Pass extra arguments to a function whose root to find using the "broyden1" solver

[ghost]

Hello,

I'm interested in passing extra arguments to a function whose root to find using the "broyden1" solver of scipy.optimize. I wasn't able to find the solution online so I hope to find it here. I'm using scipy-0.12.1-1.

The following nominal case works fine:

from numpy import cos
from scipy.optimize import broyden1
def F(x):
    return cos(x) + x[::-1] - [1, 2, 3, 4]
x = broyden1(F, [1,1,1,1], f_tol=1e-14)
print x

However, I would like to be able to define F such as:

def F(x, a, b):
    return cos(x) + x[::-1] - [1, 2, 3, 4] + a + b

The "broyden1" function doesn't have the usual arguments "args" to pass extra arguments to F. I tried to modified the "nonlin.py" source code in scipy.optimize in vain (I tried to add "args=()" in line 223 and "args" in line 273, both as argument of each function in question).

I would like to know if there is an easy fix. Thank you.

[argriffing]

You can do something like

from functools import partial
from numpy import cos
from scipy.optimize import broyden1
def F(a, b, x):
    return cos(x) + x[::-1] - [1, 2, 3, 4] + a + b
x = broyden1(partial(F, a, b), [1,1,1,1], f_tol=1e-14)
print x

[ghost]

Thanks argriffing for your quick reply! Your solution solved my problem. I didn't know about the function "partial". It is good to know.