Struct MockMultiVectorEmbedder 

pub struct MockMultiVectorEmbedder { /* private fields */ }

Mock multi-vector embedder for testing.

Generates deterministic pseudo-random embeddings based on token content. Useful for testing the retrieval pipeline without requiring a real model.

Example

use trueno_rag::multivector::MockMultiVectorEmbedder;
use trueno_rag::multivector::MultiVectorEmbedder;

let embedder = MockMultiVectorEmbedder::new(128, 512);

let emb1 = embedder.embed_tokens("hello world").unwrap();
let emb2 = embedder.embed_tokens("hello world").unwrap();

// Same input produces same output
assert_eq!(emb1.as_slice(), emb2.as_slice());

Implementations

impl MockMultiVectorEmbedder

pub fn new(dim: usize, max_tokens: usize) -> Self

Create a new mock embedder.

Arguments

• dim - Token embedding dimension (e.g., 128 for ColBERT)
• max_tokens - Maximum tokens per document

pub fn with_seed(dim: usize, max_tokens: usize, seed: u64) -> Self

Create with a custom seed for different random sequences.

Trait Implementations

impl Clone for MockMultiVectorEmbedder

fn clone(&self) -> MockMultiVectorEmbedder

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for MockMultiVectorEmbedder

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

Formats the value using the given formatter.

impl MultiVectorEmbedder for MockMultiVectorEmbedder

fn embed_tokens(&self, text: &str) -> Result<MultiVectorEmbedding>

Embed text into token-level vectors.

fn embed_tokens_batch( &self, texts: &[&str], ) -> Result<Vec<MultiVectorEmbedding>>

Batch embed multiple texts.

fn token_dimension(&self) -> usize

Get the token embedding dimension.

fn max_tokens(&self) -> usize

Get the maximum tokens per document.

fn model_id(&self) -> &str

Get the model identifier.