Enum QuantizationConfig 

pub enum QuantizationConfig {
    Int8 {
        scale: f32,
    },
}

Quantization strategies for embeddings.

Quantization reduces memory usage by converting f32 embeddings to smaller types.

Variants

Int8

8-bit integer quantization using a linear scale factor.

Values are computed as: quantized = (value * scale).clamp(-128.0, 127.0) as i8

Fields

scale: f32

The scaling factor for quantization (default: 100.0).

Implementations

impl QuantizationConfig

pub fn scale(&self) -> f32

pub fn with_scale(self, scale: f32) -> Self

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 Default for QuantizationConfig

fn default() -> Self

Returns the “default value” for a type.