WillKoehrsen/feature-selector
A Python library for automated feature selection and dimensionality reduction in machine learning datasets using importance-based and statistical methods.
Not currently ranked — collecting fresh signals.
star history
Feature Selector provides five methods to identify and remove low-value features: missing value thresholding, single unique value detection, collinearity-based removal, and importance-based filtering using LightGBM feature importances. It integrates with scikit-learn workflows and includes visualization utilities for inspecting dataset characteristics and feature importance distributions.
Frequently asked
- What is WillKoehrsen/feature-selector?
- A Python library for automated feature selection and dimensionality reduction in machine learning datasets using importance-based and statistical methods.
- Is feature-selector open source?
- Yes — WillKoehrsen/feature-selector is open source, released under the GPL-3.0 license.
- What language is feature-selector written in?
- WillKoehrsen/feature-selector is primarily written in Jupyter Notebook.
- How popular is feature-selector?
- WillKoehrsen/feature-selector has 2.2k stars on GitHub.
- Where can I find feature-selector?
- WillKoehrsen/feature-selector is on GitHub at https://github.com/WillKoehrsen/feature-selector.