-
Notifications
You must be signed in to change notification settings - Fork 3.8k
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
Implemented the Learnable Communitive Monoid Aggregation #7976
Conversation
Codecov Report
@@ Coverage Diff @@
## master #7976 +/- ##
=======================================
Coverage 89.58% 89.58%
=======================================
Files 461 462 +1
Lines 26980 27018 +38
=======================================
+ Hits 24169 24205 +36
- Misses 2811 2813 +2
... and 1 file with indirect coverage changes 📣 We’re building smart automated test selection to slash your CI/CD build times. Learn more |
overall LGTM, but lets wait on @wsad1 and/or @EdisonLeeeee reviews |
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.
Left some minor comments. Thank you :)
Co-authored-by: Jintang Li <cnljt@outlook.com>
Co-authored-by: Jintang Li <cnljt@outlook.com>
Co-authored-by: Jintang Li <cnljt@outlook.com>
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.
Wow, this is pretty cool :) Thank you so much!
Initial version of the LCM aggregation, requested in issue #7574. A possible feature to incorporate is for `forward` to additionally compute and return an associativity loss, as described in the paper. --------- Co-authored-by: Rishi Puri <puririshi98@berkeley.edu> Co-authored-by: Jintang Li <cnljt@outlook.com> Co-authored-by: rusty1s <matthias.fey@tu-dortmund.de>
Why was the associativity loss not included? |
Hi @Kroppeb! I briefly attempted to implement associativity loss, but it wasn't as straightforward as commutativity loss and I never got around to finishing it. But good point - it should be available here. |
Initial version of the LCM aggregation, requested in issue #7574. A possible feature to incorporate is for
forward
to additionally compute and return an associativity loss, as described in the paper.