Package io.github.kirstenali.deepj.nn

Class Conv2D

java.lang.Object

io.github.kirstenali.deepj.nn.Module

io.github.kirstenali.deepj.nn.Conv2D

All Implemented Interfaces:

Layer, Trainable

public final class Conv2D
extends Module

Cross-correlation of [batch, channels, height, width] inputs, using gather and matrix multiplication.

Constructor Summary

Constructors

Constructor	Description
Conv2D(int inputs, int outputs, int kernel, Random random)
Conv2D(int inputs, int outputs, Window2D window, boolean bias, Random random, ParameterFactory parameters)

Method Summary

Modifier and Type	Method	Description
Parameter	bias()
Tensor	forward(Tensor input)
List<Parameter>	parameters()
Parameter	weight()
Window2D	window()

Methods inherited from class io.github.kirstenali.deepj.nn.Module

buffers, children, eval, isTraining, setTraining, train

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.layers.Layer

forward, supportsIncrementalInference

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

zeroGrad

Constructor Details

Conv2D

public Conv2D(int inputs,
 int outputs,
 int kernel,
 Random random)

Conv2D

public Conv2D(int inputs,
 int outputs,
 Window2D window,
 boolean bias,
 Random random,
 ParameterFactory parameters)

Method Details

weight

public Parameter weight()

bias

public Parameter bias()

window

public Window2D window()

parameters

public List<Parameter> parameters()

Specified by:

parameters in interface Trainable

Overrides:

parameters in class Module

forward

public Tensor forward(Tensor input)