [21:34<03:43,  6.05s/it]
Training Epoch: 15/100, batch 426/500 completed (loss: 0.1792:  85%|████████▌ | 213/250 [21:34<03:43,  6.05s/it]
Training Epoch: 15/100, batch 426/500 completed (loss: 0.0514:  85%|████████▌ | 213/250 [21:34<03:43,  6.05s/it]
Training Epoch: 15/100, batch 426/500 completed (loss: 0.0514:  86%|████████▌ | 214/250 [21:37<03:40,  6.14s/it]
Training Epoch: 15/100, batch 426/500 completed (loss: 0.1792:  86%|████████▌ | 214/250 [21:37<03:41,  6.14s/it]
Training Epoch: 15/100, batch 426/500 completed (loss: 0.0431:  86%|████████▌ | 214/250 [21:37<03:41,  6.15s/it]
Training Epoch: 15/100, batch 426/500 completed (loss: 0.0602:  86%|████████▌ | 214/250 [21:37<03:41,  6.15s/it]
Training Epoch: 15/100, batch 427/500 completed (loss: 0.009:  86%|████████▌ | 214/250 [21:37<03:41,  6.15s/it] 
Training Epoch: 15/100, batch 427/500 completed (loss: 0.3241:  86%|████████▌ | 214/250 [21:37<03:41,  6.14s/it]
Training Epoch: 15/100, batch 427/500 completed (loss: 0.0242:  86%|████████▌ | 214/250 [21:37<03:41,  6.15s/it]
Training Epoch: 15/100, batch 427/500 completed (loss: 0.0125:  86%|████████▌ | 214/250 [21:37<03:40,  6.14s/it]
Training Epoch: 15/100, batch 428/500 completed (loss: 0.3714:  86%|████████▌ | 214/250 [21:40<03:41,  6.14s/it]
Training Epoch: 15/100, batch 428/500 completed (loss: 0.0276:  86%|████████▌ | 214/250 [21:40<03:41,  6.15s/it]
Training Epoch: 15/100, batch 428/500 completed (loss: 0.1095:  86%|████████▌ | 214/250 [21:40<03:41,  6.15s/it]
Training Epoch: 15/100, batch 428/500 completed (loss: 0.0392:  86%|████████▌ | 214/250 [21:40<03:40,  6.14s/it]
Training Epoch: 15/100, batch 428/500 completed (loss: 0.0276:  86%|████████▌ | 215/250 [21:43<03:33,  6.11s/it]
Training Epoch: 15/100, batch 428/500 completed (loss: 0.0392:  86%|████████▌ | 215/250 [21:43<03:33,  6.11s/it]
Training Epoch: 15/100, batch 428/500 completed (loss: 0.1095:  86%|████████▌ | 215/250 [21:43<03:33,  6.11s/it]
Training Epoch: 15/100, batch 428/500 completed (loss: 0.3714:  86%|████████▌ | 215/250 [21:43<03:34,  6.11s/it]
Training Epoch: 15/100, batch 429/500 completed (loss: 0.144:  86%|████████▌ | 215/250 [21:43<03:33,  6.11s/it] 
Training Epoch: 15/100, batch 429/500 completed (loss: 0.0141:  86%|████████▌ | 215/250 [21:43<03:33,  6.11s/it]
Training Epoch: 15/100, batch 429/500 completed (loss: 0.008:  86%|████████▌ | 215/250 [21:43<03:34,  6.11s/it] 
Training Epoch: 15/100, batch 429/500 completed (loss: 0.0849:  86%|████████▌ | 215/250 [21:43<03:33,  6.11s/it]
Training Epoch: 15/100, batch 430/500 completed (loss: 0.0868:  86%|████████▌ | 215/250 [21:46<03:34,  6.11s/it]
Training Epoch: 15/100, batch 430/500 completed (loss: 0.0802:  86%|████████▌ | 215/250 [21:46<03:33,  6.11s/it]
Training Epoch: 15/100, batch 430/500 completed (loss: 0.0996:  86%|████████▌ | 215/250 [21:46<03:33,  6.11s/it]
Training Epoch: 15/100, batch 430/500 completed (loss: 0.1625:  86%|████████▌ | 215/250 [21:46<03:33,  6.11s/it]
Training Epoch: 15/100, batch 430/500 completed (loss: 0.0802:  86%|████████▋ | 216/250 [21:49<03:25,  6.06s/it]
Training Epoch: 15/100, batch 430/500 completed (loss: 0.0868:  86%|████████▋ | 216/250 [21:49<03:25,  6.06s/it]
Training Epoch: 15/100, batch 430/500 completed (loss: 0.1625:  86%|████████▋ | 216/250 [21:49<03:26,  6.06s/it]
Training Epoch: 15/100, batch 430/500 completed (loss: 0.0996:  86%|████████▋ | 216/250 [21:49<03:26,  6.06s/it]
Training Epoch: 15/100, batch 431/500 completed (loss: 0.0768:  86%|████████▋ | 216/250 [21:49<03:25,  6.06s/it]
Training Epoch: 15/100, batch 431/500 completed (loss: 0.0138:  86%|████████▋ | 216/250 [21:49<03:26,  6.06s/it]
Training Epoch: 15/100, batch 431/500 completed (loss: 0.0652:  86%|████████▋ | 216/250 [21:49<03:25,  6.06s/it]
Training Epoch: 15/100, batch 431/500 completed (loss: 0.0779:  86%|████████▋ | 216/250 [21:49<03:26,  6.06s/it]
Training Epoch: 15/100, batch 432/500 completed (loss: 0.263:  86%|████████▋ | 216/250 [21:52<03:25,  6.06s/it] 
Training Epoch: 15/100, batch 432/500 completed (loss: 0.0172:  86%|████████▋ | 216/250 [21:52<03:26,  6.06s/it]
Training Epoch: 15/100, batch 432/500 completed (loss: 0.0165:  86%|████████▋ | 216/250 [21:52<03:25,  6.06s/it]
Training Epoch: 15/100, batch 432/500 completed (loss: 0.0191:  86%|████████▋ | 216/250 [21:52<03:26,  6.06s/it]
Training Epoch: 15/100, batch 432/500 completed (loss: 0.263:  87%|████████▋ | 217/250 [21:55<03:22,  6.14s/it]
Training Epoch: 15/100, batch 432/500 completed (loss: 0.0172:  87%|████████▋ | 217/250 [21:55<03:22,  6.14s/it]
Training Epoch: 15/100, batch 432/500 completed (loss: 0.0165:  87%|████████▋ | 217/250 [21:55<03:22,  6.14s/it]
Training Epoch: 15/100, batch 432/500 completed (loss: 0.0191:  87%|████████▋ | 217/250 [21:55<03:22,  6.14s/it]
Training Epoch: 15/100, batch 433/500 completed (loss: 0.0997:  87%|████████▋ | 217/250 [21:55<03:22,  6.14s/it]
Training Epoch: 15/100, batch 433/500 completed (loss: 0.1009:  87%|████████▋ | 217/250 [21:55<03:22,  6.14s/it]
Training Epoch: 15/100, batch 433/500 completed (loss: 0.0063:  87%|████████▋ | 217/250 [21:55<03:22,  6.14s/it]
Training Epoch: 15/100, batch 433/500 completed (loss: 0.0779:  87%|████████▋ | 217/250 [21:55<03:22,  6.14s/it]
Training Epoch: 15/100, batch 434/500 completed (loss: 0.1111:  87%|████████▋ | 217/250 [21:58<03:22,  6.14s/it]
Training Epoch: 15/100, batch 434/500 completed (loss: 0.1069:  87%|████████▋ | 217/250 [21:58<03:22,  6.14s/it]
Training Epoch: 15/100, batch 434/500 completed (loss: 0.0962:  87%|████████▋ | 217/250 [21:58<03:22,  6.14s/it]
Training Epoch: 15/100, batch 434/500 completed (loss: 0.1615:  87%|████████▋ | 217/250 [21:58<03:22,  6.14s/it]
Training Epoch: 15/100, batch 434/500 completed (loss: 0.0962:  87%|████████▋ | 218/250 [22:01<03:15,  6.09s/it]
Training Epoch: 15/100, batch 434/500 completed (loss: 0.1111:  87%|████████▋ | 218/250 [22:01<03:14,  6.09s/it]
Training Epoch: 15/100, batch 434/500 completed (loss: 0.1615:  87%|████████▋ | 218/250 [22:01<03:15,  6.09s/it]
Training Epoch: 15/100, batch 434/500 completed (loss: 0.1069:  87%|████████▋ | 218/250 [22:01<03:15,  6.10s/it]
Training Epoch: 15/100, batch 435/500 completed (loss: 0.0477:  87%|████████▋ | 218/250 [22:01<03:15,  6.09s/it]
Training Epoch: 15/100, batch 435/500 completed (loss: 0.0206:  87%|████████▋ | 218/250 [22:01<03:14,  6.09s/it]
Training Epoch: 15/100, batch 435/500 completed (loss: 0.0025:  87%|████████▋ | 218/250 [22:01<03:15,  6.10s/it]
Training Epoch: 15/100, batch 435/500 completed (loss: 0.0117:  87%|████████▋ | 218/250 [22:01<03:15,  6.09s/it]
Training Epoch: 15/100, batch 436/500 completed (loss: 0.1316:  87%|████████▋ | 218/250 [22:04<03:14,  6.09s/it]
Training Epoch: 15/100, batch 436/500 completed (loss: 0.1897:  87%|████████▋ | 218/250 [22:04<03:15,  6.10s/it]
Training Epoch: 15/100, batch 436/500 completed (loss: 0.109:  87%|████████▋ | 218/250 [22:04<03:15,  6.09s/it] 
Training Epoch: 15/100, batch 436/500 completed (loss: 0.0336:  87%|████████▋ | 218/250 [22:04<03:15,  6.09s/it]
Training Epoch: 15/100, batch 436/500 completed (loss: 0.0336:  88%|████████▊ | 219/250 [22:07<03:08,  6.07s/it]
Training Epoch: 15/100, batch 436/500 completed (loss: 0.109:  88%|████████▊ | 219/250 [22:07<03:08,  6.07s/it]
Training Epoch: 15/100, batch 436/500 completed (loss: 0.1316:  88%|████████▊ | 219/250 [22:07<03:08,  6.07s/it]
Training Epoch: 15/100, batch 436/500 completed (loss: 0.1897:  88%|████████▊ | 219/250 [22:07<03:08,  6.08s/it]
Training Epoch: 15/100, batch 437/500 completed (loss: 0.0823:  88%|████████▊ | 219/250 [22:07<03:08,  6.08s/it]
Training Epoch: 15/100, batch 437/500 completed (loss: 0.0581:  88%|████████▊ | 219/250 [22:07<03:08,  6.07s/it]
Training Epoch: 15/100, batch 437/500 completed (loss: 0.0248:  88%|████████▊ | 219/250 [22:07<03:08,  6.07s/it]
Training Epoch: 15/100, batch 437/500 completed (loss: 0.1958:  88%|████████▊ | 219/250 [22:07<03:08,  6.07s/it]
Training Epoch: 15/100, batch 438/500 completed (loss: 0.1696:  88%|████████▊ | 219/250 [22:11<03:08,  6.08s/it]
Training Epoch: 15/100, batch 438/500 completed (loss: 0.0762:  88%|████████▊ | 219/250 [22:11<03:08,  6.07s/it]
Training Epoch: 15/100, batch 438/500 completed (loss: 0.2249:  88%|████████▊ | 219/250 [22:11<03:08,  6.07s/it]
Training Epoch: 15/100, batch 438/500 completed (loss: 0.6571:  88%|████████▊ | 219/250 [22:11<03:08,  6.07s/it]
Training Epoch: 15/100, batch 438/500 completed (loss: 0.0762:  88%|████████▊ | 220/250 [22:13<03:03,  6.11s/it]
Training Epoch: 15/100, batch 438/500 completed (loss: 0.2249:  88%|████████▊ | 220/250 [22:13<03:03,  6.11s/it]
Training Epoch: 15/100, batch 438/500 completed (loss: 0.1696:  88%|████████▊ | 220/250 [22:13<03:03,  6.11s/it]
Training Epoch: 15/100, batch 438/500 completed (loss: 0.6571:  88%|████████▊ | 220/250 [22:13<03:03,  6.11s/it]
Training Epoch: 15/100, batch 439/500 completed (loss: 0.1647:  88%|████████▊ | 220/250 [22:13<03:03,  6.11s/it]
Training Epoch: 15/100, batch 439/500 completed (loss: 0.0481:  88%|████████▊ | 220/250 [22:13<03:03,  6.11s/it]
Training Epoch: 15/100, batch 439/500 completed (loss: 0.0084:  88%|████████▊ | 220/250 [22:13<03:03,  6.11s/it]
Training Epoch: 15/100, batch 439/500 completed (loss: 0.039:  88%|████████▊ | 220/250 [22:14<03:03,  6.11s/it] 
Training Epoch: 15/100, batch 440/500 completed (loss: 0.4886:  88%|████████▊ | 220/250 [22:16<03:03,  6.11s/it]
Training Epoch: 15/100, batch 440/500 completed (loss: 0.0189:  88%|████████▊ | 220/250 [22:16<03:03,  6.11s/it]
Training Epoch: 15/100, batch 440/500 completed (loss: 0.0664:  88%|████████▊ | 220/250 [22:16<03:03,  6.11s/it]
Training Epoch: 15/100, batch 440/500 completed (loss: 0.4817:  88%|████████▊ | 220/250 [22:16<03:03,  6.11s/it]
Training Epoch: 15/100, batch 440/500 completed (loss: 0.4817:  88%|████████▊ | 221/250 [22:20<02:56,  6.08s/it]
Training Epoch: 15/100, batch 440/500 completed (loss: 0.4886:  88%|████████▊ | 221/250 [22:19<02:56,  6.09s/it]
Training Epoch: 15/100, batch 440/500 completed (loss: 0.0189:  88%|████████▊ | 221/250 [22:19<02:56,  6.09s/it]
Training Epoch: 15/100, batch 440/500 completed (loss: 0.0664:  88%|████████▊ | 221/250 [22:19<02:56,  6.09s/it]
Training Epoch: 15/100, batch 441/500 completed (loss: 0.112:  88%|████████▊ | 221/250 [22:20<02:56,  6.09s/it] 
Training Epoch: 15/100, batch 441/500 completed (loss: 0.0237:  88%|████████▊ | 221/250 [22:20<02:56,  6.09s/it]
Training Epoch: 15/100, batch 441/500 completed (loss: 0.0175:  88%|████████▊ | 221/250 [22:20<02:56,  6.09s/it]
Training Epoch: 15/100, batch 441/500 completed (loss: 0.1215:  88%|████████▊ | 221/250 [22:20<02:56,  6.08s/it]
Training Epoch: 15/100, batch 442/500 completed (loss: 0.0394:  88%|████████▊ | 221/250 [22:23<02:56,  6.09s/it]
Training Epoch: 15/100, batch 442/500 completed (loss: 0.0053:  88%|████████▊ | 221/250 [22:23<02:56,  6.09s/it]
Training Epoch: 15/100, batch 442/500 completed (loss: 0.076:  88%|████████▊ | 221/250 [22:23<02:56,  6.09s/it] 
Training Epoch: 15/100, batch 442/500 completed (loss: 0.1444:  88%|████████▊ | 221/250 [22:23<02:56,  6.08s/it]
Training Epoch: 15/100, batch 442/500 completed (loss: 0.076:  89%|████████▉ | 222/250 [22:26<02:51,  6.11s/it]
Training Epoch: 15/100, batch 442/500 completed (loss: 0.0053:  89%|████████▉ | 222/250 [22:26<02:51,  6.11s/it]
Training Epoch: 15/100, batch 442/500 completed (loss: 0.0394:  89%|████████▉ | 222/250 [22:26<02:51,  6.11s/it]
Training Epoch: 15/100, batch 442/500 completed (loss: 0.1444:  89%|████████▉ | 222/250 [22:26<02:51,  6.12s/it]
Training Epoch: 15/100, batch 443/500 completed (loss: 0.0117:  89%|████████▉ | 222/250 [22:26<02:51,  6.11s/it]
Training Epoch: 15/100, batch 443/500 completed (loss: 0.095:  89%|████████▉ | 222/250 [22:26<02:51,  6.11s/it] 
Training Epoch: 15/100, batch 443/500 completed (loss: 0.0459:  89%|████████▉ | 222/250 [22:26<02:51,  6.11s/it]
Training Epoch: 15/100, batch 443/500 completed (loss: 0.0168:  89%|████████▉ | 222/250 [22:26<02:51,  6.12s/it]
Training Epoch: 15/100, batch 444/500 completed (loss: 0.0513:  89%|████████▉ | 222/250 [22:29<02:51,  6.11s/it]
Training Epoch: 15/100, batch 444/500 completed (loss: 0.0934:  89%|████████▉ | 222/250 [22:29<02:51,  6.11s/it]
Training Epoch: 15/100, batch 444/500 completed (loss: 0.3409:  89%|████████▉ | 222/250 [22:29<02:51,  6.11s/it]
Training Epoch: 15/100, batch 444/500 completed (loss: 0.1868:  89%|████████▉ | 222/250 [22:29<02:51,  6.12s/it]
Training Epoch: 15/100, batch 444/500 completed (loss: 0.1868:  89%|████████▉ | 223/250 [22:32<02:44,  6.08s/it]
Training Epoch: 15/100, batch 444/500 completed (loss: 0.3409:  89%|████████▉ | 223/250 [22:32<02:44,  6.08s/it]
Training Epoch: 15/100, batch 444/500 completed (loss: 0.0934:  89%|████████▉ | 223/250 [22:32<02:44,  6.08s/it]
Training Epoch: 15/100, batch 444/500 completed (loss: 0.0513:  89%|████████▉ | 223/250 [22:32<02:44,  6.08s/it]
Training Epoch: 15/100, batch 445/500 completed (loss: 0.0356:  89%|████████▉ | 223/250 [22:32<02:44,  6.08s/it]
Training Epoch: 15/100, batch 445/500 completed (loss: 0.1734:  89%|████████▉ | 223/250 [22:32<02:44,  6.08s/it]
Training Epoch: 15/100, batch 445/500 completed (loss: 0.0547:  89%|████████▉ | 223/250 [22:32<02:44,  6.08s/it]
Training Epoch: 15/100, batch 445/500 completed (loss: 0.1284:  89%|████████▉ | 223/250 [22:32<02:44,  6.08s/it]
Training Epoch: 15/100, batch 446/500 completed (loss: 0.0084:  89%|████████▉ | 223/250 [22:35<02:44,  6.08s/it]
Training Epoch: 15/100, batch 446/500 completed (loss: 0.2965:  89%|████████▉ | 223/250 [22:35<02:44,  6.08s/it]
Training Epoch: 15/100, batch 446/500 completed (loss: 0.0297:  89%|████████▉ | 223/250 [22:35<02:44,  6.08s/it]
Training Epoch: 15/100, batch 446/500 completed (loss: 0.0065:  89%|████████▉ | 223/250 [22:35<02:44,  6.08s/it]
Training Epoch: 15/100, batch 446/500 completed (loss: 0.2965:  90%|████████▉ | 224/250 [22:38<02:37,  6.05s/it]
Training Epoch: 15/100, batch 446/500 completed (loss: 0.0065:  90%|████████▉ | 224/250 [22:38<02:37,  6.06s/it]
Training Epoch: 15/100, batch 446/500 completed (loss: 0.0084:  90%|████████▉ | 224/250 [22:38<02:37,  6.06s/it]
Training Epoch: 15/100, batch 446/500 completed (loss: 0.0297:  90%|████████▉ | 224/250 [22:38<02:37,  6.06s/it]
Training Epoch: 15/100, batch 447/500 completed (loss: 0.0913:  90%|████████▉ | 224/250 [22:38<02:37,  6.06s/it]
Training Epoch: 15/100, batch 447/500 completed (loss: 0.064:  90%|████████▉ | 224/250 [22:38<02:37,  6.06s/it] 
Training Epoch: 15/100, batch 447/500 completed (loss: 0.0149:  90%|████████▉ | 224/250 [22:38<02:37,  6.05s/it]
Training Epoch: 15/100, batch 447/500 completed (loss: 0.0496:  90%|████████▉ | 224/250 [22:38<02:37,  6.06s/it]
Training Epoch: 15/100, batch 448/500 completed (loss: 0.0304:  90%|████████▉ | 224/250 [22:41<02:37,  6.06s/it]
Training Epoch: 15/100, batch 448/500 completed (loss: 0.0221:  90%|████████▉ | 224/250 [22:41<02:37,  6.05s/it]
Training Epoch: 15/100, batch 448/500 completed (loss: 0.0835:  90%|████████▉ | 224/250 [22:41<02:37,  6.06s/it]
Training Epoch: 15/100, batch 448/500 completed (loss: 0.0788:  90%|████████▉ | 224/250 [22:41<02:37,  6.06s/it]
Training Epoch: 15/100, batch 448/500 completed (loss: 0.0835:  90%|█████████ | 225/250 [22:44<02:33,  6.13s/it]
Training Epoch: 15/100, batch 448/500 completed (loss: 0.0304:  90%|█████████ | 225/250 [22:44<02:33,  6.13s/it]
Training Epoch: 15/100, batch 448/500 completed (loss: 0.0221:  90%|█████████ | 225/250 [22:44<02:33,  6.13s/it]
Training Epoch: 15/100, batch 448/500 completed (loss: 0.0788:  90%|█████████ | 225/250 [22:44<02:33,  6.13s/it]
Training Epoch: 15/100, batch 449/500 completed (loss: 0.0189:  90%|█████████ | 225/250 [22:44<02:33,  6.13s/it]
Training Epoch: 15/100, batch 449/500 completed (loss: 0.0084:  90%|█████████ | 225/250 [22:44<02:33,  6.13s/it]
Training Epoch: 15/100, batch 449/500 completed (loss: 0.3852:  90%|█████████ | 225/250 [22:44<02:33,  6.13s/it]
Training Epoch: 15/100, batch 449/500 completed (loss: 0.0194:  90%|█████████ | 225/250 [22:44<02:33,  6.13s/it]
Training Epoch: 15/100, batch 450/500 completed (loss: 0.0209:  90%|█████████ | 225/250 [22:47<02:33,  6.13s/it]
Training Epoch: 15/100, batch 450/500 completed (loss: 0.0546:  90%|█████████ | 225/250 [22:47<02:33,  6.13s/it]
Training Epoch: 15/100, batch 450/500 completed (loss: 0.0016:  90%|█████████ | 225/250 [22:47<02:33,  6.13s/it]
Training Epoch: 15/100, batch 450/500 completed (loss: 0.2076:  90%|█████████ | 225/250 [22:47<02:33,  6.13s/it]
Training Epoch: 15/100, batch 450/500 completed (loss: 0.2076:  90%|█████████ | 226/250 [22:50<02:26,  6.09s/it]
Training Epoch: 15/100, batch 450/500 completed (loss: 0.0209:  90%|█████████ | 226/250 [22:50<02:26,  6.09s/it]
Training Epoch: 15/100, batch 450/500 completed (loss: 0.0016:  90%|█████████ | 226/250 [22:50<02:26,  6.09s/it]
Training Epoch: 15/100, batch 450/500 completed (loss: 0.0546:  90%|█████████ | 226/250 [22:50<02:26,  6.09s/it]
Training Epoch: 15/100, batch 451/500 completed (loss: 0.0694:  90%|█████████ | 226/250 [22:50<02:26,  6.09s/it]
Training Epoch: 15/100, batch 451/500 completed (loss: 0.1791:  90%|█████████ | 226/250 [22:50<02:26,  6.09s/it]
Training Epoch: 15/100, batch 451/500 completed (loss: 0.2192:  90%|█████████ | 226/250 [22:50<02:26,  6.09s/it]
Training Epoch: 15/100, batch 451/500 completed (loss: 0.0136:  90%|█████████ | 226/250 [22:50<02:26,  6.09s/it]
Training Epoch: 15/100, batch 452/500 completed (loss: 0.0166:  90%|█████████ | 226/250 [22:53<02:26,  6.09s/it]
Training Epoch: 15/100, batch 452/500 completed (loss: 0.0677:  90%|█████████ | 226/250 [22:53<02:26,  6.09s/it]
Training Epoch: 15/100, batch 452/500 completed (loss: 0.0374:  90%|█████████ | 226/250 [22:53<02:26,  6.09s/it]
Training Epoch: 15/100, batch 452/500 completed (loss: 0.0289:  90%|█████████ | 226/250 [22:53<02:26,  6.09s/it]
Training Epoch: 15/100, batch 452/500 completed (loss: 0.0289:  91%|█████████ | 227/250 [22:56<02:20,  6.10s/it]
Training Epoch: 15/100, batch 452/500 completed (loss: 0.0374:  91%|█████████ | 227/250 [22:56<02:20,  6.11s/it]
Training Epoch: 15/100, batch 452/500 completed (loss: 0.0677:  91%|█████████ | 227/250 [22:56<02:20,  6.11s/it]
Training Epoch: 15/100, batch 452/500 completed (loss: 0.0166:  91%|█████████ | 227/250 [22:56<02:20,  6.12s/it]
Training Epoch: 15/100, batch 453/500 completed (loss: 0.0328:  91%|█████████ | 227/250 [22:56<02:20,  6.11s/it]
Training Epoch: 15/100, batch 453/500 completed (loss: 0.5041:  91%|█████████ | 227/250 [22:56<02:20,  6.12s/it]
Training Epoch: 15/100, batch 453/500 completed (loss: 0.011:  91%|█████████ | 227/250 [22:56<02:20,  6.11s/it] 
Training Epoch: 15/100, batch 453/500 completed (loss: 0.027:  91%|█████████ | 227/250 [22:56<02:20,  6.10s/it] 
Training Epoch: 15/100, batch 454/500 completed (loss: 0.0892:  91%|█████████ | 227/250 [22:59<02:20,  6.11s/it]
Training Epoch: 15/100, batch 454/500 completed (loss: 0.0605:  91%|█████████ | 227/250 [22:59<02:20,  6.11s/it]
Training Epoch: 15/100, batch 454/500 completed (loss: 0.0118:  91%|█████████ | 227/250 [22:59<02:20,  6.12s/it]
Training Epoch: 15/100, batch 454/500 completed (loss: 0.3875:  91%|█████████ | 227/250 [22:59<02:20,  6.10s/it]
Training Epoch: 15/100, batch 454/500 completed (loss: 0.3875:  91%|█████████ | 228/250 [23:02<02:13,  6.09s/it]
Training Epoch: 15/100, batch 454/500 completed (loss: 0.0892:  91%|█████████ | 228/250 [23:02<02:13,  6.09s/it]
Training Epoch: 15/100, batch 454/500 completed (loss: 0.0118:  91%|█████████ | 228/250 [23:02<02:13,  6.09s/it]
Training Epoch: 15/100, batch 454/500 completed (loss: 0.0605:  91%|█████████ | 228/250 [23:02<02:14,  6.09s/it]
Training Epoch: 15/100, batch 455/500 completed (loss: 0.0905:  91%|█████████ | 228/250 [23:02<02:13,  6.09s/it]
Training Epoch: 15/100, batch 455/500 completed (loss: 0.0344:  91%|█████████ | 228/250 [23:02<02:14,  6.09s/it]
Training Epoch: 15/100, batch 455/500 completed (loss: 0.0702:  91%|█████████ | 228/250 [23:02<02:13,  6.09s/it]
Training Epoch: 15/100, batch 455/500 completed (loss: 0.0061:  91%|█████████ | 228/250 [23:02<02:13,  6.09s/it]
Training Epoch: 15/100, batch 456/500 completed (loss: 0.1652:  91%|█████████ | 228/250 [23:05<02:14,  6.09s/it]
Training Epoch: 15/100, batch 456/500 completed (loss: 0.0204:  91%|█████████ | 228/250 [23:05<02:13,  6.09s/it]
Training Epoch: 15/100, batch 456/500 completed (loss: 0.057:  91%|█████████ | 228/250 [23:05<02:13,  6.09s/it] 
Training Epoch: 15/100, batch 456/500 completed (loss: 0.0421:  91%|█████████ | 228/250 [23:05<02:13,  6.09s/it]
Training Epoch: 15/100, batch 456/500 completed (loss: 0.057:  92%|█████████▏| 229/250 [23:08<02:07,  6.06s/it]
Training Epoch: 15/100, batch 456/500 completed (loss: 0.0421:  92%|█████████▏| 229/250 [23:08<02:07,  6.06s/it]
Training Epoch: 15/100, batch 456/500 completed (loss: 0.0204:  92%|█████████▏| 229/250 [23:08<02:07,  6.06s/it]
Training Epoch: 15/100, batch 456/500 completed (loss: 0.1652:  92%|█████████▏| 229/250 [23:08<02:07,  6.06s/it]
Training Epoch: 15/100, batch 457/500 completed (loss: 0.0639:  92%|█████████▏| 229/250 [23:08<02:07,  6.06s/it]
Training Epoch: 15/100, batch 457/500 completed (loss: 0.036:  92%|█████████▏| 229/250 [23:08<02:07,  6.06s/it] 
Training Epoch: 15/100, batch 457/500 completed (loss: 0.3413:  92%|█████████▏| 229/250 [23:08<02:07,  6.06s/it]
Training Epoch: 15/100, batch 457/500 completed (loss: 0.0524:  92%|█████████▏| 229/250 [23:08<02:07,  6.06s/it]
Training Epoch: 15/100, batch 458/500 completed (loss: 0.005:  92%|█████████▏| 229/250 [23:11<02:07,  6.06s/it] 
Training Epoch: 15/100, batch 458/500 completed (loss: 0.1732:  92%|█████████▏| 229/250 [23:11<02:07,  6.06s/it]
Training Epoch: 15/100, batch 458/500 completed (loss: 0.0124:  92%|█████████▏| 229/250 [23:11<02:07,  6.06s/it]
Training Epoch: 15/100, batch 458/500 completed (loss: 0.2191:  92%|█████████▏| 229/250 [23:11<02:07,  6.06s/it]
Training Epoch: 15/100, batch 458/500 completed (loss: 0.1732:  92%|█████████▏| 230/250 [23:14<02:02,  6.14s/it]
Training Epoch: 15/100, batch 458/500 completed (loss: 0.2191:  92%|█████████▏| 230/250 [23:15<02:02,  6.14s/it]
Training Epoch: 15/100, batch 458/500 completed (loss: 0.005:  92%|█████████▏| 230/250 [23:14<02:02,  6.14s/it]
Training Epoch: 15/100, batch 458/500 completed (loss: 0.0124:  92%|█████████▏| 230/250 [23:15<02:02,  6.15s/it]
Training Epoch: 15/100, batch 459/500 completed (loss: 0.0669:  92%|█████████▏| 230/250 [23:15<02:02,  6.14s/it]
Training Epoch: 15/100, batch 459/500 completed (loss: 0.0291:  92%|█████████▏| 230/250 [23:15<02:02,  6.15s/it]
Training Epoch: 15/100, batch 459/500 completed (loss: 0.0655:  92%|█████████▏| 230/250 [23:15<02:02,  6.14s/it]
Training Epoch: 15/100, batch 459/500 completed (loss: 0.1595:  92%|█████████▏| 230/250 [23:15<02:02,  6.14s/it]
Training Epoch: 15/100, batch 460/500 completed (loss: 0.3023:  92%|█████████▏| 230/250 [23:17<02:02,  6.14s/it]
Training Epoch: 15/100, batch 460/500 completed (loss: 0.0749:  92%|█████████▏| 230/250 [23:17<02:02,  6.14s/it]
Training Epoch: 15/100, batch 460/500 completed (loss: 0.2195:  92%|█████████▏| 230/250 [23:17<02:02,  6.15s/it]
Training Epoch: 15/100, batch 460/500 completed (loss: 0.1808:  92%|█████████▏| 230/250 [23:17<02:02,  6.14s/it]
Training Epoch: 15/100, batch 460/500 completed (loss: 0.1808:  92%|█████████▏| 231/250 [23:21<01:55,  6.09s/it]
Training Epoch: 15/100, batch 460/500 completed (loss: 0.0749:  92%|█████████▏| 231/250 [23:20<01:55,  6.09s/it]
Training Epoch: 15/100, batch 460/500 completed (loss: 0.2195:  92%|█████████▏| 231/250 [23:20<01:55,  6.09s/it]
Training Epoch: 15/100, batch 460/500 completed (loss: 0.3023:  92%|█████████▏| 231/250 [23:20<01:55,  6.09s/it]
Training Epoch: 15/100, batch 461/500 completed (loss: 0.0249:  92%|█████████▏| 231/250 [23:21<01:55,  6.09s/it]
Training Epoch: 15/100, batch 461/500 completed (loss: 0.0761:  92%|█████████▏| 231/250 [23:21<01:55,  6.09s/it]
Training Epoch: 15/100, batch 461/500 completed (loss: 0.0414:  92%|█████████▏| 231/250 [23:21<01:55,  6.09s/it]
Training Epoch: 15/100, batch 461/500 completed (loss: 0.0185:  92%|█████████▏| 231/250 [23:21<01:55,  6.09s/it]
Training Epoch: 15/100, batch 462/500 completed (loss: 0.2093:  92%|█████████▏| 231/250 [23:23<01:55,  6.09s/it]
Training Epoch: 15/100, batch 462/500 completed (loss: 0.4551:  92%|█████████▏| 231/250 [23:23<01:55,  6.09s/it]
Training Epoch: 15/100, batch 462/500 completed (loss: 0.023:  92%|█████████▏| 231/250 [23:23<01:55,  6.09s/it] 
Training Epoch: 15/100, batch 462/500 completed (loss: 0.0218:  92%|█████████▏| 231/250 [23:24<01:55,  6.09s/it]
Training Epoch: 15/100, batch 462/500 completed (loss: 0.0218:  93%|█████████▎| 232/250 [23:27<01:49,  6.07s/it]
Training Epoch: 15/100, batch 462/500 completed (loss: 0.023:  93%|█████████▎| 232/250 [23:26<01:49,  6.07s/it]
Training Epoch: 15/100, batch 462/500 completed (loss: 0.4551:  93%|█████████▎| 232/250 [23:27<01:49,  6.07s/it]
Training Epoch: 15/100, batch 462/500 completed (loss: 0.2093:  93%|█████████▎| 232/250 [23:26<01:49,  6.08s/it]
Training Epoch: 15/100, batch 463/500 completed (loss: 0.0737:  93%|█████████▎| 232/250 [23:27<01:49,  6.07s/it]
Training Epoch: 15/100, batch 463/500 completed (loss: 0.1177:  93%|█████████▎| 232/250 [23:27<01:49,  6.07s/it]
Training Epoch: 15/100, batch 463/500 completed (loss: 0.0722:  93%|█████████▎| 232/250 [23:27<01:49,  6.08s/it]
Training Epoch: 15/100, batch 463/500 completed (loss: 0.395:  93%|█████████▎| 232/250 [23:27<01:49,  6.07s/it] 
Training Epoch: 15/100, batch 464/500 completed (loss: 0.0061:  93%|█████████▎| 232/250 [23:30<01:49,  6.08s/it]
Training Epoch: 15/100, batch 464/500 completed (loss: 0.0155:  93%|█████████▎| 232/250 [23:30<01:49,  6.07s/it]
Training Epoch: 15/100, batch 464/500 completed (loss: 0.0305:  93%|█████████▎| 232/250 [23:30<01:49,  6.07s/it]
Training Epoch: 15/100, batch 464/500 completed (loss: 0.1205:  93%|█████████▎| 232/250 [23:30<01:49,  6.07s/it]
Training Epoch: 15/100, batch 464/500 completed (loss: 0.1205:  93%|█████████▎| 233/250 [23:33<01:43,  6.09s/it]
Training Epoch: 15/100, batch 464/500 completed (loss: 0.0061:  93%|█████████▎| 233/250 [23:33<01:43,  6.09s/it]
Training Epoch: 15/100, batch 464/500 completed (loss: 0.0305:  93%|█████████▎| 233/250 [23:33<01:43,  6.10s/it]
Training Epoch: 15/100, batch 464/500 completed (loss: 0.0155:  93%|█████████▎| 233/250 [23:33<01:43,  6.10s/it]
Training Epoch: 15/100, batch 465/500 completed (loss: 0.0312:  93%|█████████▎| 233/250 [23:33<01:43,  6.10s/it]
Training Epoch: 15/100, batch 465/500 completed (loss: 0.008:  93%|█████████▎| 233/250 [23:33<01:43,  6.09s/it] 
Training Epoch: 15/100, batch 465/500 completed (loss: 0.0258:  93%|█████████▎| 233/250 [23:33<01:43,  6.10s/it]
Training Epoch: 15/100, batch 465/500 completed (loss: 0.0423:  93%|█████████▎| 233/250 [23:33<01:43,  6.09s/it]
Training Epoch: 15/100, batch 466/500 completed (loss: 0.2395:  93%|█████████▎| 233/250 [23:36<01:43,  6.10s/it]
Training Epoch: 15/100, batch 466/500 completed (loss: 0.0568:  93%|█████████▎| 233/250 [23:36<01:43,  6.09s/it]
Training Epoch: 15/100, batch 466/500 completed (loss: 0.0105:  93%|█████████▎| 233/250 [23:36<01:43,  6.10s/it]
Training Epoch: 15/100, batch 466/500 completed (loss: 0.0699:  93%|█████████▎| 233/250 [23:36<01:43,  6.09s/it]
Training Epoch: 15/100, batch 466/500 completed (loss: 0.0699:  94%|█████████▎| 234/250 [23:39<01:37,  6.07s/it]
Training Epoch: 15/100, batch 466/500 completed (loss: 0.2395:  94%|█████████▎| 234/250 [23:39<01:37,  6.07s/it]
Training Epoch: 15/100, batch 466/500 completed (loss: 0.0105:  94%|█████████▎| 234/250 [23:39<01:37,  6.07s/it]
Training Epoch: 15/100, batch 466/500 completed (loss: 0.0568:  94%|█████████▎| 234/250 [23:39<01:37,  6.07s/it]
Training Epoch: 15/100, batch 467/500 completed (loss: 0.04:  94%|█████████▎| 234/250 [23:39<01:37,  6.07s/it]  
Training Epoch: 15/100, batch 467/500 completed (loss: 0.1351:  94%|█████████▎| 234/250 [23:39<01:37,  6.07s/it]
Training Epoch: 15/100, batch 467/500 completed (loss: 0.0416:  94%|█████████▎| 234/250 [23:39<01:37,  6.07s/it]
Training Epoch: 15/100, batch 467/500 completed (loss: 0.015:  94%|█████████▎| 234/250 [23:39<01:37,  6.07s/it] 
Training Epoch: 15/100, batch 468/500 completed (loss: 0.0303:  94%|█████████▎| 234/250 [23:42<01:37,  6.07s/it]
Training Epoch: 15/100, batch 468/500 completed (loss: 0.0515:  94%|█████████▎| 234/250 [23:42<01:37,  6.07s/it]
Training Epoch: 15/100, batch 468/500 completed (loss: 0.0383:  94%|█████████▎| 234/250 [23:42<01:37,  6.07s/it]
Training Epoch: 15/100, batch 468/500 completed (loss: 0.2201:  94%|█████████▎| 234/250 [23:42<01:37,  6.07s/it]
Training Epoch: 15/100, batch 468/500 completed (loss: 0.2201:  94%|█████████▍| 235/250 [23:45<01:31,  6.10s/it]
Training Epoch: 15/100, batch 468/500 completed (loss: 0.0383:  94%|█████████▍| 235/250 [23:45<01:31,  6.10s/it]
Training Epoch: 15/100, batch 468/500 completed (loss: 0.0303:  94%|█████████▍| 235/250 [23:45<01:31,  6.10s/it]
Training Epoch: 15/100, batch 468/500 completed (loss: 0.0515:  94%|█████████▍| 235/250 [23:45<01:31,  6.10s/it]
Training Epoch: 15/100, batch 469/500 completed (loss: 0.1903:  94%|█████████▍| 235/250 [23:45<01:31,  6.10s/it]
Training Epoch: 15/100, batch 469/500 completed (loss: 0.0128:  94%|█████████▍| 235/250 [23:45<01:31,  6.10s/it]
Training Epoch: 15/100, batch 469/500 completed (loss: 0.0029:  94%|█████████▍| 235/250 [23:45<01:31,  6.10s/it]
Training Epoch: 15/100, batch 469/500 completed (loss: 0.0229:  94%|█████████▍| 235/250 [23:45<01:31,  6.10s/it]
Training Epoch: 15/100, batch 470/500 completed (loss: 0.0502:  94%|█████████▍| 235/250 [23:48<01:31,  6.10s/it]
Training Epoch: 15/100, batch 470/500 completed (loss: 0.0627:  94%|█████████▍| 235/250 [23:48<01:31,  6.10s/it]
Training Epoch: 15/100, batch 470/500 completed (loss: 0.5215:  94%|█████████▍| 235/250 [23:48<01:31,  6.10s/it]
Training Epoch: 15/100, batch 470/500 completed (loss: 0.0496:  94%|█████████▍| 235/250 [23:48<01:31,  6.10s/it]
Training Epoch: 15/100, batch 470/500 completed (loss: 0.0627:  94%|█████████▍| 236/250 [23:51<01:25,  6.08s/it]
Training Epoch: 15/100, batch 470/500 completed (loss: 0.5215:  94%|█████████▍| 236/250 [23:51<01:25,  6.09s/it]
Training Epoch: 15/100, batch 470/500 completed (loss: 0.0502:  94%|█████████▍| 236/250 [23:51<01:25,  6.09s/it]
Training Epoch: 15/100, batch 470/500 completed (loss: 0.0496:  94%|█████████▍| 236/250 [23:51<01:25,  6.09s/it]
Training Epoch: 15/100, batch 471/500 completed (loss: 0.0275:  94%|█████████▍| 236/250 [23:51<01:25,  6.09s/it]
Training Epoch: 15/100, batch 471/500 completed (loss: 0.1112:  94%|█████████▍| 236/250 [23:51<01:25,  6.08s/it]
Training Epoch: 15/100, batch 471/500 completed (loss: 0.0463:  94%|█████████▍| 236/250 [23:51<01:25,  6.09s/it]
Training Epoch: 15/100, batch 471/500 completed (loss: 0.0728:  94%|█████████▍| 236/250 [23:51<01:25,  6.09s/it]
Training Epoch: 15/100, batch 472/500 completed (loss: 0.016:  94%|█████████▍| 236/250 [23:54<01:25,  6.08s/it] 
Training Epoch: 15/100, batch 472/500 completed (loss: 0.02:  94%|█████████▍| 236/250 [23:54<01:25,  6.09s/it]  
Training Epoch: 15/100, batch 472/500 completed (loss: 0.1869:  94%|█████████▍| 236/250 [23:54<01:25,  6.09s/it]
Training Epoch: 15/100, batch 472/500 completed (loss: 0.0904:  94%|█████████▍| 236/250 [23:54<01:25,  6.09s/it]
Training Epoch: 15/100, batch 472/500 completed (loss: 0.016:  95%|█████████▍| 237/250 [23:57<01:18,  6.06s/it]
Training Epoch: 15/100, batch 472/500 completed (loss: 0.1869:  95%|█████████▍| 237/250 [23:57<01:18,  6.06s/it]
Training Epoch: 15/100, batch 472/500 completed (loss: 0.02:  95%|█████████▍| 237/250 [23:57<01:18,  6.06s/it]
Training Epoch: 15/100, batch 472/500 completed (loss: 0.0904:  95%|█████████▍| 237/250 [23:57<01:18,  6.07s/it]
Training Epoch: 15/100, batch 473/500 completed (loss: 0.0323:  95%|█████████▍| 237/250 [23:57<01:18,  6.06s/it]
Training Epoch: 15/100, batch 473/500 completed (loss: 0.0733:  95%|█████████▍| 237/250 [23:57<01:18,  6.06s/it]
Training Epoch: 15/100, batch 473/500 completed (loss: 0.0094:  95%|█████████▍| 237/250 [23:57<01:18,  6.06s/it]
Training Epoch: 15/100, batch 473/500 completed (loss: 0.0417:  95%|█████████▍| 237/250 [23:57<01:18,  6.07s/it]
Training Epoch: 15/100, batch 474/500 completed (loss: 0.008:  95%|█████████▍| 237/250 [24:00<01:18,  6.06s/it] 
Training Epoch: 15/100, batch 474/500 completed (loss: 0.0538:  95%|█████████▍| 237/250 [24:00<01:18,  6.06s/it]
Training Epoch: 15/100, batch 474/500 completed (loss: 0.0972:  95%|█████████▍| 237/250 [24:00<01:18,  6.06s/it]
Training Epoch: 15/100, batch 474/500 completed (loss: 0.0225:  95%|█████████▍| 237/250 [24:00<01:18,  6.07s/it]
Training Epoch: 15/100, batch 474/500 completed (loss: 0.0972:  95%|█████████▌| 238/250 [24:03<01:13,  6.14s/it]
Training Epoch: 15/100, batch 474/500 completed (loss: 0.0225:  95%|█████████▌| 238/250 [24:03<01:13,  6.14s/it]
Training Epoch: 15/100, batch 474/500 completed (loss: 0.0538:  95%|█████████▌| 238/250 [24:03<01:13,  6.14s/it]
Training Epoch: 15/100, batch 474/500 completed (loss: 0.008:  95%|█████████▌| 238/250 [24:03<01:13,  6.15s/it]
Training Epoch: 15/100, batch 475/500 completed (loss: 0.0649:  95%|█████████▌| 238/250 [24:03<01:13,  6.15s/it]
Training Epoch: 15/100, batch 475/500 completed (loss: 0.0726:  95%|█████████▌| 238/250 [24:03<01:13,  6.14s/it]
Training Epoch: 15/100, batch 475/500 completed (loss: 0.0058:  95%|█████████▌| 238/250 [24:03<01:13,  6.14s/it]
Training Epoch: 15/100, batch 475/500 completed (loss: 0.003:  95%|█████████▌| 238/250 [24:03<01:13,  6.14s/it] 
Training Epoch: 15/100, batch 476/500 completed (loss: 0.0582:  95%|█████████▌| 238/250 [24:06<01:13,  6.14s/it]
Training Epoch: 15/100, batch 476/500 completed (loss: 0.0209:  95%|█████████▌| 238/250 [24:06<01:13,  6.14s/it]
Training Epoch: 15/100, batch 476/500 completed (loss: 0.1224:  95%|█████████▌| 238/250 [24:06<01:13,  6.15s/it]
Training Epoch: 15/100, batch 476/500 completed (loss: 0.0985:  95%|█████████▌| 238/250 [24:06<01:13,  6.14s/it]
Training Epoch: 15/100, batch 476/500 completed (loss: 0.1224:  96%|█████████▌| 239/250 [24:09<01:07,  6.11s/it]
Training Epoch: 15/100, batch 476/500 completed (loss: 0.0582:  96%|█████████▌| 239/250 [24:09<01:07,  6.11s/it]
Training Epoch: 15/100, batch 476/500 completed (loss: 0.0209:  96%|█████████▌| 239/250 [24:09<01:07,  6.11s/it]
Training Epoch: 15/100, batch 476/500 completed (loss: 0.0985:  96%|█████████▌| 239/250 [24:09<01:07,  6.11s/it]
Training Epoch: 15/100, batch 477/500 completed (loss: 0.0715:  96%|█████████▌| 239/250 [24:09<01:07,  6.11s/it]
Training Epoch: 15/100, batch 477/500 completed (loss: 0.0236:  96%|█████████▌| 239/250 [24:09<01:07,  6.11s/it]
Training Epoch: 15/100, batch 477/500 completed (loss: 0.0181:  96%|█████████▌| 239/250 [24:09<01:07,  6.11s/it]
Training Epoch: 15/100, batch 477/500 completed (loss: 0.0135:  96%|█████████▌| 239/250 [24:09<01:07,  6.11s/it]
Training Epoch: 15/100, batch 478/500 completed (loss: 0.0183:  96%|█████████▌| 239/250 [24:12<01:07,  6.11s/it]
Training Epoch: 15/100, batch 478/500 completed (loss: 0.0049:  96%|█████████▌| 239/250 [24:12<01:07,  6.11s/it]
Training Epoch: 15/100, batch 478/500 completed (loss: 0.1089:  96%|█████████▌| 239/250 [24:12<01:07,  6.11s/it]
Training Epoch: 15/100, batch 478/500 completed (loss: 0.0883:  96%|█████████▌| 239/250 [24:12<01:07,  6.11s/it]
Training Epoch: 15/100, batch 478/500 completed (loss: 0.0883:  96%|█████████▌| 240/250 [24:15<01:00,  6.09s/it]
Training Epoch: 15/100, batch 478/500 completed (loss: 0.0049:  96%|█████████▌| 240/250 [24:15<01:00,  6.10s/it]
Training Epoch: 15/100, batch 478/500 completed (loss: 0.0183:  96%|█████████▌| 240/250 [24:15<01:00,  6.10s/it]
Training Epoch: 15/100, batch 478/500 completed (loss: 0.1089:  96%|█████████▌| 240/250 [24:15<01:00,  6.10s/it]
Training Epoch: 15/100, batch 479/500 completed (loss: 0.0551:  96%|█████████▌| 240/250 [24:15<01:00,  6.10s/it]
Training Epoch: 15/100, batch 479/500 completed (loss: 0.0566:  96%|█████████▌| 240/250 [24:15<01:00,  6.10s/it]
Training Epoch: 15/100, batch 479/500 completed (loss: 0.0397:  96%|█████████▌| 240/250 [24:15<01:00,  6.10s/it]
Training Epoch: 15/100, batch 479/500 completed (loss: 0.0165:  96%|█████████▌| 240/250 [24:15<01:00,  6.09s/it]
Training Epoch: 15/100, batch 480/500 completed (loss: 0.1718:  96%|█████████▌| 240/250 [24:18<01:00,  6.10s/it]
Training Epoch: 15/100, batch 480/500 completed (loss: 0.3548:  96%|█████████▌| 240/250 [24:18<01:00,  6.10s/it]
Training Epoch: 15/100, batch 480/500 completed (loss: 0.0597:  96%|█████████▌| 240/250 [24:18<01:00,  6.10s/it]
Training Epoch: 15/100, batch 480/500 completed (loss: 0.0188:  96%|█████████▌| 240/250 [24:18<01:00,  6.09s/it]
Training Epoch: 15/100, batch 480/500 completed (loss: 0.3548:  96%|█████████▋| 241/250 [24:21<00:54,  6.07s/it]
Training Epoch: 15/100, batch 480/500 completed (loss: 0.1718:  96%|█████████▋| 241/250 [24:21<00:54,  6.08s/it]
Training Epoch: 15/100, batch 480/500 completed (loss: 0.0597:  96%|█████████▋| 241/250 [24:21<00:54,  6.08s/it]
Training Epoch: 15/100, batch 480/500 completed (loss: 0.0188:  96%|█████████▋| 241/250 [24:21<00:54,  6.08s/it]
Training Epoch: 15/100, batch 481/500 completed (loss: 0.2109:  96%|█████████▋| 241/250 [24:21<00:54,  6.08s/it]
Training Epoch: 15/100, batch 481/500 completed (loss: 0.2458:  96%|█████████▋| 241/250 [24:21<00:54,  6.08s/it]
Training Epoch: 15/100, batch 481/500 completed (loss: 0.1779:  96%|█████████▋| 241/250 [24:21<00:54,  6.07s/it]
Training Epoch: 15/100, batch 481/500 completed (loss: 0.0662:  96%|█████████▋| 241/250 [24:21<00:54,  6.08s/it]
Training Epoch: 15/100, batch 482/500 completed (loss: 0.0177:  96%|█████████▋| 241/250 [24:24<00:54,  6.07s/it]
Training Epoch: 15/100, batch 482/500 completed (loss: 0.1266:  96%|█████████▋| 241/250 [24:24<00:54,  6.08s/it]
Training Epoch: 15/100, batch 482/500 completed (loss: 0.0268:  96%|█████████▋| 241/250 [24:24<00:54,  6.08s/it]
Training Epoch: 15/100, batch 482/500 completed (loss: 0.0619:  96%|█████████▋| 241/250 [24:24<00:54,  6.08s/it]
Training Epoch: 15/100, batch 482/500 completed (loss: 0.1266:  97%|█████████▋| 242/250 [24:27<00:48,  6.06s/it]
Training Epoch: 15/100, batch 482/500 completed (loss: 0.0268:  97%|█████████▋| 242/250 [24:27<00:48,  6.06s/it]
Training Epoch: 15/100, batch 482/500 completed (loss: 0.0619:  97%|█████████▋| 242/250 [24:27<00:48,  6.06s/it]
Training Epoch: 15/100, batch 482/500 completed (loss: 0.0177:  97%|█████████▋| 242/250 [24:27<00:48,  6.06s/it]
Training Epoch: 15/100, batch 483/500 completed (loss: 0.0285:  97%|█████████▋| 242/250 [24:27<00:48,  6.06s/it]
Training Epoch: 15/100, batch 483/500 completed (loss: 0.4426:  97%|█████████▋| 242/250 [24:27<00:48,  6.06s/it]
Training Epoch: 15/100, batch 483/500 completed (loss: 0.098:  97%|█████████▋| 242/250 [24:27<00:48,  6.06s/it] 
Training Epoch: 15/100, batch 483/500 completed (loss: 0.3231:  97%|█████████▋| 242/250 [24:27<00:48,  6.06s/it]
Training Epoch: 15/100, batch 484/500 completed (loss: 0.113:  97%|█████████▋| 242/250 [24:31<00:48,  6.06s/it] 
Training Epoch: 15/100, batch 484/500 completed (loss: 0.1185:  97%|█████████▋| 242/250 [24:31<00:48,  6.06s/it]
Training Epoch: 15/100, batch 484/500 completed (loss: 0.0317:  97%|█████████▋| 242/250 [24:31<00:48,  6.06s/it]
Training Epoch: 15/100, batch 484/500 completed (loss: 0.0489:  97%|█████████▋| 242/250 [24:31<00:48,  6.06s/it]
Training Epoch: 15/100, batch 484/500 completed (loss: 0.113:  97%|█████████▋| 243/250 [24:34<00:42,  6.09s/it]
Training Epoch: 15/100, batch 484/500 completed (loss: 0.0317:  97%|█████████▋| 243/250 [24:34<00:42,  6.09s/it]
Training Epoch: 15/100, batch 484/500 completed (loss: 0.1185:  97%|█████████▋| 243/250 [24:34<00:42,  6.09s/it]
Training Epoch: 15/100, batch 484/500 completed (loss: 0.0489:  97%|█████████▋| 243/250 [24:34<00:42,  6.09s/it]
Training Epoch: 15/100, batch 485/500 completed (loss: 0.054:  97%|█████████▋| 243/250 [24:34<00:42,  6.09s/it] 
Training Epoch: 15/100, batch 485/500 completed (loss: 0.0347:  97%|█████████▋| 243/250 [24:34<00:42,  6.09s/it]
Training Epoch: 15/100, batch 485/500 completed (loss: 0.3173:  97%|█████████▋| 243/250 [24:34<00:42,  6.09s/it]
Training Epoch: 15/100, batch 485/500 completed (loss: 0.2832:  97%|█████████▋| 243/250 [24:34<00:42,  6.09s/it]
Training Epoch: 15/100, batch 486/500 completed (loss: 0.1834:  97%|█████████▋| 243/250 [24:37<00:42,  6.09s/it]
Training Epoch: 15/100, batch 486/500 completed (loss: 0.3129:  97%|█████████▋| 243/250 [24:37<00:42,  6.09s/it]
Training Epoch: 15/100, batch 486/500 completed (loss: 0.4362:  97%|█████████▋| 243/250 [24:37<00:42,  6.09s/it]
Training Epoch: 15/100, batch 486/500 completed (loss: 0.0145:  97%|█████████▋| 243/250 [24:37<00:42,  6.09s/it]
Training Epoch: 15/100, batch 486/500 completed (loss: 0.0145:  98%|█████████▊| 244/250 [24:40<00:36,  6.11s/it]
Training Epoch: 15/100, batch 486/500 completed (loss: 0.3129:  98%|█████████▊| 244/250 [24:40<00:36,  6.11s/it]
Training Epoch: 15/100, batch 486/500 completed (loss: 0.4362:  98%|█████████▊| 244/250 [24:40<00:36,  6.11s/it]
Training Epoch: 15/100, batch 486/500 completed (loss: 0.1834:  98%|█████████▊| 244/250 [24:40<00:36,  6.11s/it]
Training Epoch: 15/100, batch 487/500 completed (loss: 0.0155:  98%|█████████▊| 244/250 [24:40<00:36,  6.11s/it]
Training Epoch: 15/100, batch 487/500 completed (loss: 0.0385:  98%|█████████▊| 244/250 [24:40<00:36,  6.11s/it]
Training Epoch: 15/100, batch 487/500 completed (loss: 0.0184:  98%|█████████▊| 244/250 [24:40<00:36,  6.11s/it]
Training Epoch: 15/100, batch 487/500 completed (loss: 0.1683:  98%|█████████▊| 244/250 [24:40<00:36,  6.11s/it]
Training Epoch: 15/100, batch 488/500 completed (loss: 0.0211:  98%|█████████▊| 244/250 [24:43<00:36,  6.11s/it]
Training Epoch: 15/100, batch 488/500 completed (loss: 0.0691:  98%|█████████▊| 244/250 [24:43<00:36,  6.11s/it]
Training Epoch: 15/100, batch 488/500 completed (loss: 0.0485:  98%|█████████▊| 244/250 [24:43<00:36,  6.11s/it]
Training Epoch: 15/100, batch 488/500 completed (loss: 0.0881:  98%|█████████▊| 244/250 [24:43<00:36,  6.11s/it]
Training Epoch: 15/100, batch 488/500 completed (loss: 0.0881:  98%|█████████▊| 245/250 [24:46<00:30,  6.13s/it]
Training Epoch: 15/100, batch 488/500 completed (loss: 0.0485:  98%|█████████▊| 245/250 [24:46<00:30,  6.13s/it]
Training Epoch: 15/100, batch 488/500 completed (loss: 0.0211:  98%|█████████▊| 245/250 [24:46<00:30,  6.13s/it]
Training Epoch: 15/100, batch 488/500 completed (loss: 0.0691:  98%|█████████▊| 245/250 [24:46<00:30,  6.13s/it]
Training Epoch: 15/100, batch 489/500 completed (loss: 0.1226:  98%|█████████▊| 245/250 [24:46<00:30,  6.13s/it]
Training Epoch: 15/100, batch 489/500 completed (loss: 0.1476:  98%|█████████▊| 245/250 [24:46<00:30,  6.13s/it]
Training Epoch: 15/100, batch 489/500 completed (loss: 0.1524:  98%|█████████▊| 245/250 [24:46<00:30,  6.13s/it]
Training Epoch: 15/100, batch 489/500 completed (loss: 0.0351:  98%|█████████▊| 245/250 [24:46<00:30,  6.13s/it]
Training Epoch: 15/100, batch 490/500 completed (loss: 0.0119:  98%|█████████▊| 245/250 [24:49<00:30,  6.13s/it]
Training Epoch: 15/100, batch 490/500 completed (loss: 0.1978:  98%|█████████▊| 245/250 [24:49<00:30,  6.13s/it]
Training Epoch: 15/100, batch 490/500 completed (loss: 0.0415:  98%|█████████▊| 245/250 [24:49<00:30,  6.13s/it]
Training Epoch: 15/100, batch 490/500 completed (loss: 0.051:  98%|█████████▊| 245/250 [24:49<00:30,  6.13s/it] 
Training Epoch: 15/100, batch 490/500 completed (loss: 0.0119:  98%|█████████▊| 246/250 [24:52<00:24,  6.05s/it]
Training Epoch: 15/100, batch 490/500 completed (loss: 0.1978:  98%|█████████▊| 246/250 [24:52<00:24,  6.05s/it]
Training Epoch: 15/100, batch 490/500 completed (loss: 0.0415:  98%|█████████▊| 246/250 [24:52<00:24,  6.05s/it]
Training Epoch: 15/100, batch 490/500 completed (loss: 0.051:  98%|█████████▊| 246/250 [24:52<00:24,  6.05s/it]
Training Epoch: 15/100, batch 491/500 completed (loss: 0.064:  98%|█████████▊| 246/250 [24:52<00:24,  6.05s/it] 
Training Epoch: 15/100, batch 491/500 completed (loss: 0.0202:  98%|█████████▊| 246/250 [24:52<00:24,  6.05s/it]
Training Epoch: 15/100, batch 491/500 completed (loss: 0.1185:  98%|█████████▊| 246/250 [24:52<00:24,  6.05s/it]
Training Epoch: 15/100, batch 491/500 completed (loss: 0.1391:  98%|█████████▊| 246/250 [24:52<00:24,  6.05s/it]
Training Epoch: 15/100, batch 492/500 completed (loss: 0.0026:  98%|█████████▊| 246/250 [24:55<00:24,  6.05s/it]
Training Epoch: 15/100, batch 492/500 completed (loss: 0.036:  98%|█████████▊| 246/250 [24:55<00:24,  6.05s/it] 
Training Epoch: 15/100, batch 492/500 completed (loss: 0.0257:  98%|█████████▊| 246/250 [24:55<00:24,  6.05s/it]
Training Epoch: 15/100, batch 492/500 completed (loss: 0.1006:  98%|█████████▊| 246/250 [24:55<00:24,  6.05s/it]
Training Epoch: 15/100, batch 492/500 completed (loss: 0.036:  99%|█████████▉| 247/250 [24:58<00:18,  6.10s/it]
Training Epoch: 15/100, batch 492/500 completed (loss: 0.0257:  99%|█████████▉| 247/250 [24:58<00:18,  6.09s/it]
Training Epoch: 15/100, batch 492/500 completed (loss: 0.1006:  99%|█████████▉| 247/250 [24:58<00:18,  6.09s/it]
Training Epoch: 15/100, batch 492/500 completed (loss: 0.0026:  99%|█████████▉| 247/250 [24:58<00:18,  6.10s/it]
Training Epoch: 15/100, batch 493/500 completed (loss: 0.0375:  99%|█████████▉| 247/250 [24:58<00:18,  6.09s/it]
Training Epoch: 15/100, batch 493/500 completed (loss: 0.1237:  99%|█████████▉| 247/250 [24:58<00:18,  6.10s/it]
Training Epoch: 15/100, batch 493/500 completed (loss: 0.0478:  99%|█████████▉| 247/250 [24:58<00:18,  6.10s/it]
Training Epoch: 15/100, batch 493/500 completed (loss: 0.2841:  99%|█████████▉| 247/250 [24:58<00:18,  6.09s/it]
Training Epoch: 15/100, batch 494/500 completed (loss: 0.0527:  99%|█████████▉| 247/250 [25:01<00:18,  6.10s/it]
Training Epoch: 15/100, batch 494/500 completed (loss: 0.2034:  99%|█████████▉| 247/250 [25:01<00:18,  6.10s/it]
Training Epoch: 15/100, batch 494/500 completed (loss: 0.3929:  99%|█████████▉| 247/250 [25:01<00:18,  6.09s/it]
Training Epoch: 15/100, batch 494/500 completed (loss: 0.0389:  99%|█████████▉| 247/250 [25:01<00:18,  6.09s/it]
Training Epoch: 15/100, batch 494/500 completed (loss: 0.0389:  99%|█████████▉| 248/250 [25:04<00:12,  6.12s/it]
Training Epoch: 15/100, batch 494/500 completed (loss: 0.2034:  99%|█████████▉| 248/250 [25:04<00:12,  6.13s/it]
Training Epoch: 15/100, batch 494/500 completed (loss: 0.0527:  99%|█████████▉| 248/250 [25:04<00:12,  6.13s/it]
Training Epoch: 15/100, batch 494/500 completed (loss: 0.3929:  99%|█████████▉| 248/250 [25:04<00:12,  6.13s/it]
Training Epoch: 15/100, batch 495/500 completed (loss: 0.2429:  99%|█████████▉| 248/250 [25:04<00:12,  6.13s/it]
Training Epoch: 15/100, batch 495/500 completed (loss: 0.0194:  99%|█████████▉| 248/250 [25:04<00:12,  6.13s/it]
Training Epoch: 15/100, batch 495/500 completed (loss: 0.0444:  99%|█████████▉| 248/250 [25:04<00:12,  6.13s/it]
Training Epoch: 15/100, batch 495/500 completed (loss: 0.0855:  99%|█████████▉| 248/250 [25:04<00:12,  6.12s/it]
Training Epoch: 15/100, batch 496/500 completed (loss: 0.1933:  99%|█████████▉| 248/250 [25:07<00:12,  6.13s/it]
Training Epoch: 15/100, batch 496/500 completed (loss: 0.0072:  99%|█████████▉| 248/250 [25:07<00:12,  6.13s/it]
Training Epoch: 15/100, batch 496/500 completed (loss: 0.0304:  99%|█████████▉| 248/250 [25:07<00:12,  6.13s/it]
Training Epoch: 15/100, batch 496/500 completed (loss: 0.0084:  99%|█████████▉| 248/250 [25:07<00:12,  6.12s/it]
Training Epoch: 15/100, batch 496/500 completed (loss: 0.0072: 100%|█████████▉| 249/250 [25:10<00:06,  6.10s/it]
Training Epoch: 15/100, batch 496/500 completed (loss: 0.1933: 100%|█████████▉| 249/250 [25:10<00:06,  6.10s/it]
Training Epoch: 15/100, batch 496/500 completed (loss: 0.0304: 100%|█████████▉| 249/250 [25:10<00:06,  6.10s/it]
Training Epoch: 15/100, batch 496/500 completed (loss: 0.0084: 100%|█████████▉| 249/250 [25:10<00:06,  6.10s/it]
Training Epoch: 15/100, batch 497/500 completed (loss: 0.7603: 100%|█████████▉| 249/250 [25:10<00:06,  6.10s/it]
Training Epoch: 15/100, batch 497/500 completed (loss: 0.0357: 100%|█████████▉| 249/250 [25:10<00:06,  6.10s/it]
Training Epoch: 15/100, batch 497/500 completed (loss: 0.0229: 100%|█████████▉| 249/250 [25:10<00:06,  6.10s/it]
Training Epoch: 15/100, batch 497/500 completed (loss: 0.0891: 100%|█████████▉| 249/250 [25:10<00:06,  6.10s/it]
Training Epoch: 15/100, batch 498/500 completed (loss: 0.0533: 100%|█████████▉| 249/250 [25:13<00:06,  6.10s/it]
Training Epoch: 15/100, batch 498/500 completed (loss: 0.0336: 100%|█████████▉| 249/250 [25:13<00:06,  6.10s/it]
Training Epoch: 15/100, batch 498/500 completed (loss: 0.055: 100%|█████████▉| 249/250 [25:13<00:06,  6.10s/it] 
Training Epoch: 15/100, batch 498/500 completed (loss: 0.0239: 100%|█████████▉| 249/250 [25:13<00:06,  6.10s/it]
Training Epoch: 15/100, batch 498/500 completed (loss: 0.0239: 100%|██████████| 250/250 [25:16<00:00,  6.07s/it]
Training Epoch: 15/100, batch 498/500 completed (loss: 0.0336: 100%|██████████| 250/250 [25:16<00:00,  6.07s/it]
Training Epoch: 15/100, batch 498/500 completed (loss: 0.0533: 100%|██████████| 250/250 [25:16<00:00,  6.08s/it]
Training Epoch: 15/100, batch 498/500 completed (loss: 0.055: 100%|██████████| 250/250 [25:16<00:00,  6.08s/it]
Training Epoch: 15/100, batch 499/500 completed (loss: 0.031: 100%|██████████| 250/250 [25:16<00:00,  6.07s/it] 
Training Epoch: 15/100, batch 499/500 completed (loss: 0.0685: 100%|██████████| 250/250 [25:16<00:00,  6.08s/it]
Training Epoch: 15/100, batch 499/500 completed (loss: 0.0155: 100%|██████████| 250/250 [25:16<00:00,  6.08s/it]
Training Epoch: 15/100, batch 499/500 completed (loss: 0.0582: 100%|██████████| 250/250 [25:16<00:00,  6.07s/it]
Training Epoch: 15/100, batch 499/500 completed (loss: 0.0685: 100%|██████████| 250/250 [25:16<00:00,  6.07s/it]

Training Epoch: 15/100, batch 499/500 completed (loss: 0.0155: 100%|██████████| 250/250 [25:16<00:00,  6.07s/it]

Training Epoch: 15/100, batch 499/500 completed (loss: 0.031: 100%|██████████| 250/250 [25:16<00:00,  6.07s/it]

Training Epoch: 15/100, batch 499/500 completed (loss: 0.0582: 100%|██████████| 250/250 [25:17<00:00,  6.07s/it]
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/c10d_logger.py:83: UserWarning: barrier(): using the device under current context. You can specify `device_id` in `init_process_group` to mute this warning.
  return func(*args, **kwargs)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
saving model + train state.

Processing Files (0 / 0)      : |          |  0.00B /  0.00B            

New Data Upload               : |          |  0.00B /  0.00B            


  ...ve-03b89ee9/checkpoint_15:   0%|          | 82.1kB / 3.63GB            


  ...ve-03b89ee9/checkpoint_15:   0%|          | 82.1kB / 3.63GB            
Processing Files (0 / 1)      :   0%|          | 82.1kB / 3.63GB,   ???B/s  


  ...ve-03b89ee9/checkpoint_15:   0%|          | 17.7MB / 3.63GB            
Processing Files (0 / 1)      :   0%|          | 17.7MB / 3.63GB, 1.73MB/s  

New Data Upload               :  13%|█▎        | 17.6MB /  134MB, 1.72MB/s  


  ...ve-03b89ee9/checkpoint_15:   3%|▎         | 96.0MB / 3.63GB            
Processing Files (0 / 1)      :   3%|▎         | 96.0MB / 3.63GB, 9.28MB/s  

New Data Upload               :  48%|████▊     | 95.9MB /  201MB, 9.28MB/s  


  ...ve-03b89ee9/checkpoint_15:   4%|▎         |  133MB / 3.63GB            
Processing Files (0 / 1)      :   4%|▎         |  133MB / 3.63GB, 12.7MB/s  

New Data Upload               :  66%|██████▋   |  133MB /  201MB, 12.7MB/s  


  ...ve-03b89ee9/checkpoint_15:   4%|▎         |  134MB / 3.63GB            
Processing Files (0 / 1)      :   4%|▎         |  134MB / 3.63GB, 11.9MB/s  

New Data Upload               :  50%|█████     |  134MB /  268MB, 11.9MB/s  


  ...ve-03b89ee9/checkpoint_15:   6%|▌         |  205MB / 3.63GB            
Processing Files (0 / 1)      :   6%|▌         |  205MB / 3.63GB, 18.2MB/s  

New Data Upload               :  61%|██████    |  205MB /  335MB, 18.2MB/s  


  ...ve-03b89ee9/checkpoint_15:   8%|▊         |  307MB / 3.63GB            
Processing Files (0 / 1)      :   8%|▊         |  307MB / 3.63GB, 27.3MB/s  

New Data Upload               :  76%|███████▌  |  307MB /  402MB, 27.3MB/s  


  ...ve-03b89ee9/checkpoint_15:   9%|▉         |  340MB / 3.63GB            
Processing Files (0 / 1)      :   9%|▉         |  340MB / 3.63GB, 29.8MB/s  

New Data Upload               :  73%|███████▎  |  340MB /  469MB, 29.8MB/s  


  ...ve-03b89ee9/checkpoint_15:  11%|█         |  401MB / 3.63GB            
Processing Files (0 / 1)      :  11%|█         |  401MB / 3.63GB, 34.7MB/s  

New Data Upload               :  85%|████████▌ |  401MB /  469MB, 34.7MB/s  


  ...ve-03b89ee9/checkpoint_15:  11%|█         |  402MB / 3.63GB            
Processing Files (0 / 1)      :  11%|█         |  402MB / 3.63GB, 33.9MB/s  

New Data Upload               :  86%|████████▌ |  402MB /  469MB, 33.9MB/s  


  ...ve-03b89ee9/checkpoint_15:  13%|█▎        |  465MB / 3.63GB            
Processing Files (0 / 1)      :  13%|█▎        |  465MB / 3.63GB, 38.9MB/s  

New Data Upload               :  77%|███████▋  |  465MB /  603MB, 38.9MB/s  


  ...ve-03b89ee9/checkpoint_15:  16%|█▌        |  568MB / 3.63GB            
Processing Files (0 / 1)      :  16%|█▌        |  568MB / 3.63GB, 47.4MB/s  

New Data Upload               :  85%|████████▍ |  568MB /  671MB, 47.4MB/s  


  ...ve-03b89ee9/checkpoint_15:  17%|█▋        |  619MB / 3.63GB            
Processing Files (0 / 1)      :  17%|█▋        |  619MB / 3.63GB, 50.9MB/s  

New Data Upload               :  84%|████████▍ |  619MB /  738MB, 50.9MB/s  


  ...ve-03b89ee9/checkpoint_15:  20%|█▉        |  725MB / 3.63GB            
Processing Files (0 / 1)      :  20%|█▉        |  725MB / 3.63GB, 59.2MB/s  

New Data Upload               :  83%|████████▎ |  725MB /  872MB, 59.2MB/s  


  ...ve-03b89ee9/checkpoint_15:  23%|██▎       |  837MB / 3.63GB            
Processing Files (0 / 1)      :  23%|██▎       |  837MB / 3.63GB, 67.9MB/s  

New Data Upload               :  89%|████████▉ |  837MB /  939MB, 67.9MB/s  


  ...ve-03b89ee9/checkpoint_15:  25%|██▌       |  915MB / 3.63GB            
Processing Files (0 / 1)      :  25%|██▌       |  915MB / 3.63GB, 73.4MB/s  

New Data Upload               :  85%|████████▌ |  915MB / 1.07GB, 73.4MB/s  


  ...ve-03b89ee9/checkpoint_15:  28%|██▊       | 1.01GB / 3.63GB            
Processing Files (0 / 1)      :  28%|██▊       | 1.01GB / 3.63GB, 80.2MB/s  

New Data Upload               :  89%|████████▊ | 1.01GB / 1.14GB, 80.2MB/s  


  ...ve-03b89ee9/checkpoint_15:  31%|███       | 1.11GB / 3.63GB            
Processing Files (0 / 1)      :  31%|███       | 1.11GB / 3.63GB, 86.8MB/s  

New Data Upload               :  92%|█████████▏| 1.11GB / 1.21GB, 86.8MB/s  


  ...ve-03b89ee9/checkpoint_15:  33%|███▎      | 1.20GB / 3.63GB            
Processing Files (0 / 1)      :  33%|███▎      | 1.20GB / 3.63GB, 93.5MB/s  

New Data Upload               :  90%|████████▉ | 1.20GB / 1.34GB, 93.5MB/s  


  ...ve-03b89ee9/checkpoint_15:  36%|███▌      | 1.30GB / 3.63GB            
Processing Files (0 / 1)      :  36%|███▌      | 1.30GB / 3.63GB, 99.9MB/s  

New Data Upload               :  92%|█████████▏| 1.30GB / 1.41GB, 99.9MB/s  


  ...ve-03b89ee9/checkpoint_15:  38%|███▊      | 1.37GB / 3.63GB            
Processing Files (0 / 1)      :  38%|███▊      | 1.37GB / 3.63GB,  104MB/s  

New Data Upload               :  89%|████████▉ | 1.37GB / 1.54GB,  104MB/s  


  ...ve-03b89ee9/checkpoint_15:  41%|████      | 1.48GB / 3.63GB            
Processing Files (0 / 1)      :  41%|████      | 1.48GB / 3.63GB,  110MB/s  

New Data Upload               :  92%|█████████▏| 1.48GB / 1.61GB,  110MB/s  


  ...ve-03b89ee9/checkpoint_15:  42%|████▏     | 1.53GB / 3.63GB            
Processing Files (0 / 1)      :  42%|████▏     | 1.53GB / 3.63GB,  113MB/s  

New Data Upload               :  91%|█████████ | 1.53GB / 1.68GB,  113MB/s  


  ...ve-03b89ee9/checkpoint_15:  44%|████▍     | 1.60GB / 3.63GB            
Processing Files (0 / 1)      :  44%|████▍     | 1.60GB / 3.63GB,  117MB/s  

New Data Upload               :  89%|████████▊ | 1.60GB / 1.81GB,  117MB/s  


  ...ve-03b89ee9/checkpoint_15:  48%|████▊     | 1.73GB / 3.63GB            
Processing Files (0 / 1)      :  48%|████▊     | 1.73GB / 3.63GB,  125MB/s  

New Data Upload               :  92%|█████████▏| 1.73GB / 1.88GB,  125MB/s  


  ...ve-03b89ee9/checkpoint_15:  51%|█████     | 1.83GB / 3.63GB            
Processing Files (0 / 1)      :  51%|█████     | 1.83GB / 3.63GB,  131MB/s  

New Data Upload               :  94%|█████████▍| 1.83GB / 1.94GB,  131MB/s  


  ...ve-03b89ee9/checkpoint_15:  53%|█████▎    | 1.92GB / 3.63GB            
Processing Files (0 / 1)      :  53%|█████▎    | 1.92GB / 3.63GB,  136MB/s  

New Data Upload               :  93%|█████████▎| 1.92GB / 2.08GB,  136MB/s  


  ...ve-03b89ee9/checkpoint_15:  56%|█████▌    | 2.02GB / 3.63GB            
Processing Files (0 / 1)      :  56%|█████▌    | 2.02GB / 3.63GB,  141MB/s  

New Data Upload               :  94%|█████████▍| 2.02GB / 2.15GB,  141MB/s  


  ...ve-03b89ee9/checkpoint_15:  59%|█████▊    | 2.13GB / 3.63GB            
Processing Files (0 / 1)      :  59%|█████▊    | 2.13GB / 3.63GB,  147MB/s  

New Data Upload               :  93%|█████████▎| 2.13GB / 2.28GB,  147MB/s  


  ...ve-03b89ee9/checkpoint_15:  61%|██████▏   | 2.22GB / 3.63GB            
Processing Files (0 / 1)      :  61%|██████▏   | 2.22GB / 3.63GB,  152MB/s  

New Data Upload               :  95%|█████████▍| 2.22GB / 2.35GB,  152MB/s  


  ...ve-03b89ee9/checkpoint_15:  71%|███████▏  | 2.59GB / 3.63GB            
Processing Files (0 / 1)      :  71%|███████▏  | 2.59GB / 3.63GB,  180MB/s  

New Data Upload               :  98%|█████████▊| 2.31GB / 2.35GB,  156MB/s  


  ...ve-03b89ee9/checkpoint_15:  82%|████████▏ | 2.98GB / 3.63GB            
Processing Files (0 / 1)      :  82%|████████▏ | 2.98GB / 3.63GB,  208MB/s  

New Data Upload               : 100%|█████████▉| 2.35GB / 2.35GB,  156MB/s  


  ...ve-03b89ee9/checkpoint_15:  93%|█████████▎| 3.35GB / 3.63GB            
Processing Files (0 / 1)      :  93%|█████████▎| 3.35GB / 3.63GB,  235MB/s  

New Data Upload               : 100%|█████████▉| 2.35GB / 2.35GB,  153MB/s  


  ...ve-03b89ee9/checkpoint_15:  99%|█████████▉| 3.59GB / 3.63GB            
Processing Files (0 / 1)      :  99%|█████████▉| 3.59GB / 3.63GB,  249MB/s  

New Data Upload               :  98%|█████████▊| 2.35GB / 2.38GB,  150MB/s  


  ...ve-03b89ee9/checkpoint_15: 100%|█████████▉| 3.62GB / 3.63GB            
Processing Files (0 / 1)      : 100%|█████████▉| 3.62GB / 3.63GB,  247MB/s  

New Data Upload               : 100%|█████████▉| 2.38GB / 2.38GB,  150MB/s  


  ...ve-03b89ee9/checkpoint_15: 100%|█████████▉| 3.63GB / 3.63GB            
Processing Files (0 / 1)      : 100%|█████████▉| 3.63GB / 3.63GB,  242MB/s  

New Data Upload               : 100%|█████████▉| 2.38GB / 2.38GB,  147MB/s  


  ...ve-03b89ee9/checkpoint_15: 100%|██████████| 3.63GB / 3.63GB            
Processing Files (1 / 1)      : 100%|██████████| 3.63GB / 3.63GB,  237MB/s  

New Data Upload               : 100%|██████████| 2.38GB / 2.38GB,  144MB/s  
Processing Files (1 / 1)      : 100%|██████████| 3.63GB / 3.63GB,  237MB/s  

New Data Upload               : 100%|██████████| 2.38GB / 2.38GB,  144MB/s  

  ...ve-03b89ee9/checkpoint_15: 100%|██████████| 3.63GB / 3.63GB            
Uploaded checkpoint_15 to seyedparsa/thoughtformer-checkpoints/checkpoints/star-datacurr-only-4gpu-bs32-lr1e-5_n1000_qwen3-0.6b-base_thoughtformer_lr1e-5_no-reset_dc_no-tc_adaptive-03b89ee9/checkpoint_15

Processing Files (0 / 0)      : |          |  0.00B /  0.00B            

New Data Upload               : |          |  0.00B /  0.00B            


  ...e-03b89ee9/train_state_15:   0%|          | 3.68MB / 4.77GB            


  ...e-03b89ee9/train_state_15:   0%|          | 3.68MB / 4.77GB            
Processing Files (0 / 1)      :   0%|          | 3.68MB / 4.77GB,   ???B/s  

New Data Upload               :   3%|▎         | 3.68MB /  134MB,   ???B/s  


  ...e-03b89ee9/train_state_15:   2%|▏         | 88.2MB / 4.77GB            
Processing Files (0 / 1)      :   2%|▏         | 88.2MB / 4.77GB,   ???B/s  

New Data Upload               :  44%|████▍     | 88.2MB /  201MB,   ???B/s  


  ...e-03b89ee9/train_state_15:   3%|▎         |  133MB / 4.77GB            
Processing Files (0 / 1)      :   3%|▎         |  133MB / 4.77GB, 12.9MB/s  

New Data Upload               :  66%|██████▋   |  133MB /  201MB, 12.9MB/s  


  ...e-03b89ee9/train_state_15:   4%|▎         |  177MB / 4.77GB            
Processing Files (0 / 1)      :   4%|▎         |  177MB / 4.77GB, 16.2MB/s  

New Data Upload               :  53%|█████▎    |  177MB /  335MB, 16.2MB/s  


  ...e-03b89ee9/train_state_15:   6%|▌         |  296MB / 4.77GB            
Processing Files (0 / 1)      :   6%|▌         |  296MB / 4.77GB, 27.0MB/s  

New Data Upload               :  73%|███████▎  |  296MB /  402MB, 27.0MB/s  


  ...e-03b89ee9/train_state_15:   8%|▊         |  360MB / 4.77GB            
Processing Files (0 / 1)      :   8%|▊         |  360MB / 4.77GB, 32.4MB/s  

New Data Upload               :  77%|███████▋  |  360MB /  469MB, 32.4MB/s  


  ...e-03b89ee9/train_state_15:   8%|▊         |  401MB / 4.77GB            
Processing Files (0 / 1)      :   8%|▊         |  401MB / 4.77GB, 35.6MB/s  

New Data Upload               :  85%|████████▌ |  401MB /  469MB, 35.6MB/s  


  ...e-03b89ee9/train_state_15:   9%|▉         |  426MB / 4.77GB            
Processing Files (0 / 1)      :   9%|▉         |  426MB / 4.77GB, 37.1MB/s  

New Data Upload               :  79%|███████▉  |  426MB /  537MB, 37.1MB/s  


  ...e-03b89ee9/train_state_15:  11%|█         |  503MB / 4.77GB            
Processing Files (0 / 1)      :  11%|█         |  503MB / 4.77GB, 43.3MB/s  

New Data Upload               :  75%|███████▍  |  503MB /  671MB, 43.3MB/s  


  ...e-03b89ee9/train_state_15:  12%|█▏        |  596MB / 4.77GB            
Processing Files (0 / 1)      :  12%|█▏        |  596MB / 4.77GB, 50.9MB/s  

New Data Upload               :  81%|████████  |  596MB /  738MB, 50.9MB/s  


  ...e-03b89ee9/train_state_15:  14%|█▍        |  678MB / 4.77GB            
Processing Files (0 / 1)      :  14%|█▍        |  678MB / 4.77GB, 57.2MB/s  

New Data Upload               :  84%|████████▍ |  678MB /  805MB, 57.2MB/s  


  ...e-03b89ee9/train_state_15:  16%|█▌        |  771MB / 4.77GB            
Processing Files (0 / 1)      :  16%|█▌        |  771MB / 4.77GB, 64.5MB/s  

New Data Upload               :  88%|████████▊ |  771MB /  872MB, 64.5MB/s  


  ...e-03b89ee9/train_state_15:  18%|█▊        |  841MB / 4.77GB            
Processing Files (0 / 1)      :  18%|█▊        |  841MB / 4.77GB, 69.3MB/s  

New Data Upload               :  84%|████████▎ |  841MB / 1.01GB, 69.3MB/s  


  ...e-03b89ee9/train_state_15:  20%|██        |  954MB / 4.77GB            
Processing Files (0 / 1)      :  20%|██        |  954MB / 4.77GB, 77.9MB/s  

New Data Upload               :  89%|████████▉ |  954MB / 1.07GB, 77.9MB/s  


  ...e-03b89ee9/train_state_15:  22%|██▏       | 1.05GB / 4.77GB            
Processing Files (0 / 1)      :  22%|██▏       | 1.05GB / 4.77GB, 84.5MB/s  

New Data Upload               :  87%|████████▋ | 1.05GB / 1.21GB, 84.5MB/s  


  ...e-03b89ee9/train_state_15:  24%|██▍       | 1.14GB / 4.77GB            
Processing Files (0 / 1)      :  24%|██▍       | 1.14GB / 4.77GB, 91.3MB/s  

New Data Upload               :  90%|████████▉ | 1.14GB / 1.27GB, 91.3MB/s  


  ...e-03b89ee9/train_state_15:  24%|██▍       | 1.16GB / 4.77GB            
Processing Files (0 / 1)      :  24%|██▍       | 1.16GB / 4.77GB, 90.7MB/s  

New Data Upload               :  87%|████████▋ | 1.16GB / 1.34GB, 90.7MB/s  


  ...e-03b89ee9/train_state_15:  27%|██▋       | 1.28GB / 4.77GB            
Processing Files (0 / 1)      :  27%|██▋       | 1.28GB / 4.77GB, 99.0MB/s  

New Data Upload               :  87%|████████▋ | 1.28GB / 1.48GB, 99.0MB/s  


  ...e-03b89ee9/train_state_15:  29%|██▉       | 1.40GB / 4.77GB            
Processing Files (0 / 1)      :  29%|██▉       | 1.40GB / 4.77GB,  108MB/s  

New Data Upload               :  91%|█████████ | 1.40GB / 1.54GB,  108MB/s  


  ...e-03b89ee9/train_state_15:  32%|███▏      | 1.53GB / 4.77GB            
Processing Files (0 / 1)      :  32%|███▏      | 1.53GB / 4.77GB,  116MB/s  

New Data Upload               :  91%|█████████ | 1.53GB / 1.68GB,  116MB/s  


  ...e-03b89ee9/train_state_15:  34%|███▍      | 1.62GB / 4.77GB            
Processing Files (0 / 1)      :  34%|███▍      | 1.62GB / 4.77GB,  122MB/s  

New Data Upload               :  93%|█████████▎| 1.62GB / 1.74GB,  122MB/s  


  ...e-03b89ee9/train_state_15:  36%|███▌      | 1.71GB / 4.77GB            
Processing Files (0 / 1)      :  36%|███▌      | 1.71GB / 4.77GB,  127MB/s  

New Data Upload               :  91%|█████████ | 1.71GB / 1.88GB,  127MB/s  


  ...e-03b89ee9/train_state_15:  38%|███▊      | 1.82GB / 4.77GB            
Processing Files (0 / 1)      :  38%|███▊      | 1.82GB / 4.77GB,  134MB/s  

New Data Upload               :  94%|█████████▍| 1.82GB / 1.94GB,  134MB/s  


  ...e-03b89ee9/train_state_15:  40%|████      | 1.91GB / 4.77GB            
Processing Files (0 / 1)      :  40%|████      | 1.91GB / 4.77GB,  139MB/s  

New Data Upload               :  92%|█████████▏| 1.91GB / 2.08GB,  139MB/s  


  ...e-03b89ee9/train_state_15:  42%|████▏     | 1.98GB / 4.77GB            
Processing Files (0 / 1)      :  42%|████▏     | 1.98GB / 4.77GB,  142MB/s  

New Data Upload               :  92%|█████████▏| 1.98GB / 2.15GB,  142MB/s  


  ...e-03b89ee9/train_state_15:  44%|████▍     | 2.09GB / 4.77GB            
Processing Files (0 / 1)      :  44%|████▍     | 2.09GB / 4.77GB,  148MB/s  

New Data Upload               :  91%|█████████▏| 2.09GB / 2.28GB,  148MB/s  


  ...e-03b89ee9/train_state_15:  46%|████▌     | 2.20GB / 4.77GB            
Processing Files (0 / 1)      :  46%|████▌     | 2.20GB / 4.77GB,  154MB/s  

New Data Upload               :  94%|█████████▍| 2.20GB / 2.35GB,  154MB/s  


  ...e-03b89ee9/train_state_15:  48%|████▊     | 2.31GB / 4.77GB            
Processing Files (0 / 1)      :  48%|████▊     | 2.31GB / 4.77GB,  160MB/s  

New Data Upload               :  96%|█████████▌| 2.31GB / 2.41GB,  160MB/s  


  ...e-03b89ee9/train_state_15:  51%|█████     | 2.41GB / 4.77GB            
Processing Files (0 / 1)      :  51%|█████     | 2.41GB / 4.77GB,  165MB/s  

New Data Upload               :  95%|█████████▍| 2.41GB / 2.55GB,  165MB/s  


  ...e-03b89ee9/train_state_15:  52%|█████▏    | 2.49GB / 4.77GB            
Processing Files (0 / 1)      :  52%|█████▏    | 2.49GB / 4.77GB,  168MB/s  

New Data Upload               :  95%|█████████▌| 2.49GB / 2.62GB,  168MB/s  


  ...e-03b89ee9/train_state_15:  54%|█████▍    | 2.59GB / 4.77GB            
Processing Files (0 / 1)      :  54%|█████▍    | 2.59GB / 4.77GB,  173MB/s  

New Data Upload               :  97%|█████████▋| 2.59GB / 2.68GB,  173MB/s  


  ...e-03b89ee9/train_state_15:  56%|█████▌    | 2.67GB / 4.77GB            
Processing Files (0 / 1)      :  56%|█████▌    | 2.67GB / 4.77GB,  176MB/s  

New Data Upload               :  95%|█████████▍| 2.67GB / 2.82GB,  176MB/s  


  ...e-03b89ee9/train_state_15:  57%|█████▋    | 2.73GB / 4.77GB            
Processing Files (0 / 1)      :  57%|█████▋    | 2.73GB / 4.77GB,  178MB/s  

New Data Upload               :  95%|█████████▍| 2.73GB / 2.88GB,  178MB/s  


  ...e-03b89ee9/train_state_15:  60%|█████▉    | 2.84GB / 4.77GB            
Processing Files (0 / 1)      :  60%|█████▉    | 2.84GB / 4.77GB,  183MB/s  

New Data Upload               :  94%|█████████▍| 2.84GB / 3.02GB,  183MB/s  


  ...e-03b89ee9/train_state_15:  62%|██████▏   | 2.96GB / 4.77GB            
Processing Files (0 / 1)      :  62%|██████▏   | 2.96GB / 4.77GB,  189MB/s  

New Data Upload               :  96%|█████████▌| 2.96GB / 3.08GB,  189MB/s  


  ...e-03b89ee9/train_state_15:  64%|██████▍   | 3.06GB / 4.77GB            
Processing Files (0 / 1)      :  64%|██████▍   | 3.06GB / 4.77GB,  194MB/s  

New Data Upload               :  95%|█████████▌| 3.06GB / 3.22GB,  194MB/s  


  ...e-03b89ee9/train_state_15:  66%|██████▌   | 3.13GB / 4.77GB            
Processing Files (0 / 1)      :  66%|██████▌   | 3.13GB / 4.77GB,  196MB/s  

New Data Upload               :  95%|█████████▌| 3.13GB / 3.29GB,  196MB/s  


  ...e-03b89ee9/train_state_15:  68%|██████▊   | 3.23GB / 4.77GB            
Processing Files (0 / 1)      :  68%|██████▊   | 3.23GB / 4.77GB,  200MB/s  

New Data Upload               :  96%|█████████▋| 3.23GB / 3.35GB,  200MB/s  


  ...e-03b89ee9/train_state_15:  70%|██████▉   | 3.33GB / 4.77GB            
Processing Files (0 / 1)      :  70%|██████▉   | 3.33GB / 4.77GB,  204MB/s  

New Data Upload               :  96%|█████████▌| 3.33GB / 3.49GB,  204MB/s  


  ...e-03b89ee9/train_state_15:  71%|███████   | 3.38GB / 4.77GB            
Processing Files (0 / 1)      :  71%|███████   | 3.38GB / 4.77GB,  203MB/s  

New Data Upload               :  95%|█████████▌| 3.38GB / 3.55GB,  203MB/s  


  ...e-03b89ee9/train_state_15:  73%|███████▎  | 3.49GB / 4.77GB            
Processing Files (0 / 1)      :  73%|███████▎  | 3.49GB / 4.77GB,  208MB/s  

New Data Upload               :  95%|█████████▍| 3.49GB / 3.69GB,  208MB/s  


  ...e-03b89ee9/train_state_15:  76%|███████▌  | 3.62GB / 4.77GB            
Processing Files (0 / 1)      :  76%|███████▌  | 3.62GB / 4.77GB,  215MB/s  

New Data Upload               :  96%|█████████▋| 3.62GB / 3.76GB,  215MB/s  


  ...e-03b89ee9/train_state_15:  78%|███████▊  | 3.73GB / 4.77GB            
Processing Files (0 / 1)      :  78%|███████▊  | 3.73GB / 4.77GB,  219MB/s  

New Data Upload               :  96%|█████████▌| 3.73GB / 3.89GB,  219MB/s  


  ...e-03b89ee9/train_state_15:  80%|████████  | 3.83GB / 4.77GB            
Processing Files (0 / 1)      :  80%|████████  | 3.83GB / 4.77GB,  223MB/s  

New Data Upload               :  97%|█████████▋| 3.83GB / 3.96GB,  223MB/s  


  ...e-03b89ee9/train_state_15:  82%|████████▏ | 3.93GB / 4.77GB            
Processing Files (0 / 1)      :  82%|████████▏ | 3.93GB / 4.77GB,  226MB/s  

New Data Upload               :  96%|█████████▌| 3.93GB / 4.09GB,  226MB/s  


  ...e-03b89ee9/train_state_15:  85%|████████▍ | 4.03GB / 4.77GB            
Processing Files (0 / 1)      :  85%|████████▍ | 4.03GB / 4.77GB,  230MB/s  

New Data Upload               :  97%|█████████▋| 4.03GB / 4.16GB,  230MB/s  


  ...e-03b89ee9/train_state_15:  86%|████████▌ | 4.11GB / 4.77GB            
Processing Files (0 / 1)      :  86%|████████▌ | 4.11GB / 4.77GB,  231MB/s  

New Data Upload               :  96%|█████████▌| 4.11GB / 4.29GB,  231MB/s  


  ...e-03b89ee9/train_state_15:  88%|████████▊ | 4.21GB / 4.77GB            
Processing Files (0 / 1)      :  88%|████████▊ | 4.21GB / 4.77GB,  235MB/s  

New Data Upload               :  97%|█████████▋| 4.21GB / 4.36GB,  235MB/s  


  ...e-03b89ee9/train_state_15:  91%|█████████ | 4.32GB / 4.77GB            
Processing Files (0 / 1)      :  91%|█████████ | 4.32GB / 4.77GB,  239MB/s  

New Data Upload               :  96%|█████████▌| 4.32GB / 4.49GB,  239MB/s  


  ...e-03b89ee9/train_state_15:  93%|█████████▎| 4.45GB / 4.77GB            
Processing Files (0 / 1)      :  93%|█████████▎| 4.45GB / 4.77GB,  244MB/s  

New Data Upload               :  97%|█████████▋| 4.45GB / 4.56GB,  244MB/s  


  ...e-03b89ee9/train_state_15:  95%|█████████▌| 4.54GB / 4.77GB            
Processing Files (0 / 1)      :  95%|█████████▌| 4.54GB / 4.77GB,  247MB/s  

New Data Upload               :  97%|█████████▋| 4.54GB / 4.69GB,  247MB/s  


  ...e-03b89ee9/train_state_15:  97%|█████████▋| 4.62GB / 4.77GB            
Processing Files (0 / 1)      :  97%|█████████▋| 4.62GB / 4.77GB,  248MB/s  

New Data Upload               :  97%|█████████▋| 4.62GB / 4.76GB,  248MB/s  


  ...e-03b89ee9/train_state_15:  99%|█████████▉| 4.73GB / 4.77GB            
Processing Files (0 / 1)      :  99%|█████████▉| 4.73GB / 4.77GB,  252MB/s  

New Data Upload               :  99%|█████████▉| 4.73GB / 4.77GB,  252MB/s  


  ...e-03b89ee9/train_state_15: 100%|█████████▉| 4.76GB / 4.77GB            
Processing Files (0 / 1)      : 100%|█████████▉| 4.76GB / 4.77GB,  250MB/s  

New Data Upload               : 100%|█████████▉| 4.76GB / 4.77GB,  250MB/s  


  ...e-03b89ee9/train_state_15: 100%|█████████▉| 4.77GB / 4.77GB            
Processing Files (0 / 1)      : 100%|█████████▉| 4.77GB / 4.77GB,  245MB/s  

New Data Upload               : 100%|█████████▉| 4.77GB / 4.77GB,  245MB/s  


  ...e-03b89ee9/train_state_15: 100%|█████████▉| 4.77GB / 4.77GB            
Processing Files (0 / 1)      : 100%|█████████▉| 4.77GB / 4.77GB,  240MB/s  

New Data Upload               : 100%|█████████▉| 4.77GB / 4.77GB,  240MB/s  


  ...e-03b89ee9/train_state_15: 100%|█████████▉| 4.77GB / 4.77GB            
Processing Files (0 / 1)      : 100%|█████████▉| 4.77GB / 4.77GB,  231MB/s  

New Data Upload               : 100%|██████████| 4.77GB / 4.77GB,  231MB/s  
Processing Files (0 / 1)      : 100%|█████████▉| 4.77GB / 4.77GB,  231MB/s  

New Data Upload               : 100%|██████████| 4.77GB / 4.77GB,  231MB/s  

  ...e-03b89ee9/train_state_15: 100%|█████████▉| 4.77GB / 4.77GB            
Uploaded train_state_15 to seyedparsa/thoughtformer-checkpoints/checkpoints/star-datacurr-only-4gpu-bs32-lr1e-5_n1000_qwen3-0.6b-base_thoughtformer_lr1e-5_no-reset_dc_no-tc_adaptive-03b89ee9/train_state_15
Deleted old checkpoint: checkpoint_14
Deleted old checkpoint: train_state_14

Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]
Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]
Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]eval loss 0.03164251262322068

Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]
[CORRECT] Example 0 (diff=3)
  Expected : 'Claude'
  Extracted: 'Claude'
  Generated: '### Claude'

[CORRECT] Example 1 (diff=8)
  Expected : 'Guy'
  Extracted: 'Guy'
  Generated: '### Guy'

Test Accuracy:   3%|▎         | 1/32 [00:06<03:31,  6.83s/it]
Test accuracy: 1.0:   3%|▎         | 1/32 [00:06<03:31,  6.83s/it]
[CORRECT] Example 2 (diff=10)
  Expected : 'Elvis'
  Extracted: 'Elvis'
  Generated: '### Elvis'

Test Accuracy:   3%|▎         | 1/32 [00:06<03:31,  6.83s/it]
Test Accuracy:   3%|▎         | 1/32 [00:06<03:31,  6.83s/it]
Test accuracy: 1.0:   3%|▎         | 1/32 [00:06<03:31,  6.83s/it]
Test accuracy: 0.88:   3%|▎         | 1/32 [00:06<03:31,  6.83s/it]
[CORRECT] Example 3 (diff=8)
  Expected : 'Coral'
  Extracted: 'Coral'
  Generated: '### Coral'

[CORRECT] Example 4 (diff=1)
  Expected : 'Patricia'
  Extracted: 'Patricia'
  Generated: '### Patricia'

Test Accuracy:   3%|▎         | 1/32 [00:06<03:26,  6.66s/it]
Test accuracy: 1.0:   3%|▎         | 1/32 [00:06<03:26,  6.66s/it]
Test accuracy: 1.0:   6%|▋         | 2/32 [00:12<03:12,  6.43s/it]
Test accuracy: 1.0:   6%|▋         | 2/32 [00:12<03:12,  6.43s/it]
Test accuracy: 0.88:   6%|▋         | 2/32 [00:12<03:12,  6.43s/it]
Test accuracy: 1.0:   6%|▋         | 2/32 [00:12<03:12,  6.43s/it]
Test accuracy: 0.94:   6%|▋         | 2/32 [00:12<03:12,  6.43s/it]
Test accuracy: 0.94:   6%|▋         | 2/32 [00:12<03:12,  6.43s/it]
Test accuracy: 1.0:   6%|▋         | 2/32 [00:12<03:10,  6.35s/it]
Test accuracy: 1.0:   6%|▋         | 2/32 [00:12<03:10,  6.35s/it]
Test accuracy: 0.94:   9%|▉         | 3/32 [00:20<03:16,  6.79s/it]
Test accuracy: 1.0:   9%|▉         | 3/32 [00:20<03:16,  6.79s/it]
Test accuracy: 0.96:   9%|▉         | 3/32 [00:20<03:16,  6.79s/it]
Test accuracy: 0.96:   9%|▉         | 3/32 [00:20<03:16,  6.79s/it]
Test accuracy: 1.0:   9%|▉         | 3/32 [00:20<03:15,  6.75s/it]
Test accuracy: 1.0:   9%|▉         | 3/32 [00:20<03:15,  6.75s/it]
Test accuracy: 0.94:   9%|▉         | 3/32 [00:20<03:16,  6.79s/it]
Test accuracy: 0.96:   9%|▉         | 3/32 [00:20<03:16,  6.79s/it]
Test accuracy: 0.96:  12%|█▎        | 4/32 [00:26<03:02,  6.53s/it]
Test accuracy: 0.96:  12%|█▎        | 4/32 [00:26<03:02,  6.53s/it]
Test accuracy: 0.96:  12%|█▎        | 4/32 [00:26<03:02,  6.53s/it]
Test accuracy: 1.0:  12%|█▎        | 4/32 [00:26<03:02,  6.51s/it]
Test accuracy: 0.84:  12%|█▎        | 4/32 [00:26<03:02,  6.53s/it]
Test accuracy: 0.91:  12%|█▎        | 4/32 [00:26<03:02,  6.53s/it]
Test accuracy: 0.91:  12%|█▎        | 4/32 [00:26<03:02,  6.53s/it]
Test accuracy: 1.0:  12%|█▎        | 4/32 [00:26<03:02,  6.51s/it]
Test accuracy: 0.84:  16%|█▌        | 5/32 [00:32<02:54,  6.46s/it]
Test accuracy: 0.91:  16%|█▌        | 5/32 [00:32<02:54,  6.46s/it]
Test accuracy: 1.0:  16%|█▌        | 5/32 [00:32<02:54,  6.45s/it]
Test accuracy: 0.91:  16%|█▌        | 5/32 [00:32<02:54,  6.46s/it]
Test accuracy: 0.82:  16%|█▌        | 5/32 [00:32<02:54,  6.46s/it]
Test accuracy: 0.93:  16%|█▌        | 5/32 [00:32<02:54,  6.46s/it]
Test accuracy: 0.9:  16%|█▌        | 5/32 [00:32<02:54,  6.46s/it] 
Test accuracy: 1.0:  16%|█▌        | 5/32 [00:32<02:54,  6.45s/it]
Test accuracy: 1.0:  19%|█▉        | 6/32 [00:38<02:46,  6.40s/it]
Test accuracy: 0.93:  19%|█▉        | 6/32 [00:38<02:46,  6.41s/it]
Test accuracy: 0.82:  19%|█▉        | 6/32 [00:38<02:46,  6.41s/it]
Test accuracy: 0.9:  19%|█▉        | 6/32 [00:38<02:46,  6.41s/it]
Test accuracy: 0.92:  19%|█▉        | 6/32 [00:38<02:46,  6.41s/it]
Test accuracy: 0.83:  19%|█▉        | 6/32 [00:38<02:46,  6.41s/it]
Test accuracy: 1.0:  19%|█▉        | 6/32 [00:38<02:46,  6.40s/it]
Test accuracy: 0.9:  19%|█▉        | 6/32 [00:38<02:46,  6.41s/it]
Test accuracy: 0.83:  22%|██▏       | 7/32 [00:45<02:37,  6.32s/it]
Test accuracy: 0.9:  22%|██▏       | 7/32 [00:45<02:37,  6.32s/it]
Test accuracy: 0.92:  22%|██▏       | 7/32 [00:45<02:37,  6.32s/it]
Test accuracy: 0.84:  22%|██▏       | 7/32 [00:45<02:37,  6.32s/it]
Test accuracy: 0.89:  22%|██▏       | 7/32 [00:45<02:37,  6.32s/it]
Test accuracy: 0.89:  22%|██▏       | 7/32 [00:45<02:37,  6.32s/it]
Test accuracy: 1.0:  22%|██▏       | 7/32 [00:44<02:37,  6.31s/it]
Test accuracy: 0.98:  22%|██▏       | 7/32 [00:44<02:37,  6.31s/it]
Test accuracy: 0.84:  25%|██▌       | 8/32 [00:51<02:31,  6.32s/it]
Test accuracy: 0.89:  25%|██▌       | 8/32 [00:51<02:31,  6.32s/it]
Test accuracy: 0.89:  25%|██▌       | 8/32 [00:51<02:31,  6.32s/it]
Test accuracy: 0.83:  25%|██▌       | 8/32 [00:51<02:31,  6.32s/it]
Test accuracy: 0.91:  25%|██▌       | 8/32 [00:51<02:31,  6.32s/it]
Test accuracy: 0.98:  25%|██▌       | 8/32 [00:51<02:31,  6.32s/it]
Test accuracy: 0.91:  25%|██▌       | 8/32 [00:51<02:31,  6.32s/it]
Test accuracy: 0.98:  25%|██▌       | 8/32 [00:51<02:31,  6.32s/it]
Test accuracy: 0.98:  28%|██▊       | 9/32 [00:58<02:29,  6.51s/it]
Test accuracy: 0.91:  28%|██▊       | 9/32 [00:58<02:29,  6.52s/it]
Test accuracy: 0.91:  28%|██▊       | 9/32 [00:58<02:29,  6.52s/it]
Test accuracy: 0.92:  28%|██▊       | 9/32 [00:58<02:29,  6.52s/it]
Test accuracy: 0.89:  28%|██▊       | 9/32 [00:58<02:29,  6.52s/it]
Test accuracy: 0.99:  28%|██▊       | 9/32 [00:58<02:29,  6.51s/it]
Test accuracy: 0.83:  28%|██▊       | 9/32 [00:58<02:29,  6.52s/it]
Test accuracy: 0.83:  28%|██▊       | 9/32 [00:58<02:29,  6.52s/it]
Test accuracy: 0.83:  31%|███▏      | 10/32 [01:04<02:20,  6.40s/it]
Test accuracy: 0.92:  31%|███▏      | 10/32 [01:04<02:20,  6.40s/it]
Test accuracy: 0.85:  31%|███▏      | 10/32 [01:04<02:20,  6.40s/it]
Test accuracy: 0.93:  31%|███▏      | 10/32 [01:04<02:20,  6.40s/it]
Test accuracy: 0.89:  31%|███▏      | 10/32 [01:04<02:20,  6.40s/it]
Test accuracy: 0.99:  31%|███▏      | 10/32 [01:04<02:20,  6.40s/it]
Test accuracy: 0.89:  31%|███▏      | 10/32 [01:04<02:20,  6.40s/it]
Test accuracy: 0.99:  31%|███▏      | 10/32 [01:04<02:20,  6.40s/it]
Test accuracy: 0.85:  34%|███▍      | 11/32 [01:10<02:13,  6.38s/it]
Test accuracy: 0.89:  34%|███▍      | 11/32 [01:10<02:14,  6.38s/it]
Test accuracy: 0.93:  34%|███▍      | 11/32 [01:10<02:14,  6.38s/it]
Test accuracy: 0.86:  34%|███▍      | 11/32 [01:10<02:13,  6.38s/it]
Test accuracy: 0.99:  34%|███▍      | 11/32 [01:10<02:13,  6.38s/it]
Test accuracy: 0.89:  34%|███▍      | 11/32 [01:10<02:14,  6.38s/it]
Test accuracy: 0.92:  34%|███▍      | 11/32 [01:10<02:14,  6.38s/it]
Test accuracy: 0.98:  34%|███▍      | 11/32 [01:10<02:13,  6.38s/it]
Test accuracy: 0.92:  38%|███▊      | 12/32 [01:17<02:07,  6.36s/it]
Test accuracy: 0.92:  38%|███▊      | 12/32 [01:17<02:07,  6.36s/it]
Test accuracy: 0.86:  38%|███▊      | 12/32 [01:17<02:07,  6.36s/it]
Test accuracy: 0.98:  38%|███▊      | 12/32 [01:16<02:07,  6.36s/it]
Test accuracy: 0.88:  38%|███▊      | 12/32 [01:17<02:07,  6.36s/it]
Test accuracy: 0.98:  38%|███▊      | 12/32 [01:16<02:07,  6.36s/it]
Test accuracy: 0.89:  38%|███▊      | 12/32 [01:17<02:07,  6.36s/it]
Test accuracy: 0.89:  38%|███▊      | 12/32 [01:17<02:07,  6.36s/it]
Test accuracy: 0.88:  41%|████      | 13/32 [01:23<01:59,  6.29s/it]
Test accuracy: 0.89:  41%|████      | 13/32 [01:23<01:59,  6.29s/it]
Test accuracy: 0.92:  41%|████      | 13/32 [01:23<01:59,  6.29s/it]
Test accuracy: 0.98:  41%|████      | 13/32 [01:23<01:59,  6.29s/it]
Test accuracy: 0.88:  41%|████      | 13/32 [01:23<01:59,  6.29s/it]
Test accuracy: 0.87:  41%|████      | 13/32 [01:23<01:59,  6.29s/it]
Test accuracy: 0.91:  41%|████      | 13/32 [01:23<01:59,  6.29s/it]
Test accuracy: 0.97:  41%|████      | 13/32 [01:23<01:59,  6.29s/it]
Test accuracy: 0.87:  44%|████▍     | 14/32 [01:29<01:53,  6.31s/it]
Test accuracy: 0.91:  44%|████▍     | 14/32 [01:29<01:53,  6.31s/it]
Test accuracy: 0.88:  44%|████▍     | 14/32 [01:29<01:53,  6.31s/it]
Test accuracy: 0.92:  44%|████▍     | 14/32 [01:29<01:53,  6.31s/it]
Test accuracy: 0.88:  44%|████▍     | 14/32 [01:29<01:53,  6.31s/it]
Test accuracy: 0.89:  44%|████▍     | 14/32 [01:29<01:53,  6.31s/it]
Test accuracy: 0.97:  44%|████▍     | 14/32 [01:29<01:53,  6.31s/it]
Test accuracy: 0.96:  44%|████▍     | 14/32 [01:29<01:53,  6.31s/it]
Test accuracy: 0.88:  47%|████▋     | 15/32 [01:35<01:47,  6.31s/it]
Test accuracy: 0.92:  47%|████▋     | 15/32 [01:35<01:47,  6.31s/it]
Test accuracy: 0.86:  47%|████▋     | 15/32 [01:35<01:47,  6.31s/it]
Test accuracy: 0.92:  47%|████▋     | 15/32 [01:35<01:47,  6.31s/it]
Test accuracy: 0.89:  47%|████▋     | 15/32 [01:35<01:47,  6.31s/it]
Test accuracy: 0.9:  47%|████▋     | 15/32 [01:35<01:47,  6.31s/it] 
Test accuracy: 0.96:  47%|████▋     | 15/32 [01:35<01:47,  6.31s/it]
Test accuracy: 0.96:  47%|████▋     | 15/32 [01:35<01:47,  6.31s/it]
Test accuracy: 0.86:  50%|█████     | 16/32 [01:42<01:41,  6.32s/it]
Test accuracy: 0.9:  50%|█████     | 16/32 [01:42<01:41,  6.32s/it]
Test accuracy: 0.92:  50%|█████     | 16/32 [01:42<01:41,  6.32s/it]
Test accuracy: 0.86:  50%|█████     | 16/32 [01:42<01:41,  6.32s/it]
Test accuracy: 0.9:  50%|█████     | 16/32 [01:42<01:41,  6.32s/it]
Test accuracy: 0.92:  50%|█████     | 16/32 [01:42<01:41,  6.32s/it]
Test accuracy: 0.96:  50%|█████     | 16/32 [01:42<01:41,  6.32s/it]
Test accuracy: 0.96:  50%|█████     | 16/32 [01:42<01:41,  6.32s/it]
Test accuracy: 0.86:  53%|█████▎    | 17/32 [01:48<01:34,  6.27s/it]
Test accuracy: 0.96:  53%|█████▎    | 17/32 [01:48<01:34,  6.27s/it]
Test accuracy: 0.86:  53%|█████▎    | 17/32 [01:48<01:34,  6.27s/it]
Test accuracy: 0.92:  53%|█████▎    | 17/32 [01:48<01:34,  6.27s/it]
Test accuracy: 0.93:  53%|█████▎    | 17/32 [01:48<01:34,  6.27s/it]
Test accuracy: 0.96:  53%|█████▎    | 17/32 [01:48<01:34,  6.27s/it]
Test accuracy: 0.9:  53%|█████▎    | 17/32 [01:48<01:34,  6.27s/it]
Test accuracy: 0.9:  53%|█████▎    | 17/32 [01:48<01:34,  6.27s/it]
Test accuracy: 0.86:  56%|█████▋    | 18/32 [01:54<01:28,  6.29s/it]
Test accuracy: 0.86:  56%|█████▋    | 18/32 [01:54<01:28,  6.29s/it]
Test accuracy: 0.93:  56%|█████▋    | 18/32 [01:54<01:28,  6.29s/it]
Test accuracy: 0.93:  56%|█████▋    | 18/32 [01:54<01:28,  6.29s/it]
Test accuracy: 0.96:  56%|█████▋    | 18/32 [01:54<01:28,  6.29s/it]
Test accuracy: 0.96:  56%|█████▋    | 18/32 [01:54<01:28,  6.29s/it]
Test accuracy: 0.9:  56%|█████▋    | 18/32 [01:54<01:28,  6.29s/it]
Test accuracy: 0.91:  56%|█████▋    | 18/32 [01:54<01:28,  6.29s/it]
Test accuracy: 0.93:  59%|█████▉    | 19/32 [02:01<01:22,  6.31s/it]
Test accuracy: 0.86:  59%|█████▉    | 19/32 [02:01<01:22,  6.31s/it]
Test accuracy: 0.85:  59%|█████▉    | 19/32 [02:01<01:22,  6.31s/it]
Test accuracy: 0.93:  59%|█████▉    | 19/32 [02:01<01:22,  6.31s/it]
Test accuracy: 0.91:  59%|█████▉    | 19/32 [02:01<01:22,  6.31s/it]
Test accuracy: 0.9:  59%|█████▉    | 19/32 [02:01<01:22,  6.31s/it] 
Test accuracy: 0.96:  59%|█████▉    | 19/32 [02:00<01:22,  6.31s/it]
Test accuracy: 0.96:  59%|█████▉    | 19/32 [02:00<01:22,  6.31s/it]
Test accuracy: 0.96:  62%|██████▎   | 20/32 [02:07<01:15,  6.26s/it]
Test accuracy: 0.93:  62%|██████▎   | 20/32 [02:07<01:15,  6.26s/it]
Test accuracy: 0.9:  62%|██████▎   | 20/32 [02:07<01:15,  6.26s/it]
Test accuracy: 0.85:  62%|██████▎   | 20/32 [02:07<01:15,  6.26s/it]
Test accuracy: 0.9:  62%|██████▎   | 20/32 [02:07<01:15,  6.26s/it]
Test accuracy: 0.93:  62%|██████▎   | 20/32 [02:07<01:15,  6.26s/it]
Test accuracy: 0.96:  62%|██████▎   | 20/32 [02:07<01:15,  6.26s/it]
Test accuracy: 0.85:  62%|██████▎   | 20/32 [02:07<01:15,  6.26s/it]
Test accuracy: 0.85:  66%|██████▌   | 21/32 [02:14<01:11,  6.49s/it]
Test accuracy: 0.93:  66%|██████▌   | 21/32 [02:14<01:11,  6.49s/it]
Test accuracy: 0.96:  66%|██████▌   | 21/32 [02:14<01:11,  6.49s/it]
Test accuracy: 0.93:  66%|██████▌   | 21/32 [02:14<01:11,  6.49s/it]
Test accuracy: 0.85:  66%|██████▌   | 21/32 [02:14<01:11,  6.49s/it]
Test accuracy: 0.95:  66%|██████▌   | 21/32 [02:14<01:11,  6.49s/it]
Test accuracy: 0.9:  66%|██████▌   | 21/32 [02:14<01:11,  6.50s/it]
Test accuracy: 0.9:  66%|██████▌   | 21/32 [02:14<01:11,  6.50s/it]
Test accuracy: 0.9:  69%|██████▉   | 22/32 [02:20<01:04,  6.45s/it]
Test accuracy: 0.85:  69%|██████▉   | 22/32 [02:20<01:04,  6.45s/it]
Test accuracy: 0.93:  69%|██████▉   | 22/32 [02:20<01:04,  6.45s/it]
Test accuracy: 0.95:  69%|██████▉   | 22/32 [02:20<01:04,  6.45s/it]
Test accuracy: 0.89:  69%|██████▉   | 22/32 [02:20<01:04,  6.45s/it]
Test accuracy: 0.86:  69%|██████▉   | 22/32 [02:20<01:04,  6.45s/it]
Test accuracy: 0.93:  69%|██████▉   | 22/32 [02:20<01:04,  6.45s/it]
Test accuracy: 0.95:  69%|██████▉   | 22/32 [02:20<01:04,  6.45s/it]
Test accuracy: 0.86:  72%|███████▏  | 23/32 [02:26<00:57,  6.36s/it]
Test accuracy: 0.95:  72%|███████▏  | 23/32 [02:26<00:57,  6.36s/it]
Test accuracy: 0.93:  72%|███████▏  | 23/32 [02:26<00:57,  6.36s/it]
Test accuracy: 0.86:  72%|███████▏  | 23/32 [02:26<00:57,  6.36s/it]
Test accuracy: 0.93:  72%|███████▏  | 23/32 [02:26<00:57,  6.36s/it]
Test accuracy: 0.95:  72%|███████▏  | 23/32 [02:26<00:57,  6.36s/it]
Test accuracy: 0.89:  72%|███████▏  | 23/32 [02:26<00:57,  6.36s/it]
Test accuracy: 0.89:  72%|███████▏  | 23/32 [02:26<00:57,  6.36s/it]
Test accuracy: 0.86:  75%|███████▌  | 24/32 [02:33<00:52,  6.55s/it]
Test accuracy: 0.89:  75%|███████▌  | 24/32 [02:33<00:52,  6.55s/it]
Test accuracy: 0.93:  75%|███████▌  | 24/32 [02:33<00:52,  6.55s/it]
Test accuracy: 0.86:  75%|███████▌  | 24/32 [02:33<00:52,  6.55s/it]
Test accuracy: 0.9:  75%|███████▌  | 24/32 [02:33<00:52,  6.55s/it] 
Test accuracy: 0.93:  75%|███████▌  | 24/32 [02:33<00:52,  6.55s/it]
Test accuracy: 0.95:  75%|███████▌  | 24/32 [02:33<00:52,  6.55s/it]
Test accuracy: 0.94:  75%|███████▌  | 24/32 [02:33<00:52,  6.55s/it]
Test accuracy: 0.93:  78%|███████▊  | 25/32 [02:40<00:45,  6.49s/it]
Test accuracy: 0.86:  78%|███████▊  | 25/32 [02:40<00:45,  6.49s/it]
Test accuracy: 0.93:  78%|███████▊  | 25/32 [02:40<00:45,  6.49s/it]
Test accuracy: 0.87:  78%|███████▊  | 25/32 [02:40<00:45,  6.49s/it]
Test accuracy: 0.9:  78%|███████▊  | 25/32 [02:40<00:45,  6.49s/it]
Test accuracy: 0.94:  78%|███████▊  | 25/32 [02:40<00:45,  6.49s/it]
Test accuracy: 0.9:  78%|███████▊  | 25/32 [02:40<00:45,  6.49s/it]
Test accuracy: 0.94:  78%|███████▊  | 25/32 [02:40<00:45,  6.49s/it]
Test accuracy: 0.87:  81%|████████▏ | 26/32 [02:46<00:38,  6.44s/it]
Test accuracy: 0.93:  81%|████████▏ | 26/32 [02:46<00:38,  6.44s/it]
Test accuracy: 0.94:  81%|████████▏ | 26/32 [02:46<00:38,  6.44s/it]
Test accuracy: 0.9:  81%|████████▏ | 26/32 [02:46<00:38,  6.44s/it]
Test accuracy: 0.86:  81%|████████▏ | 26/32 [02:46<00:38,  6.44s/it]
Test accuracy: 0.92:  81%|████████▏ | 26/32 [02:46<00:38,  6.44s/it]
Test accuracy: 0.9:  81%|████████▏ | 26/32 [02:46<00:38,  6.44s/it]
Test accuracy: 0.94:  81%|████████▏ | 26/32 [02:46<00:38,  6.44s/it]
Test accuracy: 0.86:  84%|████████▍ | 27/32 [02:52<00:32,  6.41s/it]
Test accuracy: 0.9:  84%|████████▍ | 27/32 [02:52<00:32,  6.41s/it]
Test accuracy: 0.86:  84%|████████▍ | 27/32 [02:52<00:32,  6.41s/it]
Test accuracy: 0.92:  84%|████████▍ | 27/32 [02:52<00:32,  6.42s/it]
Test accuracy: 0.89:  84%|████████▍ | 27/32 [02:52<00:32,  6.41s/it]
Test accuracy: 0.93:  84%|████████▍ | 27/32 [02:52<00:32,  6.42s/it]
Test accuracy: 0.94:  84%|████████▍ | 27/32 [02:52<00:32,  6.41s/it]
Test accuracy: 0.94:  84%|████████▍ | 27/32 [02:52<00:32,  6.41s/it]
Test accuracy: 0.86:  88%|████████▊ | 28/32 [02:59<00:26,  6.55s/it]
Test accuracy: 0.93:  88%|████████▊ | 28/32 [02:59<00:26,  6.55s/it]
Test accuracy: 0.86:  88%|████████▊ | 28/32 [02:59<00:26,  6.55s/it]
Test accuracy: 0.94:  88%|████████▊ | 28/32 [02:59<00:26,  6.55s/it]
Test accuracy: 0.93:  88%|████████▊ | 28/32 [02:59<00:26,  6.55s/it]
Test accuracy: 0.94:  88%|████████▊ | 28/32 [02:59<00:26,  6.55s/it]
Test accuracy: 0.89:  88%|████████▊ | 28/32 [02:59<00:26,  6.55s/it]
Test accuracy: 0.89:  88%|████████▊ | 28/32 [02:59<00:26,  6.55s/it]
Test accuracy: 0.94:  91%|█████████ | 29/32 [03:05<00:19,  6.49s/it]
Test accuracy: 0.86:  91%|█████████ | 29/32 [03:06<00:19,  6.49s/it]
Test accuracy: 0.93:  91%|█████████ | 29/32 [03:06<00:19,  6.49s/it]
Test accuracy: 0.86:  91%|█████████ | 29/32 [03:06<00:19,  6.49s/it]
Test accuracy: 0.93:  91%|█████████ | 29/32 [03:06<00:19,  6.49s/it]
Test accuracy: 0.93:  91%|█████████ | 29/32 [03:05<00:19,  6.49s/it]
Test accuracy: 0.89:  91%|█████████ | 29/32 [03:06<00:19,  6.49s/it]
Test accuracy: 0.9:  91%|█████████ | 29/32 [03:06<00:19,  6.49s/it] 
Test accuracy: 0.86:  94%|█████████▍| 30/32 [03:12<00:12,  6.45s/it]
Test accuracy: 0.93:  94%|█████████▍| 30/32 [03:12<00:12,  6.45s/it]
Test accuracy: 0.93:  94%|█████████▍| 30/32 [03:12<00:12,  6.45s/it]
Test accuracy: 0.86:  94%|█████████▍| 30/32 [03:12<00:12,  6.45s/it]
Test accuracy: 0.93:  94%|█████████▍| 30/32 [03:12<00:12,  6.45s/it]
Test accuracy: 0.9:  94%|█████████▍| 30/32 [03:12<00:12,  6.45s/it]
Test accuracy: 0.93:  94%|█████████▍| 30/32 [03:12<00:12,  6.45s/it]
Test accuracy: 0.89:  94%|█████████▍| 30/32 [03:12<00:12,  6.45s/it]
Test accuracy: 0.93:  97%|█████████▋| 31/32 [03:19<00:06,  6.57s/it]
Test accuracy: 0.86:  97%|█████████▋| 31/32 [03:19<00:06,  6.57s/it]
Test accuracy: 0.93:  97%|█████████▋| 31/32 [03:19<00:06,  6.57s/it]
Test accuracy: 0.93:  97%|█████████▋| 31/32 [03:19<00:06,  6.57s/it]
Test accuracy: 0.86:  97%|█████████▋| 31/32 [03:19<00:06,  6.57s/it]
Test accuracy: 0.92:  97%|█████████▋| 31/32 [03:19<00:06,  6.57s/it]
Test accuracy: 0.89:  97%|█████████▋| 31/32 [03:19<00:06,  6.58s/it]
Test accuracy: 0.89:  97%|█████████▋| 31/32 [03:19<00:06,  6.58s/it]
Test accuracy: 0.92: 100%|██████████| 32/32 [03:21<00:00,  5.28s/it]
Test accuracy: 0.92: 100%|██████████| 32/32 [03:21<00:00,  5.28s/it]
Test accuracy: 0.89: 100%|██████████| 32/32 [03:21<00:00,  5.28s/it]
Test accuracy: 0.86: 100%|██████████| 32/32 [03:21<00:00,  5.28s/it]
Test accuracy: 0.93: 100%|██████████| 32/32 [03:21<00:00,  5.28s/it]
Test accuracy: 0.89: 100%|██████████| 32/32 [03:21<00:00,  5.28s/it]
Test accuracy: 0.86: 100%|██████████| 32/32 [03:21<00:00,  5.28s/it]
Test accuracy: 0.93: 100%|██████████| 32/32 [03:21<00:00,  5.28s/it]
Test accuracy: 0.89: 100%|██████████| 32/32 [03:21<00:00,  6.30s/it]
Device 2: Cor=222, CoT=0, Total=250

Test accuracy: 0.86: 100%|██████████| 32/32 [03:21<00:00,  6.30s/it]
Device 3: Cor=216, CoT=0, Total=250

Test accuracy: 0.93: 100%|██████████| 32/32 [03:21<00:00,  6.30s/it]
Device 1: Cor=233, CoT=0, Total=250

Test accuracy: 0.92: 100%|██████████| 32/32 [03:21<00:00,  6.30s/it]
Device 0: Cor=231, CoT=0, Total=250

============================================================
EVALUATION RESULTS (Epoch 14)
============================================================
Config: data_curriculum=ON, thought_curriculum=OFF
thought_stage=10, data_stage=8, n_latent_tokens=10
Training on difficulties: 0-8

Overall accuracy: 902 / 1000 = 0.9020

Per-difficulty accuracy:
  diff=1: 99/100 = 0.9900 (trained)
  diff=2: 99/100 = 0.9900 (trained)
  diff=3: 100/100 = 1.0000 (trained)
  diff=4: 99/100 = 0.9900 (trained)
  diff=5: 97/100 = 0.9700 (trained)
  diff=6: 100/100 = 1.0000 (trained)
  diff=7: 96/100 = 0.9600 (trained)
  diff=8: 94/100 = 0.9400 (trained)
  diff=9: 62/100 = 0.6200 (untrained)
  diff=10: 56/100 = 0.5600 (untrained)

Trained (diff 0-8): 784/800 = 0.9800
============================================================


*** ADAPTIVE STAGING: acc_trained=0.9800, acc_diff8=0.9400 >= 0.9 ***
*** Advancing to stage 9 ***


Processing Files (0 / 0)      : |          |  0.00B /  0.00B            

New Data Upload               : |          |  0.00B /  0.00B            


  ...ve-03b89ee9/checkpoint_15:  10%|▉         |  352MB / 3.63GB            


  ...ve-03b89ee9/checkpoint_15:  10%|▉         |  352MB / 3.63GB            
Processing Files (0 / 1)      :  10%|▉         |  352MB / 3.63GB,   ???B/s  


  ...ve-03b89ee9/checkpoint_15:  20%|██        |  728MB / 3.63GB            
Processing Files (0 / 1)      :  20%|██        |  728MB / 3.63GB,   ???B/s  


  ...ve-03b89ee9/checkpoint_15:  30%|███       | 1.10GB / 3.63GB            
Processing Files (0 / 1)      :  30%|███       | 1.10GB / 3.63GB,  106MB/s  


  ...ve-03b89ee9/checkpoint_15:  41%|████      | 1.47GB / 3.63GB            
Processing Files (0 / 1)      :  41%|████      | 1.47GB / 3.63GB,  139MB/s  


  ...ve-03b89ee9/checkpoint_15:  51%|█████     | 1.86GB / 3.63GB            
Processing Files (0 / 1)      :  51%|█████     | 1.86GB / 3.63GB,  173MB/s  


  ...ve-03b89ee9/checkpoint_15:  62%|██████▏   | 2.23GB / 3.63GB            
Processing Files (0 / 1)      :  62%|██████▏   | 2.23GB / 3.63GB,  205MB/s  


  ...ve-03b89ee9/checkpoint_15:  72%|███████▏  | 2.61GB / 3.63GB            
Processing Files (0 / 1)      :  72%|███████▏  | 2.61GB / 3.63GB,  236MB/s  


  ...ve-03b89ee9/checkpoint_15:  82%|████████▏ | 2.98GB / 3.63GB            
Processing Files (0 / 1)      :  82%|████████▏ | 2.98GB / 3.63GB,  266MB/s  


  ...ve-03b89ee9/checkpoint_15:  92%|█████████▏| 3.35GB / 3.63GB            
Processing Files (0 / 1)      :  92%|█████████▏| 3.35GB / 3.63GB,  294MB/s  


  ...ve-03b89ee9/checkpoint_15: 100%|██████████| 3.63GB / 3.63GB            
Processing Files (1 / 1)      : 100%|██████████| 3.63GB / 3.63GB,  313MB/s  
Processing Files (1 / 1)      : 100%|██████████| 3.63GB / 3.63GB,  313MB/s  

New Data Upload               : |          |  0.00B /  0.00B,  0.00B/s  

  ...ve-03b89ee9/checkpoint_15: 100%|██████████| 3.63GB / 3.63GB            
No files have been modified since last commit. Skipping to prevent empty commit.
Uploaded checkpoint_15 to seyedparsa/thoughtformer-checkpoints/checkpoints/star-datacurr-only-4gpu-bs32-lr1e-5_n1000_qwen3-0.6b-base_thoughtformer_lr1e-5_no-reset_dc_no-tc_adaptive-03b89ee9/checkpoint_15

============================================================
EPOCH 15/100 (thought_stage=10, data_stage=9)
============================================================
  thought_stage=10, c_thought=1, max_difficulty=9

Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):  16%|█▌        | 156/1000 [00:00<00:00, 946.88 examples/s]
Map (num_proc=4):  16%|█▌        | 156/1000 [00:00<00:00, 915.41 examples/s]
Map (num_proc=4):  16%|█▌        | 156/1000 [00:00<00:00, 893.91 examples/s]
Map (num_proc=4):  81%|████████  | 811/1000 [00:00<00:00, 3396.26 examples/s]
Map (num_proc=4):  81%|████████  | 811/1000 [00:00<00:00, 3321.59 examples/s]
Map (num_proc=4):  82%|████████▏ | 815/1000 [00:00<00:00, 2948.71 examples/s]
Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2409.24 examples/s]

Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2405.40 examples/s]

Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2430.09 examples/s]

Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Map (num_proc=4):  16%|█▌        | 156/1000 [00:00<00:01, 484.92 examples/s]
Map (num_proc=4):  81%|████████  | 811/1000 [00:00<00:00, 2075.50 examples/s]
Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:04, 1919.55 examples/s]
Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:04, 1859.65 examples/s]
Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:04, 1903.87 examples/s]
Filter (num_proc=4):  20%|██        | 2000/10000 [00:00<00:02, 3398.41 examples/s]
Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 1151.97 examples/s]

Filter (num_proc=4):  20%|██        | 2000/10000 [00:00<00:02, 3502.69 examples/s]
Filter (num_proc=4):  20%|██        | 2000/10000 [00:00<00:02, 3377.21 examples/s]
Filter (num_proc=4):  30%|███       | 3000/10000 [00:00<00:01, 4877.93 examples/s]
Filter (num_proc=4):  40%|████      | 4000/10000 [00:00<00:00, 6659.24 examples/s]
Filter (num_proc=4):  40%|████      | 4000/10000 [00:00<00:00, 6854.07 examples/s]
Filter (num_proc=4):  50%|█████     | 5000/10000 [00:00<00:00, 8372.55 examples/s]
Filter (num_proc=4):  55%|█████▌    | 5500/10000 [00:00<00:00, 8491.20 examples/s]
Filter (num_proc=4):  55%|█████▌    | 5500/10000 [00:00<00:00, 8551.52 examples/s]
Filter (num_proc=4):  70%|███████   | 7000/10000 [00:01<00:00, 9711.69 examples/s]
Filter (num_proc=4):  75%|███████▌  | 7500/10000 [00:00<00:00, 11831.93 examples/s]
Filter (num_proc=4):  75%|███████▌  | 7500/10000 [00:01<00:00, 10728.86 examples/s]
Filter (num_proc=4):  90%|█████████ | 9000/10000 [00:01<00:00, 12404.86 examples/s]
Filter (num_proc=4):  90%|█████████ | 9000/10000 [00:01<00:00, 11793.03 examples/s]
Filter (num_proc=4):  90%|█████████ | 9000/10000 [00:01<00:00, 11468.13 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 7247.47 examples/s]

Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 7359.01 examples/s]

Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 7221.97 examples/s]

Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:06, 1347.43 examples/s]
Filter (num_proc=4):  20%|██        | 2000/10000 [00:00<00:03, 2236.66 examples/s]
Filter (num_proc=4):  35%|███▌      | 3500/10000 [00:01<00:01, 4187.81 examples/s]
Filter (num_proc=4):  45%|████▌     | 4500/10000 [00:01<00:01, 4626.49 examples/s]
Filter (num_proc=4):  60%|██████    | 6000/10000 [00:01<00:00, 6388.08 examples/s]
Filter (num_proc=4):  70%|███████   | 7000/10000 [00:01<00:00, 6739.30 examples/s]
Filter (num_proc=4):  85%|████████▌ | 8500/10000 [00:01<00:00, 7082.55 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 7971.60 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:02<00:00, 4599.89 examples/s]

Map (num_proc=4):   0%|          | 0/9000 [00:00<?, ? examples/s]
Map (num_proc=4):   2%|▏         | 138/9000 [00:01<01:16, 116.39 examples/s]
Map (num_proc=4):   6%|▌         | 553/9000 [00:01<00:16, 527.33 examples/s]
Map (num_proc=4):   9%|▉         | 824/9000 [00:01<00:10, 807.05 examples/s]
Map (num_proc=4):  13%|█▎        | 1170/9000 [00:01<00:09, 804.12 examples/s]
Map (num_proc=4):  20%|█▉        | 1793/9000 [00:01<00:04, 1462.77 examples/s]
Map (num_proc=4):  27%|██▋       | 2389/9000 [00:02<00:03, 2141.86 examples/s]
Map (num_proc=4):  31%|███       | 2796/9000 [00:02<00:02, 2371.17 examples/s]
Map (num_proc=4):  36%|███▌      | 3205/9000 [00:02<00:02, 1944.44 examples/s]
Map (num_proc=4):  40%|███▉      | 3562/9000 [00:02<00:02, 1994.95 examples/s]
Map (num_proc=4):  43%|████▎     | 3835/9000 [00:02<00:02, 1979.48 examples/s]
Map (num_proc=4):  46%|████▌     | 4149/9000 [00:03<00:03, 1419.33 examples/s]
Map (num_proc=4):  57%|█████▋    | 5161/9000 [00:03<00:01, 2252.32 examples/s]
Map (num_proc=4):  65%|██████▌   | 5864/9000 [00:03<00:01, 2928.65 examples/s]
Map (num_proc=4):  73%|███████▎  | 6553/9000 [00:03<00:00, 3617.16 examples/s]
Map (num_proc=4):  79%|███████▉  | 7144/9000 [00:03<00:00, 4072.15 examples/s]
Map (num_proc=4):  85%|████████▌ | 7670/9000 [00:04<00:00, 2545.93 examples/s]
Map (num_proc=4):  92%|█████████▏| 8251/9000 [00:04<00:00, 2884.94 examples/s]
Map (num_proc=4):  97%|█████████▋| 8739/9000 [00:04<00:00, 3207.05 examples/s]
Map (num_proc=4): 100%|██████████| 9000/9000 [00:05<00:00, 1555.55 examples/s]

Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 1439.39 examples/s]
Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 1363.98 examples/s]
Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 1325.05 examples/s]
Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2723.17 examples/s]

Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2675.47 examples/s]

Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2561.07 examples/s]

Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:01, 717.10 examples/s]
Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 1356.00 examples/s]

Map (num_proc=4):   0%|          | 0/900 [00:00<?, ? examples/s]
Map (num_proc=4):  15%|█▌        | 135/900 [00:00<00:02, 360.31 examples/s]
Map (num_proc=4):  40%|███▉      | 359/900 [00:00<00:00, 857.43 examples/s]
Map (num_proc=4):  79%|███████▉  | 713/900 [00:00<00:00, 1481.66 examples/s]
Map (num_proc=4): 100%|██████████| 900/900 [00:01<00:00, 822.21 examples/s] 

Training Epoch: 16:   0%|          | 0/281 [00:00<?, ?it/s]
Training Epoch: 16:   0%|          | 0/281 [00:00<?, ?it/s]
Training Epoch: 16:   0%|          | 0/281 [00:00<?, ?it/s]
Training Epoch: 16:   0%|          | 0/281 [00:00<?, ?it/s]logging training data
rank 0 | gpu 0 | alloc=4.20 GB | reserved=45.87 GB | max=41.99 GB
rank 1 | gpu 1 | alloc=4.20 GB | reserved=45.86 GB | max=41.98 GB
rank 3 | gpu 3 | alloc=4.20 GB | reserved=45.87 GB | max=41.99 GB
rank 2 | gpu 2 | alloc=4.20 GB | reserved=52.06 GB | max=41.98 GB

