I've been using SpamAssassin for a couple weeks now, so I'm still pretty new to it. I have use_bayes 1 and auto_learn 1 in my .spamassassin/user_prefs file so shouldn't that mean that it autolearns from my email?
If I run "sa-learn --ham --mbox some_folder", it learns from the contents of the folder. When I run it again, it says that it learned from 0 messages. Well, if all these messages are new since I started using spamassassin, shouldn't it learn from 0 messages the FIRST time I run sa-learn? I read somewhere on this list that auto bayes learning only happens for emails that score above or below a certain number, so it doesn't auto-learn from emails that are in the "grey-zone" between spam and ham. Maybe that's what's happening here? I move my false positives and negatives into separate folders for bayes relearning, so in my setup at least I don't think I'd care if bayes learned some emails incorrectly as I'd correct it manually later. If anybody can tell me what's going on and what I'm missing, I'd appreciate it. Thanks, - Andrew. ----- Andrew Clarke - president, clarke.ca inc. [EMAIL PROTECTED] | http://www.clarke.ca ------------------------------------------------------- This SF.Net email sponsored by: Free pre-built ASP.NET sites including Data Reports, E-commerce, Portals, and Forums are available now. Download today and enter to win an XBOX or Visual Studio .NET. http://aspnet.click-url.com/go/psa00100003ave/direct;at.aspnet_072303_01/01 _______________________________________________ Spamassassin-talk mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/spamassassin-talk
