| { |
| "best_metric": 0.9104, |
| "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0442/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": 34.87226486206055, |
| "learning_rate": 0.000225, |
| "loss": 1.499, |
| "step": 333 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.8504, |
| "eval_loss": 0.48533451557159424, |
| "eval_runtime": 14.8295, |
| "eval_samples_per_second": 252.874, |
| "eval_steps_per_second": 3.979, |
| "step": 333 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 49.05800247192383, |
| "learning_rate": 0.00015, |
| "loss": 0.3879, |
| "step": 666 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.8762666666666666, |
| "eval_loss": 0.3949331045150757, |
| "eval_runtime": 14.8438, |
| "eval_samples_per_second": 252.63, |
| "eval_steps_per_second": 3.975, |
| "step": 666 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 47.60361862182617, |
| "learning_rate": 7.5e-05, |
| "loss": 0.2194, |
| "step": 999 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.8989333333333334, |
| "eval_loss": 0.3326835334300995, |
| "eval_runtime": 14.4892, |
| "eval_samples_per_second": 258.814, |
| "eval_steps_per_second": 4.072, |
| "step": 999 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 44.393638610839844, |
| "learning_rate": 0.0, |
| "loss": 0.136, |
| "step": 1332 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.9104, |
| "eval_loss": 0.3074781894683838, |
| "eval_runtime": 14.5551, |
| "eval_samples_per_second": 257.641, |
| "eval_steps_per_second": 4.054, |
| "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 |
| } |
|
|
