TorchRL

.. image:: https://travis-ci.com/activatedgeek/torchrl.svg?branch=master :target: https://travis-ci.com/activatedgeek/torchrl

.. image:: https://badge.fury.io/py/torchrl.svg :target: https://pypi.org/project/torchrl/

.. image:: https://img.shields.io/badge/status-beta-green.svg :target: https://pypi.org/project/torchrl/

.. image:: https://img.shields.io/pypi/pyversions/torchrl.svg :target: https://github.com/activatedgeek/torchrl

+-----------------------------------------------+---------------------------------------------------------+ | Docs: <https://torchrl.sanyamkapoor.com>_ | Github: <https://github.com/activatedgeek/torchrl>_| +-----------------------------------------------+---------------------------------------------------------+

TorchRL provides highly modular and extensible approach to experimenting with Reinforcement Learning. All this is accessible via a simple programmatic interface with experiment specifications.

Objectives

• Modularity in the RL pipeline
• Clean implementations of fundamental ideas
• Fast Experimentation
• Scalability
• Low bar and High ceiling

Install

.. code:: shell

pip install torchrl

Install from source for the latest changes that have not been published to PyPI.

.. code:: shell

pip install https://github.com/activatedgeek/torchrl/tarball/master