Upload 79 files
Browse filesThis view is limited to 50 files because it contains too many changes.
See raw diff
- swin_default_LR1e-2_addEOSToken/config__2024_01_17--11_16_51.toml +89 -0
- swin_default_LR1e-2_addEOSToken/config__2024_01_17--12_14_37.toml +89 -0
- swin_default_LR1e-2_addEOSToken/config__2024_01_17--17_16_44.toml +89 -0
- swin_default_LR1e-2_addEOSToken/swin_default_LR1e-2_addEOSToken_2024_01_17--11_16_21.log +142 -0
- swin_default_LR1e-2_addEOSToken/swin_default_LR1e-2_addEOSToken_2024_01_17--12_14_33.log +130 -0
- swin_default_LR1e-2_addEOSToken/swin_default_LR1e-2_addEOSToken_2024_01_17--17_15_22.log +203 -0
- swin_default_LR1e-2_addEOSToken/tb_log/events.out.tfevents.1705450617.gina1.1498943.0 +3 -0
- swin_default_LR1e-2_addEOSToken/tb_log/events.out.tfevents.1705450617.gina1.1498944.0 +3 -0
- swin_default_LR1e-2_addEOSToken/tb_log/events.out.tfevents.1705450617.gina1.1498945.0 +3 -0
- swin_default_LR1e-2_addEOSToken/tb_log/events.out.tfevents.1705450617.gina1.1498946.0 +3 -0
- swin_default_LR1e-2_addEOSToken/tb_log/events.out.tfevents.1705454078.gina301.610519.0 +3 -0
- swin_default_LR1e-2_addEOSToken/tb_log/events.out.tfevents.1705454078.gina301.610520.0 +3 -0
- swin_default_LR1e-2_addEOSToken/tb_log/events.out.tfevents.1705454078.gina301.610521.0 +3 -0
- swin_default_LR1e-2_addEOSToken/tb_log/events.out.tfevents.1705454078.gina301.610522.0 +3 -0
- swin_default_LR1e-2_addEOSToken/tb_log/events.out.tfevents.1705472223.gina1.1576669.0 +3 -0
- swin_default_LR1e-2_addEOSToken/tb_log/events.out.tfevents.1705472223.gina1.1576670.0 +3 -0
- swin_default_LR1e-2_addEOSToken/tb_log/events.out.tfevents.1705472223.gina1.1576671.0 +3 -0
- swin_default_LR1e-2_addEOSToken/tb_log/events.out.tfevents.1705472223.gina1.1576672.0 +3 -0
- swin_default_LR1e-3_AR-NAR/config__2024_01_18--00_22_05.toml +89 -0
- swin_default_LR1e-3_AR-NAR/config__2024_01_18--00_23_20.toml +89 -0
- swin_default_LR1e-3_AR-NAR/config__2024_01_18--00_33_25.toml +89 -0
- swin_default_LR1e-3_AR-NAR/config__2024_01_18--13_13_46.toml +89 -0
- swin_default_LR1e-3_AR-NAR/config__2024_01_18--13_14_43.toml +89 -0
- swin_default_LR1e-3_AR-NAR/swin_default_LR1e-2_AR-NAR_2024_01_18--00_21_17.log +85 -0
- swin_default_LR1e-3_AR-NAR/swin_default_LR1e-2_AR-NAR_2024_01_18--00_23_16.log +85 -0
- swin_default_LR1e-3_AR-NAR/swin_default_LR1e-2_AR-NAR_2024_01_18--00_33_02.log +515 -0
- swin_default_LR1e-3_AR-NAR/swin_default_LR1e-3_AR-NAR_2024_01_18--13_13_05.log +94 -0
- swin_default_LR1e-3_AR-NAR/swin_default_LR1e-3_AR-NAR_2024_01_18--13_14_39.log +121 -0
- swin_default_LR1e-3_AR-NAR/swin_default_LR1e-3_AR-NAR_2024_01_18--13_14_40.log +1 -0
- swin_default_LR1e-3_AR-NAR/tb_log/events.out.tfevents.1705497738.gina1.1685133.0 +3 -0
- swin_default_LR1e-3_AR-NAR/tb_log/events.out.tfevents.1705497738.gina1.1685134.0 +3 -0
- swin_default_LR1e-3_AR-NAR/tb_log/events.out.tfevents.1705497738.gina1.1685135.0 +3 -0
- swin_default_LR1e-3_AR-NAR/tb_log/events.out.tfevents.1705497738.gina1.1685136.0 +3 -0
- swin_default_LR1e-3_AR-NAR/tb_log/events.out.tfevents.1705497802.gina1.1688371.0 +3 -0
- swin_default_LR1e-3_AR-NAR/tb_log/events.out.tfevents.1705497802.gina1.1688372.0 +3 -0
- swin_default_LR1e-3_AR-NAR/tb_log/events.out.tfevents.1705497802.gina1.1688373.0 +3 -0
- swin_default_LR1e-3_AR-NAR/tb_log/events.out.tfevents.1705497802.gina1.1688374.0 +3 -0
- swin_default_LR1e-3_AR-NAR/tb_log/events.out.tfevents.1705498412.gina1.1693065.0 +3 -0
- swin_default_LR1e-3_AR-NAR/tb_log/events.out.tfevents.1705498412.gina1.1693066.0 +3 -0
- swin_default_LR1e-3_AR-NAR/tb_log/events.out.tfevents.1705498412.gina1.1693067.0 +3 -0
- swin_default_LR1e-3_AR-NAR/tb_log/events.out.tfevents.1705498412.gina1.1693068.0 +3 -0
- swin_default_LR1e-3_AR-NAR/tb_log/events.out.tfevents.1705544038.gina1.1902904.0 +3 -0
- swin_default_LR1e-3_AR-NAR/tb_log/events.out.tfevents.1705544038.gina1.1902905.0 +3 -0
- swin_default_LR1e-3_AR-NAR/tb_log/events.out.tfevents.1705544038.gina1.1902906.0 +3 -0
- swin_default_LR1e-3_AR-NAR/tb_log/events.out.tfevents.1705544038.gina1.1902907.0 +3 -0
- swin_default_LR1e-3_AR-NAR/tb_log/events.out.tfevents.1705544084.gina1.1906280.0 +3 -0
- swin_default_LR1e-3_AR-NAR/tb_log/events.out.tfevents.1705544085.gina1.1906278.0 +3 -0
- swin_default_LR1e-3_AR-NAR/tb_log/events.out.tfevents.1705544085.gina1.1906279.0 +3 -0
- swin_default_LR1e-3_AR-NAR/tb_log/events.out.tfevents.1705544085.gina1.1906281.0 +3 -0
- swin_default_LR1e-4_AR-NAR/config__2024_01_18--13_07_07.toml +89 -0
swin_default_LR1e-2_addEOSToken/config__2024_01_17--11_16_51.toml
ADDED
|
@@ -0,0 +1,89 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[[validate_dataset]]
|
| 2 |
+
path = "dataloader.Dataset"
|
| 3 |
+
|
| 4 |
+
[validate_dataset.args]
|
| 5 |
+
librispeech_dir = "~/data/LibriSpeech/LibriSpeech"
|
| 6 |
+
librispeech_metadata_fpath = "/home/xhao/proj/audiozen/recipes/librimix_sot/local/metadata/LibriSpeech/train-clean-100-24K.csv"
|
| 7 |
+
duration = 6.0
|
| 8 |
+
sr = 24000
|
| 9 |
+
num_samples = 10
|
| 10 |
+
[validate_dataset.dataloader]
|
| 11 |
+
batch_size = 1
|
| 12 |
+
num_workers = 1
|
| 13 |
+
|
| 14 |
+
[meta]
|
| 15 |
+
save_dir = "exp"
|
| 16 |
+
description = "Train a model using Generative Adversarial Networks (GANs)"
|
| 17 |
+
seed = 20220815
|
| 18 |
+
exp_id = "swin_default_LR1e-2_addEOSToken"
|
| 19 |
+
config_path = "/fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/conf/swin_default_LR1e-2_addEOSToken.toml"
|
| 20 |
+
|
| 21 |
+
[trainer]
|
| 22 |
+
path = "trainer.Trainer"
|
| 23 |
+
|
| 24 |
+
[loss_function]
|
| 25 |
+
path = "torch.nn.MSELoss"
|
| 26 |
+
|
| 27 |
+
[optimizer]
|
| 28 |
+
path = "torch.optim.AdamW"
|
| 29 |
+
|
| 30 |
+
[model]
|
| 31 |
+
path = "model.Model"
|
| 32 |
+
|
| 33 |
+
[acoustics]
|
| 34 |
+
n_fft = 512
|
| 35 |
+
hop_length = 128
|
| 36 |
+
win_length = 512
|
| 37 |
+
sr = 24000
|
| 38 |
+
|
| 39 |
+
[train_dataset]
|
| 40 |
+
path = "dataloader.Dataset"
|
| 41 |
+
|
| 42 |
+
[test_dataset]
|
| 43 |
+
path = "dataloader.Dataset"
|
| 44 |
+
|
| 45 |
+
[trainer.args]
|
| 46 |
+
debug = false
|
| 47 |
+
max_steps = 0
|
| 48 |
+
max_epochs = 1000
|
| 49 |
+
max_grad_norm = 1.0
|
| 50 |
+
save_max_score = true
|
| 51 |
+
save_ckpt_interval = 5
|
| 52 |
+
max_patience = 200
|
| 53 |
+
plot_norm = true
|
| 54 |
+
validation_interval = 200
|
| 55 |
+
max_num_checkpoints = 100
|
| 56 |
+
scheduler_name = "constant_schedule_with_warmup"
|
| 57 |
+
warmup_steps = 1000
|
| 58 |
+
warmup_ratio = 0.0
|
| 59 |
+
gradient_accumulation_steps = 1
|
| 60 |
+
|
| 61 |
+
[loss_function.args]
|
| 62 |
+
|
| 63 |
+
[optimizer.args]
|
| 64 |
+
lr = 0.0001
|
| 65 |
+
|
| 66 |
+
[model.args]
|
| 67 |
+
|
| 68 |
+
[train_dataset.args]
|
| 69 |
+
librispeech_dir = "~/data/LibriSpeech/LibriSpeech"
|
| 70 |
+
librispeech_metadata_fpath = "/home/xhao/proj/audiozen/recipes/librimix_sot/local/metadata/LibriSpeech/train-clean-100-24K.csv"
|
| 71 |
+
duration = 6.0
|
| 72 |
+
sr = 24000
|
| 73 |
+
|
| 74 |
+
[train_dataset.dataloader]
|
| 75 |
+
batch_size = 20
|
| 76 |
+
num_workers = 10
|
| 77 |
+
drop_last = true
|
| 78 |
+
pin_memory = true
|
| 79 |
+
|
| 80 |
+
[test_dataset.args]
|
| 81 |
+
librispeech_dir = "~/data/LibriSpeech/LibriSpeech"
|
| 82 |
+
librispeech_metadata_fpath = "/home/xhao/proj/audiozen/recipes/librimix_sot/local/metadata/LibriSpeech/train-clean-100-24K.csv"
|
| 83 |
+
duration = 6.0
|
| 84 |
+
sr = 24000
|
| 85 |
+
num_samples = 10
|
| 86 |
+
|
| 87 |
+
[test_dataset.dataloader]
|
| 88 |
+
batch_size = 1
|
| 89 |
+
num_workers = 1
|
swin_default_LR1e-2_addEOSToken/config__2024_01_17--12_14_37.toml
ADDED
|
@@ -0,0 +1,89 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[[validate_dataset]]
|
| 2 |
+
path = "dataloader.Dataset"
|
| 3 |
+
|
| 4 |
+
[validate_dataset.args]
|
| 5 |
+
librispeech_dir = "~/data/LibriSpeech/LibriSpeech"
|
| 6 |
+
librispeech_metadata_fpath = "/home/xhao/proj/audiozen/recipes/librimix_sot/local/metadata/LibriSpeech/train-clean-100-24K.csv"
|
| 7 |
+
duration = 6.0
|
| 8 |
+
sr = 24000
|
| 9 |
+
num_samples = 10
|
| 10 |
+
[validate_dataset.dataloader]
|
| 11 |
+
batch_size = 1
|
| 12 |
+
num_workers = 1
|
| 13 |
+
|
| 14 |
+
[meta]
|
| 15 |
+
save_dir = "exp"
|
| 16 |
+
description = "Train a model using Generative Adversarial Networks (GANs)"
|
| 17 |
+
seed = 20220815
|
| 18 |
+
exp_id = "swin_default_LR1e-2_addEOSToken"
|
| 19 |
+
config_path = "/fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/conf/swin_default_LR1e-2_addEOSToken.toml"
|
| 20 |
+
|
| 21 |
+
[trainer]
|
| 22 |
+
path = "trainer.Trainer"
|
| 23 |
+
|
| 24 |
+
[loss_function]
|
| 25 |
+
path = "torch.nn.MSELoss"
|
| 26 |
+
|
| 27 |
+
[optimizer]
|
| 28 |
+
path = "torch.optim.AdamW"
|
| 29 |
+
|
| 30 |
+
[model]
|
| 31 |
+
path = "model.Model"
|
| 32 |
+
|
| 33 |
+
[acoustics]
|
| 34 |
+
n_fft = 512
|
| 35 |
+
hop_length = 128
|
| 36 |
+
win_length = 512
|
| 37 |
+
sr = 24000
|
| 38 |
+
|
| 39 |
+
[train_dataset]
|
| 40 |
+
path = "dataloader.Dataset"
|
| 41 |
+
|
| 42 |
+
[test_dataset]
|
| 43 |
+
path = "dataloader.Dataset"
|
| 44 |
+
|
| 45 |
+
[trainer.args]
|
| 46 |
+
debug = false
|
| 47 |
+
max_steps = 0
|
| 48 |
+
max_epochs = 1000
|
| 49 |
+
max_grad_norm = 1.0
|
| 50 |
+
save_max_score = true
|
| 51 |
+
save_ckpt_interval = 5
|
| 52 |
+
max_patience = 200
|
| 53 |
+
plot_norm = true
|
| 54 |
+
validation_interval = 200
|
| 55 |
+
max_num_checkpoints = 100
|
| 56 |
+
scheduler_name = "constant_schedule_with_warmup"
|
| 57 |
+
warmup_steps = 1000
|
| 58 |
+
warmup_ratio = 0.0
|
| 59 |
+
gradient_accumulation_steps = 1
|
| 60 |
+
|
| 61 |
+
[loss_function.args]
|
| 62 |
+
|
| 63 |
+
[optimizer.args]
|
| 64 |
+
lr = 0.0001
|
| 65 |
+
|
| 66 |
+
[model.args]
|
| 67 |
+
|
| 68 |
+
[train_dataset.args]
|
| 69 |
+
librispeech_dir = "~/data/LibriSpeech/LibriSpeech"
|
| 70 |
+
librispeech_metadata_fpath = "/home/xhao/proj/audiozen/recipes/librimix_sot/local/metadata/LibriSpeech/train-clean-100-24K.csv"
|
| 71 |
+
duration = 6.0
|
| 72 |
+
sr = 24000
|
| 73 |
+
|
| 74 |
+
[train_dataset.dataloader]
|
| 75 |
+
batch_size = 20
|
| 76 |
+
num_workers = 10
|
| 77 |
+
drop_last = true
|
| 78 |
+
pin_memory = true
|
| 79 |
+
|
| 80 |
+
[test_dataset.args]
|
| 81 |
+
librispeech_dir = "~/data/LibriSpeech/LibriSpeech"
|
| 82 |
+
librispeech_metadata_fpath = "/home/xhao/proj/audiozen/recipes/librimix_sot/local/metadata/LibriSpeech/train-clean-100-24K.csv"
|
| 83 |
+
duration = 6.0
|
| 84 |
+
sr = 24000
|
| 85 |
+
num_samples = 10
|
| 86 |
+
|
| 87 |
+
[test_dataset.dataloader]
|
| 88 |
+
batch_size = 1
|
| 89 |
+
num_workers = 1
|
swin_default_LR1e-2_addEOSToken/config__2024_01_17--17_16_44.toml
ADDED
|
@@ -0,0 +1,89 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[[validate_dataset]]
|
| 2 |
+
path = "dataloader.Dataset"
|
| 3 |
+
|
| 4 |
+
[validate_dataset.args]
|
| 5 |
+
librispeech_dir = "~/data/LibriSpeech/LibriSpeech"
|
| 6 |
+
librispeech_metadata_fpath = "/home/xhao/proj/audiozen/recipes/librimix_sot/local/metadata/LibriSpeech/train-clean-100-24K.csv"
|
| 7 |
+
duration = 6.0
|
| 8 |
+
sr = 24000
|
| 9 |
+
num_samples = 10
|
| 10 |
+
[validate_dataset.dataloader]
|
| 11 |
+
batch_size = 1
|
| 12 |
+
num_workers = 1
|
| 13 |
+
|
| 14 |
+
[meta]
|
| 15 |
+
save_dir = "exp"
|
| 16 |
+
description = "Train a model using Generative Adversarial Networks (GANs)"
|
| 17 |
+
seed = 20220815
|
| 18 |
+
exp_id = "swin_default_LR1e-2_addEOSToken"
|
| 19 |
+
config_path = "/fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/conf/swin_default_LR1e-2_addEOSToken.toml"
|
| 20 |
+
|
| 21 |
+
[trainer]
|
| 22 |
+
path = "trainer.Trainer"
|
| 23 |
+
|
| 24 |
+
[loss_function]
|
| 25 |
+
path = "torch.nn.MSELoss"
|
| 26 |
+
|
| 27 |
+
[optimizer]
|
| 28 |
+
path = "torch.optim.AdamW"
|
| 29 |
+
|
| 30 |
+
[model]
|
| 31 |
+
path = "model.Model"
|
| 32 |
+
|
| 33 |
+
[acoustics]
|
| 34 |
+
n_fft = 512
|
| 35 |
+
hop_length = 128
|
| 36 |
+
win_length = 512
|
| 37 |
+
sr = 24000
|
| 38 |
+
|
| 39 |
+
[train_dataset]
|
| 40 |
+
path = "dataloader.Dataset"
|
| 41 |
+
|
| 42 |
+
[test_dataset]
|
| 43 |
+
path = "dataloader.Dataset"
|
| 44 |
+
|
| 45 |
+
[trainer.args]
|
| 46 |
+
debug = false
|
| 47 |
+
max_steps = 0
|
| 48 |
+
max_epochs = 1000
|
| 49 |
+
max_grad_norm = 1.0
|
| 50 |
+
save_max_score = true
|
| 51 |
+
save_ckpt_interval = 5
|
| 52 |
+
max_patience = 200
|
| 53 |
+
plot_norm = true
|
| 54 |
+
validation_interval = 200
|
| 55 |
+
max_num_checkpoints = 100
|
| 56 |
+
scheduler_name = "constant_schedule_with_warmup"
|
| 57 |
+
warmup_steps = 1000
|
| 58 |
+
warmup_ratio = 0.0
|
| 59 |
+
gradient_accumulation_steps = 1
|
| 60 |
+
|
| 61 |
+
[loss_function.args]
|
| 62 |
+
|
| 63 |
+
[optimizer.args]
|
| 64 |
+
lr = 0.0001
|
| 65 |
+
|
| 66 |
+
[model.args]
|
| 67 |
+
|
| 68 |
+
[train_dataset.args]
|
| 69 |
+
librispeech_dir = "~/data/LibriSpeech/LibriSpeech"
|
| 70 |
+
librispeech_metadata_fpath = "/home/xhao/proj/audiozen/recipes/librimix_sot/local/metadata/LibriSpeech/train-clean-100-24K.csv"
|
| 71 |
+
duration = 6.0
|
| 72 |
+
sr = 24000
|
| 73 |
+
|
| 74 |
+
[train_dataset.dataloader]
|
| 75 |
+
batch_size = 20
|
| 76 |
+
num_workers = 10
|
| 77 |
+
drop_last = true
|
| 78 |
+
pin_memory = true
|
| 79 |
+
|
| 80 |
+
[test_dataset.args]
|
| 81 |
+
librispeech_dir = "~/data/LibriSpeech/LibriSpeech"
|
| 82 |
+
librispeech_metadata_fpath = "/home/xhao/proj/audiozen/recipes/librimix_sot/local/metadata/LibriSpeech/train-clean-100-24K.csv"
|
| 83 |
+
duration = 6.0
|
| 84 |
+
sr = 24000
|
| 85 |
+
num_samples = 10
|
| 86 |
+
|
| 87 |
+
[test_dataset.dataloader]
|
| 88 |
+
batch_size = 1
|
| 89 |
+
num_workers = 1
|
swin_default_LR1e-2_addEOSToken/swin_default_LR1e-2_addEOSToken_2024_01_17--11_16_21.log
ADDED
|
@@ -0,0 +1,142 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
01-17 11:16:21 INFO [logger.py:80]: Initialized logger with log file in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken.
|
| 2 |
+
01-17 11:16:21 INFO [logger.py:80]: Initialized logger with log file in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken.
|
| 3 |
+
01-17 11:16:21 INFO [logger.py:80]: Initialized logger with log file in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken.
|
| 4 |
+
01-17 11:16:21 INFO [logger.py:80]: Initialized logger with log file in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken.
|
| 5 |
+
01-17 11:16:57 INFO [logging.py:61]: Configuration file is saved to /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken/config__2024_01_17--11_16_51.toml.
|
| 6 |
+
01-17 11:16:57 INFO [logging.py:61]: Environment information:
|
| 7 |
+
- `Accelerate` version: 0.26.1
|
| 8 |
+
- Platform: Linux-5.14.0-362.13.1.el9_3.x86_64-x86_64-with-glibc2.34
|
| 9 |
+
- Python version: 3.10.13
|
| 10 |
+
- Numpy version: 1.26.3
|
| 11 |
+
- PyTorch version (GPU?): 2.1.2 (True)
|
| 12 |
+
- System RAM: 503.48 GB
|
| 13 |
+
- GPU Available: True
|
| 14 |
+
- GPU IDs: 4
|
| 15 |
+
- GPU type: NVIDIA A100-SXM4-80GB
|
| 16 |
+
01-17 11:16:57 INFO [logging.py:61]:
|
| 17 |
+
===============================================================================================
|
| 18 |
+
Layer (type:depth-idx) Param #
|
| 19 |
+
===============================================================================================
|
| 20 |
+
DistributedDataParallel --
|
| 21 |
+
├─Model: 1-1 --
|
| 22 |
+
│ └─EncodecModel: 2-1 --
|
| 23 |
+
│ │ └─EncodecEncoder: 3-1 (7,425,792)
|
| 24 |
+
│ │ └─EncodecDecoder: 3-2 (7,426,018)
|
| 25 |
+
│ │ └─EncodecResidualVectorQuantizer: 3-3 --
|
| 26 |
+
│ └─ModuleList: 2-2 --
|
| 27 |
+
│ │ └─TokenEmbedding: 3-4 1,049,600
|
| 28 |
+
│ │ └─TokenEmbedding: 3-5 1,049,600
|
| 29 |
+
│ │ └─TokenEmbedding: 3-6 1,049,600
|
| 30 |
+
│ │ └─TokenEmbedding: 3-7 1,049,600
|
| 31 |
+
│ │ └─TokenEmbedding: 3-8 1,049,600
|
| 32 |
+
│ │ └─TokenEmbedding: 3-9 1,049,600
|
| 33 |
+
│ │ └─TokenEmbedding: 3-10 1,049,600
|
| 34 |
+
│ │ └─TokenEmbedding: 3-11 1,049,600
|
| 35 |
+
│ └─PreNet: 2-3 --
|
| 36 |
+
│ │ └─Sequential: 3-12 591,360
|
| 37 |
+
│ └─SinePositionalEmbedding: 2-4 1
|
| 38 |
+
│ │ └─Dropout: 3-13 --
|
| 39 |
+
│ └─TransformerEncoder: 2-5 --
|
| 40 |
+
│ │ └─ModuleList: 3-14 201,535,488
|
| 41 |
+
│ │ └─AdaptiveLayerNorm: 3-15 2,101,248
|
| 42 |
+
│ └─ModuleList: 2-6 --
|
| 43 |
+
│ │ └─Linear: 3-16 1,049,600
|
| 44 |
+
│ │ └─Linear: 3-17 1,049,600
|
| 45 |
+
│ │ └─Linear: 3-18 1,049,600
|
| 46 |
+
│ │ └─Linear: 3-19 1,049,600
|
| 47 |
+
│ │ └─Linear: 3-20 1,049,600
|
| 48 |
+
│ │ └─Linear: 3-21 1,049,600
|
| 49 |
+
│ │ └─Linear: 3-22 1,049,600
|
| 50 |
+
│ │ └─Linear: 3-23 1,049,600
|
| 51 |
+
│ └─ModuleList: 2-7 --
|
| 52 |
+
│ │ └─TokenEmbedding: 3-24 1,024
|
| 53 |
+
│ │ └─TokenEmbedding: 3-25 1,024
|
| 54 |
+
│ │ └─TokenEmbedding: 3-26 1,024
|
| 55 |
+
│ │ └─TokenEmbedding: 3-27 1,024
|
| 56 |
+
│ │ └─TokenEmbedding: 3-28 1,024
|
| 57 |
+
│ │ └─TokenEmbedding: 3-29 1,024
|
| 58 |
+
│ │ └─TokenEmbedding: 3-30 1,024
|
| 59 |
+
│ │ └─TokenEmbedding: 3-31 1,024
|
| 60 |
+
│ └─MulticlassAccuracy: 2-8 --
|
| 61 |
+
===============================================================================================
|
| 62 |
+
Total params: 235,881,699
|
| 63 |
+
Trainable params: 221,029,888
|
| 64 |
+
Non-trainable params: 14,851,811
|
| 65 |
+
===============================================================================================
|
| 66 |
+
01-17 11:16:57 INFO [logging.py:61]: Training control variables:
|
| 67 |
+
01-17 11:16:57 INFO [logging.py:61]: `steps_per_epoch`: 500
|
| 68 |
+
01-17 11:16:57 INFO [logging.py:61]: Gradient accumulation steps: 1
|
| 69 |
+
01-17 11:16:57 INFO [logging.py:61]: `update_steps_per_epoch`: 500
|
| 70 |
+
01-17 11:16:57 INFO [logging.py:61]: `max_steps`: 500000
|
| 71 |
+
01-17 11:16:57 INFO [logging.py:61]: `max_epochs`: 1000
|
| 72 |
+
01-17 11:16:57 INFO [logging.py:61]: warmup_steps=1000. warmup_ratio will be ignored.
