regarding the "Calling from MATLAB from Python" interface: how would I convert back array data returned from a matlab function as matlab.double()

I would like to refer to the example in the documentation, https://de.mathworks.com/help/matlab/matlab_external/use-matlab-arrays-in-python.html.

In the example the engine returns b, which is a 1-by-5 matlab.double array.

To further process / use that result in python it might be necessary to convert it back to a python list.

Is there a better way to do this than the following, which might be inefficient and complicated for large multidimensional arrays?

b_py = [v for v in b[0]]

Answer 1

[Elad]

Try

b._data

it should be much faster.

[Ernst Kloppenburg]

I tend to not use properties / methods marked as private (leading underscore) - you never know when this breaks.

[Elad]

That's always a good practice. However, until someone in Mathworks solve this issue, i'm not familar with more efficent way that doesn't include passing through Lists.

To transfer info the other way(and not waiting couple of minutes for a simple 3d array), I had to write even nastier piece of code :o

[Ernst Kloppenburg]

So then your answer is it for the time being. Thank you!