Enum QuantizationScale 

pub enum QuantizationScale {
    PerTensor,
    PerChannel {
        axis: u32,
    },
    PerGroup {
        group_size: u32,
    },
}

Scale metadata layout for a quantized tensor or vector.

Variants

PerTensor

One scale value for the whole buffer.

PerChannel

One scale value per slice along axis.

Fields

axis: u32

Tensor axis carrying independent scale values.

PerGroup

One scale value per contiguous group.

Fields

group_size: u32

Number of logical elements per quantization group.

Trait Implementations

impl Clone for QuantizationScale

fn clone(&self) -> QuantizationScale

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for QuantizationScale

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

Formats the value using the given formatter.

impl<'de> Deserialize<'de> for QuantizationScale

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

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl Display for QuantizationScale

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

Formats the value using the given formatter.

impl Hash for QuantizationScale

fn hash<__H: Hasher>(&self, state: &mut __H)

Feeds this value into the given Hasher.

fn hash_slice<H>(data: &[Self], state: &mut H)

where H: Hasher, Self: Sized,

Feeds a slice of this type into the given Hasher.

impl PartialEq for QuantizationScale

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

Tests for self and other values to be equal, and is used by ==.

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

Tests for !=. The default implementation is almost always sufficient, and should not be overridden without very good reason.

impl Serialize for QuantizationScale

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

where __S: Serializer,

Serialize this value into the given Serde serializer.

impl Eq for QuantizationScale

impl StructuralPartialEq for QuantizationScale