rasbt/mlxtend
Mlxtend
mlxtend is a pure Python machine learning extension library that provides additional tools for association rule mining, ensemble learning, and feature selection. It is built on numpy and pandas, with all data operations accepting and returning pandas DataFrames, and custom estimators inherit from scikit-learn’s base classes to offer a uniform fit-predict interface compatible with grid search.
The library implements the Apriori algorithm for mining frequent itemsets from transaction data and generating association rules with confidence and lift metrics. For classification, it combines multiple base models using majority voting, stacking, or bagging to produce a single ensemble prediction. Feature selection is handled through methods such as greedy sequential selection, which iteratively adds or removes features guided by cross-validated performance scores from a wrapped estimator.
All algorithms are implemented without compiled extensions, relying solely on standard Python libraries for numerical and tabular computation. The package is published on the Python Package Index and can be installed via pip.
Features
- Python Machine Learning Libraries - Pure Python library extending scikit-learn with tools for association rules, feature selection, and ensemble learning.
- Association Rule Learning - Discovers association rules from transaction data, revealing hidden co-occurrence patterns.
- Ensemble Learning Libraries - Ships an ensemble learning library for combining models via voting, stacking, and bagging.
- Frequent Itemset Mining - Implements the Apriori algorithm to mine frequent itemsets and generate association rules from transaction data.
- Majority-Vote Ensembles - Ships a majority-vote classifier that combines predictions from multiple base models.
- Model Ensembling - Combines models using voting, stacking, and bagging to improve predictive performance and robustness.
- Automated Feature Selection Tools - Selects relevant features using sequential selection algorithms guided by cross-validated performance scores.
- Minimal Feature Subset Selectors - Provides greedy sequential forward/backward feature selection with cross-validated scoring.
- Estimator Interfaces - Custom estimators inherit from scikit-learn base classes for a uniform fit-predict interface and grid-search compatibility.
- General Machine Learning - Collection of useful tools for data science tasks.
- Machine Learning - Collection of helper modules for data analysis and machine learning.
- Machine Learning Algorithms - Collection of helper modules for data analysis and machine learning workflows.
- Framework-uri de Machine Learning - Collection of useful machine learning extensions.
- Machine Learning Packages - Collection of useful tools for daily data science tasks.
- Model Ensembles - Ensemble voting and stacking regressors.
