Looping to create nested structure?

I am using code (provided to me) where data analysis consists of first converting data chunks to structures to run further functions.

I want to run multiple iterations (from different time points within the data) without manually creating a new structure for each time point.

I started with relevant data from time points into a cell array... this works

%%

nml = length(T1);

resp =cell(1,nml);

for i=1:nml;

resp{i} = BM.baselineCorrectedRespiration(1,T1(i):T2(i));

end

%%

Then I wanted to build a structure that contains all the structures corresponding to these time points for analysis

%%

Tot = struct;

for i=1:length(nml)

Tot.bm(i) = breathmetrics(resp{i},Fs,dataType);

end

%%

I tried this but it will only store one structure from the first pair of time points listed and will not continue to add my other datapoints. However, I can add manually to the structure, I'm just not sure how to automate.

Any thoughts?

Thank you!

Answer 1

[Stephen23]

"...without manually creating a new structure for each time point."

Creating individual structures by hand should definitely be avoided!

Probably the best approach would be to use one single non-scalar structure:

https://www.mathworks.com/help/matlab/matlab_prog/access-multiple-elements-of-a-nonscalar-struct-array.html

Assuming that each of the scalar structures returned by that code have the same fields, then is is very easy to create the non-scalar structure after the loop (assuming that the scalar structures are stored in a cell array named C):

S = [C{:}]

https://www.mathworks.com/help/matlab/matlab_prog/comma-separated-lists.html

https://www.mathworks.com/matlabcentral/answers/320713-how-to-operate-on-comma-separated-lists

After that you can use basic indexing to access each structure, e.g. the 2nd structure:

S(2)

[Stephen23]

Something like this:

N = numel(T1);
C = cell(1,N);
for k = 1:N
    tmp = BM.baselineCorrectedRespiration(1,T1(i):T2(i));
    C{k} = breathmetrics(tmp,Fs,dataType);
end 
S = [C{:}]

[Emma Janke]

This is exactly what I needed - thanks for your expertise!