|
{ |
|
"best_metric": 0.7043963670730591, |
|
"best_model_checkpoint": "ai-light-dance_singing3_ft_wav2vec2-large-xlsr-53-v1/checkpoint-720", |
|
"epoch": 9.996539792387543, |
|
"global_step": 720, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.14, |
|
"learning_rate": 7.000000000000001e-06, |
|
"loss": 1.7072, |
|
"step": 10 |
|
}, |
|
{ |
|
"epoch": 0.28, |
|
"learning_rate": 1.7000000000000003e-05, |
|
"loss": 1.3169, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 0.42, |
|
"learning_rate": 2.7000000000000002e-05, |
|
"loss": 1.1845, |
|
"step": 30 |
|
}, |
|
{ |
|
"epoch": 0.55, |
|
"learning_rate": 3.7e-05, |
|
"loss": 1.1996, |
|
"step": 40 |
|
}, |
|
{ |
|
"epoch": 0.69, |
|
"learning_rate": 4.7e-05, |
|
"loss": 1.0979, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 0.83, |
|
"learning_rate": 4.9477611940298504e-05, |
|
"loss": 1.095, |
|
"step": 60 |
|
}, |
|
{ |
|
"epoch": 0.97, |
|
"learning_rate": 4.8731343283582094e-05, |
|
"loss": 1.0489, |
|
"step": 70 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"eval_loss": 0.9701907634735107, |
|
"eval_runtime": 131.7628, |
|
"eval_samples_per_second": 4.038, |
|
"eval_steps_per_second": 0.508, |
|
"eval_wer": 0.6696021083756508, |
|
"step": 72 |
|
}, |
|
{ |
|
"epoch": 1.11, |
|
"learning_rate": 4.798507462686567e-05, |
|
"loss": 1.0915, |
|
"step": 80 |
|
}, |
|
{ |
|
"epoch": 1.25, |
|
"learning_rate": 4.723880597014926e-05, |
|
"loss": 1.1852, |
|
"step": 90 |
|
}, |
|
{ |
|
"epoch": 1.39, |
|
"learning_rate": 4.6492537313432837e-05, |
|
"loss": 1.1105, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 1.53, |
|
"learning_rate": 4.582089552238806e-05, |
|
"loss": 1.052, |
|
"step": 110 |
|
}, |
|
{ |
|
"epoch": 1.66, |
|
"learning_rate": 4.5074626865671645e-05, |
|
"loss": 1.0009, |
|
"step": 120 |
|
}, |
|
{ |
|
"epoch": 1.8, |
|
"learning_rate": 4.432835820895523e-05, |
|
"loss": 1.0445, |
|
"step": 130 |
|
}, |
|
{ |
|
"epoch": 1.94, |
|
"learning_rate": 4.358208955223881e-05, |
|
"loss": 0.9699, |
|
"step": 140 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"eval_loss": 0.9321106672286987, |
|
"eval_runtime": 131.1021, |
|
"eval_samples_per_second": 4.058, |
|
"eval_steps_per_second": 0.511, |
|
"eval_wer": 0.6595744680851063, |
|
"step": 144 |
|
}, |
|
{ |
|
"epoch": 2.08, |
|
"learning_rate": 4.283582089552239e-05, |
|
"loss": 1.0048, |
|
"step": 150 |
|
}, |
|
{ |
|
"epoch": 2.22, |
|
"learning_rate": 4.208955223880597e-05, |
|
"loss": 0.9799, |
|
"step": 160 |
|
}, |
|
{ |
|
"epoch": 2.36, |
|
"learning_rate": 4.1343283582089554e-05, |
|
"loss": 1.034, |
|
"step": 170 |
|
}, |
|
{ |
|
"epoch": 2.5, |
|
"learning_rate": 4.059701492537314e-05, |
|
"loss": 0.9303, |
|
"step": 180 |
|
}, |
|
{ |
|
"epoch": 2.64, |
|
