This package describes an interface for Python reinforcement learning (RL) environments. It consists of the following core components:

• dm_env.Environment: An abstract base class for RL environments.
• dm_env.TimeStep: A container class representing the outputs of the environment on each time step (transition).
• dm_env.specs: A module containing primitives that are used to describe the format of the actions consumed by an environment, as well as the observations, rewards, and discounts it returns.
• dm_env.test_utils: Tools for testing whether concrete environment implementations conform to the dm_env.Environment interface.

Please see the documentation here for more information about the semantics of the environment interface and how to use it. The examples subdirectory also contains illustrative examples of RL environments implemented using the dm_env interface.

dm_env can be installed from PyPI using pip:

pip install dm-env

You can also install it directly from our GitHub repository using pip:

pip install git+git://github.com/deepmind/dm_env.git

or alternatively by checking out a local copy of our repository and running:

pip install /path/to/local/dm_env/

To cite this repository:

@misc{dm_env2019,
  author={Alistair Muldal and
      Yotam Doron and
      John Aslanides and
      Tim Harley and
      Tom Ward and
      Siqi Liu},
  title={dm\_env: A Python interface for reinforcement learning environments},
  year={2019},
  url={http://github.com/deepmind/dm_env}
}