DamiPayne / Pong-RNN-Implementation

Using pygame to create a 2d pong game, then using gym and tensorflow to read the pixels on the screen using a CNN and then model the actions with a Qlearning RNN to beat the ai opponent

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Pong-RNN-Implementation

Using pygame to create a 2d pong game and then using gym and tensorflow to read the pixels on the screen using a CNN then model the actions with a Qlearning RNN to beat the ai opponent

Dependencies

  • Gym
  • Tensorflow (manually compile for best performance)
  • Pygame
  • opencv 3 (You need to manually compile)

About

Using pygame to create a 2d pong game, then using gym and tensorflow to read the pixels on the screen using a CNN and then model the actions with a Qlearning RNN to beat the ai opponent

License:MIT License


Languages

Language:Python 100.0%