You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
First of all, thank you for your open source
Due to my limited video memory (4090 24GB), I sometimes encounter OOM issues, but I have two graphics cards, so I want to try multi card training.
Instan NSR itself supports multiple cards, but when I simply set the GPU in the parameters to 0, 1, an error will occur in the GS branch
File "/home/ime-lab/zyb/gs/GSDF/gaussian_splatting/gaussian_renderer/__init__.py", line 266, in pr
efilter_voxel
return radii_pure > 0
RuntimeError: CUDA error: an illegal memory access was encountered
Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions.
Do you have any suggestions?
The text was updated successfully, but these errors were encountered:
First of all, thank you for your open source
Due to my limited video memory (4090 24GB), I sometimes encounter OOM issues, but I have two graphics cards, so I want to try multi card training.
Instan NSR itself supports multiple cards, but when I simply set the GPU in the parameters to 0, 1, an error will occur in the GS branch
Do you have any suggestions?
The text was updated successfully, but these errors were encountered: