blackvitriol / RL

Reinforcement Learning

Reinforcement Learning

RL is becoming my favoutire algorithm to train intelligent agents. I am developing a general AI agent that will learn from me and will be able to play any game.

This repo contains code for agents that I have trained on OpenAI's Gym or Universe and other virtual environments.

Running

OS X & Linux & Windows::

python *filename*.py

All files have been developed wholesomely, meaning that you have to run only one file for one project.

Usage example

Can be run directly or in a Jupyter Notebook.

Dependencies: Open-AI Gym

pip install gym

Results:

1. Cart-Mountain

2. Pong

2. Pendulum