Measures silhouette between clusters

Usage

silhouette(
  object,
  new_data = NULL,
  dists = NULL,
  dist_fun = philentropy::distance
)

Arguments

object: A fitted tidyclust model
new_data: A dataset to predict on. If NULL, uses trained clustering.
dists: A distance matrix. Used if new_data is NULL.
dist_fun: A function of the form function(x) that takes a data frame or matrix and returns a dist object. Defaults to philentropy::distance with Euclidean distance. See philentropy::getDistMethods() for a list of supported methods, and vignette("tuning_and_metrics", package = "tidyclust") for usage examples.

Value

A tibble giving the silhouette for each observation.

Details

silhouette_avg() is the corresponding cluster metric function that returns the average of the values given by silhouette().

Examples

kmeans_spec <- k_means(num_clusters = 5) |>
  set_engine("stats")

kmeans_fit <- fit(kmeans_spec, ~., mtcars)

dists <- mtcars |>
  as.matrix() |>
  dist()

silhouette(kmeans_fit, dists = dists)
#> # A tibble: 32 × 3
#>    cluster   neighbor  sil_width
#>    <fct>     <fct>         <dbl>
#>  1 Cluster_1 Cluster_2     0.572
#>  2 Cluster_1 Cluster_2     0.572
#>  3 Cluster_1 Cluster_2     0.752
#>  4 Cluster_2 Cluster_1     0.540
#>  5 Cluster_3 Cluster_4     0.149
#>  6 Cluster_2 Cluster_1     0.224
#>  7 Cluster_3 Cluster_4     0.649
#>  8 Cluster_1 Cluster_2     0.613
#>  9 Cluster_1 Cluster_2     0.692
#> 10 Cluster_1 Cluster_2     0.460
#> # ℹ 22 more rows