stderr.log

+ deepspeed --master_port 16182 --module safe_rlhf.finetune --train_datasets inverse-json::/home/hansirui_1st/jiayi/resist/setting3/safety_data/training/safe/safe_30k.json --model_name_or_path /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-480k-1T --max_length 2048 --trust_remote_code True --epochs 1 --per_device_train_batch_size 4 --per_device_eval_batch_size 4 --gradient_accumulation_steps 8 --gradient_checkpointing --learning_rate 1e-5 --lr_warmup_ratio 0 --weight_decay 0.0 --lr_scheduler_type constant --weight_decay 0.0 --seed 42 --output_dir /aifs4su/hansirui_1st/boyuan/resist/setting3-safety/tinyllama-1T/tinyllama-1T-s3-Q1-30k --log_type wandb --log_run_name tinyllama-1T-s3-Q1-30k --log_project Inverse_Alignment --zero_stage 3 --offload none --bf16 True --tf32 True --save_16bit
[rank4]:[W529 04:24:27.827578493 ProcessGroupNCCL.cpp:4561] [PG ID 0 PG GUID 0 Rank 4]  using GPU 4 to perform barrier as devices used by this process are currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect. Specify device_ids in barrier() to force use of a particular device, or call init_process_group() with a device_id.
[rank2]:[W529 04:24:27.827601722 ProcessGroupNCCL.cpp:4561] [PG ID 0 PG GUID 0 Rank 2]  using GPU 2 to perform barrier as devices used by this process are currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect. Specify device_ids in barrier() to force use of a particular device, or call init_process_group() with a device_id.
[rank7]:[W529 04:24:27.899692720 ProcessGroupNCCL.cpp:4561] [PG ID 0 PG GUID 0 Rank 7]  using GPU 7 to perform barrier as devices used by this process are currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect. Specify device_ids in barrier() to force use of a particular device, or call init_process_group() with a device_id.
[rank5]:[W529 04:24:27.938652087 ProcessGroupNCCL.cpp:4561] [PG ID 0 PG GUID 0 Rank 5]  using GPU 5 to perform barrier as devices used by this process are currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect. Specify device_ids in barrier() to force use of a particular device, or call init_process_group() with a device_id.
[rank6]:[W529 04:24:27.983457860 ProcessGroupNCCL.cpp:4561] [PG ID 0 PG GUID 0 Rank 6]  using GPU 6 to perform barrier as devices used by this process are currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect. Specify device_ids in barrier() to force use of a particular device, or call init_process_group() with a device_id.
[rank1]:[W529 04:24:27.991657559 ProcessGroupNCCL.cpp:4561] [PG ID 0 PG GUID 0 Rank 1]  using GPU 1 to perform barrier as devices used by this process are currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect. Specify device_ids in barrier() to force use of a particular device, or call init_process_group() with a device_id.
[rank0]:[W529 04:24:27.079396730 ProcessGroupNCCL.cpp:4561] [PG ID 0 PG GUID 0 Rank 0]  using GPU 0 to perform barrier as devices used by this process are currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect. Specify device_ids in barrier() to force use of a particular device, or call init_process_group() with a device_id.
[rank3]:[W529 04:24:27.105063926 ProcessGroupNCCL.cpp:4561] [PG ID 0 PG GUID 0 Rank 3]  using GPU 3 to perform barrier as devices used by this process are currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect. Specify device_ids in barrier() to force use of a particular device, or call init_process_group() with a device_id.
loading configuration file /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-480k-1T/config.json
loading configuration file /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-480k-1T/config.json
loading configuration file /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-480k-1T/config.json
loading configuration file /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-480k-1T/config.json
loading configuration file /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-480k-1T/config.json
loading configuration file /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-480k-1T/config.json
loading configuration file /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-480k-1T/config.json
loading configuration file /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-480k-1T/config.json
Model config LlamaConfig {
  "_name_or_path": "/aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-480k-1T",
  "architectures": [
    "LlamaForCausalLM"
  ],
  "attention_bias": false,
  "attention_dropout": 0.0,
  "bos_token_id": 1,
  "eos_token_id": 2,
  "head_dim": 64,
  "hidden_act": "silu",
  "hidden_size": 2048,
  "initializer_range": 0.02,
  "intermediate_size": 5632,
  "max_position_embeddings": 2048,
  "mlp_bias": false,
  "model_type": "llama",
  "num_attention_heads": 32,
  "num_hidden_layers": 22,
  "num_key_value_heads": 4,
  "pretraining_tp": 1,
  "rms_norm_eps": 1e-05,
  "rope_scaling": null,
  "rope_theta": 10000.0,
  "tie_word_embeddings": false,
  "torch_dtype": "float32",
  "transformers_version": "4.49.0",
  "use_cache": true,
  "vocab_size": 32000
}

Model config LlamaConfig {
  "_name_or_path": "/aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-480k-1T",
  "architectures": [
    "LlamaForCausalLM"
  ],
  "attention_bias": false,
  "attention_dropout": 0.0,
  "bos_token_id": 1,
  "eos_token_id": 2,
  "head_dim": 64,
  "hidden_act": "silu",
  "hidden_size": 2048,
  "initializer_range": 0.02,
  "intermediate_size": 5632,
  "max_position_embeddings": 2048,
  "mlp_bias": false,
  "model_type": "llama",
  "num_attention_heads": 32,
  "num_hidden_layers": 22,
  "num_key_value_heads": 4,
  "pretraining_tp": 1,
  "rms_norm_eps": 1e-05,
  "rope_scaling": null,
  "rope_theta": 10000.0,
  "tie_word_embeddings": false,
  "torch_dtype": "float32",
  "transformers_version": "4.49.0",
  "use_cache": true,
  "vocab_size": 32000
}

Model config LlamaConfig {
  "_name_or_path": "/aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-480k-1T",
  "architectures": [
    "LlamaForCausalLM"
  ],
  "attention_bias": false,
  "attention_dropout": 0.0,
  "bos_token_id": 1,
  "eos_token_id": 2,
  "head_dim": 64,
  "hidden_act": "silu",
  "hidden_size": 2048,
  "initializer_range": 0.02,
  "intermediate_size": 5632,
  "max_position_embeddings": 2048,
  "mlp_bias": false,
  "model_type": "llama",
  "num_attention_heads": 32,
  "num_hidden_layers": 22,
  "num_key_value_heads": 4,
  "pretraining_tp": 1,
  "rms_norm_eps": 1e-05,
  "rope_scaling": null,
  "rope_theta": 10000.0,
  "tie_word_embeddings": false,
  "torch_dtype": "float32",
  "transformers_version": "4.49.0",
  "use_cache": true,
  "vocab_size": 32000
}

Model config LlamaConfig {
  "_name_or_path": "/aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-480k-1T",
  "architectures": [
    "LlamaForCausalLM"
  ],
  "attention_bias": false,
  "attention_dropout": 0.0,
  "bos_token_id": 1,
  "eos_token_id": 2,
  "head_dim": 64,
  "hidden_act": "silu",
  "hidden_size": 2048,
  "initializer_range": 0.02,
  "intermediate_size": 5632,
  "max_position_embeddings": 2048,
  "mlp_bias": false,
  "model_type": "llama",
  "num_attention_heads": 32,
  "num_hidden_layers": 22,
  "num_key_value_heads": 4,
  "pretraining_tp": 1,
  "rms_norm_eps": 1e-05,
  "rope_scaling": null,
  "rope_theta": 10000.0,
  "tie_word_embeddings": false,
  "torch_dtype": "float32",
  "transformers_version": "4.49.0",
  "use_cache": true,
  "vocab_size": 32000
}

Model config LlamaConfig {
  "_name_or_path": "/aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-480k-1T",
  "architectures": [
    "LlamaForCausalLM"
  ],
  "attention_bias": false,
  "attention_dropout": 0.0,
  "bos_token_id": 1,
  "eos_token_id": 2,
  "head_dim": 64,
  "hidden_act": "silu",
  "hidden_size": 2048,
  "initializer_range": 0.02,
  "intermediate_size": 5632,
  "max_position_embeddings": 2048,
  "mlp_bias": false,
  "model_type": "llama",
  "num_attention_heads": 32,
  "num_hidden_layers": 22,
  "num_key_value_heads": 4,
  "pretraining_tp": 1,
  "rms_norm_eps": 1e-05,
  "rope_scaling": null,
  "rope_theta": 10000.0,
  "tie_word_embeddings": false,
  "torch_dtype": "float32",
  "transformers_version": "4.49.0",
  "use_cache": true,
  "vocab_size": 32000
}

Model config LlamaConfig {
  "_name_or_path": "/aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-480k-1T",
  "architectures": [
    "LlamaForCausalLM"
  ],
  "attention_bias": false,
  "attention_dropout": 0.0,
  "bos_token_id": 1,
  "eos_token_id": 2,
  "head_dim": 64,
  "hidden_act": "silu",
  "hidden_size": 2048,
  "initializer_range": 0.02,
  "intermediate_size": 5632,
  "max_position_embeddings": 2048,
  "mlp_bias": false,
  "model_type": "llama",
  "num_attention_heads": 32,
  "num_hidden_layers": 22,
  "num_key_value_heads": 4,
  "pretraining_tp": 1,
  "rms_norm_eps": 1e-05,
  "rope_scaling": null,
  "rope_theta": 10000.0,
  "tie_word_embeddings": false,
  "torch_dtype": "float32",
  "transformers_version": "4.49.0",
  "use_cache": true,
  "vocab_size": 32000
}

Model config LlamaConfig {
  "_name_or_path": "/aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-480k-1T",
  "architectures": [
    "LlamaForCausalLM"
  ],
  "attention_bias": false,
  "attention_dropout": 0.0,
  "bos_token_id": 1,
  "eos_token_id": 2,
  "head_dim": 64,
  "hidden_act": "silu",
  "hidden_size": 2048,
  "initializer_range": 0.02,
  "intermediate_size": 5632,
  "max_position_embeddings": 2048,
  "mlp_bias": false,
  "model_type": "llama",
  "num_attention_heads": 32,
  "num_hidden_layers": 22,
  "num_key_value_heads": 4,
  "pretraining_tp": 1,
  "rms_norm_eps": 1e-05,
  "rope_scaling": null,
  "rope_theta": 10000.0,
  "tie_word_embeddings": false,
  "torch_dtype": "float32",
  "transformers_version": "4.49.0",
  "use_cache": true,
  "vocab_size": 32000
}

Model config LlamaConfig {
  "_name_or_path": "/aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-480k-1T",
  "architectures": [
    "LlamaForCausalLM"
  ],
  "attention_bias": false,
  "attention_dropout": 0.0,
  "bos_token_id": 1,
  "eos_token_id": 2,
  "head_dim": 64,
  "hidden_act": "silu",
  "hidden_size": 2048,
  "initializer_range": 0.02,
  "intermediate_size": 5632,
  "max_position_embeddings": 2048,
  "mlp_bias": false,
  "model_type": "llama",
  "num_attention_heads": 32,
  "num_hidden_layers": 22,
  "num_key_value_heads": 4,
  "pretraining_tp": 1,
  "rms_norm_eps": 1e-05,
  "rope_scaling": null,
  "rope_theta": 10000.0,
  "tie_word_embeddings": false,
  "torch_dtype": "float32",
  "transformers_version": "4.49.0",
  "use_cache": true,
  "vocab_size": 32000
}

loading weights file /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-480k-1T/model.safetensors
loading weights file /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-480k-1T/model.safetensors
loading weights file /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-480k-1T/model.safetensors
loading weights file /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-480k-1T/model.safetensors
loading weights file /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-480k-1T/model.safetensors
loading weights file /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-480k-1T/model.safetensors
loading weights file /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-480k-1T/model.safetensors
Will use torch_dtype=torch.float32 as defined in model's config object
Will use torch_dtype=torch.float32 as defined in model's config object
Instantiating LlamaForCausalLM model under default dtype torch.float32.
Instantiating LlamaForCausalLM model under default dtype torch.float32.
Will use torch_dtype=torch.float32 as defined in model's config object
Will use torch_dtype=torch.float32 as defined in model's config object
Detected DeepSpeed ZeRO-3: activating zero.init() for this model
Detected DeepSpeed ZeRO-3: activating zero.init() for this model
Instantiating LlamaForCausalLM model under default dtype torch.float32.
Instantiating LlamaForCausalLM model under default dtype torch.float32.
Detected DeepSpeed ZeRO-3: activating zero.init() for this model
Detected DeepSpeed ZeRO-3: activating zero.init() for this model
Will use torch_dtype=torch.float32 as defined in model's config object
Instantiating LlamaForCausalLM model under default dtype torch.float32.
Detected DeepSpeed ZeRO-3: activating zero.init() for this model
Will use torch_dtype=torch.float32 as defined in model's config object
Instantiating LlamaForCausalLM model under default dtype torch.float32.
Detected DeepSpeed ZeRO-3: activating zero.init() for this model
Will use torch_dtype=torch.float32 as defined in model's config object
Instantiating LlamaForCausalLM model under default dtype torch.float32.
Detected DeepSpeed ZeRO-3: activating zero.init() for this model
loading weights file /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-480k-1T/model.safetensors
Generate config GenerationConfig {
  "bos_token_id": 1,
  "eos_token_id": 2
}

Generate config GenerationConfig {
  "bos_token_id": 1,
  "eos_token_id": 2
}

Generate config GenerationConfig {
  "bos_token_id": 1,
  "eos_token_id": 2
}

Generate config GenerationConfig {
  "bos_token_id": 1,
  "eos_token_id": 2
}

Generate config GenerationConfig {
  "bos_token_id": 1,
  "eos_token_id": 2
}

Generate config GenerationConfig {
  "bos_token_id": 1,
  "eos_token_id": 2
}

Generate config GenerationConfig {
  "bos_token_id": 1,
  "eos_token_id": 2
}

Will use torch_dtype=torch.float32 as defined in model's config object
Instantiating LlamaForCausalLM model under default dtype torch.float32.
Detected DeepSpeed ZeRO-3: activating zero.init() for this model
Generate config GenerationConfig {
  "bos_token_id": 1,
  "eos_token_id": 2
}

All model checkpoint weights were used when initializing LlamaForCausalLM.

All the weights of LlamaForCausalLM were initialized from the model checkpoint at /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-480k-1T.
If your task is similar to the task the model of the checkpoint was trained on, you can already use LlamaForCausalLM for predictions without further training.
All model checkpoint weights were used when initializing LlamaForCausalLM.

All the weights of LlamaForCausalLM were initialized from the model checkpoint at /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-480k-1T.
If your task is similar to the task the model of the checkpoint was trained on, you can already use LlamaForCausalLM for predictions without further training.
loading configuration file /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-480k-1T/generation_config.json
Generate config GenerationConfig {
  "bos_token_id": 1,
  "eos_token_id": 2,
  "max_length": 2048,
  "pad_token_id": 0
}

All model checkpoint weights were used when initializing LlamaForCausalLM.

All the weights of LlamaForCausalLM were initialized from the model checkpoint at /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-480k-1T.
If your task is similar to the task the model of the checkpoint was trained on, you can already use LlamaForCausalLM for predictions without further training.
All model checkpoint weights were used when initializing LlamaForCausalLM.

All the weights of LlamaForCausalLM were initialized from the model checkpoint at /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-480k-1T.
If your task is similar to the task the model of the checkpoint was trained on, you can already use LlamaForCausalLM for predictions without further training.
All model checkpoint weights were used when initializing LlamaForCausalLM.

All the weights of LlamaForCausalLM were initialized from the model checkpoint at /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-480k-1T.
If your task is similar to the task the model of the checkpoint was trained on, you can already use LlamaForCausalLM for predictions without further training.
loading configuration file /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-480k-1T/generation_config.json
Generate config GenerationConfig {
  "bos_token_id": 1,
  "eos_token_id": 2,
  "max_length": 2048,
  "pad_token_id": 0
}

All model checkpoint weights were used when initializing LlamaForCausalLM.

All the weights of LlamaForCausalLM were initialized from the model checkpoint at /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-480k-1T.
If your task is similar to the task the model of the checkpoint was trained on, you can already use LlamaForCausalLM for predictions without further training.
loading configuration file /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-480k-1T/generation_config.json
loading configuration file /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-480k-1T/generation_config.json
Generate config GenerationConfig {
  "bos_token_id": 1,
  "eos_token_id": 2,
  "max_length": 2048,
  "pad_token_id": 0
}

All model checkpoint weights were used when initializing LlamaForCausalLM.

loading configuration file /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-480k-1T/generation_config.json
All the weights of LlamaForCausalLM were initialized from the model checkpoint at /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-480k-1T.
If your task is similar to the task the model of the checkpoint was trained on, you can already use LlamaForCausalLM for predictions without further training.
Generate config GenerationConfig {
  "bos_token_id": 1,
  "eos_token_id": 2,
  "max_length": 2048,
  "pad_token_id": 0
}

Generate config GenerationConfig {
  "bos_token_id": 1,
  "eos_token_id": 2,
  "max_length": 2048,
  "pad_token_id": 0
}

loading configuration file /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-480k-1T/generation_config.json
Generate config GenerationConfig {
  "bos_token_id": 1,
  "eos_token_id": 2,
  "max_length": 2048,
  "pad_token_id": 0
}

loading configuration file /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-480k-1T/generation_config.json
Generate config GenerationConfig {
  "bos_token_id": 1,
  "eos_token_id": 2,
  "max_length": 2048,
  "pad_token_id": 0
}

loading file tokenizer.model
loading file tokenizer.json
loading file added_tokens.json
loading file special_tokens_map.json
loading file tokenizer_config.json
loading file chat_template.jinja
loading file tokenizer.model
loading file tokenizer.json
loading file added_tokens.json
loading file special_tokens_map.json
loading file tokenizer_config.json
loading file chat_template.jinja
loading file tokenizer.model
loading file tokenizer.json
loading file added_tokens.json
loading file special_tokens_map.json
loading file tokenizer_config.json
loading file chat_template.jinja
loading file tokenizer.model
loading file tokenizer.json
loading file added_tokens.json
loading file special_tokens_map.json
loading file tokenizer_config.json
loading file chat_template.jinja
loading file tokenizer.model
loading file tokenizer.json
loading file added_tokens.json
loading file special_tokens_map.json
loading file tokenizer_config.json
loading file chat_template.jinja
loading file tokenizer.model
loading file tokenizer.model
loading file tokenizer.json
loading file tokenizer.json
loading file added_tokens.json
loading file added_tokens.json
loading file special_tokens_map.json
loading file special_tokens_map.json
loading file tokenizer_config.json
loading file tokenizer_config.json
loading file chat_template.jinja
loading file chat_template.jinja
You are resizing the embedding layer without providing a `pad_to_multiple_of` parameter. This means that the new embedding dimension will be 32001. This might induce some performance reduction as *Tensor Cores* will not be available. For more details about this, or help on choosing the correct value for resizing, refer to this guide: https://docs.nvidia.com/deeplearning/performance/dl-performance-matrix-multiplication/index.html#requirements-tc
You are resizing the embedding layer without providing a `pad_to_multiple_of` parameter. This means that the new embedding dimension will be 32001. This might induce some performance reduction as *Tensor Cores* will not be available. For more details about this, or help on choosing the correct value for resizing, refer to this guide: https://docs.nvidia.com/deeplearning/performance/dl-performance-matrix-multiplication/index.html#requirements-tc
You are resizing the embedding layer without providing a `pad_to_multiple_of` parameter. This means that the new embedding dimension will be 32001. This might induce some performance reduction as *Tensor Cores* will not be available. For more details about this, or help on choosing the correct value for resizing, refer to this guide: https://docs.nvidia.com/deeplearning/performance/dl-performance-matrix-multiplication/index.html#requirements-tc
You are resizing the embedding layer without providing a `pad_to_multiple_of` parameter. This means that the new embedding dimension will be 32001. This might induce some performance reduction as *Tensor Cores* will not be available. For more details about this, or help on choosing the correct value for resizing, refer to this guide: https://docs.nvidia.com/deeplearning/performance/dl-performance-matrix-multiplication/index.html#requirements-tc
You are resizing the embedding layer without providing a `pad_to_multiple_of` parameter. This means that the new embedding dimension will be 32001. This might induce some performance reduction as *Tensor Cores* will not be available. For more details about this, or help on choosing the correct value for resizing, refer to this guide: https://docs.nvidia.com/deeplearning/performance/dl-performance-matrix-multiplication/index.html#requirements-tc
You are resizing the embedding layer without providing a `pad_to_multiple_of` parameter. This means that the new embedding dimension will be 32001. This might induce some performance reduction as *Tensor Cores* will not be available. For more details about this, or help on choosing the correct value for resizing, refer to this guide: https://docs.nvidia.com/deeplearning/performance/dl-performance-matrix-multiplication/index.html#requirements-tc
You are resizing the embedding layer without providing a `pad_to_multiple_of` parameter. This means that the new embedding dimension will be 32001. This might induce some performance reduction as *Tensor Cores* will not be available. For more details about this, or help on choosing the correct value for resizing, refer to this guide: https://docs.nvidia.com/deeplearning/performance/dl-performance-matrix-multiplication/index.html#requirements-tc
All model checkpoint weights were used when initializing LlamaForCausalLM.

All the weights of LlamaForCausalLM were initialized from the model checkpoint at /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-480k-1T.
If your task is similar to the task the model of the checkpoint was trained on, you can already use LlamaForCausalLM for predictions without further training.
loading configuration file /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-480k-1T/generation_config.json
Generate config GenerationConfig {
  "bos_token_id": 1,
  "eos_token_id": 2,
  "max_length": 2048,
  "pad_token_id": 0
}

loading file tokenizer.model
loading file tokenizer.json
loading file added_tokens.json
loading file special_tokens_map.json
loading file tokenizer_config.json
loading file chat_template.jinja
You are resizing the embedding layer without providing a `pad_to_multiple_of` parameter. This means that the new embedding dimension will be 32001. This might induce some performance reduction as *Tensor Cores* will not be available. For more details about this, or help on choosing the correct value for resizing, refer to this guide: https://docs.nvidia.com/deeplearning/performance/dl-performance-matrix-multiplication/index.html#requirements-tc
The new embeddings will be initialized from a multivariate normal distribution that has old embeddings' mean and covariance. As described in this article: https://nlp.stanford.edu/~johnhew/vocab-expansion.html. To disable this, use `mean_resizing=False`
The new embeddings will be initialized from a multivariate normal distribution that has old embeddings' mean and covariance. As described in this article: https://nlp.stanford.edu/~johnhew/vocab-expansion.html. To disable this, use `mean_resizing=False`
The new embeddings will be initialized from a multivariate normal distribution that has old embeddings' mean and covariance. As described in this article: https://nlp.stanford.edu/~johnhew/vocab-expansion.html. To disable this, use `mean_resizing=False`
The new embeddings will be initialized from a multivariate normal distribution that has old embeddings' mean and covariance. As described in this article: https://nlp.stanford.edu/~johnhew/vocab-expansion.html. To disable this, use `mean_resizing=False`
The new embeddings will be initialized from a multivariate normal distribution that has old embeddings' mean and covariance. As described in this article: https://nlp.stanford.edu/~johnhew/vocab-expansion.html. To disable this, use `mean_resizing=False`
The new embeddings will be initialized from a multivariate normal distribution that has old embeddings' mean and covariance. As described in this article: https://nlp.stanford.edu/~johnhew/vocab-expansion.html. To disable this, use `mean_resizing=False`
The new embeddings will be initialized from a multivariate normal distribution that has old embeddings' mean and covariance. As described in this article: https://nlp.stanford.edu/~johnhew/vocab-expansion.html. To disable this, use `mean_resizing=False`
The new embeddings will be initialized from a multivariate normal distribution that has old embeddings' mean and covariance. As described in this article: https://nlp.stanford.edu/~johnhew/vocab-expansion.html. To disable this, use `mean_resizing=False`
The new lm_head weights will be initialized from a multivariate normal distribution that has old embeddings' mean and covariance. As described in this article: https://nlp.stanford.edu/~johnhew/vocab-expansion.html. To disable this, use `mean_resizing=False`
The new lm_head weights will be initialized from a multivariate normal distribution that has old embeddings' mean and covariance. As described in this article: https://nlp.stanford.edu/~johnhew/vocab-expansion.html. To disable this, use `mean_resizing=False`
The new lm_head weights will be initialized from a multivariate normal distribution that has old embeddings' mean and covariance. As described in this article: https://nlp.stanford.edu/~johnhew/vocab-expansion.html. To disable this, use `mean_resizing=False`
The new lm_head weights will be initialized from a multivariate normal distribution that has old embeddings' mean and covariance. As described in this article: https://nlp.stanford.edu/~johnhew/vocab-expansion.html. To disable this, use `mean_resizing=False`
The new lm_head weights will be initialized from a multivariate normal distribution that has old embeddings' mean and covariance. As described in this article: https://nlp.stanford.edu/~johnhew/vocab-expansion.html. To disable this, use `mean_resizing=False`
The new lm_head weights will be initialized from a multivariate normal distribution that has old embeddings' mean and covariance. As described in this article: https://nlp.stanford.edu/~johnhew/vocab-expansion.html. To disable this, use `mean_resizing=False`
The new lm_head weights will be initialized from a multivariate normal distribution that has old embeddings' mean and covariance. As described in this article: https://nlp.stanford.edu/~johnhew/vocab-expansion.html. To disable this, use `mean_resizing=False`
The new lm_head weights will be initialized from a multivariate normal distribution that has old embeddings' mean and covariance. As described in this article: https://nlp.stanford.edu/~johnhew/vocab-expansion.html. To disable this, use `mean_resizing=False`
Using /home/hansirui_1st/.cache/torch_extensions/py311_cu124 as PyTorch extensions root...
Using /home/hansirui_1st/.cache/torch_extensions/py311_cu124 as PyTorch extensions root...
Using /home/hansirui_1st/.cache/torch_extensions/py311_cu124 as PyTorch extensions root...Using /home/hansirui_1st/.cache/torch_extensions/py311_cu124 as PyTorch extensions root...

Using /home/hansirui_1st/.cache/torch_extensions/py311_cu124 as PyTorch extensions root...
Using /home/hansirui_1st/.cache/torch_extensions/py311_cu124 as PyTorch extensions root...
Using /home/hansirui_1st/.cache/torch_extensions/py311_cu124 as PyTorch extensions root...
Using /home/hansirui_1st/.cache/torch_extensions/py311_cu124 as PyTorch extensions root...
Detected CUDA files, patching ldflags
Emitting ninja build file /home/hansirui_1st/.cache/torch_extensions/py311_cu124/fused_adam/build.ninja...
/aifs4su/hansirui_1st/miniconda3/envs/by-align/lib/python3.11/site-packages/torch/utils/cpp_extension.py:2059: UserWarning: TORCH_CUDA_ARCH_LIST is not set, all archs for visible cards are included for compilation. 
If this is not desired, please set os.environ['TORCH_CUDA_ARCH_LIST'].
  warnings.warn(
Building extension module fused_adam...
Allowing ninja to set a default number of workers... (overridable by setting the environment variable MAX_JOBS=N)
Loading extension module fused_adam...
Loading extension module fused_adam...
Loading extension module fused_adam...
Loading extension module fused_adam...
Loading extension module fused_adam...
Loading extension module fused_adam...
Loading extension module fused_adam...
Loading extension module fused_adam...
wandb: Using wandb-core as the SDK backend.  Please refer to https://wandb.me/wandb-core for more information.
`use_cache=True` is incompatible with gradient checkpointing. Setting `use_cache=False`.
`use_cache=True` is incompatible with gradient checkpointing. Setting `use_cache=False`.
`use_cache=True` is incompatible with gradient checkpointing. Setting `use_cache=False`.
`use_cache=True` is incompatible with gradient checkpointing. Setting `use_cache=False`.
`use_cache=True` is incompatible with gradient checkpointing. Setting `use_cache=False`.
`use_cache=True` is incompatible with gradient checkpointing. Setting `use_cache=False`.
`use_cache=True` is incompatible with gradient checkpointing. Setting `use_cache=False`.
wandb: Currently logged in as: xtom to https://api.wandb.ai. Use `wandb login --relogin` to force relogin
wandb: Tracking run with wandb version 0.19.8
wandb: Run data is saved locally in /aifs4su/hansirui_1st/boyuan/resist/setting3-safety/tinyllama-1T/tinyllama-1T-s3-Q1-30k/wandb/run-20250529_042445-lcs5hdas
wandb: Run `wandb offline` to turn off syncing.
wandb: Syncing run tinyllama-1T-s3-Q1-30k
wandb: ⭐️ View project at https://wandb.ai/xtom/Inverse_Alignment
wandb: 🚀 View run at https://wandb.ai/xtom/Inverse_Alignment/runs/lcs5hdas

Training 1/1 epoch:   0%|          | 0/938 [00:00<?, ?it/s]`use_cache=True` is incompatible with gradient checkpointing. Setting `use_cache=False`.

