| { | |
| "best_metric": 0.9592, | |
| "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0608/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": 0.35602203011512756, | |
| "learning_rate": 5.974873734152915e-05, | |
| "loss": 0.6965, | |
| "step": 333 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9349333333333333, | |
| "eval_loss": 0.24402333796024323, | |
| "eval_runtime": 11.1849, | |
| "eval_samples_per_second": 335.275, | |
| "eval_steps_per_second": 5.275, | |
| "step": 333 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.06505334377288818, | |
| "learning_rate": 3.508254927997938e-05, | |
| "loss": 0.106, | |
| "step": 666 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9445333333333333, | |
| "eval_loss": 0.19497846066951752, | |
| "eval_runtime": 11.1506, | |
| "eval_samples_per_second": 336.305, | |
| "eval_steps_per_second": 5.291, | |
| "step": 666 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 0.27592140436172485, | |
| "learning_rate": 1.0309702649891636e-05, | |
| "loss": 0.0358, | |
| "step": 999 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9546666666666667, | |
| "eval_loss": 0.16953779757022858, | |
| "eval_runtime": 11.0175, | |
| "eval_samples_per_second": 340.368, | |
| "eval_steps_per_second": 5.355, | |
| "step": 999 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 26.59088706970215, | |
| "learning_rate": 9.734847288167535e-11, | |
| "loss": 0.0161, | |
| "step": 1332 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9592, | |
| "eval_loss": 0.15777722001075745, | |
| "eval_runtime": 11.1927, | |
| "eval_samples_per_second": 335.038, | |
| "eval_steps_per_second": 5.271, | |
| "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 | |
| } | |