Trait wyrm::Node

pub trait Node: Debug + 'static {
    type Value;
    type InputGradient;
    fn forward(&self);
    fn backward(&self, _: &Ref<Self::InputGradient>);
    fn value(&self) -> Bor<Self::Value>;
    fn needs_gradient(&self) -> bool;
    fn clear(&self);
}

Trait representing a computation node. Structs implementing this trait can be used as elements of the computation graph.

Associated Types

type Value

Type of the node's value.

type InputGradient

Type of the input gradient the node receives during backpropagation.

Required Methods

fn forward(&self)

Perform the forward step. Should recursively call the forward methods of its ancestors.

fn backward(&self, _: &Ref<Self::InputGradient>)

Perform the backward step. Should recursively call the backward methods of its ancestors.

fn value(&self) -> Bor<Self::Value>

Return the value of the node.

fn needs_gradient(&self) -> bool

If the node needs to be used in the backward step.

fn clear(&self)

Reset the caches of this node and its parents.

Implementations on Foreign Types

impl Node for Rc<Node<Value = Arr, InputGradient = Arr>>

type Value = Arr

type InputGradient = Arr

fn forward(&self)

fn backward(&self, gradient: &Ref<Self::InputGradient>)

fn value(&self) -> Bor<Self::Value>

fn needs_gradient(&self) -> bool

fn clear(&self)

Implementors

• impl<LHS> Node for SparseCategoricalCrossentropyNode<LHS> where     LHS: Node<Value = Arr, InputGradient = Arr>,  type Value = Arr; type InputGradient = Arr;

• impl Node for InputNode type Value = Arr; type InputGradient = Arr;

• impl Node for ParameterNode type Value = Arr; type InputGradient = Arr;

• impl Node for IndexInputNode type Value = SmallVec<[usize; 4]>; type InputGradient = Arr;