.TH "MPSCNNUpsamplingGradient" 3 "Mon Jul 9 2018" "Version MetalPerformanceShaders-119.3" "MetalPerformanceShaders.framework" \" -*- nroff -*-
.ad l
.nh
.SH NAME
MPSCNNUpsamplingGradient
.SH SYNOPSIS
.br
.PP
.PP
\fC#import <MPSCNNUpsampling\&.h>\fP
.PP
Inherits \fBMPSCNNGradientKernel\fP\&.
.PP
Inherited by \fBMPSCNNUpsamplingBilinearGradient\fP, and \fBMPSCNNUpsamplingNearestGradient\fP\&.
.SS "Instance Methods"

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

.in +1c
.ti -1c
.RI "double \fBscaleFactorX\fP"
.br
.ti -1c
.RI "double \fBscaleFactorY\fP"
.br
.in -1c
.SS "Additional Inherited Members"
.SH "Detailed Description"
.PP 
This depends on Metal\&.framework  The \fBMPSCNNUpsamplingGradient\fP filter is used for training\&. It is the backward filter for the \fBMPSCNNUpsampling\fP filter\&. It operates on the gradient input, specifically, it reduces the size of the gradient input in the x and y dimensions\&.
.PP
The number of output feature channels remains the same as the number of input feature channels\&.
.PP
The scaleFactor must be an integer value >= 1\&. The default value is 1\&. If scaleFactor == 1, the filter acts as a copy kernel\&.
.PP
Nearest and bilinear variants are supported\&.
.PP
For example, for the nearest variant with scaleFactorX = scaleFactorY = 2, the forward pass produced the following output:
.PP
Input: Output: a a b b a b a a b b c d c c d d c c d d
.PP
To upsample the image, the input data is replicated\&.
.PP
And, the backward pass for the above froward pass is computed in the following way:
.PP
Input: Output: a1 a2 b1 b2 a2 a3 b3 b4 x y c1 c2 d1 d2 z w c3 c4 d3 d4
.PP
where x = a1 + a2 + a3 + a4 y = b1 + b2 + b3 + b4 z = c1 + c2 + c3 + c4 w = d1 + d2 + d3 + d4 
.SH "Method Documentation"
.PP 
.SS "\- (nonnull instancetype) initWithDevice: (nonnull id< MTLDevice >) device"
Standard init with default properties per filter type 
.PP
\fBParameters:\fP
.RS 4
\fIdevice\fP The device that the filter will be used on\&. May not be NULL\&. 
.RE
.PP
\fBReturns:\fP
.RS 4
\fBA\fP pointer to the newly initialized object\&. This will fail, returning nil if the device is not supported\&. Devices must be MTLFeatureSet_iOS_GPUFamily2_v1 or later\&. 
.RE
.PP

.PP
Reimplemented from \fBMPSCNNGradientKernel\fP\&.
.SH "Property Documentation"
.PP 
.SS "\- scaleFactorX\fC [read]\fP, \fC [nonatomic]\fP, \fC [assign]\fP"
The downsampling scale factor for the x dimension\&. The default value is 1\&. 
.SS "\- scaleFactorY\fC [read]\fP, \fC [nonatomic]\fP, \fC [assign]\fP"
The downsampling scale factor for the y dimension\&. The default value is 1\&. 

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