| { |
| "best_metric": 0.9381333333333334, |
| "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0724/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": 3.6546056270599365, |
| "learning_rate": 7e-05, |
| "loss": 0.6816, |
| "step": 333 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.924, |
| "eval_loss": 0.2558077573776245, |
| "eval_runtime": 11.9466, |
| "eval_samples_per_second": 313.896, |
| "eval_steps_per_second": 4.939, |
| "step": 333 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 0.10156994313001633, |
| "learning_rate": 7e-05, |
| "loss": 0.125, |
| "step": 666 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.9277333333333333, |
| "eval_loss": 0.23121190071105957, |
| "eval_runtime": 11.8494, |
| "eval_samples_per_second": 316.473, |
| "eval_steps_per_second": 4.979, |
| "step": 666 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 2.0732197761535645, |
| "learning_rate": 7e-05, |
| "loss": 0.0554, |
| "step": 999 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.928, |
| "eval_loss": 0.28154876828193665, |
| "eval_runtime": 11.7714, |
| "eval_samples_per_second": 318.568, |
| "eval_steps_per_second": 5.012, |
| "step": 999 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 0.03272829204797745, |
| "learning_rate": 7e-05, |
| "loss": 0.0324, |
| "step": 1332 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.9381333333333334, |
| "eval_loss": 0.25631949305534363, |
| "eval_runtime": 11.9523, |
| "eval_samples_per_second": 313.747, |
| "eval_steps_per_second": 4.936, |
| "step": 1332 |
| } |
| ], |
| "logging_steps": 500, |
| "max_steps": 2664, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 8, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 6.58965282490368e+18, |
| "train_batch_size": 64, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
