On Friday 14 March 2008, Ted Dunning wrote:
> Clustering implementations are notoriously hard to debug, if only because
> they are relatively robust so that broken implementations will often
> produce plausible results.

I can only second that experience. I think one way of making sanity tests 
could be to additionally implement the algorithm with a tool that is closer 
to the mathematical formulas than say Java and compare the results of each 
step.

Isabel

-- 
BOFH excuse #245:The Borg tried to assimilate your system. Resistance is 
futile.
  |\      _,,,---,,_       Web:   <http://www.isabel-drost.de>
  /,`.-'`'    -.  ;-;;,_
 |,4-  ) )-,_..;\ (  `'-'
'---''(_/--'  `-'\_) (fL)  IM:  <xmpp://[EMAIL PROTECTED]>

Attachment: signature.asc
Description: This is a digitally signed message part.

Reply via email to