Anttwo / SuGaR

[CVPR 2024] Official PyTorch implementation of SuGaR: Surface-Aligned Gaussian Splatting for Efficient 3D Mesh Reconstruction and High-Quality Mesh Rendering

Home Page:https://anttwo.github.io/sugar/

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

CUDA mismatch error while installing in docker.

jaswin90 opened this issue · comments

After doing in the .yml file installations, once we try to install the setup.py for simple-knn and diff-gaussian-rasterization, it is throwing a cuda mismatch error as nvcc is pointing to the base docker and pytorch is compiled with a different cuda.

figured out if we install cudatoolkit-11.8, a new nvcc gets installed in the conda environment and nvcc --version points to the correct one inside the conda environment.

please reopen if it doesn't work.