Mbart-50 language codes

I am having trouble understanding what the various language codes used in the model are. For example when I look at this example, I see tgt_lang="ro_RO".
Is there a list of language codes that tells me ro_RO → Romania, en_US → English (US) etc.

I see from here that this list extends to 25 languages:
langs=ar_AR,cs_CZ,de_DE,en_XX,es_XX,et_EE,fi_FI,fr_XX,gu_IN,hi_IN,it_IT,ja_XX,kk_KZ,ko_KR,lt_LT,lv_LV,my_MM,ne_NP,nl_XX,ro_RO,ru_RU,si_LK,tr_TR,vi_VN,zh_CN

Anyone can tell me what each of the above languages maps upto? And what the remaining 25 other languages codes are for mbart-50? And what languages they map upto as well?

Any help appreciated.

1 Like

Hi I am having the same issue, could you find the complete list?