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
[INTEL MKL] Add MKL-DNN quantized Matmul op with some fusions - Part2. #26910
[INTEL MKL] Add MKL-DNN quantized Matmul op with some fusions - Part2. #26910
Conversation
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.
Does this PR depend on #26909 (Part 1)?
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.
I think the PR is good to go. Once #26909 is merged, please rebase the PR and I'll approve it. Thank you again!
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.
Part 1 (#26909) has been merged. (Hasn't shown on Github yet, but it will soon.) And I realized that all three PRs are guarded with INTEL_MKL
anyway, so we probably won't have problems with the tests on Github. Will approve part 3 too.
There are four failed tests. @rthadur Could you please help pull this PR in? Thank you very much! :) |
PiperOrigin-RevId: 250721888
This PR is to replace an older PR #26271 by splitting into three. Current PR is part2. It enables MKL-DNN quantized Matmul ops through graph optimization.