Moses-support Digest, Vol 87, Issue 18

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 Reference Sentence in Tunning Phase (Asad A.Malik)
2. Google Translate - technology & paradigms behind it (Ivan Dun?er)
3. Re: Google Translate - technology & paradigms behind it
(Philipp Koehn)
4. Re: Multiple Reference Sentence in Tunning Phase (Philipp Koehn)
5. OT Giza++ test old IBM models (Per Tunedal)


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

Message: 1
Date: Wed, 8 Jan 2014 12:18:11 -0800 (PST)
From: "Asad A.Malik" <asad_12204@yahoo.com>
Subject: [Moses-support] Multiple Reference Sentence in Tunning Phase
To: "moses-support@mit.edu" <moses-support@mit.edu>
Message-ID:
<1389212291.34211.YahooMailNeo@web122201.mail.ne1.yahoo.com>
Content-Type: text/plain; charset="iso-8859-1"

Hi All,


I am trying to develop Urdu to English Translation system. I am currently in tunning phase andhaving multiple reference sentence for English. I've entered following commad:


nohup nice ~/work/mosesdecoder/scripts/training/mert-moses.pl ~/work/corpus/tunning.ur-en.true.ur ~/work/corpus/tunning.ur-en.true.en ~/work/mosesdecoder/bin/moses train/model/moses.ini --mertdir
~/work/mosesdecoder/bin/ --decoder-flags "-threads 2" &> mert.out &

?
but it isn't starting.


PS. tunning.ur-en.true.en are 4 files tunning.ur-en.true.en.0 , tunning.ur-en.true.en.1 , tunning.ur-en.true.en.2 ,tunning.ur-en.true.en.3


Regards

Asad A.Malik
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.mit.edu/mailman/private/moses-support/attachments/20140108/00329c8f/attachment-0001.htm

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

Message: 2
Date: Wed, 8 Jan 2014 21:47:43 +0100
From: Ivan Dun?er <ivandunder@gmail.com>
Subject: [Moses-support] Google Translate - technology & paradigms
behind it
To: moses-support@mit.edu
Message-ID:
<CAP1sQ+aEFJBXq166GG3hNY5UD222pdmttX7jq5OciNpz-BMEUw@mail.gmail.com>
Content-Type: text/plain; charset=ISO-8859-1

Hi everybody,

is there any reading material (e.g. scientific articles etc.) on what
MT approaches Google Translate uses?
Sorry for off-topic.

Regards,
Ivan


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

Message: 3
Date: Wed, 8 Jan 2014 21:42:54 +0000
From: Philipp Koehn <pkoehn@inf.ed.ac.uk>
Subject: Re: [Moses-support] Google Translate - technology & paradigms
behind it
To: Ivan Dun?er <ivandunder@gmail.com>
Cc: "moses-support@mit.edu" <moses-support@mit.edu>
Message-ID:
<CAAFADDCZF48bpU3rq=x0ZcLZe7AjmK-rQZ=tnsWmChCCXfr1rQ@mail.gmail.com>
Content-Type: text/plain; charset="utf-8"

Hi,

Google lists its research papers here:
http://research.google.com/pubs/MachineTranslation.html

They are generally very secretive about the specifics of their
system, but the impression is that they do something very
similar to Moses, i.e., mostly phrase-based, for some language
pairs maybe hierarchical, with pre-reordering. A key difference
is a distributed implementation that allows for very fast response
time and storage of very large models in the memory of
several machines.

-phi



On Wed, Jan 8, 2014 at 8:47 PM, Ivan Dun?er <ivandunder@gmail.com> wrote:

> Hi everybody,
>
> is there any reading material (e.g. scientific articles etc.) on what
> MT approaches Google Translate uses?
> Sorry for off-topic.
>
> Regards,
> Ivan
> _______________________________________________
> Moses-support mailing list
> Moses-support@mit.edu
> http://mailman.mit.edu/mailman/listinfo/moses-support
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.mit.edu/mailman/private/moses-support/attachments/20140108/f4bb1e62/attachment-0001.htm

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

Message: 4
Date: Wed, 8 Jan 2014 21:47:05 +0000
From: Philipp Koehn <pkoehn@inf.ed.ac.uk>
Subject: Re: [Moses-support] Multiple Reference Sentence in Tunning
Phase
To: "Asad A.Malik" <asad_12204@yahoo.com>
Cc: "moses-support@mit.edu" <moses-support@mit.edu>
Message-ID:
<CAAFADDAYU-K3ZY0fj-zFtspssoBJPLQVsFHnVX-8YRrB9EczqQ@mail.gmail.com>
Content-Type: text/plain; charset="iso-8859-1"

Hi,

try "~/work/corpus/tunning.ur-en.true.en." instead of
"~/work/corpus/tunning.ur-en.true.en"
Note the added ".".

-phi


On Wed, Jan 8, 2014 at 8:18 PM, Asad A.Malik <asad_12204@yahoo.com> wrote:

> Hi All,
>
> I am trying to develop Urdu to English Translation system. I am currently
> in tunning phase and having multiple reference sentence for English. I've
> entered following commad:
>
> *nohup nice ~/work/mosesdecoder/scripts/training/mert-moses.pl
> <http://mert-moses.pl> ~/work/corpus/tunning.ur-en.true.ur
> ~/work/corpus/tunning.ur-en.true.en ~/work/mosesdecoder/bin/moses
> train/model/moses.ini --mertdir ~/work/mosesdecoder/bin/ --decoder-flags
> "-threads 2" &> mert.out &*
>
>
> but it isn't starting.
>
> PS. tunning.ur-en.true.en are 4 files tunning.ur-en.true.en.0 , tunning.ur-en.true.en.1
> , tunning.ur-en.true.en.2 ,tunning.ur-en.true.en.3
>
>
> Regards
>
> Asad A.Malik
>
> _______________________________________________
> Moses-support mailing list
> Moses-support@mit.edu
> http://mailman.mit.edu/mailman/listinfo/moses-support
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.mit.edu/mailman/private/moses-support/attachments/20140108/384616e2/attachment-0001.htm

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

Message: 5
Date: Thu, 09 Jan 2014 10:05:21 +0100
From: Per Tunedal <per.tunedal@operamail.com>
Subject: [Moses-support] OT Giza++ test old IBM models
To: moses-support@mit.edu
Message-ID:
<1389258321.15541.68503273.38936899@webmail.messagingengine.com>
Content-Type: text/plain

Hi,
where can I find instructions how to run the old IBM models with Giza++?

I would like to try e.g. IBM model 2 to get a deeper understanding of
the basics of MT. How to do I get a translation model according to IBM
model 2 with Giza++ ?

Yours,
Per Tunedal


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

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


End of Moses-support Digest, Vol 87, Issue 18
*********************************************

Tweet