[R] different validation methods produce same results

javed khan j@vedbtk111 @end|ng |rom gm@||@com
Fri Apr 10 00:49:44 CEST 2020

I am using different validation methods (mentioned below in code) for
software defect prediction dataset. I am using them for Random and Grid
searches and all the validation methods produce same results i.e. random
search with all validation methods produce same results and same is the
case with grid search.

Am I doing something wrong?

ct_rand <- trainControl(method = "repeatedcv", number=10, repeats=10,index
= index_2, search="random")
ct_grid <- trainControl(method = "repeatedcv", number=10, repeats=10,index
= index_2, search="grid")

ct_boot1 <- trainControl(method = "boot", number=100,  index = index_2,
ct_boot2 <- trainControl(method = "boot", number=100,  index = index_2,

ct_locv <- trainControl(method = "LOOCV",  search="random")
ct_locv2 <- trainControl(method = "LOOCV",   search="grid")

ct_rand5CV <- trainControl(method = "repeatedcv", number=5,
repeats=10,index = index_2, search="random")
ct_grid5CV <- trainControl(method = "repeatedcv", number=5,
repeats=10,index = index_2, search="grid")

	[[alternative HTML version deleted]]

More information about the R-help mailing list