Norman Maurer wrote:
Am Mittwoch, den 30.08.2006, 16:08 +0200 schrieb Vincenzo Gianferrari
Pini:
To fully Support Gary's proposal we should collect a third field in the
bayesiananalysis_ham and spam tables, containing his different counter,
and the corpuses should be rebuilt from scratch.
+1 That sounds like the "right" way.
IMO in this case the problem would be for those (like me since 3 years)
that have already built a corpus on the database with the current
counters, and don't have a collection of mails to retrain from scratch.
So for instance I would not like it as a user.
But perhaps there are not so many other James users in this situation...
Vincenzo
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]