plotROC: Plots the ROC curve for a result or model

In liquidSVM: A Fast and Versatile SVM Package

Description

This can be used either using rocSVM or lsSVM

Usage

plotROC(x, correct, posValue = NULL, xlim = 0:1, ylim = 0:1,
  asp = 1, type = NULL, pch = "x", add = FALSE, ...)

Arguments

x	either the result from a test or a model
correct	either the true values or testing data for the model
posValue	the label marking the positive value. If NULL (default) then the larger value.
xlim	sets better defaults for plot.default
ylim	sets better defaults for plot.default
asp	sets better defaults for plot.default
type	sets better defaults for plot.default
pch	sets better defaults for plot.default
add	if 'FALSE' (default) produces a new plot and if 'TRUE' adds to existing plot.
...	gets passed to plot.default

See Also

rocSVM, lsSVM

rocSVM

Examples

## Not run: 
banana <- liquidData('banana-bc')
model <- rocSVM(Y~.,banana$train)

plotROC(model ,banana$test)
# or:
result <- test(model, banana$test)
plotROC(result, banana$test$Y)

model.ls <- lsSVM(Y~., banana$train)
result <- plotROC(model.ls, banana$test)

## End(Not run)

liquidSVM documentation built on Sept. 15, 2019, 1:02 a.m.