http://issues.apache.org/SpamAssassin/show_bug.cgi?id=3109
------- Additional Comments From [EMAIL PROTECTED] 2006-05-30 02:40 ------- Interesting. And that was still requiring bayes, which I bet is a major time sink. It would be interesting to see the same timing run dropping the bayes_00 and bayes_99 from the metas. I'm sure the classification results will suffer as a result. But it would be an interesting time comparison. ------- You are receiving this mail because: ------- You are the assignee for the bug, or are watching the assignee.
