Importing matlab structure to python

Question

I am very new to Python, and I have to make some changes to existing project at work. I am importing data from Matlab binary data file into my Python script.

lets say original *.m file contains something like this:

> Temperature=61.3
> 
> VibrationSamples=[76,75,76,77, ... a lot of samples, 78]
>
> save(filename.mat, 'Temperature', 'VibrationSamples', '-mat7-binary')

(This is the minimal example I can provide)

I import resulting *.mat into the python script using

> mat = scipy.io.loadmat(filename.mat)

After that I am able to access elements from my matlab structure by using

> variableA = mat('temperature')
> variableB = mat('vibrationSamples')

However I am confused with the result. If I try to print varialbeA, for given example it will come up with two-dimensional array. The output of print variableA is [[61.3]]. In *.m file it is clearly float.

Why is this happening. What is clean way to resolve it? I know I can just use

> variableA = mat('temperature')[0][0]

But this does not look appropriate.

Am I doing something wrong?

That's how MATLAB stores values isn't it? It doesn't have scalars, only arrays. — David Heffernan
– David Heffernan, Commented Jul 1, 2014 at 9:52
Oh, I forgot to mention, that I am a total n00b in MATLAB too. So what would be the appropriate solution? Address values with index[0] as I described in the question? Or do the cast like float() or something else? — Art Gertner
– Art Gertner, Commented Jul 1, 2014 at 9:54

David Heffernan · Accepted Answer · 2014-07-01 10:02:51Z

3

That's just how MATLAB stores scalars. In the MATLAB data model, scalars are represented as 1 by 1 arrays.

So in Python you would pull out the value like this:

mat['Temperature'][0,0]

More information on the MATLAB side of this issue can be found here: http://www.mathworks.co.uk/help/matlab/math/empty-matrices-scalars-and-vectors.html#f1-86433

Scalars

Any individual real or complex number is represented in MATLAB as a 1-by-1 matrix called a scalar value:
A = 5;

ndims(A)        % Check number of dimensions in A
ans =
     2

size(A)         % Check value of row and column dimensions
ans =
     1     1
Use the isscalar function to tell if a variable holds a scalar value:
isscalar(A)
ans =
    1

answered Jul 1, 2014 at 10:02

David Heffernan

616k46 gold badges1.1k silver badges1.5k bronze badges

Sign up to request clarification or add additional context in comments.

1 Comment

Art Gertner Over a year ago

Good explanation. Exactly what I needed. I will only add, that for integers and floats, casting does the same job as addressing using indexes. For example, float(mat['Temperature']) will have the same result as mat['Temperature'][0][0]. This won'e work for strings though

Collectives™ on Stack Overflow

Importing matlab structure to python

1 Answer 1

1 Comment

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

1 Comment

Your Answer

Sign up or log in

Post as a guest

Related