RE: [qmailtoaster] SA Learn Bayes

2007-10-19 Thread Sergio Minini {NETKEY}
Ok, i'll try to take a closer look. Which were the lines you told me to look for in your first reply? The ones that could tell me something about wheter SA is 'learning' or not. Thanks! -Original Message- You'll probably need to fix those issues first - sometimes spamassassin won't

Re: [qmailtoaster] SA Learn Bayes

2007-10-19 Thread Jake Vickers
Sergio Minini {NETKEY} wrote: Ok, i'll try to take a closer look. Which were the lines you told me to look for in your first reply? The ones that could tell me something about wheter SA is 'learning' or not. Thanks! -Original Message- You'll probably need to fix those

Re: [qmailtoaster] SA Learn Bayes

2007-10-19 Thread Jake Vickers
Sergio Minini {NETKEY} wrote: Jake, what should I look for? I get an endless list of items, ending with a [19469] warn: lint: 7 issues detected, please rerun with debug enabled for more information I am sure its way over 200+ msgs now, and it was working before I added the SA-script. I just

RE: [qmailtoaster] SA Learn Bayes

2007-10-19 Thread Sergio Minini {NETKEY}
, but was flagging several messages even before the cronned script. Thanks! -Original Message- From: Jake Vickers [mailto:[EMAIL PROTECTED] Sent: Friday, October 19, 2007 12:20 PM To: qmailtoaster-list@qmailtoaster.com Subject: Re: [qmailtoaster] SA Learn Bayes Sergio Minini {NETKEY

Re: [qmailtoaster] SA Learn Bayes

2007-10-19 Thread Jake Vickers
Sergio Minini {NETKEY} wrote: Hi there. I've been using the SA-learn script in the wiki for several weeks now. I haven't seen an improvement in the number of flagged emails, so I was wondering if it was working. Is there a way to check if the bayesian filters are really 'learning' what the

[qmailtoaster] SA Learn Bayes

2007-10-19 Thread Sergio Minini {NETKEY}
Hi there. I've been using the SA-learn script in the wiki for several weeks now. I haven't seen an improvement in the number of flagged emails, so I was wondering if it was working. Is there a way to check if the bayesian filters are really 'learning' what the scripts are feeding them? Thanks!