2014-08-28 5:47 GMT+02:00 Josh Wasserstein <[email protected]>: > What prior does scikit-learn use for MultinomialNB? The documentation says: > > class_prior : array-like, size (n_classes,) > Prior probabilities of the classes. If specified the priors are not adjusted > according to the data. > > For N classes, one typically uses a Dirichlet prior, but this prior would be > specified with N positive values (for the concentration parameters), not > with N probabilities (quantities bounded between 0 and 1). So MultinomialNB > seems to be using something different. What is it?
This class prior is just the P(y) in P(y|x) = (P(x|y) × P(y)) / Z. It's a simple multinomial. ------------------------------------------------------------------------------ Slashdot TV. Video for Nerds. Stuff that matters. http://tv.slashdot.org/ _______________________________________________ Scikit-learn-general mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
