Why my model is not considered MIQCP?
Ongoing
My problem is classified as MIP by gurobi, but this is its summary:
Optimize a model with 2408 rows, 2792 columns and 5191 nonzeros
Model fingerprint: 0x45d79039
Model has 210 quadratic objective terms
Model has 3006 quadratic constraints
Variable types: 624 continuous, 2168 integer (1459 binary)
is MIQCP: 0
is MIQP: 0
is MIP: 1
Why is it not solved as a MIQCP if it has quadractive objective terms and constraints?
-
Hi Lucas,
If your model is nonconvex, i.e., you have to set the NonConvex parameter to 2 then, Gurobi classifies the model as a MIP. If you have quadratic constraints, then it should also be classified as an MIQCP.
Could you share your model and a code snippet you are using to generate the output? Note that uploading files in the Community Forum is not possible but we discuss an alternative in Posting to the Community Forum.
Best regards,
Jaromił0 -
It is convex, I'm solving without setting NonConvex parameter.
I'm posting my model in mathematical notation, I guess its simpler than posting the hole code due to the indices.
Indices make no difference for this matter. Also I have painted out the fixed parameters. There is only one quadratic term, which is in the objective function. I highlighted it in red, as well as the main sequence of constraints leading this quadratic term to the constraint that uses a binary variable, b.
Why it says that has quadratic constraints when, excplicitly, there is any? And why with a explicit quadratic term in the objective function; and constraints seen as quadratic by gurobi, it doesnt classifies the model as MIQCP or, at least, MIQP?
0
Please sign in to leave a comment.
Comments
2 comments