Haibo Qiu

icon indicating a website link haibo-qiu.github.io icon indicating an email [email protected]

icon company University of Sydney icon location Beijing icon location The scariest moment is always just before you start.

Results 4 comments of Haibo Qiu

RuntimeError: The size of tensor a (719348) must match the size of tensor b (1438695) at non-singleton dimension 0

As @xindeng98 mentioned in https://github.com/guochengqian/PointNeXt/issues/18#issuecomment-1182670679, this error was caused by the fact that [test_entire_room](https://github.com/guochengqian/PointNeXt/blob/f49f3e6d9b3d9e8cfb353f4ea61502796eba4693/examples/segmentation/main.py#L358) does not support multi-gpus testing.

cannot open help pages/wiki

Hi Joel The same problem happened to me. When I clicked the wiki page, it displayed that "You do not have permission to update this wiki".

Dataset download server down

Hi there, [ScanNet benchmark](https://kaldir.vc.in.tum.de/scannet_benchmark/) was also unavailable. ![image](https://user-images.githubusercontent.com/31210561/226214878-dabf1692-e13b-42ad-997b-ff0b22b647b2.png) ![image](https://user-images.githubusercontent.com/31210561/226215027-081fe4ad-af1c-4351-ba61-b6d88611f43b.png)

How to improve the test performance on ScanNet?

Hi, I share the same concern as I achieved 74.3% on the ScanNet validation set but only managed to obtain 70.6% on the ScanNet test set.