Moses-support Digest, Vol 84, Issue 6

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: Issues in LM (Hieu Hoang)
2. Re: Mosesserver support hierarchical model? (Barry Haddow)
3. lattices with EPSILON (Hieu Hoang)


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

Message: 1
Date: Thu, 03 Oct 2013 20:16:47 +0100
From: Hieu Hoang <hieuhoang@gmail.com>
Subject: Re: [Moses-support] Issues in LM
To: moses-support@mit.edu
Message-ID: <524DC29F.4060700@gmail.com>
Content-Type: text/plain; charset="iso-8859-1"

hi karan

can you please provide the exact command you ran, and the config file
and moses.ini file you used.

i'll take a look at it & get back to you

On 02/10/2013 08:14, karan sharma wrote:
> Hey,
> I am getting a issue in LM.The error is:
>
> ERROR : number of defined LM sets (1:toy) and LM files (0:) does not
> match at experiment.perl line 2211
>
> Could anyone please provide any assist on it .
>
> Regards
>
>
> _______________________________________________
> 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/20131003/ac6216c2/attachment-0001.htm

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

Message: 2
Date: Fri, 04 Oct 2013 09:07:22 +0100
From: Barry Haddow <bhaddow@staffmail.ed.ac.uk>
Subject: Re: [Moses-support] Mosesserver support hierarchical model?
To: Lingxiao WANG <lw@similis.org>, moses-support
<moses-support@mit.edu>
Message-ID: <524E773A.5050909@staffmail.ed.ac.uk>
Content-Type: text/plain; charset="utf-8"

Hi Lingxiao

The error message shows what your problem is.
> ERROR:Language model type unknown. Probably not compiled into library
> ERROR:no LM created. We probably don't have it compiled
Your config file specifies language model type 1 (IRSTLM), but you
haven't compiled with IRSTLM. Either update your config file, or
recompile moses with '--with-irstlm=/path/to/irstlm'

