A Reinforcement Learning Library for Research and Education

Try it on Google Colab!

---

Section	Description
Goals	The philosophy of rlberry
Installation	How to install rlberry
Getting started	A quick usage guide of rlberry
Documentation	A link to the documentation
Contributing	A guide for contributing
Citation	How to cite this work

Goals

• Write detailed documentation and comprehensible tutorial/examples (Jupyter Notebook) for each implemented algorithm.

• Provide a general interface for agents, that

  • puts minimal constraints on the agent code (=> making it easy to include new algorithms and modify existing ones);

  • allows comparison between agents using a simple and unified evaluation interface (=> making it easy, for instance, to compare deep and "traditional" RL algorithms).

• Unified seeding mechanism: define only one global seed, from which all other seeds will inherit, enforcing independence of the random number generators.

• Simple interface for creating and rendering new environments.

Installation

Cloning & creating virtual environment

It is suggested to create a virtual environment using Anaconda or Miniconda:

git clone https://github.com/rlberry-py/rlberry.git
conda create -n rlberry python=3.7

Basic installation

Install without heavy libraries (e.g. pytorch).

conda activate rlberry
pip install -e .

Full installation

Install with all features,

conda activate rlberry
pip install -e .[full]

which includes:

• Numba for just-in-time compilation of algorithms based on dynamic programming,
• PyTorch for Deep RL agents,
• Optuna for hyperparameter optimization,
• ffmpeg-python for saving videos,
• PyOpenGL for more rendering options.

Getting started

Tests

To run tests, install test dependencies with pip install -e .[test] and run pytest. To run tests with coverage, install test dependencies and run bash run_testscov.sh. See coverage report in cov_html/index.html.

Documentation

The documentation is under construction and will be available here.

Contributing

Want to contribute to rlberry? Please check our contribution guidelines. A list of interesting TODO's will be available soon. If you want to add any new agents or environments, do not hesitate to open an issue!

Implementation notes

• When inheriting from the Agent class, make sure to call Agent.__init__(self, env, **kwargs) using **kwargs in case new features are added to the base class, and to make sure that copy_env and reseed_env are always an option to any agent.

• Convention for verbose in the agents:

  • verbose=0: nothing is printed
  • verbose>1: print progress messages

Errors and warnings are printed using the logging library.

Citing rlberry