hi Claude, why you just dont calculate the distance between your values for calculating similarity? is that possible?



cheers


U.


2012/10/5 Jamie Bullock <jamie@jamiebullock.com>

On 5 Oct 2012, at 10:20, Ed Kelly <morph_2016@yahoo.co.uk> wrote:

>
>
> There may be a way to use the simile algorithm (from ekext) that gives the matched-ness of two numbers based on an arbitrary window size, but it still might involve rolling your own. I have a feeling KNN is very efficient...
> Ed
>

It's efficient, but there may still be room for optimisation in the implementation if that is a priority. It's also conceptually very simple, which is one reason I like it.

best,

Jamie
_______________________________________________
Pd-list@iem.at mailing list
UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list