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.
Use the package manager pip to install the latest stable release of pyRLprob, with all its dependencies:
pip install pyrlprobTo 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.
pyRLprob has been created by Lorenzo Federici in 2021. For any problem, clarification or suggestion, you can contact the author at lorenzof@arizona.edu.
The package is under the MIT license.