| { |
| "best_metric": 0.8765333333333334, |
| "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0167/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": 98.63931274414062, |
| "learning_rate": 7e-05, |
| "loss": 3.2396, |
| "step": 333 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.6773333333333333, |
| "eval_loss": 1.6806457042694092, |
| "eval_runtime": 9.2351, |
| "eval_samples_per_second": 406.059, |
| "eval_steps_per_second": 6.389, |
| "step": 333 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 164.979248046875, |
| "learning_rate": 7e-05, |
| "loss": 0.9093, |
| "step": 666 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.8362666666666667, |
| "eval_loss": 0.5460890531539917, |
| "eval_runtime": 8.2673, |
| "eval_samples_per_second": 453.594, |
| "eval_steps_per_second": 7.137, |
| "step": 666 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 123.5877685546875, |
| "learning_rate": 7e-05, |
| "loss": 0.4802, |
| "step": 999 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.8706666666666667, |
| "eval_loss": 0.4301808476448059, |
| "eval_runtime": 8.0798, |
| "eval_samples_per_second": 464.122, |
| "eval_steps_per_second": 7.302, |
| "step": 999 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 106.66055297851562, |
| "learning_rate": 7e-05, |
| "loss": 0.3515, |
| "step": 1332 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.8765333333333334, |
| "eval_loss": 0.3952127993106842, |
| "eval_runtime": 8.1737, |
| "eval_samples_per_second": 458.786, |
| "eval_steps_per_second": 7.218, |
| "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 |
| } |
|
|
