Skip to content

SamuelDiai/SGL

Repository files navigation

Structured Graph Learning via Laplacian Spectral Constraints for Timeseries

This repository contains the work of Paul Martin and Samuel Diai for the Project 6.8 of the Course "Time Series Learning" of Laurent Oudre.

In this project, we studied a novel approach to learn a structured Graph with Laplacian constraints.

Our work is directly adapted from the article "Structured Graph Learning via Laplacian Spectral Constraints" of Sandeep Kumar, Jiaxi Ying, Jose Vinicius de M. Cardoso and Daniel P. Palomar.

alt text

The repository contains :

  • The code of the Structured Graph Learning algorithm (SGL) - SGL.py
  • Some tests made on very simple datasets (two moons, circles and blops) - basic_experiments.py
  • An experiment on an animal dataset - animals.py
  • An experiment on a cancer genome dataset - cancer.py
  • A notebook (SGL_experiments.ipynb) where we do the above experiments and plot the results.
  • A notebook (SGL_timeseries.ipynb) where we apply the SGL algorithm to timeseries data. It is directly adapted from the assignment n°6 of the course "Time Series Learning" by Charles Truong.
  • The final report of our project.

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published