[Moses-support] IWSLT 2019: 2nd call for papers - one week left

[Apologies for multiple posting] The 16th International Workshop on Spoken Language Translation IWSLT 2019 – Second Call for Participation November 2nd – 3rd, 2019 – Hong Kong http://www.iwslt.org IMPORTANT DATES Scientific papers: September 1st: paper

[Moses-support] IWSLT 2019: 1st call for papers

[Apologies for multiple posting] The 16th International Workshop on Spoken Language Translation IWSLT 2019 – First Call for Participation November 2nd – 3rd, 2019 – Hong Kong http://www.iwslt.org IMPORTANT DATES Scientific papers: September 1st: paper

[Moses-support] Mert and Moses with Word Lattice

Dear All, I want to run mert using the word lattice. Following what described in previous emails, I have added to the moses.ini: 1) input-factors equal to 2 2) InputFeature num-features=1 num-input-features=1 real-word-count=0 3) InputFeature0= 1 Then I have validated my word lattice using

[Moses-support] PhD Position in Machine Translation at Fondazione Bruno Kessler

Turchi (tur...@fbk.eu). Best Regards, Matteo Negri, Marco Turchi ___ Moses-support mailing list Moses-support@mit.edu http://mailman.mit.edu/mailman/listinfo/moses-support

Re: [Moses-support] Error compiling with --with-mm

know On 24/03/2015 15:49, marco turchi wrote: Hi All, we have fixed the problem I reported yesterday. It is not clear to me the reason of that error, but I can compile Moses with the --with-mm flag, if in all the files inside the moses/TranslationModel/UG folder, I replace

Re: [Moses-support] Error compiling with --with-mm

be useful if someone has a similar problem. Thanks a lot! Marco On Mon, Mar 23, 2015 at 5:40 PM, marco turchi marco.tur...@gmail.com wrote: Hi All, I'm compiling the last version of Moses (downloaded today from the master). Everything works fine if I do not add the --with-mm parameter

Re: [Moses-support] Error compiling with --with-mm

explain or point me to a paper? Thanks, Tom On 03/24/2015 10:49 PM, marco turchi wrote: Hi All, we have fixed the problem I reported yesterday. It is not clear to me the reason of that error, but I can compile Moses with the --with-mm flag, if in all the files inside the moses

[Moses-support] Error compiling with --with-mm

Hi All, I'm compiling the last version of Moses (downloaded today from the master). Everything works fine if I do not add the --with-mm parameter, but when I include it, moses crashes. I have looked in the mailing list but I have not found anything useful. Do you have any ideas about a possible

[Moses-support] Summer Internship Positions at FBK (Trento, Italy)

(Sorry for multiple posting) Are you a *MS or a PhD student*? Are you interested in carrying out *research on NLP* during the summer? Are you excited to spend few summer weeks in a nice city in the *middle of the Alps*? *WE ARE LOOKING FOR YOU!!!* The Human Language Technology Machine

Re: [Moses-support] Learning curves

Hi Jorg, I published these papers on this topic few years ago. I hope they can be useful. Learning performance of a machine translation system: a statistical and computational analysis. *Proceedings of the Third Workshop on Statistical Machine Translation*. Association for Computational

[Moses-support] MT Marathon 2014 - Second Call for Papers and Participation

*MT Marathon 2014 and Open Source Convention* http://www.statmt.org/mtm14 Organized and sponsored by MosesCore EU project (http://www.mosescore.eu), Machine Translation Marathon 2014 is a week-long gathering of machine translation researchers, developers, students and users. It features: - MT

[Moses-support] MT Marathon 2014 Call for Papers and Participation

MT Marathon 2014 and Open Source Convention http://www.statmt.org/mtm14 Organized and sponsored by MosesCore EU project (http://www.mosescore.eu), Machine Translation Marathon 2014 is a week-long gathering of machine translation researchers, developers, students and users. It features: - MT

[Moses-support] NLP Summer Internship at FBK (Trento, Italy)

. ... More detailed information on the the topics and financial support are available here: http://hlt.fbk.eu/en/openpositions/intgrad Further enquiries to: Vivi Nastase, e-mail: hltsip AT list.fbk eu Marco Turchi, e-mail: hltsip AT list.fbk eu ___ Moses

[Moses-support] Deadline approaching - 16 September - Multilingual Machine Translation and Text Mining Position at the Joint Research Centre - European Commission

On behalf of the Optima Team at the Joint Research Centre - European Commission == Please pass onto any potentially interested parties. Apologies for cross-posting. ==

[Moses-support] Multilingual Machine Translation and Text Mining Position at the Joint Research Centre - European Commission

On behalf of the Optima Team at the Joint Research Centre - European Commission == Please pass onto any potentially interested parties. Apologies for cross-posting. ==

Re: [Moses-support] problem using plain2snt.out for incremental training

. With mgiza I haven't had any problems. That force alignment from mgiza, is really incremental? or it only uses the old models to align new sentences. -- *De:* Hieu Hoang fishandfrol...@gmail.com *Para:* marco turchi marco.tur...@gmail.com *CC:* moses-support

[Moses-support] problem using plain2snt.out for incremental training

Hi I'm following the incremental training steps from the Moses web site and I'm running the plain2snt command: plain2snt.out newData.lower.1.fr newData.lower.1.en -txt1-vocab ../training/prepared.1/fr.vcb -txt2-vocab ../training/prepared.1/en.vcb but I get this error plain2snt.out:

[Moses-support] Incremental training

Dear all, I'm starting to use the incremental training and I was wondering if it updates the language model as well. If the answer is not, is it possible to update the language model without restarting Moses? Thanks a lot Marco ___ Moses-support mailing

Re: [Moses-support] Problem compiling Moses

autoconf automake --add-missing You should slve the problem It is possible that you have to run this sequence twice I Nicola On Nov 17, 2011, at 3:43 PM, marco turchi wrote: Hi, we are trying to compile the last version of moses, but we got this problem when we run the make

Re: [Moses-support] Problem compiling Moses

for you, let me know. If git pull is giving an error message error: Updating the following directories would lose untracked files in it: kenlm then rm -rf kenlm followed by git pull. Kenneth On 11/17/11 15:11, marco turchi wrote: Hi Nicola, thanks a lot, I'm going to try ciao Marco

Re: [Moses-support] Problem building KenLm

it? thanks a lot Marco On Mon, Oct 10, 2011 at 7:22 PM, marco turchi marco.tur...@gmail.comwrote: I see, so no chances of translating without rebuild everything :-( Thanks a lot Marco On Mon, Oct 10, 2011 at 7:05 PM, Kenneth Heafield mo...@kheafield.comwrote: ** Ah. I've changed the version

Re: [Moses-support] Problem building KenLm

::FormatLoadException' what(): File looks like it should be loaded with mmap, but the test values don't match. Was it built on a different machine or with a different compiler? I have the feeling that my moses version needs to be updated! Thanks a lot Marco On Sat, Oct 8, 2011 at 1:02 PM, marco turchi

Re: [Moses-support] Problem building KenLm

. If that doesn't work, please send me the first kilobyte of your binary file. Also, if you have Boost, can you cd kenlm make clean ./test.sh and complain if there are any test failures? Kenneth On 10/10/11 17:31, marco turchi wrote: Hi Kenneth, which number shall I use in the moses.ini 8

Re: [Moses-support] Problem building KenLm

by the tests. Kenneth On 10/10/11 17:57, marco turchi wrote: Hi what I have done is to download the last version of kenlm, built it and run the build_binary. Then I call a 2010 version of moses using the lm. I run the ./test.sh and I got these messages: util/bit_packing_test.cc:14: error

[Moses-support] Problem building KenLm

Dear All, I'm trying to build a lm using a large dataset ( 11 M sentences). I have generated the Arpa format with irstlm and now I'd like to binarize it using kenlm. I have called the build_binary to estimate memory usage, and I got this Memory estimate: type MB probing 16129 assuming -p

Re: [Moses-support] Problem building KenLm

for left state minimization. I'll fix it soon. A workaround is to pass a large -m option to build_binary. Sorry, Kenneth On 10/08/11 11:34, marco turchi wrote: Dear All, I'm trying to build a lm using a large dataset ( 11 M sentences). I have generated the Arpa format with irstlm

Re: [Moses-support] Problem building KenLm

Thanks! I'm going to update my version. Cheers Marco On Sat, Oct 8, 2011 at 1:01 PM, Kenneth Heafield mo...@kheafield.comwrote: ** Fixed in revision 4314. There's still an issue with some SRILM models failing to build that I'll get to soon. On 10/08/11 11:52, marco turchi wrote: Hi

Re: [Moses-support] analysis.perl output files

Hi Philipp, thanks a lot! it was exactly what I need. Cheers Marco On Wed, Sep 28, 2011 at 3:28 PM, Philipp Koehn pko...@inf.ed.ac.uk wrote: Hi, On Mon, Sep 26, 2011 at 5:40 PM, marco turchi marco.tur...@gmail.com wrote: corpus-coverage-summary and ttable-coverage-summary: what does

[Moses-support] analysis.perl output files

Dear All, I have had a look to the output files generated by the analysis.perl run in this way: analysis.perl -input x -input-corpus y -ttable z -dir w I have read the information available on the Moses web site http://www.statmt.org/moses/?n=FactoredTraining.EMS#ntoc25 about the analysis.perl

Re: [Moses-support] Problem training moses

AM, marco turchi marco.tur...@gmail.com wrote: Dear All, I'm training a model using 6M en-ar parallel sentences. Mgiza produces the two alignement files A3.final.gz, then the experimental enviroment crashes reporting only these messages: number of steps doable or running: 2

Re: [Moses-support] Problem training moses

that indicate errors that were found in the *.STDERR output. If the *.STDERR.digest are empty then indeed no errors were found, but then experiment.perl should not have stopped. -phi On Mon, Sep 5, 2011 at 9:21 AM, marco turchi marco.tur...@gmail.com wrote: Hi Philipp, I have checked

Re: [Moses-support] Problem training moses

5, 2011 at 2:05 PM, marco turchi marco.tur...@gmail.com wrote: Hi both the digest files contain only this message: error I can check in experiment.perl where it is printed out. Thanks a lot Marco On Mon, Sep 5, 2011 at 3:01 PM, Philipp Koehn pko...@inf.ed.ac.uk wrote: Hi

[Moses-support] Problem training moses

Dear All, I'm training a model using 6M en-ar parallel sentences. Mgiza produces the two alignement files A3.final.gz, then the experimental enviroment crashes reporting only these messages: number of steps doable or running: 2 executing

Re: [Moses-support] Moses caching questions

ISOpub), etc. Again, more to do next week. Tom On Mon, 29 Aug 2011 14:16:02 +0200, marco turchi marco.tur...@gmail.com wrote: Hi Tom, I'm running something similar to your wrapper in Java with a 16 core (thanks to hyperthreading) machines, and a common problem that I had at the beginning

Re: [Moses-support] Moses caching questions

?option=com_contentview=articleid=94:are-there-characters-that-cause-problems-in-mosescatid=30:key-conceptsItemid=57 It's missing the 5 reserved XML characters: amp; lt; gt; ' apos; quot; Tom On Tue, 30 Aug 2011 14:11:47 +0200, marco turchi marco.tur

Re: [Moses-support] Moses caching questions

Hi Tom, I'm running something similar to your wrapper in Java with a 16 core (thanks to hyperthreading) machines, and a common problem that I had at the beginning was the presence of the | characters in the source sentence. Cheers Marco On Mon, Aug 29, 2011 at 1:58 PM, Barry Haddow

Re: [Moses-support] is mert taking too much time here?

Hi Miles, I check how tuning time changes according to the tuning set size. I do not use a drastically large tuning set but I guess it can give an idea. Section 4.5 of this technical report: https://patterns.enm.bris.ac.uk/files/LearningCurveMain.pdf Cheers Marco On Mon, May 30, 2011 at 11:08

[Moses-support] Sentence mismatch error

Hi, I'm training the syntactic model, but I have some problems when I run inverse giza . I clean the data before and after running the Collins parser , but inverse-giza has fertility problems: WARNING: The following sentence pair has source/target sentence length ration more than the maximum

Re: [Moses-support] collins-parser only parse 2500 !

Hi I have the same problem with the Collins' parser. Do u know exactly what I need to change in the source code of the parser? or u have a modified version? Thanks a lot Marco On Thu, Jun 3, 2010 at 5:03 PM, Hwidong Na le...@postech.ac.kr wrote: Hi, This is not because of the wrapper script,

Re: [Moses-support] collins-parser only parse 2500 !

might use also. It splits the input into chunks of 2500 lines, It is used like this: file split-file-wrapper.py 2500 parse-en-collins outfile (But it makes the processing much slower, modifying the source would be better). -- Raphael Payen On Fri, 2010-10-15 at 14:40 +0200, marco turchi

Re: [Moses-support] Problem compiling moses

and compile your own version of libz, then persuade moses to use them. I'm afraid I don't have access to a centos install to try it out, regards Barry On Thursday 19 August 2010 08:55, marco turchi wrote: Hi Berry, looking on the Web, I found several posts http://ubuntuforums.org

[Moses-support] Compound splitter

Hi, I'd like to understand in details how the compound splitter works. Can I have more information about it or in case a pointer to a paper? Thanks a lot Marco ___ Moses-support mailing list Moses-support@mit.edu

Re: [Moses-support] is jrc acquis strange

Hi Mark, the JRC-acquis is extracted for legal documents, and in some case sentences can be strange compared to the sentence that u can find in Europarl. They can be very short, contain only numbers or references to some laws, My suggestion is to create another test set and see how your

[Moses-support] problem running mosesserver

Hi, I have installed moses adding xmlrpc-c and boost library locations to the configuration file. i can run and use properly mosesmt , but when I run mosesserver i got this error: ./mosesserver: error while loading shared libraries: libxmlrpc_server_abyss++.so.6: cannot open shared object file:

Re: [Moses-support] Binarized SRILM

Hi, I run the same test as the previous email using 1.1 M training data and also in this case I got the same bleu score and the same target sentences when I use the binary and normal SRILM language model Thanks Marco On Thu, Dec 3, 2009 at 1:31 PM, marco turchi marco.tur...@gmail.com wrote: Hi

Re: [Moses-support] Binarized SRILM

. -phi On Wed, Dec 2, 2009 at 11:44 PM, marco turchi marco.tur...@gmail.com wrote: Hi Hieu, I'll check... and I'll let u know. In case this the output of moses using a binarized srilm LM: [total=-40.478] -8.000, -74.000, 0.000, -11.209, -4.077, -2.675, -12.833, -0.519, -5.449

[Moses-support] Binarized SRILM

Hi, just a quick question, can Moses use a binary version of the srilm lm? I have built it with the option -write-binary-lm Thanks Marco ___ Moses-support mailing list Moses-support@mit.edu http://mailman.mit.edu/mailman/listinfo/moses-support

Re: [Moses-support] problem with boost library

/boost_1_41_0/include. This is probably where it is looking for them. If none of this works, then post your config.log so I can have a look at it, regards Barry On Friday 20 November 2009 18:35:34 marco turchi wrote: Hi, I'm trying to install Moses server on a linux machine at 64-bit

Re: [Moses-support] problem with boost library

is not in /usr/local/boost/lib, but it is in the boost folder in boost_1_41_0/bin.v2/libs how can I tell to moses to look for it there? thanks Marco On Mon, Nov 23, 2009 at 10:41 AM, marco turchi marco.tur...@gmail.comwrote: Hi thanks a lot for the advices... I compiled again the boost

Re: [Moses-support] problem with boost library

Hi Barry, I put all the dynamic libraries and then before to run moses I export: export LD_LIBRARY_PATH=/usr/local/ boost/lib/ and it works... Thanks again Marco On Mon, Nov 23, 2009 at 3:36 PM, Barry Haddow bhad...@inf.ed.ac.uk wrote: On Monday 23 Nov 2009 13:21:27 marco turchi wrote: Hi

[Moses-support] Moses server on the Windows version

Dear All, I have seen on the Moses web site that a Moses server and multi-threaded Moses are available. Are these services included on the Windows version of Moses? Thanks Marco ___ Moses-support mailing list Moses-support@mit.edu

Re: [Moses-support] Moses server on the Windows version

. But most of the moses developers use Linux or OSX, so you're better off using one of these platforms if possible, best regards Barry On Monday 26 October 2009 13:39, marco turchi wrote: Dear All, I have seen on the Moses web site that a Moses server and multi-threaded Moses are available

[Moses-support] Error on two halves consolidation

Hi, I have installed moses from scratch, and I test it on a small dataset, but when I run the training phase i got this error: (6.5) [0-0] consolidating the two halves @ Fri Jun 12 14:35:37 BST 2009 mismatch line 1: (! ) ne ! ) || ! ) ne ! ) || BLANK ne (0) (1) || BLANK ne (0) (1) ) mismatch

[Moses-support] Error running moses

Dear All, I'm running Moses, but I get an error when I try to translate a sentence. I save the srilm LM as binary file. I filter TM and reordering models, then I binarize the reduced TM and reordering models. When I run moses I got this: Defined parameters (per moses.ini or switch):

Re: [Moses-support] Moses-support Digest, Vol 31, Issue 2

hi, I see, but I need to run moses with the parameter -xml-input exclusive Thanks Marco On Sat, May 2, 2009 at 11:09 PM, Philipp Koehn pko...@inf.ed.ac.uk wrote: Hi, yes, but x translation=andand/x -phi On Sat, May 2, 2009 at 11:02 PM, marco turchi marco.tur...@gmail.com wrote: hi

Re: [Moses-support] Moses-support Digest, Vol 31, Issue 2

the XML markup, i.e. translate x and /x as if they were regular words. -phi On Sat, May 2, 2009 at 8:04 PM, marco turchi marco.tur...@gmail.com wrote: Hi Kevin, is it different from your suggestion if I use xthe following word/x with the parameter -xml-input ignore? Thanks a lot Marco

[Moses-support] force decoder to skip particular words

Dear All, can I force the decoder to skip particular word/phrase during the translation? I'd like that it will not look for a word/phrase inside the tables. thanks a lot Marco ___ Moses-support mailing list Moses-support@mit.edu

[Moses-support] cmert problem

Dear experts, I get this error running cmert: Hoping that scoring succeeded. We'll see if we can read the output files now. Executing: \cp -f init.opt run6.init.opt 852500 accumulated translations prev accumulated translations was : 796818 Starting cmert. Executing:

[Moses-support] Moses on Arabic data

Dear All, I'd like to use Moses to translate from Arabic to English. I have few questions: 1)Do u know where I can find a parallel corpora Arabic-English? 2)Do all the Moses tools (tokenizer, lowercaser) work for Arabic language? Thanks a lot Marco ___

Re: [Moses-support] lexical translation probability

, this was the case for a small number of phrase-pairs but I didn't worry to much because it happened rarely. Hieu Hoang www.hoang.co.uk/hieu From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of marco turchi Sent: 28 October 2008 16:30 To: moses

[Moses-support] lexical translation probability

Dear all, I'm trying to rewrite in Python the lexical translation probability code following the code inside score.cpp. In particular case, I get different results from the score.cpp output #output From Score.cpp partido de ||| of ||| (0) (0) ||| (0,1) ||| 0.67 0.75 lex0(of|partido) = 0.75 #

[Moses-support] binarize language model

Dear all, I'm using the srilm language model, but it is big and I'd like to binarize and quantize it. can I do it using the irst scripts without to set irstlm as default language model in Moses? thanks Marco ___ Moses-support mailing list

[Moses-support] Giza + Phrase extraction

Dear all, are there any checks either in Giza or during the phrase extraction for discarding bad word alignments or phrase associations? Thanks a lot Marco ___ Moses-support mailing list Moses-support@mit.edu

Re: [Moses-support] maximum Phrase length

is the maximum number of tokens in the foreign language side of an entry in the translation model. (There is no limit to the number of tokens that the target phrase may contain.) Chris On Mon, Apr 28, 2008 at 8:26 PM, marco turchi [EMAIL PROTECTED] wrote: Dear experts, one

[Moses-support] maximum Phrase length

Dear experts, one parameter of the Moses training script is the maximum phrase length. Is the maximum number of tokens of the foreign language phrase inside an entry of the translation model? Thanks Marco ___ Moses-support mailing list

Re: [Moses-support] NaN into the alignment phase

recommendation is to try to get around this by filtering your corpus based on sentence length/sentence length ratio (please, let me know if this solution works for you!). Once I confirm this is the problem, I'll look into adding trellis normalization to GIZA. --Chris On Dec 16, 2007 1:43 AM, marco

Re: [Moses-support] mert-moses on binarized moses.ini

in the configuration file. The decoder looks for the binary versions with the default extensions (.binphr.*). -phi On 11/18/07, marco turchi [EMAIL PROTECTED] wrote: Dear experts, I'm trying to run the mert-moses process using a binarized version of the phrase table and reordering table, but i got