MPSCNNConvolution(3) MetalPerformanceShaders.framework MPSCNNConvolution(3)
NAME
MPSCNNConvolution
SYNOPSIS
#import <MPSCNNConvolution.h>
Inherits MPSCNNKernel.
Inherited by MPSCNNFullyConnected.
Instance Methods
(nonnull instancetype) - initWithDevice:weights:
(nullable instancetype) - initWithCoder:device:
(nonnull instancetype) - initWithDevice:
(MPSCNNConvolutionGradientState *__nullable) - resultStateForSourceImage:sourceStates:destinationImage:
(MPSCNNConvolutionGradientStateBatch *__nullable) - resultStateBatchForSourceImage:sourceStates:destinationImage:
(MPSCNNConvolutionGradientState *__nullable) - temporaryResultStateForCommandBuffer:sourceImage:sourceStates:destinationImage:
(MPSCNNConvolutionGradientStateBatch *__nullable) - temporaryResultStateBatchForCommandBuffer:sourceImage:sourceStates:destinationImage:
(void) - reloadWeightsAndBiasesWithDataSource:
(void) - reloadWeightsAndBiasesWithCommandBuffer:state:
(MPSCNNConvolutionWeightsAndBiasesState *__nonnull) - exportWeightsAndBiasesWithCommandBuffer:resultStateCanBeTemporary:
Properties
NSUInteger inputFeatureChannels
NSUInteger outputFeatureChannels
NSUInteger groups
id< MPSCNNConvolutionDataSource > dataSource
NSUInteger subPixelScaleFactor
const MPSCNNNeuron *__nullable neuron
const MPSCNNNeuron *__nullable MPSCNNNeuronType neuronType
float neuronParameterA
float neuronParameterB
float neuronParameterC
NSUInteger channelMultiplier
MPSNNConvolutionAccumulatorPrecisionOption accumulatorPrecisionOption
Additional Inherited Members
Detailed Description
This depends on Metal.framework The MPSCNNConvolution specifies a convolution. The MPSCNNConvolution convolves the input image with a set
of filters, each producing one feature map in the output image.
Method Documentation
- (MPSCNNConvolutionWeightsAndBiasesState* __nonnull) exportWeightsAndBiasesWithCommandBuffer: (__nonnull id< MTLCommandBuffer >)
commandBuffer(BOOL) resultStateCanBeTemporary
GPU side export. Enqueue a kernel to export current weights and biases stored in MPSCNNConvoltion's internal buffers into weights and
biases MTLBuffer returned in MPSCNNConvolutionWeightsAndBiasesState.
Parameters:
commandBuffer Metal command buffer on which export kernel is enqueued.
resultStateCanBeTemporary If FALSE, state returned will be non-temporary. If TRUE, returned state may or may not be temporary.
Returns:
MPSCNNConvolutionWeightsAndBiasesState containing weights and biases buffer to which weights got exported. This state and be temporary
or non-temporary depending on the flag resultStateCanBeTemporary
- (nullable instancetype) initWithCoder: (NSCoder *__nonnull) aDecoder(nonnull id< MTLDevice >) device
NSSecureCoding compatability While the standard NSSecureCoding/NSCoding method -initWithCoder: should work, since the file can't know
which device your data is allocated on, we have to guess and may guess incorrectly. To avoid that problem, use initWithCoder:device
instead.
Parameters:
aDecoder The NSCoder subclass with your serialized MPSKernel
device The MTLDevice on which to make the MPSKernel
Returns:
A new MPSKernel object, or nil if failure.
Reimplemented from MPSCNNKernel.
Reimplemented in MPSCNNFullyConnected.
- (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 MPSCNNKernel.
Reimplemented in MPSCNNFullyConnected.
- (nonnull instancetype) initWithDevice: (nonnull id< MTLDevice >) device(nonnull id< MPSCNNConvolutionDataSource >) weights
Initializes a convolution kernel
Parameters:
device The MTLDevice on which this MPSCNNConvolution filter will be used
weights A pointer to a object that conforms to the MPSCNNConvolutionDataSource protocol. The MPSCNNConvolutionDataSource protocol
declares the methods that an instance of MPSCNNConvolution uses to obtain the weights and bias terms for the CNN convolution filter.
