makeSBM: makeSBM

In Noobivsho/ssClust: GMM for semi-supervised case a-la Mclust

Description

Generates a realization Stochastic Block Model with information for supervised labels.

Usage

makeSBM(n, m, Lam)

Arguments

n	vector of unsupervised block sizes
m	vector of supervised block sizes
Lam	interblock link probability matrix

Examples

library(ssClust)
numVert=500 #number of ambiguous vertices

#matching Li's simulation parameters
rho <- c(0.4,0.3,.3)
B1 <- matrix(c(0.5,0.3,.4,
               0.3,0.8,.6,
               .4,0.6, .3),  nrow = 3,ncol=3)
B2 <- matrix(.5 ,nrow = 3,ncol=3)
nu = .3
Lam0 = nu*B1 + (1-nu)*B2

n0 <- numVert*rho
m0<-c(20, 0, 0)
numburn0 = 10^4
numsample0 = 10^4

sbm.out = makeSBM(n0,m0,Lam0)
  A0 = as.matrix(sbm.out$A)
  observe0 = sbm.out$observe
  truth0 = sbm.out$truth

Noobivsho/ssClust documentation built on Aug. 10, 2019, 5:47 a.m.