Skip to content

DOC Add link to glossary "label indicator matrix" for classification metric docstrings #32893

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
Open
lucyleeow wants to merge 1 commit into
base: main
Choose a base branch
from
Open

DOC Add link to glossary "label indicator matrix" for classification metric docstrings #32893

lucyleeow wants to merge 1 commit into from

Conversation

@lucyleeow
Copy link
Member

@lucyleeow lucyleeow commented Dec 12, 2025

Reference Issues/PRs

Noticed while working on #32755

What does this implement/fix? Explain your changes.

Adds reference to the glossary term https://scikit-learn.org/dev/glossary.html#term-label-indicator-matrix for classification metrics

Not sure about the best term to use.

  • Avoided use of 'binary' i.e., binary indicator - binary here is talking about the indicator being 0 or 1, not about the data being binary. This may be confusing and this term is not included as an alias in the glossary so I have avoided it
  • when talking about binary data, I have used the term "label indicator format“ / ”label indicator matrix“ (depending on how best the wording fits into the sentence), to avoid use of 'multilabel' (as the data is not multilabel).
  • used 'multilabel indicator matrix' when talking about multilabel data

AI usage disclosure

I used AI assistance for:

  • Code generation (e.g., when writing an implementation or fixing a bug)
  • Test/benchmark generation
  • Documentation (including examples)
  • Research and understanding

Any other comments?

lucyleeow
lucyleeow commented Dec 12, 2025
View reviewed changes
sklearn/metrics/_ranking.py
Comment on lines -505 to +507
True labels or binary label indicators. The binary and multiclass cases
True labels or :term:`label indicator matrix`. The binary and multiclass cases
expect labels with shape (n_samples,) while the multilabel case expects
binary label indicators with shape (n_samples, n_classes).
:term:`multilabel indicator matrix` with shape (n_samples, n_classes).
Copy link
Member Author

@lucyleeow lucyleeow Dec 12, 2025

Choose a reason for hiding this comment

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

I've used both glossary aliases 'label indicator matrix' (when talking about binary data) and 'multilabel indicator matrix' (when talking about multilabel data). I think this is okay. I amended the glossary entry to clarify that it could be used for both binary and multilabel data...

lucyleeow
lucyleeow commented Dec 12, 2025
View reviewed changes
doc/glossary.rst
This format can be used to represent binary or multilabel data. Each row of
a 2d array or sparse matrix corresponds to a sample, each column
corresponds to a class, and each element is 1 if the sample is labeled
with the class and 0 if not.
Copy link
Member Author

@lucyleeow lucyleeow Dec 12, 2025 •
edited
Loading

Choose a reason for hiding this comment

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

Also wonder if we should add a short example, like we do here: https://scikit-learn.org/dev/modules/preprocessing_targets.html#multilabelbinarizer

@lucyleeow lucyleeow mentioned this pull request Dec 12, 2025
Open
4 tasks
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

No reviews

Assignees

No one assigned

Labels

Documentation module:metrics No Changelog Needed

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

1 participant

@lucyleeow