I want to develop a neural network for my data, but I have no idea if my problem should be solved with curve fitting, classification or maybe something else.

Here is my problem.

My input data are some points like [1.32 1.42 0.21 0.42 0.7 .91 1.23 1.3], which I want them to turn into [1 0 0 0 0 1 0 0]. I can guess what kind of transformation functions should I use, but, when it comes to developing the network, I've got no idea which approach should I follow.

  • 1
    $\begingroup$ Hi, welcome to AI.SE ! You should give more info if you want to get help. Where does your data come from ? Have you preprocessed-it ? How do you obtain the labels ? What are the meaning of the labels ? $\endgroup$ – 16Aghnar Aug 20 at 15:09
  • $\begingroup$ Hey there, I have 72 situations on any sample (20 samples); all situations are scored with a number from 0 to 1 which is my input for any sample. now i want to choose 5 best situations from them. I thought maybe giving 1 for the optimal situation and zero for the bad situation would be good enough; so i for target i have a vector with 72 zero elements that 5 of elements are replaced with 1. $\endgroup$ – Baldwin Aug 20 at 15:43
  • $\begingroup$ You have only 20 samples in your train set ? Do you have already labelled samples (i.e. you know the 5 best situations ) ? Also, you say the score are number from 0 to 1, but in your example above you wrote numbers higher than 1... is it correct ? $\endgroup$ – 16Aghnar Aug 21 at 8:23

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.