| { |
| "best_metric": 0.8256, |
| "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0348/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": 206.5066375732422, |
| "learning_rate": 5.974873734152915e-05, |
| "loss": 3.3459, |
| "step": 333 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.6317333333333334, |
| "eval_loss": 1.923073649406433, |
| "eval_runtime": 8.8596, |
| "eval_samples_per_second": 423.268, |
| "eval_steps_per_second": 6.659, |
| "step": 333 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 84.41645050048828, |
| "learning_rate": 3.5e-05, |
| "loss": 1.0683, |
| "step": 666 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.8045333333333333, |
| "eval_loss": 0.6832934021949768, |
| "eval_runtime": 8.8929, |
| "eval_samples_per_second": 421.685, |
| "eval_steps_per_second": 6.635, |
| "step": 666 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 95.4558334350586, |
| "learning_rate": 1.0251262658470837e-05, |
| "loss": 0.629, |
| "step": 999 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.8202666666666667, |
| "eval_loss": 0.6014976501464844, |
| "eval_runtime": 9.0168, |
| "eval_samples_per_second": 415.889, |
| "eval_steps_per_second": 6.543, |
| "step": 999 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 162.36338806152344, |
| "learning_rate": 0.0, |
| "loss": 0.5493, |
| "step": 1332 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.8256, |
| "eval_loss": 0.5849276185035706, |
| "eval_runtime": 8.8535, |
| "eval_samples_per_second": 423.561, |
| "eval_steps_per_second": 6.664, |
| "step": 1332 |
| } |
| ], |
| "logging_steps": 500, |
| "max_steps": 1332, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 4, |
| "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": 3.2705716958208e+18, |
| "train_batch_size": 64, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
