Run interval dimensionality reduction methods

Applies multiple interval DR methods (C-PCA, V-PCA, MR-PCA, SPCA, IMDS, Int-UMAP) to an interval_data object and returns the projections.

Usage

run_idr(
  x,
  methods = c("C-PCA", "V-PCA", "MR-PCA", "SPCA", "IMDS", "Int-UMAP"),
  labels = NULL,
  no_dim = 2,
  umap_config = umap_config_default(),
  verbose = TRUE
)

Arguments

x: An interval_data object (should be standardized).
methods: Character vector of methods to run. Default includes all six: "C-PCA", "V-PCA", "MR-PCA", "SPCA", "IMDS", "Int-UMAP".
labels: Optional factor of class labels. If NULL, uses x$labels.
no_dim: Integer number of target dimensions (default 2).
umap_config: List of UMAP hyperparameters. Default uses umap_config_default().
verbose: Logical; print progress messages (default TRUE).

Value

An object of class idr_projections: a named list where each element contains C (centers matrix), R (radii matrix), and type ("Point" or "Interval").

Examples

if (FALSE) { # \dontrun{
data(cars_mm)
x <- standardize(cars_mm)
proj <- run_idr(x)
} # }