| { |
| "best_metric": 0.8957333333333334, |
| "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0419/checkpoints/checkpoint-2997", |
| "epoch": 9.0, |
| "eval_steps": 500, |
| "global_step": 2997, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 1.0, |
| "grad_norm": 61.79206848144531, |
| "learning_rate": 9.698463103929542e-05, |
| "loss": 2.6757, |
| "step": 333 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.7634666666666666, |
| "eval_loss": 0.8775083422660828, |
| "eval_runtime": 13.5568, |
| "eval_samples_per_second": 276.614, |
| "eval_steps_per_second": 4.352, |
| "step": 333 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 50.82188415527344, |
| "learning_rate": 8.83022221559489e-05, |
| "loss": 0.6862, |
| "step": 666 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.8365333333333334, |
| "eval_loss": 0.518424928188324, |
| "eval_runtime": 13.7673, |
| "eval_samples_per_second": 272.385, |
| "eval_steps_per_second": 4.286, |
| "step": 666 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 96.98792266845703, |
| "learning_rate": 7.500000000000001e-05, |
| "loss": 0.4184, |
| "step": 999 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.864, |
| "eval_loss": 0.41626501083374023, |
| "eval_runtime": 13.6257, |
| "eval_samples_per_second": 275.215, |
| "eval_steps_per_second": 4.33, |
| "step": 999 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 125.59623718261719, |
| "learning_rate": 5.868240888334653e-05, |
| "loss": 0.3111, |
| "step": 1332 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.8738666666666667, |
| "eval_loss": 0.3971407115459442, |
| "eval_runtime": 13.6268, |
| "eval_samples_per_second": 275.194, |
| "eval_steps_per_second": 4.33, |
| "step": 1332 |
| }, |
| { |
| "epoch": 5.0, |
| "grad_norm": 105.28047180175781, |
| "learning_rate": 4.131759111665349e-05, |
| "loss": 0.2325, |
| "step": 1665 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_accuracy": 0.8874666666666666, |
| "eval_loss": 0.35857436060905457, |
| "eval_runtime": 13.5309, |
| "eval_samples_per_second": 277.143, |
| "eval_steps_per_second": 4.36, |
| "step": 1665 |
| }, |
| { |
| "epoch": 6.0, |
| "grad_norm": 90.8493881225586, |
| "learning_rate": 2.500000000000001e-05, |
| "loss": 0.1844, |
| "step": 1998 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_accuracy": 0.8821333333333333, |
| "eval_loss": 0.3855854868888855, |
| "eval_runtime": 13.7621, |
| "eval_samples_per_second": 272.487, |
| "eval_steps_per_second": 4.287, |
| "step": 1998 |
| }, |
| { |
| "epoch": 7.0, |
| "grad_norm": 121.79293823242188, |
| "learning_rate": 1.1697777844051105e-05, |
| "loss": 0.156, |
| "step": 2331 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_accuracy": 0.8856, |
| "eval_loss": 0.35955196619033813, |
| "eval_runtime": 13.4962, |
| "eval_samples_per_second": 277.856, |
| "eval_steps_per_second": 4.372, |
| "step": 2331 |
| }, |
| { |
| "epoch": 8.0, |
| "grad_norm": 124.54039001464844, |
| "learning_rate": 3.0153689607045845e-06, |
| "loss": 0.1431, |
| "step": 2664 |
| }, |
| { |
| "epoch": 8.0, |
| "eval_accuracy": 0.8909333333333334, |
| "eval_loss": 0.3461163640022278, |
| "eval_runtime": 13.478, |
| "eval_samples_per_second": 278.231, |
| "eval_steps_per_second": 4.377, |
| "step": 2664 |
| }, |
| { |
| "epoch": 9.0, |
| "grad_norm": 100.49520111083984, |
| "learning_rate": 0.0, |
| "loss": 0.1326, |
| "step": 2997 |
| }, |
| { |
| "epoch": 9.0, |
| "eval_accuracy": 0.8957333333333334, |
| "eval_loss": 0.3540811240673065, |
| "eval_runtime": 13.491, |
| "eval_samples_per_second": 277.963, |
| "eval_steps_per_second": 4.373, |
| "step": 2997 |
| } |
| ], |
| "logging_steps": 500, |
| "max_steps": 2997, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 9, |
| "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": 7.3587863155968e+18, |
| "train_batch_size": 64, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
