Does that imply one needs to be careful to feed non-filtered messages (i.e., without the spam assassin filters already added) when training manually with sa-learn? Or does it not make much difference?
Also, what is shown in the SA headers when bayesian classification contirbutes tot he score of a message?
Thanks,
Jonathan
Peter Kiem wrote:
Upgraded to SA 2.61, deleted the bayes_* files and relearnt from my spam and ham collections.
sa-learn --dump now reports 1513 spams and 195 hams so looks like it is happy again.
Thanks Theo :)
------------------------------------------------------- This SF.net email is sponsored by: IBM Linux Tutorials. Become an expert in LINUX or just sharpen your skills. Sign up for IBM's Free Linux Tutorials. Learn everything from the bash shell to sys admin. Click now! http://ads.osdn.com/?ad_id=1278&alloc_id=3371&op=click _______________________________________________ Spamassassin-talk mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/spamassassin-talk