Training 1/1 epoch (loss 2.3642):   0%|          | 0/938 [00:10<?, ?it/s]
Training 1/1 epoch (loss 2.3642):   0%|          | 1/938 [00:10<2:46:11, 10.64s/it]
Training 1/1 epoch (loss 2.3665):   0%|          | 1/938 [00:13<2:46:11, 10.64s/it]
Training 1/1 epoch (loss 2.3665):   0%|          | 2/938 [00:13<1:34:10,  6.04s/it]
Training 1/1 epoch (loss 2.3674):   0%|          | 2/938 [00:15<1:34:10,  6.04s/it]
Training 1/1 epoch (loss 2.3674):   0%|          | 3/938 [00:15<1:03:17,  4.06s/it]
Training 1/1 epoch (loss 2.3937):   0%|          | 3/938 [00:16<1:03:17,  4.06s/it]
Training 1/1 epoch (loss 2.3937):   0%|          | 4/938 [00:16<48:32,  3.12s/it]  
Training 1/1 epoch (loss 2.2683):   0%|          | 4/938 [00:18<48:32,  3.12s/it]
Training 1/1 epoch (loss 2.2683):   1%|          | 5/938 [00:18<40:16,  2.59s/it]
Training 1/1 epoch (loss 2.2575):   1%|          | 5/938 [00:20<40:16,  2.59s/it]
Training 1/1 epoch (loss 2.2575):   1%|          | 6/938 [00:20<34:43,  2.24s/it]
Training 1/1 epoch (loss 2.2841):   1%|          | 6/938 [00:21<34:43,  2.24s/it]
Training 1/1 epoch (loss 2.2841):   1%|          | 7/938 [00:21<30:33,  1.97s/it]
Training 1/1 epoch (loss 2.3704):   1%|          | 7/938 [00:23<30:33,  1.97s/it]
Training 1/1 epoch (loss 2.3704):   1%|          | 8/938 [00:23<30:06,  1.94s/it]
Training 1/1 epoch (loss 2.2207):   1%|          | 8/938 [00:24<30:06,  1.94s/it]
Training 1/1 epoch (loss 2.2207):   1%|          | 9/938 [00:24<25:54,  1.67s/it]
Training 1/1 epoch (loss 2.2311):   1%|          | 9/938 [00:25<25:54,  1.67s/it]
Training 1/1 epoch (loss 2.2311):   1%|          | 10/938 [00:25<22:47,  1.47s/it]
Training 1/1 epoch (loss 2.2751):   1%|          | 10/938 [00:26<22:47,  1.47s/it]
Training 1/1 epoch (loss 2.2751):   1%|          | 11/938 [00:26<22:29,  1.46s/it]
Training 1/1 epoch (loss 2.3901):   1%|          | 11/938 [00:29<22:29,  1.46s/it]
Training 1/1 epoch (loss 2.3901):   1%|▏         | 12/938 [00:29<25:54,  1.68s/it]
Training 1/1 epoch (loss 2.3838):   1%|▏         | 12/938 [00:31<25:54,  1.68s/it]
Training 1/1 epoch (loss 2.3838):   1%|▏         | 13/938 [00:31<27:37,  1.79s/it]
Training 1/1 epoch (loss 2.1834):   1%|▏         | 13/938 [00:32<27:37,  1.79s/it]
Training 1/1 epoch (loss 2.1834):   1%|▏         | 14/938 [00:32<27:48,  1.81s/it]
Training 1/1 epoch (loss 2.3654):   1%|▏         | 14/938 [00:34<27:48,  1.81s/it]
Training 1/1 epoch (loss 2.3654):   2%|▏         | 15/938 [00:34<27:09,  1.77s/it]
Training 1/1 epoch (loss 2.3359):   2%|▏         | 15/938 [00:36<27:09,  1.77s/it]
Training 1/1 epoch (loss 2.3359):   2%|▏         | 16/938 [00:36<26:31,  1.73s/it]
Training 1/1 epoch (loss 2.1469):   2%|▏         | 16/938 [00:37<26:31,  1.73s/it]
Training 1/1 epoch (loss 2.1469):   2%|▏         | 17/938 [00:37<24:45,  1.61s/it]
Training 1/1 epoch (loss 2.3188):   2%|▏         | 17/938 [00:39<24:45,  1.61s/it]
Training 1/1 epoch (loss 2.3188):   2%|▏         | 18/938 [00:39<26:38,  1.74s/it]
Training 1/1 epoch (loss 2.2120):   2%|▏         | 18/938 [00:41<26:38,  1.74s/it]
Training 1/1 epoch (loss 2.2120):   2%|▏         | 19/938 [00:41<28:07,  1.84s/it]
Training 1/1 epoch (loss 2.2399):   2%|▏         | 19/938 [00:44<28:07,  1.84s/it]
Training 1/1 epoch (loss 2.2399):   2%|▏         | 20/938 [00:44<30:25,  1.99s/it]
Training 1/1 epoch (loss 2.2323):   2%|▏         | 20/938 [00:46<30:25,  1.99s/it]
Training 1/1 epoch (loss 2.2323):   2%|▏         | 21/938 [00:46<30:18,  1.98s/it]
Training 1/1 epoch (loss 2.4383):   2%|▏         | 21/938 [00:47<30:18,  1.98s/it]
Training 1/1 epoch (loss 2.4383):   2%|▏         | 22/938 [00:47<29:47,  1.95s/it]
Training 1/1 epoch (loss 2.1563):   2%|▏         | 22/938 [00:49<29:47,  1.95s/it]
Training 1/1 epoch (loss 2.1563):   2%|▏         | 23/938 [00:49<28:58,  1.90s/it]
Training 1/1 epoch (loss 2.1938):   2%|▏         | 23/938 [00:51<28:58,  1.90s/it]
Training 1/1 epoch (loss 2.1938):   3%|▎         | 24/938 [00:51<28:44,  1.89s/it]
Training 1/1 epoch (loss 2.3226):   3%|▎         | 24/938 [00:52<28:44,  1.89s/it]
Training 1/1 epoch (loss 2.3226):   3%|▎         | 25/938 [00:52<25:03,  1.65s/it]
Training 1/1 epoch (loss 2.3257):   3%|▎         | 25/938 [00:54<25:03,  1.65s/it]
Training 1/1 epoch (loss 2.3257):   3%|▎         | 26/938 [00:54<26:30,  1.74s/it]
Training 1/1 epoch (loss 2.1772):   3%|▎         | 26/938 [00:56<26:30,  1.74s/it]
Training 1/1 epoch (loss 2.1772):   3%|▎         | 27/938 [00:56<25:38,  1.69s/it]
Training 1/1 epoch (loss 2.2153):   3%|▎         | 27/938 [00:58<25:38,  1.69s/it]
Training 1/1 epoch (loss 2.2153):   3%|▎         | 28/938 [00:58<26:36,  1.75s/it]
Training 1/1 epoch (loss 2.3130):   3%|▎         | 28/938 [00:59<26:36,  1.75s/it]
Training 1/1 epoch (loss 2.3130):   3%|▎         | 29/938 [00:59<27:27,  1.81s/it]
Training 1/1 epoch (loss 2.3894):   3%|▎         | 29/938 [01:01<27:27,  1.81s/it]
Training 1/1 epoch (loss 2.3894):   3%|▎         | 30/938 [01:01<25:25,  1.68s/it]
Training 1/1 epoch (loss 2.2545):   3%|▎         | 30/938 [01:03<25:25,  1.68s/it]
Training 1/1 epoch (loss 2.2545):   3%|▎         | 31/938 [01:03<28:00,  1.85s/it]
Training 1/1 epoch (loss 2.2425):   3%|▎         | 31/938 [01:06<28:00,  1.85s/it]
Training 1/1 epoch (loss 2.2425):   3%|▎         | 32/938 [01:06<32:50,  2.17s/it]
Training 1/1 epoch (loss 2.1617):   3%|▎         | 32/938 [01:07<32:50,  2.17s/it]
Training 1/1 epoch (loss 2.1617):   4%|▎         | 33/938 [01:07<28:34,  1.89s/it]
Training 1/1 epoch (loss 2.2476):   4%|▎         | 33/938 [01:09<28:34,  1.89s/it]
Training 1/1 epoch (loss 2.2476):   4%|▎         | 34/938 [01:09<27:23,  1.82s/it]
Training 1/1 epoch (loss 2.1486):   4%|▎         | 34/938 [01:11<27:23,  1.82s/it]
Training 1/1 epoch (loss 2.1486):   4%|▎         | 35/938 [01:11<27:44,  1.84s/it]
Training 1/1 epoch (loss 2.1386):   4%|▎         | 35/938 [01:13<27:44,  1.84s/it]
Training 1/1 epoch (loss 2.1386):   4%|▍         | 36/938 [01:13<27:02,  1.80s/it]
Training 1/1 epoch (loss 2.2049):   4%|▍         | 36/938 [01:14<27:02,  1.80s/it]
Training 1/1 epoch (loss 2.2049):   4%|▍         | 37/938 [01:14<27:01,  1.80s/it]
Training 1/1 epoch (loss 2.2077):   4%|▍         | 37/938 [01:16<27:01,  1.80s/it]
Training 1/1 epoch (loss 2.2077):   4%|▍         | 38/938 [01:16<28:34,  1.91s/it]
Training 1/1 epoch (loss 2.1831):   4%|▍         | 38/938 [01:18<28:34,  1.91s/it]
Training 1/1 epoch (loss 2.1831):   4%|▍         | 39/938 [01:18<26:01,  1.74s/it]
Training 1/1 epoch (loss 2.1193):   4%|▍         | 39/938 [01:19<26:01,  1.74s/it]
Training 1/1 epoch (loss 2.1193):   4%|▍         | 40/938 [01:19<25:09,  1.68s/it]
Training 1/1 epoch (loss 2.1246):   4%|▍         | 40/938 [01:22<25:09,  1.68s/it]
Training 1/1 epoch (loss 2.1246):   4%|▍         | 41/938 [01:22<28:28,  1.91s/it]
Training 1/1 epoch (loss 2.2744):   4%|▍         | 41/938 [01:24<28:28,  1.91s/it]
Training 1/1 epoch (loss 2.2744):   4%|▍         | 42/938 [01:24<28:57,  1.94s/it]
Training 1/1 epoch (loss 2.1316):   4%|▍         | 42/938 [01:26<28:57,  1.94s/it]
Training 1/1 epoch (loss 2.1316):   5%|▍         | 43/938 [01:26<29:48,  2.00s/it]
Training 1/1 epoch (loss 2.1229):   5%|▍         | 43/938 [01:27<29:48,  2.00s/it]
Training 1/1 epoch (loss 2.1229):   5%|▍         | 44/938 [01:27<27:33,  1.85s/it]
Training 1/1 epoch (loss 2.0377):   5%|▍         | 44/938 [01:29<27:33,  1.85s/it]
Training 1/1 epoch (loss 2.0377):   5%|▍         | 45/938 [01:29<26:31,  1.78s/it]
Training 1/1 epoch (loss 2.1367):   5%|▍         | 45/938 [01:31<26:31,  1.78s/it]
Training 1/1 epoch (loss 2.1367):   5%|▍         | 46/938 [01:31<26:31,  1.78s/it]
Training 1/1 epoch (loss 2.1473):   5%|▍         | 46/938 [01:33<26:31,  1.78s/it]
Training 1/1 epoch (loss 2.1473):   5%|▌         | 47/938 [01:33<26:49,  1.81s/it]
Training 1/1 epoch (loss 2.1490):   5%|▌         | 47/938 [01:35<26:49,  1.81s/it]
Training 1/1 epoch (loss 2.1490):   5%|▌         | 48/938 [01:35<29:15,  1.97s/it]
Training 1/1 epoch (loss 2.1981):   5%|▌         | 48/938 [01:36<29:15,  1.97s/it]
Training 1/1 epoch (loss 2.1981):   5%|▌         | 49/938 [01:36<26:14,  1.77s/it]
Training 1/1 epoch (loss 2.0060):   5%|▌         | 49/938 [01:38<26:14,  1.77s/it]
Training 1/1 epoch (loss 2.0060):   5%|▌         | 50/938 [01:38<26:21,  1.78s/it]
Training 1/1 epoch (loss 2.0804):   5%|▌         | 50/938 [01:40<26:21,  1.78s/it]
Training 1/1 epoch (loss 2.0804):   5%|▌         | 51/938 [01:40<25:26,  1.72s/it]
Training 1/1 epoch (loss 2.0610):   5%|▌         | 51/938 [01:41<25:26,  1.72s/it]
Training 1/1 epoch (loss 2.0610):   6%|▌         | 52/938 [01:41<25:11,  1.71s/it]
Training 1/1 epoch (loss 2.0236):   6%|▌         | 52/938 [01:42<25:11,  1.71s/it]
Training 1/1 epoch (loss 2.0236):   6%|▌         | 53/938 [01:42<21:06,  1.43s/it]
Training 1/1 epoch (loss 2.1761):   6%|▌         | 53/938 [01:43<21:06,  1.43s/it]
Training 1/1 epoch (loss 2.1761):   6%|▌         | 54/938 [01:43<18:32,  1.26s/it]
Training 1/1 epoch (loss 2.1312):   6%|▌         | 54/938 [01:45<18:32,  1.26s/it]
Training 1/1 epoch (loss 2.1312):   6%|▌         | 55/938 [01:45<23:08,  1.57s/it]
Training 1/1 epoch (loss 2.0028):   6%|▌         | 55/938 [01:48<23:08,  1.57s/it]
Training 1/1 epoch (loss 2.0028):   6%|▌         | 56/938 [01:48<27:39,  1.88s/it]
Training 1/1 epoch (loss 2.0148):   6%|▌         | 56/938 [01:50<27:39,  1.88s/it]
Training 1/1 epoch (loss 2.0148):   6%|▌         | 57/938 [01:50<28:20,  1.93s/it]
Training 1/1 epoch (loss 2.0791):   6%|▌         | 57/938 [01:51<28:20,  1.93s/it]
Training 1/1 epoch (loss 2.0791):   6%|▌         | 58/938 [01:51<26:07,  1.78s/it]
Training 1/1 epoch (loss 2.0438):   6%|▌         | 58/938 [01:54<26:07,  1.78s/it]
Training 1/1 epoch (loss 2.0438):   6%|▋         | 59/938 [01:54<27:39,  1.89s/it]
Training 1/1 epoch (loss 2.0343):   6%|▋         | 59/938 [01:55<27:39,  1.89s/it]
Training 1/1 epoch (loss 2.0343):   6%|▋         | 60/938 [01:55<25:59,  1.78s/it]
Training 1/1 epoch (loss 2.0962):   6%|▋         | 60/938 [01:56<25:59,  1.78s/it]
Training 1/1 epoch (loss 2.0962):   7%|▋         | 61/938 [01:56<23:13,  1.59s/it]
Training 1/1 epoch (loss 2.0811):   7%|▋         | 61/938 [01:59<23:13,  1.59s/it]
Training 1/1 epoch (loss 2.0811):   7%|▋         | 62/938 [01:59<26:39,  1.83s/it]
Training 1/1 epoch (loss 2.0461):   7%|▋         | 62/938 [02:00<26:39,  1.83s/it]
Training 1/1 epoch (loss 2.0461):   7%|▋         | 63/938 [02:00<25:42,  1.76s/it]
Training 1/1 epoch (loss 2.0340):   7%|▋         | 63/938 [02:02<25:42,  1.76s/it]
Training 1/1 epoch (loss 2.0340):   7%|▋         | 64/938 [02:02<26:20,  1.81s/it]
Training 1/1 epoch (loss 1.9828):   7%|▋         | 64/938 [02:04<26:20,  1.81s/it]
Training 1/1 epoch (loss 1.9828):   7%|▋         | 65/938 [02:04<27:38,  1.90s/it]
Training 1/1 epoch (loss 2.1134):   7%|▋         | 65/938 [02:06<27:38,  1.90s/it]
Training 1/1 epoch (loss 2.1134):   7%|▋         | 66/938 [02:06<28:26,  1.96s/it]
Training 1/1 epoch (loss 2.0507):   7%|▋         | 66/938 [02:08<28:26,  1.96s/it]
Training 1/1 epoch (loss 2.0507):   7%|▋         | 67/938 [02:08<24:49,  1.71s/it]
Training 1/1 epoch (loss 2.0059):   7%|▋         | 67/938 [02:10<24:49,  1.71s/it]
Training 1/1 epoch (loss 2.0059):   7%|▋         | 68/938 [02:10<27:56,  1.93s/it]
Training 1/1 epoch (loss 2.0370):   7%|▋         | 68/938 [02:12<27:56,  1.93s/it]
Training 1/1 epoch (loss 2.0370):   7%|▋         | 69/938 [02:12<27:44,  1.92s/it]
Training 1/1 epoch (loss 2.1034):   7%|▋         | 69/938 [02:13<27:44,  1.92s/it]
Training 1/1 epoch (loss 2.1034):   7%|▋         | 70/938 [02:13<26:27,  1.83s/it]
Training 1/1 epoch (loss 1.9836):   7%|▋         | 70/938 [02:15<26:27,  1.83s/it]
Training 1/1 epoch (loss 1.9836):   8%|▊         | 71/938 [02:15<26:27,  1.83s/it]
Training 1/1 epoch (loss 2.0255):   8%|▊         | 71/938 [02:17<26:27,  1.83s/it]
Training 1/1 epoch (loss 2.0255):   8%|▊         | 72/938 [02:17<27:37,  1.91s/it]
Training 1/1 epoch (loss 1.9502):   8%|▊         | 72/938 [02:19<27:37,  1.91s/it]
Training 1/1 epoch (loss 1.9502):   8%|▊         | 73/938 [02:19<27:50,  1.93s/it]
Training 1/1 epoch (loss 1.9574):   8%|▊         | 73/938 [02:21<27:50,  1.93s/it]
Training 1/1 epoch (loss 1.9574):   8%|▊         | 74/938 [02:21<28:20,  1.97s/it]
Training 1/1 epoch (loss 1.9175):   8%|▊         | 74/938 [02:23<28:20,  1.97s/it]
Training 1/1 epoch (loss 1.9175):   8%|▊         | 75/938 [02:23<26:07,  1.82s/it]
Training 1/1 epoch (loss 2.0255):   8%|▊         | 75/938 [02:24<26:07,  1.82s/it]
Training 1/1 epoch (loss 2.0255):   8%|▊         | 76/938 [02:24<24:19,  1.69s/it]
Training 1/1 epoch (loss 1.8727):   8%|▊         | 76/938 [02:27<24:19,  1.69s/it]
Training 1/1 epoch (loss 1.8727):   8%|▊         | 77/938 [02:27<26:42,  1.86s/it]
Training 1/1 epoch (loss 2.0836):   8%|▊         | 77/938 [02:29<26:42,  1.86s/it]
Training 1/1 epoch (loss 2.0836):   8%|▊         | 78/938 [02:29<28:52,  2.01s/it]
Training 1/1 epoch (loss 2.0942):   8%|▊         | 78/938 [02:31<28:52,  2.01s/it]
Training 1/1 epoch (loss 2.0942):   8%|▊         | 79/938 [02:31<27:55,  1.95s/it]
Training 1/1 epoch (loss 2.0371):   8%|▊         | 79/938 [02:32<27:55,  1.95s/it]
Training 1/1 epoch (loss 2.0371):   9%|▊         | 80/938 [02:32<25:13,  1.76s/it]
Training 1/1 epoch (loss 2.0046):   9%|▊         | 80/938 [02:34<25:13,  1.76s/it]
Training 1/1 epoch (loss 2.0046):   9%|▊         | 81/938 [02:34<25:25,  1.78s/it]
Training 1/1 epoch (loss 1.8833):   9%|▊         | 81/938 [02:36<25:25,  1.78s/it]
Training 1/1 epoch (loss 1.8833):   9%|▊         | 82/938 [02:36<24:48,  1.74s/it]
Training 1/1 epoch (loss 1.9973):   9%|▊         | 82/938 [02:37<24:48,  1.74s/it]
Training 1/1 epoch (loss 1.9973):   9%|▉         | 83/938 [02:37<25:32,  1.79s/it]
Training 1/1 epoch (loss 1.8684):   9%|▉         | 83/938 [02:39<25:32,  1.79s/it]
Training 1/1 epoch (loss 1.8684):   9%|▉         | 84/938 [02:39<24:09,  1.70s/it]
Training 1/1 epoch (loss 1.9141):   9%|▉         | 84/938 [02:40<24:09,  1.70s/it]
Training 1/1 epoch (loss 1.9141):   9%|▉         | 85/938 [02:40<21:32,  1.52s/it]
Training 1/1 epoch (loss 1.9408):   9%|▉         | 85/938 [02:42<21:32,  1.52s/it]
Training 1/1 epoch (loss 1.9408):   9%|▉         | 86/938 [02:42<25:04,  1.77s/it]
Training 1/1 epoch (loss 1.9982):   9%|▉         | 86/938 [02:44<25:04,  1.77s/it]
Training 1/1 epoch (loss 1.9982):   9%|▉         | 87/938 [02:44<24:40,  1.74s/it]
Training 1/1 epoch (loss 1.8733):   9%|▉         | 87/938 [02:47<24:40,  1.74s/it]
Training 1/1 epoch (loss 1.8733):   9%|▉         | 88/938 [02:47<28:08,  1.99s/it]
Training 1/1 epoch (loss 1.9091):   9%|▉         | 88/938 [02:48<28:08,  1.99s/it]
Training 1/1 epoch (loss 1.9091):   9%|▉         | 89/938 [02:48<27:13,  1.92s/it]
Training 1/1 epoch (loss 1.9175):   9%|▉         | 89/938 [02:50<27:13,  1.92s/it]
Training 1/1 epoch (loss 1.9175):  10%|▉         | 90/938 [02:50<25:08,  1.78s/it]
Training 1/1 epoch (loss 2.0434):  10%|▉         | 90/938 [02:51<25:08,  1.78s/it]
Training 1/1 epoch (loss 2.0434):  10%|▉         | 91/938 [02:51<23:17,  1.65s/it]
Training 1/1 epoch (loss 1.8274):  10%|▉         | 91/938 [02:54<23:17,  1.65s/it]
Training 1/1 epoch (loss 1.8274):  10%|▉         | 92/938 [02:54<26:40,  1.89s/it]
Training 1/1 epoch (loss 2.0520):  10%|▉         | 92/938 [02:55<26:40,  1.89s/it]
Training 1/1 epoch (loss 2.0520):  10%|▉         | 93/938 [02:55<23:05,  1.64s/it]
Training 1/1 epoch (loss 1.9133):  10%|▉         | 93/938 [02:57<23:05,  1.64s/it]
Training 1/1 epoch (loss 1.9133):  10%|█         | 94/938 [02:57<24:15,  1.72s/it]
Training 1/1 epoch (loss 1.8978):  10%|█         | 94/938 [02:58<24:15,  1.72s/it]
Training 1/1 epoch (loss 1.8978):  10%|█         | 95/938 [02:58<21:57,  1.56s/it]
Training 1/1 epoch (loss 1.9623):  10%|█         | 95/938 [03:00<21:57,  1.56s/it]
Training 1/1 epoch (loss 1.9623):  10%|█         | 96/938 [03:00<22:45,  1.62s/it]
Training 1/1 epoch (loss 1.9836):  10%|█         | 96/938 [03:01<22:45,  1.62s/it]
Training 1/1 epoch (loss 1.9836):  10%|█         | 97/938 [03:01<23:55,  1.71s/it]
Training 1/1 epoch (loss 1.8903):  10%|█         | 97/938 [03:04<23:55,  1.71s/it]
Training 1/1 epoch (loss 1.8903):  10%|█         | 98/938 [03:04<25:41,  1.83s/it]
Training 1/1 epoch (loss 1.9490):  10%|█         | 98/938 [03:05<25:41,  1.83s/it]
Training 1/1 epoch (loss 1.9490):  11%|█         | 99/938 [03:05<22:42,  1.62s/it]
Training 1/1 epoch (loss 1.7534):  11%|█         | 99/938 [03:06<22:42,  1.62s/it]
Training 1/1 epoch (loss 1.7534):  11%|█         | 100/938 [03:06<22:39,  1.62s/it]
Training 1/1 epoch (loss 1.8693):  11%|█         | 100/938 [03:08<22:39,  1.62s/it]
Training 1/1 epoch (loss 1.8693):  11%|█         | 101/938 [03:08<22:50,  1.64s/it]
Training 1/1 epoch (loss 1.8911):  11%|█         | 101/938 [03:09<22:50,  1.64s/it]
Training 1/1 epoch (loss 1.8911):  11%|█         | 102/938 [03:09<21:08,  1.52s/it]
Training 1/1 epoch (loss 1.8560):  11%|█         | 102/938 [03:11<21:08,  1.52s/it]
Training 1/1 epoch (loss 1.8560):  11%|█         | 103/938 [03:11<21:50,  1.57s/it]
Training 1/1 epoch (loss 1.9776):  11%|█         | 103/938 [03:12<21:50,  1.57s/it]
Training 1/1 epoch (loss 1.9776):  11%|█         | 104/938 [03:12<20:36,  1.48s/it]
Training 1/1 epoch (loss 1.9027):  11%|█         | 104/938 [03:13<20:36,  1.48s/it]
Training 1/1 epoch (loss 1.9027):  11%|█         | 105/938 [03:13<19:34,  1.41s/it]
Training 1/1 epoch (loss 1.9151):  11%|█         | 105/938 [03:15<19:34,  1.41s/it]
Training 1/1 epoch (loss 1.9151):  11%|█▏        | 106/938 [03:15<21:37,  1.56s/it]
Training 1/1 epoch (loss 1.9571):  11%|█▏        | 106/938 [03:17<21:37,  1.56s/it]
Training 1/1 epoch (loss 1.9571):  11%|█▏        | 107/938 [03:17<22:24,  1.62s/it]
Training 1/1 epoch (loss 2.0114):  11%|█▏        | 107/938 [03:19<22:24,  1.62s/it]
Training 1/1 epoch (loss 2.0114):  12%|█▏        | 108/938 [03:19<22:14,  1.61s/it]
Training 1/1 epoch (loss 1.9454):  12%|█▏        | 108/938 [03:21<22:14,  1.61s/it]
Training 1/1 epoch (loss 1.9454):  12%|█▏        | 109/938 [03:21<24:16,  1.76s/it]
Training 1/1 epoch (loss 1.9584):  12%|█▏        | 109/938 [03:23<24:16,  1.76s/it]
Training 1/1 epoch (loss 1.9584):  12%|█▏        | 110/938 [03:23<26:03,  1.89s/it]
Training 1/1 epoch (loss 1.8025):  12%|█▏        | 110/938 [03:25<26:03,  1.89s/it]
Training 1/1 epoch (loss 1.8025):  12%|█▏        | 111/938 [03:25<25:44,  1.87s/it]
Training 1/1 epoch (loss 1.9291):  12%|█▏        | 111/938 [03:27<25:44,  1.87s/it]
Training 1/1 epoch (loss 1.9291):  12%|█▏        | 112/938 [03:27<28:49,  2.09s/it]
Training 1/1 epoch (loss 1.8292):  12%|█▏        | 112/938 [03:29<28:49,  2.09s/it]
Training 1/1 epoch (loss 1.8292):  12%|█▏        | 113/938 [03:29<26:16,  1.91s/it]
Training 1/1 epoch (loss 1.8177):  12%|█▏        | 113/938 [03:30<26:16,  1.91s/it]
Training 1/1 epoch (loss 1.8177):  12%|█▏        | 114/938 [03:30<23:46,  1.73s/it]
Training 1/1 epoch (loss 1.8093):  12%|█▏        | 114/938 [03:32<23:46,  1.73s/it]
Training 1/1 epoch (loss 1.8093):  12%|█▏        | 115/938 [03:32<22:45,  1.66s/it]
Training 1/1 epoch (loss 1.9370):  12%|█▏        | 115/938 [03:34<22:45,  1.66s/it]
Training 1/1 epoch (loss 1.9370):  12%|█▏        | 116/938 [03:34<23:42,  1.73s/it]
Training 1/1 epoch (loss 1.8238):  12%|█▏        | 116/938 [03:35<23:42,  1.73s/it]
Training 1/1 epoch (loss 1.8238):  12%|█▏        | 117/938 [03:35<23:18,  1.70s/it]
Training 1/1 epoch (loss 1.8799):  12%|█▏        | 117/938 [03:37<23:18,  1.70s/it]
Training 1/1 epoch (loss 1.8799):  13%|█▎        | 118/938 [03:37<22:34,  1.65s/it]
Training 1/1 epoch (loss 1.9194):  13%|█▎        | 118/938 [03:38<22:34,  1.65s/it]
Training 1/1 epoch (loss 1.9194):  13%|█▎        | 119/938 [03:38<19:11,  1.41s/it]
Training 1/1 epoch (loss 1.7945):  13%|█▎        | 119/938 [03:39<19:11,  1.41s/it]
Training 1/1 epoch (loss 1.7945):  13%|█▎        | 120/938 [03:39<19:56,  1.46s/it]
Training 1/1 epoch (loss 1.8436):  13%|█▎        | 120/938 [03:41<19:56,  1.46s/it]
Training 1/1 epoch (loss 1.8436):  13%|█▎        | 121/938 [03:41<22:11,  1.63s/it]
Training 1/1 epoch (loss 1.8499):  13%|█▎        | 121/938 [03:43<22:11,  1.63s/it]
Training 1/1 epoch (loss 1.8499):  13%|█▎        | 122/938 [03:43<20:58,  1.54s/it]
Training 1/1 epoch (loss 1.8290):  13%|█▎        | 122/938 [03:45<20:58,  1.54s/it]
Training 1/1 epoch (loss 1.8290):  13%|█▎        | 123/938 [03:45<22:38,  1.67s/it]
Training 1/1 epoch (loss 1.8121):  13%|█▎        | 123/938 [03:47<22:38,  1.67s/it]
Training 1/1 epoch (loss 1.8121):  13%|█▎        | 124/938 [03:47<25:42,  1.90s/it]
Training 1/1 epoch (loss 1.9223):  13%|█▎        | 124/938 [03:49<25:42,  1.90s/it]
Training 1/1 epoch (loss 1.9223):  13%|█▎        | 125/938 [03:49<24:24,  1.80s/it]
Training 1/1 epoch (loss 1.9285):  13%|█▎        | 125/938 [03:51<24:24,  1.80s/it]
Training 1/1 epoch (loss 1.9285):  13%|█▎        | 126/938 [03:51<26:49,  1.98s/it]
Training 1/1 epoch (loss 1.7297):  13%|█▎        | 126/938 [03:53<26:49,  1.98s/it]
Training 1/1 epoch (loss 1.7297):  14%|█▎        | 127/938 [03:53<25:35,  1.89s/it]
Training 1/1 epoch (loss 1.7822):  14%|█▎        | 127/938 [03:55<25:35,  1.89s/it]
Training 1/1 epoch (loss 1.7822):  14%|█▎        | 128/938 [03:55<27:44,  2.05s/it]
Training 1/1 epoch (loss 1.8257):  14%|█▎        | 128/938 [03:57<27:44,  2.05s/it]
Training 1/1 epoch (loss 1.8257):  14%|█▍        | 129/938 [03:57<28:01,  2.08s/it]
Training 1/1 epoch (loss 1.7592):  14%|█▍        | 129/938 [03:59<28:01,  2.08s/it]
Training 1/1 epoch (loss 1.7592):  14%|█▍        | 130/938 [03:59<28:10,  2.09s/it]
Training 1/1 epoch (loss 1.8317):  14%|█▍        | 130/938 [04:01<28:10,  2.09s/it]
Training 1/1 epoch (loss 1.8317):  14%|█▍        | 131/938 [04:01<27:17,  2.03s/it]
Training 1/1 epoch (loss 1.7636):  14%|█▍        | 131/938 [04:03<27:17,  2.03s/it]
Training 1/1 epoch (loss 1.7636):  14%|█▍        | 132/938 [04:03<27:03,  2.01s/it]
Training 1/1 epoch (loss 1.8268):  14%|█▍        | 132/938 [04:04<27:03,  2.01s/it]
Training 1/1 epoch (loss 1.8268):  14%|█▍        | 133/938 [04:04<23:19,  1.74s/it]
Training 1/1 epoch (loss 1.9469):  14%|█▍        | 133/938 [04:06<23:19,  1.74s/it]
Training 1/1 epoch (loss 1.9469):  14%|█▍        | 134/938 [04:06<22:52,  1.71s/it]
Training 1/1 epoch (loss 1.7587):  14%|█▍        | 134/938 [04:07<22:52,  1.71s/it]
Training 1/1 epoch (loss 1.7587):  14%|█▍        | 135/938 [04:07<21:59,  1.64s/it]
Training 1/1 epoch (loss 1.8409):  14%|█▍        | 135/938 [04:09<21:59,  1.64s/it]
Training 1/1 epoch (loss 1.8409):  14%|█▍        | 136/938 [04:09<22:09,  1.66s/it]
Training 1/1 epoch (loss 1.8444):  14%|█▍        | 136/938 [04:11<22:09,  1.66s/it]
Training 1/1 epoch (loss 1.8444):  15%|█▍        | 137/938 [04:11<22:28,  1.68s/it]
Training 1/1 epoch (loss 1.8162):  15%|█▍        | 137/938 [04:13<22:28,  1.68s/it]
