matlab Python Module

`matlab` Python Module

The matlab Python^® module provides classes to represent arrays of MATLAB^® numeric types as Python variables. You can use these classes to pass MATLAB arrays between Python and MATLAB.

MATLAB Classes in the `matlab` Python Module

You can use MATLAB numeric arrays in Python code by importing the matlab Python package and calling the necessary constructors. For example:

import matlab
a = matlab.double([[1, 2, 3],[4, 5, 6]])

The name of the constructor indicates the MATLAB numeric type. You can pass MATLAB arrays as input arguments to MATLAB functions called from Python. When a MATLAB function returns a numeric array as an output argument, the array is returned to Python.

You can customize the array constructors as follows:

You can initialize an array with an optional initializer input argument that contains numbers. The initializer argument, which is the first positional argument, must be a Python sequence type such as a list, tuple, or range. You can specify initializer to contain multiple sequences of numbers.
You can initialize an array with an optional vector input argument that contains input of size 1-by-N. If you use vector, you cannot use initializer.
Note
When the input is of size 1-by-N, using vector is more efficient than using initializer. Python always knows the length of a one-dimensional sequence, and it can use this information to perform a single allocation of the array that will hold the output.
You can create a multidimensional array using one of the following options:
- Specify a nested sequence without specifying the size.
- Specify a nested sequence and also specify a size input argument that matches the dimensions of the nested sequence.
- Specify a one-dimensional sequence together with a multidimensional size. In this case, the sequence is assumed to represent the elements in column-major order.
You can create a MATLAB array of complex numbers by setting the optional is_complex input argument to True.
You can use custom types to create MATLAB arrays in Python. The custom type must implement the Python buffer protocol. One example is ndarray in NumPy.

You can create MATLAB arrays using these classes:

Class from `matlab` Python Package	Constructor Call in Python	Examples
`matlab.double`	matlab.double(initializer=None\|vector=None, size=None, is_complex=False)	>>> a = matlab.double(4) >>> b = matlab.double(vector=[11, 22, 33]) >>> c = matlab.double([[10, 20],[30,40]]) >>> d = matlab.double(initializer=[[10, 20],[30,40]], size=[2,2],is_complex=False) >>> e = matlab.double(vector=range(0, 20)) >>> f = matlab.double(vector=[x*x for x in range(0, 10, 2)]) >>> g = matlab.double([[1.1+2.4j, 3+4j],[5.3,6.7]], is_complex=True)
`matlab.single`	matlab.single(initializer=None\|vector=None, size=None, is_complex=False)	>>> a = matlab.single([[1.1, 2.2, 3.3],[4.4, 5.5, 6.6]]) >>> a = matlab.single(vector=[11, 22, 33], is_complex=False)
`matlab.int8`	matlab.int8(initializer=None\|vector=None, size=None, is_complex=False)	>>> a = matlab.int8([[11, 22, 33],[44, 55, 66]]) >>> a = matlab.int8(vector=[11, 22, 33], is_complex=False)
`matlab.int16`	matlab.int16(initializer=None\|vector=None, size=None, is_complex=False)	>>> e = matlab.int16([[1+2j, 3+4j],[-5,6]], is_complex=True)
`matlab.int32`	matlab.int32(initializer=None\|vector=None, size=None, is_complex=False)	>>> a = matlab.int32(initializer=[[11, 22, 33],[44, -55, 66]], size=[2,3], is_complex=False)
`matlab.int64`	matlab.int64(initializer=None\|vector=None, size=None, is_complex=False)	>>> a = matlab.int64([[11, 22, 33],[44, -55, 66]])
`matlab.uint8`	matlab.uint8(initializer=None\|vector=None, size=None, is_complex=False)	>>> a = matlab.uint8([[11, 22, 33],[44, 55, 66]]) >>> b = matlab.uint8(vector=[11, 22, 33], is_complex=False)
`matlab.uint16`	matlab.uint16(initializer=None\|vector=None, size=None, is_complex=False)	>>> a = matlab.uint16(initializer=[[11, 22, 33],[44, 55, 66]], size=[2,3], is_complex=False) >>> b = matlab.uint16(vector=[11, 22, 33], is_complex=False) >>> c = matlab.uint16([[11, 22, 33],[44, 55, 66]])
`matlab.uint32`	matlab.uint32(initializer=None\|vector=None, size=None, is_complex=False)	>>> a = matlab.uint32(vector=[11, 22, 33], is_complex=False) >>> b = matlab.uint32([[11, 22, 33],[44, 55, 66]])
`matlab.uint64`	matlab.uint64(initializer=None\|vector=None, size=None, is_complex=False)	>>> a = matlab.uint64([[11, 22, 33],[44, 55, 66]]) >>> b = matlab.uint64(vector=[11, 22, 33], is_complex=False)
`matlab.logical`	matlab.logical(initializer=None\|vector=None, size=None)^a	>>> a = matlab.logical(initializer=[[True, False, True],[True, True, True]], size=[2,3]) >>> b = matlab.logical([[True, False, True],[True, True, True]]) >>> c = matlab.logical(vector=[True, False, True]) >>> d = matlab.logical([True, False, True])
^a Logicals cannot be made into an array of complex numbers.

