plotROCseries: Plot a series of ROCs

Description Usage Arguments Details Value See Also

View source: R/rocplot.R

Description

Plot a series of ROCs

Usage

1
2
plotROCseries(prs, legend = TRUE, legend.names = NULL,
  add.stats = TRUE, col = NULL, lwd = 3, lty = NULL, ...)

Arguments

prs

a list returned by prconvert

legend

whether a legend should be plotted on the bottom right of the figure

legend.names

a vector of names to plot of the legend

add.stats

whether AUC and CI should be added to the legend

col

vector of colors of the length equal to length of prs

lwd

line width

lty

line type

...

any further arguments are passed to the rocplot function

Details

Plot a series of receiver-operator curves (ROCs). This function is a wrapper around rocplot(), allowing to plot multiple ROCs.

ROCs to be drawn are provided in the prs parameter, a list, for which each element is a data frame containing at least the following columns: reality, decision and prediction. The column reality describes the actual (real) classification; "decision" is the class assigned by the ML algorithm, and "prediction" is a numeric value describing the probability of a given sample to belong to the case class.

The graphic parameters col, lwd and lty can be either a single value, which will be applied to all lines drawn, or a vector of the same length as the number of ROCs to be drawn.

Value

A list of objects returned by the rocplot function.

See Also

rocplot, rocformat, prconvert


january3/myfuncs documentation built on April 1, 2020, 4:42 a.m.