| { |
| "best_metric": 0.8861333333333333, |
| "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0671/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": 77.22740936279297, |
| "learning_rate": 0.0001, |
| "loss": 2.8448, |
| "step": 333 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.7386666666666667, |
| "eval_loss": 0.9814813733100891, |
| "eval_runtime": 9.0161, |
| "eval_samples_per_second": 415.923, |
| "eval_steps_per_second": 6.544, |
| "step": 333 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 71.84850311279297, |
| "learning_rate": 0.0001, |
| "loss": 0.7377, |
| "step": 666 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.8506666666666667, |
| "eval_loss": 0.5300161838531494, |
| "eval_runtime": 8.9457, |
| "eval_samples_per_second": 419.195, |
| "eval_steps_per_second": 6.595, |
| "step": 666 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 137.815185546875, |
| "learning_rate": 0.0001, |
| "loss": 0.4794, |
| "step": 999 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.8789333333333333, |
| "eval_loss": 0.42663320899009705, |
| "eval_runtime": 10.3803, |
| "eval_samples_per_second": 361.261, |
| "eval_steps_per_second": 5.684, |
| "step": 999 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 70.58307647705078, |
| "learning_rate": 0.0001, |
| "loss": 0.3538, |
| "step": 1332 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.8861333333333333, |
| "eval_loss": 0.40160882472991943, |
| "eval_runtime": 8.9309, |
| "eval_samples_per_second": 419.889, |
| "eval_steps_per_second": 6.606, |
| "step": 1332 |
| } |
| ], |
| "logging_steps": 500, |
| "max_steps": 1665, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 5, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 3.2705716958208e+18, |
| "train_batch_size": 64, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
