GitHub - yinchuandong/A3C-FlappyBird: Deep Reinforcement Learning for Flappy Bird, including DQN, DRQN, A3C as benchmarks

this repository is under construction

tensorflow=1.0.0
pygame=1.9.2rc1

create config.PARALLEL_SIZE game agents in different tab
- python game_server.py 0
- python game_server.py 1
- python game_server.py 2
- python game_server.py 3
run A3C training agent
- python a3c.py

Name		Name	Last commit message	Last commit date
Latest commit History 150 Commits
assets		assets
customgame		customgame
game		game
images		images
.gitignore		.gitignore
DQN.py		DQN.py
DRQN.py		DRQN.py
README.md		README.md
a3c.py		a3c.py
a3c_actor_thread.py		a3c_actor_thread.py
a3c_network.py		a3c_network.py
config.py		config.py
dqn2.py		dqn2.py
dqn_custom.py		dqn_custom.py
game_gym_server.py		game_gym_server.py
game_server.py		game_server.py
netutil.py		netutil.py
replay_buffer.py		replay_buffer.py