> >Why the given 'variance / gap' between
>
> - testing/detecting on the part that is most times 'OK' is more safe - the
> CA should NEVER! crash assp.
>
>>Is this a 'limit' in the current CA process / detection?   >>body<<
>
> Call it a 'limit' - there are more than one 'limit'.
>
> - the currenty used HMM needs allot of memory - for example, if we would
> replace Bayesian by HMM in a default setup (5 workers, 14000 files in
> corpus), the HMM would need more than 1GB RAM
> - large HMM's will need much time in Perl
>
> Last but not least - IMHO it is better to make the base code more solid
> rock, even it takes ten times more time, than doing 'something' on the CA.
> See the CA as a little 'helper' - not as a basic feature.
>
> Thomas


Thanks Thomas,

A good explanation, I can see the 'pain + overhead' of
doing a ['little helper']  => 'CA HMM'

Possibly more than 1Gb of memory is acceptable for these
'rare CA' ASSP excursions, well -at least for those with
dedicated servers. I have 8Gb.

Anyway's, CA is a good 'debug' tool to push ASSP up!

Peter 


------------------------------------------------------------------------------
AppSumo Presents a FREE Video for the SourceForge Community by Eric 
Ries, the creator of the Lean Startup Methodology on "Lean Startup 
Secrets Revealed." This video shows you how to validate your ideas, 
optimize your ideas and identify your business strategy.
http://p.sf.net/sfu/appsumosfdev2dev
_______________________________________________
Assp-test mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/assp-test

Reply via email to