You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@spamassassin.apache.org by Justin Mason <jm...@jmason.org> on 2004/06/11 02:47:43 UTC
Re: svn commit: rev 21041 - incubator/spamassassin/trunk/masses
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
eh, why drop the FP/FN from the summary line? and it's missing
a newline ;)
quinlan@apache.org writes:
> Author: quinlan
> Date: Thu Jun 10 17:30:15 2004
> New Revision: 21041
>
> Modified:
> incubator/spamassassin/trunk/masses/logs-to-c
> Log:
> more clean-up of output
>
> Modified: incubator/spamassassin/trunk/masses/logs-to-c
> ==============================================================================
> --- incubator/spamassassin/trunk/masses/logs-to-c (original)
> +++ incubator/spamassassin/trunk/masses/logs-to-c Thu Jun 10 17:30:15 2004
> @@ -435,9 +435,6 @@
> }
>
> sub evaluate {
> - my $fprate = ($ga_ny / $num_tests) * 100.0;
> - my $fnrate = ($ga_yn / $num_tests) * 100.0;
> -
> printf ("\n# SUMMARY for threshold %3.1f:\n", $threshold);
> printf "# Correctly non-spam: %6d %4.2f%%\n",
> $ga_nn, ($ga_nn / $num_ham) * 100.0;
> @@ -467,7 +464,7 @@
>
> my $sr = ($nspamspam / $nspam) * 100.0;
> my $sp = ($nspamspam / ($nspamspam + $nlegitspam)) * 100.0;
> - printf "# TCR: %3.6f SpamRecall: %3.3f%% SpamPrec: %3.3f%% FP: %3.2f%% FN: %3.2f%%\n", $tcr, $sr, $sp, $fprate, $fnrate;
> + printf "# TCR: %3.6f SpamRecall: %3.3f%% SpamPrec: %3.3f%%", $tcr, $sr, $sp;
> }
>
> __DATA__
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (GNU/Linux)
Comment: Exmh CVS
iD8DBQFAyQEvQTcbUG5Y7woRAswFAJ0TEbOWAo5ahNE/mhEYlQA0SKK4LACghF+c
sbBpshs5hEzcRAn52YPTOPM=
=QIaX
-----END PGP SIGNATURE-----
Re: svn commit: rev 21041 - incubator/spamassassin/trunk/masses
Posted by Daniel Quinlan <qu...@pathname.com>.
Justin Mason <jm...@jmason.org> writes:
> They should be made relative to spam/ham, and reinstated. otherwise
> I'm -1 on that change.
The FP and FN numbers are still there. :-)
# SUMMARY for threshold 5.0:
# Correctly non-spam: 3643 99.62%
# Correctly spam: 24198 99.56%
# False positives: 14 0.38%
# False negatives: 106 0.44%
# TCR: 30.153846 SpamRecall: 99.564% SpamPrec: 99.942%
If you want to change the formatting and put them on the same line as
the TCR, that's okay with me. I'd favor putting FP and FN numbers on
the TCR line and moving SpamRecall and SpamPrecision up to their own
lines.
I generally look just at TCR, FP, and FN -- I find the latter two more
intuitive than Recall and Precision.
Daniel
--
Daniel Quinlan
http://www.pathname.com/~quinlan/
Re: svn commit: rev 21041 - incubator/spamassassin/trunk/masses
Posted by Daniel Quinlan <qu...@pathname.com>.
jm@jmason.org (Justin Mason) writes:
> eh, why drop the FP/FN from the summary line? and it's missing a
> newline ;)
Because those FP and FN numbers are the ones relative to the total
number of messages rather than the amount of spam or ham.
--
Daniel Quinlan
http://www.pathname.com/~quinlan/
Re: svn commit: rev 21041 - incubator/spamassassin/trunk/masses
Posted by Daniel Quinlan <qu...@pathname.com>.
jm@jmason.org (Justin Mason) writes:
> eh, why drop the FP/FN from the summary line? and it's missing a
> newline ;)
Because those FP and FN numbers are the ones relative to the total
number of messages rather than the amount of spam or ham.
--
Daniel Quinlan
http://www.pathname.com/~quinlan/
Re: svn commit: rev 21041 - incubator/spamassassin/trunk/masses
Posted by Duncan Findlay <du...@debian.org>.
On Thu, Jun 10, 2004 at 05:47:43PM -0700, Justin Mason wrote:
> eh, why drop the FP/FN from the summary line? and it's missing
> a newline ;)
>
> quinlan@apache.org writes:
> > Author: quinlan
> > Date: Thu Jun 10 17:30:15 2004
> > New Revision: 21041
> >
> > Modified:
> > incubator/spamassassin/trunk/masses/logs-to-c
> > Log:
> > more clean-up of output
> >
Aren't these changes supposed to be R-T-C?
--
Duncan Findlay
Re: svn commit: rev 21041 - incubator/spamassassin/trunk/masses
Posted by Duncan Findlay <du...@debian.org>.
On Thu, Jun 10, 2004 at 05:47:43PM -0700, Justin Mason wrote:
> eh, why drop the FP/FN from the summary line? and it's missing
> a newline ;)
>
> quinlan@apache.org writes:
> > Author: quinlan
> > Date: Thu Jun 10 17:30:15 2004
> > New Revision: 21041
> >
> > Modified:
> > incubator/spamassassin/trunk/masses/logs-to-c
> > Log:
> > more clean-up of output
> >
Aren't these changes supposed to be R-T-C?
--
Duncan Findlay