core/embedding/tokenizer_adapter library
Classes
- EmbeddingTokenizer
-
Turns raw text (with a TaskType prefix already applied) into a
TokenizedInput. One instance per loaded tokenizer file; built once by
the worker at startup and reused for every
encodecall. - TokenizedInput
- One text's tokenized form, ready for EmbeddingForwardPass.run.
Typedefs
-
EmbeddingTokenizerFactory
= Future<
EmbeddingTokenizer> Function(String tokenizerPath) - Factory that builds an EmbeddingTokenizer for a given on-disk tokenizer path.