|
{ |
|
"best_metric": 0.4499486982822418, |
|
"best_model_checkpoint": "nrshoudi/hubert-large-ls960-ft-V2-10/checkpoint-2788", |
|
"epoch": 20.0, |
|
"eval_steps": 500, |
|
"global_step": 3280, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 1.0, |
|
"learning_rate": 5e-05, |
|
"loss": 14.6873, |
|
"step": 164 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"eval_loss": 3.228632688522339, |
|
"eval_per": 1.0, |
|
"eval_runtime": 115.7091, |
|
"eval_samples_per_second": 7.026, |
|
"eval_steps_per_second": 3.517, |
|
"eval_wer": 1.0, |
|
"step": 164 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"learning_rate": 0.0001, |
|
"loss": 3.2356, |
|
"step": 328 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"eval_loss": 3.102414131164551, |
|
"eval_per": 1.0, |
|
"eval_runtime": 118.6897, |
|
"eval_samples_per_second": 6.85, |
|
"eval_steps_per_second": 3.429, |
|
"eval_wer": 1.0, |
|
"step": 328 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"learning_rate": 9.444444444444444e-05, |
|
"loss": 2.6803, |
|
"step": 492 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"eval_loss": 1.983318567276001, |
|
"eval_per": 0.9381654574578348, |
|
"eval_runtime": 119.1436, |
|
"eval_samples_per_second": 6.824, |
|
"eval_steps_per_second": 3.416, |
|
"eval_wer": 0.922402127758854, |
|
"step": 492 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"learning_rate": 8.888888888888889e-05, |
|
"loss": 1.5141, |
|
"step": 656 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"eval_loss": 1.105877161026001, |
|
"eval_per": 0.3923490059361161, |
|
"eval_runtime": 119.3145, |
|
"eval_samples_per_second": 6.814, |
|
"eval_steps_per_second": 3.411, |
|
"eval_wer": 0.409873547664598, |
|
"step": 656 |
|
}, |
|
{ |
|
"epoch": 5.0, |
|
"learning_rate": 8.333333333333334e-05, |
|
"loss": 0.7862, |
|
"step": 820 |
|
}, |
|
{ |
|
"epoch": 5.0, |
|
"eval_loss": 0.6128574013710022, |
|
"eval_per": 0.1356826533496655, |
|
"eval_runtime": 119.0095, |
|
"eval_samples_per_second": 6.831, |
|
"eval_steps_per_second": 3.42, |
|
"eval_wer": 0.15874200923895293, |
|
"step": 820 |
|
}, |
|
{ |
|
"epoch": 6.0, |
|
"learning_rate": 7.777777777777778e-05, |
|
"loss": 0.4004, |
|
"step": 984 |
|
}, |
|
{ |
|
"epoch": 6.0, |
|
"eval_loss": 0.4938863515853882, |
|
"eval_per": 0.07483746348817488, |
|
"eval_runtime": 118.5649, |
|
"eval_samples_per_second": 6.857, |
|
"eval_steps_per_second": 3.433, |
|
"eval_wer": 0.09392935467313704, |
|
"step": 984 |
|
}, |
|
{ |
|
"epoch": 7.0, |
|
"learning_rate": 7.222222222222222e-05, |
|
"loss": 0.2791, |
|
"step": 1148 |
|
}, |
|
{ |
|
"epoch": 7.0, |
|
"eval_loss": 0.48878785967826843, |
|
"eval_per": 0.06463770847074343, |
|
"eval_runtime": 121.6382, |
