|
{ |
|
"best_metric": 0.7617828961399019, |
|
"best_model_checkpoint": "./output/neuralmind/bert-large-portuguese-cased-finetune-wiki_section_disease/seed42-seq512-lr5e-05-epoch5-bs4-ts1.0-tssp1.0-cl0.5-2024-07-14_00:16:54/checkpoint-10740", |
|
"epoch": 5.0, |
|
"global_step": 13425, |
|
"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": 0.9837, |
|
"step": 2685 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"eval_EOP_f1": 0.7398610799910375, |
|
"eval_EOP_number": 2446, |
|
"eval_EOP_precision": 0.8185423896876549, |
|
"eval_EOP_recall": 0.6749795584627964, |
|
"eval_da_EOP_f1": 0.7520470291832878, |
|
"eval_da_EOP_number": 2715, |
|
"eval_da_EOP_precision": 0.87451171875, |
|
"eval_da_EOP_recall": 0.6596685082872928, |
|
"eval_da_overall_accuracy": 0.9452632554690397, |
|
"eval_da_overall_f1": 0.7520470291832878, |
|
"eval_da_overall_precision": 0.87451171875, |
|
"eval_da_overall_recall": 0.6596685082872928, |
|
"eval_loss": NaN, |
|
"eval_overall_accuracy": 0.9437145488922286, |
|
"eval_overall_f1": 0.7398610799910375, |
|
"eval_overall_precision": 0.8185423896876549, |
|
"eval_overall_recall": 0.6749795584627964, |
|
"eval_runtime": 369.4165, |
|
"eval_samples_per_second": 4.125, |
|
"eval_steps_per_second": 1.031, |
|
"step": 2685 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"learning_rate": 3e-05, |
|
"loss": 0.6831, |
|
"step": 5370 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"eval_EOP_f1": 0.7486033519553074, |
|
"eval_EOP_number": 2446, |
|
"eval_EOP_precision": 0.7889492753623188, |
|
"eval_EOP_recall": 0.7121831561733443, |
|
"eval_da_EOP_f1": 0.775566502463054, |
|
"eval_da_EOP_number": 2715, |
|
"eval_da_EOP_precision": 0.8338983050847457, |
|
"eval_da_EOP_recall": 0.7248618784530386, |
|
"eval_da_overall_accuracy": 0.9472098628105302, |
|
"eval_da_overall_f1": 0.775566502463054, |
|
"eval_da_overall_precision": 0.8338983050847457, |
|
"eval_da_overall_recall": 0.7248618784530386, |
|
"eval_loss": NaN, |
|
"eval_overall_accuracy": 0.9432782275658118, |
|
"eval_overall_f1": 0.7486033519553074, |
|
"eval_overall_precision": 0.7889492753623188, |
|
"eval_overall_recall": 0.7121831561733443, |
|
"eval_runtime": 368.9458, |
|
"eval_samples_per_second": 4.131, |
|
"eval_steps_per_second": 1.033, |
|
"step": 5370 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"learning_rate": 2e-05, |
|
"loss": 0.5066, |
|
"step": 8055 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"eval_EOP_f1": 0.7590260285474391, |
|
"eval_EOP_number": 2446, |
|
"eval_EOP_precision": 0.7799827437446074, |
|
"eval_EOP_recall": 0.7391659852820932, |
|
"eval_da_EOP_f1": 0.7927998434748582, |
|
"eval_da_EOP_number": 2715, |
|
"eval_da_EOP_precision": 0.845575959933222, |
|
"eval_da_EOP_recall": 0.7462246777163904, |
|
"eval_da_overall_accuracy": 0.9509176863181312, |
|
"eval_da_overall_f1": 0.7927998434748582, |
|
"eval_da_overall_precision": 0.845575959933222, |
|
"eval_da_overall_recall": 0.7462246777163904, |
|
"eval_loss": NaN, |
|
"eval_overall_accuracy": 0.944344790808164, |
|
"eval_overall_f1": 0.7590260285474391, |
|
"eval_overall_precision": 0.7799827437446074, |
|
"eval_overall_recall": 0.7391659852820932, |
|
"eval_runtime": 369.3175, |
|
"eval_samples_per_second": 4.127, |
|
"eval_steps_per_second": 1.032, |
|
"step": 8055 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"learning_rate": 1e-05, |
|
"loss": 0.3684, |
|
"step": 10740 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"eval_EOP_f1": 0.7617828961399019, |
|
"eval_EOP_number": 2446, |
|
"eval_EOP_precision": 0.7962550156041016, |
|
"eval_EOP_recall": 0.7301717089125103, |
|
"eval_da_EOP_f1": 0.7858523386161578, |
|
"eval_da_EOP_number": 2715, |
|
"eval_da_EOP_precision": 0.826758845058967, |
|
"eval_da_EOP_recall": 0.7488029465930018, |
|
"eval_da_overall_accuracy": 0.9486466444197256, |
|
"eval_da_overall_f1": 0.7858523386161578, |
|
"eval_da_overall_precision": 0.826758845058967, |
|
"eval_da_overall_recall": 0.7488029465930018, |
|
"eval_loss": NaN, |
|
"eval_overall_accuracy": 0.9458476753769332, |
|
"eval_overall_f1": 0.7617828961399019, |
|
"eval_overall_precision": 0.7962550156041016, |
|
"eval_overall_recall": 0.7301717089125103, |
|
"eval_runtime": 369.7052, |
|
"eval_samples_per_second": 4.122, |
|
"eval_steps_per_second": 1.031, |
|
"step": 10740 |
|
}, |
|
{ |
|
"epoch": 5.0, |
|
"learning_rate": 0.0, |
|
"loss": 0.2543, |
|
"step": 13425 |
|
}, |
|
{ |
|
"epoch": 5.0, |
|
"eval_EOP_f1": 0.7591393357008565, |
|
"eval_EOP_number": 2446, |
|
"eval_EOP_precision": 0.7761640324647586, |
|
"eval_EOP_recall": 0.7428454619787408, |
|
"eval_da_EOP_f1": 0.7873163186388245, |
|
"eval_da_EOP_number": 2715, |
|
"eval_da_EOP_precision": 0.8286528286528286, |
|
"eval_da_EOP_recall": 0.7499079189686925, |
|
"eval_da_overall_accuracy": 0.9490174267704857, |
|
"eval_da_overall_f1": 0.7873163186388245, |
|
"eval_da_overall_precision": 0.8286528286528286, |
|
"eval_da_overall_recall": 0.7499079189686925, |
|
"eval_loss": NaN, |
|
"eval_overall_accuracy": 0.9441023900712658, |
|
"eval_overall_f1": 0.7591393357008565, |
|
"eval_overall_precision": 0.7761640324647586, |
|
"eval_overall_recall": 0.7428454619787408, |
|
"eval_runtime": 370.5788, |
|
"eval_samples_per_second": 4.112, |
|
"eval_steps_per_second": 1.028, |
|
"step": 13425 |
|
}, |
|
{ |
|
"epoch": 5.0, |
|
"step": 13425, |
|
"total_flos": 1.000904429647872e+17, |
|
"train_loss": 0.5592155038844274, |
|
"train_runtime": 39662.9553, |
|
"train_samples_per_second": 1.354, |
|
"train_steps_per_second": 0.338 |
|
} |
|
], |
|
"max_steps": 13425, |
|
"num_train_epochs": 5, |
|
"total_flos": 1.000904429647872e+17, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|