| { | |
| "best_metric": 0.824, | |
| "best_model_checkpoint": "Bert_tinybert-distilled/run-11/checkpoint-1376", | |
| "epoch": 8.0, | |
| "eval_steps": 500, | |
| "global_step": 1376, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 1.173068881034851, | |
| "learning_rate": 0.00024337376165851695, | |
| "loss": 0.6049, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_AUC": 0.8273706666666667, | |
| "eval_BAC": 0.7413333333333333, | |
| "eval_MCC": 0.4850333676414723, | |
| "eval_PR_AUC": 0.7881657633117372, | |
| "eval_accuracy": 0.7413333333333333, | |
| "eval_loss": 0.5346564650535583, | |
| "eval_runtime": 6.256, | |
| "eval_samples_per_second": 239.768, | |
| "eval_steps_per_second": 7.513, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 2.4093000888824463, | |
| "learning_rate": 0.00020860608142158596, | |
| "loss": 0.5475, | |
| "step": 344 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_AUC": 0.8532853333333333, | |
| "eval_BAC": 0.7966666666666666, | |
| "eval_MCC": 0.6114050721892912, | |
| "eval_PR_AUC": 0.7888869116793986, | |
| "eval_accuracy": 0.7966666666666666, | |
| "eval_loss": 0.48248833417892456, | |
| "eval_runtime": 6.2583, | |
| "eval_samples_per_second": 239.682, | |
| "eval_steps_per_second": 7.51, | |
| "step": 344 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 3.5168275833129883, | |
| "learning_rate": 0.00017383840118465497, | |
| "loss": 0.521, | |
| "step": 516 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_AUC": 0.8562666666666667, | |
| "eval_BAC": 0.798, | |
| "eval_MCC": 0.5978527601704366, | |
| "eval_PR_AUC": 0.8147153641772997, | |
| "eval_accuracy": 0.798, | |
| "eval_loss": 0.4632047414779663, | |
| "eval_runtime": 6.2642, | |
| "eval_samples_per_second": 239.455, | |
| "eval_steps_per_second": 7.503, | |
| "step": 516 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 3.0364439487457275, | |
| "learning_rate": 0.00013907072094772397, | |
| "loss": 0.5067, | |
| "step": 688 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_AUC": 0.8710017777777778, | |
| "eval_BAC": 0.8026666666666666, | |
| "eval_MCC": 0.6086337734330317, | |
| "eval_PR_AUC": 0.8431966996463283, | |
| "eval_accuracy": 0.8026666666666666, | |
| "eval_loss": 0.4548323452472687, | |
| "eval_runtime": 6.2602, | |
| "eval_samples_per_second": 239.609, | |
| "eval_steps_per_second": 7.508, | |
| "step": 688 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 4.948005199432373, | |
| "learning_rate": 0.00010430304071079298, | |
| "loss": 0.496, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_AUC": 0.8774115555555555, | |
| "eval_BAC": 0.8133333333333332, | |
| "eval_MCC": 0.6399469848044845, | |
| "eval_PR_AUC": 0.8344867532038169, | |
| "eval_accuracy": 0.8133333333333334, | |
| "eval_loss": 0.45315295457839966, | |
| "eval_runtime": 6.257, | |
| "eval_samples_per_second": 239.731, | |
| "eval_steps_per_second": 7.512, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 7.4915666580200195, | |
| "learning_rate": 6.953536047386199e-05, | |
| "loss": 0.4845, | |
| "step": 1032 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_AUC": 0.8858737777777778, | |
| "eval_BAC": 0.8113333333333332, | |
| "eval_MCC": 0.6275096493938598, | |
| "eval_PR_AUC": 0.8524215878184646, | |
| "eval_accuracy": 0.8113333333333334, | |
| "eval_loss": 0.44440844655036926, | |
| "eval_runtime": 6.2656, | |
| "eval_samples_per_second": 239.403, | |
| "eval_steps_per_second": 7.501, | |
| "step": 1032 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 3.649087905883789, | |
| "learning_rate": 3.476768023693099e-05, | |
| "loss": 0.4771, | |
| "step": 1204 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_AUC": 0.8873937777777777, | |
| "eval_BAC": 0.8180000000000001, | |
| "eval_MCC": 0.6409466868541137, | |
| "eval_PR_AUC": 0.8535373088628392, | |
| "eval_accuracy": 0.818, | |
| "eval_loss": 0.43640071153640747, | |
| "eval_runtime": 6.2505, | |
| "eval_samples_per_second": 239.982, | |
| "eval_steps_per_second": 7.519, | |
| "step": 1204 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 2.1959824562072754, | |
| "learning_rate": 0.0, | |
| "loss": 0.4739, | |
| "step": 1376 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_AUC": 0.8897351111111111, | |
| "eval_BAC": 0.8240000000000001, | |
| "eval_MCC": 0.651352819389177, | |
| "eval_PR_AUC": 0.8590307681592729, | |
| "eval_accuracy": 0.824, | |
| "eval_loss": 0.4306699335575104, | |
| "eval_runtime": 6.2445, | |
| "eval_samples_per_second": 240.211, | |
| "eval_steps_per_second": 7.527, | |
| "step": 1376 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 1376, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 8, | |
| "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": 3603021840000.0, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.9508242983728342, | |
| "learning_rate": 0.00027814144189544795, | |
| "num_train_epochs": 8, | |
| "temperature": 2 | |
| } | |
| } | |