The script seems to have been forked just before the last changes

https://github.com/moses-smt/mosesdecoder/commit/93433cf0157ecadd19162929a077aa6179dc0d8b

If someone who uses SGE can do a merge with the latest version in github
and test that it works, I can check it in



On 2 December 2013 09:03, Thomas Meyer <thomas.me...@idiap.ch> wrote:

>  Hi all,
>
> Attached is moses-parallel.pl with some modifications (marked #thm:) to
> make it work with the SGE architecture we have here. I just ran a test
> whether it outputs the word alignment and all works normally.
>
> Here is the command I used:
>
> ./moses-parallel.pl -jobs 2 -alignment-output-file alignment.txt
> -input-file /temp/tmeyer/test.en -decoder
> /temp/tmeyer/mosesdecoder/bin/moses -config
> /temp/tmeyer/en_de/test/moses.tuned.ini > test.txt
>
> which outputs, in alignment.txt:
>
> 0-0 1-1 2-2 3-3 4-4 5-5 6-6 7-7 8-8 9-9 10-10 11-11 12-12 13-13 14-14
> 15-15 16-16 17-17 18-18 19-19 20-20 21-21 22-22 23-23 24-24
> 0-0 1-1 2-2 3-3 4-4 5-4 6-5 7-6 8-7 9-8 10-9 11-10 12-11 13-12 13-13 14-14
> 16-15 17-16 18-17 20-18 21-19 22-20 23-21 24-22 25-23 26-24 27-25 28-26
> 29-27 30-28 31-28 32-29 33-30 34-31 36-32 37-33 38-34 39-35
>
> etc.
>
> Hope it will work for you, too, Konstantinos.
>
> Best,
> Thomas
>
>
>
> On 29/11/13 21:59, Philipp Koehn wrote:
>
> Hi Thomas,
>
> in that case - can you share your modifications if they are of general
> interest and check if there is a problem with the word alignment reporting?
>
> -phi
>
> On Fri, Nov 29, 2013 at 6:52 PM, Thomas Meyer <thomas.me...@idiap.ch> 
> <thomas.me...@idiap.ch> wrote:
>
>  Hi all,
>
> I 'still' do and find it quite convenient. At least with release 1.0 and
> some modifications to moses-parallel.pl
>
> Thomas
>
>
> From: CHATZITHEODOROU Konstantinos
> Sent: Friday, 29 November 2013 7:48 PM
> To: Hieu Hoang
> Cc: moses-support@mit.edu
> Subject: Re: [Moses-support] Word-to-word alignment when moses-parallel.pl
> is used
>
> Hi Hieu,
>
> Thanks for your reply. I have a question. Why not many people still use SGE
> to run Moses?
>
> Thank you in advance.
>
> Best,
> Konstantinos
>
>
> 2013/11/29 Hieu Hoang <hieu.ho...@ed.ac.uk> <hieu.ho...@ed.ac.uk>
>
>  I'm not sure if many people still uses Sun Grid Engine to run Moses. If
> not, scripts like moses-parallel.pl may become outdated.
>
> I know at Edinburgh, we no longer use SGE. Instead we run our experiment
> on 1 large multi-core machine. Most of moses has been re-engineered to be
> multithreaded to take advantage of multicores
>
>
> On 28 November 2013 10:25, CHATZITHEODOROU Konstantinos<dinosa...@gmail.com> 
> <dinosa...@gmail.com> wrote:
>
>  Hi Philipp,
>
> Thank you your your prompt reply. I have tried the option
> "alignment-output-file" for moses-parallel.pl but didn't use it as part of
> the decoder setting. But it didn't work. Each individual sub-job finished
> with an error status stating something the sub-job is untranslated.
>
> Thank you for your time and consideration.
>
> Regards,
> Konstantinos
>
>
>
> 2013/11/27 Philipp Koehn <pko...@inf.ed.ac.uk> <pko...@inf.ed.ac.uk>
>
>  Hi,
>
> the script has an option "alignment-output-file" - did you try this?
>
> -phi
>
> On Tue, Nov 26, 2013 at 8:13 PM, CHATZITHEODOROU 
> Konstantinos<dinosa...@gmail.com> <dinosa...@gmail.com> wrote:
>
>  Hi all,
>
> Could you please give an example on how to output the word-to-word
> alignment
> during decoding when moses-parallel.pl is used?
>
> Thank you in advance for your help.
>
> Regards,
> Konstantinos
>
>
> _______________________________________________
> Moses-support mailing 
> listMoses-support@mit.eduhttp://mailman.mit.edu/mailman/listinfo/moses-support
>
>
>
>
> --
> CHATZITHEODOROU Konstantinos
>
>
> _______________________________________________
> Moses-support mailing 
> listMoses-support@mit.eduhttp://mailman.mit.edu/mailman/listinfo/moses-support
>
>
>
> --
> Hieu Hoang
> Research Associate
> University of Edinburghhttp://www.hoang.co.uk/hieu
>
>
>
> --
> CHATZITHEODOROU Konstantinos
>
>
>
> _______________________________________________
> Moses-support mailing 
> listMoses-support@mit.eduhttp://mailman.mit.edu/mailman/listinfo/moses-support
>
> _______________________________________________
> Moses-support mailing 
> listMoses-support@mit.eduhttp://mailman.mit.edu/mailman/listinfo/moses-support
>
>
>


-- 
Hieu Hoang
Research Associate
University of Edinburgh
http://www.hoang.co.uk/hieu
_______________________________________________
Moses-support mailing list
Moses-support@mit.edu
http://mailman.mit.edu/mailman/listinfo/moses-support

Reply via email to