Spaces:
Sleeping
Sleeping
File size: 48,264 Bytes
0b422fb |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 |
[i] ResNet50@Puzzle@optimal
[i] mean values is [0.485, 0.456, 0.406]
[i] std values is [0.229, 0.224, 0.225]
[i] The number of class is 20
[i] train_transform is Compose(
<tools.ai.augment_utils.RandomResize object at 0x7b9d81d50520>
<tools.ai.augment_utils.RandomHorizontalFlip object at 0x7b9d81d50490>
<tools.ai.augment_utils.Normalize object at 0x7b9d81d50730>
<tools.ai.augment_utils.RandomCrop object at 0x7b9d81d503d0>
<tools.ai.augment_utils.Transpose object at 0x7b9d81d50430>
)
[i] test_transform is Compose(
<tools.ai.augment_utils.Normalize_For_Segmentation object at 0x7b9d81d502b0>
<tools.ai.augment_utils.Top_Left_Crop_For_Segmentation object at 0x7b9d81d500d0>
<tools.ai.augment_utils.Transpose_For_Segmentation object at 0x7b9d81d50100>
)
[i] log_iteration : 66
[i] val_iteration : 661
[i] max_iteration : 9,915
[i] Architecture is resnet50
[i] Total Params: 23.55M
[i] The number of pretrained weights : 106
[i] The number of pretrained bias : 53
[i] The number of scratched weights : 1
[i] The number of scratched bias : 0
[i] iteration=66, learning_rate=0.0994, alpha=0.03, loss=0.5450, class_loss=0.2720, p_class_loss=0.2716, re_loss=0.0728, conf_loss=0.0000, time=17sec
[i] iteration=132, learning_rate=0.0988, alpha=0.08, loss=0.3240, class_loss=0.1571, p_class_loss=0.1604, re_loss=0.0800, conf_loss=0.0000, time=15sec
[i] iteration=198, learning_rate=0.0982, alpha=0.13, loss=0.2655, class_loss=0.1247, p_class_loss=0.1286, re_loss=0.0920, conf_loss=0.0000, time=15sec
[i] iteration=264, learning_rate=0.0976, alpha=0.19, loss=0.2734, class_loss=0.1248, p_class_loss=0.1296, re_loss=0.1020, conf_loss=0.0000, time=15sec
[i] iteration=330, learning_rate=0.0970, alpha=0.24, loss=0.2663, class_loss=0.1184, p_class_loss=0.1237, re_loss=0.1012, conf_loss=0.0000, time=15sec
[i] iteration=396, learning_rate=0.0964, alpha=0.29, loss=0.2637, class_loss=0.1147, p_class_loss=0.1193, re_loss=0.1016, conf_loss=0.0000, time=15sec
[i] iteration=462, learning_rate=0.0958, alpha=0.35, loss=0.2497, class_loss=0.1052, p_class_loss=0.1096, re_loss=0.1011, conf_loss=0.0000, time=15sec
[i] iteration=528, learning_rate=0.0952, alpha=0.40, loss=0.2480, class_loss=0.1024, p_class_loss=0.1073, re_loss=0.0960, conf_loss=0.0000, time=15sec
[i] iteration=594, learning_rate=0.0946, alpha=0.45, loss=0.2408, class_loss=0.0982, p_class_loss=0.1023, re_loss=0.0892, conf_loss=0.0000, time=15sec
[i] iteration=660, learning_rate=0.0940, alpha=0.51, loss=0.2340, class_loss=0.0930, p_class_loss=0.0967, re_loss=0.0879, conf_loss=0.0000, time=15sec
[i] save model
[i] iteration=661, threshold=0.10, train_mIoU=40.43%, best_train_mIoU=40.43%, time=18sec
[i] iteration=726, learning_rate=0.0934, alpha=0.56, loss=0.2258, class_loss=0.0862, p_class_loss=0.0908, re_loss=0.0873, conf_loss=0.0000, time=34sec
[i] iteration=792, learning_rate=0.0928, alpha=0.61, loss=0.2239, class_loss=0.0842, p_class_loss=0.0881, re_loss=0.0844, conf_loss=0.0000, time=15sec
[i] iteration=858, learning_rate=0.0922, alpha=0.67, loss=0.2513, class_loss=0.0957, p_class_loss=0.0998, re_loss=0.0839, conf_loss=0.0000, time=15sec
