Trait SparseEmbeddingModel 

pub trait SparseEmbeddingModel: ModelInfo {
    // Required methods
    fn embed<'life0, 'life1, 'life2, 'async_trait>(
        &'life0 self,
        texts: &'life1 [&'life2 str],
    ) -> Pin<Box<dyn Future<Output = Result<SparseEmbedResult>> + Send + 'async_trait>>
       where Self: 'async_trait,
             'life0: 'async_trait,
             'life1: 'async_trait,
             'life2: 'async_trait;
    fn vocab_size(&self) -> u32;

    // Provided method
    fn warmup<'life0, 'async_trait>(
        &'life0 self,
    ) -> Pin<Box<dyn Future<Output = Result<()>> + Send + 'async_trait>>
       where Self: Sync + 'async_trait,
             'life0: 'async_trait { ... }
}

A model that produces learned-sparse term-weight vectors from text.

Use cases include SPLADE-style retrieval and the BGE-M3 sparse head for hybrid dense + sparse search. One SparseVector is returned per input text. For scoring two sparse vectors, see sparse_dot.

Required Methods

fn embed<'life0, 'life1, 'life2, 'async_trait>( &'life0 self, texts: &'life1 [&'life2 str], ) -> Pin<Box<dyn Future<Output = Result<SparseEmbedResult>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait, 'life1: 'async_trait, 'life2: 'async_trait,

Embed a batch of text strings into learned-sparse vectors.

Returns one SparseVector per input, in input order.

Errors

Returns an error if tokenization fails, the model session errors, or the upstream API rejects the request.

fn vocab_size(&self) -> u32

The size of the term space the produced term_ids index into.

For SPLADE this is the model’s full vocabulary (term expansion); for the BGE-M3 lexical head it is the tokenizer vocabulary the input tokens map to.

Provided Methods

fn warmup<'life0, 'async_trait>( &'life0 self, ) -> Pin<Box<dyn Future<Output = Result<()>> + Send + 'async_trait>>

where Self: Sync + 'async_trait, 'life0: 'async_trait,

Optional warmup hook (e.g. load weights into memory on first access). The default is a no-op.

Errors

Returns an error if the underlying model fails to initialize.

Implementors

impl SparseEmbeddingModel for InstrumentedSparseEmbeddingModel