This is an implementation of 3 different agents in addition to a random agent that can all play the Atari game Berzerk. The following is a list of necessary pip installs:
pip install neat-python
pip install scikit-image
pip install astar
This is an implementation of the reflex algorithm to play the game. To run it, change the agent in line 45 of gym_play.py to ReflexAgent.
This is an implementation of the Neuroevolution of Augmented Topologies algorithm to play the game. To run it, change the agent in line 45 of gym_play.py to NEATAgent. To train, run gym_train_multi.py
This is an agent that just makes random moves. To run it, change the agent in line 45 of gym_play.py to RandomAgent. ``