Data Import
Importing data in MATLAB means loading data from an external file. The importdatafunction allows loading various data files of different formats. It has the following five forms:
S.N. |
Function and Description |
1 |
A=importdata(filename)
Loads data into array A from the file denoted by filename. |
2 |
A=importdata('-pastespecial')
Loads data from the system clipboard rather than from a file. |
3 |
A=importdata(___, delimiterIn)
Interprets delimiterIn as the column separator in ASCII file, filename, or the clipboard data. You can use delimiterIn with any of the input arguments in the above syntaxes. |
4 |
A= importdata(___, delimiterIn, headerlinesIn)
Loads data from ASCII file, filename, or the clipboard, reading numeric data starting from lineheaderlinesIn+1. |
5 |
[A,delimiterOut,headerlinesOut]= importdata(___)
dditionally returns the detected delimiter character for the input ASCII file in delimiterOut and the detected number of header lines in headerlinesOut, using any of the input arguments in the previous syntaxes. |
By default, Octave does not have support for importdata() function, so you will have to search and install this package to make following examples work with your Octave installation.
Example 1
Let us load and display an image file. Create a script file and type the following code in it:
filename = 'smile.jpg';
A = importdata(filename);
image(A);
When you run the file, MATLAB displays the image file. However, you must store it in the current directory.
Example 2
In this example, we import a text file and specify Delimiter and Column Header. Let us create a space-delimited ASCII file with column headers, named weeklydata.txt.
Our text file weeklydata.txt looks like this:
Create a script file and type the following code in it:
filename = 'weeklydata.txt';
delimiterIn = ' ';
headerlinesIn = 1;
A = importdata (filename,delimiterIn,headerlinesIn);
% View data
for k = [1:7]
disp(A.colheaders{1, k})
disp(A.data(:, k))
disp(' ')
end
When you run the file, it displays the following result:
SunDay
95.0100
73.1100
60.6800
48.6000
89.1300
MonDay
76.2100
45.6500
41.8500
82.1400
44.4700
TuesDay
61.5400
79.1900
92.1800
73.8200
57.6300
WednesDay
40.5700
93.5500
91.6900
41.0300
89.3600
ThursDay
55.7900
75.2900
81.3200
0.9900
13.8900
FriDay
70.2800
69.8700
90.3800
67.2200
19.8800
SaturDay
81.5300
74.6800
74.5100
93.1800
46.6000
Example 3
In this example, let us import data from clipboard.
Copy the following lines to the clipboard:
Mathematics is simple
Create a script file and type the following code:
A = importdata('-pastespecial')
When you run the file, it displays the following result:
A =
'Mathematics is simple'
Low-Level File I/O
The importdata function is a high-level function. The low-level file I/O functions in MATLAB allow the most control over reading or writing data to a file. However, these functions need more detailed information about your file to work efficiently.
MATLAB provides the following functions for read and write operations at the byte or character level:
Function |
Description |
fclose |
Close one or all open files |
feof |
Test for end-of-file |
ferror |
Information about file I/O errors |
fgetl |
Read line from file, removing newline characters |
fgets |
Read line from file, keeping newline characters |
fopen |
Open file, or obtain information about open files |
fprintf |
Write data to text file |
fread |
Read data from binary file |
frewind |
Move file position indicator to beginning of open file |
fscanf |
Read data from text file |
fseek |
Move to specified position in file |
ftell |
Position in open file |
fwrite |
Write data to binary file |
Import Text Data Files with Low-Level I/O
MATLAB provides the following functions for low-level import of text data files:
- The fscanf function reads formatted data in a text or ASCII file.
- The fgetl and fgets functions read one line of a file at a time, where a newline character separates each line.
- The fread function reads a stream of data at the byte or bit level.
Example
We have a text data file 'myfile.txt' saved in our working directory. The file stores rainfall data for three months; June, July and August for the year 2012.
The data in myfile.txt contains repeated sets of time, month and rainfall measurements at five places. The header data stores the number of months M; so we have M sets of measurements.
