| { | |
| "best_metric": 0.9288, | |
| "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0241/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": 0.1887102872133255, | |
| "learning_rate": 0.0001, | |
| "loss": 0.5695, | |
| "step": 333 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9253333333333333, | |
| "eval_loss": 0.25713443756103516, | |
| "eval_runtime": 5.2425, | |
| "eval_samples_per_second": 715.314, | |
| "eval_steps_per_second": 11.254, | |
| "step": 333 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.24801909923553467, | |
| "learning_rate": 0.0001, | |
| "loss": 0.12, | |
| "step": 666 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9237333333333333, | |
| "eval_loss": 0.26414379477500916, | |
| "eval_runtime": 5.3062, | |
| "eval_samples_per_second": 706.717, | |
| "eval_steps_per_second": 11.119, | |
| "step": 666 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 0.05407159775495529, | |
| "learning_rate": 0.0001, | |
| "loss": 0.07, | |
| "step": 999 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9224, | |
| "eval_loss": 0.28587374091148376, | |
| "eval_runtime": 4.1556, | |
| "eval_samples_per_second": 902.391, | |
| "eval_steps_per_second": 14.198, | |
| "step": 999 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 2.3713555335998535, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0471, | |
| "step": 1332 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9288, | |
| "eval_loss": 0.26511046290397644, | |
| "eval_runtime": 4.0327, | |
| "eval_samples_per_second": 929.906, | |
| "eval_steps_per_second": 14.631, | |
| "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": 6.58965282490368e+18, | |
| "train_batch_size": 64, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |