An R6 Class used for internal representation of a partially observed network
An R6 Class used for internal representation of a partially observed network
samplingRateThe percentage of observed dyads
nbNodesThe number of nodes
nbDyadsThe number of dyads
is_directedlogical indicating if the network is directed or not
netMatrixThe adjacency matrix of the network
covarArraythe array of covariates
covarMatrixthe matrix of covariates
dyadsa list of potential dyads in the network
missingDyadsarray indices of missing dyads
observedDyadsarray indices of observed dyads
samplingMatrixmatrix of observed and non-observed edges
observedNodesa vector of observed and non-observed nodes
NAsboolean for NA entries in the adjacencyMatrix
new()constructor
partlyObservedNetwork$new( adjacencyMatrix, covariates = NULL, similarity = missSBM:::l1_similarity )
adjacencyMatrixThe adjacency matrix of the network
covariatesA list with M entries (the M covariates), each of whom being either a size-N vector or N x N matrix.
similarityAn R x R -> R function to compute similarities between node covariates. Default is l1_similarity, that is, -abs(x-y).
clustering()method to cluster network data with missing value
partlyObservedNetwork$clustering(
nbBlocks,
method = c("hierarchical", "spectral", "kmeans")
)nbBlocksinteger, the chosen number of blocks
methodcharacter with a clustering method among "hierarchical", "spectral", "kmeans".
imputation()basic imputation from existing clustering
partlyObservedNetwork$imputation(clustering)
clusteringa vector with size ncol(adjacencyMatrix), providing a user-defined clustering with nbBlocks levels.
an adjacency matrix with imputed values
clone()The objects of this class are cloneable with this method.
partlyObservedNetwork$clone(deep = FALSE)
deepWhether to make a deep clone.
This class is not exported to the user