Update tokenization_SEA_BPE.py
Browse files- tokenization_SEA_BPE.py +2 -0
tokenization_SEA_BPE.py
CHANGED
@@ -43,6 +43,8 @@ class SEABPETokenizer(PreTrainedTokenizer):
|
|
43 |
more details.
|
44 |
|
45 |
"""
|
|
|
|
|
46 |
|
47 |
def __init__(self, vocab_file, unk_token='<unk>', bos_token=None, eos_token='<|endoftext|>', pad_token=None, sp_model_kwargs: Optional[Dict[str, Any]]=None, add_bos_token=False, add_eos_token=False, clean_up_tokenization_spaces=False, legacy=None, **kwargs):
|
48 |
self.sp_model_kwargs = {} if sp_model_kwargs is None else sp_model_kwargs
|
|
|
43 |
more details.
|
44 |
|
45 |
"""
|
46 |
+
|
47 |
+
vocab_files_names = VOCAB_FILES_NAMES
|
48 |
|
49 |
def __init__(self, vocab_file, unk_token='<unk>', bos_token=None, eos_token='<|endoftext|>', pad_token=None, sp_model_kwargs: Optional[Dict[str, Any]]=None, add_bos_token=False, add_eos_token=False, clean_up_tokenization_spaces=False, legacy=None, **kwargs):
|
50 |
self.sp_model_kwargs = {} if sp_model_kwargs is None else sp_model_kwargs
|