Findgroups, order of unique values in an array

Hi,

I am trying to use "findgroups" in order to later split my data into different groups. But the order at which the groups are seperated is important for me. Using "findgroups" the order of groups is not "stable". How can I change that?

I did this:

[G ID]=findgroups(labels_all);

But

ID

doesn't come in that same order as

labels_all

I know that in unique, stable can fix this

q=unique((labels_all),'stable')

here q will be in order that unique values occur in labels_all.

Thank you!

Answer 1

[Stephen23]

[ID,~,G] = unique(labels_all,'stable')

[ardeshir moinian]

Hi Stephen,

Thank you for your answer, and do you know if I use these unique groups in

mean_data=splitapply(@mean,data,G);

would the mean_data stay in the same order as the groups that I found using unique(...,'stable') or not?

Thank you!

[Stephen23]

Lets try it:

G = [2,3,1,2,1,3];
D = [4,5,0,0,2,1];
splitapply(@mean,D,G)

ans = 1×3

1 2 3

Yes, splitapply returns its outputs in the order given by the values of G.