Enum Activation 

pub enum Activation {
    Relu,
    Gelu,
    GeluTanh,
    Silu,
    Sigmoid,
    Tanh,
    None,
}

Activation function types supported by DNN kernels.

These correspond to the most common activation functions used in deep learning. Fused activation (e.g. conv + bias + ReLU) avoids extra memory round-trips and is a key optimisation target.

Variants

Relu

Rectified Linear Unit: max(0, x).

Gelu

Gaussian Error Linear Unit (exact): x * Phi(x).

GeluTanh

GELU approximated via tanh.

Silu

Sigmoid Linear Unit (SiLU / Swish): x * sigmoid(x).

Sigmoid

Logistic sigmoid: 1 / (1 + exp(-x)).

Tanh

Hyperbolic tangent.

None

Identity (no activation applied).

Trait Implementations

impl Clone for Activation

fn clone(&self) -> Activation

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Copy for Activation

impl Debug for Activation

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

Formats the value using the given formatter.

impl Eq for Activation

impl Hash for Activation

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 Activation

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