Hi Motoki, What do you mean by word alignments from a monolingual corpus?
I assume you have word alignments of a parallel corpus, and you want to extract phrase pair using your alignment file. If this is the case then you may look here http://www.statmt.org/moses/?n=FactoredTraining.TrainingParameters Scroll down to "if you may have your own method to generate a word alignment, you want to skip ....." You have to pass "--first-step 4" and "--alignment-file /your-alignment-file" to train-model.perl On Fri, Jun 3, 2016 at 12:42 AM, Motoki Wu <tokeste...@gmail.com> wrote: > Hi, > > I have the word alignments (e.g. 0-0, 0-1, 1-2...) from a monolingual > corpus using another program. > > I'd like to only use the phrase align part of the Moses pipeline directly. > > Is it possible to do so? > > Thanks > > - Motoki > _______________________________________________ > Moses-support mailing list > Moses-support@mit.edu > http://mailman.mit.edu/mailman/listinfo/moses-support > -- -Regards, Rajen Chatterjee.
_______________________________________________ Moses-support mailing list Moses-support@mit.edu http://mailman.mit.edu/mailman/listinfo/moses-support