Properties and Methods of MATLAB Classes in the `matlab` Python Package

All MATLAB arrays created with matlab package constructors have the following properties and methods:

Properties

Property Name	Description	Examples
`size`	A tuple of integers representing the dimensions of an array	>>> a = matlab.int16([[1, 2, 3],[4, 5, 6]]) >>> a.size (2, 3)
`itemsize`	An integer representing the size in bytes of an element of the array	>>> a = matlab.int16() >>> a.itemsize 2 >>> b = matlab.int32() >>> b.itemsize 4

Methods

Method Name	Purpose	Examples
`clone()`	Return a new distinct object with contents identical to the contents of the original object	>>> a = matlab.int16( [[1, 2, 3],[4, 5, 6]]) >>> b = a.clone() >>> print(b) [[1,2,3],[4,5,6]] >>> b[0][0] = 100 >>> b matlab.int16( [[100,2,3],[4,5,6]]) >>> print(a ) [[1,2,3],[4,5,6]]
`real()`	Return the real parts of elements that are complex numbers, in column-major order, as a 1-by-N array	>>> a = matlab.int16([[1 + 10j, 2 + 20j, 3 + 30j],[4, 5, 6]], is_complex=True) >>> print(a.real()) [1,4,2,5,3,6]
`imag()`	Return the imaginary parts of elements that are complex numbers, in column-major order, as a 1-by-N array	>>> a = matlab.int16([[1 + 10j, 2 + 20j, 3 + 30j],[4, 5, 6]], is_complex=True) >>> print(a.imag()) [10,0,20,0,30,0]
`noncomplex()`	Return elements that are not complex numbers, in column-major order, as a 1-by-N array	>>> a = matlab.int16( [[1, 2, 3],[4, 5, 6]]) >>> print(a.noncomplex()) [1,4,2,5,3,6]
`reshape(dim1,dim2,...,dimN)` `reshape((dim1,dim2,...,dimN))` `reshape([dim1,dim2,...,dimN])`	Reshape the array according to the dimensions and return the result	>>> a = matlab.int16( [[1, 2, 3],[4, 5, 6]]) >>> print(a) [[1,2,3],[4,5,6]] >>> a.reshape(3, 2) >>> print(a) [[1,5],[4,3],[2,6]]
`toarray()`	Return a standard Python `array.array` object constructed from the contents. Applicable for one-dimensional sequences only.	>>> a = matlab.int16( [[1, 2, 3],[4, 5, 6]]) >>> a[0].toarray() array('h', [1, 2, 3]) >>> b = matlab.int16( [[1 + 10j, 2 + 20j, 3 + 30j],[4, 5, 6]], is_complex=True) >>> b.real().toarray() array('h', [1, 4, 2, 5, 3, 6])
`tomemoryview()`	Return a standard Python `memoryview` object constructed from the contents	>>> a = matlab.int16( [[1, 2, 3],[4, 5, 6]]) >>> b = a.tomemoryview() >>> b.tolist() [[1, 2, 3], [4, 5, 6]] >>> b.shape (2, 3)

Create a MATLAB Array with N Elements

When you create an array with N elements, the size is 1-by-N because it is a MATLAB array.

import matlab
A = matlab.int8([1,2,3,4,5])
print(A.size)

(1, 5)

The initializer is a Python list containing five numbers. The MATLAB array size is 1-by-5, indicated by the tuple (1,5).

Create Multidimensional MATLAB Arrays in Python

In Python, you can create multidimensional MATLAB arrays of any numeric type. Use a nested Python list of floats to create a 2-by-5 MATLAB array of doubles.

import matlab
A = matlab.double([[1,2,3,4,5], [6,7,8,9,10]])
print(A)

[[1.0,2.0,3.0,4.0,5.0],[6.0,7.0,8.0,9.0,10.0]]

The size attribute of A shows it is a 2-by-5 array.

print(A.size)

(2, 5)

Index into MATLAB Arrays in Python

You can index into MATLAB arrays just as you can index into Python lists and tuples.

import matlab
A = matlab.int8([1,2,3,4,5])
print(A[0])

[1,2,3,4,5]

The size of the MATLAB array is (1,5); therefore, A[0] is [1,2,3,4,5]. Index into the array to get 3.

print(A[0][2])

3

Python indexing is zero-based. When you access elements of MATLAB arrays in a Python session, use zero-based indexing.

This example shows how to index into a multidimensional MATLAB array.

A = matlab.double([[1,2,3,4,5], [6,7,8,9,10]])
print(A[1][2])

8.0

Slice MATLAB Arrays in Python

You can slice MATLAB arrays just as you can slice Python lists and tuples.

import matlab
A = matlab.int8([[1,2,3,4,5]])
print(A[0][1:4])

[2,3,4]

You can assign data to a slice. This example shows an assignment from a Python list to the array.

A = matlab.double([[1,2,3,4],[5,6,7,8]])
A[0] = [10,20,30,40]
print(A)

[[10.0,20.0,30.0,40.0],[5.0,6.0,7.0,8.0]]

You can assign data from another MATLAB array, or from any Python iterable that contains numbers.

You can specify slices for assignment, as shown in this example.

A = matlab.int8([1,2,3,4,5,6,7,8])
A[0][2:4] = [30,40]
A[0][6:8] = [70,80]
print(A)

[[1,2,30,40,5,6,70,80]]

Reshape MATLAB Arrays in Python

You can reshape a MATLAB array in Python with the reshape method. The input argument, size, must be a sequence that does not change the number of elements in the array. Use reshape to change a 1-by-9 MATLAB array to 3-by-3. Elements are taken from the original array in column-major order.

import matlab
A = matlab.int8([1,2,3,4,5,6,7,8,9])
A.reshape((3,3))
print(A)

[[1,4,7],[2,5,8],[3,6,9]]

Create MATLAB Arrays Using Custom Types

You can use custom types such as ndarray in NumPy to create MATLAB arrays in Python. The custom type must implement the Python buffer protocol.

import matlab
import numpy

nf = numpy.array([[1.1, 2,2, 3.3], [4.4, 5.5, 6.6]])
md = matlab.double(nf)
ni32 = numpy.array([[1, 2, 3], [4, 5, 6]], dtype='int32')
mi32 = matlab.int32(ni32)