Evolutionary Ensemble Optimizer

Optimizer for classifier ensembles based on Genetic Algorithm techniques. Given a trained ensemble of classifiers, the Evolutionary Ensemble Classifier is capable of finding best and more accurate subset of classifiers through evolutionary methods.

Usage

The core file for the genetic optimization operation is the optimizer.py file. There is an example on how to use the system using a Bagging Classifier with Decision Trees in the file bagging_example.py. To change genetic algorithm parameters you need to open and modify bagging_example.py. In order to run the system:

$ python bagging_example.py

Dependencies

The Evolutionary Ensemble Optimizer has the following dependencies:

• Python 3
• Scikit Learn
• Colorama