Enum CalibrationMethod 

pub enum CalibrationMethod {
    MinMax,
    Percentile(f32),
    KlDivergence,
}

Calibration method for determining quantization parameters.

Variants

MinMax

Use the observed min/max values directly.

Percentile(f32)

Clip to the given percentile (e.g. 99.99) to reduce outlier impact.

KlDivergence

Find the optimal clipping threshold that minimizes KL divergence between the original and quantized distributions (TensorRT-style).

Implementations

impl CalibrationMethod

pub fn from_str_name(s: &str) -> Option<Self>

Parse a calibration method from a CLI string.

Trait Implementations

impl Clone for CalibrationMethod

fn clone(&self) -> CalibrationMethod

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for CalibrationMethod

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

Formats the value using the given formatter.

impl PartialEq for CalibrationMethod

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

Equality operator ==.

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

Inequality operator !=.

impl StructuralPartialEq for CalibrationMethod