# library("mboost")

if (FALSE) {

cvrisk2 <- function(object, folds = cv(object@weights), 

    mincriterion = seq(from = 0.6, to = 0.95, by = 0.05),

    papply = if (require("multicore")) mclapply else lapply,

    ...){

    weights <- object@weights

    if (any(weights == 0))

        warning("zero weights")

    if (is.null(folds)) {

        folds <- rmultinom(25, length(weights), weights/sum(weights))

    } else {

        stopifnot(is.matrix(folds) && nrow(folds) == length(weights))

    }

    fitfct <- object@update

    oobrisk <- matrix(0, nrow = ncol(folds), ncol = length(grid))

    dummyfct <- function(weights) {

        mod <- fitfct(weights = weights)

        sapply(mincriterion, function(m) {

            p <- predict(mod, mincriterion = m)

            if (is.factor(p)) 

                err <- p != mod@responses@variables[[1]]

            if (is.numeric(p))

                err <- (p - mod@responses@variables[[1]])^2

            mean(err[weights == 0])

        })

    }

    OOBweights <- matrix(rep(weights, ncol(folds)), ncol = ncol(folds))

    OOBweights[folds > 0] <- 0

    oobrisk <- papply(1:ncol(folds),

        function(i) dummyfct(weights = folds[, i]))

    oobrisk <- t(as.data.frame(oobrisk))

    oobrisk <- oobrisk / colSums(OOBweights)

    colnames(oobrisk) <- mincriterion

    rownames(oobrisk) <- 1:nrow(oobrisk)

    oobrisk

}

set.seed(290875)

### regression

airq <- subset(airquality, !is.na(Ozone))

airct <- ctree(Ozone ~ ., data = airq, 

               controls = ctree_control(maxsurrogate = 3, mincriterion = 0.5))

### bootstrap

cvm <- cvrisk2(airct, mincriterion = 50:95/100)

boxplot(cvm)

### 10-fold CV

cvm <- cvrisk2(airct, mincriterion = 50:95/100, cv(airct@weights, type = "kfold"))

boxplot(cvm)

}