Implementing GPU backends #32885
BishopWolf
started this conversation in
Ideas
Implementing GPU backends
#32885
Replies: 0 comments
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
Currently there are two main GPU backends that are shared by most cards and do not require too much specificities:
Less performance is better than no performance at all. GPU will always be faster than CPU!
I wouldn't suggest the CUDA nor the ROCm pats since they are very demanding with specific card logics which is against the sklearn phylosophy. But agnostic standards like Vulkan or OpenCL should be supported for sure.
Beta Was this translation helpful? Give feedback.
All reactions