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