Pointcept / PointTransformerV3

[CVPR'24 Oral] Official repository of Point Transformer V3 (PTv3)
MIT License
583 stars 30 forks source link

A question about "nan" when testing #36

Closed QWTforGithub closed 2 months ago

QWTforGithub commented 2 months ago

Thank you for sharing this very good work. I tested on ScanNet using PTv3 (Pointcept 1.5.1). However, during testing something like "[2024-04-22 23:48:31,627 INFO test.py line 230 23615] Test: scene0791_00 [65/100]-248589 Batch 16.487 (14.210) Accuracy 0.0000 (0.0000) mIoU nan (0.0000)" results like this. Is there something wrong with my settings? I am using "model_best.pth" downloaded at "https://huggingface.co/Pointcept/PointTransformerV3/tree/main/scannet-semseg-pt-v3m1-0-base". My running command is as follows: python tools/test.py --config-file configs/scannet/semseg-pt-v3m1-0-base.py --options save_path=exp/scannet/semseg-pt-v3m1-0-base weight=exp/scannet/semseg-pt-v3m1-0-base/model/model_best.pth. More detailed running information:

[2024-04-22 23:55:41,376 INFO test.py line 41 131988] => Loading config ... [2024-04-22 23:55:41,376 INFO test.py line 48 131988] => Building model ... [2024-04-22 23:55:41,836 INFO test.py line 61 131988] Num params: 46167572 [2024-04-22 23:55:42,556 INFO test.py line 68 131988] Loading weight at: ../model_best.pth [2024-04-22 23:55:42,858 INFO test.py line 80 131988] => Loaded weight '../model_best.pth' (epoch 94) [2024-04-22 23:55:42,860 INFO test.py line 53 131988] => Building test dataset & dataloader ... [2024-04-22 23:55:42,861 INFO scannet.py line 72 131988] Totally 100 x 1 samples in test set. [2024-04-22 23:55:42,862 INFO test.py line 119 131988] >>>>>>>>>>>>>>>> Start Evaluation >>>>>>>>>>>>>>>> [2024-04-22 23:55:50,217 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 0/108 [2024-04-22 23:55:50,330 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 1/108 [2024-04-22 23:55:50,443 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 2/108 [2024-04-22 23:55:50,555 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 3/108 [2024-04-22 23:55:50,668 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 4/108 [2024-04-22 23:55:50,781 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 5/108 [2024-04-22 23:55:50,894 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 6/108 [2024-04-22 23:55:51,008 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 7/108 [2024-04-22 23:55:51,121 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 8/108 [2024-04-22 23:55:51,234 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 9/108 [2024-04-22 23:55:52,020 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 10/108 [2024-04-22 23:55:52,319 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 11/108 [2024-04-22 23:55:52,663 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 12/108 [2024-04-22 23:55:52,960 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 13/108 [2024-04-22 23:55:53,076 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 14/108 [2024-04-22 23:55:53,200 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 15/108 [2024-04-22 23:55:53,321 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 16/108 [2024-04-22 23:55:53,442 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 17/108 [2024-04-22 23:55:53,560 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 18/108 [2024-04-22 23:55:53,671 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 19/108 [2024-04-22 23:55:53,787 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 20/108 [2024-04-22 23:55:53,902 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 21/108 [2024-04-22 23:55:54,022 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 22/108 [2024-04-22 23:55:54,144 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 23/108 [2024-04-22 23:55:54,260 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 24/108 [2024-04-22 23:55:54,376 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 25/108 [2024-04-22 23:55:54,486 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 26/108 [2024-04-22 23:55:54,597 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 27/108 [2024-04-22 23:55:54,899 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 28/108 [2024-04-22 23:55:56,021 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 29/108 [2024-04-22 23:55:56,292 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 30/108 [2024-04-22 23:55:56,406 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 31/108 [2024-04-22 23:55:56,514 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 32/108 [2024-04-22 23:55:56,623 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 33/108 [2024-04-22 23:55:56,730 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 34/108 [2024-04-22 23:55:56,837 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 35/108 [2024-04-22 23:55:56,946 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 36/108 [2024-04-22 23:55:57,055 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 37/108 [2024-04-22 23:55:57,163 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 38/108 [2024-04-22 23:55:57,269 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 39/108 [2024-04-22 23:55:57,375 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 40/108 [2024-04-22 23:55:57,483 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 41/108 [2024-04-22 23:55:57,590 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 42/108 [2024-04-22 23:55:57,695 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 43/108 [2024-04-22 23:55:57,801 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 44/108 [2024-04-22 23:55:57,908 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 45/108 [2024-04-22 23:55:58,015 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 46/108 [2024-04-22 23:55:58,122 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 47/108 [2024-04-22 23:55:58,229 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 48/108 [2024-04-22 23:55:58,336 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 49/108 [2024-04-22 23:55:58,442 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 50/108 [2024-04-22 23:55:58,549 