In robertness/signalgraph: Signal Processing with Graphs

Inherits from the igraph class

Signalgraph inherits from the igraph class, and all igraph methods work on a signalgraph object.

Building a signalgraph

Start with an input graph and data. The variables in the data must match vertices in the graph. However some vertices may have no data - i.e. they are hidden variables.

Here, I simulate a graph and dataset. The graph has 10 nodes, 5 of which are observed in the data. The data has 10 observations.

library(signalgraph)

g <- random_unfit_sg(10, 5, 10)
sg_viz(g, show_biases = FALSE)

The green nodes are observed and the blue nodes are hidden in the data.

data <- recover_design(g)
head(data)

robertness/signalgraph documentation built on May 27, 2019, 10:33 a.m.