Skip to main content

A python API with zero dependencies to create fully customizable environments for Reinforcement Learning projects.

Project description

The Rocket League Gym

This is a python API that can be used to treat the game Rocket League as though it were an OpenAI Gym-style environment for Reinforcement Learning projects. This API must be used with the accompanying Bakkesmod plugin.

Requirements

  • A Windows 10 PC
  • Rocket League (Both Steam and Epic are supported)
  • Bakkesmod
  • The RLGym plugin for Bakkesmod (It's installed automatically by pip)
  • Python between versions 3.7 and 3.9 (3.10 not supported).

Installation

Install the library via pip:

pip install rlgym[all]  // Installs every rlgym component

pip install rlgym  // Installs only the api

pip install rlgym[rl]  // Installs all rocket league packages

pip install rlgym[rl-sim]  // Installs only RocketSim rocket league packages

If you installed the Rocket League game distribution

Once the API is installed, you will need to enable the RLGym plugin from inside the Bakkesmod plugin manager. To do this, first launch the game, then press F2 to open the Bakkesmod menu. Navigate to the plugins tab and open the Plugin Manager. From there, scroll down until you find the RLGym plugin, and enable it. Close the game when this is done.

Testing everything works

RLGym is now installed! simply run example.py from our repo to ensure everything works.

Dependency Management

Don't specify this package as a dependency, this is an instalation only package.

You should depend directly on the package you're consuming and its corresponding extras, be it rlgym-rocket-league[sim], rlgym-api or any other in particular.

Usage

For tutorials and documentation, please visit our Wiki.

We also provide the base RLGym API in its own standalone package with no dependencies.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

rlgym-2.0.0a1.tar.gz (8.8 kB view hashes)

Uploaded Source

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page