Reference guide

Toolbox overview

This toolbox is splitted in different modules taking care of the different aspects of Graph Signal Processing.

Those modules are : Graphs, Filters, Operators and PointClouds.

You can find detailed documentation on the use of the functions in the subsequent pages.


This module implements graphs and contains predefined graphs for the most famous ones.

A graph is constructed either from its adjacency matrix, its weight matrix or any other parameter which depends on the particular graph you are trying to build. For specific information, see details here.


This module implements filters and contains predefined filters that can be directly applied to graphs.

A filter is associated to a graph and is defined with one or several function(s). We define by Filterbank a list of filters applied to a single graph. Tools for the analysis, the synthesis and the evaluation are provided to work with the filters on the graphs. For specific information, see details here.


This module implements some PointsClouds.


This module implements plotting functions for the PyGSP main objects.


This module provides optimization tools to accelarate graph signal processing as a whole.


This module implements some utilitary functions used throughout the PyGSP box.