Enum GradFlow 

pub enum GradFlow {
    Propagates,
    Severs,
    LayoutDependent,
    Unknown,
}

How an op affects gradient connectivity relative to its tensor inputs.

Variants

Propagates

Gradients flow through all tensor operands (ordinary differentiable op).

Severs

Explicitly cuts the autograd graph (detach, apply_op*_no_bwd, …).

LayoutDependent

Gradient exists only under layout/contiguity assumptions the analyzer cannot prove.

Unknown

Not enough information.

Trait Implementations

impl Clone for GradFlow

fn clone(&self) -> GradFlow

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Copy for GradFlow

impl Debug for GradFlow

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

Formats the value using the given formatter.

impl Eq for GradFlow

impl PartialEq for GradFlow

fn eq(&self, other: &GradFlow) -> bool

Equality operator ==.

fn ne(&self, other: &Rhs) -> bool

Inequality operator !=.

impl Serialize for GradFlow

fn serialize<__S>(&self, __serializer: __S) -> Result<__S::Ok, __S::Error>

where __S: Serializer,

Serialize this value into the given Serde serializer.

impl StructuralPartialEq for GradFlow