humanpose1 / MS-SVConv

Compute descriptors for 3D point cloud registration using a multi scale sparse voxel architecture

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Not able to run sample demo colab notebook

pranavgundewar opened this issue · comments

After successful installation of dependencies in the notebook, I have encountered the following error-

image


OSError Traceback (most recent call last)
in ()
6 import requests
7 import torch
----> 8 from torch_points3d.applications.pretrained_api import PretainedRegistry
9 from torch_points3d.core.data_transform import GridSampling3D, AddFeatByKey, AddOnes, Random3AxisRotation
10 from torch_points3d.datasets.registration.pair import Pair

8 frames
/usr/lib/python3.7/ctypes/init.py in init(self, name, mode, handle, use_errno, use_last_error)
362
363 if handle is None:
--> 364 self._handle = _dlopen(self._name, mode)
365 else:
366 self._handle = handle

OSError: /usr/local/lib/python3.7/dist-packages/torch_sparse/_version_cuda.so: undefined symbol: ZN3c106detail23torchInternalAssertFailEPKcS2_jS2_S2

Hi @pranavgundewar ,
Thanks for reporting the problem.
It is solved now. It was a problem of version(torch 1.9 is not compatible with the current version of torch points 3d)
I'll close the issue. If you find something else, you can reopen it.
Best,