| { | |
| "best_metric": 0.6982446363889663, | |
| "best_model_checkpoint": "vit-base-patch16-224-in21k-finetuned/checkpoint-1120", | |
| "epoch": 4.988864142538976, | |
| "eval_steps": 500, | |
| "global_step": 1120, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 4.464285714285715e-06, | |
| "loss": 1.0215, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 8.92857142857143e-06, | |
| "loss": 1.0059, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 1.3392857142857144e-05, | |
| "loss": 1.0021, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 1.785714285714286e-05, | |
| "loss": 1.0043, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 2.2321428571428575e-05, | |
| "loss": 0.975, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 2.6785714285714288e-05, | |
| "loss": 0.9459, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 3.125e-05, | |
| "loss": 0.9092, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 3.571428571428572e-05, | |
| "loss": 0.911, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 4.017857142857143e-05, | |
| "loss": 0.961, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 4.464285714285715e-05, | |
| "loss": 0.9321, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 4.910714285714286e-05, | |
| "loss": 0.9215, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 4.960317460317461e-05, | |
| "loss": 0.8922, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 4.910714285714286e-05, | |
| "loss": 0.9519, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 4.8611111111111115e-05, | |
| "loss": 0.8926, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 4.811507936507937e-05, | |
| "loss": 0.8787, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 4.761904761904762e-05, | |
| "loss": 0.8783, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 4.7123015873015876e-05, | |
| "loss": 0.8611, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 4.662698412698413e-05, | |
| "loss": 0.9162, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 4.613095238095239e-05, | |
| "loss": 0.8853, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 4.563492063492064e-05, | |
| "loss": 0.8687, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 4.5138888888888894e-05, | |
| "loss": 0.8379, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 4.464285714285715e-05, | |
| "loss": 0.8333, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6473948174979103, | |
| "eval_loss": 0.9670450687408447, | |
| "eval_runtime": 60.2016, | |
| "eval_samples_per_second": 119.233, | |
| "eval_steps_per_second": 3.737, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 4.41468253968254e-05, | |
| "loss": 0.8652, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 4.3650793650793655e-05, | |
| "loss": 0.8686, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 4.315476190476191e-05, | |
| "loss": 0.8604, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 4.265873015873016e-05, | |
| "loss": 0.8377, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 4.2162698412698416e-05, | |
| "loss": 0.8151, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 0.8818, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 4.117063492063492e-05, | |
| "loss": 0.8108, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "learning_rate": 4.067460317460318e-05, | |
| "loss": 0.8408, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 4.017857142857143e-05, | |
| "loss": 0.8801, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "learning_rate": 3.968253968253968e-05, | |
| "loss": 0.8194, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "learning_rate": 3.918650793650794e-05, | |
| "loss": 0.862, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "learning_rate": 3.8690476190476195e-05, | |
| "loss": 0.8165, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 3.8194444444444444e-05, | |
| "loss": 0.8342, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 3.76984126984127e-05, | |
| "loss": 0.8397, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "learning_rate": 3.7202380952380956e-05, | |
| "loss": 0.8318, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 3.6706349206349205e-05, | |
| "loss": 0.792, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 3.621031746031746e-05, | |
| "loss": 0.844, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 3.571428571428572e-05, | |
| "loss": 0.8179, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "learning_rate": 3.521825396825397e-05, | |
| "loss": 0.7968, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "learning_rate": 3.472222222222222e-05, | |
| "loss": 0.7903, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 3.422619047619048e-05, | |
| "loss": 0.7977, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "learning_rate": 3.3730158730158734e-05, | |
| "loss": 0.7972, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.6653663973251602, | |