|
| 73 |
+
01-17 11:16:57 INFO [logging.py:61]: ========= Epoch 1 out of 1000 =========
|
| 74 |
+
01-17 11:16:57 INFO [logging.py:61]: Begin training...
|
| 75 |
+
01-17 11:34:44 INFO [logging.py:61]: Loss 'loss' on epoch 1: 5.376262664794922
|
| 76 |
+
01-17 11:34:44 INFO [logging.py:61]: Loss 'acc' on epoch 1: 0.29953059554100037
|
| 77 |
+
01-17 11:34:44 INFO [logging.py:61]: ========= Epoch 2 out of 1000 =========
|
| 78 |
+
01-17 11:34:44 INFO [logging.py:61]: Begin training...
|
| 79 |
+
01-17 11:52:25 INFO [logging.py:61]: Loss 'loss' on epoch 2: 4.830622673034668
|
| 80 |
+
01-17 11:52:25 INFO [logging.py:61]: Loss 'acc' on epoch 2: 0.35284850001335144
|
| 81 |
+
01-17 11:52:25 INFO [logging.py:61]: ========= Epoch 3 out of 1000 =========
|
| 82 |
+
01-17 11:52:25 INFO [logging.py:61]: Begin training...
|
| 83 |
+
01-17 12:10:07 INFO [logging.py:61]: Loss 'loss' on epoch 3: 4.957332134246826
|
| 84 |
+
01-17 12:10:07 INFO [logging.py:61]: Loss 'acc' on epoch 3: 0.3474164605140686
|
| 85 |
+
01-17 12:10:07 INFO [logging.py:61]: ========= Epoch 4 out of 1000 =========
|
| 86 |
+
01-17 12:10:07 INFO [logging.py:61]: Begin training...
|
| 87 |
+
01-17 12:27:48 INFO [logging.py:61]: Loss 'loss' on epoch 4: 4.950679779052734
|
| 88 |
+
01-17 12:27:48 INFO [logging.py:61]: Loss 'acc' on epoch 4: 0.3592258095741272
|
| 89 |
+
01-17 12:27:48 INFO [logging.py:61]: ========= Epoch 5 out of 1000 =========
|
| 90 |
+
01-17 12:27:48 INFO [logging.py:61]: Begin training...
|
| 91 |
+
01-17 12:45:29 INFO [logging.py:61]: Loss 'loss' on epoch 5: 4.960457801818848
|
| 92 |
+
01-17 12:45:29 INFO [logging.py:61]: Loss 'acc' on epoch 5: 0.3581017851829529
|
| 93 |
+
01-17 12:45:29 INFO [logging.py:61]: Saving current state to /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken/checkpoints/epoch_0005
|
| 94 |
+
01-17 12:45:30 INFO [logging.py:61]: Model weights saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken/checkpoints/epoch_0005/pytorch_model.bin
|
| 95 |
+
01-17 12:45:33 INFO [logging.py:61]: Optimizer state saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken/checkpoints/epoch_0005/optimizer.bin
|
| 96 |
+
01-17 12:45:33 INFO [logging.py:61]: Scheduler state saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken/checkpoints/epoch_0005/scheduler.bin
|
| 97 |
+
01-17 12:45:33 INFO [logging.py:61]: Sampler state for dataloader 0 saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken/checkpoints/epoch_0005/sampler.bin
|
| 98 |
+
01-17 12:45:33 INFO [logging.py:61]: Sampler state for dataloader 1 saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken/checkpoints/epoch_0005/sampler_1.bin
|
| 99 |
+
01-17 12:45:33 INFO [logging.py:61]: Random states saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken/checkpoints/epoch_0005/random_states_0.pkl
|
| 100 |
+
01-17 12:45:33 INFO [logging.py:61]: Saving the state of TrainerState to /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken/checkpoints/epoch_0005/custom_checkpoint_0.pkl
|
| 101 |
+
01-17 12:45:33 INFO [logging.py:61]: ========= Epoch 6 out of 1000 =========
|
| 102 |
+
01-17 12:45:33 INFO [logging.py:61]: Begin training...
|
| 103 |
+
01-17 13:03:13 INFO [logging.py:61]: Loss 'loss' on epoch 6: 5.089817523956299
|
| 104 |
+
01-17 13:03:13 INFO [logging.py:61]: Loss 'acc' on epoch 6: 0.35066384077072144
|
| 105 |
+
01-17 13:03:13 INFO [logging.py:61]: ========= Epoch 7 out of 1000 =========
|
| 106 |
+
01-17 13:03:13 INFO [logging.py:61]: Begin training...
|
| 107 |
+
01-17 13:20:54 INFO [logging.py:61]: Loss 'loss' on epoch 7: 5.060431480407715
|
| 108 |
+
01-17 13:20:54 INFO [logging.py:61]: Loss 'acc' on epoch 7: 0.3525404930114746
|
| 109 |
+
01-17 13:20:54 INFO [logging.py:61]: ========= Epoch 8 out of 1000 =========
|
| 110 |
+
01-17 13:20:54 INFO [logging.py:61]: Begin training...
|
| 111 |
+
01-17 13:38:35 INFO [logging.py:61]: Loss 'loss' on epoch 8: 5.10891056060791
|
| 112 |
+
01-17 13:38:35 INFO [logging.py:61]: Loss 'acc' on epoch 8: 0.349361389875412
|
| 113 |
+
01-17 13:38:35 INFO [logging.py:61]: ========= Epoch 9 out of 1000 =========
|
| 114 |
+
01-17 13:38:35 INFO [logging.py:61]: Begin training...
|
| 115 |
+
01-17 13:56:16 INFO [logging.py:61]: Loss 'loss' on epoch 9: 5.178280830383301
|
| 116 |
+
01-17 13:56:16 INFO [logging.py:61]: Loss 'acc' on epoch 9: 0.34400177001953125
|
| 117 |
+
01-17 13:56:16 INFO [logging.py:61]: ========= Epoch 10 out of 1000 =========
|
| 118 |
+
01-17 13:56:16 INFO [logging.py:61]: Begin training...
|
| 119 |
+
01-17 14:13:57 INFO [logging.py:61]: Loss 'loss' on epoch 10: 5.152740001678467
|
| 120 |
+
01-17 14:13:57 INFO [logging.py:61]: Loss 'acc' on epoch 10: 0.34986352920532227
|
| 121 |
+
01-17 14:13:57 INFO [logging.py:61]: Saving current state to /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken/checkpoints/epoch_0010
|
| 122 |
+
01-17 14:13:58 INFO [logging.py:61]: Model weights saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken/checkpoints/epoch_0010/pytorch_model.bin
|
| 123 |
+
01-17 14:14:00 INFO [logging.py:61]: Optimizer state saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken/checkpoints/epoch_0010/optimizer.bin
|
| 124 |
+
01-17 14:14:00 INFO [logging.py:61]: Scheduler state saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken/checkpoints/epoch_0010/scheduler.bin
|
| 125 |
+
01-17 14:14:00 INFO [logging.py:61]: Sampler state for dataloader 0 saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken/checkpoints/epoch_0010/sampler.bin
|
| 126 |
+
01-17 14:14:00 INFO [logging.py:61]: Sampler state for dataloader 1 saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken/checkpoints/epoch_0010/sampler_1.bin
|
| 127 |
+
01-17 14:14:00 INFO [logging.py:61]: Random states saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken/checkpoints/epoch_0010/random_states_0.pkl
|
| 128 |
+
01-17 14:14:00 INFO [logging.py:61]: Saving the state of TrainerState to /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken/checkpoints/epoch_0010/custom_checkpoint_0.pkl
|
| 129 |
+
01-17 14:14:01 INFO [logging.py:61]: ========= Epoch 11 out of 1000 =========
|
| 130 |
+
01-17 14:14:01 INFO [logging.py:61]: Begin training...
|
| 131 |
+
01-17 14:31:41 INFO [logging.py:61]: Loss 'loss' on epoch 11: 5.189244270324707
|
| 132 |
+
01-17 14:31:41 INFO [logging.py:61]: Loss 'acc' on epoch 11: 0.3463723957538605
|
| 133 |
+
01-17 14:31:41 INFO [logging.py:61]: ========= Epoch 12 out of 1000 =========
|
| 134 |
+
01-17 14:31:41 INFO [logging.py:61]: Begin training...
|
| 135 |
+
01-17 14:49:22 INFO [logging.py:61]: Loss 'loss' on epoch 12: 5.161355495452881
|
| 136 |
+
01-17 14:49:22 INFO [logging.py:61]: Loss 'acc' on epoch 12: 0.347272664308548
|
| 137 |
+
01-17 14:49:22 INFO [logging.py:61]: ========= Epoch 13 out of 1000 =========
|
| 138 |
+
01-17 14:49:22 INFO [logging.py:61]: Begin training...
|
| 139 |
+
01-17 15:07:03 INFO [logging.py:61]: Loss 'loss' on epoch 13: 5.166537761688232
|
| 140 |
+
01-17 15:07:03 INFO [logging.py:61]: Loss 'acc' on epoch 13: 0.34788310527801514
|
| 141 |
+
01-17 15:07:03 INFO [logging.py:61]: ========= Epoch 14 out of 1000 =========
|
| 142 |
+
01-17 15:07:03 INFO [logging.py:61]: Begin training...
|
swin_default_LR1e-2_addEOSToken/swin_default_LR1e-2_addEOSToken_2024_01_17--12_14_33.log
ADDED
|
@@ -0,0 +1,130 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
01-17 12:14:33 INFO [logger.py:80]: Initialized logger with log file in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken.
|
| 2 |
+
01-17 12:14:33 INFO [logger.py:80]: Initialized logger with log file in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken.
|
| 3 |
+
01-17 12:14:33 INFO [logger.py:80]: Initialized logger with log file in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken.
|
| 4 |
+
01-17 12:14:33 INFO [logger.py:80]: Initialized logger with log file in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken.
|
| 5 |
+
01-17 12:14:38 INFO [logging.py:61]: Configuration file is saved to /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken/config__2024_01_17--12_14_37.toml.
|
| 6 |
+
01-17 12:14:38 INFO [logging.py:61]: Environment information:
|
| 7 |
+
- `Accelerate` version: 0.26.1
|
| 8 |
+
- Platform: Linux-5.14.0-362.13.1.el9_3.x86_64-x86_64-with-glibc2.34
|
| 9 |
+
- Python version: 3.10.13
|
| 10 |
+
- Numpy version: 1.26.3
|
| 11 |
+
- PyTorch version (GPU?): 2.1.2 (True)
|
| 12 |
+
- System RAM: 1007.48 GB
|
| 13 |
+
- GPU Available: True
|
| 14 |
+
- GPU IDs: 4
|
| 15 |
+
- GPU type: NVIDIA A100-SXM4-80GB
|
| 16 |
+
01-17 12:14:38 INFO [logging.py:61]:
|
| 17 |
+
===============================================================================================
|
| 18 |
+
Layer (type:depth-idx) Param #
|
| 19 |
+
===============================================================================================
|
| 20 |
+
DistributedDataParallel --
|
| 21 |
+
├─Model: 1-1 --
|
| 22 |
+
│ └─EncodecModel: 2-1 --
|
| 23 |
+
│ │ └─EncodecEncoder: 3-1 (7,425,792)
|
| 24 |
+
│ │ └─EncodecDecoder: 3-2 (7,426,018)
|
| 25 |
+
│ │ └─EncodecResidualVectorQuantizer: 3-3 --
|
| 26 |
+
│ └─ModuleList: 2-2 --
|
| 27 |
+
│ │ └─TokenEmbedding: 3-4 1,049,600
|
| 28 |
+
│ │ └─TokenEmbedding: 3-5 1,049,600
|
| 29 |
+
│ │ └─TokenEmbedding: 3-6 1,049,600
|
| 30 |
+
│ │ └─TokenEmbedding: 3-7 1,049,600
|
| 31 |
+
│ │ └─TokenEmbedding: 3-8 1,049,600
|
| 32 |
+
│ │ └─TokenEmbedding: 3-9 1,049,600
|
| 33 |
+
│ │ └─TokenEmbedding: 3-10 1,049,600
|
| 34 |
+
│ │ └─TokenEmbedding: 3-11 1,049,600
|
| 35 |
+
│ └─PreNet: 2-3 --
|
| 36 |
+
│ │ └─Sequential: 3-12 591,360
|
| 37 |
+
│ └─SinePositionalEmbedding: 2-4 1
|
| 38 |
+
│ │ └─Dropout: 3-13 --
|
| 39 |
+
│ └─TransformerEncoder: 2-5 --
|
| 40 |
+
│ │ └─ModuleList: 3-14 201,535,488
|
| 41 |
+
│ │ └─AdaptiveLayerNorm: 3-15 2,101,248
|
| 42 |
+
│ └─ModuleList: 2-6 --
|
| 43 |
+
│ │ └─Linear: 3-16 1,049,600
|
| 44 |
+
│ │ └─Linear: 3-17 1,049,600
|
| 45 |
+
│ │ └─Linear: 3-18 1,049,600
|
| 46 |
+
│ │ └─Linear: 3-19 1,049,600
|
| 47 |
+
│ │ └─Linear: 3-20 1,049,600
|
| 48 |
+
│ │ └─Linear: 3-21 1,049,600
|
| 49 |
+
│ │ └─Linear: 3-22 1,049,600
|
| 50 |
+
│ │ └─Linear: 3-23 1,049,600
|
| 51 |
+
│ └─ModuleList: 2-7 --
|
| 52 |
+
│ │ └─TokenEmbedding: 3-24 1,024
|
| 53 |
+
│ │ └─TokenEmbedding: 3-25 1,024
|
| 54 |
+
│ │ └─TokenEmbedding: 3-26 1,024
|
| 55 |
+
│ │ └─TokenEmbedding: 3-27 1,024
|
| 56 |
+
│ │ └─TokenEmbedding: 3-28 1,024
|
| 57 |
+
│ │ └─TokenEmbedding: 3-29 1,024
|
| 58 |
+
│ │ └─TokenEmbedding: 3-30 1,024
|
| 59 |
+
│ │ └─TokenEmbedding: 3-31 1,024
|
| 60 |
+
│ └─MulticlassAccuracy: 2-8 --
|
| 61 |
+
===============================================================================================
|
| 62 |
+
Total params: 235,881,699
|
| 63 |
+
Trainable params: 221,029,888
|
| 64 |
+
Non-trainable params: 14,851,811
|
| 65 |
+
===============================================================================================
|
| 66 |
+
01-17 12:14:38 INFO [logging.py:61]: Training control variables:
|
| 67 |
+
01-17 12:14:38 INFO [logging.py:61]: `steps_per_epoch`: 500
|
| 68 |
+
01-17 12:14:38 INFO [logging.py:61]: Gradient accumulation steps: 1
|
| 69 |
+
01-17 12:14:38 INFO [logging.py:61]: `update_steps_per_epoch`: 500
|
| 70 |
+
01-17 12:14:38 INFO [logging.py:61]: `max_steps`: 500000
|
| 71 |
+
01-17 12:14:38 INFO [logging.py:61]: `max_epochs`: 1000
|
| 72 |
+
01-17 12:14:38 INFO [logging.py:61]: warmup_steps=1000. warmup_ratio will be ignored.
|
| 73 |
+
01-17 12:14:38 INFO [logging.py:61]: ========= Epoch 1 out of 1000 =========
|
| 74 |
+
01-17 12:14:38 INFO [logging.py:61]: Begin training...
|
| 75 |
+
01-17 12:32:23 INFO [logging.py:61]: Loss 'loss' on epoch 1: 5.352734088897705
|
| 76 |
+
01-17 12:32:23 INFO [logging.py:61]: Loss 'acc' on epoch 1: 0.30075228214263916
|
| 77 |
+
01-17 12:32:23 INFO [logging.py:61]: ========= Epoch 2 out of 1000 =========
|
| 78 |
+
01-17 12:32:23 INFO [logging.py:61]: Begin training...
|
| 79 |
+
01-17 12:50:06 INFO [logging.py:61]: Loss 'loss' on epoch 2: 4.763156414031982
|
| 80 |
+
01-17 12:50:06 INFO [logging.py:61]: Loss 'acc' on epoch 2: 0.35356950759887695
|
| 81 |
+
01-17 12:50:06 INFO [logging.py:61]: ========= Epoch 3 out of 1000 =========
|
| 82 |
+
01-17 12:50:06 INFO [logging.py:61]: Begin training...
|
| 83 |
+
01-17 13:07:51 INFO [logging.py:61]: Loss 'loss' on epoch 3: 4.902694225311279
|
| 84 |
+
01-17 13:07:51 INFO [logging.py:61]: Loss 'acc' on epoch 3: 0.35302937030792236
|
| 85 |
+
01-17 13:07:51 INFO [logging.py:61]: ========= Epoch 4 out of 1000 =========
|
| 86 |
+
01-17 13:07:51 INFO [logging.py:61]: Begin training...
|
| 87 |
+
01-17 13:28:14 INFO [logging.py:61]: Loss 'loss' on epoch 4: 4.946804046630859
|
| 88 |
+
01-17 13:28:14 INFO [logging.py:61]: Loss 'acc' on epoch 4: 0.3474726378917694
|
| 89 |
+
01-17 13:28:14 INFO [logging.py:61]: ========= Epoch 5 out of 1000 =========
|
| 90 |
+
01-17 13:28:14 INFO [logging.py:61]: Begin training...
|
| 91 |
+
01-17 13:46:01 INFO [logging.py:61]: Loss 'loss' on epoch 5: 5.042134761810303
|
| 92 |
+
01-17 13:46:01 INFO [logging.py:61]: Loss 'acc' on epoch 5: 0.3472442626953125
|
| 93 |
+
01-17 13:46:01 INFO [logging.py:61]: Saving current state to /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken/checkpoints/epoch_0005
|
| 94 |
+
01-17 13:46:02 INFO [logging.py:61]: Model weights saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken/checkpoints/epoch_0005/pytorch_model.bin
|
| 95 |
+
01-17 13:46:05 INFO [logging.py:61]: Optimizer state saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken/checkpoints/epoch_0005/optimizer.bin
|
| 96 |
+
01-17 13:46:05 INFO [logging.py:61]: Scheduler state saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken/checkpoints/epoch_0005/scheduler.bin
|
| 97 |
+
01-17 13:46:05 INFO [logging.py:61]: Sampler state for dataloader 0 saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken/checkpoints/epoch_0005/sampler.bin
|
| 98 |
+
01-17 13:46:05 INFO [logging.py:61]: Sampler state for dataloader 1 saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken/checkpoints/epoch_0005/sampler_1.bin
|
| 99 |
+
01-17 13:46:05 INFO [logging.py:61]: Random states saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken/checkpoints/epoch_0005/random_states_0.pkl
|
| 100 |
+
01-17 13:46:05 INFO [logging.py:61]: Saving the state of TrainerState to /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken/checkpoints/epoch_0005/custom_checkpoint_0.pkl
|
| 101 |
+
01-17 13:46:05 INFO [logging.py:61]: ========= Epoch 6 out of 1000 =========
|
| 102 |
+
01-17 13:46:05 INFO [logging.py:61]: Begin training...
|
| 103 |
+
01-17 14:03:48 INFO [logging.py:61]: Loss 'loss' on epoch 6: 5.154253959655762
|
| 104 |
+
01-17 14:03:48 INFO [logging.py:61]: Loss 'acc' on epoch 6: 0.3447455167770386
|
| 105 |
+
01-17 14:03:48 INFO [logging.py:61]: ========= Epoch 7 out of 1000 =========
|
| 106 |
+
01-17 14:03:48 INFO [logging.py:61]: Begin training...
|
| 107 |
+
01-17 14:21:30 INFO [logging.py:61]: Loss 'loss' on epoch 7: 5.2775559425354
|
| 108 |
+
01-17 14:21:30 INFO [logging.py:61]: Loss 'acc' on epoch 7: 0.3431953191757202
|
| 109 |
+
01-17 14:21:30 INFO [logging.py:61]: ========= Epoch 8 out of 1000 =========
|
| 110 |
+
01-17 14:21:30 INFO [logging.py:61]: Begin training...
|
| 111 |
+
01-17 14:39:10 INFO [logging.py:61]: Loss 'loss' on epoch 8: 5.302144527435303
|
| 112 |
+
01-17 14:39:10 INFO [logging.py:61]: Loss 'acc' on epoch 8: 0.3417191207408905
|
| 113 |
+
01-17 14:39:10 INFO [logging.py:61]: ========= Epoch 9 out of 1000 =========
|
| 114 |
+
01-17 14:39:10 INFO [logging.py:61]: Begin training...
|
| 115 |
+
01-17 14:56:49 INFO [logging.py:61]: Loss 'loss' on epoch 9: 5.384851932525635
|
| 116 |
+
01-17 14:56:49 INFO [logging.py:61]: Loss 'acc' on epoch 9: 0.33888909220695496
|
| 117 |
+
01-17 14:56:49 INFO [logging.py:61]: ========= Epoch 10 out of 1000 =========
|
| 118 |
+
01-17 14:56:49 INFO [logging.py:61]: Begin training...
|
| 119 |
+
01-17 15:14:29 INFO [logging.py:61]: Loss 'loss' on epoch 10: 5.374612808227539
|
| 120 |
+
01-17 15:14:29 INFO [logging.py:61]: Loss 'acc' on epoch 10: 0.3413293659687042
|
| 121 |
+
01-17 15:14:29 INFO [logging.py:61]: Saving current state to /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken/checkpoints/epoch_0010
|
| 122 |
+
01-17 15:14:30 INFO [logging.py:61]: Model weights saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken/checkpoints/epoch_0010/pytorch_model.bin
|
| 123 |
+
01-17 15:14:32 INFO [logging.py:61]: Optimizer state saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken/checkpoints/epoch_0010/optimizer.bin
|
| 124 |
+
01-17 15:14:32 INFO [logging.py:61]: Scheduler state saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken/checkpoints/epoch_0010/scheduler.bin
|
| 125 |
+
01-17 15:14:32 INFO [logging.py:61]: Sampler state for dataloader 0 saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken/checkpoints/epoch_0010/sampler.bin
|
| 126 |
+
01-17 15:14:32 INFO [logging.py:61]: Sampler state for dataloader 1 saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken/checkpoints/epoch_0010/sampler_1.bin
|
| 127 |
+
01-17 15:14:32 INFO [logging.py:61]: Random states saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken/checkpoints/epoch_0010/random_states_0.pkl
|
| 128 |
+
01-17 15:14:32 INFO [logging.py:61]: Saving the state of TrainerState to /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken/checkpoints/epoch_0010/custom_checkpoint_0.pkl
|
| 129 |
+
01-17 15:14:32 INFO [logging.py:61]: ========= Epoch 11 out of 1000 =========
|
| 130 |
+
01-17 15:14:32 INFO [logging.py:61]: Begin training...
|
swin_default_LR1e-2_addEOSToken/swin_default_LR1e-2_addEOSToken_2024_01_17--17_15_22.log
ADDED
|
@@ -0,0 +1,203 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
01-17 17:15:22 INFO [logger.py:80]: Initialized logger with log file in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken.
|
| 2 |
+
01-17 17:15:22 INFO [logger.py:80]: Initialized logger with log file in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken.
|
| 3 |
+
01-17 17:15:22 INFO [logger.py:80]: Initialized logger with log file in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken.
|
| 4 |
+
01-17 17:15:22 INFO [logger.py:80]: Initialized logger with log file in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken.
|
| 5 |
+
01-17 17:17:03 INFO [logging.py:61]: Configuration file is saved to /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken/config__2024_01_17--17_16_44.toml.
