Hi Rich. The first alternative is currently not possible. If you call "fit" multiple times, the previous models will be forgotten. Only estimators with "partial_fit" methods (only SGDClassifier at the moment afaik) allow online fitting.
That being said, it would be quite easy to implement partial_fit for the Naive Bayes classifiers and I think that would be a good thing to have. If you want, we can give you some tips on how to do that if you want to take matters into your own hand. As we have pretty strict policies wrt documentation and testing this is not totally trivial if you want to get it merged. Best, Andy ------------------------------------------------------------------------------ LogMeIn Rescue: Anywhere, Anytime Remote support for IT. Free Trial Remotely access PCs and mobile devices and provide instant support Improve your efficiency, and focus on delivering more value-add services Discover what IT Professionals Know. Rescue delivers http://p.sf.net/sfu/logmein_12329d2d _______________________________________________ Scikit-learn-general mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
