rocm-pytorch-gfx803

• Copy of rocm/pytorch with gfx803 cards compiled in (see https://github.com/xuhuisheng/rocm-build/blob/develop/docs/gfx803.md)
• I made a guide for AMD GPUs in KVM using pytorch: https://github.com/jrcichra/pytorch-amd-kvm-guide
• See the original docker hub page for usage: https://hub.docker.com/r/rocm/pytorch
• And see @xuhuisheng's report: https://github.com/xuhuisheng/rocm-build/blob/master/docs/gfx803.md.

All credit for this goes to him and other's who discovered the bug!

Usage

• cd to your machine learning project directory
• sudo docker run -it -v $PWD:/projects --privileged --name pytorch --device=/dev/kfd --device=/dev/dri --group-add video jrcichra/rocm-pytorch-gfx803
• This will mount the current directory into /projects. You can navigate there and try your pytorch project. It 'should' have GPU acceleration. I checked this with:

import torch
if torch.cuda.is_available():
    device = torch.device("cuda:0")
    print("Running on the GPU")
else:
    device = torch.device("cpu")
    print("Running on the CPU")

• I get Running on the GPU
• Feel free to ctrl+d inside this container. It will stick around. If you want to go back in at a later time, just run:
• docker start pytorch ; docker exec -it pytorch bash and keep crunching models
• Pulled from my KVM + RX 580 ML guide: https://github.com/jrcichra/pytorch-amd-kvm-guide#rocmpytorch-container

gfx803 support

• @BishopWolf brought to my attention AMD is dropping support for gfx803 on rocm. See ROCm/ROCm#1353
• I've disabled Github Actions on this repo. I'm not going to update this container for now