|
| 6 |
+
01-17 17:17:03 INFO [logging.py:61]: Environment information:
|
| 7 |
+
- `Accelerate` version: 0.26.1
|
| 8 |
+
- Platform: Linux-5.14.0-362.13.1.el9_3.x86_64-x86_64-with-glibc2.34
|
| 9 |
+
- Python version: 3.10.13
|
| 10 |
+
- Numpy version: 1.26.3
|
| 11 |
+
- PyTorch version (GPU?): 2.1.2 (True)
|
| 12 |
+
- System RAM: 503.48 GB
|
| 13 |
+
- GPU Available: True
|
| 14 |
+
- GPU IDs: 4
|
| 15 |
+
- GPU type: NVIDIA A100-SXM4-80GB
|
| 16 |
+
01-17 17:17:03 INFO [logging.py:61]:
|
| 17 |
+
===============================================================================================
|
| 18 |
+
Layer (type:depth-idx) Param #
|
| 19 |
+
===============================================================================================
|
| 20 |
+
DistributedDataParallel --
|
| 21 |
+
├─Model: 1-1 --
|
| 22 |
+
│ └─EncodecModel: 2-1 --
|
| 23 |
+
│ │ └─EncodecEncoder: 3-1 (7,425,792)
|
| 24 |
+
│ │ └─EncodecDecoder: 3-2 (7,426,018)
|
| 25 |
+
│ │ └─EncodecResidualVectorQuantizer: 3-3 --
|
| 26 |
+
│ └─ModuleList: 2-2 --
|
| 27 |
+
│ │ └─TokenEmbedding: 3-4 1,049,600
|
| 28 |
+
│ │ └─TokenEmbedding: 3-5 1,049,600
|
| 29 |
+
│ │ └─TokenEmbedding: 3-6 1,049,600
|
| 30 |
+
│ │ └─TokenEmbedding: 3-7 1,049,600
|
| 31 |
+
│ │ └─TokenEmbedding: 3-8 1,049,600
|
| 32 |
+
│ │ └─TokenEmbedding: 3-9 1,049,600
|
| 33 |
+
│ │ └─TokenEmbedding: 3-10 1,049,600
|
| 34 |
+
│ │ └─TokenEmbedding: 3-11 1,049,600
|
| 35 |
+
│ └─PreNet: 2-3 --
|
| 36 |
+
│ │ └─Sequential: 3-12 591,360
|
| 37 |
+
│ └─SinePositionalEmbedding: 2-4 1
|
| 38 |
+
│ │ └─Dropout: 3-13 --
|
| 39 |
+
│ └─TransformerEncoder: 2-5 --
|
| 40 |
+
│ │ └─ModuleList: 3-14 201,535,488
|
| 41 |
+
│ │ └─AdaptiveLayerNorm: 3-15 2,101,248
|
| 42 |
+
│ └─ModuleList: 2-6 --
|
| 43 |
+
│ │ └─Linear: 3-16 1,049,600
|
| 44 |
+
│ │ └─Linear: 3-17 1,049,600
|
| 45 |
+
│ │ └─Linear: 3-18 1,049,600
|
| 46 |
+
│ │ └─Linear: 3-19 1,049,600
|
| 47 |
+
│ │ └─Linear: 3-20 1,049,600
|
| 48 |
+
│ │ └─Linear: 3-21 1,049,600
|
| 49 |
+
│ │ └─Linear: 3-22 1,049,600
|
| 50 |
+
│ │ └─Linear: 3-23 1,049,600
|
| 51 |
+
│ └─ModuleList: 2-7 --
|
| 52 |
+
│ │ └─TokenEmbedding: 3-24 1,024
|
| 53 |
+
│ │ └─TokenEmbedding: 3-25 1,024
|
| 54 |
+
│ │ └─TokenEmbedding: 3-26 1,024
|
| 55 |
+
│ │ └─TokenEmbedding: 3-27 1,024
|
| 56 |
+
│ │ └─TokenEmbedding: 3-28 1,024
|
| 57 |
+
│ │ └─TokenEmbedding: 3-29 1,024
|
| 58 |
+
│ │ └─TokenEmbedding: 3-30 1,024
|
| 59 |
+
│ │ └─TokenEmbedding: 3-31 1,024
|
| 60 |
+
│ └─MulticlassAccuracy: 2-8 --
|
| 61 |
+
===============================================================================================
|
| 62 |
+
Total params: 235,881,699
|
| 63 |
+
Trainable params: 221,029,888
|
| 64 |
+
Non-trainable params: 14,851,811
|
| 65 |
+
===============================================================================================
|
| 66 |
+
01-17 17:17:03 INFO [logging.py:61]: Training control variables:
|
| 67 |
+
01-17 17:17:03 INFO [logging.py:61]: `steps_per_epoch`: 500
|
| 68 |
+
01-17 17:17:03 INFO [logging.py:61]: Gradient accumulation steps: 1
|
| 69 |
+
01-17 17:17:03 INFO [logging.py:61]: `update_steps_per_epoch`: 500
|
| 70 |
+
01-17 17:17:03 INFO [logging.py:61]: `max_steps`: 500000
|
| 71 |
+
01-17 17:17:03 INFO [logging.py:61]: `max_epochs`: 1000
|
| 72 |
+
01-17 17:17:03 INFO [logging.py:61]: warmup_steps=1000. warmup_ratio will be ignored.
|
| 73 |
+
01-17 17:17:03 INFO [logging.py:61]: Loading states from /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken/checkpoints/epoch_0010
|
| 74 |
+
01-17 17:17:07 INFO [logging.py:61]: All model weights loaded successfully
|
| 75 |
+
01-17 17:17:13 INFO [logging.py:61]: All optimizer states loaded successfully
|
| 76 |
+
01-17 17:17:13 INFO [logging.py:61]: All scheduler states loaded successfully
|
| 77 |
+
01-17 17:17:13 INFO [logging.py:61]: All dataloader sampler states loaded successfully
|
| 78 |
+
01-17 17:17:13 INFO [logging.py:61]: All random states loaded successfully
|
| 79 |
+
01-17 17:17:13 INFO [logging.py:61]: Loading in 1 custom states
|
| 80 |
+
01-17 17:17:13 INFO [logging.py:61]: Loading the state of TrainerState from /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken/checkpoints/epoch_0010/custom_checkpoint_0.pkl
|
| 81 |
+
01-17 17:17:13 INFO [logging.py:61]: Checkpoint on epoch 10 is loaded.
|
| 82 |
+
01-17 17:17:13 INFO [logging.py:61]: ========= Epoch 11 out of 1000 =========
|
| 83 |
+
01-17 17:17:13 INFO [logging.py:61]: Begin training...
|
| 84 |
+
01-17 17:34:58 INFO [logging.py:61]: Loss 'loss' on epoch 11: 5.385993003845215
|
| 85 |
+
01-17 17:34:58 INFO [logging.py:61]: Loss 'acc' on epoch 11: 0.339417040348053
|
| 86 |
+
01-17 17:34:58 INFO [logging.py:61]: ========= Epoch 12 out of 1000 =========
|
| 87 |
+
01-17 17:34:58 INFO [logging.py:61]: Begin training...
|
| 88 |
+
01-17 17:53:05 INFO [logging.py:61]: Loss 'loss' on epoch 12: 5.376588344573975
|
| 89 |
+
01-17 17:53:06 INFO [logging.py:61]: Loss 'acc' on epoch 12: 0.33914294838905334
|
| 90 |
+
01-17 17:53:06 INFO [logging.py:61]: ========= Epoch 13 out of 1000 =========
|
| 91 |
+
01-17 17:53:06 INFO [logging.py:61]: Begin training...
|
| 92 |
+
01-17 18:11:23 INFO [logging.py:61]: Loss 'loss' on epoch 13: 5.388041973114014
|
| 93 |
+
01-17 18:11:24 INFO [logging.py:61]: Loss 'acc' on epoch 13: 0.339685320854187
|
| 94 |
+
01-17 18:11:24 INFO [logging.py:61]: ========= Epoch 14 out of 1000 =========
|
| 95 |
+
01-17 18:11:24 INFO [logging.py:61]: Begin training...
|
| 96 |
+
01-17 18:30:04 INFO [logging.py:61]: Loss 'loss' on epoch 14: 5.3795671463012695
|
| 97 |
+
01-17 18:30:04 INFO [logging.py:61]: Loss 'acc' on epoch 14: 0.3413822054862976
|
| 98 |
+
01-17 18:30:04 INFO [logging.py:61]: ========= Epoch 15 out of 1000 =========
|
| 99 |
+
01-17 18:30:04 INFO [logging.py:61]: Begin training...
|
| 100 |
+
01-17 18:48:00 INFO [logging.py:61]: Loss 'loss' on epoch 15: 5.390796661376953
|
| 101 |
+
01-17 18:48:00 INFO [logging.py:61]: Loss 'acc' on epoch 15: 0.33877530694007874
|
| 102 |
+
01-17 18:48:00 INFO [logging.py:61]: Saving current state to /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken/checkpoints/epoch_0015
|
| 103 |
+
01-17 18:48:02 INFO [logging.py:61]: Model weights saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken/checkpoints/epoch_0015/pytorch_model.bin
|
| 104 |
+
01-17 18:48:05 INFO [logging.py:61]: Optimizer state saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken/checkpoints/epoch_0015/optimizer.bin
|
| 105 |
+
01-17 18:48:05 INFO [logging.py:61]: Scheduler state saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken/checkpoints/epoch_0015/scheduler.bin
|
| 106 |
+
01-17 18:48:05 INFO [logging.py:61]: Sampler state for dataloader 0 saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken/checkpoints/epoch_0015/sampler.bin
|
| 107 |
+
01-17 18:48:05 INFO [logging.py:61]: Sampler state for dataloader 1 saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken/checkpoints/epoch_0015/sampler_1.bin
|
| 108 |
+
01-17 18:48:05 INFO [logging.py:61]: Random states saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken/checkpoints/epoch_0015/random_states_0.pkl
|
| 109 |
+
01-17 18:48:05 INFO [logging.py:61]: Saving the state of TrainerState to /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken/checkpoints/epoch_0015/custom_checkpoint_0.pkl
|
| 110 |
+
01-17 18:48:05 INFO [logging.py:61]: ========= Epoch 16 out of 1000 =========
|
| 111 |
+
01-17 18:48:05 INFO [logging.py:61]: Begin training...
|
| 112 |
+
01-17 19:06:17 INFO [logging.py:61]: Loss 'loss' on epoch 16: 5.389437198638916
|
| 113 |
+
01-17 19:06:17 INFO [logging.py:61]: Loss 'acc' on epoch 16: 0.33847111463546753
|
| 114 |
+
01-17 19:06:17 INFO [logging.py:61]: ========= Epoch 17 out of 1000 =========
|
| 115 |
+
01-17 19:06:17 INFO [logging.py:61]: Begin training...
|
| 116 |
+
01-17 19:24:08 INFO [logging.py:61]: Loss 'loss' on epoch 17: 5.380032539367676
|
| 117 |
+
01-17 19:24:08 INFO [logging.py:61]: Loss 'acc' on epoch 17: 0.339590847492218
|
| 118 |
+
01-17 19:24:08 INFO [logging.py:61]: ========= Epoch 18 out of 1000 =========
|
| 119 |
+
01-17 19:24:08 INFO [logging.py:61]: Begin training...
|
| 120 |
+
01-17 19:52:50 INFO [logging.py:61]: Loss 'loss' on epoch 18: 5.37690544128418
|
| 121 |
+
01-17 19:52:50 INFO [logging.py:61]: Loss 'acc' on epoch 18: 0.33943551778793335
|
| 122 |
+
01-17 19:52:50 INFO [logging.py:61]: ========= Epoch 19 out of 1000 =========
|
| 123 |
+
01-17 19:52:50 INFO [logging.py:61]: Begin training...
|
| 124 |
+
01-17 20:10:43 INFO [logging.py:61]: Loss 'loss' on epoch 19: 5.3855366706848145
|
| 125 |
+
01-17 20:10:43 INFO [logging.py:61]: Loss 'acc' on epoch 19: 0.33942481875419617
|
| 126 |
+
01-17 20:10:43 INFO [logging.py:61]: ========= Epoch 20 out of 1000 =========
|
| 127 |
+
01-17 20:10:43 INFO [logging.py:61]: Begin training...
|
| 128 |
+
01-17 20:29:32 INFO [logging.py:61]: Loss 'loss' on epoch 20: 5.374676704406738
|
| 129 |
+
01-17 20:29:32 INFO [logging.py:61]: Loss 'acc' on epoch 20: 0.33983924984931946
|
| 130 |
+
01-17 20:29:32 INFO [logging.py:61]: Saving current state to /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken/checkpoints/epoch_0020
|
| 131 |
+
01-17 20:29:33 INFO [logging.py:61]: Model weights saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken/checkpoints/epoch_0020/pytorch_model.bin
|
| 132 |
+
01-17 20:29:36 INFO [logging.py:61]: Optimizer state saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken/checkpoints/epoch_0020/optimizer.bin
|
| 133 |
+
01-17 20:29:36 INFO [logging.py:61]: Scheduler state saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken/checkpoints/epoch_0020/scheduler.bin
|
| 134 |
+
01-17 20:29:36 INFO [logging.py:61]: Sampler state for dataloader 0 saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken/checkpoints/epoch_0020/sampler.bin
|
| 135 |
+
01-17 20:29:36 INFO [logging.py:61]: Sampler state for dataloader 1 saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken/checkpoints/epoch_0020/sampler_1.bin
|
| 136 |
+
01-17 20:29:36 INFO [logging.py:61]: Random states saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken/checkpoints/epoch_0020/random_states_0.pkl
|
| 137 |
+
01-17 20:29:36 INFO [logging.py:61]: Saving the state of TrainerState to /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken/checkpoints/epoch_0020/custom_checkpoint_0.pkl
|
| 138 |
+
01-17 20:29:36 INFO [logging.py:61]: ========= Epoch 21 out of 1000 =========
|
| 139 |
+
01-17 20:29:36 INFO [logging.py:61]: Begin training...
|
| 140 |
+
01-17 20:47:57 INFO [logging.py:61]: Loss 'loss' on epoch 21: 5.386965274810791
|
| 141 |
+
01-17 20:47:57 INFO [logging.py:61]: Loss 'acc' on epoch 21: 0.3395291566848755
|
| 142 |
+
01-17 20:47:57 INFO [logging.py:61]: ========= Epoch 22 out of 1000 =========
|
| 143 |
+
01-17 20:47:57 INFO [logging.py:61]: Begin training...
|
| 144 |
+
01-17 21:05:45 INFO [logging.py:61]: Loss 'loss' on epoch 22: 5.365380764007568
|
| 145 |
+
01-17 21:05:45 INFO [logging.py:61]: Loss 'acc' on epoch 22: 0.34075096249580383
|
| 146 |
+
01-17 21:05:45 INFO [logging.py:61]: ========= Epoch 23 out of 1000 =========
|
| 147 |
+
01-17 21:05:45 INFO [logging.py:61]: Begin training...
|
| 148 |
+
01-17 21:23:33 INFO [logging.py:61]: Loss 'loss' on epoch 23: 5.38892936706543
|
| 149 |
+
01-17 21:23:33 INFO [logging.py:61]: Loss 'acc' on epoch 23: 0.33941057324409485
|
| 150 |
+
01-17 21:23:33 INFO [logging.py:61]: ========= Epoch 24 out of 1000 =========
|
| 151 |
+
01-17 21:23:33 INFO [logging.py:61]: Begin training...
|
| 152 |
+
01-17 21:41:21 INFO [logging.py:61]: Loss 'loss' on epoch 24: 5.370124340057373
|
| 153 |
+
01-17 21:41:21 INFO [logging.py:61]: Loss 'acc' on epoch 24: 0.3396085798740387
|
| 154 |
+
01-17 21:41:21 INFO [logging.py:61]: ========= Epoch 25 out of 1000 =========
|
| 155 |
+
01-17 21:41:21 INFO [logging.py:61]: Begin training...
|
| 156 |
+
01-17 21:59:09 INFO [logging.py:61]: Loss 'loss' on epoch 25: 5.369869232177734
|
| 157 |
+
01-17 21:59:09 INFO [logging.py:61]: Loss 'acc' on epoch 25: 0.3406093716621399
|
| 158 |
+
01-17 21:59:09 INFO [logging.py:61]: Saving current state to /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken/checkpoints/epoch_0025
|
| 159 |
+
01-17 21:59:11 INFO [logging.py:61]: Model weights saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken/checkpoints/epoch_0025/pytorch_model.bin
|
| 160 |
+
01-17 21:59:14 INFO [logging.py:61]: Optimizer state saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken/checkpoints/epoch_0025/optimizer.bin
|
| 161 |
+
01-17 21:59:14 INFO [logging.py:61]: Scheduler state saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken/checkpoints/epoch_0025/scheduler.bin
|
| 162 |
+
01-17 21:59:14 INFO [logging.py:61]: Sampler state for dataloader 0 saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken/checkpoints/epoch_0025/sampler.bin
|
| 163 |
+
01-17 21:59:14 INFO [logging.py:61]: Sampler state for dataloader 1 saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken/checkpoints/epoch_0025/sampler_1.bin
|
| 164 |
+
01-17 21:59:14 INFO [logging.py:61]: Random states saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken/checkpoints/epoch_0025/random_states_0.pkl
|
| 165 |
+
01-17 21:59:14 INFO [logging.py:61]: Saving the state of TrainerState to /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken/checkpoints/epoch_0025/custom_checkpoint_0.pkl
|
| 166 |
+
01-17 21:59:16 INFO [logging.py:61]: ========= Epoch 26 out of 1000 =========
|
| 167 |
+
01-17 21:59:16 INFO [logging.py:61]: Begin training...
|
| 168 |
+
01-17 22:17:25 INFO [logging.py:61]: Loss 'loss' on epoch 26: 5.375258445739746
|
| 169 |
+
01-17 22:17:25 INFO [logging.py:61]: Loss 'acc' on epoch 26: 0.33873483538627625
|
| 170 |
+
01-17 22:17:25 INFO [logging.py:61]: ========= Epoch 27 out of 1000 =========
|
| 171 |
+
01-17 22:17:25 INFO [logging.py:61]: Begin training...
|
| 172 |
+
01-17 22:35:12 INFO [logging.py:61]: Loss 'loss' on epoch 27: 5.3775506019592285
|
| 173 |
+
01-17 22:35:12 INFO [logging.py:61]: Loss 'acc' on epoch 27: 0.33919477462768555
|
| 174 |
+
01-17 22:35:12 INFO [logging.py:61]: ========= Epoch 28 out of 1000 =========
|
| 175 |
+
01-17 22:35:12 INFO [logging.py:61]: Begin training...
|
| 176 |
+
01-17 22:53:00 INFO [logging.py:61]: Loss 'loss' on epoch 28: 5.390237331390381
|
| 177 |
+
01-17 22:53:00 INFO [logging.py:61]: Loss 'acc' on epoch 28: 0.33835938572883606
|
| 178 |
+
01-17 22:53:00 INFO [logging.py:61]: ========= Epoch 29 out of 1000 =========
|
| 179 |
+
01-17 22:53:00 INFO [logging.py:61]: Begin training...
|
| 180 |
+
01-17 23:10:48 INFO [logging.py:61]: Loss 'loss' on epoch 29: 5.371942520141602
|
| 181 |
+
01-17 23:10:48 INFO [logging.py:61]: Loss 'acc' on epoch 29: 0.3405751883983612
|
| 182 |
+
01-17 23:10:48 INFO [logging.py:61]: ========= Epoch 30 out of 1000 =========
|
| 183 |
+
01-17 23:10:48 INFO [logging.py:61]: Begin training...
|
| 184 |
+
01-17 23:28:35 INFO [logging.py:61]: Loss 'loss' on epoch 30: 5.369765281677246
|
| 185 |
+
01-17 23:28:35 INFO [logging.py:61]: Loss 'acc' on epoch 30: 0.34033069014549255
|
| 186 |
+
01-17 23:28:35 INFO [logging.py:61]: Saving current state to /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken/checkpoints/epoch_0030
|
| 187 |
+
01-17 23:28:36 INFO [logging.py:61]: Model weights saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken/checkpoints/epoch_0030/pytorch_model.bin
|
| 188 |
+
01-17 23:28:45 INFO [logging.py:61]: Optimizer state saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken/checkpoints/epoch_0030/optimizer.bin
|
| 189 |
+
01-17 23:28:45 INFO [logging.py:61]: Scheduler state saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken/checkpoints/epoch_0030/scheduler.bin
|
| 190 |
+
01-17 23:28:45 INFO [logging.py:61]: Sampler state for dataloader 0 saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken/checkpoints/epoch_0030/sampler.bin
|
| 191 |
+
01-17 23:28:45 INFO [logging.py:61]: Sampler state for dataloader 1 saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken/checkpoints/epoch_0030/sampler_1.bin
|
| 192 |
+
01-17 23:28:45 INFO [logging.py:61]: Random states saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken/checkpoints/epoch_0030/random_states_0.pkl
|
| 193 |
+
01-17 23:28:45 INFO [logging.py:61]: Saving the state of TrainerState to /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_addEOSToken/checkpoints/epoch_0030/custom_checkpoint_0.pkl
|
| 194 |
+
01-17 23:28:45 INFO [logging.py:61]: ========= Epoch 31 out of 1000 =========
|
| 195 |
+
01-17 23:28:45 INFO [logging.py:61]: Begin training...
|
| 196 |
+
01-17 23:47:05 INFO [logging.py:61]: Loss 'loss' on epoch 31: 5.372854232788086
|
| 197 |
+
01-17 23:47:05 INFO [logging.py:61]: Loss 'acc' on epoch 31: 0.3400099575519562
|
| 198 |
+
01-17 23:47:05 INFO [logging.py:61]: ========= Epoch 32 out of 1000 =========
|
| 199 |
+
01-17 23:47:05 INFO [logging.py:61]: Begin training...
|
| 200 |
+
01-18 00:04:54 INFO [logging.py:61]: Loss 'loss' on epoch 32: 5.364051342010498
|
| 201 |
+
01-18 00:04:54 INFO [logging.py:61]: Loss 'acc' on epoch 32: 0.3399272561073303
|
| 202 |
+
01-18 00:04:54 INFO [logging.py:61]: ========= Epoch 33 out of 1000 =========
|
| 203 |
+
01-18 00:04:54 INFO [logging.py:61]: Begin training...