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 51/108 [2024-04-22 23:55:58,656 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 52/108 [2024-04-22 23:55:58,762 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 53/108 [2024-04-22 23:55:58,869 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 54/108 [2024-04-22 23:55:58,977 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 55/108 [2024-04-22 23:55:59,085 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 56/108 [2024-04-22 23:55:59,192 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 57/108 [2024-04-22 23:55:59,299 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 58/108 [2024-04-22 23:55:59,408 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 59/108 [2024-04-22 23:55:59,517 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 60/108 [2024-04-22 23:55:59,625 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 61/108 [2024-04-22 23:55:59,732 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 62/108 [2024-04-22 23:55:59,839 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 63/108 [2024-04-22 23:55:59,945 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 64/108 [2024-04-22 23:56:00,051 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 65/108 [2024-04-22 23:56:00,158 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 66/108 [2024-04-22 23:56:00,263 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 67/108 [2024-04-22 23:56:00,380 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 68/108 [2024-04-22 23:56:00,496 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 69/108 [2024-04-22 23:56:00,612 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 70/108 [2024-04-22 23:56:00,729 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 71/108 [2024-04-22 23:56:00,845 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 72/108 [2024-04-22 23:56:00,961 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 73/108 [2024-04-22 23:56:01,078 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 74/108 [2024-04-22 23:56:01,195 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 75/108 [2024-04-22 23:56:01,311 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 76/108 [2024-04-22 23:56:01,427 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 77/108 [2024-04-22 23:56:01,544 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 78/108 [2024-04-22 23:56:01,662 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 79/108 [2024-04-22 23:56:01,778 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 80/108 [2024-04-22 23:56:01,895 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 81/108 [2024-04-22 23:56:02,012 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 82/108 [2024-04-22 23:56:02,128 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 83/108 [2024-04-22 23:56:02,244 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 84/108 [2024-04-22 23:56:02,359 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 85/108 [2024-04-22 23:56:02,475 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 86/108 [2024-04-22 23:56:02,590 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 87/108 [2024-04-22 23:56:02,707 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 88/108 [2024-04-22 23:56:02,824 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 89/108 [2024-04-22 23:56:02,940 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 90/108 [2024-04-22 23:56:03,060 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 91/108 [2024-04-22 23:56:03,179 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 92/108 [2024-04-22 23:56:03,298 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 93/108 [2024-04-22 23:56:03,416 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 94/108 [2024-04-22 23:56:03,535 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 95/108 [2024-04-22 23:56:03,654 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 96/108 [2024-04-22 23:56:03,770 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 97/108 [2024-04-22 23:56:03,887 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 98/108 [2024-04-22 23:56:04,004 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 99/108 [2024-04-22 23:56:04,117 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 100/108 [2024-04-22 23:56:04,229 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 101/108 [2024-04-22 23:56:04,343 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 102/108 [2024-04-22 23:56:04,459 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 103/108 [2024-04-22 23:56:04,569 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 104/108 [2024-04-22 23:56:04,689 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 105/108 [2024-04-22 23:56:04,801 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 106/108 [2024-04-22 23:56:04,913 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 107/108 /home/liabiao/anaconda3/envs/pt3/lib/python3.8/site-packages/numpy/core/fromnumeric.py:3464: RuntimeWarning: Mean of empty slice. return _methods._mean(a, axis=axis, dtype=dtype, /home/liabiao/anaconda3/envs/pt3/lib/python3.8/site-packages/numpy/core/_methods.py:192: RuntimeWarning: invalid value encountered in scalar divide ret = ret.dtype.type(ret / rcount) [2024-04-22 23:56:04,933 INFO test.py line 230 131988] Test: scene0799_00 [1/100]-204572 Batch 16.083 (16.083) Accuracy 0.0000 (0.0000) mIoU nan (0.0000) [2024-04-22 23:56:05,225 INFO test.py line 196 131988] Test: 2/100-scene0773_00, Batch: 0/153 [2024-04-22 23:56:05,309 INFO test.py line 196 131988] Test: 2/100-scene0773_00, Batch: 1/153 [2024-04-22 23:56:05,394 INFO test.py line 196 131988] Test: 2/100-scene0773_00, Batch: 2/153 [2024-04-22 23:56:05,479 INFO test.py line 196 131988] Test: 2/100-scene0773_00, Batch: 3/153 [2024-04-22 23:56:05,564 INFO test.py line 196 131988] Test: 2/100-scene0773_00, Batch: 4/153 [2024-04-22 23:56:05,649 INFO test.py line 196 131988] Test: 2/100-scene0773_00, Batch: 5/153 [2024-04-22 23:56:05,741 INFO test.py line 196 131988] Test: 2/100-scene0773_00, Batch: 6/153 [2024-04-22 23:56:05,828 INFO test.py line 196 131988] Test: 2/100-scene0773_00, Batch: 7/153 ...

