{ "best_metric": 0.9226666666666666, "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0278/checkpoints/checkpoint-2331", "epoch": 7.0, "eval_steps": 500, "global_step": 2331, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 1.0, "grad_norm": 12.20095443725586, "learning_rate": 3e-05, "loss": 0.741, "step": 333 }, { "epoch": 1.0, "eval_accuracy": 0.9, "eval_loss": 0.32089877128601074, "eval_runtime": 4.1157, "eval_samples_per_second": 911.155, "eval_steps_per_second": 14.336, "step": 333 }, { "epoch": 2.0, "grad_norm": 19.343557357788086, "learning_rate": 3e-05, "loss": 0.1514, "step": 666 }, { "epoch": 2.0, "eval_accuracy": 0.9197333333333333, "eval_loss": 0.2810564339160919, "eval_runtime": 5.3391, "eval_samples_per_second": 702.371, "eval_steps_per_second": 11.051, "step": 666 }, { "epoch": 3.0, "grad_norm": 10.506471633911133, "learning_rate": 3e-05, "loss": 0.0787, "step": 999 }, { "epoch": 3.0, "eval_accuracy": 0.9176, "eval_loss": 0.2960298955440521, "eval_runtime": 4.225, "eval_samples_per_second": 887.58, "eval_steps_per_second": 13.965, "step": 999 }, { "epoch": 4.0, "grad_norm": Infinity, "learning_rate": 3e-05, "loss": 0.055, "step": 1332 }, { "epoch": 4.0, "eval_accuracy": 0.9224, "eval_loss": 0.29083511233329773, "eval_runtime": 4.2284, "eval_samples_per_second": 886.861, "eval_steps_per_second": 13.953, "step": 1332 }, { "epoch": 5.0, "grad_norm": 0.3717098832130432, "learning_rate": 3e-05, "loss": 0.0373, "step": 1665 }, { "epoch": 5.0, "eval_accuracy": 0.9178666666666667, "eval_loss": 0.3240320384502411, "eval_runtime": 4.675, "eval_samples_per_second": 802.148, "eval_steps_per_second": 12.62, "step": 1665 }, { "epoch": 6.0, "grad_norm": 0.04305635392665863, "learning_rate": 3e-05, "loss": 0.0303, "step": 1998 }, { "epoch": 6.0, "eval_accuracy": 0.9210666666666667, "eval_loss": 0.3584335744380951, "eval_runtime": 4.0311, "eval_samples_per_second": 930.26, "eval_steps_per_second": 14.636, "step": 1998 }, { "epoch": 7.0, "grad_norm": 0.09006813913583755, "learning_rate": 3e-05, "loss": 0.0241, "step": 2331 }, { "epoch": 7.0, "eval_accuracy": 0.9226666666666666, "eval_loss": 0.3471022844314575, "eval_runtime": 4.1621, "eval_samples_per_second": 900.978, "eval_steps_per_second": 14.175, "step": 2331 } ], "logging_steps": 500, "max_steps": 2331, "num_input_tokens_seen": 0, "num_train_epochs": 7, "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": 1.153189244358144e+19, "train_batch_size": 64, "trial_name": null, "trial_params": null }