"learning_rate": 3.985074626865672e-05, |
|
"loss": 0.9381, |
|
"step": 190 |
|
}, |
|
{ |
|
"epoch": 2.78, |
|
"learning_rate": 3.9104477611940296e-05, |
|
"loss": 0.9386, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 2.91, |
|
"learning_rate": 3.8358208955223886e-05, |
|
"loss": 0.9074, |
|
"step": 210 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"eval_loss": 0.854228138923645, |
|
"eval_runtime": 129.803, |
|
"eval_samples_per_second": 4.099, |
|
"eval_steps_per_second": 0.516, |
|
"eval_wer": 0.6138715690685865, |
|
"step": 216 |
|
}, |
|
{ |
|
"epoch": 3.06, |
|
"learning_rate": 3.761194029850746e-05, |
|
"loss": 0.9405, |
|
"step": 220 |
|
}, |
|
{ |
|
"epoch": 3.19, |
|
"learning_rate": 3.6865671641791046e-05, |
|
"loss": 0.9118, |
|
"step": 230 |
|
}, |
|
{ |
|
"epoch": 3.33, |
|
"learning_rate": 3.611940298507463e-05, |
|
"loss": 0.9953, |
|
"step": 240 |
|
}, |
|
{ |
|
"epoch": 3.47, |
|
"learning_rate": 3.537313432835821e-05, |
|
"loss": 0.9019, |
|
"step": 250 |
|
}, |
|
{ |
|
"epoch": 3.61, |
|
"learning_rate": 3.4626865671641795e-05, |
|
"loss": 0.8914, |
|
"step": 260 |
|
}, |
|
{ |
|
"epoch": 3.75, |
|
"learning_rate": 3.388059701492537e-05, |
|
"loss": 0.9089, |
|
"step": 270 |
|
}, |
|
{ |
|
"epoch": 3.89, |
|
"learning_rate": 3.3134328358208955e-05, |
|
"loss": 0.8979, |
|
"step": 280 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"eval_loss": 0.8186684846878052, |
|
"eval_runtime": 131.2574, |
|
"eval_samples_per_second": 4.053, |
|
"eval_steps_per_second": 0.51, |
|
"eval_wer": 0.5776820723789934, |
|
"step": 288 |
|
}, |
|
{ |
|
"epoch": 4.03, |
|
"learning_rate": 3.238805970149254e-05, |
|
"loss": 0.9317, |
|
"step": 290 |
|
}, |
|
{ |
|
"epoch": 4.17, |
|
"learning_rate": 3.164179104477612e-05, |
|
"loss": 0.887, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 4.3, |
|
"learning_rate": 3.0895522388059704e-05, |
|
"loss": 0.8859, |
|
"step": 310 |
|
}, |
|
{ |
|
"epoch": 4.44, |
|
"learning_rate": 3.0149253731343284e-05, |
|
"loss": 0.8687, |
|
"step": 320 |
|
}, |
|
{ |
|
"epoch": 4.58, |
|
"learning_rate": 2.9402985074626864e-05, |
|
"loss": 0.9075, |
|
"step": 330 |
|
}, |
|
{ |
|
"epoch": 4.72, |
|
"learning_rate": 2.865671641791045e-05, |
|
"loss": 0.8513, |
|
"step": 340 |
|
}, |
|
{ |
|
"epoch": 4.86, |
|
"learning_rate": 2.791044776119403e-05, |
|
"loss": 0.8585, |
|
"step": 350 |
|
}, |
|
{ |
|
"epoch": 5.0, |
|
"learning_rate": 2.716417910447761e-05, |
|
"loss": 0.8363, |
|
"step": 360 |
|
}, |
|
{ |
|
"epoch": 5.0, |
|
"eval_loss": 0.7735126614570618, |
|
"eval_runtime": 129.316, |
|
"eval_samples_per_second": 4.114, |
|
"eval_steps_per_second": 0.518, |
|
"eval_wer": 0.5551198817252684, |
|
"step": 360 |
|
}, |
|
{ |
|
"epoch": 5.14, |
|
