cgpde-lib-imbalanced

Modified Version of the CGP Library

Hybridization of Cartesian Genetic Programming and Differential Evolution to generate Artificial Neural Networks.
The methods are applied to imbalanced data classification problems by using different objective functions: accuracy, G-mean, F-score, and area under the ROC curve (AUC). It includes the CGPDE-IN, CGPDE-OUT-T, and CGPDE-OUT-V methods.

Author: Johnathan M Melo Neto
Email: jmmn.mg@gmail.com

Credits of the original work are placed below.

CGP Library

A cross platform Cartesian Genetic Programming Library written in C.

Author: Andrew James Turner
Webpage: http://www.cgplibrary.co.uk/
Email: andrew.turner@york.ac.uk
License: Lesser General Public License (LGPL)

If this library is used in published work I would greatly appreciate a citation to the following:

A. J. Turner and J. F. Miller. Introducing A Cross Platform Open Source Cartesian Genetic Programming Library. The Journal of Genetic Programming and Evolvable Machines, 2014, 16, 83-91.

To Install

On Linux

From Source

First you'll want to clone the repository:

git clone https://github.com/johnathanmelo/cgpde-lib-imbalanced.git

Once that's finished, navigate to the Root directory. In this case it would be ./cgpde-lib-imbalanced:

cd ./cgpde-lib-imbalanced

Then run Makefile:

make main

Now you can run the algorithms by running:

./main