0
$\begingroup$

I'm interested in good references on cross-validation implementations for feed-forward neural networks in pytorch from scratch.

Thanks in advance.

$\endgroup$

1 Answer 1

1
$\begingroup$

All the same considerations for cross validation apply for neural networks as for any other type of model. I.e. the usual scikit-learn (or other options for special situations like grouped+stratified CV) approaches would be used.

A common mistake with CV for neural networks is to do data augmentation before creating CV (and/or test) splits.

$\endgroup$
4
  • $\begingroup$ Thanks for the answer. Could you point out a concrete reference? $\endgroup$ May 21 at 1:54
  • 1
    $\begingroup$ General stuff onCV is covered e.g. in chapter 5 of ISLR. There's also a nice chapter in Approaching (Almost) Any Machine Learning Problem. There's solution write-ups for Kaggle competitions involving neural networks (anything with images, audio, video, text etc. will usually involve NNs nowadays), e.g. here. $\endgroup$
    – Björn
    May 21 at 7:44
  • 1
    $\begingroup$ This blog post nicely describes the dangers of random cross validation and all the considerations there should be used when constructing your test set (and all the same considerations should really be applied when constructing a CV scheme, otherwise CV is not helping). $\endgroup$
    – Björn
    May 21 at 7:45
  • $\begingroup$ Thanks a lot for the references! $\endgroup$ May 21 at 12:25

Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.