I am trying to use a pregenerated dictionary file in language modeling mode, but cannot load it in preprocessing, as the system tells me there is no -src_vocab option in preprocess.lua, although it is listed in the options. I am using OpenNMT 0.9.
[vincent@knox /home/local/opennmt/version0.9/OpenNMT]$ th preprocess.lua -data_type monotext -src_vocab /home/nobackup/DutchLM/NLM/nl.dict -train /home/nobackup/DutchLM/DutchCorpus.txt -valid /home/nobackup/DutchLM/wai-not_nieuwsbrief.tok -save_data dutchcorpus.nlm
preprocess.lua: unknown option -src_vocab
Try 'preprocess.lua -h' for more information, or browse the documentation in the docs/ directory.