Tokenizer

FunctionDescription
fromBpe configCreate from a BPE config.
fromSentencePiece configCreate from a SentencePiece config.
fromTiktoken configCreate from a Tiktoken config.
fromWordPiece configCreate from a WordPiece config.
wrap innerWrap an existing Microsoft.ML.Tokenizers.Tokenizer instance.

fromBpe

fromBpe config

Create from a BPE config.

Parameters

  • config : BpeConfig

Returns Tokenizer


fromSentencePiece

fromSentencePiece config

Create from a SentencePiece config.

Parameters

  • config : SentencePieceConfig

Returns Tokenizer


fromTiktoken

fromTiktoken config

Create from a Tiktoken config.

Parameters

  • config : TiktokenConfig

Returns Tokenizer


fromWordPiece

fromWordPiece config

Create from a WordPiece config.

Parameters

  • config : WordPieceConfig

Returns Tokenizer


wrap

wrap inner

Wrap an existing Microsoft.ML.Tokenizers.Tokenizer instance.

Parameters

  • inner : Tokenizer

Returns Tokenizer