Metadata-Version: 2.1
Name: Tensorforce
Version: 0.5.0
Summary: Tensorforce: a TensorFlow library for applied reinforcement learning
Home-page: http://github.com/tensorforce/tensorforce
Author: Alexander Kuhnle
Author-email: tensorforce.team@gmail.com
License: Apache 2.0
Download-URL: https://github.com/tensorforce/tensorforce/archive/0.5.0.tar.gz
Platform: UNKNOWN
Requires-Python: >=3.5
Description-Content-Type: text/markdown
Requires-Dist: numpy
Requires-Dist: matplotlib
Requires-Dist: pytest
Requires-Dist: tqdm
Provides-Extra: docs
Requires-Dist: m2r ; extra == 'docs'
Requires-Dist: recommonmark ; extra == 'docs'
Requires-Dist: sphinx ; extra == 'docs'
Requires-Dist: sphinx-rtd-theme ; extra == 'docs'
Provides-Extra: envs
Requires-Dist: gym[all] ; extra == 'envs'
Requires-Dist: gym-retro ; extra == 'envs'
Requires-Dist: mazeexp ; extra == 'envs'
Requires-Dist: vizdoom ; extra == 'envs'
Provides-Extra: gym
Requires-Dist: gym[all] ; extra == 'gym'
Provides-Extra: mazeexp
Requires-Dist: mazeexp ; extra == 'mazeexp'
Provides-Extra: retro
Requires-Dist: gym-retro ; extra == 'retro'
Provides-Extra: tf
Requires-Dist: tensorflow (==1.13.1) ; extra == 'tf'
Provides-Extra: tf_gpu
Requires-Dist: tensorflow-gpu (==1.13.1) ; extra == 'tf_gpu'
Provides-Extra: vizdoom
Requires-Dist: vizdoom ; extra == 'vizdoom'

# Tensorforce: a TensorFlow library for applied reinforcement learning
Tensorforce is an open-source deep reinforcement learning framework, with an emphasis on modularized flexible library design and straightforward usability for applications in research and practice. Tensorforce is built on top of [Google's TensorFlow framework](https://www.tensorflow.org/) and compatible with Python 3 (Python 2 support was dropped with version 0.5).

Tensorforce follows a set of high-level design choices which differentiate it from other similar libraries:

- **Modular component-based design**: Feature implementations, above all, strive to be as generally applicable and configurable as possible, potentially at some cost of faithfully resembling details of the introducing paper.
- **Separation of RL algorithm and application**: Algorithms are agnostic to the type and structure of inputs (states/observations) and outputs (actions/decisions), as well as the interaction with the application environment.
- **Full-on TensorFlow models**: The entire reinforcement learning logic, including control flow, is implemented in TensorFlow, to enable portable computation graphs independent of application programming language, and to facilitate the deployment of models.


