Computer Science Specialization Project focused on Reinforcement Learning.
To run this code Python3, TensorFlow, gym and matplotlib are necessary. Then the following command will run the code: $ python3.5 main_full_agent.py
It is possible to change the environment by changing the variable 'envName' in the code. It is also possible to change the value of any configuration in the main file that will allow to run different algorithm.
This code will run a Reinforcement Learning algorithm on the CartPole environment from OpenAI Gym. It will produce on the standard output the rewards per episodes. The game will also be rendered and a plot will also appear to show the evolution of the rewards