https://github.com/google-deepmind/dm_env
A Python interface for reinforcement learning environments
https://github.com/google-deepmind/dm_env
api deep-learning interface machine-learning python reinforcement-learning
Last synced: 14 days ago
JSON representation
A Python interface for reinforcement learning environments
- Host: GitHub
- URL: https://github.com/google-deepmind/dm_env
- Owner: google-deepmind
- License: apache-2.0
- Created: 2019-07-08T13:58:50.000Z (over 6 years ago)
- Default Branch: master
- Last Pushed: 2022-12-23T10:49:57.000Z (about 3 years ago)
- Last Synced: 2024-09-17T22:20:55.788Z (over 1 year ago)
- Topics: api, deep-learning, interface, machine-learning, python, reinforcement-learning
- Language: Python
- Homepage:
- Size: 104 KB
- Stars: 344
- Watchers: 14
- Forks: 43
- Open Issues: 4
-
Metadata Files:
- Readme: README.md
- Changelog: CHANGELOG.md
- Contributing: CONTRIBUTING.md
- License: LICENSE
Awesome Lists containing this project
README
# `dm_env`: The DeepMind RL Environment API
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][api_docs] 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.
## Installation
`dm_env` can be installed from PyPI using `pip`:
```bash
pip install dm-env
```
Note that from version 1.4 onwards, we support Python 3.6+ only.
You can also install it directly from our GitHub repository using `pip`:
```bash
pip install git+git://github.com/deepmind/dm_env.git
```
or alternatively by checking out a local copy of our repository and running:
```bash
pip install /path/to/local/dm_env/
```
[api_docs]: docs/index.md
[examples]: examples/
## Citing
To cite this repository:
```bibtex
@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}
}
```