I am trying to create multiple constraint functions to feed scipy.minimize.

The minimize function is:

res1 = minimize(f, x0, args, method='SLSQP', bounds=bnds, constraints=cons, options={'disp': True})

I have set cons to:

cons = [con1, con2, con3, con4]
con1 = {'type': 'eq', 'fun': constraint1}
con2 = {'type': 'eq', 'fun': constraint2}
con3 = {'type': 'eq', 'fun': constraint3}
con4 = {'type': 'eq', 'fun': constraint4}

def constraint1(x):
    return x[0]   x[1]   x[2]   x[3]   x[4]   x[5]   x[6]   x[7]   x[8] - 4321
def constraint2(x):
    return x[9]   x[10]   x[11]   x[12]   x[13]   x[14]   x[15]   x[16]   x[17] - 123
def constraint3(x):
    return x[18]   x[19]   x[20]   x[21]   x[22]   x[23]   x[24]   x[25]   x[26] - 1234
def constraint4(x):
    return x[27]   x[28]   x[29]   x[30]   x[31]   x[32]   x[33]   x[34]   x[35] - 432

How can I automate this process by using a for loop? The problem is to create function with parametric name

CodePudding user response：

You needn't give the functions names at all:

def make_constraint(i,j,c):
  def constraint(x):
    return sum(x[i:j]) c

cons=[dict(type='eq',fun=make_constraint(i*9,(i 1)*9,c))
      for i,c in enumerate([-4321,-123,-1234,-432])]

This sort of approach will not in general run quite as fast as the hand-written functions, since values like i and j must be retrieved on every call; if that matters, it is possible to use the ast module to actually create new Python functions without exec (and its associated lack of structure and security).