Understanding tuning results

[David Relson]

At 11:08 AM 6/5/03, Boris 'pi' Piwinger wrote:
>David Relson wrote:
>
> > I beg to differ.  The results say that the robs value isn't a significant
> > contributor to the result.  Based on other tests, I'd suggest using 1e-2
> > for the value.
>
>OK:
>
>algorithm=fisher
>robs=1e-2
>robx=0.415
>min_dev=0.475
>ham_cutoff=0.00
>spam_cutoff=0.503000
>spamicity_tags = Spam, Ham
>spamicity_formats = %0.3f, %0.3f
>header_format = %h: %c, spamicity=%p, version=%v/%a
>bogofilter_dir=/usr/local/pi/bogolists/.bogofilter
>
>Spam:
>    6424 test.spam
>False negatives:
>8
>   11133 test.ham
>False positives:
>0
>
>This is really sensationell!

Good.  You're seeing an improvement.

Try this patch:


[relson at osage tuning]$ diff -u ~/bogofilter/cvs/tuning/tuning.sh tuning.sh
--- /home/relson/bogofilter/cvs/tuning/tuning.sh        2003-06-05 
09:00:53.000000000 -0400
+++ tuning.sh   2003-06-05 11:19:40.000000000 -0400
@@ -125,6 +125,6 @@
  # get 10 best results (lowest false negative count)
  (echo "" ; \
  echo "Top 10 results" ; \
-grep fpos < $RESULTS | sort +14n | head -10 ) | tee -a $RESULTS
+grep fpos < $RESULTS | sort -g --key=15 | head -10 ) | tee -a $RESULTS

  date "+%m/%d %H:%M:%S"


Or, from the comand line:

cat YOUR_RESULTS | grep fpos | sort -g --key=15 | head -10