Reinforcement learning model masters atari games without prior knowledge

Built for a lab experiment, this RL agent aims to maximise rewards. The environment is created using the gym library from Open AI. The input for the agent is a 256 by 256 series of images.

Installation

pip install gym
pip install tensorflow
pip install cv2

Usage

cd into_directory
python main.py