never mind. It might be a bug of my script.
On Mon, May 9, 2011 at 7:49 PM, Shaojun Zhao <[email protected]> wrote: > Dear all, > > I am using cbayes and bayes in the naive bayes package of mahout. > > I found no performace differences between cbayes and bayes if I use > trigram. The difference is quite noticeable if I use unigram, and I > have not yet tried bigram yet. > > Does anyone know why? >
