Is there a way to use findgroups() so that it is based on order in the matrix instead of alphabetical?

I am using findgroups() as an integral part of the script I'm currently working with, but the way that findgroups() outputs its groups is causing issues.

x = {'m1' 'n1' 'k1' 'k1' 'p1' 'p1' 'b1' 'b1' 'b1'}

x = 1×9 cell array
    {'m1'}    {'n1'}    {'k1'}    {'k1'}    {'p1'}    {'p1'}    {'b1'}    {'b1'}    {'b1'}

[g,g_id] = findgroups(x)

g = 1×9

3 4 2 2 5 5 1 1 1

g_id = 1×5 cell array
    {'b1'}    {'k1'}    {'m1'}    {'n1'}    {'p1'}

What I would prefer is if g outputted as [1 2 3 3 4 4 5 5 5] and g_id as {'m1' 'n1' 'k1' 'p1' 'b1'}. Is there any way to do this?

Answer 1

[Star Strider]

There doesn’t appear to be that option for findgroups. If you only have one vector, the unique function can do thaat using the 'stable' setOrder —

x = {'m1' 'n1' 'k1' 'k1' 'p1' 'p1' 'b1' 'b1' 'b1'}

x = 1×9 cell array
    {'m1'}    {'n1'}    {'k1'}    {'k1'}    {'p1'}    {'p1'}    {'b1'}    {'b1'}    {'b1'}

[g,~,g_id] = unique(x,'stable')

g = 1×5 cell array
    {'m1'}    {'n1'}    {'k1'}    {'p1'}    {'b1'}

g_id = 9×1

1 2 3 3 4 4 5 5 5

.

[Liam]

I was wondering if something with unique() would function as a replacement, and this works perfectly, thank you!

[Star Strider]

As always, my pleasure!