0

Suppose we have 3 models M1, M2, M3. We want to compare the performance of these models using 10 fold, 10 repeated cross-validation on a train set.

What is the difference between:

CV_Folds <- createMultiFolds(y, k = 10, times = 10)
fitControl <- trainControl(method='repeatedCV',index=CV_Folds))

and

fitControl <- trainControl(method= "repeatedcv", number=10, repeats=10)
mehdi lotfi
  • 11,194
  • 18
  • 82
  • 128
user3810133
  • 31
  • 2
  • Do they return different results? Can you provide a [reproducible example](http://stackoverflow.com/questions/5963269/how-to-make-a-great-r-reproducible-example)? – MrFlick Jul 23 '14 at 17:49
  • I've been messing with that myself and I see no difference. Specifying `index = CV_folds` simply streamlines things a bit better. – Black Milk Aug 09 '14 at 08:37

1 Answers1

0

The method repeatedCV is not a valid option. A method of that name does not appear in the caret source code as of 6.0-34.

mrbcuda
  • 580
  • 7
  • 16