How to use new feature:target_subdict

my test script:

th translate.lua -gpuid 1 -replace_unk -phrase_table pb.txt -model model_epoch24_5.73.t7 -target_subdict dict.txt -src src.txt -output result.txt

although I put a lot of different words to dict.txt, but translation quality always is very low.

how to use it?

target_subdict is to be used when using a large vocabulary (I would say over 50k). Then it accelerates your decoding - to initialize it, you can use a phrase table and be very generous when picking up target words. For a 100k target vocabulary, and regular network size, if you provide ~30k target words in the sub-dictionary, you will get about 2x throughput.
You might lose a little bit in quality - but it should not be very low if you extract enough target candidates.

thanks for your answer. I used 1.3 million target words for a new practice,the results are very good, the following is my test results: