| { |
| "best_metric": 0.5306985974311829, |
| "best_model_checkpoint": "./vit-large-augmented-ph2-patch-16/checkpoint-50", |
| "epoch": 4.0, |
| "global_step": 552, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.07, |
| "learning_rate": 0.00019637681159420293, |
| "loss": 0.7939, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.14, |
| "learning_rate": 0.0001927536231884058, |
| "loss": 0.381, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.22, |
| "learning_rate": 0.0001891304347826087, |
| "loss": 0.4226, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.29, |
| "learning_rate": 0.00018586956521739133, |
| "loss": 0.4128, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.36, |
| "learning_rate": 0.00018224637681159422, |
| "loss": 0.2064, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.36, |
| "eval_accuracy": 0.8735042735042735, |
| "eval_f1": 0.8735042735042735, |
| "eval_loss": 0.5306985974311829, |
| "eval_precision": 0.8735042735042735, |
| "eval_recall": 0.8735042735042735, |
| "eval_runtime": 11.8033, |
| "eval_samples_per_second": 49.562, |
| "eval_steps_per_second": 6.269, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.43, |
| "learning_rate": 0.0001786231884057971, |
| "loss": 0.132, |
| "step": 60 |
| }, |
| { |
| "epoch": 0.51, |
| "learning_rate": 0.000175, |
| "loss": 0.1388, |
| "step": 70 |
| }, |
| { |
| "epoch": 0.58, |
| "learning_rate": 0.00017137681159420292, |
| "loss": 0.1381, |
| "step": 80 |
| }, |
| { |
| "epoch": 0.65, |
| "learning_rate": 0.0001677536231884058, |
| "loss": 0.1244, |
| "step": 90 |
| }, |
| { |
| "epoch": 0.72, |
| "learning_rate": 0.0001641304347826087, |
| "loss": 0.1145, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.72, |
| "eval_accuracy": 0.747008547008547, |
| "eval_f1": 0.747008547008547, |
| "eval_loss": 0.8837085962295532, |
| "eval_precision": 0.747008547008547, |
| "eval_recall": 0.747008547008547, |
| "eval_runtime": 11.7447, |
| "eval_samples_per_second": 49.81, |
| "eval_steps_per_second": 6.301, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.8, |
| "learning_rate": 0.0001605072463768116, |
| "loss": 0.2013, |
| "step": 110 |
| }, |
| { |
| "epoch": 0.87, |
| "learning_rate": 0.0001568840579710145, |
| "loss": 0.1908, |
| "step": 120 |
| }, |
| { |
| "epoch": 0.94, |
| "learning_rate": 0.0001532608695652174, |
| "loss": 0.1901, |
| "step": 130 |
| }, |
| { |
| "epoch": 1.01, |
| "learning_rate": 0.0001496376811594203, |
| "loss": 0.364, |
| "step": 140 |
| }, |
| { |
| "epoch": 1.09, |
| "learning_rate": 0.0001460144927536232, |
| "loss": 0.4187, |
| "step": 150 |
| }, |
| { |
| "epoch": 1.09, |
| "eval_accuracy": 0.6256410256410256, |
| "eval_f1": 0.6256410256410256, |
| "eval_loss": 0.9484676718711853, |
| "eval_precision": 0.6256410256410256, |
| "eval_recall": 0.6256410256410256, |
| "eval_runtime": 11.6896, |
| "eval_samples_per_second": 50.044, |
| "eval_steps_per_second": 6.33, |
| "step": 150 |
| }, |
| { |
| "epoch": 1.16, |
| "learning_rate": 0.00014239130434782612, |
| "loss": 0.2026, |
| "step": 160 |
| }, |
| { |
| "epoch": 1.23, |
| "learning_rate": 0.00013876811594202898, |
| "loss": 0.1367, |
| "step": 170 |
| }, |
| { |
| "epoch": 1.3, |
| "learning_rate": 0.0001351449275362319, |
| "loss": 0.0528, |
| "step": 180 |
| }, |
| { |
| "epoch": 1.38, |
| "learning_rate": 0.00013152173913043478, |
| "loss": 0.2144, |
| "step": 190 |
| }, |
| { |
| "epoch": 1.45, |
| "learning_rate": 0.00012789855072463767, |
| "loss": 0.0756, |
| "step": 200 |
| }, |
| { |
