| { | |
| "best_metric": 0.9061333333333333, | |
| "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0193/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": 58.268611907958984, | |
| "learning_rate": 0.0004, | |
| "loss": 1.375, | |
| "step": 333 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8389333333333333, | |
| "eval_loss": 0.541002094745636, | |
| "eval_runtime": 7.692, | |
| "eval_samples_per_second": 487.521, | |
| "eval_steps_per_second": 7.67, | |
| "step": 333 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 60.11704635620117, | |
| "learning_rate": 0.0003, | |
| "loss": 0.4221, | |
| "step": 666 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8832, | |
| "eval_loss": 0.38036009669303894, | |
| "eval_runtime": 7.7035, | |
| "eval_samples_per_second": 486.792, | |
| "eval_steps_per_second": 7.659, | |
| "step": 666 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 33.80007553100586, | |
| "learning_rate": 0.0002, | |
| "loss": 0.2599, | |
| "step": 999 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9002666666666667, | |
| "eval_loss": 0.34382739663124084, | |
| "eval_runtime": 7.6927, | |
| "eval_samples_per_second": 487.472, | |
| "eval_steps_per_second": 7.67, | |
| "step": 999 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 29.045394897460938, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1552, | |
| "step": 1332 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9061333333333333, | |
| "eval_loss": 0.32741138339042664, | |
| "eval_runtime": 7.7249, | |
| "eval_samples_per_second": 485.44, | |
| "eval_steps_per_second": 7.638, | |
| "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 | |
| } | |