| { | |
| "best_metric": 0.916, | |
| "best_model_checkpoint": "fine-tune-vit-L-cifar100\\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": 1.7363, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.82, | |
| "eval_loss": 0.6244344115257263, | |
| "eval_runtime": 54.6164, | |
| "eval_samples_per_second": 91.548, | |
| "eval_steps_per_second": 1.446, | |
| "step": 704 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 0.0001431818181818182, | |
| "loss": 1.1337, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8684, | |
| "eval_loss": 0.4586382806301117, | |
| "eval_runtime": 54.6154, | |
| "eval_samples_per_second": 91.549, | |
| "eval_steps_per_second": 1.446, | |
| "step": 1408 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "learning_rate": 0.00011477272727272728, | |
| "loss": 0.9411, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "learning_rate": 8.636363636363637e-05, | |
| "loss": 0.752, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8864, | |
| "eval_loss": 0.3812929093837738, | |
| "eval_runtime": 55.4776, | |
| "eval_samples_per_second": 90.126, | |
| "eval_steps_per_second": 1.424, | |
| "step": 2112 | |
| }, | |
| { | |
| "epoch": 3.55, | |
| "learning_rate": 5.7954545454545464e-05, | |
| "loss": 0.6087, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.907, | |
| "eval_loss": 0.3266417384147644, | |
| "eval_runtime": 55.3761, | |
| "eval_samples_per_second": 90.292, | |
| "eval_steps_per_second": 1.427, | |
| "step": 2816 | |
| }, | |
| { | |
| "epoch": 4.26, | |
| "learning_rate": 2.954545454545455e-05, | |
| "loss": 0.5215, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 4.97, | |
| "learning_rate": 1.1363636363636364e-06, | |
| "loss": 0.4268, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.916, | |
| "eval_loss": 0.2967301905155182, | |
| "eval_runtime": 55.5314, | |
| "eval_samples_per_second": 90.039, | |
| "eval_steps_per_second": 1.423, | |
| "step": 3520 | |
| } | |
| ], | |
| "max_steps": 3520, | |
| "num_train_epochs": 5, | |
| "total_flos": 6.16556031952896e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |