2

I have a python script like so:

import numpy as np

def my_function(x):
    return np.array([x])

And I have a MATLAB script to call it:

clear all;
clc;
if count(py.sys.path,'') == 0
    insert(py.sys.path,int32(0),'');
end

myfunction_results = py.python_matlab_test.my_function(8);
display(myfunction_results);

And it displays: 

myfunction_results = 

  Python ndarray with properties:

           T: [1×1 py.numpy.ndarray]
        base: [1×1 py.NoneType]
      ctypes: [1×1 py.numpy.core._internal._ctypes]
        data: [1×8 py.buffer]
       dtype: [1×1 py.numpy.dtype]
       flags: [1×1 py.numpy.flagsobj]
        flat: [1×1 py.numpy.flatiter]
        imag: [1×1 py.numpy.ndarray]
    itemsize: 8
      nbytes: 8
        ndim: 1
        real: [1×1 py.numpy.ndarray]
       shape: [1×1 py.tuple]
        size: 1
     strides: [1×1 py.tuple]

    [8.]

But I do not know how to actaully get the data out of this object. The type is py.numpy.ndarray, but I want to obviously use it in MATLAB as an array or matrix, or integer or something. HOw do I convert it to one of those types?

I've been looking at these: https://www.mathworks.com/help/matlab/examples/call-python-from-matlab.html https://www.mathworks.com/matlabcentral/answers/216498-passing-numpy-ndarray-from-python-to-matlab https://www.mathworks.com/help/matlab/matlab_external/use-matlab-handle-objects-in-python.html

Some of the answers suggest writing to a .mat file. I DO NOT want to write to a file. This needs to be able to run in real time and writing to a file will make it very slow for obvious reasons.

Seems like there is an answer here: "Converting" Numpy arrays to Matlab and vice versa which shows

shape = cellfun(@int64,cell(myfunction_results.shape));
ls = py.array.array('d',myfunction_results.flatten('F').tolist());
p = double(ls);

But I must say that is very cumbersome....is there an easier way?

makansij
  • 7,473
  • 28
  • 82
  • 156

1 Answers1

0

There is currently no uncomplicated way to convert a NumPy ndarray to a Matlab array, but the following Matlab function can perform the conversion:

function A = np2mat(X, tp)
% Convert NumPy ndarray to a Matlab array
if nargin < 2
    tp = 'd';
end
sz = int32(py.array.array('i', X.shape));
if strcmp(tp, 'd')
    A = reshape(double(py.array.array(tp,X.flatten('F'))), sz);
elseif strcmp(tp, 'i')
    A = reshape(int32(py.array.array(tp,X.flatten('F'))), sz);
else
    error('Unknown data type')
end

Default data type is double, specify argument tp='i' to convert an integer array.

jcbsv
  • 396
  • 2
  • 13