| "eval_loss": 0.9123018383979797, | |
| "eval_runtime": 59.7062, | |
| "eval_samples_per_second": 120.222, | |
| "eval_steps_per_second": 3.768, | |
| "step": 449 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 3.3234126984126983e-05, | |
| "loss": 0.8381, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 3.273809523809524e-05, | |
| "loss": 0.7561, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "learning_rate": 3.2242063492063495e-05, | |
| "loss": 0.7124, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "learning_rate": 3.1746031746031745e-05, | |
| "loss": 0.7322, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "learning_rate": 3.125e-05, | |
| "loss": 0.7348, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "learning_rate": 3.075396825396826e-05, | |
| "loss": 0.7468, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "learning_rate": 3.0257936507936506e-05, | |
| "loss": 0.7106, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "learning_rate": 2.9761904761904762e-05, | |
| "loss": 0.7585, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "learning_rate": 2.9265873015873018e-05, | |
| "loss": 0.7383, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "learning_rate": 2.876984126984127e-05, | |
| "loss": 0.7238, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "learning_rate": 2.8273809523809523e-05, | |
| "loss": 0.7042, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.49, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.7076, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 2.54, | |
| "learning_rate": 2.7281746031746032e-05, | |
| "loss": 0.7441, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "learning_rate": 2.6785714285714288e-05, | |
| "loss": 0.6887, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 2.63, | |
| "learning_rate": 2.628968253968254e-05, | |
| "loss": 0.7216, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "learning_rate": 2.5793650793650796e-05, | |
| "loss": 0.71, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.72, | |
| "learning_rate": 2.529761904761905e-05, | |
| "loss": 0.7265, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "learning_rate": 2.4801587301587305e-05, | |
| "loss": 0.697, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 2.81, | |
| "learning_rate": 2.4305555555555558e-05, | |
| "loss": 0.7578, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "learning_rate": 2.380952380952381e-05, | |
| "loss": 0.6838, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "learning_rate": 2.3313492063492066e-05, | |
| "loss": 0.7082, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 2.94, | |
| "learning_rate": 2.281746031746032e-05, | |
| "loss": 0.6824, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 2.98, | |
| "learning_rate": 2.2321428571428575e-05, | |
| "loss": 0.667, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.6886319308999721, | |
| "eval_loss": 0.8676984906196594, | |
| "eval_runtime": 59.6434, | |
| "eval_samples_per_second": 120.349, | |
| "eval_steps_per_second": 3.772, | |
| "step": 673 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "learning_rate": 2.1825396825396827e-05, | |
| "loss": 0.6546, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 3.07, | |
| "learning_rate": 2.132936507936508e-05, | |
| "loss": 0.6349, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "learning_rate": 2.0833333333333336e-05, | |
| "loss": 0.6229, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 3.16, | |
| "learning_rate": 2.033730158730159e-05, | |
| "loss": 0.653, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 3.21, | |
| "learning_rate": 1.984126984126984e-05, | |
| "loss": 0.6433, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 3.25, | |
| "learning_rate": 1.9345238095238097e-05, | |
| "loss": 0.6291, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 3.3, | |
| "learning_rate": 1.884920634920635e-05, | |
| "loss": 0.6621, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 3.34, | |
| "learning_rate": 1.8353174603174602e-05, | |
| "loss": 0.6116, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 3.39, | |
| "learning_rate": 1.785714285714286e-05, | |
| "loss": 0.6441, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 3.43, | |
| "learning_rate": 1.736111111111111e-05, | |
| "loss": 0.642, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 3.47, | |
| "learning_rate": 1.6865079365079367e-05, | |
| "loss": 0.6129, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 3.52, | |
| "learning_rate": 1.636904761904762e-05, | |
| "loss": 0.5959, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 3.56, | |
| "learning_rate": 1.5873015873015872e-05, | |
| "loss": 0.6257, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 3.61, | |
| "learning_rate": 1.537698412698413e-05, | |
