mlr (version 2.0)

getConfMatrix: Confusion matrix.

Description

Calculates confusion matrix for (possibly resampled) prediction. Rows indicate true classes, columns predicted classes.

Usage

getConfMatrix(pred, relative = FALSE)

Value

[matrix]. A confusion matrix.

Details

Code inspired by errormatrix.

See Also

predict.WrappedModel

Examples

## create classification task and use linear discriminant analysis for classification
task = makeClassifTask(data = iris, target = "Species")
lrn = makeLearner("classif.lda")

## set up training and test data
n = nrow(iris)
mixed.set = sample(1:n)
training.set = mixed.set[1:(n/2)]
test.set = mixed.set[(n/2 + 1):n]

## train model
mod = train(lrn, task, subset = training.set)

## get predictions and show calculate confusion matrix
pred = predict(mod, newdata = iris[test.set, ])
print(getConfMatrix(pred))
print(getConfMatrix(pred, relative = TRUE))