|
swin_default_LR1e-2_addEOSToken/tb_log/events.out.tfevents.1705450617.gina1.1498943.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e7788705d2ca707cbbd2b5ff1645866ddac63ebf122ee6a449e18cc4eb54983d
|
| 3 |
+
size 368795
|
swin_default_LR1e-2_addEOSToken/tb_log/events.out.tfevents.1705450617.gina1.1498944.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:34a26980b52523672ca790ef1d4aa6d2f2a9718cfd2764aacb44a47eeb891c8c
|
| 3 |
+
size 2170
|
swin_default_LR1e-2_addEOSToken/tb_log/events.out.tfevents.1705450617.gina1.1498945.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b68c141fba1b5301e8a56b4b075d598bb61910225e149e39870d88cbec923b4e
|
| 3 |
+
size 2170
|
swin_default_LR1e-2_addEOSToken/tb_log/events.out.tfevents.1705450617.gina1.1498946.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:33d56a85c87367dba2524854103cb0fa917e5a87b3dc9bfce8b167701fc1d23c
|
| 3 |
+
size 2170
|
swin_default_LR1e-2_addEOSToken/tb_log/events.out.tfevents.1705454078.gina301.610519.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e7edc77f10fdec037eaf9b43db84eb42657bbffa59d6250094ba607ffeea0e9a
|
| 3 |
+
size 276026
|
swin_default_LR1e-2_addEOSToken/tb_log/events.out.tfevents.1705454078.gina301.610520.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b0bdcf98a8c49f14de2ea16c197cc1bcc5a1a9c5d32f629ed617d4ca4e537ca6
|
| 3 |
+
size 2170
|
swin_default_LR1e-2_addEOSToken/tb_log/events.out.tfevents.1705454078.gina301.610521.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d0a55d85d50d9778f17cf404b752dc21123dbed4e1185acb5160704e2a7b7b5f
|
| 3 |
+
size 2170
|
swin_default_LR1e-2_addEOSToken/tb_log/events.out.tfevents.1705454078.gina301.610522.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5fbf5ffdcceff08c5a8b229bf89f47ef363ea6906f06a3b3a180b03a2f96d261
|
| 3 |
+
size 2170
|
swin_default_LR1e-2_addEOSToken/tb_log/events.out.tfevents.1705472223.gina1.1576669.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1ae26f96b329133f7afaa11b57d23661d4ccac43a2904a37ca5ff92189d11143
|
| 3 |
+
size 620690
|
swin_default_LR1e-2_addEOSToken/tb_log/events.out.tfevents.1705472223.gina1.1576670.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:af00695e81116da6aa753e531f6fefec091fdefc20ba5e0ad103bbd53362cb35
|
| 3 |
+
size 2170
|
swin_default_LR1e-2_addEOSToken/tb_log/events.out.tfevents.1705472223.gina1.1576671.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c494d29f5efe6acd4c545e367b5ff638ecf6d1b05344a5459db6b1ad76a7bf0f
|
| 3 |
+
size 2170
|
swin_default_LR1e-2_addEOSToken/tb_log/events.out.tfevents.1705472223.gina1.1576672.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5a4ec1e9390f233230307f57f7ec4d8cb489d80f77a7642c0d0b7cfba2143cca
|
| 3 |
+
size 2170
|
swin_default_LR1e-3_AR-NAR/config__2024_01_18--00_22_05.toml
ADDED
|
@@ -0,0 +1,89 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[[validate_dataset]]
|
| 2 |
+
path = "dataloader.Dataset"
|
| 3 |
+
|
| 4 |
+
[validate_dataset.args]
|
| 5 |
+
librispeech_dir = "~/data/LibriSpeech/LibriSpeech"
|
| 6 |
+
librispeech_metadata_fpath = "/home/xhao/proj/audiozen/recipes/librimix_sot/local/metadata/LibriSpeech/train-clean-100-24K.csv"
|
| 7 |
+
duration = 6.0
|
| 8 |
+
sr = 24000
|
| 9 |
+
num_samples = 10
|
| 10 |
+
[validate_dataset.dataloader]
|
| 11 |
+
batch_size = 1
|
| 12 |
+
num_workers = 1
|
| 13 |
+
|
| 14 |
+
[meta]
|
| 15 |
+
save_dir = "exp"
|
| 16 |
+
description = "Train a model using Generative Adversarial Networks (GANs)"
|
| 17 |
+
seed = 20220815
|
| 18 |
+
exp_id = "swin_default_LR1e-2_AR-NAR"
|
| 19 |
+
config_path = "/fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/conf/swin_default_LR1e-2_AR-NAR.toml"
|
| 20 |
+
|
| 21 |
+
[trainer]
|
| 22 |
+
path = "trainer.Trainer"
|
| 23 |
+
|
| 24 |
+
[loss_function]
|
| 25 |
+
path = "torch.nn.MSELoss"
|
| 26 |
+
|
| 27 |
+
[optimizer]
|
| 28 |
+
path = "torch.optim.AdamW"
|
| 29 |
+
|
| 30 |
+
[model]
|
| 31 |
+
path = "model_ar_nar.Model"
|
| 32 |
+
|
| 33 |
+
[acoustics]
|
| 34 |
+
n_fft = 512
|
| 35 |
+
hop_length = 128
|
| 36 |
+
win_length = 512
|
| 37 |
+
sr = 24000
|
| 38 |
+
|
| 39 |
+
[train_dataset]
|
| 40 |
+
path = "dataloader.Dataset"
|
| 41 |
+
|
| 42 |
+
[test_dataset]
|
| 43 |
+
path = "dataloader.Dataset"
|
| 44 |
+
|
| 45 |
+
[trainer.args]
|
| 46 |
+
debug = false
|
| 47 |
+
max_steps = 0
|
| 48 |
+
max_epochs = 1000
|
| 49 |
+
max_grad_norm = 1.0
|
| 50 |
+
save_max_score = true
|
| 51 |
+
save_ckpt_interval = 5
|
| 52 |
+
max_patience = 200
|
| 53 |
+
plot_norm = true
|
| 54 |
+
validation_interval = 200
|
| 55 |
+
max_num_checkpoints = 100
|
| 56 |
+
scheduler_name = "constant_schedule_with_warmup"
|
| 57 |
+
warmup_steps = 1000
|
| 58 |
+
warmup_ratio = 0.0
|
| 59 |
+
gradient_accumulation_steps = 1
|
| 60 |
+
|
| 61 |
+
[loss_function.args]
|
| 62 |
+
|
| 63 |
+
[optimizer.args]
|
| 64 |
+
lr = 0.001
|
| 65 |
+
|
| 66 |
+
[model.args]
|
| 67 |
+
|
| 68 |
+
[train_dataset.args]
|
| 69 |
+
librispeech_dir = "~/data/LibriSpeech/LibriSpeech"
|
| 70 |
+
librispeech_metadata_fpath = "/home/xhao/proj/audiozen/recipes/librimix_sot/local/metadata/LibriSpeech/train-clean-100-24K.csv"
|
| 71 |
+
duration = 6.0
|
| 72 |
+
sr = 24000
|
| 73 |
+
|
| 74 |
+
[train_dataset.dataloader]
|
| 75 |
+
batch_size = 20
|
| 76 |
+
num_workers = 10
|
| 77 |
+
drop_last = true
|
| 78 |
+
pin_memory = true
|
| 79 |
+
|
| 80 |
+
[test_dataset.args]
|
| 81 |
+
librispeech_dir = "~/data/LibriSpeech/LibriSpeech"
|
| 82 |
+
librispeech_metadata_fpath = "/home/xhao/proj/audiozen/recipes/librimix_sot/local/metadata/LibriSpeech/train-clean-100-24K.csv"
|
| 83 |
+
duration = 6.0
|
| 84 |
+
sr = 24000
|
| 85 |
+
num_samples = 10
|
| 86 |
+
|
| 87 |
+
[test_dataset.dataloader]
|
| 88 |
+
batch_size = 1
|
| 89 |
+
num_workers = 1
|
swin_default_LR1e-3_AR-NAR/config__2024_01_18--00_23_20.toml
ADDED
|
@@ -0,0 +1,89 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[[validate_dataset]]
|
| 2 |
+
path = "dataloader.Dataset"
|
| 3 |
+
|
| 4 |
+
[validate_dataset.args]
|
| 5 |
+
librispeech_dir = "~/data/LibriSpeech/LibriSpeech"
|
| 6 |
+
librispeech_metadata_fpath = "/home/xhao/proj/audiozen/recipes/librimix_sot/local/metadata/LibriSpeech/train-clean-100-24K.csv"
|
| 7 |
+
duration = 6.0
|
| 8 |
+
sr = 24000
|
| 9 |
+
num_samples = 10
|
| 10 |
+
[validate_dataset.dataloader]
|
| 11 |
+
batch_size = 1
|
| 12 |
+
num_workers = 1
|
| 13 |
+
|
| 14 |
+
[meta]
|
| 15 |
+
save_dir = "exp"
|
| 16 |
+
description = "Train a model using Generative Adversarial Networks (GANs)"
|
| 17 |
+
seed = 20220815
|
| 18 |
+
exp_id = "swin_default_LR1e-2_AR-NAR"
|
| 19 |
+
config_path = "/fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/conf/swin_default_LR1e-2_AR-NAR.toml"
|
| 20 |
+
|
| 21 |
+
[trainer]
|
| 22 |
+
path = "trainer.Trainer"
|
| 23 |
+
|
| 24 |
+
[loss_function]
|
| 25 |
+
path = "torch.nn.MSELoss"
|
| 26 |
+
|
| 27 |
+
[optimizer]
|
| 28 |
+
path = "torch.optim.AdamW"
|
| 29 |
+
|
| 30 |
+
[model]
|
| 31 |
+
path = "model_ar_nar.Model"
|
| 32 |
+
|
| 33 |
+
[acoustics]
|
| 34 |
+
n_fft = 512
|
| 35 |
+
hop_length = 128
|
| 36 |
+
win_length = 512
|
| 37 |
+
sr = 24000
|
| 38 |
+
|
| 39 |
+
[train_dataset]
|
| 40 |
+
path = "dataloader.Dataset"
|
| 41 |
+
|
| 42 |
+
[test_dataset]
|
| 43 |
+
path = "dataloader.Dataset"
|
| 44 |
+
|
| 45 |
+
[trainer.args]
|
| 46 |
+
debug = false
|
| 47 |
+
max_steps = 0
|
| 48 |
+
max_epochs = 1000
|
| 49 |
+
max_grad_norm = 1.0
|
| 50 |
+
save_max_score = true
|
| 51 |
+
save_ckpt_interval = 5
|
| 52 |
+
max_patience = 200
|
| 53 |
+
plot_norm = true
|
| 54 |
+
validation_interval = 200
|
| 55 |
+
max_num_checkpoints = 100
|
| 56 |
+
scheduler_name = "constant_schedule_with_warmup"
|
| 57 |
+
warmup_steps = 1000
|
| 58 |
+
warmup_ratio = 0.0
|
| 59 |
+
gradient_accumulation_steps = 1
|
| 60 |
+
|
| 61 |
+
[loss_function.args]
|
| 62 |
+
|
| 63 |
+
[optimizer.args]
|
| 64 |
+
lr = 0.001
|
| 65 |
+
|
| 66 |
+
[model.args]
|
| 67 |
+
|
| 68 |
+
[train_dataset.args]
|
| 69 |
+
librispeech_dir = "~/data/LibriSpeech/LibriSpeech"
|
| 70 |
+
librispeech_metadata_fpath = "/home/xhao/proj/audiozen/recipes/librimix_sot/local/metadata/LibriSpeech/train-clean-100-24K.csv"
|
| 71 |
+
duration = 6.0
|
| 72 |
+
sr = 24000
|
| 73 |
+
|
| 74 |
+
[train_dataset.dataloader]
|
| 75 |
+
batch_size = 20
|
| 76 |
+
num_workers = 10
|
| 77 |
+
drop_last = true
|
| 78 |
+
pin_memory = true
|
| 79 |
+
|
| 80 |
+
[test_dataset.args]
|
| 81 |
+
librispeech_dir = "~/data/LibriSpeech/LibriSpeech"
|
| 82 |
+
librispeech_metadata_fpath = "/home/xhao/proj/audiozen/recipes/librimix_sot/local/metadata/LibriSpeech/train-clean-100-24K.csv"
|
| 83 |
+
duration = 6.0
|
| 84 |
+
sr = 24000
|
| 85 |
+
num_samples = 10
|
| 86 |
+
|
| 87 |
+
[test_dataset.dataloader]
|
| 88 |
+
batch_size = 1
|
| 89 |
+
num_workers = 1
|
swin_default_LR1e-3_AR-NAR/config__2024_01_18--00_33_25.toml
ADDED
|
@@ -0,0 +1,89 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[[validate_dataset]]
|
| 2 |
+
path = "dataloader.Dataset"
|
| 3 |
+
|
| 4 |
+
[validate_dataset.args]
|
| 5 |
+
librispeech_dir = "~/data/LibriSpeech/LibriSpeech"
|
| 6 |
+
librispeech_metadata_fpath = "/home/xhao/proj/audiozen/recipes/librimix_sot/local/metadata/LibriSpeech/train-clean-100-24K.csv"
|
| 7 |
+
duration = 6.0
|
| 8 |
+
sr = 24000
|
| 9 |
+
num_samples = 10
|
| 10 |
+
[validate_dataset.dataloader]
|
| 11 |
+
batch_size = 1
|
| 12 |
+
num_workers = 1
|
| 13 |
+
|
| 14 |
+
[meta]
|
| 15 |
+
save_dir = "exp"
|
| 16 |
+
description = "Train a model using Generative Adversarial Networks (GANs)"
|
| 17 |
+
seed = 20220815
|
| 18 |
+
exp_id = "swin_default_LR1e-2_AR-NAR"
|
| 19 |
+
config_path = "/fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/conf/swin_default_LR1e-2_AR-NAR.toml"
|
| 20 |
+
|
| 21 |
+
[trainer]
|
| 22 |
+
path = "trainer.Trainer"
|
| 23 |
+
|
| 24 |
+
[loss_function]
|
| 25 |
+
path = "torch.nn.MSELoss"
|
| 26 |
+
|
| 27 |
+
[optimizer]
|
| 28 |
+
path = "torch.optim.AdamW"
|
| 29 |
+
|
| 30 |
+
[model]
|
| 31 |
+
path = "model_ar_nar.Model"
|
| 32 |
+
|
| 33 |
+
[acoustics]
|
| 34 |
+
n_fft = 512
|
| 35 |
+
hop_length = 128
|
| 36 |
+
win_length = 512
|
| 37 |
+
sr = 24000
|
| 38 |
+
|
| 39 |
+
[train_dataset]
|
| 40 |
+
path = "dataloader.Dataset"
|
| 41 |
+
|
| 42 |
+
[test_dataset]
|
| 43 |
+
path = "dataloader.Dataset"
|
| 44 |
+
|
| 45 |
+
[trainer.args]
|
| 46 |
+
debug = false
|
| 47 |
+
max_steps = 0
|
| 48 |
+
max_epochs = 1000
|
| 49 |
+
max_grad_norm = 1.0
|
| 50 |
+
save_max_score = true
|
| 51 |
+
save_ckpt_interval = 5
|
| 52 |
+
max_patience = 200
|
| 53 |
+
plot_norm = true
|
| 54 |
+
validation_interval = 200
|
| 55 |
+
max_num_checkpoints = 100
|
| 56 |
+
scheduler_name = "constant_schedule_with_warmup"
|
| 57 |
+
warmup_steps = 1000
|
| 58 |
+
warmup_ratio = 0.0
|
| 59 |
+
gradient_accumulation_steps = 1
|
| 60 |
+
|
| 61 |
+
[loss_function.args]
|
| 62 |
+
|
| 63 |
+
[optimizer.args]
|
| 64 |
+
lr = 0.001
|
| 65 |
+
|
| 66 |
+
[model.args]
|
| 67 |
+
|
| 68 |
+
[train_dataset.args]
|
| 69 |
+
librispeech_dir = "~/data/LibriSpeech/LibriSpeech"
|
| 70 |
+
librispeech_metadata_fpath = "/home/xhao/proj/audiozen/recipes/librimix_sot/local/metadata/LibriSpeech/train-clean-100-24K.csv"
|
| 71 |
+
duration = 6.0
|
| 72 |
+
sr = 24000
|
| 73 |
+
|
| 74 |
+
[train_dataset.dataloader]
|
| 75 |
+
batch_size = 20
|
| 76 |
+
num_workers = 10
|
| 77 |
+
drop_last = true
|
| 78 |
+
pin_memory = true
|
| 79 |
+
|
| 80 |
+
[test_dataset.args]
|
| 81 |
+
librispeech_dir = "~/data/LibriSpeech/LibriSpeech"
|
| 82 |
+
librispeech_metadata_fpath = "/home/xhao/proj/audiozen/recipes/librimix_sot/local/metadata/LibriSpeech/train-clean-100-24K.csv"
|
| 83 |
+
duration = 6.0
|
| 84 |
+
sr = 24000
|
| 85 |
+
num_samples = 10
|
| 86 |
+
|
| 87 |
+
[test_dataset.dataloader]
|
| 88 |
+
batch_size = 1
|
| 89 |
+
num_workers = 1
|
swin_default_LR1e-3_AR-NAR/config__2024_01_18--13_13_46.toml
ADDED
|
@@ -0,0 +1,89 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[[validate_dataset]]
|
| 2 |
+
path = "dataloader.Dataset"
|
| 3 |
+
|
| 4 |
+
[validate_dataset.args]
|
| 5 |
+
librispeech_dir = "~/data/LibriSpeech/LibriSpeech"
|
| 6 |
+
librispeech_metadata_fpath = "/home/xhao/proj/audiozen/recipes/librimix_sot/local/metadata/LibriSpeech/train-clean-100-24K.csv"
|
| 7 |
+
duration = 6.0
|
| 8 |
+
sr = 24000
|
| 9 |
+
num_samples = 10
|
| 10 |
+
[validate_dataset.dataloader]
|
| 11 |
+
batch_size = 1
|
| 12 |
+
num_workers = 1
|
| 13 |
+
|
| 14 |
+
[meta]
|
| 15 |
+
save_dir = "exp"
|
| 16 |
+
description = "Train a model using Generative Adversarial Networks (GANs)"
|
| 17 |
+
seed = 20220815
|
| 18 |
+
exp_id = "swin_default_LR1e-3_AR-NAR"
|
| 19 |
+
config_path = "/fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/conf/swin_default_LR1e-3_AR-NAR.toml"
|
| 20 |
+
|
| 21 |
+
[trainer]
|
| 22 |
+
path = "trainer.Trainer"
|
| 23 |
+
|
| 24 |
+
[loss_function]
|
| 25 |
+
path = "torch.nn.MSELoss"
|
| 26 |
+
|
| 27 |
+
[optimizer]
|
| 28 |
+
path = "torch.optim.AdamW"
|
| 29 |
+
|
| 30 |
+
[model]
|
| 31 |
+
path = "model_ar_nar.Model"
|
| 32 |
+
|
| 33 |
+
[acoustics]
|
| 34 |
+
n_fft = 512
|
| 35 |
+
hop_length = 128
|
| 36 |
+
win_length = 512
|
| 37 |
+
sr = 24000
|
| 38 |
+
|
| 39 |
+
[train_dataset]
|
| 40 |
+
path = "dataloader.Dataset"
|
| 41 |
+
|
| 42 |
+
[test_dataset]
|
| 43 |
+
path = "dataloader.Dataset"
|
| 44 |
+
|
| 45 |
+
[trainer.args]
|
| 46 |
+
debug = false
|
| 47 |
+
max_steps = 0
|
| 48 |
+
max_epochs = 1000
|
| 49 |
+
max_grad_norm = 1.0
|
| 50 |
+
save_max_score = true
|
| 51 |
+
save_ckpt_interval = 5
|
| 52 |
+
max_patience = 200
|
| 53 |
+
plot_norm = true
|
| 54 |
+
validation_interval = 200
|
| 55 |
+
max_num_checkpoints = 100
|
| 56 |
+
scheduler_name = "constant_schedule_with_warmup"
|
| 57 |
+
warmup_steps = 1000
|
| 58 |
+
warmup_ratio = 0.0
|
| 59 |
+
gradient_accumulation_steps = 1
|
| 60 |
+
|
| 61 |
+
[loss_function.args]
|
| 62 |
+
|
| 63 |
+
[optimizer.args]
|
| 64 |
+
lr = 0.001
|
| 65 |
+
|
| 66 |
+
[model.args]
|
| 67 |
+
|
| 68 |
+
[train_dataset.args]
|
| 69 |
+
librispeech_dir = "~/data/LibriSpeech/LibriSpeech"
|
| 70 |
+
librispeech_metadata_fpath = "/home/xhao/proj/audiozen/recipes/librimix_sot/local/metadata/LibriSpeech/train-clean-100-24K.csv"
|
| 71 |
+
duration = 6.0
|
| 72 |
+
sr = 24000
|
| 73 |
+
|
| 74 |
+
[train_dataset.dataloader]
|
| 75 |
+
batch_size = 20
|
| 76 |
+
num_workers = 10
|
| 77 |
+
drop_last = true
|
| 78 |
+
pin_memory = true
|
| 79 |
+
|
| 80 |
+
[test_dataset.args]
|
| 81 |
+
librispeech_dir = "~/data/LibriSpeech/LibriSpeech"
|
| 82 |
+
librispeech_metadata_fpath = "/home/xhao/proj/audiozen/recipes/librimix_sot/local/metadata/LibriSpeech/train-clean-100-24K.csv"
|
| 83 |
+
duration = 6.0
|
| 84 |
+
sr = 24000
|
| 85 |
+
num_samples = 10
|
| 86 |
+
|
| 87 |
+
[test_dataset.dataloader]
|
| 88 |
+
batch_size = 1
|
| 89 |
+
num_workers = 1
|
swin_default_LR1e-3_AR-NAR/config__2024_01_18--13_14_43.toml
ADDED
|
@@ -0,0 +1,89 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[[validate_dataset]]
|
| 2 |
+
path = "dataloader.Dataset"
|
| 3 |
+
|
| 4 |
+
[validate_dataset.args]
|
| 5 |
+
librispeech_dir = "~/data/LibriSpeech/LibriSpeech"
|
| 6 |
+
librispeech_metadata_fpath = "/home/xhao/proj/audiozen/recipes/librimix_sot/local/metadata/LibriSpeech/train-clean-100-24K.csv"
|
| 7 |
+
duration = 6.0
|
| 8 |
+
sr = 24000
|
| 9 |
+
num_samples = 10
|
| 10 |
+
[validate_dataset.dataloader]
|
| 11 |
+
batch_size = 1
|
| 12 |
+
num_workers = 1
|
| 13 |
+
|
| 14 |
+
[meta]
|
| 15 |
+
save_dir = "exp"
|
| 16 |
+
description = "Train a model using Generative Adversarial Networks (GANs)"
|
| 17 |
+
seed = 20220815
|
| 18 |
+
exp_id = "swin_default_LR1e-3_AR-NAR"
|
| 19 |
+
config_path = "/fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/conf/swin_default_LR1e-3_AR-NAR.toml"
|
| 20 |
+
|
| 21 |
+
[trainer]
|
| 22 |
+
path = "trainer.Trainer"
|
| 23 |
+
|
| 24 |
+
[loss_function]
|
| 25 |
+
path = "torch.nn.MSELoss"
|
| 26 |
+
|
| 27 |
+
[optimizer]
|
| 28 |
+
path = "torch.optim.AdamW"
|
| 29 |
+
|
| 30 |
+
[model]
|
| 31 |
+
path = "model_ar_nar.Model"
|
| 32 |
+
|
| 33 |
+
[acoustics]
|
| 34 |
+
n_fft = 512
|
| 35 |
+
hop_length = 128
|
| 36 |
+
win_length = 512
|
| 37 |
+
sr = 24000
|
| 38 |
+
|
| 39 |
+
[train_dataset]
|
| 40 |
+
path = "dataloader.Dataset"
|
| 41 |
+
|
| 42 |
+
[test_dataset]
|
| 43 |
+
path = "dataloader.Dataset"
|
| 44 |
+
|
| 45 |
+
[trainer.args]
|
| 46 |
+
debug = false
|
| 47 |
+
max_steps = 0
|
| 48 |
+
max_epochs = 1000
|
| 49 |
+
max_grad_norm = 1.0
|
| 50 |
+
save_max_score = true
|
| 51 |
+
save_ckpt_interval = 5
|
| 52 |
+
max_patience = 200
|
| 53 |
+
plot_norm = true
|
| 54 |
+
validation_interval = 200
|
| 55 |
+
max_num_checkpoints = 100
|
| 56 |
+
scheduler_name = "constant_schedule_with_warmup"
|
| 57 |
+
warmup_steps = 1000
|
| 58 |
+
warmup_ratio = 0.0
|
| 59 |
+
gradient_accumulation_steps = 1
|
| 60 |
+
|
| 61 |
+
[loss_function.args]
|
| 62 |
+
|
| 63 |
+
[optimizer.args]
|
| 64 |
+
lr = 0.001
|
| 65 |
+
|
| 66 |
+
[model.args]
|
| 67 |
+
|
| 68 |
+
[train_dataset.args]
|
| 69 |
+
librispeech_dir = "~/data/LibriSpeech/LibriSpeech"
|
| 70 |
+
librispeech_metadata_fpath = "/home/xhao/proj/audiozen/recipes/librimix_sot/local/metadata/LibriSpeech/train-clean-100-24K.csv"
|
| 71 |
+
duration = 6.0
|
| 72 |
+
sr = 24000
|
| 73 |
+
|
| 74 |
+
[train_dataset.dataloader]
|
| 75 |
+
batch_size = 20
|
| 76 |
+
num_workers = 10
|
| 77 |
+
drop_last = true
|
| 78 |
+
pin_memory = true
|
| 79 |
+
|
| 80 |
+
[test_dataset.args]
|
| 81 |
+
librispeech_dir = "~/data/LibriSpeech/LibriSpeech"
|
| 82 |
+
librispeech_metadata_fpath = "/home/xhao/proj/audiozen/recipes/librimix_sot/local/metadata/LibriSpeech/train-clean-100-24K.csv"
|
| 83 |
+
duration = 6.0
|
| 84 |
+
sr = 24000
|
| 85 |
+
num_samples = 10
|
| 86 |
+
|
| 87 |
+
[test_dataset.dataloader]
|
| 88 |
+
batch_size = 1
|
| 89 |
+
num_workers = 1
|
swin_default_LR1e-3_AR-NAR/swin_default_LR1e-2_AR-NAR_2024_01_18--00_21_17.log
ADDED
|
@@ -0,0 +1,85 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
01-18 00:21:17 INFO [logger.py:80]: Initialized logger with log file in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR.
|
| 2 |
+
01-18 00:21:17 INFO [logger.py:80]: Initialized logger with log file in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR.
|
| 3 |
+
01-18 00:21:17 INFO [logger.py:80]: Initialized logger with log file in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR.
|
| 4 |
+
01-18 00:21:17 INFO [logger.py:80]: Initialized logger with log file in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR.
|
| 5 |
+
01-18 00:22:18 INFO [logging.py:61]: Configuration file is saved to /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/config__2024_01_18--00_22_05.toml.
|
| 6 |
+
01-18 00:22:18 INFO [logging.py:61]: Environment information:
|
| 7 |
+
- `Accelerate` version: 0.26.1
|
| 8 |
+
- Platform: Linux-5.14.0-362.13.1.el9_3.x86_64-x86_64-with-glibc2.34
|
| 9 |
+
- Python version: 3.10.13
|
| 10 |
+
- Numpy version: 1.26.3
|
| 11 |
+
- PyTorch version (GPU?): 2.1.2 (True)
|
| 12 |
+
- System RAM: 503.48 GB
|
| 13 |
+
- GPU Available: True
|
| 14 |
+
- GPU IDs: 4
|
| 15 |
+
- GPU type: NVIDIA A100-SXM4-80GB
|
| 16 |
+
01-18 00:22:18 INFO [logging.py:61]:
|
| 17 |
+
===============================================================================================
|
| 18 |
+
Layer (type:depth-idx) Param #
|
| 19 |
+
===============================================================================================
|
| 20 |
+
DistributedDataParallel --
|
| 21 |
+
├─Model: 1-1 --
|
| 22 |
+
│ └─EncodecModel: 2-1 --
|
| 23 |
+
│ │ └─EncodecEncoder: 3-1 (7,425,792)
|
| 24 |
+
│ │ └─EncodecDecoder: 3-2 (7,426,018)
|
| 25 |
+
│ │ └─EncodecResidualVectorQuantizer: 3-3 --
|
| 26 |
+
│ └─TokenEmbedding: 2-2 --
|
| 27 |
+
│ │ └─Dropout: 3-4 --
|
| 28 |
+
│ │ └─Embedding: 3-5 1,049,600
|
| 29 |
+
│ └─Identity: 2-3 --
|
| 30 |
+
│ └─SinePositionalEmbedding: 2-4 1
|
| 31 |
+
│ │ └─Dropout: 3-6 --
|
| 32 |
+
│ └─TransformerEncoder: 2-5 --
|
| 33 |
+
│ │ └─ModuleList: 3-7 151,154,688
|
| 34 |
+
│ │ └─LayerNorm: 3-8 2,048
|
| 35 |
+
│ └─Linear: 2-6 1,049,600
|
| 36 |
+
│ └─MulticlassAccuracy: 2-7 --
|
| 37 |
+
│ └─TokenEmbedding: 2-8 --
|
| 38 |
+
│ │ └─Dropout: 3-9 --
|
| 39 |
+
│ │ └─Embedding: 3-10 1,048,576
|
| 40 |
+
│ └─ModuleList: 2-9 --
|
| 41 |
+
│ │ └─TokenEmbedding: 3-11 1,049,600
|
| 42 |
+
│ │ └─TokenEmbedding: 3-12 1,048,576
|
| 43 |
+
│ │ └─TokenEmbedding: 3-13 1,048,576
|
| 44 |
+
│ │ └─TokenEmbedding: 3-14 1,048,576
|
| 45 |
+
│ │ └─TokenEmbedding: 3-15 1,048,576
|
| 46 |
+
│ │ └─TokenEmbedding: 3-16 1,048,576
|
| 47 |
+
│ │ └─TokenEmbedding: 3-17 1,048,576
|
| 48 |
+
│ │ └─TokenEmbedding: 3-18 1,048,576
|
| 49 |
+
│ └─Identity: 2-10 --
|
| 50 |
+
│ └─SinePositionalEmbedding: 2-11 1
|
| 51 |
+
│ │ └─Dropout: 3-19 --
|
| 52 |
+
│ └─TransformerEncoder: 2-12 --
|
| 53 |
+
│ │ └─ModuleList: 3-20 201,535,488
|
| 54 |
+
│ │ └─AdaptiveLayerNorm: 3-21 2,101,248
|
| 55 |
+
│ └─ModuleList: 2-13 --
|
| 56 |
+
│ │ └─Linear: 3-22 1,048,576
|
| 57 |
+
│ │ └─Linear: 3-23 1,048,576
|
| 58 |
+
│ │ └─Linear: 3-24 1,048,576
|
| 59 |
+
│ │ └─Linear: 3-25 1,048,576
|
| 60 |
+
│ │ └─Linear: 3-26 1,048,576
|
| 61 |
+
│ │ └─Linear: 3-27 1,048,576
|
| 62 |
+
│ │ └─Linear: 3-28 1,048,576
|
| 63 |
+
│ └─ModuleList: 2-14 --
|
| 64 |
+
│ │ └─TokenEmbedding: 3-29 1,024
|
| 65 |
+
│ │ └─TokenEmbedding: 3-30 1,024
|
| 66 |
+
│ │ └─TokenEmbedding: 3-31 1,024
|
| 67 |
+
│ │ └─TokenEmbedding: 3-32 1,024
|
| 68 |
+
│ │ └─TokenEmbedding: 3-33 1,024
|
| 69 |
+
│ │ └─TokenEmbedding: 3-34 1,024
|
| 70 |
+
│ │ └─TokenEmbedding: 3-35 1,024
|
| 71 |
+
│ └─MulticlassAccuracy: 2-15 --
|
| 72 |
+
===============================================================================================
|
| 73 |
+
Total params: 388,529,892
|
| 74 |
+
Trainable params: 373,678,081
|
| 75 |
+
Non-trainable params: 14,851,811
|
| 76 |
+
===============================================================================================
|
| 77 |
+
01-18 00:22:18 INFO [logging.py:61]: Training control variables:
|
| 78 |
+
01-18 00:22:18 INFO [logging.py:61]: `steps_per_epoch`: 500
|
| 79 |
+
01-18 00:22:18 INFO [logging.py:61]: Gradient accumulation steps: 1
|
| 80 |
+
01-18 00:22:18 INFO [logging.py:61]: `update_steps_per_epoch`: 500
|
| 81 |
+
01-18 00:22:18 INFO [logging.py:61]: `max_steps`: 500000
|
| 82 |
+
01-18 00:22:18 INFO [logging.py:61]: `max_epochs`: 1000
|
| 83 |
+
01-18 00:22:18 INFO [logging.py:61]: warmup_steps=1000. warmup_ratio will be ignored.
