Skip to main content

A genetic AutoML system for ensemble methods

Project description


genens is an AutoML system for pipeline optimization based on developmental genetic programming.


Clone the repository.

git clone
pip install genens

Using genens

As for now, the GenensClassifier is ready to be used. It has an interface similar to other scikit-learn estimators. When fit() is called, the evolutionary optimization is run. After it finishes, predict() produces a prediction with the best of optimized pipelines. Alternatively, you can call get_best_pipelines() to get pipelines from the pareto front.

from genens import GenensClassifier
from sklearn.datasets import load_iris()

iris = load_iris()
train_X, test_X, train_y, test_y = train_test_split(,, test_size=0.25)

clf = GenensClassifier(), train_y)
... # process of evolution

pred = clf.predict(test_X)


Directory ./genens/tests contains scripts for running dataset tests and produce data about evolution process along with pickle files of best optimized pipelines. Sample config files are included in ./genens/tests/config.

  • Run genEns on a dataset specified in the config file.

python ./genens/tests/ --out OUT_DIR config CONFIG

python ./genens/tests/ --out OUT_DIR --config CONFIG

More tests are to be included in later releases.

Project details

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

genens-0.1.19.tar.gz (35.1 kB view hashes)

Uploaded Source

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page