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. Re: Best translation path in the search graph (Philipp Koehn)
2. Re: Building and applying Recasing model from Moses MES
(Philipp Koehn)
3. Segfaulting with WordTranslationFeature (Alexandra Birch)
4. Call for participation: EAMT 2015 (Felipe S?nchez Mart?nez)
5. Re: Segfaulting with WordTranslationFeature (Matthias Huck)
6. Bilm train "IndexError: list index out of range" (Marwa Refaie)
----------------------------------------------------------------------
Message: 1
Date: Fri, 17 Apr 2015 15:05:06 -0400
From: Philipp Koehn <phi@jhu.edu>
Subject: Re: [Moses-support] Best translation path in the search graph
To: Abdelfetah Boumerdas <aa_boumerdas@esi.dz>
Cc: moses-support <moses-support@mit.edu>
Message-ID:
<CAAFADDBUtk6+C0Z+af8=zns7ExJNpQ=4fkNrTww=u+jkaw9BeA@mail.gmail.com>
Content-Type: text/plain; charset=UTF-8
Hi,
running the decoder with the trace option (-t) shows how the best
translation was constructed.
-phi
On Wed, Apr 15, 2015 at 6:44 PM, Abdelfetah Boumerdas
<aa_boumerdas@esi.dz> wrote:
> Hi everyone
> I'm following the moses manual and at the end of the verbose section in the
> phrase-based tutorial, it is mentioned that "at the end, the best scoring
> final
> hypothesis is found and the hypothesis graph traversed backwards to retrieve
> the best translation". My question is: is there an option in the decoder
> that gives the best path followed in the translation graph to arrive to the
> best translation?? and if yes what is it?? please it'll help me very much if
> i have an answer.
>
> Thanks in advance.
>
> --
> BOUMERDAS Abdelfetah
> 5?me Ann?e Option Syst?mes Informatiques (SIQ)
> Ecole nationale Sup?rieure d'Informatique ESI (ex INI)
> BP 68 M Oued Smar 16309 - ALGER
>
>
>
> _______________________________________________
> Moses-support mailing list
> Moses-support@mit.edu
> http://mailman.mit.edu/mailman/listinfo/moses-support
>
------------------------------
Message: 2
Date: Fri, 17 Apr 2015 15:06:56 -0400
From: Philipp Koehn <phi@jhu.edu>
Subject: Re: [Moses-support] Building and applying Recasing model from
Moses MES
To: "Hegde, Sujay" <Sujay.Hegde@xerox.com>
Cc: "moses-support@mit.edu" <moses-support@mit.edu>
Message-ID:
<CAAFADDCMB52BSOa_=VjKmALMcd+7Cn2UK8aNR3ACjCTstGrY7w@mail.gmail.com>
Content-Type: text/plain; charset=UTF-8
Hi,
you also need to point in the EVALUATION section to the use of the
recaser by removing the comment in the line:
#recaser = $moses-script-dir/recaser/recase.perl
-phi
On Thu, Apr 16, 2015 at 3:45 AM, Hegde, Sujay <Sujay.Hegde@xerox.com> wrote:
> Dear admin,
>
>
>
> I tried to build Recasing model from EMS by removing
> IGNORE on RECASING section and then adding a config
>
> trainer = $moses-script-dir/recaser/train-recaser.perl in that section.
>
> But when I run the EMS I do not see any recasing model created in the
> model folder.
>
>
>
> Not sure what is missed out.
>
> Kindly help.
>
>
>
> Regards,
>
> Sujay
>
>
> _______________________________________________
> Moses-support mailing list
> Moses-support@mit.edu
> http://mailman.mit.edu/mailman/listinfo/moses-support
>
------------------------------
Message: 3
Date: Fri, 17 Apr 2015 22:31:45 +0100
From: Alexandra Birch <lexi.birch@gmail.com>
Subject: [Moses-support] Segfaulting with WordTranslationFeature
To: "moses-support@mit.edu" <moses-support@mit.edu>
Message-ID:
<CA+h82t4YyK0jZUvOV-_N6z1Hh8uKaFQgJZ3e8RrOQfwY6BgDUg@mail.gmail.com>
Content-Type: text/plain; charset="utf-8"
Hi there,
I have a seg fault with a normal master branch of Moses from 1 month ago,
on a normal seeming test sentence. This was an en-cs system, and it
translated the first 6000+ sentences fine. It also translates a short
version of the sentence fine.
