|
{ |
|
"best_metric": 0.8552, |
|
"best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0855/checkpoints/checkpoint-1332", |
|
"epoch": 4.0, |
|
"eval_steps": 500, |
|
"global_step": 1332, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 1.0, |
|
"grad_norm": 139.37509155273438, |
|
"learning_rate": 7e-05, |
|
"loss": 3.3077, |
|
"step": 333 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"eval_accuracy": 0.6354666666666666, |
|
"eval_loss": 1.665514349937439, |
|
"eval_runtime": 7.5046, |
|
"eval_samples_per_second": 499.696, |
|
"eval_steps_per_second": 7.862, |
|
"step": 333 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"grad_norm": 602.5709228515625, |
|
"learning_rate": 7e-05, |
|
"loss": 1.0728, |
|
"step": 666 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"eval_accuracy": 0.8, |
|
"eval_loss": 0.6992009282112122, |
|
"eval_runtime": 7.6052, |
|
"eval_samples_per_second": 493.082, |
|
"eval_steps_per_second": 7.758, |
|
"step": 666 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"grad_norm": 46.2913932800293, |
|
"learning_rate": 7e-05, |
|
"loss": 0.6245, |
|
"step": 999 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"eval_accuracy": 0.8312, |
|
"eval_loss": 0.548198401927948, |
|
"eval_runtime": 7.4769, |
|
"eval_samples_per_second": 501.544, |
|
"eval_steps_per_second": 7.891, |
|
"step": 999 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"grad_norm": 102.23623657226562, |
|
"learning_rate": 7e-05, |
|
"loss": 0.4586, |
|
"step": 1332 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"eval_accuracy": 0.8552, |
|
"eval_loss": 0.46330058574676514, |
|
"eval_runtime": 11.0104, |
|
"eval_samples_per_second": 340.588, |
|
"eval_steps_per_second": 5.359, |
|
"step": 1332 |
|
} |
|
], |
|
"logging_steps": 500, |
|
"max_steps": 1332, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 4, |
|
"save_steps": 500, |
|
"stateful_callbacks": { |
|
"TrainerControl": { |
|
"args": { |
|
"should_epoch_stop": false, |
|
"should_evaluate": false, |
|
"should_log": false, |
|
"should_save": true, |
|
"should_training_stop": true |
|
}, |
|
"attributes": {} |
|
} |
|
}, |
|
"total_flos": 3.2705716958208e+18, |
|
"train_batch_size": 64, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|