On 2016-03-05 19:58, Ayah ElMaghraby wrote: > nohup nice ~/mosesdecoder/scripts/training/train-model.perl > --target-syntax -root-dir train -corpus ~/Documents/trainData/mosesTree > -f en -e ar -hierarchical -glue-grammar -max-phrase-length 5 > --score-options="--GoodTuring" -lm > 0:3:/home/ayah/Documents/lm/truecase.en-ar.blm.ar:8 -external-bin-dir > ~/mosesdecoder/tools/mgizapp -mgiza -mgiza-cpus 2 -snt2cooc snt2cooc.pl > -parallel -sort-batch-size 253 -sort-compress gzip --giza-option > m1=5,m2=2,mh=0,m3=3,m4=3 > training.out
As far as I can tell, not understanding much of this code, the error means that there's an invalid field in one of the alignment files. The code expects each field in there to be two values, for source/target language, separated by a dash. And it's finding a field where that dash is missing. Did the file get truncated, e.g. because of a full disk? (I honestly don't know what I'm talking about here, but this little bit might help someone else find the problem more easily...) Jeroen _______________________________________________ Moses-support mailing list Moses-support@mit.edu http://mailman.mit.edu/mailman/listinfo/moses-support