{ "best_metric": 0.8239795918367347, "best_model_checkpoint": "dinov2-base-rawdata-finetuned-SkinDisease/checkpoint-70", "epoch": 10.0, "eval_steps": 500, "global_step": 70, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 1.0, "eval_accuracy": 0.29336734693877553, "eval_loss": 2.5043179988861084, "eval_runtime": 6.2443, "eval_samples_per_second": 62.777, "eval_steps_per_second": 0.641, "step": 7 }, { "epoch": 1.4285714285714286, "grad_norm": 91.53591918945312, "learning_rate": 4.761904761904762e-05, "loss": 3.0966, "step": 10 }, { "epoch": 2.0, "eval_accuracy": 0.40561224489795916, "eval_loss": 1.808035135269165, "eval_runtime": 5.5878, "eval_samples_per_second": 70.153, "eval_steps_per_second": 0.716, "step": 14 }, { "epoch": 2.857142857142857, "grad_norm": 78.80509948730469, "learning_rate": 3.968253968253968e-05, "loss": 1.6416, "step": 20 }, { "epoch": 3.0, "eval_accuracy": 0.5561224489795918, "eval_loss": 1.3656367063522339, "eval_runtime": 6.0741, "eval_samples_per_second": 64.537, "eval_steps_per_second": 0.659, "step": 21 }, { "epoch": 4.0, "eval_accuracy": 0.6632653061224489, "eval_loss": 1.0316075086593628, "eval_runtime": 6.0396, "eval_samples_per_second": 64.905, "eval_steps_per_second": 0.662, "step": 28 }, { "epoch": 4.285714285714286, "grad_norm": 47.57957077026367, "learning_rate": 3.1746031746031745e-05, "loss": 0.9959, "step": 30 }, { "epoch": 5.0, "eval_accuracy": 0.6989795918367347, "eval_loss": 0.9350685477256775, "eval_runtime": 5.5096, "eval_samples_per_second": 71.149, "eval_steps_per_second": 0.726, "step": 35 }, { "epoch": 5.714285714285714, "grad_norm": 51.770687103271484, "learning_rate": 2.380952380952381e-05, "loss": 0.7114, "step": 40 }, { "epoch": 6.0, "eval_accuracy": 0.7091836734693877, "eval_loss": 0.840167224407196, "eval_runtime": 6.6782, "eval_samples_per_second": 58.698, "eval_steps_per_second": 0.599, "step": 42 }, { "epoch": 7.0, "eval_accuracy": 0.7551020408163265, "eval_loss": 0.7250941395759583, "eval_runtime": 6.5348, "eval_samples_per_second": 59.986, "eval_steps_per_second": 0.612, "step": 49 }, { "epoch": 7.142857142857143, "grad_norm": 48.86288070678711, "learning_rate": 1.5873015873015872e-05, "loss": 0.5098, "step": 50 }, { "epoch": 8.0, "eval_accuracy": 0.7882653061224489, "eval_loss": 0.6774773597717285, "eval_runtime": 6.5943, "eval_samples_per_second": 59.445, "eval_steps_per_second": 0.607, "step": 56 }, { "epoch": 8.571428571428571, "grad_norm": 50.32086944580078, "learning_rate": 7.936507936507936e-06, "loss": 0.394, "step": 60 }, { "epoch": 9.0, "eval_accuracy": 0.7908163265306123, "eval_loss": 0.6402073502540588, "eval_runtime": 5.5562, "eval_samples_per_second": 70.552, "eval_steps_per_second": 0.72, "step": 63 }, { "epoch": 10.0, "grad_norm": 39.44784164428711, "learning_rate": 0.0, "loss": 0.336, "step": 70 }, { "epoch": 10.0, "eval_accuracy": 0.8239795918367347, "eval_loss": 0.5953671932220459, "eval_runtime": 5.497, "eval_samples_per_second": 71.312, "eval_steps_per_second": 0.728, "step": 70 }, { "epoch": 10.0, "step": 70, "total_flos": 3.601200153797591e+18, "train_loss": 1.0979123047419956, "train_runtime": 797.9064, "train_samples_per_second": 44.166, "train_steps_per_second": 0.088 } ], "logging_steps": 10, "max_steps": 70, "num_input_tokens_seen": 0, "num_train_epochs": 10, "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": 3.601200153797591e+18, "train_batch_size": 128, "trial_name": null, "trial_params": null }