Hi, On Mon, Apr 23, 2012 at 12:15 AM, Yared Mekuria <[email protected]> wrote: > Hello There, > I am on tuning part, but what is the difference of the tunning data > from training data?
Yes, there can't be any overlap between the two. The training data set should be large (100,000s of sentences, if not millions), while the tuning set relatively small (a few 1000s of sentences). -phi _______________________________________________ Moses-support mailing list [email protected] http://mailman.mit.edu/mailman/listinfo/moses-support
