laurus 0.5.0

Unified search library for lexical, vector, and semantic retrieval
Documentation 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29

//! Text and multimodal embedding support for Laurus vector search.
//!
//! - Core traits: `Embedder`, `PerFieldEmbedder`, `PrecomputedEmbedder`
//! - Feature flags: `embeddings-candle`, `embeddings-openai`, `embeddings-multimodal`, `embeddings-all`
//! - Vector dimensions must be specified explicitly in field definitions and are not inferred from the embedder
//!
// Unified embedder trait (analogous to Analyzer in lexical module)
pub mod embedder;

// Per-field embedder support (analogous to PerFieldAnalyzer)
pub mod per_field;

// Embedder for pre-computed vectors (analogous to NoOpAnalyzer)
pub mod precomputed;

// Embedder registry for creating embedders from schema definitions
pub mod registry;

// Candle implementation (requires feature flag)
#[cfg(feature = "embeddings-candle")]
pub mod candle_bert_embedder;

// OpenAI implementation (requires feature flag)
#[cfg(feature = "embeddings-openai")]
pub mod openai_embedder;

// Multimodal embedding (requires embeddings-multimodal feature flag)
#[cfg(feature = "embeddings-multimodal")]
pub mod candle_clip_embedder;