Moses-support Digest, Vol 88, Issue 52

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: ERROR: Lexical reordering scoring failed at
moses/scripts/training/train-model.perl line 1738. (Barry Haddow)
2. Re: publication for "mert-moses.pl --pro-starting-point " ?
(Philipp Koehn)
3. Constraint decoding (Saeed Farzi)
4. Moses Compilation error (priyanka)


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

Message: 1
Date: Mon, 24 Feb 2014 17:46:38 +0000
From: Barry Haddow <bhaddow@staffmail.ed.ac.uk>
Subject: Re: [Moses-support] ERROR: Lexical reordering scoring failed
at moses/scripts/training/train-model.perl line 1738.
To: =?GB2312?B?wO673bvd?= <hui.li@informatik.uni-heidelberg.de>,
moses-support@mit.edu
Message-ID: <530B857E.3080704@staffmail.ed.ac.uk>
Content-Type: text/plain; charset="gb2312"

Hi Hui

I have a vague memory that very short corpora cause problems ...

If you edit scripts/generic/extract-parallel.perl to comment out the
following line (line 223 in my version):

$cmd = "rm -rf $TMPDIR \n";

then rerun, it should retain the tmp directory created during
extraction. You could then rerun extract manually and look for the
problem. If you don't see anything odd, then upload to pastebin of
dropbox and send out the link - or just email if it's small.

cheers - Barry


On 24/02/14 16:00, ?????? wrote:
> Hi Barry,
> I tried to use absolute paths like this:
>
> hli@themisto:~$ /home/hli/MT/moses/scripts/training/train-model.perl
> /home/hli/MT -corpus /home/hli/MT/lm/1545.test.lowercased
> -external-bin-dir /home/hli/MT/bin -e de -f ade -alignment
> grow-diag-final-and -reordering msd-bidirectional-fe -lm
> 0:3:/home/hli/MT/lm/1545.test.lm >& training.out &
>
> but the same error still exists.
>
> Is that anything wrong with my corpus?
>
> Best,
> Hui
>
>
> 2014-02-24 16:48 GMT+01:00 Barry Haddow <bhaddow@staffmail.ed.ac.uk
> <mailto:bhaddow@staffmail.ed.ac.uk>>:
>
> Hi Hui
>
> It looks like your extract failed, but there is no error message.
> Do you think it ran out of disk? Could you try again, but use
> absolute paths everywhere?
>
> cheers - Barry
>
>
> On 24/02/14 09:44, ?????? wrote:
>
>
>
> Hi Barry,
> attached is my log of phrase model training. Thx a lot.
>
> Best,
> Hui
>
>
>
> 2014-02-24 10:18 GMT+01:00 Barry Haddow
> <bhaddow@staffmail.ed.ac.uk
> <mailto:bhaddow@staffmail.ed.ac.uk>
> <mailto:bhaddow@staffmail.ed.ac.uk
> <mailto:bhaddow@staffmail.ed.ac.uk>>>:
>
>
> Hi Hui
>
> Something went wrong earlier in training which meant that
> extract.o.sorted.gz was not created. You need to attach the log of
> train-model,
>
> cheers - Barry
>
>
> On 24/02/14 08:38, ?????? wrote:
>
>
>
>
> Hi all,
>
> when I was training the phrase model,
> hli@themisto:~/MT$ moses/scripts/training/train-model.perl
> home/hli/MT -corpus lm/1545.test.lowercased -external-bin-dir
> bin -e de -f ade -alignment grow-diag-final-and -reordering
> msd-bidirectional-fe -lm 0:3:/home/hli/MT/lm/1545.test.lm
>
> I obtained this error:
>
> scores lexical reordering models of several types
> (hierarchical, phrase-based and word-based-extraction
> terminate called after throwing an instance of
> 'util::ErrnoException'
> what(): util/file.cc:67 in int util::OpenReadOrThrow(const
> char*) threw ErrnoException because `-1 == (ret = open(name,
> 00))'.
> No such file or directory while opening
> ./model/extract.o.sorted.gz
> Aborted (core dumped)
> Exit code: 134
> ERROR: Lexical reordering scoring failed at
> moses/scripts/training/train-model.perl line 1738.
>
> Attached is my corpus, could you help me to find out what's
> wrong with it ?
>
> Thx a lot!
>
> Best Regards,
> Hui Li
> Institute of Computer Science
> Heidelberg
>
> *
> *
>
>
>
>
>
> _______________________________________________
> Moses-support mailing list
> Moses-support@mit.edu <mailto:Moses-support@mit.edu>
> <mailto:Moses-support@mit.edu <mailto: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 <mailto: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.
>
>

-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: not available
Url: http://mailman.mit.edu/mailman/private/moses-support/attachments/20140224/6115d567/attachment-0001.bat