cheers - Barry

On 03/10/13 14:47, Lingxiao WANG wrote:
> Hi Barry
>
> I use this command :
> nohup nice $MOSES_BIN/mosesserver -xml-input exclusive -config
> $TABLE_PATH/moses.ini --server-port 8001 --server-log
> $TABLE_PATH/requests.log > $TABLE_PATH/output.log &
>
> This is my log:
>
> Defined parameters (per moses.ini or switch):
> config: /home/marwen/GENERIC-FR-ZH/moses.ini.1
> cube-pruning-pop-limit: 1000
> input-factors: 0
> inputtype: 3
> lmodel-file: 1 0 5 /home/lingxiao/GENERIC-FR-ZH/fr-zh.binlm.1
> mapping: 0 T 0 1 T 1
> max-chart-span: 20 1000
> non-terminals: X
> search-algorithm: 3
> ttable-file: 6 0 0 5 /home/lingxiao/GENERIC-FR-ZH/phrase-table.1 6 0 0 1 /home/lingxiao/GENERIC-FR-ZH/glue-grammar.1
> ttable-limit: 20
> weight-l: 0.200959
> weight-t: 0.0380514 0.0274366 0.0451763 0.0118391 -0.160113 0.0155388
> weight-w: -0.500886
> xml-input: exclusive
> /home/lingxiao/Moses-Decoder/bin
> Loading lexical distortion models...have 0 models
> Start loading LanguageModel /home/lingxiao/GENERIC-FR-ZH/fr-zh.binlm.1 : [0.000304143] seconds
> ERROR:Language model type unknown. Probably not compiled into library
> ERROR:no LM created. We probably don't have it compiled
> Mosesserver couldn't be started.
>
>
> LX
>
>
>
> Le 1 oct. 2013 ? 22:54, Barry Haddow a ?crit :
>
>> What happens when you try it with a hierarchical model? Could you
>> give the full command you are running and the output. It *should* work.
>>
>> On 01/10/13 21:50, Similes wrote:
>>> I have tried this one, like Hieu Hoang said, only for phrase based. : (
>>>
>>> ? 1 oct. 2013?22:34?Barry Haddow <bhaddow@staffmail.ed.ac.uk
>>> <mailto:bhaddow@staffmail.ed.ac.uk>> ???
>>>
>>>> Try this
>>>> http://www.statmt.org/moses/?n=Moses.AdvancedFeatures#ntoc28
>>>>
>>>> On 01/10/13 21:31, Lingxiao WANG wrote:
>>>>>
>>>>>
>>>>> ? 1 oct. 2013?22:23?Barry Haddow <bhaddow@staffmail.ed.ac.uk
>>>>> <mailto:bhaddow@staffmail.ed.ac.uk>> ???
>>>>>
>>>>>>> But it's unlucky... It doesn't work with mosesserver : (
>>>>>>
>>>>>> As far as I know, it does work.
>>>>> It sounds a good news! Do you have some idea?
>>>>>
>>>>>
>>>>>
>>>>>> On 01/10/13 21:13, Similes wrote:
>>>>>>> Hi
>>>>>>>
>>>>>>> Not yet, I want to know that do you have some suggestions for that?
>>>>>>> Because when we take about Chinese moses, we get the answer is
>>>>>>> "Hierarchical model".
>>>>>>> But it's unlucky... It doesn't work with mosesserver : (
>>>>>>>
>>>>>>> Regards
>>>>>>>
>>>>>>> Lingxiao
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> ? 1 oct. 2013?21:49?Barry Haddow <bhaddow@staffmail.ed.ac.uk
>>>>>>> <mailto:bhaddow@staffmail.ed.ac.uk>> ???
>>>>>>>
>>>>>>>> Hi Lingziao
>>>>>>>>
>>>>>>>> Mosesserver should work with hierarchical models. Did you try
>>>>>>>> it? Was there a problem?
>>>>>>>>
>>>>>>>> cheers - Barry
>>>>>>>>
>>>>>>>> On 01/10/13 15:06, Hieu Hoang wrote:
>>>>>>>>> i think only phrase-based model are supported by the mosesserver.
>>>>>>>>>
>>>>>>>>> If you are able to extend it to hiero model, please send me
>>>>>>>>> the code and i'll check it in
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> On 27 September 2013 22:13, Similes <lw@similis.org
>>>>>>>>> <mailto:lw@similis.org>> wrote:
>>>>>>>>>
>>>>>>>>> Hi, all
>>>>>>>>>
>>>>>>>>> How could we let mosesserver to support hierarchical model?
>>>>>>>>>
>>>>>>>>> Best regards
>>>>>>>>>
>>>>>>>>> Lingxiao
>>>>>>>>> _______________________________________________
>>>>>>>>> Moses-support mailing list
>>>>>>>>> Moses-support@mit.edu <mailto:Moses-support@mit.edu>
>>>>>>>>> http://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
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> 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/20131004/9f3224b5/attachment-0001.htm

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

Message: 3
Date: Fri, 4 Oct 2013 13:52:30 +0100
From: Hieu Hoang <Hieu.Hoang@ed.ac.uk>
Subject: [Moses-support] lattices with EPSILON
To: cdec-users@googlegroups.com, joshua_technical@googlegroups.com,
moses-support <moses-support@MIT.EDU>
Message-ID:
<CAEKMkbg8mqwBDdcJwNc0xoYJxJiudFVDiqx3cZ_Phyf4y9SgHQ@mail.gmail.com>
Content-Type: text/plain; charset="iso-8859-1"

I'm just looking at the lattices decoding, as implemented in moses.

for confusion networks, it's fair to have EPSILON words (that represent
blank words). However, I don't see the point of them in lattices.

Anyone have an opinion? How is it implemented in cdec & joshua?

--
Hieu Hoang
Research Associate
University of Edinburgh
http://www.hoang.co.uk/hieu
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.mit.edu/mailman/private/moses-support/attachments/20131004/b61ed2d7/attachment.htm

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

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


End of Moses-support Digest, Vol 84, Issue 6
********************************************

Tweet