MinimiseEPL {GreedyEPL}R Documentation

MinimiseEPL

Description

Performs greedy optimisation to find the Bayes optimal decision given a collection (sample) of partitions.

Usage

MinimiseEPL(sample_of_partitions, pars = list())

Arguments

sample_of_partitions

A TxN matrix whose rows correspond to partitions of the same N data observations.

pars

A list of additional parameters. See 'Details'.

Details

See Also

CollapseLabels.

Examples

set.seed(123)
N <- 10
niter <- 50
Kup <- 3
sample_of_partitions <- matrix(sample(x = 1:Kup, size = niter*N, replace = TRUE),niter,N)
loss_type = "VI"
output <- MinimiseEPL(sample_of_partitions, list(Kup = Kup, loss_type = loss_type))
output$decision
output$EPL

[Package GreedyEPL version 1.2 Index]