| { | |
| "best_metric": 0.8213333333333334, | |
| "best_model_checkpoint": "Bert_tinybert-distilled/run-34/checkpoint-860", | |
| "epoch": 7.0, | |
| "eval_steps": 500, | |
| "global_step": 1204, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 1.2015043497085571, | |
| "learning_rate": 0.00027537877162579386, | |
| "loss": 0.5903, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_AUC": 0.8227715555555557, | |
| "eval_BAC": 0.7293333333333334, | |
| "eval_MCC": 0.462798882269463, | |
| "eval_PR_AUC": 0.7817423653297785, | |
| "eval_accuracy": 0.7293333333333333, | |
| "eval_loss": 0.524067223072052, | |
| "eval_runtime": 6.2657, | |
| "eval_samples_per_second": 239.398, | |
| "eval_steps_per_second": 7.501, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 2.673234224319458, | |
| "learning_rate": 0.00024095642517256964, | |
| "loss": 0.5391, | |
| "step": 344 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_AUC": 0.8590240000000001, | |
| "eval_BAC": 0.7893333333333333, | |
| "eval_MCC": 0.6018126794174047, | |
| "eval_PR_AUC": 0.8177888157143621, | |
| "eval_accuracy": 0.7893333333333333, | |
| "eval_loss": 0.48427751660346985, | |
| "eval_runtime": 6.2417, | |
| "eval_samples_per_second": 240.319, | |
| "eval_steps_per_second": 7.53, | |
| "step": 344 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 2.158290147781372, | |
| "learning_rate": 0.00020653407871934538, | |
| "loss": 0.5144, | |
| "step": 516 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_AUC": 0.874496, | |
| "eval_BAC": 0.8166666666666667, | |
| "eval_MCC": 0.6357253352621931, | |
| "eval_PR_AUC": 0.8445948803832817, | |
| "eval_accuracy": 0.8166666666666667, | |
| "eval_loss": 0.45355746150016785, | |
| "eval_runtime": 6.2253, | |
| "eval_samples_per_second": 240.953, | |
| "eval_steps_per_second": 7.55, | |
| "step": 516 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 3.349181890487671, | |
| "learning_rate": 0.00017211173226612116, | |
| "loss": 0.4983, | |
| "step": 688 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_AUC": 0.8846435555555555, | |
| "eval_BAC": 0.8160000000000001, | |
| "eval_MCC": 0.6361963355504687, | |
| "eval_PR_AUC": 0.8646294269782169, | |
| "eval_accuracy": 0.816, | |
| "eval_loss": 0.43939444422721863, | |
| "eval_runtime": 6.2032, | |
| "eval_samples_per_second": 241.81, | |
| "eval_steps_per_second": 7.577, | |
| "step": 688 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 3.5103800296783447, | |
| "learning_rate": 0.00013768938581289693, | |
| "loss": 0.4834, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_AUC": 0.8878346666666667, | |
| "eval_BAC": 0.8213333333333334, | |
| "eval_MCC": 0.6458179094010764, | |
| "eval_PR_AUC": 0.8644689328040056, | |
| "eval_accuracy": 0.8213333333333334, | |
| "eval_loss": 0.4339253008365631, | |
| "eval_runtime": 6.2328, | |
| "eval_samples_per_second": 240.664, | |
| "eval_steps_per_second": 7.541, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 5.336594581604004, | |
| "learning_rate": 0.00010326703935967269, | |
| "loss": 0.4746, | |
| "step": 1032 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_AUC": 0.8907893333333334, | |
| "eval_BAC": 0.8106666666666666, | |
| "eval_MCC": 0.6252673865062606, | |
| "eval_PR_AUC": 0.878022223565124, | |
| "eval_accuracy": 0.8106666666666666, | |
| "eval_loss": 0.4312085807323456, | |
| "eval_runtime": 6.2382, | |
| "eval_samples_per_second": 240.453, | |
| "eval_steps_per_second": 7.534, | |
| "step": 1032 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 4.427212715148926, | |
| "learning_rate": 6.884469290644847e-05, | |
| "loss": 0.4628, | |
| "step": 1204 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_AUC": 0.8948551111111112, | |
| "eval_BAC": 0.8200000000000001, | |
| "eval_MCC": 0.64296966214199, | |
| "eval_PR_AUC": 0.8703491534724792, | |
| "eval_accuracy": 0.82, | |
| "eval_loss": 0.4243921637535095, | |
| "eval_runtime": 6.2243, | |
| "eval_samples_per_second": 240.991, | |
| "eval_steps_per_second": 7.551, | |
| "step": 1204 | |
| } | |
| ], | |
| "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": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3152644110000.0, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.9233020385816241, | |
| "learning_rate": 0.0003098011180790181, | |
| "num_train_epochs": 9, | |
| "temperature": 6 | |
| } | |
| } | |