Struct SearchPipeline 

pub struct SearchPipeline<'a> { /* private fields */ }

A gatherer, optionally followed by a reranker.

Stages speak in DocumentKeys and each reads its own indexes, which writers anywhere move independently. A document is ranked only when every stage holds it, so a delete takes effect as soon as any index applies it and an insert once all of them have. Each search reports, as SearchResult::as_of, the oldest commit among the indexes it read.

Each stage must be able to consume the query; that is checked before any stage runs, so a query that cannot be served fails without gathering. Without a reranker the gather scores rank the results.

Implementations

impl<'a> SearchPipeline<'a>

pub fn new( gatherer: Arc<dyn CandidateGenerator + 'a>, reranker: Option<Arc<dyn Reranker + 'a>>, ) -> Self

Stages may borrow state for the life of the pipeline, so a pipeline can be built for one search over state its caller holds; 'static stages make a pipeline that lives as long as needed.

Examples found in repository

examples/stored_maxsim.rs (line 97)

fn main() -> Result<()> {
    let representation = Representation::new("toy-encoder", "1", DIMENSION, true)?;
    let directory = std::env::temp_dir().join(format!("lateweave-example-{}", std::process::id()));
    let path = directory.join("vectors");

    // The indexing side: one token per word, the axis standing in for what
    // a real encoder produces.
    let mut writer =
        VectorStoreWriter::create(&path, Encoding::Float32, "fruit", representation.clone())?;
    let embeddings = [unit(0), unit(2), unit(1), unit(2), unit(0), unit(3)].concat();
    writer.append(["a", "b", "c"], &embeddings, DIMENSION, &[2, 2, 2], None)?;
    writer.commit()?;

    // The serving side.
    let gatherer = TermGatherer {
        requires: BTreeMap::new(),
        documents: vec![("a", "red apple"), ("b", "green apple"), ("c", "red car")],
        indexed_at: SystemTime::now(),
    };
    let store = Arc::new(VectorStore::open(&path)?) as Arc<dyn MultiVectorSource>;
    let reranker = MaxSimReranker::new([store], DEFAULT_FEATURE)?;
    let pipeline = SearchPipeline::new(Arc::new(gatherer), Some(Arc::new(reranker)));

    let query = Query::new("apple").with_feature(
        DEFAULT_FEATURE,
        Feature::lazy(representation, || {
            TokenMatrix::new(unit(1).to_vec(), DIMENSION)
        }),
    );
    let request = SearchRequest::new(10, 2).with_budget(ResourceBudget::new(4096, 64, Some(1))?);

    let print = |label: &str| -> Result<()> {
        let result = pipeline.search(&query, &request)?;
        println!("{label}:");
        for document in &result.documents {
            println!(
                "  #{} {:?} score {}",
                document.rank, document.key, document.score
            );
        }
        println!("  {:?}", result.diagnostics);
        Ok(())
    };
    print("before the delete")?;

    // The indexing side deletes "b" from the vectors; the gatherer still
    // finds it, and the pipeline drops it.
    writer.delete(["b"]);
    writer.commit()?;
    print("after the delete")?;

    std::fs::remove_dir_all(directory)?;
    Ok(())
}

pub fn search( &self, query: &Query, request: &SearchRequest<'_>, ) -> Result<SearchResult>

Examples found in repository

examples/stored_maxsim.rs (line 108)

fn main() -> Result<()> {
    let representation = Representation::new("toy-encoder", "1", DIMENSION, true)?;
    let directory = std::env::temp_dir().join(format!("lateweave-example-{}", std::process::id()));
    let path = directory.join("vectors");

    // The indexing side: one token per word, the axis standing in for what
    // a real encoder produces.
    let mut writer =
        VectorStoreWriter::create(&path, Encoding::Float32, "fruit", representation.clone())?;
    let embeddings = [unit(0), unit(2), unit(1), unit(2), unit(0), unit(3)].concat();
    writer.append(["a", "b", "c"], &embeddings, DIMENSION, &[2, 2, 2], None)?;
    writer.commit()?;

    // The serving side.
    let gatherer = TermGatherer {
        requires: BTreeMap::new(),
        documents: vec![("a", "red apple"), ("b", "green apple"), ("c", "red car")],
        indexed_at: SystemTime::now(),
    };
    let store = Arc::new(VectorStore::open(&path)?) as Arc<dyn MultiVectorSource>;
    let reranker = MaxSimReranker::new([store], DEFAULT_FEATURE)?;
    let pipeline = SearchPipeline::new(Arc::new(gatherer), Some(Arc::new(reranker)));

    let query = Query::new("apple").with_feature(
        DEFAULT_FEATURE,
        Feature::lazy(representation, || {
            TokenMatrix::new(unit(1).to_vec(), DIMENSION)
        }),
    );
    let request = SearchRequest::new(10, 2).with_budget(ResourceBudget::new(4096, 64, Some(1))?);

    let print = |label: &str| -> Result<()> {
        let result = pipeline.search(&query, &request)?;
        println!("{label}:");
        for document in &result.documents {
            println!(
                "  #{} {:?} score {}",
                document.rank, document.key, document.score
            );
        }
        println!("  {:?}", result.diagnostics);
        Ok(())
    };
    print("before the delete")?;

    // The indexing side deletes "b" from the vectors; the gatherer still
    // finds it, and the pipeline drops it.
    writer.delete(["b"]);
    writer.commit()?;
    print("after the delete")?;

    std::fs::remove_dir_all(directory)?;
    Ok(())
}