| { | |
| "best_metric": 0.5597333333333333, | |
| "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0010/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": 166.27540588378906, | |
| "learning_rate": 2.5606601717798212e-05, | |
| "loss": 3.8622, | |
| "step": 333 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.28586666666666666, | |
| "eval_loss": 3.778660535812378, | |
| "eval_runtime": 15.0503, | |
| "eval_samples_per_second": 249.164, | |
| "eval_steps_per_second": 3.92, | |
| "step": 333 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 156.5283966064453, | |
| "learning_rate": 1.5e-05, | |
| "loss": 3.4487, | |
| "step": 666 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.4938666666666667, | |
| "eval_loss": 3.217481851577759, | |
| "eval_runtime": 13.5248, | |
| "eval_samples_per_second": 277.269, | |
| "eval_steps_per_second": 4.362, | |
| "step": 666 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 202.46359252929688, | |
| "learning_rate": 4.393398282201788e-06, | |
| "loss": 2.7874, | |
| "step": 999 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.5597333333333333, | |
| "eval_loss": 2.6404495239257812, | |
| "eval_runtime": 13.2578, | |
| "eval_samples_per_second": 282.853, | |
| "eval_steps_per_second": 4.45, | |
| "step": 999 | |
| } | |
| ], | |
| "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": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.4529287718656e+18, | |
| "train_batch_size": 64, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |