Cross-validation for Classification and Regression Trees

Cross-validation for Classification and Regression Trees

cv.crtree(
  object,
  K = 5,
  repeats = 1,
  cp,
  pcp = seq(0, 0.01, length.out = 11),
  seed = 1234,
  trace = TRUE,
  fun,
  ...
)

Arguments

object	Object of type "rpart" or "crtree" to use as a starting point for cross validation
K	Number of cross validation passes to use
repeats	Number of times to repeat the K cross-validation steps
cp	Complexity parameter used when building the (e.g., 0.0001)
pcp	Complexity parameter to use for pruning
seed	Random seed to use as the starting point
trace	Print progress
fun	Function to use for model evaluation (e.g., auc for classification or RMSE for regression)
...	Additional arguments to be passed to 'fun'

Value

A data.frame sorted by the mean, sd, min, and max of the performance metric

Details

See https://radiant-rstats.github.io/docs/model/crtree.html for an example in Radiant

See also

crtree to generate an initial model that can be passed to cv.crtree

Rsq to calculate an R-squared measure for a regression

RMSE to calculate the Root Mean Squared Error for a regression

MAE to calculate the Mean Absolute Error for a regression

auc to calculate the area under the ROC curve for classification

profit to calculate profits for classification at a cost/margin threshold

Examples

if (FALSE) {
result <- crtree(dvd, "buy", c("coupon", "purch", "last"))
cv.crtree(result, cp = 0.0001, pcp = seq(0, 0.01, length.out = 11))
cv.crtree(result, cp = 0.0001, pcp = c(0, 0.001, 0.002), fun = profit, cost = 1, margin = 5)
result <- crtree(diamonds, "price", c("carat", "color", "clarity"), type = "regression", cp = 0.001)
cv.crtree(result, cp = 0.001, pcp = seq(0, 0.01, length.out = 11), fun = MAE)
}