dm_control: DeepMind Infrastructure for Physics-Based Simulation.

DeepMind's software stack for physics-based simulation and Reinforcement Learning environments, using MuJoCo physics.

An introductory tutorial for this package is available as a Colaboratory notebook:

Overview

This package consists of the following "core" components:

• dm_control.mujoco: Libraries that provide Python bindings to the MuJoCo physics engine.

• dm_control.suite: A set of Python Reinforcement Learning environments powered by the MuJoCo physics engine.

• dm_control.viewer: An interactive environment viewer.

Additionally, the following components are available for the creation of more complex control tasks:

• dm_control.mjcf: A library for composing and modifying MuJoCo MJCF models in Python.

• dm_control.composer: A library for defining rich RL environments from reusable, self-contained components.

• dm_control.locomotion: Additional libraries for custom tasks.

• dm_control.locomotion.soccer: Multi-agent soccer tasks.

If you use this package, please cite our accompanying tech report:

@misc{tassa2020dmcontrol,
    title={dm_control: Software and Tasks for Continuous Control},
    author={Yuval Tassa and Saran Tunyasuvunakool and Alistair Muldal and
            Yotam Doron and Siqi Liu and Steven Bohez and Josh Merel and
            Tom Erez and Timothy Lillicrap and Nicolas Heess},
    year={2020},
    eprint={2006.12983},
    archivePrefix={arXiv},
    primaryClass={cs.RO}
}

Requirements and Installation

dm_control is regularly tested on Ubuntu 16.04 against the following Python versions:

• 3.7
• 3.8
• 3.9

Various people have been successful in getting dm_control to work on other Linux distros, macOS, and Windows. We do not provide active support for these, but will endeavour to answer questions on a best-effort basis.

Follow these steps to install dm_control:

1. Download MuJoCo 2.1.1 from the Releases page on the MuJoCo GitHub repository. MuJoCo must be installed before dm_control, since dm_control's install script generates Python ctypes bindings based on MuJoCo's header files. By default, dm_control assumes that MuJoCo is installed via the following instructions:

   • On Linux, extract the tarball into ~/.mujoco.
   • On Windows, extract the zip archive into either %HOMEPATH%\MuJoCo or %PUBLIC%\MuJoCo.
   • On macOS, either place MuJoCo.app into /Applications, or place MuJoCo.Framework into ~/.mujoco.

2. Install the dm_control Python package by running pip install dm_control. We recommend pip installing into a virtualenv, or with the --user flag to avoid interfering with system packages. At installation time, dm_control looks for the MuJoCo headers at the paths described in Step 1 by default, however this path can be configured with the headers-dir command line argument.

3. If the shared library provided by MuJoCo (i.e. libmujoco.so.2.1.1 or libmujoco.2.1.1.dylib or mujoco.dll) is installed at a non-default path, specify its location using the MJLIB_PATH environment variable. This environment variable should be set to the full path to the library file itself, e.g. export MJLIB_PATH=/path/to/libmujoco.so.2.1.1.

Note: dm_control cannot be installed in "editable" mode (i.e. pip install -e) due to the way in which we automatically generate the ctypes bindings for MuJoCo. Installing it in this way will result in import errors like:

ImportError: cannot import name 'constants' from partially initialized module 'dm_control.mujoco.wrapper.mjbindings' ...

The solution is to pip uninstall dm_control and then reinstall it without the -e flag.

Versioning

dm_control is released on a rolling basis: the latest commit on the master branch of our GitHub repository represents our latest release. Our Python package is versioned 0.0.N, where N is the number that appears in the PiperOrigin-RevId field of the commit message. We always ensure that N strictly increases between a parent commit and its children. We do not upload all versions to PyPI, and occasionally the latest version on PyPI may lag behind the latest commit on GitHub. Should this happen, you can still install the newest version available by running pip install git+git://github.com/deepmind/dm_control.git.

Rendering

The MuJoCo Python bindings support three different OpenGL rendering backends: EGL (headless, hardware-accelerated), GLFW (windowed, hardware-accelerated), and OSMesa (purely software-based). At least one of these three backends must be available in order render through dm_control.

• Hardware rendering with a windowing system is supported via GLFW and GLEW. On Linux these can be installed using your distribution's package manager. For example, on Debian and Ubuntu, this can be done by running sudo apt-get install libglfw3 libglew2.0. Please note that:

  • dm_control.viewer can only be used with GLFW.
  • GLFW will not work on headless machines.

• "Headless" hardware rendering (i.e. without a windowing system such as X11) requires EXT_platform_device support in the EGL driver. Recent Nvidia drivers support this. You will also need GLEW. On Debian and Ubuntu, this can be installed via sudo apt-get install libglew2.0.

• Software rendering requires GLX and OSMesa. On Debian and Ubuntu these can be installed using sudo apt-get install libgl1-mesa-glx libosmesa6.

By default, dm_control will attempt to use GLFW first, then EGL, then OSMesa. You can also specify a particular backend to use by setting the MUJOCO_GL= environment variable to "glfw", "egl", or "osmesa", respectively. When rendering with EGL, you can also specify which GPU to use for rendering by setting the environment variable EGL_DEVICE_ID= to the target GPU ID.

Additional instructions for Homebrew users on macOS

1. The above instructions using pip should work, provided that you use a Python interpreter that is installed by Homebrew (rather than the system-default one).

2. Before running, the DYLD_LIBRARY_PATH environment variable needs to be updated with the path to the GLFW library. This can be done by running export DYLD_LIBRARY_PATH=$(brew --prefix)/lib:$DYLD_LIBRARY_PATH.