How to get string after loading .mat file to Python

Question

I have a .mat file with variables containing numbers and strings. When I load it and get the variable containing strings I don't understand how to actually get the string out of it:

data = scipy.io.loadmat(pathName)
featurenames=data['featurenames']
print(featurenames[0:2,0])

As an output I get:

[array(['Intensity_SubsBlue_Nuclei_1_IntegratedIntensity'], dtype='<U47')
 array(['Intensity_SubsBlue_Nuclei_2_MeanIntensity'], dtype='<U41')]

How do I get to this array? I want to have just strings.

Thank you!

daniel_hck · Accepted Answer · 2018-06-01 15:50:02Z

1

I am not very familiar reading Matlab files, but you already got an array. If you want to put each string in a variable what you can do is:

string1 = featurenames[0][0]
string2 = featurenames[1][0]

Please, if this is not the answer you are looking, could you be more specific on your question? thanks!

answered Jun 1, 2018 at 15:50

daniel_hck

1,1506 gold badges22 silver badges38 bronze badges

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

Comments

hpaulj · Accepted Answer · 2018-06-01 22:06:47Z

It looks like featurenames is a cell in MATLAB. All matrices and cells are 2d in that language. Cells can contain a mix of elements, size and type. That's more like a python list than a numpy numeric array. loadmat returns that as a 2d object dtype array - containing arrays.

You selected featurenames[0:2, 0], which returns 2 of those cell elements as a 1d array.

I can recreate your array with:

In [9]: arr = np.empty(2, dtype=object)
In [11]: arr[:] = [np.array(['Intensity_SubsBlue_Nuclei_1_IntegratedIntensity'],
    ...:  dtype='<U47'),
    ...:  np.array(['Intensity_SubsBlue_Nuclei_2_MeanIntensity'], dtype='<U41')]
    ...: 
    ...:  
In [12]: arr
Out[12]: 
array([array(['Intensity_SubsBlue_Nuclei_1_IntegratedIntensity'], dtype='<U47'),
       array(['Intensity_SubsBlue_Nuclei_2_MeanIntensity'], dtype='<U41')],
      dtype=object)
In [13]: print(arr)
[array(['Intensity_SubsBlue_Nuclei_1_IntegratedIntensity'], dtype='<U47')
 array(['Intensity_SubsBlue_Nuclei_2_MeanIntensity'], dtype='<U41')]

So you have to access the elements, and then the element within each of those:

In [14]: arr[0][0]
Out[14]: 'Intensity_SubsBlue_Nuclei_1_IntegratedIntensity'
In [15]: [a.item() for a in arr]
Out[15]: 
['Intensity_SubsBlue_Nuclei_1_IntegratedIntensity',
 'Intensity_SubsBlue_Nuclei_2_MeanIntensity']

For a single element array, [0] or item() work equally well.

Or the outer elements can be joined into one array with concatenate. Note the change in dtype:

In [16]: np.concatenate(arr)
Out[16]: 
array(['Intensity_SubsBlue_Nuclei_1_IntegratedIntensity',
       'Intensity_SubsBlue_Nuclei_2_MeanIntensity'], dtype='<U47')
In [17]: _[0]
Out[17]: 'Intensity_SubsBlue_Nuclei_1_IntegratedIntensity'

jshrimp29 · Accepted Answer · 2018-06-01 15:52:16Z

0

What you appear to have is a list of scipy/numpy arrays. The problem you're experiencing is that while datatype doesn't strictly matter to Python, it definitely does to scipy and numpy, and I assume you're looking to get these to Python strings. The snippet below should do it (assuming your output is named matFileOut):

## for the zeroth array in your list, convert the scipy/numpy array to a Python list, then take the zeroth (only) element
matFileOut[0].tolist()[0]

answered Jun 1, 2018 at 15:52

jshrimp29

6537 silver badges9 bronze badges

Collectives™ on Stack Overflow

How to get string after loading .mat file to Python

3 Answers 3

Comments

Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

3 Answers 3

Comments

Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related