| { |
| "best_metric": 0.7248, |
| "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0812/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": 119.63482666015625, |
| "learning_rate": 4.267766952966369e-05, |
| "loss": 3.7365, |
| "step": 333 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.42186666666666667, |
| "eval_loss": 3.406616687774658, |
| "eval_runtime": 13.4791, |
| "eval_samples_per_second": 278.208, |
| "eval_steps_per_second": 4.377, |
| "step": 333 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 107.66165924072266, |
| "learning_rate": 2.5e-05, |
| "loss": 2.2215, |
| "step": 666 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.6381333333333333, |
| "eval_loss": 1.4104437828063965, |
| "eval_runtime": 13.681, |
| "eval_samples_per_second": 274.102, |
| "eval_steps_per_second": 4.313, |
| "step": 666 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 140.68824768066406, |
| "learning_rate": 7.3223304703363135e-06, |
| "loss": 1.1909, |
| "step": 999 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.716, |
| "eval_loss": 1.050255537033081, |
| "eval_runtime": 13.4465, |
| "eval_samples_per_second": 278.883, |
| "eval_steps_per_second": 4.388, |
| "step": 999 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 69.42652130126953, |
| "learning_rate": 0.0, |
| "loss": 1.026, |
| "step": 1332 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.7248, |
| "eval_loss": 1.0181090831756592, |
| "eval_runtime": 13.5229, |
| "eval_samples_per_second": 277.307, |
| "eval_steps_per_second": 4.363, |
| "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 |
| } |
|
|
