Re: [Moses-support] 1st CFP: EMNLP 2017 Workshop on Speech-Centric Natural Language Processing (SCNLP)

[Nick Ruiz]

THIRD CALL FOR PAPERS
EMNLP 2017 Workshop on Speech-Centric Natural Language Processing (SCNLP)

Hosted at Empirical Methods for Natural Language Processing (EMNLP)
conference
September 7, 2017, Copenhagen Denmark

SCNLP's goal is to unite the ASR and NLP communities to discuss new
frameworks for exploiting the rich information present in the speech signal
to improve the capabilities of natural language processing applications
such as conversational agents, question-answering systems, machine
translation, and search. SCNLP encourages novel contributions that revisit
the conventional NLP problems with a focus on incorporating the richness of
spoken language, as well as contributions that promote cross-fertilization
between statistical methods for ASR and NLP.

We envision SCNLP as a platform to promote collaboration between the ASR
and NLP communities and to seek ways to lower the barrier of entry for
researchers interested in working in the exciting intersection between
Speech and Natural Language Processing.

Important Dates
February 2017: First call for workshop papers
June 2 2017: Workshop papers due
June 30 2017: Notification of acceptance
July 14 2017: Camera-ready due
September 7 2017: Workshop date

*Note to MT Researchers*: Unfortunately, this workshop overlaps with Day 1
of the Second Conference on Machine Translation (WMT17), also co-located at
EMNLP. We do not wish to detract submissions from WMT, but we would
encourage your participation if you are working on the intersection of ASR
and MT.

Topics of interest include but are not limited to:

Speech translation
Spoken dialogue systems
Spoken query reformulation for Question/Answering systems
Word-sense disambiguation for speech transcripts
Information extraction on speech transcripts (combining textual corpora
with speech corpora)
Domain adaptation (Adapting textual NLP training data to speech-centric
tasks)
ASR evaluation for NLP
ASR error modeling
Speech segmentation for NLP
Punctuation insertion
Disfluency detection and correction
NLP with ASR lattices/confusion networks
Discourse and Speech Processing
Joint ASR/NLP modeling

We are excited to have a strong program committee consisting of research
leaders spanning the Speech and NLP communities.

For more information, please visit our website here:
http://speechnlp.github.io/2017/

We look forward to seeing you!

Nicholas Ruiz (Interactions, USA)
Srinivas Bangalore (Interactions, USA)
sc...@interactions.com

-- 
 

***

This e-mail and any of its attachments may contain Interactions Corporation 
proprietary information, which is privileged, confidential, or subject to 
copyright belonging to the Interactions Corporation. This e-mail is 
intended solely for the use of the individual or entity to which it is 
addressed. If you are not the intended recipient of this e-mail, you are 
hereby notified that any dissemination, distribution, copying, or action 
taken in relation to the contents of and attachments to this e-mail is 
strictly prohibited and may be unlawful. If you have received this e-mail 
in error, please notify the sender immediately and permanently delete the 
original and any copy of this e-mail and any printout. Thank You.  

*** 
 

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

[Moses-support] 1st CFP: EMNLP 2017 Workshop on Speech-Centric Natural Language Processing (SCNLP)

[Nick Ruiz]

FIRST CALL FOR PAPERS
EMNLP 2017 Workshop on Speech-Centric Natural Language Processing (SCNLP)

Hosted at Empirical Methods for Natural Language Processing (EMNLP)
conference
September 7, 2017, Copenhagen Denmark

SCNLP's goal is to unite the ASR and NLP communities to discuss new
frameworks for exploiting the rich information present in the speech signal
to improve the capabilities of natural language processing applications
such as conversational agents, question-answering systems, machine
translation, and search. SCNLP encourages novel contributions that revisit
the conventional NLP problems with a focus on incorporating the richness of
spoken language, as well as contributions that promote cross-fertilization
between statistical methods for ASR and NLP.

We envision SCNLP as a platform to promote collaboration between the ASR
and NLP communities and to seek ways to lower the barrier of entry for
researchers interested in working in the exciting intersection between
Speech and Natural Language Processing.

