dp_means: DP-means Clustering

In bearloga/dpmclust: Bayesian Nonparametric Clustering Using 'DP-means' Algorithm

Description

Perform clustering on a data matrix using a pure R implementation of the DP-means algorithm.

Usage

dp_means(x, lambda, max_iter = 100, tol = 0.001, verbose = TRUE)

Arguments

x	numeric matrix or data frame of features
lambda	threshold distance for creating new clusters
max_iter	maximum number of iterations
tol	tolerance when checking for convergence
verbose	whether to print status information after each iteration

Value

For convenience and ease of use with visualization and tidying packages, the returned object has same class (kmeans) and components as an object returned by stats::kmeans().

Examples

## Not run: 
x <- iris[, 1:4]
dp_means(x, lambda = 2)

data("wine")
x <- wine[, -1]
dp_means(x, lambda = 500)

## End(Not run)

bearloga/dpmclust documentation built on March 7, 2020, 7:11 p.m.