evo-ens - Evolutionary Ensemble Builder

A scikit-learn-compatible ensemble constructor for tabular data, powered by DEAP. evo-ens uses a Genetic Algorithm to decide which candidate models to combine and how to weight them, trading off predictive score against prediction diversity, complementing evo-gafs (feature selection) and evo-gpfe (feature engineering) further upstream in the pipeline.

Part of the evo-suite family (import name: evo_ens). Documentation: https://evo-suite.readthedocs.io/

Why evo-ens?

Installation

pip install evo-ens            # core
pip install "evo-ens[viz]"     # + matplotlib for the plotting helpers

Quickstart

from sklearn.datasets import load_breast_cancer
from evo_ens import EvoEnsConfig, EvoEnsembleClassifier

X, y = load_breast_cancer(return_X_y=True, as_frame=True)

clf = EvoEnsembleClassifier(
    config=EvoEnsConfig(population_size=60, n_generations=40, diversity_beta=0.10, verbose=False)
)
clf.fit(X, y)

print(clf.result_.summary())
print(clf.get_ensemble_info())

EvoEnsembleRegressor follows the same API for regression targets, scoring with R² by default.

Documentation & examples

• Full documentation (user guide + API reference): https://evo-suite.readthedocs.io/
• Runnable examples: the repository's examples/evo-ens/ directory.

Citation

@software{evo_ens,
  author    = {Skrauba, Axel},
  title     = {evo-ens: Evolutionary Ensemble Builder for tabular data},
  year      = {2026},
  version   = {0.1.0},
  url       = {https://github.com/AxelSkrauba/evo-suite}
}

License

MIT