QWTforGithub commented 2 months ago

Thank you for sharing this very good work. I tested on ScanNet using PTv3 (Pointcept 1.5.1). However, during testing something like "[2024-04-22 23:48:31,627 INFO test.py line 230 23615] Test: scene0791_00 [65/100]-248589 Batch 16.487 (14.210) Accuracy 0.0000 (0.0000) mIoU nan (0.0000)" results like this. Is there something wrong with my settings? I am using "model_best.pth" downloaded at "https://huggingface.co/Pointcept/PointTransformerV3/tree/main/scannet-semseg-pt-v3m1-0-base". My running command is as follows: python tools/test.py --config-file configs/scannet/semseg-pt-v3m1-0-base.py --options save_path=exp/scannet/semseg-pt-v3m1-0-base weight=exp/scannet/semseg-pt-v3m1-0-base/model/model_best.pth. More detailed running information:

[2024-04-22 23:55:41,376 INFO test.py line 41 131988] => Loading config ... [2024-04-22 23:55:41,376 INFO test.py line 48 131988] => Building model ... [2024-04-22 23:55:41,836 INFO test.py line 61 131988] Num params: 46167572 [2024-04-22 23:55:42,556 INFO test.py line 68 131988] Loading weight at: ../model_best.pth [2024-04-22 23:55:42,858 INFO test.py line 80 131988] => Loaded weight '../model_best.pth' (epoch 94) [2024-04-22 23:55:42,860 INFO test.py line 53 131988] => Building test dataset & dataloader ... [2024-04-22 23:55:42,861 INFO scannet.py line 72 131988] Totally 100 x 1 samples in test set. [2024-04-22 23:55:42,862 INFO test.py line 119 131988] >>>>>>>>>>>>>>>> Start Evaluation >>>>>>>>>>>>>>>> [2024-04-22 23:55:50,217 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 0/108 [2024-04-22 23:55:50,330 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 1/108 [2024-04-22 23:55:50,443 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 2/108 [2024-04-22 23:55:50,555 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 3/108 [2024-04-22 23:55:50,668 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 4/108 [2024-04-22 23:55:50,781 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 5/108 [2024-04-22 23:55:50,894 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 6/108 [2024-04-22 23:55:51,008 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 7/108 [2024-04-22 23:55:51,121 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 8/108 [2024-04-22 23:55:51,234 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 9/108 [2024-04-22 23:55:52,020 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 10/108 [2024-04-22 23:55:52,319 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 11/108 [2024-04-22 23:55:52,663 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 12/108 [2024-04-22 23:55:52,960 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 13/108 [2024-04-22 23:55:53,076 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 14/108 [2024-04-22 23:55:53,200 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 15/108 [2024-04-22 23:55:53,321 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 16/108 [2024-04-22 23:55:53,442 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 17/108 [2024-04-22 23:55:53,560 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 18/108 [2024-04-22 23:55:53,671 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 19/108 [2024-04-22 23:55:53,787 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 20/108 [2024-04-22 23:55:53,902 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 21/108 [2024-04-22 23:55:54,022 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 22/108 [2024-04-22 23:55:54,144 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 23/108 [2024-04-22 23:55:54,260 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 24/108 [2024-04-22 23:55:54,376 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 25/108 [2024-04-22 23:55:54,486 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 26/108 [2024-04-22 23:55:54,597 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 27/108 [2024-04-22 23:55:54,899 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 28/108 [2024-04-22 23:55:56,021 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 29/108 [2024-04-22 23:55:56,292 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 30/108 [2024-04-22 23:55:56,406 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 31/108 [2024-04-22 23:55:56,514 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 32/108 [2024-04-22 23:55:56,623 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 33/108 [2024-04-22 23:55:56,730 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 34/108 [2024-04-22 23:55:56,837 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 35/108 [2024-04-22 23:55:56,946 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 36/108 [2024-04-22 23:55:57,055 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 37/108 [2024-04-22 23:55:57,163 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 38/108 [2024-04-22 23:55:57,269 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 39/108 [2024-04-22 23:55:57,375 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 40/108 [2024-04-22 23:55:57,483 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 41/108 [2024-04-22 23:55:57,590 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 42/108 [2024-04-22 23:55:57,695 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 43/108 [2024-04-22 23:55:57,801 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 44/108 [2024-04-22 23:55:57,908 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 45/108 [2024-04-22 23:55:58,015 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 46/108 [2024-04-22 23:55:58,122 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 47/108 [2024-04-22 23:55:58,229 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 48/108 [2024-04-22 23:55:58,336 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 49/108 [2024-04-22 23:55:58,442 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 50/108 [2024-04-22 23:55:58,549 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 51/108 [2024-04-22 23:55:58,656 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 52/108 [2024-04-22 23:55:58,762 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 53/108 [2024-04-22 23:55:58,869 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 54/108 [2024-04-22 23:55:58,977 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 55/108 [2024-04-22 23:55:59,085 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 56/108 [2024-04-22 23:55:59,192 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 57/108 [2024-04-22 23:55:59,299 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 58/108 [2024-04-22 23:55:59,408 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 59/108 [2024-04-22 23:55:59,517 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 60/108 [2024-04-22 23:55:59,625 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 61/108 [2024-04-22 23:55:59,732 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 62/108 [2024-04-22 23:55:59,839 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 63/108 [2024-04-22 23:55:59,945 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 64/108 [2024-04-22 23:56:00,051 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 65/108 [2024-04-22 23:56:00,158 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 66/108 [2024-04-22 23:56:00,263 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 67/108 [2024-04-22 23:56:00,380 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 68/108 [2024-04-22 23:56:00,496 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 69/108 [2024-04-22 23:56:00,612 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 70/108 [2024-04-22 23:56:00,729 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 71/108 [2024-04-22 23:56:00,845 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 72/108 [2024-04-22 23:56:00,961 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 73/108 [2024-04-22 23:56:01,078 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 74/108 [2024-04-22 23:56:01,195 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 75/108 [2024-04-22 23:56:01,311 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 76/108 [2024-04-22 23:56:01,427 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 77/108 [2024-04-22 23:56:01,544 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 78/108 [2024-04-22 23:56:01,662 