Hi Guchun I think this happened to me before, and I just reran the giza step and it worked. It looks as though the mgiza output got corrupted (race condition?) and so merging the alignments failed,
cheers - Barry On Wednesday 14 March 2012 18:31:50 Guchun Zhang wrote: > Hi, > > I pulled the latest Moses today. Recompiled OK. Then I re-ran some > experiments using EMS to check everything. I got an error at the > run_giza/run_giza_inverse stage. > > Executing: /home/alphac/Work/moses/giza-pp/bin/merge_alignment.py > /home/alphac/Work/exp/exp1/fr-en/working/basic/sge/training/giza.1/en-fr.A > 3.final.part*>/home/alphac/Work/exp/exp1/fr-en/working/basic/sge/training/g > iza.1/en-fr.A3.final ERROR! MISSING ENTRy 5 > ERROR! MISSING ENTRy 7 > ERROR! MISSING ENTRy 8 > ERROR! MISSING ENTRy 12 > ERROR! MISSING ENTRy 14 > ERROR! MISSING ENTRy 19 > ERROR! MISSING ENTRy 21 > ERROR! MISSING ENTRy 22 > ERROR! MISSING ENTRy 23 > ..... > Combined 26 files, totally 76765 sents > > The total number of sge jobs and CPUs in the mgiza setting are both 30. > > What could be the cause of this problem? > > Many thanks, > > Guchun > -- Barry Haddow University of Edinburgh +44 (0) 131 651 3173 -- The University of Edinburgh is a charitable body, registered in Scotland, with registration number SC005336. _______________________________________________ Moses-support mailing list [email protected] http://mailman.mit.edu/mailman/listinfo/moses-support
