cluster_k_analysis: Perform the types of anaylsis which do a reduction on the...

In MalachiTimothyPhillips/ggfast: Gotta Go Fast (ggfast)

Description

NOTE: by default, this will perform the analysis on the scaled data

Usage

cluster_k_analysis(x, kmax = 10, elbowPlotFileName = "elbowPlot.jpeg",
  silhouetteFileName = "silhouette.jpeg", gapFileName = "gap.jpeg",
  iter.max = 100, nstart = 1, algorithm = "Lloyd", trace = FALSE,
  debugOutput = FALSE, debugOutFileNameBase = "sil")

Arguments

x	Data in the form of a data frame or data matrix, for safety, please use data.matrix(...) as a wrapper
kmax	(=10) Maximum number of clusters to form
elbowPlotFileName	(="elbowPlot.jpeg"), name of file for elbow plot output
silhouetteFileName	(="silhouette.jpeg"), name of file for silhouette plot output
gapFileName	(="gap.jpeg"), name of file for output of gap statistics plot
iter.max	(=100) Maximum number of iterations
nstart	How many random sets to be chosen
algorithm	What algorithm to use (default="Lloyd"). choices include: "Hartigan-Wong", "LLoyd", "MacQueen", and "Forgy"
trace	logical or integer number, currently only used in the default method ("Hartigan-Wong"): if positive (or true), tracing information on the progress of the algorithm is produced. Higher values may produce more tracing information.
debugOutput	(=FALSE) Boolean value for whether or not to output silhouette plots for each and every cluster value used. This generates a TON of output! YOU HAVE BEEN WARNED!
debugOutFileNameBase	(="sil") Base file name for debug output, if enabled.

Examples

cluster_k_analysis(data.matrix(iris), kmax=19)
cluster_k_analysis(data.matrix(iris), kmax=19, iter.max=200, debugOutput=T)

MalachiTimothyPhillips/ggfast documentation built on May 18, 2019, 11:27 p.m.