github.com web signal

TPOT AutoML tool rewritten, adds multi-objective search

By Published Updated

TL;DR

  • TPOT has been rewritten from scratch, adding genetic feature selection, multi-objective optimization, and a more flexible search-space definition.
  • The project originated at the University of Pennsylvania and the modern version is now maintained at Cedars-Sinai, with a 2025 paper in Patterns.
  • The rewritten tool supports Python 3.10 through 3.13, ships under LGPL-3.0, and is backed by NIH grants U01 AG066833 and R01 LM010098.

TPOT stands for Tree-based Pipeline Optimization Tool. The GitHub project, originally developed at the University of Pennsylvania under Randal S. Olson and Jason Moore and now maintained at Cedars-Sinai, is "a Python Automated Machine Learning tool that optimizes machine learning pipelines using genetic programming."

The maintainers have rebuilt it. "TPOT recently went through a major refactoring. The package was rewritten from scratch to improve efficiency and performance, support new features, and fix numerous bugs," the README says. New capabilities, per the project page, include "genetic feature selection, a significantly expanded and more flexible method of defining search spaces, multi-objective optimization," alongside a more modular evolutionary-algorithm framework.

The 2025 citation lists Hernandez, Saini, Ghosh and Moore in Patterns, 6(7). TPOT requires Python ">=3.10, <3.14" and ships under LGPL-3.0. Development is credited to NIH grants U01 AG066833 and R01 LM010098. Two of the researchers we track have shared the repository link into our feed.

Shared on Bluesky by 2 AI experts

Originally reported by github.com

Read the original article →

Original headline: GitHub - EpistasisLab/tpot: A Python Automated Machine Learning tool that optimizes machine learning pipelines using genetic programming.