Apply a split and/or merge strategy of the clustering in a path of models in a collection of SBM ordered by number of block. The goal is to find better initialization. This results in a "smoothing" of the ICL, that should be close to concave.
smooth(Robject, type = c("forward", "backward", "both"), control = list())an object with class missSBM_collection, i.e. an output from estimateMissSBM()
character indicating what kind of ICL smoothing should be use among "forward", "backward" or "both". Default is "forward".
a list controlling the variational EM algorithm. See details.
An invisible missSBM_collection, in which the ICL has been smoothed
The list of parameters control controls the optimization process and the variational EM algorithm, with the following entries
"iterates": integer for the number of iterations of smoothing. Default is 1.
"threshold": V-EM algorithm stops stop when an optimization step changes the objective function by less than threshold. Default is 1e-3.
"maxIter": V-EM algorithm stops when the number of iteration exceeds maxIter. Default is 100 with no covariate, 50 otherwise.
"fixPointIter": number of fix-point iterations in the V-E step. Default is 5 with no covariate, 2 otherwise.
"cores": integer for number of cores used. Default is 1.
"trace": integer for verbosity. Useless when cores > 1