FIX divide by zero in line search of GradientBoostingClassifier #28095
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
|
Note: I did not succeed to find a param setting that triggers this case for |
glemaitre
approved these changes
Jan 11, 2024
jeremiedbb
pushed a commit
to jeremiedbb/scikit-learn
that referenced
this pull request
Jan 17, 2024
jeremiedbb
pushed a commit
that referenced
this pull request
Jan 17, 2024
glemaitre
pushed a commit
to glemaitre/scikit-learn
that referenced
this pull request
Feb 10, 2024
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Outdated suggestions cannot be applied.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.
Reference Issues/PRs
This is a fix introduces in #26278 and #27312.
What does this implement/fix? Explain your changes.
This PR fixes a situation where the probability in a line search step of
GradientBoostingClassifierreaches exactly0or1leading to a division by zero.The first commit adds the test. CI will prove that it failed. After that, a fix will be added.
Any other comments?
@ogrisel #28048 (review)
Yes! I'm convinced now.
@glemaitre @lesteve This might be worth to be included in 1.4.0 or, later, in 1.4.1.
Some history
The check
denominator == 0was introduces in d63f39c.The check
abs(denominator) < 1e-150was introduces in #7970.#26278 added
_safe_dividewhich handleddenominator == 0correctly, but not on Pyodide.#27312 fixed the Pyodide thing, but introduced a
RuntimeWarnings.This PR puts back the
abs(denominator) < 1e-150without warnings.