Python re-implementation of the Diverse Niching Genetic Programming (DivNichGP) algorithm, adapted for regression and binary classification.
DivNichGP has been originally proposed by:
Shaolin Wang, Yi Mei, Mengjie Zhang. "Novel ensemble genetic programming hyper-heuristics for uncertain capacitated arc routing problem".
Proceedings of the Genetic and Evolutionary Computation Conference, July 2019, pp. 1093–1101.
This implementation was made for:
@inproceedings{virgolin2021genetic,
author = {Virgolin, Marco},
title = {Genetic Programming is Naturally Suited to Evolve Bagging Ensembles},
year = {2021},
isbn = {9781450383509},
publisher = {Association for Computing Machinery},
address = {New York, NY, USA},
doi = {10.1145/3449639.3459278},
booktitle = {Proceedings of the Genetic and Evolutionary Computation Conference},
pages = {830–839},
series = {GECCO '21}
}
If you use this code, please cite the works above.
pip install ..
numpy, scikit-learn, simplegp (see the setup.py file).