featuretools and featuristic
These are competitors—both automate feature engineering for machine learning, but Featuretools offers a mature, production-ready framework with deep integrations for time-series and relational data, while Featuristic provides an alternative approach using symbolic regression and genetic programming for interpretability-focused feature discovery.
About featuretools
alteryx/featuretools
An open source python library for automated feature engineering
This tool helps data professionals prepare their datasets for machine learning by automatically creating meaningful features from raw, structured data. You feed it multiple related tables, like customer details, transactions, and sessions, and it outputs a single, wide table of new features that can directly be used to train predictive models. It's designed for data scientists, analysts, and machine learning engineers who need to quickly generate rich features without extensive manual coding.
About featuristic
martineastwood/featuristic
Automated, interpretable feature engineering using symbolic regression and genetic programming.
This project helps data scientists and machine learning practitioners automatically create new, insightful features from their existing datasets. You provide your raw data, and it intelligently generates new, interpretable mathematical features that improve the accuracy of your predictive models. It's designed for anyone building machine learning models who wants to enhance model performance without manually crafting complex data transformations.
Related comparisons
Scores updated daily from GitHub, PyPI, and npm data. How scores work