Training Epoch: 16/100, batch 0/563 completed (loss: 0.1677:   0%|          | 0/281 [00:03<?, ?it/s]
Training Epoch: 16/100, batch 0/563 completed (loss: 0.1456:   0%|          | 0/281 [00:03<?, ?it/s]
Training Epoch: 16/100, batch 0/563 completed (loss: 0.007:   0%|          | 0/281 [00:03<?, ?it/s]
Training Epoch: 16/100, batch 0/563 completed (loss: 0.0645:   0%|          | 0/281 [00:03<?, ?it/s]
Training Epoch: 16/100, batch 0/563 completed (loss: 0.007:   0%|          | 1/281 [00:06<29:15,  6.27s/it]
Training Epoch: 16/100, batch 0/563 completed (loss: 0.1677:   0%|          | 1/281 [00:06<29:14,  6.26s/it]
Training Epoch: 16/100, batch 0/563 completed (loss: 0.0645:   0%|          | 1/281 [00:06<29:27,  6.31s/it]
Training Epoch: 16/100, batch 0/563 completed (loss: 0.1456:   0%|          | 1/281 [00:06<29:12,  6.26s/it]
Training Epoch: 16/100, batch 1/563 completed (loss: 0.0616:   0%|          | 1/281 [00:06<29:14,  6.26s/it]
Training Epoch: 16/100, batch 1/563 completed (loss: 0.1005:   0%|          | 1/281 [00:06<29:15,  6.27s/it]
Training Epoch: 16/100, batch 1/563 completed (loss: 0.0159:   0%|          | 1/281 [00:06<29:12,  6.26s/it]
Training Epoch: 16/100, batch 1/563 completed (loss: 0.0143:   0%|          | 1/281 [00:06<29:27,  6.31s/it]
Training Epoch: 16/100, batch 2/563 completed (loss: 0.0743:   0%|          | 1/281 [00:09<29:15,  6.27s/it]
Training Epoch: 16/100, batch 2/563 completed (loss: 0.1216:   0%|          | 1/281 [00:09<29:12,  6.26s/it]
Training Epoch: 16/100, batch 2/563 completed (loss: 0.0457:   0%|          | 1/281 [00:09<29:14,  6.26s/it]
Training Epoch: 16/100, batch 2/563 completed (loss: 0.0106:   0%|          | 1/281 [00:09<29:27,  6.31s/it]
Training Epoch: 16/100, batch 2/563 completed (loss: 0.0743:   1%|          | 2/281 [00:12<28:03,  6.03s/it]
Training Epoch: 16/100, batch 2/563 completed (loss: 0.0106:   1%|          | 2/281 [00:12<28:08,  6.05s/it]
Training Epoch: 16/100, batch 2/563 completed (loss: 0.0457:   1%|          | 2/281 [00:12<28:03,  6.03s/it]
Training Epoch: 16/100, batch 2/563 completed (loss: 0.1216:   1%|          | 2/281 [00:12<28:04,  6.04s/it]
Training Epoch: 16/100, batch 3/563 completed (loss: 0.4233:   1%|          | 2/281 [00:12<28:03,  6.03s/it]
Training Epoch: 16/100, batch 3/563 completed (loss: 0.0886:   1%|          | 2/281 [00:12<28:04,  6.04s/it]
Training Epoch: 16/100, batch 3/563 completed (loss: 0.0216:   1%|          | 2/281 [00:12<28:03,  6.03s/it]
Training Epoch: 16/100, batch 3/563 completed (loss: 0.0025:   1%|          | 2/281 [00:12<28:08,  6.05s/it]
Training Epoch: 16/100, batch 4/563 completed (loss: 0.2463:   1%|          | 2/281 [00:15<28:03,  6.03s/it]
Training Epoch: 16/100, batch 4/563 completed (loss: 0.0061:   1%|          | 2/281 [00:14<28:04,  6.04s/it]
Training Epoch: 16/100, batch 4/563 completed (loss: 0.0188:   1%|          | 2/281 [00:15<28:03,  6.03s/it]
Training Epoch: 16/100, batch 4/563 completed (loss: 0.0096:   1%|          | 2/281 [00:15<28:08,  6.05s/it]
Training Epoch: 16/100, batch 4/563 completed (loss: 0.0096:   1%|          | 3/281 [00:17<27:08,  5.86s/it]
Training Epoch: 16/100, batch 4/563 completed (loss: 0.0188:   1%|          | 3/281 [00:17<27:05,  5.85s/it]
Training Epoch: 16/100, batch 4/563 completed (loss: 0.2463:   1%|          | 3/281 [00:17<27:06,  5.85s/it]
Training Epoch: 16/100, batch 4/563 completed (loss: 0.0061:   1%|          | 3/281 [00:17<27:04,  5.84s/it]
Training Epoch: 16/100, batch 5/563 completed (loss: 0.0257:   1%|          | 3/281 [00:17<27:05,  5.85s/it]
Training Epoch: 16/100, batch 5/563 completed (loss: 0.0203:   1%|          | 3/281 [00:17<27:04,  5.84s/it]
Training Epoch: 16/100, batch 5/563 completed (loss: 0.0094:   1%|          | 3/281 [00:17<27:06,  5.85s/it]
Training Epoch: 16/100, batch 5/563 completed (loss: 0.0036:   1%|          | 3/281 [00:17<27:08,  5.86s/it]
Training Epoch: 16/100, batch 6/563 completed (loss: 0.0441:   1%|          | 3/281 [00:20<27:06,  5.85s/it]
Training Epoch: 16/100, batch 6/563 completed (loss: 0.0088:   1%|          | 3/281 [00:20<27:04,  5.84s/it]
Training Epoch: 16/100, batch 6/563 completed (loss: 0.0656:   1%|          | 3/281 [00:20<27:05,  5.85s/it]
Training Epoch: 16/100, batch 6/563 completed (loss: 0.1692:   1%|          | 3/281 [00:20<27:08,  5.86s/it]
Training Epoch: 16/100, batch 6/563 completed (loss: 0.1692:   1%|▏         | 4/281 [00:24<27:50,  6.03s/it]
Training Epoch: 16/100, batch 6/563 completed (loss: 0.0656:   1%|▏         | 4/281 [00:24<27:49,  6.03s/it]
Training Epoch: 16/100, batch 6/563 completed (loss: 0.0088:   1%|▏         | 4/281 [00:24<27:49,  6.03s/it]
Training Epoch: 16/100, batch 6/563 completed (loss: 0.0441:   1%|▏         | 4/281 [00:24<27:50,  6.03s/it]
Training Epoch: 16/100, batch 7/563 completed (loss: 0.0304:   1%|▏         | 4/281 [00:24<27:49,  6.03s/it]
Training Epoch: 16/100, batch 7/563 completed (loss: 0.0661:   1%|▏         | 4/281 [00:24<27:50,  6.03s/it]
Training Epoch: 16/100, batch 7/563 completed (loss: 0.5207:   1%|▏         | 4/281 [00:24<27:49,  6.03s/it]
Training Epoch: 16/100, batch 7/563 completed (loss: 0.0086:   1%|▏         | 4/281 [00:24<27:50,  6.03s/it]
Training Epoch: 16/100, batch 8/563 completed (loss: 0.0357:   1%|▏         | 4/281 [00:26<27:50,  6.03s/it]
Training Epoch: 16/100, batch 8/563 completed (loss: 0.0535:   1%|▏         | 4/281 [00:26<27:49,  6.03s/it]
Training Epoch: 16/100, batch 8/563 completed (loss: 0.018:   1%|▏         | 4/281 [00:26<27:49,  6.03s/it] 
Training Epoch: 16/100, batch 8/563 completed (loss: 0.0412:   1%|▏         | 4/281 [00:26<27:50,  6.03s/it]
Training Epoch: 16/100, batch 8/563 completed (loss: 0.018:   2%|▏         | 5/281 [00:29<27:05,  5.89s/it]
Training Epoch: 16/100, batch 8/563 completed (loss: 0.0357:   2%|▏         | 5/281 [00:29<27:06,  5.89s/it]
Training Epoch: 16/100, batch 8/563 completed (loss: 0.0412:   2%|▏         | 5/281 [00:29<27:08,  5.90s/it]
Training Epoch: 16/100, batch 8/563 completed (loss: 0.0535:   2%|▏         | 5/281 [00:29<27:07,  5.90s/it]
Training Epoch: 16/100, batch 9/563 completed (loss: 0.1329:   2%|▏         | 5/281 [00:29<27:07,  5.90s/it]
Training Epoch: 16/100, batch 9/563 completed (loss: 0.1362:   2%|▏         | 5/281 [00:29<27:05,  5.89s/it]
Training Epoch: 16/100, batch 9/563 completed (loss: 0.7558:   2%|▏         | 5/281 [00:29<27:06,  5.89s/it]
Training Epoch: 16/100, batch 9/563 completed (loss: 0.0194:   2%|▏         | 5/281 [00:29<27:08,  5.90s/it]
Training Epoch: 16/100, batch 10/563 completed (loss: 0.2133:   2%|▏         | 5/281 [00:32<27:07,  5.90s/it]
Training Epoch: 16/100, batch 10/563 completed (loss: 0.1178:   2%|▏         | 5/281 [00:32<27:05,  5.89s/it]
Training Epoch: 16/100, batch 10/563 completed (loss: 0.2275:   2%|▏         | 5/281 [00:32<27:06,  5.89s/it]
Training Epoch: 16/100, batch 10/563 completed (loss: 0.0379:   2%|▏         | 5/281 [00:32<27:08,  5.90s/it]
Training Epoch: 16/100, batch 10/563 completed (loss: 0.0379:   2%|▏         | 6/281 [00:35<26:54,  5.87s/it]
Training Epoch: 16/100, batch 10/563 completed (loss: 0.1178:   2%|▏         | 6/281 [00:35<26:52,  5.87s/it]
Training Epoch: 16/100, batch 10/563 completed (loss: 0.2275:   2%|▏         | 6/281 [00:35<26:53,  5.87s/it]
Training Epoch: 16/100, batch 10/563 completed (loss: 0.2133:   2%|▏         | 6/281 [00:35<26:53,  5.87s/it]
Training Epoch: 16/100, batch 11/563 completed (loss: 0.0513:   2%|▏         | 6/281 [00:35<26:53,  5.87s/it]
Training Epoch: 16/100, batch 11/563 completed (loss: 0.0237:   2%|▏         | 6/281 [00:35<26:52,  5.87s/it]
Training Epoch: 16/100, batch 11/563 completed (loss: 0.0075:   2%|▏         | 6/281 [00:35<26:53,  5.87s/it]
Training Epoch: 16/100, batch 11/563 completed (loss: 0.1995:   2%|▏         | 6/281 [00:35<26:54,  5.87s/it]
Training Epoch: 16/100, batch 12/563 completed (loss: 0.0141:   2%|▏         | 6/281 [00:38<26:52,  5.87s/it]
Training Epoch: 16/100, batch 12/563 completed (loss: 0.44:   2%|▏         | 6/281 [00:38<26:53,  5.87s/it]  
Training Epoch: 16/100, batch 12/563 completed (loss: 0.12:   2%|▏         | 6/281 [00:38<26:53,  5.87s/it]  
Training Epoch: 16/100, batch 12/563 completed (loss: 0.0346:   2%|▏         | 6/281 [00:38<26:54,  5.87s/it]
Training Epoch: 16/100, batch 12/563 completed (loss: 0.0346:   2%|▏         | 7/281 [00:41<27:10,  5.95s/it]
Training Epoch: 16/100, batch 12/563 completed (loss: 0.0141:   2%|▏         | 7/281 [00:41<27:10,  5.95s/it]
Training Epoch: 16/100, batch 12/563 completed (loss: 0.12:   2%|▏         | 7/281 [00:41<27:11,  5.95s/it]
Training Epoch: 16/100, batch 12/563 completed (loss: 0.44:   2%|▏         | 7/281 [00:41<27:11,  5.95s/it]
Training Epoch: 16/100, batch 13/563 completed (loss: 0.0516:   2%|▏         | 7/281 [00:41<27:11,  5.95s/it]
Training Epoch: 16/100, batch 13/563 completed (loss: 0.009:   2%|▏         | 7/281 [00:41<27:11,  5.95s/it]
Training Epoch: 16/100, batch 13/563 completed (loss: 0.0118:   2%|▏         | 7/281 [00:41<27:10,  5.95s/it]
Training Epoch: 16/100, batch 13/563 completed (loss: 0.0116:   2%|▏         | 7/281 [00:41<27:10,  5.95s/it]
Training Epoch: 16/100, batch 14/563 completed (loss: 0.0991:   2%|▏         | 7/281 [00:44<27:11,  5.95s/it]
Training Epoch: 16/100, batch 14/563 completed (loss: 0.0726:   2%|▏         | 7/281 [00:44<27:10,  5.95s/it]
Training Epoch: 16/100, batch 14/563 completed (loss: 0.0712:   2%|▏         | 7/281 [00:44<27:11,  5.95s/it]
Training Epoch: 16/100, batch 14/563 completed (loss: 0.0164:   2%|▏         | 7/281 [00:44<27:10,  5.95s/it]
Training Epoch: 16/100, batch 14/563 completed (loss: 0.0991:   3%|▎         | 8/281 [00:47<26:42,  5.87s/it]
Training Epoch: 16/100, batch 14/563 completed (loss: 0.0712:   3%|▎         | 8/281 [00:47<26:42,  5.87s/it]
Training Epoch: 16/100, batch 14/563 completed (loss: 0.0726:   3%|▎         | 8/281 [00:47<26:42,  5.87s/it]
Training Epoch: 16/100, batch 14/563 completed (loss: 0.0164:   3%|▎         | 8/281 [00:47<26:43,  5.87s/it]
Training Epoch: 16/100, batch 15/563 completed (loss: 0.1858:   3%|▎         | 8/281 [00:47<26:42,  5.87s/it]
Training Epoch: 16/100, batch 15/563 completed (loss: 0.0178:   3%|▎         | 8/281 [00:47<26:42,  5.87s/it]
Training Epoch: 16/100, batch 15/563 completed (loss: 0.0066:   3%|▎         | 8/281 [00:47<26:42,  5.87s/it]
Training Epoch: 16/100, batch 15/563 completed (loss: 0.0117:   3%|▎         | 8/281 [00:47<26:43,  5.87s/it]
Training Epoch: 16/100, batch 16/563 completed (loss: 0.0368:   3%|▎         | 8/281 [00:50<26:42,  5.87s/it]
Training Epoch: 16/100, batch 16/563 completed (loss: 0.5406:   3%|▎         | 8/281 [00:50<26:42,  5.87s/it]
Training Epoch: 16/100, batch 16/563 completed (loss: 0.0201:   3%|▎         | 8/281 [00:50<26:42,  5.87s/it]
Training Epoch: 16/100, batch 16/563 completed (loss: 0.0847:   3%|▎         | 8/281 [00:50<26:43,  5.87s/it]
Training Epoch: 16/100, batch 16/563 completed (loss: 0.0847:   3%|▎         | 9/281 [00:53<27:11,  6.00s/it]
Training Epoch: 16/100, batch 16/563 completed (loss: 0.0201:   3%|▎         | 9/281 [00:53<27:12,  6.00s/it]
Training Epoch: 16/100, batch 16/563 completed (loss: 0.5406:   3%|▎         | 9/281 [00:53<27:12,  6.00s/it]
Training Epoch: 16/100, batch 16/563 completed (loss: 0.0368:   3%|▎         | 9/281 [00:53<27:12,  6.00s/it]
Training Epoch: 16/100, batch 17/563 completed (loss: 0.0111:   3%|▎         | 9/281 [00:53<27:12,  6.00s/it]
Training Epoch: 16/100, batch 17/563 completed (loss: 0.0066:   3%|▎         | 9/281 [00:53<27:12,  6.00s/it]
Training Epoch: 16/100, batch 17/563 completed (loss: 0.011:   3%|▎         | 9/281 [00:53<27:12,  6.00s/it] 
Training Epoch: 16/100, batch 17/563 completed (loss: 0.0186:   3%|▎         | 9/281 [00:53<27:11,  6.00s/it]
Training Epoch: 16/100, batch 18/563 completed (loss: 0.0101:   3%|▎         | 9/281 [00:56<27:12,  6.00s/it]
Training Epoch: 16/100, batch 18/563 completed (loss: 0.0937:   3%|▎         | 9/281 [00:56<27:12,  6.00s/it]
Training Epoch: 16/100, batch 18/563 completed (loss: 0.2686:   3%|▎         | 9/281 [00:56<27:12,  6.00s/it]
Training Epoch: 16/100, batch 18/563 completed (loss: 0.0258:   3%|▎         | 9/281 [00:56<27:11,  6.00s/it]
Training Epoch: 16/100, batch 18/563 completed (loss: 0.0258:   4%|▎         | 10/281 [00:59<26:43,  5.92s/it]
Training Epoch: 16/100, batch 18/563 completed (loss: 0.0101:   4%|▎         | 10/281 [00:59<26:43,  5.92s/it]
Training Epoch: 16/100, batch 18/563 completed (loss: 0.2686:   4%|▎         | 10/281 [00:59<26:43,  5.92s/it]
Training Epoch: 16/100, batch 18/563 completed (loss: 0.0937:   4%|▎         | 10/281 [00:59<26:43,  5.92s/it]
Training Epoch: 16/100, batch 19/563 completed (loss: 0.0369:   4%|▎         | 10/281 [00:59<26:43,  5.92s/it]
Training Epoch: 16/100, batch 19/563 completed (loss: 0.0227:   4%|▎         | 10/281 [00:59<26:43,  5.92s/it]
Training Epoch: 16/100, batch 19/563 completed (loss: 0.0335:   4%|▎         | 10/281 [00:59<26:43,  5.92s/it]
Training Epoch: 16/100, batch 19/563 completed (loss: 0.2469:   4%|▎         | 10/281 [00:59<26:43,  5.92s/it]
Training Epoch: 16/100, batch 20/563 completed (loss: 0.0268:   4%|▎         | 10/281 [01:02<26:43,  5.92s/it]
Training Epoch: 16/100, batch 20/563 completed (loss: 0.0435:   4%|▎         | 10/281 [01:02<26:43,  5.92s/it]
Training Epoch: 16/100, batch 20/563 completed (loss: 0.0148:   4%|▎         | 10/281 [01:02<26:43,  5.92s/it]
Training Epoch: 16/100, batch 20/563 completed (loss: 0.0189:   4%|▎         | 10/281 [01:02<26:43,  5.92s/it]
Training Epoch: 16/100, batch 20/563 completed (loss: 0.0268:   4%|▍         | 11/281 [01:05<26:18,  5.85s/it]
Training Epoch: 16/100, batch 20/563 completed (loss: 0.0148:   4%|▍         | 11/281 [01:05<26:18,  5.85s/it]
Training Epoch: 16/100, batch 20/563 completed (loss: 0.0189:   4%|▍         | 11/281 [01:05<26:19,  5.85s/it]
Training Epoch: 16/100, batch 20/563 completed (loss: 0.0435:   4%|▍         | 11/281 [01:05<26:19,  5.85s/it]
Training Epoch: 16/100, batch 21/563 completed (loss: 0.0445:   4%|▍         | 11/281 [01:05<26:18,  5.85s/it]
Training Epoch: 16/100, batch 21/563 completed (loss: 0.2056:   4%|▍         | 11/281 [01:05<26:19,  5.85s/it]
Training Epoch: 16/100, batch 21/563 completed (loss: 0.0052:   4%|▍         | 11/281 [01:05<26:18,  5.85s/it]
Training Epoch: 16/100, batch 21/563 completed (loss: 0.0144:   4%|▍         | 11/281 [01:05<26:19,  5.85s/it]
Training Epoch: 16/100, batch 22/563 completed (loss: 0.0108:   4%|▍         | 11/281 [01:08<26:18,  5.85s/it]
Training Epoch: 16/100, batch 22/563 completed (loss: 0.0268:   4%|▍         | 11/281 [01:08<26:19,  5.85s/it]
Training Epoch: 16/100, batch 22/563 completed (loss: 0.0113:   4%|▍         | 11/281 [01:08<26:18,  5.85s/it]
Training Epoch: 16/100, batch 22/563 completed (loss: 0.0824:   4%|▍         | 11/281 [01:08<26:19,  5.85s/it]
Training Epoch: 16/100, batch 22/563 completed (loss: 0.0824:   4%|▍         | 12/281 [01:11<26:43,  5.96s/it]
Training Epoch: 16/100, batch 22/563 completed (loss: 0.0108:   4%|▍         | 12/281 [01:11<26:43,  5.96s/it]
Training Epoch: 16/100, batch 22/563 completed (loss: 0.0268:   4%|▍         | 12/281 [01:11<26:43,  5.96s/it]
Training Epoch: 16/100, batch 22/563 completed (loss: 0.0113:   4%|▍         | 12/281 [01:11<26:43,  5.96s/it]
Training Epoch: 16/100, batch 23/563 completed (loss: 0.0829:   4%|▍         | 12/281 [01:11<26:43,  5.96s/it]
Training Epoch: 16/100, batch 23/563 completed (loss: 0.3773:   4%|▍         | 12/281 [01:11<26:43,  5.96s/it]
Training Epoch: 16/100, batch 23/563 completed (loss: 0.1016:   4%|▍         | 12/281 [01:11<26:43,  5.96s/it]
Training Epoch: 16/100, batch 23/563 completed (loss: 0.2766:   4%|▍         | 12/281 [01:11<26:43,  5.96s/it]
Training Epoch: 16/100, batch 24/563 completed (loss: 0.1077:   4%|▍         | 12/281 [01:14<26:43,  5.96s/it]
Training Epoch: 16/100, batch 24/563 completed (loss: 0.0117:   4%|▍         | 12/281 [01:14<26:43,  5.96s/it]
Training Epoch: 16/100, batch 24/563 completed (loss: 0.0227:   4%|▍         | 12/281 [01:14<26:43,  5.96s/it]
Training Epoch: 16/100, batch 24/563 completed (loss: 0.0168:   4%|▍         | 12/281 [01:14<26:43,  5.96s/it]
Training Epoch: 16/100, batch 24/563 completed (loss: 0.0168:   5%|▍         | 13/281 [01:17<26:18,  5.89s/it]
Training Epoch: 16/100, batch 24/563 completed (loss: 0.0117:   5%|▍         | 13/281 [01:17<26:18,  5.89s/it]
Training Epoch: 16/100, batch 24/563 completed (loss: 0.0227:   5%|▍         | 13/281 [01:17<26:18,  5.89s/it]
Training Epoch: 16/100, batch 24/563 completed (loss: 0.1077:   5%|▍         | 13/281 [01:17<26:19,  5.89s/it]
Training Epoch: 16/100, batch 25/563 completed (loss: 0.0146:   5%|▍         | 13/281 [01:17<26:19,  5.89s/it]
Training Epoch: 16/100, batch 25/563 completed (loss: 0.0045:   5%|▍         | 13/281 [01:17<26:18,  5.89s/it]
Training Epoch: 16/100, batch 25/563 completed (loss: 0.3299:   5%|▍         | 13/281 [01:17<26:18,  5.89s/it]
Training Epoch: 16/100, batch 25/563 completed (loss: 0.0129:   5%|▍         | 13/281 [01:17<26:18,  5.89s/it]
Training Epoch: 16/100, batch 26/563 completed (loss: 0.0135:   5%|▍         | 13/281 [01:19<26:18,  5.89s/it]
Training Epoch: 16/100, batch 26/563 completed (loss: 0.007:   5%|▍         | 13/281 [01:19<26:19,  5.89s/it] 
Training Epoch: 16/100, batch 26/563 completed (loss: 0.0608:   5%|▍         | 13/281 [01:19<26:18,  5.89s/it]
Training Epoch: 16/100, batch 26/563 completed (loss: 0.0673:   5%|▍         | 13/281 [01:19<26:18,  5.89s/it]
Training Epoch: 16/100, batch 26/563 completed (loss: 0.0608:   5%|▍         | 14/281 [01:22<26:00,  5.85s/it]
Training Epoch: 16/100, batch 26/563 completed (loss: 0.0135:   5%|▍         | 14/281 [01:22<26:00,  5.85s/it]
Training Epoch: 16/100, batch 26/563 completed (loss: 0.007:   5%|▍         | 14/281 [01:22<26:00,  5.85s/it]
Training Epoch: 16/100, batch 26/563 completed (loss: 0.0673:   5%|▍         | 14/281 [01:22<26:01,  5.85s/it]
Training Epoch: 16/100, batch 27/563 completed (loss: 0.0712:   5%|▍         | 14/281 [01:22<26:00,  5.85s/it]
Training Epoch: 16/100, batch 27/563 completed (loss: 0.2239:   5%|▍         | 14/281 [01:22<26:00,  5.85s/it]
Training Epoch: 16/100, batch 27/563 completed (loss: 0.3338:   5%|▍         | 14/281 [01:22<26:00,  5.85s/it]
Training Epoch: 16/100, batch 27/563 completed (loss: 0.1532:   5%|▍         | 14/281 [01:22<26:01,  5.85s/it]
Training Epoch: 16/100, batch 28/563 completed (loss: 0.2053:   5%|▍         | 14/281 [01:26<26:00,  5.85s/it]
Training Epoch: 16/100, batch 28/563 completed (loss: 0.0286:   5%|▍         | 14/281 [01:26<26:00,  5.85s/it]
Training Epoch: 16/100, batch 28/563 completed (loss: 0.0749:   5%|▍         | 14/281 [01:26<26:00,  5.85s/it]
Training Epoch: 16/100, batch 28/563 completed (loss: 0.102:   5%|▍         | 14/281 [01:26<26:01,  5.85s/it] 
Training Epoch: 16/100, batch 28/563 completed (loss: 0.0749:   5%|▌         | 15/281 [01:28<26:25,  5.96s/it]
Training Epoch: 16/100, batch 28/563 completed (loss: 0.0286:   5%|▌         | 15/281 [01:28<26:25,  5.96s/it]
Training Epoch: 16/100, batch 28/563 completed (loss: 0.102:   5%|▌         | 15/281 [01:29<26:25,  5.96s/it]
Training Epoch: 16/100, batch 28/563 completed (loss: 0.2053:   5%|▌         | 15/281 [01:28<26:25,  5.96s/it]
Training Epoch: 16/100, batch 29/563 completed (loss: 0.0058:   5%|▌         | 15/281 [01:29<26:25,  5.96s/it]
Training Epoch: 16/100, batch 29/563 completed (loss: 0.0212:   5%|▌         | 15/281 [01:29<26:25,  5.96s/it]
Training Epoch: 16/100, batch 29/563 completed (loss: 0.3669:   5%|▌         | 15/281 [01:29<26:25,  5.96s/it]
Training Epoch: 16/100, batch 29/563 completed (loss: 0.007:   5%|▌         | 15/281 [01:29<26:25,  5.96s/it]
Training Epoch: 16/100, batch 30/563 completed (loss: 0.0143:   5%|▌         | 15/281 [01:31<26:25,  5.96s/it]
Training Epoch: 16/100, batch 30/563 completed (loss: 0.2837:   5%|▌         | 15/281 [01:31<26:25,  5.96s/it]
Training Epoch: 16/100, batch 30/563 completed (loss: 0.0633:   5%|▌         | 15/281 [01:31<26:25,  5.96s/it]
Training Epoch: 16/100, batch 30/563 completed (loss: 0.0201:   5%|▌         | 15/281 [01:31<26:25,  5.96s/it]
Training Epoch: 16/100, batch 30/563 completed (loss: 0.2837:   6%|▌         | 16/281 [01:34<26:03,  5.90s/it]
Training Epoch: 16/100, batch 30/563 completed (loss: 0.0633:   6%|▌         | 16/281 [01:34<26:04,  5.90s/it]
Training Epoch: 16/100, batch 30/563 completed (loss: 0.0143:   6%|▌         | 16/281 [01:34<26:04,  5.90s/it]
Training Epoch: 16/100, batch 30/563 completed (loss: 0.0201:   6%|▌         | 16/281 [01:34<26:04,  5.91s/it]
Training Epoch: 16/100, batch 31/563 completed (loss: 0.1439:   6%|▌         | 16/281 [01:34<26:04,  5.90s/it]
Training Epoch: 16/100, batch 31/563 completed (loss: 0.0077:   6%|▌         | 16/281 [01:34<26:04,  5.90s/it]
Training Epoch: 16/100, batch 31/563 completed (loss: 0.007:   6%|▌         | 16/281 [01:34<26:03,  5.90s/it] 
Training Epoch: 16/100, batch 31/563 completed (loss: 0.1139:   6%|▌         | 16/281 [01:34<26:04,  5.91s/it]
Training Epoch: 16/100, batch 32/563 completed (loss: 0.0216:   6%|▌         | 16/281 [01:37<26:04,  5.90s/it]
Training Epoch: 16/100, batch 32/563 completed (loss: 0.3153:   6%|▌         | 16/281 [01:37<26:03,  5.90s/it]
Training Epoch: 16/100, batch 32/563 completed (loss: 0.1185:   6%|▌         | 16/281 [01:37<26:04,  5.90s/it]
Training Epoch: 16/100, batch 32/563 completed (loss: 0.0359:   6%|▌         | 16/281 [01:37<26:04,  5.91s/it]
Training Epoch: 16/100, batch 32/563 completed (loss: 0.0359:   6%|▌         | 17/281 [01:40<26:19,  5.98s/it]
Training Epoch: 16/100, batch 32/563 completed (loss: 0.0216:   6%|▌         | 17/281 [01:40<26:19,  5.98s/it]
Training Epoch: 16/100, batch 32/563 completed (loss: 0.3153:   6%|▌         | 17/281 [01:40<26:19,  5.98s/it]
Training Epoch: 16/100, batch 32/563 completed (loss: 0.1185:   6%|▌         | 17/281 [01:40<26:19,  5.98s/it]
Training Epoch: 16/100, batch 33/563 completed (loss: 0.0087:   6%|▌         | 17/281 [01:40<26:19,  5.98s/it]
Training Epoch: 16/100, batch 33/563 completed (loss: 0.0539:   6%|▌         | 17/281 [01:40<26:19,  5.98s/it]
Training Epoch: 16/100, batch 33/563 completed (loss: 0.1245:   6%|▌         | 17/281 [01:40<26:19,  5.98s/it]
Training Epoch: 16/100, batch 33/563 completed (loss: 0.0323:   6%|▌         | 17/281 [01:41<26:19,  5.98s/it]
Training Epoch: 16/100, batch 34/563 completed (loss: 0.0153:   6%|▌         | 17/281 [01:43<26:19,  5.98s/it]
Training Epoch: 16/100, batch 34/563 completed (loss: 0.0505:   6%|▌         | 17/281 [01:43<26:19,  5.98s/it]
Training Epoch: 16/100, batch 34/563 completed (loss: 0.0183:   6%|▌         | 17/281 [01:43<26:19,  5.98s/it]
Training Epoch: 16/100, batch 34/563 completed (loss: 0.0103:   6%|▌         | 17/281 [01:43<26:19,  5.98s/it]
Training Epoch: 16/100, batch 34/563 completed (loss: 0.0153:   6%|▋         | 18/281 [01:46<25:55,  5.91s/it]
Training Epoch: 16/100, batch 34/563 completed (loss: 0.0103:   6%|▋         | 18/281 [01:46<25:56,  5.92s/it]
Training Epoch: 16/100, batch 34/563 completed (loss: 0.0505:   6%|▋         | 18/281 [01:46<25:56,  5.92s/it]
Training Epoch: 16/100, batch 34/563 completed (loss: 0.0183:   6%|▋         | 18/281 [01:46<25:56,  5.92s/it]
Training Epoch: 16/100, batch 35/563 completed (loss: 0.0119:   6%|▋         | 18/281 [01:46<25:56,  5.92s/it]
Training Epoch: 16/100, batch 35/563 completed (loss: 0.1742:   6%|▋         | 18/281 [01:46<25:55,  5.91s/it]
Training Epoch: 16/100, batch 35/563 completed (loss: 0.0207:   6%|▋         | 18/281 [01:46<25:56,  5.92s/it]
Training Epoch: 16/100, batch 35/563 completed (loss: 0.1385:   6%|▋         | 18/281 [01:46<25:56,  5.92s/it]
Training Epoch: 16/100, batch 36/563 completed (loss: 0.0617:   6%|▋         | 18/281 [01:49<25:56,  5.92s/it]
Training Epoch: 16/100, batch 36/563 completed (loss: 0.0541:   6%|▋         | 18/281 [01:49<25:56,  5.92s/it]
Training Epoch: 16/100, batch 36/563 completed (loss: 0.0467:   6%|▋         | 18/281 [01:49<25:55,  5.91s/it]
Training Epoch: 16/100, batch 36/563 completed (loss: 0.1287:   6%|▋         | 18/281 [01:49<25:56,  5.92s/it]
Training Epoch: 16/100, batch 36/563 completed (loss: 0.1287:   7%|▋         | 19/281 [01:52<25:38,  5.87s/it]
Training Epoch: 16/100, batch 36/563 completed (loss: 0.0617:   7%|▋         | 19/281 [01:52<25:37,  5.87s/it]
Training Epoch: 16/100, batch 36/563 completed (loss: 0.0541:   7%|▋         | 19/281 [01:52<25:38,  5.87s/it]
Training Epoch: 16/100, batch 36/563 completed (loss: 0.0467:   7%|▋         | 19/281 [01:52<25:38,  5.87s/it]
Training Epoch: 16/100, batch 37/563 completed (loss: 0.0573:   7%|▋         | 19/281 [01:52<25:38,  5.87s/it]
Training Epoch: 16/100, batch 37/563 completed (loss: 0.0693:   7%|▋         | 19/281 [01:52<25:37,  5.87s/it]
Training Epoch: 16/100, batch 37/563 completed (loss: 0.0429:   7%|▋         | 19/281 [01:52<25:38,  5.87s/it]
Training Epoch: 16/100, batch 37/563 completed (loss: 0.059:   7%|▋         | 19/281 [01:52<25:38,  5.87s/it] 
Training Epoch: 16/100, batch 38/563 completed (loss: 0.0072:   7%|▋         | 19/281 [01:55<25:38,  5.87s/it]
Training Epoch: 16/100, batch 38/563 completed (loss: 0.0537:   7%|▋         | 19/281 [01:55<25:37,  5.87s/it]
Training Epoch: 16/100, batch 38/563 completed (loss: 0.0431:   7%|▋         | 19/281 [01:55<25:38,  5.87s/it]
Training Epoch: 16/100, batch 38/563 completed (loss: 0.2342:   7%|▋         | 19/281 [01:55<25:38,  5.87s/it]
Training Epoch: 16/100, batch 38/563 completed (loss: 0.0537:   7%|▋         | 20/281 [01:58<26:04,  5.99s/it]
Training Epoch: 16/100, batch 38/563 completed (loss: 0.0431:   7%|▋         | 20/281 [01:58<26:04,  5.99s/it]
Training Epoch: 16/100, batch 38/563 completed (loss: 0.0072:   7%|▋         | 20/281 [01:58<26:04,  5.99s/it]
Training Epoch: 16/100, batch 38/563 completed (loss: 0.2342:   7%|▋         | 20/281 [01:58<26:04,  6.00s/it]
Training Epoch: 16/100, batch 39/563 completed (loss: 0.1013:   7%|▋         | 20/281 [01:58<26:04,  5.99s/it]
Training Epoch: 16/100, batch 39/563 completed (loss: 0.3301:   7%|▋         | 20/281 [01:58<26:04,  5.99s/it]
Training Epoch: 16/100, batch 39/563 completed (loss: 0.0148:   7%|▋         | 20/281 [01:58<26:04,  5.99s/it]
Training Epoch: 16/100, batch 39/563 completed (loss: 0.0027:   7%|▋         | 20/281 [01:58<26:04,  6.00s/it]
Training Epoch: 16/100, batch 40/563 completed (loss: 0.3749:   7%|▋         | 20/281 [02:01<26:04,  5.99s/it]
Training Epoch: 16/100, batch 40/563 completed (loss: 0.0182:   7%|▋         | 20/281 [02:01<26:04,  5.99s/it]
Training Epoch: 16/100, batch 40/563 completed (loss: 0.0108:   7%|▋         | 20/281 [02:01<26:04,  5.99s/it]
Training Epoch: 16/100, batch 40/563 completed (loss: 0.0565:   7%|▋         | 20/281 [02:01<26:04,  6.00s/it]
Training Epoch: 16/100, batch 40/563 completed (loss: 0.0565:   7%|▋         | 21/281 [02:04<25:42,  5.93s/it]
Training Epoch: 16/100, batch 40/563 completed (loss: 0.0182:   7%|▋         | 21/281 [02:04<25:43,  5.94s/it]
Training Epoch: 16/100, batch 40/563 completed (loss: 0.0108:   7%|▋         | 21/281 [02:04<25:43,  5.94s/it]
Training Epoch: 16/100, batch 40/563 completed (loss: 0.3749:   7%|▋         | 21/281 [02:04<25:43,  5.94s/it]
Training Epoch: 16/100, batch 41/563 completed (loss: 0.0062:   7%|▋         | 21/281 [02:04<25:43,  5.94s/it]
Training Epoch: 16/100, batch 41/563 completed (loss: 0.0633:   7%|▋         | 21/281 [02:04<25:43,  5.94s/it]
Training Epoch: 16/100, batch 41/563 completed (loss: 0.0369:   7%|▋         | 21/281 [02:04<25:43,  5.94s/it]
Training Epoch: 16/100, batch 41/563 completed (loss: 0.0292:   7%|▋         | 21/281 [02:04<25:42,  5.93s/it]
Training Epoch: 16/100, batch 42/563 completed (loss: 0.009:   7%|▋         | 21/281 [02:07<25:43,  5.94s/it] 
Training Epoch: 16/100, batch 42/563 completed (loss: 0.0512:   7%|▋         | 21/281 [02:07<25:43,  5.94s/it]
Training Epoch: 16/100, batch 42/563 completed (loss: 0.0866:   7%|▋         | 21/281 [02:07<25:43,  5.94s/it]
Training Epoch: 16/100, batch 42/563 completed (loss: 0.0102:   7%|▋         | 21/281 [02:07<25:42,  5.93s/it]
Training Epoch: 16/100, batch 42/563 completed (loss: 0.0102:   8%|▊         | 22/281 [02:10<25:50,  5.98s/it]
Training Epoch: 16/100, batch 42/563 completed (loss: 0.0512:   8%|▊         | 22/281 [02:10<25:50,  5.98s/it]
Training Epoch: 16/100, batch 42/563 completed (loss: 0.0866:   8%|▊         | 22/281 [02:10<25:50,  5.99s/it]
Training Epoch: 16/100, batch 42/563 completed (loss: 0.009:   8%|▊         | 22/281 [02:10<25:51,  5.99s/it]
Training Epoch: 16/100, batch 43/563 completed (loss: 0.0213:   8%|▊         | 22/281 [02:10<25:50,  5.99s/it]
Training Epoch: 16/100, batch 43/563 completed (loss: 0.0121:   8%|▊         | 22/281 [02:10<25:50,  5.98s/it]
Training Epoch: 16/100, batch 43/563 completed (loss: 0.0536:   8%|▊         | 22/281 [02:10<25:51,  5.99s/it]
Training Epoch: 16/100, batch 43/563 completed (loss: 0.0234:   8%|▊         | 22/281 [02:10<25:50,  5.98s/it]
Training Epoch: 16/100, batch 44/563 completed (loss: 0.0467:   8%|▊         | 22/281 [02:13<25:50,  5.99s/it]
Training Epoch: 16/100, batch 44/563 completed (loss: 0.0325:   8%|▊         | 22/281 [02:13<25:51,  5.99s/it]
Training Epoch: 16/100, batch 44/563 completed (loss: 0.1049:   8%|▊         | 22/281 [02:13<25:50,  5.98s/it]
Training Epoch: 16/100, batch 44/563 completed (loss: 0.1138:   8%|▊         | 22/281 [02:13<25:50,  5.98s/it]
Training Epoch: 16/100, batch 44/563 completed (loss: 0.1138:   8%|▊         | 23/281 [02:16<25:42,  5.98s/it]
Training Epoch: 16/100, batch 44/563 completed (loss: 0.0325:   8%|▊         | 23/281 [02:16<25:43,  5.98s/it]
Training Epoch: 16/100, batch 44/563 completed (loss: 0.1049:   8%|▊         | 23/281 [02:16<25:44,  5.99s/it]
Training Epoch: 16/100, batch 44/563 completed (loss: 0.0467:   8%|▊         | 23/281 [02:16<25:44,  5.99s/it]
Training Epoch: 16/100, batch 45/563 completed (loss: 0.0052:   8%|▊         | 23/281 [02:16<25:43,  5.98s/it]
Training Epoch: 16/100, batch 45/563 completed (loss: 0.0204:   8%|▊         | 23/281 [02:16<25:44,  5.99s/it]
Training Epoch: 16/100, batch 45/563 completed (loss: 0.0206:   8%|▊         | 23/281 [02:16<25:44,  5.99s/it]
Training Epoch: 16/100, batch 45/563 completed (loss: 0.1945:   8%|▊         | 23/281 [02:16<25:42,  5.98s/it]
Training Epoch: 16/100, batch 46/563 completed (loss: 0.0192:   8%|▊         | 23/281 [02:19<25:44,  5.99s/it]
Training Epoch: 16/100, batch 46/563 completed (loss: 0.0183:   8%|▊         | 23/281 [02:19<25:43,  5.98s/it]
Training Epoch: 16/100, batch 46/563 completed (loss: 0.036:   8%|▊         | 23/281 [02:19<25:44,  5.99s/it] 
Training Epoch: 16/100, batch 46/563 completed (loss: 0.02:   8%|▊         | 23/281 [02:19<25:42,  5.98s/it]  
Training Epoch: 16/100, batch 46/563 completed (loss: 0.02:   9%|▊         | 24/281 [02:22<25:26,  5.94s/it]
Training Epoch: 16/100, batch 46/563 completed (loss: 0.0183:   9%|▊         | 24/281 [02:22<25:26,  5.94s/it]
Training Epoch: 16/100, batch 46/563 completed (loss: 0.036:   9%|▊         | 24/281 [02:22<25:26,  5.94s/it]
Training Epoch: 16/100, batch 46/563 completed (loss: 0.0192:   9%|▊         | 24/281 [02:22<25:26,  5.94s/it]
Training Epoch: 16/100, batch 47/563 completed (loss: 0.0332:   9%|▊         | 24/281 [02:22<25:26,  5.94s/it]
Training Epoch: 16/100, batch 47/563 completed (loss: 0.0316:   9%|▊         | 24/281 [02:22<25:26,  5.94s/it]
Training Epoch: 16/100, batch 47/563 completed (loss: 0.0408:   9%|▊         | 24/281 [02:22<25:26,  5.94s/it]
Training Epoch: 16/100, batch 47/563 completed (loss: 0.0541:   9%|▊         | 24/281 [02:22<25:26,  5.94s/it]
Training Epoch: 16/100, batch 48/563 completed (loss: 0.0695:   9%|▊         | 24/281 [02:25<25:26,  5.94s/it]
Training Epoch: 16/100, batch 48/563 completed (loss: 0.1123:   9%|▊         | 24/281 [02:25<25:26,  5.94s/it]
Training Epoch: 16/100, batch 48/563 completed (loss: 0.0928:   9%|▊         | 24/281 [02:25<25:26,  5.94s/it]
Training Epoch: 16/100, batch 48/563 completed (loss: 0.2062:   9%|▊         | 24/281 [02:25<25:26,  5.94s/it]
Training Epoch: 16/100, batch 48/563 completed (loss: 0.2062:   9%|▉         | 25/281 [02:28<26:02,  6.10s/it]
Training Epoch: 16/100, batch 48/563 completed (loss: 0.0695:   9%|▉         | 25/281 [02:28<26:01,  6.10s/it]
Training Epoch: 16/100, batch 48/563 completed (loss: 0.1123:   9%|▉         | 25/281 [02:28<26:01,  6.10s/it]
Training Epoch: 16/100, batch 48/563 completed (loss: 0.0928:   9%|▉         | 25/281 [02:28<26:02,  6.10s/it]
Training Epoch: 16/100, batch 49/563 completed (loss: 0.0492:   9%|▉         | 25/281 [02:28<26:01,  6.10s/it]
Training Epoch: 16/100, batch 49/563 completed (loss: 0.2911:   9%|▉         | 25/281 [02:28<26:01,  6.10s/it]
Training Epoch: 16/100, batch 49/563 completed (loss: 0.0415:   9%|▉         | 25/281 [02:28<26:02,  6.10s/it]
Training Epoch: 16/100, batch 49/563 completed (loss: 0.0574:   9%|▉         | 25/281 [02:29<26:02,  6.10s/it]
Training Epoch: 16/100, batch 50/563 completed (loss: 0.2073:   9%|▉         | 25/281 [02:31<26:02,  6.10s/it]
Training Epoch: 16/100, batch 50/563 completed (loss: 0.0286:   9%|▉         | 25/281 [02:31<26:01,  6.10s/it]
Training Epoch: 16/100, batch 50/563 completed (loss: 0.1157:   9%|▉         | 25/281 [02:31<26:01,  6.10s/it]
Training Epoch: 16/100, batch 50/563 completed (loss: 0.2121:   9%|▉         | 25/281 [02:31<26:02,  6.10s/it]
Training Epoch: 16/100, batch 50/563 completed (loss: 0.2121:   9%|▉         | 26/281 [02:34<25:32,  6.01s/it]
Training Epoch: 16/100, batch 50/563 completed (loss: 0.1157:   9%|▉         | 26/281 [02:34<25:32,  6.01s/it]
Training Epoch: 16/100, batch 50/563 completed (loss: 0.0286:   9%|▉         | 26/281 [02:34<25:32,  6.01s/it]
Training Epoch: 16/100, batch 50/563 completed (loss: 0.2073:   9%|▉         | 26/281 [02:34<25:32,  6.01s/it]
Training Epoch: 16/100, batch 51/563 completed (loss: 0.1892:   9%|▉         | 26/281 [02:34<25:32,  6.01s/it]
Training Epoch: 16/100, batch 51/563 completed (loss: 0.0773:   9%|▉         | 26/281 [02:34<25:32,  6.01s/it]
Training Epoch: 16/100, batch 51/563 completed (loss: 0.2063:   9%|▉         | 26/281 [02:34<25:32,  6.01s/it]
Training Epoch: 16/100, batch 51/563 completed (loss: 0.0139:   9%|▉         | 26/281 [02:34<25:32,  6.01s/it]
Training Epoch: 16/100, batch 52/563 completed (loss: 0.0084:   9%|▉         | 26/281 [02:37<25:32,  6.01s/it]
Training Epoch: 16/100, batch 52/563 completed (loss: 0.082:   9%|▉         | 26/281 [02:37<25:32,  6.01s/it] 
Training Epoch: 16/100, batch 52/563 completed (loss: 0.0324:   9%|▉         | 26/281 [02:37<25:32,  6.01s/it]
Training Epoch: 16/100, batch 52/563 completed (loss: 0.0428:   9%|▉         | 26/281 [02:37<25:32,  6.01s/it]
Training Epoch: 16/100, batch 52/563 completed (loss: 0.0428:  10%|▉         | 27/281 [02:40<25:21,  5.99s/it]
Training Epoch: 16/100, batch 52/563 completed (loss: 0.082:  10%|▉         | 27/281 [02:40<25:21,  5.99s/it]
Training Epoch: 16/100, batch 52/563 completed (loss: 0.0324:  10%|▉         | 27/281 [02:40<25:21,  5.99s/it]
Training Epoch: 16/100, batch 52/563 completed (loss: 0.0084:  10%|▉         | 27/281 [02:40<25:21,  5.99s/it]
Training Epoch: 16/100, batch 53/563 completed (loss: 0.0298:  10%|▉         | 27/281 [02:40<25:21,  5.99s/it]
Training Epoch: 16/100, batch 53/563 completed (loss: 0.2667:  10%|▉         | 27/281 [02:40<25:21,  5.99s/it]
Training Epoch: 16/100, batch 53/563 completed (loss: 0.2336:  10%|▉         | 27/281 [02:40<25:21,  5.99s/it]
Training Epoch: 16/100, batch 53/563 completed (loss: 0.0134:  10%|▉         | 27/281 [02:40<25:21,  5.99s/it]
Training Epoch: 16/100, batch 54/563 completed (loss: 0.0275:  10%|▉         | 27/281 [02:44<25:21,  5.99s/it]
Training Epoch: 16/100, batch 54/563 completed (loss: 0.1126:  10%|▉         | 27/281 [02:44<25:21,  5.99s/it]
Training Epoch: 16/100, batch 54/563 completed (loss: 0.0162:  10%|▉         | 27/281 [02:44<25:21,  5.99s/it]
Training Epoch: 16/100, batch 54/563 completed (loss: 0.0257:  10%|▉         | 27/281 [02:44<25:21,  5.99s/it]
Training Epoch: 16/100, batch 54/563 completed (loss: 0.1126:  10%|▉         | 28/281 [02:46<25:34,  6.07s/it]
Training Epoch: 16/100, batch 54/563 completed (loss: 0.0275:  10%|▉         | 28/281 [02:46<25:35,  6.07s/it]
Training Epoch: 16/100, batch 54/563 completed (loss: 0.0257:  10%|▉         | 28/281 [02:46<25:36,  6.07s/it]
Training Epoch: 16/100, batch 54/563 completed (loss: 0.0162:  10%|▉         | 28/281 [02:46<25:35,  6.07s/it]
Training Epoch: 16/100, batch 55/563 completed (loss: 0.0331:  10%|▉         | 28/281 [02:46<25:34,  6.07s/it]
Training Epoch: 16/100, batch 55/563 completed (loss: 0.0104:  10%|▉         | 28/281 [02:46<25:35,  6.07s/it]
Training Epoch: 16/100, batch 55/563 completed (loss: 0.1241:  10%|▉         | 28/281 [02:46<25:35,  6.07s/it]
Training Epoch: 16/100, batch 55/563 completed (loss: 0.0317:  10%|▉         | 28/281 [02:47<25:36,  6.07s/it]
Training Epoch: 16/100, batch 56/563 completed (loss: 0.123:  10%|▉         | 28/281 [02:49<25:34,  6.07s/it] 
Training Epoch: 16/100, batch 56/563 completed (loss: 0.0267:  10%|▉         | 28/281 [02:49<25:35,  6.07s/it]
Training Epoch: 16/100, batch 56/563 completed (loss: 0.1942:  10%|▉         | 28/281 [02:49<25:35,  6.07s/it]
Training Epoch: 16/100, batch 56/563 completed (loss: 0.0742:  10%|▉         | 28/281 [02:49<25:36,  6.07s/it]
Training Epoch: 16/100, batch 56/563 completed (loss: 0.0742:  10%|█         | 29/281 [02:52<25:11,  6.00s/it]
Training Epoch: 16/100, batch 56/563 completed (loss: 0.1942:  10%|█         | 29/281 [02:52<25:11,  6.00s/it]
Training Epoch: 16/100, batch 56/563 completed (loss: 0.123:  10%|█         | 29/281 [02:52<25:11,  6.00s/it]
Training Epoch: 16/100, batch 56/563 completed (loss: 0.0267:  10%|█         | 29/281 [02:52<25:11,  6.00s/it]
Training Epoch: 16/100, batch 57/563 completed (loss: 0.051:  10%|█         | 29/281 [02:52<25:11,  6.00s/it] 
Training Epoch: 16/100, batch 57/563 completed (loss: 0.0115:  10%|█         | 29/281 [02:52<25:11,  6.00s/it]
Training Epoch: 16/100, batch 57/563 completed (loss: 0.04:  10%|█         | 29/281 [02:52<25:11,  6.00s/it]  
Training Epoch: 16/100, batch 57/563 completed (loss: 0.0019:  10%|█         | 29/281 [02:52<25:11,  6.00s/it]
Training Epoch: 16/100, batch 58/563 completed (loss: 0.2727:  10%|█         | 29/281 [02:55<25:11,  6.00s/it]
Training Epoch: 16/100, batch 58/563 completed (loss: 0.0579:  10%|█         | 29/281 [02:55<25:11,  6.00s/it]
Training Epoch: 16/100, batch 58/563 completed (loss: 0.0134:  10%|█         | 29/281 [02:55<25:11,  6.00s/it]
Training Epoch: 16/100, batch 58/563 completed (loss: 0.2358:  10%|█         | 29/281 [02:55<25:11,  6.00s/it]
Training Epoch: 16/100, batch 58/563 completed (loss: 0.0579:  11%|█         | 30/281 [02:58<25:24,  6.07s/it]
Training Epoch: 16/100, batch 58/563 completed (loss: 0.2358:  11%|█         | 30/281 [02:59<25:24,  6.08s/it]
Training Epoch: 16/100, batch 58/563 completed (loss: 0.0134:  11%|█         | 30/281 [02:59<25:24,  6.07s/it]
Training Epoch: 16/100, batch 58/563 completed (loss: 0.2727:  11%|█         | 30/281 [02:59<25:24,  6.07s/it]
Training Epoch: 16/100, batch 59/563 completed (loss: 0.0199:  11%|█         | 30/281 [02:59<25:24,  6.07s/it]
Training Epoch: 16/100, batch 59/563 completed (loss: 0.0147:  11%|█         | 30/281 [02:59<25:24,  6.07s/it]
Training Epoch: 16/100, batch 59/563 completed (loss: 0.078:  11%|█         | 30/281 [02:59<25:24,  6.07s/it] 
Training Epoch: 16/100, batch 59/563 completed (loss: 0.0111:  11%|█         | 30/281 [02:59<25:24,  6.08s/it]
Training Epoch: 16/100, batch 60/563 completed (loss: 0.0038:  11%|█         | 30/281 [03:02<25:24,  6.07s/it]
Training Epoch: 16/100, batch 60/563 completed (loss: 0.0836:  11%|█         | 30/281 [03:02<25:24,  6.07s/it]
Training Epoch: 16/100, batch 60/563 completed (loss: 0.0563:  11%|█         | 30/281 [03:02<25:24,  6.07s/it]
Training Epoch: 16/100, batch 60/563 completed (loss: 0.0577:  11%|█         | 30/281 [03:02<25:24,  6.08s/it]
Training Epoch: 16/100, batch 60/563 completed (loss: 0.0836:  11%|█         | 31/281 [03:05<25:13,  6.05s/it]
Training Epoch: 16/100, batch 60/563 completed (loss: 0.0563:  11%|█         | 31/281 [03:05<25:12,  6.05s/it]
Training Epoch: 16/100, batch 60/563 completed (loss: 0.0577:  11%|█         | 31/281 [03:05<25:13,  6.06s/it]
Training Epoch: 16/100, batch 60/563 completed (loss: 0.0038:  11%|█         | 31/281 [03:05<25:13,  6.06s/it]
Training Epoch: 16/100, batch 61/563 completed (loss: 0.0589:  11%|█         | 31/281 [03:05<25:12,  6.05s/it]
Training Epoch: 16/100, batch 61/563 completed (loss: 0.0062:  11%|█         | 31/281 [03:05<25:13,  6.06s/it]
Training Epoch: 16/100, batch 61/563 completed (loss: 0.2069:  11%|█         | 31/281 [03:05<25:13,  6.05s/it]
Training Epoch: 16/100, batch 61/563 completed (loss: 0.0127:  11%|█         | 31/281 [03:05<25:13,  6.06s/it]
Training Epoch: 16/100, batch 62/563 completed (loss: 0.1674:  11%|█         | 31/281 [03:07<25:12,  6.05s/it]
Training Epoch: 16/100, batch 62/563 completed (loss: 0.0877:  11%|█         | 31/281 [03:07<25:13,  6.06s/it]
Training Epoch: 16/100, batch 62/563 completed (loss: 0.0244:  11%|█         | 31/281 [03:07<25:13,  6.05s/it]
Training Epoch: 16/100, batch 62/563 completed (loss: 0.2625:  11%|█         | 31/281 [03:08<25:13,  6.06s/it]
Training Epoch: 16/100, batch 62/563 completed (loss: 0.1674:  11%|█▏        | 32/281 [03:10<24:50,  5.99s/it]
Training Epoch: 16/100, batch 62/563 completed (loss: 0.0877:  11%|█▏        | 32/281 [03:10<24:50,  5.99s/it]
Training Epoch: 16/100, batch 62/563 completed (loss: 0.2625:  11%|█▏        | 32/281 [03:10<24:52,  5.99s/it]
Training Epoch: 16/100, batch 62/563 completed (loss: 0.0244:  11%|█▏        | 32/281 [03:10<24:52,  5.99s/it]
Training Epoch: 16/100, batch 63/563 completed (loss: 0.0336:  11%|█▏        | 32/281 [03:10<24:50,  5.99s/it]
Training Epoch: 16/100, batch 63/563 completed (loss: 0.0303:  11%|█▏        | 32/281 [03:10<24:50,  5.99s/it]
Training Epoch: 16/100, batch 63/563 completed (loss: 0.1406:  11%|█▏        | 32/281 [03:10<24:52,  5.99s/it]
Training Epoch: 16/100, batch 63/563 completed (loss: 0.0109:  11%|█▏        | 32/281 [03:10<24:52,  5.99s/it]
Training Epoch: 16/100, batch 64/563 completed (loss: 0.0128:  11%|█▏        | 32/281 [03:14<24:50,  5.99s/it]
Training Epoch: 16/100, batch 64/563 completed (loss: 0.092:  11%|█▏        | 32/281 [03:14<24:52,  5.99s/it] 
Training Epoch: 16/100, batch 64/563 completed (loss: 0.0507:  11%|█▏        | 32/281 [03:14<24:50,  5.99s/it]
Training Epoch: 16/100, batch 64/563 completed (loss: 0.0439:  11%|█▏        | 32/281 [03:14<24:52,  5.99s/it]
Training Epoch: 16/100, batch 64/563 completed (loss: 0.0507:  12%|█▏        | 33/281 [03:17<25:16,  6.12s/it]
Training Epoch: 16/100, batch 64/563 completed (loss: 0.092:  12%|█▏        | 33/281 [03:17<25:17,  6.12s/it]
Training Epoch: 16/100, batch 64/563 completed (loss: 0.0439:  12%|█▏        | 33/281 [03:17<25:17,  6.12s/it]
Training Epoch: 16/100, batch 64/563 completed (loss: 0.0128:  12%|█▏        | 33/281 [03:17<25:17,  6.12s/it]
Training Epoch: 16/100, batch 65/563 completed (loss: 0.1215:  12%|█▏        | 33/281 [03:17<25:17,  6.12s/it]
Training Epoch: 16/100, batch 65/563 completed (loss: 0.0097:  12%|█▏        | 33/281 [03:17<25:17,  6.12s/it]
Training Epoch: 16/100, batch 65/563 completed (loss: 0.0254:  12%|█▏        | 33/281 [03:17<25:16,  6.12s/it]
Training Epoch: 16/100, batch 65/563 completed (loss: 0.0299:  12%|█▏        | 33/281 [03:17<25:17,  6.12s/it]
Training Epoch: 16/100, batch 66/563 completed (loss: 0.0142:  12%|█▏        | 33/281 [03:20<25:17,  6.12s/it]
Training Epoch: 16/100, batch 66/563 completed (loss: 0.1514:  12%|█▏        | 33/281 [03:20<25:16,  6.12s/it]
Training Epoch: 16/100, batch 66/563 completed (loss: 0.0762:  12%|█▏        | 33/281 [03:20<25:17,  6.12s/it]
Training Epoch: 16/100, batch 66/563 completed (loss: 0.0116:  12%|█▏        | 33/281 [03:20<25:17,  6.12s/it]
Training Epoch: 16/100, batch 66/563 completed (loss: 0.0116:  12%|█▏        | 34/281 [03:23<24:47,  6.02s/it]
Training Epoch: 16/100, batch 66/563 completed (loss: 0.0142:  12%|█▏        | 34/281 [03:23<24:49,  6.03s/it]
Training Epoch: 16/100, batch 66/563 completed (loss: 0.1514:  12%|█▏        | 34/281 [03:23<24:49,  6.03s/it]
Training Epoch: 16/100, batch 66/563 completed (loss: 0.0762:  12%|█▏        | 34/281 [03:23<24:50,  6.03s/it]
Training Epoch: 16/100, batch 67/563 completed (loss: 0.1217:  12%|█▏        | 34/281 [03:23<24:49,  6.03s/it]
Training Epoch: 16/100, batch 67/563 completed (loss: 0.0378:  12%|█▏        | 34/281 [03:23<24:50,  6.03s/it]
Training Epoch: 16/100, batch 67/563 completed (loss: 0.0756:  12%|█▏        | 34/281 [03:23<24:49,  6.03s/it]
Training Epoch: 16/100, batch 67/563 completed (loss: 0.0212:  12%|█▏        | 34/281 [03:23<24:47,  6.02s/it]
Training Epoch: 16/100, batch 68/563 completed (loss: 0.0101:  12%|█▏        | 34/281 [03:26<24:50,  6.03s/it]
Training Epoch: 16/100, batch 68/563 completed (loss: 0.1:  12%|█▏        | 34/281 [03:26<24:49,  6.03s/it]   
Training Epoch: 16/100, batch 68/563 completed (loss: 0.0489:  12%|█▏        | 34/281 [03:26<24:49,  6.03s/it]
Training Epoch: 16/100, batch 68/563 completed (loss: 0.0662:  12%|█▏        | 34/281 [03:26<24:47,  6.02s/it]
Training Epoch: 16/100, batch 68/563 completed (loss: 0.0662:  12%|█▏        | 35/281 [03:29<25:05,  6.12s/it]
Training Epoch: 16/100, batch 68/563 completed (loss: 0.1:  12%|█▏        | 35/281 [03:29<25:04,  6.12s/it]
Training Epoch: 16/100, batch 68/563 completed (loss: 0.0101:  12%|█▏        | 35/281 [03:29<25:04,  6.12s/it]
Training Epoch: 16/100, batch 68/563 completed (loss: 0.0489:  12%|█▏        | 35/281 [03:29<25:05,  6.12s/it]
Training Epoch: 16/100, batch 69/563 completed (loss: 0.022:  12%|█▏        | 35/281 [03:29<25:04,  6.12s/it]
Training Epoch: 16/100, batch 69/563 completed (loss: 0.0435:  12%|█▏        | 35/281 [03:29<25:04,  6.12s/it]
Training Epoch: 16/100, batch 69/563 completed (loss: 0.3325:  12%|█▏        | 35/281 [03:29<25:05,  6.12s/it]
Training Epoch: 16/100, batch 69/563 completed (loss: 0.0354:  12%|█▏        | 35/281 [03:29<25:05,  6.12s/it]
Training Epoch: 16/100, batch 70/563 completed (loss: 0.0489:  12%|█▏        | 35/281 [03:32<25:05,  6.12s/it]
Training Epoch: 16/100, batch 70/563 completed (loss: 0.0094:  12%|█▏        | 35/281 [03:32<25:04,  6.12s/it]
Training Epoch: 16/100, batch 70/563 completed (loss: 0.0346:  12%|█▏        | 35/281 [03:32<25:04,  6.12s/it]
Training Epoch: 16/100, batch 70/563 completed (loss: 0.0417:  12%|█▏        | 35/281 [03:32<25:05,  6.12s/it]
Training Epoch: 16/100, batch 70/563 completed (loss: 0.0346:  13%|█▎        | 36/281 [03:35<24:49,  6.08s/it]
Training Epoch: 16/100, batch 70/563 completed (loss: 0.0489:  13%|█▎        | 36/281 [03:35<24:50,  6.08s/it]
Training Epoch: 16/100, batch 70/563 completed (loss: 0.0094:  13%|█▎        | 36/281 [03:35<24:50,  6.08s/it]
Training Epoch: 16/100, batch 70/563 completed (loss: 0.0417:  13%|█▎        | 36/281 [03:35<24:50,  6.08s/it]
Training Epoch: 16/100, batch 71/563 completed (loss: 0.028:  13%|█▎        | 36/281 [03:35<24:50,  6.08s/it] 
Training Epoch: 16/100, batch 71/563 completed (loss: 0.0198:  13%|█▎        | 36/281 [03:35<24:49,  6.08s/it]
Training Epoch: 16/100, batch 71/563 completed (loss: 0.1599:  13%|█▎        | 36/281 [03:35<24:50,  6.08s/it]
Training Epoch: 16/100, batch 71/563 completed (loss: 0.0209:  13%|█▎        | 36/281 [03:35<24:50,  6.08s/it]
Training Epoch: 16/100, batch 72/563 completed (loss: 0.069:  13%|█▎        | 36/281 [03:38<24:49,  6.08s/it] 
Training Epoch: 16/100, batch 72/563 completed (loss: 0.0128:  13%|█▎        | 36/281 [03:38<24:50,  6.08s/it]
Training Epoch: 16/100, batch 72/563 completed (loss: 0.0395:  13%|█▎        | 36/281 [03:38<24:50,  6.08s/it]
Training Epoch: 16/100, batch 72/563 completed (loss: 0.0143:  13%|█▎        | 36/281 [03:38<24:50,  6.08s/it]
Training Epoch: 16/100, batch 72/563 completed (loss: 0.0143:  13%|█▎        | 37/281 [03:41<24:38,  6.06s/it]
Training Epoch: 16/100, batch 72/563 completed (loss: 0.0128:  13%|█▎        | 37/281 [03:41<24:39,  6.06s/it]
Training Epoch: 16/100, batch 72/563 completed (loss: 0.0395:  13%|█▎        | 37/281 [03:41<24:39,  6.07s/it]
Training Epoch: 16/100, batch 72/563 completed (loss: 0.069:  13%|█▎        | 37/281 [03:41<24:40,  6.07s/it]
Training Epoch: 16/100, batch 73/563 completed (loss: 0.1473:  13%|█▎        | 37/281 [03:41<24:39,  6.06s/it]
Training Epoch: 16/100, batch 73/563 completed (loss: 0.0241:  13%|█▎        | 37/281 [03:41<24:39,  6.07s/it]
Training Epoch: 16/100, batch 73/563 completed (loss: 0.0171:  13%|█▎        | 37/281 [03:41<24:40,  6.07s/it]
Training Epoch: 16/100, batch 73/563 completed (loss: 0.0087:  13%|█▎        | 37/281 [03:41<24:38,  6.06s/it]
Training Epoch: 16/100, batch 74/563 completed (loss: 0.0222:  13%|█▎        | 37/281 [03:44<24:40,  6.07s/it]
Training Epoch: 16/100, batch 74/563 completed (loss: 0.0078:  13%|█▎        | 37/281 [03:44<24:39,  6.07s/it]
Training Epoch: 16/100, batch 74/563 completed (loss: 0.0061:  13%|█▎        | 37/281 [03:44<24:39,  6.06s/it]
Training Epoch: 16/100, batch 74/563 completed (loss: 0.1144:  13%|█▎        | 37/281 [03:44<24:38,  6.06s/it]
Training Epoch: 16/100, batch 74/563 completed (loss: 0.1144:  14%|█▎        | 38/281 [03:47<24:50,  6.13s/it]
Training Epoch: 16/100, batch 74/563 completed (loss: 0.0222:  14%|█▎        | 38/281 [03:47<24:50,  6.13s/it]
Training Epoch: 16/100, batch 74/563 completed (loss: 0.0061:  14%|█▎        | 38/281 [03:47<24:50,  6.13s/it]
Training Epoch: 16/100, batch 74/563 completed (loss: 0.0078:  14%|█▎        | 38/281 [03:47<24:50,  6.14s/it]
Training Epoch: 16/100, batch 75/563 completed (loss: 0.0161:  14%|█▎        | 38/281 [03:47<24:50,  6.14s/it]
Training Epoch: 16/100, batch 75/563 completed (loss: 0.1217:  14%|█▎        | 38/281 [03:47<24:50,  6.13s/it]
Training Epoch: 16/100, batch 75/563 completed (loss: 0.0104:  14%|█▎        | 38/281 [03:47<24:50,  6.13s/it]
Training Epoch: 16/100, batch 75/563 completed (loss: 0.0214:  14%|█▎        | 38/281 [03:47<24:50,  6.13s/it]
Training Epoch: 16/100, batch 76/563 completed (loss: 0.0036:  14%|█▎        | 38/281 [03:50<24:50,  6.13s/it]
Training Epoch: 16/100, batch 76/563 completed (loss: 0.0359:  14%|█▎        | 38/281 [03:50<24:50,  6.14s/it]
Training Epoch: 16/100, batch 76/563 completed (loss: 0.0378:  14%|█▎        | 38/281 [03:50<24:50,  6.13s/it]
Training Epoch: 16/100, batch 76/563 completed (loss: 0.1894:  14%|█▎        | 38/281 [03:50<24:50,  6.13s/it]
Training Epoch: 16/100, batch 76/563 completed (loss: 0.1894:  14%|█▍        | 39/281 [03:53<24:34,  6.09s/it]
Training Epoch: 16/100, batch 76/563 completed (loss: 0.0359:  14%|█▍        | 39/281 [03:53<24:34,  6.09s/it]
Training Epoch: 16/100, batch 76/563 completed (loss: 0.0036:  14%|█▍        | 39/281 [03:53<24:34,  6.09s/it]
Training Epoch: 16/100, batch 76/563 completed (loss: 0.0378:  14%|█▍        | 39/281 [03:53<24:34,  6.09s/it]
Training Epoch: 16/100, batch 77/563 completed (loss: 0.0832:  14%|█▍        | 39/281 [03:53<24:34,  6.09s/it]
Training Epoch: 16/100, batch 77/563 completed (loss: 0.1202:  14%|█▍        | 39/281 [03:53<24:34,  6.09s/it]
Training Epoch: 16/100, batch 77/563 completed (loss: 0.0069:  14%|█▍        | 39/281 [03:53<24:34,  6.09s/it]
Training Epoch: 16/100, batch 77/563 completed (loss: 0.0132:  14%|█▍        | 39/281 [03:53<24:34,  6.09s/it]
Training Epoch: 16/100, batch 78/563 completed (loss: 0.0678:  14%|█▍        | 39/281 [03:56<24:34,  6.09s/it]
Training Epoch: 16/100, batch 78/563 completed (loss: 0.005:  14%|█▍        | 39/281 [03:56<24:34,  6.09s/it] 
Training Epoch: 16/100, batch 78/563 completed (loss: 0.0061:  14%|█▍        | 39/281 [03:56<24:34,  6.09s/it]
Training Epoch: 16/100, batch 78/563 completed (loss: 0.1054:  14%|█▍        | 39/281 [03:56<24:34,  6.09s/it]
Training Epoch: 16/100, batch 78/563 completed (loss: 0.1054:  14%|█▍        | 40/281 [03:59<24:33,  6.11s/it]
Training Epoch: 16/100, batch 78/563 completed (loss: 0.0678:  14%|█▍        | 40/281 [03:59<24:34,  6.12s/it]
Training Epoch: 16/100, batch 78/563 completed (loss: 0.005:  14%|█▍        | 40/281 [03:59<24:34,  6.12s/it]
Training Epoch: 16/100, batch 78/563 completed (loss: 0.0061:  14%|█▍        | 40/281 [03:59<24:35,  6.12s/it]
Training Epoch: 16/100, batch 79/563 completed (loss: 0.2946:  14%|█▍        | 40/281 [03:59<24:34,  6.12s/it]
Training Epoch: 16/100, batch 79/563 completed (loss: 0.0348:  14%|█▍        | 40/281 [03:59<24:35,  6.12s/it]
Training Epoch: 16/100, batch 79/563 completed (loss: 0.0127:  14%|█▍        | 40/281 [03:59<24:34,  6.12s/it]
Training Epoch: 16/100, batch 79/563 completed (loss: 0.0145:  14%|█▍        | 40/281 [04:00<24:33,  6.11s/it]
Training Epoch: 16/100, batch 80/563 completed (loss: 0.0285:  14%|█▍        | 40/281 [04:02<24:35,  6.12s/it]
Training Epoch: 16/100, batch 80/563 completed (loss: 0.0109:  14%|█▍        | 40/281 [04:02<24:34,  6.12s/it]
Training Epoch: 16/100, batch 80/563 completed (loss: 0.0725:  14%|█▍        | 40/281 [04:02<24:34,  6.12s/it]
Training Epoch: 16/100, batch 80/563 completed (loss: 0.0355:  14%|█▍        | 40/281 [04:03<24:33,  6.11s/it]
Training Epoch: 16/100, batch 80/563 completed (loss: 0.0109:  15%|█▍        | 41/281 [04:05<24:13,  6.06s/it]
Training Epoch: 16/100, batch 80/563 completed (loss: 0.0355:  15%|█▍        | 41/281 [04:05<24:14,  6.06s/it]
Training Epoch: 16/100, batch 80/563 completed (loss: 0.0285:  15%|█▍        | 41/281 [04:05<24:13,  6.06s/it]
Training Epoch: 16/100, batch 80/563 completed (loss: 0.0725:  15%|█▍        | 41/281 [04:05<24:13,  6.06s/it]
Training Epoch: 16/100, batch 81/563 completed (loss: 0.1912:  15%|█▍        | 41/281 [04:05<24:13,  6.06s/it]
Training Epoch: 16/100, batch 81/563 completed (loss: 0.0139:  15%|█▍        | 41/281 [04:05<24:13,  6.06s/it]
Training Epoch: 16/100, batch 81/563 completed (loss: 0.0441:  15%|█▍        | 41/281 [04:05<24:13,  6.06s/it]
Training Epoch: 16/100, batch 81/563 completed (loss: 0.0115:  15%|█▍        | 41/281 [04:05<24:14,  6.06s/it]
Training Epoch: 16/100, batch 82/563 completed (loss: 0.0733:  15%|█▍        | 41/281 [04:08<24:13,  6.06s/it]
Training Epoch: 16/100, batch 82/563 completed (loss: 0.0922:  15%|█▍        | 41/281 [04:08<24:13,  6.06s/it]
Training Epoch: 16/100, batch 82/563 completed (loss: 0.0034:  15%|█▍        | 41/281 [04:08<24:13,  6.06s/it]
Training Epoch: 16/100, batch 82/563 completed (loss: 0.0682:  15%|█▍        | 41/281 [04:08<24:14,  6.06s/it]
Training Epoch: 16/100, batch 82/563 completed (loss: 0.0034:  15%|█▍        | 42/281 [04:11<24:01,  6.03s/it]
Training Epoch: 16/100, batch 82/563 completed (loss: 0.0733:  15%|█▍        | 42/281 [04:11<24:01,  6.03s/it]
Training Epoch: 16/100, batch 82/563 completed (loss: 0.0682:  15%|█▍        | 42/281 [04:11<24:02,  6.04s/it]
Training Epoch: 16/100, batch 82/563 completed (loss: 0.0922:  15%|█▍        | 42/281 [04:11<24:01,  6.03s/it]
Training Epoch: 16/100, batch 83/563 completed (loss: 0.0096:  15%|█▍        | 42/281 [04:11<24:01,  6.03s/it]
Training Epoch: 16/100, batch 83/563 completed (loss: 0.1275:  15%|█▍        | 42/281 [04:11<24:01,  6.03s/it]
Training Epoch: 16/100, batch 83/563 completed (loss: 0.0947:  15%|█▍        | 42/281 [04:11<24:01,  6.03s/it]
Training Epoch: 16/100, batch 83/563 completed (loss: 0.1827:  15%|█▍        | 42/281 [04:11<24:02,  6.04s/it]
Training Epoch: 16/100, batch 84/563 completed (loss: 0.0102:  15%|█▍        | 42/281 [04:14<24:01,  6.03s/it]
Training Epoch: 16/100, batch 84/563 completed (loss: 0.076:  15%|█▍        | 42/281 [04:14<24:01,  6.03s/it] 
Training Epoch: 16/100, batch 84/563 completed (loss: 0.0802:  15%|█▍        | 42/281 [04:14<24:01,  6.03s/it]
Training Epoch: 16/100, batch 84/563 completed (loss: 0.1728:  15%|█▍        | 42/281 [04:15<24:02,  6.04s/it]
Training Epoch: 16/100, batch 84/563 completed (loss: 0.1728:  15%|█▌        | 43/281 [04:18<24:08,  6.09s/it]
Training Epoch: 16/100, batch 84/563 completed (loss: 0.076:  15%|█▌        | 43/281 [04:18<24:09,  6.09s/it]
Training Epoch: 16/100, batch 84/563 completed (loss: 0.0102:  15%|█▌        | 43/281 [04:18<24:09,  6.09s/it]
Training Epoch: 16/100, batch 84/563 completed (loss: 0.0802:  15%|█▌        | 43/281 [04:18<24:10,  6.09s/it]
Training Epoch: 16/100, batch 85/563 completed (loss: 0.1219:  15%|█▌        | 43/281 [04:18<24:10,  6.09s/it]
Training Epoch: 16/100, batch 85/563 completed (loss: 0.0314:  15%|█▌        | 43/281 [04:18<24:09,  6.09s/it]
Training Epoch: 16/100, batch 85/563 completed (loss: 0.0169:  15%|█▌        | 43/281 [04:18<24:09,  6.09s/it]
Training Epoch: 16/100, batch 85/563 completed (loss: 0.0083:  15%|█▌        | 43/281 [04:18<24:08,  6.09s/it]
Training Epoch: 16/100, batch 86/563 completed (loss: 0.0646:  15%|█▌        | 43/281 [04:21<24:10,  6.09s/it]
Training Epoch: 16/100, batch 86/563 completed (loss: 0.1713:  15%|█▌        | 43/281 [04:21<24:09,  6.09s/it]
Training Epoch: 16/100, batch 86/563 completed (loss: 0.0055:  15%|█▌        | 43/281 [04:21<24:09,  6.09s/it]
Training Epoch: 16/100, batch 86/563 completed (loss: 0.2354:  15%|█▌        | 43/281 [04:21<24:08,  6.09s/it]
Training Epoch: 16/100, batch 86/563 completed (loss: 0.2354:  16%|█▌        | 44/281 [04:24<23:57,  6.07s/it]
Training Epoch: 16/100, batch 86/563 completed (loss: 0.0055:  16%|█▌        | 44/281 [04:24<23:57,  6.06s/it]
Training Epoch: 16/100, batch 86/563 completed (loss: 0.0646:  16%|█▌        | 44/281 [04:24<23:57,  6.06s/it]
Training Epoch: 16/100, batch 86/563 completed (loss: 0.1713:  16%|█▌        | 44/281 [04:24<23:58,  6.07s/it]
Training Epoch: 16/100, batch 87/563 completed (loss: 0.0121:  16%|█▌        | 44/281 [04:24<23:58,  6.07s/it]
Training Epoch: 16/100, batch 87/563 completed (loss: 0.0207:  16%|█▌        | 44/281 [04:24<23:57,  6.06s/it]
Training Epoch: 16/100, batch 87/563 completed (loss: 0.0829:  16%|█▌        | 44/281 [04:24<23:57,  6.06s/it]
Training Epoch: 16/100, batch 87/563 completed (loss: 0.0842:  16%|█▌        | 44/281 [04:24<23:57,  6.07s/it]
Training Epoch: 16/100, batch 88/563 completed (loss: 0.0849:  16%|█▌        | 44/281 [04:26<23:57,  6.06s/it]
Training Epoch: 16/100, batch 88/563 completed (loss: 0.3363:  16%|█▌        | 44/281 [04:26<23:58,  6.07s/it]
Training Epoch: 16/100, batch 88/563 completed (loss: 0.0225:  16%|█▌        | 44/281 [04:26<23:57,  6.06s/it]
Training Epoch: 16/100, batch 88/563 completed (loss: 0.1069:  16%|█▌        | 44/281 [04:27<23:57,  6.07s/it]
Training Epoch: 16/100, batch 88/563 completed (loss: 0.0849:  16%|█▌        | 45/281 [04:30<23:54,  6.08s/it]
Training Epoch: 16/100, batch 88/563 completed (loss: 0.0225:  16%|█▌        | 45/281 [04:30<23:54,  6.08s/it]
Training Epoch: 16/100, batch 88/563 completed (loss: 0.1069:  16%|█▌        | 45/281 [04:30<23:54,  6.08s/it]
Training Epoch: 16/100, batch 88/563 completed (loss: 0.3363:  16%|█▌        | 45/281 [04:30<23:54,  6.08s/it]
Training Epoch: 16/100, batch 89/563 completed (loss: 0.0226:  16%|█▌        | 45/281 [04:30<23:54,  6.08s/it]
Training Epoch: 16/100, batch 89/563 completed (loss: 0.0585:  16%|█▌        | 45/281 [04:30<23:54,  6.08s/it]
Training Epoch: 16/100, batch 89/563 completed (loss: 0.0074:  16%|█▌        | 45/281 [04:30<23:54,  6.08s/it]
Training Epoch: 16/100, batch 89/563 completed (loss: 0.0089:  16%|█▌        | 45/281 [04:30<23:54,  6.08s/it]
Training Epoch: 16/100, batch 90/563 completed (loss: 0.009:  16%|█▌        | 45/281 [04:33<23:54,  6.08s/it] 
Training Epoch: 16/100, batch 90/563 completed (loss: 0.0063:  16%|█▌        | 45/281 [04:33<23:54,  6.08s/it]
Training Epoch: 16/100, batch 90/563 completed (loss: 0.0191:  16%|█▌        | 45/281 [04:33<23:54,  6.08s/it]
Training Epoch: 16/100, batch 90/563 completed (loss: 0.0177:  16%|█▌        | 45/281 [04:33<23:54,  6.08s/it]
Training Epoch: 16/100, batch 90/563 completed (loss: 0.009:  16%|█▋        | 46/281 [04:36<23:40,  6.05s/it]
Training Epoch: 16/100, batch 90/563 completed (loss: 0.0191:  16%|█▋        | 46/281 [04:36<23:41,  6.05s/it]
Training Epoch: 16/100, batch 90/563 completed (loss: 0.0177:  16%|█▋        | 46/281 [04:36<23:41,  6.05s/it]
Training Epoch: 16/100, batch 90/563 completed (loss: 0.0063:  16%|█▋        | 46/281 [04:36<23:41,  6.05s/it]
Training Epoch: 16/100, batch 91/563 completed (loss: 0.0053:  16%|█▋        | 46/281 [04:36<23:41,  6.05s/it]
Training Epoch: 16/100, batch 91/563 completed (loss: 0.1407:  16%|█▋        | 46/281 [04:36<23:41,  6.05s/it]
Training Epoch: 16/100, batch 91/563 completed (loss: 0.0966:  16%|█▋        | 46/281 [04:36<23:40,  6.05s/it]
Training Epoch: 16/100, batch 91/563 completed (loss: 0.0068:  16%|█▋        | 46/281 [04:36<23:41,  6.05s/it]
Training Epoch: 16/100, batch 92/563 completed (loss: 0.0052:  16%|█▋        | 46/281 [04:39<23:41,  6.05s/it]
Training Epoch: 16/100, batch 92/563 completed (loss: 0.0276:  16%|█▋        | 46/281 [04:39<23:40,  6.05s/it]
Training Epoch: 16/100, batch 92/563 completed (loss: 0.031:  16%|█▋        | 46/281 [04:39<23:41,  6.05s/it] 
Training Epoch: 16/100, batch 92/563 completed (loss: 0.0118:  16%|█▋        | 46/281 [04:39<23:41,  6.05s/it]
Training Epoch: 16/100, batch 92/563 completed (loss: 0.0276:  17%|█▋        | 47/281 [04:42<23:32,  6.04s/it]
Training Epoch: 16/100, batch 92/563 completed (loss: 0.0118:  17%|█▋        | 47/281 [04:42<23:32,  6.04s/it]
Training Epoch: 16/100, batch 92/563 completed (loss: 0.031:  17%|█▋        | 47/281 [04:42<23:32,  6.04s/it]
Training Epoch: 16/100, batch 92/563 completed (loss: 0.0052:  17%|█▋        | 47/281 [04:42<23:32,  6.04s/it]
Training Epoch: 16/100, batch 93/563 completed (loss: 0.0037:  17%|█▋        | 47/281 [04:42<23:32,  6.04s/it]
Training Epoch: 16/100, batch 93/563 completed (loss: 0.0697:  17%|█▋        | 47/281 [04:42<23:32,  6.04s/it]
Training Epoch: 16/100, batch 93/563 completed (loss: 0.0125:  17%|█▋        | 47/281 [04:42<23:32,  6.04s/it]
Training Epoch: 16/100, batch 93/563 completed (loss: 0.0293:  17%|█▋        | 47/281 [04:42<23:32,  6.04s/it]
Training Epoch: 16/100, batch 94/563 completed (loss: 0.0735:  17%|█▋        | 47/281 [04:45<23:32,  6.04s/it]
Training Epoch: 16/100, batch 94/563 completed (loss: 0.0407:  17%|█▋        | 47/281 [04:45<23:32,  6.04s/it]
Training Epoch: 16/100, batch 94/563 completed (loss: 0.013:  17%|█▋        | 47/281 [04:45<23:32,  6.04s/it] 
Training Epoch: 16/100, batch 94/563 completed (loss: 0.1511:  17%|█▋        | 47/281 [04:45<23:32,  6.04s/it]
Training Epoch: 16/100, batch 94/563 completed (loss: 0.1511:  17%|█▋        | 48/281 [04:48<23:46,  6.12s/it]
Training Epoch: 16/100, batch 94/563 completed (loss: 0.0735:  17%|█▋        | 48/281 [04:48<23:45,  6.12s/it]
Training Epoch: 16/100, batch 94/563 completed (loss: 0.0407:  17%|█▋        | 48/281 [04:48<23:46,  6.12s/it]
Training Epoch: 16/100, batch 94/563 completed (loss: 0.013:  17%|█▋        | 48/281 [04:48<23:46,  6.12s/it]
Training Epoch: 16/100, batch 95/563 completed (loss: 0.0088:  17%|█▋        | 48/281 [04:48<23:46,  6.12s/it]
Training Epoch: 16/100, batch 95/563 completed (loss: 0.0172:  17%|█▋        | 48/281 [04:48<23:46,  6.12s/it]
Training Epoch: 16/100, batch 95/563 completed (loss: 0.0052:  17%|█▋        | 48/281 [04:48<23:45,  6.12s/it]
Training Epoch: 16/100, batch 95/563 completed (loss: 0.0367:  17%|█▋        | 48/281 [04:48<23:46,  6.12s/it]
Training Epoch: 16/100, batch 96/563 completed (loss: 0.0288:  17%|█▋        | 48/281 [04:51<23:45,  6.12s/it]
Training Epoch: 16/100, batch 96/563 completed (loss: 0.0135:  17%|█▋        | 48/281 [04:51<23:46,  6.12s/it]
Training Epoch: 16/100, batch 96/563 completed (loss: 0.1364:  17%|█▋        | 48/281 [04:51<23:46,  6.12s/it]
Training Epoch: 16/100, batch 96/563 completed (loss: 0.1435:  17%|█▋        | 48/281 [04:51<23:46,  6.12s/it]
Training Epoch: 16/100, batch 96/563 completed (loss: 0.1364:  17%|█▋        | 49/281 [04:54<23:32,  6.09s/it]
Training Epoch: 16/100, batch 96/563 completed (loss: 0.0135:  17%|█▋        | 49/281 [04:54<23:32,  6.09s/it]
Training Epoch: 16/100, batch 96/563 completed (loss: 0.1435:  17%|█▋        | 49/281 [04:54<23:33,  6.09s/it]
Training Epoch: 16/100, batch 96/563 completed (loss: 0.0288:  17%|█▋        | 49/281 [04:54<23:33,  6.09s/it]
Training Epoch: 16/100, batch 97/563 completed (loss: 0.0111:  17%|█▋        | 49/281 [04:54<23:33,  6.09s/it]
Training Epoch: 16/100, batch 97/563 completed (loss: 0.1166:  17%|█▋        | 49/281 [04:54<23:32,  6.09s/it]
Training Epoch: 16/100, batch 97/563 completed (loss: 0.3857:  17%|█▋        | 49/281 [04:54<23:32,  6.09s/it]
Training Epoch: 16/100, batch 97/563 completed (loss: 0.0321:  17%|█▋        | 49/281 [04:54<23:33,  6.09s/it]
Training Epoch: 16/100, batch 98/563 completed (loss: 0.1965:  17%|█▋        | 49/281 [04:57<23:32,  6.09s/it]
Training Epoch: 16/100, batch 98/563 completed (loss: 0.4863:  17%|█▋        | 49/281 [04:57<23:32,  6.09s/it]
Training Epoch: 16/100, batch 98/563 completed (loss: 0.0374:  17%|█▋        | 49/281 [04:57<23:33,  6.09s/it]
Training Epoch: 16/100, batch 98/563 completed (loss: 0.0147:  17%|█▋        | 49/281 [04:57<23:33,  6.09s/it]
Training Epoch: 16/100, batch 98/563 completed (loss: 0.0147:  18%|█▊        | 50/281 [05:00<23:19,  6.06s/it]
Training Epoch: 16/100, batch 98/563 completed (loss: 0.0374:  18%|█▊        | 50/281 [05:00<23:19,  6.06s/it]
Training Epoch: 16/100, batch 98/563 completed (loss: 0.1965:  18%|█▊        | 50/281 [05:00<23:20,  6.06s/it]
Training Epoch: 16/100, batch 98/563 completed (loss: 0.4863:  18%|█▊        | 50/281 [05:00<23:20,  6.06s/it]
Training Epoch: 16/100, batch 99/563 completed (loss: 0.0619:  18%|█▊        | 50/281 [05:00<23:19,  6.06s/it]
Training Epoch: 16/100, batch 99/563 completed (loss: 0.0151:  18%|█▊        | 50/281 [05:00<23:20,  6.06s/it]
Training Epoch: 16/100, batch 99/563 completed (loss: 0.0565:  18%|█▊        | 50/281 [05:00<23:20,  6.06s/it]
Training Epoch: 16/100, batch 99/563 completed (loss: 0.0461:  18%|█▊        | 50/281 [05:00<23:19,  6.06s/it]
Training Epoch: 16/100, batch 100/563 completed (loss: 0.0783:  18%|█▊        | 50/281 [05:03<23:20,  6.06s/it]
Training Epoch: 16/100, batch 100/563 completed (loss: 0.0683:  18%|█▊        | 50/281 [05:03<23:19,  6.06s/it]
Training Epoch: 16/100, batch 100/563 completed (loss: 0.0094:  18%|█▊        | 50/281 [05:03<23:20,  6.06s/it]
Training Epoch: 16/100, batch 100/563 completed (loss: 0.0793:  18%|█▊        | 50/281 [05:03<23:19,  6.06s/it]
Training Epoch: 16/100, batch 100/563 completed (loss: 0.0683:  18%|█▊        | 51/281 [05:06<23:21,  6.09s/it]
Training Epoch: 16/100, batch 100/563 completed (loss: 0.0783:  18%|█▊        | 51/281 [05:06<23:21,  6.10s/it]
Training Epoch: 16/100, batch 100/563 completed (loss: 0.0793:  18%|█▊        | 51/281 [05:06<23:23,  6.10s/it]
Training Epoch: 16/100, batch 100/563 completed (loss: 0.0094:  18%|█▊        | 51/281 [05:06<23:22,  6.10s/it]
Training Epoch: 16/100, batch 101/563 completed (loss: 0.0048:  18%|█▊        | 51/281 [05:06<23:21,  6.09s/it]
Training Epoch: 16/100, batch 101/563 completed (loss: 0.0168:  18%|█▊        | 51/281 [05:06<23:22,  6.10s/it]
Training Epoch: 16/100, batch 101/563 completed (loss: 0.0337:  18%|█▊        | 51/281 [05:06<23:21,  6.10s/it]
Training Epoch: 16/100, batch 101/563 completed (loss: 0.106:  18%|█▊        | 51/281 [05:06<23:23,  6.10s/it] 
Training Epoch: 16/100, batch 102/563 completed (loss: 0.0887:  18%|█▊        | 51/281 [05:09<23:22,  6.10s/it]
Training Epoch: 16/100, batch 102/563 completed (loss: 0.0082:  18%|█▊        | 51/281 [05:09<23:21,  6.10s/it]
Training Epoch: 16/100, batch 102/563 completed (loss: 0.0153:  18%|█▊        | 51/281 [05:09<23:21,  6.09s/it]
Training Epoch: 16/100, batch 102/563 completed (loss: 0.5379:  18%|█▊        | 51/281 [05:09<23:23,  6.10s/it]
Training Epoch: 16/100, batch 102/563 completed (loss: 0.5379:  19%|█▊        | 52/281 [05:12<23:09,  6.07s/it]
Training Epoch: 16/100, batch 102/563 completed (loss: 0.0082:  19%|█▊        | 52/281 [05:12<23:09,  6.07s/it]
Training Epoch: 16/100, batch 102/563 completed (loss: 0.0887:  19%|█▊        | 52/281 [05:12<23:09,  6.07s/it]
Training Epoch: 16/100, batch 102/563 completed (loss: 0.0153:  19%|█▊        | 52/281 [05:12<23:09,  6.07s/it]
Training Epoch: 16/100, batch 103/563 completed (loss: 0.0062:  19%|█▊        | 52/281 [05:12<23:09,  6.07s/it]
Training Epoch: 16/100, batch 103/563 completed (loss: 0.1074:  19%|█▊        | 52/281 [05:12<23:09,  6.07s/it]
Training Epoch: 16/100, batch 103/563 completed (loss: 0.2033:  19%|█▊        | 52/281 [05:12<23:09,  6.07s/it]
Training Epoch: 16/100, batch 103/563 completed (loss: 0.0045:  19%|█▊        | 52/281 [05:12<23:09,  6.07s/it]
Training Epoch: 16/100, batch 104/563 completed (loss: 0.0235:  19%|█▊        | 52/281 [05:15<23:09,  6.07s/it]
Training Epoch: 16/100, batch 104/563 completed (loss: 0.0812:  19%|█▊        | 52/281 [05:15<23:09,  6.07s/it]
Training Epoch: 16/100, batch 104/563 completed (loss: 0.0149:  19%|█▊        | 52/281 [05:15<23:09,  6.07s/it]
Training Epoch: 16/100, batch 104/563 completed (loss: 0.0088:  19%|█▊        | 52/281 [05:15<23:09,  6.07s/it]
Training Epoch: 16/100, batch 104/563 completed (loss: 0.0088:  19%|█▉        | 53/281 [05:18<23:16,  6.13s/it]
Training Epoch: 16/100, batch 104/563 completed (loss: 0.0235:  19%|█▉        | 53/281 [05:18<23:17,  6.13s/it]
Training Epoch: 16/100, batch 104/563 completed (loss: 0.0812:  19%|█▉        | 53/281 [05:18<23:17,  6.13s/it]
Training Epoch: 16/100, batch 104/563 completed (loss: 0.0149:  19%|█▉        | 53/281 [05:18<23:18,  6.13s/it]
Training Epoch: 16/100, batch 105/563 completed (loss: 0.0193:  19%|█▉        | 53/281 [05:18<23:17,  6.13s/it]
Training Epoch: 16/100, batch 105/563 completed (loss: 0.0135:  19%|█▉        | 53/281 [05:18<23:17,  6.13s/it]
Training Epoch: 16/100, batch 105/563 completed (loss: 0.1399:  19%|█▉        | 53/281 [05:18<23:18,  6.13s/it]
Training Epoch: 16/100, batch 105/563 completed (loss: 0.031:  19%|█▉        | 53/281 [05:19<23:16,  6.13s/it] 
Training Epoch: 16/100, batch 106/563 completed (loss: 0.0209:  19%|█▉        | 53/281 [05:21<23:18,  6.13s/it]
Training Epoch: 16/100, batch 106/563 completed (loss: 0.2387:  19%|█▉        | 53/281 [05:21<23:17,  6.13s/it]
Training Epoch: 16/100, batch 106/563 completed (loss: 0.0059:  19%|█▉        | 53/281 [05:21<23:17,  6.13s/it]
Training Epoch: 16/100, batch 106/563 completed (loss: 0.0247:  19%|█▉        | 53/281 [05:21<23:16,  6.13s/it]
Training Epoch: 16/100, batch 106/563 completed (loss: 0.0247:  19%|█▉        | 54/281 [05:24<22:53,  6.05s/it]
Training Epoch: 16/100, batch 106/563 completed (loss: 0.2387:  19%|█▉        | 54/281 [05:24<22:53,  6.05s/it]
Training Epoch: 16/100, batch 106/563 completed (loss: 0.0059:  19%|█▉        | 54/281 [05:24<22:53,  6.05s/it]
Training Epoch: 16/100, batch 106/563 completed (loss: 0.0209:  19%|█▉        | 54/281 [05:24<22:53,  6.05s/it]
Training Epoch: 16/100, batch 107/563 completed (loss: 0.0147:  19%|█▉        | 54/281 [05:24<22:53,  6.05s/it]
Training Epoch: 16/100, batch 107/563 completed (loss: 0.0211:  19%|█▉        | 54/281 [05:24<22:53,  6.05s/it]
Training Epoch: 16/100, batch 107/563 completed (loss: 0.0052:  19%|█▉        | 54/281 [05:24<22:53,  6.05s/it]
Training Epoch: 16/100, batch 107/563 completed (loss: 0.1003:  19%|█▉        | 54/281 [05:24<22:53,  6.05s/it]
Training Epoch: 16/100, batch 108/563 completed (loss: 0.011:  19%|█▉        | 54/281 [05:27<22:53,  6.05s/it] 
Training Epoch: 16/100, batch 108/563 completed (loss: 0.0445:  19%|█▉        | 54/281 [05:27<22:53,  6.05s/it]
Training Epoch: 16/100, batch 108/563 completed (loss: 0.0249:  19%|█▉        | 54/281 [05:27<22:53,  6.05s/it]
Training Epoch: 16/100, batch 108/563 completed (loss: 0.0047:  19%|█▉        | 54/281 [05:27<22:53,  6.05s/it]
Training Epoch: 16/100, batch 108/563 completed (loss: 0.0445:  20%|█▉        | 55/281 [05:30<22:43,  6.03s/it]
Training Epoch: 16/100, batch 108/563 completed (loss: 0.0047:  20%|█▉        | 55/281 [05:30<22:43,  6.03s/it]
Training Epoch: 16/100, batch 108/563 completed (loss: 0.011:  20%|█▉        | 55/281 [05:30<22:43,  6.03s/it]
Training Epoch: 16/100, batch 108/563 completed (loss: 0.0249:  20%|█▉        | 55/281 [05:30<22:43,  6.03s/it]
Training Epoch: 16/100, batch 109/563 completed (loss: 0.0505:  20%|█▉        | 55/281 [05:30<22:43,  6.03s/it]
Training Epoch: 16/100, batch 109/563 completed (loss: 0.0202:  20%|█▉        | 55/281 [05:30<22:43,  6.03s/it]
Training Epoch: 16/100, batch 109/563 completed (loss: 0.0541:  20%|█▉        | 55/281 [05:30<22:43,  6.03s/it]
Training Epoch: 16/100, batch 109/563 completed (loss: 0.0392:  20%|█▉        | 55/281 [05:30<22:43,  6.03s/it]
Training Epoch: 16/100, batch 110/563 completed (loss: 0.0111:  20%|█▉        | 55/281 [05:33<22:43,  6.03s/it]
Training Epoch: 16/100, batch 110/563 completed (loss: 0.1724:  20%|█▉        | 55/281 [05:33<22:43,  6.03s/it]
Training Epoch: 16/100, batch 110/563 completed (loss: 0.0847:  20%|█▉        | 55/281 [05:33<22:43,  6.03s/it]
Training Epoch: 16/100, batch 110/563 completed (loss: 0.0055:  20%|█▉        | 55/281 [05:33<22:43,  6.03s/it]
Training Epoch: 16/100, batch 110/563 completed (loss: 0.1724:  20%|█▉        | 56/281 [05:36<22:46,  6.07s/it]
Training Epoch: 16/100, batch 110/563 completed (loss: 0.0111:  20%|█▉        | 56/281 [05:36<22:46,  6.07s/it]
Training Epoch: 16/100, batch 110/563 completed (loss: 0.0847:  20%|█▉        | 56/281 [05:36<22:46,  6.07s/it]
Training Epoch: 16/100, batch 110/563 completed (loss: 0.0055:  20%|█▉        | 56/281 [05:36<22:47,  6.08s/it]
Training Epoch: 16/100, batch 111/563 completed (loss: 0.1622:  20%|█▉        | 56/281 [05:36<22:46,  6.07s/it]
Training Epoch: 16/100, batch 111/563 completed (loss: 0.0031:  20%|█▉        | 56/281 [05:36<22:46,  6.07s/it]
Training Epoch: 16/100, batch 111/563 completed (loss: 0.0179:  20%|█▉        | 56/281 [05:36<22:46,  6.07s/it]
Training Epoch: 16/100, batch 111/563 completed (loss: 0.0238:  20%|█▉        | 56/281 [05:37<22:47,  6.08s/it]
Training Epoch: 16/100, batch 112/563 completed (loss: 0.0555:  20%|█▉        | 56/281 [05:39<22:46,  6.07s/it]
Training Epoch: 16/100, batch 112/563 completed (loss: 0.0033:  20%|█▉        | 56/281 [05:39<22:46,  6.07s/it]
Training Epoch: 16/100, batch 112/563 completed (loss: 0.0827:  20%|█▉        | 56/281 [05:39<22:46,  6.07s/it]
Training Epoch: 16/100, batch 112/563 completed (loss: 0.0391:  20%|█▉        | 56/281 [05:39<22:47,  6.08s/it]
Training Epoch: 16/100, batch 112/563 completed (loss: 0.0391:  20%|██        | 57/281 [05:42<22:25,  6.01s/it]
Training Epoch: 16/100, batch 112/563 completed (loss: 0.0033:  20%|██        | 57/281 [05:42<22:26,  6.01s/it]
Training Epoch: 16/100, batch 112/563 completed (loss: 0.0827:  20%|██        | 57/281 [05:42<22:26,  6.01s/it]
Training Epoch: 16/100, batch 112/563 completed (loss: 0.0555:  20%|██        | 57/281 [05:42<22:26,  6.01s/it]
Training Epoch: 16/100, batch 113/563 completed (loss: 0.1061:  20%|██        | 57/281 [05:42<22:26,  6.01s/it]
Training Epoch: 16/100, batch 113/563 completed (loss: 0.0126:  20%|██        | 57/281 [05:42<22:26,  6.01s/it]
Training Epoch: 16/100, batch 113/563 completed (loss: 0.0691:  20%|██        | 57/281 [05:42<22:26,  6.01s/it]
Training Epoch: 16/100, batch 113/563 completed (loss: 0.308:  20%|██        | 57/281 [05:42<22:25,  6.01s/it] 
Training Epoch: 16/100, batch 114/563 completed (loss: 0.3974:  20%|██        | 57/281 [05:45<22:26,  6.01s/it]
Training Epoch: 16/100, batch 114/563 completed (loss: 0.0942:  20%|██        | 57/281 [05:45<22:26,  6.01s/it]
Training Epoch: 16/100, batch 114/563 completed (loss: 0.0245:  20%|██        | 57/281 [05:45<22:26,  6.01s/it]
Training Epoch: 16/100, batch 114/563 completed (loss: 0.0079:  20%|██        | 57/281 [05:46<22:25,  6.01s/it]
Training Epoch: 16/100, batch 114/563 completed (loss: 0.0079:  21%|██        | 58/281 [05:49<22:41,  6.10s/it]
Training Epoch: 16/100, batch 114/563 completed (loss: 0.3974:  21%|██        | 58/281 [05:49<22:40,  6.10s/it]
Training Epoch: 16/100, batch 114/563 completed (loss: 0.0245:  21%|██        | 58/281 [05:49<22:40,  6.10s/it]
Training Epoch: 16/100, batch 114/563 completed (loss: 0.0942:  21%|██        | 58/281 [05:49<22:41,  6.11s/it]
Training Epoch: 16/100, batch 115/563 completed (loss: 0.3289:  21%|██        | 58/281 [05:49<22:40,  6.10s/it]
Training Epoch: 16/100, batch 115/563 completed (loss: 0.0261:  21%|██        | 58/281 [05:49<22:40,  6.10s/it]
Training Epoch: 16/100, batch 115/563 completed (loss: 0.0094:  21%|██        | 58/281 [05:49<22:41,  6.11s/it]
Training Epoch: 16/100, batch 115/563 completed (loss: 0.0797:  21%|██        | 58/281 [05:49<22:41,  6.10s/it]
Training Epoch: 16/100, batch 116/563 completed (loss: 0.0395:  21%|██        | 58/281 [05:52<22:41,  6.11s/it]
Training Epoch: 16/100, batch 116/563 completed (loss: 0.0051:  21%|██        | 58/281 [05:52<22:40,  6.10s/it]
Training Epoch: 16/100, batch 116/563 completed (loss: 0.0057:  21%|██        | 58/281 [05:52<22:40,  6.10s/it]
Training Epoch: 16/100, batch 116/563 completed (loss: 0.0972:  21%|██        | 58/281 [05:52<22:41,  6.10s/it]
Training Epoch: 16/100, batch 116/563 completed (loss: 0.0972:  21%|██        | 59/281 [05:55<22:29,  6.08s/it]
Training Epoch: 16/100, batch 116/563 completed (loss: 0.0057:  21%|██        | 59/281 [05:55<22:30,  6.08s/it]
Training Epoch: 16/100, batch 116/563 completed (loss: 0.0051:  21%|██        | 59/281 [05:55<22:30,  6.08s/it]
Training Epoch: 16/100, batch 116/563 completed (loss: 0.0395:  21%|██        | 59/281 [05:55<22:31,  6.09s/it]
Training Epoch: 16/100, batch 117/563 completed (loss: 0.0099:  21%|██        | 59/281 [05:55<22:30,  6.08s/it]
Training Epoch: 16/100, batch 117/563 completed (loss: 0.0375:  21%|██        | 59/281 [05:55<22:30,  6.08s/it]
Training Epoch: 16/100, batch 117/563 completed (loss: 0.0071:  21%|██        | 59/281 [05:55<22:31,  6.09s/it]
Training Epoch: 16/100, batch 117/563 completed (loss: 0.7456:  21%|██        | 59/281 [05:55<22:29,  6.08s/it]
Training Epoch: 16/100, batch 118/563 completed (loss: 0.1828:  21%|██        | 59/281 [05:58<22:31,  6.09s/it]
Training Epoch: 16/100, batch 118/563 completed (loss: 0.0126:  21%|██        | 59/281 [05:58<22:30,  6.08s/it]
Training Epoch: 16/100, batch 118/563 completed (loss: 0.0052:  21%|██        | 59/281 [05:58<22:30,  6.08s/it]
Training Epoch: 16/100, batch 118/563 completed (loss: 0.0614:  21%|██        | 59/281 [05:58<22:29,  6.08s/it]
Training Epoch: 16/100, batch 118/563 completed (loss: 0.0052:  21%|██▏       | 60/281 [06:00<22:07,  6.01s/it]
Training Epoch: 16/100, batch 118/563 completed (loss: 0.0126:  21%|██▏       | 60/281 [06:00<22:07,  6.01s/it]
Training Epoch: 16/100, batch 118/563 completed (loss: 0.0614:  21%|██▏       | 60/281 [06:01<22:08,  6.01s/it]
Training Epoch: 16/100, batch 118/563 completed (loss: 0.1828:  21%|██▏       | 60/281 [06:00<22:07,  6.01s/it]
Training Epoch: 16/100, batch 119/563 completed (loss: 0.0327:  21%|██▏       | 60/281 [06:01<22:07,  6.01s/it]
Training Epoch: 16/100, batch 119/563 completed (loss: 0.0174:  21%|██▏       | 60/281 [06:01<22:07,  6.01s/it]
Training Epoch: 16/100, batch 119/563 completed (loss: 0.0315:  21%|██▏       | 60/281 [06:01<22:07,  6.01s/it]
Training Epoch: 16/100, batch 119/563 completed (loss: 0.0133:  21%|██▏       | 60/281 [06:01<22:08,  6.01s/it]
Training Epoch: 16/100, batch 120/563 completed (loss: 0.1491:  21%|██▏       | 60/281 [06:04<22:07,  6.01s/it]
Training Epoch: 16/100, batch 120/563 completed (loss: 0.0333:  21%|██▏       | 60/281 [06:04<22:07,  6.01s/it]
Training Epoch: 16/100, batch 120/563 completed (loss: 0.0092:  21%|██▏       | 60/281 [06:04<22:07,  6.01s/it]
Training Epoch: 16/100, batch 120/563 completed (loss: 0.01:  21%|██▏       | 60/281 [06:04<22:08,  6.01s/it]  
Training Epoch: 16/100, batch 120/563 completed (loss: 0.01:  22%|██▏       | 61/281 [06:07<22:34,  6.16s/it]
Training Epoch: 16/100, batch 120/563 completed (loss: 0.0092:  22%|██▏       | 61/281 [06:07<22:34,  6.16s/it]
Training Epoch: 16/100, batch 120/563 completed (loss: 0.0333:  22%|██▏       | 61/281 [06:07<22:35,  6.16s/it]
Training Epoch: 16/100, batch 120/563 completed (loss: 0.1491:  22%|██▏       | 61/281 [06:07<22:36,  6.16s/it]
Training Epoch: 16/100, batch 121/563 completed (loss: 0.0656:  22%|██▏       | 61/281 [06:07<22:36,  6.16s/it]
Training Epoch: 16/100, batch 121/563 completed (loss: 0.0169:  22%|██▏       | 61/281 [06:07<22:34,  6.16s/it]
Training Epoch: 16/100, batch 121/563 completed (loss: 0.0205:  22%|██▏       | 61/281 [06:07<22:35,  6.16s/it]
Training Epoch: 16/100, batch 121/563 completed (loss: 0.049:  22%|██▏       | 61/281 [06:07<22:34,  6.16s/it]
Training Epoch: 16/100, batch 122/563 completed (loss: 0.0082:  22%|██▏       | 61/281 [06:10<22:35,  6.16s/it]
Training Epoch: 16/100, batch 122/563 completed (loss: 0.015:  22%|██▏       | 61/281 [06:10<22:34,  6.16s/it] 
Training Epoch: 16/100, batch 122/563 completed (loss: 0.0504:  22%|██▏       | 61/281 [06:10<22:36,  6.16s/it]
Training Epoch: 16/100, batch 122/563 completed (loss: 0.0121:  22%|██▏       | 61/281 [06:10<22:34,  6.16s/it]
Training Epoch: 16/100, batch 122/563 completed (loss: 0.0121:  22%|██▏       | 62/281 [06:13<22:11,  6.08s/it]
Training Epoch: 16/100, batch 122/563 completed (loss: 0.0504:  22%|██▏       | 62/281 [06:13<22:11,  6.08s/it]
Training Epoch: 16/100, batch 122/563 completed (loss: 0.015:  22%|██▏       | 62/281 [06:13<22:12,  6.08s/it]
Training Epoch: 16/100, batch 122/563 completed (loss: 0.0082:  22%|██▏       | 62/281 [06:13<22:11,  6.08s/it]
Training Epoch: 16/100, batch 123/563 completed (loss: 0.0026:  22%|██▏       | 62/281 [06:13<22:12,  6.08s/it]
Training Epoch: 16/100, batch 123/563 completed (loss: 0.0121:  22%|██▏       | 62/281 [06:13<22:11,  6.08s/it]
Training Epoch: 16/100, batch 123/563 completed (loss: 0.0858:  22%|██▏       | 62/281 [06:13<22:11,  6.08s/it]
Training Epoch: 16/100, batch 123/563 completed (loss: 0.0108:  22%|██▏       | 62/281 [06:13<22:11,  6.08s/it]
Training Epoch: 16/100, batch 124/563 completed (loss: 0.005:  22%|██▏       | 62/281 [06:16<22:11,  6.08s/it] 
Training Epoch: 16/100, batch 124/563 completed (loss: 0.0241:  22%|██▏       | 62/281 [06:16<22:12,  6.08s/it]
Training Epoch: 16/100, batch 124/563 completed (loss: 0.0137:  22%|██▏       | 62/281 [06:16<22:11,  6.08s/it]
Training Epoch: 16/100, batch 124/563 completed (loss: 0.0301:  22%|██▏       | 62/281 [06:16<22:11,  6.08s/it]
Training Epoch: 16/100, batch 124/563 completed (loss: 0.0241:  22%|██▏       | 63/281 [06:19<22:02,  6.07s/it]
Training Epoch: 16/100, batch 124/563 completed (loss: 0.0301:  22%|██▏       | 63/281 [06:19<22:03,  6.07s/it]
Training Epoch: 16/100, batch 124/563 completed (loss: 0.0137:  22%|██▏       | 63/281 [06:19<22:02,  6.07s/it]
Training Epoch: 16/100, batch 124/563 completed (loss: 0.005:  22%|██▏       | 63/281 [06:19<22:03,  6.07s/it]
Training Epoch: 16/100, batch 125/563 completed (loss: 0.0324:  22%|██▏       | 63/281 [06:19<22:02,  6.07s/it]
Training Epoch: 16/100, batch 125/563 completed (loss: 0.2289:  22%|██▏       | 63/281 [06:19<22:02,  6.07s/it]
Training Epoch: 16/100, batch 125/563 completed (loss: 0.0027:  22%|██▏       | 63/281 [06:19<22:03,  6.07s/it]
Training Epoch: 16/100, batch 125/563 completed (loss: 0.037:  22%|██▏       | 63/281 [06:19<22:03,  6.07s/it] 
Training Epoch: 16/100, batch 126/563 completed (loss: 0.0179:  22%|██▏       | 63/281 [06:22<22:03,  6.07s/it]
Training Epoch: 16/100, batch 126/563 completed (loss: 0.0049:  22%|██▏       | 63/281 [06:22<22:02,  6.07s/it]
Training Epoch: 16/100, batch 126/563 completed (loss: 0.0225:  22%|██▏       | 63/281 [06:22<22:02,  6.07s/it]
Training Epoch: 16/100, batch 126/563 completed (loss: 0.1382:  22%|██▏       | 63/281 [06:22<22:03,  6.07s/it]
Training Epoch: 16/100, batch 126/563 completed (loss: 0.1382:  23%|██▎       | 64/281 [06:25<22:11,  6.13s/it]
Training Epoch: 16/100, batch 126/563 completed (loss: 0.0225:  23%|██▎       | 64/281 [06:25<22:11,  6.14s/it]
Training Epoch: 16/100, batch 126/563 completed (loss: 0.0179:  23%|██▎       | 64/281 [06:25<22:12,  6.14s/it]
Training Epoch: 16/100, batch 126/563 completed (loss: 0.0049:  23%|██▎       | 64/281 [06:25<22:13,  6.15s/it]
Training Epoch: 16/100, batch 127/563 completed (loss: 0.3516:  23%|██▎       | 64/281 [06:25<22:13,  6.15s/it]
Training Epoch: 16/100, batch 127/563 completed (loss: 0.1502:  23%|██▎       | 64/281 [06:25<22:12,  6.14s/it]
Training Epoch: 16/100, batch 127/563 completed (loss: 0.0291:  23%|██▎       | 64/281 [06:25<22:11,  6.14s/it]
Training Epoch: 16/100, batch 127/563 completed (loss: 0.0264:  23%|██▎       | 64/281 [06:25<22:11,  6.13s/it]
Training Epoch: 16/100, batch 128/563 completed (loss: 0.0055:  23%|██▎       | 64/281 [06:28<22:12,  6.14s/it]
Training Epoch: 16/100, batch 128/563 completed (loss: 0.0255:  23%|██▎       | 64/281 [06:28<22:13,  6.15s/it]
Training Epoch: 16/100, batch 128/563 completed (loss: 0.0132:  23%|██▎       | 64/281 [06:28<22:11,  6.14s/it]
Training Epoch: 16/100, batch 128/563 completed (loss: 0.0083:  23%|██▎       | 64/281 [06:28<22:11,  6.13s/it]
Training Epoch: 16/100, batch 128/563 completed (loss: 0.0255:  23%|██▎       | 65/281 [06:31<21:48,  6.06s/it]
Training Epoch: 16/100, batch 128/563 completed (loss: 0.0083:  23%|██▎       | 65/281 [06:31<21:49,  6.06s/it]
Training Epoch: 16/100, batch 128/563 completed (loss: 0.0055:  23%|██▎       | 65/281 [06:31<21:48,  6.06s/it]
Training Epoch: 16/100, batch 128/563 completed (loss: 0.0132:  23%|██▎       | 65/281 [06:31<21:48,  6.06s/it]
Training Epoch: 16/100, batch 129/563 completed (loss: 0.0065:  23%|██▎       | 65/281 [06:31<21:48,  6.06s/it]
Training Epoch: 16/100, batch 129/563 completed (loss: 0.2858:  23%|██▎       | 65/281 [06:31<21:48,  6.06s/it]
Training Epoch: 16/100, batch 129/563 completed (loss: 0.1138:  23%|██▎       | 65/281 [06:31<21:48,  6.06s/it]
Training Epoch: 16/100, batch 129/563 completed (loss: 0.0145:  23%|██▎       | 65/281 [06:31<21:49,  6.06s/it]
Training Epoch: 16/100, batch 130/563 completed (loss: 0.2932:  23%|██▎       | 65/281 [06:34<21:48,  6.06s/it]
Training Epoch: 16/100, batch 130/563 completed (loss: 0.0037:  23%|██▎       | 65/281 [06:34<21:48,  6.06s/it]
Training Epoch: 16/100, batch 130/563 completed (loss: 0.0019:  23%|██▎       | 65/281 [06:34<21:48,  6.06s/it]
Training Epoch: 16/100, batch 130/563 completed (loss: 0.0349:  23%|██▎       | 65/281 [06:34<21:49,  6.06s/it]
Training Epoch: 16/100, batch 130/563 completed (loss: 0.0349:  23%|██▎       | 66/281 [06:38<22:06,  6.17s/it]
Training Epoch: 16/100, batch 130/563 completed (loss: 0.0019:  23%|██▎       | 66/281 [06:38<22:05,  6.17s/it]
Training Epoch: 16/100, batch 130/563 completed (loss: 0.2932:  23%|██▎       | 66/281 [06:38<22:05,  6.17s/it]
Training Epoch: 16/100, batch 130/563 completed (loss: 0.0037:  23%|██▎       | 66/281 [06:38<22:06,  6.17s/it]
Training Epoch: 16/100, batch 131/563 completed (loss: 0.0056:  23%|██▎       | 66/281 [06:38<22:05,  6.17s/it]
Training Epoch: 16/100, batch 131/563 completed (loss: 0.0232:  23%|██▎       | 66/281 [06:38<22:06,  6.17s/it]
Training Epoch: 16/100, batch 131/563 completed (loss: 0.0127:  23%|██▎       | 66/281 [06:38<22:05,  6.17s/it]
Training Epoch: 16/100, batch 131/563 completed (loss: 0.0072:  23%|██▎       | 66/281 [06:38<22:06,  6.17s/it]
Training Epoch: 16/100, batch 132/563 completed (loss: 0.0038:  23%|██▎       | 66/281 [06:41<22:06,  6.17s/it]
Training Epoch: 16/100, batch 132/563 completed (loss: 0.0141:  23%|██▎       | 66/281 [06:41<22:05,  6.17s/it]
Training Epoch: 16/100, batch 132/563 completed (loss: 0.0415:  23%|██▎       | 66/281 [06:40<22:05,  6.17s/it]
Training Epoch: 16/100, batch 132/563 completed (loss: 0.6131:  23%|██▎       | 66/281 [06:41<22:06,  6.17s/it]
Training Epoch: 16/100, batch 132/563 completed (loss: 0.6131:  24%|██▍       | 67/281 [06:43<21:39,  6.07s/it]
Training Epoch: 16/100, batch 132/563 completed (loss: 0.0415:  24%|██▍       | 67/281 [06:43<21:39,  6.07s/it]
Training Epoch: 16/100, batch 132/563 completed (loss: 0.0141:  24%|██▍       | 67/281 [06:43<21:41,  6.08s/it]
Training Epoch: 16/100, batch 132/563 completed (loss: 0.0038:  24%|██▍       | 67/281 [06:43<21:41,  6.08s/it]
Training Epoch: 16/100, batch 133/563 completed (loss: 0.0104:  24%|██▍       | 67/281 [06:43<21:41,  6.08s/it]
Training Epoch: 16/100, batch 133/563 completed (loss: 0.3711:  24%|██▍       | 67/281 [06:43<21:39,  6.07s/it]
Training Epoch: 16/100, batch 133/563 completed (loss: 0.4203:  24%|██▍       | 67/281 [06:43<21:41,  6.08s/it]
Training Epoch: 16/100, batch 133/563 completed (loss: 0.1703:  24%|██▍       | 67/281 [06:44<21:39,  6.07s/it]
Training Epoch: 16/100, batch 134/563 completed (loss: 0.0272:  24%|██▍       | 67/281 [06:46<21:39,  6.07s/it]
Training Epoch: 16/100, batch 134/563 completed (loss: 0.0138:  24%|██▍       | 67/281 [06:46<21:41,  6.08s/it]
Training Epoch: 16/100, batch 134/563 completed (loss: 0.0072:  24%|██▍       | 67/281 [06:46<21:41,  6.08s/it]
Training Epoch: 16/100, batch 134/563 completed (loss: 0.0016:  24%|██▍       | 67/281 [06:46<21:39,  6.07s/it]
Training Epoch: 16/100, batch 134/563 completed (loss: 0.0272:  24%|██▍       | 68/281 [06:49<21:31,  6.06s/it]
Training Epoch: 16/100, batch 134/563 completed (loss: 0.0016:  24%|██▍       | 68/281 [06:49<21:32,  6.07s/it]
Training Epoch: 16/100, batch 134/563 completed (loss: 0.0138:  24%|██▍       | 68/281 [06:49<21:30,  6.06s/it]
Training Epoch: 16/100, batch 134/563 completed (loss: 0.0072:  24%|██▍       | 68/281 [06:49<21:30,  6.06s/it]
Training Epoch: 16/100, batch 135/563 completed (loss: 0.01:  24%|██▍       | 68/281 [06:49<21:31,  6.06s/it]  
Training Epoch: 16/100, batch 135/563 completed (loss: 0.0062:  24%|██▍       | 68/281 [06:49<21:30,  6.06s/it]
Training Epoch: 16/100, batch 135/563 completed (loss: 0.0194:  24%|██▍       | 68/281 [06:49<21:30,  6.06s/it]
Training Epoch: 16/100, batch 135/563 completed (loss: 0.0116:  24%|██▍       | 68/281 [06:50<21:32,  6.07s/it]
Training Epoch: 16/100, batch 136/563 completed (loss: 0.0162:  24%|██▍       | 68/281 [06:53<21:30,  6.06s/it]
Training Epoch: 16/100, batch 136/563 completed (loss: 0.0079:  24%|██▍       | 68/281 [06:53<21:30,  6.06s/it]
Training Epoch: 16/100, batch 136/563 completed (loss: 0.0031:  24%|██▍       | 68/281 [06:53<21:31,  6.06s/it]
Training Epoch: 16/100, batch 136/563 completed (loss: 0.049:  24%|██▍       | 68/281 [06:53<21:32,  6.07s/it] 
Training Epoch: 16/100, batch 136/563 completed (loss: 0.0031:  25%|██▍       | 69/281 [06:56<21:37,  6.12s/it]
Training Epoch: 16/100, batch 136/563 completed (loss: 0.0162:  25%|██▍       | 69/281 [06:56<21:37,  6.12s/it]
Training Epoch: 16/100, batch 136/563 completed (loss: 0.049:  25%|██▍       | 69/281 [06:56<21:38,  6.13s/it]
Training Epoch: 16/100, batch 136/563 completed (loss: 0.0079:  25%|██▍       | 69/281 [06:56<21:38,  6.12s/it]
Training Epoch: 16/100, batch 137/563 completed (loss: 0.0046:  25%|██▍       | 69/281 [06:56<21:37,  6.12s/it]
Training Epoch: 16/100, batch 137/563 completed (loss: 0.0352:  25%|██▍       | 69/281 [06:56<21:38,  6.12s/it]
Training Epoch: 16/100, batch 137/563 completed (loss: 0.0099:  25%|██▍       | 69/281 [06:56<21:37,  6.12s/it]
Training Epoch: 16/100, batch 137/563 completed (loss: 0.348:  25%|██▍       | 69/281 [06:56<21:38,  6.13s/it]
Training Epoch: 16/100, batch 138/563 completed (loss: 0.0982:  25%|██▍       | 69/281 [06:59<21:37,  6.12s/it]
Training Epoch: 16/100, batch 138/563 completed (loss: 0.009:  25%|██▍       | 69/281 [06:59<21:38,  6.12s/it] 
Training Epoch: 16/100, batch 138/563 completed (loss: 0.1019:  25%|██▍       | 69/281 [06:59<21:37,  6.12s/it]
Training Epoch: 16/100, batch 138/563 completed (loss: 0.0169:  25%|██▍       | 69/281 [06:59<21:38,  6.13s/it]
Training Epoch: 16/100, batch 138/563 completed (loss: 0.0982:  25%|██▍       | 70/281 [07:02<21:15,  6.05s/it]
Training Epoch: 16/100, batch 138/563 completed (loss: 0.009:  25%|██▍       | 70/281 [07:02<21:15,  6.04s/it]
Training Epoch: 16/100, batch 138/563 completed (loss: 0.0169:  25%|██▍       | 70/281 [07:02<21:16,  6.05s/it]
Training Epoch: 16/100, batch 138/563 completed (loss: 0.1019:  25%|██▍       | 70/281 [07:02<21:16,  6.05s/it]
Training Epoch: 16/100, batch 139/563 completed (loss: 0.2175:  25%|██▍       | 70/281 [07:02<21:15,  6.04s/it]
Training Epoch: 16/100, batch 139/563 completed (loss: 0.0315:  25%|██▍       | 70/281 [07:02<21:15,  6.05s/it]
Training Epoch: 16/100, batch 139/563 completed (loss: 0.0051:  25%|██▍       | 70/281 [07:02<21:16,  6.05s/it]
Training Epoch: 16/100, batch 139/563 completed (loss: 0.008:  25%|██▍       | 70/281 [07:02<21:16,  6.05s/it] 
Training Epoch: 16/100, batch 140/563 completed (loss: 0.1069:  25%|██▍       | 70/281 [07:05<21:16,  6.05s/it]
Training Epoch: 16/100, batch 140/563 completed (loss: 0.1371:  25%|██▍       | 70/281 [07:05<21:15,  6.05s/it]
Training Epoch: 16/100, batch 140/563 completed (loss: 0.0242:  25%|██▍       | 70/281 [07:05<21:15,  6.04s/it]
Training Epoch: 16/100, batch 140/563 completed (loss: 0.0372:  25%|██▍       | 70/281 [07:05<21:16,  6.05s/it]
Training Epoch: 16/100, batch 140/563 completed (loss: 0.0372:  25%|██▌       | 71/281 [07:08<21:08,  6.04s/it]
Training Epoch: 16/100, batch 140/563 completed (loss: 0.0242:  25%|██▌       | 71/281 [07:08<21:08,  6.04s/it]
Training Epoch: 16/100, batch 140/563 completed (loss: 0.1069:  25%|██▌       | 71/281 [07:08<21:08,  6.04s/it]
Training Epoch: 16/100, batch 140/563 completed (loss: 0.1371:  25%|██▌       | 71/281 [07:08<21:10,  6.05s/it]
Training Epoch: 16/100, batch 141/563 completed (loss: 0.004:  25%|██▌       | 71/281 [07:08<21:10,  6.05s/it] 
Training Epoch: 16/100, batch 141/563 completed (loss: 0.0518:  25%|██▌       | 71/281 [07:08<21:08,  6.04s/it]
Training Epoch: 16/100, batch 141/563 completed (loss: 0.0445:  25%|██▌       | 71/281 [07:08<21:08,  6.04s/it]
Training Epoch: 16/100, batch 141/563 completed (loss: 0.0189:  25%|██▌       | 71/281 [07:08<21:08,  6.04s/it]
Training Epoch: 16/100, batch 142/563 completed (loss: 0.0262:  25%|██▌       | 71/281 [07:11<21:08,  6.04s/it]
Training Epoch: 16/100, batch 142/563 completed (loss: 0.0083:  25%|██▌       | 71/281 [07:11<21:10,  6.05s/it]
Training Epoch: 16/100, batch 142/563 completed (loss: 0.0277:  25%|██▌       | 71/281 [07:11<21:08,  6.04s/it]
Training Epoch: 16/100, batch 142/563 completed (loss: 0.0086:  25%|██▌       | 71/281 [07:11<21:08,  6.04s/it]
Training Epoch: 16/100, batch 142/563 completed (loss: 0.0086:  26%|██▌       | 72/281 [07:14<21:18,  6.12s/it]
Training Epoch: 16/100, batch 142/563 completed (loss: 0.0083:  26%|██▌       | 72/281 [07:14<21:18,  6.12s/it]
Training Epoch: 16/100, batch 142/563 completed (loss: 0.0262:  26%|██▌       | 72/281 [07:14<21:18,  6.12s/it]
Training Epoch: 16/100, batch 142/563 completed (loss: 0.0277:  26%|██▌       | 72/281 [07:14<21:19,  6.12s/it]
Training Epoch: 16/100, batch 143/563 completed (loss: 0.0645:  26%|██▌       | 72/281 [07:14<21:18,  6.12s/it]
Training Epoch: 16/100, batch 143/563 completed (loss: 0.0706:  26%|██▌       | 72/281 [07:14<21:18,  6.12s/it]
Training Epoch: 16/100, batch 143/563 completed (loss: 0.2271:  26%|██▌       | 72/281 [07:14<21:19,  6.12s/it]
Training Epoch: 16/100, batch 143/563 completed (loss: 0.0151:  26%|██▌       | 72/281 [07:14<21:18,  6.12s/it]
Training Epoch: 16/100, batch 144/563 completed (loss: 0.2058:  26%|██▌       | 72/281 [07:17<21:18,  6.12s/it]
Training Epoch: 16/100, batch 144/563 completed (loss: 0.0427:  26%|██▌       | 72/281 [07:17<21:18,  6.12s/it]
Training Epoch: 16/100, batch 144/563 completed (loss: 0.0886:  26%|██▌       | 72/281 [07:17<21:19,  6.12s/it]
Training Epoch: 16/100, batch 144/563 completed (loss: 0.0023:  26%|██▌       | 72/281 [07:17<21:18,  6.12s/it]
Training Epoch: 16/100, batch 144/563 completed (loss: 0.0023:  26%|██▌       | 73/281 [07:20<20:56,  6.04s/it]
Training Epoch: 16/100, batch 144/563 completed (loss: 0.0427:  26%|██▌       | 73/281 [07:20<20:57,  6.05s/it]
Training Epoch: 16/100, batch 144/563 completed (loss: 0.2058:  26%|██▌       | 73/281 [07:20<20:57,  6.05s/it]
Training Epoch: 16/100, batch 144/563 completed (loss: 0.0886:  26%|██▌       | 73/281 [07:20<20:57,  6.05s/it]
Training Epoch: 16/100, batch 145/563 completed (loss: 0.3898:  26%|██▌       | 73/281 [07:20<20:57,  6.05s/it]
Training Epoch: 16/100, batch 145/563 completed (loss: 0.0596:  26%|██▌       | 73/281 [07:20<20:57,  6.05s/it]
Training Epoch: 16/100, batch 145/563 completed (loss: 0.0161:  26%|██▌       | 73/281 [07:20<20:57,  6.05s/it]
Training Epoch: 16/100, batch 145/563 completed (loss: 0.0037:  26%|██▌       | 73/281 [07:20<20:56,  6.04s/it]
Training Epoch: 16/100, batch 146/563 completed (loss: 0.2994:  26%|██▌       | 73/281 [07:23<20:57,  6.05s/it]
Training Epoch: 16/100, batch 146/563 completed (loss: 0.0142:  26%|██▌       | 73/281 [07:23<20:57,  6.05s/it]
Training Epoch: 16/100, batch 146/563 completed (loss: 0.0768:  26%|██▌       | 73/281 [07:23<20:57,  6.05s/it]
Training Epoch: 16/100, batch 146/563 completed (loss: 0.1083:  26%|██▌       | 73/281 [07:23<20:56,  6.04s/it]
Training Epoch: 16/100, batch 146/563 completed (loss: 0.1083:  26%|██▋       | 74/281 [07:26<21:16,  6.17s/it]
Training Epoch: 16/100, batch 146/563 completed (loss: 0.0142:  26%|██▋       | 74/281 [07:26<21:15,  6.16s/it]
Training Epoch: 16/100, batch 146/563 completed (loss: 0.0768:  26%|██▋       | 74/281 [07:26<21:15,  6.16s/it]
Training Epoch: 16/100, batch 146/563 completed (loss: 0.2994:  26%|██▋       | 74/281 [07:26<21:16,  6.17s/it]
Training Epoch: 16/100, batch 147/563 completed (loss: 0.0144:  26%|██▋       | 74/281 [07:26<21:16,  6.17s/it]
Training Epoch: 16/100, batch 147/563 completed (loss: 0.0185:  26%|██▋       | 74/281 [07:26<21:15,  6.16s/it]
Training Epoch: 16/100, batch 147/563 completed (loss: 0.0307:  26%|██▋       | 74/281 [07:26<21:15,  6.16s/it]
Training Epoch: 16/100, batch 147/563 completed (loss: 0.0303:  26%|██▋       | 74/281 [07:26<21:16,  6.17s/it]
Training Epoch: 16/100, batch 148/563 completed (loss: 0.0228:  26%|██▋       | 74/281 [07:29<21:15,  6.16s/it]
Training Epoch: 16/100, batch 148/563 completed (loss: 0.0648:  26%|██▋       | 74/281 [07:29<21:15,  6.16s/it]
Training Epoch: 16/100, batch 148/563 completed (loss: 0.1575:  26%|██▋       | 74/281 [07:29<21:16,  6.17s/it]
Training Epoch: 16/100, batch 148/563 completed (loss: 0.0351:  26%|██▋       | 74/281 [07:29<21:16,  6.17s/it]
Training Epoch: 16/100, batch 148/563 completed (loss: 0.0351:  27%|██▋       | 75/281 [07:32<20:49,  6.06s/it]
Training Epoch: 16/100, batch 148/563 completed (loss: 0.1575:  27%|██▋       | 75/281 [07:32<20:48,  6.06s/it]
Training Epoch: 16/100, batch 148/563 completed (loss: 0.0228:  27%|██▋       | 75/281 [07:32<20:48,  6.06s/it]
Training Epoch: 16/100, batch 148/563 completed (loss: 0.0648:  27%|██▋       | 75/281 [07:32<20:48,  6.06s/it]
Training Epoch: 16/100, batch 149/563 completed (loss: 0.0101:  27%|██▋       | 75/281 [07:32<20:48,  6.06s/it]
Training Epoch: 16/100, batch 149/563 completed (loss: 0.004:  27%|██▋       | 75/281 [07:32<20:48,  6.06s/it] 
Training Epoch: 16/100, batch 149/563 completed (loss: 0.054:  27%|██▋       | 75/281 [07:32<20:48,  6.06s/it] 
Training Epoch: 16/100, batch 149/563 completed (loss: 0.0125:  27%|██▋       | 75/281 [07:32<20:49,  6.06s/it]
Training Epoch: 16/100, batch 150/563 completed (loss: 0.1408:  27%|██▋       | 75/281 [07:35<20:48,  6.06s/it]
Training Epoch: 16/100, batch 150/563 completed (loss: 0.0595:  27%|██▋       | 75/281 [07:35<20:48,  6.06s/it]
Training Epoch: 16/100, batch 150/563 completed (loss: 0.1482:  27%|██▋       | 75/281 [07:35<20:48,  6.06s/it]
Training Epoch: 16/100, batch 150/563 completed (loss: 0.0949:  27%|██▋       | 75/281 [07:35<20:49,  6.06s/it]
Training Epoch: 16/100, batch 150/563 completed (loss: 0.0595:  27%|██▋       | 76/281 [07:38<20:38,  6.04s/it]
Training Epoch: 16/100, batch 150/563 completed (loss: 0.0949:  27%|██▋       | 76/281 [07:38<20:39,  6.04s/it]
Training Epoch: 16/100, batch 150/563 completed (loss: 0.1482:  27%|██▋       | 76/281 [07:38<20:38,  6.04s/it]
Training Epoch: 16/100, batch 150/563 completed (loss: 0.1408:  27%|██▋       | 76/281 [07:38<20:38,  6.04s/it]
Training Epoch: 16/100, batch 151/563 completed (loss: 0.011:  27%|██▋       | 76/281 [07:38<20:38,  6.04s/it] 
Training Epoch: 16/100, batch 151/563 completed (loss: 0.0051:  27%|██▋       | 76/281 [07:38<20:38,  6.04s/it]
Training Epoch: 16/100, batch 151/563 completed (loss: 0.0051:  27%|██▋       | 76/281 [07:38<20:38,  6.04s/it]
Training Epoch: 16/100, batch 151/563 completed (loss: 0.0066:  27%|██▋       | 76/281 [07:38<20:39,  6.04s/it]
Training Epoch: 16/100, batch 152/563 completed (loss: 0.0346:  27%|██▋       | 76/281 [07:41<20:38,  6.04s/it]
Training Epoch: 16/100, batch 152/563 completed (loss: 0.0044:  27%|██▋       | 76/281 [07:41<20:38,  6.04s/it]
Training Epoch: 16/100, batch 152/563 completed (loss: 0.1492:  27%|██▋       | 76/281 [07:41<20:38,  6.04s/it]
Training Epoch: 16/100, batch 152/563 completed (loss: 0.0096:  27%|██▋       | 76/281 [07:41<20:39,  6.04s/it]
Training Epoch: 16/100, batch 152/563 completed (loss: 0.0096:  27%|██▋       | 77/281 [07:44<20:47,  6.12s/it]
Training Epoch: 16/100, batch 152/563 completed (loss: 0.0044:  27%|██▋       | 77/281 [07:44<20:47,  6.12s/it]
Training Epoch: 16/100, batch 152/563 completed (loss: 0.1492:  27%|██▋       | 77/281 [07:44<20:47,  6.12s/it]
Training Epoch: 16/100, batch 152/563 completed (loss: 0.0346:  27%|██▋       | 77/281 [07:44<20:47,  6.12s/it]
Training Epoch: 16/100, batch 153/563 completed (loss: 0.1222:  27%|██▋       | 77/281 [07:44<20:47,  6.12s/it]
Training Epoch: 16/100, batch 153/563 completed (loss: 0.0446:  27%|██▋       | 77/281 [07:44<20:47,  6.12s/it]
Training Epoch: 16/100, batch 153/563 completed (loss: 0.0386:  27%|██▋       | 77/281 [07:44<20:47,  6.12s/it]
Training Epoch: 16/100, batch 153/563 completed (loss: 0.0058:  27%|██▋       | 77/281 [07:44<20:47,  6.12s/it]
Training Epoch: 16/100, batch 154/563 completed (loss: 0.1625:  27%|██▋       | 77/281 [07:47<20:47,  6.12s/it]
Training Epoch: 16/100, batch 154/563 completed (loss: 0.0048:  27%|██▋       | 77/281 [07:47<20:47,  6.12s/it]
Training Epoch: 16/100, batch 154/563 completed (loss: 0.0383:  27%|██▋       | 77/281 [07:47<20:47,  6.12s/it]
Training Epoch: 16/100, batch 154/563 completed (loss: 0.0261:  27%|██▋       | 77/281 [07:47<20:47,  6.12s/it]
Training Epoch: 16/100, batch 154/563 completed (loss: 0.0261:  28%|██▊       | 78/281 [07:50<20:25,  6.04s/it]
Training Epoch: 16/100, batch 154/563 completed (loss: 0.0383:  28%|██▊       | 78/281 [07:50<20:25,  6.04s/it]
Training Epoch: 16/100, batch 154/563 completed (loss: 0.0048:  28%|██▊       | 78/281 [07:50<20:25,  6.04s/it]
Training Epoch: 16/100, batch 154/563 completed (loss: 0.1625:  28%|██▊       | 78/281 [07:50<20:25,  6.04s/it]
Training Epoch: 16/100, batch 155/563 completed (loss: 0.0085:  28%|██▊       | 78/281 [07:50<20:25,  6.04s/it]
Training Epoch: 16/100, batch 155/563 completed (loss: 0.0089:  28%|██▊       | 78/281 [07:50<20:25,  6.04s/it]
Training Epoch: 16/100, batch 155/563 completed (loss: 0.0409:  28%|██▊       | 78/281 [07:50<20:25,  6.04s/it]
Training Epoch: 16/100, batch 155/563 completed (loss: 0.1553:  28%|██▊       | 78/281 [07:50<20:25,  6.04s/it]
Training Epoch: 16/100, batch 156/563 completed (loss: 0.0367:  28%|██▊       | 78/281 [07:53<20:25,  6.04s/it]
Training Epoch: 16/100, batch 156/563 completed (loss: 0.0036:  28%|██▊       | 78/281 [07:53<20:25,  6.04s/it]
Training Epoch: 16/100, batch 156/563 completed (loss: 0.0372:  28%|██▊       | 78/281 [07:53<20:25,  6.04s/it]
Training Epoch: 16/100, batch 156/563 completed (loss: 0.0379:  28%|██▊       | 78/281 [07:53<20:25,  6.04s/it]
Training Epoch: 16/100, batch 156/563 completed (loss: 0.0379:  28%|██▊       | 79/281 [07:56<20:28,  6.08s/it]
Training Epoch: 16/100, batch 156/563 completed (loss: 0.0367:  28%|██▊       | 79/281 [07:56<20:27,  6.08s/it]
Training Epoch: 16/100, batch 156/563 completed (loss: 0.0036:  28%|██▊       | 79/281 [07:56<20:28,  6.08s/it]
Training Epoch: 16/100, batch 156/563 completed (loss: 0.0372:  28%|██▊       | 79/281 [07:56<20:28,  6.08s/it]
Training Epoch: 16/100, batch 157/563 completed (loss: 0.0118:  28%|██▊       | 79/281 [07:56<20:28,  6.08s/it]
Training Epoch: 16/100, batch 157/563 completed (loss: 0.0431:  28%|██▊       | 79/281 [07:56<20:28,  6.08s/it]
Training Epoch: 16/100, batch 157/563 completed (loss: 0.0335:  28%|██▊       | 79/281 [07:56<20:27,  6.08s/it]
Training Epoch: 16/100, batch 157/563 completed (loss: 0.0455:  28%|██▊       | 79/281 [07:56<20:28,  6.08s/it]
Training Epoch: 16/100, batch 158/563 completed (loss: 0.0016:  28%|██▊       | 79/281 [07:59<20:28,  6.08s/it]
Training Epoch: 16/100, batch 158/563 completed (loss: 0.0022:  28%|██▊       | 79/281 [07:59<20:27,  6.08s/it]
Training Epoch: 16/100, batch 158/563 completed (loss: 0.0082:  28%|██▊       | 79/281 [07:59<20:28,  6.08s/it]
Training Epoch: 16/100, batch 158/563 completed (loss: 0.2553:  28%|██▊       | 79/281 [08:00<20:28,  6.08s/it]
Training Epoch: 16/100, batch 158/563 completed (loss: 0.0016:  28%|██▊       | 80/281 [08:02<20:17,  6.06s/it]
Training Epoch: 16/100, batch 158/563 completed (loss: 0.2553:  28%|██▊       | 80/281 [08:02<20:18,  6.06s/it]
Training Epoch: 16/100, batch 158/563 completed (loss: 0.0022:  28%|██▊       | 80/281 [08:02<20:17,  6.06s/it]
Training Epoch: 16/100, batch 158/563 completed (loss: 0.0082:  28%|██▊       | 80/281 [08:02<20:18,  6.06s/it]
Training Epoch: 16/100, batch 159/563 completed (loss: 0.0363:  28%|██▊       | 80/281 [08:02<20:17,  6.06s/it]
Training Epoch: 16/100, batch 159/563 completed (loss: 0.0014:  28%|██▊       | 80/281 [08:02<20:18,  6.06s/it]
Training Epoch: 16/100, batch 159/563 completed (loss: 0.1405:  28%|██▊       | 80/281 [08:02<20:17,  6.06s/it]
Training Epoch: 16/100, batch 159/563 completed (loss: 0.0073:  28%|██▊       | 80/281 [08:02<20:18,  6.06s/it]
Training Epoch: 16/100, batch 160/563 completed (loss: 0.1642:  28%|██▊       | 80/281 [08:05<20:17,  6.06s/it]
Training Epoch: 16/100, batch 160/563 completed (loss: 0.2015:  28%|██▊       | 80/281 [08:05<20:17,  6.06s/it]
Training Epoch: 16/100, batch 160/563 completed (loss: 0.0122:  28%|██▊       | 80/281 [08:05<20:18,  6.06s/it]
Training Epoch: 16/100, batch 160/563 completed (loss: 0.0373:  28%|██▊       | 80/281 [08:05<20:18,  6.06s/it]
Training Epoch: 16/100, batch 160/563 completed (loss: 0.0122:  29%|██▉       | 81/281 [08:08<20:10,  6.05s/it]
Training Epoch: 16/100, batch 160/563 completed (loss: 0.1642:  29%|██▉       | 81/281 [08:08<20:10,  6.05s/it]
Training Epoch: 16/100, batch 160/563 completed (loss: 0.0373:  29%|██▉       | 81/281 [08:08<20:10,  6.05s/it]
Training Epoch: 16/100, batch 160/563 completed (loss: 0.2015:  29%|██▉       | 81/281 [08:08<20:10,  6.05s/it]
Training Epoch: 16/100, batch 161/563 completed (loss: 0.0231:  29%|██▉       | 81/281 [08:08<20:10,  6.05s/it]
Training Epoch: 16/100, batch 161/563 completed (loss: 0.0107:  29%|██▉       | 81/281 [08:08<20:10,  6.05s/it]
Training Epoch: 16/100, batch 161/563 completed (loss: 0.0183:  29%|██▉       | 81/281 [08:08<20:10,  6.05s/it]
Training Epoch: 16/100, batch 161/563 completed (loss: 0.0362:  29%|██▉       | 81/281 [08:08<20:10,  6.05s/it]
Training Epoch: 16/100, batch 162/563 completed (loss: 0.3487:  29%|██▉       | 81/281 [08:11<20:10,  6.05s/it]
Training Epoch: 16/100, batch 162/563 completed (loss: 0.0056:  29%|██▉       | 81/281 [08:11<20:10,  6.05s/it]
Training Epoch: 16/100, batch 162/563 completed (loss: 0.1277:  29%|██▉       | 81/281 [08:11<20:10,  6.05s/it]
Training Epoch: 16/100, batch 162/563 completed (loss: 0.1329:  29%|██▉       | 81/281 [08:11<20:10,  6.05s/it]
Training Epoch: 16/100, batch 162/563 completed (loss: 0.1329:  29%|██▉       | 82/281 [08:15<20:10,  6.08s/it]
Training Epoch: 16/100, batch 162/563 completed (loss: 0.3487:  29%|██▉       | 82/281 [08:15<20:10,  6.08s/it]
Training Epoch: 16/100, batch 162/563 completed (loss: 0.0056:  29%|██▉       | 82/281 [08:15<20:10,  6.08s/it]
Training Epoch: 16/100, batch 162/563 completed (loss: 0.1277:  29%|██▉       | 82/281 [08:15<20:10,  6.08s/it]
Training Epoch: 16/100, batch 163/563 completed (loss: 0.0194:  29%|██▉       | 82/281 [08:15<20:10,  6.08s/it]
Training Epoch: 16/100, batch 163/563 completed (loss: 0.0649:  29%|██▉       | 82/281 [08:15<20:10,  6.08s/it]
Training Epoch: 16/100, batch 163/563 completed (loss: 0.0317:  29%|██▉       | 82/281 [08:15<20:10,  6.08s/it]
Training Epoch: 16/100, batch 163/563 completed (loss: 0.0186:  29%|██▉       | 82/281 [08:15<20:10,  6.08s/it]
Training Epoch: 16/100, batch 164/563 completed (loss: 0.0126:  29%|██▉       | 82/281 [08:18<20:10,  6.08s/it]
Training Epoch: 16/100, batch 164/563 completed (loss: 0.0013:  29%|██▉       | 82/281 [08:18<20:10,  6.08s/it]
Training Epoch: 16/100, batch 164/563 completed (loss: 0.0134:  29%|██▉       | 82/281 [08:18<20:10,  6.08s/it]
Training Epoch: 16/100, batch 164/563 completed (loss: 0.0373:  29%|██▉       | 82/281 [08:18<20:10,  6.08s/it]
Training Epoch: 16/100, batch 164/563 completed (loss: 0.0373:  30%|██▉       | 83/281 [08:20<19:49,  6.01s/it]
Training Epoch: 16/100, batch 164/563 completed (loss: 0.0134:  30%|██▉       | 83/281 [08:20<19:49,  6.01s/it]
Training Epoch: 16/100, batch 164/563 completed (loss: 0.0126:  30%|██▉       | 83/281 [08:20<19:49,  6.01s/it]
Training Epoch: 16/100, batch 164/563 completed (loss: 0.0013:  30%|██▉       | 83/281 [08:20<19:49,  6.01s/it]
Training Epoch: 16/100, batch 165/563 completed (loss: 0.1437:  30%|██▉       | 83/281 [08:20<19:49,  6.01s/it]
Training Epoch: 16/100, batch 165/563 completed (loss: 0.0285:  30%|██▉       | 83/281 [08:20<19:49,  6.01s/it]
Training Epoch: 16/100, batch 165/563 completed (loss: 0.0374:  30%|██▉       | 83/281 [08:20<19:49,  6.01s/it]
Training Epoch: 16/100, batch 165/563 completed (loss: 0.0911:  30%|██▉       | 83/281 [08:20<19:49,  6.01s/it]
Training Epoch: 16/100, batch 166/563 completed (loss: 0.182:  30%|██▉       | 83/281 [08:23<19:49,  6.01s/it] 
Training Epoch: 16/100, batch 166/563 completed (loss: 0.0031:  30%|██▉       | 83/281 [08:23<19:49,  6.01s/it]
Training Epoch: 16/100, batch 166/563 completed (loss: 0.021:  30%|██▉       | 83/281 [08:23<19:49,  6.01s/it] 
Training Epoch: 16/100, batch 166/563 completed (loss: 0.0057:  30%|██▉       | 83/281 [08:23<19:49,  6.01s/it]
Training Epoch: 16/100, batch 166/563 completed (loss: 0.182:  30%|██▉       | 84/281 [08:26<19:44,  6.01s/it]
Training Epoch: 16/100, batch 166/563 completed (loss: 0.021:  30%|██▉       | 84/281 [08:26<19:44,  6.01s/it]
Training Epoch: 16/100, batch 166/563 completed (loss: 0.0031:  30%|██▉       | 84/281 [08:26<19:44,  6.01s/it]
Training Epoch: 16/100, batch 166/563 completed (loss: 0.0057:  30%|██▉       | 84/281 [08:26<19:45,  6.02s/it]
Training Epoch: 16/100, batch 167/563 completed (loss: 0.0036:  30%|██▉       | 84/281 [08:26<19:44,  6.01s/it]
Training Epoch: 16/100, batch 167/563 completed (loss: 0.4971:  30%|██▉       | 84/281 [08:26<19:44,  6.01s/it]
Training Epoch: 16/100, batch 167/563 completed (loss: 0.1366:  30%|██▉       | 84/281 [08:26<19:44,  6.01s/it]
Training Epoch: 16/100, batch 167/563 completed (loss: 0.0153:  30%|██▉       | 84/281 [08:27<19:45,  6.02s/it]
Training Epoch: 16/100, batch 168/563 completed (loss: 0.0151:  30%|██▉       | 84/281 [08:30<19:44,  6.01s/it]
Training Epoch: 16/100, batch 168/563 completed (loss: 0.0191:  30%|██▉       | 84/281 [08:30<19:44,  6.01s/it]
Training Epoch: 16/100, batch 168/563 completed (loss: 0.095:  30%|██▉       | 84/281 [08:30<19:44,  6.01s/it] 
Training Epoch: 16/100, batch 168/563 completed (loss: 0.0461:  30%|██▉       | 84/281 [08:30<19:45,  6.02s/it]
Training Epoch: 16/100, batch 168/563 completed (loss: 0.0461:  30%|███       | 85/281 [08:33<19:54,  6.10s/it]
Training Epoch: 16/100, batch 168/563 completed (loss: 0.0191:  30%|███       | 85/281 [08:33<19:54,  6.10s/it]
Training Epoch: 16/100, batch 168/563 completed (loss: 0.0151:  30%|███       | 85/281 [08:33<19:54,  6.10s/it]
Training Epoch: 16/100, batch 168/563 completed (loss: 0.095:  30%|███       | 85/281 [08:33<19:55,  6.10s/it]
Training Epoch: 16/100, batch 169/563 completed (loss: 0.0177:  30%|███       | 85/281 [08:33<19:54,  6.10s/it]
Training Epoch: 16/100, batch 169/563 completed (loss: 0.0044:  30%|███       | 85/281 [08:33<19:55,  6.10s/it]
Training Epoch: 16/100, batch 169/563 completed (loss: 0.4548:  30%|███       | 85/281 [08:33<19:54,  6.10s/it]
Training Epoch: 16/100, batch 169/563 completed (loss: 0.0362:  30%|███       | 85/281 [08:33<19:54,  6.10s/it]
Training Epoch: 16/100, batch 170/563 completed (loss: 0.3009:  30%|███       | 85/281 [08:36<19:54,  6.10s/it]
Training Epoch: 16/100, batch 170/563 completed (loss: 0.0596:  30%|███       | 85/281 [08:36<19:55,  6.10s/it]
Training Epoch: 16/100, batch 170/563 completed (loss: 0.0188:  30%|███       | 85/281 [08:36<19:54,  6.10s/it]
Training Epoch: 16/100, batch 170/563 completed (loss: 0.0028:  30%|███       | 85/281 [08:36<19:54,  6.10s/it]
Training Epoch: 16/100, batch 170/563 completed (loss: 0.0028:  31%|███       | 86/281 [08:39<19:33,  6.02s/it]
Training Epoch: 16/100, batch 170/563 completed (loss: 0.0188:  31%|███       | 86/281 [08:39<19:34,  6.02s/it]
Training Epoch: 16/100, batch 170/563 completed (loss: 0.3009:  31%|███       | 86/281 [08:39<19:34,  6.02s/it]
Training Epoch: 16/100, batch 170/563 completed (loss: 0.0596:  31%|███       | 86/281 [08:39<19:34,  6.02s/it]
Training Epoch: 16/100, batch 171/563 completed (loss: 0.0079:  31%|███       | 86/281 [08:39<19:34,  6.02s/it]
Training Epoch: 16/100, batch 171/563 completed (loss: 0.0075:  31%|███       | 86/281 [08:39<19:34,  6.02s/it]
Training Epoch: 16/100, batch 171/563 completed (loss: 0.0547:  31%|███       | 86/281 [08:39<19:34,  6.02s/it]
Training Epoch: 16/100, batch 171/563 completed (loss: 0.1551:  31%|███       | 86/281 [08:39<19:33,  6.02s/it]
Training Epoch: 16/100, batch 172/563 completed (loss: 0.4579:  31%|███       | 86/281 [08:42<19:34,  6.02s/it]
Training Epoch: 16/100, batch 172/563 completed (loss: 0.0306:  31%|███       | 86/281 [08:42<19:34,  6.02s/it]
Training Epoch: 16/100, batch 172/563 completed (loss: 0.0071:  31%|███       | 86/281 [08:42<19:34,  6.02s/it]
Training Epoch: 16/100, batch 172/563 completed (loss: 0.0125:  31%|███       | 86/281 [08:42<19:33,  6.02s/it]
Training Epoch: 16/100, batch 172/563 completed (loss: 0.0125:  31%|███       | 87/281 [08:45<19:45,  6.11s/it]
Training Epoch: 16/100, batch 172/563 completed (loss: 0.4579:  31%|███       | 87/281 [08:45<19:44,  6.11s/it]
Training Epoch: 16/100, batch 172/563 completed (loss: 0.0306:  31%|███       | 87/281 [08:45<19:44,  6.11s/it]
Training Epoch: 16/100, batch 172/563 completed (loss: 0.0071:  31%|███       | 87/281 [08:45<19:44,  6.11s/it]
Training Epoch: 16/100, batch 173/563 completed (loss: 0.0095:  31%|███       | 87/281 [08:45<19:44,  6.11s/it]
Training Epoch: 16/100, batch 173/563 completed (loss: 0.3606:  31%|███       | 87/281 [08:45<19:44,  6.11s/it]
Training Epoch: 16/100, batch 173/563 completed (loss: 0.3962:  31%|███       | 87/281 [08:45<19:44,  6.11s/it]
Training Epoch: 16/100, batch 173/563 completed (loss: 0.0216:  31%|███       | 87/281 [08:45<19:45,  6.11s/it]
Training Epoch: 16/100, batch 174/563 completed (loss: 0.026:  31%|███       | 87/281 [08:48<19:44,  6.11s/it] 
Training Epoch: 16/100, batch 174/563 completed (loss: 0.0274:  31%|███       | 87/281 [08:48<19:44,  6.11s/it]
Training Epoch: 16/100, batch 174/563 completed (loss: 0.0307:  31%|███       | 87/281 [08:48<19:44,  6.11s/it]
Training Epoch: 16/100, batch 174/563 completed (loss: 0.0059:  31%|███       | 87/281 [08:48<19:45,  6.11s/it]
Training Epoch: 16/100, batch 174/563 completed (loss: 0.0059:  31%|███▏      | 88/281 [08:51<19:24,  6.03s/it]
Training Epoch: 16/100, batch 174/563 completed (loss: 0.0307:  31%|███▏      | 88/281 [08:51<19:25,  6.04s/it]
Training Epoch: 16/100, batch 174/563 completed (loss: 0.026:  31%|███▏      | 88/281 [08:51<19:25,  6.04s/it]
Training Epoch: 16/100, batch 174/563 completed (loss: 0.0274:  31%|███▏      | 88/281 [08:51<19:27,  6.05s/it]
Training Epoch: 16/100, batch 175/563 completed (loss: 0.0289:  31%|███▏      | 88/281 [08:51<19:25,  6.04s/it]
Training Epoch: 16/100, batch 175/563 completed (loss: 0.0056:  31%|███▏      | 88/281 [08:51<19:25,  6.04s/it]
Training Epoch: 16/100, batch 175/563 completed (loss: 0.0065:  31%|███▏      | 88/281 [08:51<19:27,  6.05s/it]
Training Epoch: 16/100, batch 175/563 completed (loss: 0.6987:  31%|███▏      | 88/281 [08:51<19:24,  6.03s/it]
Training Epoch: 16/100, batch 176/563 completed (loss: 0.0373:  31%|███▏      | 88/281 [08:54<19:27,  6.05s/it]
Training Epoch: 16/100, batch 176/563 completed (loss: 0.0146:  31%|███▏      | 88/281 [08:54<19:25,  6.04s/it]
Training Epoch: 16/100, batch 176/563 completed (loss: 0.0445:  31%|███▏      | 88/281 [08:54<19:25,  6.04s/it]
Training Epoch: 16/100, batch 176/563 completed (loss: 0.0516:  31%|███▏      | 88/281 [08:54<19:24,  6.03s/it]
Training Epoch: 16/100, batch 176/563 completed (loss: 0.0146:  32%|███▏      | 89/281 [08:57<19:19,  6.04s/it]
Training Epoch: 16/100, batch 176/563 completed (loss: 0.0373:  32%|███▏      | 89/281 [08:57<19:18,  6.03s/it]
Training Epoch: 16/100, batch 176/563 completed (loss: 0.0516:  32%|███▏      | 89/281 [08:57<19:20,  6.04s/it]
Training Epoch: 16/100, batch 176/563 completed (loss: 0.0445:  32%|███▏      | 89/281 [08:57<19:19,  6.04s/it]
Training Epoch: 16/100, batch 177/563 completed (loss: 0.1848:  32%|███▏      | 89/281 [08:57<19:19,  6.04s/it]
Training Epoch: 16/100, batch 177/563 completed (loss: 0.2994:  32%|███▏      | 89/281 [08:57<19:18,  6.03s/it]
Training Epoch: 16/100, batch 177/563 completed (loss: 0.0058:  32%|███▏      | 89/281 [08:57<19:19,  6.04s/it]
Training Epoch: 16/100, batch 177/563 completed (loss: 0.0181:  32%|███▏      | 89/281 [08:57<19:20,  6.04s/it]
Training Epoch: 16/100, batch 178/563 completed (loss: 0.0101:  32%|███▏      | 89/281 [09:00<19:18,  6.03s/it]
Training Epoch: 16/100, batch 178/563 completed (loss: 0.004:  32%|███▏      | 89/281 [09:00<19:19,  6.04s/it] 
Training Epoch: 16/100, batch 178/563 completed (loss: 0.0769:  32%|███▏      | 89/281 [09:00<19:19,  6.04s/it]
Training Epoch: 16/100, batch 178/563 completed (loss: 0.0786:  32%|███▏      | 89/281 [09:00<19:20,  6.04s/it]
Training Epoch: 16/100, batch 178/563 completed (loss: 0.0786:  32%|███▏      | 90/281 [09:03<19:32,  6.14s/it]
Training Epoch: 16/100, batch 178/563 completed (loss: 0.004:  32%|███▏      | 90/281 [09:03<19:32,  6.14s/it]
Training Epoch: 16/100, batch 178/563 completed (loss: 0.0769:  32%|███▏      | 90/281 [09:03<19:33,  6.15s/it]
Training Epoch: 16/100, batch 178/563 completed (loss: 0.0101:  32%|███▏      | 90/281 [09:03<19:33,  6.14s/it]
Training Epoch: 16/100, batch 179/563 completed (loss: 0.048:  32%|███▏      | 90/281 [09:03<19:32,  6.14s/it]
Training Epoch: 16/100, batch 179/563 completed (loss: 0.2223:  32%|███▏      | 90/281 [09:03<19:33,  6.14s/it]
Training Epoch: 16/100, batch 179/563 completed (loss: 0.0118:  32%|███▏      | 90/281 [09:03<19:33,  6.15s/it]
Training Epoch: 16/100, batch 179/563 completed (loss: 0.0199:  32%|███▏      | 90/281 [09:03<19:32,  6.14s/it]
Training Epoch: 16/100, batch 180/563 completed (loss: 0.0777:  32%|███▏      | 90/281 [09:06<19:33,  6.14s/it]
Training Epoch: 16/100, batch 180/563 completed (loss: 0.0084:  32%|███▏      | 90/281 [09:06<19:33,  6.15s/it]
Training Epoch: 16/100, batch 180/563 completed (loss: 0.0263:  32%|███▏      | 90/281 [09:06<19:32,  6.14s/it]
Training Epoch: 16/100, batch 180/563 completed (loss: 0.2536:  32%|███▏      | 90/281 [09:06<19:32,  6.14s/it]
Training Epoch: 16/100, batch 180/563 completed (loss: 0.0263:  32%|███▏      | 91/281 [09:09<19:09,  6.05s/it]
Training Epoch: 16/100, batch 180/563 completed (loss: 0.2536:  32%|███▏      | 91/281 [09:09<19:10,  6.06s/it]
Training Epoch: 16/100, batch 180/563 completed (loss: 0.0084:  32%|███▏      | 91/281 [09:09<19:09,  6.05s/it]
Training Epoch: 16/100, batch 180/563 completed (loss: 0.0777:  32%|███▏      | 91/281 [09:09<19:09,  6.05s/it]
Training Epoch: 16/100, batch 181/563 completed (loss: 0.1098:  32%|███▏      | 91/281 [09:09<19:09,  6.05s/it]
Training Epoch: 16/100, batch 181/563 completed (loss: 0.1191:  32%|███▏      | 91/281 [09:09<19:09,  6.05s/it]
Training Epoch: 16/100, batch 181/563 completed (loss: 0.1624:  32%|███▏      | 91/281 [09:09<19:09,  6.05s/it]
Training Epoch: 16/100, batch 181/563 completed (loss: 0.0073:  32%|███▏      | 91/281 [09:09<19:10,  6.06s/it]
Training Epoch: 16/100, batch 182/563 completed (loss: 0.004:  32%|███▏      | 91/281 [09:12<19:09,  6.05s/it] 
Training Epoch: 16/100, batch 182/563 completed (loss: 0.0133:  32%|███▏      | 91/281 [09:12<19:09,  6.05s/it]
Training Epoch: 16/100, batch 182/563 completed (loss: 0.1015:  32%|███▏      | 91/281 [09:12<19:09,  6.05s/it]
Training Epoch: 16/100, batch 182/563 completed (loss: 0.0135:  32%|███▏      | 91/281 [09:12<19:10,  6.06s/it]
Training Epoch: 16/100, batch 182/563 completed (loss: 0.0135:  33%|███▎      | 92/281 [09:15<19:18,  6.13s/it]
Training Epoch: 16/100, batch 182/563 completed (loss: 0.1015:  33%|███▎      | 92/281 [09:15<19:17,  6.13s/it]
Training Epoch: 16/100, batch 182/563 completed (loss: 0.0133:  33%|███▎      | 92/281 [09:15<19:17,  6.13s/it]
Training Epoch: 16/100, batch 182/563 completed (loss: 0.004:  33%|███▎      | 92/281 [09:15<19:18,  6.13s/it]
Training Epoch: 16/100, batch 183/563 completed (loss: 0.0353:  33%|███▎      | 92/281 [09:15<19:18,  6.13s/it]
Training Epoch: 16/100, batch 183/563 completed (loss: 0.0447:  33%|███▎      | 92/281 [09:15<19:17,  6.13s/it]
Training Epoch: 16/100, batch 183/563 completed (loss: 0.0183:  33%|███▎      | 92/281 [09:15<19:17,  6.13s/it]
Training Epoch: 16/100, batch 183/563 completed (loss: 0.029:  33%|███▎      | 92/281 [09:15<19:18,  6.13s/it] 
Training Epoch: 16/100, batch 184/563 completed (loss: 0.0345:  33%|███▎      | 92/281 [09:18<19:17,  6.13s/it]
Training Epoch: 16/100, batch 184/563 completed (loss: 0.0443:  33%|███▎      | 92/281 [09:18<19:18,  6.13s/it]
Training Epoch: 16/100, batch 184/563 completed (loss: 0.0151:  33%|███▎      | 92/281 [09:18<19:17,  6.13s/it]
Training Epoch: 16/100, batch 184/563 completed (loss: 0.1187:  33%|███▎      | 92/281 [09:18<19:18,  6.13s/it]
Training Epoch: 16/100, batch 184/563 completed (loss: 0.1187:  33%|███▎      | 93/281 [09:21<19:01,  6.07s/it]
Training Epoch: 16/100, batch 184/563 completed (loss: 0.0443:  33%|███▎      | 93/281 [09:21<19:01,  6.07s/it]
Training Epoch: 16/100, batch 184/563 completed (loss: 0.0345:  33%|███▎      | 93/281 [09:21<19:01,  6.07s/it]
Training Epoch: 16/100, batch 184/563 completed (loss: 0.0151:  33%|███▎      | 93/281 [09:21<19:02,  6.08s/it]
Training Epoch: 16/100, batch 185/563 completed (loss: 0.0082:  33%|███▎      | 93/281 [09:21<19:01,  6.07s/it]
Training Epoch: 16/100, batch 185/563 completed (loss: 0.0122:  33%|███▎      | 93/281 [09:21<19:01,  6.07s/it]
Training Epoch: 16/100, batch 185/563 completed (loss: 0.0504:  33%|███▎      | 93/281 [09:21<19:02,  6.08s/it]
Training Epoch: 16/100, batch 185/563 completed (loss: 0.0204:  33%|███▎      | 93/281 [09:21<19:01,  6.07s/it]
Training Epoch: 16/100, batch 186/563 completed (loss: 0.3822:  33%|███▎      | 93/281 [09:24<19:01,  6.07s/it]
Training Epoch: 16/100, batch 186/563 completed (loss: 0.0354:  33%|███▎      | 93/281 [09:24<19:01,  6.07s/it]
Training Epoch: 16/100, batch 186/563 completed (loss: 0.0081:  33%|███▎      | 93/281 [09:24<19:02,  6.08s/it]
Training Epoch: 16/100, batch 186/563 completed (loss: 0.0561:  33%|███▎      | 93/281 [09:24<19:01,  6.07s/it]
Training Epoch: 16/100, batch 186/563 completed (loss: 0.3822:  33%|███▎      | 94/281 [09:27<18:43,  6.01s/it]
Training Epoch: 16/100, batch 186/563 completed (loss: 0.0081:  33%|███▎      | 94/281 [09:27<18:43,  6.01s/it]
Training Epoch: 16/100, batch 186/563 completed (loss: 0.0354:  33%|███▎      | 94/281 [09:27<18:43,  6.01s/it]
Training Epoch: 16/100, batch 186/563 completed (loss: 0.0561:  33%|███▎      | 94/281 [09:27<18:44,  6.01s/it]
Training Epoch: 16/100, batch 187/563 completed (loss: 0.0328:  33%|███▎      | 94/281 [09:27<18:43,  6.01s/it]
Training Epoch: 16/100, batch 187/563 completed (loss: 0.0156:  33%|███▎      | 94/281 [09:27<18:43,  6.01s/it]
Training Epoch: 16/100, batch 187/563 completed (loss: 0.0523:  33%|███▎      | 94/281 [09:27<18:43,  6.01s/it]
Training Epoch: 16/100, batch 187/563 completed (loss: 0.0281:  33%|███▎      | 94/281 [09:27<18:44,  6.01s/it]
Training Epoch: 16/100, batch 188/563 completed (loss: 0.0492:  33%|███▎      | 94/281 [09:31<18:43,  6.01s/it]
Training Epoch: 16/100, batch 188/563 completed (loss: 0.2604:  33%|███▎      | 94/281 [09:31<18:43,  6.01s/it]
Training Epoch: 16/100, batch 188/563 completed (loss: 0.0469:  33%|███▎      | 94/281 [09:31<18:43,  6.01s/it]
Training Epoch: 16/100, batch 188/563 completed (loss: 0.0128:  33%|███▎      | 94/281 [09:31<18:44,  6.01s/it]
Training Epoch: 16/100, batch 188/563 completed (loss: 0.2604:  34%|███▍      | 95/281 [09:34<19:02,  6.14s/it]
Training Epoch: 16/100, batch 188/563 completed (loss: 0.0492:  34%|███▍      | 95/281 [09:34<19:03,  6.15s/it]
Training Epoch: 16/100, batch 188/563 completed (loss: 0.0128:  34%|███▍      | 95/281 [09:34<19:03,  6.15s/it]
Training Epoch: 16/100, batch 188/563 completed (loss: 0.0469:  34%|███▍      | 95/281 [09:34<19:03,  6.15s/it]
Training Epoch: 16/100, batch 189/563 completed (loss: 0.1118:  34%|███▍      | 95/281 [09:34<19:03,  6.15s/it]
Training Epoch: 16/100, batch 189/563 completed (loss: 0.0196:  34%|███▍      | 95/281 [09:34<19:03,  6.15s/it]
Training Epoch: 16/100, batch 189/563 completed (loss: 0.0173:  34%|███▍      | 95/281 [09:34<19:02,  6.14s/it]
Training Epoch: 16/100, batch 189/563 completed (loss: 0.003:  34%|███▍      | 95/281 [09:34<19:03,  6.15s/it] 
Training Epoch: 16/100, batch 190/563 completed (loss: 0.0067:  34%|███▍      | 95/281 [09:37<19:03,  6.15s/it]
Training Epoch: 16/100, batch 190/563 completed (loss: 0.0103:  34%|███▍      | 95/281 [09:37<19:03,  6.15s/it]
Training Epoch: 16/100, batch 190/563 completed (loss: 0.0646:  34%|███▍      | 95/281 [09:37<19:02,  6.14s/it]
Training Epoch: 16/100, batch 190/563 completed (loss: 0.434:  34%|███▍      | 95/281 [09:37<19:03,  6.15s/it]
Training Epoch: 16/100, batch 190/563 completed (loss: 0.0646:  34%|███▍      | 96/281 [09:39<18:40,  6.06s/it]
Training Epoch: 16/100, batch 190/563 completed (loss: 0.434:  34%|███▍      | 96/281 [09:39<18:41,  6.06s/it]
Training Epoch: 16/100, batch 190/563 completed (loss: 0.0103:  34%|███▍      | 96/281 [09:39<18:40,  6.06s/it]
Training Epoch: 16/100, batch 190/563 completed (loss: 0.0067:  34%|███▍      | 96/281 [09:39<18:41,  6.06s/it]
Training Epoch: 16/100, batch 191/563 completed (loss: 0.0134:  34%|███▍      | 96/281 [09:39<18:41,  6.06s/it]
Training Epoch: 16/100, batch 191/563 completed (loss: 0.0221:  34%|███▍      | 96/281 [09:39<18:40,  6.06s/it]
Training Epoch: 16/100, batch 191/563 completed (loss: 0.0791:  34%|███▍      | 96/281 [09:39<18:40,  6.06s/it]
Training Epoch: 16/100, batch 191/563 completed (loss: 0.0384:  34%|███▍      | 96/281 [09:40<18:41,  6.06s/it]
Training Epoch: 16/100, batch 192/563 completed (loss: 0.0048:  34%|███▍      | 96/281 [09:42<18:41,  6.06s/it]
Training Epoch: 16/100, batch 192/563 completed (loss: 0.0981:  34%|███▍      | 96/281 [09:42<18:40,  6.06s/it]
Training Epoch: 16/100, batch 192/563 completed (loss: 0.0057:  34%|███▍      | 96/281 [09:42<18:40,  6.06s/it]
Training Epoch: 16/100, batch 192/563 completed (loss: 0.1542:  34%|███▍      | 96/281 [09:42<18:41,  6.06s/it]
Training Epoch: 16/100, batch 192/563 completed (loss: 0.0057:  35%|███▍      | 97/281 [09:46<18:42,  6.10s/it]
Training Epoch: 16/100, batch 192/563 completed (loss: 0.1542:  35%|███▍      | 97/281 [09:46<18:42,  6.10s/it]
Training Epoch: 16/100, batch 192/563 completed (loss: 0.0981:  35%|███▍      | 97/281 [09:46<18:42,  6.10s/it]
Training Epoch: 16/100, batch 192/563 completed (loss: 0.0048:  35%|███▍      | 97/281 [09:46<18:42,  6.10s/it]
Training Epoch: 16/100, batch 193/563 completed (loss: 0.0036:  35%|███▍      | 97/281 [09:46<18:42,  6.10s/it]
Training Epoch: 16/100, batch 193/563 completed (loss: 0.0186:  35%|███▍      | 97/281 [09:46<18:42,  6.10s/it]
Training Epoch: 16/100, batch 193/563 completed (loss: 0.0393:  35%|███▍      | 97/281 [09:46<18:42,  6.10s/it]
Training Epoch: 16/100, batch 193/563 completed (loss: 0.0025:  35%|███▍      | 97/281 [09:46<18:42,  6.10s/it]
Training Epoch: 16/100, batch 194/563 completed (loss: 0.0257:  35%|███▍      | 97/281 [09:49<18:42,  6.10s/it]
Training Epoch: 16/100, batch 194/563 completed (loss: 0.0166:  35%|███▍      | 97/281 [09:49<18:42,  6.10s/it]
Training Epoch: 16/100, batch 194/563 completed (loss: 0.0041:  35%|███▍      | 97/281 [09:49<18:42,  6.10s/it]
Training Epoch: 16/100, batch 194/563 completed (loss: 0.031:  35%|███▍      | 97/281 [09:49<18:42,  6.10s/it] 
Training Epoch: 16/100, batch 194/563 completed (loss: 0.031:  35%|███▍      | 98/281 [09:52<18:30,  6.07s/it]
Training Epoch: 16/100, batch 194/563 completed (loss: 0.0041:  35%|███▍      | 98/281 [09:52<18:30,  6.07s/it]
Training Epoch: 16/100, batch 194/563 completed (loss: 0.0166:  35%|███▍      | 98/281 [09:52<18:30,  6.07s/it]
Training Epoch: 16/100, batch 194/563 completed (loss: 0.0257:  35%|███▍      | 98/281 [09:52<18:30,  6.07s/it]
Training Epoch: 16/100, batch 195/563 completed (loss: 0.0015:  35%|███▍      | 98/281 [09:52<18:30,  6.07s/it]
Training Epoch: 16/100, batch 195/563 completed (loss: 0.0284:  35%|███▍      | 98/281 [09:52<18:30,  6.07s/it]
Training Epoch: 16/100, batch 195/563 completed (loss: 0.1211:  35%|███▍      | 98/281 [09:52<18:30,  6.07s/it]
Training Epoch: 16/100, batch 195/563 completed (loss: 0.0707:  35%|███▍      | 98/281 [09:52<18:30,  6.07s/it]
Training Epoch: 16/100, batch 196/563 completed (loss: 0.0389:  35%|███▍      | 98/281 [09:55<18:30,  6.07s/it]
Training Epoch: 16/100, batch 196/563 completed (loss: 0.0205:  35%|███▍      | 98/281 [09:55<18:30,  6.07s/it]
Training Epoch: 16/100, batch 196/563 completed (loss: 0.0227:  35%|███▍      | 98/281 [09:55<18:30,  6.07s/it]
Training Epoch: 16/100, batch 196/563 completed (loss: 0.0024:  35%|███▍      | 98/281 [09:55<18:30,  6.07s/it]
Training Epoch: 16/100, batch 196/563 completed (loss: 0.0024:  35%|███▌      | 99/281 [09:58<18:13,  6.01s/it]
Training Epoch: 16/100, batch 196/563 completed (loss: 0.0227:  35%|███▌      | 99/281 [09:58<18:14,  6.01s/it]
Training Epoch: 16/100, batch 196/563 completed (loss: 0.0389:  35%|███▌      | 99/281 [09:58<18:14,  6.01s/it]
Training Epoch: 16/100, batch 196/563 completed (loss: 0.0205:  35%|███▌      | 99/281 [09:58<18:14,  6.01s/it]
Training Epoch: 16/100, batch 197/563 completed (loss: 0.0679:  35%|███▌      | 99/281 [09:58<18:14,  6.01s/it]
Training Epoch: 16/100, batch 197/563 completed (loss: 0.0097:  35%|███▌      | 99/281 [09:58<18:14,  6.01s/it]
Training Epoch: 16/100, batch 197/563 completed (loss: 0.012:  35%|███▌      | 99/281 [09:58<18:14,  6.01s/it] 
Training Epoch: 16/100, batch 197/563 completed (loss: 0.1301:  35%|███▌      | 99/281 [09:58<18:13,  6.01s/it]
Training Epoch: 16/100, batch 198/563 completed (loss: 0.0094:  35%|███▌      | 99/281 [10:00<18:14,  6.01s/it]
Training Epoch: 16/100, batch 198/563 completed (loss: 0.0031:  35%|███▌      | 99/281 [10:00<18:14,  6.01s/it]
Training Epoch: 16/100, batch 198/563 completed (loss: 0.0194:  35%|███▌      | 99/281 [10:00<18:14,  6.01s/it]
Training Epoch: 16/100, batch 198/563 completed (loss: 0.1383:  35%|███▌      | 99/281 [10:01<18:13,  6.01s/it]
Training Epoch: 16/100, batch 198/563 completed (loss: 0.0094:  36%|███▌      | 100/281 [10:04<18:24,  6.10s/it]
Training Epoch: 16/100, batch 198/563 completed (loss: 0.0194:  36%|███▌      | 100/281 [10:04<18:24,  6.10s/it]
Training Epoch: 16/100, batch 198/563 completed (loss: 0.1383:  36%|███▌      | 100/281 [10:04<18:24,  6.10s/it]
Training Epoch: 16/100, batch 198/563 completed (loss: 0.0031:  36%|███▌      | 100/281 [10:04<18:24,  6.10s/it]
Training Epoch: 16/100, batch 199/563 completed (loss: 0.15:  36%|███▌      | 100/281 [10:04<18:24,  6.10s/it]  
Training Epoch: 16/100, batch 199/563 completed (loss: 0.0902:  36%|███▌      | 100/281 [10:04<18:24,  6.10s/it]
Training Epoch: 16/100, batch 199/563 completed (loss: 0.0567:  36%|███▌      | 100/281 [10:04<18:24,  6.10s/it]
Training Epoch: 16/100, batch 199/563 completed (loss: 0.0069:  36%|███▌      | 100/281 [10:04<18:24,  6.10s/it]
Training Epoch: 16/100, batch 200/563 completed (loss: 0.014:  36%|███▌      | 100/281 [10:07<18:24,  6.10s/it] 
Training Epoch: 16/100, batch 200/563 completed (loss: 0.0583:  36%|███▌      | 100/281 [10:07<18:24,  6.10s/it]
Training Epoch: 16/100, batch 200/563 completed (loss: 0.2146:  36%|███▌      | 100/281 [10:07<18:24,  6.10s/it]
Training Epoch: 16/100, batch 200/563 completed (loss: 0.1778:  36%|███▌      | 100/281 [10:07<18:24,  6.10s/it]
Training Epoch: 16/100, batch 200/563 completed (loss: 0.2146:  36%|███▌      | 101/281 [10:10<18:09,  6.05s/it]
Training Epoch: 16/100, batch 200/563 completed (loss: 0.0583:  36%|███▌      | 101/281 [10:10<18:08,  6.05s/it]
Training Epoch: 16/100, batch 200/563 completed (loss: 0.014:  36%|███▌      | 101/281 [10:10<18:09,  6.05s/it]
Training Epoch: 16/100, batch 200/563 completed (loss: 0.1778:  36%|███▌      | 101/281 [10:10<18:09,  6.06s/it]
Training Epoch: 16/100, batch 201/563 completed (loss: 0.0367:  36%|███▌      | 101/281 [10:10<18:08,  6.05s/it]
Training Epoch: 16/100, batch 201/563 completed (loss: 0.0322:  36%|███▌      | 101/281 [10:10<18:09,  6.05s/it]
Training Epoch: 16/100, batch 201/563 completed (loss: 0.0126:  36%|███▌      | 101/281 [10:10<18:09,  6.05s/it]
Training Epoch: 16/100, batch 201/563 completed (loss: 0.0224:  36%|███▌      | 101/281 [10:10<18:09,  6.06s/it]
Training Epoch: 16/100, batch 202/563 completed (loss: 0.005:  36%|███▌      | 101/281 [10:13<18:09,  6.05s/it] 
Training Epoch: 16/100, batch 202/563 completed (loss: 0.0113:  36%|███▌      | 101/281 [10:13<18:08,  6.05s/it]
Training Epoch: 16/100, batch 202/563 completed (loss: 0.0374:  36%|███▌      | 101/281 [10:13<18:09,  6.05s/it]
Training Epoch: 16/100, batch 202/563 completed (loss: 0.0092:  36%|███▌      | 101/281 [10:13<18:09,  6.06s/it]
Training Epoch: 16/100, batch 202/563 completed (loss: 0.0113:  36%|███▋      | 102/281 [10:16<17:52,  5.99s/it]
Training Epoch: 16/100, batch 202/563 completed (loss: 0.0374:  36%|███▋      | 102/281 [10:16<17:52,  5.99s/it]
Training Epoch: 16/100, batch 202/563 completed (loss: 0.0092:  36%|███▋      | 102/281 [10:16<17:52,  5.99s/it]
Training Epoch: 16/100, batch 202/563 completed (loss: 0.005:  36%|███▋      | 102/281 [10:16<17:52,  5.99s/it]
Training Epoch: 16/100, batch 203/563 completed (loss: 0.0052:  36%|███▋      | 102/281 [10:16<17:52,  5.99s/it]
Training Epoch: 16/100, batch 203/563 completed (loss: 0.1015:  36%|███▋      | 102/281 [10:16<17:52,  5.99s/it]
Training Epoch: 16/100, batch 203/563 completed (loss: 0.0151:  36%|███▋      | 102/281 [10:16<17:52,  5.99s/it]
Training Epoch: 16/100, batch 203/563 completed (loss: 0.1725:  36%|███▋      | 102/281 [10:16<17:52,  5.99s/it]
Training Epoch: 16/100, batch 204/563 completed (loss: 0.0907:  36%|███▋      | 102/281 [10:19<17:52,  5.99s/it]
Training Epoch: 16/100, batch 204/563 completed (loss: 0.0546:  36%|███▋      | 102/281 [10:19<17:52,  5.99s/it]
Training Epoch: 16/100, batch 204/563 completed (loss: 0.0039:  36%|███▋      | 102/281 [10:19<17:52,  5.99s/it]
Training Epoch: 16/100, batch 204/563 completed (loss: 0.0116:  36%|███▋      | 102/281 [10:19<17:52,  5.99s/it]
Training Epoch: 16/100, batch 204/563 completed (loss: 0.0116:  37%|███▋      | 103/281 [10:22<18:00,  6.07s/it]
Training Epoch: 16/100, batch 204/563 completed (loss: 0.0039:  37%|███▋      | 103/281 [10:22<18:01,  6.08s/it]
Training Epoch: 16/100, batch 204/563 completed (loss: 0.0546:  37%|███▋      | 103/281 [10:22<18:01,  6.08s/it]
Training Epoch: 16/100, batch 204/563 completed (loss: 0.0907:  37%|███▋      | 103/281 [10:22<18:01,  6.08s/it]
Training Epoch: 16/100, batch 205/563 completed (loss: 0.4096:  37%|███▋      | 103/281 [10:22<18:01,  6.08s/it]
Training Epoch: 16/100, batch 205/563 completed (loss: 0.034:  37%|███▋      | 103/281 [10:22<18:01,  6.08s/it] 
Training Epoch: 16/100, batch 205/563 completed (loss: 0.0434:  37%|███▋      | 103/281 [10:22<18:01,  6.08s/it]
Training Epoch: 16/100, batch 205/563 completed (loss: 0.0493:  37%|███▋      | 103/281 [10:22<18:00,  6.07s/it]
Training Epoch: 16/100, batch 206/563 completed (loss: 0.009:  37%|███▋      | 103/281 [10:25<18:01,  6.08s/it]
Training Epoch: 16/100, batch 206/563 completed (loss: 0.1672:  37%|███▋      | 103/281 [10:25<18:01,  6.08s/it]
Training Epoch: 16/100, batch 206/563 completed (loss: 0.0388:  37%|███▋      | 103/281 [10:25<18:01,  6.08s/it]
Training Epoch: 16/100, batch 206/563 completed (loss: 0.0365:  37%|███▋      | 103/281 [10:25<18:00,  6.07s/it]
Training Epoch: 16/100, batch 206/563 completed (loss: 0.0365:  37%|███▋      | 104/281 [10:28<17:43,  6.01s/it]
Training Epoch: 16/100, batch 206/563 completed (loss: 0.1672:  37%|███▋      | 104/281 [10:28<17:43,  6.01s/it]
Training Epoch: 16/100, batch 206/563 completed (loss: 0.009:  37%|███▋      | 104/281 [10:28<17:43,  6.01s/it]
Training Epoch: 16/100, batch 206/563 completed (loss: 0.0388:  37%|███▋      | 104/281 [10:28<17:44,  6.01s/it]
Training Epoch: 16/100, batch 207/563 completed (loss: 0.0297:  37%|███▋      | 104/281 [10:28<17:44,  6.01s/it]
Training Epoch: 16/100, batch 207/563 completed (loss: 0.0238:  37%|███▋      | 104/281 [10:28<17:43,  6.01s/it]
Training Epoch: 16/100, batch 207/563 completed (loss: 0.0056:  37%|███▋      | 104/281 [10:28<17:43,  6.01s/it]
Training Epoch: 16/100, batch 207/563 completed (loss: 0.0325:  37%|███▋      | 104/281 [10:28<17:43,  6.01s/it]
Training Epoch: 16/100, batch 208/563 completed (loss: 0.5683:  37%|███▋      | 104/281 [10:31<17:43,  6.01s/it]
Training Epoch: 16/100, batch 208/563 completed (loss: 0.0063:  37%|███▋      | 104/281 [10:31<17:43,  6.01s/it]
Training Epoch: 16/100, batch 208/563 completed (loss: 0.0992:  37%|███▋      | 104/281 [10:31<17:44,  6.01s/it]
Training Epoch: 16/100, batch 208/563 completed (loss: 0.0535:  37%|███▋      | 104/281 [10:31<17:43,  6.01s/it]
Training Epoch: 16/100, batch 208/563 completed (loss: 0.0063:  37%|███▋      | 105/281 [10:34<17:51,  6.09s/it]
Training Epoch: 16/100, batch 208/563 completed (loss: 0.0992:  37%|███▋      | 105/281 [10:34<17:51,  6.09s/it]
Training Epoch: 16/100, batch 208/563 completed (loss: 0.5683:  37%|███▋      | 105/281 [10:34<17:51,  6.09s/it]
Training Epoch: 16/100, batch 208/563 completed (loss: 0.0535:  37%|███▋      | 105/281 [10:34<17:51,  6.09s/it]
Training Epoch: 16/100, batch 209/563 completed (loss: 0.0268:  37%|███▋      | 105/281 [10:34<17:51,  6.09s/it]
Training Epoch: 16/100, batch 209/563 completed (loss: 0.1691:  37%|███▋      | 105/281 [10:34<17:51,  6.09s/it]
Training Epoch: 16/100, batch 209/563 completed (loss: 0.1066:  37%|███▋      | 105/281 [10:34<17:51,  6.09s/it]
Training Epoch: 16/100, batch 209/563 completed (loss: 0.0933:  37%|███▋      | 105/281 [10:34<17:51,  6.09s/it]
Training Epoch: 16/100, batch 210/563 completed (loss: 0.0093:  37%|███▋      | 105/281 [10:37<17:51,  6.09s/it]
Training Epoch: 16/100, batch 210/563 completed (loss: 0.0071:  37%|███▋      | 105/281 [10:37<17:51,  6.09s/it]
Training Epoch: 16/100, batch 210/563 completed (loss: 0.2423:  37%|███▋      | 105/281 [10:37<17:51,  6.09s/it]
Training Epoch: 16/100, batch 210/563 completed (loss: 0.3225:  37%|███▋      | 105/281 [10:37<17:51,  6.09s/it]
Training Epoch: 16/100, batch 210/563 completed (loss: 0.0071:  38%|███▊      | 106/281 [10:40<17:41,  6.06s/it]
Training Epoch: 16/100, batch 210/563 completed (loss: 0.3225:  38%|███▊      | 106/281 [10:40<17:41,  6.06s/it]
Training Epoch: 16/100, batch 210/563 completed (loss: 0.0093:  38%|███▊      | 106/281 [10:40<17:41,  6.06s/it]
Training Epoch: 16/100, batch 210/563 completed (loss: 0.2423:  38%|███▊      | 106/281 [10:40<17:41,  6.06s/it]
Training Epoch: 16/100, batch 211/563 completed (loss: 0.1168:  38%|███▊      | 106/281 [10:40<17:41,  6.06s/it]
Training Epoch: 16/100, batch 211/563 completed (loss: 0.3431:  38%|███▊      | 106/281 [10:40<17:41,  6.06s/it]
Training Epoch: 16/100, batch 211/563 completed (loss: 0.1391:  38%|███▊      | 106/281 [10:40<17:41,  6.06s/it]
Training Epoch: 16/100, batch 211/563 completed (loss: 0.3984:  38%|███▊      | 106/281 [10:40<17:41,  6.06s/it]
Training Epoch: 16/100, batch 212/563 completed (loss: 0.0245:  38%|███▊      | 106/281 [10:43<17:41,  6.06s/it]
Training Epoch: 16/100, batch 212/563 completed (loss: 0.571:  38%|███▊      | 106/281 [10:43<17:41,  6.06s/it] 
Training Epoch: 16/100, batch 212/563 completed (loss: 0.0124:  38%|███▊      | 106/281 [10:43<17:41,  6.06s/it]
Training Epoch: 16/100, batch 212/563 completed (loss: 0.0098:  38%|███▊      | 106/281 [10:43<17:41,  6.06s/it]
Training Epoch: 16/100, batch 212/563 completed (loss: 0.571:  38%|███▊      | 107/281 [10:46<17:22,  5.99s/it]
Training Epoch: 16/100, batch 212/563 completed (loss: 0.0098:  38%|███▊      | 107/281 [10:46<17:22,  5.99s/it]
Training Epoch: 16/100, batch 212/563 completed (loss: 0.0124:  38%|███▊      | 107/281 [10:46<17:22,  5.99s/it]
Training Epoch: 16/100, batch 212/563 completed (loss: 0.0245:  38%|███▊      | 107/281 [10:46<17:22,  5.99s/it]
Training Epoch: 16/100, batch 213/563 completed (loss: 0.1279:  38%|███▊      | 107/281 [10:46<17:22,  5.99s/it]
Training Epoch: 16/100, batch 213/563 completed (loss: 0.0466:  38%|███▊      | 107/281 [10:46<17:22,  5.99s/it]
Training Epoch: 16/100, batch 213/563 completed (loss: 0.011:  38%|███▊      | 107/281 [10:46<17:22,  5.99s/it] 
Training Epoch: 16/100, batch 213/563 completed (loss: 0.0224:  38%|███▊      | 107/281 [10:46<17:22,  5.99s/it]
Training Epoch: 16/100, batch 214/563 completed (loss: 0.0051:  38%|███▊      | 107/281 [10:49<17:22,  5.99s/it]
Training Epoch: 16/100, batch 214/563 completed (loss: 0.0129:  38%|███▊      | 107/281 [10:49<17:22,  5.99s/it]
Training Epoch: 16/100, batch 214/563 completed (loss: 0.0056:  38%|███▊      | 107/281 [10:49<17:22,  5.99s/it]
Training Epoch: 16/100, batch 214/563 completed (loss: 0.0141:  38%|███▊      | 107/281 [10:49<17:22,  5.99s/it]
Training Epoch: 16/100, batch 214/563 completed (loss: 0.0056:  38%|███▊      | 108/281 [10:52<17:40,  6.13s/it]
Training Epoch: 16/100, batch 214/563 completed (loss: 0.0141:  38%|███▊      | 108/281 [10:52<17:40,  6.13s/it]
Training Epoch: 16/100, batch 214/563 completed (loss: 0.0129:  38%|███▊      | 108/281 [10:52<17:40,  6.13s/it]
Training Epoch: 16/100, batch 214/563 completed (loss: 0.0051:  38%|███▊      | 108/281 [10:52<17:41,  6.13s/it]
Training Epoch: 16/100, batch 215/563 completed (loss: 0.0046:  38%|███▊      | 108/281 [10:52<17:40,  6.13s/it]
Training Epoch: 16/100, batch 215/563 completed (loss: 0.015:  38%|███▊      | 108/281 [10:52<17:40,  6.13s/it] 
Training Epoch: 16/100, batch 215/563 completed (loss: 0.122:  38%|███▊      | 108/281 [10:52<17:41,  6.13s/it] 
Training Epoch: 16/100, batch 215/563 completed (loss: 0.5621:  38%|███▊      | 108/281 [10:52<17:40,  6.13s/it]
Training Epoch: 16/100, batch 216/563 completed (loss: 0.2258:  38%|███▊      | 108/281 [10:55<17:40,  6.13s/it]
Training Epoch: 16/100, batch 216/563 completed (loss: 0.3141:  38%|███▊      | 108/281 [10:55<17:40,  6.13s/it]
Training Epoch: 16/100, batch 216/563 completed (loss: 0.0427:  38%|███▊      | 108/281 [10:55<17:41,  6.13s/it]
Training Epoch: 16/100, batch 216/563 completed (loss: 0.1461:  38%|███▊      | 108/281 [10:55<17:40,  6.13s/it]
Training Epoch: 16/100, batch 216/563 completed (loss: 0.1461:  39%|███▉      | 109/281 [10:58<17:19,  6.04s/it]
Training Epoch: 16/100, batch 216/563 completed (loss: 0.2258:  39%|███▉      | 109/281 [10:58<17:19,  6.04s/it]
Training Epoch: 16/100, batch 216/563 completed (loss: 0.3141:  39%|███▉      | 109/281 [10:58<17:19,  6.04s/it]
Training Epoch: 16/100, batch 216/563 completed (loss: 0.0427:  39%|███▉      | 109/281 [10:58<17:19,  6.04s/it]
Training Epoch: 16/100, batch 217/563 completed (loss: 0.0014:  39%|███▉      | 109/281 [10:58<17:19,  6.04s/it]
Training Epoch: 16/100, batch 217/563 completed (loss: 0.0127:  39%|███▉      | 109/281 [10:58<17:19,  6.04s/it]
Training Epoch: 16/100, batch 217/563 completed (loss: 0.0133:  39%|███▉      | 109/281 [10:58<17:19,  6.04s/it]
Training Epoch: 16/100, batch 217/563 completed (loss: 1.0228:  39%|███▉      | 109/281 [10:58<17:19,  6.04s/it]
Training Epoch: 16/100, batch 218/563 completed (loss: 0.0231:  39%|███▉      | 109/281 [11:01<17:19,  6.04s/it]
Training Epoch: 16/100, batch 218/563 completed (loss: 0.0035:  39%|███▉      | 109/281 [11:01<17:19,  6.04s/it]
Training Epoch: 16/100, batch 218/563 completed (loss: 0.0351:  39%|███▉      | 109/281 [11:01<17:19,  6.04s/it]
Training Epoch: 16/100, batch 218/563 completed (loss: 0.0144:  39%|███▉      | 109/281 [11:01<17:19,  6.04s/it]
Training Epoch: 16/100, batch 218/563 completed (loss: 0.0144:  39%|███▉      | 110/281 [11:04<17:17,  6.07s/it]
Training Epoch: 16/100, batch 218/563 completed (loss: 0.0035:  39%|███▉      | 110/281 [11:04<17:18,  6.07s/it]
Training Epoch: 16/100, batch 218/563 completed (loss: 0.0351:  39%|███▉      | 110/281 [11:04<17:18,  6.08s/it]
Training Epoch: 16/100, batch 218/563 completed (loss: 0.0231:  39%|███▉      | 110/281 [11:04<17:18,  6.07s/it]
Training Epoch: 16/100, batch 219/563 completed (loss: 0.0883:  39%|███▉      | 110/281 [11:04<17:18,  6.08s/it]
Training Epoch: 16/100, batch 219/563 completed (loss: 0.0602:  39%|███▉      | 110/281 [11:04<17:18,  6.07s/it]
Training Epoch: 16/100, batch 219/563 completed (loss: 0.1851:  39%|███▉      | 110/281 [11:04<17:18,  6.07s/it]
Training Epoch: 16/100, batch 219/563 completed (loss: 0.0424:  39%|███▉      | 110/281 [11:04<17:17,  6.07s/it]
Training Epoch: 16/100, batch 220/563 completed (loss: 0.0244:  39%|███▉      | 110/281 [11:07<17:18,  6.08s/it]
Training Epoch: 16/100, batch 220/563 completed (loss: 0.3608:  39%|███▉      | 110/281 [11:07<17:18,  6.07s/it]
Training Epoch: 16/100, batch 220/563 completed (loss: 0.0143:  39%|███▉      | 110/281 [11:07<17:18,  6.07s/it]
Training Epoch: 16/100, batch 220/563 completed (loss: 0.1295:  39%|███▉      | 110/281 [11:07<17:17,  6.07s/it]
Training Epoch: 16/100, batch 220/563 completed (loss: 0.1295:  40%|███▉      | 111/281 [11:10<17:15,  6.09s/it]
Training Epoch: 16/100, batch 220/563 completed (loss: 0.0244:  40%|███▉      | 111/281 [11:10<17:15,  6.09s/it]
Training Epoch: 16/100, batch 220/563 completed (loss: 0.0143:  40%|███▉      | 111/281 [11:10<17:15,  6.09s/it]
Training Epoch: 16/100, batch 220/563 completed (loss: 0.3608:  40%|███▉      | 111/281 [11:10<17:15,  6.09s/it]
Training Epoch: 16/100, batch 221/563 completed (loss: 0.0114:  40%|███▉      | 111/281 [11:10<17:15,  6.09s/it]
Training Epoch: 16/100, batch 221/563 completed (loss: 0.022:  40%|███▉      | 111/281 [11:10<17:15,  6.09s/it] 
Training Epoch: 16/100, batch 221/563 completed (loss: 0.0796:  40%|███▉      | 111/281 [11:10<17:15,  6.09s/it]
Training Epoch: 16/100, batch 221/563 completed (loss: 0.0251:  40%|███▉      | 111/281 [11:11<17:15,  6.09s/it]
Training Epoch: 16/100, batch 222/563 completed (loss: 0.0054:  40%|███▉      | 111/281 [11:13<17:15,  6.09s/it]
Training Epoch: 16/100, batch 222/563 completed (loss: 0.0508:  40%|███▉      | 111/281 [11:13<17:15,  6.09s/it]
Training Epoch: 16/100, batch 222/563 completed (loss: 0.1817:  40%|███▉      | 111/281 [11:13<17:15,  6.09s/it]
Training Epoch: 16/100, batch 222/563 completed (loss: 0.0054:  40%|███▉      | 111/281 [11:13<17:15,  6.09s/it]
Training Epoch: 16/100, batch 222/563 completed (loss: 0.0054:  40%|███▉      | 112/281 [11:16<16:57,  6.02s/it]
Training Epoch: 16/100, batch 222/563 completed (loss: 0.0508:  40%|███▉      | 112/281 [11:16<16:58,  6.02s/it]
Training Epoch: 16/100, batch 222/563 completed (loss: 0.1817:  40%|███▉      | 112/281 [11:16<16:58,  6.02s/it]
Training Epoch: 16/100, batch 222/563 completed (loss: 0.0054:  40%|███▉      | 112/281 [11:16<16:58,  6.03s/it]
Training Epoch: 16/100, batch 223/563 completed (loss: 0.0206:  40%|███▉      | 112/281 [11:16<16:58,  6.03s/it]
Training Epoch: 16/100, batch 223/563 completed (loss: 0.066:  40%|███▉      | 112/281 [11:16<16:58,  6.02s/it] 
Training Epoch: 16/100, batch 223/563 completed (loss: 0.2091:  40%|███▉      | 112/281 [11:16<16:58,  6.02s/it]
Training Epoch: 16/100, batch 223/563 completed (loss: 0.1347:  40%|███▉      | 112/281 [11:16<16:57,  6.02s/it]
Training Epoch: 16/100, batch 224/563 completed (loss: 0.0418:  40%|███▉      | 112/281 [11:19<16:58,  6.02s/it]
Training Epoch: 16/100, batch 224/563 completed (loss: 0.1659:  40%|███▉      | 112/281 [11:19<16:58,  6.03s/it]
Training Epoch: 16/100, batch 224/563 completed (loss: 0.0362:  40%|███▉      | 112/281 [11:19<16:58,  6.02s/it]
Training Epoch: 16/100, batch 224/563 completed (loss: 0.0295:  40%|███▉      | 112/281 [11:19<16:57,  6.02s/it]
Training Epoch: 16/100, batch 224/563 completed (loss: 0.1659:  40%|████      | 113/281 [11:23<17:04,  6.10s/it]
Training Epoch: 16/100, batch 224/563 completed (loss: 0.0418:  40%|████      | 113/281 [11:23<17:04,  6.10s/it]
Training Epoch: 16/100, batch 224/563 completed (loss: 0.0362:  40%|████      | 113/281 [11:23<17:04,  6.10s/it]
Training Epoch: 16/100, batch 224/563 completed (loss: 0.0295:  40%|████      | 113/281 [11:23<17:05,  6.10s/it]
Training Epoch: 16/100, batch 225/563 completed (loss: 0.1372:  40%|████      | 113/281 [11:23<17:04,  6.10s/it]
Training Epoch: 16/100, batch 225/563 completed (loss: 0.1281:  40%|████      | 113/281 [11:23<17:04,  6.10s/it]
Training Epoch: 16/100, batch 225/563 completed (loss: 0.0289:  40%|████      | 113/281 [11:23<17:04,  6.10s/it]
Training Epoch: 16/100, batch 225/563 completed (loss: 0.0214:  40%|████      | 113/281 [11:23<17:05,  6.10s/it]
Training Epoch: 16/100, batch 226/563 completed (loss: 0.0238:  40%|████      | 113/281 [11:26<17:04,  6.10s/it]
Training Epoch: 16/100, batch 226/563 completed (loss: 0.0225:  40%|████      | 113/281 [11:26<17:04,  6.10s/it]
Training Epoch: 16/100, batch 226/563 completed (loss: 0.0858:  40%|████      | 113/281 [11:26<17:04,  6.10s/it]
Training Epoch: 16/100, batch 226/563 completed (loss: 0.0747:  40%|████      | 113/281 [11:26<17:05,  6.10s/it]
Training Epoch: 16/100, batch 226/563 completed (loss: 0.0747:  41%|████      | 114/281 [11:29<16:54,  6.08s/it]
Training Epoch: 16/100, batch 226/563 completed (loss: 0.0858:  41%|████      | 114/281 [11:29<16:54,  6.08s/it]
Training Epoch: 16/100, batch 226/563 completed (loss: 0.0238:  41%|████      | 114/281 [11:29<16:54,  6.08s/it]
Training Epoch: 16/100, batch 226/563 completed (loss: 0.0225:  41%|████      | 114/281 [11:29<16:54,  6.08s/it]
Training Epoch: 16/100, batch 227/563 completed (loss: 0.3046:  41%|████      | 114/281 [11:29<16:54,  6.08s/it]
Training Epoch: 16/100, batch 227/563 completed (loss: 0.0246:  41%|████      | 114/281 [11:29<16:54,  6.08s/it]
Training Epoch: 16/100, batch 227/563 completed (loss: 0.1413:  41%|████      | 114/281 [11:29<16:54,  6.08s/it]
Training Epoch: 16/100, batch 227/563 completed (loss: 0.132:  41%|████      | 114/281 [11:29<16:54,  6.08s/it] 
Training Epoch: 16/100, batch 228/563 completed (loss: 0.0301:  41%|████      | 114/281 [11:32<16:54,  6.08s/it]
Training Epoch: 16/100, batch 228/563 completed (loss: 0.029:  41%|████      | 114/281 [11:32<16:54,  6.08s/it] 
Training Epoch: 16/100, batch 228/563 completed (loss: 0.0095:  41%|████      | 114/281 [11:32<16:54,  6.08s/it]
Training Epoch: 16/100, batch 228/563 completed (loss: 0.0252:  41%|████      | 114/281 [11:32<16:54,  6.08s/it]
Training Epoch: 16/100, batch 228/563 completed (loss: 0.0252:  41%|████      | 115/281 [11:34<16:37,  6.01s/it]
Training Epoch: 16/100, batch 228/563 completed (loss: 0.029:  41%|████      | 115/281 [11:34<16:38,  6.01s/it]
Training Epoch: 16/100, batch 228/563 completed (loss: 0.0095:  41%|████      | 115/281 [11:34<16:38,  6.02s/it]
Training Epoch: 16/100, batch 228/563 completed (loss: 0.0301:  41%|████      | 115/281 [11:34<16:38,  6.02s/it]
Training Epoch: 16/100, batch 229/563 completed (loss: 0.1176:  41%|████      | 115/281 [11:34<16:38,  6.01s/it]
Training Epoch: 16/100, batch 229/563 completed (loss: 0.0347:  41%|████      | 115/281 [11:34<16:38,  6.02s/it]
Training Epoch: 16/100, batch 229/563 completed (loss: 0.0394:  41%|████      | 115/281 [11:35<16:38,  6.02s/it]
Training Epoch: 16/100, batch 229/563 completed (loss: 0.1362:  41%|████      | 115/281 [11:35<16:37,  6.01s/it]
Training Epoch: 16/100, batch 230/563 completed (loss: 0.0061:  41%|████      | 115/281 [11:38<16:38,  6.02s/it]
Training Epoch: 16/100, batch 230/563 completed (loss: 0.0756:  41%|████      | 115/281 [11:38<16:38,  6.02s/it]
Training Epoch: 16/100, batch 230/563 completed (loss: 0.0491:  41%|████      | 115/281 [11:38<16:38,  6.01s/it]
Training Epoch: 16/100, batch 230/563 completed (loss: 0.0307:  41%|████      | 115/281 [11:38<16:37,  6.01s/it]
Training Epoch: 16/100, batch 230/563 completed (loss: 0.0491:  41%|████▏     | 116/281 [11:41<16:49,  6.12s/it]
Training Epoch: 16/100, batch 230/563 completed (loss: 0.0756:  41%|████▏     | 116/281 [11:41<16:49,  6.12s/it]
Training Epoch: 16/100, batch 230/563 completed (loss: 0.0061:  41%|████▏     | 116/281 [11:41<16:49,  6.12s/it]
Training Epoch: 16/100, batch 230/563 completed (loss: 0.0307:  41%|████▏     | 116/281 [11:41<16:50,  6.13s/it]
Training Epoch: 16/100, batch 231/563 completed (loss: 0.0193:  41%|████▏     | 116/281 [11:41<16:49,  6.12s/it]
Training Epoch: 16/100, batch 231/563 completed (loss: 0.0039:  41%|████▏     | 116/281 [11:41<16:49,  6.12s/it]
Training Epoch: 16/100, batch 231/563 completed (loss: 0.0044:  41%|████▏     | 116/281 [11:41<16:49,  6.12s/it]
Training Epoch: 16/100, batch 231/563 completed (loss: 0.254:  41%|████▏     | 116/281 [11:41<16:50,  6.13s/it] 
Training Epoch: 16/100, batch 232/563 completed (loss: 0.0836:  41%|████▏     | 116/281 [11:44<16:49,  6.12s/it]
Training Epoch: 16/100, batch 232/563 completed (loss: 0.0096:  41%|████▏     | 116/281 [11:44<16:49,  6.12s/it]
Training Epoch: 16/100, batch 232/563 completed (loss: 0.1674:  41%|████▏     | 116/281 [11:44<16:49,  6.12s/it]
Training Epoch: 16/100, batch 232/563 completed (loss: 0.0677:  41%|████▏     | 116/281 [11:44<16:50,  6.13s/it]
Training Epoch: 16/100, batch 232/563 completed (loss: 0.0836:  42%|████▏     | 117/281 [11:47<16:31,  6.05s/it]
Training Epoch: 16/100, batch 232/563 completed (loss: 0.0677:  42%|████▏     | 117/281 [11:47<16:31,  6.05s/it]
Training Epoch: 16/100, batch 232/563 completed (loss: 0.1674:  42%|████▏     | 117/281 [11:47<16:31,  6.05s/it]
Training Epoch: 16/100, batch 232/563 completed (loss: 0.0096:  42%|████▏     | 117/281 [11:47<16:32,  6.05s/it]
Training Epoch: 16/100, batch 233/563 completed (loss: 0.0068:  42%|████▏     | 117/281 [11:47<16:32,  6.05s/it]
Training Epoch: 16/100, batch 233/563 completed (loss: 0.1593:  42%|████▏     | 117/281 [11:47<16:31,  6.05s/it]
Training Epoch: 16/100, batch 233/563 completed (loss: 0.398:  42%|████▏     | 117/281 [11:47<16:31,  6.05s/it] 
Training Epoch: 16/100, batch 233/563 completed (loss: 0.7097:  42%|████▏     | 117/281 [11:47<16:31,  6.05s/it]
Training Epoch: 16/100, batch 234/563 completed (loss: 0.0211:  42%|████▏     | 117/281 [11:50<16:31,  6.05s/it]
Training Epoch: 16/100, batch 234/563 completed (loss: 0.0386:  42%|████▏     | 117/281 [11:50<16:32,  6.05s/it]
Training Epoch: 16/100, batch 234/563 completed (loss: 0.1354:  42%|████▏     | 117/281 [11:50<16:31,  6.05s/it]
Training Epoch: 16/100, batch 234/563 completed (loss: 0.0164:  42%|████▏     | 117/281 [11:50<16:31,  6.05s/it]
Training Epoch: 16/100, batch 234/563 completed (loss: 0.0211:  42%|████▏     | 118/281 [11:53<16:32,  6.09s/it]
Training Epoch: 16/100, batch 234/563 completed (loss: 0.1354:  42%|████▏     | 118/281 [11:53<16:32,  6.09s/it]
Training Epoch: 16/100, batch 234/563 completed (loss: 0.0164:  42%|████▏     | 118/281 [11:53<16:33,  6.09s/it]
Training Epoch: 16/100, batch 234/563 completed (loss: 0.0386:  42%|████▏     | 118/281 [11:53<16:33,  6.09s/it]
Training Epoch: 16/100, batch 235/563 completed (loss: 0.0901:  42%|████▏     | 118/281 [11:53<16:32,  6.09s/it]
Training Epoch: 16/100, batch 235/563 completed (loss: 0.0025:  42%|████▏     | 118/281 [11:53<16:32,  6.09s/it]
Training Epoch: 16/100, batch 235/563 completed (loss: 0.113:  42%|████▏     | 118/281 [11:53<16:33,  6.09s/it] 
Training Epoch: 16/100, batch 235/563 completed (loss: 0.0147:  42%|████▏     | 118/281 [11:53<16:33,  6.09s/it]
Training Epoch: 16/100, batch 236/563 completed (loss: 0.0464:  42%|████▏     | 118/281 [11:56<16:32,  6.09s/it]
Training Epoch: 16/100, batch 236/563 completed (loss: 0.459:  42%|████▏     | 118/281 [11:56<16:33,  6.09s/it]
Training Epoch: 16/100, batch 236/563 completed (loss: 0.1758:  42%|████▏     | 118/281 [11:56<16:32,  6.09s/it]
Training Epoch: 16/100, batch 236/563 completed (loss: 0.0431:  42%|████▏     | 118/281 [11:56<16:33,  6.09s/it]
Training Epoch: 16/100, batch 236/563 completed (loss: 0.0464:  42%|████▏     | 119/281 [11:59<16:29,  6.11s/it]
Training Epoch: 16/100, batch 236/563 completed (loss: 0.459:  42%|████▏     | 119/281 [11:59<16:29,  6.11s/it]
Training Epoch: 16/100, batch 236/563 completed (loss: 0.0431:  42%|████▏     | 119/281 [11:59<16:30,  6.11s/it]
Training Epoch: 16/100, batch 236/563 completed (loss: 0.1758:  42%|████▏     | 119/281 [11:59<16:29,  6.11s/it]
Training Epoch: 16/100, batch 237/563 completed (loss: 0.0531:  42%|████▏     | 119/281 [11:59<16:29,  6.11s/it]
Training Epoch: 16/100, batch 237/563 completed (loss: 0.075:  42%|████▏     | 119/281 [11:59<16:29,  6.11s/it] 
Training Epoch: 16/100, batch 237/563 completed (loss: 0.1819:  42%|████▏     | 119/281 [11:59<16:29,  6.11s/it]
Training Epoch: 16/100, batch 237/563 completed (loss: 0.4647:  42%|████▏     | 119/281 [11:59<16:30,  6.11s/it]
Training Epoch: 16/100, batch 238/563 completed (loss: 0.0126:  42%|████▏     | 119/281 [12:02<16:29,  6.11s/it]
Training Epoch: 16/100, batch 238/563 completed (loss: 0.0068:  42%|████▏     | 119/281 [12:02<16:29,  6.11s/it]
Training Epoch: 16/100, batch 238/563 completed (loss: 0.0054:  42%|████▏     | 119/281 [12:02<16:29,  6.11s/it]
Training Epoch: 16/100, batch 238/563 completed (loss: 0.1467:  42%|████▏     | 119/281 [12:02<16:30,  6.11s/it]
Training Epoch: 16/100, batch 238/563 completed (loss: 0.0126:  43%|████▎     | 120/281 [12:05<16:11,  6.03s/it]
Training Epoch: 16/100, batch 238/563 completed (loss: 0.0068:  43%|████▎     | 120/281 [12:05<16:10,  6.03s/it]
Training Epoch: 16/100, batch 238/563 completed (loss: 0.1467:  43%|████▎     | 120/281 [12:05<16:11,  6.03s/it]
Training Epoch: 16/100, batch 238/563 completed (loss: 0.0054:  43%|████▎     | 120/281 [12:05<16:11,  6.03s/it]
Training Epoch: 16/100, batch 239/563 completed (loss: 0.0131:  43%|████▎     | 120/281 [12:05<16:10,  6.03s/it]
Training Epoch: 16/100, batch 239/563 completed (loss: 0.4162:  43%|████▎     | 120/281 [12:05<16:11,  6.03s/it]
Training Epoch: 16/100, batch 239/563 completed (loss: 0.0286:  43%|████▎     | 120/281 [12:05<16:11,  6.03s/it]
Training Epoch: 16/100, batch 239/563 completed (loss: 0.0275:  43%|████▎     | 120/281 [12:05<16:11,  6.03s/it]
Training Epoch: 16/100, batch 240/563 completed (loss: 0.0323:  43%|████▎     | 120/281 [12:08<16:11,  6.03s/it]
Training Epoch: 16/100, batch 240/563 completed (loss: 0.1598:  43%|████▎     | 120/281 [12:08<16:11,  6.03s/it]
Training Epoch: 16/100, batch 240/563 completed (loss: 0.0645:  43%|████▎     | 120/281 [12:08<16:10,  6.03s/it]
Training Epoch: 16/100, batch 240/563 completed (loss: 0.0111:  43%|████▎     | 120/281 [12:08<16:11,  6.03s/it]
Training Epoch: 16/100, batch 240/563 completed (loss: 0.0323:  43%|████▎     | 121/281 [12:11<16:18,  6.11s/it]
Training Epoch: 16/100, batch 240/563 completed (loss: 0.1598:  43%|████▎     | 121/281 [12:11<16:17,  6.11s/it]
Training Epoch: 16/100, batch 240/563 completed (loss: 0.0645:  43%|████▎     | 121/281 [12:11<16:17,  6.11s/it]
Training Epoch: 16/100, batch 240/563 completed (loss: 0.0111:  43%|████▎     | 121/281 [12:11<16:18,  6.12s/it]
Training Epoch: 16/100, batch 241/563 completed (loss: 0.0249:  43%|████▎     | 121/281 [12:11<16:18,  6.11s/it]
Training Epoch: 16/100, batch 241/563 completed (loss: 0.0076:  43%|████▎     | 121/281 [12:11<16:17,  6.11s/it]
Training Epoch: 16/100, batch 241/563 completed (loss: 0.4156:  43%|████▎     | 121/281 [12:11<16:17,  6.11s/it]
Training Epoch: 16/100, batch 241/563 completed (loss: 0.0791:  43%|████▎     | 121/281 [12:11<16:18,  6.12s/it]
Training Epoch: 16/100, batch 242/563 completed (loss: 0.0322:  43%|████▎     | 121/281 [12:14<16:18,  6.11s/it]
Training Epoch: 16/100, batch 242/563 completed (loss: 0.0399:  43%|████▎     | 121/281 [12:14<16:17,  6.11s/it]
Training Epoch: 16/100, batch 242/563 completed (loss: 0.0781:  43%|████▎     | 121/281 [12:14<16:17,  6.11s/it]
Training Epoch: 16/100, batch 242/563 completed (loss: 0.0758:  43%|████▎     | 121/281 [12:14<16:18,  6.12s/it]
Training Epoch: 16/100, batch 242/563 completed (loss: 0.0758:  43%|████▎     | 122/281 [12:17<16:08,  6.09s/it]
Training Epoch: 16/100, batch 242/563 completed (loss: 0.0322:  43%|████▎     | 122/281 [12:17<16:09,  6.09s/it]
Training Epoch: 16/100, batch 242/563 completed (loss: 0.0781:  43%|████▎     | 122/281 [12:17<16:09,  6.09s/it]
Training Epoch: 16/100, batch 242/563 completed (loss: 0.0399:  43%|████▎     | 122/281 [12:17<16:09,  6.10s/it]
Training Epoch: 16/100, batch 243/563 completed (loss: 0.1886:  43%|████▎     | 122/281 [12:17<16:09,  6.09s/it]
Training Epoch: 16/100, batch 243/563 completed (loss: 0.0095:  43%|████▎     | 122/281 [12:17<16:09,  6.10s/it]
Training Epoch: 16/100, batch 243/563 completed (loss: 0.0138:  43%|████▎     | 122/281 [12:17<16:09,  6.09s/it]
Training Epoch: 16/100, batch 243/563 completed (loss: 0.0101:  43%|████▎     | 122/281 [12:17<16:08,  6.09s/it]
Training Epoch: 16/100, batch 244/563 completed (loss: 0.3926:  43%|████▎     | 122/281 [12:20<16:09,  6.10s/it]
Training Epoch: 16/100, batch 244/563 completed (loss: 0.0138:  43%|████▎     | 122/281 [12:20<16:09,  6.09s/it]
Training Epoch: 16/100, batch 244/563 completed (loss: 0.01:  43%|████▎     | 122/281 [12:20<16:09,  6.09s/it]  
Training Epoch: 16/100, batch 244/563 completed (loss: 0.0327:  43%|████▎     | 122/281 [12:20<16:08,  6.09s/it]
Training Epoch: 16/100, batch 244/563 completed (loss: 0.0138:  44%|████▍     | 123/281 [12:23<15:52,  6.03s/it]
Training Epoch: 16/100, batch 244/563 completed (loss: 0.01:  44%|████▍     | 123/281 [12:23<15:52,  6.03s/it]
Training Epoch: 16/100, batch 244/563 completed (loss: 0.3926:  44%|████▍     | 123/281 [12:23<15:52,  6.03s/it]
Training Epoch: 16/100, batch 244/563 completed (loss: 0.0327:  44%|████▍     | 123/281 [12:23<15:52,  6.03s/it]
Training Epoch: 16/100, batch 245/563 completed (loss: 0.0696:  44%|████▍     | 123/281 [12:23<15:52,  6.03s/it]
Training Epoch: 16/100, batch 245/563 completed (loss: 0.2162:  44%|████▍     | 123/281 [12:23<15:52,  6.03s/it]
Training Epoch: 16/100, batch 245/563 completed (loss: 0.13:  44%|████▍     | 123/281 [12:23<15:52,  6.03s/it]  
Training Epoch: 16/100, batch 245/563 completed (loss: 0.0168:  44%|████▍     | 123/281 [12:23<15:52,  6.03s/it]
Training Epoch: 16/100, batch 246/563 completed (loss: 0.1544:  44%|████▍     | 123/281 [12:26<15:52,  6.03s/it]
Training Epoch: 16/100, batch 246/563 completed (loss: 0.023:  44%|████▍     | 123/281 [12:26<15:52,  6.03s/it]
Training Epoch: 16/100, batch 246/563 completed (loss: 0.2219:  44%|████▍     | 123/281 [12:26<15:52,  6.03s/it]
Training Epoch: 16/100, batch 246/563 completed (loss: 0.0115:  44%|████▍     | 123/281 [12:27<15:52,  6.03s/it]
Training Epoch: 16/100, batch 246/563 completed (loss: 0.2219:  44%|████▍     | 124/281 [12:29<15:56,  6.09s/it]
Training Epoch: 16/100, batch 246/563 completed (loss: 0.023:  44%|████▍     | 124/281 [12:29<15:56,  6.09s/it]
Training Epoch: 16/100, batch 246/563 completed (loss: 0.1544:  44%|████▍     | 124/281 [12:29<15:56,  6.09s/it]
Training Epoch: 16/100, batch 246/563 completed (loss: 0.0115:  44%|████▍     | 124/281 [12:29<15:56,  6.09s/it]
Training Epoch: 16/100, batch 247/563 completed (loss: 0.0247:  44%|████▍     | 124/281 [12:29<15:56,  6.09s/it]
Training Epoch: 16/100, batch 247/563 completed (loss: 0.0172:  44%|████▍     | 124/281 [12:29<15:56,  6.09s/it]
Training Epoch: 16/100, batch 247/563 completed (loss: 0.0615:  44%|████▍     | 124/281 [12:29<15:56,  6.09s/it]
Training Epoch: 16/100, batch 247/563 completed (loss: 0.2179:  44%|████▍     | 124/281 [12:29<15:56,  6.09s/it]
Training Epoch: 16/100, batch 248/563 completed (loss: 0.0284:  44%|████▍     | 124/281 [12:32<15:56,  6.09s/it]
Training Epoch: 16/100, batch 248/563 completed (loss: 0.0426:  44%|████▍     | 124/281 [12:32<15:56,  6.09s/it]
Training Epoch: 16/100, batch 248/563 completed (loss: 0.2474:  44%|████▍     | 124/281 [12:32<15:56,  6.09s/it]
Training Epoch: 16/100, batch 248/563 completed (loss: 0.0107:  44%|████▍     | 124/281 [12:33<15:56,  6.09s/it]
Training Epoch: 16/100, batch 248/563 completed (loss: 0.0107:  44%|████▍     | 125/281 [12:35<15:46,  6.07s/it]
Training Epoch: 16/100, batch 248/563 completed (loss: 0.0426:  44%|████▍     | 125/281 [12:35<15:46,  6.07s/it]
Training Epoch: 16/100, batch 248/563 completed (loss: 0.2474:  44%|████▍     | 125/281 [12:35<15:47,  6.07s/it]
Training Epoch: 16/100, batch 248/563 completed (loss: 0.0284:  44%|████▍     | 125/281 [12:35<15:47,  6.07s/it]
Training Epoch: 16/100, batch 249/563 completed (loss: 0.0386:  44%|████▍     | 125/281 [12:35<15:47,  6.07s/it]
Training Epoch: 16/100, batch 249/563 completed (loss: 0.009:  44%|████▍     | 125/281 [12:35<15:46,  6.07s/it] 
Training Epoch: 16/100, batch 249/563 completed (loss: 0.0236:  44%|████▍     | 125/281 [12:35<15:47,  6.07s/it]
Training Epoch: 16/100, batch 249/563 completed (loss: 0.0645:  44%|████▍     | 125/281 [12:35<15:46,  6.07s/it]
Training Epoch: 16/100, batch 250/563 completed (loss: 0.0041:  44%|████▍     | 125/281 [12:38<15:47,  6.07s/it]
Training Epoch: 16/100, batch 250/563 completed (loss: 0.0182:  44%|████▍     | 125/281 [12:38<15:47,  6.07s/it]
Training Epoch: 16/100, batch 250/563 completed (loss: 0.03:  44%|████▍     | 125/281 [12:38<15:46,  6.07s/it] 
Training Epoch: 16/100, batch 250/563 completed (loss: 0.0542:  44%|████▍     | 125/281 [12:38<15:46,  6.07s/it]
Training Epoch: 16/100, batch 250/563 completed (loss: 0.0542:  45%|████▍     | 126/281 [12:42<15:41,  6.07s/it]
Training Epoch: 16/100, batch 250/563 completed (loss: 0.0182:  45%|████▍     | 126/281 [12:41<15:41,  6.08s/it]
Training Epoch: 16/100, batch 250/563 completed (loss: 0.03:  45%|████▍     | 126/281 [12:41<15:41,  6.08s/it]
Training Epoch: 16/100, batch 250/563 completed (loss: 0.0041:  45%|████▍     | 126/281 [12:41<15:41,  6.08s/it]
Training Epoch: 16/100, batch 251/563 completed (loss: 0.2564:  45%|████▍     | 126/281 [12:42<15:41,  6.08s/it]
Training Epoch: 16/100, batch 251/563 completed (loss: 0.1264:  45%|████▍     | 126/281 [12:42<15:41,  6.08s/it]
Training Epoch: 16/100, batch 251/563 completed (loss: 0.0035:  45%|████▍     | 126/281 [12:42<15:41,  6.08s/it]
Training Epoch: 16/100, batch 251/563 completed (loss: 0.0551:  45%|████▍     | 126/281 [12:42<15:41,  6.07s/it]
Training Epoch: 16/100, batch 252/563 completed (loss: 0.0304:  45%|████▍     | 126/281 [12:45<15:41,  6.08s/it]
Training Epoch: 16/100, batch 252/563 completed (loss: 0.0095:  45%|████▍     | 126/281 [12:45<15:41,  6.08s/it]
Training Epoch: 16/100, batch 252/563 completed (loss: 0.0635:  45%|████▍     | 126/281 [12:45<15:41,  6.08s/it]
Training Epoch: 16/100, batch 252/563 completed (loss: 0.082:  45%|████▍     | 126/281 [12:45<15:41,  6.07s/it] 
Training Epoch: 16/100, batch 252/563 completed (loss: 0.0635:  45%|████▌     | 127/281 [12:48<15:40,  6.11s/it]
Training Epoch: 16/100, batch 252/563 completed (loss: 0.0095:  45%|████▌     | 127/281 [12:48<15:40,  6.11s/it]
Training Epoch: 16/100, batch 252/563 completed (loss: 0.0304:  45%|████▌     | 127/281 [12:48<15:40,  6.11s/it]
Training Epoch: 16/100, batch 252/563 completed (loss: 0.082:  45%|████▌     | 127/281 [12:48<15:40,  6.11s/it]
Training Epoch: 16/100, batch 253/563 completed (loss: 0.0052:  45%|████▌     | 127/281 [12:48<15:40,  6.11s/it]
Training Epoch: 16/100, batch 253/563 completed (loss: 0.06:  45%|████▌     | 127/281 [12:48<15:40,  6.11s/it]  
Training Epoch: 16/100, batch 253/563 completed (loss: 0.0229:  45%|████▌     | 127/281 [12:48<15:40,  6.11s/it]
Training Epoch: 16/100, batch 253/563 completed (loss: 0.042:  45%|████▌     | 127/281 [12:48<15:40,  6.11s/it]
Training Epoch: 16/100, batch 254/563 completed (loss: 0.0466:  45%|████▌     | 127/281 [12:51<15:40,  6.11s/it]
Training Epoch: 16/100, batch 254/563 completed (loss: 0.0091:  45%|████▌     | 127/281 [12:51<15:40,  6.11s/it]
Training Epoch: 16/100, batch 254/563 completed (loss: 0.0201:  45%|████▌     | 127/281 [12:51<15:40,  6.11s/it]
Training Epoch: 16/100, batch 254/563 completed (loss: 0.0236:  45%|████▌     | 127/281 [12:51<15:40,  6.11s/it]
Training Epoch: 16/100, batch 254/563 completed (loss: 0.0236:  46%|████▌     | 128/281 [12:54<15:22,  6.03s/it]
Training Epoch: 16/100, batch 254/563 completed (loss: 0.0091:  46%|████▌     | 128/281 [12:54<15:23,  6.04s/it]
Training Epoch: 16/100, batch 254/563 completed (loss: 0.0466:  46%|████▌     | 128/281 [12:54<15:23,  6.04s/it]
Training Epoch: 16/100, batch 254/563 completed (loss: 0.0201:  46%|████▌     | 128/281 [12:54<15:24,  6.04s/it]
Training Epoch: 16/100, batch 255/563 completed (loss: 0.0168:  46%|████▌     | 128/281 [12:54<15:24,  6.04s/it]
Training Epoch: 16/100, batch 255/563 completed (loss: 0.7315:  46%|████▌     | 128/281 [12:54<15:23,  6.04s/it]
Training Epoch: 16/100, batch 255/563 completed (loss: 0.109:  46%|████▌     | 128/281 [12:54<15:23,  6.04s/it] 
Training Epoch: 16/100, batch 255/563 completed (loss: 0.038:  46%|████▌     | 128/281 [12:54<15:22,  6.03s/it] 
Training Epoch: 16/100, batch 256/563 completed (loss: 0.3315:  46%|████▌     | 128/281 [12:57<15:23,  6.04s/it]
Training Epoch: 16/100, batch 256/563 completed (loss: 0.0119:  46%|████▌     | 128/281 [12:57<15:24,  6.04s/it]
Training Epoch: 16/100, batch 256/563 completed (loss: 0.0266:  46%|████▌     | 128/281 [12:57<15:23,  6.04s/it]
Training Epoch: 16/100, batch 256/563 completed (loss: 0.3799:  46%|████▌     | 128/281 [12:57<15:22,  6.03s/it]
Training Epoch: 16/100, batch 256/563 completed (loss: 0.3315:  46%|████▌     | 129/281 [13:00<15:29,  6.12s/it]
Training Epoch: 16/100, batch 256/563 completed (loss: 0.0266:  46%|████▌     | 129/281 [13:00<15:29,  6.12s/it]
Training Epoch: 16/100, batch 256/563 completed (loss: 0.0119:  46%|████▌     | 129/281 [13:00<15:29,  6.11s/it]
Training Epoch: 16/100, batch 256/563 completed (loss: 0.3799:  46%|████▌     | 129/281 [13:00<15:30,  6.12s/it]
Training Epoch: 16/100, batch 257/563 completed (loss: 0.0216:  46%|████▌     | 129/281 [13:00<15:29,  6.11s/it]
Training Epoch: 16/100, batch 257/563 completed (loss: 0.0998:  46%|████▌     | 129/281 [13:00<15:29,  6.12s/it]
Training Epoch: 16/100, batch 257/563 completed (loss: 0.023:  46%|████▌     | 129/281 [13:00<15:29,  6.12s/it] 
Training Epoch: 16/100, batch 257/563 completed (loss: 0.0463:  46%|████▌     | 129/281 [13:00<15:30,  6.12s/it]
Training Epoch: 16/100, batch 258/563 completed (loss: 0.0491:  46%|████▌     | 129/281 [13:03<15:29,  6.11s/it]
Training Epoch: 16/100, batch 258/563 completed (loss: 0.0286:  46%|████▌     | 129/281 [13:03<15:29,  6.12s/it]
Training Epoch: 16/100, batch 258/563 completed (loss: 0.014:  46%|████▌     | 129/281 [13:03<15:29,  6.12s/it] 
Training Epoch: 16/100, batch 258/563 completed (loss: 0.0195:  46%|████▌     | 129/281 [13:03<15:30,  6.12s/it]
Training Epoch: 16/100, batch 258/563 completed (loss: 0.0286:  46%|████▋     | 130/281 [13:06<15:18,  6.08s/it]
Training Epoch: 16/100, batch 258/563 completed (loss: 0.0491:  46%|████▋     | 130/281 [13:06<15:18,  6.08s/it]
Training Epoch: 16/100, batch 258/563 completed (loss: 0.014:  46%|████▋     | 130/281 [13:06<15:18,  6.09s/it]
Training Epoch: 16/100, batch 258/563 completed (loss: 0.0195:  46%|████▋     | 130/281 [13:06<15:19,  6.09s/it]
Training Epoch: 16/100, batch 259/563 completed (loss: 0.0131:  46%|████▋     | 130/281 [13:06<15:18,  6.09s/it]
Training Epoch: 16/100, batch 259/563 completed (loss: 0.2499:  46%|████▋     | 130/281 [13:06<15:18,  6.08s/it]
Training Epoch: 16/100, batch 259/563 completed (loss: 0.0236:  46%|████▋     | 130/281 [13:06<15:18,  6.08s/it]
Training Epoch: 16/100, batch 259/563 completed (loss: 0.2447:  46%|████▋     | 130/281 [13:06<15:19,  6.09s/it]
Training Epoch: 16/100, batch 260/563 completed (loss: 0.0107:  46%|████▋     | 130/281 [13:09<15:18,  6.08s/it]
Training Epoch: 16/100, batch 260/563 completed (loss: 0.0237:  46%|████▋     | 130/281 [13:09<15:18,  6.08s/it]
Training Epoch: 16/100, batch 260/563 completed (loss: 0.1675:  46%|████▋     | 130/281 [13:09<15:18,  6.09s/it]
Training Epoch: 16/100, batch 260/563 completed (loss: 0.2313:  46%|████▋     | 130/281 [13:09<15:19,  6.09s/it]
Training Epoch: 16/100, batch 260/563 completed (loss: 0.1675:  47%|████▋     | 131/281 [13:12<15:16,  6.11s/it]
Training Epoch: 16/100, batch 260/563 completed (loss: 0.0107:  47%|████▋     | 131/281 [13:12<15:16,  6.11s/it]
Training Epoch: 16/100, batch 260/563 completed (loss: 0.0237:  47%|████▋     | 131/281 [13:12<15:17,  6.11s/it]
Training Epoch: 16/100, batch 260/563 completed (loss: 0.2313:  47%|████▋     | 131/281 [13:12<15:17,  6.11s/it]
Training Epoch: 16/100, batch 261/563 completed (loss: 0.1135:  47%|████▋     | 131/281 [13:12<15:16,  6.11s/it]
Training Epoch: 16/100, batch 261/563 completed (loss: 0.0813:  47%|████▋     | 131/281 [13:12<15:17,  6.11s/it]
Training Epoch: 16/100, batch 261/563 completed (loss: 0.0358:  47%|████▋     | 131/281 [13:12<15:16,  6.11s/it]
Training Epoch: 16/100, batch 261/563 completed (loss: 0.0028:  47%|████▋     | 131/281 [13:12<15:17,  6.11s/it]
Training Epoch: 16/100, batch 262/563 completed (loss: 0.0163:  47%|████▋     | 131/281 [13:15<15:16,  6.11s/it]
Training Epoch: 16/100, batch 262/563 completed (loss: 0.042:  47%|████▋     | 131/281 [13:15<15:16,  6.11s/it] 
Training Epoch: 16/100, batch 262/563 completed (loss: 0.0665:  47%|████▋     | 131/281 [13:15<15:17,  6.11s/it]
Training Epoch: 16/100, batch 262/563 completed (loss: 0.0806:  47%|████▋     | 131/281 [13:15<15:17,  6.11s/it]
Training Epoch: 16/100, batch 262/563 completed (loss: 0.0665:  47%|████▋     | 132/281 [13:18<15:06,  6.08s/it]
Training Epoch: 16/100, batch 262/563 completed (loss: 0.0806:  47%|████▋     | 132/281 [13:18<15:06,  6.08s/it]
Training Epoch: 16/100, batch 262/563 completed (loss: 0.0163:  47%|████▋     | 132/281 [13:18<15:06,  6.08s/it]
Training Epoch: 16/100, batch 262/563 completed (loss: 0.042:  47%|████▋     | 132/281 [13:18<15:06,  6.08s/it]
Training Epoch: 16/100, batch 263/563 completed (loss: 0.017:  47%|████▋     | 132/281 [13:18<15:06,  6.08s/it] 
Training Epoch: 16/100, batch 263/563 completed (loss: 0.0859:  47%|████▋     | 132/281 [13:18<15:06,  6.08s/it]
Training Epoch: 16/100, batch 263/563 completed (loss: 0.0132:  47%|████▋     | 132/281 [13:18<15:06,  6.08s/it]
Training Epoch: 16/100, batch 263/563 completed (loss: 0.066:  47%|████▋     | 132/281 [13:18<15:06,  6.08s/it] 
Training Epoch: 16/100, batch 264/563 completed (loss: 0.0182:  47%|████▋     | 132/281 [13:21<15:06,  6.08s/it]
Training Epoch: 16/100, batch 264/563 completed (loss: 0.0157:  47%|████▋     | 132/281 [13:21<15:06,  6.08s/it]
Training Epoch: 16/100, batch 264/563 completed (loss: 0.056:  47%|████▋     | 132/281 [13:21<15:06,  6.08s/it]
Training Epoch: 16/100, batch 264/563 completed (loss: 0.0112:  47%|████▋     | 132/281 [13:21<15:06,  6.08s/it]
Training Epoch: 16/100, batch 264/563 completed (loss: 0.0112:  47%|████▋     | 133/281 [13:24<14:57,  6.06s/it]
Training Epoch: 16/100, batch 264/563 completed (loss: 0.056:  47%|████▋     | 133/281 [13:24<14:58,  6.07s/it]
Training Epoch: 16/100, batch 264/563 completed (loss: 0.0157:  47%|████▋     | 133/281 [13:24<14:58,  6.07s/it]
Training Epoch: 16/100, batch 264/563 completed (loss: 0.0182:  47%|████▋     | 133/281 [13:24<14:59,  6.07s/it]
Training Epoch: 16/100, batch 265/563 completed (loss: 0.028:  47%|████▋     | 133/281 [13:24<14:58,  6.07s/it] 
Training Epoch: 16/100, batch 265/563 completed (loss: 0.0476:  47%|████▋     | 133/281 [13:24<14:59,  6.07s/it]
Training Epoch: 16/100, batch 265/563 completed (loss: 0.0303:  47%|████▋     | 133/281 [13:24<14:58,  6.07s/it]
Training Epoch: 16/100, batch 265/563 completed (loss: 0.1276:  47%|████▋     | 133/281 [13:24<14:57,  6.06s/it]
Training Epoch: 16/100, batch 266/563 completed (loss: 0.0393:  47%|████▋     | 133/281 [13:27<14:58,  6.07s/it]
Training Epoch: 16/100, batch 266/563 completed (loss: 0.0061:  47%|████▋     | 133/281 [13:27<14:58,  6.07s/it]
Training Epoch: 16/100, batch 266/563 completed (loss: 0.2864:  47%|████▋     | 133/281 [13:27<14:59,  6.07s/it]
Training Epoch: 16/100, batch 266/563 completed (loss: 0.0257:  47%|████▋     | 133/281 [13:27<14:57,  6.06s/it]
Training Epoch: 16/100, batch 266/563 completed (loss: 0.0061:  48%|████▊     | 134/281 [13:30<15:01,  6.13s/it]
Training Epoch: 16/100, batch 266/563 completed (loss: 0.0393:  48%|████▊     | 134/281 [13:30<15:01,  6.13s/it]
Training Epoch: 16/100, batch 266/563 completed (loss: 0.0257:  48%|████▊     | 134/281 [13:30<15:01,  6.13s/it]
Training Epoch: 16/100, batch 266/563 completed (loss: 0.2864:  48%|████▊     | 134/281 [13:30<15:01,  6.13s/it]
Training Epoch: 16/100, batch 267/563 completed (loss: 0.3664:  48%|████▊     | 134/281 [13:30<15:01,  6.13s/it]
Training Epoch: 16/100, batch 267/563 completed (loss: 0.0247:  48%|████▊     | 134/281 [13:30<15:01,  6.13s/it]
Training Epoch: 16/100, batch 267/563 completed (loss: 0.0914:  48%|████▊     | 134/281 [13:30<15:01,  6.13s/it]
Training Epoch: 16/100, batch 267/563 completed (loss: 0.0283:  48%|████▊     | 134/281 [13:30<15:01,  6.13s/it]
Training Epoch: 16/100, batch 268/563 completed (loss: 0.015:  48%|████▊     | 134/281 [13:33<15:01,  6.13s/it] 
Training Epoch: 16/100, batch 268/563 completed (loss: 0.0149:  48%|████▊     | 134/281 [13:33<15:01,  6.13s/it]
Training Epoch: 16/100, batch 268/563 completed (loss: 0.0086:  48%|████▊     | 134/281 [13:33<15:01,  6.13s/it]
Training Epoch: 16/100, batch 268/563 completed (loss: 0.0059:  48%|████▊     | 134/281 [13:33<15:01,  6.13s/it]
Training Epoch: 16/100, batch 268/563 completed (loss: 0.0086:  48%|████▊     | 135/281 [13:36<14:49,  6.09s/it]
Training Epoch: 16/100, batch 268/563 completed (loss: 0.0059:  48%|████▊     | 135/281 [13:36<14:49,  6.09s/it]
Training Epoch: 16/100, batch 268/563 completed (loss: 0.015:  48%|████▊     | 135/281 [13:36<14:49,  6.09s/it]
Training Epoch: 16/100, batch 268/563 completed (loss: 0.0149:  48%|████▊     | 135/281 [13:36<14:50,  6.10s/it]
Training Epoch: 16/100, batch 269/563 completed (loss: 0.0534:  48%|████▊     | 135/281 [13:36<14:50,  6.10s/it]
Training Epoch: 16/100, batch 269/563 completed (loss: 0.0196:  48%|████▊     | 135/281 [13:36<14:49,  6.09s/it]
Training Epoch: 16/100, batch 269/563 completed (loss: 0.0398:  48%|████▊     | 135/281 [13:36<14:49,  6.09s/it]
Training Epoch: 16/100, batch 269/563 completed (loss: 0.1089:  48%|████▊     | 135/281 [13:36<14:49,  6.09s/it]
Training Epoch: 16/100, batch 270/563 completed (loss: 0.0126:  48%|████▊     | 135/281 [13:39<14:50,  6.10s/it]
Training Epoch: 16/100, batch 270/563 completed (loss: 0.0116:  48%|████▊     | 135/281 [13:39<14:49,  6.09s/it]
Training Epoch: 16/100, batch 270/563 completed (loss: 0.2078:  48%|████▊     | 135/281 [13:39<14:49,  6.09s/it]
Training Epoch: 16/100, batch 270/563 completed (loss: 0.1632:  48%|████▊     | 135/281 [13:39<14:49,  6.09s/it]
Training Epoch: 16/100, batch 270/563 completed (loss: 0.1632:  48%|████▊     | 136/281 [13:43<14:45,  6.11s/it]
Training Epoch: 16/100, batch 270/563 completed (loss: 0.0116:  48%|████▊     | 136/281 [13:42<14:46,  6.11s/it]
Training Epoch: 16/100, batch 270/563 completed (loss: 0.0126:  48%|████▊     | 136/281 [13:42<14:46,  6.11s/it]
Training Epoch: 16/100, batch 270/563 completed (loss: 0.2078:  48%|████▊     | 136/281 [13:43<14:45,  6.11s/it]
Training Epoch: 16/100, batch 271/563 completed (loss: 0.2789:  48%|████▊     | 136/281 [13:43<14:46,  6.11s/it]
Training Epoch: 16/100, batch 271/563 completed (loss: 0.1057:  48%|████▊     | 136/281 [13:43<14:45,  6.11s/it]
Training Epoch: 16/100, batch 271/563 completed (loss: 0.1365:  48%|████▊     | 136/281 [13:43<14:46,  6.11s/it]
Training Epoch: 16/100, batch 271/563 completed (loss: 0.03:  48%|████▊     | 136/281 [13:43<14:45,  6.11s/it]  
Training Epoch: 16/100, batch 272/563 completed (loss: 0.006:  48%|████▊     | 136/281 [13:46<14:45,  6.11s/it] 
Training Epoch: 16/100, batch 272/563 completed (loss: 0.0344:  48%|████▊     | 136/281 [13:46<14:46,  6.11s/it]
Training Epoch: 16/100, batch 272/563 completed (loss: 0.1282:  48%|████▊     | 136/281 [13:46<14:46,  6.11s/it]
Training Epoch: 16/100, batch 272/563 completed (loss: 0.063:  48%|████▊     | 136/281 [13:46<14:45,  6.11s/it]
Training Epoch: 16/100, batch 272/563 completed (loss: 0.0344:  49%|████▉     | 137/281 [13:48<14:35,  6.08s/it]
Training Epoch: 16/100, batch 272/563 completed (loss: 0.006:  49%|████▉     | 137/281 [13:49<14:35,  6.08s/it]
Training Epoch: 16/100, batch 272/563 completed (loss: 0.1282:  49%|████▉     | 137/281 [13:49<14:35,  6.08s/it]
Training Epoch: 16/100, batch 272/563 completed (loss: 0.063:  49%|████▉     | 137/281 [13:49<14:35,  6.08s/it]
Training Epoch: 16/100, batch 273/563 completed (loss: 0.1659:  49%|████▉     | 137/281 [13:49<14:35,  6.08s/it]
Training Epoch: 16/100, batch 273/563 completed (loss: 0.1816:  49%|████▉     | 137/281 [13:49<14:35,  6.08s/it]
Training Epoch: 16/100, batch 273/563 completed (loss: 0.0071:  49%|████▉     | 137/281 [13:49<14:35,  6.08s/it]
Training Epoch: 16/100, batch 273/563 completed (loss: 0.0291:  49%|████▉     | 137/281 [13:49<14:35,  6.08s/it]
Training Epoch: 16/100, batch 274/563 completed (loss: 0.1186:  49%|████▉     | 137/281 [13:52<14:35,  6.08s/it]
Training Epoch: 16/100, batch 274/563 completed (loss: 0.0629:  49%|████▉     | 137/281 [13:52<14:35,  6.08s/it]
Training Epoch: 16/100, batch 274/563 completed (loss: 0.0339:  49%|████▉     | 137/281 [13:52<14:35,  6.08s/it]
Training Epoch: 16/100, batch 274/563 completed (loss: 0.0286:  49%|████▉     | 137/281 [13:52<14:35,  6.08s/it]
Training Epoch: 16/100, batch 274/563 completed (loss: 0.0286:  49%|████▉     | 138/281 [13:55<14:26,  6.06s/it]
Training Epoch: 16/100, batch 274/563 completed (loss: 0.0629:  49%|████▉     | 138/281 [13:55<14:27,  6.06s/it]
Training Epoch: 16/100, batch 274/563 completed (loss: 0.1186:  49%|████▉     | 138/281 [13:55<14:26,  6.06s/it]
Training Epoch: 16/100, batch 274/563 completed (loss: 0.0339:  49%|████▉     | 138/281 [13:55<14:27,  6.07s/it]
Training Epoch: 16/100, batch 275/563 completed (loss: 0.0253:  49%|████▉     | 138/281 [13:55<14:27,  6.06s/it]
Training Epoch: 16/100, batch 275/563 completed (loss: 0.0088:  49%|████▉     | 138/281 [13:55<14:27,  6.07s/it]
Training Epoch: 16/100, batch 275/563 completed (loss: 0.0888:  49%|████▉     | 138/281 [13:55<14:26,  6.06s/it]
Training Epoch: 16/100, batch 275/563 completed (loss: 0.118:  49%|████▉     | 138/281 [13:55<14:26,  6.06s/it] 
Training Epoch: 16/100, batch 276/563 completed (loss: 0.037:  49%|████▉     | 138/281 [13:58<14:26,  6.06s/it] 
Training Epoch: 16/100, batch 276/563 completed (loss: 0.053:  49%|████▉     | 138/281 [13:58<14:27,  6.07s/it] 
Training Epoch: 16/100, batch 276/563 completed (loss: 0.115:  49%|████▉     | 138/281 [13:58<14:27,  6.06s/it] 
Training Epoch: 16/100, batch 276/563 completed (loss: 0.0221:  49%|████▉     | 138/281 [13:58<14:26,  6.06s/it]
Training Epoch: 16/100, batch 276/563 completed (loss: 0.0221:  49%|████▉     | 139/281 [14:01<14:25,  6.10s/it]
Training Epoch: 16/100, batch 276/563 completed (loss: 0.053:  49%|████▉     | 139/281 [14:01<14:25,  6.09s/it]
Training Epoch: 16/100, batch 276/563 completed (loss: 0.115:  49%|████▉     | 139/281 [14:01<14:25,  6.10s/it]
Training Epoch: 16/100, batch 276/563 completed (loss: 0.037:  49%|████▉     | 139/281 [14:01<14:25,  6.10s/it]
Training Epoch: 16/100, batch 277/563 completed (loss: 0.1522:  49%|████▉     | 139/281 [14:01<14:25,  6.10s/it]
Training Epoch: 16/100, batch 277/563 completed (loss: 0.0264:  49%|████▉     | 139/281 [14:01<14:25,  6.10s/it]
Training Epoch: 16/100, batch 277/563 completed (loss: 0.2507:  49%|████▉     | 139/281 [14:01<14:25,  6.09s/it]
Training Epoch: 16/100, batch 277/563 completed (loss: 0.0063:  49%|████▉     | 139/281 [14:01<14:25,  6.10s/it]
Training Epoch: 16/100, batch 278/563 completed (loss: 0.0315:  49%|████▉     | 139/281 [14:04<14:25,  6.09s/it]
Training Epoch: 16/100, batch 278/563 completed (loss: 0.1555:  49%|████▉     | 139/281 [14:04<14:25,  6.10s/it]
Training Epoch: 16/100, batch 278/563 completed (loss: 0.3396:  49%|████▉     | 139/281 [14:04<14:25,  6.10s/it]
Training Epoch: 16/100, batch 278/563 completed (loss: 0.4049:  49%|████▉     | 139/281 [14:04<14:25,  6.10s/it]
Training Epoch: 16/100, batch 278/563 completed (loss: 0.4049:  50%|████▉     | 140/281 [14:07<14:15,  6.07s/it]
Training Epoch: 16/100, batch 278/563 completed (loss: 0.0315:  50%|████▉     | 140/281 [14:07<14:15,  6.07s/it]
Training Epoch: 16/100, batch 278/563 completed (loss: 0.3396:  50%|████▉     | 140/281 [14:07<14:15,  6.07s/it]
Training Epoch: 16/100, batch 278/563 completed (loss: 0.1555:  50%|████▉     | 140/281 [14:07<14:15,  6.07s/it]
Training Epoch: 16/100, batch 279/563 completed (loss: 0.1007:  50%|████▉     | 140/281 [14:07<14:15,  6.07s/it]
Training Epoch: 16/100, batch 279/563 completed (loss: 0.189:  50%|████▉     | 140/281 [14:07<14:15,  6.07s/it] 
Training Epoch: 16/100, batch 279/563 completed (loss: 0.0645:  50%|████▉     | 140/281 [14:07<14:15,  6.07s/it]
Training Epoch: 16/100, batch 279/563 completed (loss: 0.0154:  50%|████▉     | 140/281 [14:07<14:15,  6.07s/it]
Training Epoch: 16/100, batch 280/563 completed (loss: 0.0966:  50%|████▉     | 140/281 [14:10<14:15,  6.07s/it]
Training Epoch: 16/100, batch 280/563 completed (loss: 0.4742:  50%|████▉     | 140/281 [14:10<14:15,  6.07s/it]
Training Epoch: 16/100, batch 280/563 completed (loss: 0.04:  50%|████▉     | 140/281 [14:10<14:15,  6.07s/it]  
Training Epoch: 16/100, batch 280/563 completed (loss: 0.2539:  50%|████▉     | 140/281 [14:10<14:15,  6.07s/it]
Training Epoch: 16/100, batch 280/563 completed (loss: 0.0966:  50%|█████     | 141/281 [14:13<14:07,  6.06s/it]
Training Epoch: 16/100, batch 280/563 completed (loss: 0.4742:  50%|█████     | 141/281 [14:13<14:07,  6.05s/it]
Training Epoch: 16/100, batch 280/563 completed (loss: 0.04:  50%|█████     | 141/281 [14:13<14:07,  6.06s/it]
Training Epoch: 16/100, batch 280/563 completed (loss: 0.2539:  50%|█████     | 141/281 [14:13<14:08,  6.06s/it]
Training Epoch: 16/100, batch 281/563 completed (loss: 0.0138:  50%|█████     | 141/281 [14:13<14:07,  6.06s/it]
Training Epoch: 16/100, batch 281/563 completed (loss: 0.0242:  50%|█████     | 141/281 [14:13<14:07,  6.05s/it]
Training Epoch: 16/100, batch 281/563 completed (loss: 0.0242:  50%|█████     | 141/281 [14:13<14:07,  6.06s/it]
Training Epoch: 16/100, batch 281/563 completed (loss: 0.1282:  50%|█████     | 141/281 [14:13<14:08,  6.06s/it]
Training Epoch: 16/100, batch 282/563 completed (loss: 0.0262:  50%|█████     | 141/281 [14:16<14:07,  6.05s/it]
Training Epoch: 16/100, batch 282/563 completed (loss: 0.0985:  50%|█████     | 141/281 [14:16<14:07,  6.06s/it]
Training Epoch: 16/100, batch 282/563 completed (loss: 0.0122:  50%|█████     | 141/281 [14:16<14:07,  6.06s/it]
Training Epoch: 16/100, batch 282/563 completed (loss: 0.0796:  50%|█████     | 141/281 [14:16<14:08,  6.06s/it]
Training Epoch: 16/100, batch 282/563 completed (loss: 0.0796:  51%|█████     | 142/281 [14:19<14:11,  6.13s/it]
Training Epoch: 16/100, batch 282/563 completed (loss: 0.0122:  51%|█████     | 142/281 [14:19<14:11,  6.13s/it]
Training Epoch: 16/100, batch 282/563 completed (loss: 0.0262:  51%|█████     | 142/281 [14:19<14:11,  6.13s/it]
Training Epoch: 16/100, batch 282/563 completed (loss: 0.0985:  51%|█████     | 142/281 [14:19<14:11,  6.13s/it]
Training Epoch: 16/100, batch 283/563 completed (loss: 0.0539:  51%|█████     | 142/281 [14:19<14:11,  6.13s/it]
Training Epoch: 16/100, batch 283/563 completed (loss: 0.0906:  51%|█████     | 142/281 [14:19<14:11,  6.13s/it]
Training Epoch: 16/100, batch 283/563 completed (loss: 0.0253:  51%|█████     | 142/281 [14:19<14:11,  6.13s/it]
Training Epoch: 16/100, batch 283/563 completed (loss: 0.0612:  51%|█████     | 142/281 [14:19<14:11,  6.13s/it]
Training Epoch: 16/100, batch 284/563 completed (loss: 0.0332:  51%|█████     | 142/281 [14:22<14:11,  6.13s/it]
Training Epoch: 16/100, batch 284/563 completed (loss: 0.082:  51%|█████     | 142/281 [14:22<14:11,  6.13s/it] 
Training Epoch: 16/100, batch 284/563 completed (loss: 0.2384:  51%|█████     | 142/281 [14:22<14:11,  6.13s/it]
Training Epoch: 16/100, batch 284/563 completed (loss: 0.0343:  51%|█████     | 142/281 [14:22<14:11,  6.13s/it]
Training Epoch: 16/100, batch 284/563 completed (loss: 0.0332:  51%|█████     | 143/281 [14:25<14:01,  6.10s/it]
Training Epoch: 16/100, batch 284/563 completed (loss: 0.082:  51%|█████     | 143/281 [14:25<14:01,  6.10s/it]
Training Epoch: 16/100, batch 284/563 completed (loss: 0.0343:  51%|█████     | 143/281 [14:25<14:01,  6.10s/it]
Training Epoch: 16/100, batch 284/563 completed (loss: 0.2384:  51%|█████     | 143/281 [14:25<14:01,  6.10s/it]
Training Epoch: 16/100, batch 285/563 completed (loss: 0.0394:  51%|█████     | 143/281 [14:25<14:01,  6.10s/it]
Training Epoch: 16/100, batch 285/563 completed (loss: 0.0236:  51%|█████     | 143/281 [14:25<14:01,  6.10s/it]
Training Epoch: 16/100, batch 285/563 completed (loss: 0.1407:  51%|█████     | 143/281 [14:25<14:01,  6.10s/it]
Training Epoch: 16/100, batch 285/563 completed (loss: 0.0197:  51%|█████     | 143/281 [14:25<14:01,  6.10s/it]
Training Epoch: 16/100, batch 286/563 completed (loss: 0.053:  51%|█████     | 143/281 [14:28<14:01,  6.10s/it] 
Training Epoch: 16/100, batch 286/563 completed (loss: 0.0112:  51%|█████     | 143/281 [14:28<14:01,  6.10s/it]
Training Epoch: 16/100, batch 286/563 completed (loss: 0.003:  51%|█████     | 143/281 [14:28<14:01,  6.10s/it] 
Training Epoch: 16/100, batch 286/563 completed (loss: 0.0662:  51%|█████     | 143/281 [14:28<14:01,  6.10s/it]
Training Epoch: 16/100, batch 286/563 completed (loss: 0.0662:  51%|█████     | 144/281 [14:31<13:58,  6.12s/it]
Training Epoch: 16/100, batch 286/563 completed (loss: 0.003:  51%|█████     | 144/281 [14:31<13:58,  6.12s/it]
Training Epoch: 16/100, batch 286/563 completed (loss: 0.0112:  51%|█████     | 144/281 [14:31<13:58,  6.12s/it]
Training Epoch: 16/100, batch 286/563 completed (loss: 0.053:  51%|█████     | 144/281 [14:31<13:58,  6.12s/it]
Training Epoch: 16/100, batch 287/563 completed (loss: 0.0404:  51%|█████     | 144/281 [14:31<13:58,  6.12s/it]
Training Epoch: 16/100, batch 287/563 completed (loss: 0.0988:  51%|█████     | 144/281 [14:31<13:58,  6.12s/it]
Training Epoch: 16/100, batch 287/563 completed (loss: 0.063:  51%|█████     | 144/281 [14:31<13:58,  6.12s/it]
Training Epoch: 16/100, batch 287/563 completed (loss: 0.0333:  51%|█████     | 144/281 [14:31<13:58,  6.12s/it]
Training Epoch: 16/100, batch 288/563 completed (loss: 0.0022:  51%|█████     | 144/281 [14:34<13:58,  6.12s/it]
Training Epoch: 16/100, batch 288/563 completed (loss: 0.0094:  51%|█████     | 144/281 [14:34<13:58,  6.12s/it]
Training Epoch: 16/100, batch 288/563 completed (loss: 0.0792:  51%|█████     | 144/281 [14:34<13:58,  6.12s/it]
Training Epoch: 16/100, batch 288/563 completed (loss: 0.1068:  51%|█████     | 144/281 [14:34<13:58,  6.12s/it]
Training Epoch: 16/100, batch 288/563 completed (loss: 0.1068:  52%|█████▏    | 145/281 [14:37<13:47,  6.09s/it]
Training Epoch: 16/100, batch 288/563 completed (loss: 0.0022:  52%|█████▏    | 145/281 [14:37<13:47,  6.09s/it]
Training Epoch: 16/100, batch 288/563 completed (loss: 0.0792:  52%|█████▏    | 145/281 [14:37<13:47,  6.09s/it]
Training Epoch: 16/100, batch 288/563 completed (loss: 0.0094:  52%|█████▏    | 145/281 [14:37<13:47,  6.09s/it]
Training Epoch: 16/100, batch 289/563 completed (loss: 0.1159:  52%|█████▏    | 145/281 [14:37<13:47,  6.09s/it]
Training Epoch: 16/100, batch 289/563 completed (loss: 0.0839:  52%|█████▏    | 145/281 [14:37<13:47,  6.09s/it]
Training Epoch: 16/100, batch 289/563 completed (loss: 0.0186:  52%|█████▏    | 145/281 [14:37<13:47,  6.09s/it]
Training Epoch: 16/100, batch 289/563 completed (loss: 0.0555:  52%|█████▏    | 145/281 [14:37<13:47,  6.09s/it]
Training Epoch: 16/100, batch 290/563 completed (loss: 0.0312:  52%|█████▏    | 145/281 [14:40<13:47,  6.09s/it]
Training Epoch: 16/100, batch 290/563 completed (loss: 0.0081:  52%|█████▏    | 145/281 [14:40<13:47,  6.09s/it]
Training Epoch: 16/100, batch 290/563 completed (loss: 0.0954:  52%|█████▏    | 145/281 [14:40<13:47,  6.09s/it]
Training Epoch: 16/100, batch 290/563 completed (loss: 0.1478:  52%|█████▏    | 145/281 [14:40<13:47,  6.09s/it]
Training Epoch: 16/100, batch 290/563 completed (loss: 0.1478:  52%|█████▏    | 146/281 [14:43<13:44,  6.11s/it]
Training Epoch: 16/100, batch 290/563 completed (loss: 0.0312:  52%|█████▏    | 146/281 [14:43<13:45,  6.11s/it]
Training Epoch: 16/100, batch 290/563 completed (loss: 0.0954:  52%|█████▏    | 146/281 [14:43<13:45,  6.11s/it]
Training Epoch: 16/100, batch 290/563 completed (loss: 0.0081:  52%|█████▏    | 146/281 [14:43<13:45,  6.12s/it]
Training Epoch: 16/100, batch 291/563 completed (loss: 0.0208:  52%|█████▏    | 146/281 [14:43<13:45,  6.11s/it]
Training Epoch: 16/100, batch 291/563 completed (loss: 0.018:  52%|█████▏    | 146/281 [14:43<13:45,  6.11s/it] 
Training Epoch: 16/100, batch 291/563 completed (loss: 0.0139:  52%|█████▏    | 146/281 [14:43<13:45,  6.12s/it]
Training Epoch: 16/100, batch 291/563 completed (loss: 0.0517:  52%|█████▏    | 146/281 [14:44<13:44,  6.11s/it]
Training Epoch: 16/100, batch 292/563 completed (loss: 0.0178:  52%|█████▏    | 146/281 [14:47<13:45,  6.11s/it]
Training Epoch: 16/100, batch 292/563 completed (loss: 0.0448:  52%|█████▏    | 146/281 [14:47<13:45,  6.11s/it]
Training Epoch: 16/100, batch 292/563 completed (loss: 0.022:  52%|█████▏    | 146/281 [14:47<13:45,  6.12s/it] 
Training Epoch: 16/100, batch 292/563 completed (loss: 0.0418:  52%|█████▏    | 146/281 [14:47<13:44,  6.11s/it]
Training Epoch: 16/100, batch 292/563 completed (loss: 0.0418:  52%|█████▏    | 147/281 [14:49<13:35,  6.08s/it]
Training Epoch: 16/100, batch 292/563 completed (loss: 0.0448:  52%|█████▏    | 147/281 [14:49<13:34,  6.08s/it]
Training Epoch: 16/100, batch 292/563 completed (loss: 0.022:  52%|█████▏    | 147/281 [14:49<13:34,  6.08s/it]
Training Epoch: 16/100, batch 292/563 completed (loss: 0.0178:  52%|█████▏    | 147/281 [14:49<13:35,  6.08s/it]
Training Epoch: 16/100, batch 293/563 completed (loss: 0.0172:  52%|█████▏    | 147/281 [14:49<13:35,  6.08s/it]
Training Epoch: 16/100, batch 293/563 completed (loss: 0.0585:  52%|█████▏    | 147/281 [14:49<13:34,  6.08s/it]
Training Epoch: 16/100, batch 293/563 completed (loss: 0.0365:  52%|█████▏    | 147/281 [14:49<13:34,  6.08s/it]
Training Epoch: 16/100, batch 293/563 completed (loss: 0.0504:  52%|█████▏    | 147/281 [14:50<13:35,  6.08s/it]
Training Epoch: 16/100, batch 294/563 completed (loss: 0.0258:  52%|█████▏    | 147/281 [14:53<13:35,  6.08s/it]
Training Epoch: 16/100, batch 294/563 completed (loss: 0.0039:  52%|█████▏    | 147/281 [14:53<13:34,  6.08s/it]
Training Epoch: 16/100, batch 294/563 completed (loss: 0.1038:  52%|█████▏    | 147/281 [14:53<13:34,  6.08s/it]
Training Epoch: 16/100, batch 294/563 completed (loss: 0.0408:  52%|█████▏    | 147/281 [14:53<13:35,  6.08s/it]
Training Epoch: 16/100, batch 294/563 completed (loss: 0.0408:  53%|█████▎    | 148/281 [14:55<13:26,  6.06s/it]
Training Epoch: 16/100, batch 294/563 completed (loss: 0.0039:  53%|█████▎    | 148/281 [14:55<13:26,  6.06s/it]
Training Epoch: 16/100, batch 294/563 completed (loss: 0.1038:  53%|█████▎    | 148/281 [14:55<13:26,  6.06s/it]
Training Epoch: 16/100, batch 294/563 completed (loss: 0.0258:  53%|█████▎    | 148/281 [14:55<13:26,  6.06s/it]
Training Epoch: 16/100, batch 295/563 completed (loss: 0.0779:  53%|█████▎    | 148/281 [14:55<13:26,  6.06s/it]
Training Epoch: 16/100, batch 295/563 completed (loss: 0.114:  53%|█████▎    | 148/281 [14:55<13:26,  6.06s/it] 
Training Epoch: 16/100, batch 295/563 completed (loss: 0.0201:  53%|█████▎    | 148/281 [14:55<13:26,  6.06s/it]
Training Epoch: 16/100, batch 295/563 completed (loss: 0.0316:  53%|█████▎    | 148/281 [14:56<13:26,  6.06s/it]
Training Epoch: 16/100, batch 296/563 completed (loss: 0.2036:  53%|█████▎    | 148/281 [14:58<13:26,  6.06s/it]
Training Epoch: 16/100, batch 296/563 completed (loss: 0.0193:  53%|█████▎    | 148/281 [14:58<13:26,  6.06s/it]
Training Epoch: 16/100, batch 296/563 completed (loss: 0.1901:  53%|█████▎    | 148/281 [14:58<13:26,  6.06s/it]
Training Epoch: 16/100, batch 296/563 completed (loss: 0.1706:  53%|█████▎    | 148/281 [14:58<13:26,  6.06s/it]
Training Epoch: 16/100, batch 296/563 completed (loss: 0.1706:  53%|█████▎    | 149/281 [15:02<13:30,  6.14s/it]
Training Epoch: 16/100, batch 296/563 completed (loss: 0.2036:  53%|█████▎    | 149/281 [15:02<13:30,  6.14s/it]
Training Epoch: 16/100, batch 296/563 completed (loss: 0.0193:  53%|█████▎    | 149/281 [15:02<13:30,  6.14s/it]
Training Epoch: 16/100, batch 296/563 completed (loss: 0.1901:  53%|█████▎    | 149/281 [15:02<13:30,  6.14s/it]
Training Epoch: 16/100, batch 297/563 completed (loss: 0.0318:  53%|█████▎    | 149/281 [15:02<13:30,  6.14s/it]
Training Epoch: 16/100, batch 297/563 completed (loss: 0.3391:  53%|█████▎    | 149/281 [15:02<13:30,  6.14s/it]
Training Epoch: 16/100, batch 297/563 completed (loss: 0.0202:  53%|█████▎    | 149/281 [15:02<13:30,  6.14s/it]
Training Epoch: 16/100, batch 297/563 completed (loss: 0.0836:  53%|█████▎    | 149/281 [15:02<13:30,  6.14s/it]
Training Epoch: 16/100, batch 298/563 completed (loss: 0.0201:  53%|█████▎    | 149/281 [15:05<13:30,  6.14s/it]
Training Epoch: 16/100, batch 298/563 completed (loss: 0.2103:  53%|█████▎    | 149/281 [15:05<13:30,  6.14s/it]
Training Epoch: 16/100, batch 298/563 completed (loss: 0.0257:  53%|█████▎    | 149/281 [15:05<13:30,  6.14s/it]
Training Epoch: 16/100, batch 298/563 completed (loss: 0.14:  53%|█████▎    | 149/281 [15:05<13:30,  6.14s/it]  
Training Epoch: 16/100, batch 298/563 completed (loss: 0.0257:  53%|█████▎    | 150/281 [15:08<13:12,  6.05s/it]
Training Epoch: 16/100, batch 298/563 completed (loss: 0.0201:  53%|█████▎    | 150/281 [15:08<13:12,  6.05s/it]
Training Epoch: 16/100, batch 298/563 completed (loss: 0.14:  53%|█████▎    | 150/281 [15:08<13:12,  6.05s/it]
Training Epoch: 16/100, batch 298/563 completed (loss: 0.2103:  53%|█████▎    | 150/281 [15:08<13:12,  6.05s/it]
Training Epoch: 16/100, batch 299/563 completed (loss: 0.2203:  53%|█████▎    | 150/281 [15:08<13:12,  6.05s/it]
Training Epoch: 16/100, batch 299/563 completed (loss: 0.0859:  53%|█████▎    | 150/281 [15:08<13:12,  6.05s/it]
Training Epoch: 16/100, batch 299/563 completed (loss: 0.0215:  53%|█████▎    | 150/281 [15:08<13:12,  6.05s/it]
Training Epoch: 16/100, batch 299/563 completed (loss: 0.0266:  53%|█████▎    | 150/281 [15:08<13:12,  6.05s/it]
Training Epoch: 16/100, batch 300/563 completed (loss: 0.2217:  53%|█████▎    | 150/281 [15:11<13:12,  6.05s/it]
Training Epoch: 16/100, batch 300/563 completed (loss: 0.0114:  53%|█████▎    | 150/281 [15:11<13:12,  6.05s/it]
Training Epoch: 16/100, batch 300/563 completed (loss: 0.0116:  53%|█████▎    | 150/281 [15:11<13:12,  6.05s/it]
Training Epoch: 16/100, batch 300/563 completed (loss: 0.0213:  53%|█████▎    | 150/281 [15:11<13:12,  6.05s/it]
Training Epoch: 16/100, batch 300/563 completed (loss: 0.0213:  54%|█████▎    | 151/281 [15:14<13:05,  6.04s/it]
Training Epoch: 16/100, batch 300/563 completed (loss: 0.2217:  54%|█████▎    | 151/281 [15:14<13:05,  6.04s/it]
Training Epoch: 16/100, batch 300/563 completed (loss: 0.0116:  54%|█████▎    | 151/281 [15:14<13:05,  6.04s/it]
Training Epoch: 16/100, batch 300/563 completed (loss: 0.0114:  54%|█████▎    | 151/281 [15:14<13:05,  6.04s/it]
Training Epoch: 16/100, batch 301/563 completed (loss: 0.0171:  54%|█████▎    | 151/281 [15:14<13:05,  6.04s/it]
Training Epoch: 16/100, batch 301/563 completed (loss: 0.0172:  54%|█████▎    | 151/281 [15:14<13:05,  6.04s/it]
Training Epoch: 16/100, batch 301/563 completed (loss: 0.1072:  54%|█████▎    | 151/281 [15:14<13:05,  6.04s/it]
Training Epoch: 16/100, batch 301/563 completed (loss: 0.6039:  54%|█████▎    | 151/281 [15:14<13:05,  6.04s/it]
Training Epoch: 16/100, batch 302/563 completed (loss: 0.0136:  54%|█████▎    | 151/281 [15:17<13:05,  6.04s/it]
Training Epoch: 16/100, batch 302/563 completed (loss: 0.0227:  54%|█████▎    | 151/281 [15:17<13:05,  6.04s/it]
Training Epoch: 16/100, batch 302/563 completed (loss: 0.0057:  54%|█████▎    | 151/281 [15:17<13:05,  6.04s/it]
Training Epoch: 16/100, batch 302/563 completed (loss: 0.0383:  54%|█████▎    | 151/281 [15:17<13:05,  6.04s/it]
Training Epoch: 16/100, batch 302/563 completed (loss: 0.0383:  54%|█████▍    | 152/281 [15:20<13:11,  6.14s/it]
Training Epoch: 16/100, batch 302/563 completed (loss: 0.0136:  54%|█████▍    | 152/281 [15:20<13:11,  6.14s/it]
Training Epoch: 16/100, batch 302/563 completed (loss: 0.0057:  54%|█████▍    | 152/281 [15:20<13:11,  6.14s/it]
Training Epoch: 16/100, batch 302/563 completed (loss: 0.0227:  54%|█████▍    | 152/281 [15:20<13:11,  6.14s/it]
Training Epoch: 16/100, batch 303/563 completed (loss: 0.0106:  54%|█████▍    | 152/281 [15:20<13:11,  6.14s/it]
Training Epoch: 16/100, batch 303/563 completed (loss: 0.0429:  54%|█████▍    | 152/281 [15:20<13:11,  6.14s/it]
Training Epoch: 16/100, batch 303/563 completed (loss: 0.0593:  54%|█████▍    | 152/281 [15:20<13:11,  6.14s/it]
Training Epoch: 16/100, batch 303/563 completed (loss: 0.2795:  54%|█████▍    | 152/281 [15:20<13:11,  6.14s/it]
Training Epoch: 16/100, batch 304/563 completed (loss: 0.0432:  54%|█████▍    | 152/281 [15:23<13:11,  6.14s/it]
Training Epoch: 16/100, batch 304/563 completed (loss: 0.0101:  54%|█████▍    | 152/281 [15:23<13:11,  6.14s/it]
Training Epoch: 16/100, batch 304/563 completed (loss: 0.058:  54%|█████▍    | 152/281 [15:23<13:11,  6.14s/it] 
Training Epoch: 16/100, batch 304/563 completed (loss: 0.0087:  54%|█████▍    | 152/281 [15:23<13:11,  6.14s/it]
Training Epoch: 16/100, batch 304/563 completed (loss: 0.0101:  54%|█████▍    | 153/281 [15:26<13:00,  6.10s/it]
Training Epoch: 16/100, batch 304/563 completed (loss: 0.0087:  54%|█████▍    | 153/281 [15:26<13:00,  6.10s/it]
Training Epoch: 16/100, batch 304/563 completed (loss: 0.058:  54%|█████▍    | 153/281 [15:26<13:00,  6.10s/it]
Training Epoch: 16/100, batch 304/563 completed (loss: 0.0432:  54%|█████▍    | 153/281 [15:26<13:00,  6.10s/it]
Training Epoch: 16/100, batch 305/563 completed (loss: 0.1712:  54%|█████▍    | 153/281 [15:26<13:00,  6.10s/it]
Training Epoch: 16/100, batch 305/563 completed (loss: 0.0207:  54%|█████▍    | 153/281 [15:26<13:00,  6.10s/it]
Training Epoch: 16/100, batch 305/563 completed (loss: 0.0644:  54%|█████▍    | 153/281 [15:26<13:00,  6.10s/it]
Training Epoch: 16/100, batch 305/563 completed (loss: 0.1819:  54%|█████▍    | 153/281 [15:26<13:00,  6.10s/it]
Training Epoch: 16/100, batch 306/563 completed (loss: 0.0304:  54%|█████▍    | 153/281 [15:29<13:00,  6.10s/it]
Training Epoch: 16/100, batch 306/563 completed (loss: 0.0915:  54%|█████▍    | 153/281 [15:29<13:00,  6.10s/it]
Training Epoch: 16/100, batch 306/563 completed (loss: 0.1029:  54%|█████▍    | 153/281 [15:29<13:00,  6.10s/it]
Training Epoch: 16/100, batch 306/563 completed (loss: 0.0057:  54%|█████▍    | 153/281 [15:29<13:00,  6.10s/it]
Training Epoch: 16/100, batch 306/563 completed (loss: 0.0057:  55%|█████▍    | 154/281 [15:32<12:52,  6.09s/it]
Training Epoch: 16/100, batch 306/563 completed (loss: 0.1029:  55%|█████▍    | 154/281 [15:32<12:52,  6.09s/it]
Training Epoch: 16/100, batch 306/563 completed (loss: 0.0915:  55%|█████▍    | 154/281 [15:32<12:53,  6.09s/it]
Training Epoch: 16/100, batch 306/563 completed (loss: 0.0304:  55%|█████▍    | 154/281 [15:32<12:53,  6.09s/it]
Training Epoch: 16/100, batch 307/563 completed (loss: 0.0419:  55%|█████▍    | 154/281 [15:32<12:53,  6.09s/it]
Training Epoch: 16/100, batch 307/563 completed (loss: 0.1127:  55%|█████▍    | 154/281 [15:32<12:53,  6.09s/it]
Training Epoch: 16/100, batch 307/563 completed (loss: 0.0265:  55%|█████▍    | 154/281 [15:32<12:52,  6.09s/it]
Training Epoch: 16/100, batch 307/563 completed (loss: 0.0428:  55%|█████▍    | 154/281 [15:32<12:52,  6.09s/it]
Training Epoch: 16/100, batch 308/563 completed (loss: 0.0253:  55%|█████▍    | 154/281 [15:35<12:52,  6.09s/it]
Training Epoch: 16/100, batch 308/563 completed (loss: 0.0437:  55%|█████▍    | 154/281 [15:35<12:53,  6.09s/it]
Training Epoch: 16/100, batch 308/563 completed (loss: 0.0036:  55%|█████▍    | 154/281 [15:35<12:53,  6.09s/it]
Training Epoch: 16/100, batch 308/563 completed (loss: 0.0317:  55%|█████▍    | 154/281 [15:35<12:52,  6.09s/it]
Training Epoch: 16/100, batch 308/563 completed (loss: 0.0317:  55%|█████▌    | 155/281 [15:38<12:44,  6.07s/it]
Training Epoch: 16/100, batch 308/563 completed (loss: 0.0253:  55%|█████▌    | 155/281 [15:38<12:44,  6.07s/it]
Training Epoch: 16/100, batch 308/563 completed (loss: 0.0437:  55%|█████▌    | 155/281 [15:38<12:44,  6.07s/it]
Training Epoch: 16/100, batch 308/563 completed (loss: 0.0036:  55%|█████▌    | 155/281 [15:38<12:44,  6.07s/it]
Training Epoch: 16/100, batch 309/563 completed (loss: 0.0226:  55%|█████▌    | 155/281 [15:38<12:44,  6.07s/it]
Training Epoch: 16/100, batch 309/563 completed (loss: 0.1359:  55%|█████▌    | 155/281 [15:38<12:44,  6.07s/it]
Training Epoch: 16/100, batch 309/563 completed (loss: 0.1291:  55%|█████▌    | 155/281 [15:38<12:44,  6.07s/it]
Training Epoch: 16/100, batch 309/563 completed (loss: 0.0438:  55%|█████▌    | 155/281 [15:38<12:44,  6.07s/it]
Training Epoch: 16/100, batch 310/563 completed (loss: 0.0422:  55%|█████▌    | 155/281 [15:41<12:44,  6.07s/it]
Training Epoch: 16/100, batch 310/563 completed (loss: 0.0693:  55%|█████▌    | 155/281 [15:41<12:44,  6.07s/it]
Training Epoch: 16/100, batch 310/563 completed (loss: 0.0547:  55%|█████▌    | 155/281 [15:41<12:44,  6.07s/it]
Training Epoch: 16/100, batch 310/563 completed (loss: 0.0729:  55%|█████▌    | 155/281 [15:41<12:44,  6.07s/it]
Training Epoch: 16/100, batch 310/563 completed (loss: 0.0693:  56%|█████▌    | 156/281 [15:44<12:36,  6.05s/it]
Training Epoch: 16/100, batch 310/563 completed (loss: 0.0547:  56%|█████▌    | 156/281 [15:44<12:37,  6.06s/it]
Training Epoch: 16/100, batch 310/563 completed (loss: 0.0422:  56%|█████▌    | 156/281 [15:44<12:36,  6.05s/it]
Training Epoch: 16/100, batch 310/563 completed (loss: 0.0729:  56%|█████▌    | 156/281 [15:44<12:37,  6.06s/it]
Training Epoch: 16/100, batch 311/563 completed (loss: 0.0442:  56%|█████▌    | 156/281 [15:44<12:36,  6.05s/it]
Training Epoch: 16/100, batch 311/563 completed (loss: 0.0666:  56%|█████▌    | 156/281 [15:44<12:36,  6.05s/it]
Training Epoch: 16/100, batch 311/563 completed (loss: 0.0636:  56%|█████▌    | 156/281 [15:44<12:37,  6.06s/it]
Training Epoch: 16/100, batch 311/563 completed (loss: 0.0061:  56%|█████▌    | 156/281 [15:44<12:37,  6.06s/it]
Training Epoch: 16/100, batch 312/563 completed (loss: 0.0293:  56%|█████▌    | 156/281 [15:47<12:36,  6.05s/it]
Training Epoch: 16/100, batch 312/563 completed (loss: 0.0266:  56%|█████▌    | 156/281 [15:47<12:36,  6.05s/it]
Training Epoch: 16/100, batch 312/563 completed (loss: 0.322:  56%|█████▌    | 156/281 [15:47<12:37,  6.06s/it] 
Training Epoch: 16/100, batch 312/563 completed (loss: 0.4227:  56%|█████▌    | 156/281 [15:47<12:37,  6.06s/it]
Training Epoch: 16/100, batch 312/563 completed (loss: 0.0266:  56%|█████▌    | 157/281 [15:50<12:39,  6.12s/it]
Training Epoch: 16/100, batch 312/563 completed (loss: 0.322:  56%|█████▌    | 157/281 [15:50<12:39,  6.13s/it]
Training Epoch: 16/100, batch 312/563 completed (loss: 0.4227:  56%|█████▌    | 157/281 [15:50<12:39,  6.13s/it]
Training Epoch: 16/100, batch 312/563 completed (loss: 0.0293:  56%|█████▌    | 157/281 [15:50<12:39,  6.13s/it]
Training Epoch: 16/100, batch 313/563 completed (loss: 0.0054:  56%|█████▌    | 157/281 [15:50<12:39,  6.13s/it]
Training Epoch: 16/100, batch 313/563 completed (loss: 0.0828:  56%|█████▌    | 157/281 [15:50<12:39,  6.12s/it]
Training Epoch: 16/100, batch 313/563 completed (loss: 0.0397:  56%|█████▌    | 157/281 [15:50<12:39,  6.13s/it]
Training Epoch: 16/100, batch 313/563 completed (loss: 0.1492:  56%|█████▌    | 157/281 [15:50<12:39,  6.13s/it]
Training Epoch: 16/100, batch 314/563 completed (loss: 0.0965:  56%|█████▌    | 157/281 [15:53<12:39,  6.12s/it]
Training Epoch: 16/100, batch 314/563 completed (loss: 0.1889:  56%|█████▌    | 157/281 [15:53<12:39,  6.13s/it]
Training Epoch: 16/100, batch 314/563 completed (loss: 0.038:  56%|█████▌    | 157/281 [15:53<12:39,  6.13s/it] 
Training Epoch: 16/100, batch 314/563 completed (loss: 0.0162:  56%|█████▌    | 157/281 [15:53<12:39,  6.13s/it]
Training Epoch: 16/100, batch 314/563 completed (loss: 0.038:  56%|█████▌    | 158/281 [15:56<12:29,  6.09s/it]
Training Epoch: 16/100, batch 314/563 completed (loss: 0.0965:  56%|█████▌    | 158/281 [15:56<12:29,  6.10s/it]
Training Epoch: 16/100, batch 314/563 completed (loss: 0.1889:  56%|█████▌    | 158/281 [15:56<12:30,  6.10s/it]
Training Epoch: 16/100, batch 314/563 completed (loss: 0.0162:  56%|█████▌    | 158/281 [15:56<12:30,  6.10s/it]
Training Epoch: 16/100, batch 315/563 completed (loss: 0.0474:  56%|█████▌    | 158/281 [15:56<12:29,  6.10s/it]
Training Epoch: 16/100, batch 315/563 completed (loss: 0.1475:  56%|█████▌    | 158/281 [15:56<12:29,  6.09s/it]
Training Epoch: 16/100, batch 315/563 completed (loss: 0.0097:  56%|█████▌    | 158/281 [15:56<12:30,  6.10s/it]
Training Epoch: 16/100, batch 315/563 completed (loss: 0.006:  56%|█████▌    | 158/281 [15:57<12:30,  6.10s/it] 
Training Epoch: 16/100, batch 316/563 completed (loss: 0.0139:  56%|█████▌    | 158/281 [15:59<12:29,  6.09s/it]
Training Epoch: 16/100, batch 316/563 completed (loss: 0.017:  56%|█████▌    | 158/281 [15:59<12:30,  6.10s/it] 
Training Epoch: 16/100, batch 316/563 completed (loss: 0.0577:  56%|█████▌    | 158/281 [15:59<12:29,  6.10s/it]
Training Epoch: 16/100, batch 316/563 completed (loss: 0.0084:  56%|█████▌    | 158/281 [15:59<12:30,  6.10s/it]
Training Epoch: 16/100, batch 316/563 completed (loss: 0.0084:  57%|█████▋    | 159/281 [16:02<12:14,  6.02s/it]
Training Epoch: 16/100, batch 316/563 completed (loss: 0.017:  57%|█████▋    | 159/281 [16:02<12:15,  6.02s/it]
Training Epoch: 16/100, batch 316/563 completed (loss: 0.0577:  57%|█████▋    | 159/281 [16:02<12:15,  6.03s/it]
Training Epoch: 16/100, batch 316/563 completed (loss: 0.0139:  57%|█████▋    | 159/281 [16:02<12:15,  6.03s/it]
Training Epoch: 16/100, batch 317/563 completed (loss: 0.0291:  57%|█████▋    | 159/281 [16:02<12:15,  6.03s/it]
Training Epoch: 16/100, batch 317/563 completed (loss: 0.054:  57%|█████▋    | 159/281 [16:02<12:15,  6.03s/it] 
Training Epoch: 16/100, batch 317/563 completed (loss: 0.0145:  57%|█████▋    | 159/281 [16:02<12:15,  6.02s/it]
Training Epoch: 16/100, batch 317/563 completed (loss: 0.347:  57%|█████▋    | 159/281 [16:02<12:14,  6.02s/it] 
Training Epoch: 16/100, batch 318/563 completed (loss: 0.053:  57%|█████▋    | 159/281 [16:06<12:15,  6.03s/it] 
Training Epoch: 16/100, batch 318/563 completed (loss: 0.3943:  57%|█████▋    | 159/281 [16:06<12:15,  6.03s/it]
Training Epoch: 16/100, batch 318/563 completed (loss: 0.0313:  57%|█████▋    | 159/281 [16:06<12:15,  6.02s/it]
Training Epoch: 16/100, batch 318/563 completed (loss: 0.0477:  57%|█████▋    | 159/281 [16:06<12:14,  6.02s/it]
Training Epoch: 16/100, batch 318/563 completed (loss: 0.0313:  57%|█████▋    | 160/281 [16:09<12:24,  6.15s/it]
Training Epoch: 16/100, batch 318/563 completed (loss: 0.3943:  57%|█████▋    | 160/281 [16:09<12:24,  6.15s/it]
Training Epoch: 16/100, batch 318/563 completed (loss: 0.0477:  57%|█████▋    | 160/281 [16:09<12:24,  6.16s/it]
Training Epoch: 16/100, batch 318/563 completed (loss: 0.053:  57%|█████▋    | 160/281 [16:09<12:24,  6.16s/it]
Training Epoch: 16/100, batch 319/563 completed (loss: 0.0394:  57%|█████▋    | 160/281 [16:09<12:24,  6.16s/it]
Training Epoch: 16/100, batch 319/563 completed (loss: 0.2635:  57%|█████▋    | 160/281 [16:09<12:24,  6.15s/it]
Training Epoch: 16/100, batch 319/563 completed (loss: 0.0219:  57%|█████▋    | 160/281 [16:09<12:24,  6.15s/it]
Training Epoch: 16/100, batch 319/563 completed (loss: 0.0116:  57%|█████▋    | 160/281 [16:09<12:24,  6.16s/it]
Training Epoch: 16/100, batch 320/563 completed (loss: 0.0328:  57%|█████▋    | 160/281 [16:12<12:24,  6.15s/it]
Training Epoch: 16/100, batch 320/563 completed (loss: 0.0117:  57%|█████▋    | 160/281 [16:12<12:24,  6.16s/it]
Training Epoch: 16/100, batch 320/563 completed (loss: 0.0274:  57%|█████▋    | 160/281 [16:12<12:24,  6.15s/it]
Training Epoch: 16/100, batch 320/563 completed (loss: 0.005:  57%|█████▋    | 160/281 [16:12<12:24,  6.16s/it] 
Training Epoch: 16/100, batch 320/563 completed (loss: 0.0274:  57%|█████▋    | 161/281 [16:15<12:09,  6.08s/it]
Training Epoch: 16/100, batch 320/563 completed (loss: 0.0117:  57%|█████▋    | 161/281 [16:15<12:09,  6.08s/it]
Training Epoch: 16/100, batch 320/563 completed (loss: 0.0328:  57%|█████▋    | 161/281 [16:15<12:10,  6.08s/it]
Training Epoch: 16/100, batch 320/563 completed (loss: 0.005:  57%|█████▋    | 161/281 [16:15<12:09,  6.08s/it]
Training Epoch: 16/100, batch 321/563 completed (loss: 0.002:  57%|█████▋    | 161/281 [16:15<12:10,  6.08s/it] 
Training Epoch: 16/100, batch 321/563 completed (loss: 0.006:  57%|█████▋    | 161/281 [16:15<12:09,  6.08s/it] 
Training Epoch: 16/100, batch 321/563 completed (loss: 0.0394:  57%|█████▋    | 161/281 [16:15<12:09,  6.08s/it]
Training Epoch: 16/100, batch 321/563 completed (loss: 0.0133:  57%|█████▋    | 161/281 [16:15<12:09,  6.08s/it]
Training Epoch: 16/100, batch 322/563 completed (loss: 0.1536:  57%|█████▋    | 161/281 [16:18<12:09,  6.08s/it]
Training Epoch: 16/100, batch 322/563 completed (loss: 0.0317:  57%|█████▋    | 161/281 [16:18<12:10,  6.08s/it]
Training Epoch: 16/100, batch 322/563 completed (loss: 0.0862:  57%|█████▋    | 161/281 [16:18<12:09,  6.08s/it]
Training Epoch: 16/100, batch 322/563 completed (loss: 0.0773:  57%|█████▋    | 161/281 [16:18<12:09,  6.08s/it]
Training Epoch: 16/100, batch 322/563 completed (loss: 0.0773:  58%|█████▊    | 162/281 [16:21<12:06,  6.11s/it]
Training Epoch: 16/100, batch 322/563 completed (loss: 0.0862:  58%|█████▊    | 162/281 [16:21<12:06,  6.11s/it]
Training Epoch: 16/100, batch 322/563 completed (loss: 0.1536:  58%|█████▊    | 162/281 [16:21<12:07,  6.11s/it]
Training Epoch: 16/100, batch 322/563 completed (loss: 0.0317:  58%|█████▊    | 162/281 [16:21<12:07,  6.11s/it]
Training Epoch: 16/100, batch 323/563 completed (loss: 0.0232:  58%|█████▊    | 162/281 [16:21<12:07,  6.11s/it]
Training Epoch: 16/100, batch 323/563 completed (loss: 0.0595:  58%|█████▊    | 162/281 [16:21<12:06,  6.11s/it]
Training Epoch: 16/100, batch 323/563 completed (loss: 0.2209:  58%|█████▊    | 162/281 [16:21<12:07,  6.11s/it]
Training Epoch: 16/100, batch 323/563 completed (loss: 0.0047:  58%|█████▊    | 162/281 [16:21<12:06,  6.11s/it]
Training Epoch: 16/100, batch 324/563 completed (loss: 0.5478:  58%|█████▊    | 162/281 [16:24<12:07,  6.11s/it]
Training Epoch: 16/100, batch 324/563 completed (loss: 0.0157:  58%|█████▊    | 162/281 [16:24<12:06,  6.11s/it]
Training Epoch: 16/100, batch 324/563 completed (loss: 0.0416:  58%|█████▊    | 162/281 [16:24<12:07,  6.11s/it]
Training Epoch: 16/100, batch 324/563 completed (loss: 0.1735:  58%|█████▊    | 162/281 [16:24<12:06,  6.11s/it]
Training Epoch: 16/100, batch 324/563 completed (loss: 0.0157:  58%|█████▊    | 163/281 [16:27<12:02,  6.13s/it]
Training Epoch: 16/100, batch 324/563 completed (loss: 0.1735:  58%|█████▊    | 163/281 [16:27<12:02,  6.13s/it]
Training Epoch: 16/100, batch 324/563 completed (loss: 0.0416:  58%|█████▊    | 163/281 [16:27<12:03,  6.13s/it]
Training Epoch: 16/100, batch 324/563 completed (loss: 0.5478:  58%|█████▊    | 163/281 [16:27<12:02,  6.13s/it]
Training Epoch: 16/100, batch 325/563 completed (loss: 0.0624:  58%|█████▊    | 163/281 [16:27<12:03,  6.13s/it]
Training Epoch: 16/100, batch 325/563 completed (loss: 0.0353:  58%|█████▊    | 163/281 [16:27<12:02,  6.13s/it]
Training Epoch: 16/100, batch 325/563 completed (loss: 0.0406:  58%|█████▊    | 163/281 [16:27<12:02,  6.13s/it]
Training Epoch: 16/100, batch 325/563 completed (loss: 0.0049:  58%|█████▊    | 163/281 [16:27<12:02,  6.13s/it]
Training Epoch: 16/100, batch 326/563 completed (loss: 0.0072:  58%|█████▊    | 163/281 [16:30<12:02,  6.13s/it]
Training Epoch: 16/100, batch 326/563 completed (loss: 0.0343:  58%|█████▊    | 163/281 [16:30<12:02,  6.13s/it]
Training Epoch: 16/100, batch 326/563 completed (loss: 0.2785:  58%|█████▊    | 163/281 [16:30<12:03,  6.13s/it]
Training Epoch: 16/100, batch 326/563 completed (loss: 0.0137:  58%|█████▊    | 163/281 [16:30<12:02,  6.13s/it]
Training Epoch: 16/100, batch 326/563 completed (loss: 0.0137:  58%|█████▊    | 164/281 [16:33<11:47,  6.04s/it]
Training Epoch: 16/100, batch 326/563 completed (loss: 0.0072:  58%|█████▊    | 164/281 [16:33<11:47,  6.04s/it]
Training Epoch: 16/100, batch 326/563 completed (loss: 0.2785:  58%|█████▊    | 164/281 [16:33<11:47,  6.05s/it]
Training Epoch: 16/100, batch 326/563 completed (loss: 0.0343:  58%|█████▊    | 164/281 [16:33<11:47,  6.04s/it]
Training Epoch: 16/100, batch 327/563 completed (loss: 0.0272:  58%|█████▊    | 164/281 [16:33<11:47,  6.05s/it]
Training Epoch: 16/100, batch 327/563 completed (loss: 0.0046:  58%|█████▊    | 164/281 [16:33<11:47,  6.04s/it]
Training Epoch: 16/100, batch 327/563 completed (loss: 0.0035:  58%|█████▊    | 164/281 [16:33<11:47,  6.04s/it]
Training Epoch: 16/100, batch 327/563 completed (loss: 0.2266:  58%|█████▊    | 164/281 [16:33<11:47,  6.04s/it]
Training Epoch: 16/100, batch 328/563 completed (loss: 0.0101:  58%|█████▊    | 164/281 [16:36<11:47,  6.04s/it]
Training Epoch: 16/100, batch 328/563 completed (loss: 0.071:  58%|█████▊    | 164/281 [16:36<11:47,  6.05s/it] 
Training Epoch: 16/100, batch 328/563 completed (loss: 0.1425:  58%|█████▊    | 164/281 [16:36<11:47,  6.04s/it]
Training Epoch: 16/100, batch 328/563 completed (loss: 0.0129:  58%|█████▊    | 164/281 [16:36<11:47,  6.04s/it]
Training Epoch: 16/100, batch 328/563 completed (loss: 0.1425:  59%|█████▊    | 165/281 [16:39<11:49,  6.12s/it]
Training Epoch: 16/100, batch 328/563 completed (loss: 0.0129:  59%|█████▊    | 165/281 [16:39<11:50,  6.12s/it]
Training Epoch: 16/100, batch 328/563 completed (loss: 0.071:  59%|█████▊    | 165/281 [16:39<11:50,  6.12s/it]
Training Epoch: 16/100, batch 328/563 completed (loss: 0.0101:  59%|█████▊    | 165/281 [16:39<11:50,  6.12s/it]
Training Epoch: 16/100, batch 329/563 completed (loss: 0.0238:  59%|█████▊    | 165/281 [16:39<11:50,  6.12s/it]
Training Epoch: 16/100, batch 329/563 completed (loss: 0.0113:  59%|█████▊    | 165/281 [16:39<11:49,  6.12s/it]
Training Epoch: 16/100, batch 329/563 completed (loss: 0.0548:  59%|█████▊    | 165/281 [16:39<11:50,  6.12s/it]
Training Epoch: 16/100, batch 329/563 completed (loss: 0.0059:  59%|█████▊    | 165/281 [16:39<11:50,  6.12s/it]
Training Epoch: 16/100, batch 330/563 completed (loss: 0.1781:  59%|█████▊    | 165/281 [16:42<11:50,  6.12s/it]
Training Epoch: 16/100, batch 330/563 completed (loss: 0.0732:  59%|█████▊    | 165/281 [16:42<11:49,  6.12s/it]
Training Epoch: 16/100, batch 330/563 completed (loss: 0.1369:  59%|█████▊    | 165/281 [16:42<11:50,  6.12s/it]
Training Epoch: 16/100, batch 330/563 completed (loss: 0.0608:  59%|█████▊    | 165/281 [16:42<11:50,  6.12s/it]
Training Epoch: 16/100, batch 330/563 completed (loss: 0.0608:  59%|█████▉    | 166/281 [16:45<11:40,  6.09s/it]
Training Epoch: 16/100, batch 330/563 completed (loss: 0.1781:  59%|█████▉    | 166/281 [16:45<11:39,  6.09s/it]
Training Epoch: 16/100, batch 330/563 completed (loss: 0.0732:  59%|█████▉    | 166/281 [16:45<11:40,  6.09s/it]
Training Epoch: 16/100, batch 330/563 completed (loss: 0.1369:  59%|█████▉    | 166/281 [16:45<11:40,  6.09s/it]
Training Epoch: 16/100, batch 331/563 completed (loss: 0.0688:  59%|█████▉    | 166/281 [16:45<11:39,  6.09s/it]
Training Epoch: 16/100, batch 331/563 completed (loss: 0.1646:  59%|█████▉    | 166/281 [16:45<11:40,  6.09s/it]
Training Epoch: 16/100, batch 331/563 completed (loss: 0.133:  59%|█████▉    | 166/281 [16:45<11:40,  6.09s/it] 
Training Epoch: 16/100, batch 331/563 completed (loss: 0.0246:  59%|█████▉    | 166/281 [16:45<11:40,  6.09s/it]
Training Epoch: 16/100, batch 332/563 completed (loss: 0.1522:  59%|█████▉    | 166/281 [16:48<11:39,  6.09s/it]
Training Epoch: 16/100, batch 332/563 completed (loss: 0.0498:  59%|█████▉    | 166/281 [16:48<11:40,  6.09s/it]
Training Epoch: 16/100, batch 332/563 completed (loss: 0.022:  59%|█████▉    | 166/281 [16:48<11:40,  6.09s/it] 
Training Epoch: 16/100, batch 332/563 completed (loss: 0.0333:  59%|█████▉    | 166/281 [16:48<11:40,  6.09s/it]
Training Epoch: 16/100, batch 332/563 completed (loss: 0.0333:  59%|█████▉    | 167/281 [16:51<11:31,  6.07s/it]
Training Epoch: 16/100, batch 332/563 completed (loss: 0.1522:  59%|█████▉    | 167/281 [16:51<11:31,  6.07s/it]
Training Epoch: 16/100, batch 332/563 completed (loss: 0.022:  59%|█████▉    | 167/281 [16:51<11:31,  6.07s/it]
Training Epoch: 16/100, batch 332/563 completed (loss: 0.0498:  59%|█████▉    | 167/281 [16:51<11:32,  6.07s/it]
Training Epoch: 16/100, batch 333/563 completed (loss: 0.0193:  59%|█████▉    | 167/281 [16:51<11:31,  6.07s/it]
Training Epoch: 16/100, batch 333/563 completed (loss: 0.011:  59%|█████▉    | 167/281 [16:51<11:31,  6.07s/it] 
Training Epoch: 16/100, batch 333/563 completed (loss: 0.0594:  59%|█████▉    | 167/281 [16:51<11:32,  6.07s/it]
Training Epoch: 16/100, batch 333/563 completed (loss: 0.1654:  59%|█████▉    | 167/281 [16:51<11:31,  6.07s/it]
Training Epoch: 16/100, batch 334/563 completed (loss: 0.3476:  59%|█████▉    | 167/281 [16:54<11:31,  6.07s/it]
Training Epoch: 16/100, batch 334/563 completed (loss: 0.0951:  59%|█████▉    | 167/281 [16:54<11:31,  6.07s/it]
Training Epoch: 16/100, batch 334/563 completed (loss: 0.0291:  59%|█████▉    | 167/281 [16:54<11:32,  6.07s/it]
Training Epoch: 16/100, batch 334/563 completed (loss: 0.0921:  59%|█████▉    | 167/281 [16:54<11:31,  6.07s/it]
Training Epoch: 16/100, batch 334/563 completed (loss: 0.3476:  60%|█████▉    | 168/281 [16:58<11:34,  6.15s/it]
Training Epoch: 16/100, batch 334/563 completed (loss: 0.0921:  60%|█████▉    | 168/281 [16:58<11:34,  6.15s/it]
Training Epoch: 16/100, batch 334/563 completed (loss: 0.0291:  60%|█████▉    | 168/281 [16:58<11:34,  6.15s/it]
Training Epoch: 16/100, batch 334/563 completed (loss: 0.0951:  60%|█████▉    | 168/281 [16:58<11:34,  6.15s/it]
Training Epoch: 16/100, batch 335/563 completed (loss: 0.2121:  60%|█████▉    | 168/281 [16:58<11:34,  6.15s/it]
Training Epoch: 16/100, batch 335/563 completed (loss: 0.0327:  60%|█████▉    | 168/281 [16:58<11:34,  6.15s/it]
Training Epoch: 16/100, batch 335/563 completed (loss: 0.0032:  60%|█████▉    | 168/281 [16:58<11:34,  6.15s/it]
Training Epoch: 16/100, batch 335/563 completed (loss: 0.0157:  60%|█████▉    | 168/281 [16:58<11:34,  6.15s/it]
Training Epoch: 16/100, batch 336/563 completed (loss: 0.0311:  60%|█████▉    | 168/281 [17:00<11:34,  6.15s/it]
Training Epoch: 16/100, batch 336/563 completed (loss: 0.0083:  60%|█████▉    | 168/281 [17:00<11:34,  6.15s/it]
Training Epoch: 16/100, batch 336/563 completed (loss: 0.1428:  60%|█████▉    | 168/281 [17:00<11:34,  6.15s/it]
Training Epoch: 16/100, batch 336/563 completed (loss: 0.2351:  60%|█████▉    | 168/281 [17:01<11:34,  6.15s/it]
Training Epoch: 16/100, batch 336/563 completed (loss: 0.0083:  60%|██████    | 169/281 [17:03<11:18,  6.06s/it]
Training Epoch: 16/100, batch 336/563 completed (loss: 0.1428:  60%|██████    | 169/281 [17:03<11:18,  6.06s/it]
Training Epoch: 16/100, batch 336/563 completed (loss: 0.2351:  60%|██████    | 169/281 [17:03<11:19,  6.06s/it]
Training Epoch: 16/100, batch 336/563 completed (loss: 0.0311:  60%|██████    | 169/281 [17:03<11:19,  6.06s/it]
Training Epoch: 16/100, batch 337/563 completed (loss: 0.0282:  60%|██████    | 169/281 [17:03<11:19,  6.06s/it]
Training Epoch: 16/100, batch 337/563 completed (loss: 0.0316:  60%|██████    | 169/281 [17:03<11:18,  6.06s/it]
Training Epoch: 16/100, batch 337/563 completed (loss: 0.0581:  60%|██████    | 169/281 [17:03<11:18,  6.06s/it]
Training Epoch: 16/100, batch 337/563 completed (loss: 0.0081:  60%|██████    | 169/281 [17:03<11:19,  6.06s/it]
Training Epoch: 16/100, batch 338/563 completed (loss: 0.0145:  60%|██████    | 169/281 [17:07<11:19,  6.06s/it]
Training Epoch: 16/100, batch 338/563 completed (loss: 0.1821:  60%|██████    | 169/281 [17:07<11:18,  6.06s/it]
Training Epoch: 16/100, batch 338/563 completed (loss: 0.0849:  60%|██████    | 169/281 [17:07<11:18,  6.06s/it]
Training Epoch: 16/100, batch 338/563 completed (loss: 0.0713:  60%|██████    | 169/281 [17:07<11:19,  6.06s/it]
Training Epoch: 16/100, batch 338/563 completed (loss: 0.0145:  60%|██████    | 170/281 [17:10<11:16,  6.10s/it]
Training Epoch: 16/100, batch 338/563 completed (loss: 0.1821:  60%|██████    | 170/281 [17:10<11:16,  6.10s/it]
Training Epoch: 16/100, batch 338/563 completed (loss: 0.0849:  60%|██████    | 170/281 [17:10<11:16,  6.10s/it]
Training Epoch: 16/100, batch 338/563 completed (loss: 0.0713:  60%|██████    | 170/281 [17:10<11:17,  6.10s/it]
Training Epoch: 16/100, batch 339/563 completed (loss: 0.0257:  60%|██████    | 170/281 [17:10<11:16,  6.10s/it]
Training Epoch: 16/100, batch 339/563 completed (loss: 0.016:  60%|██████    | 170/281 [17:10<11:16,  6.10s/it] 
Training Epoch: 16/100, batch 339/563 completed (loss: 0.0197:  60%|██████    | 170/281 [17:10<11:16,  6.10s/it]
Training Epoch: 16/100, batch 339/563 completed (loss: 0.0699:  60%|██████    | 170/281 [17:10<11:17,  6.10s/it]
Training Epoch: 16/100, batch 340/563 completed (loss: 0.0673:  60%|██████    | 170/281 [17:13<11:16,  6.10s/it]
Training Epoch: 16/100, batch 340/563 completed (loss: 0.3142:  60%|██████    | 170/281 [17:13<11:16,  6.10s/it]
Training Epoch: 16/100, batch 340/563 completed (loss: 0.0176:  60%|██████    | 170/281 [17:13<11:16,  6.10s/it]
Training Epoch: 16/100, batch 340/563 completed (loss: 0.0225:  60%|██████    | 170/281 [17:13<11:17,  6.10s/it]
Training Epoch: 16/100, batch 340/563 completed (loss: 0.0225:  61%|██████    | 171/281 [17:16<11:13,  6.12s/it]
Training Epoch: 16/100, batch 340/563 completed (loss: 0.3142:  61%|██████    | 171/281 [17:16<11:13,  6.12s/it]
Training Epoch: 16/100, batch 340/563 completed (loss: 0.0176:  61%|██████    | 171/281 [17:16<11:13,  6.12s/it]
Training Epoch: 16/100, batch 340/563 completed (loss: 0.0673:  61%|██████    | 171/281 [17:16<11:13,  6.12s/it]
Training Epoch: 16/100, batch 341/563 completed (loss: 0.0099:  61%|██████    | 171/281 [17:16<11:13,  6.12s/it]
Training Epoch: 16/100, batch 341/563 completed (loss: 0.0536:  61%|██████    | 171/281 [17:16<11:13,  6.12s/it]
Training Epoch: 16/100, batch 341/563 completed (loss: 0.0597:  61%|██████    | 171/281 [17:16<11:13,  6.12s/it]
Training Epoch: 16/100, batch 341/563 completed (loss: 0.4448:  61%|██████    | 171/281 [17:16<11:13,  6.12s/it]
Training Epoch: 16/100, batch 342/563 completed (loss: 0.1749:  61%|██████    | 171/281 [17:19<11:13,  6.12s/it]
Training Epoch: 16/100, batch 342/563 completed (loss: 0.0065:  61%|██████    | 171/281 [17:19<11:13,  6.12s/it]
Training Epoch: 16/100, batch 342/563 completed (loss: 0.0313:  61%|██████    | 171/281 [17:19<11:13,  6.12s/it]
Training Epoch: 16/100, batch 342/563 completed (loss: 0.2532:  61%|██████    | 171/281 [17:19<11:13,  6.12s/it]
Training Epoch: 16/100, batch 342/563 completed (loss: 0.2532:  61%|██████    | 172/281 [17:22<11:03,  6.09s/it]
Training Epoch: 16/100, batch 342/563 completed (loss: 0.0313:  61%|██████    | 172/281 [17:22<11:04,  6.09s/it]
Training Epoch: 16/100, batch 342/563 completed (loss: 0.1749:  61%|██████    | 172/281 [17:22<11:04,  6.09s/it]
Training Epoch: 16/100, batch 342/563 completed (loss: 0.0065:  61%|██████    | 172/281 [17:22<11:04,  6.09s/it]
Training Epoch: 16/100, batch 343/563 completed (loss: 0.2015:  61%|██████    | 172/281 [17:22<11:04,  6.09s/it]
Training Epoch: 16/100, batch 343/563 completed (loss: 0.0344:  61%|██████    | 172/281 [17:22<11:04,  6.09s/it]
Training Epoch: 16/100, batch 343/563 completed (loss: 0.1667:  61%|██████    | 172/281 [17:22<11:04,  6.09s/it]
Training Epoch: 16/100, batch 343/563 completed (loss: 0.0531:  61%|██████    | 172/281 [17:22<11:03,  6.09s/it]
Training Epoch: 16/100, batch 344/563 completed (loss: 0.0288:  61%|██████    | 172/281 [17:25<11:04,  6.09s/it]
Training Epoch: 16/100, batch 344/563 completed (loss: 0.1063:  61%|██████    | 172/281 [17:25<11:04,  6.09s/it]
Training Epoch: 16/100, batch 344/563 completed (loss: 0.2529:  61%|██████    | 172/281 [17:25<11:04,  6.09s/it]
Training Epoch: 16/100, batch 344/563 completed (loss: 0.0777:  61%|██████    | 172/281 [17:25<11:03,  6.09s/it]
Training Epoch: 16/100, batch 344/563 completed (loss: 0.0288:  62%|██████▏   | 173/281 [17:28<10:54,  6.06s/it]
Training Epoch: 16/100, batch 344/563 completed (loss: 0.0777:  62%|██████▏   | 173/281 [17:28<10:54,  6.06s/it]
Training Epoch: 16/100, batch 344/563 completed (loss: 0.2529:  62%|██████▏   | 173/281 [17:28<10:54,  6.06s/it]
Training Epoch: 16/100, batch 344/563 completed (loss: 0.1063:  62%|██████▏   | 173/281 [17:28<10:55,  6.07s/it]
Training Epoch: 16/100, batch 345/563 completed (loss: 0.1454:  62%|██████▏   | 173/281 [17:28<10:54,  6.06s/it]
Training Epoch: 16/100, batch 345/563 completed (loss: 0.032:  62%|██████▏   | 173/281 [17:28<10:54,  6.06s/it] 
Training Epoch: 16/100, batch 345/563 completed (loss: 0.0547:  62%|██████▏   | 173/281 [17:28<10:55,  6.07s/it]
Training Epoch: 16/100, batch 345/563 completed (loss: 0.0241:  62%|██████▏   | 173/281 [17:28<10:54,  6.06s/it]
Training Epoch: 16/100, batch 346/563 completed (loss: 0.0334:  62%|██████▏   | 173/281 [17:31<10:54,  6.06s/it]
Training Epoch: 16/100, batch 346/563 completed (loss: 0.0022:  62%|██████▏   | 173/281 [17:31<10:54,  6.06s/it]
Training Epoch: 16/100, batch 346/563 completed (loss: 0.0131:  62%|██████▏   | 173/281 [17:31<10:55,  6.07s/it]
Training Epoch: 16/100, batch 346/563 completed (loss: 0.0724:  62%|██████▏   | 173/281 [17:31<10:54,  6.06s/it]
Training Epoch: 16/100, batch 346/563 completed (loss: 0.0724:  62%|██████▏   | 174/281 [17:34<10:51,  6.09s/it]
Training Epoch: 16/100, batch 346/563 completed (loss: 0.0334:  62%|██████▏   | 174/281 [17:34<10:52,  6.10s/it]
Training Epoch: 16/100, batch 346/563 completed (loss: 0.0022:  62%|██████▏   | 174/281 [17:34<10:52,  6.10s/it]
Training Epoch: 16/100, batch 346/563 completed (loss: 0.0131:  62%|██████▏   | 174/281 [17:34<10:52,  6.10s/it]
Training Epoch: 16/100, batch 347/563 completed (loss: 0.007:  62%|██████▏   | 174/281 [17:34<10:52,  6.10s/it] 
Training Epoch: 16/100, batch 347/563 completed (loss: 0.0349:  62%|██████▏   | 174/281 [17:34<10:52,  6.10s/it]
Training Epoch: 16/100, batch 347/563 completed (loss: 0.2548:  62%|██████▏   | 174/281 [17:34<10:52,  6.10s/it]
Training Epoch: 16/100, batch 347/563 completed (loss: 0.0024:  62%|██████▏   | 174/281 [17:34<10:51,  6.09s/it]
Training Epoch: 16/100, batch 348/563 completed (loss: 0.0578:  62%|██████▏   | 174/281 [17:37<10:52,  6.10s/it]
Training Epoch: 16/100, batch 348/563 completed (loss: 0.0082:  62%|██████▏   | 174/281 [17:37<10:52,  6.10s/it]
Training Epoch: 16/100, batch 348/563 completed (loss: 0.0394:  62%|██████▏   | 174/281 [17:37<10:52,  6.10s/it]
Training Epoch: 16/100, batch 348/563 completed (loss: 0.1137:  62%|██████▏   | 174/281 [17:37<10:51,  6.09s/it]
Training Epoch: 16/100, batch 348/563 completed (loss: 0.0394:  62%|██████▏   | 175/281 [17:40<10:48,  6.12s/it]
Training Epoch: 16/100, batch 348/563 completed (loss: 0.0082:  62%|██████▏   | 175/281 [17:40<10:48,  6.12s/it]
Training Epoch: 16/100, batch 348/563 completed (loss: 0.1137:  62%|██████▏   | 175/281 [17:40<10:48,  6.12s/it]
Training Epoch: 16/100, batch 348/563 completed (loss: 0.0578:  62%|██████▏   | 175/281 [17:40<10:48,  6.12s/it]
Training Epoch: 16/100, batch 349/563 completed (loss: 0.0613:  62%|██████▏   | 175/281 [17:40<10:48,  6.12s/it]
Training Epoch: 16/100, batch 349/563 completed (loss: 0.0443:  62%|██████▏   | 175/281 [17:40<10:48,  6.12s/it]
Training Epoch: 16/100, batch 349/563 completed (loss: 0.0631:  62%|██████▏   | 175/281 [17:40<10:48,  6.12s/it]
Training Epoch: 16/100, batch 349/563 completed (loss: 0.0118:  62%|██████▏   | 175/281 [17:40<10:48,  6.12s/it]
Training Epoch: 16/100, batch 350/563 completed (loss: 0.3271:  62%|██████▏   | 175/281 [17:43<10:48,  6.12s/it]
Training Epoch: 16/100, batch 350/563 completed (loss: 0.0086:  62%|██████▏   | 175/281 [17:43<10:48,  6.12s/it]
Training Epoch: 16/100, batch 350/563 completed (loss: 0.017:  62%|██████▏   | 175/281 [17:43<10:48,  6.12s/it] 
Training Epoch: 16/100, batch 350/563 completed (loss: 0.1505:  62%|██████▏   | 175/281 [17:43<10:48,  6.12s/it]
Training Epoch: 16/100, batch 350/563 completed (loss: 0.3271:  63%|██████▎   | 176/281 [17:46<10:43,  6.13s/it]
Training Epoch: 16/100, batch 350/563 completed (loss: 0.0086:  63%|██████▎   | 176/281 [17:46<10:42,  6.12s/it]
Training Epoch: 16/100, batch 350/563 completed (loss: 0.1505:  63%|██████▎   | 176/281 [17:46<10:43,  6.13s/it]
Training Epoch: 16/100, batch 350/563 completed (loss: 0.017:  63%|██████▎   | 176/281 [17:46<10:43,  6.13s/it]
Training Epoch: 16/100, batch 351/563 completed (loss: 0.044:  63%|██████▎   | 176/281 [17:46<10:42,  6.12s/it] 
Training Epoch: 16/100, batch 351/563 completed (loss: 0.138:  63%|██████▎   | 176/281 [17:46<10:43,  6.13s/it] 
Training Epoch: 16/100, batch 351/563 completed (loss: 0.0058:  63%|██████▎   | 176/281 [17:46<10:43,  6.13s/it]
Training Epoch: 16/100, batch 351/563 completed (loss: 0.0259:  63%|██████▎   | 176/281 [17:46<10:43,  6.13s/it]
Training Epoch: 16/100, batch 352/563 completed (loss: 0.0174:  63%|██████▎   | 176/281 [17:49<10:42,  6.12s/it]
Training Epoch: 16/100, batch 352/563 completed (loss: 0.0243:  63%|██████▎   | 176/281 [17:49<10:43,  6.13s/it]
Training Epoch: 16/100, batch 352/563 completed (loss: 0.0416:  63%|██████▎   | 176/281 [17:49<10:43,  6.13s/it]
Training Epoch: 16/100, batch 352/563 completed (loss: 0.0269:  63%|██████▎   | 176/281 [17:49<10:43,  6.13s/it]
Training Epoch: 16/100, batch 352/563 completed (loss: 0.0243:  63%|██████▎   | 177/281 [17:52<10:28,  6.05s/it]
Training Epoch: 16/100, batch 352/563 completed (loss: 0.0174:  63%|██████▎   | 177/281 [17:52<10:28,  6.05s/it]
Training Epoch: 16/100, batch 352/563 completed (loss: 0.0269:  63%|██████▎   | 177/281 [17:52<10:29,  6.05s/it]
Training Epoch: 16/100, batch 352/563 completed (loss: 0.0416:  63%|██████▎   | 177/281 [17:52<10:29,  6.05s/it]
Training Epoch: 16/100, batch 353/563 completed (loss: 0.0521:  63%|██████▎   | 177/281 [17:52<10:28,  6.05s/it]
Training Epoch: 16/100, batch 353/563 completed (loss: 0.0185:  63%|██████▎   | 177/281 [17:52<10:28,  6.05s/it]
Training Epoch: 16/100, batch 353/563 completed (loss: 0.1892:  63%|██████▎   | 177/281 [17:52<10:29,  6.05s/it]
Training Epoch: 16/100, batch 353/563 completed (loss: 0.0078:  63%|██████▎   | 177/281 [17:52<10:29,  6.05s/it]
Training Epoch: 16/100, batch 354/563 completed (loss: 0.1794:  63%|██████▎   | 177/281 [17:55<10:29,  6.05s/it]
Training Epoch: 16/100, batch 354/563 completed (loss: 0.0214:  63%|██████▎   | 177/281 [17:55<10:28,  6.05s/it]
Training Epoch: 16/100, batch 354/563 completed (loss: 0.016:  63%|██████▎   | 177/281 [17:55<10:28,  6.05s/it] 
Training Epoch: 16/100, batch 354/563 completed (loss: 0.0357:  63%|██████▎   | 177/281 [17:55<10:29,  6.05s/it]
Training Epoch: 16/100, batch 354/563 completed (loss: 0.0357:  63%|██████▎   | 178/281 [17:58<10:25,  6.07s/it]
Training Epoch: 16/100, batch 354/563 completed (loss: 0.0214:  63%|██████▎   | 178/281 [17:58<10:25,  6.07s/it]
Training Epoch: 16/100, batch 354/563 completed (loss: 0.016:  63%|██████▎   | 178/281 [17:58<10:25,  6.08s/it]
Training Epoch: 16/100, batch 354/563 completed (loss: 0.1794:  63%|██████▎   | 178/281 [17:58<10:25,  6.08s/it]
Training Epoch: 16/100, batch 355/563 completed (loss: 0.0127:  63%|██████▎   | 178/281 [17:58<10:25,  6.07s/it]
Training Epoch: 16/100, batch 355/563 completed (loss: 0.1735:  63%|██████▎   | 178/281 [17:58<10:25,  6.08s/it]
Training Epoch: 16/100, batch 355/563 completed (loss: 0.0265:  63%|██████▎   | 178/281 [17:58<10:25,  6.08s/it]
Training Epoch: 16/100, batch 355/563 completed (loss: 0.0085:  63%|██████▎   | 178/281 [17:58<10:25,  6.07s/it]
Training Epoch: 16/100, batch 356/563 completed (loss: 0.0435:  63%|██████▎   | 178/281 [18:01<10:25,  6.07s/it]
Training Epoch: 16/100, batch 356/563 completed (loss: 0.0774:  63%|██████▎   | 178/281 [18:01<10:25,  6.08s/it]
Training Epoch: 16/100, batch 356/563 completed (loss: 0.0352:  63%|██████▎   | 178/281 [18:01<10:25,  6.08s/it]
Training Epoch: 16/100, batch 356/563 completed (loss: 0.1123:  63%|██████▎   | 178/281 [18:01<10:25,  6.07s/it]
Training Epoch: 16/100, batch 356/563 completed (loss: 0.1123:  64%|██████▎   | 179/281 [18:04<10:22,  6.11s/it]
Training Epoch: 16/100, batch 356/563 completed (loss: 0.0352:  64%|██████▎   | 179/281 [18:04<10:22,  6.11s/it]
Training Epoch: 16/100, batch 356/563 completed (loss: 0.0774:  64%|██████▎   | 179/281 [18:04<10:22,  6.10s/it]
Training Epoch: 16/100, batch 356/563 completed (loss: 0.0435:  64%|██████▎   | 179/281 [18:04<10:22,  6.11s/it]
Training Epoch: 16/100, batch 357/563 completed (loss: 0.0102:  64%|██████▎   | 179/281 [18:04<10:22,  6.11s/it]
Training Epoch: 16/100, batch 357/563 completed (loss: 0.0632:  64%|██████▎   | 179/281 [18:04<10:22,  6.11s/it]
Training Epoch: 16/100, batch 357/563 completed (loss: 0.0028:  64%|██████▎   | 179/281 [18:04<10:22,  6.10s/it]
Training Epoch: 16/100, batch 357/563 completed (loss: 0.0055:  64%|██████▎   | 179/281 [18:05<10:22,  6.11s/it]
Training Epoch: 16/100, batch 358/563 completed (loss: 0.2544:  64%|██████▎   | 179/281 [18:07<10:22,  6.11s/it]
Training Epoch: 16/100, batch 358/563 completed (loss: 0.0222:  64%|██████▎   | 179/281 [18:07<10:22,  6.10s/it]
Training Epoch: 16/100, batch 358/563 completed (loss: 0.1728:  64%|██████▎   | 179/281 [18:07<10:22,  6.11s/it]
Training Epoch: 16/100, batch 358/563 completed (loss: 0.0052:  64%|██████▎   | 179/281 [18:07<10:22,  6.11s/it]
Training Epoch: 16/100, batch 358/563 completed (loss: 0.1728:  64%|██████▍   | 180/281 [18:11<10:18,  6.13s/it]
Training Epoch: 16/100, batch 358/563 completed (loss: 0.2544:  64%|██████▍   | 180/281 [18:11<10:18,  6.12s/it]
Training Epoch: 16/100, batch 358/563 completed (loss: 0.0222:  64%|██████▍   | 180/281 [18:11<10:18,  6.13s/it]
Training Epoch: 16/100, batch 358/563 completed (loss: 0.0052:  64%|██████▍   | 180/281 [18:11<10:19,  6.13s/it]
Training Epoch: 16/100, batch 359/563 completed (loss: 0.0124:  64%|██████▍   | 180/281 [18:11<10:18,  6.12s/it]
Training Epoch: 16/100, batch 359/563 completed (loss: 0.0492:  64%|██████▍   | 180/281 [18:11<10:18,  6.13s/it]
Training Epoch: 16/100, batch 359/563 completed (loss: 0.0533:  64%|██████▍   | 180/281 [18:11<10:18,  6.13s/it]
Training Epoch: 16/100, batch 359/563 completed (loss: 0.0948:  64%|██████▍   | 180/281 [18:11<10:19,  6.13s/it]
Training Epoch: 16/100, batch 360/563 completed (loss: 0.0506:  64%|██████▍   | 180/281 [18:14<10:18,  6.13s/it]
Training Epoch: 16/100, batch 360/563 completed (loss: 0.3076:  64%|██████▍   | 180/281 [18:14<10:18,  6.12s/it]
Training Epoch: 16/100, batch 360/563 completed (loss: 0.2082:  64%|██████▍   | 180/281 [18:14<10:18,  6.13s/it]
Training Epoch: 16/100, batch 360/563 completed (loss: 0.0023:  64%|██████▍   | 180/281 [18:14<10:19,  6.13s/it]
Training Epoch: 16/100, batch 360/563 completed (loss: 0.2082:  64%|██████▍   | 181/281 [18:17<10:13,  6.14s/it]
Training Epoch: 16/100, batch 360/563 completed (loss: 0.0023:  64%|██████▍   | 181/281 [18:17<10:13,  6.14s/it]
Training Epoch: 16/100, batch 360/563 completed (loss: 0.0506:  64%|██████▍   | 181/281 [18:17<10:13,  6.14s/it]
Training Epoch: 16/100, batch 360/563 completed (loss: 0.3076:  64%|██████▍   | 181/281 [18:17<10:14,  6.14s/it]
Training Epoch: 16/100, batch 361/563 completed (loss: 0.041:  64%|██████▍   | 181/281 [18:17<10:14,  6.14s/it] 
Training Epoch: 16/100, batch 361/563 completed (loss: 0.1181:  64%|██████▍   | 181/281 [18:17<10:13,  6.14s/it]
Training Epoch: 16/100, batch 361/563 completed (loss: 0.0078:  64%|██████▍   | 181/281 [18:17<10:13,  6.14s/it]
Training Epoch: 16/100, batch 361/563 completed (loss: 0.0256:  64%|██████▍   | 181/281 [18:17<10:13,  6.14s/it]
Training Epoch: 16/100, batch 362/563 completed (loss: 0.0142:  64%|██████▍   | 181/281 [18:20<10:13,  6.14s/it]
Training Epoch: 16/100, batch 362/563 completed (loss: 0.3525:  64%|██████▍   | 181/281 [18:20<10:14,  6.14s/it]
Training Epoch: 16/100, batch 362/563 completed (loss: 0.0241:  64%|██████▍   | 181/281 [18:20<10:13,  6.14s/it]
Training Epoch: 16/100, batch 362/563 completed (loss: 0.0527:  64%|██████▍   | 181/281 [18:20<10:13,  6.14s/it]
Training Epoch: 16/100, batch 362/563 completed (loss: 0.0142:  65%|██████▍   | 182/281 [18:23<10:00,  6.06s/it]
Training Epoch: 16/100, batch 362/563 completed (loss: 0.0527:  65%|██████▍   | 182/281 [18:23<10:00,  6.06s/it]
Training Epoch: 16/100, batch 362/563 completed (loss: 0.0241:  65%|██████▍   | 182/281 [18:23<10:00,  6.07s/it]
Training Epoch: 16/100, batch 362/563 completed (loss: 0.3525:  65%|██████▍   | 182/281 [18:23<10:00,  6.07s/it]
Training Epoch: 16/100, batch 363/563 completed (loss: 0.1363:  65%|██████▍   | 182/281 [18:23<10:00,  6.07s/it]
Training Epoch: 16/100, batch 363/563 completed (loss: 0.1906:  65%|██████▍   | 182/281 [18:23<10:00,  6.06s/it]
Training Epoch: 16/100, batch 363/563 completed (loss: 0.0607:  65%|██████▍   | 182/281 [18:23<10:00,  6.07s/it]
Training Epoch: 16/100, batch 363/563 completed (loss: 0.061:  65%|██████▍   | 182/281 [18:23<10:00,  6.06s/it] 
Training Epoch: 16/100, batch 364/563 completed (loss: 0.0114:  65%|██████▍   | 182/281 [18:26<10:00,  6.06s/it]
Training Epoch: 16/100, batch 364/563 completed (loss: 0.0108:  65%|██████▍   | 182/281 [18:26<10:00,  6.07s/it]
Training Epoch: 16/100, batch 364/563 completed (loss: 0.1497:  65%|██████▍   | 182/281 [18:26<10:00,  6.07s/it]
Training Epoch: 16/100, batch 364/563 completed (loss: 0.0505:  65%|██████▍   | 182/281 [18:26<10:00,  6.06s/it]
Training Epoch: 16/100, batch 364/563 completed (loss: 0.0505:  65%|██████▌   | 183/281 [18:29<09:57,  6.10s/it]
Training Epoch: 16/100, batch 364/563 completed (loss: 0.0114:  65%|██████▌   | 183/281 [18:29<09:57,  6.10s/it]
Training Epoch: 16/100, batch 364/563 completed (loss: 0.1497:  65%|██████▌   | 183/281 [18:29<09:57,  6.10s/it]
Training Epoch: 16/100, batch 364/563 completed (loss: 0.0108:  65%|██████▌   | 183/281 [18:29<09:57,  6.10s/it]
Training Epoch: 16/100, batch 365/563 completed (loss: 0.0033:  65%|██████▌   | 183/281 [18:29<09:57,  6.10s/it]
Training Epoch: 16/100, batch 365/563 completed (loss: 0.0134:  65%|██████▌   | 183/281 [18:29<09:57,  6.10s/it]
Training Epoch: 16/100, batch 365/563 completed (loss: 0.038:  65%|██████▌   | 183/281 [18:29<09:57,  6.10s/it] 
Training Epoch: 16/100, batch 365/563 completed (loss: 0.3515:  65%|██████▌   | 183/281 [18:29<09:57,  6.10s/it]
Training Epoch: 16/100, batch 366/563 completed (loss: 0.2196:  65%|██████▌   | 183/281 [18:32<09:57,  6.10s/it]
Training Epoch: 16/100, batch 366/563 completed (loss: 0.1025:  65%|██████▌   | 183/281 [18:32<09:57,  6.10s/it]
Training Epoch: 16/100, batch 366/563 completed (loss: 0.0395:  65%|██████▌   | 183/281 [18:32<09:57,  6.10s/it]
Training Epoch: 16/100, batch 366/563 completed (loss: 0.0226:  65%|██████▌   | 183/281 [18:32<09:57,  6.10s/it]
Training Epoch: 16/100, batch 366/563 completed (loss: 0.0226:  65%|██████▌   | 184/281 [18:35<09:53,  6.12s/it]
Training Epoch: 16/100, batch 366/563 completed (loss: 0.0395:  65%|██████▌   | 184/281 [18:35<09:53,  6.12s/it]
Training Epoch: 16/100, batch 366/563 completed (loss: 0.2196:  65%|██████▌   | 184/281 [18:35<09:53,  6.12s/it]
Training Epoch: 16/100, batch 366/563 completed (loss: 0.1025:  65%|██████▌   | 184/281 [18:35<09:53,  6.12s/it]
Training Epoch: 16/100, batch 367/563 completed (loss: 0.3703:  65%|██████▌   | 184/281 [18:35<09:53,  6.12s/it]
Training Epoch: 16/100, batch 367/563 completed (loss: 0.0074:  65%|██████▌   | 184/281 [18:35<09:53,  6.12s/it]
Training Epoch: 16/100, batch 367/563 completed (loss: 0.0164:  65%|██████▌   | 184/281 [18:35<09:53,  6.12s/it]
Training Epoch: 16/100, batch 367/563 completed (loss: 0.0363:  65%|██████▌   | 184/281 [18:35<09:53,  6.12s/it]
Training Epoch: 16/100, batch 368/563 completed (loss: 0.0563:  65%|██████▌   | 184/281 [18:38<09:53,  6.12s/it]
Training Epoch: 16/100, batch 368/563 completed (loss: 0.0135:  65%|██████▌   | 184/281 [18:38<09:53,  6.12s/it]
Training Epoch: 16/100, batch 368/563 completed (loss: 0.01:  65%|██████▌   | 184/281 [18:38<09:53,  6.12s/it]  
Training Epoch: 16/100, batch 368/563 completed (loss: 0.0136:  65%|██████▌   | 184/281 [18:38<09:53,  6.12s/it]
Training Epoch: 16/100, batch 368/563 completed (loss: 0.0563:  66%|██████▌   | 185/281 [18:41<09:44,  6.09s/it]
Training Epoch: 16/100, batch 368/563 completed (loss: 0.0136:  66%|██████▌   | 185/281 [18:41<09:44,  6.09s/it]
Training Epoch: 16/100, batch 368/563 completed (loss: 0.0135:  66%|██████▌   | 185/281 [18:41<09:44,  6.09s/it]
Training Epoch: 16/100, batch 368/563 completed (loss: 0.01:  66%|██████▌   | 185/281 [18:41<09:44,  6.09s/it]
Training Epoch: 16/100, batch 369/563 completed (loss: 0.0153:  66%|██████▌   | 185/281 [18:41<09:44,  6.09s/it]
Training Epoch: 16/100, batch 369/563 completed (loss: 0.0405:  66%|██████▌   | 185/281 [18:41<09:44,  6.09s/it]
Training Epoch: 16/100, batch 369/563 completed (loss: 0.0398:  66%|██████▌   | 185/281 [18:41<09:44,  6.09s/it]
Training Epoch: 16/100, batch 369/563 completed (loss: 0.0197:  66%|██████▌   | 185/281 [18:41<09:44,  6.09s/it]
Training Epoch: 16/100, batch 370/563 completed (loss: 0.0114:  66%|██████▌   | 185/281 [18:44<09:44,  6.09s/it]
Training Epoch: 16/100, batch 370/563 completed (loss: 0.1941:  66%|██████▌   | 185/281 [18:44<09:44,  6.09s/it]
Training Epoch: 16/100, batch 370/563 completed (loss: 0.044:  66%|██████▌   | 185/281 [18:44<09:44,  6.09s/it] 
Training Epoch: 16/100, batch 370/563 completed (loss: 0.0248:  66%|██████▌   | 185/281 [18:44<09:44,  6.09s/it]
Training Epoch: 16/100, batch 370/563 completed (loss: 0.044:  66%|██████▌   | 186/281 [18:47<09:36,  6.06s/it]
Training Epoch: 16/100, batch 370/563 completed (loss: 0.0114:  66%|██████▌   | 186/281 [18:47<09:35,  6.06s/it]
Training Epoch: 16/100, batch 370/563 completed (loss: 0.0248:  66%|██████▌   | 186/281 [18:47<09:36,  6.07s/it]
Training Epoch: 16/100, batch 370/563 completed (loss: 0.1941:  66%|██████▌   | 186/281 [18:47<09:36,  6.07s/it]
Training Epoch: 16/100, batch 371/563 completed (loss: 0.048:  66%|██████▌   | 186/281 [18:47<09:36,  6.06s/it]
Training Epoch: 16/100, batch 371/563 completed (loss: 0.0228:  66%|██████▌   | 186/281 [18:47<09:36,  6.07s/it]
Training Epoch: 16/100, batch 371/563 completed (loss: 0.0484:  66%|██████▌   | 186/281 [18:47<09:35,  6.06s/it]
Training Epoch: 16/100, batch 371/563 completed (loss: 0.1015:  66%|██████▌   | 186/281 [18:47<09:36,  6.07s/it]
Training Epoch: 16/100, batch 372/563 completed (loss: 0.0056:  66%|██████▌   | 186/281 [18:50<09:36,  6.07s/it]
Training Epoch: 16/100, batch 372/563 completed (loss: 0.0267:  66%|██████▌   | 186/281 [18:50<09:36,  6.06s/it]
Training Epoch: 16/100, batch 372/563 completed (loss: 0.0468:  66%|██████▌   | 186/281 [18:50<09:35,  6.06s/it]
Training Epoch: 16/100, batch 372/563 completed (loss: 0.0235:  66%|██████▌   | 186/281 [18:50<09:36,  6.07s/it]
Training Epoch: 16/100, batch 372/563 completed (loss: 0.0468:  67%|██████▋   | 187/281 [18:53<09:32,  6.09s/it]
Training Epoch: 16/100, batch 372/563 completed (loss: 0.0267:  67%|██████▋   | 187/281 [18:53<09:32,  6.09s/it]
Training Epoch: 16/100, batch 372/563 completed (loss: 0.0235:  67%|██████▋   | 187/281 [18:53<09:32,  6.09s/it]
Training Epoch: 16/100, batch 372/563 completed (loss: 0.0056:  67%|██████▋   | 187/281 [18:53<09:32,  6.09s/it]
Training Epoch: 16/100, batch 373/563 completed (loss: 0.161:  67%|██████▋   | 187/281 [18:53<09:32,  6.09s/it] 
Training Epoch: 16/100, batch 373/563 completed (loss: 0.0077:  67%|██████▋   | 187/281 [18:53<09:32,  6.09s/it]
Training Epoch: 16/100, batch 373/563 completed (loss: 0.0659:  67%|██████▋   | 187/281 [18:53<09:32,  6.09s/it]
Training Epoch: 16/100, batch 373/563 completed (loss: 0.1365:  67%|██████▋   | 187/281 [18:53<09:32,  6.09s/it]
Training Epoch: 16/100, batch 374/563 completed (loss: 0.0118:  67%|██████▋   | 187/281 [18:56<09:32,  6.09s/it]
Training Epoch: 16/100, batch 374/563 completed (loss: 0.0186:  67%|██████▋   | 187/281 [18:56<09:32,  6.09s/it]
Training Epoch: 16/100, batch 374/563 completed (loss: 0.209:  67%|██████▋   | 187/281 [18:56<09:32,  6.09s/it] 
Training Epoch: 16/100, batch 374/563 completed (loss: 0.0449:  67%|██████▋   | 187/281 [18:56<09:32,  6.09s/it]
Training Epoch: 16/100, batch 374/563 completed (loss: 0.0186:  67%|██████▋   | 188/281 [18:59<09:28,  6.11s/it]
Training Epoch: 16/100, batch 374/563 completed (loss: 0.209:  67%|██████▋   | 188/281 [18:59<09:28,  6.11s/it]
Training Epoch: 16/100, batch 374/563 completed (loss: 0.0449:  67%|██████▋   | 188/281 [18:59<09:28,  6.11s/it]
Training Epoch: 16/100, batch 374/563 completed (loss: 0.0118:  67%|██████▋   | 188/281 [18:59<09:28,  6.11s/it]
Training Epoch: 16/100, batch 375/563 completed (loss: 0.2972:  67%|██████▋   | 188/281 [18:59<09:28,  6.11s/it]
Training Epoch: 16/100, batch 375/563 completed (loss: 0.0075:  67%|██████▋   | 188/281 [18:59<09:28,  6.11s/it]
Training Epoch: 16/100, batch 375/563 completed (loss: 0.0687:  67%|██████▋   | 188/281 [18:59<09:28,  6.11s/it]
Training Epoch: 16/100, batch 375/563 completed (loss: 0.0253:  67%|██████▋   | 188/281 [18:59<09:28,  6.11s/it]
Training Epoch: 16/100, batch 376/563 completed (loss: 0.0816:  67%|██████▋   | 188/281 [19:02<09:28,  6.11s/it]
Training Epoch: 16/100, batch 376/563 completed (loss: 0.0121:  67%|██████▋   | 188/281 [19:02<09:28,  6.11s/it]
Training Epoch: 16/100, batch 376/563 completed (loss: 0.1881:  67%|██████▋   | 188/281 [19:02<09:28,  6.11s/it]
Training Epoch: 16/100, batch 376/563 completed (loss: 0.0472:  67%|██████▋   | 188/281 [19:02<09:28,  6.11s/it]
Training Epoch: 16/100, batch 376/563 completed (loss: 0.0816:  67%|██████▋   | 189/281 [19:05<09:19,  6.08s/it]
Training Epoch: 16/100, batch 376/563 completed (loss: 0.1881:  67%|██████▋   | 189/281 [19:05<09:19,  6.08s/it]
Training Epoch: 16/100, batch 376/563 completed (loss: 0.0472:  67%|██████▋   | 189/281 [19:05<09:19,  6.08s/it]
Training Epoch: 16/100, batch 376/563 completed (loss: 0.0121:  67%|██████▋   | 189/281 [19:05<09:19,  6.08s/it]
Training Epoch: 16/100, batch 377/563 completed (loss: 0.1504:  67%|██████▋   | 189/281 [19:05<09:19,  6.08s/it]
Training Epoch: 16/100, batch 377/563 completed (loss: 0.0389:  67%|██████▋   | 189/281 [19:05<09:19,  6.08s/it]
Training Epoch: 16/100, batch 377/563 completed (loss: 0.0122:  67%|██████▋   | 189/281 [19:05<09:19,  6.08s/it]
Training Epoch: 16/100, batch 377/563 completed (loss: 0.0031:  67%|██████▋   | 189/281 [19:05<09:19,  6.08s/it]
Training Epoch: 16/100, batch 378/563 completed (loss: 0.0211:  67%|██████▋   | 189/281 [19:08<09:19,  6.08s/it]
Training Epoch: 16/100, batch 378/563 completed (loss: 0.0081:  67%|██████▋   | 189/281 [19:08<09:19,  6.08s/it]
Training Epoch: 16/100, batch 378/563 completed (loss: 0.03:  67%|██████▋   | 189/281 [19:08<09:19,  6.08s/it]  
Training Epoch: 16/100, batch 378/563 completed (loss: 0.2122:  67%|██████▋   | 189/281 [19:09<09:19,  6.08s/it]
Training Epoch: 16/100, batch 378/563 completed (loss: 0.2122:  68%|██████▊   | 190/281 [19:11<09:11,  6.06s/it]
Training Epoch: 16/100, batch 378/563 completed (loss: 0.0081:  68%|██████▊   | 190/281 [19:11<09:12,  6.07s/it]
Training Epoch: 16/100, batch 378/563 completed (loss: 0.03:  68%|██████▊   | 190/281 [19:11<09:12,  6.07s/it]
Training Epoch: 16/100, batch 378/563 completed (loss: 0.0211:  68%|██████▊   | 190/281 [19:11<09:12,  6.07s/it]
Training Epoch: 16/100, batch 379/563 completed (loss: 0.0723:  68%|██████▊   | 190/281 [19:11<09:12,  6.07s/it]
Training Epoch: 16/100, batch 379/563 completed (loss: 0.0016:  68%|██████▊   | 190/281 [19:11<09:12,  6.07s/it]
Training Epoch: 16/100, batch 379/563 completed (loss: 0.0084:  68%|██████▊   | 190/281 [19:11<09:12,  6.07s/it]
Training Epoch: 16/100, batch 379/563 completed (loss: 0.0202:  68%|██████▊   | 190/281 [19:11<09:11,  6.06s/it]
Training Epoch: 16/100, batch 380/563 completed (loss: 0.2952:  68%|██████▊   | 190/281 [19:15<09:12,  6.07s/it]
Training Epoch: 16/100, batch 380/563 completed (loss: 0.0421:  68%|██████▊   | 190/281 [19:15<09:12,  6.07s/it]
Training Epoch: 16/100, batch 380/563 completed (loss: 0.0692:  68%|██████▊   | 190/281 [19:15<09:12,  6.07s/it]
Training Epoch: 16/100, batch 380/563 completed (loss: 0.0332:  68%|██████▊   | 190/281 [19:15<09:11,  6.06s/it]
Training Epoch: 16/100, batch 380/563 completed (loss: 0.0421:  68%|██████▊   | 191/281 [19:18<09:08,  6.10s/it]
Training Epoch: 16/100, batch 380/563 completed (loss: 0.0332:  68%|██████▊   | 191/281 [19:18<09:08,  6.10s/it]
Training Epoch: 16/100, batch 380/563 completed (loss: 0.0692:  68%|██████▊   | 191/281 [19:18<09:08,  6.10s/it]
Training Epoch: 16/100, batch 380/563 completed (loss: 0.2952:  68%|██████▊   | 191/281 [19:18<09:08,  6.10s/it]
Training Epoch: 16/100, batch 381/563 completed (loss: 0.0039:  68%|██████▊   | 191/281 [19:18<09:08,  6.10s/it]
Training Epoch: 16/100, batch 381/563 completed (loss: 0.1118:  68%|██████▊   | 191/281 [19:18<09:08,  6.10s/it]
Training Epoch: 16/100, batch 381/563 completed (loss: 0.0093:  68%|██████▊   | 191/281 [19:18<09:08,  6.10s/it]
Training Epoch: 16/100, batch 381/563 completed (loss: 0.074:  68%|██████▊   | 191/281 [19:18<09:08,  6.10s/it] 
Training Epoch: 16/100, batch 382/563 completed (loss: 0.0066:  68%|██████▊   | 191/281 [19:21<09:08,  6.10s/it]
Training Epoch: 16/100, batch 382/563 completed (loss: 0.0451:  68%|██████▊   | 191/281 [19:21<09:08,  6.10s/it]
Training Epoch: 16/100, batch 382/563 completed (loss: 0.0307:  68%|██████▊   | 191/281 [19:21<09:08,  6.10s/it]
Training Epoch: 16/100, batch 382/563 completed (loss: 0.0424:  68%|██████▊   | 191/281 [19:21<09:08,  6.10s/it]
Training Epoch: 16/100, batch 382/563 completed (loss: 0.0451:  68%|██████▊   | 192/281 [19:24<09:05,  6.13s/it]
Training Epoch: 16/100, batch 382/563 completed (loss: 0.0307:  68%|██████▊   | 192/281 [19:24<09:05,  6.12s/it]
Training Epoch: 16/100, batch 382/563 completed (loss: 0.0424:  68%|██████▊   | 192/281 [19:24<09:05,  6.13s/it]
Training Epoch: 16/100, batch 382/563 completed (loss: 0.0066:  68%|██████▊   | 192/281 [19:24<09:05,  6.13s/it]
Training Epoch: 16/100, batch 383/563 completed (loss: 0.1401:  68%|██████▊   | 192/281 [19:24<09:05,  6.13s/it]
Training Epoch: 16/100, batch 383/563 completed (loss: 0.0712:  68%|██████▊   | 192/281 [19:24<09:05,  6.13s/it]
Training Epoch: 16/100, batch 383/563 completed (loss: 0.3894:  68%|██████▊   | 192/281 [19:24<09:05,  6.12s/it]
Training Epoch: 16/100, batch 383/563 completed (loss: 0.0802:  68%|██████▊   | 192/281 [19:24<09:05,  6.13s/it]
Training Epoch: 16/100, batch 384/563 completed (loss: 0.0455:  68%|██████▊   | 192/281 [19:27<09:05,  6.12s/it]
Training Epoch: 16/100, batch 384/563 completed (loss: 0.0285:  68%|██████▊   | 192/281 [19:27<09:05,  6.13s/it]
Training Epoch: 16/100, batch 384/563 completed (loss: 0.3436:  68%|██████▊   | 192/281 [19:27<09:05,  6.13s/it]
Training Epoch: 16/100, batch 384/563 completed (loss: 0.0069:  68%|██████▊   | 192/281 [19:27<09:05,  6.13s/it]
Training Epoch: 16/100, batch 384/563 completed (loss: 0.0069:  69%|██████▊   | 193/281 [19:30<08:56,  6.10s/it]
Training Epoch: 16/100, batch 384/563 completed (loss: 0.0455:  69%|██████▊   | 193/281 [19:30<08:56,  6.10s/it]
Training Epoch: 16/100, batch 384/563 completed (loss: 0.0285:  69%|██████▊   | 193/281 [19:30<08:56,  6.10s/it]
Training Epoch: 16/100, batch 384/563 completed (loss: 0.3436:  69%|██████▊   | 193/281 [19:30<08:56,  6.10s/it]
Training Epoch: 16/100, batch 385/563 completed (loss: 0.0453:  69%|██████▊   | 193/281 [19:30<08:56,  6.10s/it]
Training Epoch: 16/100, batch 385/563 completed (loss: 0.041:  69%|██████▊   | 193/281 [19:30<08:56,  6.10s/it] 
Training Epoch: 16/100, batch 385/563 completed (loss: 0.1271:  69%|██████▊   | 193/281 [19:30<08:56,  6.10s/it]
Training Epoch: 16/100, batch 385/563 completed (loss: 0.0355:  69%|██████▊   | 193/281 [19:30<08:56,  6.10s/it]
Training Epoch: 16/100, batch 386/563 completed (loss: 0.2445:  69%|██████▊   | 193/281 [19:33<08:56,  6.10s/it]
Training Epoch: 16/100, batch 386/563 completed (loss: 0.0531:  69%|██████▊   | 193/281 [19:33<08:56,  6.10s/it]
Training Epoch: 16/100, batch 386/563 completed (loss: 0.0265:  69%|██████▊   | 193/281 [19:33<08:56,  6.10s/it]
Training Epoch: 16/100, batch 386/563 completed (loss: 0.0438:  69%|██████▊   | 193/281 [19:33<08:56,  6.10s/it]
Training Epoch: 16/100, batch 386/563 completed (loss: 0.0438:  69%|██████▉   | 194/281 [19:36<08:48,  6.08s/it]
Training Epoch: 16/100, batch 386/563 completed (loss: 0.2445:  69%|██████▉   | 194/281 [19:36<08:48,  6.08s/it]
Training Epoch: 16/100, batch 386/563 completed (loss: 0.0265:  69%|██████▉   | 194/281 [19:36<08:48,  6.08s/it]
Training Epoch: 16/100, batch 386/563 completed (loss: 0.0531:  69%|██████▉   | 194/281 [19:36<08:49,  6.09s/it]
Training Epoch: 16/100, batch 387/563 completed (loss: 0.0111:  69%|██████▉   | 194/281 [19:36<08:49,  6.09s/it]
Training Epoch: 16/100, batch 387/563 completed (loss: 0.0226:  69%|██████▉   | 194/281 [19:36<08:48,  6.08s/it]
Training Epoch: 16/100, batch 387/563 completed (loss: 0.5194:  69%|██████▉   | 194/281 [19:36<08:48,  6.08s/it]
Training Epoch: 16/100, batch 387/563 completed (loss: 0.5266:  69%|██████▉   | 194/281 [19:36<08:48,  6.08s/it]
Training Epoch: 16/100, batch 388/563 completed (loss: 0.0239:  69%|██████▉   | 194/281 [19:39<08:48,  6.08s/it]
Training Epoch: 16/100, batch 388/563 completed (loss: 0.0845:  69%|██████▉   | 194/281 [19:39<08:48,  6.08s/it]
Training Epoch: 16/100, batch 388/563 completed (loss: 0.063:  69%|██████▉   | 194/281 [19:39<08:49,  6.09s/it] 
Training Epoch: 16/100, batch 388/563 completed (loss: 0.014:  69%|██████▉   | 194/281 [19:39<08:48,  6.08s/it] 
Training Epoch: 16/100, batch 388/563 completed (loss: 0.0845:  69%|██████▉   | 195/281 [19:42<08:45,  6.11s/it]
Training Epoch: 16/100, batch 388/563 completed (loss: 0.014:  69%|██████▉   | 195/281 [19:42<08:45,  6.11s/it]
Training Epoch: 16/100, batch 388/563 completed (loss: 0.0239:  69%|██████▉   | 195/281 [19:42<08:45,  6.11s/it]
Training Epoch: 16/100, batch 388/563 completed (loss: 0.063:  69%|██████▉   | 195/281 [19:42<08:46,  6.12s/it]
Training Epoch: 16/100, batch 389/563 completed (loss: 0.0547:  69%|██████▉   | 195/281 [19:42<08:45,  6.11s/it]
Training Epoch: 16/100, batch 389/563 completed (loss: 0.0126:  69%|██████▉   | 195/281 [19:42<08:45,  6.11s/it]
Training Epoch: 16/100, batch 389/563 completed (loss: 0.0992:  69%|██████▉   | 195/281 [19:42<08:46,  6.12s/it]
Training Epoch: 16/100, batch 389/563 completed (loss: 0.1914:  69%|██████▉   | 195/281 [19:42<08:45,  6.11s/it]
Training Epoch: 16/100, batch 390/563 completed (loss: 0.0164:  69%|██████▉   | 195/281 [19:45<08:45,  6.11s/it]
Training Epoch: 16/100, batch 390/563 completed (loss: 0.0249:  69%|██████▉   | 195/281 [19:45<08:45,  6.11s/it]
Training Epoch: 16/100, batch 390/563 completed (loss: 0.0101:  69%|██████▉   | 195/281 [19:45<08:46,  6.12s/it]
Training Epoch: 16/100, batch 390/563 completed (loss: 0.0888:  69%|██████▉   | 195/281 [19:45<08:45,  6.11s/it]
Training Epoch: 16/100, batch 390/563 completed (loss: 0.0249:  70%|██████▉   | 196/281 [19:48<08:40,  6.13s/it]
Training Epoch: 16/100, batch 390/563 completed (loss: 0.0101:  70%|██████▉   | 196/281 [19:48<08:40,  6.13s/it]
Training Epoch: 16/100, batch 390/563 completed (loss: 0.0888:  70%|██████▉   | 196/281 [19:48<08:41,  6.13s/it]
Training Epoch: 16/100, batch 390/563 completed (loss: 0.0164:  70%|██████▉   | 196/281 [19:48<08:41,  6.13s/it]
Training Epoch: 16/100, batch 391/563 completed (loss: 0.1357:  70%|██████▉   | 196/281 [19:48<08:40,  6.13s/it]
Training Epoch: 16/100, batch 391/563 completed (loss: 0.0052:  70%|██████▉   | 196/281 [19:48<08:40,  6.13s/it]
Training Epoch: 16/100, batch 391/563 completed (loss: 0.111:  70%|██████▉   | 196/281 [19:48<08:41,  6.13s/it] 
Training Epoch: 16/100, batch 391/563 completed (loss: 0.0069:  70%|██████▉   | 196/281 [19:48<08:41,  6.13s/it]
Training Epoch: 16/100, batch 392/563 completed (loss: 0.0415:  70%|██████▉   | 196/281 [19:51<08:40,  6.13s/it]
Training Epoch: 16/100, batch 392/563 completed (loss: 0.0121:  70%|██████▉   | 196/281 [19:51<08:40,  6.13s/it]
Training Epoch: 16/100, batch 392/563 completed (loss: 0.1209:  70%|██████▉   | 196/281 [19:51<08:41,  6.13s/it]
Training Epoch: 16/100, batch 392/563 completed (loss: 0.0088:  70%|██████▉   | 196/281 [19:51<08:41,  6.13s/it]
Training Epoch: 16/100, batch 392/563 completed (loss: 0.1209:  70%|███████   | 197/281 [19:54<08:31,  6.09s/it]
Training Epoch: 16/100, batch 392/563 completed (loss: 0.0121:  70%|███████   | 197/281 [19:54<08:31,  6.09s/it]
Training Epoch: 16/100, batch 392/563 completed (loss: 0.0415:  70%|███████   | 197/281 [19:54<08:31,  6.09s/it]
Training Epoch: 16/100, batch 392/563 completed (loss: 0.0088:  70%|███████   | 197/281 [19:54<08:32,  6.10s/it]
Training Epoch: 16/100, batch 393/563 completed (loss: 0.0356:  70%|███████   | 197/281 [19:54<08:31,  6.09s/it]
Training Epoch: 16/100, batch 393/563 completed (loss: 0.0504:  70%|███████   | 197/281 [19:54<08:31,  6.09s/it]
Training Epoch: 16/100, batch 393/563 completed (loss: 0.0165:  70%|███████   | 197/281 [19:54<08:31,  6.09s/it]
Training Epoch: 16/100, batch 393/563 completed (loss: 0.007:  70%|███████   | 197/281 [19:54<08:32,  6.10s/it] 
Training Epoch: 16/100, batch 394/563 completed (loss: 0.0207:  70%|███████   | 197/281 [19:57<08:31,  6.09s/it]
Training Epoch: 16/100, batch 394/563 completed (loss: 0.2845:  70%|███████   | 197/281 [19:57<08:31,  6.09s/it]
Training Epoch: 16/100, batch 394/563 completed (loss: 0.452:  70%|███████   | 197/281 [19:57<08:31,  6.09s/it] 
Training Epoch: 16/100, batch 394/563 completed (loss: 0.0106:  70%|███████   | 197/281 [19:57<08:32,  6.10s/it]
Training Epoch: 16/100, batch 394/563 completed (loss: 0.0106:  70%|███████   | 198/281 [20:00<08:27,  6.12s/it]
Training Epoch: 16/100, batch 394/563 completed (loss: 0.0207:  70%|███████   | 198/281 [20:00<08:27,  6.12s/it]
Training Epoch: 16/100, batch 394/563 completed (loss: 0.2845:  70%|███████   | 198/281 [20:00<08:28,  6.12s/it]
Training Epoch: 16/100, batch 394/563 completed (loss: 0.452:  70%|███████   | 198/281 [20:00<08:28,  6.12s/it]
Training Epoch: 16/100, batch 395/563 completed (loss: 0.12:  70%|███████   | 198/281 [20:00<08:28,  6.12s/it]  
Training Epoch: 16/100, batch 395/563 completed (loss: 0.17:  70%|███████   | 198/281 [20:00<08:27,  6.12s/it]  
Training Epoch: 16/100, batch 395/563 completed (loss: 0.0135:  70%|███████   | 198/281 [20:00<08:28,  6.12s/it]
Training Epoch: 16/100, batch 395/563 completed (loss: 0.0377:  70%|███████   | 198/281 [20:00<08:27,  6.12s/it]
Training Epoch: 16/100, batch 396/563 completed (loss: 0.0068:  70%|███████   | 198/281 [20:04<08:28,  6.12s/it]
Training Epoch: 16/100, batch 396/563 completed (loss: 0.0096:  70%|███████   | 198/281 [20:03<08:27,  6.12s/it]
Training Epoch: 16/100, batch 396/563 completed (loss: 0.0937:  70%|███████   | 198/281 [20:04<08:28,  6.12s/it]
Training Epoch: 16/100, batch 396/563 completed (loss: 0.0623:  70%|███████   | 198/281 [20:04<08:27,  6.12s/it]
Training Epoch: 16/100, batch 396/563 completed (loss: 0.0068:  71%|███████   | 199/281 [20:06<08:18,  6.08s/it]
Training Epoch: 16/100, batch 396/563 completed (loss: 0.0623:  71%|███████   | 199/281 [20:06<08:19,  6.09s/it]
Training Epoch: 16/100, batch 396/563 completed (loss: 0.0096:  71%|███████   | 199/281 [20:06<08:19,  6.09s/it]
Training Epoch: 16/100, batch 396/563 completed (loss: 0.0937:  71%|███████   | 199/281 [20:06<08:19,  6.09s/it]
Training Epoch: 16/100, batch 397/563 completed (loss: 0.0694:  71%|███████   | 199/281 [20:06<08:19,  6.09s/it]
Training Epoch: 16/100, batch 397/563 completed (loss: 0.0191:  71%|███████   | 199/281 [20:06<08:19,  6.09s/it]
Training Epoch: 16/100, batch 397/563 completed (loss: 0.0799:  71%|███████   | 199/281 [20:06<08:18,  6.08s/it]
Training Epoch: 16/100, batch 397/563 completed (loss: 0.0788:  71%|███████   | 199/281 [20:06<08:19,  6.09s/it]
Training Epoch: 16/100, batch 398/563 completed (loss: 0.0059:  71%|███████   | 199/281 [20:10<08:19,  6.09s/it]
Training Epoch: 16/100, batch 398/563 completed (loss: 0.1336:  71%|███████   | 199/281 [20:10<08:19,  6.09s/it]
Training Epoch: 16/100, batch 398/563 completed (loss: 0.013:  71%|███████   | 199/281 [20:10<08:18,  6.08s/it] 
Training Epoch: 16/100, batch 398/563 completed (loss: 0.0212:  71%|███████   | 199/281 [20:10<08:19,  6.09s/it]
Training Epoch: 16/100, batch 398/563 completed (loss: 0.013:  71%|███████   | 200/281 [20:12<08:11,  6.07s/it]
Training Epoch: 16/100, batch 398/563 completed (loss: 0.0059:  71%|███████   | 200/281 [20:12<08:11,  6.07s/it]
Training Epoch: 16/100, batch 398/563 completed (loss: 0.0212:  71%|███████   | 200/281 [20:12<08:11,  6.07s/it]
Training Epoch: 16/100, batch 398/563 completed (loss: 0.1336:  71%|███████   | 200/281 [20:12<08:11,  6.07s/it]
Training Epoch: 16/100, batch 399/563 completed (loss: 0.0593:  71%|███████   | 200/281 [20:12<08:11,  6.07s/it]
Training Epoch: 16/100, batch 399/563 completed (loss: 0.2927:  71%|███████   | 200/281 [20:12<08:11,  6.07s/it]
Training Epoch: 16/100, batch 399/563 completed (loss: 0.0194:  71%|███████   | 200/281 [20:12<08:11,  6.07s/it]
Training Epoch: 16/100, batch 399/563 completed (loss: 0.0484:  71%|███████   | 200/281 [20:13<08:11,  6.07s/it]
Training Epoch: 16/100, batch 400/563 completed (loss: 0.0092:  71%|███████   | 200/281 [20:16<08:11,  6.07s/it]
Training Epoch: 16/100, batch 400/563 completed (loss: 0.0677:  71%|███████   | 200/281 [20:16<08:11,  6.07s/it]
Training Epoch: 16/100, batch 400/563 completed (loss: 0.0348:  71%|███████   | 200/281 [20:16<08:11,  6.07s/it]
Training Epoch: 16/100, batch 400/563 completed (loss: 0.0438:  71%|███████   | 200/281 [20:16<08:11,  6.07s/it]
Training Epoch: 16/100, batch 400/563 completed (loss: 0.0677:  72%|███████▏  | 201/281 [20:19<08:11,  6.14s/it]
Training Epoch: 16/100, batch 400/563 completed (loss: 0.0348:  72%|███████▏  | 201/281 [20:19<08:11,  6.14s/it]
Training Epoch: 16/100, batch 400/563 completed (loss: 0.0438:  72%|███████▏  | 201/281 [20:19<08:11,  6.14s/it]
Training Epoch: 16/100, batch 400/563 completed (loss: 0.0092:  72%|███████▏  | 201/281 [20:19<08:11,  6.14s/it]
Training Epoch: 16/100, batch 401/563 completed (loss: 0.036:  72%|███████▏  | 201/281 [20:19<08:11,  6.14s/it] 
Training Epoch: 16/100, batch 401/563 completed (loss: 0.0071:  72%|███████▏  | 201/281 [20:19<08:11,  6.14s/it]
Training Epoch: 16/100, batch 401/563 completed (loss: 0.0809:  72%|███████▏  | 201/281 [20:19<08:11,  6.14s/it]
Training Epoch: 16/100, batch 401/563 completed (loss: 0.0254:  72%|███████▏  | 201/281 [20:19<08:11,  6.14s/it]
Training Epoch: 16/100, batch 402/563 completed (loss: 0.0074:  72%|███████▏  | 201/281 [20:22<08:11,  6.14s/it]
Training Epoch: 16/100, batch 402/563 completed (loss: 0.0707:  72%|███████▏  | 201/281 [20:22<08:11,  6.14s/it]
Training Epoch: 16/100, batch 402/563 completed (loss: 0.1894:  72%|███████▏  | 201/281 [20:22<08:11,  6.14s/it]
Training Epoch: 16/100, batch 402/563 completed (loss: 0.5218:  72%|███████▏  | 201/281 [20:22<08:11,  6.14s/it]
Training Epoch: 16/100, batch 402/563 completed (loss: 0.5218:  72%|███████▏  | 202/281 [20:25<08:01,  6.09s/it]
Training Epoch: 16/100, batch 402/563 completed (loss: 0.0074:  72%|███████▏  | 202/281 [20:25<08:01,  6.09s/it]
Training Epoch: 16/100, batch 402/563 completed (loss: 0.1894:  72%|███████▏  | 202/281 [20:25<08:01,  6.09s/it]
Training Epoch: 16/100, batch 402/563 completed (loss: 0.0707:  72%|███████▏  | 202/281 [20:25<08:01,  6.10s/it]
Training Epoch: 16/100, batch 403/563 completed (loss: 0.5397:  72%|███████▏  | 202/281 [20:25<08:01,  6.10s/it]
Training Epoch: 16/100, batch 403/563 completed (loss: 0.0177:  72%|███████▏  | 202/281 [20:25<08:01,  6.09s/it]
Training Epoch: 16/100, batch 403/563 completed (loss: 0.0519:  72%|███████▏  | 202/281 [20:25<08:01,  6.09s/it]
Training Epoch: 16/100, batch 403/563 completed (loss: 0.1088:  72%|███████▏  | 202/281 [20:25<08:01,  6.09s/it]
Training Epoch: 16/100, batch 404/563 completed (loss: 0.005:  72%|███████▏  | 202/281 [20:28<08:01,  6.10s/it] 
Training Epoch: 16/100, batch 404/563 completed (loss: 0.0211:  72%|███████▏  | 202/281 [20:28<08:01,  6.09s/it]
Training Epoch: 16/100, batch 404/563 completed (loss: 0.1018:  72%|███████▏  | 202/281 [20:28<08:01,  6.09s/it]
Training Epoch: 16/100, batch 404/563 completed (loss: 0.0351:  72%|███████▏  | 202/281 [20:28<08:01,  6.09s/it]
Training Epoch: 16/100, batch 404/563 completed (loss: 0.0351:  72%|███████▏  | 203/281 [20:31<07:57,  6.12s/it]
Training Epoch: 16/100, batch 404/563 completed (loss: 0.1018:  72%|███████▏  | 203/281 [20:31<07:57,  6.12s/it]
Training Epoch: 16/100, batch 404/563 completed (loss: 0.0211:  72%|███████▏  | 203/281 [20:31<07:57,  6.12s/it]
Training Epoch: 16/100, batch 404/563 completed (loss: 0.005:  72%|███████▏  | 203/281 [20:31<07:57,  6.12s/it]
Training Epoch: 16/100, batch 405/563 completed (loss: 0.0154:  72%|███████▏  | 203/281 [20:31<07:57,  6.12s/it]
Training Epoch: 16/100, batch 405/563 completed (loss: 0.2326:  72%|███████▏  | 203/281 [20:31<07:57,  6.12s/it]
Training Epoch: 16/100, batch 405/563 completed (loss: 0.2413:  72%|███████▏  | 203/281 [20:31<07:57,  6.12s/it]
Training Epoch: 16/100, batch 405/563 completed (loss: 0.2456:  72%|███████▏  | 203/281 [20:31<07:57,  6.12s/it]
Training Epoch: 16/100, batch 406/563 completed (loss: 0.0371:  72%|███████▏  | 203/281 [20:34<07:57,  6.12s/it]
Training Epoch: 16/100, batch 406/563 completed (loss: 0.0283:  72%|███████▏  | 203/281 [20:34<07:57,  6.12s/it]
Training Epoch: 16/100, batch 406/563 completed (loss: 0.08:  72%|███████▏  | 203/281 [20:34<07:57,  6.12s/it]  
Training Epoch: 16/100, batch 406/563 completed (loss: 0.0629:  72%|███████▏  | 203/281 [20:34<07:57,  6.12s/it]
Training Epoch: 16/100, batch 406/563 completed (loss: 0.0283:  73%|███████▎  | 204/281 [20:37<07:48,  6.09s/it]
Training Epoch: 16/100, batch 406/563 completed (loss: 0.08:  73%|███████▎  | 204/281 [20:37<07:48,  6.09s/it]
Training Epoch: 16/100, batch 406/563 completed (loss: 0.0371:  73%|███████▎  | 204/281 [20:37<07:48,  6.09s/it]
Training Epoch: 16/100, batch 406/563 completed (loss: 0.0629:  73%|███████▎  | 204/281 [20:37<07:49,  6.09s/it]
Training Epoch: 16/100, batch 407/563 completed (loss: 0.0215:  73%|███████▎  | 204/281 [20:37<07:48,  6.09s/it]
Training Epoch: 16/100, batch 407/563 completed (loss: 0.0301:  73%|███████▎  | 204/281 [20:37<07:48,  6.09s/it]
Training Epoch: 16/100, batch 407/563 completed (loss: 0.1484:  73%|███████▎  | 204/281 [20:37<07:48,  6.09s/it]
Training Epoch: 16/100, batch 407/563 completed (loss: 0.0193:  73%|███████▎  | 204/281 [20:37<07:49,  6.09s/it]
Training Epoch: 16/100, batch 408/563 completed (loss: 0.2524:  73%|███████▎  | 204/281 [20:40<07:48,  6.09s/it]
Training Epoch: 16/100, batch 408/563 completed (loss: 0.1674:  73%|███████▎  | 204/281 [20:40<07:48,  6.09s/it]
Training Epoch: 16/100, batch 408/563 completed (loss: 0.0062:  73%|███████▎  | 204/281 [20:40<07:48,  6.09s/it]
Training Epoch: 16/100, batch 408/563 completed (loss: 0.1414:  73%|███████▎  | 204/281 [20:40<07:49,  6.09s/it]
Training Epoch: 16/100, batch 408/563 completed (loss: 0.2524:  73%|███████▎  | 205/281 [20:43<07:40,  6.06s/it]
Training Epoch: 16/100, batch 408/563 completed (loss: 0.0062:  73%|███████▎  | 205/281 [20:43<07:40,  6.07s/it]
Training Epoch: 16/100, batch 408/563 completed (loss: 0.1414:  73%|███████▎  | 205/281 [20:43<07:41,  6.07s/it]
Training Epoch: 16/100, batch 408/563 completed (loss: 0.1674:  73%|███████▎  | 205/281 [20:43<07:41,  6.07s/it]
Training Epoch: 16/100, batch 409/563 completed (loss: 0.1832:  73%|███████▎  | 205/281 [20:43<07:40,  6.06s/it]
Training Epoch: 16/100, batch 409/563 completed (loss: 0.1663:  73%|███████▎  | 205/281 [20:43<07:41,  6.07s/it]
Training Epoch: 16/100, batch 409/563 completed (loss: 0.0241:  73%|███████▎  | 205/281 [20:43<07:40,  6.07s/it]
Training Epoch: 16/100, batch 409/563 completed (loss: 0.0481:  73%|███████▎  | 205/281 [20:43<07:41,  6.07s/it]
Training Epoch: 16/100, batch 410/563 completed (loss: 0.0221:  73%|███████▎  | 205/281 [20:46<07:40,  6.06s/it]
Training Epoch: 16/100, batch 410/563 completed (loss: 0.0117:  73%|███████▎  | 205/281 [20:46<07:40,  6.07s/it]
Training Epoch: 16/100, batch 410/563 completed (loss: 0.0385:  73%|███████▎  | 205/281 [20:46<07:41,  6.07s/it]
Training Epoch: 16/100, batch 410/563 completed (loss: 0.3575:  73%|███████▎  | 205/281 [20:46<07:41,  6.07s/it]
Training Epoch: 16/100, batch 410/563 completed (loss: 0.3575:  73%|███████▎  | 206/281 [20:49<07:37,  6.10s/it]
Training Epoch: 16/100, batch 410/563 completed (loss: 0.0385:  73%|███████▎  | 206/281 [20:49<07:37,  6.10s/it]
Training Epoch: 16/100, batch 410/563 completed (loss: 0.0117:  73%|███████▎  | 206/281 [20:49<07:37,  6.10s/it]
Training Epoch: 16/100, batch 410/563 completed (loss: 0.0221:  73%|███████▎  | 206/281 [20:49<07:37,  6.10s/it]
Training Epoch: 16/100, batch 411/563 completed (loss: 0.1834:  73%|███████▎  | 206/281 [20:49<07:37,  6.10s/it]
Training Epoch: 16/100, batch 411/563 completed (loss: 0.4229:  73%|███████▎  | 206/281 [20:49<07:37,  6.10s/it]
Training Epoch: 16/100, batch 411/563 completed (loss: 0.0023:  73%|███████▎  | 206/281 [20:49<07:37,  6.10s/it]
Training Epoch: 16/100, batch 411/563 completed (loss: 0.1312:  73%|███████▎  | 206/281 [20:49<07:37,  6.10s/it]
Training Epoch: 16/100, batch 412/563 completed (loss: 0.0741:  73%|███████▎  | 206/281 [20:52<07:37,  6.10s/it]
Training Epoch: 16/100, batch 412/563 completed (loss: 0.0271:  73%|███████▎  | 206/281 [20:52<07:37,  6.10s/it]
Training Epoch: 16/100, batch 412/563 completed (loss: 0.2502:  73%|███████▎  | 206/281 [20:52<07:37,  6.10s/it]
Training Epoch: 16/100, batch 412/563 completed (loss: 0.0441:  73%|███████▎  | 206/281 [20:52<07:37,  6.10s/it]
Training Epoch: 16/100, batch 412/563 completed (loss: 0.0441:  74%|███████▎  | 207/281 [20:55<07:29,  6.07s/it]
Training Epoch: 16/100, batch 412/563 completed (loss: 0.0271:  74%|███████▎  | 207/281 [20:55<07:29,  6.07s/it]
Training Epoch: 16/100, batch 412/563 completed (loss: 0.0741:  74%|███████▎  | 207/281 [20:55<07:29,  6.07s/it]
Training Epoch: 16/100, batch 412/563 completed (loss: 0.2502:  74%|███████▎  | 207/281 [20:55<07:29,  6.08s/it]
Training Epoch: 16/100, batch 413/563 completed (loss: 0.2096:  74%|███████▎  | 207/281 [20:55<07:29,  6.07s/it]
Training Epoch: 16/100, batch 413/563 completed (loss: 0.249:  74%|███████▎  | 207/281 [20:55<07:29,  6.07s/it] 
Training Epoch: 16/100, batch 413/563 completed (loss: 0.0771:  74%|███████▎  | 207/281 [20:55<07:29,  6.08s/it]
Training Epoch: 16/100, batch 413/563 completed (loss: 0.0032:  74%|███████▎  | 207/281 [20:55<07:29,  6.07s/it]
Training Epoch: 16/100, batch 414/563 completed (loss: 0.0234:  74%|███████▎  | 207/281 [20:58<07:29,  6.07s/it]
Training Epoch: 16/100, batch 414/563 completed (loss: 0.0785:  74%|███████▎  | 207/281 [20:58<07:29,  6.07s/it]
Training Epoch: 16/100, batch 414/563 completed (loss: 0.2968:  74%|███████▎  | 207/281 [20:58<07:29,  6.08s/it]
Training Epoch: 16/100, batch 414/563 completed (loss: 0.0473:  74%|███████▎  | 207/281 [20:58<07:29,  6.07s/it]
Training Epoch: 16/100, batch 414/563 completed (loss: 0.2968:  74%|███████▍  | 208/281 [21:01<07:22,  6.06s/it]
Training Epoch: 16/100, batch 414/563 completed (loss: 0.0234:  74%|███████▍  | 208/281 [21:01<07:22,  6.06s/it]
Training Epoch: 16/100, batch 414/563 completed (loss: 0.0473:  74%|███████▍  | 208/281 [21:01<07:22,  6.06s/it]
Training Epoch: 16/100, batch 414/563 completed (loss: 0.0785:  74%|███████▍  | 208/281 [21:01<07:22,  6.06s/it]
Training Epoch: 16/100, batch 415/563 completed (loss: 0.1308:  74%|███████▍  | 208/281 [21:01<07:22,  6.06s/it]
Training Epoch: 16/100, batch 415/563 completed (loss: 0.0074:  74%|███████▍  | 208/281 [21:01<07:22,  6.06s/it]
Training Epoch: 16/100, batch 415/563 completed (loss: 0.0194:  74%|███████▍  | 208/281 [21:01<07:22,  6.06s/it]
Training Epoch: 16/100, batch 415/563 completed (loss: 0.0148:  74%|███████▍  | 208/281 [21:01<07:22,  6.06s/it]
Training Epoch: 16/100, batch 416/563 completed (loss: 0.1446:  74%|███████▍  | 208/281 [21:04<07:22,  6.06s/it]
Training Epoch: 16/100, batch 416/563 completed (loss: 0.3867:  74%|███████▍  | 208/281 [21:04<07:22,  6.06s/it]
Training Epoch: 16/100, batch 416/563 completed (loss: 0.0307:  74%|███████▍  | 208/281 [21:04<07:22,  6.06s/it]
Training Epoch: 16/100, batch 416/563 completed (loss: 0.2056:  74%|███████▍  | 208/281 [21:04<07:22,  6.06s/it]
Training Epoch: 16/100, batch 416/563 completed (loss: 0.0307:  74%|███████▍  | 209/281 [21:07<07:21,  6.14s/it]
Training Epoch: 16/100, batch 416/563 completed (loss: 0.2056:  74%|███████▍  | 209/281 [21:07<07:22,  6.14s/it]
Training Epoch: 16/100, batch 416/563 completed (loss: 0.1446:  74%|███████▍  | 209/281 [21:07<07:22,  6.14s/it]
Training Epoch: 16/100, batch 416/563 completed (loss: 0.3867:  74%|███████▍  | 209/281 [21:07<07:22,  6.14s/it]
Training Epoch: 16/100, batch 417/563 completed (loss: 0.024:  74%|███████▍  | 209/281 [21:07<07:22,  6.14s/it] 
Training Epoch: 16/100, batch 417/563 completed (loss: 0.0516:  74%|███████▍  | 209/281 [21:07<07:21,  6.14s/it]
Training Epoch: 16/100, batch 417/563 completed (loss: 0.0331:  74%|███████▍  | 209/281 [21:08<07:22,  6.14s/it]
Training Epoch: 16/100, batch 417/563 completed (loss: 0.039:  74%|███████▍  | 209/281 [21:08<07:22,  6.14s/it] 
Training Epoch: 16/100, batch 418/563 completed (loss: 0.0166:  74%|███████▍  | 209/281 [21:10<07:22,  6.14s/it]
Training Epoch: 16/100, batch 418/563 completed (loss: 0.0326:  74%|███████▍  | 209/281 [21:10<07:21,  6.14s/it]
Training Epoch: 16/100, batch 418/563 completed (loss: 0.0486:  74%|███████▍  | 209/281 [21:10<07:22,  6.14s/it]
Training Epoch: 16/100, batch 418/563 completed (loss: 0.0132:  74%|███████▍  | 209/281 [21:10<07:22,  6.14s/it]
Training Epoch: 16/100, batch 418/563 completed (loss: 0.0132:  75%|███████▍  | 210/281 [21:14<07:13,  6.10s/it]
Training Epoch: 16/100, batch 418/563 completed (loss: 0.0486:  75%|███████▍  | 210/281 [21:13<07:13,  6.10s/it]
Training Epoch: 16/100, batch 418/563 completed (loss: 0.0326:  75%|███████▍  | 210/281 [21:13<07:13,  6.10s/it]
Training Epoch: 16/100, batch 418/563 completed (loss: 0.0166:  75%|███████▍  | 210/281 [21:13<07:13,  6.10s/it]
Training Epoch: 16/100, batch 419/563 completed (loss: 0.0427:  75%|███████▍  | 210/281 [21:14<07:13,  6.10s/it]
Training Epoch: 16/100, batch 419/563 completed (loss: 0.1111:  75%|███████▍  | 210/281 [21:14<07:13,  6.10s/it]
Training Epoch: 16/100, batch 419/563 completed (loss: 0.0322:  75%|███████▍  | 210/281 [21:14<07:13,  6.10s/it]
Training Epoch: 16/100, batch 419/563 completed (loss: 0.0328:  75%|███████▍  | 210/281 [21:14<07:13,  6.10s/it]
Training Epoch: 16/100, batch 420/563 completed (loss: 0.0249:  75%|███████▍  | 210/281 [21:16<07:13,  6.10s/it]
Training Epoch: 16/100, batch 420/563 completed (loss: 0.0314:  75%|███████▍  | 210/281 [21:16<07:13,  6.10s/it]
Training Epoch: 16/100, batch 420/563 completed (loss: 0.0161:  75%|███████▍  | 210/281 [21:16<07:13,  6.10s/it]
Training Epoch: 16/100, batch 420/563 completed (loss: 0.4712:  75%|███████▍  | 210/281 [21:16<07:13,  6.10s/it]
Training Epoch: 16/100, batch 420/563 completed (loss: 0.0161:  75%|███████▌  | 211/281 [21:19<07:04,  6.06s/it]
Training Epoch: 16/100, batch 420/563 completed (loss: 0.0314:  75%|███████▌  | 211/281 [21:19<07:04,  6.06s/it]
Training Epoch: 16/100, batch 420/563 completed (loss: 0.4712:  75%|███████▌  | 211/281 [21:19<07:04,  6.07s/it]
Training Epoch: 16/100, batch 420/563 completed (loss: 0.0249:  75%|███████▌  | 211/281 [21:19<07:04,  6.07s/it]
Training Epoch: 16/100, batch 421/563 completed (loss: 0.0091:  75%|███████▌  | 211/281 [21:19<07:04,  6.07s/it]
Training Epoch: 16/100, batch 421/563 completed (loss: 0.0342:  75%|███████▌  | 211/281 [21:20<07:04,  6.06s/it]
Training Epoch: 16/100, batch 421/563 completed (loss: 0.0113:  75%|███████▌  | 211/281 [21:19<07:04,  6.06s/it]
Training Epoch: 16/100, batch 421/563 completed (loss: 0.0032:  75%|███████▌  | 211/281 [21:20<07:04,  6.07s/it]
Training Epoch: 16/100, batch 422/563 completed (loss: 0.0732:  75%|███████▌  | 211/281 [21:23<07:04,  6.07s/it]
Training Epoch: 16/100, batch 422/563 completed (loss: 0.0406:  75%|███████▌  | 211/281 [21:23<07:04,  6.06s/it]
Training Epoch: 16/100, batch 422/563 completed (loss: 0.0351:  75%|███████▌  | 211/281 [21:23<07:04,  6.06s/it]
Training Epoch: 16/100, batch 422/563 completed (loss: 0.2397:  75%|███████▌  | 211/281 [21:23<07:04,  6.07s/it]
Training Epoch: 16/100, batch 422/563 completed (loss: 0.0351:  75%|███████▌  | 212/281 [21:26<07:00,  6.09s/it]
Training Epoch: 16/100, batch 422/563 completed (loss: 0.0406:  75%|███████▌  | 212/281 [21:26<07:00,  6.09s/it]
Training Epoch: 16/100, batch 422/563 completed (loss: 0.2397:  75%|███████▌  | 212/281 [21:26<07:00,  6.10s/it]
Training Epoch: 16/100, batch 422/563 completed (loss: 0.0732:  75%|███████▌  | 212/281 [21:26<07:00,  6.09s/it]
Training Epoch: 16/100, batch 423/563 completed (loss: 0.0681:  75%|███████▌  | 212/281 [21:26<07:00,  6.09s/it]
Training Epoch: 16/100, batch 423/563 completed (loss: 0.019:  75%|███████▌  | 212/281 [21:26<07:00,  6.09s/it] 
Training Epoch: 16/100, batch 423/563 completed (loss: 0.2175:  75%|███████▌  | 212/281 [21:26<07:00,  6.09s/it]
Training Epoch: 16/100, batch 423/563 completed (loss: 0.0288:  75%|███████▌  | 212/281 [21:26<07:00,  6.10s/it]
Training Epoch: 16/100, batch 424/563 completed (loss: 0.4056:  75%|███████▌  | 212/281 [21:29<07:00,  6.09s/it]
Training Epoch: 16/100, batch 424/563 completed (loss: 0.0123:  75%|███████▌  | 212/281 [21:29<07:00,  6.09s/it]
Training Epoch: 16/100, batch 424/563 completed (loss: 0.0069:  75%|███████▌  | 212/281 [21:29<07:00,  6.09s/it]
Training Epoch: 16/100, batch 424/563 completed (loss: 0.0281:  75%|███████▌  | 212/281 [21:29<07:00,  6.10s/it]
Training Epoch: 16/100, batch 424/563 completed (loss: 0.0123:  76%|███████▌  | 213/281 [21:32<06:51,  6.06s/it]
Training Epoch: 16/100, batch 424/563 completed (loss: 0.4056:  76%|███████▌  | 213/281 [21:32<06:51,  6.06s/it]
Training Epoch: 16/100, batch 424/563 completed (loss: 0.0281:  76%|███████▌  | 213/281 [21:32<06:51,  6.06s/it]
Training Epoch: 16/100, batch 424/563 completed (loss: 0.0069:  76%|███████▌  | 213/281 [21:32<06:52,  6.06s/it]
Training Epoch: 16/100, batch 425/563 completed (loss: 0.0554:  76%|███████▌  | 213/281 [21:32<06:52,  6.06s/it]
Training Epoch: 16/100, batch 425/563 completed (loss: 0.0163:  76%|███████▌  | 213/281 [21:32<06:51,  6.06s/it]
Training Epoch: 16/100, batch 425/563 completed (loss: 0.1975:  76%|███████▌  | 213/281 [21:32<06:51,  6.06s/it]
Training Epoch: 16/100, batch 425/563 completed (loss: 0.0205:  76%|███████▌  | 213/281 [21:32<06:51,  6.06s/it]
Training Epoch: 16/100, batch 426/563 completed (loss: 0.1792:  76%|███████▌  | 213/281 [21:35<06:52,  6.06s/it]
Training Epoch: 16/100, batch 426/563 completed (loss: 0.0564:  76%|███████▌  | 213/281 [21:35<06:51,  6.06s/it]
Training Epoch: 16/100, batch 426/563 completed (loss: 0.1027:  76%|███████▌  | 213/281 [21:35<06:51,  6.06s/it]
Training Epoch: 16/100, batch 426/563 completed (loss: 0.0472:  76%|███████▌  | 213/281 [21:35<06:51,  6.06s/it]
Training Epoch: 16/100, batch 426/563 completed (loss: 0.0472:  76%|███████▌  | 214/281 [21:38<06:48,  6.10s/it]
Training Epoch: 16/100, batch 426/563 completed (loss: 0.1027:  76%|███████▌  | 214/281 [21:38<06:48,  6.10s/it]
Training Epoch: 16/100, batch 426/563 completed (loss: 0.0564:  76%|███████▌  | 214/281 [21:38<06:48,  6.10s/it]
Training Epoch: 16/100, batch 426/563 completed (loss: 0.1792:  76%|███████▌  | 214/281 [21:38<06:48,  6.10s/it]
Training Epoch: 16/100, batch 427/563 completed (loss: 0.0733:  76%|███████▌  | 214/281 [21:38<06:48,  6.10s/it]
Training Epoch: 16/100, batch 427/563 completed (loss: 0.0062:  76%|███████▌  | 214/281 [21:38<06:48,  6.10s/it]
Training Epoch: 16/100, batch 427/563 completed (loss: 0.0821:  76%|███████▌  | 214/281 [21:38<06:48,  6.10s/it]
Training Epoch: 16/100, batch 427/563 completed (loss: 0.0435:  76%|███████▌  | 214/281 [21:38<06:48,  6.10s/it]
Training Epoch: 16/100, batch 428/563 completed (loss: 0.3411:  76%|███████▌  | 214/281 [21:41<06:48,  6.10s/it]
Training Epoch: 16/100, batch 428/563 completed (loss: 0.2037:  76%|███████▌  | 214/281 [21:41<06:48,  6.10s/it]
Training Epoch: 16/100, batch 428/563 completed (loss: 0.0028:  76%|███████▌  | 214/281 [21:41<06:48,  6.10s/it]
Training Epoch: 16/100, batch 428/563 completed (loss: 0.0062:  76%|███████▌  | 214/281 [21:41<06:48,  6.10s/it]
Training Epoch: 16/100, batch 428/563 completed (loss: 0.2037:  77%|███████▋  | 215/281 [21:44<06:40,  6.07s/it]
Training Epoch: 16/100, batch 428/563 completed (loss: 0.3411:  77%|███████▋  | 215/281 [21:44<06:40,  6.07s/it]
Training Epoch: 16/100, batch 428/563 completed (loss: 0.0028:  77%|███████▋  | 215/281 [21:44<06:40,  6.07s/it]
Training Epoch: 16/100, batch 428/563 completed (loss: 0.0062:  77%|███████▋  | 215/281 [21:44<06:41,  6.08s/it]
Training Epoch: 16/100, batch 429/563 completed (loss: 0.0302:  77%|███████▋  | 215/281 [21:44<06:40,  6.07s/it]
Training Epoch: 16/100, batch 429/563 completed (loss: 0.6112:  77%|███████▋  | 215/281 [21:44<06:40,  6.07s/it]
Training Epoch: 16/100, batch 429/563 completed (loss: 0.0219:  77%|███████▋  | 215/281 [21:44<06:40,  6.07s/it]
Training Epoch: 16/100, batch 429/563 completed (loss: 0.0566:  77%|███████▋  | 215/281 [21:44<06:41,  6.08s/it]
Training Epoch: 16/100, batch 430/563 completed (loss: 0.3761:  77%|███████▋  | 215/281 [21:47<06:40,  6.07s/it]
Training Epoch: 16/100, batch 430/563 completed (loss: 0.1627:  77%|███████▋  | 215/281 [21:47<06:40,  6.07s/it]
Training Epoch: 16/100, batch 430/563 completed (loss: 0.3003:  77%|███████▋  | 215/281 [21:47<06:40,  6.07s/it]
Training Epoch: 16/100, batch 430/563 completed (loss: 0.0285:  77%|███████▋  | 215/281 [21:47<06:41,  6.08s/it]
Training Epoch: 16/100, batch 430/563 completed (loss: 0.3761:  77%|███████▋  | 216/281 [21:50<06:36,  6.10s/it]
Training Epoch: 16/100, batch 430/563 completed (loss: 0.3003:  77%|███████▋  | 216/281 [21:50<06:36,  6.10s/it]
Training Epoch: 16/100, batch 430/563 completed (loss: 0.1627:  77%|███████▋  | 216/281 [21:50<06:36,  6.10s/it]
Training Epoch: 16/100, batch 430/563 completed (loss: 0.0285:  77%|███████▋  | 216/281 [21:50<06:36,  6.10s/it]
Training Epoch: 16/100, batch 431/563 completed (loss: 0.1228:  77%|███████▋  | 216/281 [21:50<06:36,  6.10s/it]
Training Epoch: 16/100, batch 431/563 completed (loss: 0.0361:  77%|███████▋  | 216/281 [21:50<06:36,  6.10s/it]
Training Epoch: 16/100, batch 431/563 completed (loss: 0.0298:  77%|███████▋  | 216/281 [21:50<06:36,  6.10s/it]
Training Epoch: 16/100, batch 431/563 completed (loss: 0.0094:  77%|███████▋  | 216/281 [21:50<06:36,  6.10s/it]
Training Epoch: 16/100, batch 432/563 completed (loss: 0.0896:  77%|███████▋  | 216/281 [21:53<06:36,  6.10s/it]
Training Epoch: 16/100, batch 432/563 completed (loss: 0.0076:  77%|███████▋  | 216/281 [21:53<06:36,  6.10s/it]
Training Epoch: 16/100, batch 432/563 completed (loss: 0.5249:  77%|███████▋  | 216/281 [21:53<06:36,  6.10s/it]
Training Epoch: 16/100, batch 432/563 completed (loss: 0.011:  77%|███████▋  | 216/281 [21:53<06:36,  6.10s/it] 
Training Epoch: 16/100, batch 432/563 completed (loss: 0.5249:  77%|███████▋  | 217/281 [21:56<06:31,  6.11s/it]
Training Epoch: 16/100, batch 432/563 completed (loss: 0.0076:  77%|███████▋  | 217/281 [21:56<06:31,  6.12s/it]
Training Epoch: 16/100, batch 432/563 completed (loss: 0.0896:  77%|███████▋  | 217/281 [21:56<06:31,  6.12s/it]
Training Epoch: 16/100, batch 432/563 completed (loss: 0.011:  77%|███████▋  | 217/281 [21:56<06:31,  6.12s/it]
Training Epoch: 16/100, batch 433/563 completed (loss: 0.0056:  77%|███████▋  | 217/281 [21:56<06:31,  6.12s/it]
Training Epoch: 16/100, batch 433/563 completed (loss: 0.0949:  77%|███████▋  | 217/281 [21:56<06:31,  6.12s/it]
Training Epoch: 16/100, batch 433/563 completed (loss: 0.0996:  77%|███████▋  | 217/281 [21:56<06:31,  6.11s/it]
Training Epoch: 16/100, batch 433/563 completed (loss: 0.0074:  77%|███████▋  | 217/281 [21:56<06:31,  6.12s/it]
Training Epoch: 16/100, batch 434/563 completed (loss: 0.3463:  77%|███████▋  | 217/281 [21:59<06:31,  6.12s/it]
Training Epoch: 16/100, batch 434/563 completed (loss: 0.1312:  77%|███████▋  | 217/281 [21:59<06:31,  6.12s/it]
Training Epoch: 16/100, batch 434/563 completed (loss: 0.0239:  77%|███████▋  | 217/281 [21:59<06:31,  6.11s/it]
Training Epoch: 16/100, batch 434/563 completed (loss: 0.0173:  77%|███████▋  | 217/281 [21:59<06:31,  6.12s/it]
Training Epoch: 16/100, batch 434/563 completed (loss: 0.1312:  78%|███████▊  | 218/281 [22:02<06:23,  6.09s/it]
Training Epoch: 16/100, batch 434/563 completed (loss: 0.3463:  78%|███████▊  | 218/281 [22:02<06:23,  6.09s/it]
Training Epoch: 16/100, batch 434/563 completed (loss: 0.0173:  78%|███████▊  | 218/281 [22:02<06:23,  6.09s/it]
Training Epoch: 16/100, batch 434/563 completed (loss: 0.0239:  78%|███████▊  | 218/281 [22:02<06:23,  6.10s/it]
Training Epoch: 16/100, batch 435/563 completed (loss: 0.1051:  78%|███████▊  | 218/281 [22:02<06:23,  6.10s/it]
Training Epoch: 16/100, batch 435/563 completed (loss: 0.0333:  78%|███████▊  | 218/281 [22:02<06:23,  6.09s/it]
Training Epoch: 16/100, batch 435/563 completed (loss: 0.0299:  78%|███████▊  | 218/281 [22:02<06:23,  6.09s/it]
Training Epoch: 16/100, batch 435/563 completed (loss: 0.0235:  78%|███████▊  | 218/281 [22:02<06:23,  6.09s/it]
Training Epoch: 16/100, batch 436/563 completed (loss: 0.035:  78%|███████▊  | 218/281 [22:05<06:23,  6.10s/it] 
Training Epoch: 16/100, batch 436/563 completed (loss: 0.0237:  78%|███████▊  | 218/281 [22:05<06:23,  6.09s/it]
Training Epoch: 16/100, batch 436/563 completed (loss: 0.0058:  78%|███████▊  | 218/281 [22:05<06:23,  6.09s/it]
Training Epoch: 16/100, batch 436/563 completed (loss: 0.0531:  78%|███████▊  | 218/281 [22:05<06:23,  6.09s/it]
Training Epoch: 16/100, batch 436/563 completed (loss: 0.0531:  78%|███████▊  | 219/281 [22:08<06:18,  6.10s/it]
Training Epoch: 16/100, batch 436/563 completed (loss: 0.0058:  78%|███████▊  | 219/281 [22:08<06:18,  6.10s/it]
Training Epoch: 16/100, batch 436/563 completed (loss: 0.035:  78%|███████▊  | 219/281 [22:08<06:18,  6.10s/it]
Training Epoch: 16/100, batch 436/563 completed (loss: 0.0237:  78%|███████▊  | 219/281 [22:08<06:18,  6.11s/it]
Training Epoch: 16/100, batch 437/563 completed (loss: 0.0084:  78%|███████▊  | 219/281 [22:08<06:18,  6.11s/it]
Training Epoch: 16/100, batch 437/563 completed (loss: 0.1075:  78%|███████▊  | 219/281 [22:08<06:18,  6.10s/it]
Training Epoch: 16/100, batch 437/563 completed (loss: 0.0093:  78%|███████▊  | 219/281 [22:08<06:18,  6.10s/it]
Training Epoch: 16/100, batch 437/563 completed (loss: 0.006:  78%|███████▊  | 219/281 [22:08<06:18,  6.10s/it] 
Training Epoch: 16/100, batch 438/563 completed (loss: 0.0238:  78%|███████▊  | 219/281 [22:11<06:18,  6.11s/it]
Training Epoch: 16/100, batch 438/563 completed (loss: 0.3825:  78%|███████▊  | 219/281 [22:11<06:18,  6.10s/it]
Training Epoch: 16/100, batch 438/563 completed (loss: 0.059:  78%|███████▊  | 219/281 [22:11<06:18,  6.10s/it] 
Training Epoch: 16/100, batch 438/563 completed (loss: 0.016:  78%|███████▊  | 219/281 [22:11<06:18,  6.10s/it]
Training Epoch: 16/100, batch 438/563 completed (loss: 0.059:  78%|███████▊  | 220/281 [22:14<06:10,  6.07s/it]
Training Epoch: 16/100, batch 438/563 completed (loss: 0.016:  78%|███████▊  | 220/281 [22:14<06:10,  6.07s/it]
Training Epoch: 16/100, batch 438/563 completed (loss: 0.3825:  78%|███████▊  | 220/281 [22:14<06:10,  6.07s/it]
Training Epoch: 16/100, batch 438/563 completed (loss: 0.0238:  78%|███████▊  | 220/281 [22:14<06:10,  6.07s/it]
Training Epoch: 16/100, batch 439/563 completed (loss: 0.085:  78%|███████▊  | 220/281 [22:14<06:10,  6.07s/it] 
Training Epoch: 16/100, batch 439/563 completed (loss: 0.1847:  78%|███████▊  | 220/281 [22:14<06:10,  6.07s/it]
Training Epoch: 16/100, batch 439/563 completed (loss: 0.3068:  78%|███████▊  | 220/281 [22:14<06:10,  6.07s/it]
Training Epoch: 16/100, batch 439/563 completed (loss: 0.096:  78%|███████▊  | 220/281 [22:14<06:10,  6.07s/it]
Training Epoch: 16/100, batch 440/563 completed (loss: 0.0433:  78%|███████▊  | 220/281 [22:17<06:10,  6.07s/it]
Training Epoch: 16/100, batch 440/563 completed (loss: 0.1264:  78%|███████▊  | 220/281 [22:17<06:10,  6.07s/it]
Training Epoch: 16/100, batch 440/563 completed (loss: 0.2013:  78%|███████▊  | 220/281 [22:17<06:10,  6.07s/it]
Training Epoch: 16/100, batch 440/563 completed (loss: 0.0256:  78%|███████▊  | 220/281 [22:17<06:10,  6.07s/it]
Training Epoch: 16/100, batch 440/563 completed (loss: 0.2013:  79%|███████▊  | 221/281 [22:20<06:02,  6.05s/it]
Training Epoch: 16/100, batch 440/563 completed (loss: 0.0256:  79%|███████▊  | 221/281 [22:20<06:02,  6.05s/it]
Training Epoch: 16/100, batch 440/563 completed (loss: 0.1264:  79%|███████▊  | 221/281 [22:20<06:02,  6.05s/it]
Training Epoch: 16/100, batch 440/563 completed (loss: 0.0433:  79%|███████▊  | 221/281 [22:20<06:03,  6.05s/it]
Training Epoch: 16/100, batch 441/563 completed (loss: 0.0386:  79%|███████▊  | 221/281 [22:20<06:02,  6.05s/it]
Training Epoch: 16/100, batch 441/563 completed (loss: 0.0261:  79%|███████▊  | 221/281 [22:20<06:03,  6.05s/it]
Training Epoch: 16/100, batch 441/563 completed (loss: 0.0076:  79%|███████▊  | 221/281 [22:20<06:02,  6.05s/it]
Training Epoch: 16/100, batch 441/563 completed (loss: 0.0673:  79%|███████▊  | 221/281 [22:20<06:02,  6.05s/it]
Training Epoch: 16/100, batch 442/563 completed (loss: 0.0186:  79%|███████▊  | 221/281 [22:24<06:02,  6.05s/it]
Training Epoch: 16/100, batch 442/563 completed (loss: 0.0528:  79%|███████▊  | 221/281 [22:24<06:03,  6.05s/it]
Training Epoch: 16/100, batch 442/563 completed (loss: 0.5046:  79%|███████▊  | 221/281 [22:24<06:02,  6.05s/it]
Training Epoch: 16/100, batch 442/563 completed (loss: 0.0126:  79%|███████▊  | 221/281 [22:24<06:02,  6.05s/it]
Training Epoch: 16/100, batch 442/563 completed (loss: 0.5046:  79%|███████▉  | 222/281 [22:26<05:59,  6.09s/it]
Training Epoch: 16/100, batch 442/563 completed (loss: 0.0528:  79%|███████▉  | 222/281 [22:26<05:59,  6.09s/it]
Training Epoch: 16/100, batch 442/563 completed (loss: 0.0126:  79%|███████▉  | 222/281 [22:27<05:59,  6.09s/it]
Training Epoch: 16/100, batch 442/563 completed (loss: 0.0186:  79%|███████▉  | 222/281 [22:26<05:59,  6.09s/it]
Training Epoch: 16/100, batch 443/563 completed (loss: 0.0128:  79%|███████▉  | 222/281 [22:26<05:59,  6.09s/it]
Training Epoch: 16/100, batch 443/563 completed (loss: 0.0213:  79%|███████▉  | 222/281 [22:26<05:59,  6.09s/it]
Training Epoch: 16/100, batch 443/563 completed (loss: 0.58:  79%|███████▉  | 222/281 [22:27<05:59,  6.09s/it]  
Training Epoch: 16/100, batch 443/563 completed (loss: 0.0087:  79%|███████▉  | 222/281 [22:27<05:59,  6.09s/it]
Training Epoch: 16/100, batch 444/563 completed (loss: 0.0365:  79%|███████▉  | 222/281 [22:30<05:59,  6.09s/it]
Training Epoch: 16/100, batch 444/563 completed (loss: 0.0663:  79%|███████▉  | 222/281 [22:30<05:59,  6.09s/it]
Training Epoch: 16/100, batch 444/563 completed (loss: 0.0446:  79%|███████▉  | 222/281 [22:30<05:59,  6.09s/it]
Training Epoch: 16/100, batch 444/563 completed (loss: 0.4451:  79%|███████▉  | 222/281 [22:30<05:59,  6.09s/it]
Training Epoch: 16/100, batch 444/563 completed (loss: 0.4451:  79%|███████▉  | 223/281 [22:33<05:55,  6.12s/it]
Training Epoch: 16/100, batch 444/563 completed (loss: 0.0663:  79%|███████▉  | 223/281 [22:33<05:55,  6.12s/it]
Training Epoch: 16/100, batch 444/563 completed (loss: 0.0446:  79%|███████▉  | 223/281 [22:33<05:55,  6.13s/it]
Training Epoch: 16/100, batch 444/563 completed (loss: 0.0365:  79%|███████▉  | 223/281 [22:33<05:55,  6.13s/it]
Training Epoch: 16/100, batch 445/563 completed (loss: 0.016:  79%|███████▉  | 223/281 [22:33<05:55,  6.13s/it] 
Training Epoch: 16/100, batch 445/563 completed (loss: 0.1214:  79%|███████▉  | 223/281 [22:33<05:55,  6.12s/it]
Training Epoch: 16/100, batch 445/563 completed (loss: 0.0724:  79%|███████▉  | 223/281 [22:33<05:55,  6.13s/it]
Training Epoch: 16/100, batch 445/563 completed (loss: 0.1336:  79%|███████▉  | 223/281 [22:33<05:55,  6.12s/it]
Training Epoch: 16/100, batch 446/563 completed (loss: 0.0308:  79%|███████▉  | 223/281 [22:36<05:55,  6.13s/it]
Training Epoch: 16/100, batch 446/563 completed (loss: 0.1139:  79%|███████▉  | 223/281 [22:36<05:55,  6.13s/it]
Training Epoch: 16/100, batch 446/563 completed (loss: 0.0179:  79%|███████▉  | 223/281 [22:36<05:55,  6.12s/it]
Training Epoch: 16/100, batch 446/563 completed (loss: 0.0089:  79%|███████▉  | 223/281 [22:36<05:55,  6.12s/it]
Training Epoch: 16/100, batch 446/563 completed (loss: 0.0089:  80%|███████▉  | 224/281 [22:39<05:47,  6.10s/it]
Training Epoch: 16/100, batch 446/563 completed (loss: 0.1139:  80%|███████▉  | 224/281 [22:39<05:47,  6.10s/it]
Training Epoch: 16/100, batch 446/563 completed (loss: 0.0179:  80%|███████▉  | 224/281 [22:39<05:47,  6.10s/it]
Training Epoch: 16/100, batch 446/563 completed (loss: 0.0308:  80%|███████▉  | 224/281 [22:39<05:47,  6.10s/it]
Training Epoch: 16/100, batch 447/563 completed (loss: 0.4948:  80%|███████▉  | 224/281 [22:39<05:47,  6.10s/it]
Training Epoch: 16/100, batch 447/563 completed (loss: 0.1023:  80%|███████▉  | 224/281 [22:39<05:47,  6.10s/it]
Training Epoch: 16/100, batch 447/563 completed (loss: 0.0341:  80%|███████▉  | 224/281 [22:39<05:47,  6.10s/it]
Training Epoch: 16/100, batch 447/563 completed (loss: 0.1228:  80%|███████▉  | 224/281 [22:39<05:47,  6.10s/it]
Training Epoch: 16/100, batch 448/563 completed (loss: 0.0231:  80%|███████▉  | 224/281 [22:42<05:47,  6.10s/it]
Training Epoch: 16/100, batch 448/563 completed (loss: 0.042:  80%|███████▉  | 224/281 [22:42<05:47,  6.10s/it] 
Training Epoch: 16/100, batch 448/563 completed (loss: 0.0896:  80%|███████▉  | 224/281 [22:42<05:47,  6.10s/it]
Training Epoch: 16/100, batch 448/563 completed (loss: 0.0447:  80%|███████▉  | 224/281 [22:42<05:47,  6.10s/it]
Training Epoch: 16/100, batch 448/563 completed (loss: 0.0231:  80%|████████  | 225/281 [22:45<05:43,  6.13s/it]
Training Epoch: 16/100, batch 448/563 completed (loss: 0.0447:  80%|████████  | 225/281 [22:45<05:43,  6.13s/it]
Training Epoch: 16/100, batch 448/563 completed (loss: 0.0896:  80%|████████  | 225/281 [22:45<05:43,  6.13s/it]
Training Epoch: 16/100, batch 448/563 completed (loss: 0.042:  80%|████████  | 225/281 [22:45<05:43,  6.13s/it]
Training Epoch: 16/100, batch 449/563 completed (loss: 0.0612:  80%|████████  | 225/281 [22:45<05:43,  6.13s/it]
Training Epoch: 16/100, batch 449/563 completed (loss: 0.1252:  80%|████████  | 225/281 [22:45<05:43,  6.13s/it]
Training Epoch: 16/100, batch 449/563 completed (loss: 0.026:  80%|████████  | 225/281 [22:45<05:43,  6.13s/it] 
Training Epoch: 16/100, batch 449/563 completed (loss: 0.095:  80%|████████  | 225/281 [22:45<05:43,  6.13s/it] 
Training Epoch: 16/100, batch 450/563 completed (loss: 0.0419:  80%|████████  | 225/281 [22:48<05:43,  6.13s/it]
Training Epoch: 16/100, batch 450/563 completed (loss: 0.6417:  80%|████████  | 225/281 [22:48<05:43,  6.13s/it]
Training Epoch: 16/100, batch 450/563 completed (loss: 0.0547:  80%|████████  | 225/281 [22:48<05:43,  6.13s/it]
Training Epoch: 16/100, batch 450/563 completed (loss: 0.0183:  80%|████████  | 225/281 [22:48<05:43,  6.13s/it]
Training Epoch: 16/100, batch 450/563 completed (loss: 0.6417:  80%|████████  | 226/281 [22:51<05:35,  6.09s/it]
Training Epoch: 16/100, batch 450/563 completed (loss: 0.0183:  80%|████████  | 226/281 [22:51<05:35,  6.10s/it]
Training Epoch: 16/100, batch 450/563 completed (loss: 0.0419:  80%|████████  | 226/281 [22:51<05:35,  6.09s/it]
Training Epoch: 16/100, batch 450/563 completed (loss: 0.0547:  80%|████████  | 226/281 [22:51<05:35,  6.10s/it]
Training Epoch: 16/100, batch 451/563 completed (loss: 0.0185:  80%|████████  | 226/281 [22:51<05:35,  6.09s/it]
Training Epoch: 16/100, batch 451/563 completed (loss: 0.0052:  80%|████████  | 226/281 [22:51<05:35,  6.09s/it]
Training Epoch: 16/100, batch 451/563 completed (loss: 0.0153:  80%|████████  | 226/281 [22:51<05:35,  6.10s/it]
Training Epoch: 16/100, batch 451/563 completed (loss: 0.0107:  80%|████████  | 226/281 [22:51<05:35,  6.10s/it]
Training Epoch: 16/100, batch 452/563 completed (loss: 0.249:  80%|████████  | 226/281 [22:54<05:35,  6.10s/it] 
Training Epoch: 16/100, batch 452/563 completed (loss: 0.0224:  80%|████████  | 226/281 [22:54<05:35,  6.09s/it]
Training Epoch: 16/100, batch 452/563 completed (loss: 0.0223:  80%|████████  | 226/281 [22:54<05:35,  6.09s/it]
Training Epoch: 16/100, batch 452/563 completed (loss: 0.1272:  80%|████████  | 226/281 [22:54<05:35,  6.10s/it]
Training Epoch: 16/100, batch 452/563 completed (loss: 0.249:  81%|████████  | 227/281 [22:57<05:30,  6.12s/it]
Training Epoch: 16/100, batch 452/563 completed (loss: 0.0223:  81%|████████  | 227/281 [22:57<05:30,  6.12s/it]
Training Epoch: 16/100, batch 452/563 completed (loss: 0.0224:  81%|████████  | 227/281 [22:57<05:30,  6.12s/it]
Training Epoch: 16/100, batch 452/563 completed (loss: 0.1272:  81%|████████  | 227/281 [22:57<05:30,  6.12s/it]
Training Epoch: 16/100, batch 453/563 completed (loss: 0.0044:  81%|████████  | 227/281 [22:57<05:30,  6.12s/it]
Training Epoch: 16/100, batch 453/563 completed (loss: 0.0201:  81%|████████  | 227/281 [22:57<05:30,  6.12s/it]
Training Epoch: 16/100, batch 453/563 completed (loss: 0.0243:  81%|████████  | 227/281 [22:57<05:30,  6.12s/it]
Training Epoch: 16/100, batch 453/563 completed (loss: 0.0103:  81%|████████  | 227/281 [22:57<05:30,  6.12s/it]
Training Epoch: 16/100, batch 454/563 completed (loss: 0.0229:  81%|████████  | 227/281 [23:00<05:30,  6.12s/it]
Training Epoch: 16/100, batch 454/563 completed (loss: 0.0795:  81%|████████  | 227/281 [23:00<05:30,  6.12s/it]
Training Epoch: 16/100, batch 454/563 completed (loss: 0.0305:  81%|████████  | 227/281 [23:00<05:30,  6.12s/it]
Training Epoch: 16/100, batch 454/563 completed (loss: 0.0416:  81%|████████  | 227/281 [23:00<05:30,  6.12s/it]
Training Epoch: 16/100, batch 454/563 completed (loss: 0.0229:  81%|████████  | 228/281 [23:03<05:22,  6.09s/it]
Training Epoch: 16/100, batch 454/563 completed (loss: 0.0305:  81%|████████  | 228/281 [23:03<05:22,  6.09s/it]
Training Epoch: 16/100, batch 454/563 completed (loss: 0.0795:  81%|████████  | 228/281 [23:03<05:22,  6.08s/it]
Training Epoch: 16/100, batch 454/563 completed (loss: 0.0416:  81%|████████  | 228/281 [23:03<05:22,  6.09s/it]
Training Epoch: 16/100, batch 455/563 completed (loss: 0.0121:  81%|████████  | 228/281 [23:03<05:22,  6.09s/it]
Training Epoch: 16/100, batch 455/563 completed (loss: 0.0294:  81%|████████  | 228/281 [23:03<05:22,  6.08s/it]
Training Epoch: 16/100, batch 455/563 completed (loss: 0.0209:  81%|████████  | 228/281 [23:03<05:22,  6.09s/it]
Training Epoch: 16/100, batch 455/563 completed (loss: 0.0092:  81%|████████  | 228/281 [23:03<05:22,  6.09s/it]
Training Epoch: 16/100, batch 456/563 completed (loss: 0.0126:  81%|████████  | 228/281 [23:06<05:22,  6.08s/it]
Training Epoch: 16/100, batch 456/563 completed (loss: 0.0598:  81%|████████  | 228/281 [23:06<05:22,  6.09s/it]
Training Epoch: 16/100, batch 456/563 completed (loss: 0.0306:  81%|████████  | 228/281 [23:06<05:22,  6.09s/it]
Training Epoch: 16/100, batch 456/563 completed (loss: 0.0302:  81%|████████  | 228/281 [23:06<05:22,  6.09s/it]
Training Epoch: 16/100, batch 456/563 completed (loss: 0.0598:  81%|████████▏ | 229/281 [23:09<05:15,  6.07s/it]
Training Epoch: 16/100, batch 456/563 completed (loss: 0.0306:  81%|████████▏ | 229/281 [23:09<05:15,  6.07s/it]
Training Epoch: 16/100, batch 456/563 completed (loss: 0.0126:  81%|████████▏ | 229/281 [23:09<05:15,  6.07s/it]
Training Epoch: 16/100, batch 456/563 completed (loss: 0.0302:  81%|████████▏ | 229/281 [23:09<05:15,  6.07s/it]
Training Epoch: 16/100, batch 457/563 completed (loss: 0.0806:  81%|████████▏ | 229/281 [23:09<05:15,  6.07s/it]
Training Epoch: 16/100, batch 457/563 completed (loss: 0.0268:  81%|████████▏ | 229/281 [23:09<05:15,  6.07s/it]
Training Epoch: 16/100, batch 457/563 completed (loss: 0.0516:  81%|████████▏ | 229/281 [23:09<05:15,  6.07s/it]
Training Epoch: 16/100, batch 457/563 completed (loss: 0.0194:  81%|████████▏ | 229/281 [23:09<05:15,  6.07s/it]
Training Epoch: 16/100, batch 458/563 completed (loss: 0.0253:  81%|████████▏ | 229/281 [23:12<05:15,  6.07s/it]
Training Epoch: 16/100, batch 458/563 completed (loss: 0.1594:  81%|████████▏ | 229/281 [23:12<05:15,  6.07s/it]
Training Epoch: 16/100, batch 458/563 completed (loss: 0.1188:  81%|████████▏ | 229/281 [23:12<05:15,  6.07s/it]
Training Epoch: 16/100, batch 458/563 completed (loss: 0.0515:  81%|████████▏ | 229/281 [23:12<05:15,  6.07s/it]
Training Epoch: 16/100, batch 458/563 completed (loss: 0.1594:  82%|████████▏ | 230/281 [23:15<05:13,  6.15s/it]
Training Epoch: 16/100, batch 458/563 completed (loss: 0.1188:  82%|████████▏ | 230/281 [23:15<05:13,  6.15s/it]
Training Epoch: 16/100, batch 458/563 completed (loss: 0.0253:  82%|████████▏ | 230/281 [23:15<05:13,  6.15s/it]
Training Epoch: 16/100, batch 458/563 completed (loss: 0.0515:  82%|████████▏ | 230/281 [23:16<05:13,  6.15s/it]
Training Epoch: 16/100, batch 459/563 completed (loss: 0.083:  82%|████████▏ | 230/281 [23:15<05:13,  6.15s/it] 
Training Epoch: 16/100, batch 459/563 completed (loss: 0.0437:  82%|████████▏ | 230/281 [23:16<05:13,  6.15s/it]
Training Epoch: 16/100, batch 459/563 completed (loss: 0.0105:  82%|████████▏ | 230/281 [23:16<05:13,  6.15s/it]
Training Epoch: 16/100, batch 459/563 completed (loss: 0.2005:  82%|████████▏ | 230/281 [23:16<05:13,  6.15s/it]
Training Epoch: 16/100, batch 460/563 completed (loss: 0.0417:  82%|████████▏ | 230/281 [23:18<05:13,  6.15s/it]
Training Epoch: 16/100, batch 460/563 completed (loss: 0.0149:  82%|████████▏ | 230/281 [23:18<05:13,  6.15s/it]
Training Epoch: 16/100, batch 460/563 completed (loss: 0.0196:  82%|████████▏ | 230/281 [23:18<05:13,  6.15s/it]
Training Epoch: 16/100, batch 460/563 completed (loss: 0.0295:  82%|████████▏ | 230/281 [23:18<05:13,  6.15s/it]
Training Epoch: 16/100, batch 460/563 completed (loss: 0.0196:  82%|████████▏ | 231/281 [23:21<05:05,  6.10s/it]
Training Epoch: 16/100, batch 460/563 completed (loss: 0.0149:  82%|████████▏ | 231/281 [23:21<05:05,  6.10s/it]
Training Epoch: 16/100, batch 460/563 completed (loss: 0.0417:  82%|████████▏ | 231/281 [23:21<05:05,  6.10s/it]
Training Epoch: 16/100, batch 460/563 completed (loss: 0.0295:  82%|████████▏ | 231/281 [23:22<05:05,  6.11s/it]
Training Epoch: 16/100, batch 461/563 completed (loss: 0.039:  82%|████████▏ | 231/281 [23:22<05:05,  6.10s/it] 
Training Epoch: 16/100, batch 461/563 completed (loss: 0.0899:  82%|████████▏ | 231/281 [23:22<05:05,  6.10s/it]
Training Epoch: 16/100, batch 461/563 completed (loss: 0.1937:  82%|████████▏ | 231/281 [23:22<05:05,  6.10s/it]
Training Epoch: 16/100, batch 461/563 completed (loss: 0.15:  82%|████████▏ | 231/281 [23:22<05:05,  6.11s/it]  
Training Epoch: 16/100, batch 462/563 completed (loss: 0.158:  82%|████████▏ | 231/281 [23:24<05:05,  6.10s/it] 
Training Epoch: 16/100, batch 462/563 completed (loss: 0.0051:  82%|████████▏ | 231/281 [23:24<05:05,  6.10s/it]
Training Epoch: 16/100, batch 462/563 completed (loss: 0.03:  82%|████████▏ | 231/281 [23:24<05:05,  6.10s/it] 
Training Epoch: 16/100, batch 462/563 completed (loss: 0.0593:  82%|████████▏ | 231/281 [23:24<05:05,  6.11s/it]
Training Epoch: 16/100, batch 462/563 completed (loss: 0.0593:  83%|████████▎ | 232/281 [23:28<04:57,  6.07s/it]
Training Epoch: 16/100, batch 462/563 completed (loss: 0.0051:  83%|████████▎ | 232/281 [23:27<04:57,  6.08s/it]
Training Epoch: 16/100, batch 462/563 completed (loss: 0.158:  83%|████████▎ | 232/281 [23:27<04:57,  6.08s/it]
Training Epoch: 16/100, batch 462/563 completed (loss: 0.03:  83%|████████▎ | 232/281 [23:27<04:58,  6.08s/it]
Training Epoch: 16/100, batch 463/563 completed (loss: 0.0847:  83%|████████▎ | 232/281 [23:28<04:58,  6.08s/it]
Training Epoch: 16/100, batch 463/563 completed (loss: 0.0367:  83%|████████▎ | 232/281 [23:28<04:57,  6.08s/it]
Training Epoch: 16/100, batch 463/563 completed (loss: 0.087:  83%|████████▎ | 232/281 [23:28<04:57,  6.08s/it] 
Training Epoch: 16/100, batch 463/563 completed (loss: 0.2018:  83%|████████▎ | 232/281 [23:28<04:57,  6.07s/it]
Training Epoch: 16/100, batch 464/563 completed (loss: 0.0969:  83%|████████▎ | 232/281 [23:31<04:58,  6.08s/it]
Training Epoch: 16/100, batch 464/563 completed (loss: 0.0569:  83%|████████▎ | 232/281 [23:31<04:57,  6.08s/it]
Training Epoch: 16/100, batch 464/563 completed (loss: 0.0932:  83%|████████▎ | 232/281 [23:31<04:57,  6.08s/it]
Training Epoch: 16/100, batch 464/563 completed (loss: 0.0767:  83%|████████▎ | 232/281 [23:31<04:57,  6.07s/it]
Training Epoch: 16/100, batch 464/563 completed (loss: 0.0969:  83%|████████▎ | 233/281 [23:34<04:52,  6.10s/it]
Training Epoch: 16/100, batch 464/563 completed (loss: 0.0932:  83%|████████▎ | 233/281 [23:34<04:52,  6.10s/it]
Training Epoch: 16/100, batch 464/563 completed (loss: 0.0767:  83%|████████▎ | 233/281 [23:34<04:52,  6.10s/it]
Training Epoch: 16/100, batch 464/563 completed (loss: 0.0569:  83%|████████▎ | 233/281 [23:34<04:52,  6.10s/it]
Training Epoch: 16/100, batch 465/563 completed (loss: 0.3383:  83%|████████▎ | 233/281 [23:34<04:52,  6.10s/it]
Training Epoch: 16/100, batch 465/563 completed (loss: 0.0355:  83%|████████▎ | 233/281 [23:34<04:52,  6.10s/it]
Training Epoch: 16/100, batch 465/563 completed (loss: 0.4075:  83%|████████▎ | 233/281 [23:34<04:52,  6.10s/it]
Training Epoch: 16/100, batch 465/563 completed (loss: 0.0448:  83%|████████▎ | 233/281 [23:34<04:52,  6.10s/it]
Training Epoch: 16/100, batch 466/563 completed (loss: 0.2048:  83%|████████▎ | 233/281 [23:37<04:52,  6.10s/it]
Training Epoch: 16/100, batch 466/563 completed (loss: 0.1292:  83%|████████▎ | 233/281 [23:37<04:52,  6.10s/it]
Training Epoch: 16/100, batch 466/563 completed (loss: 0.116:  83%|████████▎ | 233/281 [23:37<04:52,  6.10s/it] 
Training Epoch: 16/100, batch 466/563 completed (loss: 0.0048:  83%|████████▎ | 233/281 [23:37<04:52,  6.10s/it]
Training Epoch: 16/100, batch 466/563 completed (loss: 0.2048:  83%|████████▎ | 234/281 [23:40<04:45,  6.07s/it]
Training Epoch: 16/100, batch 466/563 completed (loss: 0.116:  83%|████████▎ | 234/281 [23:40<04:45,  6.07s/it]
Training Epoch: 16/100, batch 466/563 completed (loss: 0.1292:  83%|████████▎ | 234/281 [23:40<04:45,  6.08s/it]
Training Epoch: 16/100, batch 466/563 completed (loss: 0.0048:  83%|████████▎ | 234/281 [23:40<04:45,  6.08s/it]
Training Epoch: 16/100, batch 467/563 completed (loss: 0.0859:  83%|████████▎ | 234/281 [23:40<04:45,  6.08s/it]
Training Epoch: 16/100, batch 467/563 completed (loss: 0.0049:  83%|████████▎ | 234/281 [23:40<04:45,  6.07s/it]
Training Epoch: 16/100, batch 467/563 completed (loss: 0.0984:  83%|████████▎ | 234/281 [23:40<04:45,  6.07s/it]
Training Epoch: 16/100, batch 467/563 completed (loss: 0.0454:  83%|████████▎ | 234/281 [23:40<04:45,  6.08s/it]
Training Epoch: 16/100, batch 468/563 completed (loss: 0.0891:  83%|████████▎ | 234/281 [23:43<04:45,  6.08s/it]
Training Epoch: 16/100, batch 468/563 completed (loss: 0.0184:  83%|████████▎ | 234/281 [23:43<04:45,  6.07s/it]
Training Epoch: 16/100, batch 468/563 completed (loss: 0.0246:  83%|████████▎ | 234/281 [23:43<04:45,  6.07s/it]
Training Epoch: 16/100, batch 468/563 completed (loss: 0.0308:  83%|████████▎ | 234/281 [23:43<04:45,  6.08s/it]
Training Epoch: 16/100, batch 468/563 completed (loss: 0.0184:  84%|████████▎ | 235/281 [23:46<04:41,  6.11s/it]
Training Epoch: 16/100, batch 468/563 completed (loss: 0.0246:  84%|████████▎ | 235/281 [23:46<04:41,  6.12s/it]
Training Epoch: 16/100, batch 468/563 completed (loss: 0.0891:  84%|████████▎ | 235/281 [23:46<04:41,  6.11s/it]
Training Epoch: 16/100, batch 468/563 completed (loss: 0.0308:  84%|████████▎ | 235/281 [23:46<04:41,  6.12s/it]
Training Epoch: 16/100, batch 469/563 completed (loss: 0.0008:  84%|████████▎ | 235/281 [23:46<04:41,  6.11s/it]
Training Epoch: 16/100, batch 469/563 completed (loss: 0.0159:  84%|████████▎ | 235/281 [23:46<04:41,  6.11s/it]
Training Epoch: 16/100, batch 469/563 completed (loss: 0.0229:  84%|████████▎ | 235/281 [23:46<04:41,  6.12s/it]
Training Epoch: 16/100, batch 469/563 completed (loss: 0.2916:  84%|████████▎ | 235/281 [23:46<04:41,  6.12s/it]
Training Epoch: 16/100, batch 470/563 completed (loss: 0.0087:  84%|████████▎ | 235/281 [23:49<04:41,  6.12s/it]
Training Epoch: 16/100, batch 470/563 completed (loss: 0.0475:  84%|████████▎ | 235/281 [23:49<04:41,  6.11s/it]
Training Epoch: 16/100, batch 470/563 completed (loss: 0.0415:  84%|████████▎ | 235/281 [23:49<04:41,  6.11s/it]
Training Epoch: 16/100, batch 470/563 completed (loss: 0.1171:  84%|████████▎ | 235/281 [23:49<04:41,  6.12s/it]
Training Epoch: 16/100, batch 470/563 completed (loss: 0.0415:  84%|████████▍ | 236/281 [23:52<04:35,  6.12s/it]
Training Epoch: 16/100, batch 470/563 completed (loss: 0.0087:  84%|████████▍ | 236/281 [23:52<04:35,  6.12s/it]
Training Epoch: 16/100, batch 470/563 completed (loss: 0.0475:  84%|████████▍ | 236/281 [23:52<04:35,  6.12s/it]
Training Epoch: 16/100, batch 470/563 completed (loss: 0.1171:  84%|████████▍ | 236/281 [23:52<04:35,  6.13s/it]
Training Epoch: 16/100, batch 471/563 completed (loss: 0.0752:  84%|████████▍ | 236/281 [23:52<04:35,  6.12s/it]
Training Epoch: 16/100, batch 471/563 completed (loss: 0.0022:  84%|████████▍ | 236/281 [23:52<04:35,  6.12s/it]
Training Epoch: 16/100, batch 471/563 completed (loss: 0.3249:  84%|████████▍ | 236/281 [23:52<04:35,  6.12s/it]
Training Epoch: 16/100, batch 471/563 completed (loss: 0.0199:  84%|████████▍ | 236/281 [23:52<04:35,  6.13s/it]
Training Epoch: 16/100, batch 472/563 completed (loss: 0.0073:  84%|████████▍ | 236/281 [23:55<04:35,  6.12s/it]
Training Epoch: 16/100, batch 472/563 completed (loss: 0.0029:  84%|████████▍ | 236/281 [23:55<04:35,  6.12s/it]
Training Epoch: 16/100, batch 472/563 completed (loss: 0.0028:  84%|████████▍ | 236/281 [23:55<04:35,  6.12s/it]
Training Epoch: 16/100, batch 472/563 completed (loss: 0.0844:  84%|████████▍ | 236/281 [23:55<04:35,  6.13s/it]
Training Epoch: 16/100, batch 472/563 completed (loss: 0.0844:  84%|████████▍ | 237/281 [23:58<04:25,  6.03s/it]
Training Epoch: 16/100, batch 472/563 completed (loss: 0.0028:  84%|████████▍ | 237/281 [23:58<04:25,  6.04s/it]
Training Epoch: 16/100, batch 472/563 completed (loss: 0.0029:  84%|████████▍ | 237/281 [23:58<04:25,  6.04s/it]
Training Epoch: 16/100, batch 472/563 completed (loss: 0.0073:  84%|████████▍ | 237/281 [23:58<04:25,  6.04s/it]
Training Epoch: 16/100, batch 473/563 completed (loss: 0.1007:  84%|████████▍ | 237/281 [23:58<04:25,  6.04s/it]
Training Epoch: 16/100, batch 473/563 completed (loss: 0.2399:  84%|████████▍ | 237/281 [23:58<04:25,  6.04s/it]
Training Epoch: 16/100, batch 473/563 completed (loss: 0.0041:  84%|████████▍ | 237/281 [23:58<04:25,  6.04s/it]
Training Epoch: 16/100, batch 473/563 completed (loss: 0.0467:  84%|████████▍ | 237/281 [23:58<04:25,  6.03s/it]
Training Epoch: 16/100, batch 474/563 completed (loss: 0.0079:  84%|████████▍ | 237/281 [24:01<04:25,  6.04s/it]
Training Epoch: 16/100, batch 474/563 completed (loss: 0.0384:  84%|████████▍ | 237/281 [24:01<04:25,  6.04s/it]
Training Epoch: 16/100, batch 474/563 completed (loss: 0.0296:  84%|████████▍ | 237/281 [24:01<04:25,  6.04s/it]
Training Epoch: 16/100, batch 474/563 completed (loss: 0.0086:  84%|████████▍ | 237/281 [24:01<04:25,  6.03s/it]
Training Epoch: 16/100, batch 474/563 completed (loss: 0.0384:  85%|████████▍ | 238/281 [24:04<04:23,  6.12s/it]
Training Epoch: 16/100, batch 474/563 completed (loss: 0.0079:  85%|████████▍ | 238/281 [24:04<04:23,  6.12s/it]
Training Epoch: 16/100, batch 474/563 completed (loss: 0.0086:  85%|████████▍ | 238/281 [24:04<04:23,  6.12s/it]
Training Epoch: 16/100, batch 474/563 completed (loss: 0.0296:  85%|████████▍ | 238/281 [24:04<04:23,  6.12s/it]
Training Epoch: 16/100, batch 475/563 completed (loss: 0.0298:  85%|████████▍ | 238/281 [24:04<04:23,  6.12s/it]
Training Epoch: 16/100, batch 475/563 completed (loss: 0.0957:  85%|████████▍ | 238/281 [24:04<04:23,  6.12s/it]
Training Epoch: 16/100, batch 475/563 completed (loss: 0.0115:  85%|████████▍ | 238/281 [24:04<04:23,  6.12s/it]
Training Epoch: 16/100, batch 475/563 completed (loss: 0.0348:  85%|████████▍ | 238/281 [24:04<04:23,  6.12s/it]
Training Epoch: 16/100, batch 476/563 completed (loss: 0.1139:  85%|████████▍ | 238/281 [24:07<04:23,  6.12s/it]
Training Epoch: 16/100, batch 476/563 completed (loss: 0.016:  85%|████████▍ | 238/281 [24:07<04:23,  6.12s/it] 
Training Epoch: 16/100, batch 476/563 completed (loss: 0.1183:  85%|████████▍ | 238/281 [24:07<04:23,  6.12s/it]
Training Epoch: 16/100, batch 476/563 completed (loss: 0.0649:  85%|████████▍ | 238/281 [24:07<04:23,  6.12s/it]
Training Epoch: 16/100, batch 476/563 completed (loss: 0.016:  85%|████████▌ | 239/281 [24:10<04:15,  6.08s/it]
Training Epoch: 16/100, batch 476/563 completed (loss: 0.1139:  85%|████████▌ | 239/281 [24:10<04:15,  6.08s/it]
Training Epoch: 16/100, batch 476/563 completed (loss: 0.1183:  85%|████████▌ | 239/281 [24:10<04:15,  6.08s/it]
Training Epoch: 16/100, batch 476/563 completed (loss: 0.0649:  85%|████████▌ | 239/281 [24:10<04:15,  6.08s/it]
Training Epoch: 16/100, batch 477/563 completed (loss: 0.0036:  85%|████████▌ | 239/281 [24:10<04:15,  6.08s/it]
Training Epoch: 16/100, batch 477/563 completed (loss: 0.1739:  85%|████████▌ | 239/281 [24:10<04:15,  6.08s/it]
Training Epoch: 16/100, batch 477/563 completed (loss: 0.035:  85%|████████▌ | 239/281 [24:10<04:15,  6.08s/it] 
Training Epoch: 16/100, batch 477/563 completed (loss: 0.0274:  85%|████████▌ | 239/281 [24:10<04:15,  6.08s/it]
Training Epoch: 16/100, batch 478/563 completed (loss: 0.0013:  85%|████████▌ | 239/281 [24:13<04:15,  6.08s/it]
Training Epoch: 16/100, batch 478/563 completed (loss: 0.0205:  85%|████████▌ | 239/281 [24:13<04:15,  6.08s/it]
Training Epoch: 16/100, batch 478/563 completed (loss: 0.1404:  85%|████████▌ | 239/281 [24:13<04:15,  6.08s/it]
Training Epoch: 16/100, batch 478/563 completed (loss: 0.0149:  85%|████████▌ | 239/281 [24:13<04:15,  6.08s/it]
Training Epoch: 16/100, batch 478/563 completed (loss: 0.0205:  85%|████████▌ | 240/281 [24:16<04:08,  6.06s/it]
Training Epoch: 16/100, batch 478/563 completed (loss: 0.0013:  85%|████████▌ | 240/281 [24:16<04:08,  6.06s/it]
Training Epoch: 16/100, batch 478/563 completed (loss: 0.0149:  85%|████████▌ | 240/281 [24:16<04:08,  6.06s/it]
Training Epoch: 16/100, batch 478/563 completed (loss: 0.1404:  85%|████████▌ | 240/281 [24:16<04:08,  6.06s/it]
Training Epoch: 16/100, batch 479/563 completed (loss: 0.0277:  85%|████████▌ | 240/281 [24:16<04:08,  6.06s/it]
Training Epoch: 16/100, batch 479/563 completed (loss: 0.0032:  85%|████████▌ | 240/281 [24:16<04:08,  6.06s/it]
Training Epoch: 16/100, batch 479/563 completed (loss: 0.0905:  85%|████████▌ | 240/281 [24:16<04:08,  6.06s/it]
Training Epoch: 16/100, batch 479/563 completed (loss: 0.0912:  85%|████████▌ | 240/281 [24:16<04:08,  6.06s/it]
Training Epoch: 16/100, batch 480/563 completed (loss: 0.0314:  85%|████████▌ | 240/281 [24:19<04:08,  6.06s/it]
Training Epoch: 16/100, batch 480/563 completed (loss: 0.3217:  85%|████████▌ | 240/281 [24:19<04:08,  6.06s/it]
Training Epoch: 16/100, batch 480/563 completed (loss: 0.0355:  85%|████████▌ | 240/281 [24:19<04:08,  6.06s/it]
Training Epoch: 16/100, batch 480/563 completed (loss: 0.0396:  85%|████████▌ | 240/281 [24:19<04:08,  6.06s/it]
Training Epoch: 16/100, batch 480/563 completed (loss: 0.3217:  86%|████████▌ | 241/281 [24:22<04:03,  6.09s/it]
Training Epoch: 16/100, batch 480/563 completed (loss: 0.0355:  86%|████████▌ | 241/281 [24:22<04:03,  6.09s/it]
Training Epoch: 16/100, batch 480/563 completed (loss: 0.0396:  86%|████████▌ | 241/281 [24:22<04:03,  6.09s/it]
Training Epoch: 16/100, batch 480/563 completed (loss: 0.0314:  86%|████████▌ | 241/281 [24:22<04:03,  6.09s/it]
Training Epoch: 16/100, batch 481/563 completed (loss: 0.0419:  86%|████████▌ | 241/281 [24:22<04:03,  6.09s/it]
Training Epoch: 16/100, batch 481/563 completed (loss: 0.0019:  86%|████████▌ | 241/281 [24:22<04:03,  6.09s/it]
Training Epoch: 16/100, batch 481/563 completed (loss: 0.0635:  86%|████████▌ | 241/281 [24:22<04:03,  6.09s/it]
Training Epoch: 16/100, batch 481/563 completed (loss: 0.0322:  86%|████████▌ | 241/281 [24:22<04:03,  6.09s/it]
Training Epoch: 16/100, batch 482/563 completed (loss: 0.1684:  86%|████████▌ | 241/281 [24:25<04:03,  6.09s/it]
Training Epoch: 16/100, batch 482/563 completed (loss: 0.0217:  86%|████████▌ | 241/281 [24:25<04:03,  6.09s/it]
Training Epoch: 16/100, batch 482/563 completed (loss: 0.1353:  86%|████████▌ | 241/281 [24:25<04:03,  6.09s/it]
Training Epoch: 16/100, batch 482/563 completed (loss: 0.1776:  86%|████████▌ | 241/281 [24:25<04:03,  6.09s/it]
Training Epoch: 16/100, batch 482/563 completed (loss: 0.1353:  86%|████████▌ | 242/281 [24:28<03:56,  6.07s/it]
Training Epoch: 16/100, batch 482/563 completed (loss: 0.1684:  86%|████████▌ | 242/281 [24:28<03:56,  6.07s/it]
Training Epoch: 16/100, batch 482/563 completed (loss: 0.1776:  86%|████████▌ | 242/281 [24:28<03:56,  6.07s/it]
Training Epoch: 16/100, batch 482/563 completed (loss: 0.0217:  86%|████████▌ | 242/281 [24:28<03:56,  6.07s/it]
Training Epoch: 16/100, batch 483/563 completed (loss: 0.1892:  86%|████████▌ | 242/281 [24:28<03:56,  6.07s/it]
Training Epoch: 16/100, batch 483/563 completed (loss: 0.5077:  86%|████████▌ | 242/281 [24:28<03:56,  6.07s/it]
Training Epoch: 16/100, batch 483/563 completed (loss: 0.0131:  86%|████████▌ | 242/281 [24:28<03:56,  6.07s/it]
Training Epoch: 16/100, batch 483/563 completed (loss: 0.0203:  86%|████████▌ | 242/281 [24:28<03:56,  6.07s/it]
Training Epoch: 16/100, batch 484/563 completed (loss: 0.2208:  86%|████████▌ | 242/281 [24:31<03:56,  6.07s/it]
Training Epoch: 16/100, batch 484/563 completed (loss: 0.0353:  86%|████████▌ | 242/281 [24:31<03:56,  6.07s/it]
Training Epoch: 16/100, batch 484/563 completed (loss: 0.1706:  86%|████████▌ | 242/281 [24:31<03:56,  6.07s/it]
Training Epoch: 16/100, batch 484/563 completed (loss: 0.1571:  86%|████████▌ | 242/281 [24:32<03:56,  6.07s/it]
Training Epoch: 16/100, batch 484/563 completed (loss: 0.2208:  86%|████████▋ | 243/281 [24:35<03:52,  6.11s/it]
Training Epoch: 16/100, batch 484/563 completed (loss: 0.1571:  86%|████████▋ | 243/281 [24:35<03:52,  6.11s/it]
Training Epoch: 16/100, batch 484/563 completed (loss: 0.0353:  86%|████████▋ | 243/281 [24:35<03:52,  6.11s/it]
Training Epoch: 16/100, batch 484/563 completed (loss: 0.1706:  86%|████████▋ | 243/281 [24:35<03:52,  6.11s/it]
Training Epoch: 16/100, batch 485/563 completed (loss: 0.0664:  86%|████████▋ | 243/281 [24:35<03:52,  6.11s/it]
Training Epoch: 16/100, batch 485/563 completed (loss: 0.4071:  86%|████████▋ | 243/281 [24:35<03:52,  6.11s/it]
Training Epoch: 16/100, batch 485/563 completed (loss: 0.0364:  86%|████████▋ | 243/281 [24:35<03:52,  6.11s/it]
Training Epoch: 16/100, batch 485/563 completed (loss: 0.0298:  86%|████████▋ | 243/281 [24:35<03:52,  6.11s/it]
Training Epoch: 16/100, batch 486/563 completed (loss: 0.0312:  86%|████████▋ | 243/281 [24:38<03:52,  6.11s/it]
Training Epoch: 16/100, batch 486/563 completed (loss: 0.1407:  86%|████████▋ | 243/281 [24:38<03:52,  6.11s/it]
Training Epoch: 16/100, batch 486/563 completed (loss: 0.0071:  86%|████████▋ | 243/281 [24:38<03:52,  6.11s/it]
Training Epoch: 16/100, batch 486/563 completed (loss: 0.0316:  86%|████████▋ | 243/281 [24:38<03:52,  6.11s/it]
Training Epoch: 16/100, batch 486/563 completed (loss: 0.1407:  87%|████████▋ | 244/281 [24:41<03:46,  6.12s/it]
Training Epoch: 16/100, batch 486/563 completed (loss: 0.0312:  87%|████████▋ | 244/281 [24:41<03:46,  6.12s/it]
Training Epoch: 16/100, batch 486/563 completed (loss: 0.0071:  87%|████████▋ | 244/281 [24:41<03:46,  6.12s/it]
Training Epoch: 16/100, batch 486/563 completed (loss: 0.0316:  87%|████████▋ | 244/281 [24:41<03:46,  6.12s/it]
Training Epoch: 16/100, batch 487/563 completed (loss: 0.0219:  87%|████████▋ | 244/281 [24:41<03:46,  6.12s/it]
Training Epoch: 16/100, batch 487/563 completed (loss: 0.0992:  87%|████████▋ | 244/281 [24:41<03:46,  6.12s/it]
Training Epoch: 16/100, batch 487/563 completed (loss: 0.1183:  87%|████████▋ | 244/281 [24:41<03:46,  6.12s/it]
Training Epoch: 16/100, batch 487/563 completed (loss: 0.3429:  87%|████████▋ | 244/281 [24:41<03:46,  6.12s/it]
Training Epoch: 16/100, batch 488/563 completed (loss: 0.0135:  87%|████████▋ | 244/281 [24:44<03:46,  6.12s/it]
Training Epoch: 16/100, batch 488/563 completed (loss: 0.0037:  87%|████████▋ | 244/281 [24:44<03:46,  6.12s/it]
Training Epoch: 16/100, batch 488/563 completed (loss: 0.4478:  87%|████████▋ | 244/281 [24:44<03:46,  6.12s/it]
Training Epoch: 16/100, batch 488/563 completed (loss: 0.0866:  87%|████████▋ | 244/281 [24:44<03:46,  6.12s/it]
Training Epoch: 16/100, batch 488/563 completed (loss: 0.0866:  87%|████████▋ | 245/281 [24:47<03:38,  6.08s/it]
Training Epoch: 16/100, batch 488/563 completed (loss: 0.0135:  87%|████████▋ | 245/281 [24:47<03:39,  6.09s/it]
Training Epoch: 16/100, batch 488/563 completed (loss: 0.0037:  87%|████████▋ | 245/281 [24:47<03:39,  6.09s/it]
Training Epoch: 16/100, batch 488/563 completed (loss: 0.4478:  87%|████████▋ | 245/281 [24:47<03:39,  6.09s/it]
Training Epoch: 16/100, batch 489/563 completed (loss: 0.012:  87%|████████▋ | 245/281 [24:47<03:39,  6.09s/it] 
Training Epoch: 16/100, batch 489/563 completed (loss: 0.0073:  87%|████████▋ | 245/281 [24:47<03:39,  6.09s/it]
Training Epoch: 16/100, batch 489/563 completed (loss: 0.0848:  87%|████████▋ | 245/281 [24:47<03:39,  6.09s/it]
Training Epoch: 16/100, batch 489/563 completed (loss: 0.0059:  87%|████████▋ | 245/281 [24:47<03:38,  6.08s/it]
Training Epoch: 16/100, batch 490/563 completed (loss: 0.3668:  87%|████████▋ | 245/281 [24:50<03:39,  6.09s/it]
Training Epoch: 16/100, batch 490/563 completed (loss: 0.0663:  87%|████████▋ | 245/281 [24:50<03:39,  6.09s/it]
Training Epoch: 16/100, batch 490/563 completed (loss: 0.1138:  87%|████████▋ | 245/281 [24:50<03:39,  6.09s/it]
Training Epoch: 16/100, batch 490/563 completed (loss: 0.1913:  87%|████████▋ | 245/281 [24:50<03:38,  6.08s/it]
Training Epoch: 16/100, batch 490/563 completed (loss: 0.1913:  88%|████████▊ | 246/281 [24:53<03:33,  6.11s/it]
Training Epoch: 16/100, batch 490/563 completed (loss: 0.0663:  88%|████████▊ | 246/281 [24:53<03:33,  6.11s/it]
Training Epoch: 16/100, batch 490/563 completed (loss: 0.3668:  88%|████████▊ | 246/281 [24:53<03:33,  6.11s/it]
Training Epoch: 16/100, batch 490/563 completed (loss: 0.1138:  88%|████████▊ | 246/281 [24:53<03:33,  6.11s/it]
Training Epoch: 16/100, batch 491/563 completed (loss: 0.0099:  88%|████████▊ | 246/281 [24:53<03:33,  6.11s/it]
Training Epoch: 16/100, batch 491/563 completed (loss: 0.0621:  88%|████████▊ | 246/281 [24:53<03:33,  6.11s/it]
Training Epoch: 16/100, batch 491/563 completed (loss: 0.255:  88%|████████▊ | 246/281 [24:53<03:33,  6.11s/it] 
Training Epoch: 16/100, batch 491/563 completed (loss: 0.0116:  88%|████████▊ | 246/281 [24:53<03:33,  6.11s/it]
Training Epoch: 16/100, batch 492/563 completed (loss: 0.0483:  88%|████████▊ | 246/281 [24:56<03:33,  6.11s/it]
Training Epoch: 16/100, batch 492/563 completed (loss: 0.0233:  88%|████████▊ | 246/281 [24:56<03:33,  6.11s/it]
Training Epoch: 16/100, batch 492/563 completed (loss: 0.021:  88%|████████▊ | 246/281 [24:56<03:33,  6.11s/it] 
Training Epoch: 16/100, batch 492/563 completed (loss: 0.042:  88%|████████▊ | 246/281 [24:56<03:33,  6.11s/it] 
Training Epoch: 16/100, batch 492/563 completed (loss: 0.021:  88%|████████▊ | 247/281 [24:59<03:26,  6.08s/it]
Training Epoch: 16/100, batch 492/563 completed (loss: 0.0483:  88%|████████▊ | 247/281 [24:59<03:26,  6.08s/it]
Training Epoch: 16/100, batch 492/563 completed (loss: 0.0233:  88%|████████▊ | 247/281 [24:59<03:26,  6.08s/it]
Training Epoch: 16/100, batch 492/563 completed (loss: 0.042:  88%|████████▊ | 247/281 [24:59<03:26,  6.08s/it]
Training Epoch: 16/100, batch 493/563 completed (loss: 0.0875:  88%|████████▊ | 247/281 [24:59<03:26,  6.08s/it]
Training Epoch: 16/100, batch 493/563 completed (loss: 0.0156:  88%|████████▊ | 247/281 [24:59<03:26,  6.08s/it]
Training Epoch: 16/100, batch 493/563 completed (loss: 0.3087:  88%|████████▊ | 247/281 [24:59<03:26,  6.08s/it]
Training Epoch: 16/100, batch 493/563 completed (loss: 0.0116:  88%|████████▊ | 247/281 [24:59<03:26,  6.08s/it]
Training Epoch: 16/100, batch 494/563 completed (loss: 0.0717:  88%|████████▊ | 247/281 [25:02<03:26,  6.08s/it]
Training Epoch: 16/100, batch 494/563 completed (loss: 0.0221:  88%|████████▊ | 247/281 [25:02<03:26,  6.08s/it]
Training Epoch: 16/100, batch 494/563 completed (loss: 0.1855:  88%|████████▊ | 247/281 [25:02<03:26,  6.08s/it]
Training Epoch: 16/100, batch 494/563 completed (loss: 0.0603:  88%|████████▊ | 247/281 [25:02<03:26,  6.08s/it]
Training Epoch: 16/100, batch 494/563 completed (loss: 0.1855:  88%|████████▊ | 248/281 [25:05<03:19,  6.06s/it]
Training Epoch: 16/100, batch 494/563 completed (loss: 0.0717:  88%|████████▊ | 248/281 [25:05<03:19,  6.06s/it]
Training Epoch: 16/100, batch 494/563 completed (loss: 0.0221:  88%|████████▊ | 248/281 [25:05<03:19,  6.06s/it]
Training Epoch: 16/100, batch 494/563 completed (loss: 0.0603:  88%|████████▊ | 248/281 [25:05<03:19,  6.06s/it]
Training Epoch: 16/100, batch 495/563 completed (loss: 0.034:  88%|████████▊ | 248/281 [25:05<03:19,  6.06s/it] 
Training Epoch: 16/100, batch 495/563 completed (loss: 0.0172:  88%|████████▊ | 248/281 [25:05<03:19,  6.06s/it]
Training Epoch: 16/100, batch 495/563 completed (loss: 0.1039:  88%|████████▊ | 248/281 [25:05<03:19,  6.06s/it]
Training Epoch: 16/100, batch 495/563 completed (loss: 0.0614:  88%|████████▊ | 248/281 [25:05<03:19,  6.06s/it]
Training Epoch: 16/100, batch 496/563 completed (loss: 0.0092:  88%|████████▊ | 248/281 [25:08<03:19,  6.06s/it]
Training Epoch: 16/100, batch 496/563 completed (loss: 0.0173:  88%|████████▊ | 248/281 [25:08<03:19,  6.06s/it]
Training Epoch: 16/100, batch 496/563 completed (loss: 0.1281:  88%|████████▊ | 248/281 [25:08<03:19,  6.06s/it]
Training Epoch: 16/100, batch 496/563 completed (loss: 0.0023:  88%|████████▊ | 248/281 [25:08<03:19,  6.06s/it]
Training Epoch: 16/100, batch 496/563 completed (loss: 0.0023:  89%|████████▊ | 249/281 [25:11<03:15,  6.10s/it]
Training Epoch: 16/100, batch 496/563 completed (loss: 0.0173:  89%|████████▊ | 249/281 [25:11<03:15,  6.10s/it]
Training Epoch: 16/100, batch 496/563 completed (loss: 0.0092:  89%|████████▊ | 249/281 [25:11<03:15,  6.10s/it]
Training Epoch: 16/100, batch 496/563 completed (loss: 0.1281:  89%|████████▊ | 249/281 [25:11<03:15,  6.10s/it]
Training Epoch: 16/100, batch 497/563 completed (loss: 0.1149:  89%|████████▊ | 249/281 [25:11<03:15,  6.10s/it]
Training Epoch: 16/100, batch 497/563 completed (loss: 0.0155:  89%|████████▊ | 249/281 [25:11<03:15,  6.10s/it]
Training Epoch: 16/100, batch 497/563 completed (loss: 0.1525:  89%|████████▊ | 249/281 [25:11<03:15,  6.10s/it]
Training Epoch: 16/100, batch 497/563 completed (loss: 0.0842:  89%|████████▊ | 249/281 [25:11<03:15,  6.10s/it]
Training Epoch: 16/100, batch 498/563 completed (loss: 0.0044:  89%|████████▊ | 249/281 [25:14<03:15,  6.10s/it]
Training Epoch: 16/100, batch 498/563 completed (loss: 0.0286:  89%|████████▊ | 249/281 [25:14<03:15,  6.10s/it]
Training Epoch: 16/100, batch 498/563 completed (loss: 0.0515:  89%|████████▊ | 249/281 [25:14<03:15,  6.10s/it]
Training Epoch: 16/100, batch 498/563 completed (loss: 0.1627:  89%|████████▊ | 249/281 [25:14<03:15,  6.10s/it]
Training Epoch: 16/100, batch 498/563 completed (loss: 0.0286:  89%|████████▉ | 250/281 [25:17<03:06,  6.03s/it]
Training Epoch: 16/100, batch 498/563 completed (loss: 0.0515:  89%|████████▉ | 250/281 [25:17<03:06,  6.03s/it]
Training Epoch: 16/100, batch 498/563 completed (loss: 0.0044:  89%|████████▉ | 250/281 [25:17<03:06,  6.03s/it]
Training Epoch: 16/100, batch 498/563 completed (loss: 0.1627:  89%|████████▉ | 250/281 [25:17<03:07,  6.03s/it]
Training Epoch: 16/100, batch 499/563 completed (loss: 0.0981:  89%|████████▉ | 250/281 [25:17<03:06,  6.03s/it]
Training Epoch: 16/100, batch 499/563 completed (loss: 0.0167:  89%|████████▉ | 250/281 [25:17<03:06,  6.03s/it]
Training Epoch: 16/100, batch 499/563 completed (loss: 0.0264:  89%|████████▉ | 250/281 [25:17<03:06,  6.03s/it]
Training Epoch: 16/100, batch 499/563 completed (loss: 0.1407:  89%|████████▉ | 250/281 [25:17<03:07,  6.03s/it]
Training Epoch: 16/100, batch 500/563 completed (loss: 0.0108:  89%|████████▉ | 250/281 [25:20<03:06,  6.03s/it]
Training Epoch: 16/100, batch 500/563 completed (loss: 0.083:  89%|████████▉ | 250/281 [25:20<03:06,  6.03s/it] 
Training Epoch: 16/100, batch 500/563 completed (loss: 0.1486:  89%|████████▉ | 250/281 [25:20<03:06,  6.03s/it]
Training Epoch: 16/100, batch 500/563 completed (loss: 0.0749:  89%|████████▉ | 250/281 [25:20<03:07,  6.03s/it]
Training Epoch: 16/100, batch 500/563 completed (loss: 0.0108:  89%|████████▉ | 251/281 [25:23<03:02,  6.07s/it]
Training Epoch: 16/100, batch 500/563 completed (loss: 0.1486:  89%|████████▉ | 251/281 [25:23<03:02,  6.07s/it]
Training Epoch: 16/100, batch 500/563 completed (loss: 0.0749:  89%|████████▉ | 251/281 [25:23<03:02,  6.07s/it]
Training Epoch: 16/100, batch 500/563 completed (loss: 0.083:  89%|████████▉ | 251/281 [25:23<03:02,  6.07s/it]
Training Epoch: 16/100, batch 501/563 completed (loss: 0.0026:  89%|████████▉ | 251/281 [25:23<03:02,  6.07s/it]
Training Epoch: 16/100, batch 501/563 completed (loss: 0.0865:  89%|████████▉ | 251/281 [25:23<03:02,  6.07s/it]
Training Epoch: 16/100, batch 501/563 completed (loss: 0.0718:  89%|████████▉ | 251/281 [25:23<03:02,  6.07s/it]
Training Epoch: 16/100, batch 501/563 completed (loss: 0.0134:  89%|████████▉ | 251/281 [25:23<03:02,  6.07s/it]
Training Epoch: 16/100, batch 502/563 completed (loss: 0.0168:  89%|████████▉ | 251/281 [25:26<03:02,  6.07s/it]
Training Epoch: 16/100, batch 502/563 completed (loss: 0.3563:  89%|████████▉ | 251/281 [25:26<03:02,  6.07s/it]
Training Epoch: 16/100, batch 502/563 completed (loss: 0.2431:  89%|████████▉ | 251/281 [25:26<03:02,  6.07s/it]
Training Epoch: 16/100, batch 502/563 completed (loss: 0.0043:  89%|████████▉ | 251/281 [25:26<03:02,  6.07s/it]
Training Epoch: 16/100, batch 502/563 completed (loss: 0.2431:  90%|████████▉ | 252/281 [25:29<02:56,  6.09s/it]
Training Epoch: 16/100, batch 502/563 completed (loss: 0.0168:  90%|████████▉ | 252/281 [25:29<02:56,  6.09s/it]
Training Epoch: 16/100, batch 502/563 completed (loss: 0.3563:  90%|████████▉ | 252/281 [25:29<02:56,  6.10s/it]
Training Epoch: 16/100, batch 502/563 completed (loss: 0.0043:  90%|████████▉ | 252/281 [25:29<02:56,  6.10s/it]
Training Epoch: 16/100, batch 503/563 completed (loss: 0.027:  90%|████████▉ | 252/281 [25:29<02:56,  6.10s/it] 
Training Epoch: 16/100, batch 503/563 completed (loss: 0.0126:  90%|████████▉ | 252/281 [25:29<02:56,  6.09s/it]
Training Epoch: 16/100, batch 503/563 completed (loss: 0.082:  90%|████████▉ | 252/281 [25:29<02:56,  6.09s/it] 
Training Epoch: 16/100, batch 503/563 completed (loss: 0.6019:  90%|████████▉ | 252/281 [25:29<02:56,  6.10s/it]
Training Epoch: 16/100, batch 504/563 completed (loss: 0.0017:  90%|████████▉ | 252/281 [25:32<02:56,  6.09s/it]
Training Epoch: 16/100, batch 504/563 completed (loss: 0.2629:  90%|████████▉ | 252/281 [25:32<02:56,  6.09s/it]
Training Epoch: 16/100, batch 504/563 completed (loss: 0.0108:  90%|████████▉ | 252/281 [25:32<02:56,  6.10s/it]
Training Epoch: 16/100, batch 504/563 completed (loss: 0.3661:  90%|████████▉ | 252/281 [25:32<02:56,  6.10s/it]
Training Epoch: 16/100, batch 504/563 completed (loss: 0.3661:  90%|█████████ | 253/281 [25:35<02:49,  6.07s/it]
Training Epoch: 16/100, batch 504/563 completed (loss: 0.0108:  90%|█████████ | 253/281 [25:35<02:50,  6.07s/it]
Training Epoch: 16/100, batch 504/563 completed (loss: 0.2629:  90%|█████████ | 253/281 [25:35<02:50,  6.08s/it]
Training Epoch: 16/100, batch 504/563 completed (loss: 0.0017:  90%|█████████ | 253/281 [25:35<02:50,  6.08s/it]
Training Epoch: 16/100, batch 505/563 completed (loss: 0.0276:  90%|█████████ | 253/281 [25:35<02:50,  6.07s/it]
Training Epoch: 16/100, batch 505/563 completed (loss: 0.0336:  90%|█████████ | 253/281 [25:35<02:50,  6.08s/it]
Training Epoch: 16/100, batch 505/563 completed (loss: 0.0075:  90%|█████████ | 253/281 [25:35<02:50,  6.08s/it]
Training Epoch: 16/100, batch 505/563 completed (loss: 0.0135:  90%|█████████ | 253/281 [25:35<02:49,  6.07s/it]
Training Epoch: 16/100, batch 506/563 completed (loss: 0.0438:  90%|█████████ | 253/281 [25:38<02:50,  6.08s/it]
Training Epoch: 16/100, batch 506/563 completed (loss: 0.1198:  90%|█████████ | 253/281 [25:38<02:50,  6.07s/it]
Training Epoch: 16/100, batch 506/563 completed (loss: 0.1258:  90%|█████████ | 253/281 [25:38<02:50,  6.08s/it]
Training Epoch: 16/100, batch 506/563 completed (loss: 0.0678:  90%|█████████ | 253/281 [25:38<02:49,  6.07s/it]
Training Epoch: 16/100, batch 506/563 completed (loss: 0.0438:  90%|█████████ | 254/281 [25:41<02:44,  6.11s/it]
Training Epoch: 16/100, batch 506/563 completed (loss: 0.1258:  90%|█████████ | 254/281 [25:41<02:44,  6.11s/it]
Training Epoch: 16/100, batch 506/563 completed (loss: 0.0678:  90%|█████████ | 254/281 [25:42<02:45,  6.11s/it]
Training Epoch: 16/100, batch 506/563 completed (loss: 0.1198:  90%|█████████ | 254/281 [25:41<02:45,  6.12s/it]
Training Epoch: 16/100, batch 507/563 completed (loss: 0.4866:  90%|█████████ | 254/281 [25:42<02:44,  6.11s/it]
Training Epoch: 16/100, batch 507/563 completed (loss: 0.0375:  90%|█████████ | 254/281 [25:42<02:45,  6.12s/it]
Training Epoch: 16/100, batch 507/563 completed (loss: 0.0425:  90%|█████████ | 254/281 [25:42<02:44,  6.11s/it]
Training Epoch: 16/100, batch 507/563 completed (loss: 0.0704:  90%|█████████ | 254/281 [25:42<02:45,  6.11s/it]
Training Epoch: 16/100, batch 508/563 completed (loss: 0.0827:  90%|█████████ | 254/281 [25:45<02:45,  6.12s/it]
Training Epoch: 16/100, batch 508/563 completed (loss: 0.0536:  90%|█████████ | 254/281 [25:45<02:44,  6.11s/it]
Training Epoch: 16/100, batch 508/563 completed (loss: 0.0627:  90%|█████████ | 254/281 [25:45<02:44,  6.11s/it]
Training Epoch: 16/100, batch 508/563 completed (loss: 0.0421:  90%|█████████ | 254/281 [25:45<02:45,  6.11s/it]
Training Epoch: 16/100, batch 508/563 completed (loss: 0.0827:  91%|█████████ | 255/281 [25:47<02:37,  6.08s/it]
Training Epoch: 16/100, batch 508/563 completed (loss: 0.0536:  91%|█████████ | 255/281 [25:47<02:38,  6.08s/it]
Training Epoch: 16/100, batch 508/563 completed (loss: 0.0627:  91%|█████████ | 255/281 [25:47<02:38,  6.08s/it]
Training Epoch: 16/100, batch 508/563 completed (loss: 0.0421:  91%|█████████ | 255/281 [25:48<02:38,  6.08s/it]
Training Epoch: 16/100, batch 509/563 completed (loss: 0.1028:  91%|█████████ | 255/281 [25:48<02:37,  6.08s/it]
Training Epoch: 16/100, batch 509/563 completed (loss: 0.0036:  91%|█████████ | 255/281 [25:48<02:38,  6.08s/it]
Training Epoch: 16/100, batch 509/563 completed (loss: 0.1981:  91%|█████████ | 255/281 [25:48<02:38,  6.08s/it]
Training Epoch: 16/100, batch 509/563 completed (loss: 0.0151:  91%|█████████ | 255/281 [25:48<02:38,  6.08s/it]
Training Epoch: 16/100, batch 510/563 completed (loss: 0.2424:  91%|█████████ | 255/281 [25:51<02:37,  6.08s/it]
Training Epoch: 16/100, batch 510/563 completed (loss: 0.0437:  91%|█████████ | 255/281 [25:51<02:38,  6.08s/it]
Training Epoch: 16/100, batch 510/563 completed (loss: 0.1898:  91%|█████████ | 255/281 [25:51<02:38,  6.08s/it]
Training Epoch: 16/100, batch 510/563 completed (loss: 0.1296:  91%|█████████ | 255/281 [25:51<02:38,  6.08s/it]
Training Epoch: 16/100, batch 510/563 completed (loss: 0.1898:  91%|█████████ | 256/281 [25:53<02:31,  6.06s/it]
Training Epoch: 16/100, batch 510/563 completed (loss: 0.0437:  91%|█████████ | 256/281 [25:53<02:31,  6.06s/it]
Training Epoch: 16/100, batch 510/563 completed (loss: 0.1296:  91%|█████████ | 256/281 [25:54<02:31,  6.06s/it]
Training Epoch: 16/100, batch 510/563 completed (loss: 0.2424:  91%|█████████ | 256/281 [25:53<02:31,  6.06s/it]
Training Epoch: 16/100, batch 511/563 completed (loss: 0.0355:  91%|█████████ | 256/281 [25:54<02:31,  6.06s/it]
Training Epoch: 16/100, batch 511/563 completed (loss: 0.0527:  91%|█████████ | 256/281 [25:54<02:31,  6.06s/it]
Training Epoch: 16/100, batch 511/563 completed (loss: 0.0532:  91%|█████████ | 256/281 [25:54<02:31,  6.06s/it]
Training Epoch: 16/100, batch 511/563 completed (loss: 0.1601:  91%|█████████ | 256/281 [25:54<02:31,  6.06s/it]
Training Epoch: 16/100, batch 512/563 completed (loss: 0.0545:  91%|█████████ | 256/281 [25:57<02:31,  6.06s/it]
Training Epoch: 16/100, batch 512/563 completed (loss: 0.0729:  91%|█████████ | 256/281 [25:57<02:31,  6.06s/it]
Training Epoch: 16/100, batch 512/563 completed (loss: 0.025:  91%|█████████ | 256/281 [25:57<02:31,  6.06s/it] 
Training Epoch: 16/100, batch 512/563 completed (loss: 0.0222:  91%|█████████ | 256/281 [25:57<02:31,  6.06s/it]
Training Epoch: 16/100, batch 512/563 completed (loss: 0.0545:  91%|█████████▏| 257/281 [26:00<02:27,  6.13s/it]
Training Epoch: 16/100, batch 512/563 completed (loss: 0.0729:  91%|█████████▏| 257/281 [26:00<02:27,  6.14s/it]
Training Epoch: 16/100, batch 512/563 completed (loss: 0.025:  91%|█████████▏| 257/281 [26:00<02:27,  6.14s/it]
Training Epoch: 16/100, batch 512/563 completed (loss: 0.0222:  91%|█████████▏| 257/281 [26:00<02:27,  6.14s/it]
Training Epoch: 16/100, batch 513/563 completed (loss: 0.0173:  91%|█████████▏| 257/281 [26:00<02:27,  6.13s/it]
Training Epoch: 16/100, batch 513/563 completed (loss: 0.2232:  91%|█████████▏| 257/281 [26:00<02:27,  6.14s/it]
Training Epoch: 16/100, batch 513/563 completed (loss: 0.0077:  91%|█████████▏| 257/281 [26:00<02:27,  6.14s/it]
Training Epoch: 16/100, batch 513/563 completed (loss: 0.5171:  91%|█████████▏| 257/281 [26:00<02:27,  6.14s/it]
Training Epoch: 16/100, batch 514/563 completed (loss: 0.4157:  91%|█████████▏| 257/281 [26:03<02:27,  6.14s/it]
Training Epoch: 16/100, batch 514/563 completed (loss: 0.1708:  91%|█████████▏| 257/281 [26:03<02:27,  6.13s/it]
Training Epoch: 16/100, batch 514/563 completed (loss: 0.0827:  91%|█████████▏| 257/281 [26:03<02:27,  6.14s/it]
Training Epoch: 16/100, batch 514/563 completed (loss: 0.1288:  91%|█████████▏| 257/281 [26:03<02:27,  6.14s/it]
Training Epoch: 16/100, batch 514/563 completed (loss: 0.1708:  92%|█████████▏| 258/281 [26:06<02:20,  6.09s/it]
Training Epoch: 16/100, batch 514/563 completed (loss: 0.0827:  92%|█████████▏| 258/281 [26:06<02:20,  6.09s/it]
Training Epoch: 16/100, batch 514/563 completed (loss: 0.1288:  92%|█████████▏| 258/281 [26:06<02:20,  6.09s/it]
Training Epoch: 16/100, batch 514/563 completed (loss: 0.4157:  92%|█████████▏| 258/281 [26:06<02:20,  6.09s/it]
Training Epoch: 16/100, batch 515/563 completed (loss: 0.0293:  92%|█████████▏| 258/281 [26:06<02:20,  6.09s/it]
Training Epoch: 16/100, batch 515/563 completed (loss: 0.2362:  92%|█████████▏| 258/281 [26:06<02:20,  6.09s/it]
Training Epoch: 16/100, batch 515/563 completed (loss: 0.0206:  92%|█████████▏| 258/281 [26:06<02:20,  6.09s/it]
Training Epoch: 16/100, batch 515/563 completed (loss: 0.0173:  92%|█████████▏| 258/281 [26:06<02:20,  6.09s/it]
Training Epoch: 16/100, batch 516/563 completed (loss: 0.081:  92%|█████████▏| 258/281 [26:09<02:20,  6.09s/it] 
Training Epoch: 16/100, batch 516/563 completed (loss: 0.0452:  92%|█████████▏| 258/281 [26:09<02:20,  6.09s/it]
Training Epoch: 16/100, batch 516/563 completed (loss: 0.0275:  92%|█████████▏| 258/281 [26:09<02:20,  6.09s/it]
Training Epoch: 16/100, batch 516/563 completed (loss: 0.0358:  92%|█████████▏| 258/281 [26:09<02:20,  6.09s/it]
Training Epoch: 16/100, batch 516/563 completed (loss: 0.081:  92%|█████████▏| 259/281 [26:12<02:13,  6.06s/it]
Training Epoch: 16/100, batch 516/563 completed (loss: 0.0452:  92%|█████████▏| 259/281 [26:12<02:13,  6.06s/it]
Training Epoch: 16/100, batch 516/563 completed (loss: 0.0358:  92%|█████████▏| 259/281 [26:12<02:13,  6.06s/it]
Training Epoch: 16/100, batch 516/563 completed (loss: 0.0275:  92%|█████████▏| 259/281 [26:12<02:13,  6.07s/it]
Training Epoch: 16/100, batch 517/563 completed (loss: 0.0145:  92%|█████████▏| 259/281 [26:12<02:13,  6.06s/it]
Training Epoch: 16/100, batch 517/563 completed (loss: 0.034:  92%|█████████▏| 259/281 [26:12<02:13,  6.07s/it] 
Training Epoch: 16/100, batch 517/563 completed (loss: 0.1622:  92%|█████████▏| 259/281 [26:12<02:13,  6.06s/it]
Training Epoch: 16/100, batch 517/563 completed (loss: 0.0024:  92%|█████████▏| 259/281 [26:12<02:13,  6.06s/it]
Training Epoch: 16/100, batch 518/563 completed (loss: 0.2488:  92%|█████████▏| 259/281 [26:15<02:13,  6.06s/it]
Training Epoch: 16/100, batch 518/563 completed (loss: 0.0482:  92%|█████████▏| 259/281 [26:15<02:13,  6.07s/it]
Training Epoch: 16/100, batch 518/563 completed (loss: 0.0816:  92%|█████████▏| 259/281 [26:15<02:13,  6.06s/it]
Training Epoch: 16/100, batch 518/563 completed (loss: 0.043:  92%|█████████▏| 259/281 [26:15<02:13,  6.06s/it] 
Training Epoch: 16/100, batch 518/563 completed (loss: 0.0816:  93%|█████████▎| 260/281 [26:18<02:07,  6.05s/it]
Training Epoch: 16/100, batch 518/563 completed (loss: 0.2488:  93%|█████████▎| 260/281 [26:18<02:07,  6.05s/it]
Training Epoch: 16/100, batch 518/563 completed (loss: 0.0482:  93%|█████████▎| 260/281 [26:18<02:07,  6.05s/it]
Training Epoch: 16/100, batch 518/563 completed (loss: 0.043:  93%|█████████▎| 260/281 [26:18<02:07,  6.05s/it]
Training Epoch: 16/100, batch 519/563 completed (loss: 0.1353:  93%|█████████▎| 260/281 [26:18<02:07,  6.05s/it]
Training Epoch: 16/100, batch 519/563 completed (loss: 0.047:  93%|█████████▎| 260/281 [26:18<02:07,  6.05s/it] 
Training Epoch: 16/100, batch 519/563 completed (loss: 0.0531:  93%|█████████▎| 260/281 [26:18<02:07,  6.05s/it]
Training Epoch: 16/100, batch 519/563 completed (loss: 0.113:  93%|█████████▎| 260/281 [26:18<02:07,  6.05s/it]
Training Epoch: 16/100, batch 520/563 completed (loss: 0.0213:  93%|█████████▎| 260/281 [26:21<02:07,  6.05s/it]
Training Epoch: 16/100, batch 520/563 completed (loss: 0.0788:  93%|█████████▎| 260/281 [26:21<02:07,  6.05s/it]
Training Epoch: 16/100, batch 520/563 completed (loss: 0.0595:  93%|█████████▎| 260/281 [26:21<02:07,  6.05s/it]
Training Epoch: 16/100, batch 520/563 completed (loss: 0.2056:  93%|█████████▎| 260/281 [26:21<02:07,  6.05s/it]
Training Epoch: 16/100, batch 520/563 completed (loss: 0.0788:  93%|█████████▎| 261/281 [26:24<02:00,  6.04s/it]
Training Epoch: 16/100, batch 520/563 completed (loss: 0.0213:  93%|█████████▎| 261/281 [26:24<02:00,  6.04s/it]
Training Epoch: 16/100, batch 520/563 completed (loss: 0.0595:  93%|█████████▎| 261/281 [26:24<02:00,  6.04s/it]
Training Epoch: 16/100, batch 520/563 completed (loss: 0.2056:  93%|█████████▎| 261/281 [26:24<02:00,  6.04s/it]
Training Epoch: 16/100, batch 521/563 completed (loss: 0.071:  93%|█████████▎| 261/281 [26:24<02:00,  6.04s/it] 
Training Epoch: 16/100, batch 521/563 completed (loss: 0.1545:  93%|█████████▎| 261/281 [26:24<02:00,  6.04s/it]
Training Epoch: 16/100, batch 521/563 completed (loss: 0.2083:  93%|█████████▎| 261/281 [26:24<02:00,  6.04s/it]
Training Epoch: 16/100, batch 521/563 completed (loss: 0.1372:  93%|█████████▎| 261/281 [26:24<02:00,  6.04s/it]
Training Epoch: 16/100, batch 522/563 completed (loss: 0.1232:  93%|█████████▎| 261/281 [26:27<02:00,  6.04s/it]
Training Epoch: 16/100, batch 522/563 completed (loss: 0.0797:  93%|█████████▎| 261/281 [26:27<02:00,  6.04s/it]
Training Epoch: 16/100, batch 522/563 completed (loss: 0.1156:  93%|█████████▎| 261/281 [26:27<02:00,  6.04s/it]
Training Epoch: 16/100, batch 522/563 completed (loss: 0.0061:  93%|█████████▎| 261/281 [26:27<02:00,  6.04s/it]
Training Epoch: 16/100, batch 522/563 completed (loss: 0.0797:  93%|█████████▎| 262/281 [26:30<01:55,  6.08s/it]
Training Epoch: 16/100, batch 522/563 completed (loss: 0.1232:  93%|█████████▎| 262/281 [26:30<01:55,  6.08s/it]
Training Epoch: 16/100, batch 522/563 completed (loss: 0.1156:  93%|█████████▎| 262/281 [26:30<01:55,  6.08s/it]
Training Epoch: 16/100, batch 522/563 completed (loss: 0.0061:  93%|█████████▎| 262/281 [26:30<01:55,  6.08s/it]
Training Epoch: 16/100, batch 523/563 completed (loss: 0.2253:  93%|█████████▎| 262/281 [26:30<01:55,  6.08s/it]
Training Epoch: 16/100, batch 523/563 completed (loss: 0.2244:  93%|█████████▎| 262/281 [26:30<01:55,  6.08s/it]
Training Epoch: 16/100, batch 523/563 completed (loss: 0.0181:  93%|█████████▎| 262/281 [26:30<01:55,  6.08s/it]
Training Epoch: 16/100, batch 523/563 completed (loss: 0.0396:  93%|█████████▎| 262/281 [26:30<01:55,  6.08s/it]
Training Epoch: 16/100, batch 524/563 completed (loss: 0.0138:  93%|█████████▎| 262/281 [26:33<01:55,  6.08s/it]
Training Epoch: 16/100, batch 524/563 completed (loss: 0.0325:  93%|█████████▎| 262/281 [26:33<01:55,  6.08s/it]
Training Epoch: 16/100, batch 524/563 completed (loss: 0.033:  93%|█████████▎| 262/281 [26:33<01:55,  6.08s/it] 
Training Epoch: 16/100, batch 524/563 completed (loss: 0.263:  93%|█████████▎| 262/281 [26:33<01:55,  6.08s/it] 
Training Epoch: 16/100, batch 524/563 completed (loss: 0.033:  94%|█████████▎| 263/281 [26:36<01:49,  6.06s/it]
Training Epoch: 16/100, batch 524/563 completed (loss: 0.0138:  94%|█████████▎| 263/281 [26:36<01:49,  6.06s/it]
Training Epoch: 16/100, batch 524/563 completed (loss: 0.0325:  94%|█████████▎| 263/281 [26:36<01:49,  6.07s/it]
Training Epoch: 16/100, batch 524/563 completed (loss: 0.263:  94%|█████████▎| 263/281 [26:36<01:49,  6.07s/it]
Training Epoch: 16/100, batch 525/563 completed (loss: 0.014:  94%|█████████▎| 263/281 [26:36<01:49,  6.07s/it] 
Training Epoch: 16/100, batch 525/563 completed (loss: 0.0906:  94%|█████████▎| 263/281 [26:36<01:49,  6.06s/it]
Training Epoch: 16/100, batch 525/563 completed (loss: 0.0108:  94%|█████████▎| 263/281 [26:36<01:49,  6.06s/it]
Training Epoch: 16/100, batch 525/563 completed (loss: 0.2995:  94%|█████████▎| 263/281 [26:36<01:49,  6.07s/it]
Training Epoch: 16/100, batch 526/563 completed (loss: 0.0064:  94%|█████████▎| 263/281 [26:39<01:49,  6.06s/it]
Training Epoch: 16/100, batch 526/563 completed (loss: 0.1216:  94%|█████████▎| 263/281 [26:39<01:49,  6.06s/it]
Training Epoch: 16/100, batch 526/563 completed (loss: 0.1238:  94%|█████████▎| 263/281 [26:39<01:49,  6.07s/it]
Training Epoch: 16/100, batch 526/563 completed (loss: 0.0023:  94%|█████████▎| 263/281 [26:39<01:49,  6.07s/it]
Training Epoch: 16/100, batch 526/563 completed (loss: 0.0064:  94%|█████████▍| 264/281 [26:42<01:43,  6.09s/it]
Training Epoch: 16/100, batch 526/563 completed (loss: 0.1216:  94%|█████████▍| 264/281 [26:42<01:43,  6.09s/it]
Training Epoch: 16/100, batch 526/563 completed (loss: 0.1238:  94%|█████████▍| 264/281 [26:42<01:43,  6.09s/it]
Training Epoch: 16/100, batch 526/563 completed (loss: 0.0023:  94%|█████████▍| 264/281 [26:42<01:43,  6.09s/it]
Training Epoch: 16/100, batch 527/563 completed (loss: 0.5627:  94%|█████████▍| 264/281 [26:42<01:43,  6.09s/it]
Training Epoch: 16/100, batch 527/563 completed (loss: 0.0256:  94%|█████████▍| 264/281 [26:42<01:43,  6.09s/it]
Training Epoch: 16/100, batch 527/563 completed (loss: 0.0447:  94%|█████████▍| 264/281 [26:42<01:43,  6.09s/it]
Training Epoch: 16/100, batch 527/563 completed (loss: 0.0278:  94%|█████████▍| 264/281 [26:42<01:43,  6.09s/it]
Training Epoch: 16/100, batch 528/563 completed (loss: 0.0178:  94%|█████████▍| 264/281 [26:45<01:43,  6.09s/it]
Training Epoch: 16/100, batch 528/563 completed (loss: 0.0582:  94%|█████████▍| 264/281 [26:45<01:43,  6.09s/it]
Training Epoch: 16/100, batch 528/563 completed (loss: 0.1072:  94%|█████████▍| 264/281 [26:45<01:43,  6.09s/it]
Training Epoch: 16/100, batch 528/563 completed (loss: 0.1325:  94%|█████████▍| 264/281 [26:45<01:43,  6.09s/it]
Training Epoch: 16/100, batch 528/563 completed (loss: 0.0582:  94%|█████████▍| 265/281 [26:48<01:37,  6.11s/it]
Training Epoch: 16/100, batch 528/563 completed (loss: 0.1072:  94%|█████████▍| 265/281 [26:48<01:37,  6.11s/it]
Training Epoch: 16/100, batch 528/563 completed (loss: 0.0178:  94%|█████████▍| 265/281 [26:48<01:37,  6.11s/it]
Training Epoch: 16/100, batch 528/563 completed (loss: 0.1325:  94%|█████████▍| 265/281 [26:48<01:37,  6.11s/it]
Training Epoch: 16/100, batch 529/563 completed (loss: 0.1634:  94%|█████████▍| 265/281 [26:48<01:37,  6.11s/it]
Training Epoch: 16/100, batch 529/563 completed (loss: 0.0351:  94%|█████████▍| 265/281 [26:48<01:37,  6.11s/it]
Training Epoch: 16/100, batch 529/563 completed (loss: 0.2083:  94%|█████████▍| 265/281 [26:48<01:37,  6.11s/it]
Training Epoch: 16/100, batch 529/563 completed (loss: 0.1628:  94%|█████████▍| 265/281 [26:48<01:37,  6.11s/it]
Training Epoch: 16/100, batch 530/563 completed (loss: 0.0489:  94%|█████████▍| 265/281 [26:51<01:37,  6.11s/it]
Training Epoch: 16/100, batch 530/563 completed (loss: 0.0553:  94%|█████████▍| 265/281 [26:51<01:37,  6.11s/it]
Training Epoch: 16/100, batch 530/563 completed (loss: 0.1085:  94%|█████████▍| 265/281 [26:51<01:37,  6.11s/it]
Training Epoch: 16/100, batch 530/563 completed (loss: 0.0354:  94%|█████████▍| 265/281 [26:51<01:37,  6.11s/it]
Training Epoch: 16/100, batch 530/563 completed (loss: 0.0553:  95%|█████████▍| 266/281 [26:54<01:31,  6.08s/it]
Training Epoch: 16/100, batch 530/563 completed (loss: 0.1085:  95%|█████████▍| 266/281 [26:54<01:31,  6.07s/it]
Training Epoch: 16/100, batch 530/563 completed (loss: 0.0489:  95%|█████████▍| 266/281 [26:54<01:31,  6.08s/it]
Training Epoch: 16/100, batch 530/563 completed (loss: 0.0354:  95%|█████████▍| 266/281 [26:54<01:31,  6.08s/it]
Training Epoch: 16/100, batch 531/563 completed (loss: 0.009:  95%|█████████▍| 266/281 [26:54<01:31,  6.08s/it] 
Training Epoch: 16/100, batch 531/563 completed (loss: 0.1001:  95%|█████████▍| 266/281 [26:54<01:31,  6.07s/it]
Training Epoch: 16/100, batch 531/563 completed (loss: 0.0094:  95%|█████████▍| 266/281 [26:54<01:31,  6.08s/it]
Training Epoch: 16/100, batch 531/563 completed (loss: 0.1224:  95%|█████████▍| 266/281 [26:54<01:31,  6.08s/it]
Training Epoch: 16/100, batch 532/563 completed (loss: 0.1:  95%|█████████▍| 266/281 [26:57<01:31,  6.08s/it]   
Training Epoch: 16/100, batch 532/563 completed (loss: 0.0041:  95%|█████████▍| 266/281 [26:57<01:31,  6.08s/it]
Training Epoch: 16/100, batch 532/563 completed (loss: 0.3656:  95%|█████████▍| 266/281 [26:57<01:31,  6.07s/it]
Training Epoch: 16/100, batch 532/563 completed (loss: 0.0271:  95%|█████████▍| 266/281 [26:57<01:31,  6.08s/it]
Training Epoch: 16/100, batch 532/563 completed (loss: 0.3656:  95%|█████████▌| 267/281 [27:00<01:24,  6.05s/it]
Training Epoch: 16/100, batch 532/563 completed (loss: 0.0041:  95%|█████████▌| 267/281 [27:00<01:24,  6.06s/it]
Training Epoch: 16/100, batch 532/563 completed (loss: 0.1:  95%|█████████▌| 267/281 [27:00<01:24,  6.06s/it]
Training Epoch: 16/100, batch 532/563 completed (loss: 0.0271:  95%|█████████▌| 267/281 [27:00<01:24,  6.06s/it]
Training Epoch: 16/100, batch 533/563 completed (loss: 0.0104:  95%|█████████▌| 267/281 [27:00<01:24,  6.05s/it]
Training Epoch: 16/100, batch 533/563 completed (loss: 0.1339:  95%|█████████▌| 267/281 [27:00<01:24,  6.06s/it]
Training Epoch: 16/100, batch 533/563 completed (loss: 0.0587:  95%|█████████▌| 267/281 [27:00<01:24,  6.06s/it]
Training Epoch: 16/100, batch 533/563 completed (loss: 0.0091:  95%|█████████▌| 267/281 [27:00<01:24,  6.06s/it]
Training Epoch: 16/100, batch 534/563 completed (loss: 0.0288:  95%|█████████▌| 267/281 [27:04<01:24,  6.06s/it]
Training Epoch: 16/100, batch 534/563 completed (loss: 0.1058:  95%|█████████▌| 267/281 [27:04<01:24,  6.06s/it]
Training Epoch: 16/100, batch 534/563 completed (loss: 0.0134:  95%|█████████▌| 267/281 [27:04<01:24,  6.05s/it]
Training Epoch: 16/100, batch 534/563 completed (loss: 0.056:  95%|█████████▌| 267/281 [27:04<01:24,  6.06s/it] 
Training Epoch: 16/100, batch 534/563 completed (loss: 0.1058:  95%|█████████▌| 268/281 [27:07<01:19,  6.09s/it]
Training Epoch: 16/100, batch 534/563 completed (loss: 0.0288:  95%|█████████▌| 268/281 [27:06<01:19,  6.09s/it]
Training Epoch: 16/100, batch 534/563 completed (loss: 0.0134:  95%|█████████▌| 268/281 [27:06<01:19,  6.09s/it]
Training Epoch: 16/100, batch 534/563 completed (loss: 0.056:  95%|█████████▌| 268/281 [27:07<01:19,  6.09s/it]
Training Epoch: 16/100, batch 535/563 completed (loss: 0.032:  95%|█████████▌| 268/281 [27:07<01:19,  6.09s/it] 
Training Epoch: 16/100, batch 535/563 completed (loss: 0.0129:  95%|█████████▌| 268/281 [27:07<01:19,  6.09s/it]
Training Epoch: 16/100, batch 535/563 completed (loss: 0.0325:  95%|█████████▌| 268/281 [27:07<01:19,  6.09s/it]
Training Epoch: 16/100, batch 535/563 completed (loss: 0.0525:  95%|█████████▌| 268/281 [27:07<01:19,  6.09s/it]
Training Epoch: 16/100, batch 536/563 completed (loss: 0.2923:  95%|█████████▌| 268/281 [27:10<01:19,  6.09s/it]
Training Epoch: 16/100, batch 536/563 completed (loss: 0.0604:  95%|█████████▌| 268/281 [27:10<01:19,  6.09s/it]
Training Epoch: 16/100, batch 536/563 completed (loss: 0.016:  95%|█████████▌| 268/281 [27:10<01:19,  6.09s/it] 
Training Epoch: 16/100, batch 536/563 completed (loss: 0.0615:  95%|█████████▌| 268/281 [27:10<01:19,  6.09s/it]
Training Epoch: 16/100, batch 536/563 completed (loss: 0.2923:  96%|█████████▌| 269/281 [27:13<01:13,  6.11s/it]
Training Epoch: 16/100, batch 536/563 completed (loss: 0.0604:  96%|█████████▌| 269/281 [27:13<01:13,  6.11s/it]
Training Epoch: 16/100, batch 536/563 completed (loss: 0.016:  96%|█████████▌| 269/281 [27:13<01:13,  6.11s/it]
Training Epoch: 16/100, batch 536/563 completed (loss: 0.0615:  96%|█████████▌| 269/281 [27:13<01:13,  6.10s/it]
Training Epoch: 16/100, batch 537/563 completed (loss: 0.1439:  96%|█████████▌| 269/281 [27:13<01:13,  6.11s/it]
Training Epoch: 16/100, batch 537/563 completed (loss: 0.0674:  96%|█████████▌| 269/281 [27:13<01:13,  6.11s/it]
Training Epoch: 16/100, batch 537/563 completed (loss: 0.016:  96%|█████████▌| 269/281 [27:13<01:13,  6.11s/it]
Training Epoch: 16/100, batch 537/563 completed (loss: 0.0847:  96%|█████████▌| 269/281 [27:13<01:13,  6.10s/it]
Training Epoch: 16/100, batch 538/563 completed (loss: 0.3426:  96%|█████████▌| 269/281 [27:16<01:13,  6.11s/it]
Training Epoch: 16/100, batch 538/563 completed (loss: 0.0369:  96%|█████████▌| 269/281 [27:16<01:13,  6.11s/it]
Training Epoch: 16/100, batch 538/563 completed (loss: 0.0147:  96%|█████████▌| 269/281 [27:16<01:13,  6.11s/it]
Training Epoch: 16/100, batch 538/563 completed (loss: 0.0072:  96%|█████████▌| 269/281 [27:16<01:13,  6.10s/it]
Training Epoch: 16/100, batch 538/563 completed (loss: 0.0147:  96%|█████████▌| 270/281 [27:19<01:06,  6.07s/it]
Training Epoch: 16/100, batch 538/563 completed (loss: 0.0369:  96%|█████████▌| 270/281 [27:19<01:06,  6.08s/it]
Training Epoch: 16/100, batch 538/563 completed (loss: 0.3426:  96%|█████████▌| 270/281 [27:19<01:06,  6.08s/it]
Training Epoch: 16/100, batch 538/563 completed (loss: 0.0072:  96%|█████████▌| 270/281 [27:19<01:06,  6.08s/it]
Training Epoch: 16/100, batch 539/563 completed (loss: 0.0747:  96%|█████████▌| 270/281 [27:19<01:06,  6.08s/it]
Training Epoch: 16/100, batch 539/563 completed (loss: 0.1326:  96%|█████████▌| 270/281 [27:19<01:06,  6.08s/it]
Training Epoch: 16/100, batch 539/563 completed (loss: 0.4052:  96%|█████████▌| 270/281 [27:19<01:06,  6.07s/it]
Training Epoch: 16/100, batch 539/563 completed (loss: 0.6902:  96%|█████████▌| 270/281 [27:19<01:06,  6.08s/it]
Training Epoch: 16/100, batch 540/563 completed (loss: 0.0282:  96%|█████████▌| 270/281 [27:22<01:06,  6.08s/it]
Training Epoch: 16/100, batch 540/563 completed (loss: 0.0626:  96%|█████████▌| 270/281 [27:22<01:06,  6.08s/it]
Training Epoch: 16/100, batch 540/563 completed (loss: 0.024:  96%|█████████▌| 270/281 [27:22<01:06,  6.07s/it] 
Training Epoch: 16/100, batch 540/563 completed (loss: 0.0434:  96%|█████████▌| 270/281 [27:22<01:06,  6.08s/it]
Training Epoch: 16/100, batch 540/563 completed (loss: 0.0434:  96%|█████████▋| 271/281 [27:25<01:01,  6.11s/it]
Training Epoch: 16/100, batch 540/563 completed (loss: 0.0626:  96%|█████████▋| 271/281 [27:25<01:01,  6.11s/it]
Training Epoch: 16/100, batch 540/563 completed (loss: 0.0282:  96%|█████████▋| 271/281 [27:25<01:01,  6.11s/it]
Training Epoch: 16/100, batch 540/563 completed (loss: 0.024:  96%|█████████▋| 271/281 [27:25<01:01,  6.12s/it]
Training Epoch: 16/100, batch 541/563 completed (loss: 0.0102:  96%|█████████▋| 271/281 [27:25<01:01,  6.11s/it]
Training Epoch: 16/100, batch 541/563 completed (loss: 0.0373:  96%|█████████▋| 271/281 [27:25<01:01,  6.11s/it]
Training Epoch: 16/100, batch 541/563 completed (loss: 0.108:  96%|█████████▋| 271/281 [27:25<01:01,  6.12s/it]
Training Epoch: 16/100, batch 541/563 completed (loss: 0.0281:  96%|█████████▋| 271/281 [27:25<01:01,  6.11s/it]
Training Epoch: 16/100, batch 542/563 completed (loss: 0.0294:  96%|█████████▋| 271/281 [27:28<01:01,  6.11s/it]
Training Epoch: 16/100, batch 542/563 completed (loss: 0.0315:  96%|█████████▋| 271/281 [27:28<01:01,  6.12s/it]
Training Epoch: 16/100, batch 542/563 completed (loss: 0.0459:  96%|█████████▋| 271/281 [27:28<01:01,  6.11s/it]
Training Epoch: 16/100, batch 542/563 completed (loss: 0.0965:  96%|█████████▋| 271/281 [27:28<01:01,  6.11s/it]
Training Epoch: 16/100, batch 542/563 completed (loss: 0.0459:  97%|█████████▋| 272/281 [27:31<00:54,  6.08s/it]
Training Epoch: 16/100, batch 542/563 completed (loss: 0.0294:  97%|█████████▋| 272/281 [27:31<00:54,  6.08s/it]
Training Epoch: 16/100, batch 542/563 completed (loss: 0.0315:  97%|█████████▋| 272/281 [27:31<00:54,  6.09s/it]
Training Epoch: 16/100, batch 542/563 completed (loss: 0.0965:  97%|█████████▋| 272/281 [27:31<00:54,  6.08s/it]
Training Epoch: 16/100, batch 543/563 completed (loss: 0.1529:  97%|█████████▋| 272/281 [27:31<00:54,  6.09s/it]
Training Epoch: 16/100, batch 543/563 completed (loss: 0.2222:  97%|█████████▋| 272/281 [27:31<00:54,  6.08s/it]
Training Epoch: 16/100, batch 543/563 completed (loss: 0.1081:  97%|█████████▋| 272/281 [27:31<00:54,  6.08s/it]
Training Epoch: 16/100, batch 543/563 completed (loss: 0.0889:  97%|█████████▋| 272/281 [27:31<00:54,  6.08s/it]
Training Epoch: 16/100, batch 544/563 completed (loss: 0.1174:  97%|█████████▋| 272/281 [27:34<00:54,  6.08s/it]
Training Epoch: 16/100, batch 544/563 completed (loss: 0.0709:  97%|█████████▋| 272/281 [27:34<00:54,  6.09s/it]
Training Epoch: 16/100, batch 544/563 completed (loss: 0.0105:  97%|█████████▋| 272/281 [27:34<00:54,  6.08s/it]
Training Epoch: 16/100, batch 544/563 completed (loss: 0.0074:  97%|█████████▋| 272/281 [27:34<00:54,  6.08s/it]
Training Epoch: 16/100, batch 544/563 completed (loss: 0.1174:  97%|█████████▋| 273/281 [27:37<00:48,  6.11s/it]
Training Epoch: 16/100, batch 544/563 completed (loss: 0.0709:  97%|█████████▋| 273/281 [27:37<00:48,  6.11s/it]
Training Epoch: 16/100, batch 544/563 completed (loss: 0.0105:  97%|█████████▋| 273/281 [27:37<00:48,  6.11s/it]
Training Epoch: 16/100, batch 544/563 completed (loss: 0.0074:  97%|█████████▋| 273/281 [27:37<00:48,  6.11s/it]
Training Epoch: 16/100, batch 545/563 completed (loss: 0.2203:  97%|█████████▋| 273/281 [27:37<00:48,  6.11s/it]
Training Epoch: 16/100, batch 545/563 completed (loss: 0.0975:  97%|█████████▋| 273/281 [27:37<00:48,  6.11s/it]
Training Epoch: 16/100, batch 545/563 completed (loss: 0.039:  97%|█████████▋| 273/281 [27:37<00:48,  6.11s/it] 
Training Epoch: 16/100, batch 545/563 completed (loss: 0.0921:  97%|█████████▋| 273/281 [27:37<00:48,  6.11s/it]
Training Epoch: 16/100, batch 546/563 completed (loss: 0.0339:  97%|█████████▋| 273/281 [27:40<00:48,  6.11s/it]
Training Epoch: 16/100, batch 546/563 completed (loss: 0.0523:  97%|█████████▋| 273/281 [27:40<00:48,  6.11s/it]
Training Epoch: 16/100, batch 546/563 completed (loss: 0.0884:  97%|█████████▋| 273/281 [27:40<00:48,  6.11s/it]
Training Epoch: 16/100, batch 546/563 completed (loss: 0.0064:  97%|█████████▋| 273/281 [27:40<00:48,  6.11s/it]
Training Epoch: 16/100, batch 546/563 completed (loss: 0.0884:  98%|█████████▊| 274/281 [27:43<00:42,  6.08s/it]
Training Epoch: 16/100, batch 546/563 completed (loss: 0.0339:  98%|█████████▊| 274/281 [27:43<00:42,  6.08s/it]
Training Epoch: 16/100, batch 546/563 completed (loss: 0.0523:  98%|█████████▊| 274/281 [27:43<00:42,  6.08s/it]
Training Epoch: 16/100, batch 546/563 completed (loss: 0.0064:  98%|█████████▊| 274/281 [27:43<00:42,  6.08s/it]
Training Epoch: 16/100, batch 547/563 completed (loss: 0.1428:  98%|█████████▊| 274/281 [27:43<00:42,  6.08s/it]
Training Epoch: 16/100, batch 547/563 completed (loss: 0.006:  98%|█████████▊| 274/281 [27:43<00:42,  6.08s/it] 
Training Epoch: 16/100, batch 547/563 completed (loss: 0.095:  98%|█████████▊| 274/281 [27:43<00:42,  6.08s/it] 
Training Epoch: 16/100, batch 547/563 completed (loss: 0.1052:  98%|█████████▊| 274/281 [27:43<00:42,  6.08s/it]
Training Epoch: 16/100, batch 548/563 completed (loss: 0.0206:  98%|█████████▊| 274/281 [27:46<00:42,  6.08s/it]
Training Epoch: 16/100, batch 548/563 completed (loss: 0.4561:  98%|█████████▊| 274/281 [27:46<00:42,  6.08s/it]
Training Epoch: 16/100, batch 548/563 completed (loss: 0.1962:  98%|█████████▊| 274/281 [27:46<00:42,  6.08s/it]
Training Epoch: 16/100, batch 548/563 completed (loss: 0.0098:  98%|█████████▊| 274/281 [27:46<00:42,  6.08s/it]
Training Epoch: 16/100, batch 548/563 completed (loss: 0.1962:  98%|█████████▊| 275/281 [27:49<00:36,  6.06s/it]
Training Epoch: 16/100, batch 548/563 completed (loss: 0.0206:  98%|█████████▊| 275/281 [27:49<00:36,  6.06s/it]
Training Epoch: 16/100, batch 548/563 completed (loss: 0.0098:  98%|█████████▊| 275/281 [27:49<00:36,  6.06s/it]
Training Epoch: 16/100, batch 548/563 completed (loss: 0.4561:  98%|█████████▊| 275/281 [27:49<00:36,  6.06s/it]
Training Epoch: 16/100, batch 549/563 completed (loss: 0.0662:  98%|█████████▊| 275/281 [27:49<00:36,  6.06s/it]
Training Epoch: 16/100, batch 549/563 completed (loss: 0.1344:  98%|█████████▊| 275/281 [27:49<00:36,  6.06s/it]
Training Epoch: 16/100, batch 549/563 completed (loss: 0.0168:  98%|█████████▊| 275/281 [27:49<00:36,  6.06s/it]
Training Epoch: 16/100, batch 549/563 completed (loss: 0.4061:  98%|█████████▊| 275/281 [27:49<00:36,  6.06s/it]
Training Epoch: 16/100, batch 550/563 completed (loss: 0.0348:  98%|█████████▊| 275/281 [27:52<00:36,  6.06s/it]
Training Epoch: 16/100, batch 550/563 completed (loss: 0.0771:  98%|█████████▊| 275/281 [27:52<00:36,  6.06s/it]
Training Epoch: 16/100, batch 550/563 completed (loss: 0.0075:  98%|█████████▊| 275/281 [27:52<00:36,  6.06s/it]
Training Epoch: 16/100, batch 550/563 completed (loss: 0.012:  98%|█████████▊| 275/281 [27:52<00:36,  6.06s/it] 
Training Epoch: 16/100, batch 550/563 completed (loss: 0.0348:  98%|█████████▊| 276/281 [27:55<00:30,  6.15s/it]
Training Epoch: 16/100, batch 550/563 completed (loss: 0.0771:  98%|█████████▊| 276/281 [27:55<00:30,  6.15s/it]
Training Epoch: 16/100, batch 550/563 completed (loss: 0.0075:  98%|█████████▊| 276/281 [27:55<00:30,  6.15s/it]
Training Epoch: 16/100, batch 550/563 completed (loss: 0.012:  98%|█████████▊| 276/281 [27:55<00:30,  6.14s/it]
Training Epoch: 16/100, batch 551/563 completed (loss: 0.1285:  98%|█████████▊| 276/281 [27:55<00:30,  6.15s/it]
Training Epoch: 16/100, batch 551/563 completed (loss: 0.0192:  98%|█████████▊| 276/281 [27:55<00:30,  6.15s/it]
Training Epoch: 16/100, batch 551/563 completed (loss: 0.1794:  98%|█████████▊| 276/281 [27:55<00:30,  6.15s/it]
Training Epoch: 16/100, batch 551/563 completed (loss: 0.0369:  98%|█████████▊| 276/281 [27:56<00:30,  6.14s/it]
Training Epoch: 16/100, batch 552/563 completed (loss: 0.0433:  98%|█████████▊| 276/281 [27:58<00:30,  6.15s/it]
Training Epoch: 16/100, batch 552/563 completed (loss: 0.0996:  98%|█████████▊| 276/281 [27:58<00:30,  6.15s/it]
Training Epoch: 16/100, batch 552/563 completed (loss: 0.2336:  98%|█████████▊| 276/281 [27:58<00:30,  6.15s/it]
Training Epoch: 16/100, batch 552/563 completed (loss: 0.0154:  98%|█████████▊| 276/281 [27:58<00:30,  6.14s/it]
Training Epoch: 16/100, batch 552/563 completed (loss: 0.0433:  99%|█████████▊| 277/281 [28:01<00:24,  6.10s/it]
Training Epoch: 16/100, batch 552/563 completed (loss: 0.2336:  99%|█████████▊| 277/281 [28:01<00:24,  6.10s/it]
Training Epoch: 16/100, batch 552/563 completed (loss: 0.0996:  99%|█████████▊| 277/281 [28:01<00:24,  6.10s/it]
Training Epoch: 16/100, batch 552/563 completed (loss: 0.0154:  99%|█████████▊| 277/281 [28:01<00:24,  6.10s/it]
Training Epoch: 16/100, batch 553/563 completed (loss: 0.0244:  99%|█████████▊| 277/281 [28:01<00:24,  6.10s/it]
Training Epoch: 16/100, batch 553/563 completed (loss: 0.0136:  99%|█████████▊| 277/281 [28:01<00:24,  6.10s/it]
Training Epoch: 16/100, batch 553/563 completed (loss: 0.0924:  99%|█████████▊| 277/281 [28:01<00:24,  6.10s/it]
Training Epoch: 16/100, batch 553/563 completed (loss: 0.1304:  99%|█████████▊| 277/281 [28:02<00:24,  6.10s/it]
Training Epoch: 16/100, batch 554/563 completed (loss: 0.0366:  99%|█████████▊| 277/281 [28:04<00:24,  6.10s/it]
Training Epoch: 16/100, batch 554/563 completed (loss: 0.0241:  99%|█████████▊| 277/281 [28:04<00:24,  6.10s/it]
Training Epoch: 16/100, batch 554/563 completed (loss: 0.0135:  99%|█████████▊| 277/281 [28:04<00:24,  6.10s/it]
Training Epoch: 16/100, batch 554/563 completed (loss: 0.0118:  99%|█████████▊| 277/281 [28:04<00:24,  6.10s/it]
Training Epoch: 16/100, batch 554/563 completed (loss: 0.0241:  99%|█████████▉| 278/281 [28:07<00:18,  6.07s/it]
Training Epoch: 16/100, batch 554/563 completed (loss: 0.0366:  99%|█████████▉| 278/281 [28:07<00:18,  6.07s/it]
Training Epoch: 16/100, batch 554/563 completed (loss: 0.0135:  99%|█████████▉| 278/281 [28:07<00:18,  6.07s/it]
Training Epoch: 16/100, batch 554/563 completed (loss: 0.0118:  99%|█████████▉| 278/281 [28:07<00:18,  6.07s/it]
Training Epoch: 16/100, batch 555/563 completed (loss: 0.0716:  99%|█████████▉| 278/281 [28:07<00:18,  6.07s/it]
Training Epoch: 16/100, batch 555/563 completed (loss: 0.2061:  99%|█████████▉| 278/281 [28:07<00:18,  6.07s/it]
Training Epoch: 16/100, batch 555/563 completed (loss: 0.0658:  99%|█████████▉| 278/281 [28:07<00:18,  6.07s/it]
Training Epoch: 16/100, batch 555/563 completed (loss: 0.0262:  99%|█████████▉| 278/281 [28:08<00:18,  6.07s/it]
Training Epoch: 16/100, batch 556/563 completed (loss: 0.019:  99%|█████████▉| 278/281 [28:11<00:18,  6.07s/it] 
Training Epoch: 16/100, batch 556/563 completed (loss: 0.2581:  99%|█████████▉| 278/281 [28:11<00:18,  6.07s/it]
Training Epoch: 16/100, batch 556/563 completed (loss: 0.3619:  99%|█████████▉| 278/281 [28:11<00:18,  6.07s/it]
Training Epoch: 16/100, batch 556/563 completed (loss: 0.0834:  99%|█████████▉| 278/281 [28:11<00:18,  6.07s/it]
Training Epoch: 16/100, batch 556/563 completed (loss: 0.2581:  99%|█████████▉| 279/281 [28:14<00:12,  6.11s/it]
Training Epoch: 16/100, batch 556/563 completed (loss: 0.3619:  99%|█████████▉| 279/281 [28:14<00:12,  6.11s/it]
Training Epoch: 16/100, batch 556/563 completed (loss: 0.019:  99%|█████████▉| 279/281 [28:14<00:12,  6.11s/it]
Training Epoch: 16/100, batch 556/563 completed (loss: 0.0834:  99%|█████████▉| 279/281 [28:14<00:12,  6.11s/it]
Training Epoch: 16/100, batch 557/563 completed (loss: 0.0499:  99%|█████████▉| 279/281 [28:14<00:12,  6.11s/it]
Training Epoch: 16/100, batch 557/563 completed (loss: 0.0189:  99%|█████████▉| 279/281 [28:14<00:12,  6.11s/it]
Training Epoch: 16/100, batch 557/563 completed (loss: 0.0426:  99%|█████████▉| 279/281 [28:14<00:12,  6.11s/it]
Training Epoch: 16/100, batch 557/563 completed (loss: 0.1819:  99%|█████████▉| 279/281 [28:14<00:12,  6.11s/it]
Training Epoch: 16/100, batch 558/563 completed (loss: 0.0632:  99%|█████████▉| 279/281 [28:17<00:12,  6.11s/it]
Training Epoch: 16/100, batch 558/563 completed (loss: 0.0065:  99%|█████████▉| 279/281 [28:17<00:12,  6.11s/it]
Training Epoch: 16/100, batch 558/563 completed (loss: 0.1018:  99%|█████████▉| 279/281 [28:17<00:12,  6.11s/it]
Training Epoch: 16/100, batch 558/563 completed (loss: 0.0189:  99%|█████████▉| 279/281 [28:17<00:12,  6.11s/it]
Training Epoch: 16/100, batch 558/563 completed (loss: 0.0189: 100%|█████████▉| 280/281 [28:20<00:06,  6.07s/it]
Training Epoch: 16/100, batch 558/563 completed (loss: 0.0632: 100%|█████████▉| 280/281 [28:20<00:06,  6.07s/it]
Training Epoch: 16/100, batch 558/563 completed (loss: 0.0065: 100%|█████████▉| 280/281 [28:20<00:06,  6.08s/it]
Training Epoch: 16/100, batch 558/563 completed (loss: 0.1018: 100%|█████████▉| 280/281 [28:20<00:06,  6.08s/it]
Training Epoch: 16/100, batch 559/563 completed (loss: 0.0591: 100%|█████████▉| 280/281 [28:20<00:06,  6.07s/it]
Training Epoch: 16/100, batch 559/563 completed (loss: 0.1855: 100%|█████████▉| 280/281 [28:20<00:06,  6.08s/it]
Training Epoch: 16/100, batch 559/563 completed (loss: 0.0447: 100%|█████████▉| 280/281 [28:20<00:06,  6.08s/it]
Training Epoch: 16/100, batch 559/563 completed (loss: 0.0094: 100%|█████████▉| 280/281 [28:20<00:06,  6.07s/it]
Training Epoch: 16/100, batch 560/563 completed (loss: 0.0753: 100%|█████████▉| 280/281 [28:23<00:06,  6.08s/it]
Training Epoch: 16/100, batch 560/563 completed (loss: 0.1062: 100%|█████████▉| 280/281 [28:23<00:06,  6.07s/it]
Training Epoch: 16/100, batch 560/563 completed (loss: 0.0294: 100%|█████████▉| 280/281 [28:23<00:06,  6.08s/it]
Training Epoch: 16/100, batch 560/563 completed (loss: 0.0292: 100%|█████████▉| 280/281 [28:23<00:06,  6.07s/it]
Training Epoch: 16/100, batch 560/563 completed (loss: 0.0753: 100%|██████████| 281/281 [28:26<00:00,  6.15s/it]
Training Epoch: 16/100, batch 560/563 completed (loss: 0.0294: 100%|██████████| 281/281 [28:26<00:00,  6.15s/it]
Training Epoch: 16/100, batch 560/563 completed (loss: 0.1062: 100%|██████████| 281/281 [28:26<00:00,  6.15s/it]
Training Epoch: 16/100, batch 560/563 completed (loss: 0.0292: 100%|██████████| 281/281 [28:26<00:00,  6.15s/it]
Training Epoch: 16/100, batch 561/563 completed (loss: 0.0284: 100%|██████████| 281/281 [28:26<00:00,  6.15s/it]
Training Epoch: 16/100, batch 561/563 completed (loss: 0.0087: 100%|██████████| 281/281 [28:26<00:00,  6.15s/it]
Training Epoch: 16/100, batch 561/563 completed (loss: 0.1054: 100%|██████████| 281/281 [28:26<00:00,  6.15s/it]
Training Epoch: 16/100, batch 561/563 completed (loss: 0.0482: 100%|██████████| 281/281 [28:26<00:00,  6.15s/it]
Training Epoch: 16/100, batch 561/563 completed (loss: 0.0284: : 282it [28:28,  4.92s/it]                       
Training Epoch: 16/100, batch 561/563 completed (loss: 0.0087: : 282it [28:28,  4.92s/it]                       
Training Epoch: 16/100, batch 561/563 completed (loss: 0.1054: : 282it [28:28,  4.92s/it]                       
Training Epoch: 16/100, batch 561/563 completed (loss: 0.0482: : 282it [28:28,  4.92s/it]                       
Training Epoch: 16/100, batch 562/563 completed (loss: 0.0073: : 282it [28:28,  4.92s/it]
Training Epoch: 16/100, batch 562/563 completed (loss: 0.0155: : 282it [28:28,  4.92s/it]
Training Epoch: 16/100, batch 562/563 completed (loss: 0.006: : 282it [28:28,  4.92s/it] 
Training Epoch: 16/100, batch 562/563 completed (loss: 0.007: : 282it [28:28,  4.92s/it] 
Training Epoch: 16/100, batch 562/563 completed (loss: 0.006: : 282it [28:28,  6.06s/it]

Training Epoch: 16/100, batch 562/563 completed (loss: 0.0073: : 282it [28:28,  6.06s/it]

Training Epoch: 16/100, batch 562/563 completed (loss: 0.0155: : 282it [28:28,  6.06s/it]

Training Epoch: 16/100, batch 562/563 completed (loss: 0.007: : 282it [28:29,  6.06s/it]
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/c10d_logger.py:83: UserWarning: barrier(): using the device under current context. You can specify `device_id` in `init_process_group` to mute this warning.
  return func(*args, **kwargs)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
saving model + train state.

