{ "best_metric": 0.9777777777777777, "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-eurosat/checkpoint-456", "epoch": 3.0, "eval_steps": 500, "global_step": 456, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.06578947368421052, "grad_norm": 6.14052677154541, "learning_rate": 1.0869565217391305e-05, "loss": 2.2902, "step": 10 }, { "epoch": 0.13157894736842105, "grad_norm": 8.597524642944336, "learning_rate": 2.173913043478261e-05, "loss": 2.0361, "step": 20 }, { "epoch": 0.19736842105263158, "grad_norm": 8.122353553771973, "learning_rate": 3.260869565217392e-05, "loss": 1.5434, "step": 30 }, { "epoch": 0.2631578947368421, "grad_norm": 17.553142547607422, "learning_rate": 4.347826086956522e-05, "loss": 0.8428, "step": 40 }, { "epoch": 0.32894736842105265, "grad_norm": 17.103687286376953, "learning_rate": 4.951219512195122e-05, "loss": 0.5853, "step": 50 }, { "epoch": 0.39473684210526316, "grad_norm": 15.302396774291992, "learning_rate": 4.829268292682927e-05, "loss": 0.5033, "step": 60 }, { "epoch": 0.4605263157894737, "grad_norm": 14.38298225402832, "learning_rate": 4.707317073170732e-05, "loss": 0.4004, "step": 70 }, { "epoch": 0.5263157894736842, "grad_norm": 13.4431791305542, "learning_rate": 4.585365853658537e-05, "loss": 0.3834, "step": 80 }, { "epoch": 0.5921052631578947, "grad_norm": 13.142807960510254, "learning_rate": 4.4634146341463416e-05, "loss": 0.4049, "step": 90 }, { "epoch": 0.6578947368421053, "grad_norm": 11.127829551696777, "learning_rate": 4.3414634146341465e-05, "loss": 0.2982, "step": 100 }, { "epoch": 0.7236842105263158, "grad_norm": 7.5213117599487305, "learning_rate": 4.2195121951219514e-05, "loss": 0.3157, "step": 110 }, { "epoch": 0.7894736842105263, "grad_norm": 24.309080123901367, "learning_rate": 4.097560975609756e-05, "loss": 0.3069, "step": 120 }, { "epoch": 0.8552631578947368, "grad_norm": 8.259692192077637, "learning_rate": 3.975609756097561e-05, "loss": 0.2633, "step": 130 }, { "epoch": 0.9210526315789473, "grad_norm": 11.893393516540527, "learning_rate": 3.853658536585366e-05, "loss": 0.2774, "step": 140 }, { "epoch": 0.9868421052631579, "grad_norm": 11.049416542053223, "learning_rate": 3.731707317073171e-05, "loss": 0.3121, "step": 150 }, { "epoch": 1.0, "eval_accuracy": 0.9532407407407407, "eval_loss": 0.14279033243656158, "eval_runtime": 13.8728, "eval_samples_per_second": 155.701, "eval_steps_per_second": 4.902, "step": 152 }, { "epoch": 1.0526315789473684, "grad_norm": 13.241842269897461, "learning_rate": 3.609756097560976e-05, "loss": 0.2404, "step": 160 }, { "epoch": 1.118421052631579, "grad_norm": 14.350299835205078, "learning_rate": 3.48780487804878e-05, "loss": 0.2475, "step": 170 }, { "epoch": 1.1842105263157894, "grad_norm": 8.458578109741211, "learning_rate": 3.365853658536586e-05, "loss": 0.2055, "step": 180 }, { "epoch": 1.25, "grad_norm": 6.782774448394775, "learning_rate": 3.2439024390243906e-05, "loss": 0.2155, "step": 190 }, { "epoch": 1.3157894736842106, "grad_norm": 10.584575653076172, "learning_rate": 3.1219512195121955e-05, "loss": 0.2179, "step": 200 }, { "epoch": 1.381578947368421, "grad_norm": 11.664501190185547, "learning_rate": 3e-05, "loss": 0.2548, "step": 210 }, { "epoch": 1.4473684210526316, "grad_norm": 14.747825622558594, "learning_rate": 2.8780487804878046e-05, "loss": 0.2451, "step": 220 }, { "epoch": 1.513157894736842, "grad_norm": 6.113654613494873, "learning_rate": 2.7560975609756102e-05, "loss": 0.2114, "step": 230 }, { "epoch": 1.5789473684210527, "grad_norm": 10.554082870483398, "learning_rate": 2.6341463414634148e-05, "loss": 0.2353, "step": 240 }, { "epoch": 1.6447368421052633, "grad_norm": 