|
| 84 |
+
01-18 00:22:18 INFO [logging.py:61]: ========= Epoch 1 out of 1000 =========
|
| 85 |
+
01-18 00:22:18 INFO [logging.py:61]: Begin training...
|
swin_default_LR1e-3_AR-NAR/swin_default_LR1e-2_AR-NAR_2024_01_18--00_23_16.log
ADDED
|
@@ -0,0 +1,85 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
01-18 00:23:16 INFO [logger.py:80]: Initialized logger with log file in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR.
|
| 2 |
+
01-18 00:23:16 INFO [logger.py:80]: Initialized logger with log file in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR.
|
| 3 |
+
01-18 00:23:16 INFO [logger.py:80]: Initialized logger with log file in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR.
|
| 4 |
+
01-18 00:23:16 INFO [logger.py:80]: Initialized logger with log file in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR.
|
| 5 |
+
01-18 00:23:22 INFO [logging.py:61]: Configuration file is saved to /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/config__2024_01_18--00_23_20.toml.
|
| 6 |
+
01-18 00:23:22 INFO [logging.py:61]: Environment information:
|
| 7 |
+
- `Accelerate` version: 0.26.1
|
| 8 |
+
- Platform: Linux-5.14.0-362.13.1.el9_3.x86_64-x86_64-with-glibc2.34
|
| 9 |
+
- Python version: 3.10.13
|
| 10 |
+
- Numpy version: 1.26.3
|
| 11 |
+
- PyTorch version (GPU?): 2.1.2 (True)
|
| 12 |
+
- System RAM: 503.48 GB
|
| 13 |
+
- GPU Available: True
|
| 14 |
+
- GPU IDs: 4
|
| 15 |
+
- GPU type: NVIDIA A100-SXM4-80GB
|
| 16 |
+
01-18 00:23:22 INFO [logging.py:61]:
|
| 17 |
+
===============================================================================================
|
| 18 |
+
Layer (type:depth-idx) Param #
|
| 19 |
+
===============================================================================================
|
| 20 |
+
DistributedDataParallel --
|
| 21 |
+
├─Model: 1-1 --
|
| 22 |
+
│ └─EncodecModel: 2-1 --
|
| 23 |
+
│ │ └─EncodecEncoder: 3-1 (7,425,792)
|
| 24 |
+
│ │ └─EncodecDecoder: 3-2 (7,426,018)
|
| 25 |
+
│ │ └─EncodecResidualVectorQuantizer: 3-3 --
|
| 26 |
+
│ └─TokenEmbedding: 2-2 --
|
| 27 |
+
│ │ └─Dropout: 3-4 --
|
| 28 |
+
│ │ └─Embedding: 3-5 524,800
|
| 29 |
+
│ └─Identity: 2-3 --
|
| 30 |
+
│ └─SinePositionalEmbedding: 2-4 1
|
| 31 |
+
│ │ └─Dropout: 3-6 --
|
| 32 |
+
│ └─TransformerEncoder: 2-5 --
|
| 33 |
+
│ │ └─ModuleList: 3-7 37,828,608
|
| 34 |
+
│ │ └─LayerNorm: 3-8 1,024
|
| 35 |
+
│ └─Linear: 2-6 524,800
|
| 36 |
+
│ └─MulticlassAccuracy: 2-7 --
|
| 37 |
+
│ └─TokenEmbedding: 2-8 --
|
| 38 |
+
│ │ └─Dropout: 3-9 --
|
| 39 |
+
│ │ └─Embedding: 3-10 524,288
|
| 40 |
+
│ └─ModuleList: 2-9 --
|
| 41 |
+
│ │ └─TokenEmbedding: 3-11 524,800
|
| 42 |
+
│ │ └─TokenEmbedding: 3-12 524,288
|
| 43 |
+
│ │ └─TokenEmbedding: 3-13 524,288
|
| 44 |
+
│ │ └─TokenEmbedding: 3-14 524,288
|
| 45 |
+
│ │ └─TokenEmbedding: 3-15 524,288
|
| 46 |
+
│ │ └─TokenEmbedding: 3-16 524,288
|
| 47 |
+
│ │ └─TokenEmbedding: 3-17 524,288
|
| 48 |
+
│ │ └─TokenEmbedding: 3-18 524,288
|
| 49 |
+
│ └─Identity: 2-10 --
|
| 50 |
+
│ └─SinePositionalEmbedding: 2-11 1
|
| 51 |
+
│ │ └─Dropout: 3-19 --
|
| 52 |
+
│ └─TransformerEncoder: 2-12 --
|
| 53 |
+
│ │ └─ModuleList: 3-20 50,436,096
|
| 54 |
+
│ │ └─AdaptiveLayerNorm: 3-21 526,336
|
| 55 |
+
│ └─ModuleList: 2-13 --
|
| 56 |
+
│ │ └─Linear: 3-22 524,288
|
| 57 |
+
│ │ └─Linear: 3-23 524,288
|
| 58 |
+
│ │ └─Linear: 3-24 524,288
|
| 59 |
+
│ │ └─Linear: 3-25 524,288
|
| 60 |
+
│ │ └─Linear: 3-26 524,288
|
| 61 |
+
��� │ └─Linear: 3-27 524,288
|
| 62 |
+
│ │ └─Linear: 3-28 524,288
|
| 63 |
+
│ └─ModuleList: 2-14 --
|
| 64 |
+
│ │ └─TokenEmbedding: 3-29 512
|
| 65 |
+
│ │ └─TokenEmbedding: 3-30 512
|
| 66 |
+
│ │ └─TokenEmbedding: 3-31 512
|
| 67 |
+
│ │ └─TokenEmbedding: 3-32 512
|
| 68 |
+
│ │ └─TokenEmbedding: 3-33 512
|
| 69 |
+
│ │ └─TokenEmbedding: 3-34 512
|
| 70 |
+
│ │ └─TokenEmbedding: 3-35 512
|
| 71 |
+
│ └─MulticlassAccuracy: 2-15 --
|
| 72 |
+
===============================================================================================
|
| 73 |
+
Total params: 113,086,180
|
| 74 |
+
Trainable params: 98,234,369
|
| 75 |
+
Non-trainable params: 14,851,811
|
| 76 |
+
===============================================================================================
|
| 77 |
+
01-18 00:23:22 INFO [logging.py:61]: Training control variables:
|
| 78 |
+
01-18 00:23:22 INFO [logging.py:61]: `steps_per_epoch`: 500
|
| 79 |
+
01-18 00:23:22 INFO [logging.py:61]: Gradient accumulation steps: 1
|
| 80 |
+
01-18 00:23:22 INFO [logging.py:61]: `update_steps_per_epoch`: 500
|
| 81 |
+
01-18 00:23:22 INFO [logging.py:61]: `max_steps`: 500000
|
| 82 |
+
01-18 00:23:22 INFO [logging.py:61]: `max_epochs`: 1000
|
| 83 |
+
01-18 00:23:22 INFO [logging.py:61]: warmup_steps=1000. warmup_ratio will be ignored.
|
| 84 |
+
01-18 00:23:22 INFO [logging.py:61]: ========= Epoch 1 out of 1000 =========
|
| 85 |
+
01-18 00:23:22 INFO [logging.py:61]: Begin training...
|
swin_default_LR1e-3_AR-NAR/swin_default_LR1e-2_AR-NAR_2024_01_18--00_33_02.log
ADDED
|
@@ -0,0 +1,515 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
01-18 00:33:02 INFO [logger.py:80]: Initialized logger with log file in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR.
|
| 2 |
+
01-18 00:33:02 INFO [logger.py:80]: Initialized logger with log file in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR.
|
| 3 |
+
01-18 00:33:02 INFO [logger.py:80]: Initialized logger with log file in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR.
|
| 4 |
+
01-18 00:33:02 INFO [logger.py:80]: Initialized logger with log file in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR.
|
| 5 |
+
01-18 00:33:32 INFO [logging.py:61]: Configuration file is saved to /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/config__2024_01_18--00_33_25.toml.
|
| 6 |
+
01-18 00:33:32 INFO [logging.py:61]: Environment information:
|
| 7 |
+
- `Accelerate` version: 0.26.1
|
| 8 |
+
- Platform: Linux-5.14.0-362.13.1.el9_3.x86_64-x86_64-with-glibc2.34
|
| 9 |
+
- Python version: 3.10.13
|
| 10 |
+
- Numpy version: 1.26.3
|
| 11 |
+
- PyTorch version (GPU?): 2.1.2 (True)
|
| 12 |
+
- System RAM: 503.48 GB
|
| 13 |
+
- GPU Available: True
|
| 14 |
+
- GPU IDs: 4
|
| 15 |
+
- GPU type: NVIDIA A100-SXM4-80GB
|
| 16 |
+
01-18 00:33:32 INFO [logging.py:61]:
|
| 17 |
+
===============================================================================================
|
| 18 |
+
Layer (type:depth-idx) Param #
|
| 19 |
+
===============================================================================================
|
| 20 |
+
DistributedDataParallel --
|
| 21 |
+
├─Model: 1-1 --
|
| 22 |
+
│ └─EncodecModel: 2-1 --
|
| 23 |
+
│ │ └─EncodecEncoder: 3-1 (7,425,792)
|
| 24 |
+
│ │ └─EncodecDecoder: 3-2 (7,426,018)
|
| 25 |
+
│ │ └─EncodecResidualVectorQuantizer: 3-3 --
|
| 26 |
+
│ └─TokenEmbedding: 2-2 --
|
| 27 |
+
│ │ └─Dropout: 3-4 --
|
| 28 |
+
│ │ └─Embedding: 3-5 524,800
|
| 29 |
+
│ └─Identity: 2-3 --
|
| 30 |
+
│ └─SinePositionalEmbedding: 2-4 1
|
| 31 |
+
│ │ └─Dropout: 3-6 --
|
| 32 |
+
│ └─TransformerEncoder: 2-5 --
|
| 33 |
+
│ │ └─ModuleList: 3-7 37,828,608
|
| 34 |
+
│ │ └─LayerNorm: 3-8 1,024
|
| 35 |
+
│ └─Linear: 2-6 524,800
|
| 36 |
+
│ └─MulticlassAccuracy: 2-7 --
|
| 37 |
+
│ └─TokenEmbedding: 2-8 --
|
| 38 |
+
│ │ └─Dropout: 3-9 --
|
| 39 |
+
│ │ └─Embedding: 3-10 524,288
|
| 40 |
+
│ └─ModuleList: 2-9 --
|
| 41 |
+
│ │ └─TokenEmbedding: 3-11 524,800
|
| 42 |
+
│ │ └─TokenEmbedding: 3-12 524,288
|
| 43 |
+
│ │ └─TokenEmbedding: 3-13 524,288
|
| 44 |
+
│ │ └─TokenEmbedding: 3-14 524,288
|
| 45 |
+
│ │ └─TokenEmbedding: 3-15 524,288
|
| 46 |
+
│ │ └─TokenEmbedding: 3-16 524,288
|
| 47 |
+
│ │ └─TokenEmbedding: 3-17 524,288
|
| 48 |
+
│ │ └─TokenEmbedding: 3-18 524,288
|
| 49 |
+
│ └─Identity: 2-10 --
|
| 50 |
+
│ └─SinePositionalEmbedding: 2-11 1
|
| 51 |
+
│ │ └─Dropout: 3-19 --
|
| 52 |
+
│ └─TransformerEncoder: 2-12 --
|
| 53 |
+
│ │ └─ModuleList: 3-20 50,436,096
|
| 54 |
+
│ │ └─AdaptiveLayerNorm: 3-21 526,336
|
| 55 |
+
│ └─ModuleList: 2-13 --
|
| 56 |
+
│ │ └─Linear: 3-22 524,288
|
| 57 |
+
│ │ └─Linear: 3-23 524,288
|
| 58 |
+
│ │ └─Linear: 3-24 524,288
|
| 59 |
+
│ │ └─Linear: 3-25 524,288
|
| 60 |
+
│ │ └─Linear: 3-26 524,288
|
| 61 |
+
│ │ └─Linear: 3-27 524,288
|
| 62 |
+
│ │ └─Linear: 3-28 524,288
|
| 63 |
+
│ └─ModuleList: 2-14 --
|
| 64 |
+
│ │ └─TokenEmbedding: 3-29 512
|
| 65 |
+
│ │ └─TokenEmbedding: 3-30 512
|
| 66 |
+
│ │ └─TokenEmbedding: 3-31 512
|
| 67 |
+
│ │ └─TokenEmbedding: 3-32 512
|
| 68 |
+
│ │ └─TokenEmbedding: 3-33 512
|
| 69 |
+
│ │ └─TokenEmbedding: 3-34 512
|
| 70 |
+
│ │ └─TokenEmbedding: 3-35 512
|
| 71 |
+
│ └─MulticlassAccuracy: 2-15 --
|
| 72 |
+
===============================================================================================
|
| 73 |
+
Total params: 113,086,180
|
| 74 |
+
Trainable params: 98,234,369
|
| 75 |
+
Non-trainable params: 14,851,811
|
| 76 |
+
===============================================================================================
|
| 77 |
+
01-18 00:33:32 INFO [logging.py:61]: Training control variables:
|
| 78 |
+
01-18 00:33:32 INFO [logging.py:61]: `steps_per_epoch`: 500
|
| 79 |
+
01-18 00:33:32 INFO [logging.py:61]: Gradient accumulation steps: 1
|
| 80 |
+
01-18 00:33:32 INFO [logging.py:61]: `update_steps_per_epoch`: 500
|
| 81 |
+
01-18 00:33:32 INFO [logging.py:61]: `max_steps`: 500000
|
| 82 |
+
01-18 00:33:32 INFO [logging.py:61]: `max_epochs`: 1000
|
| 83 |
+
01-18 00:33:32 INFO [logging.py:61]: warmup_steps=1000. warmup_ratio will be ignored.
|
| 84 |
+
01-18 00:33:32 INFO [logging.py:61]: ========= Epoch 1 out of 1000 =========
|
| 85 |
+
01-18 00:33:32 INFO [logging.py:61]: Begin training...
|
| 86 |
+
01-18 00:48:19 INFO [logging.py:61]: Loss 'loss' on epoch 1: 41.1013298034668
|
| 87 |
+
01-18 00:48:19 INFO [logging.py:61]: Loss 'ar_loss' on epoch 1: 3.8041999340057373
|
| 88 |
+
01-18 00:48:19 INFO [logging.py:61]: Loss 'nar_loss' on epoch 1: 37.2971305847168
|
| 89 |
+
01-18 00:48:19 INFO [logging.py:61]: Loss 'ar_accuracy_metric' on epoch 1: 0.5283006429672241
|
| 90 |
+
01-18 00:48:19 INFO [logging.py:61]: Loss 'nar_acc_metric' on epoch 1: 0.14399056136608124
|
| 91 |
+
01-18 00:48:19 INFO [logging.py:61]: ========= Epoch 2 out of 1000 =========
|
| 92 |
+
01-18 00:48:19 INFO [logging.py:61]: Begin training...
|
| 93 |
+
01-18 01:03:02 INFO [logging.py:61]: Loss 'loss' on epoch 2: 71.26728057861328
|
| 94 |
+
01-18 01:03:02 INFO [logging.py:61]: Loss 'ar_loss' on epoch 2: 3.0147104263305664
|
| 95 |
+
01-18 01:03:02 INFO [logging.py:61]: Loss 'nar_loss' on epoch 2: 68.25257110595703
|
| 96 |
+
01-18 01:03:02 INFO [logging.py:61]: Loss 'ar_accuracy_metric' on epoch 2: 0.6516079306602478
|
| 97 |
+
01-18 01:03:02 INFO [logging.py:61]: Loss 'nar_acc_metric' on epoch 2: 0.14630988240242004
|
| 98 |
+
01-18 01:03:02 INFO [logging.py:61]: ========= Epoch 3 out of 1000 =========
|
| 99 |
+
01-18 01:03:02 INFO [logging.py:61]: Begin training...
|
| 100 |
+
01-18 01:17:44 INFO [logging.py:61]: Loss 'loss' on epoch 3: 80.09443664550781
|
| 101 |
+
01-18 01:17:44 INFO [logging.py:61]: Loss 'ar_loss' on epoch 3: 2.8742563724517822
|
| 102 |
+
01-18 01:17:44 INFO [logging.py:61]: Loss 'nar_loss' on epoch 3: 77.22017669677734
|
| 103 |
+
01-18 01:17:44 INFO [logging.py:61]: Loss 'ar_accuracy_metric' on epoch 3: 0.6769989132881165
|
| 104 |
+
01-18 01:17:44 INFO [logging.py:61]: Loss 'nar_acc_metric' on epoch 3: 0.1532195657491684
|
| 105 |
+
01-18 01:17:44 INFO [logging.py:61]: ========= Epoch 4 out of 1000 =========
|
| 106 |
+
01-18 01:17:44 INFO [logging.py:61]: Begin training...
|
| 107 |
+
01-18 01:32:28 INFO [logging.py:61]: Loss 'loss' on epoch 4: 64.45104217529297
|
| 108 |
+
01-18 01:32:28 INFO [logging.py:61]: Loss 'ar_loss' on epoch 4: 2.756040096282959
|
| 109 |
+
01-18 01:32:28 INFO [logging.py:61]: Loss 'nar_loss' on epoch 4: 61.69499969482422
|
| 110 |
+
01-18 01:32:28 INFO [logging.py:61]: Loss 'ar_accuracy_metric' on epoch 4: 0.6978757381439209
|
| 111 |
+
01-18 01:32:28 INFO [logging.py:61]: Loss 'nar_acc_metric' on epoch 4: 0.1739574521780014
|
| 112 |
+
01-18 01:32:28 INFO [logging.py:61]: ========= Epoch 5 out of 1000 =========
|
| 113 |
+
01-18 01:32:28 INFO [logging.py:61]: Begin training...
|
| 114 |
+
01-18 01:47:11 INFO [logging.py:61]: Loss 'loss' on epoch 5: 50.534027099609375
|
| 115 |
+
01-18 01:47:11 INFO [logging.py:61]: Loss 'ar_loss' on epoch 5: 2.4202704429626465
|
| 116 |
+
01-18 01:47:11 INFO [logging.py:61]: Loss 'nar_loss' on epoch 5: 48.1137580871582
|
| 117 |
+
01-18 01:47:11 INFO [logging.py:61]: Loss 'ar_accuracy_metric' on epoch 5: 0.7520415186882019
|
| 118 |
+
01-18 01:47:11 INFO [logging.py:61]: Loss 'nar_acc_metric' on epoch 5: 0.2114027887582779
|
| 119 |
+
01-18 01:47:11 INFO [logging.py:61]: Saving current state to /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0005
|
| 120 |
+
01-18 01:47:12 INFO [logging.py:61]: Model weights saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0005/pytorch_model.bin
|
| 121 |
+
01-18 01:47:13 INFO [logging.py:61]: Optimizer state saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0005/optimizer.bin
|
| 122 |
+
01-18 01:47:13 INFO [logging.py:61]: Scheduler state saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0005/scheduler.bin
|
| 123 |
+
01-18 01:47:13 INFO [logging.py:61]: Sampler state for dataloader 0 saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0005/sampler.bin
|
| 124 |
+
01-18 01:47:13 INFO [logging.py:61]: Sampler state for dataloader 1 saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0005/sampler_1.bin
|
| 125 |
+
01-18 01:47:13 INFO [logging.py:61]: Random states saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0005/random_states_0.pkl
|
| 126 |
+
01-18 01:47:13 INFO [logging.py:61]: Saving the state of TrainerState to /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0005/custom_checkpoint_0.pkl
|
| 127 |
+
01-18 01:47:14 INFO [logging.py:61]: ========= Epoch 6 out of 1000 =========
|
| 128 |
+
01-18 01:47:14 INFO [logging.py:61]: Begin training...
|
| 129 |
+
01-18 02:02:24 INFO [logging.py:61]: Loss 'loss' on epoch 6: 37.49298858642578
|
| 130 |
+
01-18 02:02:24 INFO [logging.py:61]: Loss 'ar_loss' on epoch 6: 1.8649616241455078
|
| 131 |
+
01-18 02:02:24 INFO [logging.py:61]: Loss 'nar_loss' on epoch 6: 35.628028869628906
|
| 132 |
+
01-18 02:02:24 INFO [logging.py:61]: Loss 'ar_accuracy_metric' on epoch 6: 0.8278111219406128
|
| 133 |
+
01-18 02:02:24 INFO [logging.py:61]: Loss 'nar_acc_metric' on epoch 6: 0.22959543764591217
|
| 134 |
+
01-18 02:02:24 INFO [logging.py:61]: ========= Epoch 7 out of 1000 =========
|
| 135 |
+
01-18 02:02:24 INFO [logging.py:61]: Begin training...
|
| 136 |
+
01-18 02:17:03 INFO [logging.py:61]: Loss 'loss' on epoch 7: 28.96417236328125
|
| 137 |
+
01-18 02:17:03 INFO [logging.py:61]: Loss 'ar_loss' on epoch 7: 1.677205204963684
|
| 138 |
+
01-18 02:17:03 INFO [logging.py:61]: Loss 'nar_loss' on epoch 7: 27.286962509155273
|
| 139 |
+
01-18 02:17:03 INFO [logging.py:61]: Loss 'ar_accuracy_metric' on epoch 7: 0.8506313562393188
|
| 140 |
+
01-18 02:17:03 INFO [logging.py:61]: Loss 'nar_acc_metric' on epoch 7: 0.2341448813676834
|
| 141 |
+
01-18 02:17:03 INFO [logging.py:61]: ========= Epoch 8 out of 1000 =========
|
| 142 |
+
01-18 02:17:03 INFO [logging.py:61]: Begin training...
|
| 143 |
+
01-18 02:31:48 INFO [logging.py:61]: Loss 'loss' on epoch 8: 28.844099044799805
|
| 144 |
+
01-18 02:31:48 INFO [logging.py:61]: Loss 'ar_loss' on epoch 8: 1.6051831245422363
|
| 145 |
+
01-18 02:31:48 INFO [logging.py:61]: Loss 'nar_loss' on epoch 8: 27.238914489746094
|
| 146 |
+
01-18 02:31:48 INFO [logging.py:61]: Loss 'ar_accuracy_metric' on epoch 8: 0.85962975025177
|
| 147 |
+
01-18 02:31:48 INFO [logging.py:61]: Loss 'nar_acc_metric' on epoch 8: 0.2343209981918335
|
| 148 |
+
01-18 02:31:48 INFO [logging.py:61]: ========= Epoch 9 out of 1000 =========
|
| 149 |
+
01-18 02:31:48 INFO [logging.py:61]: Begin training...
|
| 150 |
+
01-18 02:46:32 INFO [logging.py:61]: Loss 'loss' on epoch 9: 18.423540115356445
|
| 151 |
+
01-18 02:46:32 INFO [logging.py:61]: Loss 'ar_loss' on epoch 9: 1.5607959032058716
|
| 152 |
+
01-18 02:46:32 INFO [logging.py:61]: Loss 'nar_loss' on epoch 9: 16.862743377685547
|
| 153 |
+
01-18 02:46:32 INFO [logging.py:61]: Loss 'ar_accuracy_metric' on epoch 9: 0.8656294345855713
|
| 154 |
+
01-18 02:46:32 INFO [logging.py:61]: Loss 'nar_acc_metric' on epoch 9: 0.268953800201416
|
| 155 |
+
01-18 02:46:32 INFO [logging.py:61]: ========= Epoch 10 out of 1000 =========
|
| 156 |
+
01-18 02:46:32 INFO [logging.py:61]: Begin training...
