Struct QuantizationConfig 

pub struct QuantizationConfig {
    pub qtype: QuantizationType,
    pub always_ram: bool,
    pub quantile: Option<f64>,
    pub bits: Option<f64>,
    pub compression: Option<String>,
    pub encoding: Option<String>,
    pub query_encoding: Option<String>,
    pub memory: Option<MemoryPlacement>,
}

WITH QUANTIZATION (…) settings for a vector definition.

Fields

qtype: QuantizationType

Quantization family.

always_ram: bool

Legacy flag keeping quantized vectors in RAM (prefer memory).

quantile: Option<f64>

Scalar quantization quantile in [0, 1].

bits: Option<f64>

Bits per dimension (Turbo accepts 1, 1.5, 2, or 4).

compression: Option<String>

Product quantization compression (x4–x64).

encoding: Option<String>

Binary quantization encoding (one_bit, two_bits, one_and_half_bits).

query_encoding: Option<String>

Binary query encoding (default, binary, scalar4bits, scalar8bits).

memory: Option<MemoryPlacement>

Memory placement of quantized vectors.

Trait Implementations

impl Clone for QuantizationConfig

fn clone(&self) -> QuantizationConfig

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for QuantizationConfig

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

Formats the value using the given formatter.

impl PartialEq for QuantizationConfig

fn eq(&self, other: &QuantizationConfig) -> bool

Equality operator ==.

fn ne(&self, other: &Rhs) -> bool

Inequality operator !=.

impl StructuralPartialEq for QuantizationConfig