Thanks to visit codestin.com
Credit goes to github.com

Skip to content

FIX Improves nan support in LabelEncoder #22629

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
Merged
jjerphan merged 6 commits into from
Oct 28, 2022
Merged

FIX Improves nan support in LabelEncoder #22629

jjerphan merged 6 commits into from
Oct 28, 2022

Conversation

@thomasjpfan
Copy link
Member

@thomasjpfan thomasjpfan commented Feb 28, 2022 •
edited
Loading

Reference Issues/PRs

Fixes #22628

What does this implement/fix? Explain your changes.

This makes LabelEncoder.transform convert y to the same dtype as classes_ before encoding it.

Any other comments?

On main, the behavior of nan in LabelEncoder encodes the nan in fit_transform:

from sklearn.preprocessing import LabelEncoder

print(LabelEncoder().fit_transform([1, np.nan, 2]))
# [0 2 1]

This PR makes the behavior consistent with object dtypes.

@thomasjpfan thomasjpfan mentioned this pull request Feb 28, 2022
Closed
@kabartay
Copy link

kabartay commented Mar 11, 2022

Thanks. This seems really helpful.

betatim
betatim approved these changes Oct 17, 2022
View reviewed changes
Copy link
Member

@betatim betatim left a comment

Choose a reason for hiding this comment

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

LGTM.

@thomasjpfan thomasjpfan added the Quick Review For PRs that are quick to review label Oct 25, 2022
jjerphan
jjerphan approved these changes Oct 26, 2022
View reviewed changes
Copy link
Member

@jjerphan jjerphan left a comment •
edited
Loading

Choose a reason for hiding this comment

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

LGTM. Thank you, @thomasjpfan.

Just a comment.

sklearn/utils/validation.py
Comment on lines +1151 to +1152
dtype : data-type, default=None
Data type for `y`.
Copy link
Member

@jjerphan jjerphan Oct 26, 2022

Choose a reason for hiding this comment

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

What is the behavior of column_or_1d when dtype=None? Is it this one?

Suggested change
dtype : data-type, default=None
Data type for `y`.
dtype : data-type, default=None
Data type for `y`.
When dtype is None, dtype is inferred from the elements of `y`.

Copy link
Member Author

@thomasjpfan thomasjpfan Oct 26, 2022

Choose a reason for hiding this comment

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

Yea, it is inferred from the elements of y:

XREF: np.asarray docs

@jjerphan jjerphan merged commit 9c4f023 into scikit-learn:main Oct 28, 2022
jjerphan pushed a commit to samronsin/scikit-learn that referenced this pull request Oct 28, 2022
@thomasjpfan @jjerphan
FIX Improves nan support in LabelEncoder (scikit-learn#22629)
7084b20
glemaitre pushed a commit to glemaitre/scikit-learn that referenced this pull request Oct 31, 2022
@thomasjpfan @glemaitre
FIX Improves nan support in LabelEncoder (scikit-learn#22629)
a31665a
andportnoy pushed a commit to andportnoy/scikit-learn that referenced this pull request Nov 5, 2022
@thomasjpfan @andportnoy
FIX Improves nan support in LabelEncoder (scikit-learn#22629)
3ae9fd1
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@amueller amueller amueller approved these changes

@betatim betatim betatim approved these changes

@jjerphan jjerphan jjerphan approved these changes

Assignees

No one assigned

Labels

module:preprocessing module:utils Quick Review For PRs that are quick to review

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

LabelEncoder not transforming nans as expected.

5 participants

@thomasjpfan @kabartay @amueller @betatim @jjerphan