0

I'm trying to use SCIP through python and I have installed SCIP optimization suite 3.2.1. I have problem framing my optimization question through PYSCIPOPT. As I have 2000+ variables to solve, I am wondering can I use matrix notation to frame the question in python?

Florence
  • 29
  • 6
  • 2
    Welcome to Stackoverflow! Can you please elaborate your question having your effort like code or something so that people could get your problem early and help you? Thanks! – Enamul Hassan Mar 04 '16 at 03:49
  • What exactly do you mean by matrix notation? I suppose a Matlab/Numpy- like notation to state the entire problem like 'A * x <= b'? – Gregor Mar 04 '16 at 08:35
  • yes, something like A*x<=b. Do I use coefficients? I'm also having trouble with quadratic constraints. How do I use the addCons in the python interface to write the constraint x'Qx<=0.03 for example. Where x is n*1 and Q is n*n – Florence Mar 14 '16 at 09:49

1 Answers1

0

No, this is not possible, because SCIP is constraint based and does not rely on a central matrix structure. A problem with 2000 variables is not at all large, by the way and should be processed within a a second.

This is how you would transform a quadratic constraint matrix Q of size 2:

Q = [a b;c d], x = [x1; x2]
x'Qx = ax1^2 + dx2^2 + (b+c)x1x2

This can then be passed to SCIP with the addCons() method.

mattmilten
  • 6,242
  • 3
  • 35
  • 65
  • Matt, I can show you a problem with two variables that takes forever ;) – Gregor Mar 04 '16 at 14:03
  • Thank you for your answer. Then how can I write the objective function of the form min x*QT*x-fTx in the python interface? where x is the n*1 matrix I want to solve and Q is the n*n covariance? – Florence Mar 04 '16 at 23:42
  • 1
    Gregor, I said "processed" as in "read in". I didn't mean that it can be solved that fast. – mattmilten Mar 05 '16 at 09:01
  • As manetsus already pointed out: what have you tried so far and where did you get stuck? – mattmilten Mar 05 '16 at 15:10