| { | |
| "best_metric": 0.9106666666666666, | |
| "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0905/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": 65.46792602539062, | |
| "learning_rate": 0.000375, | |
| "loss": 1.2197, | |
| "step": 333 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8584, | |
| "eval_loss": 0.45993301272392273, | |
| "eval_runtime": 13.8986, | |
| "eval_samples_per_second": 269.812, | |
| "eval_steps_per_second": 4.245, | |
| "step": 333 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 91.39410400390625, | |
| "learning_rate": 0.00025, | |
| "loss": 0.3496, | |
| "step": 666 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8938666666666667, | |
| "eval_loss": 0.3304373323917389, | |
| "eval_runtime": 13.2781, | |
| "eval_samples_per_second": 282.421, | |
| "eval_steps_per_second": 4.443, | |
| "step": 666 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 34.10322952270508, | |
| "learning_rate": 0.000125, | |
| "loss": 0.1688, | |
| "step": 999 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9037333333333334, | |
| "eval_loss": 0.32795777916908264, | |
| "eval_runtime": 13.1653, | |
| "eval_samples_per_second": 284.841, | |
| "eval_steps_per_second": 4.481, | |
| "step": 999 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 34.00672912597656, | |
| "learning_rate": 0.0, | |
| "loss": 0.073, | |
| "step": 1332 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9106666666666666, | |
| "eval_loss": 0.319932758808136, | |
| "eval_runtime": 13.1898, | |
| "eval_samples_per_second": 284.31, | |
| "eval_steps_per_second": 4.473, | |
| "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 | |
| } | |