Class Linear<T>

java.lang.Object

org.episteme.core.mathematics.ml.neural.layers.Linear<T>

All Implemented Interfaces:

Serializable, Layer<T>

public class Linear<T>
extends Object
implements Layer<T>

Fully connected (linear) layer.

See Also:

• Serialized Form

Constructor Summary

Constructors

Constructor	Description
Linear(int inFeatures, int outFeatures)

Method Summary

Modifier and Type	Method	Description
GraphNode<T>	forward(GraphNode<T> input)	Performs a forward pass using explicit autograd nodes.
Map<String, GraphNode<T>>	getParameters()	Returns the learnable parameters of this layer.

Methods inherited from class Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Methods inherited from interface Layer

forward, getGradients, setTraining

Constructor Details

Linear

public Linear(int inFeatures,
 int outFeatures)

Method Details

forward

public GraphNode<T> forward(GraphNode<T> input)

Description copied from interface: Layer

Performs a forward pass using explicit autograd nodes.

Specified by:

forward in interface Layer<T>

Parameters:

input - the input node.

Returns:

the transformed output node.

getParameters

public Map<String, GraphNode<T>> getParameters()

Description copied from interface: Layer

Returns the learnable parameters of this layer. Keys should be unique within the layer (e.g., "weights", "bias").

Specified by:

getParameters in interface Layer<T>

Returns:

a map of parameter graph nodes.