Highly Modular and Scalable Reinforcement Learning
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
docs
experiments
notebooks
tests
torchrl
.gitignore
.pylintrc
.travis.yml
LICENSE
MANIFEST.in
README.rst
_redirects
pytest.ini
requirements.txt
requirements_extra.txt
setup.py

README.rst

TorchRL

https://travis-ci.com/salmanazarr/torchrl.svg?branch=master
Docs: https://torchrl.sanyamkapoor.com Github: https://github.com/salmanazarr/torchrl

TorchRL provides highly modular and extensible approach to experimenting with Reinforcement Learning. It allows for a registry based approach to running experiments, allows easy checkpointing, and updating hyper parameter sets. All this is accessible via a programmatic interface as well as a friendly CLI.

Objectives

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

Install

pip install torchrl

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

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

This installs the torchrl package and the torchrl CLI.