ValueError using scipy.optimize on multidimensional function

Question

What am i doing wrong here?

import numpy as np
import scipy.optimize as so

def f(z):
  return z


guess=np.array([1,1])
z0=so.newton(f,guess)

Im getting the following error:

ValueError: The truth value of an array with more than one element is ambiguous. Use a.any() or a.all()

There is some info on scipy docs about vectorizing the function if its domain is multivalued, but cant quite understand what i have to do.

Are you sure you want your function to return an array as well (rather than a single value)? — Nils Gudat
– Nils Gudat, Commented Mar 1, 2019 at 15:28
Your code snippet works in python3 (giving z0 = array([0., 0.])), but not in python2. — Håkon T.
– Håkon T., Commented Mar 1, 2019 at 15:29
What version do you use of scipy (I am on 1.2.1) and numpy (1.15.4)? — Håkon T.
– Håkon T., Commented Mar 1, 2019 at 17:44

Ralf · Accepted Answer · 2019-03-01 18:21:06Z

1

If I run your code with an older version of the lib (Python 3.6.3, scipy 1.1.0), then I get the same error that you showed in your question.

When I upgrade to a newer version (same python, scipy 1.2.1) then the error goes away and the code runs fine.

answered Mar 1, 2019 at 18:21

Ralf

16.6k4 gold badges50 silver badges73 bronze badges

Sign up to request clarification or add additional context in comments.

1 Answer 1