On 09/15/2014 03:10 PM, RW wrote:
You're assuming that broad and balanced learning with a little
miss-training is necessarily worse than any kind of learning without
miss-training.
Yes and no.
Seems ppl are not worrying about missing learnt ham but an increase in
the very, VERY low scored spa
On Fri, 12 Sep 2014 16:21:48 +0200
Axb wrote:
> On 09/12/2014 03:48 PM, RW wrote:
> > There's a qualitative difference between a threshold of 0.1 and
> > -1.0. At 0.1 ham can be learned just by not hitting any spam tests,
>
> Which means that FNs get easily learnt as ham, which is what we're
> t
On 09/12/2014 03:48 PM, RW wrote:
There's a qualitative difference between a threshold of 0.1 and -1.0.
At 0.1 ham can be learned just by not hitting any spam tests,
Which means that FNs get easily learnt as ham, which is what we're
trying to avoid.
On Thu, 11 Sep 2014 16:11:33 +0200
Axb wrote:
> On 09/10/2014 11:19 PM, RW wrote:
> > On Wed, 10 Sep 2014 20:57:35 +0200
> > Axb wrote:
> >
> >
> >>> > >In practice this means that, without custom rules, ham can
> >>> > >only be autolearned if it hits a DNS whitelist rule or
> >>> > >RP_MATCHES_RC
On 09/10/2014 11:19 PM, RW wrote:
On Wed, 10 Sep 2014 20:57:35 +0200
Axb wrote:
> >In practice this means that, without custom rules, ham can only be
> >autolearned if it hits a DNS whitelist rule or RP_MATCHES_RCVD.
> >
>
>from what I'm seeing is that it takes lower scored ham to autolearn
>
On Wed, 10 Sep 2014 20:57:35 +0200
Axb wrote:
> > In practice this means that, without custom rules, ham can only be
> > autolearned if it hits a DNS whitelist rule or RP_MATCHES_RCVD.
> >
>
> from what I'm seeing is that it takes lower scored ham to autolearn
> ham. I don't use DNS whitelists a
On 09/10/2014 08:23 PM, RW wrote:
On Wed, 10 Sep 2014 15:47:48 +0200
Axb wrote:
for quite a while I've been playing with autolearn settings
SA's default is:
bayes_auto_learn_threshold_nonspam0.1
this *can* cause low scored spam to be learnt as ham.
For several months I
Hi,
SA's default is:
bayes_auto_learn_threshold_nonspam0.1
this *can* cause low scored spam to be learnt as ham.
For several months I've been using
bayes_auto_learn_threshold_nonspam -1.0
and so far no more false negatives have been learnt as ham which is
was hoping for.
If yo
Hi,
For several months I've been using
bayes_auto_learn_threshold_nonspam -1.0
Any reason you chose -1.0 rather than something a bit closer to 0,
like -0.5 or -0.2? Most of my low-scoring spam is pretty close to 0,
so I'm just wondering.
I know I made the decision years ago to l
On Wed, 10 Sep 2014 15:47:48 +0200
Axb wrote:
> for quite a while I've been playing with autolearn settings
>
> SA's default is:
> bayes_auto_learn_threshold_nonspam0.1
>
> this *can* cause low scored spam to be learnt as ham.
>
>
On Sep 10, 2014, at 7:47 AM, Axb wrote:
> For several months I've been using
> bayes_auto_learn_threshold_nonspam -1.0
Any reason you chose -1.0 rather than something a bit closer to 0, like -0.5 or
-0.2? Most of my low-scoring spam is pretty close to 0, so I'm just wo
On 09/10/2014 04:29 PM, Alex Regan wrote:
Hi,
For several months I've been using
bayes_auto_learn_threshold_nonspam -1.0
and so far no more false negatives have been learnt as ham which is
was hoping for.
If you're using autolearn, you may want to play with that threshold..
Bas
Hi,
For several months I've been using
bayes_auto_learn_threshold_nonspam -1.0
and so far no more false negatives have been learnt as ham which is
was hoping for.
If you're using autolearn, you may want to play with that threshold..
Based on your expertise with Bayes, should we
On 09/10/2014 04:05 PM, Kevin A. McGrail wrote:
On 9/10/2014 9:47 AM, Axb wrote:
for quite a while I've been playing with autolearn settings
SA's default is:
bayes_auto_learn_threshold_nonspam0.1
this *can* cause low scored spam to be learnt as ham.
For several months I
On 9/10/2014 9:47 AM, Axb wrote:
for quite a while I've been playing with autolearn settings
SA's default is:
bayes_auto_learn_threshold_nonspam0.1
this *can* cause low scored spam to be learnt as ham.
For several months I've been using
bayes_auto_learn_threshold_nonspam -
for quite a while I've been playing with autolearn settings
SA's default is:
bayes_auto_learn_threshold_nonspam0.1
this *can* cause low scored spam to be learnt as ham.
For several months I've been using
bayes_auto_learn_threshold_nonspam -1.0
and so far no more false nega
On Sat, 12 May 2007, Maciej Friedel wrote:
On 05/12/07 Abba wrote:
Is anyone using numbers higher than 1 ?
bayes_auto_learn_threshold_nonspam 3.0
bayes_auto_learn_threshold_spam 6.0
How are results of those settings? I'm just curious as I have a hard time
using auto
On 05/12/07 Abba wrote:
> Is anyone using numbers higher than 1 ?
bayes_auto_learn_threshold_nonspam 3.0
bayes_auto_learn_threshold_spam 6.0
maciek
--
|_|0|_| Maciej Friedel <[EMAIL PROTECTED]>
|_|_|0| http://wwv.pl - usługi hostingowe
|0|0|0| http://eprogram.pl - proj
Greetings
In regards to bayes_auto_learn_threshold_nonspam in local.cf
Would anyone chime in again on the use of small negative numbers on this
parameter please?
Is anyone using 0?
Is anyone using numbers between 0 and 1
Is anyone using numbers higher than 1 ?
- rh
--
Abba Communications
Chris Purves wrote:
Running "grep noautolearn /usr/share/spamassassin/*" returns the list of
tests with noautolearn set.
...
No Bayes in this list. If your bayes database is well trained, then I
don't see why it shouldn't be used to determine and train more spam or ham.
It doesn't need to be
I
don't see why it shouldn't be used to determine and train more spam or ham.
My current workaround is to set USER_IN_WHITELIST to the same value as
BAYES_00 and set large thresholds like:
bayes_auto_learn_threshold_nonspam = [0 - 5 - BAYES_00]
bayes_auto_learn_threshold_spam = [require
y? I don't see mention of this on the wiki or
list archives.)
My current workaround is to set USER_IN_WHITELIST to the same value as
BAYES_00 and set large thresholds like:
bayes_auto_learn_threshold_nonspam = [0 - 5 - BAYES_00]
bayes_auto_learn_threshold_spam = [required_score + 5 + BA
22 matches
Mail list logo
