Welcome to Autofeat’s documentation!
Autofeat
Autofeat is a Python-based, user-friendly library that incor- porates our human-in-the-loop methodology for feature discovery, designed for seamless integration in any notebook environment. The primary objective of this process is to enhance a base table by adding new features that significantly increase the predictive accuracy of a target ML model, specifically for classi- fication tasks. AutoFeat streamlines the process of selecting and integrating relevant tables from a dataset collection into the base table, thus creating an augmented table. Additionally, AutoFeat employs heuristic-based feature selection strategies to eliminate redundant or irrelevant features from this augmented table. By doing so, AutoFeat notably enhances the efficiency and accuracy of subsequent machine learning operations.
Workflow of AutoFeat
To install AutoFeat, check the Installation section.
For the entire API documentation, check the API section.
For use cases, check the Use Cases section.