Hi,
I am trying to re-write the script train-factored-phrase-model.perl to use
pgiza (parallel implementation of giza) instead of giza. The script
generates word alignment from files
./giza.fr-en/fr-en.A3.final.{bz2,gz}
./giza.en-fr/en-fr.A3.final.{bz2,gz}
(where fr was used with --f option and en was used with --e option).
However there are no files containing A3 in their names in the output of
pgiza. The output files of pgiza are:
a4.final
ahmm.final
align.final
d4.final
hhmm.final
hhmm.final.alpha
hhmm.final.beta
n4.final
p0.final
t1.final
t4.final
t4.final.inv
thmm.final
How can I get the files ./giza.fr-en/fr-en.A3.final.{bz2,gz} and
./giza.en-fr/en-fr.A3.final.{bz2,gz} from the output of pgiza ?
Thank you,
Michael.
_______________________________________________
Moses-support mailing list
[email protected]
http://mailman.mit.edu/mailman/listinfo/moses-support
