Generate Multiple Clusters

This function generates a dataset with multiple clusters.

Usage

gen_multicluster(
  n = c(200, 300, 500),
  p = 4,
  k = 3,
  loc = matrix(c(0, 0, 0, 0, 5, 9, 0, 0, 3, 4, 10, 7), nrow = 3, byrow = TRUE),
  scale = c(3, 1, 2),
  shape = c("gaussian", "bluntedcorn", "unifcube"),
  rotation = NULL,
  is_bkg = FALSE
)

Arguments

n

A numeric vector (default: c(200, 500, 300)) representing the sample sizes.

p

A numeric value (default: 4) representing the number of dimensions.

k

A numeric value (default: 3) representing the number of clusters.

loc

A numeric matrix (default: matrix(c(0, 0, 0, 0, 5, 9, 0, 0, 3, 4, 10, 7 ), nrow = 3, byrow = TRUE)) representing the locations/centroids of clusters.

scale

A numeric vector (default: c(3, 1, 2)) representing the scaling factors of clusters.

shape

A character vector (default: c("gen_gaussian", "gen_bluntedcorn", "gen_unifcube")) representing the shapes of clusters.

rotation

A numeric list which contains plane and the corresponding angle along that plane for each cluster.

is_bkg

A Boolean value (default: FALSE) representing the background noise should exist or not.

Value

A data containing same/different shaped clusters.

Examples

set.seed(20240412)
rotations_4d <- list(
cluster1 = list(
  list(plane = c(1, 2), angle = 60), # Rotation in the (1, 2) plane
  list(plane = c(3, 4), angle = 90)  # Rotation in the (3, 4) plane
  ),
cluster2 = list(
  list(plane = c(1, 3), angle = 30) # Rotation in the (1, 3) plane
  ),
cluster3 = list(
  list(plane = c(2, 4), angle = 45) # Rotation in the (2, 4) plane
  )
)
clust_data <- gen_multicluster(n = c(200, 300, 500), p = 4, k = 3,
loc = matrix(c(
  0, 0, 0, 0,
  5, 9, 0, 0,
  3, 4, 10, 7
), nrow = 3, byrow = TRUE),
scale = c(3, 1, 2),
shape = c("gaussian", "bluntedcorn", "unifcube"),
rotation = rotations_4d,
is_bkg = FALSE)
#> ✔ Data generation completed successfully! 🎉
#> Error in get(paste0("gen_", shape[i])): object 'gen_bluntedcorn' not found