Dear Philip, looking around in Moses, I have seen that there is a -max-phrase-length parameter in the decoder. I have assumed that this parameter limits the decoder to use only those entries of the translation table that contain less than max-phrase-length a number of token for each languages. I have changed this parameters (from the default value 20 to 3, the max phrase length in the training phase is fixed to 7), but the decoder continues to use the same tokens as before... shall I change anything else?
Thanks Marco On Tue, Apr 29, 2008 at 2:54 PM, Philipp Koehn <[EMAIL PROTECTED]> wrote: > Hi, > > I don't think that's correct, the phrase length is enforced > both in the foreign and English. Check > scripts/training/phrase-extract/extract.cpp > for the implementation. > > -phi > > On Tue, Apr 29, 2008 at 3:02 AM, Chris Dyer <[EMAIL PROTECTED]> wrote: > > Yes, the maximum phrase length is the maximum number of tokens in the > > foreign language side of an entry in the translation model. (There is > > no limit to the number of tokens that the target phrase may contain.) > > > > Chris > > > > > > > > On Mon, Apr 28, 2008 at 8:26 PM, marco turchi <[EMAIL PROTECTED]> > wrote: > > > Dear experts, > > > one parameter of the Moses training script is the maximum phrase > length. > > > Is the maximum number of tokens of the foreign language phrase inside > an > > > entry of the translation model? > > > > > > Thanks > > > Marco > > > > > > _______________________________________________ > > > Moses-support mailing list > > > Moses-support@mit.edu > > > http://mailman.mit.edu/mailman/listinfo/moses-support > > > > > > > > _______________________________________________ > > Moses-support mailing list > > Moses-support@mit.edu > > http://mailman.mit.edu/mailman/listinfo/moses-support > > > > >
_______________________________________________ Moses-support mailing list Moses-support@mit.edu http://mailman.mit.edu/mailman/listinfo/moses-support
