Enum Dtype 

pub enum Dtype {
    F32,
    F64,
    F16,
    BF16,
    U8,
    U16,
    U32,
    U64,
    I8,
    I16,
    I32,
    I64,
    Bool,
}

Runtime dtype tag for AnyTensor and any caller that needs to dispatch over the framework’s known tensor dtypes.

Variants

F32

32-bit IEEE 754 float — corresponds to TYPE_TENSOR_F32.

F64

64-bit IEEE 754 float — corresponds to TYPE_TENSOR_F64.

F16

16-bit IEEE 754 half — corresponds to TYPE_TENSOR_F16. Stored as half::f16.

BF16

16-bit bfloat (Google’s brain-float) — corresponds to TYPE_TENSOR_BF16. Stored as half::bf16.

U8

8-bit unsigned — corresponds to TYPE_TENSOR_U8.

U16

16-bit unsigned — corresponds to TYPE_TENSOR_U16.

U32

32-bit unsigned — corresponds to TYPE_TENSOR_U32.

U64

64-bit unsigned — corresponds to TYPE_TENSOR_U64.

I8

8-bit signed — corresponds to TYPE_TENSOR_I8 (edge quantization).

I16

16-bit signed — corresponds to TYPE_TENSOR_I16.

I32

32-bit signed — corresponds to TYPE_TENSOR_I32.

I64

64-bit signed — corresponds to TYPE_TENSOR_I64 (indices, embeddings).

Bool

1-bit boolean — corresponds to TYPE_TENSOR_BOOL (masks, predicates).

Implementations

impl Dtype

pub fn type_node(self) -> &'static TypeNode

Return the framework TypeNode static for this dtype.

Trait Implementations

impl Clone for Dtype

fn clone(&self) -> Dtype

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Copy for Dtype

impl Debug for Dtype

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

Formats the value using the given formatter.

impl Eq for Dtype

impl PartialEq for Dtype

fn eq(&self, other: &Dtype) -> 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 Dtype