Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Using ndarray as init for KMeans raises a ValueError #26657

Merged
merged 18 commits into from
Jun 26, 2023
Merged

Using ndarray as init for KMeans raises a ValueError #26657

Changes from 2 commits
Commits
Show all changes
18 commits
Select commit Hold shift + click to select a range
2b0a9bd
When init is specified as a numpy array, it errors with "ValueError: …
bnsh Jun 21, 2023
dd34560
Merge branch 'main' into main
bnsh Jun 22, 2023
7be0252
Handle callable inits as well as array like inits. Added test case.
bnsh Jun 22, 2023
6fcc539
Formatting for Black
bnsh Jun 22, 2023
79b3b05
Added n_init changes to KMeans and MiniBatchKMeans doc strings.
bnsh Jun 22, 2023
992ade2
Typo. parametrize not parameterize.
bnsh Jun 22, 2023
717cace
Black: Line too long.
bnsh Jun 22, 2023
1103a5d
More black formatting woes.
bnsh Jun 22, 2023
a6d2581
Added n_init docs for kmeans function, added a . for more formatting.
bnsh Jun 22, 2023
00b8abc
Added entry in Changelog
bnsh Jun 22, 2023
6bbd3ad
Pull Request needs to be in backquotes not double quotes.
bnsh Jun 22, 2023
1f42d0c
Comment change: unbalanced backquotes and single quotes.
bnsh Jun 23, 2023
0895f9d
"The check that init has a valid type and value is already taken care…
bnsh Jun 23, 2023
30bc7ea
Excessive detail for "What's New"
bnsh Jun 23, 2023
e4088eb
Change to docstring: Simplified initialization cases.
bnsh Jun 23, 2023
aa37c7c
Formatting. Whitespace.
bnsh Jun 23, 2023
e62bd30
Needs a period.
bnsh Jun 23, 2023
04d751d
Update v1.3.rst
jeremiedbb Jun 26, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Jump to
Jump to file
Failed to load files.
Diff view
Diff view
5 changes: 2 additions & 3 deletions sklearn/cluster/_kmeans.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -897,10 +897,9 @@ def _check_params_vs_input(self, X, default_n_init=None):
)
self._n_init = default_n_init
if self._n_init == "auto":
if self.init == "k-means++":
self._n_init = default_n_init
if (not _is_arraylike_not_scalar(self.init)) and self.init == "k-means++":
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

We should as well deal with the callable. So let's do the following:

        if self._n_init == "auto":
            if isinstance(self.init, str):
                self._n_init = 1 if self.init == "k-means++" else default_n_init
            elif callable(self.init):
                self._n_init = default_n_init
            else:  # array-like
                self._n_init = 1

Copy link
Contributor Author

@bnsh bnsh Jun 22, 2023 •
edited

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Hi @glemaitre ... OK, so I added the changes you mention.. (Tho I changed it a little bit from what you wrote.. I more explicitly demand that self.init be either "k-means++" or "random" if it's a string, and either a callable or array like otherwise, and it raises an Exception otherwise.) I also added the docstring changes and the test case, but I don't know what I'm expected to do with the doc/whats_new/v1.3.rst... Can you please tell me what exactly I should be doing there? (That seems to be the only thing that's not passing CI/CD checks...)

Thanks!

Copy link
Contributor Author

@bnsh bnsh Jun 22, 2023 •
edited

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Actually, I think I kind of figured out the Changelog thing. Or at least it's passing the CI/CD check now. I still don't know tho, if the text in there is any good..

self._n_init = 1
else:
self._n_init = default_n_init

if _is_arraylike_not_scalar(self.init) and self._n_init != 1:
warnings.warn(
Expand Down