Enum QuantParam 

pub enum QuantParam {
    F32,
    F16,
    BF16,
    UE8M0,
    UE4M3,
}

Quantization floating-point precision.

This is used to represent the floating-point precision of quantization parameters like the scale(s) or the accumulation precision used during operations like matrix multiplication.

Variants

F32

Full precision.

F16

Half precision.

BF16

bfloat16 precision.

UE8M0

unsigned floating point, e8m0 format.

UE4M3

unsigned floating point, e4m3 format.

Implementations

impl QuantParam

pub fn max_representable(&self) -> f32

The largest finite value representable by the parameter type.

A two-level scheme picks its per-tensor scale so that the largest block scale lands here, which is what keeps the block scales inside the range their type can express. That recipe only holds for a block param narrower than the scale it divides: dividing by QuantParam::F32’s or QuantParam::UE8M0’s maximum drives the per-tensor scale subnormal and the renormalized block scales to infinity. A two-level scheme has nothing to gain from those params anyway, since their block scales already reach the full range.

pub fn round_up(&self, scale: f32) -> Option<f32>

The smallest value representable by the parameter type that is not below scale.

Storing a quantization scale wants this rather than the nearest value. Rounding down puts the scale below what calibration asked for, so every value at the block maximum clips to the quantization range; rounding up costs one step of coarseness instead. Backends have to agree on this, or a tensor quantized on one reconstructs differently on another.

This is not a cast. Conversion to these types rounds to nearest, which is what a cast should do; this is the storage policy for a scale specifically.

scale must not be negative. Symmetric quantization only produces non-negative scales, and the stepping below walks away from zero for a negative input.

QuantParam::UE8M0 answers None. Its minimum is 2^-127, subnormal in f32, where the grid below no longer holds.

pub fn f32_grid(&self) -> F32Grid

The param’s grid, expressed on the f32 bit pattern. See F32Grid.

bf16 reports no subnormal range because it does not need the separate treatment: its pattern is f32’s top half all the way down, so the bit step stays right where the others stop. Its own subnormals start at 2^-133, which is subnormal in f32 too and flushed to zero by most backends.

Panics

For QuantParam::F32, which is the grid itself, and QuantParam::UE8M0, which is not yet supported.

Trait Implementations

impl Clone for QuantParam

fn clone(&self) -> QuantParam

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Copy for QuantParam

impl Debug for QuantParam

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

Formats the value using the given formatter.

impl<'de> Deserialize<'de> for QuantParam

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

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl Eq for QuantParam

impl Hash for QuantParam

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 Ord for QuantParam

fn cmp(&self, other: &QuantParam) -> Ordering

This method returns an Ordering between self and other.

fn max(self, other: Self) -> Self

where Self: Sized,

Compares and returns the maximum of two values.

fn min(self, other: Self) -> Self

where Self: Sized,

Compares and returns the minimum of two values.

fn clamp(self, min: Self, max: Self) -> Self

where Self: Sized,

Restrict a value to a certain interval.

impl PartialEq for QuantParam

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

Equality operator ==.

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

Inequality operator !=.

impl PartialOrd for QuantParam

fn partial_cmp(&self, other: &QuantParam) -> Option<Ordering>

This method returns an ordering between self and other values if one exists.

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

Tests less than (for self and other) and is used by the < operator.

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

Tests less than or equal to (for self and other) and is used by the <= operator.

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

Tests greater than (for self and other) and is used by the > operator.

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

Tests greater than or equal to (for self and other) and is used by the >= operator.

impl Serialize for QuantParam

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

where __S: Serializer,

Serialize this value into the given Serde serializer.

impl StructuralPartialEq for QuantParam