| { |
| "best_metric": 0.9050666666666667, |
| "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0736/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": 63.40135955810547, |
| "learning_rate": 0.000375, |
| "loss": 1.3158, |
| "step": 333 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.8378666666666666, |
| "eval_loss": 0.546513557434082, |
| "eval_runtime": 13.2997, |
| "eval_samples_per_second": 281.962, |
| "eval_steps_per_second": 4.436, |
| "step": 333 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 28.087303161621094, |
| "learning_rate": 0.00025, |
| "loss": 0.3818, |
| "step": 666 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.88, |
| "eval_loss": 0.41786694526672363, |
| "eval_runtime": 13.0527, |
| "eval_samples_per_second": 287.297, |
| "eval_steps_per_second": 4.52, |
| "step": 666 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 33.22799301147461, |
| "learning_rate": 0.000125, |
| "loss": 0.2073, |
| "step": 999 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.8936, |
| "eval_loss": 0.36491554975509644, |
| "eval_runtime": 13.8221, |
| "eval_samples_per_second": 271.305, |
| "eval_steps_per_second": 4.269, |
| "step": 999 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 32.74684524536133, |
| "learning_rate": 0.0, |
| "loss": 0.1165, |
| "step": 1332 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.9050666666666667, |
| "eval_loss": 0.34091874957084656, |
| "eval_runtime": 13.3715, |
| "eval_samples_per_second": 280.448, |
| "eval_steps_per_second": 4.412, |
| "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 |
| } |
|
|
