sorry, ignore this question for now. I think this is due to some non-printing characters in my dataset
On Thu, Dec 15, 2011 at 2:58 PM, Hieu Hoang <[email protected]>wrote: > hi Qin and Mosers > > has anyone encountered error when running mgizas? I'm using the Moses > scripts to run it and occasionally getting, for example > Starting MGIZA > Initializing Global Paras > DEBUG: EnterDEBUG: PrefixDEBUG: LogParsing Arguments > Opening Log File > Printing parameters > Reading vocabulary file > from:/var/www/html/experiment/english-german-McAfee.v2/baseline/training/prepared.1/de.vcb > Reading vocabulary file > from:/var/www/html/experiment/english-german-McAfee.v2/baseline/training/prepared.1/en.vcb > ERROR: reading vocabulary; 21218 pla 183 > ERROR: reading vocabulary; 27727 fa 105 > > It doesn't happen all the time, but often enough to be annoying > > For the record, this is how mgiza was run > mgizapp -CoocurrenceFile en-de.cooc -c en-de-int-train.snt -m1 5 -m2 0 > -m3 3 -m4 3 -model1dumpfrequency 1 -model4smoothfactor 0.4 -ncpus 4 > -nodumps 1 -nsmooth 4 -o en-de -onlyaldumps 1 -p0 0.999 -s de.vcb -t en.vcb > > >
_______________________________________________ Moses-support mailing list [email protected] http://mailman.mit.edu/mailman/listinfo/moses-support
