This repo is still under development. We are also actively looking for users and developers. If this sounds like you, get in touch!

Documentation#

https://raw.githubusercontent.com/jjshoots/PyFlyt/master/readme_assets/pyflyt_cover_photo.png

Installation#

We recommend installations using Python virtual environments. It is possible to install PyFlyt using conda, but YMMV.

Linux and MacOS#

Installation on Linux and MacOS is simple:

pip3 install wheel numpy
pip3 install pyflyt

numpy and wheel must be installed prior to pyflyt such that pybullet is built with numpy support.

Windows#

  1. First, install Microsoft Visual Studio Build Tools.

    • Go here, scroll down to All Downloads, expand Tools for Visual Studio, download Build Tools for Visual Studio 20XX, where XX is just the latest year available.

    • Run the installer.

    • Select Desktop development with C++, then click Install while downloading or the alternate option if you wish.

  2. Now, you can install PyFlyt as usual:

    pip3 install wheel numpy
    pip3 install pyflyt
    

Gymnasium Environments#

If all you want are the Gymnasium environments, you can skip everything and go straight to the Gymnasium Environments section.

PettingZoo Environments#

Similarly, if all you want are the PettingZoo environments, you can skip everything and go straight to the PettingZoo Environments section.

I want to do more!#

A set of helpful examples are provided in the source repository. While this documentation provides a detailed overview of PyFlyt, the examples can bring you 80% of the way to a fully fledged user in a couple hours. New users are highly encouraged to go through them first.