ggRandomForests (version 2.2.0)

gg_roc.rfsrc: ROC (Receiver operator curve) data from a classification random forest.

Description

The sensitivity and specificity of a randomForest classification object.

Usage

# S3 method for rfsrc
gg_roc(object, which.outcome, oob, ...)

Value

gg_roc

data.frame for plotting ROC curves.

Arguments

object

an rfsrc classification object

which.outcome

select the classification outcome of interest.

oob

use oob estimates (default TRUE)

...

extra arguments (not used)

See Also

plot.gg_roc rfsrc randomForest

Examples

## ------------------------------------------------------------
## classification example
## ------------------------------------------------------------
## -------- iris data
#rfsrc_iris <- rfsrc(Species ~ ., data = iris)
data(rfsrc_iris, package="ggRandomForests")

# ROC for setosa
gg_dta <- gg_roc(rfsrc_iris, which.outcome=1)
plot(gg_dta)

# ROC for versicolor
gg_dta <- gg_roc(rfsrc_iris, which.outcome=2)
plot(gg_dta)

# ROC for virginica
gg_dta <- gg_roc(rfsrc_iris, which.outcome=3)
plot(gg_dta)


rf_iris <- randomForest::randomForest(Species ~ ., data = iris)

# ROC for setosa
gg_dta <- gg_roc(rf_iris, which.outcome=1)
plot(gg_dta)

# ROC for versicolor
gg_dta <- gg_roc(rf_iris, which.outcome=2)
plot(gg_dta)

# ROC for virginica
gg_dta <- gg_roc(rf_iris, which.outcome=3)
plot(gg_dta)