if001 commited on
Commit
2faf61e
1 Parent(s): 0e4910a
Files changed (1) hide show
  1. sentencepiece_ja.py +1 -0
sentencepiece_ja.py CHANGED
@@ -28,6 +28,7 @@ class SentencePieceJA(PreTrainedTokenizer):
28
  'unk_token': unk,
29
  'mask_token': mask
30
  })
 
31
 
32
  def get_vocab(self) -> int:
33
  return self._tokenizer.get_vocab()
 
28
  'unk_token': unk,
29
  'mask_token': mask
30
  })
31
+ self._tokenizer.add_tokens([" ", " "])
32
 
33
  def get_vocab(self) -> int:
34
  return self._tokenizer.get_vocab()