Deep Q network for a simple game

This project presents how to use Double deep Q network to train an agent to play a very simple game of vacuum world.

How to

Run trainDDQN_tf.py to train a model for 1000 episodes, then run testDDQN_tf.py to test on random board configs.

Name		Name	Last commit message	Last commit date
Latest commit History 12 Commits
assets		assets
LICENSE		LICENSE
README.md		README.md
controller.py		controller.py
main.py		main.py
model_tf.py		model_tf.py
test2.gif		test2.gif
testDDQN_tf.py		testDDQN_tf.py
test_dqn.py		test_dqn.py
trainDDQN_tf.py		trainDDQN_tf.py
train_dqn.py		train_dqn.py
visulization.py		visulization.py