Training 1/1 epoch (loss 1.8162):  15%|█▍        | 138/938 [04:13<22:49,  1.71s/it]
Training 1/1 epoch (loss 1.7810):  15%|█▍        | 138/938 [04:14<22:49,  1.71s/it]
Training 1/1 epoch (loss 1.7810):  15%|█▍        | 139/938 [04:14<21:18,  1.60s/it]
Training 1/1 epoch (loss 1.7721):  15%|█▍        | 139/938 [04:16<21:18,  1.60s/it]
Training 1/1 epoch (loss 1.7721):  15%|█▍        | 140/938 [04:16<24:28,  1.84s/it]
Training 1/1 epoch (loss 1.8406):  15%|█▍        | 140/938 [04:18<24:28,  1.84s/it]
Training 1/1 epoch (loss 1.8406):  15%|█▌        | 141/938 [04:18<22:34,  1.70s/it]
Training 1/1 epoch (loss 1.8185):  15%|█▌        | 141/938 [04:20<22:34,  1.70s/it]
Training 1/1 epoch (loss 1.8185):  15%|█▌        | 142/938 [04:20<22:55,  1.73s/it]
Training 1/1 epoch (loss 1.7882):  15%|█▌        | 142/938 [04:21<22:55,  1.73s/it]
Training 1/1 epoch (loss 1.7882):  15%|█▌        | 143/938 [04:21<21:46,  1.64s/it]
Training 1/1 epoch (loss 1.8870):  15%|█▌        | 143/938 [04:23<21:46,  1.64s/it]
Training 1/1 epoch (loss 1.8870):  15%|█▌        | 144/938 [04:23<22:14,  1.68s/it]
Training 1/1 epoch (loss 1.7977):  15%|█▌        | 144/938 [04:24<22:14,  1.68s/it]
Training 1/1 epoch (loss 1.7977):  15%|█▌        | 145/938 [04:24<22:27,  1.70s/it]
Training 1/1 epoch (loss 1.7764):  15%|█▌        | 145/938 [04:27<22:27,  1.70s/it]
Training 1/1 epoch (loss 1.7764):  16%|█▌        | 146/938 [04:27<25:19,  1.92s/it]
Training 1/1 epoch (loss 1.6524):  16%|█▌        | 146/938 [04:28<25:19,  1.92s/it]
Training 1/1 epoch (loss 1.6524):  16%|█▌        | 147/938 [04:28<23:51,  1.81s/it]
Training 1/1 epoch (loss 1.9128):  16%|█▌        | 147/938 [04:30<23:51,  1.81s/it]
Training 1/1 epoch (loss 1.9128):  16%|█▌        | 148/938 [04:30<22:04,  1.68s/it]
Training 1/1 epoch (loss 1.9650):  16%|█▌        | 148/938 [04:31<22:04,  1.68s/it]
Training 1/1 epoch (loss 1.9650):  16%|█▌        | 149/938 [04:31<19:49,  1.51s/it]
Training 1/1 epoch (loss 1.7469):  16%|█▌        | 149/938 [04:32<19:49,  1.51s/it]
Training 1/1 epoch (loss 1.7469):  16%|█▌        | 150/938 [04:32<19:52,  1.51s/it]
Training 1/1 epoch (loss 1.7817):  16%|█▌        | 150/938 [04:34<19:52,  1.51s/it]
Training 1/1 epoch (loss 1.7817):  16%|█▌        | 151/938 [04:34<19:00,  1.45s/it]
Training 1/1 epoch (loss 1.7564):  16%|█▌        | 151/938 [04:37<19:00,  1.45s/it]
Training 1/1 epoch (loss 1.7564):  16%|█▌        | 152/938 [04:37<24:10,  1.85s/it]
Training 1/1 epoch (loss 1.8061):  16%|█▌        | 152/938 [04:39<24:10,  1.85s/it]
Training 1/1 epoch (loss 1.8061):  16%|█▋        | 153/938 [04:39<24:54,  1.90s/it]
Training 1/1 epoch (loss 1.6536):  16%|█▋        | 153/938 [04:41<24:54,  1.90s/it]
Training 1/1 epoch (loss 1.6536):  16%|█▋        | 154/938 [04:41<26:49,  2.05s/it]
Training 1/1 epoch (loss 1.9130):  16%|█▋        | 154/938 [04:43<26:49,  2.05s/it]
Training 1/1 epoch (loss 1.9130):  17%|█▋        | 155/938 [04:43<25:07,  1.93s/it]
Training 1/1 epoch (loss 1.7844):  17%|█▋        | 155/938 [04:44<25:07,  1.93s/it]
Training 1/1 epoch (loss 1.7844):  17%|█▋        | 156/938 [04:44<22:45,  1.75s/it]
Training 1/1 epoch (loss 1.7976):  17%|█▋        | 156/938 [04:45<22:45,  1.75s/it]
Training 1/1 epoch (loss 1.7976):  17%|█▋        | 157/938 [04:45<21:52,  1.68s/it]
Training 1/1 epoch (loss 1.7527):  17%|█▋        | 157/938 [04:48<21:52,  1.68s/it]
Training 1/1 epoch (loss 1.7527):  17%|█▋        | 158/938 [04:48<24:11,  1.86s/it]
Training 1/1 epoch (loss 1.8635):  17%|█▋        | 158/938 [04:49<24:11,  1.86s/it]
Training 1/1 epoch (loss 1.8635):  17%|█▋        | 159/938 [04:49<21:21,  1.64s/it]
Training 1/1 epoch (loss 1.8322):  17%|█▋        | 159/938 [04:51<21:21,  1.64s/it]
Training 1/1 epoch (loss 1.8322):  17%|█▋        | 160/938 [04:51<21:19,  1.65s/it]
Training 1/1 epoch (loss 1.7230):  17%|█▋        | 160/938 [04:52<21:19,  1.65s/it]
Training 1/1 epoch (loss 1.7230):  17%|█▋        | 161/938 [04:52<19:43,  1.52s/it]
Training 1/1 epoch (loss 1.7893):  17%|█▋        | 161/938 [04:53<19:43,  1.52s/it]
Training 1/1 epoch (loss 1.7893):  17%|█▋        | 162/938 [04:53<20:00,  1.55s/it]
Training 1/1 epoch (loss 1.8013):  17%|█▋        | 162/938 [04:55<20:00,  1.55s/it]
Training 1/1 epoch (loss 1.8013):  17%|█▋        | 163/938 [04:55<22:04,  1.71s/it]
Training 1/1 epoch (loss 1.7524):  17%|█▋        | 163/938 [04:57<22:04,  1.71s/it]
Training 1/1 epoch (loss 1.7524):  17%|█▋        | 164/938 [04:57<20:41,  1.60s/it]
Training 1/1 epoch (loss 1.8161):  17%|█▋        | 164/938 [04:58<20:41,  1.60s/it]
Training 1/1 epoch (loss 1.8161):  18%|█▊        | 165/938 [04:58<19:53,  1.54s/it]
Training 1/1 epoch (loss 1.7237):  18%|█▊        | 165/938 [05:00<19:53,  1.54s/it]
Training 1/1 epoch (loss 1.7237):  18%|█▊        | 166/938 [05:00<22:27,  1.74s/it]
Training 1/1 epoch (loss 1.8421):  18%|█▊        | 166/938 [05:02<22:27,  1.74s/it]
Training 1/1 epoch (loss 1.8421):  18%|█▊        | 167/938 [05:02<23:10,  1.80s/it]
Training 1/1 epoch (loss 1.6453):  18%|█▊        | 167/938 [05:05<23:10,  1.80s/it]
Training 1/1 epoch (loss 1.6453):  18%|█▊        | 168/938 [05:05<24:39,  1.92s/it]
Training 1/1 epoch (loss 1.9326):  18%|█▊        | 168/938 [05:07<24:39,  1.92s/it]
Training 1/1 epoch (loss 1.9326):  18%|█▊        | 169/938 [05:07<24:37,  1.92s/it]
Training 1/1 epoch (loss 1.7732):  18%|█▊        | 169/938 [05:09<24:37,  1.92s/it]
Training 1/1 epoch (loss 1.7732):  18%|█▊        | 170/938 [05:09<26:36,  2.08s/it]
Training 1/1 epoch (loss 1.7324):  18%|█▊        | 170/938 [05:11<26:36,  2.08s/it]
Training 1/1 epoch (loss 1.7324):  18%|█▊        | 171/938 [05:11<26:48,  2.10s/it]
Training 1/1 epoch (loss 1.7094):  18%|█▊        | 171/938 [05:13<26:48,  2.10s/it]
Training 1/1 epoch (loss 1.7094):  18%|█▊        | 172/938 [05:13<24:25,  1.91s/it]
Training 1/1 epoch (loss 1.7811):  18%|█▊        | 172/938 [05:15<24:25,  1.91s/it]
Training 1/1 epoch (loss 1.7811):  18%|█▊        | 173/938 [05:15<24:38,  1.93s/it]
Training 1/1 epoch (loss 1.8619):  18%|█▊        | 173/938 [05:17<24:38,  1.93s/it]
Training 1/1 epoch (loss 1.8619):  19%|█▊        | 174/938 [05:17<26:53,  2.11s/it]
Training 1/1 epoch (loss 1.8221):  19%|█▊        | 174/938 [05:19<26:53,  2.11s/it]
Training 1/1 epoch (loss 1.8221):  19%|█▊        | 175/938 [05:19<26:13,  2.06s/it]
Training 1/1 epoch (loss 1.6411):  19%|█▊        | 175/938 [05:21<26:13,  2.06s/it]
Training 1/1 epoch (loss 1.6411):  19%|█▉        | 176/938 [05:21<24:27,  1.93s/it]
Training 1/1 epoch (loss 1.7428):  19%|█▉        | 176/938 [05:23<24:27,  1.93s/it]
Training 1/1 epoch (loss 1.7428):  19%|█▉        | 177/938 [05:23<25:13,  1.99s/it]
Training 1/1 epoch (loss 1.8113):  19%|█▉        | 177/938 [05:25<25:13,  1.99s/it]
Training 1/1 epoch (loss 1.8113):  19%|█▉        | 178/938 [05:25<24:28,  1.93s/it]
Training 1/1 epoch (loss 1.8721):  19%|█▉        | 178/938 [05:27<24:28,  1.93s/it]
Training 1/1 epoch (loss 1.8721):  19%|█▉        | 179/938 [05:27<24:45,  1.96s/it]
Training 1/1 epoch (loss 1.7056):  19%|█▉        | 179/938 [05:28<24:45,  1.96s/it]
Training 1/1 epoch (loss 1.7056):  19%|█▉        | 180/938 [05:28<23:32,  1.86s/it]
Training 1/1 epoch (loss 1.7595):  19%|█▉        | 180/938 [05:29<23:32,  1.86s/it]
Training 1/1 epoch (loss 1.7595):  19%|█▉        | 181/938 [05:29<20:37,  1.63s/it]
Training 1/1 epoch (loss 1.9005):  19%|█▉        | 181/938 [05:31<20:37,  1.63s/it]
Training 1/1 epoch (loss 1.9005):  19%|█▉        | 182/938 [05:31<20:49,  1.65s/it]
Training 1/1 epoch (loss 1.8799):  19%|█▉        | 182/938 [05:33<20:49,  1.65s/it]
Training 1/1 epoch (loss 1.8799):  20%|█▉        | 183/938 [05:33<22:42,  1.80s/it]
Training 1/1 epoch (loss 1.7763):  20%|█▉        | 183/938 [05:35<22:42,  1.80s/it]
Training 1/1 epoch (loss 1.7763):  20%|█▉        | 184/938 [05:35<21:36,  1.72s/it]
Training 1/1 epoch (loss 1.7889):  20%|█▉        | 184/938 [05:36<21:36,  1.72s/it]
Training 1/1 epoch (loss 1.7889):  20%|█▉        | 185/938 [05:36<21:45,  1.73s/it]
Training 1/1 epoch (loss 1.7819):  20%|█▉        | 185/938 [05:39<21:45,  1.73s/it]
Training 1/1 epoch (loss 1.7819):  20%|█▉        | 186/938 [05:39<22:58,  1.83s/it]
Training 1/1 epoch (loss 1.6605):  20%|█▉        | 186/938 [05:40<22:58,  1.83s/it]
Training 1/1 epoch (loss 1.6605):  20%|█▉        | 187/938 [05:40<21:50,  1.74s/it]
Training 1/1 epoch (loss 1.7226):  20%|█▉        | 187/938 [05:42<21:50,  1.74s/it]
Training 1/1 epoch (loss 1.7226):  20%|██        | 188/938 [05:42<20:46,  1.66s/it]
Training 1/1 epoch (loss 1.7657):  20%|██        | 188/938 [05:43<20:46,  1.66s/it]
Training 1/1 epoch (loss 1.7657):  20%|██        | 189/938 [05:43<19:07,  1.53s/it]
Training 1/1 epoch (loss 2.0084):  20%|██        | 189/938 [05:45<19:07,  1.53s/it]
Training 1/1 epoch (loss 2.0084):  20%|██        | 190/938 [05:45<20:23,  1.64s/it]
Training 1/1 epoch (loss 1.7288):  20%|██        | 190/938 [05:47<20:23,  1.64s/it]
Training 1/1 epoch (loss 1.7288):  20%|██        | 191/938 [05:47<21:54,  1.76s/it]
Training 1/1 epoch (loss 1.7794):  20%|██        | 191/938 [05:49<21:54,  1.76s/it]
Training 1/1 epoch (loss 1.7794):  20%|██        | 192/938 [05:49<25:11,  2.03s/it]
Training 1/1 epoch (loss 1.7418):  20%|██        | 192/938 [05:51<25:11,  2.03s/it]
Training 1/1 epoch (loss 1.7418):  21%|██        | 193/938 [05:51<22:46,  1.83s/it]
Training 1/1 epoch (loss 1.7704):  21%|██        | 193/938 [05:53<22:46,  1.83s/it]
Training 1/1 epoch (loss 1.7704):  21%|██        | 194/938 [05:53<24:11,  1.95s/it]
Training 1/1 epoch (loss 1.7807):  21%|██        | 194/938 [05:54<24:11,  1.95s/it]
Training 1/1 epoch (loss 1.7807):  21%|██        | 195/938 [05:54<22:00,  1.78s/it]
Training 1/1 epoch (loss 1.8042):  21%|██        | 195/938 [05:57<22:00,  1.78s/it]
Training 1/1 epoch (loss 1.8042):  21%|██        | 196/938 [05:57<24:23,  1.97s/it]
Training 1/1 epoch (loss 1.7991):  21%|██        | 196/938 [05:58<24:23,  1.97s/it]
Training 1/1 epoch (loss 1.7991):  21%|██        | 197/938 [05:58<23:20,  1.89s/it]
Training 1/1 epoch (loss 1.6624):  21%|██        | 197/938 [06:00<23:20,  1.89s/it]
Training 1/1 epoch (loss 1.6624):  21%|██        | 198/938 [06:00<21:42,  1.76s/it]
Training 1/1 epoch (loss 1.7888):  21%|██        | 198/938 [06:01<21:42,  1.76s/it]
Training 1/1 epoch (loss 1.7888):  21%|██        | 199/938 [06:01<20:42,  1.68s/it]
Training 1/1 epoch (loss 1.7322):  21%|██        | 199/938 [06:04<20:42,  1.68s/it]
Training 1/1 epoch (loss 1.7322):  21%|██▏       | 200/938 [06:04<24:15,  1.97s/it]
Training 1/1 epoch (loss 1.7721):  21%|██▏       | 200/938 [06:06<24:15,  1.97s/it]
Training 1/1 epoch (loss 1.7721):  21%|██▏       | 201/938 [06:06<23:43,  1.93s/it]
Training 1/1 epoch (loss 1.7515):  21%|██▏       | 201/938 [06:08<23:43,  1.93s/it]
Training 1/1 epoch (loss 1.7515):  22%|██▏       | 202/938 [06:08<23:01,  1.88s/it]
Training 1/1 epoch (loss 1.7759):  22%|██▏       | 202/938 [06:09<23:01,  1.88s/it]
Training 1/1 epoch (loss 1.7759):  22%|██▏       | 203/938 [06:09<22:19,  1.82s/it]
Training 1/1 epoch (loss 1.7998):  22%|██▏       | 203/938 [06:11<22:19,  1.82s/it]
Training 1/1 epoch (loss 1.7998):  22%|██▏       | 204/938 [06:11<21:07,  1.73s/it]
Training 1/1 epoch (loss 1.7702):  22%|██▏       | 204/938 [06:12<21:07,  1.73s/it]
Training 1/1 epoch (loss 1.7702):  22%|██▏       | 205/938 [06:12<18:07,  1.48s/it]
Training 1/1 epoch (loss 1.8666):  22%|██▏       | 205/938 [06:14<18:07,  1.48s/it]
Training 1/1 epoch (loss 1.8666):  22%|██▏       | 206/938 [06:14<20:20,  1.67s/it]
Training 1/1 epoch (loss 1.7383):  22%|██▏       | 206/938 [06:15<20:20,  1.67s/it]
Training 1/1 epoch (loss 1.7383):  22%|██▏       | 207/938 [06:15<19:09,  1.57s/it]
Training 1/1 epoch (loss 1.8412):  22%|██▏       | 207/938 [06:16<19:09,  1.57s/it]
Training 1/1 epoch (loss 1.8412):  22%|██▏       | 208/938 [06:16<17:06,  1.41s/it]
Training 1/1 epoch (loss 1.7470):  22%|██▏       | 208/938 [06:18<17:06,  1.41s/it]
Training 1/1 epoch (loss 1.7470):  22%|██▏       | 209/938 [06:18<19:37,  1.62s/it]
Training 1/1 epoch (loss 1.7736):  22%|██▏       | 209/938 [06:21<19:37,  1.62s/it]
Training 1/1 epoch (loss 1.7736):  22%|██▏       | 210/938 [06:21<22:35,  1.86s/it]
Training 1/1 epoch (loss 1.7648):  22%|██▏       | 210/938 [06:23<22:35,  1.86s/it]
Training 1/1 epoch (loss 1.7648):  22%|██▏       | 211/938 [06:23<22:38,  1.87s/it]
Training 1/1 epoch (loss 1.6771):  22%|██▏       | 211/938 [06:24<22:38,  1.87s/it]
Training 1/1 epoch (loss 1.6771):  23%|██▎       | 212/938 [06:24<20:21,  1.68s/it]
Training 1/1 epoch (loss 1.6524):  23%|██▎       | 212/938 [06:26<20:21,  1.68s/it]
Training 1/1 epoch (loss 1.6524):  23%|██▎       | 213/938 [06:26<22:25,  1.86s/it]
Training 1/1 epoch (loss 1.7403):  23%|██▎       | 213/938 [06:28<22:25,  1.86s/it]
Training 1/1 epoch (loss 1.7403):  23%|██▎       | 214/938 [06:28<21:28,  1.78s/it]
Training 1/1 epoch (loss 1.7755):  23%|██▎       | 214/938 [06:30<21:28,  1.78s/it]
Training 1/1 epoch (loss 1.7755):  23%|██▎       | 215/938 [06:30<21:53,  1.82s/it]
Training 1/1 epoch (loss 1.7163):  23%|██▎       | 215/938 [06:31<21:53,  1.82s/it]
Training 1/1 epoch (loss 1.7163):  23%|██▎       | 216/938 [06:31<21:52,  1.82s/it]
Training 1/1 epoch (loss 1.8887):  23%|██▎       | 216/938 [06:33<21:52,  1.82s/it]
Training 1/1 epoch (loss 1.8887):  23%|██▎       | 217/938 [06:33<19:41,  1.64s/it]
Training 1/1 epoch (loss 1.7766):  23%|██▎       | 217/938 [06:34<19:41,  1.64s/it]
Training 1/1 epoch (loss 1.7766):  23%|██▎       | 218/938 [06:34<19:24,  1.62s/it]
Training 1/1 epoch (loss 1.7341):  23%|██▎       | 218/938 [06:36<19:24,  1.62s/it]
Training 1/1 epoch (loss 1.7341):  23%|██▎       | 219/938 [06:36<20:23,  1.70s/it]
Training 1/1 epoch (loss 1.6251):  23%|██▎       | 219/938 [06:38<20:23,  1.70s/it]
Training 1/1 epoch (loss 1.6251):  23%|██▎       | 220/938 [06:38<19:11,  1.60s/it]
Training 1/1 epoch (loss 1.7924):  23%|██▎       | 220/938 [06:40<19:11,  1.60s/it]
Training 1/1 epoch (loss 1.7924):  24%|██▎       | 221/938 [06:40<20:59,  1.76s/it]
Training 1/1 epoch (loss 1.7637):  24%|██▎       | 221/938 [06:41<20:59,  1.76s/it]
Training 1/1 epoch (loss 1.7637):  24%|██▎       | 222/938 [06:41<20:12,  1.69s/it]
Training 1/1 epoch (loss 1.8178):  24%|██▎       | 222/938 [06:43<20:12,  1.69s/it]
Training 1/1 epoch (loss 1.8178):  24%|██▍       | 223/938 [06:43<19:52,  1.67s/it]
Training 1/1 epoch (loss 1.8062):  24%|██▍       | 223/938 [06:45<19:52,  1.67s/it]
Training 1/1 epoch (loss 1.8062):  24%|██▍       | 224/938 [06:45<20:59,  1.76s/it]
Training 1/1 epoch (loss 1.8134):  24%|██▍       | 224/938 [06:46<20:59,  1.76s/it]
Training 1/1 epoch (loss 1.8134):  24%|██▍       | 225/938 [06:46<20:37,  1.74s/it]
Training 1/1 epoch (loss 1.8564):  24%|██▍       | 225/938 [06:49<20:37,  1.74s/it]
Training 1/1 epoch (loss 1.8564):  24%|██▍       | 226/938 [06:49<23:00,  1.94s/it]
Training 1/1 epoch (loss 1.7691):  24%|██▍       | 226/938 [06:51<23:00,  1.94s/it]
Training 1/1 epoch (loss 1.7691):  24%|██▍       | 227/938 [06:51<22:41,  1.92s/it]
Training 1/1 epoch (loss 1.7331):  24%|██▍       | 227/938 [06:52<22:41,  1.92s/it]
Training 1/1 epoch (loss 1.7331):  24%|██▍       | 228/938 [06:52<21:54,  1.85s/it]
Training 1/1 epoch (loss 1.8079):  24%|██▍       | 228/938 [06:54<21:54,  1.85s/it]
Training 1/1 epoch (loss 1.8079):  24%|██▍       | 229/938 [06:54<20:38,  1.75s/it]
Training 1/1 epoch (loss 1.7832):  24%|██▍       | 229/938 [06:56<20:38,  1.75s/it]
Training 1/1 epoch (loss 1.7832):  25%|██▍       | 230/938 [06:56<23:04,  1.95s/it]
Training 1/1 epoch (loss 1.7528):  25%|██▍       | 230/938 [06:58<23:04,  1.95s/it]
Training 1/1 epoch (loss 1.7528):  25%|██▍       | 231/938 [06:58<21:09,  1.80s/it]
Training 1/1 epoch (loss 1.7578):  25%|██▍       | 231/938 [07:00<21:09,  1.80s/it]
Training 1/1 epoch (loss 1.7578):  25%|██▍       | 232/938 [07:00<23:24,  1.99s/it]
Training 1/1 epoch (loss 1.7332):  25%|██▍       | 232/938 [07:02<23:24,  1.99s/it]
Training 1/1 epoch (loss 1.7332):  25%|██▍       | 233/938 [07:02<23:20,  1.99s/it]
Training 1/1 epoch (loss 1.7053):  25%|██▍       | 233/938 [07:03<23:20,  1.99s/it]
Training 1/1 epoch (loss 1.7053):  25%|██▍       | 234/938 [07:03<18:46,  1.60s/it]
Training 1/1 epoch (loss 1.7188):  25%|██▍       | 234/938 [07:05<18:46,  1.60s/it]
Training 1/1 epoch (loss 1.7188):  25%|██▌       | 235/938 [07:05<20:40,  1.76s/it]
Training 1/1 epoch (loss 1.6840):  25%|██▌       | 235/938 [07:07<20:40,  1.76s/it]
Training 1/1 epoch (loss 1.6840):  25%|██▌       | 236/938 [07:07<19:46,  1.69s/it]
Training 1/1 epoch (loss 1.8289):  25%|██▌       | 236/938 [07:08<19:46,  1.69s/it]
Training 1/1 epoch (loss 1.8289):  25%|██▌       | 237/938 [07:08<18:47,  1.61s/it]
Training 1/1 epoch (loss 1.6273):  25%|██▌       | 237/938 [07:09<18:47,  1.61s/it]
Training 1/1 epoch (loss 1.6273):  25%|██▌       | 238/938 [07:09<18:02,  1.55s/it]
Training 1/1 epoch (loss 1.7713):  25%|██▌       | 238/938 [07:11<18:02,  1.55s/it]
Training 1/1 epoch (loss 1.7713):  25%|██▌       | 239/938 [07:11<18:38,  1.60s/it]
Training 1/1 epoch (loss 1.7601):  25%|██▌       | 239/938 [07:13<18:38,  1.60s/it]
Training 1/1 epoch (loss 1.7601):  26%|██▌       | 240/938 [07:13<20:13,  1.74s/it]
Training 1/1 epoch (loss 1.8255):  26%|██▌       | 240/938 [07:14<20:13,  1.74s/it]
Training 1/1 epoch (loss 1.8255):  26%|██▌       | 241/938 [07:14<18:32,  1.60s/it]
Training 1/1 epoch (loss 1.7985):  26%|██▌       | 241/938 [07:15<18:32,  1.60s/it]
Training 1/1 epoch (loss 1.7985):  26%|██▌       | 242/938 [07:15<16:09,  1.39s/it]
Training 1/1 epoch (loss 1.7849):  26%|██▌       | 242/938 [07:18<16:09,  1.39s/it]
Training 1/1 epoch (loss 1.7849):  26%|██▌       | 243/938 [07:18<19:31,  1.69s/it]
Training 1/1 epoch (loss 1.6514):  26%|██▌       | 243/938 [07:19<19:31,  1.69s/it]
Training 1/1 epoch (loss 1.6514):  26%|██▌       | 244/938 [07:19<19:15,  1.66s/it]
Training 1/1 epoch (loss 1.6774):  26%|██▌       | 244/938 [07:21<19:15,  1.66s/it]
Training 1/1 epoch (loss 1.6774):  26%|██▌       | 245/938 [07:21<18:41,  1.62s/it]
Training 1/1 epoch (loss 1.7654):  26%|██▌       | 245/938 [07:23<18:41,  1.62s/it]
Training 1/1 epoch (loss 1.7654):  26%|██▌       | 246/938 [07:23<19:44,  1.71s/it]
Training 1/1 epoch (loss 1.7325):  26%|██▌       | 246/938 [07:24<19:44,  1.71s/it]
Training 1/1 epoch (loss 1.7325):  26%|██▋       | 247/938 [07:24<19:24,  1.69s/it]
Training 1/1 epoch (loss 1.6865):  26%|██▋       | 247/938 [07:27<19:24,  1.69s/it]
Training 1/1 epoch (loss 1.6865):  26%|██▋       | 248/938 [07:27<23:04,  2.01s/it]
Training 1/1 epoch (loss 1.7836):  26%|██▋       | 248/938 [07:29<23:04,  2.01s/it]
Training 1/1 epoch (loss 1.7836):  27%|██▋       | 249/938 [07:29<23:19,  2.03s/it]
Training 1/1 epoch (loss 1.6111):  27%|██▋       | 249/938 [07:32<23:19,  2.03s/it]
Training 1/1 epoch (loss 1.6111):  27%|██▋       | 250/938 [07:32<24:03,  2.10s/it]
Training 1/1 epoch (loss 1.8499):  27%|██▋       | 250/938 [07:33<24:03,  2.10s/it]
Training 1/1 epoch (loss 1.8499):  27%|██▋       | 251/938 [07:33<23:10,  2.02s/it]
Training 1/1 epoch (loss 1.7608):  27%|██▋       | 251/938 [07:35<23:10,  2.02s/it]
Training 1/1 epoch (loss 1.7608):  27%|██▋       | 252/938 [07:35<22:00,  1.93s/it]
Training 1/1 epoch (loss 1.7239):  27%|██▋       | 252/938 [07:36<22:00,  1.93s/it]
Training 1/1 epoch (loss 1.7239):  27%|██▋       | 253/938 [07:36<19:54,  1.74s/it]
Training 1/1 epoch (loss 1.7528):  27%|██▋       | 253/938 [07:38<19:54,  1.74s/it]
Training 1/1 epoch (loss 1.7528):  27%|██▋       | 254/938 [07:38<18:08,  1.59s/it]
Training 1/1 epoch (loss 1.7221):  27%|██▋       | 254/938 [07:39<18:08,  1.59s/it]
Training 1/1 epoch (loss 1.7221):  27%|██▋       | 255/938 [07:39<17:01,  1.50s/it]
Training 1/1 epoch (loss 1.6485):  27%|██▋       | 255/938 [07:40<17:01,  1.50s/it]
Training 1/1 epoch (loss 1.6485):  27%|██▋       | 256/938 [07:40<16:13,  1.43s/it]
Training 1/1 epoch (loss 1.7254):  27%|██▋       | 256/938 [07:42<16:13,  1.43s/it]
Training 1/1 epoch (loss 1.7254):  27%|██▋       | 257/938 [07:42<15:54,  1.40s/it]
Training 1/1 epoch (loss 1.7603):  27%|██▋       | 257/938 [07:43<15:54,  1.40s/it]
Training 1/1 epoch (loss 1.7603):  28%|██▊       | 258/938 [07:43<15:33,  1.37s/it]
Training 1/1 epoch (loss 1.7466):  28%|██▊       | 258/938 [07:44<15:33,  1.37s/it]
Training 1/1 epoch (loss 1.7466):  28%|██▊       | 259/938 [07:44<15:32,  1.37s/it]
Training 1/1 epoch (loss 1.7350):  28%|██▊       | 259/938 [07:46<15:32,  1.37s/it]
Training 1/1 epoch (loss 1.7350):  28%|██▊       | 260/938 [07:46<17:32,  1.55s/it]
Training 1/1 epoch (loss 1.6984):  28%|██▊       | 260/938 [07:47<17:32,  1.55s/it]
Training 1/1 epoch (loss 1.6984):  28%|██▊       | 261/938 [07:47<16:36,  1.47s/it]
Training 1/1 epoch (loss 1.7184):  28%|██▊       | 261/938 [07:49<16:36,  1.47s/it]
Training 1/1 epoch (loss 1.7184):  28%|██▊       | 262/938 [07:49<15:39,  1.39s/it]
Training 1/1 epoch (loss 1.5950):  28%|██▊       | 262/938 [07:51<15:39,  1.39s/it]
Training 1/1 epoch (loss 1.5950):  28%|██▊       | 263/938 [07:51<19:22,  1.72s/it]
Training 1/1 epoch (loss 1.8326):  28%|██▊       | 263/938 [07:53<19:22,  1.72s/it]
Training 1/1 epoch (loss 1.8326):  28%|██▊       | 264/938 [07:53<18:52,  1.68s/it]
Training 1/1 epoch (loss 1.7167):  28%|██▊       | 264/938 [07:55<18:52,  1.68s/it]
Training 1/1 epoch (loss 1.7167):  28%|██▊       | 265/938 [07:55<19:14,  1.72s/it]
Training 1/1 epoch (loss 1.7732):  28%|██▊       | 265/938 [07:56<19:14,  1.72s/it]
Training 1/1 epoch (loss 1.7732):  28%|██▊       | 266/938 [07:56<18:55,  1.69s/it]
Training 1/1 epoch (loss 1.6599):  28%|██▊       | 266/938 [07:58<18:55,  1.69s/it]
Training 1/1 epoch (loss 1.6599):  28%|██▊       | 267/938 [07:58<19:06,  1.71s/it]
Training 1/1 epoch (loss 1.6703):  28%|██▊       | 267/938 [08:00<19:06,  1.71s/it]
Training 1/1 epoch (loss 1.6703):  29%|██▊       | 268/938 [08:00<20:31,  1.84s/it]
Training 1/1 epoch (loss 1.6629):  29%|██▊       | 268/938 [08:02<20:31,  1.84s/it]
Training 1/1 epoch (loss 1.6629):  29%|██▊       | 269/938 [08:02<20:10,  1.81s/it]
Training 1/1 epoch (loss 1.7434):  29%|██▊       | 269/938 [08:03<20:10,  1.81s/it]
Training 1/1 epoch (loss 1.7434):  29%|██▉       | 270/938 [08:03<19:23,  1.74s/it]
Training 1/1 epoch (loss 1.8048):  29%|██▉       | 270/938 [08:05<19:23,  1.74s/it]
Training 1/1 epoch (loss 1.8048):  29%|██▉       | 271/938 [08:05<19:08,  1.72s/it]
Training 1/1 epoch (loss 1.6185):  29%|██▉       | 271/938 [08:07<19:08,  1.72s/it]
Training 1/1 epoch (loss 1.6185):  29%|██▉       | 272/938 [08:07<19:21,  1.74s/it]
Training 1/1 epoch (loss 1.7674):  29%|██▉       | 272/938 [08:09<19:21,  1.74s/it]
Training 1/1 epoch (loss 1.7674):  29%|██▉       | 273/938 [08:09<21:43,  1.96s/it]
Training 1/1 epoch (loss 1.7942):  29%|██▉       | 273/938 [08:11<21:43,  1.96s/it]
Training 1/1 epoch (loss 1.7942):  29%|██▉       | 274/938 [08:11<21:08,  1.91s/it]
