Declare __len__ method in PreTrainedTokenizerBase (#21210)

This commit is contained in:
Thomas Wang
2023-01-20 15:54:33 +01:00
committed by GitHub
parent ef53017520
commit 7419d807ff

View File

@@ -1578,6 +1578,9 @@ class PreTrainedTokenizerBase(SpecialTokensMixin, PushToHubMixin):
f" special_tokens={self.special_tokens_map_extended})"
)
def __len__(self) -> int:
raise NotImplementedError()
def get_vocab(self) -> Dict[str, int]:
"""
Returns the vocabulary as a dictionary of token to index.