{ "best_metric": 1.0, "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-eurosat/checkpoint-378", "epoch": 2.9822485207100593, "eval_steps": 500, "global_step": 378, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.08, "grad_norm": 12.153364181518555, "learning_rate": 1.3157894736842106e-05, "loss": 2.1615, "step": 10 }, { "epoch": 0.16, "grad_norm": 16.816362380981445, "learning_rate": 2.6315789473684212e-05, "loss": 1.8847, "step": 20 }, { "epoch": 0.24, "grad_norm": 18.271907806396484, "learning_rate": 3.9473684210526316e-05, "loss": 1.2194, "step": 30 }, { "epoch": 0.32, "grad_norm": 5.709053039550781, "learning_rate": 4.970588235294118e-05, "loss": 0.5036, "step": 40 }, { "epoch": 0.39, "grad_norm": 5.6657609939575195, "learning_rate": 4.823529411764706e-05, "loss": 0.1547, "step": 50 }, { "epoch": 0.47, "grad_norm": 4.165065288543701, "learning_rate": 4.6764705882352944e-05, "loss": 0.0731, "step": 60 }, { "epoch": 0.55, "grad_norm": 8.897403717041016, "learning_rate": 4.5294117647058826e-05, "loss": 0.0448, "step": 70 }, { "epoch": 0.63, "grad_norm": 2.7607221603393555, "learning_rate": 4.382352941176471e-05, "loss": 0.0227, "step": 80 }, { "epoch": 0.71, "grad_norm": 2.4922943115234375, "learning_rate": 4.235294117647059e-05, "loss": 0.0216, "step": 90 }, { "epoch": 0.79, "grad_norm": 1.6621469259262085, "learning_rate": 4.0882352941176474e-05, "loss": 0.0332, "step": 100 }, { "epoch": 0.87, "grad_norm": 6.23069953918457, "learning_rate": 3.9411764705882356e-05, "loss": 0.0277, "step": 110 }, { "epoch": 0.95, "grad_norm": 2.660862684249878, "learning_rate": 3.794117647058824e-05, "loss": 0.0194, "step": 120 }, { "epoch": 0.99, "eval_accuracy": 0.9994444444444445, "eval_loss": 0.002504603238776326, "eval_runtime": 853.0633, "eval_samples_per_second": 2.11, "eval_steps_per_second": 0.067, "step": 126 }, { "epoch": 1.03, "grad_norm": 3.3311333656311035, "learning_rate": 3.6470588235294114e-05, "loss": 0.0191, "step": 130 }, { "epoch": 1.1, "grad_norm": 1.546949028968811, "learning_rate": 3.5e-05, "loss": 0.0193, "step": 140 }, { "epoch": 1.18, "grad_norm": 2.039259910583496, "learning_rate": 3.352941176470588e-05, "loss": 0.0213, "step": 150 }, { "epoch": 1.26, "grad_norm": 2.8678247928619385, "learning_rate": 3.205882352941177e-05, "loss": 0.017, "step": 160 }, { "epoch": 1.34, "grad_norm": 1.6331835985183716, "learning_rate": 3.058823529411765e-05, "loss": 0.0221, "step": 170 }, { "epoch": 1.42, "grad_norm": 0.7777595520019531, "learning_rate": 2.9117647058823534e-05, "loss": 0.0206, "step": 180 }, { "epoch": 1.5, "grad_norm": 2.2770392894744873, "learning_rate": 2.7647058823529416e-05, "loss": 0.0203, "step": 190 }, { "epoch": 1.58, "grad_norm": 1.0933735370635986, "learning_rate": 2.6176470588235295e-05, "loss": 0.0233, "step": 200 }, { "epoch": 1.66, "grad_norm": 0.9536008238792419, "learning_rate": 2.4705882352941178e-05, "loss": 0.0139, "step": 210 }, { "epoch": 1.74, "grad_norm": 0.19179974496364594, "learning_rate": 2.323529411764706e-05, "loss": 0.007, "step": 220 }, { "epoch": 1.81, "grad_norm": 0.12005764245986938, "learning_rate": 2.1764705882352943e-05, "loss": 0.0105, "step": 230 }, { "epoch": 1.89, "grad_norm": 2.346966028213501, "learning_rate": 2.0294117647058825e-05, "loss": 0.0144, "step": 240 }, { "epoch": 1.97, "grad_norm": 1.2424992322921753, "learning_rate": 1.8823529411764708e-05, "loss": 0.0137, "step": 250 }, { "epoch": 2.0, "eval_accuracy": 0.9994444444444445, "eval_loss": 0.0009298253571614623, "eval_runtime": 31.8455, "eval_samples_per_second": 56.523, "eval_steps_per_second": 1.79, "step": 253 }, { "epoch": 2.05, "grad_norm": 1.2912344932556152, "learning_rate": 1.735294117647059e-05, "loss": 0.0119, "step": 260 }, { "epoch": 2.13, "grad_norm": 0.5564313530921936, "learning_rate": 1.588235294117647e-05, "loss": 0.006, "step": 270 }, { "epoch": 2.21, "grad_norm": 0.05754239112138748, "learning_rate": 1.4411764705882352e-05, "loss": 0.0068, "step": 280 }, { "epoch": 2.29, "grad_norm": 0.031185824424028397, "learning_rate": 1.2941176470588238e-05, "loss": 0.0063, "step": 290 }, { "epoch": 2.37, "grad_norm": 0.09058652073144913, "learning_rate": 1.1470588235294118e-05, "loss": 0.0119, "step": 300 }, { "epoch": 2.45, "grad_norm": 0.13693469762802124, "learning_rate": 1e-05, "loss": 0.0055, "step": 310 }, { "epoch": 2.52, "grad_norm": 2.508786201477051, "learning_rate": 8.529411764705883e-06, "loss": 0.01, "step": 320 }, { "epoch": 2.6, "grad_norm": 0.2263449877500534, "learning_rate": 7.058823529411765e-06, "loss": 0.0059, "step": 330 }, { "epoch": 2.68, "grad_norm": 0.004928161855787039, "learning_rate": 5.588235294117647e-06, "loss": 0.0064, "step": 340 }, { "epoch": 2.76, "grad_norm": 3.029522180557251, "learning_rate": 4.11764705882353e-06, "loss": 0.011, "step": 350 }, { "epoch": 2.84, "grad_norm": 0.02349434234201908, "learning_rate": 2.647058823529412e-06, "loss": 0.0114, "step": 360 }, { "epoch": 2.92, "grad_norm": 1.1028828620910645, "learning_rate": 1.1764705882352942e-06, "loss": 0.0075, "step": 370 }, { "epoch": 2.98, "eval_accuracy": 1.0, "eval_loss": 9.558848432789091e-06, "eval_runtime": 31.5287, "eval_samples_per_second": 57.091, "eval_steps_per_second": 1.808, "step": 378 }, { "epoch": 2.98, "step": 378, "total_flos": 1.2014751945240576e+18, "train_loss": 0.17176837376540616, "train_runtime": 8690.1401, "train_samples_per_second": 5.591, "train_steps_per_second": 0.043 } ], "logging_steps": 10, "max_steps": 378, "num_input_tokens_seen": 0, "num_train_epochs": 3, "save_steps": 500, "total_flos": 1.2014751945240576e+18, "train_batch_size": 32, "trial_name": null, "trial_params": null }