ํ ํฌ๋์ด์ ๋ฅผ ์ํ ์ ํธ๋ฆฌํฐ [[utilities-for-tokenizers]]
์ด ํ์ด์ง๋ ํ ํฌ๋์ด์ ์์ ์ฌ์ฉ๋๋ ๋ชจ๋ ์ ํธ๋ฆฌํฐ ํจ์๋ค์ ๋์ดํ๋ฉฐ, ์ฃผ๋ก [PreTrainedTokenizer]์ [PreTrainedTokenizerFast] ์ฌ์ด์ ๊ณตํต ๋ฉ์๋๋ฅผ ๊ตฌํํ๋ [~tokenization_utils_base.PreTrainedTokenizerBase] ํด๋์ค์ [~tokenization_utils_base.SpecialTokensMixin]์ ๋ค๋ฃน๋๋ค.
์ด ํจ์๋ค ๋๋ถ๋ถ์ ๋ผ์ด๋ธ๋ฌ๋ฆฌ์ ํ ํฌ๋์ด์ ์ฝ๋๋ฅผ ์ฐ๊ตฌํ ๋๋ง ์ ์ฉํฉ๋๋ค.
PreTrainedTokenizerBase [[transformers.PreTrainedTokenizerBase]]
[[autodoc]] tokenization_utils_base.PreTrainedTokenizerBase
- call
- all
SpecialTokensMixin [[transformers.SpecialTokensMixin]]
[[autodoc]] tokenization_utils_base.SpecialTokensMixin
Enums ๋ฐ namedtuples [[transformers.tokenization_utils_base.TruncationStrategy]]
[[autodoc]] tokenization_utils_base.TruncationStrategy
[[autodoc]] tokenization_utils_base.CharSpan
[[autodoc]] tokenization_utils_base.TokenSpan