Struct QuantizedLinearDescriptor 

pub struct QuantizedLinearDescriptor {
    pub m: i32,
    pub c_out: i32,
    pub k: i32,
    pub q_min: i32,
    pub q_max: i32,
    pub activation_element: ElementKind,
    pub weight_element: ElementKind,
}

Descriptor for a quantized_linear op.

Fields

m: i32

Number of token rows in the activation (and rows of the output).

c_out: i32

Number of output channels (rows of weight_q, cols of output).

k: i32

Inner reduction dim (cols of activation and weight_q).

q_min: i32

Activation quantization range lower bound (symmetric: -127).

q_max: i32

Activation quantization range upper bound (symmetric: 127).

activation_element: ElementKind

Activation FP element kind. Must match TIn::KIND.

weight_element: ElementKind

Weight int element kind. Today wired only for S8.

Trait Implementations

impl Clone for QuantizedLinearDescriptor

fn clone(&self) -> QuantizedLinearDescriptor

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Copy for QuantizedLinearDescriptor

impl Debug for QuantizedLinearDescriptor

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

Formats the value using the given formatter.