Processing Files (0 / 0)      : |          |  0.00B /  0.00B            

New Data Upload               : |          |  0.00B /  0.00B            


  ...ve-03b89ee9/checkpoint_16:   1%|          | 32.5MB / 3.63GB            


  ...ve-03b89ee9/checkpoint_16:   1%|          | 32.5MB / 3.63GB            
Processing Files (0 / 1)      :   1%|          | 32.5MB / 3.63GB,   ???B/s  

New Data Upload               :  24%|██▍       | 32.5MB /  134MB,   ???B/s  


  ...ve-03b89ee9/checkpoint_16:   3%|▎         |  119MB / 3.63GB            
Processing Files (0 / 1)      :   3%|▎         |  119MB / 3.63GB, 11.6MB/s  

New Data Upload               :  59%|█████▉    |  119MB /  201MB, 11.6MB/s  


  ...ve-03b89ee9/checkpoint_16:   4%|▎         |  133MB / 3.63GB            
Processing Files (0 / 1)      :   4%|▎         |  133MB / 3.63GB, 12.7MB/s  

New Data Upload               :  66%|██████▋   |  133MB /  201MB, 12.7MB/s  


  ...ve-03b89ee9/checkpoint_16:   4%|▍         |  153MB / 3.63GB            
Processing Files (0 / 1)      :   4%|▍         |  153MB / 3.63GB, 13.9MB/s  

New Data Upload               :  57%|█████▋    |  153MB /  268MB, 13.9MB/s  


  ...ve-03b89ee9/checkpoint_16:   6%|▌         |  214MB / 3.63GB            
Processing Files (0 / 1)      :   6%|▌         |  214MB / 3.63GB, 19.4MB/s  

New Data Upload               :  64%|██████▎   |  214MB /  335MB, 19.4MB/s  


  ...ve-03b89ee9/checkpoint_16:   9%|▉         |  326MB / 3.63GB            
Processing Files (0 / 1)      :   9%|▉         |  326MB / 3.63GB, 29.6MB/s  

New Data Upload               :  81%|████████  |  326MB /  402MB, 29.6MB/s  


  ...ve-03b89ee9/checkpoint_16:  10%|▉         |  356MB / 3.63GB            
Processing Files (0 / 1)      :  10%|▉         |  356MB / 3.63GB, 31.7MB/s  

New Data Upload               :  76%|███████▌  |  356MB /  469MB, 31.7MB/s  


  ...ve-03b89ee9/checkpoint_16:  11%|█         |  401MB / 3.63GB            
Processing Files (0 / 1)      :  11%|█         |  401MB / 3.63GB, 35.1MB/s  

New Data Upload               :  85%|████████▌ |  401MB /  469MB, 35.1MB/s  


  ...ve-03b89ee9/checkpoint_16:  12%|█▏        |  439MB / 3.63GB            
Processing Files (0 / 1)      :  12%|█▏        |  439MB / 3.63GB, 37.8MB/s  

New Data Upload               :  82%|████████▏ |  439MB /  536MB, 37.8MB/s  


  ...ve-03b89ee9/checkpoint_16:  14%|█▍        |  520MB / 3.63GB            
Processing Files (0 / 1)      :  14%|█▍        |  520MB / 3.63GB, 44.4MB/s  

New Data Upload               :  78%|███████▊  |  520MB /  671MB, 44.4MB/s  


  ...ve-03b89ee9/checkpoint_16:  17%|█▋        |  606MB / 3.63GB            
Processing Files (0 / 1)      :  17%|█▋        |  606MB / 3.63GB, 51.3MB/s  

New Data Upload               :  82%|████████▏ |  606MB /  738MB, 51.3MB/s  


  ...ve-03b89ee9/checkpoint_16:  19%|█▉        |  687MB / 3.63GB            
Processing Files (0 / 1)      :  19%|█▉        |  687MB / 3.63GB, 57.5MB/s  

New Data Upload               :  85%|████████▌ |  687MB /  805MB, 57.5MB/s  


  ...ve-03b89ee9/checkpoint_16:  21%|██▏       |  774MB / 3.63GB            
Processing Files (0 / 1)      :  21%|██▏       |  774MB / 3.63GB, 64.1MB/s  

New Data Upload               :  89%|████████▉ |  774MB /  872MB, 64.1MB/s  


  ...ve-03b89ee9/checkpoint_16:  24%|██▍       |  874MB / 3.63GB            
Processing Files (0 / 1)      :  24%|██▍       |  874MB / 3.63GB, 71.6MB/s  

New Data Upload               :  87%|████████▋ |  874MB / 1.01GB, 71.6MB/s  


  ...ve-03b89ee9/checkpoint_16:  26%|██▌       |  926MB / 3.63GB            
Processing Files (0 / 1)      :  26%|██▌       |  926MB / 3.63GB, 74.7MB/s  

New Data Upload               :  86%|████████▋ |  926MB / 1.07GB, 74.7MB/s  


  ...ve-03b89ee9/checkpoint_16:  29%|██▉       | 1.04GB / 3.63GB            
Processing Files (0 / 1)      :  29%|██▉       | 1.04GB / 3.63GB, 83.5MB/s  

New Data Upload               :  92%|█████████▏| 1.04GB / 1.14GB, 83.5MB/s  


  ...ve-03b89ee9/checkpoint_16:  31%|███       | 1.13GB / 3.63GB            
Processing Files (0 / 1)      :  31%|███       | 1.13GB / 3.63GB, 89.4MB/s  

New Data Upload               :  89%|████████▊ | 1.13GB / 1.27GB, 89.4MB/s  


  ...ve-03b89ee9/checkpoint_16:  33%|███▎      | 1.18GB / 3.63GB            
Processing Files (0 / 1)      :  33%|███▎      | 1.18GB / 3.63GB, 91.7MB/s  

New Data Upload               :  88%|████████▊ | 1.18GB / 1.34GB, 91.7MB/s  


  ...ve-03b89ee9/checkpoint_16:  36%|███▌      | 1.30GB / 3.63GB            
Processing Files (0 / 1)      :  36%|███▌      | 1.30GB / 3.63GB,  100MB/s  

New Data Upload               :  88%|████████▊ | 1.30GB / 1.48GB,  100MB/s  


  ...ve-03b89ee9/checkpoint_16:  38%|███▊      | 1.39GB / 3.63GB            
Processing Files (0 / 1)      :  38%|███▊      | 1.39GB / 3.63GB,  106MB/s  

New Data Upload               :  90%|████████▉ | 1.39GB / 1.54GB,  106MB/s  


  ...ve-03b89ee9/checkpoint_16:  41%|████▏     | 1.50GB / 3.63GB            
Processing Files (0 / 1)      :  41%|████▏     | 1.50GB / 3.63GB,  114MB/s  

New Data Upload               :  93%|█████████▎| 1.50GB / 1.61GB,  114MB/s  


  ...ve-03b89ee9/checkpoint_16:  44%|████▎     | 1.59GB / 3.63GB            
Processing Files (0 / 1)      :  44%|████▎     | 1.59GB / 3.63GB,  118MB/s  

New Data Upload               :  91%|█████████ | 1.59GB / 1.74GB,  118MB/s  


  ...ve-03b89ee9/checkpoint_16:  46%|████▌     | 1.67GB / 3.63GB            
Processing Files (0 / 1)      :  46%|████▌     | 1.67GB / 3.63GB,  123MB/s  

New Data Upload               :  92%|█████████▏| 1.67GB / 1.81GB,  123MB/s  


  ...ve-03b89ee9/checkpoint_16:  49%|████▉     | 1.79GB / 3.63GB            
Processing Files (0 / 1)      :  49%|████▉     | 1.79GB / 3.63GB,  130MB/s  

New Data Upload               :  92%|█████████▏| 1.79GB / 1.94GB,  130MB/s  


  ...ve-03b89ee9/checkpoint_16:  52%|█████▏    | 1.88GB / 3.63GB            
Processing Files (0 / 1)      :  52%|█████▏    | 1.88GB / 3.63GB,  135MB/s  

New Data Upload               :  93%|█████████▎| 1.88GB / 2.01GB,  135MB/s  


  ...ve-03b89ee9/checkpoint_16:  54%|█████▍    | 1.96GB / 3.63GB            
Processing Files (0 / 1)      :  54%|█████▍    | 1.96GB / 3.63GB,  139MB/s  

New Data Upload               :  94%|█████████▍| 1.96GB / 2.08GB,  139MB/s  


  ...ve-03b89ee9/checkpoint_16:  56%|█████▌    | 2.03GB / 3.63GB            
Processing Files (0 / 1)      :  56%|█████▌    | 2.03GB / 3.63GB,  142MB/s  

New Data Upload               :  92%|█████████▏| 2.03GB / 2.21GB,  142MB/s  


  ...ve-03b89ee9/checkpoint_16:  59%|█████▊    | 2.12GB / 3.63GB            
Processing Files (0 / 1)      :  59%|█████▊    | 2.12GB / 3.63GB,  147MB/s  

New Data Upload               :  93%|█████████▎| 2.12GB / 2.28GB,  147MB/s  


  ...ve-03b89ee9/checkpoint_16:  62%|██████▏   | 2.26GB / 3.63GB            
Processing Files (0 / 1)      :  62%|██████▏   | 2.26GB / 3.63GB,  156MB/s  

New Data Upload               :  96%|█████████▌| 2.25GB / 2.35GB,  155MB/s  


  ...ve-03b89ee9/checkpoint_16:  75%|███████▍  | 2.72GB / 3.63GB            
Processing Files (0 / 1)      :  75%|███████▍  | 2.72GB / 3.63GB,  191MB/s  

New Data Upload               :  99%|█████████▉| 2.33GB / 2.35GB,  158MB/s  


  ...ve-03b89ee9/checkpoint_16:  86%|████████▌ | 3.10GB / 3.63GB            
Processing Files (0 / 1)      :  86%|████████▌ | 3.10GB / 3.63GB,  219MB/s  

New Data Upload               : 100%|█████████▉| 2.35GB / 2.35GB,  156MB/s  


  ...ve-03b89ee9/checkpoint_16:  96%|█████████▌| 3.48GB / 3.63GB            
Processing Files (0 / 1)      :  96%|█████████▌| 3.48GB / 3.63GB,  246MB/s  

New Data Upload               : 100%|█████████▉| 2.35GB / 2.35GB,  153MB/s  


  ...ve-03b89ee9/checkpoint_16: 100%|█████████▉| 3.61GB / 3.63GB            
Processing Files (0 / 1)      : 100%|█████████▉| 3.61GB / 3.63GB,  251MB/s  

New Data Upload               :  99%|█████████▉| 2.37GB / 2.38GB,  152MB/s  


  ...ve-03b89ee9/checkpoint_16: 100%|█████████▉| 3.63GB / 3.63GB            
Processing Files (0 / 1)      : 100%|█████████▉| 3.63GB / 3.63GB,  247MB/s  

New Data Upload               : 100%|█████████▉| 2.38GB / 2.38GB,  150MB/s  


  ...ve-03b89ee9/checkpoint_16: 100%|█████████▉| 3.63GB / 3.63GB            
Processing Files (0 / 1)      : 100%|█████████▉| 3.63GB / 3.63GB,  237MB/s  

New Data Upload               : 100%|██████████| 2.38GB / 2.38GB,  144MB/s  
Processing Files (0 / 1)      : 100%|█████████▉| 3.63GB / 3.63GB,  237MB/s  

New Data Upload               : 100%|██████████| 2.38GB / 2.38GB,  144MB/s  

  ...ve-03b89ee9/checkpoint_16: 100%|█████████▉| 3.63GB / 3.63GB            
Uploaded checkpoint_16 to seyedparsa/thoughtformer-checkpoints/checkpoints/star-datacurr-only-4gpu-bs32-lr1e-5_n1000_qwen3-0.6b-base_thoughtformer_lr1e-5_no-reset_dc_no-tc_adaptive-03b89ee9/checkpoint_16

Processing Files (0 / 0)      : |          |  0.00B /  0.00B            

New Data Upload               : |          |  0.00B /  0.00B            


  ...e-03b89ee9/train_state_16:   0%|          | 3.69MB / 4.77GB            


  ...e-03b89ee9/train_state_16:   0%|          | 3.69MB / 4.77GB            
Processing Files (0 / 1)      :   0%|          | 3.69MB / 4.77GB,   ???B/s  

New Data Upload               :   3%|▎         | 3.69MB /  134MB,   ???B/s  


  ...e-03b89ee9/train_state_16:   1%|          | 27.9MB / 4.77GB            
Processing Files (0 / 1)      :   1%|          | 27.9MB / 4.77GB,   ???B/s  

New Data Upload               :  14%|█▍        | 27.9MB /  201MB,   ???B/s  


  ...e-03b89ee9/train_state_16:   1%|▏         | 70.5MB / 4.77GB            
Processing Files (0 / 1)      :   1%|▏         | 70.5MB / 4.77GB, 6.84MB/s  

New Data Upload               :  35%|███▌      | 70.5MB /  201MB, 6.84MB/s  


  ...e-03b89ee9/train_state_16:   2%|▏         | 72.1MB / 4.77GB            
Processing Files (0 / 1)      :   2%|▏         | 72.1MB / 4.77GB, 6.83MB/s  

New Data Upload               :  36%|███▌      | 72.1MB /  201MB, 6.83MB/s  


  ...e-03b89ee9/train_state_16:   2%|▏         | 73.7MB / 4.77GB            
Processing Files (0 / 1)      :   2%|▏         | 73.7MB / 4.77GB, 6.81MB/s  

New Data Upload               :  37%|███▋      | 73.7MB /  201MB, 6.81MB/s  


  ...e-03b89ee9/train_state_16:   2%|▏         | 80.2MB / 4.77GB            
Processing Files (0 / 1)      :   2%|▏         | 80.2MB / 4.77GB, 7.28MB/s  

New Data Upload               :  30%|██▉       | 80.2MB /  268MB, 7.28MB/s  


  ...e-03b89ee9/train_state_16:   3%|▎         |  160MB / 4.77GB            
Processing Files (0 / 1)      :   3%|▎         |  160MB / 4.77GB, 14.7MB/s  

New Data Upload               :  48%|████▊     |  160MB /  335MB, 14.7MB/s  


  ...e-03b89ee9/train_state_16:   4%|▍         |  209MB / 4.77GB            
Processing Files (0 / 1)      :   4%|▍         |  209MB / 4.77GB, 19.0MB/s  

New Data Upload               :  52%|█████▏    |  209MB /  402MB, 19.0MB/s  


  ...e-03b89ee9/train_state_16:   5%|▌         |  241MB / 4.77GB            
Processing Files (0 / 1)      :   5%|▌         |  241MB / 4.77GB, 21.6MB/s  

New Data Upload               :  60%|█████▉    |  241MB /  402MB, 21.6MB/s  


  ...e-03b89ee9/train_state_16:   6%|▌         |  283MB / 4.77GB            
Processing Files (0 / 1)      :   6%|▌         |  283MB / 4.77GB, 25.0MB/s  

New Data Upload               :  70%|███████   |  283MB /  402MB, 25.0MB/s  


  ...e-03b89ee9/train_state_16:   6%|▋         |  303MB / 4.77GB            
Processing Files (0 / 1)      :   6%|▋         |  303MB / 4.77GB, 26.2MB/s  

New Data Upload               :  65%|██████▍   |  303MB /  469MB, 26.2MB/s  


  ...e-03b89ee9/train_state_16:   8%|▊         |  398MB / 4.77GB            
Processing Files (0 / 1)      :   8%|▊         |  398MB / 4.77GB, 34.3MB/s  

New Data Upload               :  66%|██████▌   |  398MB /  603MB, 34.3MB/s  


  ...e-03b89ee9/train_state_16:  10%|█         |  490MB / 4.77GB            
Processing Files (0 / 1)      :  10%|█         |  490MB / 4.77GB, 42.0MB/s  

New Data Upload               :  73%|███████▎  |  490MB /  671MB, 42.0MB/s  


  ...e-03b89ee9/train_state_16:  12%|█▏        |  555MB / 4.77GB            
Processing Files (0 / 1)      :  12%|█▏        |  555MB / 4.77GB, 47.0MB/s  

New Data Upload               :  75%|███████▌  |  555MB /  738MB, 47.0MB/s  


  ...e-03b89ee9/train_state_16:  13%|█▎        |  632MB / 4.77GB            
Processing Files (0 / 1)      :  13%|█▎        |  632MB / 4.77GB, 52.9MB/s  

New Data Upload               :  86%|████████▌ |  632MB /  738MB, 52.9MB/s  


  ...e-03b89ee9/train_state_16:  15%|█▍        |  709MB / 4.77GB            
Processing Files (0 / 1)      :  15%|█▍        |  709MB / 4.77GB, 58.6MB/s  

New Data Upload               :  81%|████████▏ |  709MB /  872MB, 58.6MB/s  


  ...e-03b89ee9/train_state_16:  17%|█▋        |  795MB / 4.77GB            
Processing Files (0 / 1)      :  17%|█▋        |  795MB / 4.77GB, 65.0MB/s  

New Data Upload               :  85%|████████▍ |  795MB /  939MB, 65.0MB/s  


  ...e-03b89ee9/train_state_16:  19%|█▉        |  910MB / 4.77GB            
Processing Files (0 / 1)      :  19%|█▉        |  910MB / 4.77GB, 73.8MB/s  

New Data Upload               :  85%|████████▍ |  910MB / 1.07GB, 73.8MB/s  


  ...e-03b89ee9/train_state_16:  21%|██        | 1.01GB / 4.77GB            
Processing Files (0 / 1)      :  21%|██        | 1.01GB / 4.77GB, 80.6MB/s  

New Data Upload               :  88%|████████▊ | 1.01GB / 1.14GB, 80.6MB/s  


  ...e-03b89ee9/train_state_16:  23%|██▎       | 1.11GB / 4.77GB            
Processing Files (0 / 1)      :  23%|██▎       | 1.11GB / 4.77GB, 88.0MB/s  

New Data Upload               :  87%|████████▋ | 1.11GB / 1.27GB, 88.0MB/s  


  ...e-03b89ee9/train_state_16:  26%|██▌       | 1.24GB / 4.77GB            
Processing Files (0 / 1)      :  26%|██▌       | 1.24GB / 4.77GB, 97.4MB/s  

New Data Upload               :  88%|████████▊ | 1.24GB / 1.41GB, 97.4MB/s  


  ...e-03b89ee9/train_state_16:  28%|██▊       | 1.33GB / 4.77GB            
Processing Files (0 / 1)      :  28%|██▊       | 1.33GB / 4.77GB,  103MB/s  

New Data Upload               :  90%|████████▉ | 1.33GB / 1.48GB,  103MB/s  


  ...e-03b89ee9/train_state_16:  31%|███       | 1.46GB / 4.77GB            
Processing Files (0 / 1)      :  31%|███       | 1.46GB / 4.77GB,  112MB/s  

New Data Upload               :  95%|█████████▍| 1.46GB / 1.54GB,  112MB/s  


  ...e-03b89ee9/train_state_16:  32%|███▏      | 1.53GB / 4.77GB            
Processing Files (0 / 1)      :  32%|███▏      | 1.53GB / 4.77GB,  116MB/s  

New Data Upload               :  91%|█████████▏| 1.53GB / 1.68GB,  116MB/s  


  ...e-03b89ee9/train_state_16:  34%|███▍      | 1.62GB / 4.77GB            
Processing Files (0 / 1)      :  34%|███▍      | 1.62GB / 4.77GB,  122MB/s  

New Data Upload               :  90%|████████▉ | 1.62GB / 1.81GB,  122MB/s  


  ...e-03b89ee9/train_state_16:  36%|███▋      | 1.73GB / 4.77GB            
Processing Files (0 / 1)      :  36%|███▋      | 1.73GB / 4.77GB,  128MB/s  

New Data Upload               :  92%|█████████▏| 1.73GB / 1.88GB,  128MB/s  


  ...e-03b89ee9/train_state_16:  39%|███▉      | 1.85GB / 4.77GB            
Processing Files (0 / 1)      :  39%|███▉      | 1.85GB / 4.77GB,  136MB/s  

New Data Upload               :  95%|█████████▌| 1.85GB / 1.94GB,  136MB/s  


  ...e-03b89ee9/train_state_16:  40%|███▉      | 1.90GB / 4.77GB            
Processing Files (0 / 1)      :  40%|███▉      | 1.90GB / 4.77GB,  137MB/s  

New Data Upload               :  95%|█████████▍| 1.90GB / 2.01GB,  137MB/s  


  ...e-03b89ee9/train_state_16:  42%|████▏     | 2.00GB / 4.77GB            
Processing Files (0 / 1)      :  42%|████▏     | 2.00GB / 4.77GB,  143MB/s  

New Data Upload               :  93%|█████████▎| 2.00GB / 2.15GB,  143MB/s  


  ...e-03b89ee9/train_state_16:  43%|████▎     | 2.06GB / 4.77GB            
Processing Files (0 / 1)      :  43%|████▎     | 2.06GB / 4.77GB,  145MB/s  

New Data Upload               :  93%|█████████▎| 2.06GB / 2.21GB,  145MB/s  


  ...e-03b89ee9/train_state_16:  46%|████▌     | 2.19GB / 4.77GB            
Processing Files (0 / 1)      :  46%|████▌     | 2.19GB / 4.77GB,  152MB/s  

New Data Upload               :  93%|█████████▎| 2.19GB / 2.35GB,  152MB/s  


  ...e-03b89ee9/train_state_16:  47%|████▋     | 2.25GB / 4.77GB            
Processing Files (0 / 1)      :  47%|████▋     | 2.25GB / 4.77GB,  154MB/s  

New Data Upload               :  93%|█████████▎| 2.25GB / 2.41GB,  154MB/s  


  ...e-03b89ee9/train_state_16:  50%|████▉     | 2.38GB / 4.77GB            
Processing Files (0 / 1)      :  50%|████▉     | 2.38GB / 4.77GB,  162MB/s  

New Data Upload               :  96%|█████████▌| 2.38GB / 2.48GB,  162MB/s  


  ...e-03b89ee9/train_state_16:  51%|█████▏    | 2.44GB / 4.77GB            
Processing Files (0 / 1)      :  51%|█████▏    | 2.44GB / 4.77GB,  164MB/s  

New Data Upload               :  93%|█████████▎| 2.44GB / 2.62GB,  164MB/s  


  ...e-03b89ee9/train_state_16:  52%|█████▏    | 2.50GB / 4.77GB            
Processing Files (0 / 1)      :  52%|█████▏    | 2.50GB / 4.77GB,  165MB/s  

New Data Upload               :  93%|█████████▎| 2.50GB / 2.68GB,  165MB/s  


  ...e-03b89ee9/train_state_16:  54%|█████▍    | 2.59GB / 4.77GB            
Processing Files (0 / 1)      :  54%|█████▍    | 2.59GB / 4.77GB,  169MB/s  

New Data Upload               :  92%|█████████▏| 2.59GB / 2.82GB,  169MB/s  


  ...e-03b89ee9/train_state_16:  57%|█████▋    | 2.70GB / 4.77GB            
Processing Files (0 / 1)      :  57%|█████▋    | 2.70GB / 4.77GB,  175MB/s  

New Data Upload               :  94%|█████████▎| 2.70GB / 2.88GB,  175MB/s  


  ...e-03b89ee9/train_state_16:  59%|█████▊    | 2.80GB / 4.77GB            
Processing Files (0 / 1)      :  59%|█████▊    | 2.80GB / 4.77GB,  179MB/s  

New Data Upload               :  95%|█████████▍| 2.80GB / 2.95GB,  179MB/s  


  ...e-03b89ee9/train_state_16:  60%|█████▉    | 2.86GB / 4.77GB            
Processing Files (0 / 1)      :  60%|█████▉    | 2.86GB / 4.77GB,  181MB/s  

New Data Upload               :  95%|█████████▍| 2.86GB / 3.02GB,  181MB/s  


  ...e-03b89ee9/train_state_16:  63%|██████▎   | 2.99GB / 4.77GB            
Processing Files (0 / 1)      :  63%|██████▎   | 2.99GB / 4.77GB,  188MB/s  

New Data Upload               :  95%|█████████▍| 2.99GB / 3.15GB,  188MB/s  


  ...e-03b89ee9/train_state_16:  64%|██████▍   | 3.07GB / 4.77GB            
Processing Files (0 / 1)      :  64%|██████▍   | 3.07GB / 4.77GB,  190MB/s  

New Data Upload               :  95%|█████████▌| 3.07GB / 3.22GB,  190MB/s  


  ...e-03b89ee9/train_state_16:  67%|██████▋   | 3.18GB / 4.77GB            
Processing Files (0 / 1)      :  67%|██████▋   | 3.18GB / 4.77GB,  195MB/s  

New Data Upload               :  95%|█████████▍| 3.18GB / 3.35GB,  195MB/s  


  ...e-03b89ee9/train_state_16:  69%|██████▉   | 3.28GB / 4.77GB            
Processing Files (0 / 1)      :  69%|██████▉   | 3.28GB / 4.77GB,  199MB/s  

New Data Upload               :  96%|█████████▌| 3.28GB / 3.42GB,  199MB/s  


  ...e-03b89ee9/train_state_16:  70%|███████   | 3.34GB / 4.77GB            
Processing Files (0 / 1)      :  70%|███████   | 3.34GB / 4.77GB,  201MB/s  

New Data Upload               :  94%|█████████▍| 3.34GB / 3.55GB,  201MB/s  


  ...e-03b89ee9/train_state_16:  72%|███████▏  | 3.42GB / 4.77GB            
Processing Files (0 / 1)      :  72%|███████▏  | 3.42GB / 4.77GB,  202MB/s  

New Data Upload               :  94%|█████████▍| 3.42GB / 3.62GB,  202MB/s  


  ...e-03b89ee9/train_state_16:  73%|███████▎  | 3.50GB / 4.77GB            
Processing Files (0 / 1)      :  73%|███████▎  | 3.50GB / 4.77GB,  205MB/s  

New Data Upload               :  93%|█████████▎| 3.50GB / 3.76GB,  205MB/s  


  ...e-03b89ee9/train_state_16:  76%|███████▌  | 3.62GB / 4.77GB            
Processing Files (0 / 1)      :  76%|███████▌  | 3.62GB / 4.77GB,  211MB/s  

New Data Upload               :  95%|█████████▍| 3.62GB / 3.82GB,  211MB/s  


  ...e-03b89ee9/train_state_16:  78%|███████▊  | 3.74GB / 4.77GB            
Processing Files (0 / 1)      :  78%|███████▊  | 3.74GB / 4.77GB,  216MB/s  

New Data Upload               :  96%|█████████▌| 3.74GB / 3.89GB,  216MB/s  


  ...e-03b89ee9/train_state_16:  80%|███████▉  | 3.81GB / 4.77GB            
Processing Files (0 / 1)      :  80%|███████▉  | 3.81GB / 4.77GB,  217MB/s  

New Data Upload               :  95%|█████████▍| 3.81GB / 4.02GB,  217MB/s  


  ...e-03b89ee9/train_state_16:  81%|████████▏ | 3.88GB / 4.77GB            
Processing Files (0 / 1)      :  81%|████████▏ | 3.88GB / 4.77GB,  219MB/s  

New Data Upload               :  95%|█████████▍| 3.88GB / 4.09GB,  219MB/s  


  ...e-03b89ee9/train_state_16:  84%|████████▎ | 3.98GB / 4.77GB            
Processing Files (0 / 1)      :  84%|████████▎ | 3.98GB / 4.77GB,  222MB/s  

New Data Upload               :  96%|█████████▌| 3.98GB / 4.16GB,  222MB/s  


  ...e-03b89ee9/train_state_16:  85%|████████▌ | 4.06GB / 4.77GB            
Processing Files (0 / 1)      :  85%|████████▌ | 4.06GB / 4.77GB,  224MB/s  

New Data Upload               :  95%|█████████▍| 4.06GB / 4.29GB,  224MB/s  


  ...e-03b89ee9/train_state_16:  87%|████████▋ | 4.13GB / 4.77GB            
Processing Files (0 / 1)      :  87%|████████▋ | 4.13GB / 4.77GB,  225MB/s  

New Data Upload               :  95%|█████████▍| 4.13GB / 4.36GB,  225MB/s  


  ...e-03b89ee9/train_state_16:  89%|████████▉ | 4.26GB / 4.77GB            
Processing Files (0 / 1)      :  89%|████████▉ | 4.26GB / 4.77GB,  230MB/s  

New Data Upload               :  95%|█████████▍| 4.26GB / 4.49GB,  230MB/s  


  ...e-03b89ee9/train_state_16:  92%|█████████▏| 4.37GB / 4.77GB            
Processing Files (0 / 1)      :  92%|█████████▏| 4.37GB / 4.77GB,  235MB/s  

New Data Upload               :  96%|█████████▌| 4.37GB / 4.56GB,  235MB/s  


  ...e-03b89ee9/train_state_16:  94%|█████████▍| 4.48GB / 4.77GB            
Processing Files (0 / 1)      :  94%|█████████▍| 4.48GB / 4.77GB,  238MB/s  

New Data Upload               :  95%|█████████▌| 4.48GB / 4.69GB,  238MB/s  


  ...e-03b89ee9/train_state_16:  95%|█████████▌| 4.53GB / 4.77GB            
Processing Files (0 / 1)      :  95%|█████████▌| 4.53GB / 4.77GB,  238MB/s  

New Data Upload               :  95%|█████████▌| 4.53GB / 4.76GB,  238MB/s  


  ...e-03b89ee9/train_state_16:  97%|█████████▋| 4.64GB / 4.77GB            
Processing Files (0 / 1)      :  97%|█████████▋| 4.64GB / 4.77GB,  242MB/s  

New Data Upload               :  97%|█████████▋| 4.64GB / 4.77GB,  242MB/s  


  ...e-03b89ee9/train_state_16:  98%|█████████▊| 4.69GB / 4.77GB            
Processing Files (0 / 1)      :  98%|█████████▊| 4.69GB / 4.77GB,  241MB/s  

New Data Upload               :  98%|█████████▊| 4.69GB / 4.77GB,  241MB/s  


  ...e-03b89ee9/train_state_16:  99%|█████████▉| 4.71GB / 4.77GB            
Processing Files (0 / 1)      :  99%|█████████▉| 4.71GB / 4.77GB,  238MB/s  

New Data Upload               :  99%|█████████▉| 4.71GB / 4.77GB,  238MB/s  


  ...e-03b89ee9/train_state_16:  99%|█████████▉| 4.74GB / 4.77GB            
Processing Files (0 / 1)      :  99%|█████████▉| 4.74GB / 4.77GB,  236MB/s  

New Data Upload               :  99%|█████████▉| 4.74GB / 4.77GB,  236MB/s  


  ...e-03b89ee9/train_state_16: 100%|█████████▉| 4.76GB / 4.77GB            
Processing Files (0 / 1)      : 100%|█████████▉| 4.76GB / 4.77GB,  233MB/s  

New Data Upload               : 100%|█████████▉| 4.76GB / 4.77GB,  233MB/s  


  ...e-03b89ee9/train_state_16: 100%|█████████▉| 4.77GB / 4.77GB            
Processing Files (0 / 1)      : 100%|█████████▉| 4.77GB / 4.77GB,  229MB/s  

New Data Upload               : 100%|█████████▉| 4.77GB / 4.77GB,  229MB/s  


  ...e-03b89ee9/train_state_16: 100%|█████████▉| 4.77GB / 4.77GB            
Processing Files (0 / 1)      : 100%|█████████▉| 4.77GB / 4.77GB,  224MB/s  

New Data Upload               : 100%|█████████▉| 4.77GB / 4.77GB,  224MB/s  


  ...e-03b89ee9/train_state_16: 100%|█████████▉| 4.77GB / 4.77GB            
Processing Files (0 / 1)      : 100%|█████████▉| 4.77GB / 4.77GB,  220MB/s  

New Data Upload               : 100%|█████████▉| 4.77GB / 4.77GB,  220MB/s  


  ...e-03b89ee9/train_state_16: 100%|█████████▉| 4.77GB / 4.77GB            
Processing Files (0 / 1)      : 100%|█████████▉| 4.77GB / 4.77GB,  216MB/s  

New Data Upload               : 100%|█████████▉| 4.77GB / 4.77GB,  216MB/s  


  ...e-03b89ee9/train_state_16: 100%|██████████| 4.77GB / 4.77GB            
Processing Files (1 / 1)      : 100%|██████████| 4.77GB / 4.77GB,  204MB/s  

New Data Upload               : 100%|██████████| 4.77GB / 4.77GB,  204MB/s  
Processing Files (1 / 1)      : 100%|██████████| 4.77GB / 4.77GB,  204MB/s  

New Data Upload               : 100%|██████████| 4.77GB / 4.77GB,  204MB/s  

  ...e-03b89ee9/train_state_16: 100%|██████████| 4.77GB / 4.77GB            
Uploaded train_state_16 to seyedparsa/thoughtformer-checkpoints/checkpoints/star-datacurr-only-4gpu-bs32-lr1e-5_n1000_qwen3-0.6b-base_thoughtformer_lr1e-5_no-reset_dc_no-tc_adaptive-03b89ee9/train_state_16
Deleted old checkpoint: train_state_15
Deleted old checkpoint: checkpoint_15

Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]
Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]
Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]eval loss 0.02968258064305573

Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]
[CORRECT] Example 0 (diff=3)
  Expected : 'Claude'
  Extracted: 'Claude'
  Generated: '### Claude'

[CORRECT] Example 1 (diff=8)
  Expected : 'Guy'
  Extracted: 'Guy'
  Generated: '### Guy'

Test Accuracy:   3%|▎         | 1/32 [00:06<03:32,  6.86s/it]
Test Accuracy:   3%|▎         | 1/32 [00:06<03:32,  6.86s/it]
Test Accuracy:   3%|▎         | 1/32 [00:06<03:32,  6.86s/it]
Test accuracy: 1.0:   3%|▎         | 1/32 [00:06<03:32,  6.86s/it]
Test accuracy: 1.0:   3%|▎         | 1/32 [00:06<03:32,  6.86s/it]
Test accuracy: 0.88:   3%|▎         | 1/32 [00:06<03:32,  6.86s/it]
[WRONG] Example 2 (diff=10)
  Expected : 'Elvis'
  Extracted: 'Shannon'
  Generated: '### Shannon'

[CORRECT] Example 3 (diff=8)
  Expected : 'Coral'
  Extracted: 'Coral'
  Generated: '### Coral'

[CORRECT] Example 4 (diff=1)
  Expected : 'Patricia'
  Extracted: 'Patricia'
  Generated: '### Patricia'

Test Accuracy:   3%|▎         | 1/32 [00:06<03:27,  6.69s/it]
Test accuracy: 0.88:   3%|▎         | 1/32 [00:06<03:27,  6.69s/it]
Test accuracy: 1.0:   6%|▋         | 2/32 [00:13<03:17,  6.59s/it]
Test accuracy: 0.88:   6%|▋         | 2/32 [00:13<03:17,  6.59s/it]
Test accuracy: 1.0:   6%|▋         | 2/32 [00:13<03:17,  6.59s/it]
Test accuracy: 0.94:   6%|▋         | 2/32 [00:13<03:17,  6.59s/it]
Test accuracy: 0.88:   6%|▋         | 2/32 [00:13<03:17,  6.59s/it]
Test accuracy: 0.94:   6%|▋         | 2/32 [00:13<03:17,  6.59s/it]
Test accuracy: 0.88:   6%|▋         | 2/32 [00:13<03:15,  6.51s/it]
Test accuracy: 0.94:   6%|▋         | 2/32 [00:13<03:15,  6.51s/it]
Test accuracy: 0.94:   9%|▉         | 3/32 [00:20<03:13,  6.68s/it]
Test accuracy: 0.96:   9%|▉         | 3/32 [00:20<03:13,  6.68s/it]
Test accuracy: 0.88:   9%|▉         | 3/32 [00:20<03:13,  6.68s/it]
Test accuracy: 0.92:   9%|▉         | 3/32 [00:20<03:13,  6.68s/it]
Test accuracy: 0.94:   9%|▉         | 3/32 [00:19<03:12,  6.63s/it]
Test accuracy: 0.96:   9%|▉         | 3/32 [00:19<03:12,  6.63s/it]
Test accuracy: 0.94:   9%|▉         | 3/32 [00:20<03:13,  6.68s/it]
Test accuracy: 0.96:   9%|▉         | 3/32 [00:20<03:13,  6.68s/it]
Test accuracy: 0.96:  12%|█▎        | 4/32 [00:26<03:05,  6.62s/it]
Test accuracy: 0.96:  12%|█▎        | 4/32 [00:26<03:05,  6.62s/it]
Test accuracy: 0.92:  12%|█▎        | 4/32 [00:26<03:05,  6.62s/it]
Test accuracy: 0.97:  12%|█▎        | 4/32 [00:26<03:05,  6.62s/it]
Test accuracy: 0.91:  12%|█▎        | 4/32 [00:26<03:05,  6.62s/it]
Test accuracy: 0.96:  12%|█▎        | 4/32 [00:26<03:04,  6.59s/it]
Test accuracy: 0.91:  12%|█▎        | 4/32 [00:26<03:05,  6.62s/it]
Test accuracy: 0.94:  12%|█▎        | 4/32 [00:26<03:04,  6.59s/it]
Test accuracy: 0.91:  16%|█▌        | 5/32 [00:32<02:53,  6.44s/it]
Test accuracy: 0.91:  16%|█▌        | 5/32 [00:32<02:53,  6.44s/it]
Test accuracy: 0.97:  16%|█▌        | 5/32 [00:32<02:53,  6.44s/it]
Test accuracy: 0.94:  16%|█▌        | 5/32 [00:32<02:53,  6.43s/it]
Test accuracy: 0.93:  16%|█▌        | 5/32 [00:32<02:53,  6.44s/it]
Test accuracy: 0.88:  16%|█▌        | 5/32 [00:32<02:53,  6.44s/it]
Test accuracy: 0.98:  16%|█▌        | 5/32 [00:32<02:53,  6.44s/it]
Test accuracy: 0.95:  16%|█▌        | 5/32 [00:32<02:53,  6.43s/it]
Test accuracy: 0.93:  19%|█▉        | 6/32 [00:39<02:46,  6.41s/it]
Test accuracy: 0.98:  19%|█▉        | 6/32 [00:39<02:46,  6.41s/it]
Test accuracy: 0.88:  19%|█▉        | 6/32 [00:39<02:46,  6.41s/it]
Test accuracy: 0.92:  19%|█▉        | 6/32 [00:39<02:46,  6.41s/it]
Test accuracy: 0.98:  19%|█▉        | 6/32 [00:39<02:46,  6.41s/it]
Test accuracy: 0.9:  19%|█▉        | 6/32 [00:39<02:46,  6.41s/it] 
Test accuracy: 0.95:  19%|█▉        | 6/32 [00:38<02:46,  6.40s/it]
Test accuracy: 0.96:  19%|█▉        | 6/32 [00:38<02:46,  6.40s/it]
Test accuracy: 0.92:  22%|██▏       | 7/32 [00:45<02:39,  6.37s/it]
Test accuracy: 0.98:  22%|██▏       | 7/32 [00:45<02:39,  6.37s/it]
Test accuracy: 0.9:  22%|██▏       | 7/32 [00:45<02:39,  6.37s/it]
Test accuracy: 0.93:  22%|██▏       | 7/32 [00:45<02:39,  6.37s/it]
Test accuracy: 0.96:  22%|██▏       | 7/32 [00:45<02:39,  6.37s/it]
Test accuracy: 0.91:  22%|██▏       | 7/32 [00:45<02:39,  6.37s/it]
Test accuracy: 0.96:  22%|██▏       | 7/32 [00:45<02:39,  6.37s/it]
Test accuracy: 0.96:  22%|██▏       | 7/32 [00:45<02:39,  6.37s/it]
Test accuracy: 0.93:  25%|██▌       | 8/32 [00:51<02:31,  6.30s/it]
Test accuracy: 0.96:  25%|██▌       | 8/32 [00:51<02:31,  6.30s/it]
Test accuracy: 0.94:  25%|██▌       | 8/32 [00:51<02:31,  6.30s/it]
Test accuracy: 0.91:  25%|██▌       | 8/32 [00:51<02:31,  6.30s/it]
Test accuracy: 0.96:  25%|██▌       | 8/32 [00:51<02:30,  6.29s/it]
Test accuracy: 0.97:  25%|██▌       | 8/32 [00:51<02:31,  6.30s/it]
Test accuracy: 0.92:  25%|██▌       | 8/32 [00:51<02:31,  6.30s/it]
Test accuracy: 0.97:  25%|██▌       | 8/32 [00:51<02:30,  6.29s/it]
Test accuracy: 0.97:  28%|██▊       | 9/32 [00:58<02:30,  6.56s/it]
Test accuracy: 0.92:  28%|██▊       | 9/32 [00:58<02:30,  6.56s/it]
Test accuracy: 0.97:  28%|██▊       | 9/32 [00:58<02:30,  6.56s/it]
Test accuracy: 0.93:  28%|██▊       | 9/32 [00:58<02:30,  6.56s/it]
Test accuracy: 0.97:  28%|██▊       | 9/32 [00:58<02:30,  6.55s/it]
Test accuracy: 0.96:  28%|██▊       | 9/32 [00:58<02:30,  6.55s/it]
Test accuracy: 0.94:  28%|██▊       | 9/32 [00:58<02:30,  6.56s/it]
Test accuracy: 0.94:  28%|██▊       | 9/32 [00:58<02:30,  6.56s/it]
Test accuracy: 0.93:  31%|███▏      | 10/32 [01:04<02:22,  6.49s/it]
Test accuracy: 0.96:  31%|███▏      | 10/32 [01:04<02:22,  6.49s/it]
Test accuracy: 0.94:  31%|███▏      | 10/32 [01:04<02:22,  6.49s/it]
Test accuracy: 0.94:  31%|███▏      | 10/32 [01:04<02:22,  6.49s/it]
Test accuracy: 0.96:  31%|███▏      | 10/32 [01:04<02:22,  6.49s/it]
Test accuracy: 0.95:  31%|███▏      | 10/32 [01:04<02:22,  6.49s/it]
Test accuracy: 0.97:  31%|███▏      | 10/32 [01:04<02:22,  6.49s/it]
Test accuracy: 0.96:  31%|███▏      | 10/32 [01:04<02:22,  6.49s/it]
Test accuracy: 0.94:  34%|███▍      | 11/32 [01:11<02:13,  6.38s/it]
Test accuracy: 0.96:  34%|███▍      | 11/32 [01:10<02:13,  6.38s/it]
Test accuracy: 0.96:  34%|███▍      | 11/32 [01:11<02:13,  6.38s/it]
Test accuracy: 0.92:  34%|███▍      | 11/32 [01:11<02:13,  6.38s/it]
Test accuracy: 0.95:  34%|███▍      | 11/32 [01:11<02:13,  6.38s/it]
Test accuracy: 0.95:  34%|███▍      | 11/32 [01:11<02:13,  6.38s/it]
Test accuracy: 0.95:  34%|███▍      | 11/32 [01:10<02:13,  6.38s/it]
Test accuracy: 0.95:  34%|███▍      | 11/32 [01:11<02:13,  6.38s/it]
Test accuracy: 0.92:  38%|███▊      | 12/32 [01:17<02:07,  6.37s/it]
Test accuracy: 0.95:  38%|███▊      | 12/32 [01:17<02:07,  6.37s/it]
Test accuracy: 0.95:  38%|███▊      | 12/32 [01:17<02:07,  6.37s/it]
Test accuracy: 0.93:  38%|███▊      | 12/32 [01:17<02:07,  6.37s/it]
Test accuracy: 0.96:  38%|███▊      | 12/32 [01:17<02:07,  6.37s/it]
Test accuracy: 0.95:  38%|███▊      | 12/32 [01:17<02:07,  6.37s/it]
Test accuracy: 0.95:  38%|███▊      | 12/32 [01:17<02:07,  6.37s/it]
Test accuracy: 0.95:  38%|███▊      | 12/32 [01:17<02:07,  6.37s/it]
Test accuracy: 0.96:  41%|████      | 13/32 [01:23<02:00,  6.35s/it]
Test accuracy: 0.95:  41%|████      | 13/32 [01:23<02:00,  6.35s/it]
Test accuracy: 0.93:  41%|████      | 13/32 [01:23<02:00,  6.35s/it]
Test accuracy: 0.95:  41%|████      | 13/32 [01:23<02:00,  6.35s/it]
Test accuracy: 0.94:  41%|████      | 13/32 [01:23<02:00,  6.35s/it]
Test accuracy: 0.93:  41%|████      | 13/32 [01:23<02:00,  6.35s/it]
Test accuracy: 0.95:  41%|████      | 13/32 [01:23<02:00,  6.35s/it]
Test accuracy: 0.94:  41%|████      | 13/32 [01:23<02:00,  6.35s/it]
Test accuracy: 0.93:  44%|████▍     | 14/32 [01:29<01:53,  6.28s/it]
Test accuracy: 0.95:  44%|████▍     | 14/32 [01:29<01:53,  6.28s/it]
Test accuracy: 0.94:  44%|████▍     | 14/32 [01:29<01:53,  6.28s/it]
Test accuracy: 0.96:  44%|████▍     | 14/32 [01:29<01:53,  6.28s/it]
Test accuracy: 0.94:  44%|████▍     | 14/32 [01:29<01:53,  6.28s/it]
Test accuracy: 0.94:  44%|████▍     | 14/32 [01:29<01:53,  6.28s/it]
Test accuracy: 0.94:  44%|████▍     | 14/32 [01:29<01:53,  6.28s/it]
Test accuracy: 0.94:  44%|████▍     | 14/32 [01:29<01:53,  6.28s/it]
Test accuracy: 0.94:  47%|████▋     | 15/32 [01:36<01:46,  6.29s/it]
Test accuracy: 0.96:  47%|████▋     | 15/32 [01:36<01:46,  6.29s/it]
Test accuracy: 0.94:  47%|████▋     | 15/32 [01:35<01:46,  6.29s/it]
Test accuracy: 0.93:  47%|████▋     | 15/32 [01:36<01:46,  6.29s/it]
Test accuracy: 0.94:  47%|████▋     | 15/32 [01:36<01:46,  6.29s/it]
Test accuracy: 0.94:  47%|████▋     | 15/32 [01:35<01:46,  6.29s/it]
Test accuracy: 0.94:  47%|████▋     | 15/32 [01:36<01:46,  6.29s/it]
Test accuracy: 0.94:  47%|████▋     | 15/32 [01:36<01:46,  6.29s/it]
Test accuracy: 0.93:  50%|█████     | 16/32 [01:42<01:40,  6.30s/it]
Test accuracy: 0.94:  50%|█████     | 16/32 [01:42<01:40,  6.30s/it]
Test accuracy: 0.94:  50%|█████     | 16/32 [01:42<01:40,  6.30s/it]
Test accuracy: 0.94:  50%|█████     | 16/32 [01:42<01:40,  6.30s/it]
Test accuracy: 0.95:  50%|█████     | 16/32 [01:42<01:40,  6.30s/it]
Test accuracy: 0.95:  50%|█████     | 16/32 [01:42<01:40,  6.30s/it]
Test accuracy: 0.94:  50%|█████     | 16/32 [01:42<01:40,  6.30s/it]
Test accuracy: 0.94:  50%|█████     | 16/32 [01:42<01:40,  6.30s/it]
Test accuracy: 0.94:  53%|█████▎    | 17/32 [01:48<01:34,  6.32s/it]
Test accuracy: 0.95:  53%|█████▎    | 17/32 [01:48<01:34,  6.32s/it]
Test accuracy: 0.95:  53%|█████▎    | 17/32 [01:48<01:34,  6.32s/it]
Test accuracy: 0.94:  53%|█████▎    | 17/32 [01:48<01:34,  6.32s/it]
Test accuracy: 0.94:  53%|█████▎    | 17/32 [01:48<01:34,  6.32s/it]
Test accuracy: 0.95:  53%|█████▎    | 17/32 [01:48<01:34,  6.32s/it]
Test accuracy: 0.94:  53%|█████▎    | 17/32 [01:48<01:34,  6.32s/it]
Test accuracy: 0.92:  53%|█████▎    | 17/32 [01:48<01:34,  6.32s/it]
Test accuracy: 0.92:  56%|█████▋    | 18/32 [01:55<01:27,  6.27s/it]
Test accuracy: 0.94:  56%|█████▋    | 18/32 [01:54<01:27,  6.27s/it]
Test accuracy: 0.94:  56%|█████▋    | 18/32 [01:54<01:27,  6.27s/it]
Test accuracy: 0.92:  56%|█████▋    | 18/32 [01:55<01:27,  6.27s/it]
Test accuracy: 0.94:  56%|█████▋    | 18/32 [01:54<01:27,  6.27s/it]
Test accuracy: 0.94:  56%|█████▋    | 18/32 [01:54<01:27,  6.27s/it]
Test accuracy: 0.95:  56%|█████▋    | 18/32 [01:54<01:27,  6.27s/it]
Test accuracy: 0.95:  56%|█████▋    | 18/32 [01:54<01:27,  6.27s/it]
Test accuracy: 0.92:  59%|█████▉    | 19/32 [02:01<01:21,  6.29s/it]
Test accuracy: 0.94:  59%|█████▉    | 19/32 [02:01<01:21,  6.29s/it]
Test accuracy: 0.94:  59%|█████▉    | 19/32 [02:01<01:21,  6.29s/it]
Test accuracy: 0.94:  59%|█████▉    | 19/32 [02:01<01:21,  6.29s/it]
Test accuracy: 0.91:  59%|█████▉    | 19/32 [02:01<01:21,  6.29s/it]
Test accuracy: 0.95:  59%|█████▉    | 19/32 [02:01<01:21,  6.29s/it]
Test accuracy: 0.95:  59%|█████▉    | 19/32 [02:01<01:21,  6.29s/it]
Test accuracy: 0.95:  59%|█████▉    | 19/32 [02:01<01:21,  6.29s/it]
Test accuracy: 0.94:  62%|██████▎   | 20/32 [02:07<01:15,  6.31s/it]
Test accuracy: 0.95:  62%|██████▎   | 20/32 [02:07<01:15,  6.31s/it]
Test accuracy: 0.91:  62%|██████▎   | 20/32 [02:07<01:15,  6.31s/it]
Test accuracy: 0.94:  62%|██████▎   | 20/32 [02:07<01:15,  6.31s/it]
Test accuracy: 0.95:  62%|██████▎   | 20/32 [02:07<01:15,  6.31s/it]
Test accuracy: 0.95:  62%|██████▎   | 20/32 [02:07<01:15,  6.31s/it]
Test accuracy: 0.91:  62%|██████▎   | 20/32 [02:07<01:15,  6.31s/it]
Test accuracy: 0.94:  62%|██████▎   | 20/32 [02:07<01:15,  6.31s/it]
Test accuracy: 0.95:  66%|██████▌   | 21/32 [02:14<01:11,  6.47s/it]
Test accuracy: 0.94:  66%|██████▌   | 21/32 [02:14<01:11,  6.47s/it]
Test accuracy: 0.94:  66%|██████▌   | 21/32 [02:14<01:11,  6.47s/it]
Test accuracy: 0.95:  66%|██████▌   | 21/32 [02:14<01:11,  6.47s/it]
Test accuracy: 0.94:  66%|██████▌   | 21/32 [02:14<01:11,  6.47s/it]
Test accuracy: 0.93:  66%|██████▌   | 21/32 [02:14<01:11,  6.47s/it]
Test accuracy: 0.91:  66%|██████▌   | 21/32 [02:14<01:11,  6.47s/it]
Test accuracy: 0.92:  66%|██████▌   | 21/32 [02:14<01:11,  6.47s/it]
Test accuracy: 0.95:  69%|██████▉   | 22/32 [02:20<01:04,  6.43s/it]
Test accuracy: 0.93:  69%|██████▉   | 22/32 [02:20<01:04,  6.43s/it]
Test accuracy: 0.94:  69%|██████▉   | 22/32 [02:20<01:04,  6.43s/it]
Test accuracy: 0.95:  69%|██████▉   | 22/32 [02:20<01:04,  6.43s/it]
Test accuracy: 0.94:  69%|██████▉   | 22/32 [02:20<01:04,  6.43s/it]
Test accuracy: 0.94:  69%|██████▉   | 22/32 [02:20<01:04,  6.43s/it]
Test accuracy: 0.92:  69%|██████▉   | 22/32 [02:20<01:04,  6.43s/it]
Test accuracy: 0.91:  69%|██████▉   | 22/32 [02:20<01:04,  6.43s/it]
Test accuracy: 0.94:  72%|███████▏  | 23/32 [02:27<00:57,  6.40s/it]
Test accuracy: 0.91:  72%|███████▏  | 23/32 [02:27<00:57,  6.40s/it]
Test accuracy: 0.95:  72%|███████▏  | 23/32 [02:27<00:57,  6.40s/it]
Test accuracy: 0.91:  72%|███████▏  | 23/32 [02:27<00:57,  6.40s/it]
Test accuracy: 0.94:  72%|███████▏  | 23/32 [02:27<00:57,  6.40s/it]
Test accuracy: 0.96:  72%|███████▏  | 23/32 [02:27<00:57,  6.40s/it]
Test accuracy: 0.94:  72%|███████▏  | 23/32 [02:27<00:57,  6.40s/it]
Test accuracy: 0.93:  72%|███████▏  | 23/32 [02:27<00:57,  6.40s/it]
Test accuracy: 0.91:  75%|███████▌  | 24/32 [02:34<00:52,  6.59s/it]
Test accuracy: 0.94:  75%|███████▌  | 24/32 [02:34<00:52,  6.59s/it]
Test accuracy: 0.96:  75%|███████▌  | 24/32 [02:34<00:52,  6.59s/it]
Test accuracy: 0.92:  75%|███████▌  | 24/32 [02:34<00:52,  6.59s/it]
Test accuracy: 0.94:  75%|███████▌  | 24/32 [02:34<00:52,  6.59s/it]
Test accuracy: 0.96:  75%|███████▌  | 24/32 [02:34<00:52,  6.59s/it]
Test accuracy: 0.93:  75%|███████▌  | 24/32 [02:34<00:52,  6.59s/it]
Test accuracy: 0.93:  75%|███████▌  | 24/32 [02:34<00:52,  6.59s/it]
Test accuracy: 0.92:  78%|███████▊  | 25/32 [02:40<00:45,  6.52s/it]
Test accuracy: 0.94:  78%|███████▊  | 25/32 [02:40<00:45,  6.52s/it]
Test accuracy: 0.92:  78%|███████▊  | 25/32 [02:40<00:45,  6.52s/it]
Test accuracy: 0.94:  78%|███████▊  | 25/32 [02:40<00:45,  6.52s/it]
Test accuracy: 0.96:  78%|███████▊  | 25/32 [02:40<00:45,  6.52s/it]
Test accuracy: 0.96:  78%|███████▊  | 25/32 [02:40<00:45,  6.52s/it]
Test accuracy: 0.93:  78%|███████▊  | 25/32 [02:40<00:45,  6.52s/it]
Test accuracy: 0.93:  78%|███████▊  | 25/32 [02:40<00:45,  6.52s/it]
Test accuracy: 0.96:  81%|████████▏ | 26/32 [02:46<00:38,  6.41s/it]
Test accuracy: 0.94:  81%|████████▏ | 26/32 [02:46<00:38,  6.41s/it]
Test accuracy: 0.96:  81%|████████▏ | 26/32 [02:46<00:38,  6.41s/it]
Test accuracy: 0.94:  81%|████████▏ | 26/32 [02:46<00:38,  6.41s/it]
Test accuracy: 0.93:  81%|████████▏ | 26/32 [02:46<00:38,  6.41s/it]
Test accuracy: 0.92:  81%|████████▏ | 26/32 [02:46<00:38,  6.41s/it]
Test accuracy: 0.91:  81%|████████▏ | 26/32 [02:46<00:38,  6.41s/it]
Test accuracy: 0.92:  81%|████████▏ | 26/32 [02:46<00:38,  6.41s/it]
Test accuracy: 0.94:  84%|████████▍ | 27/32 [02:53<00:32,  6.44s/it]
Test accuracy: 0.96:  84%|████████▍ | 27/32 [02:53<00:32,  6.44s/it]
Test accuracy: 0.94:  84%|████████▍ | 27/32 [02:53<00:32,  6.44s/it]
Test accuracy: 0.92:  84%|████████▍ | 27/32 [02:53<00:32,  6.44s/it]
Test accuracy: 0.96:  84%|████████▍ | 27/32 [02:53<00:32,  6.44s/it]
Test accuracy: 0.92:  84%|████████▍ | 27/32 [02:53<00:32,  6.44s/it]
Test accuracy: 0.91:  84%|████████▍ | 27/32 [02:53<00:32,  6.44s/it]
Test accuracy: 0.9:  84%|████████▍ | 27/32 [02:53<00:32,  6.44s/it] 
Test accuracy: 0.94:  88%|████████▊ | 28/32 [03:00<00:26,  6.63s/it]
Test accuracy: 0.96:  88%|████████▊ | 28/32 [03:00<00:26,  6.63s/it]
Test accuracy: 0.92:  88%|████████▊ | 28/32 [03:00<00:26,  6.63s/it]
Test accuracy: 0.93:  88%|████████▊ | 28/32 [03:00<00:26,  6.63s/it]
Test accuracy: 0.96:  88%|████████▊ | 28/32 [03:00<00:26,  6.63s/it]
Test accuracy: 0.92:  88%|████████▊ | 28/32 [03:00<00:26,  6.63s/it]
Test accuracy: 0.9:  88%|████████▊ | 28/32 [03:00<00:26,  6.63s/it]
Test accuracy: 0.9:  88%|████████▊ | 28/32 [03:00<00:26,  6.63s/it]
Test accuracy: 0.93:  91%|█████████ | 29/32 [03:06<00:19,  6.49s/it]
Test accuracy: 0.93:  91%|█████████ | 29/32 [03:06<00:19,  6.49s/it]
Test accuracy: 0.96:  91%|█████████ | 29/32 [03:06<00:19,  6.49s/it]
Test accuracy: 0.92:  91%|█████████ | 29/32 [03:06<00:19,  6.49s/it]
Test accuracy: 0.96:  91%|█████████ | 29/32 [03:06<00:19,  6.49s/it]
Test accuracy: 0.92:  91%|█████████ | 29/32 [03:06<00:19,  6.49s/it]
Test accuracy: 0.9:  91%|█████████ | 29/32 [03:06<00:19,  6.49s/it]
Test accuracy: 0.91:  91%|█████████ | 29/32 [03:06<00:19,  6.49s/it]
Test accuracy: 0.93:  94%|█████████▍| 30/32 [03:12<00:12,  6.44s/it]
Test accuracy: 0.96:  94%|█████████▍| 30/32 [03:12<00:12,  6.44s/it]
Test accuracy: 0.92:  94%|█████████▍| 30/32 [03:12<00:12,  6.44s/it]
Test accuracy: 0.93:  94%|█████████▍| 30/32 [03:12<00:12,  6.44s/it]
Test accuracy: 0.96:  94%|█████████▍| 30/32 [03:12<00:12,  6.44s/it]
Test accuracy: 0.92:  94%|█████████▍| 30/32 [03:12<00:12,  6.44s/it]
Test accuracy: 0.91:  94%|█████████▍| 30/32 [03:12<00:12,  6.44s/it]
Test accuracy: 0.91:  94%|█████████▍| 30/32 [03:12<00:12,  6.44s/it]
Test accuracy: 0.93:  97%|█████████▋| 31/32 [03:19<00:06,  6.62s/it]
Test accuracy: 0.92:  97%|█████████▋| 31/32 [03:19<00:06,  6.62s/it]
Test accuracy: 0.96:  97%|█████████▋| 31/32 [03:19<00:06,  6.62s/it]
Test accuracy: 0.93:  97%|█████████▋| 31/32 [03:19<00:06,  6.62s/it]
Test accuracy: 0.96:  97%|█████████▋| 31/32 [03:19<00:06,  6.62s/it]
Test accuracy: 0.92:  97%|█████████▋| 31/32 [03:19<00:06,  6.62s/it]
Test accuracy: 0.91:  97%|█████████▋| 31/32 [03:19<00:06,  6.63s/it]
Test accuracy: 0.91:  97%|█████████▋| 31/32 [03:19<00:06,  6.63s/it]
Test accuracy: 0.92: 100%|██████████| 32/32 [03:21<00:00,  5.31s/it]
Test accuracy: 0.92: 100%|██████████| 32/32 [03:21<00:00,  5.31s/it]
Test accuracy: 0.93: 100%|██████████| 32/32 [03:22<00:00,  5.31s/it]
Test accuracy: 0.91: 100%|██████████| 32/32 [03:22<00:00,  5.31s/it]
Test accuracy: 0.93: 100%|██████████| 32/32 [03:22<00:00,  5.31s/it]
Test accuracy: 0.96: 100%|██████████| 32/32 [03:22<00:00,  5.31s/it]
Test accuracy: 0.91: 100%|██████████| 32/32 [03:22<00:00,  5.31s/it]
Test accuracy: 0.96: 100%|██████████| 32/32 [03:22<00:00,  5.31s/it]
Test accuracy: 0.91: 100%|██████████| 32/32 [03:22<00:00,  6.32s/it]
Device 2: Cor=228, CoT=0, Total=250

Test accuracy: 0.93: 100%|██████████| 32/32 [03:22<00:00,  6.32s/it]
Device 3: Cor=232, CoT=0, Total=250

Test accuracy: 0.96: 100%|██████████| 32/32 [03:22<00:00,  6.32s/it]
Device 1: Cor=240, CoT=0, Total=250

Test accuracy: 0.92: 100%|██████████| 32/32 [03:22<00:00,  6.33s/it]
Device 0: Cor=229, CoT=0, Total=250

============================================================
EVALUATION RESULTS (Epoch 15)
============================================================
Config: data_curriculum=ON, thought_curriculum=OFF
thought_stage=10, data_stage=9, n_latent_tokens=10
Training on difficulties: 0-9

Overall accuracy: 929 / 1000 = 0.9290

Per-difficulty accuracy:
  diff=1: 100/100 = 1.0000 (trained)
  diff=2: 99/100 = 0.9900 (trained)
  diff=3: 97/100 = 0.9700 (trained)
  diff=4: 100/100 = 1.0000 (trained)
  diff=5: 96/100 = 0.9600 (trained)
  diff=6: 97/100 = 0.9700 (trained)
  diff=7: 99/100 = 0.9900 (trained)
  diff=8: 97/100 = 0.9700 (trained)
  diff=9: 95/100 = 0.9500 (trained)
  diff=10: 49/100 = 0.4900 (untrained)

Trained (diff 0-9): 880/900 = 0.9778
============================================================


*** ADAPTIVE STAGING: acc_trained=0.9778, acc_diff9=0.9500 >= 0.9 ***
*** Advancing to stage 10 ***


Processing Files (0 / 0)      : |          |  0.00B /  0.00B            

New Data Upload               : |          |  0.00B /  0.00B            


  ...ve-03b89ee9/checkpoint_16:  10%|▉         |  352MB / 3.63GB            


  ...ve-03b89ee9/checkpoint_16:  10%|▉         |  352MB / 3.63GB            
Processing Files (0 / 1)      :  10%|▉         |  352MB / 3.63GB,   ???B/s  


  ...ve-03b89ee9/checkpoint_16:  20%|█▉        |  720MB / 3.63GB            
Processing Files (0 / 1)      :  20%|█▉        |  720MB / 3.63GB,   ???B/s  


  ...ve-03b89ee9/checkpoint_16:  30%|███       | 1.10GB / 3.63GB            
Processing Files (0 / 1)      :  30%|███       | 1.10GB / 3.63GB,  105MB/s  


  ...ve-03b89ee9/checkpoint_16:  41%|████      | 1.48GB / 3.63GB            
Processing Files (0 / 1)      :  41%|████      | 1.48GB / 3.63GB,  140MB/s  


  ...ve-03b89ee9/checkpoint_16:  51%|█████     | 1.86GB / 3.63GB            
Processing Files (0 / 1)      :  51%|█████     | 1.86GB / 3.63GB,  173MB/s  


  ...ve-03b89ee9/checkpoint_16:  62%|██████▏   | 2.23GB / 3.63GB            
Processing Files (0 / 1)      :  62%|██████▏   | 2.23GB / 3.63GB,  205MB/s  


  ...ve-03b89ee9/checkpoint_16:  72%|███████▏  | 2.61GB / 3.63GB            
Processing Files (0 / 1)      :  72%|███████▏  | 2.61GB / 3.63GB,  236MB/s  


  ...ve-03b89ee9/checkpoint_16:  82%|████████▏ | 2.98GB / 3.63GB            
Processing Files (0 / 1)      :  82%|████████▏ | 2.98GB / 3.63GB,  266MB/s  


  ...ve-03b89ee9/checkpoint_16:  93%|█████████▎| 3.36GB / 3.63GB            
Processing Files (0 / 1)      :  93%|█████████▎| 3.36GB / 3.63GB,  295MB/s  


  ...ve-03b89ee9/checkpoint_16: 100%|██████████| 3.63GB / 3.63GB            
Processing Files (1 / 1)      : 100%|██████████| 3.63GB / 3.63GB,  313MB/s  
Processing Files (1 / 1)      : 100%|██████████| 3.63GB / 3.63GB,  313MB/s  

New Data Upload               : |          |  0.00B /  0.00B,  0.00B/s  

  ...ve-03b89ee9/checkpoint_16: 100%|██████████| 3.63GB / 3.63GB            
No files have been modified since last commit. Skipping to prevent empty commit.
Uploaded checkpoint_16 to seyedparsa/thoughtformer-checkpoints/checkpoints/star-datacurr-only-4gpu-bs32-lr1e-5_n1000_qwen3-0.6b-base_thoughtformer_lr1e-5_no-reset_dc_no-tc_adaptive-03b89ee9/checkpoint_16

============================================================
EPOCH 16/100 (thought_stage=10, data_stage=10)
============================================================
  thought_stage=10, c_thought=1, max_difficulty=10

Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):  16%|█▌        | 156/1000 [00:00<00:00, 939.43 examples/s]
Map (num_proc=4):  16%|█▌        | 156/1000 [00:00<00:00, 927.39 examples/s]
Map (num_proc=4):  16%|█▌        | 156/1000 [00:00<00:00, 876.02 examples/s]
Map (num_proc=4):  81%|████████▏ | 814/1000 [00:00<00:00, 3403.27 examples/s]
Map (num_proc=4):  81%|████████  | 811/1000 [00:00<00:00, 3370.08 examples/s]
Map (num_proc=4):  81%|████████  | 811/1000 [00:00<00:00, 2875.76 examples/s]
Map (num_proc=4):  16%|█▌        | 157/1000 [00:00<00:01, 755.17 examples/s]
Map (num_proc=4):  81%|████████  | 812/1000 [00:00<00:00, 3040.91 examples/s]
Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 1971.04 examples/s]

Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 1977.00 examples/s]

Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 1887.83 examples/s]

Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 1745.00 examples/s]

Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:05, 1770.45 examples/s]
Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:05, 1667.64 examples/s]
Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:04, 1827.64 examples/s]
Filter (num_proc=4):  20%|██        | 2000/10000 [00:00<00:02, 3319.03 examples/s]
Filter (num_proc=4):  20%|██        | 2000/10000 [00:00<00:02, 3141.06 examples/s]
Filter (num_proc=4):  20%|██        | 2000/10000 [00:00<00:02, 3053.32 examples/s]
Filter (num_proc=4):  30%|███       | 3000/10000 [00:00<00:01, 4577.62 examples/s]
Filter (num_proc=4):  30%|███       | 3000/10000 [00:00<00:01, 4559.03 examples/s]
Filter (num_proc=4):  40%|████      | 4000/10000 [00:00<00:00, 6082.94 examples/s]
Filter (num_proc=4):  50%|█████     | 5000/10000 [00:00<00:00, 7735.07 examples/s]
Filter (num_proc=4):  45%|████▌     | 4500/10000 [00:00<00:00, 6634.98 examples/s]
Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:05, 1739.03 examples/s]
Filter (num_proc=4):  55%|█████▌    | 5500/10000 [00:00<00:00, 7869.76 examples/s]
Filter (num_proc=4):  75%|███████▌  | 7500/10000 [00:01<00:00, 11478.54 examples/s]
Filter (num_proc=4):  70%|███████   | 7000/10000 [00:01<00:00, 10125.28 examples/s]
Filter (num_proc=4):  20%|██        | 2000/10000 [00:00<00:02, 3246.58 examples/s]
Filter (num_proc=4):  70%|███████   | 7000/10000 [00:01<00:00, 9160.03 examples/s]
Filter (num_proc=4):  95%|█████████▌| 9500/10000 [00:01<00:00, 12743.51 examples/s]
Filter (num_proc=4):  30%|███       | 3000/10000 [00:00<00:01, 4603.94 examples/s]
Filter (num_proc=4):  85%|████████▌ | 8500/10000 [00:01<00:00, 10434.34 examples/s]
Filter (num_proc=4):  85%|████████▌ | 8500/10000 [00:01<00:00, 9833.83 examples/s]
Filter (num_proc=4):  50%|█████     | 5000/10000 [00:00<00:00, 7973.37 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 10577.93 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 10231.27 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 7013.56 examples/s]

Filter (num_proc=4):  75%|███████▌  | 7500/10000 [00:01<00:00, 11495.41 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 6680.49 examples/s] 

Filter (num_proc=4):  90%|█████████ | 9000/10000 [00:01<00:00, 11989.40 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 6677.15 examples/s] 

Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 6851.84 examples/s]

Map (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Map (num_proc=4):   1%|▏         | 137/10000 [00:01<01:12, 135.41 examples/s]
Map (num_proc=4):   6%|▌         | 553/10000 [00:01<00:15, 605.88 examples/s]
Map (num_proc=4):   8%|▊         | 816/10000 [00:01<00:10, 889.71 examples/s]
Map (num_proc=4):  11%|█         | 1092/10000 [00:01<00:10, 856.27 examples/s]
Map (num_proc=4):  17%|█▋        | 1749/10000 [00:01<00:04, 1667.58 examples/s]
Map (num_proc=4):  23%|██▎       | 2337/10000 [00:01<00:03, 2386.40 examples/s]
Map (num_proc=4):  29%|██▉       | 2938/10000 [00:01<00:02, 3094.78 examples/s]
Map (num_proc=4):  34%|███▍      | 3443/10000 [00:02<00:02, 2316.20 examples/s]
Map (num_proc=4):  44%|████▍     | 4380/10000 [00:02<00:01, 3515.70 examples/s]
Map (num_proc=4):  51%|█████     | 5117/10000 [00:02<00:01, 4268.14 examples/s]
Map (num_proc=4):  58%|█████▊    | 5787/10000 [00:03<00:01, 2132.22 examples/s]
Map (num_proc=4):  79%|███████▉  | 7879/10000 [00:03<00:00, 2853.60 examples/s]
Map (num_proc=4):  92%|█████████▏| 9244/10000 [00:03<00:00, 3875.03 examples/s]
Map (num_proc=4):  99%|█████████▉| 9947/10000 [00:04<00:00, 2757.04 examples/s]
Map (num_proc=4): 100%|██████████| 10000/10000 [00:05<00:00, 1934.74 examples/s]

Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 957.72 examples/s]
Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 1366.68 examples/s]
Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 1312.42 examples/s]
Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 1311.09 examples/s]
Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 1788.90 examples/s]

Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2631.64 examples/s]

Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2510.79 examples/s]

Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2514.96 examples/s]

Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):  14%|█▎        | 136/1000 [00:00<00:01, 452.01 examples/s]
Map (num_proc=4):  38%|███▊      | 385/1000 [00:00<00:00, 1054.59 examples/s]
Map (num_proc=4):  88%|████████▊ | 878/1000 [00:00<00:00, 2043.11 examples/s]
Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 1053.49 examples/s]

Training Epoch: 17:   0%|          | 0/312 [00:00<?, ?it/s]
Training Epoch: 17:   0%|          | 0/312 [00:00<?, ?it/s]
Training Epoch: 17:   0%|          | 0/312 [00:00<?, ?it/s]
Training Epoch: 17:   0%|          | 0/312 [00:00<?, ?it/s]logging training data
rank 1 | gpu 1 | alloc=4.20 GB | reserved=53.29 GB | max=41.98 GB
rank 3 | gpu 3 | alloc=4.20 GB | reserved=47.73 GB | max=41.99 GB
rank 2 | gpu 2 | alloc=4.20 GB | reserved=49.55 GB | max=41.98 GB
rank 0 | gpu 0 | alloc=4.20 GB | reserved=49.61 GB | max=41.99 GB

