search

ldkong1205 / LaserMix

[CVPR 2023 Highlight] LaserMix for Semi-Supervised LiDAR Semantic Segmentation
https://ldkong.com/LaserMix
Apache License 2.0
272 stars 17 forks source link

AttributeError: 'PointData' object has no attribute 'voxel_semantic_mask' #17

Closed molarsu closed 8 months ago

molarsu commented 8 months ago

Thanks for the great work and release the codes. When I trying to train with a single GPU and run the command “python tools/train.py configs/lasermix/lasermix_minku18_semi_semantickitti_10.py”, got a AttributeError: 'PointData' object has no attribute 'voxel_semantic_mask'. Did you mean: 'pts_semantic_mask'? What should I do to deal this problem? Thank you very much.

Xiangxu-0103 commented 8 months ago

Hi, thanks for your attention to our work. It's indeed a bug as we have refactored the structure of MinkUNet. We have updated the code. Please let us know if you face new problems.

molarsu commented 8 months ago

Thanks for your reply! But I got a TypeError: super() takes no keyword arguments from File LaserMix/mmdet3d/models/segmentors/minkunet.py in line 24 , def init(self, kwargs) -> None: super(MinkUNet).init(kwargs) And when I fix it to super().init(**kwargs), it could work.

Xiangxu-0103 commented 8 months ago

Thanks for your feedback again. It should be super(MinkUNet, self).__init__(**kwargs) and it will be fixed soon.

molarsu commented 8 months ago

Thanks for your reply and my problem has been solved!