The file looks like this:
Rainfall Data
Months: June, July, August
M=3
12:00:00
June-2012
17.21 28.52 39.78 16.55 23.67
19.15 0.35 17.57 NaN 12.01
17.92 28.49 17.40 17.06 11.09
9.59 9.33 NaN 0.31 0.23
10.46 13.17 NaN 14.89 19.33
20.97 19.50 17.65 14.45 14.00
18.23 10.34 17.95 16.46 19.34
09:10:02
July-2012
12.76 16.94 14.38 11.86 16.89
20.46 23.17 NaN 24.89 19.33
30.97 49.50 47.65 24.45 34.00
18.23 30.34 27.95 16.46 19.34
30.46 33.17 NaN 34.89 29.33
30.97 49.50 47.65 24.45 34.00
28.67 30.34 27.95 36.46 29.34
15:03:40
August-2012
17.09 16.55 19.59 17.25 19.22
17.54 11.45 13.48 22.55 24.01
NaN 21.19 25.85 25.05 27.21
26.79 24.98 12.23 16.99 18.67
17.54 11.45 13.48 22.55 24.01
NaN 21.19 25.85 25.05 27.21
26.79 24.98 12.23 16.99 18.67
We will import data from this file and display this data. Take the following steps:
- Open the file with fopen function and get the file identifier.
- Describe the data in the file with format specifiers, such as '%s' for a string, '%d' for an integer, or '%f' for a floating-point number.
- To skip literal characters in the file, include them in the format description. To skip a data field, use an asterisk ('*') in the specifier.
For example, to read the headers and return the single value for M, we write:
M = fscanf(fid, '%*s %*s\n%*s %*s %*s %*s\nM=%d\n\n', 1);
- By default, fscanf reads data according to our format description until it cannot match the description to the data, or it reaches the end of the file. Here we will use for loop for reading 3 sets of data and each time, it will read 7 rows and 5 columns.
- We will create a structure named mydata in the workspace to store data read from the file. This structure has three fields - time, month, and raindata array.
Create a script file and type the following code in it:
filename = '/data/myfile.txt';
rows = 7;
cols = 5;
% open the file
fid = fopen(filename);
% read the file headers, find M (number of months)
M = fscanf(fid, '%*s %*s\n%*s %*s %*s %*s\nM=%d\n\n', 1);
% read each set of measurements
for n = 1:M
mydata(n).time = fscanf(fid, '%s', 1);
mydata(n).month = fscanf(fid, '%s', 1);
% fscanf fills the array in column order,
% so transpose the results
mydata(n).raindata = ...
fscanf(fid, '%f', [rows, cols]);
end
for n = 1:M
disp(mydata(n).time), disp(mydata(n).month)
disp(mydata(n).raindata)
end
% close the file
fclose(fid);
When you run the file, it displays the following result:
12:00:00
June-2012
17.2100 17.5700 11.0900 13.1700 14.4500
28.5200 NaN 9.5900 NaN 14.0000
39.7800 12.0100 9.3300 14.8900 18.2300
16.5500 17.9200 NaN 19.3300 10.3400
23.6700 28.4900 0.3100 20.9700 17.9500
19.1500 17.4000 0.2300 19.5000 16.4600
0.3500 17.0600 10.4600 17.6500 19.3400
09:10:02
July-2012
12.7600 NaN 34.0000 33.1700 24.4500
16.9400 24.8900 18.2300 NaN 34.0000
14.3800 19.3300 30.3400 34.8900 28.6700
11.8600 30.9700 27.9500 29.3300 30.3400
16.8900 49.5000 16.4600 30.9700 27.9500
20.4600 47.6500 19.3400 49.5000 36.4600
23.1700 24.4500 30.4600 47.6500 29.3400
15:03:40
August-2012
17.0900 13.4800 27.2100 11.4500 25.0500
16.5500 22.5500 26.7900 13.4800 27.2100
19.5900 24.0100 24.9800 22.5500 26.7900
17.2500 NaN 12.2300 24.0100 24.9800
19.2200 21.1900 16.9900 NaN 12.2300
17.5400 25.8500 18.6700 21.1900 16.9900
11.4500 25.0500 17.5400 25.8500 18.6700
Data Export
Data export in MATLAB means to write into files. MATLAB allows you to use your data in another application that reads ASCII files. For this, MATLAB provides several data export options.
