Slow constraints addition in Python vs. C++

回答済み

Timothy Baeckelant

• 2024年04月16日 14:37

Hello,

I'm getting troubles with my model. Here is my code : 

T = model.addMVar((numnonzero,), vtype=GRB.BINARY)

for s1 in range(numnonzero):
    for s2 in range(s1+1,numnonzero):
        row1 = nonzero[s1][0]
        col1 = nonzero[s1][1]
        row2 = nonzero[s2][0]
        col2 = nonzero[s2][1]
        if (row1 == row2 or col1 == col2 or (X[row1][col2] > 0.001 and X[row2][col1] > 0.001)):
            model.addConstr(T[s1]+T[s2]<=1)

 

where nonzero is the indice set of the non-zero values of X and numnonzero is the length of this set. For some 20-by-20 matrices, it takes more than 20 seconds only to build this model (even if the optimization itself is quite faster). For example, I have the same model in C++ and the time to construct the model is almost 0, here is the code :

 

int indices[2]
double coefficients[2] = {1.0,1.0};
for (std::size_t s1 = 0; s1 < numnonzero; ++s1):
{
    for (std::size_t s2 = s1 + 1; s2 < numnonzero; ++s2)
    {
         std::size_t row1 = nonzeros[s1].row;
         std::size_t col1 = nonzeros[s1].col;
         std::size_t row2 = nonzeros[s2].row;
         std::size_t col2 = nonzeros[s2].col;
         if (row1 == row2 || col1 == col2 || (isNonzero(row1,col2) && isNonzero(row2,col1)))
         {
             indices[0]=s1;
             indices[1]=s2;
             GRBaddconstr(model,2,&indices[0],&coefficients[0], '<', 1.0, nullptr);
         }
    }
}

 

So my question is: why does it take so long in Python while I almost can get it instantly in my C++ model, and how to manage this ?

 

Have a good day,

Timothy

関連するエンティティ

• Model Building

0

• 

  Marika Karbstein

  Gurobi Staff

  • 2024年04月17日 11:50

  Hi Timothy,

  It looks like you are mixing matrix-friendly and term-based modeling in Python.
  Did you try to use Model.addVars() instead of Model.addMVar() ?

  Best regards,
  Marika

  0
• 

  Timothy Baeckelant

  • 2024年04月17日 12:47

  

  

  Hello Marika, thanks for this answer!

  I first wrote a "matrix-based modeling" but the time to build the model was even worse, so I tried like this. However, just using Model.addVars() makes the generation 50 times faster, thank you!

  0

サインインしてコメントを残してください。