{ "best_metric": 0.11360067874193192, "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": 1.413887858390808, "learning_rate": 9.649122807017545e-06, "loss": 0.5957, "step": 10 }, { "epoch": 0.3508771929824561, "grad_norm": 1.629610300064087, "learning_rate": 9.298245614035088e-06, "loss": 0.4704, "step": 20 }, { "epoch": 0.5263157894736842, "grad_norm": 1.1828052997589111, "learning_rate": 8.947368421052632e-06, "loss": 0.3788, "step": 30 }, { "epoch": 0.7017543859649122, "grad_norm": 1.2675493955612183, "learning_rate": 8.596491228070176e-06, "loss": 0.2969, "step": 40 }, { "epoch": 0.8771929824561403, "grad_norm": 1.6942648887634277, "learning_rate": 8.24561403508772e-06, "loss": 0.2623, "step": 50 }, { "epoch": 1.0526315789473684, "grad_norm": 0.7355015277862549, "learning_rate": 7.894736842105265e-06, "loss": 0.1859, "step": 60 }, { "epoch": 1.2280701754385965, "grad_norm": 0.6509683132171631, "learning_rate": 7.5438596491228074e-06, "loss": 0.1712, "step": 70 }, { "epoch": 1.4035087719298245, "grad_norm": 0.6213608384132385, "learning_rate": 7.192982456140352e-06, "loss": 0.1466, "step": 80 }, { "epoch": 1.5789473684210527, "grad_norm": 0.5607940554618835, "learning_rate": 6.842105263157896e-06, "loss": 0.1181, "step": 90 }, { "epoch": 1.7543859649122808, "grad_norm": 2.426992416381836, "learning_rate": 6.491228070175439e-06, "loss": 0.1254, "step": 100 }, { "epoch": 1.7543859649122808, "eval_accuracy": 0.9691629955947136, "eval_confusion_matrix": [ [ 59, 7 ], [ 0, 161 ] ], "eval_loss": 0.15184630453586578, "eval_runtime": 24.5664, "eval_samples_per_second": 9.24, "eval_steps_per_second": 0.326, "step": 100 }, { "epoch": 1.9298245614035088, "grad_norm": 2.1420822143554688, "learning_rate": 6.140350877192983e-06, "loss": 0.1274, "step": 110 }, { "epoch": 2.1052631578947367, "grad_norm": 0.4402596950531006, "learning_rate": 5.789473684210527e-06, "loss": 0.0958, "step": 120 }, { "epoch": 2.280701754385965, "grad_norm": 0.3711773157119751, "learning_rate": 5.438596491228071e-06, "loss": 0.1151, "step": 130 }, { "epoch": 2.456140350877193, "grad_norm": 0.4589725732803345, "learning_rate": 5.087719298245615e-06, "loss": 0.0796, "step": 140 }, { "epoch": 2.6315789473684212, "grad_norm": 0.32596296072006226, "learning_rate": 4.736842105263158e-06, "loss": 0.0935, "step": 150 }, { "epoch": 2.807017543859649, "grad_norm": 0.3621123731136322, "learning_rate": 4.385964912280702e-06, "loss": 0.0838, "step": 160 }, { "epoch": 2.982456140350877, "grad_norm": 1.6324498653411865, "learning_rate": 4.035087719298246e-06, "loss": 0.085, "step": 170 }, { "epoch": 3.1578947368421053, "grad_norm": 0.31269893050193787, "learning_rate": 3.6842105263157896e-06, "loss": 0.0898, "step": 180 }, { "epoch": 3.3333333333333335, "grad_norm": 0.279706209897995, "learning_rate": 3.3333333333333333e-06, "loss": 0.0811, "step": 190 }, { "epoch": 3.5087719298245617, "grad_norm": 0.41368523240089417, "learning_rate": 2.9824561403508774e-06, "loss": 0.0894, "step": 200 }, { "epoch": 3.5087719298245617, "eval_accuracy": 0.9691629955947136, "eval_confusion_matrix": [ [ 60, 6 ], [ 1, 160 ] ], "eval_loss": 0.11360067874193192, "eval_runtime": 24.4641, "eval_samples_per_second": 9.279, "eval_steps_per_second": 0.327, "step": 200 }, { "epoch": 3.6842105263157894, "grad_norm": 0.2807004153728485, "learning_rate": 2.631578947368421e-06, "loss": 0.1037, "step": 210 }, { "epoch": 3.8596491228070176, "grad_norm": 0.51247239112854, "learning_rate": 2.280701754385965e-06, "loss": 0.0667, "step": 220 }, { "epoch": 4.035087719298246, "grad_norm": 0.9778783321380615, "learning_rate": 1.929824561403509e-06, "loss": 0.1093, "step": 230 }, { "epoch": 4.2105263157894735, "grad_norm": 0.43163198232650757, "learning_rate": 1.5789473684210526e-06, "loss": 0.0753, "step": 240 }, { "epoch": 4.385964912280702, "grad_norm": 0.2818666994571686, "learning_rate": 1.2280701754385965e-06, "loss": 0.0563, "step": 250 }, { "epoch": 4.56140350877193, "grad_norm": 0.26998862624168396, "learning_rate": 8.771929824561404e-07, "loss": 0.0546, "step": 260 }, { "epoch": 4.7368421052631575, "grad_norm": 0.26639774441719055, "learning_rate": 5.263157894736843e-07, "loss": 0.0431, "step": 270 }, { "epoch": 4.912280701754386, "grad_norm": 0.36992037296295166, "learning_rate": 1.7543859649122808e-07, "loss": 0.0964, "step": 280 }, { "epoch": 5.0, "step": 285, "total_flos": 7.020774259082035e+17, "train_loss": 0.15240513682365417, "train_runtime": 2873.599, "train_samples_per_second": 3.174, "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 }