AM_salso: Sequentially Allocated Latent Structure Optimisation

In AntMAN: Anthology of Mixture Analysis Tools

Description

Heuristic partitioning to minimise the expected loss function with respect to a given expected adjacency matrix. This function is built upon R-package salso's implementation of the salso function. See salso \insertCitesalsoAntMAN for more details.

Usage

AM_salso(
  eam,
  loss,
  maxNClusters = 0,
  nRuns = 16,
  maxZealousAttempts = 10,
  probSequentialAllocation = 0.5,
  nCores = 0
)

Arguments

eam	a co-clustering/ clustering matrix. See salso for more information on which matrix to use.
loss	the recommended loss functions to be used are the "binder" or "VI". However, other loss functions that are supported can be found in the R-package salso's salso function.
maxNClusters	Maximum number of clusters to be considered. The actual number of clusters searched may be lower. Default is 0.
nRuns	Number of runs to try.
maxZealousAttempts	Maximum number of tries for zealous updates. See salso for more information.
probSequentialAllocation	The probability of using sequential allocation instead of random sampling via sample(1:K,ncol(x),TRUE), where K is maxNClusters. Default is 0.5. See salso for more information. argument.
nCores	Number of CPU cores to engage. Default is 0.

Value

A numeric vector describing the estimated partition. The integer values represent the cluster labels of each item respectively.

Source

David B. Dahl and Devin J. Johnson and Peter Müller (2021). salso: Search Algorithms and Loss Functions for Bayesian Clustering. R package version 0.2.15.

AntMAN documentation built on July 23, 2021, 5:08 p.m.