.TH "MPSCNNConvolutionGradientNode" 3 "Mon Jul 9 2018" "Version MetalPerformanceShaders-119.3" "MetalPerformanceShaders.framework" \" -*- nroff -*-
.ad l
.nh
.SH NAME
MPSCNNConvolutionGradientNode
.SH SYNOPSIS
.br
.PP
.PP
\fC#import <MPSNNGraphNodes\&.h>\fP
.PP
Inherits \fBMPSNNGradientFilterNode\fP, and <MPSNNTrainableNode>\&.
.SS "Instance Methods"

.in +1c
.ti -1c
.RI "(nonnull instancetype) \- \fBinitWithSourceGradient:sourceImage:convolutionGradientState:weights:\fP"
.br
.in -1c
.SS "Class Methods"

.in +1c
.ti -1c
.RI "(nonnull instancetype) + \fBnodeWithSourceGradient:sourceImage:convolutionGradientState:weights:\fP"
.br
.in -1c
.SS "Additional Inherited Members"
.SH "Method Documentation"
.PP 
.SS "\- (nonnull instancetype) initWithSourceGradient: (\fBMPSNNImageNode\fP *__nonnull) sourceGradient(\fBMPSNNImageNode\fP *__nonnull) sourceImage(\fBMPSCNNConvolutionGradientStateNode\fP *__nonnull) gradientState(nullable id< \fBMPSCNNConvolutionDataSource\fP >) weights"
\fBA\fP node to represent the gradient calculation for convolution training\&. 
.PP
\fBParameters:\fP
.RS 4
\fIsourceGradient\fP The input gradient from the 'downstream' gradient filter\&. Often that is a neuron gradient filter node\&. 
.br
\fIsourceImage\fP The input image from the forward convolution node 
.br
\fIgradientState\fP The gradient state from the forward convolution 
.br
\fIweights\fP The data source from the forward convolution\&. It may not contain an integrated neuron\&. Similary, any normalization should be broken out into a separate node\&. Pass nil to use the weights from the forward convolution pass\&. 
.RE
.PP
\fBReturns:\fP
.RS 4
\fBA\fP \fBMPSCNNConvolutionGradientNode\fP 
.RE
.PP

.SS "+ (nonnull instancetype) nodeWithSourceGradient: (\fBMPSNNImageNode\fP *__nonnull) sourceGradient(\fBMPSNNImageNode\fP *__nonnull) sourceImage(\fBMPSCNNConvolutionGradientStateNode\fP *__nonnull) gradientState(nullable id< \fBMPSCNNConvolutionDataSource\fP >) weights"
\fBA\fP node to represent the gradient calculation for convolution training\&. 
.PP
\fBParameters:\fP
.RS 4
\fIsourceGradient\fP The input gradient from the 'downstream' gradient filter\&. Often that is a neuron gradient filter node\&. 
.br
\fIsourceImage\fP The input image from the forward convolution node 
.br
\fIgradientState\fP The gradient state from the forward convolution 
.br
\fIweights\fP The data source from the forward convolution\&. It may not contain an integrated neuron\&. Similary, any normalization should be broken out into a separate node\&. Pass nil to use the weights from the forward convolution pass\&. 
.RE
.PP
\fBReturns:\fP
.RS 4
\fBA\fP \fBMPSCNNConvolutionGradientNode\fP 
.RE
.PP


.SH "Author"
.PP 
Generated automatically by Doxygen for MetalPerformanceShaders\&.framework from the source code\&.