I recently trained the basic model with Hindi-English Data (1.5m+). I obtained a BLEU Score of about 11.15 (Tilde). I know this question is very basic, but is there any way i can improve the BLEU scores by tweaking the features of OpenNMT ?
hello - there is a lot of discussion about this all around the forum: use transformer if you are not already, use more corpus with back translation, augment/normalize your corpus, etc…