{ "best_metric": 0.11670097708702087, "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": 1.2752189636230469, "learning_rate": 9.649122807017545e-06, "loss": 0.6232, "step": 10 }, { "epoch": 0.3508771929824561, "grad_norm": 1.752753496170044, "learning_rate": 9.298245614035088e-06, "loss": 0.473, "step": 20 }, { "epoch": 0.5263157894736842, "grad_norm": 1.3884831666946411, "learning_rate": 8.947368421052632e-06, "loss": 0.3834, "step": 30 }, { "epoch": 0.7017543859649122, "grad_norm": 1.6439697742462158, "learning_rate": 8.596491228070176e-06, "loss": 0.305, "step": 40 }, { "epoch": 0.8771929824561403, "grad_norm": 1.1231735944747925, "learning_rate": 8.24561403508772e-06, "loss": 0.2869, "step": 50 }, { "epoch": 1.0526315789473684, "grad_norm": 0.8180190324783325, "learning_rate": 7.894736842105265e-06, "loss": 0.2178, "step": 60 }, { "epoch": 1.2280701754385965, "grad_norm": 0.6753518581390381, "learning_rate": 7.5438596491228074e-06, "loss": 0.206, "step": 70 }, { "epoch": 1.4035087719298245, "grad_norm": 1.176467776298523, "learning_rate": 7.192982456140352e-06, "loss": 0.1828, "step": 80 }, { "epoch": 1.5789473684210527, "grad_norm": 0.7861670255661011, "learning_rate": 6.842105263157896e-06, "loss": 0.1564, "step": 90 }, { "epoch": 1.7543859649122808, "grad_norm": 0.4661114811897278, "learning_rate": 6.491228070175439e-06, "loss": 0.1497, "step": 100 }, { "epoch": 1.7543859649122808, "eval_accuracy": 0.9647577092511013, "eval_confusion_matrix": [ [ 58, 8 ], [ 0, 161 ] ], "eval_loss": 0.1807878464460373, "eval_runtime": 24.5815, "eval_samples_per_second": 9.235, "eval_steps_per_second": 0.325, "step": 100 }, { "epoch": 1.9298245614035088, "grad_norm": 5.353541851043701, "learning_rate": 6.140350877192983e-06, "loss": 0.1463, "step": 110 }, { "epoch": 2.1052631578947367, "grad_norm": 0.47052720189094543, "learning_rate": 5.789473684210527e-06, "loss": 0.1291, "step": 120 }, { "epoch": 2.280701754385965, "grad_norm": 0.43008866906166077, "learning_rate": 5.438596491228071e-06, "loss": 0.1294, "step": 130 }, { "epoch": 2.456140350877193, "grad_norm": 0.495709627866745, "learning_rate": 5.087719298245615e-06, "loss": 0.0904, "step": 140 }, { "epoch": 2.6315789473684212, "grad_norm": 0.36288028955459595, "learning_rate": 4.736842105263158e-06, "loss": 0.1004, "step": 150 }, { "epoch": 2.807017543859649, "grad_norm": 0.42618927359580994, "learning_rate": 4.385964912280702e-06, "loss": 0.0931, "step": 160 }, { "epoch": 2.982456140350877, "grad_norm": 1.7581294775009155, "learning_rate": 4.035087719298246e-06, "loss": 0.0909, "step": 170 }, { "epoch": 3.1578947368421053, "grad_norm": 0.3336179554462433, "learning_rate": 3.6842105263157896e-06, "loss": 0.0991, "step": 180 }, { "epoch": 3.3333333333333335, "grad_norm": 0.3549640476703644, "learning_rate": 3.3333333333333333e-06, "loss": 0.089, "step": 190 }, { "epoch": 3.5087719298245617, "grad_norm": 0.6302672624588013, "learning_rate": 2.9824561403508774e-06, "loss": 0.0998, "step": 200 }, { "epoch": 3.5087719298245617, "eval_accuracy": 0.9779735682819384, "eval_confusion_matrix": [ [ 61, 5 ], [ 0, 161 ] ], "eval_loss": 0.11670097708702087, "eval_runtime": 25.3798, "eval_samples_per_second": 8.944, "eval_steps_per_second": 0.315, "step": 200 }, { "epoch": 3.6842105263157894, "grad_norm": 0.3984730839729309, "learning_rate": 2.631578947368421e-06, "loss": 0.1028, "step": 210 }, { "epoch": 3.8596491228070176, "grad_norm": 0.4672948718070984, "learning_rate": 2.280701754385965e-06, "loss": 0.0746, "step": 220 }, { "epoch": 4.035087719298246, "grad_norm": 0.7369114756584167, "learning_rate": 1.929824561403509e-06, "loss": 0.1065, "step": 230 }, { "epoch": 4.2105263157894735, "grad_norm": 0.361775666475296, "learning_rate": 1.5789473684210526e-06, "loss": 0.0779, "step": 240 }, { "epoch": 4.385964912280702, "grad_norm": 0.29275935888290405, "learning_rate": 1.2280701754385965e-06, "loss": 0.0615, "step": 250 }, { "epoch": 4.56140350877193, "grad_norm": 0.30065467953681946, "learning_rate": 8.771929824561404e-07, "loss": 0.0603, "step": 260 }, { "epoch": 4.7368421052631575, "grad_norm": 0.28486543893814087, "learning_rate": 5.263157894736843e-07, "loss": 0.0523, "step": 270 }, { "epoch": 4.912280701754386, "grad_norm": 0.5634991526603699, "learning_rate": 1.7543859649122808e-07, "loss": 0.0905, "step": 280 }, { "epoch": 5.0, "step": 285, "total_flos": 7.020774259082035e+17, "train_loss": 0.16588184833526612, "train_runtime": 2879.0892, "train_samples_per_second": 3.168, "train_steps_per_second": 0.099 } ], "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 }