| { |
| "best_metric": 0.8688, |
| "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0291/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": 94.9514389038086, |
| "learning_rate": 5e-05, |
| "loss": 3.5478, |
| "step": 333 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.6005333333333334, |
| "eval_loss": 2.6279115676879883, |
| "eval_runtime": 14.2791, |
| "eval_samples_per_second": 262.621, |
| "eval_steps_per_second": 4.132, |
| "step": 333 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 107.94786071777344, |
| "learning_rate": 5e-05, |
| "loss": 1.3953, |
| "step": 666 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.8072, |
| "eval_loss": 0.772723913192749, |
| "eval_runtime": 14.3699, |
| "eval_samples_per_second": 260.962, |
| "eval_steps_per_second": 4.106, |
| "step": 666 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 110.00310516357422, |
| "learning_rate": 5e-05, |
| "loss": 0.6042, |
| "step": 999 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.8517333333333333, |
| "eval_loss": 0.5252550840377808, |
| "eval_runtime": 14.1797, |
| "eval_samples_per_second": 264.462, |
| "eval_steps_per_second": 4.161, |
| "step": 999 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 92.72199249267578, |
| "learning_rate": 5e-05, |
| "loss": 0.3999, |
| "step": 1332 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.8688, |
| "eval_loss": 0.4477456510066986, |
| "eval_runtime": 14.3199, |
| "eval_samples_per_second": 261.873, |
| "eval_steps_per_second": 4.12, |
| "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 |
| } |
|
|