Training 1/1 epoch (loss 1.8416):  29%|██▉       | 274/938 [08:13<21:08,  1.91s/it]
Training 1/1 epoch (loss 1.8416):  29%|██▉       | 275/938 [08:13<20:47,  1.88s/it]
Training 1/1 epoch (loss 1.7610):  29%|██▉       | 275/938 [08:15<20:47,  1.88s/it]
Training 1/1 epoch (loss 1.7610):  29%|██▉       | 276/938 [08:15<19:48,  1.79s/it]
Training 1/1 epoch (loss 1.8217):  29%|██▉       | 276/938 [08:17<19:48,  1.79s/it]
Training 1/1 epoch (loss 1.8217):  30%|██▉       | 277/938 [08:17<20:47,  1.89s/it]
Training 1/1 epoch (loss 1.7257):  30%|██▉       | 277/938 [08:18<20:47,  1.89s/it]
Training 1/1 epoch (loss 1.7257):  30%|██▉       | 278/938 [08:18<17:49,  1.62s/it]
Training 1/1 epoch (loss 1.7619):  30%|██▉       | 278/938 [08:19<17:49,  1.62s/it]
Training 1/1 epoch (loss 1.7619):  30%|██▉       | 279/938 [08:19<17:08,  1.56s/it]
Training 1/1 epoch (loss 1.6507):  30%|██▉       | 279/938 [08:21<17:08,  1.56s/it]
Training 1/1 epoch (loss 1.6507):  30%|██▉       | 280/938 [08:21<18:06,  1.65s/it]
Training 1/1 epoch (loss 1.7203):  30%|██▉       | 280/938 [08:22<18:06,  1.65s/it]
Training 1/1 epoch (loss 1.7203):  30%|██▉       | 281/938 [08:22<15:58,  1.46s/it]
Training 1/1 epoch (loss 1.7024):  30%|██▉       | 281/938 [08:23<15:58,  1.46s/it]
Training 1/1 epoch (loss 1.7024):  30%|███       | 282/938 [08:23<15:10,  1.39s/it]
Training 1/1 epoch (loss 1.8335):  30%|███       | 282/938 [08:25<15:10,  1.39s/it]
Training 1/1 epoch (loss 1.8335):  30%|███       | 283/938 [08:25<16:47,  1.54s/it]
Training 1/1 epoch (loss 1.6484):  30%|███       | 283/938 [08:26<16:47,  1.54s/it]
Training 1/1 epoch (loss 1.6484):  30%|███       | 284/938 [08:26<16:31,  1.52s/it]
Training 1/1 epoch (loss 1.6321):  30%|███       | 284/938 [08:28<16:31,  1.52s/it]
Training 1/1 epoch (loss 1.6321):  30%|███       | 285/938 [08:28<16:20,  1.50s/it]
Training 1/1 epoch (loss 1.7942):  30%|███       | 285/938 [08:29<16:20,  1.50s/it]
Training 1/1 epoch (loss 1.7942):  30%|███       | 286/938 [08:29<15:29,  1.43s/it]
Training 1/1 epoch (loss 1.6279):  30%|███       | 286/938 [08:32<15:29,  1.43s/it]
Training 1/1 epoch (loss 1.6279):  31%|███       | 287/938 [08:32<18:47,  1.73s/it]
Training 1/1 epoch (loss 1.8505):  31%|███       | 287/938 [08:33<18:47,  1.73s/it]
Training 1/1 epoch (loss 1.8505):  31%|███       | 288/938 [08:33<18:16,  1.69s/it]
Training 1/1 epoch (loss 1.7956):  31%|███       | 288/938 [08:35<18:16,  1.69s/it]
Training 1/1 epoch (loss 1.7956):  31%|███       | 289/938 [08:35<20:03,  1.85s/it]
Training 1/1 epoch (loss 1.6651):  31%|███       | 289/938 [08:37<20:03,  1.85s/it]
Training 1/1 epoch (loss 1.6651):  31%|███       | 290/938 [08:37<20:09,  1.87s/it]
Training 1/1 epoch (loss 1.8385):  31%|███       | 290/938 [08:39<20:09,  1.87s/it]
Training 1/1 epoch (loss 1.8385):  31%|███       | 291/938 [08:39<19:28,  1.81s/it]
Training 1/1 epoch (loss 1.6568):  31%|███       | 291/938 [08:40<19:28,  1.81s/it]
Training 1/1 epoch (loss 1.6568):  31%|███       | 292/938 [08:40<17:27,  1.62s/it]
Training 1/1 epoch (loss 1.7107):  31%|███       | 292/938 [08:42<17:27,  1.62s/it]
Training 1/1 epoch (loss 1.7107):  31%|███       | 293/938 [08:42<16:22,  1.52s/it]
Training 1/1 epoch (loss 1.7056):  31%|███       | 293/938 [08:43<16:22,  1.52s/it]
Training 1/1 epoch (loss 1.7056):  31%|███▏      | 294/938 [08:43<17:08,  1.60s/it]
Training 1/1 epoch (loss 1.5651):  31%|███▏      | 294/938 [08:45<17:08,  1.60s/it]
Training 1/1 epoch (loss 1.5651):  31%|███▏      | 295/938 [08:45<16:14,  1.52s/it]
Training 1/1 epoch (loss 1.6862):  31%|███▏      | 295/938 [08:47<16:14,  1.52s/it]
Training 1/1 epoch (loss 1.6862):  32%|███▏      | 296/938 [08:47<17:53,  1.67s/it]
Training 1/1 epoch (loss 1.7094):  32%|███▏      | 296/938 [08:49<17:53,  1.67s/it]
Training 1/1 epoch (loss 1.7094):  32%|███▏      | 297/938 [08:49<20:01,  1.87s/it]
Training 1/1 epoch (loss 1.7411):  32%|███▏      | 297/938 [08:51<20:01,  1.87s/it]
Training 1/1 epoch (loss 1.7411):  32%|███▏      | 298/938 [08:51<20:20,  1.91s/it]
Training 1/1 epoch (loss 1.7921):  32%|███▏      | 298/938 [08:52<20:20,  1.91s/it]
Training 1/1 epoch (loss 1.7921):  32%|███▏      | 299/938 [08:52<18:57,  1.78s/it]
Training 1/1 epoch (loss 1.7033):  32%|███▏      | 299/938 [08:53<18:57,  1.78s/it]
Training 1/1 epoch (loss 1.7033):  32%|███▏      | 300/938 [08:53<16:25,  1.54s/it]
Training 1/1 epoch (loss 1.7545):  32%|███▏      | 300/938 [08:56<16:25,  1.54s/it]
Training 1/1 epoch (loss 1.7545):  32%|███▏      | 301/938 [08:56<18:06,  1.71s/it]
Training 1/1 epoch (loss 1.7601):  32%|███▏      | 301/938 [08:57<18:06,  1.71s/it]
Training 1/1 epoch (loss 1.7601):  32%|███▏      | 302/938 [08:57<18:45,  1.77s/it]
Training 1/1 epoch (loss 1.6640):  32%|███▏      | 302/938 [09:00<18:45,  1.77s/it]
Training 1/1 epoch (loss 1.6640):  32%|███▏      | 303/938 [09:00<20:08,  1.90s/it]
Training 1/1 epoch (loss 1.8712):  32%|███▏      | 303/938 [09:01<20:08,  1.90s/it]
Training 1/1 epoch (loss 1.8712):  32%|███▏      | 304/938 [09:01<18:29,  1.75s/it]
Training 1/1 epoch (loss 1.6483):  32%|███▏      | 304/938 [09:03<18:29,  1.75s/it]
Training 1/1 epoch (loss 1.6483):  33%|███▎      | 305/938 [09:03<18:24,  1.74s/it]
Training 1/1 epoch (loss 1.7348):  33%|███▎      | 305/938 [09:05<18:24,  1.74s/it]
Training 1/1 epoch (loss 1.7348):  33%|███▎      | 306/938 [09:05<19:47,  1.88s/it]
Training 1/1 epoch (loss 1.7174):  33%|███▎      | 306/938 [09:07<19:47,  1.88s/it]
Training 1/1 epoch (loss 1.7174):  33%|███▎      | 307/938 [09:07<21:40,  2.06s/it]
Training 1/1 epoch (loss 1.6787):  33%|███▎      | 307/938 [09:09<21:40,  2.06s/it]
Training 1/1 epoch (loss 1.6787):  33%|███▎      | 308/938 [09:09<19:43,  1.88s/it]
Training 1/1 epoch (loss 1.5849):  33%|███▎      | 308/938 [09:11<19:43,  1.88s/it]
Training 1/1 epoch (loss 1.5849):  33%|███▎      | 309/938 [09:11<19:25,  1.85s/it]
Training 1/1 epoch (loss 1.8015):  33%|███▎      | 309/938 [09:12<19:25,  1.85s/it]
Training 1/1 epoch (loss 1.8015):  33%|███▎      | 310/938 [09:12<18:41,  1.79s/it]
Training 1/1 epoch (loss 1.5719):  33%|███▎      | 310/938 [09:15<18:41,  1.79s/it]
Training 1/1 epoch (loss 1.5719):  33%|███▎      | 311/938 [09:15<20:34,  1.97s/it]
Training 1/1 epoch (loss 1.6733):  33%|███▎      | 311/938 [09:16<20:34,  1.97s/it]
Training 1/1 epoch (loss 1.6733):  33%|███▎      | 312/938 [09:16<18:14,  1.75s/it]
Training 1/1 epoch (loss 1.9003):  33%|███▎      | 312/938 [09:18<18:14,  1.75s/it]
Training 1/1 epoch (loss 1.9003):  33%|███▎      | 313/938 [09:18<20:10,  1.94s/it]
Training 1/1 epoch (loss 1.8025):  33%|███▎      | 313/938 [09:19<20:10,  1.94s/it]
Training 1/1 epoch (loss 1.8025):  33%|███▎      | 314/938 [09:19<17:26,  1.68s/it]
Training 1/1 epoch (loss 1.6309):  33%|███▎      | 314/938 [09:21<17:26,  1.68s/it]
Training 1/1 epoch (loss 1.6309):  34%|███▎      | 315/938 [09:21<16:58,  1.63s/it]
Training 1/1 epoch (loss 1.7528):  34%|███▎      | 315/938 [09:23<16:58,  1.63s/it]
Training 1/1 epoch (loss 1.7528):  34%|███▎      | 316/938 [09:23<19:21,  1.87s/it]
Training 1/1 epoch (loss 1.6897):  34%|███▎      | 316/938 [09:25<19:21,  1.87s/it]
Training 1/1 epoch (loss 1.6897):  34%|███▍      | 317/938 [09:25<17:35,  1.70s/it]
Training 1/1 epoch (loss 1.7932):  34%|███▍      | 317/938 [09:26<17:35,  1.70s/it]
Training 1/1 epoch (loss 1.7932):  34%|███▍      | 318/938 [09:26<15:36,  1.51s/it]
Training 1/1 epoch (loss 1.6090):  34%|███▍      | 318/938 [09:28<15:36,  1.51s/it]
Training 1/1 epoch (loss 1.6090):  34%|███▍      | 319/938 [09:28<17:31,  1.70s/it]
Training 1/1 epoch (loss 1.5010):  34%|███▍      | 319/938 [09:31<17:31,  1.70s/it]
Training 1/1 epoch (loss 1.5010):  34%|███▍      | 320/938 [09:31<20:16,  1.97s/it]
Training 1/1 epoch (loss 1.6367):  34%|███▍      | 320/938 [09:33<20:16,  1.97s/it]
Training 1/1 epoch (loss 1.6367):  34%|███▍      | 321/938 [09:33<20:31,  2.00s/it]
Training 1/1 epoch (loss 1.7423):  34%|███▍      | 321/938 [09:35<20:31,  2.00s/it]
Training 1/1 epoch (loss 1.7423):  34%|███▍      | 322/938 [09:35<20:31,  2.00s/it]
Training 1/1 epoch (loss 1.8117):  34%|███▍      | 322/938 [09:36<20:31,  2.00s/it]
Training 1/1 epoch (loss 1.8117):  34%|███▍      | 323/938 [09:36<19:03,  1.86s/it]
Training 1/1 epoch (loss 1.7277):  34%|███▍      | 323/938 [09:38<19:03,  1.86s/it]
Training 1/1 epoch (loss 1.7277):  35%|███▍      | 324/938 [09:38<17:40,  1.73s/it]
Training 1/1 epoch (loss 1.6743):  35%|███▍      | 324/938 [09:40<17:40,  1.73s/it]
Training 1/1 epoch (loss 1.6743):  35%|███▍      | 325/938 [09:40<19:57,  1.95s/it]
Training 1/1 epoch (loss 1.7855):  35%|███▍      | 325/938 [09:42<19:57,  1.95s/it]
Training 1/1 epoch (loss 1.7855):  35%|███▍      | 326/938 [09:42<19:50,  1.95s/it]
Training 1/1 epoch (loss 1.6546):  35%|███▍      | 326/938 [09:43<19:50,  1.95s/it]
Training 1/1 epoch (loss 1.6546):  35%|███▍      | 327/938 [09:43<18:05,  1.78s/it]
Training 1/1 epoch (loss 1.6802):  35%|███▍      | 327/938 [09:45<18:05,  1.78s/it]
Training 1/1 epoch (loss 1.6802):  35%|███▍      | 328/938 [09:45<18:16,  1.80s/it]
Training 1/1 epoch (loss 1.6021):  35%|███▍      | 328/938 [09:47<18:16,  1.80s/it]
Training 1/1 epoch (loss 1.6021):  35%|███▌      | 329/938 [09:47<17:04,  1.68s/it]
Training 1/1 epoch (loss 1.7492):  35%|███▌      | 329/938 [09:48<17:04,  1.68s/it]
Training 1/1 epoch (loss 1.7492):  35%|███▌      | 330/938 [09:48<17:09,  1.69s/it]
Training 1/1 epoch (loss 1.7105):  35%|███▌      | 330/938 [09:50<17:09,  1.69s/it]
Training 1/1 epoch (loss 1.7105):  35%|███▌      | 331/938 [09:50<16:30,  1.63s/it]
Training 1/1 epoch (loss 1.7754):  35%|███▌      | 331/938 [09:51<16:30,  1.63s/it]
Training 1/1 epoch (loss 1.7754):  35%|███▌      | 332/938 [09:51<16:38,  1.65s/it]
Training 1/1 epoch (loss 1.7293):  35%|███▌      | 332/938 [09:53<16:38,  1.65s/it]
Training 1/1 epoch (loss 1.7293):  36%|███▌      | 333/938 [09:53<17:05,  1.69s/it]
Training 1/1 epoch (loss 1.6377):  36%|███▌      | 333/938 [09:55<17:05,  1.69s/it]
Training 1/1 epoch (loss 1.6377):  36%|███▌      | 334/938 [09:55<16:04,  1.60s/it]
Training 1/1 epoch (loss 1.7499):  36%|███▌      | 334/938 [09:56<16:04,  1.60s/it]
Training 1/1 epoch (loss 1.7499):  36%|███▌      | 335/938 [09:56<15:10,  1.51s/it]
Training 1/1 epoch (loss 1.6224):  36%|███▌      | 335/938 [09:58<15:10,  1.51s/it]
Training 1/1 epoch (loss 1.6224):  36%|███▌      | 336/938 [09:58<17:57,  1.79s/it]
Training 1/1 epoch (loss 1.7361):  36%|███▌      | 336/938 [10:00<17:57,  1.79s/it]
Training 1/1 epoch (loss 1.7361):  36%|███▌      | 337/938 [10:00<16:02,  1.60s/it]
Training 1/1 epoch (loss 1.7065):  36%|███▌      | 337/938 [10:01<16:02,  1.60s/it]
Training 1/1 epoch (loss 1.7065):  36%|███▌      | 338/938 [10:01<14:49,  1.48s/it]
Training 1/1 epoch (loss 1.7454):  36%|███▌      | 338/938 [10:02<14:49,  1.48s/it]
Training 1/1 epoch (loss 1.7454):  36%|███▌      | 339/938 [10:02<14:05,  1.41s/it]
Training 1/1 epoch (loss 1.6770):  36%|███▌      | 339/938 [10:03<14:05,  1.41s/it]
Training 1/1 epoch (loss 1.6770):  36%|███▌      | 340/938 [10:03<13:47,  1.38s/it]
Training 1/1 epoch (loss 1.7111):  36%|███▌      | 340/938 [10:05<13:47,  1.38s/it]
Training 1/1 epoch (loss 1.7111):  36%|███▋      | 341/938 [10:05<15:22,  1.54s/it]
Training 1/1 epoch (loss 1.6517):  36%|███▋      | 341/938 [10:07<15:22,  1.54s/it]
Training 1/1 epoch (loss 1.6517):  36%|███▋      | 342/938 [10:07<16:49,  1.69s/it]
Training 1/1 epoch (loss 1.8130):  36%|███▋      | 342/938 [10:09<16:49,  1.69s/it]
Training 1/1 epoch (loss 1.8130):  37%|███▋      | 343/938 [10:09<17:10,  1.73s/it]
Training 1/1 epoch (loss 1.7171):  37%|███▋      | 343/938 [10:10<17:10,  1.73s/it]
Training 1/1 epoch (loss 1.7171):  37%|███▋      | 344/938 [10:10<16:06,  1.63s/it]
Training 1/1 epoch (loss 1.6210):  37%|███▋      | 344/938 [10:12<16:06,  1.63s/it]
Training 1/1 epoch (loss 1.6210):  37%|███▋      | 345/938 [10:12<15:38,  1.58s/it]
Training 1/1 epoch (loss 1.6215):  37%|███▋      | 345/938 [10:13<15:38,  1.58s/it]
Training 1/1 epoch (loss 1.6215):  37%|███▋      | 346/938 [10:13<15:06,  1.53s/it]
Training 1/1 epoch (loss 1.6913):  37%|███▋      | 346/938 [10:15<15:06,  1.53s/it]
Training 1/1 epoch (loss 1.6913):  37%|███▋      | 347/938 [10:15<14:08,  1.44s/it]
Training 1/1 epoch (loss 1.5882):  37%|███▋      | 347/938 [10:17<14:08,  1.44s/it]
Training 1/1 epoch (loss 1.5882):  37%|███▋      | 348/938 [10:17<16:18,  1.66s/it]
Training 1/1 epoch (loss 1.7623):  37%|███▋      | 348/938 [10:18<16:18,  1.66s/it]
Training 1/1 epoch (loss 1.7623):  37%|███▋      | 349/938 [10:18<16:20,  1.66s/it]
Training 1/1 epoch (loss 1.7272):  37%|███▋      | 349/938 [10:20<16:20,  1.66s/it]
Training 1/1 epoch (loss 1.7272):  37%|███▋      | 350/938 [10:20<16:49,  1.72s/it]
Training 1/1 epoch (loss 1.7273):  37%|███▋      | 350/938 [10:22<16:49,  1.72s/it]
Training 1/1 epoch (loss 1.7273):  37%|███▋      | 351/938 [10:22<17:10,  1.76s/it]
Training 1/1 epoch (loss 1.6701):  37%|███▋      | 351/938 [10:24<17:10,  1.76s/it]
Training 1/1 epoch (loss 1.6701):  38%|███▊      | 352/938 [10:24<17:44,  1.82s/it]
Training 1/1 epoch (loss 1.8173):  38%|███▊      | 352/938 [10:26<17:44,  1.82s/it]
Training 1/1 epoch (loss 1.8173):  38%|███▊      | 353/938 [10:26<17:04,  1.75s/it]
Training 1/1 epoch (loss 1.7153):  38%|███▊      | 353/938 [10:27<17:04,  1.75s/it]
Training 1/1 epoch (loss 1.7153):  38%|███▊      | 354/938 [10:27<15:30,  1.59s/it]
Training 1/1 epoch (loss 1.7138):  38%|███▊      | 354/938 [10:29<15:30,  1.59s/it]
Training 1/1 epoch (loss 1.7138):  38%|███▊      | 355/938 [10:29<17:44,  1.83s/it]
Training 1/1 epoch (loss 1.6381):  38%|███▊      | 355/938 [10:31<17:44,  1.83s/it]
Training 1/1 epoch (loss 1.6381):  38%|███▊      | 356/938 [10:31<17:32,  1.81s/it]
Training 1/1 epoch (loss 1.6811):  38%|███▊      | 356/938 [10:32<17:32,  1.81s/it]
Training 1/1 epoch (loss 1.6811):  38%|███▊      | 357/938 [10:32<16:13,  1.68s/it]
Training 1/1 epoch (loss 1.6438):  38%|███▊      | 357/938 [10:34<16:13,  1.68s/it]
Training 1/1 epoch (loss 1.6438):  38%|███▊      | 358/938 [10:34<16:52,  1.75s/it]
Training 1/1 epoch (loss 1.6632):  38%|███▊      | 358/938 [10:37<16:52,  1.75s/it]
Training 1/1 epoch (loss 1.6632):  38%|███▊      | 359/938 [10:37<18:37,  1.93s/it]
Training 1/1 epoch (loss 1.7771):  38%|███▊      | 359/938 [10:38<18:37,  1.93s/it]
Training 1/1 epoch (loss 1.7771):  38%|███▊      | 360/938 [10:38<16:56,  1.76s/it]
Training 1/1 epoch (loss 1.8426):  38%|███▊      | 360/938 [10:40<16:56,  1.76s/it]
Training 1/1 epoch (loss 1.8426):  38%|███▊      | 361/938 [10:40<16:19,  1.70s/it]
Training 1/1 epoch (loss 1.7158):  38%|███▊      | 361/938 [10:41<16:19,  1.70s/it]
Training 1/1 epoch (loss 1.7158):  39%|███▊      | 362/938 [10:41<16:43,  1.74s/it]
Training 1/1 epoch (loss 1.7051):  39%|███▊      | 362/938 [10:43<16:43,  1.74s/it]
Training 1/1 epoch (loss 1.7051):  39%|███▊      | 363/938 [10:43<16:52,  1.76s/it]
Training 1/1 epoch (loss 1.7731):  39%|███▊      | 363/938 [10:45<16:52,  1.76s/it]
Training 1/1 epoch (loss 1.7731):  39%|███▉      | 364/938 [10:45<15:41,  1.64s/it]
Training 1/1 epoch (loss 1.6658):  39%|███▉      | 364/938 [10:46<15:41,  1.64s/it]
Training 1/1 epoch (loss 1.6658):  39%|███▉      | 365/938 [10:46<14:25,  1.51s/it]
Training 1/1 epoch (loss 1.6779):  39%|███▉      | 365/938 [10:47<14:25,  1.51s/it]
Training 1/1 epoch (loss 1.6779):  39%|███▉      | 366/938 [10:47<14:46,  1.55s/it]
Training 1/1 epoch (loss 1.7286):  39%|███▉      | 366/938 [10:48<14:46,  1.55s/it]
Training 1/1 epoch (loss 1.7286):  39%|███▉      | 367/938 [10:48<12:56,  1.36s/it]
Training 1/1 epoch (loss 1.8218):  39%|███▉      | 367/938 [10:50<12:56,  1.36s/it]
Training 1/1 epoch (loss 1.8218):  39%|███▉      | 368/938 [10:50<14:42,  1.55s/it]
Training 1/1 epoch (loss 1.6711):  39%|███▉      | 368/938 [10:52<14:42,  1.55s/it]
Training 1/1 epoch (loss 1.6711):  39%|███▉      | 369/938 [10:52<14:02,  1.48s/it]
Training 1/1 epoch (loss 1.7881):  39%|███▉      | 369/938 [10:53<14:02,  1.48s/it]
Training 1/1 epoch (loss 1.7881):  39%|███▉      | 370/938 [10:53<14:34,  1.54s/it]
Training 1/1 epoch (loss 1.6735):  39%|███▉      | 370/938 [10:55<14:34,  1.54s/it]
Training 1/1 epoch (loss 1.6735):  40%|███▉      | 371/938 [10:55<14:15,  1.51s/it]
Training 1/1 epoch (loss 1.6951):  40%|███▉      | 371/938 [10:57<14:15,  1.51s/it]
Training 1/1 epoch (loss 1.6951):  40%|███▉      | 372/938 [10:57<14:55,  1.58s/it]
Training 1/1 epoch (loss 1.6436):  40%|███▉      | 372/938 [10:58<14:55,  1.58s/it]
Training 1/1 epoch (loss 1.6436):  40%|███▉      | 373/938 [10:58<14:12,  1.51s/it]
Training 1/1 epoch (loss 1.5352):  40%|███▉      | 373/938 [11:00<14:12,  1.51s/it]
Training 1/1 epoch (loss 1.5352):  40%|███▉      | 374/938 [11:00<15:08,  1.61s/it]
Training 1/1 epoch (loss 1.7565):  40%|███▉      | 374/938 [11:02<15:08,  1.61s/it]
Training 1/1 epoch (loss 1.7565):  40%|███▉      | 375/938 [11:02<17:41,  1.89s/it]
Training 1/1 epoch (loss 1.7140):  40%|███▉      | 375/938 [11:04<17:41,  1.89s/it]
Training 1/1 epoch (loss 1.7140):  40%|████      | 376/938 [11:04<17:17,  1.85s/it]
Training 1/1 epoch (loss 1.8216):  40%|████      | 376/938 [11:06<17:17,  1.85s/it]
Training 1/1 epoch (loss 1.8216):  40%|████      | 377/938 [11:06<18:54,  2.02s/it]
Training 1/1 epoch (loss 1.6263):  40%|████      | 377/938 [11:09<18:54,  2.02s/it]
Training 1/1 epoch (loss 1.6263):  40%|████      | 378/938 [11:09<20:19,  2.18s/it]
Training 1/1 epoch (loss 1.6912):  40%|████      | 378/938 [11:11<20:19,  2.18s/it]
Training 1/1 epoch (loss 1.6912):  40%|████      | 379/938 [11:11<18:29,  1.99s/it]
Training 1/1 epoch (loss 1.6546):  40%|████      | 379/938 [11:13<18:29,  1.99s/it]
Training 1/1 epoch (loss 1.6546):  41%|████      | 380/938 [11:13<19:30,  2.10s/it]
Training 1/1 epoch (loss 1.7205):  41%|████      | 380/938 [11:15<19:30,  2.10s/it]
Training 1/1 epoch (loss 1.7205):  41%|████      | 381/938 [11:15<18:26,  1.99s/it]
Training 1/1 epoch (loss 1.5471):  41%|████      | 381/938 [11:16<18:26,  1.99s/it]
Training 1/1 epoch (loss 1.5471):  41%|████      | 382/938 [11:16<15:48,  1.71s/it]
Training 1/1 epoch (loss 1.7282):  41%|████      | 382/938 [11:17<15:48,  1.71s/it]
Training 1/1 epoch (loss 1.7282):  41%|████      | 383/938 [11:17<15:49,  1.71s/it]
Training 1/1 epoch (loss 1.7488):  41%|████      | 383/938 [11:20<15:49,  1.71s/it]
Training 1/1 epoch (loss 1.7488):  41%|████      | 384/938 [11:20<17:57,  1.95s/it]
Training 1/1 epoch (loss 1.6632):  41%|████      | 384/938 [11:22<17:57,  1.95s/it]
Training 1/1 epoch (loss 1.6632):  41%|████      | 385/938 [11:22<17:56,  1.95s/it]
Training 1/1 epoch (loss 1.6942):  41%|████      | 385/938 [11:24<17:56,  1.95s/it]
Training 1/1 epoch (loss 1.6942):  41%|████      | 386/938 [11:24<18:07,  1.97s/it]
Training 1/1 epoch (loss 1.7519):  41%|████      | 386/938 [11:25<18:07,  1.97s/it]
Training 1/1 epoch (loss 1.7519):  41%|████▏     | 387/938 [11:25<17:04,  1.86s/it]
Training 1/1 epoch (loss 1.6799):  41%|████▏     | 387/938 [11:27<17:04,  1.86s/it]
Training 1/1 epoch (loss 1.6799):  41%|████▏     | 388/938 [11:27<17:20,  1.89s/it]
Training 1/1 epoch (loss 1.7938):  41%|████▏     | 388/938 [11:30<17:20,  1.89s/it]
Training 1/1 epoch (loss 1.7938):  41%|████▏     | 389/938 [11:30<18:46,  2.05s/it]
Training 1/1 epoch (loss 1.7026):  41%|████▏     | 389/938 [11:31<18:46,  2.05s/it]
Training 1/1 epoch (loss 1.7026):  42%|████▏     | 390/938 [11:31<15:40,  1.72s/it]
Training 1/1 epoch (loss 1.6848):  42%|████▏     | 390/938 [11:32<15:40,  1.72s/it]
Training 1/1 epoch (loss 1.6848):  42%|████▏     | 391/938 [11:32<14:05,  1.54s/it]
Training 1/1 epoch (loss 1.8071):  42%|████▏     | 391/938 [11:34<14:05,  1.54s/it]
Training 1/1 epoch (loss 1.8071):  42%|████▏     | 392/938 [11:34<15:31,  1.71s/it]
Training 1/1 epoch (loss 1.6818):  42%|████▏     | 392/938 [11:35<15:31,  1.71s/it]
Training 1/1 epoch (loss 1.6818):  42%|████▏     | 393/938 [11:35<14:23,  1.58s/it]
Training 1/1 epoch (loss 1.8111):  42%|████▏     | 393/938 [11:37<14:23,  1.58s/it]
Training 1/1 epoch (loss 1.8111):  42%|████▏     | 394/938 [11:37<14:22,  1.58s/it]
Training 1/1 epoch (loss 1.6582):  42%|████▏     | 394/938 [11:39<14:22,  1.58s/it]
Training 1/1 epoch (loss 1.6582):  42%|████▏     | 395/938 [11:39<16:36,  1.84s/it]
Training 1/1 epoch (loss 1.6845):  42%|████▏     | 395/938 [11:41<16:36,  1.84s/it]
Training 1/1 epoch (loss 1.6845):  42%|████▏     | 396/938 [11:41<15:07,  1.67s/it]
Training 1/1 epoch (loss 1.6640):  42%|████▏     | 396/938 [11:43<15:07,  1.67s/it]
Training 1/1 epoch (loss 1.6640):  42%|████▏     | 397/938 [11:43<17:07,  1.90s/it]
Training 1/1 epoch (loss 1.7060):  42%|████▏     | 397/938 [11:45<17:07,  1.90s/it]
Training 1/1 epoch (loss 1.7060):  42%|████▏     | 398/938 [11:45<17:02,  1.89s/it]
Training 1/1 epoch (loss 1.6963):  42%|████▏     | 398/938 [11:47<17:02,  1.89s/it]
Training 1/1 epoch (loss 1.6963):  43%|████▎     | 399/938 [11:47<16:35,  1.85s/it]
Training 1/1 epoch (loss 1.7852):  43%|████▎     | 399/938 [11:48<16:35,  1.85s/it]
Training 1/1 epoch (loss 1.7852):  43%|████▎     | 400/938 [11:48<15:23,  1.72s/it]
Training 1/1 epoch (loss 1.7405):  43%|████▎     | 400/938 [11:50<15:23,  1.72s/it]
Training 1/1 epoch (loss 1.7405):  43%|████▎     | 401/938 [11:50<16:58,  1.90s/it]
Training 1/1 epoch (loss 1.7326):  43%|████▎     | 401/938 [11:52<16:58,  1.90s/it]
Training 1/1 epoch (loss 1.7326):  43%|████▎     | 402/938 [11:52<15:19,  1.71s/it]
Training 1/1 epoch (loss 1.7300):  43%|████▎     | 402/938 [11:54<15:19,  1.71s/it]
Training 1/1 epoch (loss 1.7300):  43%|████▎     | 403/938 [11:54<16:10,  1.81s/it]
Training 1/1 epoch (loss 1.7113):  43%|████▎     | 403/938 [11:55<16:10,  1.81s/it]
Training 1/1 epoch (loss 1.7113):  43%|████▎     | 404/938 [11:55<15:31,  1.74s/it]
Training 1/1 epoch (loss 1.6233):  43%|████▎     | 404/938 [11:57<15:31,  1.74s/it]
Training 1/1 epoch (loss 1.6233):  43%|████▎     | 405/938 [11:57<14:45,  1.66s/it]
Training 1/1 epoch (loss 1.7872):  43%|████▎     | 405/938 [11:59<14:45,  1.66s/it]
Training 1/1 epoch (loss 1.7872):  43%|████▎     | 406/938 [11:59<15:10,  1.71s/it]
Training 1/1 epoch (loss 1.7519):  43%|████▎     | 406/938 [12:00<15:10,  1.71s/it]
Training 1/1 epoch (loss 1.7519):  43%|████▎     | 407/938 [12:00<15:37,  1.77s/it]
Training 1/1 epoch (loss 1.6587):  43%|████▎     | 407/938 [12:03<15:37,  1.77s/it]
Training 1/1 epoch (loss 1.6587):  43%|████▎     | 408/938 [12:03<17:54,  2.03s/it]
Training 1/1 epoch (loss 1.7044):  43%|████▎     | 408/938 [12:05<17:54,  2.03s/it]
Training 1/1 epoch (loss 1.7044):  44%|████▎     | 409/938 [12:05<16:12,  1.84s/it]
Training 1/1 epoch (loss 1.7185):  44%|████▎     | 409/938 [12:07<16:12,  1.84s/it]
Training 1/1 epoch (loss 1.7185):  44%|████▎     | 410/938 [12:07<17:46,  2.02s/it]
