Struct ProductQuantizer 

pub struct ProductQuantizer { /* private fields */ }

Product Quantizer for high compression

Achieves 8-32x compression by dividing vectors into sub-vectors and quantizing each with a codebook.

Implementations

impl ProductQuantizer

pub fn new(dimension: usize, num_subquantizers: usize, bits: u8) -> Result<Self>

Create a new product quantizer

Arguments

• dimension - Vector dimension (must be divisible by num_subquantizers)
• num_subquantizers - Number of sub-quantizers (typically 8, 16, or 32)
• bits - Bits per code (typically 8, giving 256 centroids)

pub fn standard(dimension: usize) -> Result<Self>

Create a standard PQ with 8 sub-quantizers and 8 bits

pub fn train( &mut self, vectors: &[Vec<f32>], max_iterations: usize, ) -> Result<()>

Train the product quantizer using k-means clustering

Arguments

• vectors - Training vectors
• max_iterations - Maximum k-means iterations

pub fn quantize(&self, vector: &[f32]) -> Result<PQCode>

Quantize a vector to PQ codes

pub fn dequantize(&self, code: &PQCode) -> Result<Vec<f32>>

Dequantize PQ codes back to approximate vector

pub fn asymmetric_distance(&self, query: &[f32], code: &PQCode) -> Result<f32>

Compute asymmetric distance (query is not quantized)

This is the preferred method for search as it’s more accurate

pub fn compute_distance_table(&self, query: &[f32]) -> Result<Vec<Vec<f32>>>

Precompute distance tables for fast ADC (Asymmetric Distance Computation)

Returns a table[sq][centroid] = distance from query subvector to centroid

pub fn distance_from_table(&self, table: &[Vec<f32>], code: &PQCode) -> f32

Fast distance computation using precomputed table

pub fn compression_ratio(&self) -> f32

Get compression ratio

pub fn is_trained(&self) -> bool

Check if trained

Trait Implementations

impl Clone for ProductQuantizer

fn clone(&self) -> ProductQuantizer

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for ProductQuantizer

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

Formats the value using the given formatter.

impl<'de> Deserialize<'de> for ProductQuantizer

fn deserialize<__D>(__deserializer: __D) -> Result<Self, __D::Error>

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl Serialize for ProductQuantizer

fn serialize<__S>(&self, __serializer: __S) -> Result<__S::Ok, __S::Error>

where __S: Serializer,

Serialize this value into the given Serde serializer.