[MRG] Fix top_k_accuracy_score ignoring labels for "multiclass" case #19300
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
thomasjpfan
reviewed
Jan 29, 2021
There was a problem hiding this comment.
Thank you for the PR @flyingdutchman23 !
Please add an entry to the change log at doc/whats_new/v1.0.rst with tag |Fix|. Like the other entries there, please reference this pull request with :pr: and credit yourself (and other contributors if applicable) with :user:.
|
Thank you @thomasjpfan for reviewing this PR. |
joclement
changed the title
joclement
changed the title
joclement
changed the title
|
I accidently closed this PR, because it was on my main branch. I will open a new one with the same changes. |
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Outdated suggestions cannot be applied.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
Fixes small bug in #16625
What does this implement/fix? Explain your changes.
Details on the bug:
I stumbled upon this problem when using the LOGO cross validator. When not all the classes of a multiclass problem are contained in the
y_trueparameter, thetop_k_accuracy_scorefunction raises the following error:In order to replicate this bug I added a test, which fails, if my fix is not added. The error above is caused by that test.
The reason is that the function call
type_of_target(y_true)determines the type "binary", if the number of unique values iny_trueis<=2.I believe that this error doesn't need to happen, if the
labelsparameter is passed. Then the function can determine the classes and the type of the problem by inspecting that parameter as well.