"learning_rate": 2.6417910447761197e-05, |
|
"loss": 0.8316, |
|
"step": 370 |
|
}, |
|
{ |
|
"epoch": 5.28, |
|
"learning_rate": 2.5671641791044776e-05, |
|
"loss": 0.8286, |
|
"step": 380 |
|
}, |
|
{ |
|
"epoch": 5.42, |
|
"learning_rate": 2.492537313432836e-05, |
|
"loss": 0.8229, |
|
"step": 390 |
|
}, |
|
{ |
|
"epoch": 5.55, |
|
"learning_rate": 2.4179104477611943e-05, |
|
"loss": 0.8304, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 5.69, |
|
"learning_rate": 2.3432835820895522e-05, |
|
"loss": 0.8113, |
|
"step": 410 |
|
}, |
|
{ |
|
"epoch": 5.83, |
|
"learning_rate": 2.2686567164179106e-05, |
|
"loss": 0.8642, |
|
"step": 420 |
|
}, |
|
{ |
|
"epoch": 5.97, |
|
"learning_rate": 2.1940298507462685e-05, |
|
"loss": 0.802, |
|
"step": 430 |
|
}, |
|
{ |
|
"epoch": 6.0, |
|
"eval_loss": 0.7765986919403076, |
|
"eval_runtime": 128.8303, |
|
"eval_samples_per_second": 4.129, |
|
"eval_steps_per_second": 0.52, |
|
"eval_wer": 0.5402069807803561, |
|
"step": 432 |
|
}, |
|
{ |
|
"epoch": 6.11, |
|
"learning_rate": 2.119402985074627e-05, |
|
"loss": 0.8724, |
|
"step": 440 |
|
}, |
|
{ |
|
"epoch": 6.25, |
|
"learning_rate": 2.0447761194029852e-05, |
|
"loss": 0.7674, |
|
"step": 450 |
|
}, |
|
{ |
|
"epoch": 6.39, |
|
"learning_rate": 1.9701492537313435e-05, |
|
"loss": 0.7822, |
|
"step": 460 |
|
}, |
|
{ |
|
"epoch": 6.53, |
|
"learning_rate": 1.8955223880597015e-05, |
|
"loss": 0.8182, |
|
"step": 470 |
|
}, |
|
{ |
|
"epoch": 6.66, |
|
"learning_rate": 1.8208955223880598e-05, |
|
"loss": 0.8082, |
|
"step": 480 |
|
}, |
|
{ |
|
"epoch": 6.8, |
|
"learning_rate": 1.746268656716418e-05, |
|
"loss": 0.8026, |
|
"step": 490 |
|
}, |
|
{ |
|
"epoch": 6.94, |
|
"learning_rate": 1.6716417910447764e-05, |
|
"loss": 0.7702, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 7.0, |
|
"eval_loss": 0.7470704317092896, |
|
"eval_runtime": 130.8433, |
|
"eval_samples_per_second": 4.066, |
|
"eval_steps_per_second": 0.512, |
|
"eval_wer": 0.49713955132737675, |
|
"step": 504 |
|
}, |
|
{ |
|
"epoch": 7.08, |
|
"learning_rate": 1.5970149253731344e-05, |
|
"loss": 0.8414, |
|
"step": 510 |
|
}, |
|
{ |
|
"epoch": 7.22, |
|
"learning_rate": 1.5223880597014925e-05, |
|
"loss": 0.7644, |
|
"step": 520 |
|
}, |
|
{ |
|
"epoch": 7.36, |
|
"learning_rate": 1.4477611940298509e-05, |
|
"loss": 0.7924, |
|
"step": 530 |
|
}, |
|
{ |
|
"epoch": 7.5, |
|
"learning_rate": 1.373134328358209e-05, |
|
"loss": 0.7678, |
|
"step": 540 |
|
}, |
|
{ |
|
"epoch": 7.64, |
|
"learning_rate": 1.2985074626865673e-05, |
|
"loss": 0.7541, |
|
"step": 550 |
|
}, |
|
{ |
|
"epoch": 7.78, |
|
"learning_rate": 1.2238805970149255e-05, |
|
"loss": 0.7677, |
|
"step": 560 |
|
}, |
|
{ |
|
"epoch": 7.91, |
|
"learning_rate": 1.1492537313432836e-05, |
|