Training Epoch: 17/100, batch 0/625 completed (loss: 0.0183:   0%|          | 0/312 [00:03<?, ?it/s]
Training Epoch: 17/100, batch 0/625 completed (loss: 0.004:   0%|          | 0/312 [00:03<?, ?it/s]
Training Epoch: 17/100, batch 0/625 completed (loss: 0.0744:   0%|          | 0/312 [00:03<?, ?it/s]
Training Epoch: 17/100, batch 0/625 completed (loss: 0.0404:   0%|          | 0/312 [00:03<?, ?it/s]
Training Epoch: 17/100, batch 0/625 completed (loss: 0.0183:   0%|          | 1/312 [00:05<30:28,  5.88s/it]
Training Epoch: 17/100, batch 0/625 completed (loss: 0.0404:   0%|          | 1/312 [00:05<30:48,  5.94s/it]
Training Epoch: 17/100, batch 0/625 completed (loss: 0.004:   0%|          | 1/312 [00:05<30:36,  5.90s/it]
Training Epoch: 17/100, batch 0/625 completed (loss: 0.0744:   0%|          | 1/312 [00:05<30:38,  5.91s/it]
Training Epoch: 17/100, batch 1/625 completed (loss: 0.0636:   0%|          | 1/312 [00:05<30:38,  5.91s/it]
Training Epoch: 17/100, batch 1/625 completed (loss: 0.1421:   0%|          | 1/312 [00:05<30:28,  5.88s/it]
Training Epoch: 17/100, batch 1/625 completed (loss: 0.033:   0%|          | 1/312 [00:05<30:36,  5.90s/it]
Training Epoch: 17/100, batch 1/625 completed (loss: 0.1466:   0%|          | 1/312 [00:06<30:48,  5.94s/it]
Training Epoch: 17/100, batch 2/625 completed (loss: 0.1641:   0%|          | 1/312 [00:08<30:28,  5.88s/it]
Training Epoch: 17/100, batch 2/625 completed (loss: 0.0882:   0%|          | 1/312 [00:08<30:38,  5.91s/it]
Training Epoch: 17/100, batch 2/625 completed (loss: 0.0058:   0%|          | 1/312 [00:08<30:36,  5.90s/it]
Training Epoch: 17/100, batch 2/625 completed (loss: 0.0321:   0%|          | 1/312 [00:09<30:48,  5.94s/it]
Training Epoch: 17/100, batch 2/625 completed (loss: 0.0321:   1%|          | 2/312 [00:11<30:20,  5.87s/it]
Training Epoch: 17/100, batch 2/625 completed (loss: 0.0058:   1%|          | 2/312 [00:11<30:15,  5.86s/it]
Training Epoch: 17/100, batch 2/625 completed (loss: 0.0882:   1%|          | 2/312 [00:11<30:15,  5.86s/it]
Training Epoch: 17/100, batch 2/625 completed (loss: 0.1641:   1%|          | 2/312 [00:11<30:15,  5.86s/it]
Training Epoch: 17/100, batch 3/625 completed (loss: 0.0217:   1%|          | 2/312 [00:11<30:15,  5.86s/it]
Training Epoch: 17/100, batch 3/625 completed (loss: 0.0039:   1%|          | 2/312 [00:11<30:15,  5.86s/it]
Training Epoch: 17/100, batch 3/625 completed (loss: 0.1211:   1%|          | 2/312 [00:11<30:15,  5.86s/it]
Training Epoch: 17/100, batch 3/625 completed (loss: 0.0479:   1%|          | 2/312 [00:11<30:20,  5.87s/it]
Training Epoch: 17/100, batch 4/625 completed (loss: 0.1058:   1%|          | 2/312 [00:14<30:15,  5.86s/it]
Training Epoch: 17/100, batch 4/625 completed (loss: 0.0686:   1%|          | 2/312 [00:14<30:15,  5.86s/it]
Training Epoch: 17/100, batch 4/625 completed (loss: 0.0503:   1%|          | 2/312 [00:14<30:15,  5.86s/it]
Training Epoch: 17/100, batch 4/625 completed (loss: 0.1581:   1%|          | 2/312 [00:14<30:20,  5.87s/it]
Training Epoch: 17/100, batch 4/625 completed (loss: 0.0503:   1%|          | 3/312 [00:17<30:10,  5.86s/it]
Training Epoch: 17/100, batch 4/625 completed (loss: 0.1058:   1%|          | 3/312 [00:17<30:10,  5.86s/it]
Training Epoch: 17/100, batch 4/625 completed (loss: 0.1581:   1%|          | 3/312 [00:17<30:13,  5.87s/it]
Training Epoch: 17/100, batch 4/625 completed (loss: 0.0686:   1%|          | 3/312 [00:17<30:10,  5.86s/it]
Training Epoch: 17/100, batch 5/625 completed (loss: 0.0108:   1%|          | 3/312 [00:17<30:10,  5.86s/it]
Training Epoch: 17/100, batch 5/625 completed (loss: 0.0159:   1%|          | 3/312 [00:17<30:10,  5.86s/it]
Training Epoch: 17/100, batch 5/625 completed (loss: 0.0335:   1%|          | 3/312 [00:17<30:10,  5.86s/it]
Training Epoch: 17/100, batch 5/625 completed (loss: 0.0388:   1%|          | 3/312 [00:17<30:13,  5.87s/it]
Training Epoch: 17/100, batch 6/625 completed (loss: 0.2219:   1%|          | 3/312 [00:20<30:10,  5.86s/it]
Training Epoch: 17/100, batch 6/625 completed (loss: 0.0019:   1%|          | 3/312 [00:20<30:10,  5.86s/it]
Training Epoch: 17/100, batch 6/625 completed (loss: 0.0087:   1%|          | 3/312 [00:20<30:10,  5.86s/it]
Training Epoch: 17/100, batch 6/625 completed (loss: 0.0211:   1%|          | 3/312 [00:20<30:13,  5.87s/it]
Training Epoch: 17/100, batch 6/625 completed (loss: 0.0211:   1%|▏         | 4/312 [00:23<29:58,  5.84s/it]
Training Epoch: 17/100, batch 6/625 completed (loss: 0.0019:   1%|▏         | 4/312 [00:23<29:56,  5.83s/it]
Training Epoch: 17/100, batch 6/625 completed (loss: 0.2219:   1%|▏         | 4/312 [00:23<29:55,  5.83s/it]
Training Epoch: 17/100, batch 6/625 completed (loss: 0.0087:   1%|▏         | 4/312 [00:23<29:57,  5.84s/it]
Training Epoch: 17/100, batch 7/625 completed (loss: 0.107:   1%|▏         | 4/312 [00:23<29:55,  5.83s/it] 
Training Epoch: 17/100, batch 7/625 completed (loss: 0.0243:   1%|▏         | 4/312 [00:23<29:56,  5.83s/it]
Training Epoch: 17/100, batch 7/625 completed (loss: 0.0134:   1%|▏         | 4/312 [00:23<29:57,  5.84s/it]
Training Epoch: 17/100, batch 7/625 completed (loss: 0.014:   1%|▏         | 4/312 [00:23<29:58,  5.84s/it] 
Training Epoch: 17/100, batch 8/625 completed (loss: 0.0417:   1%|▏         | 4/312 [00:26<29:55,  5.83s/it]
Training Epoch: 17/100, batch 8/625 completed (loss: 0.0053:   1%|▏         | 4/312 [00:26<29:56,  5.83s/it]
Training Epoch: 17/100, batch 8/625 completed (loss: 0.021:   1%|▏         | 4/312 [00:26<29:57,  5.84s/it] 
Training Epoch: 17/100, batch 8/625 completed (loss: 0.2935:   1%|▏         | 4/312 [00:26<29:58,  5.84s/it]
Training Epoch: 17/100, batch 8/625 completed (loss: 0.2935:   2%|▏         | 5/312 [00:29<30:11,  5.90s/it]
Training Epoch: 17/100, batch 8/625 completed (loss: 0.0053:   2%|▏         | 5/312 [00:29<30:12,  5.90s/it]
Training Epoch: 17/100, batch 8/625 completed (loss: 0.021:   2%|▏         | 5/312 [00:29<30:11,  5.90s/it]
Training Epoch: 17/100, batch 8/625 completed (loss: 0.0417:   2%|▏         | 5/312 [00:29<30:12,  5.91s/it]
Training Epoch: 17/100, batch 9/625 completed (loss: 0.0036:   2%|▏         | 5/312 [00:29<30:12,  5.91s/it]
Training Epoch: 17/100, batch 9/625 completed (loss: 0.0113:   2%|▏         | 5/312 [00:29<30:11,  5.90s/it]
Training Epoch: 17/100, batch 9/625 completed (loss: 0.013:   2%|▏         | 5/312 [00:29<30:12,  5.90s/it] 
Training Epoch: 17/100, batch 9/625 completed (loss: 0.0262:   2%|▏         | 5/312 [00:29<30:11,  5.90s/it]
Training Epoch: 17/100, batch 10/625 completed (loss: 0.2232:   2%|▏         | 5/312 [00:32<30:12,  5.90s/it]
Training Epoch: 17/100, batch 10/625 completed (loss: 0.0065:   2%|▏         | 5/312 [00:32<30:12,  5.91s/it]
Training Epoch: 17/100, batch 10/625 completed (loss: 0.0369:   2%|▏         | 5/312 [00:32<30:11,  5.90s/it]
Training Epoch: 17/100, batch 10/625 completed (loss: 0.0051:   2%|▏         | 5/312 [00:32<30:11,  5.90s/it]
Training Epoch: 17/100, batch 10/625 completed (loss: 0.0065:   2%|▏         | 6/312 [00:35<29:42,  5.82s/it]
Training Epoch: 17/100, batch 10/625 completed (loss: 0.2232:   2%|▏         | 6/312 [00:35<29:43,  5.83s/it]
Training Epoch: 17/100, batch 10/625 completed (loss: 0.0369:   2%|▏         | 6/312 [00:35<29:43,  5.83s/it]
Training Epoch: 17/100, batch 10/625 completed (loss: 0.0051:   2%|▏         | 6/312 [00:35<29:45,  5.84s/it]
Training Epoch: 17/100, batch 11/625 completed (loss: 0.0181:   2%|▏         | 6/312 [00:35<29:43,  5.83s/it]
Training Epoch: 17/100, batch 11/625 completed (loss: 0.0368:   2%|▏         | 6/312 [00:35<29:42,  5.82s/it]
Training Epoch: 17/100, batch 11/625 completed (loss: 0.0146:   2%|▏         | 6/312 [00:35<29:43,  5.83s/it]
Training Epoch: 17/100, batch 11/625 completed (loss: 0.0079:   2%|▏         | 6/312 [00:35<29:45,  5.84s/it]
Training Epoch: 17/100, batch 12/625 completed (loss: 0.3526:   2%|▏         | 6/312 [00:37<29:42,  5.82s/it]
Training Epoch: 17/100, batch 12/625 completed (loss: 0.0338:   2%|▏         | 6/312 [00:37<29:43,  5.83s/it]
Training Epoch: 17/100, batch 12/625 completed (loss: 0.0271:   2%|▏         | 6/312 [00:37<29:43,  5.83s/it]
Training Epoch: 17/100, batch 12/625 completed (loss: 0.0704:   2%|▏         | 6/312 [00:38<29:45,  5.84s/it]
Training Epoch: 17/100, batch 12/625 completed (loss: 0.3526:   2%|▏         | 7/312 [00:40<29:42,  5.84s/it]
Training Epoch: 17/100, batch 12/625 completed (loss: 0.0271:   2%|▏         | 7/312 [00:40<29:43,  5.85s/it]
Training Epoch: 17/100, batch 12/625 completed (loss: 0.0704:   2%|▏         | 7/312 [00:41<29:43,  5.85s/it]
Training Epoch: 17/100, batch 12/625 completed (loss: 0.0338:   2%|▏         | 7/312 [00:40<29:43,  5.85s/it]
Training Epoch: 17/100, batch 13/625 completed (loss: 0.0688:   2%|▏         | 7/312 [00:41<29:43,  5.85s/it]
Training Epoch: 17/100, batch 13/625 completed (loss: 0.1941:   2%|▏         | 7/312 [00:41<29:42,  5.84s/it]
Training Epoch: 17/100, batch 13/625 completed (loss: 0.1241:   2%|▏         | 7/312 [00:41<29:43,  5.85s/it]
Training Epoch: 17/100, batch 13/625 completed (loss: 0.0241:   2%|▏         | 7/312 [00:41<29:43,  5.85s/it]
Training Epoch: 17/100, batch 14/625 completed (loss: 0.0137:   2%|▏         | 7/312 [00:44<29:43,  5.85s/it]
Training Epoch: 17/100, batch 14/625 completed (loss: 0.0087:   2%|▏         | 7/312 [00:44<29:43,  5.85s/it]
Training Epoch: 17/100, batch 14/625 completed (loss: 0.0648:   2%|▏         | 7/312 [00:44<29:42,  5.84s/it]
Training Epoch: 17/100, batch 14/625 completed (loss: 0.0169:   2%|▏         | 7/312 [00:44<29:43,  5.85s/it]
Training Epoch: 17/100, batch 14/625 completed (loss: 0.0169:   3%|▎         | 8/312 [00:47<30:06,  5.94s/it]
Training Epoch: 17/100, batch 14/625 completed (loss: 0.0087:   3%|▎         | 8/312 [00:47<30:06,  5.94s/it]
Training Epoch: 17/100, batch 14/625 completed (loss: 0.0137:   3%|▎         | 8/312 [00:47<30:07,  5.94s/it]
Training Epoch: 17/100, batch 14/625 completed (loss: 0.0648:   3%|▎         | 8/312 [00:47<30:07,  5.95s/it]
Training Epoch: 17/100, batch 15/625 completed (loss: 0.5727:   3%|▎         | 8/312 [00:47<30:06,  5.94s/it]
Training Epoch: 17/100, batch 15/625 completed (loss: 0.0362:   3%|▎         | 8/312 [00:47<30:07,  5.94s/it]
Training Epoch: 17/100, batch 15/625 completed (loss: 0.222:   3%|▎         | 8/312 [00:47<30:07,  5.95s/it] 
Training Epoch: 17/100, batch 15/625 completed (loss: 0.0207:   3%|▎         | 8/312 [00:47<30:06,  5.94s/it]
Training Epoch: 17/100, batch 16/625 completed (loss: 0.0099:   3%|▎         | 8/312 [00:50<30:07,  5.94s/it]
Training Epoch: 17/100, batch 16/625 completed (loss: 0.0162:   3%|▎         | 8/312 [00:50<30:07,  5.95s/it]
Training Epoch: 17/100, batch 16/625 completed (loss: 0.2317:   3%|▎         | 8/312 [00:50<30:06,  5.94s/it]
Training Epoch: 17/100, batch 16/625 completed (loss: 0.0195:   3%|▎         | 8/312 [00:50<30:06,  5.94s/it]
Training Epoch: 17/100, batch 16/625 completed (loss: 0.0195:   3%|▎         | 9/312 [00:52<29:38,  5.87s/it]
Training Epoch: 17/100, batch 16/625 completed (loss: 0.0099:   3%|▎         | 9/312 [00:52<29:37,  5.87s/it]
Training Epoch: 17/100, batch 16/625 completed (loss: 0.2317:   3%|▎         | 9/312 [00:52<29:37,  5.87s/it]
Training Epoch: 17/100, batch 16/625 completed (loss: 0.0162:   3%|▎         | 9/312 [00:52<29:38,  5.87s/it]
Training Epoch: 17/100, batch 17/625 completed (loss: 0.0089:   3%|▎         | 9/312 [00:52<29:38,  5.87s/it]
Training Epoch: 17/100, batch 17/625 completed (loss: 0.1179:   3%|▎         | 9/312 [00:52<29:37,  5.87s/it]
Training Epoch: 17/100, batch 17/625 completed (loss: 0.0068:   3%|▎         | 9/312 [00:52<29:37,  5.87s/it]
Training Epoch: 17/100, batch 17/625 completed (loss: 0.0731:   3%|▎         | 9/312 [00:52<29:38,  5.87s/it]
Training Epoch: 17/100, batch 18/625 completed (loss: 0.0074:   3%|▎         | 9/312 [00:55<29:37,  5.87s/it]
Training Epoch: 17/100, batch 18/625 completed (loss: 0.0425:   3%|▎         | 9/312 [00:55<29:38,  5.87s/it]
Training Epoch: 17/100, batch 18/625 completed (loss: 0.0363:   3%|▎         | 9/312 [00:55<29:37,  5.87s/it]
Training Epoch: 17/100, batch 18/625 completed (loss: 0.3246:   3%|▎         | 9/312 [00:55<29:38,  5.87s/it]
Training Epoch: 17/100, batch 18/625 completed (loss: 0.0425:   3%|▎         | 10/312 [00:58<29:32,  5.87s/it]
Training Epoch: 17/100, batch 18/625 completed (loss: 0.3246:   3%|▎         | 10/312 [00:58<29:33,  5.87s/it]
Training Epoch: 17/100, batch 18/625 completed (loss: 0.0363:   3%|▎         | 10/312 [00:58<29:33,  5.87s/it]
Training Epoch: 17/100, batch 18/625 completed (loss: 0.0074:   3%|▎         | 10/312 [00:58<29:33,  5.87s/it]
Training Epoch: 17/100, batch 19/625 completed (loss: 0.0418:   3%|▎         | 10/312 [00:58<29:33,  5.87s/it]
Training Epoch: 17/100, batch 19/625 completed (loss: 0.3746:   3%|▎         | 10/312 [00:58<29:32,  5.87s/it]
Training Epoch: 17/100, batch 19/625 completed (loss: 0.0166:   3%|▎         | 10/312 [00:58<29:33,  5.87s/it]
Training Epoch: 17/100, batch 19/625 completed (loss: 0.4322:   3%|▎         | 10/312 [00:58<29:33,  5.87s/it]
Training Epoch: 17/100, batch 20/625 completed (loss: 0.3514:   3%|▎         | 10/312 [01:02<29:32,  5.87s/it]
Training Epoch: 17/100, batch 20/625 completed (loss: 0.0379:   3%|▎         | 10/312 [01:02<29:33,  5.87s/it]
Training Epoch: 17/100, batch 20/625 completed (loss: 0.0357:   3%|▎         | 10/312 [01:02<29:33,  5.87s/it]
Training Epoch: 17/100, batch 20/625 completed (loss: 0.0109:   3%|▎         | 10/312 [01:02<29:33,  5.87s/it]
Training Epoch: 17/100, batch 20/625 completed (loss: 0.0109:   4%|▎         | 11/312 [01:04<29:53,  5.96s/it]
Training Epoch: 17/100, batch 20/625 completed (loss: 0.0357:   4%|▎         | 11/312 [01:04<29:54,  5.96s/it]
Training Epoch: 17/100, batch 20/625 completed (loss: 0.3514:   4%|▎         | 11/312 [01:04<29:54,  5.96s/it]
Training Epoch: 17/100, batch 20/625 completed (loss: 0.0379:   4%|▎         | 11/312 [01:04<29:55,  5.96s/it]
Training Epoch: 17/100, batch 21/625 completed (loss: 0.1191:   4%|▎         | 11/312 [01:04<29:54,  5.96s/it]
Training Epoch: 17/100, batch 21/625 completed (loss: 0.0068:   4%|▎         | 11/312 [01:04<29:54,  5.96s/it]
Training Epoch: 17/100, batch 21/625 completed (loss: 0.0193:   4%|▎         | 11/312 [01:04<29:55,  5.96s/it]
Training Epoch: 17/100, batch 21/625 completed (loss: 0.029:   4%|▎         | 11/312 [01:04<29:53,  5.96s/it] 
Training Epoch: 17/100, batch 22/625 completed (loss: 0.0488:   4%|▎         | 11/312 [01:07<29:54,  5.96s/it]
Training Epoch: 17/100, batch 22/625 completed (loss: 0.0059:   4%|▎         | 11/312 [01:07<29:55,  5.96s/it]
Training Epoch: 17/100, batch 22/625 completed (loss: 0.1145:   4%|▎         | 11/312 [01:07<29:54,  5.96s/it]
Training Epoch: 17/100, batch 22/625 completed (loss: 0.0289:   4%|▎         | 11/312 [01:07<29:53,  5.96s/it]
Training Epoch: 17/100, batch 22/625 completed (loss: 0.1145:   4%|▍         | 12/312 [01:10<29:40,  5.93s/it]
Training Epoch: 17/100, batch 22/625 completed (loss: 0.0059:   4%|▍         | 12/312 [01:10<29:39,  5.93s/it]
Training Epoch: 17/100, batch 22/625 completed (loss: 0.0289:   4%|▍         | 12/312 [01:10<29:40,  5.94s/it]
Training Epoch: 17/100, batch 22/625 completed (loss: 0.0488:   4%|▍         | 12/312 [01:10<29:40,  5.94s/it]
Training Epoch: 17/100, batch 23/625 completed (loss: 0.0701:   4%|▍         | 12/312 [01:10<29:40,  5.93s/it]
Training Epoch: 17/100, batch 23/625 completed (loss: 0.0497:   4%|▍         | 12/312 [01:10<29:40,  5.94s/it]
Training Epoch: 17/100, batch 23/625 completed (loss: 0.0742:   4%|▍         | 12/312 [01:10<29:39,  5.93s/it]
Training Epoch: 17/100, batch 23/625 completed (loss: 0.2211:   4%|▍         | 12/312 [01:10<29:40,  5.94s/it]
Training Epoch: 17/100, batch 24/625 completed (loss: 0.5793:   4%|▍         | 12/312 [01:13<29:39,  5.93s/it]
Training Epoch: 17/100, batch 24/625 completed (loss: 0.0029:   4%|▍         | 12/312 [01:13<29:40,  5.93s/it]
Training Epoch: 17/100, batch 24/625 completed (loss: 0.0044:   4%|▍         | 12/312 [01:13<29:40,  5.94s/it]
Training Epoch: 17/100, batch 24/625 completed (loss: 0.0132:   4%|▍         | 12/312 [01:13<29:40,  5.94s/it]
Training Epoch: 17/100, batch 24/625 completed (loss: 0.0132:   4%|▍         | 13/312 [01:16<29:55,  6.01s/it]
Training Epoch: 17/100, batch 24/625 completed (loss: 0.5793:   4%|▍         | 13/312 [01:16<29:55,  6.01s/it]
Training Epoch: 17/100, batch 24/625 completed (loss: 0.0029:   4%|▍         | 13/312 [01:16<29:56,  6.01s/it]
Training Epoch: 17/100, batch 24/625 completed (loss: 0.0044:   4%|▍         | 13/312 [01:16<29:56,  6.01s/it]
Training Epoch: 17/100, batch 25/625 completed (loss: 0.0206:   4%|▍         | 13/312 [01:16<29:55,  6.01s/it]
Training Epoch: 17/100, batch 25/625 completed (loss: 0.0266:   4%|▍         | 13/312 [01:16<29:56,  6.01s/it]
Training Epoch: 17/100, batch 25/625 completed (loss: 0.0284:   4%|▍         | 13/312 [01:16<29:56,  6.01s/it]
Training Epoch: 17/100, batch 25/625 completed (loss: 0.0144:   4%|▍         | 13/312 [01:17<29:55,  6.01s/it]
Training Epoch: 17/100, batch 26/625 completed (loss: 0.2586:   4%|▍         | 13/312 [01:19<29:55,  6.01s/it]
Training Epoch: 17/100, batch 26/625 completed (loss: 0.0031:   4%|▍         | 13/312 [01:19<29:56,  6.01s/it]
Training Epoch: 17/100, batch 26/625 completed (loss: 0.0354:   4%|▍         | 13/312 [01:19<29:56,  6.01s/it]
Training Epoch: 17/100, batch 26/625 completed (loss: 0.2512:   4%|▍         | 13/312 [01:19<29:55,  6.01s/it]
Training Epoch: 17/100, batch 26/625 completed (loss: 0.0031:   4%|▍         | 14/312 [01:22<29:25,  5.92s/it]
Training Epoch: 17/100, batch 26/625 completed (loss: 0.2512:   4%|▍         | 14/312 [01:22<29:26,  5.93s/it]
Training Epoch: 17/100, batch 26/625 completed (loss: 0.0354:   4%|▍         | 14/312 [01:22<29:25,  5.93s/it]
Training Epoch: 17/100, batch 26/625 completed (loss: 0.2586:   4%|▍         | 14/312 [01:22<29:25,  5.93s/it]
Training Epoch: 17/100, batch 27/625 completed (loss: 0.2395:   4%|▍         | 14/312 [01:22<29:25,  5.93s/it]
Training Epoch: 17/100, batch 27/625 completed (loss: 0.1076:   4%|▍         | 14/312 [01:22<29:25,  5.93s/it]
Training Epoch: 17/100, batch 27/625 completed (loss: 0.025:   4%|▍         | 14/312 [01:22<29:25,  5.92s/it] 
Training Epoch: 17/100, batch 27/625 completed (loss: 0.2498:   4%|▍         | 14/312 [01:22<29:26,  5.93s/it]
Training Epoch: 17/100, batch 28/625 completed (loss: 0.0264:   4%|▍         | 14/312 [01:25<29:25,  5.93s/it]
Training Epoch: 17/100, batch 28/625 completed (loss: 0.0114:   4%|▍         | 14/312 [01:25<29:25,  5.93s/it]
Training Epoch: 17/100, batch 28/625 completed (loss: 0.006:   4%|▍         | 14/312 [01:25<29:25,  5.92s/it]
Training Epoch: 17/100, batch 28/625 completed (loss: 0.2864:   4%|▍         | 14/312 [01:25<29:26,  5.93s/it]
Training Epoch: 17/100, batch 28/625 completed (loss: 0.0114:   5%|▍         | 15/312 [01:28<29:35,  5.98s/it]
Training Epoch: 17/100, batch 28/625 completed (loss: 0.2864:   5%|▍         | 15/312 [01:28<29:36,  5.98s/it]
Training Epoch: 17/100, batch 28/625 completed (loss: 0.006:   5%|▍         | 15/312 [01:28<29:35,  5.98s/it]
Training Epoch: 17/100, batch 28/625 completed (loss: 0.0264:   5%|▍         | 15/312 [01:28<29:35,  5.98s/it]
Training Epoch: 17/100, batch 29/625 completed (loss: 0.0186:   5%|▍         | 15/312 [01:28<29:35,  5.98s/it]
Training Epoch: 17/100, batch 29/625 completed (loss: 0.0524:   5%|▍         | 15/312 [01:28<29:35,  5.98s/it]
Training Epoch: 17/100, batch 29/625 completed (loss: 0.0879:   5%|▍         | 15/312 [01:28<29:35,  5.98s/it]
Training Epoch: 17/100, batch 29/625 completed (loss: 0.0041:   5%|▍         | 15/312 [01:28<29:36,  5.98s/it]
Training Epoch: 17/100, batch 30/625 completed (loss: 0.0407:   5%|▍         | 15/312 [01:32<29:35,  5.98s/it]
Training Epoch: 17/100, batch 30/625 completed (loss: 0.0147:   5%|▍         | 15/312 [01:31<29:35,  5.98s/it]
Training Epoch: 17/100, batch 30/625 completed (loss: 0.0185:   5%|▍         | 15/312 [01:32<29:35,  5.98s/it]
Training Epoch: 17/100, batch 30/625 completed (loss: 0.0465:   5%|▍         | 15/312 [01:32<29:36,  5.98s/it]
Training Epoch: 17/100, batch 30/625 completed (loss: 0.0147:   5%|▌         | 16/312 [01:34<29:39,  6.01s/it]
Training Epoch: 17/100, batch 30/625 completed (loss: 0.0185:   5%|▌         | 16/312 [01:34<29:40,  6.01s/it]
Training Epoch: 17/100, batch 30/625 completed (loss: 0.0407:   5%|▌         | 16/312 [01:34<29:40,  6.01s/it]
Training Epoch: 17/100, batch 30/625 completed (loss: 0.0465:   5%|▌         | 16/312 [01:34<29:40,  6.02s/it]
Training Epoch: 17/100, batch 31/625 completed (loss: 0.0696:   5%|▌         | 16/312 [01:34<29:40,  6.01s/it]
Training Epoch: 17/100, batch 31/625 completed (loss: 0.0215:   5%|▌         | 16/312 [01:34<29:39,  6.01s/it]
Training Epoch: 17/100, batch 31/625 completed (loss: 0.0097:   5%|▌         | 16/312 [01:34<29:40,  6.01s/it]
Training Epoch: 17/100, batch 31/625 completed (loss: 0.0124:   5%|▌         | 16/312 [01:34<29:40,  6.02s/it]
Training Epoch: 17/100, batch 32/625 completed (loss: 0.0039:   5%|▌         | 16/312 [01:37<29:39,  6.01s/it]
Training Epoch: 17/100, batch 32/625 completed (loss: 0.2366:   5%|▌         | 16/312 [01:37<29:40,  6.01s/it]
Training Epoch: 17/100, batch 32/625 completed (loss: 0.2963:   5%|▌         | 16/312 [01:37<29:40,  6.01s/it]
Training Epoch: 17/100, batch 32/625 completed (loss: 0.0976:   5%|▌         | 16/312 [01:37<29:40,  6.02s/it]
Training Epoch: 17/100, batch 32/625 completed (loss: 0.0976:   5%|▌         | 17/312 [01:40<29:10,  5.93s/it]
Training Epoch: 17/100, batch 32/625 completed (loss: 0.0039:   5%|▌         | 17/312 [01:40<29:10,  5.93s/it]
Training Epoch: 17/100, batch 32/625 completed (loss: 0.2963:   5%|▌         | 17/312 [01:40<29:10,  5.93s/it]
Training Epoch: 17/100, batch 32/625 completed (loss: 0.2366:   5%|▌         | 17/312 [01:40<29:10,  5.93s/it]
Training Epoch: 17/100, batch 33/625 completed (loss: 0.0061:   5%|▌         | 17/312 [01:40<29:10,  5.93s/it]
Training Epoch: 17/100, batch 33/625 completed (loss: 0.3165:   5%|▌         | 17/312 [01:40<29:10,  5.93s/it]
Training Epoch: 17/100, batch 33/625 completed (loss: 0.2962:   5%|▌         | 17/312 [01:40<29:10,  5.93s/it]
Training Epoch: 17/100, batch 33/625 completed (loss: 0.1281:   5%|▌         | 17/312 [01:40<29:10,  5.93s/it]
Training Epoch: 17/100, batch 34/625 completed (loss: 0.4426:   5%|▌         | 17/312 [01:43<29:10,  5.93s/it]
Training Epoch: 17/100, batch 34/625 completed (loss: 0.0099:   5%|▌         | 17/312 [01:43<29:10,  5.93s/it]
Training Epoch: 17/100, batch 34/625 completed (loss: 0.008:   5%|▌         | 17/312 [01:43<29:10,  5.93s/it] 
Training Epoch: 17/100, batch 34/625 completed (loss: 0.1745:   5%|▌         | 17/312 [01:43<29:10,  5.93s/it]
Training Epoch: 17/100, batch 34/625 completed (loss: 0.1745:   6%|▌         | 18/312 [01:46<29:20,  5.99s/it]
Training Epoch: 17/100, batch 34/625 completed (loss: 0.0099:   6%|▌         | 18/312 [01:46<29:21,  5.99s/it]
Training Epoch: 17/100, batch 34/625 completed (loss: 0.008:   6%|▌         | 18/312 [01:46<29:21,  5.99s/it]
Training Epoch: 17/100, batch 34/625 completed (loss: 0.4426:   6%|▌         | 18/312 [01:46<29:21,  5.99s/it]
Training Epoch: 17/100, batch 35/625 completed (loss: 0.0071:   6%|▌         | 18/312 [01:46<29:21,  5.99s/it]
Training Epoch: 17/100, batch 35/625 completed (loss: 0.0424:   6%|▌         | 18/312 [01:46<29:21,  5.99s/it]
Training Epoch: 17/100, batch 35/625 completed (loss: 0.0063:   6%|▌         | 18/312 [01:46<29:21,  5.99s/it]
Training Epoch: 17/100, batch 35/625 completed (loss: 0.1561:   6%|▌         | 18/312 [01:46<29:20,  5.99s/it]
Training Epoch: 17/100, batch 36/625 completed (loss: 0.0249:   6%|▌         | 18/312 [01:49<29:21,  5.99s/it]
Training Epoch: 17/100, batch 36/625 completed (loss: 0.008:   6%|▌         | 18/312 [01:49<29:21,  5.99s/it] 
Training Epoch: 17/100, batch 36/625 completed (loss: 0.1946:   6%|▌         | 18/312 [01:49<29:21,  5.99s/it]
Training Epoch: 17/100, batch 36/625 completed (loss: 0.0445:   6%|▌         | 18/312 [01:49<29:20,  5.99s/it]
Training Epoch: 17/100, batch 36/625 completed (loss: 0.0445:   6%|▌         | 19/312 [01:52<29:09,  5.97s/it]
Training Epoch: 17/100, batch 36/625 completed (loss: 0.0249:   6%|▌         | 19/312 [01:52<29:09,  5.97s/it]
Training Epoch: 17/100, batch 36/625 completed (loss: 0.008:   6%|▌         | 19/312 [01:52<29:09,  5.97s/it]
Training Epoch: 17/100, batch 36/625 completed (loss: 0.1946:   6%|▌         | 19/312 [01:52<29:09,  5.97s/it]
Training Epoch: 17/100, batch 37/625 completed (loss: 0.0176:   6%|▌         | 19/312 [01:52<29:09,  5.97s/it]
Training Epoch: 17/100, batch 37/625 completed (loss: 0.1245:   6%|▌         | 19/312 [01:52<29:09,  5.97s/it]
Training Epoch: 17/100, batch 37/625 completed (loss: 0.0364:   6%|▌         | 19/312 [01:52<29:09,  5.97s/it]
Training Epoch: 17/100, batch 37/625 completed (loss: 0.0153:   6%|▌         | 19/312 [01:52<29:09,  5.97s/it]
Training Epoch: 17/100, batch 38/625 completed (loss: 0.0103:   6%|▌         | 19/312 [01:55<29:09,  5.97s/it]
Training Epoch: 17/100, batch 38/625 completed (loss: 0.0778:   6%|▌         | 19/312 [01:55<29:09,  5.97s/it]
Training Epoch: 17/100, batch 38/625 completed (loss: 0.0192:   6%|▌         | 19/312 [01:55<29:09,  5.97s/it]
Training Epoch: 17/100, batch 38/625 completed (loss: 0.0139:   6%|▌         | 19/312 [01:55<29:09,  5.97s/it]
Training Epoch: 17/100, batch 38/625 completed (loss: 0.0103:   6%|▋         | 20/312 [01:58<28:48,  5.92s/it]
Training Epoch: 17/100, batch 38/625 completed (loss: 0.0778:   6%|▋         | 20/312 [01:58<28:48,  5.92s/it]
Training Epoch: 17/100, batch 38/625 completed (loss: 0.0139:   6%|▋         | 20/312 [01:58<28:50,  5.93s/it]
Training Epoch: 17/100, batch 38/625 completed (loss: 0.0192:   6%|▋         | 20/312 [01:58<28:50,  5.93s/it]
Training Epoch: 17/100, batch 39/625 completed (loss: 0.0171:   6%|▋         | 20/312 [01:58<28:48,  5.92s/it]
Training Epoch: 17/100, batch 39/625 completed (loss: 0.0186:   6%|▋         | 20/312 [01:58<28:50,  5.93s/it]
Training Epoch: 17/100, batch 39/625 completed (loss: 0.0187:   6%|▋         | 20/312 [01:58<28:48,  5.92s/it]
Training Epoch: 17/100, batch 39/625 completed (loss: 0.0345:   6%|▋         | 20/312 [01:58<28:50,  5.93s/it]
Training Epoch: 17/100, batch 40/625 completed (loss: 0.0489:   6%|▋         | 20/312 [02:01<28:48,  5.92s/it]
Training Epoch: 17/100, batch 40/625 completed (loss: 0.005:   6%|▋         | 20/312 [02:01<28:48,  5.92s/it] 
Training Epoch: 17/100, batch 40/625 completed (loss: 0.3947:   6%|▋         | 20/312 [02:01<28:50,  5.93s/it]
Training Epoch: 17/100, batch 40/625 completed (loss: 0.2171:   6%|▋         | 20/312 [02:01<28:50,  5.93s/it]
Training Epoch: 17/100, batch 40/625 completed (loss: 0.005:   7%|▋         | 21/312 [02:04<29:14,  6.03s/it]
Training Epoch: 17/100, batch 40/625 completed (loss: 0.3947:   7%|▋         | 21/312 [02:04<29:13,  6.03s/it]
Training Epoch: 17/100, batch 40/625 completed (loss: 0.0489:   7%|▋         | 21/312 [02:04<29:13,  6.03s/it]
Training Epoch: 17/100, batch 40/625 completed (loss: 0.2171:   7%|▋         | 21/312 [02:04<29:14,  6.03s/it]
Training Epoch: 17/100, batch 41/625 completed (loss: 0.0025:   7%|▋         | 21/312 [02:04<29:14,  6.03s/it]
Training Epoch: 17/100, batch 41/625 completed (loss: 0.0452:   7%|▋         | 21/312 [02:04<29:13,  6.03s/it]
Training Epoch: 17/100, batch 41/625 completed (loss: 0.2343:   7%|▋         | 21/312 [02:04<29:13,  6.03s/it]
Training Epoch: 17/100, batch 41/625 completed (loss: 0.0039:   7%|▋         | 21/312 [02:04<29:14,  6.03s/it]
Training Epoch: 17/100, batch 42/625 completed (loss: 0.3283:   7%|▋         | 21/312 [02:07<29:13,  6.03s/it]
Training Epoch: 17/100, batch 42/625 completed (loss: 0.0231:   7%|▋         | 21/312 [02:07<29:14,  6.03s/it]
Training Epoch: 17/100, batch 42/625 completed (loss: 0.0212:   7%|▋         | 21/312 [02:07<29:13,  6.03s/it]
Training Epoch: 17/100, batch 42/625 completed (loss: 0.0026:   7%|▋         | 21/312 [02:07<29:14,  6.03s/it]
Training Epoch: 17/100, batch 42/625 completed (loss: 0.0212:   7%|▋         | 22/312 [02:10<28:47,  5.96s/it]
Training Epoch: 17/100, batch 42/625 completed (loss: 0.0026:   7%|▋         | 22/312 [02:10<28:48,  5.96s/it]
Training Epoch: 17/100, batch 42/625 completed (loss: 0.0231:   7%|▋         | 22/312 [02:10<28:48,  5.96s/it]
Training Epoch: 17/100, batch 42/625 completed (loss: 0.3283:   7%|▋         | 22/312 [02:10<28:48,  5.96s/it]
Training Epoch: 17/100, batch 43/625 completed (loss: 0.0504:   7%|▋         | 22/312 [02:10<28:47,  5.96s/it]
Training Epoch: 17/100, batch 43/625 completed (loss: 0.0102:   7%|▋         | 22/312 [02:10<28:48,  5.96s/it]
Training Epoch: 17/100, batch 43/625 completed (loss: 0.0147:   7%|▋         | 22/312 [02:10<28:48,  5.96s/it]
Training Epoch: 17/100, batch 43/625 completed (loss: 0.1052:   7%|▋         | 22/312 [02:10<28:48,  5.96s/it]
Training Epoch: 17/100, batch 44/625 completed (loss: 0.0243:   7%|▋         | 22/312 [02:13<28:48,  5.96s/it]
Training Epoch: 17/100, batch 44/625 completed (loss: 0.0247:   7%|▋         | 22/312 [02:13<28:47,  5.96s/it]
Training Epoch: 17/100, batch 44/625 completed (loss: 0.515:   7%|▋         | 22/312 [02:13<28:48,  5.96s/it] 
Training Epoch: 17/100, batch 44/625 completed (loss: 0.072:   7%|▋         | 22/312 [02:13<28:48,  5.96s/it] 
Training Epoch: 17/100, batch 44/625 completed (loss: 0.0247:   7%|▋         | 23/312 [02:16<28:45,  5.97s/it]
Training Epoch: 17/100, batch 44/625 completed (loss: 0.515:   7%|▋         | 23/312 [02:16<28:45,  5.97s/it]
Training Epoch: 17/100, batch 44/625 completed (loss: 0.072:   7%|▋         | 23/312 [02:16<28:45,  5.97s/it]
Training Epoch: 17/100, batch 44/625 completed (loss: 0.0243:   7%|▋         | 23/312 [02:16<28:45,  5.97s/it]
Training Epoch: 17/100, batch 45/625 completed (loss: 0.007:   7%|▋         | 23/312 [02:16<28:45,  5.97s/it] 
Training Epoch: 17/100, batch 45/625 completed (loss: 0.0334:   7%|▋         | 23/312 [02:16<28:45,  5.97s/it]
Training Epoch: 17/100, batch 45/625 completed (loss: 0.0087:   7%|▋         | 23/312 [02:16<28:45,  5.97s/it]
Training Epoch: 17/100, batch 45/625 completed (loss: 0.029:   7%|▋         | 23/312 [02:16<28:45,  5.97s/it]
Training Epoch: 17/100, batch 46/625 completed (loss: 0.1761:   7%|▋         | 23/312 [02:19<28:45,  5.97s/it]
Training Epoch: 17/100, batch 46/625 completed (loss: 0.0446:   7%|▋         | 23/312 [02:19<28:45,  5.97s/it]
Training Epoch: 17/100, batch 46/625 completed (loss: 0.0077:   7%|▋         | 23/312 [02:19<28:45,  5.97s/it]
Training Epoch: 17/100, batch 46/625 completed (loss: 0.0152:   7%|▋         | 23/312 [02:19<28:45,  5.97s/it]
Training Epoch: 17/100, batch 46/625 completed (loss: 0.0446:   8%|▊         | 24/312 [02:22<29:02,  6.05s/it]
Training Epoch: 17/100, batch 46/625 completed (loss: 0.0077:   8%|▊         | 24/312 [02:22<29:02,  6.05s/it]
Training Epoch: 17/100, batch 46/625 completed (loss: 0.0152:   8%|▊         | 24/312 [02:22<29:02,  6.05s/it]
Training Epoch: 17/100, batch 46/625 completed (loss: 0.1761:   8%|▊         | 24/312 [02:22<29:02,  6.05s/it]
Training Epoch: 17/100, batch 47/625 completed (loss: 0.006:   8%|▊         | 24/312 [02:22<29:02,  6.05s/it] 
Training Epoch: 17/100, batch 47/625 completed (loss: 0.0284:   8%|▊         | 24/312 [02:22<29:02,  6.05s/it]
Training Epoch: 17/100, batch 47/625 completed (loss: 0.0287:   8%|▊         | 24/312 [02:22<29:02,  6.05s/it]
Training Epoch: 17/100, batch 47/625 completed (loss: 0.0166:   8%|▊         | 24/312 [02:22<29:02,  6.05s/it]
Training Epoch: 17/100, batch 48/625 completed (loss: 0.0092:   8%|▊         | 24/312 [02:25<29:02,  6.05s/it]
Training Epoch: 17/100, batch 48/625 completed (loss: 0.0234:   8%|▊         | 24/312 [02:25<29:02,  6.05s/it]
Training Epoch: 17/100, batch 48/625 completed (loss: 0.0085:   8%|▊         | 24/312 [02:25<29:02,  6.05s/it]
Training Epoch: 17/100, batch 48/625 completed (loss: 0.048:   8%|▊         | 24/312 [02:25<29:02,  6.05s/it] 
Training Epoch: 17/100, batch 48/625 completed (loss: 0.0085:   8%|▊         | 25/312 [02:28<28:36,  5.98s/it]
Training Epoch: 17/100, batch 48/625 completed (loss: 0.0092:   8%|▊         | 25/312 [02:28<28:36,  5.98s/it]
Training Epoch: 17/100, batch 48/625 completed (loss: 0.0234:   8%|▊         | 25/312 [02:28<28:36,  5.98s/it]
Training Epoch: 17/100, batch 48/625 completed (loss: 0.048:   8%|▊         | 25/312 [02:28<28:36,  5.98s/it]
Training Epoch: 17/100, batch 49/625 completed (loss: 0.0181:   8%|▊         | 25/312 [02:28<28:36,  5.98s/it]
Training Epoch: 17/100, batch 49/625 completed (loss: 0.0133:   8%|▊         | 25/312 [02:28<28:36,  5.98s/it]
Training Epoch: 17/100, batch 49/625 completed (loss: 0.0188:   8%|▊         | 25/312 [02:28<28:36,  5.98s/it]
Training Epoch: 17/100, batch 49/625 completed (loss: 0.3287:   8%|▊         | 25/312 [02:28<28:36,  5.98s/it]
Training Epoch: 17/100, batch 50/625 completed (loss: 0.0091:   8%|▊         | 25/312 [02:31<28:36,  5.98s/it]
Training Epoch: 17/100, batch 50/625 completed (loss: 0.0109:   8%|▊         | 25/312 [02:31<28:36,  5.98s/it]
Training Epoch: 17/100, batch 50/625 completed (loss: 0.025:   8%|▊         | 25/312 [02:31<28:36,  5.98s/it] 
Training Epoch: 17/100, batch 50/625 completed (loss: 0.0827:   8%|▊         | 25/312 [02:31<28:36,  5.98s/it]
Training Epoch: 17/100, batch 50/625 completed (loss: 0.0827:   8%|▊         | 26/312 [02:35<29:06,  6.11s/it]
Training Epoch: 17/100, batch 50/625 completed (loss: 0.0109:   8%|▊         | 26/312 [02:34<29:06,  6.11s/it]
Training Epoch: 17/100, batch 50/625 completed (loss: 0.025:   8%|▊         | 26/312 [02:34<29:07,  6.11s/it]
Training Epoch: 17/100, batch 50/625 completed (loss: 0.0091:   8%|▊         | 26/312 [02:34<29:06,  6.11s/it]
Training Epoch: 17/100, batch 51/625 completed (loss: 0.1891:   8%|▊         | 26/312 [02:35<29:06,  6.11s/it]
Training Epoch: 17/100, batch 51/625 completed (loss: 0.0194:   8%|▊         | 26/312 [02:35<29:06,  6.11s/it]
Training Epoch: 17/100, batch 51/625 completed (loss: 0.0386:   8%|▊         | 26/312 [02:35<29:07,  6.11s/it]
Training Epoch: 17/100, batch 51/625 completed (loss: 0.2114:   8%|▊         | 26/312 [02:35<29:06,  6.11s/it]
Training Epoch: 17/100, batch 52/625 completed (loss: 0.0517:   8%|▊         | 26/312 [02:37<29:06,  6.11s/it]
Training Epoch: 17/100, batch 52/625 completed (loss: 0.0062:   8%|▊         | 26/312 [02:37<29:07,  6.11s/it]
Training Epoch: 17/100, batch 52/625 completed (loss: 0.1138:   8%|▊         | 26/312 [02:37<29:06,  6.11s/it]
Training Epoch: 17/100, batch 52/625 completed (loss: 0.0151:   8%|▊         | 26/312 [02:37<29:06,  6.11s/it]
Training Epoch: 17/100, batch 52/625 completed (loss: 0.0062:   9%|▊         | 27/312 [02:40<28:34,  6.02s/it]
Training Epoch: 17/100, batch 52/625 completed (loss: 0.0517:   9%|▊         | 27/312 [02:40<28:34,  6.02s/it]
Training Epoch: 17/100, batch 52/625 completed (loss: 0.1138:   9%|▊         | 27/312 [02:40<28:34,  6.02s/it]
Training Epoch: 17/100, batch 52/625 completed (loss: 0.0151:   9%|▊         | 27/312 [02:40<28:34,  6.02s/it]
Training Epoch: 17/100, batch 53/625 completed (loss: 0.2967:   9%|▊         | 27/312 [02:40<28:34,  6.02s/it]
Training Epoch: 17/100, batch 53/625 completed (loss: 0.0091:   9%|▊         | 27/312 [02:40<28:34,  6.02s/it]
Training Epoch: 17/100, batch 53/625 completed (loss: 0.0468:   9%|▊         | 27/312 [02:40<28:34,  6.02s/it]
Training Epoch: 17/100, batch 53/625 completed (loss: 0.0127:   9%|▊         | 27/312 [02:40<28:34,  6.02s/it]
Training Epoch: 17/100, batch 54/625 completed (loss: 0.1147:   9%|▊         | 27/312 [02:43<28:34,  6.02s/it]
Training Epoch: 17/100, batch 54/625 completed (loss: 0.0035:   9%|▊         | 27/312 [02:43<28:34,  6.02s/it]
Training Epoch: 17/100, batch 54/625 completed (loss: 0.0088:   9%|▊         | 27/312 [02:43<28:34,  6.02s/it]
Training Epoch: 17/100, batch 54/625 completed (loss: 0.0066:   9%|▊         | 27/312 [02:43<28:34,  6.02s/it]
Training Epoch: 17/100, batch 54/625 completed (loss: 0.0088:   9%|▉         | 28/312 [02:46<28:08,  5.95s/it]
Training Epoch: 17/100, batch 54/625 completed (loss: 0.0066:   9%|▉         | 28/312 [02:46<28:09,  5.95s/it]
Training Epoch: 17/100, batch 54/625 completed (loss: 0.0035:   9%|▉         | 28/312 [02:46<28:09,  5.95s/it]
Training Epoch: 17/100, batch 54/625 completed (loss: 0.1147:   9%|▉         | 28/312 [02:46<28:09,  5.95s/it]
Training Epoch: 17/100, batch 55/625 completed (loss: 0.0083:   9%|▉         | 28/312 [02:46<28:09,  5.95s/it]
Training Epoch: 17/100, batch 55/625 completed (loss: 0.2436:   9%|▉         | 28/312 [02:46<28:08,  5.95s/it]
Training Epoch: 17/100, batch 55/625 completed (loss: 0.219:   9%|▉         | 28/312 [02:46<28:09,  5.95s/it] 
Training Epoch: 17/100, batch 55/625 completed (loss: 0.0067:   9%|▉         | 28/312 [02:46<28:09,  5.95s/it]
Training Epoch: 17/100, batch 56/625 completed (loss: 0.1253:   9%|▉         | 28/312 [02:49<28:08,  5.95s/it]
Training Epoch: 17/100, batch 56/625 completed (loss: 0.0671:   9%|▉         | 28/312 [02:49<28:09,  5.95s/it]
Training Epoch: 17/100, batch 56/625 completed (loss: 0.0179:   9%|▉         | 28/312 [02:49<28:09,  5.95s/it]
Training Epoch: 17/100, batch 56/625 completed (loss: 0.0258:   9%|▉         | 28/312 [02:50<28:09,  5.95s/it]
Training Epoch: 17/100, batch 56/625 completed (loss: 0.0258:   9%|▉         | 29/312 [02:53<28:42,  6.09s/it]
Training Epoch: 17/100, batch 56/625 completed (loss: 0.1253:   9%|▉         | 29/312 [02:52<28:42,  6.09s/it]
Training Epoch: 17/100, batch 56/625 completed (loss: 0.0179:   9%|▉         | 29/312 [02:52<28:42,  6.09s/it]
Training Epoch: 17/100, batch 56/625 completed (loss: 0.0671:   9%|▉         | 29/312 [02:52<28:42,  6.09s/it]
Training Epoch: 17/100, batch 57/625 completed (loss: 0.0158:   9%|▉         | 29/312 [02:53<28:42,  6.09s/it]
Training Epoch: 17/100, batch 57/625 completed (loss: 0.0233:   9%|▉         | 29/312 [02:53<28:42,  6.09s/it]
Training Epoch: 17/100, batch 57/625 completed (loss: 0.0131:   9%|▉         | 29/312 [02:53<28:42,  6.09s/it]
Training Epoch: 17/100, batch 57/625 completed (loss: 0.0552:   9%|▉         | 29/312 [02:53<28:42,  6.09s/it]
Training Epoch: 17/100, batch 58/625 completed (loss: 0.0616:   9%|▉         | 29/312 [02:55<28:42,  6.09s/it]
Training Epoch: 17/100, batch 58/625 completed (loss: 0.1815:   9%|▉         | 29/312 [02:55<28:42,  6.09s/it]
Training Epoch: 17/100, batch 58/625 completed (loss: 0.0146:   9%|▉         | 29/312 [02:55<28:42,  6.09s/it]
Training Epoch: 17/100, batch 58/625 completed (loss: 0.0294:   9%|▉         | 29/312 [02:56<28:42,  6.09s/it]
Training Epoch: 17/100, batch 58/625 completed (loss: 0.1815:  10%|▉         | 30/312 [02:58<28:15,  6.01s/it]
Training Epoch: 17/100, batch 58/625 completed (loss: 0.0146:  10%|▉         | 30/312 [02:58<28:15,  6.01s/it]
Training Epoch: 17/100, batch 58/625 completed (loss: 0.0616:  10%|▉         | 30/312 [02:58<28:15,  6.01s/it]
Training Epoch: 17/100, batch 58/625 completed (loss: 0.0294:  10%|▉         | 30/312 [02:58<28:16,  6.02s/it]
Training Epoch: 17/100, batch 59/625 completed (loss: 0.0093:  10%|▉         | 30/312 [02:58<28:15,  6.01s/it]
Training Epoch: 17/100, batch 59/625 completed (loss: 0.0938:  10%|▉         | 30/312 [02:58<28:15,  6.01s/it]
Training Epoch: 17/100, batch 59/625 completed (loss: 0.0108:  10%|▉         | 30/312 [02:58<28:15,  6.01s/it]
Training Epoch: 17/100, batch 59/625 completed (loss: 0.2445:  10%|▉         | 30/312 [02:58<28:16,  6.02s/it]
Training Epoch: 17/100, batch 60/625 completed (loss: 0.0458:  10%|▉         | 30/312 [03:01<28:15,  6.01s/it]
Training Epoch: 17/100, batch 60/625 completed (loss: 0.0128:  10%|▉         | 30/312 [03:01<28:15,  6.01s/it]
Training Epoch: 17/100, batch 60/625 completed (loss: 0.1875:  10%|▉         | 30/312 [03:01<28:15,  6.01s/it]
Training Epoch: 17/100, batch 60/625 completed (loss: 0.0561:  10%|▉         | 30/312 [03:01<28:16,  6.02s/it]
Training Epoch: 17/100, batch 60/625 completed (loss: 0.0128:  10%|▉         | 31/312 [03:04<28:15,  6.04s/it]
Training Epoch: 17/100, batch 60/625 completed (loss: 0.1875:  10%|▉         | 31/312 [03:04<28:15,  6.04s/it]
Training Epoch: 17/100, batch 60/625 completed (loss: 0.0458:  10%|▉         | 31/312 [03:04<28:16,  6.04s/it]
Training Epoch: 17/100, batch 60/625 completed (loss: 0.0561:  10%|▉         | 31/312 [03:04<28:16,  6.04s/it]
Training Epoch: 17/100, batch 61/625 completed (loss: 0.0363:  10%|▉         | 31/312 [03:04<28:15,  6.04s/it]
Training Epoch: 17/100, batch 61/625 completed (loss: 0.2275:  10%|▉         | 31/312 [03:04<28:16,  6.04s/it]
Training Epoch: 17/100, batch 61/625 completed (loss: 0.031:  10%|▉         | 31/312 [03:04<28:15,  6.04s/it] 
Training Epoch: 17/100, batch 61/625 completed (loss: 0.0077:  10%|▉         | 31/312 [03:05<28:16,  6.04s/it]
Training Epoch: 17/100, batch 62/625 completed (loss: 0.0641:  10%|▉         | 31/312 [03:08<28:16,  6.04s/it]
Training Epoch: 17/100, batch 62/625 completed (loss: 0.1385:  10%|▉         | 31/312 [03:08<28:15,  6.04s/it]
Training Epoch: 17/100, batch 62/625 completed (loss: 0.0555:  10%|▉         | 31/312 [03:08<28:15,  6.04s/it]
Training Epoch: 17/100, batch 62/625 completed (loss: 0.1748:  10%|▉         | 31/312 [03:08<28:16,  6.04s/it]
Training Epoch: 17/100, batch 62/625 completed (loss: 0.1748:  10%|█         | 32/312 [03:11<28:15,  6.05s/it]
Training Epoch: 17/100, batch 62/625 completed (loss: 0.0555:  10%|█         | 32/312 [03:11<28:15,  6.06s/it]
Training Epoch: 17/100, batch 62/625 completed (loss: 0.1385:  10%|█         | 32/312 [03:11<28:17,  6.06s/it]
Training Epoch: 17/100, batch 62/625 completed (loss: 0.0641:  10%|█         | 32/312 [03:11<28:17,  6.06s/it]
Training Epoch: 17/100, batch 63/625 completed (loss: 0.0557:  10%|█         | 32/312 [03:11<28:17,  6.06s/it]
Training Epoch: 17/100, batch 63/625 completed (loss: 0.0155:  10%|█         | 32/312 [03:11<28:15,  6.06s/it]
Training Epoch: 17/100, batch 63/625 completed (loss: 0.0062:  10%|█         | 32/312 [03:11<28:17,  6.06s/it]
Training Epoch: 17/100, batch 63/625 completed (loss: 0.0957:  10%|█         | 32/312 [03:11<28:15,  6.05s/it]
Training Epoch: 17/100, batch 64/625 completed (loss: 0.0846:  10%|█         | 32/312 [03:13<28:17,  6.06s/it]
Training Epoch: 17/100, batch 64/625 completed (loss: 0.0241:  10%|█         | 32/312 [03:13<28:15,  6.06s/it]
Training Epoch: 17/100, batch 64/625 completed (loss: 0.3375:  10%|█         | 32/312 [03:13<28:17,  6.06s/it]
Training Epoch: 17/100, batch 64/625 completed (loss: 0.0212:  10%|█         | 32/312 [03:14<28:15,  6.05s/it]
Training Epoch: 17/100, batch 64/625 completed (loss: 0.0241:  11%|█         | 33/312 [03:16<27:50,  5.99s/it]
Training Epoch: 17/100, batch 64/625 completed (loss: 0.0212:  11%|█         | 33/312 [03:16<27:51,  5.99s/it]
Training Epoch: 17/100, batch 64/625 completed (loss: 0.0846:  11%|█         | 33/312 [03:16<27:50,  5.99s/it]
Training Epoch: 17/100, batch 64/625 completed (loss: 0.3375:  11%|█         | 33/312 [03:16<27:51,  5.99s/it]
Training Epoch: 17/100, batch 65/625 completed (loss: 0.0139:  11%|█         | 33/312 [03:16<27:50,  5.99s/it]
Training Epoch: 17/100, batch 65/625 completed (loss: 0.0123:  11%|█         | 33/312 [03:16<27:50,  5.99s/it]
Training Epoch: 17/100, batch 65/625 completed (loss: 0.0776:  11%|█         | 33/312 [03:16<27:51,  5.99s/it]
Training Epoch: 17/100, batch 65/625 completed (loss: 0.0255:  11%|█         | 33/312 [03:16<27:51,  5.99s/it]
Training Epoch: 17/100, batch 66/625 completed (loss: 0.061:  11%|█         | 33/312 [03:20<27:50,  5.99s/it] 
Training Epoch: 17/100, batch 66/625 completed (loss: 0.0077:  11%|█         | 33/312 [03:20<27:50,  5.99s/it]
Training Epoch: 17/100, batch 66/625 completed (loss: 0.0593:  11%|█         | 33/312 [03:20<27:51,  5.99s/it]
Training Epoch: 17/100, batch 66/625 completed (loss: 0.6059:  11%|█         | 33/312 [03:20<27:51,  5.99s/it]
Training Epoch: 17/100, batch 66/625 completed (loss: 0.6059:  11%|█         | 34/312 [03:23<28:22,  6.12s/it]
Training Epoch: 17/100, batch 66/625 completed (loss: 0.0077:  11%|█         | 34/312 [03:23<28:22,  6.12s/it]
Training Epoch: 17/100, batch 66/625 completed (loss: 0.061:  11%|█         | 34/312 [03:23<28:22,  6.12s/it]
Training Epoch: 17/100, batch 66/625 completed (loss: 0.0593:  11%|█         | 34/312 [03:23<28:21,  6.12s/it]
Training Epoch: 17/100, batch 67/625 completed (loss: 0.1137:  11%|█         | 34/312 [03:23<28:22,  6.12s/it]
Training Epoch: 17/100, batch 67/625 completed (loss: 0.1003:  11%|█         | 34/312 [03:23<28:21,  6.12s/it]
Training Epoch: 17/100, batch 67/625 completed (loss: 0.0101:  11%|█         | 34/312 [03:23<28:22,  6.12s/it]
Training Epoch: 17/100, batch 67/625 completed (loss: 0.0404:  11%|█         | 34/312 [03:23<28:22,  6.12s/it]
Training Epoch: 17/100, batch 68/625 completed (loss: 0.0269:  11%|█         | 34/312 [03:26<28:21,  6.12s/it]
Training Epoch: 17/100, batch 68/625 completed (loss: 0.0047:  11%|█         | 34/312 [03:26<28:22,  6.12s/it]
Training Epoch: 17/100, batch 68/625 completed (loss: 0.0232:  11%|█         | 34/312 [03:26<28:22,  6.12s/it]
Training Epoch: 17/100, batch 68/625 completed (loss: 0.3707:  11%|█         | 34/312 [03:26<28:22,  6.12s/it]
Training Epoch: 17/100, batch 68/625 completed (loss: 0.3707:  11%|█         | 35/312 [03:29<27:53,  6.04s/it]
Training Epoch: 17/100, batch 68/625 completed (loss: 0.0232:  11%|█         | 35/312 [03:29<27:53,  6.04s/it]
Training Epoch: 17/100, batch 68/625 completed (loss: 0.0047:  11%|█         | 35/312 [03:29<27:54,  6.04s/it]
Training Epoch: 17/100, batch 68/625 completed (loss: 0.0269:  11%|█         | 35/312 [03:29<27:55,  6.05s/it]
Training Epoch: 17/100, batch 69/625 completed (loss: 0.0203:  11%|█         | 35/312 [03:29<27:55,  6.05s/it]
Training Epoch: 17/100, batch 69/625 completed (loss: 0.0163:  11%|█         | 35/312 [03:29<27:53,  6.04s/it]
Training Epoch: 17/100, batch 69/625 completed (loss: 0.0448:  11%|█         | 35/312 [03:29<27:54,  6.04s/it]
Training Epoch: 17/100, batch 69/625 completed (loss: 0.0044:  11%|█         | 35/312 [03:29<27:53,  6.04s/it]
Training Epoch: 17/100, batch 70/625 completed (loss: 0.2135:  11%|█         | 35/312 [03:32<27:53,  6.04s/it]
Training Epoch: 17/100, batch 70/625 completed (loss: 0.0619:  11%|█         | 35/312 [03:32<27:55,  6.05s/it]
Training Epoch: 17/100, batch 70/625 completed (loss: 0.0052:  11%|█         | 35/312 [03:32<27:54,  6.04s/it]
Training Epoch: 17/100, batch 70/625 completed (loss: 0.0053:  11%|█         | 35/312 [03:32<27:53,  6.04s/it]
Training Epoch: 17/100, batch 70/625 completed (loss: 0.0053:  12%|█▏        | 36/312 [03:35<27:42,  6.02s/it]
Training Epoch: 17/100, batch 70/625 completed (loss: 0.0052:  12%|█▏        | 36/312 [03:35<27:42,  6.02s/it]
Training Epoch: 17/100, batch 70/625 completed (loss: 0.0619:  12%|█▏        | 36/312 [03:35<27:41,  6.02s/it]
Training Epoch: 17/100, batch 70/625 completed (loss: 0.2135:  12%|█▏        | 36/312 [03:35<27:43,  6.03s/it]
Training Epoch: 17/100, batch 71/625 completed (loss: 0.033:  12%|█▏        | 36/312 [03:35<27:41,  6.02s/it] 
Training Epoch: 17/100, batch 71/625 completed (loss: 0.0734:  12%|█▏        | 36/312 [03:35<27:43,  6.03s/it]
Training Epoch: 17/100, batch 71/625 completed (loss: 0.0059:  12%|█▏        | 36/312 [03:35<27:42,  6.02s/it]
Training Epoch: 17/100, batch 71/625 completed (loss: 0.1858:  12%|█▏        | 36/312 [03:35<27:42,  6.02s/it]
Training Epoch: 17/100, batch 72/625 completed (loss: 0.0472:  12%|█▏        | 36/312 [03:38<27:43,  6.03s/it]
Training Epoch: 17/100, batch 72/625 completed (loss: 0.044:  12%|█▏        | 36/312 [03:38<27:41,  6.02s/it]
Training Epoch: 17/100, batch 72/625 completed (loss: 0.008:  12%|█▏        | 36/312 [03:38<27:42,  6.02s/it] 
Training Epoch: 17/100, batch 72/625 completed (loss: 0.0032:  12%|█▏        | 36/312 [03:38<27:42,  6.02s/it]
Training Epoch: 17/100, batch 72/625 completed (loss: 0.0032:  12%|█▏        | 37/312 [03:41<28:01,  6.11s/it]
Training Epoch: 17/100, batch 72/625 completed (loss: 0.0472:  12%|█▏        | 37/312 [03:41<28:01,  6.11s/it]
Training Epoch: 17/100, batch 72/625 completed (loss: 0.008:  12%|█▏        | 37/312 [03:41<28:01,  6.11s/it]
Training Epoch: 17/100, batch 72/625 completed (loss: 0.044:  12%|█▏        | 37/312 [03:41<28:02,  6.12s/it]
Training Epoch: 17/100, batch 73/625 completed (loss: 0.005:  12%|█▏        | 37/312 [03:41<28:01,  6.11s/it] 
Training Epoch: 17/100, batch 73/625 completed (loss: 0.1377:  12%|█▏        | 37/312 [03:41<28:01,  6.11s/it]
Training Epoch: 17/100, batch 73/625 completed (loss: 0.2636:  12%|█▏        | 37/312 [03:41<28:02,  6.12s/it]
Training Epoch: 17/100, batch 73/625 completed (loss: 0.1236:  12%|█▏        | 37/312 [03:41<28:01,  6.11s/it]
Training Epoch: 17/100, batch 74/625 completed (loss: 0.3706:  12%|█▏        | 37/312 [03:44<28:02,  6.12s/it]
Training Epoch: 17/100, batch 74/625 completed (loss: 0.0619:  12%|█▏        | 37/312 [03:44<28:01,  6.11s/it]
Training Epoch: 17/100, batch 74/625 completed (loss: 0.0527:  12%|█▏        | 37/312 [03:44<28:01,  6.11s/it]
Training Epoch: 17/100, batch 74/625 completed (loss: 0.0863:  12%|█▏        | 37/312 [03:44<28:01,  6.11s/it]
Training Epoch: 17/100, batch 74/625 completed (loss: 0.0527:  12%|█▏        | 38/312 [03:47<27:30,  6.03s/it]
Training Epoch: 17/100, batch 74/625 completed (loss: 0.0863:  12%|█▏        | 38/312 [03:47<27:32,  6.03s/it]
Training Epoch: 17/100, batch 74/625 completed (loss: 0.0619:  12%|█▏        | 38/312 [03:47<27:32,  6.03s/it]
Training Epoch: 17/100, batch 74/625 completed (loss: 0.3706:  12%|█▏        | 38/312 [03:47<27:32,  6.03s/it]
Training Epoch: 17/100, batch 75/625 completed (loss: 0.0825:  12%|█▏        | 38/312 [03:47<27:32,  6.03s/it]
Training Epoch: 17/100, batch 75/625 completed (loss: 0.0222:  12%|█▏        | 38/312 [03:47<27:32,  6.03s/it]
Training Epoch: 17/100, batch 75/625 completed (loss: 0.0135:  12%|█▏        | 38/312 [03:47<27:30,  6.03s/it]
Training Epoch: 17/100, batch 75/625 completed (loss: 0.4202:  12%|█▏        | 38/312 [03:47<27:32,  6.03s/it]
Training Epoch: 17/100, batch 76/625 completed (loss: 0.0317:  12%|█▏        | 38/312 [03:50<27:32,  6.03s/it]
Training Epoch: 17/100, batch 76/625 completed (loss: 0.0404:  12%|█▏        | 38/312 [03:50<27:32,  6.03s/it]
Training Epoch: 17/100, batch 76/625 completed (loss: 0.0091:  12%|█▏        | 38/312 [03:50<27:30,  6.03s/it]
Training Epoch: 17/100, batch 76/625 completed (loss: 0.0322:  12%|█▏        | 38/312 [03:50<27:32,  6.03s/it]
Training Epoch: 17/100, batch 76/625 completed (loss: 0.0322:  12%|█▎        | 39/312 [03:53<27:31,  6.05s/it]
Training Epoch: 17/100, batch 76/625 completed (loss: 0.0404:  12%|█▎        | 39/312 [03:53<27:30,  6.05s/it]
Training Epoch: 17/100, batch 76/625 completed (loss: 0.0091:  12%|█▎        | 39/312 [03:53<27:31,  6.05s/it]
Training Epoch: 17/100, batch 76/625 completed (loss: 0.0317:  12%|█▎        | 39/312 [03:53<27:30,  6.04s/it]
Training Epoch: 17/100, batch 77/625 completed (loss: 0.0052:  12%|█▎        | 39/312 [03:53<27:31,  6.05s/it]
Training Epoch: 17/100, batch 77/625 completed (loss: 0.0612:  12%|█▎        | 39/312 [03:53<27:30,  6.04s/it]
Training Epoch: 17/100, batch 77/625 completed (loss: 0.0341:  12%|█▎        | 39/312 [03:53<27:30,  6.05s/it]
Training Epoch: 17/100, batch 77/625 completed (loss: 0.0069:  12%|█▎        | 39/312 [03:53<27:31,  6.05s/it]
Training Epoch: 17/100, batch 78/625 completed (loss: 0.0272:  12%|█▎        | 39/312 [03:56<27:31,  6.05s/it]
Training Epoch: 17/100, batch 78/625 completed (loss: 0.014:  12%|█▎        | 39/312 [03:56<27:30,  6.05s/it] 
Training Epoch: 17/100, batch 78/625 completed (loss: 0.0318:  12%|█▎        | 39/312 [03:56<27:30,  6.04s/it]
Training Epoch: 17/100, batch 78/625 completed (loss: 0.2019:  12%|█▎        | 39/312 [03:56<27:31,  6.05s/it]
Training Epoch: 17/100, batch 78/625 completed (loss: 0.2019:  13%|█▎        | 40/312 [03:59<27:23,  6.04s/it]
Training Epoch: 17/100, batch 78/625 completed (loss: 0.014:  13%|█▎        | 40/312 [03:59<27:23,  6.04s/it]
Training Epoch: 17/100, batch 78/625 completed (loss: 0.0318:  13%|█▎        | 40/312 [03:59<27:24,  6.05s/it]
Training Epoch: 17/100, batch 78/625 completed (loss: 0.0272:  13%|█▎        | 40/312 [03:59<27:26,  6.05s/it]
Training Epoch: 17/100, batch 79/625 completed (loss: 0.026:  13%|█▎        | 40/312 [03:59<27:26,  6.05s/it] 
Training Epoch: 17/100, batch 79/625 completed (loss: 0.3857:  13%|█▎        | 40/312 [03:59<27:24,  6.05s/it]
Training Epoch: 17/100, batch 79/625 completed (loss: 0.0229:  13%|█▎        | 40/312 [03:59<27:23,  6.04s/it]
Training Epoch: 17/100, batch 79/625 completed (loss: 0.0241:  13%|█▎        | 40/312 [03:59<27:23,  6.04s/it]
Training Epoch: 17/100, batch 80/625 completed (loss: 0.5458:  13%|█▎        | 40/312 [04:02<27:26,  6.05s/it]
Training Epoch: 17/100, batch 80/625 completed (loss: 0.01:  13%|█▎        | 40/312 [04:02<27:23,  6.04s/it]  
Training Epoch: 17/100, batch 80/625 completed (loss: 0.2055:  13%|█▎        | 40/312 [04:02<27:24,  6.05s/it]
Training Epoch: 17/100, batch 80/625 completed (loss: 0.0367:  13%|█▎        | 40/312 [04:02<27:23,  6.04s/it]
Training Epoch: 17/100, batch 80/625 completed (loss: 0.5458:  13%|█▎        | 41/312 [04:05<27:00,  5.98s/it]
Training Epoch: 17/100, batch 80/625 completed (loss: 0.0367:  13%|█▎        | 41/312 [04:05<27:01,  5.98s/it]
Training Epoch: 17/100, batch 80/625 completed (loss: 0.2055:  13%|█▎        | 41/312 [04:05<27:00,  5.98s/it]
Training Epoch: 17/100, batch 80/625 completed (loss: 0.01:  13%|█▎        | 41/312 [04:05<27:01,  5.98s/it]
Training Epoch: 17/100, batch 81/625 completed (loss: 0.0075:  13%|█▎        | 41/312 [04:05<27:00,  5.98s/it]
Training Epoch: 17/100, batch 81/625 completed (loss: 0.1092:  13%|█▎        | 41/312 [04:05<27:00,  5.98s/it]
Training Epoch: 17/100, batch 81/625 completed (loss: 0.0115:  13%|█▎        | 41/312 [04:05<27:01,  5.98s/it]
Training Epoch: 17/100, batch 81/625 completed (loss: 0.0841:  13%|█▎        | 41/312 [04:05<27:01,  5.98s/it]
Training Epoch: 17/100, batch 82/625 completed (loss: 0.0964:  13%|█▎        | 41/312 [04:08<27:00,  5.98s/it]
Training Epoch: 17/100, batch 82/625 completed (loss: 0.3889:  13%|█▎        | 41/312 [04:08<27:01,  5.98s/it]
Training Epoch: 17/100, batch 82/625 completed (loss: 0.0431:  13%|█▎        | 41/312 [04:08<27:00,  5.98s/it]
Training Epoch: 17/100, batch 82/625 completed (loss: 0.1785:  13%|█▎        | 41/312 [04:08<27:01,  5.98s/it]
Training Epoch: 17/100, batch 82/625 completed (loss: 0.3889:  13%|█▎        | 42/312 [04:11<27:32,  6.12s/it]
Training Epoch: 17/100, batch 82/625 completed (loss: 0.1785:  13%|█▎        | 42/312 [04:11<27:33,  6.12s/it]
Training Epoch: 17/100, batch 82/625 completed (loss: 0.0431:  13%|█▎        | 42/312 [04:11<27:33,  6.12s/it]
Training Epoch: 17/100, batch 82/625 completed (loss: 0.0964:  13%|█▎        | 42/312 [04:11<27:32,  6.12s/it]
Training Epoch: 17/100, batch 83/625 completed (loss: 0.0272:  13%|█▎        | 42/312 [04:11<27:33,  6.12s/it]
Training Epoch: 17/100, batch 83/625 completed (loss: 0.0457:  13%|█▎        | 42/312 [04:11<27:32,  6.12s/it]
Training Epoch: 17/100, batch 83/625 completed (loss: 0.2188:  13%|█▎        | 42/312 [04:11<27:32,  6.12s/it]
Training Epoch: 17/100, batch 83/625 completed (loss: 0.0621:  13%|█▎        | 42/312 [04:11<27:33,  6.12s/it]
Training Epoch: 17/100, batch 84/625 completed (loss: 0.0125:  13%|█▎        | 42/312 [04:14<27:32,  6.12s/it]
Training Epoch: 17/100, batch 84/625 completed (loss: 0.0329:  13%|█▎        | 42/312 [04:14<27:32,  6.12s/it]
Training Epoch: 17/100, batch 84/625 completed (loss: 0.0585:  13%|█▎        | 42/312 [04:14<27:33,  6.12s/it]
Training Epoch: 17/100, batch 84/625 completed (loss: 0.0571:  13%|█▎        | 42/312 [04:14<27:33,  6.12s/it]
Training Epoch: 17/100, batch 84/625 completed (loss: 0.0585:  14%|█▍        | 43/312 [04:17<27:00,  6.03s/it]
Training Epoch: 17/100, batch 84/625 completed (loss: 0.0125:  14%|█▍        | 43/312 [04:17<27:00,  6.03s/it]
Training Epoch: 17/100, batch 84/625 completed (loss: 0.0571:  14%|█▍        | 43/312 [04:17<27:01,  6.03s/it]
Training Epoch: 17/100, batch 84/625 completed (loss: 0.0329:  14%|█▍        | 43/312 [04:17<27:00,  6.02s/it]
Training Epoch: 17/100, batch 85/625 completed (loss: 0.1277:  14%|█▍        | 43/312 [04:17<27:00,  6.03s/it]
Training Epoch: 17/100, batch 85/625 completed (loss: 0.0059:  14%|█▍        | 43/312 [04:17<27:00,  6.02s/it]
Training Epoch: 17/100, batch 85/625 completed (loss: 0.0506:  14%|█▍        | 43/312 [04:17<27:00,  6.03s/it]
Training Epoch: 17/100, batch 85/625 completed (loss: 0.2292:  14%|█▍        | 43/312 [04:17<27:01,  6.03s/it]
Training Epoch: 17/100, batch 86/625 completed (loss: 0.0328:  14%|█▍        | 43/312 [04:20<27:00,  6.02s/it]
Training Epoch: 17/100, batch 86/625 completed (loss: 0.0107:  14%|█▍        | 43/312 [04:20<27:00,  6.03s/it]
Training Epoch: 17/100, batch 86/625 completed (loss: 0.3596:  14%|█▍        | 43/312 [04:20<27:00,  6.03s/it]
Training Epoch: 17/100, batch 86/625 completed (loss: 0.0106:  14%|█▍        | 43/312 [04:20<27:01,  6.03s/it]
Training Epoch: 17/100, batch 86/625 completed (loss: 0.0106:  14%|█▍        | 44/312 [04:23<27:17,  6.11s/it]
Training Epoch: 17/100, batch 86/625 completed (loss: 0.3596:  14%|█▍        | 44/312 [04:23<27:17,  6.11s/it]
Training Epoch: 17/100, batch 86/625 completed (loss: 0.0107:  14%|█▍        | 44/312 [04:23<27:17,  6.11s/it]
Training Epoch: 17/100, batch 86/625 completed (loss: 0.0328:  14%|█▍        | 44/312 [04:23<27:17,  6.11s/it]
Training Epoch: 17/100, batch 87/625 completed (loss: 0.0292:  14%|█▍        | 44/312 [04:23<27:17,  6.11s/it]
Training Epoch: 17/100, batch 87/625 completed (loss: 0.0149:  14%|█▍        | 44/312 [04:23<27:17,  6.11s/it]
Training Epoch: 17/100, batch 87/625 completed (loss: 0.0136:  14%|█▍        | 44/312 [04:23<27:17,  6.11s/it]
Training Epoch: 17/100, batch 87/625 completed (loss: 0.0057:  14%|█▍        | 44/312 [04:23<27:17,  6.11s/it]
Training Epoch: 17/100, batch 88/625 completed (loss: 0.0256:  14%|█▍        | 44/312 [04:26<27:17,  6.11s/it]
Training Epoch: 17/100, batch 88/625 completed (loss: 0.0397:  14%|█▍        | 44/312 [04:26<27:17,  6.11s/it]
Training Epoch: 17/100, batch 88/625 completed (loss: 0.0172:  14%|█▍        | 44/312 [04:26<27:17,  6.11s/it]
Training Epoch: 17/100, batch 88/625 completed (loss: 0.1536:  14%|█▍        | 44/312 [04:26<27:17,  6.11s/it]
Training Epoch: 17/100, batch 88/625 completed (loss: 0.1536:  14%|█▍        | 45/312 [04:29<27:01,  6.07s/it]
Training Epoch: 17/100, batch 88/625 completed (loss: 0.0172:  14%|█▍        | 45/312 [04:29<27:01,  6.07s/it]
Training Epoch: 17/100, batch 88/625 completed (loss: 0.0256:  14%|█▍        | 45/312 [04:29<27:00,  6.07s/it]
Training Epoch: 17/100, batch 88/625 completed (loss: 0.0397:  14%|█▍        | 45/312 [04:29<27:01,  6.07s/it]
Training Epoch: 17/100, batch 89/625 completed (loss: 0.0514:  14%|█▍        | 45/312 [04:29<27:01,  6.07s/it]
Training Epoch: 17/100, batch 89/625 completed (loss: 0.026:  14%|█▍        | 45/312 [04:29<27:00,  6.07s/it] 
Training Epoch: 17/100, batch 89/625 completed (loss: 0.0633:  14%|█▍        | 45/312 [04:29<27:01,  6.07s/it]
Training Epoch: 17/100, batch 89/625 completed (loss: 0.2104:  14%|█▍        | 45/312 [04:29<27:01,  6.07s/it]
Training Epoch: 17/100, batch 90/625 completed (loss: 0.0533:  14%|█▍        | 45/312 [04:32<27:01,  6.07s/it]
Training Epoch: 17/100, batch 90/625 completed (loss: 0.0613:  14%|█▍        | 45/312 [04:32<27:01,  6.07s/it]
Training Epoch: 17/100, batch 90/625 completed (loss: 0.076:  14%|█▍        | 45/312 [04:32<27:00,  6.07s/it]
Training Epoch: 17/100, batch 90/625 completed (loss: 0.0524:  14%|█▍        | 45/312 [04:32<27:01,  6.07s/it]
Training Epoch: 17/100, batch 90/625 completed (loss: 0.0524:  15%|█▍        | 46/312 [04:35<26:35,  6.00s/it]
Training Epoch: 17/100, batch 90/625 completed (loss: 0.0613:  15%|█▍        | 46/312 [04:35<26:35,  6.00s/it]
Training Epoch: 17/100, batch 90/625 completed (loss: 0.0533:  15%|█▍        | 46/312 [04:35<26:36,  6.00s/it]
Training Epoch: 17/100, batch 90/625 completed (loss: 0.076:  15%|█▍        | 46/312 [04:35<26:37,  6.01s/it]
Training Epoch: 17/100, batch 91/625 completed (loss: 0.0198:  15%|█▍        | 46/312 [04:35<26:37,  6.01s/it]
Training Epoch: 17/100, batch 91/625 completed (loss: 0.0411:  15%|█▍        | 46/312 [04:35<26:36,  6.00s/it]
Training Epoch: 17/100, batch 91/625 completed (loss: 0.0792:  15%|█▍        | 46/312 [04:35<26:35,  6.00s/it]
Training Epoch: 17/100, batch 91/625 completed (loss: 0.0604:  15%|█▍        | 46/312 [04:35<26:35,  6.00s/it]
Training Epoch: 17/100, batch 92/625 completed (loss: 0.061:  15%|█▍        | 46/312 [04:38<26:37,  6.01s/it] 
Training Epoch: 17/100, batch 92/625 completed (loss: 0.1873:  15%|█▍        | 46/312 [04:38<26:36,  6.00s/it]
Training Epoch: 17/100, batch 92/625 completed (loss: 0.0241:  15%|█▍        | 46/312 [04:38<26:35,  6.00s/it]
Training Epoch: 17/100, batch 92/625 completed (loss: 0.0087:  15%|█▍        | 46/312 [04:38<26:35,  6.00s/it]
Training Epoch: 17/100, batch 92/625 completed (loss: 0.1873:  15%|█▌        | 47/312 [04:42<27:03,  6.13s/it]
Training Epoch: 17/100, batch 92/625 completed (loss: 0.061:  15%|█▌        | 47/312 [04:42<27:03,  6.13s/it]
Training Epoch: 17/100, batch 92/625 completed (loss: 0.0087:  15%|█▌        | 47/312 [04:42<27:04,  6.13s/it]
Training Epoch: 17/100, batch 92/625 completed (loss: 0.0241:  15%|█▌        | 47/312 [04:42<27:05,  6.13s/it]
Training Epoch: 17/100, batch 93/625 completed (loss: 0.0828:  15%|█▌        | 47/312 [04:42<27:05,  6.13s/it]
Training Epoch: 17/100, batch 93/625 completed (loss: 0.1:  15%|█▌        | 47/312 [04:42<27:03,  6.13s/it]  
Training Epoch: 17/100, batch 93/625 completed (loss: 0.0809:  15%|█▌        | 47/312 [04:42<27:03,  6.13s/it]
Training Epoch: 17/100, batch 93/625 completed (loss: 0.0281:  15%|█▌        | 47/312 [04:42<27:04,  6.13s/it]
Training Epoch: 17/100, batch 94/625 completed (loss: 0.0032:  15%|█▌        | 47/312 [04:45<27:03,  6.13s/it]
Training Epoch: 17/100, batch 94/625 completed (loss: 0.0119:  15%|█▌        | 47/312 [04:45<27:05,  6.13s/it]
Training Epoch: 17/100, batch 94/625 completed (loss: 0.0439:  15%|█▌        | 47/312 [04:45<27:03,  6.13s/it]
Training Epoch: 17/100, batch 94/625 completed (loss: 0.0263:  15%|█▌        | 47/312 [04:45<27:04,  6.13s/it]
Training Epoch: 17/100, batch 94/625 completed (loss: 0.0263:  15%|█▌        | 48/312 [04:47<26:37,  6.05s/it]
Training Epoch: 17/100, batch 94/625 completed (loss: 0.0119:  15%|█▌        | 48/312 [04:47<26:36,  6.05s/it]
Training Epoch: 17/100, batch 94/625 completed (loss: 0.0032:  15%|█▌        | 48/312 [04:47<26:36,  6.05s/it]
Training Epoch: 17/100, batch 94/625 completed (loss: 0.0439:  15%|█▌        | 48/312 [04:47<26:37,  6.05s/it]
Training Epoch: 17/100, batch 95/625 completed (loss: 0.0154:  15%|█▌        | 48/312 [04:47<26:36,  6.05s/it]
Training Epoch: 17/100, batch 95/625 completed (loss: 0.5189:  15%|█▌        | 48/312 [04:47<26:36,  6.05s/it]
Training Epoch: 17/100, batch 95/625 completed (loss: 0.0553:  15%|█▌        | 48/312 [04:47<26:37,  6.05s/it]
Training Epoch: 17/100, batch 95/625 completed (loss: 0.0822:  15%|█▌        | 48/312 [04:48<26:37,  6.05s/it]
Training Epoch: 17/100, batch 96/625 completed (loss: 0.0156:  15%|█▌        | 48/312 [04:50<26:36,  6.05s/it]
Training Epoch: 17/100, batch 96/625 completed (loss: 0.0091:  15%|█▌        | 48/312 [04:50<26:37,  6.05s/it]
Training Epoch: 17/100, batch 96/625 completed (loss: 0.0065:  15%|█▌        | 48/312 [04:50<26:36,  6.05s/it]
Training Epoch: 17/100, batch 96/625 completed (loss: 0.1323:  15%|█▌        | 48/312 [04:50<26:37,  6.05s/it]
Training Epoch: 17/100, batch 96/625 completed (loss: 0.0065:  16%|█▌        | 49/312 [04:54<26:35,  6.07s/it]
Training Epoch: 17/100, batch 96/625 completed (loss: 0.1323:  16%|█▌        | 49/312 [04:54<26:35,  6.07s/it]
Training Epoch: 17/100, batch 96/625 completed (loss: 0.0091:  16%|█▌        | 49/312 [04:54<26:35,  6.07s/it]
Training Epoch: 17/100, batch 96/625 completed (loss: 0.0156:  16%|█▌        | 49/312 [04:54<26:35,  6.07s/it]
Training Epoch: 17/100, batch 97/625 completed (loss: 0.0767:  16%|█▌        | 49/312 [04:54<26:35,  6.07s/it]
Training Epoch: 17/100, batch 97/625 completed (loss: 0.0307:  16%|█▌        | 49/312 [04:54<26:35,  6.07s/it]
Training Epoch: 17/100, batch 97/625 completed (loss: 0.0048:  16%|█▌        | 49/312 [04:54<26:35,  6.07s/it]
Training Epoch: 17/100, batch 97/625 completed (loss: 0.0245:  16%|█▌        | 49/312 [04:54<26:35,  6.07s/it]
Training Epoch: 17/100, batch 98/625 completed (loss: 0.1184:  16%|█▌        | 49/312 [04:57<26:35,  6.07s/it]
Training Epoch: 17/100, batch 98/625 completed (loss: 0.0819:  16%|█▌        | 49/312 [04:57<26:35,  6.07s/it]
Training Epoch: 17/100, batch 98/625 completed (loss: 0.0184:  16%|█▌        | 49/312 [04:57<26:35,  6.07s/it]
Training Epoch: 17/100, batch 98/625 completed (loss: 0.0531:  16%|█▌        | 49/312 [04:57<26:35,  6.07s/it]
Training Epoch: 17/100, batch 98/625 completed (loss: 0.1184:  16%|█▌        | 50/312 [05:00<26:35,  6.09s/it]
Training Epoch: 17/100, batch 98/625 completed (loss: 0.0531:  16%|█▌        | 50/312 [05:00<26:36,  6.09s/it]
Training Epoch: 17/100, batch 98/625 completed (loss: 0.0819:  16%|█▌        | 50/312 [05:00<26:36,  6.09s/it]
Training Epoch: 17/100, batch 98/625 completed (loss: 0.0184:  16%|█▌        | 50/312 [05:00<26:37,  6.10s/it]
Training Epoch: 17/100, batch 99/625 completed (loss: 0.066:  16%|█▌        | 50/312 [05:00<26:37,  6.10s/it] 
Training Epoch: 17/100, batch 99/625 completed (loss: 0.0118:  16%|█▌        | 50/312 [05:00<26:35,  6.09s/it]
Training Epoch: 17/100, batch 99/625 completed (loss: 0.021:  16%|█▌        | 50/312 [05:00<26:36,  6.09s/it] 
Training Epoch: 17/100, batch 99/625 completed (loss: 0.1218:  16%|█▌        | 50/312 [05:00<26:36,  6.09s/it]
Training Epoch: 17/100, batch 100/625 completed (loss: 0.1972:  16%|█▌        | 50/312 [05:03<26:35,  6.09s/it]
Training Epoch: 17/100, batch 100/625 completed (loss: 0.2727:  16%|█▌        | 50/312 [05:03<26:37,  6.10s/it]
Training Epoch: 17/100, batch 100/625 completed (loss: 0.0311:  16%|█▌        | 50/312 [05:03<26:36,  6.09s/it]
Training Epoch: 17/100, batch 100/625 completed (loss: 0.0151:  16%|█▌        | 50/312 [05:03<26:36,  6.09s/it]
Training Epoch: 17/100, batch 100/625 completed (loss: 0.0151:  16%|█▋        | 51/312 [05:06<26:13,  6.03s/it]
Training Epoch: 17/100, batch 100/625 completed (loss: 0.0311:  16%|█▋        | 51/312 [05:06<26:12,  6.03s/it]
Training Epoch: 17/100, batch 100/625 completed (loss: 0.1972:  16%|█▋        | 51/312 [05:06<26:13,  6.03s/it]
Training Epoch: 17/100, batch 100/625 completed (loss: 0.2727:  16%|█▋        | 51/312 [05:06<26:13,  6.03s/it]
Training Epoch: 17/100, batch 101/625 completed (loss: 0.0261:  16%|█▋        | 51/312 [05:06<26:12,  6.03s/it]
Training Epoch: 17/100, batch 101/625 completed (loss: 0.0176:  16%|█▋        | 51/312 [05:06<26:13,  6.03s/it]
Training Epoch: 17/100, batch 101/625 completed (loss: 0.0296:  16%|█▋        | 51/312 [05:06<26:13,  6.03s/it]
Training Epoch: 17/100, batch 101/625 completed (loss: 0.074:  16%|█▋        | 51/312 [05:06<26:13,  6.03s/it] 
Training Epoch: 17/100, batch 102/625 completed (loss: 0.0867:  16%|█▋        | 51/312 [05:09<26:13,  6.03s/it]
Training Epoch: 17/100, batch 102/625 completed (loss: 0.0419:  16%|█▋        | 51/312 [05:09<26:12,  6.03s/it]
Training Epoch: 17/100, batch 102/625 completed (loss: 0.0176:  16%|█▋        | 51/312 [05:09<26:13,  6.03s/it]
Training Epoch: 17/100, batch 102/625 completed (loss: 0.0139:  16%|█▋        | 51/312 [05:09<26:13,  6.03s/it]
Training Epoch: 17/100, batch 102/625 completed (loss: 0.0139:  17%|█▋        | 52/312 [05:12<26:28,  6.11s/it]
Training Epoch: 17/100, batch 102/625 completed (loss: 0.0867:  17%|█▋        | 52/312 [05:12<26:28,  6.11s/it]
Training Epoch: 17/100, batch 102/625 completed (loss: 0.0176:  17%|█▋        | 52/312 [05:12<26:28,  6.11s/it]
Training Epoch: 17/100, batch 102/625 completed (loss: 0.0419:  17%|█▋        | 52/312 [05:12<26:29,  6.11s/it]
Training Epoch: 17/100, batch 103/625 completed (loss: 0.0044:  17%|█▋        | 52/312 [05:12<26:28,  6.11s/it]
Training Epoch: 17/100, batch 103/625 completed (loss: 0.0122:  17%|█▋        | 52/312 [05:12<26:28,  6.11s/it]
Training Epoch: 17/100, batch 103/625 completed (loss: 0.0105:  17%|█▋        | 52/312 [05:12<26:29,  6.11s/it]
Training Epoch: 17/100, batch 103/625 completed (loss: 0.0067:  17%|█▋        | 52/312 [05:12<26:28,  6.11s/it]
Training Epoch: 17/100, batch 104/625 completed (loss: 0.0115:  17%|█▋        | 52/312 [05:15<26:28,  6.11s/it]
Training Epoch: 17/100, batch 104/625 completed (loss: 0.043:  17%|█▋        | 52/312 [05:15<26:29,  6.11s/it] 
Training Epoch: 17/100, batch 104/625 completed (loss: 0.0354:  17%|█▋        | 52/312 [05:15<26:28,  6.11s/it]
Training Epoch: 17/100, batch 104/625 completed (loss: 0.0151:  17%|█▋        | 52/312 [05:15<26:28,  6.11s/it]
Training Epoch: 17/100, batch 104/625 completed (loss: 0.0115:  17%|█▋        | 53/312 [05:18<26:14,  6.08s/it]
Training Epoch: 17/100, batch 104/625 completed (loss: 0.0151:  17%|█▋        | 53/312 [05:18<26:14,  6.08s/it]
Training Epoch: 17/100, batch 104/625 completed (loss: 0.043:  17%|█▋        | 53/312 [05:18<26:14,  6.08s/it]
Training Epoch: 17/100, batch 104/625 completed (loss: 0.0354:  17%|█▋        | 53/312 [05:18<26:14,  6.08s/it]
Training Epoch: 17/100, batch 105/625 completed (loss: 0.0026:  17%|█▋        | 53/312 [05:18<26:14,  6.08s/it]
Training Epoch: 17/100, batch 105/625 completed (loss: 0.1642:  17%|█▋        | 53/312 [05:18<26:14,  6.08s/it]
Training Epoch: 17/100, batch 105/625 completed (loss: 0.0074:  17%|█▋        | 53/312 [05:18<26:14,  6.08s/it]
Training Epoch: 17/100, batch 105/625 completed (loss: 0.012:  17%|█▋        | 53/312 [05:18<26:14,  6.08s/it] 
Training Epoch: 17/100, batch 106/625 completed (loss: 0.2511:  17%|█▋        | 53/312 [05:21<26:14,  6.08s/it]
Training Epoch: 17/100, batch 106/625 completed (loss: 0.018:  17%|█▋        | 53/312 [05:21<26:14,  6.08s/it] 
Training Epoch: 17/100, batch 106/625 completed (loss: 0.0186:  17%|█▋        | 53/312 [05:21<26:14,  6.08s/it]
Training Epoch: 17/100, batch 106/625 completed (loss: 0.0502:  17%|█▋        | 53/312 [05:21<26:14,  6.08s/it]
Training Epoch: 17/100, batch 106/625 completed (loss: 0.0186:  17%|█▋        | 54/312 [05:24<25:49,  6.01s/it]
Training Epoch: 17/100, batch 106/625 completed (loss: 0.0502:  17%|█▋        | 54/312 [05:24<25:50,  6.01s/it]
Training Epoch: 17/100, batch 106/625 completed (loss: 0.018:  17%|█▋        | 54/312 [05:24<25:51,  6.01s/it]
Training Epoch: 17/100, batch 106/625 completed (loss: 0.2511:  17%|█▋        | 54/312 [05:24<25:51,  6.02s/it]
Training Epoch: 17/100, batch 107/625 completed (loss: 0.0497:  17%|█▋        | 54/312 [05:24<25:51,  6.01s/it]
Training Epoch: 17/100, batch 107/625 completed (loss: 0.1273:  17%|█▋        | 54/312 [05:24<25:51,  6.02s/it]
Training Epoch: 17/100, batch 107/625 completed (loss: 0.2434:  17%|█▋        | 54/312 [05:24<25:49,  6.01s/it]
Training Epoch: 17/100, batch 107/625 completed (loss: 0.4992:  17%|█▋        | 54/312 [05:24<25:50,  6.01s/it]
Training Epoch: 17/100, batch 108/625 completed (loss: 0.2948:  17%|█▋        | 54/312 [05:27<25:49,  6.01s/it]
Training Epoch: 17/100, batch 108/625 completed (loss: 0.0104:  17%|█▋        | 54/312 [05:27<25:51,  6.01s/it]
Training Epoch: 17/100, batch 108/625 completed (loss: 0.0496:  17%|█▋        | 54/312 [05:27<25:51,  6.02s/it]
Training Epoch: 17/100, batch 108/625 completed (loss: 0.1058:  17%|█▋        | 54/312 [05:27<25:50,  6.01s/it]
Training Epoch: 17/100, batch 108/625 completed (loss: 0.2948:  18%|█▊        | 55/312 [05:30<26:23,  6.16s/it]
Training Epoch: 17/100, batch 108/625 completed (loss: 0.0104:  18%|█▊        | 55/312 [05:30<26:23,  6.16s/it]
Training Epoch: 17/100, batch 108/625 completed (loss: 0.1058:  18%|█▊        | 55/312 [05:30<26:24,  6.17s/it]
Training Epoch: 17/100, batch 108/625 completed (loss: 0.0496:  18%|█▊        | 55/312 [05:30<26:23,  6.16s/it]
Training Epoch: 17/100, batch 109/625 completed (loss: 0.0434:  18%|█▊        | 55/312 [05:30<26:23,  6.16s/it]
Training Epoch: 17/100, batch 109/625 completed (loss: 0.1152:  18%|█▊        | 55/312 [05:30<26:23,  6.16s/it]
Training Epoch: 17/100, batch 109/625 completed (loss: 0.0089:  18%|█▊        | 55/312 [05:30<26:23,  6.16s/it]
Training Epoch: 17/100, batch 109/625 completed (loss: 0.0649:  18%|█▊        | 55/312 [05:30<26:24,  6.17s/it]
Training Epoch: 17/100, batch 110/625 completed (loss: 0.1081:  18%|█▊        | 55/312 [05:33<26:23,  6.16s/it]
Training Epoch: 17/100, batch 110/625 completed (loss: 0.0443:  18%|█▊        | 55/312 [05:33<26:23,  6.16s/it]
Training Epoch: 17/100, batch 110/625 completed (loss: 0.025:  18%|█▊        | 55/312 [05:33<26:23,  6.16s/it] 
Training Epoch: 17/100, batch 110/625 completed (loss: 0.0019:  18%|█▊        | 55/312 [05:33<26:24,  6.17s/it]
Training Epoch: 17/100, batch 110/625 completed (loss: 0.025:  18%|█▊        | 56/312 [05:36<25:52,  6.06s/it]
Training Epoch: 17/100, batch 110/625 completed (loss: 0.0019:  18%|█▊        | 56/312 [05:36<25:51,  6.06s/it]
Training Epoch: 17/100, batch 110/625 completed (loss: 0.0443:  18%|█▊        | 56/312 [05:36<25:51,  6.06s/it]
Training Epoch: 17/100, batch 110/625 completed (loss: 0.1081:  18%|█▊        | 56/312 [05:36<25:51,  6.06s/it]
Training Epoch: 17/100, batch 111/625 completed (loss: 0.0344:  18%|█▊        | 56/312 [05:36<25:51,  6.06s/it]
Training Epoch: 17/100, batch 111/625 completed (loss: 0.0568:  18%|█▊        | 56/312 [05:36<25:51,  6.06s/it]
Training Epoch: 17/100, batch 111/625 completed (loss: 0.1193:  18%|█▊        | 56/312 [05:36<25:52,  6.06s/it]
Training Epoch: 17/100, batch 111/625 completed (loss: 0.0316:  18%|█▊        | 56/312 [05:36<25:51,  6.06s/it]
Training Epoch: 17/100, batch 112/625 completed (loss: 0.1204:  18%|█▊        | 56/312 [05:39<25:52,  6.06s/it]
Training Epoch: 17/100, batch 112/625 completed (loss: 0.2183:  18%|█▊        | 56/312 [05:39<25:51,  6.06s/it]
Training Epoch: 17/100, batch 112/625 completed (loss: 0.2011:  18%|█▊        | 56/312 [05:39<25:51,  6.06s/it]
Training Epoch: 17/100, batch 112/625 completed (loss: 0.0463:  18%|█▊        | 56/312 [05:39<25:51,  6.06s/it]
Training Epoch: 17/100, batch 112/625 completed (loss: 0.0463:  18%|█▊        | 57/312 [05:42<25:49,  6.08s/it]
Training Epoch: 17/100, batch 112/625 completed (loss: 0.2183:  18%|█▊        | 57/312 [05:42<25:49,  6.08s/it]
Training Epoch: 17/100, batch 112/625 completed (loss: 0.2011:  18%|█▊        | 57/312 [05:42<25:49,  6.08s/it]
Training Epoch: 17/100, batch 112/625 completed (loss: 0.1204:  18%|█▊        | 57/312 [05:42<25:50,  6.08s/it]
Training Epoch: 17/100, batch 113/625 completed (loss: 0.2991:  18%|█▊        | 57/312 [05:42<25:49,  6.08s/it]
Training Epoch: 17/100, batch 113/625 completed (loss: 0.0034:  18%|█▊        | 57/312 [05:42<25:50,  6.08s/it]
Training Epoch: 17/100, batch 113/625 completed (loss: 0.0014:  18%|█▊        | 57/312 [05:42<25:49,  6.08s/it]
Training Epoch: 17/100, batch 113/625 completed (loss: 0.0419:  18%|█▊        | 57/312 [05:42<25:49,  6.08s/it]
Training Epoch: 17/100, batch 114/625 completed (loss: 0.0081:  18%|█▊        | 57/312 [05:46<25:50,  6.08s/it]
Training Epoch: 17/100, batch 114/625 completed (loss: 0.2069:  18%|█▊        | 57/312 [05:46<25:49,  6.08s/it]
Training Epoch: 17/100, batch 114/625 completed (loss: 0.3803:  18%|█▊        | 57/312 [05:46<25:49,  6.08s/it]
Training Epoch: 17/100, batch 114/625 completed (loss: 0.3511:  18%|█▊        | 57/312 [05:46<25:49,  6.08s/it]
Training Epoch: 17/100, batch 114/625 completed (loss: 0.0081:  19%|█▊        | 58/312 [05:48<25:55,  6.12s/it]
Training Epoch: 17/100, batch 114/625 completed (loss: 0.2069:  19%|█▊        | 58/312 [05:48<25:55,  6.12s/it]
Training Epoch: 17/100, batch 114/625 completed (loss: 0.3511:  19%|█▊        | 58/312 [05:48<25:56,  6.13s/it]
Training Epoch: 17/100, batch 114/625 completed (loss: 0.3803:  19%|█▊        | 58/312 [05:48<25:56,  6.13s/it]
Training Epoch: 17/100, batch 115/625 completed (loss: 0.2058:  19%|█▊        | 58/312 [05:48<25:56,  6.13s/it]
Training Epoch: 17/100, batch 115/625 completed (loss: 0.0072:  19%|█▊        | 58/312 [05:48<25:55,  6.12s/it]
Training Epoch: 17/100, batch 115/625 completed (loss: 0.3151:  19%|█▊        | 58/312 [05:48<25:55,  6.12s/it]
Training Epoch: 17/100, batch 115/625 completed (loss: 0.2785:  19%|█▊        | 58/312 [05:49<25:56,  6.13s/it]
Training Epoch: 17/100, batch 116/625 completed (loss: 0.0218:  19%|█▊        | 58/312 [05:51<25:55,  6.12s/it]
Training Epoch: 17/100, batch 116/625 completed (loss: 0.0764:  19%|█▊        | 58/312 [05:51<25:56,  6.13s/it]
Training Epoch: 17/100, batch 116/625 completed (loss: 0.0046:  19%|█▊        | 58/312 [05:51<25:55,  6.12s/it]
Training Epoch: 17/100, batch 116/625 completed (loss: 0.0666:  19%|█▊        | 58/312 [05:51<25:56,  6.13s/it]
Training Epoch: 17/100, batch 116/625 completed (loss: 0.0666:  19%|█▉        | 59/312 [05:54<25:41,  6.09s/it]
Training Epoch: 17/100, batch 116/625 completed (loss: 0.0218:  19%|█▉        | 59/312 [05:54<25:43,  6.10s/it]
Training Epoch: 17/100, batch 116/625 completed (loss: 0.0764:  19%|█▉        | 59/312 [05:54<25:42,  6.10s/it]
Training Epoch: 17/100, batch 116/625 completed (loss: 0.0046:  19%|█▉        | 59/312 [05:54<25:42,  6.10s/it]
Training Epoch: 17/100, batch 117/625 completed (loss: 0.0504:  19%|█▉        | 59/312 [05:54<25:43,  6.10s/it]
Training Epoch: 17/100, batch 117/625 completed (loss: 0.0358:  19%|█▉        | 59/312 [05:54<25:42,  6.10s/it]
Training Epoch: 17/100, batch 117/625 completed (loss: 0.0402:  19%|█▉        | 59/312 [05:54<25:42,  6.10s/it]
Training Epoch: 17/100, batch 117/625 completed (loss: 0.3702:  19%|█▉        | 59/312 [05:55<25:41,  6.09s/it]
Training Epoch: 17/100, batch 118/625 completed (loss: 0.2674:  19%|█▉        | 59/312 [05:58<25:42,  6.10s/it]
Training Epoch: 17/100, batch 118/625 completed (loss: 0.0066:  19%|█▉        | 59/312 [05:58<25:43,  6.10s/it]
Training Epoch: 17/100, batch 118/625 completed (loss: 0.0089:  19%|█▉        | 59/312 [05:58<25:42,  6.10s/it]
Training Epoch: 17/100, batch 118/625 completed (loss: 0.1478:  19%|█▉        | 59/312 [05:58<25:41,  6.09s/it]
Training Epoch: 17/100, batch 118/625 completed (loss: 0.1478:  19%|█▉        | 60/312 [06:01<25:40,  6.11s/it]
Training Epoch: 17/100, batch 118/625 completed (loss: 0.0066:  19%|█▉        | 60/312 [06:01<25:40,  6.11s/it]
Training Epoch: 17/100, batch 118/625 completed (loss: 0.2674:  19%|█▉        | 60/312 [06:01<25:40,  6.11s/it]
Training Epoch: 17/100, batch 118/625 completed (loss: 0.0089:  19%|█▉        | 60/312 [06:01<25:40,  6.11s/it]
Training Epoch: 17/100, batch 119/625 completed (loss: 0.0886:  19%|█▉        | 60/312 [06:01<25:40,  6.11s/it]
Training Epoch: 17/100, batch 119/625 completed (loss: 0.0738:  19%|█▉        | 60/312 [06:01<25:40,  6.11s/it]
Training Epoch: 17/100, batch 119/625 completed (loss: 0.1244:  19%|█▉        | 60/312 [06:01<25:40,  6.11s/it]
Training Epoch: 17/100, batch 119/625 completed (loss: 0.0948:  19%|█▉        | 60/312 [06:01<25:40,  6.11s/it]
Training Epoch: 17/100, batch 120/625 completed (loss: 0.0154:  19%|█▉        | 60/312 [06:04<25:40,  6.11s/it]
Training Epoch: 17/100, batch 120/625 completed (loss: 0.0929:  19%|█▉        | 60/312 [06:04<25:40,  6.11s/it]
Training Epoch: 17/100, batch 120/625 completed (loss: 0.7702:  19%|█▉        | 60/312 [06:04<25:40,  6.11s/it]
Training Epoch: 17/100, batch 120/625 completed (loss: 0.0446:  19%|█▉        | 60/312 [06:04<25:40,  6.11s/it]
Training Epoch: 17/100, batch 120/625 completed (loss: 0.0446:  20%|█▉        | 61/312 [06:07<25:27,  6.09s/it]
Training Epoch: 17/100, batch 120/625 completed (loss: 0.7702:  20%|█▉        | 61/312 [06:07<25:28,  6.09s/it]
Training Epoch: 17/100, batch 120/625 completed (loss: 0.0154:  20%|█▉        | 61/312 [06:07<25:29,  6.09s/it]
Training Epoch: 17/100, batch 120/625 completed (loss: 0.0929:  20%|█▉        | 61/312 [06:07<25:29,  6.09s/it]
Training Epoch: 17/100, batch 121/625 completed (loss: 0.0203:  20%|█▉        | 61/312 [06:07<25:29,  6.09s/it]
Training Epoch: 17/100, batch 121/625 completed (loss: 0.0478:  20%|█▉        | 61/312 [06:07<25:29,  6.09s/it]
Training Epoch: 17/100, batch 121/625 completed (loss: 0.1486:  20%|█▉        | 61/312 [06:07<25:28,  6.09s/it]
Training Epoch: 17/100, batch 121/625 completed (loss: 0.0805:  20%|█▉        | 61/312 [06:07<25:27,  6.09s/it]
Training Epoch: 17/100, batch 122/625 completed (loss: 0.0321:  20%|█▉        | 61/312 [06:10<25:28,  6.09s/it]
Training Epoch: 17/100, batch 122/625 completed (loss: 0.0674:  20%|█▉        | 61/312 [06:10<25:29,  6.09s/it]
Training Epoch: 17/100, batch 122/625 completed (loss: 0.1004:  20%|█▉        | 61/312 [06:10<25:29,  6.09s/it]
Training Epoch: 17/100, batch 122/625 completed (loss: 0.0059:  20%|█▉        | 61/312 [06:10<25:27,  6.09s/it]
Training Epoch: 17/100, batch 122/625 completed (loss: 0.0059:  20%|█▉        | 62/312 [06:13<25:27,  6.11s/it]
Training Epoch: 17/100, batch 122/625 completed (loss: 0.0674:  20%|█▉        | 62/312 [06:13<25:26,  6.11s/it]
Training Epoch: 17/100, batch 122/625 completed (loss: 0.0321:  20%|█▉        | 62/312 [06:13<25:27,  6.11s/it]
Training Epoch: 17/100, batch 122/625 completed (loss: 0.1004:  20%|█▉        | 62/312 [06:13<25:28,  6.11s/it]
Training Epoch: 17/100, batch 123/625 completed (loss: 0.0052:  20%|█▉        | 62/312 [06:13<25:28,  6.11s/it]
Training Epoch: 17/100, batch 123/625 completed (loss: 0.0111:  20%|█▉        | 62/312 [06:13<25:27,  6.11s/it]
Training Epoch: 17/100, batch 123/625 completed (loss: 0.0325:  20%|█▉        | 62/312 [06:13<25:26,  6.11s/it]
Training Epoch: 17/100, batch 123/625 completed (loss: 0.2982:  20%|█▉        | 62/312 [06:13<25:27,  6.11s/it]
Training Epoch: 17/100, batch 124/625 completed (loss: 0.0044:  20%|█▉        | 62/312 [06:16<25:27,  6.11s/it]
Training Epoch: 17/100, batch 124/625 completed (loss: 0.2495:  20%|█▉        | 62/312 [06:16<25:26,  6.11s/it]
Training Epoch: 17/100, batch 124/625 completed (loss: 0.052:  20%|█▉        | 62/312 [06:16<25:28,  6.11s/it] 
Training Epoch: 17/100, batch 124/625 completed (loss: 0.1289:  20%|█▉        | 62/312 [06:16<25:27,  6.11s/it]
Training Epoch: 17/100, batch 124/625 completed (loss: 0.2495:  20%|██        | 63/312 [06:19<25:25,  6.13s/it]
Training Epoch: 17/100, batch 124/625 completed (loss: 0.052:  20%|██        | 63/312 [06:19<25:25,  6.13s/it]
Training Epoch: 17/100, batch 124/625 completed (loss: 0.1289:  20%|██        | 63/312 [06:19<25:26,  6.13s/it]
Training Epoch: 17/100, batch 124/625 completed (loss: 0.0044:  20%|██        | 63/312 [06:19<25:26,  6.13s/it]
Training Epoch: 17/100, batch 125/625 completed (loss: 0.0358:  20%|██        | 63/312 [06:19<25:25,  6.13s/it]
Training Epoch: 17/100, batch 125/625 completed (loss: 0.0445:  20%|██        | 63/312 [06:19<25:25,  6.13s/it]
Training Epoch: 17/100, batch 125/625 completed (loss: 0.0336:  20%|██        | 63/312 [06:19<25:26,  6.13s/it]
Training Epoch: 17/100, batch 125/625 completed (loss: 0.0626:  20%|██        | 63/312 [06:19<25:26,  6.13s/it]
Training Epoch: 17/100, batch 126/625 completed (loss: 0.0095:  20%|██        | 63/312 [06:22<25:26,  6.13s/it]
Training Epoch: 17/100, batch 126/625 completed (loss: 0.3034:  20%|██        | 63/312 [06:22<25:25,  6.13s/it]
Training Epoch: 17/100, batch 126/625 completed (loss: 0.005:  20%|██        | 63/312 [06:22<25:25,  6.13s/it] 
Training Epoch: 17/100, batch 126/625 completed (loss: 0.0032:  20%|██        | 63/312 [06:22<25:26,  6.13s/it]
Training Epoch: 17/100, batch 126/625 completed (loss: 0.005:  21%|██        | 64/312 [06:25<24:58,  6.04s/it]
Training Epoch: 17/100, batch 126/625 completed (loss: 0.0032:  21%|██        | 64/312 [06:25<24:59,  6.05s/it]
Training Epoch: 17/100, batch 126/625 completed (loss: 0.3034:  21%|██        | 64/312 [06:25<25:00,  6.05s/it]
Training Epoch: 17/100, batch 126/625 completed (loss: 0.0095:  21%|██        | 64/312 [06:25<25:00,  6.05s/it]
Training Epoch: 17/100, batch 127/625 completed (loss: 0.0129:  21%|██        | 64/312 [06:25<25:00,  6.05s/it]
Training Epoch: 17/100, batch 127/625 completed (loss: 0.3099:  21%|██        | 64/312 [06:25<24:58,  6.04s/it]
Training Epoch: 17/100, batch 127/625 completed (loss: 0.0532:  21%|██        | 64/312 [06:25<25:00,  6.05s/it]
Training Epoch: 17/100, batch 127/625 completed (loss: 0.0923:  21%|██        | 64/312 [06:25<24:59,  6.05s/it]
Training Epoch: 17/100, batch 128/625 completed (loss: 0.1502:  21%|██        | 64/312 [06:28<25:00,  6.05s/it]
Training Epoch: 17/100, batch 128/625 completed (loss: 0.0277:  21%|██        | 64/312 [06:28<25:00,  6.05s/it]
Training Epoch: 17/100, batch 128/625 completed (loss: 0.0079:  21%|██        | 64/312 [06:28<24:58,  6.04s/it]
Training Epoch: 17/100, batch 128/625 completed (loss: 0.2188:  21%|██        | 64/312 [06:28<24:59,  6.05s/it]
Training Epoch: 17/100, batch 128/625 completed (loss: 0.2188:  21%|██        | 65/312 [06:31<25:13,  6.13s/it]
Training Epoch: 17/100, batch 128/625 completed (loss: 0.0079:  21%|██        | 65/312 [06:31<25:13,  6.13s/it]
Training Epoch: 17/100, batch 128/625 completed (loss: 0.0277:  21%|██        | 65/312 [06:31<25:13,  6.13s/it]
Training Epoch: 17/100, batch 128/625 completed (loss: 0.1502:  21%|██        | 65/312 [06:31<25:13,  6.13s/it]
Training Epoch: 17/100, batch 129/625 completed (loss: 0.0679:  21%|██        | 65/312 [06:31<25:13,  6.13s/it]
Training Epoch: 17/100, batch 129/625 completed (loss: 0.0181:  21%|██        | 65/312 [06:31<25:13,  6.13s/it]
Training Epoch: 17/100, batch 129/625 completed (loss: 0.0064:  21%|██        | 65/312 [06:31<25:13,  6.13s/it]
Training Epoch: 17/100, batch 129/625 completed (loss: 0.1033:  21%|██        | 65/312 [06:31<25:13,  6.13s/it]
Training Epoch: 17/100, batch 130/625 completed (loss: 0.0901:  21%|██        | 65/312 [06:34<25:13,  6.13s/it]
Training Epoch: 17/100, batch 130/625 completed (loss: 0.0185:  21%|██        | 65/312 [06:34<25:13,  6.13s/it]
Training Epoch: 17/100, batch 130/625 completed (loss: 0.025:  21%|██        | 65/312 [06:34<25:13,  6.13s/it] 
Training Epoch: 17/100, batch 130/625 completed (loss: 0.0449:  21%|██        | 65/312 [06:34<25:13,  6.13s/it]
Training Epoch: 17/100, batch 130/625 completed (loss: 0.0901:  21%|██        | 66/312 [06:37<24:56,  6.09s/it]
Training Epoch: 17/100, batch 130/625 completed (loss: 0.025:  21%|██        | 66/312 [06:37<24:56,  6.08s/it]
Training Epoch: 17/100, batch 130/625 completed (loss: 0.0449:  21%|██        | 66/312 [06:37<24:57,  6.09s/it]
Training Epoch: 17/100, batch 130/625 completed (loss: 0.0185:  21%|██        | 66/312 [06:37<24:57,  6.09s/it]
Training Epoch: 17/100, batch 131/625 completed (loss: 0.0148:  21%|██        | 66/312 [06:37<24:57,  6.09s/it]
Training Epoch: 17/100, batch 131/625 completed (loss: 0.0391:  21%|██        | 66/312 [06:37<24:56,  6.09s/it]
Training Epoch: 17/100, batch 131/625 completed (loss: 0.0064:  21%|██        | 66/312 [06:37<24:56,  6.08s/it]
Training Epoch: 17/100, batch 131/625 completed (loss: 0.0285:  21%|██        | 66/312 [06:37<24:57,  6.09s/it]
Training Epoch: 17/100, batch 132/625 completed (loss: 0.0866:  21%|██        | 66/312 [06:40<24:56,  6.09s/it]
Training Epoch: 17/100, batch 132/625 completed (loss: 0.1003:  21%|██        | 66/312 [06:40<24:57,  6.09s/it]
Training Epoch: 17/100, batch 132/625 completed (loss: 0.012:  21%|██        | 66/312 [06:40<24:56,  6.08s/it] 
Training Epoch: 17/100, batch 132/625 completed (loss: 0.0327:  21%|██        | 66/312 [06:40<24:57,  6.09s/it]
Training Epoch: 17/100, batch 132/625 completed (loss: 0.012:  21%|██▏       | 67/312 [06:43<24:44,  6.06s/it]
Training Epoch: 17/100, batch 132/625 completed (loss: 0.0327:  21%|██▏       | 67/312 [06:43<24:44,  6.06s/it]
Training Epoch: 17/100, batch 132/625 completed (loss: 0.1003:  21%|██▏       | 67/312 [06:43<24:44,  6.06s/it]
Training Epoch: 17/100, batch 132/625 completed (loss: 0.0866:  21%|██▏       | 67/312 [06:43<24:45,  6.06s/it]
Training Epoch: 17/100, batch 133/625 completed (loss: 0.0606:  21%|██▏       | 67/312 [06:43<24:44,  6.06s/it]
Training Epoch: 17/100, batch 133/625 completed (loss: 0.0083:  21%|██▏       | 67/312 [06:43<24:45,  6.06s/it]
Training Epoch: 17/100, batch 133/625 completed (loss: 0.0133:  21%|██▏       | 67/312 [06:43<24:44,  6.06s/it]
Training Epoch: 17/100, batch 133/625 completed (loss: 0.0053:  21%|██▏       | 67/312 [06:43<24:44,  6.06s/it]
Training Epoch: 17/100, batch 134/625 completed (loss: 0.0264:  21%|██▏       | 67/312 [06:46<24:44,  6.06s/it]
Training Epoch: 17/100, batch 134/625 completed (loss: 0.1043:  21%|██▏       | 67/312 [06:46<24:45,  6.06s/it]
Training Epoch: 17/100, batch 134/625 completed (loss: 0.0362:  21%|██▏       | 67/312 [06:46<24:44,  6.06s/it]
Training Epoch: 17/100, batch 134/625 completed (loss: 0.0209:  21%|██▏       | 67/312 [06:46<24:44,  6.06s/it]
Training Epoch: 17/100, batch 134/625 completed (loss: 0.0209:  22%|██▏       | 68/312 [06:49<24:44,  6.09s/it]
Training Epoch: 17/100, batch 134/625 completed (loss: 0.1043:  22%|██▏       | 68/312 [06:49<24:44,  6.09s/it]
Training Epoch: 17/100, batch 134/625 completed (loss: 0.0362:  22%|██▏       | 68/312 [06:49<24:46,  6.09s/it]
Training Epoch: 17/100, batch 134/625 completed (loss: 0.0264:  22%|██▏       | 68/312 [06:49<24:46,  6.09s/it]
Training Epoch: 17/100, batch 135/625 completed (loss: 0.0083:  22%|██▏       | 68/312 [06:49<24:44,  6.09s/it]
Training Epoch: 17/100, batch 135/625 completed (loss: 0.1384:  22%|██▏       | 68/312 [06:49<24:46,  6.09s/it]
Training Epoch: 17/100, batch 135/625 completed (loss: 0.0749:  22%|██▏       | 68/312 [06:49<24:46,  6.09s/it]
Training Epoch: 17/100, batch 135/625 completed (loss: 0.2682:  22%|██▏       | 68/312 [06:49<24:44,  6.09s/it]
Training Epoch: 17/100, batch 136/625 completed (loss: 0.0238:  22%|██▏       | 68/312 [06:52<24:44,  6.09s/it]
Training Epoch: 17/100, batch 136/625 completed (loss: 0.0027:  22%|██▏       | 68/312 [06:52<24:46,  6.09s/it]
Training Epoch: 17/100, batch 136/625 completed (loss: 0.0196:  22%|██▏       | 68/312 [06:52<24:46,  6.09s/it]
Training Epoch: 17/100, batch 136/625 completed (loss: 0.4973:  22%|██▏       | 68/312 [06:52<24:44,  6.09s/it]
Training Epoch: 17/100, batch 136/625 completed (loss: 0.0238:  22%|██▏       | 69/312 [06:55<24:35,  6.07s/it]
Training Epoch: 17/100, batch 136/625 completed (loss: 0.0027:  22%|██▏       | 69/312 [06:55<24:35,  6.07s/it]
Training Epoch: 17/100, batch 136/625 completed (loss: 0.0196:  22%|██▏       | 69/312 [06:55<24:34,  6.07s/it]
Training Epoch: 17/100, batch 136/625 completed (loss: 0.4973:  22%|██▏       | 69/312 [06:55<24:36,  6.07s/it]
Training Epoch: 17/100, batch 137/625 completed (loss: 0.0329:  22%|██▏       | 69/312 [06:55<24:34,  6.07s/it]
Training Epoch: 17/100, batch 137/625 completed (loss: 0.0439:  22%|██▏       | 69/312 [06:55<24:35,  6.07s/it]
Training Epoch: 17/100, batch 137/625 completed (loss: 0.0284:  22%|██▏       | 69/312 [06:55<24:35,  6.07s/it]
Training Epoch: 17/100, batch 137/625 completed (loss: 0.0078:  22%|██▏       | 69/312 [06:55<24:36,  6.07s/it]
Training Epoch: 17/100, batch 138/625 completed (loss: 0.0361:  22%|██▏       | 69/312 [06:58<24:35,  6.07s/it]
Training Epoch: 17/100, batch 138/625 completed (loss: 0.0723:  22%|██▏       | 69/312 [06:58<24:35,  6.07s/it]
Training Epoch: 17/100, batch 138/625 completed (loss: 0.1091:  22%|██▏       | 69/312 [06:58<24:34,  6.07s/it]
Training Epoch: 17/100, batch 138/625 completed (loss: 0.0329:  22%|██▏       | 69/312 [06:58<24:36,  6.07s/it]
Training Epoch: 17/100, batch 138/625 completed (loss: 0.0329:  22%|██▏       | 70/312 [07:02<24:35,  6.10s/it]
Training Epoch: 17/100, batch 138/625 completed (loss: 0.1091:  22%|██▏       | 70/312 [07:01<24:35,  6.10s/it]
Training Epoch: 17/100, batch 138/625 completed (loss: 0.0361:  22%|██▏       | 70/312 [07:01<24:35,  6.10s/it]
Training Epoch: 17/100, batch 138/625 completed (loss: 0.0723:  22%|██▏       | 70/312 [07:01<24:36,  6.10s/it]
Training Epoch: 17/100, batch 139/625 completed (loss: 0.0308:  22%|██▏       | 70/312 [07:02<24:35,  6.10s/it]
Training Epoch: 17/100, batch 139/625 completed (loss: 0.0084:  22%|██▏       | 70/312 [07:02<24:35,  6.10s/it]
Training Epoch: 17/100, batch 139/625 completed (loss: 0.005:  22%|██▏       | 70/312 [07:02<24:36,  6.10s/it] 
Training Epoch: 17/100, batch 139/625 completed (loss: 0.0213:  22%|██▏       | 70/312 [07:02<24:35,  6.10s/it]
Training Epoch: 17/100, batch 140/625 completed (loss: 0.0762:  22%|██▏       | 70/312 [07:05<24:35,  6.10s/it]
Training Epoch: 17/100, batch 140/625 completed (loss: 0.0373:  22%|██▏       | 70/312 [07:05<24:35,  6.10s/it]
Training Epoch: 17/100, batch 140/625 completed (loss: 0.0341:  22%|██▏       | 70/312 [07:05<24:36,  6.10s/it]
Training Epoch: 17/100, batch 140/625 completed (loss: 0.1707:  22%|██▏       | 70/312 [07:05<24:35,  6.10s/it]
Training Epoch: 17/100, batch 140/625 completed (loss: 0.0373:  23%|██▎       | 71/312 [07:08<24:34,  6.12s/it]
Training Epoch: 17/100, batch 140/625 completed (loss: 0.1707:  23%|██▎       | 71/312 [07:08<24:34,  6.12s/it]
Training Epoch: 17/100, batch 140/625 completed (loss: 0.0762:  23%|██▎       | 71/312 [07:08<24:34,  6.12s/it]
Training Epoch: 17/100, batch 140/625 completed (loss: 0.0341:  23%|██▎       | 71/312 [07:08<24:35,  6.12s/it]
Training Epoch: 17/100, batch 141/625 completed (loss: 0.05:  23%|██▎       | 71/312 [07:08<24:34,  6.12s/it]  
Training Epoch: 17/100, batch 141/625 completed (loss: 0.0622:  23%|██▎       | 71/312 [07:08<24:35,  6.12s/it]
Training Epoch: 17/100, batch 141/625 completed (loss: 0.0219:  23%|██▎       | 71/312 [07:08<24:34,  6.12s/it]
Training Epoch: 17/100, batch 141/625 completed (loss: 0.0281:  23%|██▎       | 71/312 [07:08<24:34,  6.12s/it]
Training Epoch: 17/100, batch 142/625 completed (loss: 0.0214:  23%|██▎       | 71/312 [07:11<24:34,  6.12s/it]
Training Epoch: 17/100, batch 142/625 completed (loss: 0.0077:  23%|██▎       | 71/312 [07:11<24:35,  6.12s/it]
Training Epoch: 17/100, batch 142/625 completed (loss: 0.007:  23%|██▎       | 71/312 [07:11<24:34,  6.12s/it]
Training Epoch: 17/100, batch 142/625 completed (loss: 0.3636:  23%|██▎       | 71/312 [07:11<24:34,  6.12s/it]
Training Epoch: 17/100, batch 142/625 completed (loss: 0.007:  23%|██▎       | 72/312 [07:13<24:08,  6.04s/it]
Training Epoch: 17/100, batch 142/625 completed (loss: 0.0214:  23%|██▎       | 72/312 [07:13<24:08,  6.04s/it]
Training Epoch: 17/100, batch 142/625 completed (loss: 0.3636:  23%|██▎       | 72/312 [07:14<24:09,  6.04s/it]
Training Epoch: 17/100, batch 142/625 completed (loss: 0.0077:  23%|██▎       | 72/312 [07:13<24:08,  6.04s/it]
Training Epoch: 17/100, batch 143/625 completed (loss: 0.0681:  23%|██▎       | 72/312 [07:14<24:08,  6.04s/it]
Training Epoch: 17/100, batch 143/625 completed (loss: 0.1179:  23%|██▎       | 72/312 [07:14<24:08,  6.04s/it]
Training Epoch: 17/100, batch 143/625 completed (loss: 0.0065:  23%|██▎       | 72/312 [07:14<24:08,  6.04s/it]
Training Epoch: 17/100, batch 143/625 completed (loss: 0.1508:  23%|██▎       | 72/312 [07:14<24:09,  6.04s/it]
Training Epoch: 17/100, batch 144/625 completed (loss: 0.0273:  23%|██▎       | 72/312 [07:17<24:08,  6.04s/it]
Training Epoch: 17/100, batch 144/625 completed (loss: 0.0193:  23%|██▎       | 72/312 [07:17<24:08,  6.04s/it]
Training Epoch: 17/100, batch 144/625 completed (loss: 0.0179:  23%|██▎       | 72/312 [07:17<24:08,  6.04s/it]
Training Epoch: 17/100, batch 144/625 completed (loss: 0.0977:  23%|██▎       | 72/312 [07:17<24:09,  6.04s/it]
Training Epoch: 17/100, batch 144/625 completed (loss: 0.0179:  23%|██▎       | 73/312 [07:20<24:20,  6.11s/it]
Training Epoch: 17/100, batch 144/625 completed (loss: 0.0273:  23%|██▎       | 73/312 [07:20<24:20,  6.11s/it]
Training Epoch: 17/100, batch 144/625 completed (loss: 0.0193:  23%|██▎       | 73/312 [07:20<24:20,  6.11s/it]
Training Epoch: 17/100, batch 144/625 completed (loss: 0.0977:  23%|██▎       | 73/312 [07:20<24:20,  6.11s/it]
Training Epoch: 17/100, batch 145/625 completed (loss: 0.0106:  23%|██▎       | 73/312 [07:20<24:20,  6.11s/it]
Training Epoch: 17/100, batch 145/625 completed (loss: 0.0204:  23%|██▎       | 73/312 [07:20<24:20,  6.11s/it]
Training Epoch: 17/100, batch 145/625 completed (loss: 0.0314:  23%|██▎       | 73/312 [07:20<24:20,  6.11s/it]
Training Epoch: 17/100, batch 145/625 completed (loss: 0.0066:  23%|██▎       | 73/312 [07:20<24:20,  6.11s/it]
Training Epoch: 17/100, batch 146/625 completed (loss: 0.0212:  23%|██▎       | 73/312 [07:23<24:20,  6.11s/it]
Training Epoch: 17/100, batch 146/625 completed (loss: 0.0018:  23%|██▎       | 73/312 [07:23<24:20,  6.11s/it]
Training Epoch: 17/100, batch 146/625 completed (loss: 0.3267:  23%|██▎       | 73/312 [07:23<24:20,  6.11s/it]
Training Epoch: 17/100, batch 146/625 completed (loss: 0.0323:  23%|██▎       | 73/312 [07:23<24:20,  6.11s/it]
Training Epoch: 17/100, batch 146/625 completed (loss: 0.3267:  24%|██▎       | 74/312 [07:26<24:05,  6.07s/it]
Training Epoch: 17/100, batch 146/625 completed (loss: 0.0212:  24%|██▎       | 74/312 [07:26<24:05,  6.07s/it]
Training Epoch: 17/100, batch 146/625 completed (loss: 0.0323:  24%|██▎       | 74/312 [07:26<24:05,  6.08s/it]
Training Epoch: 17/100, batch 146/625 completed (loss: 0.0018:  24%|██▎       | 74/312 [07:26<24:05,  6.08s/it]
Training Epoch: 17/100, batch 147/625 completed (loss: 0.0523:  24%|██▎       | 74/312 [07:26<24:05,  6.07s/it]
Training Epoch: 17/100, batch 147/625 completed (loss: 0.1516:  24%|██▎       | 74/312 [07:26<24:05,  6.07s/it]
Training Epoch: 17/100, batch 147/625 completed (loss: 0.0046:  24%|██▎       | 74/312 [07:26<24:05,  6.08s/it]
Training Epoch: 17/100, batch 147/625 completed (loss: 0.0125:  24%|██▎       | 74/312 [07:26<24:05,  6.08s/it]
Training Epoch: 17/100, batch 148/625 completed (loss: 0.1224:  24%|██▎       | 74/312 [07:29<24:05,  6.07s/it]
Training Epoch: 17/100, batch 148/625 completed (loss: 0.0139:  24%|██▎       | 74/312 [07:29<24:05,  6.08s/it]
Training Epoch: 17/100, batch 148/625 completed (loss: 0.0106:  24%|██▎       | 74/312 [07:29<24:05,  6.07s/it]
Training Epoch: 17/100, batch 148/625 completed (loss: 0.1131:  24%|██▎       | 74/312 [07:29<24:05,  6.08s/it]
Training Epoch: 17/100, batch 148/625 completed (loss: 0.0106:  24%|██▍       | 75/312 [07:32<24:02,  6.09s/it]
Training Epoch: 17/100, batch 148/625 completed (loss: 0.1131:  24%|██▍       | 75/312 [07:32<24:02,  6.09s/it]
Training Epoch: 17/100, batch 148/625 completed (loss: 0.1224:  24%|██▍       | 75/312 [07:32<24:03,  6.09s/it]
Training Epoch: 17/100, batch 148/625 completed (loss: 0.0139:  24%|██▍       | 75/312 [07:32<24:03,  6.09s/it]
Training Epoch: 17/100, batch 149/625 completed (loss: 0.0055:  24%|██▍       | 75/312 [07:32<24:03,  6.09s/it]
Training Epoch: 17/100, batch 149/625 completed (loss: 0.0244:  24%|██▍       | 75/312 [07:32<24:03,  6.09s/it]
Training Epoch: 17/100, batch 149/625 completed (loss: 0.0282:  24%|██▍       | 75/312 [07:32<24:02,  6.09s/it]
Training Epoch: 17/100, batch 149/625 completed (loss: 0.0263:  24%|██▍       | 75/312 [07:32<24:02,  6.09s/it]
Training Epoch: 17/100, batch 150/625 completed (loss: 0.2097:  24%|██▍       | 75/312 [07:35<24:03,  6.09s/it]
Training Epoch: 17/100, batch 150/625 completed (loss: 0.0271:  24%|██▍       | 75/312 [07:35<24:02,  6.09s/it]
Training Epoch: 17/100, batch 150/625 completed (loss: 0.0711:  24%|██▍       | 75/312 [07:35<24:03,  6.09s/it]
Training Epoch: 17/100, batch 150/625 completed (loss: 0.0279:  24%|██▍       | 75/312 [07:35<24:02,  6.09s/it]
Training Epoch: 17/100, batch 150/625 completed (loss: 0.2097:  24%|██▍       | 76/312 [07:38<23:50,  6.06s/it]
Training Epoch: 17/100, batch 150/625 completed (loss: 0.0279:  24%|██▍       | 76/312 [07:38<23:50,  6.06s/it]
Training Epoch: 17/100, batch 150/625 completed (loss: 0.0271:  24%|██▍       | 76/312 [07:38<23:50,  6.06s/it]
Training Epoch: 17/100, batch 150/625 completed (loss: 0.0711:  24%|██▍       | 76/312 [07:38<23:50,  6.06s/it]
Training Epoch: 17/100, batch 151/625 completed (loss: 0.2286:  24%|██▍       | 76/312 [07:38<23:50,  6.06s/it]
Training Epoch: 17/100, batch 151/625 completed (loss: 0.1191:  24%|██▍       | 76/312 [07:38<23:50,  6.06s/it]
Training Epoch: 17/100, batch 151/625 completed (loss: 0.0117:  24%|██▍       | 76/312 [07:38<23:50,  6.06s/it]
Training Epoch: 17/100, batch 151/625 completed (loss: 0.0159:  24%|██▍       | 76/312 [07:38<23:50,  6.06s/it]
Training Epoch: 17/100, batch 152/625 completed (loss: 0.017:  24%|██▍       | 76/312 [07:41<23:50,  6.06s/it] 
Training Epoch: 17/100, batch 152/625 completed (loss: 0.0039:  24%|██▍       | 76/312 [07:41<23:50,  6.06s/it]
Training Epoch: 17/100, batch 152/625 completed (loss: 0.009:  24%|██▍       | 76/312 [07:41<23:50,  6.06s/it] 
Training Epoch: 17/100, batch 152/625 completed (loss: 0.0035:  24%|██▍       | 76/312 [07:41<23:50,  6.06s/it]
Training Epoch: 17/100, batch 152/625 completed (loss: 0.0039:  25%|██▍       | 77/312 [07:44<23:39,  6.04s/it]
Training Epoch: 17/100, batch 152/625 completed (loss: 0.0035:  25%|██▍       | 77/312 [07:44<23:39,  6.04s/it]
Training Epoch: 17/100, batch 152/625 completed (loss: 0.009:  25%|██▍       | 77/312 [07:44<23:39,  6.04s/it]
Training Epoch: 17/100, batch 152/625 completed (loss: 0.017:  25%|██▍       | 77/312 [07:44<23:40,  6.04s/it]
Training Epoch: 17/100, batch 153/625 completed (loss: 0.0464:  25%|██▍       | 77/312 [07:44<23:39,  6.04s/it]
Training Epoch: 17/100, batch 153/625 completed (loss: 0.1145:  25%|██▍       | 77/312 [07:44<23:39,  6.04s/it]
Training Epoch: 17/100, batch 153/625 completed (loss: 0.0259:  25%|██▍       | 77/312 [07:44<23:40,  6.04s/it]
Training Epoch: 17/100, batch 153/625 completed (loss: 0.0238:  25%|██▍       | 77/312 [07:44<23:39,  6.04s/it]
Training Epoch: 17/100, batch 154/625 completed (loss: 0.0137:  25%|██▍       | 77/312 [07:47<23:40,  6.04s/it]
Training Epoch: 17/100, batch 154/625 completed (loss: 0.0063:  25%|██▍       | 77/312 [07:47<23:39,  6.04s/it]
Training Epoch: 17/100, batch 154/625 completed (loss: 0.0071:  25%|██▍       | 77/312 [07:47<23:39,  6.04s/it]
Training Epoch: 17/100, batch 154/625 completed (loss: 0.1001:  25%|██▍       | 77/312 [07:47<23:39,  6.04s/it]
Training Epoch: 17/100, batch 154/625 completed (loss: 0.1001:  25%|██▌       | 78/312 [07:50<23:40,  6.07s/it]
Training Epoch: 17/100, batch 154/625 completed (loss: 0.0063:  25%|██▌       | 78/312 [07:50<23:41,  6.07s/it]
Training Epoch: 17/100, batch 154/625 completed (loss: 0.0071:  25%|██▌       | 78/312 [07:50<23:41,  6.07s/it]
Training Epoch: 17/100, batch 154/625 completed (loss: 0.0137:  25%|██▌       | 78/312 [07:50<23:41,  6.08s/it]
Training Epoch: 17/100, batch 155/625 completed (loss: 0.036:  25%|██▌       | 78/312 [07:50<23:41,  6.08s/it] 
Training Epoch: 17/100, batch 155/625 completed (loss: 0.1006:  25%|██▌       | 78/312 [07:50<23:41,  6.07s/it]
Training Epoch: 17/100, batch 155/625 completed (loss: 0.019:  25%|██▌       | 78/312 [07:50<23:41,  6.07s/it] 
Training Epoch: 17/100, batch 155/625 completed (loss: 0.0034:  25%|██▌       | 78/312 [07:50<23:40,  6.07s/it]
Training Epoch: 17/100, batch 156/625 completed (loss: 0.0126:  25%|██▌       | 78/312 [07:53<23:41,  6.08s/it]
Training Epoch: 17/100, batch 156/625 completed (loss: 0.0069:  25%|██▌       | 78/312 [07:53<23:41,  6.07s/it]
Training Epoch: 17/100, batch 156/625 completed (loss: 0.0808:  25%|██▌       | 78/312 [07:53<23:41,  6.07s/it]
Training Epoch: 17/100, batch 156/625 completed (loss: 0.0226:  25%|██▌       | 78/312 [07:53<23:40,  6.07s/it]
Training Epoch: 17/100, batch 156/625 completed (loss: 0.0226:  25%|██▌       | 79/312 [07:56<23:29,  6.05s/it]
Training Epoch: 17/100, batch 156/625 completed (loss: 0.0126:  25%|██▌       | 79/312 [07:56<23:29,  6.05s/it]
Training Epoch: 17/100, batch 156/625 completed (loss: 0.0069:  25%|██▌       | 79/312 [07:56<23:30,  6.05s/it]
Training Epoch: 17/100, batch 156/625 completed (loss: 0.0808:  25%|██▌       | 79/312 [07:56<23:30,  6.06s/it]
Training Epoch: 17/100, batch 157/625 completed (loss: 0.0222:  25%|██▌       | 79/312 [07:56<23:29,  6.05s/it]
Training Epoch: 17/100, batch 157/625 completed (loss: 0.0815:  25%|██▌       | 79/312 [07:56<23:30,  6.05s/it]
Training Epoch: 17/100, batch 157/625 completed (loss: 0.0031:  25%|██▌       | 79/312 [07:56<23:30,  6.06s/it]
Training Epoch: 17/100, batch 157/625 completed (loss: 0.0076:  25%|██▌       | 79/312 [07:56<23:29,  6.05s/it]
Training Epoch: 17/100, batch 158/625 completed (loss: 0.028:  25%|██▌       | 79/312 [07:59<23:29,  6.05s/it] 
Training Epoch: 17/100, batch 158/625 completed (loss: 0.02:  25%|██▌       | 79/312 [07:59<23:30,  6.05s/it]  
Training Epoch: 17/100, batch 158/625 completed (loss: 0.0068:  25%|██▌       | 79/312 [07:59<23:30,  6.06s/it]
Training Epoch: 17/100, batch 158/625 completed (loss: 0.0247:  25%|██▌       | 79/312 [07:59<23:29,  6.05s/it]
Training Epoch: 17/100, batch 158/625 completed (loss: 0.0247:  26%|██▌       | 80/312 [08:02<23:33,  6.09s/it]
Training Epoch: 17/100, batch 158/625 completed (loss: 0.0068:  26%|██▌       | 80/312 [08:02<23:33,  6.09s/it]
Training Epoch: 17/100, batch 158/625 completed (loss: 0.02:  26%|██▌       | 80/312 [08:02<23:33,  6.09s/it]
Training Epoch: 17/100, batch 158/625 completed (loss: 0.028:  26%|██▌       | 80/312 [08:02<23:34,  6.10s/it]
Training Epoch: 17/100, batch 159/625 completed (loss: 0.1613:  26%|██▌       | 80/312 [08:02<23:33,  6.09s/it]
Training Epoch: 17/100, batch 159/625 completed (loss: 0.0482:  26%|██▌       | 80/312 [08:02<23:34,  6.10s/it]
Training Epoch: 17/100, batch 159/625 completed (loss: 0.0359:  26%|██▌       | 80/312 [08:02<23:33,  6.09s/it]
Training Epoch: 17/100, batch 159/625 completed (loss: 0.0474:  26%|██▌       | 80/312 [08:02<23:33,  6.09s/it]
Training Epoch: 17/100, batch 160/625 completed (loss: 0.0396:  26%|██▌       | 80/312 [08:05<23:33,  6.09s/it]
Training Epoch: 17/100, batch 160/625 completed (loss: 0.3947:  26%|██▌       | 80/312 [08:05<23:33,  6.09s/it]
Training Epoch: 17/100, batch 160/625 completed (loss: 0.0059:  26%|██▌       | 80/312 [08:05<23:34,  6.10s/it]
Training Epoch: 17/100, batch 160/625 completed (loss: 0.5183:  26%|██▌       | 80/312 [08:05<23:33,  6.09s/it]
Training Epoch: 17/100, batch 160/625 completed (loss: 0.0059:  26%|██▌       | 81/312 [08:08<23:33,  6.12s/it]
Training Epoch: 17/100, batch 160/625 completed (loss: 0.5183:  26%|██▌       | 81/312 [08:08<23:34,  6.12s/it]
Training Epoch: 17/100, batch 160/625 completed (loss: 0.3947:  26%|██▌       | 81/312 [08:08<23:34,  6.12s/it]
Training Epoch: 17/100, batch 160/625 completed (loss: 0.0396:  26%|██▌       | 81/312 [08:08<23:35,  6.13s/it]
Training Epoch: 17/100, batch 161/625 completed (loss: 0.0385:  26%|██▌       | 81/312 [08:08<23:35,  6.13s/it]
Training Epoch: 17/100, batch 161/625 completed (loss: 0.0131:  26%|██▌       | 81/312 [08:08<23:33,  6.12s/it]
Training Epoch: 17/100, batch 161/625 completed (loss: 0.0646:  26%|██▌       | 81/312 [08:08<23:34,  6.12s/it]
Training Epoch: 17/100, batch 161/625 completed (loss: 0.003:  26%|██▌       | 81/312 [08:09<23:34,  6.12s/it] 
Training Epoch: 17/100, batch 162/625 completed (loss: 0.0096:  26%|██▌       | 81/312 [08:11<23:34,  6.12s/it]
Training Epoch: 17/100, batch 162/625 completed (loss: 0.0433:  26%|██▌       | 81/312 [08:11<23:35,  6.13s/it]
Training Epoch: 17/100, batch 162/625 completed (loss: 0.0162:  26%|██▌       | 81/312 [08:11<23:33,  6.12s/it]
Training Epoch: 17/100, batch 162/625 completed (loss: 0.0076:  26%|██▌       | 81/312 [08:11<23:34,  6.12s/it]
Training Epoch: 17/100, batch 162/625 completed (loss: 0.0096:  26%|██▋       | 82/312 [08:14<23:22,  6.10s/it]
Training Epoch: 17/100, batch 162/625 completed (loss: 0.0076:  26%|██▋       | 82/312 [08:14<23:22,  6.10s/it]
Training Epoch: 17/100, batch 162/625 completed (loss: 0.0433:  26%|██▋       | 82/312 [08:14<23:22,  6.10s/it]
Training Epoch: 17/100, batch 162/625 completed (loss: 0.0162:  26%|██▋       | 82/312 [08:14<23:22,  6.10s/it]
Training Epoch: 17/100, batch 163/625 completed (loss: 0.0055:  26%|██▋       | 82/312 [08:14<23:22,  6.10s/it]
Training Epoch: 17/100, batch 163/625 completed (loss: 0.1659:  26%|██▋       | 82/312 [08:14<23:22,  6.10s/it]
Training Epoch: 17/100, batch 163/625 completed (loss: 0.2324:  26%|██▋       | 82/312 [08:14<23:22,  6.10s/it]
Training Epoch: 17/100, batch 163/625 completed (loss: 0.0069:  26%|██▋       | 82/312 [08:15<23:22,  6.10s/it]
Training Epoch: 17/100, batch 164/625 completed (loss: 0.0172:  26%|██▋       | 82/312 [08:17<23:22,  6.10s/it]
Training Epoch: 17/100, batch 164/625 completed (loss: 0.0095:  26%|██▋       | 82/312 [08:17<23:22,  6.10s/it]
Training Epoch: 17/100, batch 164/625 completed (loss: 0.0009:  26%|██▋       | 82/312 [08:17<23:22,  6.10s/it]
Training Epoch: 17/100, batch 164/625 completed (loss: 0.0373:  26%|██▋       | 82/312 [08:17<23:22,  6.10s/it]
Training Epoch: 17/100, batch 164/625 completed (loss: 0.0172:  27%|██▋       | 83/312 [08:21<23:18,  6.11s/it]
Training Epoch: 17/100, batch 164/625 completed (loss: 0.0373:  27%|██▋       | 83/312 [08:21<23:18,  6.11s/it]
Training Epoch: 17/100, batch 164/625 completed (loss: 0.0095:  27%|██▋       | 83/312 [08:21<23:18,  6.11s/it]
Training Epoch: 17/100, batch 164/625 completed (loss: 0.0009:  27%|██▋       | 83/312 [08:21<23:19,  6.11s/it]
Training Epoch: 17/100, batch 165/625 completed (loss: 0.0063:  27%|██▋       | 83/312 [08:21<23:18,  6.11s/it]
Training Epoch: 17/100, batch 165/625 completed (loss: 0.0042:  27%|██▋       | 83/312 [08:21<23:19,  6.11s/it]
Training Epoch: 17/100, batch 165/625 completed (loss: 0.1813:  27%|██▋       | 83/312 [08:21<23:18,  6.11s/it]
Training Epoch: 17/100, batch 165/625 completed (loss: 0.0061:  27%|██▋       | 83/312 [08:21<23:18,  6.11s/it]
Training Epoch: 17/100, batch 166/625 completed (loss: 0.1126:  27%|██▋       | 83/312 [08:24<23:18,  6.11s/it]
Training Epoch: 17/100, batch 166/625 completed (loss: 0.0105:  27%|██▋       | 83/312 [08:24<23:18,  6.11s/it]
Training Epoch: 17/100, batch 166/625 completed (loss: 0.0191:  27%|██▋       | 83/312 [08:24<23:19,  6.11s/it]
Training Epoch: 17/100, batch 166/625 completed (loss: 0.0761:  27%|██▋       | 83/312 [08:24<23:18,  6.11s/it]
Training Epoch: 17/100, batch 166/625 completed (loss: 0.0761:  27%|██▋       | 84/312 [08:27<23:06,  6.08s/it]
Training Epoch: 17/100, batch 166/625 completed (loss: 0.1126:  27%|██▋       | 84/312 [08:27<23:05,  6.08s/it]
Training Epoch: 17/100, batch 166/625 completed (loss: 0.0191:  27%|██▋       | 84/312 [08:27<23:06,  6.08s/it]
Training Epoch: 17/100, batch 166/625 completed (loss: 0.0105:  27%|██▋       | 84/312 [08:27<23:06,  6.08s/it]
Training Epoch: 17/100, batch 167/625 completed (loss: 0.0437:  27%|██▋       | 84/312 [08:27<23:05,  6.08s/it]
Training Epoch: 17/100, batch 167/625 completed (loss: 0.1051:  27%|██▋       | 84/312 [08:27<23:06,  6.08s/it]
Training Epoch: 17/100, batch 167/625 completed (loss: 0.2485:  27%|██▋       | 84/312 [08:27<23:06,  6.08s/it]
Training Epoch: 17/100, batch 167/625 completed (loss: 0.0048:  27%|██▋       | 84/312 [08:27<23:06,  6.08s/it]
Training Epoch: 17/100, batch 168/625 completed (loss: 0.0049:  27%|██▋       | 84/312 [08:30<23:06,  6.08s/it]
Training Epoch: 17/100, batch 168/625 completed (loss: 0.2776:  27%|██▋       | 84/312 [08:30<23:05,  6.08s/it]
Training Epoch: 17/100, batch 168/625 completed (loss: 0.0091:  27%|██▋       | 84/312 [08:30<23:06,  6.08s/it]
Training Epoch: 17/100, batch 168/625 completed (loss: 0.0289:  27%|██▋       | 84/312 [08:30<23:06,  6.08s/it]
Training Epoch: 17/100, batch 168/625 completed (loss: 0.2776:  27%|██▋       | 85/312 [08:33<22:56,  6.06s/it]
Training Epoch: 17/100, batch 168/625 completed (loss: 0.0091:  27%|██▋       | 85/312 [08:33<22:56,  6.06s/it]
Training Epoch: 17/100, batch 168/625 completed (loss: 0.0289:  27%|██▋       | 85/312 [08:33<22:57,  6.07s/it]
Training Epoch: 17/100, batch 168/625 completed (loss: 0.0049:  27%|██▋       | 85/312 [08:33<22:56,  6.07s/it]
Training Epoch: 17/100, batch 169/625 completed (loss: 0.0294:  27%|██▋       | 85/312 [08:33<22:56,  6.06s/it]
Training Epoch: 17/100, batch 169/625 completed (loss: 0.0036:  27%|██▋       | 85/312 [08:33<22:56,  6.07s/it]
Training Epoch: 17/100, batch 169/625 completed (loss: 0.0198:  27%|██▋       | 85/312 [08:33<22:56,  6.06s/it]
Training Epoch: 17/100, batch 169/625 completed (loss: 0.2297:  27%|██▋       | 85/312 [08:33<22:57,  6.07s/it]
Training Epoch: 17/100, batch 170/625 completed (loss: 0.0132:  27%|██▋       | 85/312 [08:36<22:56,  6.06s/it]
Training Epoch: 17/100, batch 170/625 completed (loss: 0.0251:  27%|██▋       | 85/312 [08:36<22:56,  6.06s/it]
Training Epoch: 17/100, batch 170/625 completed (loss: 0.0056:  27%|██▋       | 85/312 [08:36<22:56,  6.07s/it]
Training Epoch: 17/100, batch 170/625 completed (loss: 0.0068:  27%|██▋       | 85/312 [08:36<22:57,  6.07s/it]
Training Epoch: 17/100, batch 170/625 completed (loss: 0.0251:  28%|██▊       | 86/312 [08:39<23:04,  6.13s/it]
Training Epoch: 17/100, batch 170/625 completed (loss: 0.0068:  28%|██▊       | 86/312 [08:39<23:05,  6.13s/it]
Training Epoch: 17/100, batch 170/625 completed (loss: 0.0056:  28%|██▊       | 86/312 [08:39<23:05,  6.13s/it]
Training Epoch: 17/100, batch 170/625 completed (loss: 0.0132:  28%|██▊       | 86/312 [08:39<23:05,  6.13s/it]
Training Epoch: 17/100, batch 171/625 completed (loss: 0.0616:  28%|██▊       | 86/312 [08:39<23:05,  6.13s/it]
Training Epoch: 17/100, batch 171/625 completed (loss: 0.1516:  28%|██▊       | 86/312 [08:39<23:05,  6.13s/it]
Training Epoch: 17/100, batch 171/625 completed (loss: 0.0565:  28%|██▊       | 86/312 [08:39<23:04,  6.13s/it]
Training Epoch: 17/100, batch 171/625 completed (loss: 0.0421:  28%|██▊       | 86/312 [08:39<23:05,  6.13s/it]
Training Epoch: 17/100, batch 172/625 completed (loss: 0.02:  28%|██▊       | 86/312 [08:42<23:04,  6.13s/it]  
Training Epoch: 17/100, batch 172/625 completed (loss: 0.0027:  28%|██▊       | 86/312 [08:42<23:05,  6.13s/it]
Training Epoch: 17/100, batch 172/625 completed (loss: 0.1762:  28%|██▊       | 86/312 [08:42<23:05,  6.13s/it]
Training Epoch: 17/100, batch 172/625 completed (loss: 0.0119:  28%|██▊       | 86/312 [08:42<23:05,  6.13s/it]
Training Epoch: 17/100, batch 172/625 completed (loss: 0.02:  28%|██▊       | 87/312 [08:45<22:40,  6.05s/it]
Training Epoch: 17/100, batch 172/625 completed (loss: 0.0027:  28%|██▊       | 87/312 [08:45<22:39,  6.04s/it]
Training Epoch: 17/100, batch 172/625 completed (loss: 0.1762:  28%|██▊       | 87/312 [08:45<22:40,  6.05s/it]
Training Epoch: 17/100, batch 172/625 completed (loss: 0.0119:  28%|██▊       | 87/312 [08:45<22:40,  6.05s/it]
Training Epoch: 17/100, batch 173/625 completed (loss: 0.0196:  28%|██▊       | 87/312 [08:45<22:40,  6.05s/it]
Training Epoch: 17/100, batch 173/625 completed (loss: 0.0043:  28%|██▊       | 87/312 [08:45<22:39,  6.04s/it]
Training Epoch: 17/100, batch 173/625 completed (loss: 0.0775:  28%|██▊       | 87/312 [08:45<22:40,  6.05s/it]
Training Epoch: 17/100, batch 173/625 completed (loss: 0.0154:  28%|██▊       | 87/312 [08:45<22:40,  6.05s/it]
Training Epoch: 17/100, batch 174/625 completed (loss: 0.0035:  28%|██▊       | 87/312 [08:48<22:39,  6.04s/it]
Training Epoch: 17/100, batch 174/625 completed (loss: 0.0031:  28%|██▊       | 87/312 [08:48<22:40,  6.05s/it]
Training Epoch: 17/100, batch 174/625 completed (loss: 0.0279:  28%|██▊       | 87/312 [08:48<22:40,  6.05s/it]
Training Epoch: 17/100, batch 174/625 completed (loss: 0.0121:  28%|██▊       | 87/312 [08:48<22:40,  6.05s/it]
Training Epoch: 17/100, batch 174/625 completed (loss: 0.0035:  28%|██▊       | 88/312 [08:51<22:31,  6.03s/it]
Training Epoch: 17/100, batch 174/625 completed (loss: 0.0121:  28%|██▊       | 88/312 [08:51<22:31,  6.03s/it]
Training Epoch: 17/100, batch 174/625 completed (loss: 0.0031:  28%|██▊       | 88/312 [08:51<22:31,  6.04s/it]
Training Epoch: 17/100, batch 174/625 completed (loss: 0.0279:  28%|██▊       | 88/312 [08:51<22:31,  6.03s/it]
Training Epoch: 17/100, batch 175/625 completed (loss: 0.0078:  28%|██▊       | 88/312 [08:51<22:31,  6.04s/it]
Training Epoch: 17/100, batch 175/625 completed (loss: 0.0706:  28%|██▊       | 88/312 [08:51<22:31,  6.03s/it]
Training Epoch: 17/100, batch 175/625 completed (loss: 0.0044:  28%|██▊       | 88/312 [08:51<22:31,  6.03s/it]
Training Epoch: 17/100, batch 175/625 completed (loss: 0.0744:  28%|██▊       | 88/312 [08:51<22:31,  6.03s/it]
Training Epoch: 17/100, batch 176/625 completed (loss: 0.0144:  28%|██▊       | 88/312 [08:54<22:31,  6.03s/it]
Training Epoch: 17/100, batch 176/625 completed (loss: 0.0853:  28%|██▊       | 88/312 [08:54<22:31,  6.04s/it]
Training Epoch: 17/100, batch 176/625 completed (loss: 0.0931:  28%|██▊       | 88/312 [08:54<22:31,  6.03s/it]
Training Epoch: 17/100, batch 176/625 completed (loss: 0.3751:  28%|██▊       | 88/312 [08:54<22:31,  6.03s/it]
Training Epoch: 17/100, batch 176/625 completed (loss: 0.0853:  29%|██▊       | 89/312 [08:57<22:42,  6.11s/it]
Training Epoch: 17/100, batch 176/625 completed (loss: 0.3751:  29%|██▊       | 89/312 [08:57<22:43,  6.11s/it]
Training Epoch: 17/100, batch 176/625 completed (loss: 0.0144:  29%|██▊       | 89/312 [08:57<22:43,  6.11s/it]
Training Epoch: 17/100, batch 176/625 completed (loss: 0.0931:  29%|██▊       | 89/312 [08:57<22:43,  6.12s/it]
Training Epoch: 17/100, batch 177/625 completed (loss: 0.0107:  29%|██▊       | 89/312 [08:57<22:42,  6.11s/it]
Training Epoch: 17/100, batch 177/625 completed (loss: 0.0747:  29%|██▊       | 89/312 [08:57<22:43,  6.11s/it]
Training Epoch: 17/100, batch 177/625 completed (loss: 0.0433:  29%|██▊       | 89/312 [08:57<22:43,  6.12s/it]
Training Epoch: 17/100, batch 177/625 completed (loss: 0.0327:  29%|██▊       | 89/312 [08:57<22:43,  6.11s/it]
Training Epoch: 17/100, batch 178/625 completed (loss: 0.0838:  29%|██▊       | 89/312 [09:00<22:43,  6.12s/it]
Training Epoch: 17/100, batch 178/625 completed (loss: 0.0019:  29%|██▊       | 89/312 [09:00<22:43,  6.11s/it]
Training Epoch: 17/100, batch 178/625 completed (loss: 0.0114:  29%|██▊       | 89/312 [09:00<22:42,  6.11s/it]
Training Epoch: 17/100, batch 178/625 completed (loss: 0.0387:  29%|██▊       | 89/312 [09:00<22:43,  6.11s/it]
Training Epoch: 17/100, batch 178/625 completed (loss: 0.0114:  29%|██▉       | 90/312 [09:03<22:19,  6.03s/it]
Training Epoch: 17/100, batch 178/625 completed (loss: 0.0387:  29%|██▉       | 90/312 [09:03<22:19,  6.03s/it]
Training Epoch: 17/100, batch 178/625 completed (loss: 0.0019:  29%|██▉       | 90/312 [09:03<22:20,  6.04s/it]
Training Epoch: 17/100, batch 178/625 completed (loss: 0.0838:  29%|██▉       | 90/312 [09:03<22:20,  6.04s/it]
Training Epoch: 17/100, batch 179/625 completed (loss: 0.0941:  29%|██▉       | 90/312 [09:03<22:19,  6.03s/it]
Training Epoch: 17/100, batch 179/625 completed (loss: 0.0158:  29%|██▉       | 90/312 [09:03<22:20,  6.04s/it]
Training Epoch: 17/100, batch 179/625 completed (loss: 0.0125:  29%|██▉       | 90/312 [09:03<22:20,  6.04s/it]
Training Epoch: 17/100, batch 179/625 completed (loss: 0.01:  29%|██▉       | 90/312 [09:03<22:19,  6.03s/it]  
Training Epoch: 17/100, batch 180/625 completed (loss: 0.0216:  29%|██▉       | 90/312 [09:06<22:20,  6.04s/it]
Training Epoch: 17/100, batch 180/625 completed (loss: 0.266:  29%|██▉       | 90/312 [09:06<22:20,  6.04s/it] 
Training Epoch: 17/100, batch 180/625 completed (loss: 0.0145:  29%|██▉       | 90/312 [09:06<22:19,  6.03s/it]
Training Epoch: 17/100, batch 180/625 completed (loss: 0.0031:  29%|██▉       | 90/312 [09:06<22:19,  6.03s/it]
Training Epoch: 17/100, batch 180/625 completed (loss: 0.0031:  29%|██▉       | 91/312 [09:09<22:32,  6.12s/it]
Training Epoch: 17/100, batch 180/625 completed (loss: 0.0145:  29%|██▉       | 91/312 [09:09<22:33,  6.12s/it]
Training Epoch: 17/100, batch 180/625 completed (loss: 0.0216:  29%|██▉       | 91/312 [09:09<22:32,  6.12s/it]
Training Epoch: 17/100, batch 180/625 completed (loss: 0.266:  29%|██▉       | 91/312 [09:09<22:33,  6.12s/it]
Training Epoch: 17/100, batch 181/625 completed (loss: 0.0218:  29%|██▉       | 91/312 [09:09<22:32,  6.12s/it]
Training Epoch: 17/100, batch 181/625 completed (loss: 0.0063:  29%|██▉       | 91/312 [09:09<22:33,  6.12s/it]
Training Epoch: 17/100, batch 181/625 completed (loss: 0.0058:  29%|██▉       | 91/312 [09:09<22:33,  6.12s/it]
Training Epoch: 17/100, batch 181/625 completed (loss: 0.3729:  29%|██▉       | 91/312 [09:09<22:32,  6.12s/it]
Training Epoch: 17/100, batch 182/625 completed (loss: 0.0552:  29%|██▉       | 91/312 [09:12<22:33,  6.12s/it]
Training Epoch: 17/100, batch 182/625 completed (loss: 0.4507:  29%|██▉       | 91/312 [09:12<22:32,  6.12s/it]
Training Epoch: 17/100, batch 182/625 completed (loss: 0.0406:  29%|██▉       | 91/312 [09:12<22:33,  6.12s/it]
Training Epoch: 17/100, batch 182/625 completed (loss: 0.3566:  29%|██▉       | 91/312 [09:12<22:32,  6.12s/it]
Training Epoch: 17/100, batch 182/625 completed (loss: 0.0552:  29%|██▉       | 92/312 [09:15<22:19,  6.09s/it]
Training Epoch: 17/100, batch 182/625 completed (loss: 0.4507:  29%|██▉       | 92/312 [09:15<22:19,  6.09s/it]
Training Epoch: 17/100, batch 182/625 completed (loss: 0.0406:  29%|██▉       | 92/312 [09:15<22:18,  6.09s/it]
Training Epoch: 17/100, batch 182/625 completed (loss: 0.3566:  29%|██▉       | 92/312 [09:15<22:19,  6.09s/it]
Training Epoch: 17/100, batch 183/625 completed (loss: 0.1641:  29%|██▉       | 92/312 [09:15<22:19,  6.09s/it]
Training Epoch: 17/100, batch 183/625 completed (loss: 0.0312:  29%|██▉       | 92/312 [09:15<22:18,  6.09s/it]
Training Epoch: 17/100, batch 183/625 completed (loss: 0.0123:  29%|██▉       | 92/312 [09:15<22:19,  6.09s/it]
Training Epoch: 17/100, batch 183/625 completed (loss: 0.303:  29%|██▉       | 92/312 [09:15<22:19,  6.09s/it] 
Training Epoch: 17/100, batch 184/625 completed (loss: 0.0606:  29%|██▉       | 92/312 [09:18<22:19,  6.09s/it]
Training Epoch: 17/100, batch 184/625 completed (loss: 0.0041:  29%|██▉       | 92/312 [09:18<22:18,  6.09s/it]
Training Epoch: 17/100, batch 184/625 completed (loss: 0.0243:  29%|██▉       | 92/312 [09:18<22:19,  6.09s/it]
Training Epoch: 17/100, batch 184/625 completed (loss: 0.0603:  29%|██▉       | 92/312 [09:18<22:19,  6.09s/it]
Training Epoch: 17/100, batch 184/625 completed (loss: 0.0603:  30%|██▉       | 93/312 [09:21<22:07,  6.06s/it]
Training Epoch: 17/100, batch 184/625 completed (loss: 0.0243:  30%|██▉       | 93/312 [09:21<22:08,  6.06s/it]
Training Epoch: 17/100, batch 184/625 completed (loss: 0.0606:  30%|██▉       | 93/312 [09:21<22:07,  6.06s/it]
Training Epoch: 17/100, batch 184/625 completed (loss: 0.0041:  30%|██▉       | 93/312 [09:21<22:07,  6.06s/it]
Training Epoch: 17/100, batch 185/625 completed (loss: 0.0042:  30%|██▉       | 93/312 [09:21<22:07,  6.06s/it]
Training Epoch: 17/100, batch 185/625 completed (loss: 0.0066:  30%|██▉       | 93/312 [09:21<22:08,  6.06s/it]
Training Epoch: 17/100, batch 185/625 completed (loss: 0.018:  30%|██▉       | 93/312 [09:21<22:07,  6.06s/it] 
Training Epoch: 17/100, batch 185/625 completed (loss: 0.2952:  30%|██▉       | 93/312 [09:21<22:07,  6.06s/it]
Training Epoch: 17/100, batch 186/625 completed (loss: 0.0034:  30%|██▉       | 93/312 [09:24<22:07,  6.06s/it]
Training Epoch: 17/100, batch 186/625 completed (loss: 0.0022:  30%|██▉       | 93/312 [09:24<22:07,  6.06s/it]
Training Epoch: 17/100, batch 186/625 completed (loss: 0.2519:  30%|██▉       | 93/312 [09:24<22:08,  6.06s/it]
Training Epoch: 17/100, batch 186/625 completed (loss: 0.0062:  30%|██▉       | 93/312 [09:24<22:07,  6.06s/it]
Training Epoch: 17/100, batch 186/625 completed (loss: 0.0034:  30%|███       | 94/312 [09:27<22:08,  6.09s/it]
Training Epoch: 17/100, batch 186/625 completed (loss: 0.2519:  30%|███       | 94/312 [09:27<22:09,  6.10s/it]
Training Epoch: 17/100, batch 186/625 completed (loss: 0.0022:  30%|███       | 94/312 [09:27<22:09,  6.10s/it]
Training Epoch: 17/100, batch 186/625 completed (loss: 0.0062:  30%|███       | 94/312 [09:27<22:09,  6.10s/it]
Training Epoch: 17/100, batch 187/625 completed (loss: 0.0047:  30%|███       | 94/312 [09:27<22:09,  6.10s/it]
Training Epoch: 17/100, batch 187/625 completed (loss: 0.0287:  30%|███       | 94/312 [09:27<22:08,  6.09s/it]
Training Epoch: 17/100, batch 187/625 completed (loss: 0.2005:  30%|███       | 94/312 [09:27<22:09,  6.10s/it]
Training Epoch: 17/100, batch 187/625 completed (loss: 0.0107:  30%|███       | 94/312 [09:28<22:09,  6.10s/it]
Training Epoch: 17/100, batch 188/625 completed (loss: 0.0369:  30%|███       | 94/312 [09:30<22:09,  6.10s/it]
Training Epoch: 17/100, batch 188/625 completed (loss: 0.0386:  30%|███       | 94/312 [09:30<22:08,  6.09s/it]
Training Epoch: 17/100, batch 188/625 completed (loss: 0.0016:  30%|███       | 94/312 [09:30<22:09,  6.10s/it]
Training Epoch: 17/100, batch 188/625 completed (loss: 0.0081:  30%|███       | 94/312 [09:30<22:09,  6.10s/it]
Training Epoch: 17/100, batch 188/625 completed (loss: 0.0016:  30%|███       | 95/312 [09:33<21:45,  6.01s/it]
Training Epoch: 17/100, batch 188/625 completed (loss: 0.0369:  30%|███       | 95/312 [09:33<21:45,  6.02s/it]
Training Epoch: 17/100, batch 188/625 completed (loss: 0.0081:  30%|███       | 95/312 [09:33<21:45,  6.02s/it]
Training Epoch: 17/100, batch 188/625 completed (loss: 0.0386:  30%|███       | 95/312 [09:33<21:46,  6.02s/it]
Training Epoch: 17/100, batch 189/625 completed (loss: 0.0059:  30%|███       | 95/312 [09:33<21:45,  6.01s/it]
Training Epoch: 17/100, batch 189/625 completed (loss: 0.0067:  30%|███       | 95/312 [09:33<21:45,  6.02s/it]
Training Epoch: 17/100, batch 189/625 completed (loss: 0.0306:  30%|███       | 95/312 [09:33<21:46,  6.02s/it]
Training Epoch: 17/100, batch 189/625 completed (loss: 0.0949:  30%|███       | 95/312 [09:33<21:45,  6.02s/it]
Training Epoch: 17/100, batch 190/625 completed (loss: 0.0167:  30%|███       | 95/312 [09:36<21:46,  6.02s/it]
Training Epoch: 17/100, batch 190/625 completed (loss: 0.0566:  30%|███       | 95/312 [09:36<21:45,  6.02s/it]
Training Epoch: 17/100, batch 190/625 completed (loss: 0.0076:  30%|███       | 95/312 [09:36<21:45,  6.01s/it]
Training Epoch: 17/100, batch 190/625 completed (loss: 0.1608:  30%|███       | 95/312 [09:36<21:45,  6.02s/it]
Training Epoch: 17/100, batch 190/625 completed (loss: 0.0566:  31%|███       | 96/312 [09:40<21:57,  6.10s/it]
Training Epoch: 17/100, batch 190/625 completed (loss: 0.0076:  31%|███       | 96/312 [09:40<21:57,  6.10s/it]
Training Epoch: 17/100, batch 190/625 completed (loss: 0.0167:  31%|███       | 96/312 [09:40<21:57,  6.10s/it]
Training Epoch: 17/100, batch 190/625 completed (loss: 0.1608:  31%|███       | 96/312 [09:40<21:58,  6.10s/it]
Training Epoch: 17/100, batch 191/625 completed (loss: 0.0124:  31%|███       | 96/312 [09:40<21:57,  6.10s/it]
Training Epoch: 17/100, batch 191/625 completed (loss: 0.1039:  31%|███       | 96/312 [09:40<21:57,  6.10s/it]
Training Epoch: 17/100, batch 191/625 completed (loss: 0.0392:  31%|███       | 96/312 [09:40<21:57,  6.10s/it]
Training Epoch: 17/100, batch 191/625 completed (loss: 0.1713:  31%|███       | 96/312 [09:40<21:58,  6.10s/it]
Training Epoch: 17/100, batch 192/625 completed (loss: 0.0023:  31%|███       | 96/312 [09:43<21:57,  6.10s/it]
Training Epoch: 17/100, batch 192/625 completed (loss: 0.0466:  31%|███       | 96/312 [09:43<21:57,  6.10s/it]
Training Epoch: 17/100, batch 192/625 completed (loss: 0.0236:  31%|███       | 96/312 [09:43<21:57,  6.10s/it]
Training Epoch: 17/100, batch 192/625 completed (loss: 0.0072:  31%|███       | 96/312 [09:43<21:58,  6.10s/it]
Training Epoch: 17/100, batch 192/625 completed (loss: 0.0023:  31%|███       | 97/312 [09:46<21:45,  6.07s/it]
Training Epoch: 17/100, batch 192/625 completed (loss: 0.0466:  31%|███       | 97/312 [09:46<21:45,  6.07s/it]
Training Epoch: 17/100, batch 192/625 completed (loss: 0.0236:  31%|███       | 97/312 [09:46<21:45,  6.07s/it]
Training Epoch: 17/100, batch 192/625 completed (loss: 0.0072:  31%|███       | 97/312 [09:46<21:45,  6.07s/it]
Training Epoch: 17/100, batch 193/625 completed (loss: 0.0094:  31%|███       | 97/312 [09:46<21:45,  6.07s/it]
Training Epoch: 17/100, batch 193/625 completed (loss: 0.043:  31%|███       | 97/312 [09:46<21:45,  6.07s/it] 
Training Epoch: 17/100, batch 193/625 completed (loss: 0.0025:  31%|███       | 97/312 [09:46<21:45,  6.07s/it]
Training Epoch: 17/100, batch 193/625 completed (loss: 0.0146:  31%|███       | 97/312 [09:46<21:45,  6.07s/it]
Training Epoch: 17/100, batch 194/625 completed (loss: 0.0572:  31%|███       | 97/312 [09:49<21:45,  6.07s/it]
Training Epoch: 17/100, batch 194/625 completed (loss: 0.05:  31%|███       | 97/312 [09:49<21:45,  6.07s/it] 
Training Epoch: 17/100, batch 194/625 completed (loss: 0.0124:  31%|███       | 97/312 [09:49<21:45,  6.07s/it]
Training Epoch: 17/100, batch 194/625 completed (loss: 0.0042:  31%|███       | 97/312 [09:49<21:45,  6.07s/it]
Training Epoch: 17/100, batch 194/625 completed (loss: 0.05:  31%|███▏      | 98/312 [09:51<21:25,  6.01s/it]
Training Epoch: 17/100, batch 194/625 completed (loss: 0.0042:  31%|███▏      | 98/312 [09:51<21:25,  6.01s/it]
Training Epoch: 17/100, batch 194/625 completed (loss: 0.0572:  31%|███▏      | 98/312 [09:51<21:25,  6.01s/it]
Training Epoch: 17/100, batch 194/625 completed (loss: 0.0124:  31%|███▏      | 98/312 [09:51<21:26,  6.01s/it]
Training Epoch: 17/100, batch 195/625 completed (loss: 0.0482:  31%|███▏      | 98/312 [09:51<21:26,  6.01s/it]
Training Epoch: 17/100, batch 195/625 completed (loss: 0.0141:  31%|███▏      | 98/312 [09:51<21:25,  6.01s/it]
Training Epoch: 17/100, batch 195/625 completed (loss: 0.0465:  31%|███▏      | 98/312 [09:51<21:25,  6.01s/it]
Training Epoch: 17/100, batch 195/625 completed (loss: 0.0165:  31%|███▏      | 98/312 [09:52<21:25,  6.01s/it]
Training Epoch: 17/100, batch 196/625 completed (loss: 0.0655:  31%|███▏      | 98/312 [09:55<21:25,  6.01s/it]
Training Epoch: 17/100, batch 196/625 completed (loss: 0.089:  31%|███▏      | 98/312 [09:55<21:26,  6.01s/it] 
Training Epoch: 17/100, batch 196/625 completed (loss: 0.0209:  31%|███▏      | 98/312 [09:55<21:25,  6.01s/it]
Training Epoch: 17/100, batch 196/625 completed (loss: 0.0087:  31%|███▏      | 98/312 [09:55<21:25,  6.01s/it]
Training Epoch: 17/100, batch 196/625 completed (loss: 0.0209:  32%|███▏      | 99/312 [09:58<21:39,  6.10s/it]
Training Epoch: 17/100, batch 196/625 completed (loss: 0.0655:  32%|███▏      | 99/312 [09:58<21:39,  6.10s/it]
Training Epoch: 17/100, batch 196/625 completed (loss: 0.089:  32%|███▏      | 99/312 [09:58<21:39,  6.10s/it]
Training Epoch: 17/100, batch 196/625 completed (loss: 0.0087:  32%|███▏      | 99/312 [09:58<21:39,  6.10s/it]
Training Epoch: 17/100, batch 197/625 completed (loss: 0.3561:  32%|███▏      | 99/312 [09:58<21:39,  6.10s/it]
Training Epoch: 17/100, batch 197/625 completed (loss: 0.0199:  32%|███▏      | 99/312 [09:58<21:39,  6.10s/it]
Training Epoch: 17/100, batch 197/625 completed (loss: 0.0052:  32%|███▏      | 99/312 [09:58<21:39,  6.10s/it]
Training Epoch: 17/100, batch 197/625 completed (loss: 0.0079:  32%|███▏      | 99/312 [09:58<21:39,  6.10s/it]
Training Epoch: 17/100, batch 198/625 completed (loss: 0.0195:  32%|███▏      | 99/312 [10:01<21:39,  6.10s/it]
Training Epoch: 17/100, batch 198/625 completed (loss: 0.0214:  32%|███▏      | 99/312 [10:01<21:39,  6.10s/it]
Training Epoch: 17/100, batch 198/625 completed (loss: 0.1579:  32%|███▏      | 99/312 [10:01<21:39,  6.10s/it]
Training Epoch: 17/100, batch 198/625 completed (loss: 0.008:  32%|███▏      | 99/312 [10:01<21:39,  6.10s/it] 
Training Epoch: 17/100, batch 198/625 completed (loss: 0.0195:  32%|███▏      | 100/312 [10:04<21:27,  6.07s/it]
Training Epoch: 17/100, batch 198/625 completed (loss: 0.0214:  32%|███▏      | 100/312 [10:04<21:26,  6.07s/it]
Training Epoch: 17/100, batch 198/625 completed (loss: 0.1579:  32%|███▏      | 100/312 [10:04<21:27,  6.07s/it]
Training Epoch: 17/100, batch 198/625 completed (loss: 0.008:  32%|███▏      | 100/312 [10:04<21:27,  6.07s/it]
Training Epoch: 17/100, batch 199/625 completed (loss: 0.0071:  32%|███▏      | 100/312 [10:04<21:26,  6.07s/it]
Training Epoch: 17/100, batch 199/625 completed (loss: 0.0152:  32%|███▏      | 100/312 [10:04<21:27,  6.07s/it]
Training Epoch: 17/100, batch 199/625 completed (loss: 0.0029:  32%|███▏      | 100/312 [10:04<21:27,  6.07s/it]
Training Epoch: 17/100, batch 199/625 completed (loss: 0.0065:  32%|███▏      | 100/312 [10:04<21:27,  6.07s/it]
Training Epoch: 17/100, batch 200/625 completed (loss: 0.202:  32%|███▏      | 100/312 [10:07<21:27,  6.07s/it] 
Training Epoch: 17/100, batch 200/625 completed (loss: 0.2332:  32%|███▏      | 100/312 [10:07<21:26,  6.07s/it]
Training Epoch: 17/100, batch 200/625 completed (loss: 0.0132:  32%|███▏      | 100/312 [10:07<21:27,  6.07s/it]
Training Epoch: 17/100, batch 200/625 completed (loss: 0.3058:  32%|███▏      | 100/312 [10:07<21:27,  6.07s/it]
Training Epoch: 17/100, batch 200/625 completed (loss: 0.0132:  32%|███▏      | 101/312 [10:10<21:33,  6.13s/it]
Training Epoch: 17/100, batch 200/625 completed (loss: 0.202:  32%|███▏      | 101/312 [10:10<21:34,  6.13s/it]
Training Epoch: 17/100, batch 200/625 completed (loss: 0.2332:  32%|███▏      | 101/312 [10:10<21:33,  6.13s/it]
Training Epoch: 17/100, batch 200/625 completed (loss: 0.3058:  32%|███▏      | 101/312 [10:10<21:33,  6.13s/it]
Training Epoch: 17/100, batch 201/625 completed (loss: 0.0783:  32%|███▏      | 101/312 [10:10<21:34,  6.13s/it]
Training Epoch: 17/100, batch 201/625 completed (loss: 0.0202:  32%|███▏      | 101/312 [10:10<21:33,  6.13s/it]
Training Epoch: 17/100, batch 201/625 completed (loss: 0.0265:  32%|███▏      | 101/312 [10:10<21:33,  6.13s/it]
Training Epoch: 17/100, batch 201/625 completed (loss: 0.0034:  32%|███▏      | 101/312 [10:10<21:33,  6.13s/it]
Training Epoch: 17/100, batch 202/625 completed (loss: 0.1376:  32%|███▏      | 101/312 [10:13<21:33,  6.13s/it]
Training Epoch: 17/100, batch 202/625 completed (loss: 0.1947:  32%|███▏      | 101/312 [10:13<21:34,  6.13s/it]
Training Epoch: 17/100, batch 202/625 completed (loss: 0.006:  32%|███▏      | 101/312 [10:13<21:33,  6.13s/it] 
Training Epoch: 17/100, batch 202/625 completed (loss: 0.0093:  32%|███▏      | 101/312 [10:13<21:33,  6.13s/it]
Training Epoch: 17/100, batch 202/625 completed (loss: 0.1947:  33%|███▎      | 102/312 [10:16<21:09,  6.04s/it]
Training Epoch: 17/100, batch 202/625 completed (loss: 0.006:  33%|███▎      | 102/312 [10:16<21:09,  6.04s/it]
Training Epoch: 17/100, batch 202/625 completed (loss: 0.0093:  33%|███▎      | 102/312 [10:16<21:08,  6.04s/it]
Training Epoch: 17/100, batch 202/625 completed (loss: 0.1376:  33%|███▎      | 102/312 [10:16<21:09,  6.04s/it]
Training Epoch: 17/100, batch 203/625 completed (loss: 0.2909:  33%|███▎      | 102/312 [10:16<21:09,  6.04s/it]
Training Epoch: 17/100, batch 203/625 completed (loss: 0.0042:  33%|███▎      | 102/312 [10:16<21:09,  6.04s/it]
Training Epoch: 17/100, batch 203/625 completed (loss: 0.1659:  33%|███▎      | 102/312 [10:16<21:09,  6.04s/it]
Training Epoch: 17/100, batch 203/625 completed (loss: 0.002:  33%|███▎      | 102/312 [10:16<21:08,  6.04s/it] 
Training Epoch: 17/100, batch 204/625 completed (loss: 0.0298:  33%|███▎      | 102/312 [10:19<21:09,  6.04s/it]
Training Epoch: 17/100, batch 204/625 completed (loss: 0.0272:  33%|███▎      | 102/312 [10:19<21:09,  6.04s/it]
Training Epoch: 17/100, batch 204/625 completed (loss: 0.0098:  33%|███▎      | 102/312 [10:19<21:09,  6.04s/it]
Training Epoch: 17/100, batch 204/625 completed (loss: 0.0051:  33%|███▎      | 102/312 [10:19<21:08,  6.04s/it]
Training Epoch: 17/100, batch 204/625 completed (loss: 0.0298:  33%|███▎      | 103/312 [10:22<20:59,  6.02s/it]
Training Epoch: 17/100, batch 204/625 completed (loss: 0.0272:  33%|███▎      | 103/312 [10:22<20:59,  6.03s/it]
Training Epoch: 17/100, batch 204/625 completed (loss: 0.0098:  33%|███▎      | 103/312 [10:22<20:59,  6.02s/it]
Training Epoch: 17/100, batch 204/625 completed (loss: 0.0051:  33%|███▎      | 103/312 [10:22<20:59,  6.03s/it]
Training Epoch: 17/100, batch 205/625 completed (loss: 0.2822:  33%|███▎      | 103/312 [10:22<20:59,  6.02s/it]
Training Epoch: 17/100, batch 205/625 completed (loss: 0.0096:  33%|███▎      | 103/312 [10:22<20:59,  6.03s/it]
Training Epoch: 17/100, batch 205/625 completed (loss: 0.021:  33%|███▎      | 103/312 [10:22<20:59,  6.02s/it] 
Training Epoch: 17/100, batch 205/625 completed (loss: 0.0013:  33%|███▎      | 103/312 [10:22<20:59,  6.03s/it]
Training Epoch: 17/100, batch 206/625 completed (loss: 0.0282:  33%|███▎      | 103/312 [10:25<20:59,  6.02s/it]
Training Epoch: 17/100, batch 206/625 completed (loss: 0.0149:  33%|███▎      | 103/312 [10:25<20:59,  6.02s/it]
Training Epoch: 17/100, batch 206/625 completed (loss: 0.004:  33%|███▎      | 103/312 [10:25<20:59,  6.03s/it] 
Training Epoch: 17/100, batch 206/625 completed (loss: 0.0948:  33%|███▎      | 103/312 [10:25<20:59,  6.03s/it]
Training Epoch: 17/100, batch 206/625 completed (loss: 0.0282:  33%|███▎      | 104/312 [10:28<21:02,  6.07s/it]
Training Epoch: 17/100, batch 206/625 completed (loss: 0.004:  33%|███▎      | 104/312 [10:28<21:02,  6.07s/it]
Training Epoch: 17/100, batch 206/625 completed (loss: 0.0149:  33%|███▎      | 104/312 [10:28<21:02,  6.07s/it]
Training Epoch: 17/100, batch 206/625 completed (loss: 0.0948:  33%|███▎      | 104/312 [10:28<21:02,  6.07s/it]
Training Epoch: 17/100, batch 207/625 completed (loss: 0.0276:  33%|███▎      | 104/312 [10:28<21:02,  6.07s/it]
Training Epoch: 17/100, batch 207/625 completed (loss: 0.2279:  33%|███▎      | 104/312 [10:28<21:02,  6.07s/it]
Training Epoch: 17/100, batch 207/625 completed (loss: 0.0079:  33%|███▎      | 104/312 [10:28<21:02,  6.07s/it]
Training Epoch: 17/100, batch 207/625 completed (loss: 0.0056:  33%|███▎      | 104/312 [10:28<21:02,  6.07s/it]
Training Epoch: 17/100, batch 208/625 completed (loss: 0.0058:  33%|███▎      | 104/312 [10:31<21:02,  6.07s/it]
Training Epoch: 17/100, batch 208/625 completed (loss: 0.014:  33%|███▎      | 104/312 [10:31<21:02,  6.07s/it] 
Training Epoch: 17/100, batch 208/625 completed (loss: 0.017:  33%|███▎      | 104/312 [10:31<21:02,  6.07s/it] 
Training Epoch: 17/100, batch 208/625 completed (loss: 0.0522:  33%|███▎      | 104/312 [10:31<21:02,  6.07s/it]
Training Epoch: 17/100, batch 208/625 completed (loss: 0.0522:  34%|███▎      | 105/312 [10:34<20:52,  6.05s/it]
Training Epoch: 17/100, batch 208/625 completed (loss: 0.014:  34%|███▎      | 105/312 [10:34<20:53,  6.05s/it]
Training Epoch: 17/100, batch 208/625 completed (loss: 0.0058:  34%|███▎      | 105/312 [10:34<20:53,  6.05s/it]
Training Epoch: 17/100, batch 208/625 completed (loss: 0.017:  34%|███▎      | 105/312 [10:34<20:53,  6.06s/it]
Training Epoch: 17/100, batch 209/625 completed (loss: 0.0053:  34%|███▎      | 105/312 [10:34<20:53,  6.06s/it]
Training Epoch: 17/100, batch 209/625 completed (loss: 0.008:  34%|███▎      | 105/312 [10:34<20:53,  6.05s/it] 
Training Epoch: 17/100, batch 209/625 completed (loss: 0.0291:  34%|███▎      | 105/312 [10:34<20:53,  6.05s/it]
Training Epoch: 17/100, batch 209/625 completed (loss: 0.0156:  34%|███▎      | 105/312 [10:34<20:52,  6.05s/it]
Training Epoch: 17/100, batch 210/625 completed (loss: 0.0048:  34%|███▎      | 105/312 [10:37<20:53,  6.05s/it]
Training Epoch: 17/100, batch 210/625 completed (loss: 0.0116:  34%|███▎      | 105/312 [10:37<20:53,  6.06s/it]
Training Epoch: 17/100, batch 210/625 completed (loss: 0.0122:  34%|███▎      | 105/312 [10:37<20:53,  6.05s/it]
Training Epoch: 17/100, batch 210/625 completed (loss: 0.0036:  34%|███▎      | 105/312 [10:37<20:52,  6.05s/it]
Training Epoch: 17/100, batch 210/625 completed (loss: 0.0116:  34%|███▍      | 106/312 [10:40<20:33,  5.99s/it]
Training Epoch: 17/100, batch 210/625 completed (loss: 0.0036:  34%|███▍      | 106/312 [10:40<20:34,  5.99s/it]
Training Epoch: 17/100, batch 210/625 completed (loss: 0.0048:  34%|███▍      | 106/312 [10:40<20:34,  5.99s/it]
Training Epoch: 17/100, batch 210/625 completed (loss: 0.0122:  34%|███▍      | 106/312 [10:40<20:34,  5.99s/it]
Training Epoch: 17/100, batch 211/625 completed (loss: 0.0223:  34%|███▍      | 106/312 [10:40<20:34,  5.99s/it]
Training Epoch: 17/100, batch 211/625 completed (loss: 0.0187:  34%|███▍      | 106/312 [10:40<20:33,  5.99s/it]
Training Epoch: 17/100, batch 211/625 completed (loss: 0.0345:  34%|███▍      | 106/312 [10:40<20:34,  5.99s/it]
Training Epoch: 17/100, batch 211/625 completed (loss: 0.0083:  34%|███▍      | 106/312 [10:40<20:34,  5.99s/it]
Training Epoch: 17/100, batch 212/625 completed (loss: 0.0092:  34%|███▍      | 106/312 [10:43<20:34,  5.99s/it]
Training Epoch: 17/100, batch 212/625 completed (loss: 0.0491:  34%|███▍      | 106/312 [10:43<20:33,  5.99s/it]
Training Epoch: 17/100, batch 212/625 completed (loss: 0.0171:  34%|███▍      | 106/312 [10:43<20:34,  5.99s/it]
Training Epoch: 17/100, batch 212/625 completed (loss: 0.0062:  34%|███▍      | 106/312 [10:43<20:34,  5.99s/it]
Training Epoch: 17/100, batch 212/625 completed (loss: 0.0062:  34%|███▍      | 107/312 [10:46<20:58,  6.14s/it]
Training Epoch: 17/100, batch 212/625 completed (loss: 0.0171:  34%|███▍      | 107/312 [10:46<20:58,  6.14s/it]
Training Epoch: 17/100, batch 212/625 completed (loss: 0.0491:  34%|███▍      | 107/312 [10:46<21:00,  6.15s/it]
Training Epoch: 17/100, batch 212/625 completed (loss: 0.0092:  34%|███▍      | 107/312 [10:46<21:00,  6.15s/it]
Training Epoch: 17/100, batch 213/625 completed (loss: 0.0552:  34%|███▍      | 107/312 [10:46<20:58,  6.14s/it]
Training Epoch: 17/100, batch 213/625 completed (loss: 0.1598:  34%|███▍      | 107/312 [10:46<21:00,  6.15s/it]
Training Epoch: 17/100, batch 213/625 completed (loss: 0.0115:  34%|███▍      | 107/312 [10:46<21:00,  6.15s/it]
Training Epoch: 17/100, batch 213/625 completed (loss: 0.003:  34%|███▍      | 107/312 [10:46<20:58,  6.14s/it] 
Training Epoch: 17/100, batch 214/625 completed (loss: 0.008:  34%|███▍      | 107/312 [10:49<20:58,  6.14s/it] 
Training Epoch: 17/100, batch 214/625 completed (loss: 0.0042:  34%|███▍      | 107/312 [10:49<21:00,  6.15s/it]
Training Epoch: 17/100, batch 214/625 completed (loss: 0.1498:  34%|███▍      | 107/312 [10:49<21:00,  6.15s/it]
Training Epoch: 17/100, batch 214/625 completed (loss: 0.0214:  34%|███▍      | 107/312 [10:49<20:58,  6.14s/it]
Training Epoch: 17/100, batch 214/625 completed (loss: 0.0042:  35%|███▍      | 108/312 [10:52<20:34,  6.05s/it]
Training Epoch: 17/100, batch 214/625 completed (loss: 0.1498:  35%|███▍      | 108/312 [10:52<20:34,  6.05s/it]
Training Epoch: 17/100, batch 214/625 completed (loss: 0.008:  35%|███▍      | 108/312 [10:52<20:35,  6.06s/it]
Training Epoch: 17/100, batch 214/625 completed (loss: 0.0214:  35%|███▍      | 108/312 [10:52<20:36,  6.06s/it]
Training Epoch: 17/100, batch 215/625 completed (loss: 0.2756:  35%|███▍      | 108/312 [10:52<20:34,  6.05s/it]
Training Epoch: 17/100, batch 215/625 completed (loss: 0.0198:  35%|███▍      | 108/312 [10:52<20:34,  6.05s/it]
Training Epoch: 17/100, batch 215/625 completed (loss: 0.0089:  35%|███▍      | 108/312 [10:52<20:35,  6.06s/it]
Training Epoch: 17/100, batch 215/625 completed (loss: 0.0114:  35%|███▍      | 108/312 [10:52<20:36,  6.06s/it]
Training Epoch: 17/100, batch 216/625 completed (loss: 0.0112:  35%|███▍      | 108/312 [10:55<20:34,  6.05s/it]
Training Epoch: 17/100, batch 216/625 completed (loss: 0.008:  35%|███▍      | 108/312 [10:55<20:34,  6.05s/it] 
Training Epoch: 17/100, batch 216/625 completed (loss: 0.0195:  35%|███▍      | 108/312 [10:55<20:35,  6.06s/it]
Training Epoch: 17/100, batch 216/625 completed (loss: 0.0997:  35%|███▍      | 108/312 [10:55<20:36,  6.06s/it]
Training Epoch: 17/100, batch 216/625 completed (loss: 0.008:  35%|███▍      | 109/312 [10:58<20:35,  6.08s/it]
Training Epoch: 17/100, batch 216/625 completed (loss: 0.0195:  35%|███▍      | 109/312 [10:58<20:35,  6.09s/it]
Training Epoch: 17/100, batch 216/625 completed (loss: 0.0112:  35%|███▍      | 109/312 [10:58<20:35,  6.09s/it]
Training Epoch: 17/100, batch 216/625 completed (loss: 0.0997:  35%|███▍      | 109/312 [10:58<20:35,  6.09s/it]
Training Epoch: 17/100, batch 217/625 completed (loss: 0.0271:  35%|███▍      | 109/312 [10:58<20:35,  6.08s/it]
Training Epoch: 17/100, batch 217/625 completed (loss: 0.0018:  35%|███▍      | 109/312 [10:58<20:35,  6.09s/it]
Training Epoch: 17/100, batch 217/625 completed (loss: 0.0729:  35%|███▍      | 109/312 [10:58<20:35,  6.09s/it]
Training Epoch: 17/100, batch 217/625 completed (loss: 0.0035:  35%|███▍      | 109/312 [10:58<20:35,  6.09s/it]
Training Epoch: 17/100, batch 218/625 completed (loss: 0.0722:  35%|███▍      | 109/312 [11:02<20:35,  6.08s/it]
Training Epoch: 17/100, batch 218/625 completed (loss: 0.008:  35%|███▍      | 109/312 [11:02<20:35,  6.09s/it] 
Training Epoch: 17/100, batch 218/625 completed (loss: 0.0606:  35%|███▍      | 109/312 [11:02<20:35,  6.09s/it]
Training Epoch: 17/100, batch 218/625 completed (loss: 0.0329:  35%|███▍      | 109/312 [11:02<20:35,  6.09s/it]
Training Epoch: 17/100, batch 218/625 completed (loss: 0.0329:  35%|███▌      | 110/312 [11:05<20:36,  6.12s/it]
Training Epoch: 17/100, batch 218/625 completed (loss: 0.008:  35%|███▌      | 110/312 [11:05<20:36,  6.12s/it]
Training Epoch: 17/100, batch 218/625 completed (loss: 0.0606:  35%|███▌      | 110/312 [11:05<20:37,  6.12s/it]
Training Epoch: 17/100, batch 218/625 completed (loss: 0.0722:  35%|███▌      | 110/312 [11:05<20:37,  6.13s/it]
Training Epoch: 17/100, batch 219/625 completed (loss: 0.0523:  35%|███▌      | 110/312 [11:05<20:37,  6.13s/it]
Training Epoch: 17/100, batch 219/625 completed (loss: 0.0914:  35%|███▌      | 110/312 [11:05<20:37,  6.12s/it]
Training Epoch: 17/100, batch 219/625 completed (loss: 0.0011:  35%|███▌      | 110/312 [11:05<20:36,  6.12s/it]
Training Epoch: 17/100, batch 219/625 completed (loss: 0.1153:  35%|███▌      | 110/312 [11:05<20:36,  6.12s/it]
Training Epoch: 17/100, batch 220/625 completed (loss: 0.0067:  35%|███▌      | 110/312 [11:08<20:37,  6.13s/it]
Training Epoch: 17/100, batch 220/625 completed (loss: 0.0127:  35%|███▌      | 110/312 [11:08<20:36,  6.12s/it]
Training Epoch: 17/100, batch 220/625 completed (loss: 0.0044:  35%|███▌      | 110/312 [11:08<20:37,  6.12s/it]
Training Epoch: 17/100, batch 220/625 completed (loss: 0.0355:  35%|███▌      | 110/312 [11:08<20:36,  6.12s/it]
Training Epoch: 17/100, batch 220/625 completed (loss: 0.0127:  36%|███▌      | 111/312 [11:10<20:14,  6.04s/it]
Training Epoch: 17/100, batch 220/625 completed (loss: 0.0067:  36%|███▌      | 111/312 [11:10<20:14,  6.04s/it]
Training Epoch: 17/100, batch 220/625 completed (loss: 0.0044:  36%|███▌      | 111/312 [11:10<20:15,  6.05s/it]
Training Epoch: 17/100, batch 220/625 completed (loss: 0.0355:  36%|███▌      | 111/312 [11:11<20:15,  6.05s/it]
Training Epoch: 17/100, batch 221/625 completed (loss: 0.4211:  36%|███▌      | 111/312 [11:10<20:14,  6.04s/it]
Training Epoch: 17/100, batch 221/625 completed (loss: 0.1341:  36%|███▌      | 111/312 [11:10<20:14,  6.04s/it]
Training Epoch: 17/100, batch 221/625 completed (loss: 0.2353:  36%|███▌      | 111/312 [11:10<20:15,  6.05s/it]
Training Epoch: 17/100, batch 221/625 completed (loss: 0.245:  36%|███▌      | 111/312 [11:11<20:15,  6.05s/it] 
Training Epoch: 17/100, batch 222/625 completed (loss: 0.1893:  36%|███▌      | 111/312 [11:14<20:14,  6.04s/it]
Training Epoch: 17/100, batch 222/625 completed (loss: 0.4011:  36%|███▌      | 111/312 [11:14<20:14,  6.04s/it]
Training Epoch: 17/100, batch 222/625 completed (loss: 0.0699:  36%|███▌      | 111/312 [11:14<20:15,  6.05s/it]
Training Epoch: 17/100, batch 222/625 completed (loss: 0.0102:  36%|███▌      | 111/312 [11:14<20:15,  6.05s/it]
Training Epoch: 17/100, batch 222/625 completed (loss: 0.0102:  36%|███▌      | 112/312 [11:17<20:33,  6.17s/it]
Training Epoch: 17/100, batch 222/625 completed (loss: 0.1893:  36%|███▌      | 112/312 [11:17<20:33,  6.17s/it]
Training Epoch: 17/100, batch 222/625 completed (loss: 0.4011:  36%|███▌      | 112/312 [11:17<20:34,  6.17s/it]
Training Epoch: 17/100, batch 222/625 completed (loss: 0.0699:  36%|███▌      | 112/312 [11:17<20:33,  6.17s/it]
Training Epoch: 17/100, batch 223/625 completed (loss: 0.0592:  36%|███▌      | 112/312 [11:17<20:34,  6.17s/it]
Training Epoch: 17/100, batch 223/625 completed (loss: 0.1411:  36%|███▌      | 112/312 [11:17<20:33,  6.17s/it]
Training Epoch: 17/100, batch 223/625 completed (loss: 0.5106:  36%|███▌      | 112/312 [11:17<20:33,  6.17s/it]
Training Epoch: 17/100, batch 223/625 completed (loss: 0.0191:  36%|███▌      | 112/312 [11:17<20:33,  6.17s/it]
Training Epoch: 17/100, batch 224/625 completed (loss: 0.0595:  36%|███▌      | 112/312 [11:20<20:34,  6.17s/it]
Training Epoch: 17/100, batch 224/625 completed (loss: 0.013:  36%|███▌      | 112/312 [11:20<20:33,  6.17s/it] 
Training Epoch: 17/100, batch 224/625 completed (loss: 0.0148:  36%|███▌      | 112/312 [11:20<20:33,  6.17s/it]
Training Epoch: 17/100, batch 224/625 completed (loss: 0.1771:  36%|███▌      | 112/312 [11:20<20:33,  6.17s/it]
Training Epoch: 17/100, batch 224/625 completed (loss: 0.1771:  36%|███▌      | 113/312 [11:23<20:08,  6.07s/it]
Training Epoch: 17/100, batch 224/625 completed (loss: 0.0595:  36%|███▌      | 113/312 [11:23<20:08,  6.07s/it]
Training Epoch: 17/100, batch 224/625 completed (loss: 0.013:  36%|███▌      | 113/312 [11:23<20:08,  6.08s/it]
Training Epoch: 17/100, batch 224/625 completed (loss: 0.0148:  36%|███▌      | 113/312 [11:23<20:09,  6.08s/it]
Training Epoch: 17/100, batch 225/625 completed (loss: 0.0263:  36%|███▌      | 113/312 [11:23<20:09,  6.08s/it]
Training Epoch: 17/100, batch 225/625 completed (loss: 0.0007:  36%|███▌      | 113/312 [11:23<20:08,  6.08s/it]
Training Epoch: 17/100, batch 225/625 completed (loss: 0.0025:  36%|███▌      | 113/312 [11:23<20:08,  6.07s/it]
Training Epoch: 17/100, batch 225/625 completed (loss: 0.0164:  36%|███▌      | 113/312 [11:23<20:08,  6.07s/it]
Training Epoch: 17/100, batch 226/625 completed (loss: 0.004:  36%|███▌      | 113/312 [11:26<20:08,  6.07s/it] 
Training Epoch: 17/100, batch 226/625 completed (loss: 0.1739:  36%|███▌      | 113/312 [11:26<20:08,  6.08s/it]
Training Epoch: 17/100, batch 226/625 completed (loss: 0.276:  36%|███▌      | 113/312 [11:26<20:09,  6.08s/it] 
Training Epoch: 17/100, batch 226/625 completed (loss: 0.0059:  36%|███▌      | 113/312 [11:26<20:08,  6.07s/it]
Training Epoch: 17/100, batch 226/625 completed (loss: 0.0059:  37%|███▋      | 114/312 [11:29<19:58,  6.05s/it]
Training Epoch: 17/100, batch 226/625 completed (loss: 0.004:  37%|███▋      | 114/312 [11:29<19:58,  6.05s/it]
Training Epoch: 17/100, batch 226/625 completed (loss: 0.276:  37%|███▋      | 114/312 [11:29<19:57,  6.05s/it]
Training Epoch: 17/100, batch 226/625 completed (loss: 0.1739:  37%|███▋      | 114/312 [11:29<19:58,  6.05s/it]
Training Epoch: 17/100, batch 227/625 completed (loss: 0.6235:  37%|███▋      | 114/312 [11:29<19:58,  6.05s/it]
Training Epoch: 17/100, batch 227/625 completed (loss: 0.0021:  37%|███▋      | 114/312 [11:29<19:57,  6.05s/it]
Training Epoch: 17/100, batch 227/625 completed (loss: 0.0366:  37%|███▋      | 114/312 [11:29<19:58,  6.05s/it]
Training Epoch: 17/100, batch 227/625 completed (loss: 0.0366:  37%|███▋      | 114/312 [11:29<19:58,  6.05s/it]
Training Epoch: 17/100, batch 228/625 completed (loss: 0.0027:  37%|███▋      | 114/312 [11:32<19:57,  6.05s/it]
Training Epoch: 17/100, batch 228/625 completed (loss: 0.0208:  37%|███▋      | 114/312 [11:32<19:58,  6.05s/it]
Training Epoch: 17/100, batch 228/625 completed (loss: 0.2185:  37%|███▋      | 114/312 [11:32<19:58,  6.05s/it]
Training Epoch: 17/100, batch 228/625 completed (loss: 0.0196:  37%|███▋      | 114/312 [11:32<19:58,  6.05s/it]
Training Epoch: 17/100, batch 228/625 completed (loss: 0.0196:  37%|███▋      | 115/312 [11:35<20:08,  6.13s/it]
Training Epoch: 17/100, batch 228/625 completed (loss: 0.0027:  37%|███▋      | 115/312 [11:35<20:07,  6.13s/it]
Training Epoch: 17/100, batch 228/625 completed (loss: 0.0208:  37%|███▋      | 115/312 [11:35<20:08,  6.14s/it]
Training Epoch: 17/100, batch 228/625 completed (loss: 0.2185:  37%|███▋      | 115/312 [11:35<20:08,  6.13s/it]
Training Epoch: 17/100, batch 229/625 completed (loss: 0.1253:  37%|███▋      | 115/312 [11:35<20:07,  6.13s/it]
Training Epoch: 17/100, batch 229/625 completed (loss: 0.0042:  37%|███▋      | 115/312 [11:35<20:08,  6.14s/it]
Training Epoch: 17/100, batch 229/625 completed (loss: 0.0507:  37%|███▋      | 115/312 [11:35<20:08,  6.13s/it]
Training Epoch: 17/100, batch 229/625 completed (loss: 0.0056:  37%|███▋      | 115/312 [11:35<20:08,  6.13s/it]
Training Epoch: 17/100, batch 230/625 completed (loss: 0.0475:  37%|███▋      | 115/312 [11:38<20:08,  6.13s/it]
Training Epoch: 17/100, batch 230/625 completed (loss: 0.014:  37%|███▋      | 115/312 [11:38<20:07,  6.13s/it] 
Training Epoch: 17/100, batch 230/625 completed (loss: 0.1608:  37%|███▋      | 115/312 [11:38<20:08,  6.14s/it]
Training Epoch: 17/100, batch 230/625 completed (loss: 0.0633:  37%|███▋      | 115/312 [11:38<20:08,  6.13s/it]
Training Epoch: 17/100, batch 230/625 completed (loss: 0.1608:  37%|███▋      | 116/312 [11:41<19:44,  6.04s/it]
Training Epoch: 17/100, batch 230/625 completed (loss: 0.0633:  37%|███▋      | 116/312 [11:41<19:45,  6.05s/it]
Training Epoch: 17/100, batch 230/625 completed (loss: 0.0475:  37%|███▋      | 116/312 [11:41<19:45,  6.05s/it]
Training Epoch: 17/100, batch 230/625 completed (loss: 0.014:  37%|███▋      | 116/312 [11:41<19:45,  6.05s/it]
Training Epoch: 17/100, batch 231/625 completed (loss: 0.0875:  37%|███▋      | 116/312 [11:41<19:45,  6.05s/it]
Training Epoch: 17/100, batch 231/625 completed (loss: 0.0654:  37%|███▋      | 116/312 [11:41<19:44,  6.04s/it]
Training Epoch: 17/100, batch 231/625 completed (loss: 0.0205:  37%|███▋      | 116/312 [11:41<19:45,  6.05s/it]
Training Epoch: 17/100, batch 231/625 completed (loss: 0.0052:  37%|███▋      | 116/312 [11:41<19:45,  6.05s/it]
Training Epoch: 17/100, batch 232/625 completed (loss: 0.4723:  37%|███▋      | 116/312 [11:44<19:45,  6.05s/it]
Training Epoch: 17/100, batch 232/625 completed (loss: 0.06:  37%|███▋      | 116/312 [11:44<19:44,  6.04s/it]  
Training Epoch: 17/100, batch 232/625 completed (loss: 0.0101:  37%|███▋      | 116/312 [11:44<19:45,  6.05s/it]
Training Epoch: 17/100, batch 232/625 completed (loss: 0.1443:  37%|███▋      | 116/312 [11:44<19:45,  6.05s/it]
Training Epoch: 17/100, batch 232/625 completed (loss: 0.06:  38%|███▊      | 117/312 [11:47<19:45,  6.08s/it]
Training Epoch: 17/100, batch 232/625 completed (loss: 0.1443:  38%|███▊      | 117/312 [11:47<19:44,  6.08s/it]
Training Epoch: 17/100, batch 232/625 completed (loss: 0.4723:  38%|███▊      | 117/312 [11:47<19:44,  6.07s/it]
Training Epoch: 17/100, batch 232/625 completed (loss: 0.0101:  38%|███▊      | 117/312 [11:47<19:44,  6.08s/it]
Training Epoch: 17/100, batch 233/625 completed (loss: 0.0069:  38%|███▊      | 117/312 [11:47<19:44,  6.08s/it]
Training Epoch: 17/100, batch 233/625 completed (loss: 0.0332:  38%|███▊      | 117/312 [11:47<19:45,  6.08s/it]
Training Epoch: 17/100, batch 233/625 completed (loss: 0.0088:  38%|███▊      | 117/312 [11:47<19:44,  6.07s/it]
Training Epoch: 17/100, batch 233/625 completed (loss: 0.3735:  38%|███▊      | 117/312 [11:47<19:44,  6.08s/it]
Training Epoch: 17/100, batch 234/625 completed (loss: 0.0879:  38%|███▊      | 117/312 [11:50<19:45,  6.08s/it]
Training Epoch: 17/100, batch 234/625 completed (loss: 0.0137:  38%|███▊      | 117/312 [11:50<19:44,  6.08s/it]
Training Epoch: 17/100, batch 234/625 completed (loss: 0.032:  38%|███▊      | 117/312 [11:50<19:44,  6.07s/it] 
Training Epoch: 17/100, batch 234/625 completed (loss: 0.0922:  38%|███▊      | 117/312 [11:50<19:44,  6.08s/it]
Training Epoch: 17/100, batch 234/625 completed (loss: 0.0879:  38%|███▊      | 118/312 [11:53<19:37,  6.07s/it]
Training Epoch: 17/100, batch 234/625 completed (loss: 0.032:  38%|███▊      | 118/312 [11:53<19:37,  6.07s/it]
Training Epoch: 17/100, batch 234/625 completed (loss: 0.0922:  38%|███▊      | 118/312 [11:53<19:38,  6.07s/it]
Training Epoch: 17/100, batch 234/625 completed (loss: 0.0137:  38%|███▊      | 118/312 [11:53<19:38,  6.07s/it]
Training Epoch: 17/100, batch 235/625 completed (loss: 0.025:  38%|███▊      | 118/312 [11:53<19:37,  6.07s/it] 
Training Epoch: 17/100, batch 235/625 completed (loss: 0.0077:  38%|███▊      | 118/312 [11:53<19:37,  6.07s/it]
Training Epoch: 17/100, batch 235/625 completed (loss: 0.0211:  38%|███▊      | 118/312 [11:53<19:38,  6.07s/it]
Training Epoch: 17/100, batch 235/625 completed (loss: 0.0133:  38%|███▊      | 118/312 [11:53<19:38,  6.07s/it]
Training Epoch: 17/100, batch 236/625 completed (loss: 0.1573:  38%|███▊      | 118/312 [11:56<19:37,  6.07s/it]
Training Epoch: 17/100, batch 236/625 completed (loss: 0.0161:  38%|███▊      | 118/312 [11:56<19:37,  6.07s/it]
Training Epoch: 17/100, batch 236/625 completed (loss: 0.1184:  38%|███▊      | 118/312 [11:56<19:38,  6.07s/it]
Training Epoch: 17/100, batch 236/625 completed (loss: 0.3353:  38%|███▊      | 118/312 [11:56<19:38,  6.07s/it]
Training Epoch: 17/100, batch 236/625 completed (loss: 0.0161:  38%|███▊      | 119/312 [11:59<19:26,  6.04s/it]
Training Epoch: 17/100, batch 236/625 completed (loss: 0.1573:  38%|███▊      | 119/312 [11:59<19:25,  6.04s/it]
Training Epoch: 17/100, batch 236/625 completed (loss: 0.1184:  38%|███▊      | 119/312 [11:59<19:25,  6.04s/it]
Training Epoch: 17/100, batch 236/625 completed (loss: 0.3353:  38%|███▊      | 119/312 [11:59<19:26,  6.04s/it]
Training Epoch: 17/100, batch 237/625 completed (loss: 0.0193:  38%|███▊      | 119/312 [11:59<19:25,  6.04s/it]
Training Epoch: 17/100, batch 237/625 completed (loss: 0.0155:  38%|███▊      | 119/312 [11:59<19:25,  6.04s/it]
Training Epoch: 17/100, batch 237/625 completed (loss: 0.0119:  38%|███▊      | 119/312 [11:59<19:26,  6.04s/it]
Training Epoch: 17/100, batch 237/625 completed (loss: 0.0044:  38%|███▊      | 119/312 [11:59<19:26,  6.04s/it]
Training Epoch: 17/100, batch 238/625 completed (loss: 0.0803:  38%|███▊      | 119/312 [12:02<19:25,  6.04s/it]
Training Epoch: 17/100, batch 238/625 completed (loss: 0.0042:  38%|███▊      | 119/312 [12:02<19:25,  6.04s/it]
Training Epoch: 17/100, batch 238/625 completed (loss: 0.1004:  38%|███▊      | 119/312 [12:02<19:26,  6.04s/it]
Training Epoch: 17/100, batch 238/625 completed (loss: 0.0075:  38%|███▊      | 119/312 [12:02<19:26,  6.04s/it]
Training Epoch: 17/100, batch 238/625 completed (loss: 0.0042:  38%|███▊      | 120/312 [12:05<19:24,  6.07s/it]
Training Epoch: 17/100, batch 238/625 completed (loss: 0.0075:  38%|███▊      | 120/312 [12:05<19:25,  6.07s/it]
Training Epoch: 17/100, batch 238/625 completed (loss: 0.0803:  38%|███▊      | 120/312 [12:05<19:25,  6.07s/it]
Training Epoch: 17/100, batch 238/625 completed (loss: 0.1004:  38%|███▊      | 120/312 [12:05<19:25,  6.07s/it]
Training Epoch: 17/100, batch 239/625 completed (loss: 0.0409:  38%|███▊      | 120/312 [12:05<19:24,  6.07s/it]
Training Epoch: 17/100, batch 239/625 completed (loss: 0.0157:  38%|███▊      | 120/312 [12:05<19:25,  6.07s/it]
Training Epoch: 17/100, batch 239/625 completed (loss: 0.1103:  38%|███▊      | 120/312 [12:05<19:25,  6.07s/it]
Training Epoch: 17/100, batch 239/625 completed (loss: 0.0092:  38%|███▊      | 120/312 [12:05<19:25,  6.07s/it]
Training Epoch: 17/100, batch 240/625 completed (loss: 0.0629:  38%|███▊      | 120/312 [12:08<19:25,  6.07s/it]
Training Epoch: 17/100, batch 240/625 completed (loss: 0.0192:  38%|███▊      | 120/312 [12:08<19:25,  6.07s/it]
Training Epoch: 17/100, batch 240/625 completed (loss: 0.0405:  38%|███▊      | 120/312 [12:08<19:24,  6.07s/it]
Training Epoch: 17/100, batch 240/625 completed (loss: 0.0127:  38%|███▊      | 120/312 [12:08<19:25,  6.07s/it]
Training Epoch: 17/100, batch 240/625 completed (loss: 0.0405:  39%|███▉      | 121/312 [12:11<19:14,  6.04s/it]
Training Epoch: 17/100, batch 240/625 completed (loss: 0.0192:  39%|███▉      | 121/312 [12:11<19:14,  6.04s/it]
Training Epoch: 17/100, batch 240/625 completed (loss: 0.0629:  39%|███▉      | 121/312 [12:11<19:14,  6.04s/it]
Training Epoch: 17/100, batch 240/625 completed (loss: 0.0127:  39%|███▉      | 121/312 [12:11<19:14,  6.04s/it]
Training Epoch: 17/100, batch 241/625 completed (loss: 0.101:  39%|███▉      | 121/312 [12:11<19:14,  6.04s/it] 
Training Epoch: 17/100, batch 241/625 completed (loss: 0.0144:  39%|███▉      | 121/312 [12:11<19:14,  6.04s/it]
Training Epoch: 17/100, batch 241/625 completed (loss: 0.0236:  39%|███▉      | 121/312 [12:11<19:14,  6.04s/it]
Training Epoch: 17/100, batch 241/625 completed (loss: 0.0268:  39%|███▉      | 121/312 [12:11<19:14,  6.04s/it]
Training Epoch: 17/100, batch 242/625 completed (loss: 0.4464:  39%|███▉      | 121/312 [12:14<19:14,  6.04s/it]
Training Epoch: 17/100, batch 242/625 completed (loss: 0.0209:  39%|███▉      | 121/312 [12:14<19:14,  6.04s/it]
Training Epoch: 17/100, batch 242/625 completed (loss: 0.0057:  39%|███▉      | 121/312 [12:14<19:14,  6.04s/it]
Training Epoch: 17/100, batch 242/625 completed (loss: 0.3263:  39%|███▉      | 121/312 [12:14<19:14,  6.04s/it]
Training Epoch: 17/100, batch 242/625 completed (loss: 0.3263:  39%|███▉      | 122/312 [12:17<19:14,  6.07s/it]
Training Epoch: 17/100, batch 242/625 completed (loss: 0.0057:  39%|███▉      | 122/312 [12:17<19:14,  6.08s/it]
Training Epoch: 17/100, batch 242/625 completed (loss: 0.0209:  39%|███▉      | 122/312 [12:17<19:14,  6.08s/it]
Training Epoch: 17/100, batch 242/625 completed (loss: 0.4464:  39%|███▉      | 122/312 [12:17<19:15,  6.08s/it]
Training Epoch: 17/100, batch 243/625 completed (loss: 0.0231:  39%|███▉      | 122/312 [12:17<19:14,  6.08s/it]
Training Epoch: 17/100, batch 243/625 completed (loss: 0.1606:  39%|███▉      | 122/312 [12:17<19:15,  6.08s/it]
Training Epoch: 17/100, batch 243/625 completed (loss: 0.0145:  39%|███▉      | 122/312 [12:17<19:14,  6.08s/it]
Training Epoch: 17/100, batch 243/625 completed (loss: 0.0072:  39%|███▉      | 122/312 [12:17<19:14,  6.07s/it]
Training Epoch: 17/100, batch 244/625 completed (loss: 0.0269:  39%|███▉      | 122/312 [12:21<19:14,  6.08s/it]
Training Epoch: 17/100, batch 244/625 completed (loss: 0.0643:  39%|███▉      | 122/312 [12:21<19:15,  6.08s/it]
Training Epoch: 17/100, batch 244/625 completed (loss: 0.0026:  39%|███▉      | 122/312 [12:21<19:14,  6.08s/it]
Training Epoch: 17/100, batch 244/625 completed (loss: 0.1534:  39%|███▉      | 122/312 [12:21<19:14,  6.07s/it]
Training Epoch: 17/100, batch 244/625 completed (loss: 0.0643:  39%|███▉      | 123/312 [12:24<19:15,  6.11s/it]
Training Epoch: 17/100, batch 244/625 completed (loss: 0.1534:  39%|███▉      | 123/312 [12:24<19:15,  6.12s/it]
Training Epoch: 17/100, batch 244/625 completed (loss: 0.0269:  39%|███▉      | 123/312 [12:24<19:16,  6.12s/it]
Training Epoch: 17/100, batch 244/625 completed (loss: 0.0026:  39%|███▉      | 123/312 [12:24<19:16,  6.12s/it]
Training Epoch: 17/100, batch 245/625 completed (loss: 0.0739:  39%|███▉      | 123/312 [12:24<19:15,  6.11s/it]
Training Epoch: 17/100, batch 245/625 completed (loss: 0.0095:  39%|███▉      | 123/312 [12:24<19:16,  6.12s/it]
Training Epoch: 17/100, batch 245/625 completed (loss: 0.2692:  39%|███▉      | 123/312 [12:24<19:16,  6.12s/it]
Training Epoch: 17/100, batch 245/625 completed (loss: 0.0099:  39%|███▉      | 123/312 [12:24<19:15,  6.12s/it]
Training Epoch: 17/100, batch 246/625 completed (loss: 0.0733:  39%|███▉      | 123/312 [12:27<19:15,  6.11s/it]
Training Epoch: 17/100, batch 246/625 completed (loss: 0.1502:  39%|███▉      | 123/312 [12:27<19:16,  6.12s/it]
Training Epoch: 17/100, batch 246/625 completed (loss: 0.0555:  39%|███▉      | 123/312 [12:27<19:16,  6.12s/it]
Training Epoch: 17/100, batch 246/625 completed (loss: 0.0194:  39%|███▉      | 123/312 [12:27<19:15,  6.12s/it]
Training Epoch: 17/100, batch 246/625 completed (loss: 0.0194:  40%|███▉      | 124/312 [12:30<19:01,  6.07s/it]
Training Epoch: 17/100, batch 246/625 completed (loss: 0.0733:  40%|███▉      | 124/312 [12:30<19:01,  6.07s/it]
Training Epoch: 17/100, batch 246/625 completed (loss: 0.0555:  40%|███▉      | 124/312 [12:30<19:02,  6.07s/it]
Training Epoch: 17/100, batch 246/625 completed (loss: 0.1502:  40%|███▉      | 124/312 [12:30<19:02,  6.08s/it]
Training Epoch: 17/100, batch 247/625 completed (loss: 0.1177:  40%|███▉      | 124/312 [12:30<19:02,  6.08s/it]
Training Epoch: 17/100, batch 247/625 completed (loss: 0.0066:  40%|███▉      | 124/312 [12:30<19:02,  6.07s/it]
Training Epoch: 17/100, batch 247/625 completed (loss: 0.0363:  40%|███▉      | 124/312 [12:30<19:01,  6.07s/it]
Training Epoch: 17/100, batch 247/625 completed (loss: 0.0688:  40%|███▉      | 124/312 [12:30<19:01,  6.07s/it]
Training Epoch: 17/100, batch 248/625 completed (loss: 0.0362:  40%|███▉      | 124/312 [12:33<19:02,  6.07s/it]
Training Epoch: 17/100, batch 248/625 completed (loss: 0.03:  40%|███▉      | 124/312 [12:33<19:01,  6.07s/it]  
Training Epoch: 17/100, batch 248/625 completed (loss: 0.0132:  40%|███▉      | 124/312 [12:33<19:02,  6.08s/it]
Training Epoch: 17/100, batch 248/625 completed (loss: 0.1662:  40%|███▉      | 124/312 [12:33<19:01,  6.07s/it]
Training Epoch: 17/100, batch 248/625 completed (loss: 0.03:  40%|████      | 125/312 [12:36<18:49,  6.04s/it]
Training Epoch: 17/100, batch 248/625 completed (loss: 0.1662:  40%|████      | 125/312 [12:36<18:50,  6.04s/it]
Training Epoch: 17/100, batch 248/625 completed (loss: 0.0132:  40%|████      | 125/312 [12:36<18:50,  6.04s/it]
Training Epoch: 17/100, batch 248/625 completed (loss: 0.0362:  40%|████      | 125/312 [12:36<18:50,  6.05s/it]
Training Epoch: 17/100, batch 249/625 completed (loss: 0.0375:  40%|████      | 125/312 [12:36<18:50,  6.05s/it]
Training Epoch: 17/100, batch 249/625 completed (loss: 0.0043:  40%|████      | 125/312 [12:36<18:50,  6.04s/it]
Training Epoch: 17/100, batch 249/625 completed (loss: 0.0292:  40%|████      | 125/312 [12:36<18:49,  6.04s/it]
Training Epoch: 17/100, batch 249/625 completed (loss: 0.0495:  40%|████      | 125/312 [12:36<18:50,  6.04s/it]
Training Epoch: 17/100, batch 250/625 completed (loss: 0.0567:  40%|████      | 125/312 [12:39<18:50,  6.05s/it]
Training Epoch: 17/100, batch 250/625 completed (loss: 0.0097:  40%|████      | 125/312 [12:39<18:50,  6.04s/it]
Training Epoch: 17/100, batch 250/625 completed (loss: 0.0058:  40%|████      | 125/312 [12:39<18:49,  6.04s/it]
Training Epoch: 17/100, batch 250/625 completed (loss: 0.0869:  40%|████      | 125/312 [12:39<18:50,  6.04s/it]
Training Epoch: 17/100, batch 250/625 completed (loss: 0.0058:  40%|████      | 126/312 [12:42<18:48,  6.07s/it]
Training Epoch: 17/100, batch 250/625 completed (loss: 0.0869:  40%|████      | 126/312 [12:42<18:49,  6.07s/it]
Training Epoch: 17/100, batch 250/625 completed (loss: 0.0567:  40%|████      | 126/312 [12:42<18:48,  6.07s/it]
Training Epoch: 17/100, batch 250/625 completed (loss: 0.0097:  40%|████      | 126/312 [12:42<18:49,  6.07s/it]
Training Epoch: 17/100, batch 251/625 completed (loss: 0.3459:  40%|████      | 126/312 [12:42<18:49,  6.07s/it]
Training Epoch: 17/100, batch 251/625 completed (loss: 0.0055:  40%|████      | 126/312 [12:42<18:48,  6.07s/it]
Training Epoch: 17/100, batch 251/625 completed (loss: 0.0134:  40%|████      | 126/312 [12:42<18:48,  6.07s/it]
Training Epoch: 17/100, batch 251/625 completed (loss: 0.3753:  40%|████      | 126/312 [12:42<18:49,  6.07s/it]
Training Epoch: 17/100, batch 252/625 completed (loss: 0.0738:  40%|████      | 126/312 [12:45<18:49,  6.07s/it]
Training Epoch: 17/100, batch 252/625 completed (loss: 0.0025:  40%|████      | 126/312 [12:45<18:48,  6.07s/it]
Training Epoch: 17/100, batch 252/625 completed (loss: 0.0612:  40%|████      | 126/312 [12:45<18:48,  6.07s/it]
Training Epoch: 17/100, batch 252/625 completed (loss: 0.0181:  40%|████      | 126/312 [12:45<18:49,  6.07s/it]
Training Epoch: 17/100, batch 252/625 completed (loss: 0.0025:  41%|████      | 127/312 [12:48<18:37,  6.04s/it]
Training Epoch: 17/100, batch 252/625 completed (loss: 0.0738:  41%|████      | 127/312 [12:48<18:37,  6.04s/it]
Training Epoch: 17/100, batch 252/625 completed (loss: 0.0181:  41%|████      | 127/312 [12:48<18:38,  6.05s/it]
Training Epoch: 17/100, batch 252/625 completed (loss: 0.0612:  41%|████      | 127/312 [12:48<18:38,  6.05s/it]
Training Epoch: 17/100, batch 253/625 completed (loss: 0.0395:  41%|████      | 127/312 [12:48<18:37,  6.04s/it]
Training Epoch: 17/100, batch 253/625 completed (loss: 0.091:  41%|████      | 127/312 [12:48<18:37,  6.04s/it] 
Training Epoch: 17/100, batch 253/625 completed (loss: 0.023:  41%|████      | 127/312 [12:48<18:38,  6.05s/it] 
Training Epoch: 17/100, batch 253/625 completed (loss: 0.0164:  41%|████      | 127/312 [12:48<18:38,  6.05s/it]
Training Epoch: 17/100, batch 254/625 completed (loss: 0.0476:  41%|████      | 127/312 [12:51<18:37,  6.04s/it]
Training Epoch: 17/100, batch 254/625 completed (loss: 0.0405:  41%|████      | 127/312 [12:51<18:37,  6.04s/it]
Training Epoch: 17/100, batch 254/625 completed (loss: 0.0163:  41%|████      | 127/312 [12:51<18:38,  6.05s/it]
Training Epoch: 17/100, batch 254/625 completed (loss: 0.0407:  41%|████      | 127/312 [12:51<18:38,  6.05s/it]
Training Epoch: 17/100, batch 254/625 completed (loss: 0.0476:  41%|████      | 128/312 [12:54<18:37,  6.07s/it]
Training Epoch: 17/100, batch 254/625 completed (loss: 0.0163:  41%|████      | 128/312 [12:54<18:37,  6.07s/it]
Training Epoch: 17/100, batch 254/625 completed (loss: 0.0407:  41%|████      | 128/312 [12:54<18:37,  6.07s/it]
Training Epoch: 17/100, batch 254/625 completed (loss: 0.0405:  41%|████      | 128/312 [12:54<18:38,  6.08s/it]
Training Epoch: 17/100, batch 255/625 completed (loss: 0.0211:  41%|████      | 128/312 [12:54<18:38,  6.08s/it]
Training Epoch: 17/100, batch 255/625 completed (loss: 0.0139:  41%|████      | 128/312 [12:54<18:37,  6.07s/it]
Training Epoch: 17/100, batch 255/625 completed (loss: 0.0505:  41%|████      | 128/312 [12:54<18:37,  6.07s/it]
Training Epoch: 17/100, batch 255/625 completed (loss: 0.0116:  41%|████      | 128/312 [12:54<18:37,  6.07s/it]
Training Epoch: 17/100, batch 256/625 completed (loss: 0.0273:  41%|████      | 128/312 [12:57<18:37,  6.07s/it]
Training Epoch: 17/100, batch 256/625 completed (loss: 0.0113:  41%|████      | 128/312 [12:57<18:38,  6.08s/it]
Training Epoch: 17/100, batch 256/625 completed (loss: 0.0021:  41%|████      | 128/312 [12:57<18:37,  6.07s/it]
Training Epoch: 17/100, batch 256/625 completed (loss: 0.019:  41%|████      | 128/312 [12:57<18:37,  6.07s/it] 
Training Epoch: 17/100, batch 256/625 completed (loss: 0.0021:  41%|████▏     | 129/312 [13:00<18:27,  6.05s/it]
Training Epoch: 17/100, batch 256/625 completed (loss: 0.019:  41%|████▏     | 129/312 [13:00<18:28,  6.06s/it]
Training Epoch: 17/100, batch 256/625 completed (loss: 0.0113:  41%|████▏     | 129/312 [13:00<18:27,  6.05s/it]
Training Epoch: 17/100, batch 256/625 completed (loss: 0.0273:  41%|████▏     | 129/312 [13:00<18:28,  6.06s/it]
Training Epoch: 17/100, batch 257/625 completed (loss: 0.0217:  41%|████▏     | 129/312 [13:00<18:28,  6.06s/it]
Training Epoch: 17/100, batch 257/625 completed (loss: 0.0753:  41%|████▏     | 129/312 [13:00<18:27,  6.05s/it]
Training Epoch: 17/100, batch 257/625 completed (loss: 0.0071:  41%|████▏     | 129/312 [13:00<18:27,  6.05s/it]
Training Epoch: 17/100, batch 257/625 completed (loss: 0.0068:  41%|████▏     | 129/312 [13:00<18:28,  6.06s/it]
Training Epoch: 17/100, batch 258/625 completed (loss: 0.0612:  41%|████▏     | 129/312 [13:03<18:27,  6.05s/it]
Training Epoch: 17/100, batch 258/625 completed (loss: 0.0176:  41%|████▏     | 129/312 [13:03<18:27,  6.05s/it]
Training Epoch: 17/100, batch 258/625 completed (loss: 0.0718:  41%|████▏     | 129/312 [13:03<18:28,  6.06s/it]
Training Epoch: 17/100, batch 258/625 completed (loss: 0.0048:  41%|████▏     | 129/312 [13:03<18:28,  6.06s/it]
Training Epoch: 17/100, batch 258/625 completed (loss: 0.0718:  42%|████▏     | 130/312 [13:06<18:26,  6.08s/it]
Training Epoch: 17/100, batch 258/625 completed (loss: 0.0176:  42%|████▏     | 130/312 [13:06<18:26,  6.08s/it]
Training Epoch: 17/100, batch 258/625 completed (loss: 0.0048:  42%|████▏     | 130/312 [13:06<18:26,  6.08s/it]
Training Epoch: 17/100, batch 258/625 completed (loss: 0.0612:  42%|████▏     | 130/312 [13:06<18:26,  6.08s/it]
Training Epoch: 17/100, batch 259/625 completed (loss: 0.0258:  42%|████▏     | 130/312 [13:06<18:26,  6.08s/it]
Training Epoch: 17/100, batch 259/625 completed (loss: 0.1435:  42%|████▏     | 130/312 [13:06<18:26,  6.08s/it]
Training Epoch: 17/100, batch 259/625 completed (loss: 0.0724:  42%|████▏     | 130/312 [13:06<18:26,  6.08s/it]
Training Epoch: 17/100, batch 259/625 completed (loss: 0.0097:  42%|████▏     | 130/312 [13:06<18:26,  6.08s/it]
Training Epoch: 17/100, batch 260/625 completed (loss: 0.0208:  42%|████▏     | 130/312 [13:09<18:26,  6.08s/it]
Training Epoch: 17/100, batch 260/625 completed (loss: 0.0006:  42%|████▏     | 130/312 [13:09<18:26,  6.08s/it]
Training Epoch: 17/100, batch 260/625 completed (loss: 0.0188:  42%|████▏     | 130/312 [13:09<18:26,  6.08s/it]
Training Epoch: 17/100, batch 260/625 completed (loss: 0.0185:  42%|████▏     | 130/312 [13:09<18:26,  6.08s/it]
Training Epoch: 17/100, batch 260/625 completed (loss: 0.0208:  42%|████▏     | 131/312 [13:12<18:23,  6.10s/it]
Training Epoch: 17/100, batch 260/625 completed (loss: 0.0185:  42%|████▏     | 131/312 [13:12<18:24,  6.10s/it]
Training Epoch: 17/100, batch 260/625 completed (loss: 0.0006:  42%|████▏     | 131/312 [13:12<18:24,  6.10s/it]
Training Epoch: 17/100, batch 260/625 completed (loss: 0.0188:  42%|████▏     | 131/312 [13:12<18:24,  6.10s/it]
Training Epoch: 17/100, batch 261/625 completed (loss: 0.0055:  42%|████▏     | 131/312 [13:12<18:24,  6.10s/it]
Training Epoch: 17/100, batch 261/625 completed (loss: 0.0599:  42%|████▏     | 131/312 [13:12<18:23,  6.10s/it]
Training Epoch: 17/100, batch 261/625 completed (loss: 0.0876:  42%|████▏     | 131/312 [13:12<18:24,  6.10s/it]
Training Epoch: 17/100, batch 261/625 completed (loss: 0.1896:  42%|████▏     | 131/312 [13:12<18:24,  6.10s/it]
Training Epoch: 17/100, batch 262/625 completed (loss: 0.1065:  42%|████▏     | 131/312 [13:15<18:24,  6.10s/it]
Training Epoch: 17/100, batch 262/625 completed (loss: 0.0268:  42%|████▏     | 131/312 [13:15<18:24,  6.10s/it]
Training Epoch: 17/100, batch 262/625 completed (loss: 0.0529:  42%|████▏     | 131/312 [13:15<18:23,  6.10s/it]
Training Epoch: 17/100, batch 262/625 completed (loss: 0.0459:  42%|████▏     | 131/312 [13:15<18:24,  6.10s/it]
Training Epoch: 17/100, batch 262/625 completed (loss: 0.1065:  42%|████▏     | 132/312 [13:18<18:11,  6.06s/it]
Training Epoch: 17/100, batch 262/625 completed (loss: 0.0459:  42%|████▏     | 132/312 [13:18<18:11,  6.06s/it]
Training Epoch: 17/100, batch 262/625 completed (loss: 0.0529:  42%|████▏     | 132/312 [13:18<18:11,  6.06s/it]
Training Epoch: 17/100, batch 262/625 completed (loss: 0.0268:  42%|████▏     | 132/312 [13:18<18:12,  6.07s/it]
Training Epoch: 17/100, batch 263/625 completed (loss: 0.027:  42%|████▏     | 132/312 [13:18<18:11,  6.06s/it] 
Training Epoch: 17/100, batch 263/625 completed (loss: 0.0604:  42%|████▏     | 132/312 [13:18<18:11,  6.06s/it]
Training Epoch: 17/100, batch 263/625 completed (loss: 0.0078:  42%|████▏     | 132/312 [13:18<18:12,  6.07s/it]
Training Epoch: 17/100, batch 263/625 completed (loss: 0.2715:  42%|████▏     | 132/312 [13:18<18:11,  6.06s/it]
Training Epoch: 17/100, batch 264/625 completed (loss: 0.0329:  42%|████▏     | 132/312 [13:21<18:12,  6.07s/it]
Training Epoch: 17/100, batch 264/625 completed (loss: 0.0432:  42%|████▏     | 132/312 [13:21<18:11,  6.06s/it]
Training Epoch: 17/100, batch 264/625 completed (loss: 0.0849:  42%|████▏     | 132/312 [13:21<18:11,  6.06s/it]
Training Epoch: 17/100, batch 264/625 completed (loss: 0.0017:  42%|████▏     | 132/312 [13:21<18:11,  6.06s/it]
Training Epoch: 17/100, batch 264/625 completed (loss: 0.0329:  43%|████▎     | 133/312 [13:24<18:09,  6.09s/it]
Training Epoch: 17/100, batch 264/625 completed (loss: 0.0017:  43%|████▎     | 133/312 [13:24<18:10,  6.09s/it]
Training Epoch: 17/100, batch 264/625 completed (loss: 0.0849:  43%|████▎     | 133/312 [13:24<18:10,  6.09s/it]
Training Epoch: 17/100, batch 264/625 completed (loss: 0.0432:  43%|████▎     | 133/312 [13:24<18:10,  6.09s/it]
Training Epoch: 17/100, batch 265/625 completed (loss: 0.0369:  43%|████▎     | 133/312 [13:24<18:10,  6.09s/it]
Training Epoch: 17/100, batch 265/625 completed (loss: 0.0254:  43%|████▎     | 133/312 [13:24<18:10,  6.09s/it]
Training Epoch: 17/100, batch 265/625 completed (loss: 0.003:  43%|████▎     | 133/312 [13:24<18:09,  6.09s/it] 
Training Epoch: 17/100, batch 265/625 completed (loss: 0.0079:  43%|████▎     | 133/312 [13:24<18:10,  6.09s/it]
Training Epoch: 17/100, batch 266/625 completed (loss: 0.1915:  43%|████▎     | 133/312 [13:27<18:10,  6.09s/it]
Training Epoch: 17/100, batch 266/625 completed (loss: 0.0288:  43%|████▎     | 133/312 [13:27<18:10,  6.09s/it]
Training Epoch: 17/100, batch 266/625 completed (loss: 0.0057:  43%|████▎     | 133/312 [13:27<18:09,  6.09s/it]
Training Epoch: 17/100, batch 266/625 completed (loss: 0.0025:  43%|████▎     | 133/312 [13:27<18:10,  6.09s/it]
Training Epoch: 17/100, batch 266/625 completed (loss: 0.0288:  43%|████▎     | 134/312 [13:30<18:00,  6.07s/it]
Training Epoch: 17/100, batch 266/625 completed (loss: 0.0057:  43%|████▎     | 134/312 [13:30<18:00,  6.07s/it]
Training Epoch: 17/100, batch 266/625 completed (loss: 0.0025:  43%|████▎     | 134/312 [13:30<18:00,  6.07s/it]
Training Epoch: 17/100, batch 266/625 completed (loss: 0.1915:  43%|████▎     | 134/312 [13:30<18:00,  6.07s/it]
Training Epoch: 17/100, batch 267/625 completed (loss: 0.0086:  43%|████▎     | 134/312 [13:30<18:00,  6.07s/it]
Training Epoch: 17/100, batch 267/625 completed (loss: 0.0043:  43%|████▎     | 134/312 [13:30<18:00,  6.07s/it]
Training Epoch: 17/100, batch 267/625 completed (loss: 0.0271:  43%|████▎     | 134/312 [13:30<18:00,  6.07s/it]
Training Epoch: 17/100, batch 267/625 completed (loss: 0.1424:  43%|████▎     | 134/312 [13:30<18:00,  6.07s/it]
Training Epoch: 17/100, batch 268/625 completed (loss: 0.0211:  43%|████▎     | 134/312 [13:33<18:00,  6.07s/it]
Training Epoch: 17/100, batch 268/625 completed (loss: 0.1472:  43%|████▎     | 134/312 [13:33<18:00,  6.07s/it]
Training Epoch: 17/100, batch 268/625 completed (loss: 0.0724:  43%|████▎     | 134/312 [13:33<18:00,  6.07s/it]
Training Epoch: 17/100, batch 268/625 completed (loss: 0.0075:  43%|████▎     | 134/312 [13:33<18:00,  6.07s/it]
Training Epoch: 17/100, batch 268/625 completed (loss: 0.1472:  43%|████▎     | 135/312 [13:36<17:58,  6.09s/it]
Training Epoch: 17/100, batch 268/625 completed (loss: 0.0211:  43%|████▎     | 135/312 [13:36<17:58,  6.09s/it]
Training Epoch: 17/100, batch 268/625 completed (loss: 0.0075:  43%|████▎     | 135/312 [13:36<17:58,  6.09s/it]
Training Epoch: 17/100, batch 268/625 completed (loss: 0.0724:  43%|████▎     | 135/312 [13:36<17:58,  6.10s/it]
Training Epoch: 17/100, batch 269/625 completed (loss: 0.0037:  43%|████▎     | 135/312 [13:36<17:58,  6.10s/it]
Training Epoch: 17/100, batch 269/625 completed (loss: 0.0223:  43%|████▎     | 135/312 [13:36<17:58,  6.09s/it]
Training Epoch: 17/100, batch 269/625 completed (loss: 0.0057:  43%|████▎     | 135/312 [13:36<17:58,  6.09s/it]
Training Epoch: 17/100, batch 269/625 completed (loss: 0.0592:  43%|████▎     | 135/312 [13:36<17:58,  6.09s/it]
Training Epoch: 17/100, batch 270/625 completed (loss: 0.0792:  43%|████▎     | 135/312 [13:40<17:58,  6.10s/it]
Training Epoch: 17/100, batch 270/625 completed (loss: 0.0054:  43%|████▎     | 135/312 [13:40<17:58,  6.09s/it]
Training Epoch: 17/100, batch 270/625 completed (loss: 0.0947:  43%|████▎     | 135/312 [13:39<17:58,  6.09s/it]
Training Epoch: 17/100, batch 270/625 completed (loss: 0.0107:  43%|████▎     | 135/312 [13:40<17:58,  6.09s/it]
Training Epoch: 17/100, batch 270/625 completed (loss: 0.0947:  44%|████▎     | 136/312 [13:42<17:46,  6.06s/it]
Training Epoch: 17/100, batch 270/625 completed (loss: 0.0054:  44%|████▎     | 136/312 [13:42<17:46,  6.06s/it]
Training Epoch: 17/100, batch 270/625 completed (loss: 0.0792:  44%|████▎     | 136/312 [13:42<17:46,  6.06s/it]
Training Epoch: 17/100, batch 270/625 completed (loss: 0.0107:  44%|████▎     | 136/312 [13:42<17:47,  6.06s/it]
Training Epoch: 17/100, batch 271/625 completed (loss: 0.0111:  44%|████▎     | 136/312 [13:42<17:46,  6.06s/it]
Training Epoch: 17/100, batch 271/625 completed (loss: 0.0342:  44%|████▎     | 136/312 [13:42<17:46,  6.06s/it]
Training Epoch: 17/100, batch 271/625 completed (loss: 0.0174:  44%|████▎     | 136/312 [13:42<17:46,  6.06s/it]
Training Epoch: 17/100, batch 271/625 completed (loss: 0.0245:  44%|████▎     | 136/312 [13:42<17:47,  6.06s/it]
Training Epoch: 17/100, batch 272/625 completed (loss: 0.0059:  44%|████▎     | 136/312 [13:46<17:46,  6.06s/it]
Training Epoch: 17/100, batch 272/625 completed (loss: 0.0228:  44%|████▎     | 136/312 [13:46<17:46,  6.06s/it]
Training Epoch: 17/100, batch 272/625 completed (loss: 0.0253:  44%|████▎     | 136/312 [13:46<17:46,  6.06s/it]
Training Epoch: 17/100, batch 272/625 completed (loss: 0.0594:  44%|████▎     | 136/312 [13:46<17:47,  6.06s/it]
Training Epoch: 17/100, batch 272/625 completed (loss: 0.0253:  44%|████▍     | 137/312 [13:49<17:46,  6.09s/it]
Training Epoch: 17/100, batch 272/625 completed (loss: 0.0059:  44%|████▍     | 137/312 [13:49<17:46,  6.09s/it]
Training Epoch: 17/100, batch 272/625 completed (loss: 0.0228:  44%|████▍     | 137/312 [13:49<17:46,  6.10s/it]
Training Epoch: 17/100, batch 272/625 completed (loss: 0.0594:  44%|████▍     | 137/312 [13:49<17:47,  6.10s/it]
Training Epoch: 17/100, batch 273/625 completed (loss: 0.0015:  44%|████▍     | 137/312 [13:49<17:46,  6.09s/it]
Training Epoch: 17/100, batch 273/625 completed (loss: 0.0151:  44%|████▍     | 137/312 [13:49<17:46,  6.10s/it]
Training Epoch: 17/100, batch 273/625 completed (loss: 0.2194:  44%|████▍     | 137/312 [13:49<17:46,  6.09s/it]
Training Epoch: 17/100, batch 273/625 completed (loss: 0.0268:  44%|████▍     | 137/312 [13:49<17:47,  6.10s/it]
Training Epoch: 17/100, batch 274/625 completed (loss: 0.2343:  44%|████▍     | 137/312 [13:52<17:46,  6.09s/it]
Training Epoch: 17/100, batch 274/625 completed (loss: 0.2344:  44%|████▍     | 137/312 [13:51<17:46,  6.09s/it]
Training Epoch: 17/100, batch 274/625 completed (loss: 0.1518:  44%|████▍     | 137/312 [13:52<17:46,  6.10s/it]
Training Epoch: 17/100, batch 274/625 completed (loss: 0.1843:  44%|████▍     | 137/312 [13:52<17:47,  6.10s/it]
Training Epoch: 17/100, batch 274/625 completed (loss: 0.1518:  44%|████▍     | 138/312 [13:55<17:42,  6.11s/it]
Training Epoch: 17/100, batch 274/625 completed (loss: 0.2344:  44%|████▍     | 138/312 [13:55<17:42,  6.11s/it]
Training Epoch: 17/100, batch 274/625 completed (loss: 0.2343:  44%|████▍     | 138/312 [13:55<17:42,  6.11s/it]
Training Epoch: 17/100, batch 274/625 completed (loss: 0.1843:  44%|████▍     | 138/312 [13:55<17:42,  6.11s/it]
Training Epoch: 17/100, batch 275/625 completed (loss: 0.1852:  44%|████▍     | 138/312 [13:55<17:42,  6.11s/it]
Training Epoch: 17/100, batch 275/625 completed (loss: 0.0392:  44%|████▍     | 138/312 [13:55<17:42,  6.11s/it]
Training Epoch: 17/100, batch 275/625 completed (loss: 0.145:  44%|████▍     | 138/312 [13:55<17:42,  6.11s/it] 
Training Epoch: 17/100, batch 275/625 completed (loss: 0.0797:  44%|████▍     | 138/312 [13:55<17:42,  6.11s/it]
Training Epoch: 17/100, batch 276/625 completed (loss: 0.0611:  44%|████▍     | 138/312 [13:58<17:42,  6.11s/it]
Training Epoch: 17/100, batch 276/625 completed (loss: 0.3691:  44%|████▍     | 138/312 [13:58<17:42,  6.11s/it]
Training Epoch: 17/100, batch 276/625 completed (loss: 0.0472:  44%|████▍     | 138/312 [13:58<17:42,  6.11s/it]
Training Epoch: 17/100, batch 276/625 completed (loss: 0.0067:  44%|████▍     | 138/312 [13:58<17:42,  6.11s/it]
Training Epoch: 17/100, batch 276/625 completed (loss: 0.0472:  45%|████▍     | 139/312 [14:01<17:33,  6.09s/it]
Training Epoch: 17/100, batch 276/625 completed (loss: 0.3691:  45%|████▍     | 139/312 [14:01<17:33,  6.09s/it]
Training Epoch: 17/100, batch 276/625 completed (loss: 0.0611:  45%|████▍     | 139/312 [14:01<17:33,  6.09s/it]
Training Epoch: 17/100, batch 276/625 completed (loss: 0.0067:  45%|████▍     | 139/312 [14:01<17:33,  6.09s/it]
Training Epoch: 17/100, batch 277/625 completed (loss: 0.029:  45%|████▍     | 139/312 [14:01<17:33,  6.09s/it] 
Training Epoch: 17/100, batch 277/625 completed (loss: 0.0015:  45%|████▍     | 139/312 [14:01<17:33,  6.09s/it]
Training Epoch: 17/100, batch 277/625 completed (loss: 0.0389:  45%|████▍     | 139/312 [14:01<17:33,  6.09s/it]
Training Epoch: 17/100, batch 277/625 completed (loss: 0.1802:  45%|████▍     | 139/312 [14:01<17:33,  6.09s/it]
Training Epoch: 17/100, batch 278/625 completed (loss: 0.0554:  45%|████▍     | 139/312 [14:04<17:33,  6.09s/it]
Training Epoch: 17/100, batch 278/625 completed (loss: 0.0216:  45%|████▍     | 139/312 [14:04<17:33,  6.09s/it]
Training Epoch: 17/100, batch 278/625 completed (loss: 0.0014:  45%|████▍     | 139/312 [14:04<17:33,  6.09s/it]
Training Epoch: 17/100, batch 278/625 completed (loss: 0.0154:  45%|████▍     | 139/312 [14:04<17:33,  6.09s/it]
Training Epoch: 17/100, batch 278/625 completed (loss: 0.0554:  45%|████▍     | 140/312 [14:07<17:24,  6.07s/it]
Training Epoch: 17/100, batch 278/625 completed (loss: 0.0014:  45%|████▍     | 140/312 [14:07<17:25,  6.08s/it]
Training Epoch: 17/100, batch 278/625 completed (loss: 0.0216:  45%|████▍     | 140/312 [14:07<17:26,  6.08s/it]
Training Epoch: 17/100, batch 278/625 completed (loss: 0.0154:  45%|████▍     | 140/312 [14:07<17:26,  6.08s/it]
Training Epoch: 17/100, batch 279/625 completed (loss: 0.0723:  45%|████▍     | 140/312 [14:07<17:25,  6.08s/it]
Training Epoch: 17/100, batch 279/625 completed (loss: 0.0105:  45%|████▍     | 140/312 [14:07<17:26,  6.08s/it]
Training Epoch: 17/100, batch 279/625 completed (loss: 0.1733:  45%|████▍     | 140/312 [14:07<17:24,  6.07s/it]
Training Epoch: 17/100, batch 279/625 completed (loss: 0.0019:  45%|████▍     | 140/312 [14:07<17:26,  6.08s/it]
Training Epoch: 17/100, batch 280/625 completed (loss: 0.0675:  45%|████▍     | 140/312 [14:10<17:26,  6.08s/it]
Training Epoch: 17/100, batch 280/625 completed (loss: 0.0188:  45%|████▍     | 140/312 [14:10<17:25,  6.08s/it]
Training Epoch: 17/100, batch 280/625 completed (loss: 0.047:  45%|████▍     | 140/312 [14:10<17:24,  6.07s/it] 
Training Epoch: 17/100, batch 280/625 completed (loss: 0.0062:  45%|████▍     | 140/312 [14:10<17:26,  6.08s/it]
Training Epoch: 17/100, batch 280/625 completed (loss: 0.047:  45%|████▌     | 141/312 [14:13<17:23,  6.10s/it]
Training Epoch: 17/100, batch 280/625 completed (loss: 0.0188:  45%|████▌     | 141/312 [14:13<17:23,  6.10s/it]
Training Epoch: 17/100, batch 280/625 completed (loss: 0.0062:  45%|████▌     | 141/312 [14:13<17:23,  6.10s/it]
Training Epoch: 17/100, batch 280/625 completed (loss: 0.0675:  45%|████▌     | 141/312 [14:13<17:23,  6.10s/it]
Training Epoch: 17/100, batch 281/625 completed (loss: 0.004:  45%|████▌     | 141/312 [14:13<17:23,  6.10s/it] 
Training Epoch: 17/100, batch 281/625 completed (loss: 0.0339:  45%|████▌     | 141/312 [14:13<17:23,  6.10s/it]
Training Epoch: 17/100, batch 281/625 completed (loss: 0.0831:  45%|████▌     | 141/312 [14:13<17:23,  6.10s/it]
Training Epoch: 17/100, batch 281/625 completed (loss: 0.0018:  45%|████▌     | 141/312 [14:13<17:23,  6.10s/it]
Training Epoch: 17/100, batch 282/625 completed (loss: 0.071:  45%|████▌     | 141/312 [14:16<17:23,  6.10s/it] 
Training Epoch: 17/100, batch 282/625 completed (loss: 0.0373:  45%|████▌     | 141/312 [14:16<17:23,  6.10s/it]
Training Epoch: 17/100, batch 282/625 completed (loss: 0.0175:  45%|████▌     | 141/312 [14:16<17:23,  6.10s/it]
Training Epoch: 17/100, batch 282/625 completed (loss: 0.0118:  45%|████▌     | 141/312 [14:16<17:23,  6.10s/it]
Training Epoch: 17/100, batch 282/625 completed (loss: 0.0118:  46%|████▌     | 142/312 [14:19<17:11,  6.07s/it]
Training Epoch: 17/100, batch 282/625 completed (loss: 0.0373:  46%|████▌     | 142/312 [14:19<17:11,  6.07s/it]
Training Epoch: 17/100, batch 282/625 completed (loss: 0.0175:  46%|████▌     | 142/312 [14:19<17:11,  6.07s/it]
Training Epoch: 17/100, batch 282/625 completed (loss: 0.071:  46%|████▌     | 142/312 [14:19<17:11,  6.07s/it]
Training Epoch: 17/100, batch 283/625 completed (loss: 0.4825:  46%|████▌     | 142/312 [14:19<17:11,  6.07s/it]
Training Epoch: 17/100, batch 283/625 completed (loss: 0.0053:  46%|████▌     | 142/312 [14:19<17:11,  6.07s/it]
Training Epoch: 17/100, batch 283/625 completed (loss: 0.0246:  46%|████▌     | 142/312 [14:19<17:11,  6.07s/it]
Training Epoch: 17/100, batch 283/625 completed (loss: 0.0095:  46%|████▌     | 142/312 [14:19<17:11,  6.07s/it]
Training Epoch: 17/100, batch 284/625 completed (loss: 0.0468:  46%|████▌     | 142/312 [14:22<17:11,  6.07s/it]
Training Epoch: 17/100, batch 284/625 completed (loss: 0.0054:  46%|████▌     | 142/312 [14:22<17:11,  6.07s/it]
Training Epoch: 17/100, batch 284/625 completed (loss: 0.1166:  46%|████▌     | 142/312 [14:22<17:11,  6.07s/it]
Training Epoch: 17/100, batch 284/625 completed (loss: 0.267:  46%|████▌     | 142/312 [14:22<17:11,  6.07s/it] 
Training Epoch: 17/100, batch 284/625 completed (loss: 0.1166:  46%|████▌     | 143/312 [14:25<17:17,  6.14s/it]
Training Epoch: 17/100, batch 284/625 completed (loss: 0.0054:  46%|████▌     | 143/312 [14:25<17:17,  6.14s/it]
Training Epoch: 17/100, batch 284/625 completed (loss: 0.0468:  46%|████▌     | 143/312 [14:25<17:17,  6.14s/it]
Training Epoch: 17/100, batch 284/625 completed (loss: 0.267:  46%|████▌     | 143/312 [14:25<17:17,  6.14s/it]
Training Epoch: 17/100, batch 285/625 completed (loss: 0.0094:  46%|████▌     | 143/312 [14:25<17:17,  6.14s/it]
Training Epoch: 17/100, batch 285/625 completed (loss: 0.0049:  46%|████▌     | 143/312 [14:25<17:17,  6.14s/it]
Training Epoch: 17/100, batch 285/625 completed (loss: 0.1256:  46%|████▌     | 143/312 [14:25<17:17,  6.14s/it]
Training Epoch: 17/100, batch 285/625 completed (loss: 0.4687:  46%|████▌     | 143/312 [14:25<17:17,  6.14s/it]
Training Epoch: 17/100, batch 286/625 completed (loss: 0.1896:  46%|████▌     | 143/312 [14:28<17:17,  6.14s/it]
Training Epoch: 17/100, batch 286/625 completed (loss: 0.4597:  46%|████▌     | 143/312 [14:28<17:17,  6.14s/it]
Training Epoch: 17/100, batch 286/625 completed (loss: 0.2461:  46%|████▌     | 143/312 [14:28<17:17,  6.14s/it]
Training Epoch: 17/100, batch 286/625 completed (loss: 0.0159:  46%|████▌     | 143/312 [14:28<17:17,  6.14s/it]
Training Epoch: 17/100, batch 286/625 completed (loss: 0.4597:  46%|████▌     | 144/312 [14:31<16:56,  6.05s/it]
Training Epoch: 17/100, batch 286/625 completed (loss: 0.1896:  46%|████▌     | 144/312 [14:31<16:57,  6.06s/it]
Training Epoch: 17/100, batch 286/625 completed (loss: 0.0159:  46%|████▌     | 144/312 [14:31<16:57,  6.06s/it]
Training Epoch: 17/100, batch 286/625 completed (loss: 0.2461:  46%|████▌     | 144/312 [14:31<16:58,  6.06s/it]
Training Epoch: 17/100, batch 287/625 completed (loss: 0.0332:  46%|████▌     | 144/312 [14:31<16:58,  6.06s/it]
Training Epoch: 17/100, batch 287/625 completed (loss: 0.0635:  46%|████▌     | 144/312 [14:31<16:56,  6.05s/it]
Training Epoch: 17/100, batch 287/625 completed (loss: 0.0185:  46%|████▌     | 144/312 [14:31<16:57,  6.06s/it]
Training Epoch: 17/100, batch 287/625 completed (loss: 0.0141:  46%|████▌     | 144/312 [14:31<16:57,  6.06s/it]
Training Epoch: 17/100, batch 288/625 completed (loss: 0.0191:  46%|████▌     | 144/312 [14:34<16:58,  6.06s/it]
Training Epoch: 17/100, batch 288/625 completed (loss: 0.0286:  46%|████▌     | 144/312 [14:34<16:56,  6.05s/it]
Training Epoch: 17/100, batch 288/625 completed (loss: 0.1969:  46%|████▌     | 144/312 [14:34<16:57,  6.06s/it]
Training Epoch: 17/100, batch 288/625 completed (loss: 0.0067:  46%|████▌     | 144/312 [14:34<16:57,  6.06s/it]
Training Epoch: 17/100, batch 288/625 completed (loss: 0.0067:  46%|████▋     | 145/312 [14:37<16:55,  6.08s/it]
Training Epoch: 17/100, batch 288/625 completed (loss: 0.0286:  46%|████▋     | 145/312 [14:37<16:56,  6.09s/it]
Training Epoch: 17/100, batch 288/625 completed (loss: 0.1969:  46%|████▋     | 145/312 [14:37<16:57,  6.09s/it]
Training Epoch: 17/100, batch 288/625 completed (loss: 0.0191:  46%|████▋     | 145/312 [14:37<16:56,  6.09s/it]
Training Epoch: 17/100, batch 289/625 completed (loss: 0.4031:  46%|████▋     | 145/312 [14:37<16:56,  6.09s/it]
Training Epoch: 17/100, batch 289/625 completed (loss: 0.011:  46%|████▋     | 145/312 [14:37<16:57,  6.09s/it] 
Training Epoch: 17/100, batch 289/625 completed (loss: 0.0515:  46%|████▋     | 145/312 [14:37<16:56,  6.09s/it]
Training Epoch: 17/100, batch 289/625 completed (loss: 0.2464:  46%|████▋     | 145/312 [14:37<16:55,  6.08s/it]
Training Epoch: 17/100, batch 290/625 completed (loss: 0.1181:  46%|████▋     | 145/312 [14:40<16:57,  6.09s/it]
Training Epoch: 17/100, batch 290/625 completed (loss: 0.0231:  46%|████▋     | 145/312 [14:40<16:56,  6.09s/it]
Training Epoch: 17/100, batch 290/625 completed (loss: 0.0058:  46%|████▋     | 145/312 [14:40<16:56,  6.09s/it]
Training Epoch: 17/100, batch 290/625 completed (loss: 0.0146:  46%|████▋     | 145/312 [14:40<16:55,  6.08s/it]
Training Epoch: 17/100, batch 290/625 completed (loss: 0.0058:  47%|████▋     | 146/312 [14:43<16:47,  6.07s/it]
Training Epoch: 17/100, batch 290/625 completed (loss: 0.0146:  47%|████▋     | 146/312 [14:43<16:47,  6.07s/it]
Training Epoch: 17/100, batch 290/625 completed (loss: 0.1181:  47%|████▋     | 146/312 [14:43<16:47,  6.07s/it]
Training Epoch: 17/100, batch 290/625 completed (loss: 0.0231:  47%|████▋     | 146/312 [14:43<16:47,  6.07s/it]
Training Epoch: 17/100, batch 291/625 completed (loss: 0.0035:  47%|████▋     | 146/312 [14:43<16:47,  6.07s/it]
Training Epoch: 17/100, batch 291/625 completed (loss: 0.0287:  47%|████▋     | 146/312 [14:43<16:47,  6.07s/it]
Training Epoch: 17/100, batch 291/625 completed (loss: 0.0192:  47%|████▋     | 146/312 [14:43<16:47,  6.07s/it]
Training Epoch: 17/100, batch 291/625 completed (loss: 0.0313:  47%|████▋     | 146/312 [14:43<16:47,  6.07s/it]
Training Epoch: 17/100, batch 292/625 completed (loss: 0.0132:  47%|████▋     | 146/312 [14:46<16:47,  6.07s/it]
Training Epoch: 17/100, batch 292/625 completed (loss: 0.1036:  47%|████▋     | 146/312 [14:46<16:47,  6.07s/it]
Training Epoch: 17/100, batch 292/625 completed (loss: 0.0226:  47%|████▋     | 146/312 [14:46<16:47,  6.07s/it]
Training Epoch: 17/100, batch 292/625 completed (loss: 0.0787:  47%|████▋     | 146/312 [14:46<16:47,  6.07s/it]
Training Epoch: 17/100, batch 292/625 completed (loss: 0.0132:  47%|████▋     | 147/312 [14:49<16:40,  6.06s/it]
Training Epoch: 17/100, batch 292/625 completed (loss: 0.1036:  47%|████▋     | 147/312 [14:49<16:40,  6.06s/it]
Training Epoch: 17/100, batch 292/625 completed (loss: 0.0226:  47%|████▋     | 147/312 [14:49<16:40,  6.06s/it]
Training Epoch: 17/100, batch 292/625 completed (loss: 0.0787:  47%|████▋     | 147/312 [14:49<16:41,  6.07s/it]
Training Epoch: 17/100, batch 293/625 completed (loss: 0.0723:  47%|████▋     | 147/312 [14:49<16:40,  6.06s/it]
Training Epoch: 17/100, batch 293/625 completed (loss: 0.0069:  47%|████▋     | 147/312 [14:49<16:40,  6.06s/it]
Training Epoch: 17/100, batch 293/625 completed (loss: 0.0238:  47%|████▋     | 147/312 [14:49<16:40,  6.06s/it]
Training Epoch: 17/100, batch 293/625 completed (loss: 0.0116:  47%|████▋     | 147/312 [14:49<16:41,  6.07s/it]
Training Epoch: 17/100, batch 294/625 completed (loss: 0.0059:  47%|████▋     | 147/312 [14:52<16:40,  6.06s/it]
Training Epoch: 17/100, batch 294/625 completed (loss: 0.4349:  47%|████▋     | 147/312 [14:52<16:40,  6.06s/it]
Training Epoch: 17/100, batch 294/625 completed (loss: 0.0534:  47%|████▋     | 147/312 [14:52<16:40,  6.06s/it]
Training Epoch: 17/100, batch 294/625 completed (loss: 0.054:  47%|████▋     | 147/312 [14:52<16:41,  6.07s/it] 
Training Epoch: 17/100, batch 294/625 completed (loss: 0.054:  47%|████▋     | 148/312 [14:56<16:37,  6.08s/it]
Training Epoch: 17/100, batch 294/625 completed (loss: 0.4349:  47%|████▋     | 148/312 [14:55<16:38,  6.09s/it]
Training Epoch: 17/100, batch 294/625 completed (loss: 0.0059:  47%|████▋     | 148/312 [14:55<16:38,  6.09s/it]
Training Epoch: 17/100, batch 294/625 completed (loss: 0.0534:  47%|████▋     | 148/312 [14:55<16:38,  6.09s/it]
Training Epoch: 17/100, batch 295/625 completed (loss: 0.0156:  47%|████▋     | 148/312 [14:56<16:38,  6.09s/it]
Training Epoch: 17/100, batch 295/625 completed (loss: 0.0365:  47%|████▋     | 148/312 [14:56<16:38,  6.09s/it]
Training Epoch: 17/100, batch 295/625 completed (loss: 0.0322:  47%|████▋     | 148/312 [14:56<16:38,  6.09s/it]
Training Epoch: 17/100, batch 295/625 completed (loss: 0.0083:  47%|████▋     | 148/312 [14:56<16:37,  6.08s/it]
Training Epoch: 17/100, batch 296/625 completed (loss: 0.0069:  47%|████▋     | 148/312 [14:59<16:38,  6.09s/it]
Training Epoch: 17/100, batch 296/625 completed (loss: 0.4425:  47%|████▋     | 148/312 [14:59<16:38,  6.09s/it]
Training Epoch: 17/100, batch 296/625 completed (loss: 0.0024:  47%|████▋     | 148/312 [14:59<16:38,  6.09s/it]
Training Epoch: 17/100, batch 296/625 completed (loss: 0.0592:  47%|████▋     | 148/312 [14:59<16:37,  6.08s/it]
Training Epoch: 17/100, batch 296/625 completed (loss: 0.0592:  48%|████▊     | 149/312 [15:02<16:27,  6.06s/it]
Training Epoch: 17/100, batch 296/625 completed (loss: 0.0069:  48%|████▊     | 149/312 [15:01<16:27,  6.06s/it]
Training Epoch: 17/100, batch 296/625 completed (loss: 0.0024:  48%|████▊     | 149/312 [15:01<16:27,  6.06s/it]
Training Epoch: 17/100, batch 296/625 completed (loss: 0.4425:  48%|████▊     | 149/312 [15:01<16:28,  6.06s/it]
Training Epoch: 17/100, batch 297/625 completed (loss: 0.0154:  48%|████▊     | 149/312 [15:02<16:27,  6.06s/it]
Training Epoch: 17/100, batch 297/625 completed (loss: 0.425:  48%|████▊     | 149/312 [15:02<16:27,  6.06s/it] 
Training Epoch: 17/100, batch 297/625 completed (loss: 0.2201:  48%|████▊     | 149/312 [15:02<16:28,  6.06s/it]
Training Epoch: 17/100, batch 297/625 completed (loss: 0.0046:  48%|████▊     | 149/312 [15:02<16:27,  6.06s/it]
Training Epoch: 17/100, batch 298/625 completed (loss: 0.0169:  48%|████▊     | 149/312 [15:04<16:27,  6.06s/it]
Training Epoch: 17/100, batch 298/625 completed (loss: 0.0586:  48%|████▊     | 149/312 [15:04<16:28,  6.06s/it]
Training Epoch: 17/100, batch 298/625 completed (loss: 0.1102:  48%|████▊     | 149/312 [15:04<16:27,  6.06s/it]
Training Epoch: 17/100, batch 298/625 completed (loss: 0.0158:  48%|████▊     | 149/312 [15:04<16:27,  6.06s/it]
Training Epoch: 17/100, batch 298/625 completed (loss: 0.0158:  48%|████▊     | 150/312 [15:08<16:27,  6.09s/it]
Training Epoch: 17/100, batch 298/625 completed (loss: 0.0169:  48%|████▊     | 150/312 [15:08<16:26,  6.09s/it]
Training Epoch: 17/100, batch 298/625 completed (loss: 0.0586:  48%|████▊     | 150/312 [15:08<16:27,  6.09s/it]
Training Epoch: 17/100, batch 298/625 completed (loss: 0.1102:  48%|████▊     | 150/312 [15:08<16:27,  6.10s/it]
Training Epoch: 17/100, batch 299/625 completed (loss: 0.0064:  48%|████▊     | 150/312 [15:08<16:27,  6.09s/it]
Training Epoch: 17/100, batch 299/625 completed (loss: 0.0213:  48%|████▊     | 150/312 [15:08<16:27,  6.10s/it]
Training Epoch: 17/100, batch 299/625 completed (loss: 0.0109:  48%|████▊     | 150/312 [15:08<16:26,  6.09s/it]
Training Epoch: 17/100, batch 299/625 completed (loss: 0.1002:  48%|████▊     | 150/312 [15:08<16:27,  6.09s/it]
Training Epoch: 17/100, batch 300/625 completed (loss: 0.0093:  48%|████▊     | 150/312 [15:11<16:26,  6.09s/it]
Training Epoch: 17/100, batch 300/625 completed (loss: 0.0579:  48%|████▊     | 150/312 [15:11<16:27,  6.09s/it]
Training Epoch: 17/100, batch 300/625 completed (loss: 0.2221:  48%|████▊     | 150/312 [15:11<16:27,  6.10s/it]
Training Epoch: 17/100, batch 300/625 completed (loss: 0.0029:  48%|████▊     | 150/312 [15:11<16:27,  6.09s/it]
Training Epoch: 17/100, batch 300/625 completed (loss: 0.0093:  48%|████▊     | 151/312 [15:14<16:24,  6.12s/it]
Training Epoch: 17/100, batch 300/625 completed (loss: 0.2221:  48%|████▊     | 151/312 [15:14<16:24,  6.12s/it]
Training Epoch: 17/100, batch 300/625 completed (loss: 0.0579:  48%|████▊     | 151/312 [15:14<16:25,  6.12s/it]
Training Epoch: 17/100, batch 300/625 completed (loss: 0.0029:  48%|████▊     | 151/312 [15:14<16:25,  6.12s/it]
Training Epoch: 17/100, batch 301/625 completed (loss: 0.3768:  48%|████▊     | 151/312 [15:14<16:24,  6.12s/it]
Training Epoch: 17/100, batch 301/625 completed (loss: 0.0421:  48%|████▊     | 151/312 [15:14<16:24,  6.12s/it]
Training Epoch: 17/100, batch 301/625 completed (loss: 0.011:  48%|████▊     | 151/312 [15:14<16:25,  6.12s/it] 
Training Epoch: 17/100, batch 301/625 completed (loss: 0.0258:  48%|████▊     | 151/312 [15:14<16:25,  6.12s/it]
Training Epoch: 17/100, batch 302/625 completed (loss: 0.0088:  48%|████▊     | 151/312 [15:17<16:24,  6.12s/it]
Training Epoch: 17/100, batch 302/625 completed (loss: 0.0158:  48%|████▊     | 151/312 [15:17<16:25,  6.12s/it]
Training Epoch: 17/100, batch 302/625 completed (loss: 0.0123:  48%|████▊     | 151/312 [15:17<16:24,  6.12s/it]
Training Epoch: 17/100, batch 302/625 completed (loss: 0.0345:  48%|████▊     | 151/312 [15:17<16:25,  6.12s/it]
Training Epoch: 17/100, batch 302/625 completed (loss: 0.0345:  49%|████▊     | 152/312 [15:20<16:04,  6.03s/it]
Training Epoch: 17/100, batch 302/625 completed (loss: 0.0088:  49%|████▊     | 152/312 [15:20<16:04,  6.03s/it]
Training Epoch: 17/100, batch 302/625 completed (loss: 0.0123:  49%|████▊     | 152/312 [15:20<16:04,  6.03s/it]
Training Epoch: 17/100, batch 302/625 completed (loss: 0.0158:  49%|████▊     | 152/312 [15:20<16:05,  6.03s/it]
Training Epoch: 17/100, batch 303/625 completed (loss: 0.0064:  49%|████▊     | 152/312 [15:20<16:05,  6.03s/it]
Training Epoch: 17/100, batch 303/625 completed (loss: 0.1527:  49%|████▊     | 152/312 [15:20<16:04,  6.03s/it]
Training Epoch: 17/100, batch 303/625 completed (loss: 0.0233:  49%|████▊     | 152/312 [15:20<16:04,  6.03s/it]
Training Epoch: 17/100, batch 303/625 completed (loss: 0.3256:  49%|████▊     | 152/312 [15:20<16:04,  6.03s/it]
Training Epoch: 17/100, batch 304/625 completed (loss: 0.1096:  49%|████▊     | 152/312 [15:23<16:04,  6.03s/it]
Training Epoch: 17/100, batch 304/625 completed (loss: 0.0135:  49%|████▊     | 152/312 [15:23<16:05,  6.03s/it]
Training Epoch: 17/100, batch 304/625 completed (loss: 0.0085:  49%|████▊     | 152/312 [15:23<16:04,  6.03s/it]
Training Epoch: 17/100, batch 304/625 completed (loss: 0.5558:  49%|████▊     | 152/312 [15:23<16:04,  6.03s/it]
Training Epoch: 17/100, batch 304/625 completed (loss: 0.1096:  49%|████▉     | 153/312 [15:26<16:04,  6.06s/it]
Training Epoch: 17/100, batch 304/625 completed (loss: 0.0135:  49%|████▉     | 153/312 [15:26<16:03,  6.06s/it]
Training Epoch: 17/100, batch 304/625 completed (loss: 0.0085:  49%|████▉     | 153/312 [15:26<16:04,  6.06s/it]
Training Epoch: 17/100, batch 304/625 completed (loss: 0.5558:  49%|████▉     | 153/312 [15:26<16:04,  6.06s/it]
Training Epoch: 17/100, batch 305/625 completed (loss: 0.0918:  49%|████▉     | 153/312 [15:26<16:04,  6.06s/it]
Training Epoch: 17/100, batch 305/625 completed (loss: 0.0534:  49%|████▉     | 153/312 [15:26<16:03,  6.06s/it]
Training Epoch: 17/100, batch 305/625 completed (loss: 0.0194:  49%|████▉     | 153/312 [15:26<16:04,  6.06s/it]
Training Epoch: 17/100, batch 305/625 completed (loss: 0.0169:  49%|████▉     | 153/312 [15:26<16:04,  6.06s/it]
Training Epoch: 17/100, batch 306/625 completed (loss: 0.0598:  49%|████▉     | 153/312 [15:29<16:03,  6.06s/it]
Training Epoch: 17/100, batch 306/625 completed (loss: 0.0572:  49%|████▉     | 153/312 [15:29<16:04,  6.06s/it]
Training Epoch: 17/100, batch 306/625 completed (loss: 0.0229:  49%|████▉     | 153/312 [15:29<16:04,  6.06s/it]
Training Epoch: 17/100, batch 306/625 completed (loss: 0.0272:  49%|████▉     | 153/312 [15:29<16:04,  6.06s/it]
Training Epoch: 17/100, batch 306/625 completed (loss: 0.0229:  49%|████▉     | 154/312 [15:32<16:05,  6.11s/it]
Training Epoch: 17/100, batch 306/625 completed (loss: 0.0598:  49%|████▉     | 154/312 [15:32<16:05,  6.11s/it]
Training Epoch: 17/100, batch 306/625 completed (loss: 0.0572:  49%|████▉     | 154/312 [15:32<16:05,  6.11s/it]
Training Epoch: 17/100, batch 306/625 completed (loss: 0.0272:  49%|████▉     | 154/312 [15:32<16:05,  6.11s/it]
Training Epoch: 17/100, batch 307/625 completed (loss: 0.4311:  49%|████▉     | 154/312 [15:32<16:05,  6.11s/it]
Training Epoch: 17/100, batch 307/625 completed (loss: 0.0077:  49%|████▉     | 154/312 [15:32<16:05,  6.11s/it]
Training Epoch: 17/100, batch 307/625 completed (loss: 0.1272:  49%|████▉     | 154/312 [15:32<16:05,  6.11s/it]
Training Epoch: 17/100, batch 307/625 completed (loss: 0.0063:  49%|████▉     | 154/312 [15:32<16:05,  6.11s/it]
Training Epoch: 17/100, batch 308/625 completed (loss: 0.0846:  49%|████▉     | 154/312 [15:35<16:05,  6.11s/it]
Training Epoch: 17/100, batch 308/625 completed (loss: 0.0797:  49%|████▉     | 154/312 [15:35<16:05,  6.11s/it]
Training Epoch: 17/100, batch 308/625 completed (loss: 0.1455:  49%|████▉     | 154/312 [15:35<16:05,  6.11s/it]
Training Epoch: 17/100, batch 308/625 completed (loss: 0.0155:  49%|████▉     | 154/312 [15:35<16:05,  6.11s/it]
Training Epoch: 17/100, batch 308/625 completed (loss: 0.1455:  50%|████▉     | 155/312 [15:38<15:56,  6.09s/it]
Training Epoch: 17/100, batch 308/625 completed (loss: 0.0846:  50%|████▉     | 155/312 [15:38<15:56,  6.09s/it]
Training Epoch: 17/100, batch 308/625 completed (loss: 0.0797:  50%|████▉     | 155/312 [15:38<15:56,  6.09s/it]
Training Epoch: 17/100, batch 308/625 completed (loss: 0.0155:  50%|████▉     | 155/312 [15:38<15:56,  6.09s/it]
Training Epoch: 17/100, batch 309/625 completed (loss: 0.0072:  50%|████▉     | 155/312 [15:38<15:56,  6.09s/it]
Training Epoch: 17/100, batch 309/625 completed (loss: 0.0078:  50%|████▉     | 155/312 [15:38<15:56,  6.09s/it]
Training Epoch: 17/100, batch 309/625 completed (loss: 0.2411:  50%|████▉     | 155/312 [15:38<15:56,  6.09s/it]
Training Epoch: 17/100, batch 309/625 completed (loss: 0.0106:  50%|████▉     | 155/312 [15:38<15:56,  6.09s/it]
Training Epoch: 17/100, batch 310/625 completed (loss: 0.0172:  50%|████▉     | 155/312 [15:41<15:56,  6.09s/it]
Training Epoch: 17/100, batch 310/625 completed (loss: 0.0147:  50%|████▉     | 155/312 [15:41<15:56,  6.09s/it]
Training Epoch: 17/100, batch 310/625 completed (loss: 0.0634:  50%|████▉     | 155/312 [15:41<15:56,  6.09s/it]
Training Epoch: 17/100, batch 310/625 completed (loss: 0.0575:  50%|████▉     | 155/312 [15:41<15:56,  6.09s/it]
Training Epoch: 17/100, batch 310/625 completed (loss: 0.0147:  50%|█████     | 156/312 [15:44<15:53,  6.12s/it]
Training Epoch: 17/100, batch 310/625 completed (loss: 0.0172:  50%|█████     | 156/312 [15:44<15:54,  6.12s/it]
Training Epoch: 17/100, batch 310/625 completed (loss: 0.0634:  50%|█████     | 156/312 [15:44<15:54,  6.12s/it]
Training Epoch: 17/100, batch 310/625 completed (loss: 0.0575:  50%|█████     | 156/312 [15:44<15:54,  6.12s/it]
Training Epoch: 17/100, batch 311/625 completed (loss: 0.0412:  50%|█████     | 156/312 [15:44<15:54,  6.12s/it]
Training Epoch: 17/100, batch 311/625 completed (loss: 0.0432:  50%|█████     | 156/312 [15:44<15:53,  6.12s/it]
Training Epoch: 17/100, batch 311/625 completed (loss: 0.0166:  50%|█████     | 156/312 [15:44<15:54,  6.12s/it]
Training Epoch: 17/100, batch 311/625 completed (loss: 0.0014:  50%|█████     | 156/312 [15:44<15:54,  6.12s/it]
Training Epoch: 17/100, batch 312/625 completed (loss: 0.0158:  50%|█████     | 156/312 [15:47<15:54,  6.12s/it]
Training Epoch: 17/100, batch 312/625 completed (loss: 0.0177:  50%|█████     | 156/312 [15:47<15:53,  6.12s/it]
Training Epoch: 17/100, batch 312/625 completed (loss: 0.3191:  50%|█████     | 156/312 [15:47<15:54,  6.12s/it]
Training Epoch: 17/100, batch 312/625 completed (loss: 0.0263:  50%|█████     | 156/312 [15:47<15:54,  6.12s/it]
Training Epoch: 17/100, batch 312/625 completed (loss: 0.0158:  50%|█████     | 157/312 [15:50<15:36,  6.04s/it]
Training Epoch: 17/100, batch 312/625 completed (loss: 0.0177:  50%|█████     | 157/312 [15:50<15:36,  6.04s/it]
Training Epoch: 17/100, batch 312/625 completed (loss: 0.3191:  50%|█████     | 157/312 [15:50<15:36,  6.04s/it]
Training Epoch: 17/100, batch 312/625 completed (loss: 0.0263:  50%|█████     | 157/312 [15:50<15:36,  6.04s/it]
Training Epoch: 17/100, batch 313/625 completed (loss: 0.0618:  50%|█████     | 157/312 [15:50<15:36,  6.04s/it]
Training Epoch: 17/100, batch 313/625 completed (loss: 0.0205:  50%|█████     | 157/312 [15:50<15:36,  6.04s/it]
Training Epoch: 17/100, batch 313/625 completed (loss: 0.0432:  50%|█████     | 157/312 [15:50<15:36,  6.04s/it]
Training Epoch: 17/100, batch 313/625 completed (loss: 0.0186:  50%|█████     | 157/312 [15:50<15:36,  6.04s/it]
Training Epoch: 17/100, batch 314/625 completed (loss: 0.0187:  50%|█████     | 157/312 [15:53<15:36,  6.04s/it]
Training Epoch: 17/100, batch 314/625 completed (loss: 0.0132:  50%|█████     | 157/312 [15:53<15:36,  6.04s/it]
Training Epoch: 17/100, batch 314/625 completed (loss: 0.0048:  50%|█████     | 157/312 [15:53<15:36,  6.04s/it]
Training Epoch: 17/100, batch 314/625 completed (loss: 0.1028:  50%|█████     | 157/312 [15:53<15:36,  6.04s/it]
Training Epoch: 17/100, batch 314/625 completed (loss: 0.0187:  51%|█████     | 158/312 [15:56<15:28,  6.03s/it]
Training Epoch: 17/100, batch 314/625 completed (loss: 0.0048:  51%|█████     | 158/312 [15:56<15:28,  6.03s/it]
Training Epoch: 17/100, batch 314/625 completed (loss: 0.0132:  51%|█████     | 158/312 [15:56<15:28,  6.03s/it]
Training Epoch: 17/100, batch 314/625 completed (loss: 0.1028:  51%|█████     | 158/312 [15:56<15:28,  6.03s/it]
Training Epoch: 17/100, batch 315/625 completed (loss: 0.0103:  51%|█████     | 158/312 [15:56<15:28,  6.03s/it]
Training Epoch: 17/100, batch 315/625 completed (loss: 0.1571:  51%|█████     | 158/312 [15:56<15:28,  6.03s/it]
Training Epoch: 17/100, batch 315/625 completed (loss: 0.018:  51%|█████     | 158/312 [15:56<15:28,  6.03s/it] 
Training Epoch: 17/100, batch 315/625 completed (loss: 0.132:  51%|█████     | 158/312 [15:56<15:28,  6.03s/it] 
Training Epoch: 17/100, batch 316/625 completed (loss: 0.0047:  51%|█████     | 158/312 [15:59<15:28,  6.03s/it]
Training Epoch: 17/100, batch 316/625 completed (loss: 0.0054:  51%|█████     | 158/312 [15:59<15:28,  6.03s/it]
Training Epoch: 17/100, batch 316/625 completed (loss: 0.118:  51%|█████     | 158/312 [15:59<15:28,  6.03s/it] 
Training Epoch: 17/100, batch 316/625 completed (loss: 0.0314:  51%|█████     | 158/312 [16:00<15:28,  6.03s/it]
Training Epoch: 17/100, batch 316/625 completed (loss: 0.118:  51%|█████     | 159/312 [16:02<15:35,  6.11s/it]
Training Epoch: 17/100, batch 316/625 completed (loss: 0.0047:  51%|█████     | 159/312 [16:02<15:35,  6.11s/it]
Training Epoch: 17/100, batch 316/625 completed (loss: 0.0054:  51%|█████     | 159/312 [16:02<15:35,  6.11s/it]
Training Epoch: 17/100, batch 316/625 completed (loss: 0.0314:  51%|█████     | 159/312 [16:02<15:35,  6.11s/it]
Training Epoch: 17/100, batch 317/625 completed (loss: 0.1418:  51%|█████     | 159/312 [16:02<15:35,  6.11s/it]
Training Epoch: 17/100, batch 317/625 completed (loss: 0.0094:  51%|█████     | 159/312 [16:02<15:35,  6.11s/it]
Training Epoch: 17/100, batch 317/625 completed (loss: 0.0361:  51%|█████     | 159/312 [16:02<15:35,  6.11s/it]
Training Epoch: 17/100, batch 317/625 completed (loss: 0.0179:  51%|█████     | 159/312 [16:02<15:35,  6.11s/it]
Training Epoch: 17/100, batch 318/625 completed (loss: 0.1253:  51%|█████     | 159/312 [16:05<15:35,  6.11s/it]
Training Epoch: 17/100, batch 318/625 completed (loss: 0.3234:  51%|█████     | 159/312 [16:05<15:35,  6.11s/it]
Training Epoch: 17/100, batch 318/625 completed (loss: 0.1296:  51%|█████     | 159/312 [16:05<15:35,  6.11s/it]
Training Epoch: 17/100, batch 318/625 completed (loss: 0.3009:  51%|█████     | 159/312 [16:05<15:35,  6.11s/it]
Training Epoch: 17/100, batch 318/625 completed (loss: 0.1296:  51%|█████▏    | 160/312 [16:08<15:15,  6.02s/it]
Training Epoch: 17/100, batch 318/625 completed (loss: 0.3234:  51%|█████▏    | 160/312 [16:08<15:15,  6.03s/it]
Training Epoch: 17/100, batch 318/625 completed (loss: 0.1253:  51%|█████▏    | 160/312 [16:08<15:15,  6.03s/it]
Training Epoch: 17/100, batch 318/625 completed (loss: 0.3009:  51%|█████▏    | 160/312 [16:08<15:15,  6.03s/it]
Training Epoch: 17/100, batch 319/625 completed (loss: 0.0188:  51%|█████▏    | 160/312 [16:08<15:15,  6.02s/it]
Training Epoch: 17/100, batch 319/625 completed (loss: 0.0771:  51%|█████▏    | 160/312 [16:08<15:15,  6.03s/it]
Training Epoch: 17/100, batch 319/625 completed (loss: 0.065:  51%|█████▏    | 160/312 [16:08<15:15,  6.03s/it] 
Training Epoch: 17/100, batch 319/625 completed (loss: 0.0048:  51%|█████▏    | 160/312 [16:08<15:15,  6.03s/it]
Training Epoch: 17/100, batch 320/625 completed (loss: 0.0035:  51%|█████▏    | 160/312 [16:11<15:15,  6.03s/it]
Training Epoch: 17/100, batch 320/625 completed (loss: 0.0304:  51%|█████▏    | 160/312 [16:11<15:15,  6.02s/it]
Training Epoch: 17/100, batch 320/625 completed (loss: 0.0381:  51%|█████▏    | 160/312 [16:11<15:15,  6.03s/it]
Training Epoch: 17/100, batch 320/625 completed (loss: 0.0145:  51%|█████▏    | 160/312 [16:11<15:15,  6.03s/it]
Training Epoch: 17/100, batch 320/625 completed (loss: 0.0145:  52%|█████▏    | 161/312 [16:15<15:23,  6.12s/it]
Training Epoch: 17/100, batch 320/625 completed (loss: 0.0035:  52%|█████▏    | 161/312 [16:15<15:24,  6.12s/it]
Training Epoch: 17/100, batch 320/625 completed (loss: 0.0304:  52%|█████▏    | 161/312 [16:15<15:24,  6.12s/it]
Training Epoch: 17/100, batch 320/625 completed (loss: 0.0381:  52%|█████▏    | 161/312 [16:15<15:24,  6.12s/it]
Training Epoch: 17/100, batch 321/625 completed (loss: 0.0055:  52%|█████▏    | 161/312 [16:15<15:24,  6.12s/it]
Training Epoch: 17/100, batch 321/625 completed (loss: 0.1634:  52%|█████▏    | 161/312 [16:15<15:24,  6.12s/it]
Training Epoch: 17/100, batch 321/625 completed (loss: 0.5221:  52%|█████▏    | 161/312 [16:15<15:24,  6.12s/it]
Training Epoch: 17/100, batch 321/625 completed (loss: 0.0163:  52%|█████▏    | 161/312 [16:15<15:23,  6.12s/it]
Training Epoch: 17/100, batch 322/625 completed (loss: 0.0038:  52%|█████▏    | 161/312 [16:18<15:24,  6.12s/it]
Training Epoch: 17/100, batch 322/625 completed (loss: 0.0205:  52%|█████▏    | 161/312 [16:18<15:24,  6.12s/it]
Training Epoch: 17/100, batch 322/625 completed (loss: 0.0679:  52%|█████▏    | 161/312 [16:18<15:24,  6.12s/it]
Training Epoch: 17/100, batch 322/625 completed (loss: 0.0457:  52%|█████▏    | 161/312 [16:18<15:23,  6.12s/it]
Training Epoch: 17/100, batch 322/625 completed (loss: 0.0038:  52%|█████▏    | 162/312 [16:20<15:04,  6.03s/it]
Training Epoch: 17/100, batch 322/625 completed (loss: 0.0205:  52%|█████▏    | 162/312 [16:20<15:04,  6.03s/it]
Training Epoch: 17/100, batch 322/625 completed (loss: 0.0679:  52%|█████▏    | 162/312 [16:20<15:04,  6.03s/it]
Training Epoch: 17/100, batch 322/625 completed (loss: 0.0457:  52%|█████▏    | 162/312 [16:20<15:04,  6.03s/it]
Training Epoch: 17/100, batch 323/625 completed (loss: 0.0566:  52%|█████▏    | 162/312 [16:20<15:04,  6.03s/it]
Training Epoch: 17/100, batch 323/625 completed (loss: 0.3607:  52%|█████▏    | 162/312 [16:20<15:04,  6.03s/it]
Training Epoch: 17/100, batch 323/625 completed (loss: 0.0164:  52%|█████▏    | 162/312 [16:20<15:04,  6.03s/it]
Training Epoch: 17/100, batch 323/625 completed (loss: 0.0143:  52%|█████▏    | 162/312 [16:20<15:04,  6.03s/it]
Training Epoch: 17/100, batch 324/625 completed (loss: 0.0324:  52%|█████▏    | 162/312 [16:23<15:04,  6.03s/it]
Training Epoch: 17/100, batch 324/625 completed (loss: 0.0204:  52%|█████▏    | 162/312 [16:23<15:04,  6.03s/it]
Training Epoch: 17/100, batch 324/625 completed (loss: 0.1621:  52%|█████▏    | 162/312 [16:23<15:04,  6.03s/it]
Training Epoch: 17/100, batch 324/625 completed (loss: 0.1879:  52%|█████▏    | 162/312 [16:23<15:04,  6.03s/it]
Training Epoch: 17/100, batch 324/625 completed (loss: 0.1879:  52%|█████▏    | 163/312 [16:26<14:56,  6.02s/it]
Training Epoch: 17/100, batch 324/625 completed (loss: 0.0324:  52%|█████▏    | 163/312 [16:26<14:57,  6.03s/it]
Training Epoch: 17/100, batch 324/625 completed (loss: 0.1621:  52%|█████▏    | 163/312 [16:26<14:58,  6.03s/it]
Training Epoch: 17/100, batch 324/625 completed (loss: 0.0204:  52%|█████▏    | 163/312 [16:26<14:58,  6.03s/it]
Training Epoch: 17/100, batch 325/625 completed (loss: 0.3337:  52%|█████▏    | 163/312 [16:26<14:58,  6.03s/it]
Training Epoch: 17/100, batch 325/625 completed (loss: 0.185:  52%|█████▏    | 163/312 [16:26<14:58,  6.03s/it] 
Training Epoch: 17/100, batch 325/625 completed (loss: 0.0856:  52%|█████▏    | 163/312 [16:26<14:57,  6.03s/it]
Training Epoch: 17/100, batch 325/625 completed (loss: 0.4869:  52%|█████▏    | 163/312 [16:27<14:56,  6.02s/it]
Training Epoch: 17/100, batch 326/625 completed (loss: 0.0547:  52%|█████▏    | 163/312 [16:30<14:58,  6.03s/it]
Training Epoch: 17/100, batch 326/625 completed (loss: 0.0084:  52%|█████▏    | 163/312 [16:30<14:57,  6.03s/it]
Training Epoch: 17/100, batch 326/625 completed (loss: 0.0221:  52%|█████▏    | 163/312 [16:30<14:58,  6.03s/it]
Training Epoch: 17/100, batch 326/625 completed (loss: 0.018:  52%|█████▏    | 163/312 [16:30<14:56,  6.02s/it] 
Training Epoch: 17/100, batch 326/625 completed (loss: 0.0084:  53%|█████▎    | 164/312 [16:33<15:04,  6.11s/it]
Training Epoch: 17/100, batch 326/625 completed (loss: 0.0547:  53%|█████▎    | 164/312 [16:33<15:04,  6.11s/it]
Training Epoch: 17/100, batch 326/625 completed (loss: 0.0221:  53%|█████▎    | 164/312 [16:33<15:04,  6.11s/it]
Training Epoch: 17/100, batch 326/625 completed (loss: 0.018:  53%|█████▎    | 164/312 [16:33<15:05,  6.12s/it]
Training Epoch: 17/100, batch 327/625 completed (loss: 0.0153:  53%|█████▎    | 164/312 [16:33<15:04,  6.11s/it]
Training Epoch: 17/100, batch 327/625 completed (loss: 0.059:  53%|█████▎    | 164/312 [16:33<15:04,  6.11s/it] 
Training Epoch: 17/100, batch 327/625 completed (loss: 0.0499:  53%|█████▎    | 164/312 [16:33<15:04,  6.11s/it]
Training Epoch: 17/100, batch 327/625 completed (loss: 0.0261:  53%|█████▎    | 164/312 [16:33<15:05,  6.12s/it]
Training Epoch: 17/100, batch 328/625 completed (loss: 0.0284:  53%|█████▎    | 164/312 [16:36<15:04,  6.11s/it]
Training Epoch: 17/100, batch 328/625 completed (loss: 0.0679:  53%|█████▎    | 164/312 [16:36<15:04,  6.11s/it]
Training Epoch: 17/100, batch 328/625 completed (loss: 0.0743:  53%|█████▎    | 164/312 [16:36<15:04,  6.11s/it]
Training Epoch: 17/100, batch 328/625 completed (loss: 0.0047:  53%|█████▎    | 164/312 [16:36<15:05,  6.12s/it]
Training Epoch: 17/100, batch 328/625 completed (loss: 0.0047:  53%|█████▎    | 165/312 [16:39<14:48,  6.04s/it]
Training Epoch: 17/100, batch 328/625 completed (loss: 0.0743:  53%|█████▎    | 165/312 [16:39<14:48,  6.04s/it]
Training Epoch: 17/100, batch 328/625 completed (loss: 0.0679:  53%|█████▎    | 165/312 [16:39<14:48,  6.04s/it]
Training Epoch: 17/100, batch 328/625 completed (loss: 0.0284:  53%|█████▎    | 165/312 [16:39<14:48,  6.05s/it]
Training Epoch: 17/100, batch 329/625 completed (loss: 0.0201:  53%|█████▎    | 165/312 [16:39<14:48,  6.04s/it]
Training Epoch: 17/100, batch 329/625 completed (loss: 0.009:  53%|█████▎    | 165/312 [16:39<14:48,  6.04s/it] 
Training Epoch: 17/100, batch 329/625 completed (loss: 0.0127:  53%|█████▎    | 165/312 [16:39<14:48,  6.05s/it]
Training Epoch: 17/100, batch 329/625 completed (loss: 0.0289:  53%|█████▎    | 165/312 [16:39<14:48,  6.04s/it]
Training Epoch: 17/100, batch 330/625 completed (loss: 0.1109:  53%|█████▎    | 165/312 [16:42<14:48,  6.05s/it]
Training Epoch: 17/100, batch 330/625 completed (loss: 0.0079:  53%|█████▎    | 165/312 [16:42<14:48,  6.04s/it]
Training Epoch: 17/100, batch 330/625 completed (loss: 0.0031:  53%|█████▎    | 165/312 [16:42<14:48,  6.04s/it]
Training Epoch: 17/100, batch 330/625 completed (loss: 0.0529:  53%|█████▎    | 165/312 [16:42<14:48,  6.04s/it]
Training Epoch: 17/100, batch 330/625 completed (loss: 0.0031:  53%|█████▎    | 166/312 [16:45<14:54,  6.13s/it]
Training Epoch: 17/100, batch 330/625 completed (loss: 0.0079:  53%|█████▎    | 166/312 [16:45<14:54,  6.13s/it]
Training Epoch: 17/100, batch 330/625 completed (loss: 0.1109:  53%|█████▎    | 166/312 [16:45<14:54,  6.13s/it]
Training Epoch: 17/100, batch 330/625 completed (loss: 0.0529:  53%|█████▎    | 166/312 [16:45<14:55,  6.13s/it]
Training Epoch: 17/100, batch 331/625 completed (loss: 0.0038:  53%|█████▎    | 166/312 [16:45<14:54,  6.13s/it]
Training Epoch: 17/100, batch 331/625 completed (loss: 0.1034:  53%|█████▎    | 166/312 [16:45<14:54,  6.13s/it]
Training Epoch: 17/100, batch 331/625 completed (loss: 0.0172:  53%|█████▎    | 166/312 [16:45<14:54,  6.13s/it]
Training Epoch: 17/100, batch 331/625 completed (loss: 0.0199:  53%|█████▎    | 166/312 [16:45<14:55,  6.13s/it]
Training Epoch: 17/100, batch 332/625 completed (loss: 0.0347:  53%|█████▎    | 166/312 [16:48<14:54,  6.13s/it]
Training Epoch: 17/100, batch 332/625 completed (loss: 0.0242:  53%|█████▎    | 166/312 [16:48<14:54,  6.13s/it]
Training Epoch: 17/100, batch 332/625 completed (loss: 0.0064:  53%|█████▎    | 166/312 [16:48<14:54,  6.13s/it]
Training Epoch: 17/100, batch 332/625 completed (loss: 0.0173:  53%|█████▎    | 166/312 [16:48<14:55,  6.13s/it]
Training Epoch: 17/100, batch 332/625 completed (loss: 0.0173:  54%|█████▎    | 167/312 [16:51<14:42,  6.09s/it]
Training Epoch: 17/100, batch 332/625 completed (loss: 0.0064:  54%|█████▎    | 167/312 [16:51<14:42,  6.09s/it]
Training Epoch: 17/100, batch 332/625 completed (loss: 0.0347:  54%|█████▎    | 167/312 [16:51<14:42,  6.09s/it]
Training Epoch: 17/100, batch 332/625 completed (loss: 0.0242:  54%|█████▎    | 167/312 [16:51<14:43,  6.09s/it]
Training Epoch: 17/100, batch 333/625 completed (loss: 0.1457:  54%|█████▎    | 167/312 [16:51<14:42,  6.09s/it]
Training Epoch: 17/100, batch 333/625 completed (loss: 0.0147:  54%|█████▎    | 167/312 [16:51<14:42,  6.09s/it]
Training Epoch: 17/100, batch 333/625 completed (loss: 0.0891:  54%|█████▎    | 167/312 [16:51<14:43,  6.09s/it]
Training Epoch: 17/100, batch 333/625 completed (loss: 0.0068:  54%|█████▎    | 167/312 [16:51<14:42,  6.09s/it]
Training Epoch: 17/100, batch 334/625 completed (loss: 0.0238:  54%|█████▎    | 167/312 [16:54<14:42,  6.09s/it]
Training Epoch: 17/100, batch 334/625 completed (loss: 0.0109:  54%|█████▎    | 167/312 [16:54<14:42,  6.09s/it]
Training Epoch: 17/100, batch 334/625 completed (loss: 0.0251:  54%|█████▎    | 167/312 [16:54<14:43,  6.09s/it]
Training Epoch: 17/100, batch 334/625 completed (loss: 0.021:  54%|█████▎    | 167/312 [16:54<14:42,  6.09s/it] 
Training Epoch: 17/100, batch 334/625 completed (loss: 0.0238:  54%|█████▍    | 168/312 [16:57<14:26,  6.02s/it]
Training Epoch: 17/100, batch 334/625 completed (loss: 0.0109:  54%|█████▍    | 168/312 [16:57<14:26,  6.02s/it]
Training Epoch: 17/100, batch 334/625 completed (loss: 0.0251:  54%|█████▍    | 168/312 [16:57<14:26,  6.02s/it]
Training Epoch: 17/100, batch 334/625 completed (loss: 0.021:  54%|█████▍    | 168/312 [16:57<14:26,  6.02s/it]
Training Epoch: 17/100, batch 335/625 completed (loss: 0.027:  54%|█████▍    | 168/312 [16:57<14:26,  6.02s/it] 
Training Epoch: 17/100, batch 335/625 completed (loss: 0.0152:  54%|█████▍    | 168/312 [16:57<14:26,  6.02s/it]
Training Epoch: 17/100, batch 335/625 completed (loss: 0.0715:  54%|█████▍    | 168/312 [16:57<14:26,  6.02s/it]
Training Epoch: 17/100, batch 335/625 completed (loss: 0.0196:  54%|█████▍    | 168/312 [16:57<14:26,  6.02s/it]
Training Epoch: 17/100, batch 336/625 completed (loss: 0.01:  54%|█████▍    | 168/312 [17:00<14:26,  6.02s/it]  
Training Epoch: 17/100, batch 336/625 completed (loss: 0.0219:  54%|█████▍    | 168/312 [17:00<14:26,  6.02s/it]
Training Epoch: 17/100, batch 336/625 completed (loss: 0.268:  54%|█████▍    | 168/312 [17:00<14:26,  6.02s/it]
Training Epoch: 17/100, batch 336/625 completed (loss: 0.0486:  54%|█████▍    | 168/312 [17:00<14:26,  6.02s/it]
Training Epoch: 17/100, batch 336/625 completed (loss: 0.01:  54%|█████▍    | 169/312 [17:03<14:35,  6.12s/it]
Training Epoch: 17/100, batch 336/625 completed (loss: 0.0219:  54%|█████▍    | 169/312 [17:03<14:35,  6.12s/it]
Training Epoch: 17/100, batch 336/625 completed (loss: 0.0486:  54%|█████▍    | 169/312 [17:03<14:35,  6.12s/it]
Training Epoch: 17/100, batch 336/625 completed (loss: 0.268:  54%|█████▍    | 169/312 [17:03<14:35,  6.12s/it]
Training Epoch: 17/100, batch 337/625 completed (loss: 0.0655:  54%|█████▍    | 169/312 [17:03<14:35,  6.12s/it]
Training Epoch: 17/100, batch 337/625 completed (loss: 0.0143:  54%|█████▍    | 169/312 [17:03<14:35,  6.12s/it]
Training Epoch: 17/100, batch 337/625 completed (loss: 0.0056:  54%|█████▍    | 169/312 [17:03<14:35,  6.12s/it]
Training Epoch: 17/100, batch 337/625 completed (loss: 0.2068:  54%|█████▍    | 169/312 [17:03<14:35,  6.12s/it]
Training Epoch: 17/100, batch 338/625 completed (loss: 0.0109:  54%|█████▍    | 169/312 [17:06<14:35,  6.12s/it]
Training Epoch: 17/100, batch 338/625 completed (loss: 0.0497:  54%|█████▍    | 169/312 [17:06<14:35,  6.12s/it]
Training Epoch: 17/100, batch 338/625 completed (loss: 0.1662:  54%|█████▍    | 169/312 [17:06<14:35,  6.12s/it]
Training Epoch: 17/100, batch 338/625 completed (loss: 0.0014:  54%|█████▍    | 169/312 [17:06<14:35,  6.12s/it]
Training Epoch: 17/100, batch 338/625 completed (loss: 0.1662:  54%|█████▍    | 170/312 [17:09<14:16,  6.03s/it]
Training Epoch: 17/100, batch 338/625 completed (loss: 0.0109:  54%|█████▍    | 170/312 [17:09<14:17,  6.04s/it]
Training Epoch: 17/100, batch 338/625 completed (loss: 0.0497:  54%|█████▍    | 170/312 [17:09<14:18,  6.04s/it]
Training Epoch: 17/100, batch 338/625 completed (loss: 0.0014:  54%|█████▍    | 170/312 [17:09<14:18,  6.04s/it]
Training Epoch: 17/100, batch 339/625 completed (loss: 0.054:  54%|█████▍    | 170/312 [17:09<14:16,  6.03s/it] 
Training Epoch: 17/100, batch 339/625 completed (loss: 0.228:  54%|█████▍    | 170/312 [17:09<14:17,  6.04s/it] 
Training Epoch: 17/100, batch 339/625 completed (loss: 0.084:  54%|█████▍    | 170/312 [17:09<14:18,  6.04s/it] 
Training Epoch: 17/100, batch 339/625 completed (loss: 0.0099:  54%|█████▍    | 170/312 [17:09<14:18,  6.04s/it]
Training Epoch: 17/100, batch 340/625 completed (loss: 0.0245:  54%|█████▍    | 170/312 [17:12<14:17,  6.04s/it]
Training Epoch: 17/100, batch 340/625 completed (loss: 0.0154:  54%|█████▍    | 170/312 [17:12<14:18,  6.04s/it]
Training Epoch: 17/100, batch 340/625 completed (loss: 0.0039:  54%|█████▍    | 170/312 [17:12<14:16,  6.03s/it]
Training Epoch: 17/100, batch 340/625 completed (loss: 0.0075:  54%|█████▍    | 170/312 [17:12<14:18,  6.04s/it]
Training Epoch: 17/100, batch 340/625 completed (loss: 0.0039:  55%|█████▍    | 171/312 [17:15<14:08,  6.02s/it]
Training Epoch: 17/100, batch 340/625 completed (loss: 0.0245:  55%|█████▍    | 171/312 [17:15<14:08,  6.02s/it]
Training Epoch: 17/100, batch 340/625 completed (loss: 0.0154:  55%|█████▍    | 171/312 [17:15<14:08,  6.02s/it]
Training Epoch: 17/100, batch 340/625 completed (loss: 0.0075:  55%|█████▍    | 171/312 [17:15<14:09,  6.02s/it]
Training Epoch: 17/100, batch 341/625 completed (loss: 0.1319:  55%|█████▍    | 171/312 [17:15<14:08,  6.02s/it]
Training Epoch: 17/100, batch 341/625 completed (loss: 0.0615:  55%|█████▍    | 171/312 [17:15<14:08,  6.02s/it]
Training Epoch: 17/100, batch 341/625 completed (loss: 0.0276:  55%|█████▍    | 171/312 [17:15<14:08,  6.02s/it]
Training Epoch: 17/100, batch 341/625 completed (loss: 0.0583:  55%|█████▍    | 171/312 [17:15<14:09,  6.02s/it]
Training Epoch: 17/100, batch 342/625 completed (loss: 0.0345:  55%|█████▍    | 171/312 [17:18<14:08,  6.02s/it]
Training Epoch: 17/100, batch 342/625 completed (loss: 0.0228:  55%|█████▍    | 171/312 [17:18<14:08,  6.02s/it]
Training Epoch: 17/100, batch 342/625 completed (loss: 0.0085:  55%|█████▍    | 171/312 [17:18<14:08,  6.02s/it]
Training Epoch: 17/100, batch 342/625 completed (loss: 0.0591:  55%|█████▍    | 171/312 [17:18<14:09,  6.02s/it]
Training Epoch: 17/100, batch 342/625 completed (loss: 0.0345:  55%|█████▌    | 172/312 [17:21<14:14,  6.11s/it]
Training Epoch: 17/100, batch 342/625 completed (loss: 0.0228:  55%|█████▌    | 172/312 [17:21<14:15,  6.11s/it]
Training Epoch: 17/100, batch 342/625 completed (loss: 0.0591:  55%|█████▌    | 172/312 [17:21<14:14,  6.11s/it]
Training Epoch: 17/100, batch 342/625 completed (loss: 0.0085:  55%|█████▌    | 172/312 [17:21<14:15,  6.11s/it]
Training Epoch: 17/100, batch 343/625 completed (loss: 0.2093:  55%|█████▌    | 172/312 [17:21<14:15,  6.11s/it]
Training Epoch: 17/100, batch 343/625 completed (loss: 0.0436:  55%|█████▌    | 172/312 [17:21<14:14,  6.11s/it]
Training Epoch: 17/100, batch 343/625 completed (loss: 0.0609:  55%|█████▌    | 172/312 [17:21<14:15,  6.11s/it]
Training Epoch: 17/100, batch 343/625 completed (loss: 0.2782:  55%|█████▌    | 172/312 [17:21<14:14,  6.11s/it]
Training Epoch: 17/100, batch 344/625 completed (loss: 0.0494:  55%|█████▌    | 172/312 [17:24<14:15,  6.11s/it]
Training Epoch: 17/100, batch 344/625 completed (loss: 0.0115:  55%|█████▌    | 172/312 [17:24<14:15,  6.11s/it]
Training Epoch: 17/100, batch 344/625 completed (loss: 0.1849:  55%|█████▌    | 172/312 [17:24<14:14,  6.11s/it]
Training Epoch: 17/100, batch 344/625 completed (loss: 0.017:  55%|█████▌    | 172/312 [17:24<14:14,  6.11s/it] 
Training Epoch: 17/100, batch 344/625 completed (loss: 0.1849:  55%|█████▌    | 173/312 [17:27<13:58,  6.04s/it]
Training Epoch: 17/100, batch 344/625 completed (loss: 0.0115:  55%|█████▌    | 173/312 [17:27<13:59,  6.04s/it]
Training Epoch: 17/100, batch 344/625 completed (loss: 0.0494:  55%|█████▌    | 173/312 [17:27<13:59,  6.04s/it]
Training Epoch: 17/100, batch 344/625 completed (loss: 0.017:  55%|█████▌    | 173/312 [17:27<13:59,  6.04s/it]
Training Epoch: 17/100, batch 345/625 completed (loss: 0.0176:  55%|█████▌    | 173/312 [17:27<13:59,  6.04s/it]
Training Epoch: 17/100, batch 345/625 completed (loss: 0.065:  55%|█████▌    | 173/312 [17:27<13:59,  6.04s/it] 
Training Epoch: 17/100, batch 345/625 completed (loss: 0.0393:  55%|█████▌    | 173/312 [17:27<13:58,  6.04s/it]
Training Epoch: 17/100, batch 345/625 completed (loss: 0.1361:  55%|█████▌    | 173/312 [17:27<13:59,  6.04s/it]
Training Epoch: 17/100, batch 346/625 completed (loss: 0.011:  55%|█████▌    | 173/312 [17:30<13:58,  6.04s/it] 
Training Epoch: 17/100, batch 346/625 completed (loss: 0.1013:  55%|█████▌    | 173/312 [17:30<13:59,  6.04s/it]
Training Epoch: 17/100, batch 346/625 completed (loss: 0.0108:  55%|█████▌    | 173/312 [17:30<13:59,  6.04s/it]
Training Epoch: 17/100, batch 346/625 completed (loss: 0.0196:  55%|█████▌    | 173/312 [17:30<13:59,  6.04s/it]
Training Epoch: 17/100, batch 346/625 completed (loss: 0.1013:  56%|█████▌    | 174/312 [17:33<14:05,  6.12s/it]
Training Epoch: 17/100, batch 346/625 completed (loss: 0.011:  56%|█████▌    | 174/312 [17:33<14:05,  6.12s/it]
Training Epoch: 17/100, batch 346/625 completed (loss: 0.0108:  56%|█████▌    | 174/312 [17:33<14:05,  6.12s/it]
Training Epoch: 17/100, batch 346/625 completed (loss: 0.0196:  56%|█████▌    | 174/312 [17:34<14:05,  6.13s/it]
Training Epoch: 17/100, batch 347/625 completed (loss: 0.0429:  56%|█████▌    | 174/312 [17:34<14:05,  6.12s/it]
Training Epoch: 17/100, batch 347/625 completed (loss: 0.0444:  56%|█████▌    | 174/312 [17:34<14:05,  6.12s/it]
Training Epoch: 17/100, batch 347/625 completed (loss: 0.0068:  56%|█████▌    | 174/312 [17:34<14:05,  6.12s/it]
Training Epoch: 17/100, batch 347/625 completed (loss: 0.0165:  56%|█████▌    | 174/312 [17:34<14:05,  6.13s/it]
Training Epoch: 17/100, batch 348/625 completed (loss: 0.0245:  56%|█████▌    | 174/312 [17:36<14:05,  6.12s/it]
Training Epoch: 17/100, batch 348/625 completed (loss: 0.0123:  56%|█████▌    | 174/312 [17:36<14:05,  6.12s/it]
Training Epoch: 17/100, batch 348/625 completed (loss: 0.0139:  56%|█████▌    | 174/312 [17:36<14:05,  6.12s/it]
Training Epoch: 17/100, batch 348/625 completed (loss: 0.0847:  56%|█████▌    | 174/312 [17:37<14:05,  6.13s/it]
Training Epoch: 17/100, batch 348/625 completed (loss: 0.0123:  56%|█████▌    | 175/312 [17:39<13:48,  6.05s/it]
Training Epoch: 17/100, batch 348/625 completed (loss: 0.0245:  56%|█████▌    | 175/312 [17:39<13:48,  6.05s/it]
Training Epoch: 17/100, batch 348/625 completed (loss: 0.0139:  56%|█████▌    | 175/312 [17:39<13:48,  6.05s/it]
Training Epoch: 17/100, batch 348/625 completed (loss: 0.0847:  56%|█████▌    | 175/312 [17:39<13:48,  6.05s/it]
Training Epoch: 17/100, batch 349/625 completed (loss: 0.017:  56%|█████▌    | 175/312 [17:39<13:48,  6.05s/it] 
Training Epoch: 17/100, batch 349/625 completed (loss: 0.0048:  56%|█████▌    | 175/312 [17:39<13:48,  6.05s/it]
Training Epoch: 17/100, batch 349/625 completed (loss: 0.0781:  56%|█████▌    | 175/312 [17:39<13:48,  6.05s/it]
Training Epoch: 17/100, batch 349/625 completed (loss: 0.0065:  56%|█████▌    | 175/312 [17:39<13:48,  6.05s/it]
Training Epoch: 17/100, batch 350/625 completed (loss: 0.0614:  56%|█████▌    | 175/312 [17:42<13:48,  6.05s/it]
Training Epoch: 17/100, batch 350/625 completed (loss: 0.0159:  56%|█████▌    | 175/312 [17:42<13:48,  6.05s/it]
Training Epoch: 17/100, batch 350/625 completed (loss: 0.0183:  56%|█████▌    | 175/312 [17:42<13:48,  6.05s/it]
Training Epoch: 17/100, batch 350/625 completed (loss: 0.3098:  56%|█████▌    | 175/312 [17:42<13:48,  6.05s/it]
Training Epoch: 17/100, batch 350/625 completed (loss: 0.0183:  56%|█████▋    | 176/312 [17:45<13:39,  6.03s/it]
Training Epoch: 17/100, batch 350/625 completed (loss: 0.0614:  56%|█████▋    | 176/312 [17:45<13:39,  6.03s/it]
Training Epoch: 17/100, batch 350/625 completed (loss: 0.0159:  56%|█████▋    | 176/312 [17:45<13:39,  6.03s/it]
Training Epoch: 17/100, batch 350/625 completed (loss: 0.3098:  56%|█████▋    | 176/312 [17:45<13:40,  6.03s/it]
Training Epoch: 17/100, batch 351/625 completed (loss: 0.0014:  56%|█████▋    | 176/312 [17:45<13:39,  6.03s/it]
Training Epoch: 17/100, batch 351/625 completed (loss: 0.0298:  56%|█████▋    | 176/312 [17:45<13:39,  6.03s/it]
Training Epoch: 17/100, batch 351/625 completed (loss: 0.08:  56%|█████▋    | 176/312 [17:45<13:39,  6.03s/it]  
Training Epoch: 17/100, batch 351/625 completed (loss: 0.0081:  56%|█████▋    | 176/312 [17:45<13:40,  6.03s/it]
Training Epoch: 17/100, batch 352/625 completed (loss: 0.4213:  56%|█████▋    | 176/312 [17:49<13:39,  6.03s/it]
Training Epoch: 17/100, batch 352/625 completed (loss: 0.0565:  56%|█████▋    | 176/312 [17:49<13:39,  6.03s/it]
Training Epoch: 17/100, batch 352/625 completed (loss: 0.1266:  56%|█████▋    | 176/312 [17:49<13:39,  6.03s/it]
Training Epoch: 17/100, batch 352/625 completed (loss: 0.0755:  56%|█████▋    | 176/312 [17:49<13:40,  6.03s/it]
Training Epoch: 17/100, batch 352/625 completed (loss: 0.4213:  57%|█████▋    | 177/312 [17:52<13:44,  6.11s/it]
Training Epoch: 17/100, batch 352/625 completed (loss: 0.0565:  57%|█████▋    | 177/312 [17:52<13:44,  6.11s/it]
Training Epoch: 17/100, batch 352/625 completed (loss: 0.1266:  57%|█████▋    | 177/312 [17:52<13:44,  6.11s/it]
Training Epoch: 17/100, batch 352/625 completed (loss: 0.0755:  57%|█████▋    | 177/312 [17:52<13:44,  6.11s/it]
Training Epoch: 17/100, batch 353/625 completed (loss: 0.3585:  57%|█████▋    | 177/312 [17:52<13:44,  6.11s/it]
Training Epoch: 17/100, batch 353/625 completed (loss: 0.1636:  57%|█████▋    | 177/312 [17:52<13:44,  6.11s/it]
Training Epoch: 17/100, batch 353/625 completed (loss: 0.161:  57%|█████▋    | 177/312 [17:52<13:44,  6.11s/it] 
Training Epoch: 17/100, batch 353/625 completed (loss: 0.0102:  57%|█████▋    | 177/312 [17:52<13:44,  6.11s/it]
Training Epoch: 17/100, batch 354/625 completed (loss: 0.2633:  57%|█████▋    | 177/312 [17:55<13:44,  6.11s/it]
Training Epoch: 17/100, batch 354/625 completed (loss: 0.0249:  57%|█████▋    | 177/312 [17:55<13:44,  6.11s/it]
Training Epoch: 17/100, batch 354/625 completed (loss: 0.0082:  57%|█████▋    | 177/312 [17:55<13:44,  6.11s/it]
Training Epoch: 17/100, batch 354/625 completed (loss: 0.1168:  57%|█████▋    | 177/312 [17:55<13:44,  6.11s/it]
Training Epoch: 17/100, batch 354/625 completed (loss: 0.2633:  57%|█████▋    | 178/312 [17:57<13:27,  6.02s/it]
Training Epoch: 17/100, batch 354/625 completed (loss: 0.0082:  57%|█████▋    | 178/312 [17:57<13:27,  6.02s/it]
Training Epoch: 17/100, batch 354/625 completed (loss: 0.0249:  57%|█████▋    | 178/312 [17:57<13:27,  6.02s/it]
Training Epoch: 17/100, batch 354/625 completed (loss: 0.1168:  57%|█████▋    | 178/312 [17:58<13:27,  6.02s/it]
Training Epoch: 17/100, batch 355/625 completed (loss: 0.0037:  57%|█████▋    | 178/312 [17:57<13:27,  6.02s/it]
Training Epoch: 17/100, batch 355/625 completed (loss: 0.0103:  57%|█████▋    | 178/312 [17:57<13:27,  6.02s/it]
Training Epoch: 17/100, batch 355/625 completed (loss: 0.0469:  57%|█████▋    | 178/312 [17:57<13:27,  6.02s/it]
Training Epoch: 17/100, batch 355/625 completed (loss: 0.0108:  57%|█████▋    | 178/312 [17:58<13:27,  6.02s/it]
Training Epoch: 17/100, batch 356/625 completed (loss: 0.0125:  57%|█████▋    | 178/312 [18:01<13:27,  6.02s/it]
Training Epoch: 17/100, batch 356/625 completed (loss: 0.0556:  57%|█████▋    | 178/312 [18:01<13:27,  6.02s/it]
Training Epoch: 17/100, batch 356/625 completed (loss: 0.0009:  57%|█████▋    | 178/312 [18:01<13:27,  6.02s/it]
Training Epoch: 17/100, batch 356/625 completed (loss: 0.019:  57%|█████▋    | 178/312 [18:01<13:27,  6.02s/it] 
Training Epoch: 17/100, batch 356/625 completed (loss: 0.0556:  57%|█████▋    | 179/312 [18:04<13:27,  6.08s/it]
Training Epoch: 17/100, batch 356/625 completed (loss: 0.019:  57%|█████▋    | 179/312 [18:04<13:27,  6.07s/it]
Training Epoch: 17/100, batch 356/625 completed (loss: 0.0009:  57%|█████▋    | 179/312 [18:04<13:28,  6.08s/it]
Training Epoch: 17/100, batch 356/625 completed (loss: 0.0125:  57%|█████▋    | 179/312 [18:04<13:29,  6.08s/it]
Training Epoch: 17/100, batch 357/625 completed (loss: 0.2043:  57%|█████▋    | 179/312 [18:04<13:28,  6.08s/it]
Training Epoch: 17/100, batch 357/625 completed (loss: 0.1728:  57%|█████▋    | 179/312 [18:04<13:29,  6.08s/it]
Training Epoch: 17/100, batch 357/625 completed (loss: 0.0598:  57%|█████▋    | 179/312 [18:04<13:27,  6.08s/it]
Training Epoch: 17/100, batch 357/625 completed (loss: 0.0891:  57%|█████▋    | 179/312 [18:04<13:27,  6.07s/it]
Training Epoch: 17/100, batch 358/625 completed (loss: 0.0292:  57%|█████▋    | 179/312 [18:07<13:27,  6.08s/it]
Training Epoch: 17/100, batch 358/625 completed (loss: 0.0444:  57%|█████▋    | 179/312 [18:07<13:29,  6.08s/it]
Training Epoch: 17/100, batch 358/625 completed (loss: 0.1822:  57%|█████▋    | 179/312 [18:07<13:28,  6.08s/it]
Training Epoch: 17/100, batch 358/625 completed (loss: 0.1075:  57%|█████▋    | 179/312 [18:07<13:27,  6.07s/it]
Training Epoch: 17/100, batch 358/625 completed (loss: 0.0292:  58%|█████▊    | 180/312 [18:10<13:18,  6.05s/it]
Training Epoch: 17/100, batch 358/625 completed (loss: 0.1822:  58%|█████▊    | 180/312 [18:10<13:18,  6.05s/it]
Training Epoch: 17/100, batch 358/625 completed (loss: 0.0444:  58%|█████▊    | 180/312 [18:10<13:18,  6.05s/it]
Training Epoch: 17/100, batch 358/625 completed (loss: 0.1075:  58%|█████▊    | 180/312 [18:10<13:18,  6.05s/it]
Training Epoch: 17/100, batch 359/625 completed (loss: 0.1165:  58%|█████▊    | 180/312 [18:10<13:18,  6.05s/it]
Training Epoch: 17/100, batch 359/625 completed (loss: 0.0149:  58%|█████▊    | 180/312 [18:10<13:18,  6.05s/it]
Training Epoch: 17/100, batch 359/625 completed (loss: 0.1057:  58%|█████▊    | 180/312 [18:10<13:18,  6.05s/it]
Training Epoch: 17/100, batch 359/625 completed (loss: 0.0128:  58%|█████▊    | 180/312 [18:10<13:18,  6.05s/it]
Training Epoch: 17/100, batch 360/625 completed (loss: 0.0023:  58%|█████▊    | 180/312 [18:13<13:18,  6.05s/it]
Training Epoch: 17/100, batch 360/625 completed (loss: 0.0373:  58%|█████▊    | 180/312 [18:13<13:18,  6.05s/it]
Training Epoch: 17/100, batch 360/625 completed (loss: 0.144:  58%|█████▊    | 180/312 [18:13<13:18,  6.05s/it] 
Training Epoch: 17/100, batch 360/625 completed (loss: 0.0315:  58%|█████▊    | 180/312 [18:13<13:18,  6.05s/it]
Training Epoch: 17/100, batch 360/625 completed (loss: 0.0373:  58%|█████▊    | 181/312 [18:15<13:03,  5.98s/it]
Training Epoch: 17/100, batch 360/625 completed (loss: 0.144:  58%|█████▊    | 181/312 [18:15<13:03,  5.98s/it]
Training Epoch: 17/100, batch 360/625 completed (loss: 0.0023:  58%|█████▊    | 181/312 [18:15<13:03,  5.98s/it]
Training Epoch: 17/100, batch 360/625 completed (loss: 0.0315:  58%|█████▊    | 181/312 [18:16<13:04,  5.99s/it]
Training Epoch: 17/100, batch 361/625 completed (loss: 0.0324:  58%|█████▊    | 181/312 [18:16<13:03,  5.98s/it]
Training Epoch: 17/100, batch 361/625 completed (loss: 0.0444:  58%|█████▊    | 181/312 [18:15<13:03,  5.98s/it]
Training Epoch: 17/100, batch 361/625 completed (loss: 0.0108:  58%|█████▊    | 181/312 [18:16<13:03,  5.98s/it]
Training Epoch: 17/100, batch 361/625 completed (loss: 0.0179:  58%|█████▊    | 181/312 [18:16<13:04,  5.99s/it]
Training Epoch: 17/100, batch 362/625 completed (loss: 0.02:  58%|█████▊    | 181/312 [18:19<13:03,  5.98s/it]  
Training Epoch: 17/100, batch 362/625 completed (loss: 0.0017:  58%|█████▊    | 181/312 [18:19<13:03,  5.98s/it]
Training Epoch: 17/100, batch 362/625 completed (loss: 0.0187:  58%|█████▊    | 181/312 [18:19<13:03,  5.98s/it]
Training Epoch: 17/100, batch 362/625 completed (loss: 0.0382:  58%|█████▊    | 181/312 [18:19<13:04,  5.99s/it]
Training Epoch: 17/100, batch 362/625 completed (loss: 0.0187:  58%|█████▊    | 182/312 [18:22<13:09,  6.07s/it]
Training Epoch: 17/100, batch 362/625 completed (loss: 0.02:  58%|█████▊    | 182/312 [18:22<13:09,  6.07s/it]
Training Epoch: 17/100, batch 362/625 completed (loss: 0.0017:  58%|█████▊    | 182/312 [18:22<13:09,  6.07s/it]
Training Epoch: 17/100, batch 362/625 completed (loss: 0.0382:  58%|█████▊    | 182/312 [18:22<13:09,  6.07s/it]
Training Epoch: 17/100, batch 363/625 completed (loss: 0.2772:  58%|█████▊    | 182/312 [18:22<13:09,  6.07s/it]
Training Epoch: 17/100, batch 363/625 completed (loss: 0.0439:  58%|█████▊    | 182/312 [18:22<13:09,  6.07s/it]
Training Epoch: 17/100, batch 363/625 completed (loss: 0.6932:  58%|█████▊    | 182/312 [18:22<13:09,  6.07s/it]
Training Epoch: 17/100, batch 363/625 completed (loss: 0.1638:  58%|█████▊    | 182/312 [18:22<13:09,  6.07s/it]
Training Epoch: 17/100, batch 364/625 completed (loss: 0.0204:  58%|█████▊    | 182/312 [18:25<13:09,  6.07s/it]
Training Epoch: 17/100, batch 364/625 completed (loss: 0.0012:  58%|█████▊    | 182/312 [18:25<13:09,  6.07s/it]
Training Epoch: 17/100, batch 364/625 completed (loss: 0.0245:  58%|█████▊    | 182/312 [18:25<13:09,  6.07s/it]
Training Epoch: 17/100, batch 364/625 completed (loss: 0.0849:  58%|█████▊    | 182/312 [18:25<13:09,  6.07s/it]
Training Epoch: 17/100, batch 364/625 completed (loss: 0.0245:  59%|█████▊    | 183/312 [18:28<13:00,  6.05s/it]
Training Epoch: 17/100, batch 364/625 completed (loss: 0.0849:  59%|█████▊    | 183/312 [18:28<13:00,  6.05s/it]
Training Epoch: 17/100, batch 364/625 completed (loss: 0.0204:  59%|█████▊    | 183/312 [18:28<13:01,  6.06s/it]
Training Epoch: 17/100, batch 364/625 completed (loss: 0.0012:  59%|█████▊    | 183/312 [18:28<13:01,  6.06s/it]
Training Epoch: 17/100, batch 365/625 completed (loss: 0.0186:  59%|█████▊    | 183/312 [18:28<13:00,  6.05s/it]
Training Epoch: 17/100, batch 365/625 completed (loss: 0.1301:  59%|█████▊    | 183/312 [18:28<13:01,  6.06s/it]
Training Epoch: 17/100, batch 365/625 completed (loss: 0.0141:  59%|█████▊    | 183/312 [18:28<13:01,  6.06s/it]
Training Epoch: 17/100, batch 365/625 completed (loss: 0.0032:  59%|█████▊    | 183/312 [18:28<13:00,  6.05s/it]
Training Epoch: 17/100, batch 366/625 completed (loss: 0.0343:  59%|█████▊    | 183/312 [18:31<13:00,  6.05s/it]
Training Epoch: 17/100, batch 366/625 completed (loss: 0.0056:  59%|█████▊    | 183/312 [18:31<13:01,  6.06s/it]
Training Epoch: 17/100, batch 366/625 completed (loss: 0.1816:  59%|█████▊    | 183/312 [18:31<13:01,  6.06s/it]
Training Epoch: 17/100, batch 366/625 completed (loss: 0.0145:  59%|█████▊    | 183/312 [18:31<13:00,  6.05s/it]
Training Epoch: 17/100, batch 366/625 completed (loss: 0.0056:  59%|█████▉    | 184/312 [18:34<12:59,  6.09s/it]
Training Epoch: 17/100, batch 366/625 completed (loss: 0.0145:  59%|█████▉    | 184/312 [18:34<12:59,  6.09s/it]
Training Epoch: 17/100, batch 366/625 completed (loss: 0.0343:  59%|█████▉    | 184/312 [18:34<13:00,  6.10s/it]
Training Epoch: 17/100, batch 366/625 completed (loss: 0.1816:  59%|█████▉    | 184/312 [18:34<13:00,  6.09s/it]
Training Epoch: 17/100, batch 367/625 completed (loss: 0.0268:  59%|█████▉    | 184/312 [18:34<13:00,  6.10s/it]
Training Epoch: 17/100, batch 367/625 completed (loss: 0.3592:  59%|█████▉    | 184/312 [18:34<12:59,  6.09s/it]
Training Epoch: 17/100, batch 367/625 completed (loss: 0.0096:  59%|█████▉    | 184/312 [18:34<13:00,  6.09s/it]
Training Epoch: 17/100, batch 367/625 completed (loss: 0.055:  59%|█████▉    | 184/312 [18:34<12:59,  6.09s/it] 
Training Epoch: 17/100, batch 368/625 completed (loss: 0.0213:  59%|█████▉    | 184/312 [18:37<13:00,  6.09s/it]
Training Epoch: 17/100, batch 368/625 completed (loss: 0.1115:  59%|█████▉    | 184/312 [18:37<13:00,  6.10s/it]
Training Epoch: 17/100, batch 368/625 completed (loss: 0.0038:  59%|█████▉    | 184/312 [18:37<12:59,  6.09s/it]
Training Epoch: 17/100, batch 368/625 completed (loss: 0.0117:  59%|█████▉    | 184/312 [18:37<12:59,  6.09s/it]
Training Epoch: 17/100, batch 368/625 completed (loss: 0.0213:  59%|█████▉    | 185/312 [18:40<12:49,  6.06s/it]
Training Epoch: 17/100, batch 368/625 completed (loss: 0.0038:  59%|█████▉    | 185/312 [18:40<12:49,  6.06s/it]
Training Epoch: 17/100, batch 368/625 completed (loss: 0.0117:  59%|█████▉    | 185/312 [18:40<12:49,  6.06s/it]
Training Epoch: 17/100, batch 368/625 completed (loss: 0.1115:  59%|█████▉    | 185/312 [18:40<12:50,  6.07s/it]
Training Epoch: 17/100, batch 369/625 completed (loss: 0.1673:  59%|█████▉    | 185/312 [18:40<12:50,  6.07s/it]
Training Epoch: 17/100, batch 369/625 completed (loss: 0.0012:  59%|█████▉    | 185/312 [18:40<12:49,  6.06s/it]
Training Epoch: 17/100, batch 369/625 completed (loss: 0.0368:  59%|█████▉    | 185/312 [18:40<12:49,  6.06s/it]
Training Epoch: 17/100, batch 369/625 completed (loss: 0.1059:  59%|█████▉    | 185/312 [18:40<12:49,  6.06s/it]
Training Epoch: 17/100, batch 370/625 completed (loss: 0.0617:  59%|█████▉    | 185/312 [18:43<12:49,  6.06s/it]
Training Epoch: 17/100, batch 370/625 completed (loss: 0.0188:  59%|█████▉    | 185/312 [18:43<12:49,  6.06s/it]
Training Epoch: 17/100, batch 370/625 completed (loss: 0.0285:  59%|█████▉    | 185/312 [18:43<12:50,  6.07s/it]
Training Epoch: 17/100, batch 370/625 completed (loss: 0.0373:  59%|█████▉    | 185/312 [18:43<12:49,  6.06s/it]
Training Epoch: 17/100, batch 370/625 completed (loss: 0.0285:  60%|█████▉    | 186/312 [18:46<12:35,  6.00s/it]
Training Epoch: 17/100, batch 370/625 completed (loss: 0.0188:  60%|█████▉    | 186/312 [18:46<12:35,  6.00s/it]
Training Epoch: 17/100, batch 370/625 completed (loss: 0.0373:  60%|█████▉    | 186/312 [18:46<12:35,  6.00s/it]
Training Epoch: 17/100, batch 370/625 completed (loss: 0.0617:  60%|█████▉    | 186/312 [18:46<12:35,  6.00s/it]
Training Epoch: 17/100, batch 371/625 completed (loss: 0.3478:  60%|█████▉    | 186/312 [18:46<12:35,  6.00s/it]
Training Epoch: 17/100, batch 371/625 completed (loss: 0.0061:  60%|█████▉    | 186/312 [18:46<12:35,  6.00s/it]
Training Epoch: 17/100, batch 371/625 completed (loss: 0.1893:  60%|█████▉    | 186/312 [18:46<12:35,  6.00s/it]
Training Epoch: 17/100, batch 371/625 completed (loss: 0.0166:  60%|█████▉    | 186/312 [18:46<12:35,  6.00s/it]
Training Epoch: 17/100, batch 372/625 completed (loss: 0.2487:  60%|█████▉    | 186/312 [18:49<12:35,  6.00s/it]
Training Epoch: 17/100, batch 372/625 completed (loss: 0.012:  60%|█████▉    | 186/312 [18:49<12:35,  6.00s/it] 
Training Epoch: 17/100, batch 372/625 completed (loss: 0.013:  60%|█████▉    | 186/312 [18:49<12:35,  6.00s/it] 
Training Epoch: 17/100, batch 372/625 completed (loss: 0.1429:  60%|█████▉    | 186/312 [18:49<12:35,  6.00s/it]
Training Epoch: 17/100, batch 372/625 completed (loss: 0.1429:  60%|█████▉    | 187/312 [18:52<12:40,  6.08s/it]
Training Epoch: 17/100, batch 372/625 completed (loss: 0.2487:  60%|█████▉    | 187/312 [18:52<12:40,  6.08s/it]
Training Epoch: 17/100, batch 372/625 completed (loss: 0.013:  60%|█████▉    | 187/312 [18:52<12:40,  6.08s/it]
Training Epoch: 17/100, batch 372/625 completed (loss: 0.012:  60%|█████▉    | 187/312 [18:52<12:40,  6.08s/it]
Training Epoch: 17/100, batch 373/625 completed (loss: 0.0071:  60%|█████▉    | 187/312 [18:52<12:40,  6.08s/it]
Training Epoch: 17/100, batch 373/625 completed (loss: 0.1339:  60%|█████▉    | 187/312 [18:52<12:40,  6.08s/it]
Training Epoch: 17/100, batch 373/625 completed (loss: 0.1588:  60%|█████▉    | 187/312 [18:52<12:40,  6.08s/it]
Training Epoch: 17/100, batch 373/625 completed (loss: 0.0189:  60%|█████▉    | 187/312 [18:52<12:40,  6.08s/it]
Training Epoch: 17/100, batch 374/625 completed (loss: 0.2293:  60%|█████▉    | 187/312 [18:55<12:40,  6.08s/it]
Training Epoch: 17/100, batch 374/625 completed (loss: 0.0306:  60%|█████▉    | 187/312 [18:55<12:40,  6.08s/it]
Training Epoch: 17/100, batch 374/625 completed (loss: 0.0121:  60%|█████▉    | 187/312 [18:55<12:40,  6.08s/it]
Training Epoch: 17/100, batch 374/625 completed (loss: 0.0888:  60%|█████▉    | 187/312 [18:55<12:40,  6.08s/it]
Training Epoch: 17/100, batch 374/625 completed (loss: 0.0888:  60%|██████    | 188/312 [18:58<12:31,  6.06s/it]
Training Epoch: 17/100, batch 374/625 completed (loss: 0.0121:  60%|██████    | 188/312 [18:58<12:31,  6.06s/it]
Training Epoch: 17/100, batch 374/625 completed (loss: 0.2293:  60%|██████    | 188/312 [18:58<12:31,  6.06s/it]
Training Epoch: 17/100, batch 374/625 completed (loss: 0.0306:  60%|██████    | 188/312 [18:58<12:32,  6.07s/it]
Training Epoch: 17/100, batch 375/625 completed (loss: 0.1833:  60%|██████    | 188/312 [18:58<12:31,  6.06s/it]
Training Epoch: 17/100, batch 375/625 completed (loss: 0.0124:  60%|██████    | 188/312 [18:58<12:32,  6.07s/it]
Training Epoch: 17/100, batch 375/625 completed (loss: 0.0759:  60%|██████    | 188/312 [18:58<12:31,  6.06s/it]
Training Epoch: 17/100, batch 375/625 completed (loss: 0.2629:  60%|██████    | 188/312 [18:58<12:31,  6.06s/it]
Training Epoch: 17/100, batch 376/625 completed (loss: 0.041:  60%|██████    | 188/312 [19:01<12:31,  6.06s/it] 
Training Epoch: 17/100, batch 376/625 completed (loss: 0.2561:  60%|██████    | 188/312 [19:01<12:32,  6.07s/it]
Training Epoch: 17/100, batch 376/625 completed (loss: 0.0236:  60%|██████    | 188/312 [19:01<12:31,  6.06s/it]
Training Epoch: 17/100, batch 376/625 completed (loss: 0.02:  60%|██████    | 188/312 [19:01<12:31,  6.06s/it]  
Training Epoch: 17/100, batch 376/625 completed (loss: 0.041:  61%|██████    | 189/312 [19:04<12:23,  6.05s/it]
Training Epoch: 17/100, batch 376/625 completed (loss: 0.02:  61%|██████    | 189/312 [19:04<12:24,  6.05s/it]
Training Epoch: 17/100, batch 376/625 completed (loss: 0.0236:  61%|██████    | 189/312 [19:04<12:24,  6.06s/it]
Training Epoch: 17/100, batch 376/625 completed (loss: 0.2561:  61%|██████    | 189/312 [19:04<12:24,  6.05s/it]
Training Epoch: 17/100, batch 377/625 completed (loss: 0.0063:  61%|██████    | 189/312 [19:04<12:24,  6.06s/it]
Training Epoch: 17/100, batch 377/625 completed (loss: 0.0092:  61%|██████    | 189/312 [19:04<12:24,  6.05s/it]
Training Epoch: 17/100, batch 377/625 completed (loss: 0.1468:  61%|██████    | 189/312 [19:04<12:23,  6.05s/it]
Training Epoch: 17/100, batch 377/625 completed (loss: 0.0158:  61%|██████    | 189/312 [19:04<12:24,  6.05s/it]
Training Epoch: 17/100, batch 378/625 completed (loss: 0.1454:  61%|██████    | 189/312 [19:07<12:24,  6.05s/it]
Training Epoch: 17/100, batch 378/625 completed (loss: 0.0357:  61%|██████    | 189/312 [19:07<12:23,  6.05s/it]
Training Epoch: 17/100, batch 378/625 completed (loss: 0.1098:  61%|██████    | 189/312 [19:07<12:24,  6.06s/it]
Training Epoch: 17/100, batch 378/625 completed (loss: 0.0049:  61%|██████    | 189/312 [19:07<12:24,  6.05s/it]
Training Epoch: 17/100, batch 378/625 completed (loss: 0.1098:  61%|██████    | 190/312 [19:10<12:22,  6.08s/it]
Training Epoch: 17/100, batch 378/625 completed (loss: 0.0049:  61%|██████    | 190/312 [19:10<12:22,  6.08s/it]
Training Epoch: 17/100, batch 378/625 completed (loss: 0.1454:  61%|██████    | 190/312 [19:10<12:22,  6.08s/it]
Training Epoch: 17/100, batch 378/625 completed (loss: 0.0357:  61%|██████    | 190/312 [19:10<12:22,  6.09s/it]
Training Epoch: 17/100, batch 379/625 completed (loss: 0.0483:  61%|██████    | 190/312 [19:10<12:22,  6.08s/it]
Training Epoch: 17/100, batch 379/625 completed (loss: 0.0053:  61%|██████    | 190/312 [19:10<12:22,  6.08s/it]
Training Epoch: 17/100, batch 379/625 completed (loss: 0.0465:  61%|██████    | 190/312 [19:10<12:22,  6.09s/it]
Training Epoch: 17/100, batch 379/625 completed (loss: 0.0433:  61%|██████    | 190/312 [19:10<12:22,  6.08s/it]
Training Epoch: 17/100, batch 380/625 completed (loss: 0.0675:  61%|██████    | 190/312 [19:13<12:22,  6.09s/it]
Training Epoch: 17/100, batch 380/625 completed (loss: 0.0721:  61%|██████    | 190/312 [19:13<12:22,  6.08s/it]
Training Epoch: 17/100, batch 380/625 completed (loss: 0.0602:  61%|██████    | 190/312 [19:13<12:22,  6.08s/it]
Training Epoch: 17/100, batch 380/625 completed (loss: 0.0056:  61%|██████    | 190/312 [19:13<12:22,  6.08s/it]
Training Epoch: 17/100, batch 380/625 completed (loss: 0.0056:  61%|██████    | 191/312 [19:16<12:12,  6.06s/it]
Training Epoch: 17/100, batch 380/625 completed (loss: 0.0602:  61%|██████    | 191/312 [19:16<12:12,  6.06s/it]
Training Epoch: 17/100, batch 380/625 completed (loss: 0.0675:  61%|██████    | 191/312 [19:16<12:13,  6.06s/it]
Training Epoch: 17/100, batch 380/625 completed (loss: 0.0721:  61%|██████    | 191/312 [19:16<12:13,  6.06s/it]
Training Epoch: 17/100, batch 381/625 completed (loss: 0.0157:  61%|██████    | 191/312 [19:16<12:13,  6.06s/it]
Training Epoch: 17/100, batch 381/625 completed (loss: 0.0811:  61%|██████    | 191/312 [19:16<12:12,  6.06s/it]
Training Epoch: 17/100, batch 381/625 completed (loss: 0.0909:  61%|██████    | 191/312 [19:16<12:13,  6.06s/it]
Training Epoch: 17/100, batch 381/625 completed (loss: 0.008:  61%|██████    | 191/312 [19:16<12:12,  6.06s/it] 
Training Epoch: 17/100, batch 382/625 completed (loss: 0.003:  61%|██████    | 191/312 [19:19<12:13,  6.06s/it] 
Training Epoch: 17/100, batch 382/625 completed (loss: 0.0319:  61%|██████    | 191/312 [19:19<12:12,  6.06s/it]
Training Epoch: 17/100, batch 382/625 completed (loss: 0.0323:  61%|██████    | 191/312 [19:19<12:13,  6.06s/it]
Training Epoch: 17/100, batch 382/625 completed (loss: 0.2953:  61%|██████    | 191/312 [19:19<12:12,  6.06s/it]
Training Epoch: 17/100, batch 382/625 completed (loss: 0.2953:  62%|██████▏   | 192/312 [19:23<12:14,  6.12s/it]
Training Epoch: 17/100, batch 382/625 completed (loss: 0.0319:  62%|██████▏   | 192/312 [19:23<12:14,  6.12s/it]
Training Epoch: 17/100, batch 382/625 completed (loss: 0.003:  62%|██████▏   | 192/312 [19:23<12:14,  6.12s/it]
Training Epoch: 17/100, batch 382/625 completed (loss: 0.0323:  62%|██████▏   | 192/312 [19:23<12:14,  6.12s/it]
Training Epoch: 17/100, batch 383/625 completed (loss: 0.0126:  62%|██████▏   | 192/312 [19:23<12:14,  6.12s/it]
Training Epoch: 17/100, batch 383/625 completed (loss: 0.0147:  62%|██████▏   | 192/312 [19:23<12:14,  6.12s/it]
Training Epoch: 17/100, batch 383/625 completed (loss: 0.0224:  62%|██████▏   | 192/312 [19:23<12:14,  6.12s/it]
Training Epoch: 17/100, batch 383/625 completed (loss: 0.0307:  62%|██████▏   | 192/312 [19:23<12:14,  6.12s/it]
Training Epoch: 17/100, batch 384/625 completed (loss: 0.0237:  62%|██████▏   | 192/312 [19:25<12:14,  6.12s/it]
Training Epoch: 17/100, batch 384/625 completed (loss: 0.0649:  62%|██████▏   | 192/312 [19:25<12:14,  6.12s/it]
Training Epoch: 17/100, batch 384/625 completed (loss: 0.0491:  62%|██████▏   | 192/312 [19:25<12:14,  6.12s/it]
Training Epoch: 17/100, batch 384/625 completed (loss: 0.0458:  62%|██████▏   | 192/312 [19:26<12:14,  6.12s/it]
Training Epoch: 17/100, batch 384/625 completed (loss: 0.0649:  62%|██████▏   | 193/312 [19:29<12:03,  6.08s/it]
Training Epoch: 17/100, batch 384/625 completed (loss: 0.0237:  62%|██████▏   | 193/312 [19:29<12:03,  6.08s/it]
Training Epoch: 17/100, batch 384/625 completed (loss: 0.0458:  62%|██████▏   | 193/312 [19:29<12:03,  6.08s/it]
Training Epoch: 17/100, batch 384/625 completed (loss: 0.0491:  62%|██████▏   | 193/312 [19:29<12:03,  6.08s/it]
Training Epoch: 17/100, batch 385/625 completed (loss: 0.005:  62%|██████▏   | 193/312 [19:29<12:03,  6.08s/it] 
Training Epoch: 17/100, batch 385/625 completed (loss: 0.0204:  62%|██████▏   | 193/312 [19:29<12:03,  6.08s/it]
Training Epoch: 17/100, batch 385/625 completed (loss: 0.0056:  62%|██████▏   | 193/312 [19:29<12:03,  6.08s/it]
Training Epoch: 17/100, batch 385/625 completed (loss: 0.0336:  62%|██████▏   | 193/312 [19:29<12:03,  6.08s/it]
Training Epoch: 17/100, batch 386/625 completed (loss: 0.3692:  62%|██████▏   | 193/312 [19:31<12:03,  6.08s/it]
Training Epoch: 17/100, batch 386/625 completed (loss: 0.1209:  62%|██████▏   | 193/312 [19:31<12:03,  6.08s/it]
Training Epoch: 17/100, batch 386/625 completed (loss: 0.0265:  62%|██████▏   | 193/312 [19:31<12:03,  6.08s/it]
Training Epoch: 17/100, batch 386/625 completed (loss: 0.0136:  62%|██████▏   | 193/312 [19:32<12:03,  6.08s/it]
Training Epoch: 17/100, batch 386/625 completed (loss: 0.3692:  62%|██████▏   | 194/312 [19:35<11:54,  6.05s/it]
Training Epoch: 17/100, batch 386/625 completed (loss: 0.1209:  62%|██████▏   | 194/312 [19:35<11:54,  6.05s/it]
Training Epoch: 17/100, batch 386/625 completed (loss: 0.0136:  62%|██████▏   | 194/312 [19:35<11:54,  6.06s/it]
Training Epoch: 17/100, batch 386/625 completed (loss: 0.0265:  62%|██████▏   | 194/312 [19:35<11:54,  6.05s/it]
Training Epoch: 17/100, batch 387/625 completed (loss: 0.0063:  62%|██████▏   | 194/312 [19:35<11:54,  6.05s/it]
Training Epoch: 17/100, batch 387/625 completed (loss: 0.011:  62%|██████▏   | 194/312 [19:35<11:54,  6.05s/it] 
Training Epoch: 17/100, batch 387/625 completed (loss: 0.0088:  62%|██████▏   | 194/312 [19:35<11:54,  6.05s/it]
Training Epoch: 17/100, batch 387/625 completed (loss: 0.1139:  62%|██████▏   | 194/312 [19:35<11:54,  6.06s/it]
Training Epoch: 17/100, batch 388/625 completed (loss: 0.0288:  62%|██████▏   | 194/312 [19:38<11:54,  6.05s/it]
Training Epoch: 17/100, batch 388/625 completed (loss: 0.1243:  62%|██████▏   | 194/312 [19:38<11:54,  6.05s/it]
Training Epoch: 17/100, batch 388/625 completed (loss: 0.434:  62%|██████▏   | 194/312 [19:38<11:54,  6.05s/it]
Training Epoch: 17/100, batch 388/625 completed (loss: 0.0064:  62%|██████▏   | 194/312 [19:38<11:54,  6.06s/it]
Training Epoch: 17/100, batch 388/625 completed (loss: 0.0064:  62%|██████▎   | 195/312 [19:41<11:53,  6.09s/it]
Training Epoch: 17/100, batch 388/625 completed (loss: 0.1243:  62%|██████▎   | 195/312 [19:41<11:53,  6.09s/it]
Training Epoch: 17/100, batch 388/625 completed (loss: 0.0288:  62%|██████▎   | 195/312 [19:41<11:53,  6.10s/it]
Training Epoch: 17/100, batch 388/625 completed (loss: 0.434:  62%|██████▎   | 195/312 [19:41<11:54,  6.10s/it]
Training Epoch: 17/100, batch 389/625 completed (loss: 0.0879:  62%|██████▎   | 195/312 [19:41<11:53,  6.09s/it]
Training Epoch: 17/100, batch 389/625 completed (loss: 0.0166:  62%|██████▎   | 195/312 [19:41<11:54,  6.10s/it]
Training Epoch: 17/100, batch 389/625 completed (loss: 0.0112:  62%|██████▎   | 195/312 [19:41<11:53,  6.10s/it]
Training Epoch: 17/100, batch 389/625 completed (loss: 0.1863:  62%|██████▎   | 195/312 [19:41<11:53,  6.09s/it]
Training Epoch: 17/100, batch 390/625 completed (loss: 0.0157:  62%|██████▎   | 195/312 [19:44<11:53,  6.09s/it]
Training Epoch: 17/100, batch 390/625 completed (loss: 0.022:  62%|██████▎   | 195/312 [19:44<11:53,  6.10s/it] 
Training Epoch: 17/100, batch 390/625 completed (loss: 0.3828:  62%|██████▎   | 195/312 [19:44<11:54,  6.10s/it]
Training Epoch: 17/100, batch 390/625 completed (loss: 0.4008:  62%|██████▎   | 195/312 [19:44<11:53,  6.09s/it]
Training Epoch: 17/100, batch 390/625 completed (loss: 0.022:  63%|██████▎   | 196/312 [19:47<11:44,  6.07s/it]
Training Epoch: 17/100, batch 390/625 completed (loss: 0.0157:  63%|██████▎   | 196/312 [19:47<11:44,  6.07s/it]
Training Epoch: 17/100, batch 390/625 completed (loss: 0.3828:  63%|██████▎   | 196/312 [19:47<11:44,  6.07s/it]
Training Epoch: 17/100, batch 390/625 completed (loss: 0.4008:  63%|██████▎   | 196/312 [19:47<11:45,  6.08s/it]
Training Epoch: 17/100, batch 391/625 completed (loss: 0.0161:  63%|██████▎   | 196/312 [19:47<11:44,  6.07s/it]
Training Epoch: 17/100, batch 391/625 completed (loss: 0.0268:  63%|██████▎   | 196/312 [19:47<11:44,  6.07s/it]
Training Epoch: 17/100, batch 391/625 completed (loss: 0.0088:  63%|██████▎   | 196/312 [19:47<11:44,  6.07s/it]
Training Epoch: 17/100, batch 391/625 completed (loss: 0.0176:  63%|██████▎   | 196/312 [19:47<11:45,  6.08s/it]
Training Epoch: 17/100, batch 392/625 completed (loss: 0.0858:  63%|██████▎   | 196/312 [19:50<11:44,  6.07s/it]
Training Epoch: 17/100, batch 392/625 completed (loss: 0.024:  63%|██████▎   | 196/312 [19:50<11:44,  6.07s/it] 
Training Epoch: 17/100, batch 392/625 completed (loss: 0.0051:  63%|██████▎   | 196/312 [19:50<11:44,  6.07s/it]
Training Epoch: 17/100, batch 392/625 completed (loss: 0.0008:  63%|██████▎   | 196/312 [19:50<11:45,  6.08s/it]
Training Epoch: 17/100, batch 392/625 completed (loss: 0.024:  63%|██████▎   | 197/312 [19:53<11:40,  6.09s/it]
Training Epoch: 17/100, batch 392/625 completed (loss: 0.0008:  63%|██████▎   | 197/312 [19:53<11:40,  6.09s/it]
Training Epoch: 17/100, batch 392/625 completed (loss: 0.0858:  63%|██████▎   | 197/312 [19:53<11:41,  6.10s/it]
Training Epoch: 17/100, batch 392/625 completed (loss: 0.0051:  63%|██████▎   | 197/312 [19:53<11:41,  6.10s/it]
Training Epoch: 17/100, batch 393/625 completed (loss: 0.0082:  63%|██████▎   | 197/312 [19:53<11:41,  6.10s/it]
Training Epoch: 17/100, batch 393/625 completed (loss: 0.0207:  63%|██████▎   | 197/312 [19:53<11:40,  6.09s/it]
Training Epoch: 17/100, batch 393/625 completed (loss: 0.4628:  63%|██████▎   | 197/312 [19:53<11:41,  6.10s/it]
Training Epoch: 17/100, batch 393/625 completed (loss: 0.0023:  63%|██████▎   | 197/312 [19:53<11:40,  6.09s/it]
Training Epoch: 17/100, batch 394/625 completed (loss: 0.5582:  63%|██████▎   | 197/312 [19:56<11:40,  6.09s/it]
Training Epoch: 17/100, batch 394/625 completed (loss: 0.0769:  63%|██████▎   | 197/312 [19:56<11:41,  6.10s/it]
Training Epoch: 17/100, batch 394/625 completed (loss: 0.0345:  63%|██████▎   | 197/312 [19:56<11:41,  6.10s/it]
Training Epoch: 17/100, batch 394/625 completed (loss: 0.0261:  63%|██████▎   | 197/312 [19:56<11:40,  6.09s/it]
Training Epoch: 17/100, batch 394/625 completed (loss: 0.0261:  63%|██████▎   | 198/312 [19:59<11:32,  6.07s/it]
Training Epoch: 17/100, batch 394/625 completed (loss: 0.0769:  63%|██████▎   | 198/312 [19:59<11:32,  6.07s/it]
Training Epoch: 17/100, batch 394/625 completed (loss: 0.5582:  63%|██████▎   | 198/312 [19:59<11:32,  6.07s/it]
Training Epoch: 17/100, batch 394/625 completed (loss: 0.0345:  63%|██████▎   | 198/312 [19:59<11:32,  6.08s/it]
Training Epoch: 17/100, batch 395/625 completed (loss: 0.0026:  63%|██████▎   | 198/312 [19:59<11:32,  6.07s/it]
Training Epoch: 17/100, batch 395/625 completed (loss: 0.0068:  63%|██████▎   | 198/312 [19:59<11:32,  6.08s/it]
Training Epoch: 17/100, batch 395/625 completed (loss: 0.0146:  63%|██████▎   | 198/312 [19:59<11:32,  6.07s/it]
Training Epoch: 17/100, batch 395/625 completed (loss: 0.0375:  63%|██████▎   | 198/312 [19:59<11:32,  6.07s/it]
Training Epoch: 17/100, batch 396/625 completed (loss: 0.0546:  63%|██████▎   | 198/312 [20:02<11:32,  6.07s/it]
Training Epoch: 17/100, batch 396/625 completed (loss: 0.0103:  63%|██████▎   | 198/312 [20:02<11:32,  6.08s/it]
Training Epoch: 17/100, batch 396/625 completed (loss: 0.206:  63%|██████▎   | 198/312 [20:02<11:32,  6.07s/it] 
Training Epoch: 17/100, batch 396/625 completed (loss: 0.0338:  63%|██████▎   | 198/312 [20:02<11:32,  6.07s/it]
Training Epoch: 17/100, batch 396/625 completed (loss: 0.0338:  64%|██████▍   | 199/312 [20:05<11:22,  6.04s/it]
Training Epoch: 17/100, batch 396/625 completed (loss: 0.0546:  64%|██████▍   | 199/312 [20:05<11:23,  6.04s/it]
Training Epoch: 17/100, batch 396/625 completed (loss: 0.206:  64%|██████▍   | 199/312 [20:05<11:23,  6.05s/it]
Training Epoch: 17/100, batch 396/625 completed (loss: 0.0103:  64%|██████▍   | 199/312 [20:05<11:23,  6.05s/it]
Training Epoch: 17/100, batch 397/625 completed (loss: 0.0076:  64%|██████▍   | 199/312 [20:05<11:23,  6.04s/it]
Training Epoch: 17/100, batch 397/625 completed (loss: 0.0499:  64%|██████▍   | 199/312 [20:05<11:23,  6.05s/it]
Training Epoch: 17/100, batch 397/625 completed (loss: 0.4058:  64%|██████▍   | 199/312 [20:05<11:23,  6.05s/it]
Training Epoch: 17/100, batch 397/625 completed (loss: 0.0265:  64%|██████▍   | 199/312 [20:05<11:22,  6.04s/it]
Training Epoch: 17/100, batch 398/625 completed (loss: 0.0203:  64%|██████▍   | 199/312 [20:08<11:23,  6.04s/it]
Training Epoch: 17/100, batch 398/625 completed (loss: 0.0535:  64%|██████▍   | 199/312 [20:08<11:23,  6.05s/it]
Training Epoch: 17/100, batch 398/625 completed (loss: 0.003:  64%|██████▍   | 199/312 [20:08<11:23,  6.05s/it] 
Training Epoch: 17/100, batch 398/625 completed (loss: 0.0274:  64%|██████▍   | 199/312 [20:08<11:22,  6.04s/it]
Training Epoch: 17/100, batch 398/625 completed (loss: 0.0203:  64%|██████▍   | 200/312 [20:11<11:26,  6.13s/it]
Training Epoch: 17/100, batch 398/625 completed (loss: 0.0535:  64%|██████▍   | 200/312 [20:11<11:27,  6.14s/it]
Training Epoch: 17/100, batch 398/625 completed (loss: 0.0274:  64%|██████▍   | 200/312 [20:11<11:27,  6.14s/it]
Training Epoch: 17/100, batch 398/625 completed (loss: 0.003:  64%|██████▍   | 200/312 [20:11<11:27,  6.14s/it]
Training Epoch: 17/100, batch 399/625 completed (loss: 0.0044:  64%|██████▍   | 200/312 [20:11<11:27,  6.14s/it]
Training Epoch: 17/100, batch 399/625 completed (loss: 0.2367:  64%|██████▍   | 200/312 [20:11<11:26,  6.13s/it]
Training Epoch: 17/100, batch 399/625 completed (loss: 0.0822:  64%|██████▍   | 200/312 [20:11<11:27,  6.14s/it]
Training Epoch: 17/100, batch 399/625 completed (loss: 0.2573:  64%|██████▍   | 200/312 [20:11<11:27,  6.14s/it]
Training Epoch: 17/100, batch 400/625 completed (loss: 0.0562:  64%|██████▍   | 200/312 [20:14<11:26,  6.13s/it]
Training Epoch: 17/100, batch 400/625 completed (loss: 0.0093:  64%|██████▍   | 200/312 [20:14<11:27,  6.14s/it]
Training Epoch: 17/100, batch 400/625 completed (loss: 0.0161:  64%|██████▍   | 200/312 [20:14<11:27,  6.14s/it]
Training Epoch: 17/100, batch 400/625 completed (loss: 0.7402:  64%|██████▍   | 200/312 [20:14<11:27,  6.14s/it]
Training Epoch: 17/100, batch 400/625 completed (loss: 0.0093:  64%|██████▍   | 201/312 [20:17<11:18,  6.11s/it]
Training Epoch: 17/100, batch 400/625 completed (loss: 0.0562:  64%|██████▍   | 201/312 [20:17<11:18,  6.11s/it]
Training Epoch: 17/100, batch 400/625 completed (loss: 0.7402:  64%|██████▍   | 201/312 [20:17<11:18,  6.11s/it]
Training Epoch: 17/100, batch 400/625 completed (loss: 0.0161:  64%|██████▍   | 201/312 [20:17<11:18,  6.11s/it]
Training Epoch: 17/100, batch 401/625 completed (loss: 0.4049:  64%|██████▍   | 201/312 [20:17<11:18,  6.11s/it]
Training Epoch: 17/100, batch 401/625 completed (loss: 0.2505:  64%|██████▍   | 201/312 [20:17<11:18,  6.11s/it]
Training Epoch: 17/100, batch 401/625 completed (loss: 0.0091:  64%|██████▍   | 201/312 [20:17<11:18,  6.11s/it]
Training Epoch: 17/100, batch 401/625 completed (loss: 0.0137:  64%|██████▍   | 201/312 [20:17<11:18,  6.11s/it]
Training Epoch: 17/100, batch 402/625 completed (loss: 0.1803:  64%|██████▍   | 201/312 [20:20<11:18,  6.11s/it]
Training Epoch: 17/100, batch 402/625 completed (loss: 0.0216:  64%|██████▍   | 201/312 [20:20<11:18,  6.11s/it]
Training Epoch: 17/100, batch 402/625 completed (loss: 0.0783:  64%|██████▍   | 201/312 [20:20<11:18,  6.11s/it]
Training Epoch: 17/100, batch 402/625 completed (loss: 0.0243:  64%|██████▍   | 201/312 [20:20<11:18,  6.11s/it]
Training Epoch: 17/100, batch 402/625 completed (loss: 0.0216:  65%|██████▍   | 202/312 [20:23<11:12,  6.12s/it]
Training Epoch: 17/100, batch 402/625 completed (loss: 0.1803:  65%|██████▍   | 202/312 [20:23<11:13,  6.12s/it]
Training Epoch: 17/100, batch 402/625 completed (loss: 0.0783:  65%|██████▍   | 202/312 [20:23<11:12,  6.11s/it]
Training Epoch: 17/100, batch 402/625 completed (loss: 0.0243:  65%|██████▍   | 202/312 [20:23<11:13,  6.12s/it]
Training Epoch: 17/100, batch 403/625 completed (loss: 0.005:  65%|██████▍   | 202/312 [20:23<11:12,  6.11s/it] 
Training Epoch: 17/100, batch 403/625 completed (loss: 0.0204:  65%|██████▍   | 202/312 [20:23<11:12,  6.12s/it]
Training Epoch: 17/100, batch 403/625 completed (loss: 0.0061:  65%|██████▍   | 202/312 [20:23<11:13,  6.12s/it]
Training Epoch: 17/100, batch 403/625 completed (loss: 0.0193:  65%|██████▍   | 202/312 [20:24<11:13,  6.12s/it]
Training Epoch: 17/100, batch 404/625 completed (loss: 0.2278:  65%|██████▍   | 202/312 [20:27<11:13,  6.12s/it]
Training Epoch: 17/100, batch 404/625 completed (loss: 0.024:  65%|██████▍   | 202/312 [20:27<11:12,  6.12s/it] 
Training Epoch: 17/100, batch 404/625 completed (loss: 0.156:  65%|██████▍   | 202/312 [20:27<11:12,  6.11s/it]
Training Epoch: 17/100, batch 404/625 completed (loss: 0.1625:  65%|██████▍   | 202/312 [20:27<11:13,  6.12s/it]
Training Epoch: 17/100, batch 404/625 completed (loss: 0.1625:  65%|██████▌   | 203/312 [20:29<11:02,  6.08s/it]
Training Epoch: 17/100, batch 404/625 completed (loss: 0.024:  65%|██████▌   | 203/312 [20:29<11:02,  6.08s/it]
Training Epoch: 17/100, batch 404/625 completed (loss: 0.2278:  65%|██████▌   | 203/312 [20:29<11:03,  6.08s/it]
Training Epoch: 17/100, batch 404/625 completed (loss: 0.156:  65%|██████▌   | 203/312 [20:29<11:02,  6.08s/it]
Training Epoch: 17/100, batch 405/625 completed (loss: 0.0093:  65%|██████▌   | 203/312 [20:29<11:02,  6.08s/it]
Training Epoch: 17/100, batch 405/625 completed (loss: 0.0989:  65%|██████▌   | 203/312 [20:29<11:03,  6.08s/it]
Training Epoch: 17/100, batch 405/625 completed (loss: 0.0545:  65%|██████▌   | 203/312 [20:29<11:02,  6.08s/it]
Training Epoch: 17/100, batch 405/625 completed (loss: 0.0075:  65%|██████▌   | 203/312 [20:29<11:02,  6.08s/it]
Training Epoch: 17/100, batch 406/625 completed (loss: 0.0186:  65%|██████▌   | 203/312 [20:33<11:03,  6.08s/it]
Training Epoch: 17/100, batch 406/625 completed (loss: 0.0119:  65%|██████▌   | 203/312 [20:33<11:02,  6.08s/it]
Training Epoch: 17/100, batch 406/625 completed (loss: 0.2236:  65%|██████▌   | 203/312 [20:33<11:02,  6.08s/it]
Training Epoch: 17/100, batch 406/625 completed (loss: 0.0255:  65%|██████▌   | 203/312 [20:33<11:02,  6.08s/it]
Training Epoch: 17/100, batch 406/625 completed (loss: 0.0186:  65%|██████▌   | 204/312 [20:35<10:54,  6.06s/it]
Training Epoch: 17/100, batch 406/625 completed (loss: 0.0119:  65%|██████▌   | 204/312 [20:35<10:55,  6.07s/it]
Training Epoch: 17/100, batch 406/625 completed (loss: 0.2236:  65%|██████▌   | 204/312 [20:35<10:54,  6.06s/it]
Training Epoch: 17/100, batch 406/625 completed (loss: 0.0255:  65%|██████▌   | 204/312 [20:35<10:55,  6.07s/it]
Training Epoch: 17/100, batch 407/625 completed (loss: 0.3507:  65%|██████▌   | 204/312 [20:35<10:54,  6.06s/it]
Training Epoch: 17/100, batch 407/625 completed (loss: 0.0032:  65%|██████▌   | 204/312 [20:35<10:54,  6.06s/it]
Training Epoch: 17/100, batch 407/625 completed (loss: 0.1191:  65%|██████▌   | 204/312 [20:35<10:55,  6.07s/it]
Training Epoch: 17/100, batch 407/625 completed (loss: 0.0132:  65%|██████▌   | 204/312 [20:36<10:55,  6.07s/it]
Training Epoch: 17/100, batch 408/625 completed (loss: 0.0419:  65%|██████▌   | 204/312 [20:39<10:55,  6.07s/it]
Training Epoch: 17/100, batch 408/625 completed (loss: 0.0189:  65%|██████▌   | 204/312 [20:39<10:54,  6.06s/it]
Training Epoch: 17/100, batch 408/625 completed (loss: 0.1676:  65%|██████▌   | 204/312 [20:39<10:54,  6.06s/it]
Training Epoch: 17/100, batch 408/625 completed (loss: 0.0049:  65%|██████▌   | 204/312 [20:39<10:55,  6.07s/it]
Training Epoch: 17/100, batch 408/625 completed (loss: 0.0049:  66%|██████▌   | 205/312 [20:42<10:56,  6.14s/it]
Training Epoch: 17/100, batch 408/625 completed (loss: 0.0419:  66%|██████▌   | 205/312 [20:42<10:56,  6.14s/it]
Training Epoch: 17/100, batch 408/625 completed (loss: 0.0189:  66%|██████▌   | 205/312 [20:42<10:57,  6.14s/it]
Training Epoch: 17/100, batch 408/625 completed (loss: 0.1676:  66%|██████▌   | 205/312 [20:42<10:56,  6.14s/it]
Training Epoch: 17/100, batch 409/625 completed (loss: 0.0518:  66%|██████▌   | 205/312 [20:42<10:57,  6.14s/it]
Training Epoch: 17/100, batch 409/625 completed (loss: 0.0397:  66%|██████▌   | 205/312 [20:42<10:56,  6.14s/it]
Training Epoch: 17/100, batch 409/625 completed (loss: 0.0234:  66%|██████▌   | 205/312 [20:42<10:56,  6.14s/it]
Training Epoch: 17/100, batch 409/625 completed (loss: 0.0198:  66%|██████▌   | 205/312 [20:42<10:56,  6.14s/it]
Training Epoch: 17/100, batch 410/625 completed (loss: 0.0117:  66%|██████▌   | 205/312 [20:45<10:57,  6.14s/it]
Training Epoch: 17/100, batch 410/625 completed (loss: 0.0106:  66%|██████▌   | 205/312 [20:45<10:56,  6.14s/it]
Training Epoch: 17/100, batch 410/625 completed (loss: 0.4881:  66%|██████▌   | 205/312 [20:45<10:56,  6.14s/it]
Training Epoch: 17/100, batch 410/625 completed (loss: 0.0322:  66%|██████▌   | 205/312 [20:45<10:56,  6.14s/it]
Training Epoch: 17/100, batch 410/625 completed (loss: 0.0106:  66%|██████▌   | 206/312 [20:48<10:45,  6.09s/it]
Training Epoch: 17/100, batch 410/625 completed (loss: 0.0117:  66%|██████▌   | 206/312 [20:48<10:45,  6.09s/it]
Training Epoch: 17/100, batch 410/625 completed (loss: 0.0322:  66%|██████▌   | 206/312 [20:48<10:45,  6.09s/it]
Training Epoch: 17/100, batch 410/625 completed (loss: 0.4881:  66%|██████▌   | 206/312 [20:48<10:45,  6.09s/it]
Training Epoch: 17/100, batch 411/625 completed (loss: 0.0743:  66%|██████▌   | 206/312 [20:48<10:45,  6.09s/it]
Training Epoch: 17/100, batch 411/625 completed (loss: 0.0225:  66%|██████▌   | 206/312 [20:48<10:45,  6.09s/it]
Training Epoch: 17/100, batch 411/625 completed (loss: 0.0158:  66%|██████▌   | 206/312 [20:48<10:45,  6.09s/it]
Training Epoch: 17/100, batch 411/625 completed (loss: 0.5817:  66%|██████▌   | 206/312 [20:48<10:45,  6.09s/it]
Training Epoch: 17/100, batch 412/625 completed (loss: 0.0203:  66%|██████▌   | 206/312 [20:51<10:45,  6.09s/it]
Training Epoch: 17/100, batch 412/625 completed (loss: 0.0285:  66%|██████▌   | 206/312 [20:51<10:45,  6.09s/it]
Training Epoch: 17/100, batch 412/625 completed (loss: 0.1361:  66%|██████▌   | 206/312 [20:51<10:45,  6.09s/it]
Training Epoch: 17/100, batch 412/625 completed (loss: 0.0142:  66%|██████▌   | 206/312 [20:51<10:45,  6.09s/it]
Training Epoch: 17/100, batch 412/625 completed (loss: 0.0142:  66%|██████▋   | 207/312 [20:54<10:38,  6.08s/it]
Training Epoch: 17/100, batch 412/625 completed (loss: 0.1361:  66%|██████▋   | 207/312 [20:54<10:38,  6.08s/it]
Training Epoch: 17/100, batch 412/625 completed (loss: 0.0203:  66%|██████▋   | 207/312 [20:54<10:38,  6.08s/it]
Training Epoch: 17/100, batch 412/625 completed (loss: 0.0285:  66%|██████▋   | 207/312 [20:54<10:38,  6.08s/it]
Training Epoch: 17/100, batch 413/625 completed (loss: 0.0503:  66%|██████▋   | 207/312 [20:54<10:38,  6.08s/it]
Training Epoch: 17/100, batch 413/625 completed (loss: 0.0617:  66%|██████▋   | 207/312 [20:54<10:38,  6.08s/it]
Training Epoch: 17/100, batch 413/625 completed (loss: 0.0255:  66%|██████▋   | 207/312 [20:54<10:38,  6.08s/it]
Training Epoch: 17/100, batch 413/625 completed (loss: 0.1962:  66%|██████▋   | 207/312 [20:54<10:38,  6.08s/it]
Training Epoch: 17/100, batch 414/625 completed (loss: 0.0228:  66%|██████▋   | 207/312 [20:57<10:38,  6.08s/it]
Training Epoch: 17/100, batch 414/625 completed (loss: 0.0161:  66%|██████▋   | 207/312 [20:57<10:38,  6.08s/it]
Training Epoch: 17/100, batch 414/625 completed (loss: 0.0069:  66%|██████▋   | 207/312 [20:57<10:38,  6.08s/it]
Training Epoch: 17/100, batch 414/625 completed (loss: 0.0098:  66%|██████▋   | 207/312 [20:57<10:38,  6.08s/it]
Training Epoch: 17/100, batch 414/625 completed (loss: 0.0098:  67%|██████▋   | 208/312 [21:00<10:35,  6.12s/it]
Training Epoch: 17/100, batch 414/625 completed (loss: 0.0161:  67%|██████▋   | 208/312 [21:00<10:36,  6.12s/it]
Training Epoch: 17/100, batch 414/625 completed (loss: 0.0069:  67%|██████▋   | 208/312 [21:00<10:36,  6.12s/it]
Training Epoch: 17/100, batch 414/625 completed (loss: 0.0228:  67%|██████▋   | 208/312 [21:00<10:36,  6.12s/it]
Training Epoch: 17/100, batch 415/625 completed (loss: 0.0398:  67%|██████▋   | 208/312 [21:00<10:36,  6.12s/it]
Training Epoch: 17/100, batch 415/625 completed (loss: 0.2281:  67%|██████▋   | 208/312 [21:00<10:36,  6.12s/it]
Training Epoch: 17/100, batch 415/625 completed (loss: 0.0182:  67%|██████▋   | 208/312 [21:00<10:36,  6.12s/it]
Training Epoch: 17/100, batch 415/625 completed (loss: 0.0372:  67%|██████▋   | 208/312 [21:00<10:35,  6.12s/it]
Training Epoch: 17/100, batch 416/625 completed (loss: 0.0392:  67%|██████▋   | 208/312 [21:03<10:36,  6.12s/it]
Training Epoch: 17/100, batch 416/625 completed (loss: 0.0102:  67%|██████▋   | 208/312 [21:03<10:36,  6.12s/it]
Training Epoch: 17/100, batch 416/625 completed (loss: 0.0715:  67%|██████▋   | 208/312 [21:03<10:36,  6.12s/it]
Training Epoch: 17/100, batch 416/625 completed (loss: 0.0104:  67%|██████▋   | 208/312 [21:03<10:35,  6.12s/it]
Training Epoch: 17/100, batch 416/625 completed (loss: 0.0392:  67%|██████▋   | 209/312 [21:06<10:25,  6.07s/it]
Training Epoch: 17/100, batch 416/625 completed (loss: 0.0715:  67%|██████▋   | 209/312 [21:06<10:25,  6.07s/it]
Training Epoch: 17/100, batch 416/625 completed (loss: 0.0102:  67%|██████▋   | 209/312 [21:06<10:25,  6.07s/it]
Training Epoch: 17/100, batch 416/625 completed (loss: 0.0104:  67%|██████▋   | 209/312 [21:06<10:25,  6.08s/it]
Training Epoch: 17/100, batch 417/625 completed (loss: 0.0292:  67%|██████▋   | 209/312 [21:06<10:25,  6.07s/it]
Training Epoch: 17/100, batch 417/625 completed (loss: 0.0563:  67%|██████▋   | 209/312 [21:06<10:25,  6.07s/it]
Training Epoch: 17/100, batch 417/625 completed (loss: 0.0163:  67%|██████▋   | 209/312 [21:06<10:25,  6.07s/it]
Training Epoch: 17/100, batch 417/625 completed (loss: 0.127:  67%|██████▋   | 209/312 [21:06<10:25,  6.08s/it] 
Training Epoch: 17/100, batch 418/625 completed (loss: 0.0113:  67%|██████▋   | 209/312 [21:09<10:25,  6.07s/it]
Training Epoch: 17/100, batch 418/625 completed (loss: 0.1048:  67%|██████▋   | 209/312 [21:09<10:25,  6.07s/it]
Training Epoch: 17/100, batch 418/625 completed (loss: 0.0094:  67%|██████▋   | 209/312 [21:09<10:25,  6.07s/it]
Training Epoch: 17/100, batch 418/625 completed (loss: 0.0069:  67%|██████▋   | 209/312 [21:09<10:25,  6.08s/it]
Training Epoch: 17/100, batch 418/625 completed (loss: 0.0069:  67%|██████▋   | 210/312 [21:12<10:18,  6.06s/it]
Training Epoch: 17/100, batch 418/625 completed (loss: 0.1048:  67%|██████▋   | 210/312 [21:12<10:18,  6.06s/it]
Training Epoch: 17/100, batch 418/625 completed (loss: 0.0113:  67%|██████▋   | 210/312 [21:12<10:18,  6.06s/it]
Training Epoch: 17/100, batch 418/625 completed (loss: 0.0094:  67%|██████▋   | 210/312 [21:12<10:19,  6.07s/it]
Training Epoch: 17/100, batch 419/625 completed (loss: 0.024:  67%|██████▋   | 210/312 [21:12<10:18,  6.06s/it] 
Training Epoch: 17/100, batch 419/625 completed (loss: 0.2875:  67%|██████▋   | 210/312 [21:12<10:18,  6.06s/it]
Training Epoch: 17/100, batch 419/625 completed (loss: 0.0101:  67%|██████▋   | 210/312 [21:12<10:19,  6.07s/it]
Training Epoch: 17/100, batch 419/625 completed (loss: 0.007:  67%|██████▋   | 210/312 [21:12<10:18,  6.06s/it] 
Training Epoch: 17/100, batch 420/625 completed (loss: 0.0305:  67%|██████▋   | 210/312 [21:15<10:18,  6.06s/it]
Training Epoch: 17/100, batch 420/625 completed (loss: 0.0271:  67%|██████▋   | 210/312 [21:15<10:19,  6.07s/it]
Training Epoch: 17/100, batch 420/625 completed (loss: 0.189:  67%|██████▋   | 210/312 [21:15<10:18,  6.06s/it]
Training Epoch: 17/100, batch 420/625 completed (loss: 0.0285:  67%|██████▋   | 210/312 [21:15<10:18,  6.06s/it]
Training Epoch: 17/100, batch 420/625 completed (loss: 0.0285:  68%|██████▊   | 211/312 [21:18<10:12,  6.07s/it]
Training Epoch: 17/100, batch 420/625 completed (loss: 0.0305:  68%|██████▊   | 211/312 [21:18<10:12,  6.07s/it]
Training Epoch: 17/100, batch 420/625 completed (loss: 0.189:  68%|██████▊   | 211/312 [21:18<10:12,  6.07s/it]
Training Epoch: 17/100, batch 420/625 completed (loss: 0.0271:  68%|██████▊   | 211/312 [21:18<10:12,  6.07s/it]
Training Epoch: 17/100, batch 421/625 completed (loss: 0.0285:  68%|██████▊   | 211/312 [21:18<10:12,  6.07s/it]
Training Epoch: 17/100, batch 421/625 completed (loss: 0.0182:  68%|██████▊   | 211/312 [21:18<10:12,  6.07s/it]
Training Epoch: 17/100, batch 421/625 completed (loss: 0.1233:  68%|██████▊   | 211/312 [21:18<10:12,  6.07s/it]
Training Epoch: 17/100, batch 421/625 completed (loss: 0.03:  68%|██████▊   | 211/312 [21:18<10:12,  6.07s/it]  
Training Epoch: 17/100, batch 422/625 completed (loss: 0.1831:  68%|██████▊   | 211/312 [21:21<10:12,  6.07s/it]
Training Epoch: 17/100, batch 422/625 completed (loss: 0.2292:  68%|██████▊   | 211/312 [21:21<10:12,  6.07s/it]
Training Epoch: 17/100, batch 422/625 completed (loss: 0.0505:  68%|██████▊   | 211/312 [21:21<10:12,  6.07s/it]
Training Epoch: 17/100, batch 422/625 completed (loss: 0.0502:  68%|██████▊   | 211/312 [21:21<10:12,  6.07s/it]
Training Epoch: 17/100, batch 422/625 completed (loss: 0.0505:  68%|██████▊   | 212/312 [21:24<10:05,  6.06s/it]
Training Epoch: 17/100, batch 422/625 completed (loss: 0.0502:  68%|██████▊   | 212/312 [21:24<10:06,  6.06s/it]
Training Epoch: 17/100, batch 422/625 completed (loss: 0.1831:  68%|██████▊   | 212/312 [21:24<10:05,  6.06s/it]
Training Epoch: 17/100, batch 422/625 completed (loss: 0.2292:  68%|██████▊   | 212/312 [21:24<10:06,  6.06s/it]
Training Epoch: 17/100, batch 423/625 completed (loss: 0.0615:  68%|██████▊   | 212/312 [21:24<10:06,  6.06s/it]
Training Epoch: 17/100, batch 423/625 completed (loss: 0.1701:  68%|██████▊   | 212/312 [21:24<10:05,  6.06s/it]
Training Epoch: 17/100, batch 423/625 completed (loss: 0.0122:  68%|██████▊   | 212/312 [21:24<10:05,  6.06s/it]
Training Epoch: 17/100, batch 423/625 completed (loss: 0.1906:  68%|██████▊   | 212/312 [21:24<10:06,  6.06s/it]
Training Epoch: 17/100, batch 424/625 completed (loss: 0.0809:  68%|██████▊   | 212/312 [21:27<10:05,  6.06s/it]
Training Epoch: 17/100, batch 424/625 completed (loss: 0.3014:  68%|██████▊   | 212/312 [21:27<10:06,  6.06s/it]
Training Epoch: 17/100, batch 424/625 completed (loss: 0.0755:  68%|██████▊   | 212/312 [21:27<10:05,  6.06s/it]
Training Epoch: 17/100, batch 424/625 completed (loss: 0.2936:  68%|██████▊   | 212/312 [21:27<10:06,  6.06s/it]
Training Epoch: 17/100, batch 424/625 completed (loss: 0.2936:  68%|██████▊   | 213/312 [21:30<10:05,  6.11s/it]
Training Epoch: 17/100, batch 424/625 completed (loss: 0.3014:  68%|██████▊   | 213/312 [21:30<10:05,  6.11s/it]
Training Epoch: 17/100, batch 424/625 completed (loss: 0.0809:  68%|██████▊   | 213/312 [21:30<10:05,  6.12s/it]
Training Epoch: 17/100, batch 424/625 completed (loss: 0.0755:  68%|██████▊   | 213/312 [21:30<10:05,  6.11s/it]
Training Epoch: 17/100, batch 425/625 completed (loss: 0.0158:  68%|██████▊   | 213/312 [21:30<10:05,  6.11s/it]
Training Epoch: 17/100, batch 425/625 completed (loss: 0.0867:  68%|██████▊   | 213/312 [21:30<10:05,  6.12s/it]
Training Epoch: 17/100, batch 425/625 completed (loss: 0.005:  68%|██████▊   | 213/312 [21:30<10:05,  6.11s/it] 
Training Epoch: 17/100, batch 425/625 completed (loss: 0.0311:  68%|██████▊   | 213/312 [21:30<10:05,  6.11s/it]
Training Epoch: 17/100, batch 426/625 completed (loss: 0.1514:  68%|██████▊   | 213/312 [21:33<10:05,  6.12s/it]
Training Epoch: 17/100, batch 426/625 completed (loss: 0.0126:  68%|██████▊   | 213/312 [21:33<10:05,  6.11s/it]
Training Epoch: 17/100, batch 426/625 completed (loss: 0.0344:  68%|██████▊   | 213/312 [21:33<10:05,  6.11s/it]
Training Epoch: 17/100, batch 426/625 completed (loss: 0.0471:  68%|██████▊   | 213/312 [21:33<10:05,  6.11s/it]
Training Epoch: 17/100, batch 426/625 completed (loss: 0.0344:  69%|██████▊   | 214/312 [21:36<09:52,  6.04s/it]
Training Epoch: 17/100, batch 426/625 completed (loss: 0.0471:  69%|██████▊   | 214/312 [21:36<09:52,  6.04s/it]
Training Epoch: 17/100, batch 426/625 completed (loss: 0.1514:  69%|██████▊   | 214/312 [21:36<09:52,  6.05s/it]
Training Epoch: 17/100, batch 426/625 completed (loss: 0.0126:  69%|██████▊   | 214/312 [21:36<09:52,  6.05s/it]
Training Epoch: 17/100, batch 427/625 completed (loss: 0.0757:  69%|██████▊   | 214/312 [21:36<09:52,  6.05s/it]
Training Epoch: 17/100, batch 427/625 completed (loss: 0.0853:  69%|██████▊   | 214/312 [21:36<09:52,  6.04s/it]
Training Epoch: 17/100, batch 427/625 completed (loss: 0.1023:  69%|██████▊   | 214/312 [21:36<09:52,  6.05s/it]
Training Epoch: 17/100, batch 427/625 completed (loss: 0.1223:  69%|██████▊   | 214/312 [21:36<09:52,  6.04s/it]
Training Epoch: 17/100, batch 428/625 completed (loss: 0.0236:  69%|██████▊   | 214/312 [21:39<09:52,  6.04s/it]
Training Epoch: 17/100, batch 428/625 completed (loss: 0.144:  69%|██████▊   | 214/312 [21:39<09:52,  6.05s/it] 
Training Epoch: 17/100, batch 428/625 completed (loss: 0.05:  69%|██████▊   | 214/312 [21:39<09:52,  6.05s/it]  
Training Epoch: 17/100, batch 428/625 completed (loss: 0.0044:  69%|██████▊   | 214/312 [21:39<09:52,  6.04s/it]
Training Epoch: 17/100, batch 428/625 completed (loss: 0.0044:  69%|██████▉   | 215/312 [21:42<09:49,  6.08s/it]
Training Epoch: 17/100, batch 428/625 completed (loss: 0.0236:  69%|██████▉   | 215/312 [21:42<09:49,  6.08s/it]
Training Epoch: 17/100, batch 428/625 completed (loss: 0.144:  69%|██████▉   | 215/312 [21:42<09:49,  6.08s/it]
Training Epoch: 17/100, batch 428/625 completed (loss: 0.05:  69%|██████▉   | 215/312 [21:42<09:49,  6.08s/it]
Training Epoch: 17/100, batch 429/625 completed (loss: 0.0101:  69%|██████▉   | 215/312 [21:42<09:49,  6.08s/it]
Training Epoch: 17/100, batch 429/625 completed (loss: 0.0731:  69%|██████▉   | 215/312 [21:42<09:49,  6.08s/it]
Training Epoch: 17/100, batch 429/625 completed (loss: 0.0114:  69%|██████▉   | 215/312 [21:42<09:49,  6.08s/it]
Training Epoch: 17/100, batch 429/625 completed (loss: 0.0282:  69%|██████▉   | 215/312 [21:43<09:49,  6.08s/it]
Training Epoch: 17/100, batch 430/625 completed (loss: 0.3404:  69%|██████▉   | 215/312 [21:46<09:49,  6.08s/it]
Training Epoch: 17/100, batch 430/625 completed (loss: 0.1043:  69%|██████▉   | 215/312 [21:46<09:49,  6.08s/it]
Training Epoch: 17/100, batch 430/625 completed (loss: 0.0846:  69%|██████▉   | 215/312 [21:46<09:49,  6.08s/it]
Training Epoch: 17/100, batch 430/625 completed (loss: 0.0138:  69%|██████▉   | 215/312 [21:46<09:49,  6.08s/it]
Training Epoch: 17/100, batch 430/625 completed (loss: 0.1043:  69%|██████▉   | 216/312 [21:49<09:45,  6.10s/it]
Training Epoch: 17/100, batch 430/625 completed (loss: 0.0138:  69%|██████▉   | 216/312 [21:49<09:45,  6.10s/it]
Training Epoch: 17/100, batch 430/625 completed (loss: 0.3404:  69%|██████▉   | 216/312 [21:49<09:45,  6.10s/it]
Training Epoch: 17/100, batch 430/625 completed (loss: 0.0846:  69%|██████▉   | 216/312 [21:49<09:45,  6.10s/it]
Training Epoch: 17/100, batch 431/625 completed (loss: 0.1779:  69%|██████▉   | 216/312 [21:49<09:45,  6.10s/it]
Training Epoch: 17/100, batch 431/625 completed (loss: 0.0213:  69%|██████▉   | 216/312 [21:49<09:45,  6.10s/it]
Training Epoch: 17/100, batch 431/625 completed (loss: 0.1315:  69%|██████▉   | 216/312 [21:49<09:45,  6.10s/it]
Training Epoch: 17/100, batch 431/625 completed (loss: 0.5091:  69%|██████▉   | 216/312 [21:49<09:45,  6.10s/it]
Training Epoch: 17/100, batch 432/625 completed (loss: 0.3435:  69%|██████▉   | 216/312 [21:52<09:45,  6.10s/it]
Training Epoch: 17/100, batch 432/625 completed (loss: 0.0051:  69%|██████▉   | 216/312 [21:52<09:45,  6.10s/it]
Training Epoch: 17/100, batch 432/625 completed (loss: 0.0242:  69%|██████▉   | 216/312 [21:52<09:45,  6.10s/it]
Training Epoch: 17/100, batch 432/625 completed (loss: 0.0087:  69%|██████▉   | 216/312 [21:52<09:45,  6.10s/it]
Training Epoch: 17/100, batch 432/625 completed (loss: 0.0087:  70%|██████▉   | 217/312 [21:55<09:36,  6.07s/it]
Training Epoch: 17/100, batch 432/625 completed (loss: 0.0051:  70%|██████▉   | 217/312 [21:55<09:36,  6.07s/it]
Training Epoch: 17/100, batch 432/625 completed (loss: 0.3435:  70%|██████▉   | 217/312 [21:55<09:36,  6.07s/it]
Training Epoch: 17/100, batch 432/625 completed (loss: 0.0242:  70%|██████▉   | 217/312 [21:55<09:36,  6.07s/it]
Training Epoch: 17/100, batch 433/625 completed (loss: 0.0212:  70%|██████▉   | 217/312 [21:55<09:36,  6.07s/it]
Training Epoch: 17/100, batch 433/625 completed (loss: 0.042:  70%|██████▉   | 217/312 [21:55<09:36,  6.07s/it] 
Training Epoch: 17/100, batch 433/625 completed (loss: 0.0326:  70%|██████▉   | 217/312 [21:55<09:36,  6.07s/it]
Training Epoch: 17/100, batch 433/625 completed (loss: 0.0157:  70%|██████▉   | 217/312 [21:55<09:36,  6.07s/it]
Training Epoch: 17/100, batch 434/625 completed (loss: 0.0216:  70%|██████▉   | 217/312 [21:58<09:36,  6.07s/it]
Training Epoch: 17/100, batch 434/625 completed (loss: 0.1118:  70%|██████▉   | 217/312 [21:58<09:36,  6.07s/it]
Training Epoch: 17/100, batch 434/625 completed (loss: 0.006:  70%|██████▉   | 217/312 [21:57<09:36,  6.07s/it] 
Training Epoch: 17/100, batch 434/625 completed (loss: 0.0108:  70%|██████▉   | 217/312 [21:58<09:36,  6.07s/it]
Training Epoch: 17/100, batch 434/625 completed (loss: 0.0108:  70%|██████▉   | 218/312 [22:01<09:36,  6.13s/it]
Training Epoch: 17/100, batch 434/625 completed (loss: 0.1118:  70%|██████▉   | 218/312 [22:01<09:36,  6.13s/it]
Training Epoch: 17/100, batch 434/625 completed (loss: 0.006:  70%|██████▉   | 218/312 [22:01<09:36,  6.13s/it]
Training Epoch: 17/100, batch 434/625 completed (loss: 0.0216:  70%|██████▉   | 218/312 [22:01<09:36,  6.13s/it]
Training Epoch: 17/100, batch 435/625 completed (loss: 0.0545:  70%|██████▉   | 218/312 [22:01<09:36,  6.13s/it]
Training Epoch: 17/100, batch 435/625 completed (loss: 0.0678:  70%|██████▉   | 218/312 [22:01<09:36,  6.13s/it]
Training Epoch: 17/100, batch 435/625 completed (loss: 0.0096:  70%|██████▉   | 218/312 [22:01<09:36,  6.13s/it]
Training Epoch: 17/100, batch 435/625 completed (loss: 0.1205:  70%|██████▉   | 218/312 [22:01<09:36,  6.13s/it]
Training Epoch: 17/100, batch 436/625 completed (loss: 0.0029:  70%|██████▉   | 218/312 [22:04<09:36,  6.13s/it]
Training Epoch: 17/100, batch 436/625 completed (loss: 0.0839:  70%|██████▉   | 218/312 [22:04<09:36,  6.13s/it]
Training Epoch: 17/100, batch 436/625 completed (loss: 0.2239:  70%|██████▉   | 218/312 [22:04<09:36,  6.13s/it]
Training Epoch: 17/100, batch 436/625 completed (loss: 0.0185:  70%|██████▉   | 218/312 [22:04<09:36,  6.13s/it]
Training Epoch: 17/100, batch 436/625 completed (loss: 0.2239:  70%|███████   | 219/312 [22:07<09:22,  6.04s/it]
Training Epoch: 17/100, batch 436/625 completed (loss: 0.0839:  70%|███████   | 219/312 [22:07<09:22,  6.04s/it]
Training Epoch: 17/100, batch 436/625 completed (loss: 0.0029:  70%|███████   | 219/312 [22:07<09:22,  6.04s/it]
Training Epoch: 17/100, batch 436/625 completed (loss: 0.0185:  70%|███████   | 219/312 [22:07<09:22,  6.05s/it]
Training Epoch: 17/100, batch 437/625 completed (loss: 0.0579:  70%|███████   | 219/312 [22:07<09:22,  6.04s/it]
Training Epoch: 17/100, batch 437/625 completed (loss: 0.5412:  70%|███████   | 219/312 [22:07<09:22,  6.04s/it]
Training Epoch: 17/100, batch 437/625 completed (loss: 0.0081:  70%|███████   | 219/312 [22:07<09:22,  6.04s/it]
Training Epoch: 17/100, batch 437/625 completed (loss: 0.0619:  70%|███████   | 219/312 [22:07<09:22,  6.05s/it]
Training Epoch: 17/100, batch 438/625 completed (loss: 0.3048:  70%|███████   | 219/312 [22:10<09:22,  6.04s/it]
Training Epoch: 17/100, batch 438/625 completed (loss: 0.0424:  70%|███████   | 219/312 [22:10<09:22,  6.04s/it]
Training Epoch: 17/100, batch 438/625 completed (loss: 0.0312:  70%|███████   | 219/312 [22:10<09:22,  6.04s/it]
Training Epoch: 17/100, batch 438/625 completed (loss: 0.0117:  70%|███████   | 219/312 [22:10<09:22,  6.05s/it]
Training Epoch: 17/100, batch 438/625 completed (loss: 0.0117:  71%|███████   | 220/312 [22:13<09:19,  6.08s/it]
Training Epoch: 17/100, batch 438/625 completed (loss: 0.0312:  71%|███████   | 220/312 [22:13<09:19,  6.08s/it]
Training Epoch: 17/100, batch 438/625 completed (loss: 0.3048:  71%|███████   | 220/312 [22:13<09:19,  6.08s/it]
Training Epoch: 17/100, batch 438/625 completed (loss: 0.0424:  71%|███████   | 220/312 [22:13<09:19,  6.09s/it]
Training Epoch: 17/100, batch 439/625 completed (loss: 0.0094:  71%|███████   | 220/312 [22:13<09:19,  6.09s/it]
Training Epoch: 17/100, batch 439/625 completed (loss: 0.0606:  71%|███████   | 220/312 [22:13<09:19,  6.08s/it]
Training Epoch: 17/100, batch 439/625 completed (loss: 0.0239:  71%|███████   | 220/312 [22:13<09:19,  6.08s/it]
Training Epoch: 17/100, batch 439/625 completed (loss: 0.2479:  71%|███████   | 220/312 [22:13<09:19,  6.08s/it]
Training Epoch: 17/100, batch 440/625 completed (loss: 0.0008:  71%|███████   | 220/312 [22:16<09:19,  6.08s/it]
Training Epoch: 17/100, batch 440/625 completed (loss: 0.0242:  71%|███████   | 220/312 [22:16<09:19,  6.08s/it]
Training Epoch: 17/100, batch 440/625 completed (loss: 0.1675:  71%|███████   | 220/312 [22:16<09:19,  6.09s/it]
Training Epoch: 17/100, batch 440/625 completed (loss: 0.0915:  71%|███████   | 220/312 [22:16<09:19,  6.08s/it]
Training Epoch: 17/100, batch 440/625 completed (loss: 0.1675:  71%|███████   | 221/312 [22:19<09:14,  6.10s/it]
Training Epoch: 17/100, batch 440/625 completed (loss: 0.0915:  71%|███████   | 221/312 [22:19<09:15,  6.10s/it]
Training Epoch: 17/100, batch 440/625 completed (loss: 0.0242:  71%|███████   | 221/312 [22:19<09:15,  6.10s/it]
Training Epoch: 17/100, batch 440/625 completed (loss: 0.0008:  71%|███████   | 221/312 [22:19<09:15,  6.10s/it]
Training Epoch: 17/100, batch 441/625 completed (loss: 0.0801:  71%|███████   | 221/312 [22:19<09:15,  6.10s/it]
Training Epoch: 17/100, batch 441/625 completed (loss: 0.0277:  71%|███████   | 221/312 [22:19<09:14,  6.10s/it]
Training Epoch: 17/100, batch 441/625 completed (loss: 0.0167:  71%|███████   | 221/312 [22:19<09:15,  6.10s/it]
Training Epoch: 17/100, batch 441/625 completed (loss: 0.0039:  71%|███████   | 221/312 [22:19<09:15,  6.10s/it]
Training Epoch: 17/100, batch 442/625 completed (loss: 0.0336:  71%|███████   | 221/312 [22:22<09:14,  6.10s/it]
Training Epoch: 17/100, batch 442/625 completed (loss: 0.0408:  71%|███████   | 221/312 [22:22<09:15,  6.10s/it]
Training Epoch: 17/100, batch 442/625 completed (loss: 0.0107:  71%|███████   | 221/312 [22:22<09:15,  6.10s/it]
Training Epoch: 17/100, batch 442/625 completed (loss: 0.014:  71%|███████   | 221/312 [22:22<09:15,  6.10s/it] 
Training Epoch: 17/100, batch 442/625 completed (loss: 0.0336:  71%|███████   | 222/312 [22:25<09:06,  6.07s/it]
Training Epoch: 17/100, batch 442/625 completed (loss: 0.014:  71%|███████   | 222/312 [22:25<09:06,  6.07s/it]
Training Epoch: 17/100, batch 442/625 completed (loss: 0.0408:  71%|███████   | 222/312 [22:25<09:06,  6.07s/it]
Training Epoch: 17/100, batch 442/625 completed (loss: 0.0107:  71%|███████   | 222/312 [22:25<09:06,  6.08s/it]
Training Epoch: 17/100, batch 443/625 completed (loss: 0.064:  71%|███████   | 222/312 [22:25<09:06,  6.08s/it] 
Training Epoch: 17/100, batch 443/625 completed (loss: 0.0066:  71%|███████   | 222/312 [22:25<09:06,  6.07s/it]
Training Epoch: 17/100, batch 443/625 completed (loss: 0.0131:  71%|███████   | 222/312 [22:25<09:06,  6.07s/it]
Training Epoch: 17/100, batch 443/625 completed (loss: 0.0741:  71%|███████   | 222/312 [22:25<09:06,  6.07s/it]
Training Epoch: 17/100, batch 444/625 completed (loss: 0.1514:  71%|███████   | 222/312 [22:28<09:06,  6.07s/it]
Training Epoch: 17/100, batch 444/625 completed (loss: 0.1352:  71%|███████   | 222/312 [22:28<09:06,  6.08s/it]
Training Epoch: 17/100, batch 444/625 completed (loss: 0.0431:  71%|███████   | 222/312 [22:28<09:06,  6.07s/it]
Training Epoch: 17/100, batch 444/625 completed (loss: 0.0526:  71%|███████   | 222/312 [22:28<09:06,  6.07s/it]
Training Epoch: 17/100, batch 444/625 completed (loss: 0.1514:  71%|███████▏  | 223/312 [22:31<08:59,  6.06s/it]
Training Epoch: 17/100, batch 444/625 completed (loss: 0.0526:  71%|███████▏  | 223/312 [22:31<08:59,  6.06s/it]
Training Epoch: 17/100, batch 444/625 completed (loss: 0.0431:  71%|███████▏  | 223/312 [22:31<08:59,  6.06s/it]
Training Epoch: 17/100, batch 444/625 completed (loss: 0.1352:  71%|███████▏  | 223/312 [22:31<08:59,  6.06s/it]
Training Epoch: 17/100, batch 445/625 completed (loss: 0.0051:  71%|███████▏  | 223/312 [22:31<08:59,  6.06s/it]
Training Epoch: 17/100, batch 445/625 completed (loss: 0.0098:  71%|███████▏  | 223/312 [22:31<08:59,  6.06s/it]
Training Epoch: 17/100, batch 445/625 completed (loss: 0.0167:  71%|███████▏  | 223/312 [22:31<08:59,  6.06s/it]
Training Epoch: 17/100, batch 445/625 completed (loss: 0.0184:  71%|███████▏  | 223/312 [22:31<08:59,  6.06s/it]
Training Epoch: 17/100, batch 446/625 completed (loss: 0.0194:  71%|███████▏  | 223/312 [22:34<08:59,  6.06s/it]
Training Epoch: 17/100, batch 446/625 completed (loss: 0.1346:  71%|███████▏  | 223/312 [22:34<08:59,  6.06s/it]
Training Epoch: 17/100, batch 446/625 completed (loss: 0.0228:  71%|███████▏  | 223/312 [22:34<08:59,  6.06s/it]
Training Epoch: 17/100, batch 446/625 completed (loss: 0.0179:  71%|███████▏  | 223/312 [22:34<08:59,  6.06s/it]
Training Epoch: 17/100, batch 446/625 completed (loss: 0.1346:  72%|███████▏  | 224/312 [22:37<08:55,  6.09s/it]
Training Epoch: 17/100, batch 446/625 completed (loss: 0.0194:  72%|███████▏  | 224/312 [22:37<08:56,  6.09s/it]
Training Epoch: 17/100, batch 446/625 completed (loss: 0.0228:  72%|███████▏  | 224/312 [22:37<08:56,  6.10s/it]
Training Epoch: 17/100, batch 446/625 completed (loss: 0.0179:  72%|███████▏  | 224/312 [22:37<08:56,  6.10s/it]
Training Epoch: 17/100, batch 447/625 completed (loss: 0.0445:  72%|███████▏  | 224/312 [22:37<08:56,  6.10s/it]
Training Epoch: 17/100, batch 447/625 completed (loss: 0.0264:  72%|███████▏  | 224/312 [22:37<08:56,  6.09s/it]
Training Epoch: 17/100, batch 447/625 completed (loss: 0.02:  72%|███████▏  | 224/312 [22:37<08:55,  6.09s/it]  
Training Epoch: 17/100, batch 447/625 completed (loss: 0.0221:  72%|███████▏  | 224/312 [22:37<08:56,  6.10s/it]
Training Epoch: 17/100, batch 448/625 completed (loss: 0.0097:  72%|███████▏  | 224/312 [22:40<08:55,  6.09s/it]
Training Epoch: 17/100, batch 448/625 completed (loss: 0.0699:  72%|███████▏  | 224/312 [22:40<08:56,  6.09s/it]
Training Epoch: 17/100, batch 448/625 completed (loss: 0.006:  72%|███████▏  | 224/312 [22:40<08:56,  6.10s/it] 
Training Epoch: 17/100, batch 448/625 completed (loss: 0.0211:  72%|███████▏  | 224/312 [22:40<08:56,  6.10s/it]
Training Epoch: 17/100, batch 448/625 completed (loss: 0.0211:  72%|███████▏  | 225/312 [22:43<08:48,  6.07s/it]
Training Epoch: 17/100, batch 448/625 completed (loss: 0.0699:  72%|███████▏  | 225/312 [22:43<08:48,  6.07s/it]
Training Epoch: 17/100, batch 448/625 completed (loss: 0.0097:  72%|███████▏  | 225/312 [22:43<08:48,  6.08s/it]
Training Epoch: 17/100, batch 448/625 completed (loss: 0.006:  72%|███████▏  | 225/312 [22:43<08:48,  6.08s/it]
Training Epoch: 17/100, batch 449/625 completed (loss: 0.0137:  72%|███████▏  | 225/312 [22:43<08:48,  6.07s/it]
Training Epoch: 17/100, batch 449/625 completed (loss: 0.0258:  72%|███████▏  | 225/312 [22:43<08:48,  6.08s/it]
Training Epoch: 17/100, batch 449/625 completed (loss: 0.0693:  72%|███████▏  | 225/312 [22:43<08:48,  6.08s/it]
Training Epoch: 17/100, batch 449/625 completed (loss: 0.0485:  72%|███████▏  | 225/312 [22:43<08:48,  6.07s/it]
Training Epoch: 17/100, batch 450/625 completed (loss: 0.0294:  72%|███████▏  | 225/312 [22:46<08:48,  6.07s/it]
Training Epoch: 17/100, batch 450/625 completed (loss: 0.0133:  72%|███████▏  | 225/312 [22:46<08:48,  6.08s/it]
Training Epoch: 17/100, batch 450/625 completed (loss: 0.082:  72%|███████▏  | 225/312 [22:46<08:48,  6.08s/it] 
Training Epoch: 17/100, batch 450/625 completed (loss: 0.0059:  72%|███████▏  | 225/312 [22:46<08:48,  6.07s/it]
Training Epoch: 17/100, batch 450/625 completed (loss: 0.0059:  72%|███████▏  | 226/312 [22:50<08:48,  6.14s/it]
Training Epoch: 17/100, batch 450/625 completed (loss: 0.0294:  72%|███████▏  | 226/312 [22:50<08:48,  6.14s/it]
Training Epoch: 17/100, batch 450/625 completed (loss: 0.082:  72%|███████▏  | 226/312 [22:50<08:48,  6.14s/it]
Training Epoch: 17/100, batch 450/625 completed (loss: 0.0133:  72%|███████▏  | 226/312 [22:50<08:47,  6.14s/it]
Training Epoch: 17/100, batch 451/625 completed (loss: 0.0112:  72%|███████▏  | 226/312 [22:50<08:48,  6.14s/it]
Training Epoch: 17/100, batch 451/625 completed (loss: 0.006:  72%|███████▏  | 226/312 [22:50<08:47,  6.14s/it] 
Training Epoch: 17/100, batch 451/625 completed (loss: 0.0553:  72%|███████▏  | 226/312 [22:50<08:48,  6.14s/it]
Training Epoch: 17/100, batch 451/625 completed (loss: 0.1124:  72%|███████▏  | 226/312 [22:50<08:48,  6.14s/it]
Training Epoch: 17/100, batch 452/625 completed (loss: 0.0283:  72%|███████▏  | 226/312 [22:52<08:48,  6.14s/it]
Training Epoch: 17/100, batch 452/625 completed (loss: 0.0659:  72%|███████▏  | 226/312 [22:52<08:48,  6.14s/it]
Training Epoch: 17/100, batch 452/625 completed (loss: 0.0354:  72%|███████▏  | 226/312 [22:52<08:47,  6.14s/it]
Training Epoch: 17/100, batch 452/625 completed (loss: 0.0039:  72%|███████▏  | 226/312 [22:53<08:48,  6.14s/it]
Training Epoch: 17/100, batch 452/625 completed (loss: 0.0659:  73%|███████▎  | 227/312 [22:55<08:33,  6.05s/it]
Training Epoch: 17/100, batch 452/625 completed (loss: 0.0354:  73%|███████▎  | 227/312 [22:55<08:33,  6.04s/it]
Training Epoch: 17/100, batch 452/625 completed (loss: 0.0039:  73%|███████▎  | 227/312 [22:55<08:33,  6.05s/it]
Training Epoch: 17/100, batch 452/625 completed (loss: 0.0283:  73%|███████▎  | 227/312 [22:55<08:34,  6.05s/it]
Training Epoch: 17/100, batch 453/625 completed (loss: 0.0031:  73%|███████▎  | 227/312 [22:55<08:34,  6.05s/it]
Training Epoch: 17/100, batch 453/625 completed (loss: 0.0301:  73%|███████▎  | 227/312 [22:55<08:33,  6.05s/it]
Training Epoch: 17/100, batch 453/625 completed (loss: 0.3063:  73%|███████▎  | 227/312 [22:55<08:33,  6.04s/it]
Training Epoch: 17/100, batch 453/625 completed (loss: 0.1284:  73%|███████▎  | 227/312 [22:55<08:33,  6.05s/it]
Training Epoch: 17/100, batch 454/625 completed (loss: 0.048:  73%|███████▎  | 227/312 [22:58<08:34,  6.05s/it] 
Training Epoch: 17/100, batch 454/625 completed (loss: 0.4703:  73%|███████▎  | 227/312 [22:58<08:33,  6.04s/it]
Training Epoch: 17/100, batch 454/625 completed (loss: 0.0182:  73%|███████▎  | 227/312 [22:58<08:33,  6.05s/it]
Training Epoch: 17/100, batch 454/625 completed (loss: 0.1572:  73%|███████▎  | 227/312 [22:59<08:33,  6.05s/it]
Training Epoch: 17/100, batch 454/625 completed (loss: 0.4703:  73%|███████▎  | 228/312 [23:02<08:30,  6.08s/it]
Training Epoch: 17/100, batch 454/625 completed (loss: 0.0182:  73%|███████▎  | 228/312 [23:01<08:30,  6.08s/it]
Training Epoch: 17/100, batch 454/625 completed (loss: 0.1572:  73%|███████▎  | 228/312 [23:02<08:30,  6.08s/it]
Training Epoch: 17/100, batch 454/625 completed (loss: 0.048:  73%|███████▎  | 228/312 [23:02<08:30,  6.08s/it]
Training Epoch: 17/100, batch 455/625 completed (loss: 0.0309:  73%|███████▎  | 228/312 [23:02<08:30,  6.08s/it]
Training Epoch: 17/100, batch 455/625 completed (loss: 0.0066:  73%|███████▎  | 228/312 [23:02<08:30,  6.08s/it]
Training Epoch: 17/100, batch 455/625 completed (loss: 0.0892:  73%|███████▎  | 228/312 [23:02<08:30,  6.08s/it]
Training Epoch: 17/100, batch 455/625 completed (loss: 0.0641:  73%|███████▎  | 228/312 [23:02<08:30,  6.08s/it]
Training Epoch: 17/100, batch 456/625 completed (loss: 0.0243:  73%|███████▎  | 228/312 [23:05<08:30,  6.08s/it]
Training Epoch: 17/100, batch 456/625 completed (loss: 0.004:  73%|███████▎  | 228/312 [23:05<08:30,  6.08s/it] 
Training Epoch: 17/100, batch 456/625 completed (loss: 0.0074:  73%|███████▎  | 228/312 [23:05<08:30,  6.08s/it]
Training Epoch: 17/100, batch 456/625 completed (loss: 0.0196:  73%|███████▎  | 228/312 [23:05<08:30,  6.08s/it]
Training Epoch: 17/100, batch 456/625 completed (loss: 0.004:  73%|███████▎  | 229/312 [23:08<08:27,  6.11s/it]
Training Epoch: 17/100, batch 456/625 completed (loss: 0.0074:  73%|███████▎  | 229/312 [23:08<08:27,  6.11s/it]
Training Epoch: 17/100, batch 456/625 completed (loss: 0.0243:  73%|███████▎  | 229/312 [23:08<08:27,  6.11s/it]
Training Epoch: 17/100, batch 456/625 completed (loss: 0.0196:  73%|███████▎  | 229/312 [23:08<08:27,  6.12s/it]
Training Epoch: 17/100, batch 457/625 completed (loss: 0.1494:  73%|███████▎  | 229/312 [23:08<08:27,  6.11s/it]
Training Epoch: 17/100, batch 457/625 completed (loss: 0.0341:  73%|███████▎  | 229/312 [23:08<08:27,  6.11s/it]
Training Epoch: 17/100, batch 457/625 completed (loss: 0.0103:  73%|███████▎  | 229/312 [23:08<08:27,  6.11s/it]
Training Epoch: 17/100, batch 457/625 completed (loss: 0.0427:  73%|███████▎  | 229/312 [23:08<08:27,  6.12s/it]
Training Epoch: 17/100, batch 458/625 completed (loss: 0.4426:  73%|███████▎  | 229/312 [23:11<08:27,  6.11s/it]
Training Epoch: 17/100, batch 458/625 completed (loss: 0.0373:  73%|███████▎  | 229/312 [23:11<08:27,  6.11s/it]
Training Epoch: 17/100, batch 458/625 completed (loss: 0.0198:  73%|███████▎  | 229/312 [23:11<08:27,  6.11s/it]
Training Epoch: 17/100, batch 458/625 completed (loss: 0.1602:  73%|███████▎  | 229/312 [23:11<08:27,  6.12s/it]
Training Epoch: 17/100, batch 458/625 completed (loss: 0.0198:  74%|███████▎  | 230/312 [23:14<08:18,  6.08s/it]
Training Epoch: 17/100, batch 458/625 completed (loss: 0.4426:  74%|███████▎  | 230/312 [23:14<08:18,  6.08s/it]
Training Epoch: 17/100, batch 458/625 completed (loss: 0.1602:  74%|███████▎  | 230/312 [23:14<08:18,  6.08s/it]
Training Epoch: 17/100, batch 458/625 completed (loss: 0.0373:  74%|███████▎  | 230/312 [23:14<08:18,  6.08s/it]
Training Epoch: 17/100, batch 459/625 completed (loss: 0.1079:  74%|███████▎  | 230/312 [23:14<08:18,  6.08s/it]
Training Epoch: 17/100, batch 459/625 completed (loss: 0.0706:  74%|███████▎  | 230/312 [23:14<08:18,  6.08s/it]
Training Epoch: 17/100, batch 459/625 completed (loss: 0.0151:  74%|███████▎  | 230/312 [23:14<08:18,  6.08s/it]
Training Epoch: 17/100, batch 459/625 completed (loss: 0.0032:  74%|███████▎  | 230/312 [23:14<08:18,  6.08s/it]
Training Epoch: 17/100, batch 460/625 completed (loss: 0.031:  74%|███████▎  | 230/312 [23:17<08:18,  6.08s/it] 
Training Epoch: 17/100, batch 460/625 completed (loss: 0.0928:  74%|███████▎  | 230/312 [23:17<08:18,  6.08s/it]
Training Epoch: 17/100, batch 460/625 completed (loss: 0.0463:  74%|███████▎  | 230/312 [23:17<08:18,  6.08s/it]
Training Epoch: 17/100, batch 460/625 completed (loss: 0.021:  74%|███████▎  | 230/312 [23:17<08:18,  6.08s/it] 
Training Epoch: 17/100, batch 460/625 completed (loss: 0.021:  74%|███████▍  | 231/312 [23:20<08:14,  6.10s/it]
Training Epoch: 17/100, batch 460/625 completed (loss: 0.031:  74%|███████▍  | 231/312 [23:20<08:14,  6.10s/it]
Training Epoch: 17/100, batch 460/625 completed (loss: 0.0928:  74%|███████▍  | 231/312 [23:20<08:14,  6.10s/it]
Training Epoch: 17/100, batch 460/625 completed (loss: 0.0463:  74%|███████▍  | 231/312 [23:20<08:14,  6.10s/it]
Training Epoch: 17/100, batch 461/625 completed (loss: 0.2109:  74%|███████▍  | 231/312 [23:20<08:14,  6.10s/it]
Training Epoch: 17/100, batch 461/625 completed (loss: 0.1065:  74%|███████▍  | 231/312 [23:20<08:14,  6.10s/it]
Training Epoch: 17/100, batch 461/625 completed (loss: 0.0383:  74%|███████▍  | 231/312 [23:20<08:14,  6.10s/it]
Training Epoch: 17/100, batch 461/625 completed (loss: 0.013:  74%|███████▍  | 231/312 [23:20<08:14,  6.10s/it]
Training Epoch: 17/100, batch 462/625 completed (loss: 0.3629:  74%|███████▍  | 231/312 [23:23<08:14,  6.10s/it]
Training Epoch: 17/100, batch 462/625 completed (loss: 0.0121:  74%|███████▍  | 231/312 [23:23<08:14,  6.10s/it]
Training Epoch: 17/100, batch 462/625 completed (loss: 0.0421:  74%|███████▍  | 231/312 [23:23<08:14,  6.10s/it]
Training Epoch: 17/100, batch 462/625 completed (loss: 0.0154:  74%|███████▍  | 231/312 [23:23<08:14,  6.10s/it]
Training Epoch: 17/100, batch 462/625 completed (loss: 0.0154:  74%|███████▍  | 232/312 [23:26<08:02,  6.03s/it]
Training Epoch: 17/100, batch 462/625 completed (loss: 0.0121:  74%|███████▍  | 232/312 [23:26<08:02,  6.03s/it]
Training Epoch: 17/100, batch 462/625 completed (loss: 0.3629:  74%|███████▍  | 232/312 [23:26<08:02,  6.03s/it]
Training Epoch: 17/100, batch 462/625 completed (loss: 0.0421:  74%|███████▍  | 232/312 [23:26<08:02,  6.03s/it]
Training Epoch: 17/100, batch 463/625 completed (loss: 0.0136:  74%|███████▍  | 232/312 [23:26<08:02,  6.03s/it]
Training Epoch: 17/100, batch 463/625 completed (loss: 0.0801:  74%|███████▍  | 232/312 [23:26<08:02,  6.03s/it]
Training Epoch: 17/100, batch 463/625 completed (loss: 0.0133:  74%|███████▍  | 232/312 [23:26<08:02,  6.03s/it]
Training Epoch: 17/100, batch 463/625 completed (loss: 0.0059:  74%|███████▍  | 232/312 [23:26<08:02,  6.03s/it]
Training Epoch: 17/100, batch 464/625 completed (loss: 0.061:  74%|███████▍  | 232/312 [23:29<08:02,  6.03s/it] 
Training Epoch: 17/100, batch 464/625 completed (loss: 0.0043:  74%|███████▍  | 232/312 [23:29<08:02,  6.03s/it]
Training Epoch: 17/100, batch 464/625 completed (loss: 0.0039:  74%|███████▍  | 232/312 [23:29<08:02,  6.03s/it]
Training Epoch: 17/100, batch 464/625 completed (loss: 0.0212:  74%|███████▍  | 232/312 [23:29<08:02,  6.03s/it]
Training Epoch: 17/100, batch 464/625 completed (loss: 0.0212:  75%|███████▍  | 233/312 [23:32<07:59,  6.07s/it]
Training Epoch: 17/100, batch 464/625 completed (loss: 0.0039:  75%|███████▍  | 233/312 [23:32<07:59,  6.07s/it]
Training Epoch: 17/100, batch 464/625 completed (loss: 0.061:  75%|███████▍  | 233/312 [23:32<07:59,  6.07s/it]
Training Epoch: 17/100, batch 464/625 completed (loss: 0.0043:  75%|███████▍  | 233/312 [23:32<08:00,  6.08s/it]
Training Epoch: 17/100, batch 465/625 completed (loss: 0.0092:  75%|███████▍  | 233/312 [23:32<07:59,  6.07s/it]
Training Epoch: 17/100, batch 465/625 completed (loss: 0.4449:  75%|███████▍  | 233/312 [23:32<08:00,  6.08s/it]
Training Epoch: 17/100, batch 465/625 completed (loss: 0.0491:  75%|███████▍  | 233/312 [23:32<07:59,  6.07s/it]
Training Epoch: 17/100, batch 465/625 completed (loss: 0.0392:  75%|███████▍  | 233/312 [23:32<07:59,  6.07s/it]
Training Epoch: 17/100, batch 466/625 completed (loss: 0.0089:  75%|███████▍  | 233/312 [23:35<08:00,  6.08s/it]
Training Epoch: 17/100, batch 466/625 completed (loss: 0.0061:  75%|███████▍  | 233/312 [23:35<07:59,  6.07s/it]
Training Epoch: 17/100, batch 466/625 completed (loss: 0.0466:  75%|███████▍  | 233/312 [23:35<07:59,  6.07s/it]
Training Epoch: 17/100, batch 466/625 completed (loss: 0.5726:  75%|███████▍  | 233/312 [23:35<07:59,  6.07s/it]
Training Epoch: 17/100, batch 466/625 completed (loss: 0.5726:  75%|███████▌  | 234/312 [23:38<07:55,  6.10s/it]
Training Epoch: 17/100, batch 466/625 completed (loss: 0.0089:  75%|███████▌  | 234/312 [23:38<07:55,  6.10s/it]
Training Epoch: 17/100, batch 466/625 completed (loss: 0.0466:  75%|███████▌  | 234/312 [23:38<07:55,  6.10s/it]
Training Epoch: 17/100, batch 466/625 completed (loss: 0.0061:  75%|███████▌  | 234/312 [23:38<07:56,  6.10s/it]
Training Epoch: 17/100, batch 467/625 completed (loss: 0.0285:  75%|███████▌  | 234/312 [23:38<07:56,  6.10s/it]
Training Epoch: 17/100, batch 467/625 completed (loss: 0.0122:  75%|███████▌  | 234/312 [23:38<07:55,  6.10s/it]
Training Epoch: 17/100, batch 467/625 completed (loss: 0.0102:  75%|███████▌  | 234/312 [23:38<07:55,  6.10s/it]
Training Epoch: 17/100, batch 467/625 completed (loss: 0.0176:  75%|███████▌  | 234/312 [23:38<07:55,  6.10s/it]
Training Epoch: 17/100, batch 468/625 completed (loss: 0.0823:  75%|███████▌  | 234/312 [23:41<07:55,  6.10s/it]
Training Epoch: 17/100, batch 468/625 completed (loss: 0.2315:  75%|███████▌  | 234/312 [23:41<07:56,  6.10s/it]
Training Epoch: 17/100, batch 468/625 completed (loss: 0.0899:  75%|███████▌  | 234/312 [23:41<07:55,  6.10s/it]
Training Epoch: 17/100, batch 468/625 completed (loss: 0.0625:  75%|███████▌  | 234/312 [23:41<07:55,  6.10s/it]
Training Epoch: 17/100, batch 468/625 completed (loss: 0.0899:  75%|███████▌  | 235/312 [23:44<07:43,  6.02s/it]
Training Epoch: 17/100, batch 468/625 completed (loss: 0.2315:  75%|███████▌  | 235/312 [23:44<07:43,  6.02s/it]
Training Epoch: 17/100, batch 468/625 completed (loss: 0.0625:  75%|███████▌  | 235/312 [23:44<07:43,  6.02s/it]
Training Epoch: 17/100, batch 468/625 completed (loss: 0.0823:  75%|███████▌  | 235/312 [23:44<07:43,  6.02s/it]
Training Epoch: 17/100, batch 469/625 completed (loss: 0.0102:  75%|███████▌  | 235/312 [23:44<07:43,  6.02s/it]
Training Epoch: 17/100, batch 469/625 completed (loss: 0.0269:  75%|███████▌  | 235/312 [23:44<07:43,  6.02s/it]
Training Epoch: 17/100, batch 469/625 completed (loss: 0.0229:  75%|███████▌  | 235/312 [23:44<07:43,  6.02s/it]
Training Epoch: 17/100, batch 469/625 completed (loss: 0.041:  75%|███████▌  | 235/312 [23:44<07:43,  6.02s/it] 
Training Epoch: 17/100, batch 470/625 completed (loss: 0.0542:  75%|███████▌  | 235/312 [23:47<07:43,  6.02s/it]
Training Epoch: 17/100, batch 470/625 completed (loss: 0.0181:  75%|███████▌  | 235/312 [23:47<07:43,  6.02s/it]
Training Epoch: 17/100, batch 470/625 completed (loss: 0.026:  75%|███████▌  | 235/312 [23:47<07:43,  6.02s/it] 
Training Epoch: 17/100, batch 470/625 completed (loss: 0.0034:  75%|███████▌  | 235/312 [23:47<07:43,  6.02s/it]
Training Epoch: 17/100, batch 470/625 completed (loss: 0.0542:  76%|███████▌  | 236/312 [23:50<07:37,  6.02s/it]
Training Epoch: 17/100, batch 470/625 completed (loss: 0.0181:  76%|███████▌  | 236/312 [23:50<07:37,  6.02s/it]
Training Epoch: 17/100, batch 470/625 completed (loss: 0.026:  76%|███████▌  | 236/312 [23:50<07:37,  6.02s/it]
Training Epoch: 17/100, batch 470/625 completed (loss: 0.0034:  76%|███████▌  | 236/312 [23:50<07:37,  6.02s/it]
Training Epoch: 17/100, batch 471/625 completed (loss: 0.2098:  76%|███████▌  | 236/312 [23:50<07:37,  6.02s/it]
Training Epoch: 17/100, batch 471/625 completed (loss: 0.0702:  76%|███████▌  | 236/312 [23:50<07:37,  6.02s/it]
Training Epoch: 17/100, batch 471/625 completed (loss: 0.1787:  76%|███████▌  | 236/312 [23:50<07:37,  6.02s/it]
Training Epoch: 17/100, batch 471/625 completed (loss: 0.0019:  76%|███████▌  | 236/312 [23:50<07:37,  6.02s/it]
Training Epoch: 17/100, batch 472/625 completed (loss: 0.0133:  76%|███████▌  | 236/312 [23:53<07:37,  6.02s/it]
Training Epoch: 17/100, batch 472/625 completed (loss: 0.1427:  76%|███████▌  | 236/312 [23:53<07:37,  6.02s/it]
Training Epoch: 17/100, batch 472/625 completed (loss: 0.0724:  76%|███████▌  | 236/312 [23:53<07:37,  6.02s/it]
Training Epoch: 17/100, batch 472/625 completed (loss: 0.1921:  76%|███████▌  | 236/312 [23:53<07:37,  6.02s/it]
Training Epoch: 17/100, batch 472/625 completed (loss: 0.1921:  76%|███████▌  | 237/312 [23:56<07:33,  6.05s/it]
Training Epoch: 17/100, batch 472/625 completed (loss: 0.1427:  76%|███████▌  | 237/312 [23:56<07:33,  6.05s/it]
Training Epoch: 17/100, batch 472/625 completed (loss: 0.0133:  76%|███████▌  | 237/312 [23:56<07:34,  6.06s/it]
Training Epoch: 17/100, batch 472/625 completed (loss: 0.0724:  76%|███████▌  | 237/312 [23:56<07:34,  6.06s/it]
Training Epoch: 17/100, batch 473/625 completed (loss: 0.0613:  76%|███████▌  | 237/312 [23:56<07:34,  6.06s/it]
Training Epoch: 17/100, batch 473/625 completed (loss: 0.0544:  76%|███████▌  | 237/312 [23:56<07:33,  6.05s/it]
Training Epoch: 17/100, batch 473/625 completed (loss: 0.2464:  76%|███████▌  | 237/312 [23:56<07:34,  6.06s/it]
Training Epoch: 17/100, batch 473/625 completed (loss: 0.1736:  76%|███████▌  | 237/312 [23:56<07:33,  6.05s/it]
Training Epoch: 17/100, batch 474/625 completed (loss: 0.2072:  76%|███████▌  | 237/312 [23:59<07:33,  6.05s/it]
Training Epoch: 17/100, batch 474/625 completed (loss: 0.0347:  76%|███████▌  | 237/312 [23:59<07:34,  6.06s/it]
Training Epoch: 17/100, batch 474/625 completed (loss: 0.0403:  76%|███████▌  | 237/312 [23:59<07:34,  6.06s/it]
Training Epoch: 17/100, batch 474/625 completed (loss: 0.0007:  76%|███████▌  | 237/312 [23:59<07:33,  6.05s/it]
Training Epoch: 17/100, batch 474/625 completed (loss: 0.0403:  76%|███████▋  | 238/312 [24:02<07:31,  6.10s/it]
Training Epoch: 17/100, batch 474/625 completed (loss: 0.2072:  76%|███████▋  | 238/312 [24:02<07:31,  6.10s/it]
Training Epoch: 17/100, batch 474/625 completed (loss: 0.0007:  76%|███████▋  | 238/312 [24:02<07:31,  6.10s/it]
Training Epoch: 17/100, batch 474/625 completed (loss: 0.0347:  76%|███████▋  | 238/312 [24:02<07:31,  6.10s/it]
Training Epoch: 17/100, batch 475/625 completed (loss: 0.0048:  76%|███████▋  | 238/312 [24:02<07:31,  6.10s/it]
Training Epoch: 17/100, batch 475/625 completed (loss: 0.0054:  76%|███████▋  | 238/312 [24:02<07:31,  6.10s/it]
Training Epoch: 17/100, batch 475/625 completed (loss: 0.0362:  76%|███████▋  | 238/312 [24:02<07:31,  6.10s/it]
Training Epoch: 17/100, batch 475/625 completed (loss: 0.0138:  76%|███████▋  | 238/312 [24:02<07:31,  6.10s/it]
Training Epoch: 17/100, batch 476/625 completed (loss: 0.1141:  76%|███████▋  | 238/312 [24:05<07:31,  6.10s/it]
Training Epoch: 17/100, batch 476/625 completed (loss: 0.1484:  76%|███████▋  | 238/312 [24:05<07:31,  6.10s/it]
Training Epoch: 17/100, batch 476/625 completed (loss: 0.1484:  76%|███████▋  | 238/312 [24:05<07:31,  6.10s/it]
Training Epoch: 17/100, batch 476/625 completed (loss: 0.0767:  76%|███████▋  | 238/312 [24:05<07:31,  6.10s/it]
Training Epoch: 17/100, batch 476/625 completed (loss: 0.1484:  77%|███████▋  | 239/312 [24:08<07:22,  6.07s/it]
Training Epoch: 17/100, batch 476/625 completed (loss: 0.0767:  77%|███████▋  | 239/312 [24:08<07:23,  6.07s/it]
Training Epoch: 17/100, batch 476/625 completed (loss: 0.1141:  77%|███████▋  | 239/312 [24:08<07:23,  6.07s/it]
Training Epoch: 17/100, batch 476/625 completed (loss: 0.1484:  77%|███████▋  | 239/312 [24:08<07:23,  6.07s/it]
Training Epoch: 17/100, batch 477/625 completed (loss: 0.0606:  77%|███████▋  | 239/312 [24:08<07:23,  6.07s/it]
Training Epoch: 17/100, batch 477/625 completed (loss: 0.052:  77%|███████▋  | 239/312 [24:08<07:22,  6.07s/it] 
Training Epoch: 17/100, batch 477/625 completed (loss: 0.0052:  77%|███████▋  | 239/312 [24:08<07:23,  6.07s/it]
Training Epoch: 17/100, batch 477/625 completed (loss: 0.0409:  77%|███████▋  | 239/312 [24:08<07:23,  6.07s/it]
Training Epoch: 17/100, batch 478/625 completed (loss: 0.0016:  77%|███████▋  | 239/312 [24:11<07:22,  6.07s/it]
Training Epoch: 17/100, batch 478/625 completed (loss: 0.102:  77%|███████▋  | 239/312 [24:11<07:23,  6.07s/it] 
Training Epoch: 17/100, batch 478/625 completed (loss: 0.0696:  77%|███████▋  | 239/312 [24:11<07:23,  6.07s/it]
Training Epoch: 17/100, batch 478/625 completed (loss: 0.0143:  77%|███████▋  | 239/312 [24:11<07:23,  6.07s/it]
Training Epoch: 17/100, batch 478/625 completed (loss: 0.0696:  77%|███████▋  | 240/312 [24:14<07:15,  6.05s/it]
Training Epoch: 17/100, batch 478/625 completed (loss: 0.0016:  77%|███████▋  | 240/312 [24:14<07:15,  6.05s/it]
Training Epoch: 17/100, batch 478/625 completed (loss: 0.102:  77%|███████▋  | 240/312 [24:14<07:15,  6.05s/it]
Training Epoch: 17/100, batch 478/625 completed (loss: 0.0143:  77%|███████▋  | 240/312 [24:14<07:15,  6.06s/it]
Training Epoch: 17/100, batch 479/625 completed (loss: 0.4703:  77%|███████▋  | 240/312 [24:14<07:15,  6.05s/it]
Training Epoch: 17/100, batch 479/625 completed (loss: 0.0109:  77%|███████▋  | 240/312 [24:14<07:15,  6.05s/it]
Training Epoch: 17/100, batch 479/625 completed (loss: 0.0244:  77%|███████▋  | 240/312 [24:14<07:15,  6.05s/it]
Training Epoch: 17/100, batch 479/625 completed (loss: 0.0022:  77%|███████▋  | 240/312 [24:14<07:15,  6.06s/it]
Training Epoch: 17/100, batch 480/625 completed (loss: 0.0812:  77%|███████▋  | 240/312 [24:17<07:15,  6.05s/it]
Training Epoch: 17/100, batch 480/625 completed (loss: 0.2629:  77%|███████▋  | 240/312 [24:17<07:15,  6.05s/it]
Training Epoch: 17/100, batch 480/625 completed (loss: 0.0155:  77%|███████▋  | 240/312 [24:17<07:15,  6.05s/it]
Training Epoch: 17/100, batch 480/625 completed (loss: 0.0603:  77%|███████▋  | 240/312 [24:17<07:15,  6.06s/it]
Training Epoch: 17/100, batch 480/625 completed (loss: 0.2629:  77%|███████▋  | 241/312 [24:20<07:11,  6.08s/it]
Training Epoch: 17/100, batch 480/625 completed (loss: 0.0603:  77%|███████▋  | 241/312 [24:20<07:11,  6.08s/it]
Training Epoch: 17/100, batch 480/625 completed (loss: 0.0155:  77%|███████▋  | 241/312 [24:20<07:12,  6.09s/it]
Training Epoch: 17/100, batch 480/625 completed (loss: 0.0812:  77%|███████▋  | 241/312 [24:20<07:12,  6.09s/it]
Training Epoch: 17/100, batch 481/625 completed (loss: 0.3659:  77%|███████▋  | 241/312 [24:20<07:11,  6.08s/it]
Training Epoch: 17/100, batch 481/625 completed (loss: 0.0034:  77%|███████▋  | 241/312 [24:20<07:12,  6.09s/it]
Training Epoch: 17/100, batch 481/625 completed (loss: 0.0958:  77%|███████▋  | 241/312 [24:20<07:12,  6.09s/it]
Training Epoch: 17/100, batch 481/625 completed (loss: 0.2158:  77%|███████▋  | 241/312 [24:21<07:11,  6.08s/it]
Training Epoch: 17/100, batch 482/625 completed (loss: 0.0531:  77%|███████▋  | 241/312 [24:24<07:12,  6.09s/it]
Training Epoch: 17/100, batch 482/625 completed (loss: 0.0119:  77%|███████▋  | 241/312 [24:24<07:11,  6.08s/it]
Training Epoch: 17/100, batch 482/625 completed (loss: 0.0598:  77%|███████▋  | 241/312 [24:24<07:12,  6.09s/it]
Training Epoch: 17/100, batch 482/625 completed (loss: 0.0287:  77%|███████▋  | 241/312 [24:24<07:11,  6.08s/it]
Training Epoch: 17/100, batch 482/625 completed (loss: 0.0287:  78%|███████▊  | 242/312 [24:27<07:07,  6.10s/it]
Training Epoch: 17/100, batch 482/625 completed (loss: 0.0598:  78%|███████▊  | 242/312 [24:27<07:07,  6.11s/it]
Training Epoch: 17/100, batch 482/625 completed (loss: 0.0119:  78%|███████▊  | 242/312 [24:27<07:07,  6.11s/it]
Training Epoch: 17/100, batch 482/625 completed (loss: 0.0531:  78%|███████▊  | 242/312 [24:27<07:07,  6.11s/it]
Training Epoch: 17/100, batch 483/625 completed (loss: 0.0395:  78%|███████▊  | 242/312 [24:27<07:07,  6.11s/it]
Training Epoch: 17/100, batch 483/625 completed (loss: 0.0076:  78%|███████▊  | 242/312 [24:27<07:07,  6.11s/it]
Training Epoch: 17/100, batch 483/625 completed (loss: 0.0553:  78%|███████▊  | 242/312 [24:27<07:07,  6.11s/it]
Training Epoch: 17/100, batch 483/625 completed (loss: 0.1935:  78%|███████▊  | 242/312 [24:27<07:07,  6.10s/it]
Training Epoch: 17/100, batch 484/625 completed (loss: 0.0008:  78%|███████▊  | 242/312 [24:30<07:07,  6.11s/it]
Training Epoch: 17/100, batch 484/625 completed (loss: 0.009:  78%|███████▊  | 242/312 [24:30<07:07,  6.11s/it] 
Training Epoch: 17/100, batch 484/625 completed (loss: 0.0138:  78%|███████▊  | 242/312 [24:30<07:07,  6.11s/it]
Training Epoch: 17/100, batch 484/625 completed (loss: 0.2143:  78%|███████▊  | 242/312 [24:30<07:07,  6.10s/it]
Training Epoch: 17/100, batch 484/625 completed (loss: 0.0008:  78%|███████▊  | 243/312 [24:32<06:56,  6.03s/it]
Training Epoch: 17/100, batch 484/625 completed (loss: 0.009:  78%|███████▊  | 243/312 [24:32<06:56,  6.03s/it]
Training Epoch: 17/100, batch 484/625 completed (loss: 0.0138:  78%|███████▊  | 243/312 [24:32<06:56,  6.03s/it]
Training Epoch: 17/100, batch 484/625 completed (loss: 0.2143:  78%|███████▊  | 243/312 [24:32<06:56,  6.04s/it]
Training Epoch: 17/100, batch 485/625 completed (loss: 0.4934:  78%|███████▊  | 243/312 [24:32<06:56,  6.03s/it]
Training Epoch: 17/100, batch 485/625 completed (loss: 0.0589:  78%|███████▊  | 243/312 [24:32<06:56,  6.03s/it]
Training Epoch: 17/100, batch 485/625 completed (loss: 0.0036:  78%|███████▊  | 243/312 [24:32<06:56,  6.03s/it]
Training Epoch: 17/100, batch 485/625 completed (loss: 0.2241:  78%|███████▊  | 243/312 [24:33<06:56,  6.04s/it]
Training Epoch: 17/100, batch 486/625 completed (loss: 0.2263:  78%|███████▊  | 243/312 [24:36<06:56,  6.03s/it]
Training Epoch: 17/100, batch 486/625 completed (loss: 0.0537:  78%|███████▊  | 243/312 [24:36<06:56,  6.03s/it]
Training Epoch: 17/100, batch 486/625 completed (loss: 0.0014:  78%|███████▊  | 243/312 [24:36<06:56,  6.03s/it]
Training Epoch: 17/100, batch 486/625 completed (loss: 0.0099:  78%|███████▊  | 243/312 [24:36<06:56,  6.04s/it]
Training Epoch: 17/100, batch 486/625 completed (loss: 0.0099:  78%|███████▊  | 244/312 [24:39<06:56,  6.13s/it]
Training Epoch: 17/100, batch 486/625 completed (loss: 0.2263:  78%|███████▊  | 244/312 [24:39<06:56,  6.13s/it]
Training Epoch: 17/100, batch 486/625 completed (loss: 0.0014:  78%|███████▊  | 244/312 [24:39<06:56,  6.13s/it]
Training Epoch: 17/100, batch 486/625 completed (loss: 0.0537:  78%|███████▊  | 244/312 [24:39<06:56,  6.13s/it]
Training Epoch: 17/100, batch 487/625 completed (loss: 0.0447:  78%|███████▊  | 244/312 [24:39<06:56,  6.13s/it]
Training Epoch: 17/100, batch 487/625 completed (loss: 0.1159:  78%|███████▊  | 244/312 [24:39<06:56,  6.13s/it]
Training Epoch: 17/100, batch 487/625 completed (loss: 0.1561:  78%|███████▊  | 244/312 [24:39<06:56,  6.13s/it]
Training Epoch: 17/100, batch 487/625 completed (loss: 0.0057:  78%|███████▊  | 244/312 [24:39<06:56,  6.13s/it]
Training Epoch: 17/100, batch 488/625 completed (loss: 0.0822:  78%|███████▊  | 244/312 [24:42<06:56,  6.13s/it]
Training Epoch: 17/100, batch 488/625 completed (loss: 0.0405:  78%|███████▊  | 244/312 [24:42<06:56,  6.13s/it]
Training Epoch: 17/100, batch 488/625 completed (loss: 0.0319:  78%|███████▊  | 244/312 [24:42<06:56,  6.13s/it]
Training Epoch: 17/100, batch 488/625 completed (loss: 0.0432:  78%|███████▊  | 244/312 [24:42<06:56,  6.13s/it]
Training Epoch: 17/100, batch 488/625 completed (loss: 0.0822:  79%|███████▊  | 245/312 [24:45<06:45,  6.05s/it]
Training Epoch: 17/100, batch 488/625 completed (loss: 0.0405:  79%|███████▊  | 245/312 [24:45<06:45,  6.05s/it]
Training Epoch: 17/100, batch 488/625 completed (loss: 0.0319:  79%|███████▊  | 245/312 [24:45<06:45,  6.05s/it]
Training Epoch: 17/100, batch 488/625 completed (loss: 0.0432:  79%|███████▊  | 245/312 [24:45<06:45,  6.05s/it]
Training Epoch: 17/100, batch 489/625 completed (loss: 0.0203:  79%|███████▊  | 245/312 [24:45<06:45,  6.05s/it]
Training Epoch: 17/100, batch 489/625 completed (loss: 0.0209:  79%|███████▊  | 245/312 [24:45<06:45,  6.05s/it]
Training Epoch: 17/100, batch 489/625 completed (loss: 0.0327:  79%|███████▊  | 245/312 [24:45<06:45,  6.05s/it]
Training Epoch: 17/100, batch 489/625 completed (loss: 0.133:  79%|███████▊  | 245/312 [24:45<06:45,  6.05s/it] 
Training Epoch: 17/100, batch 490/625 completed (loss: 0.0751:  79%|███████▊  | 245/312 [24:48<06:45,  6.05s/it]
Training Epoch: 17/100, batch 490/625 completed (loss: 0.0048:  79%|███████▊  | 245/312 [24:48<06:45,  6.05s/it]
Training Epoch: 17/100, batch 490/625 completed (loss: 0.0142:  79%|███████▊  | 245/312 [24:48<06:45,  6.05s/it]
Training Epoch: 17/100, batch 490/625 completed (loss: 0.0383:  79%|███████▊  | 245/312 [24:48<06:45,  6.05s/it]
Training Epoch: 17/100, batch 490/625 completed (loss: 0.0142:  79%|███████▉  | 246/312 [24:51<06:40,  6.08s/it]
Training Epoch: 17/100, batch 490/625 completed (loss: 0.0048:  79%|███████▉  | 246/312 [24:51<06:41,  6.08s/it]
Training Epoch: 17/100, batch 490/625 completed (loss: 0.0383:  79%|███████▉  | 246/312 [24:51<06:41,  6.08s/it]
Training Epoch: 17/100, batch 490/625 completed (loss: 0.0751:  79%|███████▉  | 246/312 [24:51<06:41,  6.08s/it]
Training Epoch: 17/100, batch 491/625 completed (loss: 0.265:  79%|███████▉  | 246/312 [24:51<06:41,  6.08s/it] 
Training Epoch: 17/100, batch 491/625 completed (loss: 0.0087:  79%|███████▉  | 246/312 [24:51<06:41,  6.08s/it]
Training Epoch: 17/100, batch 491/625 completed (loss: 0.0475:  79%|███████▉  | 246/312 [24:51<06:40,  6.08s/it]
Training Epoch: 17/100, batch 491/625 completed (loss: 0.2826:  79%|███████▉  | 246/312 [24:51<06:41,  6.08s/it]
Training Epoch: 17/100, batch 492/625 completed (loss: 0.0296:  79%|███████▉  | 246/312 [24:54<06:41,  6.08s/it]
Training Epoch: 17/100, batch 492/625 completed (loss: 0.1724:  79%|███████▉  | 246/312 [24:54<06:41,  6.08s/it]
Training Epoch: 17/100, batch 492/625 completed (loss: 0.0084:  79%|███████▉  | 246/312 [24:54<06:40,  6.08s/it]
Training Epoch: 17/100, batch 492/625 completed (loss: 0.0225:  79%|███████▉  | 246/312 [24:54<06:41,  6.08s/it]
Training Epoch: 17/100, batch 492/625 completed (loss: 0.0225:  79%|███████▉  | 247/312 [24:57<06:36,  6.10s/it]
Training Epoch: 17/100, batch 492/625 completed (loss: 0.0084:  79%|███████▉  | 247/312 [24:57<06:36,  6.10s/it]
Training Epoch: 17/100, batch 492/625 completed (loss: 0.1724:  79%|███████▉  | 247/312 [24:57<06:36,  6.10s/it]
Training Epoch: 17/100, batch 492/625 completed (loss: 0.0296:  79%|███████▉  | 247/312 [24:57<06:36,  6.10s/it]
Training Epoch: 17/100, batch 493/625 completed (loss: 0.0168:  79%|███████▉  | 247/312 [24:57<06:36,  6.10s/it]
Training Epoch: 17/100, batch 493/625 completed (loss: 0.0549:  79%|███████▉  | 247/312 [24:57<06:36,  6.10s/it]
Training Epoch: 17/100, batch 493/625 completed (loss: 0.0626:  79%|███████▉  | 247/312 [24:57<06:36,  6.10s/it]
Training Epoch: 17/100, batch 493/625 completed (loss: 0.0821:  79%|███████▉  | 247/312 [24:57<06:36,  6.10s/it]
Training Epoch: 17/100, batch 494/625 completed (loss: 0.0271:  79%|███████▉  | 247/312 [25:00<06:36,  6.10s/it]
Training Epoch: 17/100, batch 494/625 completed (loss: 0.0045:  79%|███████▉  | 247/312 [25:00<06:36,  6.10s/it]
Training Epoch: 17/100, batch 494/625 completed (loss: 0.0734:  79%|███████▉  | 247/312 [25:00<06:36,  6.10s/it]
Training Epoch: 17/100, batch 494/625 completed (loss: 0.0492:  79%|███████▉  | 247/312 [25:00<06:36,  6.10s/it]
Training Epoch: 17/100, batch 494/625 completed (loss: 0.0045:  79%|███████▉  | 248/312 [25:03<06:25,  6.02s/it]
Training Epoch: 17/100, batch 494/625 completed (loss: 0.0734:  79%|███████▉  | 248/312 [25:03<06:25,  6.02s/it]
Training Epoch: 17/100, batch 494/625 completed (loss: 0.0271:  79%|███████▉  | 248/312 [25:03<06:25,  6.02s/it]
Training Epoch: 17/100, batch 494/625 completed (loss: 0.0492:  79%|███████▉  | 248/312 [25:03<06:25,  6.02s/it]
Training Epoch: 17/100, batch 495/625 completed (loss: 0.0063:  79%|███████▉  | 248/312 [25:03<06:25,  6.02s/it]
Training Epoch: 17/100, batch 495/625 completed (loss: 0.0067:  79%|███████▉  | 248/312 [25:03<06:25,  6.02s/it]
Training Epoch: 17/100, batch 495/625 completed (loss: 0.0185:  79%|███████▉  | 248/312 [25:03<06:25,  6.02s/it]
Training Epoch: 17/100, batch 495/625 completed (loss: 0.1736:  79%|███████▉  | 248/312 [25:03<06:25,  6.02s/it]
Training Epoch: 17/100, batch 496/625 completed (loss: 0.0263:  79%|███████▉  | 248/312 [25:06<06:25,  6.02s/it]
Training Epoch: 17/100, batch 496/625 completed (loss: 0.1178:  79%|███████▉  | 248/312 [25:06<06:25,  6.02s/it]
Training Epoch: 17/100, batch 496/625 completed (loss: 0.0262:  79%|███████▉  | 248/312 [25:06<06:25,  6.02s/it]
Training Epoch: 17/100, batch 496/625 completed (loss: 0.0916:  79%|███████▉  | 248/312 [25:06<06:25,  6.02s/it]
Training Epoch: 17/100, batch 496/625 completed (loss: 0.0263:  80%|███████▉  | 249/312 [25:09<06:21,  6.06s/it]
Training Epoch: 17/100, batch 496/625 completed (loss: 0.1178:  80%|███████▉  | 249/312 [25:09<06:22,  6.06s/it]
Training Epoch: 17/100, batch 496/625 completed (loss: 0.0916:  80%|███████▉  | 249/312 [25:09<06:22,  6.06s/it]
Training Epoch: 17/100, batch 496/625 completed (loss: 0.0262:  80%|███████▉  | 249/312 [25:09<06:22,  6.07s/it]
Training Epoch: 17/100, batch 497/625 completed (loss: 0.0501:  80%|███████▉  | 249/312 [25:09<06:22,  6.06s/it]
Training Epoch: 17/100, batch 497/625 completed (loss: 0.1627:  80%|███████▉  | 249/312 [25:09<06:22,  6.07s/it]
Training Epoch: 17/100, batch 497/625 completed (loss: 0.0577:  80%|███████▉  | 249/312 [25:09<06:21,  6.06s/it]
Training Epoch: 17/100, batch 497/625 completed (loss: 0.0382:  80%|███████▉  | 249/312 [25:09<06:22,  6.06s/it]
Training Epoch: 17/100, batch 498/625 completed (loss: 0.0095:  80%|███████▉  | 249/312 [25:12<06:22,  6.07s/it]
Training Epoch: 17/100, batch 498/625 completed (loss: 0.0023:  80%|███████▉  | 249/312 [25:12<06:22,  6.06s/it]
Training Epoch: 17/100, batch 498/625 completed (loss: 0.0178:  80%|███████▉  | 249/312 [25:12<06:21,  6.06s/it]
Training Epoch: 17/100, batch 498/625 completed (loss: 0.0953:  80%|███████▉  | 249/312 [25:12<06:22,  6.06s/it]
Training Epoch: 17/100, batch 498/625 completed (loss: 0.0023:  80%|████████  | 250/312 [25:15<06:15,  6.05s/it]
Training Epoch: 17/100, batch 498/625 completed (loss: 0.0095:  80%|████████  | 250/312 [25:15<06:15,  6.05s/it]
Training Epoch: 17/100, batch 498/625 completed (loss: 0.0953:  80%|████████  | 250/312 [25:15<06:15,  6.06s/it]
Training Epoch: 17/100, batch 498/625 completed (loss: 0.0178:  80%|████████  | 250/312 [25:15<06:15,  6.06s/it]
Training Epoch: 17/100, batch 499/625 completed (loss: 0.1647:  80%|████████  | 250/312 [25:15<06:15,  6.06s/it]
Training Epoch: 17/100, batch 499/625 completed (loss: 0.1766:  80%|████████  | 250/312 [25:15<06:15,  6.05s/it]
Training Epoch: 17/100, batch 499/625 completed (loss: 0.0106:  80%|████████  | 250/312 [25:15<06:15,  6.05s/it]
Training Epoch: 17/100, batch 499/625 completed (loss: 0.0166:  80%|████████  | 250/312 [25:15<06:15,  6.06s/it]
Training Epoch: 17/100, batch 500/625 completed (loss: 0.1571:  80%|████████  | 250/312 [25:18<06:15,  6.05s/it]
Training Epoch: 17/100, batch 500/625 completed (loss: 0.0147:  80%|████████  | 250/312 [25:18<06:15,  6.05s/it]
Training Epoch: 17/100, batch 500/625 completed (loss: 0.0465:  80%|████████  | 250/312 [25:18<06:15,  6.06s/it]
Training Epoch: 17/100, batch 500/625 completed (loss: 0.0235:  80%|████████  | 250/312 [25:18<06:15,  6.06s/it]
Training Epoch: 17/100, batch 500/625 completed (loss: 0.0235:  80%|████████  | 251/312 [25:21<06:08,  6.04s/it]
Training Epoch: 17/100, batch 500/625 completed (loss: 0.0147:  80%|████████  | 251/312 [25:21<06:08,  6.04s/it]
Training Epoch: 17/100, batch 500/625 completed (loss: 0.1571:  80%|████████  | 251/312 [25:21<06:08,  6.04s/it]
Training Epoch: 17/100, batch 500/625 completed (loss: 0.0465:  80%|████████  | 251/312 [25:21<06:08,  6.04s/it]
Training Epoch: 17/100, batch 501/625 completed (loss: 0.0453:  80%|████████  | 251/312 [25:21<06:08,  6.04s/it]
Training Epoch: 17/100, batch 501/625 completed (loss: 0.0062:  80%|████████  | 251/312 [25:21<06:08,  6.04s/it]
Training Epoch: 17/100, batch 501/625 completed (loss: 0.1439:  80%|████████  | 251/312 [25:21<06:08,  6.04s/it]
Training Epoch: 17/100, batch 501/625 completed (loss: 0.0098:  80%|████████  | 251/312 [25:21<06:08,  6.04s/it]
Training Epoch: 17/100, batch 502/625 completed (loss: 0.0941:  80%|████████  | 251/312 [25:24<06:08,  6.04s/it]
Training Epoch: 17/100, batch 502/625 completed (loss: 0.0422:  80%|████████  | 251/312 [25:24<06:08,  6.04s/it]
Training Epoch: 17/100, batch 502/625 completed (loss: 0.1323:  80%|████████  | 251/312 [25:24<06:08,  6.04s/it]
Training Epoch: 17/100, batch 502/625 completed (loss: 0.1768:  80%|████████  | 251/312 [25:24<06:08,  6.04s/it]
Training Epoch: 17/100, batch 502/625 completed (loss: 0.0941:  81%|████████  | 252/312 [25:27<06:04,  6.07s/it]
Training Epoch: 17/100, batch 502/625 completed (loss: 0.1768:  81%|████████  | 252/312 [25:27<06:04,  6.07s/it]
Training Epoch: 17/100, batch 502/625 completed (loss: 0.0422:  81%|████████  | 252/312 [25:27<06:04,  6.07s/it]
Training Epoch: 17/100, batch 502/625 completed (loss: 0.1323:  81%|████████  | 252/312 [25:27<06:04,  6.07s/it]
Training Epoch: 17/100, batch 503/625 completed (loss: 0.0281:  81%|████████  | 252/312 [25:27<06:04,  6.07s/it]
Training Epoch: 17/100, batch 503/625 completed (loss: 0.0645:  81%|████████  | 252/312 [25:27<06:04,  6.07s/it]
Training Epoch: 17/100, batch 503/625 completed (loss: 0.0156:  81%|████████  | 252/312 [25:27<06:04,  6.07s/it]
Training Epoch: 17/100, batch 503/625 completed (loss: 0.0372:  81%|████████  | 252/312 [25:27<06:04,  6.07s/it]
Training Epoch: 17/100, batch 504/625 completed (loss: 0.025:  81%|████████  | 252/312 [25:30<06:04,  6.07s/it] 
Training Epoch: 17/100, batch 504/625 completed (loss: 0.1553:  81%|████████  | 252/312 [25:30<06:04,  6.07s/it]
Training Epoch: 17/100, batch 504/625 completed (loss: 0.131:  81%|████████  | 252/312 [25:30<06:04,  6.07s/it] 
Training Epoch: 17/100, batch 504/625 completed (loss: 0.1777:  81%|████████  | 252/312 [25:30<06:04,  6.07s/it]
Training Epoch: 17/100, batch 504/625 completed (loss: 0.025:  81%|████████  | 253/312 [25:33<05:56,  6.05s/it]
Training Epoch: 17/100, batch 504/625 completed (loss: 0.1553:  81%|████████  | 253/312 [25:33<05:56,  6.05s/it]
Training Epoch: 17/100, batch 504/625 completed (loss: 0.131:  81%|████████  | 253/312 [25:33<05:56,  6.04s/it]
Training Epoch: 17/100, batch 504/625 completed (loss: 0.1777:  81%|████████  | 253/312 [25:33<05:56,  6.05s/it]
Training Epoch: 17/100, batch 505/625 completed (loss: 0.0971:  81%|████████  | 253/312 [25:33<05:56,  6.05s/it]
Training Epoch: 17/100, batch 505/625 completed (loss: 0.1811:  81%|████████  | 253/312 [25:33<05:56,  6.05s/it]
Training Epoch: 17/100, batch 505/625 completed (loss: 0.0963:  81%|████████  | 253/312 [25:33<05:56,  6.04s/it]
Training Epoch: 17/100, batch 505/625 completed (loss: 0.0046:  81%|████████  | 253/312 [25:33<05:56,  6.05s/it]
Training Epoch: 17/100, batch 506/625 completed (loss: 0.089:  81%|████████  | 253/312 [25:36<05:56,  6.05s/it] 
Training Epoch: 17/100, batch 506/625 completed (loss: 0.0028:  81%|████████  | 253/312 [25:36<05:56,  6.05s/it]
Training Epoch: 17/100, batch 506/625 completed (loss: 0.0027:  81%|████████  | 253/312 [25:36<05:56,  6.04s/it]
Training Epoch: 17/100, batch 506/625 completed (loss: 0.2503:  81%|████████  | 253/312 [25:36<05:56,  6.05s/it]
Training Epoch: 17/100, batch 506/625 completed (loss: 0.089:  81%|████████▏ | 254/312 [25:39<05:49,  6.03s/it]
Training Epoch: 17/100, batch 506/625 completed (loss: 0.0028:  81%|████████▏ | 254/312 [25:39<05:49,  6.03s/it]
Training Epoch: 17/100, batch 506/625 completed (loss: 0.0027:  81%|████████▏ | 254/312 [25:39<05:49,  6.03s/it]
Training Epoch: 17/100, batch 506/625 completed (loss: 0.2503:  81%|████████▏ | 254/312 [25:39<05:49,  6.03s/it]
Training Epoch: 17/100, batch 507/625 completed (loss: 0.0067:  81%|████████▏ | 254/312 [25:39<05:49,  6.03s/it]
Training Epoch: 17/100, batch 507/625 completed (loss: 0.0111:  81%|████████▏ | 254/312 [25:39<05:49,  6.03s/it]
Training Epoch: 17/100, batch 507/625 completed (loss: 0.1897:  81%|████████▏ | 254/312 [25:39<05:49,  6.03s/it]
Training Epoch: 17/100, batch 507/625 completed (loss: 0.018:  81%|████████▏ | 254/312 [25:39<05:49,  6.03s/it] 
Training Epoch: 17/100, batch 508/625 completed (loss: 0.2283:  81%|████████▏ | 254/312 [25:43<05:49,  6.03s/it]
Training Epoch: 17/100, batch 508/625 completed (loss: 0.026:  81%|████████▏ | 254/312 [25:43<05:49,  6.03s/it] 
Training Epoch: 17/100, batch 508/625 completed (loss: 0.0501:  81%|████████▏ | 254/312 [25:43<05:49,  6.03s/it]
Training Epoch: 17/100, batch 508/625 completed (loss: 0.0011:  81%|████████▏ | 254/312 [25:43<05:49,  6.03s/it]
Training Epoch: 17/100, batch 508/625 completed (loss: 0.0501:  82%|████████▏ | 255/312 [25:45<05:48,  6.11s/it]
Training Epoch: 17/100, batch 508/625 completed (loss: 0.2283:  82%|████████▏ | 255/312 [25:45<05:48,  6.11s/it]
Training Epoch: 17/100, batch 508/625 completed (loss: 0.026:  82%|████████▏ | 255/312 [25:45<05:48,  6.11s/it]
Training Epoch: 17/100, batch 508/625 completed (loss: 0.0011:  82%|████████▏ | 255/312 [25:45<05:48,  6.11s/it]
Training Epoch: 17/100, batch 509/625 completed (loss: 0.0555:  82%|████████▏ | 255/312 [25:45<05:48,  6.11s/it]
Training Epoch: 17/100, batch 509/625 completed (loss: 0.0156:  82%|████████▏ | 255/312 [25:45<05:48,  6.11s/it]
Training Epoch: 17/100, batch 509/625 completed (loss: 0.0025:  82%|████████▏ | 255/312 [25:45<05:48,  6.11s/it]
Training Epoch: 17/100, batch 509/625 completed (loss: 0.066:  82%|████████▏ | 255/312 [25:45<05:48,  6.11s/it] 
Training Epoch: 17/100, batch 510/625 completed (loss: 0.0445:  82%|████████▏ | 255/312 [25:48<05:48,  6.11s/it]
Training Epoch: 17/100, batch 510/625 completed (loss: 0.0091:  82%|████████▏ | 255/312 [25:48<05:48,  6.11s/it]
Training Epoch: 17/100, batch 510/625 completed (loss: 0.1122:  82%|████████▏ | 255/312 [25:48<05:48,  6.11s/it]
Training Epoch: 17/100, batch 510/625 completed (loss: 0.0481:  82%|████████▏ | 255/312 [25:48<05:48,  6.11s/it]
Training Epoch: 17/100, batch 510/625 completed (loss: 0.0091:  82%|████████▏ | 256/312 [25:51<05:37,  6.02s/it]
Training Epoch: 17/100, batch 510/625 completed (loss: 0.0481:  82%|████████▏ | 256/312 [25:51<05:37,  6.02s/it]
Training Epoch: 17/100, batch 510/625 completed (loss: 0.0445:  82%|████████▏ | 256/312 [25:51<05:37,  6.03s/it]
Training Epoch: 17/100, batch 510/625 completed (loss: 0.1122:  82%|████████▏ | 256/312 [25:51<05:37,  6.03s/it]
Training Epoch: 17/100, batch 511/625 completed (loss: 0.0353:  82%|████████▏ | 256/312 [25:51<05:37,  6.03s/it]
Training Epoch: 17/100, batch 511/625 completed (loss: 0.1556:  82%|████████▏ | 256/312 [25:51<05:37,  6.03s/it]
Training Epoch: 17/100, batch 511/625 completed (loss: 0.0031:  82%|████████▏ | 256/312 [25:51<05:37,  6.02s/it]
Training Epoch: 17/100, batch 511/625 completed (loss: 0.0095:  82%|████████▏ | 256/312 [25:51<05:37,  6.02s/it]
Training Epoch: 17/100, batch 512/625 completed (loss: 0.0751:  82%|████████▏ | 256/312 [25:54<05:37,  6.03s/it]
Training Epoch: 17/100, batch 512/625 completed (loss: 0.347:  82%|████████▏ | 256/312 [25:54<05:37,  6.02s/it] 
Training Epoch: 17/100, batch 512/625 completed (loss: 0.0199:  82%|████████▏ | 256/312 [25:54<05:37,  6.03s/it]
Training Epoch: 17/100, batch 512/625 completed (loss: 0.1072:  82%|████████▏ | 256/312 [25:54<05:37,  6.02s/it]
Training Epoch: 17/100, batch 512/625 completed (loss: 0.1072:  82%|████████▏ | 257/312 [25:58<05:36,  6.11s/it]
Training Epoch: 17/100, batch 512/625 completed (loss: 0.0751:  82%|████████▏ | 257/312 [25:58<05:36,  6.12s/it]
Training Epoch: 17/100, batch 512/625 completed (loss: 0.347:  82%|████████▏ | 257/312 [25:58<05:36,  6.12s/it]
Training Epoch: 17/100, batch 512/625 completed (loss: 0.0199:  82%|████████▏ | 257/312 [25:58<05:36,  6.12s/it]
Training Epoch: 17/100, batch 513/625 completed (loss: 0.0033:  82%|████████▏ | 257/312 [25:58<05:36,  6.12s/it]
Training Epoch: 17/100, batch 513/625 completed (loss: 0.0565:  82%|████████▏ | 257/312 [25:58<05:36,  6.12s/it]
Training Epoch: 17/100, batch 513/625 completed (loss: 0.0682:  82%|████████▏ | 257/312 [25:58<05:36,  6.12s/it]
Training Epoch: 17/100, batch 513/625 completed (loss: 0.0481:  82%|████████▏ | 257/312 [25:58<05:36,  6.11s/it]
Training Epoch: 17/100, batch 514/625 completed (loss: 0.027:  82%|████████▏ | 257/312 [26:01<05:36,  6.12s/it] 
Training Epoch: 17/100, batch 514/625 completed (loss: 0.0587:  82%|████████▏ | 257/312 [26:01<05:36,  6.12s/it]
Training Epoch: 17/100, batch 514/625 completed (loss: 0.0054:  82%|████████▏ | 257/312 [26:01<05:36,  6.12s/it]
Training Epoch: 17/100, batch 514/625 completed (loss: 0.0328:  82%|████████▏ | 257/312 [26:01<05:36,  6.11s/it]
Training Epoch: 17/100, batch 514/625 completed (loss: 0.0054:  83%|████████▎ | 258/312 [26:04<05:28,  6.09s/it]
Training Epoch: 17/100, batch 514/625 completed (loss: 0.027:  83%|████████▎ | 258/312 [26:04<05:28,  6.09s/it]
Training Epoch: 17/100, batch 514/625 completed (loss: 0.0587:  83%|████████▎ | 258/312 [26:04<05:28,  6.09s/it]
Training Epoch: 17/100, batch 514/625 completed (loss: 0.0328:  83%|████████▎ | 258/312 [26:04<05:28,  6.09s/it]
Training Epoch: 17/100, batch 515/625 completed (loss: 0.0245:  83%|████████▎ | 258/312 [26:04<05:28,  6.09s/it]
Training Epoch: 17/100, batch 515/625 completed (loss: 0.0196:  83%|████████▎ | 258/312 [26:04<05:28,  6.09s/it]
Training Epoch: 17/100, batch 515/625 completed (loss: 0.0139:  83%|████████▎ | 258/312 [26:04<05:28,  6.09s/it]
Training Epoch: 17/100, batch 515/625 completed (loss: 0.3876:  83%|████████▎ | 258/312 [26:04<05:28,  6.09s/it]
Training Epoch: 17/100, batch 516/625 completed (loss: 0.0238:  83%|████████▎ | 258/312 [26:07<05:28,  6.09s/it]
Training Epoch: 17/100, batch 516/625 completed (loss: 0.3034:  83%|████████▎ | 258/312 [26:07<05:28,  6.09s/it]
Training Epoch: 17/100, batch 516/625 completed (loss: 0.0059:  83%|████████▎ | 258/312 [26:07<05:28,  6.09s/it]
Training Epoch: 17/100, batch 516/625 completed (loss: 0.0045:  83%|████████▎ | 258/312 [26:07<05:28,  6.09s/it]
Training Epoch: 17/100, batch 516/625 completed (loss: 0.0238:  83%|████████▎ | 259/312 [26:10<05:21,  6.06s/it]
Training Epoch: 17/100, batch 516/625 completed (loss: 0.3034:  83%|████████▎ | 259/312 [26:10<05:21,  6.06s/it]
Training Epoch: 17/100, batch 516/625 completed (loss: 0.0045:  83%|████████▎ | 259/312 [26:10<05:21,  6.06s/it]
Training Epoch: 17/100, batch 516/625 completed (loss: 0.0059:  83%|████████▎ | 259/312 [26:10<05:21,  6.07s/it]
Training Epoch: 17/100, batch 517/625 completed (loss: 0.0178:  83%|████████▎ | 259/312 [26:10<05:21,  6.06s/it]
Training Epoch: 17/100, batch 517/625 completed (loss: 0.02:  83%|████████▎ | 259/312 [26:10<05:21,  6.07s/it]  
Training Epoch: 17/100, batch 517/625 completed (loss: 0.0174:  83%|████████▎ | 259/312 [26:10<05:21,  6.06s/it]
Training Epoch: 17/100, batch 517/625 completed (loss: 0.0231:  83%|████████▎ | 259/312 [26:10<05:21,  6.06s/it]
Training Epoch: 17/100, batch 518/625 completed (loss: 0.031:  83%|████████▎ | 259/312 [26:13<05:21,  6.07s/it]
Training Epoch: 17/100, batch 518/625 completed (loss: 0.0196:  83%|████████▎ | 259/312 [26:13<05:21,  6.06s/it]
Training Epoch: 17/100, batch 518/625 completed (loss: 0.0196:  83%|████████▎ | 259/312 [26:13<05:21,  6.06s/it]
Training Epoch: 17/100, batch 518/625 completed (loss: 0.045:  83%|████████▎ | 259/312 [26:13<05:21,  6.06s/it] 
Training Epoch: 17/100, batch 518/625 completed (loss: 0.031:  83%|████████▎ | 260/312 [26:16<05:19,  6.13s/it]
Training Epoch: 17/100, batch 518/625 completed (loss: 0.0196:  83%|████████▎ | 260/312 [26:16<05:19,  6.14s/it]
Training Epoch: 17/100, batch 518/625 completed (loss: 0.0196:  83%|████████▎ | 260/312 [26:16<05:19,  6.14s/it]
Training Epoch: 17/100, batch 518/625 completed (loss: 0.045:  83%|████████▎ | 260/312 [26:16<05:19,  6.14s/it]
Training Epoch: 17/100, batch 519/625 completed (loss: 0.0041:  83%|████████▎ | 260/312 [26:16<05:19,  6.14s/it]
Training Epoch: 17/100, batch 519/625 completed (loss: 0.0029:  83%|████████▎ | 260/312 [26:16<05:19,  6.13s/it]
Training Epoch: 17/100, batch 519/625 completed (loss: 0.1201:  83%|████████▎ | 260/312 [26:16<05:19,  6.14s/it]
Training Epoch: 17/100, batch 519/625 completed (loss: 0.0095:  83%|████████▎ | 260/312 [26:16<05:19,  6.14s/it]
Training Epoch: 17/100, batch 520/625 completed (loss: 0.0297:  83%|████████▎ | 260/312 [26:19<05:19,  6.13s/it]
Training Epoch: 17/100, batch 520/625 completed (loss: 0.0028:  83%|████████▎ | 260/312 [26:19<05:19,  6.14s/it]
Training Epoch: 17/100, batch 520/625 completed (loss: 0.0007:  83%|████████▎ | 260/312 [26:19<05:19,  6.14s/it]
Training Epoch: 17/100, batch 520/625 completed (loss: 0.0106:  83%|████████▎ | 260/312 [26:19<05:19,  6.14s/it]
Training Epoch: 17/100, batch 520/625 completed (loss: 0.0028:  84%|████████▎ | 261/312 [26:22<05:08,  6.05s/it]
Training Epoch: 17/100, batch 520/625 completed (loss: 0.0007:  84%|████████▎ | 261/312 [26:22<05:08,  6.05s/it]
Training Epoch: 17/100, batch 520/625 completed (loss: 0.0297:  84%|████████▎ | 261/312 [26:22<05:08,  6.05s/it]
Training Epoch: 17/100, batch 520/625 completed (loss: 0.0106:  84%|████████▎ | 261/312 [26:22<05:08,  6.05s/it]
Training Epoch: 17/100, batch 521/625 completed (loss: 0.1305:  84%|████████▎ | 261/312 [26:22<05:08,  6.05s/it]
Training Epoch: 17/100, batch 521/625 completed (loss: 0.1779:  84%|████████▎ | 261/312 [26:22<05:08,  6.05s/it]
Training Epoch: 17/100, batch 521/625 completed (loss: 0.0126:  84%|████████▎ | 261/312 [26:22<05:08,  6.05s/it]
Training Epoch: 17/100, batch 521/625 completed (loss: 0.1167:  84%|████████▎ | 261/312 [26:22<05:08,  6.05s/it]
Training Epoch: 17/100, batch 522/625 completed (loss: 0.5783:  84%|████████▎ | 261/312 [26:25<05:08,  6.05s/it]
Training Epoch: 17/100, batch 522/625 completed (loss: 0.0068:  84%|████████▎ | 261/312 [26:25<05:08,  6.05s/it]
Training Epoch: 17/100, batch 522/625 completed (loss: 0.0033:  84%|████████▎ | 261/312 [26:25<05:08,  6.05s/it]
Training Epoch: 17/100, batch 522/625 completed (loss: 0.0105:  84%|████████▎ | 261/312 [26:25<05:08,  6.05s/it]
Training Epoch: 17/100, batch 522/625 completed (loss: 0.5783:  84%|████████▍ | 262/312 [26:28<05:01,  6.04s/it]
Training Epoch: 17/100, batch 522/625 completed (loss: 0.0033:  84%|████████▍ | 262/312 [26:28<05:02,  6.04s/it]
Training Epoch: 17/100, batch 522/625 completed (loss: 0.0105:  84%|████████▍ | 262/312 [26:28<05:02,  6.04s/it]
Training Epoch: 17/100, batch 522/625 completed (loss: 0.0068:  84%|████████▍ | 262/312 [26:28<05:02,  6.04s/it]
Training Epoch: 17/100, batch 523/625 completed (loss: 0.0783:  84%|████████▍ | 262/312 [26:28<05:01,  6.04s/it]
Training Epoch: 17/100, batch 523/625 completed (loss: 0.003:  84%|████████▍ | 262/312 [26:28<05:02,  6.04s/it] 
Training Epoch: 17/100, batch 523/625 completed (loss: 0.0319:  84%|████████▍ | 262/312 [26:28<05:02,  6.04s/it]
Training Epoch: 17/100, batch 523/625 completed (loss: 0.021:  84%|████████▍ | 262/312 [26:28<05:02,  6.04s/it] 
Training Epoch: 17/100, batch 524/625 completed (loss: 0.0578:  84%|████████▍ | 262/312 [26:31<05:02,  6.04s/it]
Training Epoch: 17/100, batch 524/625 completed (loss: 0.1408:  84%|████████▍ | 262/312 [26:31<05:01,  6.04s/it]
Training Epoch: 17/100, batch 524/625 completed (loss: 0.006:  84%|████████▍ | 262/312 [26:31<05:02,  6.04s/it]
Training Epoch: 17/100, batch 524/625 completed (loss: 0.0066:  84%|████████▍ | 262/312 [26:31<05:02,  6.04s/it]
Training Epoch: 17/100, batch 524/625 completed (loss: 0.0578:  84%|████████▍ | 263/312 [26:34<04:59,  6.12s/it]
Training Epoch: 17/100, batch 524/625 completed (loss: 0.006:  84%|████████▍ | 263/312 [26:34<04:59,  6.12s/it]
Training Epoch: 17/100, batch 524/625 completed (loss: 0.1408:  84%|████████▍ | 263/312 [26:34<04:59,  6.12s/it]
Training Epoch: 17/100, batch 524/625 completed (loss: 0.0066:  84%|████████▍ | 263/312 [26:34<04:59,  6.12s/it]
Training Epoch: 17/100, batch 525/625 completed (loss: 0.0047:  84%|████████▍ | 263/312 [26:34<04:59,  6.12s/it]
Training Epoch: 17/100, batch 525/625 completed (loss: 0.0267:  84%|████████▍ | 263/312 [26:34<04:59,  6.12s/it]
Training Epoch: 17/100, batch 525/625 completed (loss: 0.1463:  84%|████████▍ | 263/312 [26:34<04:59,  6.12s/it]
Training Epoch: 17/100, batch 525/625 completed (loss: 0.0325:  84%|████████▍ | 263/312 [26:34<04:59,  6.12s/it]
Training Epoch: 17/100, batch 526/625 completed (loss: 0.0479:  84%|████████▍ | 263/312 [26:37<04:59,  6.12s/it]
Training Epoch: 17/100, batch 526/625 completed (loss: 0.0525:  84%|████████▍ | 263/312 [26:37<04:59,  6.12s/it]
Training Epoch: 17/100, batch 526/625 completed (loss: 0.0193:  84%|████████▍ | 263/312 [26:37<04:59,  6.12s/it]
Training Epoch: 17/100, batch 526/625 completed (loss: 0.3485:  84%|████████▍ | 263/312 [26:37<04:59,  6.12s/it]
Training Epoch: 17/100, batch 526/625 completed (loss: 0.0193:  85%|████████▍ | 264/312 [26:40<04:51,  6.08s/it]
Training Epoch: 17/100, batch 526/625 completed (loss: 0.0479:  85%|████████▍ | 264/312 [26:40<04:51,  6.08s/it]
Training Epoch: 17/100, batch 526/625 completed (loss: 0.0525:  85%|████████▍ | 264/312 [26:40<04:51,  6.08s/it]
Training Epoch: 17/100, batch 526/625 completed (loss: 0.3485:  85%|████████▍ | 264/312 [26:40<04:52,  6.08s/it]
Training Epoch: 17/100, batch 527/625 completed (loss: 0.0192:  85%|████████▍ | 264/312 [26:40<04:51,  6.08s/it]
Training Epoch: 17/100, batch 527/625 completed (loss: 0.0194:  85%|████████▍ | 264/312 [26:40<04:51,  6.08s/it]
Training Epoch: 17/100, batch 527/625 completed (loss: 0.0043:  85%|████████▍ | 264/312 [26:40<04:51,  6.08s/it]
Training Epoch: 17/100, batch 527/625 completed (loss: 0.1037:  85%|████████▍ | 264/312 [26:40<04:52,  6.08s/it]
Training Epoch: 17/100, batch 528/625 completed (loss: 0.0069:  85%|████████▍ | 264/312 [26:43<04:51,  6.08s/it]
Training Epoch: 17/100, batch 528/625 completed (loss: 0.2266:  85%|████████▍ | 264/312 [26:43<04:51,  6.08s/it]
Training Epoch: 17/100, batch 528/625 completed (loss: 0.0099:  85%|████████▍ | 264/312 [26:43<04:51,  6.08s/it]
Training Epoch: 17/100, batch 528/625 completed (loss: 0.0089:  85%|████████▍ | 264/312 [26:43<04:52,  6.08s/it]
Training Epoch: 17/100, batch 528/625 completed (loss: 0.0069:  85%|████████▍ | 265/312 [26:46<04:48,  6.14s/it]
Training Epoch: 17/100, batch 528/625 completed (loss: 0.0099:  85%|████████▍ | 265/312 [26:46<04:48,  6.14s/it]
Training Epoch: 17/100, batch 528/625 completed (loss: 0.2266:  85%|████████▍ | 265/312 [26:46<04:48,  6.14s/it]
Training Epoch: 17/100, batch 528/625 completed (loss: 0.0089:  85%|████████▍ | 265/312 [26:46<04:48,  6.14s/it]
Training Epoch: 17/100, batch 529/625 completed (loss: 0.0713:  85%|████████▍ | 265/312 [26:46<04:48,  6.14s/it]
Training Epoch: 17/100, batch 529/625 completed (loss: 0.0018:  85%|████████▍ | 265/312 [26:46<04:48,  6.14s/it]
Training Epoch: 17/100, batch 529/625 completed (loss: 0.0069:  85%|████████▍ | 265/312 [26:46<04:48,  6.14s/it]
Training Epoch: 17/100, batch 529/625 completed (loss: 0.0076:  85%|████████▍ | 265/312 [26:46<04:48,  6.14s/it]
Training Epoch: 17/100, batch 530/625 completed (loss: 0.0087:  85%|████████▍ | 265/312 [26:49<04:48,  6.14s/it]
Training Epoch: 17/100, batch 530/625 completed (loss: 0.1174:  85%|████████▍ | 265/312 [26:49<04:48,  6.14s/it]
Training Epoch: 17/100, batch 530/625 completed (loss: 0.1564:  85%|████████▍ | 265/312 [26:49<04:48,  6.14s/it]
Training Epoch: 17/100, batch 530/625 completed (loss: 0.0015:  85%|████████▍ | 265/312 [26:49<04:48,  6.14s/it]
Training Epoch: 17/100, batch 530/625 completed (loss: 0.0087:  85%|████████▌ | 266/312 [26:52<04:38,  6.06s/it]
Training Epoch: 17/100, batch 530/625 completed (loss: 0.1174:  85%|████████▌ | 266/312 [26:52<04:38,  6.06s/it]
Training Epoch: 17/100, batch 530/625 completed (loss: 0.1564:  85%|████████▌ | 266/312 [26:52<04:39,  6.07s/it]
Training Epoch: 17/100, batch 530/625 completed (loss: 0.0015:  85%|████████▌ | 266/312 [26:52<04:39,  6.07s/it]
Training Epoch: 17/100, batch 531/625 completed (loss: 0.0061:  85%|████████▌ | 266/312 [26:52<04:39,  6.07s/it]
Training Epoch: 17/100, batch 531/625 completed (loss: 0.0163:  85%|████████▌ | 266/312 [26:52<04:38,  6.06s/it]
Training Epoch: 17/100, batch 531/625 completed (loss: 0.0097:  85%|████████▌ | 266/312 [26:52<04:38,  6.06s/it]
Training Epoch: 17/100, batch 531/625 completed (loss: 0.0149:  85%|████████▌ | 266/312 [26:52<04:39,  6.07s/it]
Training Epoch: 17/100, batch 532/625 completed (loss: 0.479:  85%|████████▌ | 266/312 [26:55<04:38,  6.06s/it] 
Training Epoch: 17/100, batch 532/625 completed (loss: 0.0115:  85%|████████▌ | 266/312 [26:55<04:38,  6.06s/it]
Training Epoch: 17/100, batch 532/625 completed (loss: 0.007:  85%|████████▌ | 266/312 [26:55<04:39,  6.07s/it] 
Training Epoch: 17/100, batch 532/625 completed (loss: 0.0597:  85%|████████▌ | 266/312 [26:55<04:39,  6.07s/it]
Training Epoch: 17/100, batch 532/625 completed (loss: 0.479:  86%|████████▌ | 267/312 [26:58<04:32,  6.05s/it]
Training Epoch: 17/100, batch 532/625 completed (loss: 0.007:  86%|████████▌ | 267/312 [26:58<04:32,  6.05s/it]
Training Epoch: 17/100, batch 532/625 completed (loss: 0.0115:  86%|████████▌ | 267/312 [26:58<04:32,  6.05s/it]
Training Epoch: 17/100, batch 532/625 completed (loss: 0.0597:  86%|████████▌ | 267/312 [26:58<04:32,  6.05s/it]
Training Epoch: 17/100, batch 533/625 completed (loss: 0.0067:  86%|████████▌ | 267/312 [26:58<04:32,  6.05s/it]
Training Epoch: 17/100, batch 533/625 completed (loss: 0.0369:  86%|████████▌ | 267/312 [26:58<04:32,  6.05s/it]
Training Epoch: 17/100, batch 533/625 completed (loss: 0.036:  86%|████████▌ | 267/312 [26:58<04:32,  6.05s/it]
Training Epoch: 17/100, batch 533/625 completed (loss: 0.0809:  86%|████████▌ | 267/312 [26:58<04:32,  6.05s/it]
Training Epoch: 17/100, batch 534/625 completed (loss: 0.0237:  86%|████████▌ | 267/312 [27:01<04:32,  6.05s/it]
Training Epoch: 17/100, batch 534/625 completed (loss: 0.3293:  86%|████████▌ | 267/312 [27:01<04:32,  6.05s/it]
Training Epoch: 17/100, batch 534/625 completed (loss: 0.0178:  86%|████████▌ | 267/312 [27:01<04:32,  6.05s/it]
Training Epoch: 17/100, batch 534/625 completed (loss: 0.0764:  86%|████████▌ | 267/312 [27:01<04:32,  6.05s/it]
Training Epoch: 17/100, batch 534/625 completed (loss: 0.3293:  86%|████████▌ | 268/312 [27:05<04:29,  6.13s/it]
Training Epoch: 17/100, batch 534/625 completed (loss: 0.0237:  86%|████████▌ | 268/312 [27:05<04:29,  6.13s/it]
Training Epoch: 17/100, batch 534/625 completed (loss: 0.0178:  86%|████████▌ | 268/312 [27:05<04:29,  6.13s/it]
Training Epoch: 17/100, batch 534/625 completed (loss: 0.0764:  86%|████████▌ | 268/312 [27:05<04:29,  6.13s/it]
Training Epoch: 17/100, batch 535/625 completed (loss: 0.015:  86%|████████▌ | 268/312 [27:05<04:29,  6.13s/it] 
Training Epoch: 17/100, batch 535/625 completed (loss: 0.0991:  86%|████████▌ | 268/312 [27:05<04:29,  6.13s/it]
Training Epoch: 17/100, batch 535/625 completed (loss: 0.0065:  86%|████████▌ | 268/312 [27:05<04:29,  6.13s/it]
Training Epoch: 17/100, batch 535/625 completed (loss: 0.0107:  86%|████████▌ | 268/312 [27:05<04:29,  6.13s/it]
Training Epoch: 17/100, batch 536/625 completed (loss: 0.2409:  86%|████████▌ | 268/312 [27:08<04:29,  6.13s/it]
Training Epoch: 17/100, batch 536/625 completed (loss: 0.1137:  86%|████████▌ | 268/312 [27:08<04:29,  6.13s/it]
Training Epoch: 17/100, batch 536/625 completed (loss: 0.1028:  86%|████████▌ | 268/312 [27:08<04:29,  6.13s/it]
Training Epoch: 17/100, batch 536/625 completed (loss: 0.0111:  86%|████████▌ | 268/312 [27:08<04:29,  6.13s/it]
Training Epoch: 17/100, batch 536/625 completed (loss: 0.1137:  86%|████████▌ | 269/312 [27:11<04:22,  6.10s/it]
Training Epoch: 17/100, batch 536/625 completed (loss: 0.2409:  86%|████████▌ | 269/312 [27:11<04:22,  6.10s/it]
Training Epoch: 17/100, batch 536/625 completed (loss: 0.0111:  86%|████████▌ | 269/312 [27:11<04:22,  6.09s/it]
Training Epoch: 17/100, batch 536/625 completed (loss: 0.1028:  86%|████████▌ | 269/312 [27:11<04:22,  6.10s/it]
Training Epoch: 17/100, batch 537/625 completed (loss: 0.0138:  86%|████████▌ | 269/312 [27:11<04:22,  6.10s/it]
Training Epoch: 17/100, batch 537/625 completed (loss: 0.0075:  86%|████████▌ | 269/312 [27:11<04:22,  6.10s/it]
Training Epoch: 17/100, batch 537/625 completed (loss: 0.0156:  86%|████████▌ | 269/312 [27:11<04:22,  6.10s/it]
Training Epoch: 17/100, batch 537/625 completed (loss: 0.0544:  86%|████████▌ | 269/312 [27:11<04:22,  6.09s/it]
Training Epoch: 17/100, batch 538/625 completed (loss: 0.07:  86%|████████▌ | 269/312 [27:14<04:22,  6.10s/it]  
Training Epoch: 17/100, batch 538/625 completed (loss: 0.0662:  86%|████████▌ | 269/312 [27:14<04:22,  6.10s/it]
Training Epoch: 17/100, batch 538/625 completed (loss: 0.0213:  86%|████████▌ | 269/312 [27:14<04:22,  6.10s/it]
Training Epoch: 17/100, batch 538/625 completed (loss: 0.3658:  86%|████████▌ | 269/312 [27:14<04:22,  6.09s/it]
Training Epoch: 17/100, batch 538/625 completed (loss: 0.0662:  87%|████████▋ | 270/312 [27:17<04:15,  6.07s/it]
Training Epoch: 17/100, batch 538/625 completed (loss: 0.3658:  87%|████████▋ | 270/312 [27:17<04:14,  6.07s/it]
Training Epoch: 17/100, batch 538/625 completed (loss: 0.0213:  87%|████████▋ | 270/312 [27:17<04:15,  6.08s/it]
Training Epoch: 17/100, batch 538/625 completed (loss: 0.07:  87%|████████▋ | 270/312 [27:17<04:15,  6.08s/it]
Training Epoch: 17/100, batch 539/625 completed (loss: 0.0156:  87%|████████▋ | 270/312 [27:17<04:15,  6.07s/it]
Training Epoch: 17/100, batch 539/625 completed (loss: 0.1992:  87%|████████▋ | 270/312 [27:17<04:15,  6.08s/it]
Training Epoch: 17/100, batch 539/625 completed (loss: 0.3648:  87%|████████▋ | 270/312 [27:17<04:15,  6.08s/it]
Training Epoch: 17/100, batch 539/625 completed (loss: 0.0202:  87%|████████▋ | 270/312 [27:17<04:14,  6.07s/it]
Training Epoch: 17/100, batch 540/625 completed (loss: 0.0348:  87%|████████▋ | 270/312 [27:20<04:15,  6.07s/it]
Training Epoch: 17/100, batch 540/625 completed (loss: 0.0676:  87%|████████▋ | 270/312 [27:20<04:15,  6.08s/it]
Training Epoch: 17/100, batch 540/625 completed (loss: 0.1735:  87%|████████▋ | 270/312 [27:20<04:15,  6.08s/it]
Training Epoch: 17/100, batch 540/625 completed (loss: 0.0161:  87%|████████▋ | 270/312 [27:20<04:14,  6.07s/it]
Training Epoch: 17/100, batch 540/625 completed (loss: 0.0348:  87%|████████▋ | 271/312 [27:23<04:10,  6.10s/it]
Training Epoch: 17/100, batch 540/625 completed (loss: 0.0161:  87%|████████▋ | 271/312 [27:23<04:10,  6.10s/it]
Training Epoch: 17/100, batch 540/625 completed (loss: 0.1735:  87%|████████▋ | 271/312 [27:23<04:10,  6.10s/it]
Training Epoch: 17/100, batch 540/625 completed (loss: 0.0676:  87%|████████▋ | 271/312 [27:23<04:10,  6.10s/it]
Training Epoch: 17/100, batch 541/625 completed (loss: 0.0375:  87%|████████▋ | 271/312 [27:23<04:10,  6.10s/it]
Training Epoch: 17/100, batch 541/625 completed (loss: 0.0255:  87%|████████▋ | 271/312 [27:23<04:10,  6.10s/it]
Training Epoch: 17/100, batch 541/625 completed (loss: 0.2174:  87%|████████▋ | 271/312 [27:23<04:10,  6.10s/it]
Training Epoch: 17/100, batch 541/625 completed (loss: 0.1066:  87%|████████▋ | 271/312 [27:23<04:10,  6.10s/it]
Training Epoch: 17/100, batch 542/625 completed (loss: 0.0034:  87%|████████▋ | 271/312 [27:26<04:10,  6.10s/it]
Training Epoch: 17/100, batch 542/625 completed (loss: 0.0211:  87%|████████▋ | 271/312 [27:26<04:10,  6.10s/it]
Training Epoch: 17/100, batch 542/625 completed (loss: 0.2917:  87%|████████▋ | 271/312 [27:26<04:10,  6.10s/it]
Training Epoch: 17/100, batch 542/625 completed (loss: 0.015:  87%|████████▋ | 271/312 [27:26<04:10,  6.10s/it] 
Training Epoch: 17/100, batch 542/625 completed (loss: 0.0034:  87%|████████▋ | 272/312 [27:29<04:02,  6.07s/it]
Training Epoch: 17/100, batch 542/625 completed (loss: 0.0211:  87%|████████▋ | 272/312 [27:29<04:02,  6.07s/it]
Training Epoch: 17/100, batch 542/625 completed (loss: 0.2917:  87%|████████▋ | 272/312 [27:29<04:02,  6.07s/it]
Training Epoch: 17/100, batch 542/625 completed (loss: 0.015:  87%|████████▋ | 272/312 [27:29<04:02,  6.07s/it]
Training Epoch: 17/100, batch 543/625 completed (loss: 0.122:  87%|████████▋ | 272/312 [27:29<04:02,  6.07s/it] 
Training Epoch: 17/100, batch 543/625 completed (loss: 0.0469:  87%|████████▋ | 272/312 [27:29<04:02,  6.07s/it]
Training Epoch: 17/100, batch 543/625 completed (loss: 0.0097:  87%|████████▋ | 272/312 [27:29<04:02,  6.07s/it]
Training Epoch: 17/100, batch 543/625 completed (loss: 0.0998:  87%|████████▋ | 272/312 [27:29<04:02,  6.07s/it]
Training Epoch: 17/100, batch 544/625 completed (loss: 0.0082:  87%|████████▋ | 272/312 [27:32<04:02,  6.07s/it]
Training Epoch: 17/100, batch 544/625 completed (loss: 0.1077:  87%|████████▋ | 272/312 [27:32<04:02,  6.07s/it]
Training Epoch: 17/100, batch 544/625 completed (loss: 0.0729:  87%|████████▋ | 272/312 [27:32<04:02,  6.07s/it]
Training Epoch: 17/100, batch 544/625 completed (loss: 0.0064:  87%|████████▋ | 272/312 [27:32<04:02,  6.07s/it]
Training Epoch: 17/100, batch 544/625 completed (loss: 0.1077:  88%|████████▊ | 273/312 [27:35<03:57,  6.09s/it]
Training Epoch: 17/100, batch 544/625 completed (loss: 0.0082:  88%|████████▊ | 273/312 [27:35<03:57,  6.09s/it]
Training Epoch: 17/100, batch 544/625 completed (loss: 0.0064:  88%|████████▊ | 273/312 [27:35<03:57,  6.09s/it]
Training Epoch: 17/100, batch 544/625 completed (loss: 0.0729:  88%|████████▊ | 273/312 [27:35<03:57,  6.09s/it]
Training Epoch: 17/100, batch 545/625 completed (loss: 0.0169:  88%|████████▊ | 273/312 [27:35<03:57,  6.09s/it]
Training Epoch: 17/100, batch 545/625 completed (loss: 0.0202:  88%|████████▊ | 273/312 [27:35<03:57,  6.09s/it]
Training Epoch: 17/100, batch 545/625 completed (loss: 0.1228:  88%|████████▊ | 273/312 [27:35<03:57,  6.09s/it]
Training Epoch: 17/100, batch 545/625 completed (loss: 0.0225:  88%|████████▊ | 273/312 [27:35<03:57,  6.09s/it]
Training Epoch: 17/100, batch 546/625 completed (loss: 0.1051:  88%|████████▊ | 273/312 [27:38<03:57,  6.09s/it]
Training Epoch: 17/100, batch 546/625 completed (loss: 0.0142:  88%|████████▊ | 273/312 [27:38<03:57,  6.09s/it]
Training Epoch: 17/100, batch 546/625 completed (loss: 0.1535:  88%|████████▊ | 273/312 [27:38<03:57,  6.09s/it]
Training Epoch: 17/100, batch 546/625 completed (loss: 0.0038:  88%|████████▊ | 273/312 [27:38<03:57,  6.09s/it]
Training Epoch: 17/100, batch 546/625 completed (loss: 0.1535:  88%|████████▊ | 274/312 [27:41<03:50,  6.06s/it]
Training Epoch: 17/100, batch 546/625 completed (loss: 0.0038:  88%|████████▊ | 274/312 [27:41<03:50,  6.06s/it]
Training Epoch: 17/100, batch 546/625 completed (loss: 0.1051:  88%|████████▊ | 274/312 [27:41<03:50,  6.06s/it]
Training Epoch: 17/100, batch 546/625 completed (loss: 0.0142:  88%|████████▊ | 274/312 [27:41<03:50,  6.06s/it]
Training Epoch: 17/100, batch 547/625 completed (loss: 0.0396:  88%|████████▊ | 274/312 [27:41<03:50,  6.06s/it]
Training Epoch: 17/100, batch 547/625 completed (loss: 0.0197:  88%|████████▊ | 274/312 [27:41<03:50,  6.06s/it]
Training Epoch: 17/100, batch 547/625 completed (loss: 0.2484:  88%|████████▊ | 274/312 [27:41<03:50,  6.06s/it]
Training Epoch: 17/100, batch 547/625 completed (loss: 0.0094:  88%|████████▊ | 274/312 [27:41<03:50,  6.06s/it]
Training Epoch: 17/100, batch 548/625 completed (loss: 0.0099:  88%|████████▊ | 274/312 [27:44<03:50,  6.06s/it]
Training Epoch: 17/100, batch 548/625 completed (loss: 0.0103:  88%|████████▊ | 274/312 [27:44<03:50,  6.06s/it]
Training Epoch: 17/100, batch 548/625 completed (loss: 0.0196:  88%|████████▊ | 274/312 [27:44<03:50,  6.06s/it]
Training Epoch: 17/100, batch 548/625 completed (loss: 0.0307:  88%|████████▊ | 274/312 [27:44<03:50,  6.06s/it]
Training Epoch: 17/100, batch 548/625 completed (loss: 0.0307:  88%|████████▊ | 275/312 [27:47<03:45,  6.09s/it]
Training Epoch: 17/100, batch 548/625 completed (loss: 0.0103:  88%|████████▊ | 275/312 [27:47<03:45,  6.09s/it]
Training Epoch: 17/100, batch 548/625 completed (loss: 0.0099:  88%|████████▊ | 275/312 [27:47<03:45,  6.09s/it]
Training Epoch: 17/100, batch 548/625 completed (loss: 0.0196:  88%|████████▊ | 275/312 [27:47<03:45,  6.09s/it]
Training Epoch: 17/100, batch 549/625 completed (loss: 0.0457:  88%|████████▊ | 275/312 [27:47<03:45,  6.09s/it]
Training Epoch: 17/100, batch 549/625 completed (loss: 0.0411:  88%|████████▊ | 275/312 [27:47<03:45,  6.09s/it]
Training Epoch: 17/100, batch 549/625 completed (loss: 0.0466:  88%|████████▊ | 275/312 [27:47<03:45,  6.09s/it]
Training Epoch: 17/100, batch 549/625 completed (loss: 0.021:  88%|████████▊ | 275/312 [27:47<03:45,  6.09s/it] 
Training Epoch: 17/100, batch 550/625 completed (loss: 0.1152:  88%|████████▊ | 275/312 [27:50<03:45,  6.09s/it]
Training Epoch: 17/100, batch 550/625 completed (loss: 0.0402:  88%|████████▊ | 275/312 [27:50<03:45,  6.09s/it]
Training Epoch: 17/100, batch 550/625 completed (loss: 0.0083:  88%|████████▊ | 275/312 [27:50<03:45,  6.09s/it]
Training Epoch: 17/100, batch 550/625 completed (loss: 0.1468:  88%|████████▊ | 275/312 [27:50<03:45,  6.09s/it]
Training Epoch: 17/100, batch 550/625 completed (loss: 0.1152:  88%|████████▊ | 276/312 [27:53<03:40,  6.12s/it]
Training Epoch: 17/100, batch 550/625 completed (loss: 0.1468:  88%|████████▊ | 276/312 [27:53<03:40,  6.12s/it]
Training Epoch: 17/100, batch 550/625 completed (loss: 0.0402:  88%|████████▊ | 276/312 [27:53<03:40,  6.12s/it]
Training Epoch: 17/100, batch 550/625 completed (loss: 0.0083:  88%|████████▊ | 276/312 [27:53<03:40,  6.13s/it]
Training Epoch: 17/100, batch 551/625 completed (loss: 0.0637:  88%|████████▊ | 276/312 [27:53<03:40,  6.12s/it]
Training Epoch: 17/100, batch 551/625 completed (loss: 0.0125:  88%|████████▊ | 276/312 [27:53<03:40,  6.13s/it]
Training Epoch: 17/100, batch 551/625 completed (loss: 0.3264:  88%|████████▊ | 276/312 [27:53<03:40,  6.12s/it]
Training Epoch: 17/100, batch 551/625 completed (loss: 0.011:  88%|████████▊ | 276/312 [27:53<03:40,  6.12s/it] 
Training Epoch: 17/100, batch 552/625 completed (loss: 0.2051:  88%|████████▊ | 276/312 [27:56<03:40,  6.12s/it]
Training Epoch: 17/100, batch 552/625 completed (loss: 0.2198:  88%|████████▊ | 276/312 [27:56<03:40,  6.12s/it]
Training Epoch: 17/100, batch 552/625 completed (loss: 0.1129:  88%|████████▊ | 276/312 [27:56<03:40,  6.13s/it]
Training Epoch: 17/100, batch 552/625 completed (loss: 0.0167:  88%|████████▊ | 276/312 [27:56<03:40,  6.12s/it]
Training Epoch: 17/100, batch 552/625 completed (loss: 0.2198:  89%|████████▉ | 277/312 [27:59<03:32,  6.08s/it]
Training Epoch: 17/100, batch 552/625 completed (loss: 0.1129:  89%|████████▉ | 277/312 [27:59<03:32,  6.08s/it]
Training Epoch: 17/100, batch 552/625 completed (loss: 0.2051:  89%|████████▉ | 277/312 [27:59<03:32,  6.08s/it]
Training Epoch: 17/100, batch 552/625 completed (loss: 0.0167:  89%|████████▉ | 277/312 [27:59<03:33,  6.09s/it]
Training Epoch: 17/100, batch 553/625 completed (loss: 0.0391:  89%|████████▉ | 277/312 [27:59<03:32,  6.08s/it]
Training Epoch: 17/100, batch 553/625 completed (loss: 0.1686:  89%|████████▉ | 277/312 [27:59<03:32,  6.08s/it]
Training Epoch: 17/100, batch 553/625 completed (loss: 0.1198:  89%|████████▉ | 277/312 [27:59<03:32,  6.08s/it]
Training Epoch: 17/100, batch 553/625 completed (loss: 0.0195:  89%|████████▉ | 277/312 [27:59<03:33,  6.09s/it]
Training Epoch: 17/100, batch 554/625 completed (loss: 0.0048:  89%|████████▉ | 277/312 [28:02<03:32,  6.08s/it]
Training Epoch: 17/100, batch 554/625 completed (loss: 0.0238:  89%|████████▉ | 277/312 [28:02<03:32,  6.08s/it]
Training Epoch: 17/100, batch 554/625 completed (loss: 0.0184:  89%|████████▉ | 277/312 [28:02<03:32,  6.08s/it]
Training Epoch: 17/100, batch 554/625 completed (loss: 0.004:  89%|████████▉ | 277/312 [28:02<03:33,  6.09s/it] 
Training Epoch: 17/100, batch 554/625 completed (loss: 0.0184:  89%|████████▉ | 278/312 [28:05<03:27,  6.11s/it]
Training Epoch: 17/100, batch 554/625 completed (loss: 0.004:  89%|████████▉ | 278/312 [28:05<03:27,  6.11s/it]
Training Epoch: 17/100, batch 554/625 completed (loss: 0.0238:  89%|████████▉ | 278/312 [28:05<03:27,  6.11s/it]
Training Epoch: 17/100, batch 554/625 completed (loss: 0.0048:  89%|████████▉ | 278/312 [28:05<03:27,  6.11s/it]
Training Epoch: 17/100, batch 555/625 completed (loss: 0.0148:  89%|████████▉ | 278/312 [28:05<03:27,  6.11s/it]
Training Epoch: 17/100, batch 555/625 completed (loss: 0.0213:  89%|████████▉ | 278/312 [28:05<03:27,  6.11s/it]
Training Epoch: 17/100, batch 555/625 completed (loss: 0.0487:  89%|████████▉ | 278/312 [28:05<03:27,  6.11s/it]
Training Epoch: 17/100, batch 555/625 completed (loss: 0.2149:  89%|████████▉ | 278/312 [28:06<03:27,  6.11s/it]
Training Epoch: 17/100, batch 556/625 completed (loss: 0.0428:  89%|████████▉ | 278/312 [28:09<03:27,  6.11s/it]
Training Epoch: 17/100, batch 556/625 completed (loss: 0.011:  89%|████████▉ | 278/312 [28:09<03:27,  6.11s/it] 
Training Epoch: 17/100, batch 556/625 completed (loss: 0.0055:  89%|████████▉ | 278/312 [28:09<03:27,  6.11s/it]
Training Epoch: 17/100, batch 556/625 completed (loss: 0.0402:  89%|████████▉ | 278/312 [28:09<03:27,  6.11s/it]
Training Epoch: 17/100, batch 556/625 completed (loss: 0.0055:  89%|████████▉ | 279/312 [28:12<03:22,  6.14s/it]
Training Epoch: 17/100, batch 556/625 completed (loss: 0.0428:  89%|████████▉ | 279/312 [28:12<03:22,  6.14s/it]
Training Epoch: 17/100, batch 556/625 completed (loss: 0.011:  89%|████████▉ | 279/312 [28:12<03:22,  6.14s/it]
Training Epoch: 17/100, batch 556/625 completed (loss: 0.0402:  89%|████████▉ | 279/312 [28:12<03:22,  6.14s/it]
Training Epoch: 17/100, batch 557/625 completed (loss: 0.0153:  89%|████████▉ | 279/312 [28:12<03:22,  6.14s/it]
Training Epoch: 17/100, batch 557/625 completed (loss: 0.0753:  89%|████████▉ | 279/312 [28:12<03:22,  6.14s/it]
Training Epoch: 17/100, batch 557/625 completed (loss: 0.0367:  89%|████████▉ | 279/312 [28:12<03:22,  6.14s/it]
Training Epoch: 17/100, batch 557/625 completed (loss: 0.0037:  89%|████████▉ | 279/312 [28:12<03:22,  6.14s/it]
Training Epoch: 17/100, batch 558/625 completed (loss: 0.0561:  89%|████████▉ | 279/312 [28:15<03:22,  6.14s/it]
Training Epoch: 17/100, batch 558/625 completed (loss: 0.006:  89%|████████▉ | 279/312 [28:15<03:22,  6.14s/it] 
Training Epoch: 17/100, batch 558/625 completed (loss: 0.0065:  89%|████████▉ | 279/312 [28:15<03:22,  6.14s/it]
Training Epoch: 17/100, batch 558/625 completed (loss: 0.0335:  89%|████████▉ | 279/312 [28:15<03:22,  6.14s/it]
Training Epoch: 17/100, batch 558/625 completed (loss: 0.006:  90%|████████▉ | 280/312 [28:17<03:13,  6.05s/it]
Training Epoch: 17/100, batch 558/625 completed (loss: 0.0065:  90%|████████▉ | 280/312 [28:17<03:13,  6.05s/it]
Training Epoch: 17/100, batch 558/625 completed (loss: 0.0561:  90%|████████▉ | 280/312 [28:17<03:13,  6.05s/it]
Training Epoch: 17/100, batch 558/625 completed (loss: 0.0335:  90%|████████▉ | 280/312 [28:18<03:13,  6.05s/it]
Training Epoch: 17/100, batch 559/625 completed (loss: 0.0854:  90%|████████▉ | 280/312 [28:18<03:13,  6.05s/it]
Training Epoch: 17/100, batch 559/625 completed (loss: 0.5201:  90%|████████▉ | 280/312 [28:18<03:13,  6.05s/it]
Training Epoch: 17/100, batch 559/625 completed (loss: 0.0091:  90%|████████▉ | 280/312 [28:17<03:13,  6.05s/it]
Training Epoch: 17/100, batch 559/625 completed (loss: 0.018:  90%|████████▉ | 280/312 [28:18<03:13,  6.05s/it] 
Training Epoch: 17/100, batch 560/625 completed (loss: 0.0253:  90%|████████▉ | 280/312 [28:21<03:13,  6.05s/it]
Training Epoch: 17/100, batch 560/625 completed (loss: 0.0238:  90%|████████▉ | 280/312 [28:21<03:13,  6.05s/it]
Training Epoch: 17/100, batch 560/625 completed (loss: 0.025:  90%|████████▉ | 280/312 [28:21<03:13,  6.05s/it] 
Training Epoch: 17/100, batch 560/625 completed (loss: 0.2085:  90%|████████▉ | 280/312 [28:21<03:13,  6.05s/it]
Training Epoch: 17/100, batch 560/625 completed (loss: 0.0238:  90%|█████████ | 281/312 [28:23<03:07,  6.05s/it]
Training Epoch: 17/100, batch 560/625 completed (loss: 0.2085:  90%|█████████ | 281/312 [28:24<03:07,  6.05s/it]
Training Epoch: 17/100, batch 560/625 completed (loss: 0.025:  90%|█████████ | 281/312 [28:24<03:07,  6.05s/it]
Training Epoch: 17/100, batch 560/625 completed (loss: 0.0253:  90%|█████████ | 281/312 [28:24<03:07,  6.05s/it]
Training Epoch: 17/100, batch 561/625 completed (loss: 0.0148:  90%|█████████ | 281/312 [28:24<03:07,  6.05s/it]
Training Epoch: 17/100, batch 561/625 completed (loss: 0.2079:  90%|█████████ | 281/312 [28:24<03:07,  6.05s/it]
Training Epoch: 17/100, batch 561/625 completed (loss: 0.0061:  90%|█████████ | 281/312 [28:24<03:07,  6.05s/it]
Training Epoch: 17/100, batch 561/625 completed (loss: 0.01:  90%|█████████ | 281/312 [28:24<03:07,  6.05s/it]  
Training Epoch: 17/100, batch 562/625 completed (loss: 0.3582:  90%|█████████ | 281/312 [28:27<03:07,  6.05s/it]
Training Epoch: 17/100, batch 562/625 completed (loss: 0.0201:  90%|█████████ | 281/312 [28:27<03:07,  6.05s/it]
Training Epoch: 17/100, batch 562/625 completed (loss: 0.3185:  90%|█████████ | 281/312 [28:27<03:07,  6.05s/it]
Training Epoch: 17/100, batch 562/625 completed (loss: 0.1192:  90%|█████████ | 281/312 [28:27<03:07,  6.05s/it]
Training Epoch: 17/100, batch 562/625 completed (loss: 0.3185:  90%|█████████ | 282/312 [28:30<03:02,  6.09s/it]
Training Epoch: 17/100, batch 562/625 completed (loss: 0.0201:  90%|█████████ | 282/312 [28:30<03:02,  6.09s/it]
Training Epoch: 17/100, batch 562/625 completed (loss: 0.3582:  90%|█████████ | 282/312 [28:30<03:02,  6.09s/it]
Training Epoch: 17/100, batch 562/625 completed (loss: 0.1192:  90%|█████████ | 282/312 [28:30<03:02,  6.09s/it]
Training Epoch: 17/100, batch 563/625 completed (loss: 0.0119:  90%|█████████ | 282/312 [28:30<03:02,  6.09s/it]
Training Epoch: 17/100, batch 563/625 completed (loss: 0.0344:  90%|█████████ | 282/312 [28:30<03:02,  6.09s/it]
Training Epoch: 17/100, batch 563/625 completed (loss: 0.011:  90%|█████████ | 282/312 [28:30<03:02,  6.09s/it] 
Training Epoch: 17/100, batch 563/625 completed (loss: 0.0415:  90%|█████████ | 282/312 [28:30<03:02,  6.09s/it]
Training Epoch: 17/100, batch 564/625 completed (loss: 0.0073:  90%|█████████ | 282/312 [28:33<03:02,  6.09s/it]
Training Epoch: 17/100, batch 564/625 completed (loss: 0.3234:  90%|█████████ | 282/312 [28:33<03:02,  6.09s/it]
Training Epoch: 17/100, batch 564/625 completed (loss: 0.0354:  90%|█████████ | 282/312 [28:33<03:02,  6.09s/it]
Training Epoch: 17/100, batch 564/625 completed (loss: 0.0322:  90%|█████████ | 282/312 [28:33<03:02,  6.09s/it]
Training Epoch: 17/100, batch 564/625 completed (loss: 0.3234:  91%|█████████ | 283/312 [28:36<02:56,  6.08s/it]
Training Epoch: 17/100, batch 564/625 completed (loss: 0.0322:  91%|█████████ | 283/312 [28:36<02:56,  6.08s/it]
Training Epoch: 17/100, batch 564/625 completed (loss: 0.0073:  91%|█████████ | 283/312 [28:36<02:56,  6.08s/it]
Training Epoch: 17/100, batch 564/625 completed (loss: 0.0354:  91%|█████████ | 283/312 [28:36<02:56,  6.08s/it]
Training Epoch: 17/100, batch 565/625 completed (loss: 0.0449:  91%|█████████ | 283/312 [28:36<02:56,  6.08s/it]
Training Epoch: 17/100, batch 565/625 completed (loss: 0.1021:  91%|█████████ | 283/312 [28:36<02:56,  6.08s/it]
Training Epoch: 17/100, batch 565/625 completed (loss: 0.175:  91%|█████████ | 283/312 [28:36<02:56,  6.08s/it] 
Training Epoch: 17/100, batch 565/625 completed (loss: 0.0532:  91%|█████████ | 283/312 [28:36<02:56,  6.08s/it]
Training Epoch: 17/100, batch 566/625 completed (loss: 0.1437:  91%|█████████ | 283/312 [28:39<02:56,  6.08s/it]
Training Epoch: 17/100, batch 566/625 completed (loss: 0.0064:  91%|█████████ | 283/312 [28:39<02:56,  6.08s/it]
Training Epoch: 17/100, batch 566/625 completed (loss: 0.0931:  91%|█████████ | 283/312 [28:39<02:56,  6.08s/it]
Training Epoch: 17/100, batch 566/625 completed (loss: 0.0633:  91%|█████████ | 283/312 [28:39<02:56,  6.08s/it]
Training Epoch: 17/100, batch 566/625 completed (loss: 0.0064:  91%|█████████ | 284/312 [28:42<02:48,  6.01s/it]
Training Epoch: 17/100, batch 566/625 completed (loss: 0.0633:  91%|█████████ | 284/312 [28:42<02:48,  6.01s/it]
Training Epoch: 17/100, batch 566/625 completed (loss: 0.1437:  91%|█████████ | 284/312 [28:42<02:48,  6.01s/it]
Training Epoch: 17/100, batch 566/625 completed (loss: 0.0931:  91%|█████████ | 284/312 [28:42<02:48,  6.02s/it]
Training Epoch: 17/100, batch 567/625 completed (loss: 0.0284:  91%|█████████ | 284/312 [28:42<02:48,  6.01s/it]
Training Epoch: 17/100, batch 567/625 completed (loss: 0.0471:  91%|█████████ | 284/312 [28:42<02:48,  6.02s/it]
Training Epoch: 17/100, batch 567/625 completed (loss: 0.3098:  91%|█████████ | 284/312 [28:42<02:48,  6.01s/it]
Training Epoch: 17/100, batch 567/625 completed (loss: 0.1355:  91%|█████████ | 284/312 [28:42<02:48,  6.01s/it]
Training Epoch: 17/100, batch 568/625 completed (loss: 0.0085:  91%|█████████ | 284/312 [28:45<02:48,  6.01s/it]
Training Epoch: 17/100, batch 568/625 completed (loss: 0.0652:  91%|█████████ | 284/312 [28:45<02:48,  6.02s/it]
Training Epoch: 17/100, batch 568/625 completed (loss: 0.0037:  91%|█████████ | 284/312 [28:45<02:48,  6.01s/it]
Training Epoch: 17/100, batch 568/625 completed (loss: 0.4848:  91%|█████████ | 284/312 [28:45<02:48,  6.01s/it]
Training Epoch: 17/100, batch 568/625 completed (loss: 0.0085:  91%|█████████▏| 285/312 [28:48<02:43,  6.05s/it]
Training Epoch: 17/100, batch 568/625 completed (loss: 0.0652:  91%|█████████▏| 285/312 [28:48<02:43,  6.05s/it]
Training Epoch: 17/100, batch 568/625 completed (loss: 0.0037:  91%|█████████▏| 285/312 [28:48<02:43,  6.05s/it]
Training Epoch: 17/100, batch 568/625 completed (loss: 0.4848:  91%|█████████▏| 285/312 [28:48<02:43,  6.05s/it]
Training Epoch: 17/100, batch 569/625 completed (loss: 0.1202:  91%|█████████▏| 285/312 [28:48<02:43,  6.05s/it]
Training Epoch: 17/100, batch 569/625 completed (loss: 0.3725:  91%|█████████▏| 285/312 [28:48<02:43,  6.05s/it]
Training Epoch: 17/100, batch 569/625 completed (loss: 0.6613:  91%|█████████▏| 285/312 [28:48<02:43,  6.05s/it]
Training Epoch: 17/100, batch 569/625 completed (loss: 0.0249:  91%|█████████▏| 285/312 [28:48<02:43,  6.05s/it]
Training Epoch: 17/100, batch 570/625 completed (loss: 0.0148:  91%|█████████▏| 285/312 [28:51<02:43,  6.05s/it]
Training Epoch: 17/100, batch 570/625 completed (loss: 0.019:  91%|█████████▏| 285/312 [28:51<02:43,  6.05s/it] 
Training Epoch: 17/100, batch 570/625 completed (loss: 0.0184:  91%|█████████▏| 285/312 [28:51<02:43,  6.05s/it]
Training Epoch: 17/100, batch 570/625 completed (loss: 0.0187:  91%|█████████▏| 285/312 [28:51<02:43,  6.05s/it]
Training Epoch: 17/100, batch 570/625 completed (loss: 0.0148:  92%|█████████▏| 286/312 [28:54<02:38,  6.08s/it]
Training Epoch: 17/100, batch 570/625 completed (loss: 0.0184:  92%|█████████▏| 286/312 [28:54<02:38,  6.08s/it]
Training Epoch: 17/100, batch 570/625 completed (loss: 0.0187:  92%|█████████▏| 286/312 [28:54<02:38,  6.08s/it]
Training Epoch: 17/100, batch 570/625 completed (loss: 0.019:  92%|█████████▏| 286/312 [28:54<02:38,  6.08s/it]
Training Epoch: 17/100, batch 571/625 completed (loss: 0.0025:  92%|█████████▏| 286/312 [28:54<02:38,  6.08s/it]
Training Epoch: 17/100, batch 571/625 completed (loss: 0.5281:  92%|█████████▏| 286/312 [28:54<02:38,  6.08s/it]
Training Epoch: 17/100, batch 571/625 completed (loss: 0.0052:  92%|█████████▏| 286/312 [28:54<02:38,  6.08s/it]
Training Epoch: 17/100, batch 571/625 completed (loss: 0.0118:  92%|█████████▏| 286/312 [28:54<02:38,  6.08s/it]
Training Epoch: 17/100, batch 572/625 completed (loss: 0.0492:  92%|█████████▏| 286/312 [28:57<02:38,  6.08s/it]
Training Epoch: 17/100, batch 572/625 completed (loss: 0.013:  92%|█████████▏| 286/312 [28:57<02:38,  6.08s/it] 
Training Epoch: 17/100, batch 572/625 completed (loss: 0.2938:  92%|█████████▏| 286/312 [28:57<02:38,  6.08s/it]
Training Epoch: 17/100, batch 572/625 completed (loss: 0.0714:  92%|█████████▏| 286/312 [28:57<02:38,  6.08s/it]
Training Epoch: 17/100, batch 572/625 completed (loss: 0.0714:  92%|█████████▏| 287/312 [29:00<02:32,  6.11s/it]
Training Epoch: 17/100, batch 572/625 completed (loss: 0.013:  92%|█████████▏| 287/312 [29:00<02:32,  6.11s/it]
Training Epoch: 17/100, batch 572/625 completed (loss: 0.0492:  92%|█████████▏| 287/312 [29:00<02:32,  6.11s/it]
Training Epoch: 17/100, batch 572/625 completed (loss: 0.2938:  92%|█████████▏| 287/312 [29:00<02:32,  6.12s/it]
Training Epoch: 17/100, batch 573/625 completed (loss: 0.0709:  92%|█████████▏| 287/312 [29:00<02:32,  6.11s/it]
Training Epoch: 17/100, batch 573/625 completed (loss: 0.0723:  92%|█████████▏| 287/312 [29:00<02:32,  6.12s/it]
Training Epoch: 17/100, batch 573/625 completed (loss: 0.0123:  92%|█████████▏| 287/312 [29:00<02:32,  6.11s/it]
Training Epoch: 17/100, batch 573/625 completed (loss: 0.0496:  92%|█████████▏| 287/312 [29:00<02:32,  6.11s/it]
Training Epoch: 17/100, batch 574/625 completed (loss: 0.2035:  92%|█████████▏| 287/312 [29:03<02:32,  6.12s/it]
Training Epoch: 17/100, batch 574/625 completed (loss: 0.0058:  92%|█████████▏| 287/312 [29:03<02:32,  6.11s/it]
Training Epoch: 17/100, batch 574/625 completed (loss: 0.0072:  92%|█████████▏| 287/312 [29:03<02:32,  6.11s/it]
Training Epoch: 17/100, batch 574/625 completed (loss: 0.0054:  92%|█████████▏| 287/312 [29:03<02:32,  6.11s/it]
Training Epoch: 17/100, batch 574/625 completed (loss: 0.0072:  92%|█████████▏| 288/312 [29:06<02:25,  6.07s/it]
Training Epoch: 17/100, batch 574/625 completed (loss: 0.2035:  92%|█████████▏| 288/312 [29:06<02:25,  6.07s/it]
Training Epoch: 17/100, batch 574/625 completed (loss: 0.0054:  92%|█████████▏| 288/312 [29:06<02:25,  6.07s/it]
Training Epoch: 17/100, batch 574/625 completed (loss: 0.0058:  92%|█████████▏| 288/312 [29:06<02:25,  6.07s/it]
Training Epoch: 17/100, batch 575/625 completed (loss: 0.0255:  92%|█████████▏| 288/312 [29:06<02:25,  6.07s/it]
Training Epoch: 17/100, batch 575/625 completed (loss: 0.0128:  92%|█████████▏| 288/312 [29:06<02:25,  6.07s/it]
Training Epoch: 17/100, batch 575/625 completed (loss: 0.0088:  92%|█████████▏| 288/312 [29:06<02:25,  6.07s/it]
Training Epoch: 17/100, batch 575/625 completed (loss: 0.0132:  92%|█████████▏| 288/312 [29:06<02:25,  6.07s/it]
Training Epoch: 17/100, batch 576/625 completed (loss: 0.3293:  92%|█████████▏| 288/312 [29:09<02:25,  6.07s/it]
Training Epoch: 17/100, batch 576/625 completed (loss: 0.0078:  92%|█████████▏| 288/312 [29:09<02:25,  6.07s/it]
Training Epoch: 17/100, batch 576/625 completed (loss: 0.0351:  92%|█████████▏| 288/312 [29:09<02:25,  6.07s/it]
Training Epoch: 17/100, batch 576/625 completed (loss: 0.5303:  92%|█████████▏| 288/312 [29:09<02:25,  6.07s/it]
Training Epoch: 17/100, batch 576/625 completed (loss: 0.0078:  93%|█████████▎| 289/312 [29:12<02:19,  6.06s/it]
Training Epoch: 17/100, batch 576/625 completed (loss: 0.3293:  93%|█████████▎| 289/312 [29:12<02:19,  6.06s/it]
Training Epoch: 17/100, batch 576/625 completed (loss: 0.5303:  93%|█████████▎| 289/312 [29:12<02:19,  6.07s/it]
Training Epoch: 17/100, batch 576/625 completed (loss: 0.0351:  93%|█████████▎| 289/312 [29:12<02:19,  6.07s/it]
Training Epoch: 17/100, batch 577/625 completed (loss: 0.4543:  93%|█████████▎| 289/312 [29:12<02:19,  6.06s/it]
Training Epoch: 17/100, batch 577/625 completed (loss: 0.0226:  93%|█████████▎| 289/312 [29:12<02:19,  6.07s/it]
Training Epoch: 17/100, batch 577/625 completed (loss: 0.0064:  93%|█████████▎| 289/312 [29:12<02:19,  6.06s/it]
Training Epoch: 17/100, batch 577/625 completed (loss: 0.012:  93%|█████████▎| 289/312 [29:12<02:19,  6.07s/it] 
Training Epoch: 17/100, batch 578/625 completed (loss: 0.0441:  93%|█████████▎| 289/312 [29:15<02:19,  6.06s/it]
Training Epoch: 17/100, batch 578/625 completed (loss: 0.0092:  93%|█████████▎| 289/312 [29:15<02:19,  6.06s/it]
Training Epoch: 17/100, batch 578/625 completed (loss: 0.0141:  93%|█████████▎| 289/312 [29:15<02:19,  6.07s/it]
Training Epoch: 17/100, batch 578/625 completed (loss: 0.0391:  93%|█████████▎| 289/312 [29:15<02:19,  6.07s/it]
Training Epoch: 17/100, batch 578/625 completed (loss: 0.0141:  93%|█████████▎| 290/312 [29:18<02:14,  6.09s/it]
Training Epoch: 17/100, batch 578/625 completed (loss: 0.0092:  93%|█████████▎| 290/312 [29:18<02:14,  6.09s/it]
Training Epoch: 17/100, batch 578/625 completed (loss: 0.0441:  93%|█████████▎| 290/312 [29:18<02:14,  6.10s/it]
Training Epoch: 17/100, batch 578/625 completed (loss: 0.0391:  93%|█████████▎| 290/312 [29:18<02:14,  6.10s/it]
Training Epoch: 17/100, batch 579/625 completed (loss: 0.2426:  93%|█████████▎| 290/312 [29:18<02:14,  6.09s/it]
Training Epoch: 17/100, batch 579/625 completed (loss: 0.0169:  93%|█████████▎| 290/312 [29:18<02:14,  6.09s/it]
Training Epoch: 17/100, batch 579/625 completed (loss: 0.0795:  93%|█████████▎| 290/312 [29:18<02:14,  6.10s/it]
Training Epoch: 17/100, batch 579/625 completed (loss: 0.1322:  93%|█████████▎| 290/312 [29:18<02:14,  6.10s/it]
Training Epoch: 17/100, batch 580/625 completed (loss: 0.0359:  93%|█████████▎| 290/312 [29:21<02:14,  6.10s/it]
Training Epoch: 17/100, batch 580/625 completed (loss: 0.0294:  93%|█████████▎| 290/312 [29:21<02:14,  6.09s/it]
Training Epoch: 17/100, batch 580/625 completed (loss: 0.1423:  93%|█████████▎| 290/312 [29:21<02:14,  6.09s/it]
Training Epoch: 17/100, batch 580/625 completed (loss: 0.1227:  93%|█████████▎| 290/312 [29:21<02:14,  6.10s/it]
Training Epoch: 17/100, batch 580/625 completed (loss: 0.1423:  93%|█████████▎| 291/312 [29:24<02:07,  6.09s/it]
Training Epoch: 17/100, batch 580/625 completed (loss: 0.0359:  93%|█████████▎| 291/312 [29:24<02:07,  6.09s/it]
Training Epoch: 17/100, batch 580/625 completed (loss: 0.1227:  93%|█████████▎| 291/312 [29:24<02:07,  6.09s/it]
Training Epoch: 17/100, batch 580/625 completed (loss: 0.0294:  93%|█████████▎| 291/312 [29:24<02:08,  6.10s/it]
Training Epoch: 17/100, batch 581/625 completed (loss: 0.054:  93%|█████████▎| 291/312 [29:24<02:07,  6.09s/it] 
Training Epoch: 17/100, batch 581/625 completed (loss: 0.0654:  93%|█████████▎| 291/312 [29:24<02:07,  6.09s/it]
Training Epoch: 17/100, batch 581/625 completed (loss: 0.0389:  93%|█████████▎| 291/312 [29:24<02:08,  6.10s/it]
Training Epoch: 17/100, batch 581/625 completed (loss: 0.0582:  93%|█████████▎| 291/312 [29:24<02:07,  6.09s/it]
Training Epoch: 17/100, batch 582/625 completed (loss: 0.01:  93%|█████████▎| 291/312 [29:27<02:08,  6.10s/it]  
Training Epoch: 17/100, batch 582/625 completed (loss: 0.0119:  93%|█████████▎| 291/312 [29:27<02:07,  6.09s/it]
Training Epoch: 17/100, batch 582/625 completed (loss: 0.0099:  93%|█████████▎| 291/312 [29:27<02:07,  6.09s/it]
Training Epoch: 17/100, batch 582/625 completed (loss: 0.0089:  93%|█████████▎| 291/312 [29:27<02:07,  6.09s/it]
Training Epoch: 17/100, batch 582/625 completed (loss: 0.0099:  94%|█████████▎| 292/312 [29:30<02:00,  6.02s/it]
Training Epoch: 17/100, batch 582/625 completed (loss: 0.01:  94%|█████████▎| 292/312 [29:30<02:00,  6.02s/it]
Training Epoch: 17/100, batch 582/625 completed (loss: 0.0119:  94%|█████████▎| 292/312 [29:30<02:00,  6.03s/it]
Training Epoch: 17/100, batch 582/625 completed (loss: 0.0089:  94%|█████████▎| 292/312 [29:30<02:00,  6.03s/it]
Training Epoch: 17/100, batch 583/625 completed (loss: 0.0048:  94%|█████████▎| 292/312 [29:30<02:00,  6.02s/it]
Training Epoch: 17/100, batch 583/625 completed (loss: 0.009:  94%|█████████▎| 292/312 [29:30<02:00,  6.02s/it] 
Training Epoch: 17/100, batch 583/625 completed (loss: 0.0065:  94%|█████████▎| 292/312 [29:30<02:00,  6.03s/it]
Training Epoch: 17/100, batch 583/625 completed (loss: 0.0799:  94%|█████████▎| 292/312 [29:30<02:00,  6.03s/it]
Training Epoch: 17/100, batch 584/625 completed (loss: 0.0383:  94%|█████████▎| 292/312 [29:33<02:00,  6.03s/it]
Training Epoch: 17/100, batch 584/625 completed (loss: 0.0038:  94%|█████████▎| 292/312 [29:33<02:00,  6.02s/it]
Training Epoch: 17/100, batch 584/625 completed (loss: 0.0344:  94%|█████████▎| 292/312 [29:33<02:00,  6.02s/it]
Training Epoch: 17/100, batch 584/625 completed (loss: 0.0088:  94%|█████████▎| 292/312 [29:33<02:00,  6.03s/it]
Training Epoch: 17/100, batch 584/625 completed (loss: 0.0038:  94%|█████████▍| 293/312 [29:36<01:55,  6.09s/it]
Training Epoch: 17/100, batch 584/625 completed (loss: 0.0383:  94%|█████████▍| 293/312 [29:36<01:55,  6.09s/it]
Training Epoch: 17/100, batch 584/625 completed (loss: 0.0088:  94%|█████████▍| 293/312 [29:37<01:55,  6.09s/it]
Training Epoch: 17/100, batch 584/625 completed (loss: 0.0344:  94%|█████████▍| 293/312 [29:36<01:55,  6.10s/it]
Training Epoch: 17/100, batch 585/625 completed (loss: 0.145:  94%|█████████▍| 293/312 [29:37<01:55,  6.09s/it] 
Training Epoch: 17/100, batch 585/625 completed (loss: 0.0045:  94%|█████████▍| 293/312 [29:37<01:55,  6.10s/it]
Training Epoch: 17/100, batch 585/625 completed (loss: 0.0722:  94%|█████████▍| 293/312 [29:37<01:55,  6.09s/it]
Training Epoch: 17/100, batch 585/625 completed (loss: 0.1427:  94%|█████████▍| 293/312 [29:37<01:55,  6.09s/it]
Training Epoch: 17/100, batch 586/625 completed (loss: 0.051:  94%|█████████▍| 293/312 [29:40<01:55,  6.09s/it] 
Training Epoch: 17/100, batch 586/625 completed (loss: 0.0397:  94%|█████████▍| 293/312 [29:40<01:55,  6.09s/it]
Training Epoch: 17/100, batch 586/625 completed (loss: 0.2203:  94%|█████████▍| 293/312 [29:40<01:55,  6.10s/it]
Training Epoch: 17/100, batch 586/625 completed (loss: 0.0078:  94%|█████████▍| 293/312 [29:40<01:55,  6.09s/it]
Training Epoch: 17/100, batch 586/625 completed (loss: 0.0397:  94%|█████████▍| 294/312 [29:43<01:49,  6.08s/it]
Training Epoch: 17/100, batch 586/625 completed (loss: 0.051:  94%|█████████▍| 294/312 [29:43<01:49,  6.08s/it]
Training Epoch: 17/100, batch 586/625 completed (loss: 0.2203:  94%|█████████▍| 294/312 [29:43<01:49,  6.08s/it]
Training Epoch: 17/100, batch 586/625 completed (loss: 0.0078:  94%|█████████▍| 294/312 [29:43<01:49,  6.08s/it]
Training Epoch: 17/100, batch 587/625 completed (loss: 0.0455:  94%|█████████▍| 294/312 [29:43<01:49,  6.08s/it]
Training Epoch: 17/100, batch 587/625 completed (loss: 0.0388:  94%|█████████▍| 294/312 [29:43<01:49,  6.08s/it]
Training Epoch: 17/100, batch 587/625 completed (loss: 0.0509:  94%|█████████▍| 294/312 [29:43<01:49,  6.08s/it]
Training Epoch: 17/100, batch 587/625 completed (loss: 0.0675:  94%|█████████▍| 294/312 [29:43<01:49,  6.08s/it]
Training Epoch: 17/100, batch 588/625 completed (loss: 0.0395:  94%|█████████▍| 294/312 [29:46<01:49,  6.08s/it]
Training Epoch: 17/100, batch 588/625 completed (loss: 0.0925:  94%|█████████▍| 294/312 [29:45<01:49,  6.08s/it]
Training Epoch: 17/100, batch 588/625 completed (loss: 0.0017:  94%|█████████▍| 294/312 [29:46<01:49,  6.08s/it]
Training Epoch: 17/100, batch 588/625 completed (loss: 0.0383:  94%|█████████▍| 294/312 [29:46<01:49,  6.08s/it]
Training Epoch: 17/100, batch 588/625 completed (loss: 0.0395:  95%|█████████▍| 295/312 [29:49<01:43,  6.11s/it]
Training Epoch: 17/100, batch 588/625 completed (loss: 0.0925:  95%|█████████▍| 295/312 [29:49<01:43,  6.11s/it]
Training Epoch: 17/100, batch 588/625 completed (loss: 0.0017:  95%|█████████▍| 295/312 [29:49<01:43,  6.11s/it]
Training Epoch: 17/100, batch 588/625 completed (loss: 0.0383:  95%|█████████▍| 295/312 [29:49<01:43,  6.11s/it]
Training Epoch: 17/100, batch 589/625 completed (loss: 0.004:  95%|█████████▍| 295/312 [29:49<01:43,  6.11s/it] 
Training Epoch: 17/100, batch 589/625 completed (loss: 0.344:  95%|█████████▍| 295/312 [29:49<01:43,  6.11s/it] 
Training Epoch: 17/100, batch 589/625 completed (loss: 0.1004:  95%|█████████▍| 295/312 [29:49<01:43,  6.11s/it]
Training Epoch: 17/100, batch 589/625 completed (loss: 0.0017:  95%|█████████▍| 295/312 [29:49<01:43,  6.11s/it]
Training Epoch: 17/100, batch 590/625 completed (loss: 0.034:  95%|█████████▍| 295/312 [29:52<01:43,  6.11s/it]
Training Epoch: 17/100, batch 590/625 completed (loss: 0.0146:  95%|█████████▍| 295/312 [29:52<01:43,  6.11s/it]
Training Epoch: 17/100, batch 590/625 completed (loss: 0.0123:  95%|█████████▍| 295/312 [29:52<01:43,  6.11s/it]
Training Epoch: 17/100, batch 590/625 completed (loss: 0.0219:  95%|█████████▍| 295/312 [29:52<01:43,  6.11s/it]
Training Epoch: 17/100, batch 590/625 completed (loss: 0.0219:  95%|█████████▍| 296/312 [29:55<01:37,  6.12s/it]
Training Epoch: 17/100, batch 590/625 completed (loss: 0.0123:  95%|█████████▍| 296/312 [29:55<01:37,  6.12s/it]
Training Epoch: 17/100, batch 590/625 completed (loss: 0.034:  95%|█████████▍| 296/312 [29:55<01:37,  6.12s/it]
Training Epoch: 17/100, batch 590/625 completed (loss: 0.0146:  95%|█████████▍| 296/312 [29:55<01:37,  6.12s/it]
Training Epoch: 17/100, batch 591/625 completed (loss: 0.0091:  95%|█████████▍| 296/312 [29:55<01:37,  6.12s/it]
Training Epoch: 17/100, batch 591/625 completed (loss: 0.0206:  95%|█████████▍| 296/312 [29:55<01:37,  6.12s/it]
Training Epoch: 17/100, batch 591/625 completed (loss: 0.3148:  95%|█████████▍| 296/312 [29:55<01:37,  6.12s/it]
Training Epoch: 17/100, batch 591/625 completed (loss: 0.0048:  95%|█████████▍| 296/312 [29:55<01:37,  6.12s/it]
Training Epoch: 17/100, batch 592/625 completed (loss: 0.0342:  95%|█████████▍| 296/312 [29:58<01:37,  6.12s/it]
Training Epoch: 17/100, batch 592/625 completed (loss: 0.2809:  95%|█████████▍| 296/312 [29:58<01:37,  6.12s/it]
Training Epoch: 17/100, batch 592/625 completed (loss: 0.0813:  95%|█████████▍| 296/312 [29:58<01:37,  6.12s/it]
Training Epoch: 17/100, batch 592/625 completed (loss: 0.0326:  95%|█████████▍| 296/312 [29:58<01:37,  6.12s/it]
Training Epoch: 17/100, batch 592/625 completed (loss: 0.2809:  95%|█████████▌| 297/312 [30:01<01:30,  6.04s/it]
Training Epoch: 17/100, batch 592/625 completed (loss: 0.0326:  95%|█████████▌| 297/312 [30:01<01:30,  6.04s/it]
Training Epoch: 17/100, batch 592/625 completed (loss: 0.0342:  95%|█████████▌| 297/312 [30:01<01:30,  6.04s/it]
Training Epoch: 17/100, batch 592/625 completed (loss: 0.0813:  95%|█████████▌| 297/312 [30:01<01:30,  6.04s/it]
Training Epoch: 17/100, batch 593/625 completed (loss: 0.0204:  95%|█████████▌| 297/312 [30:01<01:30,  6.04s/it]
Training Epoch: 17/100, batch 593/625 completed (loss: 0.0212:  95%|█████████▌| 297/312 [30:01<01:30,  6.04s/it]
Training Epoch: 17/100, batch 593/625 completed (loss: 0.0197:  95%|█████████▌| 297/312 [30:01<01:30,  6.04s/it]
Training Epoch: 17/100, batch 593/625 completed (loss: 0.0237:  95%|█████████▌| 297/312 [30:01<01:30,  6.04s/it]
Training Epoch: 17/100, batch 594/625 completed (loss: 0.0279:  95%|█████████▌| 297/312 [30:04<01:30,  6.04s/it]
Training Epoch: 17/100, batch 594/625 completed (loss: 0.0171:  95%|█████████▌| 297/312 [30:04<01:30,  6.04s/it]
Training Epoch: 17/100, batch 594/625 completed (loss: 0.012:  95%|█████████▌| 297/312 [30:04<01:30,  6.04s/it] 
Training Epoch: 17/100, batch 594/625 completed (loss: 0.0023:  95%|█████████▌| 297/312 [30:04<01:30,  6.04s/it]
Training Epoch: 17/100, batch 594/625 completed (loss: 0.0279:  96%|█████████▌| 298/312 [30:07<01:24,  6.07s/it]
Training Epoch: 17/100, batch 594/625 completed (loss: 0.0023:  96%|█████████▌| 298/312 [30:07<01:24,  6.07s/it]
Training Epoch: 17/100, batch 594/625 completed (loss: 0.012:  96%|█████████▌| 298/312 [30:07<01:25,  6.07s/it]
Training Epoch: 17/100, batch 594/625 completed (loss: 0.0171:  96%|█████████▌| 298/312 [30:07<01:25,  6.07s/it]
Training Epoch: 17/100, batch 595/625 completed (loss: 0.0337:  96%|█████████▌| 298/312 [30:07<01:25,  6.07s/it]
Training Epoch: 17/100, batch 595/625 completed (loss: 0.0722:  96%|█████████▌| 298/312 [30:07<01:25,  6.07s/it]
Training Epoch: 17/100, batch 595/625 completed (loss: 0.2068:  96%|█████████▌| 298/312 [30:07<01:24,  6.07s/it]
Training Epoch: 17/100, batch 595/625 completed (loss: 0.054:  96%|█████████▌| 298/312 [30:07<01:24,  6.07s/it] 
Training Epoch: 17/100, batch 596/625 completed (loss: 0.022:  96%|█████████▌| 298/312 [30:10<01:25,  6.07s/it] 
Training Epoch: 17/100, batch 596/625 completed (loss: 0.0483:  96%|█████████▌| 298/312 [30:10<01:25,  6.07s/it]
Training Epoch: 17/100, batch 596/625 completed (loss: 0.1759:  96%|█████████▌| 298/312 [30:10<01:24,  6.07s/it]
Training Epoch: 17/100, batch 596/625 completed (loss: 0.0023:  96%|█████████▌| 298/312 [30:10<01:24,  6.07s/it]
Training Epoch: 17/100, batch 596/625 completed (loss: 0.1759:  96%|█████████▌| 299/312 [30:13<01:19,  6.09s/it]
Training Epoch: 17/100, batch 596/625 completed (loss: 0.022:  96%|█████████▌| 299/312 [30:13<01:19,  6.09s/it]
Training Epoch: 17/100, batch 596/625 completed (loss: 0.0483:  96%|█████████▌| 299/312 [30:13<01:19,  6.09s/it]
Training Epoch: 17/100, batch 596/625 completed (loss: 0.0023:  96%|█████████▌| 299/312 [30:13<01:19,  6.10s/it]
Training Epoch: 17/100, batch 597/625 completed (loss: 0.0074:  96%|█████████▌| 299/312 [30:13<01:19,  6.09s/it]
Training Epoch: 17/100, batch 597/625 completed (loss: 0.0446:  96%|█████████▌| 299/312 [30:13<01:19,  6.09s/it]
Training Epoch: 17/100, batch 597/625 completed (loss: 0.0426:  96%|█████████▌| 299/312 [30:13<01:19,  6.09s/it]
Training Epoch: 17/100, batch 597/625 completed (loss: 0.0336:  96%|█████████▌| 299/312 [30:13<01:19,  6.10s/it]
Training Epoch: 17/100, batch 598/625 completed (loss: 0.0569:  96%|█████████▌| 299/312 [30:16<01:19,  6.09s/it]
Training Epoch: 17/100, batch 598/625 completed (loss: 0.0452:  96%|█████████▌| 299/312 [30:16<01:19,  6.09s/it]
Training Epoch: 17/100, batch 598/625 completed (loss: 0.0085:  96%|█████████▌| 299/312 [30:16<01:19,  6.09s/it]
Training Epoch: 17/100, batch 598/625 completed (loss: 0.0068:  96%|█████████▌| 299/312 [30:16<01:19,  6.10s/it]
Training Epoch: 17/100, batch 598/625 completed (loss: 0.0452:  96%|█████████▌| 300/312 [30:19<01:12,  6.08s/it]
Training Epoch: 17/100, batch 598/625 completed (loss: 0.0085:  96%|█████████▌| 300/312 [30:19<01:12,  6.08s/it]
Training Epoch: 17/100, batch 598/625 completed (loss: 0.0569:  96%|█████████▌| 300/312 [30:19<01:13,  6.08s/it]
Training Epoch: 17/100, batch 598/625 completed (loss: 0.0068:  96%|█████████▌| 300/312 [30:19<01:13,  6.08s/it]
Training Epoch: 17/100, batch 599/625 completed (loss: 0.0129:  96%|█████████▌| 300/312 [30:19<01:12,  6.08s/it]
Training Epoch: 17/100, batch 599/625 completed (loss: 0.0212:  96%|█████████▌| 300/312 [30:19<01:12,  6.08s/it]
Training Epoch: 17/100, batch 599/625 completed (loss: 0.0175:  96%|█████████▌| 300/312 [30:19<01:13,  6.08s/it]
Training Epoch: 17/100, batch 599/625 completed (loss: 0.0823:  96%|█████████▌| 300/312 [30:19<01:13,  6.08s/it]
Training Epoch: 17/100, batch 600/625 completed (loss: 0.0336:  96%|█████████▌| 300/312 [30:22<01:13,  6.08s/it]
Training Epoch: 17/100, batch 600/625 completed (loss: 0.0908:  96%|█████████▌| 300/312 [30:22<01:12,  6.08s/it]
Training Epoch: 17/100, batch 600/625 completed (loss: 0.0094:  96%|█████████▌| 300/312 [30:22<01:12,  6.08s/it]
Training Epoch: 17/100, batch 600/625 completed (loss: 0.0094:  96%|█████████▌| 300/312 [30:22<01:13,  6.08s/it]
Training Epoch: 17/100, batch 600/625 completed (loss: 0.0094:  96%|█████████▋| 301/312 [30:25<01:07,  6.15s/it]
Training Epoch: 17/100, batch 600/625 completed (loss: 0.0094:  96%|█████████▋| 301/312 [30:25<01:07,  6.15s/it]
Training Epoch: 17/100, batch 600/625 completed (loss: 0.0336:  96%|█████████▋| 301/312 [30:25<01:07,  6.15s/it]
Training Epoch: 17/100, batch 600/625 completed (loss: 0.0908:  96%|█████████▋| 301/312 [30:25<01:07,  6.15s/it]
Training Epoch: 17/100, batch 601/625 completed (loss: 0.0336:  96%|█████████▋| 301/312 [30:25<01:07,  6.15s/it]
Training Epoch: 17/100, batch 601/625 completed (loss: 0.0184:  96%|█████████▋| 301/312 [30:25<01:07,  6.15s/it]
Training Epoch: 17/100, batch 601/625 completed (loss: 0.0104:  96%|█████████▋| 301/312 [30:25<01:07,  6.15s/it]
Training Epoch: 17/100, batch 601/625 completed (loss: 0.003:  96%|█████████▋| 301/312 [30:25<01:07,  6.15s/it] 
Training Epoch: 17/100, batch 602/625 completed (loss: 0.031:  96%|█████████▋| 301/312 [30:28<01:07,  6.15s/it] 
Training Epoch: 17/100, batch 602/625 completed (loss: 0.0057:  96%|█████████▋| 301/312 [30:28<01:07,  6.15s/it]
Training Epoch: 17/100, batch 602/625 completed (loss: 0.0807:  96%|█████████▋| 301/312 [30:28<01:07,  6.15s/it]
Training Epoch: 17/100, batch 602/625 completed (loss: 0.0029:  96%|█████████▋| 301/312 [30:28<01:07,  6.15s/it]
Training Epoch: 17/100, batch 602/625 completed (loss: 0.0029:  97%|█████████▋| 302/312 [30:31<01:00,  6.05s/it]
Training Epoch: 17/100, batch 602/625 completed (loss: 0.031:  97%|█████████▋| 302/312 [30:31<01:00,  6.06s/it]
Training Epoch: 17/100, batch 602/625 completed (loss: 0.0057:  97%|█████████▋| 302/312 [30:31<01:00,  6.06s/it]
Training Epoch: 17/100, batch 602/625 completed (loss: 0.0807:  97%|█████████▋| 302/312 [30:31<01:00,  6.05s/it]
Training Epoch: 17/100, batch 603/625 completed (loss: 0.1147:  97%|█████████▋| 302/312 [30:31<01:00,  6.05s/it]
Training Epoch: 17/100, batch 603/625 completed (loss: 0.0073:  97%|█████████▋| 302/312 [30:31<01:00,  6.06s/it]
Training Epoch: 17/100, batch 603/625 completed (loss: 0.0281:  97%|█████████▋| 302/312 [30:31<01:00,  6.06s/it]
Training Epoch: 17/100, batch 603/625 completed (loss: 0.173:  97%|█████████▋| 302/312 [30:31<01:00,  6.05s/it] 
Training Epoch: 17/100, batch 604/625 completed (loss: 0.0782:  97%|█████████▋| 302/312 [30:34<01:00,  6.06s/it]
Training Epoch: 17/100, batch 604/625 completed (loss: 0.0357:  97%|█████████▋| 302/312 [30:34<01:00,  6.06s/it]
Training Epoch: 17/100, batch 604/625 completed (loss: 0.088:  97%|█████████▋| 302/312 [30:34<01:00,  6.05s/it] 
Training Epoch: 17/100, batch 604/625 completed (loss: 0.3309:  97%|█████████▋| 302/312 [30:34<01:00,  6.05s/it]
Training Epoch: 17/100, batch 604/625 completed (loss: 0.3309:  97%|█████████▋| 303/312 [30:37<00:54,  6.03s/it]
Training Epoch: 17/100, batch 604/625 completed (loss: 0.088:  97%|█████████▋| 303/312 [30:37<00:54,  6.04s/it]
Training Epoch: 17/100, batch 604/625 completed (loss: 0.0782:  97%|█████████▋| 303/312 [30:37<00:54,  6.04s/it]
Training Epoch: 17/100, batch 604/625 completed (loss: 0.0357:  97%|█████████▋| 303/312 [30:37<00:54,  6.04s/it]
Training Epoch: 17/100, batch 605/625 completed (loss: 0.206:  97%|█████████▋| 303/312 [30:37<00:54,  6.04s/it]
Training Epoch: 17/100, batch 605/625 completed (loss: 0.0108:  97%|█████████▋| 303/312 [30:37<00:54,  6.04s/it]
Training Epoch: 17/100, batch 605/625 completed (loss: 0.3384:  97%|█████████▋| 303/312 [30:37<00:54,  6.04s/it]
Training Epoch: 17/100, batch 605/625 completed (loss: 0.0029:  97%|█████████▋| 303/312 [30:37<00:54,  6.03s/it]
Training Epoch: 17/100, batch 606/625 completed (loss: 0.0788:  97%|█████████▋| 303/312 [30:41<00:54,  6.04s/it]
Training Epoch: 17/100, batch 606/625 completed (loss: 0.057:  97%|█████████▋| 303/312 [30:41<00:54,  6.04s/it] 
Training Epoch: 17/100, batch 606/625 completed (loss: 0.0017:  97%|█████████▋| 303/312 [30:41<00:54,  6.04s/it]
Training Epoch: 17/100, batch 606/625 completed (loss: 0.0951:  97%|█████████▋| 303/312 [30:41<00:54,  6.03s/it]
Training Epoch: 17/100, batch 606/625 completed (loss: 0.0788:  97%|█████████▋| 304/312 [30:43<00:48,  6.12s/it]
Training Epoch: 17/100, batch 606/625 completed (loss: 0.0951:  97%|█████████▋| 304/312 [30:44<00:48,  6.12s/it]
Training Epoch: 17/100, batch 606/625 completed (loss: 0.057:  97%|█████████▋| 304/312 [30:44<00:49,  6.13s/it]
Training Epoch: 17/100, batch 606/625 completed (loss: 0.0017:  97%|█████████▋| 304/312 [30:44<00:49,  6.13s/it]
Training Epoch: 17/100, batch 607/625 completed (loss: 0.2674:  97%|█████████▋| 304/312 [30:44<00:49,  6.13s/it]
Training Epoch: 17/100, batch 607/625 completed (loss: 0.0789:  97%|█████████▋| 304/312 [30:44<00:48,  6.12s/it]
Training Epoch: 17/100, batch 607/625 completed (loss: 0.0384:  97%|█████████▋| 304/312 [30:44<00:49,  6.13s/it]
Training Epoch: 17/100, batch 607/625 completed (loss: 0.0125:  97%|█████████▋| 304/312 [30:44<00:48,  6.12s/it]
Training Epoch: 17/100, batch 608/625 completed (loss: 0.0166:  97%|█████████▋| 304/312 [30:46<00:49,  6.13s/it]
Training Epoch: 17/100, batch 608/625 completed (loss: 0.0256:  97%|█████████▋| 304/312 [30:46<00:48,  6.12s/it]
Training Epoch: 17/100, batch 608/625 completed (loss: 0.0205:  97%|█████████▋| 304/312 [30:46<00:49,  6.13s/it]
Training Epoch: 17/100, batch 608/625 completed (loss: 0.0071:  97%|█████████▋| 304/312 [30:47<00:48,  6.12s/it]
Training Epoch: 17/100, batch 608/625 completed (loss: 0.0256:  98%|█████████▊| 305/312 [30:49<00:42,  6.04s/it]
Training Epoch: 17/100, batch 608/625 completed (loss: 0.0166:  98%|█████████▊| 305/312 [30:49<00:42,  6.04s/it]
Training Epoch: 17/100, batch 608/625 completed (loss: 0.0071:  98%|█████████▊| 305/312 [30:49<00:42,  6.04s/it]
Training Epoch: 17/100, batch 608/625 completed (loss: 0.0205:  98%|█████████▊| 305/312 [30:49<00:42,  6.04s/it]
Training Epoch: 17/100, batch 609/625 completed (loss: 0.022:  98%|█████████▊| 305/312 [30:49<00:42,  6.04s/it] 
Training Epoch: 17/100, batch 609/625 completed (loss: 0.0324:  98%|█████████▊| 305/312 [30:49<00:42,  6.04s/it]
Training Epoch: 17/100, batch 609/625 completed (loss: 0.0066:  98%|█████████▊| 305/312 [30:49<00:42,  6.04s/it]
Training Epoch: 17/100, batch 609/625 completed (loss: 0.0317:  98%|█████████▊| 305/312 [30:49<00:42,  6.04s/it]
Training Epoch: 17/100, batch 610/625 completed (loss: 0.1076:  98%|█████████▊| 305/312 [30:52<00:42,  6.04s/it]
Training Epoch: 17/100, batch 610/625 completed (loss: 0.1484:  98%|█████████▊| 305/312 [30:52<00:42,  6.04s/it]
Training Epoch: 17/100, batch 610/625 completed (loss: 0.016:  98%|█████████▊| 305/312 [30:52<00:42,  6.04s/it]
Training Epoch: 17/100, batch 610/625 completed (loss: 0.0756:  98%|█████████▊| 305/312 [30:53<00:42,  6.04s/it]
Training Epoch: 17/100, batch 610/625 completed (loss: 0.1076:  98%|█████████▊| 306/312 [30:56<00:36,  6.12s/it]
Training Epoch: 17/100, batch 610/625 completed (loss: 0.016:  98%|█████████▊| 306/312 [30:56<00:36,  6.12s/it]
Training Epoch: 17/100, batch 610/625 completed (loss: 0.1484:  98%|█████████▊| 306/312 [30:56<00:36,  6.12s/it]
Training Epoch: 17/100, batch 610/625 completed (loss: 0.0756:  98%|█████████▊| 306/312 [30:56<00:36,  6.12s/it]
Training Epoch: 17/100, batch 611/625 completed (loss: 0.0349:  98%|█████████▊| 306/312 [30:56<00:36,  6.12s/it]
Training Epoch: 17/100, batch 611/625 completed (loss: 0.3573:  98%|█████████▊| 306/312 [30:56<00:36,  6.12s/it]
Training Epoch: 17/100, batch 611/625 completed (loss: 0.1561:  98%|█████████▊| 306/312 [30:56<00:36,  6.12s/it]
Training Epoch: 17/100, batch 611/625 completed (loss: 0.0394:  98%|█████████▊| 306/312 [30:56<00:36,  6.12s/it]
Training Epoch: 17/100, batch 612/625 completed (loss: 0.0772:  98%|█████████▊| 306/312 [30:59<00:36,  6.12s/it]
Training Epoch: 17/100, batch 612/625 completed (loss: 0.1107:  98%|█████████▊| 306/312 [30:59<00:36,  6.12s/it]
Training Epoch: 17/100, batch 612/625 completed (loss: 0.2421:  98%|█████████▊| 306/312 [30:59<00:36,  6.12s/it]
Training Epoch: 17/100, batch 612/625 completed (loss: 0.0775:  98%|█████████▊| 306/312 [30:59<00:36,  6.12s/it]
Training Epoch: 17/100, batch 612/625 completed (loss: 0.1107:  98%|█████████▊| 307/312 [31:02<00:30,  6.10s/it]
Training Epoch: 17/100, batch 612/625 completed (loss: 0.2421:  98%|█████████▊| 307/312 [31:02<00:30,  6.10s/it]
Training Epoch: 17/100, batch 612/625 completed (loss: 0.0772:  98%|█████████▊| 307/312 [31:02<00:30,  6.10s/it]
Training Epoch: 17/100, batch 612/625 completed (loss: 0.0775:  98%|█████████▊| 307/312 [31:02<00:30,  6.10s/it]
Training Epoch: 17/100, batch 613/625 completed (loss: 0.0354:  98%|█████████▊| 307/312 [31:02<00:30,  6.10s/it]
Training Epoch: 17/100, batch 613/625 completed (loss: 0.0434:  98%|█████████▊| 307/312 [31:02<00:30,  6.10s/it]
Training Epoch: 17/100, batch 613/625 completed (loss: 0.2074:  98%|█████████▊| 307/312 [31:02<00:30,  6.10s/it]
Training Epoch: 17/100, batch 613/625 completed (loss: 0.1356:  98%|█████████▊| 307/312 [31:02<00:30,  6.10s/it]
Training Epoch: 17/100, batch 614/625 completed (loss: 0.0727:  98%|█████████▊| 307/312 [31:05<00:30,  6.10s/it]
Training Epoch: 17/100, batch 614/625 completed (loss: 0.0023:  98%|█████████▊| 307/312 [31:05<00:30,  6.10s/it]
Training Epoch: 17/100, batch 614/625 completed (loss: 0.0035:  98%|█████████▊| 307/312 [31:05<00:30,  6.10s/it]
Training Epoch: 17/100, batch 614/625 completed (loss: 0.0088:  98%|█████████▊| 307/312 [31:05<00:30,  6.10s/it]
Training Epoch: 17/100, batch 614/625 completed (loss: 0.0727:  99%|█████████▊| 308/312 [31:08<00:24,  6.07s/it]
Training Epoch: 17/100, batch 614/625 completed (loss: 0.0023:  99%|█████████▊| 308/312 [31:08<00:24,  6.07s/it]
Training Epoch: 17/100, batch 614/625 completed (loss: 0.0035:  99%|█████████▊| 308/312 [31:08<00:24,  6.07s/it]
Training Epoch: 17/100, batch 614/625 completed (loss: 0.0088:  99%|█████████▊| 308/312 [31:08<00:24,  6.07s/it]
Training Epoch: 17/100, batch 615/625 completed (loss: 0.033:  99%|█████████▊| 308/312 [31:08<00:24,  6.07s/it] 
Training Epoch: 17/100, batch 615/625 completed (loss: 0.0155:  99%|█████████▊| 308/312 [31:08<00:24,  6.07s/it]
Training Epoch: 17/100, batch 615/625 completed (loss: 0.0235:  99%|█████████▊| 308/312 [31:08<00:24,  6.07s/it]
Training Epoch: 17/100, batch 615/625 completed (loss: 0.1559:  99%|█████████▊| 308/312 [31:08<00:24,  6.07s/it]
Training Epoch: 17/100, batch 616/625 completed (loss: 0.013:  99%|█████████▊| 308/312 [31:11<00:24,  6.07s/it] 
Training Epoch: 17/100, batch 616/625 completed (loss: 0.0403:  99%|█████████▊| 308/312 [31:11<00:24,  6.07s/it]
Training Epoch: 17/100, batch 616/625 completed (loss: 0.129:  99%|█████████▊| 308/312 [31:11<00:24,  6.07s/it]
Training Epoch: 17/100, batch 616/625 completed (loss: 0.0107:  99%|█████████▊| 308/312 [31:11<00:24,  6.07s/it]
Training Epoch: 17/100, batch 616/625 completed (loss: 0.013:  99%|█████████▉| 309/312 [31:14<00:18,  6.11s/it]
Training Epoch: 17/100, batch 616/625 completed (loss: 0.0403:  99%|█████████▉| 309/312 [31:14<00:18,  6.11s/it]
Training Epoch: 17/100, batch 616/625 completed (loss: 0.129:  99%|█████████▉| 309/312 [31:14<00:18,  6.11s/it]
Training Epoch: 17/100, batch 616/625 completed (loss: 0.0107:  99%|█████████▉| 309/312 [31:14<00:18,  6.12s/it]
Training Epoch: 17/100, batch 617/625 completed (loss: 0.0312:  99%|█████████▉| 309/312 [31:14<00:18,  6.11s/it]
Training Epoch: 17/100, batch 617/625 completed (loss: 0.0176:  99%|█████████▉| 309/312 [31:14<00:18,  6.11s/it]
Training Epoch: 17/100, batch 617/625 completed (loss: 0.0581:  99%|█████████▉| 309/312 [31:14<00:18,  6.11s/it]
Training Epoch: 17/100, batch 617/625 completed (loss: 0.0168:  99%|█████████▉| 309/312 [31:14<00:18,  6.12s/it]
Training Epoch: 17/100, batch 618/625 completed (loss: 0.0156:  99%|█████████▉| 309/312 [31:17<00:18,  6.11s/it]
Training Epoch: 17/100, batch 618/625 completed (loss: 0.0856:  99%|█████████▉| 309/312 [31:17<00:18,  6.11s/it]
Training Epoch: 17/100, batch 618/625 completed (loss: 0.008:  99%|█████████▉| 309/312 [31:17<00:18,  6.11s/it] 
Training Epoch: 17/100, batch 618/625 completed (loss: 0.0393:  99%|█████████▉| 309/312 [31:17<00:18,  6.12s/it]
Training Epoch: 17/100, batch 618/625 completed (loss: 0.0156:  99%|█████████▉| 310/312 [31:20<00:12,  6.04s/it]
Training Epoch: 17/100, batch 618/625 completed (loss: 0.0393:  99%|█████████▉| 310/312 [31:20<00:12,  6.03s/it]
Training Epoch: 17/100, batch 618/625 completed (loss: 0.008:  99%|█████████▉| 310/312 [31:20<00:12,  6.04s/it]
Training Epoch: 17/100, batch 618/625 completed (loss: 0.0856:  99%|█████████▉| 310/312 [31:20<00:12,  6.04s/it]
Training Epoch: 17/100, batch 619/625 completed (loss: 0.0764:  99%|█████████▉| 310/312 [31:20<00:12,  6.04s/it]
Training Epoch: 17/100, batch 619/625 completed (loss: 0.173:  99%|█████████▉| 310/312 [31:20<00:12,  6.04s/it] 
Training Epoch: 17/100, batch 619/625 completed (loss: 0.012:  99%|█████████▉| 310/312 [31:20<00:12,  6.04s/it]
Training Epoch: 17/100, batch 619/625 completed (loss: 0.0131:  99%|█████████▉| 310/312 [31:20<00:12,  6.03s/it]
Training Epoch: 17/100, batch 620/625 completed (loss: 0.0191:  99%|█████████▉| 310/312 [31:23<00:12,  6.04s/it]
Training Epoch: 17/100, batch 620/625 completed (loss: 0.0269:  99%|█████████▉| 310/312 [31:23<00:12,  6.04s/it]
Training Epoch: 17/100, batch 620/625 completed (loss: 0.0975:  99%|█████████▉| 310/312 [31:23<00:12,  6.04s/it]
Training Epoch: 17/100, batch 620/625 completed (loss: 0.039:  99%|█████████▉| 310/312 [31:23<00:12,  6.03s/it] 
Training Epoch: 17/100, batch 620/625 completed (loss: 0.0975: 100%|█████████▉| 311/312 [31:26<00:06,  6.03s/it]
Training Epoch: 17/100, batch 620/625 completed (loss: 0.0269: 100%|█████████▉| 311/312 [31:26<00:06,  6.03s/it]
Training Epoch: 17/100, batch 620/625 completed (loss: 0.039: 100%|█████████▉| 311/312 [31:26<00:06,  6.03s/it]
Training Epoch: 17/100, batch 620/625 completed (loss: 0.0191: 100%|█████████▉| 311/312 [31:26<00:06,  6.03s/it]
Training Epoch: 17/100, batch 621/625 completed (loss: 0.1474: 100%|█████████▉| 311/312 [31:26<00:06,  6.03s/it]
Training Epoch: 17/100, batch 621/625 completed (loss: 0.1272: 100%|█████████▉| 311/312 [31:26<00:06,  6.03s/it]
Training Epoch: 17/100, batch 621/625 completed (loss: 0.3189: 100%|█████████▉| 311/312 [31:26<00:06,  6.03s/it]
Training Epoch: 17/100, batch 621/625 completed (loss: 0.1943: 100%|█████████▉| 311/312 [31:26<00:06,  6.03s/it]
Training Epoch: 17/100, batch 622/625 completed (loss: 0.2776: 100%|█████████▉| 311/312 [31:29<00:06,  6.03s/it]
Training Epoch: 17/100, batch 622/625 completed (loss: 0.1155: 100%|█████████▉| 311/312 [31:29<00:06,  6.03s/it]
Training Epoch: 17/100, batch 622/625 completed (loss: 0.005: 100%|█████████▉| 311/312 [31:29<00:06,  6.03s/it] 
Training Epoch: 17/100, batch 622/625 completed (loss: 0.0423: 100%|█████████▉| 311/312 [31:29<00:06,  6.03s/it]
Training Epoch: 17/100, batch 622/625 completed (loss: 0.005: 100%|██████████| 312/312 [31:32<00:00,  6.11s/it]
Training Epoch: 17/100, batch 622/625 completed (loss: 0.1155: 100%|██████████| 312/312 [31:32<00:00,  6.11s/it]
Training Epoch: 17/100, batch 622/625 completed (loss: 0.2776: 100%|██████████| 312/312 [31:32<00:00,  6.10s/it]
Training Epoch: 17/100, batch 622/625 completed (loss: 0.0423: 100%|██████████| 312/312 [31:32<00:00,  6.11s/it]
Training Epoch: 17/100, batch 623/625 completed (loss: 0.0178: 100%|██████████| 312/312 [31:32<00:00,  6.10s/it]
Training Epoch: 17/100, batch 623/625 completed (loss: 0.1644: 100%|██████████| 312/312 [31:32<00:00,  6.11s/it]
Training Epoch: 17/100, batch 623/625 completed (loss: 0.0787: 100%|██████████| 312/312 [31:32<00:00,  6.11s/it]
Training Epoch: 17/100, batch 623/625 completed (loss: 0.0369: 100%|██████████| 312/312 [31:32<00:00,  6.11s/it]
Training Epoch: 17/100, batch 623/625 completed (loss: 0.0787: : 313it [31:35,  5.16s/it]                       
Training Epoch: 17/100, batch 623/625 completed (loss: 0.0369: : 313it [31:35,  5.15s/it]                       
Training Epoch: 17/100, batch 623/625 completed (loss: 0.1644: : 313it [31:35,  5.16s/it]                       
Training Epoch: 17/100, batch 623/625 completed (loss: 0.0178: : 313it [31:35,  5.16s/it]                       
Training Epoch: 17/100, batch 624/625 completed (loss: 0.0214: : 313it [31:35,  5.16s/it]
Training Epoch: 17/100, batch 624/625 completed (loss: 0.0345: : 313it [31:35,  5.16s/it]
Training Epoch: 17/100, batch 624/625 completed (loss: 0.1026: : 313it [31:35,  5.16s/it]
Training Epoch: 17/100, batch 624/625 completed (loss: 0.1809: : 313it [31:35,  5.15s/it]
Training Epoch: 17/100, batch 624/625 completed (loss: 0.0345: : 313it [31:35,  6.06s/it]

Training Epoch: 17/100, batch 624/625 completed (loss: 0.0214: : 313it [31:35,  6.06s/it]

Training Epoch: 17/100, batch 624/625 completed (loss: 0.1026: : 313it [31:35,  6.06s/it]

Training Epoch: 17/100, batch 624/625 completed (loss: 0.1809: : 313it [31:35,  6.06s/it]
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/c10d_logger.py:83: UserWarning: barrier(): using the device under current context. You can specify `device_id` in `init_process_group` to mute this warning.
  return func(*args, **kwargs)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
saving model + train state.

