The PyPI package does not install <a href="https://pytorch.org/get-started/locally/" target="_blank" rel="noopener noreferrer">Pytorch</a> because there is no option to build that package specifying exactly the CUDA version you want to use. There are a few solutions to set up ``pyproject.toml`` with poetry and specify the CUDA version, as discussed <a href="https://github.com/python-poetry/poetry/issues/6409" target="_blank" rel="noopener noreferrer">here</a>, but then PyPI package can not be built (as stated <a href="https://peps.python.org/pep-0440/#direct-references" target="_blank" rel="noopener noreferrer">here</a>).
0 commit comments