AttributeError: 'generator' object has no attribute 'GetCoeffs'

[mlgolden07]

I am getting this error:

<-- My error -->

Traceback (most recent call last):

  File "my_projects/MyKnapsack.py", line 40, in <module>

    main()

  File "my_projects/MyKnapsac.py", line 27, in main

    solver.Maximize(solver.Sum(values[i] * x[i]) for i in range(num_items))

  File "/Library/Python/2.7/site-packages/ortools-5.0.3919-py2.7-macosx-10.6-intel.egg/ortools/linear_solver/pywraplp.py", line 326, in Maximize

    coeffs = expr.GetCoeffs()

AttributeError: 'generator' object has no attribute 'GetCoeffs'

make: *** [rpy] Error 1

<-- End Error-->

I have no clue what is going on. I am new to OR tools and python.

I am trying to build a simple program to solve Integer programming (decision variable is either 0 or 1).

Please let me know what went wrong with my code (this is a simple knapsack problem). I appreciate in advance:

from __future__ import print_function

from ortools.linear_solver import pywraplp

def main():

  # Instantiate a mixed-integer solver, naming it SolveIntegerProblem.

  solver = pywraplp.Solver('SolveIntegerProblem',

                           pywraplp.Solver.CBC_MIXED_INTEGER_PROGRAMMING)

  

  #value data

  values = [100,20,50,25,50,75]

  weight = [200,70,20,25,30,150]

  maxweight = 350

  #create variables

  x = {}

  num_items = len(values)

  

  for i in range(num_items):

    x[i] = solver.IntVar(0,1, 'x[%i]' % (i))

  #constraints

  solver.Add(solver.Sum([weight[i] * x[i] for i in range(num_items)]) <= maxweight)

  #objective

  solver.Maximize(solver.Sum(values[i] * x[i]) for i in range(num_items))

  sol = solver.Solve()

  print('Total Cost = ', solver.Objective().value())

  print()

  for i in range(num_items):

    if x[i].solution_value() > 0:

      print('Weight', i, ' Cost = ', values[i])

  print()

  print("Time = ", solver.WallTime(), "milliseconds")

if __name__ == '__main__':

  main()

[Driss Lahlou]

Hello,

I was looking at linear_programming.py and figured out that you didn't set the coefficients correctly in your example.

from __future__ import print_function

from ortools.linear_solver import pywraplp

def main():

# Instantiate a mixed-integer solver, naming it SolveIntegerProblem.

solver = pywraplp.Solver('SolveIntegerProblem',

pywraplp.Solver.CBC_MIXED_INTEGER_PROGRAMMING)

  #value data

values = [100,20,50,25,50,75]

weight = [200,70,20,25,30,150]

maxweight = 350

#create variables

x = {}

num_items = len(values)

for i in range(num_items):

x[i] = solver.IntVar(0,1, 'x[%i]' % (i))

#constraints

c = solver.Constraint(-solver.infinity(), maxweight, 'c')

for i in range(num_items):

c.SetCoefficient(x[i], weight[i])

#solver.Add(solver.Sum([weight[i] * x[i] for i in range(num_items)]) <= maxweight)

#objective

objective = solver.Objective()

objective.SetCoefficient(x[i], values[i])

objective.SetMaximization()

#solver.Maximize(solver.Sum(values[i] * x[i]) for i in range(num_items))

sol = solver.Solve()

print('Total Cost = ', solver.Objective().Value())

print()

for i in range(num_items):

if x[i].solution_value() > 0:

print('Weight', i, ' Cost = ', values[i])

print()

print("Time = ", solver.WallTime(), "milliseconds")

if __name__ == '__main__':

main()

I hope this helps.

Driss

[mlgolden07]

Thanks Driss,

It doesn't give me an error and found a solution. However, the solver only picks an item that is the heaviest, instead of selecting multiple items to reach the maxweight of 350.

That is why in my constraint I tried to sum all the weight to be less than or equal to maxweight.

I followed the MIP example from this page: Comparing MIP and CP, but I don't know how to change my syntax to solve my problem.

[mlgolden07]

I have figured out. It is actually the objective functions that needs to be changed.

From 

  solver.Maximize(solver.Sum(values[i] * x[i]) for i in range(num_items))

To

  solver.Maximize(solver.Sum([values[i] * x[i] for i in range(num_items)]))

it is the location of the closing bracket ']' that was incorrect.

On Monday, January 16, 2017 at 8:50:11 PM UTC-5, mlgolden07 wrote: