matlab load python scripts have different result when directly run in python

I load a python user-defined scripts in Matlab but I got different results with Python.

In python, I have a function to calculate the numerical intergral of an intergral function.

In matlab load this python scripts by:

system('python Userdefined.py');
A = py.Userdefined.Intergral(0.2);

But here 'A' is different with python result when I directly run 'Intergral(0.2)' in python.

I have confirmed that, Matlab result is wrong but I don't know why!

I just load it in Matlab.

[KOU DU]

By the way, because the python script has a very complicated coding. It wastes time to change it to Matlab. I just load it in Matlab but got different results. I am so confused.

[the cyclist]

Are you able to upload the script? I don't see how to help without it.

[Mohammad Sami]

what is the purpose of the system command you issue before calling the function ?

[KOU DU]

I use system to load the python user-defined scripts.

Here I put an example of python script which is named 'Intergral.py':

#! /usr/bin/env python
import math
import sympy
from sympy import *
import numpy as np
from numpy import *
import scipy
from scipy import special
from scipy import *
from scipy.integrate import quad
from scipy import integrate

def da_theta_phi_np(p,theta,phi):
    twop=2*p ; one_over_p=1./p ; one_over_twop=1./(2*p)
    ct=np.cos(theta) ; st=np.sin(theta)
    cp=np.cos(phi) ; sp=np.sin(phi)
    r0=ct**twop+(st*cp)**twop+(st*sp)**twop
    r=r0**(-one_over_twop)
    tpm=twop-1.
    drdtheta=(st*ct**tpm-ct*st**tpm*(cp**twop+sp**twop))\
        *r0**(-one_over_twop-1)
    drdphi=(st**twop*(cp*sp**tpm-sp*cp**tpm))\
        *r0**(-one_over_twop-1)
    st=np.sin(theta)
    return r*np.sqrt((r*st)**2+(drdtheta*st)**2+drdphi**2)

def da_np(p,phi):
    aux = quad(lambda theta: da_theta_phi_np(p,theta,phi), 0., acos(1./np.sqrt(2.+np.tan(phi)**2)))
    res=aux[0]
    return res
                           
def surfacearea_np(p):
    aux = quad(lambda phi: da_np(p,phi), 0., math.pi/4.)
    res=aux[0]
    return 48.*res

When I command in Matlab:

    py.Intergral.surfacearea_np(0.2)
    ans =
    1.731011361308392

But I got 5.73153081591379 when I run in python. All results when I give input p<0.5 are different.

[Ishan]

A call to the surfacearea_np with 0.2 as input gave an output of 1.7310113613083917 which is what MATLAB also gives. Essentially MATLAB just creates a call to the given function in the file you named 'Integral.py'. Do make sure you are calling the same function in your python script. Hope this helps!