{ "best_metric": 0.9737037037037037, "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-eurosat/checkpoint-285", "epoch": 3.0, "eval_steps": 500, "global_step": 285, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.10526315789473684, "grad_norm": 5.289836406707764, "learning_rate": 1.7241379310344828e-05, "loss": 2.2434, "step": 10 }, { "epoch": 0.21052631578947367, "grad_norm": 5.927619457244873, "learning_rate": 3.4482758620689657e-05, "loss": 1.7979, "step": 20 }, { "epoch": 0.3157894736842105, "grad_norm": 8.896160125732422, "learning_rate": 4.9804687500000004e-05, "loss": 1.032, "step": 30 }, { "epoch": 0.42105263157894735, "grad_norm": 22.719310760498047, "learning_rate": 4.78515625e-05, "loss": 0.5576, "step": 40 }, { "epoch": 0.5263157894736842, "grad_norm": 13.316530227661133, "learning_rate": 4.58984375e-05, "loss": 0.4798, "step": 50 }, { "epoch": 0.631578947368421, "grad_norm": 13.046772003173828, "learning_rate": 4.3945312500000005e-05, "loss": 0.3895, "step": 60 }, { "epoch": 0.7368421052631579, "grad_norm": 10.398356437683105, "learning_rate": 4.1992187500000003e-05, "loss": 0.342, "step": 70 }, { "epoch": 0.8421052631578947, "grad_norm": 11.918193817138672, "learning_rate": 4.00390625e-05, "loss": 0.3263, "step": 80 }, { "epoch": 0.9473684210526315, "grad_norm": 7.351226329803467, "learning_rate": 3.80859375e-05, "loss": 0.281, "step": 90 }, { "epoch": 1.0, "eval_accuracy": 0.9562962962962963, "eval_loss": 0.15288086235523224, "eval_runtime": 19.6267, "eval_samples_per_second": 137.568, "eval_steps_per_second": 2.191, "step": 95 }, { "epoch": 1.0526315789473684, "grad_norm": 8.774168968200684, "learning_rate": 3.6132812500000005e-05, "loss": 0.2871, "step": 100 }, { "epoch": 1.1578947368421053, "grad_norm": 8.237377166748047, "learning_rate": 3.41796875e-05, "loss": 0.2664, "step": 110 }, { "epoch": 1.263157894736842, "grad_norm": 12.695280075073242, "learning_rate": 3.22265625e-05, "loss": 0.2378, "step": 120 }, { "epoch": 1.368421052631579, "grad_norm": 8.310792922973633, "learning_rate": 3.02734375e-05, "loss": 0.2349, "step": 130 }, { "epoch": 1.4736842105263157, "grad_norm": 10.032830238342285, "learning_rate": 2.83203125e-05, "loss": 0.2248, "step": 140 }, { "epoch": 1.5789473684210527, "grad_norm": 7.90118932723999, "learning_rate": 2.63671875e-05, "loss": 0.2076, "step": 150 }, { "epoch": 1.6842105263157894, "grad_norm": 15.789655685424805, "learning_rate": 2.44140625e-05, "loss": 0.1953, "step": 160 }, { "epoch": 1.7894736842105263, "grad_norm": 8.423171043395996, "learning_rate": 2.24609375e-05, "loss": 0.1949, "step": 170 }, { "epoch": 1.8947368421052633, "grad_norm": 10.328082084655762, "learning_rate": 2.05078125e-05, "loss": 0.1943, "step": 180 }, { "epoch": 2.0, "grad_norm": 14.849194526672363, "learning_rate": 1.85546875e-05, "loss": 0.1976, "step": 190 }, { "epoch": 2.0, "eval_accuracy": 0.9722222222222222, "eval_loss": 0.09291564673185349, "eval_runtime": 13.9085, "eval_samples_per_second": 194.125, "eval_steps_per_second": 3.092, "step": 190 }, { "epoch": 2.1052631578947367, "grad_norm": 9.324716567993164, "learning_rate": 1.66015625e-05, "loss": 0.1966, "step": 200 }, { "epoch": 2.2105263157894735, "grad_norm": 8.505390167236328, "learning_rate": 1.4648437500000001e-05, "loss": 0.2039, "step": 210 }, { "epoch": 2.3157894736842106, "grad_norm": 6.697166919708252, "learning_rate": 1.2695312500000001e-05, "loss": 0.1835, "step": 220 }, { "epoch": 2.4210526315789473, "grad_norm": 14.402400016784668, "learning_rate": 1.0742187500000001e-05, "loss": 0.1704, "step": 230 }, { "epoch": 2.526315789473684, "grad_norm": 8.63730239868164, "learning_rate": 8.789062500000001e-06, "loss": 0.1799, "step": 240 }, { "epoch": 2.6315789473684212, "grad_norm": 6.670950412750244, "learning_rate": 6.8359375e-06, "loss": 0.1442, "step": 250 }, { "epoch": 2.736842105263158, "grad_norm": 8.355303764343262, "learning_rate": 4.8828125e-06, "loss": 0.1708, "step": 260 }, { "epoch": 2.8421052631578947, "grad_norm": 6.931336879730225, "learning_rate": 2.9296875e-06, "loss": 0.1881, "step": 270 }, { "epoch": 2.9473684210526314, "grad_norm": 6.851834297180176, "learning_rate": 9.765625e-07, "loss": 0.1805, "step": 280 }, { "epoch": 3.0, "eval_accuracy": 0.9737037037037037, "eval_loss": 0.07975314557552338, "eval_runtime": 14.0525, "eval_samples_per_second": 192.136, "eval_steps_per_second": 3.06, "step": 285 }, { "epoch": 3.0, "step": 285, "total_flos": 1.8124066505760768e+18, "train_loss": 0.39992518445901704, "train_runtime": 819.9071, "train_samples_per_second": 88.913, "train_steps_per_second": 0.348 } ], "logging_steps": 10, "max_steps": 285, "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": 1.8124066505760768e+18, "train_batch_size": 64, "trial_name": null, "trial_params": null }