| { |
| "best_metric": 0.8602666666666666, |
| "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0073/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": 68.89601135253906, |
| "learning_rate": 0.0005, |
| "loss": 1.4115, |
| "step": 333 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.8098666666666666, |
| "eval_loss": 0.6413070559501648, |
| "eval_runtime": 7.7331, |
| "eval_samples_per_second": 484.929, |
| "eval_steps_per_second": 7.63, |
| "step": 333 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 28.683923721313477, |
| "learning_rate": 0.0005, |
| "loss": 0.5265, |
| "step": 666 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.8448, |
| "eval_loss": 0.5106350183486938, |
| "eval_runtime": 7.7559, |
| "eval_samples_per_second": 483.502, |
| "eval_steps_per_second": 7.607, |
| "step": 666 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 23.327964782714844, |
| "learning_rate": 0.0005, |
| "loss": 0.3462, |
| "step": 999 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.8496, |
| "eval_loss": 0.5320701003074646, |
| "eval_runtime": 7.5219, |
| "eval_samples_per_second": 498.545, |
| "eval_steps_per_second": 7.844, |
| "step": 999 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 28.17293930053711, |
| "learning_rate": 0.0005, |
| "loss": 0.2385, |
| "step": 1332 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.8602666666666666, |
| "eval_loss": 0.5266678333282471, |
| "eval_runtime": 7.5351, |
| "eval_samples_per_second": 497.67, |
| "eval_steps_per_second": 7.83, |
| "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 |
| } |
|
|
