1
$\begingroup$

I have seen classification CNNs that train with numerous images for a subset of labels (i.e. Number of images >> Number of labels), however, is it still possible to use CNNs when the number of images = Number of labels?

specifically consider: having N settings that you can control to generate a unique image. Is it possible to make a CNN that can describe the mapping? (Is CNN the right architecture to use?)

$\endgroup$

1 Answer 1

0
$\begingroup$

If you have single training data for each image (Number of image = Number of labels), no architecture is a good architecture. All good image models are deep learning model and are data hungry. If you have just a single image for each level it will not be able to learn anything

$\endgroup$
1
  • $\begingroup$ Is it fair to say then that one should approach it as a regression problem, and in this case maybe we can get okay results with a minimal data set as I mentioned above? The point you make is fair and is noted, but in the case where you cannot generate a lot of data are there any tricks to be played that can help us? $\endgroup$
    – Akash
    Apr 4 at 22:03

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.