A framework for learning about and experimenting with reinforcement learning algorithms
A framework for learning about and experimenting with reinforcement learning algorithms. It is built on top of TensorFlow and TFLearn and is interfaces with the OpenAI gym (universe should work, too). It aims to be as modular as possible so that new algorithms and ideas can easily be tested. I started it to gain a better understanding of core RL algorithms and maybe it can be useful for others as well.
Algorithms (future algorithms italicized):
Temporal Difference Learning
Policy gradient Q-learning
Vanilla policy gradient
Deterministic policy gradient
Natural policy gradient
Cross entropy method
Function approximators (defined by TFLearn model):
Works with any OpenAI gym environment.
Improved TensorBoard logging
Improved model snapshotting to include exploration states, memories, etc.
Errors / warnings on TensorFlow session save
Free software: MIT license
First release on PyPI.
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.