0
$\begingroup$

I have a very limited dataset and have used 80% of it to pre-train an autoencoder.

Now, I attached the enconder part to a SVM. In order to train the SVM, is ok to train it using the exactly same (80%) dataset?

$\endgroup$
0
$\begingroup$

I think you should be fine. The auto-encoder only performs a pre-selection of the features, so it has nothing to do with the actual model. Just make sure you never mix data to train and to test/validate a model!

$\endgroup$
  • $\begingroup$ have you find this statement in any book/paper? $\endgroup$ – Helder Jun 6 '18 at 11:58

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.