将Matlab数据文件读入Python，需要导出为CSV

Question

I have read a Matlab file containing a large amount of arrays as a dataset into Python storing the Matlab Dictionary under the variable name mat using the command:

mat = loadmat('Sample Matlab Extract.mat')

Is there a way I can then use Python's write to csv functionality to save this Matlab dictionary variable I read into Python as a comma separated file?

with open('mycsvfile.csv','wb') as f:
   w = csv.writer(f)
   w.writerows(mat.items())
   f.close()

creates a CSV file with one column containing array names within the dictionary and then another column containing the first element of each corresponding array. Is there a way to utilize a command similar to this to obtain all corresponding elements within the arrays inside of the 'mat' dictionary variable?

Answer 1

The function scipy.io.loadmat generates a dictionary looking something like this:

{'__globals__': [],
 '__header__': 'MATLAB 5.0 MAT-file, Platform: MACI, Created on: Wed Sep 24 16:11:51 2014',
 '__version__': '1.0',
 'a': array([[1, 2, 3]], dtype=uint8),
 'b': array([[4, 5, 6]], dtype=uint8)}

It sounds like what you want to do is make a .csv file with the keys "a", "b", etc. as the column names and their corresponding arrays as data associated with each column. If so, I would recommend using pandas to make a nicely formatted dataset that can be exported to a .csv file. First, you need to clean out the commentary members of your dictionary (all the keys beginning with "__"). Then, you want to turn each item value in your dictionary into a pandas.Series object. The dictionary can then be turned into a pandas.DataFrame object, which can also be saved as a .csv file. Your code would look like this:

import scipy.io
import pandas as pd

mat = scipy.io.loadmat('matex.mat')
mat = {k:v for k, v in mat.items() if k[0] != '_'}
data = pd.DataFrame({k: pd.Series(v[0]) for k, v in mat.items()}) # compatible for both python 2.x and python 3.x

data.to_csv("example.csv")

Answer 2

This is correct solution for converting any .mat file into .csv file. Try it

   import scipy.io
   import numpy as np
   data = scipy.io.loadmat("file.mat")

   for i in data:
        if '__' not in i and 'readme' not in i:
              np.savetxt(("file.csv"),data[i],delimiter=',')

Answer 3

import scipy.io
import pandas as pd
import numpy as np
from datetime import datetime, timedelta

class MatDataToCSV():

    def init(self):

        pass

    def convert_mat_tocsv(self):

        mat = scipy.io.loadmat('wiki.mat')

        instances = mat['wiki'][0][0][0].shape[1]
        columns = ["dob", "photo_taken", "full_path", "gender",\
                "name", "face_location", "face_score", "second_face_score"]
        df = pd.DataFrame(index = range(0,instances), columns = columns)

        for i in mat:
            if i == "wiki":
                current_array = mat[i][0][0]
                for j in range(len(current_array)):
                    df[columns[j]] = pd.DataFrame(current_array[j][0])
        return df

Answer 4

reading a matfile (.MAT) with the below code data = scipy.io.loadmat(files[0])

gives a dictionary of values and keys

and " ' header ', ' version ', ' globals '" these are some of the default values which we need to remove

cols=[]
for i in data:
    if '__' not in i :
       cols.append(i)
temp_df=pd.DataFrame(columns=cols)
for i in data:
    if '__' not in i :
       temp_df[i]=(data[i]).ravel()

we remove the unwanted header values using "if '__' not in i:" and then make a dataframe using the rest of the headers and finally assign the column values to respective column headers