| { |
| "best_metric": 0.8696, |
| "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0285/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": 82.36766815185547, |
| "learning_rate": 7.681980515339464e-05, |
| "loss": 2.811, |
| "step": 333 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.7496, |
| "eval_loss": 1.007891058921814, |
| "eval_runtime": 7.4231, |
| "eval_samples_per_second": 505.181, |
| "eval_steps_per_second": 7.948, |
| "step": 333 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 166.72305297851562, |
| "learning_rate": 4.5e-05, |
| "loss": 0.7142, |
| "step": 666 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.8512, |
| "eval_loss": 0.503613293170929, |
| "eval_runtime": 7.0685, |
| "eval_samples_per_second": 530.521, |
| "eval_steps_per_second": 8.347, |
| "step": 666 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 175.9752197265625, |
| "learning_rate": 1.3180194846605365e-05, |
| "loss": 0.4589, |
| "step": 999 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.8677333333333334, |
| "eval_loss": 0.4431128203868866, |
| "eval_runtime": 7.2878, |
| "eval_samples_per_second": 514.555, |
| "eval_steps_per_second": 8.096, |
| "step": 999 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 74.36175537109375, |
| "learning_rate": 0.0, |
| "loss": 0.3914, |
| "step": 1332 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.8696, |
| "eval_loss": 0.43353191018104553, |
| "eval_runtime": 7.2128, |
| "eval_samples_per_second": 519.907, |
| "eval_steps_per_second": 8.18, |
| "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 |
| } |
|
|
