You are viewing a plain text version of this content. The canonical link for it is here.
Posted to users@spamassassin.apache.org by Philipp Ewald <ph...@digionline.de> on 2019/12/04 16:37:08 UTC
bayes_auto_learn_on_error
Hi guys,
autolearning will be performed only when a bayes classifier had a different opinion from what the autolearner is now trying to teach it
i thought bayes only learn on error like:
score > 5 && Bayes_00 or score < -1 && bayes_99 (+bayes_999)
i dont get it:
score=-1.9 tagged_above=-9999 required=5 tests=[BAYES_00=-1.9] autolearn=ham autolearn_force=no
score=0.813 tagged_above=-9999 required=5 tests=[BAYES_50=0.8, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, HEADER_FROM_DIFFERENT_DOMAINS=0.001, HTML_MESSAGE=0.001, LOTS_OF_MONEY=0.001, T_KAM_HTML_FONT_INVALID=0.01] autolearn=ham autolearn_force=no
score=-1.898 tagged_above=-9999 required=5 tests=[BAYES_00=-1.9, FSL_HELO_NON_FQDN_1=0.001, TVD_SPACE_RATIO=0.001] autolearn=ham autolearn_force=no
score=-1.899 tagged_above=-9999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_NONE=-0.0001, TVD_SPACE_RATIO=0.001] autolearn=ham autolearn_force=no
score=-1.899 tagged_above=-9999 required=5 tests=[BAYES_00=-1.9, TVD_SPACE_RATIO=0.001] autolearn=ham autolearn_force=no
score=0.8 tagged_above=-9999 required=5 tests=[BAYES_50=0.8, DKIMWL_WL_MED=-0.001, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_NONE=-0.0001] autolearn=ham autolearn_force=no
score=-2.808 tagged_above=-9999 required=5 tests=[BAYES_00=-1.9, PYZOR_CHECK=1.392, RCVD_IN_DNSWL_MED=-2.3] autolearn=ham autolearn_force=no
score=-1.998 tagged_above=-9999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, HTML_IMAGE_RATIO_08=0.001, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_NONE=-0.0001] autolearn=ham autolearn_force=no
[...]
score=-3.042 tagged_above=-9999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, HTML_TAG_BALANCE_BODY=1.157, RCVD_IN_DNSWL_MED=-2.3] autolearn=ham autolearn_force=no
score=-6.599 tagged_above=-9999 required=5 tests=[BAYES_05=-0.5, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, HEADER_FROM_DIFFERENT_DOMAINS=0.001, MAILING_LIST_MULTI=-1, RCVD_IN_DNSWL_HI=-5] autolearn=ham autolearn_force=no
can someone explain me this option?
Thanks!
kind regards
Philipp
--
Philipp Ewald
Administrator
DigiOnline GmbH, Probsteigasse 15 - 19, 50670 Köln
Telefon: +49 221 6500-532, Fax: +49 221 6500-690, E-Mail: philipp.ewald@digionline.de
AG Köln HRB 27711, St.-Nr. 5215 5811 0640
Geschäftsführer: Werner Grafenhain
Informationen zum Datenschutz: www.digionline.de/ds
Re: bayes_auto_learn_on_error
Posted by RW <rw...@googlemail.com>.
On Wed, 4 Dec 2019 17:37:08 +0100
Philipp Ewald wrote:
> Hi guys,
>
> autolearning will be performed only when a bayes classifier had a
> different opinion from what the autolearner is now trying to teach it
>
> i thought bayes only learn on error like:
> score > 5 && Bayes_00
Sort of, it's more like
suitable_for_autotraining_as_spam && ! BAYES_99
> ... tests=[BAYES_00=-1.9] ... autolearn=ham autolearn_force=no
That's not supposed to happen.