| { | |
| "best_metric": 0.8997333333333334, | |
| "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0979/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": 36.88692855834961, | |
| "learning_rate": 0.000225, | |
| "loss": 1.4991, | |
| "step": 333 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8357333333333333, | |
| "eval_loss": 0.5326321721076965, | |
| "eval_runtime": 7.4838, | |
| "eval_samples_per_second": 501.084, | |
| "eval_steps_per_second": 7.884, | |
| "step": 333 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 45.86236572265625, | |
| "learning_rate": 0.00015, | |
| "loss": 0.4179, | |
| "step": 666 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8770666666666667, | |
| "eval_loss": 0.3863641321659088, | |
| "eval_runtime": 7.2227, | |
| "eval_samples_per_second": 519.199, | |
| "eval_steps_per_second": 8.169, | |
| "step": 666 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 45.01603317260742, | |
| "learning_rate": 7.5e-05, | |
| "loss": 0.2378, | |
| "step": 999 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.896, | |
| "eval_loss": 0.3457378149032593, | |
| "eval_runtime": 7.1823, | |
| "eval_samples_per_second": 522.118, | |
| "eval_steps_per_second": 8.215, | |
| "step": 999 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 72.08895111083984, | |
| "learning_rate": 0.0, | |
| "loss": 0.1492, | |
| "step": 1332 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8997333333333334, | |
| "eval_loss": 0.33335596323013306, | |
| "eval_runtime": 7.1659, | |
| "eval_samples_per_second": 523.311, | |
| "eval_steps_per_second": 8.233, | |
| "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 | |
| } | |