Important Dates
February 2017: First call for workshop papers
June 2 2017: Workshop papers due
June 30 2017: Notification of acceptance
July 14 2017: Camera-ready due
September 7 2017: Workshop date

*Note to MT Researchers*: Unfortunately, this workshop overlaps with Day 1
of the Second Conference on Machine Translation (WMT17), also co-located at
EMNLP. We do not wish to detract submissions from WMT, but we would
encourage your participation if you are working on the intersection of ASR
and MT.

Topics of interest include but are not limited to:

Speech translation
Spoken dialogue systems
Spoken query reformulation for Question/Answering systems
Word-sense disambiguation for speech transcripts
Information extraction on speech transcripts (combining textual corpora
with speech corpora)
Domain adaptation (Adapting textual NLP training data to speech-centric
tasks)
ASR evaluation for NLP
ASR error modeling
Speech segmentation for NLP
Punctuation insertion
Disfluency detection and correction
NLP with ASR lattices/confusion networks
Discourse and Speech Processing
Joint ASR/NLP modeling

We are excited to have a strong program committee consisting of research
leaders spanning the Speech and NLP communities.

For more information, please visit our website here:
http://speechnlp.github.io/2017/

We look forward to seeing you!

Nicholas Ruiz (Interactions, USA)
Srinivas Bangalore (Interactions, USA)
sc...@interactions.com

-- 
 

***

This e-mail and any of its attachments may contain Interactions Corporation 
proprietary information, which is privileged, confidential, or subject to 
copyright belonging to the Interactions Corporation. This e-mail is 
intended solely for the use of the individual or entity to which it is 
addressed. If you are not the intended recipient of this e-mail, you are 
hereby notified that any dissemination, distribution, copying, or action 
taken in relation to the contents of and attachments to this e-mail is 
strictly prohibited and may be unlawful. If you have received this e-mail 
in error, please notify the sender immediately and permanently delete the 
original and any copy of this e-mail and any printout. Thank You.  

*** 
 

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

Re: [Moses-support] Problem with linking libraries

[Nick Ruiz]

Hi all,

I saw this message from last month that doesn't look like it's been 
addressed. This is somewhat of a common problem for people writing code 
in Moses and new to compiling with bjam. Does anyone have any advice?


Thanks,
Nick

On 04/29/2013 04:16 PM, Prashant Mathur wrote:

Hi all,

I have implemented a feature in moses and now I am trying to use the 
implementation of MIRA optimiser in mira/ directory.

For this I just add these few lines to the code.

OnlineLearner.cpp
[code]
#include "mira/Optimiser.h"
using namespace Mira;
...
MiraOptimiser* optimiser = NULL;
...
optimiser = new MiraOptimiser(slack, scale_margin, scale_margin_precision,
scale_update, scale_update_precision, boost, 
normaliseMargin, sigmoidParam);

[/code]

This is the exact code from mira/Main.cpp except the fact that the 
code itself is in mira/ directory.
The code was perfectly compiling before and even now the OnlineLearner 
compiles. But later there is some problem with linking.


gcc.link 
moses-chart-cmd/bin/gcc-4.4.7/release/debug-symbols-on/link-static/threading-multi/moses_chart
moses/bin/gcc-4.4.7/release/debug-symbols-on/link-static/threading-multi/libmoses.a(OnlineLearner.o): 
In function `MiraOptimiser':
/research/hlt/prashant/Moses/moses_RELEASE-v1/./mira/Optimiser.h:80: 
*undefined reference to `vtable for Mira::MiraOptimiser'*

collect2: ld returned 1 exit status

"g++" -L"/research/hlt/prashant/MyInstallation/lib" 
-L"/research/hlt/prashant/MyInstallation/lib64" -Wl,-R 
-Wl,"/research/hlt/prashant/MyInstallation/lib" -Wl,-R 
-Wl,"/research/hlt/prashant/MyInstallation/lib64" -Wl,-rpath-link 
-Wl,"/research/hlt/prashant/MyInstallation/lib" -Wl,-rpath-link 
-Wl,"/research/hlt/prashant/MyInstallation/lib64" -o 
"moses-chart-cmd/bin/gcc-4.4.7/release/debug-symbols-on/link-static/threading-multi/moses_chart" 
-Wl,--start-group 
"moses-chart-cmd/bin/gcc-4.4.7/release/debug-symbols-on/link-static/threading-multi/Main.o" 
"moses-chart-cmd/bin/gcc-4.4.7/release/debug-symbols-on/link-static/threading-multi/mbr.o" 
"moses-chart-cmd/bin/gcc-4.4.7/release/debug-symbols-on/link-static/threading-multi/IOWrapper.o" 
"moses-chart-cmd/bin/gcc-4.4.7/release/debug-symbols-on/link-static/threading-multi/TranslationAnalysis.o" 
"moses/bin/gcc-4.4.7/release/debug-symbols-on/link-static/threading-multi/libmoses.a"  
-Wl,-Bstatic -lirstlm -Wl,-Bdynamic -lrt -llzma -lbz2 
-lboost_system-mt -lboost_thread-mt -lSegFault -lz -lrt 
-Wl,--end-group -g -pthread


I take that there is some problem with linking the libraries. For my 
code to work I would need to use libmira_lib.a, but when I try adding 
/mira/ to moses/Jamfile .. it cannot use it because the library is 
currently being built.


Any solution to this problem?
Is there a way to include /Optimiser.h/ without changing the jamfiles?
I am pretty new to the linking stuff, so please pardon if I am wrong 
somewhere.


Thanks a lot,
--
Prashant


___
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

Re: [Moses-support] mosese decoder android and ios porting

[Nick Ruiz]

Porting Moses to Android would be interesting. I think that the Android 
NDK is improving so it could be more feasible nowadays. There are 
several speech translation apps available on the market (e.g. Jibbigo) 
that demonstrate that MT is certainly viable on Android (and iOS) 
devices without a client/server model. And particularly looking at the 
current generation phones (quad core, 2 gb RAM), it's certainly possible 
to have an Android-specific decoder. However, the Moses decoder has lots 
of built-in functionality that may be a bit heavy to port (and test).


Kenneth Heafield mentioned flashing his phone to Gentoo Linux and 
installing Moses on it about a year ago. While this doesn't address the 
NDK and JNI issues, perhaps he might be able to answer a few questions 
about compiling Moses on ARM architectures.


Nick

On 11/27/2012 07:32 AM, Le Truong Vinh Phu wrote:

Hi MG Jang,

I'm not sure there is a way to port moses decoder to run on Android, 
but I'm curious, why do you need to do that at the first place?


In my opinion, decoding is quite a heavy task, time and memory 
consuming that I'm not sure if an Android device is able to handle. So 
far the translation apps I've seen follow the client-server model, 
where the decoding takes place on server side.


Maybe your task is supposed to be extremely light-weight?

LTVP

On Nov 27, 2012, at 1:47 PM, myounggun jang > wrote:



I'm trying to port the moses decoder for android
I'll make tables in linux system. and then I just use it in android 
system. I just need to port moses decoder.
I'm trying to find article about moses android porting. however I 
didn't find useful article.
I found data for android 
http://www.mail-archive.com/moses-support@mit.edu/msg06392.html

however there are missing parts.
is there any aticle or some ported source for android and ios?
Thanks in advance.
MG Jang
___
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


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

Re: [Moses-support] Moses installation

[Nick Ruiz]

Hi Federico,

Please check the compilation instructions in BUILD-INSTRUCTIONS.txt -- 
it has a list of comprehensive steps to compile Moses, and also gives 
instructions on how to install dependencies (e.g. Boost).


