Genetic algorithm generating AI capable to play Atari2600 games.
## Getting Started
These instructions will get you a copy of the project up and running on your local machine for development or training purposes.
### Prerequisites
In order to use this algorithm you need a compiler that support c++ 14 and the boost library (`apt-get install libboost-all-dev`).
### Installing
Once this git repository is cloned, you need to set the environment variable LD_LIBRARY_PATH.
Writing this line in your .bashrc is a good way to achieve it : `export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/my/path/to/ale/directory`
Compile the algorithm by using `make` in the src subfolder.
## Training agents
Once you have set the value `algoGen` to true in the config file, type `./atari -m` in a terminal to launch the master. Type `./iatari -s` in other terminals to launch slaves.
## Exhib agents
Set `algoGen` to false and `displayScreen` to true in the config file and start the algorithm without arguments.
## Documentation
The documentation is available as a website in the subfolder doc/html.
## Build With
* This algorithm use the Arcade Learning Environment : [ALE repository](https://github.com/mgbellemare/Arcade-Learning-Environment)
* The documentation is build with Doxygen : [Doxygen website](http://www.doxygen.nl)