Training 1/1 epoch (loss 1.5486):  44%|████▎     | 410/938 [12:08<17:46,  2.02s/it]
Training 1/1 epoch (loss 1.5486):  44%|████▍     | 411/938 [12:08<15:53,  1.81s/it]
Training 1/1 epoch (loss 1.7297):  44%|████▍     | 411/938 [12:10<15:53,  1.81s/it]
Training 1/1 epoch (loss 1.7297):  44%|████▍     | 412/938 [12:10<14:40,  1.67s/it]
Training 1/1 epoch (loss 1.6108):  44%|████▍     | 412/938 [12:12<14:40,  1.67s/it]
Training 1/1 epoch (loss 1.6108):  44%|████▍     | 413/938 [12:12<16:02,  1.83s/it]
Training 1/1 epoch (loss 1.6902):  44%|████▍     | 413/938 [12:14<16:02,  1.83s/it]
Training 1/1 epoch (loss 1.6902):  44%|████▍     | 414/938 [12:14<17:15,  1.98s/it]
Training 1/1 epoch (loss 1.6453):  44%|████▍     | 414/938 [12:16<17:15,  1.98s/it]
Training 1/1 epoch (loss 1.6453):  44%|████▍     | 415/938 [12:16<15:32,  1.78s/it]
Training 1/1 epoch (loss 1.6007):  44%|████▍     | 415/938 [12:17<15:32,  1.78s/it]
Training 1/1 epoch (loss 1.6007):  44%|████▍     | 416/938 [12:17<14:45,  1.70s/it]
Training 1/1 epoch (loss 1.5869):  44%|████▍     | 416/938 [12:18<14:45,  1.70s/it]
Training 1/1 epoch (loss 1.5869):  44%|████▍     | 417/938 [12:18<13:50,  1.59s/it]
Training 1/1 epoch (loss 1.5915):  44%|████▍     | 417/938 [12:21<13:50,  1.59s/it]
Training 1/1 epoch (loss 1.5915):  45%|████▍     | 418/938 [12:21<15:58,  1.84s/it]
Training 1/1 epoch (loss 1.6989):  45%|████▍     | 418/938 [12:23<15:58,  1.84s/it]
Training 1/1 epoch (loss 1.6989):  45%|████▍     | 419/938 [12:23<17:00,  1.97s/it]
Training 1/1 epoch (loss 1.7330):  45%|████▍     | 419/938 [12:24<17:00,  1.97s/it]
Training 1/1 epoch (loss 1.7330):  45%|████▍     | 420/938 [12:24<15:20,  1.78s/it]
Training 1/1 epoch (loss 1.6063):  45%|████▍     | 420/938 [12:26<15:20,  1.78s/it]
Training 1/1 epoch (loss 1.6063):  45%|████▍     | 421/938 [12:26<14:57,  1.74s/it]
Training 1/1 epoch (loss 1.6435):  45%|████▍     | 421/938 [12:27<14:57,  1.74s/it]
Training 1/1 epoch (loss 1.6435):  45%|████▍     | 422/938 [12:27<13:22,  1.56s/it]
Training 1/1 epoch (loss 1.6234):  45%|████▍     | 422/938 [12:28<13:22,  1.56s/it]
Training 1/1 epoch (loss 1.6234):  45%|████▌     | 423/938 [12:28<12:44,  1.48s/it]
Training 1/1 epoch (loss 1.5341):  45%|████▌     | 423/938 [12:30<12:44,  1.48s/it]
Training 1/1 epoch (loss 1.5341):  45%|████▌     | 424/938 [12:30<13:50,  1.62s/it]
Training 1/1 epoch (loss 1.7040):  45%|████▌     | 424/938 [12:32<13:50,  1.62s/it]
Training 1/1 epoch (loss 1.7040):  45%|████▌     | 425/938 [12:32<15:06,  1.77s/it]
Training 1/1 epoch (loss 1.6856):  45%|████▌     | 425/938 [12:35<15:06,  1.77s/it]
Training 1/1 epoch (loss 1.6856):  45%|████▌     | 426/938 [12:35<15:41,  1.84s/it]
Training 1/1 epoch (loss 1.7521):  45%|████▌     | 426/938 [12:36<15:41,  1.84s/it]
Training 1/1 epoch (loss 1.7521):  46%|████▌     | 427/938 [12:36<14:19,  1.68s/it]
Training 1/1 epoch (loss 1.6158):  46%|████▌     | 427/938 [12:38<14:19,  1.68s/it]
Training 1/1 epoch (loss 1.6158):  46%|████▌     | 428/938 [12:38<14:34,  1.71s/it]
Training 1/1 epoch (loss 1.6760):  46%|████▌     | 428/938 [12:39<14:34,  1.71s/it]
Training 1/1 epoch (loss 1.6760):  46%|████▌     | 429/938 [12:39<13:30,  1.59s/it]
Training 1/1 epoch (loss 1.7262):  46%|████▌     | 429/938 [12:40<13:30,  1.59s/it]
Training 1/1 epoch (loss 1.7262):  46%|████▌     | 430/938 [12:40<13:12,  1.56s/it]
Training 1/1 epoch (loss 1.6388):  46%|████▌     | 430/938 [12:42<13:12,  1.56s/it]
Training 1/1 epoch (loss 1.6388):  46%|████▌     | 431/938 [12:42<14:26,  1.71s/it]
Training 1/1 epoch (loss 1.7167):  46%|████▌     | 431/938 [12:44<14:26,  1.71s/it]
Training 1/1 epoch (loss 1.7167):  46%|████▌     | 432/938 [12:44<14:58,  1.77s/it]
Training 1/1 epoch (loss 1.6577):  46%|████▌     | 432/938 [12:46<14:58,  1.77s/it]
Training 1/1 epoch (loss 1.6577):  46%|████▌     | 433/938 [12:46<14:44,  1.75s/it]
Training 1/1 epoch (loss 1.7561):  46%|████▌     | 433/938 [12:48<14:44,  1.75s/it]
Training 1/1 epoch (loss 1.7561):  46%|████▋     | 434/938 [12:48<14:42,  1.75s/it]
Training 1/1 epoch (loss 1.6844):  46%|████▋     | 434/938 [12:50<14:42,  1.75s/it]
Training 1/1 epoch (loss 1.6844):  46%|████▋     | 435/938 [12:50<16:41,  1.99s/it]
Training 1/1 epoch (loss 1.7679):  46%|████▋     | 435/938 [12:52<16:41,  1.99s/it]
Training 1/1 epoch (loss 1.7679):  46%|████▋     | 436/938 [12:52<16:29,  1.97s/it]
Training 1/1 epoch (loss 1.7295):  46%|████▋     | 436/938 [12:55<16:29,  1.97s/it]
Training 1/1 epoch (loss 1.7295):  47%|████▋     | 437/938 [12:55<17:39,  2.11s/it]
Training 1/1 epoch (loss 1.5883):  47%|████▋     | 437/938 [12:57<17:39,  2.11s/it]
Training 1/1 epoch (loss 1.5883):  47%|████▋     | 438/938 [12:57<17:29,  2.10s/it]
Training 1/1 epoch (loss 1.7476):  47%|████▋     | 438/938 [12:59<17:29,  2.10s/it]
Training 1/1 epoch (loss 1.7476):  47%|████▋     | 439/938 [12:59<17:53,  2.15s/it]
Training 1/1 epoch (loss 1.6734):  47%|████▋     | 439/938 [13:01<17:53,  2.15s/it]
Training 1/1 epoch (loss 1.6734):  47%|████▋     | 440/938 [13:01<16:47,  2.02s/it]
Training 1/1 epoch (loss 1.6561):  47%|████▋     | 440/938 [13:02<16:47,  2.02s/it]
Training 1/1 epoch (loss 1.6561):  47%|████▋     | 441/938 [13:02<14:57,  1.81s/it]
Training 1/1 epoch (loss 1.5984):  47%|████▋     | 441/938 [13:03<14:57,  1.81s/it]
Training 1/1 epoch (loss 1.5984):  47%|████▋     | 442/938 [13:03<13:46,  1.67s/it]
Training 1/1 epoch (loss 1.6607):  47%|████▋     | 442/938 [13:05<13:46,  1.67s/it]
Training 1/1 epoch (loss 1.6607):  47%|████▋     | 443/938 [13:05<13:15,  1.61s/it]
Training 1/1 epoch (loss 1.7436):  47%|████▋     | 443/938 [13:07<13:15,  1.61s/it]
Training 1/1 epoch (loss 1.7436):  47%|████▋     | 444/938 [13:07<13:28,  1.64s/it]
Training 1/1 epoch (loss 1.5258):  47%|████▋     | 444/938 [13:08<13:28,  1.64s/it]
Training 1/1 epoch (loss 1.5258):  47%|████▋     | 445/938 [13:08<13:41,  1.67s/it]
Training 1/1 epoch (loss 1.5784):  47%|████▋     | 445/938 [13:10<13:41,  1.67s/it]
Training 1/1 epoch (loss 1.5784):  48%|████▊     | 446/938 [13:10<12:44,  1.55s/it]
Training 1/1 epoch (loss 1.7454):  48%|████▊     | 446/938 [13:12<12:44,  1.55s/it]
Training 1/1 epoch (loss 1.7454):  48%|████▊     | 447/938 [13:12<14:57,  1.83s/it]
Training 1/1 epoch (loss 1.6974):  48%|████▊     | 447/938 [13:14<14:57,  1.83s/it]
Training 1/1 epoch (loss 1.6974):  48%|████▊     | 448/938 [13:14<13:49,  1.69s/it]
Training 1/1 epoch (loss 1.7923):  48%|████▊     | 448/938 [13:15<13:49,  1.69s/it]
Training 1/1 epoch (loss 1.7923):  48%|████▊     | 449/938 [13:15<14:28,  1.78s/it]
Training 1/1 epoch (loss 1.5356):  48%|████▊     | 449/938 [13:17<14:28,  1.78s/it]
Training 1/1 epoch (loss 1.5356):  48%|████▊     | 450/938 [13:17<14:22,  1.77s/it]
Training 1/1 epoch (loss 1.6472):  48%|████▊     | 450/938 [13:19<14:22,  1.77s/it]
Training 1/1 epoch (loss 1.6472):  48%|████▊     | 451/938 [13:19<13:17,  1.64s/it]
Training 1/1 epoch (loss 1.6526):  48%|████▊     | 451/938 [13:20<13:17,  1.64s/it]
Training 1/1 epoch (loss 1.6526):  48%|████▊     | 452/938 [13:20<13:40,  1.69s/it]
Training 1/1 epoch (loss 1.6675):  48%|████▊     | 452/938 [13:22<13:40,  1.69s/it]
Training 1/1 epoch (loss 1.6675):  48%|████▊     | 453/938 [13:22<13:03,  1.62s/it]
Training 1/1 epoch (loss 1.6228):  48%|████▊     | 453/938 [13:23<13:03,  1.62s/it]
Training 1/1 epoch (loss 1.6228):  48%|████▊     | 454/938 [13:23<12:45,  1.58s/it]
Training 1/1 epoch (loss 1.7333):  48%|████▊     | 454/938 [13:26<12:45,  1.58s/it]
Training 1/1 epoch (loss 1.7333):  49%|████▊     | 455/938 [13:26<14:47,  1.84s/it]
Training 1/1 epoch (loss 1.6566):  49%|████▊     | 455/938 [13:28<14:47,  1.84s/it]
Training 1/1 epoch (loss 1.6566):  49%|████▊     | 456/938 [13:28<15:38,  1.95s/it]
Training 1/1 epoch (loss 1.7602):  49%|████▊     | 456/938 [13:29<15:38,  1.95s/it]
Training 1/1 epoch (loss 1.7602):  49%|████▊     | 457/938 [13:29<14:32,  1.81s/it]
Training 1/1 epoch (loss 1.7448):  49%|████▊     | 457/938 [13:31<14:32,  1.81s/it]
Training 1/1 epoch (loss 1.7448):  49%|████▉     | 458/938 [13:31<14:25,  1.80s/it]
Training 1/1 epoch (loss 1.7527):  49%|████▉     | 458/938 [13:33<14:25,  1.80s/it]
Training 1/1 epoch (loss 1.7527):  49%|████▉     | 459/938 [13:33<15:25,  1.93s/it]
Training 1/1 epoch (loss 1.7173):  49%|████▉     | 459/938 [13:35<15:25,  1.93s/it]
Training 1/1 epoch (loss 1.7173):  49%|████▉     | 460/938 [13:35<14:07,  1.77s/it]
Training 1/1 epoch (loss 1.6367):  49%|████▉     | 460/938 [13:36<14:07,  1.77s/it]
Training 1/1 epoch (loss 1.6367):  49%|████▉     | 461/938 [13:36<12:52,  1.62s/it]
Training 1/1 epoch (loss 1.6336):  49%|████▉     | 461/938 [13:37<12:52,  1.62s/it]
Training 1/1 epoch (loss 1.6336):  49%|████▉     | 462/938 [13:37<11:43,  1.48s/it]
Training 1/1 epoch (loss 1.6535):  49%|████▉     | 462/938 [13:39<11:43,  1.48s/it]
Training 1/1 epoch (loss 1.6535):  49%|████▉     | 463/938 [13:39<13:16,  1.68s/it]
Training 1/1 epoch (loss 1.6838):  49%|████▉     | 463/938 [13:42<13:16,  1.68s/it]
Training 1/1 epoch (loss 1.6838):  49%|████▉     | 464/938 [13:42<14:16,  1.81s/it]
Training 1/1 epoch (loss 1.6924):  49%|████▉     | 464/938 [13:43<14:16,  1.81s/it]
Training 1/1 epoch (loss 1.6924):  50%|████▉     | 465/938 [13:43<12:48,  1.62s/it]
Training 1/1 epoch (loss 1.7381):  50%|████▉     | 465/938 [13:44<12:48,  1.62s/it]
Training 1/1 epoch (loss 1.7381):  50%|████▉     | 466/938 [13:44<13:08,  1.67s/it]
Training 1/1 epoch (loss 1.7371):  50%|████▉     | 466/938 [13:47<13:08,  1.67s/it]
Training 1/1 epoch (loss 1.7371):  50%|████▉     | 467/938 [13:47<15:10,  1.93s/it]
Training 1/1 epoch (loss 1.7783):  50%|████▉     | 467/938 [13:48<15:10,  1.93s/it]
Training 1/1 epoch (loss 1.7783):  50%|████▉     | 468/938 [13:48<13:13,  1.69s/it]
Training 1/1 epoch (loss 1.7299):  50%|████▉     | 468/938 [13:50<13:13,  1.69s/it]
Training 1/1 epoch (loss 1.7299):  50%|█████     | 469/938 [13:50<14:12,  1.82s/it]
Training 1/1 epoch (loss 1.5889):  50%|█████     | 469/938 [13:52<14:12,  1.82s/it]
Training 1/1 epoch (loss 1.5889):  50%|█████     | 470/938 [13:52<14:12,  1.82s/it]
Training 1/1 epoch (loss 1.6707):  50%|█████     | 470/938 [13:54<14:12,  1.82s/it]
Training 1/1 epoch (loss 1.6707):  50%|█████     | 471/938 [13:54<15:28,  1.99s/it]
Training 1/1 epoch (loss 1.7085):  50%|█████     | 471/938 [13:57<15:28,  1.99s/it]
Training 1/1 epoch (loss 1.7085):  50%|█████     | 472/938 [13:57<17:34,  2.26s/it]
Training 1/1 epoch (loss 1.5644):  50%|█████     | 472/938 [13:59<17:34,  2.26s/it]
Training 1/1 epoch (loss 1.5644):  50%|█████     | 473/938 [13:59<15:20,  1.98s/it]
Training 1/1 epoch (loss 1.7177):  50%|█████     | 473/938 [14:01<15:20,  1.98s/it]
Training 1/1 epoch (loss 1.7177):  51%|█████     | 474/938 [14:01<15:55,  2.06s/it]
Training 1/1 epoch (loss 1.6041):  51%|█████     | 474/938 [14:03<15:55,  2.06s/it]
Training 1/1 epoch (loss 1.6041):  51%|█████     | 475/938 [14:03<15:36,  2.02s/it]
Training 1/1 epoch (loss 1.6741):  51%|█████     | 475/938 [14:05<15:36,  2.02s/it]
Training 1/1 epoch (loss 1.6741):  51%|█████     | 476/938 [14:05<16:30,  2.14s/it]
Training 1/1 epoch (loss 1.6186):  51%|█████     | 476/938 [14:07<16:30,  2.14s/it]
Training 1/1 epoch (loss 1.6186):  51%|█████     | 477/938 [14:07<15:49,  2.06s/it]
Training 1/1 epoch (loss 1.5926):  51%|█████     | 477/938 [14:08<15:49,  2.06s/it]
Training 1/1 epoch (loss 1.5926):  51%|█████     | 478/938 [14:08<14:03,  1.83s/it]
Training 1/1 epoch (loss 1.7135):  51%|█████     | 478/938 [14:10<14:03,  1.83s/it]
Training 1/1 epoch (loss 1.7135):  51%|█████     | 479/938 [14:10<13:18,  1.74s/it]
Training 1/1 epoch (loss 1.6665):  51%|█████     | 479/938 [14:13<13:18,  1.74s/it]
Training 1/1 epoch (loss 1.6665):  51%|█████     | 480/938 [14:13<15:00,  1.97s/it]
Training 1/1 epoch (loss 1.7033):  51%|█████     | 480/938 [14:14<15:00,  1.97s/it]
Training 1/1 epoch (loss 1.7033):  51%|█████▏    | 481/938 [14:14<13:59,  1.84s/it]
Training 1/1 epoch (loss 1.6576):  51%|█████▏    | 481/938 [14:16<13:59,  1.84s/it]
Training 1/1 epoch (loss 1.6576):  51%|█████▏    | 482/938 [14:16<14:30,  1.91s/it]
Training 1/1 epoch (loss 1.5951):  51%|█████▏    | 482/938 [14:18<14:30,  1.91s/it]
Training 1/1 epoch (loss 1.5951):  51%|█████▏    | 483/938 [14:18<14:01,  1.85s/it]
Training 1/1 epoch (loss 1.8186):  51%|█████▏    | 483/938 [14:20<14:01,  1.85s/it]
Training 1/1 epoch (loss 1.8186):  52%|█████▏    | 484/938 [14:20<14:41,  1.94s/it]
Training 1/1 epoch (loss 1.7995):  52%|█████▏    | 484/938 [14:21<14:41,  1.94s/it]
Training 1/1 epoch (loss 1.7995):  52%|█████▏    | 485/938 [14:21<12:24,  1.64s/it]
Training 1/1 epoch (loss 1.6765):  52%|█████▏    | 485/938 [14:23<12:24,  1.64s/it]
Training 1/1 epoch (loss 1.6765):  52%|█████▏    | 486/938 [14:23<13:06,  1.74s/it]
Training 1/1 epoch (loss 1.6843):  52%|█████▏    | 486/938 [14:24<13:06,  1.74s/it]
Training 1/1 epoch (loss 1.6843):  52%|█████▏    | 487/938 [14:24<12:23,  1.65s/it]
Training 1/1 epoch (loss 1.6547):  52%|█████▏    | 487/938 [14:26<12:23,  1.65s/it]
Training 1/1 epoch (loss 1.6547):  52%|█████▏    | 488/938 [14:26<12:29,  1.66s/it]
Training 1/1 epoch (loss 1.6791):  52%|█████▏    | 488/938 [14:27<12:29,  1.66s/it]
Training 1/1 epoch (loss 1.6791):  52%|█████▏    | 489/938 [14:27<11:05,  1.48s/it]
Training 1/1 epoch (loss 1.7183):  52%|█████▏    | 489/938 [14:30<11:05,  1.48s/it]
Training 1/1 epoch (loss 1.7183):  52%|█████▏    | 490/938 [14:30<13:11,  1.77s/it]
Training 1/1 epoch (loss 1.7868):  52%|█████▏    | 490/938 [14:32<13:11,  1.77s/it]
Training 1/1 epoch (loss 1.7868):  52%|█████▏    | 491/938 [14:32<13:44,  1.84s/it]
Training 1/1 epoch (loss 1.7571):  52%|█████▏    | 491/938 [14:33<13:44,  1.84s/it]
Training 1/1 epoch (loss 1.7571):  52%|█████▏    | 492/938 [14:33<12:47,  1.72s/it]
Training 1/1 epoch (loss 1.6114):  52%|█████▏    | 492/938 [14:35<12:47,  1.72s/it]
Training 1/1 epoch (loss 1.6114):  53%|█████▎    | 493/938 [14:35<12:48,  1.73s/it]
Training 1/1 epoch (loss 1.6650):  53%|█████▎    | 493/938 [14:36<12:48,  1.73s/it]
Training 1/1 epoch (loss 1.6650):  53%|█████▎    | 494/938 [14:36<12:12,  1.65s/it]
Training 1/1 epoch (loss 1.6952):  53%|█████▎    | 494/938 [14:37<12:12,  1.65s/it]
Training 1/1 epoch (loss 1.6952):  53%|█████▎    | 495/938 [14:37<10:40,  1.45s/it]
Training 1/1 epoch (loss 1.5615):  53%|█████▎    | 495/938 [14:39<10:40,  1.45s/it]
Training 1/1 epoch (loss 1.5615):  53%|█████▎    | 496/938 [14:39<12:22,  1.68s/it]
Training 1/1 epoch (loss 1.7291):  53%|█████▎    | 496/938 [14:40<12:22,  1.68s/it]
Training 1/1 epoch (loss 1.7291):  53%|█████▎    | 497/938 [14:40<10:50,  1.48s/it]
Training 1/1 epoch (loss 1.6759):  53%|█████▎    | 497/938 [14:43<10:50,  1.48s/it]
Training 1/1 epoch (loss 1.6759):  53%|█████▎    | 498/938 [14:43<12:56,  1.76s/it]
Training 1/1 epoch (loss 1.6221):  53%|█████▎    | 498/938 [14:44<12:56,  1.76s/it]
Training 1/1 epoch (loss 1.6221):  53%|█████▎    | 499/938 [14:44<12:03,  1.65s/it]
Training 1/1 epoch (loss 1.5733):  53%|█████▎    | 499/938 [14:46<12:03,  1.65s/it]
Training 1/1 epoch (loss 1.5733):  53%|█████▎    | 500/938 [14:46<11:52,  1.63s/it]
Training 1/1 epoch (loss 1.7118):  53%|█████▎    | 500/938 [14:47<11:52,  1.63s/it]
Training 1/1 epoch (loss 1.7118):  53%|█████▎    | 501/938 [14:47<11:04,  1.52s/it]
Training 1/1 epoch (loss 1.7094):  53%|█████▎    | 501/938 [14:49<11:04,  1.52s/it]
Training 1/1 epoch (loss 1.7094):  54%|█████▎    | 502/938 [14:49<12:09,  1.67s/it]
Training 1/1 epoch (loss 1.6651):  54%|█████▎    | 502/938 [14:51<12:09,  1.67s/it]
Training 1/1 epoch (loss 1.6651):  54%|█████▎    | 503/938 [14:51<12:07,  1.67s/it]
Training 1/1 epoch (loss 1.5667):  54%|█████▎    | 503/938 [14:53<12:07,  1.67s/it]
Training 1/1 epoch (loss 1.5667):  54%|█████▎    | 504/938 [14:53<13:28,  1.86s/it]
Training 1/1 epoch (loss 1.6659):  54%|█████▎    | 504/938 [14:55<13:28,  1.86s/it]
Training 1/1 epoch (loss 1.6659):  54%|█████▍    | 505/938 [14:55<13:42,  1.90s/it]
Training 1/1 epoch (loss 1.6712):  54%|█████▍    | 505/938 [14:56<13:42,  1.90s/it]
Training 1/1 epoch (loss 1.6712):  54%|█████▍    | 506/938 [14:56<12:03,  1.67s/it]
Training 1/1 epoch (loss 1.7060):  54%|█████▍    | 506/938 [14:58<12:03,  1.67s/it]
Training 1/1 epoch (loss 1.7060):  54%|█████▍    | 507/938 [14:58<12:30,  1.74s/it]
Training 1/1 epoch (loss 1.6946):  54%|█████▍    | 507/938 [15:00<12:30,  1.74s/it]
Training 1/1 epoch (loss 1.6946):  54%|█████▍    | 508/938 [15:00<13:23,  1.87s/it]
Training 1/1 epoch (loss 1.6668):  54%|█████▍    | 508/938 [15:01<13:23,  1.87s/it]
Training 1/1 epoch (loss 1.6668):  54%|█████▍    | 509/938 [15:01<11:29,  1.61s/it]
Training 1/1 epoch (loss 1.7390):  54%|█████▍    | 509/938 [15:04<11:29,  1.61s/it]
Training 1/1 epoch (loss 1.7390):  54%|█████▍    | 510/938 [15:04<13:07,  1.84s/it]
Training 1/1 epoch (loss 1.8071):  54%|█████▍    | 510/938 [15:05<13:07,  1.84s/it]
Training 1/1 epoch (loss 1.8071):  54%|█████▍    | 511/938 [15:05<12:41,  1.78s/it]
Training 1/1 epoch (loss 1.6757):  54%|█████▍    | 511/938 [15:08<12:41,  1.78s/it]
Training 1/1 epoch (loss 1.6757):  55%|█████▍    | 512/938 [15:08<14:29,  2.04s/it]
Training 1/1 epoch (loss 1.7170):  55%|█████▍    | 512/938 [15:09<14:29,  2.04s/it]
Training 1/1 epoch (loss 1.7170):  55%|█████▍    | 513/938 [15:09<12:30,  1.77s/it]
Training 1/1 epoch (loss 1.6939):  55%|█████▍    | 513/938 [15:10<12:30,  1.77s/it]
Training 1/1 epoch (loss 1.6939):  55%|█████▍    | 514/938 [15:10<10:59,  1.56s/it]
Training 1/1 epoch (loss 1.6355):  55%|█████▍    | 514/938 [15:12<10:59,  1.56s/it]
Training 1/1 epoch (loss 1.6355):  55%|█████▍    | 515/938 [15:12<10:55,  1.55s/it]
Training 1/1 epoch (loss 1.6799):  55%|█████▍    | 515/938 [15:13<10:55,  1.55s/it]
Training 1/1 epoch (loss 1.6799):  55%|█████▌    | 516/938 [15:13<10:58,  1.56s/it]
Training 1/1 epoch (loss 1.7207):  55%|█████▌    | 516/938 [15:14<10:58,  1.56s/it]
Training 1/1 epoch (loss 1.7207):  55%|█████▌    | 517/938 [15:14<10:12,  1.46s/it]
Training 1/1 epoch (loss 1.6529):  55%|█████▌    | 517/938 [15:16<10:12,  1.46s/it]
Training 1/1 epoch (loss 1.6529):  55%|█████▌    | 518/938 [15:16<10:32,  1.51s/it]
Training 1/1 epoch (loss 1.7066):  55%|█████▌    | 518/938 [15:18<10:32,  1.51s/it]
Training 1/1 epoch (loss 1.7066):  55%|█████▌    | 519/938 [15:18<12:07,  1.74s/it]
Training 1/1 epoch (loss 1.7686):  55%|█████▌    | 519/938 [15:21<12:07,  1.74s/it]
Training 1/1 epoch (loss 1.7686):  55%|█████▌    | 520/938 [15:21<14:09,  2.03s/it]
Training 1/1 epoch (loss 1.5733):  55%|█████▌    | 520/938 [15:23<14:09,  2.03s/it]
Training 1/1 epoch (loss 1.5733):  56%|█████▌    | 521/938 [15:23<13:43,  1.97s/it]
Training 1/1 epoch (loss 1.6656):  56%|█████▌    | 521/938 [15:25<13:43,  1.97s/it]
Training 1/1 epoch (loss 1.6656):  56%|█████▌    | 522/938 [15:25<13:17,  1.92s/it]
Training 1/1 epoch (loss 1.6536):  56%|█████▌    | 522/938 [15:26<13:17,  1.92s/it]
Training 1/1 epoch (loss 1.6536):  56%|█████▌    | 523/938 [15:26<11:21,  1.64s/it]
Training 1/1 epoch (loss 1.5721):  56%|█████▌    | 523/938 [15:27<11:21,  1.64s/it]
Training 1/1 epoch (loss 1.5721):  56%|█████▌    | 524/938 [15:27<11:28,  1.66s/it]
Training 1/1 epoch (loss 1.6322):  56%|█████▌    | 524/938 [15:29<11:28,  1.66s/it]
Training 1/1 epoch (loss 1.6322):  56%|█████▌    | 525/938 [15:29<12:08,  1.77s/it]
Training 1/1 epoch (loss 1.6944):  56%|█████▌    | 525/938 [15:31<12:08,  1.77s/it]
Training 1/1 epoch (loss 1.6944):  56%|█████▌    | 526/938 [15:31<10:58,  1.60s/it]
Training 1/1 epoch (loss 1.6409):  56%|█████▌    | 526/938 [15:32<10:58,  1.60s/it]
Training 1/1 epoch (loss 1.6409):  56%|█████▌    | 527/938 [15:32<11:12,  1.64s/it]
Training 1/1 epoch (loss 1.6971):  56%|█████▌    | 527/938 [15:35<11:12,  1.64s/it]
Training 1/1 epoch (loss 1.6971):  56%|█████▋    | 528/938 [15:35<13:16,  1.94s/it]
Training 1/1 epoch (loss 1.6193):  56%|█████▋    | 528/938 [15:37<13:16,  1.94s/it]
Training 1/1 epoch (loss 1.6193):  56%|█████▋    | 529/938 [15:37<12:39,  1.86s/it]
Training 1/1 epoch (loss 1.6399):  56%|█████▋    | 529/938 [15:39<12:39,  1.86s/it]
Training 1/1 epoch (loss 1.6399):  57%|█████▋    | 530/938 [15:39<12:48,  1.88s/it]
Training 1/1 epoch (loss 1.6076):  57%|█████▋    | 530/938 [15:40<12:48,  1.88s/it]
Training 1/1 epoch (loss 1.6076):  57%|█████▋    | 531/938 [15:40<11:24,  1.68s/it]
Training 1/1 epoch (loss 1.7347):  57%|█████▋    | 531/938 [15:41<11:24,  1.68s/it]
Training 1/1 epoch (loss 1.7347):  57%|█████▋    | 532/938 [15:41<11:05,  1.64s/it]
Training 1/1 epoch (loss 1.6753):  57%|█████▋    | 532/938 [15:43<11:05,  1.64s/it]
Training 1/1 epoch (loss 1.6753):  57%|█████▋    | 533/938 [15:43<11:17,  1.67s/it]
Training 1/1 epoch (loss 1.6207):  57%|█████▋    | 533/938 [15:45<11:17,  1.67s/it]
Training 1/1 epoch (loss 1.6207):  57%|█████▋    | 534/938 [15:45<11:53,  1.77s/it]
Training 1/1 epoch (loss 1.6286):  57%|█████▋    | 534/938 [15:46<11:53,  1.77s/it]
Training 1/1 epoch (loss 1.6286):  57%|█████▋    | 535/938 [15:46<10:55,  1.63s/it]
Training 1/1 epoch (loss 1.5639):  57%|█████▋    | 535/938 [15:49<10:55,  1.63s/it]
Training 1/1 epoch (loss 1.5639):  57%|█████▋    | 536/938 [15:49<13:04,  1.95s/it]
Training 1/1 epoch (loss 1.6974):  57%|█████▋    | 536/938 [15:51<13:04,  1.95s/it]
Training 1/1 epoch (loss 1.6974):  57%|█████▋    | 537/938 [15:51<11:58,  1.79s/it]
Training 1/1 epoch (loss 1.5899):  57%|█████▋    | 537/938 [15:53<11:58,  1.79s/it]
Training 1/1 epoch (loss 1.5899):  57%|█████▋    | 538/938 [15:53<13:10,  1.98s/it]
Training 1/1 epoch (loss 1.6678):  57%|█████▋    | 538/938 [15:55<13:10,  1.98s/it]
Training 1/1 epoch (loss 1.6678):  57%|█████▋    | 539/938 [15:55<13:57,  2.10s/it]
Training 1/1 epoch (loss 1.6711):  57%|█████▋    | 539/938 [15:57<13:57,  2.10s/it]
Training 1/1 epoch (loss 1.6711):  58%|█████▊    | 540/938 [15:57<12:07,  1.83s/it]
Training 1/1 epoch (loss 1.5466):  58%|█████▊    | 540/938 [15:58<12:07,  1.83s/it]
Training 1/1 epoch (loss 1.5466):  58%|█████▊    | 541/938 [15:58<11:57,  1.81s/it]
Training 1/1 epoch (loss 1.6184):  58%|█████▊    | 541/938 [15:59<11:57,  1.81s/it]
Training 1/1 epoch (loss 1.6184):  58%|█████▊    | 542/938 [15:59<10:44,  1.63s/it]
Training 1/1 epoch (loss 1.6934):  58%|█████▊    | 542/938 [16:02<10:44,  1.63s/it]
Training 1/1 epoch (loss 1.6934):  58%|█████▊    | 543/938 [16:02<11:32,  1.75s/it]
Training 1/1 epoch (loss 1.5949):  58%|█████▊    | 543/938 [16:03<11:32,  1.75s/it]
Training 1/1 epoch (loss 1.5949):  58%|█████▊    | 544/938 [16:03<11:40,  1.78s/it]
Training 1/1 epoch (loss 1.5527):  58%|█████▊    | 544/938 [16:06<11:40,  1.78s/it]
Training 1/1 epoch (loss 1.5527):  58%|█████▊    | 545/938 [16:06<12:57,  1.98s/it]
Training 1/1 epoch (loss 1.5874):  58%|█████▊    | 545/938 [16:07<12:57,  1.98s/it]
Training 1/1 epoch (loss 1.5874):  58%|█████▊    | 546/938 [16:07<11:18,  1.73s/it]
Training 1/1 epoch (loss 1.5654):  58%|█████▊    | 546/938 [16:09<11:18,  1.73s/it]
