See here for complete documentation.
lrl is a Python package for applying (and hopefully, learning!) basic Reinforcement Learning algorithms. It is intended to be an early stepping stone for someone trying to understanding the basic concepts of planning and learning, providing out-of-the-box implementations of some simple environments and algorithms in a well documented, readable, and digestible way to give someone platform from which to build understanding.
The overall goal of the author in writing this package was to provide people interested in Reinforcement Learning a starting point and handrail to help them as they began learning. The fastest, most efficient implementation is less important here than code which can be read and learned from by someone new to the topic and with intermediate Python skills.
lrl is accessible using pip
pip install lrl
or, you can pull the source to your working directory so you can play along at home
git clone https://github.com/ca-scribner/lrl.git lrl
pip install -e lrl