Error in Setting Objective

回答済み

Prachi Agrawal

2022年12月14日 10:12

• 
• 
• 

Dear Sir, 

I am trying to run this code in which I am using an MPS data file. I am extracting the coefficient matrix, the right-hand side vector, and the cost vector for an optimization problem from the MPS file. It shows some errors in setting the objective function. I do not understand the errors in the objective function and the constraints. Please help me in this regard. I will be incredibly grateful to you. The error is as follows:

from gurobipy import *
from numpy import genfromtxt
import numpy as np
import scipy as sp
import pandas as pd
from sklearn.preprocessing import normalize

linear = Model("LP")
m = read("mpsfile.mps")
B = m.getA()
B1 = np.array(B)

r = m.getAttr("rhs", m.getConstrs())
b = np.array(r)

obj = m.getAttr("Obj", m.getVars())
c = np.array(obj)

N = len(c)

sen = m.getAttr("Sense", m.getConstrs())
optr = np.array(sen)

x = m.addMVar(N, vtype=GRB.CONTINUOUS, lb=0, name='x')



obj_linear = (c @ x)

linear.setObjective(obj_linear, sense=GRB.MINIMIZE)

linear.addMConstr(B1, x, optr, b)
m.optimize()
m.update()

関連

• Matrix API

0

• 

  Jaromił Najman

  • Gurobi Staff

  2022年12月14日 14:14

  Hi Prachi,

  The issue you encounter is because you are adding the variables to Model \(\texttt{m}\) instead of Model \(\texttt{linear}\). The variable addition step should read

  x = linear.addMVar(N, vtype=GRB.CONTINUOUS, lb=0, name='x')

  Moreover, the matrix returned by the getA method is already of type \(\texttt{scipy.sparse}\) in CSR format. Thus, you don't need the line where you convert it to an \(\texttt{np.array}\)

  linear = Model("LP")
  m = read("myLP.lp")
  B = m.getA()
  [...]
  linear.addMConstr(B, x, optr, b)

  Best regards, 
  Jaromił

  0

サインインしてコメントを残してください。