Training 1/1 epoch (loss 1.5654):  58%|█████▊    | 547/938 [16:09<11:57,  1.84s/it]
Training 1/1 epoch (loss 1.6996):  58%|█████▊    | 547/938 [16:11<11:57,  1.84s/it]
Training 1/1 epoch (loss 1.6996):  58%|█████▊    | 548/938 [16:11<11:22,  1.75s/it]
Training 1/1 epoch (loss 1.5919):  58%|█████▊    | 548/938 [16:12<11:22,  1.75s/it]
Training 1/1 epoch (loss 1.5919):  59%|█████▊    | 549/938 [16:12<11:23,  1.76s/it]
Training 1/1 epoch (loss 1.7445):  59%|█████▊    | 549/938 [16:14<11:23,  1.76s/it]
Training 1/1 epoch (loss 1.7445):  59%|█████▊    | 550/938 [16:14<12:04,  1.87s/it]
Training 1/1 epoch (loss 1.6699):  59%|█████▊    | 550/938 [16:16<12:04,  1.87s/it]
Training 1/1 epoch (loss 1.6699):  59%|█████▊    | 551/938 [16:16<10:34,  1.64s/it]
Training 1/1 epoch (loss 1.6124):  59%|█████▊    | 551/938 [16:17<10:34,  1.64s/it]
Training 1/1 epoch (loss 1.6124):  59%|█████▉    | 552/938 [16:17<10:50,  1.69s/it]
Training 1/1 epoch (loss 1.7060):  59%|█████▉    | 552/938 [16:19<10:50,  1.69s/it]
Training 1/1 epoch (loss 1.7060):  59%|█████▉    | 553/938 [16:19<10:53,  1.70s/it]
Training 1/1 epoch (loss 1.6753):  59%|█████▉    | 553/938 [16:21<10:53,  1.70s/it]
Training 1/1 epoch (loss 1.6753):  59%|█████▉    | 554/938 [16:21<11:31,  1.80s/it]
Training 1/1 epoch (loss 1.5931):  59%|█████▉    | 554/938 [16:23<11:31,  1.80s/it]
Training 1/1 epoch (loss 1.5931):  59%|█████▉    | 555/938 [16:23<12:27,  1.95s/it]
Training 1/1 epoch (loss 1.7042):  59%|█████▉    | 555/938 [16:25<12:27,  1.95s/it]
Training 1/1 epoch (loss 1.7042):  59%|█████▉    | 556/938 [16:25<12:04,  1.90s/it]
Training 1/1 epoch (loss 1.6807):  59%|█████▉    | 556/938 [16:28<12:04,  1.90s/it]
Training 1/1 epoch (loss 1.6807):  59%|█████▉    | 557/938 [16:28<13:03,  2.06s/it]
Training 1/1 epoch (loss 1.7135):  59%|█████▉    | 557/938 [16:29<13:03,  2.06s/it]
Training 1/1 epoch (loss 1.7135):  59%|█████▉    | 558/938 [16:29<12:21,  1.95s/it]
Training 1/1 epoch (loss 1.6850):  59%|█████▉    | 558/938 [16:31<12:21,  1.95s/it]
Training 1/1 epoch (loss 1.6850):  60%|█████▉    | 559/938 [16:31<11:19,  1.79s/it]
Training 1/1 epoch (loss 1.5869):  60%|█████▉    | 559/938 [16:32<11:19,  1.79s/it]
Training 1/1 epoch (loss 1.5869):  60%|█████▉    | 560/938 [16:32<10:07,  1.61s/it]
Training 1/1 epoch (loss 1.6969):  60%|█████▉    | 560/938 [16:34<10:07,  1.61s/it]
Training 1/1 epoch (loss 1.6969):  60%|█████▉    | 561/938 [16:34<10:57,  1.74s/it]
Training 1/1 epoch (loss 1.7370):  60%|█████▉    | 561/938 [16:36<10:57,  1.74s/it]
Training 1/1 epoch (loss 1.7370):  60%|█████▉    | 562/938 [16:36<10:34,  1.69s/it]
Training 1/1 epoch (loss 1.5801):  60%|█████▉    | 562/938 [16:37<10:34,  1.69s/it]
Training 1/1 epoch (loss 1.5801):  60%|██████    | 563/938 [16:37<10:30,  1.68s/it]
Training 1/1 epoch (loss 1.7125):  60%|██████    | 563/938 [16:39<10:30,  1.68s/it]
Training 1/1 epoch (loss 1.7125):  60%|██████    | 564/938 [16:39<10:47,  1.73s/it]
Training 1/1 epoch (loss 1.6552):  60%|██████    | 564/938 [16:41<10:47,  1.73s/it]
Training 1/1 epoch (loss 1.6552):  60%|██████    | 565/938 [16:41<10:25,  1.68s/it]
Training 1/1 epoch (loss 1.6371):  60%|██████    | 565/938 [16:42<10:25,  1.68s/it]
Training 1/1 epoch (loss 1.6371):  60%|██████    | 566/938 [16:42<09:24,  1.52s/it]
Training 1/1 epoch (loss 1.6378):  60%|██████    | 566/938 [16:44<09:24,  1.52s/it]
Training 1/1 epoch (loss 1.6378):  60%|██████    | 567/938 [16:44<11:01,  1.78s/it]
Training 1/1 epoch (loss 1.7080):  60%|██████    | 567/938 [16:46<11:01,  1.78s/it]
Training 1/1 epoch (loss 1.7080):  61%|██████    | 568/938 [16:46<11:56,  1.94s/it]
Training 1/1 epoch (loss 1.7744):  61%|██████    | 568/938 [16:48<11:56,  1.94s/it]
Training 1/1 epoch (loss 1.7744):  61%|██████    | 569/938 [16:48<11:48,  1.92s/it]
Training 1/1 epoch (loss 1.7244):  61%|██████    | 569/938 [16:50<11:48,  1.92s/it]
Training 1/1 epoch (loss 1.7244):  61%|██████    | 570/938 [16:50<10:45,  1.75s/it]
Training 1/1 epoch (loss 1.7050):  61%|██████    | 570/938 [16:51<10:45,  1.75s/it]
Training 1/1 epoch (loss 1.7050):  61%|██████    | 571/938 [16:51<10:38,  1.74s/it]
Training 1/1 epoch (loss 1.7081):  61%|██████    | 571/938 [16:54<10:38,  1.74s/it]
Training 1/1 epoch (loss 1.7081):  61%|██████    | 572/938 [16:54<11:24,  1.87s/it]
Training 1/1 epoch (loss 1.7232):  61%|██████    | 572/938 [16:55<11:24,  1.87s/it]
Training 1/1 epoch (loss 1.7232):  61%|██████    | 573/938 [16:55<11:03,  1.82s/it]
Training 1/1 epoch (loss 1.6579):  61%|██████    | 573/938 [16:58<11:03,  1.82s/it]
Training 1/1 epoch (loss 1.6579):  61%|██████    | 574/938 [16:58<11:53,  1.96s/it]
Training 1/1 epoch (loss 1.6486):  61%|██████    | 574/938 [16:59<11:53,  1.96s/it]
Training 1/1 epoch (loss 1.6486):  61%|██████▏   | 575/938 [16:59<11:14,  1.86s/it]
Training 1/1 epoch (loss 1.6540):  61%|██████▏   | 575/938 [17:01<11:14,  1.86s/it]
Training 1/1 epoch (loss 1.6540):  61%|██████▏   | 576/938 [17:01<11:06,  1.84s/it]
Training 1/1 epoch (loss 1.6416):  61%|██████▏   | 576/938 [17:03<11:06,  1.84s/it]
Training 1/1 epoch (loss 1.6416):  62%|██████▏   | 577/938 [17:03<11:10,  1.86s/it]
Training 1/1 epoch (loss 1.7497):  62%|██████▏   | 577/938 [17:05<11:10,  1.86s/it]
Training 1/1 epoch (loss 1.7497):  62%|██████▏   | 578/938 [17:05<12:14,  2.04s/it]
Training 1/1 epoch (loss 1.7072):  62%|██████▏   | 578/938 [17:07<12:14,  2.04s/it]
Training 1/1 epoch (loss 1.7072):  62%|██████▏   | 579/938 [17:07<12:00,  2.01s/it]
Training 1/1 epoch (loss 1.7092):  62%|██████▏   | 579/938 [17:09<12:00,  2.01s/it]
Training 1/1 epoch (loss 1.7092):  62%|██████▏   | 580/938 [17:09<11:13,  1.88s/it]
Training 1/1 epoch (loss 1.6372):  62%|██████▏   | 580/938 [17:10<11:13,  1.88s/it]
Training 1/1 epoch (loss 1.6372):  62%|██████▏   | 581/938 [17:10<09:52,  1.66s/it]
Training 1/1 epoch (loss 1.5802):  62%|██████▏   | 581/938 [17:12<09:52,  1.66s/it]
Training 1/1 epoch (loss 1.5802):  62%|██████▏   | 582/938 [17:12<10:04,  1.70s/it]
Training 1/1 epoch (loss 1.7043):  62%|██████▏   | 582/938 [17:14<10:04,  1.70s/it]
Training 1/1 epoch (loss 1.7043):  62%|██████▏   | 583/938 [17:14<10:30,  1.78s/it]
Training 1/1 epoch (loss 1.5734):  62%|██████▏   | 583/938 [17:15<10:30,  1.78s/it]
Training 1/1 epoch (loss 1.5734):  62%|██████▏   | 584/938 [17:15<08:52,  1.50s/it]
Training 1/1 epoch (loss 1.5659):  62%|██████▏   | 584/938 [17:16<08:52,  1.50s/it]
Training 1/1 epoch (loss 1.5659):  62%|██████▏   | 585/938 [17:16<08:37,  1.47s/it]
Training 1/1 epoch (loss 1.7004):  62%|██████▏   | 585/938 [17:17<08:37,  1.47s/it]
Training 1/1 epoch (loss 1.7004):  62%|██████▏   | 586/938 [17:17<08:32,  1.46s/it]
Training 1/1 epoch (loss 1.6783):  62%|██████▏   | 586/938 [17:20<08:32,  1.46s/it]
Training 1/1 epoch (loss 1.6783):  63%|██████▎   | 587/938 [17:20<10:03,  1.72s/it]
Training 1/1 epoch (loss 1.6910):  63%|██████▎   | 587/938 [17:21<10:03,  1.72s/it]
Training 1/1 epoch (loss 1.6910):  63%|██████▎   | 588/938 [17:21<09:13,  1.58s/it]
Training 1/1 epoch (loss 1.7185):  63%|██████▎   | 588/938 [17:23<09:13,  1.58s/it]
Training 1/1 epoch (loss 1.7185):  63%|██████▎   | 589/938 [17:23<10:38,  1.83s/it]
Training 1/1 epoch (loss 1.7102):  63%|██████▎   | 589/938 [17:25<10:38,  1.83s/it]
Training 1/1 epoch (loss 1.7102):  63%|██████▎   | 590/938 [17:25<09:55,  1.71s/it]
Training 1/1 epoch (loss 1.6582):  63%|██████▎   | 590/938 [17:26<09:55,  1.71s/it]
Training 1/1 epoch (loss 1.6582):  63%|██████▎   | 591/938 [17:26<09:13,  1.59s/it]
Training 1/1 epoch (loss 1.5853):  63%|██████▎   | 591/938 [17:28<09:13,  1.59s/it]
Training 1/1 epoch (loss 1.5853):  63%|██████▎   | 592/938 [17:28<09:16,  1.61s/it]
Training 1/1 epoch (loss 1.7383):  63%|██████▎   | 592/938 [17:30<09:16,  1.61s/it]
Training 1/1 epoch (loss 1.7383):  63%|██████▎   | 593/938 [17:30<10:45,  1.87s/it]
Training 1/1 epoch (loss 1.7039):  63%|██████▎   | 593/938 [17:32<10:45,  1.87s/it]
Training 1/1 epoch (loss 1.7039):  63%|██████▎   | 594/938 [17:32<10:25,  1.82s/it]
Training 1/1 epoch (loss 1.6169):  63%|██████▎   | 594/938 [17:34<10:25,  1.82s/it]
Training 1/1 epoch (loss 1.6169):  63%|██████▎   | 595/938 [17:34<09:59,  1.75s/it]
Training 1/1 epoch (loss 1.7103):  63%|██████▎   | 595/938 [17:35<09:59,  1.75s/it]
Training 1/1 epoch (loss 1.7103):  64%|██████▎   | 596/938 [17:35<09:56,  1.75s/it]
Training 1/1 epoch (loss 1.6287):  64%|██████▎   | 596/938 [17:37<09:56,  1.75s/it]
Training 1/1 epoch (loss 1.6287):  64%|██████▎   | 597/938 [17:37<09:13,  1.62s/it]
Training 1/1 epoch (loss 1.7467):  64%|██████▎   | 597/938 [17:38<09:13,  1.62s/it]
Training 1/1 epoch (loss 1.7467):  64%|██████▍   | 598/938 [17:38<08:46,  1.55s/it]
Training 1/1 epoch (loss 1.7030):  64%|██████▍   | 598/938 [17:41<08:46,  1.55s/it]
Training 1/1 epoch (loss 1.7030):  64%|██████▍   | 599/938 [17:41<10:14,  1.81s/it]
Training 1/1 epoch (loss 1.6192):  64%|██████▍   | 599/938 [17:42<10:14,  1.81s/it]
Training 1/1 epoch (loss 1.6192):  64%|██████▍   | 600/938 [17:42<10:00,  1.78s/it]
Training 1/1 epoch (loss 1.6988):  64%|██████▍   | 600/938 [17:44<10:00,  1.78s/it]
Training 1/1 epoch (loss 1.6988):  64%|██████▍   | 601/938 [17:44<09:20,  1.66s/it]
Training 1/1 epoch (loss 1.5549):  64%|██████▍   | 601/938 [17:45<09:20,  1.66s/it]
Training 1/1 epoch (loss 1.5549):  64%|██████▍   | 602/938 [17:45<09:23,  1.68s/it]
Training 1/1 epoch (loss 1.6579):  64%|██████▍   | 602/938 [17:48<09:23,  1.68s/it]
Training 1/1 epoch (loss 1.6579):  64%|██████▍   | 603/938 [17:48<10:49,  1.94s/it]
Training 1/1 epoch (loss 1.7234):  64%|██████▍   | 603/938 [17:50<10:49,  1.94s/it]
Training 1/1 epoch (loss 1.7234):  64%|██████▍   | 604/938 [17:50<10:24,  1.87s/it]
Training 1/1 epoch (loss 1.6337):  64%|██████▍   | 604/938 [17:51<10:24,  1.87s/it]
Training 1/1 epoch (loss 1.6337):  64%|██████▍   | 605/938 [17:51<09:49,  1.77s/it]
Training 1/1 epoch (loss 1.6888):  64%|██████▍   | 605/938 [17:53<09:49,  1.77s/it]
Training 1/1 epoch (loss 1.6888):  65%|██████▍   | 606/938 [17:53<10:33,  1.91s/it]
Training 1/1 epoch (loss 1.6501):  65%|██████▍   | 606/938 [17:55<10:33,  1.91s/it]
Training 1/1 epoch (loss 1.6501):  65%|██████▍   | 607/938 [17:55<09:38,  1.75s/it]
Training 1/1 epoch (loss 1.7369):  65%|██████▍   | 607/938 [17:56<09:38,  1.75s/it]
Training 1/1 epoch (loss 1.7369):  65%|██████▍   | 608/938 [17:56<09:19,  1.70s/it]
Training 1/1 epoch (loss 1.6555):  65%|██████▍   | 608/938 [17:58<09:19,  1.70s/it]
Training 1/1 epoch (loss 1.6555):  65%|██████▍   | 609/938 [17:58<09:56,  1.81s/it]
Training 1/1 epoch (loss 1.6312):  65%|██████▍   | 609/938 [18:00<09:56,  1.81s/it]
Training 1/1 epoch (loss 1.6312):  65%|██████▌   | 610/938 [18:00<09:34,  1.75s/it]
Training 1/1 epoch (loss 1.7105):  65%|██████▌   | 610/938 [18:02<09:34,  1.75s/it]
Training 1/1 epoch (loss 1.7105):  65%|██████▌   | 611/938 [18:02<10:11,  1.87s/it]
Training 1/1 epoch (loss 1.6007):  65%|██████▌   | 611/938 [18:04<10:11,  1.87s/it]
Training 1/1 epoch (loss 1.6007):  65%|██████▌   | 612/938 [18:04<10:23,  1.91s/it]
Training 1/1 epoch (loss 1.4909):  65%|██████▌   | 612/938 [18:06<10:23,  1.91s/it]
Training 1/1 epoch (loss 1.4909):  65%|██████▌   | 613/938 [18:06<10:54,  2.01s/it]
Training 1/1 epoch (loss 1.7684):  65%|██████▌   | 613/938 [18:08<10:54,  2.01s/it]
Training 1/1 epoch (loss 1.7684):  65%|██████▌   | 614/938 [18:08<10:48,  2.00s/it]
Training 1/1 epoch (loss 1.5471):  65%|██████▌   | 614/938 [18:11<10:48,  2.00s/it]
Training 1/1 epoch (loss 1.5471):  66%|██████▌   | 615/938 [18:11<11:22,  2.11s/it]
Training 1/1 epoch (loss 1.6809):  66%|██████▌   | 615/938 [18:12<11:22,  2.11s/it]
Training 1/1 epoch (loss 1.6809):  66%|██████▌   | 616/938 [18:12<10:11,  1.90s/it]
Training 1/1 epoch (loss 1.5531):  66%|██████▌   | 616/938 [18:13<10:11,  1.90s/it]
Training 1/1 epoch (loss 1.5531):  66%|██████▌   | 617/938 [18:13<08:58,  1.68s/it]
Training 1/1 epoch (loss 1.6196):  66%|██████▌   | 617/938 [18:15<08:58,  1.68s/it]
Training 1/1 epoch (loss 1.6196):  66%|██████▌   | 618/938 [18:15<09:06,  1.71s/it]
Training 1/1 epoch (loss 1.6438):  66%|██████▌   | 618/938 [18:17<09:06,  1.71s/it]
Training 1/1 epoch (loss 1.6438):  66%|██████▌   | 619/938 [18:17<09:24,  1.77s/it]
Training 1/1 epoch (loss 1.5402):  66%|██████▌   | 619/938 [18:19<09:24,  1.77s/it]
Training 1/1 epoch (loss 1.5402):  66%|██████▌   | 620/938 [18:19<10:26,  1.97s/it]
Training 1/1 epoch (loss 1.5305):  66%|██████▌   | 620/938 [18:22<10:26,  1.97s/it]
Training 1/1 epoch (loss 1.5305):  66%|██████▌   | 621/938 [18:22<10:40,  2.02s/it]
Training 1/1 epoch (loss 1.7508):  66%|██████▌   | 621/938 [18:24<10:40,  2.02s/it]
Training 1/1 epoch (loss 1.7508):  66%|██████▋   | 622/938 [18:24<10:36,  2.01s/it]
Training 1/1 epoch (loss 1.5797):  66%|██████▋   | 622/938 [18:25<10:36,  2.01s/it]
Training 1/1 epoch (loss 1.5797):  66%|██████▋   | 623/938 [18:25<09:28,  1.80s/it]
Training 1/1 epoch (loss 1.6022):  66%|██████▋   | 623/938 [18:26<09:28,  1.80s/it]
Training 1/1 epoch (loss 1.6022):  67%|██████▋   | 624/938 [18:26<08:55,  1.71s/it]
Training 1/1 epoch (loss 1.4827):  67%|██████▋   | 624/938 [18:28<08:55,  1.71s/it]
Training 1/1 epoch (loss 1.4827):  67%|██████▋   | 625/938 [18:28<08:51,  1.70s/it]
Training 1/1 epoch (loss 1.6315):  67%|██████▋   | 625/938 [18:30<08:51,  1.70s/it]
Training 1/1 epoch (loss 1.6315):  67%|██████▋   | 626/938 [18:30<09:10,  1.76s/it]
Training 1/1 epoch (loss 1.6479):  67%|██████▋   | 626/938 [18:31<09:10,  1.76s/it]
Training 1/1 epoch (loss 1.6479):  67%|██████▋   | 627/938 [18:31<08:40,  1.67s/it]
Training 1/1 epoch (loss 1.6270):  67%|██████▋   | 627/938 [18:34<08:40,  1.67s/it]
Training 1/1 epoch (loss 1.6270):  67%|██████▋   | 628/938 [18:34<09:24,  1.82s/it]
Training 1/1 epoch (loss 1.6150):  67%|██████▋   | 628/938 [18:35<09:24,  1.82s/it]
Training 1/1 epoch (loss 1.6150):  67%|██████▋   | 629/938 [18:35<08:04,  1.57s/it]
Training 1/1 epoch (loss 1.6671):  67%|██████▋   | 629/938 [18:37<08:04,  1.57s/it]
Training 1/1 epoch (loss 1.6671):  67%|██████▋   | 630/938 [18:37<09:05,  1.77s/it]
Training 1/1 epoch (loss 1.5844):  67%|██████▋   | 630/938 [18:38<09:05,  1.77s/it]
Training 1/1 epoch (loss 1.5844):  67%|██████▋   | 631/938 [18:38<08:48,  1.72s/it]
Training 1/1 epoch (loss 1.7203):  67%|██████▋   | 631/938 [18:40<08:48,  1.72s/it]
Training 1/1 epoch (loss 1.7203):  67%|██████▋   | 632/938 [18:40<08:30,  1.67s/it]
Training 1/1 epoch (loss 1.5241):  67%|██████▋   | 632/938 [18:41<08:30,  1.67s/it]
Training 1/1 epoch (loss 1.5241):  67%|██████▋   | 633/938 [18:41<08:04,  1.59s/it]
Training 1/1 epoch (loss 1.6496):  67%|██████▋   | 633/938 [18:43<08:04,  1.59s/it]
Training 1/1 epoch (loss 1.6496):  68%|██████▊   | 634/938 [18:43<08:49,  1.74s/it]
Training 1/1 epoch (loss 1.6195):  68%|██████▊   | 634/938 [18:45<08:49,  1.74s/it]
Training 1/1 epoch (loss 1.6195):  68%|██████▊   | 635/938 [18:45<09:05,  1.80s/it]
Training 1/1 epoch (loss 1.5745):  68%|██████▊   | 635/938 [18:47<09:05,  1.80s/it]
Training 1/1 epoch (loss 1.5745):  68%|██████▊   | 636/938 [18:47<09:03,  1.80s/it]
Training 1/1 epoch (loss 1.6842):  68%|██████▊   | 636/938 [18:48<09:03,  1.80s/it]
Training 1/1 epoch (loss 1.6842):  68%|██████▊   | 637/938 [18:48<08:04,  1.61s/it]
Training 1/1 epoch (loss 1.6555):  68%|██████▊   | 637/938 [18:50<08:04,  1.61s/it]
Training 1/1 epoch (loss 1.6555):  68%|██████▊   | 638/938 [18:50<08:13,  1.65s/it]
Training 1/1 epoch (loss 1.5324):  68%|██████▊   | 638/938 [18:52<08:13,  1.65s/it]
Training 1/1 epoch (loss 1.5324):  68%|██████▊   | 639/938 [18:52<07:59,  1.60s/it]
Training 1/1 epoch (loss 1.7007):  68%|██████▊   | 639/938 [18:53<07:59,  1.60s/it]
Training 1/1 epoch (loss 1.7007):  68%|██████▊   | 640/938 [18:53<08:15,  1.66s/it]
Training 1/1 epoch (loss 1.5156):  68%|██████▊   | 640/938 [18:55<08:15,  1.66s/it]
Training 1/1 epoch (loss 1.5156):  68%|██████▊   | 641/938 [18:55<08:46,  1.77s/it]
Training 1/1 epoch (loss 1.6029):  68%|██████▊   | 641/938 [18:57<08:46,  1.77s/it]
Training 1/1 epoch (loss 1.6029):  68%|██████▊   | 642/938 [18:57<08:30,  1.72s/it]
Training 1/1 epoch (loss 1.7046):  68%|██████▊   | 642/938 [18:58<08:30,  1.72s/it]
Training 1/1 epoch (loss 1.7046):  69%|██████▊   | 643/938 [18:58<07:35,  1.54s/it]
Training 1/1 epoch (loss 1.5233):  69%|██████▊   | 643/938 [19:01<07:35,  1.54s/it]
Training 1/1 epoch (loss 1.5233):  69%|██████▊   | 644/938 [19:01<08:50,  1.81s/it]
Training 1/1 epoch (loss 1.6974):  69%|██████▊   | 644/938 [19:02<08:50,  1.81s/it]
Training 1/1 epoch (loss 1.6974):  69%|██████▉   | 645/938 [19:02<08:27,  1.73s/it]
Training 1/1 epoch (loss 1.7108):  69%|██████▉   | 645/938 [19:04<08:27,  1.73s/it]
Training 1/1 epoch (loss 1.7108):  69%|██████▉   | 646/938 [19:04<08:38,  1.78s/it]
Training 1/1 epoch (loss 1.6923):  69%|██████▉   | 646/938 [19:05<08:38,  1.78s/it]
Training 1/1 epoch (loss 1.6923):  69%|██████▉   | 647/938 [19:05<07:43,  1.59s/it]
Training 1/1 epoch (loss 1.6771):  69%|██████▉   | 647/938 [19:07<07:43,  1.59s/it]
Training 1/1 epoch (loss 1.6771):  69%|██████▉   | 648/938 [19:07<08:05,  1.68s/it]
Training 1/1 epoch (loss 1.6607):  69%|██████▉   | 648/938 [19:09<08:05,  1.68s/it]
Training 1/1 epoch (loss 1.6607):  69%|██████▉   | 649/938 [19:09<08:09,  1.69s/it]
Training 1/1 epoch (loss 1.5992):  69%|██████▉   | 649/938 [19:10<08:09,  1.69s/it]
Training 1/1 epoch (loss 1.5992):  69%|██████▉   | 650/938 [19:10<07:51,  1.64s/it]
Training 1/1 epoch (loss 1.6316):  69%|██████▉   | 650/938 [19:12<07:51,  1.64s/it]
Training 1/1 epoch (loss 1.6316):  69%|██████▉   | 651/938 [19:12<07:51,  1.64s/it]
Training 1/1 epoch (loss 1.5788):  69%|██████▉   | 651/938 [19:14<07:51,  1.64s/it]
Training 1/1 epoch (loss 1.5788):  70%|██████▉   | 652/938 [19:14<08:08,  1.71s/it]
Training 1/1 epoch (loss 1.7553):  70%|██████▉   | 652/938 [19:15<08:08,  1.71s/it]
Training 1/1 epoch (loss 1.7553):  70%|██████▉   | 653/938 [19:15<07:28,  1.57s/it]
Training 1/1 epoch (loss 1.6977):  70%|██████▉   | 653/938 [19:16<07:28,  1.57s/it]
Training 1/1 epoch (loss 1.6977):  70%|██████▉   | 654/938 [19:16<07:03,  1.49s/it]
Training 1/1 epoch (loss 1.5974):  70%|██████▉   | 654/938 [19:17<07:03,  1.49s/it]
Training 1/1 epoch (loss 1.5974):  70%|██████▉   | 655/938 [19:17<06:29,  1.38s/it]
Training 1/1 epoch (loss 1.6643):  70%|██████▉   | 655/938 [19:20<06:29,  1.38s/it]
Training 1/1 epoch (loss 1.6643):  70%|██████▉   | 656/938 [19:20<08:00,  1.70s/it]
Training 1/1 epoch (loss 1.7435):  70%|██████▉   | 656/938 [19:21<08:00,  1.70s/it]
Training 1/1 epoch (loss 1.7435):  70%|███████   | 657/938 [19:21<07:18,  1.56s/it]
Training 1/1 epoch (loss 1.6602):  70%|███████   | 657/938 [19:23<07:18,  1.56s/it]
Training 1/1 epoch (loss 1.6602):  70%|███████   | 658/938 [19:23<07:43,  1.66s/it]
Training 1/1 epoch (loss 1.6943):  70%|███████   | 658/938 [19:24<07:43,  1.66s/it]
Training 1/1 epoch (loss 1.6943):  70%|███████   | 659/938 [19:24<07:11,  1.55s/it]
Training 1/1 epoch (loss 1.6833):  70%|███████   | 659/938 [19:26<07:11,  1.55s/it]
Training 1/1 epoch (loss 1.6833):  70%|███████   | 660/938 [19:26<07:40,  1.66s/it]
Training 1/1 epoch (loss 1.7015):  70%|███████   | 660/938 [19:28<07:40,  1.66s/it]
Training 1/1 epoch (loss 1.7015):  70%|███████   | 661/938 [19:28<07:37,  1.65s/it]
Training 1/1 epoch (loss 1.6166):  70%|███████   | 661/938 [19:29<07:37,  1.65s/it]
Training 1/1 epoch (loss 1.6166):  71%|███████   | 662/938 [19:29<07:30,  1.63s/it]
Training 1/1 epoch (loss 1.5359):  71%|███████   | 662/938 [19:31<07:30,  1.63s/it]
Training 1/1 epoch (loss 1.5359):  71%|███████   | 663/938 [19:31<06:55,  1.51s/it]
Training 1/1 epoch (loss 1.6401):  71%|███████   | 663/938 [19:33<06:55,  1.51s/it]
Training 1/1 epoch (loss 1.6401):  71%|███████   | 664/938 [19:33<07:54,  1.73s/it]
Training 1/1 epoch (loss 1.6338):  71%|███████   | 664/938 [19:34<07:54,  1.73s/it]
Training 1/1 epoch (loss 1.6338):  71%|███████   | 665/938 [19:34<07:36,  1.67s/it]
Training 1/1 epoch (loss 1.6029):  71%|███████   | 665/938 [19:36<07:36,  1.67s/it]
Training 1/1 epoch (loss 1.6029):  71%|███████   | 666/938 [19:36<07:42,  1.70s/it]
Training 1/1 epoch (loss 1.6205):  71%|███████   | 666/938 [19:38<07:42,  1.70s/it]
Training 1/1 epoch (loss 1.6205):  71%|███████   | 667/938 [19:38<07:07,  1.58s/it]
Training 1/1 epoch (loss 1.5328):  71%|███████   | 667/938 [19:39<07:07,  1.58s/it]
Training 1/1 epoch (loss 1.5328):  71%|███████   | 668/938 [19:39<06:27,  1.43s/it]
Training 1/1 epoch (loss 1.6066):  71%|███████   | 668/938 [19:41<06:27,  1.43s/it]
Training 1/1 epoch (loss 1.6066):  71%|███████▏  | 669/938 [19:41<07:32,  1.68s/it]
Training 1/1 epoch (loss 1.5146):  71%|███████▏  | 669/938 [19:42<07:32,  1.68s/it]
Training 1/1 epoch (loss 1.5146):  71%|███████▏  | 670/938 [19:42<07:18,  1.64s/it]
Training 1/1 epoch (loss 1.6841):  71%|███████▏  | 670/938 [19:44<07:18,  1.64s/it]
Training 1/1 epoch (loss 1.6841):  72%|███████▏  | 671/938 [19:44<07:46,  1.75s/it]
Training 1/1 epoch (loss 1.7960):  72%|███████▏  | 671/938 [19:47<07:46,  1.75s/it]
Training 1/1 epoch (loss 1.7960):  72%|███████▏  | 672/938 [19:47<08:15,  1.86s/it]
Training 1/1 epoch (loss 1.6020):  72%|███████▏  | 672/938 [19:48<08:15,  1.86s/it]
Training 1/1 epoch (loss 1.6020):  72%|███████▏  | 673/938 [19:48<08:14,  1.87s/it]
Training 1/1 epoch (loss 1.6291):  72%|███████▏  | 673/938 [19:50<08:14,  1.87s/it]
Training 1/1 epoch (loss 1.6291):  72%|███████▏  | 674/938 [19:50<07:28,  1.70s/it]
Training 1/1 epoch (loss 1.6105):  72%|███████▏  | 674/938 [19:51<07:28,  1.70s/it]
Training 1/1 epoch (loss 1.6105):  72%|███████▏  | 675/938 [19:51<07:17,  1.66s/it]
Training 1/1 epoch (loss 1.6379):  72%|███████▏  | 675/938 [19:53<07:17,  1.66s/it]
Training 1/1 epoch (loss 1.6379):  72%|███████▏  | 676/938 [19:53<07:26,  1.71s/it]
Training 1/1 epoch (loss 1.7485):  72%|███████▏  | 676/938 [19:55<07:26,  1.71s/it]
Training 1/1 epoch (loss 1.7485):  72%|███████▏  | 677/938 [19:55<07:05,  1.63s/it]
Training 1/1 epoch (loss 1.6121):  72%|███████▏  | 677/938 [19:56<07:05,  1.63s/it]
Training 1/1 epoch (loss 1.6121):  72%|███████▏  | 678/938 [19:56<06:13,  1.44s/it]
Training 1/1 epoch (loss 1.6232):  72%|███████▏  | 678/938 [19:57<06:13,  1.44s/it]
Training 1/1 epoch (loss 1.6232):  72%|███████▏  | 679/938 [19:57<06:02,  1.40s/it]
Training 1/1 epoch (loss 1.6441):  72%|███████▏  | 679/938 [19:59<06:02,  1.40s/it]
Training 1/1 epoch (loss 1.6441):  72%|███████▏  | 680/938 [19:59<07:26,  1.73s/it]
Training 1/1 epoch (loss 1.6391):  72%|███████▏  | 680/938 [20:01<07:26,  1.73s/it]
Training 1/1 epoch (loss 1.6391):  73%|███████▎  | 681/938 [20:01<07:48,  1.82s/it]
Training 1/1 epoch (loss 1.6488):  73%|███████▎  | 681/938 [20:03<07:48,  1.82s/it]
Training 1/1 epoch (loss 1.6488):  73%|███████▎  | 682/938 [20:03<07:11,  1.68s/it]
Training 1/1 epoch (loss 1.6706):  73%|███████▎  | 682/938 [20:04<07:11,  1.68s/it]
