reinforcement-tetris

Using reinforcement learning and Keras to learn to play Tetris

Todo

Add GUI to really visualize it playing Tetris
Test on some other metrics besides number of deaths
Add support for previous time steps (this seems like it could be tricky - could be done with persistent states in RNN layers maybe)

Name		Name	Last commit message	Last commit date
Latest commit History 11 Commits
.gitignore		.gitignore
README.md		README.md
analytic.py		analytic.py
engine.py		engine.py
train.py		train.py
utils.py		utils.py