Hi, it's really hard to say why GIZA++ is failing; it is typically robust (but not that easy to use).
Try using absolute paths, to eliminate that as a source of error. Try it on a small subset of the corpus first, then use larger subsets. Look carefully at the STDERR and STDOUT to see if there are any hints. Make sure that you are not running out of memory. -phi On Mon, Apr 30, 2012 at 3:15 PM, Info Ic <[email protected]> wrote: > > > Nobody has an answer ???? am really hopeless I couldn't find answers > anywhere else... I reviewed my corpus but the error keeps showing up > ...could it be a problem in GIZA++ ??? or maybe cygwin ???? I don't know > where am'I supposed to look for an answer Am running out of time > ________________________________ > From: [email protected] > To: [email protected] > Subject: Error : died with signal 11, with coredump > Date: Sun, 29 Apr 2012 15:34:58 +0000 > > > Hello everyone > > Hmm: Iteration 1 > ERROR: Execution of: /home/user/bin/GIZA++ -CoocurrenceFile > ./giza.DIA-STAND/DIA-STAND.cooc -c ./corpus/DIA-STAND-int-train.snt -m1 5 > -m2 0 -m3 3 -m4 3 -model1dumpfrequency 1 -model4smoothfactor 0.4 -nodumps 1 > -nsmooth 4 -o ./giza.DIA-STAND/DIA-STAND -onlyaldumps 1 -p0 0.999 -s > ./corpus/STAND.vcb -t ./corpus/DIA.vcb > died with signal 11, with coredump > > > Can Anyone pleaaaaase explain to me what does this error mean??? everytime > I get it I try to verify wether I left some blank lines in my corpus or not > but obviously it is non about that > > _______________________________________________ > Moses-support mailing list > [email protected] > http://mailman.mit.edu/mailman/listinfo/moses-support > _______________________________________________ Moses-support mailing list [email protected] http://mailman.mit.edu/mailman/listinfo/moses-support