12.61452579498291, "learning_rate": 2.5121951219512197e-05, "loss": 0.2162, "step": 250 }, { "epoch": 1.7105263157894737, "grad_norm": 11.650195121765137, "learning_rate": 2.3902439024390243e-05, "loss": 0.2318, "step": 260 }, { "epoch": 1.776315789473684, "grad_norm": 4.788726329803467, "learning_rate": 2.2682926829268295e-05, "loss": 0.2193, "step": 270 }, { "epoch": 1.8421052631578947, "grad_norm": 15.522054672241211, "learning_rate": 2.146341463414634e-05, "loss": 0.1898, "step": 280 }, { "epoch": 1.9078947368421053, "grad_norm": 9.786781311035156, "learning_rate": 2.0243902439024393e-05, "loss": 0.2029, "step": 290 }, { "epoch": 1.973684210526316, "grad_norm": 18.139461517333984, "learning_rate": 1.902439024390244e-05, "loss": 0.2403, "step": 300 }, { "epoch": 2.0, "eval_accuracy": 0.9652777777777778, "eval_loss": 0.09587351232767105, "eval_runtime": 13.2305, "eval_samples_per_second": 163.259, "eval_steps_per_second": 5.14, "step": 304 }, { "epoch": 2.039473684210526, "grad_norm": 7.916225910186768, "learning_rate": 1.7804878048780488e-05, "loss": 0.1847, "step": 310 }, { "epoch": 2.1052631578947367, "grad_norm": 6.741857051849365, "learning_rate": 1.6585365853658537e-05, "loss": 0.1565, "step": 320 }, { "epoch": 2.1710526315789473, "grad_norm": 9.491477012634277, "learning_rate": 1.5365853658536586e-05, "loss": 0.1449, "step": 330 }, { "epoch": 2.236842105263158, "grad_norm": 11.790963172912598, "learning_rate": 1.4146341463414633e-05, "loss": 0.1701, "step": 340 }, { "epoch": 2.3026315789473686, "grad_norm": 17.536788940429688, "learning_rate": 1.2926829268292684e-05, "loss": 0.1653, "step": 350 }, { "epoch": 2.3684210526315788, "grad_norm": 10.477115631103516, "learning_rate": 1.1707317073170733e-05, "loss": 0.1704, "step": 360 }, { "epoch": 2.4342105263157894, "grad_norm": 7.794534206390381, "learning_rate": 1.048780487804878e-05, "loss": 0.1763, "step": 370 }, { "epoch": 2.5, "grad_norm": 11.786982536315918, "learning_rate": 9.26829268292683e-06, "loss": 0.1705, "step": 380 }, { "epoch": 2.5657894736842106, "grad_norm": 8.611421585083008, "learning_rate": 8.048780487804879e-06, "loss": 0.1597, "step": 390 }, { "epoch": 2.6315789473684212, "grad_norm": 4.161876678466797, "learning_rate": 6.829268292682928e-06, "loss": 0.158, "step": 400 }, { "epoch": 2.6973684210526314, "grad_norm": 6.414247989654541, "learning_rate": 5.609756097560976e-06, "loss": 0.1447, "step": 410 }, { "epoch": 2.763157894736842, "grad_norm": 8.693388938903809, "learning_rate": 4.390243902439024e-06, "loss": 0.1711, "step": 420 }, { "epoch": 2.8289473684210527, "grad_norm": 9.579256057739258, "learning_rate": 3.1707317073170736e-06, "loss": 0.1379, "step": 430 }, { "epoch": 2.8947368421052633, "grad_norm": 7.8827080726623535, "learning_rate": 1.951219512195122e-06, "loss": 0.1847, "step": 440 }, { "epoch": 2.9605263157894735, "grad_norm": 14.064638137817383, "learning_rate": 7.317073170731708e-07, "loss": 0.1688, "step": 450 }, { "epoch": 3.0, "eval_accuracy": 0.9777777777777777, "eval_loss": 0.07008983194828033, "eval_runtime": 13.5293, "eval_samples_per_second": 159.653, "eval_steps_per_second": 5.026, "step": 456 }, { "epoch": 3.0, "step": 456, "total_flos": 1.4499253204608614e+18, "train_loss": 0.36645709030460893, "train_runtime": 903.5275, "train_samples_per_second": 64.547, "train_steps_per_second": 0.505 } ], "logging_steps": 10, "max_steps": 456, "num_input_tokens_seen": 0, "num_train_epochs": 3, "save_steps": 500, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": true }, "attributes": {} } }, "total_flos": 1.4499253204608614e+18, "train_batch_size": 32, "trial_name": null, "trial_params": null }