shuffle_gm_sim: Shuffle graph matching simulation for one graph run in...

In dpmcsuss/gmmle:

Description

For a given graph and probability p, this procedure randomly flips edges with probability p and then matches the noisy graph to the original starting at the "true" correspondence. This procedure is run nmc times for each p in p_grid. See local_gm_errors. par_run_all_gm runs this once for each graph in the package, corresponding to the simulation in the manuscript.

Usage

shuffle_gm_sim(
  graph,
  cl = NULL,
  p_grid = 10^(-seq(0.5, 4, length.out = 8)),
  nmc = 30,
  ask = TRUE
)

par_run_all_gm(
  data_df = NULL,
  nmc = 30,
  p_grid = 10^(-seq(0.5, 4, length.out = 8)),
  ask = TRUE
)

Arguments

graph	Graph to run on
cl	Cluster generated by parallel::makeCluster. If NULL then not run in parallel.
p_grid	Grid of error probabilities
nmc	number of Monte Carlo replicates
ask	Whether to warn about long run times.
data_df	A tibble that is a subset of gmmle_data.

Value

A list of lists of graph mathching results for each p in p_grid

Functions

• par_run_all_gm:

Examples

shuffle_gm_sim(karate, p_grid = c(0.01, 0.1),
  nmc = 2, ask = FALSE)

dpmcsuss/gmmle documentation built on July 2, 2020, 6:24 p.m.