Struct BasicBPETokenizer 

pub struct BasicBPETokenizer { /* private fields */ }

A basic Byte-Pair Encoding tokenizer.

Implementations

impl BasicBPETokenizer

pub fn new() -> Self

Creates a new BPE tokenizer.

pub fn train(&mut self, text: &str, num_merges: usize)

Trains the BPE tokenizer on text.

pub fn get_vocab(&self) -> &[String]

Returns the vocabulary.

Trait Implementations

impl Clone for BasicBPETokenizer

fn clone(&self) -> BasicBPETokenizer

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for BasicBPETokenizer

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl Default for BasicBPETokenizer

fn default() -> Self

Returns the “default value” for a type.

impl Tokenizer for BasicBPETokenizer

fn tokenize(&self, text: &str) -> Vec<String>

Tokenizes a string into tokens.

fn encode(&self, text: &str, vocab: &Vocab) -> Vec<usize>

Tokenizes and encodes to indices using a vocabulary.