各位专家，我想用caret包实现支持向量回归+RFE，用的是包内自带的iris数据，代码如下：
library(caret)

x<-iris[,-1]
y<-iris[,1]
subsets <- c(1:5, 10, 15, 20, 25,30,35,40,45,50,55)
set.seed(24)
fiveStats <- function(...) c(twoClassSummary(...), defaultSummary(...))
caretFuncs$summary <- fiveStats
fitControl <- trainControl(method = "cv", repeats = 5, classProbs = F, summaryFunction = fiveStats, search = "grid",savePredictions = T)
tunegrid <- expand.grid(.sigma = 2^(-10:4), .C=2^(-4:10))
rfe.full <- rfeControl(functions = caretFuncs,   method = "cv", repeats = 5,  verbose = FALSE)
svmProfile <- rfe(x,y,  sizes = subsets, metric = "RMSE",  trControl = fitControl,  rfeControl = rfe.full)


这段代码运行结果报错：Error in `$<-.data.frame`(`*tmp*`, "y", value = numeric(0)) : replacement has 0 rows, data has 10


请教专家，应该怎样解决呢，谢谢！