rlgraph 0.5.5

A Framework for Modular Deep Reinforcement Learning

RLgraph is a framework to quickly prototype, define and execute reinforcement learning algorithms both in research and practice. RLgraph supports both TensorFlow (or static graphs in general) and Pytorch (eager/define-by run execution) through a single component based interface. An introductory blogpost can be found here: https://rlgraph.github.io/rlgraph/2019/01/04/introducing-rlgraph.html