Can anyone here run PyTorch on ROCm with an AMD APU on a laptop? If so, what is the brand/model of your laptop? How is the performance?

From a quick research online, it looks like this should be possible, but there are a lot of pitfalls.

See here: https://community.amd.com/t5/knowledge-base/amd-rocm-hardware-and-software-support-document/ta-p/489937

> The integrated GPUs in AMD APUs are not officially supported targets for ROCm. As described [below](#limited-support), "Carrizo", "Bristol Ridge", and "Raven Ridge" APUs are enabled in our upstream drivers and the ROCm OpenCL runtime.

> However, they are not enabled in the HIP runtime, and may not work due to motherboard or OEM hardware limitations.

> As such, they are not yet officially supported targets for ROCm.

Also this discussion: https://github.com/RadeonOpenCompute/ROCm/issues/608

In summary, it is possible to use AMD APUs (at least the desktop versions) for machine learning with PyTorch and ROCm, but it needs a bios which is properly built, which apparently is hard to find…

I need a very portable solution to do ML with PyTorch, and small laptops with Ryzen APUs look like a good solution, but I’d like to make sure before I invest my money in that.