|
| 157 |
+
01-18 03:01:15 INFO [logging.py:61]: Loss 'loss' on epoch 10: 17.509979248046875
|
| 158 |
+
01-18 03:01:15 INFO [logging.py:61]: Loss 'ar_loss' on epoch 10: 1.516535997390747
|
| 159 |
+
01-18 03:01:15 INFO [logging.py:61]: Loss 'nar_loss' on epoch 10: 15.993441581726074
|
| 160 |
+
01-18 03:01:15 INFO [logging.py:61]: Loss 'ar_accuracy_metric' on epoch 10: 0.8709884881973267
|
| 161 |
+
01-18 03:01:15 INFO [logging.py:61]: Loss 'nar_acc_metric' on epoch 10: 0.27383264899253845
|
| 162 |
+
01-18 03:01:15 INFO [logging.py:61]: Saving current state to /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0010
|
| 163 |
+
01-18 03:01:16 INFO [logging.py:61]: Model weights saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0010/pytorch_model.bin
|
| 164 |
+
01-18 03:01:18 INFO [logging.py:61]: Optimizer state saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0010/optimizer.bin
|
| 165 |
+
01-18 03:01:18 INFO [logging.py:61]: Scheduler state saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0010/scheduler.bin
|
| 166 |
+
01-18 03:01:18 INFO [logging.py:61]: Sampler state for dataloader 0 saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0010/sampler.bin
|
| 167 |
+
01-18 03:01:18 INFO [logging.py:61]: Sampler state for dataloader 1 saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0010/sampler_1.bin
|
| 168 |
+
01-18 03:01:18 INFO [logging.py:61]: Random states saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0010/random_states_0.pkl
|
| 169 |
+
01-18 03:01:18 INFO [logging.py:61]: Saving the state of TrainerState to /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0010/custom_checkpoint_0.pkl
|
| 170 |
+
01-18 03:01:18 INFO [logging.py:61]: ========= Epoch 11 out of 1000 =========
|
| 171 |
+
01-18 03:01:18 INFO [logging.py:61]: Begin training...
|
| 172 |
+
01-18 03:16:19 INFO [logging.py:61]: Loss 'loss' on epoch 11: 20.488168716430664
|
| 173 |
+
01-18 03:16:19 INFO [logging.py:61]: Loss 'ar_loss' on epoch 11: 1.492824673652649
|
| 174 |
+
01-18 03:16:19 INFO [logging.py:61]: Loss 'nar_loss' on epoch 11: 18.995346069335938
|
| 175 |
+
01-18 03:16:19 INFO [logging.py:61]: Loss 'ar_accuracy_metric' on epoch 11: 0.8740840554237366
|
| 176 |
+
01-18 03:16:19 INFO [logging.py:61]: Loss 'nar_acc_metric' on epoch 11: 0.2761929929256439
|
| 177 |
+
01-18 03:16:19 INFO [logging.py:61]: ========= Epoch 12 out of 1000 =========
|
| 178 |
+
01-18 03:16:19 INFO [logging.py:61]: Begin training...
|
| 179 |
+
01-18 03:31:06 INFO [logging.py:61]: Loss 'loss' on epoch 12: 12.799497604370117
|
| 180 |
+
01-18 03:31:06 INFO [logging.py:61]: Loss 'ar_loss' on epoch 12: 1.4752053022384644
|
| 181 |
+
01-18 03:31:06 INFO [logging.py:61]: Loss 'nar_loss' on epoch 12: 11.324292182922363
|
| 182 |
+
01-18 03:31:06 INFO [logging.py:61]: Loss 'ar_accuracy_metric' on epoch 12: 0.8769260048866272
|
| 183 |
+
01-18 03:31:06 INFO [logging.py:61]: Loss 'nar_acc_metric' on epoch 12: 0.29501354694366455
|
| 184 |
+
01-18 03:31:06 INFO [logging.py:61]: ========= Epoch 13 out of 1000 =========
|
| 185 |
+
01-18 03:31:06 INFO [logging.py:61]: Begin training...
|
| 186 |
+
01-18 03:45:51 INFO [logging.py:61]: Loss 'loss' on epoch 13: 9.103281021118164
|
| 187 |
+
01-18 03:45:51 INFO [logging.py:61]: Loss 'ar_loss' on epoch 13: 1.456277847290039
|
| 188 |
+
01-18 03:45:51 INFO [logging.py:61]: Loss 'nar_loss' on epoch 13: 7.647003650665283
|
| 189 |
+
01-18 03:45:51 INFO [logging.py:61]: Loss 'ar_accuracy_metric' on epoch 13: 0.8797268867492676
|
| 190 |
+
01-18 03:45:51 INFO [logging.py:61]: Loss 'nar_acc_metric' on epoch 13: 0.3158681094646454
|
| 191 |
+
01-18 03:45:51 INFO [logging.py:61]: ========= Epoch 14 out of 1000 =========
|
| 192 |
+
01-18 03:45:51 INFO [logging.py:61]: Begin training...
|
| 193 |
+
01-18 04:00:36 INFO [logging.py:61]: Loss 'loss' on epoch 14: 13.451669692993164
|
| 194 |
+
01-18 04:00:36 INFO [logging.py:61]: Loss 'ar_loss' on epoch 14: 1.4416464567184448
|
| 195 |
+
01-18 04:00:36 INFO [logging.py:61]: Loss 'nar_loss' on epoch 14: 12.01002311706543
|
| 196 |
+
01-18 04:00:36 INFO [logging.py:61]: Loss 'ar_accuracy_metric' on epoch 14: 0.8816895484924316
|
| 197 |
+
01-18 04:00:36 INFO [logging.py:61]: Loss 'nar_acc_metric' on epoch 14: 0.2968295216560364
|
| 198 |
+
01-18 04:00:36 INFO [logging.py:61]: ========= Epoch 15 out of 1000 =========
|
| 199 |
+
01-18 04:00:36 INFO [logging.py:61]: Begin training...
|
| 200 |
+
01-18 04:15:15 INFO [logging.py:61]: Loss 'loss' on epoch 15: 10.637478828430176
|
| 201 |
+
01-18 04:15:15 INFO [logging.py:61]: Loss 'ar_loss' on epoch 15: 1.4250620603561401
|
| 202 |
+
01-18 04:15:15 INFO [logging.py:61]: Loss 'nar_loss' on epoch 15: 9.21241569519043
|
| 203 |
+
01-18 04:15:15 INFO [logging.py:61]: Loss 'ar_accuracy_metric' on epoch 15: 0.8841312527656555
|
| 204 |
+
01-18 04:15:15 INFO [logging.py:61]: Loss 'nar_acc_metric' on epoch 15: 0.3035244345664978
|
| 205 |
+
01-18 04:15:15 INFO [logging.py:61]: Saving current state to /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0015
|
| 206 |
+
01-18 04:15:16 INFO [logging.py:61]: Model weights saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0015/pytorch_model.bin
|
| 207 |
+
01-18 04:15:17 INFO [logging.py:61]: Optimizer state saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0015/optimizer.bin
|
| 208 |
+
01-18 04:15:17 INFO [logging.py:61]: Scheduler state saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0015/scheduler.bin
|
| 209 |
+
01-18 04:15:17 INFO [logging.py:61]: Sampler state for dataloader 0 saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0015/sampler.bin
|
| 210 |
+
01-18 04:15:17 INFO [logging.py:61]: Sampler state for dataloader 1 saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0015/sampler_1.bin
|
| 211 |
+
01-18 04:15:17 INFO [logging.py:61]: Random states saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0015/random_states_0.pkl
|
| 212 |
+
01-18 04:15:17 INFO [logging.py:61]: Saving the state of TrainerState to /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0015/custom_checkpoint_0.pkl
|
| 213 |
+
01-18 04:15:17 INFO [logging.py:61]: ========= Epoch 16 out of 1000 =========
|
| 214 |
+
01-18 04:15:17 INFO [logging.py:61]: Begin training...
|
| 215 |
+
01-18 04:30:27 INFO [logging.py:61]: Loss 'loss' on epoch 16: 7.80450439453125
|
| 216 |
+
01-18 04:30:27 INFO [logging.py:61]: Loss 'ar_loss' on epoch 16: 1.4189400672912598
|
| 217 |
+
01-18 04:30:27 INFO [logging.py:61]: Loss 'nar_loss' on epoch 16: 6.385564804077148
|
| 218 |
+
01-18 04:30:27 INFO [logging.py:61]: Loss 'ar_accuracy_metric' on epoch 16: 0.8853126764297485
|
| 219 |
+
01-18 04:30:27 INFO [logging.py:61]: Loss 'nar_acc_metric' on epoch 16: 0.3228910267353058
|
| 220 |
+
01-18 04:30:27 INFO [logging.py:61]: ========= Epoch 17 out of 1000 =========
|
| 221 |
+
01-18 04:30:27 INFO [logging.py:61]: Begin training...
|
| 222 |
+
01-18 04:45:12 INFO [logging.py:61]: Loss 'loss' on epoch 17: 7.98758602142334
|
| 223 |
+
01-18 04:45:12 INFO [logging.py:61]: Loss 'ar_loss' on epoch 17: 1.402923583984375
|
| 224 |
+
01-18 04:45:12 INFO [logging.py:61]: Loss 'nar_loss' on epoch 17: 6.584662914276123
|
| 225 |
+
01-18 04:45:12 INFO [logging.py:61]: Loss 'ar_accuracy_metric' on epoch 17: 0.8873386383056641
|
| 226 |
+
01-18 04:45:12 INFO [logging.py:61]: Loss 'nar_acc_metric' on epoch 17: 0.3134208917617798
|
| 227 |
+
01-18 04:45:12 INFO [logging.py:61]: ========= Epoch 18 out of 1000 =========
|
| 228 |
+
01-18 04:45:12 INFO [logging.py:61]: Begin training...
|
| 229 |
+
01-18 04:59:55 INFO [logging.py:61]: Loss 'loss' on epoch 18: 8.954412460327148
|
| 230 |
+
01-18 04:59:55 INFO [logging.py:61]: Loss 'ar_loss' on epoch 18: 1.3915390968322754
|
| 231 |
+
01-18 04:59:55 INFO [logging.py:61]: Loss 'nar_loss' on epoch 18: 7.562872886657715
|
| 232 |
+
01-18 04:59:55 INFO [logging.py:61]: Loss 'ar_accuracy_metric' on epoch 18: 0.8887759447097778
|
| 233 |
+
01-18 04:59:55 INFO [logging.py:61]: Loss 'nar_acc_metric' on epoch 18: 0.3137945532798767
|
| 234 |
+
01-18 04:59:55 INFO [logging.py:61]: ========= Epoch 19 out of 1000 =========
|
| 235 |
+
01-18 04:59:55 INFO [logging.py:61]: Begin training...
|
| 236 |
+
01-18 05:14:30 INFO [logging.py:61]: Loss 'loss' on epoch 19: 6.779003620147705
|
| 237 |
+
01-18 05:14:30 INFO [logging.py:61]: Loss 'ar_loss' on epoch 19: 1.3836216926574707
|
| 238 |
+
01-18 05:14:30 INFO [logging.py:61]: Loss 'nar_loss' on epoch 19: 5.395382404327393
|
| 239 |
+
01-18 05:14:30 INFO [logging.py:61]: Loss 'ar_accuracy_metric' on epoch 19: 0.8901618719100952
|
| 240 |
+
01-18 05:14:30 INFO [logging.py:61]: Loss 'nar_acc_metric' on epoch 19: 0.3347846567630768
|
| 241 |
+
01-18 05:14:30 INFO [logging.py:61]: ========= Epoch 20 out of 1000 =========
|
| 242 |
+
01-18 05:14:30 INFO [logging.py:61]: Begin training...
|
| 243 |
+
01-18 05:29:19 INFO [logging.py:61]: Loss 'loss' on epoch 20: 8.779007911682129
|
| 244 |
+
01-18 05:29:19 INFO [logging.py:61]: Loss 'ar_loss' on epoch 20: 1.3702212572097778
|
| 245 |
+
01-18 05:29:19 INFO [logging.py:61]: Loss 'nar_loss' on epoch 20: 7.408785343170166
|
| 246 |
+
01-18 05:29:19 INFO [logging.py:61]: Loss 'ar_accuracy_metric' on epoch 20: 0.8919013738632202
|
| 247 |
+
01-18 05:29:19 INFO [logging.py:61]: Loss 'nar_acc_metric' on epoch 20: 0.3212190866470337
|
| 248 |
+
01-18 05:29:19 INFO [logging.py:61]: Saving current state to /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0020
|
| 249 |
+
01-18 05:29:19 INFO [logging.py:61]: Model weights saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0020/pytorch_model.bin
|
| 250 |
+
01-18 05:29:21 INFO [logging.py:61]: Optimizer state saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0020/optimizer.bin
|
| 251 |
+
01-18 05:29:21 INFO [logging.py:61]: Scheduler state saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0020/scheduler.bin
|
| 252 |
+
01-18 05:29:21 INFO [logging.py:61]: Sampler state for dataloader 0 saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0020/sampler.bin
|
| 253 |
+
01-18 05:29:21 INFO [logging.py:61]: Sampler state for dataloader 1 saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0020/sampler_1.bin
|
| 254 |
+
01-18 05:29:21 INFO [logging.py:61]: Random states saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0020/random_states_0.pkl
|
| 255 |
+
01-18 05:29:21 INFO [logging.py:61]: Saving the state of TrainerState to /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0020/custom_checkpoint_0.pkl
|
| 256 |
+
01-18 05:29:21 INFO [logging.py:61]: ========= Epoch 21 out of 1000 =========
|
| 257 |
+
01-18 05:29:21 INFO [logging.py:61]: Begin training...
|
| 258 |
+
01-18 05:44:34 INFO [logging.py:61]: Loss 'loss' on epoch 21: 6.515992164611816
|
| 259 |
+
01-18 05:44:34 INFO [logging.py:61]: Loss 'ar_loss' on epoch 21: 1.363883376121521
|
| 260 |
+
01-18 05:44:34 INFO [logging.py:61]: Loss 'nar_loss' on epoch 21: 5.152108669281006
|
| 261 |
+
01-18 05:44:34 INFO [logging.py:61]: Loss 'ar_accuracy_metric' on epoch 21: 0.8928155899047852
|
| 262 |
+
01-18 05:44:34 INFO [logging.py:61]: Loss 'nar_acc_metric' on epoch 21: 0.3437054753303528
|
| 263 |
+
01-18 05:44:34 INFO [logging.py:61]: ========= Epoch 22 out of 1000 =========
|
| 264 |
+
01-18 05:44:34 INFO [logging.py:61]: Begin training...
|
| 265 |
+
01-18 05:59:19 INFO [logging.py:61]: Loss 'loss' on epoch 22: 6.411158084869385
|
| 266 |
+
01-18 05:59:19 INFO [logging.py:61]: Loss 'ar_loss' on epoch 22: 1.3568339347839355
|
| 267 |
+
01-18 05:59:19 INFO [logging.py:61]: Loss 'nar_loss' on epoch 22: 5.054324150085449
|
| 268 |
+
01-18 05:59:19 INFO [logging.py:61]: Loss 'ar_accuracy_metric' on epoch 22: 0.8938457369804382
|
| 269 |
+
01-18 05:59:19 INFO [logging.py:61]: Loss 'nar_acc_metric' on epoch 22: 0.3443514406681061
|
| 270 |
+
01-18 05:59:19 INFO [logging.py:61]: ========= Epoch 23 out of 1000 =========
|
| 271 |
+
01-18 05:59:19 INFO [logging.py:61]: Begin training...
|
| 272 |
+
01-18 06:14:01 INFO [logging.py:61]: Loss 'loss' on epoch 23: 6.536004066467285
|
| 273 |
+
01-18 06:14:01 INFO [logging.py:61]: Loss 'ar_loss' on epoch 23: 1.3483623266220093
|
| 274 |
+
01-18 06:14:01 INFO [logging.py:61]: Loss 'nar_loss' on epoch 23: 5.187641620635986
|
| 275 |
+
01-18 06:14:01 INFO [logging.py:61]: Loss 'ar_accuracy_metric' on epoch 23: 0.8951892256736755
|
| 276 |
+
01-18 06:14:01 INFO [logging.py:61]: Loss 'nar_acc_metric' on epoch 23: 0.3462466299533844
|
| 277 |
+
01-18 06:14:01 INFO [logging.py:61]: ========= Epoch 24 out of 1000 =========
|
| 278 |
+
01-18 06:14:01 INFO [logging.py:61]: Begin training...
|
| 279 |
+
01-18 06:28:46 INFO [logging.py:61]: Loss 'loss' on epoch 24: 6.32097053527832
|
| 280 |
+
01-18 06:28:46 INFO [logging.py:61]: Loss 'ar_loss' on epoch 24: 1.3559930324554443
|
| 281 |
+
01-18 06:28:46 INFO [logging.py:61]: Loss 'nar_loss' on epoch 24: 4.9649786949157715
|
| 282 |
+
01-18 06:28:46 INFO [logging.py:61]: Loss 'ar_accuracy_metric' on epoch 24: 0.8945856094360352
|
| 283 |
+
01-18 06:28:46 INFO [logging.py:61]: Loss 'nar_acc_metric' on epoch 24: 0.3438716530799866
|
| 284 |
+
01-18 06:28:46 INFO [logging.py:61]: ========= Epoch 25 out of 1000 =========
|
| 285 |
+
01-18 06:28:46 INFO [logging.py:61]: Begin training...
|
| 286 |
+
01-18 06:43:30 INFO [logging.py:61]: Loss 'loss' on epoch 25: 6.4575934410095215
|
| 287 |
+
01-18 06:43:30 INFO [logging.py:61]: Loss 'ar_loss' on epoch 25: 1.3440008163452148
|
| 288 |
+
01-18 06:43:30 INFO [logging.py:61]: Loss 'nar_loss' on epoch 25: 5.113592147827148
|
| 289 |
+
01-18 06:43:30 INFO [logging.py:61]: Loss 'ar_accuracy_metric' on epoch 25: 0.8960744142532349
|
| 290 |
+
01-18 06:43:30 INFO [logging.py:61]: Loss 'nar_acc_metric' on epoch 25: 0.3424939811229706
|
| 291 |
+
01-18 06:43:30 INFO [logging.py:61]: Saving current state to /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0025
|
| 292 |
+
01-18 06:43:31 INFO [logging.py:61]: Model weights saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0025/pytorch_model.bin
|
| 293 |
+
01-18 06:43:32 INFO [logging.py:61]: Optimizer state saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0025/optimizer.bin
|
| 294 |
+
01-18 06:43:32 INFO [logging.py:61]: Scheduler state saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0025/scheduler.bin
|
| 295 |
+
01-18 06:43:32 INFO [logging.py:61]: Sampler state for dataloader 0 saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0025/sampler.bin
|
| 296 |
+
01-18 06:43:32 INFO [logging.py:61]: Sampler state for dataloader 1 saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0025/sampler_1.bin
|
| 297 |
+
01-18 06:43:32 INFO [logging.py:61]: Random states saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0025/random_states_0.pkl
|
| 298 |
+
01-18 06:43:32 INFO [logging.py:61]: Saving the state of TrainerState to /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0025/custom_checkpoint_0.pkl
|
| 299 |
+
01-18 06:43:32 INFO [logging.py:61]: ========= Epoch 26 out of 1000 =========
|
| 300 |
+
01-18 06:43:32 INFO [logging.py:61]: Begin training...
|
| 301 |
+
01-18 06:59:08 INFO [logging.py:61]: Loss 'loss' on epoch 26: 5.9814558029174805
|
| 302 |
+
01-18 06:59:08 INFO [logging.py:61]: Loss 'ar_loss' on epoch 26: 1.3335331678390503
|
| 303 |
+
01-18 06:59:08 INFO [logging.py:61]: Loss 'nar_loss' on epoch 26: 4.647922992706299
|
| 304 |
+
01-18 06:59:08 INFO [logging.py:61]: Loss 'ar_accuracy_metric' on epoch 26: 0.8972698450088501
|
| 305 |
+
01-18 06:59:08 INFO [logging.py:61]: Loss 'nar_acc_metric' on epoch 26: 0.3597102165222168
|
| 306 |
+
01-18 06:59:08 INFO [logging.py:61]: ========= Epoch 27 out of 1000 =========
|
| 307 |
+
01-18 06:59:08 INFO [logging.py:61]: Begin training...
|
| 308 |
+
01-18 07:13:48 INFO [logging.py:61]: Loss 'loss' on epoch 27: 5.9211649894714355
|
| 309 |
+
01-18 07:13:48 INFO [logging.py:61]: Loss 'ar_loss' on epoch 27: 1.3315753936767578
|
| 310 |
+
01-18 07:13:48 INFO [logging.py:61]: Loss 'nar_loss' on epoch 27: 4.589588642120361
|
| 311 |
+
01-18 07:13:48 INFO [logging.py:61]: Loss 'ar_accuracy_metric' on epoch 27: 0.8978341817855835
|
| 312 |
+
01-18 07:13:48 INFO [logging.py:61]: Loss 'nar_acc_metric' on epoch 27: 0.3617432415485382
|
| 313 |
+
01-18 07:13:48 INFO [logging.py:61]: ========= Epoch 28 out of 1000 =========
|
| 314 |
+
01-18 07:13:48 INFO [logging.py:61]: Begin training...
|
| 315 |
+
01-18 07:28:33 INFO [logging.py:61]: Loss 'loss' on epoch 28: 5.8742170333862305
|
| 316 |
+
01-18 07:28:33 INFO [logging.py:61]: Loss 'ar_loss' on epoch 28: 1.3296784162521362
|
| 317 |
+
01-18 07:28:33 INFO [logging.py:61]: Loss 'nar_loss' on epoch 28: 4.544538497924805
|
| 318 |
+
01-18 07:28:33 INFO [logging.py:61]: Loss 'ar_accuracy_metric' on epoch 28: 0.8983495831489563
|
| 319 |
+
01-18 07:28:33 INFO [logging.py:61]: Loss 'nar_acc_metric' on epoch 28: 0.366417795419693
|
| 320 |
+
01-18 07:28:33 INFO [logging.py:61]: ========= Epoch 29 out of 1000 =========
|
| 321 |
+
01-18 07:28:33 INFO [logging.py:61]: Begin training...
|
| 322 |
+
01-18 07:43:17 INFO [logging.py:61]: Loss 'loss' on epoch 29: 5.860039710998535
|
| 323 |
+
01-18 07:43:17 INFO [logging.py:61]: Loss 'ar_loss' on epoch 29: 1.3166179656982422
|
| 324 |
+
01-18 07:43:17 INFO [logging.py:61]: Loss 'nar_loss' on epoch 29: 4.543421268463135
|
| 325 |
+
01-18 07:43:17 INFO [logging.py:61]: Loss 'ar_accuracy_metric' on epoch 29: 0.8997495174407959
|
| 326 |
+
01-18 07:43:17 INFO [logging.py:61]: Loss 'nar_acc_metric' on epoch 29: 0.3649654686450958
|
| 327 |
+
01-18 07:43:17 INFO [logging.py:61]: ========= Epoch 30 out of 1000 =========
|
| 328 |
+
01-18 07:43:17 INFO [logging.py:61]: Begin training...
|
| 329 |
+
01-18 07:58:02 INFO [logging.py:61]: Loss 'loss' on epoch 30: 5.796247959136963
|
| 330 |
+
01-18 07:58:02 INFO [logging.py:61]: Loss 'ar_loss' on epoch 30: 1.3036959171295166
|
| 331 |
+
01-18 07:58:02 INFO [logging.py:61]: Loss 'nar_loss' on epoch 30: 4.492552280426025
|
| 332 |
+
01-18 07:58:02 INFO [logging.py:61]: Loss 'ar_accuracy_metric' on epoch 30: 0.90115886926651
|
| 333 |
+
01-18 07:58:02 INFO [logging.py:61]: Loss 'nar_acc_metric' on epoch 30: 0.37101566791534424
|
| 334 |
+
01-18 07:58:02 INFO [logging.py:61]: Saving current state to /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0030
|
| 335 |
+
01-18 07:58:03 INFO [logging.py:61]: Model weights saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0030/pytorch_model.bin
|
| 336 |
+
01-18 07:58:04 INFO [logging.py:61]: Optimizer state saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0030/optimizer.bin
|
| 337 |
+
01-18 07:58:04 INFO [logging.py:61]: Scheduler state saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0030/scheduler.bin
|
| 338 |
+
01-18 07:58:04 INFO [logging.py:61]: Sampler state for dataloader 0 saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0030/sampler.bin
|
| 339 |
+
01-18 07:58:04 INFO [logging.py:61]: Sampler state for dataloader 1 saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0030/sampler_1.bin
|
| 340 |
+
01-18 07:58:04 INFO [logging.py:61]: Random states saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0030/random_states_0.pkl
|
| 341 |
+
01-18 07:58:04 INFO [logging.py:61]: Saving the state of TrainerState to /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0030/custom_checkpoint_0.pkl
|
| 342 |
+
01-18 07:58:04 INFO [logging.py:61]: ========= Epoch 31 out of 1000 =========
|
| 343 |
+
01-18 07:58:04 INFO [logging.py:61]: Begin training...
|
| 344 |
+
01-18 08:12:58 INFO [logging.py:61]: Loss 'loss' on epoch 31: 5.9744038581848145
|
| 345 |
+
01-18 08:12:58 INFO [logging.py:61]: Loss 'ar_loss' on epoch 31: 1.311854362487793
|
| 346 |
+
01-18 08:12:58 INFO [logging.py:61]: Loss 'nar_loss' on epoch 31: 4.662549018859863
|
| 347 |
+
01-18 08:12:58 INFO [logging.py:61]: Loss 'ar_accuracy_metric' on epoch 31: 0.9008342623710632
|
| 348 |
+
01-18 08:12:58 INFO [logging.py:61]: Loss 'nar_acc_metric' on epoch 31: 0.366784006357193
|
| 349 |
+
01-18 08:12:58 INFO [logging.py:61]: ========= Epoch 32 out of 1000 =========
|
| 350 |
+
01-18 08:12:58 INFO [logging.py:61]: Begin training...
