When I add this command to Dockerfile, the compilation breaks. This is due to in setup.py, there are multiple places, it calls torch.cuda, leaving the error message: "Unknown compute capability. Specify the target compute capabilities in the TCNN_CUDA_ARCHITECTURES environment variable or install PyTorch with the CUDA backend to detect it automatically."
How can I install the bindings through Dockerfile?
wzds2015
commented
5 months ago
I am able to install the python bindings on the server. pip install git+https://github.com/NVlabs/tiny-cuda-nn/#subdirectory=bindings/torch
When I add this command to Dockerfile, the compilation breaks. This is due to in setup.py, there are multiple places, it calls torch.cuda, leaving the error message: "Unknown compute capability. Specify the target compute capabilities in the TCNN_CUDA_ARCHITECTURES environment variable or install PyTorch with the CUDA backend to detect it automatically."
How can I install the bindings through Dockerfile?