[i] iteration=924, learning_rate=0.0916, alpha=0.72, loss=0.2326, class_loss=0.0871, p_class_loss=0.0914, re_loss=0.0753, conf_loss=0.0000, time=15sec
[i] iteration=990, learning_rate=0.0910, alpha=0.77, loss=0.2415, class_loss=0.0887, p_class_loss=0.0934, re_loss=0.0771, conf_loss=0.0000, time=15sec
[i] iteration=1,056, learning_rate=0.0904, alpha=0.83, loss=0.2436, class_loss=0.0887, p_class_loss=0.0931, re_loss=0.0749, conf_loss=0.0000, time=15sec
[i] iteration=1,122, learning_rate=0.0898, alpha=0.88, loss=0.2431, class_loss=0.0885, p_class_loss=0.0923, re_loss=0.0709, conf_loss=0.0000, time=15sec
[i] iteration=1,188, learning_rate=0.0892, alpha=0.93, loss=0.2383, class_loss=0.0846, p_class_loss=0.0887, re_loss=0.0698, conf_loss=0.0000, time=15sec
[i] iteration=1,254, learning_rate=0.0886, alpha=0.98, loss=0.2512, class_loss=0.0905, p_class_loss=0.0949, re_loss=0.0668, conf_loss=0.0000, time=15sec
[i] iteration=1,320, learning_rate=0.0879, alpha=1.04, loss=0.2436, class_loss=0.0860, p_class_loss=0.0902, re_loss=0.0649, conf_loss=0.0000, time=15sec
[i] save model
[i] iteration=1,322, threshold=0.10, train_mIoU=43.46%, best_train_mIoU=43.46%, time=18sec
[i] iteration=1,386, learning_rate=0.0873, alpha=1.09, loss=0.2324, class_loss=0.0796, p_class_loss=0.0843, re_loss=0.0628, conf_loss=0.0000, time=34sec
[i] iteration=1,452, learning_rate=0.0867, alpha=1.14, loss=0.2405, class_loss=0.0829, p_class_loss=0.0869, re_loss=0.0618, conf_loss=0.0000, time=15sec
[i] iteration=1,518, learning_rate=0.0861, alpha=1.20, loss=0.2409, class_loss=0.0820, p_class_loss=0.0863, re_loss=0.0606, conf_loss=0.0000, time=15sec
[i] iteration=1,584, learning_rate=0.0855, alpha=1.25, loss=0.2384, class_loss=0.0817, p_class_loss=0.0860, re_loss=0.0564, conf_loss=0.0000, time=15sec
[i] iteration=1,650, learning_rate=0.0849, alpha=1.30, loss=0.2454, class_loss=0.0832, p_class_loss=0.0869, re_loss=0.0578, conf_loss=0.0000, time=15sec
[i] iteration=1,716, learning_rate=0.0843, alpha=1.36, loss=0.2452, class_loss=0.0819, p_class_loss=0.0863, re_loss=0.0568, conf_loss=0.0000, time=15sec
[i] iteration=1,782, learning_rate=0.0837, alpha=1.41, loss=0.2589, class_loss=0.0876, p_class_loss=0.0915, re_loss=0.0565, conf_loss=0.0000, time=15sec
[i] iteration=1,848, learning_rate=0.0831, alpha=1.46, loss=0.2510, class_loss=0.0851, p_class_loss=0.0891, re_loss=0.0525, conf_loss=0.0000, time=15sec
[i] iteration=1,914, learning_rate=0.0825, alpha=1.52, loss=0.2478, class_loss=0.0831, p_class_loss=0.0872, re_loss=0.0511, conf_loss=0.0000, time=15sec
[i] iteration=1,980, learning_rate=0.0818, alpha=1.57, loss=0.2568, class_loss=0.0858, p_class_loss=0.0899, re_loss=0.0516, conf_loss=0.0000, time=15sec
[i] iteration=1,983, threshold=0.10, train_mIoU=43.18%, best_train_mIoU=43.46%, time=17sec
[i] iteration=2,046, learning_rate=0.0812, alpha=1.62, loss=0.2431, class_loss=0.0796, p_class_loss=0.0835, re_loss=0.0493, conf_loss=0.0000, time=34sec
[i] iteration=2,112, learning_rate=0.0806, alpha=1.68, loss=0.2583, class_loss=0.0855, p_class_loss=0.0897, re_loss=0.0496, conf_loss=0.0000, time=15sec
