How do I call a custom python function from matlab that uses an import statement?

I have a custom python script called testmat.py with two functions:

# testmat.py
  import numpy

def createarray(n):
    P = numpy.ones(2);
    print(P)
    return P

def createarray2(n):
    P = 2;
    print(P)
    return P

def createarray3():
    P = 3.14;
    print(numpy.sin(P))
    return P

I can call createarray2 from matlab by using these commands: testmat2 = py.importlib.import_module('testmat2');

>> testmat2.createarray2(2)
2

ans =

    int64

     2

However when I call createarray, matlab can't handle this request:

   >> testmat2.createarray(2)
  Error using numeric>ones (line 192)
  Python Error: TypeError: 'float' object cannot be interpreted as an index

Error in testmat2>createarray (line 24)
    P = numpy.ones(2);

How do I get matlab to use functions that use import statements.

Answer 1

[Robert Snoeberger]

Based on the error message, "TypeError: 'float' object cannot be interpreted as an index" you are passing a double when you should be passing an integer, such as in32. Try this:

>> testmat2.createarray(int32(2))

Note that the literal 2 is a double in MATLAB. When it is passed to Python, it is automatically converted to a Python float [1]. That is what the error message is complaining about.

[1] https://www.mathworks.com/help/matlab/matlab_external/passing-data-to-python.html