Skip to content

SubwordTokenizer

Defined in: parser.ts:31

Minimal sub-word tokenizer contract (satisfied by @huggingface/transformers).

SubwordTokenizer(text, opts): object

Defined in: parser.ts:32

Minimal sub-word tokenizer contract (satisfied by @huggingface/transformers).

string

boolean

object

input_ids: object

data: ArrayLike<number | bigint>

optional cls_token_id?: number

Defined in: parser.ts:33


optional unk_token_id?: number

Defined in: parser.ts:34