Moses-support Digest, Vol 86, Issue 33

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. (no subject) (Kalyani Baruah)
2. single-input-sentence-problem-EMS (nadeem khan)
3. Re: Increasing context scope during training (Dimitris Mavroeidis)
4. Re: (no subject) (Philipp Koehn)
5. Re: Increasing context scope during training (R?dolfs Mazurs)
6. Re: EMS-Decoder-Problem (Barry Haddow)


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

Message: 1
Date: Tue, 10 Dec 2013 16:57:46 +0530
From: Kalyani Baruah <kajubaruah04@gmail.com>
Subject: [Moses-support] (no subject)
To: "moses-support@mit.edu" <moses-support@mit.edu>
Message-ID:
<CAJZ5LDcabt9oDV1BFK3gAA4NCic3EkJ0EX7pa5MXY6DGVkXPpg@mail.gmail.com>
Content-Type: text/plain; charset="iso-8859-1"

hii
Can you provide me with a ppt(power point presntation ) regarding
statistical translation using moses toolkit






Regards,


*Kalyanee Kanchan Baruah*
Department of Information Technology,
Institute of Science and Technology,
Gauhati University,Guwahati,India
Phone- +91-9706242124
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.mit.edu/mailman/private/moses-support/attachments/20131210/3f456f46/attachment-0001.htm

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

Message: 2
Date: Tue, 10 Dec 2013 04:06:59 -0800 (PST)
From: nadeem khan <nad_star06@yahoo.com>
Subject: [Moses-support] single-input-sentence-problem-EMS
To: "moses-support@mit.edu" <moses-support@mit.edu>
Message-ID:
<1386677219.3424.YahooMailNeo@web162406.mail.bf1.yahoo.com>
Content-Type: text/plain; charset="us-ascii"

I ran EMS on config.toy with toydata successfully all runs just fine there was not a single error but when it comes to my own data of hindi and experiment I am getting stuck with BLEU and BLEU-c Crashed and in input file produced by EMS have only 1 single sentence from my above 900 sentences.
When I invistaged the problem there is only 1 single Input Segment in test.input.tc.1. why and how the EMS taking only 1 segment from my input test-src.sgm file? and when I investigated further there is a fatal error under EVALUATION_test_nist-bleu-c.1.STDERR of no id in srcset. why I am getting that???? as I am giving it the complete sgm frame for wrapping out the output.
Help out please
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.mit.edu/mailman/private/moses-support/attachments/20131210/be3becb7/attachment-0001.htm

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

Message: 3
Date: Tue, 10 Dec 2013 15:12:56 +0200
From: Dimitris Mavroeidis <dmavroeidis@csri.gr>
Subject: Re: [Moses-support] Increasing context scope during training
To: R?dolfs Mazurs <rudolfs.mazurs@gmail.com>, "moses-support@mit.edu"
<moses-support@mit.edu>
Message-ID: <52A71358.2000702@csri.gr>
Content-Type: text/plain; charset=UTF-8; format=flowed

Dear R?dolfs,

You must be referring to the language model's n-gram size. If you are
using EMS, then you can set "order" in the "LM" portion of the
configuration file.

Setting a higher n-gram order (not more than 5) usually helps, but that
depends on various factors, especially the target language, the size of
your corpus, etc. Just give it a try and see what order gives the best
results for your situation.

Best regards,
Dimitris

On 09/12/2013 11:21 ??, R?dolfs Mazurs wrote:
> Hi all,
>
> I am looking to improve quality of translation on my limited corpus.
> During training process I noticed that ngrams only go up to 3. Is there
> a way to increase the upper limit on ngram count? And is there a chance
> it would improve results of translations?
>



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

Message: 4
Date: Tue, 10 Dec 2013 15:02:15 +0000
From: Philipp Koehn <pkoehn@inf.ed.ac.uk>
Subject: Re: [Moses-support] (no subject)
To: Kalyani Baruah <kajubaruah04@gmail.com>
Cc: "moses-support@mit.edu" <moses-support@mit.edu>
Message-ID:
<CAAFADDAScQKtYCeTE2X3iBvE=yEH1Et6Wv3sV=PZc5_SrhtbmA@mail.gmail.com>
Content-Type: text/plain; charset=ISO-8859-1

Hi,

TAUS put together a basic slide presentation:
https://www.taus.net/press-releases/free-open-source-machine-translation-tutorial-is-made-available-by-taus

-phi

On Tue, Dec 10, 2013 at 11:27 AM, Kalyani Baruah <kajubaruah04@gmail.com> wrote:
> hii
> Can you provide me with a ppt(power point presntation ) regarding
> statistical translation using moses toolkit
>
>
>
>
>
>
> Regards,
>
>
> Kalyanee Kanchan Baruah
> Department of Information Technology,
> Institute of Science and Technology,
> Gauhati University,Guwahati,India
> Phone- +91-9706242124
>
> _______________________________________________
> Moses-support mailing list
> Moses-support@mit.edu
> http://mailman.mit.edu/mailman/listinfo/moses-support
>


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

Message: 5
Date: Tue, 10 Dec 2013 17:07:19 +0200
From: R?dolfs Mazurs <rudolfs.mazurs@gmail.com>
Subject: Re: [Moses-support] Increasing context scope during training
To: Dimitris Mavroeidis <dmavroeidis@csri.gr>
Cc: "moses-support@mit.edu" <moses-support@mit.edu>
Message-ID: <1386688039.7339.4.camel@steps>
Content-Type: text/plain; charset="UTF-8"

Thanks for the pointer, Dimitri!

Although I don't use EMS, I guess that script irstlm/bin/build-lm.sh is
responsible for LM part and the option is
-n Order of language model (default 3)

Thanks again!

On O , 2013-12-10 at 15:12 +0200, Dimitris Mavroeidis wrote:
> Dear R?dolfs,
>
> You must be referring to the language model's n-gram size. If you are
> using EMS, then you can set "order" in the "LM" portion of the
> configuration file.
>
> Setting a higher n-gram order (not more than 5) usually helps, but that
> depends on various factors, especially the target language, the size of
> your corpus, etc. Just give it a try and see what order gives the best
> results for your situation.
>
> Best regards,
> Dimitris
>
> On 09/12/2013 11:21 ??, R?dolfs Mazurs wrote:
> > Hi all,
> >
> > I am looking to improve quality of translation on my limited corpus.
> > During training process I noticed that ngrams only go up to 3. Is there
> > a way to increase the upper limit on ngram count? And is there a chance
> > it would improve results of translations?
> >
>




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

Message: 6
Date: Tue, 10 Dec 2013 16:51:53 +0000
From: Barry Haddow <bhaddow@staffmail.ed.ac.uk>
Subject: Re: [Moses-support] EMS-Decoder-Problem
To: nadeem khan <nad_star06@yahoo.com>
Cc: "moses-support@mit.edu" <moses-support@mit.edu>
Message-ID: <52A746A9.9010807@staffmail.ed.ac.uk>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed

Hi Nadeem

It looks like something went wrong earlier in the EVALUATION section,
possibly in the input-from-sgm step. I would check all the steps in this
section for errors.

It is also not clear to me that the truecaser will work with Hindi as it
is designed for languages written in the latin script,

cheers - Barry

On 07/12/13 18:51, nadeem khan wrote:
>
>
>
> Hello Sir;
>
> I am using EMS now and getting into a problem with my data of hindi
> language.
> I ran EMS on config.toy just fine there was not a single error but
> when it comes to my own data and experiment I am getting stuck with
> BLEU and BLEU-c Crashed.
> When I invistaged the problem there is only 1 single Input Segment in
> test.input.tc.1. why and how the EMS taking only 1 segment from my
> input test-src.sgm file? and when I investigated further there is a
> fatal error under EVALUATION_test_nist-bleu-c.1.STDERR of no id in
> srcset. why I am getting that???? as I am giving it the complete sgm
> frame for wrapping out the output.
>
> I am sending you my those testdata sgm file as well as the input and
> output generated by EMS for my dataset.
> Please have a look at it and Reply with your kind comments to resolve
> these issues
> Waiting for your kind response
>
> THANK YOU
> Regards
> nadeem
>
>
>
>
> _______________________________________________
> 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.



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

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


End of Moses-support Digest, Vol 86, Issue 33
*********************************************

Tweet