Schlably

A Python-Based Scheduling Laboratory for Reinforcement Learning Solutions

Schlably is a Python-Based framework for experiments on scheduling problems with Deep Reinforcement Learning (DRL). It features an extendable gym environment and DRL-Agents along with code for data generation, training and testing.

Schlably was developed such that modules may be used as they are, but also may be customized to fit the needs of the user. While the framework works out of the box and can be adjusted through config files, some changes are intentionally only possible through changes to the code. We believe that this makes it easier to apply small changes without having to deal with complex multi-level inheritances.

Also check out our:

Github Repository

ArXiv paper

Main Features

• Python-based

• Includes JSSP, FJSSP and the option for resource constrainted scheduling problems

• Includes DRL-Agents

• Easy to use

• Easy to extend and adjust

User Guide

• Installation
• Quickstart
  • Data Generation
  • Training
  • Testing
  • Advanced config handling
• Tutorials
  • Custom Problem Settings
  • Looping over Configuration Parameters
  • Weights & Biases sweeps
• API Reference
  • Reinforcement Learning Agents
  • Reinforcement Learning Functions
  • Environments
  • Visuals Generator
  • Data Generator
  • Solvers
  • Benchmark Datasets

Indices and tables

• Index

• Module Index

• Search Page