Struct HashEmbedder 

pub struct HashEmbedder { /* private fields */ }

SimHash-based embedder that produces deterministic embeddings locally.

Algorithm

For each word in the input text:

1. Hash the word with FnvHasher (FNV-1a, deterministic cross-platform).
2. For each dimension i in 0..dimension, check bit (i % 64) of the hash XORed with i. If the bit is set, add +1.0 to that dimension; otherwise add -1.0.
3. After processing all words, normalize the vector to unit length (L2 norm).

The default dimension is 384.

Implementations

impl HashEmbedder

pub fn new(dimension: usize) -> Self

Create a new HashEmbedder with the specified embedding dimension.

pub fn default_dimension() -> Self

Create a HashEmbedder with the default dimension of 384.

pub fn compute_embedding(&self, text: &str) -> Vec<f32>

Compute the SimHash embedding synchronously.

This is the underlying synchronous implementation. The async Embedder::embed trait method delegates to this.

Trait Implementations

impl Embedder for HashEmbedder

fn embed<'life0, 'life1, 'async_trait>( &'life0 self, text: &'life1 str, ) -> Pin<Box<dyn Future<Output = Result<Vec<f32>, EmbeddingError>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait, 'life1: 'async_trait,

Generate a vector embedding for the given text.

fn embed_batch<'life0, 'life1, 'async_trait>( &'life0 self, texts: &'life1 [String], ) -> Pin<Box<dyn Future<Output = Result<Vec<Vec<f32>>, EmbeddingError>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait, 'life1: 'async_trait,

Batch embed multiple texts.

fn dimension(&self) -> usize

Return the dimensionality of embeddings produced by this embedder.

fn name(&self) -> &str

Return the name/identifier of this embedder (e.g. “hash”, “openai-text-embedding-3-small”).