multipartiteBMFixedModel {GREMLINS} | R Documentation |
Model selection and estimation of multipartite blockmodels
Description
Estimate the parameters and give the clustering for given numbers of blocks
Usage
multipartiteBMFixedModel(
list_Net,
v_distrib,
namesFG,
v_K,
classifInit = NULL,
nbCores = NULL,
maxiterVE = NULL,
maxiterVEM = NULL,
verbose = TRUE
)
Arguments
list_Net |
A list of network (defined via the function DefineNetwork) |
v_distrib |
Type of proababilistic distributions in each network : if 0/1 then bernoulli, if counting then poisson, gaussian or Zero Inflated Gaussian (ZIgaussian) My default = Bernoulli. Must give a vector whose length is the number of networks in list_Net |
namesFG |
Names of functional groups (must correspond to names in listNet) |
v_K |
A vector with the numbers of blocks per functional group |
classifInit |
A list of initial classification for each functional group in the same order as in namesFG |
nbCores |
Number or cores used for the estimation. Not parallelized on windows. By default : half of the cores |
maxiterVE |
Maximum number of iterations in the VE step of the VEM algorithm. Default value = 1000 |
maxiterVEM |
Maximum number of iterations of the VEM algorithm. Default value = 1000 |
verbose |
Set to TRUE to display the current step of the search algorithm |
Value
Estimated parameters and a classification
Examples
namesFG <- c('A','B')
list_pi <- list(c(0.5,0.5),c(0.3,0.7)) # prop of blocks in each FG
E <- rbind(c(1,2),c(2,2)) # architecture of the multipartite net.
typeInter <- c( "inc","diradj")
v_distrib <- c('poisson','bernoulli')
list_theta <- list()
list_theta[[1]] <- matrix(c(6.1, 8.9, 6.6, 3), 2, 2)
list_theta[[2]] <- matrix(c(0.7,1.0, 0.4, 0.6),2, 2)
list_Net <- rMBM(v_NQ = c(20,20),E , typeInter, v_distrib, list_pi,
list_theta, namesFG = namesFG, seed = 2)$list_Net
#res_MBMsimu_fixed <- multipartiteBMFixedModel(list_Net, v_distrib,
# namesFG = namesFG,
# v_K = c(1,2),
# nbCores = 2)