| { | |
| "best_metric": 0.8213333333333334, | |
| "best_model_checkpoint": "Bert_tinybert-distilled/run-17/checkpoint-1376", | |
| "epoch": 9.0, | |
| "eval_steps": 500, | |
| "global_step": 1548, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.8996825218200684, | |
| "learning_rate": 0.00018817928905452526, | |
| "loss": 0.4453, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_AUC": 0.8236924444444444, | |
| "eval_BAC": 0.7266666666666667, | |
| "eval_MCC": 0.4600832969449537, | |
| "eval_PR_AUC": 0.7804386215311028, | |
| "eval_accuracy": 0.7266666666666667, | |
| "eval_loss": 0.41169798374176025, | |
| "eval_runtime": 6.2662, | |
| "eval_samples_per_second": 239.379, | |
| "eval_steps_per_second": 7.501, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 1.0351688861846924, | |
| "learning_rate": 0.0001646568779227096, | |
| "loss": 0.4213, | |
| "step": 344 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_AUC": 0.8474275555555556, | |
| "eval_BAC": 0.7453333333333334, | |
| "eval_MCC": 0.5440525853779636, | |
| "eval_PR_AUC": 0.8071261622681414, | |
| "eval_accuracy": 0.7453333333333333, | |
| "eval_loss": 0.41122254729270935, | |
| "eval_runtime": 6.2642, | |
| "eval_samples_per_second": 239.457, | |
| "eval_steps_per_second": 7.503, | |
| "step": 344 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 1.5609855651855469, | |
| "learning_rate": 0.00014113446679089392, | |
| "loss": 0.4096, | |
| "step": 516 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_AUC": 0.8763626666666667, | |
| "eval_BAC": 0.8093333333333333, | |
| "eval_MCC": 0.6314240225841081, | |
| "eval_PR_AUC": 0.841535983625851, | |
| "eval_accuracy": 0.8093333333333333, | |
| "eval_loss": 0.3761481046676636, | |
| "eval_runtime": 6.277, | |
| "eval_samples_per_second": 238.967, | |
| "eval_steps_per_second": 7.488, | |
| "step": 516 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 2.15095591545105, | |
| "learning_rate": 0.00011761205565907828, | |
| "loss": 0.3997, | |
| "step": 688 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_AUC": 0.8826080000000001, | |
| "eval_BAC": 0.8093333333333332, | |
| "eval_MCC": 0.6297353307925535, | |
| "eval_PR_AUC": 0.8579321059753027, | |
| "eval_accuracy": 0.8093333333333333, | |
| "eval_loss": 0.3721379339694977, | |
| "eval_runtime": 6.2809, | |
| "eval_samples_per_second": 238.818, | |
| "eval_steps_per_second": 7.483, | |
| "step": 688 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 1.7977828979492188, | |
| "learning_rate": 9.408964452726263e-05, | |
| "loss": 0.3901, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_AUC": 0.8896995555555555, | |
| "eval_BAC": 0.8146666666666667, | |
| "eval_MCC": 0.636998520566478, | |
| "eval_PR_AUC": 0.8700939507087682, | |
| "eval_accuracy": 0.8146666666666667, | |
| "eval_loss": 0.3646060526371002, | |
| "eval_runtime": 6.2599, | |
| "eval_samples_per_second": 239.621, | |
| "eval_steps_per_second": 7.508, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 2.0664055347442627, | |
| "learning_rate": 7.056723339544696e-05, | |
| "loss": 0.3846, | |
| "step": 1032 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_AUC": 0.8885244444444446, | |
| "eval_BAC": 0.812, | |
| "eval_MCC": 0.6345192563324739, | |
| "eval_PR_AUC": 0.8498945081379672, | |
| "eval_accuracy": 0.812, | |
| "eval_loss": 0.36806994676589966, | |
| "eval_runtime": 6.2717, | |
| "eval_samples_per_second": 239.169, | |
| "eval_steps_per_second": 7.494, | |
| "step": 1032 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 2.351468801498413, | |
| "learning_rate": 4.7044822263631315e-05, | |
| "loss": 0.3791, | |
| "step": 1204 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_AUC": 0.8900231111111112, | |
| "eval_BAC": 0.8180000000000001, | |
| "eval_MCC": 0.6418466082815267, | |
| "eval_PR_AUC": 0.8644464782061326, | |
| "eval_accuracy": 0.818, | |
| "eval_loss": 0.3628422021865845, | |
| "eval_runtime": 6.244, | |
| "eval_samples_per_second": 240.232, | |
| "eval_steps_per_second": 7.527, | |
| "step": 1204 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 1.438085675239563, | |
| "learning_rate": 2.3522411131815657e-05, | |
| "loss": 0.376, | |
| "step": 1376 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_AUC": 0.8940835555555554, | |
| "eval_BAC": 0.8213333333333334, | |
| "eval_MCC": 0.6463542230804561, | |
| "eval_PR_AUC": 0.8707386681458926, | |
| "eval_accuracy": 0.8213333333333334, | |
| "eval_loss": 0.3610929250717163, | |
| "eval_runtime": 6.2579, | |
| "eval_samples_per_second": 239.696, | |
| "eval_steps_per_second": 7.51, | |
| "step": 1376 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 1.2869964838027954, | |
| "learning_rate": 0.0, | |
| "loss": 0.3747, | |
| "step": 1548 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_AUC": 0.8969119999999999, | |
| "eval_BAC": 0.8140000000000001, | |
| "eval_MCC": 0.6394011437485282, | |
| "eval_PR_AUC": 0.8712273615291846, | |
| "eval_accuracy": 0.814, | |
| "eval_loss": 0.3661109209060669, | |
| "eval_runtime": 6.2498, | |
| "eval_samples_per_second": 240.007, | |
| "eval_steps_per_second": 7.52, | |
| "step": 1548 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 1548, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 4053399570000.0, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.6759429935635236, | |
| "learning_rate": 0.0002117017001863409, | |
| "num_train_epochs": 9, | |
| "temperature": 2 | |
| } | |
| } | |