| { | |
| "best_metric": 0.944, | |
| "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0866/checkpoints/checkpoint-999", | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 999, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 7.2543864250183105, | |
| "learning_rate": 0.000375, | |
| "loss": 0.712, | |
| "step": 333 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8562666666666666, | |
| "eval_loss": 0.489900141954422, | |
| "eval_runtime": 11.5625, | |
| "eval_samples_per_second": 324.324, | |
| "eval_steps_per_second": 5.103, | |
| "step": 333 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.03380799666047096, | |
| "learning_rate": 0.00012500000000000006, | |
| "loss": 0.2279, | |
| "step": 666 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9301333333333334, | |
| "eval_loss": 0.23566731810569763, | |
| "eval_runtime": 12.1121, | |
| "eval_samples_per_second": 309.607, | |
| "eval_steps_per_second": 4.871, | |
| "step": 666 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 0.03999759256839752, | |
| "learning_rate": 0.0, | |
| "loss": 0.038, | |
| "step": 999 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.944, | |
| "eval_loss": 0.20463323593139648, | |
| "eval_runtime": 12.5318, | |
| "eval_samples_per_second": 299.238, | |
| "eval_steps_per_second": 4.708, | |
| "step": 999 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 999, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "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": 4.94223961867776e+18, | |
| "train_batch_size": 64, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |