rasbt/mlxtend

Health v2 · maintenance only

Flags: no_license

Adoption not part of the score

5166 stars · 911 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

mlxtend (machine learning extensions) is a Python library of helper tools for day-to-day data science and machine learning tasks. It provides ensemble methods like stacking and voting classifiers, feature selection/extraction, model evaluation utilities, plotting helpers, and frequent pattern mining such as the Apriori algorithm.

Use cases

• stack multiple scikit-learn classifiers into an ensemble
• mine frequent itemsets and association rules with apriori
• plot decision regions of a classifier
• run statistical tests to compare classifiers like mcnemar or 5x2cv F-test
• perform feature selection with sequential feature selection
• decompose bias and variance of a model
• bootstrap model evaluation scores
• load MNIST or iris datasets for experiments

When to choose

• you use scikit-learn and need stacking/voting ensembles or evaluation utilities it lacks
• you need association rule mining (Apriori, FP-Growth) in Python
• you want plotting helpers for decision boundaries and confusion matrices
• you need statistical model comparison tests like McNemar's test or Cochran's Q

When to avoid

• you need deep learning or GPU-accelerated training - use PyTorch or TensorFlow instead
• you need large-scale distributed frequent pattern mining - use Spark MLlib
• you want a full AutoML or end-to-end ML pipeline framework
• your project depends only on scikit-learn features already covering your needs

Facets

library · maturity active

machine-learning data-science data-visualization parser machine-learning data-science python cross-platform association-rules apriori frequent-pattern-mining ensemble-learning stacking feature-selection scikit-learn model-evaluation algorithms

Member repositories

For agents

markdown · JSON · MCP: product_card(name="rasbt/mlxtend")