sherpa supports both CPU and CUDA. To install a CUDA version of sherpa, all you need to do is to install PyTorch with CUDA support.
Supported operating systems, Python versions, PyTorch versions, and CUDA versions are listed below:
We provide the following methods to install sherpa. You can choose the one that works best for you.
from-source