| { | |
| "best_metric": 0.8904, | |
| "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0276/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": 40.414615631103516, | |
| "learning_rate": 0.0003, | |
| "loss": 1.4702, | |
| "step": 333 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8501333333333333, | |
| "eval_loss": 0.491788387298584, | |
| "eval_runtime": 7.0869, | |
| "eval_samples_per_second": 529.143, | |
| "eval_steps_per_second": 8.325, | |
| "step": 333 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 51.90673828125, | |
| "learning_rate": 0.0003, | |
| "loss": 0.4067, | |
| "step": 666 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.864, | |
| "eval_loss": 0.43237975239753723, | |
| "eval_runtime": 7.0938, | |
| "eval_samples_per_second": 528.63, | |
| "eval_steps_per_second": 8.317, | |
| "step": 666 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 47.88104248046875, | |
| "learning_rate": 0.0003, | |
| "loss": 0.2326, | |
| "step": 999 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8770666666666667, | |
| "eval_loss": 0.4306904375553131, | |
| "eval_runtime": 7.0237, | |
| "eval_samples_per_second": 533.908, | |
| "eval_steps_per_second": 8.4, | |
| "step": 999 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 75.5455093383789, | |
| "learning_rate": 0.0003, | |
| "loss": 0.1469, | |
| "step": 1332 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8904, | |
| "eval_loss": 0.4010811448097229, | |
| "eval_runtime": 7.0615, | |
| "eval_samples_per_second": 531.045, | |
| "eval_steps_per_second": 8.355, | |
| "step": 1332 | |
| } | |
| ], | |
| "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": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.2705716958208e+18, | |
| "train_batch_size": 64, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |