Skip to content

Latest commit

 

History

History
15 lines (8 loc) · 548 Bytes

File metadata and controls

15 lines (8 loc) · 548 Bytes

DiffSG: Diffusion for Solution Generation in Network Optimization

Introduction

This repository contains code for exploratory experiments using generative diffusion models as solution generators for network optimization problems (classifier-free guidance DDPM, in torch).

Files

baselines: The implementation of the baselines used in the article.

datasets: Dataset creation tools and some ready-made datasets.

ddpm_opt: Neural network, model, training and evaluation scripts.

utils: Dataset processing tools, etc.