So
"Daniel , the previous owner"
translates fine but:
"Daniel , the previous owner , supported the author cinema on the complex
premises after having himself financed its construction"
segfaults!
If I remove the WordTranslation feature, so I delete:
< [feature]
< WordTranslationFeature name=WT input-factor=0 output-factor=0 simple=1
source-context=0 target-context=0
from the moses.ini file, then it stops segfaulting.
Any had this happen to them? Does anyone know how much this feature helps?
Lexi
--
------------------------------------------------------------------------------------------
School of Informatics
University of Edinburgh
Phone +44 (0)131 650-8286
--
The University of Edinburgh is a charitable body, registered in
Scotland, with registration number SC005336.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.mit.edu/mailman/private/moses-support/attachments/20150417/76741879/attachment-0001.htm
------------------------------
Message: 4
Date: Fri, 17 Apr 2015 23:48:18 +0200
From: Felipe S?nchez Mart?nez <fsanchez@dlsi.ua.es>
Subject: [Moses-support] Call for participation: EAMT 2015
To: mt-list@eamt.org, corpora@uib.no, elsnet-list@elsnet.org,
moses-support <moses-support@mit.edu>, apertium-stuff
<apertium-stuff@lists.sourceforge.net>
Message-ID: <55317FA2.9010709@dlsi.ua.es>
Content-Type: text/plain; charset=utf-8; format=flowed
---------------------------------------------------------------------
Call for Participation
18th Annual Conference of the European Association for Machine
Translation (EAMT 2015; Antalya, Turkey)
---------------------------------------------------------------------
The European Association for Machine Translation
(EAMT,http://www.eamt.org) invites everyone interested in machine
translation, translation-related tools and resources to attend to this
conference. The 18th Annual Conference of the European Association for
Machine Translation, will be held in Antalya, Turkey from 11 to 13 May
2015.
This year's programme includes oral and poster presentations of both
user papers and research papers, a poster session about project and
product descriptions, an invited talk by Olga Beregovaya, and an
exciting social programme.
The early registration deadline has been extended to April 22nd, and
there is an excellent package (including all-in food and drink for ?35
for the whole week) for those who stay in the 5-star conference
hotel.
The programme is available at
http://www.eamt2015.org/files/downloads/EAMT2015-conference-program.pdf
The list of accepted papers is available at
http://www.eamt2015.org/en/ACCEPTED-PAPERS.html
More information on the invited talk is available at
http://www.eamt2015.org/en/INVITED-SPEAKER.html
For more information, please visit the conference web page at
http://www.eamt2015.org/
We look forward to see you in Antalya.
Conference organisers
-------------------------------------------------------------
General Chair:
Andy Way (Dublin City University, Ireland)
Track Chairs:
Felipe S?nchez-Mart?nez - Research programme chair
(Universitat d?Alacant, Spain)
Gema Ram?rez-Sanchez - User programme co-chair
(Prompsit Language Engineering, Spain)
Fred Hollowood - User programme co-chair
(Fred Hollowood Consulting, Ireland)
Local Organisation Chair:
?lknur Durgar El-Kahlout (T?bitak, Turkey)
------------------------------
Message: 5
Date: Fri, 17 Apr 2015 23:00:27 +0100
From: Matthias Huck <mhuck@inf.ed.ac.uk>
Subject: Re: [Moses-support] Segfaulting with WordTranslationFeature
To: Alexandra Birch <lexi.birch@gmail.com>
Cc: "moses-support@mit.edu" <moses-support@mit.edu>
Message-ID: <1429308027.2310.11.camel@inf.ed.ac.uk>
Content-Type: text/plain; charset="UTF-8"
Hi Lexi,
The feature most likely won't be particularly important.