Processing Files (0 / 0)      : |          |  0.00B /  0.00B            

New Data Upload               : |          |  0.00B /  0.00B            


  ...ve-03b89ee9/checkpoint_17:   1%|          | 18.7MB / 3.63GB            


  ...ve-03b89ee9/checkpoint_17:   1%|          | 18.7MB / 3.63GB            
Processing Files (0 / 1)      :   1%|          | 18.7MB / 3.63GB,   ???B/s  

New Data Upload               :  14%|█▍        | 18.7MB /  134MB,   ???B/s  


  ...ve-03b89ee9/checkpoint_17:   2%|▏         | 89.7MB / 3.63GB            
Processing Files (0 / 1)      :   2%|▏         | 89.7MB / 3.63GB, 8.74MB/s  

New Data Upload               :  45%|████▍     | 89.7MB /  201MB, 8.74MB/s  


  ...ve-03b89ee9/checkpoint_17:   4%|▎         |  133MB / 3.63GB            
Processing Files (0 / 1)      :   4%|▎         |  133MB / 3.63GB, 12.7MB/s  

New Data Upload               :  66%|██████▌   |  133MB /  201MB, 12.7MB/s  


  ...ve-03b89ee9/checkpoint_17:   4%|▎         |  134MB / 3.63GB            
Processing Files (0 / 1)      :   4%|▎         |  134MB / 3.63GB, 12.2MB/s  

New Data Upload               :  50%|████▉     |  134MB /  268MB, 12.2MB/s  


  ...ve-03b89ee9/checkpoint_17:   6%|▌         |  203MB / 3.63GB            
Processing Files (0 / 1)      :   6%|▌         |  203MB / 3.63GB, 18.5MB/s  

New Data Upload               :  61%|██████    |  203MB /  335MB, 18.5MB/s  


  ...ve-03b89ee9/checkpoint_17:   8%|▊         |  298MB / 3.63GB            
Processing Files (0 / 1)      :   8%|▊         |  298MB / 3.63GB, 27.1MB/s  

New Data Upload               :  74%|███████▍  |  298MB /  402MB, 27.1MB/s  


  ...ve-03b89ee9/checkpoint_17:   9%|▉         |  342MB / 3.63GB            
Processing Files (0 / 1)      :   9%|▉         |  342MB / 3.63GB, 30.5MB/s  

New Data Upload               :  73%|███████▎  |  342MB /  469MB, 30.5MB/s  


  ...ve-03b89ee9/checkpoint_17:  11%|█         |  401MB / 3.63GB            
Processing Files (0 / 1)      :  11%|█         |  401MB / 3.63GB, 35.3MB/s  

New Data Upload               :  86%|████████▌ |  401MB /  469MB, 35.3MB/s  


  ...ve-03b89ee9/checkpoint_17:  12%|█▏        |  448MB / 3.63GB            
Processing Files (0 / 1)      :  12%|█▏        |  448MB / 3.63GB, 38.9MB/s  

New Data Upload               :  84%|████████▎ |  448MB /  536MB, 38.9MB/s  


  ...ve-03b89ee9/checkpoint_17:  14%|█▍        |  521MB / 3.63GB            
Processing Files (0 / 1)      :  14%|█▍        |  521MB / 3.63GB, 44.7MB/s  

New Data Upload               :  78%|███████▊  |  521MB /  671MB, 44.7MB/s  


  ...ve-03b89ee9/checkpoint_17:  17%|█▋        |  611MB / 3.63GB            
Processing Files (0 / 1)      :  17%|█▋        |  611MB / 3.63GB, 51.9MB/s  

New Data Upload               :  83%|████████▎ |  611MB /  738MB, 51.9MB/s  


  ...ve-03b89ee9/checkpoint_17:  19%|█▉        |  687MB / 3.63GB            
Processing Files (0 / 1)      :  19%|█▉        |  687MB / 3.63GB, 57.6MB/s  

New Data Upload               :  85%|████████▌ |  687MB /  805MB, 57.6MB/s  


  ...ve-03b89ee9/checkpoint_17:  22%|██▏       |  783MB / 3.63GB            
Processing Files (0 / 1)      :  22%|██▏       |  783MB / 3.63GB, 65.0MB/s  

New Data Upload               :  90%|████████▉ |  783MB /  872MB, 65.0MB/s  


  ...ve-03b89ee9/checkpoint_17:  24%|██▎       |  860MB / 3.63GB            
Processing Files (0 / 1)      :  24%|██▎       |  860MB / 3.63GB, 70.5MB/s  

New Data Upload               :  92%|█████████▏|  860MB /  939MB, 70.5MB/s  


  ...ve-03b89ee9/checkpoint_17:  25%|██▌       |  924MB / 3.63GB            
Processing Files (0 / 1)      :  25%|██▌       |  924MB / 3.63GB, 74.7MB/s  

New Data Upload               :  86%|████████▌ |  924MB / 1.07GB, 74.7MB/s  


  ...ve-03b89ee9/checkpoint_17:  28%|██▊       | 1.01GB / 3.63GB            
Processing Files (0 / 1)      :  28%|██▊       | 1.01GB / 3.63GB, 80.8MB/s  

New Data Upload               :  89%|████████▉ | 1.01GB / 1.14GB, 80.8MB/s  


  ...ve-03b89ee9/checkpoint_17:  31%|███       | 1.12GB / 3.63GB            
Processing Files (0 / 1)      :  31%|███       | 1.12GB / 3.63GB, 88.4MB/s  

New Data Upload               :  88%|████████▊ | 1.12GB / 1.27GB, 88.4MB/s  


  ...ve-03b89ee9/checkpoint_17:  34%|███▎      | 1.22GB / 3.63GB            
Processing Files (0 / 1)      :  34%|███▎      | 1.22GB / 3.63GB, 95.3MB/s  

New Data Upload               :  91%|█████████ | 1.22GB / 1.34GB, 95.3MB/s  


  ...ve-03b89ee9/checkpoint_17:  36%|███▌      | 1.30GB / 3.63GB            
Processing Files (0 / 1)      :  36%|███▌      | 1.30GB / 3.63GB,  101MB/s  

New Data Upload               :  92%|█████████▏| 1.30GB / 1.41GB,  101MB/s  


  ...ve-03b89ee9/checkpoint_17:  38%|███▊      | 1.37GB / 3.63GB            
Processing Files (0 / 1)      :  38%|███▊      | 1.37GB / 3.63GB,  104MB/s  

New Data Upload               :  89%|████████▉ | 1.37GB / 1.54GB,  104MB/s  


  ...ve-03b89ee9/checkpoint_17:  41%|████      | 1.48GB / 3.63GB            
Processing Files (0 / 1)      :  41%|████      | 1.48GB / 3.63GB,  112MB/s  

New Data Upload               :  92%|█████████▏| 1.48GB / 1.61GB,  112MB/s  


  ...ve-03b89ee9/checkpoint_17:  44%|████▎     | 1.58GB / 3.63GB            
Processing Files (0 / 1)      :  44%|████▎     | 1.58GB / 3.63GB,  118MB/s  

New Data Upload               :  94%|█████████▍| 1.58GB / 1.68GB,  118MB/s  


  ...ve-03b89ee9/checkpoint_17:  46%|████▌     | 1.65GB / 3.63GB            
Processing Files (0 / 1)      :  46%|████▌     | 1.65GB / 3.63GB,  121MB/s  

New Data Upload               :  95%|█████████▍| 1.65GB / 1.74GB,  121MB/s  


  ...ve-03b89ee9/checkpoint_17:  48%|████▊     | 1.75GB / 3.63GB            
Processing Files (0 / 1)      :  48%|████▊     | 1.75GB / 3.63GB,  127MB/s  

New Data Upload               :  93%|█████████▎| 1.75GB / 1.88GB,  127MB/s  


  ...ve-03b89ee9/checkpoint_17:  50%|████▉     | 1.81GB / 3.63GB            
Processing Files (0 / 1)      :  50%|████▉     | 1.81GB / 3.63GB,  130MB/s  

New Data Upload               :  93%|█████████▎| 1.81GB / 1.94GB,  130MB/s  


  ...ve-03b89ee9/checkpoint_17:  53%|█████▎    | 1.92GB / 3.63GB            
Processing Files (0 / 1)      :  53%|█████▎    | 1.92GB / 3.63GB,  136MB/s  

New Data Upload               :  95%|█████████▌| 1.92GB / 2.01GB,  136MB/s  


  ...ve-03b89ee9/checkpoint_17:  55%|█████▌    | 2.00GB / 3.63GB            
Processing Files (0 / 1)      :  55%|█████▌    | 2.00GB / 3.63GB,  141MB/s  

New Data Upload               :  93%|█████████▎| 2.00GB / 2.15GB,  141MB/s  


  ...ve-03b89ee9/checkpoint_17:  57%|█████▋    | 2.08GB / 3.63GB            
Processing Files (0 / 1)      :  57%|█████▋    | 2.08GB / 3.63GB,  144MB/s  

New Data Upload               :  94%|█████████▍| 2.08GB / 2.21GB,  144MB/s  


  ...ve-03b89ee9/checkpoint_17:  59%|█████▉    | 2.16GB / 3.63GB            
Processing Files (0 / 1)      :  59%|█████▉    | 2.16GB / 3.63GB,  147MB/s  

New Data Upload               :  95%|█████████▍| 2.16GB / 2.28GB,  147MB/s  


  ...ve-03b89ee9/checkpoint_17:  65%|██████▌   | 2.37GB / 3.63GB            
Processing Files (0 / 1)      :  65%|██████▌   | 2.37GB / 3.63GB,  162MB/s  

New Data Upload               :  97%|█████████▋| 2.27GB / 2.35GB,  154MB/s  


  ...ve-03b89ee9/checkpoint_17:  78%|███████▊  | 2.83GB / 3.63GB            
Processing Files (0 / 1)      :  78%|███████▊  | 2.83GB / 3.63GB,  197MB/s  

New Data Upload               : 100%|█████████▉| 2.35GB / 2.35GB,  157MB/s  


  ...ve-03b89ee9/checkpoint_17:  88%|████████▊ | 3.20GB / 3.63GB            
Processing Files (0 / 1)      :  88%|████████▊ | 3.20GB / 3.63GB,  224MB/s  

New Data Upload               : 100%|█████████▉| 2.35GB / 2.35GB,  154MB/s  


  ...ve-03b89ee9/checkpoint_17:  99%|█████████▊| 3.58GB / 3.63GB            
Processing Files (0 / 1)      :  99%|█████████▊| 3.58GB / 3.63GB,  251MB/s  


  ...ve-03b89ee9/checkpoint_17: 100%|█████████▉| 3.63GB / 3.63GB            
Processing Files (0 / 1)      : 100%|█████████▉| 3.63GB / 3.63GB,  249MB/s  

New Data Upload               : 100%|█████████▉| 2.38GB / 2.38GB,  150MB/s  


  ...ve-03b89ee9/checkpoint_17: 100%|█████████▉| 3.63GB / 3.63GB            
Processing Files (0 / 1)      : 100%|█████████▉| 3.63GB / 3.63GB,  244MB/s  

New Data Upload               : 100%|█████████▉| 2.38GB / 2.38GB,  147MB/s  


  ...ve-03b89ee9/checkpoint_17: 100%|█████████▉| 3.63GB / 3.63GB            
Processing Files (0 / 1)      : 100%|█████████▉| 3.63GB / 3.63GB,  239MB/s  

New Data Upload               : 100%|█████████▉| 2.38GB / 2.38GB,  144MB/s  


  ...ve-03b89ee9/checkpoint_17: 100%|██████████| 3.63GB / 3.63GB            
Processing Files (1 / 1)      : 100%|██████████| 3.63GB / 3.63GB,  234MB/s  

New Data Upload               : 100%|██████████| 2.38GB / 2.38GB,  141MB/s  
Processing Files (1 / 1)      : 100%|██████████| 3.63GB / 3.63GB,  234MB/s  

New Data Upload               : 100%|██████████| 2.38GB / 2.38GB,  141MB/s  

  ...ve-03b89ee9/checkpoint_17: 100%|██████████| 3.63GB / 3.63GB            
Uploaded checkpoint_17 to seyedparsa/thoughtformer-checkpoints/checkpoints/star-datacurr-only-4gpu-bs32-lr1e-5_n1000_qwen3-0.6b-base_thoughtformer_lr1e-5_no-reset_dc_no-tc_adaptive-03b89ee9/checkpoint_17

Processing Files (0 / 0)      : |          |  0.00B /  0.00B            

New Data Upload               : |          |  0.00B /  0.00B            


  ...e-03b89ee9/train_state_17:   0%|          | 3.69MB / 4.77GB            


  ...e-03b89ee9/train_state_17:   0%|          | 3.69MB / 4.77GB            
Processing Files (0 / 1)      :   0%|          | 3.69MB / 4.77GB,   ???B/s  

New Data Upload               :   3%|▎         | 3.69MB /  134MB,   ???B/s  


  ...e-03b89ee9/train_state_17:   2%|▏         | 93.2MB / 4.77GB            
Processing Files (0 / 1)      :   2%|▏         | 93.2MB / 4.77GB,   ???B/s  

New Data Upload               :  46%|████▋     | 93.2MB /  201MB,   ???B/s  


  ...e-03b89ee9/train_state_17:   3%|▎         |  134MB / 4.77GB            
Processing Files (0 / 1)      :   3%|▎         |  134MB / 4.77GB, 12.9MB/s  

New Data Upload               :  66%|██████▋   |  134MB /  201MB, 12.9MB/s  


  ...e-03b89ee9/train_state_17:   3%|▎         |  162MB / 4.77GB            
Processing Files (0 / 1)      :   3%|▎         |  162MB / 4.77GB, 14.7MB/s  

New Data Upload               :  48%|████▊     |  162MB /  335MB, 14.7MB/s  


  ...e-03b89ee9/train_state_17:   5%|▌         |  251MB / 4.77GB            
Processing Files (0 / 1)      :   5%|▌         |  251MB / 4.77GB, 22.8MB/s  

New Data Upload               :  62%|██████▏   |  251MB /  402MB, 22.8MB/s  


  ...e-03b89ee9/train_state_17:   7%|▋         |  338MB / 4.77GB            
Processing Files (0 / 1)      :   7%|▋         |  338MB / 4.77GB, 30.5MB/s  

New Data Upload               :  84%|████████▍ |  338MB /  402MB, 30.5MB/s  


  ...e-03b89ee9/train_state_17:   8%|▊         |  370MB / 4.77GB            
Processing Files (0 / 1)      :   8%|▊         |  370MB / 4.77GB, 32.8MB/s  

New Data Upload               :  79%|███████▉  |  370MB /  469MB, 32.8MB/s  


  ...e-03b89ee9/train_state_17:   9%|▊         |  405MB / 4.77GB            
Processing Files (0 / 1)      :   9%|▊         |  405MB / 4.77GB, 35.3MB/s  

New Data Upload               :  76%|███████▌  |  405MB /  536MB, 35.3MB/s  


  ...e-03b89ee9/train_state_17:  10%|▉         |  475MB / 4.77GB            
Processing Files (0 / 1)      :  10%|▉         |  475MB / 4.77GB, 40.9MB/s  

New Data Upload               :  79%|███████▊  |  475MB /  604MB, 40.9MB/s  


  ...e-03b89ee9/train_state_17:  12%|█▏        |  551MB / 4.77GB            
Processing Files (0 / 1)      :  12%|█▏        |  551MB / 4.77GB, 47.0MB/s  

New Data Upload               :  75%|███████▍  |  551MB /  738MB, 47.0MB/s  


  ...e-03b89ee9/train_state_17:  14%|█▍        |  656MB / 4.77GB            
Processing Files (0 / 1)      :  14%|█▍        |  656MB / 4.77GB, 55.5MB/s  

New Data Upload               :  81%|████████▏ |  656MB /  805MB, 55.5MB/s  


  ...e-03b89ee9/train_state_17:  16%|█▌        |  765MB / 4.77GB            
Processing Files (0 / 1)      :  16%|█▌        |  765MB / 4.77GB, 64.2MB/s  

New Data Upload               :  88%|████████▊ |  765MB /  872MB, 64.2MB/s  


  ...e-03b89ee9/train_state_17:  18%|█▊        |  844MB / 4.77GB            
Processing Files (0 / 1)      :  18%|█▊        |  844MB / 4.77GB, 69.9MB/s  

New Data Upload               :  84%|████████▍ |  844MB / 1.01GB, 69.9MB/s  


  ...e-03b89ee9/train_state_17:  20%|█▉        |  941MB / 4.77GB            
Processing Files (0 / 1)      :  20%|█▉        |  941MB / 4.77GB, 77.1MB/s  

New Data Upload               :  88%|████████▊ |  941MB / 1.07GB, 77.1MB/s  


  ...e-03b89ee9/train_state_17:  22%|██▏       | 1.03GB / 4.77GB            
Processing Files (0 / 1)      :  22%|██▏       | 1.03GB / 4.77GB, 83.5MB/s  

New Data Upload               :  85%|████████▌ | 1.03GB / 1.21GB, 83.5MB/s  


  ...e-03b89ee9/train_state_17:  24%|██▍       | 1.15GB / 4.77GB            
Processing Files (0 / 1)      :  24%|██▍       | 1.15GB / 4.77GB, 92.5MB/s  

New Data Upload               :  91%|█████████ | 1.15GB / 1.27GB, 92.5MB/s  


  ...e-03b89ee9/train_state_17:  26%|██▌       | 1.22GB / 4.77GB            
Processing Files (0 / 1)      :  26%|██▌       | 1.22GB / 4.77GB, 96.4MB/s  

New Data Upload               :  91%|█████████ | 1.22GB / 1.34GB, 96.4MB/s  


  ...e-03b89ee9/train_state_17:  27%|██▋       | 1.31GB / 4.77GB            
Processing Files (0 / 1)      :  27%|██▋       | 1.31GB / 4.77GB,  102MB/s  

New Data Upload               :  89%|████████▊ | 1.31GB / 1.48GB,  102MB/s  


  ...e-03b89ee9/train_state_17:  30%|███       | 1.43GB / 4.77GB            
Processing Files (0 / 1)      :  30%|███       | 1.43GB / 4.77GB,  111MB/s  

New Data Upload               :  93%|█████████▎| 1.43GB / 1.54GB,  111MB/s  


  ...e-03b89ee9/train_state_17:  32%|███▏      | 1.55GB / 4.77GB            
Processing Files (0 / 1)      :  32%|███▏      | 1.55GB / 4.77GB,  118MB/s  

New Data Upload               :  92%|█████████▏| 1.55GB / 1.68GB,  118MB/s  


  ...e-03b89ee9/train_state_17:  34%|███▍      | 1.62GB / 4.77GB            
Processing Files (0 / 1)      :  34%|███▍      | 1.62GB / 4.77GB,  122MB/s  

New Data Upload               :  93%|█████████▎| 1.62GB / 1.74GB,  122MB/s  


  ...e-03b89ee9/train_state_17:  36%|███▌      | 1.71GB / 4.77GB            
Processing Files (0 / 1)      :  36%|███▌      | 1.71GB / 4.77GB,  127MB/s  

New Data Upload               :  91%|█████████ | 1.71GB / 1.88GB,  127MB/s  


  ...e-03b89ee9/train_state_17:  39%|███▊      | 1.84GB / 4.77GB            
Processing Files (0 / 1)      :  39%|███▊      | 1.84GB / 4.77GB,  136MB/s  

New Data Upload               :  95%|█████████▍| 1.84GB / 1.94GB,  136MB/s  


  ...e-03b89ee9/train_state_17:  41%|████      | 1.94GB / 4.77GB            
Processing Files (0 / 1)      :  41%|████      | 1.94GB / 4.77GB,  141MB/s  

New Data Upload               :  93%|█████████▎| 1.94GB / 2.08GB,  141MB/s  


  ...e-03b89ee9/train_state_17:  42%|████▏     | 2.00GB / 4.77GB            
Processing Files (0 / 1)      :  42%|████▏     | 2.00GB / 4.77GB,  144MB/s  

New Data Upload               :  93%|█████████▎| 2.00GB / 2.15GB,  144MB/s  


  ...e-03b89ee9/train_state_17:  44%|████▍     | 2.09GB / 4.77GB            
Processing Files (0 / 1)      :  44%|████▍     | 2.09GB / 4.77GB,  148MB/s  

New Data Upload               :  92%|█████████▏| 2.09GB / 2.28GB,  148MB/s  


  ...e-03b89ee9/train_state_17:  46%|████▋     | 2.21GB / 4.77GB            
Processing Files (0 / 1)      :  46%|████▋     | 2.21GB / 4.77GB,  155MB/s  

New Data Upload               :  94%|█████████▍| 2.21GB / 2.35GB,  155MB/s  


  ...e-03b89ee9/train_state_17:  49%|████▉     | 2.32GB / 4.77GB            
Processing Files (0 / 1)      :  49%|████▉     | 2.32GB / 4.77GB,  161MB/s  

New Data Upload               :  94%|█████████▎| 2.32GB / 2.48GB,  161MB/s  


  ...e-03b89ee9/train_state_17:  51%|█████     | 2.42GB / 4.77GB            
Processing Files (0 / 1)      :  51%|█████     | 2.42GB / 4.77GB,  166MB/s  

New Data Upload               :  95%|█████████▍| 2.42GB / 2.55GB,  166MB/s  


  ...e-03b89ee9/train_state_17:  53%|█████▎    | 2.52GB / 4.77GB            
Processing Files (0 / 1)      :  53%|█████▎    | 2.52GB / 4.77GB,  171MB/s  

New Data Upload               :  94%|█████████▍| 2.52GB / 2.68GB,  171MB/s  


  ...e-03b89ee9/train_state_17:  54%|█████▍    | 2.58GB / 4.77GB            
Processing Files (0 / 1)      :  54%|█████▍    | 2.58GB / 4.77GB,  172MB/s  

New Data Upload               :  94%|█████████▍| 2.58GB / 2.75GB,  172MB/s  


  ...e-03b89ee9/train_state_17:  56%|█████▌    | 2.68GB / 4.77GB            
Processing Files (0 / 1)      :  56%|█████▌    | 2.68GB / 4.77GB,  177MB/s  

New Data Upload               :  95%|█████████▌| 2.68GB / 2.82GB,  177MB/s  


  ...e-03b89ee9/train_state_17:  58%|█████▊    | 2.78GB / 4.77GB            
Processing Files (0 / 1)      :  58%|█████▊    | 2.78GB / 4.77GB,  181MB/s  

New Data Upload               :  94%|█████████▍| 2.78GB / 2.95GB,  181MB/s  


  ...e-03b89ee9/train_state_17:  61%|██████    | 2.90GB / 4.77GB            
Processing Files (0 / 1)      :  61%|██████    | 2.90GB / 4.77GB,  188MB/s  

New Data Upload               :  96%|█████████▌| 2.90GB / 3.02GB,  188MB/s  


  ...e-03b89ee9/train_state_17:  63%|██████▎   | 2.99GB / 4.77GB            
Processing Files (0 / 1)      :  63%|██████▎   | 2.99GB / 4.77GB,  191MB/s  

New Data Upload               :  95%|█████████▍| 2.99GB / 3.15GB,  191MB/s  


  ...e-03b89ee9/train_state_17:  65%|██████▍   | 3.09GB / 4.77GB            
Processing Files (0 / 1)      :  65%|██████▍   | 3.09GB / 4.77GB,  196MB/s  

New Data Upload               :  96%|█████████▌| 3.09GB / 3.22GB,  196MB/s  


  ...e-03b89ee9/train_state_17:  67%|██████▋   | 3.17GB / 4.77GB            
Processing Files (0 / 1)      :  67%|██████▋   | 3.17GB / 4.77GB,  198MB/s  

New Data Upload               :  95%|█████████▍| 3.17GB / 3.35GB,  198MB/s  


  ...e-03b89ee9/train_state_17:  68%|██████▊   | 3.25GB / 4.77GB            
Processing Files (0 / 1)      :  68%|██████▊   | 3.25GB / 4.77GB,  201MB/s  

New Data Upload               :  95%|█████████▌| 3.25GB / 3.42GB,  201MB/s  


  ...e-03b89ee9/train_state_17:  70%|███████   | 3.35GB / 4.77GB            
Processing Files (0 / 1)      :  70%|███████   | 3.35GB / 4.77GB,  205MB/s  

New Data Upload               :  94%|█████████▍| 3.35GB / 3.55GB,  205MB/s  


  ...e-03b89ee9/train_state_17:  73%|███████▎  | 3.48GB / 4.77GB            
Processing Files (0 / 1)      :  73%|███████▎  | 3.48GB / 4.77GB,  211MB/s  

New Data Upload               :  96%|█████████▌| 3.48GB / 3.62GB,  211MB/s  


  ...e-03b89ee9/train_state_17:  75%|███████▌  | 3.59GB / 4.77GB            
Processing Files (0 / 1)      :  75%|███████▌  | 3.59GB / 4.77GB,  216MB/s  

New Data Upload               :  96%|█████████▌| 3.59GB / 3.76GB,  216MB/s  


  ...e-03b89ee9/train_state_17:  77%|███████▋  | 3.67GB / 4.77GB            
Processing Files (0 / 1)      :  77%|███████▋  | 3.67GB / 4.77GB,  218MB/s  

New Data Upload               :  96%|█████████▌| 3.67GB / 3.82GB,  218MB/s  


  ...e-03b89ee9/train_state_17:  79%|███████▉  | 3.78GB / 4.77GB            
Processing Files (0 / 1)      :  79%|███████▉  | 3.78GB / 4.77GB,  223MB/s  

New Data Upload               :  97%|█████████▋| 3.78GB / 3.89GB,  223MB/s  


  ...e-03b89ee9/train_state_17:  81%|████████▏ | 3.88GB / 4.77GB            
Processing Files (0 / 1)      :  81%|████████▏ | 3.88GB / 4.77GB,  226MB/s  

New Data Upload               :  96%|█████████▋| 3.88GB / 4.02GB,  226MB/s  


  ...e-03b89ee9/train_state_17:  83%|████████▎ | 3.95GB / 4.77GB            
Processing Files (0 / 1)      :  83%|████████▎ | 3.95GB / 4.77GB,  227MB/s  

New Data Upload               :  97%|█████████▋| 3.95GB / 4.09GB,  227MB/s  


  ...e-03b89ee9/train_state_17:  86%|████████▌ | 4.08GB / 4.77GB            
Processing Files (0 / 1)      :  86%|████████▌ | 4.08GB / 4.77GB,  233MB/s  

New Data Upload               :  97%|█████████▋| 4.08GB / 4.22GB,  233MB/s  


  ...e-03b89ee9/train_state_17:  87%|████████▋ | 4.16GB / 4.77GB            
Processing Files (0 / 1)      :  87%|████████▋ | 4.16GB / 4.77GB,  235MB/s  

New Data Upload               :  96%|█████████▌| 4.16GB / 4.36GB,  235MB/s  


  ...e-03b89ee9/train_state_17:  89%|████████▉ | 4.26GB / 4.77GB            
Processing Files (0 / 1)      :  89%|████████▉ | 4.26GB / 4.77GB,  238MB/s  

New Data Upload               :  96%|█████████▋| 4.26GB / 4.43GB,  238MB/s  


  ...e-03b89ee9/train_state_17:  92%|█████████▏| 4.37GB / 4.77GB            
Processing Files (0 / 1)      :  92%|█████████▏| 4.37GB / 4.77GB,  242MB/s  

New Data Upload               :  96%|█████████▌| 4.37GB / 4.56GB,  242MB/s  


  ...e-03b89ee9/train_state_17:  94%|█████████▍| 4.48GB / 4.77GB            
Processing Files (0 / 1)      :  94%|█████████▍| 4.48GB / 4.77GB,  246MB/s  

New Data Upload               :  97%|█████████▋| 4.48GB / 4.63GB,  246MB/s  


  ...e-03b89ee9/train_state_17:  96%|█████████▌| 4.58GB / 4.77GB            
Processing Files (0 / 1)      :  96%|█████████▌| 4.58GB / 4.77GB,  249MB/s  

New Data Upload               :  96%|█████████▌| 4.58GB / 4.76GB,  249MB/s  


  ...e-03b89ee9/train_state_17:  99%|█████████▊| 4.70GB / 4.77GB            
Processing Files (0 / 1)      :  99%|█████████▊| 4.70GB / 4.77GB,  253MB/s  

New Data Upload               :  99%|█████████▊| 4.70GB / 4.77GB,  253MB/s  


  ...e-03b89ee9/train_state_17: 100%|█████████▉| 4.77GB / 4.77GB            
Processing Files (0 / 1)      : 100%|█████████▉| 4.77GB / 4.77GB,  253MB/s  

New Data Upload               : 100%|█████████▉| 4.77GB / 4.77GB,  253MB/s  


  ...e-03b89ee9/train_state_17: 100%|█████████▉| 4.77GB / 4.77GB            
Processing Files (0 / 1)      : 100%|█████████▉| 4.77GB / 4.77GB,  249MB/s  

New Data Upload               : 100%|█████████▉| 4.77GB / 4.77GB,  249MB/s  


  ...e-03b89ee9/train_state_17: 100%|█████████▉| 4.77GB / 4.77GB            
Processing Files (0 / 1)      : 100%|█████████▉| 4.77GB / 4.77GB,  244MB/s  

New Data Upload               : 100%|█████████▉| 4.77GB / 4.77GB,  244MB/s  


  ...e-03b89ee9/train_state_17: 100%|██████████| 4.77GB / 4.77GB            
Processing Files (1 / 1)      : 100%|██████████| 4.77GB / 4.77GB,  239MB/s  

New Data Upload               : 100%|██████████| 4.77GB / 4.77GB,  239MB/s  
Processing Files (1 / 1)      : 100%|██████████| 4.77GB / 4.77GB,  239MB/s  

New Data Upload               : 100%|██████████| 4.77GB / 4.77GB,  239MB/s  

  ...e-03b89ee9/train_state_17: 100%|██████████| 4.77GB / 4.77GB            
Uploaded train_state_17 to seyedparsa/thoughtformer-checkpoints/checkpoints/star-datacurr-only-4gpu-bs32-lr1e-5_n1000_qwen3-0.6b-base_thoughtformer_lr1e-5_no-reset_dc_no-tc_adaptive-03b89ee9/train_state_17
Deleted old checkpoint: train_state_16
Deleted old checkpoint: checkpoint_16

Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]
Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]
Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]eval loss 0.024477771945327284

Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]
[CORRECT] Example 0 (diff=3)
  Expected : 'Claude'
  Extracted: 'Claude'
  Generated: '### Claude'

[CORRECT] Example 1 (diff=8)
  Expected : 'Guy'
  Extracted: 'Guy'
  Generated: '### Guy'

Test Accuracy:   3%|▎         | 1/32 [00:06<03:21,  6.51s/it]
Test Accuracy:   3%|▎         | 1/32 [00:06<03:21,  6.51s/it]
Test Accuracy:   3%|▎         | 1/32 [00:06<03:21,  6.51s/it]
Test accuracy: 1.0:   3%|▎         | 1/32 [00:06<03:21,  6.51s/it]
Test accuracy: 0.88:   3%|▎         | 1/32 [00:06<03:21,  6.51s/it]
Test accuracy: 1.0:   3%|▎         | 1/32 [00:06<03:21,  6.51s/it]
[CORRECT] Example 2 (diff=10)
  Expected : 'Elvis'
  Extracted: 'Elvis'
  Generated: '### Elvis'

[CORRECT] Example 3 (diff=8)
  Expected : 'Coral'
  Extracted: 'Coral'
  Generated: '### Coral'

[CORRECT] Example 4 (diff=1)
  Expected : 'Patricia'
  Extracted: 'Patricia'
  Generated: '### Patricia'

Test Accuracy:   3%|▎         | 1/32 [00:06<03:17,  6.38s/it]
Test accuracy: 1.0:   3%|▎         | 1/32 [00:06<03:17,  6.38s/it]
Test accuracy: 1.0:   6%|▋         | 2/32 [00:12<03:13,  6.44s/it]
Test accuracy: 1.0:   6%|▋         | 2/32 [00:12<03:13,  6.44s/it]
Test accuracy: 0.88:   6%|▋         | 2/32 [00:12<03:13,  6.44s/it]
Test accuracy: 1.0:   6%|▋         | 2/32 [00:12<03:13,  6.44s/it]
Test accuracy: 0.94:   6%|▋         | 2/32 [00:12<03:13,  6.44s/it]
Test accuracy: 0.94:   6%|▋         | 2/32 [00:12<03:13,  6.44s/it]
Test accuracy: 1.0:   6%|▋         | 2/32 [00:12<03:11,  6.39s/it]
Test accuracy: 1.0:   6%|▋         | 2/32 [00:12<03:11,  6.39s/it]
Test accuracy: 0.94:   9%|▉         | 3/32 [00:19<03:14,  6.70s/it]
Test accuracy: 0.96:   9%|▉         | 3/32 [00:19<03:14,  6.70s/it]
Test accuracy: 1.0:   9%|▉         | 3/32 [00:19<03:14,  6.70s/it]
Test accuracy: 1.0:   9%|▉         | 3/32 [00:19<03:14,  6.70s/it]
Test accuracy: 1.0:   9%|▉         | 3/32 [00:19<03:13,  6.67s/it]
Test accuracy: 1.0:   9%|▉         | 3/32 [00:19<03:13,  6.67s/it]
Test accuracy: 0.94:   9%|▉         | 3/32 [00:19<03:14,  6.70s/it]
Test accuracy: 0.96:   9%|▉         | 3/32 [00:19<03:14,  6.70s/it]
Test accuracy: 0.96:  12%|█▎        | 4/32 [00:26<03:01,  6.48s/it]
Test accuracy: 1.0:  12%|█▎        | 4/32 [00:26<03:01,  6.48s/it]
Test accuracy: 0.96:  12%|█▎        | 4/32 [00:26<03:01,  6.48s/it]
Test accuracy: 0.97:  12%|█▎        | 4/32 [00:26<03:01,  6.48s/it]
Test accuracy: 1.0:  12%|█▎        | 4/32 [00:26<03:01,  6.48s/it]
Test accuracy: 0.97:  12%|█▎        | 4/32 [00:26<03:01,  6.48s/it]
Test accuracy: 1.0:  12%|█▎        | 4/32 [00:25<03:00,  6.46s/it]
Test accuracy: 1.0:  12%|█▎        | 4/32 [00:25<03:00,  6.46s/it]
Test accuracy: 0.97:  16%|█▌        | 5/32 [00:32<02:53,  6.41s/it]
Test accuracy: 1.0:  16%|█▌        | 5/32 [00:32<02:53,  6.41s/it]
Test accuracy: 1.0:  16%|█▌        | 5/32 [00:32<02:53,  6.41s/it]
Test accuracy: 0.95:  16%|█▌        | 5/32 [00:32<02:53,  6.41s/it]
Test accuracy: 0.97:  16%|█▌        | 5/32 [00:32<02:53,  6.41s/it]
Test accuracy: 0.98:  16%|█▌        | 5/32 [00:32<02:53,  6.41s/it]
Test accuracy: 1.0:  16%|█▌        | 5/32 [00:32<02:52,  6.40s/it]
Test accuracy: 1.0:  16%|█▌        | 5/32 [00:32<02:52,  6.40s/it]
Test accuracy: 0.95:  19%|█▉        | 6/32 [00:38<02:45,  6.36s/it]
Test accuracy: 0.98:  19%|█▉        | 6/32 [00:38<02:45,  6.36s/it]
Test accuracy: 1.0:  19%|█▉        | 6/32 [00:38<02:45,  6.36s/it]
Test accuracy: 0.94:  19%|█▉        | 6/32 [00:38<02:45,  6.36s/it]
Test accuracy: 0.98:  19%|█▉        | 6/32 [00:38<02:45,  6.36s/it]
Test accuracy: 0.96:  19%|█▉        | 6/32 [00:38<02:45,  6.36s/it]
Test accuracy: 1.0:  19%|█▉        | 6/32 [00:38<02:45,  6.36s/it]
Test accuracy: 1.0:  19%|█▉        | 6/32 [00:38<02:45,  6.36s/it]
Test accuracy: 0.94:  22%|██▏       | 7/32 [00:44<02:38,  6.36s/it]
Test accuracy: 0.96:  22%|██▏       | 7/32 [00:44<02:38,  6.36s/it]
Test accuracy: 0.98:  22%|██▏       | 7/32 [00:44<02:38,  6.36s/it]
Test accuracy: 0.95:  22%|██▏       | 7/32 [00:44<02:38,  6.36s/it]
Test accuracy: 0.96:  22%|██▏       | 7/32 [00:44<02:38,  6.36s/it]
Test accuracy: 0.98:  22%|██▏       | 7/32 [00:44<02:38,  6.36s/it]
Test accuracy: 1.0:  22%|██▏       | 7/32 [00:44<02:38,  6.35s/it]
Test accuracy: 1.0:  22%|██▏       | 7/32 [00:44<02:38,  6.35s/it]
Test accuracy: 0.96:  25%|██▌       | 8/32 [00:51<02:32,  6.34s/it]
Test accuracy: 0.98:  25%|██▌       | 8/32 [00:51<02:32,  6.34s/it]
Test accuracy: 0.95:  25%|██▌       | 8/32 [00:51<02:32,  6.34s/it]
Test accuracy: 0.97:  25%|██▌       | 8/32 [00:51<02:32,  6.34s/it]
Test accuracy: 0.98:  25%|██▌       | 8/32 [00:51<02:32,  6.34s/it]
Test accuracy: 1.0:  25%|██▌       | 8/32 [00:51<02:32,  6.33s/it]
Test accuracy: 0.95:  25%|██▌       | 8/32 [00:51<02:32,  6.34s/it]
Test accuracy: 1.0:  25%|██▌       | 8/32 [00:51<02:32,  6.33s/it]
Test accuracy: 0.95:  28%|██▊       | 9/32 [00:58<02:28,  6.47s/it]
Test accuracy: 0.98:  28%|██▊       | 9/32 [00:58<02:28,  6.47s/it]
Test accuracy: 0.94:  28%|██▊       | 9/32 [00:58<02:28,  6.47s/it]
Test accuracy: 0.99:  28%|██▊       | 9/32 [00:58<02:28,  6.47s/it]
Test accuracy: 1.0:  28%|██▊       | 9/32 [00:57<02:28,  6.47s/it]
Test accuracy: 1.0:  28%|██▊       | 9/32 [00:57<02:28,  6.47s/it]
Test accuracy: 0.97:  28%|██▊       | 9/32 [00:58<02:28,  6.47s/it]
Test accuracy: 0.97:  28%|██▊       | 9/32 [00:58<02:28,  6.47s/it]
Test accuracy: 0.99:  31%|███▏      | 10/32 [01:04<02:21,  6.43s/it]
Test accuracy: 0.94:  31%|███▏      | 10/32 [01:04<02:21,  6.43s/it]
Test accuracy: 0.97:  31%|███▏      | 10/32 [01:04<02:21,  6.43s/it]
Test accuracy: 0.99:  31%|███▏      | 10/32 [01:04<02:21,  6.43s/it]
Test accuracy: 0.96:  31%|███▏      | 10/32 [01:04<02:21,  6.43s/it]
Test accuracy: 0.95:  31%|███▏      | 10/32 [01:04<02:21,  6.43s/it]
Test accuracy: 1.0:  31%|███▏      | 10/32 [01:04<02:21,  6.43s/it]
Test accuracy: 1.0:  31%|███▏      | 10/32 [01:04<02:21,  6.43s/it]
Test accuracy: 0.95:  34%|███▍      | 11/32 [01:10<02:14,  6.39s/it]
Test accuracy: 0.99:  34%|███▍      | 11/32 [01:10<02:14,  6.39s/it]
Test accuracy: 0.96:  34%|███▍      | 11/32 [01:10<02:14,  6.39s/it]
Test accuracy: 0.99:  34%|███▍      | 11/32 [01:10<02:14,  6.39s/it]
Test accuracy: 0.95:  34%|███▍      | 11/32 [01:10<02:14,  6.39s/it]
Test accuracy: 1.0:  34%|███▍      | 11/32 [01:10<02:14,  6.39s/it]
Test accuracy: 0.97:  34%|███▍      | 11/32 [01:10<02:14,  6.39s/it]
Test accuracy: 1.0:  34%|███▍      | 11/32 [01:10<02:14,  6.39s/it]
Test accuracy: 0.95:  38%|███▊      | 12/32 [01:16<02:06,  6.31s/it]
Test accuracy: 0.99:  38%|███▊      | 12/32 [01:16<02:06,  6.31s/it]
Test accuracy: 0.97:  38%|███▊      | 12/32 [01:16<02:06,  6.31s/it]
Test accuracy: 1.0:  38%|███▊      | 12/32 [01:16<02:06,  6.31s/it]
Test accuracy: 0.96:  38%|███▊      | 12/32 [01:16<02:06,  6.31s/it]
Test accuracy: 0.99:  38%|███▊      | 12/32 [01:16<02:06,  6.31s/it]
Test accuracy: 0.97:  38%|███▊      | 12/32 [01:16<02:06,  6.31s/it]
Test accuracy: 1.0:  38%|███▊      | 12/32 [01:16<02:06,  6.31s/it]
Test accuracy: 0.96:  41%|████      | 13/32 [01:23<02:00,  6.32s/it]
Test accuracy: 0.97:  41%|████      | 13/32 [01:23<02:00,  6.32s/it]
Test accuracy: 0.99:  41%|████      | 13/32 [01:23<02:00,  6.32s/it]
Test accuracy: 0.96:  41%|████      | 13/32 [01:23<02:00,  6.32s/it]
Test accuracy: 0.96:  41%|████      | 13/32 [01:23<02:00,  6.32s/it]
Test accuracy: 0.99:  41%|████      | 13/32 [01:23<02:00,  6.32s/it]
Test accuracy: 1.0:  41%|████      | 13/32 [01:22<02:00,  6.32s/it]
Test accuracy: 1.0:  41%|████      | 13/32 [01:22<02:00,  6.32s/it]
Test accuracy: 0.96:  44%|████▍     | 14/32 [01:29<01:53,  6.31s/it]
Test accuracy: 0.96:  44%|████▍     | 14/32 [01:29<01:53,  6.31s/it]
Test accuracy: 0.99:  44%|████▍     | 14/32 [01:29<01:53,  6.31s/it]
Test accuracy: 1.0:  44%|████▍     | 14/32 [01:29<01:53,  6.31s/it]
Test accuracy: 0.96:  44%|████▍     | 14/32 [01:29<01:53,  6.31s/it]
Test accuracy: 0.96:  44%|████▍     | 14/32 [01:29<01:53,  6.31s/it]
Test accuracy: 0.99:  44%|████▍     | 14/32 [01:29<01:53,  6.31s/it]
Test accuracy: 1.0:  44%|████▍     | 14/32 [01:29<01:53,  6.31s/it]
Test accuracy: 0.96:  47%|████▋     | 15/32 [01:35<01:46,  6.26s/it]
Test accuracy: 0.96:  47%|████▋     | 15/32 [01:35<01:46,  6.26s/it]
Test accuracy: 0.97:  47%|████▋     | 15/32 [01:35<01:46,  6.26s/it]
Test accuracy: 1.0:  47%|████▋     | 15/32 [01:35<01:46,  6.26s/it]
Test accuracy: 0.96:  47%|████▋     | 15/32 [01:35<01:46,  6.26s/it]
Test accuracy: 0.99:  47%|████▋     | 15/32 [01:35<01:46,  6.26s/it]
Test accuracy: 1.0:  47%|████▋     | 15/32 [01:35<01:46,  6.26s/it]
Test accuracy: 0.99:  47%|████▋     | 15/32 [01:35<01:46,  6.26s/it]
Test accuracy: 0.96:  50%|█████     | 16/32 [01:41<01:40,  6.28s/it]
Test accuracy: 0.96:  50%|█████     | 16/32 [01:41<01:40,  6.28s/it]
Test accuracy: 0.99:  50%|█████     | 16/32 [01:41<01:40,  6.28s/it]
Test accuracy: 0.98:  50%|█████     | 16/32 [01:41<01:40,  6.28s/it]
Test accuracy: 0.97:  50%|█████     | 16/32 [01:41<01:40,  6.28s/it]
Test accuracy: 0.97:  50%|█████     | 16/32 [01:41<01:40,  6.28s/it]
Test accuracy: 1.0:  50%|█████     | 16/32 [01:41<01:40,  6.28s/it]
Test accuracy: 1.0:  50%|█████     | 16/32 [01:41<01:40,  6.28s/it]
Test accuracy: 0.98:  53%|█████▎    | 17/32 [01:48<01:34,  6.29s/it]
Test accuracy: 1.0:  53%|█████▎    | 17/32 [01:48<01:34,  6.29s/it]
Test accuracy: 0.99:  53%|█████▎    | 17/32 [01:48<01:34,  6.29s/it]
Test accuracy: 0.96:  53%|█████▎    | 17/32 [01:48<01:34,  6.29s/it]
Test accuracy: 0.99:  53%|█████▎    | 17/32 [01:48<01:34,  6.29s/it]
Test accuracy: 0.96:  53%|█████▎    | 17/32 [01:48<01:34,  6.29s/it]
Test accuracy: 0.97:  53%|█████▎    | 17/32 [01:48<01:34,  6.29s/it]
Test accuracy: 0.97:  53%|█████▎    | 17/32 [01:48<01:34,  6.29s/it]
Test accuracy: 0.97:  56%|█████▋    | 18/32 [01:54<01:28,  6.31s/it]
Test accuracy: 0.96:  56%|█████▋    | 18/32 [01:54<01:28,  6.31s/it]
Test accuracy: 0.99:  56%|█████▋    | 18/32 [01:54<01:28,  6.31s/it]
Test accuracy: 0.96:  56%|█████▋    | 18/32 [01:54<01:28,  6.31s/it]
Test accuracy: 0.97:  56%|█████▋    | 18/32 [01:54<01:28,  6.31s/it]
Test accuracy: 0.98:  56%|█████▋    | 18/32 [01:54<01:28,  6.31s/it]
Test accuracy: 0.99:  56%|█████▋    | 18/32 [01:54<01:28,  6.31s/it]
Test accuracy: 0.99:  56%|█████▋    | 18/32 [01:54<01:28,  6.31s/it]
Test accuracy: 0.98:  59%|█████▉    | 19/32 [02:00<01:21,  6.26s/it]
Test accuracy: 0.96:  59%|█████▉    | 19/32 [02:00<01:21,  6.26s/it]
Test accuracy: 0.98:  59%|█████▉    | 19/32 [02:00<01:21,  6.26s/it]
Test accuracy: 0.97:  59%|█████▉    | 19/32 [02:00<01:21,  6.26s/it]
Test accuracy: 0.96:  59%|█████▉    | 19/32 [02:00<01:21,  6.26s/it]
Test accuracy: 0.97:  59%|█████▉    | 19/32 [02:00<01:21,  6.26s/it]
Test accuracy: 0.99:  59%|█████▉    | 19/32 [02:00<01:21,  6.26s/it]
Test accuracy: 0.99:  59%|█████▉    | 19/32 [02:00<01:21,  6.26s/it]
Test accuracy: 0.96:  62%|██████▎   | 20/32 [02:07<01:15,  6.28s/it]
Test accuracy: 0.97:  62%|██████▎   | 20/32 [02:07<01:15,  6.28s/it]
Test accuracy: 0.98:  62%|██████▎   | 20/32 [02:07<01:15,  6.28s/it]
Test accuracy: 0.99:  62%|██████▎   | 20/32 [02:06<01:15,  6.28s/it]
Test accuracy: 0.96:  62%|██████▎   | 20/32 [02:07<01:15,  6.28s/it]
Test accuracy: 0.98:  62%|██████▎   | 20/32 [02:07<01:15,  6.28s/it]
Test accuracy: 0.98:  62%|██████▎   | 20/32 [02:07<01:15,  6.28s/it]
Test accuracy: 0.99:  62%|██████▎   | 20/32 [02:06<01:15,  6.28s/it]
Test accuracy: 0.96:  66%|██████▌   | 21/32 [02:14<01:11,  6.51s/it]
Test accuracy: 0.98:  66%|██████▌   | 21/32 [02:14<01:11,  6.51s/it]
Test accuracy: 0.96:  66%|██████▌   | 21/32 [02:14<01:11,  6.51s/it]
Test accuracy: 0.98:  66%|██████▌   | 21/32 [02:14<01:11,  6.51s/it]
Test accuracy: 0.99:  66%|██████▌   | 21/32 [02:13<01:11,  6.51s/it]
Test accuracy: 0.99:  66%|██████▌   | 21/32 [02:13<01:11,  6.51s/it]
Test accuracy: 0.98:  66%|██████▌   | 21/32 [02:14<01:11,  6.51s/it]
Test accuracy: 0.98:  66%|██████▌   | 21/32 [02:14<01:11,  6.51s/it]
Test accuracy: 0.98:  69%|██████▉   | 22/32 [02:20<01:04,  6.40s/it]
Test accuracy: 0.98:  69%|██████▉   | 22/32 [02:20<01:04,  6.40s/it]
Test accuracy: 0.98:  69%|██████▉   | 22/32 [02:20<01:04,  6.40s/it]
Test accuracy: 0.98:  69%|██████▉   | 22/32 [02:20<01:04,  6.40s/it]
Test accuracy: 0.99:  69%|██████▉   | 22/32 [02:20<01:04,  6.40s/it]
Test accuracy: 0.96:  69%|██████▉   | 22/32 [02:20<01:04,  6.40s/it]
Test accuracy: 0.96:  69%|██████▉   | 22/32 [02:20<01:04,  6.40s/it]
Test accuracy: 0.99:  69%|██████▉   | 22/32 [02:20<01:04,  6.40s/it]
Test accuracy: 0.96:  72%|███████▏  | 23/32 [02:26<00:57,  6.38s/it]
Test accuracy: 0.98:  72%|███████▏  | 23/32 [02:26<00:57,  6.38s/it]
Test accuracy: 0.98:  72%|███████▏  | 23/32 [02:26<00:57,  6.38s/it]
Test accuracy: 0.96:  72%|███████▏  | 23/32 [02:26<00:57,  6.38s/it]
Test accuracy: 0.99:  72%|███████▏  | 23/32 [02:26<00:57,  6.38s/it]
Test accuracy: 0.99:  72%|███████▏  | 23/32 [02:26<00:57,  6.38s/it]
Test accuracy: 0.98:  72%|███████▏  | 23/32 [02:26<00:57,  6.38s/it]
Test accuracy: 0.98:  72%|███████▏  | 23/32 [02:26<00:57,  6.38s/it]
Test accuracy: 0.98:  75%|███████▌  | 24/32 [02:33<00:52,  6.59s/it]
Test accuracy: 0.96:  75%|███████▌  | 24/32 [02:33<00:52,  6.59s/it]
Test accuracy: 0.98:  75%|███████▌  | 24/32 [02:33<00:52,  6.59s/it]
Test accuracy: 0.96:  75%|███████▌  | 24/32 [02:33<00:52,  6.59s/it]
Test accuracy: 0.97:  75%|███████▌  | 24/32 [02:33<00:52,  6.59s/it]
Test accuracy: 0.98:  75%|███████▌  | 24/32 [02:33<00:52,  6.59s/it]
Test accuracy: 0.99:  75%|███████▌  | 24/32 [02:33<00:52,  6.59s/it]
Test accuracy: 0.99:  75%|███████▌  | 24/32 [02:33<00:52,  6.59s/it]
Test accuracy: 0.97:  78%|███████▊  | 25/32 [02:39<00:45,  6.51s/it]
Test accuracy: 0.96:  78%|███████▊  | 25/32 [02:39<00:45,  6.51s/it]
Test accuracy: 0.97:  78%|███████▊  | 25/32 [02:39<00:45,  6.51s/it]
Test accuracy: 0.96:  78%|███████▊  | 25/32 [02:39<00:45,  6.51s/it]
Test accuracy: 0.98:  78%|███████▊  | 25/32 [02:39<00:45,  6.51s/it]
Test accuracy: 0.98:  78%|███████▊  | 25/32 [02:39<00:45,  6.51s/it]
Test accuracy: 0.99:  78%|███████▊  | 25/32 [02:39<00:45,  6.51s/it]
Test accuracy: 0.99:  78%|███████▊  | 25/32 [02:39<00:45,  6.51s/it]
Test accuracy: 0.97:  81%|████████▏ | 26/32 [02:46<00:38,  6.46s/it]
Test accuracy: 0.96:  81%|████████▏ | 26/32 [02:46<00:38,  6.46s/it]
Test accuracy: 0.98:  81%|████████▏ | 26/32 [02:46<00:38,  6.46s/it]
Test accuracy: 0.97:  81%|████████▏ | 26/32 [02:46<00:38,  6.46s/it]
Test accuracy: 0.96:  81%|████████▏ | 26/32 [02:46<00:38,  6.46s/it]
Test accuracy: 0.98:  81%|████████▏ | 26/32 [02:46<00:38,  6.46s/it]
Test accuracy: 0.99:  81%|████████▏ | 26/32 [02:46<00:38,  6.46s/it]
Test accuracy: 0.99:  81%|████████▏ | 26/32 [02:46<00:38,  6.46s/it]
Test accuracy: 0.96:  84%|████████▍ | 27/32 [02:52<00:31,  6.37s/it]
Test accuracy: 0.97:  84%|████████▍ | 27/32 [02:52<00:31,  6.37s/it]
Test accuracy: 0.96:  84%|████████▍ | 27/32 [02:52<00:31,  6.37s/it]
Test accuracy: 0.97:  84%|████████▍ | 27/32 [02:52<00:31,  6.37s/it]
Test accuracy: 0.98:  84%|████████▍ | 27/32 [02:52<00:31,  6.37s/it]
Test accuracy: 0.99:  84%|████████▍ | 27/32 [02:52<00:31,  6.37s/it]
Test accuracy: 0.98:  84%|████████▍ | 27/32 [02:52<00:31,  6.37s/it]
Test accuracy: 0.99:  84%|████████▍ | 27/32 [02:52<00:31,  6.37s/it]
Test accuracy: 0.97:  88%|████████▊ | 28/32 [02:59<00:26,  6.56s/it]
Test accuracy: 0.96:  88%|████████▊ | 28/32 [02:59<00:26,  6.56s/it]
Test accuracy: 0.97:  88%|████████▊ | 28/32 [02:59<00:26,  6.56s/it]
Test accuracy: 0.96:  88%|████████▊ | 28/32 [02:59<00:26,  6.56s/it]
Test accuracy: 0.99:  88%|████████▊ | 28/32 [02:59<00:26,  6.56s/it]
Test accuracy: 0.99:  88%|████████▊ | 28/32 [02:59<00:26,  6.56s/it]
Test accuracy: 0.98:  88%|████████▊ | 28/32 [02:59<00:26,  6.57s/it]
Test accuracy: 0.98:  88%|████████▊ | 28/32 [02:59<00:26,  6.57s/it]
Test accuracy: 0.96:  91%|█████████ | 29/32 [03:05<00:19,  6.50s/it]
Test accuracy: 0.98:  91%|█████████ | 29/32 [03:05<00:19,  6.50s/it]
Test accuracy: 0.97:  91%|█████████ | 29/32 [03:05<00:19,  6.50s/it]
Test accuracy: 0.98:  91%|█████████ | 29/32 [03:05<00:19,  6.50s/it]
Test accuracy: 0.97:  91%|█████████ | 29/32 [03:05<00:19,  6.50s/it]
Test accuracy: 0.97:  91%|█████████ | 29/32 [03:05<00:19,  6.50s/it]
Test accuracy: 0.99:  91%|█████████ | 29/32 [03:05<00:19,  6.50s/it]
Test accuracy: 0.99:  91%|█████████ | 29/32 [03:05<00:19,  6.50s/it]
Test accuracy: 0.98:  94%|█████████▍| 30/32 [03:11<00:12,  6.39s/it]
Test accuracy: 0.97:  94%|█████████▍| 30/32 [03:11<00:12,  6.39s/it]
Test accuracy: 0.97:  94%|█████████▍| 30/32 [03:11<00:12,  6.39s/it]
Test accuracy: 0.98:  94%|█████████▍| 30/32 [03:11<00:12,  6.39s/it]
Test accuracy: 0.96:  94%|█████████▍| 30/32 [03:11<00:12,  6.39s/it]
Test accuracy: 0.97:  94%|█████████▍| 30/32 [03:11<00:12,  6.39s/it]
Test accuracy: 0.99:  94%|█████████▍| 30/32 [03:11<00:12,  6.39s/it]
Test accuracy: 0.99:  94%|█████████▍| 30/32 [03:11<00:12,  6.39s/it]
Test accuracy: 0.97:  97%|█████████▋| 31/32 [03:18<00:06,  6.58s/it]
Test accuracy: 0.97:  97%|█████████▋| 31/32 [03:18<00:06,  6.58s/it]
Test accuracy: 0.99:  97%|█████████▋| 31/32 [03:18<00:06,  6.58s/it]
Test accuracy: 0.96:  97%|█████████▋| 31/32 [03:18<00:06,  6.58s/it]
Test accuracy: 0.96:  97%|█████████▋| 31/32 [03:18<00:06,  6.58s/it]
Test accuracy: 0.98:  97%|█████████▋| 31/32 [03:18<00:06,  6.58s/it]
Test accuracy: 0.98:  97%|█████████▋| 31/32 [03:18<00:06,  6.59s/it]
Test accuracy: 0.98:  97%|█████████▋| 31/32 [03:18<00:06,  6.59s/it]
Test accuracy: 0.98: 100%|██████████| 32/32 [03:21<00:00,  5.28s/it]
Test accuracy: 0.98: 100%|██████████| 32/32 [03:21<00:00,  5.28s/it]
Test accuracy: 0.96: 100%|██████████| 32/32 [03:21<00:00,  5.28s/it]
Test accuracy: 0.98: 100%|██████████| 32/32 [03:21<00:00,  5.28s/it]
Test accuracy: 0.97: 100%|██████████| 32/32 [03:21<00:00,  5.28s/it]
Test accuracy: 0.96: 100%|██████████| 32/32 [03:21<00:00,  5.28s/it]
Test accuracy: 0.98: 100%|██████████| 32/32 [03:21<00:00,  5.28s/it]
Test accuracy: 0.97: 100%|██████████| 32/32 [03:21<00:00,  5.28s/it]
Test accuracy: 0.98: 100%|██████████| 32/32 [03:21<00:00,  6.29s/it]
Device 2: Cor=246, CoT=0, Total=250

Test accuracy: 0.96: 100%|██████████| 32/32 [03:21<00:00,  6.29s/it]
Device 3: Cor=241, CoT=0, Total=250

Test accuracy: 0.97: 100%|██████████| 32/32 [03:21<00:00,  6.29s/it]
Device 1: Cor=242, CoT=0, Total=250

Test accuracy: 0.98: 100%|██████████| 32/32 [03:21<00:00,  6.29s/it]
Device 0: Cor=246, CoT=0, Total=250

============================================================
EVALUATION RESULTS (Epoch 16)
============================================================
Config: data_curriculum=ON, thought_curriculum=OFF
thought_stage=10, data_stage=10, n_latent_tokens=10
Training on difficulties: 0-10

Overall accuracy: 975 / 1000 = 0.9750

Per-difficulty accuracy:
  diff=1: 98/100 = 0.9800 (trained)
  diff=2: 98/100 = 0.9800 (trained)
  diff=3: 96/100 = 0.9600 (trained)
  diff=4: 100/100 = 1.0000 (trained)
  diff=5: 100/100 = 1.0000 (trained)
  diff=6: 98/100 = 0.9800 (trained)
  diff=7: 99/100 = 0.9900 (trained)
  diff=8: 96/100 = 0.9600 (trained)
  diff=9: 94/100 = 0.9400 (trained)
  diff=10: 96/100 = 0.9600 (trained)

Trained (diff 0-10): 975/1000 = 0.9750
============================================================


Processing Files (0 / 0)      : |          |  0.00B /  0.00B            

New Data Upload               : |          |  0.00B /  0.00B            


  ...ve-03b89ee9/checkpoint_17:  10%|█         |  368MB / 3.63GB            


  ...ve-03b89ee9/checkpoint_17:  10%|█         |  368MB / 3.63GB            
Processing Files (0 / 1)      :  10%|█         |  368MB / 3.63GB,   ???B/s  


  ...ve-03b89ee9/checkpoint_17:  20%|██        |  728MB / 3.63GB            
Processing Files (0 / 1)      :  20%|██        |  728MB / 3.63GB,   ???B/s  


  ...ve-03b89ee9/checkpoint_17:  30%|███       | 1.10GB / 3.63GB            
Processing Files (0 / 1)      :  30%|███       | 1.10GB / 3.63GB,  105MB/s  


  ...ve-03b89ee9/checkpoint_17:  41%|████      | 1.47GB / 3.63GB            
Processing Files (0 / 1)      :  41%|████      | 1.47GB / 3.63GB,  139MB/s  


  ...ve-03b89ee9/checkpoint_17:  51%|█████     | 1.85GB / 3.63GB            
Processing Files (0 / 1)      :  51%|█████     | 1.85GB / 3.63GB,  172MB/s  


  ...ve-03b89ee9/checkpoint_17:  61%|██████▏   | 2.22GB / 3.63GB            
Processing Files (0 / 1)      :  61%|██████▏   | 2.22GB / 3.63GB,  204MB/s  


  ...ve-03b89ee9/checkpoint_17:  72%|███████▏  | 2.60GB / 3.63GB            
Processing Files (0 / 1)      :  72%|███████▏  | 2.60GB / 3.63GB,  235MB/s  


  ...ve-03b89ee9/checkpoint_17:  82%|████████▏ | 2.97GB / 3.63GB            
Processing Files (0 / 1)      :  82%|████████▏ | 2.97GB / 3.63GB,  264MB/s  


  ...ve-03b89ee9/checkpoint_17:  92%|█████████▏| 3.34GB / 3.63GB            
Processing Files (0 / 1)      :  92%|█████████▏| 3.34GB / 3.63GB,  292MB/s  


  ...ve-03b89ee9/checkpoint_17: 100%|██████████| 3.63GB / 3.63GB            
Processing Files (1 / 1)      : 100%|██████████| 3.63GB / 3.63GB,  313MB/s  
Processing Files (1 / 1)      : 100%|██████████| 3.63GB / 3.63GB,  313MB/s  

New Data Upload               : |          |  0.00B /  0.00B,  0.00B/s  

  ...ve-03b89ee9/checkpoint_17: 100%|██████████| 3.63GB / 3.63GB            
No files have been modified since last commit. Skipping to prevent empty commit.
Uploaded checkpoint_17 to seyedparsa/thoughtformer-checkpoints/checkpoints/star-datacurr-only-4gpu-bs32-lr1e-5_n1000_qwen3-0.6b-base_thoughtformer_lr1e-5_no-reset_dc_no-tc_adaptive-03b89ee9/checkpoint_17

============================================================
EPOCH 17/100 (thought_stage=10, data_stage=10)
============================================================
  thought_stage=10, c_thought=1, max_difficulty=10

Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):  16%|█▌        | 156/1000 [00:00<00:00, 935.56 examples/s]
Map (num_proc=4):  16%|█▌        | 156/1000 [00:00<00:00, 896.04 examples/s]
Map (num_proc=4):  16%|█▌        | 156/1000 [00:00<00:00, 904.60 examples/s]
Map (num_proc=4):  81%|████████  | 811/1000 [00:00<00:00, 3404.36 examples/s]
Map (num_proc=4):  81%|████████  | 812/1000 [00:00<00:00, 3389.52 examples/s]
Map (num_proc=4):  81%|████████  | 811/1000 [00:00<00:00, 2955.44 examples/s]
Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2487.33 examples/s]

Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2359.75 examples/s]

Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2375.97 examples/s]

Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):  13%|█▎        | 134/1000 [00:00<00:01, 470.68 examples/s]
Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Map (num_proc=4):  70%|███████   | 701/1000 [00:00<00:00, 2190.49 examples/s]
Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 1202.85 examples/s]

Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:04, 1862.97 examples/s]
Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:05, 1787.85 examples/s]
Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:05, 1625.47 examples/s]
Filter (num_proc=4):  20%|██        | 2000/10000 [00:00<00:02, 3099.03 examples/s]
Filter (num_proc=4):  20%|██        | 2000/10000 [00:00<00:02, 3080.33 examples/s]
Filter (num_proc=4):  20%|██        | 2000/10000 [00:00<00:02, 3016.08 examples/s]
Filter (num_proc=4):  40%|████      | 4000/10000 [00:00<00:00, 6054.14 examples/s]
Filter (num_proc=4):  30%|███       | 3000/10000 [00:00<00:01, 4338.67 examples/s]
Filter (num_proc=4):  40%|████      | 4000/10000 [00:00<00:01, 5842.95 examples/s]
Filter (num_proc=4):  65%|██████▌   | 6500/10000 [00:00<00:00, 9398.32 examples/s]
Filter (num_proc=4):  50%|█████     | 5000/10000 [00:00<00:00, 7569.12 examples/s]
Filter (num_proc=4):  65%|██████▌   | 6500/10000 [00:01<00:00, 9176.47 examples/s]
Filter (num_proc=4):  70%|███████   | 7000/10000 [00:01<00:00, 10101.57 examples/s]
Filter (num_proc=4):  80%|████████  | 8000/10000 [00:01<00:00, 10262.90 examples/s]
Filter (num_proc=4):  85%|████████▌ | 8500/10000 [00:01<00:00, 9386.36 examples/s]
Filter (num_proc=4):  85%|████████▌ | 8500/10000 [00:01<00:00, 10524.93 examples/s]
Filter (num_proc=4):  95%|█████████▌| 9500/10000 [00:01<00:00, 10514.17 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 9801.86 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 10735.98 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 6742.26 examples/s]

Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 6671.48 examples/s] 

Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 6540.93 examples/s]

Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:06, 1355.45 examples/s]
Filter (num_proc=4):  20%|██        | 2000/10000 [00:00<00:03, 2508.47 examples/s]
Filter (num_proc=4):  40%|████      | 4000/10000 [00:01<00:01, 5024.54 examples/s]
Filter (num_proc=4):  65%|██████▌   | 6500/10000 [00:01<00:00, 7863.37 examples/s]
Filter (num_proc=4):  85%|████████▌ | 8500/10000 [00:01<00:00, 8664.80 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 8905.09 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 5419.05 examples/s]

Map (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Map (num_proc=4):   1%|▏         | 129/10000 [00:01<01:29, 110.88 examples/s]
Map (num_proc=4):   5%|▌         | 545/10000 [00:01<00:17, 530.46 examples/s]
Map (num_proc=4):   8%|▊         | 820/10000 [00:01<00:11, 819.15 examples/s]
Map (num_proc=4):  11%|█         | 1085/10000 [00:01<00:10, 815.33 examples/s]
Map (num_proc=4):  15%|█▍        | 1484/10000 [00:01<00:06, 1263.80 examples/s]
Map (num_proc=4):  19%|█▉        | 1937/10000 [00:01<00:04, 1803.29 examples/s]
Map (num_proc=4):  25%|██▌       | 2520/10000 [00:02<00:02, 2561.48 examples/s]
Map (num_proc=4):  30%|██▉       | 2992/10000 [00:02<00:03, 2074.29 examples/s]
Map (num_proc=4):  37%|███▋      | 3718/10000 [00:02<00:02, 2931.94 examples/s]
Map (num_proc=4):  46%|████▋     | 4630/10000 [00:02<00:01, 4145.62 examples/s]
Map (num_proc=4):  54%|█████▎    | 5363/10000 [00:02<00:01, 4009.37 examples/s]
Map (num_proc=4):  59%|█████▉    | 5908/10000 [00:03<00:01, 2968.04 examples/s]
Map (num_proc=4):  64%|██████▍   | 6389/10000 [00:03<00:01, 2430.66 examples/s]
Map (num_proc=4):  79%|███████▊  | 7864/10000 [00:03<00:00, 3343.86 examples/s]
Map (num_proc=4):  83%|████████▎ | 8282/10000 [00:03<00:00, 2778.64 examples/s]
Map (num_proc=4):  95%|█████████▌| 9500/10000 [00:04<00:00, 3044.57 examples/s]
Map (num_proc=4):  99%|█████████▉| 9948/10000 [00:04<00:00, 2676.84 examples/s]
Map (num_proc=4): 100%|██████████| 10000/10000 [00:05<00:00, 1829.67 examples/s]

Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 1309.80 examples/s]
Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 1349.49 examples/s]
Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 1335.07 examples/s]
Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 778.70 examples/s]
Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2589.77 examples/s]

Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2493.21 examples/s]

Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2565.24 examples/s]

Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 1456.72 examples/s]

Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):  14%|█▎        | 137/1000 [00:00<00:02, 387.08 examples/s]
Map (num_proc=4):  39%|███▊      | 387/1000 [00:00<00:00, 956.28 examples/s]
Map (num_proc=4):  85%|████████▌ | 853/1000 [00:00<00:00, 1874.51 examples/s]
Map (num_proc=4): 100%|██████████| 1000/1000 [00:01<00:00, 945.14 examples/s]

Training Epoch: 18:   0%|          | 0/312 [00:00<?, ?it/s]
Training Epoch: 18:   0%|          | 0/312 [00:00<?, ?it/s]
Training Epoch: 18:   0%|          | 0/312 [00:00<?, ?it/s]
Training Epoch: 18:   0%|          | 0/312 [00:00<?, ?it/s]logging training data
rank 0 | gpu 0 | alloc=4.20 GB | reserved=47.73 GB | max=41.99 GB
rank 1 | gpu 1 | alloc=4.20 GB | reserved=47.73 GB | max=41.98 GB
rank 3 | gpu 3 | alloc=4.20 GB | reserved=47.73 GB | max=41.99 GB
rank 2 | gpu 2 | alloc=4.20 GB | reserved=49.61 GB | max=41.98 GB

