plotAlgoCorMatrix {aslib}R Documentation

Plots the correlation matrix of the algorithms.

Description

If NAs occur, they are imputed (before aggregation) by base + 0.3 * range. base is the cutoff value for runtimes scenarios with cutoff or the worst performance for all others.

Stochastic replications are aggregated by the mean value.

Usage

plotAlgoCorMatrix(asscenario, measure, order.method = "hclust",
  hclust.method = "ward.D2", cor.method = "spearman")

Arguments

asscenario

[ASScenario]
Algorithm selection scenario.

measure

[character(1)]
Measure to plot. Default is first measure in scenario.

order.method

[character(1)]
Method for ordering the algorithms within the plot. Possible values are “hclust” (for hierarchical clustering order), “FPC” (first principal component order), “AOE” (angular order of eigenvectors), “original” (original order) and “alphabet” (alphabetical order). See corrMatOrder. Default is “hclust”.

hclust.method

[character(1)]
Method for hierarchical clustering. Only useful, when order.method is set to “hclust”, otherwise ignored. Possible values are: “ward.D2”, “single”, “complete”, “average”, “mcquitty”, “median” and “centroid”. See corrMatOrder. Default is “ward.D2”.

cor.method

[character(1)]
Method to be used for calculating the correlation between the algorithms. Possible values are “pearson”, “kendall” and “spearman”. See cor. Default is “spearman”.

Value

See corrplot.


[Package aslib version 0.1.1 Index]