KmeansParam-class: K-means clustering

In LTLA/bluster: Clustering Algorithms for Bioconductor

K-means clustering

Description

Run the base kmeans function with the specified number of centers within clusterRows.

Usage

KmeansParam(centers, iter.max = NULL, nstart = NULL, algorithm = NULL)

## S4 method for signature 'ANY,KmeansParam'
clusterRows(x, BLUSPARAM, full = FALSE)

Arguments

centers	An integer scalar specifying the number of centers. Alternatively, a function that takes the number of observations and returns the number of centers.
iter.max, nstart, algorithm	Further arguments to pass to kmeans. Set to the kmeans defaults if not supplied.
x	A numeric matrix-like object where rows represent observations and columns represent variables.
BLUSPARAM	A KmeansParam object.
full	Logical scalar indicating whether the full k-means statistics should be returned.

Details

This class usually requires the user to specify the number of clusters beforehand. However, we can also allow the number of clusters to vary as a function of the number of observations. The latter is occasionally useful, e.g., to allow the clustering to automatically become more granular for large datasets.

To modify an existing KmeansParam object x, users can simply call x[[i]] or x[[i]] <- value where i is any argument used in the constructor.

Value

The KmeansParam constructor will return a KmeansParam object with the specified parameters.

The clusterRows method will return a factor of length equal to nrow(x) containing the cluster assignments. If full=TRUE, a list is returned with clusters (the factor, as above) and objects (a list containing kmeans, the direct output of kmeans).

Author(s)

Aaron Lun

See Also

kmeans, which actually does all the heavy lifting.

MbkmeansParam, for a faster but more approximate version of the k-means algorithm.

Examples

clusterRows(iris[,1:4], KmeansParam(centers=4))
clusterRows(iris[,1:4], KmeansParam(centers=4, algorithm="Lloyd"))
clusterRows(iris[,1:4], KmeansParam(centers=sqrt))

LTLA/bluster documentation built on Aug. 20, 2023, 5:39 a.m.