Training Epoch: 18/100, batch 0/625 completed (loss: 0.0216:   0%|          | 0/312 [00:03<?, ?it/s]
Training Epoch: 18/100, batch 0/625 completed (loss: 0.0044:   0%|          | 0/312 [00:03<?, ?it/s]
Training Epoch: 18/100, batch 0/625 completed (loss: 0.0043:   0%|          | 0/312 [00:03<?, ?it/s]
Training Epoch: 18/100, batch 0/625 completed (loss: 0.0021:   0%|          | 0/312 [00:03<?, ?it/s]
Training Epoch: 18/100, batch 0/625 completed (loss: 0.0021:   0%|          | 1/312 [00:06<32:36,  6.29s/it]
Training Epoch: 18/100, batch 0/625 completed (loss: 0.0216:   0%|          | 1/312 [00:06<32:18,  6.23s/it]
Training Epoch: 18/100, batch 0/625 completed (loss: 0.0044:   0%|          | 1/312 [00:06<32:23,  6.25s/it]
Training Epoch: 18/100, batch 0/625 completed (loss: 0.0043:   0%|          | 1/312 [00:06<32:24,  6.25s/it]
Training Epoch: 18/100, batch 1/625 completed (loss: 0.0731:   0%|          | 1/312 [00:06<32:23,  6.25s/it]
Training Epoch: 18/100, batch 1/625 completed (loss: 0.004:   0%|          | 1/312 [00:06<32:24,  6.25s/it] 
Training Epoch: 18/100, batch 1/625 completed (loss: 0.0021:   0%|          | 1/312 [00:06<32:18,  6.23s/it]
Training Epoch: 18/100, batch 1/625 completed (loss: 0.008:   0%|          | 1/312 [00:06<32:36,  6.29s/it] 
Training Epoch: 18/100, batch 2/625 completed (loss: 0.0011:   0%|          | 1/312 [00:09<32:18,  6.23s/it]
Training Epoch: 18/100, batch 2/625 completed (loss: 0.0024:   0%|          | 1/312 [00:09<32:23,  6.25s/it]
Training Epoch: 18/100, batch 2/625 completed (loss: 0.0102:   0%|          | 1/312 [00:09<32:24,  6.25s/it]
Training Epoch: 18/100, batch 2/625 completed (loss: 0.0157:   0%|          | 1/312 [00:09<32:36,  6.29s/it]
Training Epoch: 18/100, batch 2/625 completed (loss: 0.0024:   1%|          | 2/312 [00:11<30:20,  5.87s/it]
Training Epoch: 18/100, batch 2/625 completed (loss: 0.0157:   1%|          | 2/312 [00:11<30:27,  5.89s/it]
Training Epoch: 18/100, batch 2/625 completed (loss: 0.0011:   1%|          | 2/312 [00:11<30:19,  5.87s/it]
Training Epoch: 18/100, batch 2/625 completed (loss: 0.0102:   1%|          | 2/312 [00:11<30:21,  5.88s/it]
Training Epoch: 18/100, batch 3/625 completed (loss: 0.0853:   1%|          | 2/312 [00:11<30:21,  5.88s/it]
Training Epoch: 18/100, batch 3/625 completed (loss: 0.0042:   1%|          | 2/312 [00:11<30:20,  5.87s/it]
Training Epoch: 18/100, batch 3/625 completed (loss: 0.0116:   1%|          | 2/312 [00:11<30:19,  5.87s/it]
Training Epoch: 18/100, batch 3/625 completed (loss: 0.0084:   1%|          | 2/312 [00:11<30:27,  5.89s/it]
Training Epoch: 18/100, batch 4/625 completed (loss: 0.017:   1%|          | 2/312 [00:14<30:20,  5.87s/it] 
Training Epoch: 18/100, batch 4/625 completed (loss: 0.0017:   1%|          | 2/312 [00:14<30:19,  5.87s/it]
Training Epoch: 18/100, batch 4/625 completed (loss: 0.0137:   1%|          | 2/312 [00:14<30:21,  5.88s/it]
Training Epoch: 18/100, batch 4/625 completed (loss: 0.0227:   1%|          | 2/312 [00:14<30:27,  5.89s/it]
Training Epoch: 18/100, batch 4/625 completed (loss: 0.0137:   1%|          | 3/312 [00:17<30:43,  5.96s/it]
Training Epoch: 18/100, batch 4/625 completed (loss: 0.0017:   1%|          | 3/312 [00:17<30:42,  5.96s/it]
Training Epoch: 18/100, batch 4/625 completed (loss: 0.017:   1%|          | 3/312 [00:17<30:43,  5.97s/it]
Training Epoch: 18/100, batch 4/625 completed (loss: 0.0227:   1%|          | 3/312 [00:17<30:47,  5.98s/it]
Training Epoch: 18/100, batch 5/625 completed (loss: 0.0149:   1%|          | 3/312 [00:17<30:43,  5.97s/it]
Training Epoch: 18/100, batch 5/625 completed (loss: 0.024:   1%|          | 3/312 [00:17<30:42,  5.96s/it] 
Training Epoch: 18/100, batch 5/625 completed (loss: 0.0102:   1%|          | 3/312 [00:17<30:43,  5.96s/it]
Training Epoch: 18/100, batch 5/625 completed (loss: 0.0033:   1%|          | 3/312 [00:18<30:47,  5.98s/it]
Training Epoch: 18/100, batch 6/625 completed (loss: 0.0399:   1%|          | 3/312 [00:20<30:43,  5.96s/it]
Training Epoch: 18/100, batch 6/625 completed (loss: 0.0014:   1%|          | 3/312 [00:20<30:42,  5.96s/it]
Training Epoch: 18/100, batch 6/625 completed (loss: 0.0819:   1%|          | 3/312 [00:20<30:43,  5.97s/it]
Training Epoch: 18/100, batch 6/625 completed (loss: 0.0027:   1%|          | 3/312 [00:20<30:47,  5.98s/it]
Training Epoch: 18/100, batch 6/625 completed (loss: 0.0819:   1%|▏         | 4/312 [00:23<29:56,  5.83s/it]
Training Epoch: 18/100, batch 6/625 completed (loss: 0.0014:   1%|▏         | 4/312 [00:23<29:56,  5.83s/it]
Training Epoch: 18/100, batch 6/625 completed (loss: 0.0399:   1%|▏         | 4/312 [00:23<29:56,  5.83s/it]
Training Epoch: 18/100, batch 6/625 completed (loss: 0.0027:   1%|▏         | 4/312 [00:23<29:58,  5.84s/it]
Training Epoch: 18/100, batch 7/625 completed (loss: 0.0105:   1%|▏         | 4/312 [00:23<29:56,  5.83s/it]
Training Epoch: 18/100, batch 7/625 completed (loss: 0.005:   1%|▏         | 4/312 [00:23<29:56,  5.83s/it] 
Training Epoch: 18/100, batch 7/625 completed (loss: 0.0265:   1%|▏         | 4/312 [00:23<29:56,  5.83s/it]
Training Epoch: 18/100, batch 7/625 completed (loss: 0.0012:   1%|▏         | 4/312 [00:23<29:58,  5.84s/it]
Training Epoch: 18/100, batch 8/625 completed (loss: 0.0122:   1%|▏         | 4/312 [00:26<29:56,  5.83s/it]
Training Epoch: 18/100, batch 8/625 completed (loss: 0.0011:   1%|▏         | 4/312 [00:26<29:56,  5.83s/it]
Training Epoch: 18/100, batch 8/625 completed (loss: 0.0055:   1%|▏         | 4/312 [00:26<29:56,  5.83s/it]
Training Epoch: 18/100, batch 8/625 completed (loss: 0.0119:   1%|▏         | 4/312 [00:26<29:58,  5.84s/it]
Training Epoch: 18/100, batch 8/625 completed (loss: 0.0011:   2%|▏         | 5/312 [00:29<29:50,  5.83s/it]
Training Epoch: 18/100, batch 8/625 completed (loss: 0.0122:   2%|▏         | 5/312 [00:29<29:49,  5.83s/it]
Training Epoch: 18/100, batch 8/625 completed (loss: 0.0119:   2%|▏         | 5/312 [00:29<29:51,  5.84s/it]
Training Epoch: 18/100, batch 8/625 completed (loss: 0.0055:   2%|▏         | 5/312 [00:29<29:50,  5.83s/it]
Training Epoch: 18/100, batch 9/625 completed (loss: 0.029:   2%|▏         | 5/312 [00:29<29:50,  5.83s/it] 
Training Epoch: 18/100, batch 9/625 completed (loss: 0.0007:   2%|▏         | 5/312 [00:29<29:49,  5.83s/it]
Training Epoch: 18/100, batch 9/625 completed (loss: 0.0029:   2%|▏         | 5/312 [00:29<29:50,  5.83s/it]
Training Epoch: 18/100, batch 9/625 completed (loss: 0.0032:   2%|▏         | 5/312 [00:29<29:51,  5.84s/it]
Training Epoch: 18/100, batch 10/625 completed (loss: 0.0359:   2%|▏         | 5/312 [00:32<29:50,  5.83s/it]
Training Epoch: 18/100, batch 10/625 completed (loss: 0.0871:   2%|▏         | 5/312 [00:32<29:49,  5.83s/it]
Training Epoch: 18/100, batch 10/625 completed (loss: 0.0176:   2%|▏         | 5/312 [00:32<29:50,  5.83s/it]
Training Epoch: 18/100, batch 10/625 completed (loss: 0.0043:   2%|▏         | 5/312 [00:32<29:51,  5.84s/it]
Training Epoch: 18/100, batch 10/625 completed (loss: 0.0359:   2%|▏         | 6/312 [00:35<29:59,  5.88s/it]
Training Epoch: 18/100, batch 10/625 completed (loss: 0.0871:   2%|▏         | 6/312 [00:35<29:59,  5.88s/it]
Training Epoch: 18/100, batch 10/625 completed (loss: 0.0043:   2%|▏         | 6/312 [00:35<30:00,  5.88s/it]
Training Epoch: 18/100, batch 10/625 completed (loss: 0.0176:   2%|▏         | 6/312 [00:35<30:00,  5.88s/it]
Training Epoch: 18/100, batch 11/625 completed (loss: 0.0261:   2%|▏         | 6/312 [00:35<30:00,  5.88s/it]
Training Epoch: 18/100, batch 11/625 completed (loss: 0.0234:   2%|▏         | 6/312 [00:35<29:59,  5.88s/it]
Training Epoch: 18/100, batch 11/625 completed (loss: 0.0016:   2%|▏         | 6/312 [00:35<29:59,  5.88s/it]
Training Epoch: 18/100, batch 11/625 completed (loss: 0.002:   2%|▏         | 6/312 [00:35<30:00,  5.88s/it] 
Training Epoch: 18/100, batch 12/625 completed (loss: 0.0017:   2%|▏         | 6/312 [00:38<30:00,  5.88s/it]
Training Epoch: 18/100, batch 12/625 completed (loss: 0.0146:   2%|▏         | 6/312 [00:38<29:59,  5.88s/it]
Training Epoch: 18/100, batch 12/625 completed (loss: 0.0095:   2%|▏         | 6/312 [00:38<29:59,  5.88s/it]
Training Epoch: 18/100, batch 12/625 completed (loss: 0.0054:   2%|▏         | 6/312 [00:38<30:00,  5.88s/it]
Training Epoch: 18/100, batch 12/625 completed (loss: 0.0054:   2%|▏         | 7/312 [00:41<29:32,  5.81s/it]
Training Epoch: 18/100, batch 12/625 completed (loss: 0.0095:   2%|▏         | 7/312 [00:41<29:32,  5.81s/it]
Training Epoch: 18/100, batch 12/625 completed (loss: 0.0146:   2%|▏         | 7/312 [00:41<29:33,  5.81s/it]
Training Epoch: 18/100, batch 12/625 completed (loss: 0.0017:   2%|▏         | 7/312 [00:41<29:33,  5.81s/it]
Training Epoch: 18/100, batch 13/625 completed (loss: 0.0308:   2%|▏         | 7/312 [00:41<29:33,  5.81s/it]
Training Epoch: 18/100, batch 13/625 completed (loss: 0.0035:   2%|▏         | 7/312 [00:41<29:33,  5.81s/it]
Training Epoch: 18/100, batch 13/625 completed (loss: 0.0026:   2%|▏         | 7/312 [00:41<29:32,  5.81s/it]
Training Epoch: 18/100, batch 13/625 completed (loss: 0.0196:   2%|▏         | 7/312 [00:41<29:32,  5.81s/it]
Training Epoch: 18/100, batch 14/625 completed (loss: 0.0138:   2%|▏         | 7/312 [00:43<29:33,  5.81s/it]
Training Epoch: 18/100, batch 14/625 completed (loss: 0.045:   2%|▏         | 7/312 [00:43<29:32,  5.81s/it] 
Training Epoch: 18/100, batch 14/625 completed (loss: 0.0009:   2%|▏         | 7/312 [00:43<29:33,  5.81s/it]
Training Epoch: 18/100, batch 14/625 completed (loss: 0.004:   2%|▏         | 7/312 [00:43<29:32,  5.81s/it] 
Training Epoch: 18/100, batch 14/625 completed (loss: 0.0138:   3%|▎         | 8/312 [00:46<29:30,  5.82s/it]
Training Epoch: 18/100, batch 14/625 completed (loss: 0.045:   3%|▎         | 8/312 [00:46<29:31,  5.83s/it]
Training Epoch: 18/100, batch 14/625 completed (loss: 0.004:   3%|▎         | 8/312 [00:46<29:31,  5.83s/it]
Training Epoch: 18/100, batch 14/625 completed (loss: 0.0009:   3%|▎         | 8/312 [00:46<29:32,  5.83s/it]
Training Epoch: 18/100, batch 15/625 completed (loss: 0.4453:   3%|▎         | 8/312 [00:46<29:32,  5.83s/it]
Training Epoch: 18/100, batch 15/625 completed (loss: 0.0022:   3%|▎         | 8/312 [00:46<29:31,  5.83s/it]
Training Epoch: 18/100, batch 15/625 completed (loss: 0.1156:   3%|▎         | 8/312 [00:46<29:30,  5.82s/it]
Training Epoch: 18/100, batch 15/625 completed (loss: 0.0445:   3%|▎         | 8/312 [00:46<29:31,  5.83s/it]
Training Epoch: 18/100, batch 16/625 completed (loss: 0.0023:   3%|▎         | 8/312 [00:50<29:30,  5.82s/it]
Training Epoch: 18/100, batch 16/625 completed (loss: 0.0029:   3%|▎         | 8/312 [00:50<29:31,  5.83s/it]
Training Epoch: 18/100, batch 16/625 completed (loss: 0.0009:   3%|▎         | 8/312 [00:50<29:32,  5.83s/it]
Training Epoch: 18/100, batch 16/625 completed (loss: 0.0151:   3%|▎         | 8/312 [00:50<29:31,  5.83s/it]
Training Epoch: 18/100, batch 16/625 completed (loss: 0.0151:   3%|▎         | 9/312 [00:52<29:42,  5.88s/it]
Training Epoch: 18/100, batch 16/625 completed (loss: 0.0009:   3%|▎         | 9/312 [00:52<29:41,  5.88s/it]
Training Epoch: 18/100, batch 16/625 completed (loss: 0.0023:   3%|▎         | 9/312 [00:52<29:42,  5.88s/it]
Training Epoch: 18/100, batch 16/625 completed (loss: 0.0029:   3%|▎         | 9/312 [00:52<29:42,  5.88s/it]
Training Epoch: 18/100, batch 17/625 completed (loss: 0.0573:   3%|▎         | 9/312 [00:52<29:41,  5.88s/it]
Training Epoch: 18/100, batch 17/625 completed (loss: 0.0124:   3%|▎         | 9/312 [00:52<29:42,  5.88s/it]
Training Epoch: 18/100, batch 17/625 completed (loss: 0.0037:   3%|▎         | 9/312 [00:52<29:42,  5.88s/it]
Training Epoch: 18/100, batch 17/625 completed (loss: 0.0085:   3%|▎         | 9/312 [00:52<29:42,  5.88s/it]
Training Epoch: 18/100, batch 18/625 completed (loss: 0.0037:   3%|▎         | 9/312 [00:55<29:42,  5.88s/it]
Training Epoch: 18/100, batch 18/625 completed (loss: 0.0039:   3%|▎         | 9/312 [00:55<29:42,  5.88s/it]
Training Epoch: 18/100, batch 18/625 completed (loss: 0.0004:   3%|▎         | 9/312 [00:55<29:41,  5.88s/it]
Training Epoch: 18/100, batch 18/625 completed (loss: 0.0121:   3%|▎         | 9/312 [00:55<29:42,  5.88s/it]
Training Epoch: 18/100, batch 18/625 completed (loss: 0.0004:   3%|▎         | 10/312 [00:58<29:19,  5.83s/it]
Training Epoch: 18/100, batch 18/625 completed (loss: 0.0037:   3%|▎         | 10/312 [00:58<29:20,  5.83s/it]
Training Epoch: 18/100, batch 18/625 completed (loss: 0.0039:   3%|▎         | 10/312 [00:58<29:20,  5.83s/it]
Training Epoch: 18/100, batch 18/625 completed (loss: 0.0121:   3%|▎         | 10/312 [00:58<29:21,  5.83s/it]
Training Epoch: 18/100, batch 19/625 completed (loss: 0.0017:   3%|▎         | 10/312 [00:58<29:20,  5.83s/it]
Training Epoch: 18/100, batch 19/625 completed (loss: 0.0059:   3%|▎         | 10/312 [00:58<29:20,  5.83s/it]
Training Epoch: 18/100, batch 19/625 completed (loss: 0.0012:   3%|▎         | 10/312 [00:58<29:19,  5.83s/it]
Training Epoch: 18/100, batch 19/625 completed (loss: 0.021:   3%|▎         | 10/312 [00:58<29:21,  5.83s/it] 
Training Epoch: 18/100, batch 20/625 completed (loss: 0.0023:   3%|▎         | 10/312 [01:01<29:19,  5.83s/it]
Training Epoch: 18/100, batch 20/625 completed (loss: 0.0118:   3%|▎         | 10/312 [01:01<29:20,  5.83s/it]
Training Epoch: 18/100, batch 20/625 completed (loss: 0.0021:   3%|▎         | 10/312 [01:01<29:20,  5.83s/it]
Training Epoch: 18/100, batch 20/625 completed (loss: 0.0109:   3%|▎         | 10/312 [01:01<29:21,  5.83s/it]
Training Epoch: 18/100, batch 20/625 completed (loss: 0.0109:   4%|▎         | 11/312 [01:05<30:03,  5.99s/it]
Training Epoch: 18/100, batch 20/625 completed (loss: 0.0023:   4%|▎         | 11/312 [01:04<30:03,  5.99s/it]
Training Epoch: 18/100, batch 20/625 completed (loss: 0.0118:   4%|▎         | 11/312 [01:04<30:03,  5.99s/it]
Training Epoch: 18/100, batch 20/625 completed (loss: 0.0021:   4%|▎         | 11/312 [01:04<30:03,  5.99s/it]
Training Epoch: 18/100, batch 21/625 completed (loss: 0.064:   4%|▎         | 11/312 [01:05<30:03,  5.99s/it] 
Training Epoch: 18/100, batch 21/625 completed (loss: 0.0124:   4%|▎         | 11/312 [01:05<30:03,  5.99s/it]
Training Epoch: 18/100, batch 21/625 completed (loss: 0.003:   4%|▎         | 11/312 [01:05<30:03,  5.99s/it] 
Training Epoch: 18/100, batch 21/625 completed (loss: 0.0355:   4%|▎         | 11/312 [01:05<30:03,  5.99s/it]
Training Epoch: 18/100, batch 22/625 completed (loss: 0.0175:   4%|▎         | 11/312 [01:07<30:03,  5.99s/it]
Training Epoch: 18/100, batch 22/625 completed (loss: 0.0252:   4%|▎         | 11/312 [01:07<30:03,  5.99s/it]
Training Epoch: 18/100, batch 22/625 completed (loss: 0.002:   4%|▎         | 11/312 [01:07<30:03,  5.99s/it]
Training Epoch: 18/100, batch 22/625 completed (loss: 0.0409:   4%|▎         | 11/312 [01:07<30:03,  5.99s/it]
Training Epoch: 18/100, batch 22/625 completed (loss: 0.0409:   4%|▍         | 12/312 [01:10<29:36,  5.92s/it]
Training Epoch: 18/100, batch 22/625 completed (loss: 0.0252:   4%|▍         | 12/312 [01:10<29:36,  5.92s/it]
Training Epoch: 18/100, batch 22/625 completed (loss: 0.002:   4%|▍         | 12/312 [01:10<29:37,  5.92s/it]
Training Epoch: 18/100, batch 22/625 completed (loss: 0.0175:   4%|▍         | 12/312 [01:10<29:37,  5.92s/it]
Training Epoch: 18/100, batch 23/625 completed (loss: 0.0265:   4%|▍         | 12/312 [01:10<29:37,  5.92s/it]
Training Epoch: 18/100, batch 23/625 completed (loss: 0.0548:   4%|▍         | 12/312 [01:10<29:37,  5.92s/it]
Training Epoch: 18/100, batch 23/625 completed (loss: 0.0018:   4%|▍         | 12/312 [01:10<29:36,  5.92s/it]
Training Epoch: 18/100, batch 23/625 completed (loss: 0.0073:   4%|▍         | 12/312 [01:10<29:36,  5.92s/it]
Training Epoch: 18/100, batch 24/625 completed (loss: 0.0182:   4%|▍         | 12/312 [01:13<29:36,  5.92s/it]
Training Epoch: 18/100, batch 24/625 completed (loss: 0.0045:   4%|▍         | 12/312 [01:13<29:37,  5.92s/it]
Training Epoch: 18/100, batch 24/625 completed (loss: 0.0106:   4%|▍         | 12/312 [01:13<29:37,  5.92s/it]
Training Epoch: 18/100, batch 24/625 completed (loss: 0.0046:   4%|▍         | 12/312 [01:13<29:36,  5.92s/it]
Training Epoch: 18/100, batch 24/625 completed (loss: 0.0106:   4%|▍         | 13/312 [01:16<29:15,  5.87s/it]
Training Epoch: 18/100, batch 24/625 completed (loss: 0.0045:   4%|▍         | 13/312 [01:16<29:15,  5.87s/it]
Training Epoch: 18/100, batch 24/625 completed (loss: 0.0182:   4%|▍         | 13/312 [01:16<29:15,  5.87s/it]
Training Epoch: 18/100, batch 24/625 completed (loss: 0.0046:   4%|▍         | 13/312 [01:16<29:16,  5.87s/it]
Training Epoch: 18/100, batch 25/625 completed (loss: 0.0054:   4%|▍         | 13/312 [01:16<29:15,  5.87s/it]
Training Epoch: 18/100, batch 25/625 completed (loss: 0.1488:   4%|▍         | 13/312 [01:16<29:15,  5.87s/it]
Training Epoch: 18/100, batch 25/625 completed (loss: 0.0019:   4%|▍         | 13/312 [01:16<29:15,  5.87s/it]
Training Epoch: 18/100, batch 25/625 completed (loss: 0.3914:   4%|▍         | 13/312 [01:16<29:16,  5.87s/it]
Training Epoch: 18/100, batch 26/625 completed (loss: 0.0239:   4%|▍         | 13/312 [01:20<29:15,  5.87s/it]
Training Epoch: 18/100, batch 26/625 completed (loss: 0.0007:   4%|▍         | 13/312 [01:20<29:15,  5.87s/it]
Training Epoch: 18/100, batch 26/625 completed (loss: 0.0395:   4%|▍         | 13/312 [01:20<29:15,  5.87s/it]
Training Epoch: 18/100, batch 26/625 completed (loss: 0.0048:   4%|▍         | 13/312 [01:20<29:16,  5.87s/it]
Training Epoch: 18/100, batch 26/625 completed (loss: 0.0048:   4%|▍         | 14/312 [01:22<29:53,  6.02s/it]
Training Epoch: 18/100, batch 26/625 completed (loss: 0.0239:   4%|▍         | 14/312 [01:22<29:53,  6.02s/it]
Training Epoch: 18/100, batch 26/625 completed (loss: 0.0007:   4%|▍         | 14/312 [01:22<29:54,  6.02s/it]
Training Epoch: 18/100, batch 26/625 completed (loss: 0.0395:   4%|▍         | 14/312 [01:22<29:54,  6.02s/it]
Training Epoch: 18/100, batch 27/625 completed (loss: 0.0619:   4%|▍         | 14/312 [01:22<29:53,  6.02s/it]
Training Epoch: 18/100, batch 27/625 completed (loss: 0.0072:   4%|▍         | 14/312 [01:22<29:54,  6.02s/it]
Training Epoch: 18/100, batch 27/625 completed (loss: 0.0011:   4%|▍         | 14/312 [01:22<29:54,  6.02s/it]
Training Epoch: 18/100, batch 27/625 completed (loss: 0.0022:   4%|▍         | 14/312 [01:22<29:53,  6.02s/it]
Training Epoch: 18/100, batch 28/625 completed (loss: 0.0163:   4%|▍         | 14/312 [01:25<29:54,  6.02s/it]
Training Epoch: 18/100, batch 28/625 completed (loss: 0.0052:   4%|▍         | 14/312 [01:25<29:54,  6.02s/it]
Training Epoch: 18/100, batch 28/625 completed (loss: 0.0723:   4%|▍         | 14/312 [01:25<29:53,  6.02s/it]
Training Epoch: 18/100, batch 28/625 completed (loss: 0.1005:   4%|▍         | 14/312 [01:25<29:53,  6.02s/it]
Training Epoch: 18/100, batch 28/625 completed (loss: 0.0723:   5%|▍         | 15/312 [01:28<29:20,  5.93s/it]
Training Epoch: 18/100, batch 28/625 completed (loss: 0.1005:   5%|▍         | 15/312 [01:28<29:21,  5.93s/it]
Training Epoch: 18/100, batch 28/625 completed (loss: 0.0052:   5%|▍         | 15/312 [01:28<29:20,  5.93s/it]
Training Epoch: 18/100, batch 28/625 completed (loss: 0.0163:   5%|▍         | 15/312 [01:28<29:21,  5.93s/it]
Training Epoch: 18/100, batch 29/625 completed (loss: 0.0012:   5%|▍         | 15/312 [01:28<29:20,  5.93s/it]
Training Epoch: 18/100, batch 29/625 completed (loss: 0.0762:   5%|▍         | 15/312 [01:28<29:20,  5.93s/it]
Training Epoch: 18/100, batch 29/625 completed (loss: 0.0061:   5%|▍         | 15/312 [01:28<29:21,  5.93s/it]
Training Epoch: 18/100, batch 29/625 completed (loss: 0.0682:   5%|▍         | 15/312 [01:28<29:21,  5.93s/it]
Training Epoch: 18/100, batch 30/625 completed (loss: 0.0104:   5%|▍         | 15/312 [01:31<29:20,  5.93s/it]
Training Epoch: 18/100, batch 30/625 completed (loss: 0.0083:   5%|▍         | 15/312 [01:31<29:21,  5.93s/it]
Training Epoch: 18/100, batch 30/625 completed (loss: 0.0061:   5%|▍         | 15/312 [01:31<29:20,  5.93s/it]
Training Epoch: 18/100, batch 30/625 completed (loss: 0.0003:   5%|▍         | 15/312 [01:31<29:21,  5.93s/it]
Training Epoch: 18/100, batch 30/625 completed (loss: 0.0003:   5%|▌         | 16/312 [01:34<29:44,  6.03s/it]
Training Epoch: 18/100, batch 30/625 completed (loss: 0.0083:   5%|▌         | 16/312 [01:34<29:44,  6.03s/it]
Training Epoch: 18/100, batch 30/625 completed (loss: 0.0104:   5%|▌         | 16/312 [01:34<29:44,  6.03s/it]
Training Epoch: 18/100, batch 30/625 completed (loss: 0.0061:   5%|▌         | 16/312 [01:34<29:44,  6.03s/it]
Training Epoch: 18/100, batch 31/625 completed (loss: 0.0289:   5%|▌         | 16/312 [01:34<29:44,  6.03s/it]
Training Epoch: 18/100, batch 31/625 completed (loss: 0.0122:   5%|▌         | 16/312 [01:34<29:44,  6.03s/it]
Training Epoch: 18/100, batch 31/625 completed (loss: 0.0219:   5%|▌         | 16/312 [01:34<29:44,  6.03s/it]
Training Epoch: 18/100, batch 31/625 completed (loss: 0.0288:   5%|▌         | 16/312 [01:34<29:44,  6.03s/it]
Training Epoch: 18/100, batch 32/625 completed (loss: 0.0078:   5%|▌         | 16/312 [01:37<29:44,  6.03s/it]
Training Epoch: 18/100, batch 32/625 completed (loss: 0.0218:   5%|▌         | 16/312 [01:37<29:44,  6.03s/it]
Training Epoch: 18/100, batch 32/625 completed (loss: 0.0178:   5%|▌         | 16/312 [01:37<29:44,  6.03s/it]
Training Epoch: 18/100, batch 32/625 completed (loss: 0.0041:   5%|▌         | 16/312 [01:37<29:44,  6.03s/it]
Training Epoch: 18/100, batch 32/625 completed (loss: 0.0178:   5%|▌         | 17/312 [01:40<29:26,  5.99s/it]
Training Epoch: 18/100, batch 32/625 completed (loss: 0.0218:   5%|▌         | 17/312 [01:40<29:26,  5.99s/it]
Training Epoch: 18/100, batch 32/625 completed (loss: 0.0041:   5%|▌         | 17/312 [01:40<29:27,  5.99s/it]
Training Epoch: 18/100, batch 32/625 completed (loss: 0.0078:   5%|▌         | 17/312 [01:40<29:27,  5.99s/it]
Training Epoch: 18/100, batch 33/625 completed (loss: 0.0675:   5%|▌         | 17/312 [01:40<29:27,  5.99s/it]
Training Epoch: 18/100, batch 33/625 completed (loss: 0.0196:   5%|▌         | 17/312 [01:40<29:26,  5.99s/it]
Training Epoch: 18/100, batch 33/625 completed (loss: 0.0094:   5%|▌         | 17/312 [01:40<29:26,  5.99s/it]
Training Epoch: 18/100, batch 33/625 completed (loss: 0.0074:   5%|▌         | 17/312 [01:40<29:27,  5.99s/it]
Training Epoch: 18/100, batch 34/625 completed (loss: 0.0026:   5%|▌         | 17/312 [01:43<29:26,  5.99s/it]
Training Epoch: 18/100, batch 34/625 completed (loss: 0.0027:   5%|▌         | 17/312 [01:43<29:27,  5.99s/it]
Training Epoch: 18/100, batch 34/625 completed (loss: 0.0011:   5%|▌         | 17/312 [01:43<29:26,  5.99s/it]
Training Epoch: 18/100, batch 34/625 completed (loss: 0.0823:   5%|▌         | 17/312 [01:43<29:27,  5.99s/it]
Training Epoch: 18/100, batch 34/625 completed (loss: 0.0011:   6%|▌         | 18/312 [01:46<29:02,  5.93s/it]
Training Epoch: 18/100, batch 34/625 completed (loss: 0.0027:   6%|▌         | 18/312 [01:46<29:02,  5.93s/it]
Training Epoch: 18/100, batch 34/625 completed (loss: 0.0823:   6%|▌         | 18/312 [01:46<29:03,  5.93s/it]
Training Epoch: 18/100, batch 34/625 completed (loss: 0.0026:   6%|▌         | 18/312 [01:46<29:03,  5.93s/it]
Training Epoch: 18/100, batch 35/625 completed (loss: 0.0005:   6%|▌         | 18/312 [01:46<29:02,  5.93s/it]
Training Epoch: 18/100, batch 35/625 completed (loss: 0.0228:   6%|▌         | 18/312 [01:46<29:03,  5.93s/it]
Training Epoch: 18/100, batch 35/625 completed (loss: 0.0249:   6%|▌         | 18/312 [01:46<29:02,  5.93s/it]
Training Epoch: 18/100, batch 35/625 completed (loss: 0.0098:   6%|▌         | 18/312 [01:46<29:03,  5.93s/it]
Training Epoch: 18/100, batch 36/625 completed (loss: 0.0089:   6%|▌         | 18/312 [01:49<29:03,  5.93s/it]
Training Epoch: 18/100, batch 36/625 completed (loss: 0.0254:   6%|▌         | 18/312 [01:49<29:02,  5.93s/it]
Training Epoch: 18/100, batch 36/625 completed (loss: 0.0005:   6%|▌         | 18/312 [01:49<29:02,  5.93s/it]
Training Epoch: 18/100, batch 36/625 completed (loss: 0.0301:   6%|▌         | 18/312 [01:49<29:03,  5.93s/it]
Training Epoch: 18/100, batch 36/625 completed (loss: 0.0301:   6%|▌         | 19/312 [01:52<29:29,  6.04s/it]
Training Epoch: 18/100, batch 36/625 completed (loss: 0.0005:   6%|▌         | 19/312 [01:52<29:29,  6.04s/it]
Training Epoch: 18/100, batch 36/625 completed (loss: 0.0089:   6%|▌         | 19/312 [01:52<29:29,  6.04s/it]
Training Epoch: 18/100, batch 36/625 completed (loss: 0.0254:   6%|▌         | 19/312 [01:52<29:30,  6.04s/it]
Training Epoch: 18/100, batch 37/625 completed (loss: 0.0041:   6%|▌         | 19/312 [01:52<29:29,  6.04s/it]
Training Epoch: 18/100, batch 37/625 completed (loss: 0.0101:   6%|▌         | 19/312 [01:52<29:29,  6.04s/it]
Training Epoch: 18/100, batch 37/625 completed (loss: 0.0129:   6%|▌         | 19/312 [01:52<29:30,  6.04s/it]
Training Epoch: 18/100, batch 37/625 completed (loss: 0.0024:   6%|▌         | 19/312 [01:52<29:29,  6.04s/it]
Training Epoch: 18/100, batch 38/625 completed (loss: 0.0131:   6%|▌         | 19/312 [01:55<29:29,  6.04s/it]
Training Epoch: 18/100, batch 38/625 completed (loss: 0.0567:   6%|▌         | 19/312 [01:55<29:30,  6.04s/it]
Training Epoch: 18/100, batch 38/625 completed (loss: 0.0013:   6%|▌         | 19/312 [01:55<29:29,  6.04s/it]
Training Epoch: 18/100, batch 38/625 completed (loss: 0.0068:   6%|▌         | 19/312 [01:55<29:29,  6.04s/it]
Training Epoch: 18/100, batch 38/625 completed (loss: 0.0013:   6%|▋         | 20/312 [01:58<29:04,  5.97s/it]
Training Epoch: 18/100, batch 38/625 completed (loss: 0.0068:   6%|▋         | 20/312 [01:58<29:05,  5.98s/it]
Training Epoch: 18/100, batch 38/625 completed (loss: 0.0131:   6%|▋         | 20/312 [01:58<29:05,  5.98s/it]
Training Epoch: 18/100, batch 38/625 completed (loss: 0.0567:   6%|▋         | 20/312 [01:58<29:06,  5.98s/it]
Training Epoch: 18/100, batch 39/625 completed (loss: 0.0026:   6%|▋         | 20/312 [01:58<29:04,  5.97s/it]
Training Epoch: 18/100, batch 39/625 completed (loss: 0.0264:   6%|▋         | 20/312 [01:58<29:06,  5.98s/it]
Training Epoch: 18/100, batch 39/625 completed (loss: 0.0177:   6%|▋         | 20/312 [01:58<29:05,  5.98s/it]
Training Epoch: 18/100, batch 39/625 completed (loss: 0.003:   6%|▋         | 20/312 [01:58<29:05,  5.98s/it] 
Training Epoch: 18/100, batch 40/625 completed (loss: 0.0004:   6%|▋         | 20/312 [02:01<29:04,  5.97s/it]
Training Epoch: 18/100, batch 40/625 completed (loss: 0.0207:   6%|▋         | 20/312 [02:01<29:06,  5.98s/it]
Training Epoch: 18/100, batch 40/625 completed (loss: 0.0007:   6%|▋         | 20/312 [02:01<29:05,  5.98s/it]
Training Epoch: 18/100, batch 40/625 completed (loss: 0.0124:   6%|▋         | 20/312 [02:01<29:05,  5.98s/it]
Training Epoch: 18/100, batch 40/625 completed (loss: 0.0207:   7%|▋         | 21/312 [02:04<28:56,  5.97s/it]
Training Epoch: 18/100, batch 40/625 completed (loss: 0.0007:   7%|▋         | 21/312 [02:04<28:56,  5.97s/it]
Training Epoch: 18/100, batch 40/625 completed (loss: 0.0124:   7%|▋         | 21/312 [02:04<28:56,  5.97s/it]
Training Epoch: 18/100, batch 40/625 completed (loss: 0.0004:   7%|▋         | 21/312 [02:04<28:57,  5.97s/it]
Training Epoch: 18/100, batch 41/625 completed (loss: 0.0017:   7%|▋         | 21/312 [02:04<28:56,  5.97s/it]
Training Epoch: 18/100, batch 41/625 completed (loss: 0.0461:   7%|▋         | 21/312 [02:04<28:56,  5.97s/it]
Training Epoch: 18/100, batch 41/625 completed (loss: 0.2372:   7%|▋         | 21/312 [02:04<28:57,  5.97s/it]
Training Epoch: 18/100, batch 41/625 completed (loss: 0.0107:   7%|▋         | 21/312 [02:04<28:56,  5.97s/it]
Training Epoch: 18/100, batch 42/625 completed (loss: 0.083:   7%|▋         | 21/312 [02:07<28:56,  5.97s/it] 
Training Epoch: 18/100, batch 42/625 completed (loss: 0.0071:   7%|▋         | 21/312 [02:07<28:56,  5.97s/it]
Training Epoch: 18/100, batch 42/625 completed (loss: 0.0049:   7%|▋         | 21/312 [02:07<28:57,  5.97s/it]
Training Epoch: 18/100, batch 42/625 completed (loss: 0.0208:   7%|▋         | 21/312 [02:07<28:56,  5.97s/it]
Training Epoch: 18/100, batch 42/625 completed (loss: 0.0071:   7%|▋         | 22/312 [02:10<29:05,  6.02s/it]
Training Epoch: 18/100, batch 42/625 completed (loss: 0.083:   7%|▋         | 22/312 [02:10<29:05,  6.02s/it]
Training Epoch: 18/100, batch 42/625 completed (loss: 0.0049:   7%|▋         | 22/312 [02:10<29:05,  6.02s/it]
Training Epoch: 18/100, batch 42/625 completed (loss: 0.0208:   7%|▋         | 22/312 [02:10<29:05,  6.02s/it]
Training Epoch: 18/100, batch 43/625 completed (loss: 0.0049:   7%|▋         | 22/312 [02:10<29:05,  6.02s/it]
Training Epoch: 18/100, batch 43/625 completed (loss: 0.079:   7%|▋         | 22/312 [02:10<29:05,  6.02s/it] 
Training Epoch: 18/100, batch 43/625 completed (loss: 0.0007:   7%|▋         | 22/312 [02:10<29:05,  6.02s/it]
Training Epoch: 18/100, batch 43/625 completed (loss: 0.0083:   7%|▋         | 22/312 [02:10<29:05,  6.02s/it]
Training Epoch: 18/100, batch 44/625 completed (loss: 0.0282:   7%|▋         | 22/312 [02:13<29:05,  6.02s/it]
Training Epoch: 18/100, batch 44/625 completed (loss: 0.0074:   7%|▋         | 22/312 [02:13<29:05,  6.02s/it]
Training Epoch: 18/100, batch 44/625 completed (loss: 0.0027:   7%|▋         | 22/312 [02:13<29:05,  6.02s/it]
Training Epoch: 18/100, batch 44/625 completed (loss: 0.0097:   7%|▋         | 22/312 [02:13<29:05,  6.02s/it]
Training Epoch: 18/100, batch 44/625 completed (loss: 0.0027:   7%|▋         | 23/312 [02:16<28:42,  5.96s/it]
Training Epoch: 18/100, batch 44/625 completed (loss: 0.0282:   7%|▋         | 23/312 [02:16<28:42,  5.96s/it]
Training Epoch: 18/100, batch 44/625 completed (loss: 0.0074:   7%|▋         | 23/312 [02:16<28:43,  5.96s/it]
Training Epoch: 18/100, batch 44/625 completed (loss: 0.0097:   7%|▋         | 23/312 [02:16<28:43,  5.96s/it]
Training Epoch: 18/100, batch 45/625 completed (loss: 0.0035:   7%|▋         | 23/312 [02:16<28:43,  5.96s/it]
Training Epoch: 18/100, batch 45/625 completed (loss: 0.0031:   7%|▋         | 23/312 [02:16<28:42,  5.96s/it]
Training Epoch: 18/100, batch 45/625 completed (loss: 0.049:   7%|▋         | 23/312 [02:16<28:42,  5.96s/it] 
Training Epoch: 18/100, batch 45/625 completed (loss: 0.0049:   7%|▋         | 23/312 [02:16<28:43,  5.96s/it]
Training Epoch: 18/100, batch 46/625 completed (loss: 0.021:   7%|▋         | 23/312 [02:19<28:43,  5.96s/it] 
Training Epoch: 18/100, batch 46/625 completed (loss: 0.006:   7%|▋         | 23/312 [02:19<28:42,  5.96s/it] 
Training Epoch: 18/100, batch 46/625 completed (loss: 0.0205:   7%|▋         | 23/312 [02:19<28:42,  5.96s/it]
Training Epoch: 18/100, batch 46/625 completed (loss: 0.0037:   7%|▋         | 23/312 [02:19<28:43,  5.96s/it]
Training Epoch: 18/100, batch 46/625 completed (loss: 0.0037:   8%|▊         | 24/312 [02:22<28:50,  6.01s/it]
Training Epoch: 18/100, batch 46/625 completed (loss: 0.021:   8%|▊         | 24/312 [02:22<28:50,  6.01s/it]
Training Epoch: 18/100, batch 46/625 completed (loss: 0.0205:   8%|▊         | 24/312 [02:22<28:51,  6.01s/it]
Training Epoch: 18/100, batch 46/625 completed (loss: 0.006:   8%|▊         | 24/312 [02:22<28:51,  6.01s/it]
Training Epoch: 18/100, batch 47/625 completed (loss: 0.0021:   8%|▊         | 24/312 [02:22<28:50,  6.01s/it]
Training Epoch: 18/100, batch 47/625 completed (loss: 0.0008:   8%|▊         | 24/312 [02:22<28:51,  6.01s/it]
Training Epoch: 18/100, batch 47/625 completed (loss: 0.0053:   8%|▊         | 24/312 [02:22<28:51,  6.01s/it]
Training Epoch: 18/100, batch 47/625 completed (loss: 0.008:   8%|▊         | 24/312 [02:22<28:50,  6.01s/it] 
Training Epoch: 18/100, batch 48/625 completed (loss: 0.0072:   8%|▊         | 24/312 [02:25<28:50,  6.01s/it]
Training Epoch: 18/100, batch 48/625 completed (loss: 0.003:   8%|▊         | 24/312 [02:25<28:51,  6.01s/it] 
Training Epoch: 18/100, batch 48/625 completed (loss: 0.007:   8%|▊         | 24/312 [02:25<28:51,  6.01s/it] 
Training Epoch: 18/100, batch 48/625 completed (loss: 0.002:   8%|▊         | 24/312 [02:25<28:50,  6.01s/it]
Training Epoch: 18/100, batch 48/625 completed (loss: 0.003:   8%|▊         | 25/312 [02:28<28:41,  6.00s/it]
Training Epoch: 18/100, batch 48/625 completed (loss: 0.0072:   8%|▊         | 25/312 [02:28<28:41,  6.00s/it]
Training Epoch: 18/100, batch 48/625 completed (loss: 0.002:   8%|▊         | 25/312 [02:28<28:41,  6.00s/it]
Training Epoch: 18/100, batch 48/625 completed (loss: 0.007:   8%|▊         | 25/312 [02:28<28:42,  6.00s/it]
Training Epoch: 18/100, batch 49/625 completed (loss: 0.0032:   8%|▊         | 25/312 [02:28<28:41,  6.00s/it]
Training Epoch: 18/100, batch 49/625 completed (loss: 0.0116:   8%|▊         | 25/312 [02:28<28:41,  6.00s/it]
Training Epoch: 18/100, batch 49/625 completed (loss: 0.0105:   8%|▊         | 25/312 [02:28<28:42,  6.00s/it]
Training Epoch: 18/100, batch 49/625 completed (loss: 0.0049:   8%|▊         | 25/312 [02:28<28:41,  6.00s/it]
Training Epoch: 18/100, batch 50/625 completed (loss: 0.023:   8%|▊         | 25/312 [02:31<28:42,  6.00s/it] 
Training Epoch: 18/100, batch 50/625 completed (loss: 0.0023:   8%|▊         | 25/312 [02:31<28:41,  6.00s/it]
Training Epoch: 18/100, batch 50/625 completed (loss: 0.0133:   8%|▊         | 25/312 [02:31<28:41,  6.00s/it]
Training Epoch: 18/100, batch 50/625 completed (loss: 0.0341:   8%|▊         | 25/312 [02:31<28:41,  6.00s/it]
Training Epoch: 18/100, batch 50/625 completed (loss: 0.0133:   8%|▊         | 26/312 [02:34<28:19,  5.94s/it]
Training Epoch: 18/100, batch 50/625 completed (loss: 0.0023:   8%|▊         | 26/312 [02:34<28:20,  5.94s/it]
Training Epoch: 18/100, batch 50/625 completed (loss: 0.0341:   8%|▊         | 26/312 [02:34<28:20,  5.94s/it]
Training Epoch: 18/100, batch 50/625 completed (loss: 0.023:   8%|▊         | 26/312 [02:34<28:20,  5.94s/it]
Training Epoch: 18/100, batch 51/625 completed (loss: 0.0112:   8%|▊         | 26/312 [02:34<28:20,  5.94s/it]
Training Epoch: 18/100, batch 51/625 completed (loss: 0.0008:   8%|▊         | 26/312 [02:34<28:19,  5.94s/it]
Training Epoch: 18/100, batch 51/625 completed (loss: 0.0196:   8%|▊         | 26/312 [02:34<28:20,  5.94s/it]
Training Epoch: 18/100, batch 51/625 completed (loss: 0.0454:   8%|▊         | 26/312 [02:34<28:20,  5.94s/it]
Training Epoch: 18/100, batch 52/625 completed (loss: 0.0133:   8%|▊         | 26/312 [02:37<28:20,  5.94s/it]
Training Epoch: 18/100, batch 52/625 completed (loss: 0.0565:   8%|▊         | 26/312 [02:37<28:20,  5.94s/it]
Training Epoch: 18/100, batch 52/625 completed (loss: 0.0178:   8%|▊         | 26/312 [02:37<28:19,  5.94s/it]
Training Epoch: 18/100, batch 52/625 completed (loss: 0.0251:   8%|▊         | 26/312 [02:37<28:20,  5.94s/it]
Training Epoch: 18/100, batch 52/625 completed (loss: 0.0178:   9%|▊         | 27/312 [02:40<28:54,  6.09s/it]
Training Epoch: 18/100, batch 52/625 completed (loss: 0.0251:   9%|▊         | 27/312 [02:40<28:54,  6.09s/it]
Training Epoch: 18/100, batch 52/625 completed (loss: 0.0133:   9%|▊         | 27/312 [02:40<28:54,  6.09s/it]
Training Epoch: 18/100, batch 52/625 completed (loss: 0.0565:   9%|▊         | 27/312 [02:40<28:54,  6.09s/it]
Training Epoch: 18/100, batch 53/625 completed (loss: 0.0033:   9%|▊         | 27/312 [02:40<28:54,  6.09s/it]
Training Epoch: 18/100, batch 53/625 completed (loss: 0.0021:   9%|▊         | 27/312 [02:40<28:54,  6.09s/it]
Training Epoch: 18/100, batch 53/625 completed (loss: 0.0105:   9%|▊         | 27/312 [02:40<28:54,  6.09s/it]
Training Epoch: 18/100, batch 53/625 completed (loss: 0.0009:   9%|▊         | 27/312 [02:40<28:54,  6.09s/it]
Training Epoch: 18/100, batch 54/625 completed (loss: 0.0312:   9%|▊         | 27/312 [02:43<28:54,  6.09s/it]
Training Epoch: 18/100, batch 54/625 completed (loss: 0.036:   9%|▊         | 27/312 [02:43<28:54,  6.09s/it] 
Training Epoch: 18/100, batch 54/625 completed (loss: 0.0025:   9%|▊         | 27/312 [02:43<28:54,  6.09s/it]
Training Epoch: 18/100, batch 54/625 completed (loss: 0.0124:   9%|▊         | 27/312 [02:43<28:54,  6.09s/it]
Training Epoch: 18/100, batch 54/625 completed (loss: 0.0124:   9%|▉         | 28/312 [02:46<28:26,  6.01s/it]
Training Epoch: 18/100, batch 54/625 completed (loss: 0.0312:   9%|▉         | 28/312 [02:46<28:26,  6.01s/it]
Training Epoch: 18/100, batch 54/625 completed (loss: 0.0025:   9%|▉         | 28/312 [02:46<28:26,  6.01s/it]
Training Epoch: 18/100, batch 54/625 completed (loss: 0.036:   9%|▉         | 28/312 [02:46<28:26,  6.01s/it]
Training Epoch: 18/100, batch 55/625 completed (loss: 0.0121:   9%|▉         | 28/312 [02:46<28:26,  6.01s/it]
Training Epoch: 18/100, batch 55/625 completed (loss: 0.1494:   9%|▉         | 28/312 [02:46<28:26,  6.01s/it]
Training Epoch: 18/100, batch 55/625 completed (loss: 0.0758:   9%|▉         | 28/312 [02:46<28:26,  6.01s/it]
Training Epoch: 18/100, batch 55/625 completed (loss: 0.0024:   9%|▉         | 28/312 [02:46<28:26,  6.01s/it]
Training Epoch: 18/100, batch 56/625 completed (loss: 0.058:   9%|▉         | 28/312 [02:49<28:26,  6.01s/it] 
Training Epoch: 18/100, batch 56/625 completed (loss: 0.0048:   9%|▉         | 28/312 [02:49<28:26,  6.01s/it]
Training Epoch: 18/100, batch 56/625 completed (loss: 0.0137:   9%|▉         | 28/312 [02:49<28:26,  6.01s/it]
Training Epoch: 18/100, batch 56/625 completed (loss: 0.0056:   9%|▉         | 28/312 [02:49<28:26,  6.01s/it]
Training Epoch: 18/100, batch 56/625 completed (loss: 0.058:   9%|▉         | 29/312 [02:52<28:15,  5.99s/it]
Training Epoch: 18/100, batch 56/625 completed (loss: 0.0137:   9%|▉         | 29/312 [02:52<28:15,  5.99s/it]
Training Epoch: 18/100, batch 56/625 completed (loss: 0.0048:   9%|▉         | 29/312 [02:52<28:15,  5.99s/it]
Training Epoch: 18/100, batch 56/625 completed (loss: 0.0056:   9%|▉         | 29/312 [02:52<28:15,  5.99s/it]
Training Epoch: 18/100, batch 57/625 completed (loss: 0.0169:   9%|▉         | 29/312 [02:52<28:15,  5.99s/it]
Training Epoch: 18/100, batch 57/625 completed (loss: 0.0112:   9%|▉         | 29/312 [02:52<28:15,  5.99s/it]
Training Epoch: 18/100, batch 57/625 completed (loss: 0.001:   9%|▉         | 29/312 [02:52<28:15,  5.99s/it] 
Training Epoch: 18/100, batch 57/625 completed (loss: 0.0102:   9%|▉         | 29/312 [02:52<28:15,  5.99s/it]
Training Epoch: 18/100, batch 58/625 completed (loss: 0.0046:   9%|▉         | 29/312 [02:56<28:15,  5.99s/it]
Training Epoch: 18/100, batch 58/625 completed (loss: 0.0015:   9%|▉         | 29/312 [02:56<28:15,  5.99s/it]
Training Epoch: 18/100, batch 58/625 completed (loss: 0.0052:   9%|▉         | 29/312 [02:56<28:15,  5.99s/it]
Training Epoch: 18/100, batch 58/625 completed (loss: 0.0026:   9%|▉         | 29/312 [02:56<28:15,  5.99s/it]
Training Epoch: 18/100, batch 58/625 completed (loss: 0.0015:  10%|▉         | 30/312 [02:58<28:34,  6.08s/it]
Training Epoch: 18/100, batch 58/625 completed (loss: 0.0046:  10%|▉         | 30/312 [02:58<28:34,  6.08s/it]
Training Epoch: 18/100, batch 58/625 completed (loss: 0.0052:  10%|▉         | 30/312 [02:58<28:35,  6.08s/it]
Training Epoch: 18/100, batch 58/625 completed (loss: 0.0026:  10%|▉         | 30/312 [02:59<28:34,  6.08s/it]
Training Epoch: 18/100, batch 59/625 completed (loss: 0.0023:  10%|▉         | 30/312 [02:58<28:34,  6.08s/it]
Training Epoch: 18/100, batch 59/625 completed (loss: 0.0137:  10%|▉         | 30/312 [02:59<28:35,  6.08s/it]
Training Epoch: 18/100, batch 59/625 completed (loss: 0.0035:  10%|▉         | 30/312 [02:59<28:34,  6.08s/it]
Training Epoch: 18/100, batch 59/625 completed (loss: 0.0658:  10%|▉         | 30/312 [02:59<28:34,  6.08s/it]
Training Epoch: 18/100, batch 60/625 completed (loss: 0.1332:  10%|▉         | 30/312 [03:01<28:35,  6.08s/it]
Training Epoch: 18/100, batch 60/625 completed (loss: 0.0018:  10%|▉         | 30/312 [03:01<28:34,  6.08s/it]
Training Epoch: 18/100, batch 60/625 completed (loss: 0.008:  10%|▉         | 30/312 [03:01<28:34,  6.08s/it] 
Training Epoch: 18/100, batch 60/625 completed (loss: 0.0038:  10%|▉         | 30/312 [03:01<28:34,  6.08s/it]
Training Epoch: 18/100, batch 60/625 completed (loss: 0.1332:  10%|▉         | 31/312 [03:04<28:07,  6.01s/it]
Training Epoch: 18/100, batch 60/625 completed (loss: 0.0038:  10%|▉         | 31/312 [03:04<28:07,  6.01s/it]
Training Epoch: 18/100, batch 60/625 completed (loss: 0.0018:  10%|▉         | 31/312 [03:04<28:08,  6.01s/it]
Training Epoch: 18/100, batch 60/625 completed (loss: 0.008:  10%|▉         | 31/312 [03:04<28:09,  6.01s/it]
Training Epoch: 18/100, batch 61/625 completed (loss: 0.0233:  10%|▉         | 31/312 [03:04<28:07,  6.01s/it]
Training Epoch: 18/100, batch 61/625 completed (loss: 0.0023:  10%|▉         | 31/312 [03:04<28:09,  6.01s/it]
Training Epoch: 18/100, batch 61/625 completed (loss: 0.0043:  10%|▉         | 31/312 [03:04<28:08,  6.01s/it]
Training Epoch: 18/100, batch 61/625 completed (loss: 0.011:  10%|▉         | 31/312 [03:04<28:07,  6.01s/it] 
Training Epoch: 18/100, batch 62/625 completed (loss: 0.0303:  10%|▉         | 31/312 [03:07<28:08,  6.01s/it]
Training Epoch: 18/100, batch 62/625 completed (loss: 0.0032:  10%|▉         | 31/312 [03:07<28:07,  6.01s/it]
Training Epoch: 18/100, batch 62/625 completed (loss: 0.0404:  10%|▉         | 31/312 [03:07<28:09,  6.01s/it]
Training Epoch: 18/100, batch 62/625 completed (loss: 0.004:  10%|▉         | 31/312 [03:07<28:07,  6.01s/it]
Training Epoch: 18/100, batch 62/625 completed (loss: 0.004:  10%|█         | 32/312 [03:10<28:11,  6.04s/it]
Training Epoch: 18/100, batch 62/625 completed (loss: 0.0032:  10%|█         | 32/312 [03:10<28:11,  6.04s/it]
Training Epoch: 18/100, batch 62/625 completed (loss: 0.0404:  10%|█         | 32/312 [03:10<28:11,  6.04s/it]
Training Epoch: 18/100, batch 62/625 completed (loss: 0.0303:  10%|█         | 32/312 [03:10<28:12,  6.04s/it]
Training Epoch: 18/100, batch 63/625 completed (loss: 0.015:  10%|█         | 32/312 [03:10<28:12,  6.04s/it] 
Training Epoch: 18/100, batch 63/625 completed (loss: 0.0194:  10%|█         | 32/312 [03:10<28:11,  6.04s/it]
Training Epoch: 18/100, batch 63/625 completed (loss: 0.0057:  10%|█         | 32/312 [03:10<28:11,  6.04s/it]
Training Epoch: 18/100, batch 63/625 completed (loss: 0.0088:  10%|█         | 32/312 [03:11<28:11,  6.04s/it]
Training Epoch: 18/100, batch 64/625 completed (loss: 0.0835:  10%|█         | 32/312 [03:14<28:11,  6.04s/it]
Training Epoch: 18/100, batch 64/625 completed (loss: 0.0127:  10%|█         | 32/312 [03:14<28:11,  6.04s/it]
Training Epoch: 18/100, batch 64/625 completed (loss: 0.0025:  10%|█         | 32/312 [03:14<28:12,  6.04s/it]
Training Epoch: 18/100, batch 64/625 completed (loss: 0.0206:  10%|█         | 32/312 [03:14<28:11,  6.04s/it]
Training Epoch: 18/100, batch 64/625 completed (loss: 0.0835:  11%|█         | 33/312 [03:16<28:00,  6.02s/it]
Training Epoch: 18/100, batch 64/625 completed (loss: 0.0206:  11%|█         | 33/312 [03:16<28:01,  6.03s/it]
Training Epoch: 18/100, batch 64/625 completed (loss: 0.0025:  11%|█         | 33/312 [03:16<28:01,  6.03s/it]
Training Epoch: 18/100, batch 64/625 completed (loss: 0.0127:  11%|█         | 33/312 [03:16<28:01,  6.03s/it]
Training Epoch: 18/100, batch 65/625 completed (loss: 0.0099:  11%|█         | 33/312 [03:16<28:01,  6.03s/it]
Training Epoch: 18/100, batch 65/625 completed (loss: 0.0296:  11%|█         | 33/312 [03:16<28:01,  6.03s/it]
Training Epoch: 18/100, batch 65/625 completed (loss: 0.0659:  11%|█         | 33/312 [03:16<28:00,  6.02s/it]
Training Epoch: 18/100, batch 65/625 completed (loss: 0.1086:  11%|█         | 33/312 [03:17<28:01,  6.03s/it]
Training Epoch: 18/100, batch 66/625 completed (loss: 0.0377:  11%|█         | 33/312 [03:19<28:01,  6.03s/it]
Training Epoch: 18/100, batch 66/625 completed (loss: 0.047:  11%|█         | 33/312 [03:19<28:00,  6.02s/it] 
Training Epoch: 18/100, batch 66/625 completed (loss: 0.0028:  11%|█         | 33/312 [03:19<28:01,  6.03s/it]
Training Epoch: 18/100, batch 66/625 completed (loss: 0.0721:  11%|█         | 33/312 [03:19<28:01,  6.03s/it]
Training Epoch: 18/100, batch 66/625 completed (loss: 0.0028:  11%|█         | 34/312 [03:22<27:39,  5.97s/it]
Training Epoch: 18/100, batch 66/625 completed (loss: 0.0721:  11%|█         | 34/312 [03:22<27:39,  5.97s/it]
Training Epoch: 18/100, batch 66/625 completed (loss: 0.047:  11%|█         | 34/312 [03:22<27:39,  5.97s/it]
Training Epoch: 18/100, batch 66/625 completed (loss: 0.0377:  11%|█         | 34/312 [03:22<27:40,  5.97s/it]
Training Epoch: 18/100, batch 67/625 completed (loss: 0.0012:  11%|█         | 34/312 [03:22<27:40,  5.97s/it]
Training Epoch: 18/100, batch 67/625 completed (loss: 0.0177:  11%|█         | 34/312 [03:22<27:39,  5.97s/it]
Training Epoch: 18/100, batch 67/625 completed (loss: 0.0024:  11%|█         | 34/312 [03:22<27:39,  5.97s/it]
Training Epoch: 18/100, batch 67/625 completed (loss: 0.0106:  11%|█         | 34/312 [03:22<27:39,  5.97s/it]
Training Epoch: 18/100, batch 68/625 completed (loss: 0.0066:  11%|█         | 34/312 [03:26<27:40,  5.97s/it]
Training Epoch: 18/100, batch 68/625 completed (loss: 0.0023:  11%|█         | 34/312 [03:26<27:39,  5.97s/it]
Training Epoch: 18/100, batch 68/625 completed (loss: 0.0045:  11%|█         | 34/312 [03:26<27:39,  5.97s/it]
Training Epoch: 18/100, batch 68/625 completed (loss: 0.0156:  11%|█         | 34/312 [03:26<27:39,  5.97s/it]
Training Epoch: 18/100, batch 68/625 completed (loss: 0.0023:  11%|█         | 35/312 [03:29<28:10,  6.10s/it]
Training Epoch: 18/100, batch 68/625 completed (loss: 0.0066:  11%|█         | 35/312 [03:29<28:10,  6.10s/it]
Training Epoch: 18/100, batch 68/625 completed (loss: 0.0156:  11%|█         | 35/312 [03:29<28:11,  6.11s/it]
Training Epoch: 18/100, batch 68/625 completed (loss: 0.0045:  11%|█         | 35/312 [03:29<28:11,  6.11s/it]
Training Epoch: 18/100, batch 69/625 completed (loss: 0.0015:  11%|█         | 35/312 [03:29<28:10,  6.10s/it]
Training Epoch: 18/100, batch 69/625 completed (loss: 0.018:  11%|█         | 35/312 [03:29<28:11,  6.11s/it] 
Training Epoch: 18/100, batch 69/625 completed (loss: 0.0008:  11%|█         | 35/312 [03:29<28:10,  6.10s/it]
Training Epoch: 18/100, batch 69/625 completed (loss: 0.0467:  11%|█         | 35/312 [03:29<28:11,  6.11s/it]
Training Epoch: 18/100, batch 70/625 completed (loss: 0.005:  11%|█         | 35/312 [03:32<28:11,  6.11s/it]
Training Epoch: 18/100, batch 70/625 completed (loss: 0.0031:  11%|█         | 35/312 [03:32<28:10,  6.10s/it]
Training Epoch: 18/100, batch 70/625 completed (loss: 0.0035:  11%|█         | 35/312 [03:32<28:10,  6.10s/it]
Training Epoch: 18/100, batch 70/625 completed (loss: 0.0245:  11%|█         | 35/312 [03:32<28:11,  6.11s/it]
Training Epoch: 18/100, batch 70/625 completed (loss: 0.005:  12%|█▏        | 36/312 [03:34<27:41,  6.02s/it]
Training Epoch: 18/100, batch 70/625 completed (loss: 0.0035:  12%|█▏        | 36/312 [03:34<27:41,  6.02s/it]
Training Epoch: 18/100, batch 70/625 completed (loss: 0.0245:  12%|█▏        | 36/312 [03:35<27:42,  6.02s/it]
Training Epoch: 18/100, batch 70/625 completed (loss: 0.0031:  12%|█▏        | 36/312 [03:34<27:41,  6.02s/it]
Training Epoch: 18/100, batch 71/625 completed (loss: 0.0164:  12%|█▏        | 36/312 [03:35<27:41,  6.02s/it]
Training Epoch: 18/100, batch 71/625 completed (loss: 0.0124:  12%|█▏        | 36/312 [03:35<27:41,  6.02s/it]
Training Epoch: 18/100, batch 71/625 completed (loss: 0.0203:  12%|█▏        | 36/312 [03:35<27:41,  6.02s/it]
Training Epoch: 18/100, batch 71/625 completed (loss: 0.019:  12%|█▏        | 36/312 [03:35<27:42,  6.02s/it] 
Training Epoch: 18/100, batch 72/625 completed (loss: 0.0015:  12%|█▏        | 36/312 [03:37<27:41,  6.02s/it]
Training Epoch: 18/100, batch 72/625 completed (loss: 0.034:  12%|█▏        | 36/312 [03:37<27:41,  6.02s/it] 
Training Epoch: 18/100, batch 72/625 completed (loss: 0.0046:  12%|█▏        | 36/312 [03:37<27:41,  6.02s/it]
Training Epoch: 18/100, batch 72/625 completed (loss: 0.0136:  12%|█▏        | 36/312 [03:38<27:42,  6.02s/it]
Training Epoch: 18/100, batch 72/625 completed (loss: 0.034:  12%|█▏        | 37/312 [03:41<27:49,  6.07s/it]
Training Epoch: 18/100, batch 72/625 completed (loss: 0.0046:  12%|█▏        | 37/312 [03:41<27:49,  6.07s/it]
Training Epoch: 18/100, batch 72/625 completed (loss: 0.0015:  12%|█▏        | 37/312 [03:41<27:49,  6.07s/it]
Training Epoch: 18/100, batch 72/625 completed (loss: 0.0136:  12%|█▏        | 37/312 [03:41<27:49,  6.07s/it]
Training Epoch: 18/100, batch 73/625 completed (loss: 0.0239:  12%|█▏        | 37/312 [03:41<27:49,  6.07s/it]
Training Epoch: 18/100, batch 73/625 completed (loss: 0.0138:  12%|█▏        | 37/312 [03:41<27:49,  6.07s/it]
Training Epoch: 18/100, batch 73/625 completed (loss: 0.0074:  12%|█▏        | 37/312 [03:41<27:49,  6.07s/it]
Training Epoch: 18/100, batch 73/625 completed (loss: 0.0172:  12%|█▏        | 37/312 [03:41<27:49,  6.07s/it]
Training Epoch: 18/100, batch 74/625 completed (loss: 0.0045:  12%|█▏        | 37/312 [03:44<27:49,  6.07s/it]
Training Epoch: 18/100, batch 74/625 completed (loss: 0.0463:  12%|█▏        | 37/312 [03:44<27:49,  6.07s/it]
Training Epoch: 18/100, batch 74/625 completed (loss: 0.0016:  12%|█▏        | 37/312 [03:44<27:49,  6.07s/it]
Training Epoch: 18/100, batch 74/625 completed (loss: 0.0309:  12%|█▏        | 37/312 [03:44<27:49,  6.07s/it]
Training Epoch: 18/100, batch 74/625 completed (loss: 0.0016:  12%|█▏        | 38/312 [03:47<27:37,  6.05s/it]
Training Epoch: 18/100, batch 74/625 completed (loss: 0.0463:  12%|█▏        | 38/312 [03:47<27:37,  6.05s/it]
Training Epoch: 18/100, batch 74/625 completed (loss: 0.0309:  12%|█▏        | 38/312 [03:47<27:37,  6.05s/it]
Training Epoch: 18/100, batch 74/625 completed (loss: 0.0045:  12%|█▏        | 38/312 [03:47<27:38,  6.05s/it]
Training Epoch: 18/100, batch 75/625 completed (loss: 0.0039:  12%|█▏        | 38/312 [03:47<27:37,  6.05s/it]
Training Epoch: 18/100, batch 75/625 completed (loss: 0.0181:  12%|█▏        | 38/312 [03:47<27:37,  6.05s/it]
Training Epoch: 18/100, batch 75/625 completed (loss: 0.013:  12%|█▏        | 38/312 [03:47<27:38,  6.05s/it] 
Training Epoch: 18/100, batch 75/625 completed (loss: 0.0227:  12%|█▏        | 38/312 [03:47<27:37,  6.05s/it]
Training Epoch: 18/100, batch 76/625 completed (loss: 0.0062:  12%|█▏        | 38/312 [03:50<27:37,  6.05s/it]
Training Epoch: 18/100, batch 76/625 completed (loss: 0.083:  12%|█▏        | 38/312 [03:50<27:37,  6.05s/it] 
Training Epoch: 18/100, batch 76/625 completed (loss: 0.0003:  12%|█▏        | 38/312 [03:50<27:38,  6.05s/it]
Training Epoch: 18/100, batch 76/625 completed (loss: 0.0051:  12%|█▏        | 38/312 [03:50<27:37,  6.05s/it]
Training Epoch: 18/100, batch 76/625 completed (loss: 0.0051:  12%|█▎        | 39/312 [03:53<27:13,  5.98s/it]
Training Epoch: 18/100, batch 76/625 completed (loss: 0.083:  12%|█▎        | 39/312 [03:53<27:13,  5.98s/it]
Training Epoch: 18/100, batch 76/625 completed (loss: 0.0003:  12%|█▎        | 39/312 [03:52<27:13,  5.98s/it]
Training Epoch: 18/100, batch 76/625 completed (loss: 0.0062:  12%|█▎        | 39/312 [03:53<27:14,  5.99s/it]
Training Epoch: 18/100, batch 77/625 completed (loss: 0.0312:  12%|█▎        | 39/312 [03:53<27:13,  5.98s/it]
Training Epoch: 18/100, batch 77/625 completed (loss: 0.0174:  12%|█▎        | 39/312 [03:53<27:13,  5.98s/it]
Training Epoch: 18/100, batch 77/625 completed (loss: 0.0014:  12%|█▎        | 39/312 [03:53<27:14,  5.99s/it]
Training Epoch: 18/100, batch 77/625 completed (loss: 0.0006:  12%|█▎        | 39/312 [03:53<27:13,  5.98s/it]
Training Epoch: 18/100, batch 78/625 completed (loss: 0.0006:  12%|█▎        | 39/312 [03:56<27:14,  5.99s/it]
Training Epoch: 18/100, batch 78/625 completed (loss: 0.0014:  12%|█▎        | 39/312 [03:56<27:13,  5.98s/it]
Training Epoch: 18/100, batch 78/625 completed (loss: 0.0017:  12%|█▎        | 39/312 [03:56<27:13,  5.98s/it]
Training Epoch: 18/100, batch 78/625 completed (loss: 0.0049:  12%|█▎        | 39/312 [03:56<27:13,  5.98s/it]
Training Epoch: 18/100, batch 78/625 completed (loss: 0.0006:  13%|█▎        | 40/312 [03:59<27:44,  6.12s/it]
Training Epoch: 18/100, batch 78/625 completed (loss: 0.0017:  13%|█▎        | 40/312 [03:59<27:44,  6.12s/it]
Training Epoch: 18/100, batch 78/625 completed (loss: 0.0049:  13%|█▎        | 40/312 [03:59<27:44,  6.12s/it]
Training Epoch: 18/100, batch 78/625 completed (loss: 0.0014:  13%|█▎        | 40/312 [03:59<27:44,  6.12s/it]
Training Epoch: 18/100, batch 79/625 completed (loss: 0.0322:  13%|█▎        | 40/312 [03:59<27:44,  6.12s/it]
Training Epoch: 18/100, batch 79/625 completed (loss: 0.0068:  13%|█▎        | 40/312 [03:59<27:44,  6.12s/it]
Training Epoch: 18/100, batch 79/625 completed (loss: 0.0032:  13%|█▎        | 40/312 [03:59<27:44,  6.12s/it]
Training Epoch: 18/100, batch 79/625 completed (loss: 0.0216:  13%|█▎        | 40/312 [03:59<27:44,  6.12s/it]
Training Epoch: 18/100, batch 80/625 completed (loss: 0.0099:  13%|█▎        | 40/312 [04:02<27:44,  6.12s/it]
Training Epoch: 18/100, batch 80/625 completed (loss: 0.0134:  13%|█▎        | 40/312 [04:02<27:44,  6.12s/it]
Training Epoch: 18/100, batch 80/625 completed (loss: 0.0003:  13%|█▎        | 40/312 [04:02<27:44,  6.12s/it]
Training Epoch: 18/100, batch 80/625 completed (loss: 0.0064:  13%|█▎        | 40/312 [04:02<27:44,  6.12s/it]
Training Epoch: 18/100, batch 80/625 completed (loss: 0.0134:  13%|█▎        | 41/312 [04:05<27:14,  6.03s/it]
Training Epoch: 18/100, batch 80/625 completed (loss: 0.0003:  13%|█▎        | 41/312 [04:05<27:14,  6.03s/it]
Training Epoch: 18/100, batch 80/625 completed (loss: 0.0064:  13%|█▎        | 41/312 [04:05<27:14,  6.03s/it]
Training Epoch: 18/100, batch 80/625 completed (loss: 0.0099:  13%|█▎        | 41/312 [04:05<27:14,  6.03s/it]
Training Epoch: 18/100, batch 81/625 completed (loss: 0.029:  13%|█▎        | 41/312 [04:05<27:14,  6.03s/it] 
Training Epoch: 18/100, batch 81/625 completed (loss: 0.0031:  13%|█▎        | 41/312 [04:05<27:14,  6.03s/it]
Training Epoch: 18/100, batch 81/625 completed (loss: 0.0087:  13%|█▎        | 41/312 [04:05<27:14,  6.03s/it]
Training Epoch: 18/100, batch 81/625 completed (loss: 0.0139:  13%|█▎        | 41/312 [04:05<27:14,  6.03s/it]
Training Epoch: 18/100, batch 82/625 completed (loss: 0.0079:  13%|█▎        | 41/312 [04:08<27:14,  6.03s/it]
Training Epoch: 18/100, batch 82/625 completed (loss: 0.0095:  13%|█▎        | 41/312 [04:08<27:14,  6.03s/it]
Training Epoch: 18/100, batch 82/625 completed (loss: 0.0378:  13%|█▎        | 41/312 [04:08<27:14,  6.03s/it]
Training Epoch: 18/100, batch 82/625 completed (loss: 0.0786:  13%|█▎        | 41/312 [04:08<27:14,  6.03s/it]
Training Epoch: 18/100, batch 82/625 completed (loss: 0.0378:  13%|█▎        | 42/312 [04:11<26:52,  5.97s/it]
Training Epoch: 18/100, batch 82/625 completed (loss: 0.0095:  13%|█▎        | 42/312 [04:11<26:52,  5.97s/it]
Training Epoch: 18/100, batch 82/625 completed (loss: 0.0786:  13%|█▎        | 42/312 [04:11<26:52,  5.97s/it]
Training Epoch: 18/100, batch 82/625 completed (loss: 0.0079:  13%|█▎        | 42/312 [04:11<26:53,  5.98s/it]
Training Epoch: 18/100, batch 83/625 completed (loss: 0.0268:  13%|█▎        | 42/312 [04:11<26:53,  5.98s/it]
Training Epoch: 18/100, batch 83/625 completed (loss: 0.0016:  13%|█▎        | 42/312 [04:11<26:52,  5.97s/it]
Training Epoch: 18/100, batch 83/625 completed (loss: 0.0009:  13%|█▎        | 42/312 [04:11<26:52,  5.97s/it]
Training Epoch: 18/100, batch 83/625 completed (loss: 0.002:  13%|█▎        | 42/312 [04:11<26:52,  5.97s/it] 
Training Epoch: 18/100, batch 84/625 completed (loss: 0.0009:  13%|█▎        | 42/312 [04:14<26:52,  5.97s/it]
Training Epoch: 18/100, batch 84/625 completed (loss: 0.0047:  13%|█▎        | 42/312 [04:14<26:52,  5.97s/it]
Training Epoch: 18/100, batch 84/625 completed (loss: 0.0131:  13%|█▎        | 42/312 [04:14<26:53,  5.98s/it]
Training Epoch: 18/100, batch 84/625 completed (loss: 0.0003:  13%|█▎        | 42/312 [04:14<26:52,  5.97s/it]
Training Epoch: 18/100, batch 84/625 completed (loss: 0.0131:  14%|█▍        | 43/312 [04:17<27:25,  6.12s/it]
Training Epoch: 18/100, batch 84/625 completed (loss: 0.0003:  14%|█▍        | 43/312 [04:17<27:26,  6.12s/it]
Training Epoch: 18/100, batch 84/625 completed (loss: 0.0009:  14%|█▍        | 43/312 [04:17<27:26,  6.12s/it]
Training Epoch: 18/100, batch 84/625 completed (loss: 0.0047:  14%|█▍        | 43/312 [04:17<27:27,  6.12s/it]
Training Epoch: 18/100, batch 85/625 completed (loss: 0.0029:  14%|█▍        | 43/312 [04:17<27:26,  6.12s/it]
Training Epoch: 18/100, batch 85/625 completed (loss: 0.0266:  14%|█▍        | 43/312 [04:17<27:25,  6.12s/it]
Training Epoch: 18/100, batch 85/625 completed (loss: 0.0019:  14%|█▍        | 43/312 [04:17<27:27,  6.12s/it]
Training Epoch: 18/100, batch 85/625 completed (loss: 0.0013:  14%|█▍        | 43/312 [04:17<27:26,  6.12s/it]
Training Epoch: 18/100, batch 86/625 completed (loss: 0.0259:  14%|█▍        | 43/312 [04:20<27:27,  6.12s/it]
Training Epoch: 18/100, batch 86/625 completed (loss: 0.0015:  14%|█▍        | 43/312 [04:20<27:26,  6.12s/it]
Training Epoch: 18/100, batch 86/625 completed (loss: 0.174:  14%|█▍        | 43/312 [04:20<27:25,  6.12s/it] 
Training Epoch: 18/100, batch 86/625 completed (loss: 0.0054:  14%|█▍        | 43/312 [04:20<27:26,  6.12s/it]
Training Epoch: 18/100, batch 86/625 completed (loss: 0.174:  14%|█▍        | 44/312 [04:23<26:56,  6.03s/it]
Training Epoch: 18/100, batch 86/625 completed (loss: 0.0054:  14%|█▍        | 44/312 [04:23<26:56,  6.03s/it]
Training Epoch: 18/100, batch 86/625 completed (loss: 0.0259:  14%|█▍        | 44/312 [04:23<26:56,  6.03s/it]
Training Epoch: 18/100, batch 86/625 completed (loss: 0.0015:  14%|█▍        | 44/312 [04:23<26:56,  6.03s/it]
Training Epoch: 18/100, batch 87/625 completed (loss: 0.0109:  14%|█▍        | 44/312 [04:23<26:56,  6.03s/it]
Training Epoch: 18/100, batch 87/625 completed (loss: 0.0028:  14%|█▍        | 44/312 [04:23<26:56,  6.03s/it]
Training Epoch: 18/100, batch 87/625 completed (loss: 0.0191:  14%|█▍        | 44/312 [04:23<26:56,  6.03s/it]
Training Epoch: 18/100, batch 87/625 completed (loss: 0.0067:  14%|█▍        | 44/312 [04:23<26:56,  6.03s/it]
Training Epoch: 18/100, batch 88/625 completed (loss: 0.0283:  14%|█▍        | 44/312 [04:26<26:56,  6.03s/it]
Training Epoch: 18/100, batch 88/625 completed (loss: 0.0034:  14%|█▍        | 44/312 [04:26<26:56,  6.03s/it]
Training Epoch: 18/100, batch 88/625 completed (loss: 0.0021:  14%|█▍        | 44/312 [04:26<26:56,  6.03s/it]
Training Epoch: 18/100, batch 88/625 completed (loss: 0.0042:  14%|█▍        | 44/312 [04:26<26:56,  6.03s/it]
Training Epoch: 18/100, batch 88/625 completed (loss: 0.0283:  14%|█▍        | 45/312 [04:29<27:15,  6.13s/it]
Training Epoch: 18/100, batch 88/625 completed (loss: 0.0034:  14%|█▍        | 45/312 [04:29<27:15,  6.13s/it]
Training Epoch: 18/100, batch 88/625 completed (loss: 0.0042:  14%|█▍        | 45/312 [04:29<27:15,  6.13s/it]
Training Epoch: 18/100, batch 88/625 completed (loss: 0.0021:  14%|█▍        | 45/312 [04:29<27:15,  6.13s/it]
Training Epoch: 18/100, batch 89/625 completed (loss: 0.0244:  14%|█▍        | 45/312 [04:29<27:15,  6.13s/it]
Training Epoch: 18/100, batch 89/625 completed (loss: 0.0023:  14%|█▍        | 45/312 [04:29<27:15,  6.13s/it]
Training Epoch: 18/100, batch 89/625 completed (loss: 0.09:  14%|█▍        | 45/312 [04:29<27:15,  6.13s/it]  
Training Epoch: 18/100, batch 89/625 completed (loss: 0.0043:  14%|█▍        | 45/312 [04:29<27:15,  6.13s/it]
Training Epoch: 18/100, batch 90/625 completed (loss: 0.0011:  14%|█▍        | 45/312 [04:32<27:15,  6.13s/it]
Training Epoch: 18/100, batch 90/625 completed (loss: 0.0457:  14%|█▍        | 45/312 [04:32<27:15,  6.13s/it]
Training Epoch: 18/100, batch 90/625 completed (loss: 0.0031:  14%|█▍        | 45/312 [04:32<27:15,  6.13s/it]
Training Epoch: 18/100, batch 90/625 completed (loss: 0.0633:  14%|█▍        | 45/312 [04:32<27:15,  6.13s/it]
Training Epoch: 18/100, batch 90/625 completed (loss: 0.0457:  15%|█▍        | 46/312 [04:35<27:02,  6.10s/it]
Training Epoch: 18/100, batch 90/625 completed (loss: 0.0031:  15%|█▍        | 46/312 [04:35<27:02,  6.10s/it]
Training Epoch: 18/100, batch 90/625 completed (loss: 0.0633:  15%|█▍        | 46/312 [04:35<27:02,  6.10s/it]
Training Epoch: 18/100, batch 90/625 completed (loss: 0.0011:  15%|█▍        | 46/312 [04:35<27:02,  6.10s/it]
Training Epoch: 18/100, batch 91/625 completed (loss: 0.0062:  15%|█▍        | 46/312 [04:35<27:02,  6.10s/it]
Training Epoch: 18/100, batch 91/625 completed (loss: 0.0155:  15%|█▍        | 46/312 [04:35<27:02,  6.10s/it]
Training Epoch: 18/100, batch 91/625 completed (loss: 0.0019:  15%|█▍        | 46/312 [04:35<27:02,  6.10s/it]
Training Epoch: 18/100, batch 91/625 completed (loss: 0.1163:  15%|█▍        | 46/312 [04:35<27:02,  6.10s/it]
Training Epoch: 18/100, batch 92/625 completed (loss: 0.0279:  15%|█▍        | 46/312 [04:38<27:02,  6.10s/it]
Training Epoch: 18/100, batch 92/625 completed (loss: 0.0003:  15%|█▍        | 46/312 [04:38<27:02,  6.10s/it]
Training Epoch: 18/100, batch 92/625 completed (loss: 0.0121:  15%|█▍        | 46/312 [04:38<27:02,  6.10s/it]
Training Epoch: 18/100, batch 92/625 completed (loss: 0.0074:  15%|█▍        | 46/312 [04:38<27:02,  6.10s/it]
Training Epoch: 18/100, batch 92/625 completed (loss: 0.0003:  15%|█▌        | 47/312 [04:41<26:51,  6.08s/it]
Training Epoch: 18/100, batch 92/625 completed (loss: 0.0121:  15%|█▌        | 47/312 [04:41<26:51,  6.08s/it]
Training Epoch: 18/100, batch 92/625 completed (loss: 0.0074:  15%|█▌        | 47/312 [04:41<26:51,  6.08s/it]
Training Epoch: 18/100, batch 92/625 completed (loss: 0.0279:  15%|█▌        | 47/312 [04:41<26:51,  6.08s/it]
Training Epoch: 18/100, batch 93/625 completed (loss: 0.005:  15%|█▌        | 47/312 [04:41<26:51,  6.08s/it] 
Training Epoch: 18/100, batch 93/625 completed (loss: 0.0217:  15%|█▌        | 47/312 [04:41<26:51,  6.08s/it]
Training Epoch: 18/100, batch 93/625 completed (loss: 0.0023:  15%|█▌        | 47/312 [04:41<26:51,  6.08s/it]
Training Epoch: 18/100, batch 93/625 completed (loss: 0.0002:  15%|█▌        | 47/312 [04:41<26:51,  6.08s/it]
Training Epoch: 18/100, batch 94/625 completed (loss: 0.0012:  15%|█▌        | 47/312 [04:44<26:51,  6.08s/it]
Training Epoch: 18/100, batch 94/625 completed (loss: 0.0204:  15%|█▌        | 47/312 [04:44<26:51,  6.08s/it]
Training Epoch: 18/100, batch 94/625 completed (loss: 0.0014:  15%|█▌        | 47/312 [04:44<26:51,  6.08s/it]
Training Epoch: 18/100, batch 94/625 completed (loss: 0.0059:  15%|█▌        | 47/312 [04:44<26:51,  6.08s/it]
Training Epoch: 18/100, batch 94/625 completed (loss: 0.0014:  15%|█▌        | 48/312 [04:47<26:37,  6.05s/it]
Training Epoch: 18/100, batch 94/625 completed (loss: 0.0204:  15%|█▌        | 48/312 [04:47<26:37,  6.05s/it]
Training Epoch: 18/100, batch 94/625 completed (loss: 0.0012:  15%|█▌        | 48/312 [04:47<26:37,  6.05s/it]
Training Epoch: 18/100, batch 94/625 completed (loss: 0.0059:  15%|█▌        | 48/312 [04:47<26:37,  6.05s/it]
Training Epoch: 18/100, batch 95/625 completed (loss: 0.0238:  15%|█▌        | 48/312 [04:47<26:37,  6.05s/it]
Training Epoch: 18/100, batch 95/625 completed (loss: 0.0046:  15%|█▌        | 48/312 [04:47<26:37,  6.05s/it]
Training Epoch: 18/100, batch 95/625 completed (loss: 0.0363:  15%|█▌        | 48/312 [04:47<26:37,  6.05s/it]
Training Epoch: 18/100, batch 95/625 completed (loss: 0.0063:  15%|█▌        | 48/312 [04:47<26:37,  6.05s/it]
Training Epoch: 18/100, batch 96/625 completed (loss: 0.0128:  15%|█▌        | 48/312 [04:50<26:37,  6.05s/it]
Training Epoch: 18/100, batch 96/625 completed (loss: 0.0005:  15%|█▌        | 48/312 [04:50<26:37,  6.05s/it]
Training Epoch: 18/100, batch 96/625 completed (loss: 0.0006:  15%|█▌        | 48/312 [04:50<26:37,  6.05s/it]
Training Epoch: 18/100, batch 96/625 completed (loss: 0.0053:  15%|█▌        | 48/312 [04:51<26:37,  6.05s/it]
Training Epoch: 18/100, batch 96/625 completed (loss: 0.0005:  16%|█▌        | 49/312 [04:53<26:31,  6.05s/it]
Training Epoch: 18/100, batch 96/625 completed (loss: 0.0053:  16%|█▌        | 49/312 [04:53<26:31,  6.05s/it]
Training Epoch: 18/100, batch 96/625 completed (loss: 0.0128:  16%|█▌        | 49/312 [04:53<26:32,  6.05s/it]
Training Epoch: 18/100, batch 96/625 completed (loss: 0.0006:  16%|█▌        | 49/312 [04:53<26:32,  6.05s/it]
Training Epoch: 18/100, batch 97/625 completed (loss: 0.0095:  16%|█▌        | 49/312 [04:53<26:32,  6.05s/it]
Training Epoch: 18/100, batch 97/625 completed (loss: 0.051:  16%|█▌        | 49/312 [04:53<26:32,  6.05s/it] 
Training Epoch: 18/100, batch 97/625 completed (loss: 0.0019:  16%|█▌        | 49/312 [04:53<26:31,  6.05s/it]
Training Epoch: 18/100, batch 97/625 completed (loss: 0.0024:  16%|█▌        | 49/312 [04:53<26:31,  6.05s/it]
Training Epoch: 18/100, batch 98/625 completed (loss: 0.0074:  16%|█▌        | 49/312 [04:56<26:32,  6.05s/it]
Training Epoch: 18/100, batch 98/625 completed (loss: 0.0145:  16%|█▌        | 49/312 [04:56<26:31,  6.05s/it]
Training Epoch: 18/100, batch 98/625 completed (loss: 0.0037:  16%|█▌        | 49/312 [04:56<26:32,  6.05s/it]
Training Epoch: 18/100, batch 98/625 completed (loss: 0.003:  16%|█▌        | 49/312 [04:56<26:31,  6.05s/it] 
Training Epoch: 18/100, batch 98/625 completed (loss: 0.0037:  16%|█▌        | 50/312 [05:00<26:43,  6.12s/it]
Training Epoch: 18/100, batch 98/625 completed (loss: 0.0145:  16%|█▌        | 50/312 [05:00<26:44,  6.12s/it]
Training Epoch: 18/100, batch 98/625 completed (loss: 0.003:  16%|█▌        | 50/312 [05:00<26:44,  6.12s/it]
Training Epoch: 18/100, batch 98/625 completed (loss: 0.0074:  16%|█▌        | 50/312 [05:00<26:44,  6.13s/it]
Training Epoch: 18/100, batch 99/625 completed (loss: 0.0101:  16%|█▌        | 50/312 [05:00<26:44,  6.13s/it]
Training Epoch: 18/100, batch 99/625 completed (loss: 0.0041:  16%|█▌        | 50/312 [05:00<26:43,  6.12s/it]
Training Epoch: 18/100, batch 99/625 completed (loss: 0.0003:  16%|█▌        | 50/312 [05:00<26:44,  6.12s/it]
Training Epoch: 18/100, batch 99/625 completed (loss: 0.0051:  16%|█▌        | 50/312 [05:00<26:44,  6.12s/it]
Training Epoch: 18/100, batch 100/625 completed (loss: 0.0049:  16%|█▌        | 50/312 [05:03<26:44,  6.12s/it]
Training Epoch: 18/100, batch 100/625 completed (loss: 0.0078:  16%|█▌        | 50/312 [05:03<26:43,  6.12s/it]
Training Epoch: 18/100, batch 100/625 completed (loss: 0.0016:  16%|█▌        | 50/312 [05:03<26:44,  6.13s/it]
Training Epoch: 18/100, batch 100/625 completed (loss: 0.0028:  16%|█▌        | 50/312 [05:03<26:44,  6.12s/it]
Training Epoch: 18/100, batch 100/625 completed (loss: 0.0078:  16%|█▋        | 51/312 [05:06<26:29,  6.09s/it]
Training Epoch: 18/100, batch 100/625 completed (loss: 0.0028:  16%|█▋        | 51/312 [05:06<26:28,  6.09s/it]
Training Epoch: 18/100, batch 100/625 completed (loss: 0.0049:  16%|█▋        | 51/312 [05:06<26:29,  6.09s/it]
Training Epoch: 18/100, batch 100/625 completed (loss: 0.0016:  16%|█▋        | 51/312 [05:06<26:28,  6.09s/it]
Training Epoch: 18/100, batch 101/625 completed (loss: 0.0128:  16%|█▋        | 51/312 [05:06<26:28,  6.09s/it]
Training Epoch: 18/100, batch 101/625 completed (loss: 0.0406:  16%|█▋        | 51/312 [05:06<26:29,  6.09s/it]
Training Epoch: 18/100, batch 101/625 completed (loss: 0.0007:  16%|█▋        | 51/312 [05:06<26:29,  6.09s/it]
Training Epoch: 18/100, batch 101/625 completed (loss: 0.0681:  16%|█▋        | 51/312 [05:06<26:28,  6.09s/it]
Training Epoch: 18/100, batch 102/625 completed (loss: 0.0107:  16%|█▋        | 51/312 [05:09<26:29,  6.09s/it]
Training Epoch: 18/100, batch 102/625 completed (loss: 0.0033:  16%|█▋        | 51/312 [05:09<26:28,  6.09s/it]
Training Epoch: 18/100, batch 102/625 completed (loss: 0.015:  16%|█▋        | 51/312 [05:09<26:29,  6.09s/it] 
Training Epoch: 18/100, batch 102/625 completed (loss: 0.0269:  16%|█▋        | 51/312 [05:09<26:28,  6.09s/it]
Training Epoch: 18/100, batch 102/625 completed (loss: 0.015:  17%|█▋        | 52/312 [05:12<26:04,  6.02s/it]
Training Epoch: 18/100, batch 102/625 completed (loss: 0.0269:  17%|█▋        | 52/312 [05:12<26:04,  6.02s/it]
Training Epoch: 18/100, batch 102/625 completed (loss: 0.0033:  17%|█▋        | 52/312 [05:11<26:04,  6.02s/it]
Training Epoch: 18/100, batch 102/625 completed (loss: 0.0107:  17%|█▋        | 52/312 [05:12<26:05,  6.02s/it]
Training Epoch: 18/100, batch 103/625 completed (loss: 0.0095:  17%|█▋        | 52/312 [05:12<26:04,  6.02s/it]
Training Epoch: 18/100, batch 103/625 completed (loss: 0.0081:  17%|█▋        | 52/312 [05:12<26:04,  6.02s/it]
Training Epoch: 18/100, batch 103/625 completed (loss: 0.0084:  17%|█▋        | 52/312 [05:12<26:05,  6.02s/it]
Training Epoch: 18/100, batch 103/625 completed (loss: 0.0005:  17%|█▋        | 52/312 [05:12<26:04,  6.02s/it]
Training Epoch: 18/100, batch 104/625 completed (loss: 0.0074:  17%|█▋        | 52/312 [05:15<26:05,  6.02s/it]
Training Epoch: 18/100, batch 104/625 completed (loss: 0.0116:  17%|█▋        | 52/312 [05:15<26:04,  6.02s/it]
Training Epoch: 18/100, batch 104/625 completed (loss: 0.0097:  17%|█▋        | 52/312 [05:15<26:04,  6.02s/it]
Training Epoch: 18/100, batch 104/625 completed (loss: 0.0001:  17%|█▋        | 52/312 [05:15<26:04,  6.02s/it]
Training Epoch: 18/100, batch 104/625 completed (loss: 0.0116:  17%|█▋        | 53/312 [05:18<26:20,  6.10s/it]
Training Epoch: 18/100, batch 104/625 completed (loss: 0.0001:  17%|█▋        | 53/312 [05:18<26:21,  6.11s/it]
Training Epoch: 18/100, batch 104/625 completed (loss: 0.0074:  17%|█▋        | 53/312 [05:18<26:21,  6.11s/it]
Training Epoch: 18/100, batch 104/625 completed (loss: 0.0097:  17%|█▋        | 53/312 [05:18<26:22,  6.11s/it]
Training Epoch: 18/100, batch 105/625 completed (loss: 0.0034:  17%|█▋        | 53/312 [05:18<26:20,  6.10s/it]
Training Epoch: 18/100, batch 105/625 completed (loss: 0.0046:  17%|█▋        | 53/312 [05:18<26:21,  6.11s/it]
Training Epoch: 18/100, batch 105/625 completed (loss: 0.0029:  17%|█▋        | 53/312 [05:18<26:22,  6.11s/it]
Training Epoch: 18/100, batch 105/625 completed (loss: 0.0619:  17%|█▋        | 53/312 [05:18<26:21,  6.11s/it]
Training Epoch: 18/100, batch 106/625 completed (loss: 0.0049:  17%|█▋        | 53/312 [05:21<26:21,  6.11s/it]
Training Epoch: 18/100, batch 106/625 completed (loss: 0.015:  17%|█▋        | 53/312 [05:21<26:20,  6.10s/it] 
Training Epoch: 18/100, batch 106/625 completed (loss: 0.0211:  17%|█▋        | 53/312 [05:21<26:22,  6.11s/it]
Training Epoch: 18/100, batch 106/625 completed (loss: 0.0006:  17%|█▋        | 53/312 [05:21<26:21,  6.11s/it]
Training Epoch: 18/100, batch 106/625 completed (loss: 0.0211:  17%|█▋        | 54/312 [05:24<26:06,  6.07s/it]
Training Epoch: 18/100, batch 106/625 completed (loss: 0.015:  17%|█▋        | 54/312 [05:24<26:07,  6.07s/it]
Training Epoch: 18/100, batch 106/625 completed (loss: 0.0049:  17%|█▋        | 54/312 [05:24<26:06,  6.07s/it]
Training Epoch: 18/100, batch 106/625 completed (loss: 0.0006:  17%|█▋        | 54/312 [05:24<26:07,  6.07s/it]
Training Epoch: 18/100, batch 107/625 completed (loss: 0.0012:  17%|█▋        | 54/312 [05:24<26:06,  6.07s/it]
Training Epoch: 18/100, batch 107/625 completed (loss: 0.0047:  17%|█▋        | 54/312 [05:24<26:07,  6.07s/it]
Training Epoch: 18/100, batch 107/625 completed (loss: 0.0027:  17%|█▋        | 54/312 [05:24<26:06,  6.07s/it]
Training Epoch: 18/100, batch 107/625 completed (loss: 0.0838:  17%|█▋        | 54/312 [05:24<26:07,  6.07s/it]
Training Epoch: 18/100, batch 108/625 completed (loss: 0.0248:  17%|█▋        | 54/312 [05:27<26:06,  6.07s/it]
Training Epoch: 18/100, batch 108/625 completed (loss: 0.0012:  17%|█▋        | 54/312 [05:27<26:07,  6.07s/it]
Training Epoch: 18/100, batch 108/625 completed (loss: 0.0004:  17%|█▋        | 54/312 [05:27<26:06,  6.07s/it]
Training Epoch: 18/100, batch 108/625 completed (loss: 0.0004:  17%|█▋        | 54/312 [05:27<26:07,  6.07s/it]
Training Epoch: 18/100, batch 108/625 completed (loss: 0.0012:  18%|█▊        | 55/312 [05:30<25:53,  6.04s/it]
Training Epoch: 18/100, batch 108/625 completed (loss: 0.0004:  18%|█▊        | 55/312 [05:30<25:53,  6.04s/it]
Training Epoch: 18/100, batch 108/625 completed (loss: 0.0004:  18%|█▊        | 55/312 [05:30<25:53,  6.04s/it]
Training Epoch: 18/100, batch 108/625 completed (loss: 0.0248:  18%|█▊        | 55/312 [05:30<25:53,  6.04s/it]
Training Epoch: 18/100, batch 109/625 completed (loss: 0.0103:  18%|█▊        | 55/312 [05:30<25:53,  6.04s/it]
Training Epoch: 18/100, batch 109/625 completed (loss: 0.0018:  18%|█▊        | 55/312 [05:30<25:53,  6.04s/it]
Training Epoch: 18/100, batch 109/625 completed (loss: 0.0024:  18%|█▊        | 55/312 [05:30<25:53,  6.04s/it]
Training Epoch: 18/100, batch 109/625 completed (loss: 0.0067:  18%|█▊        | 55/312 [05:30<25:53,  6.04s/it]
Training Epoch: 18/100, batch 110/625 completed (loss: 0.0002:  18%|█▊        | 55/312 [05:33<25:53,  6.04s/it]
Training Epoch: 18/100, batch 110/625 completed (loss: 0.003:  18%|█▊        | 55/312 [05:33<25:53,  6.04s/it] 
Training Epoch: 18/100, batch 110/625 completed (loss: 0.0484:  18%|█▊        | 55/312 [05:33<25:53,  6.04s/it]
Training Epoch: 18/100, batch 110/625 completed (loss: 0.0105:  18%|█▊        | 55/312 [05:33<25:53,  6.04s/it]
Training Epoch: 18/100, batch 110/625 completed (loss: 0.0484:  18%|█▊        | 56/312 [05:36<25:43,  6.03s/it]
Training Epoch: 18/100, batch 110/625 completed (loss: 0.003:  18%|█▊        | 56/312 [05:36<25:43,  6.03s/it]
Training Epoch: 18/100, batch 110/625 completed (loss: 0.0002:  18%|█▊        | 56/312 [05:36<25:43,  6.03s/it]
Training Epoch: 18/100, batch 110/625 completed (loss: 0.0105:  18%|█▊        | 56/312 [05:36<25:43,  6.03s/it]
Training Epoch: 18/100, batch 111/625 completed (loss: 0.0029:  18%|█▊        | 56/312 [05:36<25:43,  6.03s/it]
Training Epoch: 18/100, batch 111/625 completed (loss: 0.005:  18%|█▊        | 56/312 [05:36<25:43,  6.03s/it] 
Training Epoch: 18/100, batch 111/625 completed (loss: 0.0005:  18%|█▊        | 56/312 [05:36<25:43,  6.03s/it]
Training Epoch: 18/100, batch 111/625 completed (loss: 0.0042:  18%|█▊        | 56/312 [05:36<25:43,  6.03s/it]
Training Epoch: 18/100, batch 112/625 completed (loss: 0.0043:  18%|█▊        | 56/312 [05:39<25:43,  6.03s/it]
Training Epoch: 18/100, batch 112/625 completed (loss: 0.0214:  18%|█▊        | 56/312 [05:39<25:43,  6.03s/it]
Training Epoch: 18/100, batch 112/625 completed (loss: 0.0006:  18%|█▊        | 56/312 [05:39<25:43,  6.03s/it]
Training Epoch: 18/100, batch 112/625 completed (loss: 0.0544:  18%|█▊        | 56/312 [05:39<25:43,  6.03s/it]
Training Epoch: 18/100, batch 112/625 completed (loss: 0.0214:  18%|█▊        | 57/312 [05:42<25:36,  6.03s/it]
Training Epoch: 18/100, batch 112/625 completed (loss: 0.0043:  18%|█▊        | 57/312 [05:42<25:36,  6.03s/it]
Training Epoch: 18/100, batch 112/625 completed (loss: 0.0006:  18%|█▊        | 57/312 [05:42<25:36,  6.03s/it]
Training Epoch: 18/100, batch 112/625 completed (loss: 0.0544:  18%|█▊        | 57/312 [05:42<25:36,  6.03s/it]
Training Epoch: 18/100, batch 113/625 completed (loss: 0.0105:  18%|█▊        | 57/312 [05:42<25:36,  6.03s/it]
Training Epoch: 18/100, batch 113/625 completed (loss: 0.0119:  18%|█▊        | 57/312 [05:42<25:36,  6.03s/it]
Training Epoch: 18/100, batch 113/625 completed (loss: 0.0254:  18%|█▊        | 57/312 [05:42<25:36,  6.03s/it]
Training Epoch: 18/100, batch 113/625 completed (loss: 0.0002:  18%|█▊        | 57/312 [05:42<25:36,  6.03s/it]
Training Epoch: 18/100, batch 114/625 completed (loss: 0.0349:  18%|█▊        | 57/312 [05:45<25:36,  6.03s/it]
Training Epoch: 18/100, batch 114/625 completed (loss: 0.0001:  18%|█▊        | 57/312 [05:45<25:36,  6.03s/it]
Training Epoch: 18/100, batch 114/625 completed (loss: 0.0018:  18%|█▊        | 57/312 [05:45<25:36,  6.03s/it]
Training Epoch: 18/100, batch 114/625 completed (loss: 0.0164:  18%|█▊        | 57/312 [05:45<25:36,  6.03s/it]
Training Epoch: 18/100, batch 114/625 completed (loss: 0.0349:  19%|█▊        | 58/312 [05:48<25:52,  6.11s/it]
Training Epoch: 18/100, batch 114/625 completed (loss: 0.0018:  19%|█▊        | 58/312 [05:48<25:52,  6.11s/it]
Training Epoch: 18/100, batch 114/625 completed (loss: 0.0164:  19%|█▊        | 58/312 [05:48<25:52,  6.11s/it]
Training Epoch: 18/100, batch 114/625 completed (loss: 0.0001:  19%|█▊        | 58/312 [05:48<25:52,  6.11s/it]
Training Epoch: 18/100, batch 115/625 completed (loss: 0.0002:  19%|█▊        | 58/312 [05:48<25:52,  6.11s/it]
Training Epoch: 18/100, batch 115/625 completed (loss: 0.0082:  19%|█▊        | 58/312 [05:48<25:52,  6.11s/it]
Training Epoch: 18/100, batch 115/625 completed (loss: 0.0008:  19%|█▊        | 58/312 [05:48<25:52,  6.11s/it]
Training Epoch: 18/100, batch 115/625 completed (loss: 0.0005:  19%|█▊        | 58/312 [05:48<25:52,  6.11s/it]
Training Epoch: 18/100, batch 116/625 completed (loss: 0.0034:  19%|█▊        | 58/312 [05:51<25:52,  6.11s/it]
Training Epoch: 18/100, batch 116/625 completed (loss: 0.0003:  19%|█▊        | 58/312 [05:51<25:52,  6.11s/it]
Training Epoch: 18/100, batch 116/625 completed (loss: 0.0006:  19%|█▊        | 58/312 [05:51<25:52,  6.11s/it]
Training Epoch: 18/100, batch 116/625 completed (loss: 0.0003:  19%|█▊        | 58/312 [05:51<25:52,  6.11s/it]
Training Epoch: 18/100, batch 116/625 completed (loss: 0.0006:  19%|█▉        | 59/312 [05:54<25:26,  6.03s/it]
Training Epoch: 18/100, batch 116/625 completed (loss: 0.0003:  19%|█▉        | 59/312 [05:54<25:26,  6.03s/it]
Training Epoch: 18/100, batch 116/625 completed (loss: 0.0003:  19%|█▉        | 59/312 [05:54<25:26,  6.03s/it]
Training Epoch: 18/100, batch 116/625 completed (loss: 0.0034:  19%|█▉        | 59/312 [05:54<25:26,  6.03s/it]
Training Epoch: 18/100, batch 117/625 completed (loss: 0.0005:  19%|█▉        | 59/312 [05:54<25:26,  6.03s/it]
Training Epoch: 18/100, batch 117/625 completed (loss: 0.0001:  19%|█▉        | 59/312 [05:54<25:26,  6.03s/it]
Training Epoch: 18/100, batch 117/625 completed (loss: 0.0059:  19%|█▉        | 59/312 [05:54<25:26,  6.03s/it]
Training Epoch: 18/100, batch 117/625 completed (loss: 0.0003:  19%|█▉        | 59/312 [05:54<25:26,  6.03s/it]
Training Epoch: 18/100, batch 118/625 completed (loss: 0.0113:  19%|█▉        | 59/312 [05:57<25:26,  6.03s/it]
Training Epoch: 18/100, batch 118/625 completed (loss: 0.0021:  19%|█▉        | 59/312 [05:57<25:26,  6.03s/it]
Training Epoch: 18/100, batch 118/625 completed (loss: 0.0:  19%|█▉        | 59/312 [05:57<25:26,  6.03s/it]   
Training Epoch: 18/100, batch 118/625 completed (loss: 0.1268:  19%|█▉        | 59/312 [05:57<25:26,  6.03s/it]
Training Epoch: 18/100, batch 118/625 completed (loss: 0.0113:  19%|█▉        | 60/312 [06:00<25:35,  6.09s/it]
Training Epoch: 18/100, batch 118/625 completed (loss: 0.1268:  19%|█▉        | 60/312 [06:00<25:35,  6.09s/it]
Training Epoch: 18/100, batch 118/625 completed (loss: 0.0021:  19%|█▉        | 60/312 [06:00<25:36,  6.10s/it]
Training Epoch: 18/100, batch 118/625 completed (loss: 0.0:  19%|█▉        | 60/312 [06:00<25:36,  6.10s/it]
Training Epoch: 18/100, batch 119/625 completed (loss: 0.0004:  19%|█▉        | 60/312 [06:00<25:35,  6.09s/it]
Training Epoch: 18/100, batch 119/625 completed (loss: 0.002:  19%|█▉        | 60/312 [06:00<25:36,  6.10s/it]
Training Epoch: 18/100, batch 119/625 completed (loss: 0.0042:  19%|█▉        | 60/312 [06:00<25:36,  6.10s/it]
Training Epoch: 18/100, batch 119/625 completed (loss: 0.0008:  19%|█▉        | 60/312 [06:00<25:35,  6.09s/it]
Training Epoch: 18/100, batch 120/625 completed (loss: 0.0124:  19%|█▉        | 60/312 [06:03<25:36,  6.10s/it]
Training Epoch: 18/100, batch 120/625 completed (loss: 0.0012:  19%|█▉        | 60/312 [06:03<25:35,  6.09s/it]
Training Epoch: 18/100, batch 120/625 completed (loss: 0.0025:  19%|█▉        | 60/312 [06:03<25:36,  6.10s/it]
Training Epoch: 18/100, batch 120/625 completed (loss: 0.0022:  19%|█▉        | 60/312 [06:03<25:35,  6.09s/it]
Training Epoch: 18/100, batch 120/625 completed (loss: 0.0124:  20%|█▉        | 61/312 [06:06<25:20,  6.06s/it]
Training Epoch: 18/100, batch 120/625 completed (loss: 0.0022:  20%|█▉        | 61/312 [06:06<25:20,  6.06s/it]
Training Epoch: 18/100, batch 120/625 completed (loss: 0.0025:  20%|█▉        | 61/312 [06:06<25:20,  6.06s/it]
Training Epoch: 18/100, batch 120/625 completed (loss: 0.0012:  20%|█▉        | 61/312 [06:06<25:20,  6.06s/it]
Training Epoch: 18/100, batch 121/625 completed (loss: 0.0005:  20%|█▉        | 61/312 [06:06<25:20,  6.06s/it]
Training Epoch: 18/100, batch 121/625 completed (loss: 0.133:  20%|█▉        | 61/312 [06:06<25:20,  6.06s/it] 
Training Epoch: 18/100, batch 121/625 completed (loss: 0.0038:  20%|█▉        | 61/312 [06:06<25:20,  6.06s/it]
Training Epoch: 18/100, batch 121/625 completed (loss: 0.0074:  20%|█▉        | 61/312 [06:06<25:20,  6.06s/it]
Training Epoch: 18/100, batch 122/625 completed (loss: 0.0266:  20%|█▉        | 61/312 [06:09<25:20,  6.06s/it]
Training Epoch: 18/100, batch 122/625 completed (loss: 0.0017:  20%|█▉        | 61/312 [06:09<25:20,  6.06s/it]
Training Epoch: 18/100, batch 122/625 completed (loss: 0.0031:  20%|█▉        | 61/312 [06:09<25:20,  6.06s/it]
Training Epoch: 18/100, batch 122/625 completed (loss: 0.0691:  20%|█▉        | 61/312 [06:09<25:20,  6.06s/it]
Training Epoch: 18/100, batch 122/625 completed (loss: 0.0031:  20%|█▉        | 62/312 [06:12<25:10,  6.04s/it]
Training Epoch: 18/100, batch 122/625 completed (loss: 0.0017:  20%|█▉        | 62/312 [06:12<25:10,  6.04s/it]
Training Epoch: 18/100, batch 122/625 completed (loss: 0.0691:  20%|█▉        | 62/312 [06:12<25:11,  6.04s/it]
Training Epoch: 18/100, batch 122/625 completed (loss: 0.0266:  20%|█▉        | 62/312 [06:12<25:11,  6.05s/it]
Training Epoch: 18/100, batch 123/625 completed (loss: 0.0085:  20%|█▉        | 62/312 [06:12<25:11,  6.05s/it]
Training Epoch: 18/100, batch 123/625 completed (loss: 0.0039:  20%|█▉        | 62/312 [06:12<25:10,  6.04s/it]
Training Epoch: 18/100, batch 123/625 completed (loss: 0.0011:  20%|█▉        | 62/312 [06:12<25:10,  6.04s/it]
Training Epoch: 18/100, batch 123/625 completed (loss: 0.0008:  20%|█▉        | 62/312 [06:12<25:11,  6.04s/it]
Training Epoch: 18/100, batch 124/625 completed (loss: 0.0013:  20%|█▉        | 62/312 [06:15<25:10,  6.04s/it]
Training Epoch: 18/100, batch 124/625 completed (loss: 0.0018:  20%|█▉        | 62/312 [06:15<25:10,  6.04s/it]
Training Epoch: 18/100, batch 124/625 completed (loss: 0.0021:  20%|█▉        | 62/312 [06:15<25:11,  6.05s/it]
Training Epoch: 18/100, batch 124/625 completed (loss: 0.0108:  20%|█▉        | 62/312 [06:15<25:11,  6.04s/it]
Training Epoch: 18/100, batch 124/625 completed (loss: 0.0018:  20%|██        | 63/312 [06:18<25:22,  6.12s/it]
Training Epoch: 18/100, batch 124/625 completed (loss: 0.0021:  20%|██        | 63/312 [06:18<25:22,  6.12s/it]
Training Epoch: 18/100, batch 124/625 completed (loss: 0.0013:  20%|██        | 63/312 [06:18<25:23,  6.12s/it]
Training Epoch: 18/100, batch 124/625 completed (loss: 0.0108:  20%|██        | 63/312 [06:19<25:23,  6.12s/it]
Training Epoch: 18/100, batch 125/625 completed (loss: 0.0087:  20%|██        | 63/312 [06:19<25:22,  6.12s/it]
Training Epoch: 18/100, batch 125/625 completed (loss: 0.0594:  20%|██        | 63/312 [06:19<25:22,  6.12s/it]
Training Epoch: 18/100, batch 125/625 completed (loss: 0.0284:  20%|██        | 63/312 [06:19<25:23,  6.12s/it]
Training Epoch: 18/100, batch 125/625 completed (loss: 0.0028:  20%|██        | 63/312 [06:19<25:23,  6.12s/it]
Training Epoch: 18/100, batch 126/625 completed (loss: 0.0055:  20%|██        | 63/312 [06:21<25:23,  6.12s/it]
Training Epoch: 18/100, batch 126/625 completed (loss: 0.0062:  20%|██        | 63/312 [06:21<25:22,  6.12s/it]
Training Epoch: 18/100, batch 126/625 completed (loss: 0.0009:  20%|██        | 63/312 [06:21<25:22,  6.12s/it]
Training Epoch: 18/100, batch 126/625 completed (loss: 0.0063:  20%|██        | 63/312 [06:21<25:23,  6.12s/it]
Training Epoch: 18/100, batch 126/625 completed (loss: 0.0062:  21%|██        | 64/312 [06:24<24:55,  6.03s/it]
Training Epoch: 18/100, batch 126/625 completed (loss: 0.0055:  21%|██        | 64/312 [06:24<24:55,  6.03s/it]
Training Epoch: 18/100, batch 126/625 completed (loss: 0.0063:  21%|██        | 64/312 [06:24<24:55,  6.03s/it]
Training Epoch: 18/100, batch 126/625 completed (loss: 0.0009:  21%|██        | 64/312 [06:24<24:55,  6.03s/it]
Training Epoch: 18/100, batch 127/625 completed (loss: 0.0177:  21%|██        | 64/312 [06:24<24:55,  6.03s/it]
Training Epoch: 18/100, batch 127/625 completed (loss: 0.0008:  21%|██        | 64/312 [06:24<24:55,  6.03s/it]
Training Epoch: 18/100, batch 127/625 completed (loss: 0.005:  21%|██        | 64/312 [06:24<24:55,  6.03s/it] 
Training Epoch: 18/100, batch 127/625 completed (loss: 0.0013:  21%|██        | 64/312 [06:24<24:55,  6.03s/it]
Training Epoch: 18/100, batch 128/625 completed (loss: 0.0119:  21%|██        | 64/312 [06:27<24:55,  6.03s/it]
Training Epoch: 18/100, batch 128/625 completed (loss: 0.0046:  21%|██        | 64/312 [06:27<24:55,  6.03s/it]
Training Epoch: 18/100, batch 128/625 completed (loss: 0.0114:  21%|██        | 64/312 [06:27<24:55,  6.03s/it]
Training Epoch: 18/100, batch 128/625 completed (loss: 0.0057:  21%|██        | 64/312 [06:28<24:55,  6.03s/it]
Training Epoch: 18/100, batch 128/625 completed (loss: 0.0057:  21%|██        | 65/312 [06:31<24:58,  6.07s/it]
Training Epoch: 18/100, batch 128/625 completed (loss: 0.0046:  21%|██        | 65/312 [06:30<24:58,  6.07s/it]
Training Epoch: 18/100, batch 128/625 completed (loss: 0.0119:  21%|██        | 65/312 [06:30<24:58,  6.07s/it]
Training Epoch: 18/100, batch 128/625 completed (loss: 0.0114:  21%|██        | 65/312 [06:30<24:58,  6.07s/it]
Training Epoch: 18/100, batch 129/625 completed (loss: 0.001:  21%|██        | 65/312 [06:30<24:58,  6.07s/it] 
Training Epoch: 18/100, batch 129/625 completed (loss: 0.0012:  21%|██        | 65/312 [06:31<24:58,  6.07s/it]
Training Epoch: 18/100, batch 129/625 completed (loss: 0.0013:  21%|██        | 65/312 [06:31<24:58,  6.07s/it]
Training Epoch: 18/100, batch 129/625 completed (loss: 0.0009:  21%|██        | 65/312 [06:31<24:58,  6.07s/it]
Training Epoch: 18/100, batch 130/625 completed (loss: 0.0227:  21%|██        | 65/312 [06:34<24:58,  6.07s/it]
Training Epoch: 18/100, batch 130/625 completed (loss: 0.015:  21%|██        | 65/312 [06:34<24:58,  6.07s/it] 
Training Epoch: 18/100, batch 130/625 completed (loss: 0.0061:  21%|██        | 65/312 [06:34<24:58,  6.07s/it]
Training Epoch: 18/100, batch 130/625 completed (loss: 0.0002:  21%|██        | 65/312 [06:34<24:58,  6.07s/it]
Training Epoch: 18/100, batch 130/625 completed (loss: 0.0061:  21%|██        | 66/312 [06:37<24:59,  6.10s/it]
Training Epoch: 18/100, batch 130/625 completed (loss: 0.015:  21%|██        | 66/312 [06:37<24:59,  6.10s/it]
Training Epoch: 18/100, batch 130/625 completed (loss: 0.0002:  21%|██        | 66/312 [06:37<24:59,  6.10s/it]
Training Epoch: 18/100, batch 130/625 completed (loss: 0.0227:  21%|██        | 66/312 [06:37<25:01,  6.10s/it]
Training Epoch: 18/100, batch 131/625 completed (loss: 0.0106:  21%|██        | 66/312 [06:37<25:01,  6.10s/it]
Training Epoch: 18/100, batch 131/625 completed (loss: 0.0005:  21%|██        | 66/312 [06:37<24:59,  6.10s/it]
Training Epoch: 18/100, batch 131/625 completed (loss: 0.0009:  21%|██        | 66/312 [06:37<24:59,  6.10s/it]
Training Epoch: 18/100, batch 131/625 completed (loss: 0.0118:  21%|██        | 66/312 [06:37<24:59,  6.10s/it]
Training Epoch: 18/100, batch 132/625 completed (loss: 0.0005:  21%|██        | 66/312 [06:40<24:59,  6.10s/it]
Training Epoch: 18/100, batch 132/625 completed (loss: 0.0017:  21%|██        | 66/312 [06:40<24:59,  6.10s/it]
Training Epoch: 18/100, batch 132/625 completed (loss: 0.0042:  21%|██        | 66/312 [06:40<25:01,  6.10s/it]
Training Epoch: 18/100, batch 132/625 completed (loss: 0.0104:  21%|██        | 66/312 [06:40<24:59,  6.10s/it]
Training Epoch: 18/100, batch 132/625 completed (loss: 0.0042:  21%|██▏       | 67/312 [06:42<24:36,  6.03s/it]
Training Epoch: 18/100, batch 132/625 completed (loss: 0.0017:  21%|██▏       | 67/312 [06:42<24:37,  6.03s/it]
Training Epoch: 18/100, batch 132/625 completed (loss: 0.0104:  21%|██▏       | 67/312 [06:43<24:37,  6.03s/it]
Training Epoch: 18/100, batch 132/625 completed (loss: 0.0005:  21%|██▏       | 67/312 [06:42<24:37,  6.03s/it]
Training Epoch: 18/100, batch 133/625 completed (loss: 0.0105:  21%|██▏       | 67/312 [06:43<24:37,  6.03s/it]
Training Epoch: 18/100, batch 133/625 completed (loss: 0.0028:  21%|██▏       | 67/312 [06:43<24:37,  6.03s/it]
Training Epoch: 18/100, batch 133/625 completed (loss: 0.0008:  21%|██▏       | 67/312 [06:43<24:36,  6.03s/it]
Training Epoch: 18/100, batch 133/625 completed (loss: 0.0006:  21%|██▏       | 67/312 [06:43<24:37,  6.03s/it]
Training Epoch: 18/100, batch 134/625 completed (loss: 0.0025:  21%|██▏       | 67/312 [06:46<24:37,  6.03s/it]
Training Epoch: 18/100, batch 134/625 completed (loss: 0.0021:  21%|██▏       | 67/312 [06:46<24:36,  6.03s/it]
Training Epoch: 18/100, batch 134/625 completed (loss: 0.0007:  21%|██▏       | 67/312 [06:46<24:37,  6.03s/it]
Training Epoch: 18/100, batch 134/625 completed (loss: 0.0041:  21%|██▏       | 67/312 [06:46<24:37,  6.03s/it]
Training Epoch: 18/100, batch 134/625 completed (loss: 0.0041:  22%|██▏       | 68/312 [06:49<25:02,  6.16s/it]
Training Epoch: 18/100, batch 134/625 completed (loss: 0.0007:  22%|██▏       | 68/312 [06:49<25:03,  6.16s/it]
Training Epoch: 18/100, batch 134/625 completed (loss: 0.0025:  22%|██▏       | 68/312 [06:49<25:03,  6.16s/it]
Training Epoch: 18/100, batch 134/625 completed (loss: 0.0021:  22%|██▏       | 68/312 [06:49<25:03,  6.16s/it]
Training Epoch: 18/100, batch 135/625 completed (loss: 0.0002:  22%|██▏       | 68/312 [06:49<25:03,  6.16s/it]
Training Epoch: 18/100, batch 135/625 completed (loss: 0.0072:  22%|██▏       | 68/312 [06:49<25:03,  6.16s/it]
Training Epoch: 18/100, batch 135/625 completed (loss: 0.0004:  22%|██▏       | 68/312 [06:49<25:03,  6.16s/it]
Training Epoch: 18/100, batch 135/625 completed (loss: 0.006:  22%|██▏       | 68/312 [06:49<25:02,  6.16s/it] 
Training Epoch: 18/100, batch 136/625 completed (loss: 0.0003:  22%|██▏       | 68/312 [06:52<25:03,  6.16s/it]
Training Epoch: 18/100, batch 136/625 completed (loss: 0.0003:  22%|██▏       | 68/312 [06:52<25:03,  6.16s/it]
Training Epoch: 18/100, batch 136/625 completed (loss: 0.0052:  22%|██▏       | 68/312 [06:52<25:03,  6.16s/it]
Training Epoch: 18/100, batch 136/625 completed (loss: 0.0005:  22%|██▏       | 68/312 [06:52<25:02,  6.16s/it]
Training Epoch: 18/100, batch 136/625 completed (loss: 0.0052:  22%|██▏       | 69/312 [06:55<24:42,  6.10s/it]
Training Epoch: 18/100, batch 136/625 completed (loss: 0.0003:  22%|██▏       | 69/312 [06:55<24:43,  6.10s/it]
Training Epoch: 18/100, batch 136/625 completed (loss: 0.0003:  22%|██▏       | 69/312 [06:55<24:42,  6.10s/it]
Training Epoch: 18/100, batch 136/625 completed (loss: 0.0005:  22%|██▏       | 69/312 [06:55<24:43,  6.10s/it]
Training Epoch: 18/100, batch 137/625 completed (loss: 0.0244:  22%|██▏       | 69/312 [06:55<24:42,  6.10s/it]
Training Epoch: 18/100, batch 137/625 completed (loss: 0.0039:  22%|██▏       | 69/312 [06:55<24:42,  6.10s/it]
Training Epoch: 18/100, batch 137/625 completed (loss: 0.0002:  22%|██▏       | 69/312 [06:55<24:43,  6.10s/it]
Training Epoch: 18/100, batch 137/625 completed (loss: 0.0092:  22%|██▏       | 69/312 [06:55<24:43,  6.10s/it]
Training Epoch: 18/100, batch 138/625 completed (loss: 0.0008:  22%|██▏       | 69/312 [06:58<24:43,  6.10s/it]
Training Epoch: 18/100, batch 138/625 completed (loss: 0.0005:  22%|██▏       | 69/312 [06:58<24:42,  6.10s/it]
Training Epoch: 18/100, batch 138/625 completed (loss: 0.0023:  22%|██▏       | 69/312 [06:58<24:42,  6.10s/it]
Training Epoch: 18/100, batch 138/625 completed (loss: 0.0006:  22%|██▏       | 69/312 [06:58<24:43,  6.10s/it]
Training Epoch: 18/100, batch 138/625 completed (loss: 0.0008:  22%|██▏       | 70/312 [07:01<24:29,  6.07s/it]
Training Epoch: 18/100, batch 138/625 completed (loss: 0.0005:  22%|██▏       | 70/312 [07:01<24:29,  6.07s/it]
Training Epoch: 18/100, batch 138/625 completed (loss: 0.0006:  22%|██▏       | 70/312 [07:01<24:30,  6.07s/it]
Training Epoch: 18/100, batch 138/625 completed (loss: 0.0023:  22%|██▏       | 70/312 [07:01<24:30,  6.08s/it]
Training Epoch: 18/100, batch 139/625 completed (loss: 0.0002:  22%|██▏       | 70/312 [07:01<24:29,  6.07s/it]
Training Epoch: 18/100, batch 139/625 completed (loss: 0.0004:  22%|██▏       | 70/312 [07:01<24:30,  6.08s/it]
Training Epoch: 18/100, batch 139/625 completed (loss: 0.0007:  22%|██▏       | 70/312 [07:01<24:29,  6.07s/it]
Training Epoch: 18/100, batch 139/625 completed (loss: 0.0133:  22%|██▏       | 70/312 [07:01<24:30,  6.07s/it]
Training Epoch: 18/100, batch 140/625 completed (loss: 0.003:  22%|██▏       | 70/312 [07:04<24:29,  6.07s/it] 
Training Epoch: 18/100, batch 140/625 completed (loss: 0.0258:  22%|██▏       | 70/312 [07:04<24:30,  6.08s/it]
Training Epoch: 18/100, batch 140/625 completed (loss: 0.0026:  22%|██▏       | 70/312 [07:04<24:29,  6.07s/it]
Training Epoch: 18/100, batch 140/625 completed (loss: 0.0132:  22%|██▏       | 70/312 [07:04<24:30,  6.07s/it]
Training Epoch: 18/100, batch 140/625 completed (loss: 0.0258:  23%|██▎       | 71/312 [07:07<24:39,  6.14s/it]
Training Epoch: 18/100, batch 140/625 completed (loss: 0.003:  23%|██▎       | 71/312 [07:07<24:40,  6.14s/it]
Training Epoch: 18/100, batch 140/625 completed (loss: 0.0026:  23%|██▎       | 71/312 [07:07<24:40,  6.14s/it]
Training Epoch: 18/100, batch 140/625 completed (loss: 0.0132:  23%|██▎       | 71/312 [07:07<24:40,  6.14s/it]
Training Epoch: 18/100, batch 141/625 completed (loss: 0.0006:  23%|██▎       | 71/312 [07:07<24:39,  6.14s/it]
Training Epoch: 18/100, batch 141/625 completed (loss: 0.0233:  23%|██▎       | 71/312 [07:07<24:40,  6.14s/it]
Training Epoch: 18/100, batch 141/625 completed (loss: 0.0023:  23%|██▎       | 71/312 [07:07<24:40,  6.14s/it]
Training Epoch: 18/100, batch 141/625 completed (loss: 0.0018:  23%|██▎       | 71/312 [07:07<24:40,  6.14s/it]
Training Epoch: 18/100, batch 142/625 completed (loss: 0.0166:  23%|██▎       | 71/312 [07:10<24:40,  6.14s/it]
Training Epoch: 18/100, batch 142/625 completed (loss: 0.0:  23%|██▎       | 71/312 [07:10<24:39,  6.14s/it]   
Training Epoch: 18/100, batch 142/625 completed (loss: 0.0112:  23%|██▎       | 71/312 [07:10<24:40,  6.14s/it]
Training Epoch: 18/100, batch 142/625 completed (loss: 0.0005:  23%|██▎       | 71/312 [07:10<24:40,  6.14s/it]
Training Epoch: 18/100, batch 142/625 completed (loss: 0.0112:  23%|██▎       | 72/312 [07:13<24:12,  6.05s/it]
Training Epoch: 18/100, batch 142/625 completed (loss: 0.0005:  23%|██▎       | 72/312 [07:13<24:12,  6.05s/it]
Training Epoch: 18/100, batch 142/625 completed (loss: 0.0:  23%|██▎       | 72/312 [07:13<24:12,  6.05s/it]
Training Epoch: 18/100, batch 142/625 completed (loss: 0.0166:  23%|██▎       | 72/312 [07:13<24:13,  6.06s/it]
Training Epoch: 18/100, batch 143/625 completed (loss: 0.0081:  23%|██▎       | 72/312 [07:13<24:12,  6.05s/it]
Training Epoch: 18/100, batch 143/625 completed (loss: 0.0019:  23%|██▎       | 72/312 [07:13<24:13,  6.06s/it]
Training Epoch: 18/100, batch 143/625 completed (loss: 0.0347:  23%|██▎       | 72/312 [07:13<24:12,  6.05s/it]
Training Epoch: 18/100, batch 143/625 completed (loss: 0.0009:  23%|██▎       | 72/312 [07:13<24:12,  6.05s/it]
Training Epoch: 18/100, batch 144/625 completed (loss: 0.0099:  23%|██▎       | 72/312 [07:16<24:12,  6.05s/it]
Training Epoch: 18/100, batch 144/625 completed (loss: 0.0083:  23%|██▎       | 72/312 [07:16<24:13,  6.06s/it]
Training Epoch: 18/100, batch 144/625 completed (loss: 0.0473:  23%|██▎       | 72/312 [07:16<24:12,  6.05s/it]
Training Epoch: 18/100, batch 144/625 completed (loss: 0.007:  23%|██▎       | 72/312 [07:16<24:12,  6.05s/it] 
Training Epoch: 18/100, batch 144/625 completed (loss: 0.0473:  23%|██▎       | 73/312 [07:19<24:23,  6.12s/it]
Training Epoch: 18/100, batch 144/625 completed (loss: 0.0099:  23%|██▎       | 73/312 [07:19<24:23,  6.12s/it]
Training Epoch: 18/100, batch 144/625 completed (loss: 0.0083:  23%|██▎       | 73/312 [07:19<24:23,  6.12s/it]
Training Epoch: 18/100, batch 144/625 completed (loss: 0.007:  23%|██▎       | 73/312 [07:19<24:23,  6.13s/it]
Training Epoch: 18/100, batch 145/625 completed (loss: 0.0541:  23%|██▎       | 73/312 [07:19<24:23,  6.12s/it]
Training Epoch: 18/100, batch 145/625 completed (loss: 0.0001:  23%|██▎       | 73/312 [07:19<24:23,  6.12s/it]
Training Epoch: 18/100, batch 145/625 completed (loss: 0.0034:  23%|██▎       | 73/312 [07:19<24:23,  6.12s/it]
Training Epoch: 18/100, batch 145/625 completed (loss: 0.0012:  23%|██▎       | 73/312 [07:19<24:23,  6.13s/it]
Training Epoch: 18/100, batch 146/625 completed (loss: 0.0011:  23%|██▎       | 73/312 [07:22<24:23,  6.12s/it]
Training Epoch: 18/100, batch 146/625 completed (loss: 0.0147:  23%|██▎       | 73/312 [07:22<24:23,  6.12s/it]
Training Epoch: 18/100, batch 146/625 completed (loss: 0.0002:  23%|██▎       | 73/312 [07:22<24:23,  6.12s/it]
Training Epoch: 18/100, batch 146/625 completed (loss: 0.0013:  23%|██▎       | 73/312 [07:23<24:23,  6.13s/it]
Training Epoch: 18/100, batch 146/625 completed (loss: 0.0013:  24%|██▎       | 74/312 [07:25<24:08,  6.09s/it]
Training Epoch: 18/100, batch 146/625 completed (loss: 0.0147:  24%|██▎       | 74/312 [07:25<24:08,  6.09s/it]
Training Epoch: 18/100, batch 146/625 completed (loss: 0.0011:  24%|██▎       | 74/312 [07:25<24:08,  6.09s/it]
Training Epoch: 18/100, batch 146/625 completed (loss: 0.0002:  24%|██▎       | 74/312 [07:25<24:09,  6.09s/it]
Training Epoch: 18/100, batch 147/625 completed (loss: 0.0002:  24%|██▎       | 74/312 [07:25<24:09,  6.09s/it]
Training Epoch: 18/100, batch 147/625 completed (loss: 0.0006:  24%|██▎       | 74/312 [07:25<24:08,  6.09s/it]
Training Epoch: 18/100, batch 147/625 completed (loss: 0.0016:  24%|██▎       | 74/312 [07:25<24:08,  6.09s/it]
Training Epoch: 18/100, batch 147/625 completed (loss: 0.0698:  24%|██▎       | 74/312 [07:25<24:08,  6.09s/it]
Training Epoch: 18/100, batch 148/625 completed (loss: 0.012:  24%|██▎       | 74/312 [07:28<24:09,  6.09s/it] 
Training Epoch: 18/100, batch 148/625 completed (loss: 0.0064:  24%|██▎       | 74/312 [07:28<24:08,  6.09s/it]
Training Epoch: 18/100, batch 148/625 completed (loss: 0.0032:  24%|██▎       | 74/312 [07:28<24:08,  6.09s/it]
Training Epoch: 18/100, batch 148/625 completed (loss: 0.0:  24%|██▎       | 74/312 [07:28<24:08,  6.09s/it]   
Training Epoch: 18/100, batch 148/625 completed (loss: 0.012:  24%|██▍       | 75/312 [07:31<23:45,  6.02s/it]
Training Epoch: 18/100, batch 148/625 completed (loss: 0.0:  24%|██▍       | 75/312 [07:31<23:46,  6.02s/it]
Training Epoch: 18/100, batch 148/625 completed (loss: 0.0064:  24%|██▍       | 75/312 [07:31<23:46,  6.02s/it]
Training Epoch: 18/100, batch 148/625 completed (loss: 0.0032:  24%|██▍       | 75/312 [07:31<23:46,  6.02s/it]
Training Epoch: 18/100, batch 149/625 completed (loss: 0.0002:  24%|██▍       | 75/312 [07:31<23:46,  6.02s/it]
Training Epoch: 18/100, batch 149/625 completed (loss: 0.0002:  24%|██▍       | 75/312 [07:31<23:45,  6.02s/it]
Training Epoch: 18/100, batch 149/625 completed (loss: 0.0008:  24%|██▍       | 75/312 [07:31<23:46,  6.02s/it]
Training Epoch: 18/100, batch 149/625 completed (loss: 0.012:  24%|██▍       | 75/312 [07:31<23:46,  6.02s/it]
Training Epoch: 18/100, batch 150/625 completed (loss: 0.0012:  24%|██▍       | 75/312 [07:35<23:45,  6.02s/it]
Training Epoch: 18/100, batch 150/625 completed (loss: 0.0001:  24%|██▍       | 75/312 [07:35<23:46,  6.02s/it]
Training Epoch: 18/100, batch 150/625 completed (loss: 0.1078:  24%|██▍       | 75/312 [07:35<23:46,  6.02s/it]
Training Epoch: 18/100, batch 150/625 completed (loss: 0.0675:  24%|██▍       | 75/312 [07:35<23:46,  6.02s/it]
Training Epoch: 18/100, batch 150/625 completed (loss: 0.0001:  24%|██▍       | 76/312 [07:38<24:12,  6.15s/it]
Training Epoch: 18/100, batch 150/625 completed (loss: 0.1078:  24%|██▍       | 76/312 [07:38<24:12,  6.15s/it]
Training Epoch: 18/100, batch 150/625 completed (loss: 0.0012:  24%|██▍       | 76/312 [07:38<24:12,  6.15s/it]
Training Epoch: 18/100, batch 150/625 completed (loss: 0.0675:  24%|██▍       | 76/312 [07:38<24:12,  6.16s/it]
Training Epoch: 18/100, batch 151/625 completed (loss: 0.0003:  24%|██▍       | 76/312 [07:38<24:12,  6.15s/it]
Training Epoch: 18/100, batch 151/625 completed (loss: 0.0706:  24%|██▍       | 76/312 [07:38<24:12,  6.15s/it]
Training Epoch: 18/100, batch 151/625 completed (loss: 0.0024:  24%|██▍       | 76/312 [07:38<24:12,  6.15s/it]
Training Epoch: 18/100, batch 151/625 completed (loss: 0.0003:  24%|██▍       | 76/312 [07:38<24:12,  6.16s/it]
Training Epoch: 18/100, batch 152/625 completed (loss: 0.0026:  24%|██▍       | 76/312 [07:41<24:12,  6.15s/it]
Training Epoch: 18/100, batch 152/625 completed (loss: 0.0067:  24%|██▍       | 76/312 [07:41<24:12,  6.15s/it]
Training Epoch: 18/100, batch 152/625 completed (loss: 0.0041:  24%|██▍       | 76/312 [07:41<24:12,  6.15s/it]
Training Epoch: 18/100, batch 152/625 completed (loss: 0.0006:  24%|██▍       | 76/312 [07:41<24:12,  6.16s/it]
Training Epoch: 18/100, batch 152/625 completed (loss: 0.0041:  25%|██▍       | 77/312 [07:44<23:44,  6.06s/it]
Training Epoch: 18/100, batch 152/625 completed (loss: 0.0067:  25%|██▍       | 77/312 [07:44<23:45,  6.06s/it]
Training Epoch: 18/100, batch 152/625 completed (loss: 0.0006:  25%|██▍       | 77/312 [07:44<23:45,  6.06s/it]
Training Epoch: 18/100, batch 152/625 completed (loss: 0.0026:  25%|██▍       | 77/312 [07:44<23:45,  6.07s/it]
Training Epoch: 18/100, batch 153/625 completed (loss: 0.0002:  25%|██▍       | 77/312 [07:44<23:45,  6.07s/it]
Training Epoch: 18/100, batch 153/625 completed (loss: 0.0033:  25%|██▍       | 77/312 [07:44<23:44,  6.06s/it]
Training Epoch: 18/100, batch 153/625 completed (loss: 0.0014:  25%|██▍       | 77/312 [07:44<23:45,  6.06s/it]
Training Epoch: 18/100, batch 153/625 completed (loss: 0.0026:  25%|██▍       | 77/312 [07:44<23:45,  6.06s/it]
Training Epoch: 18/100, batch 154/625 completed (loss: 0.0101:  25%|██▍       | 77/312 [07:47<23:44,  6.06s/it]
Training Epoch: 18/100, batch 154/625 completed (loss: 0.0004:  25%|██▍       | 77/312 [07:47<23:45,  6.07s/it]
Training Epoch: 18/100, batch 154/625 completed (loss: 0.0068:  25%|██▍       | 77/312 [07:46<23:45,  6.06s/it]
Training Epoch: 18/100, batch 154/625 completed (loss: 0.0015:  25%|██▍       | 77/312 [07:47<23:45,  6.06s/it]
Training Epoch: 18/100, batch 154/625 completed (loss: 0.0015:  25%|██▌       | 78/312 [07:50<23:52,  6.12s/it]
Training Epoch: 18/100, batch 154/625 completed (loss: 0.0004:  25%|██▌       | 78/312 [07:50<23:52,  6.12s/it]
Training Epoch: 18/100, batch 154/625 completed (loss: 0.0101:  25%|██▌       | 78/312 [07:50<23:53,  6.13s/it]
Training Epoch: 18/100, batch 154/625 completed (loss: 0.0068:  25%|██▌       | 78/312 [07:50<23:53,  6.13s/it]
Training Epoch: 18/100, batch 155/625 completed (loss: 0.0046:  25%|██▌       | 78/312 [07:50<23:53,  6.13s/it]
Training Epoch: 18/100, batch 155/625 completed (loss: 0.0007:  25%|██▌       | 78/312 [07:50<23:52,  6.12s/it]
Training Epoch: 18/100, batch 155/625 completed (loss: 0.0435:  25%|██▌       | 78/312 [07:50<23:53,  6.13s/it]
Training Epoch: 18/100, batch 155/625 completed (loss: 0.0269:  25%|██▌       | 78/312 [07:50<23:52,  6.12s/it]
Training Epoch: 18/100, batch 156/625 completed (loss: 0.0307:  25%|██▌       | 78/312 [07:53<23:52,  6.12s/it]
Training Epoch: 18/100, batch 156/625 completed (loss: 0.0018:  25%|██▌       | 78/312 [07:53<23:53,  6.13s/it]
Training Epoch: 18/100, batch 156/625 completed (loss: 0.0108:  25%|██▌       | 78/312 [07:53<23:53,  6.13s/it]
Training Epoch: 18/100, batch 156/625 completed (loss: 0.0011:  25%|██▌       | 78/312 [07:53<23:52,  6.12s/it]
Training Epoch: 18/100, batch 156/625 completed (loss: 0.0011:  25%|██▌       | 79/312 [07:56<23:40,  6.10s/it]
Training Epoch: 18/100, batch 156/625 completed (loss: 0.0307:  25%|██▌       | 79/312 [07:56<23:40,  6.10s/it]
Training Epoch: 18/100, batch 156/625 completed (loss: 0.0108:  25%|██▌       | 79/312 [07:56<23:40,  6.10s/it]
Training Epoch: 18/100, batch 156/625 completed (loss: 0.0018:  25%|██▌       | 79/312 [07:56<23:40,  6.10s/it]
Training Epoch: 18/100, batch 157/625 completed (loss: 0.0282:  25%|██▌       | 79/312 [07:56<23:40,  6.10s/it]
Training Epoch: 18/100, batch 157/625 completed (loss: 0.002:  25%|██▌       | 79/312 [07:56<23:40,  6.10s/it] 
Training Epoch: 18/100, batch 157/625 completed (loss: 0.0037:  25%|██▌       | 79/312 [07:56<23:40,  6.10s/it]
Training Epoch: 18/100, batch 157/625 completed (loss: 0.0011:  25%|██▌       | 79/312 [07:56<23:40,  6.10s/it]
Training Epoch: 18/100, batch 158/625 completed (loss: 0.0008:  25%|██▌       | 79/312 [07:59<23:40,  6.10s/it]
Training Epoch: 18/100, batch 158/625 completed (loss: 0.0005:  25%|██▌       | 79/312 [07:59<23:40,  6.10s/it]
Training Epoch: 18/100, batch 158/625 completed (loss: 0.0033:  25%|██▌       | 79/312 [07:59<23:40,  6.10s/it]
Training Epoch: 18/100, batch 158/625 completed (loss: 0.0003:  25%|██▌       | 79/312 [07:59<23:40,  6.10s/it]
Training Epoch: 18/100, batch 158/625 completed (loss: 0.0033:  26%|██▌       | 80/312 [08:02<23:18,  6.03s/it]
Training Epoch: 18/100, batch 158/625 completed (loss: 0.0008:  26%|██▌       | 80/312 [08:02<23:19,  6.03s/it]
Training Epoch: 18/100, batch 158/625 completed (loss: 0.0003:  26%|██▌       | 80/312 [08:02<23:19,  6.03s/it]
Training Epoch: 18/100, batch 158/625 completed (loss: 0.0005:  26%|██▌       | 80/312 [08:02<23:20,  6.04s/it]
Training Epoch: 18/100, batch 159/625 completed (loss: 0.0538:  26%|██▌       | 80/312 [08:02<23:19,  6.03s/it]
Training Epoch: 18/100, batch 159/625 completed (loss: 0.0296:  26%|██▌       | 80/312 [08:02<23:20,  6.04s/it]
Training Epoch: 18/100, batch 159/625 completed (loss: 0.0015:  26%|██▌       | 80/312 [08:02<23:18,  6.03s/it]
Training Epoch: 18/100, batch 159/625 completed (loss: 0.0009:  26%|██▌       | 80/312 [08:02<23:19,  6.03s/it]
Training Epoch: 18/100, batch 160/625 completed (loss: 0.0039:  26%|██▌       | 80/312 [08:05<23:18,  6.03s/it]
Training Epoch: 18/100, batch 160/625 completed (loss: 0.0003:  26%|██▌       | 80/312 [08:05<23:20,  6.04s/it]
Training Epoch: 18/100, batch 160/625 completed (loss: 0.0017:  26%|██▌       | 80/312 [08:05<23:19,  6.03s/it]
Training Epoch: 18/100, batch 160/625 completed (loss: 0.04:  26%|██▌       | 80/312 [08:05<23:19,  6.03s/it]  
Training Epoch: 18/100, batch 160/625 completed (loss: 0.0039:  26%|██▌       | 81/312 [08:08<23:35,  6.13s/it]
Training Epoch: 18/100, batch 160/625 completed (loss: 0.0017:  26%|██▌       | 81/312 [08:08<23:36,  6.13s/it]
Training Epoch: 18/100, batch 160/625 completed (loss: 0.0003:  26%|██▌       | 81/312 [08:08<23:37,  6.13s/it]
Training Epoch: 18/100, batch 160/625 completed (loss: 0.04:  26%|██▌       | 81/312 [08:08<23:37,  6.14s/it]
Training Epoch: 18/100, batch 161/625 completed (loss: 0.0083:  26%|██▌       | 81/312 [08:08<23:37,  6.13s/it]
Training Epoch: 18/100, batch 161/625 completed (loss: 0.0034:  26%|██▌       | 81/312 [08:08<23:36,  6.13s/it]
Training Epoch: 18/100, batch 161/625 completed (loss: 0.0003:  26%|██▌       | 81/312 [08:08<23:35,  6.13s/it]
Training Epoch: 18/100, batch 161/625 completed (loss: 0.0009:  26%|██▌       | 81/312 [08:08<23:37,  6.14s/it]
Training Epoch: 18/100, batch 162/625 completed (loss: 0.0003:  26%|██▌       | 81/312 [08:11<23:37,  6.13s/it]
Training Epoch: 18/100, batch 162/625 completed (loss: 0.0005:  26%|██▌       | 81/312 [08:11<23:35,  6.13s/it]
Training Epoch: 18/100, batch 162/625 completed (loss: 0.0178:  26%|██▌       | 81/312 [08:11<23:36,  6.13s/it]
Training Epoch: 18/100, batch 162/625 completed (loss: 0.0018:  26%|██▌       | 81/312 [08:11<23:37,  6.14s/it]
Training Epoch: 18/100, batch 162/625 completed (loss: 0.0005:  26%|██▋       | 82/312 [08:14<23:12,  6.05s/it]
Training Epoch: 18/100, batch 162/625 completed (loss: 0.0178:  26%|██▋       | 82/312 [08:14<23:12,  6.05s/it]
Training Epoch: 18/100, batch 162/625 completed (loss: 0.0018:  26%|██▋       | 82/312 [08:14<23:12,  6.05s/it]
Training Epoch: 18/100, batch 162/625 completed (loss: 0.0003:  26%|██▋       | 82/312 [08:14<23:11,  6.05s/it]
Training Epoch: 18/100, batch 163/625 completed (loss: 0.0061:  26%|██▋       | 82/312 [08:14<23:12,  6.05s/it]
Training Epoch: 18/100, batch 163/625 completed (loss: 0.0005:  26%|██▋       | 82/312 [08:14<23:11,  6.05s/it]
Training Epoch: 18/100, batch 163/625 completed (loss: 0.0245:  26%|██▋       | 82/312 [08:14<23:12,  6.05s/it]
Training Epoch: 18/100, batch 163/625 completed (loss: 0.0029:  26%|██▋       | 82/312 [08:14<23:12,  6.05s/it]
Training Epoch: 18/100, batch 164/625 completed (loss: 0.0176:  26%|██▋       | 82/312 [08:17<23:12,  6.05s/it]
Training Epoch: 18/100, batch 164/625 completed (loss: 0.0005:  26%|██▋       | 82/312 [08:17<23:12,  6.05s/it]
Training Epoch: 18/100, batch 164/625 completed (loss: 0.0015:  26%|██▋       | 82/312 [08:17<23:11,  6.05s/it]
Training Epoch: 18/100, batch 164/625 completed (loss: 0.0004:  26%|██▋       | 82/312 [08:17<23:12,  6.05s/it]
Training Epoch: 18/100, batch 164/625 completed (loss: 0.0176:  27%|██▋       | 83/312 [08:20<23:04,  6.05s/it]
Training Epoch: 18/100, batch 164/625 completed (loss: 0.0004:  27%|██▋       | 83/312 [08:20<23:04,  6.04s/it]
Training Epoch: 18/100, batch 164/625 completed (loss: 0.0015:  27%|██▋       | 83/312 [08:20<23:04,  6.04s/it]
Training Epoch: 18/100, batch 164/625 completed (loss: 0.0005:  27%|██▋       | 83/312 [08:20<23:04,  6.05s/it]
Training Epoch: 18/100, batch 165/625 completed (loss: 0.0004:  27%|██▋       | 83/312 [08:20<23:04,  6.05s/it]
Training Epoch: 18/100, batch 165/625 completed (loss: 0.0012:  27%|██▋       | 83/312 [08:20<23:04,  6.05s/it]
Training Epoch: 18/100, batch 165/625 completed (loss: 0.0012:  27%|██▋       | 83/312 [08:20<23:04,  6.04s/it]
Training Epoch: 18/100, batch 165/625 completed (loss: 0.0001:  27%|██▋       | 83/312 [08:20<23:04,  6.04s/it]
Training Epoch: 18/100, batch 166/625 completed (loss: 0.0004:  27%|██▋       | 83/312 [08:23<23:04,  6.04s/it]
Training Epoch: 18/100, batch 166/625 completed (loss: 0.0038:  27%|██▋       | 83/312 [08:23<23:04,  6.05s/it]
Training Epoch: 18/100, batch 166/625 completed (loss: 0.0012:  27%|██▋       | 83/312 [08:23<23:04,  6.05s/it]
Training Epoch: 18/100, batch 166/625 completed (loss: 0.0032:  27%|██▋       | 83/312 [08:23<23:04,  6.04s/it]
Training Epoch: 18/100, batch 166/625 completed (loss: 0.0038:  27%|██▋       | 84/312 [08:26<23:16,  6.13s/it]
Training Epoch: 18/100, batch 166/625 completed (loss: 0.0012:  27%|██▋       | 84/312 [08:26<23:16,  6.13s/it]
Training Epoch: 18/100, batch 166/625 completed (loss: 0.0004:  27%|██▋       | 84/312 [08:26<23:16,  6.13s/it]
Training Epoch: 18/100, batch 166/625 completed (loss: 0.0032:  27%|██▋       | 84/312 [08:26<23:17,  6.13s/it]
Training Epoch: 18/100, batch 167/625 completed (loss: 0.0087:  27%|██▋       | 84/312 [08:26<23:16,  6.13s/it]
Training Epoch: 18/100, batch 167/625 completed (loss: 0.0004:  27%|██▋       | 84/312 [08:26<23:16,  6.13s/it]
Training Epoch: 18/100, batch 167/625 completed (loss: 0.0058:  27%|██▋       | 84/312 [08:26<23:16,  6.13s/it]
Training Epoch: 18/100, batch 167/625 completed (loss: 0.0296:  27%|██▋       | 84/312 [08:26<23:17,  6.13s/it]
Training Epoch: 18/100, batch 168/625 completed (loss: 0.028:  27%|██▋       | 84/312 [08:29<23:16,  6.13s/it] 
Training Epoch: 18/100, batch 168/625 completed (loss: 0.0037:  27%|██▋       | 84/312 [08:29<23:16,  6.13s/it]
Training Epoch: 18/100, batch 168/625 completed (loss: 0.015:  27%|██▋       | 84/312 [08:29<23:16,  6.13s/it] 
Training Epoch: 18/100, batch 168/625 completed (loss: 0.0007:  27%|██▋       | 84/312 [08:29<23:17,  6.13s/it]
Training Epoch: 18/100, batch 168/625 completed (loss: 0.0007:  27%|██▋       | 85/312 [08:32<22:55,  6.06s/it]
Training Epoch: 18/100, batch 168/625 completed (loss: 0.0037:  27%|██▋       | 85/312 [08:32<22:56,  6.06s/it]
Training Epoch: 18/100, batch 168/625 completed (loss: 0.028:  27%|██▋       | 85/312 [08:32<22:56,  6.06s/it]
Training Epoch: 18/100, batch 168/625 completed (loss: 0.015:  27%|██▋       | 85/312 [08:32<22:56,  6.06s/it]
Training Epoch: 18/100, batch 169/625 completed (loss: 0.0006:  27%|██▋       | 85/312 [08:32<22:56,  6.06s/it]
Training Epoch: 18/100, batch 169/625 completed (loss: 0.0093:  27%|██▋       | 85/312 [08:32<22:56,  6.06s/it]
Training Epoch: 18/100, batch 169/625 completed (loss: 0.0012:  27%|██▋       | 85/312 [08:32<22:56,  6.06s/it]
Training Epoch: 18/100, batch 169/625 completed (loss: 0.0004:  27%|██▋       | 85/312 [08:32<22:55,  6.06s/it]
Training Epoch: 18/100, batch 170/625 completed (loss: 0.0486:  27%|██▋       | 85/312 [08:35<22:56,  6.06s/it]
Training Epoch: 18/100, batch 170/625 completed (loss: 0.0186:  27%|██▋       | 85/312 [08:35<22:56,  6.06s/it]
Training Epoch: 18/100, batch 170/625 completed (loss: 0.0036:  27%|██▋       | 85/312 [08:35<22:56,  6.06s/it]
Training Epoch: 18/100, batch 170/625 completed (loss: 0.0048:  27%|██▋       | 85/312 [08:35<22:55,  6.06s/it]
Training Epoch: 18/100, batch 170/625 completed (loss: 0.0486:  28%|██▊       | 86/312 [08:38<22:53,  6.08s/it]
Training Epoch: 18/100, batch 170/625 completed (loss: 0.0186:  28%|██▊       | 86/312 [08:38<22:53,  6.08s/it]
Training Epoch: 18/100, batch 170/625 completed (loss: 0.0036:  28%|██▊       | 86/312 [08:38<22:54,  6.08s/it]
Training Epoch: 18/100, batch 170/625 completed (loss: 0.0048:  28%|██▊       | 86/312 [08:38<22:53,  6.08s/it]
Training Epoch: 18/100, batch 171/625 completed (loss: 0.0533:  28%|██▊       | 86/312 [08:38<22:53,  6.08s/it]
Training Epoch: 18/100, batch 171/625 completed (loss: 0.0021:  28%|██▊       | 86/312 [08:38<22:54,  6.08s/it]
Training Epoch: 18/100, batch 171/625 completed (loss: 0.002:  28%|██▊       | 86/312 [08:38<22:53,  6.08s/it] 
Training Epoch: 18/100, batch 171/625 completed (loss: 0.0057:  28%|██▊       | 86/312 [08:38<22:53,  6.08s/it]
Training Epoch: 18/100, batch 172/625 completed (loss: 0.0004:  28%|██▊       | 86/312 [08:42<22:54,  6.08s/it]
Training Epoch: 18/100, batch 172/625 completed (loss: 0.0076:  28%|██▊       | 86/312 [08:42<22:53,  6.08s/it]
Training Epoch: 18/100, batch 172/625 completed (loss: 0.001:  28%|██▊       | 86/312 [08:42<22:53,  6.08s/it]
Training Epoch: 18/100, batch 172/625 completed (loss: 0.0009:  28%|██▊       | 86/312 [08:42<22:53,  6.08s/it]
Training Epoch: 18/100, batch 172/625 completed (loss: 0.0004:  28%|██▊       | 87/312 [08:44<22:46,  6.07s/it]
Training Epoch: 18/100, batch 172/625 completed (loss: 0.0009:  28%|██▊       | 87/312 [08:44<22:46,  6.07s/it]
Training Epoch: 18/100, batch 172/625 completed (loss: 0.001:  28%|██▊       | 87/312 [08:44<22:46,  6.07s/it]
Training Epoch: 18/100, batch 172/625 completed (loss: 0.0076:  28%|██▊       | 87/312 [08:44<22:46,  6.07s/it]
Training Epoch: 18/100, batch 173/625 completed (loss: 0.0018:  28%|██▊       | 87/312 [08:44<22:46,  6.07s/it]
Training Epoch: 18/100, batch 173/625 completed (loss: 0.0025:  28%|██▊       | 87/312 [08:44<22:46,  6.07s/it]
Training Epoch: 18/100, batch 173/625 completed (loss: 0.0012:  28%|██▊       | 87/312 [08:44<22:46,  6.07s/it]
Training Epoch: 18/100, batch 173/625 completed (loss: 0.0005:  28%|██▊       | 87/312 [08:45<22:46,  6.07s/it]
Training Epoch: 18/100, batch 174/625 completed (loss: 0.0007:  28%|██▊       | 87/312 [08:47<22:46,  6.07s/it]
Training Epoch: 18/100, batch 174/625 completed (loss: 0.0025:  28%|██▊       | 87/312 [08:47<22:46,  6.07s/it]
Training Epoch: 18/100, batch 174/625 completed (loss: 0.0159:  28%|██▊       | 87/312 [08:47<22:46,  6.07s/it]
Training Epoch: 18/100, batch 174/625 completed (loss: 0.0007:  28%|██▊       | 87/312 [08:47<22:46,  6.07s/it]
Training Epoch: 18/100, batch 174/625 completed (loss: 0.0025:  28%|██▊       | 88/312 [08:50<22:24,  6.00s/it]
Training Epoch: 18/100, batch 174/625 completed (loss: 0.0159:  28%|██▊       | 88/312 [08:50<22:25,  6.00s/it]
Training Epoch: 18/100, batch 174/625 completed (loss: 0.0007:  28%|██▊       | 88/312 [08:50<22:25,  6.01s/it]
Training Epoch: 18/100, batch 174/625 completed (loss: 0.0007:  28%|██▊       | 88/312 [08:50<22:25,  6.01s/it]
Training Epoch: 18/100, batch 175/625 completed (loss: 0.0749:  28%|██▊       | 88/312 [08:50<22:25,  6.00s/it]
Training Epoch: 18/100, batch 175/625 completed (loss: 0.2364:  28%|██▊       | 88/312 [08:50<22:24,  6.00s/it]
Training Epoch: 18/100, batch 175/625 completed (loss: 0.0003:  28%|██▊       | 88/312 [08:50<22:25,  6.01s/it]
Training Epoch: 18/100, batch 175/625 completed (loss: 0.0024:  28%|██▊       | 88/312 [08:50<22:25,  6.01s/it]
Training Epoch: 18/100, batch 176/625 completed (loss: 0.0063:  28%|██▊       | 88/312 [08:54<22:24,  6.00s/it]
Training Epoch: 18/100, batch 176/625 completed (loss: 0.0014:  28%|██▊       | 88/312 [08:54<22:25,  6.01s/it]
Training Epoch: 18/100, batch 176/625 completed (loss: 0.0025:  28%|██▊       | 88/312 [08:54<22:25,  6.00s/it]
Training Epoch: 18/100, batch 176/625 completed (loss: 0.0006:  28%|██▊       | 88/312 [08:54<22:25,  6.01s/it]
Training Epoch: 18/100, batch 176/625 completed (loss: 0.0014:  29%|██▊       | 89/312 [08:57<22:50,  6.15s/it]
Training Epoch: 18/100, batch 176/625 completed (loss: 0.0025:  29%|██▊       | 89/312 [08:57<22:50,  6.15s/it]
Training Epoch: 18/100, batch 176/625 completed (loss: 0.0063:  29%|██▊       | 89/312 [08:57<22:50,  6.15s/it]
Training Epoch: 18/100, batch 176/625 completed (loss: 0.0006:  29%|██▊       | 89/312 [08:57<22:50,  6.15s/it]
Training Epoch: 18/100, batch 177/625 completed (loss: 0.0368:  29%|██▊       | 89/312 [08:57<22:50,  6.15s/it]
Training Epoch: 18/100, batch 177/625 completed (loss: 0.0067:  29%|██▊       | 89/312 [08:57<22:50,  6.15s/it]
Training Epoch: 18/100, batch 177/625 completed (loss: 0.0032:  29%|██▊       | 89/312 [08:57<22:50,  6.15s/it]
Training Epoch: 18/100, batch 177/625 completed (loss: 0.0005:  29%|██▊       | 89/312 [08:57<22:50,  6.15s/it]
Training Epoch: 18/100, batch 178/625 completed (loss: 0.0076:  29%|██▊       | 89/312 [09:00<22:50,  6.15s/it]
Training Epoch: 18/100, batch 178/625 completed (loss: 0.0001:  29%|██▊       | 89/312 [09:00<22:50,  6.15s/it]
Training Epoch: 18/100, batch 178/625 completed (loss: 0.0237:  29%|██▊       | 89/312 [09:00<22:50,  6.15s/it]
Training Epoch: 18/100, batch 178/625 completed (loss: 0.0025:  29%|██▊       | 89/312 [09:00<22:50,  6.15s/it]
Training Epoch: 18/100, batch 178/625 completed (loss: 0.0001:  29%|██▉       | 90/312 [09:03<22:23,  6.05s/it]
Training Epoch: 18/100, batch 178/625 completed (loss: 0.0076:  29%|██▉       | 90/312 [09:03<22:23,  6.05s/it]
Training Epoch: 18/100, batch 178/625 completed (loss: 0.0025:  29%|██▉       | 90/312 [09:03<22:23,  6.05s/it]
Training Epoch: 18/100, batch 178/625 completed (loss: 0.0237:  29%|██▉       | 90/312 [09:03<22:23,  6.05s/it]
Training Epoch: 18/100, batch 179/625 completed (loss: 0.0023:  29%|██▉       | 90/312 [09:03<22:23,  6.05s/it]
Training Epoch: 18/100, batch 179/625 completed (loss: 0.0004:  29%|██▉       | 90/312 [09:03<22:23,  6.05s/it]
Training Epoch: 18/100, batch 179/625 completed (loss: 0.0011:  29%|██▉       | 90/312 [09:03<22:23,  6.05s/it]
Training Epoch: 18/100, batch 179/625 completed (loss: 0.0028:  29%|██▉       | 90/312 [09:03<22:23,  6.05s/it]
Training Epoch: 18/100, batch 180/625 completed (loss: 0.0001:  29%|██▉       | 90/312 [09:06<22:23,  6.05s/it]
Training Epoch: 18/100, batch 180/625 completed (loss: 0.001:  29%|██▉       | 90/312 [09:06<22:23,  6.05s/it] 
Training Epoch: 18/100, batch 180/625 completed (loss: 0.0014:  29%|██▉       | 90/312 [09:06<22:23,  6.05s/it]
Training Epoch: 18/100, batch 180/625 completed (loss: 0.003:  29%|██▉       | 90/312 [09:06<22:23,  6.05s/it] 
Training Epoch: 18/100, batch 180/625 completed (loss: 0.001:  29%|██▉       | 91/312 [09:09<22:12,  6.03s/it]
Training Epoch: 18/100, batch 180/625 completed (loss: 0.0001:  29%|██▉       | 91/312 [09:09<22:13,  6.03s/it]
Training Epoch: 18/100, batch 180/625 completed (loss: 0.003:  29%|██▉       | 91/312 [09:09<22:13,  6.03s/it]
Training Epoch: 18/100, batch 180/625 completed (loss: 0.0014:  29%|██▉       | 91/312 [09:09<22:14,  6.04s/it]
Training Epoch: 18/100, batch 181/625 completed (loss: 0.0093:  29%|██▉       | 91/312 [09:09<22:12,  6.03s/it]
Training Epoch: 18/100, batch 181/625 completed (loss: 0.0012:  29%|██▉       | 91/312 [09:09<22:13,  6.03s/it]
Training Epoch: 18/100, batch 181/625 completed (loss: 0.1532:  29%|██▉       | 91/312 [09:09<22:14,  6.04s/it]
Training Epoch: 18/100, batch 181/625 completed (loss: 0.0037:  29%|██▉       | 91/312 [09:09<22:13,  6.03s/it]
Training Epoch: 18/100, batch 182/625 completed (loss: 0.0004:  29%|██▉       | 91/312 [09:12<22:14,  6.04s/it]
Training Epoch: 18/100, batch 182/625 completed (loss: 0.0001:  29%|██▉       | 91/312 [09:12<22:12,  6.03s/it]
Training Epoch: 18/100, batch 182/625 completed (loss: 0.0215:  29%|██▉       | 91/312 [09:12<22:13,  6.03s/it]
Training Epoch: 18/100, batch 182/625 completed (loss: 0.0:  29%|██▉       | 91/312 [09:12<22:13,  6.03s/it]   
Training Epoch: 18/100, batch 182/625 completed (loss: 0.0215:  29%|██▉       | 92/312 [09:15<22:26,  6.12s/it]
Training Epoch: 18/100, batch 182/625 completed (loss: 0.0004:  29%|██▉       | 92/312 [09:15<22:26,  6.12s/it]
Training Epoch: 18/100, batch 182/625 completed (loss: 0.0:  29%|██▉       | 92/312 [09:15<22:27,  6.12s/it]
Training Epoch: 18/100, batch 182/625 completed (loss: 0.0001:  29%|██▉       | 92/312 [09:15<22:28,  6.13s/it]
Training Epoch: 18/100, batch 183/625 completed (loss: 0.001:  29%|██▉       | 92/312 [09:15<22:28,  6.13s/it] 
Training Epoch: 18/100, batch 183/625 completed (loss: 0.003:  29%|██▉       | 92/312 [09:15<22:26,  6.12s/it] 
Training Epoch: 18/100, batch 183/625 completed (loss: 0.0062:  29%|██▉       | 92/312 [09:15<22:26,  6.12s/it]
Training Epoch: 18/100, batch 183/625 completed (loss: 0.0018:  29%|██▉       | 92/312 [09:15<22:27,  6.12s/it]
Training Epoch: 18/100, batch 184/625 completed (loss: 0.0001:  29%|██▉       | 92/312 [09:18<22:28,  6.13s/it]
Training Epoch: 18/100, batch 184/625 completed (loss: 0.0005:  29%|██▉       | 92/312 [09:18<22:26,  6.12s/it]
Training Epoch: 18/100, batch 184/625 completed (loss: 0.0073:  29%|██▉       | 92/312 [09:18<22:26,  6.12s/it]
Training Epoch: 18/100, batch 184/625 completed (loss: 0.0001:  29%|██▉       | 92/312 [09:18<22:27,  6.12s/it]
Training Epoch: 18/100, batch 184/625 completed (loss: 0.0001:  30%|██▉       | 93/312 [09:21<22:03,  6.04s/it]
Training Epoch: 18/100, batch 184/625 completed (loss: 0.0005:  30%|██▉       | 93/312 [09:21<22:03,  6.04s/it]
Training Epoch: 18/100, batch 184/625 completed (loss: 0.0073:  30%|██▉       | 93/312 [09:21<22:03,  6.05s/it]
Training Epoch: 18/100, batch 184/625 completed (loss: 0.0001:  30%|██▉       | 93/312 [09:21<22:04,  6.05s/it]
Training Epoch: 18/100, batch 185/625 completed (loss: 0.0007:  30%|██▉       | 93/312 [09:21<22:03,  6.05s/it]
Training Epoch: 18/100, batch 185/625 completed (loss: 0.005:  30%|██▉       | 93/312 [09:21<22:03,  6.04s/it] 
Training Epoch: 18/100, batch 185/625 completed (loss: 0.0154:  30%|██▉       | 93/312 [09:21<22:03,  6.04s/it]
Training Epoch: 18/100, batch 185/625 completed (loss: 0.0164:  30%|██▉       | 93/312 [09:21<22:04,  6.05s/it]
Training Epoch: 18/100, batch 186/625 completed (loss: 0.0314:  30%|██▉       | 93/312 [09:24<22:03,  6.05s/it]
Training Epoch: 18/100, batch 186/625 completed (loss: 0.0015:  30%|██▉       | 93/312 [09:24<22:03,  6.04s/it]
Training Epoch: 18/100, batch 186/625 completed (loss: 0.0014:  30%|██▉       | 93/312 [09:24<22:03,  6.04s/it]
Training Epoch: 18/100, batch 186/625 completed (loss: 0.0001:  30%|██▉       | 93/312 [09:24<22:04,  6.05s/it]
Training Epoch: 18/100, batch 186/625 completed (loss: 0.0015:  30%|███       | 94/312 [09:27<22:08,  6.09s/it]
Training Epoch: 18/100, batch 186/625 completed (loss: 0.0014:  30%|███       | 94/312 [09:27<22:08,  6.09s/it]
Training Epoch: 18/100, batch 186/625 completed (loss: 0.0314:  30%|███       | 94/312 [09:27<22:08,  6.09s/it]
Training Epoch: 18/100, batch 186/625 completed (loss: 0.0001:  30%|███       | 94/312 [09:27<22:08,  6.09s/it]
Training Epoch: 18/100, batch 187/625 completed (loss: 0.0008:  30%|███       | 94/312 [09:27<22:08,  6.09s/it]
Training Epoch: 18/100, batch 187/625 completed (loss: 0.0076:  30%|███       | 94/312 [09:27<22:08,  6.09s/it]
Training Epoch: 18/100, batch 187/625 completed (loss: 0.0001:  30%|███       | 94/312 [09:27<22:08,  6.09s/it]
Training Epoch: 18/100, batch 187/625 completed (loss: 0.0006:  30%|███       | 94/312 [09:27<22:08,  6.09s/it]
Training Epoch: 18/100, batch 188/625 completed (loss: 0.002:  30%|███       | 94/312 [09:30<22:08,  6.09s/it] 
Training Epoch: 18/100, batch 188/625 completed (loss: 0.0:  30%|███       | 94/312 [09:30<22:08,  6.09s/it]   
Training Epoch: 18/100, batch 188/625 completed (loss: 0.0064:  30%|███       | 94/312 [09:30<22:08,  6.09s/it]
Training Epoch: 18/100, batch 188/625 completed (loss: 0.0114:  30%|███       | 94/312 [09:30<22:08,  6.09s/it]
Training Epoch: 18/100, batch 188/625 completed (loss: 0.0064:  30%|███       | 95/312 [09:33<21:58,  6.08s/it]
Training Epoch: 18/100, batch 188/625 completed (loss: 0.0114:  30%|███       | 95/312 [09:33<21:58,  6.08s/it]
Training Epoch: 18/100, batch 188/625 completed (loss: 0.0:  30%|███       | 95/312 [09:33<21:59,  6.08s/it]
Training Epoch: 18/100, batch 188/625 completed (loss: 0.002:  30%|███       | 95/312 [09:33<21:59,  6.08s/it]
Training Epoch: 18/100, batch 189/625 completed (loss: 0.0014:  30%|███       | 95/312 [09:33<21:59,  6.08s/it]
Training Epoch: 18/100, batch 189/625 completed (loss: 0.0355:  30%|███       | 95/312 [09:33<21:59,  6.08s/it]
Training Epoch: 18/100, batch 189/625 completed (loss: 0.0001:  30%|███       | 95/312 [09:33<21:58,  6.08s/it]
Training Epoch: 18/100, batch 189/625 completed (loss: 0.0027:  30%|███       | 95/312 [09:33<21:58,  6.08s/it]
Training Epoch: 18/100, batch 190/625 completed (loss: 0.0003:  30%|███       | 95/312 [09:36<21:59,  6.08s/it]
Training Epoch: 18/100, batch 190/625 completed (loss: 0.0005:  30%|███       | 95/312 [09:36<21:59,  6.08s/it]
Training Epoch: 18/100, batch 190/625 completed (loss: 0.0365:  30%|███       | 95/312 [09:36<21:58,  6.08s/it]
Training Epoch: 18/100, batch 190/625 completed (loss: 0.0058:  30%|███       | 95/312 [09:36<21:58,  6.08s/it]
Training Epoch: 18/100, batch 190/625 completed (loss: 0.0005:  31%|███       | 96/312 [09:39<21:49,  6.06s/it]
Training Epoch: 18/100, batch 190/625 completed (loss: 0.0003:  31%|███       | 96/312 [09:39<21:50,  6.07s/it]
Training Epoch: 18/100, batch 190/625 completed (loss: 0.0058:  31%|███       | 96/312 [09:39<21:50,  6.07s/it]
Training Epoch: 18/100, batch 190/625 completed (loss: 0.0365:  31%|███       | 96/312 [09:39<21:50,  6.07s/it]
Training Epoch: 18/100, batch 191/625 completed (loss: 0.001:  31%|███       | 96/312 [09:39<21:50,  6.07s/it] 
Training Epoch: 18/100, batch 191/625 completed (loss: 0.0006:  31%|███       | 96/312 [09:39<21:49,  6.06s/it]
Training Epoch: 18/100, batch 191/625 completed (loss: 0.0215:  31%|███       | 96/312 [09:39<21:50,  6.07s/it]
Training Epoch: 18/100, batch 191/625 completed (loss: 0.0143:  31%|███       | 96/312 [09:39<21:50,  6.07s/it]
Training Epoch: 18/100, batch 192/625 completed (loss: 0.001:  31%|███       | 96/312 [09:42<21:50,  6.07s/it] 
Training Epoch: 18/100, batch 192/625 completed (loss: 0.0014:  31%|███       | 96/312 [09:42<21:49,  6.06s/it]
Training Epoch: 18/100, batch 192/625 completed (loss: 0.061:  31%|███       | 96/312 [09:42<21:50,  6.07s/it]
Training Epoch: 18/100, batch 192/625 completed (loss: 0.0009:  31%|███       | 96/312 [09:42<21:50,  6.07s/it]
Training Epoch: 18/100, batch 192/625 completed (loss: 0.0014:  31%|███       | 97/312 [09:45<22:02,  6.15s/it]
Training Epoch: 18/100, batch 192/625 completed (loss: 0.0009:  31%|███       | 97/312 [09:45<22:02,  6.15s/it]
Training Epoch: 18/100, batch 192/625 completed (loss: 0.061:  31%|███       | 97/312 [09:45<22:02,  6.15s/it]
Training Epoch: 18/100, batch 192/625 completed (loss: 0.001:  31%|███       | 97/312 [09:45<22:02,  6.15s/it]
Training Epoch: 18/100, batch 193/625 completed (loss: 0.0014:  31%|███       | 97/312 [09:45<22:02,  6.15s/it]
Training Epoch: 18/100, batch 193/625 completed (loss: 0.0086:  31%|███       | 97/312 [09:45<22:02,  6.15s/it]
Training Epoch: 18/100, batch 193/625 completed (loss: 0.0091:  31%|███       | 97/312 [09:45<22:02,  6.15s/it]
Training Epoch: 18/100, batch 193/625 completed (loss: 0.0:  31%|███       | 97/312 [09:45<22:02,  6.15s/it]   
Training Epoch: 18/100, batch 194/625 completed (loss: 0.0028:  31%|███       | 97/312 [09:48<22:02,  6.15s/it]
Training Epoch: 18/100, batch 194/625 completed (loss: 0.0006:  31%|███       | 97/312 [09:48<22:02,  6.15s/it]
Training Epoch: 18/100, batch 194/625 completed (loss: 0.0586:  31%|███       | 97/312 [09:48<22:02,  6.15s/it]
Training Epoch: 18/100, batch 194/625 completed (loss: 0.0058:  31%|███       | 97/312 [09:48<22:02,  6.15s/it]
Training Epoch: 18/100, batch 194/625 completed (loss: 0.0006:  31%|███▏      | 98/312 [09:51<21:36,  6.06s/it]
Training Epoch: 18/100, batch 194/625 completed (loss: 0.0586:  31%|███▏      | 98/312 [09:51<21:36,  6.06s/it]
Training Epoch: 18/100, batch 194/625 completed (loss: 0.0058:  31%|███▏      | 98/312 [09:51<21:36,  6.06s/it]
Training Epoch: 18/100, batch 194/625 completed (loss: 0.0028:  31%|███▏      | 98/312 [09:51<21:36,  6.06s/it]
Training Epoch: 18/100, batch 195/625 completed (loss: 0.0007:  31%|███▏      | 98/312 [09:51<21:36,  6.06s/it]
Training Epoch: 18/100, batch 195/625 completed (loss: 0.0024:  31%|███▏      | 98/312 [09:51<21:36,  6.06s/it]
Training Epoch: 18/100, batch 195/625 completed (loss: 0.0282:  31%|███▏      | 98/312 [09:51<21:36,  6.06s/it]
Training Epoch: 18/100, batch 195/625 completed (loss: 0.0013:  31%|███▏      | 98/312 [09:51<21:36,  6.06s/it]
Training Epoch: 18/100, batch 196/625 completed (loss: 0.0111:  31%|███▏      | 98/312 [09:54<21:36,  6.06s/it]
Training Epoch: 18/100, batch 196/625 completed (loss: 0.0049:  31%|███▏      | 98/312 [09:54<21:36,  6.06s/it]
Training Epoch: 18/100, batch 196/625 completed (loss: 0.0016:  31%|███▏      | 98/312 [09:54<21:36,  6.06s/it]
Training Epoch: 18/100, batch 196/625 completed (loss: 0.0003:  31%|███▏      | 98/312 [09:54<21:36,  6.06s/it]
Training Epoch: 18/100, batch 196/625 completed (loss: 0.0111:  32%|███▏      | 99/312 [09:58<21:45,  6.13s/it]
Training Epoch: 18/100, batch 196/625 completed (loss: 0.0003:  32%|███▏      | 99/312 [09:58<21:45,  6.13s/it]
Training Epoch: 18/100, batch 196/625 completed (loss: 0.0016:  32%|███▏      | 99/312 [09:58<21:45,  6.13s/it]
Training Epoch: 18/100, batch 196/625 completed (loss: 0.0049:  32%|███▏      | 99/312 [09:58<21:46,  6.13s/it]
Training Epoch: 18/100, batch 197/625 completed (loss: 0.0005:  32%|███▏      | 99/312 [09:58<21:46,  6.13s/it]
Training Epoch: 18/100, batch 197/625 completed (loss: 0.0011:  32%|███▏      | 99/312 [09:58<21:45,  6.13s/it]
Training Epoch: 18/100, batch 197/625 completed (loss: 0.0008:  32%|███▏      | 99/312 [09:58<21:45,  6.13s/it]
Training Epoch: 18/100, batch 197/625 completed (loss: 0.0008:  32%|███▏      | 99/312 [09:58<21:45,  6.13s/it]
Training Epoch: 18/100, batch 198/625 completed (loss: 0.0022:  32%|███▏      | 99/312 [10:00<21:45,  6.13s/it]
Training Epoch: 18/100, batch 198/625 completed (loss: 0.0001:  32%|███▏      | 99/312 [10:00<21:46,  6.13s/it]
Training Epoch: 18/100, batch 198/625 completed (loss: 0.0198:  32%|███▏      | 99/312 [10:00<21:45,  6.13s/it]
Training Epoch: 18/100, batch 198/625 completed (loss: 0.0012:  32%|███▏      | 99/312 [10:01<21:45,  6.13s/it]
Training Epoch: 18/100, batch 198/625 completed (loss: 0.0198:  32%|███▏      | 100/312 [10:04<21:34,  6.10s/it]
Training Epoch: 18/100, batch 198/625 completed (loss: 0.0001:  32%|███▏      | 100/312 [10:04<21:34,  6.11s/it]
Training Epoch: 18/100, batch 198/625 completed (loss: 0.0012:  32%|███▏      | 100/312 [10:04<21:34,  6.11s/it]
Training Epoch: 18/100, batch 198/625 completed (loss: 0.0022:  32%|███▏      | 100/312 [10:04<21:35,  6.11s/it]
Training Epoch: 18/100, batch 199/625 completed (loss: 0.0013:  32%|███▏      | 100/312 [10:04<21:35,  6.11s/it]
Training Epoch: 18/100, batch 199/625 completed (loss: 0.0002:  32%|███▏      | 100/312 [10:04<21:34,  6.10s/it]
Training Epoch: 18/100, batch 199/625 completed (loss: 0.0025:  32%|███▏      | 100/312 [10:04<21:34,  6.11s/it]
Training Epoch: 18/100, batch 199/625 completed (loss: 0.0046:  32%|███▏      | 100/312 [10:04<21:34,  6.11s/it]
Training Epoch: 18/100, batch 200/625 completed (loss: 0.0007:  32%|███▏      | 100/312 [10:07<21:34,  6.10s/it]
Training Epoch: 18/100, batch 200/625 completed (loss: 0.0007:  32%|███▏      | 100/312 [10:07<21:34,  6.11s/it]
Training Epoch: 18/100, batch 200/625 completed (loss: 0.0125:  32%|███▏      | 100/312 [10:07<21:35,  6.11s/it]
Training Epoch: 18/100, batch 200/625 completed (loss: 0.0018:  32%|███▏      | 100/312 [10:07<21:34,  6.11s/it]
Training Epoch: 18/100, batch 200/625 completed (loss: 0.0007:  32%|███▏      | 101/312 [10:09<21:11,  6.03s/it]
Training Epoch: 18/100, batch 200/625 completed (loss: 0.0007:  32%|███▏      | 101/312 [10:09<21:12,  6.03s/it]
Training Epoch: 18/100, batch 200/625 completed (loss: 0.0125:  32%|███▏      | 101/312 [10:09<21:12,  6.03s/it]
Training Epoch: 18/100, batch 200/625 completed (loss: 0.0018:  32%|███▏      | 101/312 [10:09<21:12,  6.03s/it]
Training Epoch: 18/100, batch 201/625 completed (loss: 0.0197:  32%|███▏      | 101/312 [10:09<21:12,  6.03s/it]
Training Epoch: 18/100, batch 201/625 completed (loss: 0.0005:  32%|███▏      | 101/312 [10:09<21:11,  6.03s/it]
Training Epoch: 18/100, batch 201/625 completed (loss: 0.0016:  32%|███▏      | 101/312 [10:09<21:12,  6.03s/it]
Training Epoch: 18/100, batch 201/625 completed (loss: 0.0176:  32%|███▏      | 101/312 [10:10<21:12,  6.03s/it]
Training Epoch: 18/100, batch 202/625 completed (loss: 0.0045:  32%|███▏      | 101/312 [10:13<21:11,  6.03s/it]
Training Epoch: 18/100, batch 202/625 completed (loss: 0.0017:  32%|███▏      | 101/312 [10:13<21:12,  6.03s/it]
Training Epoch: 18/100, batch 202/625 completed (loss: 0.0335:  32%|███▏      | 101/312 [10:13<21:12,  6.03s/it]
Training Epoch: 18/100, batch 202/625 completed (loss: 0.0018:  32%|███▏      | 101/312 [10:13<21:12,  6.03s/it]
Training Epoch: 18/100, batch 202/625 completed (loss: 0.0017:  33%|███▎      | 102/312 [10:16<21:24,  6.12s/it]
Training Epoch: 18/100, batch 202/625 completed (loss: 0.0335:  33%|███▎      | 102/312 [10:16<21:25,  6.12s/it]
Training Epoch: 18/100, batch 202/625 completed (loss: 0.0018:  33%|███▎      | 102/312 [10:16<21:25,  6.12s/it]
Training Epoch: 18/100, batch 202/625 completed (loss: 0.0045:  33%|███▎      | 102/312 [10:16<21:25,  6.12s/it]
Training Epoch: 18/100, batch 203/625 completed (loss: 0.0087:  33%|███▎      | 102/312 [10:16<21:25,  6.12s/it]
Training Epoch: 18/100, batch 203/625 completed (loss: 0.0021:  33%|███▎      | 102/312 [10:16<21:25,  6.12s/it]
Training Epoch: 18/100, batch 203/625 completed (loss: 0.0002:  33%|███▎      | 102/312 [10:16<21:24,  6.12s/it]
Training Epoch: 18/100, batch 203/625 completed (loss: 0.0008:  33%|███▎      | 102/312 [10:16<21:25,  6.12s/it]
Training Epoch: 18/100, batch 204/625 completed (loss: 0.0145:  33%|███▎      | 102/312 [10:19<21:25,  6.12s/it]
Training Epoch: 18/100, batch 204/625 completed (loss: 0.002:  33%|███▎      | 102/312 [10:19<21:25,  6.12s/it] 
Training Epoch: 18/100, batch 204/625 completed (loss: 0.0597:  33%|███▎      | 102/312 [10:19<21:24,  6.12s/it]
Training Epoch: 18/100, batch 204/625 completed (loss: 0.0097:  33%|███▎      | 102/312 [10:19<21:25,  6.12s/it]
Training Epoch: 18/100, batch 204/625 completed (loss: 0.0597:  33%|███▎      | 103/312 [10:22<21:13,  6.09s/it]
Training Epoch: 18/100, batch 204/625 completed (loss: 0.0145:  33%|███▎      | 103/312 [10:22<21:13,  6.09s/it]
Training Epoch: 18/100, batch 204/625 completed (loss: 0.002:  33%|███▎      | 103/312 [10:22<21:13,  6.09s/it]
Training Epoch: 18/100, batch 204/625 completed (loss: 0.0097:  33%|███▎      | 103/312 [10:22<21:13,  6.09s/it]
Training Epoch: 18/100, batch 205/625 completed (loss: 0.0082:  33%|███▎      | 103/312 [10:22<21:13,  6.09s/it]
Training Epoch: 18/100, batch 205/625 completed (loss: 0.0073:  33%|███▎      | 103/312 [10:22<21:13,  6.09s/it]
Training Epoch: 18/100, batch 205/625 completed (loss: 0.0029:  33%|███▎      | 103/312 [10:22<21:13,  6.09s/it]
Training Epoch: 18/100, batch 205/625 completed (loss: 0.0033:  33%|███▎      | 103/312 [10:22<21:13,  6.09s/it]
Training Epoch: 18/100, batch 206/625 completed (loss: 0.0174:  33%|███▎      | 103/312 [10:25<21:13,  6.09s/it]
Training Epoch: 18/100, batch 206/625 completed (loss: 0.0081:  33%|███▎      | 103/312 [10:25<21:13,  6.09s/it]
Training Epoch: 18/100, batch 206/625 completed (loss: 0.0019:  33%|███▎      | 103/312 [10:25<21:13,  6.09s/it]
Training Epoch: 18/100, batch 206/625 completed (loss: 0.0158:  33%|███▎      | 103/312 [10:25<21:13,  6.09s/it]
Training Epoch: 18/100, batch 206/625 completed (loss: 0.0081:  33%|███▎      | 104/312 [10:28<21:22,  6.16s/it]
Training Epoch: 18/100, batch 206/625 completed (loss: 0.0019:  33%|███▎      | 104/312 [10:28<21:22,  6.16s/it]
Training Epoch: 18/100, batch 206/625 completed (loss: 0.0174:  33%|███▎      | 104/312 [10:28<21:22,  6.17s/it]
Training Epoch: 18/100, batch 206/625 completed (loss: 0.0158:  33%|███▎      | 104/312 [10:28<21:23,  6.17s/it]
Training Epoch: 18/100, batch 207/625 completed (loss: 0.0336:  33%|███▎      | 104/312 [10:28<21:22,  6.16s/it]
Training Epoch: 18/100, batch 207/625 completed (loss: 0.0115:  33%|███▎      | 104/312 [10:28<21:22,  6.16s/it]
Training Epoch: 18/100, batch 207/625 completed (loss: 0.0625:  33%|███▎      | 104/312 [10:28<21:22,  6.17s/it]
Training Epoch: 18/100, batch 207/625 completed (loss: 0.0069:  33%|███▎      | 104/312 [10:28<21:23,  6.17s/it]
Training Epoch: 18/100, batch 208/625 completed (loss: 0.0036:  33%|███▎      | 104/312 [10:31<21:22,  6.16s/it]
Training Epoch: 18/100, batch 208/625 completed (loss: 0.0038:  33%|███▎      | 104/312 [10:31<21:22,  6.17s/it]
Training Epoch: 18/100, batch 208/625 completed (loss: 0.0368:  33%|███▎      | 104/312 [10:31<21:22,  6.16s/it]
Training Epoch: 18/100, batch 208/625 completed (loss: 0.0294:  33%|███▎      | 104/312 [10:31<21:23,  6.17s/it]
Training Epoch: 18/100, batch 208/625 completed (loss: 0.0038:  34%|███▎      | 105/312 [10:34<20:57,  6.08s/it]
Training Epoch: 18/100, batch 208/625 completed (loss: 0.0294:  34%|███▎      | 105/312 [10:34<20:57,  6.07s/it]
Training Epoch: 18/100, batch 208/625 completed (loss: 0.0036:  34%|███▎      | 105/312 [10:34<20:57,  6.08s/it]
Training Epoch: 18/100, batch 208/625 completed (loss: 0.0368:  34%|███▎      | 105/312 [10:34<20:58,  6.08s/it]
Training Epoch: 18/100, batch 209/625 completed (loss: 0.0009:  34%|███▎      | 105/312 [10:34<20:57,  6.08s/it]
Training Epoch: 18/100, batch 209/625 completed (loss: 0.0066:  34%|███▎      | 105/312 [10:34<20:58,  6.08s/it]
Training Epoch: 18/100, batch 209/625 completed (loss: 0.0021:  34%|███▎      | 105/312 [10:34<20:57,  6.08s/it]
Training Epoch: 18/100, batch 209/625 completed (loss: 0.0065:  34%|███▎      | 105/312 [10:34<20:57,  6.07s/it]
Training Epoch: 18/100, batch 210/625 completed (loss: 0.0051:  34%|███▎      | 105/312 [10:37<20:57,  6.08s/it]
Training Epoch: 18/100, batch 210/625 completed (loss: 0.0207:  34%|███▎      | 105/312 [10:37<20:58,  6.08s/it]
Training Epoch: 18/100, batch 210/625 completed (loss: 0.0016:  34%|███▎      | 105/312 [10:37<20:57,  6.08s/it]
Training Epoch: 18/100, batch 210/625 completed (loss: 0.0003:  34%|███▎      | 105/312 [10:37<20:57,  6.07s/it]
Training Epoch: 18/100, batch 210/625 completed (loss: 0.0207:  34%|███▍      | 106/312 [10:40<20:47,  6.06s/it]
Training Epoch: 18/100, batch 210/625 completed (loss: 0.0016:  34%|███▍      | 106/312 [10:40<20:47,  6.06s/it]
Training Epoch: 18/100, batch 210/625 completed (loss: 0.0051:  34%|███▍      | 106/312 [10:40<20:47,  6.06s/it]
Training Epoch: 18/100, batch 210/625 completed (loss: 0.0003:  34%|███▍      | 106/312 [10:40<20:47,  6.06s/it]
Training Epoch: 18/100, batch 211/625 completed (loss: 0.0147:  34%|███▍      | 106/312 [10:40<20:47,  6.06s/it]
Training Epoch: 18/100, batch 211/625 completed (loss: 0.0184:  34%|███▍      | 106/312 [10:40<20:47,  6.06s/it]
Training Epoch: 18/100, batch 211/625 completed (loss: 0.0011:  34%|███▍      | 106/312 [10:40<20:47,  6.06s/it]
Training Epoch: 18/100, batch 211/625 completed (loss: 0.0048:  34%|███▍      | 106/312 [10:40<20:47,  6.06s/it]
Training Epoch: 18/100, batch 212/625 completed (loss: 0.0047:  34%|███▍      | 106/312 [10:43<20:47,  6.06s/it]
Training Epoch: 18/100, batch 212/625 completed (loss: 0.0052:  34%|███▍      | 106/312 [10:43<20:47,  6.06s/it]
Training Epoch: 18/100, batch 212/625 completed (loss: 0.0046:  34%|███▍      | 106/312 [10:43<20:47,  6.06s/it]
Training Epoch: 18/100, batch 212/625 completed (loss: 0.0014:  34%|███▍      | 106/312 [10:43<20:47,  6.06s/it]
Training Epoch: 18/100, batch 212/625 completed (loss: 0.0047:  34%|███▍      | 107/312 [10:46<20:55,  6.13s/it]
Training Epoch: 18/100, batch 212/625 completed (loss: 0.0052:  34%|███▍      | 107/312 [10:46<20:55,  6.13s/it]
Training Epoch: 18/100, batch 212/625 completed (loss: 0.0046:  34%|███▍      | 107/312 [10:46<20:55,  6.13s/it]
Training Epoch: 18/100, batch 212/625 completed (loss: 0.0014:  34%|███▍      | 107/312 [10:46<20:55,  6.13s/it]
Training Epoch: 18/100, batch 213/625 completed (loss: 0.0229:  34%|███▍      | 107/312 [10:46<20:55,  6.13s/it]
Training Epoch: 18/100, batch 213/625 completed (loss: 0.0177:  34%|███▍      | 107/312 [10:46<20:55,  6.13s/it]
Training Epoch: 18/100, batch 213/625 completed (loss: 0.0006:  34%|███▍      | 107/312 [10:46<20:55,  6.13s/it]
Training Epoch: 18/100, batch 213/625 completed (loss: 0.0516:  34%|███▍      | 107/312 [10:46<20:55,  6.13s/it]
Training Epoch: 18/100, batch 214/625 completed (loss: 0.0044:  34%|███▍      | 107/312 [10:49<20:55,  6.13s/it]
Training Epoch: 18/100, batch 214/625 completed (loss: 0.0028:  34%|███▍      | 107/312 [10:49<20:55,  6.13s/it]
Training Epoch: 18/100, batch 214/625 completed (loss: 0.0005:  34%|███▍      | 107/312 [10:49<20:55,  6.13s/it]
Training Epoch: 18/100, batch 214/625 completed (loss: 0.0006:  34%|███▍      | 107/312 [10:49<20:55,  6.13s/it]
Training Epoch: 18/100, batch 214/625 completed (loss: 0.0044:  35%|███▍      | 108/312 [10:52<20:42,  6.09s/it]
Training Epoch: 18/100, batch 214/625 completed (loss: 0.0005:  35%|███▍      | 108/312 [10:52<20:42,  6.09s/it]
Training Epoch: 18/100, batch 214/625 completed (loss: 0.0028:  35%|███▍      | 108/312 [10:52<20:42,  6.09s/it]
Training Epoch: 18/100, batch 214/625 completed (loss: 0.0006:  35%|███▍      | 108/312 [10:52<20:42,  6.09s/it]
Training Epoch: 18/100, batch 215/625 completed (loss: 0.0002:  35%|███▍      | 108/312 [10:52<20:42,  6.09s/it]
Training Epoch: 18/100, batch 215/625 completed (loss: 0.001:  35%|███▍      | 108/312 [10:52<20:42,  6.09s/it] 
Training Epoch: 18/100, batch 215/625 completed (loss: 0.2049:  35%|███▍      | 108/312 [10:52<20:42,  6.09s/it]
Training Epoch: 18/100, batch 215/625 completed (loss: 0.0061:  35%|███▍      | 108/312 [10:52<20:42,  6.09s/it]
Training Epoch: 18/100, batch 216/625 completed (loss: 0.0063:  35%|███▍      | 108/312 [10:55<20:42,  6.09s/it]
Training Epoch: 18/100, batch 216/625 completed (loss: 0.0436:  35%|███▍      | 108/312 [10:55<20:42,  6.09s/it]
Training Epoch: 18/100, batch 216/625 completed (loss: 0.0001:  35%|███▍      | 108/312 [10:55<20:42,  6.09s/it]
Training Epoch: 18/100, batch 216/625 completed (loss: 0.0002:  35%|███▍      | 108/312 [10:55<20:42,  6.09s/it]
Training Epoch: 18/100, batch 216/625 completed (loss: 0.0002:  35%|███▍      | 109/312 [10:58<20:31,  6.07s/it]
Training Epoch: 18/100, batch 216/625 completed (loss: 0.0063:  35%|███▍      | 109/312 [10:58<20:32,  6.07s/it]
Training Epoch: 18/100, batch 216/625 completed (loss: 0.0001:  35%|███▍      | 109/312 [10:58<20:32,  6.07s/it]
Training Epoch: 18/100, batch 216/625 completed (loss: 0.0436:  35%|███▍      | 109/312 [10:58<20:33,  6.07s/it]
Training Epoch: 18/100, batch 217/625 completed (loss: 0.0144:  35%|███▍      | 109/312 [10:58<20:32,  6.07s/it]
Training Epoch: 18/100, batch 217/625 completed (loss: 0.0121:  35%|███▍      | 109/312 [10:58<20:32,  6.07s/it]
Training Epoch: 18/100, batch 217/625 completed (loss: 0.0008:  35%|███▍      | 109/312 [10:58<20:33,  6.07s/it]
Training Epoch: 18/100, batch 217/625 completed (loss: 0.0033:  35%|███▍      | 109/312 [10:58<20:31,  6.07s/it]
Training Epoch: 18/100, batch 218/625 completed (loss: 0.0008:  35%|███▍      | 109/312 [11:01<20:32,  6.07s/it]
Training Epoch: 18/100, batch 218/625 completed (loss: 0.001:  35%|███▍      | 109/312 [11:01<20:33,  6.07s/it] 
Training Epoch: 18/100, batch 218/625 completed (loss: 0.001:  35%|███▍      | 109/312 [11:01<20:32,  6.07s/it] 
Training Epoch: 18/100, batch 218/625 completed (loss: 0.0792:  35%|███▍      | 109/312 [11:02<20:31,  6.07s/it]
Training Epoch: 18/100, batch 218/625 completed (loss: 0.001:  35%|███▌      | 110/312 [11:04<20:22,  6.05s/it]
Training Epoch: 18/100, batch 218/625 completed (loss: 0.0008:  35%|███▌      | 110/312 [11:04<20:22,  6.05s/it]
Training Epoch: 18/100, batch 218/625 completed (loss: 0.001:  35%|███▌      | 110/312 [11:04<20:22,  6.05s/it]
Training Epoch: 18/100, batch 218/625 completed (loss: 0.0792:  35%|███▌      | 110/312 [11:04<20:22,  6.05s/it]
Training Epoch: 18/100, batch 219/625 completed (loss: 0.0172:  35%|███▌      | 110/312 [11:04<20:22,  6.05s/it]
Training Epoch: 18/100, batch 219/625 completed (loss: 0.0062:  35%|███▌      | 110/312 [11:04<20:22,  6.05s/it]
Training Epoch: 18/100, batch 219/625 completed (loss: 0.0001:  35%|███▌      | 110/312 [11:04<20:22,  6.05s/it]
Training Epoch: 18/100, batch 219/625 completed (loss: 0.0076:  35%|███▌      | 110/312 [11:04<20:22,  6.05s/it]
Training Epoch: 18/100, batch 220/625 completed (loss: 0.0049:  35%|███▌      | 110/312 [11:07<20:22,  6.05s/it]
Training Epoch: 18/100, batch 220/625 completed (loss: 0.0248:  35%|███▌      | 110/312 [11:07<20:22,  6.05s/it]
Training Epoch: 18/100, batch 220/625 completed (loss: 0.1227:  35%|███▌      | 110/312 [11:07<20:22,  6.05s/it]
Training Epoch: 18/100, batch 220/625 completed (loss: 0.0011:  35%|███▌      | 110/312 [11:07<20:22,  6.05s/it]
Training Epoch: 18/100, batch 220/625 completed (loss: 0.0049:  36%|███▌      | 111/312 [11:10<20:13,  6.03s/it]
Training Epoch: 18/100, batch 220/625 completed (loss: 0.0248:  36%|███▌      | 111/312 [11:10<20:13,  6.04s/it]
Training Epoch: 18/100, batch 220/625 completed (loss: 0.1227:  36%|███▌      | 111/312 [11:10<20:13,  6.04s/it]
Training Epoch: 18/100, batch 220/625 completed (loss: 0.0011:  36%|███▌      | 111/312 [11:10<20:13,  6.04s/it]
Training Epoch: 18/100, batch 221/625 completed (loss: 0.0003:  36%|███▌      | 111/312 [11:10<20:13,  6.04s/it]
Training Epoch: 18/100, batch 221/625 completed (loss: 0.0007:  36%|███▌      | 111/312 [11:10<20:13,  6.04s/it]
Training Epoch: 18/100, batch 221/625 completed (loss: 0.0027:  36%|███▌      | 111/312 [11:10<20:13,  6.03s/it]
Training Epoch: 18/100, batch 221/625 completed (loss: 0.0005:  36%|███▌      | 111/312 [11:10<20:13,  6.04s/it]
Training Epoch: 18/100, batch 222/625 completed (loss: 0.0002:  36%|███▌      | 111/312 [11:13<20:13,  6.03s/it]
Training Epoch: 18/100, batch 222/625 completed (loss: 0.0004:  36%|███▌      | 111/312 [11:13<20:13,  6.04s/it]
Training Epoch: 18/100, batch 222/625 completed (loss: 0.0017:  36%|███▌      | 111/312 [11:13<20:13,  6.04s/it]
Training Epoch: 18/100, batch 222/625 completed (loss: 0.0001:  36%|███▌      | 111/312 [11:13<20:13,  6.04s/it]
Training Epoch: 18/100, batch 222/625 completed (loss: 0.0001:  36%|███▌      | 112/312 [11:17<20:15,  6.08s/it]
Training Epoch: 18/100, batch 222/625 completed (loss: 0.0002:  36%|███▌      | 112/312 [11:16<20:15,  6.08s/it]
Training Epoch: 18/100, batch 222/625 completed (loss: 0.0017:  36%|███▌      | 112/312 [11:17<20:16,  6.08s/it]
Training Epoch: 18/100, batch 222/625 completed (loss: 0.0004:  36%|███▌      | 112/312 [11:17<20:16,  6.08s/it]
Training Epoch: 18/100, batch 223/625 completed (loss: 0.0015:  36%|███▌      | 112/312 [11:17<20:16,  6.08s/it]
Training Epoch: 18/100, batch 223/625 completed (loss: 0.0002:  36%|███▌      | 112/312 [11:17<20:15,  6.08s/it]
Training Epoch: 18/100, batch 223/625 completed (loss: 0.2805:  36%|███▌      | 112/312 [11:17<20:16,  6.08s/it]
Training Epoch: 18/100, batch 223/625 completed (loss: 0.0005:  36%|███▌      | 112/312 [11:17<20:15,  6.08s/it]
Training Epoch: 18/100, batch 224/625 completed (loss: 0.0059:  36%|███▌      | 112/312 [11:19<20:15,  6.08s/it]
Training Epoch: 18/100, batch 224/625 completed (loss: 0.0008:  36%|███▌      | 112/312 [11:19<20:16,  6.08s/it]
Training Epoch: 18/100, batch 224/625 completed (loss: 0.0002:  36%|███▌      | 112/312 [11:19<20:16,  6.08s/it]
Training Epoch: 18/100, batch 224/625 completed (loss: 0.0017:  36%|███▌      | 112/312 [11:20<20:15,  6.08s/it]
Training Epoch: 18/100, batch 224/625 completed (loss: 0.0002:  36%|███▌      | 113/312 [11:22<19:54,  6.00s/it]
Training Epoch: 18/100, batch 224/625 completed (loss: 0.0008:  36%|███▌      | 113/312 [11:22<19:55,  6.01s/it]
Training Epoch: 18/100, batch 224/625 completed (loss: 0.0059:  36%|███▌      | 113/312 [11:22<19:55,  6.01s/it]
Training Epoch: 18/100, batch 224/625 completed (loss: 0.0017:  36%|███▌      | 113/312 [11:22<19:55,  6.01s/it]
Training Epoch: 18/100, batch 225/625 completed (loss: 0.016:  36%|███▌      | 113/312 [11:22<19:54,  6.00s/it] 
Training Epoch: 18/100, batch 225/625 completed (loss: 0.0622:  36%|███▌      | 113/312 [11:22<19:55,  6.01s/it]
Training Epoch: 18/100, batch 225/625 completed (loss: 0.0343:  36%|███▌      | 113/312 [11:22<19:55,  6.01s/it]
Training Epoch: 18/100, batch 225/625 completed (loss: 0.0147:  36%|███▌      | 113/312 [11:22<19:55,  6.01s/it]
Training Epoch: 18/100, batch 226/625 completed (loss: 0.0007:  36%|███▌      | 113/312 [11:25<19:54,  6.00s/it]
Training Epoch: 18/100, batch 226/625 completed (loss: 0.0015:  36%|███▌      | 113/312 [11:25<19:55,  6.01s/it]
Training Epoch: 18/100, batch 226/625 completed (loss: 0.0038:  36%|███▌      | 113/312 [11:25<19:55,  6.01s/it]
Training Epoch: 18/100, batch 226/625 completed (loss: 0.0047:  36%|███▌      | 113/312 [11:26<19:55,  6.01s/it]
Training Epoch: 18/100, batch 226/625 completed (loss: 0.0015:  37%|███▋      | 114/312 [11:28<19:49,  6.01s/it]
Training Epoch: 18/100, batch 226/625 completed (loss: 0.0047:  37%|███▋      | 114/312 [11:28<19:49,  6.01s/it]
Training Epoch: 18/100, batch 226/625 completed (loss: 0.0007:  37%|███▋      | 114/312 [11:28<19:49,  6.01s/it]
Training Epoch: 18/100, batch 226/625 completed (loss: 0.0038:  37%|███▋      | 114/312 [11:28<19:50,  6.01s/it]
Training Epoch: 18/100, batch 227/625 completed (loss: 0.0003:  37%|███▋      | 114/312 [11:28<19:49,  6.01s/it]
Training Epoch: 18/100, batch 227/625 completed (loss: 0.0122:  37%|███▋      | 114/312 [11:28<19:50,  6.01s/it]
Training Epoch: 18/100, batch 227/625 completed (loss: 0.0012:  37%|███▋      | 114/312 [11:28<19:49,  6.01s/it]
Training Epoch: 18/100, batch 227/625 completed (loss: 0.0118:  37%|███▋      | 114/312 [11:28<19:49,  6.01s/it]
Training Epoch: 18/100, batch 228/625 completed (loss: 0.0031:  37%|███▋      | 114/312 [11:32<19:49,  6.01s/it]
Training Epoch: 18/100, batch 228/625 completed (loss: 0.0173:  37%|███▋      | 114/312 [11:32<19:49,  6.01s/it]
Training Epoch: 18/100, batch 228/625 completed (loss: 0.0059:  37%|███▋      | 114/312 [11:32<19:50,  6.01s/it]
Training Epoch: 18/100, batch 228/625 completed (loss: 0.001:  37%|███▋      | 114/312 [11:32<19:49,  6.01s/it] 
Training Epoch: 18/100, batch 228/625 completed (loss: 0.0173:  37%|███▋      | 115/312 [11:35<19:59,  6.09s/it]
Training Epoch: 18/100, batch 228/625 completed (loss: 0.0059:  37%|███▋      | 115/312 [11:35<19:59,  6.09s/it]
Training Epoch: 18/100, batch 228/625 completed (loss: 0.0031:  37%|███▋      | 115/312 [11:35<19:59,  6.09s/it]
Training Epoch: 18/100, batch 228/625 completed (loss: 0.001:  37%|███▋      | 115/312 [11:35<20:00,  6.09s/it]
Training Epoch: 18/100, batch 229/625 completed (loss: 0.0018:  37%|███▋      | 115/312 [11:35<19:59,  6.09s/it]
Training Epoch: 18/100, batch 229/625 completed (loss: 0.0023:  37%|███▋      | 115/312 [11:35<19:59,  6.09s/it]
Training Epoch: 18/100, batch 229/625 completed (loss: 0.0041:  37%|███▋      | 115/312 [11:35<19:59,  6.09s/it]
Training Epoch: 18/100, batch 229/625 completed (loss: 0.0034:  37%|███▋      | 115/312 [11:35<20:00,  6.09s/it]
Training Epoch: 18/100, batch 230/625 completed (loss: 0.0095:  37%|███▋      | 115/312 [11:38<19:59,  6.09s/it]
Training Epoch: 18/100, batch 230/625 completed (loss: 0.0364:  37%|███▋      | 115/312 [11:38<19:59,  6.09s/it]
Training Epoch: 18/100, batch 230/625 completed (loss: 0.0021:  37%|███▋      | 115/312 [11:38<19:59,  6.09s/it]
Training Epoch: 18/100, batch 230/625 completed (loss: 0.001:  37%|███▋      | 115/312 [11:38<20:00,  6.09s/it] 
Training Epoch: 18/100, batch 230/625 completed (loss: 0.0021:  37%|███▋      | 116/312 [11:40<19:38,  6.01s/it]
Training Epoch: 18/100, batch 230/625 completed (loss: 0.001:  37%|███▋      | 116/312 [11:41<19:38,  6.01s/it]
Training Epoch: 18/100, batch 230/625 completed (loss: 0.0095:  37%|███▋      | 116/312 [11:40<19:38,  6.01s/it]
Training Epoch: 18/100, batch 230/625 completed (loss: 0.0364:  37%|███▋      | 116/312 [11:40<19:38,  6.01s/it]
Training Epoch: 18/100, batch 231/625 completed (loss: 0.0021:  37%|███▋      | 116/312 [11:41<19:38,  6.01s/it]
Training Epoch: 18/100, batch 231/625 completed (loss: 0.0147:  37%|███▋      | 116/312 [11:41<19:38,  6.01s/it]
Training Epoch: 18/100, batch 231/625 completed (loss: 0.002:  37%|███▋      | 116/312 [11:41<19:38,  6.01s/it] 
Training Epoch: 18/100, batch 231/625 completed (loss: 0.0022:  37%|███▋      | 116/312 [11:41<19:38,  6.01s/it]
Training Epoch: 18/100, batch 232/625 completed (loss: 0.0018:  37%|███▋      | 116/312 [11:44<19:38,  6.01s/it]
Training Epoch: 18/100, batch 232/625 completed (loss: 0.0061:  37%|███▋      | 116/312 [11:44<19:38,  6.01s/it]
Training Epoch: 18/100, batch 232/625 completed (loss: 0.0017:  37%|███▋      | 116/312 [11:44<19:38,  6.01s/it]
Training Epoch: 18/100, batch 232/625 completed (loss: 0.0006:  37%|███▋      | 116/312 [11:44<19:38,  6.01s/it]
Training Epoch: 18/100, batch 232/625 completed (loss: 0.0006:  38%|███▊      | 117/312 [11:47<19:50,  6.10s/it]
Training Epoch: 18/100, batch 232/625 completed (loss: 0.0017:  38%|███▊      | 117/312 [11:47<19:50,  6.10s/it]
Training Epoch: 18/100, batch 232/625 completed (loss: 0.0061:  38%|███▊      | 117/312 [11:47<19:50,  6.10s/it]
Training Epoch: 18/100, batch 232/625 completed (loss: 0.0018:  38%|███▊      | 117/312 [11:47<19:50,  6.11s/it]
Training Epoch: 18/100, batch 233/625 completed (loss: 0.0438:  38%|███▊      | 117/312 [11:47<19:50,  6.10s/it]
Training Epoch: 18/100, batch 233/625 completed (loss: 0.0036:  38%|███▊      | 117/312 [11:47<19:50,  6.11s/it]
Training Epoch: 18/100, batch 233/625 completed (loss: 0.009:  38%|███▊      | 117/312 [11:47<19:50,  6.10s/it] 
Training Epoch: 18/100, batch 233/625 completed (loss: 0.0038:  38%|███▊      | 117/312 [11:47<19:50,  6.10s/it]
Training Epoch: 18/100, batch 234/625 completed (loss: 0.0255:  38%|███▊      | 117/312 [11:50<19:50,  6.10s/it]
Training Epoch: 18/100, batch 234/625 completed (loss: 0.0008:  38%|███▊      | 117/312 [11:50<19:50,  6.11s/it]
Training Epoch: 18/100, batch 234/625 completed (loss: 0.0322:  38%|███▊      | 117/312 [11:50<19:50,  6.10s/it]
Training Epoch: 18/100, batch 234/625 completed (loss: 0.0173:  38%|███▊      | 117/312 [11:50<19:50,  6.10s/it]
Training Epoch: 18/100, batch 234/625 completed (loss: 0.0173:  38%|███▊      | 118/312 [11:53<19:29,  6.03s/it]
Training Epoch: 18/100, batch 234/625 completed (loss: 0.0008:  38%|███▊      | 118/312 [11:53<19:29,  6.03s/it]
Training Epoch: 18/100, batch 234/625 completed (loss: 0.0322:  38%|███▊      | 118/312 [11:53<19:29,  6.03s/it]
Training Epoch: 18/100, batch 234/625 completed (loss: 0.0255:  38%|███▊      | 118/312 [11:53<19:29,  6.03s/it]
Training Epoch: 18/100, batch 235/625 completed (loss: 0.0409:  38%|███▊      | 118/312 [11:53<19:29,  6.03s/it]
Training Epoch: 18/100, batch 235/625 completed (loss: 0.0001:  38%|███▊      | 118/312 [11:53<19:29,  6.03s/it]
Training Epoch: 18/100, batch 235/625 completed (loss: 0.0003:  38%|███▊      | 118/312 [11:53<19:29,  6.03s/it]
Training Epoch: 18/100, batch 235/625 completed (loss: 0.0266:  38%|███▊      | 118/312 [11:53<19:29,  6.03s/it]
Training Epoch: 18/100, batch 236/625 completed (loss: 0.0126:  38%|███▊      | 118/312 [11:56<19:29,  6.03s/it]
Training Epoch: 18/100, batch 236/625 completed (loss: 0.0143:  38%|███▊      | 118/312 [11:56<19:29,  6.03s/it]
Training Epoch: 18/100, batch 236/625 completed (loss: 0.0372:  38%|███▊      | 118/312 [11:56<19:29,  6.03s/it]
Training Epoch: 18/100, batch 236/625 completed (loss: 0.0015:  38%|███▊      | 118/312 [11:56<19:29,  6.03s/it]
Training Epoch: 18/100, batch 236/625 completed (loss: 0.0143:  38%|███▊      | 119/312 [11:59<19:23,  6.03s/it]
Training Epoch: 18/100, batch 236/625 completed (loss: 0.0126:  38%|███▊      | 119/312 [11:59<19:23,  6.03s/it]
Training Epoch: 18/100, batch 236/625 completed (loss: 0.0372:  38%|███▊      | 119/312 [11:59<19:23,  6.03s/it]
Training Epoch: 18/100, batch 236/625 completed (loss: 0.0015:  38%|███▊      | 119/312 [11:59<19:23,  6.03s/it]
Training Epoch: 18/100, batch 237/625 completed (loss: 0.0018:  38%|███▊      | 119/312 [11:59<19:23,  6.03s/it]
Training Epoch: 18/100, batch 237/625 completed (loss: 0.0072:  38%|███▊      | 119/312 [11:59<19:23,  6.03s/it]
Training Epoch: 18/100, batch 237/625 completed (loss: 0.0022:  38%|███▊      | 119/312 [11:59<19:23,  6.03s/it]
Training Epoch: 18/100, batch 237/625 completed (loss: 0.0001:  38%|███▊      | 119/312 [11:59<19:23,  6.03s/it]
Training Epoch: 18/100, batch 238/625 completed (loss: 0.0018:  38%|███▊      | 119/312 [12:02<19:23,  6.03s/it]
Training Epoch: 18/100, batch 238/625 completed (loss: 0.0027:  38%|███▊      | 119/312 [12:02<19:23,  6.03s/it]
Training Epoch: 18/100, batch 238/625 completed (loss: 0.0009:  38%|███▊      | 119/312 [12:02<19:23,  6.03s/it]
Training Epoch: 18/100, batch 238/625 completed (loss: 0.0005:  38%|███▊      | 119/312 [12:02<19:23,  6.03s/it]
Training Epoch: 18/100, batch 238/625 completed (loss: 0.0005:  38%|███▊      | 120/312 [12:05<19:31,  6.10s/it]
Training Epoch: 18/100, batch 238/625 completed (loss: 0.0009:  38%|███▊      | 120/312 [12:05<19:32,  6.10s/it]
Training Epoch: 18/100, batch 238/625 completed (loss: 0.0018:  38%|███▊      | 120/312 [12:05<19:32,  6.10s/it]
Training Epoch: 18/100, batch 238/625 completed (loss: 0.0027:  38%|███▊      | 120/312 [12:05<19:33,  6.11s/it]
Training Epoch: 18/100, batch 239/625 completed (loss: 0.0028:  38%|███▊      | 120/312 [12:05<19:32,  6.10s/it]
Training Epoch: 18/100, batch 239/625 completed (loss: 0.0153:  38%|███▊      | 120/312 [12:05<19:32,  6.10s/it]
Training Epoch: 18/100, batch 239/625 completed (loss: 0.0264:  38%|███▊      | 120/312 [12:05<19:33,  6.11s/it]
Training Epoch: 18/100, batch 239/625 completed (loss: 0.0057:  38%|███▊      | 120/312 [12:05<19:31,  6.10s/it]
Training Epoch: 18/100, batch 240/625 completed (loss: 0.0664:  38%|███▊      | 120/312 [12:08<19:32,  6.10s/it]
Training Epoch: 18/100, batch 240/625 completed (loss: 0.0004:  38%|███▊      | 120/312 [12:08<19:32,  6.10s/it]
Training Epoch: 18/100, batch 240/625 completed (loss: 0.0142:  38%|███▊      | 120/312 [12:08<19:33,  6.11s/it]
Training Epoch: 18/100, batch 240/625 completed (loss: 0.0084:  38%|███▊      | 120/312 [12:08<19:31,  6.10s/it]
Training Epoch: 18/100, batch 240/625 completed (loss: 0.0004:  39%|███▉      | 121/312 [12:11<19:10,  6.03s/it]
Training Epoch: 18/100, batch 240/625 completed (loss: 0.0142:  39%|███▉      | 121/312 [12:11<19:10,  6.03s/it]
Training Epoch: 18/100, batch 240/625 completed (loss: 0.0084:  39%|███▉      | 121/312 [12:11<19:11,  6.03s/it]
Training Epoch: 18/100, batch 240/625 completed (loss: 0.0664:  39%|███▉      | 121/312 [12:11<19:11,  6.03s/it]
Training Epoch: 18/100, batch 241/625 completed (loss: 0.0008:  39%|███▉      | 121/312 [12:11<19:10,  6.03s/it]
Training Epoch: 18/100, batch 241/625 completed (loss: 0.0001:  39%|███▉      | 121/312 [12:11<19:10,  6.03s/it]
Training Epoch: 18/100, batch 241/625 completed (loss: 0.0055:  39%|███▉      | 121/312 [12:11<19:11,  6.03s/it]
Training Epoch: 18/100, batch 241/625 completed (loss: 0.0064:  39%|███▉      | 121/312 [12:11<19:11,  6.03s/it]
Training Epoch: 18/100, batch 242/625 completed (loss: 0.0006:  39%|███▉      | 121/312 [12:14<19:11,  6.03s/it]
Training Epoch: 18/100, batch 242/625 completed (loss: 0.0132:  39%|███▉      | 121/312 [12:14<19:10,  6.03s/it]
Training Epoch: 18/100, batch 242/625 completed (loss: 0.0009:  39%|███▉      | 121/312 [12:14<19:10,  6.03s/it]
Training Epoch: 18/100, batch 242/625 completed (loss: 0.0015:  39%|███▉      | 121/312 [12:14<19:11,  6.03s/it]
Training Epoch: 18/100, batch 242/625 completed (loss: 0.0015:  39%|███▉      | 122/312 [12:17<19:17,  6.09s/it]
Training Epoch: 18/100, batch 242/625 completed (loss: 0.0006:  39%|███▉      | 122/312 [12:17<19:17,  6.09s/it]
Training Epoch: 18/100, batch 242/625 completed (loss: 0.0009:  39%|███▉      | 122/312 [12:17<19:18,  6.10s/it]
Training Epoch: 18/100, batch 242/625 completed (loss: 0.0132:  39%|███▉      | 122/312 [12:17<19:18,  6.10s/it]
Training Epoch: 18/100, batch 243/625 completed (loss: 0.0002:  39%|███▉      | 122/312 [12:17<19:18,  6.10s/it]
Training Epoch: 18/100, batch 243/625 completed (loss: 0.0004:  39%|███▉      | 122/312 [12:17<19:17,  6.09s/it]
Training Epoch: 18/100, batch 243/625 completed (loss: 0.0892:  39%|███▉      | 122/312 [12:17<19:18,  6.10s/it]
Training Epoch: 18/100, batch 243/625 completed (loss: 0.0015:  39%|███▉      | 122/312 [12:17<19:17,  6.09s/it]
Training Epoch: 18/100, batch 244/625 completed (loss: 0.0006:  39%|███▉      | 122/312 [12:20<19:18,  6.10s/it]
Training Epoch: 18/100, batch 244/625 completed (loss: 0.0016:  39%|███▉      | 122/312 [12:20<19:18,  6.10s/it]
Training Epoch: 18/100, batch 244/625 completed (loss: 0.0107:  39%|███▉      | 122/312 [12:20<19:17,  6.09s/it]
Training Epoch: 18/100, batch 244/625 completed (loss: 0.0008:  39%|███▉      | 122/312 [12:20<19:17,  6.09s/it]
Training Epoch: 18/100, batch 244/625 completed (loss: 0.0107:  39%|███▉      | 123/312 [12:23<18:57,  6.02s/it]
Training Epoch: 18/100, batch 244/625 completed (loss: 0.0008:  39%|███▉      | 123/312 [12:23<18:57,  6.02s/it]
Training Epoch: 18/100, batch 244/625 completed (loss: 0.0006:  39%|███▉      | 123/312 [12:23<18:57,  6.02s/it]
Training Epoch: 18/100, batch 244/625 completed (loss: 0.0016:  39%|███▉      | 123/312 [12:23<18:57,  6.02s/it]
Training Epoch: 18/100, batch 245/625 completed (loss: 0.0101:  39%|███▉      | 123/312 [12:23<18:57,  6.02s/it]
Training Epoch: 18/100, batch 245/625 completed (loss: 0.0014:  39%|███▉      | 123/312 [12:23<18:57,  6.02s/it]
Training Epoch: 18/100, batch 245/625 completed (loss: 0.0036:  39%|███▉      | 123/312 [12:23<18:57,  6.02s/it]
Training Epoch: 18/100, batch 245/625 completed (loss: 0.0025:  39%|███▉      | 123/312 [12:23<18:57,  6.02s/it]
Training Epoch: 18/100, batch 246/625 completed (loss: 0.0012:  39%|███▉      | 123/312 [12:26<18:57,  6.02s/it]
Training Epoch: 18/100, batch 246/625 completed (loss: 0.0005:  39%|███▉      | 123/312 [12:26<18:57,  6.02s/it]
Training Epoch: 18/100, batch 246/625 completed (loss: 0.0002:  39%|███▉      | 123/312 [12:26<18:57,  6.02s/it]
Training Epoch: 18/100, batch 246/625 completed (loss: 0.0001:  39%|███▉      | 123/312 [12:26<18:57,  6.02s/it]
Training Epoch: 18/100, batch 246/625 completed (loss: 0.0012:  40%|███▉      | 124/312 [12:29<18:43,  5.97s/it]
Training Epoch: 18/100, batch 246/625 completed (loss: 0.0005:  40%|███▉      | 124/312 [12:29<18:43,  5.98s/it]
Training Epoch: 18/100, batch 246/625 completed (loss: 0.0002:  40%|███▉      | 124/312 [12:29<18:44,  5.98s/it]
Training Epoch: 18/100, batch 246/625 completed (loss: 0.0001:  40%|███▉      | 124/312 [12:29<18:45,  5.99s/it]
Training Epoch: 18/100, batch 247/625 completed (loss: 0.0185:  40%|███▉      | 124/312 [12:29<18:44,  5.98s/it]
Training Epoch: 18/100, batch 247/625 completed (loss: 0.0748:  40%|███▉      | 124/312 [12:29<18:43,  5.98s/it]
Training Epoch: 18/100, batch 247/625 completed (loss: 0.0042:  40%|███▉      | 124/312 [12:29<18:43,  5.97s/it]
Training Epoch: 18/100, batch 247/625 completed (loss: 0.0073:  40%|███▉      | 124/312 [12:29<18:45,  5.99s/it]
Training Epoch: 18/100, batch 248/625 completed (loss: 0.0045:  40%|███▉      | 124/312 [12:32<18:44,  5.98s/it]
Training Epoch: 18/100, batch 248/625 completed (loss: 0.0115:  40%|███▉      | 124/312 [12:32<18:43,  5.98s/it]
Training Epoch: 18/100, batch 248/625 completed (loss: 0.0003:  40%|███▉      | 124/312 [12:32<18:43,  5.97s/it]
Training Epoch: 18/100, batch 248/625 completed (loss: 0.0341:  40%|███▉      | 124/312 [12:32<18:45,  5.99s/it]
Training Epoch: 18/100, batch 248/625 completed (loss: 0.0045:  40%|████      | 125/312 [12:35<19:05,  6.12s/it]
Training Epoch: 18/100, batch 248/625 completed (loss: 0.0341:  40%|████      | 125/312 [12:35<19:05,  6.13s/it]
Training Epoch: 18/100, batch 248/625 completed (loss: 0.0115:  40%|████      | 125/312 [12:35<19:05,  6.13s/it]
Training Epoch: 18/100, batch 248/625 completed (loss: 0.0003:  40%|████      | 125/312 [12:35<19:06,  6.13s/it]
Training Epoch: 18/100, batch 249/625 completed (loss: 0.0181:  40%|████      | 125/312 [12:35<19:05,  6.12s/it]
Training Epoch: 18/100, batch 249/625 completed (loss: 0.0001:  40%|████      | 125/312 [12:35<19:06,  6.13s/it]
Training Epoch: 18/100, batch 249/625 completed (loss: 0.0051:  40%|████      | 125/312 [12:35<19:05,  6.13s/it]
Training Epoch: 18/100, batch 249/625 completed (loss: 0.1794:  40%|████      | 125/312 [12:35<19:05,  6.13s/it]
Training Epoch: 18/100, batch 250/625 completed (loss: 0.0395:  40%|████      | 125/312 [12:38<19:06,  6.13s/it]
Training Epoch: 18/100, batch 250/625 completed (loss: 0.0048:  40%|████      | 125/312 [12:38<19:05,  6.12s/it]
Training Epoch: 18/100, batch 250/625 completed (loss: 0.0002:  40%|████      | 125/312 [12:38<19:05,  6.13s/it]
Training Epoch: 18/100, batch 250/625 completed (loss: 0.0023:  40%|████      | 125/312 [12:38<19:05,  6.13s/it]
Training Epoch: 18/100, batch 250/625 completed (loss: 0.0023:  40%|████      | 126/312 [12:41<18:45,  6.05s/it]
Training Epoch: 18/100, batch 250/625 completed (loss: 0.0002:  40%|████      | 126/312 [12:41<18:46,  6.05s/it]
Training Epoch: 18/100, batch 250/625 completed (loss: 0.0395:  40%|████      | 126/312 [12:41<18:46,  6.06s/it]
Training Epoch: 18/100, batch 250/625 completed (loss: 0.0048:  40%|████      | 126/312 [12:41<18:47,  6.06s/it]
Training Epoch: 18/100, batch 251/625 completed (loss: 0.0024:  40%|████      | 126/312 [12:41<18:46,  6.06s/it]
Training Epoch: 18/100, batch 251/625 completed (loss: 0.0831:  40%|████      | 126/312 [12:41<18:46,  6.05s/it]
Training Epoch: 18/100, batch 251/625 completed (loss: 0.0004:  40%|████      | 126/312 [12:41<18:47,  6.06s/it]
Training Epoch: 18/100, batch 251/625 completed (loss: 0.0008:  40%|████      | 126/312 [12:41<18:45,  6.05s/it]
Training Epoch: 18/100, batch 252/625 completed (loss: 0.0038:  40%|████      | 126/312 [12:44<18:46,  6.06s/it]
Training Epoch: 18/100, batch 252/625 completed (loss: 0.0017:  40%|████      | 126/312 [12:44<18:47,  6.06s/it]
Training Epoch: 18/100, batch 252/625 completed (loss: 0.0002:  40%|████      | 126/312 [12:44<18:46,  6.05s/it]
Training Epoch: 18/100, batch 252/625 completed (loss: 0.0066:  40%|████      | 126/312 [12:44<18:45,  6.05s/it]
Training Epoch: 18/100, batch 252/625 completed (loss: 0.0017:  41%|████      | 127/312 [12:47<18:36,  6.03s/it]
Training Epoch: 18/100, batch 252/625 completed (loss: 0.0038:  41%|████      | 127/312 [12:47<18:37,  6.04s/it]
Training Epoch: 18/100, batch 252/625 completed (loss: 0.0002:  41%|████      | 127/312 [12:47<18:37,  6.04s/it]
Training Epoch: 18/100, batch 252/625 completed (loss: 0.0066:  41%|████      | 127/312 [12:47<18:37,  6.04s/it]
Training Epoch: 18/100, batch 253/625 completed (loss: 0.0003:  41%|████      | 127/312 [12:47<18:36,  6.03s/it]
Training Epoch: 18/100, batch 253/625 completed (loss: 0.0027:  41%|████      | 127/312 [12:47<18:37,  6.04s/it]
Training Epoch: 18/100, batch 253/625 completed (loss: 0.0012:  41%|████      | 127/312 [12:47<18:37,  6.04s/it]
Training Epoch: 18/100, batch 253/625 completed (loss: 0.0025:  41%|████      | 127/312 [12:47<18:37,  6.04s/it]
Training Epoch: 18/100, batch 254/625 completed (loss: 0.0028:  41%|████      | 127/312 [12:51<18:37,  6.04s/it]
Training Epoch: 18/100, batch 254/625 completed (loss: 0.0017:  41%|████      | 127/312 [12:51<18:37,  6.04s/it]
Training Epoch: 18/100, batch 254/625 completed (loss: 0.0118:  41%|████      | 127/312 [12:51<18:36,  6.03s/it]
Training Epoch: 18/100, batch 254/625 completed (loss: 0.0074:  41%|████      | 127/312 [12:51<18:37,  6.04s/it]
Training Epoch: 18/100, batch 254/625 completed (loss: 0.0028:  41%|████      | 128/312 [12:53<18:46,  6.12s/it]
Training Epoch: 18/100, batch 254/625 completed (loss: 0.0118:  41%|████      | 128/312 [12:53<18:46,  6.12s/it]
Training Epoch: 18/100, batch 254/625 completed (loss: 0.0017:  41%|████      | 128/312 [12:53<18:46,  6.12s/it]
Training Epoch: 18/100, batch 254/625 completed (loss: 0.0074:  41%|████      | 128/312 [12:54<18:47,  6.13s/it]
Training Epoch: 18/100, batch 255/625 completed (loss: 0.0599:  41%|████      | 128/312 [12:53<18:46,  6.12s/it]
Training Epoch: 18/100, batch 255/625 completed (loss: 0.0017:  41%|████      | 128/312 [12:54<18:46,  6.12s/it]
Training Epoch: 18/100, batch 255/625 completed (loss: 0.0019:  41%|████      | 128/312 [12:54<18:46,  6.12s/it]
Training Epoch: 18/100, batch 255/625 completed (loss: 0.0007:  41%|████      | 128/312 [12:54<18:47,  6.13s/it]
Training Epoch: 18/100, batch 256/625 completed (loss: 0.0005:  41%|████      | 128/312 [12:56<18:46,  6.12s/it]
Training Epoch: 18/100, batch 256/625 completed (loss: 0.0008:  41%|████      | 128/312 [12:56<18:46,  6.12s/it]
Training Epoch: 18/100, batch 256/625 completed (loss: 0.0006:  41%|████      | 128/312 [12:56<18:46,  6.12s/it]
Training Epoch: 18/100, batch 256/625 completed (loss: 0.0001:  41%|████      | 128/312 [12:56<18:47,  6.13s/it]
Training Epoch: 18/100, batch 256/625 completed (loss: 0.0006:  41%|████▏     | 129/312 [12:59<18:26,  6.05s/it]
Training Epoch: 18/100, batch 256/625 completed (loss: 0.0008:  41%|████▏     | 129/312 [12:59<18:26,  6.04s/it]
Training Epoch: 18/100, batch 256/625 completed (loss: 0.0005:  41%|████▏     | 129/312 [12:59<18:26,  6.05s/it]
Training Epoch: 18/100, batch 256/625 completed (loss: 0.0001:  41%|████▏     | 129/312 [12:59<18:26,  6.05s/it]
Training Epoch: 18/100, batch 257/625 completed (loss: 0.018:  41%|████▏     | 129/312 [12:59<18:26,  6.05s/it] 
Training Epoch: 18/100, batch 257/625 completed (loss: 0.0103:  41%|████▏     | 129/312 [12:59<18:26,  6.05s/it]
Training Epoch: 18/100, batch 257/625 completed (loss: 0.0041:  41%|████▏     | 129/312 [12:59<18:26,  6.04s/it]
Training Epoch: 18/100, batch 257/625 completed (loss: 0.0006:  41%|████▏     | 129/312 [12:59<18:26,  6.05s/it]
Training Epoch: 18/100, batch 258/625 completed (loss: 0.0003:  41%|████▏     | 129/312 [13:02<18:26,  6.05s/it]
Training Epoch: 18/100, batch 258/625 completed (loss: 0.0003:  41%|████▏     | 129/312 [13:02<18:26,  6.05s/it]
Training Epoch: 18/100, batch 258/625 completed (loss: 0.0108:  41%|████▏     | 129/312 [13:02<18:26,  6.04s/it]
Training Epoch: 18/100, batch 258/625 completed (loss: 0.0007:  41%|████▏     | 129/312 [13:02<18:26,  6.05s/it]
Training Epoch: 18/100, batch 258/625 completed (loss: 0.0003:  42%|████▏     | 130/312 [13:06<18:36,  6.13s/it]
Training Epoch: 18/100, batch 258/625 completed (loss: 0.0108:  42%|████▏     | 130/312 [13:06<18:35,  6.13s/it]
Training Epoch: 18/100, batch 258/625 completed (loss: 0.0003:  42%|████▏     | 130/312 [13:06<18:36,  6.13s/it]
Training Epoch: 18/100, batch 258/625 completed (loss: 0.0007:  42%|████▏     | 130/312 [13:06<18:36,  6.13s/it]
Training Epoch: 18/100, batch 259/625 completed (loss: 0.0126:  42%|████▏     | 130/312 [13:06<18:35,  6.13s/it]
Training Epoch: 18/100, batch 259/625 completed (loss: 0.0014:  42%|████▏     | 130/312 [13:06<18:36,  6.13s/it]
Training Epoch: 18/100, batch 259/625 completed (loss: 0.0042:  42%|████▏     | 130/312 [13:06<18:36,  6.13s/it]
Training Epoch: 18/100, batch 259/625 completed (loss: 0.0068:  42%|████▏     | 130/312 [13:06<18:36,  6.13s/it]
Training Epoch: 18/100, batch 260/625 completed (loss: 0.0014:  42%|████▏     | 130/312 [13:09<18:35,  6.13s/it]
Training Epoch: 18/100, batch 260/625 completed (loss: 0.0006:  42%|████▏     | 130/312 [13:09<18:36,  6.13s/it]
Training Epoch: 18/100, batch 260/625 completed (loss: 0.0031:  42%|████▏     | 130/312 [13:09<18:36,  6.13s/it]
Training Epoch: 18/100, batch 260/625 completed (loss: 0.003:  42%|████▏     | 130/312 [13:09<18:36,  6.13s/it] 
Training Epoch: 18/100, batch 260/625 completed (loss: 0.0031:  42%|████▏     | 131/312 [13:11<18:14,  6.05s/it]
Training Epoch: 18/100, batch 260/625 completed (loss: 0.0014:  42%|████▏     | 131/312 [13:11<18:14,  6.05s/it]
Training Epoch: 18/100, batch 260/625 completed (loss: 0.003:  42%|████▏     | 131/312 [13:12<18:14,  6.05s/it]
Training Epoch: 18/100, batch 260/625 completed (loss: 0.0006:  42%|████▏     | 131/312 [13:12<18:15,  6.05s/it]
Training Epoch: 18/100, batch 261/625 completed (loss: 0.0046:  42%|████▏     | 131/312 [13:12<18:14,  6.05s/it]
Training Epoch: 18/100, batch 261/625 completed (loss: 0.0006:  42%|████▏     | 131/312 [13:12<18:14,  6.05s/it]
Training Epoch: 18/100, batch 261/625 completed (loss: 0.0012:  42%|████▏     | 131/312 [13:12<18:15,  6.05s/it]
Training Epoch: 18/100, batch 261/625 completed (loss: 0.0305:  42%|████▏     | 131/312 [13:12<18:14,  6.05s/it]
Training Epoch: 18/100, batch 262/625 completed (loss: 0.0006:  42%|████▏     | 131/312 [13:14<18:15,  6.05s/it]
Training Epoch: 18/100, batch 262/625 completed (loss: 0.0001:  42%|████▏     | 131/312 [13:14<18:14,  6.05s/it]
Training Epoch: 18/100, batch 262/625 completed (loss: 0.0011:  42%|████▏     | 131/312 [13:14<18:14,  6.05s/it]
Training Epoch: 18/100, batch 262/625 completed (loss: 0.0013:  42%|████▏     | 131/312 [13:15<18:14,  6.05s/it]
Training Epoch: 18/100, batch 262/625 completed (loss: 0.0006:  42%|████▏     | 132/312 [13:17<17:56,  5.98s/it]
Training Epoch: 18/100, batch 262/625 completed (loss: 0.0001:  42%|████▏     | 132/312 [13:17<17:56,  5.98s/it]
Training Epoch: 18/100, batch 262/625 completed (loss: 0.0013:  42%|████▏     | 132/312 [13:17<17:56,  5.98s/it]
Training Epoch: 18/100, batch 262/625 completed (loss: 0.0011:  42%|████▏     | 132/312 [13:17<17:56,  5.98s/it]
Training Epoch: 18/100, batch 263/625 completed (loss: 0.0042:  42%|████▏     | 132/312 [13:17<17:56,  5.98s/it]
Training Epoch: 18/100, batch 263/625 completed (loss: 0.0002:  42%|████▏     | 132/312 [13:17<17:56,  5.98s/it]
Training Epoch: 18/100, batch 263/625 completed (loss: 0.0002:  42%|████▏     | 132/312 [13:17<17:56,  5.98s/it]
Training Epoch: 18/100, batch 263/625 completed (loss: 0.001:  42%|████▏     | 132/312 [13:17<17:56,  5.98s/it] 
Training Epoch: 18/100, batch 264/625 completed (loss: 0.0011:  42%|████▏     | 132/312 [13:21<17:56,  5.98s/it]
Training Epoch: 18/100, batch 264/625 completed (loss: 0.001:  42%|████▏     | 132/312 [13:21<17:56,  5.98s/it] 
Training Epoch: 18/100, batch 264/625 completed (loss: 0.0033:  42%|████▏     | 132/312 [13:21<17:56,  5.98s/it]
Training Epoch: 18/100, batch 264/625 completed (loss: 0.0012:  42%|████▏     | 132/312 [13:21<17:56,  5.98s/it]
Training Epoch: 18/100, batch 264/625 completed (loss: 0.0011:  43%|████▎     | 133/312 [13:24<18:15,  6.12s/it]
Training Epoch: 18/100, batch 264/625 completed (loss: 0.0012:  43%|████▎     | 133/312 [13:24<18:15,  6.12s/it]
Training Epoch: 18/100, batch 264/625 completed (loss: 0.0033:  43%|████▎     | 133/312 [13:24<18:16,  6.13s/it]
Training Epoch: 18/100, batch 264/625 completed (loss: 0.001:  43%|████▎     | 133/312 [13:24<18:16,  6.12s/it]
Training Epoch: 18/100, batch 265/625 completed (loss: 0.0017:  43%|████▎     | 133/312 [13:24<18:16,  6.12s/it]
Training Epoch: 18/100, batch 265/625 completed (loss: 0.0169:  43%|████▎     | 133/312 [13:24<18:16,  6.13s/it]
Training Epoch: 18/100, batch 265/625 completed (loss: 0.0343:  43%|████▎     | 133/312 [13:24<18:15,  6.12s/it]
Training Epoch: 18/100, batch 265/625 completed (loss: 0.0015:  43%|████▎     | 133/312 [13:24<18:15,  6.12s/it]
Training Epoch: 18/100, batch 266/625 completed (loss: 0.0004:  43%|████▎     | 133/312 [13:27<18:15,  6.12s/it]
Training Epoch: 18/100, batch 266/625 completed (loss: 0.0067:  43%|████▎     | 133/312 [13:27<18:16,  6.12s/it]
Training Epoch: 18/100, batch 266/625 completed (loss: 0.001:  43%|████▎     | 133/312 [13:27<18:16,  6.13s/it] 
Training Epoch: 18/100, batch 266/625 completed (loss: 0.0056:  43%|████▎     | 133/312 [13:27<18:15,  6.12s/it]
Training Epoch: 18/100, batch 266/625 completed (loss: 0.001:  43%|████▎     | 134/312 [13:30<17:55,  6.04s/it]
Training Epoch: 18/100, batch 266/625 completed (loss: 0.0067:  43%|████▎     | 134/312 [13:30<17:55,  6.04s/it]
Training Epoch: 18/100, batch 266/625 completed (loss: 0.0004:  43%|████▎     | 134/312 [13:30<17:55,  6.04s/it]
Training Epoch: 18/100, batch 266/625 completed (loss: 0.0056:  43%|████▎     | 134/312 [13:30<17:55,  6.04s/it]
Training Epoch: 18/100, batch 267/625 completed (loss: 0.0063:  43%|████▎     | 134/312 [13:30<17:55,  6.04s/it]
Training Epoch: 18/100, batch 267/625 completed (loss: 0.0038:  43%|████▎     | 134/312 [13:30<17:55,  6.04s/it]
Training Epoch: 18/100, batch 267/625 completed (loss: 0.0153:  43%|████▎     | 134/312 [13:30<17:55,  6.04s/it]
Training Epoch: 18/100, batch 267/625 completed (loss: 0.0006:  43%|████▎     | 134/312 [13:30<17:55,  6.04s/it]
Training Epoch: 18/100, batch 268/625 completed (loss: 0.0064:  43%|████▎     | 134/312 [13:33<17:55,  6.04s/it]
Training Epoch: 18/100, batch 268/625 completed (loss: 0.0003:  43%|████▎     | 134/312 [13:33<17:55,  6.04s/it]
Training Epoch: 18/100, batch 268/625 completed (loss: 0.0018:  43%|████▎     | 134/312 [13:33<17:55,  6.04s/it]
Training Epoch: 18/100, batch 268/625 completed (loss: 0.0004:  43%|████▎     | 134/312 [13:33<17:55,  6.04s/it]
Training Epoch: 18/100, batch 268/625 completed (loss: 0.0018:  43%|████▎     | 135/312 [13:36<17:55,  6.08s/it]
Training Epoch: 18/100, batch 268/625 completed (loss: 0.0004:  43%|████▎     | 135/312 [13:36<17:55,  6.08s/it]
Training Epoch: 18/100, batch 268/625 completed (loss: 0.0064:  43%|████▎     | 135/312 [13:36<17:56,  6.08s/it]
Training Epoch: 18/100, batch 268/625 completed (loss: 0.0003:  43%|████▎     | 135/312 [13:36<17:56,  6.08s/it]
Training Epoch: 18/100, batch 269/625 completed (loss: 0.0006:  43%|████▎     | 135/312 [13:36<17:56,  6.08s/it]
Training Epoch: 18/100, batch 269/625 completed (loss: 0.0005:  43%|████▎     | 135/312 [13:36<17:56,  6.08s/it]
Training Epoch: 18/100, batch 269/625 completed (loss: 0.0029:  43%|████▎     | 135/312 [13:36<17:55,  6.08s/it]
Training Epoch: 18/100, batch 269/625 completed (loss: 0.0055:  43%|████▎     | 135/312 [13:36<17:55,  6.08s/it]
Training Epoch: 18/100, batch 270/625 completed (loss: 0.0023:  43%|████▎     | 135/312 [13:39<17:56,  6.08s/it]
Training Epoch: 18/100, batch 270/625 completed (loss: 0.0015:  43%|████▎     | 135/312 [13:39<17:55,  6.08s/it]
Training Epoch: 18/100, batch 270/625 completed (loss: 0.0049:  43%|████▎     | 135/312 [13:39<17:56,  6.08s/it]
Training Epoch: 18/100, batch 270/625 completed (loss: 0.0005:  43%|████▎     | 135/312 [13:39<17:55,  6.08s/it]
Training Epoch: 18/100, batch 270/625 completed (loss: 0.0015:  44%|████▎     | 136/312 [13:42<17:53,  6.10s/it]
Training Epoch: 18/100, batch 270/625 completed (loss: 0.0049:  44%|████▎     | 136/312 [13:42<17:53,  6.10s/it]
Training Epoch: 18/100, batch 270/625 completed (loss: 0.0023:  44%|████▎     | 136/312 [13:42<17:53,  6.10s/it]
Training Epoch: 18/100, batch 270/625 completed (loss: 0.0005:  44%|████▎     | 136/312 [13:42<17:54,  6.10s/it]
Training Epoch: 18/100, batch 271/625 completed (loss: 0.0001:  44%|████▎     | 136/312 [13:42<17:53,  6.10s/it]
Training Epoch: 18/100, batch 271/625 completed (loss: 0.0013:  44%|████▎     | 136/312 [13:42<17:53,  6.10s/it]
Training Epoch: 18/100, batch 271/625 completed (loss: 0.0024:  44%|████▎     | 136/312 [13:42<17:53,  6.10s/it]
Training Epoch: 18/100, batch 271/625 completed (loss: 0.0119:  44%|████▎     | 136/312 [13:42<17:54,  6.10s/it]
Training Epoch: 18/100, batch 272/625 completed (loss: 0.0801:  44%|████▎     | 136/312 [13:45<17:53,  6.10s/it]
Training Epoch: 18/100, batch 272/625 completed (loss: 0.0035:  44%|████▎     | 136/312 [13:45<17:53,  6.10s/it]
Training Epoch: 18/100, batch 272/625 completed (loss: 0.0006:  44%|████▎     | 136/312 [13:45<17:53,  6.10s/it]
Training Epoch: 18/100, batch 272/625 completed (loss: 0.0019:  44%|████▎     | 136/312 [13:45<17:54,  6.10s/it]
Training Epoch: 18/100, batch 272/625 completed (loss: 0.0006:  44%|████▍     | 137/312 [13:48<17:33,  6.02s/it]
Training Epoch: 18/100, batch 272/625 completed (loss: 0.0019:  44%|████▍     | 137/312 [13:48<17:33,  6.02s/it]
Training Epoch: 18/100, batch 272/625 completed (loss: 0.0801:  44%|████▍     | 137/312 [13:48<17:33,  6.02s/it]
Training Epoch: 18/100, batch 272/625 completed (loss: 0.0035:  44%|████▍     | 137/312 [13:48<17:33,  6.02s/it]
Training Epoch: 18/100, batch 273/625 completed (loss: 0.0003:  44%|████▍     | 137/312 [13:48<17:33,  6.02s/it]
Training Epoch: 18/100, batch 273/625 completed (loss: 0.001:  44%|████▍     | 137/312 [13:48<17:33,  6.02s/it] 
Training Epoch: 18/100, batch 273/625 completed (loss: 0.0256:  44%|████▍     | 137/312 [13:48<17:33,  6.02s/it]
Training Epoch: 18/100, batch 273/625 completed (loss: 0.0028:  44%|████▍     | 137/312 [13:48<17:33,  6.02s/it]
Training Epoch: 18/100, batch 274/625 completed (loss: 0.0131:  44%|████▍     | 137/312 [13:51<17:33,  6.02s/it]
Training Epoch: 18/100, batch 274/625 completed (loss: 0.0057:  44%|████▍     | 137/312 [13:51<17:33,  6.02s/it]
Training Epoch: 18/100, batch 274/625 completed (loss: 0.0012:  44%|████▍     | 137/312 [13:51<17:33,  6.02s/it]
Training Epoch: 18/100, batch 274/625 completed (loss: 0.001:  44%|████▍     | 137/312 [13:51<17:33,  6.02s/it] 
Training Epoch: 18/100, batch 274/625 completed (loss: 0.0057:  44%|████▍     | 138/312 [13:54<17:40,  6.09s/it]
Training Epoch: 18/100, batch 274/625 completed (loss: 0.0012:  44%|████▍     | 138/312 [13:54<17:40,  6.09s/it]
Training Epoch: 18/100, batch 274/625 completed (loss: 0.0131:  44%|████▍     | 138/312 [13:54<17:40,  6.10s/it]
Training Epoch: 18/100, batch 274/625 completed (loss: 0.001:  44%|████▍     | 138/312 [13:54<17:40,  6.10s/it]
Training Epoch: 18/100, batch 275/625 completed (loss: 0.0004:  44%|████▍     | 138/312 [13:54<17:40,  6.10s/it]
Training Epoch: 18/100, batch 275/625 completed (loss: 0.001:  44%|████▍     | 138/312 [13:54<17:40,  6.09s/it] 
Training Epoch: 18/100, batch 275/625 completed (loss: 0.0041:  44%|████▍     | 138/312 [13:54<17:40,  6.09s/it]
Training Epoch: 18/100, batch 275/625 completed (loss: 0.0006:  44%|████▍     | 138/312 [13:54<17:40,  6.10s/it]
Training Epoch: 18/100, batch 276/625 completed (loss: 0.0121:  44%|████▍     | 138/312 [13:57<17:40,  6.09s/it]
Training Epoch: 18/100, batch 276/625 completed (loss: 0.0228:  44%|████▍     | 138/312 [13:57<17:40,  6.10s/it]
Training Epoch: 18/100, batch 276/625 completed (loss: 0.0168:  44%|████▍     | 138/312 [13:57<17:40,  6.09s/it]
Training Epoch: 18/100, batch 276/625 completed (loss: 0.0006:  44%|████▍     | 138/312 [13:57<17:40,  6.10s/it]
Training Epoch: 18/100, batch 276/625 completed (loss: 0.0006:  45%|████▍     | 139/312 [14:00<17:30,  6.07s/it]
Training Epoch: 18/100, batch 276/625 completed (loss: 0.0168:  45%|████▍     | 139/312 [14:00<17:32,  6.08s/it]
Training Epoch: 18/100, batch 276/625 completed (loss: 0.0121:  45%|████▍     | 139/312 [14:00<17:32,  6.08s/it]
Training Epoch: 18/100, batch 276/625 completed (loss: 0.0228:  45%|████▍     | 139/312 [14:00<17:32,  6.08s/it]
Training Epoch: 18/100, batch 277/625 completed (loss: 0.0024:  45%|████▍     | 139/312 [14:00<17:32,  6.08s/it]
Training Epoch: 18/100, batch 277/625 completed (loss: 0.0044:  45%|████▍     | 139/312 [14:00<17:32,  6.08s/it]
Training Epoch: 18/100, batch 277/625 completed (loss: 0.004:  45%|████▍     | 139/312 [14:00<17:32,  6.08s/it] 
Training Epoch: 18/100, batch 277/625 completed (loss: 0.0089:  45%|████▍     | 139/312 [14:00<17:30,  6.07s/it]
Training Epoch: 18/100, batch 278/625 completed (loss: 0.008:  45%|████▍     | 139/312 [14:03<17:32,  6.08s/it] 
Training Epoch: 18/100, batch 278/625 completed (loss: 0.0005:  45%|████▍     | 139/312 [14:03<17:32,  6.08s/it]
Training Epoch: 18/100, batch 278/625 completed (loss: 0.001:  45%|████▍     | 139/312 [14:03<17:32,  6.08s/it]
Training Epoch: 18/100, batch 278/625 completed (loss: 0.004:  45%|████▍     | 139/312 [14:03<17:30,  6.07s/it] 
Training Epoch: 18/100, batch 278/625 completed (loss: 0.001:  45%|████▍     | 140/312 [14:06<17:20,  6.05s/it]
Training Epoch: 18/100, batch 278/625 completed (loss: 0.004:  45%|████▍     | 140/312 [14:06<17:20,  6.05s/it]
Training Epoch: 18/100, batch 278/625 completed (loss: 0.0005:  45%|████▍     | 140/312 [14:06<17:20,  6.05s/it]
Training Epoch: 18/100, batch 278/625 completed (loss: 0.008:  45%|████▍     | 140/312 [14:06<17:20,  6.05s/it]
Training Epoch: 18/100, batch 279/625 completed (loss: 0.0001:  45%|████▍     | 140/312 [14:06<17:20,  6.05s/it]
Training Epoch: 18/100, batch 279/625 completed (loss: 0.0008:  45%|████▍     | 140/312 [14:06<17:20,  6.05s/it]
Training Epoch: 18/100, batch 279/625 completed (loss: 0.0051:  45%|████▍     | 140/312 [14:06<17:20,  6.05s/it]
Training Epoch: 18/100, batch 279/625 completed (loss: 0.0009:  45%|████▍     | 140/312 [14:06<17:20,  6.05s/it]
Training Epoch: 18/100, batch 280/625 completed (loss: 0.0093:  45%|████▍     | 140/312 [14:09<17:20,  6.05s/it]
Training Epoch: 18/100, batch 280/625 completed (loss: 0.0016:  45%|████▍     | 140/312 [14:09<17:20,  6.05s/it]
Training Epoch: 18/100, batch 280/625 completed (loss: 0.0101:  45%|████▍     | 140/312 [14:09<17:20,  6.05s/it]
Training Epoch: 18/100, batch 280/625 completed (loss: 0.0004:  45%|████▍     | 140/312 [14:09<17:20,  6.05s/it]
Training Epoch: 18/100, batch 280/625 completed (loss: 0.0093:  45%|████▌     | 141/312 [14:12<17:28,  6.13s/it]
Training Epoch: 18/100, batch 280/625 completed (loss: 0.0101:  45%|████▌     | 141/312 [14:12<17:27,  6.13s/it]
Training Epoch: 18/100, batch 280/625 completed (loss: 0.0004:  45%|████▌     | 141/312 [14:12<17:28,  6.13s/it]
Training Epoch: 18/100, batch 280/625 completed (loss: 0.0016:  45%|████▌     | 141/312 [14:12<17:28,  6.13s/it]
Training Epoch: 18/100, batch 281/625 completed (loss: 0.001:  45%|████▌     | 141/312 [14:12<17:28,  6.13s/it] 
Training Epoch: 18/100, batch 281/625 completed (loss: 0.0005:  45%|████▌     | 141/312 [14:12<17:28,  6.13s/it]
Training Epoch: 18/100, batch 281/625 completed (loss: 0.0123:  45%|████▌     | 141/312 [14:12<17:27,  6.13s/it]
Training Epoch: 18/100, batch 281/625 completed (loss: 0.0002:  45%|████▌     | 141/312 [14:12<17:28,  6.13s/it]
Training Epoch: 18/100, batch 282/625 completed (loss: 0.0051:  45%|████▌     | 141/312 [14:15<17:27,  6.13s/it]
Training Epoch: 18/100, batch 282/625 completed (loss: 0.0015:  45%|████▌     | 141/312 [14:15<17:28,  6.13s/it]
Training Epoch: 18/100, batch 282/625 completed (loss: 0.043:  45%|████▌     | 141/312 [14:15<17:28,  6.13s/it]
Training Epoch: 18/100, batch 282/625 completed (loss: 0.0031:  45%|████▌     | 141/312 [14:15<17:28,  6.13s/it]
Training Epoch: 18/100, batch 282/625 completed (loss: 0.043:  46%|████▌     | 142/312 [14:18<17:07,  6.05s/it]
Training Epoch: 18/100, batch 282/625 completed (loss: 0.0031:  46%|████▌     | 142/312 [14:18<17:08,  6.05s/it]
Training Epoch: 18/100, batch 282/625 completed (loss: 0.0015:  46%|████▌     | 142/312 [14:18<17:08,  6.05s/it]
Training Epoch: 18/100, batch 282/625 completed (loss: 0.0051:  46%|████▌     | 142/312 [14:18<17:07,  6.05s/it]
Training Epoch: 18/100, batch 283/625 completed (loss: 0.0004:  46%|████▌     | 142/312 [14:18<17:07,  6.05s/it]
Training Epoch: 18/100, batch 283/625 completed (loss: 0.0106:  46%|████▌     | 142/312 [14:18<17:08,  6.05s/it]
Training Epoch: 18/100, batch 283/625 completed (loss: 0.0017:  46%|████▌     | 142/312 [14:18<17:07,  6.05s/it]
Training Epoch: 18/100, batch 283/625 completed (loss: 0.001:  46%|████▌     | 142/312 [14:18<17:08,  6.05s/it] 
Training Epoch: 18/100, batch 284/625 completed (loss: 0.0332:  46%|████▌     | 142/312 [14:21<17:08,  6.05s/it]
Training Epoch: 18/100, batch 284/625 completed (loss: 0.011:  46%|████▌     | 142/312 [14:21<17:07,  6.05s/it] 
Training Epoch: 18/100, batch 284/625 completed (loss: 0.0031:  46%|████▌     | 142/312 [14:21<17:07,  6.05s/it]
Training Epoch: 18/100, batch 284/625 completed (loss: 0.0001:  46%|████▌     | 142/312 [14:21<17:08,  6.05s/it]
Training Epoch: 18/100, batch 284/625 completed (loss: 0.0332:  46%|████▌     | 143/312 [14:25<17:14,  6.12s/it]
Training Epoch: 18/100, batch 284/625 completed (loss: 0.0031:  46%|████▌     | 143/312 [14:25<17:13,  6.12s/it]
Training Epoch: 18/100, batch 284/625 completed (loss: 0.011:  46%|████▌     | 143/312 [14:25<17:14,  6.12s/it]
Training Epoch: 18/100, batch 284/625 completed (loss: 0.0001:  46%|████▌     | 143/312 [14:25<17:15,  6.12s/it]
Training Epoch: 18/100, batch 285/625 completed (loss: 0.0046:  46%|████▌     | 143/312 [14:25<17:13,  6.12s/it]
Training Epoch: 18/100, batch 285/625 completed (loss: 0.0004:  46%|████▌     | 143/312 [14:25<17:14,  6.12s/it]
Training Epoch: 18/100, batch 285/625 completed (loss: 0.0014:  46%|████▌     | 143/312 [14:25<17:14,  6.12s/it]
Training Epoch: 18/100, batch 285/625 completed (loss: 0.0034:  46%|████▌     | 143/312 [14:25<17:15,  6.12s/it]
Training Epoch: 18/100, batch 286/625 completed (loss: 0.0005:  46%|████▌     | 143/312 [14:28<17:13,  6.12s/it]
Training Epoch: 18/100, batch 286/625 completed (loss: 0.0121:  46%|████▌     | 143/312 [14:28<17:14,  6.12s/it]
Training Epoch: 18/100, batch 286/625 completed (loss: 0.0081:  46%|████▌     | 143/312 [14:28<17:14,  6.12s/it]
Training Epoch: 18/100, batch 286/625 completed (loss: 0.0029:  46%|████▌     | 143/312 [14:28<17:15,  6.12s/it]
Training Epoch: 18/100, batch 286/625 completed (loss: 0.0081:  46%|████▌     | 144/312 [14:31<17:00,  6.07s/it]
Training Epoch: 18/100, batch 286/625 completed (loss: 0.0029:  46%|████▌     | 144/312 [14:31<17:00,  6.07s/it]
Training Epoch: 18/100, batch 286/625 completed (loss: 0.0005:  46%|████▌     | 144/312 [14:31<17:00,  6.08s/it]
Training Epoch: 18/100, batch 286/625 completed (loss: 0.0121:  46%|████▌     | 144/312 [14:31<17:00,  6.08s/it]
Training Epoch: 18/100, batch 287/625 completed (loss: 0.0133:  46%|████▌     | 144/312 [14:31<17:00,  6.08s/it]
Training Epoch: 18/100, batch 287/625 completed (loss: 0.0005:  46%|████▌     | 144/312 [14:31<17:00,  6.08s/it]
Training Epoch: 18/100, batch 287/625 completed (loss: 0.0037:  46%|████▌     | 144/312 [14:31<17:00,  6.07s/it]
Training Epoch: 18/100, batch 287/625 completed (loss: 0.0005:  46%|████▌     | 144/312 [14:31<17:00,  6.07s/it]
Training Epoch: 18/100, batch 288/625 completed (loss: 0.0001:  46%|████▌     | 144/312 [14:33<17:00,  6.08s/it]
Training Epoch: 18/100, batch 288/625 completed (loss: 0.0017:  46%|████▌     | 144/312 [14:33<17:00,  6.08s/it]
Training Epoch: 18/100, batch 288/625 completed (loss: 0.0037:  46%|████▌     | 144/312 [14:33<17:00,  6.07s/it]
Training Epoch: 18/100, batch 288/625 completed (loss: 0.0184:  46%|████▌     | 144/312 [14:34<17:00,  6.07s/it]
Training Epoch: 18/100, batch 288/625 completed (loss: 0.0037:  46%|████▋     | 145/312 [14:36<16:49,  6.05s/it]
Training Epoch: 18/100, batch 288/625 completed (loss: 0.0184:  46%|████▋     | 145/312 [14:37<16:49,  6.05s/it]
Training Epoch: 18/100, batch 288/625 completed (loss: 0.0017:  46%|████▋     | 145/312 [14:36<16:49,  6.05s/it]
Training Epoch: 18/100, batch 288/625 completed (loss: 0.0001:  46%|████▋     | 145/312 [14:36<16:50,  6.05s/it]
Training Epoch: 18/100, batch 289/625 completed (loss: 0.0092:  46%|████▋     | 145/312 [14:37<16:49,  6.05s/it]
Training Epoch: 18/100, batch 289/625 completed (loss: 0.0071:  46%|████▋     | 145/312 [14:37<16:50,  6.05s/it]
Training Epoch: 18/100, batch 289/625 completed (loss: 0.0049:  46%|████▋     | 145/312 [14:37<16:49,  6.05s/it]
Training Epoch: 18/100, batch 289/625 completed (loss: 0.0014:  46%|████▋     | 145/312 [14:37<16:49,  6.05s/it]
Training Epoch: 18/100, batch 290/625 completed (loss: 0.045:  46%|████▋     | 145/312 [14:40<16:49,  6.05s/it] 
Training Epoch: 18/100, batch 290/625 completed (loss: 0.0014:  46%|████▋     | 145/312 [14:40<16:50,  6.05s/it]
Training Epoch: 18/100, batch 290/625 completed (loss: 0.0005:  46%|████▋     | 145/312 [14:40<16:49,  6.05s/it]
Training Epoch: 18/100, batch 290/625 completed (loss: 0.0042:  46%|████▋     | 145/312 [14:40<16:49,  6.05s/it]
Training Epoch: 18/100, batch 290/625 completed (loss: 0.0042:  47%|████▋     | 146/312 [14:43<16:49,  6.08s/it]
Training Epoch: 18/100, batch 290/625 completed (loss: 0.045:  47%|████▋     | 146/312 [14:43<16:49,  6.08s/it]
Training Epoch: 18/100, batch 290/625 completed (loss: 0.0005:  47%|████▋     | 146/312 [14:43<16:49,  6.08s/it]
Training Epoch: 18/100, batch 290/625 completed (loss: 0.0014:  47%|████▋     | 146/312 [14:43<16:49,  6.08s/it]
Training Epoch: 18/100, batch 291/625 completed (loss: 0.0011:  47%|████▋     | 146/312 [14:43<16:49,  6.08s/it]
Training Epoch: 18/100, batch 291/625 completed (loss: 0.0006:  47%|████▋     | 146/312 [14:43<16:49,  6.08s/it]
Training Epoch: 18/100, batch 291/625 completed (loss: 0.0:  47%|████▋     | 146/312 [14:43<16:49,  6.08s/it]   
Training Epoch: 18/100, batch 291/625 completed (loss: 0.0297:  47%|████▋     | 146/312 [14:43<16:49,  6.08s/it]
Training Epoch: 18/100, batch 292/625 completed (loss: 0.0006:  47%|████▋     | 146/312 [14:46<16:49,  6.08s/it]
Training Epoch: 18/100, batch 292/625 completed (loss: 0.0124:  47%|████▋     | 146/312 [14:46<16:49,  6.08s/it]
Training Epoch: 18/100, batch 292/625 completed (loss: 0.0007:  47%|████▋     | 146/312 [14:46<16:49,  6.08s/it]
Training Epoch: 18/100, batch 292/625 completed (loss: 0.0452:  47%|████▋     | 146/312 [14:46<16:49,  6.08s/it]
Training Epoch: 18/100, batch 292/625 completed (loss: 0.0452:  47%|████▋     | 147/312 [14:49<16:37,  6.04s/it]
Training Epoch: 18/100, batch 292/625 completed (loss: 0.0006:  47%|████▋     | 147/312 [14:49<16:37,  6.05s/it]
Training Epoch: 18/100, batch 292/625 completed (loss: 0.0007:  47%|████▋     | 147/312 [14:49<16:37,  6.05s/it]
Training Epoch: 18/100, batch 292/625 completed (loss: 0.0124:  47%|████▋     | 147/312 [14:49<16:37,  6.05s/it]
Training Epoch: 18/100, batch 293/625 completed (loss: 0.0007:  47%|████▋     | 147/312 [14:49<16:37,  6.05s/it]
Training Epoch: 18/100, batch 293/625 completed (loss: 0.0005:  47%|████▋     | 147/312 [14:49<16:37,  6.05s/it]
Training Epoch: 18/100, batch 293/625 completed (loss: 0.0023:  47%|████▋     | 147/312 [14:49<16:37,  6.05s/it]
Training Epoch: 18/100, batch 293/625 completed (loss: 0.0061:  47%|████▋     | 147/312 [14:49<16:37,  6.04s/it]
Training Epoch: 18/100, batch 294/625 completed (loss: 0.3069:  47%|████▋     | 147/312 [14:52<16:37,  6.05s/it]
Training Epoch: 18/100, batch 294/625 completed (loss: 0.0031:  47%|████▋     | 147/312 [14:52<16:37,  6.05s/it]
Training Epoch: 18/100, batch 294/625 completed (loss: 0.0015:  47%|████▋     | 147/312 [14:52<16:37,  6.05s/it]
Training Epoch: 18/100, batch 294/625 completed (loss: 0.0016:  47%|████▋     | 147/312 [14:52<16:37,  6.04s/it]
Training Epoch: 18/100, batch 294/625 completed (loss: 0.0015:  47%|████▋     | 148/312 [14:55<16:36,  6.07s/it]
Training Epoch: 18/100, batch 294/625 completed (loss: 0.0016:  47%|████▋     | 148/312 [14:55<16:36,  6.07s/it]
Training Epoch: 18/100, batch 294/625 completed (loss: 0.3069:  47%|████▋     | 148/312 [14:55<16:35,  6.07s/it]
Training Epoch: 18/100, batch 294/625 completed (loss: 0.0031:  47%|████▋     | 148/312 [14:55<16:36,  6.08s/it]
Training Epoch: 18/100, batch 295/625 completed (loss: 0.0044:  47%|████▋     | 148/312 [14:55<16:36,  6.07s/it]
Training Epoch: 18/100, batch 295/625 completed (loss: 0.0016:  47%|████▋     | 148/312 [14:55<16:36,  6.08s/it]
Training Epoch: 18/100, batch 295/625 completed (loss: 0.0018:  47%|████▋     | 148/312 [14:55<16:35,  6.07s/it]
Training Epoch: 18/100, batch 295/625 completed (loss: 0.0011:  47%|████▋     | 148/312 [14:55<16:36,  6.07s/it]
Training Epoch: 18/100, batch 296/625 completed (loss: 0.0069:  47%|████▋     | 148/312 [14:58<16:36,  6.08s/it]
Training Epoch: 18/100, batch 296/625 completed (loss: 0.0096:  47%|████▋     | 148/312 [14:58<16:35,  6.07s/it]
Training Epoch: 18/100, batch 296/625 completed (loss: 0.0028:  47%|████▋     | 148/312 [14:58<16:36,  6.07s/it]
Training Epoch: 18/100, batch 296/625 completed (loss: 0.0023:  47%|████▋     | 148/312 [14:58<16:36,  6.07s/it]
Training Epoch: 18/100, batch 296/625 completed (loss: 0.0096:  48%|████▊     | 149/312 [15:01<16:33,  6.09s/it]
Training Epoch: 18/100, batch 296/625 completed (loss: 0.0028:  48%|████▊     | 149/312 [15:01<16:33,  6.10s/it]
Training Epoch: 18/100, batch 296/625 completed (loss: 0.0069:  48%|████▊     | 149/312 [15:01<16:33,  6.09s/it]
Training Epoch: 18/100, batch 296/625 completed (loss: 0.0023:  48%|████▊     | 149/312 [15:01<16:33,  6.10s/it]
Training Epoch: 18/100, batch 297/625 completed (loss: 0.0252:  48%|████▊     | 149/312 [15:01<16:33,  6.09s/it]
Training Epoch: 18/100, batch 297/625 completed (loss: 0.0001:  48%|████▊     | 149/312 [15:01<16:33,  6.09s/it]
Training Epoch: 18/100, batch 297/625 completed (loss: 0.0071:  48%|████▊     | 149/312 [15:01<16:33,  6.10s/it]
Training Epoch: 18/100, batch 297/625 completed (loss: 0.0871:  48%|████▊     | 149/312 [15:01<16:33,  6.10s/it]
Training Epoch: 18/100, batch 298/625 completed (loss: 0.003:  48%|████▊     | 149/312 [15:04<16:33,  6.09s/it] 
Training Epoch: 18/100, batch 298/625 completed (loss: 0.0012:  48%|████▊     | 149/312 [15:04<16:33,  6.10s/it]
Training Epoch: 18/100, batch 298/625 completed (loss: 0.0003:  48%|████▊     | 149/312 [15:04<16:33,  6.09s/it]
Training Epoch: 18/100, batch 298/625 completed (loss: 0.0023:  48%|████▊     | 149/312 [15:04<16:33,  6.10s/it]
Training Epoch: 18/100, batch 298/625 completed (loss: 0.0003:  48%|████▊     | 150/312 [15:07<16:14,  6.01s/it]
Training Epoch: 18/100, batch 298/625 completed (loss: 0.0023:  48%|████▊     | 150/312 [15:07<16:14,  6.02s/it]
Training Epoch: 18/100, batch 298/625 completed (loss: 0.0012:  48%|████▊     | 150/312 [15:07<16:14,  6.02s/it]
Training Epoch: 18/100, batch 298/625 completed (loss: 0.003:  48%|████▊     | 150/312 [15:07<16:14,  6.02s/it]
Training Epoch: 18/100, batch 299/625 completed (loss: 0.0004:  48%|████▊     | 150/312 [15:07<16:14,  6.02s/it]
Training Epoch: 18/100, batch 299/625 completed (loss: 0.0012:  48%|████▊     | 150/312 [15:07<16:14,  6.01s/it]
Training Epoch: 18/100, batch 299/625 completed (loss: 0.013:  48%|████▊     | 150/312 [15:07<16:14,  6.02s/it]
Training Epoch: 18/100, batch 299/625 completed (loss: 0.0265:  48%|████▊     | 150/312 [15:07<16:14,  6.02s/it]
Training Epoch: 18/100, batch 300/625 completed (loss: 0.0093:  48%|████▊     | 150/312 [15:10<16:14,  6.02s/it]
Training Epoch: 18/100, batch 300/625 completed (loss: 0.0189:  48%|████▊     | 150/312 [15:10<16:14,  6.02s/it]
Training Epoch: 18/100, batch 300/625 completed (loss: 0.0216:  48%|████▊     | 150/312 [15:10<16:14,  6.01s/it]
Training Epoch: 18/100, batch 300/625 completed (loss: 0.0:  48%|████▊     | 150/312 [15:10<16:14,  6.02s/it]   
Training Epoch: 18/100, batch 300/625 completed (loss: 0.0:  48%|████▊     | 151/312 [15:13<16:21,  6.10s/it]
Training Epoch: 18/100, batch 300/625 completed (loss: 0.0093:  48%|████▊     | 151/312 [15:13<16:22,  6.10s/it]
Training Epoch: 18/100, batch 300/625 completed (loss: 0.0216:  48%|████▊     | 151/312 [15:13<16:21,  6.10s/it]
Training Epoch: 18/100, batch 300/625 completed (loss: 0.0189:  48%|████▊     | 151/312 [15:13<16:22,  6.10s/it]
Training Epoch: 18/100, batch 301/625 completed (loss: 0.0002:  48%|████▊     | 151/312 [15:13<16:22,  6.10s/it]
Training Epoch: 18/100, batch 301/625 completed (loss: 0.0007:  48%|████▊     | 151/312 [15:13<16:21,  6.10s/it]
Training Epoch: 18/100, batch 301/625 completed (loss: 0.0141:  48%|████▊     | 151/312 [15:13<16:22,  6.10s/it]
Training Epoch: 18/100, batch 301/625 completed (loss: 0.0021:  48%|████▊     | 151/312 [15:13<16:21,  6.10s/it]
Training Epoch: 18/100, batch 302/625 completed (loss: 0.0078:  48%|████▊     | 151/312 [15:16<16:22,  6.10s/it]
Training Epoch: 18/100, batch 302/625 completed (loss: 0.003:  48%|████▊     | 151/312 [15:16<16:22,  6.10s/it] 
Training Epoch: 18/100, batch 302/625 completed (loss: 0.0042:  48%|████▊     | 151/312 [15:16<16:21,  6.10s/it]
Training Epoch: 18/100, batch 302/625 completed (loss: 0.0002:  48%|████▊     | 151/312 [15:16<16:21,  6.10s/it]
Training Epoch: 18/100, batch 302/625 completed (loss: 0.0078:  49%|████▊     | 152/312 [15:19<16:11,  6.07s/it]
Training Epoch: 18/100, batch 302/625 completed (loss: 0.0042:  49%|████▊     | 152/312 [15:19<16:11,  6.07s/it]
Training Epoch: 18/100, batch 302/625 completed (loss: 0.003:  49%|████▊     | 152/312 [15:19<16:11,  6.07s/it]
Training Epoch: 18/100, batch 302/625 completed (loss: 0.0002:  49%|████▊     | 152/312 [15:19<16:12,  6.08s/it]
Training Epoch: 18/100, batch 303/625 completed (loss: 0.0005:  49%|████▊     | 152/312 [15:19<16:11,  6.07s/it]
Training Epoch: 18/100, batch 303/625 completed (loss: 0.0102:  49%|████▊     | 152/312 [15:19<16:11,  6.07s/it]
Training Epoch: 18/100, batch 303/625 completed (loss: 0.0185:  49%|████▊     | 152/312 [15:19<16:11,  6.07s/it]
Training Epoch: 18/100, batch 303/625 completed (loss: 0.0377:  49%|████▊     | 152/312 [15:19<16:12,  6.08s/it]
Training Epoch: 18/100, batch 304/625 completed (loss: 0.0004:  49%|████▊     | 152/312 [15:22<16:11,  6.07s/it]
Training Epoch: 18/100, batch 304/625 completed (loss: 0.0003:  49%|████▊     | 152/312 [15:22<16:11,  6.07s/it]
Training Epoch: 18/100, batch 304/625 completed (loss: 0.0008:  49%|████▊     | 152/312 [15:22<16:11,  6.07s/it]
Training Epoch: 18/100, batch 304/625 completed (loss: 0.0105:  49%|████▊     | 152/312 [15:22<16:12,  6.08s/it]
Training Epoch: 18/100, batch 304/625 completed (loss: 0.0004:  49%|████▉     | 153/312 [15:25<16:08,  6.09s/it]
Training Epoch: 18/100, batch 304/625 completed (loss: 0.0105:  49%|████▉     | 153/312 [15:25<16:08,  6.09s/it]
Training Epoch: 18/100, batch 304/625 completed (loss: 0.0003:  49%|████▉     | 153/312 [15:25<16:08,  6.09s/it]
Training Epoch: 18/100, batch 304/625 completed (loss: 0.0008:  49%|████▉     | 153/312 [15:25<16:09,  6.10s/it]
Training Epoch: 18/100, batch 305/625 completed (loss: 0.001:  49%|████▉     | 153/312 [15:25<16:08,  6.09s/it] 
Training Epoch: 18/100, batch 305/625 completed (loss: 0.0029:  49%|████▉     | 153/312 [15:25<16:08,  6.09s/it]
Training Epoch: 18/100, batch 305/625 completed (loss: 0.0009:  49%|████▉     | 153/312 [15:25<16:09,  6.10s/it]
Training Epoch: 18/100, batch 305/625 completed (loss: 0.0018:  49%|████▉     | 153/312 [15:25<16:08,  6.09s/it]
Training Epoch: 18/100, batch 306/625 completed (loss: 0.0022:  49%|████▉     | 153/312 [15:28<16:08,  6.09s/it]
Training Epoch: 18/100, batch 306/625 completed (loss: 0.0979:  49%|████▉     | 153/312 [15:28<16:08,  6.09s/it]
Training Epoch: 18/100, batch 306/625 completed (loss: 0.0049:  49%|████▉     | 153/312 [15:28<16:09,  6.10s/it]
Training Epoch: 18/100, batch 306/625 completed (loss: 0.0126:  49%|████▉     | 153/312 [15:28<16:08,  6.09s/it]
Training Epoch: 18/100, batch 306/625 completed (loss: 0.0979:  49%|████▉     | 154/312 [15:31<15:59,  6.08s/it]
Training Epoch: 18/100, batch 306/625 completed (loss: 0.0126:  49%|████▉     | 154/312 [15:31<16:00,  6.08s/it]
Training Epoch: 18/100, batch 306/625 completed (loss: 0.0022:  49%|████▉     | 154/312 [15:31<16:00,  6.08s/it]
Training Epoch: 18/100, batch 306/625 completed (loss: 0.0049:  49%|████▉     | 154/312 [15:31<16:01,  6.08s/it]
Training Epoch: 18/100, batch 307/625 completed (loss: 0.0003:  49%|████▉     | 154/312 [15:31<16:00,  6.08s/it]
Training Epoch: 18/100, batch 307/625 completed (loss: 0.0016:  49%|████▉     | 154/312 [15:31<15:59,  6.08s/it]
Training Epoch: 18/100, batch 307/625 completed (loss: 0.1286:  49%|████▉     | 154/312 [15:31<16:01,  6.08s/it]
Training Epoch: 18/100, batch 307/625 completed (loss: 0.0391:  49%|████▉     | 154/312 [15:31<16:00,  6.08s/it]
Training Epoch: 18/100, batch 308/625 completed (loss: 0.0014:  49%|████▉     | 154/312 [15:34<16:00,  6.08s/it]
Training Epoch: 18/100, batch 308/625 completed (loss: 0.0052:  49%|████▉     | 154/312 [15:34<16:01,  6.08s/it]
Training Epoch: 18/100, batch 308/625 completed (loss: 0.0057:  49%|████▉     | 154/312 [15:34<15:59,  6.08s/it]
Training Epoch: 18/100, batch 308/625 completed (loss: 0.0063:  49%|████▉     | 154/312 [15:34<16:00,  6.08s/it]
Training Epoch: 18/100, batch 308/625 completed (loss: 0.0063:  50%|████▉     | 155/312 [15:37<15:52,  6.07s/it]
Training Epoch: 18/100, batch 308/625 completed (loss: 0.0057:  50%|████▉     | 155/312 [15:37<15:52,  6.07s/it]
Training Epoch: 18/100, batch 308/625 completed (loss: 0.0014:  50%|████▉     | 155/312 [15:37<15:53,  6.07s/it]
Training Epoch: 18/100, batch 308/625 completed (loss: 0.0052:  50%|████▉     | 155/312 [15:37<15:53,  6.07s/it]
Training Epoch: 18/100, batch 309/625 completed (loss: 0.0006:  50%|████▉     | 155/312 [15:37<15:53,  6.07s/it]
Training Epoch: 18/100, batch 309/625 completed (loss: 0.0014:  50%|████▉     | 155/312 [15:37<15:52,  6.07s/it]
Training Epoch: 18/100, batch 309/625 completed (loss: 0.0021:  50%|████▉     | 155/312 [15:37<15:53,  6.07s/it]
Training Epoch: 18/100, batch 309/625 completed (loss: 0.0111:  50%|████▉     | 155/312 [15:37<15:52,  6.07s/it]
Training Epoch: 18/100, batch 310/625 completed (loss: 0.0059:  50%|████▉     | 155/312 [15:40<15:52,  6.07s/it]
Training Epoch: 18/100, batch 310/625 completed (loss: 0.0071:  50%|████▉     | 155/312 [15:40<15:53,  6.07s/it]
Training Epoch: 18/100, batch 310/625 completed (loss: 0.0452:  50%|████▉     | 155/312 [15:40<15:53,  6.07s/it]
Training Epoch: 18/100, batch 310/625 completed (loss: 0.0084:  50%|████▉     | 155/312 [15:40<15:52,  6.07s/it]
Training Epoch: 18/100, batch 310/625 completed (loss: 0.0071:  50%|█████     | 156/312 [15:43<15:51,  6.10s/it]
Training Epoch: 18/100, batch 310/625 completed (loss: 0.0084:  50%|█████     | 156/312 [15:43<15:51,  6.10s/it]
Training Epoch: 18/100, batch 310/625 completed (loss: 0.0059:  50%|█████     | 156/312 [15:43<15:51,  6.10s/it]
Training Epoch: 18/100, batch 310/625 completed (loss: 0.0452:  50%|█████     | 156/312 [15:43<15:51,  6.10s/it]
Training Epoch: 18/100, batch 311/625 completed (loss: 0.0076:  50%|█████     | 156/312 [15:43<15:51,  6.10s/it]
Training Epoch: 18/100, batch 311/625 completed (loss: 0.0009:  50%|█████     | 156/312 [15:43<15:51,  6.10s/it]
Training Epoch: 18/100, batch 311/625 completed (loss: 0.0001:  50%|█████     | 156/312 [15:43<15:51,  6.10s/it]
Training Epoch: 18/100, batch 311/625 completed (loss: 0.0306:  50%|█████     | 156/312 [15:44<15:51,  6.10s/it]
Training Epoch: 18/100, batch 312/625 completed (loss: 0.0112:  50%|█████     | 156/312 [15:47<15:51,  6.10s/it]
Training Epoch: 18/100, batch 312/625 completed (loss: 0.0034:  50%|█████     | 156/312 [15:47<15:51,  6.10s/it]
Training Epoch: 18/100, batch 312/625 completed (loss: 0.0035:  50%|█████     | 156/312 [15:47<15:51,  6.10s/it]
Training Epoch: 18/100, batch 312/625 completed (loss: 0.0004:  50%|█████     | 156/312 [15:47<15:51,  6.10s/it]
Training Epoch: 18/100, batch 312/625 completed (loss: 0.0004:  50%|█████     | 157/312 [15:50<15:48,  6.12s/it]
Training Epoch: 18/100, batch 312/625 completed (loss: 0.0035:  50%|█████     | 157/312 [15:50<15:48,  6.12s/it]
Training Epoch: 18/100, batch 312/625 completed (loss: 0.0112:  50%|█████     | 157/312 [15:50<15:48,  6.12s/it]
Training Epoch: 18/100, batch 312/625 completed (loss: 0.0034:  50%|█████     | 157/312 [15:50<15:47,  6.11s/it]
Training Epoch: 18/100, batch 313/625 completed (loss: 0.0:  50%|█████     | 157/312 [15:50<15:47,  6.11s/it]   
Training Epoch: 18/100, batch 313/625 completed (loss: 0.0009:  50%|█████     | 157/312 [15:50<15:48,  6.12s/it]
Training Epoch: 18/100, batch 313/625 completed (loss: 0.0089:  50%|█████     | 157/312 [15:50<15:48,  6.12s/it]
Training Epoch: 18/100, batch 313/625 completed (loss: 0.0114:  50%|█████     | 157/312 [15:50<15:48,  6.12s/it]
Training Epoch: 18/100, batch 314/625 completed (loss: 0.0063:  50%|█████     | 157/312 [15:53<15:48,  6.12s/it]
Training Epoch: 18/100, batch 314/625 completed (loss: 0.0171:  50%|█████     | 157/312 [15:53<15:47,  6.11s/it]
Training Epoch: 18/100, batch 314/625 completed (loss: 0.0063:  50%|█████     | 157/312 [15:53<15:48,  6.12s/it]
Training Epoch: 18/100, batch 314/625 completed (loss: 0.0305:  50%|█████     | 157/312 [15:53<15:48,  6.12s/it]
Training Epoch: 18/100, batch 314/625 completed (loss: 0.0063:  51%|█████     | 158/312 [15:56<15:37,  6.08s/it]
Training Epoch: 18/100, batch 314/625 completed (loss: 0.0063:  51%|█████     | 158/312 [15:56<15:37,  6.09s/it]
Training Epoch: 18/100, batch 314/625 completed (loss: 0.0305:  51%|█████     | 158/312 [15:56<15:37,  6.09s/it]
Training Epoch: 18/100, batch 314/625 completed (loss: 0.0171:  51%|█████     | 158/312 [15:56<15:37,  6.09s/it]
Training Epoch: 18/100, batch 315/625 completed (loss: 0.002:  51%|█████     | 158/312 [15:56<15:37,  6.09s/it] 
Training Epoch: 18/100, batch 315/625 completed (loss: 0.0052:  51%|█████     | 158/312 [15:56<15:37,  6.08s/it]
Training Epoch: 18/100, batch 315/625 completed (loss: 0.0086:  51%|█████     | 158/312 [15:56<15:37,  6.09s/it]
Training Epoch: 18/100, batch 315/625 completed (loss: 0.0071:  51%|█████     | 158/312 [15:56<15:37,  6.09s/it]
Training Epoch: 18/100, batch 316/625 completed (loss: 0.0043:  51%|█████     | 158/312 [15:59<15:37,  6.08s/it]
Training Epoch: 18/100, batch 316/625 completed (loss: 0.0019:  51%|█████     | 158/312 [15:59<15:37,  6.09s/it]
Training Epoch: 18/100, batch 316/625 completed (loss: 0.003:  51%|█████     | 158/312 [15:59<15:37,  6.09s/it]
Training Epoch: 18/100, batch 316/625 completed (loss: 0.0001:  51%|█████     | 158/312 [15:59<15:37,  6.09s/it]
Training Epoch: 18/100, batch 316/625 completed (loss: 0.003:  51%|█████     | 159/312 [16:02<15:33,  6.10s/it]
Training Epoch: 18/100, batch 316/625 completed (loss: 0.0043:  51%|█████     | 159/312 [16:02<15:33,  6.10s/it]
Training Epoch: 18/100, batch 316/625 completed (loss: 0.0001:  51%|█████     | 159/312 [16:02<15:33,  6.10s/it]
Training Epoch: 18/100, batch 316/625 completed (loss: 0.0019:  51%|█████     | 159/312 [16:02<15:33,  6.10s/it]
Training Epoch: 18/100, batch 317/625 completed (loss: 0.0002:  51%|█████     | 159/312 [16:02<15:33,  6.10s/it]
Training Epoch: 18/100, batch 317/625 completed (loss: 0.0154:  51%|█████     | 159/312 [16:02<15:33,  6.10s/it]
Training Epoch: 18/100, batch 317/625 completed (loss: 0.0002:  51%|█████     | 159/312 [16:02<15:33,  6.10s/it]
Training Epoch: 18/100, batch 317/625 completed (loss: 0.0059:  51%|█████     | 159/312 [16:02<15:33,  6.10s/it]
Training Epoch: 18/100, batch 318/625 completed (loss: 0.0563:  51%|█████     | 159/312 [16:05<15:33,  6.10s/it]
Training Epoch: 18/100, batch 318/625 completed (loss: 0.0013:  51%|█████     | 159/312 [16:05<15:33,  6.10s/it]
Training Epoch: 18/100, batch 318/625 completed (loss: 0.016:  51%|█████     | 159/312 [16:05<15:33,  6.10s/it] 
Training Epoch: 18/100, batch 318/625 completed (loss: 0.0028:  51%|█████     | 159/312 [16:05<15:33,  6.10s/it]
Training Epoch: 18/100, batch 318/625 completed (loss: 0.0563:  51%|█████▏    | 160/312 [16:08<15:24,  6.08s/it]
Training Epoch: 18/100, batch 318/625 completed (loss: 0.0028:  51%|█████▏    | 160/312 [16:08<15:24,  6.08s/it]
Training Epoch: 18/100, batch 318/625 completed (loss: 0.016:  51%|█████▏    | 160/312 [16:08<15:24,  6.08s/it]
Training Epoch: 18/100, batch 318/625 completed (loss: 0.0013:  51%|█████▏    | 160/312 [16:08<15:23,  6.08s/it]
Training Epoch: 18/100, batch 319/625 completed (loss: 0.0007:  51%|█████▏    | 160/312 [16:08<15:23,  6.08s/it]
Training Epoch: 18/100, batch 319/625 completed (loss: 0.0022:  51%|█████▏    | 160/312 [16:08<15:24,  6.08s/it]
Training Epoch: 18/100, batch 319/625 completed (loss: 0.0004:  51%|█████▏    | 160/312 [16:08<15:24,  6.08s/it]
Training Epoch: 18/100, batch 319/625 completed (loss: 0.0007:  51%|█████▏    | 160/312 [16:08<15:24,  6.08s/it]
Training Epoch: 18/100, batch 320/625 completed (loss: 0.0022:  51%|█████▏    | 160/312 [16:11<15:23,  6.08s/it]
Training Epoch: 18/100, batch 320/625 completed (loss: 0.0538:  51%|█████▏    | 160/312 [16:11<15:24,  6.08s/it]
Training Epoch: 18/100, batch 320/625 completed (loss: 0.0029:  51%|█████▏    | 160/312 [16:11<15:24,  6.08s/it]
Training Epoch: 18/100, batch 320/625 completed (loss: 0.0129:  51%|█████▏    | 160/312 [16:11<15:24,  6.08s/it]
Training Epoch: 18/100, batch 320/625 completed (loss: 0.0129:  52%|█████▏    | 161/312 [16:14<15:20,  6.10s/it]
Training Epoch: 18/100, batch 320/625 completed (loss: 0.0538:  52%|█████▏    | 161/312 [16:14<15:20,  6.10s/it]
Training Epoch: 18/100, batch 320/625 completed (loss: 0.0022:  52%|█████▏    | 161/312 [16:14<15:20,  6.10s/it]
Training Epoch: 18/100, batch 320/625 completed (loss: 0.0029:  52%|█████▏    | 161/312 [16:14<15:21,  6.10s/it]
Training Epoch: 18/100, batch 321/625 completed (loss: 0.0023:  52%|█████▏    | 161/312 [16:14<15:21,  6.10s/it]
Training Epoch: 18/100, batch 321/625 completed (loss: 0.0011:  52%|█████▏    | 161/312 [16:14<15:20,  6.10s/it]
Training Epoch: 18/100, batch 321/625 completed (loss: 0.0003:  52%|█████▏    | 161/312 [16:14<15:20,  6.10s/it]
Training Epoch: 18/100, batch 321/625 completed (loss: 0.0045:  52%|█████▏    | 161/312 [16:14<15:20,  6.10s/it]
Training Epoch: 18/100, batch 322/625 completed (loss: 0.0022:  52%|█████▏    | 161/312 [16:17<15:21,  6.10s/it]
Training Epoch: 18/100, batch 322/625 completed (loss: 0.0067:  52%|█████▏    | 161/312 [16:17<15:20,  6.10s/it]
Training Epoch: 18/100, batch 322/625 completed (loss: 0.0004:  52%|█████▏    | 161/312 [16:17<15:20,  6.10s/it]
Training Epoch: 18/100, batch 322/625 completed (loss: 0.0:  52%|█████▏    | 161/312 [16:17<15:20,  6.10s/it]   
Training Epoch: 18/100, batch 322/625 completed (loss: 0.0067:  52%|█████▏    | 162/312 [16:20<15:10,  6.07s/it]
Training Epoch: 18/100, batch 322/625 completed (loss: 0.0:  52%|█████▏    | 162/312 [16:20<15:10,  6.07s/it]
Training Epoch: 18/100, batch 322/625 completed (loss: 0.0022:  52%|█████▏    | 162/312 [16:20<15:10,  6.07s/it]
Training Epoch: 18/100, batch 322/625 completed (loss: 0.0004:  52%|█████▏    | 162/312 [16:20<15:10,  6.07s/it]
Training Epoch: 18/100, batch 323/625 completed (loss: 0.0023:  52%|█████▏    | 162/312 [16:20<15:10,  6.07s/it]
Training Epoch: 18/100, batch 323/625 completed (loss: 0.0016:  52%|█████▏    | 162/312 [16:20<15:10,  6.07s/it]
Training Epoch: 18/100, batch 323/625 completed (loss: 0.0015:  52%|█████▏    | 162/312 [16:20<15:10,  6.07s/it]
Training Epoch: 18/100, batch 323/625 completed (loss: 0.0131:  52%|█████▏    | 162/312 [16:20<15:10,  6.07s/it]
Training Epoch: 18/100, batch 324/625 completed (loss: 0.0103:  52%|█████▏    | 162/312 [16:23<15:10,  6.07s/it]
Training Epoch: 18/100, batch 324/625 completed (loss: 0.035:  52%|█████▏    | 162/312 [16:23<15:10,  6.07s/it] 
Training Epoch: 18/100, batch 324/625 completed (loss: 0.0075:  52%|█████▏    | 162/312 [16:23<15:10,  6.07s/it]
Training Epoch: 18/100, batch 324/625 completed (loss: 0.0847:  52%|█████▏    | 162/312 [16:23<15:10,  6.07s/it]
Training Epoch: 18/100, batch 324/625 completed (loss: 0.0847:  52%|█████▏    | 163/312 [16:26<15:09,  6.10s/it]
Training Epoch: 18/100, batch 324/625 completed (loss: 0.035:  52%|█████▏    | 163/312 [16:26<15:09,  6.10s/it]
Training Epoch: 18/100, batch 324/625 completed (loss: 0.0075:  52%|█████▏    | 163/312 [16:26<15:09,  6.11s/it]
Training Epoch: 18/100, batch 324/625 completed (loss: 0.0103:  52%|█████▏    | 163/312 [16:26<15:09,  6.10s/it]
Training Epoch: 18/100, batch 325/625 completed (loss: 0.0154:  52%|█████▏    | 163/312 [16:26<15:09,  6.10s/it]
Training Epoch: 18/100, batch 325/625 completed (loss: 0.0006:  52%|█████▏    | 163/312 [16:26<15:09,  6.10s/it]
Training Epoch: 18/100, batch 325/625 completed (loss: 0.0038:  52%|█████▏    | 163/312 [16:26<15:09,  6.11s/it]
Training Epoch: 18/100, batch 325/625 completed (loss: 0.0041:  52%|█████▏    | 163/312 [16:26<15:09,  6.10s/it]
Training Epoch: 18/100, batch 326/625 completed (loss: 0.0038:  52%|█████▏    | 163/312 [16:29<15:09,  6.11s/it]
Training Epoch: 18/100, batch 326/625 completed (loss: 0.011:  52%|█████▏    | 163/312 [16:29<15:09,  6.10s/it] 
Training Epoch: 18/100, batch 326/625 completed (loss: 0.0077:  52%|█████▏    | 163/312 [16:29<15:09,  6.10s/it]
Training Epoch: 18/100, batch 326/625 completed (loss: 0.0167:  52%|█████▏    | 163/312 [16:29<15:09,  6.10s/it]
Training Epoch: 18/100, batch 326/625 completed (loss: 0.011:  53%|█████▎    | 164/312 [16:32<15:05,  6.12s/it]
Training Epoch: 18/100, batch 326/625 completed (loss: 0.0038:  53%|█████▎    | 164/312 [16:32<15:06,  6.12s/it]
Training Epoch: 18/100, batch 326/625 completed (loss: 0.0077:  53%|█████▎    | 164/312 [16:32<15:05,  6.12s/it]
Training Epoch: 18/100, batch 326/625 completed (loss: 0.0167:  53%|█████▎    | 164/312 [16:32<15:06,  6.13s/it]
Training Epoch: 18/100, batch 327/625 completed (loss: 0.0213:  53%|█████▎    | 164/312 [16:32<15:05,  6.12s/it]
Training Epoch: 18/100, batch 327/625 completed (loss: 0.0014:  53%|█████▎    | 164/312 [16:32<15:05,  6.12s/it]
Training Epoch: 18/100, batch 327/625 completed (loss: 0.001:  53%|█████▎    | 164/312 [16:32<15:06,  6.12s/it] 
Training Epoch: 18/100, batch 327/625 completed (loss: 0.0014:  53%|█████▎    | 164/312 [16:32<15:06,  6.13s/it]
Training Epoch: 18/100, batch 328/625 completed (loss: 0.0066:  53%|█████▎    | 164/312 [16:35<15:06,  6.12s/it]
Training Epoch: 18/100, batch 328/625 completed (loss: 0.0022:  53%|█████▎    | 164/312 [16:35<15:05,  6.12s/it]
Training Epoch: 18/100, batch 328/625 completed (loss: 0.137:  53%|█████▎    | 164/312 [16:35<15:05,  6.12s/it] 
Training Epoch: 18/100, batch 328/625 completed (loss: 0.0001:  53%|█████▎    | 164/312 [16:35<15:06,  6.13s/it]
Training Epoch: 18/100, batch 328/625 completed (loss: 0.0022:  53%|█████▎    | 165/312 [16:38<14:55,  6.09s/it]
Training Epoch: 18/100, batch 328/625 completed (loss: 0.0001:  53%|█████▎    | 165/312 [16:38<14:55,  6.09s/it]
Training Epoch: 18/100, batch 328/625 completed (loss: 0.137:  53%|█████▎    | 165/312 [16:38<14:55,  6.09s/it]
Training Epoch: 18/100, batch 328/625 completed (loss: 0.0066:  53%|█████▎    | 165/312 [16:38<14:55,  6.09s/it]
Training Epoch: 18/100, batch 329/625 completed (loss: 0.0019:  53%|█████▎    | 165/312 [16:38<14:55,  6.09s/it]
Training Epoch: 18/100, batch 329/625 completed (loss: 0.0105:  53%|█████▎    | 165/312 [16:38<14:55,  6.09s/it]
Training Epoch: 18/100, batch 329/625 completed (loss: 0.0177:  53%|█████▎    | 165/312 [16:38<14:55,  6.09s/it]
Training Epoch: 18/100, batch 329/625 completed (loss: 0.0002:  53%|█████▎    | 165/312 [16:38<14:55,  6.09s/it]
Training Epoch: 18/100, batch 330/625 completed (loss: 0.0301:  53%|█████▎    | 165/312 [16:41<14:55,  6.09s/it]
Training Epoch: 18/100, batch 330/625 completed (loss: 0.0081:  53%|█████▎    | 165/312 [16:41<14:55,  6.09s/it]
Training Epoch: 18/100, batch 330/625 completed (loss: 0.0024:  53%|█████▎    | 165/312 [16:41<14:55,  6.09s/it]
Training Epoch: 18/100, batch 330/625 completed (loss: 0.0004:  53%|█████▎    | 165/312 [16:41<14:55,  6.09s/it]
Training Epoch: 18/100, batch 330/625 completed (loss: 0.0081:  53%|█████▎    | 166/312 [16:44<14:46,  6.07s/it]
Training Epoch: 18/100, batch 330/625 completed (loss: 0.0301:  53%|█████▎    | 166/312 [16:44<14:46,  6.07s/it]
Training Epoch: 18/100, batch 330/625 completed (loss: 0.0004:  53%|█████▎    | 166/312 [16:44<14:46,  6.07s/it]
Training Epoch: 18/100, batch 330/625 completed (loss: 0.0024:  53%|█████▎    | 166/312 [16:44<14:46,  6.07s/it]
Training Epoch: 18/100, batch 331/625 completed (loss: 0.0081:  53%|█████▎    | 166/312 [16:44<14:46,  6.07s/it]
Training Epoch: 18/100, batch 331/625 completed (loss: 0.0004:  53%|█████▎    | 166/312 [16:44<14:46,  6.07s/it]
Training Epoch: 18/100, batch 331/625 completed (loss: 0.0002:  53%|█████▎    | 166/312 [16:44<14:46,  6.07s/it]
Training Epoch: 18/100, batch 331/625 completed (loss: 0.052:  53%|█████▎    | 166/312 [16:44<14:46,  6.07s/it] 
Training Epoch: 18/100, batch 332/625 completed (loss: 0.0061:  53%|█████▎    | 166/312 [16:47<14:46,  6.07s/it]
Training Epoch: 18/100, batch 332/625 completed (loss: 0.0034:  53%|█████▎    | 166/312 [16:47<14:46,  6.07s/it]
Training Epoch: 18/100, batch 332/625 completed (loss: 0.0004:  53%|█████▎    | 166/312 [16:47<14:46,  6.07s/it]
Training Epoch: 18/100, batch 332/625 completed (loss: 0.0019:  53%|█████▎    | 166/312 [16:48<14:46,  6.07s/it]
Training Epoch: 18/100, batch 332/625 completed (loss: 0.0061:  54%|█████▎    | 167/312 [16:50<14:39,  6.06s/it]
Training Epoch: 18/100, batch 332/625 completed (loss: 0.0019:  54%|█████▎    | 167/312 [16:50<14:39,  6.06s/it]
Training Epoch: 18/100, batch 332/625 completed (loss: 0.0004:  54%|█████▎    | 167/312 [16:50<14:39,  6.07s/it]
Training Epoch: 18/100, batch 332/625 completed (loss: 0.0034:  54%|█████▎    | 167/312 [16:50<14:39,  6.07s/it]
Training Epoch: 18/100, batch 333/625 completed (loss: 0.0002:  54%|█████▎    | 167/312 [16:50<14:39,  6.06s/it]
Training Epoch: 18/100, batch 333/625 completed (loss: 0.0118:  54%|█████▎    | 167/312 [16:50<14:39,  6.07s/it]
Training Epoch: 18/100, batch 333/625 completed (loss: 0.0063:  54%|█████▎    | 167/312 [16:50<14:39,  6.07s/it]
Training Epoch: 18/100, batch 333/625 completed (loss: 0.0008:  54%|█████▎    | 167/312 [16:50<14:39,  6.06s/it]
Training Epoch: 18/100, batch 334/625 completed (loss: 0.0654:  54%|█████▎    | 167/312 [16:53<14:39,  6.07s/it]
Training Epoch: 18/100, batch 334/625 completed (loss: 0.0009:  54%|█████▎    | 167/312 [16:53<14:39,  6.07s/it]
Training Epoch: 18/100, batch 334/625 completed (loss: 0.0033:  54%|█████▎    | 167/312 [16:53<14:39,  6.06s/it]
Training Epoch: 18/100, batch 334/625 completed (loss: 0.0033:  54%|█████▎    | 167/312 [16:53<14:39,  6.06s/it]
Training Epoch: 18/100, batch 334/625 completed (loss: 0.0033:  54%|█████▍    | 168/312 [16:56<14:31,  6.05s/it]
Training Epoch: 18/100, batch 334/625 completed (loss: 0.0009:  54%|█████▍    | 168/312 [16:56<14:30,  6.05s/it]
Training Epoch: 18/100, batch 334/625 completed (loss: 0.0654:  54%|█████▍    | 168/312 [16:56<14:30,  6.05s/it]
Training Epoch: 18/100, batch 334/625 completed (loss: 0.0033:  54%|█████▍    | 168/312 [16:56<14:31,  6.05s/it]
Training Epoch: 18/100, batch 335/625 completed (loss: 0.0001:  54%|█████▍    | 168/312 [16:56<14:31,  6.05s/it]
Training Epoch: 18/100, batch 335/625 completed (loss: 0.0002:  54%|█████▍    | 168/312 [16:56<14:30,  6.05s/it]
Training Epoch: 18/100, batch 335/625 completed (loss: 0.0002:  54%|█████▍    | 168/312 [16:56<14:30,  6.05s/it]
Training Epoch: 18/100, batch 335/625 completed (loss: 0.0217:  54%|█████▍    | 168/312 [16:56<14:31,  6.05s/it]
Training Epoch: 18/100, batch 336/625 completed (loss: 0.0028:  54%|█████▍    | 168/312 [16:59<14:31,  6.05s/it]
Training Epoch: 18/100, batch 336/625 completed (loss: 0.0113:  54%|█████▍    | 168/312 [16:59<14:30,  6.05s/it]
Training Epoch: 18/100, batch 336/625 completed (loss: 0.0001:  54%|█████▍    | 168/312 [16:59<14:30,  6.05s/it]
Training Epoch: 18/100, batch 336/625 completed (loss: 0.0037:  54%|█████▍    | 168/312 [17:00<14:31,  6.05s/it]
Training Epoch: 18/100, batch 336/625 completed (loss: 0.0028:  54%|█████▍    | 169/312 [17:03<14:35,  6.12s/it]
Training Epoch: 18/100, batch 336/625 completed (loss: 0.0001:  54%|█████▍    | 169/312 [17:03<14:34,  6.12s/it]
Training Epoch: 18/100, batch 336/625 completed (loss: 0.0113:  54%|█████▍    | 169/312 [17:03<14:34,  6.12s/it]
Training Epoch: 18/100, batch 336/625 completed (loss: 0.0037:  54%|█████▍    | 169/312 [17:03<14:34,  6.12s/it]
Training Epoch: 18/100, batch 337/625 completed (loss: 0.0034:  54%|█████▍    | 169/312 [17:03<14:34,  6.12s/it]
Training Epoch: 18/100, batch 337/625 completed (loss: 0.0005:  54%|█████▍    | 169/312 [17:03<14:35,  6.12s/it]
Training Epoch: 18/100, batch 337/625 completed (loss: 0.0033:  54%|█████▍    | 169/312 [17:03<14:34,  6.12s/it]
Training Epoch: 18/100, batch 337/625 completed (loss: 0.0013:  54%|█████▍    | 169/312 [17:03<14:34,  6.12s/it]
Training Epoch: 18/100, batch 338/625 completed (loss: 0.001:  54%|█████▍    | 169/312 [17:06<14:34,  6.12s/it] 
Training Epoch: 18/100, batch 338/625 completed (loss: 0.0177:  54%|█████▍    | 169/312 [17:06<14:34,  6.12s/it]
Training Epoch: 18/100, batch 338/625 completed (loss: 0.0036:  54%|█████▍    | 169/312 [17:06<14:35,  6.12s/it]
Training Epoch: 18/100, batch 338/625 completed (loss: 0.0031:  54%|█████▍    | 169/312 [17:06<14:34,  6.12s/it]
Training Epoch: 18/100, batch 338/625 completed (loss: 0.0036:  54%|█████▍    | 170/312 [17:08<14:16,  6.03s/it]
Training Epoch: 18/100, batch 338/625 completed (loss: 0.0177:  54%|█████▍    | 170/312 [17:08<14:16,  6.03s/it]
Training Epoch: 18/100, batch 338/625 completed (loss: 0.0031:  54%|█████▍    | 170/312 [17:09<14:16,  6.03s/it]
Training Epoch: 18/100, batch 338/625 completed (loss: 0.001:  54%|█████▍    | 170/312 [17:08<14:16,  6.03s/it]
Training Epoch: 18/100, batch 339/625 completed (loss: 0.0269:  54%|█████▍    | 170/312 [17:09<14:16,  6.03s/it]
Training Epoch: 18/100, batch 339/625 completed (loss: 0.0004:  54%|█████▍    | 170/312 [17:09<14:16,  6.03s/it]
Training Epoch: 18/100, batch 339/625 completed (loss: 0.0039:  54%|█████▍    | 170/312 [17:09<14:16,  6.03s/it]
Training Epoch: 18/100, batch 339/625 completed (loss: 0.0023:  54%|█████▍    | 170/312 [17:09<14:16,  6.03s/it]
Training Epoch: 18/100, batch 340/625 completed (loss: 0.0008:  54%|█████▍    | 170/312 [17:12<14:16,  6.03s/it]
Training Epoch: 18/100, batch 340/625 completed (loss: 0.2562:  54%|█████▍    | 170/312 [17:12<14:16,  6.03s/it]
Training Epoch: 18/100, batch 340/625 completed (loss: 0.0007:  54%|█████▍    | 170/312 [17:12<14:16,  6.03s/it]
Training Epoch: 18/100, batch 340/625 completed (loss: 0.0007:  54%|█████▍    | 170/312 [17:12<14:16,  6.03s/it]
Training Epoch: 18/100, batch 340/625 completed (loss: 0.0007:  55%|█████▍    | 171/312 [17:15<14:10,  6.04s/it]
Training Epoch: 18/100, batch 340/625 completed (loss: 0.2562:  55%|█████▍    | 171/312 [17:15<14:11,  6.04s/it]
Training Epoch: 18/100, batch 340/625 completed (loss: 0.0007:  55%|█████▍    | 171/312 [17:15<14:11,  6.04s/it]
Training Epoch: 18/100, batch 340/625 completed (loss: 0.0008:  55%|█████▍    | 171/312 [17:15<14:11,  6.04s/it]
Training Epoch: 18/100, batch 341/625 completed (loss: 0.0518:  55%|█████▍    | 171/312 [17:15<14:11,  6.04s/it]
Training Epoch: 18/100, batch 341/625 completed (loss: 0.155:  55%|█████▍    | 171/312 [17:15<14:11,  6.04s/it] 
Training Epoch: 18/100, batch 341/625 completed (loss: 0.0009:  55%|█████▍    | 171/312 [17:15<14:11,  6.04s/it]
Training Epoch: 18/100, batch 341/625 completed (loss: 0.001:  55%|█████▍    | 171/312 [17:15<14:10,  6.04s/it] 
Training Epoch: 18/100, batch 342/625 completed (loss: 0.0022:  55%|█████▍    | 171/312 [17:18<14:11,  6.04s/it]
Training Epoch: 18/100, batch 342/625 completed (loss: 0.0003:  55%|█████▍    | 171/312 [17:18<14:11,  6.04s/it]
Training Epoch: 18/100, batch 342/625 completed (loss: 0.0004:  55%|█████▍    | 171/312 [17:18<14:11,  6.04s/it]
Training Epoch: 18/100, batch 342/625 completed (loss: 0.0041:  55%|█████▍    | 171/312 [17:18<14:10,  6.04s/it]
Training Epoch: 18/100, batch 342/625 completed (loss: 0.0041:  55%|█████▌    | 172/312 [17:21<14:16,  6.12s/it]
Training Epoch: 18/100, batch 342/625 completed (loss: 0.0004:  55%|█████▌    | 172/312 [17:21<14:17,  6.12s/it]
Training Epoch: 18/100, batch 342/625 completed (loss: 0.0022:  55%|█████▌    | 172/312 [17:21<14:16,  6.12s/it]
Training Epoch: 18/100, batch 342/625 completed (loss: 0.0003:  55%|█████▌    | 172/312 [17:21<14:17,  6.12s/it]
Training Epoch: 18/100, batch 343/625 completed (loss: 0.0111:  55%|█████▌    | 172/312 [17:21<14:17,  6.12s/it]
Training Epoch: 18/100, batch 343/625 completed (loss: 0.0231:  55%|█████▌    | 172/312 [17:21<14:16,  6.12s/it]
Training Epoch: 18/100, batch 343/625 completed (loss: 0.0042:  55%|█████▌    | 172/312 [17:21<14:17,  6.12s/it]
Training Epoch: 18/100, batch 343/625 completed (loss: 0.011:  55%|█████▌    | 172/312 [17:21<14:16,  6.12s/it] 
Training Epoch: 18/100, batch 344/625 completed (loss: 0.0026:  55%|█████▌    | 172/312 [17:24<14:17,  6.12s/it]
Training Epoch: 18/100, batch 344/625 completed (loss: 0.0003:  55%|█████▌    | 172/312 [17:24<14:16,  6.12s/it]
Training Epoch: 18/100, batch 344/625 completed (loss: 0.0078:  55%|█████▌    | 172/312 [17:24<14:17,  6.12s/it]
Training Epoch: 18/100, batch 344/625 completed (loss: 0.0005:  55%|█████▌    | 172/312 [17:24<14:16,  6.12s/it]
Training Epoch: 18/100, batch 344/625 completed (loss: 0.0026:  55%|█████▌    | 173/312 [17:27<14:06,  6.09s/it]
Training Epoch: 18/100, batch 344/625 completed (loss: 0.0078:  55%|█████▌    | 173/312 [17:27<14:06,  6.09s/it]
Training Epoch: 18/100, batch 344/625 completed (loss: 0.0005:  55%|█████▌    | 173/312 [17:27<14:06,  6.09s/it]
Training Epoch: 18/100, batch 344/625 completed (loss: 0.0003:  55%|█████▌    | 173/312 [17:27<14:06,  6.09s/it]
Training Epoch: 18/100, batch 345/625 completed (loss: 0.1192:  55%|█████▌    | 173/312 [17:27<14:06,  6.09s/it]
Training Epoch: 18/100, batch 345/625 completed (loss: 0.0016:  55%|█████▌    | 173/312 [17:27<14:06,  6.09s/it]
Training Epoch: 18/100, batch 345/625 completed (loss: 0.0898:  55%|█████▌    | 173/312 [17:27<14:06,  6.09s/it]
Training Epoch: 18/100, batch 345/625 completed (loss: 0.0009:  55%|█████▌    | 173/312 [17:27<14:06,  6.09s/it]
Training Epoch: 18/100, batch 346/625 completed (loss: 0.0001:  55%|█████▌    | 173/312 [17:30<14:06,  6.09s/it]
Training Epoch: 18/100, batch 346/625 completed (loss: 0.0077:  55%|█████▌    | 173/312 [17:30<14:06,  6.09s/it]
Training Epoch: 18/100, batch 346/625 completed (loss: 0.0003:  55%|█████▌    | 173/312 [17:30<14:06,  6.09s/it]
Training Epoch: 18/100, batch 346/625 completed (loss: 0.2375:  55%|█████▌    | 173/312 [17:30<14:06,  6.09s/it]
Training Epoch: 18/100, batch 346/625 completed (loss: 0.0003:  56%|█████▌    | 174/312 [17:33<14:03,  6.12s/it]
Training Epoch: 18/100, batch 346/625 completed (loss: 0.0001:  56%|█████▌    | 174/312 [17:33<14:03,  6.11s/it]
Training Epoch: 18/100, batch 346/625 completed (loss: 0.0077:  56%|█████▌    | 174/312 [17:33<14:03,  6.11s/it]
Training Epoch: 18/100, batch 346/625 completed (loss: 0.2375:  56%|█████▌    | 174/312 [17:33<14:04,  6.12s/it]
Training Epoch: 18/100, batch 347/625 completed (loss: 0.003:  56%|█████▌    | 174/312 [17:33<14:03,  6.11s/it] 
Training Epoch: 18/100, batch 347/625 completed (loss: 0.0036:  56%|█████▌    | 174/312 [17:33<14:03,  6.11s/it]
Training Epoch: 18/100, batch 347/625 completed (loss: 0.0048:  56%|█████▌    | 174/312 [17:33<14:03,  6.12s/it]
Training Epoch: 18/100, batch 347/625 completed (loss: 0.0241:  56%|█████▌    | 174/312 [17:33<14:04,  6.12s/it]
Training Epoch: 18/100, batch 348/625 completed (loss: 0.0042:  56%|█████▌    | 174/312 [17:36<14:03,  6.11s/it]
Training Epoch: 18/100, batch 348/625 completed (loss: 0.0017:  56%|█████▌    | 174/312 [17:36<14:03,  6.12s/it]
Training Epoch: 18/100, batch 348/625 completed (loss: 0.0025:  56%|█████▌    | 174/312 [17:36<14:03,  6.11s/it]
Training Epoch: 18/100, batch 348/625 completed (loss: 0.0013:  56%|█████▌    | 174/312 [17:36<14:04,  6.12s/it]
Training Epoch: 18/100, batch 348/625 completed (loss: 0.0042:  56%|█████▌    | 175/312 [17:39<13:52,  6.08s/it]
Training Epoch: 18/100, batch 348/625 completed (loss: 0.0017:  56%|█████▌    | 175/312 [17:39<13:53,  6.08s/it]
Training Epoch: 18/100, batch 348/625 completed (loss: 0.0025:  56%|█████▌    | 175/312 [17:39<13:53,  6.08s/it]
Training Epoch: 18/100, batch 348/625 completed (loss: 0.0013:  56%|█████▌    | 175/312 [17:39<13:53,  6.08s/it]
Training Epoch: 18/100, batch 349/625 completed (loss: 0.0863:  56%|█████▌    | 175/312 [17:39<13:53,  6.08s/it]
Training Epoch: 18/100, batch 349/625 completed (loss: 0.0037:  56%|█████▌    | 175/312 [17:39<13:52,  6.08s/it]
Training Epoch: 18/100, batch 349/625 completed (loss: 0.0632:  56%|█████▌    | 175/312 [17:39<13:53,  6.08s/it]
Training Epoch: 18/100, batch 349/625 completed (loss: 0.002:  56%|█████▌    | 175/312 [17:39<13:53,  6.08s/it] 
Training Epoch: 18/100, batch 350/625 completed (loss: 0.0049:  56%|█████▌    | 175/312 [17:42<13:53,  6.08s/it]
Training Epoch: 18/100, batch 350/625 completed (loss: 0.0292:  56%|█████▌    | 175/312 [17:42<13:52,  6.08s/it]
Training Epoch: 18/100, batch 350/625 completed (loss: 0.0004:  56%|█████▌    | 175/312 [17:42<13:53,  6.08s/it]
Training Epoch: 18/100, batch 350/625 completed (loss: 0.0025:  56%|█████▌    | 175/312 [17:42<13:53,  6.08s/it]
Training Epoch: 18/100, batch 350/625 completed (loss: 0.0049:  56%|█████▋    | 176/312 [17:45<13:44,  6.06s/it]
Training Epoch: 18/100, batch 350/625 completed (loss: 0.0292:  56%|█████▋    | 176/312 [17:45<13:44,  6.06s/it]
Training Epoch: 18/100, batch 350/625 completed (loss: 0.0025:  56%|█████▋    | 176/312 [17:45<13:45,  6.07s/it]
Training Epoch: 18/100, batch 350/625 completed (loss: 0.0004:  56%|█████▋    | 176/312 [17:45<13:45,  6.07s/it]
Training Epoch: 18/100, batch 351/625 completed (loss: 0.0007:  56%|█████▋    | 176/312 [17:45<13:44,  6.06s/it]
Training Epoch: 18/100, batch 351/625 completed (loss: 0.0005:  56%|█████▋    | 176/312 [17:45<13:44,  6.06s/it]
Training Epoch: 18/100, batch 351/625 completed (loss: 0.001:  56%|█████▋    | 176/312 [17:45<13:45,  6.07s/it] 
Training Epoch: 18/100, batch 351/625 completed (loss: 0.0006:  56%|█████▋    | 176/312 [17:45<13:45,  6.07s/it]
Training Epoch: 18/100, batch 352/625 completed (loss: 0.0012:  56%|█████▋    | 176/312 [17:48<13:44,  6.06s/it]
Training Epoch: 18/100, batch 352/625 completed (loss: 0.0015:  56%|█████▋    | 176/312 [17:48<13:45,  6.07s/it]
Training Epoch: 18/100, batch 352/625 completed (loss: 0.0004:  56%|█████▋    | 176/312 [17:48<13:44,  6.06s/it]
Training Epoch: 18/100, batch 352/625 completed (loss: 0.0006:  56%|█████▋    | 176/312 [17:48<13:45,  6.07s/it]
Training Epoch: 18/100, batch 352/625 completed (loss: 0.0012:  57%|█████▋    | 177/312 [17:51<13:41,  6.09s/it]
Training Epoch: 18/100, batch 352/625 completed (loss: 0.0004:  57%|█████▋    | 177/312 [17:51<13:41,  6.09s/it]
Training Epoch: 18/100, batch 352/625 completed (loss: 0.0006:  57%|█████▋    | 177/312 [17:51<13:41,  6.09s/it]
Training Epoch: 18/100, batch 352/625 completed (loss: 0.0015:  57%|█████▋    | 177/312 [17:51<13:41,  6.09s/it]
Training Epoch: 18/100, batch 353/625 completed (loss: 0.0305:  57%|█████▋    | 177/312 [17:51<13:41,  6.09s/it]
Training Epoch: 18/100, batch 353/625 completed (loss: 0.0007:  57%|█████▋    | 177/312 [17:51<13:41,  6.09s/it]
Training Epoch: 18/100, batch 353/625 completed (loss: 0.005:  57%|█████▋    | 177/312 [17:51<13:41,  6.09s/it] 
Training Epoch: 18/100, batch 353/625 completed (loss: 0.018:  57%|█████▋    | 177/312 [17:51<13:41,  6.09s/it] 
Training Epoch: 18/100, batch 354/625 completed (loss: 0.0024:  57%|█████▋    | 177/312 [17:54<13:41,  6.09s/it]
Training Epoch: 18/100, batch 354/625 completed (loss: 0.0031:  57%|█████▋    | 177/312 [17:54<13:41,  6.09s/it]
Training Epoch: 18/100, batch 354/625 completed (loss: 0.0017:  57%|█████▋    | 177/312 [17:54<13:41,  6.09s/it]
Training Epoch: 18/100, batch 354/625 completed (loss: 0.0085:  57%|█████▋    | 177/312 [17:54<13:41,  6.09s/it]
Training Epoch: 18/100, batch 354/625 completed (loss: 0.0024:  57%|█████▋    | 178/312 [17:57<13:31,  6.06s/it]
Training Epoch: 18/100, batch 354/625 completed (loss: 0.0017:  57%|█████▋    | 178/312 [17:57<13:31,  6.06s/it]
Training Epoch: 18/100, batch 354/625 completed (loss: 0.0085:  57%|█████▋    | 178/312 [17:57<13:31,  6.06s/it]
Training Epoch: 18/100, batch 354/625 completed (loss: 0.0031:  57%|█████▋    | 178/312 [17:57<13:31,  6.06s/it]
Training Epoch: 18/100, batch 355/625 completed (loss: 0.0026:  57%|█████▋    | 178/312 [17:57<13:31,  6.06s/it]
Training Epoch: 18/100, batch 355/625 completed (loss: 0.0108:  57%|█████▋    | 178/312 [17:57<13:31,  6.06s/it]
Training Epoch: 18/100, batch 355/625 completed (loss: 0.0131:  57%|█████▋    | 178/312 [17:57<13:31,  6.06s/it]
Training Epoch: 18/100, batch 355/625 completed (loss: 0.0032:  57%|█████▋    | 178/312 [17:57<13:31,  6.06s/it]
Training Epoch: 18/100, batch 356/625 completed (loss: 0.0002:  57%|█████▋    | 178/312 [18:00<13:31,  6.06s/it]
Training Epoch: 18/100, batch 356/625 completed (loss: 0.0003:  57%|█████▋    | 178/312 [18:00<13:31,  6.06s/it]
Training Epoch: 18/100, batch 356/625 completed (loss: 0.0369:  57%|█████▋    | 178/312 [18:00<13:31,  6.06s/it]
Training Epoch: 18/100, batch 356/625 completed (loss: 0.0019:  57%|█████▋    | 178/312 [18:00<13:31,  6.06s/it]
Training Epoch: 18/100, batch 356/625 completed (loss: 0.0003:  57%|█████▋    | 179/312 [18:03<13:30,  6.09s/it]
Training Epoch: 18/100, batch 356/625 completed (loss: 0.0002:  57%|█████▋    | 179/312 [18:03<13:30,  6.09s/it]
Training Epoch: 18/100, batch 356/625 completed (loss: 0.0369:  57%|█████▋    | 179/312 [18:03<13:30,  6.10s/it]
Training Epoch: 18/100, batch 356/625 completed (loss: 0.0019:  57%|█████▋    | 179/312 [18:03<13:30,  6.09s/it]
Training Epoch: 18/100, batch 357/625 completed (loss: 0.0987:  57%|█████▋    | 179/312 [18:03<13:30,  6.09s/it]
Training Epoch: 18/100, batch 357/625 completed (loss: 0.0003:  57%|█████▋    | 179/312 [18:03<13:30,  6.10s/it]
Training Epoch: 18/100, batch 357/625 completed (loss: 0.0061:  57%|█████▋    | 179/312 [18:03<13:30,  6.09s/it]
Training Epoch: 18/100, batch 357/625 completed (loss: 0.0021:  57%|█████▋    | 179/312 [18:03<13:30,  6.09s/it]
Training Epoch: 18/100, batch 358/625 completed (loss: 0.0001:  57%|█████▋    | 179/312 [18:06<13:30,  6.09s/it]
Training Epoch: 18/100, batch 358/625 completed (loss: 0.0003:  57%|█████▋    | 179/312 [18:07<13:30,  6.10s/it]
Training Epoch: 18/100, batch 358/625 completed (loss: 0.0001:  57%|█████▋    | 179/312 [18:07<13:30,  6.09s/it]
Training Epoch: 18/100, batch 358/625 completed (loss: 0.0004:  57%|█████▋    | 179/312 [18:07<13:30,  6.09s/it]
Training Epoch: 18/100, batch 358/625 completed (loss: 0.0003:  58%|█████▊    | 180/312 [18:10<13:26,  6.11s/it]
Training Epoch: 18/100, batch 358/625 completed (loss: 0.0001:  58%|█████▊    | 180/312 [18:10<13:27,  6.11s/it]
Training Epoch: 18/100, batch 358/625 completed (loss: 0.0001:  58%|█████▊    | 180/312 [18:10<13:26,  6.11s/it]
Training Epoch: 18/100, batch 358/625 completed (loss: 0.0004:  58%|█████▊    | 180/312 [18:10<13:27,  6.12s/it]
Training Epoch: 18/100, batch 359/625 completed (loss: 0.2506:  58%|█████▊    | 180/312 [18:10<13:26,  6.11s/it]
Training Epoch: 18/100, batch 359/625 completed (loss: 0.0011:  58%|█████▊    | 180/312 [18:10<13:26,  6.11s/it]
Training Epoch: 18/100, batch 359/625 completed (loss: 0.0143:  58%|█████▊    | 180/312 [18:10<13:27,  6.11s/it]
Training Epoch: 18/100, batch 359/625 completed (loss: 0.0028:  58%|█████▊    | 180/312 [18:10<13:27,  6.12s/it]
Training Epoch: 18/100, batch 360/625 completed (loss: 0.0644:  58%|█████▊    | 180/312 [18:13<13:26,  6.11s/it]
Training Epoch: 18/100, batch 360/625 completed (loss: 0.0021:  58%|█████▊    | 180/312 [18:13<13:26,  6.11s/it]
Training Epoch: 18/100, batch 360/625 completed (loss: 0.0525:  58%|█████▊    | 180/312 [18:13<13:27,  6.11s/it]
Training Epoch: 18/100, batch 360/625 completed (loss: 0.0008:  58%|█████▊    | 180/312 [18:13<13:27,  6.12s/it]
Training Epoch: 18/100, batch 360/625 completed (loss: 0.0644:  58%|█████▊    | 181/312 [18:15<13:11,  6.05s/it]
Training Epoch: 18/100, batch 360/625 completed (loss: 0.0525:  58%|█████▊    | 181/312 [18:15<13:11,  6.04s/it]
Training Epoch: 18/100, batch 360/625 completed (loss: 0.0021:  58%|█████▊    | 181/312 [18:15<13:11,  6.04s/it]
Training Epoch: 18/100, batch 360/625 completed (loss: 0.0008:  58%|█████▊    | 181/312 [18:15<13:12,  6.05s/it]
Training Epoch: 18/100, batch 361/625 completed (loss: 0.0403:  58%|█████▊    | 181/312 [18:15<13:11,  6.04s/it]
Training Epoch: 18/100, batch 361/625 completed (loss: 0.0005:  58%|█████▊    | 181/312 [18:15<13:11,  6.04s/it]
Training Epoch: 18/100, batch 361/625 completed (loss: 0.0001:  58%|█████▊    | 181/312 [18:15<13:11,  6.05s/it]
Training Epoch: 18/100, batch 361/625 completed (loss: 0.0017:  58%|█████▊    | 181/312 [18:16<13:12,  6.05s/it]
Training Epoch: 18/100, batch 362/625 completed (loss: 0.0004:  58%|█████▊    | 181/312 [18:19<13:11,  6.04s/it]
Training Epoch: 18/100, batch 362/625 completed (loss: 0.0174:  58%|█████▊    | 181/312 [18:19<13:11,  6.05s/it]
Training Epoch: 18/100, batch 362/625 completed (loss: 0.005:  58%|█████▊    | 181/312 [18:19<13:11,  6.04s/it] 
Training Epoch: 18/100, batch 362/625 completed (loss: 0.0007:  58%|█████▊    | 181/312 [18:19<13:12,  6.05s/it]
Training Epoch: 18/100, batch 362/625 completed (loss: 0.0004:  58%|█████▊    | 182/312 [18:22<13:15,  6.12s/it]
Training Epoch: 18/100, batch 362/625 completed (loss: 0.0174:  58%|█████▊    | 182/312 [18:22<13:15,  6.12s/it]
Training Epoch: 18/100, batch 362/625 completed (loss: 0.005:  58%|█████▊    | 182/312 [18:22<13:16,  6.12s/it]
Training Epoch: 18/100, batch 362/625 completed (loss: 0.0007:  58%|█████▊    | 182/312 [18:22<13:15,  6.12s/it]
Training Epoch: 18/100, batch 363/625 completed (loss: 0.001:  58%|█████▊    | 182/312 [18:22<13:15,  6.12s/it] 
Training Epoch: 18/100, batch 363/625 completed (loss: 0.0015:  58%|█████▊    | 182/312 [18:22<13:16,  6.12s/it]
Training Epoch: 18/100, batch 363/625 completed (loss: 0.0056:  58%|█████▊    | 182/312 [18:22<13:15,  6.12s/it]
Training Epoch: 18/100, batch 363/625 completed (loss: 0.0002:  58%|█████▊    | 182/312 [18:22<13:15,  6.12s/it]
Training Epoch: 18/100, batch 364/625 completed (loss: 0.0039:  58%|█████▊    | 182/312 [18:25<13:15,  6.12s/it]
Training Epoch: 18/100, batch 364/625 completed (loss: 0.0073:  58%|█████▊    | 182/312 [18:25<13:15,  6.12s/it]
Training Epoch: 18/100, batch 364/625 completed (loss: 0.0024:  58%|█████▊    | 182/312 [18:25<13:16,  6.12s/it]
Training Epoch: 18/100, batch 364/625 completed (loss: 0.0062:  58%|█████▊    | 182/312 [18:25<13:15,  6.12s/it]
Training Epoch: 18/100, batch 364/625 completed (loss: 0.0039:  59%|█████▊    | 183/312 [18:28<13:05,  6.09s/it]
Training Epoch: 18/100, batch 364/625 completed (loss: 0.0073:  59%|█████▊    | 183/312 [18:28<13:06,  6.09s/it]
Training Epoch: 18/100, batch 364/625 completed (loss: 0.0062:  59%|█████▊    | 183/312 [18:28<13:06,  6.09s/it]
Training Epoch: 18/100, batch 364/625 completed (loss: 0.0024:  59%|█████▊    | 183/312 [18:28<13:06,  6.10s/it]
Training Epoch: 18/100, batch 365/625 completed (loss: 0.0003:  59%|█████▊    | 183/312 [18:28<13:06,  6.09s/it]
Training Epoch: 18/100, batch 365/625 completed (loss: 0.0007:  59%|█████▊    | 183/312 [18:28<13:06,  6.10s/it]
Training Epoch: 18/100, batch 365/625 completed (loss: 0.0026:  59%|█████▊    | 183/312 [18:28<13:05,  6.09s/it]
Training Epoch: 18/100, batch 365/625 completed (loss: 0.1925:  59%|█████▊    | 183/312 [18:28<13:06,  6.09s/it]
Training Epoch: 18/100, batch 366/625 completed (loss: 0.0006:  59%|█████▊    | 183/312 [18:31<13:05,  6.09s/it]
Training Epoch: 18/100, batch 366/625 completed (loss: 0.0257:  59%|█████▊    | 183/312 [18:31<13:06,  6.10s/it]
Training Epoch: 18/100, batch 366/625 completed (loss: 0.1166:  59%|█████▊    | 183/312 [18:31<13:06,  6.09s/it]
Training Epoch: 18/100, batch 366/625 completed (loss: 0.0086:  59%|█████▊    | 183/312 [18:31<13:06,  6.09s/it]
Training Epoch: 18/100, batch 366/625 completed (loss: 0.0006:  59%|█████▉    | 184/312 [18:34<13:03,  6.12s/it]
Training Epoch: 18/100, batch 366/625 completed (loss: 0.1166:  59%|█████▉    | 184/312 [18:34<13:03,  6.12s/it]
Training Epoch: 18/100, batch 366/625 completed (loss: 0.0086:  59%|█████▉    | 184/312 [18:34<13:03,  6.12s/it]
Training Epoch: 18/100, batch 366/625 completed (loss: 0.0257:  59%|█████▉    | 184/312 [18:34<13:03,  6.12s/it]
Training Epoch: 18/100, batch 367/625 completed (loss: 0.0009:  59%|█████▉    | 184/312 [18:34<13:03,  6.12s/it]
Training Epoch: 18/100, batch 367/625 completed (loss: 0.0111:  59%|█████▉    | 184/312 [18:34<13:03,  6.12s/it]
Training Epoch: 18/100, batch 367/625 completed (loss: 0.0003:  59%|█████▉    | 184/312 [18:34<13:03,  6.12s/it]
Training Epoch: 18/100, batch 367/625 completed (loss: 0.0001:  59%|█████▉    | 184/312 [18:34<13:03,  6.12s/it]
Training Epoch: 18/100, batch 368/625 completed (loss: 0.004:  59%|█████▉    | 184/312 [18:37<13:03,  6.12s/it] 
Training Epoch: 18/100, batch 368/625 completed (loss: 0.0039:  59%|█████▉    | 184/312 [18:37<13:03,  6.12s/it]
Training Epoch: 18/100, batch 368/625 completed (loss: 0.0008:  59%|█████▉    | 184/312 [18:37<13:03,  6.12s/it]
Training Epoch: 18/100, batch 368/625 completed (loss: 0.0597:  59%|█████▉    | 184/312 [18:37<13:03,  6.12s/it]
Training Epoch: 18/100, batch 368/625 completed (loss: 0.004:  59%|█████▉    | 185/312 [18:40<13:00,  6.14s/it]
Training Epoch: 18/100, batch 368/625 completed (loss: 0.0008:  59%|█████▉    | 185/312 [18:40<13:00,  6.15s/it]
Training Epoch: 18/100, batch 368/625 completed (loss: 0.0597:  59%|█████▉    | 185/312 [18:40<13:00,  6.14s/it]
Training Epoch: 18/100, batch 368/625 completed (loss: 0.0039:  59%|█████▉    | 185/312 [18:40<13:00,  6.15s/it]
Training Epoch: 18/100, batch 369/625 completed (loss: 0.0592:  59%|█████▉    | 185/312 [18:40<13:00,  6.15s/it]
Training Epoch: 18/100, batch 369/625 completed (loss: 0.0044:  59%|█████▉    | 185/312 [18:40<13:00,  6.15s/it]
Training Epoch: 18/100, batch 369/625 completed (loss: 0.0007:  59%|█████▉    | 185/312 [18:40<13:00,  6.14s/it]
Training Epoch: 18/100, batch 369/625 completed (loss: 0.0017:  59%|█████▉    | 185/312 [18:40<13:00,  6.14s/it]
Training Epoch: 18/100, batch 370/625 completed (loss: 0.0007:  59%|█████▉    | 185/312 [18:43<13:00,  6.15s/it]
Training Epoch: 18/100, batch 370/625 completed (loss: 0.0001:  59%|█████▉    | 185/312 [18:43<13:00,  6.14s/it]
Training Epoch: 18/100, batch 370/625 completed (loss: 0.0042:  59%|█████▉    | 185/312 [18:43<13:00,  6.15s/it]
Training Epoch: 18/100, batch 370/625 completed (loss: 0.002:  59%|█████▉    | 185/312 [18:43<13:00,  6.14s/it] 
Training Epoch: 18/100, batch 370/625 completed (loss: 0.002:  60%|█████▉    | 186/312 [18:46<12:43,  6.06s/it]
Training Epoch: 18/100, batch 370/625 completed (loss: 0.0042:  60%|█████▉    | 186/312 [18:46<12:43,  6.06s/it]
Training Epoch: 18/100, batch 370/625 completed (loss: 0.0001:  60%|█████▉    | 186/312 [18:46<12:44,  6.06s/it]
Training Epoch: 18/100, batch 370/625 completed (loss: 0.0007:  60%|█████▉    | 186/312 [18:46<12:43,  6.06s/it]
Training Epoch: 18/100, batch 371/625 completed (loss: 0.0063:  60%|█████▉    | 186/312 [18:46<12:44,  6.06s/it]
Training Epoch: 18/100, batch 371/625 completed (loss: 0.0001:  60%|█████▉    | 186/312 [18:46<12:43,  6.06s/it]
Training Epoch: 18/100, batch 371/625 completed (loss: 0.0164:  60%|█████▉    | 186/312 [18:46<12:43,  6.06s/it]
Training Epoch: 18/100, batch 371/625 completed (loss: 0.0053:  60%|█████▉    | 186/312 [18:46<12:43,  6.06s/it]
Training Epoch: 18/100, batch 372/625 completed (loss: 0.016:  60%|█████▉    | 186/312 [18:49<12:43,  6.06s/it] 
Training Epoch: 18/100, batch 372/625 completed (loss: 0.0001:  60%|█████▉    | 186/312 [18:49<12:44,  6.06s/it]
Training Epoch: 18/100, batch 372/625 completed (loss: 0.0003:  60%|█████▉    | 186/312 [18:49<12:43,  6.06s/it]
Training Epoch: 18/100, batch 372/625 completed (loss: 0.0261:  60%|█████▉    | 186/312 [18:49<12:43,  6.06s/it]
Training Epoch: 18/100, batch 372/625 completed (loss: 0.0003:  60%|█████▉    | 187/312 [18:52<12:45,  6.13s/it]
Training Epoch: 18/100, batch 372/625 completed (loss: 0.016:  60%|█████▉    | 187/312 [18:52<12:45,  6.12s/it]
Training Epoch: 18/100, batch 372/625 completed (loss: 0.0261:  60%|█████▉    | 187/312 [18:52<12:45,  6.12s/it]
Training Epoch: 18/100, batch 372/625 completed (loss: 0.0001:  60%|█████▉    | 187/312 [18:52<12:45,  6.13s/it]
Training Epoch: 18/100, batch 373/625 completed (loss: 0.0076:  60%|█████▉    | 187/312 [18:52<12:45,  6.13s/it]
Training Epoch: 18/100, batch 373/625 completed (loss: 0.001:  60%|█████▉    | 187/312 [18:52<12:45,  6.13s/it] 
Training Epoch: 18/100, batch 373/625 completed (loss: 0.001:  60%|█████▉    | 187/312 [18:52<12:45,  6.12s/it]
Training Epoch: 18/100, batch 373/625 completed (loss: 0.0047:  60%|█████▉    | 187/312 [18:52<12:45,  6.12s/it]
Training Epoch: 18/100, batch 374/625 completed (loss: 0.0421:  60%|█████▉    | 187/312 [18:55<12:45,  6.12s/it]
Training Epoch: 18/100, batch 374/625 completed (loss: 0.0006:  60%|█████▉    | 187/312 [18:55<12:45,  6.13s/it]
Training Epoch: 18/100, batch 374/625 completed (loss: 0.005:  60%|█████▉    | 187/312 [18:55<12:45,  6.13s/it] 
Training Epoch: 18/100, batch 374/625 completed (loss: 0.0042:  60%|█████▉    | 187/312 [18:55<12:45,  6.12s/it]
Training Epoch: 18/100, batch 374/625 completed (loss: 0.0042:  60%|██████    | 188/312 [18:58<12:35,  6.09s/it]
Training Epoch: 18/100, batch 374/625 completed (loss: 0.005:  60%|██████    | 188/312 [18:58<12:35,  6.09s/it]
Training Epoch: 18/100, batch 374/625 completed (loss: 0.0006:  60%|██████    | 188/312 [18:58<12:35,  6.09s/it]
Training Epoch: 18/100, batch 374/625 completed (loss: 0.0421:  60%|██████    | 188/312 [18:58<12:35,  6.09s/it]
Training Epoch: 18/100, batch 375/625 completed (loss: 0.0709:  60%|██████    | 188/312 [18:58<12:35,  6.09s/it]
Training Epoch: 18/100, batch 375/625 completed (loss: 0.0082:  60%|██████    | 188/312 [18:58<12:35,  6.09s/it]
Training Epoch: 18/100, batch 375/625 completed (loss: 0.0081:  60%|██████    | 188/312 [18:58<12:35,  6.09s/it]
Training Epoch: 18/100, batch 375/625 completed (loss: 0.0003:  60%|██████    | 188/312 [18:58<12:35,  6.09s/it]
Training Epoch: 18/100, batch 376/625 completed (loss: 0.0009:  60%|██████    | 188/312 [19:01<12:35,  6.09s/it]
Training Epoch: 18/100, batch 376/625 completed (loss: 0.002:  60%|██████    | 188/312 [19:01<12:35,  6.09s/it] 
Training Epoch: 18/100, batch 376/625 completed (loss: 0.0323:  60%|██████    | 188/312 [19:01<12:35,  6.09s/it]
Training Epoch: 18/100, batch 376/625 completed (loss: 0.0105:  60%|██████    | 188/312 [19:01<12:35,  6.09s/it]
Training Epoch: 18/100, batch 376/625 completed (loss: 0.0105:  61%|██████    | 189/312 [19:04<12:25,  6.06s/it]
Training Epoch: 18/100, batch 376/625 completed (loss: 0.002:  61%|██████    | 189/312 [19:04<12:26,  6.07s/it]
Training Epoch: 18/100, batch 376/625 completed (loss: 0.0323:  61%|██████    | 189/312 [19:04<12:26,  6.07s/it]
Training Epoch: 18/100, batch 376/625 completed (loss: 0.0009:  61%|██████    | 189/312 [19:04<12:26,  6.07s/it]
Training Epoch: 18/100, batch 377/625 completed (loss: 0.0007:  61%|██████    | 189/312 [19:04<12:26,  6.07s/it]
Training Epoch: 18/100, batch 377/625 completed (loss: 0.0102:  61%|██████    | 189/312 [19:04<12:26,  6.07s/it]
Training Epoch: 18/100, batch 377/625 completed (loss: 0.0232:  61%|██████    | 189/312 [19:04<12:26,  6.07s/it]
Training Epoch: 18/100, batch 377/625 completed (loss: 0.0638:  61%|██████    | 189/312 [19:04<12:25,  6.06s/it]
Training Epoch: 18/100, batch 378/625 completed (loss: 0.0055:  61%|██████    | 189/312 [19:07<12:26,  6.07s/it]
Training Epoch: 18/100, batch 378/625 completed (loss: 0.032:  61%|██████    | 189/312 [19:07<12:26,  6.07s/it] 
Training Epoch: 18/100, batch 378/625 completed (loss: 0.0166:  61%|██████    | 189/312 [19:07<12:26,  6.07s/it]
Training Epoch: 18/100, batch 378/625 completed (loss: 0.0244:  61%|██████    | 189/312 [19:08<12:25,  6.06s/it]
Training Epoch: 18/100, batch 378/625 completed (loss: 0.0055:  61%|██████    | 190/312 [19:10<12:24,  6.10s/it]
Training Epoch: 18/100, batch 378/625 completed (loss: 0.0244:  61%|██████    | 190/312 [19:11<12:24,  6.11s/it]
Training Epoch: 18/100, batch 378/625 completed (loss: 0.0166:  61%|██████    | 190/312 [19:11<12:24,  6.11s/it]
Training Epoch: 18/100, batch 378/625 completed (loss: 0.032:  61%|██████    | 190/312 [19:11<12:24,  6.11s/it]
Training Epoch: 18/100, batch 379/625 completed (loss: 0.0022:  61%|██████    | 190/312 [19:11<12:24,  6.11s/it]
Training Epoch: 18/100, batch 379/625 completed (loss: 0.0002:  61%|██████    | 190/312 [19:11<12:24,  6.10s/it]
Training Epoch: 18/100, batch 379/625 completed (loss: 0.0004:  61%|██████    | 190/312 [19:11<12:24,  6.11s/it]
Training Epoch: 18/100, batch 379/625 completed (loss: 0.0004:  61%|██████    | 190/312 [19:11<12:24,  6.11s/it]
Training Epoch: 18/100, batch 380/625 completed (loss: 0.0291:  61%|██████    | 190/312 [19:14<12:24,  6.10s/it]
Training Epoch: 18/100, batch 380/625 completed (loss: 0.0041:  61%|██████    | 190/312 [19:14<12:24,  6.11s/it]
Training Epoch: 18/100, batch 380/625 completed (loss: 0.0014:  61%|██████    | 190/312 [19:13<12:24,  6.11s/it]
Training Epoch: 18/100, batch 380/625 completed (loss: 0.0034:  61%|██████    | 190/312 [19:14<12:24,  6.11s/it]
Training Epoch: 18/100, batch 380/625 completed (loss: 0.0291:  61%|██████    | 191/312 [19:16<12:10,  6.04s/it]
Training Epoch: 18/100, batch 380/625 completed (loss: 0.0041:  61%|██████    | 191/312 [19:16<12:10,  6.04s/it]
Training Epoch: 18/100, batch 380/625 completed (loss: 0.0034:  61%|██████    | 191/312 [19:16<12:10,  6.04s/it]
Training Epoch: 18/100, batch 380/625 completed (loss: 0.0014:  61%|██████    | 191/312 [19:16<12:10,  6.03s/it]
Training Epoch: 18/100, batch 381/625 completed (loss: 0.0196:  61%|██████    | 191/312 [19:16<12:10,  6.04s/it]
Training Epoch: 18/100, batch 381/625 completed (loss: 0.0012:  61%|██████    | 191/312 [19:16<12:10,  6.04s/it]
Training Epoch: 18/100, batch 381/625 completed (loss: 0.0005:  61%|██████    | 191/312 [19:16<12:10,  6.03s/it]
Training Epoch: 18/100, batch 381/625 completed (loss: 0.0015:  61%|██████    | 191/312 [19:16<12:10,  6.04s/it]
Training Epoch: 18/100, batch 382/625 completed (loss: 0.0003:  61%|██████    | 191/312 [19:19<12:10,  6.04s/it]
Training Epoch: 18/100, batch 382/625 completed (loss: 0.0012:  61%|██████    | 191/312 [19:19<12:10,  6.03s/it]
Training Epoch: 18/100, batch 382/625 completed (loss: 0.0111:  61%|██████    | 191/312 [19:19<12:10,  6.04s/it]
Training Epoch: 18/100, batch 382/625 completed (loss: 0.0011:  61%|██████    | 191/312 [19:19<12:10,  6.04s/it]
Training Epoch: 18/100, batch 382/625 completed (loss: 0.0111:  62%|██████▏   | 192/312 [19:23<12:15,  6.13s/it]
Training Epoch: 18/100, batch 382/625 completed (loss: 0.0003:  62%|██████▏   | 192/312 [19:23<12:15,  6.13s/it]
Training Epoch: 18/100, batch 382/625 completed (loss: 0.0011:  62%|██████▏   | 192/312 [19:23<12:15,  6.13s/it]
Training Epoch: 18/100, batch 382/625 completed (loss: 0.0012:  62%|██████▏   | 192/312 [19:23<12:15,  6.13s/it]
Training Epoch: 18/100, batch 383/625 completed (loss: 0.0004:  62%|██████▏   | 192/312 [19:23<12:15,  6.13s/it]
Training Epoch: 18/100, batch 383/625 completed (loss: 0.0016:  62%|██████▏   | 192/312 [19:23<12:15,  6.13s/it]
Training Epoch: 18/100, batch 383/625 completed (loss: 0.0132:  62%|██████▏   | 192/312 [19:23<12:15,  6.13s/it]
Training Epoch: 18/100, batch 383/625 completed (loss: 0.0023:  62%|██████▏   | 192/312 [19:23<12:15,  6.13s/it]
Training Epoch: 18/100, batch 384/625 completed (loss: 0.0006:  62%|██████▏   | 192/312 [19:26<12:15,  6.13s/it]
Training Epoch: 18/100, batch 384/625 completed (loss: 0.0069:  62%|██████▏   | 192/312 [19:26<12:15,  6.13s/it]
Training Epoch: 18/100, batch 384/625 completed (loss: 0.0444:  62%|██████▏   | 192/312 [19:26<12:15,  6.13s/it]
Training Epoch: 18/100, batch 384/625 completed (loss: 0.0038:  62%|██████▏   | 192/312 [19:26<12:15,  6.13s/it]
Training Epoch: 18/100, batch 384/625 completed (loss: 0.0444:  62%|██████▏   | 193/312 [19:29<12:04,  6.09s/it]
Training Epoch: 18/100, batch 384/625 completed (loss: 0.0038:  62%|██████▏   | 193/312 [19:29<12:05,  6.09s/it]
Training Epoch: 18/100, batch 384/625 completed (loss: 0.0069:  62%|██████▏   | 193/312 [19:29<12:05,  6.09s/it]
Training Epoch: 18/100, batch 384/625 completed (loss: 0.0006:  62%|██████▏   | 193/312 [19:29<12:05,  6.09s/it]
Training Epoch: 18/100, batch 385/625 completed (loss: 0.0024:  62%|██████▏   | 193/312 [19:29<12:04,  6.09s/it]
Training Epoch: 18/100, batch 385/625 completed (loss: 0.0058:  62%|██████▏   | 193/312 [19:29<12:05,  6.09s/it]
Training Epoch: 18/100, batch 385/625 completed (loss: 0.0192:  62%|██████▏   | 193/312 [19:29<12:05,  6.09s/it]
Training Epoch: 18/100, batch 385/625 completed (loss: 0.3688:  62%|██████▏   | 193/312 [19:29<12:05,  6.09s/it]
Training Epoch: 18/100, batch 386/625 completed (loss: 0.0099:  62%|██████▏   | 193/312 [19:32<12:05,  6.09s/it]
Training Epoch: 18/100, batch 386/625 completed (loss: 0.0093:  62%|██████▏   | 193/312 [19:32<12:04,  6.09s/it]
Training Epoch: 18/100, batch 386/625 completed (loss: 0.0062:  62%|██████▏   | 193/312 [19:32<12:05,  6.09s/it]
Training Epoch: 18/100, batch 386/625 completed (loss: 0.035:  62%|██████▏   | 193/312 [19:32<12:05,  6.09s/it] 
Training Epoch: 18/100, batch 386/625 completed (loss: 0.0062:  62%|██████▏   | 194/312 [19:35<11:50,  6.02s/it]
Training Epoch: 18/100, batch 386/625 completed (loss: 0.0093:  62%|██████▏   | 194/312 [19:35<11:50,  6.02s/it]
Training Epoch: 18/100, batch 386/625 completed (loss: 0.0099:  62%|██████▏   | 194/312 [19:35<11:50,  6.02s/it]
Training Epoch: 18/100, batch 386/625 completed (loss: 0.035:  62%|██████▏   | 194/312 [19:35<11:50,  6.02s/it]
Training Epoch: 18/100, batch 387/625 completed (loss: 0.0014:  62%|██████▏   | 194/312 [19:35<11:50,  6.02s/it]
Training Epoch: 18/100, batch 387/625 completed (loss: 0.0007:  62%|██████▏   | 194/312 [19:35<11:50,  6.02s/it]
Training Epoch: 18/100, batch 387/625 completed (loss: 0.0056:  62%|██████▏   | 194/312 [19:35<11:50,  6.02s/it]
Training Epoch: 18/100, batch 387/625 completed (loss: 0.0005:  62%|██████▏   | 194/312 [19:35<11:50,  6.02s/it]
Training Epoch: 18/100, batch 388/625 completed (loss: 0.0023:  62%|██████▏   | 194/312 [19:38<11:50,  6.02s/it]
Training Epoch: 18/100, batch 388/625 completed (loss: 0.0159:  62%|██████▏   | 194/312 [19:38<11:50,  6.02s/it]
Training Epoch: 18/100, batch 388/625 completed (loss: 0.0015:  62%|██████▏   | 194/312 [19:38<11:50,  6.02s/it]
Training Epoch: 18/100, batch 388/625 completed (loss: 0.0017:  62%|██████▏   | 194/312 [19:38<11:50,  6.02s/it]
Training Epoch: 18/100, batch 388/625 completed (loss: 0.0023:  62%|██████▎   | 195/312 [19:41<11:56,  6.13s/it]
Training Epoch: 18/100, batch 388/625 completed (loss: 0.0017:  62%|██████▎   | 195/312 [19:41<11:56,  6.13s/it]
Training Epoch: 18/100, batch 388/625 completed (loss: 0.0015:  62%|██████▎   | 195/312 [19:41<11:57,  6.13s/it]
Training Epoch: 18/100, batch 388/625 completed (loss: 0.0159:  62%|██████▎   | 195/312 [19:41<11:57,  6.13s/it]
Training Epoch: 18/100, batch 389/625 completed (loss: 0.0169:  62%|██████▎   | 195/312 [19:41<11:57,  6.13s/it]
Training Epoch: 18/100, batch 389/625 completed (loss: 0.0016:  62%|██████▎   | 195/312 [19:41<11:57,  6.13s/it]
Training Epoch: 18/100, batch 389/625 completed (loss: 0.0025:  62%|██████▎   | 195/312 [19:41<11:56,  6.13s/it]
Training Epoch: 18/100, batch 389/625 completed (loss: 0.0092:  62%|██████▎   | 195/312 [19:41<11:56,  6.13s/it]
Training Epoch: 18/100, batch 390/625 completed (loss: 0.008:  62%|██████▎   | 195/312 [19:44<11:56,  6.13s/it] 
Training Epoch: 18/100, batch 390/625 completed (loss: 0.0298:  62%|██████▎   | 195/312 [19:44<11:57,  6.13s/it]
Training Epoch: 18/100, batch 390/625 completed (loss: 0.0604:  62%|██████▎   | 195/312 [19:44<11:57,  6.13s/it]
Training Epoch: 18/100, batch 390/625 completed (loss: 0.0348:  62%|██████▎   | 195/312 [19:44<11:56,  6.13s/it]
Training Epoch: 18/100, batch 390/625 completed (loss: 0.008:  63%|██████▎   | 196/312 [19:47<11:42,  6.06s/it]
Training Epoch: 18/100, batch 390/625 completed (loss: 0.0348:  63%|██████▎   | 196/312 [19:47<11:42,  6.06s/it]
Training Epoch: 18/100, batch 390/625 completed (loss: 0.0298:  63%|██████▎   | 196/312 [19:47<11:43,  6.06s/it]
Training Epoch: 18/100, batch 390/625 completed (loss: 0.0604:  63%|██████▎   | 196/312 [19:47<11:42,  6.06s/it]
Training Epoch: 18/100, batch 391/625 completed (loss: 0.005:  63%|██████▎   | 196/312 [19:47<11:42,  6.06s/it]
Training Epoch: 18/100, batch 391/625 completed (loss: 0.0003:  63%|██████▎   | 196/312 [19:47<11:42,  6.06s/it]
Training Epoch: 18/100, batch 391/625 completed (loss: 0.0035:  63%|██████▎   | 196/312 [19:47<11:43,  6.06s/it]
Training Epoch: 18/100, batch 391/625 completed (loss: 0.0586:  63%|██████▎   | 196/312 [19:47<11:42,  6.06s/it]
Training Epoch: 18/100, batch 392/625 completed (loss: 0.0142:  63%|██████▎   | 196/312 [19:50<11:42,  6.06s/it]
Training Epoch: 18/100, batch 392/625 completed (loss: 0.0067:  63%|██████▎   | 196/312 [19:50<11:42,  6.06s/it]
Training Epoch: 18/100, batch 392/625 completed (loss: 0.0003:  63%|██████▎   | 196/312 [19:50<11:43,  6.06s/it]
Training Epoch: 18/100, batch 392/625 completed (loss: 0.0025:  63%|██████▎   | 196/312 [19:50<11:42,  6.06s/it]
Training Epoch: 18/100, batch 392/625 completed (loss: 0.0003:  63%|██████▎   | 197/312 [19:53<11:40,  6.09s/it]
Training Epoch: 18/100, batch 392/625 completed (loss: 0.0142:  63%|██████▎   | 197/312 [19:53<11:40,  6.09s/it]
Training Epoch: 18/100, batch 392/625 completed (loss: 0.0067:  63%|██████▎   | 197/312 [19:53<11:40,  6.09s/it]
Training Epoch: 18/100, batch 392/625 completed (loss: 0.0025:  63%|██████▎   | 197/312 [19:53<11:40,  6.09s/it]
Training Epoch: 18/100, batch 393/625 completed (loss: 0.0015:  63%|██████▎   | 197/312 [19:53<11:40,  6.09s/it]
Training Epoch: 18/100, batch 393/625 completed (loss: 0.0056:  63%|██████▎   | 197/312 [19:53<11:40,  6.09s/it]
Training Epoch: 18/100, batch 393/625 completed (loss: 0.3532:  63%|██████▎   | 197/312 [19:53<11:40,  6.09s/it]
Training Epoch: 18/100, batch 393/625 completed (loss: 0.0064:  63%|██████▎   | 197/312 [19:53<11:40,  6.09s/it]
Training Epoch: 18/100, batch 394/625 completed (loss: 0.0113:  63%|██████▎   | 197/312 [19:56<11:40,  6.09s/it]
Training Epoch: 18/100, batch 394/625 completed (loss: 0.0025:  63%|██████▎   | 197/312 [19:56<11:40,  6.09s/it]
Training Epoch: 18/100, batch 394/625 completed (loss: 0.0202:  63%|██████▎   | 197/312 [19:56<11:40,  6.09s/it]
Training Epoch: 18/100, batch 394/625 completed (loss: 0.0008:  63%|██████▎   | 197/312 [19:56<11:40,  6.09s/it]
Training Epoch: 18/100, batch 394/625 completed (loss: 0.0025:  63%|██████▎   | 198/312 [19:59<11:36,  6.11s/it]
Training Epoch: 18/100, batch 394/625 completed (loss: 0.0008:  63%|██████▎   | 198/312 [19:59<11:36,  6.11s/it]
Training Epoch: 18/100, batch 394/625 completed (loss: 0.0202:  63%|██████▎   | 198/312 [19:59<11:36,  6.11s/it]
Training Epoch: 18/100, batch 394/625 completed (loss: 0.0113:  63%|██████▎   | 198/312 [19:59<11:36,  6.11s/it]
Training Epoch: 18/100, batch 395/625 completed (loss: 0.0027:  63%|██████▎   | 198/312 [19:59<11:36,  6.11s/it]
Training Epoch: 18/100, batch 395/625 completed (loss: 0.025:  63%|██████▎   | 198/312 [19:59<11:36,  6.11s/it] 
Training Epoch: 18/100, batch 395/625 completed (loss: 0.0031:  63%|██████▎   | 198/312 [19:59<11:36,  6.11s/it]
Training Epoch: 18/100, batch 395/625 completed (loss: 0.0025:  63%|██████▎   | 198/312 [19:59<11:36,  6.11s/it]
Training Epoch: 18/100, batch 396/625 completed (loss: 0.0006:  63%|██████▎   | 198/312 [20:02<11:36,  6.11s/it]
Training Epoch: 18/100, batch 396/625 completed (loss: 0.0052:  63%|██████▎   | 198/312 [20:02<11:36,  6.11s/it]
Training Epoch: 18/100, batch 396/625 completed (loss: 0.0032:  63%|██████▎   | 198/312 [20:02<11:36,  6.11s/it]
Training Epoch: 18/100, batch 396/625 completed (loss: 0.0063:  63%|██████▎   | 198/312 [20:02<11:36,  6.11s/it]
Training Epoch: 18/100, batch 396/625 completed (loss: 0.0063:  64%|██████▍   | 199/312 [20:05<11:22,  6.04s/it]
Training Epoch: 18/100, batch 396/625 completed (loss: 0.0006:  64%|██████▍   | 199/312 [20:05<11:23,  6.04s/it]
Training Epoch: 18/100, batch 396/625 completed (loss: 0.0052:  64%|██████▍   | 199/312 [20:05<11:22,  6.04s/it]
Training Epoch: 18/100, batch 396/625 completed (loss: 0.0032:  64%|██████▍   | 199/312 [20:05<11:23,  6.04s/it]
Training Epoch: 18/100, batch 397/625 completed (loss: 0.0076:  64%|██████▍   | 199/312 [20:05<11:23,  6.04s/it]
Training Epoch: 18/100, batch 397/625 completed (loss: 0.0154:  64%|██████▍   | 199/312 [20:05<11:22,  6.04s/it]
Training Epoch: 18/100, batch 397/625 completed (loss: 0.0732:  64%|██████▍   | 199/312 [20:05<11:23,  6.04s/it]
Training Epoch: 18/100, batch 397/625 completed (loss: 0.002:  64%|██████▍   | 199/312 [20:05<11:22,  6.04s/it] 
Training Epoch: 18/100, batch 398/625 completed (loss: 0.0098:  64%|██████▍   | 199/312 [20:08<11:23,  6.04s/it]
Training Epoch: 18/100, batch 398/625 completed (loss: 0.0074:  64%|██████▍   | 199/312 [20:08<11:22,  6.04s/it]
Training Epoch: 18/100, batch 398/625 completed (loss: 0.0001:  64%|██████▍   | 199/312 [20:08<11:23,  6.04s/it]
Training Epoch: 18/100, batch 398/625 completed (loss: 0.0029:  64%|██████▍   | 199/312 [20:08<11:22,  6.04s/it]
Training Epoch: 18/100, batch 398/625 completed (loss: 0.0029:  64%|██████▍   | 200/312 [20:11<11:20,  6.08s/it]
Training Epoch: 18/100, batch 398/625 completed (loss: 0.0098:  64%|██████▍   | 200/312 [20:11<11:20,  6.08s/it]
Training Epoch: 18/100, batch 398/625 completed (loss: 0.0074:  64%|██████▍   | 200/312 [20:11<11:21,  6.08s/it]
Training Epoch: 18/100, batch 398/625 completed (loss: 0.0001:  64%|██████▍   | 200/312 [20:11<11:21,  6.08s/it]
Training Epoch: 18/100, batch 399/625 completed (loss: 0.0003:  64%|██████▍   | 200/312 [20:11<11:21,  6.08s/it]
Training Epoch: 18/100, batch 399/625 completed (loss: 0.0015:  64%|██████▍   | 200/312 [20:11<11:21,  6.08s/it]
Training Epoch: 18/100, batch 399/625 completed (loss: 0.014:  64%|██████▍   | 200/312 [20:11<11:20,  6.08s/it] 
Training Epoch: 18/100, batch 399/625 completed (loss: 0.0001:  64%|██████▍   | 200/312 [20:11<11:20,  6.08s/it]
Training Epoch: 18/100, batch 400/625 completed (loss: 0.0238:  64%|██████▍   | 200/312 [20:14<11:21,  6.08s/it]
Training Epoch: 18/100, batch 400/625 completed (loss: 0.0015:  64%|██████▍   | 200/312 [20:14<11:20,  6.08s/it]
Training Epoch: 18/100, batch 400/625 completed (loss: 0.0039:  64%|██████▍   | 200/312 [20:14<11:21,  6.08s/it]
Training Epoch: 18/100, batch 400/625 completed (loss: 0.0057:  64%|██████▍   | 200/312 [20:14<11:20,  6.08s/it]
Training Epoch: 18/100, batch 400/625 completed (loss: 0.0057:  64%|██████▍   | 201/312 [20:17<11:14,  6.08s/it]
Training Epoch: 18/100, batch 400/625 completed (loss: 0.0039:  64%|██████▍   | 201/312 [20:17<11:14,  6.07s/it]
Training Epoch: 18/100, batch 400/625 completed (loss: 0.0015:  64%|██████▍   | 201/312 [20:17<11:14,  6.08s/it]
Training Epoch: 18/100, batch 400/625 completed (loss: 0.0238:  64%|██████▍   | 201/312 [20:17<11:14,  6.07s/it]
Training Epoch: 18/100, batch 401/625 completed (loss: 0.0057:  64%|██████▍   | 201/312 [20:17<11:14,  6.08s/it]
Training Epoch: 18/100, batch 401/625 completed (loss: 0.0002:  64%|██████▍   | 201/312 [20:17<11:14,  6.07s/it]
Training Epoch: 18/100, batch 401/625 completed (loss: 0.0008:  64%|██████▍   | 201/312 [20:17<11:14,  6.07s/it]
Training Epoch: 18/100, batch 401/625 completed (loss: 0.0017:  64%|██████▍   | 201/312 [20:17<11:14,  6.08s/it]
Training Epoch: 18/100, batch 402/625 completed (loss: 0.0015:  64%|██████▍   | 201/312 [20:20<11:14,  6.07s/it]
Training Epoch: 18/100, batch 402/625 completed (loss: 0.001:  64%|██████▍   | 201/312 [20:20<11:14,  6.08s/it] 
Training Epoch: 18/100, batch 402/625 completed (loss: 0.3877:  64%|██████▍   | 201/312 [20:20<11:14,  6.07s/it]
Training Epoch: 18/100, batch 402/625 completed (loss: 0.002:  64%|██████▍   | 201/312 [20:20<11:14,  6.08s/it] 
Training Epoch: 18/100, batch 402/625 completed (loss: 0.0015:  65%|██████▍   | 202/312 [20:23<11:05,  6.05s/it]
Training Epoch: 18/100, batch 402/625 completed (loss: 0.002:  65%|██████▍   | 202/312 [20:23<11:05,  6.05s/it]
Training Epoch: 18/100, batch 402/625 completed (loss: 0.001:  65%|██████▍   | 202/312 [20:23<11:06,  6.06s/it]
Training Epoch: 18/100, batch 402/625 completed (loss: 0.3877:  65%|██████▍   | 202/312 [20:23<11:05,  6.05s/it]
Training Epoch: 18/100, batch 403/625 completed (loss: 0.0457:  65%|██████▍   | 202/312 [20:23<11:05,  6.05s/it]
Training Epoch: 18/100, batch 403/625 completed (loss: 0.075:  65%|██████▍   | 202/312 [20:23<11:06,  6.06s/it]
Training Epoch: 18/100, batch 403/625 completed (loss: 0.0009:  65%|██████▍   | 202/312 [20:23<11:05,  6.05s/it]
Training Epoch: 18/100, batch 403/625 completed (loss: 0.0124:  65%|██████▍   | 202/312 [20:23<11:05,  6.05s/it]
Training Epoch: 18/100, batch 404/625 completed (loss: 0.0005:  65%|██████▍   | 202/312 [20:26<11:05,  6.05s/it]
Training Epoch: 18/100, batch 404/625 completed (loss: 0.0072:  65%|██████▍   | 202/312 [20:26<11:05,  6.05s/it]
Training Epoch: 18/100, batch 404/625 completed (loss: 0.0012:  65%|██████▍   | 202/312 [20:26<11:06,  6.06s/it]
Training Epoch: 18/100, batch 404/625 completed (loss: 0.0043:  65%|██████▍   | 202/312 [20:26<11:05,  6.05s/it]
Training Epoch: 18/100, batch 404/625 completed (loss: 0.0072:  65%|██████▌   | 203/312 [20:30<11:07,  6.13s/it]
Training Epoch: 18/100, batch 404/625 completed (loss: 0.0043:  65%|██████▌   | 203/312 [20:30<11:07,  6.13s/it]
Training Epoch: 18/100, batch 404/625 completed (loss: 0.0012:  65%|██████▌   | 203/312 [20:30<11:07,  6.13s/it]
Training Epoch: 18/100, batch 404/625 completed (loss: 0.0005:  65%|██████▌   | 203/312 [20:30<11:07,  6.13s/it]
Training Epoch: 18/100, batch 405/625 completed (loss: 0.0004:  65%|██████▌   | 203/312 [20:30<11:07,  6.13s/it]
Training Epoch: 18/100, batch 405/625 completed (loss: 0.0011:  65%|██████▌   | 203/312 [20:30<11:07,  6.13s/it]
Training Epoch: 18/100, batch 405/625 completed (loss: 0.0014:  65%|██████▌   | 203/312 [20:30<11:07,  6.13s/it]
Training Epoch: 18/100, batch 405/625 completed (loss: 0.0011:  65%|██████▌   | 203/312 [20:30<11:07,  6.13s/it]
Training Epoch: 18/100, batch 406/625 completed (loss: 0.0017:  65%|██████▌   | 203/312 [20:33<11:07,  6.13s/it]
Training Epoch: 18/100, batch 406/625 completed (loss: 0.0038:  65%|██████▌   | 203/312 [20:33<11:07,  6.13s/it]
Training Epoch: 18/100, batch 406/625 completed (loss: 0.0016:  65%|██████▌   | 203/312 [20:33<11:07,  6.13s/it]
Training Epoch: 18/100, batch 406/625 completed (loss: 0.0049:  65%|██████▌   | 203/312 [20:33<11:07,  6.13s/it]
Training Epoch: 18/100, batch 406/625 completed (loss: 0.0049:  65%|██████▌   | 204/312 [20:36<10:53,  6.05s/it]
Training Epoch: 18/100, batch 406/625 completed (loss: 0.0016:  65%|██████▌   | 204/312 [20:35<10:53,  6.05s/it]
Training Epoch: 18/100, batch 406/625 completed (loss: 0.0038:  65%|██████▌   | 204/312 [20:35<10:53,  6.05s/it]
Training Epoch: 18/100, batch 406/625 completed (loss: 0.0017:  65%|██████▌   | 204/312 [20:35<10:53,  6.05s/it]
Training Epoch: 18/100, batch 407/625 completed (loss: 0.0083:  65%|██████▌   | 204/312 [20:36<10:53,  6.05s/it]
Training Epoch: 18/100, batch 407/625 completed (loss: 0.0072:  65%|██████▌   | 204/312 [20:36<10:53,  6.05s/it]
Training Epoch: 18/100, batch 407/625 completed (loss: 0.0005:  65%|██████▌   | 204/312 [20:36<10:53,  6.05s/it]
Training Epoch: 18/100, batch 407/625 completed (loss: 0.0004:  65%|██████▌   | 204/312 [20:36<10:53,  6.05s/it]
Training Epoch: 18/100, batch 408/625 completed (loss: 0.0001:  65%|██████▌   | 204/312 [20:39<10:53,  6.05s/it]
Training Epoch: 18/100, batch 408/625 completed (loss: 0.0006:  65%|██████▌   | 204/312 [20:39<10:53,  6.05s/it]
Training Epoch: 18/100, batch 408/625 completed (loss: 0.002:  65%|██████▌   | 204/312 [20:39<10:53,  6.05s/it] 
Training Epoch: 18/100, batch 408/625 completed (loss: 0.0009:  65%|██████▌   | 204/312 [20:39<10:53,  6.05s/it]
Training Epoch: 18/100, batch 408/625 completed (loss: 0.0009:  66%|██████▌   | 205/312 [20:42<10:51,  6.09s/it]
Training Epoch: 18/100, batch 408/625 completed (loss: 0.0006:  66%|██████▌   | 205/312 [20:42<10:52,  6.09s/it]
Training Epoch: 18/100, batch 408/625 completed (loss: 0.0001:  66%|██████▌   | 205/312 [20:42<10:52,  6.09s/it]
Training Epoch: 18/100, batch 408/625 completed (loss: 0.002:  66%|██████▌   | 205/312 [20:42<10:52,  6.10s/it]
Training Epoch: 18/100, batch 409/625 completed (loss: 0.0014:  66%|██████▌   | 205/312 [20:42<10:52,  6.10s/it]
Training Epoch: 18/100, batch 409/625 completed (loss: 0.0025:  66%|██████▌   | 205/312 [20:42<10:52,  6.09s/it]
Training Epoch: 18/100, batch 409/625 completed (loss: 0.0011:  66%|██████▌   | 205/312 [20:42<10:52,  6.09s/it]
Training Epoch: 18/100, batch 409/625 completed (loss: 0.0051:  66%|██████▌   | 205/312 [20:42<10:51,  6.09s/it]
Training Epoch: 18/100, batch 410/625 completed (loss: 0.0003:  66%|██████▌   | 205/312 [20:45<10:52,  6.09s/it]
Training Epoch: 18/100, batch 410/625 completed (loss: 0.0203:  66%|██████▌   | 205/312 [20:45<10:52,  6.10s/it]
Training Epoch: 18/100, batch 410/625 completed (loss: 0.0002:  66%|██████▌   | 205/312 [20:45<10:52,  6.09s/it]
Training Epoch: 18/100, batch 410/625 completed (loss: 0.0032:  66%|██████▌   | 205/312 [20:45<10:51,  6.09s/it]
Training Epoch: 18/100, batch 410/625 completed (loss: 0.0003:  66%|██████▌   | 206/312 [20:48<10:49,  6.13s/it]
Training Epoch: 18/100, batch 410/625 completed (loss: 0.0203:  66%|██████▌   | 206/312 [20:48<10:49,  6.13s/it]
Training Epoch: 18/100, batch 410/625 completed (loss: 0.0032:  66%|██████▌   | 206/312 [20:48<10:50,  6.13s/it]
Training Epoch: 18/100, batch 410/625 completed (loss: 0.0002:  66%|██████▌   | 206/312 [20:48<10:50,  6.14s/it]
Training Epoch: 18/100, batch 411/625 completed (loss: 0.0011:  66%|██████▌   | 206/312 [20:48<10:50,  6.14s/it]
Training Epoch: 18/100, batch 411/625 completed (loss: 0.0024:  66%|██████▌   | 206/312 [20:48<10:49,  6.13s/it]
Training Epoch: 18/100, batch 411/625 completed (loss: 0.0132:  66%|██████▌   | 206/312 [20:48<10:49,  6.13s/it]
Training Epoch: 18/100, batch 411/625 completed (loss: 0.0041:  66%|██████▌   | 206/312 [20:48<10:50,  6.13s/it]
Training Epoch: 18/100, batch 412/625 completed (loss: 0.0429:  66%|██████▌   | 206/312 [20:51<10:49,  6.13s/it]
Training Epoch: 18/100, batch 412/625 completed (loss: 0.0015:  66%|██████▌   | 206/312 [20:51<10:50,  6.14s/it]
Training Epoch: 18/100, batch 412/625 completed (loss: 0.0067:  66%|██████▌   | 206/312 [20:51<10:49,  6.13s/it]
Training Epoch: 18/100, batch 412/625 completed (loss: 0.0007:  66%|██████▌   | 206/312 [20:51<10:50,  6.13s/it]
Training Epoch: 18/100, batch 412/625 completed (loss: 0.0067:  66%|██████▋   | 207/312 [20:54<10:36,  6.06s/it]
Training Epoch: 18/100, batch 412/625 completed (loss: 0.0007:  66%|██████▋   | 207/312 [20:54<10:36,  6.06s/it]
Training Epoch: 18/100, batch 412/625 completed (loss: 0.0429:  66%|██████▋   | 207/312 [20:54<10:36,  6.06s/it]
Training Epoch: 18/100, batch 412/625 completed (loss: 0.0015:  66%|██████▋   | 207/312 [20:54<10:36,  6.06s/it]
Training Epoch: 18/100, batch 413/625 completed (loss: 0.0007:  66%|██████▋   | 207/312 [20:54<10:36,  6.06s/it]
Training Epoch: 18/100, batch 413/625 completed (loss: 0.0015:  66%|██████▋   | 207/312 [20:54<10:36,  6.06s/it]
Training Epoch: 18/100, batch 413/625 completed (loss: 0.0307:  66%|██████▋   | 207/312 [20:54<10:36,  6.06s/it]
Training Epoch: 18/100, batch 413/625 completed (loss: 0.0102:  66%|██████▋   | 207/312 [20:54<10:36,  6.06s/it]
Training Epoch: 18/100, batch 414/625 completed (loss: 0.0002:  66%|██████▋   | 207/312 [20:57<10:36,  6.06s/it]
Training Epoch: 18/100, batch 414/625 completed (loss: 0.0046:  66%|██████▋   | 207/312 [20:57<10:36,  6.06s/it]
Training Epoch: 18/100, batch 414/625 completed (loss: 0.0324:  66%|██████▋   | 207/312 [20:57<10:36,  6.06s/it]
Training Epoch: 18/100, batch 414/625 completed (loss: 0.0066:  66%|██████▋   | 207/312 [20:57<10:36,  6.06s/it]
Training Epoch: 18/100, batch 414/625 completed (loss: 0.0324:  67%|██████▋   | 208/312 [21:00<10:42,  6.18s/it]
Training Epoch: 18/100, batch 414/625 completed (loss: 0.0046:  67%|██████▋   | 208/312 [21:00<10:42,  6.18s/it]
Training Epoch: 18/100, batch 414/625 completed (loss: 0.0066:  67%|██████▋   | 208/312 [21:00<10:42,  6.18s/it]
Training Epoch: 18/100, batch 414/625 completed (loss: 0.0002:  67%|██████▋   | 208/312 [21:00<10:42,  6.18s/it]
Training Epoch: 18/100, batch 415/625 completed (loss: 0.0004:  67%|██████▋   | 208/312 [21:00<10:42,  6.18s/it]
Training Epoch: 18/100, batch 415/625 completed (loss: 0.0013:  67%|██████▋   | 208/312 [21:00<10:42,  6.18s/it]
Training Epoch: 18/100, batch 415/625 completed (loss: 0.0002:  67%|██████▋   | 208/312 [21:00<10:42,  6.18s/it]
Training Epoch: 18/100, batch 415/625 completed (loss: 0.0107:  67%|██████▋   | 208/312 [21:00<10:42,  6.18s/it]
Training Epoch: 18/100, batch 416/625 completed (loss: 0.0002:  67%|██████▋   | 208/312 [21:03<10:42,  6.18s/it]
Training Epoch: 18/100, batch 416/625 completed (loss: 0.0089:  67%|██████▋   | 208/312 [21:03<10:42,  6.18s/it]
Training Epoch: 18/100, batch 416/625 completed (loss: 0.0011:  67%|██████▋   | 208/312 [21:03<10:42,  6.18s/it]
Training Epoch: 18/100, batch 416/625 completed (loss: 0.0014:  67%|██████▋   | 208/312 [21:03<10:42,  6.18s/it]
Training Epoch: 18/100, batch 416/625 completed (loss: 0.0002:  67%|██████▋   | 209/312 [21:06<10:27,  6.09s/it]
Training Epoch: 18/100, batch 416/625 completed (loss: 0.0089:  67%|██████▋   | 209/312 [21:06<10:27,  6.09s/it]
Training Epoch: 18/100, batch 416/625 completed (loss: 0.0011:  67%|██████▋   | 209/312 [21:06<10:27,  6.09s/it]
Training Epoch: 18/100, batch 416/625 completed (loss: 0.0014:  67%|██████▋   | 209/312 [21:06<10:27,  6.09s/it]
Training Epoch: 18/100, batch 417/625 completed (loss: 0.0033:  67%|██████▋   | 209/312 [21:06<10:27,  6.09s/it]
Training Epoch: 18/100, batch 417/625 completed (loss: 0.0023:  67%|██████▋   | 209/312 [21:06<10:27,  6.09s/it]
Training Epoch: 18/100, batch 417/625 completed (loss: 0.0143:  67%|██████▋   | 209/312 [21:06<10:27,  6.09s/it]
Training Epoch: 18/100, batch 417/625 completed (loss: 0.0185:  67%|██████▋   | 209/312 [21:06<10:27,  6.09s/it]
Training Epoch: 18/100, batch 418/625 completed (loss: 0.0002:  67%|██████▋   | 209/312 [21:09<10:27,  6.09s/it]
Training Epoch: 18/100, batch 418/625 completed (loss: 0.0346:  67%|██████▋   | 209/312 [21:09<10:27,  6.09s/it]
Training Epoch: 18/100, batch 418/625 completed (loss: 0.0067:  67%|██████▋   | 209/312 [21:09<10:27,  6.09s/it]
Training Epoch: 18/100, batch 418/625 completed (loss: 0.0011:  67%|██████▋   | 209/312 [21:09<10:27,  6.09s/it]
Training Epoch: 18/100, batch 418/625 completed (loss: 0.0067:  67%|██████▋   | 210/312 [21:12<10:23,  6.11s/it]
Training Epoch: 18/100, batch 418/625 completed (loss: 0.0346:  67%|██████▋   | 210/312 [21:12<10:23,  6.11s/it]
Training Epoch: 18/100, batch 418/625 completed (loss: 0.0011:  67%|██████▋   | 210/312 [21:12<10:23,  6.12s/it]
Training Epoch: 18/100, batch 418/625 completed (loss: 0.0002:  67%|██████▋   | 210/312 [21:12<10:24,  6.12s/it]
Training Epoch: 18/100, batch 419/625 completed (loss: 0.0028:  67%|██████▋   | 210/312 [21:12<10:23,  6.11s/it]
Training Epoch: 18/100, batch 419/625 completed (loss: 0.0021:  67%|██████▋   | 210/312 [21:12<10:23,  6.11s/it]
Training Epoch: 18/100, batch 419/625 completed (loss: 0.1307:  67%|██████▋   | 210/312 [21:12<10:24,  6.12s/it]
Training Epoch: 18/100, batch 419/625 completed (loss: 0.0527:  67%|██████▋   | 210/312 [21:12<10:23,  6.12s/it]
Training Epoch: 18/100, batch 420/625 completed (loss: 0.0035:  67%|██████▋   | 210/312 [21:15<10:23,  6.11s/it]
Training Epoch: 18/100, batch 420/625 completed (loss: 0.001:  67%|██████▋   | 210/312 [21:15<10:23,  6.11s/it] 
Training Epoch: 18/100, batch 420/625 completed (loss: 0.0027:  67%|██████▋   | 210/312 [21:15<10:24,  6.12s/it]
Training Epoch: 18/100, batch 420/625 completed (loss: 0.0019:  67%|██████▋   | 210/312 [21:15<10:23,  6.12s/it]
Training Epoch: 18/100, batch 420/625 completed (loss: 0.001:  68%|██████▊   | 211/312 [21:18<10:14,  6.09s/it]
Training Epoch: 18/100, batch 420/625 completed (loss: 0.0027:  68%|██████▊   | 211/312 [21:18<10:15,  6.09s/it]
Training Epoch: 18/100, batch 420/625 completed (loss: 0.0019:  68%|██████▊   | 211/312 [21:18<10:15,  6.09s/it]
Training Epoch: 18/100, batch 420/625 completed (loss: 0.0035:  68%|██████▊   | 211/312 [21:18<10:15,  6.09s/it]
Training Epoch: 18/100, batch 421/625 completed (loss: 0.0012:  68%|██████▊   | 211/312 [21:18<10:15,  6.09s/it]
Training Epoch: 18/100, batch 421/625 completed (loss: 0.0032:  68%|██████▊   | 211/312 [21:18<10:15,  6.09s/it]
Training Epoch: 18/100, batch 421/625 completed (loss: 0.0008:  68%|██████▊   | 211/312 [21:18<10:14,  6.09s/it]
Training Epoch: 18/100, batch 421/625 completed (loss: 0.0028:  68%|██████▊   | 211/312 [21:18<10:15,  6.09s/it]
Training Epoch: 18/100, batch 422/625 completed (loss: 0.0515:  68%|██████▊   | 211/312 [21:21<10:15,  6.09s/it]
Training Epoch: 18/100, batch 422/625 completed (loss: 0.0025:  68%|██████▊   | 211/312 [21:21<10:14,  6.09s/it]
Training Epoch: 18/100, batch 422/625 completed (loss: 0.001:  68%|██████▊   | 211/312 [21:21<10:15,  6.09s/it] 
Training Epoch: 18/100, batch 422/625 completed (loss: 0.0001:  68%|██████▊   | 211/312 [21:21<10:15,  6.09s/it]
Training Epoch: 18/100, batch 422/625 completed (loss: 0.001:  68%|██████▊   | 212/312 [21:24<10:01,  6.01s/it]
Training Epoch: 18/100, batch 422/625 completed (loss: 0.0515:  68%|██████▊   | 212/312 [21:24<10:01,  6.02s/it]
Training Epoch: 18/100, batch 422/625 completed (loss: 0.0025:  68%|██████▊   | 212/312 [21:24<10:01,  6.02s/it]
Training Epoch: 18/100, batch 422/625 completed (loss: 0.0001:  68%|██████▊   | 212/312 [21:24<10:01,  6.02s/it]
Training Epoch: 18/100, batch 423/625 completed (loss: 0.0018:  68%|██████▊   | 212/312 [21:24<10:01,  6.01s/it]
Training Epoch: 18/100, batch 423/625 completed (loss: 0.0006:  68%|██████▊   | 212/312 [21:24<10:01,  6.02s/it]
Training Epoch: 18/100, batch 423/625 completed (loss: 0.0033:  68%|██████▊   | 212/312 [21:24<10:01,  6.02s/it]
Training Epoch: 18/100, batch 423/625 completed (loss: 0.0001:  68%|██████▊   | 212/312 [21:24<10:01,  6.02s/it]
Training Epoch: 18/100, batch 424/625 completed (loss: 0.0488:  68%|██████▊   | 212/312 [21:27<10:01,  6.02s/it]
Training Epoch: 18/100, batch 424/625 completed (loss: 0.0018:  68%|██████▊   | 212/312 [21:27<10:01,  6.01s/it]
Training Epoch: 18/100, batch 424/625 completed (loss: 0.0212:  68%|██████▊   | 212/312 [21:27<10:01,  6.02s/it]
Training Epoch: 18/100, batch 424/625 completed (loss: 0.001:  68%|██████▊   | 212/312 [21:27<10:01,  6.02s/it] 
Training Epoch: 18/100, batch 424/625 completed (loss: 0.0488:  68%|██████▊   | 213/312 [21:30<10:00,  6.06s/it]
Training Epoch: 18/100, batch 424/625 completed (loss: 0.0018:  68%|██████▊   | 213/312 [21:30<10:00,  6.06s/it]
Training Epoch: 18/100, batch 424/625 completed (loss: 0.001:  68%|██████▊   | 213/312 [21:30<10:00,  6.06s/it]
Training Epoch: 18/100, batch 424/625 completed (loss: 0.0212:  68%|██████▊   | 213/312 [21:30<10:00,  6.06s/it]
Training Epoch: 18/100, batch 425/625 completed (loss: 0.0054:  68%|██████▊   | 213/312 [21:30<10:00,  6.06s/it]
Training Epoch: 18/100, batch 425/625 completed (loss: 0.0328:  68%|██████▊   | 213/312 [21:30<10:00,  6.06s/it]
Training Epoch: 18/100, batch 425/625 completed (loss: 0.0092:  68%|██████▊   | 213/312 [21:30<10:00,  6.06s/it]
Training Epoch: 18/100, batch 425/625 completed (loss: 0.0102:  68%|██████▊   | 213/312 [21:30<10:00,  6.06s/it]
Training Epoch: 18/100, batch 426/625 completed (loss: 0.0007:  68%|██████▊   | 213/312 [21:34<10:00,  6.06s/it]
Training Epoch: 18/100, batch 426/625 completed (loss: 0.0009:  68%|██████▊   | 213/312 [21:34<10:00,  6.06s/it]
Training Epoch: 18/100, batch 426/625 completed (loss: 0.0062:  68%|██████▊   | 213/312 [21:34<10:00,  6.06s/it]
Training Epoch: 18/100, batch 426/625 completed (loss: 0.0169:  68%|██████▊   | 213/312 [21:34<10:00,  6.06s/it]
Training Epoch: 18/100, batch 426/625 completed (loss: 0.0007:  69%|██████▊   | 214/312 [21:37<09:57,  6.09s/it]
Training Epoch: 18/100, batch 426/625 completed (loss: 0.0062:  69%|██████▊   | 214/312 [21:37<09:57,  6.09s/it]
Training Epoch: 18/100, batch 426/625 completed (loss: 0.0169:  69%|██████▊   | 214/312 [21:37<09:57,  6.09s/it]
Training Epoch: 18/100, batch 426/625 completed (loss: 0.0009:  69%|██████▊   | 214/312 [21:37<09:57,  6.09s/it]
Training Epoch: 18/100, batch 427/625 completed (loss: 0.003:  69%|██████▊   | 214/312 [21:37<09:57,  6.09s/it] 
Training Epoch: 18/100, batch 427/625 completed (loss: 0.0643:  69%|██████▊   | 214/312 [21:37<09:57,  6.09s/it]
Training Epoch: 18/100, batch 427/625 completed (loss: 0.0003:  69%|██████▊   | 214/312 [21:37<09:57,  6.09s/it]
Training Epoch: 18/100, batch 427/625 completed (loss: 0.0041:  69%|██████▊   | 214/312 [21:37<09:57,  6.09s/it]
Training Epoch: 18/100, batch 428/625 completed (loss: 0.0145:  69%|██████▊   | 214/312 [21:39<09:57,  6.09s/it]
Training Epoch: 18/100, batch 428/625 completed (loss: 0.0263:  69%|██████▊   | 214/312 [21:39<09:57,  6.09s/it]
Training Epoch: 18/100, batch 428/625 completed (loss: 0.0009:  69%|██████▊   | 214/312 [21:39<09:57,  6.09s/it]
Training Epoch: 18/100, batch 428/625 completed (loss: 0.0027:  69%|██████▊   | 214/312 [21:40<09:57,  6.09s/it]
Training Epoch: 18/100, batch 428/625 completed (loss: 0.0263:  69%|██████▉   | 215/312 [21:42<09:47,  6.06s/it]
Training Epoch: 18/100, batch 428/625 completed (loss: 0.0145:  69%|██████▉   | 215/312 [21:43<09:47,  6.06s/it]
Training Epoch: 18/100, batch 428/625 completed (loss: 0.0009:  69%|██████▉   | 215/312 [21:43<09:47,  6.06s/it]
Training Epoch: 18/100, batch 428/625 completed (loss: 0.0027:  69%|██████▉   | 215/312 [21:43<09:48,  6.06s/it]
Training Epoch: 18/100, batch 429/625 completed (loss: 0.0015:  69%|██████▉   | 215/312 [21:43<09:47,  6.06s/it]
Training Epoch: 18/100, batch 429/625 completed (loss: 0.0055:  69%|██████▉   | 215/312 [21:43<09:47,  6.06s/it]
Training Epoch: 18/100, batch 429/625 completed (loss: 0.0011:  69%|██████▉   | 215/312 [21:43<09:47,  6.06s/it]
Training Epoch: 18/100, batch 429/625 completed (loss: 0.0131:  69%|██████▉   | 215/312 [21:43<09:48,  6.06s/it]
Training Epoch: 18/100, batch 430/625 completed (loss: 0.0058:  69%|██████▉   | 215/312 [21:46<09:47,  6.06s/it]
Training Epoch: 18/100, batch 430/625 completed (loss: 0.0006:  69%|██████▉   | 215/312 [21:46<09:47,  6.06s/it]
Training Epoch: 18/100, batch 430/625 completed (loss: 0.0012:  69%|██████▉   | 215/312 [21:46<09:47,  6.06s/it]
Training Epoch: 18/100, batch 430/625 completed (loss: 0.0019:  69%|██████▉   | 215/312 [21:46<09:48,  6.06s/it]
Training Epoch: 18/100, batch 430/625 completed (loss: 0.0006:  69%|██████▉   | 216/312 [21:49<09:40,  6.04s/it]
Training Epoch: 18/100, batch 430/625 completed (loss: 0.0019:  69%|██████▉   | 216/312 [21:49<09:40,  6.04s/it]
Training Epoch: 18/100, batch 430/625 completed (loss: 0.0058:  69%|██████▉   | 216/312 [21:49<09:40,  6.05s/it]
Training Epoch: 18/100, batch 430/625 completed (loss: 0.0012:  69%|██████▉   | 216/312 [21:49<09:40,  6.05s/it]
Training Epoch: 18/100, batch 431/625 completed (loss: 0.0046:  69%|██████▉   | 216/312 [21:49<09:40,  6.04s/it]
Training Epoch: 18/100, batch 431/625 completed (loss: 0.0007:  69%|██████▉   | 216/312 [21:49<09:40,  6.05s/it]
Training Epoch: 18/100, batch 431/625 completed (loss: 0.0003:  69%|██████▉   | 216/312 [21:49<09:40,  6.05s/it]
Training Epoch: 18/100, batch 431/625 completed (loss: 0.0013:  69%|██████▉   | 216/312 [21:49<09:40,  6.04s/it]
Training Epoch: 18/100, batch 432/625 completed (loss: 0.0109:  69%|██████▉   | 216/312 [21:52<09:40,  6.05s/it]
Training Epoch: 18/100, batch 432/625 completed (loss: 0.0365:  69%|██████▉   | 216/312 [21:52<09:40,  6.05s/it]
Training Epoch: 18/100, batch 432/625 completed (loss: 0.0003:  69%|██████▉   | 216/312 [21:52<09:40,  6.04s/it]
Training Epoch: 18/100, batch 432/625 completed (loss: 0.0031:  69%|██████▉   | 216/312 [21:52<09:40,  6.04s/it]
Training Epoch: 18/100, batch 432/625 completed (loss: 0.0003:  70%|██████▉   | 217/312 [21:55<09:35,  6.06s/it]
Training Epoch: 18/100, batch 432/625 completed (loss: 0.0365:  70%|██████▉   | 217/312 [21:55<09:35,  6.06s/it]
Training Epoch: 18/100, batch 432/625 completed (loss: 0.0031:  70%|██████▉   | 217/312 [21:55<09:35,  6.06s/it]
Training Epoch: 18/100, batch 432/625 completed (loss: 0.0109:  70%|██████▉   | 217/312 [21:55<09:35,  6.06s/it]
Training Epoch: 18/100, batch 433/625 completed (loss: 0.0214:  70%|██████▉   | 217/312 [21:55<09:35,  6.06s/it]
Training Epoch: 18/100, batch 433/625 completed (loss: 0.0004:  70%|██████▉   | 217/312 [21:55<09:35,  6.06s/it]
Training Epoch: 18/100, batch 433/625 completed (loss: 0.003:  70%|██████▉   | 217/312 [21:55<09:35,  6.06s/it] 
Training Epoch: 18/100, batch 433/625 completed (loss: 0.0001:  70%|██████▉   | 217/312 [21:55<09:35,  6.06s/it]
Training Epoch: 18/100, batch 434/625 completed (loss: 0.0032:  70%|██████▉   | 217/312 [21:58<09:35,  6.06s/it]
Training Epoch: 18/100, batch 434/625 completed (loss: 0.0002:  70%|██████▉   | 217/312 [21:58<09:35,  6.06s/it]
Training Epoch: 18/100, batch 434/625 completed (loss: 0.0404:  70%|██████▉   | 217/312 [21:58<09:35,  6.06s/it]
Training Epoch: 18/100, batch 434/625 completed (loss: 0.0301:  70%|██████▉   | 217/312 [21:58<09:35,  6.06s/it]
Training Epoch: 18/100, batch 434/625 completed (loss: 0.0032:  70%|██████▉   | 218/312 [22:01<09:32,  6.09s/it]
Training Epoch: 18/100, batch 434/625 completed (loss: 0.0002:  70%|██████▉   | 218/312 [22:01<09:32,  6.09s/it]
Training Epoch: 18/100, batch 434/625 completed (loss: 0.0301:  70%|██████▉   | 218/312 [22:01<09:32,  6.09s/it]
Training Epoch: 18/100, batch 434/625 completed (loss: 0.0404:  70%|██████▉   | 218/312 [22:01<09:32,  6.09s/it]
Training Epoch: 18/100, batch 435/625 completed (loss: 0.1623:  70%|██████▉   | 218/312 [22:01<09:32,  6.09s/it]
Training Epoch: 18/100, batch 435/625 completed (loss: 0.0227:  70%|██████▉   | 218/312 [22:01<09:32,  6.09s/it]
Training Epoch: 18/100, batch 435/625 completed (loss: 0.0005:  70%|██████▉   | 218/312 [22:01<09:32,  6.09s/it]
Training Epoch: 18/100, batch 435/625 completed (loss: 0.0016:  70%|██████▉   | 218/312 [22:01<09:32,  6.09s/it]
Training Epoch: 18/100, batch 436/625 completed (loss: 0.014:  70%|██████▉   | 218/312 [22:04<09:32,  6.09s/it] 
Training Epoch: 18/100, batch 436/625 completed (loss: 0.0133:  70%|██████▉   | 218/312 [22:04<09:32,  6.09s/it]
Training Epoch: 18/100, batch 436/625 completed (loss: 0.0018:  70%|██████▉   | 218/312 [22:04<09:32,  6.09s/it]
Training Epoch: 18/100, batch 436/625 completed (loss: 0.0015:  70%|██████▉   | 218/312 [22:04<09:32,  6.09s/it]
Training Epoch: 18/100, batch 436/625 completed (loss: 0.0133:  70%|███████   | 219/312 [22:07<09:26,  6.10s/it]
Training Epoch: 18/100, batch 436/625 completed (loss: 0.014:  70%|███████   | 219/312 [22:07<09:26,  6.10s/it]
Training Epoch: 18/100, batch 436/625 completed (loss: 0.0018:  70%|███████   | 219/312 [22:07<09:26,  6.10s/it]
Training Epoch: 18/100, batch 436/625 completed (loss: 0.0015:  70%|███████   | 219/312 [22:07<09:27,  6.10s/it]
Training Epoch: 18/100, batch 437/625 completed (loss: 0.0004:  70%|███████   | 219/312 [22:07<09:26,  6.10s/it]
Training Epoch: 18/100, batch 437/625 completed (loss: 0.0082:  70%|███████   | 219/312 [22:07<09:26,  6.10s/it]
Training Epoch: 18/100, batch 437/625 completed (loss: 0.052:  70%|███████   | 219/312 [22:07<09:26,  6.10s/it]
Training Epoch: 18/100, batch 437/625 completed (loss: 0.0006:  70%|███████   | 219/312 [22:07<09:27,  6.10s/it]
Training Epoch: 18/100, batch 438/625 completed (loss: 0.0004:  70%|███████   | 219/312 [22:10<09:26,  6.10s/it]
Training Epoch: 18/100, batch 438/625 completed (loss: 0.0049:  70%|███████   | 219/312 [22:10<09:26,  6.10s/it]
Training Epoch: 18/100, batch 438/625 completed (loss: 0.0015:  70%|███████   | 219/312 [22:10<09:26,  6.10s/it]
Training Epoch: 18/100, batch 438/625 completed (loss: 0.0056:  70%|███████   | 219/312 [22:10<09:27,  6.10s/it]
Training Epoch: 18/100, batch 438/625 completed (loss: 0.0004:  71%|███████   | 220/312 [22:13<09:18,  6.07s/it]
Training Epoch: 18/100, batch 438/625 completed (loss: 0.0056:  71%|███████   | 220/312 [22:13<09:18,  6.07s/it]
Training Epoch: 18/100, batch 438/625 completed (loss: 0.0015:  71%|███████   | 220/312 [22:13<09:18,  6.07s/it]
Training Epoch: 18/100, batch 438/625 completed (loss: 0.0049:  71%|███████   | 220/312 [22:13<09:18,  6.07s/it]
Training Epoch: 18/100, batch 439/625 completed (loss: 0.0009:  71%|███████   | 220/312 [22:13<09:18,  6.07s/it]
Training Epoch: 18/100, batch 439/625 completed (loss: 0.0024:  71%|███████   | 220/312 [22:13<09:18,  6.07s/it]
Training Epoch: 18/100, batch 439/625 completed (loss: 0.0002:  71%|███████   | 220/312 [22:13<09:18,  6.07s/it]
Training Epoch: 18/100, batch 439/625 completed (loss: 0.0016:  71%|███████   | 220/312 [22:13<09:18,  6.07s/it]
Training Epoch: 18/100, batch 440/625 completed (loss: 0.0892:  71%|███████   | 220/312 [22:16<09:18,  6.07s/it]
Training Epoch: 18/100, batch 440/625 completed (loss: 0.0324:  71%|███████   | 220/312 [22:16<09:18,  6.07s/it]
Training Epoch: 18/100, batch 440/625 completed (loss: 0.0004:  71%|███████   | 220/312 [22:16<09:18,  6.07s/it]
Training Epoch: 18/100, batch 440/625 completed (loss: 0.0035:  71%|███████   | 220/312 [22:16<09:18,  6.07s/it]
Training Epoch: 18/100, batch 440/625 completed (loss: 0.0035:  71%|███████   | 221/312 [22:19<09:10,  6.05s/it]
Training Epoch: 18/100, batch 440/625 completed (loss: 0.0324:  71%|███████   | 221/312 [22:19<09:11,  6.06s/it]
Training Epoch: 18/100, batch 440/625 completed (loss: 0.0892:  71%|███████   | 221/312 [22:19<09:12,  6.07s/it]
Training Epoch: 18/100, batch 440/625 completed (loss: 0.0004:  71%|███████   | 221/312 [22:19<09:12,  6.07s/it]
Training Epoch: 18/100, batch 441/625 completed (loss: 0.0001:  71%|███████   | 221/312 [22:19<09:11,  6.06s/it]
Training Epoch: 18/100, batch 441/625 completed (loss: 0.002:  71%|███████   | 221/312 [22:19<09:12,  6.07s/it] 
Training Epoch: 18/100, batch 441/625 completed (loss: 0.0002:  71%|███████   | 221/312 [22:19<09:12,  6.07s/it]
Training Epoch: 18/100, batch 441/625 completed (loss: 0.0022:  71%|███████   | 221/312 [22:19<09:10,  6.05s/it]
Training Epoch: 18/100, batch 442/625 completed (loss: 0.0026:  71%|███████   | 221/312 [22:22<09:12,  6.07s/it]
Training Epoch: 18/100, batch 442/625 completed (loss: 0.0134:  71%|███████   | 221/312 [22:22<09:11,  6.06s/it]
Training Epoch: 18/100, batch 442/625 completed (loss: 0.0109:  71%|███████   | 221/312 [22:22<09:12,  6.07s/it]
Training Epoch: 18/100, batch 442/625 completed (loss: 0.1308:  71%|███████   | 221/312 [22:22<09:10,  6.05s/it]
Training Epoch: 18/100, batch 442/625 completed (loss: 0.0134:  71%|███████   | 222/312 [22:25<09:09,  6.10s/it]
Training Epoch: 18/100, batch 442/625 completed (loss: 0.0026:  71%|███████   | 222/312 [22:25<09:08,  6.10s/it]
Training Epoch: 18/100, batch 442/625 completed (loss: 0.1308:  71%|███████   | 222/312 [22:25<09:09,  6.10s/it]
Training Epoch: 18/100, batch 442/625 completed (loss: 0.0109:  71%|███████   | 222/312 [22:25<09:09,  6.10s/it]
Training Epoch: 18/100, batch 443/625 completed (loss: 0.0082:  71%|███████   | 222/312 [22:25<09:09,  6.10s/it]
Training Epoch: 18/100, batch 443/625 completed (loss: 0.0001:  71%|███████   | 222/312 [22:25<09:08,  6.10s/it]
Training Epoch: 18/100, batch 443/625 completed (loss: 0.0019:  71%|███████   | 222/312 [22:25<09:09,  6.10s/it]
Training Epoch: 18/100, batch 443/625 completed (loss: 0.0058:  71%|███████   | 222/312 [22:25<09:09,  6.10s/it]
Training Epoch: 18/100, batch 444/625 completed (loss: 0.0131:  71%|███████   | 222/312 [22:28<09:08,  6.10s/it]
Training Epoch: 18/100, batch 444/625 completed (loss: 0.0942:  71%|███████   | 222/312 [22:28<09:09,  6.10s/it]
Training Epoch: 18/100, batch 444/625 completed (loss: 0.0058:  71%|███████   | 222/312 [22:28<09:09,  6.10s/it]
Training Epoch: 18/100, batch 444/625 completed (loss: 0.0005:  71%|███████   | 222/312 [22:28<09:09,  6.10s/it]
Training Epoch: 18/100, batch 444/625 completed (loss: 0.0058:  71%|███████▏  | 223/312 [22:31<09:00,  6.07s/it]
Training Epoch: 18/100, batch 444/625 completed (loss: 0.0942:  71%|███████▏  | 223/312 [22:31<09:00,  6.07s/it]
Training Epoch: 18/100, batch 444/625 completed (loss: 0.0131:  71%|███████▏  | 223/312 [22:31<09:00,  6.07s/it]
Training Epoch: 18/100, batch 444/625 completed (loss: 0.0005:  71%|███████▏  | 223/312 [22:31<09:00,  6.08s/it]
Training Epoch: 18/100, batch 445/625 completed (loss: 0.0193:  71%|███████▏  | 223/312 [22:31<09:00,  6.07s/it]
Training Epoch: 18/100, batch 445/625 completed (loss: 0.0001:  71%|███████▏  | 223/312 [22:31<09:00,  6.07s/it]
Training Epoch: 18/100, batch 445/625 completed (loss: 0.0002:  71%|███████▏  | 223/312 [22:31<09:00,  6.07s/it]
Training Epoch: 18/100, batch 445/625 completed (loss: 0.0005:  71%|███████▏  | 223/312 [22:31<09:00,  6.08s/it]
Training Epoch: 18/100, batch 446/625 completed (loss: 0.0004:  71%|███████▏  | 223/312 [22:34<09:00,  6.07s/it]
Training Epoch: 18/100, batch 446/625 completed (loss: 0.0047:  71%|███████▏  | 223/312 [22:34<09:00,  6.07s/it]
Training Epoch: 18/100, batch 446/625 completed (loss: 0.0626:  71%|███████▏  | 223/312 [22:34<09:00,  6.07s/it]
Training Epoch: 18/100, batch 446/625 completed (loss: 0.0521:  71%|███████▏  | 223/312 [22:34<09:00,  6.08s/it]
Training Epoch: 18/100, batch 446/625 completed (loss: 0.0047:  72%|███████▏  | 224/312 [22:37<08:57,  6.11s/it]
Training Epoch: 18/100, batch 446/625 completed (loss: 0.0521:  72%|███████▏  | 224/312 [22:37<08:57,  6.11s/it]
Training Epoch: 18/100, batch 446/625 completed (loss: 0.0004:  72%|███████▏  | 224/312 [22:37<08:57,  6.11s/it]
Training Epoch: 18/100, batch 446/625 completed (loss: 0.0626:  72%|███████▏  | 224/312 [22:37<08:58,  6.12s/it]
Training Epoch: 18/100, batch 447/625 completed (loss: 0.0121:  72%|███████▏  | 224/312 [22:37<08:57,  6.11s/it]
Training Epoch: 18/100, batch 447/625 completed (loss: 0.0007:  72%|███████▏  | 224/312 [22:37<08:57,  6.11s/it]
Training Epoch: 18/100, batch 447/625 completed (loss: 0.0163:  72%|███████▏  | 224/312 [22:37<08:58,  6.12s/it]
Training Epoch: 18/100, batch 447/625 completed (loss: 0.0236:  72%|███████▏  | 224/312 [22:37<08:57,  6.11s/it]
Training Epoch: 18/100, batch 448/625 completed (loss: 0.0003:  72%|███████▏  | 224/312 [22:40<08:57,  6.11s/it]
Training Epoch: 18/100, batch 448/625 completed (loss: 0.0067:  72%|███████▏  | 224/312 [22:40<08:57,  6.11s/it]
Training Epoch: 18/100, batch 448/625 completed (loss: 0.0024:  72%|███████▏  | 224/312 [22:40<08:58,  6.12s/it]
Training Epoch: 18/100, batch 448/625 completed (loss: 0.0016:  72%|███████▏  | 224/312 [22:41<08:57,  6.11s/it]
Training Epoch: 18/100, batch 448/625 completed (loss: 0.0024:  72%|███████▏  | 225/312 [22:43<08:48,  6.08s/it]
Training Epoch: 18/100, batch 448/625 completed (loss: 0.0003:  72%|███████▏  | 225/312 [22:43<08:49,  6.08s/it]
Training Epoch: 18/100, batch 448/625 completed (loss: 0.0016:  72%|███████▏  | 225/312 [22:43<08:49,  6.09s/it]
Training Epoch: 18/100, batch 448/625 completed (loss: 0.0067:  72%|███████▏  | 225/312 [22:43<08:49,  6.08s/it]
Training Epoch: 18/100, batch 449/625 completed (loss: 0.0161:  72%|███████▏  | 225/312 [22:43<08:48,  6.08s/it]
Training Epoch: 18/100, batch 449/625 completed (loss: 0.0018:  72%|███████▏  | 225/312 [22:43<08:49,  6.08s/it]
Training Epoch: 18/100, batch 449/625 completed (loss: 0.0008:  72%|███████▏  | 225/312 [22:43<08:49,  6.08s/it]
Training Epoch: 18/100, batch 449/625 completed (loss: 0.0018:  72%|███████▏  | 225/312 [22:43<08:49,  6.09s/it]
Training Epoch: 18/100, batch 450/625 completed (loss: 0.0005:  72%|███████▏  | 225/312 [22:46<08:48,  6.08s/it]
Training Epoch: 18/100, batch 450/625 completed (loss: 0.008:  72%|███████▏  | 225/312 [22:46<08:49,  6.08s/it] 
Training Epoch: 18/100, batch 450/625 completed (loss: 0.0125:  72%|███████▏  | 225/312 [22:46<08:49,  6.08s/it]
Training Epoch: 18/100, batch 450/625 completed (loss: 0.0033:  72%|███████▏  | 225/312 [22:47<08:49,  6.09s/it]
Training Epoch: 18/100, batch 450/625 completed (loss: 0.008:  72%|███████▏  | 226/312 [22:49<08:44,  6.10s/it]
Training Epoch: 18/100, batch 450/625 completed (loss: 0.0033:  72%|███████▏  | 226/312 [22:50<08:44,  6.10s/it]
Training Epoch: 18/100, batch 450/625 completed (loss: 0.0125:  72%|███████▏  | 226/312 [22:49<08:44,  6.10s/it]
Training Epoch: 18/100, batch 450/625 completed (loss: 0.0005:  72%|███████▏  | 226/312 [22:49<08:44,  6.10s/it]
Training Epoch: 18/100, batch 451/625 completed (loss: 0.0007:  72%|███████▏  | 226/312 [22:50<08:44,  6.10s/it]
Training Epoch: 18/100, batch 451/625 completed (loss: 0.0018:  72%|███████▏  | 226/312 [22:50<08:44,  6.10s/it]
Training Epoch: 18/100, batch 451/625 completed (loss: 0.0014:  72%|███████▏  | 226/312 [22:50<08:44,  6.10s/it]
Training Epoch: 18/100, batch 451/625 completed (loss: 0.0004:  72%|███████▏  | 226/312 [22:50<08:44,  6.10s/it]
Training Epoch: 18/100, batch 452/625 completed (loss: 0.0199:  72%|███████▏  | 226/312 [22:53<08:44,  6.10s/it]
Training Epoch: 18/100, batch 452/625 completed (loss: 0.0186:  72%|███████▏  | 226/312 [22:53<08:44,  6.10s/it]
Training Epoch: 18/100, batch 452/625 completed (loss: 0.003:  72%|███████▏  | 226/312 [22:53<08:44,  6.10s/it] 
Training Epoch: 18/100, batch 452/625 completed (loss: 0.0001:  72%|███████▏  | 226/312 [22:53<08:44,  6.10s/it]
Training Epoch: 18/100, batch 452/625 completed (loss: 0.0199:  73%|███████▎  | 227/312 [22:56<08:37,  6.09s/it]
Training Epoch: 18/100, batch 452/625 completed (loss: 0.003:  73%|███████▎  | 227/312 [22:56<08:37,  6.08s/it]
Training Epoch: 18/100, batch 452/625 completed (loss: 0.0001:  73%|███████▎  | 227/312 [22:56<08:37,  6.09s/it]
Training Epoch: 18/100, batch 452/625 completed (loss: 0.0186:  73%|███████▎  | 227/312 [22:56<08:37,  6.09s/it]
Training Epoch: 18/100, batch 453/625 completed (loss: 0.0006:  73%|███████▎  | 227/312 [22:56<08:37,  6.09s/it]
Training Epoch: 18/100, batch 453/625 completed (loss: 0.0018:  73%|███████▎  | 227/312 [22:56<08:37,  6.08s/it]
Training Epoch: 18/100, batch 453/625 completed (loss: 0.0018:  73%|███████▎  | 227/312 [22:56<08:37,  6.09s/it]
Training Epoch: 18/100, batch 453/625 completed (loss: 0.0023:  73%|███████▎  | 227/312 [22:56<08:37,  6.09s/it]
Training Epoch: 18/100, batch 454/625 completed (loss: 0.0236:  73%|███████▎  | 227/312 [22:59<08:37,  6.08s/it]
Training Epoch: 18/100, batch 454/625 completed (loss: 0.0001:  73%|███████▎  | 227/312 [22:59<08:37,  6.09s/it]
Training Epoch: 18/100, batch 454/625 completed (loss: 0.0008:  73%|███████▎  | 227/312 [22:59<08:37,  6.09s/it]
Training Epoch: 18/100, batch 454/625 completed (loss: 0.0023:  73%|███████▎  | 227/312 [22:59<08:37,  6.09s/it]
Training Epoch: 18/100, batch 454/625 completed (loss: 0.0008:  73%|███████▎  | 228/312 [23:02<08:33,  6.11s/it]
Training Epoch: 18/100, batch 454/625 completed (loss: 0.0001:  73%|███████▎  | 228/312 [23:02<08:32,  6.10s/it]
Training Epoch: 18/100, batch 454/625 completed (loss: 0.0236:  73%|███████▎  | 228/312 [23:02<08:32,  6.11s/it]
Training Epoch: 18/100, batch 454/625 completed (loss: 0.0023:  73%|███████▎  | 228/312 [23:02<08:33,  6.11s/it]
Training Epoch: 18/100, batch 455/625 completed (loss: 0.0412:  73%|███████▎  | 228/312 [23:02<08:32,  6.10s/it]
Training Epoch: 18/100, batch 455/625 completed (loss: 0.0342:  73%|███████▎  | 228/312 [23:02<08:33,  6.11s/it]
Training Epoch: 18/100, batch 455/625 completed (loss: 0.0199:  73%|███████▎  | 228/312 [23:02<08:32,  6.11s/it]
Training Epoch: 18/100, batch 455/625 completed (loss: 0.0058:  73%|███████▎  | 228/312 [23:02<08:33,  6.11s/it]
Training Epoch: 18/100, batch 456/625 completed (loss: 0.0031:  73%|███████▎  | 228/312 [23:05<08:32,  6.11s/it]
Training Epoch: 18/100, batch 456/625 completed (loss: 0.0235:  73%|███████▎  | 228/312 [23:05<08:32,  6.10s/it]
Training Epoch: 18/100, batch 456/625 completed (loss: 0.0005:  73%|███████▎  | 228/312 [23:05<08:33,  6.11s/it]
Training Epoch: 18/100, batch 456/625 completed (loss: 0.0013:  73%|███████▎  | 228/312 [23:05<08:33,  6.11s/it]
Training Epoch: 18/100, batch 456/625 completed (loss: 0.0005:  73%|███████▎  | 229/312 [23:08<08:24,  6.08s/it]
Training Epoch: 18/100, batch 456/625 completed (loss: 0.0013:  73%|███████▎  | 229/312 [23:08<08:24,  6.08s/it]
Training Epoch: 18/100, batch 456/625 completed (loss: 0.0031:  73%|███████▎  | 229/312 [23:08<08:25,  6.09s/it]
Training Epoch: 18/100, batch 456/625 completed (loss: 0.0235:  73%|███████▎  | 229/312 [23:08<08:25,  6.09s/it]
Training Epoch: 18/100, batch 457/625 completed (loss: 0.0012:  73%|███████▎  | 229/312 [23:08<08:25,  6.09s/it]
Training Epoch: 18/100, batch 457/625 completed (loss: 0.056:  73%|███████▎  | 229/312 [23:08<08:24,  6.08s/it] 
Training Epoch: 18/100, batch 457/625 completed (loss: 0.0029:  73%|███████▎  | 229/312 [23:08<08:25,  6.09s/it]
Training Epoch: 18/100, batch 457/625 completed (loss: 0.0025:  73%|███████▎  | 229/312 [23:08<08:24,  6.08s/it]
Training Epoch: 18/100, batch 458/625 completed (loss: 0.0206:  73%|███████▎  | 229/312 [23:11<08:25,  6.09s/it]
Training Epoch: 18/100, batch 458/625 completed (loss: 0.0005:  73%|███████▎  | 229/312 [23:11<08:24,  6.08s/it]
Training Epoch: 18/100, batch 458/625 completed (loss: 0.0194:  73%|███████▎  | 229/312 [23:11<08:25,  6.09s/it]
Training Epoch: 18/100, batch 458/625 completed (loss: 0.0043:  73%|███████▎  | 229/312 [23:11<08:24,  6.08s/it]
Training Epoch: 18/100, batch 458/625 completed (loss: 0.0194:  74%|███████▎  | 230/312 [23:14<08:20,  6.11s/it]
Training Epoch: 18/100, batch 458/625 completed (loss: 0.0005:  74%|███████▎  | 230/312 [23:14<08:21,  6.11s/it]
Training Epoch: 18/100, batch 458/625 completed (loss: 0.0043:  74%|███████▎  | 230/312 [23:14<08:21,  6.11s/it]
Training Epoch: 18/100, batch 458/625 completed (loss: 0.0206:  74%|███████▎  | 230/312 [23:14<08:21,  6.11s/it]
Training Epoch: 18/100, batch 459/625 completed (loss: 0.0008:  74%|███████▎  | 230/312 [23:14<08:21,  6.11s/it]
Training Epoch: 18/100, batch 459/625 completed (loss: 0.0033:  74%|███████▎  | 230/312 [23:14<08:20,  6.11s/it]
Training Epoch: 18/100, batch 459/625 completed (loss: 0.0013:  74%|███████▎  | 230/312 [23:14<08:21,  6.11s/it]
Training Epoch: 18/100, batch 459/625 completed (loss: 0.0004:  74%|███████▎  | 230/312 [23:14<08:21,  6.11s/it]
Training Epoch: 18/100, batch 460/625 completed (loss: 0.001:  74%|███████▎  | 230/312 [23:17<08:21,  6.11s/it] 
Training Epoch: 18/100, batch 460/625 completed (loss: 0.0003:  74%|███████▎  | 230/312 [23:17<08:20,  6.11s/it]
Training Epoch: 18/100, batch 460/625 completed (loss: 0.0117:  74%|███████▎  | 230/312 [23:17<08:21,  6.11s/it]
Training Epoch: 18/100, batch 460/625 completed (loss: 0.0001:  74%|███████▎  | 230/312 [23:17<08:21,  6.11s/it]
Training Epoch: 18/100, batch 460/625 completed (loss: 0.0117:  74%|███████▍  | 231/312 [23:20<08:15,  6.12s/it]
Training Epoch: 18/100, batch 460/625 completed (loss: 0.001:  74%|███████▍  | 231/312 [23:20<08:16,  6.13s/it]
Training Epoch: 18/100, batch 460/625 completed (loss: 0.0001:  74%|███████▍  | 231/312 [23:20<08:16,  6.13s/it]
Training Epoch: 18/100, batch 460/625 completed (loss: 0.0003:  74%|███████▍  | 231/312 [23:20<08:16,  6.13s/it]
Training Epoch: 18/100, batch 461/625 completed (loss: 0.0045:  74%|███████▍  | 231/312 [23:20<08:16,  6.13s/it]
Training Epoch: 18/100, batch 461/625 completed (loss: 0.0134:  74%|███████▍  | 231/312 [23:20<08:15,  6.12s/it]
Training Epoch: 18/100, batch 461/625 completed (loss: 0.0025:  74%|███████▍  | 231/312 [23:20<08:16,  6.13s/it]
Training Epoch: 18/100, batch 461/625 completed (loss: 0.0013:  74%|███████▍  | 231/312 [23:20<08:16,  6.13s/it]
Training Epoch: 18/100, batch 462/625 completed (loss: 0.0014:  74%|███████▍  | 231/312 [23:23<08:16,  6.13s/it]
Training Epoch: 18/100, batch 462/625 completed (loss: 0.0014:  74%|███████▍  | 231/312 [23:23<08:15,  6.12s/it]
Training Epoch: 18/100, batch 462/625 completed (loss: 0.004:  74%|███████▍  | 231/312 [23:23<08:16,  6.13s/it] 
Training Epoch: 18/100, batch 462/625 completed (loss: 0.0059:  74%|███████▍  | 231/312 [23:23<08:16,  6.13s/it]
Training Epoch: 18/100, batch 462/625 completed (loss: 0.004:  74%|███████▍  | 232/312 [23:26<08:08,  6.10s/it]
Training Epoch: 18/100, batch 462/625 completed (loss: 0.0059:  74%|███████▍  | 232/312 [23:26<08:08,  6.10s/it]
Training Epoch: 18/100, batch 462/625 completed (loss: 0.0014:  74%|███████▍  | 232/312 [23:26<08:08,  6.10s/it]
Training Epoch: 18/100, batch 462/625 completed (loss: 0.0014:  74%|███████▍  | 232/312 [23:26<08:08,  6.10s/it]
Training Epoch: 18/100, batch 463/625 completed (loss: 0.0017:  74%|███████▍  | 232/312 [23:26<08:08,  6.10s/it]
Training Epoch: 18/100, batch 463/625 completed (loss: 0.0001:  74%|███████▍  | 232/312 [23:26<08:08,  6.10s/it]
Training Epoch: 18/100, batch 463/625 completed (loss: 0.0073:  74%|███████▍  | 232/312 [23:26<08:08,  6.10s/it]
Training Epoch: 18/100, batch 463/625 completed (loss: 0.0004:  74%|███████▍  | 232/312 [23:26<08:08,  6.10s/it]
Training Epoch: 18/100, batch 464/625 completed (loss: 0.0215:  74%|███████▍  | 232/312 [23:29<08:08,  6.10s/it]
Training Epoch: 18/100, batch 464/625 completed (loss: 0.0006:  74%|███████▍  | 232/312 [23:29<08:08,  6.10s/it]
Training Epoch: 18/100, batch 464/625 completed (loss: 0.0626:  74%|███████▍  | 232/312 [23:29<08:08,  6.10s/it]
Training Epoch: 18/100, batch 464/625 completed (loss: 0.0003:  74%|███████▍  | 232/312 [23:29<08:08,  6.10s/it]
Training Epoch: 18/100, batch 464/625 completed (loss: 0.0003:  75%|███████▍  | 233/312 [23:32<08:00,  6.08s/it]
Training Epoch: 18/100, batch 464/625 completed (loss: 0.0215:  75%|███████▍  | 233/312 [23:32<08:00,  6.09s/it]
Training Epoch: 18/100, batch 464/625 completed (loss: 0.0006:  75%|███████▍  | 233/312 [23:32<08:00,  6.08s/it]
Training Epoch: 18/100, batch 464/625 completed (loss: 0.0626:  75%|███████▍  | 233/312 [23:32<08:01,  6.09s/it]
Training Epoch: 18/100, batch 465/625 completed (loss: 0.0082:  75%|███████▍  | 233/312 [23:32<08:01,  6.09s/it]
Training Epoch: 18/100, batch 465/625 completed (loss: 0.0015:  75%|███████▍  | 233/312 [23:32<08:00,  6.08s/it]
Training Epoch: 18/100, batch 465/625 completed (loss: 0.0001:  75%|███████▍  | 233/312 [23:32<08:00,  6.09s/it]
Training Epoch: 18/100, batch 465/625 completed (loss: 0.0031:  75%|███████▍  | 233/312 [23:32<08:00,  6.08s/it]
Training Epoch: 18/100, batch 466/625 completed (loss: 0.0045:  75%|███████▍  | 233/312 [23:35<08:00,  6.08s/it]
Training Epoch: 18/100, batch 466/625 completed (loss: 0.0043:  75%|███████▍  | 233/312 [23:35<08:00,  6.09s/it]
Training Epoch: 18/100, batch 466/625 completed (loss: 0.0023:  75%|███████▍  | 233/312 [23:35<08:01,  6.09s/it]
Training Epoch: 18/100, batch 466/625 completed (loss: 0.02:  75%|███████▍  | 233/312 [23:35<08:00,  6.08s/it]  
Training Epoch: 18/100, batch 466/625 completed (loss: 0.0043:  75%|███████▌  | 234/312 [23:38<07:56,  6.11s/it]
Training Epoch: 18/100, batch 466/625 completed (loss: 0.02:  75%|███████▌  | 234/312 [23:38<07:56,  6.11s/it]
Training Epoch: 18/100, batch 466/625 completed (loss: 0.0045:  75%|███████▌  | 234/312 [23:38<07:56,  6.11s/it]
Training Epoch: 18/100, batch 466/625 completed (loss: 0.0023:  75%|███████▌  | 234/312 [23:38<07:56,  6.11s/it]
Training Epoch: 18/100, batch 467/625 completed (loss: 0.002:  75%|███████▌  | 234/312 [23:38<07:56,  6.11s/it] 
Training Epoch: 18/100, batch 467/625 completed (loss: 0.0004:  75%|███████▌  | 234/312 [23:38<07:56,  6.11s/it]
Training Epoch: 18/100, batch 467/625 completed (loss: 0.006:  75%|███████▌  | 234/312 [23:38<07:56,  6.11s/it] 
Training Epoch: 18/100, batch 467/625 completed (loss: 0.0025:  75%|███████▌  | 234/312 [23:38<07:56,  6.11s/it]
Training Epoch: 18/100, batch 468/625 completed (loss: 0.0059:  75%|███████▌  | 234/312 [23:41<07:56,  6.11s/it]
Training Epoch: 18/100, batch 468/625 completed (loss: 0.1158:  75%|███████▌  | 234/312 [23:41<07:56,  6.11s/it]
Training Epoch: 18/100, batch 468/625 completed (loss: 0.0003:  75%|███████▌  | 234/312 [23:41<07:56,  6.11s/it]
Training Epoch: 18/100, batch 468/625 completed (loss: 0.0544:  75%|███████▌  | 234/312 [23:42<07:56,  6.11s/it]
Training Epoch: 18/100, batch 468/625 completed (loss: 0.0059:  75%|███████▌  | 235/312 [23:44<07:48,  6.08s/it]
Training Epoch: 18/100, batch 468/625 completed (loss: 0.0003:  75%|███████▌  | 235/312 [23:44<07:48,  6.09s/it]
Training Epoch: 18/100, batch 468/625 completed (loss: 0.1158:  75%|███████▌  | 235/312 [23:44<07:48,  6.08s/it]
Training Epoch: 18/100, batch 468/625 completed (loss: 0.0544:  75%|███████▌  | 235/312 [23:44<07:48,  6.09s/it]
Training Epoch: 18/100, batch 469/625 completed (loss: 0.0011:  75%|███████▌  | 235/312 [23:44<07:48,  6.08s/it]
Training Epoch: 18/100, batch 469/625 completed (loss: 0.0134:  75%|███████▌  | 235/312 [23:44<07:48,  6.08s/it]
Training Epoch: 18/100, batch 469/625 completed (loss: 0.0086:  75%|███████▌  | 235/312 [23:44<07:48,  6.09s/it]
Training Epoch: 18/100, batch 469/625 completed (loss: 0.0005:  75%|███████▌  | 235/312 [23:44<07:48,  6.09s/it]
Training Epoch: 18/100, batch 470/625 completed (loss: 0.0031:  75%|███████▌  | 235/312 [23:47<07:48,  6.09s/it]
Training Epoch: 18/100, batch 470/625 completed (loss: 0.0012:  75%|███████▌  | 235/312 [23:47<07:48,  6.08s/it]
Training Epoch: 18/100, batch 470/625 completed (loss: 0.0514:  75%|███████▌  | 235/312 [23:47<07:48,  6.08s/it]
Training Epoch: 18/100, batch 470/625 completed (loss: 0.001:  75%|███████▌  | 235/312 [23:48<07:48,  6.09s/it] 
Training Epoch: 18/100, batch 470/625 completed (loss: 0.0514:  76%|███████▌  | 236/312 [23:51<07:44,  6.11s/it]
Training Epoch: 18/100, batch 470/625 completed (loss: 0.001:  76%|███████▌  | 236/312 [23:51<07:44,  6.12s/it]
Training Epoch: 18/100, batch 470/625 completed (loss: 0.0031:  76%|███████▌  | 236/312 [23:51<07:44,  6.12s/it]
Training Epoch: 18/100, batch 470/625 completed (loss: 0.0012:  76%|███████▌  | 236/312 [23:51<07:44,  6.12s/it]
Training Epoch: 18/100, batch 471/625 completed (loss: 0.0007:  76%|███████▌  | 236/312 [23:51<07:44,  6.12s/it]
Training Epoch: 18/100, batch 471/625 completed (loss: 0.026:  76%|███████▌  | 236/312 [23:51<07:44,  6.11s/it] 
Training Epoch: 18/100, batch 471/625 completed (loss: 0.0185:  76%|███████▌  | 236/312 [23:51<07:44,  6.12s/it]
Training Epoch: 18/100, batch 471/625 completed (loss: 0.0696:  76%|███████▌  | 236/312 [23:51<07:44,  6.12s/it]
Training Epoch: 18/100, batch 472/625 completed (loss: 0.0001:  76%|███████▌  | 236/312 [23:54<07:44,  6.12s/it]
Training Epoch: 18/100, batch 472/625 completed (loss: 0.0052:  76%|███████▌  | 236/312 [23:54<07:44,  6.12s/it]
Training Epoch: 18/100, batch 472/625 completed (loss: 0.0002:  76%|███████▌  | 236/312 [23:54<07:44,  6.11s/it]
Training Epoch: 18/100, batch 472/625 completed (loss: 0.1817:  76%|███████▌  | 236/312 [23:54<07:44,  6.12s/it]
Training Epoch: 18/100, batch 472/625 completed (loss: 0.1817:  76%|███████▌  | 237/312 [23:57<07:39,  6.12s/it]
Training Epoch: 18/100, batch 472/625 completed (loss: 0.0001:  76%|███████▌  | 237/312 [23:57<07:39,  6.13s/it]
Training Epoch: 18/100, batch 472/625 completed (loss: 0.0002:  76%|███████▌  | 237/312 [23:57<07:39,  6.13s/it]
Training Epoch: 18/100, batch 472/625 completed (loss: 0.0052:  76%|███████▌  | 237/312 [23:57<07:39,  6.13s/it]
Training Epoch: 18/100, batch 473/625 completed (loss: 0.0009:  76%|███████▌  | 237/312 [23:57<07:39,  6.13s/it]
Training Epoch: 18/100, batch 473/625 completed (loss: 0.0023:  76%|███████▌  | 237/312 [23:57<07:39,  6.13s/it]
Training Epoch: 18/100, batch 473/625 completed (loss: 0.0012:  76%|███████▌  | 237/312 [23:57<07:39,  6.13s/it]
Training Epoch: 18/100, batch 473/625 completed (loss: 0.0008:  76%|███████▌  | 237/312 [23:57<07:39,  6.12s/it]
Training Epoch: 18/100, batch 474/625 completed (loss: 0.0011:  76%|███████▌  | 237/312 [24:00<07:39,  6.13s/it]
Training Epoch: 18/100, batch 474/625 completed (loss: 0.0008:  76%|███████▌  | 237/312 [24:00<07:39,  6.13s/it]
Training Epoch: 18/100, batch 474/625 completed (loss: 0.01:  76%|███████▌  | 237/312 [24:00<07:39,  6.13s/it]  
Training Epoch: 18/100, batch 474/625 completed (loss: 0.0302:  76%|███████▌  | 237/312 [24:00<07:39,  6.12s/it]
Training Epoch: 18/100, batch 474/625 completed (loss: 0.0011:  76%|███████▋  | 238/312 [24:03<07:28,  6.06s/it]
Training Epoch: 18/100, batch 474/625 completed (loss: 0.01:  76%|███████▋  | 238/312 [24:03<07:27,  6.05s/it]
Training Epoch: 18/100, batch 474/625 completed (loss: 0.0008:  76%|███████▋  | 238/312 [24:03<07:28,  6.06s/it]
Training Epoch: 18/100, batch 474/625 completed (loss: 0.0302:  76%|███████▋  | 238/312 [24:03<07:28,  6.06s/it]
Training Epoch: 18/100, batch 475/625 completed (loss: 0.0007:  76%|███████▋  | 238/312 [24:03<07:27,  6.05s/it]
Training Epoch: 18/100, batch 475/625 completed (loss: 0.0023:  76%|███████▋  | 238/312 [24:03<07:28,  6.06s/it]
Training Epoch: 18/100, batch 475/625 completed (loss: 0.004:  76%|███████▋  | 238/312 [24:03<07:28,  6.06s/it] 
Training Epoch: 18/100, batch 475/625 completed (loss: 0.0222:  76%|███████▋  | 238/312 [24:03<07:28,  6.06s/it]
Training Epoch: 18/100, batch 476/625 completed (loss: 0.0111:  76%|███████▋  | 238/312 [24:06<07:28,  6.06s/it]
Training Epoch: 18/100, batch 476/625 completed (loss: 0.0062:  76%|███████▋  | 238/312 [24:06<07:27,  6.05s/it]
Training Epoch: 18/100, batch 476/625 completed (loss: 0.0048:  76%|███████▋  | 238/312 [24:06<07:28,  6.06s/it]
Training Epoch: 18/100, batch 476/625 completed (loss: 0.0043:  76%|███████▋  | 238/312 [24:06<07:28,  6.06s/it]
Training Epoch: 18/100, batch 476/625 completed (loss: 0.0048:  77%|███████▋  | 239/312 [24:09<07:27,  6.14s/it]
Training Epoch: 18/100, batch 476/625 completed (loss: 0.0062:  77%|███████▋  | 239/312 [24:09<07:27,  6.13s/it]
Training Epoch: 18/100, batch 476/625 completed (loss: 0.0043:  77%|███████▋  | 239/312 [24:09<07:28,  6.14s/it]
Training Epoch: 18/100, batch 476/625 completed (loss: 0.0111:  77%|███████▋  | 239/312 [24:09<07:28,  6.14s/it]
Training Epoch: 18/100, batch 477/625 completed (loss: 0.0001:  77%|███████▋  | 239/312 [24:09<07:28,  6.14s/it]
Training Epoch: 18/100, batch 477/625 completed (loss: 0.006:  77%|███████▋  | 239/312 [24:09<07:27,  6.13s/it] 
Training Epoch: 18/100, batch 477/625 completed (loss: 0.0023:  77%|███████▋  | 239/312 [24:09<07:27,  6.14s/it]
Training Epoch: 18/100, batch 477/625 completed (loss: 0.0002:  77%|███████▋  | 239/312 [24:09<07:28,  6.14s/it]
Training Epoch: 18/100, batch 478/625 completed (loss: 0.0908:  77%|███████▋  | 239/312 [24:12<07:28,  6.14s/it]
Training Epoch: 18/100, batch 478/625 completed (loss: 0.0073:  77%|███████▋  | 239/312 [24:12<07:27,  6.13s/it]
Training Epoch: 18/100, batch 478/625 completed (loss: 0.0002:  77%|███████▋  | 239/312 [24:12<07:27,  6.14s/it]
Training Epoch: 18/100, batch 478/625 completed (loss: 0.0006:  77%|███████▋  | 239/312 [24:12<07:28,  6.14s/it]
Training Epoch: 18/100, batch 478/625 completed (loss: 0.0908:  77%|███████▋  | 240/312 [24:15<07:19,  6.10s/it]
Training Epoch: 18/100, batch 478/625 completed (loss: 0.0002:  77%|███████▋  | 240/312 [24:15<07:19,  6.10s/it]
Training Epoch: 18/100, batch 478/625 completed (loss: 0.0006:  77%|███████▋  | 240/312 [24:15<07:19,  6.10s/it]
Training Epoch: 18/100, batch 478/625 completed (loss: 0.0073:  77%|███████▋  | 240/312 [24:15<07:19,  6.10s/it]
Training Epoch: 18/100, batch 479/625 completed (loss: 0.0144:  77%|███████▋  | 240/312 [24:15<07:19,  6.10s/it]
Training Epoch: 18/100, batch 479/625 completed (loss: 0.0056:  77%|███████▋  | 240/312 [24:15<07:19,  6.10s/it]
Training Epoch: 18/100, batch 479/625 completed (loss: 0.0007:  77%|███████▋  | 240/312 [24:15<07:19,  6.10s/it]
Training Epoch: 18/100, batch 479/625 completed (loss: 0.0022:  77%|███████▋  | 240/312 [24:15<07:19,  6.10s/it]
Training Epoch: 18/100, batch 480/625 completed (loss: 0.0001:  77%|███████▋  | 240/312 [24:18<07:19,  6.10s/it]
Training Epoch: 18/100, batch 480/625 completed (loss: 0.0004:  77%|███████▋  | 240/312 [24:18<07:19,  6.10s/it]
Training Epoch: 18/100, batch 480/625 completed (loss: 0.0008:  77%|███████▋  | 240/312 [24:18<07:19,  6.10s/it]
Training Epoch: 18/100, batch 480/625 completed (loss: 0.0759:  77%|███████▋  | 240/312 [24:18<07:19,  6.10s/it]
Training Epoch: 18/100, batch 480/625 completed (loss: 0.0008:  77%|███████▋  | 241/312 [24:21<07:11,  6.08s/it]
Training Epoch: 18/100, batch 480/625 completed (loss: 0.0004:  77%|███████▋  | 241/312 [24:21<07:11,  6.08s/it]
Training Epoch: 18/100, batch 480/625 completed (loss: 0.0759:  77%|███████▋  | 241/312 [24:21<07:11,  6.08s/it]
Training Epoch: 18/100, batch 480/625 completed (loss: 0.0001:  77%|███████▋  | 241/312 [24:21<07:11,  6.08s/it]
Training Epoch: 18/100, batch 481/625 completed (loss: 0.1596:  77%|███████▋  | 241/312 [24:21<07:11,  6.08s/it]
Training Epoch: 18/100, batch 481/625 completed (loss: 0.0206:  77%|███████▋  | 241/312 [24:21<07:11,  6.08s/it]
Training Epoch: 18/100, batch 481/625 completed (loss: 0.0001:  77%|███████▋  | 241/312 [24:21<07:11,  6.08s/it]
Training Epoch: 18/100, batch 481/625 completed (loss: 0.0113:  77%|███████▋  | 241/312 [24:21<07:11,  6.08s/it]
Training Epoch: 18/100, batch 482/625 completed (loss: 0.0201:  77%|███████▋  | 241/312 [24:24<07:11,  6.08s/it]
Training Epoch: 18/100, batch 482/625 completed (loss: 0.0051:  77%|███████▋  | 241/312 [24:24<07:11,  6.08s/it]
Training Epoch: 18/100, batch 482/625 completed (loss: 0.0025:  77%|███████▋  | 241/312 [24:24<07:11,  6.08s/it]
Training Epoch: 18/100, batch 482/625 completed (loss: 0.0011:  77%|███████▋  | 241/312 [24:24<07:11,  6.08s/it]
Training Epoch: 18/100, batch 482/625 completed (loss: 0.0201:  78%|███████▊  | 242/312 [24:27<07:06,  6.10s/it]
Training Epoch: 18/100, batch 482/625 completed (loss: 0.0025:  78%|███████▊  | 242/312 [24:27<07:06,  6.10s/it]
Training Epoch: 18/100, batch 482/625 completed (loss: 0.0011:  78%|███████▊  | 242/312 [24:27<07:06,  6.10s/it]
Training Epoch: 18/100, batch 482/625 completed (loss: 0.0051:  78%|███████▊  | 242/312 [24:27<07:06,  6.10s/it]
Training Epoch: 18/100, batch 483/625 completed (loss: 0.0001:  78%|███████▊  | 242/312 [24:27<07:06,  6.10s/it]
Training Epoch: 18/100, batch 483/625 completed (loss: 0.0222:  78%|███████▊  | 242/312 [24:27<07:06,  6.10s/it]
Training Epoch: 18/100, batch 483/625 completed (loss: 0.0021:  78%|███████▊  | 242/312 [24:27<07:06,  6.10s/it]
Training Epoch: 18/100, batch 483/625 completed (loss: 0.0163:  78%|███████▊  | 242/312 [24:27<07:06,  6.10s/it]
Training Epoch: 18/100, batch 484/625 completed (loss: 0.0351:  78%|███████▊  | 242/312 [24:30<07:06,  6.10s/it]
Training Epoch: 18/100, batch 484/625 completed (loss: 0.0015:  78%|███████▊  | 242/312 [24:30<07:06,  6.10s/it]
Training Epoch: 18/100, batch 484/625 completed (loss: 0.0045:  78%|███████▊  | 242/312 [24:30<07:06,  6.10s/it]
Training Epoch: 18/100, batch 484/625 completed (loss: 0.002:  78%|███████▊  | 242/312 [24:30<07:06,  6.10s/it] 
Training Epoch: 18/100, batch 484/625 completed (loss: 0.002:  78%|███████▊  | 243/312 [24:33<06:59,  6.07s/it]
Training Epoch: 18/100, batch 484/625 completed (loss: 0.0045:  78%|███████▊  | 243/312 [24:33<06:59,  6.07s/it]
Training Epoch: 18/100, batch 484/625 completed (loss: 0.0015:  78%|███████▊  | 243/312 [24:33<06:59,  6.08s/it]
Training Epoch: 18/100, batch 484/625 completed (loss: 0.0351:  78%|███████▊  | 243/312 [24:33<06:59,  6.08s/it]
Training Epoch: 18/100, batch 485/625 completed (loss: 0.003:  78%|███████▊  | 243/312 [24:33<06:59,  6.07s/it] 
Training Epoch: 18/100, batch 485/625 completed (loss: 0.0035:  78%|███████▊  | 243/312 [24:33<06:59,  6.08s/it]
Training Epoch: 18/100, batch 485/625 completed (loss: 0.0015:  78%|███████▊  | 243/312 [24:33<06:59,  6.08s/it]
Training Epoch: 18/100, batch 485/625 completed (loss: 0.0049:  78%|███████▊  | 243/312 [24:33<06:59,  6.07s/it]
Training Epoch: 18/100, batch 486/625 completed (loss: 0.0081:  78%|███████▊  | 243/312 [24:36<06:59,  6.07s/it]
Training Epoch: 18/100, batch 486/625 completed (loss: 0.0005:  78%|███████▊  | 243/312 [24:36<06:59,  6.08s/it]
Training Epoch: 18/100, batch 486/625 completed (loss: 0.0052:  78%|███████▊  | 243/312 [24:36<06:59,  6.08s/it]
Training Epoch: 18/100, batch 486/625 completed (loss: 0.0006:  78%|███████▊  | 243/312 [24:36<06:59,  6.07s/it]
Training Epoch: 18/100, batch 486/625 completed (loss: 0.0052:  78%|███████▊  | 244/312 [24:39<06:56,  6.12s/it]
Training Epoch: 18/100, batch 486/625 completed (loss: 0.0081:  78%|███████▊  | 244/312 [24:39<06:56,  6.12s/it]
Training Epoch: 18/100, batch 486/625 completed (loss: 0.0005:  78%|███████▊  | 244/312 [24:39<06:56,  6.12s/it]
Training Epoch: 18/100, batch 486/625 completed (loss: 0.0006:  78%|███████▊  | 244/312 [24:39<06:56,  6.12s/it]
Training Epoch: 18/100, batch 487/625 completed (loss: 0.0121:  78%|███████▊  | 244/312 [24:39<06:56,  6.12s/it]
Training Epoch: 18/100, batch 487/625 completed (loss: 0.0003:  78%|███████▊  | 244/312 [24:39<06:56,  6.12s/it]
Training Epoch: 18/100, batch 487/625 completed (loss: 0.0011:  78%|███████▊  | 244/312 [24:39<06:56,  6.12s/it]
Training Epoch: 18/100, batch 487/625 completed (loss: 0.0024:  78%|███████▊  | 244/312 [24:39<06:56,  6.12s/it]
Training Epoch: 18/100, batch 488/625 completed (loss: 0.0029:  78%|███████▊  | 244/312 [24:42<06:56,  6.12s/it]
Training Epoch: 18/100, batch 488/625 completed (loss: 0.0117:  78%|███████▊  | 244/312 [24:42<06:56,  6.12s/it]
Training Epoch: 18/100, batch 488/625 completed (loss: 0.002:  78%|███████▊  | 244/312 [24:42<06:56,  6.12s/it] 
Training Epoch: 18/100, batch 488/625 completed (loss: 0.0012:  78%|███████▊  | 244/312 [24:42<06:56,  6.12s/it]
Training Epoch: 18/100, batch 488/625 completed (loss: 0.0029:  79%|███████▊  | 245/312 [24:45<06:48,  6.10s/it]
Training Epoch: 18/100, batch 488/625 completed (loss: 0.002:  79%|███████▊  | 245/312 [24:45<06:48,  6.10s/it]
Training Epoch: 18/100, batch 488/625 completed (loss: 0.0117:  79%|███████▊  | 245/312 [24:45<06:48,  6.10s/it]
Training Epoch: 18/100, batch 488/625 completed (loss: 0.0012:  79%|███████▊  | 245/312 [24:45<06:48,  6.10s/it]
Training Epoch: 18/100, batch 489/625 completed (loss: 0.0064:  79%|███████▊  | 245/312 [24:45<06:48,  6.10s/it]
Training Epoch: 18/100, batch 489/625 completed (loss: 0.0384:  79%|███████▊  | 245/312 [24:45<06:48,  6.10s/it]
Training Epoch: 18/100, batch 489/625 completed (loss: 0.0001:  79%|███████▊  | 245/312 [24:45<06:48,  6.10s/it]
Training Epoch: 18/100, batch 489/625 completed (loss: 0.1062:  79%|███████▊  | 245/312 [24:45<06:48,  6.10s/it]
Training Epoch: 18/100, batch 490/625 completed (loss: 0.0037:  79%|███████▊  | 245/312 [24:48<06:48,  6.10s/it]
Training Epoch: 18/100, batch 490/625 completed (loss: 0.007:  79%|███████▊  | 245/312 [24:48<06:48,  6.10s/it] 
Training Epoch: 18/100, batch 490/625 completed (loss: 0.0017:  79%|███████▊  | 245/312 [24:48<06:48,  6.10s/it]
Training Epoch: 18/100, batch 490/625 completed (loss: 0.0021:  79%|███████▊  | 245/312 [24:48<06:48,  6.10s/it]
Training Epoch: 18/100, batch 490/625 completed (loss: 0.007:  79%|███████▉  | 246/312 [24:51<06:38,  6.04s/it]
Training Epoch: 18/100, batch 490/625 completed (loss: 0.0037:  79%|███████▉  | 246/312 [24:51<06:38,  6.04s/it]
Training Epoch: 18/100, batch 490/625 completed (loss: 0.0021:  79%|███████▉  | 246/312 [24:51<06:38,  6.04s/it]
Training Epoch: 18/100, batch 490/625 completed (loss: 0.0017:  79%|███████▉  | 246/312 [24:51<06:38,  6.04s/it]
Training Epoch: 18/100, batch 491/625 completed (loss: 0.0018:  79%|███████▉  | 246/312 [24:51<06:38,  6.04s/it]
Training Epoch: 18/100, batch 491/625 completed (loss: 0.0018:  79%|███████▉  | 246/312 [24:51<06:38,  6.04s/it]
Training Epoch: 18/100, batch 491/625 completed (loss: 0.0075:  79%|███████▉  | 246/312 [24:51<06:38,  6.04s/it]
Training Epoch: 18/100, batch 491/625 completed (loss: 0.0016:  79%|███████▉  | 246/312 [24:51<06:38,  6.04s/it]
Training Epoch: 18/100, batch 492/625 completed (loss: 0.0044:  79%|███████▉  | 246/312 [24:55<06:38,  6.04s/it]
Training Epoch: 18/100, batch 492/625 completed (loss: 0.0039:  79%|███████▉  | 246/312 [24:55<06:38,  6.04s/it]
Training Epoch: 18/100, batch 492/625 completed (loss: 0.0059:  79%|███████▉  | 246/312 [24:55<06:38,  6.04s/it]
Training Epoch: 18/100, batch 492/625 completed (loss: 0.0001:  79%|███████▉  | 246/312 [24:55<06:38,  6.04s/it]
Training Epoch: 18/100, batch 492/625 completed (loss: 0.0039:  79%|███████▉  | 247/312 [24:58<06:37,  6.11s/it]
Training Epoch: 18/100, batch 492/625 completed (loss: 0.0044:  79%|███████▉  | 247/312 [24:58<06:37,  6.12s/it]
Training Epoch: 18/100, batch 492/625 completed (loss: 0.0001:  79%|███████▉  | 247/312 [24:58<06:37,  6.12s/it]
Training Epoch: 18/100, batch 492/625 completed (loss: 0.0059:  79%|███████▉  | 247/312 [24:58<06:37,  6.12s/it]
Training Epoch: 18/100, batch 493/625 completed (loss: 0.0044:  79%|███████▉  | 247/312 [24:58<06:37,  6.12s/it]
Training Epoch: 18/100, batch 493/625 completed (loss: 0.0002:  79%|███████▉  | 247/312 [24:58<06:37,  6.12s/it]
Training Epoch: 18/100, batch 493/625 completed (loss: 0.0027:  79%|███████▉  | 247/312 [24:58<06:37,  6.11s/it]
Training Epoch: 18/100, batch 493/625 completed (loss: 0.0023:  79%|███████▉  | 247/312 [24:58<06:37,  6.12s/it]
Training Epoch: 18/100, batch 494/625 completed (loss: 0.0041:  79%|███████▉  | 247/312 [25:01<06:37,  6.12s/it]
Training Epoch: 18/100, batch 494/625 completed (loss: 0.0066:  79%|███████▉  | 247/312 [25:01<06:37,  6.12s/it]
Training Epoch: 18/100, batch 494/625 completed (loss: 0.0449:  79%|███████▉  | 247/312 [25:01<06:37,  6.11s/it]
Training Epoch: 18/100, batch 494/625 completed (loss: 0.0003:  79%|███████▉  | 247/312 [25:01<06:37,  6.12s/it]
Training Epoch: 18/100, batch 494/625 completed (loss: 0.0066:  79%|███████▉  | 248/312 [25:04<06:29,  6.09s/it]
Training Epoch: 18/100, batch 494/625 completed (loss: 0.0041:  79%|███████▉  | 248/312 [25:04<06:29,  6.09s/it]
Training Epoch: 18/100, batch 494/625 completed (loss: 0.0003:  79%|███████▉  | 248/312 [25:04<06:29,  6.09s/it]
Training Epoch: 18/100, batch 494/625 completed (loss: 0.0449:  79%|███████▉  | 248/312 [25:04<06:29,  6.09s/it]
Training Epoch: 18/100, batch 495/625 completed (loss: 0.0569:  79%|███████▉  | 248/312 [25:04<06:29,  6.09s/it]
Training Epoch: 18/100, batch 495/625 completed (loss: 0.0043:  79%|███████▉  | 248/312 [25:04<06:29,  6.09s/it]
Training Epoch: 18/100, batch 495/625 completed (loss: 0.0003:  79%|███████▉  | 248/312 [25:04<06:29,  6.09s/it]
Training Epoch: 18/100, batch 495/625 completed (loss: 0.008:  79%|███████▉  | 248/312 [25:04<06:29,  6.09s/it] 
Training Epoch: 18/100, batch 496/625 completed (loss: 0.0054:  79%|███████▉  | 248/312 [25:07<06:29,  6.09s/it]
Training Epoch: 18/100, batch 496/625 completed (loss: 0.0003:  79%|███████▉  | 248/312 [25:07<06:29,  6.09s/it]
Training Epoch: 18/100, batch 496/625 completed (loss: 0.0005:  79%|███████▉  | 248/312 [25:07<06:29,  6.09s/it]
Training Epoch: 18/100, batch 496/625 completed (loss: 0.0034:  79%|███████▉  | 248/312 [25:07<06:29,  6.09s/it]
Training Epoch: 18/100, batch 496/625 completed (loss: 0.0005:  80%|███████▉  | 249/312 [25:10<06:22,  6.07s/it]
Training Epoch: 18/100, batch 496/625 completed (loss: 0.0003:  80%|███████▉  | 249/312 [25:10<06:22,  6.07s/it]
Training Epoch: 18/100, batch 496/625 completed (loss: 0.0054:  80%|███████▉  | 249/312 [25:10<06:22,  6.07s/it]
Training Epoch: 18/100, batch 496/625 completed (loss: 0.0034:  80%|███████▉  | 249/312 [25:10<06:22,  6.07s/it]
Training Epoch: 18/100, batch 497/625 completed (loss: 0.0003:  80%|███████▉  | 249/312 [25:10<06:22,  6.07s/it]
Training Epoch: 18/100, batch 497/625 completed (loss: 0.0064:  80%|███████▉  | 249/312 [25:10<06:22,  6.07s/it]
Training Epoch: 18/100, batch 497/625 completed (loss: 0.0141:  80%|███████▉  | 249/312 [25:10<06:22,  6.07s/it]
Training Epoch: 18/100, batch 497/625 completed (loss: 0.004:  80%|███████▉  | 249/312 [25:10<06:22,  6.07s/it] 
Training Epoch: 18/100, batch 498/625 completed (loss: 0.0014:  80%|███████▉  | 249/312 [25:13<06:22,  6.07s/it]
Training Epoch: 18/100, batch 498/625 completed (loss: 0.0029:  80%|███████▉  | 249/312 [25:13<06:22,  6.07s/it]
Training Epoch: 18/100, batch 498/625 completed (loss: 0.008:  80%|███████▉  | 249/312 [25:13<06:22,  6.07s/it] 
Training Epoch: 18/100, batch 498/625 completed (loss: 0.0035:  80%|███████▉  | 249/312 [25:13<06:22,  6.07s/it]
Training Epoch: 18/100, batch 498/625 completed (loss: 0.0014:  80%|████████  | 250/312 [25:16<06:19,  6.13s/it]
Training Epoch: 18/100, batch 498/625 completed (loss: 0.0035:  80%|████████  | 250/312 [25:16<06:19,  6.13s/it]
Training Epoch: 18/100, batch 498/625 completed (loss: 0.0029:  80%|████████  | 250/312 [25:16<06:20,  6.13s/it]
Training Epoch: 18/100, batch 498/625 completed (loss: 0.008:  80%|████████  | 250/312 [25:16<06:20,  6.13s/it]
Training Epoch: 18/100, batch 499/625 completed (loss: 0.2359:  80%|████████  | 250/312 [25:16<06:20,  6.13s/it]
Training Epoch: 18/100, batch 499/625 completed (loss: 0.0003:  80%|████████  | 250/312 [25:16<06:19,  6.13s/it]
Training Epoch: 18/100, batch 499/625 completed (loss: 0.0346:  80%|████████  | 250/312 [25:16<06:20,  6.13s/it]
Training Epoch: 18/100, batch 499/625 completed (loss: 0.0012:  80%|████████  | 250/312 [25:16<06:19,  6.13s/it]
Training Epoch: 18/100, batch 500/625 completed (loss: 0.0002:  80%|████████  | 250/312 [25:19<06:20,  6.13s/it]
Training Epoch: 18/100, batch 500/625 completed (loss: 0.0186:  80%|████████  | 250/312 [25:19<06:19,  6.13s/it]
Training Epoch: 18/100, batch 500/625 completed (loss: 0.0005:  80%|████████  | 250/312 [25:19<06:20,  6.13s/it]
Training Epoch: 18/100, batch 500/625 completed (loss: 0.0043:  80%|████████  | 250/312 [25:19<06:19,  6.13s/it]
Training Epoch: 18/100, batch 500/625 completed (loss: 0.0043:  80%|████████  | 251/312 [25:22<06:09,  6.05s/it]
Training Epoch: 18/100, batch 500/625 completed (loss: 0.0005:  80%|████████  | 251/312 [25:22<06:09,  6.05s/it]
Training Epoch: 18/100, batch 500/625 completed (loss: 0.0186:  80%|████████  | 251/312 [25:22<06:09,  6.05s/it]
Training Epoch: 18/100, batch 500/625 completed (loss: 0.0002:  80%|████████  | 251/312 [25:22<06:09,  6.06s/it]
Training Epoch: 18/100, batch 501/625 completed (loss: 0.0004:  80%|████████  | 251/312 [25:22<06:09,  6.05s/it]
Training Epoch: 18/100, batch 501/625 completed (loss: 0.0067:  80%|████████  | 251/312 [25:22<06:09,  6.05s/it]
Training Epoch: 18/100, batch 501/625 completed (loss: 0.0075:  80%|████████  | 251/312 [25:22<06:09,  6.06s/it]
Training Epoch: 18/100, batch 501/625 completed (loss: 0.009:  80%|████████  | 251/312 [25:22<06:09,  6.05s/it] 
Training Epoch: 18/100, batch 502/625 completed (loss: 0.0189:  80%|████████  | 251/312 [25:25<06:09,  6.05s/it]
Training Epoch: 18/100, batch 502/625 completed (loss: 0.0015:  80%|████████  | 251/312 [25:25<06:09,  6.05s/it]
Training Epoch: 18/100, batch 502/625 completed (loss: 0.0019:  80%|████████  | 251/312 [25:25<06:09,  6.06s/it]
Training Epoch: 18/100, batch 502/625 completed (loss: 0.0:  80%|████████  | 251/312 [25:25<06:09,  6.05s/it]  
Training Epoch: 18/100, batch 502/625 completed (loss: 0.0019:  81%|████████  | 252/312 [25:28<06:08,  6.13s/it]
Training Epoch: 18/100, batch 502/625 completed (loss: 0.0015:  81%|████████  | 252/312 [25:28<06:08,  6.14s/it]
Training Epoch: 18/100, batch 502/625 completed (loss: 0.0189:  81%|████████  | 252/312 [25:28<06:08,  6.14s/it]
Training Epoch: 18/100, batch 502/625 completed (loss: 0.0:  81%|████████  | 252/312 [25:28<06:08,  6.14s/it]
Training Epoch: 18/100, batch 503/625 completed (loss: 0.0094:  81%|████████  | 252/312 [25:28<06:08,  6.14s/it]
Training Epoch: 18/100, batch 503/625 completed (loss: 0.0018:  81%|████████  | 252/312 [25:28<06:08,  6.13s/it]
Training Epoch: 18/100, batch 503/625 completed (loss: 0.0029:  81%|████████  | 252/312 [25:28<06:08,  6.14s/it]
Training Epoch: 18/100, batch 503/625 completed (loss: 0.0077:  81%|████████  | 252/312 [25:28<06:08,  6.14s/it]
Training Epoch: 18/100, batch 504/625 completed (loss: 0.0286:  81%|████████  | 252/312 [25:31<06:08,  6.14s/it]
Training Epoch: 18/100, batch 504/625 completed (loss: 0.0036:  81%|████████  | 252/312 [25:31<06:08,  6.13s/it]
Training Epoch: 18/100, batch 504/625 completed (loss: 0.0096:  81%|████████  | 252/312 [25:31<06:08,  6.14s/it]
Training Epoch: 18/100, batch 504/625 completed (loss: 0.0002:  81%|████████  | 252/312 [25:31<06:08,  6.14s/it]
Training Epoch: 18/100, batch 504/625 completed (loss: 0.0286:  81%|████████  | 253/312 [25:34<06:00,  6.11s/it]
Training Epoch: 18/100, batch 504/625 completed (loss: 0.0002:  81%|████████  | 253/312 [25:34<06:00,  6.11s/it]
Training Epoch: 18/100, batch 504/625 completed (loss: 0.0096:  81%|████████  | 253/312 [25:34<06:00,  6.11s/it]
Training Epoch: 18/100, batch 504/625 completed (loss: 0.0036:  81%|████████  | 253/312 [25:34<06:00,  6.11s/it]
Training Epoch: 18/100, batch 505/625 completed (loss: 0.0003:  81%|████████  | 253/312 [25:34<06:00,  6.11s/it]
Training Epoch: 18/100, batch 505/625 completed (loss: 0.0237:  81%|████████  | 253/312 [25:34<06:00,  6.11s/it]
Training Epoch: 18/100, batch 505/625 completed (loss: 0.0013:  81%|████████  | 253/312 [25:34<06:00,  6.11s/it]
Training Epoch: 18/100, batch 505/625 completed (loss: 0.0058:  81%|████████  | 253/312 [25:34<06:00,  6.11s/it]
Training Epoch: 18/100, batch 506/625 completed (loss: 0.0009:  81%|████████  | 253/312 [25:37<06:00,  6.11s/it]
Training Epoch: 18/100, batch 506/625 completed (loss: 0.0012:  81%|████████  | 253/312 [25:37<06:00,  6.11s/it]
Training Epoch: 18/100, batch 506/625 completed (loss: 0.2053:  81%|████████  | 253/312 [25:37<06:00,  6.11s/it]
Training Epoch: 18/100, batch 506/625 completed (loss: 0.0203:  81%|████████  | 253/312 [25:37<06:00,  6.11s/it]
Training Epoch: 18/100, batch 506/625 completed (loss: 0.2053:  81%|████████▏ | 254/312 [25:40<05:52,  6.08s/it]
Training Epoch: 18/100, batch 506/625 completed (loss: 0.0203:  81%|████████▏ | 254/312 [25:40<05:52,  6.08s/it]
Training Epoch: 18/100, batch 506/625 completed (loss: 0.0009:  81%|████████▏ | 254/312 [25:40<05:52,  6.09s/it]
Training Epoch: 18/100, batch 506/625 completed (loss: 0.0012:  81%|████████▏ | 254/312 [25:40<05:53,  6.09s/it]
Training Epoch: 18/100, batch 507/625 completed (loss: 0.004:  81%|████████▏ | 254/312 [25:40<05:53,  6.09s/it] 
Training Epoch: 18/100, batch 507/625 completed (loss: 0.0001:  81%|████████▏ | 254/312 [25:40<05:52,  6.08s/it]
Training Epoch: 18/100, batch 507/625 completed (loss: 0.0338:  81%|████████▏ | 254/312 [25:40<05:52,  6.09s/it]
Training Epoch: 18/100, batch 507/625 completed (loss: 0.0003:  81%|████████▏ | 254/312 [25:40<05:52,  6.08s/it]
Training Epoch: 18/100, batch 508/625 completed (loss: 0.0229:  81%|████████▏ | 254/312 [25:43<05:52,  6.09s/it]
Training Epoch: 18/100, batch 508/625 completed (loss: 0.0002:  81%|████████▏ | 254/312 [25:43<05:53,  6.09s/it]
Training Epoch: 18/100, batch 508/625 completed (loss: 0.001:  81%|████████▏ | 254/312 [25:43<05:52,  6.08s/it] 
Training Epoch: 18/100, batch 508/625 completed (loss: 0.0003:  81%|████████▏ | 254/312 [25:44<05:52,  6.08s/it]
Training Epoch: 18/100, batch 508/625 completed (loss: 0.0002:  82%|████████▏ | 255/312 [25:46<05:47,  6.10s/it]
Training Epoch: 18/100, batch 508/625 completed (loss: 0.0003:  82%|████████▏ | 255/312 [25:46<05:48,  6.11s/it]
Training Epoch: 18/100, batch 508/625 completed (loss: 0.0229:  82%|████████▏ | 255/312 [25:46<05:48,  6.11s/it]
Training Epoch: 18/100, batch 508/625 completed (loss: 0.001:  82%|████████▏ | 255/312 [25:46<05:48,  6.11s/it]
Training Epoch: 18/100, batch 509/625 completed (loss: 0.0026:  82%|████████▏ | 255/312 [25:46<05:48,  6.11s/it]
Training Epoch: 18/100, batch 509/625 completed (loss: 0.0102:  82%|████████▏ | 255/312 [25:46<05:47,  6.10s/it]
Training Epoch: 18/100, batch 509/625 completed (loss: 0.0078:  82%|████████▏ | 255/312 [25:46<05:48,  6.11s/it]
Training Epoch: 18/100, batch 509/625 completed (loss: 0.0079:  82%|████████▏ | 255/312 [25:46<05:48,  6.11s/it]
Training Epoch: 18/100, batch 510/625 completed (loss: 0.0042:  82%|████████▏ | 255/312 [25:49<05:48,  6.11s/it]
Training Epoch: 18/100, batch 510/625 completed (loss: 0.0061:  82%|████████▏ | 255/312 [25:49<05:48,  6.11s/it]
Training Epoch: 18/100, batch 510/625 completed (loss: 0.0033:  82%|████████▏ | 255/312 [25:49<05:47,  6.10s/it]
Training Epoch: 18/100, batch 510/625 completed (loss: 0.0029:  82%|████████▏ | 255/312 [25:50<05:48,  6.11s/it]
Training Epoch: 18/100, batch 510/625 completed (loss: 0.0042:  82%|████████▏ | 256/312 [25:52<05:40,  6.08s/it]
Training Epoch: 18/100, batch 510/625 completed (loss: 0.0061:  82%|████████▏ | 256/312 [25:52<05:40,  6.08s/it]
Training Epoch: 18/100, batch 510/625 completed (loss: 0.0029:  82%|████████▏ | 256/312 [25:52<05:40,  6.08s/it]
Training Epoch: 18/100, batch 510/625 completed (loss: 0.0033:  82%|████████▏ | 256/312 [25:52<05:40,  6.08s/it]
Training Epoch: 18/100, batch 511/625 completed (loss: 0.0001:  82%|████████▏ | 256/312 [25:52<05:40,  6.08s/it]
Training Epoch: 18/100, batch 511/625 completed (loss: 0.001:  82%|████████▏ | 256/312 [25:52<05:40,  6.08s/it] 
Training Epoch: 18/100, batch 511/625 completed (loss: 0.0001:  82%|████████▏ | 256/312 [25:52<05:40,  6.08s/it]
Training Epoch: 18/100, batch 511/625 completed (loss: 0.0002:  82%|████████▏ | 256/312 [25:52<05:40,  6.08s/it]
Training Epoch: 18/100, batch 512/625 completed (loss: 0.0014:  82%|████████▏ | 256/312 [25:55<05:40,  6.08s/it]
Training Epoch: 18/100, batch 512/625 completed (loss: 0.002:  82%|████████▏ | 256/312 [25:55<05:40,  6.08s/it]
Training Epoch: 18/100, batch 512/625 completed (loss: 0.0032:  82%|████████▏ | 256/312 [25:55<05:40,  6.08s/it]
Training Epoch: 18/100, batch 512/625 completed (loss: 0.0008:  82%|████████▏ | 256/312 [25:55<05:40,  6.08s/it]
Training Epoch: 18/100, batch 512/625 completed (loss: 0.002:  82%|████████▏ | 257/312 [25:58<05:35,  6.09s/it]
Training Epoch: 18/100, batch 512/625 completed (loss: 0.0014:  82%|████████▏ | 257/312 [25:58<05:35,  6.09s/it]
Training Epoch: 18/100, batch 512/625 completed (loss: 0.0032:  82%|████████▏ | 257/312 [25:58<05:35,  6.10s/it]
Training Epoch: 18/100, batch 512/625 completed (loss: 0.0008:  82%|████████▏ | 257/312 [25:59<05:35,  6.10s/it]
Training Epoch: 18/100, batch 513/625 completed (loss: 0.0031:  82%|████████▏ | 257/312 [25:59<05:35,  6.09s/it]
Training Epoch: 18/100, batch 513/625 completed (loss: 0.0004:  82%|████████▏ | 257/312 [25:59<05:35,  6.09s/it]
Training Epoch: 18/100, batch 513/625 completed (loss: 0.0009:  82%|████████▏ | 257/312 [25:59<05:35,  6.10s/it]
Training Epoch: 18/100, batch 513/625 completed (loss: 0.0011:  82%|████████▏ | 257/312 [25:59<05:35,  6.10s/it]
Training Epoch: 18/100, batch 514/625 completed (loss: 0.0046:  82%|████████▏ | 257/312 [26:02<05:35,  6.09s/it]
Training Epoch: 18/100, batch 514/625 completed (loss: 0.0005:  82%|████████▏ | 257/312 [26:02<05:35,  6.09s/it]
Training Epoch: 18/100, batch 514/625 completed (loss: 0.001:  82%|████████▏ | 257/312 [26:02<05:35,  6.10s/it] 
Training Epoch: 18/100, batch 514/625 completed (loss: 0.0001:  82%|████████▏ | 257/312 [26:02<05:35,  6.10s/it]
Training Epoch: 18/100, batch 514/625 completed (loss: 0.0001:  83%|████████▎ | 258/312 [26:05<05:30,  6.12s/it]
Training Epoch: 18/100, batch 514/625 completed (loss: 0.001:  83%|████████▎ | 258/312 [26:05<05:30,  6.13s/it]
Training Epoch: 18/100, batch 514/625 completed (loss: 0.0046:  83%|████████▎ | 258/312 [26:05<05:30,  6.13s/it]
Training Epoch: 18/100, batch 514/625 completed (loss: 0.0005:  83%|████████▎ | 258/312 [26:05<05:30,  6.13s/it]
Training Epoch: 18/100, batch 515/625 completed (loss: 0.0256:  83%|████████▎ | 258/312 [26:05<05:30,  6.13s/it]
Training Epoch: 18/100, batch 515/625 completed (loss: 0.3401:  83%|████████▎ | 258/312 [26:05<05:30,  6.13s/it]
Training Epoch: 18/100, batch 515/625 completed (loss: 0.0061:  83%|████████▎ | 258/312 [26:05<05:30,  6.13s/it]
Training Epoch: 18/100, batch 515/625 completed (loss: 0.013:  83%|████████▎ | 258/312 [26:05<05:30,  6.12s/it] 
Training Epoch: 18/100, batch 516/625 completed (loss: 0.0095:  83%|████████▎ | 258/312 [26:08<05:30,  6.13s/it]
Training Epoch: 18/100, batch 516/625 completed (loss: 0.0015:  83%|████████▎ | 258/312 [26:08<05:30,  6.13s/it]
Training Epoch: 18/100, batch 516/625 completed (loss: 0.0092:  83%|████████▎ | 258/312 [26:08<05:30,  6.13s/it]
Training Epoch: 18/100, batch 516/625 completed (loss: 0.0315:  83%|████████▎ | 258/312 [26:08<05:30,  6.12s/it]
Training Epoch: 18/100, batch 516/625 completed (loss: 0.0015:  83%|████████▎ | 259/312 [26:11<05:20,  6.05s/it]
Training Epoch: 18/100, batch 516/625 completed (loss: 0.0095:  83%|████████▎ | 259/312 [26:11<05:20,  6.06s/it]
Training Epoch: 18/100, batch 516/625 completed (loss: 0.0315:  83%|████████▎ | 259/312 [26:11<05:21,  6.06s/it]
Training Epoch: 18/100, batch 516/625 completed (loss: 0.0092:  83%|████████▎ | 259/312 [26:11<05:21,  6.06s/it]
Training Epoch: 18/100, batch 517/625 completed (loss: 0.0021:  83%|████████▎ | 259/312 [26:11<05:20,  6.05s/it]
Training Epoch: 18/100, batch 517/625 completed (loss: 0.0007:  83%|████████▎ | 259/312 [26:11<05:21,  6.06s/it]
Training Epoch: 18/100, batch 517/625 completed (loss: 0.0004:  83%|████████▎ | 259/312 [26:11<05:20,  6.06s/it]
Training Epoch: 18/100, batch 517/625 completed (loss: 0.0198:  83%|████████▎ | 259/312 [26:11<05:21,  6.06s/it]
Training Epoch: 18/100, batch 518/625 completed (loss: 0.0047:  83%|████████▎ | 259/312 [26:14<05:20,  6.06s/it]
Training Epoch: 18/100, batch 518/625 completed (loss: 0.0035:  83%|████████▎ | 259/312 [26:14<05:21,  6.06s/it]
Training Epoch: 18/100, batch 518/625 completed (loss: 0.0011:  83%|████████▎ | 259/312 [26:14<05:20,  6.05s/it]
Training Epoch: 18/100, batch 518/625 completed (loss: 0.0031:  83%|████████▎ | 259/312 [26:14<05:21,  6.06s/it]
Training Epoch: 18/100, batch 518/625 completed (loss: 0.0047:  83%|████████▎ | 260/312 [26:17<05:16,  6.09s/it]
Training Epoch: 18/100, batch 518/625 completed (loss: 0.0035:  83%|████████▎ | 260/312 [26:17<05:16,  6.09s/it]
Training Epoch: 18/100, batch 518/625 completed (loss: 0.0011:  83%|████████▎ | 260/312 [26:17<05:16,  6.09s/it]
Training Epoch: 18/100, batch 518/625 completed (loss: 0.0031:  83%|████████▎ | 260/312 [26:17<05:16,  6.09s/it]
Training Epoch: 18/100, batch 519/625 completed (loss: 0.0025:  83%|████████▎ | 260/312 [26:17<05:16,  6.09s/it]
Training Epoch: 18/100, batch 519/625 completed (loss: 0.0014:  83%|████████▎ | 260/312 [26:17<05:16,  6.09s/it]
Training Epoch: 18/100, batch 519/625 completed (loss: 0.0005:  83%|████████▎ | 260/312 [26:17<05:16,  6.09s/it]
Training Epoch: 18/100, batch 519/625 completed (loss: 0.0021:  83%|████████▎ | 260/312 [26:17<05:16,  6.09s/it]
Training Epoch: 18/100, batch 520/625 completed (loss: 0.0001:  83%|████████▎ | 260/312 [26:20<05:16,  6.09s/it]
Training Epoch: 18/100, batch 520/625 completed (loss: 0.0068:  83%|████████▎ | 260/312 [26:20<05:16,  6.09s/it]
Training Epoch: 18/100, batch 520/625 completed (loss: 0.003:  83%|████████▎ | 260/312 [26:20<05:16,  6.09s/it] 
Training Epoch: 18/100, batch 520/625 completed (loss: 0.0001:  83%|████████▎ | 260/312 [26:20<05:16,  6.09s/it]
Training Epoch: 18/100, batch 520/625 completed (loss: 0.0001:  84%|████████▎ | 261/312 [26:23<05:11,  6.11s/it]
Training Epoch: 18/100, batch 520/625 completed (loss: 0.003:  84%|████████▎ | 261/312 [26:23<05:11,  6.11s/it]
Training Epoch: 18/100, batch 520/625 completed (loss: 0.0001:  84%|████████▎ | 261/312 [26:23<05:11,  6.11s/it]
Training Epoch: 18/100, batch 520/625 completed (loss: 0.0068:  84%|████████▎ | 261/312 [26:23<05:12,  6.12s/it]
Training Epoch: 18/100, batch 521/625 completed (loss: 0.0349:  84%|████████▎ | 261/312 [26:23<05:12,  6.12s/it]
Training Epoch: 18/100, batch 521/625 completed (loss: 0.003:  84%|████████▎ | 261/312 [26:23<05:11,  6.11s/it]
Training Epoch: 18/100, batch 521/625 completed (loss: 0.0007:  84%|████████▎ | 261/312 [26:23<05:11,  6.11s/it]
Training Epoch: 18/100, batch 521/625 completed (loss: 0.022:  84%|████████▎ | 261/312 [26:23<05:11,  6.11s/it] 
Training Epoch: 18/100, batch 522/625 completed (loss: 0.1887:  84%|████████▎ | 261/312 [26:26<05:12,  6.12s/it]
Training Epoch: 18/100, batch 522/625 completed (loss: 0.0021:  84%|████████▎ | 261/312 [26:26<05:11,  6.11s/it]
Training Epoch: 18/100, batch 522/625 completed (loss: 0.0016:  84%|████████▎ | 261/312 [26:26<05:11,  6.11s/it]
Training Epoch: 18/100, batch 522/625 completed (loss: 0.0006:  84%|████████▎ | 261/312 [26:26<05:11,  6.11s/it]
Training Epoch: 18/100, batch 522/625 completed (loss: 0.0021:  84%|████████▍ | 262/312 [26:29<05:06,  6.13s/it]
Training Epoch: 18/100, batch 522/625 completed (loss: 0.0006:  84%|████████▍ | 262/312 [26:29<05:06,  6.13s/it]
Training Epoch: 18/100, batch 522/625 completed (loss: 0.0016:  84%|████████▍ | 262/312 [26:29<05:06,  6.13s/it]
Training Epoch: 18/100, batch 522/625 completed (loss: 0.1887:  84%|████████▍ | 262/312 [26:29<05:06,  6.13s/it]
Training Epoch: 18/100, batch 523/625 completed (loss: 0.002:  84%|████████▍ | 262/312 [26:29<05:06,  6.13s/it] 
Training Epoch: 18/100, batch 523/625 completed (loss: 0.0004:  84%|████████▍ | 262/312 [26:29<05:06,  6.13s/it]
Training Epoch: 18/100, batch 523/625 completed (loss: 0.0033:  84%|████████▍ | 262/312 [26:29<05:06,  6.13s/it]
Training Epoch: 18/100, batch 523/625 completed (loss: 0.0008:  84%|████████▍ | 262/312 [26:29<05:06,  6.13s/it]
Training Epoch: 18/100, batch 524/625 completed (loss: 0.0018:  84%|████████▍ | 262/312 [26:32<05:06,  6.13s/it]
Training Epoch: 18/100, batch 524/625 completed (loss: 0.0021:  84%|████████▍ | 262/312 [26:32<05:06,  6.13s/it]
Training Epoch: 18/100, batch 524/625 completed (loss: 0.004:  84%|████████▍ | 262/312 [26:32<05:06,  6.13s/it] 
Training Epoch: 18/100, batch 524/625 completed (loss: 0.0036:  84%|████████▍ | 262/312 [26:32<05:06,  6.13s/it]
Training Epoch: 18/100, batch 524/625 completed (loss: 0.004:  84%|████████▍ | 263/312 [26:35<05:01,  6.16s/it]
Training Epoch: 18/100, batch 524/625 completed (loss: 0.0036:  84%|████████▍ | 263/312 [26:35<05:01,  6.16s/it]
Training Epoch: 18/100, batch 524/625 completed (loss: 0.0018:  84%|████████▍ | 263/312 [26:35<05:01,  6.16s/it]
Training Epoch: 18/100, batch 524/625 completed (loss: 0.0021:  84%|████████▍ | 263/312 [26:35<05:02,  6.17s/it]
Training Epoch: 18/100, batch 525/625 completed (loss: 0.0005:  84%|████████▍ | 263/312 [26:35<05:01,  6.16s/it]
Training Epoch: 18/100, batch 525/625 completed (loss: 0.0121:  84%|████████▍ | 263/312 [26:35<05:01,  6.16s/it]
Training Epoch: 18/100, batch 525/625 completed (loss: 0.0022:  84%|████████▍ | 263/312 [26:35<05:02,  6.17s/it]
Training Epoch: 18/100, batch 525/625 completed (loss: 0.0054:  84%|████████▍ | 263/312 [26:35<05:01,  6.16s/it]
Training Epoch: 18/100, batch 526/625 completed (loss: 0.0015:  84%|████████▍ | 263/312 [26:38<05:01,  6.16s/it]
Training Epoch: 18/100, batch 526/625 completed (loss: 0.0007:  84%|████████▍ | 263/312 [26:38<05:01,  6.16s/it]
Training Epoch: 18/100, batch 526/625 completed (loss: 0.0132:  84%|████████▍ | 263/312 [26:38<05:02,  6.17s/it]
Training Epoch: 18/100, batch 526/625 completed (loss: 0.0036:  84%|████████▍ | 263/312 [26:38<05:01,  6.16s/it]
Training Epoch: 18/100, batch 526/625 completed (loss: 0.0007:  85%|████████▍ | 264/312 [26:41<04:51,  6.08s/it]
Training Epoch: 18/100, batch 526/625 completed (loss: 0.0015:  85%|████████▍ | 264/312 [26:41<04:51,  6.08s/it]
Training Epoch: 18/100, batch 526/625 completed (loss: 0.0036:  85%|████████▍ | 264/312 [26:41<04:51,  6.08s/it]
Training Epoch: 18/100, batch 526/625 completed (loss: 0.0132:  85%|████████▍ | 264/312 [26:41<04:52,  6.09s/it]
Training Epoch: 18/100, batch 527/625 completed (loss: 0.0003:  85%|████████▍ | 264/312 [26:41<04:52,  6.09s/it]
Training Epoch: 18/100, batch 527/625 completed (loss: 0.0009:  85%|████████▍ | 264/312 [26:41<04:51,  6.08s/it]
Training Epoch: 18/100, batch 527/625 completed (loss: 0.0014:  85%|████████▍ | 264/312 [26:41<04:51,  6.08s/it]
Training Epoch: 18/100, batch 527/625 completed (loss: 0.0036:  85%|████████▍ | 264/312 [26:41<04:51,  6.08s/it]
Training Epoch: 18/100, batch 528/625 completed (loss: 0.001:  85%|████████▍ | 264/312 [26:44<04:51,  6.08s/it] 
Training Epoch: 18/100, batch 528/625 completed (loss: 0.0005:  85%|████████▍ | 264/312 [26:44<04:52,  6.09s/it]
Training Epoch: 18/100, batch 528/625 completed (loss: 0.007:  85%|████████▍ | 264/312 [26:44<04:51,  6.08s/it] 
Training Epoch: 18/100, batch 528/625 completed (loss: 0.0031:  85%|████████▍ | 264/312 [26:44<04:51,  6.08s/it]
Training Epoch: 18/100, batch 528/625 completed (loss: 0.007:  85%|████████▍ | 265/312 [26:47<04:46,  6.11s/it]
Training Epoch: 18/100, batch 528/625 completed (loss: 0.001:  85%|████████▍ | 265/312 [26:47<04:46,  6.10s/it]
Training Epoch: 18/100, batch 528/625 completed (loss: 0.0005:  85%|████████▍ | 265/312 [26:47<04:46,  6.10s/it]
Training Epoch: 18/100, batch 528/625 completed (loss: 0.0031:  85%|████████▍ | 265/312 [26:47<04:47,  6.11s/it]
Training Epoch: 18/100, batch 529/625 completed (loss: 0.0015:  85%|████████▍ | 265/312 [26:47<04:46,  6.11s/it]
Training Epoch: 18/100, batch 529/625 completed (loss: 0.0006:  85%|████████▍ | 265/312 [26:47<04:46,  6.10s/it]
Training Epoch: 18/100, batch 529/625 completed (loss: 0.0003:  85%|████████▍ | 265/312 [26:47<04:46,  6.10s/it]
Training Epoch: 18/100, batch 529/625 completed (loss: 0.0179:  85%|████████▍ | 265/312 [26:47<04:47,  6.11s/it]
Training Epoch: 18/100, batch 530/625 completed (loss: 0.0027:  85%|████████▍ | 265/312 [26:50<04:46,  6.10s/it]
Training Epoch: 18/100, batch 530/625 completed (loss: 0.0098:  85%|████████▍ | 265/312 [26:51<04:46,  6.10s/it]
Training Epoch: 18/100, batch 530/625 completed (loss: 0.1165:  85%|████████▍ | 265/312 [26:51<04:46,  6.11s/it]
Training Epoch: 18/100, batch 530/625 completed (loss: 0.0005:  85%|████████▍ | 265/312 [26:51<04:47,  6.11s/it]
Training Epoch: 18/100, batch 530/625 completed (loss: 0.0027:  85%|████████▌ | 266/312 [26:54<04:41,  6.12s/it]
Training Epoch: 18/100, batch 530/625 completed (loss: 0.0098:  85%|████████▌ | 266/312 [26:54<04:41,  6.12s/it]
Training Epoch: 18/100, batch 530/625 completed (loss: 0.0005:  85%|████████▌ | 266/312 [26:54<04:41,  6.12s/it]
Training Epoch: 18/100, batch 530/625 completed (loss: 0.1165:  85%|████████▌ | 266/312 [26:54<04:41,  6.12s/it]
Training Epoch: 18/100, batch 531/625 completed (loss: 0.0554:  85%|████████▌ | 266/312 [26:54<04:41,  6.12s/it]
Training Epoch: 18/100, batch 531/625 completed (loss: 0.0186:  85%|████████▌ | 266/312 [26:54<04:41,  6.12s/it]
Training Epoch: 18/100, batch 531/625 completed (loss: 0.0009:  85%|████████▌ | 266/312 [26:54<04:41,  6.12s/it]
Training Epoch: 18/100, batch 531/625 completed (loss: 0.0147:  85%|████████▌ | 266/312 [26:54<04:41,  6.12s/it]
Training Epoch: 18/100, batch 532/625 completed (loss: 0.0002:  85%|████████▌ | 266/312 [26:57<04:41,  6.12s/it]
Training Epoch: 18/100, batch 532/625 completed (loss: 0.0206:  85%|████████▌ | 266/312 [26:57<04:41,  6.12s/it]
Training Epoch: 18/100, batch 532/625 completed (loss: 0.0003:  85%|████████▌ | 266/312 [26:57<04:41,  6.12s/it]
Training Epoch: 18/100, batch 532/625 completed (loss: 0.0005:  85%|████████▌ | 266/312 [26:57<04:41,  6.12s/it]
Training Epoch: 18/100, batch 532/625 completed (loss: 0.0206:  86%|████████▌ | 267/312 [27:00<04:33,  6.08s/it]
Training Epoch: 18/100, batch 532/625 completed (loss: 0.0002:  86%|████████▌ | 267/312 [27:00<04:33,  6.08s/it]
Training Epoch: 18/100, batch 532/625 completed (loss: 0.0005:  86%|████████▌ | 267/312 [27:00<04:33,  6.09s/it]
Training Epoch: 18/100, batch 532/625 completed (loss: 0.0003:  86%|████████▌ | 267/312 [27:00<04:33,  6.09s/it]
Training Epoch: 18/100, batch 533/625 completed (loss: 0.0044:  86%|████████▌ | 267/312 [27:00<04:33,  6.08s/it]
Training Epoch: 18/100, batch 533/625 completed (loss: 0.0031:  86%|████████▌ | 267/312 [27:00<04:33,  6.08s/it]
Training Epoch: 18/100, batch 533/625 completed (loss: 0.0114:  86%|████████▌ | 267/312 [27:00<04:33,  6.09s/it]
Training Epoch: 18/100, batch 533/625 completed (loss: 0.0006:  86%|████████▌ | 267/312 [27:00<04:33,  6.09s/it]
Training Epoch: 18/100, batch 534/625 completed (loss: 0.001:  86%|████████▌ | 267/312 [27:03<04:33,  6.08s/it] 
Training Epoch: 18/100, batch 534/625 completed (loss: 0.003:  86%|████████▌ | 267/312 [27:03<04:33,  6.08s/it] 
Training Epoch: 18/100, batch 534/625 completed (loss: 0.0004:  86%|████████▌ | 267/312 [27:03<04:33,  6.09s/it]
Training Epoch: 18/100, batch 534/625 completed (loss: 0.0019:  86%|████████▌ | 267/312 [27:03<04:33,  6.09s/it]
Training Epoch: 18/100, batch 534/625 completed (loss: 0.0019:  86%|████████▌ | 268/312 [27:06<04:26,  6.06s/it]
Training Epoch: 18/100, batch 534/625 completed (loss: 0.0004:  86%|████████▌ | 268/312 [27:06<04:26,  6.06s/it]
Training Epoch: 18/100, batch 534/625 completed (loss: 0.003:  86%|████████▌ | 268/312 [27:06<04:26,  6.06s/it]
Training Epoch: 18/100, batch 534/625 completed (loss: 0.001:  86%|████████▌ | 268/312 [27:06<04:26,  6.07s/it]
Training Epoch: 18/100, batch 535/625 completed (loss: 0.0028:  86%|████████▌ | 268/312 [27:06<04:26,  6.06s/it]
Training Epoch: 18/100, batch 535/625 completed (loss: 0.0226:  86%|████████▌ | 268/312 [27:06<04:26,  6.06s/it]
Training Epoch: 18/100, batch 535/625 completed (loss: 0.0409:  86%|████████▌ | 268/312 [27:06<04:26,  6.07s/it]
Training Epoch: 18/100, batch 535/625 completed (loss: 0.0057:  86%|████████▌ | 268/312 [27:06<04:26,  6.06s/it]
Training Epoch: 18/100, batch 536/625 completed (loss: 0.0146:  86%|████████▌ | 268/312 [27:09<04:26,  6.06s/it]
Training Epoch: 18/100, batch 536/625 completed (loss: 0.0005:  86%|████████▌ | 268/312 [27:09<04:26,  6.06s/it]
Training Epoch: 18/100, batch 536/625 completed (loss: 0.0038:  86%|████████▌ | 268/312 [27:09<04:26,  6.07s/it]
Training Epoch: 18/100, batch 536/625 completed (loss: 0.0013:  86%|████████▌ | 268/312 [27:09<04:26,  6.06s/it]
Training Epoch: 18/100, batch 536/625 completed (loss: 0.0146:  86%|████████▌ | 269/312 [27:12<04:22,  6.10s/it]
Training Epoch: 18/100, batch 536/625 completed (loss: 0.0038:  86%|████████▌ | 269/312 [27:12<04:22,  6.10s/it]
Training Epoch: 18/100, batch 536/625 completed (loss: 0.0005:  86%|████████▌ | 269/312 [27:12<04:22,  6.10s/it]
Training Epoch: 18/100, batch 536/625 completed (loss: 0.0013:  86%|████████▌ | 269/312 [27:12<04:22,  6.11s/it]
Training Epoch: 18/100, batch 537/625 completed (loss: 0.0197:  86%|████████▌ | 269/312 [27:12<04:22,  6.10s/it]
Training Epoch: 18/100, batch 537/625 completed (loss: 0.0006:  86%|████████▌ | 269/312 [27:12<04:22,  6.10s/it]
Training Epoch: 18/100, batch 537/625 completed (loss: 0.0051:  86%|████████▌ | 269/312 [27:12<04:22,  6.10s/it]
Training Epoch: 18/100, batch 537/625 completed (loss: 0.0004:  86%|████████▌ | 269/312 [27:12<04:22,  6.11s/it]
Training Epoch: 18/100, batch 538/625 completed (loss: 0.0013:  86%|████████▌ | 269/312 [27:15<04:22,  6.10s/it]
Training Epoch: 18/100, batch 538/625 completed (loss: 0.0017:  86%|████████▌ | 269/312 [27:15<04:22,  6.10s/it]
Training Epoch: 18/100, batch 538/625 completed (loss: 0.0207:  86%|████████▌ | 269/312 [27:15<04:22,  6.10s/it]
Training Epoch: 18/100, batch 538/625 completed (loss: 0.0021:  86%|████████▌ | 269/312 [27:15<04:22,  6.11s/it]
Training Epoch: 18/100, batch 538/625 completed (loss: 0.0017:  87%|████████▋ | 270/312 [27:18<04:16,  6.12s/it]
Training Epoch: 18/100, batch 538/625 completed (loss: 0.0013:  87%|████████▋ | 270/312 [27:18<04:16,  6.12s/it]
Training Epoch: 18/100, batch 538/625 completed (loss: 0.0021:  87%|████████▋ | 270/312 [27:18<04:16,  6.12s/it]
Training Epoch: 18/100, batch 538/625 completed (loss: 0.0207:  87%|████████▋ | 270/312 [27:18<04:17,  6.12s/it]
Training Epoch: 18/100, batch 539/625 completed (loss: 0.0129:  87%|████████▋ | 270/312 [27:18<04:16,  6.12s/it]
Training Epoch: 18/100, batch 539/625 completed (loss: 0.0103:  87%|████████▋ | 270/312 [27:18<04:16,  6.12s/it]
Training Epoch: 18/100, batch 539/625 completed (loss: 0.0275:  87%|████████▋ | 270/312 [27:18<04:17,  6.12s/it]
Training Epoch: 18/100, batch 539/625 completed (loss: 0.001:  87%|████████▋ | 270/312 [27:18<04:16,  6.12s/it] 
Training Epoch: 18/100, batch 540/625 completed (loss: 0.0125:  87%|████████▋ | 270/312 [27:21<04:17,  6.12s/it]
Training Epoch: 18/100, batch 540/625 completed (loss: 0.0843:  87%|████████▋ | 270/312 [27:21<04:16,  6.12s/it]
Training Epoch: 18/100, batch 540/625 completed (loss: 0.0008:  87%|████████▋ | 270/312 [27:21<04:16,  6.12s/it]
Training Epoch: 18/100, batch 540/625 completed (loss: 0.0056:  87%|████████▋ | 270/312 [27:21<04:16,  6.12s/it]
Training Epoch: 18/100, batch 540/625 completed (loss: 0.0008:  87%|████████▋ | 271/312 [27:24<04:10,  6.10s/it]
Training Epoch: 18/100, batch 540/625 completed (loss: 0.0125:  87%|████████▋ | 271/312 [27:24<04:10,  6.10s/it]
Training Epoch: 18/100, batch 540/625 completed (loss: 0.0843:  87%|████████▋ | 271/312 [27:24<04:10,  6.10s/it]
Training Epoch: 18/100, batch 540/625 completed (loss: 0.0056:  87%|████████▋ | 271/312 [27:24<04:10,  6.10s/it]
Training Epoch: 18/100, batch 541/625 completed (loss: 0.0002:  87%|████████▋ | 271/312 [27:24<04:10,  6.10s/it]
Training Epoch: 18/100, batch 541/625 completed (loss: 0.0033:  87%|████████▋ | 271/312 [27:24<04:10,  6.10s/it]
Training Epoch: 18/100, batch 541/625 completed (loss: 0.0002:  87%|████████▋ | 271/312 [27:24<04:10,  6.10s/it]
Training Epoch: 18/100, batch 541/625 completed (loss: 0.0013:  87%|████████▋ | 271/312 [27:24<04:10,  6.10s/it]
Training Epoch: 18/100, batch 542/625 completed (loss: 0.0115:  87%|████████▋ | 271/312 [27:27<04:10,  6.10s/it]
Training Epoch: 18/100, batch 542/625 completed (loss: 0.0016:  87%|████████▋ | 271/312 [27:27<04:10,  6.10s/it]
Training Epoch: 18/100, batch 542/625 completed (loss: 0.0013:  87%|████████▋ | 271/312 [27:27<04:10,  6.10s/it]
Training Epoch: 18/100, batch 542/625 completed (loss: 0.0004:  87%|████████▋ | 271/312 [27:27<04:10,  6.10s/it]
Training Epoch: 18/100, batch 542/625 completed (loss: 0.0013:  87%|████████▋ | 272/312 [27:30<04:00,  6.02s/it]
Training Epoch: 18/100, batch 542/625 completed (loss: 0.0016:  87%|████████▋ | 272/312 [27:30<04:00,  6.02s/it]
Training Epoch: 18/100, batch 542/625 completed (loss: 0.0115:  87%|████████▋ | 272/312 [27:30<04:00,  6.02s/it]
Training Epoch: 18/100, batch 542/625 completed (loss: 0.0004:  87%|████████▋ | 272/312 [27:30<04:01,  6.03s/it]
Training Epoch: 18/100, batch 543/625 completed (loss: 0.0042:  87%|████████▋ | 272/312 [27:30<04:00,  6.02s/it]
Training Epoch: 18/100, batch 543/625 completed (loss: 0.0101:  87%|████████▋ | 272/312 [27:30<04:00,  6.02s/it]
Training Epoch: 18/100, batch 543/625 completed (loss: 0.0002:  87%|████████▋ | 272/312 [27:30<04:00,  6.02s/it]
Training Epoch: 18/100, batch 543/625 completed (loss: 0.0006:  87%|████████▋ | 272/312 [27:30<04:01,  6.03s/it]
Training Epoch: 18/100, batch 544/625 completed (loss: 0.0011:  87%|████████▋ | 272/312 [27:33<04:00,  6.02s/it]
Training Epoch: 18/100, batch 544/625 completed (loss: 0.0028:  87%|████████▋ | 272/312 [27:33<04:00,  6.02s/it]
Training Epoch: 18/100, batch 544/625 completed (loss: 0.0004:  87%|████████▋ | 272/312 [27:33<04:00,  6.02s/it]
Training Epoch: 18/100, batch 544/625 completed (loss: 0.0005:  87%|████████▋ | 272/312 [27:33<04:01,  6.03s/it]
Training Epoch: 18/100, batch 544/625 completed (loss: 0.0005:  88%|████████▊ | 273/312 [27:36<03:56,  6.07s/it]
Training Epoch: 18/100, batch 544/625 completed (loss: 0.0011:  88%|████████▊ | 273/312 [27:36<03:56,  6.07s/it]
Training Epoch: 18/100, batch 544/625 completed (loss: 0.0004:  88%|████████▊ | 273/312 [27:36<03:56,  6.07s/it]
Training Epoch: 18/100, batch 544/625 completed (loss: 0.0028:  88%|████████▊ | 273/312 [27:36<03:56,  6.07s/it]
Training Epoch: 18/100, batch 545/625 completed (loss: 0.0002:  88%|████████▊ | 273/312 [27:36<03:56,  6.07s/it]
Training Epoch: 18/100, batch 545/625 completed (loss: 0.0014:  88%|████████▊ | 273/312 [27:36<03:56,  6.07s/it]
Training Epoch: 18/100, batch 545/625 completed (loss: 0.0005:  88%|████████▊ | 273/312 [27:36<03:56,  6.07s/it]
Training Epoch: 18/100, batch 545/625 completed (loss: 0.0062:  88%|████████▊ | 273/312 [27:36<03:56,  6.07s/it]
Training Epoch: 18/100, batch 546/625 completed (loss: 0.0018:  88%|████████▊ | 273/312 [27:39<03:56,  6.07s/it]
Training Epoch: 18/100, batch 546/625 completed (loss: 0.031:  88%|████████▊ | 273/312 [27:39<03:56,  6.07s/it] 
Training Epoch: 18/100, batch 546/625 completed (loss: 0.0025:  88%|████████▊ | 273/312 [27:39<03:56,  6.07s/it]
Training Epoch: 18/100, batch 546/625 completed (loss: 0.0357:  88%|████████▊ | 273/312 [27:39<03:56,  6.07s/it]
Training Epoch: 18/100, batch 546/625 completed (loss: 0.0357:  88%|████████▊ | 274/312 [27:42<03:51,  6.10s/it]
Training Epoch: 18/100, batch 546/625 completed (loss: 0.0025:  88%|████████▊ | 274/312 [27:42<03:51,  6.10s/it]
Training Epoch: 18/100, batch 546/625 completed (loss: 0.0018:  88%|████████▊ | 274/312 [27:42<03:51,  6.10s/it]
Training Epoch: 18/100, batch 546/625 completed (loss: 0.031:  88%|████████▊ | 274/312 [27:42<03:52,  6.11s/it]
Training Epoch: 18/100, batch 547/625 completed (loss: 0.0069:  88%|████████▊ | 274/312 [27:42<03:52,  6.11s/it]
Training Epoch: 18/100, batch 547/625 completed (loss: 0.0121:  88%|████████▊ | 274/312 [27:42<03:51,  6.10s/it]
Training Epoch: 18/100, batch 547/625 completed (loss: 0.0037:  88%|████████▊ | 274/312 [27:42<03:51,  6.10s/it]
Training Epoch: 18/100, batch 547/625 completed (loss: 0.0028:  88%|████████▊ | 274/312 [27:42<03:51,  6.10s/it]
Training Epoch: 18/100, batch 548/625 completed (loss: 0.0014:  88%|████████▊ | 274/312 [27:45<03:51,  6.10s/it]
Training Epoch: 18/100, batch 548/625 completed (loss: 0.0:  88%|████████▊ | 274/312 [27:45<03:52,  6.11s/it]   
Training Epoch: 18/100, batch 548/625 completed (loss: 0.0009:  88%|████████▊ | 274/312 [27:45<03:51,  6.10s/it]
Training Epoch: 18/100, batch 548/625 completed (loss: 0.0166:  88%|████████▊ | 274/312 [27:45<03:51,  6.10s/it]
Training Epoch: 18/100, batch 548/625 completed (loss: 0.0009:  88%|████████▊ | 275/312 [27:48<03:44,  6.07s/it]
Training Epoch: 18/100, batch 548/625 completed (loss: 0.0166:  88%|████████▊ | 275/312 [27:48<03:44,  6.07s/it]
Training Epoch: 18/100, batch 548/625 completed (loss: 0.0:  88%|████████▊ | 275/312 [27:48<03:44,  6.07s/it]
Training Epoch: 18/100, batch 548/625 completed (loss: 0.0014:  88%|████████▊ | 275/312 [27:48<03:44,  6.07s/it]
Training Epoch: 18/100, batch 549/625 completed (loss: 0.0004:  88%|████████▊ | 275/312 [27:48<03:44,  6.07s/it]
Training Epoch: 18/100, batch 549/625 completed (loss: 0.0006:  88%|████████▊ | 275/312 [27:48<03:44,  6.07s/it]
Training Epoch: 18/100, batch 549/625 completed (loss: 0.0003:  88%|████████▊ | 275/312 [27:48<03:44,  6.07s/it]
Training Epoch: 18/100, batch 549/625 completed (loss: 0.0021:  88%|████████▊ | 275/312 [27:48<03:44,  6.07s/it]
Training Epoch: 18/100, batch 550/625 completed (loss: 0.0003:  88%|████████▊ | 275/312 [27:51<03:44,  6.07s/it]
Training Epoch: 18/100, batch 550/625 completed (loss: 0.0026:  88%|████████▊ | 275/312 [27:51<03:44,  6.07s/it]
Training Epoch: 18/100, batch 550/625 completed (loss: 0.0025:  88%|████████▊ | 275/312 [27:51<03:44,  6.07s/it]
Training Epoch: 18/100, batch 550/625 completed (loss: 0.0118:  88%|████████▊ | 275/312 [27:51<03:44,  6.07s/it]
Training Epoch: 18/100, batch 550/625 completed (loss: 0.0025:  88%|████████▊ | 276/312 [27:54<03:37,  6.05s/it]
Training Epoch: 18/100, batch 550/625 completed (loss: 0.0118:  88%|████████▊ | 276/312 [27:54<03:37,  6.05s/it]
Training Epoch: 18/100, batch 550/625 completed (loss: 0.0026:  88%|████████▊ | 276/312 [27:54<03:37,  6.05s/it]
Training Epoch: 18/100, batch 550/625 completed (loss: 0.0003:  88%|████████▊ | 276/312 [27:54<03:37,  6.05s/it]
Training Epoch: 18/100, batch 551/625 completed (loss: 0.0:  88%|████████▊ | 276/312 [27:54<03:37,  6.05s/it]   
Training Epoch: 18/100, batch 551/625 completed (loss: 0.0005:  88%|████████▊ | 276/312 [27:54<03:37,  6.05s/it]
Training Epoch: 18/100, batch 551/625 completed (loss: 0.0005:  88%|████████▊ | 276/312 [27:54<03:37,  6.05s/it]
Training Epoch: 18/100, batch 551/625 completed (loss: 0.0039:  88%|████████▊ | 276/312 [27:54<03:37,  6.05s/it]
Training Epoch: 18/100, batch 552/625 completed (loss: 0.0018:  88%|████████▊ | 276/312 [27:57<03:37,  6.05s/it]
Training Epoch: 18/100, batch 552/625 completed (loss: 0.0008:  88%|████████▊ | 276/312 [27:57<03:37,  6.05s/it]
Training Epoch: 18/100, batch 552/625 completed (loss: 0.084:  88%|████████▊ | 276/312 [27:57<03:37,  6.05s/it]
Training Epoch: 18/100, batch 552/625 completed (loss: 0.0008:  88%|████████▊ | 276/312 [27:57<03:37,  6.05s/it]
Training Epoch: 18/100, batch 552/625 completed (loss: 0.0018:  89%|████████▉ | 277/312 [28:00<03:32,  6.07s/it]
Training Epoch: 18/100, batch 552/625 completed (loss: 0.084:  89%|████████▉ | 277/312 [28:00<03:32,  6.07s/it]
Training Epoch: 18/100, batch 552/625 completed (loss: 0.0008:  89%|████████▉ | 277/312 [28:00<03:32,  6.07s/it]
Training Epoch: 18/100, batch 552/625 completed (loss: 0.0008:  89%|████████▉ | 277/312 [28:00<03:32,  6.07s/it]
Training Epoch: 18/100, batch 553/625 completed (loss: 0.0177:  89%|████████▉ | 277/312 [28:00<03:32,  6.07s/it]
Training Epoch: 18/100, batch 553/625 completed (loss: 0.0042:  89%|████████▉ | 277/312 [28:00<03:32,  6.07s/it]
Training Epoch: 18/100, batch 553/625 completed (loss: 0.0223:  89%|████████▉ | 277/312 [28:00<03:32,  6.07s/it]
Training Epoch: 18/100, batch 553/625 completed (loss: 0.0024:  89%|████████▉ | 277/312 [28:00<03:32,  6.07s/it]
Training Epoch: 18/100, batch 554/625 completed (loss: 0.0643:  89%|████████▉ | 277/312 [28:03<03:32,  6.07s/it]
Training Epoch: 18/100, batch 554/625 completed (loss: 0.0228:  89%|████████▉ | 277/312 [28:03<03:32,  6.07s/it]
Training Epoch: 18/100, batch 554/625 completed (loss: 0.0099:  89%|████████▉ | 277/312 [28:03<03:32,  6.07s/it]
Training Epoch: 18/100, batch 554/625 completed (loss: 0.0022:  89%|████████▉ | 277/312 [28:03<03:32,  6.07s/it]
Training Epoch: 18/100, batch 554/625 completed (loss: 0.0022:  89%|████████▉ | 278/312 [28:06<03:27,  6.09s/it]
Training Epoch: 18/100, batch 554/625 completed (loss: 0.0643:  89%|████████▉ | 278/312 [28:06<03:27,  6.09s/it]
Training Epoch: 18/100, batch 554/625 completed (loss: 0.0228:  89%|████████▉ | 278/312 [28:06<03:27,  6.09s/it]
Training Epoch: 18/100, batch 554/625 completed (loss: 0.0099:  89%|████████▉ | 278/312 [28:06<03:27,  6.10s/it]
Training Epoch: 18/100, batch 555/625 completed (loss: 0.0048:  89%|████████▉ | 278/312 [28:06<03:27,  6.09s/it]
Training Epoch: 18/100, batch 555/625 completed (loss: 0.0001:  89%|████████▉ | 278/312 [28:06<03:27,  6.10s/it]
Training Epoch: 18/100, batch 555/625 completed (loss: 0.0002:  89%|████████▉ | 278/312 [28:06<03:27,  6.09s/it]
Training Epoch: 18/100, batch 555/625 completed (loss: 0.0014:  89%|████████▉ | 278/312 [28:07<03:27,  6.09s/it]
Training Epoch: 18/100, batch 556/625 completed (loss: 0.0014:  89%|████████▉ | 278/312 [28:09<03:27,  6.10s/it]
Training Epoch: 18/100, batch 556/625 completed (loss: 0.0034:  89%|████████▉ | 278/312 [28:09<03:27,  6.09s/it]
Training Epoch: 18/100, batch 556/625 completed (loss: 0.012:  89%|████████▉ | 278/312 [28:09<03:27,  6.09s/it] 
Training Epoch: 18/100, batch 556/625 completed (loss: 0.0002:  89%|████████▉ | 278/312 [28:09<03:27,  6.09s/it]
Training Epoch: 18/100, batch 556/625 completed (loss: 0.0002:  89%|████████▉ | 279/312 [28:13<03:21,  6.11s/it]
Training Epoch: 18/100, batch 556/625 completed (loss: 0.0034:  89%|████████▉ | 279/312 [28:13<03:21,  6.11s/it]
Training Epoch: 18/100, batch 556/625 completed (loss: 0.0014:  89%|████████▉ | 279/312 [28:13<03:21,  6.11s/it]
Training Epoch: 18/100, batch 556/625 completed (loss: 0.012:  89%|████████▉ | 279/312 [28:13<03:21,  6.11s/it]
Training Epoch: 18/100, batch 557/625 completed (loss: 0.0002:  89%|████████▉ | 279/312 [28:13<03:21,  6.11s/it]
Training Epoch: 18/100, batch 557/625 completed (loss: 0.0003:  89%|████████▉ | 279/312 [28:13<03:21,  6.11s/it]
Training Epoch: 18/100, batch 557/625 completed (loss: 0.0035:  89%|████████▉ | 279/312 [28:13<03:21,  6.11s/it]
Training Epoch: 18/100, batch 557/625 completed (loss: 0.003:  89%|████████▉ | 279/312 [28:13<03:21,  6.11s/it] 
Training Epoch: 18/100, batch 558/625 completed (loss: 0.0016:  89%|████████▉ | 279/312 [28:16<03:21,  6.11s/it]
Training Epoch: 18/100, batch 558/625 completed (loss: 0.0021:  89%|████████▉ | 279/312 [28:16<03:21,  6.11s/it]
Training Epoch: 18/100, batch 558/625 completed (loss: 0.0058:  89%|████████▉ | 279/312 [28:16<03:21,  6.11s/it]
Training Epoch: 18/100, batch 558/625 completed (loss: 0.0057:  89%|████████▉ | 279/312 [28:16<03:21,  6.11s/it]
Training Epoch: 18/100, batch 558/625 completed (loss: 0.0057:  90%|████████▉ | 280/312 [28:19<03:14,  6.08s/it]
Training Epoch: 18/100, batch 558/625 completed (loss: 0.0021:  90%|████████▉ | 280/312 [28:19<03:14,  6.08s/it]
Training Epoch: 18/100, batch 558/625 completed (loss: 0.0058:  90%|████████▉ | 280/312 [28:19<03:14,  6.08s/it]
Training Epoch: 18/100, batch 558/625 completed (loss: 0.0016:  90%|████████▉ | 280/312 [28:19<03:14,  6.08s/it]
Training Epoch: 18/100, batch 559/625 completed (loss: 0.0027:  90%|████████▉ | 280/312 [28:19<03:14,  6.08s/it]
Training Epoch: 18/100, batch 559/625 completed (loss: 0.0065:  90%|████████▉ | 280/312 [28:19<03:14,  6.08s/it]
Training Epoch: 18/100, batch 559/625 completed (loss: 0.0153:  90%|████████▉ | 280/312 [28:19<03:14,  6.08s/it]
Training Epoch: 18/100, batch 559/625 completed (loss: 0.0029:  90%|████████▉ | 280/312 [28:19<03:14,  6.08s/it]
Training Epoch: 18/100, batch 560/625 completed (loss: 0.0005:  90%|████████▉ | 280/312 [28:22<03:14,  6.08s/it]
Training Epoch: 18/100, batch 560/625 completed (loss: 0.0003:  90%|████████▉ | 280/312 [28:22<03:14,  6.08s/it]
Training Epoch: 18/100, batch 560/625 completed (loss: 0.004:  90%|████████▉ | 280/312 [28:22<03:14,  6.08s/it] 
Training Epoch: 18/100, batch 560/625 completed (loss: 0.0009:  90%|████████▉ | 280/312 [28:22<03:14,  6.08s/it]
Training Epoch: 18/100, batch 560/625 completed (loss: 0.0009:  90%|█████████ | 281/312 [28:25<03:07,  6.06s/it]
Training Epoch: 18/100, batch 560/625 completed (loss: 0.0003:  90%|█████████ | 281/312 [28:25<03:07,  6.06s/it]
Training Epoch: 18/100, batch 560/625 completed (loss: 0.0005:  90%|█████████ | 281/312 [28:25<03:07,  6.06s/it]
Training Epoch: 18/100, batch 560/625 completed (loss: 0.004:  90%|█████████ | 281/312 [28:25<03:07,  6.06s/it]
Training Epoch: 18/100, batch 561/625 completed (loss: 0.0176:  90%|█████████ | 281/312 [28:25<03:07,  6.06s/it]
Training Epoch: 18/100, batch 561/625 completed (loss: 0.0002:  90%|█████████ | 281/312 [28:25<03:07,  6.06s/it]
Training Epoch: 18/100, batch 561/625 completed (loss: 0.0001:  90%|█████████ | 281/312 [28:25<03:07,  6.06s/it]
Training Epoch: 18/100, batch 561/625 completed (loss: 0.0002:  90%|█████████ | 281/312 [28:25<03:07,  6.06s/it]
Training Epoch: 18/100, batch 562/625 completed (loss: 0.0043:  90%|█████████ | 281/312 [28:28<03:07,  6.06s/it]
Training Epoch: 18/100, batch 562/625 completed (loss: 0.0006:  90%|█████████ | 281/312 [28:28<03:07,  6.06s/it]
Training Epoch: 18/100, batch 562/625 completed (loss: 0.001:  90%|█████████ | 281/312 [28:28<03:07,  6.06s/it] 
Training Epoch: 18/100, batch 562/625 completed (loss: 0.0013:  90%|█████████ | 281/312 [28:28<03:07,  6.06s/it]
Training Epoch: 18/100, batch 562/625 completed (loss: 0.0043:  90%|█████████ | 282/312 [28:31<03:02,  6.10s/it]
Training Epoch: 18/100, batch 562/625 completed (loss: 0.0006:  90%|█████████ | 282/312 [28:31<03:02,  6.10s/it]
Training Epoch: 18/100, batch 562/625 completed (loss: 0.0013:  90%|█████████ | 282/312 [28:31<03:03,  6.10s/it]
Training Epoch: 18/100, batch 562/625 completed (loss: 0.001:  90%|█████████ | 282/312 [28:31<03:03,  6.10s/it]
Training Epoch: 18/100, batch 563/625 completed (loss: 0.0092:  90%|█████████ | 282/312 [28:31<03:02,  6.10s/it]
Training Epoch: 18/100, batch 563/625 completed (loss: 0.0:  90%|█████████ | 282/312 [28:31<03:03,  6.10s/it]  
Training Epoch: 18/100, batch 563/625 completed (loss: 0.0354:  90%|█████████ | 282/312 [28:31<03:02,  6.10s/it]
Training Epoch: 18/100, batch 563/625 completed (loss: 0.0004:  90%|█████████ | 282/312 [28:31<03:03,  6.10s/it]
Training Epoch: 18/100, batch 564/625 completed (loss: 0.0006:  90%|█████████ | 282/312 [28:34<03:03,  6.10s/it]
Training Epoch: 18/100, batch 564/625 completed (loss: 0.0009:  90%|█████████ | 282/312 [28:34<03:02,  6.10s/it]
Training Epoch: 18/100, batch 564/625 completed (loss: 0.0239:  90%|█████████ | 282/312 [28:34<03:02,  6.10s/it]
Training Epoch: 18/100, batch 564/625 completed (loss: 0.0004:  90%|█████████ | 282/312 [28:34<03:03,  6.10s/it]
Training Epoch: 18/100, batch 564/625 completed (loss: 0.0004:  91%|█████████ | 283/312 [28:37<02:55,  6.07s/it]
Training Epoch: 18/100, batch 564/625 completed (loss: 0.0239:  91%|█████████ | 283/312 [28:37<02:56,  6.07s/it]
Training Epoch: 18/100, batch 564/625 completed (loss: 0.0009:  91%|█████████ | 283/312 [28:37<02:56,  6.07s/it]
Training Epoch: 18/100, batch 564/625 completed (loss: 0.0006:  91%|█████████ | 283/312 [28:37<02:55,  6.07s/it]
Training Epoch: 18/100, batch 565/625 completed (loss: 0.0228:  91%|█████████ | 283/312 [28:37<02:56,  6.07s/it]
Training Epoch: 18/100, batch 565/625 completed (loss: 0.0006:  91%|█████████ | 283/312 [28:37<02:55,  6.07s/it]
Training Epoch: 18/100, batch 565/625 completed (loss: 0.0106:  91%|█████████ | 283/312 [28:37<02:56,  6.07s/it]
Training Epoch: 18/100, batch 565/625 completed (loss: 0.025:  91%|█████████ | 283/312 [28:37<02:55,  6.07s/it] 
Training Epoch: 18/100, batch 566/625 completed (loss: 0.0039:  91%|█████████ | 283/312 [28:40<02:56,  6.07s/it]
Training Epoch: 18/100, batch 566/625 completed (loss: 0.0008:  91%|█████████ | 283/312 [28:40<02:55,  6.07s/it]
Training Epoch: 18/100, batch 566/625 completed (loss: 0.0109:  91%|█████████ | 283/312 [28:40<02:56,  6.07s/it]
Training Epoch: 18/100, batch 566/625 completed (loss: 0.0001:  91%|█████████ | 283/312 [28:40<02:55,  6.07s/it]
Training Epoch: 18/100, batch 566/625 completed (loss: 0.0039:  91%|█████████ | 284/312 [28:43<02:50,  6.09s/it]
Training Epoch: 18/100, batch 566/625 completed (loss: 0.0001:  91%|█████████ | 284/312 [28:43<02:50,  6.10s/it]
Training Epoch: 18/100, batch 566/625 completed (loss: 0.0008:  91%|█████████ | 284/312 [28:43<02:50,  6.10s/it]
Training Epoch: 18/100, batch 566/625 completed (loss: 0.0109:  91%|█████████ | 284/312 [28:43<02:50,  6.11s/it]
Training Epoch: 18/100, batch 567/625 completed (loss: 0.0016:  91%|█████████ | 284/312 [28:43<02:50,  6.09s/it]
Training Epoch: 18/100, batch 567/625 completed (loss: 0.0074:  91%|█████████ | 284/312 [28:43<02:50,  6.11s/it]
Training Epoch: 18/100, batch 567/625 completed (loss: 0.0038:  91%|█████████ | 284/312 [28:43<02:50,  6.10s/it]
Training Epoch: 18/100, batch 567/625 completed (loss: 0.0177:  91%|█████████ | 284/312 [28:43<02:50,  6.10s/it]
Training Epoch: 18/100, batch 568/625 completed (loss: 0.0013:  91%|█████████ | 284/312 [28:46<02:50,  6.11s/it]
Training Epoch: 18/100, batch 568/625 completed (loss: 0.0009:  91%|█████████ | 284/312 [28:46<02:50,  6.09s/it]
Training Epoch: 18/100, batch 568/625 completed (loss: 0.0009:  91%|█████████ | 284/312 [28:46<02:50,  6.10s/it]
Training Epoch: 18/100, batch 568/625 completed (loss: 0.0035:  91%|█████████ | 284/312 [28:46<02:50,  6.10s/it]
Training Epoch: 18/100, batch 568/625 completed (loss: 0.0009:  91%|█████████▏| 285/312 [28:49<02:44,  6.08s/it]
Training Epoch: 18/100, batch 568/625 completed (loss: 0.0035:  91%|█████████▏| 285/312 [28:49<02:44,  6.09s/it]
Training Epoch: 18/100, batch 568/625 completed (loss: 0.0009:  91%|█████████▏| 285/312 [28:49<02:44,  6.09s/it]
Training Epoch: 18/100, batch 568/625 completed (loss: 0.0013:  91%|█████████▏| 285/312 [28:49<02:44,  6.09s/it]
Training Epoch: 18/100, batch 569/625 completed (loss: 0.0008:  91%|█████████▏| 285/312 [28:49<02:44,  6.09s/it]
Training Epoch: 18/100, batch 569/625 completed (loss: 0.0003:  91%|█████████▏| 285/312 [28:49<02:44,  6.08s/it]
Training Epoch: 18/100, batch 569/625 completed (loss: 0.0035:  91%|█████████▏| 285/312 [28:49<02:44,  6.09s/it]
Training Epoch: 18/100, batch 569/625 completed (loss: 0.0006:  91%|█████████▏| 285/312 [28:49<02:44,  6.09s/it]
Training Epoch: 18/100, batch 570/625 completed (loss: 0.0:  91%|█████████▏| 285/312 [28:52<02:44,  6.09s/it]   
Training Epoch: 18/100, batch 570/625 completed (loss: 0.0008:  91%|█████████▏| 285/312 [28:52<02:44,  6.09s/it]
Training Epoch: 18/100, batch 570/625 completed (loss: 0.0023:  91%|█████████▏| 285/312 [28:52<02:44,  6.08s/it]
Training Epoch: 18/100, batch 570/625 completed (loss: 0.0074:  91%|█████████▏| 285/312 [28:52<02:44,  6.09s/it]
Training Epoch: 18/100, batch 570/625 completed (loss: 0.0074:  92%|█████████▏| 286/312 [28:55<02:38,  6.11s/it]
Training Epoch: 18/100, batch 570/625 completed (loss: 0.0:  92%|█████████▏| 286/312 [28:55<02:38,  6.11s/it]
Training Epoch: 18/100, batch 570/625 completed (loss: 0.0008:  92%|█████████▏| 286/312 [28:55<02:38,  6.11s/it]
Training Epoch: 18/100, batch 570/625 completed (loss: 0.0023:  92%|█████████▏| 286/312 [28:55<02:38,  6.11s/it]
Training Epoch: 18/100, batch 571/625 completed (loss: 0.0055:  92%|█████████▏| 286/312 [28:55<02:38,  6.11s/it]
Training Epoch: 18/100, batch 571/625 completed (loss: 0.0381:  92%|█████████▏| 286/312 [28:55<02:38,  6.11s/it]
Training Epoch: 18/100, batch 571/625 completed (loss: 0.1824:  92%|█████████▏| 286/312 [28:55<02:38,  6.11s/it]
Training Epoch: 18/100, batch 571/625 completed (loss: 0.0144:  92%|█████████▏| 286/312 [28:55<02:38,  6.11s/it]
Training Epoch: 18/100, batch 572/625 completed (loss: 0.0273:  92%|█████████▏| 286/312 [28:58<02:38,  6.11s/it]
Training Epoch: 18/100, batch 572/625 completed (loss: 0.0023:  92%|█████████▏| 286/312 [28:58<02:38,  6.11s/it]
Training Epoch: 18/100, batch 572/625 completed (loss: 0.0118:  92%|█████████▏| 286/312 [28:58<02:38,  6.11s/it]
Training Epoch: 18/100, batch 572/625 completed (loss: 0.0016:  92%|█████████▏| 286/312 [28:58<02:38,  6.11s/it]
Training Epoch: 18/100, batch 572/625 completed (loss: 0.0118:  92%|█████████▏| 287/312 [29:01<02:33,  6.13s/it]
Training Epoch: 18/100, batch 572/625 completed (loss: 0.0016:  92%|█████████▏| 287/312 [29:01<02:33,  6.14s/it]
Training Epoch: 18/100, batch 572/625 completed (loss: 0.0273:  92%|█████████▏| 287/312 [29:01<02:33,  6.13s/it]
Training Epoch: 18/100, batch 572/625 completed (loss: 0.0023:  92%|█████████▏| 287/312 [29:01<02:33,  6.14s/it]
Training Epoch: 18/100, batch 573/625 completed (loss: 0.0005:  92%|█████████▏| 287/312 [29:01<02:33,  6.14s/it]
Training Epoch: 18/100, batch 573/625 completed (loss: 0.0017:  92%|█████████▏| 287/312 [29:01<02:33,  6.13s/it]
Training Epoch: 18/100, batch 573/625 completed (loss: 0.037:  92%|█████████▏| 287/312 [29:01<02:33,  6.13s/it] 
Training Epoch: 18/100, batch 573/625 completed (loss: 0.0001:  92%|█████████▏| 287/312 [29:01<02:33,  6.14s/it]
Training Epoch: 18/100, batch 574/625 completed (loss: 0.0006:  92%|█████████▏| 287/312 [29:04<02:33,  6.14s/it]
Training Epoch: 18/100, batch 574/625 completed (loss: 0.0362:  92%|█████████▏| 287/312 [29:04<02:33,  6.13s/it]
Training Epoch: 18/100, batch 574/625 completed (loss: 0.0007:  92%|█████████▏| 287/312 [29:04<02:33,  6.13s/it]
Training Epoch: 18/100, batch 574/625 completed (loss: 0.0006:  92%|█████████▏| 287/312 [29:04<02:33,  6.14s/it]
Training Epoch: 18/100, batch 574/625 completed (loss: 0.0006:  92%|█████████▏| 288/312 [29:07<02:26,  6.10s/it]
Training Epoch: 18/100, batch 574/625 completed (loss: 0.0362:  92%|█████████▏| 288/312 [29:07<02:26,  6.10s/it]
Training Epoch: 18/100, batch 574/625 completed (loss: 0.0007:  92%|█████████▏| 288/312 [29:07<02:26,  6.10s/it]
Training Epoch: 18/100, batch 574/625 completed (loss: 0.0006:  92%|█████████▏| 288/312 [29:07<02:26,  6.10s/it]
Training Epoch: 18/100, batch 575/625 completed (loss: 0.0003:  92%|█████████▏| 288/312 [29:07<02:26,  6.10s/it]
Training Epoch: 18/100, batch 575/625 completed (loss: 0.0014:  92%|█████████▏| 288/312 [29:07<02:26,  6.10s/it]
Training Epoch: 18/100, batch 575/625 completed (loss: 0.0001:  92%|█████████▏| 288/312 [29:07<02:26,  6.10s/it]
Training Epoch: 18/100, batch 575/625 completed (loss: 0.0004:  92%|█████████▏| 288/312 [29:07<02:26,  6.10s/it]
Training Epoch: 18/100, batch 576/625 completed (loss: 0.0002:  92%|█████████▏| 288/312 [29:10<02:26,  6.10s/it]
Training Epoch: 18/100, batch 576/625 completed (loss: 0.1361:  92%|█████████▏| 288/312 [29:11<02:26,  6.10s/it]
Training Epoch: 18/100, batch 576/625 completed (loss: 0.0001:  92%|█████████▏| 288/312 [29:11<02:26,  6.10s/it]
Training Epoch: 18/100, batch 576/625 completed (loss: 0.0211:  92%|█████████▏| 288/312 [29:11<02:26,  6.10s/it]
Training Epoch: 18/100, batch 576/625 completed (loss: 0.0211:  93%|█████████▎| 289/312 [29:13<02:19,  6.07s/it]
Training Epoch: 18/100, batch 576/625 completed (loss: 0.0001:  93%|█████████▎| 289/312 [29:13<02:19,  6.08s/it]
Training Epoch: 18/100, batch 576/625 completed (loss: 0.1361:  93%|█████████▎| 289/312 [29:13<02:19,  6.08s/it]
Training Epoch: 18/100, batch 576/625 completed (loss: 0.0002:  93%|█████████▎| 289/312 [29:13<02:19,  6.08s/it]
Training Epoch: 18/100, batch 577/625 completed (loss: 0.0398:  93%|█████████▎| 289/312 [29:13<02:19,  6.08s/it]
Training Epoch: 18/100, batch 577/625 completed (loss: 0.0014:  93%|█████████▎| 289/312 [29:13<02:19,  6.08s/it]
Training Epoch: 18/100, batch 577/625 completed (loss: 0.0009:  93%|█████████▎| 289/312 [29:13<02:19,  6.08s/it]
Training Epoch: 18/100, batch 577/625 completed (loss: 0.0075:  93%|█████████▎| 289/312 [29:13<02:19,  6.07s/it]
Training Epoch: 18/100, batch 578/625 completed (loss: 0.0024:  93%|█████████▎| 289/312 [29:17<02:19,  6.08s/it]
Training Epoch: 18/100, batch 578/625 completed (loss: 0.0086:  93%|█████████▎| 289/312 [29:17<02:19,  6.08s/it]
Training Epoch: 18/100, batch 578/625 completed (loss: 0.0174:  93%|█████████▎| 289/312 [29:17<02:19,  6.08s/it]
Training Epoch: 18/100, batch 578/625 completed (loss: 0.0006:  93%|█████████▎| 289/312 [29:17<02:19,  6.07s/it]
Training Epoch: 18/100, batch 578/625 completed (loss: 0.0086:  93%|█████████▎| 290/312 [29:20<02:14,  6.10s/it]
Training Epoch: 18/100, batch 578/625 completed (loss: 0.0006:  93%|█████████▎| 290/312 [29:20<02:14,  6.11s/it]
Training Epoch: 18/100, batch 578/625 completed (loss: 0.0174:  93%|█████████▎| 290/312 [29:20<02:14,  6.10s/it]
Training Epoch: 18/100, batch 578/625 completed (loss: 0.0024:  93%|█████████▎| 290/312 [29:20<02:14,  6.11s/it]
Training Epoch: 18/100, batch 579/625 completed (loss: 0.0004:  93%|█████████▎| 290/312 [29:20<02:14,  6.10s/it]
Training Epoch: 18/100, batch 579/625 completed (loss: 0.085:  93%|█████████▎| 290/312 [29:20<02:14,  6.10s/it] 
Training Epoch: 18/100, batch 579/625 completed (loss: 0.0022:  93%|█████████▎| 290/312 [29:20<02:14,  6.11s/it]
Training Epoch: 18/100, batch 579/625 completed (loss: 0.0039:  93%|█████████▎| 290/312 [29:20<02:14,  6.11s/it]
Training Epoch: 18/100, batch 580/625 completed (loss: 0.0066:  93%|█████████▎| 290/312 [29:23<02:14,  6.11s/it]
Training Epoch: 18/100, batch 580/625 completed (loss: 0.0009:  93%|█████████▎| 290/312 [29:23<02:14,  6.10s/it]
Training Epoch: 18/100, batch 580/625 completed (loss: 0.0069:  93%|█████████▎| 290/312 [29:23<02:14,  6.10s/it]
Training Epoch: 18/100, batch 580/625 completed (loss: 0.0002:  93%|█████████▎| 290/312 [29:23<02:14,  6.11s/it]
Training Epoch: 18/100, batch 580/625 completed (loss: 0.0009:  93%|█████████▎| 291/312 [29:26<02:08,  6.12s/it]
Training Epoch: 18/100, batch 580/625 completed (loss: 0.0069:  93%|█████████▎| 291/312 [29:26<02:08,  6.12s/it]
Training Epoch: 18/100, batch 580/625 completed (loss: 0.0066:  93%|█████████▎| 291/312 [29:26<02:08,  6.12s/it]
Training Epoch: 18/100, batch 580/625 completed (loss: 0.0002:  93%|█████████▎| 291/312 [29:26<02:08,  6.13s/it]
Training Epoch: 18/100, batch 581/625 completed (loss: 0.0062:  93%|█████████▎| 291/312 [29:26<02:08,  6.12s/it]
Training Epoch: 18/100, batch 581/625 completed (loss: 0.0126:  93%|█████████▎| 291/312 [29:26<02:08,  6.12s/it]
Training Epoch: 18/100, batch 581/625 completed (loss: 0.0011:  93%|█████████▎| 291/312 [29:26<02:08,  6.12s/it]
Training Epoch: 18/100, batch 581/625 completed (loss: 0.0007:  93%|█████████▎| 291/312 [29:26<02:08,  6.13s/it]
Training Epoch: 18/100, batch 582/625 completed (loss: 0.0001:  93%|█████████▎| 291/312 [29:29<02:08,  6.12s/it]
Training Epoch: 18/100, batch 582/625 completed (loss: 0.0003:  93%|█████████▎| 291/312 [29:29<02:08,  6.12s/it]
Training Epoch: 18/100, batch 582/625 completed (loss: 0.0089:  93%|█████████▎| 291/312 [29:29<02:08,  6.12s/it]
Training Epoch: 18/100, batch 582/625 completed (loss: 0.001:  93%|█████████▎| 291/312 [29:29<02:08,  6.13s/it] 
Training Epoch: 18/100, batch 582/625 completed (loss: 0.001:  94%|█████████▎| 292/312 [29:32<02:02,  6.10s/it]
Training Epoch: 18/100, batch 582/625 completed (loss: 0.0001:  94%|█████████▎| 292/312 [29:32<02:02,  6.10s/it]
Training Epoch: 18/100, batch 582/625 completed (loss: 0.0003:  94%|█████████▎| 292/312 [29:32<02:02,  6.11s/it]
Training Epoch: 18/100, batch 582/625 completed (loss: 0.0089:  94%|█████████▎| 292/312 [29:32<02:02,  6.11s/it]
Training Epoch: 18/100, batch 583/625 completed (loss: 0.014:  94%|█████████▎| 292/312 [29:32<02:02,  6.11s/it] 
Training Epoch: 18/100, batch 583/625 completed (loss: 0.0021:  94%|█████████▎| 292/312 [29:32<02:02,  6.11s/it]
Training Epoch: 18/100, batch 583/625 completed (loss: 0.0004:  94%|█████████▎| 292/312 [29:32<02:02,  6.10s/it]
Training Epoch: 18/100, batch 583/625 completed (loss: 0.0054:  94%|█████████▎| 292/312 [29:32<02:02,  6.10s/it]
Training Epoch: 18/100, batch 584/625 completed (loss: 0.0003:  94%|█████████▎| 292/312 [29:35<02:02,  6.10s/it]
Training Epoch: 18/100, batch 584/625 completed (loss: 0.0085:  94%|█████████▎| 292/312 [29:35<02:02,  6.11s/it]
Training Epoch: 18/100, batch 584/625 completed (loss: 0.0002:  94%|█████████▎| 292/312 [29:35<02:02,  6.11s/it]
Training Epoch: 18/100, batch 584/625 completed (loss: 0.0183:  94%|█████████▎| 292/312 [29:35<02:02,  6.10s/it]
Training Epoch: 18/100, batch 584/625 completed (loss: 0.0183:  94%|█████████▍| 293/312 [29:38<01:56,  6.14s/it]
Training Epoch: 18/100, batch 584/625 completed (loss: 0.0085:  94%|█████████▍| 293/312 [29:38<01:56,  6.14s/it]
Training Epoch: 18/100, batch 584/625 completed (loss: 0.0003:  94%|█████████▍| 293/312 [29:38<01:56,  6.14s/it]
Training Epoch: 18/100, batch 584/625 completed (loss: 0.0002:  94%|█████████▍| 293/312 [29:38<01:56,  6.14s/it]
Training Epoch: 18/100, batch 585/625 completed (loss: 0.1332:  94%|█████████▍| 293/312 [29:38<01:56,  6.14s/it]
Training Epoch: 18/100, batch 585/625 completed (loss: 0.0004:  94%|█████████▍| 293/312 [29:38<01:56,  6.14s/it]
Training Epoch: 18/100, batch 585/625 completed (loss: 0.0025:  94%|█████████▍| 293/312 [29:38<01:56,  6.14s/it]
Training Epoch: 18/100, batch 585/625 completed (loss: 0.0257:  94%|█████████▍| 293/312 [29:38<01:56,  6.14s/it]
Training Epoch: 18/100, batch 586/625 completed (loss: 0.0002:  94%|█████████▍| 293/312 [29:41<01:56,  6.14s/it]
Training Epoch: 18/100, batch 586/625 completed (loss: 0.0007:  94%|█████████▍| 293/312 [29:41<01:56,  6.14s/it]
Training Epoch: 18/100, batch 586/625 completed (loss: 0.0:  94%|█████████▍| 293/312 [29:41<01:56,  6.14s/it]   
Training Epoch: 18/100, batch 586/625 completed (loss: 0.0009:  94%|█████████▍| 293/312 [29:41<01:56,  6.14s/it]
Training Epoch: 18/100, batch 586/625 completed (loss: 0.0002:  94%|█████████▍| 294/312 [29:44<01:49,  6.10s/it]
Training Epoch: 18/100, batch 586/625 completed (loss: 0.0009:  94%|█████████▍| 294/312 [29:44<01:49,  6.11s/it]
Training Epoch: 18/100, batch 586/625 completed (loss: 0.0:  94%|█████████▍| 294/312 [29:44<01:49,  6.10s/it]
Training Epoch: 18/100, batch 586/625 completed (loss: 0.0007:  94%|█████████▍| 294/312 [29:44<01:49,  6.11s/it]
Training Epoch: 18/100, batch 587/625 completed (loss: 0.0002:  94%|█████████▍| 294/312 [29:44<01:49,  6.10s/it]
Training Epoch: 18/100, batch 587/625 completed (loss: 0.0003:  94%|█████████▍| 294/312 [29:44<01:49,  6.11s/it]
Training Epoch: 18/100, batch 587/625 completed (loss: 0.0005:  94%|█████████▍| 294/312 [29:44<01:49,  6.10s/it]
Training Epoch: 18/100, batch 587/625 completed (loss: 0.0539:  94%|█████████▍| 294/312 [29:44<01:49,  6.11s/it]
Training Epoch: 18/100, batch 588/625 completed (loss: 0.0024:  94%|█████████▍| 294/312 [29:47<01:49,  6.10s/it]
Training Epoch: 18/100, batch 588/625 completed (loss: 0.0833:  94%|█████████▍| 294/312 [29:47<01:49,  6.11s/it]
Training Epoch: 18/100, batch 588/625 completed (loss: 0.0059:  94%|█████████▍| 294/312 [29:47<01:49,  6.10s/it]
Training Epoch: 18/100, batch 588/625 completed (loss: 0.0013:  94%|█████████▍| 294/312 [29:47<01:49,  6.11s/it]
Training Epoch: 18/100, batch 588/625 completed (loss: 0.0013:  95%|█████████▍| 295/312 [29:50<01:44,  6.13s/it]
Training Epoch: 18/100, batch 588/625 completed (loss: 0.0024:  95%|█████████▍| 295/312 [29:50<01:44,  6.13s/it]
Training Epoch: 18/100, batch 588/625 completed (loss: 0.0833:  95%|█████████▍| 295/312 [29:50<01:44,  6.13s/it]
Training Epoch: 18/100, batch 588/625 completed (loss: 0.0059:  95%|█████████▍| 295/312 [29:50<01:44,  6.14s/it]
Training Epoch: 18/100, batch 589/625 completed (loss: 0.0128:  95%|█████████▍| 295/312 [29:50<01:44,  6.13s/it]
Training Epoch: 18/100, batch 589/625 completed (loss: 0.0048:  95%|█████████▍| 295/312 [29:50<01:44,  6.13s/it]
Training Epoch: 18/100, batch 589/625 completed (loss: 0.0139:  95%|█████████▍| 295/312 [29:50<01:44,  6.14s/it]
Training Epoch: 18/100, batch 589/625 completed (loss: 0.0016:  95%|█████████▍| 295/312 [29:50<01:44,  6.13s/it]
Training Epoch: 18/100, batch 590/625 completed (loss: 0.0002:  95%|█████████▍| 295/312 [29:53<01:44,  6.13s/it]
Training Epoch: 18/100, batch 590/625 completed (loss: 0.0093:  95%|█████████▍| 295/312 [29:53<01:44,  6.13s/it]
Training Epoch: 18/100, batch 590/625 completed (loss: 0.0006:  95%|█████████▍| 295/312 [29:53<01:44,  6.14s/it]
Training Epoch: 18/100, batch 590/625 completed (loss: 0.0075:  95%|█████████▍| 295/312 [29:53<01:44,  6.13s/it]
Training Epoch: 18/100, batch 590/625 completed (loss: 0.0093:  95%|█████████▍| 296/312 [29:56<01:37,  6.09s/it]
Training Epoch: 18/100, batch 590/625 completed (loss: 0.0002:  95%|█████████▍| 296/312 [29:56<01:37,  6.09s/it]
Training Epoch: 18/100, batch 590/625 completed (loss: 0.0075:  95%|█████████▍| 296/312 [29:56<01:37,  6.09s/it]
Training Epoch: 18/100, batch 590/625 completed (loss: 0.0006:  95%|█████████▍| 296/312 [29:56<01:37,  6.09s/it]
Training Epoch: 18/100, batch 591/625 completed (loss: 0.0011:  95%|█████████▍| 296/312 [29:56<01:37,  6.09s/it]
Training Epoch: 18/100, batch 591/625 completed (loss: 0.0282:  95%|█████████▍| 296/312 [29:56<01:37,  6.09s/it]
Training Epoch: 18/100, batch 591/625 completed (loss: 0.0085:  95%|█████████▍| 296/312 [29:56<01:37,  6.09s/it]
Training Epoch: 18/100, batch 591/625 completed (loss: 0.0315:  95%|█████████▍| 296/312 [29:56<01:37,  6.09s/it]
Training Epoch: 18/100, batch 592/625 completed (loss: 0.0114:  95%|█████████▍| 296/312 [29:59<01:37,  6.09s/it]
Training Epoch: 18/100, batch 592/625 completed (loss: 0.0015:  95%|█████████▍| 296/312 [29:59<01:37,  6.09s/it]
Training Epoch: 18/100, batch 592/625 completed (loss: 0.0006:  95%|█████████▍| 296/312 [29:59<01:37,  6.09s/it]
Training Epoch: 18/100, batch 592/625 completed (loss: 0.0001:  95%|█████████▍| 296/312 [29:59<01:37,  6.09s/it]
Training Epoch: 18/100, batch 592/625 completed (loss: 0.0015:  95%|█████████▌| 297/312 [30:02<01:30,  6.06s/it]
Training Epoch: 18/100, batch 592/625 completed (loss: 0.0001:  95%|█████████▌| 297/312 [30:02<01:30,  6.06s/it]
Training Epoch: 18/100, batch 592/625 completed (loss: 0.0114:  95%|█████████▌| 297/312 [30:02<01:31,  6.07s/it]
Training Epoch: 18/100, batch 592/625 completed (loss: 0.0006:  95%|█████████▌| 297/312 [30:02<01:31,  6.07s/it]
Training Epoch: 18/100, batch 593/625 completed (loss: 0.0003:  95%|█████████▌| 297/312 [30:02<01:30,  6.06s/it]
Training Epoch: 18/100, batch 593/625 completed (loss: 0.0061:  95%|█████████▌| 297/312 [30:02<01:31,  6.07s/it]
Training Epoch: 18/100, batch 593/625 completed (loss: 0.0002:  95%|█████████▌| 297/312 [30:02<01:31,  6.07s/it]
Training Epoch: 18/100, batch 593/625 completed (loss: 0.0065:  95%|█████████▌| 297/312 [30:02<01:30,  6.06s/it]
Training Epoch: 18/100, batch 594/625 completed (loss: 0.0007:  95%|█████████▌| 297/312 [30:05<01:30,  6.06s/it]
Training Epoch: 18/100, batch 594/625 completed (loss: 0.0032:  95%|█████████▌| 297/312 [30:05<01:31,  6.07s/it]
Training Epoch: 18/100, batch 594/625 completed (loss: 0.0007:  95%|█████████▌| 297/312 [30:05<01:31,  6.07s/it]
Training Epoch: 18/100, batch 594/625 completed (loss: 0.0064:  95%|█████████▌| 297/312 [30:05<01:30,  6.06s/it]
Training Epoch: 18/100, batch 594/625 completed (loss: 0.0032:  96%|█████████▌| 298/312 [30:08<01:24,  6.06s/it]
Training Epoch: 18/100, batch 594/625 completed (loss: 0.0064:  96%|█████████▌| 298/312 [30:08<01:24,  6.06s/it]
Training Epoch: 18/100, batch 594/625 completed (loss: 0.0007:  96%|█████████▌| 298/312 [30:08<01:24,  6.07s/it]
Training Epoch: 18/100, batch 594/625 completed (loss: 0.0007:  96%|█████████▌| 298/312 [30:08<01:24,  6.06s/it]
Training Epoch: 18/100, batch 595/625 completed (loss: 0.0003:  96%|█████████▌| 298/312 [30:08<01:24,  6.06s/it]
Training Epoch: 18/100, batch 595/625 completed (loss: 0.0081:  96%|█████████▌| 298/312 [30:08<01:24,  6.07s/it]
Training Epoch: 18/100, batch 595/625 completed (loss: 0.002:  96%|█████████▌| 298/312 [30:08<01:24,  6.06s/it] 
Training Epoch: 18/100, batch 595/625 completed (loss: 0.0002:  96%|█████████▌| 298/312 [30:08<01:24,  6.06s/it]
Training Epoch: 18/100, batch 596/625 completed (loss: 0.0001:  96%|█████████▌| 298/312 [30:11<01:24,  6.06s/it]
Training Epoch: 18/100, batch 596/625 completed (loss: 0.0002:  96%|█████████▌| 298/312 [30:11<01:24,  6.06s/it]
Training Epoch: 18/100, batch 596/625 completed (loss: 0.0001:  96%|█████████▌| 298/312 [30:11<01:24,  6.07s/it]
Training Epoch: 18/100, batch 596/625 completed (loss: 0.005:  96%|█████████▌| 298/312 [30:11<01:24,  6.06s/it] 
Training Epoch: 18/100, batch 596/625 completed (loss: 0.005:  96%|█████████▌| 299/312 [30:15<01:19,  6.09s/it]
Training Epoch: 18/100, batch 596/625 completed (loss: 0.0002:  96%|█████████▌| 299/312 [30:14<01:19,  6.09s/it]
Training Epoch: 18/100, batch 596/625 completed (loss: 0.0001:  96%|█████████▌| 299/312 [30:14<01:19,  6.09s/it]
Training Epoch: 18/100, batch 596/625 completed (loss: 0.0001:  96%|█████████▌| 299/312 [30:14<01:19,  6.09s/it]
Training Epoch: 18/100, batch 597/625 completed (loss: 0.0002:  96%|█████████▌| 299/312 [30:14<01:19,  6.09s/it]
Training Epoch: 18/100, batch 597/625 completed (loss: 0.0018:  96%|█████████▌| 299/312 [30:15<01:19,  6.09s/it]
Training Epoch: 18/100, batch 597/625 completed (loss: 0.0015:  96%|█████████▌| 299/312 [30:15<01:19,  6.09s/it]
Training Epoch: 18/100, batch 597/625 completed (loss: 0.0048:  96%|█████████▌| 299/312 [30:15<01:19,  6.09s/it]
Training Epoch: 18/100, batch 598/625 completed (loss: 0.0034:  96%|█████████▌| 299/312 [30:17<01:19,  6.09s/it]
Training Epoch: 18/100, batch 598/625 completed (loss: 0.1266:  96%|█████████▌| 299/312 [30:17<01:19,  6.09s/it]
Training Epoch: 18/100, batch 598/625 completed (loss: 0.0114:  96%|█████████▌| 299/312 [30:17<01:19,  6.09s/it]
Training Epoch: 18/100, batch 598/625 completed (loss: 0.0019:  96%|█████████▌| 299/312 [30:18<01:19,  6.09s/it]
Training Epoch: 18/100, batch 598/625 completed (loss: 0.1266:  96%|█████████▌| 300/312 [30:21<01:12,  6.07s/it]
Training Epoch: 18/100, batch 598/625 completed (loss: 0.0019:  96%|█████████▌| 300/312 [30:21<01:12,  6.07s/it]
Training Epoch: 18/100, batch 598/625 completed (loss: 0.0114:  96%|█████████▌| 300/312 [30:20<01:12,  6.07s/it]
Training Epoch: 18/100, batch 598/625 completed (loss: 0.0034:  96%|█████████▌| 300/312 [30:21<01:12,  6.07s/it]
Training Epoch: 18/100, batch 599/625 completed (loss: 0.0004:  96%|█████████▌| 300/312 [30:21<01:12,  6.07s/it]
Training Epoch: 18/100, batch 599/625 completed (loss: 0.0001:  96%|█████████▌| 300/312 [30:21<01:12,  6.07s/it]
Training Epoch: 18/100, batch 599/625 completed (loss: 0.0046:  96%|█████████▌| 300/312 [30:21<01:12,  6.07s/it]
Training Epoch: 18/100, batch 599/625 completed (loss: 0.0151:  96%|█████████▌| 300/312 [30:21<01:12,  6.07s/it]
Training Epoch: 18/100, batch 600/625 completed (loss: 0.0242:  96%|█████████▌| 300/312 [30:24<01:12,  6.07s/it]
Training Epoch: 18/100, batch 600/625 completed (loss: 0.2646:  96%|█████████▌| 300/312 [30:24<01:12,  6.07s/it]
Training Epoch: 18/100, batch 600/625 completed (loss: 0.0152:  96%|█████████▌| 300/312 [30:24<01:12,  6.07s/it]
Training Epoch: 18/100, batch 600/625 completed (loss: 0.0081:  96%|█████████▌| 300/312 [30:24<01:12,  6.07s/it]
Training Epoch: 18/100, batch 600/625 completed (loss: 0.2646:  96%|█████████▋| 301/312 [30:27<01:07,  6.10s/it]
Training Epoch: 18/100, batch 600/625 completed (loss: 0.0152:  96%|█████████▋| 301/312 [30:27<01:07,  6.10s/it]
Training Epoch: 18/100, batch 600/625 completed (loss: 0.0242:  96%|█████████▋| 301/312 [30:27<01:07,  6.10s/it]
Training Epoch: 18/100, batch 600/625 completed (loss: 0.0081:  96%|█████████▋| 301/312 [30:27<01:07,  6.11s/it]
Training Epoch: 18/100, batch 601/625 completed (loss: 0.0057:  96%|█████████▋| 301/312 [30:27<01:07,  6.10s/it]
Training Epoch: 18/100, batch 601/625 completed (loss: 0.057:  96%|█████████▋| 301/312 [30:27<01:07,  6.10s/it] 
Training Epoch: 18/100, batch 601/625 completed (loss: 0.0023:  96%|█████████▋| 301/312 [30:27<01:07,  6.10s/it]
Training Epoch: 18/100, batch 601/625 completed (loss: 0.0881:  96%|█████████▋| 301/312 [30:27<01:07,  6.11s/it]
Training Epoch: 18/100, batch 602/625 completed (loss: 0.0019:  96%|█████████▋| 301/312 [30:30<01:07,  6.10s/it]
Training Epoch: 18/100, batch 602/625 completed (loss: 0.0185:  96%|█████████▋| 301/312 [30:30<01:07,  6.10s/it]
Training Epoch: 18/100, batch 602/625 completed (loss: 0.0127:  96%|█████████▋| 301/312 [30:30<01:07,  6.10s/it]
Training Epoch: 18/100, batch 602/625 completed (loss: 0.0003:  96%|█████████▋| 301/312 [30:30<01:07,  6.11s/it]
Training Epoch: 18/100, batch 602/625 completed (loss: 0.0185:  97%|█████████▋| 302/312 [30:33<01:00,  6.08s/it]
Training Epoch: 18/100, batch 602/625 completed (loss: 0.0003:  97%|█████████▋| 302/312 [30:33<01:00,  6.08s/it]
Training Epoch: 18/100, batch 602/625 completed (loss: 0.0127:  97%|█████████▋| 302/312 [30:33<01:00,  6.08s/it]
Training Epoch: 18/100, batch 602/625 completed (loss: 0.0019:  97%|█████████▋| 302/312 [30:33<01:00,  6.09s/it]
Training Epoch: 18/100, batch 603/625 completed (loss: 0.0112:  97%|█████████▋| 302/312 [30:33<01:00,  6.08s/it]
Training Epoch: 18/100, batch 603/625 completed (loss: 0.001:  97%|█████████▋| 302/312 [30:33<01:00,  6.08s/it] 
Training Epoch: 18/100, batch 603/625 completed (loss: 0.0057:  97%|█████████▋| 302/312 [30:33<01:00,  6.09s/it]
Training Epoch: 18/100, batch 603/625 completed (loss: 0.0014:  97%|█████████▋| 302/312 [30:33<01:00,  6.08s/it]
Training Epoch: 18/100, batch 604/625 completed (loss: 0.0175:  97%|█████████▋| 302/312 [30:36<01:00,  6.09s/it]
Training Epoch: 18/100, batch 604/625 completed (loss: 0.0002:  97%|█████████▋| 302/312 [30:36<01:00,  6.08s/it]
Training Epoch: 18/100, batch 604/625 completed (loss: 0.0046:  97%|█████████▋| 302/312 [30:36<01:00,  6.08s/it]
Training Epoch: 18/100, batch 604/625 completed (loss: 0.0005:  97%|█████████▋| 302/312 [30:36<01:00,  6.08s/it]
Training Epoch: 18/100, batch 604/625 completed (loss: 0.0046:  97%|█████████▋| 303/312 [30:39<00:55,  6.12s/it]
Training Epoch: 18/100, batch 604/625 completed (loss: 0.0005:  97%|█████████▋| 303/312 [30:39<00:55,  6.12s/it]
Training Epoch: 18/100, batch 604/625 completed (loss: 0.0175:  97%|█████████▋| 303/312 [30:39<00:55,  6.12s/it]
Training Epoch: 18/100, batch 604/625 completed (loss: 0.0002:  97%|█████████▋| 303/312 [30:39<00:55,  6.12s/it]
Training Epoch: 18/100, batch 605/625 completed (loss: 0.0152:  97%|█████████▋| 303/312 [30:39<00:55,  6.12s/it]
Training Epoch: 18/100, batch 605/625 completed (loss: 0.0003:  97%|█████████▋| 303/312 [30:39<00:55,  6.12s/it]
Training Epoch: 18/100, batch 605/625 completed (loss: 0.0027:  97%|█████████▋| 303/312 [30:39<00:55,  6.12s/it]
Training Epoch: 18/100, batch 605/625 completed (loss: 0.0002:  97%|█████████▋| 303/312 [30:39<00:55,  6.12s/it]
Training Epoch: 18/100, batch 606/625 completed (loss: 0.05:  97%|█████████▋| 303/312 [30:42<00:55,  6.12s/it]  
Training Epoch: 18/100, batch 606/625 completed (loss: 0.0013:  97%|█████████▋| 303/312 [30:42<00:55,  6.12s/it]
Training Epoch: 18/100, batch 606/625 completed (loss: 0.0002:  97%|█████████▋| 303/312 [30:42<00:55,  6.12s/it]
Training Epoch: 18/100, batch 606/625 completed (loss: 0.0061:  97%|█████████▋| 303/312 [30:42<00:55,  6.12s/it]
Training Epoch: 18/100, batch 606/625 completed (loss: 0.0061:  97%|█████████▋| 304/312 [30:45<00:49,  6.13s/it]
Training Epoch: 18/100, batch 606/625 completed (loss: 0.05:  97%|█████████▋| 304/312 [30:45<00:49,  6.13s/it]
Training Epoch: 18/100, batch 606/625 completed (loss: 0.0013:  97%|█████████▋| 304/312 [30:45<00:49,  6.13s/it]
Training Epoch: 18/100, batch 606/625 completed (loss: 0.0002:  97%|█████████▋| 304/312 [30:45<00:49,  6.13s/it]
Training Epoch: 18/100, batch 607/625 completed (loss: 0.5536:  97%|█████████▋| 304/312 [30:45<00:49,  6.13s/it]
Training Epoch: 18/100, batch 607/625 completed (loss: 0.0268:  97%|█████████▋| 304/312 [30:45<00:49,  6.13s/it]
Training Epoch: 18/100, batch 607/625 completed (loss: 0.0051:  97%|█████████▋| 304/312 [30:45<00:49,  6.13s/it]
Training Epoch: 18/100, batch 607/625 completed (loss: 0.0352:  97%|█████████▋| 304/312 [30:45<00:49,  6.13s/it]
Training Epoch: 18/100, batch 608/625 completed (loss: 0.0171:  97%|█████████▋| 304/312 [30:48<00:49,  6.13s/it]
Training Epoch: 18/100, batch 608/625 completed (loss: 0.0084:  97%|█████████▋| 304/312 [30:48<00:49,  6.13s/it]
Training Epoch: 18/100, batch 608/625 completed (loss: 0.0036:  97%|█████████▋| 304/312 [30:48<00:49,  6.13s/it]
Training Epoch: 18/100, batch 608/625 completed (loss: 0.0345:  97%|█████████▋| 304/312 [30:48<00:49,  6.13s/it]
Training Epoch: 18/100, batch 608/625 completed (loss: 0.0345:  98%|█████████▊| 305/312 [30:51<00:42,  6.09s/it]
Training Epoch: 18/100, batch 608/625 completed (loss: 0.0036:  98%|█████████▊| 305/312 [30:51<00:42,  6.09s/it]
Training Epoch: 18/100, batch 608/625 completed (loss: 0.0171:  98%|█████████▊| 305/312 [30:51<00:42,  6.09s/it]
Training Epoch: 18/100, batch 608/625 completed (loss: 0.0084:  98%|█████████▊| 305/312 [30:51<00:42,  6.09s/it]
Training Epoch: 18/100, batch 609/625 completed (loss: 0.0432:  98%|█████████▊| 305/312 [30:51<00:42,  6.09s/it]
Training Epoch: 18/100, batch 609/625 completed (loss: 0.0053:  98%|█████████▊| 305/312 [30:51<00:42,  6.09s/it]
Training Epoch: 18/100, batch 609/625 completed (loss: 0.0017:  98%|█████████▊| 305/312 [30:51<00:42,  6.09s/it]
Training Epoch: 18/100, batch 609/625 completed (loss: 0.0003:  98%|█████████▊| 305/312 [30:51<00:42,  6.09s/it]
Training Epoch: 18/100, batch 610/625 completed (loss: 0.0523:  98%|█████████▊| 305/312 [30:54<00:42,  6.09s/it]
Training Epoch: 18/100, batch 610/625 completed (loss: 0.0239:  98%|█████████▊| 305/312 [30:54<00:42,  6.09s/it]
Training Epoch: 18/100, batch 610/625 completed (loss: 0.0283:  98%|█████████▊| 305/312 [30:54<00:42,  6.09s/it]
Training Epoch: 18/100, batch 610/625 completed (loss: 0.0001:  98%|█████████▊| 305/312 [30:54<00:42,  6.09s/it]
Training Epoch: 18/100, batch 610/625 completed (loss: 0.0523:  98%|█████████▊| 306/312 [30:57<00:36,  6.08s/it]
Training Epoch: 18/100, batch 610/625 completed (loss: 0.0001:  98%|█████████▊| 306/312 [30:57<00:36,  6.08s/it]
Training Epoch: 18/100, batch 610/625 completed (loss: 0.0239:  98%|█████████▊| 306/312 [30:57<00:36,  6.08s/it]
Training Epoch: 18/100, batch 610/625 completed (loss: 0.0283:  98%|█████████▊| 306/312 [30:57<00:36,  6.08s/it]
Training Epoch: 18/100, batch 611/625 completed (loss: 0.0024:  98%|█████████▊| 306/312 [30:57<00:36,  6.08s/it]
Training Epoch: 18/100, batch 611/625 completed (loss: 0.0056:  98%|█████████▊| 306/312 [30:57<00:36,  6.08s/it]
Training Epoch: 18/100, batch 611/625 completed (loss: 0.0362:  98%|█████████▊| 306/312 [30:57<00:36,  6.08s/it]
Training Epoch: 18/100, batch 611/625 completed (loss: 0.0101:  98%|█████████▊| 306/312 [30:57<00:36,  6.08s/it]
Training Epoch: 18/100, batch 612/625 completed (loss: 0.007:  98%|█████████▊| 306/312 [31:00<00:36,  6.08s/it] 
Training Epoch: 18/100, batch 612/625 completed (loss: 0.0103:  98%|█████████▊| 306/312 [31:00<00:36,  6.08s/it]
Training Epoch: 18/100, batch 612/625 completed (loss: 0.0002:  98%|█████████▊| 306/312 [31:00<00:36,  6.08s/it]
Training Epoch: 18/100, batch 612/625 completed (loss: 0.0079:  98%|█████████▊| 306/312 [31:00<00:36,  6.08s/it]
Training Epoch: 18/100, batch 612/625 completed (loss: 0.0002:  98%|█████████▊| 307/312 [31:03<00:30,  6.09s/it]
Training Epoch: 18/100, batch 612/625 completed (loss: 0.007:  98%|█████████▊| 307/312 [31:03<00:30,  6.09s/it]
Training Epoch: 18/100, batch 612/625 completed (loss: 0.0103:  98%|█████████▊| 307/312 [31:03<00:30,  6.09s/it]
Training Epoch: 18/100, batch 612/625 completed (loss: 0.0079:  98%|█████████▊| 307/312 [31:03<00:30,  6.10s/it]
Training Epoch: 18/100, batch 613/625 completed (loss: 0.0002:  98%|█████████▊| 307/312 [31:03<00:30,  6.09s/it]
Training Epoch: 18/100, batch 613/625 completed (loss: 0.0003:  98%|█████████▊| 307/312 [31:03<00:30,  6.09s/it]
Training Epoch: 18/100, batch 613/625 completed (loss: 0.0029:  98%|█████████▊| 307/312 [31:03<00:30,  6.09s/it]
Training Epoch: 18/100, batch 613/625 completed (loss: 0.003:  98%|█████████▊| 307/312 [31:03<00:30,  6.10s/it] 
Training Epoch: 18/100, batch 614/625 completed (loss: 0.0058:  98%|█████████▊| 307/312 [31:06<00:30,  6.09s/it]
Training Epoch: 18/100, batch 614/625 completed (loss: 0.0007:  98%|█████████▊| 307/312 [31:06<00:30,  6.09s/it]
Training Epoch: 18/100, batch 614/625 completed (loss: 0.0017:  98%|█████████▊| 307/312 [31:06<00:30,  6.09s/it]
Training Epoch: 18/100, batch 614/625 completed (loss: 0.0117:  98%|█████████▊| 307/312 [31:06<00:30,  6.10s/it]
Training Epoch: 18/100, batch 614/625 completed (loss: 0.0117:  99%|█████████▊| 308/312 [31:09<00:24,  6.07s/it]
Training Epoch: 18/100, batch 614/625 completed (loss: 0.0058:  99%|█████████▊| 308/312 [31:09<00:24,  6.08s/it]
Training Epoch: 18/100, batch 614/625 completed (loss: 0.0017:  99%|█████████▊| 308/312 [31:09<00:24,  6.08s/it]
Training Epoch: 18/100, batch 614/625 completed (loss: 0.0007:  99%|█████████▊| 308/312 [31:09<00:24,  6.08s/it]
Training Epoch: 18/100, batch 615/625 completed (loss: 0.0018:  99%|█████████▊| 308/312 [31:09<00:24,  6.08s/it]
Training Epoch: 18/100, batch 615/625 completed (loss: 0.0043:  99%|█████████▊| 308/312 [31:09<00:24,  6.08s/it]
Training Epoch: 18/100, batch 615/625 completed (loss: 0.0004:  99%|█████████▊| 308/312 [31:09<00:24,  6.08s/it]
Training Epoch: 18/100, batch 615/625 completed (loss: 0.0052:  99%|█████████▊| 308/312 [31:09<00:24,  6.07s/it]
Training Epoch: 18/100, batch 616/625 completed (loss: 0.0006:  99%|█████████▊| 308/312 [31:13<00:24,  6.08s/it]
Training Epoch: 18/100, batch 616/625 completed (loss: 0.0017:  99%|█████████▊| 308/312 [31:13<00:24,  6.08s/it]
Training Epoch: 18/100, batch 616/625 completed (loss: 0.0038:  99%|█████████▊| 308/312 [31:13<00:24,  6.08s/it]
Training Epoch: 18/100, batch 616/625 completed (loss: 0.0538:  99%|█████████▊| 308/312 [31:13<00:24,  6.07s/it]
Training Epoch: 18/100, batch 616/625 completed (loss: 0.0538:  99%|█████████▉| 309/312 [31:16<00:18,  6.11s/it]
Training Epoch: 18/100, batch 616/625 completed (loss: 0.0017:  99%|█████████▉| 309/312 [31:15<00:18,  6.11s/it]
Training Epoch: 18/100, batch 616/625 completed (loss: 0.0006:  99%|█████████▉| 309/312 [31:15<00:18,  6.11s/it]
Training Epoch: 18/100, batch 616/625 completed (loss: 0.0038:  99%|█████████▉| 309/312 [31:15<00:18,  6.11s/it]
Training Epoch: 18/100, batch 617/625 completed (loss: 0.0482:  99%|█████████▉| 309/312 [31:15<00:18,  6.11s/it]
Training Epoch: 18/100, batch 617/625 completed (loss: 0.0024:  99%|█████████▉| 309/312 [31:16<00:18,  6.11s/it]
Training Epoch: 18/100, batch 617/625 completed (loss: 0.0083:  99%|█████████▉| 309/312 [31:16<00:18,  6.11s/it]
Training Epoch: 18/100, batch 617/625 completed (loss: 0.0053:  99%|█████████▉| 309/312 [31:16<00:18,  6.11s/it]
Training Epoch: 18/100, batch 618/625 completed (loss: 0.0032:  99%|█████████▉| 309/312 [31:19<00:18,  6.11s/it]
Training Epoch: 18/100, batch 618/625 completed (loss: 0.0028:  99%|█████████▉| 309/312 [31:19<00:18,  6.11s/it]
Training Epoch: 18/100, batch 618/625 completed (loss: 0.0044:  99%|█████████▉| 309/312 [31:19<00:18,  6.11s/it]
Training Epoch: 18/100, batch 618/625 completed (loss: 0.013:  99%|█████████▉| 309/312 [31:19<00:18,  6.11s/it] 
Training Epoch: 18/100, batch 618/625 completed (loss: 0.0032:  99%|█████████▉| 310/312 [31:21<00:12,  6.08s/it]
Training Epoch: 18/100, batch 618/625 completed (loss: 0.0028:  99%|█████████▉| 310/312 [31:21<00:12,  6.08s/it]
Training Epoch: 18/100, batch 618/625 completed (loss: 0.0044:  99%|█████████▉| 310/312 [31:21<00:12,  6.08s/it]
Training Epoch: 18/100, batch 618/625 completed (loss: 0.013:  99%|█████████▉| 310/312 [31:22<00:12,  6.09s/it]
Training Epoch: 18/100, batch 619/625 completed (loss: 0.1405:  99%|█████████▉| 310/312 [31:22<00:12,  6.08s/it]
Training Epoch: 18/100, batch 619/625 completed (loss: 0.0052:  99%|█████████▉| 310/312 [31:22<00:12,  6.08s/it]
Training Epoch: 18/100, batch 619/625 completed (loss: 0.0011:  99%|█████████▉| 310/312 [31:22<00:12,  6.08s/it]
Training Epoch: 18/100, batch 619/625 completed (loss: 0.0161:  99%|█████████▉| 310/312 [31:22<00:12,  6.09s/it]
Training Epoch: 18/100, batch 620/625 completed (loss: 0.0029:  99%|█████████▉| 310/312 [31:25<00:12,  6.08s/it]
Training Epoch: 18/100, batch 620/625 completed (loss: 0.002:  99%|█████████▉| 310/312 [31:25<00:12,  6.08s/it] 
Training Epoch: 18/100, batch 620/625 completed (loss: 0.0644:  99%|█████████▉| 310/312 [31:25<00:12,  6.08s/it]
Training Epoch: 18/100, batch 620/625 completed (loss: 0.0013:  99%|█████████▉| 310/312 [31:25<00:12,  6.09s/it]
Training Epoch: 18/100, batch 620/625 completed (loss: 0.0644: 100%|█████████▉| 311/312 [31:28<00:06,  6.12s/it]
Training Epoch: 18/100, batch 620/625 completed (loss: 0.0013: 100%|█████████▉| 311/312 [31:28<00:06,  6.12s/it]
Training Epoch: 18/100, batch 620/625 completed (loss: 0.0029: 100%|█████████▉| 311/312 [31:28<00:06,  6.12s/it]
Training Epoch: 18/100, batch 620/625 completed (loss: 0.002: 100%|█████████▉| 311/312 [31:28<00:06,  6.12s/it]
Training Epoch: 18/100, batch 621/625 completed (loss: 0.0067: 100%|█████████▉| 311/312 [31:28<00:06,  6.12s/it]
Training Epoch: 18/100, batch 621/625 completed (loss: 0.0013: 100%|█████████▉| 311/312 [31:28<00:06,  6.12s/it]
Training Epoch: 18/100, batch 621/625 completed (loss: 0.0041: 100%|█████████▉| 311/312 [31:28<00:06,  6.12s/it]
Training Epoch: 18/100, batch 621/625 completed (loss: 0.0077: 100%|█████████▉| 311/312 [31:28<00:06,  6.12s/it]
Training Epoch: 18/100, batch 622/625 completed (loss: 0.0337: 100%|█████████▉| 311/312 [31:31<00:06,  6.12s/it]
Training Epoch: 18/100, batch 622/625 completed (loss: 0.0012: 100%|█████████▉| 311/312 [31:31<00:06,  6.12s/it]
Training Epoch: 18/100, batch 622/625 completed (loss: 0.0004: 100%|█████████▉| 311/312 [31:31<00:06,  6.12s/it]
Training Epoch: 18/100, batch 622/625 completed (loss: 0.01: 100%|█████████▉| 311/312 [31:31<00:06,  6.12s/it]  
Training Epoch: 18/100, batch 622/625 completed (loss: 0.0337: 100%|██████████| 312/312 [31:34<00:00,  6.13s/it]
Training Epoch: 18/100, batch 622/625 completed (loss: 0.01: 100%|██████████| 312/312 [31:34<00:00,  6.13s/it]
Training Epoch: 18/100, batch 622/625 completed (loss: 0.0004: 100%|██████████| 312/312 [31:34<00:00,  6.13s/it]
Training Epoch: 18/100, batch 622/625 completed (loss: 0.0012: 100%|██████████| 312/312 [31:34<00:00,  6.13s/it]
Training Epoch: 18/100, batch 623/625 completed (loss: 0.0006: 100%|██████████| 312/312 [31:34<00:00,  6.13s/it]
Training Epoch: 18/100, batch 623/625 completed (loss: 0.0014: 100%|██████████| 312/312 [31:34<00:00,  6.13s/it]
Training Epoch: 18/100, batch 623/625 completed (loss: 0.0203: 100%|██████████| 312/312 [31:34<00:00,  6.13s/it]
Training Epoch: 18/100, batch 623/625 completed (loss: 0.0059: 100%|██████████| 312/312 [31:34<00:00,  6.13s/it]
Training Epoch: 18/100, batch 623/625 completed (loss: 0.0203: : 313it [31:37,  5.17s/it]                       
Training Epoch: 18/100, batch 623/625 completed (loss: 0.0014: : 313it [31:37,  5.16s/it]                       
Training Epoch: 18/100, batch 623/625 completed (loss: 0.0059: : 313it [31:37,  5.17s/it]                       
Training Epoch: 18/100, batch 623/625 completed (loss: 0.0006: : 313it [31:37,  5.17s/it]                       
Training Epoch: 18/100, batch 624/625 completed (loss: 0.0035: : 313it [31:37,  5.16s/it]
Training Epoch: 18/100, batch 624/625 completed (loss: 0.0194: : 313it [31:37,  5.17s/it]
Training Epoch: 18/100, batch 624/625 completed (loss: 0.0014: : 313it [31:37,  5.17s/it]
Training Epoch: 18/100, batch 624/625 completed (loss: 0.0009: : 313it [31:37,  5.17s/it]
Training Epoch: 18/100, batch 624/625 completed (loss: 0.0014: : 313it [31:37,  6.06s/it]

Training Epoch: 18/100, batch 624/625 completed (loss: 0.0194: : 313it [31:37,  6.06s/it]
Training Epoch: 18/100, batch 624/625 completed (loss: 0.0035: : 313it [31:37,  6.06s/it]


Training Epoch: 18/100, batch 624/625 completed (loss: 0.0009: : 313it [31:37,  6.06s/it]
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/c10d_logger.py:83: UserWarning: barrier(): using the device under current context. You can specify `device_id` in `init_process_group` to mute this warning.
  return func(*args, **kwargs)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
saving model + train state.

