This is a version of NeuroEvolution of Augmenting Topologies (NEAT) made by me. It's far from done but you can try it out using the pong sample anyway.
The Pong Sample will evolve the networks to play Pong. It will continue evolving te networks until one achieves a perfect 10/10 game, and then shows that network on the screen playing matches.
- A Linux distro
- Code::Blocks
- SFML
- Install SFML
sudo apt-get install libsfml-dev - Install Code::Blocks
sudo apt-get install codeblocks - Download or clone the repository
- Navigate into the main folder of the repository
- Open My_Neat.cbp
- Compile and run in Code::Blocks
NEAT is a method proposed by Ken Stanley et al. for optimizing neural networks. It adjusts the topology of the neural networks as well as their weights.