Sort a n array for maximum number of recurring values and storing the indexes corresponding to erased components

Hi,

i'm working with an array of thousands of elements and i've to limit the repeated values to 10.

Let say:

a=[0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,2,0,0,0,0,0,5,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,9];

has to become:

[0,0,0,0,0,0,0,0,0,0,2,0,0,0,0,0,5,0,0,0,0,0,0,0,0,0,0,0,9].

And i want to store the indexes corresponding to the erased 10+-times recurrent values.

Any suggestion would be appreciated.

Thanks in advance,

Alessandro

[Matt J]

And i want to store the indexes corresponding to the erased 10+-times recurrent values.

If I have a repitiion of a number, say [2 2 2 2 2] and I reduce it to [2,2,2,2] clearly I have erased one of the 2's. But which one? The first? The last? One of them in the middle?

[Alessandro Togni]

The 10th, 11th, etc of each list of 10+ occurrences!

Could be useful to say that i have to edit other vector of the same lenght in the same way on the one we're talking about.

Answer 1

[Matt J]

a=[0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,2,0,0,0,0,0,5,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,9];

s=diff([inf,a])~=0;

b=ones(size(a));
b(s)=b(s)-[0,diff(find(s))];

discard=(cumsum(b)>10); %indices to discard

a_short=a(~discard) %truncated version of a

a_short = 1×28

0 0 0 0 0 0 0 0 0 0 2 0 0 0 0 0 5 0 0 0 0 0 0 0 0 0 0 9