|
"eval_samples_per_second": 6.684, |
|
"eval_steps_per_second": 3.346, |
|
"eval_wer": 0.08151742802482385, |
|
"step": 1148 |
|
}, |
|
{ |
|
"epoch": 8.0, |
|
"learning_rate": 6.666666666666667e-05, |
|
"loss": 0.2168, |
|
"step": 1312 |
|
}, |
|
{ |
|
"epoch": 8.0, |
|
"eval_loss": 0.5082865953445435, |
|
"eval_per": 0.06623951757278809, |
|
"eval_runtime": 120.0011, |
|
"eval_samples_per_second": 6.775, |
|
"eval_steps_per_second": 3.392, |
|
"eval_wer": 0.08296393075451448, |
|
"step": 1312 |
|
}, |
|
{ |
|
"epoch": 9.0, |
|
"learning_rate": 6.111111111111112e-05, |
|
"loss": 0.1726, |
|
"step": 1476 |
|
}, |
|
{ |
|
"epoch": 9.0, |
|
"eval_loss": 0.47479337453842163, |
|
"eval_per": 0.05957316498633751, |
|
"eval_runtime": 120.2413, |
|
"eval_samples_per_second": 6.761, |
|
"eval_steps_per_second": 3.385, |
|
"eval_wer": 0.0748915122952732, |
|
"step": 1476 |
|
}, |
|
{ |
|
"epoch": 10.0, |
|
"learning_rate": 5.555555555555556e-05, |
|
"loss": 0.1412, |
|
"step": 1640 |
|
}, |
|
{ |
|
"epoch": 10.0, |
|
"eval_loss": 0.49549877643585205, |
|
"eval_per": 0.0575237915763686, |
|
"eval_runtime": 120.5928, |
|
"eval_samples_per_second": 6.742, |
|
"eval_steps_per_second": 3.375, |
|
"eval_wer": 0.07419159161961644, |
|
"step": 1640 |
|
}, |
|
{ |
|
"epoch": 11.0, |
|
"learning_rate": 5e-05, |
|
"loss": 0.1156, |
|
"step": 1804 |
|
}, |
|
{ |
|
"epoch": 11.0, |
|
"eval_loss": 0.49861228466033936, |
|
"eval_per": 0.056393102798454726, |
|
"eval_runtime": 120.178, |
|
"eval_samples_per_second": 6.765, |
|
"eval_steps_per_second": 3.387, |
|
"eval_wer": 0.07148523167374364, |
|
"step": 1804 |
|
}, |
|
{ |
|
"epoch": 12.0, |
|
"learning_rate": 4.4444444444444447e-05, |
|
"loss": 0.1321, |
|
"step": 1968 |
|
}, |
|
{ |
|
"epoch": 12.0, |
|
"eval_loss": 0.5101034045219421, |
|
"eval_per": 0.05516818995571469, |
|
"eval_runtime": 119.6692, |
|
"eval_samples_per_second": 6.794, |
|
"eval_steps_per_second": 3.401, |
|
"eval_wer": 0.07027203583593859, |
|
"step": 1968 |
|
}, |
|
{ |
|
"epoch": 13.0, |
|
"learning_rate": 3.888888888888889e-05, |
|
"loss": 0.103, |
|
"step": 2132 |
|
}, |
|
{ |
|
"epoch": 13.0, |
|
"eval_loss": 0.4728148877620697, |
|
"eval_per": 0.052718364270234616, |
|
"eval_runtime": 120.0092, |
|
"eval_samples_per_second": 6.774, |
|
"eval_steps_per_second": 3.391, |
|
"eval_wer": 0.0667724324576548, |
|
"step": 2132 |
|
}, |
|
{ |
|
"epoch": 14.0, |
|
"learning_rate": 3.3333333333333335e-05, |
|
"loss": 0.0772, |
|
"step": 2296 |
|
}, |
|
{ |
|
"epoch": 14.0, |
|
"eval_loss": 0.48323920369148254, |
|
"eval_per": 0.050880995006124564, |
|
"eval_runtime": 120.2949, |
|
"eval_samples_per_second": 6.758, |
|
"eval_steps_per_second": 3.383, |
|
