Commit
•
10f8197
1
Parent(s):
ad15617
Update tokenization_bn.py
Browse files- tokenization_bn.py +2 -1
tokenization_bn.py
CHANGED
@@ -48,10 +48,11 @@ class BNTokenizer(PreTrainedTokenizer):
|
|
48 |
|
49 |
def __init__(self, vocab_file, bos_token=None, eos_token='</s>', unk_token='<unk>', pad_token='<|reserved001|>', sep_token=None, sp_model_kwargs: Optional[Dict[str, Any]]=None, **kwargs) -> None:
|
50 |
self.sp_model_kwargs = {} if sp_model_kwargs is None else sp_model_kwargs
|
51 |
-
super().__init__(bos_token=bos_token, eos_token=eos_token, unk_token=unk_token, pad_token=pad_token, sep_token=sep_token, sp_model_kwargs=self.sp_model_kwargs, **kwargs)
|
52 |
self.vocab_file = vocab_file
|
53 |
self.sp_model = spm.SentencePieceProcessor(**self.sp_model_kwargs)
|
54 |
self.sp_model.Load(vocab_file)
|
|
|
|
|
55 |
|
56 |
@property
|
57 |
def vocab_size(self):
|
|
|
48 |
|
49 |
def __init__(self, vocab_file, bos_token=None, eos_token='</s>', unk_token='<unk>', pad_token='<|reserved001|>', sep_token=None, sp_model_kwargs: Optional[Dict[str, Any]]=None, **kwargs) -> None:
|
50 |
self.sp_model_kwargs = {} if sp_model_kwargs is None else sp_model_kwargs
|
|
|
51 |
self.vocab_file = vocab_file
|
52 |
self.sp_model = spm.SentencePieceProcessor(**self.sp_model_kwargs)
|
53 |
self.sp_model.Load(vocab_file)
|
54 |
+
super().__init__(bos_token=bos_token, eos_token=eos_token, unk_token=unk_token, pad_token=pad_token, sep_token=sep_token, sp_model_kwargs=self.sp_model_kwargs, **kwargs)
|
55 |
+
|
56 |
|
57 |
@property
|
58 |
def vocab_size(self):
|