| "epoch": 1.45, |
| "eval_accuracy": 0.8324786324786325, |
| "eval_f1": 0.8324786324786324, |
| "eval_loss": 0.6959302425384521, |
| "eval_precision": 0.8324786324786325, |
| "eval_recall": 0.8324786324786325, |
| "eval_runtime": 12.2171, |
| "eval_samples_per_second": 47.884, |
| "eval_steps_per_second": 6.057, |
| "step": 200 |
| }, |
| { |
| "epoch": 1.52, |
| "learning_rate": 0.0001242753623188406, |
| "loss": 0.106, |
| "step": 210 |
| }, |
| { |
| "epoch": 1.59, |
| "learning_rate": 0.00012065217391304348, |
| "loss": 0.1342, |
| "step": 220 |
| }, |
| { |
| "epoch": 1.67, |
| "learning_rate": 0.0001170289855072464, |
| "loss": 0.1399, |
| "step": 230 |
| }, |
| { |
| "epoch": 1.74, |
| "learning_rate": 0.00011340579710144927, |
| "loss": 0.1476, |
| "step": 240 |
| }, |
| { |
| "epoch": 1.81, |
| "learning_rate": 0.00010978260869565218, |
| "loss": 0.0696, |
| "step": 250 |
| }, |
| { |
| "epoch": 1.81, |
| "eval_accuracy": 0.8170940170940171, |
| "eval_f1": 0.8170940170940171, |
| "eval_loss": 0.7697324156761169, |
| "eval_precision": 0.8170940170940171, |
| "eval_recall": 0.8170940170940171, |
| "eval_runtime": 11.717, |
| "eval_samples_per_second": 49.927, |
| "eval_steps_per_second": 6.316, |
| "step": 250 |
| }, |
| { |
| "epoch": 1.88, |
| "learning_rate": 0.00010615942028985508, |
| "loss": 0.2838, |
| "step": 260 |
| }, |
| { |
| "epoch": 1.96, |
| "learning_rate": 0.00010253623188405797, |
| "loss": 0.0627, |
| "step": 270 |
| }, |
| { |
| "epoch": 2.03, |
| "learning_rate": 9.891304347826087e-05, |
| "loss": 0.1012, |
| "step": 280 |
| }, |
| { |
| "epoch": 2.1, |
| "learning_rate": 9.528985507246378e-05, |
| "loss": 0.0562, |
| "step": 290 |
| }, |
| { |
| "epoch": 2.17, |
| "learning_rate": 9.166666666666667e-05, |
| "loss": 0.0251, |
| "step": 300 |
| }, |
| { |
| "epoch": 2.17, |
| "eval_accuracy": 0.8324786324786325, |
| "eval_f1": 0.8324786324786324, |
| "eval_loss": 0.7361409664154053, |
| "eval_precision": 0.8324786324786325, |
| "eval_recall": 0.8324786324786325, |
| "eval_runtime": 11.7369, |
| "eval_samples_per_second": 49.843, |
| "eval_steps_per_second": 6.305, |
| "step": 300 |
| }, |
| { |
| "epoch": 2.25, |
| "learning_rate": 8.804347826086957e-05, |
| "loss": 0.0332, |
| "step": 310 |
| }, |
| { |
| "epoch": 2.32, |
| "learning_rate": 8.442028985507246e-05, |
| "loss": 0.0024, |
| "step": 320 |
| }, |
| { |
| "epoch": 2.39, |
| "learning_rate": 8.079710144927536e-05, |
| "loss": 0.002, |
| "step": 330 |
| }, |
| { |
| "epoch": 2.46, |
| "learning_rate": 7.717391304347827e-05, |
| "loss": 0.0019, |
| "step": 340 |
| }, |
| { |
| "epoch": 2.54, |
| "learning_rate": 7.355072463768117e-05, |
| "loss": 0.0604, |
| "step": 350 |
| }, |
| { |
| "epoch": 2.54, |
| "eval_accuracy": 0.8427350427350427, |
| "eval_f1": 0.8427350427350426, |
| "eval_loss": 0.9344781637191772, |
| "eval_precision": 0.8427350427350427, |
| "eval_recall": 0.8427350427350427, |
| "eval_runtime": 11.7927, |
| "eval_samples_per_second": 49.607, |
| "eval_steps_per_second": 6.275, |
| "step": 350 |
| }, |
| { |
| "epoch": 2.61, |
| "learning_rate": 6.992753623188406e-05, |
| "loss": 0.0097, |
| "step": 360 |
| }, |
| { |
| "epoch": 2.68, |
| "learning_rate": 6.630434782608695e-05, |
| "loss": 0.0034, |
| "step": 370 |
| }, |
| { |
| "epoch": 2.75, |
| "learning_rate": 6.268115942028985e-05, |
| "loss": 0.0225, |
| "step": 380 |
| }, |
| { |
| "epoch": 2.83, |
| "learning_rate": 5.905797101449275e-05, |
| "loss": 0.0351, |
| "step": 390 |
| }, |
| { |
| "epoch": 2.9, |
