I forked Pytorch from GitHub and I followed the following steps
- conda install numpy ninja pyyaml mkl mkl-include setuptools cmake cffi
- git clone --recursive GitHub - pytorch/pytorch: Tensors and Dynamic neural networks in Python with strong GPU acceleration
- cd pytorch
- python tools/amd_build/build_amd.py
- python setup.py install
I successfully completed the PyTorch build process. However, when I try to run PyTorch, it is unable to detect my AMD GPU. Additionally, PyTorch is unable to assign streams or execute any functions that involve CUDA. This issue is quite perplexing as the GPU-related functionality seems to be completely non-functional.