I am given a dataset where the class labels are continuous values between [-1,1]. Based on this, I have few questions:

  1. Can I use Back-Propagation Neural Network (BPNN) for this problem? Previously, I had a different dataset where the labels are binary classes but for this dataset I am not sure since it is now a regression problem.
  2. In case neural network can work with this dataset, what activation function should I use? Sigmoid, hyperbolic tan function (tanh), or rectified linear unit (relu)?

Thank you.

  • $\begingroup$ What do you mean class labels are in this range? Are they supposed to be independent of each other? $\endgroup$ – Jakub Bartczuk Oct 26 '17 at 21:31

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.