MPSNNGradientFilterNode(3) MetalPerformanceShaders.framework MPSNNGradientFilterNode(3)
NAME
MPSNNGradientFilterNode
SYNOPSIS
#import <MPSNNGraphNodes.h>
Inherits MPSNNFilterNode.
Inherited by MPSCNNBatchNormalizationGradientNode, MPSCNNConvolutionGradientNode, MPSCNNCrossChannelNormalizationGradientNode,
MPSCNNDropoutGradientNode, MPSCNNInstanceNormalizationGradientNode, MPSCNNLocalContrastNormalizationGradientNode,
MPSCNNLogSoftMaxGradientNode, MPSCNNNeuronGradientNode, MPSCNNPoolingGradientNode, MPSCNNSoftMaxGradientNode,
MPSCNNSpatialNormalizationGradientNode, MPSCNNUpsamplingBilinearGradientNode, MPSCNNUpsamplingNearestGradientNode,
MPSNNArithmeticGradientNode, and MPSNNConcatenationGradientNode.
Instance Methods
(MPSNNGradientFilterNode *__nonnull) - gradientFilterWithSources:
(NSArray< MPSNNGradientFilterNode * > *__nonnull) - gradientFiltersWithSources:
(MPSNNGradientFilterNode *__nonnull) - gradientFilterWithSource:
(NSArray< MPSNNGradientFilterNode * > *__nonnull) - gradientFiltersWithSource:
Additional Inherited Members
Detailed Description
For each MPSNNFilterNode, there is a corresponding MPSNNGradientFilterNode used for training that back propagates image gradients to refine
the various parameters in each node. Generally, it takes as input a gradient corresponding to the result image from the MPSNNFilterNode and
returns a gradient image corresponding to the source image of the MPSNNFilterNode. In addition, there is generally a MPSNNState produced by
the MPSNNFilterNode that is consumed by the MPSNNGradientNode and the MPSNNGradientNode generally needs to look at the MPSNNFilterNode
source image.
If you have a simple method to traverse your inference graph backwards, then -[MPSNNFilterNode gradientFilterWithSource:] is a simple way
to construct these.
Method Documentation
- (NSArray <MPSNNGradientFilterNode*> * __nonnull) gradientFiltersWithSource: (MPSNNImageNode *__nonnull) gradientImage
Return multiple gradient versions of the filter MPSNNFilters that consume multiple inputs generally result in multiple conjugate filters
for the gradient computation at the end of training. For example, a single concatenation operation that concatenates multple images will
result in an array of slice operators that carve out subsections of the input gradient image.
Reimplemented from MPSNNFilterNode.
- (NSArray <MPSNNGradientFilterNode*> * __nonnull) gradientFiltersWithSources: (NSArray< MPSNNImageNode * > *__nonnull) gradientImages
Return multiple gradient versions of the filter MPSNNFilters that consume multiple inputs generally result in multiple conjugate filters
for the gradient computation at the end of training. For example, a single concatenation operation that concatenates multple images will
result in an array of slice operators that carve out subsections of the input gradient image.
Reimplemented from MPSNNFilterNode.
- (MPSNNGradientFilterNode*__nonnull) gradientFilterWithSource: (MPSNNImageNode *__nonnull) gradientImage
Return the gradient (backwards) version of this filter. The backwards training version of the filter will be returned. The non-gradient
image and state arguments for the filter are automatically obtained from the target.
Parameters:
gradientImage The gradient images corresponding with the resultImage of the target
Reimplemented from MPSNNFilterNode.
- (MPSNNGradientFilterNode*__nonnull) gradientFilterWithSources: (NSArray< MPSNNImageNode * > *__nonnull) gradientImages
Return the gradient (backwards) version of this filter. The backwards training version of the filter will be returned. The non-gradient
image and state arguments for the filter are automatically obtained from the target.
Parameters:
gradientImages The gradient images corresponding with the resultImage of the target
Reimplemented from MPSNNFilterNode.
Author
Generated automatically by Doxygen for MetalPerformanceShaders.framework from the source code.
Version MetalPerformanceShaders-100 Thu Feb 8 2018 MPSNNGradientFilterNode(3)