scikit-learn-contrib/boruta_py

Health v2 · maintenance only

Adoption not part of the score

1627 stars · 264 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

A Python implementation of the Boruta all-relevant feature selection method, built on scikit-learn. It identifies all features carrying predictive information rather than just a minimal optimal subset, with a scikit-learn-compatible fit/transform interface.

Use cases

• find all features relevant for prediction in a dataset
• feature selection before training a machine learning model
• understand which variables explain a phenomenon in my data
• rank features by importance using random forests
• select features for biological or genomic data analysis

When to choose

• you want all relevant features, not just a minimal subset
• you work within the scikit-learn ecosystem
• you need feature importance ranking with statistical rigor

When to avoid

• you need a minimal optimal feature subset for a specific classifier
• your data is too large for iterative random-forest fitting
• you need non-tree-based feature selection methods

Facets

library · maturity active

machine-learning data-science machine-learning data-science python feature-selection scikit-learn boruta random-forest all-relevant-feature-selection

Member repositories

For agents

markdown · JSON · MCP: product_card(name="scikit-learn-contrib/boruta_py")