[i] iteration=2,178, learning_rate=0.0800, alpha=1.73, loss=0.2435, class_loss=0.0797, p_class_loss=0.0837, re_loss=0.0463, conf_loss=0.0000, time=15sec
[i] iteration=2,244, learning_rate=0.0794, alpha=1.78, loss=0.2551, class_loss=0.0834, p_class_loss=0.0877, re_loss=0.0471, conf_loss=0.0000, time=15sec
[i] iteration=2,310, learning_rate=0.0788, alpha=1.84, loss=0.2540, class_loss=0.0833, p_class_loss=0.0879, re_loss=0.0450, conf_loss=0.0000, time=15sec
[i] iteration=2,376, learning_rate=0.0782, alpha=1.89, loss=0.2583, class_loss=0.0841, p_class_loss=0.0890, re_loss=0.0451, conf_loss=0.0000, time=15sec
[i] iteration=2,442, learning_rate=0.0775, alpha=1.94, loss=0.2598, class_loss=0.0858, p_class_loss=0.0898, re_loss=0.0433, conf_loss=0.0000, time=15sec
[i] iteration=2,508, learning_rate=0.0769, alpha=2.00, loss=0.2635, class_loss=0.0871, p_class_loss=0.0911, re_loss=0.0427, conf_loss=0.0000, time=15sec
[i] iteration=2,574, learning_rate=0.0763, alpha=2.05, loss=0.2546, class_loss=0.0815, p_class_loss=0.0855, re_loss=0.0427, conf_loss=0.0000, time=15sec
[i] iteration=2,640, learning_rate=0.0757, alpha=2.10, loss=0.2583, class_loss=0.0855, p_class_loss=0.0894, re_loss=0.0396, conf_loss=0.0000, time=15sec
[i] save model
[i] iteration=2,644, threshold=0.10, train_mIoU=44.51%, best_train_mIoU=44.51%, time=18sec
[i] iteration=2,706, learning_rate=0.0751, alpha=2.16, loss=0.2611, class_loss=0.0843, p_class_loss=0.0881, re_loss=0.0411, conf_loss=0.0000, time=34sec
[i] iteration=2,772, learning_rate=0.0745, alpha=2.21, loss=0.2525, class_loss=0.0810, p_class_loss=0.0847, re_loss=0.0393, conf_loss=0.0000, time=15sec
[i] iteration=2,838, learning_rate=0.0738, alpha=2.26, loss=0.2540, class_loss=0.0816, p_class_loss=0.0857, re_loss=0.0383, conf_loss=0.0000, time=15sec
[i] iteration=2,904, learning_rate=0.0732, alpha=2.32, loss=0.2652, class_loss=0.0844, p_class_loss=0.0884, re_loss=0.0399, conf_loss=0.0000, time=15sec
[i] iteration=2,970, learning_rate=0.0726, alpha=2.37, loss=0.2607, class_loss=0.0836, p_class_loss=0.0875, re_loss=0.0378, conf_loss=0.0000, time=15sec
[i] iteration=3,036, learning_rate=0.0720, alpha=2.42, loss=0.2755, class_loss=0.0893, p_class_loss=0.0935, re_loss=0.0383, conf_loss=0.0000, time=15sec
[i] iteration=3,102, learning_rate=0.0714, alpha=2.48, loss=0.2690, class_loss=0.0872, p_class_loss=0.0913, re_loss=0.0366, conf_loss=0.0000, time=15sec
[i] iteration=3,168, learning_rate=0.0707, alpha=2.53, loss=0.2728, class_loss=0.0901, p_class_loss=0.0939, re_loss=0.0351, conf_loss=0.0000, time=15sec
[i] iteration=3,234, learning_rate=0.0701, alpha=2.58, loss=0.2591, class_loss=0.0822, p_class_loss=0.0865, re_loss=0.0350, conf_loss=0.0000, time=15sec
[i] iteration=3,300, learning_rate=0.0695, alpha=2.64, loss=0.2650, class_loss=0.0854, p_class_loss=0.0892, re_loss=0.0343, conf_loss=0.0000, time=15sec
[i] save model
[i] iteration=3,305, threshold=0.10, train_mIoU=45.10%, best_train_mIoU=45.10%, time=18sec
[i] iteration=3,366, learning_rate=0.0689, alpha=2.69, loss=0.2514, class_loss=0.0783, p_class_loss=0.0823, re_loss=0.0338, conf_loss=0.0000, time=34sec
[i] iteration=3,432, learning_rate=0.0682, alpha=2.74, loss=0.2762, class_loss=0.0893, p_class_loss=0.0933, re_loss=0.0341, conf_loss=0.0000, time=15sec
[i] iteration=3,498, learning_rate=0.0676, alpha=2.80, loss=0.2591, class_loss=0.0813, p_class_loss=0.0847, re_loss=0.0333, conf_loss=0.0000, time=15sec
[i] iteration=3,564, learning_rate=0.0670, alpha=2.85, loss=0.2858, class_loss=0.0919, p_class_loss=0.0962, re_loss=0.0343, conf_loss=0.0000, time=15sec
[i] iteration=3,630, learning_rate=0.0664, alpha=2.90, loss=0.2693, class_loss=0.0856, p_class_loss=0.0902, re_loss=0.0322, conf_loss=0.0000, time=15sec
[i] iteration=3,696, learning_rate=0.0657, alpha=2.96, loss=0.2648, class_loss=0.0845, p_class_loss=0.0881, re_loss=0.0312, conf_loss=0.0000, time=15sec
[i] iteration=3,762, learning_rate=0.0651, alpha=3.01, loss=0.2704, class_loss=0.0865, p_class_loss=0.0907, re_loss=0.0310, conf_loss=0.0000, time=15sec
[i] iteration=3,828, learning_rate=0.0645, alpha=3.06, loss=0.2752, class_loss=0.0883, p_class_loss=0.0921, re_loss=0.0310, conf_loss=0.0000, time=15sec
[i] iteration=3,894, learning_rate=0.0638, alpha=3.11, loss=0.2768, class_loss=0.0883, p_class_loss=0.0921, re_loss=0.0309, conf_loss=0.0000, time=15sec
[i] iteration=3,960, learning_rate=0.0632, alpha=3.17, loss=0.2772, class_loss=0.0894, p_class_loss=0.0932, re_loss=0.0299, conf_loss=0.0000, time=15sec
[i] save model
[i] iteration=3,966, threshold=0.10, train_mIoU=46.22%, best_train_mIoU=46.22%, time=18sec
[i] iteration=4,026, learning_rate=0.0626, alpha=3.22, loss=0.2772, class_loss=0.0880, p_class_loss=0.0916, re_loss=0.0303, conf_loss=0.0000, time=34sec
[i] iteration=4,092, learning_rate=0.0619, alpha=3.27, loss=0.2782, class_loss=0.0884, p_class_loss=0.0925, re_loss=0.0297, conf_loss=0.0000, time=15sec
[i] iteration=4,158, learning_rate=0.0613, alpha=3.33, loss=0.2727, class_loss=0.0879, p_class_loss=0.0918, re_loss=0.0280, conf_loss=0.0000, time=15sec
[i] iteration=4,224, learning_rate=0.0607, alpha=3.38, loss=0.2739, class_loss=0.0872, p_class_loss=0.0913, re_loss=0.0282, conf_loss=0.0000, time=15sec
[i] iteration=4,290, learning_rate=0.0601, alpha=3.43, loss=0.2813, class_loss=0.0892, p_class_loss=0.0932, re_loss=0.0288, conf_loss=0.0000, time=15sec
[i] iteration=4,356, learning_rate=0.0594, alpha=3.49, loss=0.2800, class_loss=0.0912, p_class_loss=0.0951, re_loss=0.0269, conf_loss=0.0000, time=15sec
[i] iteration=4,422, learning_rate=0.0588, alpha=3.54, loss=0.2824, class_loss=0.0909, p_class_loss=0.0948, re_loss=0.0273, conf_loss=0.0000, time=15sec
[i] iteration=4,488, learning_rate=0.0581, alpha=3.59, loss=0.2726, class_loss=0.0877, p_class_loss=0.0913, re_loss=0.0261, conf_loss=0.0000, time=15sec
[i] iteration=4,554, learning_rate=0.0575, alpha=3.65, loss=0.2724, class_loss=0.0873, p_class_loss=0.0913, re_loss=0.0257, conf_loss=0.0000, time=15sec
[i] iteration=4,620, learning_rate=0.0569, alpha=3.70, loss=0.2819, class_loss=0.0911, p_class_loss=0.0948, re_loss=0.0259, conf_loss=0.0000, time=15sec
[i] iteration=4,627, threshold=0.10, train_mIoU=45.99%, best_train_mIoU=46.22%, time=17sec
[i] iteration=4,686, learning_rate=0.0562, alpha=3.75, loss=0.2867, class_loss=0.0922, p_class_loss=0.0960, re_loss=0.0262, conf_loss=0.0000, time=34sec
[i] iteration=4,752, learning_rate=0.0556, alpha=3.81, loss=0.2750, class_loss=0.0880, p_class_loss=0.0918, re_loss=0.0250, conf_loss=0.0000, time=15sec
[i] iteration=4,818, learning_rate=0.0550, alpha=3.86, loss=0.2699, class_loss=0.0854, p_class_loss=0.0889, re_loss=0.0248, conf_loss=0.0000, time=15sec
[i] iteration=4,884, learning_rate=0.0543, alpha=3.91, loss=0.2864, class_loss=0.0915, p_class_loss=0.0951, re_loss=0.0255, conf_loss=0.0000, time=15sec
[i] iteration=4,950, learning_rate=0.0537, alpha=3.97, loss=0.2827, class_loss=0.0912, p_class_loss=0.0949, re_loss=0.0243, conf_loss=0.0000, time=15sec
[i] iteration=5,016, learning_rate=0.0530, alpha=4.00, loss=0.2900, class_loss=0.0932, p_class_loss=0.0970, re_loss=0.0249, conf_loss=0.0000, time=15sec
[i] iteration=5,082, learning_rate=0.0524, alpha=4.00, loss=0.2810, class_loss=0.0896, p_class_loss=0.0935, re_loss=0.0245, conf_loss=0.0000, time=15sec
[i] iteration=5,148, learning_rate=0.0517, alpha=4.00, loss=0.2826, class_loss=0.0911, p_class_loss=0.0947, re_loss=0.0242, conf_loss=0.0000, time=15sec
[i] iteration=5,214, learning_rate=0.0511, alpha=4.00, loss=0.2820, class_loss=0.0909, p_class_loss=0.0948, re_loss=0.0241, conf_loss=0.0000, time=15sec
[i] iteration=5,280, learning_rate=0.0505, alpha=4.00, loss=0.2832, class_loss=0.0922, p_class_loss=0.0959, re_loss=0.0238, conf_loss=0.0000, time=15sec
[i] iteration=5,288, threshold=0.10, train_mIoU=45.94%, best_train_mIoU=46.22%, time=18sec
[i] iteration=5,346, learning_rate=0.0498, alpha=4.00, loss=0.2745, class_loss=0.0878, p_class_loss=0.0916, re_loss=0.0238, conf_loss=0.0000, time=34sec
[i] iteration=5,412, learning_rate=0.0492, alpha=4.00, loss=0.2867, class_loss=0.0933, p_class_loss=0.0971, re_loss=0.0241, conf_loss=0.0000, time=15sec
[i] iteration=5,478, learning_rate=0.0485, alpha=4.00, loss=0.2826, class_loss=0.0931, p_class_loss=0.0968, re_loss=0.0232, conf_loss=0.0000, time=15sec
[i] iteration=5,544, learning_rate=0.0479, alpha=4.00, loss=0.2708, class_loss=0.0865, p_class_loss=0.0905, re_loss=0.0234, conf_loss=0.0000, time=15sec
[i] iteration=5,610, learning_rate=0.0472, alpha=4.00, loss=0.2724, class_loss=0.0871, p_class_loss=0.0911, re_loss=0.0236, conf_loss=0.0000, time=15sec
[i] iteration=5,676, learning_rate=0.0466, alpha=4.00, loss=0.2745, class_loss=0.0871, p_class_loss=0.0910, re_loss=0.0241, conf_loss=0.0000, time=15sec
[i] iteration=5,742, learning_rate=0.0459, alpha=4.00, loss=0.2714, class_loss=0.0877, p_class_loss=0.0913, re_loss=0.0231, conf_loss=0.0000, time=15sec
[i] iteration=5,808, learning_rate=0.0452, alpha=4.00, loss=0.2797, class_loss=0.0911, p_class_loss=0.0949, re_loss=0.0234, conf_loss=0.0000, time=15sec
[i] iteration=5,874, learning_rate=0.0446, alpha=4.00, loss=0.2769, class_loss=0.0914, p_class_loss=0.0949, re_loss=0.0227, conf_loss=0.0000, time=15sec
[i] iteration=5,940, learning_rate=0.0439, alpha=4.00, loss=0.2750, class_loss=0.0889, p_class_loss=0.0927, re_loss=0.0233, conf_loss=0.0000, time=15sec
[i] iteration=5,949, threshold=0.10, train_mIoU=46.19%, best_train_mIoU=46.22%, time=17sec
[i] iteration=6,006, learning_rate=0.0433, alpha=4.00, loss=0.2710, class_loss=0.0880, p_class_loss=0.0917, re_loss=0.0228, conf_loss=0.0000, time=34sec
