Fitting of custom non-linear function

MarkusBJMarkusBJ Member Posts: 1 Contributor I
edited November 2018 in Help
I have some data (~100 examples) : 3 attributes (att1, att2, att3) and the label (lab1)

Theory says that
lab1 = p1*att1^2 + p2*(att2^1/2) + p3*exp^(p4*att3) + p5,
where p1..p5 - some parameters

Is it possible to find these parameters by using Rapidminer (non-linear least squares or something)?
Which operators should I use?


  • Options
    earmijoearmijo Member Posts: 270 Unicorn
    Had it not been for p4, I think you could have transform the attributes and run a simple linear regression. I don't know how to do it in RM.

    You might want to check R or Formulize (http://www.nutonian.com/eureqa-ii/)
Sign In or Register to comment.