| { | |
| "best_metric": 0.9836132732486685, | |
| "best_model_checkpoint": "./output/neuralmind/bert-large-portuguese-cased-finetune-wiki_section_disease/seed42-seq512-lr5e-05-epoch5-bs4-ts1.0-tssp1.0-cl0.5-2024-05-30_21:01:41/checkpoint-8040", | |
| "epoch": 5.0, | |
| "global_step": 13400, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 4e-05, | |
| "loss": 0.5631, | |
| "step": 2680 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_EOP_f1": 0.9767820773930753, | |
| "eval_EOP_number": 2446, | |
| "eval_EOP_precision": 0.9732142857142857, | |
| "eval_EOP_recall": 0.9803761242845462, | |
| "eval_da_EOP_f1": 0.9276059988890947, | |
| "eval_da_EOP_number": 2736, | |
| "eval_da_EOP_precision": 0.9399624765478424, | |
| "eval_da_EOP_recall": 0.9155701754385965, | |
| "eval_da_overall_accuracy": 0.9829540500479553, | |
| "eval_da_overall_f1": 0.9276059988890947, | |
| "eval_da_overall_precision": 0.9399624765478424, | |
| "eval_da_overall_recall": 0.9155701754385965, | |
| "eval_loss": NaN, | |
| "eval_overall_accuracy": 0.9944503943140882, | |
| "eval_overall_f1": 0.9767820773930753, | |
| "eval_overall_precision": 0.9732142857142857, | |
| "eval_overall_recall": 0.9803761242845462, | |
| "eval_runtime": 350.1554, | |
| "eval_samples_per_second": 4.361, | |
| "eval_steps_per_second": 1.091, | |
| "step": 2680 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4083, | |
| "step": 5360 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_EOP_f1": 0.9791751735402207, | |
| "eval_EOP_number": 2446, | |
| "eval_EOP_precision": 0.9779771615008157, | |
| "eval_EOP_recall": 0.9803761242845462, | |
| "eval_da_EOP_f1": 0.9395747855277882, | |
| "eval_da_EOP_number": 2736, | |
| "eval_da_EOP_precision": 0.9592536176694593, | |
| "eval_da_EOP_recall": 0.9206871345029239, | |
| "eval_da_overall_accuracy": 0.985874967303165, | |
| "eval_da_overall_f1": 0.9395747855277882, | |
| "eval_da_overall_precision": 0.9592536176694593, | |
| "eval_da_overall_recall": 0.9206871345029239, | |
| "eval_loss": NaN, | |
| "eval_overall_accuracy": 0.9950345633336579, | |
| "eval_overall_f1": 0.9791751735402207, | |
| "eval_overall_precision": 0.9779771615008157, | |
| "eval_overall_recall": 0.9803761242845462, | |
| "eval_runtime": 351.0792, | |
| "eval_samples_per_second": 4.349, | |
| "eval_steps_per_second": 1.088, | |
| "step": 5360 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 2e-05, | |
| "loss": 0.3595, | |
| "step": 8040 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_EOP_f1": 0.9836132732486685, | |
| "eval_EOP_number": 2446, | |
| "eval_EOP_precision": 0.985632183908046, | |
| "eval_EOP_recall": 0.9816026165167621, | |
| "eval_da_EOP_f1": 0.9373584905660378, | |
| "eval_da_EOP_number": 2736, | |
| "eval_da_EOP_precision": 0.968798751950078, | |
| "eval_da_EOP_recall": 0.9078947368421053, | |
| "eval_da_overall_accuracy": 0.985526201063737, | |
| "eval_da_overall_f1": 0.9373584905660378, | |
| "eval_da_overall_precision": 0.968798751950078, | |
| "eval_da_overall_recall": 0.9078947368421053, | |
| "eval_loss": NaN, | |
| "eval_overall_accuracy": 0.9961055398695355, | |
| "eval_overall_f1": 0.9836132732486685, | |
| "eval_overall_precision": 0.985632183908046, | |
| "eval_overall_recall": 0.9816026165167621, | |
| "eval_runtime": 350.0594, | |
| "eval_samples_per_second": 4.362, | |
| "eval_steps_per_second": 1.091, | |
| "step": 8040 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 1e-05, | |
| "loss": 0.315, | |
| "step": 10720 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_EOP_f1": 0.981170691772411, | |
| "eval_EOP_number": 2446, | |
| "eval_EOP_precision": 0.9823770491803279, | |
| "eval_EOP_recall": 0.9799672935404743, | |
| "eval_da_EOP_f1": 0.9371944339977436, | |
| "eval_da_EOP_number": 2736, | |
| "eval_da_EOP_precision": 0.965143299767622, | |
| "eval_da_EOP_recall": 0.9108187134502924, | |
| "eval_da_overall_accuracy": 0.98543900950388, | |
| "eval_da_overall_f1": 0.9371944339977436, | |
| "eval_da_overall_precision": 0.965143299767622, | |
| "eval_da_overall_recall": 0.9108187134502924, | |
| "eval_loss": NaN, | |
| "eval_overall_accuracy": 0.995521370849966, | |
| "eval_overall_f1": 0.981170691772411, | |
| "eval_overall_precision": 0.9823770491803279, | |
| "eval_overall_recall": 0.9799672935404743, | |
| "eval_runtime": 351.0283, | |
| "eval_samples_per_second": 4.35, | |
| "eval_steps_per_second": 1.088, | |
| "step": 10720 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 0.0, | |
| "loss": 0.2695, | |
| "step": 13400 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_EOP_f1": 0.9823553549446039, | |
| "eval_EOP_number": 2446, | |
| "eval_EOP_precision": 0.985996705107084, | |
| "eval_EOP_recall": 0.9787408013082584, | |
| "eval_da_EOP_f1": 0.9391206313416008, | |
| "eval_da_EOP_number": 2736, | |
| "eval_da_EOP_precision": 0.9663573085846868, | |
| "eval_da_EOP_recall": 0.9133771929824561, | |
| "eval_da_overall_accuracy": 0.985874967303165, | |
| "eval_da_overall_f1": 0.9391206313416008, | |
| "eval_da_overall_precision": 0.9663573085846868, | |
| "eval_da_overall_recall": 0.9133771929824561, | |
| "eval_loss": NaN, | |
| "eval_overall_accuracy": 0.9958134553597507, | |
| "eval_overall_f1": 0.9823553549446039, | |
| "eval_overall_precision": 0.985996705107084, | |
| "eval_overall_recall": 0.9787408013082584, | |
| "eval_runtime": 351.0391, | |
| "eval_samples_per_second": 4.35, | |
| "eval_steps_per_second": 1.088, | |
| "step": 13400 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 13400, | |
| "total_flos": 9.99040548028416e+16, | |
| "train_loss": 0.3830805900915345, | |
| "train_runtime": 38367.8373, | |
| "train_samples_per_second": 1.397, | |
| "train_steps_per_second": 0.349 | |
| } | |
| ], | |
| "max_steps": 13400, | |
| "num_train_epochs": 5, | |
| "total_flos": 9.99040548028416e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |