I'm trying to understand how shgo has to be used. In the following example, I would expect that no solution can be found, because the constraint function always returns 1.0 even though it's supposed to be equal to zero. However, shgo terminates successfully:
>>> from scipy.optimize import shgo
>>> shgo(func=lambda x: -x.sum(), bounds=[(0, 1)]*5, constraints=({'type':'eq','fun': lambda x: 1.0}))
fun: -5.0
funl: array([-5.])
message: 'Optimization terminated successfully.'
nfev: 39
nit: 2
nlfev: 6
nlhev: 0
nljev: 1
success: True
x: array([1., 1., 1., 1., 1.])
xl: array([[1., 1., 1., 1., 1.]])
How are constraint functions meant to be used?
