Struct TosaFp8ToBfloat16Case 

pub struct TosaFp8ToBfloat16Case {
    pub name: &'static str,
    pub input_dtype: PackedDType,
    pub artifact: &'static [u8],
    pub input: PackedTensor,
    pub output: Bfloat16Tensor,
}

A stable explicit FP8 → BF16 CAST with a bit-exact output oracle.

Fields

name: &'static str

Diagnostic case name.

input_dtype: PackedDType

Source FP8 storage encoding.

artifact: &'static [u8]

TOSA 1.0 FlatBuffer payload.

input: PackedTensor

Single graph-visible FP8 input.

output: Bfloat16Tensor

Single graph-visible BF16 output.

Implementations

impl TosaFp8ToBfloat16Case

pub fn output_matches(self, actual: &[u16]) -> bool

Compare BF16 output bits exactly, allowing only NaN payload canonicalization.

Trait Implementations

impl Clone for TosaFp8ToBfloat16Case

fn clone(&self) -> TosaFp8ToBfloat16Case

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Copy for TosaFp8ToBfloat16Case

impl Debug for TosaFp8ToBfloat16Case

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

Formats the value using the given formatter.

impl Eq for TosaFp8ToBfloat16Case

impl PartialEq for TosaFp8ToBfloat16Case

fn eq(&self, other: &TosaFp8ToBfloat16Case) -> 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 StructuralPartialEq for TosaFp8ToBfloat16Case