Struct HydraByteTokenizer 

pub struct HydraByteTokenizer { /* private fields */ }

Byte-level tokenizer that matches Hydra’s training tokenizer.

Uses the same encoding as the Python SimpleTokenizer:

• PAD = 0, EOS = 1, BOS = 2
• Byte values 0-255 map to token IDs 3-258
• Sequences are wrapped with BOS and EOS tokens

Implementations

impl HydraByteTokenizer

pub const PAD_TOKEN_ID: u32 = 0

PAD token ID

pub const EOS_TOKEN_ID: u32 = 1

EOS token ID

pub const BOS_TOKEN_ID: u32 = 2

BOS token ID

pub const BYTE_OFFSET: u32 = 3

Offset for byte values (first 3 IDs reserved for special tokens)

pub fn new() -> Self

Create new Hydra byte tokenizer with default max length (512).

pub fn with_max_length(max_length: usize) -> Self

Create tokenizer with custom max length.

Trait Implementations

impl Clone for HydraByteTokenizer

fn clone(&self) -> HydraByteTokenizer

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for HydraByteTokenizer

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

Formats the value using the given formatter.

impl Default for HydraByteTokenizer

fn default() -> Self

Returns the “default value” for a type.

impl HydraTokenizer for HydraByteTokenizer

fn encode(&self, text: &str) -> Result<Vec<u32>>

Encode text to token IDs.

fn decode(&self, tokens: &[u32]) -> Result<String>

Decode token IDs back to text.

fn vocab_size(&self) -> usize

Get the vocabulary size.

fn tokenizer_type(&self) -> TokenizerType

Get the tokenizer type.

fn truncate(&self, tokens: Vec<u32>) -> Vec<u32>

Truncate tokens to maximum length for Hydra.

fn encode_for_hydra(&self, text: &str) -> Result<Vec<u32>>

Encode and truncate for Hydra input.