{ "best_metric": 0.10852827876806259, "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": 1.0400904417037964, "learning_rate": 9.649122807017545e-06, "loss": 0.6273, "step": 10 }, { "epoch": 0.3508771929824561, "grad_norm": 2.321244239807129, "learning_rate": 9.298245614035088e-06, "loss": 0.5016, "step": 20 }, { "epoch": 0.5263157894736842, "grad_norm": 1.2707430124282837, "learning_rate": 8.947368421052632e-06, "loss": 0.3722, "step": 30 }, { "epoch": 0.7017543859649122, "grad_norm": 1.238302230834961, "learning_rate": 8.596491228070176e-06, "loss": 0.2875, "step": 40 }, { "epoch": 0.8771929824561403, "grad_norm": 0.7120734453201294, "learning_rate": 8.24561403508772e-06, "loss": 0.2431, "step": 50 }, { "epoch": 1.0526315789473684, "grad_norm": 0.7913864850997925, "learning_rate": 7.894736842105265e-06, "loss": 0.1649, "step": 60 }, { "epoch": 1.2280701754385965, "grad_norm": 0.5288364887237549, "learning_rate": 7.5438596491228074e-06, "loss": 0.1515, "step": 70 }, { "epoch": 1.4035087719298245, "grad_norm": 0.541506290435791, "learning_rate": 7.192982456140352e-06, "loss": 0.1479, "step": 80 }, { "epoch": 1.5789473684210527, "grad_norm": 0.5459899306297302, "learning_rate": 6.842105263157896e-06, "loss": 0.1204, "step": 90 }, { "epoch": 1.7543859649122808, "grad_norm": 0.38489124178886414, "learning_rate": 6.491228070175439e-06, "loss": 0.1191, "step": 100 }, { "epoch": 1.7543859649122808, "eval_accuracy": 0.9647577092511013, "eval_confusion_matrix": [ [ 58, 8 ], [ 0, 161 ] ], "eval_loss": 0.15112297236919403, "eval_runtime": 25.4447, "eval_samples_per_second": 8.921, "eval_steps_per_second": 0.314, "step": 100 }, { "epoch": 1.9298245614035088, "grad_norm": 3.6338634490966797, "learning_rate": 6.140350877192983e-06, "loss": 0.1244, "step": 110 }, { "epoch": 2.1052631578947367, "grad_norm": 0.5916204452514648, "learning_rate": 5.789473684210527e-06, "loss": 0.1078, "step": 120 }, { "epoch": 2.280701754385965, "grad_norm": 0.43088626861572266, "learning_rate": 5.438596491228071e-06, "loss": 0.1147, "step": 130 }, { "epoch": 2.456140350877193, "grad_norm": 0.355512410402298, "learning_rate": 5.087719298245615e-06, "loss": 0.0797, "step": 140 }, { "epoch": 2.6315789473684212, "grad_norm": 0.33613306283950806, "learning_rate": 4.736842105263158e-06, "loss": 0.0903, "step": 150 }, { "epoch": 2.807017543859649, "grad_norm": 1.4403668642044067, "learning_rate": 4.385964912280702e-06, "loss": 0.084, "step": 160 }, { "epoch": 2.982456140350877, "grad_norm": 1.5403963327407837, "learning_rate": 4.035087719298246e-06, "loss": 0.0825, "step": 170 }, { "epoch": 3.1578947368421053, "grad_norm": 0.3081526458263397, "learning_rate": 3.6842105263157896e-06, "loss": 0.0893, "step": 180 }, { "epoch": 3.3333333333333335, "grad_norm": 0.3143157958984375, "learning_rate": 3.3333333333333333e-06, "loss": 0.0865, "step": 190 }, { "epoch": 3.5087719298245617, "grad_norm": 0.49979138374328613, "learning_rate": 2.9824561403508774e-06, "loss": 0.09, "step": 200 }, { "epoch": 3.5087719298245617, "eval_accuracy": 0.973568281938326, "eval_confusion_matrix": [ [ 60, 6 ], [ 0, 161 ] ], "eval_loss": 0.10852827876806259, "eval_runtime": 24.9381, "eval_samples_per_second": 9.103, "eval_steps_per_second": 0.321, "step": 200 }, { "epoch": 3.6842105263157894, "grad_norm": 0.26619821786880493, "learning_rate": 2.631578947368421e-06, "loss": 0.0989, "step": 210 }, { "epoch": 3.8596491228070176, "grad_norm": 0.3666665256023407, "learning_rate": 2.280701754385965e-06, "loss": 0.0687, "step": 220 }, { "epoch": 4.035087719298246, "grad_norm": 0.7419259548187256, "learning_rate": 1.929824561403509e-06, "loss": 0.1107, "step": 230 }, { "epoch": 4.2105263157894735, "grad_norm": 0.3317352831363678, "learning_rate": 1.5789473684210526e-06, "loss": 0.075, "step": 240 }, { "epoch": 4.385964912280702, "grad_norm": 0.27071481943130493, "learning_rate": 1.2280701754385965e-06, "loss": 0.0571, "step": 250 }, { "epoch": 4.56140350877193, "grad_norm": 0.2539863586425781, "learning_rate": 8.771929824561404e-07, "loss": 0.0559, "step": 260 }, { "epoch": 4.7368421052631575, "grad_norm": 0.24817486107349396, "learning_rate": 5.263157894736843e-07, "loss": 0.0483, "step": 270 }, { "epoch": 4.912280701754386, "grad_norm": 0.696506917476654, "learning_rate": 1.7543859649122808e-07, "loss": 0.0915, "step": 280 }, { "epoch": 5.0, "step": 285, "total_flos": 7.020774259082035e+17, "train_loss": 0.15217331428276865, "train_runtime": 2935.6937, "train_samples_per_second": 3.107, "train_steps_per_second": 0.097 } ], "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 }