Returns:
A valid MPSCNNConvolution object or nil, if failure.
Reimplemented in MPSCNNFullyConnected.
- (void) reloadWeightsAndBiasesWithCommandBuffer: (__nonnull id< MTLCommandBuffer >) commandBuffer(MPSCNNConvolutionWeightsAndBiasesState
*__nonnull) state
GPU side reload. Reload the updated weights and biases from update buffer produced by application enqueued metal kernel into internal
weights and biases buffer. Weights and biases gradients needed for update are obtained from MPSCNNConvolutionGradientState object's
gradientForWeights and gradientForBiases metal buffer.
Parameters:
commandBuffer Metal command buffer on which application update kernel was enqueued consuming MPSCNNConvolutionGradientState's
gradientForWeights and gradientForBiases buffers and producing updateBuffer metal buffer.
state MPSCNNConvolutionWeightsAndBiasesState containing weights and biases buffers which have updated weights produced by application's
update kernel. The state readcount will be decremented.
- (void) reloadWeightsAndBiasesWithDataSource: (__nonnull id< MPSCNNConvolutionDataSource >) dataSource
CPU side reload. Reload the updated weights and biases from data provider into internal weights and bias buffers. Weights and biases
gradients needed for update are obtained from MPSCNNConvolutionGradientState object.
Parameters:
dataSource The data source which has been updated with weights and biases gradeint from MPSCNNConvolutionGradientState object passed
into MPSCNNConvolution and MPSCNNConvolutionGradient encode calls.
- (MPSCNNConvolutionGradientStateBatch * __nullable) resultStateBatchForSourceImage: (MPSImageBatch *__nonnull) sourceImage(NSArray<
MPSStateBatch * > *__nullable) sourceStates(MPSImageBatch *_Nonnull) destinationImage
- (MPSCNNConvolutionGradientState * __nullable) resultStateForSourceImage: (MPSImage *__nonnull) sourceImage(NSArray< MPSState * > *__nullable)
sourceStates(MPSImage *__nonnull) destinationImage
Allocate a MPCNNConvolutionGradientSState to hold the results from a -encodeBatchToCommandBuffer... operation
Parameters:
sourceImage The MPSImage consumed by the associated -encode call.
sourceStates The list of MPSStates consumed by the associated -encode call, for a batch size of 1.
Returns:
The list of states produced by the -encode call for batch size of 1. -isResultStateReusedAcrossBatch returns YES for MPSCNNConvolution
so same state is used across entire batch. State object is not reusasable across batches.
Reimplemented from MPSCNNKernel.
- (MPSCNNConvolutionGradientStateBatch * __nullable) temporaryResultStateBatchForCommandBuffer: (nonnull id< MTLCommandBuffer >)
commandBuffer(MPSImageBatch *__nonnull) sourceImage(NSArray< MPSStateBatch * > *__nullable) sourceStates(MPSImageBatch *__nonnull)
destinationImage
Reimplemented from MPSCNNKernel.
- (MPSCNNConvolutionGradientState * __nullable) temporaryResultStateForCommandBuffer: (nonnull id< MTLCommandBuffer >) commandBuffer(MPSImage
*__nonnull) sourceImage(NSArray< MPSState * > *__nullable) sourceStates(MPSImage *__nonnull) destinationImage
Allocate a temporary MPSState (subclass) to hold the results from a -encodeBatchToCommandBuffer... operation A graph may need to allocate
storage up front before executing. This may be necessary to avoid using too much memory and to manage large batches. The function should
allocate any MPSState objects that will be produced by an -encode call with the indicated sourceImages and sourceStates inputs. Though the
states can be further adjusted in the ensuing -encode call, the states should be initialized with all important data and all MTLResource
storage allocated. The data stored in the MTLResource need not be initialized, unless the ensuing -encode call expects it to be.
The MTLDevice used by the result is derived from the command buffer. The padding policy will be applied to the filter before this is called
to give it the chance to configure any properties like MPSCNNKernel.offset.
