Struct LocalOnnxProvider 

pub struct LocalOnnxProvider { /* private fields */ }

Local ONNX-based embedding provider

This is a stub implementation. In production, this would use:

• ort (ONNX Runtime) for model inference
• fastembed-rs for pre-packaged models
• tokenizers for text preprocessing

Implementations

impl LocalOnnxProvider

pub fn new(config: EmbeddingConfig) -> EmbeddingResult<Self>

Create a new local ONNX provider

pub fn load_pretrained(model_name: &str) -> EmbeddingResult<Self>

Load a pre-trained model by name

Trait Implementations

impl Debug for LocalOnnxProvider

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

Formats the value using the given formatter.

impl EmbeddingProvider for LocalOnnxProvider

fn model_name(&self) -> &str

Get the model name

fn dimension(&self) -> usize

Get the embedding dimension

fn max_length(&self) -> usize

Maximum text length (in characters or tokens)

fn embed(&self, text: &str) -> EmbeddingResult<Vec<f32>>

Generate embedding for a single text

fn embed_batch(&self, texts: &[&str]) -> EmbeddingResult<Vec<Vec<f32>>>

Generate embeddings for multiple texts (batch)

fn normalize(&self, embedding: &mut [f32])

Normalize an embedding vector (L2 normalization)