> I have read the suggestions about occasional retraining. > Is it possible to automate this? Because I have only found docs that > talks about manually doing that. > > I was thinking about running a cron job that dumps all training data, > except the training data for the last X days. > Is this possible and if yes, how?
You can use the sb_mboxtrain.py script (or training via sb_filter). You'll need to have a collection of ham and spam that is is managed by the cron job (i.e. it deletes old files), in one of the formats that sb_mboxtrain understands (e.g. mbox). You can then use sb_mboxtrain to create a fresh database every X days, and mv that to wherever SpamBayes expects to find your database (while SpamBayes isn't running, ideally). =Tony.Meyer -- Please always include the list (spambayes at python.org) in your replies (reply-all), and please don't send me personal mail about SpamBayes. http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this. _______________________________________________ [email protected] http://mail.python.org/mailman/listinfo/spambayes Check the FAQ before asking: http://spambayes.sf.net/faq.html