|
| 351 |
+
01-18 08:27:46 INFO [logging.py:61]: Loss 'loss' on epoch 32: 5.735446453094482
|
| 352 |
+
01-18 08:27:46 INFO [logging.py:61]: Loss 'ar_loss' on epoch 32: 1.3046653270721436
|
| 353 |
+
01-18 08:27:46 INFO [logging.py:61]: Loss 'nar_loss' on epoch 32: 4.43078088760376
|
| 354 |
+
01-18 08:27:46 INFO [logging.py:61]: Loss 'ar_accuracy_metric' on epoch 32: 0.9014154076576233
|
| 355 |
+
01-18 08:27:46 INFO [logging.py:61]: Loss 'nar_acc_metric' on epoch 32: 0.37591466307640076
|
| 356 |
+
01-18 08:27:46 INFO [logging.py:61]: ========= Epoch 33 out of 1000 =========
|
| 357 |
+
01-18 08:27:46 INFO [logging.py:61]: Begin training...
|
| 358 |
+
01-18 08:42:30 INFO [logging.py:61]: Loss 'loss' on epoch 33: 5.872527122497559
|
| 359 |
+
01-18 08:42:30 INFO [logging.py:61]: Loss 'ar_loss' on epoch 33: 1.2990329265594482
|
| 360 |
+
01-18 08:42:30 INFO [logging.py:61]: Loss 'nar_loss' on epoch 33: 4.5734944343566895
|
| 361 |
+
01-18 08:42:30 INFO [logging.py:61]: Loss 'ar_accuracy_metric' on epoch 33: 0.9022935032844543
|
| 362 |
+
01-18 08:42:30 INFO [logging.py:61]: Loss 'nar_acc_metric' on epoch 33: 0.37081286311149597
|
| 363 |
+
01-18 08:42:30 INFO [logging.py:61]: ========= Epoch 34 out of 1000 =========
|
| 364 |
+
01-18 08:42:30 INFO [logging.py:61]: Begin training...
|
| 365 |
+
01-18 08:57:15 INFO [logging.py:61]: Loss 'loss' on epoch 34: 5.6954474449157715
|
| 366 |
+
01-18 08:57:15 INFO [logging.py:61]: Loss 'ar_loss' on epoch 34: 1.3018174171447754
|
| 367 |
+
01-18 08:57:15 INFO [logging.py:61]: Loss 'nar_loss' on epoch 34: 4.393630504608154
|
| 368 |
+
01-18 08:57:15 INFO [logging.py:61]: Loss 'ar_accuracy_metric' on epoch 34: 0.9023982882499695
|
| 369 |
+
01-18 08:57:15 INFO [logging.py:61]: Loss 'nar_acc_metric' on epoch 34: 0.38097846508026123
|
| 370 |
+
01-18 08:57:15 INFO [logging.py:61]: ========= Epoch 35 out of 1000 =========
|
| 371 |
+
01-18 08:57:15 INFO [logging.py:61]: Begin training...
|
| 372 |
+
01-18 09:11:52 INFO [logging.py:61]: Loss 'loss' on epoch 35: 5.682955741882324
|
| 373 |
+
01-18 09:11:52 INFO [logging.py:61]: Loss 'ar_loss' on epoch 35: 1.2978262901306152
|
| 374 |
+
01-18 09:11:52 INFO [logging.py:61]: Loss 'nar_loss' on epoch 35: 4.385129451751709
|
| 375 |
+
01-18 09:11:52 INFO [logging.py:61]: Loss 'ar_accuracy_metric' on epoch 35: 0.9027104377746582
|
| 376 |
+
01-18 09:11:52 INFO [logging.py:61]: Loss 'nar_acc_metric' on epoch 35: 0.3820667564868927
|
| 377 |
+
01-18 09:11:52 INFO [logging.py:61]: Saving current state to /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0035
|
| 378 |
+
01-18 09:11:53 INFO [logging.py:61]: Model weights saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0035/pytorch_model.bin
|
| 379 |
+
01-18 09:11:54 INFO [logging.py:61]: Optimizer state saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0035/optimizer.bin
|
| 380 |
+
01-18 09:11:54 INFO [logging.py:61]: Scheduler state saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0035/scheduler.bin
|
| 381 |
+
01-18 09:11:54 INFO [logging.py:61]: Sampler state for dataloader 0 saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0035/sampler.bin
|
| 382 |
+
01-18 09:11:54 INFO [logging.py:61]: Sampler state for dataloader 1 saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0035/sampler_1.bin
|
| 383 |
+
01-18 09:11:54 INFO [logging.py:61]: Random states saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0035/random_states_0.pkl
|
| 384 |
+
01-18 09:11:54 INFO [logging.py:61]: Saving the state of TrainerState to /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0035/custom_checkpoint_0.pkl
|
| 385 |
+
01-18 09:11:54 INFO [logging.py:61]: ========= Epoch 36 out of 1000 =========
|
| 386 |
+
01-18 09:11:54 INFO [logging.py:61]: Begin training...
|
| 387 |
+
01-18 09:27:23 INFO [logging.py:61]: Loss 'loss' on epoch 36: 5.689979553222656
|
| 388 |
+
01-18 09:27:23 INFO [logging.py:61]: Loss 'ar_loss' on epoch 36: 1.2983156442642212
|
| 389 |
+
01-18 09:27:23 INFO [logging.py:61]: Loss 'nar_loss' on epoch 36: 4.391663551330566
|
| 390 |
+
01-18 09:27:23 INFO [logging.py:61]: Loss 'ar_accuracy_metric' on epoch 36: 0.90284264087677
|
| 391 |
+
01-18 09:27:23 INFO [logging.py:61]: Loss 'nar_acc_metric' on epoch 36: 0.3806702196598053
|
| 392 |
+
01-18 09:27:23 INFO [logging.py:61]: ========= Epoch 37 out of 1000 =========
|
| 393 |
+
01-18 09:27:23 INFO [logging.py:61]: Begin training...
|
| 394 |
+
01-18 09:42:07 INFO [logging.py:61]: Loss 'loss' on epoch 37: 5.661297798156738
|
| 395 |
+
01-18 09:42:07 INFO [logging.py:61]: Loss 'ar_loss' on epoch 37: 1.2922143936157227
|
| 396 |
+
01-18 09:42:07 INFO [logging.py:61]: Loss 'nar_loss' on epoch 37: 4.369083404541016
|
| 397 |
+
01-18 09:42:07 INFO [logging.py:61]: Loss 'ar_accuracy_metric' on epoch 37: 0.9038951396942139
|
| 398 |
+
01-18 09:42:07 INFO [logging.py:61]: Loss 'nar_acc_metric' on epoch 37: 0.3834979236125946
|
| 399 |
+
01-18 09:42:07 INFO [logging.py:61]: ========= Epoch 38 out of 1000 =========
|
| 400 |
+
01-18 09:42:07 INFO [logging.py:61]: Begin training...
|
| 401 |
+
01-18 09:57:28 INFO [logging.py:61]: Loss 'loss' on epoch 38: 5.638166427612305
|
| 402 |
+
01-18 09:57:28 INFO [logging.py:61]: Loss 'ar_loss' on epoch 38: 1.2806264162063599
|
| 403 |
+
01-18 09:57:28 INFO [logging.py:61]: Loss 'nar_loss' on epoch 38: 4.357540130615234
|
| 404 |
+
01-18 09:57:28 INFO [logging.py:61]: Loss 'ar_accuracy_metric' on epoch 38: 0.9050056338310242
|
| 405 |
+
01-18 09:57:28 INFO [logging.py:61]: Loss 'nar_acc_metric' on epoch 38: 0.38519880175590515
|
| 406 |
+
01-18 09:57:28 INFO [logging.py:61]: ========= Epoch 39 out of 1000 =========
|
| 407 |
+
01-18 09:57:28 INFO [logging.py:61]: Begin training...
|
| 408 |
+
01-18 10:12:01 INFO [logging.py:61]: Loss 'loss' on epoch 39: 5.627877712249756
|
| 409 |
+
01-18 10:12:01 INFO [logging.py:61]: Loss 'ar_loss' on epoch 39: 1.2748196125030518
|
| 410 |
+
01-18 10:12:01 INFO [logging.py:61]: Loss 'nar_loss' on epoch 39: 4.353058338165283
|
| 411 |
+
01-18 10:12:01 INFO [logging.py:61]: Loss 'ar_accuracy_metric' on epoch 39: 0.9056810140609741
|
| 412 |
+
01-18 10:12:01 INFO [logging.py:61]: Loss 'nar_acc_metric' on epoch 39: 0.3871024549007416
|
| 413 |
+
01-18 10:12:01 INFO [logging.py:61]: ========= Epoch 40 out of 1000 =========
|
| 414 |
+
01-18 10:12:01 INFO [logging.py:61]: Begin training...
|
| 415 |
+
01-18 10:26:42 INFO [logging.py:61]: Loss 'loss' on epoch 40: 5.619797229766846
|
| 416 |
+
01-18 10:26:42 INFO [logging.py:61]: Loss 'ar_loss' on epoch 40: 1.2779885530471802
|
| 417 |
+
01-18 10:26:42 INFO [logging.py:61]: Loss 'nar_loss' on epoch 40: 4.341809272766113
|
| 418 |
+
01-18 10:26:42 INFO [logging.py:61]: Loss 'ar_accuracy_metric' on epoch 40: 0.9054821729660034
|
| 419 |
+
01-18 10:26:42 INFO [logging.py:61]: Loss 'nar_acc_metric' on epoch 40: 0.38915959000587463
|
| 420 |
+
01-18 10:26:42 INFO [logging.py:61]: Saving current state to /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0040
|
| 421 |
+
01-18 10:26:43 INFO [logging.py:61]: Model weights saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0040/pytorch_model.bin
|
| 422 |
+
01-18 10:26:45 INFO [logging.py:61]: Optimizer state saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0040/optimizer.bin
|
| 423 |
+
01-18 10:26:45 INFO [logging.py:61]: Scheduler state saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0040/scheduler.bin
|
| 424 |
+
01-18 10:26:45 INFO [logging.py:61]: Sampler state for dataloader 0 saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0040/sampler.bin
|
| 425 |
+
01-18 10:26:45 INFO [logging.py:61]: Sampler state for dataloader 1 saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0040/sampler_1.bin
|
| 426 |
+
01-18 10:26:45 INFO [logging.py:61]: Random states saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0040/random_states_0.pkl
|
| 427 |
+
01-18 10:26:45 INFO [logging.py:61]: Saving the state of TrainerState to /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0040/custom_checkpoint_0.pkl
|
| 428 |
+
01-18 10:26:45 INFO [logging.py:61]: ========= Epoch 41 out of 1000 =========
|
| 429 |
+
01-18 10:26:45 INFO [logging.py:61]: Begin training...
|
| 430 |
+
01-18 10:42:20 INFO [logging.py:61]: Loss 'loss' on epoch 41: 5.582706928253174
|
| 431 |
+
01-18 10:42:20 INFO [logging.py:61]: Loss 'ar_loss' on epoch 41: 1.2725111246109009
|
| 432 |
+
01-18 10:42:20 INFO [logging.py:61]: Loss 'nar_loss' on epoch 41: 4.3101959228515625
|
| 433 |
+
01-18 10:42:20 INFO [logging.py:61]: Loss 'ar_accuracy_metric' on epoch 41: 0.9060384035110474
|
| 434 |
+
01-18 10:42:20 INFO [logging.py:61]: Loss 'nar_acc_metric' on epoch 41: 0.39349165558815
|
| 435 |
+
01-18 10:42:20 INFO [logging.py:61]: ========= Epoch 42 out of 1000 =========
|
| 436 |
+
01-18 10:42:20 INFO [logging.py:61]: Begin training...
|
| 437 |
+
01-18 10:57:24 INFO [logging.py:61]: Loss 'loss' on epoch 42: 5.580837726593018
|
| 438 |
+
01-18 10:57:24 INFO [logging.py:61]: Loss 'ar_loss' on epoch 42: 1.2710316181182861
|
| 439 |
+
01-18 10:57:24 INFO [logging.py:61]: Loss 'nar_loss' on epoch 42: 4.309806823730469
|
| 440 |
+
01-18 10:57:24 INFO [logging.py:61]: Loss 'ar_accuracy_metric' on epoch 42: 0.9061719179153442
|
| 441 |
+
01-18 10:57:24 INFO [logging.py:61]: Loss 'nar_acc_metric' on epoch 42: 0.39457419514656067
|
| 442 |
+
01-18 10:57:24 INFO [logging.py:61]: ========= Epoch 43 out of 1000 =========
|
| 443 |
+
01-18 10:57:24 INFO [logging.py:61]: Begin training...
|
| 444 |
+
01-18 11:13:05 INFO [logging.py:61]: Loss 'loss' on epoch 43: 5.569582939147949
|
| 445 |
+
01-18 11:13:05 INFO [logging.py:61]: Loss 'ar_loss' on epoch 43: 1.2762430906295776
|
| 446 |
+
01-18 11:13:05 INFO [logging.py:61]: Loss 'nar_loss' on epoch 43: 4.293339252471924
|
| 447 |
+
01-18 11:13:05 INFO [logging.py:61]: Loss 'ar_accuracy_metric' on epoch 43: 0.9061148166656494
|
| 448 |
+
01-18 11:13:05 INFO [logging.py:61]: Loss 'nar_acc_metric' on epoch 43: 0.39825910329818726
|
| 449 |
+
01-18 11:13:05 INFO [logging.py:61]: ========= Epoch 44 out of 1000 =========
|
| 450 |
+
01-18 11:13:05 INFO [logging.py:61]: Begin training...
|
| 451 |
+
01-18 11:27:38 INFO [logging.py:61]: Loss 'loss' on epoch 44: 5.717195987701416
|
| 452 |
+
01-18 11:27:38 INFO [logging.py:61]: Loss 'ar_loss' on epoch 44: 1.2707628011703491
|
| 453 |
+
01-18 11:27:38 INFO [logging.py:61]: Loss 'nar_loss' on epoch 44: 4.446432590484619
|
| 454 |
+
01-18 11:27:38 INFO [logging.py:61]: Loss 'ar_accuracy_metric' on epoch 44: 0.9066540598869324
|
| 455 |
+
01-18 11:27:38 INFO [logging.py:61]: Loss 'nar_acc_metric' on epoch 44: 0.3859277665615082
|
| 456 |
+
01-18 11:27:38 INFO [logging.py:61]: ========= Epoch 45 out of 1000 =========
|
| 457 |
+
01-18 11:27:38 INFO [logging.py:61]: Begin training...
|
| 458 |
+
01-18 11:42:11 INFO [logging.py:61]: Loss 'loss' on epoch 45: 5.571887969970703
|
| 459 |
+
01-18 11:42:11 INFO [logging.py:61]: Loss 'ar_loss' on epoch 45: 1.2563157081604004
|
| 460 |
+
01-18 11:42:11 INFO [logging.py:61]: Loss 'nar_loss' on epoch 45: 4.315572261810303
|
| 461 |
+
01-18 11:42:11 INFO [logging.py:61]: Loss 'ar_accuracy_metric' on epoch 45: 0.9081159830093384
|
| 462 |
+
01-18 11:42:11 INFO [logging.py:61]: Loss 'nar_acc_metric' on epoch 45: 0.39590778946876526
|
| 463 |
+
01-18 11:42:11 INFO [logging.py:61]: Saving current state to /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0045
|
| 464 |
+
01-18 11:42:12 INFO [logging.py:61]: Model weights saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0045/pytorch_model.bin
|
| 465 |
+
01-18 11:42:13 INFO [logging.py:61]: Optimizer state saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0045/optimizer.bin
|
| 466 |
+
01-18 11:42:13 INFO [logging.py:61]: Scheduler state saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0045/scheduler.bin
|
| 467 |
+
01-18 11:42:13 INFO [logging.py:61]: Sampler state for dataloader 0 saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0045/sampler.bin
|
| 468 |
+
01-18 11:42:13 INFO [logging.py:61]: Sampler state for dataloader 1 saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0045/sampler_1.bin
|
| 469 |
+
01-18 11:42:13 INFO [logging.py:61]: Random states saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0045/random_states_0.pkl
|
| 470 |
+
01-18 11:42:13 INFO [logging.py:61]: Saving the state of TrainerState to /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0045/custom_checkpoint_0.pkl
|
| 471 |
+
01-18 11:42:13 INFO [logging.py:61]: ========= Epoch 46 out of 1000 =========
|
| 472 |
+
01-18 11:42:13 INFO [logging.py:61]: Begin training...
|
| 473 |
+
01-18 11:56:47 INFO [logging.py:61]: Loss 'loss' on epoch 46: 5.5488386154174805
|
| 474 |
+
01-18 11:56:47 INFO [logging.py:61]: Loss 'ar_loss' on epoch 46: 1.270042061805725
|
| 475 |
+
01-18 11:56:47 INFO [logging.py:61]: Loss 'nar_loss' on epoch 46: 4.278796195983887
|
| 476 |
+
01-18 11:56:47 INFO [logging.py:61]: Loss 'ar_accuracy_metric' on epoch 46: 0.9069695472717285
|
| 477 |
+
01-18 11:56:47 INFO [logging.py:61]: Loss 'nar_acc_metric' on epoch 46: 0.40045443177223206
|
| 478 |
+
01-18 11:56:47 INFO [logging.py:61]: ========= Epoch 47 out of 1000 =========
|
| 479 |
+
01-18 11:56:47 INFO [logging.py:61]: Begin training...
|
| 480 |
+
01-18 12:11:19 INFO [logging.py:61]: Loss 'loss' on epoch 47: 5.518341541290283
|
| 481 |
+
01-18 12:11:19 INFO [logging.py:61]: Loss 'ar_loss' on epoch 47: 1.2568832635879517
|
| 482 |
+
01-18 12:11:19 INFO [logging.py:61]: Loss 'nar_loss' on epoch 47: 4.261458396911621
|
| 483 |
+
01-18 12:11:19 INFO [logging.py:61]: Loss 'ar_accuracy_metric' on epoch 47: 0.9085185527801514
|
| 484 |
+
01-18 12:11:19 INFO [logging.py:61]: Loss 'nar_acc_metric' on epoch 47: 0.4032013416290283
|
| 485 |
+
01-18 12:11:19 INFO [logging.py:61]: ========= Epoch 48 out of 1000 =========
|
| 486 |
+
01-18 12:11:19 INFO [logging.py:61]: Begin training...
|
| 487 |
+
01-18 12:25:53 INFO [logging.py:61]: Loss 'loss' on epoch 48: 5.5060343742370605
|
| 488 |
+
01-18 12:25:54 INFO [logging.py:61]: Loss 'ar_loss' on epoch 48: 1.258938193321228
|
| 489 |
+
01-18 12:25:54 INFO [logging.py:61]: Loss 'nar_loss' on epoch 48: 4.247096538543701
|
| 490 |
+
01-18 12:25:54 INFO [logging.py:61]: Loss 'ar_accuracy_metric' on epoch 48: 0.9083271622657776
|
| 491 |
+
01-18 12:25:54 INFO [logging.py:61]: Loss 'nar_acc_metric' on epoch 48: 0.40573224425315857
|
| 492 |
+
01-18 12:25:54 INFO [logging.py:61]: ========= Epoch 49 out of 1000 =========
|
| 493 |
+
01-18 12:25:54 INFO [logging.py:61]: Begin training...
|
| 494 |
+
01-18 12:40:27 INFO [logging.py:61]: Loss 'loss' on epoch 49: 5.473796844482422
|
| 495 |
+
01-18 12:40:27 INFO [logging.py:61]: Loss 'ar_loss' on epoch 49: 1.2464219331741333
|
| 496 |
+
01-18 12:40:27 INFO [logging.py:61]: Loss 'nar_loss' on epoch 49: 4.22737455368042
|
| 497 |
+
01-18 12:40:27 INFO [logging.py:61]: Loss 'ar_accuracy_metric' on epoch 49: 0.9096682071685791
|
| 498 |
+
01-18 12:40:27 INFO [logging.py:61]: Loss 'nar_acc_metric' on epoch 49: 0.40877988934516907
|
| 499 |
+
01-18 12:40:27 INFO [logging.py:61]: ========= Epoch 50 out of 1000 =========
|
| 500 |
+
01-18 12:40:27 INFO [logging.py:61]: Begin training...
|
| 501 |
+
01-18 12:54:59 INFO [logging.py:61]: Loss 'loss' on epoch 50: 5.809966564178467
|
| 502 |
+
01-18 12:54:59 INFO [logging.py:61]: Loss 'ar_loss' on epoch 50: 1.2504818439483643
|
| 503 |
+
01-18 12:54:59 INFO [logging.py:61]: Loss 'nar_loss' on epoch 50: 4.559484958648682
|
| 504 |
+
01-18 12:54:59 INFO [logging.py:61]: Loss 'ar_accuracy_metric' on epoch 50: 0.9094463586807251
|
| 505 |
+
01-18 12:54:59 INFO [logging.py:61]: Loss 'nar_acc_metric' on epoch 50: 0.38511520624160767
|
| 506 |
+
01-18 12:54:59 INFO [logging.py:61]: Saving current state to /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0050
|
| 507 |
+
01-18 12:55:00 INFO [logging.py:61]: Model weights saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0050/pytorch_model.bin
|
| 508 |
+
01-18 12:55:01 INFO [logging.py:61]: Optimizer state saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0050/optimizer.bin
|
| 509 |
+
01-18 12:55:01 INFO [logging.py:61]: Scheduler state saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0050/scheduler.bin
|
| 510 |
+
01-18 12:55:01 INFO [logging.py:61]: Sampler state for dataloader 0 saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0050/sampler.bin
|
| 511 |
+
01-18 12:55:01 INFO [logging.py:61]: Sampler state for dataloader 1 saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0050/sampler_1.bin
|
| 512 |
+
01-18 12:55:01 INFO [logging.py:61]: Random states saved in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0050/random_states_0.pkl
|
| 513 |
+
01-18 12:55:01 INFO [logging.py:61]: Saving the state of TrainerState to /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-2_AR-NAR/checkpoints/epoch_0050/custom_checkpoint_0.pkl
|
| 514 |
+
01-18 12:55:01 INFO [logging.py:61]: ========= Epoch 51 out of 1000 =========
|
| 515 |
+
01-18 12:55:01 INFO [logging.py:61]: Begin training...
|
swin_default_LR1e-3_AR-NAR/swin_default_LR1e-3_AR-NAR_2024_01_18--13_13_05.log
ADDED
|
@@ -0,0 +1,94 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
01-18 13:13:05 INFO [logger.py:80]: Initialized logger with log file in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-3_AR-NAR.
|
| 2 |
+
01-18 13:13:05 INFO [logger.py:80]: Initialized logger with log file in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-3_AR-NAR.
|
| 3 |
+
01-18 13:13:05 INFO [logger.py:80]: Initialized logger with log file in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-3_AR-NAR.
|
| 4 |
+
01-18 13:13:05 INFO [logger.py:80]: Initialized logger with log file in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-3_AR-NAR.
|
| 5 |
+
01-18 13:13:58 INFO [logging.py:61]: Configuration file is saved to /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-3_AR-NAR/config__2024_01_18--13_13_46.toml.
|
| 6 |
+
01-18 13:13:58 INFO [logging.py:61]: Environment information:
|
| 7 |
+
- `Accelerate` version: 0.26.1
|
| 8 |
+
- Platform: Linux-5.14.0-362.13.1.el9_3.x86_64-x86_64-with-glibc2.34
|
| 9 |
+
- Python version: 3.10.13
|
| 10 |
+
- Numpy version: 1.26.3
|
| 11 |
+
- PyTorch version (GPU?): 2.1.2 (True)
|
| 12 |
+
- System RAM: 503.48 GB
|
| 13 |
+
- GPU Available: True
|
| 14 |
+
- GPU IDs: 4
|
| 15 |
+
- GPU type: NVIDIA A100-SXM4-80GB
|
| 16 |
+
01-18 13:13:58 INFO [logging.py:61]:
|
| 17 |
+
===============================================================================================
|
| 18 |
+
Layer (type:depth-idx) Param #
|
| 19 |
+
===============================================================================================
|
| 20 |
+
DistributedDataParallel --
|
| 21 |
+
├─Model: 1-1 --
|
| 22 |
+
│ └─EncodecModel: 2-1 --
|
| 23 |
+
│ │ └─EncodecEncoder: 3-1 (7,425,792)
|
| 24 |
+
│ │ └─EncodecDecoder: 3-2 (7,426,018)
|
| 25 |
+
│ │ └─EncodecResidualVectorQuantizer: 3-3 --
|
| 26 |
+
│ └─TokenEmbedding: 2-2 --
|
| 27 |
+
│ │ └─Dropout: 3-4 --
|
| 28 |
+
│ │ └─Embedding: 3-5 524,800
|
| 29 |
+
│ └─Identity: 2-3 --
|
| 30 |
+
│ └─SinePositionalEmbedding: 2-4 1
|
| 31 |
+
│ │ └─Dropout: 3-6 --
|
| 32 |
+
│ └─TransformerEncoder: 2-5 --
|
| 33 |
+
│ │ └─ModuleList: 3-7 37,828,608
|
| 34 |
+
│ │ └─LayerNorm: 3-8 1,024
|
| 35 |
+
│ └─Linear: 2-6 524,800
|
| 36 |
+
│ └─MulticlassAccuracy: 2-7 --
|
| 37 |
+
│ └─TokenEmbedding: 2-8 --
|
| 38 |
+
│ │ └─Dropout: 3-9 --
|
| 39 |
+
│ │ └─Embedding: 3-10 524,288
|
| 40 |
+
│ └─ModuleList: 2-9 --
|
| 41 |
+
│ │ └─TokenEmbedding: 3-11 524,800
|
| 42 |
+
│ │ └─TokenEmbedding: 3-12 524,288
|
| 43 |
+
│ │ └─TokenEmbedding: 3-13 524,288
|
| 44 |
+
│ │ └─TokenEmbedding: 3-14 524,288
|
| 45 |
+
│ │ └─TokenEmbedding: 3-15 524,288
|
| 46 |
+
│ │ └─TokenEmbedding: 3-16 524,288
|
| 47 |
+
│ │ └─TokenEmbedding: 3-17 524,288
|
| 48 |
+
│ │ └─TokenEmbedding: 3-18 524,288
|
| 49 |
+
│ └─Identity: 2-10 --
|
| 50 |
+
│ └─SinePositionalEmbedding: 2-11 1
|
| 51 |
+
│ │ └─Dropout: 3-19 --
|
| 52 |
+
│ └─TransformerEncoder: 2-12 --
|
| 53 |
+
│ │ └─ModuleList: 3-20 50,436,096
|
| 54 |
+
│ │ └─AdaptiveLayerNorm: 3-21 526,336
|
| 55 |
+
│ └─ModuleList: 2-13 --
|
| 56 |
+
│ │ └─Linear: 3-22 524,288
|
| 57 |
+
│ │ └─Linear: 3-23 524,288
|
| 58 |
+
│ │ └─Linear: 3-24 524,288
|
| 59 |
+
│ │ └─Linear: 3-25 524,288
|
| 60 |
+
│ │ └─Linear: 3-26 524,288
|
| 61 |
+
��� │ └─Linear: 3-27 524,288
|
| 62 |
+
│ │ └─Linear: 3-28 524,288
|
| 63 |
+
│ └─ModuleList: 2-14 --
|
| 64 |
+
│ │ └─TokenEmbedding: 3-29 512
|
| 65 |
+
│ │ └─TokenEmbedding: 3-30 512
|
| 66 |
+
│ │ └─TokenEmbedding: 3-31 512
|
| 67 |
+
│ │ └─TokenEmbedding: 3-32 512
|
| 68 |
+
│ │ └─TokenEmbedding: 3-33 512
|
| 69 |
+
│ │ └─TokenEmbedding: 3-34 512
|
| 70 |
+
│ │ └─TokenEmbedding: 3-35 512
|
| 71 |
+
│ └─MulticlassAccuracy: 2-15 --
|
| 72 |
+
===============================================================================================
|
| 73 |
+
Total params: 113,086,180
|
| 74 |
+
Trainable params: 98,234,369
|
| 75 |
+
Non-trainable params: 14,851,811
|
| 76 |
+
===============================================================================================
|
| 77 |
+
01-18 13:13:58 INFO [logging.py:61]: Training control variables:
|
| 78 |
+
01-18 13:13:58 INFO [logging.py:61]: `steps_per_epoch`: 500
|
| 79 |
+
01-18 13:13:58 INFO [logging.py:61]: Gradient accumulation steps: 1
|
| 80 |
+
01-18 13:13:58 INFO [logging.py:61]: `update_steps_per_epoch`: 500
|
| 81 |
+
01-18 13:13:58 INFO [logging.py:61]: `max_steps`: 500000
|
| 82 |
+
01-18 13:13:58 INFO [logging.py:61]: `max_epochs`: 1000
|
| 83 |
+
01-18 13:13:58 INFO [logging.py:61]: warmup_steps=1000. warmup_ratio will be ignored.
