Back to Gap Map Infrastructure / Classic ML & computer vision

TPOT

Epistasis Lab
open source / Overall score: 3.2

TPOT, the Tree-based Pipeline Optimization Tool, is an AutoML library that uses genetic programming to search over scikit-learn pipelines of preprocessing steps, feature selectors and models, and returns the best pipeline it finds for a classification or regression task. Version 1.0 folded in the rewritten TPOT2, adding multi-objective search and configurable search spaces. It is developed at Cedars-Sinai; the original version came from the University of Pennsylvania.

Openness

5 high confidence
5.0
license
LGPL-3.0 (OSI; version 3 or any later version)
source
public (github.com/EpistasisLab/tpot)
core features withheld
no — academic project from Cedars-Sinai; nothing sold

TPOT is under the GNU Lesser General Public License and builds from the public repository. It is an academic project from Cedars-Sinai and the University of Pennsylvania, and nothing is sold beside it.

Adoption

2 high confidence
2.0

Measured on monthly PyPI downloads of the tpot package.

Capability

5 medium confidence
5.0

Given training data and a target, TPOT searches for a whole pipeline and fits the best one itself, which is the automation AutoGluon provides. Unlike AutoGluon it bundles no pretrained model.

  • https://raw.githubusercontent.com/EpistasisLab/tpot/HEAD/README.md recorded 2026-11-12

    README: "TPOT is a Python Automated Machine Learning tool that optimizes machine learning pipelines using genetic programming"; the rewrite adds "genetic feature selection, a significantly expanded and more flexible method of defining search spaces, multi-objective optimization"; usage "est = tpot.TPOTClassifier() est.fit(X,y)".

Verified 2026-11-12