i’m trying to solve an optimization problem by defining all the possible combinations of costs included in an array called values, to solve the problem. To do this, i used the following method:

for i in range(len(values)):
    for j in range(i+1, len(values)+1):
        result.append(values[i:j])

The resulting array will have “len(values) + (len(values)-1) + (…) + 1” rows and if printed will be visualized as:

[1]
[1, 2]
[1, 2, 3]
[1, 2, 3, 4]
...
[1, 2, 3, 4, ..., len(values)]
...
[3]
[3, 4]
...
[3, 4, ..., len(values)]
etc...

I want to add a constraint to impose that the sum of the values contained in every group that I select from the array is lesser or equal than the thrshld I have set. Previously I initialized a binary array x[] with the same size of values. After using m.optimize(), the elements of x[] will be true if their index is the same as the last position in one of the combinations of values i have chosen as part of the optimal solution, and false otherwise…

To help me choose the right index of x[], I defined the following function:

def chooseindex(a, c, k):
  for j in range(a):
    c += a-j
    if k <= c:
      return j

in which a will be the length of the original array, c is the optional offset and k is the index of the element of result that is being considered

My solution was:

m.add_constr((sum(result[i])*(1 - x[len(result[i]) + chooseindex(len(values), 0, i) - 1]) for i in range(len(result)) <= (thrshld - 0.1))

The compiler returned the following error:

'<=' not supported between instances of 'generator' and 'float'

Afterwards I tried rearranging the terms as:

m.add_constr(sum(result[i])*(1 - x[len(result[i]) + chooseindex(len(values), 0, i) - 1])<= (thrshld - 0.1) for i in range(len(result))

but in return I got:

'generator' object has no attribute 'expr'

I also tried replacing thrshld with a list, but with no luck. Any suggestions on how to make this work?