|
{ |
|
"best_metric": 0.7399236317352567, |
|
"best_model_checkpoint": "./outputs/xlm-roberta-large-output/vua_metaphor_32_0.1_0.00003_07-19-22_17-18/checkpoint-903", |
|
"epoch": 3.998341625207297, |
|
"global_step": 1204, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 1.0, |
|
"learning_rate": 2.3952254641909816e-05, |
|
"loss": 0.2302, |
|
"step": 301 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"eval_accuracy": 0.9587389893370422, |
|
"eval_f1": 0.6235381935804927, |
|
"eval_loss": 0.1028517335653305, |
|
"eval_precision": 0.7607771706132362, |
|
"eval_recall": 0.5282462057335582, |
|
"eval_runtime": 10.4111, |
|
"eval_samples_per_second": 231.484, |
|
"eval_steps_per_second": 29.008, |
|
"step": 301 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"learning_rate": 1.596816976127321e-05, |
|
"loss": 0.0825, |
|
"step": 602 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"eval_accuracy": 0.9657749052333033, |
|
"eval_f1": 0.7058823529411765, |
|
"eval_loss": 0.08832709491252899, |
|
"eval_precision": 0.7947229551451187, |
|
"eval_recall": 0.6349072512647554, |
|
"eval_runtime": 10.4364, |
|
"eval_samples_per_second": 230.922, |
|
"eval_steps_per_second": 28.937, |
|
"step": 602 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"learning_rate": 7.984084880636604e-06, |
|
"loss": 0.053, |
|
"step": 903 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"eval_accuracy": 0.9665657639968366, |
|
"eval_f1": 0.7399236317352567, |
|
"eval_loss": 0.09772537648677826, |
|
"eval_precision": 0.7446626814688301, |
|
"eval_recall": 0.7352445193929174, |
|
"eval_runtime": 10.4481, |
|
"eval_samples_per_second": 230.664, |
|
"eval_steps_per_second": 28.905, |
|
"step": 903 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"learning_rate": 0.0, |
|
"loss": 0.0302, |
|
"step": 1204 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"eval_accuracy": 0.9673838937522158, |
|
"eval_f1": 0.7341040462427745, |
|
"eval_loss": 0.11126415431499481, |
|
"eval_precision": 0.7765757290686736, |
|
"eval_recall": 0.6960370994940978, |
|
"eval_runtime": 10.506, |
|
"eval_samples_per_second": 229.394, |
|
"eval_steps_per_second": 28.746, |
|
"step": 1204 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"step": 1204, |
|
"total_flos": 4407052490588292.0, |
|
"train_loss": 0.0989648171041495, |
|
"train_runtime": 907.5206, |
|
"train_samples_per_second": 42.459, |
|
"train_steps_per_second": 1.327 |
|
} |
|
], |
|
"max_steps": 1204, |
|
"num_train_epochs": 4, |
|
"total_flos": 4407052490588292.0, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|