| { |
| "best_metric": 0.9456, |
| "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0726/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": 0.9036150574684143, |
| "learning_rate": 3.7500000000000003e-05, |
| "loss": 0.9009, |
| "step": 333 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.928, |
| "eval_loss": 0.2804586887359619, |
| "eval_runtime": 11.5842, |
| "eval_samples_per_second": 323.716, |
| "eval_steps_per_second": 5.093, |
| "step": 333 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 0.21802140772342682, |
| "learning_rate": 2.5e-05, |
| "loss": 0.1433, |
| "step": 666 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.9397333333333333, |
| "eval_loss": 0.22649814188480377, |
| "eval_runtime": 11.7608, |
| "eval_samples_per_second": 318.856, |
| "eval_steps_per_second": 5.017, |
| "step": 666 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 0.07849038392305374, |
| "learning_rate": 1.25e-05, |
| "loss": 0.0561, |
| "step": 999 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.9352, |
| "eval_loss": 0.24769245088100433, |
| "eval_runtime": 12.8082, |
| "eval_samples_per_second": 292.78, |
| "eval_steps_per_second": 4.606, |
| "step": 999 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 0.047373317182064056, |
| "learning_rate": 0.0, |
| "loss": 0.0249, |
| "step": 1332 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.9456, |
| "eval_loss": 0.21177205443382263, |
| "eval_runtime": 12.066, |
| "eval_samples_per_second": 310.791, |
| "eval_steps_per_second": 4.89, |
| "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 |
| } |
|
|
