-
Notifications
You must be signed in to change notification settings - Fork 5.7k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Experimental: allow fp16 in mps
#961
Conversation
It appears to not longer be necessary.
The documentation is not available anymore as the PR was closed or merged. |
The RC of PyTorch 1.13 is ready: pytorch/pytorch#86312 (comment) This is now ready for review. I removed workarounds for fixed issues, applied performance optimizations and updated the documentation. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Feel free to merge whenever :-)
* Docs: refer to pre-RC version of PyTorch 1.13.0. * Remove temporary workaround for unavailable op. * Update comment to make it less ambiguous. * Remove use of contiguous in mps. It appears to not longer be necessary. * Special case: use einsum for much better performance in mps * Update mps docs. * MPS: make pipeline work in half precision.
The pipeline works if this change is applied, but results differ from those obtained when using full precision.
This is intended to be applied after #926.
Fixes #660.