But this might be a completely different issue than you think. You
should debug this. Can you print the phrase pair that is applied when
the error occurs?
I recently came across a segfault that seemed to be caused by the OSM
feature. Then it turned out that a collision in the compact phrase table
entailed out-of-bounds word alignments. The solution (recommended by
Marcin) was to switch to modified parameters for phrase table
binarization, thereby avoiding the collision. The modified binarization
parameters are now default:
https://github.com/moses-smt/mosesdecoder/commit/506427368fbb9b980784ed55a68777be43896e8a
Cheers,
Matthias
On Fri, 2015-04-17 at 22:31 +0100, Alexandra Birch wrote:
> Hi there,
>
>
> I have a seg fault with a normal master branch of Moses from 1 month
> ago, on a normal seeming test sentence. This was an en-cs system, and
> it translated the first 6000+ sentences fine. It also translates a
> short version of the sentence fine.
>
>
> So
> "Daniel , the previous owner"
>
>
> translates fine but:
>
>
> "Daniel , the previous owner , supported the author cinema on the
> complex premises after having himself financed its construction"
>
>
>
> segfaults!
>
>
> If I remove the WordTranslation feature, so I delete:
>
>
> < [feature]
> < WordTranslationFeature name=WT input-factor=0 output-factor=0
> simple=1 source-context=0 target-context=0
>
>
>
> from the moses.ini file, then it stops segfaulting.
>
>
> Any had this happen to them? Does anyone know how much this feature
> helps?
>
>
> Lexi
>
>
>
>
>
>
> --
> ------------------------------------------------------------------------------------------
> School of Informatics
> University of Edinburgh
> Phone +44 (0)131 650-8286
>
> --
> The University of Edinburgh is a charitable body, registered in
> Scotland, with registration number SC005336.
>
> _______________________________________________
> Moses-support mailing list
> Moses-support@mit.edu
> http://mailman.mit.edu/mailman/listinfo/moses-support
--
The University of Edinburgh is a charitable body, registered in
Scotland, with registration number SC005336.
------------------------------
Message: 6
Date: Fri, 17 Apr 2015 22:48:29 +0000
From: Marwa Refaie <basmallah@hotmail.com>
Subject: [Moses-support] Bilm train "IndexError: list index out of
range"
To: "moses-support@mit.edu" <moses-support@mit.edu>
Message-ID: <DUB118-W37DA2758EB89BC899E5633BAE30@phx.gbl>
Content-Type: text/plain; charset="windows-1256"
Hi all,
I trained BiLM with English-Arabic corpus , & it give better Blue score. But when work with different bigger corpus I have this error !! target_aligns[tpos].append(spos)IndexError: list index out of range
I have file .ngrams which stopped after small no of lines by this error ??? any help plz ??
2015-04-18 00:42:19 INFO: Vocabulary size: 111432015-04-18 00:42:19 INFO: Reading vocabulary from ai2.true.ar....................2015-04-18 00:42:19 INFO: Vocabulary size: 162402015-04-18 00:42:19 INFO: Extracting ngramsTraceback (most recent call last): File "/home/marwa/mosesdecoder/scripts/training/bilingual-lm/extract_training.py", line 167, in <module> main() File "/home/marwa/mosesdecoder/scripts/training/bilingual-lm/extract_training.py", line 107, in main ofh) File "/home/marwa/mosesdecoder/scripts/training/bilingual-lm/extract.py", line 74, in get_ngrams target_aligns[tpos].append(spos)IndexError: list index out of range
Marwa N. Refaie
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.mit.edu/mailman/private/moses-support/attachments/20150417/260225f1/attachment.htm
------------------------------
_______________________________________________
Moses-support mailing list
Moses-support@mit.edu
http://mailman.mit.edu/mailman/listinfo/moses-support
End of Moses-support Digest, Vol 102, Issue 37
**********************************************
Subscribe to:
Post Comments (Atom)
0 Response to "Moses-support Digest, Vol 102, Issue 37"
Post a Comment