dissimMethod: Sample caret methods to maximize dissimilarity in the...

In jknowles/EWStools: Tools for automating the testing and evaluation of education early warning system models

Description

Create a subset of caret methods in an effort to maximize dissimilarity between them along the features identified in the caretTags dataset

Usage

dissimMethod(methods, k, distancemethod, n, what = c("tags", "matrix"))

Arguments

methods	a vector of character representing method names in caret
k	the number of clusters
distancemethod	the method used to calculate the dissimilarity
n	number methods to sample per cluster
what	what to return, either "tags" for the caretTags data or "matrix" for the dissimilarity matrix

Value

Returns either a dataframe or a matrix with the dissimilarity values for the methods sampled and the names of the methods.

Examples

require(EWStools)
data(caretTags)
dissimMethod(sample(caretTags$method, 30), k = 4, distancemethod = "maximum", n = 2, what = "tags")

jknowles/EWStools documentation built on May 19, 2019, 11:42 a.m.