[i] iteration=6,072, learning_rate=0.0426, alpha=4.00, loss=0.2694, class_loss=0.0884, p_class_loss=0.0919, re_loss=0.0223, conf_loss=0.0000, time=15sec
[i] iteration=6,138, learning_rate=0.0420, alpha=4.00, loss=0.2777, class_loss=0.0895, p_class_loss=0.0933, re_loss=0.0237, conf_loss=0.0000, time=15sec
[i] iteration=6,204, learning_rate=0.0413, alpha=4.00, loss=0.2700, class_loss=0.0880, p_class_loss=0.0919, re_loss=0.0225, conf_loss=0.0000, time=15sec
[i] iteration=6,270, learning_rate=0.0406, alpha=4.00, loss=0.2749, class_loss=0.0894, p_class_loss=0.0929, re_loss=0.0231, conf_loss=0.0000, time=15sec
[i] iteration=6,336, learning_rate=0.0400, alpha=4.00, loss=0.2713, class_loss=0.0875, p_class_loss=0.0914, re_loss=0.0231, conf_loss=0.0000, time=15sec
[i] iteration=6,402, learning_rate=0.0393, alpha=4.00, loss=0.2701, class_loss=0.0871, p_class_loss=0.0906, re_loss=0.0231, conf_loss=0.0000, time=15sec
[i] iteration=6,468, learning_rate=0.0386, alpha=4.00, loss=0.2666, class_loss=0.0865, p_class_loss=0.0901, re_loss=0.0225, conf_loss=0.0000, time=15sec
[i] iteration=6,534, learning_rate=0.0380, alpha=4.00, loss=0.2677, class_loss=0.0849, p_class_loss=0.0887, re_loss=0.0235, conf_loss=0.0000, time=15sec
[i] iteration=6,600, learning_rate=0.0373, alpha=4.00, loss=0.2730, class_loss=0.0907, p_class_loss=0.0944, re_loss=0.0220, conf_loss=0.0000, time=15sec
[i] iteration=6,610, threshold=0.10, train_mIoU=45.70%, best_train_mIoU=46.22%, time=18sec
[i] iteration=6,666, learning_rate=0.0366, alpha=4.00, loss=0.2714, class_loss=0.0878, p_class_loss=0.0917, re_loss=0.0230, conf_loss=0.0000, time=34sec
[i] iteration=6,732, learning_rate=0.0360, alpha=4.00, loss=0.2584, class_loss=0.0832, p_class_loss=0.0867, re_loss=0.0222, conf_loss=0.0000, time=15sec
[i] iteration=6,798, learning_rate=0.0353, alpha=4.00, loss=0.2658, class_loss=0.0863, p_class_loss=0.0902, re_loss=0.0223, conf_loss=0.0000, time=15sec
[i] iteration=6,864, learning_rate=0.0346, alpha=4.00, loss=0.2640, class_loss=0.0841, p_class_loss=0.0881, re_loss=0.0229, conf_loss=0.0000, time=15sec
[i] iteration=6,930, learning_rate=0.0340, alpha=4.00, loss=0.2637, class_loss=0.0854, p_class_loss=0.0890, re_loss=0.0223, conf_loss=0.0000, time=15sec
[i] iteration=6,996, learning_rate=0.0333, alpha=4.00, loss=0.2721, class_loss=0.0877, p_class_loss=0.0917, re_loss=0.0232, conf_loss=0.0000, time=15sec
[i] iteration=7,062, learning_rate=0.0326, alpha=4.00, loss=0.2568, class_loss=0.0822, p_class_loss=0.0862, re_loss=0.0221, conf_loss=0.0000, time=15sec
[i] iteration=7,128, learning_rate=0.0319, alpha=4.00, loss=0.2660, class_loss=0.0868, p_class_loss=0.0903, re_loss=0.0222, conf_loss=0.0000, time=15sec
[i] iteration=7,194, learning_rate=0.0312, alpha=4.00, loss=0.2583, class_loss=0.0835, p_class_loss=0.0868, re_loss=0.0220, conf_loss=0.0000, time=15sec
[i] iteration=7,260, learning_rate=0.0306, alpha=4.00, loss=0.2688, class_loss=0.0869, p_class_loss=0.0906, re_loss=0.0228, conf_loss=0.0000, time=15sec
[i] iteration=7,271, threshold=0.10, train_mIoU=45.85%, best_train_mIoU=46.22%, time=18sec
[i] iteration=7,326, learning_rate=0.0299, alpha=4.00, loss=0.2663, class_loss=0.0865, p_class_loss=0.0900, re_loss=0.0225, conf_loss=0.0000, time=34sec