Training 1/1 epoch (loss 1.6706):  73%|███████▎  | 683/938 [20:04<06:32,  1.54s/it]
Training 1/1 epoch (loss 1.6623):  73%|███████▎  | 683/938 [20:05<06:32,  1.54s/it]
Training 1/1 epoch (loss 1.6623):  73%|███████▎  | 684/938 [20:05<06:26,  1.52s/it]
Training 1/1 epoch (loss 1.6514):  73%|███████▎  | 684/938 [20:07<06:26,  1.52s/it]
Training 1/1 epoch (loss 1.6514):  73%|███████▎  | 685/938 [20:07<06:16,  1.49s/it]
Training 1/1 epoch (loss 1.7450):  73%|███████▎  | 685/938 [20:08<06:16,  1.49s/it]
Training 1/1 epoch (loss 1.7450):  73%|███████▎  | 686/938 [20:08<06:04,  1.45s/it]
Training 1/1 epoch (loss 1.6288):  73%|███████▎  | 686/938 [20:10<06:04,  1.45s/it]
Training 1/1 epoch (loss 1.6288):  73%|███████▎  | 687/938 [20:10<06:29,  1.55s/it]
Training 1/1 epoch (loss 1.6545):  73%|███████▎  | 687/938 [20:12<06:29,  1.55s/it]
Training 1/1 epoch (loss 1.6545):  73%|███████▎  | 688/938 [20:12<06:34,  1.58s/it]
Training 1/1 epoch (loss 1.6777):  73%|███████▎  | 688/938 [20:13<06:34,  1.58s/it]
Training 1/1 epoch (loss 1.6777):  73%|███████▎  | 689/938 [20:13<06:38,  1.60s/it]
Training 1/1 epoch (loss 1.6914):  73%|███████▎  | 689/938 [20:15<06:38,  1.60s/it]
Training 1/1 epoch (loss 1.6914):  74%|███████▎  | 690/938 [20:15<06:49,  1.65s/it]
Training 1/1 epoch (loss 1.7137):  74%|███████▎  | 690/938 [20:16<06:49,  1.65s/it]
Training 1/1 epoch (loss 1.7137):  74%|███████▎  | 691/938 [20:16<06:10,  1.50s/it]
Training 1/1 epoch (loss 1.6512):  74%|███████▎  | 691/938 [20:19<06:10,  1.50s/it]
Training 1/1 epoch (loss 1.6512):  74%|███████▍  | 692/938 [20:19<07:21,  1.79s/it]
Training 1/1 epoch (loss 1.5983):  74%|███████▍  | 692/938 [20:21<07:21,  1.79s/it]
Training 1/1 epoch (loss 1.5983):  74%|███████▍  | 693/938 [20:21<07:21,  1.80s/it]
Training 1/1 epoch (loss 1.5683):  74%|███████▍  | 693/938 [20:22<07:21,  1.80s/it]
Training 1/1 epoch (loss 1.5683):  74%|███████▍  | 694/938 [20:22<07:05,  1.74s/it]
Training 1/1 epoch (loss 1.6326):  74%|███████▍  | 694/938 [20:23<07:05,  1.74s/it]
Training 1/1 epoch (loss 1.6326):  74%|███████▍  | 695/938 [20:23<06:13,  1.54s/it]
Training 1/1 epoch (loss 1.4904):  74%|███████▍  | 695/938 [20:25<06:13,  1.54s/it]
Training 1/1 epoch (loss 1.4904):  74%|███████▍  | 696/938 [20:25<06:30,  1.61s/it]
Training 1/1 epoch (loss 1.6970):  74%|███████▍  | 696/938 [20:26<06:30,  1.61s/it]
Training 1/1 epoch (loss 1.6970):  74%|███████▍  | 697/938 [20:26<06:19,  1.57s/it]
Training 1/1 epoch (loss 1.6090):  74%|███████▍  | 697/938 [20:29<06:19,  1.57s/it]
Training 1/1 epoch (loss 1.6090):  74%|███████▍  | 698/938 [20:29<06:56,  1.74s/it]
Training 1/1 epoch (loss 1.5457):  74%|███████▍  | 698/938 [20:30<06:56,  1.74s/it]
Training 1/1 epoch (loss 1.5457):  75%|███████▍  | 699/938 [20:30<06:09,  1.54s/it]
Training 1/1 epoch (loss 1.6609):  75%|███████▍  | 699/938 [20:32<06:09,  1.54s/it]
Training 1/1 epoch (loss 1.6609):  75%|███████▍  | 700/938 [20:32<07:07,  1.80s/it]
Training 1/1 epoch (loss 1.5759):  75%|███████▍  | 700/938 [20:34<07:07,  1.80s/it]
Training 1/1 epoch (loss 1.5759):  75%|███████▍  | 701/938 [20:34<07:21,  1.86s/it]
Training 1/1 epoch (loss 1.5403):  75%|███████▍  | 701/938 [20:36<07:21,  1.86s/it]
Training 1/1 epoch (loss 1.5403):  75%|███████▍  | 702/938 [20:36<07:42,  1.96s/it]
Training 1/1 epoch (loss 1.6956):  75%|███████▍  | 702/938 [20:38<07:42,  1.96s/it]
Training 1/1 epoch (loss 1.6956):  75%|███████▍  | 703/938 [20:38<07:30,  1.92s/it]
Training 1/1 epoch (loss 1.5778):  75%|███████▍  | 703/938 [20:40<07:30,  1.92s/it]
Training 1/1 epoch (loss 1.5778):  75%|███████▌  | 704/938 [20:40<07:03,  1.81s/it]
Training 1/1 epoch (loss 1.6077):  75%|███████▌  | 704/938 [20:42<07:03,  1.81s/it]
Training 1/1 epoch (loss 1.6077):  75%|███████▌  | 705/938 [20:42<07:25,  1.91s/it]
Training 1/1 epoch (loss 1.5508):  75%|███████▌  | 705/938 [20:44<07:25,  1.91s/it]
Training 1/1 epoch (loss 1.5508):  75%|███████▌  | 706/938 [20:44<07:25,  1.92s/it]
Training 1/1 epoch (loss 1.6312):  75%|███████▌  | 706/938 [20:45<07:25,  1.92s/it]
Training 1/1 epoch (loss 1.6312):  75%|███████▌  | 707/938 [20:45<06:56,  1.80s/it]
Training 1/1 epoch (loss 1.6783):  75%|███████▌  | 707/938 [20:47<06:56,  1.80s/it]
Training 1/1 epoch (loss 1.6783):  75%|███████▌  | 708/938 [20:47<06:24,  1.67s/it]
Training 1/1 epoch (loss 1.7160):  75%|███████▌  | 708/938 [20:48<06:24,  1.67s/it]
Training 1/1 epoch (loss 1.7160):  76%|███████▌  | 709/938 [20:48<06:19,  1.66s/it]
Training 1/1 epoch (loss 1.6439):  76%|███████▌  | 709/938 [20:50<06:19,  1.66s/it]
Training 1/1 epoch (loss 1.6439):  76%|███████▌  | 710/938 [20:50<06:02,  1.59s/it]
Training 1/1 epoch (loss 1.7015):  76%|███████▌  | 710/938 [20:51<06:02,  1.59s/it]
Training 1/1 epoch (loss 1.7015):  76%|███████▌  | 711/938 [20:51<06:13,  1.65s/it]
Training 1/1 epoch (loss 1.5984):  76%|███████▌  | 711/938 [20:53<06:13,  1.65s/it]
Training 1/1 epoch (loss 1.5984):  76%|███████▌  | 712/938 [20:53<06:02,  1.60s/it]
Training 1/1 epoch (loss 1.6589):  76%|███████▌  | 712/938 [20:54<06:02,  1.60s/it]
Training 1/1 epoch (loss 1.6589):  76%|███████▌  | 713/938 [20:54<05:43,  1.52s/it]
Training 1/1 epoch (loss 1.6155):  76%|███████▌  | 713/938 [20:57<05:43,  1.52s/it]
Training 1/1 epoch (loss 1.6155):  76%|███████▌  | 714/938 [20:57<06:49,  1.83s/it]
Training 1/1 epoch (loss 1.6212):  76%|███████▌  | 714/938 [20:58<06:49,  1.83s/it]
Training 1/1 epoch (loss 1.6212):  76%|███████▌  | 715/938 [20:58<06:24,  1.72s/it]
Training 1/1 epoch (loss 1.5857):  76%|███████▌  | 715/938 [21:00<06:24,  1.72s/it]
Training 1/1 epoch (loss 1.5857):  76%|███████▋  | 716/938 [21:00<06:07,  1.66s/it]
Training 1/1 epoch (loss 1.6549):  76%|███████▋  | 716/938 [21:02<06:07,  1.66s/it]
Training 1/1 epoch (loss 1.6549):  76%|███████▋  | 717/938 [21:02<06:13,  1.69s/it]
Training 1/1 epoch (loss 1.5916):  76%|███████▋  | 717/938 [21:03<06:13,  1.69s/it]
Training 1/1 epoch (loss 1.5916):  77%|███████▋  | 718/938 [21:03<06:16,  1.71s/it]
Training 1/1 epoch (loss 1.6353):  77%|███████▋  | 718/938 [21:05<06:16,  1.71s/it]
Training 1/1 epoch (loss 1.6353):  77%|███████▋  | 719/938 [21:05<06:30,  1.78s/it]
Training 1/1 epoch (loss 1.5956):  77%|███████▋  | 719/938 [21:08<06:30,  1.78s/it]
Training 1/1 epoch (loss 1.5956):  77%|███████▋  | 720/938 [21:08<07:13,  1.99s/it]
Training 1/1 epoch (loss 1.5728):  77%|███████▋  | 720/938 [21:09<07:13,  1.99s/it]
Training 1/1 epoch (loss 1.5728):  77%|███████▋  | 721/938 [21:09<06:38,  1.84s/it]
Training 1/1 epoch (loss 1.5926):  77%|███████▋  | 721/938 [21:11<06:38,  1.84s/it]
Training 1/1 epoch (loss 1.5926):  77%|███████▋  | 722/938 [21:11<06:26,  1.79s/it]
Training 1/1 epoch (loss 1.4899):  77%|███████▋  | 722/938 [21:13<06:26,  1.79s/it]
Training 1/1 epoch (loss 1.4899):  77%|███████▋  | 723/938 [21:13<06:22,  1.78s/it]
Training 1/1 epoch (loss 1.5451):  77%|███████▋  | 723/938 [21:14<06:22,  1.78s/it]
Training 1/1 epoch (loss 1.5451):  77%|███████▋  | 724/938 [21:14<05:37,  1.58s/it]
Training 1/1 epoch (loss 1.5769):  77%|███████▋  | 724/938 [21:16<05:37,  1.58s/it]
Training 1/1 epoch (loss 1.5769):  77%|███████▋  | 725/938 [21:16<06:25,  1.81s/it]
Training 1/1 epoch (loss 1.5607):  77%|███████▋  | 725/938 [21:18<06:25,  1.81s/it]
Training 1/1 epoch (loss 1.5607):  77%|███████▋  | 726/938 [21:18<05:55,  1.68s/it]
Training 1/1 epoch (loss 1.6548):  77%|███████▋  | 726/938 [21:19<05:55,  1.68s/it]
Training 1/1 epoch (loss 1.6548):  78%|███████▊  | 727/938 [21:19<05:47,  1.65s/it]
Training 1/1 epoch (loss 1.5169):  78%|███████▊  | 727/938 [21:22<05:47,  1.65s/it]
Training 1/1 epoch (loss 1.5169):  78%|███████▊  | 728/938 [21:22<06:50,  1.96s/it]
Training 1/1 epoch (loss 1.5861):  78%|███████▊  | 728/938 [21:24<06:50,  1.96s/it]
Training 1/1 epoch (loss 1.5861):  78%|███████▊  | 729/938 [21:24<06:53,  1.98s/it]
Training 1/1 epoch (loss 1.6404):  78%|███████▊  | 729/938 [21:26<06:53,  1.98s/it]
Training 1/1 epoch (loss 1.6404):  78%|███████▊  | 730/938 [21:26<06:38,  1.91s/it]
Training 1/1 epoch (loss 1.6701):  78%|███████▊  | 730/938 [21:27<06:38,  1.91s/it]
Training 1/1 epoch (loss 1.6701):  78%|███████▊  | 731/938 [21:27<05:46,  1.67s/it]
Training 1/1 epoch (loss 1.6185):  78%|███████▊  | 731/938 [21:28<05:46,  1.67s/it]
Training 1/1 epoch (loss 1.6185):  78%|███████▊  | 732/938 [21:28<05:49,  1.70s/it]
Training 1/1 epoch (loss 1.7940):  78%|███████▊  | 732/938 [21:30<05:49,  1.70s/it]
Training 1/1 epoch (loss 1.7940):  78%|███████▊  | 733/938 [21:30<06:02,  1.77s/it]
Training 1/1 epoch (loss 1.6340):  78%|███████▊  | 733/938 [21:32<06:02,  1.77s/it]
Training 1/1 epoch (loss 1.6340):  78%|███████▊  | 734/938 [21:32<06:11,  1.82s/it]
Training 1/1 epoch (loss 1.6850):  78%|███████▊  | 734/938 [21:34<06:11,  1.82s/it]
Training 1/1 epoch (loss 1.6850):  78%|███████▊  | 735/938 [21:34<06:17,  1.86s/it]
Training 1/1 epoch (loss 1.6135):  78%|███████▊  | 735/938 [21:37<06:17,  1.86s/it]
Training 1/1 epoch (loss 1.6135):  78%|███████▊  | 736/938 [21:37<07:20,  2.18s/it]
Training 1/1 epoch (loss 1.6239):  78%|███████▊  | 736/938 [21:39<07:20,  2.18s/it]
Training 1/1 epoch (loss 1.6239):  79%|███████▊  | 737/938 [21:39<07:02,  2.10s/it]
Training 1/1 epoch (loss 1.6005):  79%|███████▊  | 737/938 [21:41<07:02,  2.10s/it]
Training 1/1 epoch (loss 1.6005):  79%|███████▊  | 738/938 [21:41<06:31,  1.96s/it]
Training 1/1 epoch (loss 1.5277):  79%|███████▊  | 738/938 [21:42<06:31,  1.96s/it]
Training 1/1 epoch (loss 1.5277):  79%|███████▉  | 739/938 [21:42<05:49,  1.76s/it]
Training 1/1 epoch (loss 1.6243):  79%|███████▉  | 739/938 [21:44<05:49,  1.76s/it]
Training 1/1 epoch (loss 1.6243):  79%|███████▉  | 740/938 [21:44<05:42,  1.73s/it]
Training 1/1 epoch (loss 1.6485):  79%|███████▉  | 740/938 [21:45<05:42,  1.73s/it]
Training 1/1 epoch (loss 1.6485):  79%|███████▉  | 741/938 [21:45<05:36,  1.71s/it]
Training 1/1 epoch (loss 1.6249):  79%|███████▉  | 741/938 [21:47<05:36,  1.71s/it]
Training 1/1 epoch (loss 1.6249):  79%|███████▉  | 742/938 [21:47<05:19,  1.63s/it]
Training 1/1 epoch (loss 1.6400):  79%|███████▉  | 742/938 [21:49<05:19,  1.63s/it]
Training 1/1 epoch (loss 1.6400):  79%|███████▉  | 743/938 [21:49<05:29,  1.69s/it]
Training 1/1 epoch (loss 1.6029):  79%|███████▉  | 743/938 [21:51<05:29,  1.69s/it]
Training 1/1 epoch (loss 1.6029):  79%|███████▉  | 744/938 [21:51<05:39,  1.75s/it]
Training 1/1 epoch (loss 1.6937):  79%|███████▉  | 744/938 [21:52<05:39,  1.75s/it]
Training 1/1 epoch (loss 1.6937):  79%|███████▉  | 745/938 [21:52<05:15,  1.64s/it]
Training 1/1 epoch (loss 1.5794):  79%|███████▉  | 745/938 [21:54<05:15,  1.64s/it]
Training 1/1 epoch (loss 1.5794):  80%|███████▉  | 746/938 [21:54<05:23,  1.68s/it]
Training 1/1 epoch (loss 1.7296):  80%|███████▉  | 746/938 [21:56<05:23,  1.68s/it]
Training 1/1 epoch (loss 1.7296):  80%|███████▉  | 747/938 [21:56<05:47,  1.82s/it]
Training 1/1 epoch (loss 1.5961):  80%|███████▉  | 747/938 [21:57<05:47,  1.82s/it]
Training 1/1 epoch (loss 1.5961):  80%|███████▉  | 748/938 [21:57<05:24,  1.71s/it]
Training 1/1 epoch (loss 1.5481):  80%|███████▉  | 748/938 [21:59<05:24,  1.71s/it]
Training 1/1 epoch (loss 1.5481):  80%|███████▉  | 749/938 [21:59<05:34,  1.77s/it]
Training 1/1 epoch (loss 1.5574):  80%|███████▉  | 749/938 [22:01<05:34,  1.77s/it]
Training 1/1 epoch (loss 1.5574):  80%|███████▉  | 750/938 [22:01<05:09,  1.64s/it]
Training 1/1 epoch (loss 1.6031):  80%|███████▉  | 750/938 [22:03<05:09,  1.64s/it]
Training 1/1 epoch (loss 1.6031):  80%|████████  | 751/938 [22:03<05:37,  1.80s/it]
Training 1/1 epoch (loss 1.6198):  80%|████████  | 751/938 [22:04<05:37,  1.80s/it]
Training 1/1 epoch (loss 1.6198):  80%|████████  | 752/938 [22:04<05:29,  1.77s/it]
Training 1/1 epoch (loss 1.6611):  80%|████████  | 752/938 [22:06<05:29,  1.77s/it]
Training 1/1 epoch (loss 1.6611):  80%|████████  | 753/938 [22:06<05:39,  1.83s/it]
Training 1/1 epoch (loss 1.6502):  80%|████████  | 753/938 [22:08<05:39,  1.83s/it]
Training 1/1 epoch (loss 1.6502):  80%|████████  | 754/938 [22:08<05:03,  1.65s/it]
Training 1/1 epoch (loss 1.6332):  80%|████████  | 754/938 [22:09<05:03,  1.65s/it]
Training 1/1 epoch (loss 1.6332):  80%|████████  | 755/938 [22:09<04:44,  1.55s/it]
Training 1/1 epoch (loss 1.6245):  80%|████████  | 755/938 [22:10<04:44,  1.55s/it]
Training 1/1 epoch (loss 1.6245):  81%|████████  | 756/938 [22:10<04:28,  1.47s/it]
Training 1/1 epoch (loss 1.5710):  81%|████████  | 756/938 [22:12<04:28,  1.47s/it]
Training 1/1 epoch (loss 1.5710):  81%|████████  | 757/938 [22:12<04:53,  1.62s/it]
Training 1/1 epoch (loss 1.7167):  81%|████████  | 757/938 [22:14<04:53,  1.62s/it]
Training 1/1 epoch (loss 1.7167):  81%|████████  | 758/938 [22:14<04:56,  1.65s/it]
Training 1/1 epoch (loss 1.6289):  81%|████████  | 758/938 [22:15<04:56,  1.65s/it]
Training 1/1 epoch (loss 1.6289):  81%|████████  | 759/938 [22:15<04:45,  1.60s/it]
Training 1/1 epoch (loss 1.4599):  81%|████████  | 759/938 [22:17<04:45,  1.60s/it]
Training 1/1 epoch (loss 1.4599):  81%|████████  | 760/938 [22:17<04:56,  1.66s/it]
Training 1/1 epoch (loss 1.7722):  81%|████████  | 760/938 [22:19<04:56,  1.66s/it]
Training 1/1 epoch (loss 1.7722):  81%|████████  | 761/938 [22:19<04:37,  1.57s/it]
Training 1/1 epoch (loss 1.5909):  81%|████████  | 761/938 [22:20<04:37,  1.57s/it]
Training 1/1 epoch (loss 1.5909):  81%|████████  | 762/938 [22:20<04:43,  1.61s/it]
Training 1/1 epoch (loss 1.7242):  81%|████████  | 762/938 [22:23<04:43,  1.61s/it]
Training 1/1 epoch (loss 1.7242):  81%|████████▏ | 763/938 [22:23<05:24,  1.86s/it]
Training 1/1 epoch (loss 1.4632):  81%|████████▏ | 763/938 [22:24<05:24,  1.86s/it]
Training 1/1 epoch (loss 1.4632):  81%|████████▏ | 764/938 [22:24<05:09,  1.78s/it]
Training 1/1 epoch (loss 1.6840):  81%|████████▏ | 764/938 [22:27<05:09,  1.78s/it]
Training 1/1 epoch (loss 1.6840):  82%|████████▏ | 765/938 [22:27<05:32,  1.92s/it]
Training 1/1 epoch (loss 1.6136):  82%|████████▏ | 765/938 [22:28<05:32,  1.92s/it]
Training 1/1 epoch (loss 1.6136):  82%|████████▏ | 766/938 [22:28<05:04,  1.77s/it]
Training 1/1 epoch (loss 1.6828):  82%|████████▏ | 766/938 [22:30<05:04,  1.77s/it]
Training 1/1 epoch (loss 1.6828):  82%|████████▏ | 767/938 [22:30<05:40,  1.99s/it]
Training 1/1 epoch (loss 1.5523):  82%|████████▏ | 767/938 [22:32<05:40,  1.99s/it]
Training 1/1 epoch (loss 1.5523):  82%|████████▏ | 768/938 [22:32<05:19,  1.88s/it]
Training 1/1 epoch (loss 1.8273):  82%|████████▏ | 768/938 [22:34<05:19,  1.88s/it]
Training 1/1 epoch (loss 1.8273):  82%|████████▏ | 769/938 [22:34<04:56,  1.76s/it]
Training 1/1 epoch (loss 1.6319):  82%|████████▏ | 769/938 [22:35<04:56,  1.76s/it]
Training 1/1 epoch (loss 1.6319):  82%|████████▏ | 770/938 [22:35<04:33,  1.63s/it]
Training 1/1 epoch (loss 1.5363):  82%|████████▏ | 770/938 [22:36<04:33,  1.63s/it]
Training 1/1 epoch (loss 1.5363):  82%|████████▏ | 771/938 [22:36<04:22,  1.57s/it]
Training 1/1 epoch (loss 1.5323):  82%|████████▏ | 771/938 [22:37<04:22,  1.57s/it]
Training 1/1 epoch (loss 1.5323):  82%|████████▏ | 772/938 [22:37<03:55,  1.42s/it]
Training 1/1 epoch (loss 1.5366):  82%|████████▏ | 772/938 [22:39<03:55,  1.42s/it]
Training 1/1 epoch (loss 1.5366):  82%|████████▏ | 773/938 [22:39<03:54,  1.42s/it]
Training 1/1 epoch (loss 1.5064):  82%|████████▏ | 773/938 [22:41<03:54,  1.42s/it]
Training 1/1 epoch (loss 1.5064):  83%|████████▎ | 774/938 [22:41<04:25,  1.62s/it]
Training 1/1 epoch (loss 1.6167):  83%|████████▎ | 774/938 [22:43<04:25,  1.62s/it]
Training 1/1 epoch (loss 1.6167):  83%|████████▎ | 775/938 [22:43<05:03,  1.86s/it]
Training 1/1 epoch (loss 1.7102):  83%|████████▎ | 775/938 [22:46<05:03,  1.86s/it]
Training 1/1 epoch (loss 1.7102):  83%|████████▎ | 776/938 [22:46<05:25,  2.01s/it]
Training 1/1 epoch (loss 1.6233):  83%|████████▎ | 776/938 [22:48<05:25,  2.01s/it]
Training 1/1 epoch (loss 1.6233):  83%|████████▎ | 777/938 [22:48<05:30,  2.05s/it]
Training 1/1 epoch (loss 1.5811):  83%|████████▎ | 777/938 [22:50<05:30,  2.05s/it]
Training 1/1 epoch (loss 1.5811):  83%|████████▎ | 778/938 [22:50<05:37,  2.11s/it]
Training 1/1 epoch (loss 1.5642):  83%|████████▎ | 778/938 [22:52<05:37,  2.11s/it]
Training 1/1 epoch (loss 1.5642):  83%|████████▎ | 779/938 [22:52<05:16,  1.99s/it]
Training 1/1 epoch (loss 1.7016):  83%|████████▎ | 779/938 [22:54<05:16,  1.99s/it]
Training 1/1 epoch (loss 1.7016):  83%|████████▎ | 780/938 [22:54<05:23,  2.05s/it]
Training 1/1 epoch (loss 1.6875):  83%|████████▎ | 780/938 [22:56<05:23,  2.05s/it]
Training 1/1 epoch (loss 1.6875):  83%|████████▎ | 781/938 [22:56<05:22,  2.05s/it]
Training 1/1 epoch (loss 1.5724):  83%|████████▎ | 781/938 [22:57<05:22,  2.05s/it]
Training 1/1 epoch (loss 1.5724):  83%|████████▎ | 782/938 [22:57<04:53,  1.88s/it]
Training 1/1 epoch (loss 1.6424):  83%|████████▎ | 782/938 [22:59<04:53,  1.88s/it]
Training 1/1 epoch (loss 1.6424):  83%|████████▎ | 783/938 [22:59<04:33,  1.76s/it]
Training 1/1 epoch (loss 1.6039):  83%|████████▎ | 783/938 [23:01<04:33,  1.76s/it]
Training 1/1 epoch (loss 1.6039):  84%|████████▎ | 784/938 [23:01<04:36,  1.80s/it]
Training 1/1 epoch (loss 1.6263):  84%|████████▎ | 784/938 [23:03<04:36,  1.80s/it]
Training 1/1 epoch (loss 1.6263):  84%|████████▎ | 785/938 [23:03<04:53,  1.92s/it]
Training 1/1 epoch (loss 1.6376):  84%|████████▎ | 785/938 [23:05<04:53,  1.92s/it]
Training 1/1 epoch (loss 1.6376):  84%|████████▍ | 786/938 [23:05<04:42,  1.86s/it]
Training 1/1 epoch (loss 1.5653):  84%|████████▍ | 786/938 [23:06<04:42,  1.86s/it]
Training 1/1 epoch (loss 1.5653):  84%|████████▍ | 787/938 [23:06<04:02,  1.60s/it]
Training 1/1 epoch (loss 1.7475):  84%|████████▍ | 787/938 [23:07<04:02,  1.60s/it]
Training 1/1 epoch (loss 1.7475):  84%|████████▍ | 788/938 [23:07<03:42,  1.49s/it]
Training 1/1 epoch (loss 1.5016):  84%|████████▍ | 788/938 [23:09<03:42,  1.49s/it]
Training 1/1 epoch (loss 1.5016):  84%|████████▍ | 789/938 [23:09<04:10,  1.68s/it]
Training 1/1 epoch (loss 1.5116):  84%|████████▍ | 789/938 [23:11<04:10,  1.68s/it]
Training 1/1 epoch (loss 1.5116):  84%|████████▍ | 790/938 [23:11<04:27,  1.81s/it]
Training 1/1 epoch (loss 1.6499):  84%|████████▍ | 790/938 [23:13<04:27,  1.81s/it]
Training 1/1 epoch (loss 1.6499):  84%|████████▍ | 791/938 [23:13<04:23,  1.79s/it]
Training 1/1 epoch (loss 1.5790):  84%|████████▍ | 791/938 [23:15<04:23,  1.79s/it]
Training 1/1 epoch (loss 1.5790):  84%|████████▍ | 792/938 [23:15<04:41,  1.93s/it]
Training 1/1 epoch (loss 1.6589):  84%|████████▍ | 792/938 [23:17<04:41,  1.93s/it]
Training 1/1 epoch (loss 1.6589):  85%|████████▍ | 793/938 [23:17<04:21,  1.81s/it]
Training 1/1 epoch (loss 1.6847):  85%|████████▍ | 793/938 [23:19<04:21,  1.81s/it]
Training 1/1 epoch (loss 1.6847):  85%|████████▍ | 794/938 [23:19<04:33,  1.90s/it]
Training 1/1 epoch (loss 1.5705):  85%|████████▍ | 794/938 [23:21<04:33,  1.90s/it]
Training 1/1 epoch (loss 1.5705):  85%|████████▍ | 795/938 [23:21<04:23,  1.85s/it]
Training 1/1 epoch (loss 1.5955):  85%|████████▍ | 795/938 [23:22<04:23,  1.85s/it]
Training 1/1 epoch (loss 1.5955):  85%|████████▍ | 796/938 [23:22<04:17,  1.81s/it]
Training 1/1 epoch (loss 1.5617):  85%|████████▍ | 796/938 [23:24<04:17,  1.81s/it]
Training 1/1 epoch (loss 1.5617):  85%|████████▍ | 797/938 [23:24<04:02,  1.72s/it]
Training 1/1 epoch (loss 1.6692):  85%|████████▍ | 797/938 [23:26<04:02,  1.72s/it]
Training 1/1 epoch (loss 1.6692):  85%|████████▌ | 798/938 [23:26<04:07,  1.77s/it]
Training 1/1 epoch (loss 1.6722):  85%|████████▌ | 798/938 [23:28<04:07,  1.77s/it]
Training 1/1 epoch (loss 1.6722):  85%|████████▌ | 799/938 [23:28<04:37,  1.99s/it]
Training 1/1 epoch (loss 1.6106):  85%|████████▌ | 799/938 [23:30<04:37,  1.99s/it]
Training 1/1 epoch (loss 1.6106):  85%|████████▌ | 800/938 [23:30<04:25,  1.92s/it]
Training 1/1 epoch (loss 1.6886):  85%|████████▌ | 800/938 [23:31<04:25,  1.92s/it]
Training 1/1 epoch (loss 1.6886):  85%|████████▌ | 801/938 [23:31<04:02,  1.77s/it]
Training 1/1 epoch (loss 1.6853):  85%|████████▌ | 801/938 [23:32<04:02,  1.77s/it]
Training 1/1 epoch (loss 1.6853):  86%|████████▌ | 802/938 [23:32<03:32,  1.56s/it]
Training 1/1 epoch (loss 1.6326):  86%|████████▌ | 802/938 [23:34<03:32,  1.56s/it]
Training 1/1 epoch (loss 1.6326):  86%|████████▌ | 803/938 [23:34<03:19,  1.48s/it]
Training 1/1 epoch (loss 1.5779):  86%|████████▌ | 803/938 [23:36<03:19,  1.48s/it]
Training 1/1 epoch (loss 1.5779):  86%|████████▌ | 804/938 [23:36<03:47,  1.70s/it]
Training 1/1 epoch (loss 1.6227):  86%|████████▌ | 804/938 [23:38<03:47,  1.70s/it]
Training 1/1 epoch (loss 1.6227):  86%|████████▌ | 805/938 [23:38<04:18,  1.94s/it]
Training 1/1 epoch (loss 1.5845):  86%|████████▌ | 805/938 [23:40<04:18,  1.94s/it]
Training 1/1 epoch (loss 1.5845):  86%|████████▌ | 806/938 [23:40<03:57,  1.80s/it]
Training 1/1 epoch (loss 1.6335):  86%|████████▌ | 806/938 [23:42<03:57,  1.80s/it]
Training 1/1 epoch (loss 1.6335):  86%|████████▌ | 807/938 [23:42<04:10,  1.91s/it]
Training 1/1 epoch (loss 1.6343):  86%|████████▌ | 807/938 [23:44<04:10,  1.91s/it]
Training 1/1 epoch (loss 1.6343):  86%|████████▌ | 808/938 [23:44<04:14,  1.96s/it]
Training 1/1 epoch (loss 1.5133):  86%|████████▌ | 808/938 [23:46<04:14,  1.96s/it]
Training 1/1 epoch (loss 1.5133):  86%|████████▌ | 809/938 [23:46<04:09,  1.94s/it]
Training 1/1 epoch (loss 1.6376):  86%|████████▌ | 809/938 [23:47<04:09,  1.94s/it]
Training 1/1 epoch (loss 1.6376):  86%|████████▋ | 810/938 [23:47<03:40,  1.72s/it]
Training 1/1 epoch (loss 1.5454):  86%|████████▋ | 810/938 [23:50<03:40,  1.72s/it]
Training 1/1 epoch (loss 1.5454):  86%|████████▋ | 811/938 [23:50<04:05,  1.93s/it]
Training 1/1 epoch (loss 1.6229):  86%|████████▋ | 811/938 [23:51<04:05,  1.93s/it]
Training 1/1 epoch (loss 1.6229):  87%|████████▋ | 812/938 [23:51<03:31,  1.68s/it]
Training 1/1 epoch (loss 1.6652):  87%|████████▋ | 812/938 [23:53<03:31,  1.68s/it]
Training 1/1 epoch (loss 1.6652):  87%|████████▋ | 813/938 [23:53<03:45,  1.80s/it]
Training 1/1 epoch (loss 1.6262):  87%|████████▋ | 813/938 [23:54<03:45,  1.80s/it]
Training 1/1 epoch (loss 1.6262):  87%|████████▋ | 814/938 [23:54<03:24,  1.65s/it]
Training 1/1 epoch (loss 1.6030):  87%|████████▋ | 814/938 [23:56<03:24,  1.65s/it]
Training 1/1 epoch (loss 1.6030):  87%|████████▋ | 815/938 [23:56<03:44,  1.82s/it]
Training 1/1 epoch (loss 1.6373):  87%|████████▋ | 815/938 [23:58<03:44,  1.82s/it]
Training 1/1 epoch (loss 1.6373):  87%|████████▋ | 816/938 [23:58<03:50,  1.89s/it]
Training 1/1 epoch (loss 1.6218):  87%|████████▋ | 816/938 [24:00<03:50,  1.89s/it]
Training 1/1 epoch (loss 1.6218):  87%|████████▋ | 817/938 [24:00<03:27,  1.71s/it]
Training 1/1 epoch (loss 1.5518):  87%|████████▋ | 817/938 [24:02<03:27,  1.71s/it]
Training 1/1 epoch (loss 1.5518):  87%|████████▋ | 818/938 [24:02<03:29,  1.74s/it]
Training 1/1 epoch (loss 1.7403):  87%|████████▋ | 818/938 [24:03<03:29,  1.74s/it]
Training 1/1 epoch (loss 1.7403):  87%|████████▋ | 819/938 [24:03<03:22,  1.70s/it]
