Obviously The error is due to the usage of a different corpus in building the language model (at least that's what happened with me )
2012/4/19 sara hamza <sarahamz...@gmail.com> > Can Anyone tell me what this message mean please ?? is it a bad > configuration of GIZA++ ?? > > *Hmm: Iteration 1* > *ERROR: Execution of: /home/user/bin/GIZA++ -CoocurrenceFile > ./giza.stand-dia/stand-dia.cooc -c ./corpus/stand-dia-int-train.snt -m1 5 > -m2 0 -m3 3 -m4 3 -model1dumpfrequency 1 -model4smoothfactor 0.4 -nodumps 1 > -nsmooth 4 -o ./giza.stand-dia/stand-dia -onlyaldumps 1 -p0 0.999 -s > ./corpus/dia.vcb -t ./corpus/stand.vcb* > * died with signal 11, with coredump* > > > 2012/4/19 sara hamza <sarahamz...@gmail.com> > >> Oh thank you lixianhua . >> >> >> 2012/4/19 lixianhua <lixian...@cn.fujitsu.com> >> >>> ** ** >>> >>> Hi Sara,**** >>> >>> ** ** >>> >>> I think either to train language model on a new corpus or on the >>> training data is OK.**** >>> >>> ** ** >>> >>> *发件人:* moses-support-boun...@mit.edu [mailto: >>> moses-support-boun...@mit.edu] *代表 *sara hamza >>> *发送时间:* 2012年4月19日 17:17 >>> *收件人:* Moses-support@mit.edu >>> *主题:* [Moses-support] Language Model**** >>> >>> ** ** >>> >>> Hello Guys, **** >>> >>> **** >>> >>> In order to build a language model , Am I supposed to use a different >>> corpus from the one I will use in training the translation model ?? Thank >>> you in advance**** >>> >> >> >
_______________________________________________ Moses-support mailing list Moses-support@mit.edu http://mailman.mit.edu/mailman/listinfo/moses-support
