.TH "MPSRNNSingleGateDescriptor" 3 "Mon Jul 9 2018" "Version MetalPerformanceShaders-119.3" "MetalPerformanceShaders.framework" \" -*- nroff -*-
.ad l
.nh
.SH NAME
MPSRNNSingleGateDescriptor
.SH SYNOPSIS
.br
.PP
.PP
\fC#import <MPSRNNLayer\&.h>\fP
.PP
Inherits \fBMPSRNNDescriptor\fP\&.
.SS "Class Methods"

.in +1c
.ti -1c
.RI "(nonnull instancetype) + \fBcreateRNNSingleGateDescriptorWithInputFeatureChannels:outputFeatureChannels:\fP"
.br
.in -1c
.SS "Properties"

.in +1c
.ti -1c
.RI "id< \fBMPSCNNConvolutionDataSource\fP > \fBinputWeights\fP"
.br
.ti -1c
.RI "id< \fBMPSCNNConvolutionDataSource\fP > \fBrecurrentWeights\fP"
.br
.in -1c
.SH "Detailed Description"
.PP 
This depends on Metal\&.framework  The \fBMPSRNNSingleGateDescriptor\fP specifies a simple recurrent block/layer descriptor\&. The RNN layer initialized with a \fBMPSRNNSingleGateDescriptor\fP transforms the input data (image or matrix), and previous output with a set of filters, each producing one feature map in the new output data\&. The user may provide the RNN unit a single input or a sequence of inputs\&. 
.PP
.nf
Description of operation:

.fi
.PP
.PP
Let x_j be the input data (at time index t of sequence, j index containing quadruplet: batch index, x,y and feature index (x=y=0 for matrices))\&. Let h0_j be the recurrent input (previous output) data from previous time step (at time index t-1 of sequence)\&. Let h1_i be the output data produced at this time step\&.
.PP
Let W_ij, U_ij be the weights for input and recurrent input data respectively Let b_i be a bias term
.PP
Let gi(x) be a neuron activation function
.PP
Then the new output image h1_i data is computed as follows: 
.PP
.nf
h1_i = gi( W_ij * x_j + U_ij * h0_j  + b_i )

.fi
.PP
.PP
The '*' stands for convolution (see \fBMPSRNNImageInferenceLayer\fP) or matrix-vector/matrix multiplication (see \fBMPSRNNMatrixInferenceLayer\fP)\&. Summation is over index j (except for the batch index), but there is no summation over repeated index i - the output index\&. Note that for validity all intermediate images have to be of same size and the U matrix has to be square (ie\&. outputFeatureChannels == inputFeatureChannels in those)\&. Also the bias terms are scalars wrt\&. spatial dimensions\&. 
.SH "Method Documentation"
.PP 
.SS "+ (nonnull instancetype) createRNNSingleGateDescriptorWithInputFeatureChannels: (NSUInteger) inputFeatureChannels(NSUInteger) outputFeatureChannels"
Creates a \fBMPSRNNSingleGateDescriptor\fP 
.PP
\fBParameters:\fP
.RS 4
\fIinputFeatureChannels\fP The number of feature channels in the input image/matrix\&. Must be >= 1\&. 
.br
\fIoutputFeatureChannels\fP The number of feature channels in the output image/matrix\&. Must be >= 1\&. 
.RE
.PP
\fBReturns:\fP
.RS 4
\fBA\fP valid \fBMPSRNNSingleGateDescriptor\fP object or nil, if failure\&. 
.RE
.PP

.SH "Property Documentation"
.PP 
.SS "\- inputWeights\fC [read]\fP, \fC [write]\fP, \fC [nonatomic]\fP, \fC [retain]\fP"
Contains weights 'W_ij', bias 'b_i' and neuron 'gi' from the simple RNN layer formula\&. If nil then assumed zero weights, bias and no neuron (identity mapping)\&. Defaults to nil\&. 
.SS "\- recurrentWeights\fC [read]\fP, \fC [write]\fP, \fC [nonatomic]\fP, \fC [retain]\fP"
Contains weights 'U_ij' from the simple RNN layer formula\&. If nil then assumed zero weights\&. Defaults to nil\&. 

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