scikit-learn-stubs

0.0.3 verified Fri May 01 auth: no python

Type stubs for scikit-learn, providing static type information for type checkers like mypy and pyright. Current version 0.0.3, updated 2025; low release cadence. Maintained by Microsoft via the python-type-stubs repository, forked by hoel-bagard.

pip install scikit-learn-stubs

Common errors

error Cannot find implementation or library stub for module 'sklearn' ↓

cause The stubs package is not installed or not compatible with the installed scikit-learn version.

fix

Run: pip install scikit-learn-stubs

error Module 'sklearn.linear_model' has no attribute 'LogisticRegression' ↓

cause Outdated stub files missing some estimators; or incorrect scikit-learn version.

fix

Upgrade scikit-learn and stubs: pip install --upgrade scikit-learn scikit-learn-stubs

Warnings

gotcha Stubs may not cover all scikit-learn modules or parameters, especially experimental ones. Use strict type checking cautiously. ↓

fix Check the stub coverage in the repository before relying on types for rarely used estimators.

deprecated The original Microsoft python-type-stubs repository is archived. This fork may become outdated relative to new scikit-learn releases. ↓

fix Consider using the official scikit-learn type stubs from scikit-learn itself (if available in future) or contribute to this fork.

breaking Version 0.0.x is pre-1.0. Breaking changes in stub types may occur between minor versions. ↓

fix Pin exact version in requirements: scikit-learn-stubs==0.0.3

Imports

sklearn.linear_model.LogisticRegression
```
from sklearn.linear_model import LogisticRegression
```
Correct import; stubs do not change import paths.

Quickstart

Basic scikit-learn usage with type stubs enabled. The stubs are automatically discovered by mypy/pyright when installed.

from sklearn.linear_model import LogisticRegression
import numpy as np

# Example usage with type hints (stubs enable type checking)
model: LogisticRegression = LogisticRegression()
X = np.array([[0, 0], [1, 1]])
y = np.array([0, 1])
model.fit(X, y)
print(model.predict([[0.5, 0.5]]))