Testing fisher

[Boris 'pi' Piwinger]

Greg Louis wrote:

> Looks like min_dev of 0.025 and cutoff 0.6 worked quite well for you
> with the robx and robs values you employ.  That's interesting; I was
> doing some tuning yesterday and found that 0.25 worked best for me

I did also test this in the meantime:

algorithm=fisher
min_dev=0.25
spam_cutoff = 0.60
Spam:
   4335 test.spam
False negatives:
191
Ham:
  15362 test.ham
False positives:
0

So it gives me a lot more false negatives.

pi