gym_tf

DQN algorithm implementation with openai gym and tensorflow

Papers

DQN
Double DQN
Prioritized Experience Replay(removed)
Dueling Network

Prioritized Replay does not work
Model based DQN seems not working
Tensoflow is on native backend(on my laptop 850m), should be on the server

Name		Name	Last commit message	Last commit date
Latest commit History 56 Commits
DQN		DQN
flappy_bird		flappy_bird
.gitattributes		.gitattributes
.gitignore		.gitignore
README.md		README.md
main.py		main.py