AM_salso {AntMAN} | R Documentation |
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 (Dahl et al. 2021) for more details.
AM_salso( eam, loss, maxNClusters = 0, nRuns = 16, maxZealousAttempts = 10, probSequentialAllocation = 0.5, nCores = 0 )
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. |
A numeric vector describing the estimated partition. The integer values represent the cluster labels of each item respectively.
David B. Dahl and Devin J. Johnson and Peter MÃ¼ller (2021). salso: Search Algorithms and Loss Functions for Bayesian Clustering. R package version 0.2.15.