"eval_wer": 0.06485931594419299, |
|
"step": 2296 |
|
}, |
|
{ |
|
"epoch": 15.0, |
|
"learning_rate": 2.777777777777778e-05, |
|
"loss": 0.0858, |
|
"step": 2460 |
|
}, |
|
{ |
|
"epoch": 15.0, |
|
"eval_loss": 0.4829542934894562, |
|
"eval_per": 0.05045698671440686, |
|
"eval_runtime": 120.6615, |
|
"eval_samples_per_second": 6.738, |
|
"eval_steps_per_second": 3.373, |
|
"eval_wer": 0.06490597732257011, |
|
"step": 2460 |
|
}, |
|
{ |
|
"epoch": 16.0, |
|
"learning_rate": 2.2222222222222223e-05, |
|
"loss": 0.0874, |
|
"step": 2624 |
|
}, |
|
{ |
|
"epoch": 16.0, |
|
"eval_loss": 0.4697466194629669, |
|
"eval_per": 0.050928107038537646, |
|
"eval_runtime": 120.1647, |
|
"eval_samples_per_second": 6.766, |
|
"eval_steps_per_second": 3.387, |
|
"eval_wer": 0.06420605664691335, |
|
"step": 2624 |
|
}, |
|
{ |
|
"epoch": 17.0, |
|
"learning_rate": 1.6666666666666667e-05, |
|
"loss": 0.0784, |
|
"step": 2788 |
|
}, |
|
{ |
|
"epoch": 17.0, |
|
"eval_loss": 0.4499486982822418, |
|
"eval_per": 0.05163478752473382, |
|
"eval_runtime": 120.3424, |
|
"eval_samples_per_second": 6.756, |
|
"eval_steps_per_second": 3.382, |
|
"eval_wer": 0.06523260697120993, |
|
"step": 2788 |
|
}, |
|
{ |
|
"epoch": 18.0, |
|
"learning_rate": 1.1111111111111112e-05, |
|
"loss": 0.0703, |
|
"step": 2952 |
|
}, |
|
{ |
|
"epoch": 18.0, |
|
"eval_loss": 0.4699481725692749, |
|
"eval_per": 0.04998586639027608, |
|
"eval_runtime": 119.817, |
|
"eval_samples_per_second": 6.785, |
|
"eval_steps_per_second": 3.397, |
|
"eval_wer": 0.06383276561989641, |
|
"step": 2952 |
|
}, |
|
{ |
|
"epoch": 19.0, |
|
"learning_rate": 5.555555555555556e-06, |
|
"loss": 0.062, |
|
"step": 3116 |
|
}, |
|
{ |
|
"epoch": 19.0, |
|
"eval_loss": 0.47566303610801697, |
|
"eval_per": 0.04956185809855837, |
|
"eval_runtime": 120.5098, |
|
"eval_samples_per_second": 6.746, |
|
"eval_steps_per_second": 3.377, |
|
"eval_wer": 0.06336615183612523, |
|
"step": 3116 |
|
}, |
|
{ |
|
"epoch": 20.0, |
|
"learning_rate": 0.0, |
|
"loss": 0.0588, |
|
"step": 3280 |
|
}, |
|
{ |
|
"epoch": 20.0, |
|
"eval_loss": 0.4727163016796112, |
|
"eval_per": 0.04904362574201451, |
|
"eval_runtime": 120.1063, |
|
"eval_samples_per_second": 6.769, |
|
"eval_steps_per_second": 3.389, |
|
"eval_wer": 0.06280621529559983, |
|
"step": 3280 |
|
}, |
|
{ |
|
"epoch": 20.0, |
|
"step": 3280, |
|
"total_flos": 1.036683517600052e+18, |
|
"train_loss": 1.2492202968132204, |
|
"train_runtime": 4352.7187, |
|
"train_samples_per_second": 1.503, |
|
"train_steps_per_second": 0.754 |
|
} |
|
], |
|
"logging_steps": 500, |
|
"max_steps": 3280, |
|
"num_train_epochs": 20, |
|
"save_steps": 500, |
|
"total_flos": 1.036683517600052e+18, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|