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: monotone-at-ponctuation questions (Philipp Koehn)
2. Re: Problem in extracting rules for syntax based model
(Khyati Gupta)
3. Re: monotone-at-ponctuation questions (Vito Mandorino)
----------------------------------------------------------------------
Message: 1
Date: Mon, 15 Feb 2016 15:26:49 -0500
From: Philipp Koehn <phi@jhu.edu>
Subject: Re: [Moses-support] monotone-at-ponctuation questions
To: Kekwal <vito.mandorino@linguacustodia.com>
Cc: moses-support <moses-support@mit.edu>
Message-ID:
<CAAFADDD7t+7Zt=s8ggDTzxurHX6y8wFjfaLntdBvQUeK512DUw@mail.gmail.com>
Content-Type: text/plain; charset="utf-8"
Hi,
On Mon, Feb 15, 2016 at 7:43 AM, Kekwal <vito.mandorino@linguacustodia.com>
wrote:
> Dear all,
>
> I have some questions on the -mp option of the decoder:
>
> 1. Is the -mp option integrated by default in the decoder? I have compared
> more than 20000 segments translated with and without the option and there
> are no differences in the outputs.
>
No, it is not turned on by default. I do not have a good explanation why
there is no difference in your segments, but it is likely due to very
little reordering in general.
2. Is the -mp option compatible with mosesserver?
>
It should be, but I have not tested it personally.
3. According to the manual, -mp is equivalent to adding <wall /> tags
> before and after the following ponctuation signs
> ,.!?:;"
> like in
> , -----> <wall /> , <wall />
> Could it be useful to add walls also before and after brackets? Which
> files in the repository should be modified in order to achieve this?
> ( ----> <wall /> ( <wall />
> ) ----> <wall /> ) <wall />
>
You could just add these walls into the xml markup of the input sentence.
The code is in moses/ReorderingConstraint.cpp
You may also use zones:
this a <zone> ( <wall/> small <wall/> ) </zone> test .
This allows movement of the whole parenthetical expression, with local
walls within to ensure proper placement of the parentheses.
-phi
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.mit.edu/mailman/private/moses-support/attachments/20160215/3dd985c3/attachment-0001.html
------------------------------
Message: 2
Date: Tue, 16 Feb 2016 09:01:53 +0000 (UTC)
From: Khyati Gupta <gkhyati@rocketmail.com>
Subject: Re: [Moses-support] Problem in extracting rules for syntax
based model
To: Hieu Hoang <hieuhoang@gmail.com>, "moses-support@mit.edu"
<moses-support@mit.edu>, Philipp Koehn <phi@jhu.edu>
Cc: "invincible.roy@gmail.com" <invincible.roy@gmail.com>
Message-ID:
<430605842.4929913.1455613313060.JavaMail.yahoo@mail.yahoo.com>
Content-Type: text/plain; charset="utf-8"
Sir,
I had already placed this switch while training which gave the output-
42800 [X] ||| <tree [SC] ||| 0.00103349 1 0.0103349 1 ||| 0-0 ||| 10 1 1 |||
Does the syntactic tags have a effect on moses translation ? Does moses support Penn Treebank only??Enju parser doesnt use Penn Treebank..
Pls do reply.
?Thanks
On Saturday, February 13, 2016 4:09 AM, Hieu Hoang <hieuhoang@gmail.com> wrote:
when you extract the rules during training, you must tell the extract program that there is parse information on the target side
??? .../extract-rules --TargetSyntax? ....
On 08/02/16 10:31, Khyati Gupta wrote:
Hi,
I have completed string to tree syntax based model. And have converted the output of enju parser to format accepted by moses for synatx based training. But the rule table created consist of phrases like <tree label ............ as a result the translated text also contains these phrases.
Output of enju- <sentence id="s0" parse_status="success" fom="0.991145"><cons id="c0" cat="NP" xcat="" head="c1" sem_head="c1" schema="empty_spec_head"><cons id="c1" cat="NX" xcat="" head="t0" sem_head="t0"><tok id="t0" cat="N" pos="NN" base="broadcasting" lexentry="[D<N.3sg>]"pred="noun_arg0">broadcasting</tok></cons></cons></sentence>
Converted format of enju to format accepted by moses- <tree label="TOP"> <tree label="NP"> <tree label="NX"> <tree label="N"> broadcasting </tree> </tree> </tree> </tree>?
Rule table created- 42800 [X] ||| <tree [SC] ||| 0.00103349 1 0.0103349 1 ||| 0-0 ||| 10 1 1 |||
[X][ADJP] ?????? [X] ||| <tree [X][ADJP] <tree [NP] ||| 0.00729519 1 0.00226863 1 ||| 0-1 1-2 ||| 0.485714 1.5619 0.342857 |||
Where am i going wrong?? Is there a problem in XML Format. ??? ? Thanks Khyati Gupta Trainee CDAC Noida
--
Hieu Hoang
http://www.hoang.co.uk/hieu
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.mit.edu/mailman/private/moses-support/attachments/20160216/67e31726/attachment-0001.html
------------------------------
Message: 3
Date: Tue, 16 Feb 2016 13:53:58 +0100
From: Vito Mandorino <vito.mandorino@linguacustodia.com>
Subject: Re: [Moses-support] monotone-at-ponctuation questions
To: Philipp Koehn <phi@jhu.edu>
Cc: moses-support <moses-support@mit.edu>
Message-ID:
<CA+8mSmEU_18kXKWQOiFqnc4AeGx1iZ0UsQ=EaSESffnbu=-J4Q@mail.gmail.com>
Content-Type: text/plain; charset="utf-8"
Thank you. The <zone> tag seems to be more appropriate for brackets indeed.
I have done some decoding tests with verbose option:
/mosesdecoder/bin/moses -mp --verbose 3 -f moses.ini.
In the log, there is no mention of XML wall tags using the current moses
repository on github. Instead, there do appear the line
XML TAG IS: wall/
when decoding with an ancien version of moses or, alternatively, when
manually inserting <wall/> tags in the source segment.
Best regards,
Vito
2016-02-15 21:26 GMT+01:00 Philipp Koehn <phi@jhu.edu>:
> Hi,
>
> On Mon, Feb 15, 2016 at 7:43 AM, Kekwal <vito.mandorino@linguacustodia.com
> > wrote:
>
>> Dear all,
>>
>> I have some questions on the -mp option of the decoder:
>>
>> 1. Is the -mp option integrated by default in the decoder? I have
>> compared more than 20000 segments translated with and without the option
>> and there are no differences in the outputs.
>>
>
> No, it is not turned on by default. I do not have a good explanation why
> there is no difference in your segments, but it is likely due to very
> little reordering in general.
>
> 2. Is the -mp option compatible with mosesserver?
>>
>
> It should be, but I have not tested it personally.
>
> 3. According to the manual, -mp is equivalent to adding <wall /> tags
>> before and after the following ponctuation signs
>> ,.!?:;"
>> like in
>> , -----> <wall /> , <wall />
>> Could it be useful to add walls also before and after brackets? Which
>> files in the repository should be modified in order to achieve this?
>> ( ----> <wall /> ( <wall />
>> ) ----> <wall /> ) <wall />
>>
>
> You could just add these walls into the xml markup of the input sentence.
> The code is in moses/ReorderingConstraint.cpp
>
> You may also use zones:
>
> this a <zone> ( <wall/> small <wall/> ) </zone> test .
>
> This allows movement of the whole parenthetical expression, with local
> walls within to ensure proper placement of the parentheses.
>
> -phi
>
>
--
*M**. Vito MANDORINO -- Chief Scientist*
[image: Description : Description : lingua_custodia_final full logo]
*The Translation Trustee*
*1, Place Charles de Gaulle, **78180 Montigny-le-Bretonneux*
*Tel : +33 1 30 44 04 23 Mobile : +33 6 84 65 68 89*
*Email :* *vito.mandorino@linguacustodia.com
<massinissa.ahmim@linguacustodia.com>*
*Website :* *www.linguacustodia.com <http://www.linguacustodia.com/> -
www.thetranslationtrustee.com <http://www.thetranslationtrustee.com/>*
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.mit.edu/mailman/private/moses-support/attachments/20160216/ce43fdfe/attachment.html
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image001.jpg
Type: image/jpeg
Size: 4421 bytes
Desc: not available
Url : http://mailman.mit.edu/mailman/private/moses-support/attachments/20160216/ce43fdfe/attachment.jpg
------------------------------
_______________________________________________
Moses-support mailing list
Moses-support@mit.edu
http://mailman.mit.edu/mailman/listinfo/moses-support
End of Moses-support Digest, Vol 112, Issue 30
**********************************************
Subscribe to:
Post Comments (Atom)
0 Response to "Moses-support Digest, Vol 112, Issue 30"
Post a Comment