"loss": 0.7483, |
|
"step": 570 |
|
}, |
|
{ |
|
"epoch": 8.0, |
|
"eval_loss": 0.7225961089134216, |
|
"eval_runtime": 129.9737, |
|
"eval_samples_per_second": 4.093, |
|
"eval_steps_per_second": 0.515, |
|
"eval_wer": 0.49810374750915987, |
|
"step": 576 |
|
}, |
|
{ |
|
"epoch": 8.06, |
|
"learning_rate": 1.0746268656716418e-05, |
|
"loss": 0.7906, |
|
"step": 580 |
|
}, |
|
{ |
|
"epoch": 8.19, |
|
"learning_rate": 1e-05, |
|
"loss": 0.7329, |
|
"step": 590 |
|
}, |
|
{ |
|
"epoch": 8.33, |
|
"learning_rate": 9.253731343283582e-06, |
|
"loss": 0.7563, |
|
"step": 600 |
|
}, |
|
{ |
|
"epoch": 8.47, |
|
"learning_rate": 8.507462686567165e-06, |
|
"loss": 0.7423, |
|
"step": 610 |
|
}, |
|
{ |
|
"epoch": 8.61, |
|
"learning_rate": 7.761194029850747e-06, |
|
"loss": 0.7615, |
|
"step": 620 |
|
}, |
|
{ |
|
"epoch": 8.75, |
|
"learning_rate": 7.014925373134329e-06, |
|
"loss": 0.7527, |
|
"step": 630 |
|
}, |
|
{ |
|
"epoch": 8.89, |
|
"learning_rate": 6.26865671641791e-06, |
|
"loss": 0.7196, |
|
"step": 640 |
|
}, |
|
{ |
|
"epoch": 9.0, |
|
"eval_loss": 0.709452748298645, |
|
"eval_runtime": 130.8637, |
|
"eval_samples_per_second": 4.065, |
|
"eval_steps_per_second": 0.512, |
|
"eval_wer": 0.4842193224914829, |
|
"step": 648 |
|
}, |
|
{ |
|
"epoch": 9.03, |
|
"learning_rate": 5.522388059701493e-06, |
|
"loss": 0.7642, |
|
"step": 650 |
|
}, |
|
{ |
|
"epoch": 9.17, |
|
"learning_rate": 4.7761194029850745e-06, |
|
"loss": 0.7204, |
|
"step": 660 |
|
}, |
|
{ |
|
"epoch": 9.3, |
|
"learning_rate": 4.029850746268657e-06, |
|
"loss": 0.7626, |
|
"step": 670 |
|
}, |
|
{ |
|
"epoch": 9.44, |
|
"learning_rate": 3.2835820895522387e-06, |
|
"loss": 0.7175, |
|
"step": 680 |
|
}, |
|
{ |
|
"epoch": 9.58, |
|
"learning_rate": 2.537313432835821e-06, |
|
"loss": 0.7452, |
|
"step": 690 |
|
}, |
|
{ |
|
"epoch": 9.72, |
|
"learning_rate": 1.7910447761194032e-06, |
|
"loss": 0.7173, |
|
"step": 700 |
|
}, |
|
{ |
|
"epoch": 9.86, |
|
"learning_rate": 1.0447761194029853e-06, |
|
"loss": 0.743, |
|
"step": 710 |
|
}, |
|
{ |
|
"epoch": 10.0, |
|
"learning_rate": 2.9850746268656716e-07, |
|
"loss": 0.731, |
|
"step": 720 |
|
}, |
|
{ |
|
"epoch": 10.0, |
|
"eval_loss": 0.7043963670730591, |
|
"eval_runtime": 129.9462, |
|
"eval_samples_per_second": 4.094, |
|
"eval_steps_per_second": 0.516, |
|
"eval_wer": 0.47573439609179147, |
|
"step": 720 |
|
}, |
|
{ |
|
"epoch": 10.0, |
|
"step": 720, |
|
"total_flos": 2.7197958559499194e+19, |
|
"train_loss": 0.8933021876547071, |
|
"train_runtime": 19496.7314, |
|
"train_samples_per_second": 2.368, |
|
"train_steps_per_second": 0.037 |
|
} |
|
], |
|
"max_steps": 720, |
|
"num_train_epochs": 10, |
|
"total_flos": 2.7197958559499194e+19, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|