For CART model, caret seems to only provide tunning of complexity parameter. Is there way to tune other parameters such as minbucket?
可以将文章内容翻译成中文,广告屏蔽插件可能会导致该功能失效(如失效,请关闭广告屏蔽插件后再试):
问题:
回答1:
Arguments passed to the classification or regression routines are included in the dots parameter.
As you want to include minbucket
, parameter control
should be included inside train
.
As example:
library("caret")
train(Kyphosis ~ Age + Number + Start, data = kyphosis,
method = "rpart",
tuneGrid = data.frame(cp = c(0.01, 0.05)),
control = rpart.control(minsplit = 10, minbucket = 5))
If you want to tune minbucket
you will have to do it manually.