[i] iteration=7,392, learning_rate=0.0292, alpha=4.00, loss=0.2671, class_loss=0.0863, p_class_loss=0.0901, re_loss=0.0227, conf_loss=0.0000, time=15sec
[i] iteration=7,458, learning_rate=0.0285, alpha=4.00, loss=0.2605, class_loss=0.0839, p_class_loss=0.0877, re_loss=0.0222, conf_loss=0.0000, time=15sec
[i] iteration=7,524, learning_rate=0.0278, alpha=4.00, loss=0.2510, class_loss=0.0811, p_class_loss=0.0846, re_loss=0.0213, conf_loss=0.0000, time=15sec
[i] iteration=7,590, learning_rate=0.0271, alpha=4.00, loss=0.2630, class_loss=0.0848, p_class_loss=0.0884, re_loss=0.0224, conf_loss=0.0000, time=15sec
[i] iteration=7,656, learning_rate=0.0264, alpha=4.00, loss=0.2555, class_loss=0.0814, p_class_loss=0.0850, re_loss=0.0223, conf_loss=0.0000, time=15sec
[i] iteration=7,722, learning_rate=0.0257, alpha=4.00, loss=0.2520, class_loss=0.0803, p_class_loss=0.0842, re_loss=0.0219, conf_loss=0.0000, time=15sec
[i] iteration=7,788, learning_rate=0.0250, alpha=4.00, loss=0.2517, class_loss=0.0800, p_class_loss=0.0837, re_loss=0.0220, conf_loss=0.0000, time=15sec
[i] iteration=7,854, learning_rate=0.0243, alpha=4.00, loss=0.2637, class_loss=0.0844, p_class_loss=0.0883, re_loss=0.0227, conf_loss=0.0000, time=15sec
[i] iteration=7,920, learning_rate=0.0236, alpha=4.00, loss=0.2664, class_loss=0.0849, p_class_loss=0.0884, re_loss=0.0233, conf_loss=0.0000, time=15sec
[i] save model
[i] iteration=7,932, threshold=0.10, train_mIoU=46.27%, best_train_mIoU=46.27%, time=18sec
[i] iteration=7,986, learning_rate=0.0229, alpha=4.00, loss=0.2546, class_loss=0.0811, p_class_loss=0.0846, re_loss=0.0222, conf_loss=0.0000, time=34sec
[i] iteration=8,052, learning_rate=0.0222, alpha=4.00, loss=0.2614, class_loss=0.0840, p_class_loss=0.0878, re_loss=0.0224, conf_loss=0.0000, time=15sec
[i] iteration=8,118, learning_rate=0.0215, alpha=4.00, loss=0.2555, class_loss=0.0818, p_class_loss=0.0855, re_loss=0.0220, conf_loss=0.0000, time=15sec
[i] iteration=8,184, learning_rate=0.0208, alpha=4.00, loss=0.2465, class_loss=0.0785, p_class_loss=0.0823, re_loss=0.0214, conf_loss=0.0000, time=15sec
[i] iteration=8,250, learning_rate=0.0201, alpha=4.00, loss=0.2620, class_loss=0.0845, p_class_loss=0.0882, re_loss=0.0223, conf_loss=0.0000, time=15sec
[i] iteration=8,316, learning_rate=0.0194, alpha=4.00, loss=0.2562, class_loss=0.0832, p_class_loss=0.0868, re_loss=0.0216, conf_loss=0.0000, time=15sec
[i] iteration=8,382, learning_rate=0.0186, alpha=4.00, loss=0.2568, class_loss=0.0826, p_class_loss=0.0861, re_loss=0.0220, conf_loss=0.0000, time=15sec
[i] iteration=8,448, learning_rate=0.0179, alpha=4.00, loss=0.2609, class_loss=0.0831, p_class_loss=0.0871, re_loss=0.0227, conf_loss=0.0000, time=15sec
[i] iteration=8,514, learning_rate=0.0172, alpha=4.00, loss=0.2572, class_loss=0.0822, p_class_loss=0.0862, re_loss=0.0222, conf_loss=0.0000, time=15sec
[i] iteration=8,580, learning_rate=0.0165, alpha=4.00, loss=0.2527, class_loss=0.0802, p_class_loss=0.0843, re_loss=0.0220, conf_loss=0.0000, time=15sec
[i] save model
[i] iteration=8,593, threshold=0.10, train_mIoU=46.42%, best_train_mIoU=46.42%, time=18sec
