| { | |
| "best_metric": 0.7346666666666667, | |
| "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0251/checkpoints/checkpoint-999", | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 999, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 199.42501831054688, | |
| "learning_rate": 5.2499999999999995e-05, | |
| "loss": 3.4337, | |
| "step": 333 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5674666666666667, | |
| "eval_loss": 2.3823511600494385, | |
| "eval_runtime": 14.2116, | |
| "eval_samples_per_second": 263.868, | |
| "eval_steps_per_second": 4.152, | |
| "step": 333 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 87.78582000732422, | |
| "learning_rate": 1.7500000000000005e-05, | |
| "loss": 1.5686, | |
| "step": 666 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7213333333333334, | |
| "eval_loss": 1.129364252090454, | |
| "eval_runtime": 14.1938, | |
| "eval_samples_per_second": 264.201, | |
| "eval_steps_per_second": 4.157, | |
| "step": 666 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 120.55499267578125, | |
| "learning_rate": 0.0, | |
| "loss": 1.0735, | |
| "step": 999 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.7346666666666667, | |
| "eval_loss": 1.0167325735092163, | |
| "eval_runtime": 14.2751, | |
| "eval_samples_per_second": 262.696, | |
| "eval_steps_per_second": 4.133, | |
| "step": 999 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 999, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "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": 2.4529287718656e+18, | |
| "train_batch_size": 64, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |