uses rpart instead of randomForest for faster example

R/FeatureEffect.R

@@ -56,8 +56,8 @@
 #' @examples
 #' # We train a random forest on the Boston dataset:
 #' data("Boston", package = "MASS")
-#' library("randomForest")
-#' rf <- randomForest(medv ~ ., data = Boston, ntree = 50)
+#' library("rpart")
+#' rf <- rpart(medv ~ ., data = Boston)
 #' mod <- Predictor$new(rf, data = Boston)
 #'
 #' # Compute the accumulated local effects for the first feature
@@ -109,7 +109,7 @@
 #' plot(eff)
 #'
 #' # FeatureEffect plots also works with multiclass classification
-#' rf <- randomForest(Species ~ ., data = iris, ntree = 50)
+#' rf <- rpart(Species ~ ., data = iris)
 #' mod <- Predictor$new(rf, data = iris, type = "prob")
 #'
 #' # For some models we have to specify additional arguments for the predict