plotlyROC: Use 'plotly' to produce a ROC plot.

In WinVector/WVPlots: Common Plots for Analysis

Use plotly to produce a ROC plot.

Description

Note: any arrange_ warning is a version incompatibility between plotly and dplyr.

Usage

plotlyROC(
  d,
  predCol,
  outcomeCol,
  outcomeTarget,
  title,
  ...,
  estimate_sig = FALSE
)

Arguments

d	dataframe
predCol	name of column with numeric predictions
outcomeCol	name of column with truth
outcomeTarget	value considered true
title	character title for plot
...	no unnamed argument, added to force named binding of later arguments.
estimate_sig	logical, if TRUE estimate and display significance of difference from AUC 0.5.

Value

plotly plot

See Also

ROCPlot

Examples

if(FALSE && requireNamespace("plotly", quietly = TRUE)) {
   if (requireNamespace('data.table', quietly = TRUE)) {
	   # don't multi-thread during CRAN checks
		   data.table::setDTthreads(1)
   }
   set.seed(34903490)
   x = rnorm(50)
   y = 0.5*x^2 + 2*x + rnorm(length(x))
   frm = data.frame(x=x,yC=y>=as.numeric(quantile(y,probs=0.8)))
   plotlyROC(frm, 'x', 'yC', TRUE, 'example plot', estimate_sig = TRUE)
}

WinVector/WVPlots documentation built on April 23, 2024, 4:51 a.m.