Train Gym-derived environments in Python/C++ through Ray RLlib
Project description
PyRLprob is an open-source python library for training, evaluation, and postprocessing of Gym-based environments, written in Python, through Ray-RLlib reinforcement learning library.
Installation
After installing Anaconda, you can create a virtual environment with a specific Python version (3.9.x) by using conda:
conda create --name myenv python=3.9
where myenv is the name of the environment, and activate the environment with:
conda activate myenv
Then, use the package manager pip to install the latest stable release of pyRLprob, with all its dependencies:
pip install pyrlprob
To test if the package is installed correctly, run the following tests:
from pyrlprob.tests import *
test_train_eval_py()
If the code exits without errors, a folder named results/ with the test results will be created in your current directory.
User Guide
Credits
pyRLprob has been created by Lorenzo Federici in 2021. For any problem, clarification or suggestion, you can contact the author at lorenzof@arizona.edu.
License
The package is under the MIT license.
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.