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

Message: 2
Date: Mon, 24 Feb 2014 14:18:58 -0500
From: Philipp Koehn <pkoehn@inf.ed.ac.uk>
Subject: Re: [Moses-support] publication for "mert-moses.pl
--pro-starting-point " ?
To: Marcin Junczys-Dowmunt <junczys@amu.edu.pl>
Cc: moses-support <moses-support@mit.edu>, Barry Haddow
<bhaddow@staffmail.ed.ac.uk>
Message-ID:
<CAAFADDCLL39==DkuGs8HQygvThOuCZgsOuXZktD0z-o+GOighQ@mail.gmail.com>
Content-Type: text/plain; charset=ISO-8859-1

Hi,

in the experiments that I did, I found that kbmira is a little bit more stable
(i.e., less variance between tuning runs) than MERT. It runs pretty much
always to the maximum of 25 iterations.

-phi

On Mon, Feb 24, 2014 at 10:01 AM, Marcin Junczys-Dowmunt
<junczys@amu.edu.pl> wrote:
> Thanks,
> what's your general experience with pro/kbmira and BLEU. For my metric,
> it seems both methods do not really converge too well, although they
> find improvements from time to time. Do you have a similar experience
> with BLEU? If yes, do you let the optimization hit the 25 iterations limit?
> Best,
> Marcin
>
> W dniu 23.02.2014 17:55, Barry Haddow pisze:
>> Hi Marcin
>>
>> It's mentioned here:
>> http://www.aclweb.org/anthology/W/W12/W12-3139.pdf
>> (section 4.2) but it's not something we spent much time experimenting
>> with,
>>
>> cheers - Barry
>>
>> On 23/02/14 15:12, Marcin Junczys-Dowmunt wrote:
>>> Hi,
>>> Is there a publication that describes the idea behind
>>> "--pro-starting-point" option in mert-moses.pl? This seems to be working
>>> really well for me.
>>> Best,
>>> Marcin
>>> _______________________________________________
>>> Moses-support mailing list
>>> Moses-support@mit.edu
>>> http://mailman.mit.edu/mailman/listinfo/moses-support
>>>
>>
>
> _______________________________________________
> Moses-support mailing list
> Moses-support@mit.edu
> http://mailman.mit.edu/mailman/listinfo/moses-support


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

Message: 3
Date: Tue, 25 Feb 2014 10:20:36 +0330
From: Saeed Farzi <saeedfarzi@gmail.com>
Subject: [Moses-support] Constraint decoding
To: moses-support <moses-support@mit.edu>
Message-ID:
<CAOhrzg7wDLpcxdQo7WuJHBABOEGakhS3kFc7L6FZRdiYgiRctQ@mail.gmail.com>
Content-Type: text/plain; charset="iso-8859-1"

Dear all,

I am trying to use constraint decoding with moses, Any body knows how do
this?

By the way, are there differences between constraint decoding and forced
decoding?


Any guidance would be highly appreciated.
cheers
--
S.Farzi, Ph.D. Student
Natural Language Processing Lab,
School of Electrical and Computer Eng.,
Tehran University
Tel: +9821-6111-9719
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.mit.edu/mailman/private/moses-support/attachments/20140225/dad69af3/attachment-0001.htm

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

Message: 4
Date: Tue, 25 Feb 2014 13:06:38 +0530
From: priyanka <priyanka@cse.iitb.ac.in>
Subject: [Moses-support] Moses Compilation error
To: <moses-support@mit.edu>
Message-ID: <ab022c19b915b27b9c79ea784c073d83@cse.iitb.ac.in>
Content-Type: text/plain; charset="utf-8"

Hello everyone,

I am getting gcc.link error while compiling the moses. I tried
compiling with boost_1_51_0, as suggested in the thread
http://comments.gmane.org/gmane.comp.nlp.moses.user/8140
Still I am getting the same error. I have attached the error log with
the mail.
My system is Ubuntu 12.04 64 bit
Command used for moses compilation:
/usr/bin/bjam -a --with-srilm=/home/piyu/MTech/MTech1-Sem2/tools/srilm
--with-giza=/home/piyu/MTech/MTech1-Sem2/tools/bin
--with-boost=/usr/include/boost link=shared -j2

Will really appreciate your help.

Thanks,
Priyanka
MTech1, IIT Bombay
-------------- next part --------------
A non-text attachment was scrubbed...
Name: build.log.zip
Type: application/zip
Size: 33585 bytes
Desc: not available
Url : http://mailman.mit.edu/mailman/private/moses-support/attachments/20140225/efe18a61/attachment.zip

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

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


End of Moses-support Digest, Vol 88, Issue 52
*********************************************

Tweet