[Issue]: Radeon PRO W6800 Pro Graphics native Windows Pytorch acceleration support via ROCm / HIP or some other way?
AshleyT3 opened this issue · comments
Problem Description
Is there a plan to eventually support Radeon PRO W6800 Pro Graphics native Windows Pytorch acceleration support via ROCm / HIP or some other way? Today, it seems Pytorch is not supported natively on Windows aside from using DirectML. DirectML is usable but in at least one case it requires customizations to pre-existing Pytorch code so does not insert GPU acceleration support seamlessly in such a case.
I've seen some online commentary about ROCm 6.1 possibly supporting native Windows ROCm / Pytorch support but it's not official AMD commentary. If there is no official public word on this, it would be great to get some FAQ or comment on this.
Operating System
Windows 11
CPU
5800X
GPU
AMD Radeon Pro W6800
ROCm Version
ROCm 6.0.0, ROCm 5.7.1
ROCm Component
No response
Steps to Reproduce
No response
(Optional for Linux users) Output of /opt/rocm/bin/rocminfo --support
No response
Additional Information
No response