Moses-support Digest, Vol 85, Issue 24

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. Just running Model1 using giza++ (rohit dholakia)


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

Message: 1
Date: Fri, 8 Nov 2013 12:45:59 -0800
From: rohit dholakia <rdholaki@sfu.ca>
Subject: [Moses-support] Just running Model1 using giza++
To: moses-support@mit.edu
Message-ID:
<CAA==LguVVx4fWOU_z9=koKLjM7kstYd=A4GhVVEgvj_nxOs9pw@mail.gmail.com>
Content-Type: text/plain; charset="iso-8859-1"

Hi,

If I just run model1 using --giza-option
m1=5,m2=0,m3=0,m4=0,t1=5,nodumps=1 and take a look at the src-tgt.A1.5
file, or the tgt-src.A1.5 file, this is what I see

# Sentence pair (1) source length 3 target length 1 *alignment score :
0.134457*
!
NULL ({ }) ! ({ 1 }) ( ({ }) source ({ })



I was wondering what that _alignment score_ represents in this case,
because without the advanced models, there cannot be a viterbi alignment
score. I checked the training.out file and it only is running Model 1, 5
iterations. Is the alignment score = *(1/(l+1)**m) * max(t(f|e))* ?

thanks !
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.mit.edu/mailman/private/moses-support/attachments/20131108/d23b4ae9/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 85, Issue 24
*********************************************

Tweet