TPOT
epistasislab.github.io
- Free plan
- macOS
- Linux
- Self-hosted
What TPOT is
TPOT is a free Python automated machine learning tool that uses genetic programming to optimize machine learning pipelines. Its current package includes genetic feature selection, flexible search space definitions, multi-objective optimization and a modular evolutionary algorithm framework. Documentation covers classifiers and regressors, as well as graph, sequential, tree and union pipeline types. When preprocessing is enabled, TPOT imputes missing values, one-hot encodes categorical features and standardizes data. It uses Dask for parallel processing and integrates with scikit-learn; an optional installation choice provides extra scikit-learn extensions. TPOT requires Python and can be installed manually or in a conda environment, with documented versions from 3.10 up to but not including 3.14. The project began in 2016 and says it was developed at Cedars-Sinai's Artificial Intelligence Innovation Lab with NIH funding. It is distributed under the GNU Lesser General Public License version 3 or later, without warranty. The project directs users to GitHub issues for bug reports, enhancement suggestions and discussion.
Who it suits
TPOT suits Python users who want to automate machine learning pipeline optimization and can work with its documented Python range. It may be useful to practitioners building classifier or regressor pipelines with scikit-learn.
Good to know, and what to watch
- Automates pipeline optimization using genetic programming.
- Supports classifiers, regressors and several pipeline types.
- Can impute, encode and standardize data.
- Uses Dask for parallel processing.
- Assumes input data suits the chosen operators.
- Preprocessing occurs before cross-validation splitting.
- Scikit-learn extensions may have issues on Arm CPUs.
Plans
Every plan and priceTPOT
Free plan
The verdict
TPOT provides a free, flexible approach to automated pipeline search for Python users. Review its data-format assumptions and preprocessing behavior before applying it to a validation workflow.
What the maker publishes
- Data handling limit
- TPOT does not check whether input data is correctly formatted and assumes the chosen operators can handle the supplied data. source
- Features
- The current package includes genetic feature selection, flexible search space definitions, multi-objective optimization, and a modular evolutionary algorithm framework. source
- Founded
- 2016 source
- Installation
- TPOT requires Python and can be installed in a conda environment or manually; the documented Python range is >=3.10 and <3.14. source
- Integrations
- The installation documentation lists scikit-learn and offers extra scikit-learn extensions through the tpot[sklearnex] installation option. source
- License
- TPOT is free software distributed under the GNU Lesser General Public License version 3 or later, without warranty. source
- Parallel processing
- TPOT uses Dask for parallel processing and recommends guarding script code with an if __name__ == "__main__" block. source
- Pipeline types
- The documentation includes classifiers and regressors, as well as graph, sequential, tree, and union pipeline types. source
- Platform caveat
- The documentation warns that scikit-learn extensions may have compatibility or performance issues on Arm-based CPUs such as M1 Macs. source
- Preprocessing
- With preprocessing enabled, TPOT imputes missing values, one-hot encodes categorical features, and standardizes data. source
- Preprocessing limit
- The documentation says preprocessing is currently fitted and transformed on the entire training set before cross-validation splitting. source
- Purpose
- TPOT is a Python automated machine learning tool that optimizes machine learning pipelines using genetic programming. source
- Research focus
- The project says TPOT was developed in the Artificial Intelligence Innovation Lab at Cedars-Sinai with NIH funding. source
- Support
- The project directs users to its GitHub issues to report bugs or suggest enhancements and to discuss extensions. source
Checked against the maker's pages on 9 Oct 2026.