| { |
| "best_metric": 0.9818, |
| "best_model_checkpoint": "fine-tune-Deit-cifar10\\checkpoint-3520", |
| "epoch": 5.0, |
| "global_step": 3520, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.71, |
| "learning_rate": 0.00017159090909090908, |
| "loss": 0.6047, |
| "step": 500 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.9498, |
| "eval_loss": 0.1472097635269165, |
| "eval_runtime": 26.7701, |
| "eval_samples_per_second": 186.776, |
| "eval_steps_per_second": 2.951, |
| "step": 704 |
| }, |
| { |
| "epoch": 1.42, |
| "learning_rate": 0.0001431818181818182, |
| "loss": 0.4259, |
| "step": 1000 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.9574, |
| "eval_loss": 0.1324978768825531, |
| "eval_runtime": 26.5889, |
| "eval_samples_per_second": 188.048, |
| "eval_steps_per_second": 2.971, |
| "step": 1408 |
| }, |
| { |
| "epoch": 2.13, |
| "learning_rate": 0.00011477272727272728, |
| "loss": 0.3563, |
| "step": 1500 |
| }, |
| { |
| "epoch": 2.84, |
| "learning_rate": 8.636363636363637e-05, |
| "loss": 0.2994, |
| "step": 2000 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.969, |
| "eval_loss": 0.08403348177671432, |
| "eval_runtime": 26.6795, |
| "eval_samples_per_second": 187.41, |
| "eval_steps_per_second": 2.961, |
| "step": 2112 |
| }, |
| { |
| "epoch": 3.55, |
| "learning_rate": 5.7954545454545464e-05, |
| "loss": 0.2518, |
| "step": 2500 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.9758, |
| "eval_loss": 0.07257051765918732, |
| "eval_runtime": 25.7546, |
| "eval_samples_per_second": 194.14, |
| "eval_steps_per_second": 3.067, |
| "step": 2816 |
| }, |
| { |
| "epoch": 4.26, |
| "learning_rate": 2.954545454545455e-05, |
| "loss": 0.2112, |
| "step": 3000 |
| }, |
| { |
| "epoch": 4.97, |
| "learning_rate": 1.1363636363636364e-06, |
| "loss": 0.1888, |
| "step": 3500 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_accuracy": 0.9818, |
| "eval_loss": 0.05922615900635719, |
| "eval_runtime": 25.7526, |
| "eval_samples_per_second": 194.155, |
| "eval_steps_per_second": 3.068, |
| "step": 3520 |
| } |
| ], |
| "max_steps": 3520, |
| "num_train_epochs": 5, |
| "total_flos": 1.74369478284288e+19, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
