Split observations by k-means clustering groups.

Aligns and groups observations based on k-means clustering, enabling observation splits by cluster groups.

Usage

align_kmeans(
  centers,
  ...,
  data = NULL,
  active = NULL,
  set_context = deprecated(),
  name = deprecated()
)

Arguments

centers

either the number of clusters, say \(k\), or a set of initial (distinct) cluster centres. If a number, a random set of (distinct) rows in x is chosen as the initial centres.

...

Arguments passed on to stats::kmeans

iter.max

the maximum number of iterations allowed.

nstart

if centers is a number, how many random sets should be chosen?

algorithm

character: may be abbreviated. Note that "Lloyd" and "Forgy" are alternative names for one algorithm.

trace

logical or integer number, currently only used in the default method ("Hartigan-Wong"): if positive (or true), tracing information on the progress of the algorithm is produced. Higher values may produce more tracing information.

data

A matrix-like object. By default, it inherits from the layout matrix.

active

A active() object that defines the context settings when added to a layout.

set_context

Please use active argument instead.

name

Please use active argument instead.

Value

A "AlignKmeans" object.

Axis Alignment for Observations

It is important to note that we consider rows as observations, meaning vec_size(data)/NROW(data) must match the number of observations along the axis used for alignment (x-axis for a vertical stack layout, y-axis for a horizontal stack layout).

• quad_layout()/ggheatmap(): For column annotation, the layout matrix will be transposed before use (if data is a function, it is applied to the transposed matrix), as column annotation uses columns as observations but alignment requires rows.

• stack_layout(): The layout matrix is used as is, aligning all plots along a single axis.

Examples

ggheatmap(matrix(rnorm(81), nrow = 9)) +
    anno_top() +
    align_kmeans(3L)
#> → heatmap built with `geom_tile()`