Login or Register to Ask a Question and Join Our Community

Linux and UNIX Man Pages

Linux & Unix Commands - Search Man Pages

mpscnnupsamplingnearestgradientnode(3) [mojave man page]

MPSCNNUpsamplingNearestGradientNode(3)			 MetalPerformanceShaders.framework		    MPSCNNUpsamplingNearestGradientNode(3)

NAME

       MPSCNNUpsamplingNearestGradientNode

SYNOPSIS

       #import <MPSNNGraphNodes.h>

       Inherits MPSNNGradientFilterNode.

   Instance Methods
       (nonnull instancetype) - initWithSourceGradient:sourceImage:gradientState:scaleFactorX:scaleFactorY:

   Class Methods
       (nonnull instancetype) + nodeWithSourceGradient:sourceImage:gradientState:scaleFactorX:scaleFactorY:

   Properties
       double scaleFactorX
       double scaleFactorY

Detailed Description
       Node representing a MPSCNNUpsamplingNearest kernel

Method Documentation
   - (nonnull instancetype) initWithSourceGradient: (MPSNNImageNode *__nonnull) sourceGradient(MPSNNImageNode *__nonnull)
       sourceImage(MPSNNGradientStateNode *__nonnull) gradientState(double) scaleFactorX(double) scaleFactorY
       A node to represent the gradient calculation for nearest upsampling training.  [forwardFilter gradientFilterWithSources:] is a more
       convient way to do this.

       Parameters:
	   sourceGradient The input gradient from the 'downstream' gradient filter.
	   sourceImage The input image from the forward filter node
	   gradientState The gradient state from the forward filter
	   scaleFactorX The X scale factor from the forward pass
	   scaleFactorY The Y scale factor from the forward pass

       Returns:
	   A MPSCNNUpsamplingNearestGradientNode

   + (nonnull instancetype) nodeWithSourceGradient: (MPSNNImageNode *__nonnull) sourceGradient(MPSNNImageNode *__nonnull)
       sourceImage(MPSNNGradientStateNode *__nonnull) gradientState(double) scaleFactorX(double) scaleFactorY
       A node to represent the gradient calculation for nearest upsampling training.  [forwardFilter gradientFilterWithSources:] is a more
       convient way to do this.

       Parameters:
	   sourceGradient The input gradient from the 'downstream' gradient filter.
	   sourceImage The input image from the forward filter node
	   gradientState The gradient state from the forward filter
	   scaleFactorX The X scale factor from the forward pass
	   scaleFactorY The Y scale factor from the forward pass

       Returns:
	   A MPSCNNUpsamplingNearestGradientNode

Property Documentation
   - (double) scaleFactorX [read],  [nonatomic],  [assign]
   - (double) scaleFactorY [read],  [nonatomic],  [assign]
Author
       Generated automatically by Doxygen for MetalPerformanceShaders.framework from the source code.

Version MetalPerformanceShaders-100				  Thu Feb 8 2018			    MPSCNNUpsamplingNearestGradientNode(3)

Check Out this Related Man Page

MPSCNNLocalContrastNormalizationGradientNode(3) 	 MetalPerformanceShaders.framework	   MPSCNNLocalContrastNormalizationGradientNode(3)

NAME

       MPSCNNLocalContrastNormalizationGradientNode

SYNOPSIS

       #import <MPSNNGraphNodes.h>

       Inherits MPSNNGradientFilterNode.

   Instance Methods
       (nonnull instancetype) - initWithSourceGradient:sourceImage:gradientState:kernelWidth:kernelHeight:

   Class Methods
       (nonnull instancetype) + nodeWithSourceGradient:sourceImage:gradientState:kernelWidth:kernelHeight:

   Properties
       float alpha
       float beta
       float delta
       float p0
       float pm
       float ps
       NSUInteger kernelWidth
       NSUInteger kernelHeight

Method Documentation
   - (nonnull instancetype) initWithSourceGradient: (MPSNNImageNode *__nonnull) sourceGradient(MPSNNImageNode *__nonnull)
       sourceImage(MPSNNGradientStateNode *__nonnull) gradientState(NSUInteger) kernelWidth(NSUInteger) kernelHeight
   + (nonnull instancetype) nodeWithSourceGradient: (MPSNNImageNode *__nonnull) sourceGradient(MPSNNImageNode *__nonnull)
       sourceImage(MPSNNGradientStateNode *__nonnull) gradientState(NSUInteger) kernelWidth(NSUInteger) kernelHeight
Property Documentation
   - alpha [read],  [write],  [nonatomic],  [assign]
       The value of alpha. Default is 0.0  The default value 0.0 is not recommended and is preserved for backwards compatibility. With alpha 0, it
       performs a local mean subtraction. The MPSCNNLocalContrastNormalizationNode used with the MPSNNGraph uses 1.0 as a default.

   - beta [read],  [write],  [nonatomic],  [assign]
       The value of beta. Default is 0.5

   - delta [read],  [write],  [nonatomic],  [assign]
       The value of delta. Default is 1/1024

   - (NSUInteger) kernelHeight [read],	[nonatomic],  [assign]
   - (NSUInteger) kernelWidth [read],  [nonatomic],  [assign]
   - p0 [read],  [write],  [nonatomic],  [assign]
       The value of p0. Default is 1.0

   - pm [read],  [write],  [nonatomic],  [assign]
       The value of pm. Default is 0.0

   - ps [read],  [write],  [nonatomic],  [assign]
       The value of ps. Default is 1.0

Author
       Generated automatically by Doxygen for MetalPerformanceShaders.framework from the source code.

Version MetalPerformanceShaders-100				  Thu Feb 8 2018		   MPSCNNLocalContrastNormalizationGradientNode(3)
Man Page