| { |
| "best_metric": 0.8941333333333333, |
| "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0338/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": 117.53186798095703, |
| "learning_rate": 9e-05, |
| "loss": 2.7909, |
| "step": 333 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.7666666666666667, |
| "eval_loss": 0.9297710657119751, |
| "eval_runtime": 4.923, |
| "eval_samples_per_second": 761.725, |
| "eval_steps_per_second": 11.984, |
| "step": 333 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 90.52135467529297, |
| "learning_rate": 9e-05, |
| "loss": 0.6609, |
| "step": 666 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.8629333333333333, |
| "eval_loss": 0.47415679693222046, |
| "eval_runtime": 4.8903, |
| "eval_samples_per_second": 766.827, |
| "eval_steps_per_second": 12.065, |
| "step": 666 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 121.64726257324219, |
| "learning_rate": 9e-05, |
| "loss": 0.3619, |
| "step": 999 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.8730666666666667, |
| "eval_loss": 0.4215207099914551, |
| "eval_runtime": 5.3895, |
| "eval_samples_per_second": 695.796, |
| "eval_steps_per_second": 10.947, |
| "step": 999 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 57.37247848510742, |
| "learning_rate": 9e-05, |
| "loss": 0.2325, |
| "step": 1332 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.8941333333333333, |
| "eval_loss": 0.37584030628204346, |
| "eval_runtime": 4.8949, |
| "eval_samples_per_second": 766.105, |
| "eval_steps_per_second": 12.053, |
| "step": 1332 |
| } |
| ], |
| "logging_steps": 500, |
| "max_steps": 1665, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 5, |
| "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": 3.2705716958208e+18, |
| "train_batch_size": 64, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
