BKPC (version 1.0.1)

plot.marginalRelevance: Plot marginalRelevance Objects

Description

Plots marginal relevance scores for features of a given data matrix. The default plot shows: the marginal relevance score (MR score) of each feature. The "pairs" and "parallelcoord" show scatterplot matrix and the parallel coordinates plot of features ordered by their MR score.

Usage

# S3 method for marginalRelevance
plot(x, newdata = NULL, n.feat = NULL, type = "default", ...)

Arguments

x

a marginalRelevance object.

newdata

a matrix containing the new input data.

n.feat

the number of features with highest MR score to plot. Default is all features.

type

"parallelcoord", "pairs" or default.

…

options directly passed to the plot function.

Details

If newdata is omitted the predictions are based on the data used for deriving the MR score.

See Also

marginalRelevance

Examples

# NOT RUN {
data(microarray)

profiles <- as.matrix(microarray[, -2309])
tumourType <-  microarray[, 2309]
 
 margRelv <- marginalRelevance(profiles, tumourType)


# plot 30 gene profiles with highest marginal relevance score
 plot(margRelv, type = "parallelcoord", n.feat = 50, col = tumourType )


# }
# NOT RUN {
library(kernlab)
data(spam)

test <- sample(1:4601,2000)


dt <- as.matrix(spam[-test ,-58])
labels <- spam[-test , 58]

 
margRelv <- marginalRelevance(dt, labels)

#plot MR scores
plot(margRelv)

plot(margRelv , col = labels, type = "pairs", n.feat = 5)
plot(margRelv , col = labels, type = "parallelcoord", n.feat = 30)


# test set
plot(margRelv , as.matrix(spam[test ,-58]), col = spam[test , 58], 
type = "pairs", n.feat = 5)

plot(margRelv , as.matrix(spam[test ,-58]), col = spam[test , 58], 
type = "parallelcoord", n.feat = 30)



 

 
# }