| "loss": 0.5942, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 3.65, | |
| "learning_rate": 1.4880952380952381e-05, | |
| "loss": 0.5848, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 3.7, | |
| "learning_rate": 1.4384920634920635e-05, | |
| "loss": 0.6071, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 3.74, | |
| "learning_rate": 1.388888888888889e-05, | |
| "loss": 0.6009, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 3.79, | |
| "learning_rate": 1.3392857142857144e-05, | |
| "loss": 0.6541, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 3.83, | |
| "learning_rate": 1.2896825396825398e-05, | |
| "loss": 0.6306, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 3.88, | |
| "learning_rate": 1.2400793650793652e-05, | |
| "loss": 0.6398, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 3.92, | |
| "learning_rate": 1.1904761904761905e-05, | |
| "loss": 0.6285, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 3.96, | |
| "learning_rate": 1.140873015873016e-05, | |
| "loss": 0.5729, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.6937865700752298, | |
| "eval_loss": 0.8486846685409546, | |
| "eval_runtime": 59.646, | |
| "eval_samples_per_second": 120.343, | |
| "eval_steps_per_second": 3.772, | |
| "step": 898 | |
| }, | |
| { | |
| "epoch": 4.01, | |
| "learning_rate": 1.0912698412698414e-05, | |
| "loss": 0.5634, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 4.05, | |
| "learning_rate": 1.0416666666666668e-05, | |
| "loss": 0.5553, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 4.1, | |
| "learning_rate": 9.92063492063492e-06, | |
| "loss": 0.5474, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 4.14, | |
| "learning_rate": 9.424603174603175e-06, | |
| "loss": 0.5299, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 4.19, | |
| "learning_rate": 8.92857142857143e-06, | |
| "loss": 0.5541, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 4.23, | |
| "learning_rate": 8.432539682539684e-06, | |
| "loss": 0.5341, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 4.28, | |
| "learning_rate": 7.936507936507936e-06, | |
| "loss": 0.5315, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 4.32, | |
| "learning_rate": 7.4404761904761905e-06, | |
| "loss": 0.5265, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 4.37, | |
| "learning_rate": 6.944444444444445e-06, | |
| "loss": 0.5063, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 4.41, | |
| "learning_rate": 6.448412698412699e-06, | |
| "loss": 0.53, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 4.45, | |
| "learning_rate": 5.9523809523809525e-06, | |
| "loss": 0.5322, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.5, | |
| "learning_rate": 5.456349206349207e-06, | |
| "loss": 0.5397, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 4.54, | |
| "learning_rate": 4.96031746031746e-06, | |
| "loss": 0.533, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 4.59, | |
| "learning_rate": 4.464285714285715e-06, | |
| "loss": 0.6067, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 4.63, | |
| "learning_rate": 3.968253968253968e-06, | |
| "loss": 0.568, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 4.68, | |
| "learning_rate": 3.4722222222222224e-06, | |
| "loss": 0.5507, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 4.72, | |
| "learning_rate": 2.9761904761904763e-06, | |
| "loss": 0.5993, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 4.77, | |
| "learning_rate": 2.48015873015873e-06, | |
| "loss": 0.5343, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 4.81, | |
| "learning_rate": 1.984126984126984e-06, | |
| "loss": 0.5301, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 4.86, | |
| "learning_rate": 1.4880952380952381e-06, | |
| "loss": 0.5446, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 4.9, | |
| "learning_rate": 9.92063492063492e-07, | |
| "loss": 0.5496, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 4.94, | |
| "learning_rate": 4.96031746031746e-07, | |
| "loss": 0.5198, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 4.99, | |
| "learning_rate": 0.0, | |
| "loss": 0.5347, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 4.99, | |
| "eval_accuracy": 0.6982446363889663, | |
| "eval_loss": 0.8523625135421753, | |
| "eval_runtime": 60.8207, | |
| "eval_samples_per_second": 118.019, | |
| "eval_steps_per_second": 3.699, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 4.99, | |
| "step": 1120, | |
| "total_flos": 1.1101379964762415e+19, | |
| "train_loss": 0.7277007077421461, | |
| "train_runtime": 3200.5261, | |
| "train_samples_per_second": 44.85, | |
| "train_steps_per_second": 0.35 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 1120, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "total_flos": 1.1101379964762415e+19, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |