Enum ActivationKind 

pub enum ActivationKind {
    Relu,
    Gelu,
    Silu,
}

Activation functions implemented by the Bias*Activation EpilogueKind variants. Surfaced for telemetry and selector logic; the kernel selection itself is driven by the enum variant.

Intentionally NOT #[non_exhaustive] — paired with EpilogueKind which is also left exhaustive. Adding a new activation requires shipping a matching Bias<Activation> epilogue kernel, which is a deliberate breaking-change event.

Variants

Relu

relu(x) = max(x, 0).

Gelu

Exact (erf-based) Gaussian Error Linear Unit. Matches PyTorch’s default nn.GELU().

Silu

silu(x) = x · sigmoid(x). Also known as Swish-1.

Trait Implementations

impl Clone for ActivationKind

fn clone(&self) -> ActivationKind

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Copy for ActivationKind

impl Debug for ActivationKind

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

Formats the value using the given formatter.

impl Eq for ActivationKind

impl Hash for ActivationKind

fn hash<__H>(&self, state: &mut __H)

where __H: Hasher,

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 ActivationKind

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