Skip to content

[python-package] Replace bare Any with specific types in sklearn helpers #12111

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

Open
PavelGuzenfeld wants to merge 2 commits into
base: master
Choose a base branch
from
Open

[python-package] Replace bare Any with specific types in sklearn helpers #12111

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
eeb0f68
[python-package] Replace bare Any with specific types in sklearn.py
PavelGuzenfeld Mar 19, 2026
1b1a4ee
Keep return type as Any since DaskDMatrix does not extend DMatrix
PavelGuzenfeld Mar 19, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
26 changes: 13 additions & 13 deletions python-package/xgboost/sklearn.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -649,7 +649,7 @@ def get_model_categories(


def pick_ref_categories(
X: Any,
X: ArrayLike,
model_cats: Optional[Union[FeatureTypes, Categories]],
Xy_cats: Optional[Categories],
) -> Optional[Union[FeatureTypes, Categories]]:
Expand Down Expand Up @@ -682,18 +682,18 @@ def pick_ref_categories(
def _wrap_evaluation_matrices(
*,
missing: float,
X: Any,
y: Any,
group: Optional[Any],
qid: Optional[Any],
sample_weight: Optional[Any],
base_margin: Optional[Any],
X: ArrayLike,
y: ArrayLike,
group: Optional[ArrayLike],
qid: Optional[ArrayLike],
sample_weight: Optional[ArrayLike],
base_margin: Optional[ArrayLike],
feature_weights: Optional[ArrayLike],
eval_set: Optional[Sequence[Tuple[Any, Any]]],
sample_weight_eval_set: Optional[Sequence[Any]],
base_margin_eval_set: Optional[Sequence[Any]],
eval_group: Optional[Sequence[Any]],
eval_qid: Optional[Sequence[Any]],
eval_set: Optional[Sequence[Tuple[ArrayLike, ArrayLike]]],
sample_weight_eval_set: Optional[Sequence[ArrayLike]],
base_margin_eval_set: Optional[Sequence[ArrayLike]],
eval_group: Optional[Sequence[ArrayLike]],
eval_qid: Optional[Sequence[ArrayLike]],
create_dmatrix: Callable,
enable_categorical: bool,
feature_types: Optional[Union[FeatureTypes, Categories]],
Expand Down Expand Up @@ -1209,7 +1209,7 @@ def custom_metric(m: Callable) -> Metric:
wrapped = _metric_decorator(m)
return wrapped

def invalid_type(m: Any) -> None:
def invalid_type(m: object) -> None:
Copy link
Copy Markdown
Member

@trivialfis trivialfis Mar 20, 2026

Choose a reason for hiding this comment

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

why?

Copy link
Copy Markdown
Contributor Author

@PavelGuzenfeld PavelGuzenfeld Mar 20, 2026

Choose a reason for hiding this comment

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

object is the proper top type in Python — it means "accepts anything" while keeping type safety. Any disables type checking entirely (both co- and contra-variant), so mypy silently skips any misuse. Since invalid_type only calls type(m), which works on every object, object is the accurate annotation.

That said, this is a minor stylistic point — happy to revert if you prefer keeping Any here.

msg = f"Invalid type for the `eval_metric`: {type(m)}"
raise TypeError(msg)

Expand Down
Loading