Skip to main content

VectorStore

lance_index::vector::storage

Trait VectorStore 

Source 
pub trait VectorStore:
    Send
    + Sync
    + Sized
    + Clone {
    type DistanceCalculator<'a>: DistCalculator
       where Self: 'a;


Show 14 methods    // Required methods
    fn as_any(&self) -> &dyn Any;
    fn schema(&self) -> &SchemaRef;
    fn to_batches(&self) -> Result<impl Iterator<Item = RecordBatch> + Send>;
    fn len(&self) -> usize;
    fn distance_type(&self) -> DistanceType;
    fn row_id(&self, id: u32) -> u64;
    fn row_ids(&self) -> impl Iterator<Item = &u64>;
    fn append_batch(
        &self,
        batch: RecordBatch,
        vector_column: &str,
    ) -> Result<Self>;
    fn dist_calculator(
        &self,
        query: ArrayRef,
        dist_q_c: f32,
    ) -> Self::DistanceCalculator<'_>;
    fn dist_calculator_from_id(&self, id: u32) -> Self::DistanceCalculator<'_>;

    // Provided methods
    fn is_empty(&self) -> bool { ... }
    fn dist_calculator_with_scratch<'a>(
        &'a self,
        query: ArrayRef,
        dist_q_c: f32,
        _residual: Option<QueryResidual<'a>>,
        _f32_scratch: &'a mut Vec<f32>,
        _options: DistanceCalculatorOptions,
    ) -> Self::DistanceCalculator<'a> { ... }
    fn dist_between(&self, u: u32, v: u32) -> f32 { ... }
    fn prefers_candidate(
        &self,
        candidate: &OrderedNode,
        selected: &[OrderedNode],
    ) -> bool { ... }

}
Expand description

Vector Storage is the abstraction to store the vectors.

It can be in-memory or on-disk, raw vector or quantized vectors.

It abstracts away the logic to compute the distance between vectors.

TODO: should we rename this to “VectorDistance”?;

Internal API

API stability is not guaranteed

Required Associated Types§

Source

type DistanceCalculator<'a>: DistCalculator where Self: 'a

Required Methods§

Source

fn as_any(&self) -> &dyn Any

Source

fn schema(&self) -> &SchemaRef

Source

fn to_batches(&self) -> Result<impl Iterator<Item = RecordBatch> + Send>

Source

fn len(&self) -> usize

Source

fn distance_type(&self) -> DistanceType

Return DistanceType.

Source

fn row_id(&self, id: u32) -> u64

Get the lance ROW ID from one vector.

Source

fn row_ids(&self) -> impl Iterator<Item = &u64>

Source

fn append_batch(&self, batch: RecordBatch, vector_column: &str) -> Result<Self>

Append Raw RecordBatch into the Storage. The storage implement will perform quantization if necessary.

Source

fn dist_calculator( &self, query: ArrayRef, dist_q_c: f32, ) -> Self::DistanceCalculator<'_>

Create a DistCalculator to compute the distance between the query.

Using dist calculator can be more efficient as it can pre-compute some values.

Source

fn dist_calculator_from_id(&self, id: u32) -> Self::DistanceCalculator<'_>

Provided Methods§

Source

fn is_empty(&self) -> bool

Returns true if this graph is empty.

Source

fn dist_calculator_with_scratch<'a>( &'a self, query: ArrayRef, dist_q_c: f32, _residual: Option<QueryResidual<'a>>, _f32_scratch: &'a mut Vec<f32>, _options: DistanceCalculatorOptions, ) -> Self::DistanceCalculator<'a>

Create a DistCalculator, reusing caller-owned scratch for query-time precomputed state when the storage supports it.

Source

fn dist_between(&self, u: u32, v: u32) -> f32

Source

fn prefers_candidate( &self, candidate: &OrderedNode, selected: &[OrderedNode], ) -> bool

Dyn Compatibility§

This trait is not dyn compatible.

In older versions of Rust, dyn compatibility was called "object safety".

Implementors§