MATLAB Answers

How to load all the data in one folder

32 views (last 30 days)
Sikai Wu
Sikai Wu on 19 Feb 2021
Edited: Stephen Cobeldick on 19 Feb 2021
Path = 'D:\50-350\';
File = dir( fullfile(Path,'*.dat'));
FileNames = {File.name}';
Length_Names = size(FileNames,1);
for i=1:Length_Names
filename=strcat(Path, FileNames(i));
eval(['Data',num2str(i),'=','load(filename{1,1})',';']);
end
I have 3 targets here:
1、Find out all the files with '.dat' in folder '50-350';
2、Read the data;
2、Rename the data from ‘Data1 Data2 Data3 Data4...’ one by one;
By the codes above, I successfuly achieve the goals, but I find that the size of the data change.
For example: I load the ' 31.dat ' by the above codes, and rename it as ' Data31'. The size of Data31 is 3353×10 . If I read '31.dat' directly, the size is 2501×10. I am wonder why it happened.
  2 Comments
Stephen Cobeldick
Stephen Cobeldick on 19 Feb 2021
This comment was flagged by Sikai Wu
For robustness you should preallocate Data before the loop.
Also fullfile is recommended instead of string concatenation.
You will probably find that the files are not imported in numeric order: to get numeric order based on numbered filenames (without leading zeros, as your example shows) you will need to do either of these:
  • sort the names alphanumerically.
  • generate the filenames (rather than using DIR), as shown here:

Sign in to comment.

Answers (1)

Stephen Cobeldick
Stephen Cobeldick on 19 Feb 2021
Edited: Stephen Cobeldick on 19 Feb 2021
"I am wonder why it happened."
Because you always load exactly the same file data (note the indexing you used):
filename{1,1}
Your code always loads the first file and ignores all the other files.
In any case your approach is not robust. Here is a simpler and more robust way to import that data:
P = 'D:\50-350\';
S = dir(fullfile(P,'*.dat'));
for k = 1:numel(S)
F = dir(fullfile(P,S(k).name));
S(k).data = readmatrix(F); % much better than EVAL and LOAD.
end
Depending on how the files are named, you might also find this useful:
  2 Comments
Stephen Cobeldick
Stephen Cobeldick on 19 Feb 2021
For 2017a you can try dlmread or cvsread or textscan or whatever suits your file format.

Sign in to comment.

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by