Play Super Mario by Reinforcement Learning

I am learing reinforcement learning and applying it to the Super Mario.

The performance of the project is still terrible.

Name		Name	Last commit message	Last commit date
Latest commit History 3 Commits
project		project
README.md		README.md
data.py		data.py
env.py		env.py
loss.py		loss.py
model.py		model.py
storage.py		storage.py
test.py		test.py
train.py		train.py

Provide feedback