Jump to content

Wrote a guide to get Tensorflow accelerated PI tools accelerated on AMD GPUs


Recommended Posts

Right now, it works for Ubuntu, I am building a Fedora 40 workstation to build on as Fedora "Vendors" ROCm now, so that one shouldn't be too hard to target.  AMD has been promising some PyTorch updates on WIndows with MIOPEN finally getting ported to Windows, so over the next few months there may be a chance to see Windows. If Windows support for ROCm doesn't mature, I may try and revive a newer version of Tensorflow DirectML .

https://rockchucksummit.com/pixinsight-gpu-acceleration-for-amd/

This works for the excellent RC-Astro tools and StarNet++ - on my 7900xtx it makes working in 61MP a breeze.

  • Like 1
Link to comment
Share on other sites

  • 2 weeks later...

Please sign in to comment

You will be able to leave a comment after signing in



Sign In Now
×
×
  • Create New...

Important Information

By using this site you agree to our Guidelines