Unable to pass multiple constraints using a for-loop

Asked Oct 21 '17 at 11:42

Active Oct 21 '17 at 11:42

Viewed 140 times

I am trying to minimize a function with two constraints: x[0] > 1, x[1] > 1

If I pass them as a tuple (commented out in the code below) I am getting the right answer (which is (1,1)).

However, if I pass them in a for loop, it seems that the 1st constraint (x[0] > 1) is being ignored. I am getting the answer as (0,1).

What am I missing?

from scipy.optimize import minimize

def costFunction(x):
    f = x[0]**2 + x[1]**2
    return f

cons = []
n = 2
for i in range(n):
    cons.append({'type': 'ineq', 'fun': lambda x: x[i] - 1})

# cons = ({'type': 'ineq', 'fun': lambda x: x[0] - 1},
#         {'type': 'ineq', 'fun': lambda x: x[1] - 1})

x0 = [0,0]
sol = minimize(costFunction, x0, constraints = cons)
print(sol['x'])

Thanks.

asked Oct 21 '17 at 11:42

Suvo

That's because the `i` in the lambda is not bounded to the lambda. – Willem Van Onsem Oct 21 '17 at 11:42
you can write it as: `lambda x, j=i: x[j] - 1` – Willem Van Onsem Oct 21 '17 at 11:43
1

There is the same question here - https://stackoverflow.com/questions/19837486/python-lambda-in-a-loop – Igor Oct 21 '17 at 11:49

Unable to pass multiple constraints using a for-loop

0 Answers0