You can create the following type of files:
- Rectangular, delimited ASCII data file from an array.
- Diary (or log) file of keystrokes and the resulting text output.
- Specialized ASCII file using low-level functions such as fprintf.
- MEX-file to access your C/C++ or Fortran routine that writes to a particular text file format.
Apart from this, you can also export data to spreadsheets.
There are two ways to export a numeric array as a delimited ASCII data file:
- Using the save function and specifying the -ASCII qualifier
- Using the dlmwrite function
Syntax for using the save function is:
save my_data.out num_array -ASCII
where, my_data.out is the delimited ASCII data file created, num_array is a numeric array and �ASCII is the specifier.
Syntax for using the dlmwrite function is:
dlmwrite('my_data.out', num_array, 'dlm_char')
where, my_data.out is the delimited ASCII data file created, num_array is a numeric array and dlm_charis the delimiter character.
Example
The following example demonstrates the concept. Create a script file and type the following code:
num_array = [ 1 2 3 4 ; 4 5 6 7; 7 8 9 0];
save array_data1.out num_array -ASCII;
type array_data1.out
dlmwrite('array_data2.out', num_array, ' ');
type array_data2.out
When you run the file, it displays the following result:
1.0000000e+00 2.0000000e+00 3.0000000e+00 4.0000000e+00
4.0000000e+00 5.0000000e+00 6.0000000e+00 7.0000000e+00
7.0000000e+00 8.0000000e+00 9.0000000e+00 0.0000000e+00
1 2 3 4
4 5 6 7
7 8 9 0
Please note that the save -ascii command and the dlmwrite command does not work with cell arrays as input. To create a delimited ASCII file from the contents of a cell array, you can
- Either, convert the cell array to a matrix using the cell2mat function
- Or export the cell array using low-level file I/O functions.
If you use the save function to write a character array to an ASCII file, it writes the ASCII equivalent of the characters to the file.
For example, let us write the word 'hello' to a file:
h = 'hello';
save textdata.out h -ascii
type textdata.out
MATLAB executes the above statements and displays the following result:
1.0400000e+02 1.0100000e+02 1.0800000e+02 1.0800000e+02 1.1100000e+02
Which are the characters of the string 'hello' in 8-digit ASCII format.
Writing to Diary Files
Diary files are activity logs of your MATLAB session. The diary function creates an exact copy of your session in a disk file, excluding graphics.
To turn on the diary function, type:
diary
Optionally, you can give the name of the log file, say:
diary logdata.out
To turn off the diary function:
diary off
You can open the diary file in a text editor.
Exporting Data to Text Data Files with Low-Level I/O
So far, we have exported numeric arrays. However, you may need to create other text files, including combinations of numeric and character data, nonrectangular output files, or files with non-ASCII encoding schemes. For these purposes, MATLAB provides the low-level fprintf function.
As in low-level I/O file activities, before exporting, you need to open or create a file with the fopenfunction and get the file identifier. By default, fopen opens a file for read-only access. You should specify the permission to write or append, such as 'w' or 'a'.
After processing the file, you need to close it with fclose(fid) function.
The following example demonstrates the concept:
Example
Create a script file and type the following code in it:
% create a matrix y, with two rows
x = 0:10:100;
y = [x; log(x)];
% open a file for writing
fid = fopen('logtable.txt', 'w');
% Table Header
fprintf(fid, 'Log Function\n\n');
% print values in column order
% two values appear on each row of the file
fprintf(fid, '%f %f\n', y);
fclose(fid);
% display the file created
type logtable.txt
When you run the file, it displays the following result:
Log Function
0.000000 -Inf
10.000000 2.302585
20.000000 2.995732
30.000000 3.401197
40.000000 3.688879
50.000000 3.912023
60.000000 4.094345
70.000000 4.248495
80.000000 4.382027
90.000000 4.499810
100.000000 4.605170
MATLAB is available in both commercial and academic versions with new releases biannually e. g. R2013a (released around March 2013), and R2011b (released around September 2013). MATLAB itself is the core product and is augmented by additional toolboxes, many of which have to be purchased separately.
