How to sum repeated adjacent values in a vector

Hi guys,

I have a random vector with -1 and 1.

for example:

X = [1, 1, -1, 1, 1, 1, -1, -1, 1, -1, 1, 1, 1, 1, -1, -1]

I wish to creat a vector that sums repeated adjacent values.

so for the above example:

Y = [2, -1, 3, -2, 1, -1, 4, -2]

any idea?

Answer 1

[Cedric Wannaz]

Here is one way:

Y = splitapply( @sum, X, [0, cumsum( diff(X) ~= 0 )] + 1 ) ;

or, split into two expressions that make it easier to understand:

groupId = [0, cumsum( diff(X) ~= 0 )] + 1 ;
Y = splitapply( @sum, X, groupId ) ;

[Andrei Bobrov]

+1.

accumarray(cumsum(diff([0;X(:)])~=0),X(:));

[Galgool]

it does the job. thanks.