how to extract (lopp) double variables from cell array?

Hello guys, I am novice with matlab and I can not solve this problem. I have two cell arrays one of 16x1 containing double variables [33x1000] and another 16x1 containing labels. My purpose is to extract the double variables from the cell array and rename these variables with the labels contained in the other cell array, with a lopp.

Do you have any idea how I can proceed?

Thanks you!!

[Stephen23]

"My purpose is to extract the double variables from the cell array and rename these variables with the labels contained in the other cell array..."

What exactly is the goal of this? Why do you need to do this?

http://xyproblem.info/

https://en.wikipedia.org/wiki/XY_problem

Don't tell us about the code you are trying to write, explain what you are trying to achieve and how you want to process your data.

Answer 1

[Ameer Hamza]

Not a good idea to name variables dynamically: https://www.mathworks.com/matlabcentral/answers/304528-tutorial-why-variables-should-not-be-named-dynamically-eval. Your current data structure is fine. You might benefit by creating a table to further organize your data

C1; % cell array of matrices
C2; % cell array of labels
T = table(C1, C2)

[Stephen23]

alfonso gravante's "Answer" moved here:

Thanks for the answer, I realize it’s a bad idea to proceed like this, but what you suggested doesn’t work for me, I try to explain.

I have a series of excel files that contain several sheets. My goal is to extract the contents of these sheets for each excel file and save them individually with their name. I created this script that allows me to extract the contents of sheets from individual files, but separate data and names into two distinct cell arrays.My goal would therefore be to obtain from a single excel file a single matlab file, renamed with the same name as the excel one (eg. pSpk_001a_sh_raster_data.m).

I repeat, I know that is not the correct way to proceed, but with my little experience I cannot find a better solution.

thank you so much for any help

folder='C:\Users\Desktop\on_the_fly\matrix_decoding';
ff='pSpk_*_sh_raster_data.xls'
names=dir(fullfile(folder,ff))
filename = {names.name}
for ii = 1:numel(filename)
  f=fullfile(folder,filename{ii});
  [status,sheets] = xlsfinfo(f);
  result=[];
  for k=1:numel(sheets)
      result{k,1}=xlsread(f,sheets{k});
  end
  out{ii}=result;
  out=out';
end
for ii = length(out):-1:1 
a{ii}=cell2mat(out{1,1});
end

[Stephen23]

"... I cannot find a better solution."

.. to what problem?

It appears that your code imports the data nicely into a cell array, which can be easily accessed using indexing.

[Stephen23]

alfonso gravante's "Answer" moved here:

thanks for the quick answer.

I followed your advice, indexing I reached my goal. However, I wonder if I can automate my script more.

My final goal is to get a.mat file organized so:

This part of the script creates a cell array (raster_data_all) that contains all raster_data as double variables.

folder='C:\Users\Stefano\Desktop\on_the_fly\matrix_decoding';
ff='N*_SPK*.xls';
names=dir(fullfile(folder,ff));
filename = {names.name}';
for ii = 1:numel(filename);
  f=fullfile(folder,filename{ii});
  [status,sheets] = xlsfinfo(f);
  result=[];
  for k=1:numel(sheets)
      result{k,1}=xlsread(f,sheets{k});
  end
  out{ii}=result;
end

for ii = length(out):-1:1 ;
raster_data_all{ii}=cell2mat(out{1,1});
raster_data_all=raster_data_all';
end

this second part creates the struct (raster_labels) which contains two cell arrays (stimulus_ID and stimulus_position).

%for create a raster_labels structure 
cond1=repmat({'Cue_G_cube'},1,10);
cond2 =repmat({'Cue_G_cylinder'},1,10);
cond3=repmat({'Cue_G_sphere'},1,3);
stimulus_ID = {cond1{:}, cond2{:}, cond3{:}};
clear ('cond1', 'cond2', 'cond3')

%for create a raster_labels structure 
pos1=repmat({'center'},1,10);
pos2 = repmat({'upper'},1,10);
pos3=repmat({'lower'},1,10);
stimulus_position = {pos1{:}, pos2{:}, pos3{:}};
clear ('pos1', 'pos2', 'pos3')

raster_labels.stimulus_ID = stimulus_ID;
raster_labels.stimulus_position = stimulus_position;

finally, I follow your advice and index to access the data and save them

raster_data=raster_data_all{1};
save('N001a_SPK.mat', 'raster_data', 'raster_labels');

The script works that way, but I have to create hundreds of flie (spk001a, spk001b...), all those contained in raster_data_all.

It's possible automate the last step?

I can create, according to you, a loop that does these steps:

1) Individually extract all the doubles contained in the cell array (raster_data_all)

2) rename them as "raster_data

3) save them individually in a mat file (the name of the.mat file takes it from the cell array (filename)

4)add to this file.mat also the struct (raster_labels)

thank you very much for any suggestion

[Stephen23]

"It's possible automate the last step?"

Yes. As long as you can access some data using indexing or fieldnames (or similar) then you can easily process them in a loop. Exporting data to multiple files is a very common task, and the MATLAB documentation shows the basic approach:

https://www.mathworks.com/help/matlab/import_export/process-a-sequence-of-files.html

Use the sprintf to generate the filename, and then save your data using an appropriate function.

[alfonso gravante]

thanks a lot for the suggestions!!