FIX Improve best run detection in kmeans when n_init > 1 #21195
Merged
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
|
It seems to fix it indeed on my machine:
|
|
I guess it would be great to have a few tests for |
I also observe this on my machine. |
jjerphan
approved these changes
Sep 30, 2021
|
LGTM, one thing I am wondering is whether the changelog should go in v1.0.rst or v1.1.rst ... |
v1.0 because the next bug fix release will be 1.0.1 |
|
Merging, thanks! |
glemaitre
pushed a commit
to glemaitre/scikit-learn
that referenced
this pull request
Oct 23, 2021
glemaitre
pushed a commit
that referenced
this pull request
Oct 25, 2021
samronsin
pushed a commit
to samronsin/scikit-learn
that referenced
this pull request
Nov 30, 2021
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Outdated suggestions cannot be applied.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.
Fixes #21160
Instead of relying of some tolerance to take rouding errors into account, we can simply check that the clustering of the new run is the same as the clustering of the best run so far (up to a permutation of the labels).
The check is really quick: ~200ms for 100_000_000 samples, which is negligible compared to a complete run of kmeans.
The non regression test is already there: test_k_means_fit_predict (see the linked issue), even if it does not fail in the CI because it's an effect quite hard to trigger. I think checking that it now passes is enough. @lesteve could you confirm that ?