[i] iteration=8,646, learning_rate=0.0157, alpha=4.00, loss=0.2529, class_loss=0.0817, p_class_loss=0.0854, re_loss=0.0215, conf_loss=0.0000, time=34sec
[i] iteration=8,712, learning_rate=0.0150, alpha=4.00, loss=0.2517, class_loss=0.0801, p_class_loss=0.0839, re_loss=0.0219, conf_loss=0.0000, time=15sec
[i] iteration=8,778, learning_rate=0.0143, alpha=4.00, loss=0.2487, class_loss=0.0797, p_class_loss=0.0836, re_loss=0.0214, conf_loss=0.0000, time=15sec
[i] iteration=8,844, learning_rate=0.0135, alpha=4.00, loss=0.2572, class_loss=0.0829, p_class_loss=0.0869, re_loss=0.0218, conf_loss=0.0000, time=15sec
[i] iteration=8,910, learning_rate=0.0128, alpha=4.00, loss=0.2498, class_loss=0.0794, p_class_loss=0.0829, re_loss=0.0219, conf_loss=0.0000, time=15sec
[i] iteration=8,976, learning_rate=0.0120, alpha=4.00, loss=0.2437, class_loss=0.0769, p_class_loss=0.0807, re_loss=0.0215, conf_loss=0.0000, time=15sec
[i] iteration=9,042, learning_rate=0.0112, alpha=4.00, loss=0.2490, class_loss=0.0789, p_class_loss=0.0825, re_loss=0.0219, conf_loss=0.0000, time=15sec
[i] iteration=9,108, learning_rate=0.0105, alpha=4.00, loss=0.2652, class_loss=0.0854, p_class_loss=0.0893, re_loss=0.0226, conf_loss=0.0000, time=15sec
[i] iteration=9,174, learning_rate=0.0097, alpha=4.00, loss=0.2582, class_loss=0.0836, p_class_loss=0.0872, re_loss=0.0219, conf_loss=0.0000, time=15sec
[i] iteration=9,240, learning_rate=0.0089, alpha=4.00, loss=0.2455, class_loss=0.0778, p_class_loss=0.0813, re_loss=0.0216, conf_loss=0.0000, time=15sec
[i] iteration=9,254, threshold=0.10, train_mIoU=45.94%, best_train_mIoU=46.42%, time=17sec
[i] iteration=9,306, learning_rate=0.0081, alpha=4.00, loss=0.2416, class_loss=0.0761, p_class_loss=0.0797, re_loss=0.0214, conf_loss=0.0000, time=34sec
[i] iteration=9,372, learning_rate=0.0073, alpha=4.00, loss=0.2461, class_loss=0.0781, p_class_loss=0.0819, re_loss=0.0215, conf_loss=0.0000, time=15sec
[i] iteration=9,438, learning_rate=0.0065, alpha=4.00, loss=0.2505, class_loss=0.0789, p_class_loss=0.0825, re_loss=0.0223, conf_loss=0.0000, time=15sec
[i] iteration=9,504, learning_rate=0.0057, alpha=4.00, loss=0.2517, class_loss=0.0796, p_class_loss=0.0837, re_loss=0.0221, conf_loss=0.0000, time=15sec
[i] iteration=9,570, learning_rate=0.0049, alpha=4.00, loss=0.2550, class_loss=0.0811, p_class_loss=0.0850, re_loss=0.0222, conf_loss=0.0000, time=15sec
[i] iteration=9,636, learning_rate=0.0040, alpha=4.00, loss=0.2542, class_loss=0.0802, p_class_loss=0.0843, re_loss=0.0224, conf_loss=0.0000, time=15sec
[i] iteration=9,702, learning_rate=0.0032, alpha=4.00, loss=0.2487, class_loss=0.0785, p_class_loss=0.0823, re_loss=0.0220, conf_loss=0.0000, time=15sec
[i] iteration=9,768, learning_rate=0.0023, alpha=4.00, loss=0.2435, class_loss=0.0775, p_class_loss=0.0807, re_loss=0.0213, conf_loss=0.0000, time=15sec
[i] iteration=9,834, learning_rate=0.0013, alpha=4.00, loss=0.2416, class_loss=0.0764, p_class_loss=0.0797, re_loss=0.0214, conf_loss=0.0000, time=15sec
[i] iteration=9,900, learning_rate=0.0003, alpha=4.00, loss=0.2595, class_loss=0.0842, p_class_loss=0.0882, re_loss=0.0218, conf_loss=0.0000, time=15sec
[i] iteration=9,915, threshold=0.10, train_mIoU=46.11%, best_train_mIoU=46.42%, time=18sec
|