| { | |
| "best_metric": 0.7189333333333333, | |
| "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0051/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": 125.03839874267578, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 3.7205, | |
| "step": 333 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5061333333333333, | |
| "eval_loss": 3.345376491546631, | |
| "eval_runtime": 13.0639, | |
| "eval_samples_per_second": 287.051, | |
| "eval_steps_per_second": 4.516, | |
| "step": 333 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 133.34735107421875, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 2.5919, | |
| "step": 666 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.6698666666666667, | |
| "eval_loss": 1.8750858306884766, | |
| "eval_runtime": 12.5979, | |
| "eval_samples_per_second": 297.668, | |
| "eval_steps_per_second": 4.683, | |
| "step": 666 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 128.18321228027344, | |
| "learning_rate": 0.0, | |
| "loss": 1.5887, | |
| "step": 999 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.7189333333333333, | |
| "eval_loss": 1.5659562349319458, | |
| "eval_runtime": 12.385, | |
| "eval_samples_per_second": 302.786, | |
| "eval_steps_per_second": 4.764, | |
| "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 | |
| } | |