{ "best_metric": 0.11440654844045639, "best_model_checkpoint": "/home/projects/Project - Fundus Photo Glaucoma Detection (AI)//model_results/vit-base-gpu/checkpoint-200", "epoch": 5.0, "eval_steps": 100, "global_step": 285, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.17543859649122806, "grad_norm": 1.0955568552017212, "learning_rate": 9.649122807017545e-06, "loss": 0.6216, "step": 10 }, { "epoch": 0.3508771929824561, "grad_norm": 2.5903913974761963, "learning_rate": 9.298245614035088e-06, "loss": 0.5035, "step": 20 }, { "epoch": 0.5263157894736842, "grad_norm": 1.378684639930725, "learning_rate": 8.947368421052632e-06, "loss": 0.3913, "step": 30 }, { "epoch": 0.7017543859649122, "grad_norm": 1.2637547254562378, "learning_rate": 8.596491228070176e-06, "loss": 0.3097, "step": 40 }, { "epoch": 0.8771929824561403, "grad_norm": 1.0544745922088623, "learning_rate": 8.24561403508772e-06, "loss": 0.2689, "step": 50 }, { "epoch": 1.0526315789473684, "grad_norm": 0.7577498555183411, "learning_rate": 7.894736842105265e-06, "loss": 0.1876, "step": 60 }, { "epoch": 1.2280701754385965, "grad_norm": 0.7630872130393982, "learning_rate": 7.5438596491228074e-06, "loss": 0.1723, "step": 70 }, { "epoch": 1.4035087719298245, "grad_norm": 0.5983202457427979, "learning_rate": 7.192982456140352e-06, "loss": 0.1496, "step": 80 }, { "epoch": 1.5789473684210527, "grad_norm": 1.4984568357467651, "learning_rate": 6.842105263157896e-06, "loss": 0.1248, "step": 90 }, { "epoch": 1.7543859649122808, "grad_norm": 1.8406482934951782, "learning_rate": 6.491228070175439e-06, "loss": 0.1284, "step": 100 }, { "epoch": 1.7543859649122808, "eval_accuracy": 0.9647577092511013, "eval_confusion_matrix": [ [ 58, 8 ], [ 0, 161 ] ], "eval_loss": 0.1682671159505844, "eval_runtime": 24.1717, "eval_samples_per_second": 9.391, "eval_steps_per_second": 0.331, "step": 100 }, { "epoch": 1.9298245614035088, "grad_norm": 4.00988245010376, "learning_rate": 6.140350877192983e-06, "loss": 0.1316, "step": 110 }, { "epoch": 2.1052631578947367, "grad_norm": 3.0652315616607666, "learning_rate": 5.789473684210527e-06, "loss": 0.1133, "step": 120 }, { "epoch": 2.280701754385965, "grad_norm": 0.3155815303325653, "learning_rate": 5.438596491228071e-06, "loss": 0.1197, "step": 130 }, { "epoch": 2.456140350877193, "grad_norm": 0.848831057548523, "learning_rate": 5.087719298245615e-06, "loss": 0.0831, "step": 140 }, { "epoch": 2.6315789473684212, "grad_norm": 0.3472151756286621, "learning_rate": 4.736842105263158e-06, "loss": 0.0943, "step": 150 }, { "epoch": 2.807017543859649, "grad_norm": 0.5705087780952454, "learning_rate": 4.385964912280702e-06, "loss": 0.0886, "step": 160 }, { "epoch": 2.982456140350877, "grad_norm": 3.6621124744415283, "learning_rate": 4.035087719298246e-06, "loss": 0.091, "step": 170 }, { "epoch": 3.1578947368421053, "grad_norm": 0.34632256627082825, "learning_rate": 3.6842105263157896e-06, "loss": 0.0906, "step": 180 }, { "epoch": 3.3333333333333335, "grad_norm": 0.31107649207115173, "learning_rate": 3.3333333333333333e-06, "loss": 0.0893, "step": 190 }, { "epoch": 3.5087719298245617, "grad_norm": 1.2312792539596558, "learning_rate": 2.9824561403508774e-06, "loss": 0.0943, "step": 200 }, { "epoch": 3.5087719298245617, "eval_accuracy": 0.973568281938326, "eval_confusion_matrix": [ [ 60, 6 ], [ 0, 161 ] ], "eval_loss": 0.11440654844045639, "eval_runtime": 24.1834, "eval_samples_per_second": 9.387, "eval_steps_per_second": 0.331, "step": 200 }, { "epoch": 3.6842105263157894, "grad_norm": 0.2930678725242615, "learning_rate": 2.631578947368421e-06, "loss": 0.1119, "step": 210 }, { "epoch": 3.8596491228070176, "grad_norm": 0.34262707829475403, "learning_rate": 2.280701754385965e-06, "loss": 0.0739, "step": 220 }, { "epoch": 4.035087719298246, "grad_norm": 0.6521685123443604, "learning_rate": 1.929824561403509e-06, "loss": 0.1082, "step": 230 }, { "epoch": 4.2105263157894735, "grad_norm": 0.42672786116600037, "learning_rate": 1.5789473684210526e-06, "loss": 0.0797, "step": 240 }, { "epoch": 4.385964912280702, "grad_norm": 0.2844175100326538, "learning_rate": 1.2280701754385965e-06, "loss": 0.0613, "step": 250 }, { "epoch": 4.56140350877193, "grad_norm": 0.2799729108810425, "learning_rate": 8.771929824561404e-07, "loss": 0.0594, "step": 260 }, { "epoch": 4.7368421052631575, "grad_norm": 0.27341070771217346, "learning_rate": 5.263157894736843e-07, "loss": 0.0479, "step": 270 }, { "epoch": 4.912280701754386, "grad_norm": 0.501814603805542, "learning_rate": 1.7543859649122808e-07, "loss": 0.1039, "step": 280 }, { "epoch": 5.0, "step": 285, "total_flos": 7.020774259082035e+17, "train_loss": 0.15960428672924376, "train_runtime": 2897.6063, "train_samples_per_second": 3.147, "train_steps_per_second": 0.098 } ], "logging_steps": 10, "max_steps": 285, "num_input_tokens_seen": 0, "num_train_epochs": 5, "save_steps": 100, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": true }, "attributes": {} } }, "total_flos": 7.020774259082035e+17, "train_batch_size": 32, "trial_name": null, "trial_params": null }