Enum executorch_sys::torch::executor::TensorShapeDynamism

#[repr(u8)]
pub enum TensorShapeDynamism {
    STATIC = 0,
    DYNAMIC_BOUND = 1,
    DYNAMIC_UNBOUND = 2,
}

Rank can never change in executorch. But shape sometimes can. This enum is used to help provide better safety in kernels on what tensors are resizable

WARNING: This abstraction is only temporary. The long term vision is that this wont exist in the runtime. Instead the runtime will support a debug mode that allows for similar fail early patterns without having to pay the runtime cost of directly embedding this abstraction in tensor and performing checks against it during resizing. Adding this debug mode is non trivial though so for the short term this abstraction helps us move fast. TODO(jakeszwe): T134528146

Variants

STATIC = 0

Cannot change shape

DYNAMIC_BOUND = 1

shape cannot exceed initial capacity

DYNAMIC_UNBOUND = 2

No restriction on shape and capacity

Trait Implementations

impl Clone for TensorShapeDynamism

fn clone(&self) -> TensorShapeDynamism

Returns a copy of the value.

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

Performs copy-assignment from source.

impl Debug for TensorShapeDynamism

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

Formats the value using the given formatter.

impl Hash for TensorShapeDynamism

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 TensorShapeDynamism

fn eq(&self, other: &TensorShapeDynamism) -> 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 Eq for TensorShapeDynamism

impl StructuralPartialEq for TensorShapeDynamism