Make AUCPR work with multiple query groups #4436
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This was referenced May 3, 2019
Closed
|
I'm merging this, since it involves the change of a single assertion and doesn't change the actual implementation of AUCPR. |
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Fixes #4431.
I read the paper referenced by the AUCPR implementation, and as far as I can tell, the
EvalAucPRseems to be consistent with the paper.The issue was that
EvalAucPRwas assuming the existence of only a single query group. More precisely, it throws an exception whensum_auc >= 1.0, where it should have checkedsum_auc >= ngroupinstead. Since all tests for AUCPR was performing binary classification (where there is only one group), the bug went undetected.Fix. Have a proper check on the AUC value, in presence of multiple query groups. Also add a test with learning-to-rank example.