spuds_datagen: Simulation of Data Sets for Clustering

In DavidHofmeyr/spuds: Spectral Partitioning Using Density Separation

Description

Generates data from a Gaussian mixture, and then applies a perturbation to each point to render non-convex clusters.

Usage

  spuds_datagen(nits, nclust, dim, scale, curve)

Arguments

nits	integer number of data points.
nclust	integer number of clusters.
dim	integer number of dimensions.
scale	(optional) numeric. Larger values result in greater cluster overlap. Default is 1.
curve	(optional) numeric. Larger values result in greater perturbation from Gaussian. Default is number of dimensions.

Value

a named list with fields $x: the matrix of data $c: the vector of class labels

DavidHofmeyr/spuds documentation built on May 20, 2019, 9:40 a.m.