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 79/108 [2024-04-22 23:56:01,778 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 80/108 [2024-04-22 23:56:01,895 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 81/108 [2024-04-22 23:56:02,012 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 82/108 [2024-04-22 23:56:02,128 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 83/108 [2024-04-22 23:56:02,244 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 84/108 [2024-04-22 23:56:02,359 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 85/108 [2024-04-22 23:56:02,475 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 86/108 [2024-04-22 23:56:02,590 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 87/108 [2024-04-22 23:56:02,707 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 88/108 [2024-04-22 23:56:02,824 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 89/108 [2024-04-22 23:56:02,940 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 90/108 [2024-04-22 23:56:03,060 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 91/108 [2024-04-22 23:56:03,179 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 92/108 [2024-04-22 23:56:03,298 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 93/108 [2024-04-22 23:56:03,416 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 94/108 [2024-04-22 23:56:03,535 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 95/108 [2024-04-22 23:56:03,654 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 96/108 [2024-04-22 23:56:03,770 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 97/108 [2024-04-22 23:56:03,887 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 98/108 [2024-04-22 23:56:04,004 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 99/108 [2024-04-22 23:56:04,117 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 100/108 [2024-04-22 23:56:04,229 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 101/108 [2024-04-22 23:56:04,343 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 102/108 [2024-04-22 23:56:04,459 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 103/108 [2024-04-22 23:56:04,569 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 104/108 [2024-04-22 23:56:04,689 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 105/108 [2024-04-22 23:56:04,801 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 106/108 [2024-04-22 23:56:04,913 INFO test.py line 196 131988] Test: 1/100-scene0799_00, Batch: 107/108 /home/liabiao/anaconda3/envs/pt3/lib/python3.8/site-packages/numpy/core/fromnumeric.py:3464: RuntimeWarning: Mean of empty slice. return _methods._mean(a, axis=axis, dtype=dtype, /home/liabiao/anaconda3/envs/pt3/lib/python3.8/site-packages/numpy/core/_methods.py:192: RuntimeWarning: invalid value encountered in scalar divide ret = ret.dtype.type(ret / rcount) [2024-04-22 23:56:04,933 INFO test.py line 230 131988] Test: scene0799_00 [1/100]-204572 Batch 16.083 (16.083) Accuracy 0.0000 (0.0000) mIoU nan (0.0000) [2024-04-22 23:56:05,225 INFO test.py line 196 131988] Test: 2/100-scene0773_00, Batch: 0/153 [2024-04-22 23:56:05,309 INFO test.py line 196 131988] Test: 2/100-scene0773_00, Batch: 1/153 [2024-04-22 23:56:05,394 INFO test.py line 196 131988] Test: 2/100-scene0773_00, Batch: 2/153 [2024-04-22 23:56:05,479 INFO test.py line 196 131988] Test: 2/100-scene0773_00, Batch: 3/153 [2024-04-22 23:56:05,564 INFO test.py line 196 131988] Test: 2/100-scene0773_00, Batch: 4/153 [2024-04-22 23:56:05,649 INFO test.py line 196 131988] Test: 2/100-scene0773_00, Batch: 5/153 [2024-04-22 23:56:05,741 INFO test.py line 196 131988] Test: 2/100-scene0773_00, Batch: 6/153 [2024-04-22 23:56:05,828 INFO test.py line 196 131988] Test: 2/100-scene0773_00, Batch: 7/153 ...

I've found the reason. I'm very sorry, I just came into contact with the point cloud segmentation task, so I don't know much about ScanNet. ScanNet officials did not directly give ScanNet test set (100 scenes) labels. I tested it on the test set (the code assigned the label value to -1 by default), which resulted in "nan" appearing.

QWTforGithub commented 2 months ago

I think this issue can be closed.