CAUTION: The kernel must have all properties set to values that will ultimately be passed to the -encode call that writes to the state,
before -resultStateForSourceImages:sourceStates:destinationImage: is called or behavior is undefined. Please note that
-destinationImageDescriptorForSourceImages:sourceStates:destinationImage: will alter some of these properties automatically based on the
padding policy. If you intend to call that to make the destination image, then you should call that before
-resultStateForSourceImages:sourceStates:destinationImage:. This will ensure the properties used in the encode call and in the destination
image creation match those used to configure the state.
The following order is recommended:
// Configure MPSCNNKernel properties first
kernel.edgeMode = MPSImageEdgeModeZero;
kernel.destinationFeatureChannelOffset = 128; // concatenation without the copy
// ALERT: will change MPSCNNKernel properties
MPSImageDescriptor * d = [kernel destinationImageDescriptorForSourceImage: source
sourceStates: states];
MPSTemporaryImage * dest = [MPSTemporaryImage temporaryImageWithCommandBuffer: cmdBuf
imageDescriptor: d];
// Now that all properties are configured properly, we can make the result state
// and call encode.
MPSState * __nullable destState = [kernel temporaryResultStateForCommandBuffer: cmdBuf
sourceImage: source
sourceStates: states];
// This form of -encode will be declared by the MPSCNNKernel subclass
[kernel encodeToCommandBuffer: cmdBuf
sourceImage: source
destinationState: destState
destinationImage: dest ];
Default: returns nil
Parameters:
commandBuffer The command buffer to allocate the temporary storage against The state will only be valid on this command buffer.
sourceImage The MPSImage consumed by the associated -encode call.
sourceStates The list of MPSStates consumed by the associated -encode call, for a batch size of 1.
destinationImage The destination image for the encode call
Returns:
The list of states produced by the -encode call for batch size of 1. When the batch size is not 1, this function will be called
repeatedly unless -isResultStateReusedAcrossBatch returns YES. If -isResultStateReusedAcrossBatch returns YES, then it will be called
once per batch and the MPSStateBatch array will contain MPSStateBatch.length references to the same object.
Reimplemented from MPSCNNKernel.
Property Documentation
- (MPSNNConvolutionAccumulatorPrecisionOption) accumulatorPrecisionOption [read], [write], [nonatomic], [assign]
Precision of accumulator used in convolution. See MPSNeuralNetworkTypes.h for discussion. Default is
MPSNNConvolutionAccumulatorPrecisionOptionHalf.
- (NSUInteger) channelMultiplier [read], [nonatomic], [assign]
Channel multiplier. For convolution created with MPSCNNDepthWiseConvolutionDescriptor, it is the number of output feature channels for
each input channel. See MPSCNNDepthWiseConvolutionDescriptor for more details. Default is 0 which means regular CNN convolution.
- dataSource [read], [nonatomic], [retain]
dataSource with which convolution object was created
- groups [read], [nonatomic], [assign]
Number of groups input and output channels are divided into.
- inputFeatureChannels [read], [nonatomic], [assign]
The number of feature channels per pixel in the input image.
- neuron [read], [nonatomic], [assign]
MPSCNNNeuron filter to be applied as part of convolution. Can be nil in wich case no neuron activation fuction is applied.
- (float) neuronParameterA [read], [nonatomic], [assign]
Parameter 'a' for the neuron. Default: 1.0f Please see class description for interpretation of a.
- (float) neuronParameterB [read], [nonatomic], [assign]
Parameter 'b' for the neuron. Default: 1.0f Please see class description for interpretation of b.
- (float) neuronParameterC [read], [nonatomic], [assign]
Parameter 'c' for the neuron. Default: 1.0f Please see class description for interpretation of c.
- (const MPSCNNNeuron* __nullable MPSCNNNeuronType) neuronType [read], [nonatomic], [assign]
The type of neuron to append to the convolution Please see class description for a full list. Default is MPSCNNNeuronTypeNone.
- outputFeatureChannels [read], [nonatomic], [assign]
The number of feature channels per pixel in the output image.
- subPixelScaleFactor [read], [nonatomic], [assign]
Sub pixel scale factor which was passed in as part of MPSCNNConvolutionDescriptor when creating this MPSCNNConvolution object.
Author
Generated automatically by Doxygen for MetalPerformanceShaders.framework from the source code.
Version MetalPerformanceShaders-100 Thu Feb 8 2018 MPSCNNConvolution(3)