Repository navigation
Respect positive_label in binary metric counts - #3111
Open
akshitkhokhani wants to merge 1 commit into
Open
akshitkhokhani wants to merge 1 commit into
akshitkhokhani wants to merge 1 commit into
Conversation
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
What does this PR do?
Make
BinaryMetricStatsuse its configuredpositive_labelwhen calculating confusion counts and derived metrics. Threshold estimation already uses that label, but the confusion calculation treats the original labels as a 0/1 mask. A perfectly separated classifier with positive class encoded as0consequently reports F1=0; other encodings can produce negative false-positive counts.The change creates a local binary truth mask without changing stored labels, threshold selection, or the DER formula. Regression coverage checks equivalent encodings with fixed and automatic thresholds, integer and floating labels, explicit confusion counts, repeated summaries, and unchanged labels.
Validation: the new regression fails in eight custom-label cases on the original source while four default-label controls pass. With the fix, CPU metric unit tests and module doctests pass (29 tests), and all repository pre-commit checks pass. No new dependency.