Yes, I noticed that, too. It is odd.

On Fri, Aug 24, 2012 at 9:49 AM, Nadi Tomeh <[email protected]> wrote:
> Hi all,
>
> I was wondering why the -no-lexical-weighting training option prevents
> computing the vocabulary files in train-model.pl
>
> sub get_vocabulary {
>      return unless $___LEXICAL_WEIGHTING;
>
> while they are needed to train the alignment models?
>
> --
> Nadi Tomeh
> Univ. Paris XI, LIMSI-CNRS
> BP 133 F-91403 ORSAY CEDEX
> Tél : +33/0 1 69 85 80 43
> [email protected]

_______________________________________________
Moses-support mailing list
[email protected]
http://mailman.mit.edu/mailman/listinfo/moses-support

Reply via email to