{ "best_metric": 0.10691466927528381, "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": 1.0408214330673218, "learning_rate": 9.649122807017545e-06, "loss": 0.6273, "step": 10 }, { "epoch": 0.3508771929824561, "grad_norm": 2.3301267623901367, "learning_rate": 9.298245614035088e-06, "loss": 0.5016, "step": 20 }, { "epoch": 0.5263157894736842, "grad_norm": 1.2897900342941284, "learning_rate": 8.947368421052632e-06, "loss": 0.3726, "step": 30 }, { "epoch": 0.7017543859649122, "grad_norm": 1.2317367792129517, "learning_rate": 8.596491228070176e-06, "loss": 0.2889, "step": 40 }, { "epoch": 0.8771929824561403, "grad_norm": 0.6814886331558228, "learning_rate": 8.24561403508772e-06, "loss": 0.2442, "step": 50 }, { "epoch": 1.0526315789473684, "grad_norm": 0.7818458080291748, "learning_rate": 7.894736842105265e-06, "loss": 0.1655, "step": 60 }, { "epoch": 1.2280701754385965, "grad_norm": 0.5145708322525024, "learning_rate": 7.5438596491228074e-06, "loss": 0.1518, "step": 70 }, { "epoch": 1.4035087719298245, "grad_norm": 0.5460637807846069, "learning_rate": 7.192982456140352e-06, "loss": 0.1489, "step": 80 }, { "epoch": 1.5789473684210527, "grad_norm": 0.5549706816673279, "learning_rate": 6.842105263157896e-06, "loss": 0.1214, "step": 90 }, { "epoch": 1.7543859649122808, "grad_norm": 0.3951248824596405, "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": 0.152234748005867, "eval_runtime": 25.6232, "eval_samples_per_second": 8.859, "eval_steps_per_second": 0.312, "step": 100 }, { "epoch": 1.9298245614035088, "grad_norm": 3.548973798751831, "learning_rate": 6.140350877192983e-06, "loss": 0.1237, "step": 110 }, { "epoch": 2.1052631578947367, "grad_norm": 0.8018013834953308, "learning_rate": 5.789473684210527e-06, "loss": 0.107, "step": 120 }, { "epoch": 2.280701754385965, "grad_norm": 0.44410237669944763, "learning_rate": 5.438596491228071e-06, "loss": 0.1164, "step": 130 }, { "epoch": 2.456140350877193, "grad_norm": 0.38201433420181274, "learning_rate": 5.087719298245615e-06, "loss": 0.0798, "step": 140 }, { "epoch": 2.6315789473684212, "grad_norm": 0.32631874084472656, "learning_rate": 4.736842105263158e-06, "loss": 0.0899, "step": 150 }, { "epoch": 2.807017543859649, "grad_norm": 0.7354286313056946, "learning_rate": 4.385964912280702e-06, "loss": 0.0839, "step": 160 }, { "epoch": 2.982456140350877, "grad_norm": 0.7380229234695435, "learning_rate": 4.035087719298246e-06, "loss": 0.0815, "step": 170 }, { "epoch": 3.1578947368421053, "grad_norm": 0.2998797297477722, "learning_rate": 3.6842105263157896e-06, "loss": 0.0888, "step": 180 }, { "epoch": 3.3333333333333335, "grad_norm": 0.324789434671402, "learning_rate": 3.3333333333333333e-06, "loss": 0.0868, "step": 190 }, { "epoch": 3.5087719298245617, "grad_norm": 0.5307926535606384, "learning_rate": 2.9824561403508774e-06, "loss": 0.0899, "step": 200 }, { "epoch": 3.5087719298245617, "eval_accuracy": 0.9779735682819384, "eval_confusion_matrix": [ [ 61, 5 ], [ 0, 161 ] ], "eval_loss": 0.10691466927528381, "eval_runtime": 24.2322, "eval_samples_per_second": 9.368, "eval_steps_per_second": 0.33, "step": 200 }, { "epoch": 3.6842105263157894, "grad_norm": 0.2680317461490631, "learning_rate": 2.631578947368421e-06, "loss": 0.0987, "step": 210 }, { "epoch": 3.8596491228070176, "grad_norm": 0.373812198638916, "learning_rate": 2.280701754385965e-06, "loss": 0.0691, "step": 220 }, { "epoch": 4.035087719298246, "grad_norm": 0.6366022825241089, "learning_rate": 1.929824561403509e-06, "loss": 0.1113, "step": 230 }, { "epoch": 4.2105263157894735, "grad_norm": 0.3165888786315918, "learning_rate": 1.5789473684210526e-06, "loss": 0.075, "step": 240 }, { "epoch": 4.385964912280702, "grad_norm": 0.2567109167575836, "learning_rate": 1.2280701754385965e-06, "loss": 0.0567, "step": 250 }, { "epoch": 4.56140350877193, "grad_norm": 0.24947808682918549, "learning_rate": 8.771929824561404e-07, "loss": 0.0557, "step": 260 }, { "epoch": 4.7368421052631575, "grad_norm": 0.2490372359752655, "learning_rate": 5.263157894736843e-07, "loss": 0.0488, "step": 270 }, { "epoch": 4.912280701754386, "grad_norm": 0.6002548336982727, "learning_rate": 1.7543859649122808e-07, "loss": 0.0924, "step": 280 }, { "epoch": 5.0, "step": 285, "total_flos": 7.020774259082035e+17, "train_loss": 0.15237279649366411, "train_runtime": 2884.1273, "train_samples_per_second": 3.162, "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 }