> > > Hello all > > I have been working with Berkeley Aligner for a while now. I had a > question > regarding it's configuration file. I am trying to align English-Hindi > parallel corpus as provided > at the following link. http://www.cse.unt.edu/~rada/wpt05/#def . This > link provided me with 3 datasets namely:- > > 1)training data (size = 3440 parallel sentence pairs), > 2)development data and (size = 25 parallel sentence pairs), > 3)test data. (size = 90 parallel sentence pairs), > > I have downloaded all three of them and out of these three the development > and testdata have the file containing human world alignments, and the > training data don't have it. On the basis of these datasets I have made > the > following .conf file for Berkeley aligner(attached in this mail) However > on > running the aligner I don't know why it is giving me 3441 lines of > alignment > in output. The name of output file is also train.align. I wanted it to > output the file with alignments of the parallel corpus in the test data. > Can > anyone please tell me what is wrong with my configuration file? > > regards > Debu >
align.conf
Description: Binary data
_______________________________________________ Moses-support mailing list Moses-support@mit.edu http://mailman.mit.edu/mailman/listinfo/moses-support