Re: K means.

[Juan Pablo Carbajal]

On Fri, Dec 9, 2011 at 9:26 AM, Søren Hauberg <address@hidden> wrote:
> fre, 09 12 2011 kl. 08:00 +0100, skrev Søren Hauberg:
>> My current version works like
>>
>>   clusters = kmeans (data, initial_clusters);
>>
>> whereas Matlab's work like
>>
>>   clusters = kmeans (data, number_of_clusters);
>>
>> I think we should at least match this before putting it into the
>> statistics package. But otherwise, I see no problems with not having all
>> the options available that Matlab has.
>
> The attached version has a more compatible API. It's not perfect, but it
> seems to work fairly well.
>
> Søren
>
> _______________________________________________
> Help-octave mailing list
> address@hidden
> https://mailman.cae.wisc.edu/listinfo/help-octave
>

Hi all,

We could use the feature request forum for working functions requiring
 a little review (like missing help, etc...). Once they are acceptable
we can put them in the package. Of course this doesn't make them
visible to user (which are potential curators of the function), but if
the fixing required is small (as seem to be the case here) the forum
may work well.

http://sourceforge.net/tracker/?func=browse&group_id=2888&atid=352888

-- 
M. Sc. Juan Pablo Carbajal
-----
PhD Student
University of Zürich
http://ailab.ifi.uzh.ch/carbajal/