Calling Matlab Functions from Python

Calling MATLAB functions from python

PyMat looks like it's been abandoned.

I'm assuming you are on windows so you could always do the simplest approach and use Matlab's COM interface:

>>> import win32com.client
>>> h = win32com.client.Dispatch('matlab.application')
>>> h.Execute ("plot([0 18], [7 23])")
>>> h.Execute ("1+1")
u'\nans =\n\n     2\n\n'

More info here

calling matlab from python script always displays matlab information even with nodisplay flag

As stated in the comments
-nodisplay is just to run Matlab headless without graphical output and does not disable the commandline or shell output.

-If you don't want any output at all from matlab you can use &>/dev/null as stated here to dump the output or write it to a file.

-If you want to remove the Matlab "welcome Header" forever, you can edit the matlabrc.m file and remove it there, wich might be against the matlab Terms & Conditions. For more information please look here.

-A 3rd Option is to remove the Header using the tail Function of Linux like described here.

This is not part of your Question, but could be handy if you plan to use more Matlab Funcions in Python: There is a Matlab Engine API, that makes it possible to run Matlab funcions without using system. (And also is a lot faster, as you don't need to start Matlab every time you want to call a Matlab function)

calling MATLAB function from Python, what is Python's equivalent of MATLAB 'matrix' datatype?

The answer turned out to be:

import matlab

myMatrix = matlab.double([[1,2],[3,4]])

myMatrix can then be passed to a MATLAB function. MATLAB will even get True if it tests its type: ismatrix(myMatrix).

Calling Matlab function in python3.6 is ten times slower than runing the same function directly in Matlab

That's just the overhead for calling MATLAb from python and not doing it directly.

Maybe you could improve performance by using parquet, that's a format to exchange code between Python and Matlab. If you have large column-oriented data that makes the transfer more efficient. Check it out here:
https://www.mathworks.com/help/matlab/parquet-files.html

You can also find information about this here: https://blogs.mathworks.com/racing-lounge/2020/09/14/using-matlab-and-python-together/
It's about using Python from MATLAB but the conversion format is the same

---

Related Topics