|
{ |
|
"best_metric": 0.09438519924879074, |
|
"best_model_checkpoint": "./SonicCharacterClassifier\\checkpoint-1400", |
|
"epoch": 4.093567251461988, |
|
"global_step": 1400, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.29, |
|
"learning_rate": 4.853801169590643e-05, |
|
"loss": 0.3188, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 0.58, |
|
"learning_rate": 4.707602339181287e-05, |
|
"loss": 0.2087, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 0.58, |
|
"eval_accuracy": 0.9407407407407361, |
|
"eval_loss": 0.16877754032611847, |
|
"eval_precision": 0.6539702845075653, |
|
"eval_recall": 0.5742313120602596, |
|
"eval_runtime": 161.9664, |
|
"eval_samples_per_second": 8.446, |
|
"eval_steps_per_second": 1.056, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 0.88, |
|
"learning_rate": 4.56140350877193e-05, |
|
"loss": 0.1663, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 1.17, |
|
"learning_rate": 4.4152046783625734e-05, |
|
"loss": 0.1155, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 1.17, |
|
"eval_accuracy": 0.9615009746588669, |
|
"eval_loss": 0.12015020102262497, |
|
"eval_precision": 0.8305708768976313, |
|
"eval_recall": 0.7801112001221648, |
|
"eval_runtime": 161.8616, |
|
"eval_samples_per_second": 8.452, |
|
"eval_steps_per_second": 1.056, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 1.46, |
|
"learning_rate": 4.269005847953216e-05, |
|
"loss": 0.095, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 1.75, |
|
"learning_rate": 4.12280701754386e-05, |
|
"loss": 0.0932, |
|
"step": 600 |
|
}, |
|
{ |
|
"epoch": 1.75, |
|
"eval_accuracy": 0.964278752436644, |
|
"eval_loss": 0.1112869530916214, |
|
"eval_precision": 0.8358102748453625, |
|
"eval_recall": 0.7573138768314205, |
|
"eval_runtime": 160.6953, |
|
"eval_samples_per_second": 8.513, |
|
"eval_steps_per_second": 1.064, |
|
"step": 600 |
|
}, |
|
{ |
|
"epoch": 2.05, |
|
"learning_rate": 3.976608187134503e-05, |
|
"loss": 0.087, |
|
"step": 700 |
|
}, |
|
{ |
|
"epoch": 2.34, |
|
"learning_rate": 3.8304093567251465e-05, |
|
"loss": 0.0489, |
|
"step": 800 |
|
}, |
|
{ |
|
"epoch": 2.34, |
|
"eval_accuracy": 0.9684697855750467, |
|
"eval_loss": 0.09839322417974472, |
|
"eval_precision": 0.877861029889539, |
|
"eval_recall": 0.8274951210258222, |
|
"eval_runtime": 160.5107, |
|
"eval_samples_per_second": 8.523, |
|
"eval_steps_per_second": 1.065, |
|
"step": 800 |
|
}, |
|
{ |
|
"epoch": 2.63, |
|
"learning_rate": 3.6842105263157895e-05, |
|
"loss": 0.0506, |
|
"step": 900 |
|
}, |
|
{ |
|
"epoch": 2.92, |
|
"learning_rate": 3.538011695906433e-05, |
|
"loss": 0.0503, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 2.92, |
|
"eval_accuracy": 0.9692495126705621, |
|
"eval_loss": 0.09646110981702805, |
|
"eval_precision": 0.8755529497305813, |
|
"eval_recall": 0.8374415545468176, |
|
"eval_runtime": 161.798, |
|
"eval_samples_per_second": 8.455, |
|
"eval_steps_per_second": 1.057, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 3.22, |
|
"learning_rate": 3.391812865497076e-05, |
|
"loss": 0.0297, |
|
"step": 1100 |
|
}, |
|
{ |
|
"epoch": 3.51, |
|
"learning_rate": 3.24561403508772e-05, |
|
"loss": 0.0245, |
|
"step": 1200 |
|
}, |
|
{ |
|
"epoch": 3.51, |
|
"eval_accuracy": 0.9697855750487306, |
|
"eval_loss": 0.09691683202981949, |
|
"eval_precision": 0.8796307635671673, |
|
"eval_recall": 0.8392919296428067, |
|
"eval_runtime": 163.6772, |
|
"eval_samples_per_second": 8.358, |
|
"eval_steps_per_second": 1.045, |
|
"step": 1200 |
|
}, |
|
{ |
|
"epoch": 3.8, |
|
"learning_rate": 3.0994152046783626e-05, |
|
"loss": 0.026, |
|
"step": 1300 |
|
}, |
|
{ |
|
"epoch": 4.09, |
|
"learning_rate": 2.9532163742690062e-05, |
|
"loss": 0.0231, |
|
"step": 1400 |
|
}, |
|
{ |
|
"epoch": 4.09, |
|
"eval_accuracy": 0.9713937621832336, |
|
"eval_loss": 0.09438519924879074, |
|
"eval_precision": 0.8850769600769599, |
|
"eval_recall": 0.8511938435184049, |
|
"eval_runtime": 161.9688, |
|
"eval_samples_per_second": 8.446, |
|
"eval_steps_per_second": 1.056, |
|
"step": 1400 |
|
} |
|
], |
|
"max_steps": 3420, |
|
"num_train_epochs": 10, |
|
"total_flos": 5.207113357118669e+18, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|