MPSMatrixNeuronGradient(3) MetalPerformanceShaders.framework MPSMatrixNeuronGradient(3)

MPSMatrixNeuronGradient

#import <MPSMatrixNeuron.h>

Inherits MPSMatrixBinaryKernel.


(void) - setNeuronType:parameterA:parameterB:parameterC:
(MPSCNNNeuronType) - neuronType
(float) - neuronParameterA
(float) - neuronParameterB
(float) - neuronParameterC
(void) - setNeuronToPReLUWithParametersA:
(nonnull instancetype) - initWithDevice:
(void) - encodeToCommandBuffer:gradientMatrix:inputMatrix:biasVector:resultGradientForDataMatrix:resultGradientForBiasVector:
(nullable instancetype) - initWithCoder:device:
(nonnull instancetype) - copyWithZone:device:


NSUInteger sourceNumberOfFeatureVectors
NSUInteger sourceInputFeatureChannels
double alpha

This depends on Metal.framework.

A neuron gradient activation kernel that operates on matrices.

A MPSMatrixNeuronGradient object computes the results of backpropagating the gradients of a loss function with respect to the outputs of an MPSMatrixNeuron object. The corresponding properties and data used by the MPSMatrixNeuronGradient object should correspond to those used by the forward MPSMatrixNeuron object for which the gradient is being computed.

- (nonnull instancetype) copyWithZone: (nullable NSZone *) zone(nullable id< MTLDevice >) device

Make a copy of this kernel for a new device -

See also:

MPSKernel

Parameters:

zone The NSZone in which to allocate the object
device The device for the new MPSKernel. If nil, then use self.device.

Returns:

A pointer to a copy of this MPSKernel. This will fail, returning nil if the device is not supported. Devices must be MTLFeatureSet_iOS_GPUFamily2_v1 or later.

Reimplemented from MPSKernel.

- (void) encodeToCommandBuffer: (nonnull id< MTLCommandBuffer >) commandBuffer(MPSMatrix *__nonnull) gradientMatrix(MPSMatrix *__nonnull) inputMatrix(MPSVector *__nullable) biasVector(MPSMatrix *__nonnull) resultGradientForDataMatrix(MPSVector *__nullable) resultGradientForBiasVector

Encode a MPSMatrixNeuronGradient object to a command buffer and compute its gradient with respect to its input data.

Parameters:

commandBuffer The commandBuffer on which to encode the operation.
gradientMatrix A matrix whose values represent the gradient of a loss function with respect to the results of a forward MPSMatrixNeuron operation.
inputMatrix A matrix containing the inputs to a forward MPSMatrixNeuron operation for which the gradient values are to be computed.
biasVector A vector containing the bias terms.
resultGradientForDataMatrix The matrix containing the resulting gradient values.
resultGradientForBiasVector If non-NULL the vector containing gradients for the bias terms.

- (nullable instancetype) initWithCoder: (NSCoder *__nonnull) aDecoder(nonnull id< MTLDevice >) device

NSSecureCoding compatability See MPSKernel::initWithCoder.

Parameters:

aDecoder The NSCoder subclass with your serialized MPSMatrixNeuronGradient
device The MTLDevice on which to make the MPSMatrixNeuronGradient object.

Returns:

A new MPSMatrixNeuronGradient object, or nil if failure.

Reimplemented from MPSKernel.

- (nonnull instancetype) initWithDevice: (nonnull id< MTLDevice >) device

Standard init with default properties per filter type

Parameters:

device The device that the filter will be used on. May not be NULL.

Returns:

a 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.

Reimplemented from MPSKernel.

- (float) neuronParameterA

Getter funtion for neuronType set using setNeuronType:parameterA:parameterB:parameterC method

- (float) neuronParameterB

Getter funtion for neuronType set using setNeuronType:parameterA:parameterB:parameterC method

- (float) neuronParameterC

Getter funtion for neuronType set using setNeuronType:parameterA:parameterB:parameterC method

- (MPSCNNNeuronType) neuronType

Getter funtion for neuronType set using setNeuronType:parameterA:parameterB:parameterC method

- (void) setNeuronToPReLUWithParametersA: (NSData *__nonnull) A

Add per output value neuron parameters A for PReLu neuron activation functions.

This method sets the neuron to PReLU, zeros parameters A and B and sets the per output value neuron parameters A to an array containing a unique value of A for each output value.

If the neuron function is f(v,a,b), it will apply



   resultMatrix(i, j) = f( input(i, j), A[j], B[j] )
where j in [0, sourceInputFeatureChannels]

See https://arxiv.org/pdf/1502.01852.pdf for details.

All other neuron types, where parameter A and parameter B are shared across output values must be set using -setNeuronType:parameterA:parameterB:

Parameters:

A An array containing float values for neuron parameter A. Number of entries must be equal to MIN(inputMatrix.columns - sourceMatrixOrigin.y, sourceInputFeatureChannels)

- (void) setNeuronType: (MPSCNNNeuronType) neuronType(float) parameterA(float) parameterB(float) parameterC

Specifies a neuron activation function to be used.

This method can be used to add a neuron activation funtion of given type with associated scalar parameters A, B, and C that are shared across all output values. Note that this method can only be used to specify neurons which are specified by three (or fewer) parameters shared across all output values (or channels, in CNN nomenclature). It is an error to call this method for neuron activation functions like MPSCNNNeuronTypePReLU, which require per-channel parameter values. For those kind of neuron activation functions, use appropriate setter functions. An MPSMatrixNeuron kernel is initialized with a default neuron function of MPSCNNNeuronTypeNone.

Parameters:

neuronType Type of neuron activation function. For full list see MPSCNNNeuronType.h
parameterA parameterA of neuron activation that is shared across all output values.
parameterB parameterB of neuron activation that is shared across all output values.
parameterC parameterC of neuron activation that is shared across all output values.

- alpha [read], [write], [nonatomic], [assign]

The scale factor to apply to the input.

- sourceInputFeatureChannels [read], [write], [nonatomic], [assign]

The number of feature channels in the input vectors.

- sourceNumberOfFeatureVectors [read], [write], [nonatomic], [assign]

The number of input vectors which make up the input array.

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

Mon Jul 9 2018 Version MetalPerformanceShaders-119.3