Struct F16 

pub struct F16(pub u16);

IEEE 754 binary16 half-precision float.

Stored as u16. All arithmetic is performed by promoting to f64, computing, then narrowing back. This ensures deterministic behavior regardless of platform, since the f64 path is well-defined.

Tuple Fields

0: u16

Implementations

impl F16

pub const ZERO: F16

Positive zero.

pub const NEG_ZERO: F16

Negative zero.

pub const INFINITY: F16

Positive infinity.

pub const NEG_INFINITY: F16

Negative infinity.

pub const NAN: F16

Canonical NaN.

pub const MAX: F16

Maximum finite value: 65504.0.

pub const MIN_POSITIVE_SUBNORMAL: F16

Minimum positive subnormal.

pub fn to_f64(self) -> f64

Convert f16 to f64.

Handles normals, subnormals, zeros, infinities, and NaNs.

pub fn from_f64(value: f64) -> Self

Convert f64 to f16 (round-to-nearest-even).

Handles overflow to infinity, underflow to zero, and subnormals.

pub fn from_f32(value: f32) -> Self

Convert f32 to f16.

pub fn to_f32(self) -> f32

Convert f16 to f32.

pub fn is_nan(self) -> bool

Check if this is NaN.

pub fn is_infinite(self) -> bool

Check if this is infinite.

pub fn is_finite(self) -> bool

Check if this is finite (not NaN or Inf).

pub fn is_subnormal(self) -> bool

Check if this is subnormal (exponent == 0, mantissa != 0).

pub fn add(self, rhs: Self) -> Self

Add two f16 values. Promotes both to f64, adds, then narrows back.

pub fn sub(self, rhs: Self) -> Self

Subtract rhs from self. Promotes both to f64, subtracts, then narrows back.

pub fn mul(self, rhs: Self) -> Self

Multiply two f16 values. Promotes both to f64, multiplies, then narrows back.

pub fn div(self, rhs: Self) -> Self

Divide self by rhs. Promotes both to f64, divides, then narrows back.

pub fn neg(self) -> Self

Negate by toggling the sign bit. Does not promote to f64.

Trait Implementations

impl Clone for F16

fn clone(&self) -> F16

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for F16

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

Formats the value using the given formatter.

impl Display for F16

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

Formats the value using the given formatter.

impl Hash for F16

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 F16

fn eq(&self, other: &F16) -> 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 Copy for F16

impl Eq for F16

impl StructuralPartialEq for F16