| { |
| "best_metric": 0.8613333333333333, |
| "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0225/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": 77.33866882324219, |
| "learning_rate": 9e-05, |
| "loss": 2.8519, |
| "step": 333 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.6872, |
| "eval_loss": 1.1847468614578247, |
| "eval_runtime": 13.3356, |
| "eval_samples_per_second": 281.202, |
| "eval_steps_per_second": 4.424, |
| "step": 333 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 131.59158325195312, |
| "learning_rate": 9e-05, |
| "loss": 0.7942, |
| "step": 666 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.812, |
| "eval_loss": 0.6346172094345093, |
| "eval_runtime": 13.2826, |
| "eval_samples_per_second": 282.325, |
| "eval_steps_per_second": 4.442, |
| "step": 666 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 143.20480346679688, |
| "learning_rate": 9e-05, |
| "loss": 0.4949, |
| "step": 999 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.8581333333333333, |
| "eval_loss": 0.4704890549182892, |
| "eval_runtime": 13.202, |
| "eval_samples_per_second": 284.048, |
| "eval_steps_per_second": 4.469, |
| "step": 999 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 49.974769592285156, |
| "learning_rate": 9e-05, |
| "loss": 0.3621, |
| "step": 1332 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.8613333333333333, |
| "eval_loss": 0.44395729899406433, |
| "eval_runtime": 13.2569, |
| "eval_samples_per_second": 282.872, |
| "eval_steps_per_second": 4.451, |
| "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 |
| } |
|
|
