0

I am trying to work on building a relationship in Keras, between X and Y where X= (1,30) and Y= (1,10).

I made model, to predict the Y from X. I am working in the biology field, and I want to predict the binary vector of Y through X with accuracy (90%) .

I will explain mode: X is a binary vector of length n (e.g., X(1,:) = [0/1, 0/1, ..., 0/1]). Y is a binary vector of length m, where m < n (e.g., Y(1,:) = [0/1, 0/1, ..., 0/1]).

For each X => Y the data is like that : for example a sample :

X = [1,0,1,1,1,0,1,0,1,0,1,1,0,1] and its Y=[0,1,1,1,1,0,1]

My objective is to develop a machine learning model M that can predict the vector Y from the vector X with an accuracy greater than 90%. I accept only an error of 1 bit ! not more.

My question, what is the best cost function and loss function should I use ?

stevGates
  • 111
  • This sounds like a fairly routine multi-label problem. Do you see it any different? – Dave Jan 15 '24 at 17:19
  • I think it is a regression – stevGates Jan 15 '24 at 17:28
  • If you are sure that accuracy is what you want (it often isn't), then by all means, go ahead and use that. Let's hope that biology obliges your goals by being sufficiently predictable (it often isn't). – Stephan Kolassa Jan 15 '24 at 17:33
  • I want predect Y correctly that s what I am looking for So I don't know is it classification or regression ( I want make it regression ) – stevGates Jan 15 '24 at 17:39
  • Could you please say what you mean when you use the terms classification and regression? To me, this is kind a “would a rose by any other name” situation, but I’d like the communication to be as smooth as possible. – Dave Jan 15 '24 at 19:28

0 Answers0