{ "best_metric": 0.1123642697930336, "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": 1.4138169288635254, "learning_rate": 9.649122807017545e-06, "loss": 0.5957, "step": 10 }, { "epoch": 0.3508771929824561, "grad_norm": 1.6343547105789185, "learning_rate": 9.298245614035088e-06, "loss": 0.4704, "step": 20 }, { "epoch": 0.5263157894736842, "grad_norm": 1.1984628438949585, "learning_rate": 8.947368421052632e-06, "loss": 0.3792, "step": 30 }, { "epoch": 0.7017543859649122, "grad_norm": 1.2228361368179321, "learning_rate": 8.596491228070176e-06, "loss": 0.3013, "step": 40 }, { "epoch": 0.8771929824561403, "grad_norm": 1.6350319385528564, "learning_rate": 8.24561403508772e-06, "loss": 0.267, "step": 50 }, { "epoch": 1.0526315789473684, "grad_norm": 0.7450676560401917, "learning_rate": 7.894736842105265e-06, "loss": 0.189, "step": 60 }, { "epoch": 1.2280701754385965, "grad_norm": 0.6679130792617798, "learning_rate": 7.5438596491228074e-06, "loss": 0.1731, "step": 70 }, { "epoch": 1.4035087719298245, "grad_norm": 0.6309726238250732, "learning_rate": 7.192982456140352e-06, "loss": 0.1479, "step": 80 }, { "epoch": 1.5789473684210527, "grad_norm": 0.5439296960830688, "learning_rate": 6.842105263157896e-06, "loss": 0.1194, "step": 90 }, { "epoch": 1.7543859649122808, "grad_norm": 2.2208399772644043, "learning_rate": 6.491228070175439e-06, "loss": 0.1265, "step": 100 }, { "epoch": 1.7543859649122808, "eval_accuracy": 0.9691629955947136, "eval_confusion_matrix": [ [ 59, 7 ], [ 0, 161 ] ], "eval_loss": 0.15117686986923218, "eval_runtime": 24.8749, "eval_samples_per_second": 9.126, "eval_steps_per_second": 0.322, "step": 100 }, { "epoch": 1.9298245614035088, "grad_norm": 4.325542449951172, "learning_rate": 6.140350877192983e-06, "loss": 0.1281, "step": 110 }, { "epoch": 2.1052631578947367, "grad_norm": 0.43080058693885803, "learning_rate": 5.789473684210527e-06, "loss": 0.0971, "step": 120 }, { "epoch": 2.280701754385965, "grad_norm": 0.3750275671482086, "learning_rate": 5.438596491228071e-06, "loss": 0.1155, "step": 130 }, { "epoch": 2.456140350877193, "grad_norm": 0.42711108922958374, "learning_rate": 5.087719298245615e-06, "loss": 0.0805, "step": 140 }, { "epoch": 2.6315789473684212, "grad_norm": 0.3187861740589142, "learning_rate": 4.736842105263158e-06, "loss": 0.0936, "step": 150 }, { "epoch": 2.807017543859649, "grad_norm": 0.36738303303718567, "learning_rate": 4.385964912280702e-06, "loss": 0.084, "step": 160 }, { "epoch": 2.982456140350877, "grad_norm": 2.0587196350097656, "learning_rate": 4.035087719298246e-06, "loss": 0.0855, "step": 170 }, { "epoch": 3.1578947368421053, "grad_norm": 0.31759971380233765, "learning_rate": 3.6842105263157896e-06, "loss": 0.0896, "step": 180 }, { "epoch": 3.3333333333333335, "grad_norm": 0.2914780378341675, "learning_rate": 3.3333333333333333e-06, "loss": 0.0814, "step": 190 }, { "epoch": 3.5087719298245617, "grad_norm": 0.33958643674850464, "learning_rate": 2.9824561403508774e-06, "loss": 0.0892, "step": 200 }, { "epoch": 3.5087719298245617, "eval_accuracy": 0.9691629955947136, "eval_confusion_matrix": [ [ 60, 6 ], [ 1, 160 ] ], "eval_loss": 0.1123642697930336, "eval_runtime": 24.2578, "eval_samples_per_second": 9.358, "eval_steps_per_second": 0.33, "step": 200 }, { "epoch": 3.6842105263157894, "grad_norm": 0.27888378500938416, "learning_rate": 2.631578947368421e-06, "loss": 0.1024, "step": 210 }, { "epoch": 3.8596491228070176, "grad_norm": 0.526310920715332, "learning_rate": 2.280701754385965e-06, "loss": 0.067, "step": 220 }, { "epoch": 4.035087719298246, "grad_norm": 1.0086055994033813, "learning_rate": 1.929824561403509e-06, "loss": 0.1094, "step": 230 }, { "epoch": 4.2105263157894735, "grad_norm": 0.41842520236968994, "learning_rate": 1.5789473684210526e-06, "loss": 0.0748, "step": 240 }, { "epoch": 4.385964912280702, "grad_norm": 0.2806910574436188, "learning_rate": 1.2280701754385965e-06, "loss": 0.0563, "step": 250 }, { "epoch": 4.56140350877193, "grad_norm": 0.2728028893470764, "learning_rate": 8.771929824561404e-07, "loss": 0.055, "step": 260 }, { "epoch": 4.7368421052631575, "grad_norm": 0.2732146978378296, "learning_rate": 5.263157894736843e-07, "loss": 0.0434, "step": 270 }, { "epoch": 4.912280701754386, "grad_norm": 0.3774060010910034, "learning_rate": 1.7543859649122808e-07, "loss": 0.0962, "step": 280 }, { "epoch": 5.0, "step": 285, "total_flos": 7.020774259082035e+17, "train_loss": 0.15312039318837617, "train_runtime": 2880.1419, "train_samples_per_second": 3.167, "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 }