Training 1/1 epoch (loss 1.5657):  87%|████████▋ | 819/938 [24:04<03:22,  1.70s/it]
Training 1/1 epoch (loss 1.5657):  87%|████████▋ | 820/938 [24:04<03:01,  1.54s/it]
Training 1/1 epoch (loss 1.6311):  87%|████████▋ | 820/938 [24:06<03:01,  1.54s/it]
Training 1/1 epoch (loss 1.6311):  88%|████████▊ | 821/938 [24:06<03:04,  1.58s/it]
Training 1/1 epoch (loss 1.7462):  88%|████████▊ | 821/938 [24:07<03:04,  1.58s/it]
Training 1/1 epoch (loss 1.7462):  88%|████████▊ | 822/938 [24:07<02:53,  1.49s/it]
Training 1/1 epoch (loss 1.5769):  88%|████████▊ | 822/938 [24:10<02:53,  1.49s/it]
Training 1/1 epoch (loss 1.5769):  88%|████████▊ | 823/938 [24:10<03:26,  1.79s/it]
Training 1/1 epoch (loss 1.5556):  88%|████████▊ | 823/938 [24:12<03:26,  1.79s/it]
Training 1/1 epoch (loss 1.5556):  88%|████████▊ | 824/938 [24:12<03:48,  2.01s/it]
Training 1/1 epoch (loss 1.5467):  88%|████████▊ | 824/938 [24:14<03:48,  2.01s/it]
Training 1/1 epoch (loss 1.5467):  88%|████████▊ | 825/938 [24:14<03:31,  1.87s/it]
Training 1/1 epoch (loss 1.6188):  88%|████████▊ | 825/938 [24:15<03:31,  1.87s/it]
Training 1/1 epoch (loss 1.6188):  88%|████████▊ | 826/938 [24:15<03:11,  1.71s/it]
Training 1/1 epoch (loss 1.6148):  88%|████████▊ | 826/938 [24:17<03:11,  1.71s/it]
Training 1/1 epoch (loss 1.6148):  88%|████████▊ | 827/938 [24:17<03:18,  1.79s/it]
Training 1/1 epoch (loss 1.6826):  88%|████████▊ | 827/938 [24:18<03:18,  1.79s/it]
Training 1/1 epoch (loss 1.6826):  88%|████████▊ | 828/938 [24:18<02:51,  1.56s/it]
Training 1/1 epoch (loss 1.6352):  88%|████████▊ | 828/938 [24:19<02:51,  1.56s/it]
Training 1/1 epoch (loss 1.6352):  88%|████████▊ | 829/938 [24:19<02:41,  1.48s/it]
Training 1/1 epoch (loss 1.6645):  88%|████████▊ | 829/938 [24:21<02:41,  1.48s/it]
Training 1/1 epoch (loss 1.6645):  88%|████████▊ | 830/938 [24:21<02:50,  1.58s/it]
Training 1/1 epoch (loss 1.7079):  88%|████████▊ | 830/938 [24:23<02:50,  1.58s/it]
Training 1/1 epoch (loss 1.7079):  89%|████████▊ | 831/938 [24:23<03:05,  1.73s/it]
Training 1/1 epoch (loss 1.6748):  89%|████████▊ | 831/938 [24:25<03:05,  1.73s/it]
Training 1/1 epoch (loss 1.6748):  89%|████████▊ | 832/938 [24:25<02:51,  1.61s/it]
Training 1/1 epoch (loss 1.6288):  89%|████████▊ | 832/938 [24:27<02:51,  1.61s/it]
Training 1/1 epoch (loss 1.6288):  89%|████████▉ | 833/938 [24:27<03:13,  1.84s/it]
Training 1/1 epoch (loss 1.6097):  89%|████████▉ | 833/938 [24:29<03:13,  1.84s/it]
Training 1/1 epoch (loss 1.6097):  89%|████████▉ | 834/938 [24:29<03:10,  1.84s/it]
Training 1/1 epoch (loss 1.5737):  89%|████████▉ | 834/938 [24:31<03:10,  1.84s/it]
Training 1/1 epoch (loss 1.5737):  89%|████████▉ | 835/938 [24:31<03:09,  1.84s/it]
Training 1/1 epoch (loss 1.5851):  89%|████████▉ | 835/938 [24:32<03:09,  1.84s/it]
Training 1/1 epoch (loss 1.5851):  89%|████████▉ | 836/938 [24:32<02:51,  1.68s/it]
Training 1/1 epoch (loss 1.7441):  89%|████████▉ | 836/938 [24:34<02:51,  1.68s/it]
Training 1/1 epoch (loss 1.7441):  89%|████████▉ | 837/938 [24:34<02:44,  1.63s/it]
Training 1/1 epoch (loss 1.5846):  89%|████████▉ | 837/938 [24:36<02:44,  1.63s/it]
Training 1/1 epoch (loss 1.5846):  89%|████████▉ | 838/938 [24:36<02:52,  1.73s/it]
Training 1/1 epoch (loss 1.5801):  89%|████████▉ | 838/938 [24:38<02:52,  1.73s/it]
Training 1/1 epoch (loss 1.5801):  89%|████████▉ | 839/938 [24:38<03:10,  1.93s/it]
Training 1/1 epoch (loss 1.5368):  89%|████████▉ | 839/938 [24:40<03:10,  1.93s/it]
Training 1/1 epoch (loss 1.5368):  90%|████████▉ | 840/938 [24:40<03:20,  2.04s/it]
Training 1/1 epoch (loss 1.6890):  90%|████████▉ | 840/938 [24:43<03:20,  2.04s/it]
Training 1/1 epoch (loss 1.6890):  90%|████████▉ | 841/938 [24:43<03:32,  2.19s/it]
Training 1/1 epoch (loss 1.5645):  90%|████████▉ | 841/938 [24:44<03:32,  2.19s/it]
Training 1/1 epoch (loss 1.5645):  90%|████████▉ | 842/938 [24:44<03:03,  1.91s/it]
Training 1/1 epoch (loss 1.6793):  90%|████████▉ | 842/938 [24:46<03:03,  1.91s/it]
Training 1/1 epoch (loss 1.6793):  90%|████████▉ | 843/938 [24:46<03:06,  1.96s/it]
Training 1/1 epoch (loss 1.5022):  90%|████████▉ | 843/938 [24:49<03:06,  1.96s/it]
Training 1/1 epoch (loss 1.5022):  90%|████████▉ | 844/938 [24:49<03:16,  2.10s/it]
Training 1/1 epoch (loss 1.6861):  90%|████████▉ | 844/938 [24:51<03:16,  2.10s/it]
Training 1/1 epoch (loss 1.6861):  90%|█████████ | 845/938 [24:51<03:12,  2.07s/it]
Training 1/1 epoch (loss 1.7327):  90%|█████████ | 845/938 [24:52<03:12,  2.07s/it]
Training 1/1 epoch (loss 1.7327):  90%|█████████ | 846/938 [24:52<02:49,  1.84s/it]
Training 1/1 epoch (loss 1.5982):  90%|█████████ | 846/938 [24:54<02:49,  1.84s/it]
Training 1/1 epoch (loss 1.5982):  90%|█████████ | 847/938 [24:54<02:47,  1.84s/it]
Training 1/1 epoch (loss 1.5562):  90%|█████████ | 847/938 [24:55<02:47,  1.84s/it]
Training 1/1 epoch (loss 1.5562):  90%|█████████ | 848/938 [24:55<02:37,  1.75s/it]
Training 1/1 epoch (loss 1.6165):  90%|█████████ | 848/938 [24:57<02:37,  1.75s/it]
Training 1/1 epoch (loss 1.6165):  91%|█████████ | 849/938 [24:57<02:41,  1.81s/it]
Training 1/1 epoch (loss 1.5538):  91%|█████████ | 849/938 [24:58<02:41,  1.81s/it]
Training 1/1 epoch (loss 1.5538):  91%|█████████ | 850/938 [24:58<02:25,  1.66s/it]
Training 1/1 epoch (loss 1.6753):  91%|█████████ | 850/938 [25:01<02:25,  1.66s/it]
Training 1/1 epoch (loss 1.6753):  91%|█████████ | 851/938 [25:01<02:38,  1.83s/it]
Training 1/1 epoch (loss 1.7139):  91%|█████████ | 851/938 [25:02<02:38,  1.83s/it]
Training 1/1 epoch (loss 1.7139):  91%|█████████ | 852/938 [25:02<02:32,  1.78s/it]
Training 1/1 epoch (loss 1.6367):  91%|█████████ | 852/938 [25:04<02:32,  1.78s/it]
Training 1/1 epoch (loss 1.6367):  91%|█████████ | 853/938 [25:04<02:37,  1.86s/it]
Training 1/1 epoch (loss 1.5918):  91%|█████████ | 853/938 [25:06<02:37,  1.86s/it]
Training 1/1 epoch (loss 1.5918):  91%|█████████ | 854/938 [25:06<02:17,  1.64s/it]
Training 1/1 epoch (loss 1.6636):  91%|█████████ | 854/938 [25:07<02:17,  1.64s/it]
Training 1/1 epoch (loss 1.6636):  91%|█████████ | 855/938 [25:07<02:21,  1.71s/it]
Training 1/1 epoch (loss 1.6128):  91%|█████████ | 855/938 [25:09<02:21,  1.71s/it]
Training 1/1 epoch (loss 1.6128):  91%|█████████▏| 856/938 [25:09<02:14,  1.64s/it]
Training 1/1 epoch (loss 1.6664):  91%|█████████▏| 856/938 [25:11<02:14,  1.64s/it]
Training 1/1 epoch (loss 1.6664):  91%|█████████▏| 857/938 [25:11<02:32,  1.89s/it]
Training 1/1 epoch (loss 1.5991):  91%|█████████▏| 857/938 [25:12<02:32,  1.89s/it]
Training 1/1 epoch (loss 1.5991):  91%|█████████▏| 858/938 [25:12<02:12,  1.66s/it]
Training 1/1 epoch (loss 1.6348):  91%|█████████▏| 858/938 [25:14<02:12,  1.66s/it]
Training 1/1 epoch (loss 1.6348):  92%|█████████▏| 859/938 [25:14<02:04,  1.58s/it]
Training 1/1 epoch (loss 1.5268):  92%|█████████▏| 859/938 [25:15<02:04,  1.58s/it]
Training 1/1 epoch (loss 1.5268):  92%|█████████▏| 860/938 [25:15<01:47,  1.38s/it]
Training 1/1 epoch (loss 1.6037):  92%|█████████▏| 860/938 [25:17<01:47,  1.38s/it]
Training 1/1 epoch (loss 1.6037):  92%|█████████▏| 861/938 [25:17<02:00,  1.57s/it]
Training 1/1 epoch (loss 1.6754):  92%|█████████▏| 861/938 [25:19<02:00,  1.57s/it]
Training 1/1 epoch (loss 1.6754):  92%|█████████▏| 862/938 [25:19<02:02,  1.62s/it]
Training 1/1 epoch (loss 1.7184):  92%|█████████▏| 862/938 [25:21<02:02,  1.62s/it]
Training 1/1 epoch (loss 1.7184):  92%|█████████▏| 863/938 [25:21<02:19,  1.86s/it]
Training 1/1 epoch (loss 1.5992):  92%|█████████▏| 863/938 [25:22<02:19,  1.86s/it]
Training 1/1 epoch (loss 1.5992):  92%|█████████▏| 864/938 [25:22<02:07,  1.73s/it]
Training 1/1 epoch (loss 1.6667):  92%|█████████▏| 864/938 [25:24<02:07,  1.73s/it]
Training 1/1 epoch (loss 1.6667):  92%|█████████▏| 865/938 [25:24<02:00,  1.65s/it]
Training 1/1 epoch (loss 1.4673):  92%|█████████▏| 865/938 [25:26<02:00,  1.65s/it]
Training 1/1 epoch (loss 1.4673):  92%|█████████▏| 866/938 [25:26<02:04,  1.73s/it]
Training 1/1 epoch (loss 1.4345):  92%|█████████▏| 866/938 [25:27<02:04,  1.73s/it]
Training 1/1 epoch (loss 1.4345):  92%|█████████▏| 867/938 [25:27<01:50,  1.55s/it]
Training 1/1 epoch (loss 1.5307):  92%|█████████▏| 867/938 [25:28<01:50,  1.55s/it]
Training 1/1 epoch (loss 1.5307):  93%|█████████▎| 868/938 [25:28<01:47,  1.54s/it]
Training 1/1 epoch (loss 1.5825):  93%|█████████▎| 868/938 [25:31<01:47,  1.54s/it]
Training 1/1 epoch (loss 1.5825):  93%|█████████▎| 869/938 [25:31<02:05,  1.82s/it]
Training 1/1 epoch (loss 1.7159):  93%|█████████▎| 869/938 [25:33<02:05,  1.82s/it]
Training 1/1 epoch (loss 1.7159):  93%|█████████▎| 870/938 [25:33<02:01,  1.79s/it]
Training 1/1 epoch (loss 1.5632):  93%|█████████▎| 870/938 [25:34<02:01,  1.79s/it]
Training 1/1 epoch (loss 1.5632):  93%|█████████▎| 871/938 [25:34<01:56,  1.73s/it]
Training 1/1 epoch (loss 1.6163):  93%|█████████▎| 871/938 [25:36<01:56,  1.73s/it]
Training 1/1 epoch (loss 1.6163):  93%|█████████▎| 872/938 [25:36<01:51,  1.68s/it]
Training 1/1 epoch (loss 1.7090):  93%|█████████▎| 872/938 [25:37<01:51,  1.68s/it]
Training 1/1 epoch (loss 1.7090):  93%|█████████▎| 873/938 [25:37<01:45,  1.62s/it]
Training 1/1 epoch (loss 1.5975):  93%|█████████▎| 873/938 [25:39<01:45,  1.62s/it]
Training 1/1 epoch (loss 1.5975):  93%|█████████▎| 874/938 [25:39<01:49,  1.71s/it]
Training 1/1 epoch (loss 1.6858):  93%|█████████▎| 874/938 [25:41<01:49,  1.71s/it]
Training 1/1 epoch (loss 1.6858):  93%|█████████▎| 875/938 [25:41<01:41,  1.62s/it]
Training 1/1 epoch (loss 1.6241):  93%|█████████▎| 875/938 [25:42<01:41,  1.62s/it]
Training 1/1 epoch (loss 1.6241):  93%|█████████▎| 876/938 [25:42<01:38,  1.59s/it]
Training 1/1 epoch (loss 1.6043):  93%|█████████▎| 876/938 [25:44<01:38,  1.59s/it]
Training 1/1 epoch (loss 1.6043):  93%|█████████▎| 877/938 [25:44<01:33,  1.53s/it]
Training 1/1 epoch (loss 1.5642):  93%|█████████▎| 877/938 [25:45<01:33,  1.53s/it]
Training 1/1 epoch (loss 1.5642):  94%|█████████▎| 878/938 [25:45<01:26,  1.44s/it]
Training 1/1 epoch (loss 1.5492):  94%|█████████▎| 878/938 [25:47<01:26,  1.44s/it]
Training 1/1 epoch (loss 1.5492):  94%|█████████▎| 879/938 [25:47<01:30,  1.54s/it]
Training 1/1 epoch (loss 1.7440):  94%|█████████▎| 879/938 [25:48<01:30,  1.54s/it]
Training 1/1 epoch (loss 1.7440):  94%|█████████▍| 880/938 [25:48<01:33,  1.62s/it]
Training 1/1 epoch (loss 1.6050):  94%|█████████▍| 880/938 [25:51<01:33,  1.62s/it]
Training 1/1 epoch (loss 1.6050):  94%|█████████▍| 881/938 [25:51<01:42,  1.80s/it]
Training 1/1 epoch (loss 1.6441):  94%|█████████▍| 881/938 [25:52<01:42,  1.80s/it]
Training 1/1 epoch (loss 1.6441):  94%|█████████▍| 882/938 [25:52<01:30,  1.62s/it]
Training 1/1 epoch (loss 1.6818):  94%|█████████▍| 882/938 [25:54<01:30,  1.62s/it]
Training 1/1 epoch (loss 1.6818):  94%|█████████▍| 883/938 [25:54<01:35,  1.74s/it]
Training 1/1 epoch (loss 1.6821):  94%|█████████▍| 883/938 [25:56<01:35,  1.74s/it]
Training 1/1 epoch (loss 1.6821):  94%|█████████▍| 884/938 [25:56<01:34,  1.75s/it]
Training 1/1 epoch (loss 1.6445):  94%|█████████▍| 884/938 [25:58<01:34,  1.75s/it]
Training 1/1 epoch (loss 1.6445):  94%|█████████▍| 885/938 [25:58<01:43,  1.95s/it]
Training 1/1 epoch (loss 1.5409):  94%|█████████▍| 885/938 [25:59<01:43,  1.95s/it]
Training 1/1 epoch (loss 1.5409):  94%|█████████▍| 886/938 [25:59<01:33,  1.80s/it]
Training 1/1 epoch (loss 1.6092):  94%|█████████▍| 886/938 [26:01<01:33,  1.80s/it]
Training 1/1 epoch (loss 1.6092):  95%|█████████▍| 887/938 [26:01<01:26,  1.71s/it]
Training 1/1 epoch (loss 1.5736):  95%|█████████▍| 887/938 [26:03<01:26,  1.71s/it]
Training 1/1 epoch (loss 1.5736):  95%|█████████▍| 888/938 [26:03<01:37,  1.94s/it]
Training 1/1 epoch (loss 1.6421):  95%|█████████▍| 888/938 [26:06<01:37,  1.94s/it]
Training 1/1 epoch (loss 1.6421):  95%|█████████▍| 889/938 [26:06<01:42,  2.09s/it]
Training 1/1 epoch (loss 1.5172):  95%|█████████▍| 889/938 [26:08<01:42,  2.09s/it]
Training 1/1 epoch (loss 1.5172):  95%|█████████▍| 890/938 [26:08<01:40,  2.10s/it]
Training 1/1 epoch (loss 1.6062):  95%|█████████▍| 890/938 [26:10<01:40,  2.10s/it]
Training 1/1 epoch (loss 1.6062):  95%|█████████▍| 891/938 [26:10<01:40,  2.14s/it]
Training 1/1 epoch (loss 1.6629):  95%|█████████▍| 891/938 [26:12<01:40,  2.14s/it]
Training 1/1 epoch (loss 1.6629):  95%|█████████▌| 892/938 [26:12<01:27,  1.90s/it]
Training 1/1 epoch (loss 1.6867):  95%|█████████▌| 892/938 [26:13<01:27,  1.90s/it]
Training 1/1 epoch (loss 1.6867):  95%|█████████▌| 893/938 [26:13<01:17,  1.73s/it]
Training 1/1 epoch (loss 1.5468):  95%|█████████▌| 893/938 [26:15<01:17,  1.73s/it]
Training 1/1 epoch (loss 1.5468):  95%|█████████▌| 894/938 [26:15<01:16,  1.73s/it]
Training 1/1 epoch (loss 1.6193):  95%|█████████▌| 894/938 [26:16<01:16,  1.73s/it]
Training 1/1 epoch (loss 1.6193):  95%|█████████▌| 895/938 [26:16<01:09,  1.62s/it]
Training 1/1 epoch (loss 1.5666):  95%|█████████▌| 895/938 [26:18<01:09,  1.62s/it]
Training 1/1 epoch (loss 1.5666):  96%|█████████▌| 896/938 [26:18<01:11,  1.71s/it]
Training 1/1 epoch (loss 1.5504):  96%|█████████▌| 896/938 [26:20<01:11,  1.71s/it]
Training 1/1 epoch (loss 1.5504):  96%|█████████▌| 897/938 [26:20<01:12,  1.78s/it]
Training 1/1 epoch (loss 1.6028):  96%|█████████▌| 897/938 [26:21<01:12,  1.78s/it]
Training 1/1 epoch (loss 1.6028):  96%|█████████▌| 898/938 [26:21<01:04,  1.60s/it]
Training 1/1 epoch (loss 1.6251):  96%|█████████▌| 898/938 [26:23<01:04,  1.60s/it]
Training 1/1 epoch (loss 1.6251):  96%|█████████▌| 899/938 [26:23<01:11,  1.84s/it]
Training 1/1 epoch (loss 1.6016):  96%|█████████▌| 899/938 [26:25<01:11,  1.84s/it]
Training 1/1 epoch (loss 1.6016):  96%|█████████▌| 900/938 [26:25<01:05,  1.74s/it]
Training 1/1 epoch (loss 1.5155):  96%|█████████▌| 900/938 [26:27<01:05,  1.74s/it]
Training 1/1 epoch (loss 1.5155):  96%|█████████▌| 901/938 [26:27<01:11,  1.93s/it]
Training 1/1 epoch (loss 1.5319):  96%|█████████▌| 901/938 [26:28<01:11,  1.93s/it]
Training 1/1 epoch (loss 1.5319):  96%|█████████▌| 902/938 [26:28<01:01,  1.72s/it]
Training 1/1 epoch (loss 1.5948):  96%|█████████▌| 902/938 [26:30<01:01,  1.72s/it]
Training 1/1 epoch (loss 1.5948):  96%|█████████▋| 903/938 [26:30<01:01,  1.74s/it]
Training 1/1 epoch (loss 1.6712):  96%|█████████▋| 903/938 [26:32<01:01,  1.74s/it]
Training 1/1 epoch (loss 1.6712):  96%|█████████▋| 904/938 [26:32<01:01,  1.81s/it]
Training 1/1 epoch (loss 1.6584):  96%|█████████▋| 904/938 [26:33<01:01,  1.81s/it]
Training 1/1 epoch (loss 1.6584):  96%|█████████▋| 905/938 [26:33<00:50,  1.54s/it]
Training 1/1 epoch (loss 1.6454):  96%|█████████▋| 905/938 [26:35<00:50,  1.54s/it]
Training 1/1 epoch (loss 1.6454):  97%|█████████▋| 906/938 [26:35<00:50,  1.57s/it]
Training 1/1 epoch (loss 1.5435):  97%|█████████▋| 906/938 [26:37<00:50,  1.57s/it]
Training 1/1 epoch (loss 1.5435):  97%|█████████▋| 907/938 [26:37<00:57,  1.85s/it]
Training 1/1 epoch (loss 1.6056):  97%|█████████▋| 907/938 [26:39<00:57,  1.85s/it]
Training 1/1 epoch (loss 1.6056):  97%|█████████▋| 908/938 [26:39<00:56,  1.88s/it]
Training 1/1 epoch (loss 1.5718):  97%|█████████▋| 908/938 [26:40<00:56,  1.88s/it]
Training 1/1 epoch (loss 1.5718):  97%|█████████▋| 909/938 [26:40<00:48,  1.67s/it]
Training 1/1 epoch (loss 1.6430):  97%|█████████▋| 909/938 [26:42<00:48,  1.67s/it]
Training 1/1 epoch (loss 1.6430):  97%|█████████▋| 910/938 [26:42<00:45,  1.63s/it]
Training 1/1 epoch (loss 1.6004):  97%|█████████▋| 910/938 [26:44<00:45,  1.63s/it]
Training 1/1 epoch (loss 1.6004):  97%|█████████▋| 911/938 [26:44<00:43,  1.60s/it]
Training 1/1 epoch (loss 1.6304):  97%|█████████▋| 911/938 [26:46<00:43,  1.60s/it]
Training 1/1 epoch (loss 1.6304):  97%|█████████▋| 912/938 [26:46<00:44,  1.73s/it]
Training 1/1 epoch (loss 1.6232):  97%|█████████▋| 912/938 [26:47<00:44,  1.73s/it]
Training 1/1 epoch (loss 1.6232):  97%|█████████▋| 913/938 [26:47<00:38,  1.55s/it]
Training 1/1 epoch (loss 1.6282):  97%|█████████▋| 913/938 [26:48<00:38,  1.55s/it]
Training 1/1 epoch (loss 1.6282):  97%|█████████▋| 914/938 [26:48<00:37,  1.56s/it]
Training 1/1 epoch (loss 1.5943):  97%|█████████▋| 914/938 [26:51<00:37,  1.56s/it]
Training 1/1 epoch (loss 1.5943):  98%|█████████▊| 915/938 [26:51<00:41,  1.82s/it]
Training 1/1 epoch (loss 1.6319):  98%|█████████▊| 915/938 [26:52<00:41,  1.82s/it]
Training 1/1 epoch (loss 1.6319):  98%|█████████▊| 916/938 [26:52<00:38,  1.75s/it]
Training 1/1 epoch (loss 1.6005):  98%|█████████▊| 916/938 [26:55<00:38,  1.75s/it]
Training 1/1 epoch (loss 1.6005):  98%|█████████▊| 917/938 [26:55<00:41,  2.00s/it]
Training 1/1 epoch (loss 1.5844):  98%|█████████▊| 917/938 [26:57<00:41,  2.00s/it]
Training 1/1 epoch (loss 1.5844):  98%|█████████▊| 918/938 [26:57<00:39,  1.96s/it]
Training 1/1 epoch (loss 1.5915):  98%|█████████▊| 918/938 [26:59<00:39,  1.96s/it]
Training 1/1 epoch (loss 1.5915):  98%|█████████▊| 919/938 [26:59<00:37,  2.00s/it]
Training 1/1 epoch (loss 1.5817):  98%|█████████▊| 919/938 [27:00<00:37,  2.00s/it]
Training 1/1 epoch (loss 1.5817):  98%|█████████▊| 920/938 [27:00<00:33,  1.88s/it]
Training 1/1 epoch (loss 1.5877):  98%|█████████▊| 920/938 [27:02<00:33,  1.88s/it]
Training 1/1 epoch (loss 1.5877):  98%|█████████▊| 921/938 [27:02<00:31,  1.83s/it]
Training 1/1 epoch (loss 1.5393):  98%|█████████▊| 921/938 [27:05<00:31,  1.83s/it]
Training 1/1 epoch (loss 1.5393):  98%|█████████▊| 922/938 [27:05<00:32,  2.02s/it]
Training 1/1 epoch (loss 1.5404):  98%|█████████▊| 922/938 [27:06<00:32,  2.02s/it]
Training 1/1 epoch (loss 1.5404):  98%|█████████▊| 923/938 [27:06<00:28,  1.89s/it]
Training 1/1 epoch (loss 1.6016):  98%|█████████▊| 923/938 [27:07<00:28,  1.89s/it]
Training 1/1 epoch (loss 1.6016):  99%|█████████▊| 924/938 [27:07<00:22,  1.63s/it]
Training 1/1 epoch (loss 1.6482):  99%|█████████▊| 924/938 [27:09<00:22,  1.63s/it]
Training 1/1 epoch (loss 1.6482):  99%|█████████▊| 925/938 [27:09<00:21,  1.65s/it]
Training 1/1 epoch (loss 1.4983):  99%|█████████▊| 925/938 [27:10<00:21,  1.65s/it]
Training 1/1 epoch (loss 1.4983):  99%|█████████▊| 926/938 [27:10<00:19,  1.64s/it]
Training 1/1 epoch (loss 1.6209):  99%|█████████▊| 926/938 [27:11<00:19,  1.64s/it]
Training 1/1 epoch (loss 1.6209):  99%|█████████▉| 927/938 [27:11<00:15,  1.43s/it]
Training 1/1 epoch (loss 1.5850):  99%|█████████▉| 927/938 [27:13<00:15,  1.43s/it]
Training 1/1 epoch (loss 1.5850):  99%|█████████▉| 928/938 [27:13<00:14,  1.43s/it]
Training 1/1 epoch (loss 1.5641):  99%|█████████▉| 928/938 [27:15<00:14,  1.43s/it]
Training 1/1 epoch (loss 1.5641):  99%|█████████▉| 929/938 [27:15<00:14,  1.59s/it]
Training 1/1 epoch (loss 1.5814):  99%|█████████▉| 929/938 [27:17<00:14,  1.59s/it]
Training 1/1 epoch (loss 1.5814):  99%|█████████▉| 930/938 [27:17<00:13,  1.69s/it]
Training 1/1 epoch (loss 1.5632):  99%|█████████▉| 930/938 [27:18<00:13,  1.69s/it]
Training 1/1 epoch (loss 1.5632):  99%|█████████▉| 931/938 [27:18<00:10,  1.56s/it]
Training 1/1 epoch (loss 1.6210):  99%|█████████▉| 931/938 [27:20<00:10,  1.56s/it]
Training 1/1 epoch (loss 1.6210):  99%|█████████▉| 932/938 [27:20<00:09,  1.57s/it]
Training 1/1 epoch (loss 1.6144):  99%|█████████▉| 932/938 [27:21<00:09,  1.57s/it]
Training 1/1 epoch (loss 1.6144):  99%|█████████▉| 933/938 [27:21<00:08,  1.60s/it]
Training 1/1 epoch (loss 1.6112):  99%|█████████▉| 933/938 [27:23<00:08,  1.60s/it]
Training 1/1 epoch (loss 1.6112): 100%|█████████▉| 934/938 [27:23<00:06,  1.70s/it]
Training 1/1 epoch (loss 1.7452): 100%|█████████▉| 934/938 [27:25<00:06,  1.70s/it]
Training 1/1 epoch (loss 1.7452): 100%|█████████▉| 935/938 [27:25<00:04,  1.59s/it]
Training 1/1 epoch (loss 1.6199): 100%|█████████▉| 935/938 [27:26<00:04,  1.59s/it]
Training 1/1 epoch (loss 1.6199): 100%|█████████▉| 936/938 [27:26<00:03,  1.66s/it]
Training 1/1 epoch (loss 1.5855): 100%|█████████▉| 936/938 [27:28<00:03,  1.66s/it]
Training 1/1 epoch (loss 1.5855): 100%|█████████▉| 937/938 [27:28<00:01,  1.63s/it]
Training 1/1 epoch (loss 1.6134): 100%|█████████▉| 937/938 [27:30<00:01,  1.63s/it]
Training 1/1 epoch (loss 1.6134): 100%|██████████| 938/938 [27:30<00:00,  1.69s/it]
Training 1/1 epoch (loss 1.6134): 100%|██████████| 938/938 [27:30<00:00,  1.76s/it]
tokenizer config file saved in /aifs4su/hansirui_1st/boyuan/resist/setting3-safety/tinyllama-1T/tinyllama-1T-s3-Q1-30k/tokenizer_config.json
Special tokens file saved in /aifs4su/hansirui_1st/boyuan/resist/setting3-safety/tinyllama-1T/tinyllama-1T-s3-Q1-30k/special_tokens_map.json
wandb: ERROR Problem finishing run
Exception ignored in atexit callback: <bound method rank_zero_only.<locals>.wrapper of <safe_rlhf.logger.Logger object at 0x15505f9e3a90>>
Traceback (most recent call last):
  File "/home/hansirui_1st/jiayi/resist/setting3/safe_rlhf/utils.py", line 212, in wrapper
    return func(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^
  File "/home/hansirui_1st/jiayi/resist/setting3/safe_rlhf/logger.py", line 183, in close
    self.wandb.finish()
  File "/aifs4su/hansirui_1st/miniconda3/envs/by-align/lib/python3.11/site-packages/wandb/sdk/wandb_run.py", line 449, in wrapper
    return func(self, *args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/aifs4su/hansirui_1st/miniconda3/envs/by-align/lib/python3.11/site-packages/wandb/sdk/wandb_run.py", line 391, in wrapper
    return func(self, *args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/aifs4su/hansirui_1st/miniconda3/envs/by-align/lib/python3.11/site-packages/wandb/sdk/wandb_run.py", line 2106, in finish
    return self._finish(exit_code)
           ^^^^^^^^^^^^^^^^^^^^^^^
  File "/aifs4su/hansirui_1st/miniconda3/envs/by-align/lib/python3.11/site-packages/wandb/sdk/wandb_run.py", line 2127, in _finish
    self._atexit_cleanup(exit_code=exit_code)
  File "/aifs4su/hansirui_1st/miniconda3/envs/by-align/lib/python3.11/site-packages/wandb/sdk/wandb_run.py", line 2352, in _atexit_cleanup
    self._on_finish()
  File "/aifs4su/hansirui_1st/miniconda3/envs/by-align/lib/python3.11/site-packages/wandb/sdk/wandb_run.py", line 2609, in _on_finish
    wait_with_progress(
  File "/aifs4su/hansirui_1st/miniconda3/envs/by-align/lib/python3.11/site-packages/wandb/sdk/mailbox/wait_with_progress.py", line 24, in wait_with_progress
    return wait_all_with_progress(
           ^^^^^^^^^^^^^^^^^^^^^^^
  File "/aifs4su/hansirui_1st/miniconda3/envs/by-align/lib/python3.11/site-packages/wandb/sdk/mailbox/wait_with_progress.py", line 87, in wait_all_with_progress
    return asyncio_compat.run(progress_loop_with_timeout)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/aifs4su/hansirui_1st/miniconda3/envs/by-align/lib/python3.11/site-packages/wandb/sdk/lib/asyncio_compat.py", line 27, in run
    future = executor.submit(runner.run, fn)
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/aifs4su/hansirui_1st/miniconda3/envs/by-align/lib/python3.11/concurrent/futures/thread.py", line 169, in submit
    raise RuntimeError('cannot schedule new futures after '
RuntimeError: cannot schedule new futures after interpreter shutdown