Re: Using database of an old SpamProbe version

Brian Burton wrote:
> Thomas Schürger wrote:
>>
>> Is it possible to use a previous v0.9h database file with v1.2? Or
>> would it be better to start from scratch with an empty database?
>
> My advice would be to back up your database and then score some emails 
> using 1.2.  Also use spamprobe dump|more to see if spamprobe is able to 
> read the database successfull.  I think it ought to work fine.
> 
> If you find that 1.2 is not working properly with your database then use 
> the export/import trick mentioned above.

Can I change my advice? :-)

I think the database format is probably the same but if you aren't too 
opposed to the idea it might be worthwhile to build a new database using 
the auto-train feature.  The reason is that the new parser finds 
different terms than the old one so you might get better initial 
accuracy using a new database with 1.2.

auto-train is a little slow but it seems to do a good job of creating an 
initial database from a set of ham and spam files.  It's similar to 
using train-good and train-spam manually except that you tell it all of 
your files of both types on the command line and it takes care of 
training on messages of both types simultaneously to build a more 
optimal database.

All the best,
++Brian


-------------------------------------------------------
SF.Net email is sponsored by: Discover Easy Linux Migration Strategies
from IBM. Find simple to follow Roadmaps, straightforward articles,
informative Webcasts and more! Get everything you need to get up to
speed, fast. http://ads.osdn.com/?ad_idt77&alloc_id492&op=click