| "learning_rate": 5.5434782608695654e-05, |
| "loss": 0.0005, |
| "step": 400 |
| }, |
| { |
| "epoch": 2.9, |
| "eval_accuracy": 0.8512820512820513, |
| "eval_f1": 0.8512820512820513, |
| "eval_loss": 0.9581409692764282, |
| "eval_precision": 0.8512820512820513, |
| "eval_recall": 0.8512820512820513, |
| "eval_runtime": 11.7794, |
| "eval_samples_per_second": 49.663, |
| "eval_steps_per_second": 6.282, |
| "step": 400 |
| }, |
| { |
| "epoch": 2.97, |
| "learning_rate": 5.181159420289855e-05, |
| "loss": 0.0011, |
| "step": 410 |
| }, |
| { |
| "epoch": 3.04, |
| "learning_rate": 4.818840579710145e-05, |
| "loss": 0.0008, |
| "step": 420 |
| }, |
| { |
| "epoch": 3.12, |
| "learning_rate": 4.456521739130435e-05, |
| "loss": 0.0004, |
| "step": 430 |
| }, |
| { |
| "epoch": 3.19, |
| "learning_rate": 4.094202898550725e-05, |
| "loss": 0.0011, |
| "step": 440 |
| }, |
| { |
| "epoch": 3.26, |
| "learning_rate": 3.7318840579710144e-05, |
| "loss": 0.0005, |
| "step": 450 |
| }, |
| { |
| "epoch": 3.26, |
| "eval_accuracy": 0.8444444444444444, |
| "eval_f1": 0.8444444444444444, |
| "eval_loss": 1.0674240589141846, |
| "eval_precision": 0.8444444444444444, |
| "eval_recall": 0.8444444444444444, |
| "eval_runtime": 11.718, |
| "eval_samples_per_second": 49.923, |
| "eval_steps_per_second": 6.315, |
| "step": 450 |
| }, |
| { |
| "epoch": 3.33, |
| "learning_rate": 3.369565217391305e-05, |
| "loss": 0.0004, |
| "step": 460 |
| }, |
| { |
| "epoch": 3.41, |
| "learning_rate": 3.0072463768115944e-05, |
| "loss": 0.001, |
| "step": 470 |
| }, |
| { |
| "epoch": 3.48, |
| "learning_rate": 2.6449275362318844e-05, |
| "loss": 0.0003, |
| "step": 480 |
| }, |
| { |
| "epoch": 3.55, |
| "learning_rate": 2.282608695652174e-05, |
| "loss": 0.0002, |
| "step": 490 |
| }, |
| { |
| "epoch": 3.62, |
| "learning_rate": 1.9202898550724637e-05, |
| "loss": 0.005, |
| "step": 500 |
| }, |
| { |
| "epoch": 3.62, |
| "eval_accuracy": 0.8564102564102564, |
| "eval_f1": 0.8564102564102564, |
| "eval_loss": 0.9463948607444763, |
| "eval_precision": 0.8564102564102564, |
| "eval_recall": 0.8564102564102564, |
| "eval_runtime": 11.7058, |
| "eval_samples_per_second": 49.975, |
| "eval_steps_per_second": 6.322, |
| "step": 500 |
| }, |
| { |
| "epoch": 3.7, |
| "learning_rate": 1.5579710144927537e-05, |
| "loss": 0.0002, |
| "step": 510 |
| }, |
| { |
| "epoch": 3.77, |
| "learning_rate": 1.1956521739130435e-05, |
| "loss": 0.0002, |
| "step": 520 |
| }, |
| { |
| "epoch": 3.84, |
| "learning_rate": 8.333333333333334e-06, |
| "loss": 0.0002, |
| "step": 530 |
| }, |
| { |
| "epoch": 3.91, |
| "learning_rate": 4.710144927536232e-06, |
| "loss": 0.0002, |
| "step": 540 |
| }, |
| { |
| "epoch": 3.99, |
| "learning_rate": 1.0869565217391306e-06, |
| "loss": 0.0002, |
| "step": 550 |
| }, |
| { |
| "epoch": 3.99, |
| "eval_accuracy": 0.8564102564102564, |
| "eval_f1": 0.8564102564102564, |
| "eval_loss": 0.9574871063232422, |
| "eval_precision": 0.8564102564102564, |
| "eval_recall": 0.8564102564102564, |
| "eval_runtime": 11.6893, |
| "eval_samples_per_second": 50.046, |
| "eval_steps_per_second": 6.331, |
| "step": 550 |
| }, |
| { |
| "epoch": 4.0, |
| "step": 552, |
| "total_flos": 2.410620782808269e+18, |
| "train_loss": 0.11269745317058287, |
| "train_runtime": 784.6004, |
| "train_samples_per_second": 11.216, |
| "train_steps_per_second": 0.704 |
| } |
| ], |
| "max_steps": 552, |
| "num_train_epochs": 4, |
| "total_flos": 2.410620782808269e+18, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
