Question about tokenizer
#1
by
LukeYang
- opened
May I ask how did you generate the .spm file for the MarianTokenizer? I'm trying to train a google/sentencepiece model, but it only returns the .model and .vocab file. Should I convert them manually? or is there any other methods.
LukeYang
changed discussion status to
closed
LukeYang
changed discussion status to
open
LukeYang
changed discussion status to
closed