scikit-learn-contrib/category_encoders

Health v2 · maintenance only

Adoption not part of the score

2499 stars · 411 forks observed · 2026-08-28

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

A scikit-learn-contrib library providing a collection of sklearn-compatible transformers for encoding categorical variables into numeric form using techniques like one-hot, binary, hashing, target, and CatBoost encoding. It supports pandas DataFrames, works in sklearn pipelines, and includes both supervised and unsupervised encoders.

Use cases

• encode categorical columns for machine learning models
• target encode high-cardinality categorical features
• one-hot encode pandas dataframes in sklearn pipelines
• convert string categories to numbers for sklearn
• avoid target leakage when encoding categories
• pickle and reuse a fitted categorical encoder

When to choose

• you need sklearn-pipeline-compatible categorical transformers
• you work with pandas DataFrames and want column-level control
• you need supervised encoders like target, CatBoost, or James-Stein encoding

When to avoid

• your data has no categorical features
• you need deep-learning-native embedding layers instead of tabular encoding
• you are outside the Python/sklearn ecosystem

Facets

library · maturity active

machine-learning data-science etl machine-learning data-science python categorical-encoding sklearn-transformers feature-engineering target-encoding pandas

Member repositories

For agents

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