bogus bogotuning

[Boris 'pi' Piwinger]

"Peter Bishop" <pgb at adelard.com> wrote:

>> > From the tests it appears to perform well on small numbers of email 
>> > messages and does not need a large corpus.
>> 
>> Good to know. I did not test that myself.
>
>In that case I will withdraw the remark.
>As I thought I remembered **you** posting some
>comparaitve  performance tests 
>using bogominitrain, full training, random training etc.
>test

Yes, I did, but not with a small number of messages. I
assume this also works, though.

pi