Playing-paddle-ball-game-using-Deep-Q-Learning-algorithm

date: 2019.07.09

Reference source

https://github.com/azhar2205/paddle-ball-using-dqlearn

http://edersantana.github.io/articles/keras_rl/

Thank azhar2205 who posted code on GitHub.

I changed the reward value and tested training model, it's more effective for training.

Training and Testing video

https://youtu.be/0N6VUMZLGzA

Pre-requisites

• Python (tested on 3.6)
• Keras
• Theano/Tensorflow
• pygame

How to training and testing

1.Run "train_paddle_ball.py"

2.Run "test_paddle_ball.py"

There is a backup final model in Folder"Final model".