{ "best_metric": 0.11656656116247177, "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": 1.2793735265731812, "learning_rate": 9.649122807017545e-06, "loss": 0.6232, "step": 10 }, { "epoch": 0.3508771929824561, "grad_norm": 1.7450703382492065, "learning_rate": 9.298245614035088e-06, "loss": 0.4729, "step": 20 }, { "epoch": 0.5263157894736842, "grad_norm": 1.4280253648757935, "learning_rate": 8.947368421052632e-06, "loss": 0.3831, "step": 30 }, { "epoch": 0.7017543859649122, "grad_norm": 1.5422084331512451, "learning_rate": 8.596491228070176e-06, "loss": 0.3052, "step": 40 }, { "epoch": 0.8771929824561403, "grad_norm": 1.118567943572998, "learning_rate": 8.24561403508772e-06, "loss": 0.2872, "step": 50 }, { "epoch": 1.0526315789473684, "grad_norm": 0.8227060437202454, "learning_rate": 7.894736842105265e-06, "loss": 0.2179, "step": 60 }, { "epoch": 1.2280701754385965, "grad_norm": 0.6690755486488342, "learning_rate": 7.5438596491228074e-06, "loss": 0.2056, "step": 70 }, { "epoch": 1.4035087719298245, "grad_norm": 1.198415756225586, "learning_rate": 7.192982456140352e-06, "loss": 0.1825, "step": 80 }, { "epoch": 1.5789473684210527, "grad_norm": 0.763924777507782, "learning_rate": 6.842105263157896e-06, "loss": 0.1564, "step": 90 }, { "epoch": 1.7543859649122808, "grad_norm": 0.4554818570613861, "learning_rate": 6.491228070175439e-06, "loss": 0.1496, "step": 100 }, { "epoch": 1.7543859649122808, "eval_accuracy": 0.9647577092511013, "eval_confusion_matrix": [ [ 58, 8 ], [ 0, 161 ] ], "eval_loss": 0.17874090373516083, "eval_runtime": 24.3384, "eval_samples_per_second": 9.327, "eval_steps_per_second": 0.329, "step": 100 }, { "epoch": 1.9298245614035088, "grad_norm": 4.51282262802124, "learning_rate": 6.140350877192983e-06, "loss": 0.1444, "step": 110 }, { "epoch": 2.1052631578947367, "grad_norm": 0.47525253891944885, "learning_rate": 5.789473684210527e-06, "loss": 0.1274, "step": 120 }, { "epoch": 2.280701754385965, "grad_norm": 0.4660364091396332, "learning_rate": 5.438596491228071e-06, "loss": 0.1291, "step": 130 }, { "epoch": 2.456140350877193, "grad_norm": 0.5069414973258972, "learning_rate": 5.087719298245615e-06, "loss": 0.0901, "step": 140 }, { "epoch": 2.6315789473684212, "grad_norm": 0.36069270968437195, "learning_rate": 4.736842105263158e-06, "loss": 0.0998, "step": 150 }, { "epoch": 2.807017543859649, "grad_norm": 0.4235830008983612, "learning_rate": 4.385964912280702e-06, "loss": 0.0929, "step": 160 }, { "epoch": 2.982456140350877, "grad_norm": 1.640547275543213, "learning_rate": 4.035087719298246e-06, "loss": 0.0905, "step": 170 }, { "epoch": 3.1578947368421053, "grad_norm": 0.33297139406204224, "learning_rate": 3.6842105263157896e-06, "loss": 0.0982, "step": 180 }, { "epoch": 3.3333333333333335, "grad_norm": 0.3522562086582184, "learning_rate": 3.3333333333333333e-06, "loss": 0.0889, "step": 190 }, { "epoch": 3.5087719298245617, "grad_norm": 0.5093055367469788, "learning_rate": 2.9824561403508774e-06, "loss": 0.1002, "step": 200 }, { "epoch": 3.5087719298245617, "eval_accuracy": 0.9779735682819384, "eval_confusion_matrix": [ [ 61, 5 ], [ 0, 161 ] ], "eval_loss": 0.11656656116247177, "eval_runtime": 24.5633, "eval_samples_per_second": 9.241, "eval_steps_per_second": 0.326, "step": 200 }, { "epoch": 3.6842105263157894, "grad_norm": 0.3924790918827057, "learning_rate": 2.631578947368421e-06, "loss": 0.1029, "step": 210 }, { "epoch": 3.8596491228070176, "grad_norm": 0.43185240030288696, "learning_rate": 2.280701754385965e-06, "loss": 0.0743, "step": 220 }, { "epoch": 4.035087719298246, "grad_norm": 0.7460317015647888, "learning_rate": 1.929824561403509e-06, "loss": 0.1062, "step": 230 }, { "epoch": 4.2105263157894735, "grad_norm": 0.36178895831108093, "learning_rate": 1.5789473684210526e-06, "loss": 0.0778, "step": 240 }, { "epoch": 4.385964912280702, "grad_norm": 0.2968393862247467, "learning_rate": 1.2280701754385965e-06, "loss": 0.0613, "step": 250 }, { "epoch": 4.56140350877193, "grad_norm": 0.3041611611843109, "learning_rate": 8.771929824561404e-07, "loss": 0.0602, "step": 260 }, { "epoch": 4.7368421052631575, "grad_norm": 0.2833913564682007, "learning_rate": 5.263157894736843e-07, "loss": 0.053, "step": 270 }, { "epoch": 4.912280701754386, "grad_norm": 0.5582696199417114, "learning_rate": 1.7543859649122808e-07, "loss": 0.0901, "step": 280 }, { "epoch": 5.0, "step": 285, "total_flos": 7.020774259082035e+17, "train_loss": 0.1656334604087629, "train_runtime": 2889.8629, "train_samples_per_second": 3.156, "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 }