kmeans_cuda: K-Means Clustering using CUDA

In kmcudaR: 'Yingyang' K-Means and K-NN using NVIDIA CUDA

Description

Performs k-means clustering on a numeric matrix using a NVIDIA GPU via CUDA

Usage

kmeans_cuda(samples, clusters, tolerance = 0.01, init = "k-means++",
  yinyang_t = 0.1, metric = "L2", average_distance = FALSE,
  seed = NULL, device = 0L, verbosity = 0L)

Arguments

samples	A numeric matrix
clusters	the number of clusters
tolerance	if the relative number of reassignments drops below this value the algorithm stops
init	A character vector or numeric matrix, sets the method for centroids initialization. Options include "k-means++", "afk-mc2", "random" or numeric matrix of shape [clusters, number of features]. Default = "kmeans++"
yinyang_t	numeric value defining relative number of cluster groups. Usually 0.1 but 0 disables Yinyang refinement.
metric	Character vector specifying distance metric to use. The default is Euclidean (L2), it can be changed to "cos" for Sphereical K-means with angular distance. NOTE - the samples must be normalized in the latter case.
average_distance	logical indicating whether to calculate the average distance between cluster elements and the corresponding centroids. Useful for finding the best 'K'. Returned as third list element
seed	random generator seed for reproducible results [deprecated]
device	integer defining device to use. 1 = first device, 2 = second device, 3 = first & second devices, 0 = use all devices. Default = 0
verbosity	Integer indicating amount of output to see. 0 = silence, 1 = progress logging, 2 = all output

Value

a list consisting of

centroids	Cluster centroids
assignments	integer vector of sample-cluster associations
average_distance	average distance between cluster elements

kmcudaR documentation built on May 2, 2019, 9:17 a.m.