How to create groups by using labels?

I have two arrays, one with values and one with labels, both of the same length.

For example :

V = [ 1 2 3 4 5 6 7 8]

L = { 'A', 'A', 'A', 'C', 'C', 'C', 'A', 'B' }

now I want to use to labels to split V into three groups like:

A = [1 2 3 7]

B = [8]

C = [4 5 6]

How do I do that? Cant find it anywhere

Answer 1

[the cyclist]

Here is one way:

V = [ 1 2 3 4 5 6 7 8];
L = { 'A', 'A', 'A', 'C', 'C', 'C', 'A', 'B' };

[tf,loc] = ismember(L,unique(L));

for ni = 1:max(loc)
    VV{ni} = V(loc==ni)
end

Note that I used a cell array to store the output, rather than the alphabetic sequence variables.

[Jim]

Thank you very much!

Answer 2

[Daniel M]

I don't recommend making variables dynamically for various reasons, but making a struct is fine to do, and can even make fieldnames dynamically:

[uL,~,iL] = unique(L);
for f = 1:length(uL)
    s.(uL{f}) = V(iL==f);
end

The output struct s will have fieldnames that are the same as the labels in L.

[Jim]

Thanks, will try that instead.

[Jim]

This doesnt work if the labels are numbers.

Do you have any idea how to make it work for numberlabels aswell?

[Daniel M]

Well, how would you have done it manually? In your original post you wanted to make variables named A = [1 2 3 7], etc. How would you have done it if the label was a number instead?

You can try prepending a letter before any labels that start with numbers.

L = {'A','A','B','2','C','1','x1','1x'}
s = ~cellfun(@isempty, regexp(L,'^\d'),'UniformOutput',true) % finds the labels that start with a number
L(s) = cellfun(@(x) ['n',x],L(s),'UniformOutput',false); % preprend with the letter n
L

with output

L =

  1×8 cell array

    {'A'}    {'A'}    {'B'}    {'2'}    {'C'}    {'1'}    {'x1'}    {'1x'}
    
s =

  1×8 logical array

   0   0   0   1   0   1   0   1
   
L =

  1×8 cell array

    {'A'}    {'A'}    {'B'}    {'n2'}    {'C'}    {'n1'}    {'x1'}    {'n1x'}