Moses-support Digest, Vol 100, Issue 30

Send Moses-support mailing list submissions to
moses-support@mit.edu

To subscribe or unsubscribe via the World Wide Web, visit
http://mailman.mit.edu/mailman/listinfo/moses-support
or, via email, send a message with subject or body 'help' to
moses-support-request@mit.edu

You can reach the person managing the list at
moses-support-owner@mit.edu

When replying, please edit your Subject line so it is more specific
than "Re: Contents of Moses-support digest..."


Today's Topics:

1. multiple decoding paths (Jorg Tiedemann)


----------------------------------------------------------------------

Message: 1
Date: Sun, 8 Feb 2015 14:18:19 +0100
From: Jorg Tiedemann <tiedeman@gmail.com>
Subject: [Moses-support] multiple decoding paths
To: moses-support <moses-support@mit.edu>
Message-ID: <04878ECE-8EC1-418C-A19F-3D1DD41150C5@gmail.com>
Content-Type: text/plain; charset="iso-8859-1"


I have a question on decoding with multiple decoding paths. Can the various decoding paths be mixed in the final complete hypothesis or does the score of one decoding path compete with the one from the others?

I use different TMs in different decoding paths and I have seen that my output has scores set for all of them for a given sentence. This suggests that the final hypothesis uses options from both paths but is that really the case?

Another question: Can I use factors in lattice input? Is it even possible to have XML markup in input word lattices?

Thanks!
J?rg



-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.mit.edu/mailman/private/moses-support/attachments/20150208/13b9c7ee/attachment-0001.htm

------------------------------

_______________________________________________
Moses-support mailing list
Moses-support@mit.edu
http://mailman.mit.edu/mailman/listinfo/moses-support


End of Moses-support Digest, Vol 100, Issue 30
**********************************************

Tweet