|
{ |
|
"best_metric": 0.5861096817040292, |
|
"best_model_checkpoint": "models/spanish-offensive-language-bert-base-spanish-wwm-cased/checkpoint-664", |
|
"epoch": 5.0, |
|
"global_step": 830, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 1.0, |
|
"learning_rate": 1.6000000000000003e-05, |
|
"loss": 0.6383, |
|
"step": 166 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"eval_f1": 0.49156525826500225, |
|
"eval_loss": 0.6105424165725708, |
|
"eval_runtime": 4.1796, |
|
"eval_samples_per_second": 211.743, |
|
"eval_steps_per_second": 4.546, |
|
"step": 166 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"learning_rate": 1.2e-05, |
|
"loss": 0.5779, |
|
"step": 332 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"eval_f1": 0.5531108537740815, |
|
"eval_loss": 0.6026707887649536, |
|
"eval_runtime": 4.3621, |
|
"eval_samples_per_second": 202.885, |
|
"eval_steps_per_second": 4.356, |
|
"step": 332 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"learning_rate": 8.000000000000001e-06, |
|
"loss": 0.4791, |
|
"step": 498 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"eval_f1": 0.5851959878843573, |
|
"eval_loss": 0.6503026485443115, |
|
"eval_runtime": 4.2153, |
|
"eval_samples_per_second": 209.95, |
|
"eval_steps_per_second": 4.507, |
|
"step": 498 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"learning_rate": 4.000000000000001e-06, |
|
"loss": 0.3626, |
|
"step": 664 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"eval_f1": 0.5861096817040292, |
|
"eval_loss": 0.7753247022628784, |
|
"eval_runtime": 4.374, |
|
"eval_samples_per_second": 202.334, |
|
"eval_steps_per_second": 4.344, |
|
"step": 664 |
|
}, |
|
{ |
|
"epoch": 5.0, |
|
"learning_rate": 0.0, |
|
"loss": 0.2758, |
|
"step": 830 |
|
}, |
|
{ |
|
"epoch": 5.0, |
|
"eval_f1": 0.5855743933442549, |
|
"eval_loss": 0.8416333794593811, |
|
"eval_runtime": 4.2197, |
|
"eval_samples_per_second": 209.729, |
|
"eval_steps_per_second": 4.503, |
|
"step": 830 |
|
}, |
|
{ |
|
"epoch": 5.0, |
|
"step": 830, |
|
"total_flos": 1.04770822244352e+16, |
|
"train_loss": 0.4667576571544969, |
|
"train_runtime": 491.7868, |
|
"train_samples_per_second": 80.97, |
|
"train_steps_per_second": 1.688 |
|
} |
|
], |
|
"max_steps": 830, |
|
"num_train_epochs": 5, |
|
"total_flos": 1.04770822244352e+16, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|