Tokenizer class final

Encode/decode between strings and token ids using a LlamaVocab.

Stateless and cheap to construct. Hold one alongside the model.

Constructors

Tokenizer(LlamaVocab vocab)
const

Properties

hashCode int
The hash code for this object.
no setterinherited
runtimeType Type
A representation of the runtime type of the object.
no setterinherited
vocab LlamaVocab
final

Methods

decode(int token, {bool unparseSpecial = true}) String
Detokenize a single token into its UTF-8 piece.
decodeAll(List<int> tokens, {bool removeSpecial = false, bool unparseSpecial = true}) String
Detokenize a list of tokens into a UTF-8 string.
encode(String text, {bool addSpecial = true, bool parseSpecial = true}) List<int>
Tokenize text.
encodeToken(int token, {bool special = true}) Uint8List
Raw byte form of a single token's piece — useful when streaming, where piece boundaries do not align with UTF-8 codepoints.
noSuchMethod(Invocation invocation) → dynamic
Invoked when a nonexistent method or property is accessed.
inherited
toString() String
A string representation of this object.
inherited

Operators

operator ==(Object other) bool
The equality operator.
inherited