Minimum pairwise distance within a selection (T_k = k-centre objective)

Returns the minimum pairwise distance among selected points, the canonical k-centre objective \(T_k\). Higher values indicate a more spread-out (better) selection.

Usage

TkScore(d = NULL, idx, points = NULL)

Arguments

d

Pairwise distance matrix or dist object. Ignored when points is supplied.

idx

Integer vector of selected row/col indices.

points

Optional N x dim numeric coordinate matrix. When supplied, the score is computed from stats::dist() on the selected sub-coordinates only (k x k), never the full N x N matrix (d is then unused). For Euclidean data the result is identical to the matrix path.

Value

Numeric scalar; NA_real_ if length(idx) < 2.

Examples

set.seed(1)
pts <- matrix(rnorm(60), ncol = 2)
d <- dist(pts)
TkScore(d, Gonzalez(d, 5L))
#> [1] 1.765223