Saving variables as .mat or xls files without overwriting them with every iteration of a loop

I have written a script that loops through 15 datasets. I want to save some variables as mat files or xls files without them being overwritten on every iteration.

Any tips?

%[ALLEEG EEG CURRENTSET ALLCOM] = eeglab;
for i = 1:15

 name = ['sub', num2str(i), '.set'];
 EEG = pop_loadset('filename',name,'filepath','C:\\kristin\\EXAM2\\EEGrawData\\');
    
%% RECODING TRIGGERS
EEG = recodingTriggers(EEG);
%This function recodes the triggers into the following categories;
%correctGo, incorrectGo, falseAlarm, correctStop;


%% FILTERING
%For now, I am using the Matlab function movmean(A, k, dim) to run a
%three-point average filtering of the data. k = width of points to average.
%
 EEG.data = movmean(EEG.data,3,2);
 
%% RE-REFERENCING;
 commonAvgRef = mean(EEG.data);
 EEG.data = bsxfun(@minus, EEG.data, commonAvgRef);

% Here, I first compute the average value of all electrodes, simply by computing the mean of each
% column accross the rows in the EEG.data file. Then i subtracted the vector containing 
%common average reference from each corresponding column in the EEG.data matrix.  

%% EPOCHING
 run epochingData.m
 run RTs.m
 
%% BASELINE CORRECTION
 
 epochs_CG = baselineCorr(epochs_CG, -200);
 epochs_CS = baselineCorr(epochs_CS, -200);
 epochs_FA = baselineCorr(epochs_FA, -200);


%% ARTIFACT REJECTION

run artifactRejection.m

%% EXTRACTING ERPs
ERP_CG = mean(epochs_CG,3);
ERP_CS = mean(epochs_CS,3);
ERP_FA = mean(epochs_FA,3);


%% extracting and saving mean RTs

meanRT_CG = mean(RT_CG);
meanRT_CS = mean(RT_CS);
meanRT_FA = mean(RT_FA);

meanRT_CG = array2table(meanRT_CG);
meanRT_CS = array2table(meanRT_CS);
meanRT_FA = array2table(meanRT_FA);

writetable(meanRT_CG, 'meanRT_CS.xls'
writetable(meanRT_CS, 'meanRT_CS.xls');
writetable(meanRT_FA, 'meanRT_FA.xls');
%% Writing output

% ???

clear all
clc
end 

I want to save the variables epochs_CG, epochs_FA and epochsCS (3D matrices) as separate mat files for each iteration, and the RTs as xls files. The xls files are saved, but overwritten on each iteration. Any tips?

[Adam]

Just create a filename with i appended to it each iteration e.g.

writetable(meanRT_CG, sprintf( 'meanRT_CS_%i.xls', i ) )

and do the same for mat files.

[User48765]

Perfect! thank you! Sorry for the dumb question, I'm a matlab novice

[Adam]

Everyone learns somewhere ;-)

[Adam Danz]

I wrote this up before seeing Adam's suggestion above but here's a nearly identical solution that uses an anonymous function to create the file name. By the way, why not save all of the matlab variables to the same file?

% Create an anonymous function that creates filenames
% s is the name of the file; example: 'meanRT_CS'
% d is the iteration number (integer)
% e is the file extension without the dot; example: xls
fnameFcn = @(s,d,e)sprintf('%s_%d.%s', s,d,e); 

for i = 1:15
    
    % Save mat files (but why not save all vars to same 
    save(fnameFcn('epochs_CS',i,'mat'), 'epochs_CS')
    save(fnameFcn('epochs_FA ',i,'mat'), 'epochs_FA')
    save(fnameFcn('epochs_CG ',i,'mat'), 'epochs_CG')
    
    % Save xls files
    writetable(meanRT_CG, fnameFcn('meanRT_CG',i,'xls'))
    writetable(meanRT_CS, fnameFcn('meanRT_CS',i,'xls'))
    writetable(meanRT_FA, fnameFcn('meanRT_FA',i,'xls'))
    
end

[User48765]

Hi, when I try to save it, I get this error;

fnameFcn = @(s,d,e)sprintf('%s_%d.%s', s,d,e); 
for i = 1:15    
    save(fnameFcn('epochs_CG',i,'mat'), epochs_CG)
    save(fnameFcn('epochs_FA ',i,'mat'), epochs_CG)
    save(fnameFcn('epochs_CG ',i,'mat'), epochs_CG)
    
    Error using save
    Must be a string scalar or character vector.
    

Did I write it in wrong? Can't seem to get it to work.

[Stephen23]

save unfortunately only accepts the names of variables, so this is what you need to provide it:

save(fnameFcn('epochs_CG',i,'mat'), 'epochs_CG')
                                    ^         ^ must be the variable NAME!

[Adam Danz]

Oops, I should have enclosed the variable names in single quotes in my examples. I'll edit it now.

[Image Analyst]

Maybe put it as an official Answer below so you can get credit for it.

Answer 1

[Adam Danz]

(continued from comments section above).

Here's an implementation of Adam's suggestion.

% Create an anonymous function that creates filenames
% s is the name of the file; example: 'meanRT_CS'
% d is the iteration number (integer)
% e is the file extension without the dot; example: xls
fnameFcn = @(s,d,e)sprintf('%s_%d.%s', s,d,e); 
for i = 1:15
    
    % Save mat files (but why not save all vars to same 
    save(fnameFcn('epochs_CS',i,'mat'), 'epochs_CS')
    save(fnameFcn('epochs_FA ',i,'mat'), 'epochs_FA')
    save(fnameFcn('epochs_CG ',i,'mat'), 'epochs_CG')
    
    % Save xls files
    writetable(meanRT_CG, fnameFcn('meanRT_CG',i,'xls'))
    writetable(meanRT_CS, fnameFcn('meanRT_CS',i,'xls'))
    writetable(meanRT_FA, fnameFcn('meanRT_FA',i,'xls'))
    
end

[User48765]

Thank you!