bipartite.match: Bipartite graph matching
In zdebruine/LSMF: Least squares matrix factorization

Description Usage Arguments Details Value Examples

View source: R/bipartite.match.R

Hungarian algorithm for matching samples in a bipartite graph from a distance ("cost") matrix

1	bipartite.match(costMatrix)

costMatrix

A distance matrix giving the cost of each possible pairing

This algorithm was cloned from RcppHungarian, an Rcpp wrapper for the original C implementation by Cong Ma (2016).

List of components "cost" and "parings", with pairings given as an n x 2 matrix

/dontrun{
data(moca7k)
model1 <- lsmf(moca7k[,1000:2000], k = 10)
model2 <- lsmf(moca7k[,2000:300], k = 10)
costMatrix <- 1 + 1e-10 - sparse.cos(model1$W, model2$W)
matched <- bipartite.match(costMatrix)
model2W.reordered <- model2$W[,matched$pairs[,2]]
}