DQN

Implements a Double Dueling Deep Q-Network (DQN) in PyTorch.

The following agent was trained on the pixel input of the Atari 2600 Pong environment from OpenAI gym:

Name		Name	Last commit message	Last commit date
Latest commit History 8 Commits
.gitignore		.gitignore
README.md		README.md
atari_wrappers.py		atari_wrappers.py
dqn.py		dqn.py
evaluation.ipynb		evaluation.ipynb
model.py		model.py
pong_500k.gif		pong_500k.gif
training.ipynb		training.ipynb
utils.py		utils.py

Provide feedback