-
Notifications
You must be signed in to change notification settings - Fork 377
pointnet2 installation failed #64
Comments
Hi! I am faced with exactly the same issue as yours. Installing PyTorch and Tensorflow, then running |
Hello, i have solved out this problem. I used another platform with GPU 1660ti and CUDA10.0, then the problem disappear. So the bug may be related to the CUDA version i guess. |
@StevenZzz07 Yes, I solved this issue, I have tensorflow-gpu==1.14 and pytorch==1.1.0 and CUDA 10.0 installed, I fixed it by install original Pointnet_Pytorch code the repo borrowed from and set the PYTHONPATH= |
For me, the problem was solved by installing the correct versions of tensorflow(1.14) and pytorch (1.1). However, another problem popped up afterwards. |
For me the problem vanished after upgrading from |
I have a similar problem. Solved by install CUDA 10.0. tensorflow1.14, pytorch 1.2 |
After I succeed running
python setup.py install
, I failed to runpython models/votenet.py
as follows:I can
import torch; import pointnet2
without errors, but not whenimport torch; import pointnet2._ext as _ext
The text was updated successfully, but these errors were encountered: