| { | |
| "best_metric": 0.854691743850708, | |
| "best_model_checkpoint": "nrshoudi/hubert-large-ls960-ft-V2-50/checkpoint-11452", | |
| "epoch": 20.0, | |
| "eval_steps": 500, | |
| "global_step": 16360, | |
| "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": 12.8849, | |
| "step": 818 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 4.592483043670654, | |
| "eval_per": 0.9649250918684632, | |
| "eval_runtime": 118.1886, | |
| "eval_samples_per_second": 6.879, | |
| "eval_steps_per_second": 3.444, | |
| "eval_wer": 0.9551117540012132, | |
| "step": 818 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 0.0001, | |
| "loss": 2.7511, | |
| "step": 1636 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.7072813510894775, | |
| "eval_per": 0.4559738057099783, | |
| "eval_runtime": 117.681, | |
| "eval_samples_per_second": 6.909, | |
| "eval_steps_per_second": 3.459, | |
| "eval_wer": 0.4692734823386683, | |
| "step": 1636 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 9.444444444444444e-05, | |
| "loss": 1.1653, | |
| "step": 2454 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 1.1204495429992676, | |
| "eval_per": 0.1317016866107604, | |
| "eval_runtime": 118.0673, | |
| "eval_samples_per_second": 6.886, | |
| "eval_steps_per_second": 3.447, | |
| "eval_wer": 0.15342261210396155, | |
| "step": 2454 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 8.888888888888889e-05, | |
| "loss": 0.7529, | |
| "step": 3272 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 1.033559799194336, | |
| "eval_per": 0.08411853387355131, | |
| "eval_runtime": 118.2209, | |
| "eval_samples_per_second": 6.877, | |
| "eval_steps_per_second": 3.443, | |
| "eval_wer": 0.10554803788903924, | |
| "step": 3272 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 8.333333333333334e-05, | |
| "loss": 0.6309, | |
| "step": 4090 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 1.0015147924423218, | |
| "eval_per": 0.08166870818807123, | |
| "eval_runtime": 118.3158, | |
| "eval_samples_per_second": 6.871, | |
| "eval_steps_per_second": 3.44, | |
| "eval_wer": 0.10228174140264103, | |
| "step": 4090 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 7.777777777777778e-05, | |
| "loss": 0.5354, | |
| "step": 4908 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 1.0387498140335083, | |
| "eval_per": 0.07768774144916611, | |
| "eval_runtime": 118.2614, | |
| "eval_samples_per_second": 6.875, | |
| "eval_steps_per_second": 3.442, | |
| "eval_wer": 0.09924875180812841, | |
| "step": 4908 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 7.222222222222222e-05, | |
| "loss": 0.4907, | |
| "step": 5726 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 0.9956713318824768, | |
| "eval_per": 0.08925374540657684, | |
| "eval_runtime": 117.9222, | |
| "eval_samples_per_second": 6.894, | |
| "eval_steps_per_second": 3.451, | |
| "eval_wer": 0.10872101161868322, | |
| "step": 5726 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 6.666666666666667e-05, | |
| "loss": 0.4326, | |
| "step": 6544 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.8882040977478027, | |
| "eval_per": 0.08440120606802977, | |
| "eval_runtime": 118.2986, | |
| "eval_samples_per_second": 6.872, | |
| "eval_steps_per_second": 3.44, | |
| "eval_wer": 0.10914096402407727, | |
| "step": 6544 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 6.111111111111112e-05, | |
| "loss": 0.4148, | |
| "step": 7362 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 0.9541579484939575, | |
| "eval_per": 0.06376613587110148, | |
| "eval_runtime": 118.3333, | |
| "eval_samples_per_second": 6.87, | |
| "eval_steps_per_second": 3.439, | |
| "eval_wer": 0.08301059213289161, | |
| "step": 7362 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 5.555555555555556e-05, | |
| "loss": 0.3779, | |
| "step": 8180 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 0.9479207396507263, | |
| "eval_per": 0.05010364647130877, | |
| "eval_runtime": 118.3528, | |
| "eval_samples_per_second": 6.869, | |
| "eval_steps_per_second": 3.439, | |
| "eval_wer": 0.06901217861975643, | |
| "step": 8180 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "learning_rate": 5e-05, | |
| "loss": 0.3502, | |
