I'm doing nested cross-validation on a classification problem (inner loop to tune hyperparameters, outer loop to select algorithm).

I've often heard that a reasonable number of folds for standard (I.e.: not nested) cross-validation is 5 or 10.

Is there a similar rule of thumb for nested cross-validation? Should I have more folds in the inner or the outer? How many?

I'm not looking for a definitive answer, I'm just curious to hear what people usually go for. Thanks.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.