Processing Files (0 / 0)      : |          |  0.00B /  0.00B            

New Data Upload               : |          |  0.00B /  0.00B            


  ...ve-03b89ee9/checkpoint_18:   0%|          | 16.5MB / 3.63GB            


  ...ve-03b89ee9/checkpoint_18:   0%|          | 16.5MB / 3.63GB            
Processing Files (0 / 1)      :   0%|          | 16.5MB / 3.63GB,   ???B/s  

New Data Upload               :  12%|█▏        | 16.5MB /  134MB,   ???B/s  


  ...ve-03b89ee9/checkpoint_18:   3%|▎         |  109MB / 3.63GB            
Processing Files (0 / 1)      :   3%|▎         |  109MB / 3.63GB, 10.6MB/s  

New Data Upload               :  54%|█████▍    |  109MB /  201MB, 10.6MB/s  


  ...ve-03b89ee9/checkpoint_18:   4%|▎         |  133MB / 3.63GB            
Processing Files (0 / 1)      :   4%|▎         |  133MB / 3.63GB, 12.7MB/s  

New Data Upload               :  66%|██████▋   |  133MB /  201MB, 12.7MB/s  


  ...ve-03b89ee9/checkpoint_18:   4%|▎         |  134MB / 3.63GB            
Processing Files (0 / 1)      :   4%|▎         |  134MB / 3.63GB, 12.2MB/s  

New Data Upload               :  66%|██████▋   |  134MB /  201MB, 12.2MB/s  


  ...ve-03b89ee9/checkpoint_18:   5%|▌         |  186MB / 3.63GB            
Processing Files (0 / 1)      :   5%|▌         |  186MB / 3.63GB, 16.8MB/s  

New Data Upload               :  55%|█████▌    |  186MB /  335MB, 16.8MB/s  


  ...ve-03b89ee9/checkpoint_18:   8%|▊         |  279MB / 3.63GB            
Processing Files (0 / 1)      :   8%|▊         |  279MB / 3.63GB, 25.2MB/s  

New Data Upload               :  69%|██████▉   |  279MB /  402MB, 25.2MB/s  


  ...ve-03b89ee9/checkpoint_18:   9%|▉         |  335MB / 3.63GB            
Processing Files (0 / 1)      :   9%|▉         |  335MB / 3.63GB, 29.9MB/s  

New Data Upload               :  71%|███████▏  |  335MB /  470MB, 29.9MB/s  


  ...ve-03b89ee9/checkpoint_18:  11%|█         |  392MB / 3.63GB            
Processing Files (0 / 1)      :  11%|█         |  392MB / 3.63GB, 34.5MB/s  

New Data Upload               :  83%|████████▎ |  392MB /  470MB, 34.5MB/s  


  ...ve-03b89ee9/checkpoint_18:  12%|█▏        |  432MB / 3.63GB            
Processing Files (0 / 1)      :  12%|█▏        |  432MB / 3.63GB, 37.4MB/s  

New Data Upload               :  81%|████████  |  432MB /  537MB, 37.4MB/s  


  ...ve-03b89ee9/checkpoint_18:  14%|█▍        |  501MB / 3.63GB            
Processing Files (0 / 1)      :  14%|█▍        |  501MB / 3.63GB, 42.9MB/s  

New Data Upload               :  83%|████████▎ |  501MB /  604MB, 42.9MB/s  


  ...ve-03b89ee9/checkpoint_18:  17%|█▋        |  603MB / 3.63GB            
Processing Files (0 / 1)      :  17%|█▋        |  603MB / 3.63GB, 51.2MB/s  

New Data Upload               :  90%|████████▉ |  603MB /  671MB, 51.2MB/s  


  ...ve-03b89ee9/checkpoint_18:  18%|█▊        |  661MB / 3.63GB            
Processing Files (0 / 1)      :  18%|█▊        |  661MB / 3.63GB, 55.4MB/s  

New Data Upload               :  82%|████████▏ |  661MB /  805MB, 55.4MB/s  


  ...ve-03b89ee9/checkpoint_18:  21%|██        |  749MB / 3.63GB            
Processing Files (0 / 1)      :  21%|██        |  749MB / 3.63GB, 62.1MB/s  

New Data Upload               :  86%|████████▌ |  749MB /  872MB, 62.1MB/s  


  ...ve-03b89ee9/checkpoint_18:  23%|██▎       |  841MB / 3.63GB            
Processing Files (0 / 1)      :  23%|██▎       |  841MB / 3.63GB, 69.0MB/s  

New Data Upload               :  90%|████████▉ |  841MB /  939MB, 69.0MB/s  


  ...ve-03b89ee9/checkpoint_18:  26%|██▌       |  929MB / 3.63GB            
Processing Files (0 / 1)      :  26%|██▌       |  929MB / 3.63GB, 75.4MB/s  

New Data Upload               :  92%|█████████▏|  929MB / 1.01GB, 75.4MB/s  


  ...ve-03b89ee9/checkpoint_18:  28%|██▊       |  998MB / 3.63GB            
Processing Files (0 / 1)      :  28%|██▊       |  998MB / 3.63GB, 79.8MB/s  

New Data Upload               :  88%|████████▊ |  998MB / 1.14GB, 79.8MB/s  


  ...ve-03b89ee9/checkpoint_18:  30%|██▉       | 1.09GB / 3.63GB            
Processing Files (0 / 1)      :  30%|██▉       | 1.09GB / 3.63GB, 85.8MB/s  

New Data Upload               :  90%|████████▉ | 1.09GB / 1.21GB, 85.8MB/s  


  ...ve-03b89ee9/checkpoint_18:  33%|███▎      | 1.20GB / 3.63GB            
Processing Files (0 / 1)      :  33%|███▎      | 1.20GB / 3.63GB, 94.1MB/s  

New Data Upload               :  90%|████████▉ | 1.20GB / 1.34GB, 94.1MB/s  


  ...ve-03b89ee9/checkpoint_18:  36%|███▌      | 1.29GB / 3.63GB            
Processing Files (0 / 1)      :  36%|███▌      | 1.29GB / 3.63GB, 99.9MB/s  

New Data Upload               :  92%|█████████▏| 1.29GB / 1.41GB, 99.9MB/s  


  ...ve-03b89ee9/checkpoint_18:  38%|███▊      | 1.37GB / 3.63GB            
Processing Files (0 / 1)      :  38%|███▊      | 1.37GB / 3.63GB,  104MB/s  

New Data Upload               :  93%|█████████▎| 1.37GB / 1.48GB,  104MB/s  


  ...ve-03b89ee9/checkpoint_18:  41%|████      | 1.47GB / 3.63GB            
Processing Files (0 / 1)      :  41%|████      | 1.47GB / 3.63GB,  111MB/s  

New Data Upload               :  92%|█████████▏| 1.47GB / 1.61GB,  111MB/s  


  ...ve-03b89ee9/checkpoint_18:  43%|████▎     | 1.58GB / 3.63GB            
Processing Files (0 / 1)      :  43%|████▎     | 1.58GB / 3.63GB,  118MB/s  

New Data Upload               :  94%|█████████▍| 1.58GB / 1.68GB,  118MB/s  


  ...ve-03b89ee9/checkpoint_18:  45%|████▌     | 1.65GB / 3.63GB            
Processing Files (0 / 1)      :  45%|████▌     | 1.65GB / 3.63GB,  122MB/s  

New Data Upload               :  95%|█████████▍| 1.65GB / 1.74GB,  122MB/s  


  ...ve-03b89ee9/checkpoint_18:  48%|████▊     | 1.74GB / 3.63GB            
Processing Files (0 / 1)      :  48%|████▊     | 1.74GB / 3.63GB,  127MB/s  

New Data Upload               :  93%|█████████▎| 1.74GB / 1.88GB,  127MB/s  


  ...ve-03b89ee9/checkpoint_18:  50%|█████     | 1.82GB / 3.63GB            
Processing Files (0 / 1)      :  50%|█████     | 1.82GB / 3.63GB,  131MB/s  

New Data Upload               :  93%|█████████▎| 1.82GB / 1.94GB,  131MB/s  


  ...ve-03b89ee9/checkpoint_18:  53%|█████▎    | 1.92GB / 3.63GB            
Processing Files (0 / 1)      :  53%|█████▎    | 1.92GB / 3.63GB,  136MB/s  

New Data Upload               :  92%|█████████▏| 1.92GB / 2.08GB,  136MB/s  


  ...ve-03b89ee9/checkpoint_18:  56%|█████▌    | 2.02GB / 3.63GB            
Processing Files (0 / 1)      :  56%|█████▌    | 2.02GB / 3.63GB,  142MB/s  

New Data Upload               :  94%|█████████▍| 2.02GB / 2.15GB,  142MB/s  


  ...ve-03b89ee9/checkpoint_18:  58%|█████▊    | 2.11GB / 3.63GB            
Processing Files (0 / 1)      :  58%|█████▊    | 2.11GB / 3.63GB,  147MB/s  

New Data Upload               :  93%|█████████▎| 2.11GB / 2.28GB,  147MB/s  


  ...ve-03b89ee9/checkpoint_18:  61%|██████    | 2.21GB / 3.63GB            
Processing Files (0 / 1)      :  61%|██████    | 2.21GB / 3.63GB,  152MB/s  

New Data Upload               :  94%|█████████▍| 2.21GB / 2.35GB,  152MB/s  


  ...ve-03b89ee9/checkpoint_18:  71%|███████▏  | 2.58GB / 3.63GB            
Processing Files (0 / 1)      :  71%|███████▏  | 2.58GB / 3.63GB,  180MB/s  

New Data Upload               :  98%|█████████▊| 2.31GB / 2.35GB,  157MB/s  


  ...ve-03b89ee9/checkpoint_18:  82%|████████▏ | 2.98GB / 3.63GB            
Processing Files (0 / 1)      :  82%|████████▏ | 2.98GB / 3.63GB,  210MB/s  

New Data Upload               : 100%|█████████▉| 2.34GB / 2.35GB,  157MB/s  


  ...ve-03b89ee9/checkpoint_18:  93%|█████████▎| 3.36GB / 3.63GB            
Processing Files (0 / 1)      :  93%|█████████▎| 3.36GB / 3.63GB,  237MB/s  

New Data Upload               : 100%|█████████▉| 2.35GB / 2.35GB,  154MB/s  


  ...ve-03b89ee9/checkpoint_18:  99%|█████████▉| 3.59GB / 3.63GB            
Processing Files (0 / 1)      :  99%|█████████▉| 3.59GB / 3.63GB,  251MB/s  

New Data Upload               :  98%|█████████▊| 2.35GB / 2.38GB,  151MB/s  


  ...ve-03b89ee9/checkpoint_18: 100%|█████████▉| 3.63GB / 3.63GB            
Processing Files (0 / 1)      : 100%|█████████▉| 3.63GB / 3.63GB,  249MB/s  

New Data Upload               : 100%|█████████▉| 2.38GB / 2.38GB,  151MB/s  


  ...ve-03b89ee9/checkpoint_18: 100%|██████████| 3.63GB / 3.63GB            
Processing Files (1 / 1)      : 100%|██████████| 3.63GB / 3.63GB,  239MB/s  

New Data Upload               : 100%|██████████| 2.38GB / 2.38GB,  145MB/s  
Processing Files (1 / 1)      : 100%|██████████| 3.63GB / 3.63GB,  239MB/s  

New Data Upload               : 100%|██████████| 2.38GB / 2.38GB,  145MB/s  

  ...ve-03b89ee9/checkpoint_18: 100%|██████████| 3.63GB / 3.63GB            
Uploaded checkpoint_18 to seyedparsa/thoughtformer-checkpoints/checkpoints/star-datacurr-only-4gpu-bs32-lr1e-5_n1000_qwen3-0.6b-base_thoughtformer_lr1e-5_no-reset_dc_no-tc_adaptive-03b89ee9/checkpoint_18

Processing Files (0 / 0)      : |          |  0.00B /  0.00B            

New Data Upload               : |          |  0.00B /  0.00B            


  ...e-03b89ee9/train_state_18:   0%|          | 3.68MB / 4.77GB            


  ...e-03b89ee9/train_state_18:   0%|          | 3.68MB / 4.77GB            
Processing Files (0 / 1)      :   0%|          | 3.68MB / 4.77GB,   ???B/s  

New Data Upload               :   3%|▎         | 3.68MB /  134MB,   ???B/s  


  ...e-03b89ee9/train_state_18:   2%|▏         | 78.3MB / 4.77GB            
Processing Files (0 / 1)      :   2%|▏         | 78.3MB / 4.77GB,   ???B/s  

New Data Upload               :  39%|███▉      | 78.3MB /  201MB,   ???B/s  


  ...e-03b89ee9/train_state_18:   3%|▎         |  133MB / 4.77GB            
Processing Files (0 / 1)      :   3%|▎         |  133MB / 4.77GB, 12.9MB/s  

New Data Upload               :  66%|██████▋   |  133MB /  201MB, 12.9MB/s  


  ...e-03b89ee9/train_state_18:   3%|▎         |  152MB / 4.77GB            
Processing Files (0 / 1)      :   3%|▎         |  152MB / 4.77GB, 13.8MB/s  

New Data Upload               :  45%|████▌     |  152MB /  335MB, 13.8MB/s  


  ...e-03b89ee9/train_state_18:   6%|▌         |  272MB / 4.77GB            
Processing Files (0 / 1)      :   6%|▌         |  272MB / 4.77GB, 24.9MB/s  

New Data Upload               :  68%|██████▊   |  272MB /  402MB, 24.9MB/s  


  ...e-03b89ee9/train_state_18:   7%|▋         |  334MB / 4.77GB            
Processing Files (0 / 1)      :   7%|▋         |  334MB / 4.77GB, 30.1MB/s  

New Data Upload               :  83%|████████▎ |  334MB /  402MB, 30.1MB/s  


  ...e-03b89ee9/train_state_18:   7%|▋         |  334MB / 4.77GB            
Processing Files (0 / 1)      :   7%|▋         |  334MB / 4.77GB, 28.8MB/s  

New Data Upload               :  83%|████████▎ |  334MB /  402MB, 28.8MB/s  


  ...e-03b89ee9/train_state_18:   9%|▊         |  413MB / 4.77GB            
Processing Files (0 / 1)      :   9%|▊         |  413MB / 4.77GB, 35.4MB/s  

New Data Upload               :  69%|██████▊   |  413MB /  603MB, 35.4MB/s  


  ...e-03b89ee9/train_state_18:  11%|█▏        |  543MB / 4.77GB            
Processing Files (0 / 1)      :  11%|█▏        |  543MB / 4.77GB, 46.6MB/s  

New Data Upload               :  81%|████████  |  543MB /  670MB, 46.6MB/s  


  ...e-03b89ee9/train_state_18:  13%|█▎        |  638MB / 4.77GB            
Processing Files (0 / 1)      :  13%|█▎        |  638MB / 4.77GB, 54.2MB/s  

New Data Upload               :  87%|████████▋ |  638MB /  738MB, 54.2MB/s  


  ...e-03b89ee9/train_state_18:  14%|█▍        |  669MB / 4.77GB            
Processing Files (0 / 1)      :  14%|█▍        |  669MB / 4.77GB, 55.8MB/s  

New Data Upload               :  91%|█████████ |  669MB /  738MB, 55.8MB/s  


  ...e-03b89ee9/train_state_18:  14%|█▍        |  689MB / 4.77GB            
Processing Files (0 / 1)      :  14%|█▍        |  689MB / 4.77GB, 56.3MB/s  

New Data Upload               :  86%|████████▌ |  689MB /  805MB, 56.3MB/s  


  ...e-03b89ee9/train_state_18:  16%|█▌        |  770MB / 4.77GB            
Processing Files (0 / 1)      :  16%|█▌        |  770MB / 4.77GB, 62.3MB/s  

New Data Upload               :  88%|████████▊ |  770MB /  872MB, 62.3MB/s  


  ...e-03b89ee9/train_state_18:  18%|█▊        |  869MB / 4.77GB            
Processing Files (0 / 1)      :  18%|█▊        |  869MB / 4.77GB, 69.8MB/s  

New Data Upload               :  86%|████████▋ |  869MB / 1.01GB, 69.8MB/s  


  ...e-03b89ee9/train_state_18:  20%|██        |  964MB / 4.77GB            
Processing Files (0 / 1)      :  20%|██        |  964MB / 4.77GB, 76.7MB/s  

New Data Upload               :  90%|████████▉ |  964MB / 1.07GB, 76.7MB/s  


  ...e-03b89ee9/train_state_18:  22%|██▏       | 1.05GB / 4.77GB            
Processing Files (0 / 1)      :  22%|██▏       | 1.05GB / 4.77GB, 83.0MB/s  

New Data Upload               :  87%|████████▋ | 1.05GB / 1.21GB, 83.0MB/s  


  ...e-03b89ee9/train_state_18:  24%|██▍       | 1.16GB / 4.77GB            
Processing Files (0 / 1)      :  24%|██▍       | 1.16GB / 4.77GB, 90.4MB/s  

New Data Upload               :  86%|████████▋ | 1.16GB / 1.34GB, 90.4MB/s  


  ...e-03b89ee9/train_state_18:  27%|██▋       | 1.28GB / 4.77GB            
Processing Files (0 / 1)      :  27%|██▋       | 1.28GB / 4.77GB, 98.7MB/s  

New Data Upload               :  91%|█████████ | 1.28GB / 1.41GB, 98.7MB/s  


  ...e-03b89ee9/train_state_18:  29%|██▉       | 1.37GB / 4.77GB            
Processing Files (0 / 1)      :  29%|██▉       | 1.37GB / 4.77GB,  105MB/s  

New Data Upload               :  89%|████████▉ | 1.37GB / 1.54GB,  105MB/s  


  ...e-03b89ee9/train_state_18:  31%|███       | 1.47GB / 4.77GB            
Processing Files (0 / 1)      :  31%|███       | 1.47GB / 4.77GB,  112MB/s  

New Data Upload               :  92%|█████████▏| 1.47GB / 1.61GB,  112MB/s  


  ...e-03b89ee9/train_state_18:  33%|███▎      | 1.57GB / 4.77GB            
Processing Files (0 / 1)      :  33%|███▎      | 1.57GB / 4.77GB,  118MB/s  

New Data Upload               :  94%|█████████▍| 1.57GB / 1.68GB,  118MB/s  


  ...e-03b89ee9/train_state_18:  35%|███▌      | 1.67GB / 4.77GB            
Processing Files (0 / 1)      :  35%|███▌      | 1.67GB / 4.77GB,  124MB/s  

New Data Upload               :  92%|█████████▏| 1.67GB / 1.81GB,  124MB/s  


  ...e-03b89ee9/train_state_18:  37%|███▋      | 1.77GB / 4.77GB            
Processing Files (0 / 1)      :  37%|███▋      | 1.77GB / 4.77GB,  129MB/s  

New Data Upload               :  94%|█████████▍| 1.77GB / 1.88GB,  129MB/s  


  ...e-03b89ee9/train_state_18:  39%|███▉      | 1.86GB / 4.77GB            
Processing Files (0 / 1)      :  39%|███▉      | 1.86GB / 4.77GB,  134MB/s  

New Data Upload               :  92%|█████████▏| 1.86GB / 2.01GB,  134MB/s  


  ...e-03b89ee9/train_state_18:  41%|████      | 1.96GB / 4.77GB            
Processing Files (0 / 1)      :  41%|████      | 1.96GB / 4.77GB,  140MB/s  

New Data Upload               :  94%|█████████▍| 1.96GB / 2.08GB,  140MB/s  


  ...e-03b89ee9/train_state_18:  43%|████▎     | 2.05GB / 4.77GB            
Processing Files (0 / 1)      :  43%|████▎     | 2.05GB / 4.77GB,  145MB/s  

New Data Upload               :  93%|█████████▎| 2.05GB / 2.21GB,  145MB/s  


  ...e-03b89ee9/train_state_18:  45%|████▍     | 2.14GB / 4.77GB            
Processing Files (0 / 1)      :  45%|████▍     | 2.14GB / 4.77GB,  149MB/s  

New Data Upload               :  94%|█████████▎| 2.14GB / 2.28GB,  149MB/s  


  ...e-03b89ee9/train_state_18:  47%|████▋     | 2.25GB / 4.77GB            
Processing Files (0 / 1)      :  47%|████▋     | 2.25GB / 4.77GB,  155MB/s  

New Data Upload               :  96%|█████████▌| 2.25GB / 2.35GB,  155MB/s  


  ...e-03b89ee9/train_state_18:  49%|████▉     | 2.33GB / 4.77GB            
Processing Files (0 / 1)      :  49%|████▉     | 2.33GB / 4.77GB,  159MB/s  

New Data Upload               :  94%|█████████▍| 2.33GB / 2.48GB,  159MB/s  


  ...e-03b89ee9/train_state_18:  51%|█████     | 2.44GB / 4.77GB            
Processing Files (0 / 1)      :  51%|█████     | 2.44GB / 4.77GB,  164MB/s  

New Data Upload               :  96%|█████████▌| 2.44GB / 2.55GB,  164MB/s  


  ...e-03b89ee9/train_state_18:  53%|█████▎    | 2.53GB / 4.77GB            
Processing Files (0 / 1)      :  53%|█████▎    | 2.53GB / 4.77GB,  168MB/s  

New Data Upload               :  97%|█████████▋| 2.53GB / 2.62GB,  168MB/s  


  ...e-03b89ee9/train_state_18:  55%|█████▍    | 2.62GB / 4.77GB            
Processing Files (0 / 1)      :  55%|█████▍    | 2.62GB / 4.77GB,  172MB/s  

New Data Upload               :  95%|█████████▌| 2.62GB / 2.75GB,  172MB/s  


  ...e-03b89ee9/train_state_18:  57%|█████▋    | 2.71GB / 4.77GB            
Processing Files (0 / 1)      :  57%|█████▋    | 2.71GB / 4.77GB,  177MB/s  

New Data Upload               :  96%|█████████▋| 2.71GB / 2.82GB,  177MB/s  


  ...e-03b89ee9/train_state_18:  58%|█████▊    | 2.78GB / 4.77GB            
Processing Files (0 / 1)      :  58%|█████▊    | 2.78GB / 4.77GB,  179MB/s  

New Data Upload               :  97%|█████████▋| 2.78GB / 2.88GB,  179MB/s  


  ...e-03b89ee9/train_state_18:  61%|██████    | 2.89GB / 4.77GB            
Processing Files (0 / 1)      :  61%|██████    | 2.89GB / 4.77GB,  184MB/s  

New Data Upload               :  96%|█████████▌| 2.89GB / 3.02GB,  184MB/s  


  ...e-03b89ee9/train_state_18:  63%|██████▎   | 2.99GB / 4.77GB            
Processing Files (0 / 1)      :  63%|██████▎   | 2.99GB / 4.77GB,  188MB/s  

New Data Upload               :  95%|█████████▍| 2.99GB / 3.15GB,  188MB/s  


  ...e-03b89ee9/train_state_18:  65%|██████▍   | 3.08GB / 4.77GB            
Processing Files (0 / 1)      :  65%|██████▍   | 3.08GB / 4.77GB,  192MB/s  

New Data Upload               :  96%|█████████▌| 3.08GB / 3.22GB,  192MB/s  


  ...e-03b89ee9/train_state_18:  67%|██████▋   | 3.17GB / 4.77GB            
Processing Files (0 / 1)      :  67%|██████▋   | 3.17GB / 4.77GB,  196MB/s  

New Data Upload               :  97%|█████████▋| 3.17GB / 3.29GB,  196MB/s  


  ...e-03b89ee9/train_state_18:  69%|██████▊   | 3.28GB / 4.77GB            
Processing Files (0 / 1)      :  69%|██████▊   | 3.28GB / 4.77GB,  200MB/s  

New Data Upload               :  96%|█████████▌| 3.28GB / 3.42GB,  200MB/s  


  ...e-03b89ee9/train_state_18:  70%|███████   | 3.36GB / 4.77GB            
Processing Files (0 / 1)      :  70%|███████   | 3.36GB / 4.77GB,  203MB/s  

New Data Upload               :  96%|█████████▋| 3.36GB / 3.49GB,  203MB/s  


  ...e-03b89ee9/train_state_18:  72%|███████▏  | 3.42GB / 4.77GB            
Processing Files (0 / 1)      :  72%|███████▏  | 3.42GB / 4.77GB,  203MB/s  

New Data Upload               :  94%|█████████▍| 3.42GB / 3.62GB,  203MB/s  


  ...e-03b89ee9/train_state_18:  74%|███████▍  | 3.52GB / 4.77GB            
Processing Files (0 / 1)      :  74%|███████▍  | 3.52GB / 4.77GB,  208MB/s  

New Data Upload               :  95%|█████████▌| 3.52GB / 3.69GB,  208MB/s  


  ...e-03b89ee9/train_state_18:  76%|███████▌  | 3.63GB / 4.77GB            
Processing Files (0 / 1)      :  76%|███████▌  | 3.63GB / 4.77GB,  212MB/s  

New Data Upload               :  95%|█████████▍| 3.63GB / 3.82GB,  212MB/s  


  ...e-03b89ee9/train_state_18:  78%|███████▊  | 3.72GB / 4.77GB            
Processing Files (0 / 1)      :  78%|███████▊  | 3.72GB / 4.77GB,  215MB/s  

New Data Upload               :  96%|█████████▌| 3.72GB / 3.89GB,  215MB/s  


  ...e-03b89ee9/train_state_18:  80%|████████  | 3.82GB / 4.77GB            
Processing Files (0 / 1)      :  80%|████████  | 3.82GB / 4.77GB,  219MB/s  

New Data Upload               :  97%|█████████▋| 3.82GB / 3.96GB,  219MB/s  


  ...e-03b89ee9/train_state_18:  82%|████████▏ | 3.92GB / 4.77GB            
Processing Files (0 / 1)      :  82%|████████▏ | 3.92GB / 4.77GB,  222MB/s  

New Data Upload               :  96%|█████████▌| 3.92GB / 4.09GB,  222MB/s  


  ...e-03b89ee9/train_state_18:  84%|████████▍ | 4.02GB / 4.77GB            
Processing Files (0 / 1)      :  84%|████████▍ | 4.02GB / 4.77GB,  226MB/s  

New Data Upload               :  97%|█████████▋| 4.02GB / 4.16GB,  226MB/s  


  ...e-03b89ee9/train_state_18:  86%|████████▋ | 4.11GB / 4.77GB            
Processing Files (0 / 1)      :  86%|████████▋ | 4.11GB / 4.77GB,  229MB/s  

New Data Upload               :  96%|█████████▌| 4.11GB / 4.29GB,  229MB/s  


  ...e-03b89ee9/train_state_18:  89%|████████▊ | 4.22GB / 4.77GB            
Processing Files (0 / 1)      :  89%|████████▊ | 4.22GB / 4.77GB,  233MB/s  

New Data Upload               :  97%|█████████▋| 4.22GB / 4.36GB,  233MB/s  


  ...e-03b89ee9/train_state_18:  91%|█████████ | 4.33GB / 4.77GB            
Processing Files (0 / 1)      :  91%|█████████ | 4.33GB / 4.77GB,  237MB/s  

New Data Upload               :  96%|█████████▋| 4.33GB / 4.49GB,  237MB/s  


  ...e-03b89ee9/train_state_18:  93%|█████████▎| 4.45GB / 4.77GB            
Processing Files (0 / 1)      :  93%|█████████▎| 4.45GB / 4.77GB,  241MB/s  

New Data Upload               :  98%|█████████▊| 4.45GB / 4.56GB,  241MB/s  


  ...e-03b89ee9/train_state_18:  96%|█████████▌| 4.56GB / 4.77GB            
Processing Files (0 / 1)      :  96%|█████████▌| 4.56GB / 4.77GB,  245MB/s  

New Data Upload               :  97%|█████████▋| 4.56GB / 4.69GB,  245MB/s  


  ...e-03b89ee9/train_state_18:  98%|█████████▊| 4.65GB / 4.77GB            
Processing Files (0 / 1)      :  98%|█████████▊| 4.65GB / 4.77GB,  248MB/s  

New Data Upload               :  98%|█████████▊| 4.65GB / 4.77GB,  248MB/s  


  ...e-03b89ee9/train_state_18:  99%|█████████▉| 4.72GB / 4.77GB            
Processing Files (0 / 1)      :  99%|█████████▉| 4.72GB / 4.77GB,  248MB/s  

New Data Upload               :  99%|█████████▉| 4.72GB / 4.77GB,  248MB/s  


  ...e-03b89ee9/train_state_18: 100%|█████████▉| 4.77GB / 4.77GB            
Processing Files (0 / 1)      : 100%|█████████▉| 4.77GB / 4.77GB,  247MB/s  

New Data Upload               : 100%|█████████▉| 4.77GB / 4.77GB,  247MB/s  


  ...e-03b89ee9/train_state_18: 100%|█████████▉| 4.77GB / 4.77GB            
Processing Files (0 / 1)      : 100%|█████████▉| 4.77GB / 4.77GB,  242MB/s  

New Data Upload               : 100%|█████████▉| 4.77GB / 4.77GB,  242MB/s  


  ...e-03b89ee9/train_state_18: 100%|█████████▉| 4.77GB / 4.77GB            
Processing Files (0 / 1)      : 100%|█████████▉| 4.77GB / 4.77GB,  238MB/s  

New Data Upload               : 100%|█████████▉| 4.77GB / 4.77GB,  238MB/s  


  ...e-03b89ee9/train_state_18: 100%|██████████| 4.77GB / 4.77GB            
Processing Files (1 / 1)      : 100%|██████████| 4.77GB / 4.77GB,  233MB/s  

New Data Upload               : 100%|██████████| 4.77GB / 4.77GB,  233MB/s  
Processing Files (1 / 1)      : 100%|██████████| 4.77GB / 4.77GB,  233MB/s  

New Data Upload               : 100%|██████████| 4.77GB / 4.77GB,  233MB/s  

  ...e-03b89ee9/train_state_18: 100%|██████████| 4.77GB / 4.77GB            
Uploaded train_state_18 to seyedparsa/thoughtformer-checkpoints/checkpoints/star-datacurr-only-4gpu-bs32-lr1e-5_n1000_qwen3-0.6b-base_thoughtformer_lr1e-5_no-reset_dc_no-tc_adaptive-03b89ee9/train_state_18
Deleted old checkpoint: checkpoint_17
Deleted old checkpoint: train_state_17

Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]
Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]
Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]eval loss 0.010729017657771086

Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]
[CORRECT] Example 0 (diff=3)
  Expected : 'Claude'
  Extracted: 'Claude'
  Generated: '### Claude'

[CORRECT] Example 1 (diff=8)
  Expected : 'Guy'
  Extracted: 'Guy'
  Generated: '### Guy'

Test Accuracy:   3%|▎         | 1/32 [00:06<03:25,  6.62s/it]
Test Accuracy:   3%|▎         | 1/32 [00:06<03:25,  6.62s/it]
Test Accuracy:   3%|▎         | 1/32 [00:06<03:25,  6.62s/it]
[CORRECT] Example 2 (diff=10)
  Expected : 'Elvis'
  Extracted: 'Elvis'
  Generated: '### Elvis'

Test accuracy: 0.88:   3%|▎         | 1/32 [00:06<03:25,  6.62s/it]
Test accuracy: 1.0:   3%|▎         | 1/32 [00:06<03:25,  6.62s/it]
Test accuracy: 1.0:   3%|▎         | 1/32 [00:06<03:25,  6.62s/it]
[CORRECT] Example 3 (diff=8)
  Expected : 'Coral'
  Extracted: 'Coral'
  Generated: '### Coral'

[CORRECT] Example 4 (diff=1)
  Expected : 'Patricia'
  Extracted: 'Patricia'
  Generated: '### Patricia'

Test Accuracy:   3%|▎         | 1/32 [00:06<03:19,  6.43s/it]
Test accuracy: 1.0:   3%|▎         | 1/32 [00:06<03:19,  6.43s/it]
Test accuracy: 1.0:   6%|▋         | 2/32 [00:13<03:14,  6.49s/it]
Test accuracy: 0.88:   6%|▋         | 2/32 [00:13<03:14,  6.49s/it]
Test accuracy: 1.0:   6%|▋         | 2/32 [00:13<03:14,  6.49s/it]
Test accuracy: 1.0:   6%|▋         | 2/32 [00:13<03:14,  6.49s/it]
Test accuracy: 0.94:   6%|▋         | 2/32 [00:13<03:14,  6.49s/it]
Test accuracy: 1.0:   6%|▋         | 2/32 [00:13<03:14,  6.49s/it]
Test accuracy: 1.0:   6%|▋         | 2/32 [00:12<03:12,  6.41s/it]
Test accuracy: 1.0:   6%|▋         | 2/32 [00:12<03:12,  6.41s/it]
Test accuracy: 0.94:   9%|▉         | 3/32 [00:20<03:14,  6.72s/it]
Test accuracy: 1.0:   9%|▉         | 3/32 [00:20<03:14,  6.72s/it]
Test accuracy: 1.0:   9%|▉         | 3/32 [00:19<03:13,  6.68s/it]
Test accuracy: 0.96:   9%|▉         | 3/32 [00:20<03:14,  6.72s/it]
Test accuracy: 1.0:   9%|▉         | 3/32 [00:20<03:14,  6.72s/it]
Test accuracy: 1.0:   9%|▉         | 3/32 [00:19<03:13,  6.68s/it]
Test accuracy: 1.0:   9%|▉         | 3/32 [00:20<03:15,  6.73s/it]
Test accuracy: 1.0:   9%|▉         | 3/32 [00:20<03:15,  6.73s/it]
Test accuracy: 1.0:  12%|█▎        | 4/32 [00:26<03:01,  6.49s/it]
Test accuracy: 1.0:  12%|█▎        | 4/32 [00:26<03:01,  6.49s/it]
Test accuracy: 1.0:  12%|█▎        | 4/32 [00:26<03:01,  6.49s/it]
Test accuracy: 0.96:  12%|█▎        | 4/32 [00:26<03:01,  6.49s/it]
Test accuracy: 1.0:  12%|█▎        | 4/32 [00:26<03:01,  6.49s/it]
Test accuracy: 0.97:  12%|█▎        | 4/32 [00:26<03:01,  6.49s/it]
Test accuracy: 1.0:  12%|█▎        | 4/32 [00:25<03:01,  6.47s/it]
Test accuracy: 1.0:  12%|█▎        | 4/32 [00:25<03:01,  6.47s/it]
Test accuracy: 0.97:  16%|█▌        | 5/32 [00:32<02:53,  6.42s/it]
Test accuracy: 1.0:  16%|█▌        | 5/32 [00:32<02:53,  6.42s/it]
Test accuracy: 0.98:  16%|█▌        | 5/32 [00:32<02:53,  6.42s/it]
Test accuracy: 1.0:  16%|█▌        | 5/32 [00:32<02:53,  6.42s/it]
Test accuracy: 1.0:  16%|█▌        | 5/32 [00:32<02:52,  6.41s/it]
Test accuracy: 1.0:  16%|█▌        | 5/32 [00:32<02:53,  6.42s/it]
Test accuracy: 1.0:  16%|█▌        | 5/32 [00:32<02:53,  6.42s/it]
Test accuracy: 1.0:  16%|█▌        | 5/32 [00:32<02:52,  6.41s/it]
Test accuracy: 0.98:  19%|█▉        | 6/32 [00:38<02:45,  6.37s/it]
Test accuracy: 1.0:  19%|█▉        | 6/32 [00:38<02:45,  6.37s/it]
Test accuracy: 1.0:  19%|█▉        | 6/32 [00:38<02:45,  6.37s/it]
Test accuracy: 1.0:  19%|█▉        | 6/32 [00:38<02:45,  6.37s/it]
Test accuracy: 0.98:  19%|█▉        | 6/32 [00:38<02:45,  6.37s/it]
Test accuracy: 1.0:  19%|█▉        | 6/32 [00:38<02:45,  6.37s/it]
Test accuracy: 1.0:  19%|█▉        | 6/32 [00:38<02:45,  6.36s/it]
Test accuracy: 1.0:  19%|█▉        | 6/32 [00:38<02:45,  6.36s/it]
Test accuracy: 1.0:  22%|██▏       | 7/32 [00:45<02:39,  6.36s/it]
Test accuracy: 0.98:  22%|██▏       | 7/32 [00:45<02:39,  6.36s/it]
Test accuracy: 1.0:  22%|██▏       | 7/32 [00:45<02:39,  6.36s/it]
Test accuracy: 1.0:  22%|██▏       | 7/32 [00:45<02:39,  6.36s/it]
Test accuracy: 0.98:  22%|██▏       | 7/32 [00:45<02:39,  6.36s/it]
Test accuracy: 1.0:  22%|██▏       | 7/32 [00:45<02:39,  6.36s/it]
Test accuracy: 1.0:  22%|██▏       | 7/32 [00:44<02:38,  6.35s/it]
Test accuracy: 1.0:  22%|██▏       | 7/32 [00:44<02:38,  6.35s/it]
Test accuracy: 1.0:  25%|██▌       | 8/32 [00:51<02:32,  6.34s/it]
Test accuracy: 1.0:  25%|██▌       | 8/32 [00:51<02:32,  6.34s/it]
Test accuracy: 0.98:  25%|██▌       | 8/32 [00:51<02:32,  6.34s/it]
Test accuracy: 0.98:  25%|██▌       | 8/32 [00:51<02:32,  6.34s/it]
Test accuracy: 1.0:  25%|██▌       | 8/32 [00:51<02:32,  6.34s/it]
Test accuracy: 0.98:  25%|██▌       | 8/32 [00:51<02:32,  6.34s/it]
Test accuracy: 1.0:  25%|██▌       | 8/32 [00:51<02:32,  6.34s/it]
Test accuracy: 1.0:  25%|██▌       | 8/32 [00:51<02:32,  6.34s/it]
Test accuracy: 1.0:  28%|██▊       | 9/32 [00:58<02:28,  6.47s/it]
Test accuracy: 0.98:  28%|██▊       | 9/32 [00:58<02:28,  6.47s/it]
Test accuracy: 1.0:  28%|██▊       | 9/32 [00:58<02:28,  6.47s/it]
Test accuracy: 0.99:  28%|██▊       | 9/32 [00:58<02:28,  6.47s/it]
Test accuracy: 1.0:  28%|██▊       | 9/32 [00:57<02:28,  6.47s/it]
Test accuracy: 1.0:  28%|██▊       | 9/32 [00:57<02:28,  6.47s/it]
Test accuracy: 0.98:  28%|██▊       | 9/32 [00:58<02:28,  6.48s/it]
Test accuracy: 0.99:  28%|██▊       | 9/32 [00:58<02:28,  6.48s/it]
Test accuracy: 0.99:  31%|███▏      | 10/32 [01:04<02:21,  6.43s/it]
Test accuracy: 0.99:  31%|███▏      | 10/32 [01:04<02:21,  6.43s/it]
Test accuracy: 1.0:  31%|███▏      | 10/32 [01:04<02:21,  6.43s/it]
Test accuracy: 0.99:  31%|███▏      | 10/32 [01:04<02:21,  6.43s/it]
Test accuracy: 1.0:  31%|███▏      | 10/32 [01:04<02:21,  6.43s/it]
Test accuracy: 0.99:  31%|███▏      | 10/32 [01:04<02:21,  6.43s/it]
Test accuracy: 1.0:  31%|███▏      | 10/32 [01:04<02:21,  6.43s/it]
Test accuracy: 1.0:  31%|███▏      | 10/32 [01:04<02:21,  6.43s/it]
Test accuracy: 0.99:  34%|███▍      | 11/32 [01:10<02:14,  6.40s/it]
Test accuracy: 1.0:  34%|███▍      | 11/32 [01:10<02:14,  6.40s/it]
Test accuracy: 0.99:  34%|███▍      | 11/32 [01:10<02:14,  6.40s/it]
Test accuracy: 1.0:  34%|███▍      | 11/32 [01:10<02:14,  6.40s/it]
Test accuracy: 0.99:  34%|███▍      | 11/32 [01:10<02:14,  6.40s/it]
Test accuracy: 1.0:  34%|███▍      | 11/32 [01:10<02:14,  6.39s/it]
Test accuracy: 0.99:  34%|███▍      | 11/32 [01:10<02:14,  6.40s/it]
Test accuracy: 1.0:  34%|███▍      | 11/32 [01:10<02:14,  6.39s/it]
Test accuracy: 0.99:  38%|███▊      | 12/32 [01:16<02:06,  6.31s/it]
Test accuracy: 0.99:  38%|███▊      | 12/32 [01:16<02:06,  6.31s/it]
Test accuracy: 1.0:  38%|███▊      | 12/32 [01:16<02:06,  6.31s/it]
Test accuracy: 0.99:  38%|███▊      | 12/32 [01:16<02:06,  6.31s/it]
Test accuracy: 1.0:  38%|███▊      | 12/32 [01:16<02:06,  6.31s/it]
Test accuracy: 0.99:  38%|███▊      | 12/32 [01:16<02:06,  6.31s/it]
Test accuracy: 1.0:  38%|███▊      | 12/32 [01:16<02:06,  6.31s/it]
Test accuracy: 1.0:  38%|███▊      | 12/32 [01:16<02:06,  6.31s/it]
Test accuracy: 1.0:  41%|████      | 13/32 [01:23<02:00,  6.32s/it]
Test accuracy: 0.99:  41%|████      | 13/32 [01:23<02:00,  6.32s/it]
Test accuracy: 0.99:  41%|████      | 13/32 [01:23<02:00,  6.32s/it]
Test accuracy: 1.0:  41%|████      | 13/32 [01:23<02:00,  6.32s/it]
Test accuracy: 0.99:  41%|████      | 13/32 [01:23<02:00,  6.32s/it]
Test accuracy: 0.99:  41%|████      | 13/32 [01:23<02:00,  6.32s/it]
Test accuracy: 1.0:  41%|████      | 13/32 [01:23<02:00,  6.32s/it]
Test accuracy: 1.0:  41%|████      | 13/32 [01:23<02:00,  6.32s/it]
Test accuracy: 1.0:  44%|████▍     | 14/32 [01:29<01:53,  6.32s/it]
Test accuracy: 1.0:  44%|████▍     | 14/32 [01:29<01:53,  6.32s/it]
Test accuracy: 0.99:  44%|████▍     | 14/32 [01:29<01:53,  6.32s/it]
Test accuracy: 1.0:  44%|████▍     | 14/32 [01:29<01:53,  6.32s/it]
Test accuracy: 0.99:  44%|████▍     | 14/32 [01:29<01:53,  6.32s/it]
Test accuracy: 0.99:  44%|████▍     | 14/32 [01:29<01:53,  6.32s/it]
Test accuracy: 1.0:  44%|████▍     | 14/32 [01:29<01:53,  6.32s/it]
Test accuracy: 0.99:  44%|████▍     | 14/32 [01:29<01:53,  6.32s/it]
Test accuracy: 0.99:  47%|████▋     | 15/32 [01:35<01:46,  6.26s/it]
Test accuracy: 1.0:  47%|████▋     | 15/32 [01:35<01:46,  6.26s/it]
Test accuracy: 1.0:  47%|████▋     | 15/32 [01:35<01:46,  6.26s/it]
Test accuracy: 0.99:  47%|████▋     | 15/32 [01:35<01:46,  6.26s/it]
Test accuracy: 0.98:  47%|████▋     | 15/32 [01:35<01:46,  6.26s/it]
Test accuracy: 1.0:  47%|████▋     | 15/32 [01:35<01:46,  6.26s/it]
Test accuracy: 0.99:  47%|████▋     | 15/32 [01:35<01:46,  6.26s/it]
Test accuracy: 1.0:  47%|████▋     | 15/32 [01:35<01:46,  6.26s/it]
Test accuracy: 0.98:  50%|█████     | 16/32 [01:42<01:40,  6.28s/it]
Test accuracy: 1.0:  50%|█████     | 16/32 [01:42<01:40,  6.28s/it]
Test accuracy: 1.0:  50%|█████     | 16/32 [01:41<01:40,  6.28s/it]
Test accuracy: 0.99:  50%|█████     | 16/32 [01:42<01:40,  6.28s/it]
Test accuracy: 0.98:  50%|█████     | 16/32 [01:42<01:40,  6.28s/it]
Test accuracy: 1.0:  50%|█████     | 16/32 [01:42<01:40,  6.28s/it]
Test accuracy: 0.99:  50%|█████     | 16/32 [01:42<01:40,  6.28s/it]
Test accuracy: 1.0:  50%|█████     | 16/32 [01:41<01:40,  6.28s/it]
Test accuracy: 1.0:  53%|█████▎    | 17/32 [01:48<01:34,  6.29s/it]
Test accuracy: 0.98:  53%|█████▎    | 17/32 [01:48<01:34,  6.29s/it]
Test accuracy: 0.99:  53%|█████▎    | 17/32 [01:48<01:34,  6.29s/it]
Test accuracy: 1.0:  53%|█████▎    | 17/32 [01:48<01:34,  6.29s/it]
Test accuracy: 0.99:  53%|█████▎    | 17/32 [01:48<01:34,  6.29s/it]
Test accuracy: 1.0:  53%|█████▎    | 17/32 [01:48<01:34,  6.29s/it]
Test accuracy: 0.99:  53%|█████▎    | 17/32 [01:48<01:34,  6.29s/it]
Test accuracy: 1.0:  53%|█████▎    | 17/32 [01:48<01:34,  6.29s/it]
Test accuracy: 0.99:  56%|█████▋    | 18/32 [01:54<01:28,  6.31s/it]
Test accuracy: 0.99:  56%|█████▋    | 18/32 [01:54<01:28,  6.31s/it]
Test accuracy: 0.99:  56%|█████▋    | 18/32 [01:54<01:28,  6.31s/it]
Test accuracy: 0.99:  56%|█████▋    | 18/32 [01:54<01:28,  6.31s/it]
Test accuracy: 1.0:  56%|█████▋    | 18/32 [01:54<01:28,  6.31s/it]
Test accuracy: 1.0:  56%|█████▋    | 18/32 [01:54<01:28,  6.31s/it]
Test accuracy: 1.0:  56%|█████▋    | 18/32 [01:54<01:28,  6.31s/it]
Test accuracy: 1.0:  56%|█████▋    | 18/32 [01:54<01:28,  6.31s/it]
Test accuracy: 1.0:  59%|█████▉    | 19/32 [02:00<01:21,  6.26s/it]
Test accuracy: 0.99:  59%|█████▉    | 19/32 [02:00<01:21,  6.26s/it]
Test accuracy: 0.99:  59%|█████▉    | 19/32 [02:00<01:21,  6.26s/it]
Test accuracy: 1.0:  59%|█████▉    | 19/32 [02:00<01:21,  6.26s/it]
Test accuracy: 0.99:  59%|█████▉    | 19/32 [02:00<01:21,  6.26s/it]
Test accuracy: 1.0:  59%|█████▉    | 19/32 [02:00<01:21,  6.26s/it]
Test accuracy: 0.99:  59%|█████▉    | 19/32 [02:00<01:21,  6.26s/it]
Test accuracy: 1.0:  59%|█████▉    | 19/32 [02:00<01:21,  6.26s/it]
Test accuracy: 0.99:  62%|██████▎   | 20/32 [02:07<01:15,  6.28s/it]
Test accuracy: 0.98:  62%|██████▎   | 20/32 [02:07<01:15,  6.28s/it]
Test accuracy: 0.99:  62%|██████▎   | 20/32 [02:07<01:15,  6.28s/it]
Test accuracy: 1.0:  62%|██████▎   | 20/32 [02:06<01:15,  6.28s/it]
Test accuracy: 0.99:  62%|██████▎   | 20/32 [02:07<01:15,  6.28s/it]
Test accuracy: 1.0:  62%|██████▎   | 20/32 [02:06<01:15,  6.28s/it]
Test accuracy: 1.0:  62%|██████▎   | 20/32 [02:07<01:15,  6.28s/it]
Test accuracy: 1.0:  62%|██████▎   | 20/32 [02:07<01:15,  6.28s/it]
Test accuracy: 0.98:  66%|██████▌   | 21/32 [02:14<01:11,  6.51s/it]
Test accuracy: 0.99:  66%|██████▌   | 21/32 [02:14<01:11,  6.51s/it]
Test accuracy: 0.99:  66%|██████▌   | 21/32 [02:14<01:11,  6.51s/it]
Test accuracy: 0.98:  66%|██████▌   | 21/32 [02:14<01:11,  6.51s/it]
Test accuracy: 1.0:  66%|██████▌   | 21/32 [02:14<01:11,  6.51s/it]
Test accuracy: 1.0:  66%|██████▌   | 21/32 [02:14<01:11,  6.51s/it]
Test accuracy: 1.0:  66%|██████▌   | 21/32 [02:14<01:11,  6.51s/it]
Test accuracy: 1.0:  66%|██████▌   | 21/32 [02:14<01:11,  6.51s/it]
Test accuracy: 1.0:  69%|██████▉   | 22/32 [02:20<01:04,  6.40s/it]
Test accuracy: 0.98:  69%|██████▉   | 22/32 [02:20<01:04,  6.40s/it]
Test accuracy: 0.99:  69%|██████▉   | 22/32 [02:20<01:04,  6.40s/it]
Test accuracy: 1.0:  69%|██████▉   | 22/32 [02:20<01:04,  6.40s/it]
Test accuracy: 0.98:  69%|██████▉   | 22/32 [02:20<01:04,  6.40s/it]
Test accuracy: 0.99:  69%|██████▉   | 22/32 [02:20<01:04,  6.40s/it]
Test accuracy: 1.0:  69%|██████▉   | 22/32 [02:20<01:04,  6.40s/it]
Test accuracy: 1.0:  69%|██████▉   | 22/32 [02:20<01:04,  6.40s/it]
Test accuracy: 0.98:  72%|███████▏  | 23/32 [02:26<00:57,  6.38s/it]
Test accuracy: 0.99:  72%|███████▏  | 23/32 [02:26<00:57,  6.38s/it]
Test accuracy: 0.99:  72%|███████▏  | 23/32 [02:26<00:57,  6.38s/it]
Test accuracy: 0.98:  72%|███████▏  | 23/32 [02:26<00:57,  6.38s/it]
Test accuracy: 1.0:  72%|███████▏  | 23/32 [02:26<00:57,  6.38s/it]
Test accuracy: 1.0:  72%|███████▏  | 23/32 [02:26<00:57,  6.38s/it]
Test accuracy: 1.0:  72%|███████▏  | 23/32 [02:26<00:57,  6.38s/it]
Test accuracy: 1.0:  72%|███████▏  | 23/32 [02:26<00:57,  6.38s/it]
Test accuracy: 1.0:  75%|███████▌  | 24/32 [02:33<00:52,  6.58s/it]
Test accuracy: 1.0:  75%|███████▌  | 24/32 [02:33<00:52,  6.58s/it]
Test accuracy: 0.98:  75%|███████▌  | 24/32 [02:33<00:52,  6.58s/it]
Test accuracy: 0.99:  75%|███████▌  | 24/32 [02:33<00:52,  6.58s/it]
Test accuracy: 0.98:  75%|███████▌  | 24/32 [02:33<00:52,  6.58s/it]
Test accuracy: 0.99:  75%|███████▌  | 24/32 [02:33<00:52,  6.58s/it]
Test accuracy: 1.0:  75%|███████▌  | 24/32 [02:33<00:52,  6.59s/it]
Test accuracy: 1.0:  75%|███████▌  | 24/32 [02:33<00:52,  6.59s/it]
Test accuracy: 0.98:  78%|███████▊  | 25/32 [02:40<00:45,  6.51s/it]
Test accuracy: 1.0:  78%|███████▊  | 25/32 [02:40<00:45,  6.51s/it]
Test accuracy: 0.99:  78%|███████▊  | 25/32 [02:40<00:45,  6.51s/it]
Test accuracy: 1.0:  78%|███████▊  | 25/32 [02:40<00:45,  6.51s/it]
Test accuracy: 0.99:  78%|███████▊  | 25/32 [02:40<00:45,  6.51s/it]
Test accuracy: 0.99:  78%|███████▊  | 25/32 [02:40<00:45,  6.51s/it]
Test accuracy: 1.0:  78%|███████▊  | 25/32 [02:39<00:45,  6.51s/it]
Test accuracy: 1.0:  78%|███████▊  | 25/32 [02:39<00:45,  6.51s/it]
Test accuracy: 0.99:  81%|████████▏ | 26/32 [02:46<00:38,  6.46s/it]
Test accuracy: 0.99:  81%|████████▏ | 26/32 [02:46<00:38,  6.46s/it]
Test accuracy: 0.99:  81%|████████▏ | 26/32 [02:46<00:38,  6.46s/it]
Test accuracy: 1.0:  81%|████████▏ | 26/32 [02:46<00:38,  6.46s/it]
Test accuracy: 0.99:  81%|████████▏ | 26/32 [02:46<00:38,  6.46s/it]
Test accuracy: 1.0:  81%|████████▏ | 26/32 [02:46<00:38,  6.46s/it]
Test accuracy: 1.0:  81%|████████▏ | 26/32 [02:46<00:38,  6.46s/it]
Test accuracy: 1.0:  81%|████████▏ | 26/32 [02:46<00:38,  6.46s/it]
Test accuracy: 0.99:  84%|████████▍ | 27/32 [02:52<00:31,  6.36s/it]
Test accuracy: 1.0:  84%|████████▍ | 27/32 [02:52<00:31,  6.36s/it]
Test accuracy: 0.99:  84%|████████▍ | 27/32 [02:52<00:31,  6.36s/it]
Test accuracy: 1.0:  84%|████████▍ | 27/32 [02:52<00:31,  6.36s/it]
Test accuracy: 0.99:  84%|████████▍ | 27/32 [02:52<00:31,  6.36s/it]
Test accuracy: 1.0:  84%|████████▍ | 27/32 [02:52<00:31,  6.36s/it]
Test accuracy: 0.99:  84%|████████▍ | 27/32 [02:52<00:31,  6.36s/it]
Test accuracy: 1.0:  84%|████████▍ | 27/32 [02:52<00:31,  6.36s/it]
Test accuracy: 0.99:  88%|████████▊ | 28/32 [02:59<00:26,  6.56s/it]
Test accuracy: 0.99:  88%|████████▊ | 28/32 [02:59<00:26,  6.56s/it]
Test accuracy: 1.0:  88%|████████▊ | 28/32 [02:59<00:26,  6.56s/it]
Test accuracy: 0.99:  88%|████████▊ | 28/32 [02:59<00:26,  6.56s/it]
Test accuracy: 0.98:  88%|████████▊ | 28/32 [02:59<00:26,  6.56s/it]
Test accuracy: 1.0:  88%|████████▊ | 28/32 [02:59<00:26,  6.56s/it]
Test accuracy: 1.0:  88%|████████▊ | 28/32 [02:59<00:26,  6.56s/it]
Test accuracy: 1.0:  88%|████████▊ | 28/32 [02:59<00:26,  6.56s/it]
Test accuracy: 0.99:  91%|█████████ | 29/32 [03:05<00:19,  6.50s/it]
Test accuracy: 0.98:  91%|█████████ | 29/32 [03:05<00:19,  6.50s/it]
Test accuracy: 0.99:  91%|█████████ | 29/32 [03:05<00:19,  6.50s/it]
Test accuracy: 1.0:  91%|█████████ | 29/32 [03:05<00:19,  6.50s/it]
Test accuracy: 0.98:  91%|█████████ | 29/32 [03:05<00:19,  6.50s/it]
Test accuracy: 1.0:  91%|█████████ | 29/32 [03:05<00:19,  6.50s/it]
Test accuracy: 1.0:  91%|█████████ | 29/32 [03:05<00:19,  6.50s/it]
Test accuracy: 1.0:  91%|█████████ | 29/32 [03:05<00:19,  6.50s/it]
Test accuracy: 0.98:  94%|█████████▍| 30/32 [03:12<00:12,  6.39s/it]
Test accuracy: 1.0:  94%|█████████▍| 30/32 [03:12<00:12,  6.39s/it]
Test accuracy: 0.99:  94%|█████████▍| 30/32 [03:12<00:12,  6.39s/it]
Test accuracy: 1.0:  94%|█████████▍| 30/32 [03:11<00:12,  6.39s/it]
Test accuracy: 1.0:  94%|█████████▍| 30/32 [03:12<00:12,  6.39s/it]
Test accuracy: 0.98:  94%|█████████▍| 30/32 [03:12<00:12,  6.39s/it]
Test accuracy: 0.99:  94%|█████████▍| 30/32 [03:12<00:12,  6.39s/it]
Test accuracy: 1.0:  94%|█████████▍| 30/32 [03:11<00:12,  6.39s/it]
Test accuracy: 0.98:  97%|█████████▋| 31/32 [03:19<00:06,  6.58s/it]
Test accuracy: 0.99:  97%|█████████▋| 31/32 [03:19<00:06,  6.58s/it]
Test accuracy: 1.0:  97%|█████████▋| 31/32 [03:18<00:06,  6.58s/it]
Test accuracy: 0.99:  97%|█████████▋| 31/32 [03:19<00:06,  6.58s/it]
Test accuracy: 0.98:  97%|█████████▋| 31/32 [03:19<00:06,  6.58s/it]
Test accuracy: 1.0:  97%|█████████▋| 31/32 [03:18<00:06,  6.58s/it]
Test accuracy: 1.0:  97%|█████████▋| 31/32 [03:19<00:06,  6.58s/it]
Test accuracy: 1.0:  97%|█████████▋| 31/32 [03:19<00:06,  6.58s/it]
Test accuracy: 1.0: 100%|██████████| 32/32 [03:21<00:00,  5.28s/it]
Test accuracy: 1.0: 100%|██████████| 32/32 [03:21<00:00,  5.28s/it]
Test accuracy: 0.98: 100%|██████████| 32/32 [03:21<00:00,  5.28s/it]
Test accuracy: 1.0: 100%|██████████| 32/32 [03:21<00:00,  5.28s/it]
Test accuracy: 0.99: 100%|██████████| 32/32 [03:21<00:00,  5.28s/it]
Test accuracy: 0.98: 100%|██████████| 32/32 [03:21<00:00,  5.28s/it]
Test accuracy: 1.0: 100%|██████████| 32/32 [03:21<00:00,  5.28s/it]
Test accuracy: 0.99: 100%|██████████| 32/32 [03:21<00:00,  5.28s/it]
Test accuracy: 1.0: 100%|██████████| 32/32 [03:21<00:00,  6.30s/it]
Device 2: Cor=250, CoT=0, Total=250

Test accuracy: 0.98: 100%|██████████| 32/32 [03:21<00:00,  6.30s/it]
Device 3: Cor=246, CoT=0, Total=250

Test accuracy: 0.99: 100%|██████████| 32/32 [03:21<00:00,  6.30s/it]
Device 1: Cor=247, CoT=0, Total=250

Test accuracy: 1.0: 100%|██████████| 32/32 [03:21<00:00,  6.30s/it]
Device 0: Cor=249, CoT=0, Total=250

============================================================
EVALUATION RESULTS (Epoch 17)
============================================================
Config: data_curriculum=ON, thought_curriculum=OFF
thought_stage=10, data_stage=10, n_latent_tokens=10
Training on difficulties: 0-10

Overall accuracy: 992 / 1000 = 0.9920

Per-difficulty accuracy:
  diff=1: 100/100 = 1.0000 (trained)
  diff=2: 99/100 = 0.9900 (trained)
  diff=3: 99/100 = 0.9900 (trained)
  diff=4: 99/100 = 0.9900 (trained)
  diff=5: 99/100 = 0.9900 (trained)
  diff=6: 99/100 = 0.9900 (trained)
  diff=7: 100/100 = 1.0000 (trained)
  diff=8: 100/100 = 1.0000 (trained)
  diff=9: 99/100 = 0.9900 (trained)
  diff=10: 98/100 = 0.9800 (trained)

Trained (diff 0-10): 992/1000 = 0.9920
============================================================


Processing Files (0 / 0)      : |          |  0.00B /  0.00B            

New Data Upload               : |          |  0.00B /  0.00B            


  ...ve-03b89ee9/checkpoint_18:  10%|█         |  368MB / 3.63GB            


  ...ve-03b89ee9/checkpoint_18:  10%|█         |  368MB / 3.63GB            
Processing Files (0 / 1)      :  10%|█         |  368MB / 3.63GB,   ???B/s  


  ...ve-03b89ee9/checkpoint_18:  20%|██        |  736MB / 3.63GB            
Processing Files (0 / 1)      :  20%|██        |  736MB / 3.63GB,   ???B/s  


  ...ve-03b89ee9/checkpoint_18:  31%|███       | 1.12GB / 3.63GB            
Processing Files (0 / 1)      :  31%|███       | 1.12GB / 3.63GB,  108MB/s  


  ...ve-03b89ee9/checkpoint_18:  41%|████▏     | 1.50GB / 3.63GB            
Processing Files (0 / 1)      :  41%|████▏     | 1.50GB / 3.63GB,  142MB/s  


  ...ve-03b89ee9/checkpoint_18:  52%|█████▏    | 1.88GB / 3.63GB            
Processing Files (0 / 1)      :  52%|█████▏    | 1.88GB / 3.63GB,  175MB/s  


  ...ve-03b89ee9/checkpoint_18:  62%|██████▏   | 2.26GB / 3.63GB            
Processing Files (0 / 1)      :  62%|██████▏   | 2.26GB / 3.63GB,  207MB/s  


  ...ve-03b89ee9/checkpoint_18:  73%|███████▎  | 2.64GB / 3.63GB            
Processing Files (0 / 1)      :  73%|███████▎  | 2.64GB / 3.63GB,  238MB/s  


  ...ve-03b89ee9/checkpoint_18:  83%|████████▎ | 3.02GB / 3.63GB            
Processing Files (0 / 1)      :  83%|████████▎ | 3.02GB / 3.63GB,  268MB/s  


  ...ve-03b89ee9/checkpoint_18:  93%|█████████▎| 3.38GB / 3.63GB            
Processing Files (0 / 1)      :  93%|█████████▎| 3.38GB / 3.63GB,  296MB/s  


  ...ve-03b89ee9/checkpoint_18: 100%|██████████| 3.63GB / 3.63GB            
Processing Files (1 / 1)      : 100%|██████████| 3.63GB / 3.63GB,  312MB/s  
Processing Files (1 / 1)      : 100%|██████████| 3.63GB / 3.63GB,  312MB/s  

New Data Upload               : |          |  0.00B /  0.00B,  0.00B/s  

  ...ve-03b89ee9/checkpoint_18: 100%|██████████| 3.63GB / 3.63GB            
No files have been modified since last commit. Skipping to prevent empty commit.
Uploaded checkpoint_18 to seyedparsa/thoughtformer-checkpoints/checkpoints/star-datacurr-only-4gpu-bs32-lr1e-5_n1000_qwen3-0.6b-base_thoughtformer_lr1e-5_no-reset_dc_no-tc_adaptive-03b89ee9/checkpoint_18

============================================================
EPOCH 18/100 (thought_stage=10, data_stage=10)
============================================================
  thought_stage=10, c_thought=1, max_difficulty=10

Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):  16%|█▌        | 158/1000 [00:00<00:00, 942.05 examples/s]
Map (num_proc=4):  16%|█▌        | 156/1000 [00:00<00:00, 903.70 examples/s]
Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):  16%|█▌        | 156/1000 [00:00<00:00, 857.52 examples/s]
Map (num_proc=4):  81%|████████▏ | 813/1000 [00:00<00:00, 3288.21 examples/s]
Map (num_proc=4):  81%|████████  | 811/1000 [00:00<00:00, 3302.11 examples/s]
Map (num_proc=4):  81%|████████  | 811/1000 [00:00<00:00, 3141.58 examples/s]
Map (num_proc=4):  16%|█▌        | 157/1000 [00:00<00:01, 727.01 examples/s]
Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 1941.75 examples/s]

Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 1930.61 examples/s]

Map (num_proc=4):  81%|████████▏ | 814/1000 [00:00<00:00, 2726.48 examples/s]
Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 1821.16 examples/s]

Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2218.38 examples/s]

Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:04, 1802.73 examples/s]
Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:05, 1612.65 examples/s]
Filter (num_proc=4):  20%|██        | 2000/10000 [00:00<00:02, 3376.62 examples/s]
Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:05, 1544.62 examples/s]
Filter (num_proc=4):  20%|██        | 2000/10000 [00:00<00:02, 2899.00 examples/s]
Filter (num_proc=4):  30%|███       | 3000/10000 [00:00<00:01, 4782.50 examples/s]
Filter (num_proc=4):  20%|██        | 2000/10000 [00:00<00:02, 2741.28 examples/s]
Filter (num_proc=4):  50%|█████     | 5000/10000 [00:00<00:00, 8097.98 examples/s]
Filter (num_proc=4):  40%|████      | 4000/10000 [00:00<00:01, 5721.96 examples/s]
Filter (num_proc=4):  10%|█         | 1000/10000 [00:00<00:05, 1505.49 examples/s]
Filter (num_proc=4):  75%|███████▌  | 7500/10000 [00:01<00:00, 11436.95 examples/s]
Filter (num_proc=4):  40%|████      | 4000/10000 [00:00<00:01, 5160.53 examples/s]
Filter (num_proc=4):  65%|██████▌   | 6500/10000 [00:01<00:00, 8764.71 examples/s]
Filter (num_proc=4):  90%|█████████ | 9000/10000 [00:01<00:00, 11856.05 examples/s]
Filter (num_proc=4):  20%|██        | 2000/10000 [00:00<00:02, 2691.41 examples/s]
Filter (num_proc=4):  65%|██████▌   | 6500/10000 [00:01<00:00, 7852.19 examples/s]
Filter (num_proc=4):  85%|████████▌ | 8500/10000 [00:01<00:00, 9178.32 examples/s]
Filter (num_proc=4):  40%|████      | 4000/10000 [00:00<00:01, 5325.01 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 9445.89 examples/s]
Filter (num_proc=4):  85%|████████▌ | 8500/10000 [00:01<00:00, 8735.41 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 7115.35 examples/s]

Filter (num_proc=4):  65%|██████▌   | 6500/10000 [00:01<00:00, 8220.47 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 6374.07 examples/s]

Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 8742.55 examples/s]
Filter (num_proc=4):  85%|████████▌ | 8500/10000 [00:01<00:00, 8804.65 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 5938.42 examples/s]

Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 9028.82 examples/s]
Filter (num_proc=4): 100%|██████████| 10000/10000 [00:01<00:00, 5877.43 examples/s]

Map (num_proc=4):   0%|          | 0/10000 [00:00<?, ? examples/s]
Map (num_proc=4):   1%|▏         | 128/10000 [00:01<01:38, 100.38 examples/s]
Map (num_proc=4):   5%|▌         | 545/10000 [00:01<00:19, 491.78 examples/s]
Map (num_proc=4):   8%|▊         | 822/10000 [00:01<00:11, 769.15 examples/s]
Map (num_proc=4):  12%|█▏        | 1163/10000 [00:01<00:11, 791.88 examples/s]
Map (num_proc=4):  20%|██        | 2028/10000 [00:02<00:04, 1700.92 examples/s]
Map (num_proc=4):  26%|██▌       | 2566/10000 [00:02<00:03, 2239.75 examples/s]
Map (num_proc=4):  30%|███       | 3000/10000 [00:02<00:03, 1763.93 examples/s]
Map (num_proc=4):  42%|████▏     | 4200/10000 [00:02<00:01, 3066.99 examples/s]
Map (num_proc=4):  48%|████▊     | 4790/10000 [00:02<00:01, 3528.36 examples/s]
Map (num_proc=4):  53%|█████▎    | 5339/10000 [00:02<00:01, 3841.33 examples/s]
Map (num_proc=4):  59%|█████▉    | 5902/10000 [00:03<00:01, 2220.94 examples/s]
Map (num_proc=4):  77%|███████▋  | 7685/10000 [00:03<00:00, 4127.44 examples/s]
Map (num_proc=4):  84%|████████▎ | 8363/10000 [00:04<00:00, 2721.39 examples/s]
Map (num_proc=4):  94%|█████████▎| 9364/10000 [00:04<00:00, 3467.05 examples/s]
Map (num_proc=4): 100%|██████████| 10000/10000 [00:04<00:00, 2218.96 examples/s]
Map (num_proc=4): 100%|██████████| 10000/10000 [00:05<00:00, 1778.83 examples/s]

Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 909.10 examples/s]
Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 1280.59 examples/s]
Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 1351.64 examples/s]
Filter (num_proc=4):  25%|██▌       | 250/1000 [00:00<00:00, 1316.61 examples/s]
Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 1697.76 examples/s]

Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2641.41 examples/s]

Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2439.04 examples/s]

Filter (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 2562.58 examples/s]

Map (num_proc=4):   0%|          | 0/1000 [00:00<?, ? examples/s]
Map (num_proc=4):  14%|█▍        | 138/1000 [00:00<00:01, 442.86 examples/s]
Map (num_proc=4):  39%|███▊      | 387/1000 [00:00<00:00, 1040.77 examples/s]
Map (num_proc=4):  78%|███████▊  | 775/1000 [00:00<00:00, 1810.69 examples/s]
Map (num_proc=4): 100%|██████████| 1000/1000 [00:00<00:00, 1036.69 examples/s]

Training Epoch: 19:   0%|          | 0/312 [00:00<?, ?it/s]
Training Epoch: 19:   0%|          | 0/312 [00:00<?, ?it/s]
Training Epoch: 19:   0%|          | 0/312 [00:00<?, ?it/s]
Training Epoch: 19:   0%|          | 0/312 [00:00<?, ?it/s]logging training data
rank 1 | gpu 1 | alloc=4.20 GB | reserved=55.20 GB | max=41.98 GB
rank 3 | gpu 3 | alloc=4.20 GB | reserved=47.73 GB | max=41.99 GB
rank 2 | gpu 2 | alloc=4.20 GB | reserved=47.74 GB | max=41.98 GB
rank 0 | gpu 0 | alloc=4.20 GB | reserved=47.74 GB | max=41.99 GB

Training Epoch: 19/100, batch 0/625 completed (loss: 0.0122:   0%|          | 0/312 [00:03<?, ?it/s]
Training Epoch: 19/100, batch 0/625 completed (loss: 0.0002:   0%|          | 0/312 [00:03<?, ?it/s]
Training Epoch: 19/100, batch 0/625 completed (loss: 0.0008:   0%|          | 0/312 [00:03<?, ?it/s]
Training Epoch: 19/100, batch 0/625 completed (loss: 0.0003:   0%|          | 0/312 [00:03<?, ?it/s]
Training Epoch: 19/100, batch 0/625 completed (loss: 0.0122:   0%|          | 1/312 [00:06<32:07,  6.20s/it]
Training Epoch: 19/100, batch 0/625 completed (loss: 0.0003:   0%|          | 1/312 [00:06<32:29,  6.27s/it]
Training Epoch: 19/100, batch 0/625 completed (loss: 0.0002:   0%|          | 1/312 [00:06<32:16,  6.23s/it]
Training Epoch: 19/100, batch 0/625 completed (loss: 0.0008:   0%|          | 1/312 [00:06<32:17,  6.23s/it]
Training Epoch: 19/100, batch 1/625 completed (loss: 0.0007:   0%|          | 1/312 [00:06<32:16,  6.23s/it]
Training Epoch: 19/100, batch 1/625 completed (loss: 0.001:   0%|          | 1/312 [00:06<32:17,  6.23s/it] 
Training Epoch: 19/100, batch 1/625 completed (loss: 0.0003:   0%|          | 1/312 [00:06<32:07,  6.20s/it]
Training Epoch: 19/100, batch 1/625 completed (loss: 0.0051:   0%|          | 1/312 [00:06<32:29,  6.27s/it]
Training Epoch: 19/100, batch 2/625 completed (loss: 0.0004:   0%|          | 1/312 [00:09<32:16,  6.23s/it]
Training Epoch: 19/100, batch 2/625 completed (loss: 0.0006:   0%|          | 1/312 [00:09<32:07,  6.20s/it]
Training Epoch: 19/100, batch 2/625 completed (loss: 0.0032:   0%|          | 1/312 [00:09<32:17,  6.23s/it]
Training Epoch: 19/100, batch 2/625 completed (loss: 0.0149:   0%|          | 1/312 [00:09<32:29,  6.27s/it]
Training Epoch: 19/100, batch 2/625 completed (loss: 0.0006:   1%|          | 2/312 [00:11<30:16,  5.86s/it]
Training Epoch: 19/100, batch 2/625 completed (loss: 0.0004:   1%|          | 2/312 [00:11<30:17,  5.86s/it]
Training Epoch: 19/100, batch 2/625 completed (loss: 0.0149:   1%|          | 2/312 [00:11<30:23,  5.88s/it]
Training Epoch: 19/100, batch 2/625 completed (loss: 0.0032:   1%|          | 2/312 [00:11<30:18,  5.87s/it]
Training Epoch: 19/100, batch 3/625 completed (loss: 0.0013:   1%|          | 2/312 [00:11<30:17,  5.86s/it]
Training Epoch: 19/100, batch 3/625 completed (loss: 0.017:   1%|          | 2/312 [00:11<30:18,  5.87s/it] 
Training Epoch: 19/100, batch 3/625 completed (loss: 0.0303:   1%|          | 2/312 [00:11<30:16,  5.86s/it]
Training Epoch: 19/100, batch 3/625 completed (loss: 0.01:   1%|          | 2/312 [00:11<30:23,  5.88s/it]  
Training Epoch: 19/100, batch 4/625 completed (loss: 0.0006:   1%|          | 2/312 [00:14<30:16,  5.86s/it]
Training Epoch: 19/100, batch 4/625 completed (loss: 0.0049:   1%|          | 2/312 [00:14<30:17,  5.86s/it]
Training Epoch: 19/100, batch 4/625 completed (loss: 0.0009:   1%|          | 2/312 [00:14<30:18,  5.87s/it]
Training Epoch: 19/100, batch 4/625 completed (loss: 0.0007:   1%|          | 2/312 [00:14<30:23,  5.88s/it]
Training Epoch: 19/100, batch 4/625 completed (loss: 0.0007:   1%|          | 3/312 [00:17<30:41,  5.96s/it]
Training Epoch: 19/100, batch 4/625 completed (loss: 0.0009:   1%|          | 3/312 [00:17<30:38,  5.95s/it]
Training Epoch: 19/100, batch 4/625 completed (loss: 0.0049:   1%|          | 3/312 [00:17<30:39,  5.95s/it]
Training Epoch: 19/100, batch 4/625 completed (loss: 0.0006:   1%|          | 3/312 [00:17<30:39,  5.95s/it]
Training Epoch: 19/100, batch 5/625 completed (loss: 0.0009:   1%|          | 3/312 [00:17<30:39,  5.95s/it]
Training Epoch: 19/100, batch 5/625 completed (loss: 0.0002:   1%|          | 3/312 [00:17<30:39,  5.95s/it]
Training Epoch: 19/100, batch 5/625 completed (loss: 0.0002:   1%|          | 3/312 [00:17<30:38,  5.95s/it]
Training Epoch: 19/100, batch 5/625 completed (loss: 0.0004:   1%|          | 3/312 [00:17<30:41,  5.96s/it]
Training Epoch: 19/100, batch 6/625 completed (loss: 0.0016:   1%|          | 3/312 [00:20<30:39,  5.95s/it]
Training Epoch: 19/100, batch 6/625 completed (loss: 0.0001:   1%|          | 3/312 [00:20<30:39,  5.95s/it]
Training Epoch: 19/100, batch 6/625 completed (loss: 0.0009:   1%|          | 3/312 [00:20<30:38,  5.95s/it]
Training Epoch: 19/100, batch 6/625 completed (loss: 0.001:   1%|          | 3/312 [00:20<30:41,  5.96s/it] 
Training Epoch: 19/100, batch 6/625 completed (loss: 0.001:   1%|▏         | 4/312 [00:23<29:56,  5.83s/it]
Training Epoch: 19/100, batch 6/625 completed (loss: 0.0001:   1%|▏         | 4/312 [00:23<29:55,  5.83s/it]
Training Epoch: 19/100, batch 6/625 completed (loss: 0.0016:   1%|▏         | 4/312 [00:23<29:56,  5.83s/it]
Training Epoch: 19/100, batch 6/625 completed (loss: 0.0009:   1%|▏         | 4/312 [00:23<29:56,  5.83s/it]
Training Epoch: 19/100, batch 7/625 completed (loss: 0.0015:   1%|▏         | 4/312 [00:23<29:56,  5.83s/it]
Training Epoch: 19/100, batch 7/625 completed (loss: 0.0053:   1%|▏         | 4/312 [00:23<29:55,  5.83s/it]
Training Epoch: 19/100, batch 7/625 completed (loss: 0.0014:   1%|▏         | 4/312 [00:23<29:56,  5.83s/it]
Training Epoch: 19/100, batch 7/625 completed (loss: 0.0007:   1%|▏         | 4/312 [00:23<29:56,  5.83s/it]
Training Epoch: 19/100, batch 8/625 completed (loss: 0.0337:   1%|▏         | 4/312 [00:26<29:56,  5.83s/it]
Training Epoch: 19/100, batch 8/625 completed (loss: 0.0005:   1%|▏         | 4/312 [00:26<29:55,  5.83s/it]
Training Epoch: 19/100, batch 8/625 completed (loss: 0.0003:   1%|▏         | 4/312 [00:26<29:56,  5.83s/it]
Training Epoch: 19/100, batch 8/625 completed (loss: 0.0022:   1%|▏         | 4/312 [00:26<29:56,  5.83s/it]
Training Epoch: 19/100, batch 8/625 completed (loss: 0.0003:   2%|▏         | 5/312 [00:29<29:49,  5.83s/it]
Training Epoch: 19/100, batch 8/625 completed (loss: 0.0022:   2%|▏         | 5/312 [00:29<29:51,  5.84s/it]
Training Epoch: 19/100, batch 8/625 completed (loss: 0.0337:   2%|▏         | 5/312 [00:29<29:50,  5.83s/it]
Training Epoch: 19/100, batch 8/625 completed (loss: 0.0005:   2%|▏         | 5/312 [00:29<29:51,  5.83s/it]
Training Epoch: 19/100, batch 9/625 completed (loss: 0.0002:   2%|▏         | 5/312 [00:29<29:51,  5.83s/it]
Training Epoch: 19/100, batch 9/625 completed (loss: 0.0006:   2%|▏         | 5/312 [00:29<29:50,  5.83s/it]
Training Epoch: 19/100, batch 9/625 completed (loss: 0.0036:   2%|▏         | 5/312 [00:29<29:49,  5.83s/it]
Training Epoch: 19/100, batch 9/625 completed (loss: 0.0005:   2%|▏         | 5/312 [00:29<29:51,  5.84s/it]
Training Epoch: 19/100, batch 10/625 completed (loss: 0.0011:   2%|▏         | 5/312 [00:32<29:49,  5.83s/it]
Training Epoch: 19/100, batch 10/625 completed (loss: 0.0003:   2%|▏         | 5/312 [00:32<29:51,  5.83s/it]
Training Epoch: 19/100, batch 10/625 completed (loss: 0.0004:   2%|▏         | 5/312 [00:32<29:50,  5.83s/it]
Training Epoch: 19/100, batch 10/625 completed (loss: 0.0027:   2%|▏         | 5/312 [00:32<29:51,  5.84s/it]
Training Epoch: 19/100, batch 10/625 completed (loss: 0.0027:   2%|▏         | 6/312 [00:35<30:03,  5.89s/it]
Training Epoch: 19/100, batch 10/625 completed (loss: 0.0003:   2%|▏         | 6/312 [00:35<30:03,  5.89s/it]
Training Epoch: 19/100, batch 10/625 completed (loss: 0.0011:   2%|▏         | 6/312 [00:35<30:03,  5.89s/it]
Training Epoch: 19/100, batch 10/625 completed (loss: 0.0004:   2%|▏         | 6/312 [00:35<30:03,  5.90s/it]
Training Epoch: 19/100, batch 11/625 completed (loss: 0.0004:   2%|▏         | 6/312 [00:35<30:03,  5.89s/it]
Training Epoch: 19/100, batch 11/625 completed (loss: 0.0018:   2%|▏         | 6/312 [00:35<30:03,  5.89s/it]
Training Epoch: 19/100, batch 11/625 completed (loss: 0.0003:   2%|▏         | 6/312 [00:35<30:03,  5.90s/it]
Training Epoch: 19/100, batch 11/625 completed (loss: 0.0013:   2%|▏         | 6/312 [00:35<30:03,  5.89s/it]
Training Epoch: 19/100, batch 12/625 completed (loss: 0.0002:   2%|▏         | 6/312 [00:38<30:03,  5.90s/it]
Training Epoch: 19/100, batch 12/625 completed (loss: 0.003:   2%|▏         | 6/312 [00:38<30:03,  5.89s/it] 
Training Epoch: 19/100, batch 12/625 completed (loss: 0.0012:   2%|▏         | 6/312 [00:38<30:03,  5.89s/it]
Training Epoch: 19/100, batch 12/625 completed (loss: 0.0007:   2%|▏         | 6/312 [00:38<30:03,  5.89s/it]
Training Epoch: 19/100, batch 12/625 completed (loss: 0.0007:   2%|▏         | 7/312 [00:41<29:37,  5.83s/it]
Training Epoch: 19/100, batch 12/625 completed (loss: 0.003:   2%|▏         | 7/312 [00:41<29:36,  5.83s/it]
Training Epoch: 19/100, batch 12/625 completed (loss: 0.0012:   2%|▏         | 7/312 [00:41<29:37,  5.83s/it]
Training Epoch: 19/100, batch 12/625 completed (loss: 0.0002:   2%|▏         | 7/312 [00:41<29:41,  5.84s/it]
Training Epoch: 19/100, batch 13/625 completed (loss: 0.0007:   2%|▏         | 7/312 [00:41<29:36,  5.83s/it]
Training Epoch: 19/100, batch 13/625 completed (loss: 0.0019:   2%|▏         | 7/312 [00:41<29:41,  5.84s/it]
Training Epoch: 19/100, batch 13/625 completed (loss: 0.0029:   2%|▏         | 7/312 [00:41<29:37,  5.83s/it]
Training Epoch: 19/100, batch 13/625 completed (loss: 0.0017:   2%|▏         | 7/312 [00:41<29:37,  5.83s/it]
Training Epoch: 19/100, batch 14/625 completed (loss: 0.0037:   2%|▏         | 7/312 [00:43<29:41,  5.84s/it]
Training Epoch: 19/100, batch 14/625 completed (loss: 0.002:   2%|▏         | 7/312 [00:43<29:37,  5.83s/it] 
Training Epoch: 19/100, batch 14/625 completed (loss: 0.0018:   2%|▏         | 7/312 [00:43<29:36,  5.83s/it]
Training Epoch: 19/100, batch 14/625 completed (loss: 0.0006:   2%|▏         | 7/312 [00:44<29:37,  5.83s/it]
Training Epoch: 19/100, batch 14/625 completed (loss: 0.0037:   3%|▎         | 8/312 [00:46<29:35,  5.84s/it]
Training Epoch: 19/100, batch 14/625 completed (loss: 0.0006:   3%|▎         | 8/312 [00:47<29:37,  5.85s/it]
Training Epoch: 19/100, batch 14/625 completed (loss: 0.002:   3%|▎         | 8/312 [00:46<29:37,  5.85s/it]
Training Epoch: 19/100, batch 14/625 completed (loss: 0.0018:   3%|▎         | 8/312 [00:46<29:38,  5.85s/it]
Training Epoch: 19/100, batch 15/625 completed (loss: 0.0222:   3%|▎         | 8/312 [00:47<29:38,  5.85s/it]
Training Epoch: 19/100, batch 15/625 completed (loss: 0.0046:   3%|▎         | 8/312 [00:47<29:35,  5.84s/it]
Training Epoch: 19/100, batch 15/625 completed (loss: 0.0322:   3%|▎         | 8/312 [00:47<29:37,  5.85s/it]
Training Epoch: 19/100, batch 15/625 completed (loss: 0.0001:   3%|▎         | 8/312 [00:47<29:37,  5.85s/it]
Training Epoch: 19/100, batch 16/625 completed (loss: 0.0003:   3%|▎         | 8/312 [00:50<29:35,  5.84s/it]
Training Epoch: 19/100, batch 16/625 completed (loss: 0.0008:   3%|▎         | 8/312 [00:50<29:37,  5.85s/it]
Training Epoch: 19/100, batch 16/625 completed (loss: 0.002:   3%|▎         | 8/312 [00:50<29:38,  5.85s/it] 
Training Epoch: 19/100, batch 16/625 completed (loss: 0.0001:   3%|▎         | 8/312 [00:50<29:37,  5.85s/it]
Training Epoch: 19/100, batch 16/625 completed (loss: 0.002:   3%|▎         | 9/312 [00:53<30:01,  5.95s/it]
Training Epoch: 19/100, batch 16/625 completed (loss: 0.0001:   3%|▎         | 9/312 [00:53<30:03,  5.95s/it]
Training Epoch: 19/100, batch 16/625 completed (loss: 0.0008:   3%|▎         | 9/312 [00:53<30:03,  5.95s/it]
Training Epoch: 19/100, batch 16/625 completed (loss: 0.0003:   3%|▎         | 9/312 [00:53<30:03,  5.95s/it]
Training Epoch: 19/100, batch 17/625 completed (loss: 0.002:   3%|▎         | 9/312 [00:53<30:03,  5.95s/it] 
Training Epoch: 19/100, batch 17/625 completed (loss: 0.0003:   3%|▎         | 9/312 [00:53<30:03,  5.95s/it]
Training Epoch: 19/100, batch 17/625 completed (loss: 0.0013:   3%|▎         | 9/312 [00:53<30:01,  5.95s/it]
Training Epoch: 19/100, batch 17/625 completed (loss: 0.0003:   3%|▎         | 9/312 [00:53<30:03,  5.95s/it]
Training Epoch: 19/100, batch 18/625 completed (loss: 0.0001:   3%|▎         | 9/312 [00:56<30:03,  5.95s/it]
Training Epoch: 19/100, batch 18/625 completed (loss: 0.0009:   3%|▎         | 9/312 [00:56<30:01,  5.95s/it]
Training Epoch: 19/100, batch 18/625 completed (loss: 0.0086:   3%|▎         | 9/312 [00:56<30:03,  5.95s/it]
Training Epoch: 19/100, batch 18/625 completed (loss: 0.0004:   3%|▎         | 9/312 [00:56<30:03,  5.95s/it]
Training Epoch: 19/100, batch 18/625 completed (loss: 0.0001:   3%|▎         | 10/312 [00:58<29:34,  5.88s/it]
Training Epoch: 19/100, batch 18/625 completed (loss: 0.0004:   3%|▎         | 10/312 [00:58<29:36,  5.88s/it]
Training Epoch: 19/100, batch 18/625 completed (loss: 0.0009:   3%|▎         | 10/312 [00:58<29:36,  5.88s/it]
Training Epoch: 19/100, batch 18/625 completed (loss: 0.0086:   3%|▎         | 10/312 [00:58<29:36,  5.88s/it]
Training Epoch: 19/100, batch 19/625 completed (loss: 0.0002:   3%|▎         | 10/312 [00:58<29:36,  5.88s/it]
Training Epoch: 19/100, batch 19/625 completed (loss: 0.0098:   3%|▎         | 10/312 [00:58<29:34,  5.88s/it]
Training Epoch: 19/100, batch 19/625 completed (loss: 0.0173:   3%|▎         | 10/312 [00:58<29:36,  5.88s/it]
Training Epoch: 19/100, batch 19/625 completed (loss: 0.0004:   3%|▎         | 10/312 [00:58<29:36,  5.88s/it]
Training Epoch: 19/100, batch 20/625 completed (loss: 0.0001:   3%|▎         | 10/312 [01:01<29:36,  5.88s/it]
Training Epoch: 19/100, batch 20/625 completed (loss: 0.0455:   3%|▎         | 10/312 [01:01<29:34,  5.88s/it]
Training Epoch: 19/100, batch 20/625 completed (loss: 0.0002:   3%|▎         | 10/312 [01:01<29:36,  5.88s/it]
Training Epoch: 19/100, batch 20/625 completed (loss: 0.0012:   3%|▎         | 10/312 [01:01<29:36,  5.88s/it]
Training Epoch: 19/100, batch 20/625 completed (loss: 0.0012:   4%|▎         | 11/312 [01:05<30:01,  5.99s/it]
Training Epoch: 19/100, batch 20/625 completed (loss: 0.0455:   4%|▎         | 11/312 [01:05<30:01,  5.99s/it]
Training Epoch: 19/100, batch 20/625 completed (loss: 0.0002:   4%|▎         | 11/312 [01:05<30:01,  5.99s/it]
Training Epoch: 19/100, batch 20/625 completed (loss: 0.0001:   4%|▎         | 11/312 [01:05<30:01,  5.99s/it]
Training Epoch: 19/100, batch 21/625 completed (loss: 0.0004:   4%|▎         | 11/312 [01:05<30:01,  5.99s/it]
Training Epoch: 19/100, batch 21/625 completed (loss: 0.0004:   4%|▎         | 11/312 [01:05<30:01,  5.99s/it]
Training Epoch: 19/100, batch 21/625 completed (loss: 0.0006:   4%|▎         | 11/312 [01:05<30:01,  5.99s/it]
Training Epoch: 19/100, batch 21/625 completed (loss: 0.0027:   4%|▎         | 11/312 [01:05<30:01,  5.99s/it]
Training Epoch: 19/100, batch 22/625 completed (loss: 0.0002:   4%|▎         | 11/312 [01:07<30:01,  5.99s/it]
Training Epoch: 19/100, batch 22/625 completed (loss: 0.0007:   4%|▎         | 11/312 [01:07<30:01,  5.99s/it]
Training Epoch: 19/100, batch 22/625 completed (loss: 0.0084:   4%|▎         | 11/312 [01:07<30:01,  5.99s/it]
Training Epoch: 19/100, batch 22/625 completed (loss: 0.0002:   4%|▎         | 11/312 [01:08<30:01,  5.99s/it]
Training Epoch: 19/100, batch 22/625 completed (loss: 0.0002:   4%|▍         | 12/312 [01:10<29:29,  5.90s/it]
Training Epoch: 19/100, batch 22/625 completed (loss: 0.0002:   4%|▍         | 12/312 [01:10<29:31,  5.90s/it]
Training Epoch: 19/100, batch 22/625 completed (loss: 0.0007:   4%|▍         | 12/312 [01:10<29:31,  5.90s/it]
Training Epoch: 19/100, batch 22/625 completed (loss: 0.0084:   4%|▍         | 12/312 [01:10<29:31,  5.90s/it]
Training Epoch: 19/100, batch 23/625 completed (loss: 0.0003:   4%|▍         | 12/312 [01:10<29:31,  5.90s/it]
Training Epoch: 19/100, batch 23/625 completed (loss: 0.0009:   4%|▍         | 12/312 [01:10<29:29,  5.90s/it]
Training Epoch: 19/100, batch 23/625 completed (loss: 0.0005:   4%|▍         | 12/312 [01:10<29:31,  5.90s/it]
Training Epoch: 19/100, batch 23/625 completed (loss: 0.0046:   4%|▍         | 12/312 [01:10<29:31,  5.90s/it]
Training Epoch: 19/100, batch 24/625 completed (loss: 0.0005:   4%|▍         | 12/312 [01:13<29:31,  5.90s/it]
Training Epoch: 19/100, batch 24/625 completed (loss: 0.0002:   4%|▍         | 12/312 [01:13<29:31,  5.90s/it]
Training Epoch: 19/100, batch 24/625 completed (loss: 0.0001:   4%|▍         | 12/312 [01:13<29:29,  5.90s/it]
Training Epoch: 19/100, batch 24/625 completed (loss: 0.0002:   4%|▍         | 12/312 [01:13<29:31,  5.90s/it]
Training Epoch: 19/100, batch 24/625 completed (loss: 0.0002:   4%|▍         | 13/312 [01:16<29:10,  5.85s/it]
Training Epoch: 19/100, batch 24/625 completed (loss: 0.0001:   4%|▍         | 13/312 [01:16<29:10,  5.85s/it]
Training Epoch: 19/100, batch 24/625 completed (loss: 0.0002:   4%|▍         | 13/312 [01:16<29:11,  5.86s/it]
Training Epoch: 19/100, batch 24/625 completed (loss: 0.0005:   4%|▍         | 13/312 [01:16<29:11,  5.86s/it]
Training Epoch: 19/100, batch 25/625 completed (loss: 0.0034:   4%|▍         | 13/312 [01:16<29:10,  5.85s/it]
Training Epoch: 19/100, batch 25/625 completed (loss: 0.0016:   4%|▍         | 13/312 [01:16<29:10,  5.85s/it]
Training Epoch: 19/100, batch 25/625 completed (loss: 0.1739:   4%|▍         | 13/312 [01:16<29:11,  5.86s/it]
Training Epoch: 19/100, batch 25/625 completed (loss: 0.0005:   4%|▍         | 13/312 [01:16<29:11,  5.86s/it]
Training Epoch: 19/100, batch 26/625 completed (loss: 0.0011:   4%|▍         | 13/312 [01:19<29:10,  5.85s/it]
Training Epoch: 19/100, batch 26/625 completed (loss: 0.001:   4%|▍         | 13/312 [01:19<29:11,  5.86s/it] 
Training Epoch: 19/100, batch 26/625 completed (loss: 0.0018:   4%|▍         | 13/312 [01:19<29:10,  5.85s/it]
Training Epoch: 19/100, batch 26/625 completed (loss: 0.0255:   4%|▍         | 13/312 [01:20<29:11,  5.86s/it]
Training Epoch: 19/100, batch 26/625 completed (loss: 0.0255:   4%|▍         | 14/312 [01:22<29:52,  6.02s/it]
Training Epoch: 19/100, batch 26/625 completed (loss: 0.001:   4%|▍         | 14/312 [01:22<29:52,  6.01s/it]
Training Epoch: 19/100, batch 26/625 completed (loss: 0.0011:   4%|▍         | 14/312 [01:22<29:52,  6.02s/it]
Training Epoch: 19/100, batch 26/625 completed (loss: 0.0018:   4%|▍         | 14/312 [01:22<29:53,  6.02s/it]
Training Epoch: 19/100, batch 27/625 completed (loss: 0.0006:   4%|▍         | 14/312 [01:22<29:52,  6.02s/it]
Training Epoch: 19/100, batch 27/625 completed (loss: 0.0021:   4%|▍         | 14/312 [01:22<29:52,  6.01s/it]
Training Epoch: 19/100, batch 27/625 completed (loss: 0.0017:   4%|▍         | 14/312 [01:22<29:53,  6.02s/it]
Training Epoch: 19/100, batch 27/625 completed (loss: 0.0118:   4%|▍         | 14/312 [01:23<29:52,  6.02s/it]
Training Epoch: 19/100, batch 28/625 completed (loss: 0.0001:   4%|▍         | 14/312 [01:25<29:52,  6.01s/it]
Training Epoch: 19/100, batch 28/625 completed (loss: 0.0001:   4%|▍         | 14/312 [01:25<29:52,  6.02s/it]
Training Epoch: 19/100, batch 28/625 completed (loss: 0.0025:   4%|▍         | 14/312 [01:25<29:53,  6.02s/it]
Training Epoch: 19/100, batch 28/625 completed (loss: 0.0776:   4%|▍         | 14/312 [01:25<29:52,  6.02s/it]
Training Epoch: 19/100, batch 28/625 completed (loss: 0.0776:   5%|▍         | 15/312 [01:28<29:21,  5.93s/it]
Training Epoch: 19/100, batch 28/625 completed (loss: 0.0001:   5%|▍         | 15/312 [01:28<29:22,  5.93s/it]
Training Epoch: 19/100, batch 28/625 completed (loss: 0.0025:   5%|▍         | 15/312 [01:28<29:22,  5.93s/it]
Training Epoch: 19/100, batch 28/625 completed (loss: 0.0001:   5%|▍         | 15/312 [01:28<29:23,  5.94s/it]
Training Epoch: 19/100, batch 29/625 completed (loss: 0.0035:   5%|▍         | 15/312 [01:28<29:22,  5.93s/it]
Training Epoch: 19/100, batch 29/625 completed (loss: 0.0131:   5%|▍         | 15/312 [01:28<29:23,  5.94s/it]
Training Epoch: 19/100, batch 29/625 completed (loss: 0.0063:   5%|▍         | 15/312 [01:28<29:22,  5.93s/it]
Training Epoch: 19/100, batch 29/625 completed (loss: 0.0006:   5%|▍         | 15/312 [01:28<29:21,  5.93s/it]
Training Epoch: 19/100, batch 30/625 completed (loss: 0.0004:   5%|▍         | 15/312 [01:31<29:22,  5.93s/it]
Training Epoch: 19/100, batch 30/625 completed (loss: 0.0113:   5%|▍         | 15/312 [01:31<29:22,  5.93s/it]
Training Epoch: 19/100, batch 30/625 completed (loss: 0.0:   5%|▍         | 15/312 [01:31<29:23,  5.94s/it]   
Training Epoch: 19/100, batch 30/625 completed (loss: 0.0005:   5%|▍         | 15/312 [01:31<29:21,  5.93s/it]
Training Epoch: 19/100, batch 30/625 completed (loss: 0.0005:   5%|▌         | 16/312 [01:34<29:28,  5.98s/it]
Training Epoch: 19/100, batch 30/625 completed (loss: 0.0:   5%|▌         | 16/312 [01:34<29:28,  5.97s/it]
Training Epoch: 19/100, batch 30/625 completed (loss: 0.0004:   5%|▌         | 16/312 [01:34<29:28,  5.97s/it]
Training Epoch: 19/100, batch 30/625 completed (loss: 0.0113:   5%|▌         | 16/312 [01:34<29:28,  5.98s/it]
Training Epoch: 19/100, batch 31/625 completed (loss: 0.0011:   5%|▌         | 16/312 [01:34<29:28,  5.97s/it]
Training Epoch: 19/100, batch 31/625 completed (loss: 0.0006:   5%|▌         | 16/312 [01:34<29:28,  5.98s/it]
Training Epoch: 19/100, batch 31/625 completed (loss: 0.0001:   5%|▌         | 16/312 [01:34<29:28,  5.97s/it]
Training Epoch: 19/100, batch 31/625 completed (loss: 0.0238:   5%|▌         | 16/312 [01:34<29:28,  5.98s/it]
Training Epoch: 19/100, batch 32/625 completed (loss: 0.0015:   5%|▌         | 16/312 [01:37<29:28,  5.98s/it]
Training Epoch: 19/100, batch 32/625 completed (loss: 0.0114:   5%|▌         | 16/312 [01:37<29:28,  5.97s/it]
Training Epoch: 19/100, batch 32/625 completed (loss: 0.0955:   5%|▌         | 16/312 [01:37<29:28,  5.97s/it]
Training Epoch: 19/100, batch 32/625 completed (loss: 0.0001:   5%|▌         | 16/312 [01:38<29:28,  5.98s/it]
Training Epoch: 19/100, batch 32/625 completed (loss: 0.0001:   5%|▌         | 17/312 [01:40<29:30,  6.00s/it]
Training Epoch: 19/100, batch 32/625 completed (loss: 0.0114:   5%|▌         | 17/312 [01:40<29:31,  6.00s/it]
Training Epoch: 19/100, batch 32/625 completed (loss: 0.0955:   5%|▌         | 17/312 [01:40<29:33,  6.01s/it]
Training Epoch: 19/100, batch 32/625 completed (loss: 0.0015:   5%|▌         | 17/312 [01:40<29:33,  6.01s/it]
Training Epoch: 19/100, batch 33/625 completed (loss: 0.0001:   5%|▌         | 17/312 [01:40<29:33,  6.01s/it]
Training Epoch: 19/100, batch 33/625 completed (loss: 0.0001:   5%|▌         | 17/312 [01:40<29:33,  6.01s/it]
Training Epoch: 19/100, batch 33/625 completed (loss: 0.0003:   5%|▌         | 17/312 [01:40<29:31,  6.00s/it]
Training Epoch: 19/100, batch 33/625 completed (loss: 0.0007:   5%|▌         | 17/312 [01:40<29:30,  6.00s/it]
Training Epoch: 19/100, batch 34/625 completed (loss: 0.0002:   5%|▌         | 17/312 [01:43<29:33,  6.01s/it]
Training Epoch: 19/100, batch 34/625 completed (loss: 0.0028:   5%|▌         | 17/312 [01:43<29:33,  6.01s/it]
Training Epoch: 19/100, batch 34/625 completed (loss: 0.0003:   5%|▌         | 17/312 [01:43<29:31,  6.00s/it]
Training Epoch: 19/100, batch 34/625 completed (loss: 0.0017:   5%|▌         | 17/312 [01:43<29:30,  6.00s/it]
Training Epoch: 19/100, batch 34/625 completed (loss: 0.0017:   6%|▌         | 18/312 [01:46<29:07,  5.94s/it]
Training Epoch: 19/100, batch 34/625 completed (loss: 0.0028:   6%|▌         | 18/312 [01:46<29:06,  5.94s/it]
Training Epoch: 19/100, batch 34/625 completed (loss: 0.0002:   6%|▌         | 18/312 [01:46<29:06,  5.94s/it]
Training Epoch: 19/100, batch 34/625 completed (loss: 0.0003:   6%|▌         | 18/312 [01:46<29:07,  5.95s/it]
Training Epoch: 19/100, batch 35/625 completed (loss: 0.006:   6%|▌         | 18/312 [01:46<29:06,  5.94s/it] 
Training Epoch: 19/100, batch 35/625 completed (loss: 0.0001:   6%|▌         | 18/312 [01:46<29:07,  5.95s/it]
Training Epoch: 19/100, batch 35/625 completed (loss: 0.0004:   6%|▌         | 18/312 [01:46<29:06,  5.94s/it]
Training Epoch: 19/100, batch 35/625 completed (loss: 0.0003:   6%|▌         | 18/312 [01:46<29:07,  5.94s/it]
Training Epoch: 19/100, batch 36/625 completed (loss: 0.0046:   6%|▌         | 18/312 [01:49<29:06,  5.94s/it]
Training Epoch: 19/100, batch 36/625 completed (loss: 0.012:   6%|▌         | 18/312 [01:49<29:07,  5.95s/it] 
Training Epoch: 19/100, batch 36/625 completed (loss: 0.0007:   6%|▌         | 18/312 [01:49<29:06,  5.94s/it]
Training Epoch: 19/100, batch 36/625 completed (loss: 0.0006:   6%|▌         | 18/312 [01:49<29:07,  5.94s/it]
Training Epoch: 19/100, batch 36/625 completed (loss: 0.012:   6%|▌         | 19/312 [01:52<29:17,  6.00s/it]
Training Epoch: 19/100, batch 36/625 completed (loss: 0.0006:   6%|▌         | 19/312 [01:52<29:19,  6.00s/it]
Training Epoch: 19/100, batch 36/625 completed (loss: 0.0046:   6%|▌         | 19/312 [01:52<29:18,  6.00s/it]
Training Epoch: 19/100, batch 36/625 completed (loss: 0.0007:   6%|▌         | 19/312 [01:52<29:18,  6.00s/it]
Training Epoch: 19/100, batch 37/625 completed (loss: 0.0354:   6%|▌         | 19/312 [01:52<29:18,  6.00s/it]
Training Epoch: 19/100, batch 37/625 completed (loss: 0.0001:   6%|▌         | 19/312 [01:52<29:18,  6.00s/it]
Training Epoch: 19/100, batch 37/625 completed (loss: 0.0:   6%|▌         | 19/312 [01:52<29:17,  6.00s/it]  
Training Epoch: 19/100, batch 37/625 completed (loss: 0.0029:   6%|▌         | 19/312 [01:52<29:19,  6.00s/it]
Training Epoch: 19/100, batch 38/625 completed (loss: 0.0044:   6%|▌         | 19/312 [01:55<29:18,  6.00s/it]
Training Epoch: 19/100, batch 38/625 completed (loss: 0.0013:   6%|▌         | 19/312 [01:55<29:17,  6.00s/it]
Training Epoch: 19/100, batch 38/625 completed (loss: 0.0035:   6%|▌         | 19/312 [01:55<29:18,  6.00s/it]
Training Epoch: 19/100, batch 38/625 completed (loss: 0.0014:   6%|▌         | 19/312 [01:55<29:19,  6.00s/it]
Training Epoch: 19/100, batch 38/625 completed (loss: 0.0014:   6%|▋         | 20/312 [01:58<29:06,  5.98s/it]
Training Epoch: 19/100, batch 38/625 completed (loss: 0.0035:   6%|▋         | 20/312 [01:58<29:05,  5.98s/it]
Training Epoch: 19/100, batch 38/625 completed (loss: 0.0013:   6%|▋         | 20/312 [01:58<29:06,  5.98s/it]
Training Epoch: 19/100, batch 38/625 completed (loss: 0.0044:   6%|▋         | 20/312 [01:58<29:06,  5.98s/it]
Training Epoch: 19/100, batch 39/625 completed (loss: 0.0004:   6%|▋         | 20/312 [01:58<29:06,  5.98s/it]
Training Epoch: 19/100, batch 39/625 completed (loss: 0.0015:   6%|▋         | 20/312 [01:58<29:05,  5.98s/it]
Training Epoch: 19/100, batch 39/625 completed (loss: 0.0001:   6%|▋         | 20/312 [01:58<29:06,  5.98s/it]
Training Epoch: 19/100, batch 39/625 completed (loss: 0.002:   6%|▋         | 20/312 [01:58<29:06,  5.98s/it] 
Training Epoch: 19/100, batch 40/625 completed (loss: 0.0024:   6%|▋         | 20/312 [02:01<29:05,  5.98s/it]
Training Epoch: 19/100, batch 40/625 completed (loss: 0.0015:   6%|▋         | 20/312 [02:01<29:06,  5.98s/it]
Training Epoch: 19/100, batch 40/625 completed (loss: 0.0003:   6%|▋         | 20/312 [02:01<29:06,  5.98s/it]
Training Epoch: 19/100, batch 40/625 completed (loss: 0.0006:   6%|▋         | 20/312 [02:01<29:06,  5.98s/it]
Training Epoch: 19/100, batch 40/625 completed (loss: 0.0006:   7%|▋         | 21/312 [02:04<28:45,  5.93s/it]
Training Epoch: 19/100, batch 40/625 completed (loss: 0.0024:   7%|▋         | 21/312 [02:04<28:44,  5.93s/it]
Training Epoch: 19/100, batch 40/625 completed (loss: 0.0003:   7%|▋         | 21/312 [02:04<28:44,  5.93s/it]
Training Epoch: 19/100, batch 40/625 completed (loss: 0.0015:   7%|▋         | 21/312 [02:04<28:45,  5.93s/it]
Training Epoch: 19/100, batch 41/625 completed (loss: 0.0009:   7%|▋         | 21/312 [02:04<28:45,  5.93s/it]
Training Epoch: 19/100, batch 41/625 completed (loss: 0.0021:   7%|▋         | 21/312 [02:04<28:44,  5.93s/it]
Training Epoch: 19/100, batch 41/625 completed (loss: 0.0002:   7%|▋         | 21/312 [02:04<28:44,  5.93s/it]
Training Epoch: 19/100, batch 41/625 completed (loss: 0.0023:   7%|▋         | 21/312 [02:04<28:45,  5.93s/it]
Training Epoch: 19/100, batch 42/625 completed (loss: 0.0002:   7%|▋         | 21/312 [02:07<28:45,  5.93s/it]
Training Epoch: 19/100, batch 42/625 completed (loss: 0.0096:   7%|▋         | 21/312 [02:07<28:44,  5.93s/it]
Training Epoch: 19/100, batch 42/625 completed (loss: 0.0025:   7%|▋         | 21/312 [02:07<28:44,  5.93s/it]
Training Epoch: 19/100, batch 42/625 completed (loss: 0.0037:   7%|▋         | 21/312 [02:07<28:45,  5.93s/it]
Training Epoch: 19/100, batch 42/625 completed (loss: 0.0025:   7%|▋         | 22/312 [02:10<29:15,  6.05s/it]
Training Epoch: 19/100, batch 42/625 completed (loss: 0.0037:   7%|▋         | 22/312 [02:10<29:16,  6.06s/it]
Training Epoch: 19/100, batch 42/625 completed (loss: 0.0002:   7%|▋         | 22/312 [02:10<29:16,  6.06s/it]
Training Epoch: 19/100, batch 42/625 completed (loss: 0.0096:   7%|▋         | 22/312 [02:10<29:16,  6.06s/it]
Training Epoch: 19/100, batch 43/625 completed (loss: 0.0001:   7%|▋         | 22/312 [02:10<29:16,  6.06s/it]
Training Epoch: 19/100, batch 43/625 completed (loss: 0.0005:   7%|▋         | 22/312 [02:10<29:15,  6.05s/it]
Training Epoch: 19/100, batch 43/625 completed (loss: 0.0071:   7%|▋         | 22/312 [02:10<29:16,  6.06s/it]
Training Epoch: 19/100, batch 43/625 completed (loss: 0.0085:   7%|▋         | 22/312 [02:10<29:16,  6.06s/it]
Training Epoch: 19/100, batch 44/625 completed (loss: 0.0014:   7%|▋         | 22/312 [02:13<29:16,  6.06s/it]
Training Epoch: 19/100, batch 44/625 completed (loss: 0.0037:   7%|▋         | 22/312 [02:13<29:15,  6.05s/it]
Training Epoch: 19/100, batch 44/625 completed (loss: 0.0021:   7%|▋         | 22/312 [02:13<29:16,  6.06s/it]
Training Epoch: 19/100, batch 44/625 completed (loss: 0.0022:   7%|▋         | 22/312 [02:13<29:16,  6.06s/it]
Training Epoch: 19/100, batch 44/625 completed (loss: 0.0022:   7%|▋         | 23/312 [02:16<28:47,  5.98s/it]
Training Epoch: 19/100, batch 44/625 completed (loss: 0.0021:   7%|▋         | 23/312 [02:16<28:46,  5.98s/it]
Training Epoch: 19/100, batch 44/625 completed (loss: 0.0014:   7%|▋         | 23/312 [02:16<28:46,  5.97s/it]
Training Epoch: 19/100, batch 44/625 completed (loss: 0.0037:   7%|▋         | 23/312 [02:16<28:47,  5.98s/it]
Training Epoch: 19/100, batch 45/625 completed (loss: 0.0014:   7%|▋         | 23/312 [02:16<28:46,  5.98s/it]
Training Epoch: 19/100, batch 45/625 completed (loss: 0.0005:   7%|▋         | 23/312 [02:16<28:47,  5.98s/it]
Training Epoch: 19/100, batch 45/625 completed (loss: 0.0013:   7%|▋         | 23/312 [02:16<28:46,  5.97s/it]
Training Epoch: 19/100, batch 45/625 completed (loss: 0.0048:   7%|▋         | 23/312 [02:16<28:47,  5.98s/it]
Training Epoch: 19/100, batch 46/625 completed (loss: 0.0339:   7%|▋         | 23/312 [02:19<28:46,  5.98s/it]
Training Epoch: 19/100, batch 46/625 completed (loss: 0.0026:   7%|▋         | 23/312 [02:19<28:47,  5.98s/it]
Training Epoch: 19/100, batch 46/625 completed (loss: 0.0:   7%|▋         | 23/312 [02:19<28:46,  5.97s/it]   
Training Epoch: 19/100, batch 46/625 completed (loss: 0.0005:   7%|▋         | 23/312 [02:19<28:47,  5.98s/it]
Training Epoch: 19/100, batch 46/625 completed (loss: 0.0339:   8%|▊         | 24/312 [02:22<29:11,  6.08s/it]
Training Epoch: 19/100, batch 46/625 completed (loss: 0.0026:   8%|▊         | 24/312 [02:22<29:11,  6.08s/it]
Training Epoch: 19/100, batch 46/625 completed (loss: 0.0005:   8%|▊         | 24/312 [02:23<29:12,  6.08s/it]
Training Epoch: 19/100, batch 46/625 completed (loss: 0.0:   8%|▊         | 24/312 [02:22<29:12,  6.08s/it]
Training Epoch: 19/100, batch 47/625 completed (loss: 0.0007:   8%|▊         | 24/312 [02:23<29:11,  6.08s/it]
Training Epoch: 19/100, batch 47/625 completed (loss: 0.0029:   8%|▊         | 24/312 [02:23<29:11,  6.08s/it]
Training Epoch: 19/100, batch 47/625 completed (loss: 0.0155:   8%|▊         | 24/312 [02:23<29:12,  6.08s/it]
Training Epoch: 19/100, batch 47/625 completed (loss: 0.0007:   8%|▊         | 24/312 [02:23<29:12,  6.08s/it]
Training Epoch: 19/100, batch 48/625 completed (loss: 0.0002:   8%|▊         | 24/312 [02:26<29:11,  6.08s/it]
Training Epoch: 19/100, batch 48/625 completed (loss: 0.0009:   8%|▊         | 24/312 [02:26<29:11,  6.08s/it]
Training Epoch: 19/100, batch 48/625 completed (loss: 0.0011:   8%|▊         | 24/312 [02:26<29:12,  6.08s/it]
Training Epoch: 19/100, batch 48/625 completed (loss: 0.0125:   8%|▊         | 24/312 [02:26<29:12,  6.08s/it]
Training Epoch: 19/100, batch 48/625 completed (loss: 0.0009:   8%|▊         | 25/312 [02:28<28:57,  6.05s/it]
Training Epoch: 19/100, batch 48/625 completed (loss: 0.0002:   8%|▊         | 25/312 [02:28<28:57,  6.06s/it]
Training Epoch: 19/100, batch 48/625 completed (loss: 0.0011:   8%|▊         | 25/312 [02:28<28:57,  6.06s/it]
Training Epoch: 19/100, batch 48/625 completed (loss: 0.0125:   8%|▊         | 25/312 [02:29<28:58,  6.06s/it]
Training Epoch: 19/100, batch 49/625 completed (loss: 0.0005:   8%|▊         | 25/312 [02:29<28:57,  6.06s/it]
Training Epoch: 19/100, batch 49/625 completed (loss: 0.001:   8%|▊         | 25/312 [02:29<28:57,  6.06s/it] 
Training Epoch: 19/100, batch 49/625 completed (loss: 0.0013:   8%|▊         | 25/312 [02:29<28:57,  6.05s/it]
Training Epoch: 19/100, batch 49/625 completed (loss: 0.0663:   8%|▊         | 25/312 [02:29<28:58,  6.06s/it]
Training Epoch: 19/100, batch 50/625 completed (loss: 0.0006:   8%|▊         | 25/312 [02:31<28:57,  6.06s/it]
Training Epoch: 19/100, batch 50/625 completed (loss: 0.0389:   8%|▊         | 25/312 [02:31<28:57,  6.06s/it]
Training Epoch: 19/100, batch 50/625 completed (loss: 0.0037:   8%|▊         | 25/312 [02:31<28:57,  6.05s/it]
Training Epoch: 19/100, batch 50/625 completed (loss: 0.0557:   8%|▊         | 25/312 [02:31<28:58,  6.06s/it]
Training Epoch: 19/100, batch 50/625 completed (loss: 0.0037:   8%|▊         | 26/312 [02:34<28:30,  5.98s/it]
Training Epoch: 19/100, batch 50/625 completed (loss: 0.0389:   8%|▊         | 26/312 [02:34<28:30,  5.98s/it]
Training Epoch: 19/100, batch 50/625 completed (loss: 0.0557:   8%|▊         | 26/312 [02:34<28:31,  5.98s/it]
Training Epoch: 19/100, batch 50/625 completed (loss: 0.0006:   8%|▊         | 26/312 [02:34<28:31,  5.98s/it]
Training Epoch: 19/100, batch 51/625 completed (loss: 0.0:   8%|▊         | 26/312 [02:34<28:30,  5.98s/it]   
Training Epoch: 19/100, batch 51/625 completed (loss: 0.0018:   8%|▊         | 26/312 [02:34<28:31,  5.98s/it]
Training Epoch: 19/100, batch 51/625 completed (loss: 0.0014:   8%|▊         | 26/312 [02:34<28:30,  5.98s/it]
Training Epoch: 19/100, batch 51/625 completed (loss: 0.0093:   8%|▊         | 26/312 [02:34<28:31,  5.98s/it]
Training Epoch: 19/100, batch 52/625 completed (loss: 0.0038:   8%|▊         | 26/312 [02:38<28:31,  5.98s/it]
Training Epoch: 19/100, batch 52/625 completed (loss: 0.0006:   8%|▊         | 26/312 [02:38<28:30,  5.98s/it]
Training Epoch: 19/100, batch 52/625 completed (loss: 0.003:   8%|▊         | 26/312 [02:38<28:30,  5.98s/it] 
Training Epoch: 19/100, batch 52/625 completed (loss: 0.0056:   8%|▊         | 26/312 [02:38<28:31,  5.98s/it]
Training Epoch: 19/100, batch 52/625 completed (loss: 0.0006:   9%|▊         | 27/312 [02:41<29:01,  6.11s/it]
Training Epoch: 19/100, batch 52/625 completed (loss: 0.0056:   9%|▊         | 27/312 [02:41<29:01,  6.11s/it]
Training Epoch: 19/100, batch 52/625 completed (loss: 0.003:   9%|▊         | 27/312 [02:41<29:02,  6.11s/it]
Training Epoch: 19/100, batch 52/625 completed (loss: 0.0038:   9%|▊         | 27/312 [02:41<29:02,  6.11s/it]
Training Epoch: 19/100, batch 53/625 completed (loss: 0.0005:   9%|▊         | 27/312 [02:41<29:01,  6.11s/it]
Training Epoch: 19/100, batch 53/625 completed (loss: 0.0006:   9%|▊         | 27/312 [02:41<29:02,  6.11s/it]
Training Epoch: 19/100, batch 53/625 completed (loss: 0.0007:   9%|▊         | 27/312 [02:41<29:02,  6.11s/it]
Training Epoch: 19/100, batch 53/625 completed (loss: 0.0005:   9%|▊         | 27/312 [02:41<29:01,  6.11s/it]
Training Epoch: 19/100, batch 54/625 completed (loss: 0.0003:   9%|▊         | 27/312 [02:44<29:02,  6.11s/it]
Training Epoch: 19/100, batch 54/625 completed (loss: 0.0001:   9%|▊         | 27/312 [02:44<29:01,  6.11s/it]
Training Epoch: 19/100, batch 54/625 completed (loss: 0.0003:   9%|▊         | 27/312 [02:44<29:02,  6.11s/it]
Training Epoch: 19/100, batch 54/625 completed (loss: 0.0012:   9%|▊         | 27/312 [02:44<29:01,  6.11s/it]
Training Epoch: 19/100, batch 54/625 completed (loss: 0.0003:   9%|▉         | 28/312 [02:47<28:31,  6.03s/it]
Training Epoch: 19/100, batch 54/625 completed (loss: 0.0001:   9%|▉         | 28/312 [02:47<28:32,  6.03s/it]
Training Epoch: 19/100, batch 54/625 completed (loss: 0.0012:   9%|▉         | 28/312 [02:47<28:32,  6.03s/it]
Training Epoch: 19/100, batch 54/625 completed (loss: 0.0003:   9%|▉         | 28/312 [02:47<28:33,  6.03s/it]
Training Epoch: 19/100, batch 55/625 completed (loss: 0.011:   9%|▉         | 28/312 [02:47<28:31,  6.03s/it] 
Training Epoch: 19/100, batch 55/625 completed (loss: 0.0268:   9%|▉         | 28/312 [02:47<28:33,  6.03s/it]
Training Epoch: 19/100, batch 55/625 completed (loss: 0.0011:   9%|▉         | 28/312 [02:47<28:32,  6.03s/it]
Training Epoch: 19/100, batch 55/625 completed (loss: 0.0023:   9%|▉         | 28/312 [02:47<28:32,  6.03s/it]
Training Epoch: 19/100, batch 56/625 completed (loss: 0.0208:   9%|▉         | 28/312 [02:49<28:31,  6.03s/it]
Training Epoch: 19/100, batch 56/625 completed (loss: 0.0001:   9%|▉         | 28/312 [02:49<28:33,  6.03s/it]
Training Epoch: 19/100, batch 56/625 completed (loss: 0.0001:   9%|▉         | 28/312 [02:49<28:32,  6.03s/it]
Training Epoch: 19/100, batch 56/625 completed (loss: 0.0024:   9%|▉         | 28/312 [02:50<28:32,  6.03s/it]
Training Epoch: 19/100, batch 56/625 completed (loss: 0.0024:   9%|▉         | 29/312 [02:53<28:21,  6.01s/it]
Training Epoch: 19/100, batch 56/625 completed (loss: 0.0001:   9%|▉         | 29/312 [02:53<28:21,  6.01s/it]
Training Epoch: 19/100, batch 56/625 completed (loss: 0.0208:   9%|▉         | 29/312 [02:53<28:22,  6.02s/it]
Training Epoch: 19/100, batch 56/625 completed (loss: 0.0001:   9%|▉         | 29/312 [02:53<28:21,  6.01s/it]
Training Epoch: 19/100, batch 57/625 completed (loss: 0.0013:   9%|▉         | 29/312 [02:53<28:21,  6.01s/it]
Training Epoch: 19/100, batch 57/625 completed (loss: 0.001:   9%|▉         | 29/312 [02:53<28:22,  6.02s/it] 
Training Epoch: 19/100, batch 57/625 completed (loss: 0.0016:   9%|▉         | 29/312 [02:53<28:21,  6.01s/it]
Training Epoch: 19/100, batch 57/625 completed (loss: 0.0005:   9%|▉         | 29/312 [02:53<28:21,  6.01s/it]
Training Epoch: 19/100, batch 58/625 completed (loss: 0.0007:   9%|▉         | 29/312 [02:56<28:21,  6.01s/it]
Training Epoch: 19/100, batch 58/625 completed (loss: 0.0042:   9%|▉         | 29/312 [02:56<28:22,  6.02s/it]
Training Epoch: 19/100, batch 58/625 completed (loss: 0.006:   9%|▉         | 29/312 [02:56<28:21,  6.01s/it] 
Training Epoch: 19/100, batch 58/625 completed (loss: 0.0017:   9%|▉         | 29/312 [02:56<28:21,  6.01s/it]
Training Epoch: 19/100, batch 58/625 completed (loss: 0.0042:  10%|▉         | 30/312 [02:59<28:29,  6.06s/it]
Training Epoch: 19/100, batch 58/625 completed (loss: 0.0007:  10%|▉         | 30/312 [02:59<28:29,  6.06s/it]
Training Epoch: 19/100, batch 58/625 completed (loss: 0.006:  10%|▉         | 30/312 [02:59<28:29,  6.06s/it]
Training Epoch: 19/100, batch 58/625 completed (loss: 0.0017:  10%|▉         | 30/312 [02:59<28:30,  6.06s/it]
Training Epoch: 19/100, batch 59/625 completed (loss: 0.0074:  10%|▉         | 30/312 [02:59<28:29,  6.06s/it]
Training Epoch: 19/100, batch 59/625 completed (loss: 0.0004:  10%|▉         | 30/312 [02:59<28:29,  6.06s/it]
Training Epoch: 19/100, batch 59/625 completed (loss: 0.001:  10%|▉         | 30/312 [02:59<28:29,  6.06s/it] 
Training Epoch: 19/100, batch 59/625 completed (loss: 0.0008:  10%|▉         | 30/312 [02:59<28:30,  6.06s/it]
Training Epoch: 19/100, batch 60/625 completed (loss: 0.0005:  10%|▉         | 30/312 [03:02<28:29,  6.06s/it]
Training Epoch: 19/100, batch 60/625 completed (loss: 0.0001:  10%|▉         | 30/312 [03:02<28:29,  6.06s/it]
Training Epoch: 19/100, batch 60/625 completed (loss: 0.0002:  10%|▉         | 30/312 [03:02<28:29,  6.06s/it]
Training Epoch: 19/100, batch 60/625 completed (loss: 0.0002:  10%|▉         | 30/312 [03:02<28:30,  6.06s/it]
Training Epoch: 19/100, batch 60/625 completed (loss: 0.0001:  10%|▉         | 31/312 [03:04<28:01,  5.98s/it]
Training Epoch: 19/100, batch 60/625 completed (loss: 0.0002:  10%|▉         | 31/312 [03:04<28:00,  5.98s/it]
Training Epoch: 19/100, batch 60/625 completed (loss: 0.0005:  10%|▉         | 31/312 [03:04<28:01,  5.99s/it]
Training Epoch: 19/100, batch 60/625 completed (loss: 0.0002:  10%|▉         | 31/312 [03:05<28:01,  5.98s/it]
Training Epoch: 19/100, batch 61/625 completed (loss: 0.0009:  10%|▉         | 31/312 [03:05<28:01,  5.99s/it]
Training Epoch: 19/100, batch 61/625 completed (loss: 0.0001:  10%|▉         | 31/312 [03:05<28:00,  5.98s/it]
Training Epoch: 19/100, batch 61/625 completed (loss: 0.0008:  10%|▉         | 31/312 [03:05<28:01,  5.98s/it]
Training Epoch: 19/100, batch 61/625 completed (loss: 0.0009:  10%|▉         | 31/312 [03:05<28:01,  5.98s/it]
Training Epoch: 19/100, batch 62/625 completed (loss: 0.0013:  10%|▉         | 31/312 [03:08<28:01,  5.99s/it]
Training Epoch: 19/100, batch 62/625 completed (loss: 0.0013:  10%|▉         | 31/312 [03:08<28:01,  5.98s/it]
Training Epoch: 19/100, batch 62/625 completed (loss: 0.003:  10%|▉         | 31/312 [03:08<28:00,  5.98s/it] 
Training Epoch: 19/100, batch 62/625 completed (loss: 0.001:  10%|▉         | 31/312 [03:08<28:01,  5.98s/it] 
Training Epoch: 19/100, batch 62/625 completed (loss: 0.0013:  10%|█         | 32/312 [03:11<28:32,  6.12s/it]
Training Epoch: 19/100, batch 62/625 completed (loss: 0.0013:  10%|█         | 32/312 [03:11<28:32,  6.12s/it]
Training Epoch: 19/100, batch 62/625 completed (loss: 0.003:  10%|█         | 32/312 [03:11<28:32,  6.12s/it]
Training Epoch: 19/100, batch 62/625 completed (loss: 0.001:  10%|█         | 32/312 [03:11<28:32,  6.12s/it]
Training Epoch: 19/100, batch 63/625 completed (loss: 0.004:  10%|█         | 32/312 [03:11<28:32,  6.12s/it] 
Training Epoch: 19/100, batch 63/625 completed (loss: 0.0002:  10%|█         | 32/312 [03:11<28:32,  6.12s/it]
Training Epoch: 19/100, batch 63/625 completed (loss: 0.0008:  10%|█         | 32/312 [03:11<28:32,  6.12s/it]
Training Epoch: 19/100, batch 63/625 completed (loss: 0.0034:  10%|█         | 32/312 [03:11<28:32,  6.12s/it]
Training Epoch: 19/100, batch 64/625 completed (loss: 0.0001:  10%|█         | 32/312 [03:14<28:32,  6.12s/it]
Training Epoch: 19/100, batch 64/625 completed (loss: 0.0049:  10%|█         | 32/312 [03:14<28:32,  6.12s/it]
Training Epoch: 19/100, batch 64/625 completed (loss: 0.0003:  10%|█         | 32/312 [03:14<28:32,  6.12s/it]
Training Epoch: 19/100, batch 64/625 completed (loss: 0.0008:  10%|█         | 32/312 [03:14<28:32,  6.12s/it]
Training Epoch: 19/100, batch 64/625 completed (loss: 0.0001:  11%|█         | 33/312 [03:17<28:03,  6.03s/it]
Training Epoch: 19/100, batch 64/625 completed (loss: 0.0003:  11%|█         | 33/312 [03:17<28:03,  6.03s/it]
Training Epoch: 19/100, batch 64/625 completed (loss: 0.0008:  11%|█         | 33/312 [03:17<28:03,  6.04s/it]
Training Epoch: 19/100, batch 64/625 completed (loss: 0.0049:  11%|█         | 33/312 [03:17<28:03,  6.04s/it]
Training Epoch: 19/100, batch 65/625 completed (loss: 0.0003:  11%|█         | 33/312 [03:17<28:03,  6.03s/it]
Training Epoch: 19/100, batch 65/625 completed (loss: 0.0004:  11%|█         | 33/312 [03:17<28:03,  6.03s/it]
Training Epoch: 19/100, batch 65/625 completed (loss: 0.0003:  11%|█         | 33/312 [03:17<28:03,  6.04s/it]
Training Epoch: 19/100, batch 65/625 completed (loss: 0.0005:  11%|█         | 33/312 [03:17<28:03,  6.04s/it]
Training Epoch: 19/100, batch 66/625 completed (loss: 0.0003:  11%|█         | 33/312 [03:20<28:03,  6.04s/it]
Training Epoch: 19/100, batch 66/625 completed (loss: 0.0011:  11%|█         | 33/312 [03:20<28:03,  6.03s/it]
Training Epoch: 19/100, batch 66/625 completed (loss: 0.0496:  11%|█         | 33/312 [03:20<28:03,  6.03s/it]
Training Epoch: 19/100, batch 66/625 completed (loss: 0.0007:  11%|█         | 33/312 [03:20<28:03,  6.04s/it]
Training Epoch: 19/100, batch 66/625 completed (loss: 0.0007:  11%|█         | 34/312 [03:23<27:41,  5.98s/it]
Training Epoch: 19/100, batch 66/625 completed (loss: 0.0011:  11%|█         | 34/312 [03:23<27:43,  5.98s/it]
Training Epoch: 19/100, batch 66/625 completed (loss: 0.0496:  11%|█         | 34/312 [03:23<27:44,  5.99s/it]
Training Epoch: 19/100, batch 66/625 completed (loss: 0.0003:  11%|█         | 34/312 [03:23<27:43,  5.99s/it]
Training Epoch: 19/100, batch 67/625 completed (loss: 0.0024:  11%|█         | 34/312 [03:23<27:44,  5.99s/it]
Training Epoch: 19/100, batch 67/625 completed (loss: 0.0012:  11%|█         | 34/312 [03:23<27:43,  5.98s/it]
Training Epoch: 19/100, batch 67/625 completed (loss: 0.0003:  11%|█         | 34/312 [03:23<27:43,  5.99s/it]
Training Epoch: 19/100, batch 67/625 completed (loss: 0.0004:  11%|█         | 34/312 [03:23<27:41,  5.98s/it]
Training Epoch: 19/100, batch 68/625 completed (loss: 0.0001:  11%|█         | 34/312 [03:26<27:44,  5.99s/it]
Training Epoch: 19/100, batch 68/625 completed (loss: 0.008:  11%|█         | 34/312 [03:26<27:43,  5.99s/it] 
Training Epoch: 19/100, batch 68/625 completed (loss: 0.0014:  11%|█         | 34/312 [03:26<27:43,  5.98s/it]
Training Epoch: 19/100, batch 68/625 completed (loss: 0.0016:  11%|█         | 34/312 [03:26<27:41,  5.98s/it]
Training Epoch: 19/100, batch 68/625 completed (loss: 0.0001:  11%|█         | 35/312 [03:29<28:05,  6.08s/it]
Training Epoch: 19/100, batch 68/625 completed (loss: 0.0014:  11%|█         | 35/312 [03:29<28:05,  6.09s/it]
Training Epoch: 19/100, batch 68/625 completed (loss: 0.0016:  11%|█         | 35/312 [03:29<28:05,  6.09s/it]
Training Epoch: 19/100, batch 68/625 completed (loss: 0.008:  11%|█         | 35/312 [03:29<28:05,  6.09s/it]
Training Epoch: 19/100, batch 69/625 completed (loss: 0.0005:  11%|█         | 35/312 [03:29<28:05,  6.08s/it]
Training Epoch: 19/100, batch 69/625 completed (loss: 0.0004:  11%|█         | 35/312 [03:29<28:05,  6.09s/it]
Training Epoch: 19/100, batch 69/625 completed (loss: 0.0001:  11%|█         | 35/312 [03:29<28:05,  6.09s/it]
Training Epoch: 19/100, batch 69/625 completed (loss: 0.0154:  11%|█         | 35/312 [03:29<28:05,  6.09s/it]
Training Epoch: 19/100, batch 70/625 completed (loss: 0.0033:  11%|█         | 35/312 [03:32<28:05,  6.08s/it]
Training Epoch: 19/100, batch 70/625 completed (loss: 0.0002:  11%|█         | 35/312 [03:32<28:05,  6.09s/it]
Training Epoch: 19/100, batch 70/625 completed (loss: 0.002:  11%|█         | 35/312 [03:32<28:05,  6.09s/it] 
Training Epoch: 19/100, batch 70/625 completed (loss: 0.0004:  11%|█         | 35/312 [03:32<28:05,  6.09s/it]
Training Epoch: 19/100, batch 70/625 completed (loss: 0.0033:  12%|█▏        | 36/312 [03:35<27:37,  6.01s/it]
Training Epoch: 19/100, batch 70/625 completed (loss: 0.002:  12%|█▏        | 36/312 [03:35<27:38,  6.01s/it]
Training Epoch: 19/100, batch 70/625 completed (loss: 0.0002:  12%|█▏        | 36/312 [03:35<27:37,  6.01s/it]
Training Epoch: 19/100, batch 70/625 completed (loss: 0.0004:  12%|█▏        | 36/312 [03:35<27:38,  6.01s/it]
Training Epoch: 19/100, batch 71/625 completed (loss: 0.0011:  12%|█▏        | 36/312 [03:35<27:37,  6.01s/it]
Training Epoch: 19/100, batch 71/625 completed (loss: 0.0001:  12%|█▏        | 36/312 [03:35<27:37,  6.01s/it]
Training Epoch: 19/100, batch 71/625 completed (loss: 0.0044:  12%|█▏        | 36/312 [03:35<27:38,  6.01s/it]
Training Epoch: 19/100, batch 71/625 completed (loss: 0.0234:  12%|█▏        | 36/312 [03:35<27:38,  6.01s/it]
Training Epoch: 19/100, batch 72/625 completed (loss: 0.0017:  12%|█▏        | 36/312 [03:38<27:37,  6.01s/it]
Training Epoch: 19/100, batch 72/625 completed (loss: 0.0003:  12%|█▏        | 36/312 [03:38<27:37,  6.01s/it]
Training Epoch: 19/100, batch 72/625 completed (loss: 0.0048:  12%|█▏        | 36/312 [03:38<27:38,  6.01s/it]
Training Epoch: 19/100, batch 72/625 completed (loss: 0.0022:  12%|█▏        | 36/312 [03:38<27:38,  6.01s/it]
Training Epoch: 19/100, batch 72/625 completed (loss: 0.0048:  12%|█▏        | 37/312 [03:41<27:40,  6.04s/it]
Training Epoch: 19/100, batch 72/625 completed (loss: 0.0017:  12%|█▏        | 37/312 [03:41<27:40,  6.04s/it]
Training Epoch: 19/100, batch 72/625 completed (loss: 0.0003:  12%|█▏        | 37/312 [03:41<27:40,  6.04s/it]
Training Epoch: 19/100, batch 72/625 completed (loss: 0.0022:  12%|█▏        | 37/312 [03:41<27:40,  6.04s/it]
Training Epoch: 19/100, batch 73/625 completed (loss: 0.002:  12%|█▏        | 37/312 [03:41<27:40,  6.04s/it] 
Training Epoch: 19/100, batch 73/625 completed (loss: 0.0014:  12%|█▏        | 37/312 [03:41<27:40,  6.04s/it]
Training Epoch: 19/100, batch 73/625 completed (loss: 0.0001:  12%|█▏        | 37/312 [03:41<27:40,  6.04s/it]
Training Epoch: 19/100, batch 73/625 completed (loss: 0.019:  12%|█▏        | 37/312 [03:41<27:40,  6.04s/it] 
Training Epoch: 19/100, batch 74/625 completed (loss: 0.0019:  12%|█▏        | 37/312 [03:44<27:40,  6.04s/it]
Training Epoch: 19/100, batch 74/625 completed (loss: 0.0025:  12%|█▏        | 37/312 [03:44<27:40,  6.04s/it]
Training Epoch: 19/100, batch 74/625 completed (loss: 0.0001:  12%|█▏        | 37/312 [03:44<27:40,  6.04s/it]
Training Epoch: 19/100, batch 74/625 completed (loss: 0.0006:  12%|█▏        | 37/312 [03:44<27:40,  6.04s/it]
Training Epoch: 19/100, batch 74/625 completed (loss: 0.0006:  12%|█▏        | 38/312 [03:47<27:45,  6.08s/it]
Training Epoch: 19/100, batch 74/625 completed (loss: 0.0019:  12%|█▏        | 38/312 [03:47<27:46,  6.08s/it]
Training Epoch: 19/100, batch 74/625 completed (loss: 0.0001:  12%|█▏        | 38/312 [03:47<27:47,  6.08s/it]
Training Epoch: 19/100, batch 74/625 completed (loss: 0.0025:  12%|█▏        | 38/312 [03:47<27:46,  6.08s/it]
Training Epoch: 19/100, batch 75/625 completed (loss: 0.0007:  12%|█▏        | 38/312 [03:47<27:47,  6.08s/it]
Training Epoch: 19/100, batch 75/625 completed (loss: 0.0002:  12%|█▏        | 38/312 [03:47<27:46,  6.08s/it]
Training Epoch: 19/100, batch 75/625 completed (loss: 0.0259:  12%|█▏        | 38/312 [03:47<27:46,  6.08s/it]
Training Epoch: 19/100, batch 75/625 completed (loss: 0.0002:  12%|█▏        | 38/312 [03:47<27:45,  6.08s/it]
Training Epoch: 19/100, batch 76/625 completed (loss: 0.0016:  12%|█▏        | 38/312 [03:50<27:47,  6.08s/it]
Training Epoch: 19/100, batch 76/625 completed (loss: 0.0003:  12%|█▏        | 38/312 [03:50<27:46,  6.08s/it]
Training Epoch: 19/100, batch 76/625 completed (loss: 0.0334:  12%|█▏        | 38/312 [03:50<27:46,  6.08s/it]
Training Epoch: 19/100, batch 76/625 completed (loss: 0.009:  12%|█▏        | 38/312 [03:50<27:45,  6.08s/it] 
Training Epoch: 19/100, batch 76/625 completed (loss: 0.0016:  12%|█▎        | 39/312 [03:53<27:22,  6.02s/it]
Training Epoch: 19/100, batch 76/625 completed (loss: 0.009:  12%|█▎        | 39/312 [03:53<27:22,  6.02s/it]
Training Epoch: 19/100, batch 76/625 completed (loss: 0.0003:  12%|█▎        | 39/312 [03:53<27:22,  6.02s/it]
Training Epoch: 19/100, batch 76/625 completed (loss: 0.0334:  12%|█▎        | 39/312 [03:53<27:23,  6.02s/it]
Training Epoch: 19/100, batch 77/625 completed (loss: 0.0015:  12%|█▎        | 39/312 [03:53<27:22,  6.02s/it]
Training Epoch: 19/100, batch 77/625 completed (loss: 0.0003:  12%|█▎        | 39/312 [03:53<27:22,  6.02s/it]
Training Epoch: 19/100, batch 77/625 completed (loss: 0.0072:  12%|█▎        | 39/312 [03:53<27:23,  6.02s/it]
Training Epoch: 19/100, batch 77/625 completed (loss: 0.0002:  12%|█▎        | 39/312 [03:53<27:22,  6.02s/it]
Training Epoch: 19/100, batch 78/625 completed (loss: 0.0008:  12%|█▎        | 39/312 [03:56<27:22,  6.02s/it]
Training Epoch: 19/100, batch 78/625 completed (loss: 0.0004:  12%|█▎        | 39/312 [03:56<27:22,  6.02s/it]
Training Epoch: 19/100, batch 78/625 completed (loss: 0.002:  12%|█▎        | 39/312 [03:56<27:23,  6.02s/it] 
Training Epoch: 19/100, batch 78/625 completed (loss: 0.0012:  12%|█▎        | 39/312 [03:56<27:22,  6.02s/it]
Training Epoch: 19/100, batch 78/625 completed (loss: 0.0004:  13%|█▎        | 40/312 [03:59<27:20,  6.03s/it]
Training Epoch: 19/100, batch 78/625 completed (loss: 0.0012:  13%|█▎        | 40/312 [03:59<27:20,  6.03s/it]
Training Epoch: 19/100, batch 78/625 completed (loss: 0.002:  13%|█▎        | 40/312 [03:59<27:20,  6.03s/it]
Training Epoch: 19/100, batch 78/625 completed (loss: 0.0008:  13%|█▎        | 40/312 [03:59<27:20,  6.03s/it]
Training Epoch: 19/100, batch 79/625 completed (loss: 0.0004:  13%|█▎        | 40/312 [03:59<27:20,  6.03s/it]
Training Epoch: 19/100, batch 79/625 completed (loss: 0.0007:  13%|█▎        | 40/312 [03:59<27:20,  6.03s/it]
Training Epoch: 19/100, batch 79/625 completed (loss: 0.0005:  13%|█▎        | 40/312 [03:59<27:20,  6.03s/it]
Training Epoch: 19/100, batch 79/625 completed (loss: 0.002:  13%|█▎        | 40/312 [03:59<27:20,  6.03s/it] 
Training Epoch: 19/100, batch 80/625 completed (loss: 0.0003:  13%|█▎        | 40/312 [04:02<27:20,  6.03s/it]
Training Epoch: 19/100, batch 80/625 completed (loss: 0.0032:  13%|█▎        | 40/312 [04:02<27:20,  6.03s/it]
Training Epoch: 19/100, batch 80/625 completed (loss: 0.0044:  13%|█▎        | 40/312 [04:02<27:20,  6.03s/it]
Training Epoch: 19/100, batch 80/625 completed (loss: 0.0001:  13%|█▎        | 40/312 [04:02<27:20,  6.03s/it]
Training Epoch: 19/100, batch 80/625 completed (loss: 0.0032:  13%|█▎        | 41/312 [04:05<27:11,  6.02s/it]
Training Epoch: 19/100, batch 80/625 completed (loss: 0.0044:  13%|█▎        | 41/312 [04:05<27:11,  6.02s/it]
Training Epoch: 19/100, batch 80/625 completed (loss: 0.0001:  13%|█▎        | 41/312 [04:05<27:12,  6.02s/it]
Training Epoch: 19/100, batch 80/625 completed (loss: 0.0003:  13%|█▎        | 41/312 [04:05<27:12,  6.03s/it]
Training Epoch: 19/100, batch 81/625 completed (loss: 0.0005:  13%|█▎        | 41/312 [04:05<27:11,  6.02s/it]
Training Epoch: 19/100, batch 81/625 completed (loss: 0.0034:  13%|█▎        | 41/312 [04:05<27:11,  6.02s/it]
Training Epoch: 19/100, batch 81/625 completed (loss: 0.038:  13%|█▎        | 41/312 [04:05<27:12,  6.03s/it] 
Training Epoch: 19/100, batch 81/625 completed (loss: 0.0108:  13%|█▎        | 41/312 [04:05<27:12,  6.02s/it]
Training Epoch: 19/100, batch 82/625 completed (loss: 0.0012:  13%|█▎        | 41/312 [04:08<27:12,  6.03s/it]
Training Epoch: 19/100, batch 82/625 completed (loss: 0.0003:  13%|█▎        | 41/312 [04:08<27:11,  6.02s/it]
Training Epoch: 19/100, batch 82/625 completed (loss: 0.0002:  13%|█▎        | 41/312 [04:08<27:11,  6.02s/it]
Training Epoch: 19/100, batch 82/625 completed (loss: 0.0031:  13%|█▎        | 41/312 [04:08<27:12,  6.02s/it]
Training Epoch: 19/100, batch 82/625 completed (loss: 0.0031:  13%|█▎        | 42/312 [04:11<27:02,  6.01s/it]
Training Epoch: 19/100, batch 82/625 completed (loss: 0.0002:  13%|█▎        | 42/312 [04:11<27:03,  6.01s/it]
Training Epoch: 19/100, batch 82/625 completed (loss: 0.0003:  13%|█▎        | 42/312 [04:11<27:03,  6.01s/it]
Training Epoch: 19/100, batch 82/625 completed (loss: 0.0012:  13%|█▎        | 42/312 [04:11<27:03,  6.01s/it]
Training Epoch: 19/100, batch 83/625 completed (loss: 0.0039:  13%|█▎        | 42/312 [04:11<27:03,  6.01s/it]
Training Epoch: 19/100, batch 83/625 completed (loss: 0.0261:  13%|█▎        | 42/312 [04:11<27:03,  6.01s/it]
Training Epoch: 19/100, batch 83/625 completed (loss: 0.0045:  13%|█▎        | 42/312 [04:11<27:03,  6.01s/it]
Training Epoch: 19/100, batch 83/625 completed (loss: 0.0008:  13%|█▎        | 42/312 [04:11<27:02,  6.01s/it]
Training Epoch: 19/100, batch 84/625 completed (loss: 0.0007:  13%|█▎        | 42/312 [04:14<27:03,  6.01s/it]
Training Epoch: 19/100, batch 84/625 completed (loss: 0.003:  13%|█▎        | 42/312 [04:14<27:03,  6.01s/it] 
Training Epoch: 19/100, batch 84/625 completed (loss: 0.0027:  13%|█▎        | 42/312 [04:14<27:03,  6.01s/it]
Training Epoch: 19/100, batch 84/625 completed (loss: 0.0006:  13%|█▎        | 42/312 [04:14<27:02,  6.01s/it]
Training Epoch: 19/100, batch 84/625 completed (loss: 0.003:  14%|█▍        | 43/312 [04:17<27:07,  6.05s/it]
Training Epoch: 19/100, batch 84/625 completed (loss: 0.0007:  14%|█▍        | 43/312 [04:17<27:08,  6.05s/it]
Training Epoch: 19/100, batch 84/625 completed (loss: 0.0006:  14%|█▍        | 43/312 [04:17<27:08,  6.05s/it]
Training Epoch: 19/100, batch 84/625 completed (loss: 0.0027:  14%|█▍        | 43/312 [04:17<27:08,  6.05s/it]
Training Epoch: 19/100, batch 85/625 completed (loss: 0.0:  14%|█▍        | 43/312 [04:17<27:08,  6.05s/it]   
Training Epoch: 19/100, batch 85/625 completed (loss: 0.0004:  14%|█▍        | 43/312 [04:17<27:07,  6.05s/it]
Training Epoch: 19/100, batch 85/625 completed (loss: 0.0006:  14%|█▍        | 43/312 [04:17<27:08,  6.05s/it]
Training Epoch: 19/100, batch 85/625 completed (loss: 0.0009:  14%|█▍        | 43/312 [04:17<27:08,  6.05s/it]
Training Epoch: 19/100, batch 86/625 completed (loss: 0.0006:  14%|█▍        | 43/312 [04:20<27:07,  6.05s/it]
Training Epoch: 19/100, batch 86/625 completed (loss: 0.0001:  14%|█▍        | 43/312 [04:20<27:08,  6.05s/it]
Training Epoch: 19/100, batch 86/625 completed (loss: 0.0003:  14%|█▍        | 43/312 [04:20<27:08,  6.05s/it]
Training Epoch: 19/100, batch 86/625 completed (loss: 0.0033:  14%|█▍        | 43/312 [04:20<27:08,  6.05s/it]
Training Epoch: 19/100, batch 86/625 completed (loss: 0.0033:  14%|█▍        | 44/312 [04:23<26:57,  6.03s/it]
Training Epoch: 19/100, batch 86/625 completed (loss: 0.0003:  14%|█▍        | 44/312 [04:23<26:57,  6.04s/it]
Training Epoch: 19/100, batch 86/625 completed (loss: 0.0006:  14%|█▍        | 44/312 [04:23<26:57,  6.04s/it]
Training Epoch: 19/100, batch 86/625 completed (loss: 0.0001:  14%|█▍        | 44/312 [04:23<26:57,  6.03s/it]
Training Epoch: 19/100, batch 87/625 completed (loss: 0.006:  14%|█▍        | 44/312 [04:23<26:57,  6.03s/it] 
Training Epoch: 19/100, batch 87/625 completed (loss: 0.0011:  14%|█▍        | 44/312 [04:23<26:57,  6.04s/it]
Training Epoch: 19/100, batch 87/625 completed (loss: 0.0001:  14%|█▍        | 44/312 [04:23<26:57,  6.04s/it]
Training Epoch: 19/100, batch 87/625 completed (loss: 0.0062:  14%|█▍        | 44/312 [04:23<26:57,  6.03s/it]
Training Epoch: 19/100, batch 88/625 completed (loss: 0.007:  14%|█▍        | 44/312 [04:26<26:57,  6.04s/it] 
Training Epoch: 19/100, batch 88/625 completed (loss: 0.0058:  14%|█▍        | 44/312 [04:26<26:57,  6.03s/it]
Training Epoch: 19/100, batch 88/625 completed (loss: 0.0001:  14%|█▍        | 44/312 [04:26<26:57,  6.04s/it]
Training Epoch: 19/100, batch 88/625 completed (loss: 0.002:  14%|█▍        | 44/312 [04:26<26:57,  6.03s/it] 
Training Epoch: 19/100, batch 88/625 completed (loss: 0.007:  14%|█▍        | 45/312 [04:29<27:01,  6.07s/it]
Training Epoch: 19/100, batch 88/625 completed (loss: 0.002:  14%|█▍        | 45/312 [04:29<27:02,  6.08s/it]
Training Epoch: 19/100, batch 88/625 completed (loss: 0.0058:  14%|█▍        | 45/312 [04:29<27:01,  6.07s/it]
Training Epoch: 19/100, batch 88/625 completed (loss: 0.0001:  14%|█▍        | 45/312 [04:29<27:02,  6.08s/it]
Training Epoch: 19/100, batch 89/625 completed (loss: 0.0045:  14%|█▍        | 45/312 [04:29<27:02,  6.08s/it]
Training Epoch: 19/100, batch 89/625 completed (loss: 0.0035:  14%|█▍        | 45/312 [04:29<27:01,  6.07s/it]
Training Epoch: 19/100, batch 89/625 completed (loss: 0.0027:  14%|█▍        | 45/312 [04:29<27:01,  6.07s/it]
Training Epoch: 19/100, batch 89/625 completed (loss: 0.7877:  14%|█▍        | 45/312 [04:29<27:02,  6.08s/it]
Training Epoch: 19/100, batch 90/625 completed (loss: 0.0147:  14%|█▍        | 45/312 [04:32<27:01,  6.07s/it]
Training Epoch: 19/100, batch 90/625 completed (loss: 0.001:  14%|█▍        | 45/312 [04:32<27:01,  6.07s/it] 
Training Epoch: 19/100, batch 90/625 completed (loss: 0.0061:  14%|█▍        | 45/312 [04:32<27:02,  6.08s/it]
Training Epoch: 19/100, batch 90/625 completed (loss: 0.0021:  14%|█▍        | 45/312 [04:32<27:02,  6.08s/it]
Training Epoch: 19/100, batch 90/625 completed (loss: 0.001:  15%|█▍        | 46/312 [04:35<26:49,  6.05s/it]
Training Epoch: 19/100, batch 90/625 completed (loss: 0.0147:  15%|█▍        | 46/312 [04:35<26:49,  6.05s/it]
Training Epoch: 19/100, batch 90/625 completed (loss: 0.0021:  15%|█▍        | 46/312 [04:35<26:49,  6.05s/it]
Training Epoch: 19/100, batch 90/625 completed (loss: 0.0061:  15%|█▍        | 46/312 [04:35<26:49,  6.05s/it]
Training Epoch: 19/100, batch 91/625 completed (loss: 0.0048:  15%|█▍        | 46/312 [04:35<26:49,  6.05s/it]
Training Epoch: 19/100, batch 91/625 completed (loss: 0.0011:  15%|█▍        | 46/312 [04:35<26:49,  6.05s/it]
Training Epoch: 19/100, batch 91/625 completed (loss: 0.0001:  15%|█▍        | 46/312 [04:35<26:49,  6.05s/it]
Training Epoch: 19/100, batch 91/625 completed (loss: 0.0037:  15%|█▍        | 46/312 [04:35<26:49,  6.05s/it]
Training Epoch: 19/100, batch 92/625 completed (loss: 0.0025:  15%|█▍        | 46/312 [04:38<26:49,  6.05s/it]
Training Epoch: 19/100, batch 92/625 completed (loss: 0.0011:  15%|█▍        | 46/312 [04:38<26:49,  6.05s/it]
Training Epoch: 19/100, batch 92/625 completed (loss: 0.0041:  15%|█▍        | 46/312 [04:38<26:49,  6.05s/it]
Training Epoch: 19/100, batch 92/625 completed (loss: 0.0006:  15%|█▍        | 46/312 [04:38<26:49,  6.05s/it]
Training Epoch: 19/100, batch 92/625 completed (loss: 0.0011:  15%|█▌        | 47/312 [04:41<26:27,  5.99s/it]
Training Epoch: 19/100, batch 92/625 completed (loss: 0.0041:  15%|█▌        | 47/312 [04:41<26:27,  5.99s/it]
Training Epoch: 19/100, batch 92/625 completed (loss: 0.0025:  15%|█▌        | 47/312 [04:41<26:27,  5.99s/it]
Training Epoch: 19/100, batch 92/625 completed (loss: 0.0006:  15%|█▌        | 47/312 [04:41<26:28,  5.99s/it]
Training Epoch: 19/100, batch 93/625 completed (loss: 0.0:  15%|█▌        | 47/312 [04:41<26:27,  5.99s/it]   
Training Epoch: 19/100, batch 93/625 completed (loss: 0.0003:  15%|█▌        | 47/312 [04:41<26:27,  5.99s/it]
Training Epoch: 19/100, batch 93/625 completed (loss: 0.0014:  15%|█▌        | 47/312 [04:41<26:27,  5.99s/it]
Training Epoch: 19/100, batch 93/625 completed (loss: 0.0005:  15%|█▌        | 47/312 [04:41<26:28,  5.99s/it]
Training Epoch: 19/100, batch 94/625 completed (loss: 0.0016:  15%|█▌        | 47/312 [04:44<26:27,  5.99s/it]
Training Epoch: 19/100, batch 94/625 completed (loss: 0.0003:  15%|█▌        | 47/312 [04:44<26:27,  5.99s/it]
Training Epoch: 19/100, batch 94/625 completed (loss: 0.0003:  15%|█▌        | 47/312 [04:44<26:27,  5.99s/it]
Training Epoch: 19/100, batch 94/625 completed (loss: 0.0006:  15%|█▌        | 47/312 [04:44<26:28,  5.99s/it]
Training Epoch: 19/100, batch 94/625 completed (loss: 0.0003:  15%|█▌        | 48/312 [04:47<26:45,  6.08s/it]
Training Epoch: 19/100, batch 94/625 completed (loss: 0.0006:  15%|█▌        | 48/312 [04:47<26:45,  6.08s/it]
Training Epoch: 19/100, batch 94/625 completed (loss: 0.0016:  15%|█▌        | 48/312 [04:47<26:45,  6.08s/it]
Training Epoch: 19/100, batch 94/625 completed (loss: 0.0003:  15%|█▌        | 48/312 [04:47<26:46,  6.08s/it]
Training Epoch: 19/100, batch 95/625 completed (loss: 0.0:  15%|█▌        | 48/312 [04:47<26:45,  6.08s/it]   
Training Epoch: 19/100, batch 95/625 completed (loss: 0.01:  15%|█▌        | 48/312 [04:47<26:46,  6.08s/it]  
Training Epoch: 19/100, batch 95/625 completed (loss: 0.0004:  15%|█▌        | 48/312 [04:47<26:45,  6.08s/it]
Training Epoch: 19/100, batch 95/625 completed (loss: 0.0021:  15%|█▌        | 48/312 [04:48<26:45,  6.08s/it]
Training Epoch: 19/100, batch 96/625 completed (loss: 0.0066:  15%|█▌        | 48/312 [04:50<26:45,  6.08s/it]
Training Epoch: 19/100, batch 96/625 completed (loss: 0.0008:  15%|█▌        | 48/312 [04:50<26:46,  6.08s/it]
Training Epoch: 19/100, batch 96/625 completed (loss: 0.0008:  15%|█▌        | 48/312 [04:50<26:45,  6.08s/it]
Training Epoch: 19/100, batch 96/625 completed (loss: 0.0004:  15%|█▌        | 48/312 [04:50<26:45,  6.08s/it]
Training Epoch: 19/100, batch 96/625 completed (loss: 0.0004:  16%|█▌        | 49/312 [04:53<26:31,  6.05s/it]
Training Epoch: 19/100, batch 96/625 completed (loss: 0.0008:  16%|█▌        | 49/312 [04:53<26:30,  6.05s/it]
Training Epoch: 19/100, batch 96/625 completed (loss: 0.0066:  16%|█▌        | 49/312 [04:53<26:31,  6.05s/it]
Training Epoch: 19/100, batch 96/625 completed (loss: 0.0008:  16%|█▌        | 49/312 [04:53<26:31,  6.05s/it]
Training Epoch: 19/100, batch 97/625 completed (loss: 0.0007:  16%|█▌        | 49/312 [04:53<26:31,  6.05s/it]
Training Epoch: 19/100, batch 97/625 completed (loss: 0.0003:  16%|█▌        | 49/312 [04:53<26:31,  6.05s/it]
Training Epoch: 19/100, batch 97/625 completed (loss: 0.0014:  16%|█▌        | 49/312 [04:53<26:30,  6.05s/it]
Training Epoch: 19/100, batch 97/625 completed (loss: 0.0003:  16%|█▌        | 49/312 [04:54<26:31,  6.05s/it]
Training Epoch: 19/100, batch 98/625 completed (loss: 0.0003:  16%|█▌        | 49/312 [04:56<26:31,  6.05s/it]
Training Epoch: 19/100, batch 98/625 completed (loss: 0.0002:  16%|█▌        | 49/312 [04:56<26:31,  6.05s/it]
Training Epoch: 19/100, batch 98/625 completed (loss: 0.0074:  16%|█▌        | 49/312 [04:56<26:30,  6.05s/it]
Training Epoch: 19/100, batch 98/625 completed (loss: 0.0011:  16%|█▌        | 49/312 [04:56<26:31,  6.05s/it]
Training Epoch: 19/100, batch 98/625 completed (loss: 0.0011:  16%|█▌        | 50/312 [05:00<26:34,  6.09s/it]
Training Epoch: 19/100, batch 98/625 completed (loss: 0.0074:  16%|█▌        | 50/312 [05:00<26:35,  6.09s/it]
Training Epoch: 19/100, batch 98/625 completed (loss: 0.0003:  16%|█▌        | 50/312 [05:00<26:35,  6.09s/it]
Training Epoch: 19/100, batch 98/625 completed (loss: 0.0002:  16%|█▌        | 50/312 [05:00<26:36,  6.09s/it]
Training Epoch: 19/100, batch 99/625 completed (loss: 0.0009:  16%|█▌        | 50/312 [05:00<26:36,  6.09s/it]
Training Epoch: 19/100, batch 99/625 completed (loss: 0.0053:  16%|█▌        | 50/312 [05:00<26:35,  6.09s/it]
Training Epoch: 19/100, batch 99/625 completed (loss: 0.0:  16%|█▌        | 50/312 [05:00<26:35,  6.09s/it]   
Training Epoch: 19/100, batch 99/625 completed (loss: 0.0007:  16%|█▌        | 50/312 [05:00<26:34,  6.09s/it]
Training Epoch: 19/100, batch 100/625 completed (loss: 0.1849:  16%|█▌        | 50/312 [05:03<26:36,  6.09s/it]
Training Epoch: 19/100, batch 100/625 completed (loss: 0.0037:  16%|█▌        | 50/312 [05:03<26:35,  6.09s/it]
Training Epoch: 19/100, batch 100/625 completed (loss: 0.0004:  16%|█▌        | 50/312 [05:03<26:35,  6.09s/it]
Training Epoch: 19/100, batch 100/625 completed (loss: 0.0002:  16%|█▌        | 50/312 [05:03<26:34,  6.09s/it]
Training Epoch: 19/100, batch 100/625 completed (loss: 0.0004:  16%|█▋        | 51/312 [05:05<26:09,  6.01s/it]
Training Epoch: 19/100, batch 100/625 completed (loss: 0.0002:  16%|█▋        | 51/312 [05:05<26:09,  6.01s/it]
Training Epoch: 19/100, batch 100/625 completed (loss: 0.0037:  16%|█▋        | 51/312 [05:05<26:10,  6.02s/it]
Training Epoch: 19/100, batch 100/625 completed (loss: 0.1849:  16%|█▋        | 51/312 [05:05<26:10,  6.02s/it]
Training Epoch: 19/100, batch 101/625 completed (loss: 0.0055:  16%|█▋        | 51/312 [05:05<26:10,  6.02s/it]
Training Epoch: 19/100, batch 101/625 completed (loss: 0.0006:  16%|█▋        | 51/312 [05:05<26:09,  6.01s/it]
Training Epoch: 19/100, batch 101/625 completed (loss: 0.0012:  16%|█▋        | 51/312 [05:05<26:10,  6.02s/it]
Training Epoch: 19/100, batch 101/625 completed (loss: 0.0038:  16%|█▋        | 51/312 [05:06<26:09,  6.01s/it]
Training Epoch: 19/100, batch 102/625 completed (loss: 0.0006:  16%|█▋        | 51/312 [05:09<26:10,  6.02s/it]
Training Epoch: 19/100, batch 102/625 completed (loss: 0.0003:  16%|█▋        | 51/312 [05:09<26:10,  6.02s/it]
Training Epoch: 19/100, batch 102/625 completed (loss: 0.0005:  16%|█▋        | 51/312 [05:09<26:09,  6.01s/it]
Training Epoch: 19/100, batch 102/625 completed (loss: 0.0001:  16%|█▋        | 51/312 [05:09<26:09,  6.01s/it]
Training Epoch: 19/100, batch 102/625 completed (loss: 0.0006:  17%|█▋        | 52/312 [05:11<26:04,  6.02s/it]
Training Epoch: 19/100, batch 102/625 completed (loss: 0.0003:  17%|█▋        | 52/312 [05:11<26:05,  6.02s/it]
Training Epoch: 19/100, batch 102/625 completed (loss: 0.0005:  17%|█▋        | 52/312 [05:11<26:06,  6.02s/it]
Training Epoch: 19/100, batch 102/625 completed (loss: 0.0001:  17%|█▋        | 52/312 [05:12<26:06,  6.02s/it]
Training Epoch: 19/100, batch 103/625 completed (loss: 0.0664:  17%|█▋        | 52/312 [05:12<26:05,  6.02s/it]
Training Epoch: 19/100, batch 103/625 completed (loss: 0.0001:  17%|█▋        | 52/312 [05:12<26:04,  6.02s/it]
Training Epoch: 19/100, batch 103/625 completed (loss: 0.0002:  17%|█▋        | 52/312 [05:12<26:06,  6.02s/it]
Training Epoch: 19/100, batch 103/625 completed (loss: 0.0001:  17%|█▋        | 52/312 [05:12<26:06,  6.02s/it]
Training Epoch: 19/100, batch 104/625 completed (loss: 0.0015:  17%|█▋        | 52/312 [05:15<26:05,  6.02s/it]
Training Epoch: 19/100, batch 104/625 completed (loss: 0.0007:  17%|█▋        | 52/312 [05:15<26:06,  6.02s/it]
Training Epoch: 19/100, batch 104/625 completed (loss: 0.0038:  17%|█▋        | 52/312 [05:15<26:04,  6.02s/it]
Training Epoch: 19/100, batch 104/625 completed (loss: 0.0014:  17%|█▋        | 52/312 [05:15<26:06,  6.02s/it]
Training Epoch: 19/100, batch 104/625 completed (loss: 0.0014:  17%|█▋        | 53/312 [05:18<26:20,  6.10s/it]
Training Epoch: 19/100, batch 104/625 completed (loss: 0.0038:  17%|█▋        | 53/312 [05:18<26:20,  6.10s/it]
Training Epoch: 19/100, batch 104/625 completed (loss: 0.0007:  17%|█▋        | 53/312 [05:18<26:21,  6.11s/it]
Training Epoch: 19/100, batch 104/625 completed (loss: 0.0015:  17%|█▋        | 53/312 [05:18<26:21,  6.10s/it]
Training Epoch: 19/100, batch 105/625 completed (loss: 0.0005:  17%|█▋        | 53/312 [05:18<26:20,  6.10s/it]
Training Epoch: 19/100, batch 105/625 completed (loss: 0.0001:  17%|█▋        | 53/312 [05:18<26:21,  6.10s/it]
Training Epoch: 19/100, batch 105/625 completed (loss: 0.0605:  17%|█▋        | 53/312 [05:18<26:21,  6.11s/it]
Training Epoch: 19/100, batch 105/625 completed (loss: 0.0047:  17%|█▋        | 53/312 [05:18<26:20,  6.10s/it]
Training Epoch: 19/100, batch 106/625 completed (loss: 0.0048:  17%|█▋        | 53/312 [05:21<26:20,  6.10s/it]
Training Epoch: 19/100, batch 106/625 completed (loss: 0.0079:  17%|█▋        | 53/312 [05:21<26:21,  6.11s/it]
Training Epoch: 19/100, batch 106/625 completed (loss: 0.0016:  17%|█▋        | 53/312 [05:21<26:21,  6.10s/it]
Training Epoch: 19/100, batch 106/625 completed (loss: 0.0005:  17%|█▋        | 53/312 [05:21<26:20,  6.10s/it]
Training Epoch: 19/100, batch 106/625 completed (loss: 0.0016:  17%|█▋        | 54/312 [05:24<26:11,  6.09s/it]
Training Epoch: 19/100, batch 106/625 completed (loss: 0.0048:  17%|█▋        | 54/312 [05:24<26:11,  6.09s/it]
Training Epoch: 19/100, batch 106/625 completed (loss: 0.0005:  17%|█▋        | 54/312 [05:24<26:12,  6.09s/it]
Training Epoch: 19/100, batch 106/625 completed (loss: 0.0079:  17%|█▋        | 54/312 [05:24<26:11,  6.09s/it]
Training Epoch: 19/100, batch 107/625 completed (loss: 0.002:  17%|█▋        | 54/312 [05:24<26:11,  6.09s/it] 
Training Epoch: 19/100, batch 107/625 completed (loss: 0.0001:  17%|█▋        | 54/312 [05:24<26:11,  6.09s/it]
Training Epoch: 19/100, batch 107/625 completed (loss: 0.0004:  17%|█▋        | 54/312 [05:24<26:11,  6.09s/it]
Training Epoch: 19/100, batch 107/625 completed (loss: 0.0022:  17%|█▋        | 54/312 [05:24<26:12,  6.09s/it]
Training Epoch: 19/100, batch 108/625 completed (loss: 0.0012:  17%|█▋        | 54/312 [05:27<26:11,  6.09s/it]
Training Epoch: 19/100, batch 108/625 completed (loss: 0.0018:  17%|█▋        | 54/312 [05:27<26:11,  6.09s/it]
Training Epoch: 19/100, batch 108/625 completed (loss: 0.0042:  17%|█▋        | 54/312 [05:27<26:11,  6.09s/it]
Training Epoch: 19/100, batch 108/625 completed (loss: 0.0025:  17%|█▋        | 54/312 [05:27<26:12,  6.09s/it]
Training Epoch: 19/100, batch 108/625 completed (loss: 0.0025:  18%|█▊        | 55/312 [05:30<25:58,  6.07s/it]
Training Epoch: 19/100, batch 108/625 completed (loss: 0.0018:  18%|█▊        | 55/312 [05:30<25:58,  6.06s/it]
Training Epoch: 19/100, batch 108/625 completed (loss: 0.0042:  18%|█▊        | 55/312 [05:30<25:59,  6.07s/it]
Training Epoch: 19/100, batch 108/625 completed (loss: 0.0012:  18%|█▊        | 55/312 [05:30<25:59,  6.07s/it]
Training Epoch: 19/100, batch 109/625 completed (loss: 0.0001:  18%|█▊        | 55/312 [05:30<25:58,  6.06s/it]
Training Epoch: 19/100, batch 109/625 completed (loss: 0.0005:  18%|█▊        | 55/312 [05:30<25:59,  6.07s/it]
Training Epoch: 19/100, batch 109/625 completed (loss: 0.0005:  18%|█▊        | 55/312 [05:30<25:59,  6.07s/it]
Training Epoch: 19/100, batch 109/625 completed (loss: 0.0048:  18%|█▊        | 55/312 [05:30<25:58,  6.07s/it]
Training Epoch: 19/100, batch 110/625 completed (loss: 0.0:  18%|█▊        | 55/312 [05:33<25:58,  6.06s/it]   
Training Epoch: 19/100, batch 110/625 completed (loss: 0.0007:  18%|█▊        | 55/312 [05:33<25:59,  6.07s/it]
Training Epoch: 19/100, batch 110/625 completed (loss: 0.0:  18%|█▊        | 55/312 [05:33<25:59,  6.07s/it]   
Training Epoch: 19/100, batch 110/625 completed (loss: 0.0021:  18%|█▊        | 55/312 [05:33<25:58,  6.07s/it]
Training Epoch: 19/100, batch 110/625 completed (loss: 0.0:  18%|█▊        | 56/312 [05:36<25:59,  6.09s/it]
Training Epoch: 19/100, batch 110/625 completed (loss: 0.0:  18%|█▊        | 56/312 [05:36<25:59,  6.09s/it]
Training Epoch: 19/100, batch 110/625 completed (loss: 0.0007:  18%|█▊        | 56/312 [05:36<25:59,  6.09s/it]
Training Epoch: 19/100, batch 110/625 completed (loss: 0.0021:  18%|█▊        | 56/312 [05:36<25:59,  6.09s/it]
Training Epoch: 19/100, batch 111/625 completed (loss: 0.0008:  18%|█▊        | 56/312 [05:36<25:59,  6.09s/it]
Training Epoch: 19/100, batch 111/625 completed (loss: 0.0006:  18%|█▊        | 56/312 [05:36<25:59,  6.09s/it]
Training Epoch: 19/100, batch 111/625 completed (loss: 0.0012:  18%|█▊        | 56/312 [05:36<25:59,  6.09s/it]
Training Epoch: 19/100, batch 111/625 completed (loss: 0.0021:  18%|█▊        | 56/312 [05:36<25:59,  6.09s/it]
Training Epoch: 19/100, batch 112/625 completed (loss: 0.0012:  18%|█▊        | 56/312 [05:39<25:59,  6.09s/it]
Training Epoch: 19/100, batch 112/625 completed (loss: 0.0017:  18%|█▊        | 56/312 [05:39<25:59,  6.09s/it]
Training Epoch: 19/100, batch 112/625 completed (loss: 0.0015:  18%|█▊        | 56/312 [05:39<25:59,  6.09s/it]
Training Epoch: 19/100, batch 112/625 completed (loss: 0.0035:  18%|█▊        | 56/312 [05:39<25:59,  6.09s/it]
Training Epoch: 19/100, batch 112/625 completed (loss: 0.0035:  18%|█▊        | 57/312 [05:42<25:48,  6.07s/it]
Training Epoch: 19/100, batch 112/625 completed (loss: 0.0015:  18%|█▊        | 57/312 [05:42<25:48,  6.07s/it]
Training Epoch: 19/100, batch 112/625 completed (loss: 0.0012:  18%|█▊        | 57/312 [05:42<25:48,  6.07s/it]
Training Epoch: 19/100, batch 112/625 completed (loss: 0.0017:  18%|█▊        | 57/312 [05:42<25:49,  6.08s/it]
Training Epoch: 19/100, batch 113/625 completed (loss: 0.0001:  18%|█▊        | 57/312 [05:42<25:48,  6.07s/it]
Training Epoch: 19/100, batch 113/625 completed (loss: 0.0002:  18%|█▊        | 57/312 [05:42<25:48,  6.07s/it]
Training Epoch: 19/100, batch 113/625 completed (loss: 0.0003:  18%|█▊        | 57/312 [05:42<25:49,  6.08s/it]
Training Epoch: 19/100, batch 113/625 completed (loss: 0.0001:  18%|█▊        | 57/312 [05:42<25:48,  6.07s/it]
Training Epoch: 19/100, batch 114/625 completed (loss: 0.0099:  18%|█▊        | 57/312 [05:45<25:48,  6.07s/it]
Training Epoch: 19/100, batch 114/625 completed (loss: 0.0002:  18%|█▊        | 57/312 [05:45<25:49,  6.08s/it]
Training Epoch: 19/100, batch 114/625 completed (loss: 0.0005:  18%|█▊        | 57/312 [05:45<25:48,  6.07s/it]
Training Epoch: 19/100, batch 114/625 completed (loss: 0.0028:  18%|█▊        | 57/312 [05:45<25:48,  6.07s/it]
Training Epoch: 19/100, batch 114/625 completed (loss: 0.0099:  19%|█▊        | 58/312 [05:48<25:58,  6.14s/it]
Training Epoch: 19/100, batch 114/625 completed (loss: 0.0005:  19%|█▊        | 58/312 [05:48<25:58,  6.14s/it]
Training Epoch: 19/100, batch 114/625 completed (loss: 0.0002:  19%|█▊        | 58/312 [05:48<25:58,  6.14s/it]
Training Epoch: 19/100, batch 114/625 completed (loss: 0.0028:  19%|█▊        | 58/312 [05:48<25:59,  6.14s/it]
Training Epoch: 19/100, batch 115/625 completed (loss: 0.0006:  19%|█▊        | 58/312 [05:48<25:58,  6.14s/it]
Training Epoch: 19/100, batch 115/625 completed (loss: 0.0034:  19%|█▊        | 58/312 [05:48<25:58,  6.14s/it]
Training Epoch: 19/100, batch 115/625 completed (loss: 0.0006:  19%|█▊        | 58/312 [05:48<25:58,  6.14s/it]
Training Epoch: 19/100, batch 115/625 completed (loss: 0.0001:  19%|█▊        | 58/312 [05:48<25:59,  6.14s/it]
Training Epoch: 19/100, batch 116/625 completed (loss: 0.0007:  19%|█▊        | 58/312 [05:51<25:58,  6.14s/it]
Training Epoch: 19/100, batch 116/625 completed (loss: 0.0006:  19%|█▊        | 58/312 [05:51<25:58,  6.14s/it]
Training Epoch: 19/100, batch 116/625 completed (loss: 0.0004:  19%|█▊        | 58/312 [05:51<25:58,  6.14s/it]
Training Epoch: 19/100, batch 116/625 completed (loss: 0.0005:  19%|█▊        | 58/312 [05:51<25:59,  6.14s/it]
Training Epoch: 19/100, batch 116/625 completed (loss: 0.0005:  19%|█▉        | 59/312 [05:54<25:27,  6.04s/it]
Training Epoch: 19/100, batch 116/625 completed (loss: 0.0004:  19%|█▉        | 59/312 [05:54<25:28,  6.04s/it]
Training Epoch: 19/100, batch 116/625 completed (loss: 0.0006:  19%|█▉        | 59/312 [05:54<25:28,  6.04s/it]
Training Epoch: 19/100, batch 116/625 completed (loss: 0.0007:  19%|█▉        | 59/312 [05:54<25:28,  6.04s/it]
Training Epoch: 19/100, batch 117/625 completed (loss: 0.0013:  19%|█▉        | 59/312 [05:54<25:28,  6.04s/it]
Training Epoch: 19/100, batch 117/625 completed (loss: 0.0012:  19%|█▉        | 59/312 [05:54<25:28,  6.04s/it]
Training Epoch: 19/100, batch 117/625 completed (loss: 0.0001:  19%|█▉        | 59/312 [05:54<25:28,  6.04s/it]
Training Epoch: 19/100, batch 117/625 completed (loss: 0.0003:  19%|█▉        | 59/312 [05:54<25:27,  6.04s/it]
Training Epoch: 19/100, batch 118/625 completed (loss: 0.0004:  19%|█▉        | 59/312 [05:57<25:28,  6.04s/it]
Training Epoch: 19/100, batch 118/625 completed (loss: 0.0006:  19%|█▉        | 59/312 [05:57<25:28,  6.04s/it]
Training Epoch: 19/100, batch 118/625 completed (loss: 0.0006:  19%|█▉        | 59/312 [05:57<25:28,  6.04s/it]
Training Epoch: 19/100, batch 118/625 completed (loss: 0.0028:  19%|█▉        | 59/312 [05:57<25:27,  6.04s/it]
Training Epoch: 19/100, batch 118/625 completed (loss: 0.0028:  19%|█▉        | 60/312 [06:00<25:32,  6.08s/it]
Training Epoch: 19/100, batch 118/625 completed (loss: 0.0006:  19%|█▉        | 60/312 [06:00<25:32,  6.08s/it]
Training Epoch: 19/100, batch 118/625 completed (loss: 0.0006:  19%|█▉        | 60/312 [06:00<25:32,  6.08s/it]
Training Epoch: 19/100, batch 118/625 completed (loss: 0.0004:  19%|█▉        | 60/312 [06:00<25:32,  6.08s/it]
Training Epoch: 19/100, batch 119/625 completed (loss: 0.0003:  19%|█▉        | 60/312 [06:00<25:32,  6.08s/it]
Training Epoch: 19/100, batch 119/625 completed (loss: 0.0037:  19%|█▉        | 60/312 [06:00<25:32,  6.08s/it]
Training Epoch: 19/100, batch 119/625 completed (loss: 0.0002:  19%|█▉        | 60/312 [06:00<25:32,  6.08s/it]
Training Epoch: 19/100, batch 119/625 completed (loss: 0.0148:  19%|█▉        | 60/312 [06:00<25:32,  6.08s/it]
Training Epoch: 19/100, batch 120/625 completed (loss: 0.0002:  19%|█▉        | 60/312 [06:03<25:32,  6.08s/it]
Training Epoch: 19/100, batch 120/625 completed (loss: 0.0005:  19%|█▉        | 60/312 [06:03<25:32,  6.08s/it]
Training Epoch: 19/100, batch 120/625 completed (loss: 0.0003:  19%|█▉        | 60/312 [06:03<25:32,  6.08s/it]
Training Epoch: 19/100, batch 120/625 completed (loss: 0.0001:  19%|█▉        | 60/312 [06:03<25:32,  6.08s/it]
Training Epoch: 19/100, batch 120/625 completed (loss: 0.0001:  20%|█▉        | 61/312 [06:06<25:30,  6.10s/it]
Training Epoch: 19/100, batch 120/625 completed (loss: 0.0005:  20%|█▉        | 61/312 [06:06<25:31,  6.10s/it]
Training Epoch: 19/100, batch 120/625 completed (loss: 0.0002:  20%|█▉        | 61/312 [06:06<25:31,  6.10s/it]
Training Epoch: 19/100, batch 120/625 completed (loss: 0.0003:  20%|█▉        | 61/312 [06:06<25:32,  6.10s/it]
Training Epoch: 19/100, batch 121/625 completed (loss: 0.0071:  20%|█▉        | 61/312 [06:06<25:31,  6.10s/it]
Training Epoch: 19/100, batch 121/625 completed (loss: 0.0002:  20%|█▉        | 61/312 [06:06<25:32,  6.10s/it]
Training Epoch: 19/100, batch 121/625 completed (loss: 0.0004:  20%|█▉        | 61/312 [06:06<25:31,  6.10s/it]
Training Epoch: 19/100, batch 121/625 completed (loss: 0.0003:  20%|█▉        | 61/312 [06:07<25:30,  6.10s/it]
Training Epoch: 19/100, batch 122/625 completed (loss: 0.0004:  20%|█▉        | 61/312 [06:09<25:32,  6.10s/it]
Training Epoch: 19/100, batch 122/625 completed (loss: 0.0002:  20%|█▉        | 61/312 [06:09<25:31,  6.10s/it]
Training Epoch: 19/100, batch 122/625 completed (loss: 0.0003:  20%|█▉        | 61/312 [06:09<25:31,  6.10s/it]
Training Epoch: 19/100, batch 122/625 completed (loss: 0.0006:  20%|█▉        | 61/312 [06:09<25:30,  6.10s/it]
Training Epoch: 19/100, batch 122/625 completed (loss: 0.0004:  20%|█▉        | 62/312 [06:12<25:03,  6.01s/it]
Training Epoch: 19/100, batch 122/625 completed (loss: 0.0002:  20%|█▉        | 62/312 [06:12<25:04,  6.02s/it]
Training Epoch: 19/100, batch 122/625 completed (loss: 0.0003:  20%|█▉        | 62/312 [06:12<25:03,  6.02s/it]
Training Epoch: 19/100, batch 122/625 completed (loss: 0.0006:  20%|█▉        | 62/312 [06:12<25:04,  6.02s/it]
Training Epoch: 19/100, batch 123/625 completed (loss: 0.0003:  20%|█▉        | 62/312 [06:12<25:03,  6.02s/it]
Training Epoch: 19/100, batch 123/625 completed (loss: 0.0003:  20%|█▉        | 62/312 [06:12<25:03,  6.01s/it]
Training Epoch: 19/100, batch 123/625 completed (loss: 0.0013:  20%|█▉        | 62/312 [06:12<25:04,  6.02s/it]
Training Epoch: 19/100, batch 123/625 completed (loss: 0.0005:  20%|█▉        | 62/312 [06:12<25:04,  6.02s/it]
Training Epoch: 19/100, batch 124/625 completed (loss: 0.0001:  20%|█▉        | 62/312 [06:16<25:03,  6.01s/it]
Training Epoch: 19/100, batch 124/625 completed (loss: 0.0124:  20%|█▉        | 62/312 [06:16<25:03,  6.02s/it]
Training Epoch: 19/100, batch 124/625 completed (loss: 0.0002:  20%|█▉        | 62/312 [06:16<25:04,  6.02s/it]
Training Epoch: 19/100, batch 124/625 completed (loss: 0.0002:  20%|█▉        | 62/312 [06:16<25:04,  6.02s/it]
Training Epoch: 19/100, batch 124/625 completed (loss: 0.0001:  20%|██        | 63/312 [06:19<25:29,  6.14s/it]
Training Epoch: 19/100, batch 124/625 completed (loss: 0.0124:  20%|██        | 63/312 [06:19<25:29,  6.14s/it]
Training Epoch: 19/100, batch 124/625 completed (loss: 0.0002:  20%|██        | 63/312 [06:19<25:29,  6.14s/it]
Training Epoch: 19/100, batch 124/625 completed (loss: 0.0002:  20%|██        | 63/312 [06:19<25:29,  6.14s/it]
Training Epoch: 19/100, batch 125/625 completed (loss: 0.0052:  20%|██        | 63/312 [06:19<25:29,  6.14s/it]
Training Epoch: 19/100, batch 125/625 completed (loss: 0.0002:  20%|██        | 63/312 [06:19<25:29,  6.14s/it]
Training Epoch: 19/100, batch 125/625 completed (loss: 0.0005:  20%|██        | 63/312 [06:19<25:29,  6.14s/it]
Training Epoch: 19/100, batch 125/625 completed (loss: 0.0004:  20%|██        | 63/312 [06:19<25:29,  6.14s/it]
Training Epoch: 19/100, batch 126/625 completed (loss: 0.0:  20%|██        | 63/312 [06:22<25:29,  6.14s/it]   
Training Epoch: 19/100, batch 126/625 completed (loss: 0.004:  20%|██        | 63/312 [06:22<25:29,  6.14s/it] 
Training Epoch: 19/100, batch 126/625 completed (loss: 0.0003:  20%|██        | 63/312 [06:22<25:29,  6.14s/it]
Training Epoch: 19/100, batch 126/625 completed (loss: 0.0059:  20%|██        | 63/312 [06:22<25:29,  6.14s/it]
Training Epoch: 19/100, batch 126/625 completed (loss: 0.004:  21%|██        | 64/312 [06:25<25:02,  6.06s/it]
Training Epoch: 19/100, batch 126/625 completed (loss: 0.0:  21%|██        | 64/312 [06:25<25:02,  6.06s/it]
Training Epoch: 19/100, batch 126/625 completed (loss: 0.0003:  21%|██        | 64/312 [06:25<25:03,  6.06s/it]
Training Epoch: 19/100, batch 126/625 completed (loss: 0.0059:  21%|██        | 64/312 [06:25<25:03,  6.06s/it]
Training Epoch: 19/100, batch 127/625 completed (loss: 0.0011:  21%|██        | 64/312 [06:25<25:02,  6.06s/it]
Training Epoch: 19/100, batch 127/625 completed (loss: 0.0011:  21%|██        | 64/312 [06:25<25:03,  6.06s/it]
Training Epoch: 19/100, batch 127/625 completed (loss: 0.0004:  21%|██        | 64/312 [06:25<25:02,  6.06s/it]
Training Epoch: 19/100, batch 127/625 completed (loss: 0.0055:  21%|██        | 64/312 [06:25<25:03,  6.06s/it]
Training Epoch: 19/100, batch 128/625 completed (loss: 0.0131:  21%|██        | 64/312 [06:28<25:02,  6.06s/it]
Training Epoch: 19/100, batch 128/625 completed (loss: 0.0004:  21%|██        | 64/312 [06:28<25:03,  6.06s/it]
Training Epoch: 19/100, batch 128/625 completed (loss: 0.0013:  21%|██        | 64/312 [06:28<25:02,  6.06s/it]
Training Epoch: 19/100, batch 128/625 completed (loss: 0.0034:  21%|██        | 64/312 [06:28<25:03,  6.06s/it]
Training Epoch: 19/100, batch 128/625 completed (loss: 0.0013:  21%|██        | 65/312 [06:31<24:53,  6.05s/it]
Training Epoch: 19/100, batch 128/625 completed (loss: 0.0131:  21%|██        | 65/312 [06:31<24:53,  6.05s/it]
Training Epoch: 19/100, batch 128/625 completed (loss: 0.0034:  21%|██        | 65/312 [06:31<24:53,  6.05s/it]
Training Epoch: 19/100, batch 128/625 completed (loss: 0.0004:  21%|██        | 65/312 [06:31<24:53,  6.05s/it]
Training Epoch: 19/100, batch 129/625 completed (loss: 0.005:  21%|██        | 65/312 [06:31<24:53,  6.05s/it] 
Training Epoch: 19/100, batch 129/625 completed (loss: 0.0005:  21%|██        | 65/312 [06:31<24:53,  6.05s/it]
Training Epoch: 19/100, batch 129/625 completed (loss: 0.0002:  21%|██        | 65/312 [06:31<24:53,  6.05s/it]
Training Epoch: 19/100, batch 129/625 completed (loss: 0.0:  21%|██        | 65/312 [06:31<24:53,  6.05s/it]   
Training Epoch: 19/100, batch 130/625 completed (loss: 0.0016:  21%|██        | 65/312 [06:34<24:53,  6.05s/it]
Training Epoch: 19/100, batch 130/625 completed (loss: 0.0033:  21%|██        | 65/312 [06:34<24:53,  6.05s/it]
Training Epoch: 19/100, batch 130/625 completed (loss: 0.0001:  21%|██        | 65/312 [06:34<24:53,  6.05s/it]
Training Epoch: 19/100, batch 130/625 completed (loss: 0.0003:  21%|██        | 65/312 [06:34<24:53,  6.05s/it]
Training Epoch: 19/100, batch 130/625 completed (loss: 0.0016:  21%|██        | 66/312 [06:37<24:56,  6.08s/it]
Training Epoch: 19/100, batch 130/625 completed (loss: 0.0033:  21%|██        | 66/312 [06:37<24:56,  6.08s/it]
Training Epoch: 19/100, batch 130/625 completed (loss: 0.0001:  21%|██        | 66/312 [06:37<24:56,  6.08s/it]
Training Epoch: 19/100, batch 130/625 completed (loss: 0.0003:  21%|██        | 66/312 [06:37<24:56,  6.09s/it]
Training Epoch: 19/100, batch 131/625 completed (loss: 0.0004:  21%|██        | 66/312 [06:37<24:56,  6.08s/it]
Training Epoch: 19/100, batch 131/625 completed (loss: 0.0013:  21%|██        | 66/312 [06:37<24:56,  6.08s/it]
Training Epoch: 19/100, batch 131/625 completed (loss: 0.0002:  21%|██        | 66/312 [06:37<24:56,  6.08s/it]
Training Epoch: 19/100, batch 131/625 completed (loss: 0.0027:  21%|██        | 66/312 [06:37<24:56,  6.09s/it]
Training Epoch: 19/100, batch 132/625 completed (loss: 0.0016:  21%|██        | 66/312 [06:40<24:56,  6.08s/it]
Training Epoch: 19/100, batch 132/625 completed (loss: 0.0005:  21%|██        | 66/312 [06:40<24:56,  6.08s/it]
Training Epoch: 19/100, batch 132/625 completed (loss: 0.0024:  21%|██        | 66/312 [06:40<24:56,  6.08s/it]
Training Epoch: 19/100, batch 132/625 completed (loss: 0.0002:  21%|██        | 66/312 [06:40<24:56,  6.09s/it]
Training Epoch: 19/100, batch 132/625 completed (loss: 0.0005:  21%|██▏       | 67/312 [06:43<24:30,  6.00s/it]
Training Epoch: 19/100, batch 132/625 completed (loss: 0.0002:  21%|██▏       | 67/312 [06:43<24:30,  6.00s/it]
Training Epoch: 19/100, batch 132/625 completed (loss: 0.0016:  21%|██▏       | 67/312 [06:43<24:30,  6.00s/it]
Training Epoch: 19/100, batch 132/625 completed (loss: 0.0024:  21%|██▏       | 67/312 [06:43<24:31,  6.00s/it]
Training Epoch: 19/100, batch 133/625 completed (loss: 0.0017:  21%|██▏       | 67/312 [06:43<24:30,  6.00s/it]
Training Epoch: 19/100, batch 133/625 completed (loss: 0.026:  21%|██▏       | 67/312 [06:43<24:31,  6.00s/it] 
Training Epoch: 19/100, batch 133/625 completed (loss: 0.0004:  21%|██▏       | 67/312 [06:43<24:30,  6.00s/it]
Training Epoch: 19/100, batch 133/625 completed (loss: 0.0016:  21%|██▏       | 67/312 [06:43<24:30,  6.00s/it]
Training Epoch: 19/100, batch 134/625 completed (loss: 0.0002:  21%|██▏       | 67/312 [06:46<24:30,  6.00s/it]
Training Epoch: 19/100, batch 134/625 completed (loss: 0.0013:  21%|██▏       | 67/312 [06:46<24:31,  6.00s/it]
Training Epoch: 19/100, batch 134/625 completed (loss: 0.0025:  21%|██▏       | 67/312 [06:46<24:30,  6.00s/it]
Training Epoch: 19/100, batch 134/625 completed (loss: 0.0002:  21%|██▏       | 67/312 [06:46<24:30,  6.00s/it]
Training Epoch: 19/100, batch 134/625 completed (loss: 0.0025:  22%|██▏       | 68/312 [06:49<24:44,  6.08s/it]
Training Epoch: 19/100, batch 134/625 completed (loss: 0.0013:  22%|██▏       | 68/312 [06:49<24:44,  6.08s/it]
Training Epoch: 19/100, batch 134/625 completed (loss: 0.0002:  22%|██▏       | 68/312 [06:49<24:44,  6.08s/it]
Training Epoch: 19/100, batch 134/625 completed (loss: 0.0002:  22%|██▏       | 68/312 [06:49<24:44,  6.09s/it]
Training Epoch: 19/100, batch 135/625 completed (loss: 0.0077:  22%|██▏       | 68/312 [06:49<24:44,  6.08s/it]
Training Epoch: 19/100, batch 135/625 completed (loss: 0.002:  22%|██▏       | 68/312 [06:49<24:44,  6.08s/it] 
Training Epoch: 19/100, batch 135/625 completed (loss: 0.0009:  22%|██▏       | 68/312 [06:49<24:44,  6.08s/it]
Training Epoch: 19/100, batch 135/625 completed (loss: 0.0002:  22%|██▏       | 68/312 [06:49<24:44,  6.09s/it]
Training Epoch: 19/100, batch 136/625 completed (loss: 0.0018:  22%|██▏       | 68/312 [06:52<24:44,  6.08s/it]
Training Epoch: 19/100, batch 136/625 completed (loss: 0.0043:  22%|██▏       | 68/312 [06:52<24:44,  6.08s/it]
Training Epoch: 19/100, batch 136/625 completed (loss: 0.0064:  22%|██▏       | 68/312 [06:52<24:44,  6.08s/it]
Training Epoch: 19/100, batch 136/625 completed (loss: 0.0001:  22%|██▏       | 68/312 [06:52<24:44,  6.09s/it]
Training Epoch: 19/100, batch 136/625 completed (loss: 0.0064:  22%|██▏       | 69/312 [06:55<24:20,  6.01s/it]
Training Epoch: 19/100, batch 136/625 completed (loss: 0.0018:  22%|██▏       | 69/312 [06:55<24:20,  6.01s/it]
Training Epoch: 19/100, batch 136/625 completed (loss: 0.0001:  22%|██▏       | 69/312 [06:55<24:20,  6.01s/it]
Training Epoch: 19/100, batch 136/625 completed (loss: 0.0043:  22%|██▏       | 69/312 [06:55<24:21,  6.01s/it]
Training Epoch: 19/100, batch 137/625 completed (loss: 0.0003:  22%|██▏       | 69/312 [06:55<24:21,  6.01s/it]
Training Epoch: 19/100, batch 137/625 completed (loss: 0.003:  22%|██▏       | 69/312 [06:55<24:20,  6.01s/it] 
Training Epoch: 19/100, batch 137/625 completed (loss: 0.005:  22%|██▏       | 69/312 [06:55<24:20,  6.01s/it] 
Training Epoch: 19/100, batch 137/625 completed (loss: 0.0003:  22%|██▏       | 69/312 [06:55<24:20,  6.01s/it]
Training Epoch: 19/100, batch 138/625 completed (loss: 0.0523:  22%|██▏       | 69/312 [06:58<24:21,  6.01s/it]
Training Epoch: 19/100, batch 138/625 completed (loss: 0.0156:  22%|██▏       | 69/312 [06:58<24:20,  6.01s/it]
Training Epoch: 19/100, batch 138/625 completed (loss: 0.0006:  22%|██▏       | 69/312 [06:58<24:20,  6.01s/it]
Training Epoch: 19/100, batch 138/625 completed (loss: 0.0005:  22%|██▏       | 69/312 [06:58<24:20,  6.01s/it]
Training Epoch: 19/100, batch 138/625 completed (loss: 0.0005:  22%|██▏       | 70/312 [07:01<24:00,  5.95s/it]
Training Epoch: 19/100, batch 138/625 completed (loss: 0.0523:  22%|██▏       | 70/312 [07:01<24:00,  5.95s/it]
Training Epoch: 19/100, batch 138/625 completed (loss: 0.0156:  22%|██▏       | 70/312 [07:00<24:00,  5.95s/it]
Training Epoch: 19/100, batch 138/625 completed (loss: 0.0006:  22%|██▏       | 70/312 [07:01<24:00,  5.95s/it]
Training Epoch: 19/100, batch 139/625 completed (loss: 0.0014:  22%|██▏       | 70/312 [07:01<24:00,  5.95s/it]
Training Epoch: 19/100, batch 139/625 completed (loss: 0.0001:  22%|██▏       | 70/312 [07:01<24:00,  5.95s/it]
Training Epoch: 19/100, batch 139/625 completed (loss: 0.0004:  22%|██▏       | 70/312 [07:01<24:00,  5.95s/it]
Training Epoch: 19/100, batch 139/625 completed (loss: 0.0032:  22%|██▏       | 70/312 [07:01<24:00,  5.95s/it]
Training Epoch: 19/100, batch 140/625 completed (loss: 0.0004:  22%|██▏       | 70/312 [07:04<24:00,  5.95s/it]
Training Epoch: 19/100, batch 140/625 completed (loss: 0.0067:  22%|██▏       | 70/312 [07:04<24:00,  5.95s/it]
Training Epoch: 19/100, batch 140/625 completed (loss: 0.0011:  22%|██▏       | 70/312 [07:04<24:00,  5.95s/it]
Training Epoch: 19/100, batch 140/625 completed (loss: 0.0003:  22%|██▏       | 70/312 [07:04<24:00,  5.95s/it]
Training Epoch: 19/100, batch 140/625 completed (loss: 0.0003:  23%|██▎       | 71/312 [07:07<24:21,  6.07s/it]
Training Epoch: 19/100, batch 140/625 completed (loss: 0.0067:  23%|██▎       | 71/312 [07:07<24:22,  6.07s/it]
Training Epoch: 19/100, batch 140/625 completed (loss: 0.0004:  23%|██▎       | 71/312 [07:07<24:22,  6.07s/it]
Training Epoch: 19/100, batch 140/625 completed (loss: 0.0011:  23%|██▎       | 71/312 [07:07<24:22,  6.07s/it]
Training Epoch: 19/100, batch 141/625 completed (loss: 0.001:  23%|██▎       | 71/312 [07:07<24:22,  6.07s/it] 
Training Epoch: 19/100, batch 141/625 completed (loss: 0.0006:  23%|██▎       | 71/312 [07:07<24:22,  6.07s/it]
Training Epoch: 19/100, batch 141/625 completed (loss: 0.0034:  23%|██▎       | 71/312 [07:07<24:22,  6.07s/it]
Training Epoch: 19/100, batch 141/625 completed (loss: 0.0006:  23%|██▎       | 71/312 [07:07<24:21,  6.07s/it]
Training Epoch: 19/100, batch 142/625 completed (loss: 0.0012:  23%|██▎       | 71/312 [07:10<24:22,  6.07s/it]
Training Epoch: 19/100, batch 142/625 completed (loss: 0.0188:  23%|██▎       | 71/312 [07:10<24:22,  6.07s/it]
Training Epoch: 19/100, batch 142/625 completed (loss: 0.0022:  23%|██▎       | 71/312 [07:10<24:22,  6.07s/it]
Training Epoch: 19/100, batch 142/625 completed (loss: 0.0002:  23%|██▎       | 71/312 [07:10<24:21,  6.07s/it]
Training Epoch: 19/100, batch 142/625 completed (loss: 0.0002:  23%|██▎       | 72/312 [07:13<23:59,  6.00s/it]
Training Epoch: 19/100, batch 142/625 completed (loss: 0.0012:  23%|██▎       | 72/312 [07:13<23:59,  6.00s/it]
Training Epoch: 19/100, batch 142/625 completed (loss: 0.0188:  23%|██▎       | 72/312 [07:13<24:00,  6.00s/it]
Training Epoch: 19/100, batch 142/625 completed (loss: 0.0022:  23%|██▎       | 72/312 [07:13<24:00,  6.00s/it]
Training Epoch: 19/100, batch 143/625 completed (loss: 0.0026:  23%|██▎       | 72/312 [07:13<24:00,  6.00s/it]
Training Epoch: 19/100, batch 143/625 completed (loss: 0.0219:  23%|██▎       | 72/312 [07:13<24:00,  6.00s/it]
Training Epoch: 19/100, batch 143/625 completed (loss: 0.0023:  23%|██▎       | 72/312 [07:13<23:59,  6.00s/it]
Training Epoch: 19/100, batch 143/625 completed (loss: 0.0007:  23%|██▎       | 72/312 [07:13<23:59,  6.00s/it]
Training Epoch: 19/100, batch 144/625 completed (loss: 0.343:  23%|██▎       | 72/312 [07:16<24:00,  6.00s/it] 
Training Epoch: 19/100, batch 144/625 completed (loss: 0.0126:  23%|██▎       | 72/312 [07:16<23:59,  6.00s/it]
Training Epoch: 19/100, batch 144/625 completed (loss: 0.0241:  23%|██▎       | 72/312 [07:16<24:00,  6.00s/it]
Training Epoch: 19/100, batch 144/625 completed (loss: 0.0001:  23%|██▎       | 72/312 [07:16<23:59,  6.00s/it]
Training Epoch: 19/100, batch 144/625 completed (loss: 0.0126:  23%|██▎       | 73/312 [07:19<24:07,  6.06s/it]
Training Epoch: 19/100, batch 144/625 completed (loss: 0.0241:  23%|██▎       | 73/312 [07:19<24:07,  6.06s/it]
Training Epoch: 19/100, batch 144/625 completed (loss: 0.0001:  23%|██▎       | 73/312 [07:19<24:07,  6.06s/it]
Training Epoch: 19/100, batch 144/625 completed (loss: 0.343:  23%|██▎       | 73/312 [07:19<24:07,  6.06s/it]
Training Epoch: 19/100, batch 145/625 completed (loss: 0.0028:  23%|██▎       | 73/312 [07:19<24:07,  6.06s/it]
Training Epoch: 19/100, batch 145/625 completed (loss: 0.0013:  23%|██▎       | 73/312 [07:19<24:07,  6.06s/it]
Training Epoch: 19/100, batch 145/625 completed (loss: 0.0021:  23%|██▎       | 73/312 [07:19<24:07,  6.06s/it]
Training Epoch: 19/100, batch 145/625 completed (loss: 0.0158:  23%|██▎       | 73/312 [07:19<24:07,  6.06s/it]
Training Epoch: 19/100, batch 146/625 completed (loss: 0.0003:  23%|██▎       | 73/312 [07:22<24:07,  6.06s/it]
Training Epoch: 19/100, batch 146/625 completed (loss: 0.0295:  23%|██▎       | 73/312 [07:22<24:07,  6.06s/it]
Training Epoch: 19/100, batch 146/625 completed (loss: 0.0022:  23%|██▎       | 73/312 [07:22<24:07,  6.06s/it]
Training Epoch: 19/100, batch 146/625 completed (loss: 0.0001:  23%|██▎       | 73/312 [07:22<24:07,  6.06s/it]
Training Epoch: 19/100, batch 146/625 completed (loss: 0.0022:  24%|██▎       | 74/312 [07:25<24:06,  6.08s/it]
Training Epoch: 19/100, batch 146/625 completed (loss: 0.0003:  24%|██▎       | 74/312 [07:25<24:07,  6.08s/it]
Training Epoch: 19/100, batch 146/625 completed (loss: 0.0295:  24%|██▎       | 74/312 [07:25<24:07,  6.08s/it]
Training Epoch: 19/100, batch 146/625 completed (loss: 0.0001:  24%|██▎       | 74/312 [07:25<24:08,  6.08s/it]
Training Epoch: 19/100, batch 147/625 completed (loss: 0.0186:  24%|██▎       | 74/312 [07:25<24:07,  6.08s/it]
Training Epoch: 19/100, batch 147/625 completed (loss: 0.0179:  24%|██▎       | 74/312 [07:25<24:07,  6.08s/it]
Training Epoch: 19/100, batch 147/625 completed (loss: 0.0209:  24%|██▎       | 74/312 [07:25<24:06,  6.08s/it]
Training Epoch: 19/100, batch 147/625 completed (loss: 0.038:  24%|██▎       | 74/312 [07:25<24:08,  6.08s/it] 
Training Epoch: 19/100, batch 148/625 completed (loss: 0.0145:  24%|██▎       | 74/312 [07:28<24:07,  6.08s/it]
Training Epoch: 19/100, batch 148/625 completed (loss: 0.0138:  24%|██▎       | 74/312 [07:28<24:07,  6.08s/it]
Training Epoch: 19/100, batch 148/625 completed (loss: 0.0037:  24%|██▎       | 74/312 [07:28<24:06,  6.08s/it]
Training Epoch: 19/100, batch 148/625 completed (loss: 0.0227:  24%|██▎       | 74/312 [07:28<24:08,  6.08s/it]
Training Epoch: 19/100, batch 148/625 completed (loss: 0.0227:  24%|██▍       | 75/312 [07:31<23:42,  6.00s/it]
Training Epoch: 19/100, batch 148/625 completed (loss: 0.0037:  24%|██▍       | 75/312 [07:31<23:42,  6.00s/it]
Training Epoch: 19/100, batch 148/625 completed (loss: 0.0145:  24%|██▍       | 75/312 [07:31<23:42,  6.00s/it]
Training Epoch: 19/100, batch 148/625 completed (loss: 0.0138:  24%|██▍       | 75/312 [07:31<23:42,  6.00s/it]
Training Epoch: 19/100, batch 149/625 completed (loss: 0.0:  24%|██▍       | 75/312 [07:31<23:42,  6.00s/it]   
Training Epoch: 19/100, batch 149/625 completed (loss: 0.002:  24%|██▍       | 75/312 [07:31<23:42,  6.00s/it] 
Training Epoch: 19/100, batch 149/625 completed (loss: 0.0054:  24%|██▍       | 75/312 [07:31<23:42,  6.00s/it]
Training Epoch: 19/100, batch 149/625 completed (loss: 0.0308:  24%|██▍       | 75/312 [07:31<23:42,  6.00s/it]
Training Epoch: 19/100, batch 150/625 completed (loss: 0.0003:  24%|██▍       | 75/312 [07:34<23:42,  6.00s/it]
Training Epoch: 19/100, batch 150/625 completed (loss: 0.004:  24%|██▍       | 75/312 [07:34<23:42,  6.00s/it]
Training Epoch: 19/100, batch 150/625 completed (loss: 0.0002:  24%|██▍       | 75/312 [07:34<23:42,  6.00s/it]
Training Epoch: 19/100, batch 150/625 completed (loss: 0.0093:  24%|██▍       | 75/312 [07:34<23:42,  6.00s/it]
Training Epoch: 19/100, batch 150/625 completed (loss: 0.0002:  24%|██▍       | 76/312 [07:37<24:00,  6.10s/it]
Training Epoch: 19/100, batch 150/625 completed (loss: 0.004:  24%|██▍       | 76/312 [07:37<24:00,  6.10s/it]
Training Epoch: 19/100, batch 150/625 completed (loss: 0.0093:  24%|██▍       | 76/312 [07:37<24:00,  6.10s/it]
Training Epoch: 19/100, batch 150/625 completed (loss: 0.0003:  24%|██▍       | 76/312 [07:37<24:00,  6.10s/it]
Training Epoch: 19/100, batch 151/625 completed (loss: 0.0084:  24%|██▍       | 76/312 [07:37<24:00,  6.10s/it]
Training Epoch: 19/100, batch 151/625 completed (loss: 0.0011:  24%|██▍       | 76/312 [07:37<24:00,  6.10s/it]
Training Epoch: 19/100, batch 151/625 completed (loss: 0.0049:  24%|██▍       | 76/312 [07:37<24:00,  6.10s/it]
Training Epoch: 19/100, batch 151/625 completed (loss: 0.0006:  24%|██▍       | 76/312 [07:37<24:00,  6.10s/it]
Training Epoch: 19/100, batch 152/625 completed (loss: 0.0011:  24%|██▍       | 76/312 [07:40<24:00,  6.10s/it]
Training Epoch: 19/100, batch 152/625 completed (loss: 0.0001:  24%|██▍       | 76/312 [07:40<24:00,  6.10s/it]
Training Epoch: 19/100, batch 152/625 completed (loss: 0.0015:  24%|██▍       | 76/312 [07:40<24:00,  6.10s/it]
Training Epoch: 19/100, batch 152/625 completed (loss: 0.0022:  24%|██▍       | 76/312 [07:40<24:00,  6.10s/it]
Training Epoch: 19/100, batch 152/625 completed (loss: 0.0011:  25%|██▍       | 77/312 [07:43<23:35,  6.02s/it]
Training Epoch: 19/100, batch 152/625 completed (loss: 0.0022:  25%|██▍       | 77/312 [07:43<23:35,  6.02s/it]
Training Epoch: 19/100, batch 152/625 completed (loss: 0.0015:  25%|██▍       | 77/312 [07:43<23:35,  6.02s/it]
Training Epoch: 19/100, batch 152/625 completed (loss: 0.0001:  25%|██▍       | 77/312 [07:43<23:36,  6.03s/it]
Training Epoch: 19/100, batch 153/625 completed (loss: 0.0046:  25%|██▍       | 77/312 [07:43<23:35,  6.02s/it]
Training Epoch: 19/100, batch 153/625 completed (loss: 0.0008:  25%|██▍       | 77/312 [07:43<23:35,  6.02s/it]
Training Epoch: 19/100, batch 153/625 completed (loss: 0.0031:  25%|██▍       | 77/312 [07:43<23:36,  6.03s/it]
Training Epoch: 19/100, batch 153/625 completed (loss: 0.0012:  25%|██▍       | 77/312 [07:43<23:35,  6.02s/it]
Training Epoch: 19/100, batch 154/625 completed (loss: 0.0065:  25%|██▍       | 77/312 [07:46<23:35,  6.02s/it]
Training Epoch: 19/100, batch 154/625 completed (loss: 0.0001:  25%|██▍       | 77/312 [07:46<23:35,  6.02s/it]
Training Epoch: 19/100, batch 154/625 completed (loss: 0.0033:  25%|██▍       | 77/312 [07:46<23:36,  6.03s/it]
Training Epoch: 19/100, batch 154/625 completed (loss: 0.0007:  25%|██▍       | 77/312 [07:46<23:35,  6.02s/it]
Training Epoch: 19/100, batch 154/625 completed (loss: 0.0007:  25%|██▌       | 78/312 [07:49<23:28,  6.02s/it]
Training Epoch: 19/100, batch 154/625 completed (loss: 0.0033:  25%|██▌       | 78/312 [07:49<23:28,  6.02s/it]
Training Epoch: 19/100, batch 154/625 completed (loss: 0.0001:  25%|██▌       | 78/312 [07:49<23:29,  6.02s/it]
Training Epoch: 19/100, batch 154/625 completed (loss: 0.0065:  25%|██▌       | 78/312 [07:49<23:29,  6.02s/it]
Training Epoch: 19/100, batch 155/625 completed (loss: 0.007:  25%|██▌       | 78/312 [07:49<23:29,  6.02s/it] 
Training Epoch: 19/100, batch 155/625 completed (loss: 0.0005:  25%|██▌       | 78/312 [07:49<23:29,  6.02s/it]
Training Epoch: 19/100, batch 155/625 completed (loss: 0.0019:  25%|██▌       | 78/312 [07:49<23:28,  6.02s/it]
Training Epoch: 19/100, batch 155/625 completed (loss: 0.0016:  25%|██▌       | 78/312 [07:49<23:28,  6.02s/it]
Training Epoch: 19/100, batch 156/625 completed (loss: 0.0002:  25%|██▌       | 78/312 [07:52<23:28,  6.02s/it]
Training Epoch: 19/100, batch 156/625 completed (loss: 0.0006:  25%|██▌       | 78/312 [07:52<23:29,  6.02s/it]
Training Epoch: 19/100, batch 156/625 completed (loss: 0.0002:  25%|██▌       | 78/312 [07:52<23:29,  6.02s/it]
Training Epoch: 19/100, batch 156/625 completed (loss: 0.0002:  25%|██▌       | 78/312 [07:52<23:28,  6.02s/it]
Training Epoch: 19/100, batch 156/625 completed (loss: 0.0002:  25%|██▌       | 79/312 [07:55<23:41,  6.10s/it]
Training Epoch: 19/100, batch 156/625 completed (loss: 0.0002:  25%|██▌       | 79/312 [07:55<23:41,  6.10s/it]
Training Epoch: 19/100, batch 156/625 completed (loss: 0.0006:  25%|██▌       | 79/312 [07:55<23:42,  6.10s/it]
Training Epoch: 19/100, batch 156/625 completed (loss: 0.0002:  25%|██▌       | 79/312 [07:55<23:42,  6.11s/it]
Training Epoch: 19/100, batch 157/625 completed (loss: 0.0071:  25%|██▌       | 79/312 [07:55<23:41,  6.10s/it]
Training Epoch: 19/100, batch 157/625 completed (loss: 0.0034:  25%|██▌       | 79/312 [07:55<23:41,  6.10s/it]
Training Epoch: 19/100, batch 157/625 completed (loss: 0.0083:  25%|██▌       | 79/312 [07:55<23:42,  6.10s/it]
Training Epoch: 19/100, batch 157/625 completed (loss: 0.0003:  25%|██▌       | 79/312 [07:55<23:42,  6.11s/it]
Training Epoch: 19/100, batch 158/625 completed (loss: 0.0029:  25%|██▌       | 79/312 [07:58<23:42,  6.10s/it]
Training Epoch: 19/100, batch 158/625 completed (loss: 0.0004:  25%|██▌       | 79/312 [07:58<23:41,  6.10s/it]
Training Epoch: 19/100, batch 158/625 completed (loss: 0.0008:  25%|██▌       | 79/312 [07:58<23:41,  6.10s/it]
Training Epoch: 19/100, batch 158/625 completed (loss: 0.0002:  25%|██▌       | 79/312 [07:58<23:42,  6.11s/it]
Training Epoch: 19/100, batch 158/625 completed (loss: 0.0004:  26%|██▌       | 80/312 [08:01<23:16,  6.02s/it]
Training Epoch: 19/100, batch 158/625 completed (loss: 0.0002:  26%|██▌       | 80/312 [08:01<23:16,  6.02s/it]
Training Epoch: 19/100, batch 158/625 completed (loss: 0.0008:  26%|██▌       | 80/312 [08:01<23:16,  6.02s/it]
Training Epoch: 19/100, batch 158/625 completed (loss: 0.0029:  26%|██▌       | 80/312 [08:01<23:16,  6.02s/it]
Training Epoch: 19/100, batch 159/625 completed (loss: 0.0008:  26%|██▌       | 80/312 [08:01<23:16,  6.02s/it]
Training Epoch: 19/100, batch 159/625 completed (loss: 0.0012:  26%|██▌       | 80/312 [08:01<23:16,  6.02s/it]
Training Epoch: 19/100, batch 159/625 completed (loss: 0.0007:  26%|██▌       | 80/312 [08:01<23:16,  6.02s/it]
Training Epoch: 19/100, batch 159/625 completed (loss: 0.0009:  26%|██▌       | 80/312 [08:01<23:16,  6.02s/it]
Training Epoch: 19/100, batch 160/625 completed (loss: 0.0009:  26%|██▌       | 80/312 [08:04<23:16,  6.02s/it]
Training Epoch: 19/100, batch 160/625 completed (loss: 0.004:  26%|██▌       | 80/312 [08:04<23:16,  6.02s/it] 
Training Epoch: 19/100, batch 160/625 completed (loss: 0.003:  26%|██▌       | 80/312 [08:04<23:16,  6.02s/it] 
Training Epoch: 19/100, batch 160/625 completed (loss: 0.0013:  26%|██▌       | 80/312 [08:04<23:16,  6.02s/it]
Training Epoch: 19/100, batch 160/625 completed (loss: 0.0013:  26%|██▌       | 81/312 [08:07<23:21,  6.07s/it]
Training Epoch: 19/100, batch 160/625 completed (loss: 0.003:  26%|██▌       | 81/312 [08:07<23:21,  6.07s/it]
Training Epoch: 19/100, batch 160/625 completed (loss: 0.0009:  26%|██▌       | 81/312 [08:07<23:21,  6.07s/it]
Training Epoch: 19/100, batch 160/625 completed (loss: 0.004:  26%|██▌       | 81/312 [08:07<23:22,  6.07s/it]
Training Epoch: 19/100, batch 161/625 completed (loss: 0.0:  26%|██▌       | 81/312 [08:07<23:22,  6.07s/it]  
Training Epoch: 19/100, batch 161/625 completed (loss: 0.0023:  26%|██▌       | 81/312 [08:07<23:21,  6.07s/it]
Training Epoch: 19/100, batch 161/625 completed (loss: 0.0016:  26%|██▌       | 81/312 [08:07<23:21,  6.07s/it]
Training Epoch: 19/100, batch 161/625 completed (loss: 0.0007:  26%|██▌       | 81/312 [08:07<23:21,  6.07s/it]
Training Epoch: 19/100, batch 162/625 completed (loss: 0.0034:  26%|██▌       | 81/312 [08:11<23:22,  6.07s/it]
Training Epoch: 19/100, batch 162/625 completed (loss: 0.0009:  26%|██▌       | 81/312 [08:11<23:21,  6.07s/it]
Training Epoch: 19/100, batch 162/625 completed (loss: 0.0002:  26%|██▌       | 81/312 [08:11<23:21,  6.07s/it]
Training Epoch: 19/100, batch 162/625 completed (loss: 0.0056:  26%|██▌       | 81/312 [08:11<23:21,  6.07s/it]
Training Epoch: 19/100, batch 162/625 completed (loss: 0.0034:  26%|██▋       | 82/312 [08:13<23:19,  6.08s/it]
Training Epoch: 19/100, batch 162/625 completed (loss: 0.0009:  26%|██▋       | 82/312 [08:13<23:19,  6.09s/it]
Training Epoch: 19/100, batch 162/625 completed (loss: 0.0056:  26%|██▋       | 82/312 [08:13<23:19,  6.09s/it]
Training Epoch: 19/100, batch 162/625 completed (loss: 0.0002:  26%|██▋       | 82/312 [08:13<23:19,  6.09s/it]
Training Epoch: 19/100, batch 163/625 completed (loss: 0.0009:  26%|██▋       | 82/312 [08:13<23:19,  6.08s/it]
Training Epoch: 19/100, batch 163/625 completed (loss: 0.0001:  26%|██▋       | 82/312 [08:13<23:19,  6.09s/it]
Training Epoch: 19/100, batch 163/625 completed (loss: 0.0029:  26%|██▋       | 82/312 [08:13<23:19,  6.09s/it]
Training Epoch: 19/100, batch 163/625 completed (loss: 0.0022:  26%|██▋       | 82/312 [08:14<23:19,  6.09s/it]
Training Epoch: 19/100, batch 164/625 completed (loss: 0.0004:  26%|██▋       | 82/312 [08:16<23:19,  6.09s/it]
Training Epoch: 19/100, batch 164/625 completed (loss: 0.0045:  26%|██▋       | 82/312 [08:16<23:19,  6.09s/it]
Training Epoch: 19/100, batch 164/625 completed (loss: 0.0014:  26%|██▋       | 82/312 [08:16<23:19,  6.08s/it]
Training Epoch: 19/100, batch 164/625 completed (loss: 0.0013:  26%|██▋       | 82/312 [08:16<23:19,  6.09s/it]
Training Epoch: 19/100, batch 164/625 completed (loss: 0.0013:  27%|██▋       | 83/312 [08:19<23:07,  6.06s/it]
Training Epoch: 19/100, batch 164/625 completed (loss: 0.0004:  27%|██▋       | 83/312 [08:19<23:07,  6.06s/it]
Training Epoch: 19/100, batch 164/625 completed (loss: 0.0045:  27%|██▋       | 83/312 [08:19<23:07,  6.06s/it]
Training Epoch: 19/100, batch 164/625 completed (loss: 0.0014:  27%|██▋       | 83/312 [08:19<23:07,  6.06s/it]
Training Epoch: 19/100, batch 165/625 completed (loss: 0.0001:  27%|██▋       | 83/312 [08:19<23:07,  6.06s/it]
Training Epoch: 19/100, batch 165/625 completed (loss: 0.007:  27%|██▋       | 83/312 [08:19<23:07,  6.06s/it] 
Training Epoch: 19/100, batch 165/625 completed (loss: 0.0024:  27%|██▋       | 83/312 [08:19<23:07,  6.06s/it]
Training Epoch: 19/100, batch 165/625 completed (loss: 0.0001:  27%|██▋       | 83/312 [08:20<23:07,  6.06s/it]
Training Epoch: 19/100, batch 166/625 completed (loss: 0.001:  27%|██▋       | 83/312 [08:23<23:07,  6.06s/it]
Training Epoch: 19/100, batch 166/625 completed (loss: 0.0011:  27%|██▋       | 83/312 [08:23<23:07,  6.06s/it]
Training Epoch: 19/100, batch 166/625 completed (loss: 0.0104:  27%|██▋       | 83/312 [08:23<23:07,  6.06s/it]
Training Epoch: 19/100, batch 166/625 completed (loss: 0.0041:  27%|██▋       | 83/312 [08:23<23:07,  6.06s/it]
Training Epoch: 19/100, batch 166/625 completed (loss: 0.0041:  27%|██▋       | 84/312 [08:26<23:18,  6.13s/it]
Training Epoch: 19/100, batch 166/625 completed (loss: 0.0104:  27%|██▋       | 84/312 [08:26<23:18,  6.13s/it]
Training Epoch: 19/100, batch 166/625 completed (loss: 0.001:  27%|██▋       | 84/312 [08:26<23:18,  6.13s/it]
Training Epoch: 19/100, batch 166/625 completed (loss: 0.0011:  27%|██▋       | 84/312 [08:26<23:18,  6.13s/it]
Training Epoch: 19/100, batch 167/625 completed (loss: 0.0013:  27%|██▋       | 84/312 [08:26<23:18,  6.13s/it]
Training Epoch: 19/100, batch 167/625 completed (loss: 0.0007:  27%|██▋       | 84/312 [08:26<23:18,  6.13s/it]
Training Epoch: 19/100, batch 167/625 completed (loss: 0.0102:  27%|██▋       | 84/312 [08:26<23:18,  6.13s/it]
Training Epoch: 19/100, batch 167/625 completed (loss: 0.0003:  27%|██▋       | 84/312 [08:26<23:18,  6.13s/it]
Training Epoch: 19/100, batch 168/625 completed (loss: 0.0001:  27%|██▋       | 84/312 [08:29<23:18,  6.13s/it]
Training Epoch: 19/100, batch 168/625 completed (loss: 0.0003:  27%|██▋       | 84/312 [08:29<23:18,  6.13s/it]
Training Epoch: 19/100, batch 168/625 completed (loss: 0.0004:  27%|██▋       | 84/312 [08:29<23:18,  6.13s/it]
Training Epoch: 19/100, batch 168/625 completed (loss: 0.0003:  27%|██▋       | 84/312 [08:29<23:18,  6.13s/it]
Training Epoch: 19/100, batch 168/625 completed (loss: 0.0001:  27%|██▋       | 85/312 [08:32<22:53,  6.05s/it]
Training Epoch: 19/100, batch 168/625 completed (loss: 0.0003:  27%|██▋       | 85/312 [08:32<22:53,  6.05s/it]
Training Epoch: 19/100, batch 168/625 completed (loss: 0.0004:  27%|██▋       | 85/312 [08:32<22:53,  6.05s/it]
Training Epoch: 19/100, batch 168/625 completed (loss: 0.0003:  27%|██▋       | 85/312 [08:32<22:54,  6.05s/it]
Training Epoch: 19/100, batch 169/625 completed (loss: 0.0052:  27%|██▋       | 85/312 [08:32<22:53,  6.05s/it]
Training Epoch: 19/100, batch 169/625 completed (loss: 0.0605:  27%|██▋       | 85/312 [08:32<22:53,  6.05s/it]
Training Epoch: 19/100, batch 169/625 completed (loss: 0.0014:  27%|██▋       | 85/312 [08:32<22:53,  6.05s/it]
Training Epoch: 19/100, batch 169/625 completed (loss: 0.0015:  27%|██▋       | 85/312 [08:32<22:54,  6.05s/it]
Training Epoch: 19/100, batch 170/625 completed (loss: 0.0129:  27%|██▋       | 85/312 [08:35<22:53,  6.05s/it]
Training Epoch: 19/100, batch 170/625 completed (loss: 0.0004:  27%|██▋       | 85/312 [08:35<22:53,  6.05s/it]
Training Epoch: 19/100, batch 170/625 completed (loss: 0.0004:  27%|██▋       | 85/312 [08:35<22:53,  6.05s/it]
Training Epoch: 19/100, batch 170/625 completed (loss: 0.0006:  27%|██▋       | 85/312 [08:35<22:54,  6.05s/it]
Training Epoch: 19/100, batch 170/625 completed (loss: 0.0004:  28%|██▊       | 86/312 [08:38<22:55,  6.09s/it]
Training Epoch: 19/100, batch 170/625 completed (loss: 0.0006:  28%|██▊       | 86/312 [08:38<22:55,  6.09s/it]
Training Epoch: 19/100, batch 170/625 completed (loss: 0.0004:  28%|██▊       | 86/312 [08:38<22:56,  6.09s/it]
Training Epoch: 19/100, batch 170/625 completed (loss: 0.0129:  28%|██▊       | 86/312 [08:38<22:56,  6.09s/it]
Training Epoch: 19/100, batch 171/625 completed (loss: 0.0002:  28%|██▊       | 86/312 [08:38<22:56,  6.09s/it]
Training Epoch: 19/100, batch 171/625 completed (loss: 0.0046:  28%|██▊       | 86/312 [08:38<22:55,  6.09s/it]
Training Epoch: 19/100, batch 171/625 completed (loss: 0.0034:  28%|██▊       | 86/312 [08:38<22:56,  6.09s/it]
Training Epoch: 19/100, batch 171/625 completed (loss: 0.0022:  28%|██▊       | 86/312 [08:38<22:55,  6.09s/it]
Training Epoch: 19/100, batch 172/625 completed (loss: 0.0009:  28%|██▊       | 86/312 [08:41<22:56,  6.09s/it]
Training Epoch: 19/100, batch 172/625 completed (loss: 0.0005:  28%|██▊       | 86/312 [08:41<22:55,  6.09s/it]
Training Epoch: 19/100, batch 172/625 completed (loss: 0.0004:  28%|██▊       | 86/312 [08:41<22:56,  6.09s/it]
Training Epoch: 19/100, batch 172/625 completed (loss: 0.0025:  28%|██▊       | 86/312 [08:41<22:55,  6.09s/it]
Training Epoch: 19/100, batch 172/625 completed (loss: 0.0005:  28%|██▊       | 87/312 [08:44<22:54,  6.11s/it]
Training Epoch: 19/100, batch 172/625 completed (loss: 0.0025:  28%|██▊       | 87/312 [08:44<22:55,  6.11s/it]
Training Epoch: 19/100, batch 172/625 completed (loss: 0.0004:  28%|██▊       | 87/312 [08:44<22:55,  6.11s/it]
Training Epoch: 19/100, batch 172/625 completed (loss: 0.0009:  28%|██▊       | 87/312 [08:44<22:55,  6.11s/it]
Training Epoch: 19/100, batch 173/625 completed (loss: 0.0004:  28%|██▊       | 87/312 [08:44<22:55,  6.11s/it]
Training Epoch: 19/100, batch 173/625 completed (loss: 0.0:  28%|██▊       | 87/312 [08:44<22:55,  6.11s/it]   
Training Epoch: 19/100, batch 173/625 completed (loss: 0.0003:  28%|██▊       | 87/312 [08:44<22:54,  6.11s/it]
Training Epoch: 19/100, batch 173/625 completed (loss: 0.0043:  28%|██▊       | 87/312 [08:44<22:55,  6.11s/it]
Training Epoch: 19/100, batch 174/625 completed (loss: 0.0002:  28%|██▊       | 87/312 [08:47<22:55,  6.11s/it]
Training Epoch: 19/100, batch 174/625 completed (loss: 0.0001:  28%|██▊       | 87/312 [08:47<22:55,  6.11s/it]
Training Epoch: 19/100, batch 174/625 completed (loss: 0.0011:  28%|██▊       | 87/312 [08:47<22:54,  6.11s/it]
Training Epoch: 19/100, batch 174/625 completed (loss: 0.0003:  28%|██▊       | 87/312 [08:47<22:55,  6.11s/it]
Training Epoch: 19/100, batch 174/625 completed (loss: 0.0003:  28%|██▊       | 88/312 [08:50<22:42,  6.08s/it]
Training Epoch: 19/100, batch 174/625 completed (loss: 0.0002:  28%|██▊       | 88/312 [08:50<22:42,  6.08s/it]
Training Epoch: 19/100, batch 174/625 completed (loss: 0.0011:  28%|██▊       | 88/312 [08:50<22:43,  6.09s/it]
Training Epoch: 19/100, batch 174/625 completed (loss: 0.0001:  28%|██▊       | 88/312 [08:50<22:42,  6.08s/it]
Training Epoch: 19/100, batch 175/625 completed (loss: 0.0002:  28%|██▊       | 88/312 [08:50<22:43,  6.09s/it]
Training Epoch: 19/100, batch 175/625 completed (loss: 0.0001:  28%|██▊       | 88/312 [08:50<22:42,  6.08s/it]
Training Epoch: 19/100, batch 175/625 completed (loss: 0.0002:  28%|██▊       | 88/312 [08:50<22:42,  6.08s/it]
Training Epoch: 19/100, batch 175/625 completed (loss: 0.0077:  28%|██▊       | 88/312 [08:50<22:42,  6.08s/it]
Training Epoch: 19/100, batch 176/625 completed (loss: 0.0046:  28%|██▊       | 88/312 [08:53<22:43,  6.09s/it]
Training Epoch: 19/100, batch 176/625 completed (loss: 0.0006:  28%|██▊       | 88/312 [08:53<22:42,  6.08s/it]
Training Epoch: 19/100, batch 176/625 completed (loss: 0.0006:  28%|██▊       | 88/312 [08:53<22:42,  6.08s/it]
Training Epoch: 19/100, batch 176/625 completed (loss: 0.0003:  28%|██▊       | 88/312 [08:53<22:42,  6.08s/it]
Training Epoch: 19/100, batch 176/625 completed (loss: 0.0046:  29%|██▊       | 89/312 [08:56<22:40,  6.10s/it]
Training Epoch: 19/100, batch 176/625 completed (loss: 0.0006:  29%|██▊       | 89/312 [08:56<22:40,  6.10s/it]
Training Epoch: 19/100, batch 176/625 completed (loss: 0.0003:  29%|██▊       | 89/312 [08:56<22:40,  6.10s/it]
Training Epoch: 19/100, batch 176/625 completed (loss: 0.0006:  29%|██▊       | 89/312 [08:56<22:40,  6.10s/it]
Training Epoch: 19/100, batch 177/625 completed (loss: 0.0008:  29%|██▊       | 89/312 [08:56<22:40,  6.10s/it]
Training Epoch: 19/100, batch 177/625 completed (loss: 0.0003:  29%|██▊       | 89/312 [08:56<22:40,  6.10s/it]
Training Epoch: 19/100, batch 177/625 completed (loss: 0.0016:  29%|██▊       | 89/312 [08:56<22:40,  6.10s/it]
Training Epoch: 19/100, batch 177/625 completed (loss: 0.0001:  29%|██▊       | 89/312 [08:56<22:40,  6.10s/it]
Training Epoch: 19/100, batch 178/625 completed (loss: 0.0005:  29%|██▊       | 89/312 [08:59<22:40,  6.10s/it]
Training Epoch: 19/100, batch 178/625 completed (loss: 0.0:  29%|██▊       | 89/312 [08:59<22:40,  6.10s/it]   
Training Epoch: 19/100, batch 178/625 completed (loss: 0.0003:  29%|██▊       | 89/312 [08:59<22:40,  6.10s/it]
Training Epoch: 19/100, batch 178/625 completed (loss: 0.007:  29%|██▊       | 89/312 [08:59<22:40,  6.10s/it] 
Training Epoch: 19/100, batch 178/625 completed (loss: 0.0:  29%|██▉       | 90/312 [09:02<22:29,  6.08s/it]
Training Epoch: 19/100, batch 178/625 completed (loss: 0.0003:  29%|██▉       | 90/312 [09:02<22:28,  6.08s/it]
Training Epoch: 19/100, batch 178/625 completed (loss: 0.007:  29%|██▉       | 90/312 [09:02<22:29,  6.08s/it]
Training Epoch: 19/100, batch 178/625 completed (loss: 0.0005:  29%|██▉       | 90/312 [09:02<22:29,  6.08s/it]
Training Epoch: 19/100, batch 179/625 completed (loss: 0.0002:  29%|██▉       | 90/312 [09:02<22:29,  6.08s/it]
Training Epoch: 19/100, batch 179/625 completed (loss: 0.0001:  29%|██▉       | 90/312 [09:02<22:28,  6.08s/it]
Training Epoch: 19/100, batch 179/625 completed (loss: 0.0353:  29%|██▉       | 90/312 [09:02<22:29,  6.08s/it]
Training Epoch: 19/100, batch 179/625 completed (loss: 0.0003:  29%|██▉       | 90/312 [09:02<22:29,  6.08s/it]
Training Epoch: 19/100, batch 180/625 completed (loss: 0.0006:  29%|██▉       | 90/312 [09:05<22:29,  6.08s/it]
Training Epoch: 19/100, batch 180/625 completed (loss: 0.0122:  29%|██▉       | 90/312 [09:05<22:29,  6.08s/it]
Training Epoch: 19/100, batch 180/625 completed (loss: 0.0016:  29%|██▉       | 90/312 [09:05<22:28,  6.08s/it]
Training Epoch: 19/100, batch 180/625 completed (loss: 0.002:  29%|██▉       | 90/312 [09:05<22:29,  6.08s/it] 
Training Epoch: 19/100, batch 180/625 completed (loss: 0.002:  29%|██▉       | 91/312 [09:08<22:19,  6.06s/it]
Training Epoch: 19/100, batch 180/625 completed (loss: 0.0122:  29%|██▉       | 91/312 [09:08<22:20,  6.06s/it]
Training Epoch: 19/100, batch 180/625 completed (loss: 0.0016:  29%|██▉       | 91/312 [09:08<22:21,  6.07s/it]
Training Epoch: 19/100, batch 180/625 completed (loss: 0.0006:  29%|██▉       | 91/312 [09:08<22:21,  6.07s/it]
Training Epoch: 19/100, batch 181/625 completed (loss: 0.0001:  29%|██▉       | 91/312 [09:08<22:20,  6.06s/it]
Training Epoch: 19/100, batch 181/625 completed (loss: 0.0043:  29%|██▉       | 91/312 [09:08<22:21,  6.07s/it]
Training Epoch: 19/100, batch 181/625 completed (loss: 0.0018:  29%|██▉       | 91/312 [09:08<22:21,  6.07s/it]
Training Epoch: 19/100, batch 181/625 completed (loss: 0.0012:  29%|██▉       | 91/312 [09:08<22:19,  6.06s/it]
Training Epoch: 19/100, batch 182/625 completed (loss: 0.0009:  29%|██▉       | 91/312 [09:11<22:20,  6.06s/it]
Training Epoch: 19/100, batch 182/625 completed (loss: 0.0077:  29%|██▉       | 91/312 [09:11<22:21,  6.07s/it]
Training Epoch: 19/100, batch 182/625 completed (loss: 0.0029:  29%|██▉       | 91/312 [09:11<22:21,  6.07s/it]
Training Epoch: 19/100, batch 182/625 completed (loss: 0.0147:  29%|██▉       | 91/312 [09:11<22:19,  6.06s/it]
Training Epoch: 19/100, batch 182/625 completed (loss: 0.0077:  29%|██▉       | 92/312 [09:14<22:29,  6.14s/it]
Training Epoch: 19/100, batch 182/625 completed (loss: 0.0009:  29%|██▉       | 92/312 [09:14<22:30,  6.14s/it]
Training Epoch: 19/100, batch 182/625 completed (loss: 0.0147:  29%|██▉       | 92/312 [09:15<22:31,  6.14s/it]
Training Epoch: 19/100, batch 182/625 completed (loss: 0.0029:  29%|██▉       | 92/312 [09:14<22:30,  6.14s/it]
Training Epoch: 19/100, batch 183/625 completed (loss: 0.0423:  29%|██▉       | 92/312 [09:15<22:30,  6.14s/it]
Training Epoch: 19/100, batch 183/625 completed (loss: 0.0004:  29%|██▉       | 92/312 [09:15<22:29,  6.14s/it]
Training Epoch: 19/100, batch 183/625 completed (loss: 0.0031:  29%|██▉       | 92/312 [09:15<22:30,  6.14s/it]
Training Epoch: 19/100, batch 183/625 completed (loss: 0.0016:  29%|██▉       | 92/312 [09:15<22:31,  6.14s/it]
Training Epoch: 19/100, batch 184/625 completed (loss: 0.0009:  29%|██▉       | 92/312 [09:17<22:30,  6.14s/it]
Training Epoch: 19/100, batch 184/625 completed (loss: 0.0011:  29%|██▉       | 92/312 [09:17<22:30,  6.14s/it]
Training Epoch: 19/100, batch 184/625 completed (loss: 0.0003:  29%|██▉       | 92/312 [09:17<22:29,  6.14s/it]
Training Epoch: 19/100, batch 184/625 completed (loss: 0.0019:  29%|██▉       | 92/312 [09:17<22:31,  6.14s/it]
Training Epoch: 19/100, batch 184/625 completed (loss: 0.0019:  30%|██▉       | 93/312 [09:20<22:05,  6.05s/it]
Training Epoch: 19/100, batch 184/625 completed (loss: 0.0003:  30%|██▉       | 93/312 [09:20<22:05,  6.05s/it]
Training Epoch: 19/100, batch 184/625 completed (loss: 0.0011:  30%|██▉       | 93/312 [09:20<22:05,  6.05s/it]
Training Epoch: 19/100, batch 184/625 completed (loss: 0.0009:  30%|██▉       | 93/312 [09:20<22:05,  6.05s/it]
Training Epoch: 19/100, batch 185/625 completed (loss: 0.0018:  30%|██▉       | 93/312 [09:20<22:05,  6.05s/it]
Training Epoch: 19/100, batch 185/625 completed (loss: 0.0003:  30%|██▉       | 93/312 [09:20<22:05,  6.05s/it]
Training Epoch: 19/100, batch 185/625 completed (loss: 0.0004:  30%|██▉       | 93/312 [09:20<22:05,  6.05s/it]
Training Epoch: 19/100, batch 185/625 completed (loss: 0.0001:  30%|██▉       | 93/312 [09:20<22:05,  6.05s/it]
Training Epoch: 19/100, batch 186/625 completed (loss: 0.0087:  30%|██▉       | 93/312 [09:23<22:05,  6.05s/it]
Training Epoch: 19/100, batch 186/625 completed (loss: 0.0027:  30%|██▉       | 93/312 [09:23<22:05,  6.05s/it]
Training Epoch: 19/100, batch 186/625 completed (loss: 0.0007:  30%|██▉       | 93/312 [09:23<22:05,  6.05s/it]
Training Epoch: 19/100, batch 186/625 completed (loss: 0.0003:  30%|██▉       | 93/312 [09:24<22:05,  6.05s/it]
Training Epoch: 19/100, batch 186/625 completed (loss: 0.0003:  30%|███       | 94/312 [09:27<22:07,  6.09s/it]
Training Epoch: 19/100, batch 186/625 completed (loss: 0.0027:  30%|███       | 94/312 [09:27<22:07,  6.09s/it]
Training Epoch: 19/100, batch 186/625 completed (loss: 0.0087:  30%|███       | 94/312 [09:27<22:07,  6.09s/it]
Training Epoch: 19/100, batch 186/625 completed (loss: 0.0007:  30%|███       | 94/312 [09:27<22:07,  6.09s/it]
Training Epoch: 19/100, batch 187/625 completed (loss: 0.0011:  30%|███       | 94/312 [09:27<22:07,  6.09s/it]
Training Epoch: 19/100, batch 187/625 completed (loss: 0.0008:  30%|███       | 94/312 [09:27<22:07,  6.09s/it]
Training Epoch: 19/100, batch 187/625 completed (loss: 0.0048:  30%|███       | 94/312 [09:27<22:07,  6.09s/it]
Training Epoch: 19/100, batch 187/625 completed (loss: 0.0013:  30%|███       | 94/312 [09:27<22:07,  6.09s/it]
Training Epoch: 19/100, batch 188/625 completed (loss: 0.0008:  30%|███       | 94/312 [09:29<22:07,  6.09s/it]
Training Epoch: 19/100, batch 188/625 completed (loss: 0.0004:  30%|███       | 94/312 [09:29<22:07,  6.09s/it]
Training Epoch: 19/100, batch 188/625 completed (loss: 0.0001:  30%|███       | 94/312 [09:29<22:07,  6.09s/it]
Training Epoch: 19/100, batch 188/625 completed (loss: 0.0038:  30%|███       | 94/312 [09:30<22:07,  6.09s/it]
Training Epoch: 19/100, batch 188/625 completed (loss: 0.0038:  30%|███       | 95/312 [09:33<21:55,  6.06s/it]
Training Epoch: 19/100, batch 188/625 completed (loss: 0.0001:  30%|███       | 95/312 [09:33<21:55,  6.06s/it]
Training Epoch: 19/100, batch 188/625 completed (loss: 0.0008:  30%|███       | 95/312 [09:33<21:55,  6.06s/it]
Training Epoch: 19/100, batch 188/625 completed (loss: 0.0004:  30%|███       | 95/312 [09:33<21:55,  6.06s/it]
Training Epoch: 19/100, batch 189/625 completed (loss: 0.0005:  30%|███       | 95/312 [09:33<21:55,  6.06s/it]
Training Epoch: 19/100, batch 189/625 completed (loss: 0.0056:  30%|███       | 95/312 [09:33<21:55,  6.06s/it]
Training Epoch: 19/100, batch 189/625 completed (loss: 0.0013:  30%|███       | 95/312 [09:33<21:55,  6.06s/it]
Training Epoch: 19/100, batch 189/625 completed (loss: 0.0004:  30%|███       | 95/312 [09:33<21:55,  6.06s/it]
Training Epoch: 19/100, batch 190/625 completed (loss: 0.0004:  30%|███       | 95/312 [09:35<21:55,  6.06s/it]
Training Epoch: 19/100, batch 190/625 completed (loss: 0.0077:  30%|███       | 95/312 [09:35<21:55,  6.06s/it]
Training Epoch: 19/100, batch 190/625 completed (loss: 0.0027:  30%|███       | 95/312 [09:35<21:55,  6.06s/it]
Training Epoch: 19/100, batch 190/625 completed (loss: 0.0003:  30%|███       | 95/312 [09:36<21:55,  6.06s/it]
Training Epoch: 19/100, batch 190/625 completed (loss: 0.0003:  31%|███       | 96/312 [09:39<21:46,  6.05s/it]
Training Epoch: 19/100, batch 190/625 completed (loss: 0.0004:  31%|███       | 96/312 [09:39<21:46,  6.05s/it]
Training Epoch: 19/100, batch 190/625 completed (loss: 0.0027:  31%|███       | 96/312 [09:39<21:46,  6.05s/it]
Training Epoch: 19/100, batch 190/625 completed (loss: 0.0077:  31%|███       | 96/312 [09:39<21:46,  6.05s/it]
Training Epoch: 19/100, batch 191/625 completed (loss: 0.0178:  31%|███       | 96/312 [09:39<21:46,  6.05s/it]
Training Epoch: 19/100, batch 191/625 completed (loss: 0.0009:  31%|███       | 96/312 [09:39<21:46,  6.05s/it]
Training Epoch: 19/100, batch 191/625 completed (loss: 0.0012:  31%|███       | 96/312 [09:39<21:46,  6.05s/it]
Training Epoch: 19/100, batch 191/625 completed (loss: 0.0054:  31%|███       | 96/312 [09:39<21:46,  6.05s/it]
Training Epoch: 19/100, batch 192/625 completed (loss: 0.0032:  31%|███       | 96/312 [09:42<21:46,  6.05s/it]
Training Epoch: 19/100, batch 192/625 completed (loss: 0.0008:  31%|███       | 96/312 [09:42<21:46,  6.05s/it]
Training Epoch: 19/100, batch 192/625 completed (loss: 0.0062:  31%|███       | 96/312 [09:42<21:46,  6.05s/it]
Training Epoch: 19/100, batch 192/625 completed (loss: 0.0002:  31%|███       | 96/312 [09:42<21:46,  6.05s/it]
Training Epoch: 19/100, batch 192/625 completed (loss: 0.0002:  31%|███       | 97/312 [09:45<21:46,  6.08s/it]
Training Epoch: 19/100, batch 192/625 completed (loss: 0.0062:  31%|███       | 97/312 [09:45<21:46,  6.08s/it]
Training Epoch: 19/100, batch 192/625 completed (loss: 0.0008:  31%|███       | 97/312 [09:45<21:46,  6.08s/it]
Training Epoch: 19/100, batch 192/625 completed (loss: 0.0032:  31%|███       | 97/312 [09:45<21:46,  6.08s/it]
Training Epoch: 19/100, batch 193/625 completed (loss: 0.0001:  31%|███       | 97/312 [09:45<21:46,  6.08s/it]
Training Epoch: 19/100, batch 193/625 completed (loss: 0.002:  31%|███       | 97/312 [09:45<21:46,  6.08s/it] 
Training Epoch: 19/100, batch 193/625 completed (loss: 0.0027:  31%|███       | 97/312 [09:45<21:46,  6.08s/it]
Training Epoch: 19/100, batch 193/625 completed (loss: 0.0004:  31%|███       | 97/312 [09:45<21:46,  6.08s/it]
Training Epoch: 19/100, batch 194/625 completed (loss: 0.0013:  31%|███       | 97/312 [09:48<21:46,  6.08s/it]
Training Epoch: 19/100, batch 194/625 completed (loss: 0.0247:  31%|███       | 97/312 [09:48<21:46,  6.08s/it]
Training Epoch: 19/100, batch 194/625 completed (loss: 0.0001:  31%|███       | 97/312 [09:48<21:46,  6.08s/it]
Training Epoch: 19/100, batch 194/625 completed (loss: 0.0015:  31%|███       | 97/312 [09:48<21:46,  6.08s/it]
Training Epoch: 19/100, batch 194/625 completed (loss: 0.0001:  31%|███▏      | 98/312 [09:51<21:37,  6.06s/it]
Training Epoch: 19/100, batch 194/625 completed (loss: 0.0015:  31%|███▏      | 98/312 [09:51<21:38,  6.07s/it]
Training Epoch: 19/100, batch 194/625 completed (loss: 0.0013:  31%|███▏      | 98/312 [09:51<21:37,  6.06s/it]
Training Epoch: 19/100, batch 194/625 completed (loss: 0.0247:  31%|███▏      | 98/312 [09:51<21:37,  6.07s/it]
Training Epoch: 19/100, batch 195/625 completed (loss: 0.0107:  31%|███▏      | 98/312 [09:51<21:37,  6.06s/it]
Training Epoch: 19/100, batch 195/625 completed (loss: 0.0123:  31%|███▏      | 98/312 [09:51<21:37,  6.07s/it]
Training Epoch: 19/100, batch 195/625 completed (loss: 0.0007:  31%|███▏      | 98/312 [09:51<21:37,  6.06s/it]
Training Epoch: 19/100, batch 195/625 completed (loss: 0.0143:  31%|███▏      | 98/312 [09:51<21:38,  6.07s/it]
Training Epoch: 19/100, batch 196/625 completed (loss: 0.0265:  31%|███▏      | 98/312 [09:54<21:37,  6.06s/it]
Training Epoch: 19/100, batch 196/625 completed (loss: 0.0036:  31%|███▏      | 98/312 [09:54<21:37,  6.07s/it]
Training Epoch: 19/100, batch 196/625 completed (loss: 0.0002:  31%|███▏      | 98/312 [09:54<21:37,  6.06s/it]
Training Epoch: 19/100, batch 196/625 completed (loss: 0.0007:  31%|███▏      | 98/312 [09:54<21:38,  6.07s/it]
Training Epoch: 19/100, batch 196/625 completed (loss: 0.0265:  32%|███▏      | 99/312 [09:57<21:40,  6.11s/it]
Training Epoch: 19/100, batch 196/625 completed (loss: 0.0002:  32%|███▏      | 99/312 [09:57<21:40,  6.11s/it]
Training Epoch: 19/100, batch 196/625 completed (loss: 0.0036:  32%|███▏      | 99/312 [09:57<21:40,  6.11s/it]
Training Epoch: 19/100, batch 196/625 completed (loss: 0.0007:  32%|███▏      | 99/312 [09:57<21:41,  6.11s/it]
Training Epoch: 19/100, batch 197/625 completed (loss: 0.0003:  32%|███▏      | 99/312 [09:57<21:40,  6.11s/it]
Training Epoch: 19/100, batch 197/625 completed (loss: 0.0005:  32%|███▏      | 99/312 [09:57<21:40,  6.11s/it]
Training Epoch: 19/100, batch 197/625 completed (loss: 0.0012:  32%|███▏      | 99/312 [09:57<21:40,  6.11s/it]
Training Epoch: 19/100, batch 197/625 completed (loss: 0.0015:  32%|███▏      | 99/312 [09:57<21:41,  6.11s/it]
Training Epoch: 19/100, batch 198/625 completed (loss: 0.0017:  32%|███▏      | 99/312 [10:00<21:40,  6.11s/it]
Training Epoch: 19/100, batch 198/625 completed (loss: 0.0002:  32%|███▏      | 99/312 [10:00<21:40,  6.11s/it]
Training Epoch: 19/100, batch 198/625 completed (loss: 0.0004:  32%|███▏      | 99/312 [10:00<21:40,  6.11s/it]
Training Epoch: 19/100, batch 198/625 completed (loss: 0.0003:  32%|███▏      | 99/312 [10:00<21:41,  6.11s/it]
Training Epoch: 19/100, batch 198/625 completed (loss: 0.0017:  32%|███▏      | 100/312 [10:03<21:38,  6.13s/it]
Training Epoch: 19/100, batch 198/625 completed (loss: 0.0004:  32%|███▏      | 100/312 [10:03<21:39,  6.13s/it]
Training Epoch: 19/100, batch 198/625 completed (loss: 0.0003:  32%|███▏      | 100/312 [10:03<21:39,  6.13s/it]
Training Epoch: 19/100, batch 198/625 completed (loss: 0.0002:  32%|███▏      | 100/312 [10:03<21:39,  6.13s/it]
Training Epoch: 19/100, batch 199/625 completed (loss: 0.0012:  32%|███▏      | 100/312 [10:03<21:38,  6.13s/it]
Training Epoch: 19/100, batch 199/625 completed (loss: 0.0009:  32%|███▏      | 100/312 [10:03<21:39,  6.13s/it]
Training Epoch: 19/100, batch 199/625 completed (loss: 0.0042:  32%|███▏      | 100/312 [10:03<21:39,  6.13s/it]
Training Epoch: 19/100, batch 199/625 completed (loss: 0.0003:  32%|███▏      | 100/312 [10:03<21:39,  6.13s/it]
Training Epoch: 19/100, batch 200/625 completed (loss: 0.1879:  32%|███▏      | 100/312 [10:06<21:39,  6.13s/it]
Training Epoch: 19/100, batch 200/625 completed (loss: 0.0002:  32%|███▏      | 100/312 [10:06<21:38,  6.13s/it]
Training Epoch: 19/100, batch 200/625 completed (loss: 0.0:  32%|███▏      | 100/312 [10:06<21:39,  6.13s/it]   
Training Epoch: 19/100, batch 200/625 completed (loss: 0.0001:  32%|███▏      | 100/312 [10:06<21:39,  6.13s/it]
Training Epoch: 19/100, batch 200/625 completed (loss: 0.1879:  32%|███▏      | 101/312 [10:09<21:15,  6.05s/it]
Training Epoch: 19/100, batch 200/625 completed (loss: 0.0:  32%|███▏      | 101/312 [10:09<21:15,  6.05s/it]
Training Epoch: 19/100, batch 200/625 completed (loss: 0.0001:  32%|███▏      | 101/312 [10:09<21:15,  6.05s/it]
Training Epoch: 19/100, batch 200/625 completed (loss: 0.0002:  32%|███▏      | 101/312 [10:09<21:16,  6.05s/it]
Training Epoch: 19/100, batch 201/625 completed (loss: 0.0034:  32%|███▏      | 101/312 [10:09<21:15,  6.05s/it]
Training Epoch: 19/100, batch 201/625 completed (loss: 0.0004:  32%|███▏      | 101/312 [10:09<21:16,  6.05s/it]
Training Epoch: 19/100, batch 201/625 completed (loss: 0.0021:  32%|███▏      | 101/312 [10:09<21:15,  6.05s/it]
Training Epoch: 19/100, batch 201/625 completed (loss: 0.1405:  32%|███▏      | 101/312 [10:09<21:15,  6.05s/it]
Training Epoch: 19/100, batch 202/625 completed (loss: 0.0003:  32%|███▏      | 101/312 [10:12<21:15,  6.05s/it]
Training Epoch: 19/100, batch 202/625 completed (loss: 0.0006:  32%|███▏      | 101/312 [10:12<21:15,  6.05s/it]
Training Epoch: 19/100, batch 202/625 completed (loss: 0.003:  32%|███▏      | 101/312 [10:12<21:16,  6.05s/it] 
Training Epoch: 19/100, batch 202/625 completed (loss: 0.0002:  32%|███▏      | 101/312 [10:12<21:15,  6.05s/it]
Training Epoch: 19/100, batch 202/625 completed (loss: 0.003:  33%|███▎      | 102/312 [10:15<21:24,  6.12s/it]
Training Epoch: 19/100, batch 202/625 completed (loss: 0.0003:  33%|███▎      | 102/312 [10:15<21:24,  6.12s/it]
Training Epoch: 19/100, batch 202/625 completed (loss: 0.0002:  33%|███▎      | 102/312 [10:15<21:24,  6.12s/it]
Training Epoch: 19/100, batch 202/625 completed (loss: 0.0006:  33%|███▎      | 102/312 [10:15<21:24,  6.12s/it]
Training Epoch: 19/100, batch 203/625 completed (loss: 0.0022:  33%|███▎      | 102/312 [10:15<21:24,  6.12s/it]
Training Epoch: 19/100, batch 203/625 completed (loss: 0.0229:  33%|███▎      | 102/312 [10:15<21:24,  6.12s/it]
Training Epoch: 19/100, batch 203/625 completed (loss: 0.0007:  33%|███▎      | 102/312 [10:15<21:24,  6.12s/it]
Training Epoch: 19/100, batch 203/625 completed (loss: 0.0:  33%|███▎      | 102/312 [10:15<21:24,  6.12s/it]   
Training Epoch: 19/100, batch 204/625 completed (loss: 0.0002:  33%|███▎      | 102/312 [10:18<21:24,  6.12s/it]
Training Epoch: 19/100, batch 204/625 completed (loss: 0.0029:  33%|███▎      | 102/312 [10:18<21:24,  6.12s/it]
Training Epoch: 19/100, batch 204/625 completed (loss: 0.0001:  33%|███▎      | 102/312 [10:18<21:24,  6.12s/it]
Training Epoch: 19/100, batch 204/625 completed (loss: 0.0009:  33%|███▎      | 102/312 [10:18<21:24,  6.12s/it]
Training Epoch: 19/100, batch 204/625 completed (loss: 0.0001:  33%|███▎      | 103/312 [10:21<21:12,  6.09s/it]
Training Epoch: 19/100, batch 204/625 completed (loss: 0.0002:  33%|███▎      | 103/312 [10:21<21:12,  6.09s/it]
Training Epoch: 19/100, batch 204/625 completed (loss: 0.0029:  33%|███▎      | 103/312 [10:21<21:12,  6.09s/it]
Training Epoch: 19/100, batch 204/625 completed (loss: 0.0009:  33%|███▎      | 103/312 [10:21<21:12,  6.09s/it]
Training Epoch: 19/100, batch 205/625 completed (loss: 0.0161:  33%|███▎      | 103/312 [10:21<21:12,  6.09s/it]
Training Epoch: 19/100, batch 205/625 completed (loss: 0.0004:  33%|███▎      | 103/312 [10:21<21:12,  6.09s/it]
Training Epoch: 19/100, batch 205/625 completed (loss: 0.0001:  33%|███▎      | 103/312 [10:21<21:12,  6.09s/it]
Training Epoch: 19/100, batch 205/625 completed (loss: 0.0006:  33%|███▎      | 103/312 [10:21<21:12,  6.09s/it]
Training Epoch: 19/100, batch 206/625 completed (loss: 0.0063:  33%|███▎      | 103/312 [10:24<21:12,  6.09s/it]
Training Epoch: 19/100, batch 206/625 completed (loss: 0.0005:  33%|███▎      | 103/312 [10:24<21:12,  6.09s/it]
Training Epoch: 19/100, batch 206/625 completed (loss: 0.0001:  33%|███▎      | 103/312 [10:24<21:12,  6.09s/it]
Training Epoch: 19/100, batch 206/625 completed (loss: 0.0004:  33%|███▎      | 103/312 [10:24<21:12,  6.09s/it]
Training Epoch: 19/100, batch 206/625 completed (loss: 0.0005:  33%|███▎      | 104/312 [10:27<21:00,  6.06s/it]
Training Epoch: 19/100, batch 206/625 completed (loss: 0.0004:  33%|███▎      | 104/312 [10:27<21:00,  6.06s/it]
Training Epoch: 19/100, batch 206/625 completed (loss: 0.0063:  33%|███▎      | 104/312 [10:27<21:01,  6.07s/it]
Training Epoch: 19/100, batch 206/625 completed (loss: 0.0001:  33%|███▎      | 104/312 [10:27<21:02,  6.07s/it]
Training Epoch: 19/100, batch 207/625 completed (loss: 0.0004:  33%|███▎      | 104/312 [10:27<21:00,  6.06s/it]
Training Epoch: 19/100, batch 207/625 completed (loss: 0.0009:  33%|███▎      | 104/312 [10:27<21:02,  6.07s/it]
Training Epoch: 19/100, batch 207/625 completed (loss: 0.0003:  33%|███▎      | 104/312 [10:27<21:01,  6.07s/it]
Training Epoch: 19/100, batch 207/625 completed (loss: 0.0092:  33%|███▎      | 104/312 [10:27<21:00,  6.06s/it]
Training Epoch: 19/100, batch 208/625 completed (loss: 0.0009:  33%|███▎      | 104/312 [10:31<21:00,  6.06s/it]
Training Epoch: 19/100, batch 208/625 completed (loss: 0.0023:  33%|███▎      | 104/312 [10:31<21:01,  6.07s/it]
Training Epoch: 19/100, batch 208/625 completed (loss: 0.006:  33%|███▎      | 104/312 [10:31<21:02,  6.07s/it] 
Training Epoch: 19/100, batch 208/625 completed (loss: 0.0009:  33%|███▎      | 104/312 [10:31<21:00,  6.06s/it]
Training Epoch: 19/100, batch 208/625 completed (loss: 0.0009:  34%|███▎      | 105/312 [10:33<20:59,  6.08s/it]
Training Epoch: 19/100, batch 208/625 completed (loss: 0.006:  34%|███▎      | 105/312 [10:33<20:59,  6.08s/it]
Training Epoch: 19/100, batch 208/625 completed (loss: 0.0023:  34%|███▎      | 105/312 [10:33<20:59,  6.08s/it]
Training Epoch: 19/100, batch 208/625 completed (loss: 0.0009:  34%|███▎      | 105/312 [10:33<20:59,  6.08s/it]
Training Epoch: 19/100, batch 209/625 completed (loss: 0.0003:  34%|███▎      | 105/312 [10:33<20:59,  6.08s/it]
Training Epoch: 19/100, batch 209/625 completed (loss: 0.0356:  34%|███▎      | 105/312 [10:33<20:59,  6.08s/it]
Training Epoch: 19/100, batch 209/625 completed (loss: 0.0019:  34%|███▎      | 105/312 [10:33<20:59,  6.08s/it]
Training Epoch: 19/100, batch 209/625 completed (loss: 0.0014:  34%|███▎      | 105/312 [10:33<20:59,  6.08s/it]
Training Epoch: 19/100, batch 210/625 completed (loss: 0.1572:  34%|███▎      | 105/312 [10:37<20:59,  6.08s/it]
Training Epoch: 19/100, batch 210/625 completed (loss: 0.0063:  34%|███▎      | 105/312 [10:37<20:59,  6.08s/it]
Training Epoch: 19/100, batch 210/625 completed (loss: 0.0169:  34%|███▎      | 105/312 [10:37<20:59,  6.08s/it]
Training Epoch: 19/100, batch 210/625 completed (loss: 0.0006:  34%|███▎      | 105/312 [10:37<20:59,  6.08s/it]
Training Epoch: 19/100, batch 210/625 completed (loss: 0.0169:  34%|███▍      | 106/312 [10:39<20:48,  6.06s/it]
Training Epoch: 19/100, batch 210/625 completed (loss: 0.0063:  34%|███▍      | 106/312 [10:39<20:48,  6.06s/it]
Training Epoch: 19/100, batch 210/625 completed (loss: 0.1572:  34%|███▍      | 106/312 [10:39<20:48,  6.06s/it]
Training Epoch: 19/100, batch 210/625 completed (loss: 0.0006:  34%|███▍      | 106/312 [10:39<20:48,  6.06s/it]
Training Epoch: 19/100, batch 211/625 completed (loss: 0.0007:  34%|███▍      | 106/312 [10:39<20:48,  6.06s/it]
Training Epoch: 19/100, batch 211/625 completed (loss: 0.0443:  34%|███▍      | 106/312 [10:39<20:48,  6.06s/it]
Training Epoch: 19/100, batch 211/625 completed (loss: 0.0007:  34%|███▍      | 106/312 [10:39<20:48,  6.06s/it]
Training Epoch: 19/100, batch 211/625 completed (loss: 0.0034:  34%|███▍      | 106/312 [10:39<20:48,  6.06s/it]
Training Epoch: 19/100, batch 212/625 completed (loss: 0.0004:  34%|███▍      | 106/312 [10:43<20:48,  6.06s/it]
Training Epoch: 19/100, batch 212/625 completed (loss: 0.0024:  34%|███▍      | 106/312 [10:43<20:48,  6.06s/it]
Training Epoch: 19/100, batch 212/625 completed (loss: 0.0027:  34%|███▍      | 106/312 [10:43<20:48,  6.06s/it]
Training Epoch: 19/100, batch 212/625 completed (loss: 0.0009:  34%|███▍      | 106/312 [10:43<20:48,  6.06s/it]
Training Epoch: 19/100, batch 212/625 completed (loss: 0.0004:  34%|███▍      | 107/312 [10:46<20:47,  6.09s/it]
Training Epoch: 19/100, batch 212/625 completed (loss: 0.0027:  34%|███▍      | 107/312 [10:46<20:48,  6.09s/it]
Training Epoch: 19/100, batch 212/625 completed (loss: 0.0024:  34%|███▍      | 107/312 [10:46<20:47,  6.09s/it]
Training Epoch: 19/100, batch 212/625 completed (loss: 0.0009:  34%|███▍      | 107/312 [10:46<20:48,  6.09s/it]
Training Epoch: 19/100, batch 213/625 completed (loss: 0.0019:  34%|███▍      | 107/312 [10:46<20:48,  6.09s/it]
Training Epoch: 19/100, batch 213/625 completed (loss: 0.0003:  34%|███▍      | 107/312 [10:46<20:47,  6.09s/it]
Training Epoch: 19/100, batch 213/625 completed (loss: 0.0002:  34%|███▍      | 107/312 [10:46<20:47,  6.09s/it]
Training Epoch: 19/100, batch 213/625 completed (loss: 0.0009:  34%|███▍      | 107/312 [10:46<20:48,  6.09s/it]
Training Epoch: 19/100, batch 214/625 completed (loss: 0.0049:  34%|███▍      | 107/312 [10:49<20:47,  6.09s/it]
Training Epoch: 19/100, batch 214/625 completed (loss: 0.0514:  34%|███▍      | 107/312 [10:49<20:47,  6.09s/it]
Training Epoch: 19/100, batch 214/625 completed (loss: 0.0009:  34%|███▍      | 107/312 [10:49<20:48,  6.09s/it]
Training Epoch: 19/100, batch 214/625 completed (loss: 0.0031:  34%|███▍      | 107/312 [10:49<20:48,  6.09s/it]
Training Epoch: 19/100, batch 214/625 completed (loss: 0.0009:  35%|███▍      | 108/312 [10:52<20:36,  6.06s/it]
Training Epoch: 19/100, batch 214/625 completed (loss: 0.0031:  35%|███▍      | 108/312 [10:52<20:37,  6.07s/it]
Training Epoch: 19/100, batch 214/625 completed (loss: 0.0049:  35%|███▍      | 108/312 [10:52<20:37,  6.07s/it]
Training Epoch: 19/100, batch 214/625 completed (loss: 0.0514:  35%|███▍      | 108/312 [10:52<20:37,  6.07s/it]
Training Epoch: 19/100, batch 215/625 completed (loss: 0.0004:  35%|███▍      | 108/312 [10:52<20:37,  6.07s/it]
Training Epoch: 19/100, batch 215/625 completed (loss: 0.0001:  35%|███▍      | 108/312 [10:52<20:36,  6.06s/it]
Training Epoch: 19/100, batch 215/625 completed (loss: 0.002:  35%|███▍      | 108/312 [10:52<20:37,  6.07s/it] 
Training Epoch: 19/100, batch 215/625 completed (loss: 0.0034:  35%|███▍      | 108/312 [10:52<20:37,  6.07s/it]
Training Epoch: 19/100, batch 216/625 completed (loss: 0.007:  35%|███▍      | 108/312 [10:55<20:36,  6.06s/it] 
Training Epoch: 19/100, batch 216/625 completed (loss: 0.0006:  35%|███▍      | 108/312 [10:55<20:37,  6.07s/it]
Training Epoch: 19/100, batch 216/625 completed (loss: 0.0004:  35%|███▍      | 108/312 [10:55<20:37,  6.07s/it]
Training Epoch: 19/100, batch 216/625 completed (loss: 0.0029:  35%|███▍      | 108/312 [10:55<20:37,  6.07s/it]
Training Epoch: 19/100, batch 216/625 completed (loss: 0.0006:  35%|███▍      | 109/312 [10:58<20:37,  6.09s/it]
Training Epoch: 19/100, batch 216/625 completed (loss: 0.0004:  35%|███▍      | 109/312 [10:58<20:37,  6.09s/it]
Training Epoch: 19/100, batch 216/625 completed (loss: 0.007:  35%|███▍      | 109/312 [10:58<20:38,  6.10s/it]
Training Epoch: 19/100, batch 216/625 completed (loss: 0.0029:  35%|███▍      | 109/312 [10:58<20:37,  6.10s/it]
Training Epoch: 19/100, batch 217/625 completed (loss: 0.0111:  35%|███▍      | 109/312 [10:58<20:38,  6.10s/it]
Training Epoch: 19/100, batch 217/625 completed (loss: 0.0001:  35%|███▍      | 109/312 [10:58<20:37,  6.09s/it]
Training Epoch: 19/100, batch 217/625 completed (loss: 0.0015:  35%|███▍      | 109/312 [10:58<20:37,  6.09s/it]
Training Epoch: 19/100, batch 217/625 completed (loss: 0.0005:  35%|███▍      | 109/312 [10:58<20:37,  6.10s/it]
Training Epoch: 19/100, batch 218/625 completed (loss: 0.0153:  35%|███▍      | 109/312 [11:01<20:38,  6.10s/it]
Training Epoch: 19/100, batch 218/625 completed (loss: 0.0005:  35%|███▍      | 109/312 [11:01<20:37,  6.09s/it]
Training Epoch: 19/100, batch 218/625 completed (loss: 0.0007:  35%|███▍      | 109/312 [11:01<20:37,  6.09s/it]
Training Epoch: 19/100, batch 218/625 completed (loss: 0.0679:  35%|███▍      | 109/312 [11:01<20:37,  6.10s/it]
Training Epoch: 19/100, batch 218/625 completed (loss: 0.0153:  35%|███▌      | 110/312 [11:04<20:34,  6.11s/it]
Training Epoch: 19/100, batch 218/625 completed (loss: 0.0005:  35%|███▌      | 110/312 [11:04<20:34,  6.11s/it]
Training Epoch: 19/100, batch 218/625 completed (loss: 0.0007:  35%|███▌      | 110/312 [11:04<20:34,  6.11s/it]
Training Epoch: 19/100, batch 218/625 completed (loss: 0.0679:  35%|███▌      | 110/312 [11:04<20:34,  6.11s/it]
Training Epoch: 19/100, batch 219/625 completed (loss: 0.0008:  35%|███▌      | 110/312 [11:04<20:34,  6.11s/it]
Training Epoch: 19/100, batch 219/625 completed (loss: 0.0064:  35%|███▌      | 110/312 [11:04<20:34,  6.11s/it]
Training Epoch: 19/100, batch 219/625 completed (loss: 0.0192:  35%|███▌      | 110/312 [11:04<20:34,  6.11s/it]
Training Epoch: 19/100, batch 219/625 completed (loss: 0.0012:  35%|███▌      | 110/312 [11:04<20:34,  6.11s/it]
Training Epoch: 19/100, batch 220/625 completed (loss: 0.0005:  35%|███▌      | 110/312 [11:07<20:34,  6.11s/it]
Training Epoch: 19/100, batch 220/625 completed (loss: 0.0021:  35%|███▌      | 110/312 [11:07<20:34,  6.11s/it]
Training Epoch: 19/100, batch 220/625 completed (loss: 0.0003:  35%|███▌      | 110/312 [11:07<20:34,  6.11s/it]
Training Epoch: 19/100, batch 220/625 completed (loss: 0.0025:  35%|███▌      | 110/312 [11:07<20:34,  6.11s/it]
Training Epoch: 19/100, batch 220/625 completed (loss: 0.0003:  36%|███▌      | 111/312 [11:10<20:22,  6.08s/it]
Training Epoch: 19/100, batch 220/625 completed (loss: 0.0021:  36%|███▌      | 111/312 [11:10<20:22,  6.08s/it]
Training Epoch: 19/100, batch 220/625 completed (loss: 0.0025:  36%|███▌      | 111/312 [11:10<20:22,  6.08s/it]
Training Epoch: 19/100, batch 220/625 completed (loss: 0.0005:  36%|███▌      | 111/312 [11:10<20:22,  6.08s/it]
Training Epoch: 19/100, batch 221/625 completed (loss: 0.0002:  36%|███▌      | 111/312 [11:10<20:22,  6.08s/it]
Training Epoch: 19/100, batch 221/625 completed (loss: 0.0104:  36%|███▌      | 111/312 [11:10<20:22,  6.08s/it]
Training Epoch: 19/100, batch 221/625 completed (loss: 0.0038:  36%|███▌      | 111/312 [11:10<20:22,  6.08s/it]
Training Epoch: 19/100, batch 221/625 completed (loss: 0.0008:  36%|███▌      | 111/312 [11:10<20:22,  6.08s/it]
Training Epoch: 19/100, batch 222/625 completed (loss: 0.0061:  36%|███▌      | 111/312 [11:13<20:22,  6.08s/it]
Training Epoch: 19/100, batch 222/625 completed (loss: 0.0016:  36%|███▌      | 111/312 [11:13<20:22,  6.08s/it]
Training Epoch: 19/100, batch 222/625 completed (loss: 0.0005:  36%|███▌      | 111/312 [11:13<20:22,  6.08s/it]
Training Epoch: 19/100, batch 222/625 completed (loss: 0.0002:  36%|███▌      | 111/312 [11:13<20:22,  6.08s/it]
Training Epoch: 19/100, batch 222/625 completed (loss: 0.0016:  36%|███▌      | 112/312 [11:16<20:22,  6.11s/it]
Training Epoch: 19/100, batch 222/625 completed (loss: 0.0061:  36%|███▌      | 112/312 [11:16<20:23,  6.12s/it]
Training Epoch: 19/100, batch 222/625 completed (loss: 0.0005:  36%|███▌      | 112/312 [11:16<20:23,  6.12s/it]
Training Epoch: 19/100, batch 222/625 completed (loss: 0.0002:  36%|███▌      | 112/312 [11:16<20:23,  6.12s/it]
Training Epoch: 19/100, batch 223/625 completed (loss: 0.0006:  36%|███▌      | 112/312 [11:16<20:22,  6.11s/it]
Training Epoch: 19/100, batch 223/625 completed (loss: 0.0006:  36%|███▌      | 112/312 [11:16<20:23,  6.12s/it]
Training Epoch: 19/100, batch 223/625 completed (loss: 0.0009:  36%|███▌      | 112/312 [11:16<20:23,  6.12s/it]
Training Epoch: 19/100, batch 223/625 completed (loss: 0.0:  36%|███▌      | 112/312 [11:16<20:23,  6.12s/it]   
Training Epoch: 19/100, batch 224/625 completed (loss: 0.0077:  36%|███▌      | 112/312 [11:19<20:23,  6.12s/it]
Training Epoch: 19/100, batch 224/625 completed (loss: 0.0014:  36%|███▌      | 112/312 [11:19<20:22,  6.11s/it]
Training Epoch: 19/100, batch 224/625 completed (loss: 0.0014:  36%|███▌      | 112/312 [11:19<20:23,  6.12s/it]
Training Epoch: 19/100, batch 224/625 completed (loss: 0.0002:  36%|███▌      | 112/312 [11:19<20:23,  6.12s/it]
Training Epoch: 19/100, batch 224/625 completed (loss: 0.0014:  36%|███▌      | 113/312 [11:22<20:10,  6.08s/it]
Training Epoch: 19/100, batch 224/625 completed (loss: 0.0002:  36%|███▌      | 113/312 [11:22<20:09,  6.08s/it]
Training Epoch: 19/100, batch 224/625 completed (loss: 0.0014:  36%|███▌      | 113/312 [11:22<20:11,  6.09s/it]
Training Epoch: 19/100, batch 224/625 completed (loss: 0.0077:  36%|███▌      | 113/312 [11:22<20:10,  6.08s/it]
Training Epoch: 19/100, batch 225/625 completed (loss: 0.0043:  36%|███▌      | 113/312 [11:22<20:10,  6.08s/it]
Training Epoch: 19/100, batch 225/625 completed (loss: 0.0035:  36%|███▌      | 113/312 [11:22<20:10,  6.08s/it]
Training Epoch: 19/100, batch 225/625 completed (loss: 0.0031:  36%|███▌      | 113/312 [11:22<20:11,  6.09s/it]
Training Epoch: 19/100, batch 225/625 completed (loss: 0.0006:  36%|███▌      | 113/312 [11:22<20:09,  6.08s/it]
Training Epoch: 19/100, batch 226/625 completed (loss: 0.0015:  36%|███▌      | 113/312 [11:25<20:10,  6.08s/it]
Training Epoch: 19/100, batch 226/625 completed (loss: 0.0017:  36%|███▌      | 113/312 [11:25<20:10,  6.08s/it]
Training Epoch: 19/100, batch 226/625 completed (loss: 0.0096:  36%|███▌      | 113/312 [11:25<20:11,  6.09s/it]
Training Epoch: 19/100, batch 226/625 completed (loss: 0.0067:  36%|███▌      | 113/312 [11:25<20:09,  6.08s/it]
Training Epoch: 19/100, batch 226/625 completed (loss: 0.0096:  37%|███▋      | 114/312 [11:28<20:08,  6.10s/it]
Training Epoch: 19/100, batch 226/625 completed (loss: 0.0015:  37%|███▋      | 114/312 [11:28<20:09,  6.11s/it]
Training Epoch: 19/100, batch 226/625 completed (loss: 0.0067:  37%|███▋      | 114/312 [11:28<20:08,  6.10s/it]
Training Epoch: 19/100, batch 226/625 completed (loss: 0.0017:  37%|███▋      | 114/312 [11:28<20:09,  6.11s/it]
Training Epoch: 19/100, batch 227/625 completed (loss: 0.0002:  37%|███▋      | 114/312 [11:28<20:09,  6.11s/it]
Training Epoch: 19/100, batch 227/625 completed (loss: 0.0088:  37%|███▋      | 114/312 [11:28<20:09,  6.11s/it]
Training Epoch: 19/100, batch 227/625 completed (loss: 0.0008:  37%|███▋      | 114/312 [11:28<20:08,  6.10s/it]
Training Epoch: 19/100, batch 227/625 completed (loss: 0.0028:  37%|███▋      | 114/312 [11:28<20:08,  6.10s/it]
Training Epoch: 19/100, batch 228/625 completed (loss: 0.0025:  37%|███▋      | 114/312 [11:31<20:09,  6.11s/it]
Training Epoch: 19/100, batch 228/625 completed (loss: 0.0042:  37%|███▋      | 114/312 [11:31<20:08,  6.10s/it]
Training Epoch: 19/100, batch 228/625 completed (loss: 0.0064:  37%|███▋      | 114/312 [11:31<20:09,  6.11s/it]
Training Epoch: 19/100, batch 228/625 completed (loss: 0.001:  37%|███▋      | 114/312 [11:31<20:08,  6.10s/it] 
Training Epoch: 19/100, batch 228/625 completed (loss: 0.001:  37%|███▋      | 115/312 [11:34<20:06,  6.13s/it]
Training Epoch: 19/100, batch 228/625 completed (loss: 0.0042:  37%|███▋      | 115/312 [11:34<20:07,  6.13s/it]
Training Epoch: 19/100, batch 228/625 completed (loss: 0.0025:  37%|███▋      | 115/312 [11:34<20:07,  6.13s/it]
Training Epoch: 19/100, batch 228/625 completed (loss: 0.0064:  37%|███▋      | 115/312 [11:34<20:07,  6.13s/it]
Training Epoch: 19/100, batch 229/625 completed (loss: 0.0008:  37%|███▋      | 115/312 [11:34<20:07,  6.13s/it]
Training Epoch: 19/100, batch 229/625 completed (loss: 0.0031:  37%|███▋      | 115/312 [11:34<20:07,  6.13s/it]
Training Epoch: 19/100, batch 229/625 completed (loss: 0.0:  37%|███▋      | 115/312 [11:34<20:07,  6.13s/it]   
Training Epoch: 19/100, batch 229/625 completed (loss: 0.0002:  37%|███▋      | 115/312 [11:35<20:06,  6.13s/it]
Training Epoch: 19/100, batch 230/625 completed (loss: 0.0106:  37%|███▋      | 115/312 [11:37<20:07,  6.13s/it]
Training Epoch: 19/100, batch 230/625 completed (loss: 0.0006:  37%|███▋      | 115/312 [11:37<20:07,  6.13s/it]
Training Epoch: 19/100, batch 230/625 completed (loss: 0.0034:  37%|███▋      | 115/312 [11:37<20:07,  6.13s/it]
Training Epoch: 19/100, batch 230/625 completed (loss: 0.0006:  37%|███▋      | 115/312 [11:37<20:06,  6.13s/it]
Training Epoch: 19/100, batch 230/625 completed (loss: 0.0006:  37%|███▋      | 116/312 [11:40<19:45,  6.05s/it]
Training Epoch: 19/100, batch 230/625 completed (loss: 0.0106:  37%|███▋      | 116/312 [11:40<19:45,  6.05s/it]
Training Epoch: 19/100, batch 230/625 completed (loss: 0.0006:  37%|███▋      | 116/312 [11:40<19:46,  6.05s/it]
Training Epoch: 19/100, batch 230/625 completed (loss: 0.0034:  37%|███▋      | 116/312 [11:40<19:46,  6.05s/it]
Training Epoch: 19/100, batch 231/625 completed (loss: 0.0002:  37%|███▋      | 116/312 [11:40<19:45,  6.05s/it]
Training Epoch: 19/100, batch 231/625 completed (loss: 0.0022:  37%|███▋      | 116/312 [11:40<19:46,  6.05s/it]
Training Epoch: 19/100, batch 231/625 completed (loss: 0.0003:  37%|███▋      | 116/312 [11:40<19:45,  6.05s/it]
Training Epoch: 19/100, batch 231/625 completed (loss: 0.0004:  37%|███▋      | 116/312 [11:40<19:46,  6.05s/it]
Training Epoch: 19/100, batch 232/625 completed (loss: 0.0072:  37%|███▋      | 116/312 [11:43<19:45,  6.05s/it]
Training Epoch: 19/100, batch 232/625 completed (loss: 0.0018:  37%|███▋      | 116/312 [11:43<19:45,  6.05s/it]
Training Epoch: 19/100, batch 232/625 completed (loss: 0.0:  37%|███▋      | 116/312 [11:43<19:46,  6.05s/it]   
Training Epoch: 19/100, batch 232/625 completed (loss: 0.0008:  37%|███▋      | 116/312 [11:43<19:46,  6.05s/it]
Training Epoch: 19/100, batch 232/625 completed (loss: 0.0008:  38%|███▊      | 117/312 [11:47<19:47,  6.09s/it]
Training Epoch: 19/100, batch 232/625 completed (loss: 0.0072:  38%|███▊      | 117/312 [11:46<19:47,  6.09s/it]
Training Epoch: 19/100, batch 232/625 completed (loss: 0.0:  38%|███▊      | 117/312 [11:46<19:47,  6.09s/it]
Training Epoch: 19/100, batch 232/625 completed (loss: 0.0018:  38%|███▊      | 117/312 [11:46<19:48,  6.10s/it]
Training Epoch: 19/100, batch 233/625 completed (loss: 0.0003:  38%|███▊      | 117/312 [11:47<19:47,  6.09s/it]
Training Epoch: 19/100, batch 233/625 completed (loss: 0.0019:  38%|███▊      | 117/312 [11:47<19:48,  6.10s/it]
Training Epoch: 19/100, batch 233/625 completed (loss: 0.0053:  38%|███▊      | 117/312 [11:47<19:47,  6.09s/it]
Training Epoch: 19/100, batch 233/625 completed (loss: 0.0041:  38%|███▊      | 117/312 [11:47<19:47,  6.09s/it]
Training Epoch: 19/100, batch 234/625 completed (loss: 0.0001:  38%|███▊      | 117/312 [11:49<19:47,  6.09s/it]
Training Epoch: 19/100, batch 234/625 completed (loss: 0.0003:  38%|███▊      | 117/312 [11:49<19:47,  6.09s/it]
Training Epoch: 19/100, batch 234/625 completed (loss: 0.0006:  38%|███▊      | 117/312 [11:49<19:48,  6.10s/it]
Training Epoch: 19/100, batch 234/625 completed (loss: 0.0012:  38%|███▊      | 117/312 [11:50<19:47,  6.09s/it]
Training Epoch: 19/100, batch 234/625 completed (loss: 0.0012:  38%|███▊      | 118/312 [11:53<19:39,  6.08s/it]
Training Epoch: 19/100, batch 234/625 completed (loss: 0.0006:  38%|███▊      | 118/312 [11:53<19:39,  6.08s/it]
Training Epoch: 19/100, batch 234/625 completed (loss: 0.0001:  38%|███▊      | 118/312 [11:53<19:40,  6.08s/it]
Training Epoch: 19/100, batch 234/625 completed (loss: 0.0003:  38%|███▊      | 118/312 [11:53<19:39,  6.08s/it]
Training Epoch: 19/100, batch 235/625 completed (loss: 0.0001:  38%|███▊      | 118/312 [11:53<19:39,  6.08s/it]
Training Epoch: 19/100, batch 235/625 completed (loss: 0.0007:  38%|███▊      | 118/312 [11:53<19:39,  6.08s/it]
Training Epoch: 19/100, batch 235/625 completed (loss: 0.0004:  38%|███▊      | 118/312 [11:53<19:40,  6.08s/it]
Training Epoch: 19/100, batch 235/625 completed (loss: 0.0005:  38%|███▊      | 118/312 [11:53<19:39,  6.08s/it]
Training Epoch: 19/100, batch 236/625 completed (loss: 0.0001:  38%|███▊      | 118/312 [11:56<19:39,  6.08s/it]
Training Epoch: 19/100, batch 236/625 completed (loss: 0.0005:  38%|███▊      | 118/312 [11:56<19:39,  6.08s/it]
Training Epoch: 19/100, batch 236/625 completed (loss: 0.0097:  38%|███▊      | 118/312 [11:56<19:40,  6.08s/it]
Training Epoch: 19/100, batch 236/625 completed (loss: 0.0018:  38%|███▊      | 118/312 [11:56<19:39,  6.08s/it]
Training Epoch: 19/100, batch 236/625 completed (loss: 0.0001:  38%|███▊      | 119/312 [11:59<19:37,  6.10s/it]
Training Epoch: 19/100, batch 236/625 completed (loss: 0.0005:  38%|███▊      | 119/312 [11:59<19:37,  6.10s/it]
Training Epoch: 19/100, batch 236/625 completed (loss: 0.0097:  38%|███▊      | 119/312 [11:59<19:37,  6.10s/it]
Training Epoch: 19/100, batch 236/625 completed (loss: 0.0018:  38%|███▊      | 119/312 [11:59<19:38,  6.11s/it]
Training Epoch: 19/100, batch 237/625 completed (loss: 0.0002:  38%|███▊      | 119/312 [11:59<19:37,  6.10s/it]
Training Epoch: 19/100, batch 237/625 completed (loss: 0.0057:  38%|███▊      | 119/312 [11:59<19:37,  6.10s/it]
Training Epoch: 19/100, batch 237/625 completed (loss: 0.0018:  38%|███▊      | 119/312 [11:59<19:37,  6.10s/it]
Training Epoch: 19/100, batch 237/625 completed (loss: 0.0001:  38%|███▊      | 119/312 [11:59<19:38,  6.11s/it]
Training Epoch: 19/100, batch 238/625 completed (loss: 0.002:  38%|███▊      | 119/312 [12:02<19:37,  6.10s/it] 
Training Epoch: 19/100, batch 238/625 completed (loss: 0.033:  38%|███▊      | 119/312 [12:02<19:37,  6.10s/it] 
Training Epoch: 19/100, batch 238/625 completed (loss: 0.0013:  38%|███▊      | 119/312 [12:02<19:37,  6.10s/it]
Training Epoch: 19/100, batch 238/625 completed (loss: 0.0104:  38%|███▊      | 119/312 [12:02<19:38,  6.11s/it]
Training Epoch: 19/100, batch 238/625 completed (loss: 0.0104:  38%|███▊      | 120/312 [12:05<19:27,  6.08s/it]
Training Epoch: 19/100, batch 238/625 completed (loss: 0.002:  38%|███▊      | 120/312 [12:05<19:28,  6.08s/it]
Training Epoch: 19/100, batch 238/625 completed (loss: 0.033:  38%|███▊      | 120/312 [12:05<19:28,  6.09s/it]
Training Epoch: 19/100, batch 238/625 completed (loss: 0.0013:  38%|███▊      | 120/312 [12:05<19:28,  6.09s/it]
Training Epoch: 19/100, batch 239/625 completed (loss: 0.0006:  38%|███▊      | 120/312 [12:05<19:28,  6.08s/it]
Training Epoch: 19/100, batch 239/625 completed (loss: 0.0538:  38%|███▊      | 120/312 [12:05<19:28,  6.09s/it]
Training Epoch: 19/100, batch 239/625 completed (loss: 0.0003:  38%|███▊      | 120/312 [12:05<19:28,  6.09s/it]
Training Epoch: 19/100, batch 239/625 completed (loss: 0.0004:  38%|███▊      | 120/312 [12:05<19:27,  6.08s/it]
Training Epoch: 19/100, batch 240/625 completed (loss: 0.001:  38%|███▊      | 120/312 [12:08<19:28,  6.09s/it] 
Training Epoch: 19/100, batch 240/625 completed (loss: 0.0013:  38%|███▊      | 120/312 [12:08<19:28,  6.08s/it]
Training Epoch: 19/100, batch 240/625 completed (loss: 0.0018:  38%|███▊      | 120/312 [12:08<19:28,  6.09s/it]
Training Epoch: 19/100, batch 240/625 completed (loss: 0.0003:  38%|███▊      | 120/312 [12:08<19:27,  6.08s/it]
Training Epoch: 19/100, batch 240/625 completed (loss: 0.0013:  39%|███▉      | 121/312 [12:11<19:17,  6.06s/it]
Training Epoch: 19/100, batch 240/625 completed (loss: 0.001:  39%|███▉      | 121/312 [12:11<19:17,  6.06s/it]
Training Epoch: 19/100, batch 240/625 completed (loss: 0.0018:  39%|███▉      | 121/312 [12:11<19:18,  6.06s/it]
Training Epoch: 19/100, batch 240/625 completed (loss: 0.0003:  39%|███▉      | 121/312 [12:11<19:19,  6.07s/it]
Training Epoch: 19/100, batch 241/625 completed (loss: 0.0071:  39%|███▉      | 121/312 [12:11<19:18,  6.06s/it]
Training Epoch: 19/100, batch 241/625 completed (loss: 0.0005:  39%|███▉      | 121/312 [12:11<19:17,  6.06s/it]
Training Epoch: 19/100, batch 241/625 completed (loss: 0.0002:  39%|███▉      | 121/312 [12:11<19:17,  6.06s/it]
Training Epoch: 19/100, batch 241/625 completed (loss: 0.0001:  39%|███▉      | 121/312 [12:11<19:19,  6.07s/it]
Training Epoch: 19/100, batch 242/625 completed (loss: 0.0035:  39%|███▉      | 121/312 [12:14<19:17,  6.06s/it]
Training Epoch: 19/100, batch 242/625 completed (loss: 0.0007:  39%|███▉      | 121/312 [12:14<19:18,  6.06s/it]
Training Epoch: 19/100, batch 242/625 completed (loss: 0.0005:  39%|███▉      | 121/312 [12:14<19:17,  6.06s/it]
Training Epoch: 19/100, batch 242/625 completed (loss: 0.0001:  39%|███▉      | 121/312 [12:14<19:19,  6.07s/it]
Training Epoch: 19/100, batch 242/625 completed (loss: 0.0005:  39%|███▉      | 122/312 [12:17<19:16,  6.09s/it]
Training Epoch: 19/100, batch 242/625 completed (loss: 0.0035:  39%|███▉      | 122/312 [12:17<19:16,  6.09s/it]
Training Epoch: 19/100, batch 242/625 completed (loss: 0.0007:  39%|███▉      | 122/312 [12:17<19:16,  6.09s/it]
Training Epoch: 19/100, batch 242/625 completed (loss: 0.0001:  39%|███▉      | 122/312 [12:17<19:16,  6.09s/it]
Training Epoch: 19/100, batch 243/625 completed (loss: 0.0329:  39%|███▉      | 122/312 [12:17<19:16,  6.09s/it]
Training Epoch: 19/100, batch 243/625 completed (loss: 0.0008:  39%|███▉      | 122/312 [12:17<19:16,  6.09s/it]
Training Epoch: 19/100, batch 243/625 completed (loss: 0.0003:  39%|███▉      | 122/312 [12:17<19:16,  6.09s/it]
Training Epoch: 19/100, batch 243/625 completed (loss: 0.0079:  39%|███▉      | 122/312 [12:17<19:16,  6.09s/it]
Training Epoch: 19/100, batch 244/625 completed (loss: 0.0066:  39%|███▉      | 122/312 [12:20<19:16,  6.09s/it]
Training Epoch: 19/100, batch 244/625 completed (loss: 0.0007:  39%|███▉      | 122/312 [12:20<19:16,  6.09s/it]
Training Epoch: 19/100, batch 244/625 completed (loss: 0.0005:  39%|███▉      | 122/312 [12:20<19:16,  6.09s/it]
Training Epoch: 19/100, batch 244/625 completed (loss: 0.0044:  39%|███▉      | 122/312 [12:20<19:16,  6.09s/it]
Training Epoch: 19/100, batch 244/625 completed (loss: 0.0007:  39%|███▉      | 123/312 [12:23<19:13,  6.10s/it]
Training Epoch: 19/100, batch 244/625 completed (loss: 0.0066:  39%|███▉      | 123/312 [12:23<19:13,  6.10s/it]
Training Epoch: 19/100, batch 244/625 completed (loss: 0.0044:  39%|███▉      | 123/312 [12:23<19:13,  6.10s/it]
Training Epoch: 19/100, batch 244/625 completed (loss: 0.0005:  39%|███▉      | 123/312 [12:23<19:13,  6.10s/it]
Training Epoch: 19/100, batch 245/625 completed (loss: 0.0022:  39%|███▉      | 123/312 [12:23<19:13,  6.10s/it]
Training Epoch: 19/100, batch 245/625 completed (loss: 0.0038:  39%|███▉      | 123/312 [12:23<19:13,  6.10s/it]
Training Epoch: 19/100, batch 245/625 completed (loss: 0.0026:  39%|███▉      | 123/312 [12:23<19:13,  6.10s/it]
Training Epoch: 19/100, batch 245/625 completed (loss: 0.0069:  39%|███▉      | 123/312 [12:23<19:13,  6.10s/it]
Training Epoch: 19/100, batch 246/625 completed (loss: 0.0004:  39%|███▉      | 123/312 [12:26<19:13,  6.10s/it]
Training Epoch: 19/100, batch 246/625 completed (loss: 0.0003:  39%|███▉      | 123/312 [12:26<19:13,  6.10s/it]
Training Epoch: 19/100, batch 246/625 completed (loss: 0.0001:  39%|███▉      | 123/312 [12:26<19:13,  6.10s/it]
Training Epoch: 19/100, batch 246/625 completed (loss: 0.0002:  39%|███▉      | 123/312 [12:26<19:13,  6.10s/it]
Training Epoch: 19/100, batch 246/625 completed (loss: 0.0003:  40%|███▉      | 124/312 [12:29<18:53,  6.03s/it]
Training Epoch: 19/100, batch 246/625 completed (loss: 0.0002:  40%|███▉      | 124/312 [12:29<18:53,  6.03s/it]
Training Epoch: 19/100, batch 246/625 completed (loss: 0.0001:  40%|███▉      | 124/312 [12:29<18:54,  6.03s/it]
Training Epoch: 19/100, batch 246/625 completed (loss: 0.0004:  40%|███▉      | 124/312 [12:29<18:54,  6.03s/it]
Training Epoch: 19/100, batch 247/625 completed (loss: 0.0007:  40%|███▉      | 124/312 [12:29<18:53,  6.03s/it]
Training Epoch: 19/100, batch 247/625 completed (loss: 0.0098:  40%|███▉      | 124/312 [12:29<18:54,  6.03s/it]
Training Epoch: 19/100, batch 247/625 completed (loss: 0.0013:  40%|███▉      | 124/312 [12:29<18:54,  6.03s/it]
Training Epoch: 19/100, batch 247/625 completed (loss: 0.0008:  40%|███▉      | 124/312 [12:29<18:53,  6.03s/it]
Training Epoch: 19/100, batch 248/625 completed (loss: 0.0019:  40%|███▉      | 124/312 [12:32<18:53,  6.03s/it]
Training Epoch: 19/100, batch 248/625 completed (loss: 0.0128:  40%|███▉      | 124/312 [12:32<18:54,  6.03s/it]
Training Epoch: 19/100, batch 248/625 completed (loss: 0.0001:  40%|███▉      | 124/312 [12:32<18:54,  6.03s/it]
Training Epoch: 19/100, batch 248/625 completed (loss: 0.0013:  40%|███▉      | 124/312 [12:32<18:53,  6.03s/it]
Training Epoch: 19/100, batch 248/625 completed (loss: 0.0128:  40%|████      | 125/312 [12:35<19:04,  6.12s/it]
Training Epoch: 19/100, batch 248/625 completed (loss: 0.0019:  40%|████      | 125/312 [12:35<19:04,  6.12s/it]
Training Epoch: 19/100, batch 248/625 completed (loss: 0.0001:  40%|████      | 125/312 [12:35<19:04,  6.12s/it]
Training Epoch: 19/100, batch 248/625 completed (loss: 0.0013:  40%|████      | 125/312 [12:35<19:04,  6.12s/it]
Training Epoch: 19/100, batch 249/625 completed (loss: 0.0006:  40%|████      | 125/312 [12:35<19:04,  6.12s/it]
Training Epoch: 19/100, batch 249/625 completed (loss: 0.0:  40%|████      | 125/312 [12:35<19:04,  6.12s/it]   
Training Epoch: 19/100, batch 249/625 completed (loss: 0.0024:  40%|████      | 125/312 [12:35<19:04,  6.12s/it]
Training Epoch: 19/100, batch 249/625 completed (loss: 0.0006:  40%|████      | 125/312 [12:35<19:04,  6.12s/it]
Training Epoch: 19/100, batch 250/625 completed (loss: 0.0004:  40%|████      | 125/312 [12:38<19:04,  6.12s/it]
Training Epoch: 19/100, batch 250/625 completed (loss: 0.0003:  40%|████      | 125/312 [12:38<19:04,  6.12s/it]
Training Epoch: 19/100, batch 250/625 completed (loss: 0.0024:  40%|████      | 125/312 [12:38<19:04,  6.12s/it]
Training Epoch: 19/100, batch 250/625 completed (loss: 0.0107:  40%|████      | 125/312 [12:38<19:04,  6.12s/it]
Training Epoch: 19/100, batch 250/625 completed (loss: 0.0003:  40%|████      | 126/312 [12:41<18:51,  6.08s/it]
Training Epoch: 19/100, batch 250/625 completed (loss: 0.0024:  40%|████      | 126/312 [12:41<18:51,  6.09s/it]
Training Epoch: 19/100, batch 250/625 completed (loss: 0.0107:  40%|████      | 126/312 [12:41<18:51,  6.08s/it]
Training Epoch: 19/100, batch 250/625 completed (loss: 0.0004:  40%|████      | 126/312 [12:41<18:53,  6.09s/it]
Training Epoch: 19/100, batch 251/625 completed (loss: 0.005:  40%|████      | 126/312 [12:41<18:53,  6.09s/it] 
Training Epoch: 19/100, batch 251/625 completed (loss: 0.0018:  40%|████      | 126/312 [12:41<18:51,  6.09s/it]
Training Epoch: 19/100, batch 251/625 completed (loss: 0.0006:  40%|████      | 126/312 [12:41<18:51,  6.08s/it]
Training Epoch: 19/100, batch 251/625 completed (loss: 0.0029:  40%|████      | 126/312 [12:41<18:51,  6.08s/it]
Training Epoch: 19/100, batch 252/625 completed (loss: 0.0041:  40%|████      | 126/312 [12:44<18:51,  6.08s/it]
Training Epoch: 19/100, batch 252/625 completed (loss: 0.0023:  40%|████      | 126/312 [12:44<18:53,  6.09s/it]
Training Epoch: 19/100, batch 252/625 completed (loss: 0.0003:  40%|████      | 126/312 [12:44<18:51,  6.09s/it]
Training Epoch: 19/100, batch 252/625 completed (loss: 0.0002:  40%|████      | 126/312 [12:44<18:51,  6.08s/it]
Training Epoch: 19/100, batch 252/625 completed (loss: 0.0023:  41%|████      | 127/312 [12:47<18:40,  6.06s/it]
Training Epoch: 19/100, batch 252/625 completed (loss: 0.0003:  41%|████      | 127/312 [12:47<18:41,  6.06s/it]
Training Epoch: 19/100, batch 252/625 completed (loss: 0.0002:  41%|████      | 127/312 [12:47<18:41,  6.06s/it]
Training Epoch: 19/100, batch 252/625 completed (loss: 0.0041:  41%|████      | 127/312 [12:47<18:41,  6.06s/it]
Training Epoch: 19/100, batch 253/625 completed (loss: 0.0001:  41%|████      | 127/312 [12:47<18:41,  6.06s/it]
Training Epoch: 19/100, batch 253/625 completed (loss: 0.0009:  41%|████      | 127/312 [12:47<18:40,  6.06s/it]
Training Epoch: 19/100, batch 253/625 completed (loss: 0.0058:  41%|████      | 127/312 [12:47<18:41,  6.06s/it]
Training Epoch: 19/100, batch 253/625 completed (loss: 0.0001:  41%|████      | 127/312 [12:47<18:41,  6.06s/it]
Training Epoch: 19/100, batch 254/625 completed (loss: 0.0011:  41%|████      | 127/312 [12:50<18:41,  6.06s/it]
Training Epoch: 19/100, batch 254/625 completed (loss: 0.0017:  41%|████      | 127/312 [12:50<18:40,  6.06s/it]
Training Epoch: 19/100, batch 254/625 completed (loss: 0.002:  41%|████      | 127/312 [12:50<18:41,  6.06s/it] 
Training Epoch: 19/100, batch 254/625 completed (loss: 0.0002:  41%|████      | 127/312 [12:50<18:41,  6.06s/it]
Training Epoch: 19/100, batch 254/625 completed (loss: 0.0017:  41%|████      | 128/312 [12:53<18:40,  6.09s/it]
Training Epoch: 19/100, batch 254/625 completed (loss: 0.0002:  41%|████      | 128/312 [12:53<18:40,  6.09s/it]
Training Epoch: 19/100, batch 254/625 completed (loss: 0.0011:  41%|████      | 128/312 [12:53<18:41,  6.09s/it]
Training Epoch: 19/100, batch 254/625 completed (loss: 0.002:  41%|████      | 128/312 [12:53<18:41,  6.10s/it]
Training Epoch: 19/100, batch 255/625 completed (loss: 0.0003:  41%|████      | 128/312 [12:53<18:41,  6.10s/it]
Training Epoch: 19/100, batch 255/625 completed (loss: 0.0011:  41%|████      | 128/312 [12:53<18:41,  6.09s/it]
Training Epoch: 19/100, batch 255/625 completed (loss: 0.0006:  41%|████      | 128/312 [12:53<18:40,  6.09s/it]
Training Epoch: 19/100, batch 255/625 completed (loss: 0.0119:  41%|████      | 128/312 [12:53<18:40,  6.09s/it]
Training Epoch: 19/100, batch 256/625 completed (loss: 0.0574:  41%|████      | 128/312 [12:57<18:41,  6.09s/it]
Training Epoch: 19/100, batch 256/625 completed (loss: 0.001:  41%|████      | 128/312 [12:57<18:40,  6.09s/it] 
Training Epoch: 19/100, batch 256/625 completed (loss: 0.0002:  41%|████      | 128/312 [12:57<18:41,  6.10s/it]
Training Epoch: 19/100, batch 256/625 completed (loss: 0.0024:  41%|████      | 128/312 [12:57<18:40,  6.09s/it]
Training Epoch: 19/100, batch 256/625 completed (loss: 0.0574:  41%|████▏     | 129/312 [12:59<18:28,  6.06s/it]
Training Epoch: 19/100, batch 256/625 completed (loss: 0.0002:  41%|████▏     | 129/312 [12:59<18:28,  6.06s/it]
Training Epoch: 19/100, batch 256/625 completed (loss: 0.001:  41%|████▏     | 129/312 [12:59<18:29,  6.06s/it]
Training Epoch: 19/100, batch 256/625 completed (loss: 0.0024:  41%|████▏     | 129/312 [12:59<18:29,  6.06s/it]
Training Epoch: 19/100, batch 257/625 completed (loss: 0.0009:  41%|████▏     | 129/312 [12:59<18:29,  6.06s/it]
Training Epoch: 19/100, batch 257/625 completed (loss: 0.003:  41%|████▏     | 129/312 [12:59<18:28,  6.06s/it] 
Training Epoch: 19/100, batch 257/625 completed (loss: 0.0026:  41%|████▏     | 129/312 [12:59<18:28,  6.06s/it]
Training Epoch: 19/100, batch 257/625 completed (loss: 0.001:  41%|████▏     | 129/312 [12:59<18:29,  6.06s/it] 
Training Epoch: 19/100, batch 258/625 completed (loss: 0.0025:  41%|████▏     | 129/312 [13:02<18:28,  6.06s/it]
Training Epoch: 19/100, batch 258/625 completed (loss: 0.0007:  41%|████▏     | 129/312 [13:02<18:29,  6.06s/it]
Training Epoch: 19/100, batch 258/625 completed (loss: 0.0031:  41%|████▏     | 129/312 [13:02<18:28,  6.06s/it]
Training Epoch: 19/100, batch 258/625 completed (loss: 0.0006:  41%|████▏     | 129/312 [13:03<18:29,  6.06s/it]
Training Epoch: 19/100, batch 258/625 completed (loss: 0.0007:  42%|████▏     | 130/312 [13:06<18:28,  6.09s/it]
Training Epoch: 19/100, batch 258/625 completed (loss: 0.0031:  42%|████▏     | 130/312 [13:06<18:29,  6.10s/it]
Training Epoch: 19/100, batch 258/625 completed (loss: 0.0006:  42%|████▏     | 130/312 [13:06<18:29,  6.10s/it]
Training Epoch: 19/100, batch 258/625 completed (loss: 0.0025:  42%|████▏     | 130/312 [13:06<18:30,  6.10s/it]
Training Epoch: 19/100, batch 259/625 completed (loss: 0.0002:  42%|████▏     | 130/312 [13:06<18:30,  6.10s/it]
Training Epoch: 19/100, batch 259/625 completed (loss: 0.0012:  42%|████▏     | 130/312 [13:06<18:29,  6.10s/it]
Training Epoch: 19/100, batch 259/625 completed (loss: 0.0008:  42%|████▏     | 130/312 [13:06<18:28,  6.09s/it]
Training Epoch: 19/100, batch 259/625 completed (loss: 0.0004:  42%|████▏     | 130/312 [13:06<18:29,  6.10s/it]
Training Epoch: 19/100, batch 260/625 completed (loss: 0.0003:  42%|████▏     | 130/312 [13:09<18:30,  6.10s/it]
Training Epoch: 19/100, batch 260/625 completed (loss: 0.0012:  42%|████▏     | 130/312 [13:09<18:29,  6.10s/it]
Training Epoch: 19/100, batch 260/625 completed (loss: 0.0011:  42%|████▏     | 130/312 [13:09<18:28,  6.09s/it]
Training Epoch: 19/100, batch 260/625 completed (loss: 0.0016:  42%|████▏     | 130/312 [13:09<18:29,  6.10s/it]
Training Epoch: 19/100, batch 260/625 completed (loss: 0.0003:  42%|████▏     | 131/312 [13:12<18:26,  6.11s/it]
Training Epoch: 19/100, batch 260/625 completed (loss: 0.0011:  42%|████▏     | 131/312 [13:12<18:26,  6.12s/it]
Training Epoch: 19/100, batch 260/625 completed (loss: 0.0016:  42%|████▏     | 131/312 [13:12<18:26,  6.11s/it]
Training Epoch: 19/100, batch 260/625 completed (loss: 0.0012:  42%|████▏     | 131/312 [13:12<18:26,  6.12s/it]
Training Epoch: 19/100, batch 261/625 completed (loss: 0.0001:  42%|████▏     | 131/312 [13:12<18:26,  6.12s/it]
Training Epoch: 19/100, batch 261/625 completed (loss: 0.0157:  42%|████▏     | 131/312 [13:12<18:26,  6.11s/it]
Training Epoch: 19/100, batch 261/625 completed (loss: 0.0004:  42%|████▏     | 131/312 [13:12<18:26,  6.12s/it]
Training Epoch: 19/100, batch 261/625 completed (loss: 0.0017:  42%|████▏     | 131/312 [13:12<18:26,  6.11s/it]
Training Epoch: 19/100, batch 262/625 completed (loss: 0.0018:  42%|████▏     | 131/312 [13:15<18:26,  6.12s/it]
Training Epoch: 19/100, batch 262/625 completed (loss: 0.0008:  42%|████▏     | 131/312 [13:15<18:26,  6.12s/it]
Training Epoch: 19/100, batch 262/625 completed (loss: 0.0006:  42%|████▏     | 131/312 [13:15<18:26,  6.11s/it]
Training Epoch: 19/100, batch 262/625 completed (loss: 0.0007:  42%|████▏     | 131/312 [13:15<18:26,  6.11s/it]
Training Epoch: 19/100, batch 262/625 completed (loss: 0.0007:  42%|████▏     | 132/312 [13:18<18:13,  6.08s/it]
Training Epoch: 19/100, batch 262/625 completed (loss: 0.0006:  42%|████▏     | 132/312 [13:18<18:13,  6.08s/it]
Training Epoch: 19/100, batch 262/625 completed (loss: 0.0018:  42%|████▏     | 132/312 [13:18<18:13,  6.08s/it]
Training Epoch: 19/100, batch 262/625 completed (loss: 0.0008:  42%|████▏     | 132/312 [13:18<18:15,  6.08s/it]
Training Epoch: 19/100, batch 263/625 completed (loss: 0.0005:  42%|████▏     | 132/312 [13:18<18:13,  6.08s/it]
Training Epoch: 19/100, batch 263/625 completed (loss: 0.0005:  42%|████▏     | 132/312 [13:18<18:15,  6.08s/it]
Training Epoch: 19/100, batch 263/625 completed (loss: 0.0035:  42%|████▏     | 132/312 [13:18<18:13,  6.08s/it]
Training Epoch: 19/100, batch 263/625 completed (loss: 0.0018:  42%|████▏     | 132/312 [13:18<18:13,  6.08s/it]
Training Epoch: 19/100, batch 264/625 completed (loss: 0.0042:  42%|████▏     | 132/312 [13:21<18:15,  6.08s/it]
Training Epoch: 19/100, batch 264/625 completed (loss: 0.0006:  42%|████▏     | 132/312 [13:21<18:13,  6.08s/it]
Training Epoch: 19/100, batch 264/625 completed (loss: 0.0003:  42%|████▏     | 132/312 [13:21<18:13,  6.08s/it]
Training Epoch: 19/100, batch 264/625 completed (loss: 0.0014:  42%|████▏     | 132/312 [13:21<18:13,  6.08s/it]
Training Epoch: 19/100, batch 264/625 completed (loss: 0.0014:  43%|████▎     | 133/312 [13:24<18:11,  6.10s/it]
Training Epoch: 19/100, batch 264/625 completed (loss: 0.0042:  43%|████▎     | 133/312 [13:24<18:11,  6.10s/it]
Training Epoch: 19/100, batch 264/625 completed (loss: 0.0006:  43%|████▎     | 133/312 [13:24<18:12,  6.10s/it]
Training Epoch: 19/100, batch 264/625 completed (loss: 0.0003:  43%|████▎     | 133/312 [13:24<18:12,  6.10s/it]
Training Epoch: 19/100, batch 265/625 completed (loss: 0.1016:  43%|████▎     | 133/312 [13:24<18:12,  6.10s/it]
Training Epoch: 19/100, batch 265/625 completed (loss: 0.0035:  43%|████▎     | 133/312 [13:24<18:11,  6.10s/it]
Training Epoch: 19/100, batch 265/625 completed (loss: 0.0033:  43%|████▎     | 133/312 [13:24<18:12,  6.10s/it]
Training Epoch: 19/100, batch 265/625 completed (loss: 0.0018:  43%|████▎     | 133/312 [13:24<18:11,  6.10s/it]
Training Epoch: 19/100, batch 266/625 completed (loss: 0.0064:  43%|████▎     | 133/312 [13:27<18:12,  6.10s/it]
Training Epoch: 19/100, batch 266/625 completed (loss: 0.0015:  43%|████▎     | 133/312 [13:27<18:12,  6.10s/it]
Training Epoch: 19/100, batch 266/625 completed (loss: 0.003:  43%|████▎     | 133/312 [13:27<18:11,  6.10s/it] 
Training Epoch: 19/100, batch 266/625 completed (loss: 0.0:  43%|████▎     | 133/312 [13:27<18:11,  6.10s/it]   
Training Epoch: 19/100, batch 266/625 completed (loss: 0.0064:  43%|████▎     | 134/312 [13:30<18:01,  6.07s/it]
Training Epoch: 19/100, batch 266/625 completed (loss: 0.003:  43%|████▎     | 134/312 [13:30<18:01,  6.07s/it]
Training Epoch: 19/100, batch 266/625 completed (loss: 0.0015:  43%|████▎     | 134/312 [13:30<18:01,  6.07s/it]
Training Epoch: 19/100, batch 266/625 completed (loss: 0.0:  43%|████▎     | 134/312 [13:30<18:01,  6.08s/it]
Training Epoch: 19/100, batch 267/625 completed (loss: 0.0009:  43%|████▎     | 134/312 [13:30<18:01,  6.07s/it]
Training Epoch: 19/100, batch 267/625 completed (loss: 0.0013:  43%|████▎     | 134/312 [13:30<18:01,  6.07s/it]
Training Epoch: 19/100, batch 267/625 completed (loss: 0.0013:  43%|████▎     | 134/312 [13:30<18:01,  6.07s/it]
Training Epoch: 19/100, batch 267/625 completed (loss: 0.0009:  43%|████▎     | 134/312 [13:30<18:01,  6.08s/it]
Training Epoch: 19/100, batch 268/625 completed (loss: 0.003:  43%|████▎     | 134/312 [13:33<18:01,  6.07s/it] 
Training Epoch: 19/100, batch 268/625 completed (loss: 0.0001:  43%|████▎     | 134/312 [13:33<18:01,  6.07s/it]
Training Epoch: 19/100, batch 268/625 completed (loss: 0.014:  43%|████▎     | 134/312 [13:33<18:01,  6.07s/it] 
Training Epoch: 19/100, batch 268/625 completed (loss: 0.0004:  43%|████▎     | 134/312 [13:33<18:01,  6.08s/it]
Training Epoch: 19/100, batch 268/625 completed (loss: 0.0001:  43%|████▎     | 135/312 [13:36<17:51,  6.05s/it]
Training Epoch: 19/100, batch 268/625 completed (loss: 0.003:  43%|████▎     | 135/312 [13:36<17:51,  6.06s/it]
Training Epoch: 19/100, batch 268/625 completed (loss: 0.0004:  43%|████▎     | 135/312 [13:36<17:52,  6.06s/it]
Training Epoch: 19/100, batch 268/625 completed (loss: 0.014:  43%|████▎     | 135/312 [13:36<17:52,  6.06s/it]
Training Epoch: 19/100, batch 269/625 completed (loss: 0.0114:  43%|████▎     | 135/312 [13:36<17:51,  6.06s/it]
Training Epoch: 19/100, batch 269/625 completed (loss: 0.0004:  43%|████▎     | 135/312 [13:36<17:51,  6.05s/it]
Training Epoch: 19/100, batch 269/625 completed (loss: 0.0001:  43%|████▎     | 135/312 [13:36<17:52,  6.06s/it]
Training Epoch: 19/100, batch 269/625 completed (loss: 0.0021:  43%|████▎     | 135/312 [13:36<17:52,  6.06s/it]
Training Epoch: 19/100, batch 270/625 completed (loss: 0.0107:  43%|████▎     | 135/312 [13:39<17:51,  6.06s/it]
Training Epoch: 19/100, batch 270/625 completed (loss: 0.0003:  43%|████▎     | 135/312 [13:39<17:51,  6.05s/it]
Training Epoch: 19/100, batch 270/625 completed (loss: 0.0319:  43%|████▎     | 135/312 [13:39<17:52,  6.06s/it]
Training Epoch: 19/100, batch 270/625 completed (loss: 0.0085:  43%|████▎     | 135/312 [13:39<17:52,  6.06s/it]
Training Epoch: 19/100, batch 270/625 completed (loss: 0.0003:  44%|████▎     | 136/312 [13:42<17:59,  6.13s/it]
Training Epoch: 19/100, batch 270/625 completed (loss: 0.0107:  44%|████▎     | 136/312 [13:42<17:59,  6.13s/it]
Training Epoch: 19/100, batch 270/625 completed (loss: 0.0319:  44%|████▎     | 136/312 [13:42<17:59,  6.13s/it]
Training Epoch: 19/100, batch 270/625 completed (loss: 0.0085:  44%|████▎     | 136/312 [13:42<18:00,  6.14s/it]
Training Epoch: 19/100, batch 271/625 completed (loss: 0.0035:  44%|████▎     | 136/312 [13:42<17:59,  6.13s/it]
Training Epoch: 19/100, batch 271/625 completed (loss: 0.0002:  44%|████▎     | 136/312 [13:42<17:59,  6.13s/it]
Training Epoch: 19/100, batch 271/625 completed (loss: 0.0009:  44%|████▎     | 136/312 [13:42<17:59,  6.13s/it]
Training Epoch: 19/100, batch 271/625 completed (loss: 0.0113:  44%|████▎     | 136/312 [13:42<18:00,  6.14s/it]
Training Epoch: 19/100, batch 272/625 completed (loss: 0.0003:  44%|████▎     | 136/312 [13:45<17:59,  6.13s/it]
Training Epoch: 19/100, batch 272/625 completed (loss: 0.001:  44%|████▎     | 136/312 [13:45<17:59,  6.13s/it] 
Training Epoch: 19/100, batch 272/625 completed (loss: 0.0001:  44%|████▎     | 136/312 [13:45<17:59,  6.13s/it]
Training Epoch: 19/100, batch 272/625 completed (loss: 0.0009:  44%|████▎     | 136/312 [13:45<18:00,  6.14s/it]
Training Epoch: 19/100, batch 272/625 completed (loss: 0.0003:  44%|████▍     | 137/312 [13:48<17:45,  6.09s/it]
Training Epoch: 19/100, batch 272/625 completed (loss: 0.0009:  44%|████▍     | 137/312 [13:48<17:45,  6.09s/it]
Training Epoch: 19/100, batch 272/625 completed (loss: 0.0001:  44%|████▍     | 137/312 [13:48<17:46,  6.09s/it]
Training Epoch: 19/100, batch 272/625 completed (loss: 0.001:  44%|████▍     | 137/312 [13:48<17:46,  6.09s/it]
Training Epoch: 19/100, batch 273/625 completed (loss: 0.0005:  44%|████▍     | 137/312 [13:48<17:46,  6.09s/it]
Training Epoch: 19/100, batch 273/625 completed (loss: 0.2473:  44%|████▍     | 137/312 [13:48<17:45,  6.09s/it]
Training Epoch: 19/100, batch 273/625 completed (loss: 0.0045:  44%|████▍     | 137/312 [13:48<17:46,  6.09s/it]
Training Epoch: 19/100, batch 273/625 completed (loss: 0.0011:  44%|████▍     | 137/312 [13:48<17:45,  6.09s/it]
Training Epoch: 19/100, batch 274/625 completed (loss: 0.0002:  44%|████▍     | 137/312 [13:51<17:46,  6.09s/it]
Training Epoch: 19/100, batch 274/625 completed (loss: 0.0:  44%|████▍     | 137/312 [13:51<17:46,  6.09s/it]   
Training Epoch: 19/100, batch 274/625 completed (loss: 0.0003:  44%|████▍     | 137/312 [13:51<17:45,  6.09s/it]
Training Epoch: 19/100, batch 274/625 completed (loss: 0.0001:  44%|████▍     | 137/312 [13:51<17:45,  6.09s/it]
Training Epoch: 19/100, batch 274/625 completed (loss: 0.0:  44%|████▍     | 138/312 [13:54<17:34,  6.06s/it]
Training Epoch: 19/100, batch 274/625 completed (loss: 0.0003:  44%|████▍     | 138/312 [13:54<17:34,  6.06s/it]
Training Epoch: 19/100, batch 274/625 completed (loss: 0.0002:  44%|████▍     | 138/312 [13:54<17:34,  6.06s/it]
Training Epoch: 19/100, batch 274/625 completed (loss: 0.0001:  44%|████▍     | 138/312 [13:54<17:34,  6.06s/it]
Training Epoch: 19/100, batch 275/625 completed (loss: 0.0033:  44%|████▍     | 138/312 [13:54<17:34,  6.06s/it]
Training Epoch: 19/100, batch 275/625 completed (loss: 0.0034:  44%|████▍     | 138/312 [13:54<17:34,  6.06s/it]
Training Epoch: 19/100, batch 275/625 completed (loss: 0.0005:  44%|████▍     | 138/312 [13:54<17:34,  6.06s/it]
Training Epoch: 19/100, batch 275/625 completed (loss: 0.001:  44%|████▍     | 138/312 [13:54<17:34,  6.06s/it] 
Training Epoch: 19/100, batch 276/625 completed (loss: 0.0001:  44%|████▍     | 138/312 [13:57<17:34,  6.06s/it]
Training Epoch: 19/100, batch 276/625 completed (loss: 0.0009:  44%|████▍     | 138/312 [13:57<17:34,  6.06s/it]
Training Epoch: 19/100, batch 276/625 completed (loss: 0.0012:  44%|████▍     | 138/312 [13:57<17:34,  6.06s/it]
Training Epoch: 19/100, batch 276/625 completed (loss: 0.0008:  44%|████▍     | 138/312 [13:58<17:34,  6.06s/it]
Training Epoch: 19/100, batch 276/625 completed (loss: 0.0001:  45%|████▍     | 139/312 [14:00<17:32,  6.09s/it]
Training Epoch: 19/100, batch 276/625 completed (loss: 0.0009:  45%|████▍     | 139/312 [14:00<17:33,  6.09s/it]
Training Epoch: 19/100, batch 276/625 completed (loss: 0.0012:  45%|████▍     | 139/312 [14:00<17:33,  6.09s/it]
Training Epoch: 19/100, batch 276/625 completed (loss: 0.0008:  45%|████▍     | 139/312 [14:00<17:33,  6.09s/it]
Training Epoch: 19/100, batch 277/625 completed (loss: 0.0032:  45%|████▍     | 139/312 [14:00<17:33,  6.09s/it]
Training Epoch: 19/100, batch 277/625 completed (loss: 0.0014:  45%|████▍     | 139/312 [14:00<17:32,  6.09s/it]
Training Epoch: 19/100, batch 277/625 completed (loss: 0.0144:  45%|████▍     | 139/312 [14:00<17:33,  6.09s/it]
Training Epoch: 19/100, batch 277/625 completed (loss: 0.0006:  45%|████▍     | 139/312 [14:00<17:33,  6.09s/it]
Training Epoch: 19/100, batch 278/625 completed (loss: 0.0022:  45%|████▍     | 139/312 [14:03<17:33,  6.09s/it]
Training Epoch: 19/100, batch 278/625 completed (loss: 0.0015:  45%|████▍     | 139/312 [14:03<17:33,  6.09s/it]
Training Epoch: 19/100, batch 278/625 completed (loss: 0.0002:  45%|████▍     | 139/312 [14:03<17:32,  6.09s/it]
Training Epoch: 19/100, batch 278/625 completed (loss: 0.0004:  45%|████▍     | 139/312 [14:04<17:33,  6.09s/it]
Training Epoch: 19/100, batch 278/625 completed (loss: 0.0015:  45%|████▍     | 140/312 [14:06<17:23,  6.07s/it]
Training Epoch: 19/100, batch 278/625 completed (loss: 0.0022:  45%|████▍     | 140/312 [14:06<17:24,  6.07s/it]
Training Epoch: 19/100, batch 278/625 completed (loss: 0.0002:  45%|████▍     | 140/312 [14:06<17:24,  6.07s/it]
Training Epoch: 19/100, batch 278/625 completed (loss: 0.0004:  45%|████▍     | 140/312 [14:06<17:24,  6.07s/it]
Training Epoch: 19/100, batch 279/625 completed (loss: 0.0007:  45%|████▍     | 140/312 [14:06<17:23,  6.07s/it]
Training Epoch: 19/100, batch 279/625 completed (loss: 0.004:  45%|████▍     | 140/312 [14:06<17:24,  6.07s/it] 
Training Epoch: 19/100, batch 279/625 completed (loss: 0.0006:  45%|████▍     | 140/312 [14:06<17:24,  6.07s/it]
Training Epoch: 19/100, batch 279/625 completed (loss: 0.0003:  45%|████▍     | 140/312 [14:06<17:24,  6.07s/it]
Training Epoch: 19/100, batch 280/625 completed (loss: 0.0:  45%|████▍     | 140/312 [14:09<17:24,  6.07s/it]   
Training Epoch: 19/100, batch 280/625 completed (loss: 0.0008:  45%|████▍     | 140/312 [14:09<17:24,  6.07s/it]
Training Epoch: 19/100, batch 280/625 completed (loss: 0.0002:  45%|████▍     | 140/312 [14:09<17:23,  6.07s/it]
Training Epoch: 19/100, batch 280/625 completed (loss: 0.0004:  45%|████▍     | 140/312 [14:09<17:24,  6.07s/it]
Training Epoch: 19/100, batch 280/625 completed (loss: 0.0008:  45%|████▌     | 141/312 [14:13<17:21,  6.09s/it]
Training Epoch: 19/100, batch 280/625 completed (loss: 0.0002:  45%|████▌     | 141/312 [14:13<17:21,  6.09s/it]
Training Epoch: 19/100, batch 280/625 completed (loss: 0.0:  45%|████▌     | 141/312 [14:13<17:21,  6.09s/it]
Training Epoch: 19/100, batch 280/625 completed (loss: 0.0004:  45%|████▌     | 141/312 [14:13<17:21,  6.09s/it]
Training Epoch: 19/100, batch 281/625 completed (loss: 0.0001:  45%|████▌     | 141/312 [14:13<17:21,  6.09s/it]
Training Epoch: 19/100, batch 281/625 completed (loss: 0.0096:  45%|████▌     | 141/312 [14:13<17:21,  6.09s/it]
Training Epoch: 19/100, batch 281/625 completed (loss: 0.0003:  45%|████▌     | 141/312 [14:13<17:21,  6.09s/it]
Training Epoch: 19/100, batch 281/625 completed (loss: 0.0014:  45%|████▌     | 141/312 [14:13<17:21,  6.09s/it]
Training Epoch: 19/100, batch 282/625 completed (loss: 0.0004:  45%|████▌     | 141/312 [14:16<17:21,  6.09s/it]
Training Epoch: 19/100, batch 282/625 completed (loss: 0.0013:  45%|████▌     | 141/312 [14:16<17:21,  6.09s/it]
Training Epoch: 19/100, batch 282/625 completed (loss: 0.0326:  45%|████▌     | 141/312 [14:16<17:21,  6.09s/it]
Training Epoch: 19/100, batch 282/625 completed (loss: 0.0005:  45%|████▌     | 141/312 [14:16<17:21,  6.09s/it]
Training Epoch: 19/100, batch 282/625 completed (loss: 0.0326:  46%|████▌     | 142/312 [14:19<17:10,  6.06s/it]
Training Epoch: 19/100, batch 282/625 completed (loss: 0.0013:  46%|████▌     | 142/312 [14:19<17:11,  6.07s/it]
Training Epoch: 19/100, batch 282/625 completed (loss: 0.0005:  46%|████▌     | 142/312 [14:19<17:11,  6.07s/it]
Training Epoch: 19/100, batch 282/625 completed (loss: 0.0004:  46%|████▌     | 142/312 [14:19<17:11,  6.07s/it]
Training Epoch: 19/100, batch 283/625 completed (loss: 0.0009:  46%|████▌     | 142/312 [14:19<17:11,  6.07s/it]
Training Epoch: 19/100, batch 283/625 completed (loss: 0.0002:  46%|████▌     | 142/312 [14:19<17:10,  6.06s/it]
Training Epoch: 19/100, batch 283/625 completed (loss: 0.0001:  46%|████▌     | 142/312 [14:19<17:11,  6.07s/it]
Training Epoch: 19/100, batch 283/625 completed (loss: 0.0046:  46%|████▌     | 142/312 [14:19<17:11,  6.07s/it]
Training Epoch: 19/100, batch 284/625 completed (loss: 0.0006:  46%|████▌     | 142/312 [14:22<17:11,  6.07s/it]
Training Epoch: 19/100, batch 284/625 completed (loss: 0.0002:  46%|████▌     | 142/312 [14:22<17:10,  6.06s/it]
Training Epoch: 19/100, batch 284/625 completed (loss: 0.0004:  46%|████▌     | 142/312 [14:22<17:11,  6.07s/it]
Training Epoch: 19/100, batch 284/625 completed (loss: 0.0031:  46%|████▌     | 142/312 [14:22<17:11,  6.07s/it]
Training Epoch: 19/100, batch 284/625 completed (loss: 0.0006:  46%|████▌     | 143/312 [14:25<17:03,  6.05s/it]
Training Epoch: 19/100, batch 284/625 completed (loss: 0.0002:  46%|████▌     | 143/312 [14:25<17:03,  6.06s/it]
Training Epoch: 19/100, batch 284/625 completed (loss: 0.0031:  46%|████▌     | 143/312 [14:25<17:03,  6.05s/it]
Training Epoch: 19/100, batch 284/625 completed (loss: 0.0004:  46%|████▌     | 143/312 [14:25<17:03,  6.06s/it]
Training Epoch: 19/100, batch 285/625 completed (loss: 0.0004:  46%|████▌     | 143/312 [14:25<17:03,  6.05s/it]
Training Epoch: 19/100, batch 285/625 completed (loss: 0.0089:  46%|████▌     | 143/312 [14:25<17:03,  6.06s/it]
Training Epoch: 19/100, batch 285/625 completed (loss: 0.0197:  46%|████▌     | 143/312 [14:25<17:03,  6.06s/it]
Training Epoch: 19/100, batch 285/625 completed (loss: 0.0008:  46%|████▌     | 143/312 [14:25<17:03,  6.05s/it]
Training Epoch: 19/100, batch 286/625 completed (loss: 0.0155:  46%|████▌     | 143/312 [14:28<17:03,  6.06s/it]
Training Epoch: 19/100, batch 286/625 completed (loss: 0.0001:  46%|████▌     | 143/312 [14:28<17:03,  6.05s/it]
Training Epoch: 19/100, batch 286/625 completed (loss: 0.0009:  46%|████▌     | 143/312 [14:28<17:03,  6.06s/it]
Training Epoch: 19/100, batch 286/625 completed (loss: 0.0195:  46%|████▌     | 143/312 [14:28<17:03,  6.05s/it]
Training Epoch: 19/100, batch 286/625 completed (loss: 0.0009:  46%|████▌     | 144/312 [14:31<17:02,  6.09s/it]
Training Epoch: 19/100, batch 286/625 completed (loss: 0.0155:  46%|████▌     | 144/312 [14:31<17:02,  6.09s/it]
Training Epoch: 19/100, batch 286/625 completed (loss: 0.0195:  46%|████▌     | 144/312 [14:31<17:02,  6.09s/it]
Training Epoch: 19/100, batch 286/625 completed (loss: 0.0001:  46%|████▌     | 144/312 [14:31<17:03,  6.09s/it]
Training Epoch: 19/100, batch 287/625 completed (loss: 0.0001:  46%|████▌     | 144/312 [14:31<17:02,  6.09s/it]
Training Epoch: 19/100, batch 287/625 completed (loss: 0.0038:  46%|████▌     | 144/312 [14:31<17:03,  6.09s/it]
Training Epoch: 19/100, batch 287/625 completed (loss: 0.0001:  46%|████▌     | 144/312 [14:31<17:02,  6.09s/it]
Training Epoch: 19/100, batch 287/625 completed (loss: 0.0037:  46%|████▌     | 144/312 [14:31<17:02,  6.09s/it]
Training Epoch: 19/100, batch 288/625 completed (loss: 0.0001:  46%|████▌     | 144/312 [14:34<17:03,  6.09s/it]
Training Epoch: 19/100, batch 288/625 completed (loss: 0.0052:  46%|████▌     | 144/312 [14:34<17:02,  6.09s/it]
Training Epoch: 19/100, batch 288/625 completed (loss: 0.0002:  46%|████▌     | 144/312 [14:34<17:02,  6.09s/it]
Training Epoch: 19/100, batch 288/625 completed (loss: 0.0004:  46%|████▌     | 144/312 [14:34<17:02,  6.09s/it]
Training Epoch: 19/100, batch 288/625 completed (loss: 0.0052:  46%|████▋     | 145/312 [14:37<16:53,  6.07s/it]
Training Epoch: 19/100, batch 288/625 completed (loss: 0.0002:  46%|████▋     | 145/312 [14:37<16:53,  6.07s/it]
Training Epoch: 19/100, batch 288/625 completed (loss: 0.0004:  46%|████▋     | 145/312 [14:37<16:53,  6.07s/it]
Training Epoch: 19/100, batch 288/625 completed (loss: 0.0001:  46%|████▋     | 145/312 [14:37<16:53,  6.07s/it]
Training Epoch: 19/100, batch 289/625 completed (loss: 0.0181:  46%|████▋     | 145/312 [14:37<16:53,  6.07s/it]
Training Epoch: 19/100, batch 289/625 completed (loss: 0.0308:  46%|████▋     | 145/312 [14:37<16:53,  6.07s/it]
Training Epoch: 19/100, batch 289/625 completed (loss: 0.0013:  46%|████▋     | 145/312 [14:37<16:53,  6.07s/it]
Training Epoch: 19/100, batch 289/625 completed (loss: 0.0:  46%|████▋     | 145/312 [14:37<16:53,  6.07s/it]   
Training Epoch: 19/100, batch 290/625 completed (loss: 0.0024:  46%|████▋     | 145/312 [14:40<16:53,  6.07s/it]
Training Epoch: 19/100, batch 290/625 completed (loss: 0.001:  46%|████▋     | 145/312 [14:40<16:53,  6.07s/it] 
Training Epoch: 19/100, batch 290/625 completed (loss: 0.0004:  46%|████▋     | 145/312 [14:40<16:53,  6.07s/it]
Training Epoch: 19/100, batch 290/625 completed (loss: 0.0001:  46%|████▋     | 145/312 [14:40<16:53,  6.07s/it]
Training Epoch: 19/100, batch 290/625 completed (loss: 0.0001:  47%|████▋     | 146/312 [14:43<16:44,  6.05s/it]
Training Epoch: 19/100, batch 290/625 completed (loss: 0.0004:  47%|████▋     | 146/312 [14:43<16:45,  6.05s/it]
Training Epoch: 19/100, batch 290/625 completed (loss: 0.001:  47%|████▋     | 146/312 [14:43<16:45,  6.06s/it]
Training Epoch: 19/100, batch 290/625 completed (loss: 0.0024:  47%|████▋     | 146/312 [14:43<16:45,  6.06s/it]
Training Epoch: 19/100, batch 291/625 completed (loss: 0.0:  47%|████▋     | 146/312 [14:43<16:45,  6.05s/it]   
Training Epoch: 19/100, batch 291/625 completed (loss: 0.0003:  47%|████▋     | 146/312 [14:43<16:45,  6.06s/it]
Training Epoch: 19/100, batch 291/625 completed (loss: 0.0002:  47%|████▋     | 146/312 [14:43<16:45,  6.06s/it]
Training Epoch: 19/100, batch 291/625 completed (loss: 0.0056:  47%|████▋     | 146/312 [14:43<16:44,  6.05s/it]
Training Epoch: 19/100, batch 292/625 completed (loss: 0.0044:  47%|████▋     | 146/312 [14:46<16:45,  6.06s/it]
Training Epoch: 19/100, batch 292/625 completed (loss: 0.0027:  47%|████▋     | 146/312 [14:46<16:45,  6.06s/it]
Training Epoch: 19/100, batch 292/625 completed (loss: 0.0:  47%|████▋     | 146/312 [14:46<16:45,  6.05s/it]
Training Epoch: 19/100, batch 292/625 completed (loss: 0.0188:  47%|████▋     | 146/312 [14:46<16:44,  6.05s/it]
Training Epoch: 19/100, batch 292/625 completed (loss: 0.0027:  47%|████▋     | 147/312 [14:49<16:43,  6.08s/it]
Training Epoch: 19/100, batch 292/625 completed (loss: 0.0044:  47%|████▋     | 147/312 [14:49<16:43,  6.08s/it]
Training Epoch: 19/100, batch 292/625 completed (loss: 0.0:  47%|████▋     | 147/312 [14:49<16:43,  6.08s/it]
Training Epoch: 19/100, batch 292/625 completed (loss: 0.0188:  47%|████▋     | 147/312 [14:49<16:43,  6.08s/it]
Training Epoch: 19/100, batch 293/625 completed (loss: 0.0001:  47%|████▋     | 147/312 [14:49<16:43,  6.08s/it]
Training Epoch: 19/100, batch 293/625 completed (loss: 0.0086:  47%|████▋     | 147/312 [14:49<16:43,  6.08s/it]
Training Epoch: 19/100, batch 293/625 completed (loss: 0.0003:  47%|████▋     | 147/312 [14:49<16:43,  6.08s/it]
Training Epoch: 19/100, batch 293/625 completed (loss: 0.0001:  47%|████▋     | 147/312 [14:49<16:43,  6.08s/it]
Training Epoch: 19/100, batch 294/625 completed (loss: 0.0005:  47%|████▋     | 147/312 [14:52<16:43,  6.08s/it]
Training Epoch: 19/100, batch 294/625 completed (loss: 0.0015:  47%|████▋     | 147/312 [14:52<16:43,  6.08s/it]
Training Epoch: 19/100, batch 294/625 completed (loss: 0.0013:  47%|████▋     | 147/312 [14:52<16:43,  6.08s/it]
Training Epoch: 19/100, batch 294/625 completed (loss: 0.0024:  47%|████▋     | 147/312 [14:52<16:43,  6.08s/it]
Training Epoch: 19/100, batch 294/625 completed (loss: 0.0015:  47%|████▋     | 148/312 [14:55<16:33,  6.06s/it]
Training Epoch: 19/100, batch 294/625 completed (loss: 0.0005:  47%|████▋     | 148/312 [14:55<16:33,  6.06s/it]
Training Epoch: 19/100, batch 294/625 completed (loss: 0.0013:  47%|████▋     | 148/312 [14:55<16:33,  6.06s/it]
Training Epoch: 19/100, batch 294/625 completed (loss: 0.0024:  47%|████▋     | 148/312 [14:55<16:33,  6.06s/it]
Training Epoch: 19/100, batch 295/625 completed (loss: 0.0005:  47%|████▋     | 148/312 [14:55<16:33,  6.06s/it]
Training Epoch: 19/100, batch 295/625 completed (loss: 0.0111:  47%|████▋     | 148/312 [14:55<16:33,  6.06s/it]
Training Epoch: 19/100, batch 295/625 completed (loss: 0.0006:  47%|████▋     | 148/312 [14:55<16:33,  6.06s/it]
Training Epoch: 19/100, batch 295/625 completed (loss: 0.0024:  47%|████▋     | 148/312 [14:55<16:33,  6.06s/it]
Training Epoch: 19/100, batch 296/625 completed (loss: 0.001:  47%|████▋     | 148/312 [14:58<16:33,  6.06s/it] 
Training Epoch: 19/100, batch 296/625 completed (loss: 0.0142:  47%|████▋     | 148/312 [14:58<16:33,  6.06s/it]
Training Epoch: 19/100, batch 296/625 completed (loss: 0.0093:  47%|████▋     | 148/312 [14:58<16:33,  6.06s/it]
Training Epoch: 19/100, batch 296/625 completed (loss: 0.0007:  47%|████▋     | 148/312 [14:58<16:33,  6.06s/it]
Training Epoch: 19/100, batch 296/625 completed (loss: 0.0142:  48%|████▊     | 149/312 [15:01<16:31,  6.08s/it]
Training Epoch: 19/100, batch 296/625 completed (loss: 0.001:  48%|████▊     | 149/312 [15:01<16:32,  6.09s/it]
Training Epoch: 19/100, batch 296/625 completed (loss: 0.0093:  48%|████▊     | 149/312 [15:01<16:32,  6.09s/it]
Training Epoch: 19/100, batch 296/625 completed (loss: 0.0007:  48%|████▊     | 149/312 [15:01<16:32,  6.09s/it]
Training Epoch: 19/100, batch 297/625 completed (loss: 0.0079:  48%|████▊     | 149/312 [15:01<16:32,  6.09s/it]
Training Epoch: 19/100, batch 297/625 completed (loss: 0.0002:  48%|████▊     | 149/312 [15:01<16:32,  6.09s/it]
Training Epoch: 19/100, batch 297/625 completed (loss: 0.0006:  48%|████▊     | 149/312 [15:01<16:31,  6.08s/it]
Training Epoch: 19/100, batch 297/625 completed (loss: 0.0:  48%|████▊     | 149/312 [15:01<16:32,  6.09s/it]   
Training Epoch: 19/100, batch 298/625 completed (loss: 0.0022:  48%|████▊     | 149/312 [15:04<16:32,  6.09s/it]
Training Epoch: 19/100, batch 298/625 completed (loss: 0.0003:  48%|████▊     | 149/312 [15:04<16:32,  6.09s/it]
Training Epoch: 19/100, batch 298/625 completed (loss: 0.0008:  48%|████▊     | 149/312 [15:04<16:31,  6.08s/it]
Training Epoch: 19/100, batch 298/625 completed (loss: 0.001:  48%|████▊     | 149/312 [15:04<16:32,  6.09s/it]
Training Epoch: 19/100, batch 298/625 completed (loss: 0.0003:  48%|████▊     | 150/312 [15:07<16:30,  6.12s/it]
Training Epoch: 19/100, batch 298/625 completed (loss: 0.0022:  48%|████▊     | 150/312 [15:07<16:30,  6.12s/it]
Training Epoch: 19/100, batch 298/625 completed (loss: 0.001:  48%|████▊     | 150/312 [15:07<16:30,  6.12s/it]
Training Epoch: 19/100, batch 298/625 completed (loss: 0.0008:  48%|████▊     | 150/312 [15:07<16:31,  6.12s/it]
Training Epoch: 19/100, batch 299/625 completed (loss: 0.0014:  48%|████▊     | 150/312 [15:07<16:30,  6.12s/it]
Training Epoch: 19/100, batch 299/625 completed (loss: 0.0019:  48%|████▊     | 150/312 [15:07<16:30,  6.12s/it]
Training Epoch: 19/100, batch 299/625 completed (loss: 0.0053:  48%|████▊     | 150/312 [15:07<16:31,  6.12s/it]
Training Epoch: 19/100, batch 299/625 completed (loss: 0.0059:  48%|████▊     | 150/312 [15:07<16:30,  6.12s/it]
Training Epoch: 19/100, batch 300/625 completed (loss: 0.0042:  48%|████▊     | 150/312 [15:10<16:30,  6.12s/it]
Training Epoch: 19/100, batch 300/625 completed (loss: 0.0008:  48%|████▊     | 150/312 [15:10<16:30,  6.12s/it]
Training Epoch: 19/100, batch 300/625 completed (loss: 0.0038:  48%|████▊     | 150/312 [15:10<16:31,  6.12s/it]
Training Epoch: 19/100, batch 300/625 completed (loss: 0.0002:  48%|████▊     | 150/312 [15:10<16:30,  6.12s/it]
Training Epoch: 19/100, batch 300/625 completed (loss: 0.0042:  48%|████▊     | 151/312 [15:13<16:10,  6.03s/it]
Training Epoch: 19/100, batch 300/625 completed (loss: 0.0038:  48%|████▊     | 151/312 [15:13<16:10,  6.03s/it]
Training Epoch: 19/100, batch 300/625 completed (loss: 0.0008:  48%|████▊     | 151/312 [15:13<16:10,  6.03s/it]
Training Epoch: 19/100, batch 300/625 completed (loss: 0.0002:  48%|████▊     | 151/312 [15:13<16:11,  6.03s/it]
Training Epoch: 19/100, batch 301/625 completed (loss: 0.001:  48%|████▊     | 151/312 [15:13<16:10,  6.03s/it] 
Training Epoch: 19/100, batch 301/625 completed (loss: 0.0002:  48%|████▊     | 151/312 [15:13<16:10,  6.03s/it]
Training Epoch: 19/100, batch 301/625 completed (loss: 0.0038:  48%|████▊     | 151/312 [15:13<16:10,  6.03s/it]
Training Epoch: 19/100, batch 301/625 completed (loss: 0.0053:  48%|████▊     | 151/312 [15:13<16:11,  6.03s/it]
Training Epoch: 19/100, batch 302/625 completed (loss: 0.0218:  48%|████▊     | 151/312 [15:16<16:10,  6.03s/it]
Training Epoch: 19/100, batch 302/625 completed (loss: 0.0003:  48%|████▊     | 151/312 [15:16<16:10,  6.03s/it]
Training Epoch: 19/100, batch 302/625 completed (loss: 0.0023:  48%|████▊     | 151/312 [15:16<16:10,  6.03s/it]
Training Epoch: 19/100, batch 302/625 completed (loss: 0.0139:  48%|████▊     | 151/312 [15:16<16:11,  6.03s/it]
Training Epoch: 19/100, batch 302/625 completed (loss: 0.0023:  49%|████▊     | 152/312 [15:19<16:17,  6.11s/it]
Training Epoch: 19/100, batch 302/625 completed (loss: 0.0218:  49%|████▊     | 152/312 [15:19<16:17,  6.11s/it]
Training Epoch: 19/100, batch 302/625 completed (loss: 0.0139:  49%|████▊     | 152/312 [15:19<16:17,  6.11s/it]
Training Epoch: 19/100, batch 302/625 completed (loss: 0.0003:  49%|████▊     | 152/312 [15:19<16:18,  6.12s/it]
Training Epoch: 19/100, batch 303/625 completed (loss: 0.0074:  49%|████▊     | 152/312 [15:19<16:17,  6.11s/it]
Training Epoch: 19/100, batch 303/625 completed (loss: 0.0006:  49%|████▊     | 152/312 [15:19<16:18,  6.12s/it]
Training Epoch: 19/100, batch 303/625 completed (loss: 0.0002:  49%|████▊     | 152/312 [15:19<16:17,  6.11s/it]
Training Epoch: 19/100, batch 303/625 completed (loss: 0.0002:  49%|████▊     | 152/312 [15:19<16:17,  6.11s/it]
Training Epoch: 19/100, batch 304/625 completed (loss: 0.0008:  49%|████▊     | 152/312 [15:23<16:17,  6.11s/it]
Training Epoch: 19/100, batch 304/625 completed (loss: 0.0006:  49%|████▊     | 152/312 [15:23<16:18,  6.12s/it]
Training Epoch: 19/100, batch 304/625 completed (loss: 0.0205:  49%|████▊     | 152/312 [15:23<16:17,  6.11s/it]
Training Epoch: 19/100, batch 304/625 completed (loss: 0.0005:  49%|████▊     | 152/312 [15:23<16:17,  6.11s/it]
Training Epoch: 19/100, batch 304/625 completed (loss: 0.0006:  49%|████▉     | 153/312 [15:25<16:08,  6.09s/it]
Training Epoch: 19/100, batch 304/625 completed (loss: 0.0005:  49%|████▉     | 153/312 [15:25<16:08,  6.09s/it]
Training Epoch: 19/100, batch 304/625 completed (loss: 0.0008:  49%|████▉     | 153/312 [15:25<16:09,  6.10s/it]
Training Epoch: 19/100, batch 304/625 completed (loss: 0.0205:  49%|████▉     | 153/312 [15:25<16:09,  6.10s/it]
Training Epoch: 19/100, batch 305/625 completed (loss: 0.0011:  49%|████▉     | 153/312 [15:25<16:08,  6.09s/it]
Training Epoch: 19/100, batch 305/625 completed (loss: 0.0094:  49%|████▉     | 153/312 [15:25<16:09,  6.10s/it]
Training Epoch: 19/100, batch 305/625 completed (loss: 0.001:  49%|████▉     | 153/312 [15:25<16:09,  6.10s/it] 
Training Epoch: 19/100, batch 305/625 completed (loss: 0.0008:  49%|████▉     | 153/312 [15:26<16:08,  6.09s/it]
Training Epoch: 19/100, batch 306/625 completed (loss: 0.003:  49%|████▉     | 153/312 [15:28<16:09,  6.10s/it]
Training Epoch: 19/100, batch 306/625 completed (loss: 0.0011:  49%|████▉     | 153/312 [15:28<16:08,  6.09s/it]
Training Epoch: 19/100, batch 306/625 completed (loss: 0.0031:  49%|████▉     | 153/312 [15:28<16:09,  6.10s/it]
Training Epoch: 19/100, batch 306/625 completed (loss: 0.0091:  49%|████▉     | 153/312 [15:28<16:08,  6.09s/it]
Training Epoch: 19/100, batch 306/625 completed (loss: 0.0091:  49%|████▉     | 154/312 [15:32<16:05,  6.11s/it]
Training Epoch: 19/100, batch 306/625 completed (loss: 0.0011:  49%|████▉     | 154/312 [15:32<16:05,  6.11s/it]
Training Epoch: 19/100, batch 306/625 completed (loss: 0.003:  49%|████▉     | 154/312 [15:32<16:06,  6.12s/it]
Training Epoch: 19/100, batch 306/625 completed (loss: 0.0031:  49%|████▉     | 154/312 [15:32<16:06,  6.12s/it]
Training Epoch: 19/100, batch 307/625 completed (loss: 0.0002:  49%|████▉     | 154/312 [15:32<16:06,  6.12s/it]
Training Epoch: 19/100, batch 307/625 completed (loss: 0.0002:  49%|████▉     | 154/312 [15:32<16:05,  6.11s/it]
Training Epoch: 19/100, batch 307/625 completed (loss: 0.0004:  49%|████▉     | 154/312 [15:32<16:06,  6.12s/it]
Training Epoch: 19/100, batch 307/625 completed (loss: 0.0026:  49%|████▉     | 154/312 [15:32<16:05,  6.11s/it]
Training Epoch: 19/100, batch 308/625 completed (loss: 0.001:  49%|████▉     | 154/312 [15:35<16:06,  6.12s/it] 
Training Epoch: 19/100, batch 308/625 completed (loss: 0.0044:  49%|████▉     | 154/312 [15:35<16:06,  6.12s/it]
Training Epoch: 19/100, batch 308/625 completed (loss: 0.0015:  49%|████▉     | 154/312 [15:35<16:05,  6.11s/it]
Training Epoch: 19/100, batch 308/625 completed (loss: 0.0007:  49%|████▉     | 154/312 [15:35<16:05,  6.11s/it]
Training Epoch: 19/100, batch 308/625 completed (loss: 0.0015:  50%|████▉     | 155/312 [15:38<16:02,  6.13s/it]
Training Epoch: 19/100, batch 308/625 completed (loss: 0.0044:  50%|████▉     | 155/312 [15:38<16:02,  6.13s/it]
Training Epoch: 19/100, batch 308/625 completed (loss: 0.0007:  50%|████▉     | 155/312 [15:38<16:02,  6.13s/it]
Training Epoch: 19/100, batch 308/625 completed (loss: 0.001:  50%|████▉     | 155/312 [15:38<16:02,  6.13s/it]
Training Epoch: 19/100, batch 309/625 completed (loss: 0.0016:  50%|████▉     | 155/312 [15:38<16:02,  6.13s/it]
Training Epoch: 19/100, batch 309/625 completed (loss: 0.0057:  50%|████▉     | 155/312 [15:38<16:02,  6.13s/it]
Training Epoch: 19/100, batch 309/625 completed (loss: 0.0008:  50%|████▉     | 155/312 [15:38<16:02,  6.13s/it]
Training Epoch: 19/100, batch 309/625 completed (loss: 0.0001:  50%|████▉     | 155/312 [15:38<16:02,  6.13s/it]
Training Epoch: 19/100, batch 310/625 completed (loss: 0.0005:  50%|████▉     | 155/312 [15:41<16:02,  6.13s/it]
Training Epoch: 19/100, batch 310/625 completed (loss: 0.0006:  50%|████▉     | 155/312 [15:41<16:02,  6.13s/it]
Training Epoch: 19/100, batch 310/625 completed (loss: 0.0041:  50%|████▉     | 155/312 [15:41<16:02,  6.13s/it]
Training Epoch: 19/100, batch 310/625 completed (loss: 0.0007:  50%|████▉     | 155/312 [15:41<16:02,  6.13s/it]
Training Epoch: 19/100, batch 310/625 completed (loss: 0.0041:  50%|█████     | 156/312 [15:44<15:43,  6.05s/it]
Training Epoch: 19/100, batch 310/625 completed (loss: 0.0007:  50%|█████     | 156/312 [15:44<15:43,  6.05s/it]
Training Epoch: 19/100, batch 310/625 completed (loss: 0.0005:  50%|█████     | 156/312 [15:44<15:44,  6.05s/it]
Training Epoch: 19/100, batch 310/625 completed (loss: 0.0006:  50%|█████     | 156/312 [15:44<15:43,  6.05s/it]
Training Epoch: 19/100, batch 311/625 completed (loss: 0.0012:  50%|█████     | 156/312 [15:44<15:43,  6.05s/it]
Training Epoch: 19/100, batch 311/625 completed (loss: 0.0013:  50%|█████     | 156/312 [15:44<15:44,  6.05s/it]
Training Epoch: 19/100, batch 311/625 completed (loss: 0.001:  50%|█████     | 156/312 [15:44<15:43,  6.05s/it] 
Training Epoch: 19/100, batch 311/625 completed (loss: 0.0021:  50%|█████     | 156/312 [15:44<15:43,  6.05s/it]
Training Epoch: 19/100, batch 312/625 completed (loss: 0.001:  50%|█████     | 156/312 [15:47<15:44,  6.05s/it] 
Training Epoch: 19/100, batch 312/625 completed (loss: 0.0001:  50%|█████     | 156/312 [15:47<15:43,  6.05s/it]
Training Epoch: 19/100, batch 312/625 completed (loss: 0.0001:  50%|█████     | 156/312 [15:47<15:43,  6.05s/it]
Training Epoch: 19/100, batch 312/625 completed (loss: 0.0024:  50%|█████     | 156/312 [15:47<15:43,  6.05s/it]
Training Epoch: 19/100, batch 312/625 completed (loss: 0.0001:  50%|█████     | 157/312 [15:50<15:50,  6.13s/it]
Training Epoch: 19/100, batch 312/625 completed (loss: 0.0001:  50%|█████     | 157/312 [15:50<15:50,  6.13s/it]
Training Epoch: 19/100, batch 312/625 completed (loss: 0.0024:  50%|█████     | 157/312 [15:50<15:50,  6.13s/it]
Training Epoch: 19/100, batch 312/625 completed (loss: 0.001:  50%|█████     | 157/312 [15:50<15:50,  6.14s/it]
Training Epoch: 19/100, batch 313/625 completed (loss: 0.0557:  50%|█████     | 157/312 [15:50<15:50,  6.14s/it]
Training Epoch: 19/100, batch 313/625 completed (loss: 0.0004:  50%|█████     | 157/312 [15:50<15:50,  6.13s/it]
Training Epoch: 19/100, batch 313/625 completed (loss: 0.0086:  50%|█████     | 157/312 [15:50<15:50,  6.13s/it]
Training Epoch: 19/100, batch 313/625 completed (loss: 0.0006:  50%|█████     | 157/312 [15:50<15:50,  6.13s/it]
Training Epoch: 19/100, batch 314/625 completed (loss: 0.0001:  50%|█████     | 157/312 [15:53<15:50,  6.13s/it]
Training Epoch: 19/100, batch 314/625 completed (loss: 0.0018:  50%|█████     | 157/312 [15:53<15:50,  6.13s/it]
Training Epoch: 19/100, batch 314/625 completed (loss: 0.0009:  50%|█████     | 157/312 [15:53<15:50,  6.14s/it]
Training Epoch: 19/100, batch 314/625 completed (loss: 0.0005:  50%|█████     | 157/312 [15:53<15:50,  6.13s/it]
Training Epoch: 19/100, batch 314/625 completed (loss: 0.0009:  51%|█████     | 158/312 [15:56<15:38,  6.09s/it]
Training Epoch: 19/100, batch 314/625 completed (loss: 0.0018:  51%|█████     | 158/312 [15:56<15:38,  6.09s/it]
Training Epoch: 19/100, batch 314/625 completed (loss: 0.0005:  51%|█████     | 158/312 [15:56<15:38,  6.09s/it]
Training Epoch: 19/100, batch 314/625 completed (loss: 0.0001:  51%|█████     | 158/312 [15:56<15:39,  6.10s/it]
Training Epoch: 19/100, batch 315/625 completed (loss: 0.0001:  51%|█████     | 158/312 [15:56<15:39,  6.10s/it]
Training Epoch: 19/100, batch 315/625 completed (loss: 0.0005:  51%|█████     | 158/312 [15:56<15:38,  6.09s/it]
Training Epoch: 19/100, batch 315/625 completed (loss: 0.0004:  51%|█████     | 158/312 [15:56<15:38,  6.09s/it]
Training Epoch: 19/100, batch 315/625 completed (loss: 0.0155:  51%|█████     | 158/312 [15:56<15:38,  6.09s/it]
Training Epoch: 19/100, batch 316/625 completed (loss: 0.0009:  51%|█████     | 158/312 [15:59<15:38,  6.09s/it]
Training Epoch: 19/100, batch 316/625 completed (loss: 0.0047:  51%|█████     | 158/312 [15:59<15:38,  6.09s/it]
Training Epoch: 19/100, batch 316/625 completed (loss: 0.0004:  51%|█████     | 158/312 [15:59<15:39,  6.10s/it]
Training Epoch: 19/100, batch 316/625 completed (loss: 0.0001:  51%|█████     | 158/312 [15:59<15:38,  6.09s/it]
Training Epoch: 19/100, batch 316/625 completed (loss: 0.0004:  51%|█████     | 159/312 [16:02<15:28,  6.07s/it]
Training Epoch: 19/100, batch 316/625 completed (loss: 0.0009:  51%|█████     | 159/312 [16:02<15:28,  6.07s/it]
Training Epoch: 19/100, batch 316/625 completed (loss: 0.0047:  51%|█████     | 159/312 [16:02<15:29,  6.07s/it]
Training Epoch: 19/100, batch 316/625 completed (loss: 0.0001:  51%|█████     | 159/312 [16:02<15:29,  6.07s/it]
Training Epoch: 19/100, batch 317/625 completed (loss: 0.0073:  51%|█████     | 159/312 [16:02<15:28,  6.07s/it]
Training Epoch: 19/100, batch 317/625 completed (loss: 0.0011:  51%|█████     | 159/312 [16:02<15:28,  6.07s/it]
Training Epoch: 19/100, batch 317/625 completed (loss: 0.0011:  51%|█████     | 159/312 [16:02<15:29,  6.07s/it]
Training Epoch: 19/100, batch 317/625 completed (loss: 0.0002:  51%|█████     | 159/312 [16:02<15:29,  6.07s/it]
Training Epoch: 19/100, batch 318/625 completed (loss: 0.0001:  51%|█████     | 159/312 [16:05<15:29,  6.07s/it]
Training Epoch: 19/100, batch 318/625 completed (loss: 0.0002:  51%|█████     | 159/312 [16:05<15:28,  6.07s/it]
Training Epoch: 19/100, batch 318/625 completed (loss: 0.0002:  51%|█████     | 159/312 [16:05<15:28,  6.07s/it]
Training Epoch: 19/100, batch 318/625 completed (loss: 0.0072:  51%|█████     | 159/312 [16:05<15:29,  6.07s/it]
Training Epoch: 19/100, batch 318/625 completed (loss: 0.0002:  51%|█████▏    | 160/312 [16:08<15:27,  6.10s/it]
Training Epoch: 19/100, batch 318/625 completed (loss: 0.0002:  51%|█████▏    | 160/312 [16:08<15:27,  6.10s/it]
Training Epoch: 19/100, batch 318/625 completed (loss: 0.0001:  51%|█████▏    | 160/312 [16:08<15:27,  6.10s/it]
Training Epoch: 19/100, batch 318/625 completed (loss: 0.0072:  51%|█████▏    | 160/312 [16:08<15:27,  6.10s/it]
Training Epoch: 19/100, batch 319/625 completed (loss: 0.002:  51%|█████▏    | 160/312 [16:08<15:27,  6.10s/it] 
Training Epoch: 19/100, batch 319/625 completed (loss: 0.0004:  51%|█████▏    | 160/312 [16:08<15:27,  6.10s/it]
Training Epoch: 19/100, batch 319/625 completed (loss: 0.0004:  51%|█████▏    | 160/312 [16:08<15:27,  6.10s/it]
Training Epoch: 19/100, batch 319/625 completed (loss: 0.0117:  51%|█████▏    | 160/312 [16:08<15:27,  6.10s/it]
Training Epoch: 19/100, batch 320/625 completed (loss: 0.0012:  51%|█████▏    | 160/312 [16:11<15:27,  6.10s/it]
Training Epoch: 19/100, batch 320/625 completed (loss: 0.0003:  51%|█████▏    | 160/312 [16:11<15:27,  6.10s/it]
Training Epoch: 19/100, batch 320/625 completed (loss: 0.0016:  51%|█████▏    | 160/312 [16:11<15:27,  6.10s/it]
Training Epoch: 19/100, batch 320/625 completed (loss: 0.0025:  51%|█████▏    | 160/312 [16:11<15:27,  6.10s/it]
Training Epoch: 19/100, batch 320/625 completed (loss: 0.0025:  52%|█████▏    | 161/312 [16:14<15:16,  6.07s/it]
Training Epoch: 19/100, batch 320/625 completed (loss: 0.0016:  52%|█████▏    | 161/312 [16:14<15:17,  6.07s/it]
Training Epoch: 19/100, batch 320/625 completed (loss: 0.0003:  52%|█████▏    | 161/312 [16:14<15:17,  6.08s/it]
Training Epoch: 19/100, batch 320/625 completed (loss: 0.0012:  52%|█████▏    | 161/312 [16:14<15:17,  6.07s/it]
Training Epoch: 19/100, batch 321/625 completed (loss: 0.0003:  52%|█████▏    | 161/312 [16:14<15:17,  6.07s/it]
Training Epoch: 19/100, batch 321/625 completed (loss: 0.0008:  52%|█████▏    | 161/312 [16:14<15:17,  6.08s/it]
Training Epoch: 19/100, batch 321/625 completed (loss: 0.0009:  52%|█████▏    | 161/312 [16:14<15:17,  6.07s/it]
Training Epoch: 19/100, batch 321/625 completed (loss: 0.0012:  52%|█████▏    | 161/312 [16:14<15:16,  6.07s/it]
Training Epoch: 19/100, batch 322/625 completed (loss: 0.0017:  52%|█████▏    | 161/312 [16:17<15:17,  6.07s/it]
Training Epoch: 19/100, batch 322/625 completed (loss: 0.0002:  52%|█████▏    | 161/312 [16:17<15:17,  6.07s/it]
Training Epoch: 19/100, batch 322/625 completed (loss: 0.0026:  52%|█████▏    | 161/312 [16:17<15:17,  6.08s/it]
Training Epoch: 19/100, batch 322/625 completed (loss: 0.0006:  52%|█████▏    | 161/312 [16:17<15:16,  6.07s/it]
Training Epoch: 19/100, batch 322/625 completed (loss: 0.0026:  52%|█████▏    | 162/312 [16:20<15:13,  6.09s/it]
Training Epoch: 19/100, batch 322/625 completed (loss: 0.0002:  52%|█████▏    | 162/312 [16:20<15:13,  6.09s/it]
Training Epoch: 19/100, batch 322/625 completed (loss: 0.0006:  52%|█████▏    | 162/312 [16:20<15:13,  6.09s/it]
Training Epoch: 19/100, batch 322/625 completed (loss: 0.0017:  52%|█████▏    | 162/312 [16:20<15:14,  6.09s/it]
Training Epoch: 19/100, batch 323/625 completed (loss: 0.0002:  52%|█████▏    | 162/312 [16:20<15:13,  6.09s/it]
Training Epoch: 19/100, batch 323/625 completed (loss: 0.0005:  52%|█████▏    | 162/312 [16:20<15:13,  6.09s/it]
Training Epoch: 19/100, batch 323/625 completed (loss: 0.0028:  52%|█████▏    | 162/312 [16:20<15:14,  6.09s/it]
Training Epoch: 19/100, batch 323/625 completed (loss: 0.0026:  52%|█████▏    | 162/312 [16:20<15:13,  6.09s/it]
Training Epoch: 19/100, batch 324/625 completed (loss: 0.0122:  52%|█████▏    | 162/312 [16:23<15:14,  6.09s/it]
Training Epoch: 19/100, batch 324/625 completed (loss: 0.0001:  52%|█████▏    | 162/312 [16:23<15:13,  6.09s/it]
Training Epoch: 19/100, batch 324/625 completed (loss: 0.0006:  52%|█████▏    | 162/312 [16:23<15:13,  6.09s/it]
Training Epoch: 19/100, batch 324/625 completed (loss: 0.0003:  52%|█████▏    | 162/312 [16:23<15:13,  6.09s/it]
Training Epoch: 19/100, batch 324/625 completed (loss: 0.0001:  52%|█████▏    | 163/312 [16:27<15:14,  6.14s/it]
Training Epoch: 19/100, batch 324/625 completed (loss: 0.0006:  52%|█████▏    | 163/312 [16:27<15:14,  6.14s/it]
Training Epoch: 19/100, batch 324/625 completed (loss: 0.0003:  52%|█████▏    | 163/312 [16:27<15:14,  6.14s/it]
Training Epoch: 19/100, batch 324/625 completed (loss: 0.0122:  52%|█████▏    | 163/312 [16:27<15:14,  6.14s/it]
Training Epoch: 19/100, batch 325/625 completed (loss: 0.001:  52%|█████▏    | 163/312 [16:27<15:14,  6.14s/it] 
Training Epoch: 19/100, batch 325/625 completed (loss: 0.0044:  52%|█████▏    | 163/312 [16:27<15:14,  6.14s/it]
Training Epoch: 19/100, batch 325/625 completed (loss: 0.0119:  52%|█████▏    | 163/312 [16:27<15:14,  6.14s/it]
Training Epoch: 19/100, batch 325/625 completed (loss: 0.0022:  52%|█████▏    | 163/312 [16:27<15:14,  6.14s/it]
Training Epoch: 19/100, batch 326/625 completed (loss: 0.0012:  52%|█████▏    | 163/312 [16:30<15:14,  6.14s/it]
Training Epoch: 19/100, batch 326/625 completed (loss: 0.0014:  52%|█████▏    | 163/312 [16:30<15:14,  6.14s/it]
Training Epoch: 19/100, batch 326/625 completed (loss: 0.0:  52%|█████▏    | 163/312 [16:30<15:14,  6.14s/it]   
Training Epoch: 19/100, batch 326/625 completed (loss: 0.0007:  52%|█████▏    | 163/312 [16:30<15:14,  6.14s/it]
Training Epoch: 19/100, batch 326/625 completed (loss: 0.0014:  53%|█████▎    | 164/312 [16:32<14:56,  6.06s/it]
Training Epoch: 19/100, batch 326/625 completed (loss: 0.0007:  53%|█████▎    | 164/312 [16:32<14:56,  6.06s/it]
Training Epoch: 19/100, batch 326/625 completed (loss: 0.0012:  53%|█████▎    | 164/312 [16:32<14:57,  6.06s/it]
Training Epoch: 19/100, batch 326/625 completed (loss: 0.0:  53%|█████▎    | 164/312 [16:32<14:57,  6.07s/it]
Training Epoch: 19/100, batch 327/625 completed (loss: 0.0045:  53%|█████▎    | 164/312 [16:32<14:57,  6.06s/it]
Training Epoch: 19/100, batch 327/625 completed (loss: 0.0048:  53%|█████▎    | 164/312 [16:32<14:56,  6.06s/it]
Training Epoch: 19/100, batch 327/625 completed (loss: 0.0118:  53%|█████▎    | 164/312 [16:32<14:57,  6.07s/it]
Training Epoch: 19/100, batch 327/625 completed (loss: 0.0025:  53%|█████▎    | 164/312 [16:33<14:56,  6.06s/it]
Training Epoch: 19/100, batch 328/625 completed (loss: 0.0003:  53%|█████▎    | 164/312 [16:36<14:56,  6.06s/it]
Training Epoch: 19/100, batch 328/625 completed (loss: 0.0004:  53%|█████▎    | 164/312 [16:36<14:57,  6.07s/it]
Training Epoch: 19/100, batch 328/625 completed (loss: 0.0003:  53%|█████▎    | 164/312 [16:36<14:57,  6.06s/it]
Training Epoch: 19/100, batch 328/625 completed (loss: 0.0111:  53%|█████▎    | 164/312 [16:36<14:56,  6.06s/it]
Training Epoch: 19/100, batch 328/625 completed (loss: 0.0003:  53%|█████▎    | 165/312 [16:39<14:56,  6.10s/it]
Training Epoch: 19/100, batch 328/625 completed (loss: 0.0004:  53%|█████▎    | 165/312 [16:39<14:56,  6.10s/it]
Training Epoch: 19/100, batch 328/625 completed (loss: 0.0003:  53%|█████▎    | 165/312 [16:39<14:56,  6.10s/it]
Training Epoch: 19/100, batch 328/625 completed (loss: 0.0111:  53%|█████▎    | 165/312 [16:39<14:57,  6.10s/it]
Training Epoch: 19/100, batch 329/625 completed (loss: 0.0036:  53%|█████▎    | 165/312 [16:39<14:56,  6.10s/it]
Training Epoch: 19/100, batch 329/625 completed (loss: 0.0034:  53%|█████▎    | 165/312 [16:39<14:56,  6.10s/it]
Training Epoch: 19/100, batch 329/625 completed (loss: 0.0002:  53%|█████▎    | 165/312 [16:39<14:56,  6.10s/it]
Training Epoch: 19/100, batch 329/625 completed (loss: 0.0012:  53%|█████▎    | 165/312 [16:39<14:57,  6.10s/it]
Training Epoch: 19/100, batch 330/625 completed (loss: 0.0013:  53%|█████▎    | 165/312 [16:42<14:56,  6.10s/it]
Training Epoch: 19/100, batch 330/625 completed (loss: 0.0006:  53%|█████▎    | 165/312 [16:42<14:56,  6.10s/it]
Training Epoch: 19/100, batch 330/625 completed (loss: 0.0003:  53%|█████▎    | 165/312 [16:42<14:56,  6.10s/it]
Training Epoch: 19/100, batch 330/625 completed (loss: 0.0001:  53%|█████▎    | 165/312 [16:42<14:57,  6.10s/it]
Training Epoch: 19/100, batch 330/625 completed (loss: 0.0013:  53%|█████▎    | 166/312 [16:45<14:53,  6.12s/it]
Training Epoch: 19/100, batch 330/625 completed (loss: 0.0003:  53%|█████▎    | 166/312 [16:45<14:53,  6.12s/it]
Training Epoch: 19/100, batch 330/625 completed (loss: 0.0006:  53%|█████▎    | 166/312 [16:45<14:53,  6.12s/it]
Training Epoch: 19/100, batch 330/625 completed (loss: 0.0001:  53%|█████▎    | 166/312 [16:45<14:53,  6.12s/it]
Training Epoch: 19/100, batch 331/625 completed (loss: 0.0002:  53%|█████▎    | 166/312 [16:45<14:53,  6.12s/it]
Training Epoch: 19/100, batch 331/625 completed (loss: 0.0009:  53%|█████▎    | 166/312 [16:45<14:53,  6.12s/it]
Training Epoch: 19/100, batch 331/625 completed (loss: 0.0009:  53%|█████▎    | 166/312 [16:45<14:53,  6.12s/it]
Training Epoch: 19/100, batch 331/625 completed (loss: 0.0003:  53%|█████▎    | 166/312 [16:45<14:53,  6.12s/it]
Training Epoch: 19/100, batch 332/625 completed (loss: 0.0002:  53%|█████▎    | 166/312 [16:48<14:53,  6.12s/it]
Training Epoch: 19/100, batch 332/625 completed (loss: 0.0002:  53%|█████▎    | 166/312 [16:48<14:53,  6.12s/it]
Training Epoch: 19/100, batch 332/625 completed (loss: 0.0012:  53%|█████▎    | 166/312 [16:48<14:53,  6.12s/it]
Training Epoch: 19/100, batch 332/625 completed (loss: 0.0008:  53%|█████▎    | 166/312 [16:48<14:53,  6.12s/it]
Training Epoch: 19/100, batch 332/625 completed (loss: 0.0012:  54%|█████▎    | 167/312 [16:51<14:41,  6.08s/it]
Training Epoch: 19/100, batch 332/625 completed (loss: 0.0002:  54%|█████▎    | 167/312 [16:51<14:41,  6.08s/it]
Training Epoch: 19/100, batch 332/625 completed (loss: 0.0002:  54%|█████▎    | 167/312 [16:51<14:41,  6.08s/it]
Training Epoch: 19/100, batch 332/625 completed (loss: 0.0008:  54%|█████▎    | 167/312 [16:51<14:42,  6.09s/it]
Training Epoch: 19/100, batch 333/625 completed (loss: 0.0126:  54%|█████▎    | 167/312 [16:51<14:41,  6.08s/it]
Training Epoch: 19/100, batch 333/625 completed (loss: 0.0016:  54%|█████▎    | 167/312 [16:51<14:41,  6.08s/it]
Training Epoch: 19/100, batch 333/625 completed (loss: 0.0008:  54%|█████▎    | 167/312 [16:51<14:41,  6.08s/it]
Training Epoch: 19/100, batch 333/625 completed (loss: 0.0049:  54%|█████▎    | 167/312 [16:51<14:42,  6.09s/it]
Training Epoch: 19/100, batch 334/625 completed (loss: 0.0022:  54%|█████▎    | 167/312 [16:54<14:41,  6.08s/it]
Training Epoch: 19/100, batch 334/625 completed (loss: 0.0039:  54%|█████▎    | 167/312 [16:54<14:41,  6.08s/it]
Training Epoch: 19/100, batch 334/625 completed (loss: 0.0:  54%|█████▎    | 167/312 [16:54<14:41,  6.08s/it]   
Training Epoch: 19/100, batch 334/625 completed (loss: 0.0046:  54%|█████▎    | 167/312 [16:54<14:42,  6.09s/it]
Training Epoch: 19/100, batch 334/625 completed (loss: 0.0046:  54%|█████▍    | 168/312 [16:57<14:38,  6.10s/it]
Training Epoch: 19/100, batch 334/625 completed (loss: 0.0039:  54%|█████▍    | 168/312 [16:57<14:38,  6.10s/it]
Training Epoch: 19/100, batch 334/625 completed (loss: 0.0022:  54%|█████▍    | 168/312 [16:57<14:39,  6.11s/it]
Training Epoch: 19/100, batch 334/625 completed (loss: 0.0:  54%|█████▍    | 168/312 [16:57<14:39,  6.11s/it]
Training Epoch: 19/100, batch 335/625 completed (loss: 0.0022:  54%|█████▍    | 168/312 [16:57<14:39,  6.11s/it]
Training Epoch: 19/100, batch 335/625 completed (loss: 0.0005:  54%|█████▍    | 168/312 [16:57<14:39,  6.11s/it]
Training Epoch: 19/100, batch 335/625 completed (loss: 0.0003:  54%|█████▍    | 168/312 [16:57<14:38,  6.10s/it]
Training Epoch: 19/100, batch 335/625 completed (loss: 0.0008:  54%|█████▍    | 168/312 [16:57<14:38,  6.10s/it]
Training Epoch: 19/100, batch 336/625 completed (loss: 0.0013:  54%|█████▍    | 168/312 [17:00<14:39,  6.11s/it]
Training Epoch: 19/100, batch 336/625 completed (loss: 0.0004:  54%|█████▍    | 168/312 [17:00<14:39,  6.11s/it]
Training Epoch: 19/100, batch 336/625 completed (loss: 0.0009:  54%|█████▍    | 168/312 [17:00<14:38,  6.10s/it]
Training Epoch: 19/100, batch 336/625 completed (loss: 0.0146:  54%|█████▍    | 168/312 [17:00<14:38,  6.10s/it]
Training Epoch: 19/100, batch 336/625 completed (loss: 0.0146:  54%|█████▍    | 169/312 [17:03<14:28,  6.07s/it]
Training Epoch: 19/100, batch 336/625 completed (loss: 0.0009:  54%|█████▍    | 169/312 [17:03<14:28,  6.08s/it]
Training Epoch: 19/100, batch 336/625 completed (loss: 0.0013:  54%|█████▍    | 169/312 [17:03<14:28,  6.08s/it]
Training Epoch: 19/100, batch 336/625 completed (loss: 0.0004:  54%|█████▍    | 169/312 [17:03<14:29,  6.08s/it]
Training Epoch: 19/100, batch 337/625 completed (loss: 0.0075:  54%|█████▍    | 169/312 [17:03<14:29,  6.08s/it]
Training Epoch: 19/100, batch 337/625 completed (loss: 0.0002:  54%|█████▍    | 169/312 [17:03<14:28,  6.08s/it]
Training Epoch: 19/100, batch 337/625 completed (loss: 0.0025:  54%|█████▍    | 169/312 [17:03<14:28,  6.08s/it]
Training Epoch: 19/100, batch 337/625 completed (loss: 0.0008:  54%|█████▍    | 169/312 [17:03<14:28,  6.07s/it]
Training Epoch: 19/100, batch 338/625 completed (loss: 0.001:  54%|█████▍    | 169/312 [17:06<14:28,  6.08s/it] 
Training Epoch: 19/100, batch 338/625 completed (loss: 0.0003:  54%|█████▍    | 169/312 [17:06<14:29,  6.08s/it]
Training Epoch: 19/100, batch 338/625 completed (loss: 0.0145:  54%|█████▍    | 169/312 [17:06<14:28,  6.08s/it]
Training Epoch: 19/100, batch 338/625 completed (loss: 0.0019:  54%|█████▍    | 169/312 [17:06<14:28,  6.07s/it]
Training Epoch: 19/100, batch 338/625 completed (loss: 0.0145:  54%|█████▍    | 170/312 [17:09<14:20,  6.06s/it]
Training Epoch: 19/100, batch 338/625 completed (loss: 0.0003:  54%|█████▍    | 170/312 [17:09<14:20,  6.06s/it]
Training Epoch: 19/100, batch 338/625 completed (loss: 0.0019:  54%|█████▍    | 170/312 [17:09<14:20,  6.06s/it]
Training Epoch: 19/100, batch 338/625 completed (loss: 0.001:  54%|█████▍    | 170/312 [17:09<14:20,  6.06s/it]
Training Epoch: 19/100, batch 339/625 completed (loss: 0.0018:  54%|█████▍    | 170/312 [17:09<14:20,  6.06s/it]
Training Epoch: 19/100, batch 339/625 completed (loss: 0.0015:  54%|█████▍    | 170/312 [17:09<14:20,  6.06s/it]
Training Epoch: 19/100, batch 339/625 completed (loss: 0.0002:  54%|█████▍    | 170/312 [17:09<14:20,  6.06s/it]
Training Epoch: 19/100, batch 339/625 completed (loss: 0.0192:  54%|█████▍    | 170/312 [17:09<14:20,  6.06s/it]
Training Epoch: 19/100, batch 340/625 completed (loss: 0.0005:  54%|█████▍    | 170/312 [17:12<14:20,  6.06s/it]
Training Epoch: 19/100, batch 340/625 completed (loss: 0.0001:  54%|█████▍    | 170/312 [17:12<14:20,  6.06s/it]
Training Epoch: 19/100, batch 340/625 completed (loss: 0.0023:  54%|█████▍    | 170/312 [17:12<14:20,  6.06s/it]
Training Epoch: 19/100, batch 340/625 completed (loss: 0.0156:  54%|█████▍    | 170/312 [17:12<14:20,  6.06s/it]
Training Epoch: 19/100, batch 340/625 completed (loss: 0.0005:  55%|█████▍    | 171/312 [17:15<14:19,  6.10s/it]
Training Epoch: 19/100, batch 340/625 completed (loss: 0.0001:  55%|█████▍    | 171/312 [17:15<14:19,  6.10s/it]
Training Epoch: 19/100, batch 340/625 completed (loss: 0.0023:  55%|█████▍    | 171/312 [17:15<14:19,  6.10s/it]
Training Epoch: 19/100, batch 340/625 completed (loss: 0.0156:  55%|█████▍    | 171/312 [17:15<14:20,  6.10s/it]
Training Epoch: 19/100, batch 341/625 completed (loss: 0.0002:  55%|█████▍    | 171/312 [17:15<14:19,  6.10s/it]
Training Epoch: 19/100, batch 341/625 completed (loss: 0.0004:  55%|█████▍    | 171/312 [17:15<14:19,  6.10s/it]
Training Epoch: 19/100, batch 341/625 completed (loss: 0.0006:  55%|█████▍    | 171/312 [17:15<14:19,  6.10s/it]
Training Epoch: 19/100, batch 341/625 completed (loss: 0.0:  55%|█████▍    | 171/312 [17:15<14:20,  6.10s/it]   
Training Epoch: 19/100, batch 342/625 completed (loss: 0.0001:  55%|█████▍    | 171/312 [17:18<14:19,  6.10s/it]
Training Epoch: 19/100, batch 342/625 completed (loss: 0.0001:  55%|█████▍    | 171/312 [17:18<14:19,  6.10s/it]
Training Epoch: 19/100, batch 342/625 completed (loss: 0.0004:  55%|█████▍    | 171/312 [17:18<14:19,  6.10s/it]
Training Epoch: 19/100, batch 342/625 completed (loss: 0.0011:  55%|█████▍    | 171/312 [17:18<14:20,  6.10s/it]
Training Epoch: 19/100, batch 342/625 completed (loss: 0.0004:  55%|█████▌    | 172/312 [17:21<14:02,  6.02s/it]
Training Epoch: 19/100, batch 342/625 completed (loss: 0.0011:  55%|█████▌    | 172/312 [17:21<14:02,  6.02s/it]
Training Epoch: 19/100, batch 342/625 completed (loss: 0.0001:  55%|█████▌    | 172/312 [17:21<14:03,  6.02s/it]
Training Epoch: 19/100, batch 342/625 completed (loss: 0.0001:  55%|█████▌    | 172/312 [17:21<14:03,  6.02s/it]
Training Epoch: 19/100, batch 343/625 completed (loss: 0.0019:  55%|█████▌    | 172/312 [17:21<14:02,  6.02s/it]
Training Epoch: 19/100, batch 343/625 completed (loss: 0.0002:  55%|█████▌    | 172/312 [17:21<14:03,  6.02s/it]
Training Epoch: 19/100, batch 343/625 completed (loss: 0.0001:  55%|█████▌    | 172/312 [17:21<14:03,  6.02s/it]
Training Epoch: 19/100, batch 343/625 completed (loss: 0.0013:  55%|█████▌    | 172/312 [17:21<14:02,  6.02s/it]
Training Epoch: 19/100, batch 344/625 completed (loss: 0.0006:  55%|█████▌    | 172/312 [17:24<14:02,  6.02s/it]
Training Epoch: 19/100, batch 344/625 completed (loss: 0.0007:  55%|█████▌    | 172/312 [17:24<14:03,  6.02s/it]
Training Epoch: 19/100, batch 344/625 completed (loss: 0.0074:  55%|█████▌    | 172/312 [17:24<14:03,  6.02s/it]
Training Epoch: 19/100, batch 344/625 completed (loss: 0.0002:  55%|█████▌    | 172/312 [17:24<14:02,  6.02s/it]
Training Epoch: 19/100, batch 344/625 completed (loss: 0.0006:  55%|█████▌    | 173/312 [17:27<14:01,  6.05s/it]
Training Epoch: 19/100, batch 344/625 completed (loss: 0.0007:  55%|█████▌    | 173/312 [17:27<14:00,  6.05s/it]
Training Epoch: 19/100, batch 344/625 completed (loss: 0.0074:  55%|█████▌    | 173/312 [17:27<14:01,  6.05s/it]
Training Epoch: 19/100, batch 344/625 completed (loss: 0.0002:  55%|█████▌    | 173/312 [17:27<14:01,  6.05s/it]
Training Epoch: 19/100, batch 345/625 completed (loss: 0.0001:  55%|█████▌    | 173/312 [17:27<14:00,  6.05s/it]
Training Epoch: 19/100, batch 345/625 completed (loss: 0.0003:  55%|█████▌    | 173/312 [17:27<14:01,  6.05s/it]
Training Epoch: 19/100, batch 345/625 completed (loss: 0.0013:  55%|█████▌    | 173/312 [17:27<14:01,  6.05s/it]
Training Epoch: 19/100, batch 345/625 completed (loss: 0.0127:  55%|█████▌    | 173/312 [17:27<14:01,  6.05s/it]
Training Epoch: 19/100, batch 346/625 completed (loss: 0.0005:  55%|█████▌    | 173/312 [17:30<14:00,  6.05s/it]
Training Epoch: 19/100, batch 346/625 completed (loss: 0.0004:  55%|█████▌    | 173/312 [17:30<14:01,  6.05s/it]
Training Epoch: 19/100, batch 346/625 completed (loss: 0.0002:  55%|█████▌    | 173/312 [17:30<14:01,  6.05s/it]
Training Epoch: 19/100, batch 346/625 completed (loss: 0.003:  55%|█████▌    | 173/312 [17:30<14:01,  6.05s/it] 
Training Epoch: 19/100, batch 346/625 completed (loss: 0.0005:  56%|█████▌    | 174/312 [17:33<14:01,  6.10s/it]
Training Epoch: 19/100, batch 346/625 completed (loss: 0.0004:  56%|█████▌    | 174/312 [17:33<14:01,  6.10s/it]
Training Epoch: 19/100, batch 346/625 completed (loss: 0.0002:  56%|█████▌    | 174/312 [17:33<14:01,  6.10s/it]
Training Epoch: 19/100, batch 346/625 completed (loss: 0.003:  56%|█████▌    | 174/312 [17:33<14:01,  6.10s/it]
Training Epoch: 19/100, batch 347/625 completed (loss: 0.0002:  56%|█████▌    | 174/312 [17:33<14:01,  6.10s/it]
Training Epoch: 19/100, batch 347/625 completed (loss: 0.0004:  56%|█████▌    | 174/312 [17:33<14:01,  6.10s/it]
Training Epoch: 19/100, batch 347/625 completed (loss: 0.0043:  56%|█████▌    | 174/312 [17:33<14:01,  6.10s/it]
Training Epoch: 19/100, batch 347/625 completed (loss: 0.0014:  56%|█████▌    | 174/312 [17:33<14:01,  6.10s/it]
Training Epoch: 19/100, batch 348/625 completed (loss: 0.0005:  56%|█████▌    | 174/312 [17:36<14:01,  6.10s/it]
Training Epoch: 19/100, batch 348/625 completed (loss: 0.0007:  56%|█████▌    | 174/312 [17:36<14:01,  6.10s/it]
Training Epoch: 19/100, batch 348/625 completed (loss: 0.0005:  56%|█████▌    | 174/312 [17:36<14:01,  6.10s/it]
Training Epoch: 19/100, batch 348/625 completed (loss: 0.0011:  56%|█████▌    | 174/312 [17:36<14:01,  6.10s/it]
Training Epoch: 19/100, batch 348/625 completed (loss: 0.0005:  56%|█████▌    | 175/312 [17:39<13:58,  6.12s/it]
Training Epoch: 19/100, batch 348/625 completed (loss: 0.0005:  56%|█████▌    | 175/312 [17:39<13:58,  6.12s/it]
Training Epoch: 19/100, batch 348/625 completed (loss: 0.0007:  56%|█████▌    | 175/312 [17:39<13:58,  6.12s/it]
Training Epoch: 19/100, batch 348/625 completed (loss: 0.0011:  56%|█████▌    | 175/312 [17:40<13:58,  6.12s/it]
Training Epoch: 19/100, batch 349/625 completed (loss: 0.0121:  56%|█████▌    | 175/312 [17:40<13:58,  6.12s/it]
Training Epoch: 19/100, batch 349/625 completed (loss: 0.0017:  56%|█████▌    | 175/312 [17:40<13:58,  6.12s/it]
Training Epoch: 19/100, batch 349/625 completed (loss: 0.0002:  56%|█████▌    | 175/312 [17:40<13:58,  6.12s/it]
Training Epoch: 19/100, batch 349/625 completed (loss: 0.0003:  56%|█████▌    | 175/312 [17:40<13:58,  6.12s/it]
Training Epoch: 19/100, batch 350/625 completed (loss: 0.0056:  56%|█████▌    | 175/312 [17:42<13:58,  6.12s/it]
Training Epoch: 19/100, batch 350/625 completed (loss: 0.0019:  56%|█████▌    | 175/312 [17:42<13:58,  6.12s/it]
Training Epoch: 19/100, batch 350/625 completed (loss: 0.0077:  56%|█████▌    | 175/312 [17:42<13:58,  6.12s/it]
Training Epoch: 19/100, batch 350/625 completed (loss: 0.0007:  56%|█████▌    | 175/312 [17:43<13:58,  6.12s/it]
Training Epoch: 19/100, batch 350/625 completed (loss: 0.0007:  56%|█████▋    | 176/312 [17:46<13:47,  6.09s/it]
Training Epoch: 19/100, batch 350/625 completed (loss: 0.0019:  56%|█████▋    | 176/312 [17:45<13:48,  6.09s/it]
Training Epoch: 19/100, batch 350/625 completed (loss: 0.0056:  56%|█████▋    | 176/312 [17:45<13:48,  6.09s/it]
Training Epoch: 19/100, batch 350/625 completed (loss: 0.0077:  56%|█████▋    | 176/312 [17:45<13:48,  6.09s/it]
Training Epoch: 19/100, batch 351/625 completed (loss: 0.0007:  56%|█████▋    | 176/312 [17:46<13:48,  6.09s/it]
Training Epoch: 19/100, batch 351/625 completed (loss: 0.0004:  56%|█████▋    | 176/312 [17:46<13:48,  6.09s/it]
Training Epoch: 19/100, batch 351/625 completed (loss: 0.0008:  56%|█████▋    | 176/312 [17:46<13:48,  6.09s/it]
Training Epoch: 19/100, batch 351/625 completed (loss: 0.0018:  56%|█████▋    | 176/312 [17:46<13:47,  6.09s/it]
Training Epoch: 19/100, batch 352/625 completed (loss: 0.0046:  56%|█████▋    | 176/312 [17:49<13:48,  6.09s/it]
Training Epoch: 19/100, batch 352/625 completed (loss: 0.0007:  56%|█████▋    | 176/312 [17:49<13:48,  6.09s/it]
Training Epoch: 19/100, batch 352/625 completed (loss: 0.0011:  56%|█████▋    | 176/312 [17:49<13:48,  6.09s/it]
Training Epoch: 19/100, batch 352/625 completed (loss: 0.0:  56%|█████▋    | 176/312 [17:49<13:47,  6.09s/it]   
Training Epoch: 19/100, batch 352/625 completed (loss: 0.0046:  57%|█████▋    | 177/312 [17:52<13:38,  6.07s/it]
Training Epoch: 19/100, batch 352/625 completed (loss: 0.0:  57%|█████▋    | 177/312 [17:52<13:39,  6.07s/it]
Training Epoch: 19/100, batch 352/625 completed (loss: 0.0011:  57%|█████▋    | 177/312 [17:52<13:39,  6.07s/it]
Training Epoch: 19/100, batch 352/625 completed (loss: 0.0007:  57%|█████▋    | 177/312 [17:52<13:39,  6.07s/it]
Training Epoch: 19/100, batch 353/625 completed (loss: 0.0018:  57%|█████▋    | 177/312 [17:52<13:39,  6.07s/it]
Training Epoch: 19/100, batch 353/625 completed (loss: 0.0005:  57%|█████▋    | 177/312 [17:52<13:39,  6.07s/it]
Training Epoch: 19/100, batch 353/625 completed (loss: 0.0008:  57%|█████▋    | 177/312 [17:52<13:38,  6.07s/it]
Training Epoch: 19/100, batch 353/625 completed (loss: 0.0005:  57%|█████▋    | 177/312 [17:52<13:39,  6.07s/it]
Training Epoch: 19/100, batch 354/625 completed (loss: 0.0227:  57%|█████▋    | 177/312 [17:55<13:39,  6.07s/it]
Training Epoch: 19/100, batch 354/625 completed (loss: 0.0:  57%|█████▋    | 177/312 [17:55<13:38,  6.07s/it]   
Training Epoch: 19/100, batch 354/625 completed (loss: 0.0001:  57%|█████▋    | 177/312 [17:55<13:39,  6.07s/it]
Training Epoch: 19/100, batch 354/625 completed (loss: 0.0007:  57%|█████▋    | 177/312 [17:55<13:39,  6.07s/it]
Training Epoch: 19/100, batch 354/625 completed (loss: 0.0:  57%|█████▋    | 178/312 [17:58<13:36,  6.09s/it]
Training Epoch: 19/100, batch 354/625 completed (loss: 0.0227:  57%|█████▋    | 178/312 [17:58<13:36,  6.09s/it]
Training Epoch: 19/100, batch 354/625 completed (loss: 0.0001:  57%|█████▋    | 178/312 [17:58<13:36,  6.09s/it]
Training Epoch: 19/100, batch 354/625 completed (loss: 0.0007:  57%|█████▋    | 178/312 [17:58<13:36,  6.09s/it]
Training Epoch: 19/100, batch 355/625 completed (loss: 0.0001:  57%|█████▋    | 178/312 [17:58<13:36,  6.09s/it]
Training Epoch: 19/100, batch 355/625 completed (loss: 0.0002:  57%|█████▋    | 178/312 [17:58<13:36,  6.09s/it]
Training Epoch: 19/100, batch 355/625 completed (loss: 0.0004:  57%|█████▋    | 178/312 [17:58<13:36,  6.09s/it]
Training Epoch: 19/100, batch 355/625 completed (loss: 0.0002:  57%|█████▋    | 178/312 [17:58<13:36,  6.09s/it]
Training Epoch: 19/100, batch 356/625 completed (loss: 0.0002:  57%|█████▋    | 178/312 [18:01<13:36,  6.09s/it]
Training Epoch: 19/100, batch 356/625 completed (loss: 0.0001:  57%|█████▋    | 178/312 [18:01<13:36,  6.09s/it]
Training Epoch: 19/100, batch 356/625 completed (loss: 0.0001:  57%|█████▋    | 178/312 [18:01<13:36,  6.09s/it]
Training Epoch: 19/100, batch 356/625 completed (loss: 0.0005:  57%|█████▋    | 178/312 [18:01<13:36,  6.09s/it]
Training Epoch: 19/100, batch 356/625 completed (loss: 0.0002:  57%|█████▋    | 179/312 [18:04<13:32,  6.11s/it]
Training Epoch: 19/100, batch 356/625 completed (loss: 0.0001:  57%|█████▋    | 179/312 [18:04<13:33,  6.11s/it]
Training Epoch: 19/100, batch 356/625 completed (loss: 0.0001:  57%|█████▋    | 179/312 [18:04<13:33,  6.11s/it]
Training Epoch: 19/100, batch 356/625 completed (loss: 0.0005:  57%|█████▋    | 179/312 [18:04<13:33,  6.11s/it]
Training Epoch: 19/100, batch 357/625 completed (loss: 0.0002:  57%|█████▋    | 179/312 [18:04<13:32,  6.11s/it]
Training Epoch: 19/100, batch 357/625 completed (loss: 0.0001:  57%|█████▋    | 179/312 [18:04<13:33,  6.11s/it]
Training Epoch: 19/100, batch 357/625 completed (loss: 0.01:  57%|█████▋    | 179/312 [18:04<13:33,  6.11s/it]  
Training Epoch: 19/100, batch 357/625 completed (loss: 0.0001:  57%|█████▋    | 179/312 [18:04<13:33,  6.11s/it]
Training Epoch: 19/100, batch 358/625 completed (loss: 0.0017:  57%|█████▋    | 179/312 [18:07<13:33,  6.11s/it]
Training Epoch: 19/100, batch 358/625 completed (loss: 0.0005:  57%|█████▋    | 179/312 [18:07<13:32,  6.11s/it]
Training Epoch: 19/100, batch 358/625 completed (loss: 0.0012:  57%|█████▋    | 179/312 [18:07<13:33,  6.11s/it]
Training Epoch: 19/100, batch 358/625 completed (loss: 0.0002:  57%|█████▋    | 179/312 [18:07<13:33,  6.11s/it]
Training Epoch: 19/100, batch 358/625 completed (loss: 0.0012:  58%|█████▊    | 180/312 [18:10<13:21,  6.08s/it]
Training Epoch: 19/100, batch 358/625 completed (loss: 0.0005:  58%|█████▊    | 180/312 [18:10<13:21,  6.07s/it]
Training Epoch: 19/100, batch 358/625 completed (loss: 0.0017:  58%|█████▊    | 180/312 [18:10<13:21,  6.07s/it]
Training Epoch: 19/100, batch 358/625 completed (loss: 0.0002:  58%|█████▊    | 180/312 [18:10<13:22,  6.08s/it]
Training Epoch: 19/100, batch 359/625 completed (loss: 0.0009:  58%|█████▊    | 180/312 [18:10<13:21,  6.07s/it]
Training Epoch: 19/100, batch 359/625 completed (loss: 0.0149:  58%|█████▊    | 180/312 [18:10<13:21,  6.07s/it]
Training Epoch: 19/100, batch 359/625 completed (loss: 0.0004:  58%|█████▊    | 180/312 [18:10<13:21,  6.08s/it]
Training Epoch: 19/100, batch 359/625 completed (loss: 0.0002:  58%|█████▊    | 180/312 [18:10<13:22,  6.08s/it]
Training Epoch: 19/100, batch 360/625 completed (loss: 0.0027:  58%|█████▊    | 180/312 [18:13<13:21,  6.07s/it]
Training Epoch: 19/100, batch 360/625 completed (loss: 0.0004:  58%|█████▊    | 180/312 [18:13<13:21,  6.07s/it]
Training Epoch: 19/100, batch 360/625 completed (loss: 0.0002:  58%|█████▊    | 180/312 [18:13<13:21,  6.08s/it]
Training Epoch: 19/100, batch 360/625 completed (loss: 0.0099:  58%|█████▊    | 180/312 [18:13<13:22,  6.08s/it]
Training Epoch: 19/100, batch 360/625 completed (loss: 0.0027:  58%|█████▊    | 181/312 [18:16<13:13,  6.06s/it]
Training Epoch: 19/100, batch 360/625 completed (loss: 0.0004:  58%|█████▊    | 181/312 [18:16<13:13,  6.06s/it]
Training Epoch: 19/100, batch 360/625 completed (loss: 0.0002:  58%|█████▊    | 181/312 [18:16<13:13,  6.06s/it]
Training Epoch: 19/100, batch 360/625 completed (loss: 0.0099:  58%|█████▊    | 181/312 [18:16<13:13,  6.06s/it]
Training Epoch: 19/100, batch 361/625 completed (loss: 0.0008:  58%|█████▊    | 181/312 [18:16<13:13,  6.06s/it]
Training Epoch: 19/100, batch 361/625 completed (loss: 0.0017:  58%|█████▊    | 181/312 [18:16<13:13,  6.06s/it]
Training Epoch: 19/100, batch 361/625 completed (loss: 0.0003:  58%|█████▊    | 181/312 [18:16<13:13,  6.06s/it]
Training Epoch: 19/100, batch 361/625 completed (loss: 0.0002:  58%|█████▊    | 181/312 [18:16<13:13,  6.06s/it]
Training Epoch: 19/100, batch 362/625 completed (loss: 0.0002:  58%|█████▊    | 181/312 [18:19<13:13,  6.06s/it]
Training Epoch: 19/100, batch 362/625 completed (loss: 0.0036:  58%|█████▊    | 181/312 [18:19<13:13,  6.06s/it]
Training Epoch: 19/100, batch 362/625 completed (loss: 0.0031:  58%|█████▊    | 181/312 [18:19<13:13,  6.06s/it]
Training Epoch: 19/100, batch 362/625 completed (loss: 0.0003:  58%|█████▊    | 181/312 [18:19<13:13,  6.06s/it]
Training Epoch: 19/100, batch 362/625 completed (loss: 0.0002:  58%|█████▊    | 182/312 [18:22<13:12,  6.09s/it]
Training Epoch: 19/100, batch 362/625 completed (loss: 0.0036:  58%|█████▊    | 182/312 [18:22<13:12,  6.09s/it]
Training Epoch: 19/100, batch 362/625 completed (loss: 0.0031:  58%|█████▊    | 182/312 [18:22<13:12,  6.10s/it]
Training Epoch: 19/100, batch 362/625 completed (loss: 0.0003:  58%|█████▊    | 182/312 [18:22<13:12,  6.10s/it]
Training Epoch: 19/100, batch 363/625 completed (loss: 0.0001:  58%|█████▊    | 182/312 [18:22<13:12,  6.09s/it]
Training Epoch: 19/100, batch 363/625 completed (loss: 0.0004:  58%|█████▊    | 182/312 [18:22<13:12,  6.10s/it]
Training Epoch: 19/100, batch 363/625 completed (loss: 0.0076:  58%|█████▊    | 182/312 [18:22<13:12,  6.09s/it]
Training Epoch: 19/100, batch 363/625 completed (loss: 0.0002:  58%|█████▊    | 182/312 [18:22<13:12,  6.10s/it]
Training Epoch: 19/100, batch 364/625 completed (loss: 0.001:  58%|█████▊    | 182/312 [18:25<13:12,  6.09s/it] 
Training Epoch: 19/100, batch 364/625 completed (loss: 0.0031:  58%|█████▊    | 182/312 [18:25<13:12,  6.09s/it]
Training Epoch: 19/100, batch 364/625 completed (loss: 0.001:  58%|█████▊    | 182/312 [18:25<13:12,  6.10s/it] 
Training Epoch: 19/100, batch 364/625 completed (loss: 0.026:  58%|█████▊    | 182/312 [18:25<13:12,  6.10s/it] 
Training Epoch: 19/100, batch 364/625 completed (loss: 0.0031:  59%|█████▊    | 183/312 [18:28<13:07,  6.10s/it]
Training Epoch: 19/100, batch 364/625 completed (loss: 0.001:  59%|█████▊    | 183/312 [18:28<13:07,  6.10s/it]
Training Epoch: 19/100, batch 364/625 completed (loss: 0.001:  59%|█████▊    | 183/312 [18:28<13:07,  6.10s/it]
Training Epoch: 19/100, batch 364/625 completed (loss: 0.026:  59%|█████▊    | 183/312 [18:28<13:07,  6.10s/it]
Training Epoch: 19/100, batch 365/625 completed (loss: 0.0002:  59%|█████▊    | 183/312 [18:28<13:07,  6.10s/it]
Training Epoch: 19/100, batch 365/625 completed (loss: 0.0017:  59%|█████▊    | 183/312 [18:28<13:07,  6.10s/it]
Training Epoch: 19/100, batch 365/625 completed (loss: 0.0002:  59%|█████▊    | 183/312 [18:28<13:07,  6.10s/it]
Training Epoch: 19/100, batch 365/625 completed (loss: 0.0034:  59%|█████▊    | 183/312 [18:28<13:07,  6.10s/it]
Training Epoch: 19/100, batch 366/625 completed (loss: 0.0003:  59%|█████▊    | 183/312 [18:31<13:07,  6.10s/it]
Training Epoch: 19/100, batch 366/625 completed (loss: 0.0003:  59%|█████▊    | 183/312 [18:31<13:07,  6.10s/it]
Training Epoch: 19/100, batch 366/625 completed (loss: 0.0:  59%|█████▊    | 183/312 [18:31<13:07,  6.10s/it]   
Training Epoch: 19/100, batch 366/625 completed (loss: 0.002:  59%|█████▊    | 183/312 [18:31<13:07,  6.10s/it] 
Training Epoch: 19/100, batch 366/625 completed (loss: 0.0003:  59%|█████▉    | 184/312 [18:34<12:57,  6.08s/it]
Training Epoch: 19/100, batch 366/625 completed (loss: 0.0003:  59%|█████▉    | 184/312 [18:34<12:57,  6.08s/it]
Training Epoch: 19/100, batch 366/625 completed (loss: 0.0:  59%|█████▉    | 184/312 [18:34<12:58,  6.08s/it]
Training Epoch: 19/100, batch 366/625 completed (loss: 0.002:  59%|█████▉    | 184/312 [18:34<12:57,  6.08s/it]
Training Epoch: 19/100, batch 367/625 completed (loss: 0.0012:  59%|█████▉    | 184/312 [18:34<12:57,  6.08s/it]
Training Epoch: 19/100, batch 367/625 completed (loss: 0.042:  59%|█████▉    | 184/312 [18:34<12:58,  6.08s/it]
Training Epoch: 19/100, batch 367/625 completed (loss: 0.001:  59%|█████▉    | 184/312 [18:34<12:57,  6.08s/it] 
Training Epoch: 19/100, batch 367/625 completed (loss: 0.0079:  59%|█████▉    | 184/312 [18:34<12:57,  6.08s/it]
Training Epoch: 19/100, batch 368/625 completed (loss: 0.0001:  59%|█████▉    | 184/312 [18:37<12:58,  6.08s/it]
Training Epoch: 19/100, batch 368/625 completed (loss: 0.0003:  59%|█████▉    | 184/312 [18:37<12:57,  6.08s/it]
Training Epoch: 19/100, batch 368/625 completed (loss: 0.0001:  59%|█████▉    | 184/312 [18:37<12:57,  6.08s/it]
Training Epoch: 19/100, batch 368/625 completed (loss: 0.0006:  59%|█████▉    | 184/312 [18:37<12:57,  6.08s/it]
Training Epoch: 19/100, batch 368/625 completed (loss: 0.0001:  59%|█████▉    | 185/312 [18:40<12:49,  6.06s/it]
Training Epoch: 19/100, batch 368/625 completed (loss: 0.0003:  59%|█████▉    | 185/312 [18:40<12:49,  6.06s/it]
Training Epoch: 19/100, batch 368/625 completed (loss: 0.0001:  59%|█████▉    | 185/312 [18:40<12:49,  6.06s/it]
Training Epoch: 19/100, batch 368/625 completed (loss: 0.0006:  59%|█████▉    | 185/312 [18:40<12:49,  6.06s/it]
Training Epoch: 19/100, batch 369/625 completed (loss: 0.0004:  59%|█████▉    | 185/312 [18:40<12:49,  6.06s/it]
Training Epoch: 19/100, batch 369/625 completed (loss: 0.0004:  59%|█████▉    | 185/312 [18:40<12:49,  6.06s/it]
Training Epoch: 19/100, batch 369/625 completed (loss: 0.0014:  59%|█████▉    | 185/312 [18:40<12:49,  6.06s/it]
Training Epoch: 19/100, batch 369/625 completed (loss: 0.0003:  59%|█████▉    | 185/312 [18:40<12:49,  6.06s/it]
Training Epoch: 19/100, batch 370/625 completed (loss: 0.0024:  59%|█████▉    | 185/312 [18:43<12:49,  6.06s/it]
Training Epoch: 19/100, batch 370/625 completed (loss: 0.0025:  59%|█████▉    | 185/312 [18:43<12:49,  6.06s/it]
Training Epoch: 19/100, batch 370/625 completed (loss: 0.0009:  59%|█████▉    | 185/312 [18:43<12:49,  6.06s/it]
Training Epoch: 19/100, batch 370/625 completed (loss: 0.0001:  59%|█████▉    | 185/312 [18:44<12:49,  6.06s/it]
Training Epoch: 19/100, batch 370/625 completed (loss: 0.0001:  60%|█████▉    | 186/312 [18:47<12:53,  6.14s/it]
Training Epoch: 19/100, batch 370/625 completed (loss: 0.0009:  60%|█████▉    | 186/312 [18:47<12:53,  6.14s/it]
Training Epoch: 19/100, batch 370/625 completed (loss: 0.0025:  60%|█████▉    | 186/312 [18:46<12:54,  6.14s/it]
Training Epoch: 19/100, batch 370/625 completed (loss: 0.0024:  60%|█████▉    | 186/312 [18:47<12:53,  6.14s/it]
Training Epoch: 19/100, batch 371/625 completed (loss: 0.0002:  60%|█████▉    | 186/312 [18:47<12:53,  6.14s/it]
Training Epoch: 19/100, batch 371/625 completed (loss: 0.004:  60%|█████▉    | 186/312 [18:47<12:53,  6.14s/it] 
Training Epoch: 19/100, batch 371/625 completed (loss: 0.0304:  60%|█████▉    | 186/312 [18:47<12:54,  6.14s/it]
Training Epoch: 19/100, batch 371/625 completed (loss: 0.0004:  60%|█████▉    | 186/312 [18:47<12:53,  6.14s/it]
Training Epoch: 19/100, batch 372/625 completed (loss: 0.0001:  60%|█████▉    | 186/312 [18:49<12:54,  6.14s/it]
Training Epoch: 19/100, batch 372/625 completed (loss: 0.0056:  60%|█████▉    | 186/312 [18:50<12:53,  6.14s/it]
Training Epoch: 19/100, batch 372/625 completed (loss: 0.0001:  60%|█████▉    | 186/312 [18:50<12:53,  6.14s/it]
Training Epoch: 19/100, batch 372/625 completed (loss: 0.0007:  60%|█████▉    | 186/312 [18:50<12:53,  6.14s/it]
Training Epoch: 19/100, batch 372/625 completed (loss: 0.0001:  60%|█████▉    | 187/312 [18:52<12:37,  6.06s/it]
Training Epoch: 19/100, batch 372/625 completed (loss: 0.0001:  60%|█████▉    | 187/312 [18:52<12:37,  6.06s/it]
Training Epoch: 19/100, batch 372/625 completed (loss: 0.0056:  60%|█████▉    | 187/312 [18:52<12:37,  6.06s/it]
Training Epoch: 19/100, batch 372/625 completed (loss: 0.0007:  60%|█████▉    | 187/312 [18:52<12:38,  6.07s/it]
Training Epoch: 19/100, batch 373/625 completed (loss: 0.0003:  60%|█████▉    | 187/312 [18:52<12:37,  6.06s/it]
Training Epoch: 19/100, batch 373/625 completed (loss: 0.0001:  60%|█████▉    | 187/312 [18:52<12:37,  6.06s/it]
Training Epoch: 19/100, batch 373/625 completed (loss: 0.0013:  60%|█████▉    | 187/312 [18:52<12:37,  6.06s/it]
Training Epoch: 19/100, batch 373/625 completed (loss: 0.0002:  60%|█████▉    | 187/312 [18:52<12:38,  6.07s/it]
Training Epoch: 19/100, batch 374/625 completed (loss: 0.0005:  60%|█████▉    | 187/312 [18:56<12:37,  6.06s/it]
Training Epoch: 19/100, batch 374/625 completed (loss: 0.0002:  60%|█████▉    | 187/312 [18:56<12:37,  6.06s/it]
Training Epoch: 19/100, batch 374/625 completed (loss: 0.0014:  60%|█████▉    | 187/312 [18:56<12:37,  6.06s/it]
Training Epoch: 19/100, batch 374/625 completed (loss: 0.0003:  60%|█████▉    | 187/312 [18:56<12:38,  6.07s/it]
Training Epoch: 19/100, batch 374/625 completed (loss: 0.0005:  60%|██████    | 188/312 [18:59<12:41,  6.15s/it]
Training Epoch: 19/100, batch 374/625 completed (loss: 0.0014:  60%|██████    | 188/312 [18:59<12:41,  6.14s/it]
Training Epoch: 19/100, batch 374/625 completed (loss: 0.0002:  60%|██████    | 188/312 [18:59<12:42,  6.15s/it]
Training Epoch: 19/100, batch 374/625 completed (loss: 0.0003:  60%|██████    | 188/312 [18:59<12:42,  6.15s/it]
Training Epoch: 19/100, batch 375/625 completed (loss: 0.0004:  60%|██████    | 188/312 [18:59<12:41,  6.15s/it]
Training Epoch: 19/100, batch 375/625 completed (loss: 0.0038:  60%|██████    | 188/312 [18:59<12:42,  6.15s/it]
Training Epoch: 19/100, batch 375/625 completed (loss: 0.0001:  60%|██████    | 188/312 [18:59<12:41,  6.14s/it]
Training Epoch: 19/100, batch 375/625 completed (loss: 0.0018:  60%|██████    | 188/312 [18:59<12:42,  6.15s/it]
Training Epoch: 19/100, batch 376/625 completed (loss: 0.0014:  60%|██████    | 188/312 [19:02<12:41,  6.14s/it]
Training Epoch: 19/100, batch 376/625 completed (loss: 0.0015:  60%|██████    | 188/312 [19:02<12:41,  6.15s/it]
Training Epoch: 19/100, batch 376/625 completed (loss: 0.0039:  60%|██████    | 188/312 [19:02<12:42,  6.15s/it]
Training Epoch: 19/100, batch 376/625 completed (loss: 0.0006:  60%|██████    | 188/312 [19:02<12:42,  6.15s/it]
Training Epoch: 19/100, batch 376/625 completed (loss: 0.0014:  61%|██████    | 189/312 [19:05<12:30,  6.10s/it]
Training Epoch: 19/100, batch 376/625 completed (loss: 0.0039:  61%|██████    | 189/312 [19:05<12:30,  6.10s/it]
Training Epoch: 19/100, batch 376/625 completed (loss: 0.0015:  61%|██████    | 189/312 [19:05<12:30,  6.10s/it]
Training Epoch: 19/100, batch 376/625 completed (loss: 0.0006:  61%|██████    | 189/312 [19:05<12:30,  6.10s/it]
Training Epoch: 19/100, batch 377/625 completed (loss: 0.0001:  61%|██████    | 189/312 [19:05<12:30,  6.10s/it]
Training Epoch: 19/100, batch 377/625 completed (loss: 0.0007:  61%|██████    | 189/312 [19:05<12:30,  6.10s/it]
Training Epoch: 19/100, batch 377/625 completed (loss: 0.0006:  61%|██████    | 189/312 [19:05<12:30,  6.10s/it]
Training Epoch: 19/100, batch 377/625 completed (loss: 0.0012:  61%|██████    | 189/312 [19:05<12:30,  6.10s/it]
Training Epoch: 19/100, batch 378/625 completed (loss: 0.0005:  61%|██████    | 189/312 [19:08<12:30,  6.10s/it]
Training Epoch: 19/100, batch 378/625 completed (loss: 0.0008:  61%|██████    | 189/312 [19:08<12:30,  6.10s/it]
Training Epoch: 19/100, batch 378/625 completed (loss: 0.0009:  61%|██████    | 189/312 [19:08<12:30,  6.10s/it]
Training Epoch: 19/100, batch 378/625 completed (loss: 0.0029:  61%|██████    | 189/312 [19:08<12:30,  6.10s/it]
Training Epoch: 19/100, batch 378/625 completed (loss: 0.0005:  61%|██████    | 190/312 [19:11<12:21,  6.08s/it]
Training Epoch: 19/100, batch 378/625 completed (loss: 0.0008:  61%|██████    | 190/312 [19:11<12:21,  6.08s/it]
Training Epoch: 19/100, batch 378/625 completed (loss: 0.0009:  61%|██████    | 190/312 [19:11<12:21,  6.08s/it]
Training Epoch: 19/100, batch 378/625 completed (loss: 0.0029:  61%|██████    | 190/312 [19:11<12:21,  6.08s/it]
Training Epoch: 19/100, batch 379/625 completed (loss: 0.0091:  61%|██████    | 190/312 [19:11<12:21,  6.08s/it]
Training Epoch: 19/100, batch 379/625 completed (loss: 0.0006:  61%|██████    | 190/312 [19:11<12:21,  6.08s/it]
Training Epoch: 19/100, batch 379/625 completed (loss: 0.0097:  61%|██████    | 190/312 [19:11<12:21,  6.08s/it]
Training Epoch: 19/100, batch 379/625 completed (loss: 0.0002:  61%|██████    | 190/312 [19:11<12:21,  6.08s/it]
Training Epoch: 19/100, batch 380/625 completed (loss: 0.0003:  61%|██████    | 190/312 [19:14<12:21,  6.08s/it]
Training Epoch: 19/100, batch 380/625 completed (loss: 0.0002:  61%|██████    | 190/312 [19:14<12:21,  6.08s/it]
Training Epoch: 19/100, batch 380/625 completed (loss: 0.0001:  61%|██████    | 190/312 [19:14<12:21,  6.08s/it]
Training Epoch: 19/100, batch 380/625 completed (loss: 0.0005:  61%|██████    | 190/312 [19:14<12:21,  6.08s/it]
Training Epoch: 19/100, batch 380/625 completed (loss: 0.0001:  61%|██████    | 191/312 [19:17<12:17,  6.10s/it]
Training Epoch: 19/100, batch 380/625 completed (loss: 0.0003:  61%|██████    | 191/312 [19:17<12:17,  6.10s/it]
Training Epoch: 19/100, batch 380/625 completed (loss: 0.0002:  61%|██████    | 191/312 [19:17<12:17,  6.10s/it]
Training Epoch: 19/100, batch 380/625 completed (loss: 0.0005:  61%|██████    | 191/312 [19:17<12:18,  6.10s/it]
Training Epoch: 19/100, batch 381/625 completed (loss: 0.0017:  61%|██████    | 191/312 [19:17<12:17,  6.10s/it]
Training Epoch: 19/100, batch 381/625 completed (loss: 0.0021:  61%|██████    | 191/312 [19:17<12:17,  6.10s/it]
Training Epoch: 19/100, batch 381/625 completed (loss: 0.0008:  61%|██████    | 191/312 [19:17<12:17,  6.10s/it]
Training Epoch: 19/100, batch 381/625 completed (loss: 0.0005:  61%|██████    | 191/312 [19:17<12:18,  6.10s/it]
Training Epoch: 19/100, batch 382/625 completed (loss: 0.0035:  61%|██████    | 191/312 [19:20<12:17,  6.10s/it]
Training Epoch: 19/100, batch 382/625 completed (loss: 0.0138:  61%|██████    | 191/312 [19:20<12:17,  6.10s/it]
Training Epoch: 19/100, batch 382/625 completed (loss: 0.0009:  61%|██████    | 191/312 [19:20<12:17,  6.10s/it]
Training Epoch: 19/100, batch 382/625 completed (loss: 0.0001:  61%|██████    | 191/312 [19:20<12:18,  6.10s/it]
Training Epoch: 19/100, batch 382/625 completed (loss: 0.0009:  62%|██████▏   | 192/312 [19:23<12:08,  6.07s/it]
Training Epoch: 19/100, batch 382/625 completed (loss: 0.0035:  62%|██████▏   | 192/312 [19:23<12:08,  6.07s/it]
Training Epoch: 19/100, batch 382/625 completed (loss: 0.0138:  62%|██████▏   | 192/312 [19:23<12:08,  6.07s/it]
Training Epoch: 19/100, batch 382/625 completed (loss: 0.0001:  62%|██████▏   | 192/312 [19:23<12:08,  6.07s/it]
Training Epoch: 19/100, batch 383/625 completed (loss: 0.0001:  62%|██████▏   | 192/312 [19:23<12:08,  6.07s/it]
Training Epoch: 19/100, batch 383/625 completed (loss: 0.0006:  62%|██████▏   | 192/312 [19:23<12:08,  6.07s/it]
Training Epoch: 19/100, batch 383/625 completed (loss: 0.0017:  62%|██████▏   | 192/312 [19:23<12:08,  6.07s/it]
Training Epoch: 19/100, batch 383/625 completed (loss: 0.001:  62%|██████▏   | 192/312 [19:23<12:08,  6.07s/it] 
Training Epoch: 19/100, batch 384/625 completed (loss: 0.001:  62%|██████▏   | 192/312 [19:26<12:08,  6.07s/it] 
Training Epoch: 19/100, batch 384/625 completed (loss: 0.0014:  62%|██████▏   | 192/312 [19:26<12:08,  6.07s/it]
Training Epoch: 19/100, batch 384/625 completed (loss: 0.0028:  62%|██████▏   | 192/312 [19:26<12:08,  6.07s/it]
Training Epoch: 19/100, batch 384/625 completed (loss: 0.006:  62%|██████▏   | 192/312 [19:26<12:08,  6.07s/it]
Training Epoch: 19/100, batch 384/625 completed (loss: 0.001:  62%|██████▏   | 193/312 [19:29<12:10,  6.14s/it]
Training Epoch: 19/100, batch 384/625 completed (loss: 0.0028:  62%|██████▏   | 193/312 [19:29<12:10,  6.14s/it]
Training Epoch: 19/100, batch 384/625 completed (loss: 0.0014:  62%|██████▏   | 193/312 [19:29<12:10,  6.14s/it]
Training Epoch: 19/100, batch 384/625 completed (loss: 0.006:  62%|██████▏   | 193/312 [19:29<12:10,  6.14s/it]
Training Epoch: 19/100, batch 385/625 completed (loss: 0.0014:  62%|██████▏   | 193/312 [19:29<12:10,  6.14s/it]
Training Epoch: 19/100, batch 385/625 completed (loss: 0.0014:  62%|██████▏   | 193/312 [19:29<12:10,  6.14s/it]
Training Epoch: 19/100, batch 385/625 completed (loss: 0.0:  62%|██████▏   | 193/312 [19:29<12:10,  6.14s/it]   
Training Epoch: 19/100, batch 385/625 completed (loss: 0.001:  62%|██████▏   | 193/312 [19:29<12:10,  6.14s/it]
Training Epoch: 19/100, batch 386/625 completed (loss: 0.0004:  62%|██████▏   | 193/312 [19:32<12:10,  6.14s/it]
Training Epoch: 19/100, batch 386/625 completed (loss: 0.0003:  62%|██████▏   | 193/312 [19:32<12:10,  6.14s/it]
Training Epoch: 19/100, batch 386/625 completed (loss: 0.0002:  62%|██████▏   | 193/312 [19:32<12:10,  6.14s/it]
Training Epoch: 19/100, batch 386/625 completed (loss: 0.0251:  62%|██████▏   | 193/312 [19:32<12:10,  6.14s/it]
Training Epoch: 19/100, batch 386/625 completed (loss: 0.0003:  62%|██████▏   | 194/312 [19:35<12:00,  6.10s/it]
Training Epoch: 19/100, batch 386/625 completed (loss: 0.0004:  62%|██████▏   | 194/312 [19:35<12:00,  6.10s/it]
Training Epoch: 19/100, batch 386/625 completed (loss: 0.0002:  62%|██████▏   | 194/312 [19:35<12:00,  6.10s/it]
Training Epoch: 19/100, batch 386/625 completed (loss: 0.0251:  62%|██████▏   | 194/312 [19:35<12:00,  6.10s/it]
Training Epoch: 19/100, batch 387/625 completed (loss: 0.0001:  62%|██████▏   | 194/312 [19:35<12:00,  6.10s/it]
Training Epoch: 19/100, batch 387/625 completed (loss: 0.001:  62%|██████▏   | 194/312 [19:35<12:00,  6.10s/it] 
Training Epoch: 19/100, batch 387/625 completed (loss: 0.0025:  62%|██████▏   | 194/312 [19:35<12:00,  6.10s/it]
Training Epoch: 19/100, batch 387/625 completed (loss: 0.0004:  62%|██████▏   | 194/312 [19:35<12:00,  6.10s/it]
Training Epoch: 19/100, batch 388/625 completed (loss: 0.0002:  62%|██████▏   | 194/312 [19:38<12:00,  6.10s/it]
Training Epoch: 19/100, batch 388/625 completed (loss: 0.0002:  62%|██████▏   | 194/312 [19:38<12:00,  6.10s/it]
Training Epoch: 19/100, batch 388/625 completed (loss: 0.0002:  62%|██████▏   | 194/312 [19:38<12:00,  6.10s/it]
Training Epoch: 19/100, batch 388/625 completed (loss: 0.001:  62%|██████▏   | 194/312 [19:38<12:00,  6.10s/it] 
Training Epoch: 19/100, batch 388/625 completed (loss: 0.0002:  62%|██████▎   | 195/312 [19:41<11:46,  6.03s/it]
Training Epoch: 19/100, batch 388/625 completed (loss: 0.0002:  62%|██████▎   | 195/312 [19:41<11:46,  6.04s/it]
Training Epoch: 19/100, batch 388/625 completed (loss: 0.001:  62%|██████▎   | 195/312 [19:41<11:46,  6.04s/it]
Training Epoch: 19/100, batch 388/625 completed (loss: 0.0002:  62%|██████▎   | 195/312 [19:41<11:46,  6.04s/it]
Training Epoch: 19/100, batch 389/625 completed (loss: 0.0005:  62%|██████▎   | 195/312 [19:41<11:46,  6.03s/it]
Training Epoch: 19/100, batch 389/625 completed (loss: 0.0011:  62%|██████▎   | 195/312 [19:41<11:46,  6.04s/it]
Training Epoch: 19/100, batch 389/625 completed (loss: 0.0004:  62%|██████▎   | 195/312 [19:41<11:46,  6.04s/it]
Training Epoch: 19/100, batch 389/625 completed (loss: 0.0:  62%|██████▎   | 195/312 [19:41<11:46,  6.04s/it]  
Training Epoch: 19/100, batch 390/625 completed (loss: 0.0018:  62%|██████▎   | 195/312 [19:44<11:46,  6.04s/it]
Training Epoch: 19/100, batch 390/625 completed (loss: 0.0001:  62%|██████▎   | 195/312 [19:44<11:46,  6.04s/it]
Training Epoch: 19/100, batch 390/625 completed (loss: 0.0002:  62%|██████▎   | 195/312 [19:44<11:46,  6.03s/it]
Training Epoch: 19/100, batch 390/625 completed (loss: 0.0113:  62%|██████▎   | 195/312 [19:45<11:46,  6.04s/it]
Training Epoch: 19/100, batch 390/625 completed (loss: 0.0001:  63%|██████▎   | 196/312 [19:47<11:47,  6.10s/it]
Training Epoch: 19/100, batch 390/625 completed (loss: 0.0002:  63%|██████▎   | 196/312 [19:47<11:47,  6.10s/it]
Training Epoch: 19/100, batch 390/625 completed (loss: 0.0018:  63%|██████▎   | 196/312 [19:47<11:47,  6.10s/it]
Training Epoch: 19/100, batch 390/625 completed (loss: 0.0113:  63%|██████▎   | 196/312 [19:47<11:47,  6.10s/it]
Training Epoch: 19/100, batch 391/625 completed (loss: 0.0019:  63%|██████▎   | 196/312 [19:47<11:47,  6.10s/it]
Training Epoch: 19/100, batch 391/625 completed (loss: 0.0021:  63%|██████▎   | 196/312 [19:47<11:47,  6.10s/it]
Training Epoch: 19/100, batch 391/625 completed (loss: 0.0005:  63%|██████▎   | 196/312 [19:47<11:47,  6.10s/it]
Training Epoch: 19/100, batch 391/625 completed (loss: 0.0:  63%|██████▎   | 196/312 [19:47<11:47,  6.10s/it]   
Training Epoch: 19/100, batch 392/625 completed (loss: 0.0018:  63%|██████▎   | 196/312 [19:50<11:47,  6.10s/it]
Training Epoch: 19/100, batch 392/625 completed (loss: 0.0149:  63%|██████▎   | 196/312 [19:50<11:47,  6.10s/it]
Training Epoch: 19/100, batch 392/625 completed (loss: 0.0001:  63%|██████▎   | 196/312 [19:50<11:47,  6.10s/it]
Training Epoch: 19/100, batch 392/625 completed (loss: 0.0006:  63%|██████▎   | 196/312 [19:51<11:47,  6.10s/it]
Training Epoch: 19/100, batch 392/625 completed (loss: 0.0018:  63%|██████▎   | 197/312 [19:53<11:37,  6.07s/it]
Training Epoch: 19/100, batch 392/625 completed (loss: 0.0001:  63%|██████▎   | 197/312 [19:53<11:37,  6.06s/it]
Training Epoch: 19/100, batch 392/625 completed (loss: 0.0149:  63%|██████▎   | 197/312 [19:53<11:37,  6.07s/it]
Training Epoch: 19/100, batch 392/625 completed (loss: 0.0006:  63%|██████▎   | 197/312 [19:53<11:37,  6.07s/it]
Training Epoch: 19/100, batch 393/625 completed (loss: 0.0023:  63%|██████▎   | 197/312 [19:53<11:37,  6.07s/it]
Training Epoch: 19/100, batch 393/625 completed (loss: 0.0005:  63%|██████▎   | 197/312 [19:53<11:37,  6.06s/it]
Training Epoch: 19/100, batch 393/625 completed (loss: 0.0:  63%|██████▎   | 197/312 [19:53<11:37,  6.07s/it]   
Training Epoch: 19/100, batch 393/625 completed (loss: 0.0003:  63%|██████▎   | 197/312 [19:53<11:37,  6.07s/it]
Training Epoch: 19/100, batch 394/625 completed (loss: 0.0001:  63%|██████▎   | 197/312 [19:56<11:37,  6.07s/it]
Training Epoch: 19/100, batch 394/625 completed (loss: 0.0005:  63%|██████▎   | 197/312 [19:56<11:37,  6.07s/it]
Training Epoch: 19/100, batch 394/625 completed (loss: 0.0001:  63%|██████▎   | 197/312 [19:56<11:37,  6.06s/it]
Training Epoch: 19/100, batch 394/625 completed (loss: 0.0007:  63%|██████▎   | 197/312 [19:56<11:37,  6.07s/it]
Training Epoch: 19/100, batch 394/625 completed (loss: 0.0007:  63%|██████▎   | 198/312 [19:59<11:29,  6.05s/it]
Training Epoch: 19/100, batch 394/625 completed (loss: 0.0001:  63%|██████▎   | 198/312 [19:59<11:30,  6.06s/it]
Training Epoch: 19/100, batch 394/625 completed (loss: 0.0001:  63%|██████▎   | 198/312 [19:59<11:30,  6.06s/it]
Training Epoch: 19/100, batch 394/625 completed (loss: 0.0005:  63%|██████▎   | 198/312 [19:59<11:30,  6.06s/it]
Training Epoch: 19/100, batch 395/625 completed (loss: 0.0006:  63%|██████▎   | 198/312 [19:59<11:30,  6.06s/it]
Training Epoch: 19/100, batch 395/625 completed (loss: 0.0001:  63%|██████▎   | 198/312 [19:59<11:30,  6.06s/it]
Training Epoch: 19/100, batch 395/625 completed (loss: 0.0002:  63%|██████▎   | 198/312 [19:59<11:30,  6.06s/it]
Training Epoch: 19/100, batch 395/625 completed (loss: 0.0004:  63%|██████▎   | 198/312 [19:59<11:29,  6.05s/it]
Training Epoch: 19/100, batch 396/625 completed (loss: 0.0074:  63%|██████▎   | 198/312 [20:03<11:30,  6.06s/it]
Training Epoch: 19/100, batch 396/625 completed (loss: 0.0012:  63%|██████▎   | 198/312 [20:03<11:30,  6.06s/it]
Training Epoch: 19/100, batch 396/625 completed (loss: 0.004:  63%|██████▎   | 198/312 [20:03<11:30,  6.06s/it] 
Training Epoch: 19/100, batch 396/625 completed (loss: 0.0001:  63%|██████▎   | 198/312 [20:03<11:29,  6.05s/it]
Training Epoch: 19/100, batch 396/625 completed (loss: 0.004:  64%|██████▍   | 199/312 [20:05<11:23,  6.05s/it]
Training Epoch: 19/100, batch 396/625 completed (loss: 0.0074:  64%|██████▍   | 199/312 [20:05<11:23,  6.05s/it]
Training Epoch: 19/100, batch 396/625 completed (loss: 0.0012:  64%|██████▍   | 199/312 [20:05<11:23,  6.05s/it]
Training Epoch: 19/100, batch 396/625 completed (loss: 0.0001:  64%|██████▍   | 199/312 [20:05<11:24,  6.05s/it]
Training Epoch: 19/100, batch 397/625 completed (loss: 0.0001:  64%|██████▍   | 199/312 [20:05<11:23,  6.05s/it]
Training Epoch: 19/100, batch 397/625 completed (loss: 0.0007:  64%|██████▍   | 199/312 [20:05<11:23,  6.05s/it]
Training Epoch: 19/100, batch 397/625 completed (loss: 0.0413:  64%|██████▍   | 199/312 [20:05<11:23,  6.05s/it]
Training Epoch: 19/100, batch 397/625 completed (loss: 0.0021:  64%|██████▍   | 199/312 [20:05<11:24,  6.05s/it]
Training Epoch: 19/100, batch 398/625 completed (loss: 0.0112:  64%|██████▍   | 199/312 [20:09<11:23,  6.05s/it]
Training Epoch: 19/100, batch 398/625 completed (loss: 0.0:  64%|██████▍   | 199/312 [20:09<11:23,  6.05s/it]   
Training Epoch: 19/100, batch 398/625 completed (loss: 0.0008:  64%|██████▍   | 199/312 [20:09<11:23,  6.05s/it]
Training Epoch: 19/100, batch 398/625 completed (loss: 0.0035:  64%|██████▍   | 199/312 [20:09<11:24,  6.05s/it]
Training Epoch: 19/100, batch 398/625 completed (loss: 0.0112:  64%|██████▍   | 200/312 [20:11<11:16,  6.04s/it]
Training Epoch: 19/100, batch 398/625 completed (loss: 0.0:  64%|██████▍   | 200/312 [20:11<11:16,  6.04s/it]
Training Epoch: 19/100, batch 398/625 completed (loss: 0.0035:  64%|██████▍   | 200/312 [20:11<11:16,  6.04s/it]
Training Epoch: 19/100, batch 398/625 completed (loss: 0.0008:  64%|██████▍   | 200/312 [20:11<11:16,  6.04s/it]
Training Epoch: 19/100, batch 399/625 completed (loss: 0.0008:  64%|██████▍   | 200/312 [20:11<11:16,  6.04s/it]
Training Epoch: 19/100, batch 399/625 completed (loss: 0.0001:  64%|██████▍   | 200/312 [20:11<11:16,  6.04s/it]
Training Epoch: 19/100, batch 399/625 completed (loss: 0.0024:  64%|██████▍   | 200/312 [20:11<11:16,  6.04s/it]
Training Epoch: 19/100, batch 399/625 completed (loss: 0.0005:  64%|██████▍   | 200/312 [20:12<11:16,  6.04s/it]
Training Epoch: 19/100, batch 400/625 completed (loss: 0.0012:  64%|██████▍   | 200/312 [20:15<11:16,  6.04s/it]
Training Epoch: 19/100, batch 400/625 completed (loss: 0.0009:  64%|██████▍   | 200/312 [20:15<11:16,  6.04s/it]
Training Epoch: 19/100, batch 400/625 completed (loss: 0.0004:  64%|██████▍   | 200/312 [20:15<11:16,  6.04s/it]
Training Epoch: 19/100, batch 400/625 completed (loss: 0.0009:  64%|██████▍   | 200/312 [20:15<11:16,  6.04s/it]
Training Epoch: 19/100, batch 400/625 completed (loss: 0.0009:  64%|██████▍   | 201/312 [20:18<11:19,  6.13s/it]
Training Epoch: 19/100, batch 400/625 completed (loss: 0.0004:  64%|██████▍   | 201/312 [20:18<11:20,  6.13s/it]
Training Epoch: 19/100, batch 400/625 completed (loss: 0.0012:  64%|██████▍   | 201/312 [20:18<11:20,  6.13s/it]
Training Epoch: 19/100, batch 400/625 completed (loss: 0.0009:  64%|██████▍   | 201/312 [20:18<11:20,  6.13s/it]
Training Epoch: 19/100, batch 401/625 completed (loss: 0.0004:  64%|██████▍   | 201/312 [20:18<11:20,  6.13s/it]
Training Epoch: 19/100, batch 401/625 completed (loss: 0.0023:  64%|██████▍   | 201/312 [20:18<11:20,  6.13s/it]
Training Epoch: 19/100, batch 401/625 completed (loss: 0.0009:  64%|██████▍   | 201/312 [20:18<11:20,  6.13s/it]
Training Epoch: 19/100, batch 401/625 completed (loss: 0.0001:  64%|██████▍   | 201/312 [20:18<11:19,  6.13s/it]
Training Epoch: 19/100, batch 402/625 completed (loss: 0.0005:  64%|██████▍   | 201/312 [20:21<11:20,  6.13s/it]
Training Epoch: 19/100, batch 402/625 completed (loss: 0.0002:  64%|██████▍   | 201/312 [20:21<11:20,  6.13s/it]
Training Epoch: 19/100, batch 402/625 completed (loss: 0.0006:  64%|██████▍   | 201/312 [20:21<11:20,  6.13s/it]
Training Epoch: 19/100, batch 402/625 completed (loss: 0.0053:  64%|██████▍   | 201/312 [20:21<11:19,  6.13s/it]
Training Epoch: 19/100, batch 402/625 completed (loss: 0.0053:  65%|██████▍   | 202/312 [20:24<11:10,  6.09s/it]
Training Epoch: 19/100, batch 402/625 completed (loss: 0.0002:  65%|██████▍   | 202/312 [20:24<11:10,  6.09s/it]
Training Epoch: 19/100, batch 402/625 completed (loss: 0.0006:  65%|██████▍   | 202/312 [20:24<11:10,  6.09s/it]
Training Epoch: 19/100, batch 402/625 completed (loss: 0.0005:  65%|██████▍   | 202/312 [20:24<11:10,  6.10s/it]
Training Epoch: 19/100, batch 403/625 completed (loss: 0.0005:  65%|██████▍   | 202/312 [20:24<11:10,  6.09s/it]
Training Epoch: 19/100, batch 403/625 completed (loss: 0.0003:  65%|██████▍   | 202/312 [20:24<11:10,  6.10s/it]
Training Epoch: 19/100, batch 403/625 completed (loss: 0.0016:  65%|██████▍   | 202/312 [20:24<11:10,  6.09s/it]
Training Epoch: 19/100, batch 403/625 completed (loss: 0.0032:  65%|██████▍   | 202/312 [20:24<11:10,  6.09s/it]
Training Epoch: 19/100, batch 404/625 completed (loss: 0.0005:  65%|██████▍   | 202/312 [20:27<11:10,  6.10s/it]
Training Epoch: 19/100, batch 404/625 completed (loss: 0.0038:  65%|██████▍   | 202/312 [20:27<11:10,  6.09s/it]
Training Epoch: 19/100, batch 404/625 completed (loss: 0.0011:  65%|██████▍   | 202/312 [20:27<11:10,  6.09s/it]
Training Epoch: 19/100, batch 404/625 completed (loss: 0.0003:  65%|██████▍   | 202/312 [20:27<11:10,  6.09s/it]
Training Epoch: 19/100, batch 404/625 completed (loss: 0.0011:  65%|██████▌   | 203/312 [20:30<11:01,  6.07s/it]
Training Epoch: 19/100, batch 404/625 completed (loss: 0.0005:  65%|██████▌   | 203/312 [20:30<11:01,  6.07s/it]
Training Epoch: 19/100, batch 404/625 completed (loss: 0.0038:  65%|██████▌   | 203/312 [20:30<11:01,  6.07s/it]
Training Epoch: 19/100, batch 404/625 completed (loss: 0.0003:  65%|██████▌   | 203/312 [20:30<11:01,  6.07s/it]
Training Epoch: 19/100, batch 405/625 completed (loss: 0.0035:  65%|██████▌   | 203/312 [20:30<11:01,  6.07s/it]
Training Epoch: 19/100, batch 405/625 completed (loss: 0.0011:  65%|██████▌   | 203/312 [20:30<11:01,  6.07s/it]
Training Epoch: 19/100, batch 405/625 completed (loss: 0.0004:  65%|██████▌   | 203/312 [20:30<11:01,  6.07s/it]
Training Epoch: 19/100, batch 405/625 completed (loss: 0.0003:  65%|██████▌   | 203/312 [20:30<11:01,  6.07s/it]
Training Epoch: 19/100, batch 406/625 completed (loss: 0.0002:  65%|██████▌   | 203/312 [20:33<11:01,  6.07s/it]
Training Epoch: 19/100, batch 406/625 completed (loss: 0.0003:  65%|██████▌   | 203/312 [20:33<11:01,  6.07s/it]
Training Epoch: 19/100, batch 406/625 completed (loss: 0.0015:  65%|██████▌   | 203/312 [20:33<11:01,  6.07s/it]
Training Epoch: 19/100, batch 406/625 completed (loss: 0.0004:  65%|██████▌   | 203/312 [20:33<11:01,  6.07s/it]
Training Epoch: 19/100, batch 406/625 completed (loss: 0.0002:  65%|██████▌   | 204/312 [20:36<10:53,  6.05s/it]
Training Epoch: 19/100, batch 406/625 completed (loss: 0.0015:  65%|██████▌   | 204/312 [20:36<10:53,  6.05s/it]
Training Epoch: 19/100, batch 406/625 completed (loss: 0.0003:  65%|██████▌   | 204/312 [20:36<10:53,  6.05s/it]
Training Epoch: 19/100, batch 406/625 completed (loss: 0.0004:  65%|██████▌   | 204/312 [20:36<10:53,  6.05s/it]
Training Epoch: 19/100, batch 407/625 completed (loss: 0.0005:  65%|██████▌   | 204/312 [20:36<10:53,  6.05s/it]
Training Epoch: 19/100, batch 407/625 completed (loss: 0.0008:  65%|██████▌   | 204/312 [20:36<10:53,  6.05s/it]
Training Epoch: 19/100, batch 407/625 completed (loss: 0.0007:  65%|██████▌   | 204/312 [20:36<10:53,  6.05s/it]
Training Epoch: 19/100, batch 407/625 completed (loss: 0.0001:  65%|██████▌   | 204/312 [20:36<10:53,  6.05s/it]
Training Epoch: 19/100, batch 408/625 completed (loss: 0.0002:  65%|██████▌   | 204/312 [20:39<10:53,  6.05s/it]
Training Epoch: 19/100, batch 408/625 completed (loss: 0.0061:  65%|██████▌   | 204/312 [20:39<10:53,  6.05s/it]
Training Epoch: 19/100, batch 408/625 completed (loss: 0.0015:  65%|██████▌   | 204/312 [20:39<10:53,  6.05s/it]
Training Epoch: 19/100, batch 408/625 completed (loss: 0.0066:  65%|██████▌   | 204/312 [20:39<10:53,  6.05s/it]
Training Epoch: 19/100, batch 408/625 completed (loss: 0.0015:  66%|██████▌   | 205/312 [20:42<10:45,  6.03s/it]
Training Epoch: 19/100, batch 408/625 completed (loss: 0.0066:  66%|██████▌   | 205/312 [20:42<10:45,  6.03s/it]
Training Epoch: 19/100, batch 408/625 completed (loss: 0.0002:  66%|██████▌   | 205/312 [20:42<10:45,  6.03s/it]
Training Epoch: 19/100, batch 408/625 completed (loss: 0.0061:  66%|██████▌   | 205/312 [20:42<10:45,  6.03s/it]
Training Epoch: 19/100, batch 409/625 completed (loss: 0.0014:  66%|██████▌   | 205/312 [20:42<10:45,  6.03s/it]
Training Epoch: 19/100, batch 409/625 completed (loss: 0.0009:  66%|██████▌   | 205/312 [20:42<10:45,  6.03s/it]
Training Epoch: 19/100, batch 409/625 completed (loss: 0.0057:  66%|██████▌   | 205/312 [20:42<10:45,  6.03s/it]
Training Epoch: 19/100, batch 409/625 completed (loss: 0.002:  66%|██████▌   | 205/312 [20:42<10:45,  6.03s/it] 
Training Epoch: 19/100, batch 410/625 completed (loss: 0.012:  66%|██████▌   | 205/312 [20:45<10:45,  6.03s/it] 
Training Epoch: 19/100, batch 410/625 completed (loss: 0.0009:  66%|██████▌   | 205/312 [20:45<10:45,  6.03s/it]
Training Epoch: 19/100, batch 410/625 completed (loss: 0.0004:  66%|██████▌   | 205/312 [20:45<10:45,  6.03s/it]
Training Epoch: 19/100, batch 410/625 completed (loss: 0.0005:  66%|██████▌   | 205/312 [20:45<10:45,  6.03s/it]
Training Epoch: 19/100, batch 410/625 completed (loss: 0.0004:  66%|██████▌   | 206/312 [20:48<10:43,  6.07s/it]
Training Epoch: 19/100, batch 410/625 completed (loss: 0.0009:  66%|██████▌   | 206/312 [20:48<10:43,  6.07s/it]
Training Epoch: 19/100, batch 410/625 completed (loss: 0.0005:  66%|██████▌   | 206/312 [20:48<10:43,  6.07s/it]
Training Epoch: 19/100, batch 410/625 completed (loss: 0.012:  66%|██████▌   | 206/312 [20:48<10:43,  6.07s/it]
Training Epoch: 19/100, batch 411/625 completed (loss: 0.0005:  66%|██████▌   | 206/312 [20:48<10:43,  6.07s/it]
Training Epoch: 19/100, batch 411/625 completed (loss: 0.0528:  66%|██████▌   | 206/312 [20:48<10:43,  6.07s/it]
Training Epoch: 19/100, batch 411/625 completed (loss: 0.0011:  66%|██████▌   | 206/312 [20:48<10:43,  6.07s/it]
Training Epoch: 19/100, batch 411/625 completed (loss: 0.0003:  66%|██████▌   | 206/312 [20:48<10:43,  6.07s/it]
Training Epoch: 19/100, batch 412/625 completed (loss: 0.003:  66%|██████▌   | 206/312 [20:51<10:43,  6.07s/it] 
Training Epoch: 19/100, batch 412/625 completed (loss: 0.0006:  66%|██████▌   | 206/312 [20:51<10:43,  6.07s/it]
Training Epoch: 19/100, batch 412/625 completed (loss: 0.0001:  66%|██████▌   | 206/312 [20:51<10:43,  6.07s/it]
Training Epoch: 19/100, batch 412/625 completed (loss: 0.0012:  66%|██████▌   | 206/312 [20:51<10:43,  6.07s/it]
Training Epoch: 19/100, batch 412/625 completed (loss: 0.0006:  66%|██████▋   | 207/312 [20:54<10:40,  6.10s/it]
Training Epoch: 19/100, batch 412/625 completed (loss: 0.003:  66%|██████▋   | 207/312 [20:54<10:40,  6.10s/it]
Training Epoch: 19/100, batch 412/625 completed (loss: 0.0012:  66%|██████▋   | 207/312 [20:54<10:40,  6.10s/it]
Training Epoch: 19/100, batch 412/625 completed (loss: 0.0001:  66%|██████▋   | 207/312 [20:54<10:40,  6.10s/it]
Training Epoch: 19/100, batch 413/625 completed (loss: 0.005:  66%|██████▋   | 207/312 [20:54<10:40,  6.10s/it] 
Training Epoch: 19/100, batch 413/625 completed (loss: 0.0004:  66%|██████▋   | 207/312 [20:54<10:40,  6.10s/it]
Training Epoch: 19/100, batch 413/625 completed (loss: 0.0002:  66%|██████▋   | 207/312 [20:54<10:40,  6.10s/it]
Training Epoch: 19/100, batch 413/625 completed (loss: 0.0001:  66%|██████▋   | 207/312 [20:54<10:40,  6.10s/it]
Training Epoch: 19/100, batch 414/625 completed (loss: 0.0006:  66%|██████▋   | 207/312 [20:57<10:40,  6.10s/it]
Training Epoch: 19/100, batch 414/625 completed (loss: 0.0006:  66%|██████▋   | 207/312 [20:57<10:40,  6.10s/it]
Training Epoch: 19/100, batch 414/625 completed (loss: 0.0001:  66%|██████▋   | 207/312 [20:57<10:40,  6.10s/it]
Training Epoch: 19/100, batch 414/625 completed (loss: 0.0004:  66%|██████▋   | 207/312 [20:57<10:40,  6.10s/it]
Training Epoch: 19/100, batch 414/625 completed (loss: 0.0006:  67%|██████▋   | 208/312 [21:00<10:26,  6.02s/it]
Training Epoch: 19/100, batch 414/625 completed (loss: 0.0004:  67%|██████▋   | 208/312 [21:00<10:26,  6.02s/it]
Training Epoch: 19/100, batch 414/625 completed (loss: 0.0006:  67%|██████▋   | 208/312 [21:00<10:26,  6.03s/it]
Training Epoch: 19/100, batch 414/625 completed (loss: 0.0001:  67%|██████▋   | 208/312 [21:00<10:27,  6.03s/it]
Training Epoch: 19/100, batch 415/625 completed (loss: 0.0039:  67%|██████▋   | 208/312 [21:00<10:27,  6.03s/it]
Training Epoch: 19/100, batch 415/625 completed (loss: 0.0128:  67%|██████▋   | 208/312 [21:00<10:26,  6.03s/it]
Training Epoch: 19/100, batch 415/625 completed (loss: 0.0002:  67%|██████▋   | 208/312 [21:00<10:26,  6.02s/it]
Training Epoch: 19/100, batch 415/625 completed (loss: 0.0027:  67%|██████▋   | 208/312 [21:00<10:26,  6.02s/it]
Training Epoch: 19/100, batch 416/625 completed (loss: 0.0055:  67%|██████▋   | 208/312 [21:03<10:27,  6.03s/it]
Training Epoch: 19/100, batch 416/625 completed (loss: 0.0008:  67%|██████▋   | 208/312 [21:03<10:26,  6.02s/it]
Training Epoch: 19/100, batch 416/625 completed (loss: 0.0001:  67%|██████▋   | 208/312 [21:03<10:26,  6.03s/it]
Training Epoch: 19/100, batch 416/625 completed (loss: 0.0006:  67%|██████▋   | 208/312 [21:03<10:26,  6.02s/it]
Training Epoch: 19/100, batch 416/625 completed (loss: 0.0006:  67%|██████▋   | 209/312 [21:06<10:30,  6.12s/it]
Training Epoch: 19/100, batch 416/625 completed (loss: 0.0001:  67%|██████▋   | 209/312 [21:06<10:30,  6.12s/it]
Training Epoch: 19/100, batch 416/625 completed (loss: 0.0008:  67%|██████▋   | 209/312 [21:06<10:30,  6.12s/it]
Training Epoch: 19/100, batch 416/625 completed (loss: 0.0055:  67%|██████▋   | 209/312 [21:06<10:30,  6.12s/it]
Training Epoch: 19/100, batch 417/625 completed (loss: 0.0:  67%|██████▋   | 209/312 [21:06<10:30,  6.12s/it]   
Training Epoch: 19/100, batch 417/625 completed (loss: 0.0034:  67%|██████▋   | 209/312 [21:06<10:30,  6.12s/it]
Training Epoch: 19/100, batch 417/625 completed (loss: 0.0001:  67%|██████▋   | 209/312 [21:06<10:30,  6.12s/it]
Training Epoch: 19/100, batch 417/625 completed (loss: 0.0006:  67%|██████▋   | 209/312 [21:06<10:30,  6.12s/it]
Training Epoch: 19/100, batch 418/625 completed (loss: 0.0003:  67%|██████▋   | 209/312 [21:09<10:30,  6.12s/it]
Training Epoch: 19/100, batch 418/625 completed (loss: 0.0016:  67%|██████▋   | 209/312 [21:09<10:30,  6.12s/it]
Training Epoch: 19/100, batch 418/625 completed (loss: 0.0015:  67%|██████▋   | 209/312 [21:09<10:30,  6.12s/it]
Training Epoch: 19/100, batch 418/625 completed (loss: 0.0002:  67%|██████▋   | 209/312 [21:09<10:30,  6.12s/it]
Training Epoch: 19/100, batch 418/625 completed (loss: 0.0003:  67%|██████▋   | 210/312 [21:12<10:21,  6.09s/it]
Training Epoch: 19/100, batch 418/625 completed (loss: 0.0002:  67%|██████▋   | 210/312 [21:12<10:21,  6.09s/it]
Training Epoch: 19/100, batch 418/625 completed (loss: 0.0015:  67%|██████▋   | 210/312 [21:12<10:21,  6.09s/it]
Training Epoch: 19/100, batch 418/625 completed (loss: 0.0016:  67%|██████▋   | 210/312 [21:12<10:21,  6.09s/it]
Training Epoch: 19/100, batch 419/625 completed (loss: 0.0005:  67%|██████▋   | 210/312 [21:12<10:21,  6.09s/it]
Training Epoch: 19/100, batch 419/625 completed (loss: 0.0003:  67%|██████▋   | 210/312 [21:12<10:21,  6.09s/it]
Training Epoch: 19/100, batch 419/625 completed (loss: 0.0006:  67%|██████▋   | 210/312 [21:12<10:21,  6.09s/it]
Training Epoch: 19/100, batch 419/625 completed (loss: 0.0015:  67%|██████▋   | 210/312 [21:12<10:21,  6.09s/it]
Training Epoch: 19/100, batch 420/625 completed (loss: 0.0004:  67%|██████▋   | 210/312 [21:15<10:21,  6.09s/it]
Training Epoch: 19/100, batch 420/625 completed (loss: 0.0016:  67%|██████▋   | 210/312 [21:15<10:21,  6.09s/it]
Training Epoch: 19/100, batch 420/625 completed (loss: 0.0004:  67%|██████▋   | 210/312 [21:15<10:21,  6.09s/it]
Training Epoch: 19/100, batch 420/625 completed (loss: 0.0001:  67%|██████▋   | 210/312 [21:15<10:21,  6.09s/it]
Training Epoch: 19/100, batch 420/625 completed (loss: 0.0004:  68%|██████▊   | 211/312 [21:18<10:09,  6.03s/it]
Training Epoch: 19/100, batch 420/625 completed (loss: 0.0004:  68%|██████▊   | 211/312 [21:18<10:09,  6.03s/it]
Training Epoch: 19/100, batch 420/625 completed (loss: 0.0001:  68%|██████▊   | 211/312 [21:18<10:09,  6.04s/it]
Training Epoch: 19/100, batch 420/625 completed (loss: 0.0016:  68%|██████▊   | 211/312 [21:18<10:09,  6.04s/it]
Training Epoch: 19/100, batch 421/625 completed (loss: 0.004:  68%|██████▊   | 211/312 [21:18<10:09,  6.04s/it] 
Training Epoch: 19/100, batch 421/625 completed (loss: 0.0245:  68%|██████▊   | 211/312 [21:18<10:09,  6.03s/it]
Training Epoch: 19/100, batch 421/625 completed (loss: 0.0011:  68%|██████▊   | 211/312 [21:18<10:09,  6.03s/it]
Training Epoch: 19/100, batch 421/625 completed (loss: 0.0003:  68%|██████▊   | 211/312 [21:18<10:09,  6.04s/it]
Training Epoch: 19/100, batch 422/625 completed (loss: 0.0012:  68%|██████▊   | 211/312 [21:22<10:09,  6.04s/it]
Training Epoch: 19/100, batch 422/625 completed (loss: 0.0009:  68%|██████▊   | 211/312 [21:22<10:09,  6.03s/it]
Training Epoch: 19/100, batch 422/625 completed (loss: 0.0004:  68%|██████▊   | 211/312 [21:22<10:09,  6.03s/it]
Training Epoch: 19/100, batch 422/625 completed (loss: 0.001:  68%|██████▊   | 211/312 [21:22<10:09,  6.04s/it] 
Training Epoch: 19/100, batch 422/625 completed (loss: 0.0004:  68%|██████▊   | 212/312 [21:24<10:08,  6.09s/it]
Training Epoch: 19/100, batch 422/625 completed (loss: 0.0009:  68%|██████▊   | 212/312 [21:24<10:08,  6.09s/it]
Training Epoch: 19/100, batch 422/625 completed (loss: 0.001:  68%|██████▊   | 212/312 [21:24<10:08,  6.09s/it]
Training Epoch: 19/100, batch 422/625 completed (loss: 0.0012:  68%|██████▊   | 212/312 [21:24<10:08,  6.09s/it]
Training Epoch: 19/100, batch 423/625 completed (loss: 0.0009:  68%|██████▊   | 212/312 [21:24<10:08,  6.09s/it]
Training Epoch: 19/100, batch 423/625 completed (loss: 0.0036:  68%|██████▊   | 212/312 [21:24<10:08,  6.09s/it]
Training Epoch: 19/100, batch 423/625 completed (loss: 0.0104:  68%|██████▊   | 212/312 [21:24<10:08,  6.09s/it]
Training Epoch: 19/100, batch 423/625 completed (loss: 0.0679:  68%|██████▊   | 212/312 [21:25<10:08,  6.09s/it]
Training Epoch: 19/100, batch 424/625 completed (loss: 0.0011:  68%|██████▊   | 212/312 [21:28<10:08,  6.09s/it]
Training Epoch: 19/100, batch 424/625 completed (loss: 0.001:  68%|██████▊   | 212/312 [21:28<10:08,  6.09s/it] 
Training Epoch: 19/100, batch 424/625 completed (loss: 0.0005:  68%|██████▊   | 212/312 [21:28<10:08,  6.09s/it]
Training Epoch: 19/100, batch 424/625 completed (loss: 0.0002:  68%|██████▊   | 212/312 [21:28<10:08,  6.09s/it]
Training Epoch: 19/100, batch 424/625 completed (loss: 0.0011:  68%|██████▊   | 213/312 [21:30<10:01,  6.07s/it]
Training Epoch: 19/100, batch 424/625 completed (loss: 0.0002:  68%|██████▊   | 213/312 [21:31<10:01,  6.07s/it]
Training Epoch: 19/100, batch 424/625 completed (loss: 0.0005:  68%|██████▊   | 213/312 [21:30<10:01,  6.08s/it]
Training Epoch: 19/100, batch 424/625 completed (loss: 0.001:  68%|██████▊   | 213/312 [21:30<10:01,  6.08s/it]
Training Epoch: 19/100, batch 425/625 completed (loss: 0.0025:  68%|██████▊   | 213/312 [21:31<10:01,  6.07s/it]
Training Epoch: 19/100, batch 425/625 completed (loss: 0.0143:  68%|██████▊   | 213/312 [21:31<10:01,  6.08s/it]
Training Epoch: 19/100, batch 425/625 completed (loss: 0.0008:  68%|██████▊   | 213/312 [21:31<10:01,  6.08s/it]
Training Epoch: 19/100, batch 425/625 completed (loss: 0.0002:  68%|██████▊   | 213/312 [21:31<10:01,  6.07s/it]
Training Epoch: 19/100, batch 426/625 completed (loss: 0.0003:  68%|██████▊   | 213/312 [21:33<10:01,  6.08s/it]
Training Epoch: 19/100, batch 426/625 completed (loss: 0.0008:  68%|██████▊   | 213/312 [21:33<10:01,  6.08s/it]
Training Epoch: 19/100, batch 426/625 completed (loss: 0.0048:  68%|██████▊   | 213/312 [21:33<10:01,  6.07s/it]
Training Epoch: 19/100, batch 426/625 completed (loss: 0.0013:  68%|██████▊   | 213/312 [21:34<10:01,  6.07s/it]
Training Epoch: 19/100, batch 426/625 completed (loss: 0.0013:  69%|██████▊   | 214/312 [21:37<09:58,  6.10s/it]
Training Epoch: 19/100, batch 426/625 completed (loss: 0.0048:  69%|██████▊   | 214/312 [21:37<09:58,  6.11s/it]
Training Epoch: 19/100, batch 426/625 completed (loss: 0.0003:  69%|██████▊   | 214/312 [21:37<09:58,  6.11s/it]
Training Epoch: 19/100, batch 426/625 completed (loss: 0.0008:  69%|██████▊   | 214/312 [21:37<09:58,  6.11s/it]
Training Epoch: 19/100, batch 427/625 completed (loss: 0.0003:  69%|██████▊   | 214/312 [21:37<09:58,  6.11s/it]
Training Epoch: 19/100, batch 427/625 completed (loss: 0.0008:  69%|██████▊   | 214/312 [21:37<09:58,  6.11s/it]
Training Epoch: 19/100, batch 427/625 completed (loss: 0.0006:  69%|██████▊   | 214/312 [21:37<09:58,  6.11s/it]
Training Epoch: 19/100, batch 427/625 completed (loss: 0.0285:  69%|██████▊   | 214/312 [21:37<09:58,  6.10s/it]
Training Epoch: 19/100, batch 428/625 completed (loss: 0.0002:  69%|██████▊   | 214/312 [21:40<09:58,  6.11s/it]
Training Epoch: 19/100, batch 428/625 completed (loss: 0.0011:  69%|██████▊   | 214/312 [21:40<09:58,  6.11s/it]
Training Epoch: 19/100, batch 428/625 completed (loss: 0.0025:  69%|██████▊   | 214/312 [21:40<09:58,  6.11s/it]
Training Epoch: 19/100, batch 428/625 completed (loss: 0.0002:  69%|██████▊   | 214/312 [21:40<09:58,  6.10s/it]
Training Epoch: 19/100, batch 428/625 completed (loss: 0.0025:  69%|██████▉   | 215/312 [21:43<09:55,  6.13s/it]
Training Epoch: 19/100, batch 428/625 completed (loss: 0.0011:  69%|██████▉   | 215/312 [21:43<09:54,  6.13s/it]
Training Epoch: 19/100, batch 428/625 completed (loss: 0.0002:  69%|██████▉   | 215/312 [21:43<09:55,  6.13s/it]
Training Epoch: 19/100, batch 428/625 completed (loss: 0.0002:  69%|██████▉   | 215/312 [21:43<09:55,  6.14s/it]
Training Epoch: 19/100, batch 429/625 completed (loss: 0.0001:  69%|██████▉   | 215/312 [21:43<09:55,  6.13s/it]
Training Epoch: 19/100, batch 429/625 completed (loss: 0.0016:  69%|██████▉   | 215/312 [21:43<09:55,  6.13s/it]
Training Epoch: 19/100, batch 429/625 completed (loss: 0.0171:  69%|██████▉   | 215/312 [21:43<09:54,  6.13s/it]
Training Epoch: 19/100, batch 429/625 completed (loss: 0.0224:  69%|██████▉   | 215/312 [21:43<09:55,  6.14s/it]
Training Epoch: 19/100, batch 430/625 completed (loss: 0.0188:  69%|██████▉   | 215/312 [21:46<09:55,  6.13s/it]
Training Epoch: 19/100, batch 430/625 completed (loss: 0.0037:  69%|██████▉   | 215/312 [21:46<09:54,  6.13s/it]
Training Epoch: 19/100, batch 430/625 completed (loss: 0.0007:  69%|██████▉   | 215/312 [21:46<09:55,  6.13s/it]
Training Epoch: 19/100, batch 430/625 completed (loss: 0.0021:  69%|██████▉   | 215/312 [21:46<09:55,  6.14s/it]
Training Epoch: 19/100, batch 430/625 completed (loss: 0.0007:  69%|██████▉   | 216/312 [21:49<09:41,  6.05s/it]
Training Epoch: 19/100, batch 430/625 completed (loss: 0.0037:  69%|██████▉   | 216/312 [21:49<09:41,  6.05s/it]
Training Epoch: 19/100, batch 430/625 completed (loss: 0.0021:  69%|██████▉   | 216/312 [21:49<09:41,  6.06s/it]
Training Epoch: 19/100, batch 430/625 completed (loss: 0.0188:  69%|██████▉   | 216/312 [21:49<09:41,  6.06s/it]
Training Epoch: 19/100, batch 431/625 completed (loss: 0.0003:  69%|██████▉   | 216/312 [21:49<09:41,  6.06s/it]
Training Epoch: 19/100, batch 431/625 completed (loss: 0.0024:  69%|██████▉   | 216/312 [21:49<09:41,  6.05s/it]
Training Epoch: 19/100, batch 431/625 completed (loss: 0.0001:  69%|██████▉   | 216/312 [21:49<09:41,  6.05s/it]
Training Epoch: 19/100, batch 431/625 completed (loss: 0.0001:  69%|██████▉   | 216/312 [21:49<09:41,  6.06s/it]
Training Epoch: 19/100, batch 432/625 completed (loss: 0.0002:  69%|██████▉   | 216/312 [21:52<09:41,  6.05s/it]
Training Epoch: 19/100, batch 432/625 completed (loss: 0.0002:  69%|██████▉   | 216/312 [21:52<09:41,  6.06s/it]
Training Epoch: 19/100, batch 432/625 completed (loss: 0.0006:  69%|██████▉   | 216/312 [21:52<09:41,  6.05s/it]
Training Epoch: 19/100, batch 432/625 completed (loss: 0.0021:  69%|██████▉   | 216/312 [21:52<09:41,  6.06s/it]
Training Epoch: 19/100, batch 432/625 completed (loss: 0.0002:  70%|██████▉   | 217/312 [21:55<09:37,  6.08s/it]
Training Epoch: 19/100, batch 432/625 completed (loss: 0.0006:  70%|██████▉   | 217/312 [21:55<09:37,  6.08s/it]
Training Epoch: 19/100, batch 432/625 completed (loss: 0.0002:  70%|██████▉   | 217/312 [21:55<09:37,  6.08s/it]
Training Epoch: 19/100, batch 432/625 completed (loss: 0.0021:  70%|██████▉   | 217/312 [21:55<09:37,  6.08s/it]
Training Epoch: 19/100, batch 433/625 completed (loss: 0.0004:  70%|██████▉   | 217/312 [21:55<09:37,  6.08s/it]
Training Epoch: 19/100, batch 433/625 completed (loss: 0.0003:  70%|██████▉   | 217/312 [21:55<09:37,  6.08s/it]
Training Epoch: 19/100, batch 433/625 completed (loss: 0.0002:  70%|██████▉   | 217/312 [21:55<09:37,  6.08s/it]
Training Epoch: 19/100, batch 433/625 completed (loss: 0.0001:  70%|██████▉   | 217/312 [21:55<09:37,  6.08s/it]
Training Epoch: 19/100, batch 434/625 completed (loss: 0.0:  70%|██████▉   | 217/312 [21:58<09:37,  6.08s/it]   
Training Epoch: 19/100, batch 434/625 completed (loss: 0.0116:  70%|██████▉   | 217/312 [21:58<09:37,  6.08s/it]
Training Epoch: 19/100, batch 434/625 completed (loss: 0.0004:  70%|██████▉   | 217/312 [21:58<09:37,  6.08s/it]
Training Epoch: 19/100, batch 434/625 completed (loss: 0.0142:  70%|██████▉   | 217/312 [21:58<09:37,  6.08s/it]
Training Epoch: 19/100, batch 434/625 completed (loss: 0.0004:  70%|██████▉   | 218/312 [22:01<09:35,  6.12s/it]
Training Epoch: 19/100, batch 434/625 completed (loss: 0.0142:  70%|██████▉   | 218/312 [22:01<09:34,  6.12s/it]
Training Epoch: 19/100, batch 434/625 completed (loss: 0.0116:  70%|██████▉   | 218/312 [22:01<09:35,  6.12s/it]
Training Epoch: 19/100, batch 434/625 completed (loss: 0.0:  70%|██████▉   | 218/312 [22:01<09:35,  6.12s/it]
Training Epoch: 19/100, batch 435/625 completed (loss: 0.0139:  70%|██████▉   | 218/312 [22:01<09:35,  6.12s/it]
Training Epoch: 19/100, batch 435/625 completed (loss: 0.0005:  70%|██████▉   | 218/312 [22:01<09:35,  6.12s/it]
Training Epoch: 19/100, batch 435/625 completed (loss: 0.0005:  70%|██████▉   | 218/312 [22:01<09:35,  6.12s/it]
Training Epoch: 19/100, batch 435/625 completed (loss: 0.0007:  70%|██████▉   | 218/312 [22:01<09:34,  6.12s/it]
Training Epoch: 19/100, batch 436/625 completed (loss: 0.0006:  70%|██████▉   | 218/312 [22:04<09:35,  6.12s/it]
Training Epoch: 19/100, batch 436/625 completed (loss: 0.0004:  70%|██████▉   | 218/312 [22:04<09:35,  6.12s/it]
Training Epoch: 19/100, batch 436/625 completed (loss: 0.0024:  70%|██████▉   | 218/312 [22:04<09:35,  6.12s/it]
Training Epoch: 19/100, batch 436/625 completed (loss: 0.0004:  70%|██████▉   | 218/312 [22:04<09:34,  6.12s/it]
Training Epoch: 19/100, batch 436/625 completed (loss: 0.0004:  70%|███████   | 219/312 [22:07<09:22,  6.05s/it]
Training Epoch: 19/100, batch 436/625 completed (loss: 0.0024:  70%|███████   | 219/312 [22:07<09:22,  6.05s/it]
Training Epoch: 19/100, batch 436/625 completed (loss: 0.0004:  70%|███████   | 219/312 [22:07<09:22,  6.05s/it]
Training Epoch: 19/100, batch 436/625 completed (loss: 0.0006:  70%|███████   | 219/312 [22:07<09:22,  6.05s/it]
Training Epoch: 19/100, batch 437/625 completed (loss: 0.0:  70%|███████   | 219/312 [22:07<09:22,  6.05s/it]   
Training Epoch: 19/100, batch 437/625 completed (loss: 0.0:  70%|███████   | 219/312 [22:07<09:22,  6.05s/it]   
Training Epoch: 19/100, batch 437/625 completed (loss: 0.0005:  70%|███████   | 219/312 [22:07<09:22,  6.05s/it]
Training Epoch: 19/100, batch 437/625 completed (loss: 0.0463:  70%|███████   | 219/312 [22:07<09:22,  6.05s/it]
Training Epoch: 19/100, batch 438/625 completed (loss: 0.0004:  70%|███████   | 219/312 [22:10<09:22,  6.05s/it]
Training Epoch: 19/100, batch 438/625 completed (loss: 0.0001:  70%|███████   | 219/312 [22:10<09:22,  6.05s/it]
Training Epoch: 19/100, batch 438/625 completed (loss: 0.0001:  70%|███████   | 219/312 [22:10<09:22,  6.05s/it]
Training Epoch: 19/100, batch 438/625 completed (loss: 0.0023:  70%|███████   | 219/312 [22:10<09:22,  6.05s/it]
Training Epoch: 19/100, batch 438/625 completed (loss: 0.0001:  71%|███████   | 220/312 [22:13<09:25,  6.15s/it]
Training Epoch: 19/100, batch 438/625 completed (loss: 0.0004:  71%|███████   | 220/312 [22:13<09:25,  6.15s/it]
Training Epoch: 19/100, batch 438/625 completed (loss: 0.0001:  71%|███████   | 220/312 [22:13<09:25,  6.15s/it]
Training Epoch: 19/100, batch 438/625 completed (loss: 0.0023:  71%|███████   | 220/312 [22:13<09:25,  6.15s/it]
Training Epoch: 19/100, batch 439/625 completed (loss: 0.0001:  71%|███████   | 220/312 [22:13<09:25,  6.15s/it]
Training Epoch: 19/100, batch 439/625 completed (loss: 0.0001:  71%|███████   | 220/312 [22:13<09:25,  6.15s/it]
Training Epoch: 19/100, batch 439/625 completed (loss: 0.0012:  71%|███████   | 220/312 [22:13<09:25,  6.15s/it]
Training Epoch: 19/100, batch 439/625 completed (loss: 0.0004:  71%|███████   | 220/312 [22:13<09:25,  6.15s/it]
Training Epoch: 19/100, batch 440/625 completed (loss: 0.0009:  71%|███████   | 220/312 [22:16<09:25,  6.15s/it]
Training Epoch: 19/100, batch 440/625 completed (loss: 0.0003:  71%|███████   | 220/312 [22:16<09:25,  6.15s/it]
Training Epoch: 19/100, batch 440/625 completed (loss: 0.0002:  71%|███████   | 220/312 [22:16<09:25,  6.15s/it]
Training Epoch: 19/100, batch 440/625 completed (loss: 0.0002:  71%|███████   | 220/312 [22:17<09:25,  6.15s/it]
Training Epoch: 19/100, batch 440/625 completed (loss: 0.0003:  71%|███████   | 221/312 [22:19<09:15,  6.11s/it]
Training Epoch: 19/100, batch 440/625 completed (loss: 0.0002:  71%|███████   | 221/312 [22:19<09:15,  6.10s/it]
Training Epoch: 19/100, batch 440/625 completed (loss: 0.0009:  71%|███████   | 221/312 [22:19<09:15,  6.11s/it]
Training Epoch: 19/100, batch 440/625 completed (loss: 0.0002:  71%|███████   | 221/312 [22:19<09:15,  6.10s/it]
Training Epoch: 19/100, batch 441/625 completed (loss: 0.0045:  71%|███████   | 221/312 [22:19<09:15,  6.11s/it]
Training Epoch: 19/100, batch 441/625 completed (loss: 0.0006:  71%|███████   | 221/312 [22:19<09:15,  6.10s/it]
Training Epoch: 19/100, batch 441/625 completed (loss: 0.0:  71%|███████   | 221/312 [22:19<09:15,  6.11s/it]   
Training Epoch: 19/100, batch 441/625 completed (loss: 0.0001:  71%|███████   | 221/312 [22:19<09:15,  6.10s/it]
Training Epoch: 19/100, batch 442/625 completed (loss: 0.0006:  71%|███████   | 221/312 [22:22<09:15,  6.11s/it]
Training Epoch: 19/100, batch 442/625 completed (loss: 0.0013:  71%|███████   | 221/312 [22:22<09:15,  6.10s/it]
Training Epoch: 19/100, batch 442/625 completed (loss: 0.0223:  71%|███████   | 221/312 [22:22<09:15,  6.11s/it]
Training Epoch: 19/100, batch 442/625 completed (loss: 0.0015:  71%|███████   | 221/312 [22:23<09:15,  6.10s/it]
Training Epoch: 19/100, batch 442/625 completed (loss: 0.0006:  71%|███████   | 222/312 [22:25<09:10,  6.12s/it]
Training Epoch: 19/100, batch 442/625 completed (loss: 0.0223:  71%|███████   | 222/312 [22:25<09:10,  6.11s/it]
Training Epoch: 19/100, batch 442/625 completed (loss: 0.0013:  71%|███████   | 222/312 [22:26<09:10,  6.12s/it]
Training Epoch: 19/100, batch 442/625 completed (loss: 0.0015:  71%|███████   | 222/312 [22:26<09:10,  6.12s/it]
Training Epoch: 19/100, batch 443/625 completed (loss: 0.0014:  71%|███████   | 222/312 [22:26<09:10,  6.12s/it]
Training Epoch: 19/100, batch 443/625 completed (loss: 0.0:  71%|███████   | 222/312 [22:26<09:10,  6.12s/it]   
Training Epoch: 19/100, batch 443/625 completed (loss: 0.0014:  71%|███████   | 222/312 [22:26<09:10,  6.11s/it]
Training Epoch: 19/100, batch 443/625 completed (loss: 0.0044:  71%|███████   | 222/312 [22:26<09:10,  6.12s/it]
Training Epoch: 19/100, batch 444/625 completed (loss: 0.0009:  71%|███████   | 222/312 [22:28<09:10,  6.12s/it]
Training Epoch: 19/100, batch 444/625 completed (loss: 0.0056:  71%|███████   | 222/312 [22:28<09:10,  6.12s/it]
Training Epoch: 19/100, batch 444/625 completed (loss: 0.0002:  71%|███████   | 222/312 [22:28<09:10,  6.11s/it]
Training Epoch: 19/100, batch 444/625 completed (loss: 0.0002:  71%|███████   | 222/312 [22:29<09:10,  6.12s/it]
Training Epoch: 19/100, batch 444/625 completed (loss: 0.0009:  71%|███████▏  | 223/312 [22:32<09:06,  6.14s/it]
Training Epoch: 19/100, batch 444/625 completed (loss: 0.0002:  71%|███████▏  | 223/312 [22:32<09:06,  6.14s/it]
Training Epoch: 19/100, batch 444/625 completed (loss: 0.0002:  71%|███████▏  | 223/312 [22:32<09:06,  6.14s/it]
Training Epoch: 19/100, batch 444/625 completed (loss: 0.0056:  71%|███████▏  | 223/312 [22:32<09:07,  6.15s/it]
Training Epoch: 19/100, batch 445/625 completed (loss: 0.0003:  71%|███████▏  | 223/312 [22:32<09:07,  6.15s/it]
Training Epoch: 19/100, batch 445/625 completed (loss: 0.0001:  71%|███████▏  | 223/312 [22:32<09:06,  6.14s/it]
Training Epoch: 19/100, batch 445/625 completed (loss: 0.0013:  71%|███████▏  | 223/312 [22:32<09:06,  6.14s/it]
Training Epoch: 19/100, batch 445/625 completed (loss: 0.0113:  71%|███████▏  | 223/312 [22:32<09:06,  6.14s/it]
Training Epoch: 19/100, batch 446/625 completed (loss: 0.001:  71%|███████▏  | 223/312 [22:35<09:07,  6.15s/it] 
Training Epoch: 19/100, batch 446/625 completed (loss: 0.0021:  71%|███████▏  | 223/312 [22:35<09:06,  6.14s/it]
Training Epoch: 19/100, batch 446/625 completed (loss: 0.0012:  71%|███████▏  | 223/312 [22:35<09:06,  6.14s/it]
Training Epoch: 19/100, batch 446/625 completed (loss: 0.0001:  71%|███████▏  | 223/312 [22:35<09:06,  6.14s/it]
Training Epoch: 19/100, batch 446/625 completed (loss: 0.0012:  72%|███████▏  | 224/312 [22:38<08:53,  6.06s/it]
Training Epoch: 19/100, batch 446/625 completed (loss: 0.0021:  72%|███████▏  | 224/312 [22:38<08:53,  6.06s/it]
Training Epoch: 19/100, batch 446/625 completed (loss: 0.0001:  72%|███████▏  | 224/312 [22:38<08:53,  6.06s/it]
Training Epoch: 19/100, batch 446/625 completed (loss: 0.001:  72%|███████▏  | 224/312 [22:38<08:53,  6.06s/it]
Training Epoch: 19/100, batch 447/625 completed (loss: 0.0005:  72%|███████▏  | 224/312 [22:38<08:53,  6.06s/it]
Training Epoch: 19/100, batch 447/625 completed (loss: 0.0001:  72%|███████▏  | 224/312 [22:38<08:53,  6.06s/it]
Training Epoch: 19/100, batch 447/625 completed (loss: 0.0001:  72%|███████▏  | 224/312 [22:38<08:53,  6.06s/it]
Training Epoch: 19/100, batch 447/625 completed (loss: 0.0002:  72%|███████▏  | 224/312 [22:38<08:53,  6.06s/it]
Training Epoch: 19/100, batch 448/625 completed (loss: 0.0011:  72%|███████▏  | 224/312 [22:41<08:53,  6.06s/it]
Training Epoch: 19/100, batch 448/625 completed (loss: 0.0025:  72%|███████▏  | 224/312 [22:41<08:53,  6.06s/it]
Training Epoch: 19/100, batch 448/625 completed (loss: 0.017:  72%|███████▏  | 224/312 [22:41<08:53,  6.06s/it] 
Training Epoch: 19/100, batch 448/625 completed (loss: 0.001:  72%|███████▏  | 224/312 [22:41<08:53,  6.06s/it] 
Training Epoch: 19/100, batch 448/625 completed (loss: 0.0025:  72%|███████▏  | 225/312 [22:44<08:49,  6.08s/it]
Training Epoch: 19/100, batch 448/625 completed (loss: 0.0011:  72%|███████▏  | 225/312 [22:44<08:49,  6.09s/it]
Training Epoch: 19/100, batch 448/625 completed (loss: 0.001:  72%|███████▏  | 225/312 [22:44<08:49,  6.08s/it]
Training Epoch: 19/100, batch 448/625 completed (loss: 0.017:  72%|███████▏  | 225/312 [22:44<08:49,  6.09s/it]
Training Epoch: 19/100, batch 449/625 completed (loss: 0.0014:  72%|███████▏  | 225/312 [22:44<08:49,  6.09s/it]
Training Epoch: 19/100, batch 449/625 completed (loss: 0.0002:  72%|███████▏  | 225/312 [22:44<08:49,  6.08s/it]
Training Epoch: 19/100, batch 449/625 completed (loss: 0.0119:  72%|███████▏  | 225/312 [22:44<08:49,  6.09s/it]
Training Epoch: 19/100, batch 449/625 completed (loss: 0.0004:  72%|███████▏  | 225/312 [22:44<08:49,  6.08s/it]
Training Epoch: 19/100, batch 450/625 completed (loss: 0.0:  72%|███████▏  | 225/312 [22:47<08:49,  6.09s/it]   
Training Epoch: 19/100, batch 450/625 completed (loss: 0.0072:  72%|███████▏  | 225/312 [22:47<08:49,  6.09s/it]
Training Epoch: 19/100, batch 450/625 completed (loss: 0.0012:  72%|███████▏  | 225/312 [22:47<08:49,  6.08s/it]
Training Epoch: 19/100, batch 450/625 completed (loss: 0.0001:  72%|███████▏  | 225/312 [22:47<08:49,  6.08s/it]
Training Epoch: 19/100, batch 450/625 completed (loss: 0.0012:  72%|███████▏  | 226/312 [22:50<08:42,  6.08s/it]
Training Epoch: 19/100, batch 450/625 completed (loss: 0.0001:  72%|███████▏  | 226/312 [22:50<08:42,  6.08s/it]
Training Epoch: 19/100, batch 450/625 completed (loss: 0.0072:  72%|███████▏  | 226/312 [22:50<08:42,  6.08s/it]
Training Epoch: 19/100, batch 450/625 completed (loss: 0.0:  72%|███████▏  | 226/312 [22:50<08:42,  6.08s/it]
Training Epoch: 19/100, batch 451/625 completed (loss: 0.0065:  72%|███████▏  | 226/312 [22:50<08:42,  6.08s/it]
Training Epoch: 19/100, batch 451/625 completed (loss: 0.0005:  72%|███████▏  | 226/312 [22:50<08:42,  6.08s/it]
Training Epoch: 19/100, batch 451/625 completed (loss: 0.0059:  72%|███████▏  | 226/312 [22:50<08:42,  6.08s/it]
Training Epoch: 19/100, batch 451/625 completed (loss: 0.0001:  72%|███████▏  | 226/312 [22:50<08:42,  6.08s/it]
Training Epoch: 19/100, batch 452/625 completed (loss: 0.0011:  72%|███████▏  | 226/312 [22:53<08:42,  6.08s/it]
Training Epoch: 19/100, batch 452/625 completed (loss: 0.0014:  72%|███████▏  | 226/312 [22:53<08:42,  6.08s/it]
Training Epoch: 19/100, batch 452/625 completed (loss: 0.0015:  72%|███████▏  | 226/312 [22:53<08:42,  6.08s/it]
Training Epoch: 19/100, batch 452/625 completed (loss: 0.0004:  72%|███████▏  | 226/312 [22:53<08:42,  6.08s/it]
Training Epoch: 19/100, batch 452/625 completed (loss: 0.0015:  73%|███████▎  | 227/312 [22:56<08:36,  6.08s/it]
Training Epoch: 19/100, batch 452/625 completed (loss: 0.0014:  73%|███████▎  | 227/312 [22:56<08:36,  6.08s/it]
Training Epoch: 19/100, batch 452/625 completed (loss: 0.0011:  73%|███████▎  | 227/312 [22:56<08:36,  6.08s/it]
Training Epoch: 19/100, batch 452/625 completed (loss: 0.0004:  73%|███████▎  | 227/312 [22:56<08:37,  6.08s/it]
Training Epoch: 19/100, batch 453/625 completed (loss: 0.0002:  73%|███████▎  | 227/312 [22:56<08:36,  6.08s/it]
Training Epoch: 19/100, batch 453/625 completed (loss: 0.0002:  73%|███████▎  | 227/312 [22:56<08:36,  6.08s/it]
Training Epoch: 19/100, batch 453/625 completed (loss: 0.0031:  73%|███████▎  | 227/312 [22:56<08:36,  6.08s/it]
Training Epoch: 19/100, batch 453/625 completed (loss: 0.0001:  73%|███████▎  | 227/312 [22:56<08:37,  6.08s/it]
Training Epoch: 19/100, batch 454/625 completed (loss: 0.0005:  73%|███████▎  | 227/312 [22:59<08:36,  6.08s/it]
Training Epoch: 19/100, batch 454/625 completed (loss: 0.0001:  73%|███████▎  | 227/312 [22:59<08:36,  6.08s/it]
Training Epoch: 19/100, batch 454/625 completed (loss: 0.002:  73%|███████▎  | 227/312 [22:59<08:36,  6.08s/it] 
Training Epoch: 19/100, batch 454/625 completed (loss: 0.0052:  73%|███████▎  | 227/312 [22:59<08:37,  6.08s/it]
Training Epoch: 19/100, batch 454/625 completed (loss: 0.0001:  73%|███████▎  | 228/312 [23:02<08:30,  6.08s/it]
Training Epoch: 19/100, batch 454/625 completed (loss: 0.0005:  73%|███████▎  | 228/312 [23:02<08:30,  6.08s/it]
Training Epoch: 19/100, batch 454/625 completed (loss: 0.0052:  73%|███████▎  | 228/312 [23:02<08:30,  6.08s/it]
Training Epoch: 19/100, batch 454/625 completed (loss: 0.002:  73%|███████▎  | 228/312 [23:02<08:30,  6.07s/it]
Training Epoch: 19/100, batch 455/625 completed (loss: 0.0004:  73%|███████▎  | 228/312 [23:02<08:30,  6.08s/it]
Training Epoch: 19/100, batch 455/625 completed (loss: 0.0002:  73%|███████▎  | 228/312 [23:02<08:30,  6.08s/it]
Training Epoch: 19/100, batch 455/625 completed (loss: 0.0015:  73%|███████▎  | 228/312 [23:02<08:30,  6.07s/it]
Training Epoch: 19/100, batch 455/625 completed (loss: 0.0007:  73%|███████▎  | 228/312 [23:02<08:30,  6.08s/it]
Training Epoch: 19/100, batch 456/625 completed (loss: 0.0003:  73%|███████▎  | 228/312 [23:05<08:30,  6.08s/it]
Training Epoch: 19/100, batch 456/625 completed (loss: 0.0008:  73%|███████▎  | 228/312 [23:05<08:30,  6.08s/it]
Training Epoch: 19/100, batch 456/625 completed (loss: 0.0084:  73%|███████▎  | 228/312 [23:05<08:30,  6.07s/it]
Training Epoch: 19/100, batch 456/625 completed (loss: 0.0004:  73%|███████▎  | 228/312 [23:05<08:30,  6.08s/it]
Training Epoch: 19/100, batch 456/625 completed (loss: 0.0008:  73%|███████▎  | 229/312 [23:08<08:18,  6.01s/it]
Training Epoch: 19/100, batch 456/625 completed (loss: 0.0003:  73%|███████▎  | 229/312 [23:08<08:18,  6.01s/it]
Training Epoch: 19/100, batch 456/625 completed (loss: 0.0084:  73%|███████▎  | 229/312 [23:08<08:18,  6.01s/it]
Training Epoch: 19/100, batch 456/625 completed (loss: 0.0004:  73%|███████▎  | 229/312 [23:08<08:19,  6.01s/it]
Training Epoch: 19/100, batch 457/625 completed (loss: 0.0:  73%|███████▎  | 229/312 [23:08<08:18,  6.01s/it]   
Training Epoch: 19/100, batch 457/625 completed (loss: 0.0005:  73%|███████▎  | 229/312 [23:08<08:18,  6.01s/it]
Training Epoch: 19/100, batch 457/625 completed (loss: 0.0028:  73%|███████▎  | 229/312 [23:08<08:18,  6.01s/it]
Training Epoch: 19/100, batch 457/625 completed (loss: 0.0004:  73%|███████▎  | 229/312 [23:08<08:19,  6.01s/it]
Training Epoch: 19/100, batch 458/625 completed (loss: 0.0008:  73%|███████▎  | 229/312 [23:11<08:18,  6.01s/it]
Training Epoch: 19/100, batch 458/625 completed (loss: 0.0002:  73%|███████▎  | 229/312 [23:11<08:18,  6.01s/it]
Training Epoch: 19/100, batch 458/625 completed (loss: 0.0001:  73%|███████▎  | 229/312 [23:11<08:18,  6.01s/it]
Training Epoch: 19/100, batch 458/625 completed (loss: 0.0018:  73%|███████▎  | 229/312 [23:11<08:19,  6.01s/it]
Training Epoch: 19/100, batch 458/625 completed (loss: 0.0008:  74%|███████▎  | 230/312 [23:14<08:16,  6.05s/it]
Training Epoch: 19/100, batch 458/625 completed (loss: 0.0002:  74%|███████▎  | 230/312 [23:14<08:16,  6.05s/it]
Training Epoch: 19/100, batch 458/625 completed (loss: 0.0001:  74%|███████▎  | 230/312 [23:14<08:16,  6.05s/it]
Training Epoch: 19/100, batch 458/625 completed (loss: 0.0018:  74%|███████▎  | 230/312 [23:14<08:16,  6.05s/it]
Training Epoch: 19/100, batch 459/625 completed (loss: 0.0009:  74%|███████▎  | 230/312 [23:14<08:16,  6.05s/it]
Training Epoch: 19/100, batch 459/625 completed (loss: 0.0:  74%|███████▎  | 230/312 [23:14<08:16,  6.05s/it]   
Training Epoch: 19/100, batch 459/625 completed (loss: 0.0002:  74%|███████▎  | 230/312 [23:14<08:16,  6.05s/it]
Training Epoch: 19/100, batch 459/625 completed (loss: 0.0024:  74%|███████▎  | 230/312 [23:14<08:16,  6.05s/it]
Training Epoch: 19/100, batch 460/625 completed (loss: 0.0005:  74%|███████▎  | 230/312 [23:17<08:16,  6.05s/it]
Training Epoch: 19/100, batch 460/625 completed (loss: 0.0002:  74%|███████▎  | 230/312 [23:17<08:16,  6.05s/it]
Training Epoch: 19/100, batch 460/625 completed (loss: 0.0001:  74%|███████▎  | 230/312 [23:17<08:16,  6.05s/it]
Training Epoch: 19/100, batch 460/625 completed (loss: 0.0003:  74%|███████▎  | 230/312 [23:17<08:16,  6.05s/it]
Training Epoch: 19/100, batch 460/625 completed (loss: 0.0005:  74%|███████▍  | 231/312 [23:20<08:14,  6.11s/it]
Training Epoch: 19/100, batch 460/625 completed (loss: 0.0001:  74%|███████▍  | 231/312 [23:20<08:14,  6.11s/it]
Training Epoch: 19/100, batch 460/625 completed (loss: 0.0002:  74%|███████▍  | 231/312 [23:20<08:14,  6.10s/it]
Training Epoch: 19/100, batch 460/625 completed (loss: 0.0003:  74%|███████▍  | 231/312 [23:20<08:14,  6.11s/it]
Training Epoch: 19/100, batch 461/625 completed (loss: 0.0003:  74%|███████▍  | 231/312 [23:20<08:14,  6.11s/it]
Training Epoch: 19/100, batch 461/625 completed (loss: 0.0:  74%|███████▍  | 231/312 [23:20<08:14,  6.11s/it]   
Training Epoch: 19/100, batch 461/625 completed (loss: 0.0002:  74%|███████▍  | 231/312 [23:20<08:14,  6.10s/it]
Training Epoch: 19/100, batch 461/625 completed (loss: 0.0004:  74%|███████▍  | 231/312 [23:20<08:14,  6.11s/it]
Training Epoch: 19/100, batch 462/625 completed (loss: 0.0001:  74%|███████▍  | 231/312 [23:23<08:14,  6.11s/it]
Training Epoch: 19/100, batch 462/625 completed (loss: 0.0004:  74%|███████▍  | 231/312 [23:23<08:14,  6.11s/it]
Training Epoch: 19/100, batch 462/625 completed (loss: 0.0003:  74%|███████▍  | 231/312 [23:23<08:14,  6.10s/it]
Training Epoch: 19/100, batch 462/625 completed (loss: 0.0031:  74%|███████▍  | 231/312 [23:23<08:14,  6.11s/it]
Training Epoch: 19/100, batch 462/625 completed (loss: 0.0004:  74%|███████▍  | 232/312 [23:26<08:09,  6.12s/it]
Training Epoch: 19/100, batch 462/625 completed (loss: 0.0001:  74%|███████▍  | 232/312 [23:26<08:09,  6.12s/it]
Training Epoch: 19/100, batch 462/625 completed (loss: 0.0003:  74%|███████▍  | 232/312 [23:26<08:09,  6.12s/it]
Training Epoch: 19/100, batch 462/625 completed (loss: 0.0031:  74%|███████▍  | 232/312 [23:26<08:09,  6.12s/it]
Training Epoch: 19/100, batch 463/625 completed (loss: 0.0297:  74%|███████▍  | 232/312 [23:26<08:09,  6.12s/it]
Training Epoch: 19/100, batch 463/625 completed (loss: 0.0013:  74%|███████▍  | 232/312 [23:26<08:09,  6.12s/it]
Training Epoch: 19/100, batch 463/625 completed (loss: 0.0011:  74%|███████▍  | 232/312 [23:26<08:09,  6.12s/it]
Training Epoch: 19/100, batch 463/625 completed (loss: 0.0016:  74%|███████▍  | 232/312 [23:26<08:09,  6.12s/it]
Training Epoch: 19/100, batch 464/625 completed (loss: 0.0002:  74%|███████▍  | 232/312 [23:29<08:09,  6.12s/it]
Training Epoch: 19/100, batch 464/625 completed (loss: 0.0004:  74%|███████▍  | 232/312 [23:29<08:09,  6.12s/it]
Training Epoch: 19/100, batch 464/625 completed (loss: 0.0004:  74%|███████▍  | 232/312 [23:29<08:09,  6.12s/it]
Training Epoch: 19/100, batch 464/625 completed (loss: 0.0008:  74%|███████▍  | 232/312 [23:29<08:09,  6.12s/it]
Training Epoch: 19/100, batch 464/625 completed (loss: 0.0002:  75%|███████▍  | 233/312 [23:32<07:57,  6.04s/it]
Training Epoch: 19/100, batch 464/625 completed (loss: 0.0004:  75%|███████▍  | 233/312 [23:32<07:57,  6.04s/it]
Training Epoch: 19/100, batch 464/625 completed (loss: 0.0004:  75%|███████▍  | 233/312 [23:32<07:57,  6.04s/it]
Training Epoch: 19/100, batch 464/625 completed (loss: 0.0008:  75%|███████▍  | 233/312 [23:32<07:57,  6.04s/it]
Training Epoch: 19/100, batch 465/625 completed (loss: 0.0004:  75%|███████▍  | 233/312 [23:32<07:57,  6.04s/it]
Training Epoch: 19/100, batch 465/625 completed (loss: 0.0014:  75%|███████▍  | 233/312 [23:32<07:57,  6.04s/it]
Training Epoch: 19/100, batch 465/625 completed (loss: 0.0:  75%|███████▍  | 233/312 [23:32<07:57,  6.04s/it]   
Training Epoch: 19/100, batch 465/625 completed (loss: 0.0003:  75%|███████▍  | 233/312 [23:32<07:57,  6.04s/it]
Training Epoch: 19/100, batch 466/625 completed (loss: 0.0005:  75%|███████▍  | 233/312 [23:35<07:57,  6.04s/it]
Training Epoch: 19/100, batch 466/625 completed (loss: 0.0:  75%|███████▍  | 233/312 [23:35<07:57,  6.04s/it]
Training Epoch: 19/100, batch 466/625 completed (loss: 0.0014:  75%|███████▍  | 233/312 [23:35<07:57,  6.04s/it]
Training Epoch: 19/100, batch 466/625 completed (loss: 0.0004:  75%|███████▍  | 233/312 [23:36<07:57,  6.04s/it]
Training Epoch: 19/100, batch 466/625 completed (loss: 0.0:  75%|███████▌  | 234/312 [23:38<07:54,  6.08s/it]
Training Epoch: 19/100, batch 466/625 completed (loss: 0.0014:  75%|███████▌  | 234/312 [23:38<07:54,  6.08s/it]
Training Epoch: 19/100, batch 466/625 completed (loss: 0.0005:  75%|███████▌  | 234/312 [23:38<07:54,  6.08s/it]
Training Epoch: 19/100, batch 466/625 completed (loss: 0.0004:  75%|███████▌  | 234/312 [23:38<07:54,  6.08s/it]
Training Epoch: 19/100, batch 467/625 completed (loss: 0.0002:  75%|███████▌  | 234/312 [23:38<07:54,  6.08s/it]
Training Epoch: 19/100, batch 467/625 completed (loss: 0.0003:  75%|███████▌  | 234/312 [23:38<07:54,  6.08s/it]
Training Epoch: 19/100, batch 467/625 completed (loss: 0.0009:  75%|███████▌  | 234/312 [23:38<07:54,  6.08s/it]
Training Epoch: 19/100, batch 467/625 completed (loss: 0.0003:  75%|███████▌  | 234/312 [23:38<07:54,  6.08s/it]
Training Epoch: 19/100, batch 468/625 completed (loss: 0.0003:  75%|███████▌  | 234/312 [23:41<07:54,  6.08s/it]
Training Epoch: 19/100, batch 468/625 completed (loss: 0.0006:  75%|███████▌  | 234/312 [23:41<07:54,  6.08s/it]
Training Epoch: 19/100, batch 468/625 completed (loss: 0.0014:  75%|███████▌  | 234/312 [23:41<07:54,  6.08s/it]
Training Epoch: 19/100, batch 468/625 completed (loss: 0.0003:  75%|███████▌  | 234/312 [23:42<07:54,  6.08s/it]
Training Epoch: 19/100, batch 468/625 completed (loss: 0.0006:  75%|███████▌  | 235/312 [23:44<07:48,  6.09s/it]
Training Epoch: 19/100, batch 468/625 completed (loss: 0.0003:  75%|███████▌  | 235/312 [23:44<07:49,  6.09s/it]
Training Epoch: 19/100, batch 468/625 completed (loss: 0.0014:  75%|███████▌  | 235/312 [23:44<07:49,  6.09s/it]
Training Epoch: 19/100, batch 468/625 completed (loss: 0.0003:  75%|███████▌  | 235/312 [23:45<07:49,  6.09s/it]
Training Epoch: 19/100, batch 469/625 completed (loss: 0.0018:  75%|███████▌  | 235/312 [23:44<07:49,  6.09s/it]
Training Epoch: 19/100, batch 469/625 completed (loss: 0.0098:  75%|███████▌  | 235/312 [23:45<07:49,  6.09s/it]
Training Epoch: 19/100, batch 469/625 completed (loss: 0.0014:  75%|███████▌  | 235/312 [23:45<07:48,  6.09s/it]
Training Epoch: 19/100, batch 469/625 completed (loss: 0.0001:  75%|███████▌  | 235/312 [23:45<07:49,  6.09s/it]
Training Epoch: 19/100, batch 470/625 completed (loss: 0.0021:  75%|███████▌  | 235/312 [23:47<07:49,  6.09s/it]
Training Epoch: 19/100, batch 470/625 completed (loss: 0.0035:  75%|███████▌  | 235/312 [23:47<07:49,  6.09s/it]
Training Epoch: 19/100, batch 470/625 completed (loss: 0.008:  75%|███████▌  | 235/312 [23:47<07:48,  6.09s/it] 
Training Epoch: 19/100, batch 470/625 completed (loss: 0.0009:  75%|███████▌  | 235/312 [23:48<07:49,  6.09s/it]
Training Epoch: 19/100, batch 470/625 completed (loss: 0.0035:  76%|███████▌  | 236/312 [23:50<07:41,  6.07s/it]
Training Epoch: 19/100, batch 470/625 completed (loss: 0.008:  76%|███████▌  | 236/312 [23:50<07:41,  6.07s/it]
Training Epoch: 19/100, batch 470/625 completed (loss: 0.0021:  76%|███████▌  | 236/312 [23:50<07:41,  6.08s/it]
Training Epoch: 19/100, batch 470/625 completed (loss: 0.0009:  76%|███████▌  | 236/312 [23:51<07:41,  6.08s/it]
Training Epoch: 19/100, batch 471/625 completed (loss: 0.0014:  76%|███████▌  | 236/312 [23:51<07:41,  6.07s/it]
Training Epoch: 19/100, batch 471/625 completed (loss: 0.0012:  76%|███████▌  | 236/312 [23:51<07:41,  6.08s/it]
Training Epoch: 19/100, batch 471/625 completed (loss: 0.0012:  76%|███████▌  | 236/312 [23:51<07:41,  6.07s/it]
Training Epoch: 19/100, batch 471/625 completed (loss: 0.0002:  76%|███████▌  | 236/312 [23:51<07:41,  6.08s/it]
Training Epoch: 19/100, batch 472/625 completed (loss: 0.0289:  76%|███████▌  | 236/312 [23:54<07:41,  6.07s/it]
Training Epoch: 19/100, batch 472/625 completed (loss: 0.0001:  76%|███████▌  | 236/312 [23:54<07:41,  6.08s/it]
Training Epoch: 19/100, batch 472/625 completed (loss: 0.078:  76%|███████▌  | 236/312 [23:54<07:41,  6.07s/it] 
Training Epoch: 19/100, batch 472/625 completed (loss: 0.0027:  76%|███████▌  | 236/312 [23:54<07:41,  6.08s/it]
Training Epoch: 19/100, batch 472/625 completed (loss: 0.078:  76%|███████▌  | 237/312 [23:57<07:38,  6.11s/it]
Training Epoch: 19/100, batch 472/625 completed (loss: 0.0001:  76%|███████▌  | 237/312 [23:57<07:38,  6.11s/it]
Training Epoch: 19/100, batch 472/625 completed (loss: 0.0027:  76%|███████▌  | 237/312 [23:57<07:38,  6.11s/it]
Training Epoch: 19/100, batch 472/625 completed (loss: 0.0289:  76%|███████▌  | 237/312 [23:57<07:38,  6.12s/it]
Training Epoch: 19/100, batch 473/625 completed (loss: 0.0002:  76%|███████▌  | 237/312 [23:57<07:38,  6.12s/it]
Training Epoch: 19/100, batch 473/625 completed (loss: 0.0001:  76%|███████▌  | 237/312 [23:57<07:38,  6.11s/it]
Training Epoch: 19/100, batch 473/625 completed (loss: 0.0002:  76%|███████▌  | 237/312 [23:57<07:38,  6.11s/it]
Training Epoch: 19/100, batch 473/625 completed (loss: 0.0003:  76%|███████▌  | 237/312 [23:57<07:38,  6.11s/it]
Training Epoch: 19/100, batch 474/625 completed (loss: 0.0011:  76%|███████▌  | 237/312 [24:00<07:38,  6.11s/it]
Training Epoch: 19/100, batch 474/625 completed (loss: 0.0015:  76%|███████▌  | 237/312 [24:00<07:38,  6.11s/it]
Training Epoch: 19/100, batch 474/625 completed (loss: 0.0272:  76%|███████▌  | 237/312 [24:00<07:38,  6.12s/it]
Training Epoch: 19/100, batch 474/625 completed (loss: 0.0014:  76%|███████▌  | 237/312 [24:00<07:38,  6.11s/it]
Training Epoch: 19/100, batch 474/625 completed (loss: 0.0015:  76%|███████▋  | 238/312 [24:03<07:30,  6.08s/it]
Training Epoch: 19/100, batch 474/625 completed (loss: 0.0011:  76%|███████▋  | 238/312 [24:03<07:30,  6.08s/it]
Training Epoch: 19/100, batch 474/625 completed (loss: 0.0014:  76%|███████▋  | 238/312 [24:03<07:30,  6.08s/it]
Training Epoch: 19/100, batch 474/625 completed (loss: 0.0272:  76%|███████▋  | 238/312 [24:03<07:30,  6.09s/it]
Training Epoch: 19/100, batch 475/625 completed (loss: 0.0015:  76%|███████▋  | 238/312 [24:03<07:30,  6.09s/it]
Training Epoch: 19/100, batch 475/625 completed (loss: 0.0009:  76%|███████▋  | 238/312 [24:03<07:30,  6.08s/it]
Training Epoch: 19/100, batch 475/625 completed (loss: 0.0004:  76%|███████▋  | 238/312 [24:03<07:30,  6.08s/it]
Training Epoch: 19/100, batch 475/625 completed (loss: 0.0008:  76%|███████▋  | 238/312 [24:03<07:30,  6.08s/it]
Training Epoch: 19/100, batch 476/625 completed (loss: 0.0002:  76%|███████▋  | 238/312 [24:06<07:30,  6.08s/it]
Training Epoch: 19/100, batch 476/625 completed (loss: 0.0024:  76%|███████▋  | 238/312 [24:06<07:30,  6.09s/it]
Training Epoch: 19/100, batch 476/625 completed (loss: 0.0003:  76%|███████▋  | 238/312 [24:06<07:30,  6.08s/it]
Training Epoch: 19/100, batch 476/625 completed (loss: 0.0435:  76%|███████▋  | 238/312 [24:06<07:30,  6.08s/it]
Training Epoch: 19/100, batch 476/625 completed (loss: 0.0002:  77%|███████▋  | 239/312 [24:09<07:27,  6.13s/it]
Training Epoch: 19/100, batch 476/625 completed (loss: 0.0003:  77%|███████▋  | 239/312 [24:09<07:27,  6.13s/it]
Training Epoch: 19/100, batch 476/625 completed (loss: 0.0435:  77%|███████▋  | 239/312 [24:09<07:27,  6.13s/it]
Training Epoch: 19/100, batch 476/625 completed (loss: 0.0024:  77%|███████▋  | 239/312 [24:09<07:27,  6.13s/it]
Training Epoch: 19/100, batch 477/625 completed (loss: 0.0018:  77%|███████▋  | 239/312 [24:09<07:27,  6.13s/it]
Training Epoch: 19/100, batch 477/625 completed (loss: 0.0007:  77%|███████▋  | 239/312 [24:09<07:27,  6.13s/it]
Training Epoch: 19/100, batch 477/625 completed (loss: 0.0013:  77%|███████▋  | 239/312 [24:09<07:27,  6.13s/it]
Training Epoch: 19/100, batch 477/625 completed (loss: 0.0001:  77%|███████▋  | 239/312 [24:09<07:27,  6.13s/it]
Training Epoch: 19/100, batch 478/625 completed (loss: 0.0023:  77%|███████▋  | 239/312 [24:12<07:27,  6.13s/it]
Training Epoch: 19/100, batch 478/625 completed (loss: 0.0007:  77%|███████▋  | 239/312 [24:12<07:27,  6.13s/it]
Training Epoch: 19/100, batch 478/625 completed (loss: 0.0001:  77%|███████▋  | 239/312 [24:12<07:27,  6.13s/it]
Training Epoch: 19/100, batch 478/625 completed (loss: 0.0001:  77%|███████▋  | 239/312 [24:12<07:27,  6.13s/it]
Training Epoch: 19/100, batch 478/625 completed (loss: 0.0023:  77%|███████▋  | 240/312 [24:15<07:21,  6.14s/it]
Training Epoch: 19/100, batch 478/625 completed (loss: 0.0001:  77%|███████▋  | 240/312 [24:15<07:22,  6.14s/it]
Training Epoch: 19/100, batch 478/625 completed (loss: 0.0007:  77%|███████▋  | 240/312 [24:15<07:21,  6.14s/it]
Training Epoch: 19/100, batch 478/625 completed (loss: 0.0001:  77%|███████▋  | 240/312 [24:15<07:22,  6.14s/it]
Training Epoch: 19/100, batch 479/625 completed (loss: 0.0014:  77%|███████▋  | 240/312 [24:15<07:21,  6.14s/it]
Training Epoch: 19/100, batch 479/625 completed (loss: 0.0002:  77%|███████▋  | 240/312 [24:15<07:22,  6.14s/it]
Training Epoch: 19/100, batch 479/625 completed (loss: 0.0012:  77%|███████▋  | 240/312 [24:15<07:21,  6.14s/it]
Training Epoch: 19/100, batch 479/625 completed (loss: 0.0032:  77%|███████▋  | 240/312 [24:15<07:22,  6.14s/it]
Training Epoch: 19/100, batch 480/625 completed (loss: 0.0002:  77%|███████▋  | 240/312 [24:18<07:21,  6.14s/it]
Training Epoch: 19/100, batch 480/625 completed (loss: 0.0003:  77%|███████▋  | 240/312 [24:18<07:21,  6.14s/it]
Training Epoch: 19/100, batch 480/625 completed (loss: 0.0006:  77%|███████▋  | 240/312 [24:18<07:22,  6.14s/it]
Training Epoch: 19/100, batch 480/625 completed (loss: 0.005:  77%|███████▋  | 240/312 [24:18<07:22,  6.14s/it] 
Training Epoch: 19/100, batch 480/625 completed (loss: 0.0003:  77%|███████▋  | 241/312 [24:21<07:12,  6.09s/it]
Training Epoch: 19/100, batch 480/625 completed (loss: 0.0002:  77%|███████▋  | 241/312 [24:21<07:12,  6.10s/it]
Training Epoch: 19/100, batch 480/625 completed (loss: 0.0006:  77%|███████▋  | 241/312 [24:21<07:13,  6.10s/it]
Training Epoch: 19/100, batch 480/625 completed (loss: 0.005:  77%|███████▋  | 241/312 [24:21<07:12,  6.10s/it]
Training Epoch: 19/100, batch 481/625 completed (loss: 0.0007:  77%|███████▋  | 241/312 [24:21<07:12,  6.10s/it]
Training Epoch: 19/100, batch 481/625 completed (loss: 0.0001:  77%|███████▋  | 241/312 [24:21<07:12,  6.09s/it]
Training Epoch: 19/100, batch 481/625 completed (loss: 0.001:  77%|███████▋  | 241/312 [24:21<07:13,  6.10s/it] 
Training Epoch: 19/100, batch 481/625 completed (loss: 0.0062:  77%|███████▋  | 241/312 [24:21<07:12,  6.10s/it]
Training Epoch: 19/100, batch 482/625 completed (loss: 0.011:  77%|███████▋  | 241/312 [24:24<07:12,  6.09s/it] 
Training Epoch: 19/100, batch 482/625 completed (loss: 0.001:  77%|███████▋  | 241/312 [24:24<07:13,  6.10s/it]
Training Epoch: 19/100, batch 482/625 completed (loss: 0.0001:  77%|███████▋  | 241/312 [24:24<07:12,  6.10s/it]
Training Epoch: 19/100, batch 482/625 completed (loss: 0.0037:  77%|███████▋  | 241/312 [24:24<07:12,  6.10s/it]
Training Epoch: 19/100, batch 482/625 completed (loss: 0.001:  78%|███████▊  | 242/312 [24:27<07:05,  6.08s/it]
Training Epoch: 19/100, batch 482/625 completed (loss: 0.0001:  78%|███████▊  | 242/312 [24:27<07:05,  6.08s/it]
Training Epoch: 19/100, batch 482/625 completed (loss: 0.0037:  78%|███████▊  | 242/312 [24:27<07:05,  6.08s/it]
Training Epoch: 19/100, batch 482/625 completed (loss: 0.011:  78%|███████▊  | 242/312 [24:27<07:05,  6.08s/it]
Training Epoch: 19/100, batch 483/625 completed (loss: 0.0115:  78%|███████▊  | 242/312 [24:27<07:05,  6.08s/it]
Training Epoch: 19/100, batch 483/625 completed (loss: 0.0004:  78%|███████▊  | 242/312 [24:27<07:05,  6.08s/it]
Training Epoch: 19/100, batch 483/625 completed (loss: 0.0102:  78%|███████▊  | 242/312 [24:27<07:05,  6.08s/it]
Training Epoch: 19/100, batch 483/625 completed (loss: 0.0:  78%|███████▊  | 242/312 [24:27<07:05,  6.08s/it]   
Training Epoch: 19/100, batch 484/625 completed (loss: 0.0029:  78%|███████▊  | 242/312 [24:30<07:05,  6.08s/it]
Training Epoch: 19/100, batch 484/625 completed (loss: 0.1157:  78%|███████▊  | 242/312 [24:30<07:05,  6.08s/it]
Training Epoch: 19/100, batch 484/625 completed (loss: 0.0123:  78%|███████▊  | 242/312 [24:30<07:05,  6.08s/it]
Training Epoch: 19/100, batch 484/625 completed (loss: 0.0002:  78%|███████▊  | 242/312 [24:30<07:05,  6.08s/it]
Training Epoch: 19/100, batch 484/625 completed (loss: 0.0029:  78%|███████▊  | 243/312 [24:33<07:00,  6.10s/it]
Training Epoch: 19/100, batch 484/625 completed (loss: 0.0123:  78%|███████▊  | 243/312 [24:33<07:00,  6.10s/it]
Training Epoch: 19/100, batch 484/625 completed (loss: 0.1157:  78%|███████▊  | 243/312 [24:33<07:01,  6.10s/it]
Training Epoch: 19/100, batch 484/625 completed (loss: 0.0002:  78%|███████▊  | 243/312 [24:33<07:00,  6.10s/it]
Training Epoch: 19/100, batch 485/625 completed (loss: 0.0002:  78%|███████▊  | 243/312 [24:33<07:01,  6.10s/it]
Training Epoch: 19/100, batch 485/625 completed (loss: 0.0025:  78%|███████▊  | 243/312 [24:33<07:00,  6.10s/it]
Training Epoch: 19/100, batch 485/625 completed (loss: 0.0001:  78%|███████▊  | 243/312 [24:33<07:00,  6.10s/it]
Training Epoch: 19/100, batch 485/625 completed (loss: 0.0002:  78%|███████▊  | 243/312 [24:33<07:00,  6.10s/it]
Training Epoch: 19/100, batch 486/625 completed (loss: 0.0008:  78%|███████▊  | 243/312 [24:36<07:01,  6.10s/it]
Training Epoch: 19/100, batch 486/625 completed (loss: 0.0017:  78%|███████▊  | 243/312 [24:36<07:00,  6.10s/it]
Training Epoch: 19/100, batch 486/625 completed (loss: 0.007:  78%|███████▊  | 243/312 [24:36<07:00,  6.10s/it] 
Training Epoch: 19/100, batch 486/625 completed (loss: 0.0002:  78%|███████▊  | 243/312 [24:36<07:00,  6.10s/it]
Training Epoch: 19/100, batch 486/625 completed (loss: 0.0008:  78%|███████▊  | 244/312 [24:39<06:53,  6.08s/it]
Training Epoch: 19/100, batch 486/625 completed (loss: 0.007:  78%|███████▊  | 244/312 [24:39<06:53,  6.08s/it]
Training Epoch: 19/100, batch 486/625 completed (loss: 0.0017:  78%|███████▊  | 244/312 [24:39<06:53,  6.08s/it]
Training Epoch: 19/100, batch 486/625 completed (loss: 0.0002:  78%|███████▊  | 244/312 [24:39<06:53,  6.08s/it]
Training Epoch: 19/100, batch 487/625 completed (loss: 0.0069:  78%|███████▊  | 244/312 [24:39<06:53,  6.08s/it]
Training Epoch: 19/100, batch 487/625 completed (loss: 0.0019:  78%|███████▊  | 244/312 [24:39<06:53,  6.08s/it]
Training Epoch: 19/100, batch 487/625 completed (loss: 0.0028:  78%|███████▊  | 244/312 [24:39<06:53,  6.08s/it]
Training Epoch: 19/100, batch 487/625 completed (loss: 0.0002:  78%|███████▊  | 244/312 [24:39<06:53,  6.08s/it]
Training Epoch: 19/100, batch 488/625 completed (loss: 0.0013:  78%|███████▊  | 244/312 [24:42<06:53,  6.08s/it]
Training Epoch: 19/100, batch 488/625 completed (loss: 0.0001:  78%|███████▊  | 244/312 [24:42<06:53,  6.08s/it]
Training Epoch: 19/100, batch 488/625 completed (loss: 0.0001:  78%|███████▊  | 244/312 [24:42<06:53,  6.08s/it]
Training Epoch: 19/100, batch 488/625 completed (loss: 0.0001:  78%|███████▊  | 244/312 [24:42<06:53,  6.08s/it]
Training Epoch: 19/100, batch 488/625 completed (loss: 0.0013:  79%|███████▊  | 245/312 [24:46<06:51,  6.14s/it]
Training Epoch: 19/100, batch 488/625 completed (loss: 0.0001:  79%|███████▊  | 245/312 [24:46<06:51,  6.14s/it]
Training Epoch: 19/100, batch 488/625 completed (loss: 0.0001:  79%|███████▊  | 245/312 [24:46<06:51,  6.14s/it]
Training Epoch: 19/100, batch 488/625 completed (loss: 0.0001:  79%|███████▊  | 245/312 [24:46<06:51,  6.14s/it]
Training Epoch: 19/100, batch 489/625 completed (loss: 0.0063:  79%|███████▊  | 245/312 [24:46<06:51,  6.14s/it]
Training Epoch: 19/100, batch 489/625 completed (loss: 0.0014:  79%|███████▊  | 245/312 [24:46<06:51,  6.14s/it]
Training Epoch: 19/100, batch 489/625 completed (loss: 0.0006:  79%|███████▊  | 245/312 [24:46<06:51,  6.14s/it]
Training Epoch: 19/100, batch 489/625 completed (loss: 0.001:  79%|███████▊  | 245/312 [24:46<06:51,  6.14s/it] 
Training Epoch: 19/100, batch 490/625 completed (loss: 0.0012:  79%|███████▊  | 245/312 [24:49<06:51,  6.14s/it]
Training Epoch: 19/100, batch 490/625 completed (loss: 0.0034:  79%|███████▊  | 245/312 [24:49<06:51,  6.14s/it]
Training Epoch: 19/100, batch 490/625 completed (loss: 0.0004:  79%|███████▊  | 245/312 [24:49<06:51,  6.14s/it]
Training Epoch: 19/100, batch 490/625 completed (loss: 0.0009:  79%|███████▊  | 245/312 [24:49<06:51,  6.14s/it]
Training Epoch: 19/100, batch 490/625 completed (loss: 0.0034:  79%|███████▉  | 246/312 [24:51<06:39,  6.06s/it]
Training Epoch: 19/100, batch 490/625 completed (loss: 0.0004:  79%|███████▉  | 246/312 [24:51<06:39,  6.06s/it]
Training Epoch: 19/100, batch 490/625 completed (loss: 0.0012:  79%|███████▉  | 246/312 [24:51<06:40,  6.06s/it]
Training Epoch: 19/100, batch 490/625 completed (loss: 0.0009:  79%|███████▉  | 246/312 [24:52<06:39,  6.06s/it]
Training Epoch: 19/100, batch 491/625 completed (loss: 0.0126:  79%|███████▉  | 246/312 [24:52<06:39,  6.06s/it]
Training Epoch: 19/100, batch 491/625 completed (loss: 0.0005:  79%|███████▉  | 246/312 [24:52<06:40,  6.06s/it]
Training Epoch: 19/100, batch 491/625 completed (loss: 0.0035:  79%|███████▉  | 246/312 [24:52<06:39,  6.06s/it]
Training Epoch: 19/100, batch 491/625 completed (loss: 0.0037:  79%|███████▉  | 246/312 [24:52<06:39,  6.06s/it]
Training Epoch: 19/100, batch 492/625 completed (loss: 0.0003:  79%|███████▉  | 246/312 [24:55<06:39,  6.06s/it]
Training Epoch: 19/100, batch 492/625 completed (loss: 0.0022:  79%|███████▉  | 246/312 [24:55<06:39,  6.06s/it]
Training Epoch: 19/100, batch 492/625 completed (loss: 0.0006:  79%|███████▉  | 246/312 [24:55<06:40,  6.06s/it]
Training Epoch: 19/100, batch 492/625 completed (loss: 0.0005:  79%|███████▉  | 246/312 [24:55<06:39,  6.06s/it]
Training Epoch: 19/100, batch 492/625 completed (loss: 0.0006:  79%|███████▉  | 247/312 [24:57<06:33,  6.05s/it]
Training Epoch: 19/100, batch 492/625 completed (loss: 0.0003:  79%|███████▉  | 247/312 [24:57<06:33,  6.05s/it]
Training Epoch: 19/100, batch 492/625 completed (loss: 0.0022:  79%|███████▉  | 247/312 [24:58<06:33,  6.05s/it]
Training Epoch: 19/100, batch 492/625 completed (loss: 0.0005:  79%|███████▉  | 247/312 [24:58<06:33,  6.05s/it]
Training Epoch: 19/100, batch 493/625 completed (loss: 0.0001:  79%|███████▉  | 247/312 [24:58<06:33,  6.05s/it]
Training Epoch: 19/100, batch 493/625 completed (loss: 0.0007:  79%|███████▉  | 247/312 [24:58<06:33,  6.05s/it]
Training Epoch: 19/100, batch 493/625 completed (loss: 0.0018:  79%|███████▉  | 247/312 [24:58<06:33,  6.05s/it]
Training Epoch: 19/100, batch 493/625 completed (loss: 0.0017:  79%|███████▉  | 247/312 [24:58<06:33,  6.05s/it]
Training Epoch: 19/100, batch 494/625 completed (loss: 0.006:  79%|███████▉  | 247/312 [25:01<06:33,  6.05s/it] 
Training Epoch: 19/100, batch 494/625 completed (loss: 0.0002:  79%|███████▉  | 247/312 [25:01<06:33,  6.05s/it]
Training Epoch: 19/100, batch 494/625 completed (loss: 0.0046:  79%|███████▉  | 247/312 [25:01<06:33,  6.05s/it]
Training Epoch: 19/100, batch 494/625 completed (loss: 0.0043:  79%|███████▉  | 247/312 [25:01<06:33,  6.05s/it]
Training Epoch: 19/100, batch 494/625 completed (loss: 0.0002:  79%|███████▉  | 248/312 [25:04<06:30,  6.10s/it]
Training Epoch: 19/100, batch 494/625 completed (loss: 0.0043:  79%|███████▉  | 248/312 [25:04<06:30,  6.10s/it]
Training Epoch: 19/100, batch 494/625 completed (loss: 0.006:  79%|███████▉  | 248/312 [25:04<06:30,  6.10s/it]
Training Epoch: 19/100, batch 494/625 completed (loss: 0.0046:  79%|███████▉  | 248/312 [25:04<06:30,  6.11s/it]
Training Epoch: 19/100, batch 495/625 completed (loss: 0.0002:  79%|███████▉  | 248/312 [25:04<06:30,  6.10s/it]
Training Epoch: 19/100, batch 495/625 completed (loss: 0.0001:  79%|███████▉  | 248/312 [25:04<06:30,  6.10s/it]
Training Epoch: 19/100, batch 495/625 completed (loss: 0.0135:  79%|███████▉  | 248/312 [25:04<06:30,  6.11s/it]
Training Epoch: 19/100, batch 495/625 completed (loss: 0.0004:  79%|███████▉  | 248/312 [25:04<06:30,  6.10s/it]
Training Epoch: 19/100, batch 496/625 completed (loss: 0.0001:  79%|███████▉  | 248/312 [25:07<06:30,  6.10s/it]
Training Epoch: 19/100, batch 496/625 completed (loss: 0.0007:  79%|███████▉  | 248/312 [25:07<06:30,  6.11s/it]
Training Epoch: 19/100, batch 496/625 completed (loss: 0.0001:  79%|███████▉  | 248/312 [25:07<06:30,  6.10s/it]
Training Epoch: 19/100, batch 496/625 completed (loss: 0.001:  79%|███████▉  | 248/312 [25:07<06:30,  6.10s/it] 
Training Epoch: 19/100, batch 496/625 completed (loss: 0.0007:  80%|███████▉  | 249/312 [25:10<06:23,  6.08s/it]
Training Epoch: 19/100, batch 496/625 completed (loss: 0.0001:  80%|███████▉  | 249/312 [25:10<06:23,  6.08s/it]
Training Epoch: 19/100, batch 496/625 completed (loss: 0.0001:  80%|███████▉  | 249/312 [25:10<06:23,  6.08s/it]
Training Epoch: 19/100, batch 496/625 completed (loss: 0.001:  80%|███████▉  | 249/312 [25:10<06:23,  6.08s/it]
Training Epoch: 19/100, batch 497/625 completed (loss: 0.0007:  80%|███████▉  | 249/312 [25:10<06:23,  6.08s/it]
Training Epoch: 19/100, batch 497/625 completed (loss: 0.0:  80%|███████▉  | 249/312 [25:10<06:23,  6.08s/it]   
Training Epoch: 19/100, batch 497/625 completed (loss: 0.0008:  80%|███████▉  | 249/312 [25:10<06:23,  6.08s/it]
Training Epoch: 19/100, batch 497/625 completed (loss: 0.0018:  80%|███████▉  | 249/312 [25:10<06:23,  6.08s/it]
Training Epoch: 19/100, batch 498/625 completed (loss: 0.0001:  80%|███████▉  | 249/312 [25:13<06:23,  6.08s/it]
Training Epoch: 19/100, batch 498/625 completed (loss: 0.0002:  80%|███████▉  | 249/312 [25:13<06:23,  6.08s/it]
Training Epoch: 19/100, batch 498/625 completed (loss: 0.0:  80%|███████▉  | 249/312 [25:13<06:23,  6.08s/it]
Training Epoch: 19/100, batch 498/625 completed (loss: 0.0003:  80%|███████▉  | 249/312 [25:13<06:23,  6.08s/it]
Training Epoch: 19/100, batch 498/625 completed (loss: 0.0002:  80%|████████  | 250/312 [25:16<06:18,  6.10s/it]
Training Epoch: 19/100, batch 498/625 completed (loss: 0.0:  80%|████████  | 250/312 [25:16<06:18,  6.11s/it]
Training Epoch: 19/100, batch 498/625 completed (loss: 0.0003:  80%|████████  | 250/312 [25:16<06:18,  6.11s/it]
Training Epoch: 19/100, batch 498/625 completed (loss: 0.0001:  80%|████████  | 250/312 [25:16<06:18,  6.11s/it]
Training Epoch: 19/100, batch 499/625 completed (loss: 0.0008:  80%|████████  | 250/312 [25:16<06:18,  6.11s/it]
Training Epoch: 19/100, batch 499/625 completed (loss: 0.0008:  80%|████████  | 250/312 [25:16<06:18,  6.10s/it]
Training Epoch: 19/100, batch 499/625 completed (loss: 0.0003:  80%|████████  | 250/312 [25:16<06:18,  6.11s/it]
Training Epoch: 19/100, batch 499/625 completed (loss: 0.0002:  80%|████████  | 250/312 [25:16<06:18,  6.11s/it]
Training Epoch: 19/100, batch 500/625 completed (loss: 0.0012:  80%|████████  | 250/312 [25:19<06:18,  6.11s/it]
Training Epoch: 19/100, batch 500/625 completed (loss: 0.0121:  80%|████████  | 250/312 [25:19<06:18,  6.10s/it]
Training Epoch: 19/100, batch 500/625 completed (loss: 0.0001:  80%|████████  | 250/312 [25:19<06:18,  6.11s/it]
Training Epoch: 19/100, batch 500/625 completed (loss: 0.0004:  80%|████████  | 250/312 [25:19<06:18,  6.11s/it]
Training Epoch: 19/100, batch 500/625 completed (loss: 0.0012:  80%|████████  | 251/312 [25:22<06:10,  6.08s/it]
Training Epoch: 19/100, batch 500/625 completed (loss: 0.0121:  80%|████████  | 251/312 [25:22<06:10,  6.08s/it]
Training Epoch: 19/100, batch 500/625 completed (loss: 0.0001:  80%|████████  | 251/312 [25:22<06:10,  6.08s/it]
Training Epoch: 19/100, batch 500/625 completed (loss: 0.0004:  80%|████████  | 251/312 [25:22<06:10,  6.08s/it]
Training Epoch: 19/100, batch 501/625 completed (loss: 0.0004:  80%|████████  | 251/312 [25:22<06:10,  6.08s/it]
Training Epoch: 19/100, batch 501/625 completed (loss: 0.0119:  80%|████████  | 251/312 [25:22<06:10,  6.08s/it]
Training Epoch: 19/100, batch 501/625 completed (loss: 0.002:  80%|████████  | 251/312 [25:22<06:10,  6.08s/it] 
Training Epoch: 19/100, batch 501/625 completed (loss: 0.0002:  80%|████████  | 251/312 [25:22<06:10,  6.08s/it]
Training Epoch: 19/100, batch 502/625 completed (loss: 0.0:  80%|████████  | 251/312 [25:25<06:10,  6.08s/it]   
Training Epoch: 19/100, batch 502/625 completed (loss: 0.001:  80%|████████  | 251/312 [25:25<06:10,  6.08s/it] 
Training Epoch: 19/100, batch 502/625 completed (loss: 0.0031:  80%|████████  | 251/312 [25:25<06:10,  6.08s/it]
Training Epoch: 19/100, batch 502/625 completed (loss: 0.001:  80%|████████  | 251/312 [25:25<06:10,  6.08s/it] 
Training Epoch: 19/100, batch 502/625 completed (loss: 0.0:  81%|████████  | 252/312 [25:28<06:04,  6.07s/it]
Training Epoch: 19/100, batch 502/625 completed (loss: 0.001:  81%|████████  | 252/312 [25:28<06:04,  6.08s/it]
Training Epoch: 19/100, batch 502/625 completed (loss: 0.001:  81%|████████  | 252/312 [25:28<06:04,  6.07s/it]
Training Epoch: 19/100, batch 502/625 completed (loss: 0.0031:  81%|████████  | 252/312 [25:28<06:04,  6.08s/it]
Training Epoch: 19/100, batch 503/625 completed (loss: 0.0004:  81%|████████  | 252/312 [25:28<06:04,  6.07s/it]
Training Epoch: 19/100, batch 503/625 completed (loss: 0.0017:  81%|████████  | 252/312 [25:28<06:04,  6.08s/it]
Training Epoch: 19/100, batch 503/625 completed (loss: 0.0006:  81%|████████  | 252/312 [25:28<06:04,  6.08s/it]
Training Epoch: 19/100, batch 503/625 completed (loss: 0.0001:  81%|████████  | 252/312 [25:28<06:04,  6.07s/it]
Training Epoch: 19/100, batch 504/625 completed (loss: 0.0004:  81%|████████  | 252/312 [25:31<06:04,  6.08s/it]
Training Epoch: 19/100, batch 504/625 completed (loss: 0.0008:  81%|████████  | 252/312 [25:31<06:04,  6.07s/it]
Training Epoch: 19/100, batch 504/625 completed (loss: 0.006:  81%|████████  | 252/312 [25:31<06:04,  6.08s/it] 
Training Epoch: 19/100, batch 504/625 completed (loss: 0.0025:  81%|████████  | 252/312 [25:31<06:04,  6.07s/it]
Training Epoch: 19/100, batch 504/625 completed (loss: 0.0004:  81%|████████  | 253/312 [25:34<06:02,  6.15s/it]
Training Epoch: 19/100, batch 504/625 completed (loss: 0.0025:  81%|████████  | 253/312 [25:34<06:02,  6.15s/it]
Training Epoch: 19/100, batch 504/625 completed (loss: 0.0008:  81%|████████  | 253/312 [25:34<06:03,  6.15s/it]
Training Epoch: 19/100, batch 504/625 completed (loss: 0.006:  81%|████████  | 253/312 [25:34<06:03,  6.15s/it]
Training Epoch: 19/100, batch 505/625 completed (loss: 0.0002:  81%|████████  | 253/312 [25:34<06:03,  6.15s/it]
Training Epoch: 19/100, batch 505/625 completed (loss: 0.0005:  81%|████████  | 253/312 [25:34<06:03,  6.15s/it]
Training Epoch: 19/100, batch 505/625 completed (loss: 0.0002:  81%|████████  | 253/312 [25:34<06:02,  6.15s/it]
Training Epoch: 19/100, batch 505/625 completed (loss: 0.0004:  81%|████████  | 253/312 [25:34<06:02,  6.15s/it]
Training Epoch: 19/100, batch 506/625 completed (loss: 0.0035:  81%|████████  | 253/312 [25:37<06:03,  6.15s/it]
Training Epoch: 19/100, batch 506/625 completed (loss: 0.0019:  81%|████████  | 253/312 [25:37<06:02,  6.15s/it]
Training Epoch: 19/100, batch 506/625 completed (loss: 0.0009:  81%|████████  | 253/312 [25:37<06:03,  6.15s/it]
Training Epoch: 19/100, batch 506/625 completed (loss: 0.0004:  81%|████████  | 253/312 [25:37<06:02,  6.15s/it]
Training Epoch: 19/100, batch 506/625 completed (loss: 0.0019:  81%|████████▏ | 254/312 [25:40<05:54,  6.12s/it]
Training Epoch: 19/100, batch 506/625 completed (loss: 0.0009:  81%|████████▏ | 254/312 [25:40<05:54,  6.12s/it]
Training Epoch: 19/100, batch 506/625 completed (loss: 0.0004:  81%|████████▏ | 254/312 [25:40<05:54,  6.12s/it]
Training Epoch: 19/100, batch 506/625 completed (loss: 0.0035:  81%|████████▏ | 254/312 [25:40<05:54,  6.12s/it]
Training Epoch: 19/100, batch 507/625 completed (loss: 0.0006:  81%|████████▏ | 254/312 [25:40<05:54,  6.12s/it]
Training Epoch: 19/100, batch 507/625 completed (loss: 0.0012:  81%|████████▏ | 254/312 [25:40<05:54,  6.12s/it]
Training Epoch: 19/100, batch 507/625 completed (loss: 0.0001:  81%|████████▏ | 254/312 [25:40<05:54,  6.12s/it]
Training Epoch: 19/100, batch 507/625 completed (loss: 0.0005:  81%|████████▏ | 254/312 [25:40<05:54,  6.12s/it]
Training Epoch: 19/100, batch 508/625 completed (loss: 0.0131:  81%|████████▏ | 254/312 [25:43<05:54,  6.12s/it]
Training Epoch: 19/100, batch 508/625 completed (loss: 0.0011:  81%|████████▏ | 254/312 [25:43<05:54,  6.12s/it]
Training Epoch: 19/100, batch 508/625 completed (loss: 0.0019:  81%|████████▏ | 254/312 [25:43<05:54,  6.12s/it]
Training Epoch: 19/100, batch 508/625 completed (loss: 0.0009:  81%|████████▏ | 254/312 [25:43<05:54,  6.12s/it]
Training Epoch: 19/100, batch 508/625 completed (loss: 0.0019:  82%|████████▏ | 255/312 [25:46<05:46,  6.08s/it]
Training Epoch: 19/100, batch 508/625 completed (loss: 0.0131:  82%|████████▏ | 255/312 [25:46<05:46,  6.08s/it]
Training Epoch: 19/100, batch 508/625 completed (loss: 0.0009:  82%|████████▏ | 255/312 [25:46<05:46,  6.08s/it]
Training Epoch: 19/100, batch 508/625 completed (loss: 0.0011:  82%|████████▏ | 255/312 [25:46<05:46,  6.08s/it]
Training Epoch: 19/100, batch 509/625 completed (loss: 0.0003:  82%|████████▏ | 255/312 [25:46<05:46,  6.08s/it]
Training Epoch: 19/100, batch 509/625 completed (loss: 0.0015:  82%|████████▏ | 255/312 [25:46<05:46,  6.08s/it]
Training Epoch: 19/100, batch 509/625 completed (loss: 0.0036:  82%|████████▏ | 255/312 [25:46<05:46,  6.08s/it]
Training Epoch: 19/100, batch 509/625 completed (loss: 0.0003:  82%|████████▏ | 255/312 [25:46<05:46,  6.08s/it]
Training Epoch: 19/100, batch 510/625 completed (loss: 0.0003:  82%|████████▏ | 255/312 [25:50<05:46,  6.08s/it]
Training Epoch: 19/100, batch 510/625 completed (loss: 0.0044:  82%|████████▏ | 255/312 [25:49<05:46,  6.08s/it]
Training Epoch: 19/100, batch 510/625 completed (loss: 0.0005:  82%|████████▏ | 255/312 [25:50<05:46,  6.08s/it]
Training Epoch: 19/100, batch 510/625 completed (loss: 0.0081:  82%|████████▏ | 255/312 [25:50<05:46,  6.08s/it]
Training Epoch: 19/100, batch 510/625 completed (loss: 0.0044:  82%|████████▏ | 256/312 [25:53<05:41,  6.11s/it]
Training Epoch: 19/100, batch 510/625 completed (loss: 0.0081:  82%|████████▏ | 256/312 [25:53<05:41,  6.10s/it]
Training Epoch: 19/100, batch 510/625 completed (loss: 0.0003:  82%|████████▏ | 256/312 [25:53<05:41,  6.11s/it]
Training Epoch: 19/100, batch 510/625 completed (loss: 0.0005:  82%|████████▏ | 256/312 [25:53<05:42,  6.11s/it]
Training Epoch: 19/100, batch 511/625 completed (loss: 0.0033:  82%|████████▏ | 256/312 [25:53<05:42,  6.11s/it]
Training Epoch: 19/100, batch 511/625 completed (loss: 0.0032:  82%|████████▏ | 256/312 [25:53<05:41,  6.11s/it]
Training Epoch: 19/100, batch 511/625 completed (loss: 0.0008:  82%|████████▏ | 256/312 [25:53<05:41,  6.11s/it]
Training Epoch: 19/100, batch 511/625 completed (loss: 0.0004:  82%|████████▏ | 256/312 [25:53<05:41,  6.10s/it]
Training Epoch: 19/100, batch 512/625 completed (loss: 0.0001:  82%|████████▏ | 256/312 [25:56<05:42,  6.11s/it]
Training Epoch: 19/100, batch 512/625 completed (loss: 0.001:  82%|████████▏ | 256/312 [25:56<05:41,  6.11s/it] 
Training Epoch: 19/100, batch 512/625 completed (loss: 0.0092:  82%|████████▏ | 256/312 [25:56<05:41,  6.11s/it]
Training Epoch: 19/100, batch 512/625 completed (loss: 0.0:  82%|████████▏ | 256/312 [25:56<05:41,  6.10s/it]   
Training Epoch: 19/100, batch 512/625 completed (loss: 0.001:  82%|████████▏ | 257/312 [25:59<05:34,  6.08s/it]
Training Epoch: 19/100, batch 512/625 completed (loss: 0.0092:  82%|████████▏ | 257/312 [25:59<05:34,  6.08s/it]
Training Epoch: 19/100, batch 512/625 completed (loss: 0.0:  82%|████████▏ | 257/312 [25:59<05:34,  6.08s/it]
Training Epoch: 19/100, batch 512/625 completed (loss: 0.0001:  82%|████████▏ | 257/312 [25:59<05:34,  6.08s/it]
Training Epoch: 19/100, batch 513/625 completed (loss: 0.0:  82%|████████▏ | 257/312 [25:59<05:34,  6.08s/it]   
Training Epoch: 19/100, batch 513/625 completed (loss: 0.0:  82%|████████▏ | 257/312 [25:59<05:34,  6.08s/it]   
Training Epoch: 19/100, batch 513/625 completed (loss: 0.0003:  82%|████████▏ | 257/312 [25:59<05:34,  6.08s/it]
Training Epoch: 19/100, batch 513/625 completed (loss: 0.0044:  82%|████████▏ | 257/312 [25:59<05:34,  6.08s/it]
Training Epoch: 19/100, batch 514/625 completed (loss: 0.0007:  82%|████████▏ | 257/312 [26:02<05:34,  6.08s/it]
Training Epoch: 19/100, batch 514/625 completed (loss: 0.0023:  82%|████████▏ | 257/312 [26:02<05:34,  6.08s/it]
Training Epoch: 19/100, batch 514/625 completed (loss: 0.004:  82%|████████▏ | 257/312 [26:02<05:34,  6.08s/it]
Training Epoch: 19/100, batch 514/625 completed (loss: 0.0034:  82%|████████▏ | 257/312 [26:02<05:34,  6.08s/it]
Training Epoch: 19/100, batch 514/625 completed (loss: 0.0023:  83%|████████▎ | 258/312 [26:05<05:29,  6.11s/it]
Training Epoch: 19/100, batch 514/625 completed (loss: 0.0034:  83%|████████▎ | 258/312 [26:05<05:29,  6.11s/it]
Training Epoch: 19/100, batch 514/625 completed (loss: 0.0007:  83%|████████▎ | 258/312 [26:05<05:30,  6.11s/it]
Training Epoch: 19/100, batch 514/625 completed (loss: 0.004:  83%|████████▎ | 258/312 [26:05<05:30,  6.11s/it]
Training Epoch: 19/100, batch 515/625 completed (loss: 0.0017:  83%|████████▎ | 258/312 [26:05<05:30,  6.11s/it]
Training Epoch: 19/100, batch 515/625 completed (loss: 0.0002:  83%|████████▎ | 258/312 [26:05<05:29,  6.11s/it]
Training Epoch: 19/100, batch 515/625 completed (loss: 0.0037:  83%|████████▎ | 258/312 [26:05<05:30,  6.11s/it]
Training Epoch: 19/100, batch 515/625 completed (loss: 0.0053:  83%|████████▎ | 258/312 [26:05<05:29,  6.11s/it]
Training Epoch: 19/100, batch 516/625 completed (loss: 0.0016:  83%|████████▎ | 258/312 [26:08<05:30,  6.11s/it]
Training Epoch: 19/100, batch 516/625 completed (loss: 0.001:  83%|████████▎ | 258/312 [26:08<05:29,  6.11s/it] 
Training Epoch: 19/100, batch 516/625 completed (loss: 0.0115:  83%|████████▎ | 258/312 [26:08<05:30,  6.11s/it]
Training Epoch: 19/100, batch 516/625 completed (loss: 0.0004:  83%|████████▎ | 258/312 [26:08<05:29,  6.11s/it]
Training Epoch: 19/100, batch 516/625 completed (loss: 0.0004:  83%|████████▎ | 259/312 [26:11<05:22,  6.08s/it]
Training Epoch: 19/100, batch 516/625 completed (loss: 0.001:  83%|████████▎ | 259/312 [26:11<05:22,  6.08s/it]
Training Epoch: 19/100, batch 516/625 completed (loss: 0.0115:  83%|████████▎ | 259/312 [26:11<05:22,  6.09s/it]
Training Epoch: 19/100, batch 516/625 completed (loss: 0.0016:  83%|████████▎ | 259/312 [26:11<05:22,  6.09s/it]
Training Epoch: 19/100, batch 517/625 completed (loss: 0.0002:  83%|████████▎ | 259/312 [26:11<05:22,  6.08s/it]
Training Epoch: 19/100, batch 517/625 completed (loss: 0.0:  83%|████████▎ | 259/312 [26:11<05:22,  6.09s/it]   
Training Epoch: 19/100, batch 517/625 completed (loss: 0.0001:  83%|████████▎ | 259/312 [26:11<05:22,  6.09s/it]
Training Epoch: 19/100, batch 517/625 completed (loss: 0.0002:  83%|████████▎ | 259/312 [26:11<05:22,  6.08s/it]
Training Epoch: 19/100, batch 518/625 completed (loss: 0.0004:  83%|████████▎ | 259/312 [26:14<05:22,  6.09s/it]
Training Epoch: 19/100, batch 518/625 completed (loss: 0.0016:  83%|████████▎ | 259/312 [26:14<05:22,  6.09s/it]
Training Epoch: 19/100, batch 518/625 completed (loss: 0.0008:  83%|████████▎ | 259/312 [26:14<05:22,  6.08s/it]
Training Epoch: 19/100, batch 518/625 completed (loss: 0.0005:  83%|████████▎ | 259/312 [26:14<05:22,  6.08s/it]
Training Epoch: 19/100, batch 518/625 completed (loss: 0.0004:  83%|████████▎ | 260/312 [26:17<05:15,  6.06s/it]
Training Epoch: 19/100, batch 518/625 completed (loss: 0.0008:  83%|████████▎ | 260/312 [26:17<05:15,  6.06s/it]
Training Epoch: 19/100, batch 518/625 completed (loss: 0.0005:  83%|████████▎ | 260/312 [26:17<05:15,  6.06s/it]
Training Epoch: 19/100, batch 518/625 completed (loss: 0.0016:  83%|████████▎ | 260/312 [26:17<05:15,  6.06s/it]
Training Epoch: 19/100, batch 519/625 completed (loss: 0.0039:  83%|████████▎ | 260/312 [26:17<05:15,  6.06s/it]
Training Epoch: 19/100, batch 519/625 completed (loss: 0.0029:  83%|████████▎ | 260/312 [26:17<05:15,  6.06s/it]
Training Epoch: 19/100, batch 519/625 completed (loss: 0.0024:  83%|████████▎ | 260/312 [26:17<05:15,  6.06s/it]
Training Epoch: 19/100, batch 519/625 completed (loss: 0.0017:  83%|████████▎ | 260/312 [26:17<05:15,  6.06s/it]
Training Epoch: 19/100, batch 520/625 completed (loss: 0.0008:  83%|████████▎ | 260/312 [26:20<05:15,  6.06s/it]
Training Epoch: 19/100, batch 520/625 completed (loss: 0.0004:  83%|████████▎ | 260/312 [26:20<05:15,  6.06s/it]
Training Epoch: 19/100, batch 520/625 completed (loss: 0.0056:  83%|████████▎ | 260/312 [26:20<05:15,  6.06s/it]
Training Epoch: 19/100, batch 520/625 completed (loss: 0.0001:  83%|████████▎ | 260/312 [26:20<05:15,  6.06s/it]
Training Epoch: 19/100, batch 520/625 completed (loss: 0.0004:  84%|████████▎ | 261/312 [26:23<05:12,  6.13s/it]
Training Epoch: 19/100, batch 520/625 completed (loss: 0.0001:  84%|████████▎ | 261/312 [26:23<05:12,  6.13s/it]
Training Epoch: 19/100, batch 520/625 completed (loss: 0.0008:  84%|████████▎ | 261/312 [26:23<05:12,  6.13s/it]
Training Epoch: 19/100, batch 520/625 completed (loss: 0.0056:  84%|████████▎ | 261/312 [26:23<05:12,  6.13s/it]
Training Epoch: 19/100, batch 521/625 completed (loss: 0.0011:  84%|████████▎ | 261/312 [26:23<05:12,  6.13s/it]
Training Epoch: 19/100, batch 521/625 completed (loss: 0.0009:  84%|████████▎ | 261/312 [26:23<05:12,  6.13s/it]
Training Epoch: 19/100, batch 521/625 completed (loss: 0.0001:  84%|████████▎ | 261/312 [26:23<05:12,  6.13s/it]
Training Epoch: 19/100, batch 521/625 completed (loss: 0.0003:  84%|████████▎ | 261/312 [26:23<05:12,  6.13s/it]
Training Epoch: 19/100, batch 522/625 completed (loss: 0.0008:  84%|████████▎ | 261/312 [26:26<05:12,  6.13s/it]
Training Epoch: 19/100, batch 522/625 completed (loss: 0.0005:  84%|████████▎ | 261/312 [26:26<05:12,  6.13s/it]
Training Epoch: 19/100, batch 522/625 completed (loss: 0.0004:  84%|████████▎ | 261/312 [26:26<05:12,  6.13s/it]
Training Epoch: 19/100, batch 522/625 completed (loss: 0.0049:  84%|████████▎ | 261/312 [26:26<05:12,  6.13s/it]
Training Epoch: 19/100, batch 522/625 completed (loss: 0.0049:  84%|████████▍ | 262/312 [26:29<05:04,  6.10s/it]
Training Epoch: 19/100, batch 522/625 completed (loss: 0.0004:  84%|████████▍ | 262/312 [26:29<05:05,  6.10s/it]
Training Epoch: 19/100, batch 522/625 completed (loss: 0.0005:  84%|████████▍ | 262/312 [26:29<05:05,  6.10s/it]
Training Epoch: 19/100, batch 522/625 completed (loss: 0.0008:  84%|████████▍ | 262/312 [26:29<05:05,  6.10s/it]
Training Epoch: 19/100, batch 523/625 completed (loss: 0.0013:  84%|████████▍ | 262/312 [26:29<05:05,  6.10s/it]
Training Epoch: 19/100, batch 523/625 completed (loss: 0.0001:  84%|████████▍ | 262/312 [26:29<05:05,  6.10s/it]
Training Epoch: 19/100, batch 523/625 completed (loss: 0.0005:  84%|████████▍ | 262/312 [26:29<05:05,  6.10s/it]
Training Epoch: 19/100, batch 523/625 completed (loss: 0.0021:  84%|████████▍ | 262/312 [26:29<05:04,  6.10s/it]
Training Epoch: 19/100, batch 524/625 completed (loss: 0.0008:  84%|████████▍ | 262/312 [26:32<05:05,  6.10s/it]
Training Epoch: 19/100, batch 524/625 completed (loss: 0.0002:  84%|████████▍ | 262/312 [26:32<05:05,  6.10s/it]
Training Epoch: 19/100, batch 524/625 completed (loss: 0.0007:  84%|████████▍ | 262/312 [26:32<05:05,  6.10s/it]
Training Epoch: 19/100, batch 524/625 completed (loss: 0.0238:  84%|████████▍ | 262/312 [26:32<05:04,  6.10s/it]
Training Epoch: 19/100, batch 524/625 completed (loss: 0.0002:  84%|████████▍ | 263/312 [26:35<04:58,  6.08s/it]
Training Epoch: 19/100, batch 524/625 completed (loss: 0.0008:  84%|████████▍ | 263/312 [26:35<04:58,  6.08s/it]
Training Epoch: 19/100, batch 524/625 completed (loss: 0.0238:  84%|████████▍ | 263/312 [26:35<04:58,  6.09s/it]
Training Epoch: 19/100, batch 524/625 completed (loss: 0.0007:  84%|████████▍ | 263/312 [26:35<04:58,  6.09s/it]
Training Epoch: 19/100, batch 525/625 completed (loss: 0.0:  84%|████████▍ | 263/312 [26:35<04:58,  6.08s/it]   
Training Epoch: 19/100, batch 525/625 completed (loss: 0.0001:  84%|████████▍ | 263/312 [26:35<04:58,  6.08s/it]
Training Epoch: 19/100, batch 525/625 completed (loss: 0.0001:  84%|████████▍ | 263/312 [26:35<04:58,  6.09s/it]
Training Epoch: 19/100, batch 525/625 completed (loss: 0.0011:  84%|████████▍ | 263/312 [26:35<04:58,  6.09s/it]
Training Epoch: 19/100, batch 526/625 completed (loss: 0.0001:  84%|████████▍ | 263/312 [26:38<04:58,  6.08s/it]
Training Epoch: 19/100, batch 526/625 completed (loss: 0.02:  84%|████████▍ | 263/312 [26:38<04:58,  6.09s/it]  
Training Epoch: 19/100, batch 526/625 completed (loss: 0.0005:  84%|████████▍ | 263/312 [26:38<04:58,  6.08s/it]
Training Epoch: 19/100, batch 526/625 completed (loss: 0.0003:  84%|████████▍ | 263/312 [26:38<04:58,  6.09s/it]
Training Epoch: 19/100, batch 526/625 completed (loss: 0.0001:  85%|████████▍ | 264/312 [26:41<04:53,  6.11s/it]
Training Epoch: 19/100, batch 526/625 completed (loss: 0.02:  85%|████████▍ | 264/312 [26:41<04:53,  6.11s/it]
Training Epoch: 19/100, batch 526/625 completed (loss: 0.0003:  85%|████████▍ | 264/312 [26:41<04:53,  6.11s/it]
Training Epoch: 19/100, batch 526/625 completed (loss: 0.0005:  85%|████████▍ | 264/312 [26:41<04:53,  6.11s/it]
Training Epoch: 19/100, batch 527/625 completed (loss: 0.0006:  85%|████████▍ | 264/312 [26:41<04:53,  6.11s/it]
Training Epoch: 19/100, batch 527/625 completed (loss: 0.0019:  85%|████████▍ | 264/312 [26:41<04:53,  6.11s/it]
Training Epoch: 19/100, batch 527/625 completed (loss: 0.0215:  85%|████████▍ | 264/312 [26:41<04:53,  6.11s/it]
Training Epoch: 19/100, batch 527/625 completed (loss: 0.0005:  85%|████████▍ | 264/312 [26:41<04:53,  6.11s/it]
Training Epoch: 19/100, batch 528/625 completed (loss: 0.0007:  85%|████████▍ | 264/312 [26:44<04:53,  6.11s/it]
Training Epoch: 19/100, batch 528/625 completed (loss: 0.0029:  85%|████████▍ | 264/312 [26:44<04:53,  6.11s/it]
Training Epoch: 19/100, batch 528/625 completed (loss: 0.0003:  85%|████████▍ | 264/312 [26:44<04:53,  6.11s/it]
Training Epoch: 19/100, batch 528/625 completed (loss: 0.0005:  85%|████████▍ | 264/312 [26:44<04:53,  6.11s/it]
Training Epoch: 19/100, batch 528/625 completed (loss: 0.0029:  85%|████████▍ | 265/312 [26:47<04:45,  6.07s/it]
Training Epoch: 19/100, batch 528/625 completed (loss: 0.0005:  85%|████████▍ | 265/312 [26:47<04:45,  6.07s/it]
Training Epoch: 19/100, batch 528/625 completed (loss: 0.0007:  85%|████████▍ | 265/312 [26:47<04:45,  6.07s/it]
Training Epoch: 19/100, batch 528/625 completed (loss: 0.0003:  85%|████████▍ | 265/312 [26:47<04:45,  6.07s/it]
Training Epoch: 19/100, batch 529/625 completed (loss: 0.0001:  85%|████████▍ | 265/312 [26:47<04:45,  6.07s/it]
Training Epoch: 19/100, batch 529/625 completed (loss: 0.0042:  85%|████████▍ | 265/312 [26:47<04:45,  6.07s/it]
Training Epoch: 19/100, batch 529/625 completed (loss: 0.0004:  85%|████████▍ | 265/312 [26:47<04:45,  6.07s/it]
Training Epoch: 19/100, batch 529/625 completed (loss: 0.0156:  85%|████████▍ | 265/312 [26:47<04:45,  6.07s/it]
Training Epoch: 19/100, batch 530/625 completed (loss: 0.0008:  85%|████████▍ | 265/312 [26:50<04:45,  6.07s/it]
Training Epoch: 19/100, batch 530/625 completed (loss: 0.0001:  85%|████████▍ | 265/312 [26:50<04:45,  6.07s/it]
Training Epoch: 19/100, batch 530/625 completed (loss: 0.0002:  85%|████████▍ | 265/312 [26:50<04:45,  6.07s/it]
Training Epoch: 19/100, batch 530/625 completed (loss: 0.0016:  85%|████████▍ | 265/312 [26:50<04:45,  6.07s/it]
Training Epoch: 19/100, batch 530/625 completed (loss: 0.0016:  85%|████████▌ | 266/312 [26:54<04:42,  6.14s/it]
Training Epoch: 19/100, batch 530/625 completed (loss: 0.0002:  85%|████████▌ | 266/312 [26:54<04:42,  6.15s/it]
Training Epoch: 19/100, batch 530/625 completed (loss: 0.0001:  85%|████████▌ | 266/312 [26:54<04:42,  6.15s/it]
Training Epoch: 19/100, batch 530/625 completed (loss: 0.0008:  85%|████████▌ | 266/312 [26:54<04:42,  6.15s/it]
Training Epoch: 19/100, batch 531/625 completed (loss: 0.0:  85%|████████▌ | 266/312 [26:54<04:42,  6.15s/it]   
Training Epoch: 19/100, batch 531/625 completed (loss: 0.0001:  85%|████████▌ | 266/312 [26:54<04:42,  6.15s/it]
Training Epoch: 19/100, batch 531/625 completed (loss: 0.0036:  85%|████████▌ | 266/312 [26:54<04:42,  6.15s/it]
Training Epoch: 19/100, batch 531/625 completed (loss: 0.0046:  85%|████████▌ | 266/312 [26:54<04:42,  6.14s/it]
Training Epoch: 19/100, batch 532/625 completed (loss: 0.0003:  85%|████████▌ | 266/312 [26:57<04:42,  6.15s/it]
Training Epoch: 19/100, batch 532/625 completed (loss: 0.0003:  85%|████████▌ | 266/312 [26:57<04:42,  6.15s/it]
Training Epoch: 19/100, batch 532/625 completed (loss: 0.0006:  85%|████████▌ | 266/312 [26:57<04:42,  6.15s/it]
Training Epoch: 19/100, batch 532/625 completed (loss: 0.0003:  85%|████████▌ | 266/312 [26:57<04:42,  6.14s/it]
Training Epoch: 19/100, batch 532/625 completed (loss: 0.0003:  86%|████████▌ | 267/312 [26:59<04:32,  6.06s/it]
Training Epoch: 19/100, batch 532/625 completed (loss: 0.0006:  86%|████████▌ | 267/312 [26:59<04:32,  6.06s/it]
Training Epoch: 19/100, batch 532/625 completed (loss: 0.0003:  86%|████████▌ | 267/312 [27:00<04:32,  6.06s/it]
Training Epoch: 19/100, batch 532/625 completed (loss: 0.0003:  86%|████████▌ | 267/312 [26:59<04:32,  6.06s/it]
Training Epoch: 19/100, batch 533/625 completed (loss: 0.0058:  86%|████████▌ | 267/312 [27:00<04:32,  6.06s/it]
Training Epoch: 19/100, batch 533/625 completed (loss: 0.0027:  86%|████████▌ | 267/312 [27:00<04:32,  6.06s/it]
Training Epoch: 19/100, batch 533/625 completed (loss: 0.0005:  86%|████████▌ | 267/312 [27:00<04:32,  6.06s/it]
Training Epoch: 19/100, batch 533/625 completed (loss: 0.0006:  86%|████████▌ | 267/312 [27:00<04:32,  6.06s/it]
Training Epoch: 19/100, batch 534/625 completed (loss: 0.0:  86%|████████▌ | 267/312 [27:03<04:32,  6.06s/it]   
Training Epoch: 19/100, batch 534/625 completed (loss: 0.0003:  86%|████████▌ | 267/312 [27:03<04:32,  6.06s/it]
Training Epoch: 19/100, batch 534/625 completed (loss: 0.002:  86%|████████▌ | 267/312 [27:03<04:32,  6.06s/it] 
Training Epoch: 19/100, batch 534/625 completed (loss: 0.006:  86%|████████▌ | 267/312 [27:03<04:32,  6.06s/it] 
Training Epoch: 19/100, batch 534/625 completed (loss: 0.0:  86%|████████▌ | 268/312 [27:06<04:28,  6.10s/it]
Training Epoch: 19/100, batch 534/625 completed (loss: 0.006:  86%|████████▌ | 268/312 [27:06<04:28,  6.10s/it]
Training Epoch: 19/100, batch 534/625 completed (loss: 0.002:  86%|████████▌ | 268/312 [27:06<04:28,  6.10s/it]
Training Epoch: 19/100, batch 534/625 completed (loss: 0.0003:  86%|████████▌ | 268/312 [27:06<04:28,  6.11s/it]
Training Epoch: 19/100, batch 535/625 completed (loss: 0.0035:  86%|████████▌ | 268/312 [27:06<04:28,  6.11s/it]
Training Epoch: 19/100, batch 535/625 completed (loss: 0.0082:  86%|████████▌ | 268/312 [27:06<04:28,  6.10s/it]
Training Epoch: 19/100, batch 535/625 completed (loss: 0.0008:  86%|████████▌ | 268/312 [27:06<04:28,  6.10s/it]
Training Epoch: 19/100, batch 535/625 completed (loss: 0.0:  86%|████████▌ | 268/312 [27:06<04:28,  6.10s/it]  
Training Epoch: 19/100, batch 536/625 completed (loss: 0.0004:  86%|████████▌ | 268/312 [27:09<04:28,  6.11s/it]
Training Epoch: 19/100, batch 536/625 completed (loss: 0.0001:  86%|████████▌ | 268/312 [27:09<04:28,  6.10s/it]
Training Epoch: 19/100, batch 536/625 completed (loss: 0.0017:  86%|████████▌ | 268/312 [27:09<04:28,  6.10s/it]
Training Epoch: 19/100, batch 536/625 completed (loss: 0.0051:  86%|████████▌ | 268/312 [27:09<04:28,  6.10s/it]
Training Epoch: 19/100, batch 536/625 completed (loss: 0.0001:  86%|████████▌ | 269/312 [27:12<04:23,  6.12s/it]
Training Epoch: 19/100, batch 536/625 completed (loss: 0.0051:  86%|████████▌ | 269/312 [27:12<04:23,  6.12s/it]
Training Epoch: 19/100, batch 536/625 completed (loss: 0.0017:  86%|████████▌ | 269/312 [27:12<04:23,  6.12s/it]
Training Epoch: 19/100, batch 536/625 completed (loss: 0.0004:  86%|████████▌ | 269/312 [27:12<04:23,  6.12s/it]
Training Epoch: 19/100, batch 537/625 completed (loss: 0.0003:  86%|████████▌ | 269/312 [27:12<04:23,  6.12s/it]
Training Epoch: 19/100, batch 537/625 completed (loss: 0.0003:  86%|████████▌ | 269/312 [27:12<04:23,  6.12s/it]
Training Epoch: 19/100, batch 537/625 completed (loss: 0.0196:  86%|████████▌ | 269/312 [27:12<04:23,  6.12s/it]
Training Epoch: 19/100, batch 537/625 completed (loss: 0.0029:  86%|████████▌ | 269/312 [27:12<04:23,  6.12s/it]
Training Epoch: 19/100, batch 538/625 completed (loss: 0.0465:  86%|████████▌ | 269/312 [27:15<04:23,  6.12s/it]
Training Epoch: 19/100, batch 538/625 completed (loss: 0.0001:  86%|████████▌ | 269/312 [27:15<04:23,  6.12s/it]
Training Epoch: 19/100, batch 538/625 completed (loss: 0.0003:  86%|████████▌ | 269/312 [27:15<04:23,  6.12s/it]
Training Epoch: 19/100, batch 538/625 completed (loss: 0.0002:  86%|████████▌ | 269/312 [27:15<04:23,  6.12s/it]
Training Epoch: 19/100, batch 538/625 completed (loss: 0.0003:  87%|████████▋ | 270/312 [27:18<04:15,  6.09s/it]
Training Epoch: 19/100, batch 538/625 completed (loss: 0.0002:  87%|████████▋ | 270/312 [27:18<04:15,  6.09s/it]
Training Epoch: 19/100, batch 538/625 completed (loss: 0.0001:  87%|████████▋ | 270/312 [27:18<04:15,  6.09s/it]
Training Epoch: 19/100, batch 538/625 completed (loss: 0.0465:  87%|████████▋ | 270/312 [27:18<04:16,  6.10s/it]
Training Epoch: 19/100, batch 539/625 completed (loss: 0.001:  87%|████████▋ | 270/312 [27:18<04:16,  6.10s/it] 
Training Epoch: 19/100, batch 539/625 completed (loss: 0.0013:  87%|████████▋ | 270/312 [27:18<04:15,  6.09s/it]
Training Epoch: 19/100, batch 539/625 completed (loss: 0.0025:  87%|████████▋ | 270/312 [27:18<04:15,  6.09s/it]
Training Epoch: 19/100, batch 539/625 completed (loss: 0.0005:  87%|████████▋ | 270/312 [27:18<04:15,  6.09s/it]
Training Epoch: 19/100, batch 540/625 completed (loss: 0.0029:  87%|████████▋ | 270/312 [27:21<04:16,  6.10s/it]
Training Epoch: 19/100, batch 540/625 completed (loss: 0.0171:  87%|████████▋ | 270/312 [27:21<04:15,  6.09s/it]
Training Epoch: 19/100, batch 540/625 completed (loss: 0.0027:  87%|████████▋ | 270/312 [27:21<04:15,  6.09s/it]
Training Epoch: 19/100, batch 540/625 completed (loss: 0.001:  87%|████████▋ | 270/312 [27:21<04:15,  6.09s/it] 
Training Epoch: 19/100, batch 540/625 completed (loss: 0.0027:  87%|████████▋ | 271/312 [27:24<04:10,  6.11s/it]
Training Epoch: 19/100, batch 540/625 completed (loss: 0.001:  87%|████████▋ | 271/312 [27:24<04:10,  6.10s/it]
Training Epoch: 19/100, batch 540/625 completed (loss: 0.0171:  87%|████████▋ | 271/312 [27:24<04:10,  6.11s/it]
Training Epoch: 19/100, batch 540/625 completed (loss: 0.0029:  87%|████████▋ | 271/312 [27:24<04:10,  6.11s/it]
Training Epoch: 19/100, batch 541/625 completed (loss: 0.0001:  87%|████████▋ | 271/312 [27:24<04:10,  6.11s/it]
Training Epoch: 19/100, batch 541/625 completed (loss: 0.0041:  87%|████████▋ | 271/312 [27:24<04:10,  6.11s/it]
Training Epoch: 19/100, batch 541/625 completed (loss: 0.0:  87%|████████▋ | 271/312 [27:24<04:10,  6.11s/it]   
Training Epoch: 19/100, batch 541/625 completed (loss: 0.0007:  87%|████████▋ | 271/312 [27:24<04:10,  6.10s/it]
Training Epoch: 19/100, batch 542/625 completed (loss: 0.0005:  87%|████████▋ | 271/312 [27:27<04:10,  6.11s/it]
Training Epoch: 19/100, batch 542/625 completed (loss: 0.0247:  87%|████████▋ | 271/312 [27:27<04:10,  6.11s/it]
Training Epoch: 19/100, batch 542/625 completed (loss: 0.0008:  87%|████████▋ | 271/312 [27:27<04:10,  6.11s/it]
Training Epoch: 19/100, batch 542/625 completed (loss: 0.0022:  87%|████████▋ | 271/312 [27:27<04:10,  6.10s/it]
Training Epoch: 19/100, batch 542/625 completed (loss: 0.0008:  87%|████████▋ | 272/312 [27:30<04:02,  6.07s/it]
Training Epoch: 19/100, batch 542/625 completed (loss: 0.0022:  87%|████████▋ | 272/312 [27:30<04:02,  6.07s/it]
Training Epoch: 19/100, batch 542/625 completed (loss: 0.0247:  87%|████████▋ | 272/312 [27:30<04:02,  6.07s/it]
Training Epoch: 19/100, batch 542/625 completed (loss: 0.0005:  87%|████████▋ | 272/312 [27:30<04:03,  6.08s/it]
Training Epoch: 19/100, batch 543/625 completed (loss: 0.0019:  87%|████████▋ | 272/312 [27:30<04:02,  6.07s/it]
Training Epoch: 19/100, batch 543/625 completed (loss: 0.0007:  87%|████████▋ | 272/312 [27:30<04:02,  6.07s/it]
Training Epoch: 19/100, batch 543/625 completed (loss: 0.0022:  87%|████████▋ | 272/312 [27:30<04:03,  6.08s/it]
Training Epoch: 19/100, batch 543/625 completed (loss: 0.0093:  87%|████████▋ | 272/312 [27:30<04:02,  6.07s/it]
Training Epoch: 19/100, batch 544/625 completed (loss: 0.0005:  87%|████████▋ | 272/312 [27:33<04:03,  6.08s/it]
Training Epoch: 19/100, batch 544/625 completed (loss: 0.0008:  87%|████████▋ | 272/312 [27:33<04:02,  6.07s/it]
Training Epoch: 19/100, batch 544/625 completed (loss: 0.0013:  87%|████████▋ | 272/312 [27:33<04:02,  6.07s/it]
Training Epoch: 19/100, batch 544/625 completed (loss: 0.001:  87%|████████▋ | 272/312 [27:33<04:02,  6.07s/it] 
Training Epoch: 19/100, batch 544/625 completed (loss: 0.0008:  88%|████████▊ | 273/312 [27:36<03:57,  6.10s/it]
Training Epoch: 19/100, batch 544/625 completed (loss: 0.0005:  88%|████████▊ | 273/312 [27:36<03:57,  6.09s/it]
Training Epoch: 19/100, batch 544/625 completed (loss: 0.001:  88%|████████▊ | 273/312 [27:36<03:57,  6.09s/it]
Training Epoch: 19/100, batch 544/625 completed (loss: 0.0013:  88%|████████▊ | 273/312 [27:36<03:57,  6.09s/it]
Training Epoch: 19/100, batch 545/625 completed (loss: 0.0003:  88%|████████▊ | 273/312 [27:36<03:57,  6.10s/it]
Training Epoch: 19/100, batch 545/625 completed (loss: 0.0008:  88%|████████▊ | 273/312 [27:36<03:57,  6.09s/it]
Training Epoch: 19/100, batch 545/625 completed (loss: 0.0008:  88%|████████▊ | 273/312 [27:36<03:57,  6.09s/it]
Training Epoch: 19/100, batch 545/625 completed (loss: 0.0003:  88%|████████▊ | 273/312 [27:36<03:57,  6.09s/it]
Training Epoch: 19/100, batch 546/625 completed (loss: 0.0032:  88%|████████▊ | 273/312 [27:39<03:57,  6.09s/it]
Training Epoch: 19/100, batch 546/625 completed (loss: 0.0005:  88%|████████▊ | 273/312 [27:39<03:57,  6.09s/it]
Training Epoch: 19/100, batch 546/625 completed (loss: 0.0177:  88%|████████▊ | 273/312 [27:39<03:57,  6.10s/it]
Training Epoch: 19/100, batch 546/625 completed (loss: 0.0032:  88%|████████▊ | 273/312 [27:39<03:57,  6.09s/it]
Training Epoch: 19/100, batch 546/625 completed (loss: 0.0032:  88%|████████▊ | 274/312 [27:42<03:52,  6.11s/it]
Training Epoch: 19/100, batch 546/625 completed (loss: 0.0005:  88%|████████▊ | 274/312 [27:42<03:52,  6.11s/it]
Training Epoch: 19/100, batch 546/625 completed (loss: 0.0177:  88%|████████▊ | 274/312 [27:42<03:52,  6.12s/it]
Training Epoch: 19/100, batch 546/625 completed (loss: 0.0032:  88%|████████▊ | 274/312 [27:42<03:52,  6.11s/it]
Training Epoch: 19/100, batch 547/625 completed (loss: 0.0082:  88%|████████▊ | 274/312 [27:42<03:52,  6.12s/it]
Training Epoch: 19/100, batch 547/625 completed (loss: 0.0005:  88%|████████▊ | 274/312 [27:42<03:52,  6.11s/it]
Training Epoch: 19/100, batch 547/625 completed (loss: 0.0012:  88%|████████▊ | 274/312 [27:42<03:52,  6.11s/it]
Training Epoch: 19/100, batch 547/625 completed (loss: 0.0001:  88%|████████▊ | 274/312 [27:42<03:52,  6.11s/it]
Training Epoch: 19/100, batch 548/625 completed (loss: 0.0006:  88%|████████▊ | 274/312 [27:45<03:52,  6.11s/it]
Training Epoch: 19/100, batch 548/625 completed (loss: 0.0014:  88%|████████▊ | 274/312 [27:45<03:52,  6.12s/it]
Training Epoch: 19/100, batch 548/625 completed (loss: 0.0014:  88%|████████▊ | 274/312 [27:45<03:52,  6.11s/it]
Training Epoch: 19/100, batch 548/625 completed (loss: 0.0025:  88%|████████▊ | 274/312 [27:45<03:52,  6.11s/it]
Training Epoch: 19/100, batch 548/625 completed (loss: 0.0025:  88%|████████▊ | 275/312 [27:48<03:45,  6.09s/it]
Training Epoch: 19/100, batch 548/625 completed (loss: 0.0014:  88%|████████▊ | 275/312 [27:48<03:45,  6.09s/it]
Training Epoch: 19/100, batch 548/625 completed (loss: 0.0014:  88%|████████▊ | 275/312 [27:48<03:45,  6.09s/it]
Training Epoch: 19/100, batch 548/625 completed (loss: 0.0006:  88%|████████▊ | 275/312 [27:48<03:45,  6.09s/it]
Training Epoch: 19/100, batch 549/625 completed (loss: 0.0003:  88%|████████▊ | 275/312 [27:48<03:45,  6.09s/it]
Training Epoch: 19/100, batch 549/625 completed (loss: 0.0028:  88%|████████▊ | 275/312 [27:48<03:45,  6.09s/it]
Training Epoch: 19/100, batch 549/625 completed (loss: 0.0001:  88%|████████▊ | 275/312 [27:48<03:45,  6.09s/it]
Training Epoch: 19/100, batch 549/625 completed (loss: 0.0028:  88%|████████▊ | 275/312 [27:48<03:45,  6.09s/it]
Training Epoch: 19/100, batch 550/625 completed (loss: 0.0005:  88%|████████▊ | 275/312 [27:51<03:45,  6.09s/it]
Training Epoch: 19/100, batch 550/625 completed (loss: 0.0002:  88%|████████▊ | 275/312 [27:51<03:45,  6.09s/it]
Training Epoch: 19/100, batch 550/625 completed (loss: 0.0002:  88%|████████▊ | 275/312 [27:51<03:45,  6.09s/it]
Training Epoch: 19/100, batch 550/625 completed (loss: 0.0005:  88%|████████▊ | 275/312 [27:52<03:45,  6.09s/it]
Training Epoch: 19/100, batch 550/625 completed (loss: 0.0005:  88%|████████▊ | 276/312 [27:54<03:38,  6.06s/it]
Training Epoch: 19/100, batch 550/625 completed (loss: 0.0002:  88%|████████▊ | 276/312 [27:54<03:38,  6.06s/it]
Training Epoch: 19/100, batch 550/625 completed (loss: 0.0005:  88%|████████▊ | 276/312 [27:54<03:38,  6.06s/it]
Training Epoch: 19/100, batch 550/625 completed (loss: 0.0002:  88%|████████▊ | 276/312 [27:54<03:38,  6.06s/it]
Training Epoch: 19/100, batch 551/625 completed (loss: 0.0138:  88%|████████▊ | 276/312 [27:54<03:38,  6.06s/it]
Training Epoch: 19/100, batch 551/625 completed (loss: 0.0005:  88%|████████▊ | 276/312 [27:54<03:38,  6.06s/it]
Training Epoch: 19/100, batch 551/625 completed (loss: 0.001:  88%|████████▊ | 276/312 [27:54<03:38,  6.06s/it] 
Training Epoch: 19/100, batch 551/625 completed (loss: 0.0012:  88%|████████▊ | 276/312 [27:54<03:38,  6.06s/it]
Training Epoch: 19/100, batch 552/625 completed (loss: 0.0001:  88%|████████▊ | 276/312 [27:57<03:38,  6.06s/it]
Training Epoch: 19/100, batch 552/625 completed (loss: 0.0007:  88%|████████▊ | 276/312 [27:57<03:38,  6.06s/it]
Training Epoch: 19/100, batch 552/625 completed (loss: 0.0001:  88%|████████▊ | 276/312 [27:57<03:38,  6.06s/it]
Training Epoch: 19/100, batch 552/625 completed (loss: 0.0004:  88%|████████▊ | 276/312 [27:57<03:38,  6.06s/it]
Training Epoch: 19/100, batch 552/625 completed (loss: 0.0001:  89%|████████▉ | 277/312 [28:00<03:33,  6.09s/it]
Training Epoch: 19/100, batch 552/625 completed (loss: 0.0007:  89%|████████▉ | 277/312 [28:00<03:33,  6.09s/it]
Training Epoch: 19/100, batch 552/625 completed (loss: 0.0001:  89%|████████▉ | 277/312 [28:00<03:32,  6.08s/it]
Training Epoch: 19/100, batch 552/625 completed (loss: 0.0004:  89%|████████▉ | 277/312 [28:01<03:33,  6.09s/it]
Training Epoch: 19/100, batch 553/625 completed (loss: 0.0473:  89%|████████▉ | 277/312 [28:01<03:33,  6.09s/it]
Training Epoch: 19/100, batch 553/625 completed (loss: 0.0076:  89%|████████▉ | 277/312 [28:01<03:32,  6.08s/it]
Training Epoch: 19/100, batch 553/625 completed (loss: 0.0004:  89%|████████▉ | 277/312 [28:01<03:33,  6.09s/it]
Training Epoch: 19/100, batch 553/625 completed (loss: 0.0004:  89%|████████▉ | 277/312 [28:01<03:33,  6.09s/it]
Training Epoch: 19/100, batch 554/625 completed (loss: 0.0006:  89%|████████▉ | 277/312 [28:04<03:33,  6.09s/it]
Training Epoch: 19/100, batch 554/625 completed (loss: 0.0004:  89%|████████▉ | 277/312 [28:04<03:33,  6.09s/it]
Training Epoch: 19/100, batch 554/625 completed (loss: 0.0001:  89%|████████▉ | 277/312 [28:04<03:32,  6.08s/it]
Training Epoch: 19/100, batch 554/625 completed (loss: 0.0005:  89%|████████▉ | 277/312 [28:04<03:33,  6.09s/it]
Training Epoch: 19/100, batch 554/625 completed (loss: 0.0006:  89%|████████▉ | 278/312 [28:07<03:27,  6.12s/it]
Training Epoch: 19/100, batch 554/625 completed (loss: 0.0005:  89%|████████▉ | 278/312 [28:07<03:27,  6.12s/it]
Training Epoch: 19/100, batch 554/625 completed (loss: 0.0001:  89%|████████▉ | 278/312 [28:07<03:27,  6.11s/it]
Training Epoch: 19/100, batch 554/625 completed (loss: 0.0004:  89%|████████▉ | 278/312 [28:07<03:28,  6.12s/it]
Training Epoch: 19/100, batch 555/625 completed (loss: 0.0004:  89%|████████▉ | 278/312 [28:07<03:28,  6.12s/it]
Training Epoch: 19/100, batch 555/625 completed (loss: 0.0002:  89%|████████▉ | 278/312 [28:07<03:27,  6.12s/it]
Training Epoch: 19/100, batch 555/625 completed (loss: 0.0001:  89%|████████▉ | 278/312 [28:07<03:27,  6.11s/it]
Training Epoch: 19/100, batch 555/625 completed (loss: 0.0002:  89%|████████▉ | 278/312 [28:07<03:27,  6.12s/it]
Training Epoch: 19/100, batch 556/625 completed (loss: 0.0007:  89%|████████▉ | 278/312 [28:10<03:27,  6.12s/it]
Training Epoch: 19/100, batch 556/625 completed (loss: 0.001:  89%|████████▉ | 278/312 [28:10<03:27,  6.11s/it] 
Training Epoch: 19/100, batch 556/625 completed (loss: 0.0004:  89%|████████▉ | 278/312 [28:10<03:28,  6.12s/it]
Training Epoch: 19/100, batch 556/625 completed (loss: 0.0019:  89%|████████▉ | 278/312 [28:10<03:27,  6.12s/it]
Training Epoch: 19/100, batch 556/625 completed (loss: 0.0007:  89%|████████▉ | 279/312 [28:13<03:22,  6.13s/it]
Training Epoch: 19/100, batch 556/625 completed (loss: 0.0019:  89%|████████▉ | 279/312 [28:13<03:22,  6.14s/it]
Training Epoch: 19/100, batch 556/625 completed (loss: 0.0004:  89%|████████▉ | 279/312 [28:13<03:22,  6.14s/it]
Training Epoch: 19/100, batch 556/625 completed (loss: 0.001:  89%|████████▉ | 279/312 [28:13<03:22,  6.14s/it]
Training Epoch: 19/100, batch 557/625 completed (loss: 0.0004:  89%|████████▉ | 279/312 [28:13<03:22,  6.13s/it]
Training Epoch: 19/100, batch 557/625 completed (loss: 0.0106:  89%|████████▉ | 279/312 [28:13<03:22,  6.14s/it]
Training Epoch: 19/100, batch 557/625 completed (loss: 0.0045:  89%|████████▉ | 279/312 [28:13<03:22,  6.14s/it]
Training Epoch: 19/100, batch 557/625 completed (loss: 0.0001:  89%|████████▉ | 279/312 [28:13<03:22,  6.14s/it]
Training Epoch: 19/100, batch 558/625 completed (loss: 0.0015:  89%|████████▉ | 279/312 [28:16<03:22,  6.14s/it]
Training Epoch: 19/100, batch 558/625 completed (loss: 0.0066:  89%|████████▉ | 279/312 [28:16<03:22,  6.14s/it]
Training Epoch: 19/100, batch 558/625 completed (loss: 0.0001:  89%|████████▉ | 279/312 [28:16<03:22,  6.13s/it]
Training Epoch: 19/100, batch 558/625 completed (loss: 0.0001:  89%|████████▉ | 279/312 [28:16<03:22,  6.14s/it]
Training Epoch: 19/100, batch 558/625 completed (loss: 0.0001:  90%|████████▉ | 280/312 [28:19<03:15,  6.11s/it]
Training Epoch: 19/100, batch 558/625 completed (loss: 0.0015:  90%|████████▉ | 280/312 [28:19<03:15,  6.11s/it]
Training Epoch: 19/100, batch 558/625 completed (loss: 0.0001:  90%|████████▉ | 280/312 [28:19<03:15,  6.12s/it]
Training Epoch: 19/100, batch 558/625 completed (loss: 0.0066:  90%|████████▉ | 280/312 [28:19<03:15,  6.11s/it]
Training Epoch: 19/100, batch 559/625 completed (loss: 0.0003:  90%|████████▉ | 280/312 [28:19<03:15,  6.12s/it]
Training Epoch: 19/100, batch 559/625 completed (loss: 0.0003:  90%|████████▉ | 280/312 [28:19<03:15,  6.11s/it]
Training Epoch: 19/100, batch 559/625 completed (loss: 0.0001:  90%|████████▉ | 280/312 [28:19<03:15,  6.11s/it]
Training Epoch: 19/100, batch 559/625 completed (loss: 0.0022:  90%|████████▉ | 280/312 [28:19<03:15,  6.11s/it]
Training Epoch: 19/100, batch 560/625 completed (loss: 0.0009:  90%|████████▉ | 280/312 [28:22<03:15,  6.12s/it]
Training Epoch: 19/100, batch 560/625 completed (loss: 0.0026:  90%|████████▉ | 280/312 [28:22<03:15,  6.11s/it]
Training Epoch: 19/100, batch 560/625 completed (loss: 0.0008:  90%|████████▉ | 280/312 [28:22<03:15,  6.11s/it]
Training Epoch: 19/100, batch 560/625 completed (loss: 0.0012:  90%|████████▉ | 280/312 [28:22<03:15,  6.11s/it]
Training Epoch: 19/100, batch 560/625 completed (loss: 0.0008:  90%|█████████ | 281/312 [28:25<03:08,  6.08s/it]
Training Epoch: 19/100, batch 560/625 completed (loss: 0.0026:  90%|█████████ | 281/312 [28:25<03:08,  6.08s/it]
Training Epoch: 19/100, batch 560/625 completed (loss: 0.0012:  90%|█████████ | 281/312 [28:25<03:08,  6.08s/it]
Training Epoch: 19/100, batch 560/625 completed (loss: 0.0009:  90%|█████████ | 281/312 [28:25<03:08,  6.08s/it]
Training Epoch: 19/100, batch 561/625 completed (loss: 0.0032:  90%|█████████ | 281/312 [28:25<03:08,  6.08s/it]
Training Epoch: 19/100, batch 561/625 completed (loss: 0.0005:  90%|█████████ | 281/312 [28:25<03:08,  6.08s/it]
Training Epoch: 19/100, batch 561/625 completed (loss: 0.0008:  90%|█████████ | 281/312 [28:25<03:08,  6.08s/it]
Training Epoch: 19/100, batch 561/625 completed (loss: 0.0001:  90%|█████████ | 281/312 [28:25<03:08,  6.08s/it]
Training Epoch: 19/100, batch 562/625 completed (loss: 0.0003:  90%|█████████ | 281/312 [28:28<03:08,  6.08s/it]
Training Epoch: 19/100, batch 562/625 completed (loss: 0.003:  90%|█████████ | 281/312 [28:28<03:08,  6.08s/it] 
Training Epoch: 19/100, batch 562/625 completed (loss: 0.0026:  90%|█████████ | 281/312 [28:28<03:08,  6.08s/it]
Training Epoch: 19/100, batch 562/625 completed (loss: 0.0047:  90%|█████████ | 281/312 [28:28<03:08,  6.08s/it]
Training Epoch: 19/100, batch 562/625 completed (loss: 0.0003:  90%|█████████ | 282/312 [28:31<03:02,  6.07s/it]
Training Epoch: 19/100, batch 562/625 completed (loss: 0.0047:  90%|█████████ | 282/312 [28:31<03:02,  6.07s/it]
Training Epoch: 19/100, batch 562/625 completed (loss: 0.003:  90%|█████████ | 282/312 [28:31<03:02,  6.07s/it]
Training Epoch: 19/100, batch 562/625 completed (loss: 0.0026:  90%|█████████ | 282/312 [28:31<03:02,  6.07s/it]
Training Epoch: 19/100, batch 563/625 completed (loss: 0.0001:  90%|█████████ | 282/312 [28:31<03:02,  6.07s/it]
Training Epoch: 19/100, batch 563/625 completed (loss: 0.0012:  90%|█████████ | 282/312 [28:31<03:02,  6.07s/it]
Training Epoch: 19/100, batch 563/625 completed (loss: 0.0004:  90%|█████████ | 282/312 [28:31<03:02,  6.07s/it]
Training Epoch: 19/100, batch 563/625 completed (loss: 0.0018:  90%|█████████ | 282/312 [28:31<03:02,  6.07s/it]
Training Epoch: 19/100, batch 564/625 completed (loss: 0.0004:  90%|█████████ | 282/312 [28:34<03:02,  6.07s/it]
Training Epoch: 19/100, batch 564/625 completed (loss: 0.0011:  90%|█████████ | 282/312 [28:34<03:02,  6.07s/it]
Training Epoch: 19/100, batch 564/625 completed (loss: 0.0002:  90%|█████████ | 282/312 [28:34<03:02,  6.07s/it]
Training Epoch: 19/100, batch 564/625 completed (loss: 0.0001:  90%|█████████ | 282/312 [28:34<03:02,  6.07s/it]
Training Epoch: 19/100, batch 564/625 completed (loss: 0.0001:  91%|█████████ | 283/312 [28:37<02:56,  6.10s/it]
Training Epoch: 19/100, batch 564/625 completed (loss: 0.0002:  91%|█████████ | 283/312 [28:37<02:56,  6.10s/it]
Training Epoch: 19/100, batch 564/625 completed (loss: 0.0004:  91%|█████████ | 283/312 [28:37<02:56,  6.10s/it]
Training Epoch: 19/100, batch 564/625 completed (loss: 0.0011:  91%|█████████ | 283/312 [28:37<02:56,  6.10s/it]
Training Epoch: 19/100, batch 565/625 completed (loss: 0.0007:  91%|█████████ | 283/312 [28:37<02:56,  6.10s/it]
Training Epoch: 19/100, batch 565/625 completed (loss: 0.0002:  91%|█████████ | 283/312 [28:37<02:56,  6.10s/it]
Training Epoch: 19/100, batch 565/625 completed (loss: 0.0:  91%|█████████ | 283/312 [28:37<02:56,  6.10s/it]   
Training Epoch: 19/100, batch 565/625 completed (loss: 0.0027:  91%|█████████ | 283/312 [28:37<02:56,  6.10s/it]
Training Epoch: 19/100, batch 566/625 completed (loss: 0.0001:  91%|█████████ | 283/312 [28:40<02:56,  6.10s/it]
Training Epoch: 19/100, batch 566/625 completed (loss: 0.0011:  91%|█████████ | 283/312 [28:40<02:56,  6.10s/it]
Training Epoch: 19/100, batch 566/625 completed (loss: 0.0012:  91%|█████████ | 283/312 [28:40<02:56,  6.10s/it]
Training Epoch: 19/100, batch 566/625 completed (loss: 0.0002:  91%|█████████ | 283/312 [28:40<02:56,  6.10s/it]
Training Epoch: 19/100, batch 566/625 completed (loss: 0.0011:  91%|█████████ | 284/312 [28:43<02:48,  6.03s/it]
Training Epoch: 19/100, batch 566/625 completed (loss: 0.0002:  91%|█████████ | 284/312 [28:43<02:48,  6.03s/it]
Training Epoch: 19/100, batch 566/625 completed (loss: 0.0001:  91%|█████████ | 284/312 [28:43<02:49,  6.04s/it]
Training Epoch: 19/100, batch 566/625 completed (loss: 0.0012:  91%|█████████ | 284/312 [28:43<02:49,  6.04s/it]
Training Epoch: 19/100, batch 567/625 completed (loss: 0.0001:  91%|█████████ | 284/312 [28:43<02:49,  6.04s/it]
Training Epoch: 19/100, batch 567/625 completed (loss: 0.0062:  91%|█████████ | 284/312 [28:43<02:49,  6.04s/it]
Training Epoch: 19/100, batch 567/625 completed (loss: 0.0069:  91%|█████████ | 284/312 [28:43<02:48,  6.03s/it]
Training Epoch: 19/100, batch 567/625 completed (loss: 0.0003:  91%|█████████ | 284/312 [28:43<02:48,  6.03s/it]
Training Epoch: 19/100, batch 568/625 completed (loss: 0.0052:  91%|█████████ | 284/312 [28:46<02:49,  6.04s/it]
Training Epoch: 19/100, batch 568/625 completed (loss: 0.0003:  91%|█████████ | 284/312 [28:46<02:48,  6.03s/it]
Training Epoch: 19/100, batch 568/625 completed (loss: 0.0001:  91%|█████████ | 284/312 [28:46<02:49,  6.04s/it]
Training Epoch: 19/100, batch 568/625 completed (loss: 0.0004:  91%|█████████ | 284/312 [28:46<02:48,  6.03s/it]
Training Epoch: 19/100, batch 568/625 completed (loss: 0.0003:  91%|█████████▏| 285/312 [28:49<02:42,  6.04s/it]
Training Epoch: 19/100, batch 568/625 completed (loss: 0.0004:  91%|█████████▏| 285/312 [28:49<02:43,  6.04s/it]
Training Epoch: 19/100, batch 568/625 completed (loss: 0.0001:  91%|█████████▏| 285/312 [28:49<02:42,  6.04s/it]
Training Epoch: 19/100, batch 568/625 completed (loss: 0.0052:  91%|█████████▏| 285/312 [28:49<02:42,  6.03s/it]
Training Epoch: 19/100, batch 569/625 completed (loss: 0.0583:  91%|█████████▏| 285/312 [28:49<02:42,  6.03s/it]
Training Epoch: 19/100, batch 569/625 completed (loss: 0.0001:  91%|█████████▏| 285/312 [28:49<02:42,  6.04s/it]
Training Epoch: 19/100, batch 569/625 completed (loss: 0.0053:  91%|█████████▏| 285/312 [28:49<02:42,  6.04s/it]
Training Epoch: 19/100, batch 569/625 completed (loss: 0.0199:  91%|█████████▏| 285/312 [28:49<02:43,  6.04s/it]
Training Epoch: 19/100, batch 570/625 completed (loss: 0.0015:  91%|█████████▏| 285/312 [28:52<02:42,  6.04s/it]
Training Epoch: 19/100, batch 570/625 completed (loss: 0.006:  91%|█████████▏| 285/312 [28:52<02:42,  6.04s/it] 
Training Epoch: 19/100, batch 570/625 completed (loss: 0.0001:  91%|█████████▏| 285/312 [28:52<02:42,  6.03s/it]
Training Epoch: 19/100, batch 570/625 completed (loss: 0.0044:  91%|█████████▏| 285/312 [28:52<02:43,  6.04s/it]
Training Epoch: 19/100, batch 570/625 completed (loss: 0.0044:  92%|█████████▏| 286/312 [28:55<02:38,  6.08s/it]
Training Epoch: 19/100, batch 570/625 completed (loss: 0.0001:  92%|█████████▏| 286/312 [28:55<02:38,  6.08s/it]
Training Epoch: 19/100, batch 570/625 completed (loss: 0.006:  92%|█████████▏| 286/312 [28:55<02:38,  6.09s/it]
Training Epoch: 19/100, batch 570/625 completed (loss: 0.0015:  92%|█████████▏| 286/312 [28:55<02:38,  6.09s/it]
Training Epoch: 19/100, batch 571/625 completed (loss: 0.0029:  92%|█████████▏| 286/312 [28:55<02:38,  6.09s/it]
Training Epoch: 19/100, batch 571/625 completed (loss: 0.0:  92%|█████████▏| 286/312 [28:55<02:38,  6.08s/it]   
Training Epoch: 19/100, batch 571/625 completed (loss: 0.0006:  92%|█████████▏| 286/312 [28:55<02:38,  6.09s/it]
Training Epoch: 19/100, batch 571/625 completed (loss: 0.0235:  92%|█████████▏| 286/312 [28:55<02:38,  6.08s/it]
Training Epoch: 19/100, batch 572/625 completed (loss: 0.0008:  92%|█████████▏| 286/312 [28:58<02:38,  6.09s/it]
Training Epoch: 19/100, batch 572/625 completed (loss: 0.0007:  92%|█████████▏| 286/312 [28:58<02:38,  6.09s/it]
Training Epoch: 19/100, batch 572/625 completed (loss: 0.0002:  92%|█████████▏| 286/312 [28:58<02:38,  6.08s/it]
Training Epoch: 19/100, batch 572/625 completed (loss: 0.001:  92%|█████████▏| 286/312 [28:58<02:38,  6.08s/it] 
Training Epoch: 19/100, batch 572/625 completed (loss: 0.0002:  92%|█████████▏| 287/312 [29:01<02:30,  6.02s/it]
Training Epoch: 19/100, batch 572/625 completed (loss: 0.0008:  92%|█████████▏| 287/312 [29:01<02:30,  6.02s/it]
Training Epoch: 19/100, batch 572/625 completed (loss: 0.001:  92%|█████████▏| 287/312 [29:01<02:30,  6.02s/it]
Training Epoch: 19/100, batch 572/625 completed (loss: 0.0007:  92%|█████████▏| 287/312 [29:01<02:30,  6.02s/it]
Training Epoch: 19/100, batch 573/625 completed (loss: 0.0001:  92%|█████████▏| 287/312 [29:01<02:30,  6.02s/it]
Training Epoch: 19/100, batch 573/625 completed (loss: 0.0013:  92%|█████████▏| 287/312 [29:01<02:30,  6.02s/it]
Training Epoch: 19/100, batch 573/625 completed (loss: 0.0021:  92%|█████████▏| 287/312 [29:01<02:30,  6.02s/it]
Training Epoch: 19/100, batch 573/625 completed (loss: 0.0003:  92%|█████████▏| 287/312 [29:01<02:30,  6.02s/it]
Training Epoch: 19/100, batch 574/625 completed (loss: 0.0006:  92%|█████████▏| 287/312 [29:04<02:30,  6.02s/it]
Training Epoch: 19/100, batch 574/625 completed (loss: 0.0001:  92%|█████████▏| 287/312 [29:04<02:30,  6.02s/it]
Training Epoch: 19/100, batch 574/625 completed (loss: 0.0064:  92%|█████████▏| 287/312 [29:04<02:30,  6.02s/it]
Training Epoch: 19/100, batch 574/625 completed (loss: 0.0003:  92%|█████████▏| 287/312 [29:04<02:30,  6.02s/it]
Training Epoch: 19/100, batch 574/625 completed (loss: 0.0064:  92%|█████████▏| 288/312 [29:07<02:25,  6.07s/it]
Training Epoch: 19/100, batch 574/625 completed (loss: 0.0006:  92%|█████████▏| 288/312 [29:07<02:25,  6.07s/it]
Training Epoch: 19/100, batch 574/625 completed (loss: 0.0003:  92%|█████████▏| 288/312 [29:07<02:25,  6.07s/it]
Training Epoch: 19/100, batch 574/625 completed (loss: 0.0001:  92%|█████████▏| 288/312 [29:07<02:25,  6.08s/it]
Training Epoch: 19/100, batch 575/625 completed (loss: 0.0025:  92%|█████████▏| 288/312 [29:07<02:25,  6.08s/it]
Training Epoch: 19/100, batch 575/625 completed (loss: 0.0004:  92%|█████████▏| 288/312 [29:07<02:25,  6.07s/it]
Training Epoch: 19/100, batch 575/625 completed (loss: 0.0001:  92%|█████████▏| 288/312 [29:07<02:25,  6.07s/it]
Training Epoch: 19/100, batch 575/625 completed (loss: 0.0002:  92%|█████████▏| 288/312 [29:07<02:25,  6.07s/it]
Training Epoch: 19/100, batch 576/625 completed (loss: 0.0017:  92%|█████████▏| 288/312 [29:10<02:25,  6.07s/it]
Training Epoch: 19/100, batch 576/625 completed (loss: 0.021:  92%|█████████▏| 288/312 [29:10<02:25,  6.08s/it] 
Training Epoch: 19/100, batch 576/625 completed (loss: 0.0016:  92%|█████████▏| 288/312 [29:10<02:25,  6.07s/it]
Training Epoch: 19/100, batch 576/625 completed (loss: 0.0001:  92%|█████████▏| 288/312 [29:10<02:25,  6.07s/it]
Training Epoch: 19/100, batch 576/625 completed (loss: 0.0001:  93%|█████████▎| 289/312 [29:13<02:19,  6.05s/it]
Training Epoch: 19/100, batch 576/625 completed (loss: 0.0016:  93%|█████████▎| 289/312 [29:13<02:19,  6.05s/it]
Training Epoch: 19/100, batch 576/625 completed (loss: 0.0017:  93%|█████████▎| 289/312 [29:13<02:19,  6.05s/it]
Training Epoch: 19/100, batch 576/625 completed (loss: 0.021:  93%|█████████▎| 289/312 [29:13<02:19,  6.05s/it]
Training Epoch: 19/100, batch 577/625 completed (loss: 0.0003:  93%|█████████▎| 289/312 [29:13<02:19,  6.05s/it]
Training Epoch: 19/100, batch 577/625 completed (loss: 0.0004:  93%|█████████▎| 289/312 [29:13<02:19,  6.05s/it]
Training Epoch: 19/100, batch 577/625 completed (loss: 0.0013:  93%|█████████▎| 289/312 [29:13<02:19,  6.05s/it]
Training Epoch: 19/100, batch 577/625 completed (loss: 0.0137:  93%|█████████▎| 289/312 [29:13<02:19,  6.05s/it]
Training Epoch: 19/100, batch 578/625 completed (loss: 0.0005:  93%|█████████▎| 289/312 [29:16<02:19,  6.05s/it]
Training Epoch: 19/100, batch 578/625 completed (loss: 0.0002:  93%|█████████▎| 289/312 [29:16<02:19,  6.05s/it]
Training Epoch: 19/100, batch 578/625 completed (loss: 0.0041:  93%|█████████▎| 289/312 [29:16<02:19,  6.05s/it]
Training Epoch: 19/100, batch 578/625 completed (loss: 0.0005:  93%|█████████▎| 289/312 [29:16<02:19,  6.05s/it]
Training Epoch: 19/100, batch 578/625 completed (loss: 0.0041:  93%|█████████▎| 290/312 [29:19<02:13,  6.07s/it]
Training Epoch: 19/100, batch 578/625 completed (loss: 0.0002:  93%|█████████▎| 290/312 [29:19<02:13,  6.07s/it]
Training Epoch: 19/100, batch 578/625 completed (loss: 0.0005:  93%|█████████▎| 290/312 [29:19<02:13,  6.07s/it]
Training Epoch: 19/100, batch 578/625 completed (loss: 0.0005:  93%|█████████▎| 290/312 [29:19<02:13,  6.07s/it]
Training Epoch: 19/100, batch 579/625 completed (loss: 0.0009:  93%|█████████▎| 290/312 [29:19<02:13,  6.07s/it]
Training Epoch: 19/100, batch 579/625 completed (loss: 0.001:  93%|█████████▎| 290/312 [29:19<02:13,  6.07s/it] 
Training Epoch: 19/100, batch 579/625 completed (loss: 0.0009:  93%|█████████▎| 290/312 [29:19<02:13,  6.07s/it]
Training Epoch: 19/100, batch 579/625 completed (loss: 0.0003:  93%|█████████▎| 290/312 [29:20<02:13,  6.07s/it]
Training Epoch: 19/100, batch 580/625 completed (loss: 0.0002:  93%|█████████▎| 290/312 [29:23<02:13,  6.07s/it]
Training Epoch: 19/100, batch 580/625 completed (loss: 0.0004:  93%|█████████▎| 290/312 [29:23<02:13,  6.07s/it]
Training Epoch: 19/100, batch 580/625 completed (loss: 0.0135:  93%|█████████▎| 290/312 [29:23<02:13,  6.07s/it]
Training Epoch: 19/100, batch 580/625 completed (loss: 0.0004:  93%|█████████▎| 290/312 [29:23<02:13,  6.07s/it]
Training Epoch: 19/100, batch 580/625 completed (loss: 0.0004:  93%|█████████▎| 291/312 [29:26<02:07,  6.10s/it]
Training Epoch: 19/100, batch 580/625 completed (loss: 0.0002:  93%|█████████▎| 291/312 [29:26<02:08,  6.10s/it]
Training Epoch: 19/100, batch 580/625 completed (loss: 0.0135:  93%|█████████▎| 291/312 [29:26<02:07,  6.09s/it]
Training Epoch: 19/100, batch 580/625 completed (loss: 0.0004:  93%|█████████▎| 291/312 [29:26<02:08,  6.10s/it]
Training Epoch: 19/100, batch 581/625 completed (loss: 0.001:  93%|█████████▎| 291/312 [29:26<02:07,  6.09s/it] 
Training Epoch: 19/100, batch 581/625 completed (loss: 0.0002:  93%|█████████▎| 291/312 [29:26<02:08,  6.10s/it]
Training Epoch: 19/100, batch 581/625 completed (loss: 0.0002:  93%|█████████▎| 291/312 [29:26<02:08,  6.10s/it]
Training Epoch: 19/100, batch 581/625 completed (loss: 0.0011:  93%|█████████▎| 291/312 [29:26<02:07,  6.10s/it]
Training Epoch: 19/100, batch 582/625 completed (loss: 0.0004:  93%|█████████▎| 291/312 [29:29<02:08,  6.10s/it]
Training Epoch: 19/100, batch 582/625 completed (loss: 0.0019:  93%|█████████▎| 291/312 [29:29<02:08,  6.10s/it]
Training Epoch: 19/100, batch 582/625 completed (loss: 0.0001:  93%|█████████▎| 291/312 [29:29<02:07,  6.09s/it]
Training Epoch: 19/100, batch 582/625 completed (loss: 0.0096:  93%|█████████▎| 291/312 [29:29<02:07,  6.10s/it]
Training Epoch: 19/100, batch 582/625 completed (loss: 0.0004:  94%|█████████▎| 292/312 [29:31<02:00,  6.02s/it]
Training Epoch: 19/100, batch 582/625 completed (loss: 0.0096:  94%|█████████▎| 292/312 [29:31<02:00,  6.02s/it]
Training Epoch: 19/100, batch 582/625 completed (loss: 0.0019:  94%|█████████▎| 292/312 [29:31<02:00,  6.02s/it]
Training Epoch: 19/100, batch 582/625 completed (loss: 0.0001:  94%|█████████▎| 292/312 [29:31<02:00,  6.02s/it]
Training Epoch: 19/100, batch 583/625 completed (loss: 0.0:  94%|█████████▎| 292/312 [29:31<02:00,  6.02s/it]   
Training Epoch: 19/100, batch 583/625 completed (loss: 0.0001:  94%|█████████▎| 292/312 [29:31<02:00,  6.02s/it]
Training Epoch: 19/100, batch 583/625 completed (loss: 0.0015:  94%|█████████▎| 292/312 [29:31<02:00,  6.02s/it]
Training Epoch: 19/100, batch 583/625 completed (loss: 0.0035:  94%|█████████▎| 292/312 [29:32<02:00,  6.02s/it]
Training Epoch: 19/100, batch 584/625 completed (loss: 0.0003:  94%|█████████▎| 292/312 [29:35<02:00,  6.02s/it]
Training Epoch: 19/100, batch 584/625 completed (loss: 0.0008:  94%|█████████▎| 292/312 [29:35<02:00,  6.02s/it]
Training Epoch: 19/100, batch 584/625 completed (loss: 0.0003:  94%|█████████▎| 292/312 [29:35<02:00,  6.02s/it]
Training Epoch: 19/100, batch 584/625 completed (loss: 0.0001:  94%|█████████▎| 292/312 [29:35<02:00,  6.02s/it]
Training Epoch: 19/100, batch 584/625 completed (loss: 0.0008:  94%|█████████▍| 293/312 [29:38<01:56,  6.11s/it]
Training Epoch: 19/100, batch 584/625 completed (loss: 0.0003:  94%|█████████▍| 293/312 [29:38<01:56,  6.11s/it]
Training Epoch: 19/100, batch 584/625 completed (loss: 0.0003:  94%|█████████▍| 293/312 [29:38<01:56,  6.11s/it]
Training Epoch: 19/100, batch 584/625 completed (loss: 0.0001:  94%|█████████▍| 293/312 [29:38<01:56,  6.11s/it]
Training Epoch: 19/100, batch 585/625 completed (loss: 0.0001:  94%|█████████▍| 293/312 [29:38<01:56,  6.11s/it]
Training Epoch: 19/100, batch 585/625 completed (loss: 0.0009:  94%|█████████▍| 293/312 [29:38<01:56,  6.11s/it]
Training Epoch: 19/100, batch 585/625 completed (loss: 0.0005:  94%|█████████▍| 293/312 [29:38<01:56,  6.11s/it]
Training Epoch: 19/100, batch 585/625 completed (loss: 0.0002:  94%|█████████▍| 293/312 [29:38<01:56,  6.11s/it]
Training Epoch: 19/100, batch 586/625 completed (loss: 0.0001:  94%|█████████▍| 293/312 [29:41<01:56,  6.11s/it]
Training Epoch: 19/100, batch 586/625 completed (loss: 0.0034:  94%|█████████▍| 293/312 [29:41<01:56,  6.11s/it]
Training Epoch: 19/100, batch 586/625 completed (loss: 0.0041:  94%|█████████▍| 293/312 [29:41<01:56,  6.11s/it]
Training Epoch: 19/100, batch 586/625 completed (loss: 0.0002:  94%|█████████▍| 293/312 [29:41<01:56,  6.11s/it]
Training Epoch: 19/100, batch 586/625 completed (loss: 0.0034:  94%|█████████▍| 294/312 [29:44<01:49,  6.08s/it]
Training Epoch: 19/100, batch 586/625 completed (loss: 0.0002:  94%|█████████▍| 294/312 [29:44<01:49,  6.08s/it]
Training Epoch: 19/100, batch 586/625 completed (loss: 0.0001:  94%|█████████▍| 294/312 [29:44<01:49,  6.08s/it]
Training Epoch: 19/100, batch 586/625 completed (loss: 0.0041:  94%|█████████▍| 294/312 [29:44<01:49,  6.08s/it]
Training Epoch: 19/100, batch 587/625 completed (loss: 0.0002:  94%|█████████▍| 294/312 [29:44<01:49,  6.08s/it]
Training Epoch: 19/100, batch 587/625 completed (loss: 0.0001:  94%|█████████▍| 294/312 [29:44<01:49,  6.08s/it]
Training Epoch: 19/100, batch 587/625 completed (loss: 0.0014:  94%|█████████▍| 294/312 [29:44<01:49,  6.08s/it]
Training Epoch: 19/100, batch 587/625 completed (loss: 0.0008:  94%|█████████▍| 294/312 [29:44<01:49,  6.08s/it]
Training Epoch: 19/100, batch 588/625 completed (loss: 0.0004:  94%|█████████▍| 294/312 [29:47<01:49,  6.08s/it]
Training Epoch: 19/100, batch 588/625 completed (loss: 0.0003:  94%|█████████▍| 294/312 [29:47<01:49,  6.08s/it]
Training Epoch: 19/100, batch 588/625 completed (loss: 0.0022:  94%|█████████▍| 294/312 [29:47<01:49,  6.08s/it]
Training Epoch: 19/100, batch 588/625 completed (loss: 0.0003:  94%|█████████▍| 294/312 [29:47<01:49,  6.08s/it]
Training Epoch: 19/100, batch 588/625 completed (loss: 0.0003:  95%|█████████▍| 295/312 [29:50<01:43,  6.11s/it]
Training Epoch: 19/100, batch 588/625 completed (loss: 0.0022:  95%|█████████▍| 295/312 [29:50<01:43,  6.11s/it]
Training Epoch: 19/100, batch 588/625 completed (loss: 0.0003:  95%|█████████▍| 295/312 [29:50<01:43,  6.11s/it]
Training Epoch: 19/100, batch 588/625 completed (loss: 0.0004:  95%|█████████▍| 295/312 [29:50<01:43,  6.11s/it]
Training Epoch: 19/100, batch 589/625 completed (loss: 0.0008:  95%|█████████▍| 295/312 [29:50<01:43,  6.11s/it]
Training Epoch: 19/100, batch 589/625 completed (loss: 0.0007:  95%|█████████▍| 295/312 [29:50<01:43,  6.11s/it]
Training Epoch: 19/100, batch 589/625 completed (loss: 0.0004:  95%|█████████▍| 295/312 [29:50<01:43,  6.11s/it]
Training Epoch: 19/100, batch 589/625 completed (loss: 0.0014:  95%|█████████▍| 295/312 [29:50<01:43,  6.11s/it]
Training Epoch: 19/100, batch 590/625 completed (loss: 0.0011:  95%|█████████▍| 295/312 [29:53<01:43,  6.11s/it]
Training Epoch: 19/100, batch 590/625 completed (loss: 0.0004:  95%|█████████▍| 295/312 [29:53<01:43,  6.11s/it]
Training Epoch: 19/100, batch 590/625 completed (loss: 0.0003:  95%|█████████▍| 295/312 [29:53<01:43,  6.11s/it]
Training Epoch: 19/100, batch 590/625 completed (loss: 0.0007:  95%|█████████▍| 295/312 [29:53<01:43,  6.11s/it]
Training Epoch: 19/100, batch 590/625 completed (loss: 0.0003:  95%|█████████▍| 296/312 [29:56<01:37,  6.12s/it]
Training Epoch: 19/100, batch 590/625 completed (loss: 0.0011:  95%|█████████▍| 296/312 [29:56<01:38,  6.13s/it]
Training Epoch: 19/100, batch 590/625 completed (loss: 0.0007:  95%|█████████▍| 296/312 [29:56<01:37,  6.12s/it]
Training Epoch: 19/100, batch 590/625 completed (loss: 0.0004:  95%|█████████▍| 296/312 [29:56<01:38,  6.13s/it]
Training Epoch: 19/100, batch 591/625 completed (loss: 0.0019:  95%|█████████▍| 296/312 [29:56<01:37,  6.12s/it]
Training Epoch: 19/100, batch 591/625 completed (loss: 0.002:  95%|█████████▍| 296/312 [29:56<01:38,  6.13s/it] 
Training Epoch: 19/100, batch 591/625 completed (loss: 0.0002:  95%|█████████▍| 296/312 [29:56<01:38,  6.13s/it]
Training Epoch: 19/100, batch 591/625 completed (loss: 0.0006:  95%|█████████▍| 296/312 [29:56<01:37,  6.12s/it]
Training Epoch: 19/100, batch 592/625 completed (loss: 0.0013:  95%|█████████▍| 296/312 [29:59<01:38,  6.13s/it]
Training Epoch: 19/100, batch 592/625 completed (loss: 0.0008:  95%|█████████▍| 296/312 [29:59<01:37,  6.12s/it]
Training Epoch: 19/100, batch 592/625 completed (loss: 0.0012:  95%|█████████▍| 296/312 [29:59<01:38,  6.13s/it]
Training Epoch: 19/100, batch 592/625 completed (loss: 0.0008:  95%|█████████▍| 296/312 [29:59<01:37,  6.12s/it]
Training Epoch: 19/100, batch 592/625 completed (loss: 0.0008:  95%|█████████▌| 297/312 [30:02<01:30,  6.04s/it]
Training Epoch: 19/100, batch 592/625 completed (loss: 0.0013:  95%|█████████▌| 297/312 [30:02<01:30,  6.04s/it]
Training Epoch: 19/100, batch 592/625 completed (loss: 0.0012:  95%|█████████▌| 297/312 [30:02<01:30,  6.04s/it]
Training Epoch: 19/100, batch 592/625 completed (loss: 0.0008:  95%|█████████▌| 297/312 [30:02<01:30,  6.05s/it]
Training Epoch: 19/100, batch 593/625 completed (loss: 0.0014:  95%|█████████▌| 297/312 [30:02<01:30,  6.05s/it]
Training Epoch: 19/100, batch 593/625 completed (loss: 0.0018:  95%|█████████▌| 297/312 [30:02<01:30,  6.04s/it]
Training Epoch: 19/100, batch 593/625 completed (loss: 0.0099:  95%|█████████▌| 297/312 [30:02<01:30,  6.04s/it]
Training Epoch: 19/100, batch 593/625 completed (loss: 0.0044:  95%|█████████▌| 297/312 [30:02<01:30,  6.04s/it]
Training Epoch: 19/100, batch 594/625 completed (loss: 0.0008:  95%|█████████▌| 297/312 [30:05<01:30,  6.04s/it]
Training Epoch: 19/100, batch 594/625 completed (loss: 0.0022:  95%|█████████▌| 297/312 [30:05<01:30,  6.04s/it]
Training Epoch: 19/100, batch 594/625 completed (loss: 0.0:  95%|█████████▌| 297/312 [30:05<01:30,  6.05s/it]   
Training Epoch: 19/100, batch 594/625 completed (loss: 0.0005:  95%|█████████▌| 297/312 [30:05<01:30,  6.04s/it]
Training Epoch: 19/100, batch 594/625 completed (loss: 0.0022:  96%|█████████▌| 298/312 [30:08<01:25,  6.13s/it]
Training Epoch: 19/100, batch 594/625 completed (loss: 0.0005:  96%|█████████▌| 298/312 [30:08<01:25,  6.13s/it]
Training Epoch: 19/100, batch 594/625 completed (loss: 0.0:  96%|█████████▌| 298/312 [30:08<01:25,  6.13s/it]
Training Epoch: 19/100, batch 594/625 completed (loss: 0.0008:  96%|█████████▌| 298/312 [30:08<01:25,  6.13s/it]
Training Epoch: 19/100, batch 595/625 completed (loss: 0.0007:  96%|█████████▌| 298/312 [30:08<01:25,  6.13s/it]
Training Epoch: 19/100, batch 595/625 completed (loss: 0.0007:  96%|█████████▌| 298/312 [30:08<01:25,  6.13s/it]
Training Epoch: 19/100, batch 595/625 completed (loss: 0.0061:  96%|█████████▌| 298/312 [30:08<01:25,  6.13s/it]
Training Epoch: 19/100, batch 595/625 completed (loss: 0.0013:  96%|█████████▌| 298/312 [30:08<01:25,  6.13s/it]
Training Epoch: 19/100, batch 596/625 completed (loss: 0.0006:  96%|█████████▌| 298/312 [30:11<01:25,  6.13s/it]
Training Epoch: 19/100, batch 596/625 completed (loss: 0.0437:  96%|█████████▌| 298/312 [30:11<01:25,  6.13s/it]
Training Epoch: 19/100, batch 596/625 completed (loss: 0.0016:  96%|█████████▌| 298/312 [30:11<01:25,  6.13s/it]
Training Epoch: 19/100, batch 596/625 completed (loss: 0.0002:  96%|█████████▌| 298/312 [30:11<01:25,  6.13s/it]
Training Epoch: 19/100, batch 596/625 completed (loss: 0.0437:  96%|█████████▌| 299/312 [30:14<01:19,  6.09s/it]
Training Epoch: 19/100, batch 596/625 completed (loss: 0.0016:  96%|█████████▌| 299/312 [30:14<01:19,  6.09s/it]
Training Epoch: 19/100, batch 596/625 completed (loss: 0.0006:  96%|█████████▌| 299/312 [30:14<01:19,  6.09s/it]
Training Epoch: 19/100, batch 596/625 completed (loss: 0.0002:  96%|█████████▌| 299/312 [30:14<01:19,  6.09s/it]
Training Epoch: 19/100, batch 597/625 completed (loss: 0.0036:  96%|█████████▌| 299/312 [30:14<01:19,  6.09s/it]
Training Epoch: 19/100, batch 597/625 completed (loss: 0.0004:  96%|█████████▌| 299/312 [30:14<01:19,  6.09s/it]
Training Epoch: 19/100, batch 597/625 completed (loss: 0.0004:  96%|█████████▌| 299/312 [30:14<01:19,  6.09s/it]
Training Epoch: 19/100, batch 597/625 completed (loss: 0.0004:  96%|█████████▌| 299/312 [30:14<01:19,  6.09s/it]
Training Epoch: 19/100, batch 598/625 completed (loss: 0.0006:  96%|█████████▌| 299/312 [30:17<01:19,  6.09s/it]
Training Epoch: 19/100, batch 598/625 completed (loss: 0.0006:  96%|█████████▌| 299/312 [30:17<01:19,  6.09s/it]
Training Epoch: 19/100, batch 598/625 completed (loss: 0.0:  96%|█████████▌| 299/312 [30:17<01:19,  6.09s/it]   
Training Epoch: 19/100, batch 598/625 completed (loss: 0.0001:  96%|█████████▌| 299/312 [30:17<01:19,  6.09s/it]
Training Epoch: 19/100, batch 598/625 completed (loss: 0.0001:  96%|█████████▌| 300/312 [30:20<01:12,  6.07s/it]
Training Epoch: 19/100, batch 598/625 completed (loss: 0.0:  96%|█████████▌| 300/312 [30:20<01:12,  6.07s/it]
Training Epoch: 19/100, batch 598/625 completed (loss: 0.0006:  96%|█████████▌| 300/312 [30:20<01:12,  6.08s/it]
Training Epoch: 19/100, batch 598/625 completed (loss: 0.0006:  96%|█████████▌| 300/312 [30:20<01:12,  6.08s/it]
Training Epoch: 19/100, batch 599/625 completed (loss: 0.001:  96%|█████████▌| 300/312 [30:20<01:12,  6.08s/it] 
Training Epoch: 19/100, batch 599/625 completed (loss: 0.0001:  96%|█████████▌| 300/312 [30:20<01:12,  6.08s/it]
Training Epoch: 19/100, batch 599/625 completed (loss: 0.0002:  96%|█████████▌| 300/312 [30:20<01:12,  6.07s/it]
Training Epoch: 19/100, batch 599/625 completed (loss: 0.0128:  96%|█████████▌| 300/312 [30:20<01:12,  6.07s/it]
Training Epoch: 19/100, batch 600/625 completed (loss: 0.0001:  96%|█████████▌| 300/312 [30:24<01:12,  6.07s/it]
Training Epoch: 19/100, batch 600/625 completed (loss: 0.0066:  96%|█████████▌| 300/312 [30:24<01:12,  6.08s/it]
Training Epoch: 19/100, batch 600/625 completed (loss: 0.0001:  96%|█████████▌| 300/312 [30:24<01:12,  6.08s/it]
Training Epoch: 19/100, batch 600/625 completed (loss: 0.0:  96%|█████████▌| 300/312 [30:24<01:12,  6.07s/it]   
Training Epoch: 19/100, batch 600/625 completed (loss: 0.0:  96%|█████████▋| 301/312 [30:27<01:07,  6.10s/it]
Training Epoch: 19/100, batch 600/625 completed (loss: 0.0001:  96%|█████████▋| 301/312 [30:26<01:07,  6.10s/it]
Training Epoch: 19/100, batch 600/625 completed (loss: 0.0001:  96%|█████████▋| 301/312 [30:26<01:07,  6.10s/it]
Training Epoch: 19/100, batch 600/625 completed (loss: 0.0066:  96%|█████████▋| 301/312 [30:26<01:07,  6.11s/it]
Training Epoch: 19/100, batch 601/625 completed (loss: 0.0004:  96%|█████████▋| 301/312 [30:27<01:07,  6.10s/it]
Training Epoch: 19/100, batch 601/625 completed (loss: 0.0:  96%|█████████▋| 301/312 [30:27<01:07,  6.10s/it]   
Training Epoch: 19/100, batch 601/625 completed (loss: 0.0009:  96%|█████████▋| 301/312 [30:27<01:07,  6.11s/it]
Training Epoch: 19/100, batch 601/625 completed (loss: 0.0029:  96%|█████████▋| 301/312 [30:27<01:07,  6.10s/it]
Training Epoch: 19/100, batch 602/625 completed (loss: 0.0007:  96%|█████████▋| 301/312 [30:30<01:07,  6.10s/it]
Training Epoch: 19/100, batch 602/625 completed (loss: 0.0012:  96%|█████████▋| 301/312 [30:30<01:07,  6.10s/it]
Training Epoch: 19/100, batch 602/625 completed (loss: 0.0001:  96%|█████████▋| 301/312 [30:30<01:07,  6.11s/it]
Training Epoch: 19/100, batch 602/625 completed (loss: 0.0023:  96%|█████████▋| 301/312 [30:30<01:07,  6.10s/it]
Training Epoch: 19/100, batch 602/625 completed (loss: 0.0001:  97%|█████████▋| 302/312 [30:32<01:00,  6.08s/it]
Training Epoch: 19/100, batch 602/625 completed (loss: 0.0023:  97%|█████████▋| 302/312 [30:33<01:00,  6.08s/it]
Training Epoch: 19/100, batch 602/625 completed (loss: 0.0007:  97%|█████████▋| 302/312 [30:32<01:00,  6.08s/it]
Training Epoch: 19/100, batch 602/625 completed (loss: 0.0012:  97%|█████████▋| 302/312 [30:33<01:00,  6.08s/it]
Training Epoch: 19/100, batch 603/625 completed (loss: 0.0001:  97%|█████████▋| 302/312 [30:33<01:00,  6.08s/it]
Training Epoch: 19/100, batch 603/625 completed (loss: 0.0039:  97%|█████████▋| 302/312 [30:33<01:00,  6.08s/it]
Training Epoch: 19/100, batch 603/625 completed (loss: 0.0001:  97%|█████████▋| 302/312 [30:33<01:00,  6.08s/it]
Training Epoch: 19/100, batch 603/625 completed (loss: 0.0021:  97%|█████████▋| 302/312 [30:33<01:00,  6.08s/it]
Training Epoch: 19/100, batch 604/625 completed (loss: 0.0014:  97%|█████████▋| 302/312 [30:36<01:00,  6.08s/it]
Training Epoch: 19/100, batch 604/625 completed (loss: 0.0024:  97%|█████████▋| 302/312 [30:36<01:00,  6.08s/it]
Training Epoch: 19/100, batch 604/625 completed (loss: 0.0003:  97%|█████████▋| 302/312 [30:36<01:00,  6.08s/it]
Training Epoch: 19/100, batch 604/625 completed (loss: 0.0032:  97%|█████████▋| 302/312 [30:36<01:00,  6.08s/it]
Training Epoch: 19/100, batch 604/625 completed (loss: 0.0024:  97%|█████████▋| 303/312 [30:39<00:55,  6.12s/it]
Training Epoch: 19/100, batch 604/625 completed (loss: 0.0032:  97%|█████████▋| 303/312 [30:39<00:55,  6.12s/it]
Training Epoch: 19/100, batch 604/625 completed (loss: 0.0003:  97%|█████████▋| 303/312 [30:39<00:55,  6.12s/it]
Training Epoch: 19/100, batch 604/625 completed (loss: 0.0014:  97%|█████████▋| 303/312 [30:39<00:55,  6.12s/it]
Training Epoch: 19/100, batch 605/625 completed (loss: 0.001:  97%|█████████▋| 303/312 [30:39<00:55,  6.12s/it] 
Training Epoch: 19/100, batch 605/625 completed (loss: 0.0023:  97%|█████████▋| 303/312 [30:39<00:55,  6.12s/it]
Training Epoch: 19/100, batch 605/625 completed (loss: 0.0:  97%|█████████▋| 303/312 [30:39<00:55,  6.12s/it]   
Training Epoch: 19/100, batch 605/625 completed (loss: 0.0003:  97%|█████████▋| 303/312 [30:39<00:55,  6.12s/it]
Training Epoch: 19/100, batch 606/625 completed (loss: 0.0002:  97%|█████████▋| 303/312 [30:42<00:55,  6.12s/it]
Training Epoch: 19/100, batch 606/625 completed (loss: 0.0018:  97%|█████████▋| 303/312 [30:42<00:55,  6.12s/it]
Training Epoch: 19/100, batch 606/625 completed (loss: 0.0009:  97%|█████████▋| 303/312 [30:42<00:55,  6.12s/it]
Training Epoch: 19/100, batch 606/625 completed (loss: 0.001:  97%|█████████▋| 303/312 [30:42<00:55,  6.12s/it] 
Training Epoch: 19/100, batch 606/625 completed (loss: 0.001:  97%|█████████▋| 304/312 [30:45<00:49,  6.14s/it]
Training Epoch: 19/100, batch 606/625 completed (loss: 0.0018:  97%|█████████▋| 304/312 [30:45<00:49,  6.14s/it]
Training Epoch: 19/100, batch 606/625 completed (loss: 0.0009:  97%|█████████▋| 304/312 [30:45<00:49,  6.14s/it]
Training Epoch: 19/100, batch 606/625 completed (loss: 0.0002:  97%|█████████▋| 304/312 [30:45<00:49,  6.14s/it]
Training Epoch: 19/100, batch 607/625 completed (loss: 0.0002:  97%|█████████▋| 304/312 [30:45<00:49,  6.14s/it]
Training Epoch: 19/100, batch 607/625 completed (loss: 0.0192:  97%|█████████▋| 304/312 [30:45<00:49,  6.14s/it]
Training Epoch: 19/100, batch 607/625 completed (loss: 0.0013:  97%|█████████▋| 304/312 [30:45<00:49,  6.14s/it]
Training Epoch: 19/100, batch 607/625 completed (loss: 0.0002:  97%|█████████▋| 304/312 [30:45<00:49,  6.14s/it]
Training Epoch: 19/100, batch 608/625 completed (loss: 0.0009:  97%|█████████▋| 304/312 [30:48<00:49,  6.14s/it]
Training Epoch: 19/100, batch 608/625 completed (loss: 0.0006:  97%|█████████▋| 304/312 [30:48<00:49,  6.14s/it]
Training Epoch: 19/100, batch 608/625 completed (loss: 0.0001:  97%|█████████▋| 304/312 [30:48<00:49,  6.14s/it]
Training Epoch: 19/100, batch 608/625 completed (loss: 0.0001:  97%|█████████▋| 304/312 [30:48<00:49,  6.14s/it]
Training Epoch: 19/100, batch 608/625 completed (loss: 0.0001:  98%|█████████▊| 305/312 [30:51<00:42,  6.06s/it]
Training Epoch: 19/100, batch 608/625 completed (loss: 0.0001:  98%|█████████▊| 305/312 [30:51<00:42,  6.06s/it]
Training Epoch: 19/100, batch 608/625 completed (loss: 0.0009:  98%|█████████▊| 305/312 [30:51<00:42,  6.06s/it]
Training Epoch: 19/100, batch 608/625 completed (loss: 0.0006:  98%|█████████▊| 305/312 [30:51<00:42,  6.06s/it]
Training Epoch: 19/100, batch 609/625 completed (loss: 0.004:  98%|█████████▊| 305/312 [30:51<00:42,  6.06s/it] 
Training Epoch: 19/100, batch 609/625 completed (loss: 0.0039:  98%|█████████▊| 305/312 [30:51<00:42,  6.06s/it]
Training Epoch: 19/100, batch 609/625 completed (loss: 0.0008:  98%|█████████▊| 305/312 [30:51<00:42,  6.06s/it]
Training Epoch: 19/100, batch 609/625 completed (loss: 0.0012:  98%|█████████▊| 305/312 [30:51<00:42,  6.06s/it]
Training Epoch: 19/100, batch 610/625 completed (loss: 0.0001:  98%|█████████▊| 305/312 [30:54<00:42,  6.06s/it]
Training Epoch: 19/100, batch 610/625 completed (loss: 0.0002:  98%|█████████▊| 305/312 [30:54<00:42,  6.06s/it]
Training Epoch: 19/100, batch 610/625 completed (loss: 0.0024:  98%|█████████▊| 305/312 [30:54<00:42,  6.06s/it]
Training Epoch: 19/100, batch 610/625 completed (loss: 0.003:  98%|█████████▊| 305/312 [30:54<00:42,  6.06s/it] 
Training Epoch: 19/100, batch 610/625 completed (loss: 0.0001:  98%|█████████▊| 306/312 [30:57<00:37,  6.18s/it]
Training Epoch: 19/100, batch 610/625 completed (loss: 0.003:  98%|█████████▊| 306/312 [30:57<00:37,  6.18s/it]
Training Epoch: 19/100, batch 610/625 completed (loss: 0.0024:  98%|█████████▊| 306/312 [30:57<00:37,  6.18s/it]
Training Epoch: 19/100, batch 610/625 completed (loss: 0.0002:  98%|█████████▊| 306/312 [30:57<00:37,  6.18s/it]
Training Epoch: 19/100, batch 611/625 completed (loss: 0.0031:  98%|█████████▊| 306/312 [30:57<00:37,  6.18s/it]
Training Epoch: 19/100, batch 611/625 completed (loss: 0.0078:  98%|█████████▊| 306/312 [30:57<00:37,  6.18s/it]
Training Epoch: 19/100, batch 611/625 completed (loss: 0.0623:  98%|█████████▊| 306/312 [30:57<00:37,  6.18s/it]
Training Epoch: 19/100, batch 611/625 completed (loss: 0.0003:  98%|█████████▊| 306/312 [30:57<00:37,  6.18s/it]
Training Epoch: 19/100, batch 612/625 completed (loss: 0.0014:  98%|█████████▊| 306/312 [31:00<00:37,  6.18s/it]
Training Epoch: 19/100, batch 612/625 completed (loss: 0.0001:  98%|█████████▊| 306/312 [31:00<00:37,  6.18s/it]
Training Epoch: 19/100, batch 612/625 completed (loss: 0.0003:  98%|█████████▊| 306/312 [31:00<00:37,  6.18s/it]
Training Epoch: 19/100, batch 612/625 completed (loss: 0.0013:  98%|█████████▊| 306/312 [31:00<00:37,  6.18s/it]
Training Epoch: 19/100, batch 612/625 completed (loss: 0.0014:  98%|█████████▊| 307/312 [31:03<00:30,  6.13s/it]
Training Epoch: 19/100, batch 612/625 completed (loss: 0.0013:  98%|█████████▊| 307/312 [31:03<00:30,  6.13s/it]
Training Epoch: 19/100, batch 612/625 completed (loss: 0.0001:  98%|█████████▊| 307/312 [31:03<00:30,  6.13s/it]
Training Epoch: 19/100, batch 612/625 completed (loss: 0.0003:  98%|█████████▊| 307/312 [31:03<00:30,  6.13s/it]
Training Epoch: 19/100, batch 613/625 completed (loss: 0.0011:  98%|█████████▊| 307/312 [31:03<00:30,  6.13s/it]
Training Epoch: 19/100, batch 613/625 completed (loss: 0.0008:  98%|█████████▊| 307/312 [31:03<00:30,  6.13s/it]
Training Epoch: 19/100, batch 613/625 completed (loss: 0.0002:  98%|█████████▊| 307/312 [31:03<00:30,  6.13s/it]
Training Epoch: 19/100, batch 613/625 completed (loss: 0.003:  98%|█████████▊| 307/312 [31:03<00:30,  6.13s/it] 
Training Epoch: 19/100, batch 614/625 completed (loss: 0.0005:  98%|█████████▊| 307/312 [31:06<00:30,  6.13s/it]
Training Epoch: 19/100, batch 614/625 completed (loss: 0.0008:  98%|█████████▊| 307/312 [31:06<00:30,  6.13s/it]
Training Epoch: 19/100, batch 614/625 completed (loss: 0.0035:  98%|█████████▊| 307/312 [31:06<00:30,  6.13s/it]
Training Epoch: 19/100, batch 614/625 completed (loss: 0.0043:  98%|█████████▊| 307/312 [31:06<00:30,  6.13s/it]
Training Epoch: 19/100, batch 614/625 completed (loss: 0.0008:  99%|█████████▊| 308/312 [31:09<00:24,  6.09s/it]
Training Epoch: 19/100, batch 614/625 completed (loss: 0.0035:  99%|█████████▊| 308/312 [31:09<00:24,  6.08s/it]
Training Epoch: 19/100, batch 614/625 completed (loss: 0.0005:  99%|█████████▊| 308/312 [31:09<00:24,  6.09s/it]
Training Epoch: 19/100, batch 614/625 completed (loss: 0.0043:  99%|█████████▊| 308/312 [31:09<00:24,  6.09s/it]
Training Epoch: 19/100, batch 615/625 completed (loss: 0.0029:  99%|█████████▊| 308/312 [31:09<00:24,  6.08s/it]
Training Epoch: 19/100, batch 615/625 completed (loss: 0.0007:  99%|█████████▊| 308/312 [31:09<00:24,  6.09s/it]
Training Epoch: 19/100, batch 615/625 completed (loss: 0.0001:  99%|█████████▊| 308/312 [31:09<00:24,  6.09s/it]
Training Epoch: 19/100, batch 615/625 completed (loss: 0.0022:  99%|█████████▊| 308/312 [31:09<00:24,  6.09s/it]
Training Epoch: 19/100, batch 616/625 completed (loss: 0.0017:  99%|█████████▊| 308/312 [31:12<00:24,  6.09s/it]
Training Epoch: 19/100, batch 616/625 completed (loss: 0.002:  99%|█████████▊| 308/312 [31:12<00:24,  6.08s/it] 
Training Epoch: 19/100, batch 616/625 completed (loss: 0.0007:  99%|█████████▊| 308/312 [31:12<00:24,  6.09s/it]
Training Epoch: 19/100, batch 616/625 completed (loss: 0.0007:  99%|█████████▊| 308/312 [31:12<00:24,  6.09s/it]
Training Epoch: 19/100, batch 616/625 completed (loss: 0.0017:  99%|█████████▉| 309/312 [31:15<00:18,  6.13s/it]
Training Epoch: 19/100, batch 616/625 completed (loss: 0.002:  99%|█████████▉| 309/312 [31:15<00:18,  6.12s/it]
Training Epoch: 19/100, batch 616/625 completed (loss: 0.0007:  99%|█████████▉| 309/312 [31:15<00:18,  6.13s/it]
Training Epoch: 19/100, batch 616/625 completed (loss: 0.0007:  99%|█████████▉| 309/312 [31:15<00:18,  6.13s/it]
Training Epoch: 19/100, batch 617/625 completed (loss: 0.0018:  99%|█████████▉| 309/312 [31:15<00:18,  6.13s/it]
Training Epoch: 19/100, batch 617/625 completed (loss: 0.0007:  99%|█████████▉| 309/312 [31:15<00:18,  6.13s/it]
Training Epoch: 19/100, batch 617/625 completed (loss: 0.0061:  99%|█████████▉| 309/312 [31:15<00:18,  6.12s/it]
Training Epoch: 19/100, batch 617/625 completed (loss: 0.0002:  99%|█████████▉| 309/312 [31:16<00:18,  6.13s/it]
Training Epoch: 19/100, batch 618/625 completed (loss: 0.0002:  99%|█████████▉| 309/312 [31:18<00:18,  6.13s/it]
Training Epoch: 19/100, batch 618/625 completed (loss: 0.0001:  99%|█████████▉| 309/312 [31:18<00:18,  6.13s/it]
Training Epoch: 19/100, batch 618/625 completed (loss: 0.0002:  99%|█████████▉| 309/312 [31:18<00:18,  6.12s/it]
Training Epoch: 19/100, batch 618/625 completed (loss: 0.0003:  99%|█████████▉| 309/312 [31:18<00:18,  6.13s/it]
Training Epoch: 19/100, batch 618/625 completed (loss: 0.0001:  99%|█████████▉| 310/312 [31:21<00:12,  6.04s/it]
Training Epoch: 19/100, batch 618/625 completed (loss: 0.0003:  99%|█████████▉| 310/312 [31:21<00:12,  6.04s/it]
Training Epoch: 19/100, batch 618/625 completed (loss: 0.0002:  99%|█████████▉| 310/312 [31:21<00:12,  6.05s/it]
Training Epoch: 19/100, batch 618/625 completed (loss: 0.0002:  99%|█████████▉| 310/312 [31:21<00:12,  6.05s/it]
Training Epoch: 19/100, batch 619/625 completed (loss: 0.0011:  99%|█████████▉| 310/312 [31:21<00:12,  6.05s/it]
Training Epoch: 19/100, batch 619/625 completed (loss: 0.0:  99%|█████████▉| 310/312 [31:21<00:12,  6.05s/it]   
Training Epoch: 19/100, batch 619/625 completed (loss: 0.0016:  99%|█████████▉| 310/312 [31:21<00:12,  6.04s/it]
Training Epoch: 19/100, batch 619/625 completed (loss: 0.039:  99%|█████████▉| 310/312 [31:21<00:12,  6.04s/it] 
Training Epoch: 19/100, batch 620/625 completed (loss: 0.0001:  99%|█████████▉| 310/312 [31:24<00:12,  6.05s/it]
Training Epoch: 19/100, batch 620/625 completed (loss: 0.0003:  99%|█████████▉| 310/312 [31:24<00:12,  6.04s/it]
Training Epoch: 19/100, batch 620/625 completed (loss: 0.0197:  99%|█████████▉| 310/312 [31:24<00:12,  6.05s/it]
Training Epoch: 19/100, batch 620/625 completed (loss: 0.0008:  99%|█████████▉| 310/312 [31:24<00:12,  6.04s/it]
Training Epoch: 19/100, batch 620/625 completed (loss: 0.0003: 100%|█████████▉| 311/312 [31:27<00:06,  6.08s/it]
Training Epoch: 19/100, batch 620/625 completed (loss: 0.0008: 100%|█████████▉| 311/312 [31:28<00:06,  6.08s/it]
Training Epoch: 19/100, batch 620/625 completed (loss: 0.0001: 100%|█████████▉| 311/312 [31:27<00:06,  6.08s/it]
Training Epoch: 19/100, batch 620/625 completed (loss: 0.0197: 100%|█████████▉| 311/312 [31:27<00:06,  6.08s/it]
Training Epoch: 19/100, batch 621/625 completed (loss: 0.0047: 100%|█████████▉| 311/312 [31:28<00:06,  6.08s/it]
Training Epoch: 19/100, batch 621/625 completed (loss: 0.007: 100%|█████████▉| 311/312 [31:28<00:06,  6.08s/it] 
Training Epoch: 19/100, batch 621/625 completed (loss: 0.001: 100%|█████████▉| 311/312 [31:28<00:06,  6.08s/it] 
Training Epoch: 19/100, batch 621/625 completed (loss: 0.0008: 100%|█████████▉| 311/312 [31:28<00:06,  6.08s/it]
Training Epoch: 19/100, batch 622/625 completed (loss: 0.0006: 100%|█████████▉| 311/312 [31:31<00:06,  6.08s/it]
Training Epoch: 19/100, batch 622/625 completed (loss: 0.0004: 100%|█████████▉| 311/312 [31:31<00:06,  6.08s/it]
Training Epoch: 19/100, batch 622/625 completed (loss: 0.0: 100%|█████████▉| 311/312 [31:31<00:06,  6.08s/it]  
Training Epoch: 19/100, batch 622/625 completed (loss: 0.0001: 100%|█████████▉| 311/312 [31:31<00:06,  6.08s/it]
Training Epoch: 19/100, batch 622/625 completed (loss: 0.0006: 100%|██████████| 312/312 [31:34<00:00,  6.11s/it]
Training Epoch: 19/100, batch 622/625 completed (loss: 0.0: 100%|██████████| 312/312 [31:34<00:00,  6.11s/it]
Training Epoch: 19/100, batch 622/625 completed (loss: 0.0004: 100%|██████████| 312/312 [31:34<00:00,  6.11s/it]
Training Epoch: 19/100, batch 622/625 completed (loss: 0.0001: 100%|██████████| 312/312 [31:34<00:00,  6.11s/it]
Training Epoch: 19/100, batch 623/625 completed (loss: 0.0035: 100%|██████████| 312/312 [31:34<00:00,  6.11s/it]
Training Epoch: 19/100, batch 623/625 completed (loss: 0.0007: 100%|██████████| 312/312 [31:34<00:00,  6.11s/it]
Training Epoch: 19/100, batch 623/625 completed (loss: 0.001: 100%|██████████| 312/312 [31:34<00:00,  6.11s/it] 
Training Epoch: 19/100, batch 623/625 completed (loss: 0.0005: 100%|██████████| 312/312 [31:34<00:00,  6.11s/it]
Training Epoch: 19/100, batch 623/625 completed (loss: 0.001: : 313it [31:37,  5.15s/it]                       
Training Epoch: 19/100, batch 623/625 completed (loss: 0.0007: : 313it [31:37,  5.15s/it]                       
Training Epoch: 19/100, batch 623/625 completed (loss: 0.0005: : 313it [31:37,  5.16s/it]                       
Training Epoch: 19/100, batch 623/625 completed (loss: 0.0035: : 313it [31:37,  5.16s/it]                       
Training Epoch: 19/100, batch 624/625 completed (loss: 0.0002: : 313it [31:37,  5.16s/it]
Training Epoch: 19/100, batch 624/625 completed (loss: 0.0003: : 313it [31:37,  5.15s/it]
Training Epoch: 19/100, batch 624/625 completed (loss: 0.0005: : 313it [31:37,  5.15s/it]
Training Epoch: 19/100, batch 624/625 completed (loss: 0.0597: : 313it [31:37,  5.16s/it]
Training Epoch: 19/100, batch 624/625 completed (loss: 0.0005: : 313it [31:37,  6.06s/it]

