Re: Mixed 64-bit system GerH binaries / BillYscripts --> two-sided training? YES!

   From: Paolo <[email protected]>

   > Hard proof I don't have, to be sure, but my feeling is that the extra 
   > info would still save same serious space when compared to 2 (or 

   easy and already proposed  / discussed for OSB:

   current:
   cssfile_1             cssfile_2                 cssfile_n
   [--hash--][--value--] [--hash--][--value--] ... [--hash--][--value--]

   multiclass format:

   [--hash--][--value_1--][--value_2--]...[--value_n--]

   the multiclass needs to have some % bincount more than current per-class.
   Of course, if 'value' is just [01], that'd be a bargain: up to 32 classes
   for a 32b [value], or up to 4 if 0..255 is enough, etc.

Sadly, it appears not; sometimes the classifiers are really operating
in the 100,000 event realms (not for mail, but other stuff).

So, my plan is to build for 32 bit hashes and featurecounts and
have it defined in a way that it's easy to retrench down to 16 bits
if needed.  Below that the fractional win is small anyway.

     - Bill

-------------------------------------------------------------------------
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/