/UNEAT

An implementation of the NeuroEvolution of Augmenting Topologies algorithm which aims at both building and training artificial neural networks in order to solve the widest possible range of problems.

Primary LanguagePython

UNEAT

First CS project, old (~2015) and ugly code !

An implementation from scratch of a general-purpose algorithm performing NeuroEvolution through Augmenting Topologies (NEAT) in python. It's capable of both generating a neural network whose architecture's evolves during the training process.

Refactoring attempt of this.

Credits

The NEAT algorithm that inspired this project was introduced by Kenneth O. Stanley and Risto Miikkulainen in 2002 in this paper.