Best,
Nick

On 11/13/2012 12:18 PM, f.fancellu wrote:

Hi,
I am trying to run the moses decoder the first time as written in the 
guidelines but I am encountering a problem when I get to this point.

cd ~/mosesdecoder/sample-models
  ~/mosesdecoder/bin/moses -f phrase-model/moses.ini<  phrase-model/in>  out

The problem is that there is no ~/mosesdecoder/bin folder in the mosesdecoder I 
have cloned in my home folder.
Have I missed something along the way?

I was wondering if someone could help me with this.
Thank you,
Federico Fancellu.


___
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

Re: [Moses-support] IRSTLM ERROR

[Nick Ruiz]

Hi Clement,

The installation instructions for IRSTLM can be found here:
http://sourceforge.net/apps/mediawiki/irstlm/index.php?title=Installation_Guidelines
also, here is the instruction manual:
http://sourceforge.net/apps/mediawiki/irstlm/index.php?title=User_Manual

First, try calling the following:
bash regenerate-makefiles.sh
(depending on the system, you may need to specify which shell 
interpreter is needed)


If all else fails, do you have GNU autotools installed?

Best,
Nick

--
Nick Ruiz
Fondazione Bruno Kessler
FBK-Irst - Human Language Technology Unit
38050 Povo (TN) Italy

On 11/12/2012 06:04 PM, OYELEKE ODOJE wrote:

Hi Barry,

While installing IRSTLM file version irstlm-5.70.04.
the command line "./regenerate-makefiles.sh" generated the error below

mrodoje@ubuntu:~/irstlm-5.70.04$ ./regenerate-makefiles.sh
Calling
Calling ...
./regenerate-makefiles.sh: line 52: -I: command not found aclocal failed

what do you think could be done to configure the irstlm for baseline 
system?


 Mosesdecoder and Giza-pp,  have been installed.

Clement Odoje
Department of Linguistics and African Languages

University of Ibadan,

Ibadan, Nigeria

+2348032387999
What you do today becomes history tomorrow, what will you be 
remembered for?



___
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

Re: [Moses-support] SRILM vs IRSTLM

[Nick Ruiz]

Hi Marcin,

Have you done any perplexity tests on your trained LMs? For example, can 
you compute the perplexity on your evaluation set using IRSTLM and also 
using SRILM and compare the results? Also, keep in mind that IRSTLM 
reserves out-of-vocabulary probabilities based on the predefined 
vocabulary size of the LM. This is done using the `dub` parameter. I 
typically only use IRSTLM in Moses, so I'm not sure if the `dictionary 
upper-bound` could be something missing from KenLM that could cause a 
performance hit on IRSTLM-trained models. Just a guess.

Best,
Nick


On 11/08/2012 08:22 AM, Marcin Junczys-Dowmunt wrote:
> Hi Pratyush,
> Thanks for the hint. That solved the problem I had with the arpa files
> when using -lm=msb and KenLM. Unfortunately, this does not seem to
> improve performance of IRSTLM much when compared to SRILM. So I guess I
> will have to stick with SRILM for now.
>
> Kenneth, weren't you working on your own tool to produce language models?
> Best,
> Marcin
>
> W dniu 07.11.2012 11:18, Pratyush Banerjee pisze:
>> Hi Marcin,
>>
>> I have used msb with irstlm... but seems to have worked fine for me...
>>
>> You mentioned faulty arpa files for 5-grams... is it because KenLM
>> complains of missing 4-grams, 3-grams etc ?
>> Have you tried using -ps=no option with tlm ?
>>
>> IRSTLM is known to prune singletons n-grams in order to reduce the
>> size of the LM... (tlm has it on by default..)
>>
>> If you use this option, usually KenLM does not complain... I have also
>> used such LMs with SRILM for further mixing and it went fine...
>>
>> I am sure somebody from the IRSTLM community could confirm this...
>>
>> Hope this resolves the issue...
>>
>> Thanks and Regards,
>>
>> Pratyush
>>
>>
>> On Tue, Nov 6, 2012 at 9:26 PM, Marcin Junczys-Dowmunt
>> mailto:junc...@amu.edu.pl>>  wrote:
>>
>>  On the irstlm page it says:
>>
>>  'Modified shift-beta, also known as “improved kneser-ney smoothing”'
>>
>>  Unfortunately I cannot use "msb" because it seems to produce
>>  faulty arpa
>>  files for 5-grams. So I am trying only "shift-beta" whatever that
>>  means.
>>  Maybe that's the main problem?
>>  Also, my data sets are not that small, the plain arpa files currently
>>  exceed 20 GB.
>>
>>  Best,
>>  Marcin
>>
>>  W dniu 06.11.2012 22:15, Jonathan Clark pisze:
>>  >  As far as I know, exact modified Kneser-Ney smoothing (the current
>>  >  state of the art) is not supported by IRSTLM. IRSTLM instead
>>  >  implements modified shift-beta smoothing, which isn't quite as
>>  >  effective -- especially on smaller data sets.
>>  >
>>  >  Cheers,
>>  >  Jon
>>  >
>>  >
>>  >  On Tue, Nov 6, 2012 at 1:08 PM, Marcin Junczys-Dowmunt
>>  >  mailto:junc...@amu.edu.pl>>  wrote:
>>  >>  Hi,
>>  >>  Slightly off-topic, but I am out of ideas. I am trying to
>>  figure out
>>  >>  what set of parameters I have to use with IRSTLM to creates LMs
>>  that are
>>  >>  equivalent to language models created with SRILM using the
>>  following
>>  >>  command:
>>  >>
>>  >>  (SRILM:) ngram-count -order 5 -unk -interpolate -kndiscount -text
>>  >>  input.en -lm lm.en.arpa
>>  >>
>>  >>  Up to now, I am using this chain of commands for IRSTLM:
>>  >>
>>  >>  perl -C -pe 'chomp; $_ = "  $_\n"'<  input.en>
>>  input.en.sb
>>  >>  ngt -i=input.en.sb  -n=5 -b=yes -o=lm.en.bin
>>  >>  tlm -tr=lm.en.bin -lm=sb -bo=yes -n=5 -o=lm.en.arpa
>>  >>
>>  >>  I know this is not quite the same, but it comes closest in terms of
>>  >>  quality and size. The translation results, however, are still
>>  >>  consistently worse than with SRILM models, differences in BLEU
>>  are up to
>>  >>  1%.
>>  >>
>>  >>  I use KenLM with Moses to binarize the resulting arpa files, so
>>  this is
>>  >>  not a code issue.
>>  >>
>>  >>  Also it seems IRSTLM has a bug with the modified shift beta
>>  option. At
>>  >>  least KenLM complains that not all 4-grams are present although
>>  there
>>  >>  are 5-grams that contain them.
>>  >>
>>  >>  Any ideas?
>>  >>  Thanks,
>>  >>  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
>>
>>
>
> ___
> Moses-support mailing list
> Moses-support@mit.edu
> http://mailman.mit.edu/mailman/listinfo/moses-support

___
Moses-support mailing list
Moses-suppo

Re: [Moses-support] wrap-xml.perl

[Nick Ruiz]

Hi Cyrene,

Did you try adding the language (en) as a command-line parameter?

/wrap-xml.perl /home/cyrine/Bureau/testetvalidation/test-en.En.sgm en 
/home/cyrine/Bureau/testetvalidation/testocHtok.sgm


Let us know if you still have problems.

Best,
Nick

On 11/06/2012 09:24 PM, Cyrine NASRI wrote:

Hi,
Thank you Christian  for your reply
i read the tuto
i do like that :

/wrap-xml.perl /home/cyrine/Bureau/testetvalidation/test-en.En.sgm 
/home/cyrine/Bureau/testetvalidation/testocHtok.sgm


but i got this message :

Use of uninitialized value $src in -e at ./wrap-xml.perl line 6.
Use of uninitialized value $src in concatenation (.) or string at 
./wrap-xml.perl line 6.

wrapping frame not found () at ./wrap-xml.perl line 6.


Any idea please?

thank you
Bests

Cyrine



2012/11/6 Christian Federmann >


Hi Cyrine,

the following excerpt from WMT09's website should help:

Wrap the output in SGML
$ scripts/wrap-xml.perl wmt08/devtest/devtest2006-ref.en.sgm en <
working-dir/evaluation/devtest2006.output.detokenized >
working-dir/evaluation/devtest2006.output.sgm

Source: http://www.statmt.org/wmt09/baseline.html

Cheers and best,
   Christian



On Nov 6, 2012, at 3:57 PM, Cyrine NASRI wrote:

> Hello all,
>
> I'd like to know how to use the script wrap-xml.perl to change a
texte file to a xml format. If possible the general syntax.
>
> Bests
>
> --
> Cyrine
> Ph.D. Student in Computer Science
>
> ___
> Moses-support mailing list
> Moses-support@mit.edu 
> http://mailman.mit.edu/mailman/listinfo/moses-support




--
Dipl.-Inf. Christian Federmann, Researcher, Language Technology Lab
Office +1.09 -- Phone +49-681/857-75-5353
,  Fax +49-681/857-75-5338

DFKI GmbH,  Campus D3 2,  Stuhlsatzenhausweg 3,  66123 Saarbruecken
http://www.dfki.de/~cfedermann 

---
Deutsches Forschungszentrum fuer Kuenstliche Intelligenz GmbH
Trippstadter Strasse 122, D-67663 Kaiserslautern, Germany
Geschaeftsfuehrung:
Prof. Dr. Dr. h.c. mult. Wolfgang Wahlster (Vorsitzender)
Dr. Walter Olthoff
Vorsitzender des Aufsichtsrats:
Prof. Dr. h.c. Hans A. Aukes
Amtsgericht Kaiserslautern, HRB 2313
---






--
/Cyrine
Ph.D. Student in Computer Science/


___
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

[Moses-support] Building models with multiple translations of parallel corpora

[Nick Ruiz]

Hi everyone,

I'm curious if anyone has tried something like this: say you have a 
parallel corpus with multiple translations available in either the 
source or target language. One good example would be the Bible. Many 
translations exist in English, Italian, etc., but virtually all come 
from original sources with a Hebrew Old Testament and a Greek New 
Testament. Has anyone tried building phrase tables with a single 
translation of the source corpus and multiple translations of the target 
corpus (e.g. Greek New Testament + KJV/NIV/ESV/... English New 
Testaments)? The purpose of such a phrase table would be to increase the 
number of valid translations of a phrase (or prune out phrase 
translations that don't agree among the various translations).

I'm aware of various techniques for using multiple phrase tables, but 
what about combining all corpora into a single phrase table? I assume 
that a concatenation approach would require duplicating the source 
language corpus N times for each target parallel corpus and treating the 
task as if your target corpus did not have parallel versions. You could 
also mix PTs trained on each source/target corpus -- one way to combine 
into a single PT would be to use the Fill-Up technique (Nakov, 2008; 
Bisazza et al. 2011). Or you could use ensemble decoding (Razmara et 
al., 2012) or the standard (log-) linear combinations.

If you have considered this type of scenario before, what approaches 
have you tried?

Best regards,
Nick Ruiz
___
Moses-support mailing list
Moses-support@mit.edu
http://mailman.mit.edu/mailman/listinfo/moses-support

Re: [Moses-support] using IRSTLM to build a tri-gram language model.

[Nick Ruiz]

Hi Yared,

Your command syntax looks fine. Is this your first time using IRSTLM to 
train a LM on this machine? If not, please make sure you followed the 
installation guidelines listed here:

http://sourceforge.net/apps/mediawiki/irstlm/index.php?title=Installation_Guidelines

The redundant error is probably because build-lm is splitting the LM 
estimation into multiple jobs and the training command is not executing.


Nick

On 04/10/2012 03:04 PM, Yared Mekuria wrote:
hello there, I am using welsh and english parallel corpus to test the 
moses translation.

I tokenize and lower case the training data and while
I am using IRSTLM to build a tri-gram language model using the command

export IRSTLM=/home/guest/tools/irstlm
tools/irstlm/bin/build-lm.sh -t /tmp -i work/lm/pnaw-full.lowercased.en -o 
work/lm/pnaw-full.en.lm
I GET THE THE REDUNDANT ERROR SAYING

[: 180: 2: unexpexted operatator
[: 180: 2: unexpexted operatator
[: 180: 2: unexpexted operatator
[: 180: 2: unexpexted operatator ..

Please suggest me any ideas what I should do to solve this.
Thank you.


___
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

[Moses-support] Moses Wiki?

[Nick Ruiz]

Hi everyone,

Could we create an open wiki for Moses documentation? With all of the 
changes that have taken place recently, several of the tutorials are 
out-of-date and lead to code that won't compile. These tutorials are 
much appreciated, because they provide a starting point for new 
developers, but ideally it would be beneficial to create a wiki where 
all developers can fill in the missing parts of the documentation and 
hopefully make the documentation process easier and faster.

Nobody likes to write documentation, so maybe this approach could 
lighten the load.

Thanks for the consideration.
Nick Ruiz
___
Moses-support mailing list
Moses-support@mit.edu
http://mailman.mit.edu/mailman/listinfo/moses-support

Re: [Moses-support] Boost-build and adding new feature functions

[Nick Ruiz]

Thanks, Kenneth. You were right, I was using the src/LM directory, since 
my feature is language model-related, so I had thought it would be 
logical place for the source. Unfortunately, the src/LM directory isn't 
swept in the same way as the src directory, according to the Jamfiles. 
So, rather than coupling my feature with the code in the LM dir, I 
simply moved my sources up to the parent directory and everything 
compiles smoothly.

Nick

On 03/19/2012 10:13 PM, Kenneth Heafield wrote:
> Hi,
>
>   The equivalent is Jamfile.  Take a look at mert/Jamfile; the basics are
> fairly easy.  Some Jamfiles use a glob e.g. moses/src/Jamfile:
>
> lib moses :
> #All cpp files except those listed
> [ glob *.cpp DynSAInclude/*.cpp : ThreadPool.cpp
> SyntacticLanguageModel.cpp ]
> synlm ThreadPool CYKPlusParser//CYKPlusParser LM//LM
> RuleTable//RuleTable Scope3Parser//Scope3Parser headers ../..//z
> ../../OnDiskPt//OnDiskPt ;
>
>   Since it seems you made a new directory, you can either add to the glob
> like DynSAInclude/*.cpp or create a Jamfile for your directory and
> reference it as a library like CYKPlusParser does.
>
> Kenneth
>
> On 03/19/2012 04:42 PM, Nick Ruiz wrote:
>> Hi everyone,
>>
>> I have a question about adding new feature functions using bjam. I
>> created a new .h and .cpp file and incorporated the feature in
>> StaticData.h/.cpp; however, running bjam causes the friendly "undefined
>> reference" error when I try to call my new feature's constructor. The
>> problem is that the feature's source files aren't included in the build
>> process.
>>
>> How do I add the new files in the build process with Boost? I assumed
>> that the subdirectories would automatically be scanned, since I didn't
>> immediately see something like a Makefile to work from. Or maybe I just
>> missed it. :)
>>
>> Thanks,
>> Nick
>> ___
>> 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

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

[Moses-support] Boost-build and adding new feature functions

[Nick Ruiz]

Hi everyone,

I have a question about adding new feature functions using bjam. I 
created a new .h and .cpp file and incorporated the feature in 
StaticData.h/.cpp; however, running bjam causes the friendly "undefined 
reference" error when I try to call my new feature's constructor. The 
problem is that the feature's source files aren't included in the build 
process.

How do I add the new files in the build process with Boost? I assumed 
that the subdirectories would automatically be scanned, since I didn't 
immediately see something like a Makefile to work from. Or maybe I just 
missed it. :)

Thanks,
Nick
___
Moses-support mailing list
Moses-support@mit.edu
http://mailman.mit.edu/mailman/listinfo/moses-support