| { |
| "best_metric": 0.9125333333333333, |
| "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0447/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": 73.5950698852539, |
| "learning_rate": 0.000225, |
| "loss": 1.4862, |
| "step": 333 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.8608, |
| "eval_loss": 0.47239360213279724, |
| "eval_runtime": 7.8017, |
| "eval_samples_per_second": 480.665, |
| "eval_steps_per_second": 7.562, |
| "step": 333 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 133.64695739746094, |
| "learning_rate": 0.00015, |
| "loss": 0.369, |
| "step": 666 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.8962666666666667, |
| "eval_loss": 0.35309845209121704, |
| "eval_runtime": 7.8516, |
| "eval_samples_per_second": 477.607, |
| "eval_steps_per_second": 7.514, |
| "step": 666 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 89.02796936035156, |
| "learning_rate": 7.5e-05, |
| "loss": 0.2036, |
| "step": 999 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.9096, |
| "eval_loss": 0.321846067905426, |
| "eval_runtime": 7.8789, |
| "eval_samples_per_second": 475.957, |
| "eval_steps_per_second": 7.488, |
| "step": 999 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 29.980043411254883, |
| "learning_rate": 0.0, |
| "loss": 0.1146, |
| "step": 1332 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.9125333333333333, |
| "eval_loss": 0.29979413747787476, |
| "eval_runtime": 7.9893, |
| "eval_samples_per_second": 469.38, |
| "eval_steps_per_second": 7.385, |
| "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 |
| } |
|
|