|
| 84 |
+
01-18 13:13:58 INFO [logging.py:61]: Loading states from /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-3_AR-NAR/checkpoints/epoch_0050
|
| 85 |
+
01-18 13:13:58 INFO [logging.py:61]: All model weights loaded successfully
|
| 86 |
+
01-18 13:13:59 INFO [logging.py:61]: All optimizer states loaded successfully
|
| 87 |
+
01-18 13:13:59 INFO [logging.py:61]: All scheduler states loaded successfully
|
| 88 |
+
01-18 13:13:59 INFO [logging.py:61]: All dataloader sampler states loaded successfully
|
| 89 |
+
01-18 13:13:59 INFO [logging.py:61]: All random states loaded successfully
|
| 90 |
+
01-18 13:14:00 INFO [logging.py:61]: Loading in 1 custom states
|
| 91 |
+
01-18 13:14:00 INFO [logging.py:61]: Loading the state of TrainerState from /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-3_AR-NAR/checkpoints/epoch_0050/custom_checkpoint_0.pkl
|
| 92 |
+
01-18 13:14:00 INFO [logging.py:61]: Checkpoint on epoch 50 is loaded.
|
| 93 |
+
01-18 13:14:00 INFO [logging.py:61]: ========= Epoch 51 out of 1000 =========
|
| 94 |
+
01-18 13:14:00 INFO [logging.py:61]: Begin training...
|
swin_default_LR1e-3_AR-NAR/swin_default_LR1e-3_AR-NAR_2024_01_18--13_14_39.log
ADDED
|
@@ -0,0 +1,121 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
01-18 13:14:39 INFO [logger.py:80]: Initialized logger with log file in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-3_AR-NAR.
|
| 2 |
+
01-18 13:14:39 INFO [logger.py:80]: Initialized logger with log file in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-3_AR-NAR.
|
| 3 |
+
01-18 13:14:39 INFO [logger.py:80]: Initialized logger with log file in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-3_AR-NAR.
|
| 4 |
+
01-18 13:14:45 INFO [logging.py:61]: Configuration file is saved to /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-3_AR-NAR/config__2024_01_18--13_14_43.toml.
|
| 5 |
+
01-18 13:14:45 INFO [logging.py:61]: Environment information:
|
| 6 |
+
- `Accelerate` version: 0.26.1
|
| 7 |
+
- Platform: Linux-5.14.0-362.13.1.el9_3.x86_64-x86_64-with-glibc2.34
|
| 8 |
+
- Python version: 3.10.13
|
| 9 |
+
- Numpy version: 1.26.3
|
| 10 |
+
- PyTorch version (GPU?): 2.1.2 (True)
|
| 11 |
+
- System RAM: 503.48 GB
|
| 12 |
+
- GPU Available: True
|
| 13 |
+
- GPU IDs: 4
|
| 14 |
+
- GPU type: NVIDIA A100-SXM4-80GB
|
| 15 |
+
01-18 13:14:45 INFO [logging.py:61]:
|
| 16 |
+
===============================================================================================
|
| 17 |
+
Layer (type:depth-idx) Param #
|
| 18 |
+
===============================================================================================
|
| 19 |
+
DistributedDataParallel --
|
| 20 |
+
├─Model: 1-1 --
|
| 21 |
+
│ └─EncodecModel: 2-1 --
|
| 22 |
+
│ │ └─EncodecEncoder: 3-1 (7,425,792)
|
| 23 |
+
│ │ └─EncodecDecoder: 3-2 (7,426,018)
|
| 24 |
+
│ │ └─EncodecResidualVectorQuantizer: 3-3 --
|
| 25 |
+
│ └─TokenEmbedding: 2-2 --
|
| 26 |
+
│ │ └─Dropout: 3-4 --
|
| 27 |
+
│ │ └─Embedding: 3-5 524,800
|
| 28 |
+
│ └─Identity: 2-3 --
|
| 29 |
+
│ └─SinePositionalEmbedding: 2-4 1
|
| 30 |
+
│ │ └─Dropout: 3-6 --
|
| 31 |
+
│ └─TransformerEncoder: 2-5 --
|
| 32 |
+
│ │ └─ModuleList: 3-7 37,828,608
|
| 33 |
+
│ │ └─LayerNorm: 3-8 1,024
|
| 34 |
+
│ └─Linear: 2-6 524,800
|
| 35 |
+
│ └─MulticlassAccuracy: 2-7 --
|
| 36 |
+
│ └─TokenEmbedding: 2-8 --
|
| 37 |
+
│ │ └─Dropout: 3-9 --
|
| 38 |
+
│ │ └─Embedding: 3-10 524,288
|
| 39 |
+
│ └─ModuleList: 2-9 --
|
| 40 |
+
│ │ └─TokenEmbedding: 3-11 524,800
|
| 41 |
+
│ │ └─TokenEmbedding: 3-12 524,288
|
| 42 |
+
│ │ └─TokenEmbedding: 3-13 524,288
|
| 43 |
+
│ │ └─TokenEmbedding: 3-14 524,288
|
| 44 |
+
│ │ └─TokenEmbedding: 3-15 524,288
|
| 45 |
+
│ │ └─TokenEmbedding: 3-16 524,288
|
| 46 |
+
│ │ └─TokenEmbedding: 3-17 524,288
|
| 47 |
+
│ │ └─TokenEmbedding: 3-18 524,288
|
| 48 |
+
│ └─Identity: 2-10 --
|
| 49 |
+
│ └─SinePositionalEmbedding: 2-11 1
|
| 50 |
+
│ │ └─Dropout: 3-19 --
|
| 51 |
+
│ └─TransformerEncoder: 2-12 --
|
| 52 |
+
│ │ └─ModuleList: 3-20 50,436,096
|
| 53 |
+
│ │ └─AdaptiveLayerNorm: 3-21 526,336
|
| 54 |
+
│ └─ModuleList: 2-13 --
|
| 55 |
+
│ │ └─Linear: 3-22 524,288
|
| 56 |
+
│ │ └─Linear: 3-23 524,288
|
| 57 |
+
│ │ └─Linear: 3-24 524,288
|
| 58 |
+
│ │ └─Linear: 3-25 524,288
|
| 59 |
+
│ │ └─Linear: 3-26 524,288
|
| 60 |
+
│ │ └─Linear: 3-27 524,288
|
| 61 |
+
│ │ └─Linear: 3-28 524,288
|
| 62 |
+
│ └─ModuleList: 2-14 --
|
| 63 |
+
│ │ └─TokenEmbedding: 3-29 512
|
| 64 |
+
│ │ └─TokenEmbedding: 3-30 512
|
| 65 |
+
│ │ └─TokenEmbedding: 3-31 512
|
| 66 |
+
│ │ └─TokenEmbedding: 3-32 512
|
| 67 |
+
│ │ └─TokenEmbedding: 3-33 512
|
| 68 |
+
│ │ └─TokenEmbedding: 3-34 512
|
| 69 |
+
│ │ └─TokenEmbedding: 3-35 512
|
| 70 |
+
│ └─MulticlassAccuracy: 2-15 --
|
| 71 |
+
===============================================================================================
|
| 72 |
+
Total params: 113,086,180
|
| 73 |
+
Trainable params: 98,234,369
|
| 74 |
+
Non-trainable params: 14,851,811
|
| 75 |
+
===============================================================================================
|
| 76 |
+
01-18 13:14:45 INFO [logging.py:61]: Training control variables:
|
| 77 |
+
01-18 13:14:45 INFO [logging.py:61]: `steps_per_epoch`: 500
|
| 78 |
+
01-18 13:14:45 INFO [logging.py:61]: Gradient accumulation steps: 1
|
| 79 |
+
01-18 13:14:45 INFO [logging.py:61]: `update_steps_per_epoch`: 500
|
| 80 |
+
01-18 13:14:45 INFO [logging.py:61]: `max_steps`: 500000
|
| 81 |
+
01-18 13:14:45 INFO [logging.py:61]: `max_epochs`: 1000
|
| 82 |
+
01-18 13:14:45 INFO [logging.py:61]: warmup_steps=1000. warmup_ratio will be ignored.
|
| 83 |
+
01-18 13:14:45 INFO [logging.py:61]: Loading states from /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-3_AR-NAR/checkpoints/epoch_0050
|
| 84 |
+
01-18 13:14:45 INFO [logging.py:61]: All model weights loaded successfully
|
| 85 |
+
01-18 13:14:46 INFO [logging.py:61]: All optimizer states loaded successfully
|
| 86 |
+
01-18 13:14:46 INFO [logging.py:61]: All scheduler states loaded successfully
|
| 87 |
+
01-18 13:14:46 INFO [logging.py:61]: All dataloader sampler states loaded successfully
|
| 88 |
+
01-18 13:14:46 INFO [logging.py:61]: All random states loaded successfully
|
| 89 |
+
01-18 13:14:46 INFO [logging.py:61]: Loading in 1 custom states
|
| 90 |
+
01-18 13:14:46 INFO [logging.py:61]: Loading the state of TrainerState from /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-3_AR-NAR/checkpoints/epoch_0050/custom_checkpoint_0.pkl
|
| 91 |
+
01-18 13:14:46 INFO [logging.py:61]: Checkpoint on epoch 50 is loaded.
|
| 92 |
+
01-18 13:14:46 INFO [logging.py:61]: ========= Epoch 51 out of 1000 =========
|
| 93 |
+
01-18 13:14:46 INFO [logging.py:61]: Begin training...
|
| 94 |
+
01-18 13:29:20 INFO [logging.py:61]: Loss 'loss' on epoch 51: 5.497116565704346
|
| 95 |
+
01-18 13:29:20 INFO [logging.py:61]: Loss 'ar_loss' on epoch 51: 1.253153920173645
|
| 96 |
+
01-18 13:29:20 INFO [logging.py:61]: Loss 'nar_loss' on epoch 51: 4.243962287902832
|
| 97 |
+
01-18 13:29:20 INFO [logging.py:61]: Loss 'ar_accuracy_metric' on epoch 51: 0.9091926217079163
|
| 98 |
+
01-18 13:29:20 INFO [logging.py:61]: Loss 'nar_acc_metric' on epoch 51: 0.40614935755729675
|
| 99 |
+
01-18 13:29:20 INFO [logging.py:61]: ========= Epoch 52 out of 1000 =========
|
| 100 |
+
01-18 13:29:20 INFO [logging.py:61]: Begin training...
|
| 101 |
+
01-18 13:43:53 INFO [logging.py:61]: Loss 'loss' on epoch 52: 5.452788829803467
|
| 102 |
+
01-18 13:43:53 INFO [logging.py:61]: Loss 'ar_loss' on epoch 52: 1.245349645614624
|
| 103 |
+
01-18 13:43:53 INFO [logging.py:61]: Loss 'nar_loss' on epoch 52: 4.207438945770264
|
| 104 |
+
01-18 13:43:53 INFO [logging.py:61]: Loss 'ar_accuracy_metric' on epoch 52: 0.9099061489105225
|
| 105 |
+
01-18 13:43:53 INFO [logging.py:61]: Loss 'nar_acc_metric' on epoch 52: 0.41280823945999146
|
| 106 |
+
01-18 13:43:53 INFO [logging.py:61]: ========= Epoch 53 out of 1000 =========
|
| 107 |
+
01-18 13:43:53 INFO [logging.py:61]: Begin training...
|
| 108 |
+
01-18 13:58:30 INFO [logging.py:61]: Loss 'loss' on epoch 53: 5.5037336349487305
|
| 109 |
+
01-18 13:58:30 INFO [logging.py:61]: Loss 'ar_loss' on epoch 53: 1.2453527450561523
|
| 110 |
+
01-18 13:58:30 INFO [logging.py:61]: Loss 'nar_loss' on epoch 53: 4.25838041305542
|
| 111 |
+
01-18 13:58:30 INFO [logging.py:61]: Loss 'ar_accuracy_metric' on epoch 53: 0.9102101922035217
|
| 112 |
+
01-18 13:58:30 INFO [logging.py:61]: Loss 'nar_acc_metric' on epoch 53: 0.40620309114456177
|
| 113 |
+
01-18 13:58:30 INFO [logging.py:61]: ========= Epoch 54 out of 1000 =========
|
| 114 |
+
01-18 13:58:30 INFO [logging.py:61]: Begin training...
|
| 115 |
+
01-18 14:13:06 INFO [logging.py:61]: Loss 'loss' on epoch 54: 5.462146282196045
|
| 116 |
+
01-18 14:13:06 INFO [logging.py:61]: Loss 'ar_loss' on epoch 54: 1.2472673654556274
|
| 117 |
+
01-18 14:13:06 INFO [logging.py:61]: Loss 'nar_loss' on epoch 54: 4.214879512786865
|
| 118 |
+
01-18 14:13:06 INFO [logging.py:61]: Loss 'ar_accuracy_metric' on epoch 54: 0.9099826216697693
|
| 119 |
+
01-18 14:13:06 INFO [logging.py:61]: Loss 'nar_acc_metric' on epoch 54: 0.4112021327018738
|
| 120 |
+
01-18 14:13:06 INFO [logging.py:61]: ========= Epoch 55 out of 1000 =========
|
| 121 |
+
01-18 14:13:06 INFO [logging.py:61]: Begin training...
|
swin_default_LR1e-3_AR-NAR/swin_default_LR1e-3_AR-NAR_2024_01_18--13_14_40.log
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
01-18 13:14:40 INFO [logger.py:80]: Initialized logger with log file in /fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/exp/swin_default_LR1e-3_AR-NAR.
|
swin_default_LR1e-3_AR-NAR/tb_log/events.out.tfevents.1705497738.gina1.1685133.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0db7b9d9f6a021d3820b3843c34b06bc51d57a21200540d977ae644206471754
|
| 3 |
+
size 2166
|
swin_default_LR1e-3_AR-NAR/tb_log/events.out.tfevents.1705497738.gina1.1685134.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c455eb1988f60f1059a5d69e417cc6f3bfd31662340202195fcfd908dd2b4517
|
| 3 |
+
size 2166
|
swin_default_LR1e-3_AR-NAR/tb_log/events.out.tfevents.1705497738.gina1.1685135.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5589a30e959b9cc127b91c8fd0778037d7efcd6c7d84e5de3df1c9de60d79f6a
|
| 3 |
+
size 2166
|
swin_default_LR1e-3_AR-NAR/tb_log/events.out.tfevents.1705497738.gina1.1685136.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bc2cf6cf8a87551e6cdaa13aaccd57b997c5523d5fcf74189141c71117e8a2b2
|
| 3 |
+
size 2166
|
swin_default_LR1e-3_AR-NAR/tb_log/events.out.tfevents.1705497802.gina1.1688371.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ad2246b8f994474bf05d7f34f921f5191105657fff5ab6f1c1ea723c76b824d7
|
| 3 |
+
size 16886
|
swin_default_LR1e-3_AR-NAR/tb_log/events.out.tfevents.1705497802.gina1.1688372.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:76f9a5e4dfe677d7a51b782cfcf642d1119c3d9ddd55f61a7d4a0fb71dc6e09e
|
| 3 |
+
size 2166
|
swin_default_LR1e-3_AR-NAR/tb_log/events.out.tfevents.1705497802.gina1.1688373.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2f5f21e4c295d18007c964da32d7b1a8ca5173b09af0ce62d0032c2d204c5390
|
| 3 |
+
size 2166
|
swin_default_LR1e-3_AR-NAR/tb_log/events.out.tfevents.1705497802.gina1.1688374.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fa97cba9197532f36bf3d5c158492e5793eac1bc64d0d3f4efbcecafd97c6854
|
| 3 |
+
size 2166
|
swin_default_LR1e-3_AR-NAR/tb_log/events.out.tfevents.1705498412.gina1.1693065.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c8f451a4e9894799899b82c2ba35bb8330502198e1ad3cdd9e6d6e0990667497
|
| 3 |
+
size 1392147
|
swin_default_LR1e-3_AR-NAR/tb_log/events.out.tfevents.1705498412.gina1.1693066.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9d9c9627dc4c32e365485f25015c3e55e198e23cd38ae374c22c50bbd71f2fe2
|
| 3 |
+
size 2166
|
swin_default_LR1e-3_AR-NAR/tb_log/events.out.tfevents.1705498412.gina1.1693067.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:45b2fa15d31fe66ac3e0ddb66e2b906fb54281a5dd9aa5f4f8fdd2631668ec0c
|
| 3 |
+
size 2166
|
swin_default_LR1e-3_AR-NAR/tb_log/events.out.tfevents.1705498412.gina1.1693068.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1ae622a26435c40efe71de00c971eb5022a6d469d2187073028ef5d338368ae9
|
| 3 |
+
size 2166
|
swin_default_LR1e-3_AR-NAR/tb_log/events.out.tfevents.1705544038.gina1.1902904.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:78c031ae9f3d415f23c57dbcf2f45c8a2971a83dad389a2c3c1bcbf750c29eda
|
| 3 |
+
size 3266
|
swin_default_LR1e-3_AR-NAR/tb_log/events.out.tfevents.1705544038.gina1.1902905.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:731d7d744187d5c858d68909f304294a1215bf76660c0a33bf0c481bbae2a6b8
|
| 3 |
+
size 2166
|
swin_default_LR1e-3_AR-NAR/tb_log/events.out.tfevents.1705544038.gina1.1902906.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4d84abd50de308ddc97b1d78aaf91cec87d6d38e3b771b6aeb05c922a61fd56b
|
| 3 |
+
size 2166
|
swin_default_LR1e-3_AR-NAR/tb_log/events.out.tfevents.1705544038.gina1.1902907.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c9d207613129eb1b3e74702a3e039ed92d6a2af1d52c45653aff2c278cb14d0a
|
| 3 |
+
size 2166
|
swin_default_LR1e-3_AR-NAR/tb_log/events.out.tfevents.1705544084.gina1.1906280.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7c2fd6ce73bd363b25ad60f086960873063a6fc723aeb8443f56ddbaf34caade
|
| 3 |
+
size 2166
|
swin_default_LR1e-3_AR-NAR/tb_log/events.out.tfevents.1705544085.gina1.1906278.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6014bdba79c64effbc71783539675bb5cffdf946d22ecbad3082683ae55c32bc
|
| 3 |
+
size 137240
|
swin_default_LR1e-3_AR-NAR/tb_log/events.out.tfevents.1705544085.gina1.1906279.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d6a1da535825cde6ff0bd5b73f77dbc9daba972a296229ec8caf86877bbe1ccb
|
| 3 |
+
size 2166
|
swin_default_LR1e-3_AR-NAR/tb_log/events.out.tfevents.1705544085.gina1.1906281.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1e2cf4c3e804063dc307695f33ebb3484035892070eb43b4cfd9026081fa18d0
|
| 3 |
+
size 2166
|
swin_default_LR1e-4_AR-NAR/config__2024_01_18--13_07_07.toml
ADDED
|
@@ -0,0 +1,89 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[[validate_dataset]]
|
| 2 |
+
path = "dataloader.Dataset"
|
| 3 |
+
|
| 4 |
+
[validate_dataset.args]
|
| 5 |
+
librispeech_dir = "~/data/LibriSpeech/LibriSpeech"
|
| 6 |
+
librispeech_metadata_fpath = "/home/xhao/proj/audiozen/recipes/librimix_sot/local/metadata/LibriSpeech/train-clean-100-24K.csv"
|
| 7 |
+
duration = 6.0
|
| 8 |
+
sr = 24000
|
| 9 |
+
num_samples = 10
|
| 10 |
+
[validate_dataset.dataloader]
|
| 11 |
+
batch_size = 1
|
| 12 |
+
num_workers = 1
|
| 13 |
+
|
| 14 |
+
[meta]
|
| 15 |
+
save_dir = "exp"
|
| 16 |
+
description = "Train a model using Generative Adversarial Networks (GANs)"
|
| 17 |
+
seed = 20220815
|
| 18 |
+
exp_id = "swin_default_LR1e-4_AR-NAR"
|
| 19 |
+
config_path = "/fred/oz325/xhao/proj/audiozen/recipes/librimix_sot/tokenizer_separation/conf/swin_default_LR1e-4_AR-NAR.toml"
|
| 20 |
+
|
| 21 |
+
[trainer]
|
| 22 |
+
path = "trainer.Trainer"
|
| 23 |
+
|
| 24 |
+
[loss_function]
|
| 25 |
+
path = "torch.nn.MSELoss"
|
| 26 |
+
|
| 27 |
+
[optimizer]
|
| 28 |
+
path = "torch.optim.AdamW"
|
| 29 |
+
|
| 30 |
+
[model]
|
| 31 |
+
path = "model_ar_nar.Model"
|
| 32 |
+
|
| 33 |
+
[acoustics]
|
| 34 |
+
n_fft = 512
|
| 35 |
+
hop_length = 128
|
| 36 |
+
win_length = 512
|
| 37 |
+
sr = 24000
|
| 38 |
+
|
| 39 |
+
[train_dataset]
|
| 40 |
+
path = "dataloader.Dataset"
|
| 41 |
+
|
| 42 |
+
[test_dataset]
|
| 43 |
+
path = "dataloader.Dataset"
|
| 44 |
+
|
| 45 |
+
[trainer.args]
|
| 46 |
+
debug = false
|
| 47 |
+
max_steps = 0
|
| 48 |
+
max_epochs = 1000
|
| 49 |
+
max_grad_norm = 1.0
|
| 50 |
+
save_max_score = true
|
| 51 |
+
save_ckpt_interval = 5
|
| 52 |
+
max_patience = 200
|
| 53 |
+
plot_norm = true
|
| 54 |
+
validation_interval = 200
|
| 55 |
+
max_num_checkpoints = 100
|
| 56 |
+
scheduler_name = "constant_schedule_with_warmup"
|
| 57 |
+
warmup_steps = 1000
|
| 58 |
+
warmup_ratio = 0.0
|
| 59 |
+
gradient_accumulation_steps = 1
|
| 60 |
+
|
| 61 |
+
[loss_function.args]
|
| 62 |
+
|
| 63 |
+
[optimizer.args]
|
| 64 |
+
lr = 0.0001
|
| 65 |
+
|
| 66 |
+
[model.args]
|
| 67 |
+
|
| 68 |
+
[train_dataset.args]
|
| 69 |
+
librispeech_dir = "~/data/LibriSpeech/LibriSpeech"
|
| 70 |
+
librispeech_metadata_fpath = "/home/xhao/proj/audiozen/recipes/librimix_sot/local/metadata/LibriSpeech/train-clean-100-24K.csv"
|
| 71 |
+
duration = 6.0
|
| 72 |
+
sr = 24000
|
| 73 |
+
|
| 74 |
+
[train_dataset.dataloader]
|
| 75 |
+
batch_size = 20
|
| 76 |
+
num_workers = 10
|
| 77 |
+
drop_last = true
|
| 78 |
+
pin_memory = true
|
| 79 |
+
|
| 80 |
+
[test_dataset.args]
|
| 81 |
+
librispeech_dir = "~/data/LibriSpeech/LibriSpeech"
|
| 82 |
+
librispeech_metadata_fpath = "/home/xhao/proj/audiozen/recipes/librimix_sot/local/metadata/LibriSpeech/train-clean-100-24K.csv"
|
| 83 |
+
duration = 6.0
|
| 84 |
+
sr = 24000
|
| 85 |
+
num_samples = 10
|
| 86 |
+
|
| 87 |
+
[test_dataset.dataloader]
|
| 88 |
+
batch_size = 1
|
| 89 |
+
num_workers = 1
|