when is Bayes scoring used?
Dene Ulmschneider
dene at DATATECHIE.COM
Wed May 7 14:41:46 IST 2003
Julian-
there was a message that I sent yesterday (clipped it out of last email to
list) that showed the complete output of the command "sa-learn -D
--rebuild". That's how I know the system says there are "Only 87 spam(s) in
Bayes_db < 200".
I cannot figure out why the learn.spam.log is always counting "learned
from" messages but it is not increasing the number when I run the "sa-learn
-D --rebuild".
Any ideas?
Dene
At 01:50 PM 5/7/2003 +0100, you wrote:
>At 13:32 07/05/2003, you wrote:
>>OK - this is really getting a little confusing...
>>
>>I checked my "learn.spam.log" this morning and a found the following entries:
>>
>>Wed May 7 00:01:01 EDT 2003
>>Learned from 4 messages.
>>Learned from 1 messages.
>>Wed May 7 01:01:01 EDT 2003
>>
>>The problem is that when I run sa-learn -D --rebuild I still get the
>>message that says:
>>Only 87 spam(s) in Bayes_db < 200
>>(it should be AT LEAST the 87 form yesterday plus the ones listed above -
>>right?)
>>
>>Can anyone tell me how to fix this? The Bayes files on /root/.spamassassin
>>are all being updated multiple times per day so I know it is working,
>>unless of course the sa-learn command is reading Bayes info from another
>>directory that really DOES only have 87 spam(s).
>>
>>Is there a way to run sa-learn and have it tell you the path that it is
>>reading the Bayes info from?
>
>Have you tried
> sa-learn -D
>? I just ran "sa-learn" on its own and it prints the usage for you.
>
>
>>Thank for any assistance.
>>
>>Dene
>>
>>At 02:38 PM 5/6/2003 -0400, you wrote:
>>>something else to add...
>>>
>>>According the script that Julian provided to run sa-learn through cron,
>>>my log is called "learn.spam.log"
>>>
>>>When I checked that file - I added up all of the "learned form XX
>>>messages" and the total number was 447.
>>>
>>>Is the "learned from" referring to spam and ham? Is it possible that I
>>>have 87 spam and the rest of them a ham? I thought I was pretty sure that
>>>more spam was getting processed than ham - but I could be wrong.
>>>
>>>Can anyone shed a little light?
>>>
>>>Dene
>
>--
>Julian Field
>www.MailScanner.info
>MailScanner thanks transtec Computers for their support
More information about the MailScanner
mailing list