|
{ |
|
"best_metric": 0.0, |
|
"best_model_checkpoint": "logs/eurlex/roberta-base/seed_1/checkpoint-215", |
|
"epoch": 2.0, |
|
"eval_steps": 500, |
|
"global_step": 430, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 1.0, |
|
"eval_loss": 0.17694051563739777, |
|
"eval_macro-f1": 0.0, |
|
"eval_micro-f1": 0.0, |
|
"eval_runtime": 3.944, |
|
"eval_samples_per_second": 1267.741, |
|
"eval_steps_per_second": 5.071, |
|
"step": 215 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"eval_loss": 0.17320719361305237, |
|
"eval_macro-f1": 0.0, |
|
"eval_micro-f1": 0.0, |
|
"eval_runtime": 3.964, |
|
"eval_samples_per_second": 1261.34, |
|
"eval_steps_per_second": 5.045, |
|
"step": 430 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"step": 430, |
|
"total_flos": 2.898875076784947e+16, |
|
"train_loss": 0.1999838718148165, |
|
"train_runtime": 196.4851, |
|
"train_samples_per_second": 559.839, |
|
"train_steps_per_second": 2.188 |
|
} |
|
], |
|
"logging_steps": 500, |
|
"max_steps": 430, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 2, |
|
"save_steps": 500, |
|
"stateful_callbacks": { |
|
"EarlyStoppingCallback": { |
|
"args": { |
|
"early_stopping_patience": 3, |
|
"early_stopping_threshold": 0.0 |
|
}, |
|
"attributes": { |
|
"early_stopping_patience_counter": 0 |
|
} |
|
}, |
|
"TrainerControl": { |
|
"args": { |
|
"should_epoch_stop": false, |
|
"should_evaluate": false, |
|
"should_log": false, |
|
"should_save": true, |
|
"should_training_stop": true |
|
}, |
|
"attributes": {} |
|
} |
|
}, |
|
"total_flos": 2.898875076784947e+16, |
|
"train_batch_size": 128, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|