| { |
| "best_metric": 0.8786666666666667, |
| "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0923/checkpoints/checkpoint-2997", |
| "epoch": 9.0, |
| "eval_steps": 500, |
| "global_step": 2997, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 1.0, |
| "grad_norm": 46.075042724609375, |
| "learning_rate": 0.0002909538931178862, |
| "loss": 1.7306, |
| "step": 333 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.7946666666666666, |
| "eval_loss": 0.6476795673370361, |
| "eval_runtime": 9.5619, |
| "eval_samples_per_second": 392.18, |
| "eval_steps_per_second": 6.17, |
| "step": 333 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 84.4308853149414, |
| "learning_rate": 0.00026490666646784665, |
| "loss": 0.5711, |
| "step": 666 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.8269333333333333, |
| "eval_loss": 0.5388044714927673, |
| "eval_runtime": 9.5014, |
| "eval_samples_per_second": 394.678, |
| "eval_steps_per_second": 6.21, |
| "step": 666 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 25.506851196289062, |
| "learning_rate": 0.000225, |
| "loss": 0.392, |
| "step": 999 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.8557333333333333, |
| "eval_loss": 0.4795023798942566, |
| "eval_runtime": 9.514, |
| "eval_samples_per_second": 394.156, |
| "eval_steps_per_second": 6.201, |
| "step": 999 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 24.1665096282959, |
| "learning_rate": 0.00017604722665003956, |
| "loss": 0.2688, |
| "step": 1332 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.8592, |
| "eval_loss": 0.48038893938064575, |
| "eval_runtime": 9.452, |
| "eval_samples_per_second": 396.742, |
| "eval_steps_per_second": 6.242, |
| "step": 1332 |
| }, |
| { |
| "epoch": 5.0, |
| "grad_norm": 31.360639572143555, |
| "learning_rate": 0.00012395277334996044, |
| "loss": 0.1874, |
| "step": 1665 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_accuracy": 0.8669333333333333, |
| "eval_loss": 0.4494073987007141, |
| "eval_runtime": 9.0698, |
| "eval_samples_per_second": 413.459, |
| "eval_steps_per_second": 6.505, |
| "step": 1665 |
| }, |
| { |
| "epoch": 6.0, |
| "grad_norm": 34.85376739501953, |
| "learning_rate": 7.500000000000002e-05, |
| "loss": 0.138, |
| "step": 1998 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_accuracy": 0.8690666666666667, |
| "eval_loss": 0.4560053050518036, |
| "eval_runtime": 9.0941, |
| "eval_samples_per_second": 412.355, |
| "eval_steps_per_second": 6.488, |
| "step": 1998 |
| }, |
| { |
| "epoch": 7.0, |
| "grad_norm": 41.313575744628906, |
| "learning_rate": 3.509333353215331e-05, |
| "loss": 0.0995, |
| "step": 2331 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_accuracy": 0.8725333333333334, |
| "eval_loss": 0.4581080675125122, |
| "eval_runtime": 9.0359, |
| "eval_samples_per_second": 415.012, |
| "eval_steps_per_second": 6.53, |
| "step": 2331 |
| }, |
| { |
| "epoch": 8.0, |
| "grad_norm": 28.095003128051758, |
| "learning_rate": 9.046106882113751e-06, |
| "loss": 0.079, |
| "step": 2664 |
| }, |
| { |
| "epoch": 8.0, |
| "eval_accuracy": 0.8784, |
| "eval_loss": 0.4646308124065399, |
| "eval_runtime": 9.1761, |
| "eval_samples_per_second": 408.67, |
| "eval_steps_per_second": 6.43, |
| "step": 2664 |
| }, |
| { |
| "epoch": 9.0, |
| "grad_norm": 71.67350769042969, |
| "learning_rate": 0.0, |
| "loss": 0.071, |
| "step": 2997 |
| }, |
| { |
| "epoch": 9.0, |
| "eval_accuracy": 0.8786666666666667, |
| "eval_loss": 0.4452505111694336, |
| "eval_runtime": 9.0748, |
| "eval_samples_per_second": 413.231, |
| "eval_steps_per_second": 6.501, |
| "step": 2997 |
| } |
| ], |
| "logging_steps": 500, |
| "max_steps": 2997, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 9, |
| "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": 7.3587863155968e+18, |
| "train_batch_size": 64, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