Training Epoch: 19/100, batch 624/625 completed (loss: 0.0003: : 313it [31:37,  6.06s/it]

Training Epoch: 19/100, batch 624/625 completed (loss: 0.0002: : 313it [31:37,  6.06s/it]

Training Epoch: 19/100, batch 624/625 completed (loss: 0.0597: : 313it [31:37,  6.06s/it]
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/c10d_logger.py:83: UserWarning: barrier(): using the device under current context. You can specify `device_id` in `init_process_group` to mute this warning.
  return func(*args, **kwargs)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:821: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  prev_state_dict_settings = FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/_optim_utils.py:1172: UserWarning: `_get_pg_default_device` will be deprecated, it only stays for backward-compatibility reason. If you need to find a device for object collectives, please use `_get_object_coll_device`. If you need to query the device types supported by group, please use `_device_capability(group)`. 
  device = _get_pg_default_device(group)
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
/home/ubuntu/thoughtformer/.venv/lib/python3.12/site-packages/torch/distributed/fsdp/fully_sharded_data_parallel.py:828: FutureWarning: FSDP.state_dict_type() and FSDP.set_state_dict_type() are being deprecated. Please use APIs, get_state_dict() and set_state_dict(), which can support different parallelisms, FSDP1, FSDP2, DDP. API doc: https://pytorch.org/docs/stable/distributed.checkpoint.html#torch.distributed.checkpoint.state_dict.get_state_dict .Tutorial: https://pytorch.org/tutorials/recipes/distributed_checkpoint_recipe.html .
  FullyShardedDataParallel.set_state_dict_type(
saving model + train state.

Processing Files (0 / 0)      : |          |  0.00B /  0.00B            

New Data Upload               : |          |  0.00B /  0.00B            


  ...ve-03b89ee9/checkpoint_19:   1%|          | 18.7MB / 3.63GB            


  ...ve-03b89ee9/checkpoint_19:   1%|          | 18.7MB / 3.63GB            
Processing Files (0 / 1)      :   1%|          | 18.7MB / 3.63GB,   ???B/s  

New Data Upload               :  14%|█▍        | 18.7MB /  134MB,   ???B/s  


  ...ve-03b89ee9/checkpoint_19:   3%|▎         | 99.2MB / 3.63GB            
Processing Files (0 / 1)      :   3%|▎         | 99.2MB / 3.63GB, 9.66MB/s  

New Data Upload               :  49%|████▉     | 99.2MB /  201MB, 9.66MB/s  


  ...ve-03b89ee9/checkpoint_19:   4%|▎         |  133MB / 3.63GB            
Processing Files (0 / 1)      :   4%|▎         |  133MB / 3.63GB, 12.7MB/s  

New Data Upload               :  66%|██████▋   |  133MB /  201MB, 12.7MB/s  


  ...ve-03b89ee9/checkpoint_19:   4%|▍         |  137MB / 3.63GB            
Processing Files (0 / 1)      :   4%|▍         |  137MB / 3.63GB, 12.5MB/s  

New Data Upload               :  51%|█████▏    |  137MB /  268MB, 12.5MB/s  


  ...ve-03b89ee9/checkpoint_19:   6%|▌         |  207MB / 3.63GB            
Processing Files (0 / 1)      :   6%|▌         |  207MB / 3.63GB, 18.8MB/s  

New Data Upload               :  62%|██████▏   |  207MB /  335MB, 18.8MB/s  


  ...ve-03b89ee9/checkpoint_19:   9%|▊         |  312MB / 3.63GB            
Processing Files (0 / 1)      :   9%|▊         |  312MB / 3.63GB, 28.3MB/s  

New Data Upload               :  77%|███████▋  |  312MB /  402MB, 28.3MB/s  


  ...ve-03b89ee9/checkpoint_19:  10%|▉         |  348MB / 3.63GB            
Processing Files (0 / 1)      :  10%|▉         |  348MB / 3.63GB, 31.0MB/s  

New Data Upload               :  74%|███████▍  |  348MB /  469MB, 31.0MB/s  


  ...ve-03b89ee9/checkpoint_19:  11%|█         |  401MB / 3.63GB            
Processing Files (0 / 1)      :  11%|█         |  401MB / 3.63GB, 35.2MB/s  

New Data Upload               :  85%|████████▌ |  401MB /  469MB, 35.2MB/s  


  ...ve-03b89ee9/checkpoint_19:  12%|█▏        |  421MB / 3.63GB            
Processing Files (0 / 1)      :  12%|█▏        |  421MB / 3.63GB, 36.3MB/s  

New Data Upload               :  78%|███████▊  |  421MB /  536MB, 36.3MB/s  


  ...ve-03b89ee9/checkpoint_19:  14%|█▍        |  503MB / 3.63GB            
Processing Files (0 / 1)      :  14%|█▍        |  503MB / 3.63GB, 43.0MB/s  

New Data Upload               :  83%|████████▎ |  503MB /  603MB, 43.0MB/s  


  ...ve-03b89ee9/checkpoint_19:  17%|█▋        |  607MB / 3.63GB            
Processing Files (0 / 1)      :  17%|█▋        |  607MB / 3.63GB, 51.5MB/s  

New Data Upload               :  82%|████████▏ |  607MB /  738MB, 51.5MB/s  


  ...ve-03b89ee9/checkpoint_19:  18%|█▊        |  670MB / 3.63GB            
Processing Files (0 / 1)      :  18%|█▊        |  670MB / 3.63GB, 56.1MB/s  

New Data Upload               :  83%|████████▎ |  670MB /  805MB, 56.1MB/s  


  ...ve-03b89ee9/checkpoint_19:  21%|██        |  759MB / 3.63GB            
Processing Files (0 / 1)      :  21%|██        |  759MB / 3.63GB, 62.9MB/s  

New Data Upload               :  81%|████████  |  759MB /  939MB, 62.9MB/s  


  ...ve-03b89ee9/checkpoint_19:  25%|██▍       |  892MB / 3.63GB            
Processing Files (0 / 1)      :  25%|██▍       |  892MB / 3.63GB, 73.5MB/s  

New Data Upload               :  89%|████████▊ |  892MB / 1.01GB, 73.5MB/s  


  ...ve-03b89ee9/checkpoint_19:  27%|██▋       |  987MB / 3.63GB            
Processing Files (0 / 1)      :  27%|██▋       |  987MB / 3.63GB, 80.4MB/s  

New Data Upload               :  92%|█████████▏|  987MB / 1.07GB, 80.4MB/s  


  ...ve-03b89ee9/checkpoint_19:  29%|██▉       | 1.05GB / 3.63GB            
Processing Files (0 / 1)      :  29%|██▉       | 1.05GB / 3.63GB, 84.5MB/s  

New Data Upload               :  87%|████████▋ | 1.05GB / 1.21GB, 84.5MB/s  


  ...ve-03b89ee9/checkpoint_19:  32%|███▏      | 1.16GB / 3.63GB            
Processing Files (0 / 1)      :  32%|███▏      | 1.16GB / 3.63GB, 92.2MB/s  

New Data Upload               :  91%|█████████ | 1.16GB / 1.27GB, 92.2MB/s  


  ...ve-03b89ee9/checkpoint_19:  34%|███▍      | 1.24GB / 3.63GB            
Processing Files (0 / 1)      :  34%|███▍      | 1.24GB / 3.63GB, 97.3MB/s  

New Data Upload               :  93%|█████████▎| 1.24GB / 1.34GB, 97.3MB/s  


  ...ve-03b89ee9/checkpoint_19:  37%|███▋      | 1.34GB / 3.63GB            
Processing Files (0 / 1)      :  37%|███▋      | 1.34GB / 3.63GB,  104MB/s  

New Data Upload               :  91%|█████████ | 1.34GB / 1.48GB,  104MB/s  


  ...ve-03b89ee9/checkpoint_19:  40%|███▉      | 1.44GB / 3.63GB            
Processing Files (0 / 1)      :  40%|███▉      | 1.44GB / 3.63GB,  110MB/s  

New Data Upload               :  93%|█████████▎| 1.44GB / 1.54GB,  110MB/s  


  ...ve-03b89ee9/checkpoint_19:  41%|████▏     | 1.50GB / 3.63GB            
Processing Files (0 / 1)      :  41%|████▏     | 1.50GB / 3.63GB,  113MB/s  

New Data Upload               :  90%|████████▉ | 1.50GB / 1.68GB,  113MB/s  


  ...ve-03b89ee9/checkpoint_19:  45%|████▍     | 1.63GB / 3.63GB            
Processing Files (0 / 1)      :  45%|████▍     | 1.63GB / 3.63GB,  121MB/s  

New Data Upload               :  93%|█████████▎| 1.63GB / 1.74GB,  121MB/s  


  ...ve-03b89ee9/checkpoint_19:  47%|████▋     | 1.72GB / 3.63GB            
Processing Files (0 / 1)      :  47%|████▋     | 1.72GB / 3.63GB,  127MB/s  

New Data Upload               :  95%|█████████▍| 1.72GB / 1.81GB,  127MB/s  


  ...ve-03b89ee9/checkpoint_19:  50%|████▉     | 1.81GB / 3.63GB            
Processing Files (0 / 1)      :  50%|████▉     | 1.81GB / 3.63GB,  132MB/s  

New Data Upload               :  93%|█████████▎| 1.81GB / 1.94GB,  132MB/s  


  ...ve-03b89ee9/checkpoint_19:  52%|█████▏    | 1.89GB / 3.63GB            
Processing Files (0 / 1)      :  52%|█████▏    | 1.89GB / 3.63GB,  136MB/s  

New Data Upload               :  94%|█████████▍| 1.89GB / 2.01GB,  136MB/s  


  ...ve-03b89ee9/checkpoint_19:  55%|█████▍    | 1.99GB / 3.63GB            
Processing Files (0 / 1)      :  55%|█████▍    | 1.99GB / 3.63GB,  142MB/s  

New Data Upload               :  93%|█████████▎| 1.99GB / 2.15GB,  142MB/s  


  ...ve-03b89ee9/checkpoint_19:  57%|█████▋    | 2.08GB / 3.63GB            
Processing Files (0 / 1)      :  57%|█████▋    | 2.08GB / 3.63GB,  146MB/s  

New Data Upload               :  94%|█████████▍| 2.08GB / 2.21GB,  146MB/s  


  ...ve-03b89ee9/checkpoint_19:  60%|██████    | 2.18GB / 3.63GB            
Processing Files (0 / 1)      :  60%|██████    | 2.18GB / 3.63GB,  151MB/s  

New Data Upload               :  95%|█████████▌| 2.18GB / 2.28GB,  151MB/s  


  ...ve-03b89ee9/checkpoint_19:  67%|██████▋   | 2.43GB / 3.63GB            
Processing Files (0 / 1)      :  67%|██████▋   | 2.43GB / 3.63GB,  170MB/s  

New Data Upload               :  98%|█████████▊| 2.29GB / 2.35GB,  158MB/s  


  ...ve-03b89ee9/checkpoint_19:  79%|███████▉  | 2.86GB / 3.63GB            
Processing Files (0 / 1)      :  79%|███████▉  | 2.86GB / 3.63GB,  202MB/s  

New Data Upload               : 100%|█████████▉| 2.35GB / 2.35GB,  159MB/s  


  ...ve-03b89ee9/checkpoint_19:  89%|████████▉ | 3.23GB / 3.63GB            
Processing Files (0 / 1)      :  89%|████████▉ | 3.23GB / 3.63GB,  228MB/s  

New Data Upload               : 100%|█████████▉| 2.35GB / 2.35GB,  156MB/s  


  ...ve-03b89ee9/checkpoint_19:  94%|█████████▎| 3.40GB / 3.63GB            
Processing Files (0 / 1)      :  94%|█████████▎| 3.40GB / 3.63GB,  238MB/s  

New Data Upload               : 100%|█████████▉| 2.35GB / 2.35GB,  153MB/s  


  ...ve-03b89ee9/checkpoint_19:  99%|█████████▉| 3.59GB / 3.63GB            
Processing Files (0 / 1)      :  99%|█████████▉| 3.59GB / 3.63GB,  249MB/s  

New Data Upload               :  99%|█████████▊| 2.35GB / 2.38GB,  150MB/s  


  ...ve-03b89ee9/checkpoint_19: 100%|█████████▉| 3.63GB / 3.63GB            
Processing Files (0 / 1)      : 100%|█████████▉| 3.63GB / 3.63GB,  247MB/s  

New Data Upload               : 100%|█████████▉| 2.38GB / 2.38GB,  149MB/s  


  ...ve-03b89ee9/checkpoint_19: 100%|█████████▉| 3.63GB / 3.63GB            
Processing Files (0 / 1)      : 100%|█████████▉| 3.63GB / 3.63GB,  241MB/s  

New Data Upload               : 100%|█████████▉| 2.38GB / 2.38GB,  146MB/s  


  ...ve-03b89ee9/checkpoint_19: 100%|██████████| 3.63GB / 3.63GB            
Processing Files (1 / 1)      : 100%|██████████| 3.63GB / 3.63GB,  237MB/s  

New Data Upload               : 100%|██████████| 2.38GB / 2.38GB,  143MB/s  
Processing Files (1 / 1)      : 100%|██████████| 3.63GB / 3.63GB,  237MB/s  

New Data Upload               : 100%|██████████| 2.38GB / 2.38GB,  143MB/s  

  ...ve-03b89ee9/checkpoint_19: 100%|██████████| 3.63GB / 3.63GB            
Uploaded checkpoint_19 to seyedparsa/thoughtformer-checkpoints/checkpoints/star-datacurr-only-4gpu-bs32-lr1e-5_n1000_qwen3-0.6b-base_thoughtformer_lr1e-5_no-reset_dc_no-tc_adaptive-03b89ee9/checkpoint_19

Processing Files (0 / 0)      : |          |  0.00B /  0.00B            

New Data Upload               : |          |  0.00B /  0.00B            


  ...e-03b89ee9/train_state_19:   0%|          | 3.68MB / 4.77GB            


  ...e-03b89ee9/train_state_19:   0%|          | 3.68MB / 4.77GB            
Processing Files (0 / 1)      :   0%|          | 3.68MB / 4.77GB,   ???B/s  

New Data Upload               :   3%|▎         | 3.68MB /  134MB,   ???B/s  


  ...e-03b89ee9/train_state_19:   2%|▏         | 92.4MB / 4.77GB            
Processing Files (0 / 1)      :   2%|▏         | 92.4MB / 4.77GB,   ???B/s  

New Data Upload               :  46%|████▌     | 92.4MB /  201MB,   ???B/s  


  ...e-03b89ee9/train_state_19:   3%|▎         |  133MB / 4.77GB            
Processing Files (0 / 1)      :   3%|▎         |  133MB / 4.77GB, 12.9MB/s  

New Data Upload               :  66%|██████▋   |  133MB /  201MB, 12.9MB/s  


  ...e-03b89ee9/train_state_19:   3%|▎         |  145MB / 4.77GB            
Processing Files (0 / 1)      :   3%|▎         |  145MB / 4.77GB, 12.8MB/s  

New Data Upload               :  54%|█████▍    |  145MB /  268MB, 12.8MB/s  


  ...e-03b89ee9/train_state_19:   5%|▌         |  259MB / 4.77GB            
Processing Files (0 / 1)      :   5%|▌         |  259MB / 4.77GB, 23.3MB/s  

New Data Upload               :  64%|██████▍   |  259MB /  402MB, 23.3MB/s  


  ...e-03b89ee9/train_state_19:   7%|▋         |  338MB / 4.77GB            
Processing Files (0 / 1)      :   7%|▋         |  338MB / 4.77GB, 30.1MB/s  

New Data Upload               :  72%|███████▏  |  338MB /  470MB, 30.1MB/s  


  ...e-03b89ee9/train_state_19:   8%|▊         |  395MB / 4.77GB            
Processing Files (0 / 1)      :   8%|▊         |  395MB / 4.77GB, 34.8MB/s  

New Data Upload               :  84%|████████▍ |  395MB /  470MB, 34.8MB/s  


  ...e-03b89ee9/train_state_19:   8%|▊         |  401MB / 4.77GB            
Processing Files (0 / 1)      :   8%|▊         |  401MB / 4.77GB, 34.5MB/s  

New Data Upload               :  85%|████████▌ |  401MB /  470MB, 34.5MB/s  


  ...e-03b89ee9/train_state_19:   8%|▊         |  402MB / 4.77GB            
Processing Files (0 / 1)      :   8%|▊         |  402MB / 4.77GB, 33.7MB/s  

New Data Upload               :  86%|████████▌ |  402MB /  470MB, 33.7MB/s  


  ...e-03b89ee9/train_state_19:  10%|█         |  494MB / 4.77GB            
Processing Files (0 / 1)      :  10%|█         |  494MB / 4.77GB, 41.4MB/s  

New Data Upload               :  82%|████████▏ |  494MB /  604MB, 41.4MB/s  


  ...e-03b89ee9/train_state_19:  12%|█▏        |  581MB / 4.77GB            
Processing Files (0 / 1)      :  12%|█▏        |  581MB / 4.77GB, 48.4MB/s  

New Data Upload               :  79%|███████▉  |  581MB /  738MB, 48.4MB/s  


  ...e-03b89ee9/train_state_19:  14%|█▍        |  689MB / 4.77GB            
Processing Files (0 / 1)      :  14%|█▍        |  689MB / 4.77GB, 57.1MB/s  

New Data Upload               :  86%|████████▌ |  689MB /  805MB, 57.1MB/s  


  ...e-03b89ee9/train_state_19:  16%|█▋        |  775MB / 4.77GB            
Processing Files (0 / 1)      :  16%|█▋        |  775MB / 4.77GB, 63.6MB/s  

New Data Upload               :  89%|████████▉ |  775MB /  872MB, 63.6MB/s  


  ...e-03b89ee9/train_state_19:  18%|█▊        |  854MB / 4.77GB            
Processing Files (0 / 1)      :  18%|█▊        |  854MB / 4.77GB, 69.2MB/s  

New Data Upload               :  85%|████████▍ |  854MB / 1.01GB, 69.2MB/s  


  ...e-03b89ee9/train_state_19:  20%|██        |  954MB / 4.77GB            
Processing Files (0 / 1)      :  20%|██        |  954MB / 4.77GB, 76.6MB/s  

New Data Upload               :  89%|████████▉ |  954MB / 1.07GB, 76.6MB/s  


  ...e-03b89ee9/train_state_19:  22%|██▏       | 1.06GB / 4.77GB            
Processing Files (0 / 1)      :  22%|██▏       | 1.06GB / 4.77GB, 83.9MB/s  

New Data Upload               :  87%|████████▋ | 1.06GB / 1.21GB, 83.9MB/s  


  ...e-03b89ee9/train_state_19:  25%|██▍       | 1.18GB / 4.77GB            
Processing Files (0 / 1)      :  25%|██▍       | 1.18GB / 4.77GB, 92.8MB/s  

New Data Upload               :  92%|█████████▏| 1.18GB / 1.27GB, 92.8MB/s  


  ...e-03b89ee9/train_state_19:  27%|██▋       | 1.28GB / 4.77GB            
Processing Files (0 / 1)      :  27%|██▋       | 1.28GB / 4.77GB, 99.5MB/s  

New Data Upload               :  91%|█████████ | 1.28GB / 1.41GB, 99.5MB/s  


  ...e-03b89ee9/train_state_19:  29%|██▉       | 1.37GB / 4.77GB            
Processing Files (0 / 1)      :  29%|██▉       | 1.37GB / 4.77GB,  106MB/s  

New Data Upload               :  93%|█████████▎| 1.37GB / 1.48GB,  106MB/s  


  ...e-03b89ee9/train_state_19:  31%|███       | 1.47GB / 4.77GB            
Processing Files (0 / 1)      :  31%|███       | 1.47GB / 4.77GB,  112MB/s  

New Data Upload               :  91%|█████████ | 1.47GB / 1.61GB,  112MB/s  


  ...e-03b89ee9/train_state_19:  33%|███▎      | 1.57GB / 4.77GB            
Processing Files (0 / 1)      :  33%|███▎      | 1.57GB / 4.77GB,  118MB/s  

New Data Upload               :  94%|█████████▍| 1.57GB / 1.68GB,  118MB/s  


  ...e-03b89ee9/train_state_19:  35%|███▍      | 1.66GB / 4.77GB            
Processing Files (0 / 1)      :  35%|███▍      | 1.66GB / 4.77GB,  123MB/s  

New Data Upload               :  95%|█████████▌| 1.66GB / 1.74GB,  123MB/s  


  ...e-03b89ee9/train_state_19:  37%|███▋      | 1.75GB / 4.77GB            
Processing Files (0 / 1)      :  37%|███▋      | 1.75GB / 4.77GB,  128MB/s  

New Data Upload               :  93%|█████████▎| 1.75GB / 1.88GB,  128MB/s  


  ...e-03b89ee9/train_state_19:  38%|███▊      | 1.83GB / 4.77GB            
Processing Files (0 / 1)      :  38%|███▊      | 1.83GB / 4.77GB,  133MB/s  

New Data Upload               :  94%|█████████▍| 1.83GB / 1.94GB,  133MB/s  


  ...e-03b89ee9/train_state_19:  40%|████      | 1.92GB / 4.77GB            
Processing Files (0 / 1)      :  40%|████      | 1.92GB / 4.77GB,  137MB/s  

New Data Upload               :  95%|█████████▌| 1.92GB / 2.01GB,  137MB/s  


  ...e-03b89ee9/train_state_19:  42%|████▏     | 2.01GB / 4.77GB            
Processing Files (0 / 1)      :  42%|████▏     | 2.01GB / 4.77GB,  142MB/s  

New Data Upload               :  94%|█████████▍| 2.01GB / 2.15GB,  142MB/s  


  ...e-03b89ee9/train_state_19:  44%|████▍     | 2.12GB / 4.77GB            
Processing Files (0 / 1)      :  44%|████▍     | 2.12GB / 4.77GB,  148MB/s  

New Data Upload               :  96%|█████████▌| 2.12GB / 2.21GB,  148MB/s  


  ...e-03b89ee9/train_state_19:  46%|████▌     | 2.19GB / 4.77GB            
Processing Files (0 / 1)      :  46%|████▌     | 2.19GB / 4.77GB,  151MB/s  

New Data Upload               :  93%|█████████▎| 2.19GB / 2.35GB,  151MB/s  


  ...e-03b89ee9/train_state_19:  48%|████▊     | 2.29GB / 4.77GB            
Processing Files (0 / 1)      :  48%|████▊     | 2.29GB / 4.77GB,  157MB/s  

New Data Upload               :  95%|█████████▍| 2.29GB / 2.41GB,  157MB/s  


  ...e-03b89ee9/train_state_19:  50%|█████     | 2.39GB / 4.77GB            
Processing Files (0 / 1)      :  50%|█████     | 2.39GB / 4.77GB,  161MB/s  

New Data Upload               :  96%|█████████▌| 2.39GB / 2.48GB,  161MB/s  


  ...e-03b89ee9/train_state_19:  52%|█████▏    | 2.50GB / 4.77GB            
Processing Files (0 / 1)      :  52%|█████▏    | 2.50GB / 4.77GB,  167MB/s  

New Data Upload               :  95%|█████████▌| 2.50GB / 2.62GB,  167MB/s  


  ...e-03b89ee9/train_state_19:  54%|█████▍    | 2.57GB / 4.77GB            
Processing Files (0 / 1)      :  54%|█████▍    | 2.57GB / 4.77GB,  169MB/s  

New Data Upload               :  96%|█████████▌| 2.57GB / 2.68GB,  169MB/s  


  ...e-03b89ee9/train_state_19:  56%|█████▌    | 2.66GB / 4.77GB            
Processing Files (0 / 1)      :  56%|█████▌    | 2.66GB / 4.77GB,  174MB/s  

New Data Upload               :  97%|█████████▋| 2.66GB / 2.75GB,  174MB/s  


  ...e-03b89ee9/train_state_19:  58%|█████▊    | 2.75GB / 4.77GB            
Processing Files (0 / 1)      :  58%|█████▊    | 2.75GB / 4.77GB,  177MB/s  

New Data Upload               :  95%|█████████▌| 2.75GB / 2.88GB,  177MB/s  


  ...e-03b89ee9/train_state_19:  59%|█████▉    | 2.83GB / 4.77GB            
Processing Files (0 / 1)      :  59%|█████▉    | 2.83GB / 4.77GB,  180MB/s  

New Data Upload               :  96%|█████████▌| 2.83GB / 2.95GB,  180MB/s  


  ...e-03b89ee9/train_state_19:  61%|██████▏   | 2.93GB / 4.77GB            
Processing Files (0 / 1)      :  61%|██████▏   | 2.93GB / 4.77GB,  185MB/s  

New Data Upload               :  97%|█████████▋| 2.93GB / 3.02GB,  185MB/s  


  ...e-03b89ee9/train_state_19:  63%|██████▎   | 3.00GB / 4.77GB            
Processing Files (0 / 1)      :  63%|██████▎   | 3.00GB / 4.77GB,  187MB/s  

New Data Upload               :  95%|█████████▌| 3.00GB / 3.15GB,  187MB/s  


  ...e-03b89ee9/train_state_19:  65%|██████▌   | 3.11GB / 4.77GB            
Processing Files (0 / 1)      :  65%|██████▌   | 3.11GB / 4.77GB,  192MB/s  

New Data Upload               :  97%|█████████▋| 3.11GB / 3.22GB,  192MB/s  


  ...e-03b89ee9/train_state_19:  67%|██████▋   | 3.22GB / 4.77GB            
Processing Files (0 / 1)      :  67%|██████▋   | 3.22GB / 4.77GB,  196MB/s  

New Data Upload               :  96%|█████████▌| 3.22GB / 3.35GB,  196MB/s  


  ...e-03b89ee9/train_state_19:  70%|██████▉   | 3.31GB / 4.77GB            
Processing Files (0 / 1)      :  70%|██████▉   | 3.31GB / 4.77GB,  200MB/s  

New Data Upload               :  97%|█████████▋| 3.31GB / 3.42GB,  200MB/s  


  ...e-03b89ee9/train_state_19:  71%|███████▏  | 3.41GB / 4.77GB            
Processing Files (0 / 1)      :  71%|███████▏  | 3.41GB / 4.77GB,  204MB/s  

New Data Upload               :  96%|█████████▌| 3.41GB / 3.55GB,  204MB/s  


  ...e-03b89ee9/train_state_19:  74%|███████▎  | 3.51GB / 4.77GB            
Processing Files (0 / 1)      :  74%|███████▎  | 3.51GB / 4.77GB,  208MB/s  

New Data Upload               :  97%|█████████▋| 3.51GB / 3.62GB,  208MB/s  


  ...e-03b89ee9/train_state_19:  75%|███████▌  | 3.58GB / 4.77GB            
Processing Files (0 / 1)      :  75%|███████▌  | 3.58GB / 4.77GB,  210MB/s  

New Data Upload               :  95%|█████████▌| 3.58GB / 3.76GB,  210MB/s  


  ...e-03b89ee9/train_state_19:  78%|███████▊  | 3.71GB / 4.77GB            
Processing Files (0 / 1)      :  78%|███████▊  | 3.71GB / 4.77GB,  215MB/s  

New Data Upload               :  97%|█████████▋| 3.71GB / 3.82GB,  215MB/s  


  ...e-03b89ee9/train_state_19:  79%|███████▉  | 3.79GB / 4.77GB            
Processing Files (0 / 1)      :  79%|███████▉  | 3.79GB / 4.77GB,  218MB/s  

New Data Upload               :  97%|█████████▋| 3.79GB / 3.89GB,  218MB/s  


  ...e-03b89ee9/train_state_19:  81%|████████  | 3.87GB / 4.77GB            
Processing Files (0 / 1)      :  81%|████████  | 3.87GB / 4.77GB,  220MB/s  

New Data Upload               :  96%|█████████▌| 3.87GB / 4.02GB,  220MB/s  


  ...e-03b89ee9/train_state_19:  84%|████████▎ | 3.99GB / 4.77GB            
Processing Files (0 / 1)      :  84%|████████▎ | 3.99GB / 4.77GB,  225MB/s  

New Data Upload               :  98%|█████████▊| 3.99GB / 4.09GB,  225MB/s  


  ...e-03b89ee9/train_state_19:  86%|████████▌ | 4.08GB / 4.77GB            
Processing Files (0 / 1)      :  86%|████████▌ | 4.08GB / 4.77GB,  228MB/s  

New Data Upload               :  97%|█████████▋| 4.08GB / 4.23GB,  228MB/s  


  ...e-03b89ee9/train_state_19:  88%|████████▊ | 4.18GB / 4.77GB            
Processing Files (0 / 1)      :  88%|████████▊ | 4.18GB / 4.77GB,  231MB/s  

New Data Upload               :  97%|█████████▋| 4.18GB / 4.29GB,  231MB/s  


  ...e-03b89ee9/train_state_19:  90%|████████▉ | 4.27GB / 4.77GB            
Processing Files (0 / 1)      :  90%|████████▉ | 4.27GB / 4.77GB,  234MB/s  

New Data Upload               :  96%|█████████▋| 4.27GB / 4.43GB,  234MB/s  


  ...e-03b89ee9/train_state_19:  92%|█████████▏| 4.36GB / 4.77GB            
Processing Files (0 / 1)      :  92%|█████████▏| 4.36GB / 4.77GB,  236MB/s  

New Data Upload               :  97%|█████████▋| 4.36GB / 4.49GB,  236MB/s  


  ...e-03b89ee9/train_state_19:  94%|█████████▍| 4.48GB / 4.77GB            
Processing Files (0 / 1)      :  94%|█████████▍| 4.48GB / 4.77GB,  241MB/s  

New Data Upload               :  97%|█████████▋| 4.48GB / 4.63GB,  241MB/s  


  ...e-03b89ee9/train_state_19:  96%|█████████▌| 4.58GB / 4.77GB            
Processing Files (0 / 1)      :  96%|█████████▌| 4.58GB / 4.77GB,  244MB/s  

New Data Upload               :  98%|█████████▊| 4.58GB / 4.69GB,  244MB/s  


  ...e-03b89ee9/train_state_19:  98%|█████████▊| 4.68GB / 4.77GB            
Processing Files (0 / 1)      :  98%|█████████▊| 4.68GB / 4.77GB,  247MB/s  

New Data Upload               :  98%|█████████▊| 4.68GB / 4.77GB,  247MB/s  


  ...e-03b89ee9/train_state_19: 100%|█████████▉| 4.77GB / 4.77GB            
Processing Files (0 / 1)      : 100%|█████████▉| 4.77GB / 4.77GB,  249MB/s  

New Data Upload               : 100%|█████████▉| 4.77GB / 4.77GB,  249MB/s  


  ...e-03b89ee9/train_state_19: 100%|█████████▉| 4.77GB / 4.77GB            
Processing Files (0 / 1)      : 100%|█████████▉| 4.77GB / 4.77GB,  244MB/s  

New Data Upload               : 100%|█████████▉| 4.77GB / 4.77GB,  244MB/s  


  ...e-03b89ee9/train_state_19: 100%|█████████▉| 4.77GB / 4.77GB            
Processing Files (0 / 1)      : 100%|█████████▉| 4.77GB / 4.77GB,  239MB/s  

New Data Upload               : 100%|█████████▉| 4.77GB / 4.77GB,  239MB/s  


  ...e-03b89ee9/train_state_19: 100%|█████████▉| 4.77GB / 4.77GB            
Processing Files (0 / 1)      : 100%|█████████▉| 4.77GB / 4.77GB,  235MB/s  

New Data Upload               : 100%|█████████▉| 4.77GB / 4.77GB,  235MB/s  


  ...e-03b89ee9/train_state_19: 100%|██████████| 4.77GB / 4.77GB            
Processing Files (1 / 1)      : 100%|██████████| 4.77GB / 4.77GB,  230MB/s  

New Data Upload               : 100%|██████████| 4.77GB / 4.77GB,  230MB/s  
Processing Files (1 / 1)      : 100%|██████████| 4.77GB / 4.77GB,  230MB/s  

New Data Upload               : 100%|██████████| 4.77GB / 4.77GB,  230MB/s  

  ...e-03b89ee9/train_state_19: 100%|██████████| 4.77GB / 4.77GB            
Uploaded train_state_19 to seyedparsa/thoughtformer-checkpoints/checkpoints/star-datacurr-only-4gpu-bs32-lr1e-5_n1000_qwen3-0.6b-base_thoughtformer_lr1e-5_no-reset_dc_no-tc_adaptive-03b89ee9/train_state_19
Deleted old checkpoint: train_state_18
Deleted old checkpoint: checkpoint_18

Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]
Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]
Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]eval loss 0.0006427034358715727

Test Accuracy:   0%|          | 0/32 [00:00<?, ?it/s]
[CORRECT] Example 0 (diff=3)
  Expected : 'Claude'
  Extracted: 'Claude'
  Generated: '### Claude'

[CORRECT] Example 1 (diff=8)
  Expected : 'Guy'
  Extracted: 'Guy'
  Generated: '### Guy'
Test Accuracy:   3%|▎         | 1/32 [00:06<03:22,  6.52s/it]

Test Accuracy:   3%|▎         | 1/32 [00:06<03:22,  6.53s/it]
Test Accuracy:   3%|▎         | 1/32 [00:06<03:22,  6.52s/it]
Test accuracy: 1.0:   3%|▎         | 1/32 [00:06<03:22,  6.52s/it]
Test accuracy: 1.0:   3%|▎         | 1/32 [00:06<03:22,  6.53s/it]
Test accuracy: 1.0:   3%|▎         | 1/32 [00:06<03:22,  6.52s/it]
[CORRECT] Example 2 (diff=10)
  Expected : 'Elvis'
  Extracted: 'Elvis'
  Generated: '### Elvis'

[CORRECT] Example 3 (diff=8)
  Expected : 'Coral'
  Extracted: 'Coral'
  Generated: '### Coral'

[CORRECT] Example 4 (diff=1)
  Expected : 'Patricia'
  Extracted: 'Patricia'
  Generated: '### Patricia'

Test Accuracy:   3%|▎         | 1/32 [00:06<03:18,  6.39s/it]
Test accuracy: 1.0:   3%|▎         | 1/32 [00:06<03:18,  6.39s/it]
Test accuracy: 1.0:   6%|▋         | 2/32 [00:12<03:13,  6.45s/it]
Test accuracy: 1.0:   6%|▋         | 2/32 [00:12<03:13,  6.45s/it]
Test accuracy: 1.0:   6%|▋         | 2/32 [00:12<03:13,  6.45s/it]
Test accuracy: 1.0:   6%|▋         | 2/32 [00:12<03:13,  6.45s/it]
Test accuracy: 1.0:   6%|▋         | 2/32 [00:12<03:11,  6.39s/it]
Test accuracy: 1.0:   6%|▋         | 2/32 [00:12<03:13,  6.45s/it]
Test accuracy: 1.0:   6%|▋         | 2/32 [00:12<03:13,  6.45s/it]
Test accuracy: 1.0:   6%|▋         | 2/32 [00:12<03:11,  6.39s/it]
Test accuracy: 1.0:   9%|▉         | 3/32 [00:19<03:14,  6.71s/it]
Test accuracy: 1.0:   9%|▉         | 3/32 [00:19<03:14,  6.71s/it]
Test accuracy: 1.0:   9%|▉         | 3/32 [00:19<03:14,  6.71s/it]
Test accuracy: 1.0:   9%|▉         | 3/32 [00:19<03:14,  6.71s/it]
Test accuracy: 1.0:   9%|▉         | 3/32 [00:19<03:13,  6.68s/it]
Test accuracy: 1.0:   9%|▉         | 3/32 [00:19<03:13,  6.68s/it]
Test accuracy: 1.0:   9%|▉         | 3/32 [00:19<03:14,  6.71s/it]
Test accuracy: 1.0:   9%|▉         | 3/32 [00:19<03:14,  6.71s/it]
Test accuracy: 1.0:  12%|█▎        | 4/32 [00:26<03:01,  6.48s/it]
Test accuracy: 1.0:  12%|█▎        | 4/32 [00:26<03:01,  6.48s/it]
Test accuracy: 1.0:  12%|█▎        | 4/32 [00:26<03:01,  6.48s/it]
Test accuracy: 1.0:  12%|█▎        | 4/32 [00:26<03:01,  6.48s/it]
Test accuracy: 1.0:  12%|█▎        | 4/32 [00:26<03:01,  6.48s/it]
Test accuracy: 1.0:  12%|█▎        | 4/32 [00:26<03:01,  6.48s/it]
Test accuracy: 1.0:  12%|█▎        | 4/32 [00:25<03:00,  6.46s/it]
Test accuracy: 1.0:  12%|█▎        | 4/32 [00:25<03:00,  6.46s/it]
Test accuracy: 1.0:  16%|█▌        | 5/32 [00:32<02:53,  6.42s/it]
Test accuracy: 1.0:  16%|█▌        | 5/32 [00:32<02:53,  6.42s/it]
Test accuracy: 1.0:  16%|█▌        | 5/32 [00:32<02:53,  6.42s/it]
Test accuracy: 1.0:  16%|█▌        | 5/32 [00:32<02:53,  6.42s/it]
Test accuracy: 1.0:  16%|█▌        | 5/32 [00:32<02:53,  6.42s/it]
Test accuracy: 1.0:  16%|█▌        | 5/32 [00:32<02:53,  6.42s/it]
Test accuracy: 1.0:  16%|█▌        | 5/32 [00:32<02:53,  6.41s/it]
Test accuracy: 1.0:  16%|█▌        | 5/32 [00:32<02:53,  6.41s/it]
Test accuracy: 1.0:  19%|█▉        | 6/32 [00:38<02:45,  6.38s/it]
Test accuracy: 1.0:  19%|█▉        | 6/32 [00:38<02:45,  6.38s/it]
Test accuracy: 1.0:  19%|█▉        | 6/32 [00:38<02:45,  6.38s/it]
Test accuracy: 1.0:  19%|█▉        | 6/32 [00:38<02:45,  6.38s/it]
Test accuracy: 1.0:  19%|█▉        | 6/32 [00:38<02:45,  6.38s/it]
Test accuracy: 1.0:  19%|█▉        | 6/32 [00:38<02:45,  6.38s/it]
Test accuracy: 1.0:  19%|█▉        | 6/32 [00:38<02:45,  6.37s/it]
Test accuracy: 1.0:  19%|█▉        | 6/32 [00:38<02:45,  6.37s/it]
Test accuracy: 1.0:  22%|██▏       | 7/32 [00:45<02:39,  6.36s/it]
Test accuracy: 1.0:  22%|██▏       | 7/32 [00:45<02:39,  6.36s/it]
Test accuracy: 1.0:  22%|██▏       | 7/32 [00:45<02:39,  6.36s/it]
Test accuracy: 1.0:  22%|██▏       | 7/32 [00:44<02:38,  6.36s/it]
Test accuracy: 1.0:  22%|██▏       | 7/32 [00:45<02:39,  6.36s/it]
Test accuracy: 1.0:  22%|██▏       | 7/32 [00:45<02:39,  6.36s/it]
Test accuracy: 1.0:  22%|██▏       | 7/32 [00:45<02:39,  6.36s/it]
Test accuracy: 1.0:  22%|██▏       | 7/32 [00:44<02:38,  6.36s/it]
Test accuracy: 1.0:  25%|██▌       | 8/32 [00:51<02:32,  6.34s/it]
Test accuracy: 1.0:  25%|██▌       | 8/32 [00:51<02:32,  6.34s/it]
Test accuracy: 1.0:  25%|██▌       | 8/32 [00:51<02:32,  6.34s/it]
Test accuracy: 1.0:  25%|██▌       | 8/32 [00:51<02:32,  6.34s/it]
Test accuracy: 1.0:  25%|██▌       | 8/32 [00:51<02:32,  6.34s/it]
Test accuracy: 1.0:  25%|██▌       | 8/32 [00:51<02:32,  6.34s/it]
Test accuracy: 1.0:  25%|██▌       | 8/32 [00:51<02:32,  6.34s/it]
Test accuracy: 1.0:  25%|██▌       | 8/32 [00:51<02:32,  6.34s/it]
Test accuracy: 1.0:  28%|██▊       | 9/32 [00:58<02:28,  6.47s/it]
Test accuracy: 1.0:  28%|██▊       | 9/32 [00:58<02:28,  6.48s/it]
Test accuracy: 1.0:  28%|██▊       | 9/32 [00:58<02:28,  6.47s/it]
Test accuracy: 1.0:  28%|██▊       | 9/32 [00:58<02:28,  6.48s/it]
Test accuracy: 1.0:  28%|██▊       | 9/32 [00:57<02:28,  6.47s/it]
Test accuracy: 1.0:  28%|██▊       | 9/32 [00:57<02:28,  6.47s/it]
Test accuracy: 1.0:  28%|██▊       | 9/32 [00:58<02:28,  6.48s/it]
Test accuracy: 1.0:  28%|██▊       | 9/32 [00:58<02:28,  6.48s/it]
Test accuracy: 1.0:  31%|███▏      | 10/32 [01:04<02:21,  6.44s/it]
Test accuracy: 1.0:  31%|███▏      | 10/32 [01:04<02:21,  6.44s/it]
Test accuracy: 1.0:  31%|███▏      | 10/32 [01:04<02:21,  6.44s/it]
Test accuracy: 1.0:  31%|███▏      | 10/32 [01:04<02:21,  6.44s/it]
Test accuracy: 1.0:  31%|███▏      | 10/32 [01:04<02:21,  6.44s/it]
Test accuracy: 1.0:  31%|███▏      | 10/32 [01:04<02:21,  6.44s/it]
Test accuracy: 1.0:  31%|███▏      | 10/32 [01:04<02:21,  6.43s/it]
Test accuracy: 1.0:  31%|███▏      | 10/32 [01:04<02:21,  6.43s/it]
Test accuracy: 1.0:  34%|███▍      | 11/32 [01:10<02:14,  6.41s/it]
Test accuracy: 1.0:  34%|███▍      | 11/32 [01:10<02:14,  6.41s/it]
Test accuracy: 1.0:  34%|███▍      | 11/32 [01:10<02:14,  6.41s/it]
Test accuracy: 1.0:  34%|███▍      | 11/32 [01:10<02:14,  6.41s/it]
Test accuracy: 1.0:  34%|███▍      | 11/32 [01:10<02:14,  6.40s/it]
Test accuracy: 1.0:  34%|███▍      | 11/32 [01:10<02:14,  6.41s/it]
Test accuracy: 1.0:  34%|███▍      | 11/32 [01:10<02:14,  6.41s/it]
Test accuracy: 1.0:  34%|███▍      | 11/32 [01:10<02:14,  6.40s/it]
Test accuracy: 1.0:  38%|███▊      | 12/32 [01:16<02:06,  6.32s/it]
Test accuracy: 1.0:  38%|███▊      | 12/32 [01:16<02:06,  6.32s/it]
Test accuracy: 1.0:  38%|███▊      | 12/32 [01:16<02:06,  6.32s/it]
Test accuracy: 1.0:  38%|███▊      | 12/32 [01:16<02:06,  6.32s/it]
Test accuracy: 1.0:  38%|███▊      | 12/32 [01:16<02:06,  6.32s/it]
Test accuracy: 1.0:  38%|███▊      | 12/32 [01:16<02:06,  6.32s/it]
Test accuracy: 1.0:  38%|███▊      | 12/32 [01:16<02:06,  6.32s/it]
Test accuracy: 1.0:  38%|███▊      | 12/32 [01:16<02:06,  6.32s/it]
Test accuracy: 1.0:  41%|████      | 13/32 [01:23<02:00,  6.33s/it]
Test accuracy: 1.0:  41%|████      | 13/32 [01:23<02:00,  6.33s/it]
Test accuracy: 1.0:  41%|████      | 13/32 [01:23<02:00,  6.33s/it]
Test accuracy: 1.0:  41%|████      | 13/32 [01:23<02:00,  6.33s/it]
Test accuracy: 1.0:  41%|████      | 13/32 [01:23<02:00,  6.33s/it]
Test accuracy: 1.0:  41%|████      | 13/32 [01:23<02:00,  6.33s/it]
Test accuracy: 1.0:  41%|████      | 13/32 [01:23<02:00,  6.33s/it]
Test accuracy: 1.0:  41%|████      | 13/32 [01:23<02:00,  6.33s/it]
Test accuracy: 1.0:  44%|████▍     | 14/32 [01:29<01:53,  6.32s/it]
Test accuracy: 1.0:  44%|████▍     | 14/32 [01:29<01:53,  6.32s/it]
Test accuracy: 1.0:  44%|████▍     | 14/32 [01:29<01:53,  6.32s/it]
Test accuracy: 1.0:  44%|████▍     | 14/32 [01:29<01:53,  6.32s/it]
Test accuracy: 1.0:  44%|████▍     | 14/32 [01:29<01:53,  6.32s/it]
Test accuracy: 1.0:  44%|████▍     | 14/32 [01:29<01:53,  6.32s/it]
Test accuracy: 1.0:  44%|████▍     | 14/32 [01:29<01:53,  6.32s/it]
Test accuracy: 1.0:  44%|████▍     | 14/32 [01:29<01:53,  6.32s/it]
Test accuracy: 1.0:  47%|████▋     | 15/32 [01:35<01:46,  6.27s/it]
Test accuracy: 1.0:  47%|████▋     | 15/32 [01:35<01:46,  6.27s/it]
Test accuracy: 1.0:  47%|████▋     | 15/32 [01:35<01:46,  6.27s/it]
Test accuracy: 1.0:  47%|████▋     | 15/32 [01:35<01:46,  6.27s/it]
Test accuracy: 1.0:  47%|████▋     | 15/32 [01:35<01:46,  6.27s/it]
Test accuracy: 1.0:  47%|████▋     | 15/32 [01:35<01:46,  6.27s/it]
Test accuracy: 1.0:  47%|████▋     | 15/32 [01:35<01:46,  6.27s/it]
Test accuracy: 1.0:  47%|████▋     | 15/32 [01:35<01:46,  6.27s/it]
Test accuracy: 1.0:  50%|█████     | 16/32 [01:42<01:40,  6.28s/it]
Test accuracy: 1.0:  50%|█████     | 16/32 [01:42<01:40,  6.28s/it]
Test accuracy: 1.0:  50%|█████     | 16/32 [01:42<01:40,  6.28s/it]
Test accuracy: 1.0:  50%|█████     | 16/32 [01:42<01:40,  6.28s/it]
Test accuracy: 1.0:  50%|█████     | 16/32 [01:42<01:40,  6.28s/it]
Test accuracy: 1.0:  50%|█████     | 16/32 [01:41<01:40,  6.28s/it]
Test accuracy: 1.0:  50%|█████     | 16/32 [01:42<01:40,  6.28s/it]
Test accuracy: 1.0:  50%|█████     | 16/32 [01:41<01:40,  6.28s/it]
Test accuracy: 1.0:  53%|█████▎    | 17/32 [01:48<01:34,  6.30s/it]
Test accuracy: 1.0:  53%|█████▎    | 17/32 [01:48<01:34,  6.30s/it]
Test accuracy: 1.0:  53%|█████▎    | 17/32 [01:48<01:34,  6.30s/it]
Test accuracy: 1.0:  53%|█████▎    | 17/32 [01:48<01:34,  6.30s/it]
Test accuracy: 1.0:  53%|█████▎    | 17/32 [01:48<01:34,  6.30s/it]
Test accuracy: 1.0:  53%|█████▎    | 17/32 [01:48<01:34,  6.30s/it]
Test accuracy: 1.0:  53%|█████▎    | 17/32 [01:48<01:34,  6.30s/it]
Test accuracy: 1.0:  53%|█████▎    | 17/32 [01:48<01:34,  6.30s/it]
Test accuracy: 1.0:  56%|█████▋    | 18/32 [01:54<01:28,  6.31s/it]
Test accuracy: 1.0:  56%|█████▋    | 18/32 [01:54<01:28,  6.31s/it]
Test accuracy: 1.0:  56%|█████▋    | 18/32 [01:54<01:28,  6.31s/it]
Test accuracy: 1.0:  56%|█████▋    | 18/32 [01:54<01:28,  6.31s/it]
Test accuracy: 1.0:  56%|█████▋    | 18/32 [01:54<01:28,  6.31s/it]
Test accuracy: 1.0:  56%|█████▋    | 18/32 [01:54<01:28,  6.31s/it]
Test accuracy: 1.0:  56%|█████▋    | 18/32 [01:54<01:28,  6.31s/it]
Test accuracy: 1.0:  56%|█████▋    | 18/32 [01:54<01:28,  6.31s/it]
Test accuracy: 1.0:  59%|█████▉    | 19/32 [02:00<01:21,  6.26s/it]
Test accuracy: 1.0:  59%|█████▉    | 19/32 [02:00<01:21,  6.26s/it]
Test accuracy: 1.0:  59%|█████▉    | 19/32 [02:00<01:21,  6.26s/it]
Test accuracy: 1.0:  59%|█████▉    | 19/32 [02:00<01:21,  6.26s/it]
Test accuracy: 1.0:  59%|█████▉    | 19/32 [02:00<01:21,  6.26s/it]
Test accuracy: 1.0:  59%|█████▉    | 19/32 [02:00<01:21,  6.26s/it]
Test accuracy: 1.0:  59%|█████▉    | 19/32 [02:00<01:21,  6.26s/it]
Test accuracy: 1.0:  59%|█████▉    | 19/32 [02:00<01:21,  6.26s/it]
Test accuracy: 1.0:  62%|██████▎   | 20/32 [02:07<01:15,  6.28s/it]
Test accuracy: 1.0:  62%|██████▎   | 20/32 [02:07<01:15,  6.28s/it]
Test accuracy: 1.0:  62%|██████▎   | 20/32 [02:07<01:15,  6.28s/it]
Test accuracy: 1.0:  62%|██████▎   | 20/32 [02:07<01:15,  6.28s/it]
Test accuracy: 1.0:  62%|██████▎   | 20/32 [02:07<01:15,  6.28s/it]
Test accuracy: 1.0:  62%|██████▎   | 20/32 [02:07<01:15,  6.28s/it]
Test accuracy: 1.0:  62%|██████▎   | 20/32 [02:07<01:15,  6.28s/it]
Test accuracy: 1.0:  62%|██████▎   | 20/32 [02:07<01:15,  6.28s/it]
Test accuracy: 1.0:  66%|██████▌   | 21/32 [02:14<01:11,  6.52s/it]
Test accuracy: 1.0:  66%|██████▌   | 21/32 [02:14<01:11,  6.52s/it]
Test accuracy: 1.0:  66%|██████▌   | 21/32 [02:14<01:11,  6.52s/it]
Test accuracy: 1.0:  66%|██████▌   | 21/32 [02:14<01:11,  6.52s/it]
Test accuracy: 1.0:  66%|██████▌   | 21/32 [02:14<01:11,  6.52s/it]
Test accuracy: 1.0:  66%|██████▌   | 21/32 [02:14<01:11,  6.52s/it]
Test accuracy: 1.0:  66%|██████▌   | 21/32 [02:14<01:11,  6.52s/it]
Test accuracy: 1.0:  66%|██████▌   | 21/32 [02:14<01:11,  6.52s/it]
Test accuracy: 1.0:  69%|██████▉   | 22/32 [02:20<01:04,  6.41s/it]
Test accuracy: 1.0:  69%|██████▉   | 22/32 [02:20<01:04,  6.41s/it]
Test accuracy: 1.0:  69%|██████▉   | 22/32 [02:20<01:04,  6.41s/it]
Test accuracy: 1.0:  69%|██████▉   | 22/32 [02:20<01:04,  6.41s/it]
Test accuracy: 1.0:  69%|██████▉   | 22/32 [02:20<01:04,  6.41s/it]
Test accuracy: 1.0:  69%|██████▉   | 22/32 [02:20<01:04,  6.41s/it]
Test accuracy: 1.0:  69%|██████▉   | 22/32 [02:20<01:04,  6.41s/it]
Test accuracy: 1.0:  69%|██████▉   | 22/32 [02:20<01:04,  6.41s/it]
Test accuracy: 1.0:  72%|███████▏  | 23/32 [02:26<00:57,  6.39s/it]
Test accuracy: 1.0:  72%|███████▏  | 23/32 [02:26<00:57,  6.39s/it]
Test accuracy: 1.0:  72%|███████▏  | 23/32 [02:26<00:57,  6.39s/it]
Test accuracy: 1.0:  72%|███████▏  | 23/32 [02:26<00:57,  6.39s/it]
Test accuracy: 1.0:  72%|███████▏  | 23/32 [02:26<00:57,  6.39s/it]
Test accuracy: 1.0:  72%|███████▏  | 23/32 [02:26<00:57,  6.39s/it]
Test accuracy: 1.0:  72%|███████▏  | 23/32 [02:26<00:57,  6.39s/it]
Test accuracy: 1.0:  72%|███████▏  | 23/32 [02:26<00:57,  6.39s/it]
Test accuracy: 1.0:  75%|███████▌  | 24/32 [02:33<00:52,  6.58s/it]
Test accuracy: 1.0:  75%|███████▌  | 24/32 [02:33<00:52,  6.59s/it]
Test accuracy: 1.0:  75%|███████▌  | 24/32 [02:33<00:52,  6.59s/it]
Test accuracy: 1.0:  75%|███████▌  | 24/32 [02:33<00:52,  6.58s/it]
Test accuracy: 1.0:  75%|███████▌  | 24/32 [02:33<00:52,  6.59s/it]
Test accuracy: 1.0:  75%|███████▌  | 24/32 [02:33<00:52,  6.59s/it]
Test accuracy: 1.0:  75%|███████▌  | 24/32 [02:33<00:52,  6.59s/it]
Test accuracy: 1.0:  75%|███████▌  | 24/32 [02:33<00:52,  6.59s/it]
Test accuracy: 1.0:  78%|███████▊  | 25/32 [02:40<00:45,  6.50s/it]
Test accuracy: 1.0:  78%|███████▊  | 25/32 [02:40<00:45,  6.50s/it]
Test accuracy: 1.0:  78%|███████▊  | 25/32 [02:40<00:45,  6.50s/it]
Test accuracy: 1.0:  78%|███████▊  | 25/32 [02:40<00:45,  6.50s/it]
Test accuracy: 1.0:  78%|███████▊  | 25/32 [02:40<00:45,  6.50s/it]
Test accuracy: 1.0:  78%|███████▊  | 25/32 [02:40<00:45,  6.50s/it]
Test accuracy: 1.0:  78%|███████▊  | 25/32 [02:39<00:45,  6.50s/it]
Test accuracy: 1.0:  78%|███████▊  | 25/32 [02:39<00:45,  6.50s/it]
Test accuracy: 1.0:  81%|████████▏ | 26/32 [02:46<00:38,  6.46s/it]
Test accuracy: 1.0:  81%|████████▏ | 26/32 [02:46<00:38,  6.46s/it]
Test accuracy: 1.0:  81%|████████▏ | 26/32 [02:46<00:38,  6.46s/it]
Test accuracy: 1.0:  81%|████████▏ | 26/32 [02:46<00:38,  6.46s/it]
Test accuracy: 1.0:  81%|████████▏ | 26/32 [02:46<00:38,  6.46s/it]
Test accuracy: 1.0:  81%|████████▏ | 26/32 [02:46<00:38,  6.46s/it]
Test accuracy: 1.0:  81%|████████▏ | 26/32 [02:46<00:38,  6.46s/it]
Test accuracy: 1.0:  81%|████████▏ | 26/32 [02:46<00:38,  6.46s/it]
Test accuracy: 1.0:  84%|████████▍ | 27/32 [02:52<00:31,  6.37s/it]
Test accuracy: 1.0:  84%|████████▍ | 27/32 [02:52<00:31,  6.37s/it]
Test accuracy: 1.0:  84%|████████▍ | 27/32 [02:52<00:31,  6.37s/it]
Test accuracy: 1.0:  84%|████████▍ | 27/32 [02:52<00:31,  6.37s/it]
Test accuracy: 1.0:  84%|████████▍ | 27/32 [02:52<00:31,  6.37s/it]
Test accuracy: 1.0:  84%|████████▍ | 27/32 [02:52<00:31,  6.37s/it]
Test accuracy: 1.0:  84%|████████▍ | 27/32 [02:52<00:31,  6.37s/it]
Test accuracy: 1.0:  84%|████████▍ | 27/32 [02:52<00:31,  6.37s/it]
Test accuracy: 1.0:  88%|████████▊ | 28/32 [02:59<00:26,  6.57s/it]
Test accuracy: 1.0:  88%|████████▊ | 28/32 [02:59<00:26,  6.57s/it]
Test accuracy: 1.0:  88%|████████▊ | 28/32 [02:59<00:26,  6.57s/it]
Test accuracy: 1.0:  88%|████████▊ | 28/32 [02:59<00:26,  6.57s/it]
Test accuracy: 1.0:  88%|████████▊ | 28/32 [02:59<00:26,  6.57s/it]
Test accuracy: 1.0:  88%|████████▊ | 28/32 [02:59<00:26,  6.57s/it]
Test accuracy: 1.0:  88%|████████▊ | 28/32 [02:59<00:26,  6.57s/it]
Test accuracy: 1.0:  88%|████████▊ | 28/32 [02:59<00:26,  6.57s/it]
Test accuracy: 1.0:  91%|█████████ | 29/32 [03:06<00:19,  6.51s/it]
Test accuracy: 1.0:  91%|█████████ | 29/32 [03:06<00:19,  6.51s/it]
Test accuracy: 1.0:  91%|█████████ | 29/32 [03:06<00:19,  6.51s/it]
Test accuracy: 1.0:  91%|█████████ | 29/32 [03:06<00:19,  6.51s/it]
Test accuracy: 1.0:  91%|█████████ | 29/32 [03:06<00:19,  6.51s/it]
Test accuracy: 1.0:  91%|█████████ | 29/32 [03:05<00:19,  6.51s/it]
Test accuracy: 1.0:  91%|█████████ | 29/32 [03:06<00:19,  6.51s/it]
Test accuracy: 1.0:  91%|█████████ | 29/32 [03:05<00:19,  6.51s/it]
Test accuracy: 1.0:  94%|█████████▍| 30/32 [03:12<00:12,  6.40s/it]
Test accuracy: 1.0:  94%|█████████▍| 30/32 [03:12<00:12,  6.40s/it]
Test accuracy: 1.0:  94%|█████████▍| 30/32 [03:12<00:12,  6.40s/it]
Test accuracy: 1.0:  94%|█████████▍| 30/32 [03:12<00:12,  6.40s/it]
Test accuracy: 1.0:  94%|█████████▍| 30/32 [03:12<00:12,  6.40s/it]
Test accuracy: 1.0:  94%|█████████▍| 30/32 [03:12<00:12,  6.40s/it]
Test accuracy: 1.0:  94%|█████████▍| 30/32 [03:12<00:12,  6.40s/it]
Test accuracy: 1.0:  94%|█████████▍| 30/32 [03:12<00:12,  6.40s/it]
Test accuracy: 1.0:  97%|█████████▋| 31/32 [03:19<00:06,  6.59s/it]
Test accuracy: 1.0:  97%|█████████▋| 31/32 [03:19<00:06,  6.59s/it]
Test accuracy: 1.0:  97%|█████████▋| 31/32 [03:19<00:06,  6.59s/it]
Test accuracy: 1.0:  97%|█████████▋| 31/32 [03:19<00:06,  6.59s/it]
Test accuracy: 1.0:  97%|█████████▋| 31/32 [03:19<00:06,  6.59s/it]
Test accuracy: 1.0:  97%|█████████▋| 31/32 [03:19<00:06,  6.59s/it]
Test accuracy: 1.0:  97%|█████████▋| 31/32 [03:19<00:06,  6.60s/it]
Test accuracy: 1.0:  97%|█████████▋| 31/32 [03:19<00:06,  6.60s/it]
Test accuracy: 1.0: 100%|██████████| 32/32 [03:21<00:00,  5.29s/it]
Test accuracy: 1.0: 100%|██████████| 32/32 [03:21<00:00,  5.29s/it]
Test accuracy: 1.0: 100%|██████████| 32/32 [03:21<00:00,  5.29s/it]
Test accuracy: 1.0: 100%|██████████| 32/32 [03:21<00:00,  5.29s/it]
Test accuracy: 1.0: 100%|██████████| 32/32 [03:21<00:00,  5.29s/it]
Test accuracy: 1.0: 100%|██████████| 32/32 [03:21<00:00,  5.29s/it]
Test accuracy: 1.0: 100%|██████████| 32/32 [03:21<00:00,  5.29s/it]
Test accuracy: 1.0: 100%|██████████| 32/32 [03:21<00:00,  5.29s/it]
Test accuracy: 1.0: 100%|██████████| 32/32 [03:21<00:00,  6.30s/it]
Device 2: Cor=250, CoT=0, Total=250

Test accuracy: 1.0: 100%|██████████| 32/32 [03:21<00:00,  6.30s/it]
Device 3: Cor=250, CoT=0, Total=250

Test accuracy: 1.0: 100%|██████████| 32/32 [03:21<00:00,  6.30s/it]
Device 1: Cor=250, CoT=0, Total=250

Test accuracy: 1.0: 100%|██████████| 32/32 [03:21<00:00,  6.30s/it]
Device 0: Cor=250, CoT=0, Total=250

============================================================
EVALUATION RESULTS (Epoch 18)
============================================================
Config: data_curriculum=ON, thought_curriculum=OFF
thought_stage=10, data_stage=10, n_latent_tokens=10
Training on difficulties: 0-10

Overall accuracy: 1000 / 1000 = 1.0000

Per-difficulty accuracy:
  diff=1: 100/100 = 1.0000 (trained)
  diff=2: 100/100 = 1.0000 (trained)
  diff=3: 100/100 = 1.0000 (trained)
  diff=4: 100/100 = 1.0000 (trained)
  diff=5: 100/100 = 1.0000 (trained)
  diff=6: 100/100 = 1.0000 (trained)
  diff=7: 100/100 = 1.0000 (trained)
  diff=8: 100/100 = 1.0000 (trained)
  diff=9: 100/100 = 1.0000 (trained)
  diff=10: 100/100 = 1.0000 (trained)

Trained (diff 0-10): 1000/1000 = 1.0000
============================================================


*** EARLY STOP: 100% accuracy at epoch 19, all stages reached ***

wandb: 
wandb: 🚀 View run star-datacurr-only-4gpu-bs32-lr1e-5_n1000_qwen3-0.6b-base_thoughtformer_lr1e-5_no-reset_dc_no-tc_adaptive-03b89ee9 at: https://wandb.ai/seyedparsa/thoughtformer/runs/efv4go42
wandb: Find logs at: wandb/run-20260530_221859-efv4go42/logs
[rank0]:[W531 04:37:12.821203020 ProcessGroupNCCL.cpp:1647] Warning: WARNING: destroy_process_group() was not called before program exit, which can leak resources. For more info, please see https://pytorch.org/docs/stable/distributed.html#shutdown (function operator())