| { | |
| "best_metric": 0.16344858705997467, | |
| "best_model_checkpoint": "pubmedbert_caption+ref_RTX6000/checkpoint-1230", | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 3690, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.4065040650406504, | |
| "grad_norm": 7.820664405822754, | |
| "learning_rate": 2.75609756097561e-05, | |
| "loss": 0.2074, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.8130081300813008, | |
| "grad_norm": 2.407073974609375, | |
| "learning_rate": 2.5121951219512197e-05, | |
| "loss": 0.1646, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9358961251405786, | |
| "eval_f1": 0.9313383960406174, | |
| "eval_f1_medical": 0.9490285342655068, | |
| "eval_f1_non_medical": 0.9136482578157279, | |
| "eval_loss": 0.16344858705997467, | |
| "eval_precision": 0.9352860582386011, | |
| "eval_recall": 0.9279290617594869, | |
| "eval_runtime": 93.0813, | |
| "eval_samples_per_second": 105.08, | |
| "eval_steps_per_second": 3.287, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 1.2195121951219512, | |
| "grad_norm": 2.0749707221984863, | |
| "learning_rate": 2.2682926829268295e-05, | |
| "loss": 0.131, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.6260162601626016, | |
| "grad_norm": 3.7739999294281006, | |
| "learning_rate": 2.024390243902439e-05, | |
| "loss": 0.1145, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9353849299662611, | |
| "eval_f1": 0.9315011393590601, | |
| "eval_f1_medical": 0.9478117258464079, | |
| "eval_f1_non_medical": 0.9151905528717122, | |
| "eval_loss": 0.1923774778842926, | |
| "eval_precision": 0.9306285500446541, | |
| "eval_recall": 0.9324104388612393, | |
| "eval_runtime": 92.7173, | |
| "eval_samples_per_second": 105.493, | |
| "eval_steps_per_second": 3.3, | |
| "step": 2460 | |
| }, | |
| { | |
| "epoch": 2.032520325203252, | |
| "grad_norm": 9.552169799804688, | |
| "learning_rate": 1.7804878048780488e-05, | |
| "loss": 0.1044, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.4390243902439024, | |
| "grad_norm": 0.6083847284317017, | |
| "learning_rate": 1.5365853658536586e-05, | |
| "loss": 0.0656, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.845528455284553, | |
| "grad_norm": 2.998614549636841, | |
| "learning_rate": 1.2926829268292684e-05, | |
| "loss": 0.0744, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9359983641754421, | |
| "eval_f1": 0.9318359818739486, | |
| "eval_f1_medical": 0.9486801114936875, | |
| "eval_f1_non_medical": 0.9149918522542096, | |
| "eval_loss": 0.23488560318946838, | |
| "eval_precision": 0.9329783496995749, | |
| "eval_recall": 0.9307471811921053, | |
| "eval_runtime": 92.6853, | |
| "eval_samples_per_second": 105.529, | |
| "eval_steps_per_second": 3.301, | |
| "step": 3690 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 6150, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "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": 3.106719963933312e+16, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |