| { | |
| "best_metric": 0.8394666666666667, | |
| "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0018/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": 90.74603271484375, | |
| "learning_rate": 5e-05, | |
| "loss": 3.6459, | |
| "step": 333 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5472, | |
| "eval_loss": 2.9521560668945312, | |
| "eval_runtime": 5.3795, | |
| "eval_samples_per_second": 697.085, | |
| "eval_steps_per_second": 10.967, | |
| "step": 333 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 215.66773986816406, | |
| "learning_rate": 5e-05, | |
| "loss": 1.652, | |
| "step": 666 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.764, | |
| "eval_loss": 0.9053675532341003, | |
| "eval_runtime": 5.3254, | |
| "eval_samples_per_second": 704.174, | |
| "eval_steps_per_second": 11.079, | |
| "step": 666 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 132.63424682617188, | |
| "learning_rate": 5e-05, | |
| "loss": 0.7872, | |
| "step": 999 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8208, | |
| "eval_loss": 0.6271626353263855, | |
| "eval_runtime": 5.3985, | |
| "eval_samples_per_second": 694.635, | |
| "eval_steps_per_second": 10.929, | |
| "step": 999 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 250.75367736816406, | |
| "learning_rate": 5e-05, | |
| "loss": 0.5824, | |
| "step": 1332 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8394666666666667, | |
| "eval_loss": 0.5343669056892395, | |
| "eval_runtime": 5.3888, | |
| "eval_samples_per_second": 695.885, | |
| "eval_steps_per_second": 10.949, | |
| "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 | |
| } | |