clusterer: Function to Cluster Non-Conformity Scores Using Either...

In pintervals: Model Agnostic Prediction Intervals

Function to Cluster Non-Conformity Scores Using Either Kolmogorov-Smirnov or K-Means Clustering

Description

Function to Cluster Non-Conformity Scores Using Either Kolmogorov-Smirnov or K-Means Clustering

Usage

clusterer(
  ncs,
  m,
  class_vec,
  maxit = 100,
  method = c("ks", "kmeans"),
  q = seq(0.1, 0.9, by = 0.1),
  min_class_size = 10
)

Arguments

ncs	Vector of non-conformity scores
m	Number of clusters to form
class_vec	Vector of class labels
maxit	Maximum number of iterations for the clustering algorithm
method	Clustering method to use, either 'ks' for Kolmogorov-Smirnov or 'kmeans' for K-means clustering
q	Quantiles to use for K-means clustering, default is a sequence from 0.1 to 0.9 in steps of 0.1
min_class_size	Minimum number of observations required in a class to be included in clustering

Value

A vector of cluster assignments, with attributes containing the clusters, coverage gaps, method used, number of clusters, and Calibrated Clustering index

pintervals documentation built on March 3, 2026, 5:06 p.m.