deepboof.impl.backward.standard

Class DSpatialWindowChannel<T extends deepboof.Tensor<T>,P extends DSpatialPadding2D<T>>

java.lang.Object

deepboof.impl.forward.standard.BaseFunction<T>

deepboof.impl.forward.standard.BaseSpatialWindow<T,VT>

deepboof.impl.forward.standard.SpatialWindowChannel<T,P>

deepboof.impl.backward.standard.DSpatialWindowChannel<T,P>

All Implemented Interfaces:

deepboof.DFunction<T>, deepboof.Function<T>

Direct Known Subclasses:

DSpatialMaxPooling_F64

public abstract class DSpatialWindowChannel<T extends deepboof.Tensor<T>,P extends DSpatialPadding2D<T>>
extends deepboof.impl.forward.standard.SpatialWindowChannel<T,P>
implements deepboof.DFunction<T>

Backwards functions for operations which convolve a window across the input spatial tensor and process the image in a BCHW (batch, channel, (row, column)) order, e.g. one channel at a time.

Field Summary

Fields

Modifier and Type	Field and Description
protected T	dpadding
protected boolean	learningMode

Fields inherited from class deepboof.impl.forward.standard.SpatialWindowChannel

output

Fields inherited from class deepboof.impl.forward.standard.BaseSpatialWindow

C, config, H, HH, Ho, Hp, N, padding, W, Wo, Wp, WW

Fields inherited from class deepboof.impl.forward.standard.BaseFunction

miniBatchSize, parameters, shapeInput, shapeOutput, shapeParameters

Constructor Summary

Constructors

Constructor and Description
DSpatialWindowChannel(deepboof.forward.ConfigSpatial config, P padding)

Method Summary

Modifier and Type	Method and Description
protected abstract void	_backwards(T input, T dout, T gradientInput, java.util.List<T> gradientParameters)
void	backwards(T input, T dout, T gradientInput, java.util.List<T> gradientParameters)
protected abstract void	backwardsAt_border(P padded, int batch, int channel, int padY, int padX, int outY, int outX) Applies the backwards local window operation.
protected abstract void	backwardsAt_inner(T input, int batch, int channel, int inY, int inX, int outY, int outX) Applies the backwards local window operation.
void	backwardsChannel(T input, T gradientInput) Convolve window across 'input' spatial tensor and compute its gradient
void	evaluating()
java.lang.Class<T>	getTensorType()
boolean	isLearning()
void	learning()

Methods inherited from class deepboof.impl.forward.standard.SpatialWindowChannel

forwardAt_border, forwardAt_inner, forwardChannel

Methods inherited from class deepboof.impl.forward.standard.BaseSpatialWindow

_initialize, getPadding, innerLowerExtent, innerUpperExtent, isEntirelyBorder

Methods inherited from class deepboof.impl.forward.standard.BaseFunction

_forward, _setParameters, forward, getOutputShape, getParameters, getParameterShapes, initialize, setParameters

Methods inherited from class java.lang.Object

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

Methods inherited from interface deepboof.Function

forward, getOutputShape, getParameters, getParameterShapes, initialize, setParameters

Field Detail

learningMode

protected boolean learningMode

dpadding

protected T extends deepboof.Tensor<T> dpadding

Constructor Detail

DSpatialWindowChannel

public DSpatialWindowChannel(deepboof.forward.ConfigSpatial config,
                             P padding)

Method Detail

backwards

public void backwards(T input,
                      T dout,
                      T gradientInput,
                      java.util.List<T> gradientParameters)

Specified by:

backwards in interface deepboof.DFunction<T extends deepboof.Tensor<T>>

_backwards

protected abstract void _backwards(T input,
                                   T dout,
                                   T gradientInput,
                                   java.util.List<T> gradientParameters)

backwardsChannel

public void backwardsChannel(T input,
                             T gradientInput)

Convolve window across 'input' spatial tensor and compute its gradient

Parameters:

input - Input spatial tensor

gradientInput - Storage for input's gradient

backwardsAt_inner

protected abstract void backwardsAt_inner(T input,
                                          int batch,
                                          int channel,
                                          int inY,
                                          int inX,
                                          int outY,
                                          int outX)

Applies the backwards local window operation. The padded gradient (dpadding) should be computed here. The input gradient will be found later on

Parameters:

input - Input spatial tensor

batch - Index of input in mini-batch that is being processed

channel - Channel

inY - y-axis lower extent, in input tensor coordinates

inX - x-axis lower extent, in input tensor coordinates

outY - y-axis output coordinate

outX - x-axis output coordinate

backwardsAt_border

protected abstract void backwardsAt_border(P padded,
                                           int batch,
                                           int channel,
                                           int padY,
                                           int padX,
                                           int outY,
                                           int outX)

Applies the backwards local window operation. The padded gradient (dpadding) should be computed here. The input gradient will be found later on

Parameters:

padded - Input spatial virtual tensor

batch - Index of input in mini-batch that is being processed

channel - Channel

padY - y-axis lower extent, in padded tensor coordinates

padX - x-axis lower extent, in padded tensor coordinates

outY - y-axis output coordinate

outX - x-axis output coordinate

learning

public void learning()

Specified by:

learning in interface deepboof.DFunction<T extends deepboof.Tensor<T>>

evaluating

public void evaluating()

Specified by:

evaluating in interface deepboof.DFunction<T extends deepboof.Tensor<T>>

isLearning

public boolean isLearning()

Specified by:

isLearning in interface deepboof.DFunction<T extends deepboof.Tensor<T>>

getTensorType

public java.lang.Class<T> getTensorType()

Specified by:

getTensorType in interface deepboof.Function<T extends deepboof.Tensor<T>>