Package io.github.kirstenali.deepj.layers

Class Linear

java.lang.Object

io.github.kirstenali.deepj.layers.Linear

All Implemented Interfaces:

Layer, Projection, Trainable

public final class Linear
extends Object
implements Projection

Fully-connected layer: y = xW + b x: [n x dIn], W: [dIn x dOut], b: [1 x dOut]

Constructor Summary

Constructors

Constructor	Description
Linear(int dIn, int dOut, Random rnd)

Method Summary

Modifier and Type	Method	Description
Tensor	backward(Tensor gradY)
Parameter	bias()
Tensor	forward(Tensor x)
List<Parameter>	parameters()
Parameter	weight()

Methods inherited from class java.lang.Object

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

Methods inherited from interface io.github.kirstenali.deepj.training.Trainable

zeroGrad

Constructor Details

Linear

public Linear(int dIn,
 int dOut,
 Random rnd)

Method Details

forward

public Tensor forward(Tensor x)

Specified by:

forward in interface Layer

backward

public Tensor backward(Tensor gradY)

Specified by:

backward in interface Layer

parameters

public List<Parameter> parameters()

Specified by:

parameters in interface Trainable

weight

public Parameter weight()

bias

public Parameter bias()