| "step": 8998 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_loss": 0.9840469360351562, | |
| "eval_per": 0.04906718175822105, | |
| "eval_runtime": 118.5969, | |
| "eval_samples_per_second": 6.855, | |
| "eval_steps_per_second": 3.432, | |
| "eval_wer": 0.06887219448462507, | |
| "step": 8998 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 0.3294, | |
| "step": 9816 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 1.087670922279358, | |
| "eval_per": 0.04906718175822105, | |
| "eval_runtime": 118.227, | |
| "eval_samples_per_second": 6.877, | |
| "eval_steps_per_second": 3.443, | |
| "eval_wer": 0.06943213102515049, | |
| "step": 9816 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "learning_rate": 3.888888888888889e-05, | |
| "loss": 0.3239, | |
| "step": 10634 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_loss": 0.8955483436584473, | |
| "eval_per": 0.05335437670781117, | |
| "eval_runtime": 118.657, | |
| "eval_samples_per_second": 6.852, | |
| "eval_steps_per_second": 3.43, | |
| "eval_wer": 0.07311837991694274, | |
| "step": 10634 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.3069, | |
| "step": 11452 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_loss": 0.854691743850708, | |
| "eval_per": 0.05804202393291247, | |
| "eval_runtime": 119.0248, | |
| "eval_samples_per_second": 6.831, | |
| "eval_steps_per_second": 3.419, | |
| "eval_wer": 0.077597872241146, | |
| "step": 11452 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.2689, | |
| "step": 12270 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_loss": 0.968323290348053, | |
| "eval_per": 0.05245924809196269, | |
| "eval_runtime": 118.5754, | |
| "eval_samples_per_second": 6.856, | |
| "eval_steps_per_second": 3.432, | |
| "eval_wer": 0.07199850683589193, | |
| "step": 12270 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 0.2486, | |
| "step": 13088 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_loss": 0.9282185435295105, | |
| "eval_per": 0.05189390370300575, | |
| "eval_runtime": 118.5748, | |
| "eval_samples_per_second": 6.856, | |
| "eval_steps_per_second": 3.432, | |
| "eval_wer": 0.07041201997106994, | |
| "step": 13088 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.2291, | |
| "step": 13906 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_loss": 0.900354266166687, | |
| "eval_per": 0.048101385093752946, | |
| "eval_runtime": 118.0532, | |
| "eval_samples_per_second": 6.887, | |
| "eval_steps_per_second": 3.448, | |
| "eval_wer": 0.06714572348467174, | |
| "step": 13906 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "learning_rate": 1.1111111111111112e-05, | |
| "loss": 0.2294, | |
| "step": 14724 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_loss": 0.9242432117462158, | |
| "eval_per": 0.05472062564779045, | |
| "eval_runtime": 118.8448, | |
| "eval_samples_per_second": 6.841, | |
| "eval_steps_per_second": 3.425, | |
| "eval_wer": 0.07470486678176473, | |
| "step": 14724 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "learning_rate": 5.555555555555556e-06, | |
| "loss": 0.2151, | |
| "step": 15542 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_loss": 0.9400033354759216, | |
| "eval_per": 0.05540375011778008, | |
| "eval_runtime": 118.591, | |
| "eval_samples_per_second": 6.855, | |
| "eval_steps_per_second": 3.432, | |
| "eval_wer": 0.07470486678176473, | |
| "step": 15542 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 0.0, | |
| "loss": 0.2109, | |
| "step": 16360 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_loss": 0.9222491979598999, | |
| "eval_per": 0.05396683312918119, | |
| "eval_runtime": 119.3623, | |
| "eval_samples_per_second": 6.811, | |
| "eval_steps_per_second": 3.41, | |
| "eval_wer": 0.07321170267369698, | |
| "step": 16360 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 16360, | |
| "total_flos": 5.249198170404504e+18, | |
| "train_loss": 1.15744112849527, | |
| "train_runtime": 11016.1558, | |
| "train_samples_per_second": 2.97, | |
| "train_steps_per_second": 1.485 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 16360, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "total_flos": 5.249198170404504e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |