Hi Cuong, the latest version of the MTeval script is v13a: - ftp://jaguar.ncsl.nist.gov/mt/resources/mteval-v13a.pl
You don't have to write an own converter, just use wrap-xml.perl: - http://statmt.org/wmt11/wrap-xml.perl There are usage instructions on the WMT 2012 website; under the "Test Set Submission" section: - http://statmt.org/wmt12/translation-task.html Hope that helps, Christian PS: the error message you're receiving tells you that there are no <seg id="123">some sentence text</seg> segments (= sentences) in your .sgm file; again, wrap-xml.perl should be helpful. On Jan 2, 2013, at 7:44 AM, Cuong Hoang wrote: > Hi Everyone, > I try to write a text-to-SGML format converter, > My First lines: > > <srcset setid="newsdev2009" srclang="any"> > t<doc docid="aktualne.cz-2010-12-02-60918" genre="news" origlang="en"> > > I don't know what is the docid here means and set the name is: > aktualne.cz-2010-12-02-60918 > However, when I run mteval-v11b.pl, I'm stuck with the error: > > FATAL INPUT ERROR: no segments in document 'aktualne.cz-2010-12-02-60918' in > file ... > > So how to overcome this problem? > Thanks and best regards, > > > > On Sun, Dec 30, 2012 at 11:37 AM, Cuong Hoang <hoangcuong2...@gmail.com> > wrote: > Hi all, > I have a task in which I need to compare to WER and PER, NIST. > I have three questions here, hope someone would help me. > 1. Do you know where I can download WER and PER code? > 2. NIST implementation from NIST mteval script requires that text is wrapped > into a SGML format. However, I still got a problem to convert my text files > to SGML format. > For example, I have three files: testCorpus.en or testCorpus.fr and an OUTPUT > file. But I follow the example from WMT tutorial is quite complicating: > • scripts/wrap-xml.perl dev/newstest2010-ref.en.sgm en < > newstest2010.output.detokenized > > working-dir/evaluation/newstest2010.output.sgm > In this case, what is newstest2010-ref.en.sgm, and what is > newstest2010.output.sgm in my case? (Sorry if this very is a naive question) > > Thanks > C. Hoang > > -- > Best Regards, > C. Hoang > > {Mimosa, SMT}@Addict > > > > -- > Best Regards, > C. Hoang > > {Mimosa, SMT}@Addict > _______________________________________________ > 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 ------------------------------------------------------------------- _______________________________________________ Moses-support mailing list Moses-support@mit.edu http://mailman.mit.edu/mailman/listinfo/moses-support
