| { | |
| "best_metric": 0.8800495553294101, | |
| "best_model_checkpoint": "./output/neuralmind/bert-large-portuguese-cased-finetune-wiki_section_city/seed42-seq512-lr5e-05-epoch5-bs4-ts1.0-tssp1.0-cl0.5-2024-07-03_20:57:35/checkpoint-52000", | |
| "epoch": 4.999961539940772, | |
| "global_step": 65000, | |
| "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": 0.5927, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_EOP_f1": 0.8637108561810692, | |
| "eval_EOP_number": 11519, | |
| "eval_EOP_precision": 0.8817145957677699, | |
| "eval_EOP_recall": 0.8464276412883063, | |
| "eval_da_EOP_f1": 0.8848200312989045, | |
| "eval_da_EOP_number": 13045, | |
| "eval_da_EOP_precision": 0.9035557331202557, | |
| "eval_da_EOP_recall": 0.8668455346876198, | |
| "eval_da_overall_accuracy": 0.974613248710829, | |
| "eval_da_overall_f1": 0.8848200312989045, | |
| "eval_da_overall_precision": 0.9035557331202557, | |
| "eval_da_overall_recall": 0.8668455346876198, | |
| "eval_loss": NaN, | |
| "eval_overall_accuracy": 0.9725439457481931, | |
| "eval_overall_f1": 0.8637108561810692, | |
| "eval_overall_precision": 0.8817145957677699, | |
| "eval_overall_recall": 0.8464276412883063, | |
| "eval_runtime": 1878.3511, | |
| "eval_samples_per_second": 4.113, | |
| "eval_steps_per_second": 1.029, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4209, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_EOP_f1": 0.8753550015777848, | |
| "eval_EOP_number": 11519, | |
| "eval_EOP_precision": 0.9104463615903976, | |
| "eval_EOP_recall": 0.8428683045403247, | |
| "eval_da_EOP_f1": 0.9015245488375816, | |
| "eval_da_EOP_number": 13045, | |
| "eval_da_EOP_precision": 0.94299347061778, | |
| "eval_da_EOP_recall": 0.8635492525871982, | |
| "eval_da_overall_accuracy": 0.9787782625942086, | |
| "eval_da_overall_f1": 0.9015245488375816, | |
| "eval_da_overall_precision": 0.94299347061778, | |
| "eval_da_overall_recall": 0.8635492525871982, | |
| "eval_loss": NaN, | |
| "eval_overall_accuracy": 0.9753279200499688, | |
| "eval_overall_f1": 0.8753550015777848, | |
| "eval_overall_precision": 0.9104463615903976, | |
| "eval_overall_recall": 0.8428683045403247, | |
| "eval_runtime": 1878.6106, | |
| "eval_samples_per_second": 4.112, | |
| "eval_steps_per_second": 1.028, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 2e-05, | |
| "loss": 0.3373, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_EOP_f1": 0.879935463631067, | |
| "eval_EOP_number": 11519, | |
| "eval_EOP_precision": 0.9094867518992032, | |
| "eval_EOP_recall": 0.8522441184130567, | |
| "eval_da_EOP_f1": 0.9077958053027304, | |
| "eval_da_EOP_number": 13045, | |
| "eval_da_EOP_precision": 0.938241308793456, | |
| "eval_da_EOP_recall": 0.8792640858566501, | |
| "eval_da_overall_accuracy": 0.9799079040408396, | |
| "eval_da_overall_f1": 0.9077958053027304, | |
| "eval_da_overall_precision": 0.938241308793456, | |
| "eval_da_overall_recall": 0.8792640858566501, | |
| "eval_loss": NaN, | |
| "eval_overall_accuracy": 0.9760952975818685, | |
| "eval_overall_f1": 0.879935463631067, | |
| "eval_overall_precision": 0.9094867518992032, | |
| "eval_overall_recall": 0.8522441184130567, | |
| "eval_runtime": 1879.4394, | |
| "eval_samples_per_second": 4.11, | |
| "eval_steps_per_second": 1.028, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 1e-05, | |
| "loss": 0.259, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_EOP_f1": 0.8800495553294101, | |
| "eval_EOP_number": 11519, | |
| "eval_EOP_precision": 0.8974011911207364, | |
| "eval_EOP_recall": 0.8633561941140724, | |
| "eval_da_EOP_f1": 0.9105588594383838, | |
| "eval_da_EOP_number": 13045, | |
| "eval_da_EOP_precision": 0.9363356552729629, | |
| "eval_da_EOP_recall": 0.8861632809505557, | |
| "eval_da_overall_accuracy": 0.9804166738526809, | |
| "eval_da_overall_f1": 0.9105588594383838, | |
| "eval_da_overall_precision": 0.9363356552729629, | |
| "eval_da_overall_recall": 0.8861632809505557, | |
| "eval_loss": NaN, | |
| "eval_overall_accuracy": 0.9758097617560453, | |
| "eval_overall_f1": 0.8800495553294101, | |
| "eval_overall_precision": 0.8974011911207364, | |
| "eval_overall_recall": 0.8633561941140724, | |
| "eval_runtime": 1877.3417, | |
| "eval_samples_per_second": 4.115, | |
| "eval_steps_per_second": 1.029, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 0.0, | |
| "loss": 0.1861, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_EOP_f1": 0.8782045673287252, | |
| "eval_EOP_number": 11519, | |
| "eval_EOP_precision": 0.885175059529059, | |
| "eval_EOP_recall": 0.8713429985241774, | |
| "eval_da_EOP_f1": 0.9094169005300903, | |
| "eval_da_EOP_number": 13045, | |
| "eval_da_EOP_precision": 0.9250654190785822, | |
| "eval_da_EOP_recall": 0.8942889996167114, | |
| "eval_da_overall_accuracy": 0.9799596433437386, | |
| "eval_da_overall_f1": 0.9094169005300903, | |
| "eval_da_overall_precision": 0.9250654190785822, | |
| "eval_da_overall_recall": 0.8942889996167114, | |
| "eval_loss": NaN, | |
| "eval_overall_accuracy": 0.9751583831533863, | |
| "eval_overall_f1": 0.8782045673287252, | |
| "eval_overall_precision": 0.885175059529059, | |
| "eval_overall_recall": 0.8713429985241774, | |
| "eval_runtime": 1877.6728, | |
| "eval_samples_per_second": 4.114, | |
| "eval_steps_per_second": 1.029, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 65000, | |
| "total_flos": 4.846166765850378e+17, | |
| "train_loss": 0.3591941669170673, | |
| "train_runtime": 192526.8268, | |
| "train_samples_per_second": 1.35, | |
| "train_steps_per_second": 0.338 | |
| } | |
| ], | |
| "max_steps": 65000, | |
| "num_train_epochs": 5, | |
| "total_flos": 4.846166765850378e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |