Extension to OpenAI Gym interface for building energy optimisation allowing diverse controllers, including RL and MPC.
Project description
A toolbox for benchmarking reinforcement learning (RL) algorithms on building energy optimisation (BEO) problems. Beobench does not replace existing libraries defining BEO problems (such as BOPTEST) — instead it makes working with them easier.
Features
Wide range of RL algorithms: test the most common RL algorithms on BEO problems without re-implementing by using beobench’s Ray RLlib integration.
Experiment logging: log experiment results in a reproducible and sharable manner via Weights and Biases.
Hyperparameter tuning: easily tune hyperparameters using the extensive Ray Tune syntax.
Installers: avoid having to manage messy Python namespaces yourself — just install beobench via pip and use its pre-configured docker containers to take care of managing other BEO packages and their dependencies.
Documentation
License
MIT license
Credits
This package was originally created with Cookiecutter and the audreyr/cookiecutter-pypackage project template.
History
0.1.0 (2021-10-28)
First release on PyPI.
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Hashes for beobench-0.1.0-py2.py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 5b3d86b5ede6dc7a873e2ddfe816f4972b104f6c462790223a14ed5c8530456c |
|
MD5 | 570d82255af192975b3d9b6d71662ce0 |
|
BLAKE2b-256 | 9e66a28c4ff81704c5140c1f5bacb77fae27f9901f6f4e8f06ec843051059ab1 |