| { | |
| "best_metric": 0.8856, | |
| "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0299/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": 1.8594378232955933, | |
| "learning_rate": 5.974873734152915e-05, | |
| "loss": 1.6075, | |
| "step": 333 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7949333333333334, | |
| "eval_loss": 0.7638483047485352, | |
| "eval_runtime": 12.3227, | |
| "eval_samples_per_second": 304.315, | |
| "eval_steps_per_second": 4.788, | |
| "step": 333 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 2.857916831970215, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.511, | |
| "step": 666 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8538666666666667, | |
| "eval_loss": 0.5260677337646484, | |
| "eval_runtime": 12.461, | |
| "eval_samples_per_second": 300.938, | |
| "eval_steps_per_second": 4.735, | |
| "step": 666 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 0.861556887626648, | |
| "learning_rate": 1.0251262658470837e-05, | |
| "loss": 0.2556, | |
| "step": 999 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8802666666666666, | |
| "eval_loss": 0.42199862003326416, | |
| "eval_runtime": 12.5036, | |
| "eval_samples_per_second": 299.913, | |
| "eval_steps_per_second": 4.719, | |
| "step": 999 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 8.748467445373535, | |
| "learning_rate": 0.0, | |
| "loss": 0.1355, | |
| "step": 1332 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8856, | |
| "eval_loss": 0.40191107988357544, | |
| "eval_runtime": 12.3985, | |
| "eval_samples_per_second": 302.456, | |
| "eval_steps_per_second": 4.759, | |
| "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 | |
| } | |