An optimized RL approach to learn and simulate the Job-Shop Scheduling problem using Constraint Programming
Project description
A Constraint Programming Based Job-Shop Scheduling Environment
This is a Constraint Programming (CP) based Job-Shop Scheduling (JSS) Environment that can be combined with Reinforcement-Learning (RL) to train an agent to solve a scheduling problem.
Compared to other environments, this CP based one is made to be fast and scalable.
If you install it using setup.py, it will automatically compile using MyPyC.
Also, this environment has been developed for an end-to-end approach, there is no pre-defined reward function and the observation are simply the raw IntervalVariable representation.
We recommend the reader to check the paper for more information.
If you need to define a reward function or a different observation, you can do so by forking the environment and modifying the step function.
Installation
pip install jss_cp
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 job_shop_cp_env-1.0.0-cp310-cp310-macosx_11_0_arm64.whl
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 | d11f21aa3a39ae873a902d5d452909efbffdb29e0b94f506fe3343a1886c549b |
|
| MD5 | a7b00c6f59373d5ef43ede11cabba705 |
|
| BLAKE2b-256 | b7ac1042bc3618306a65c591b32963490e142594ec8dbdb82288877b65e98f71 |
