public class MPSMatrixFullyConnected extends MPSMatrixBinaryKernel
[@dependency] This depends on Metal.framework.
Applies a fully connected neural network layer by performing a a matrix multiplication, adding a bias vector, scaling, and applying a neuron activation function.
A MPSMatrixFullyConnected object computes:
y = neuron(alpha * x * W + bias)
y is the output matrix, x and W are input matrices corresponding to a collection of input vectors and weights respectively, and bias is a vector which is broadcast and accumulated to each row of the product. alpha is a scale factor applied to the product.
neuron() is a pointwise function applied to the intermediate result.
NSObject.Function_instanceMethodForSelector_ret, NSObject.Function_methodForSelector_ret| Modifier | Constructor and Description |
|---|---|
protected |
MPSMatrixFullyConnected(org.moe.natj.general.Pointer peer) |
| Modifier and Type | Method and Description |
|---|---|
boolean |
_supportsSecureCoding()
This property must return YES on all classes that allow secure coding.
|
static boolean |
accessInstanceVariablesDirectly() |
static MPSMatrixFullyConnected |
alloc() |
static MPSMatrixFullyConnected |
allocWithZone(org.moe.natj.general.ptr.VoidPtr zone) |
double |
alpha()
[@property] alpha
|
static boolean |
automaticallyNotifiesObserversForKey(java.lang.String key) |
static void |
cancelPreviousPerformRequestsWithTarget(java.lang.Object aTarget) |
static void |
cancelPreviousPerformRequestsWithTargetSelectorObject(java.lang.Object aTarget,
org.moe.natj.objc.SEL aSelector,
java.lang.Object anArgument) |
static NSArray<java.lang.String> |
classFallbacksForKeyedArchiver() |
static org.moe.natj.objc.Class |
classForKeyedUnarchiver() |
MPSMatrixFullyConnected |
copyWithZoneDevice(org.moe.natj.general.ptr.VoidPtr zone,
MTLDevice device)
Make a copy of this kernel for a new device - @see MPSKernel
|
static java.lang.String |
debugDescription_static() |
static java.lang.String |
description_static() |
void |
encodeToCommandBufferInputMatrixWeightMatrixBiasVectorResultMatrix(MTLCommandBuffer commandBuffer,
MPSMatrix inputMatrix,
MPSMatrix weightMatrix,
MPSVector biasVector,
MPSMatrix resultMatrix)
Encode a MPSMatrixFullyConnected object to a command buffer.
|
static long |
hash_static() |
MPSMatrixFullyConnected |
init() |
MPSMatrixFullyConnected |
initWithCoder(NSCoder aDecoder)
NS_DESIGNATED_INITIALIZER
|
MPSMatrixFullyConnected |
initWithCoderDevice(NSCoder aDecoder,
java.lang.Object device)
NSSecureCoding compatability
|
MPSMatrixFullyConnected |
initWithDevice(java.lang.Object device)
Standard init with default properties per filter type
|
static NSObject.Function_instanceMethodForSelector_ret |
instanceMethodForSelector(org.moe.natj.objc.SEL aSelector) |
static NSMethodSignature |
instanceMethodSignatureForSelector(org.moe.natj.objc.SEL aSelector) |
static boolean |
instancesRespondToSelector(org.moe.natj.objc.SEL aSelector) |
static boolean |
isSubclassOfClass(org.moe.natj.objc.Class aClass) |
static NSSet<java.lang.String> |
keyPathsForValuesAffectingValueForKey(java.lang.String key) |
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
|
int |
neuronType()
Getter funtion for neuronType set using setNeuronType:parameterA:parameterB:parameterC method
|
static MPSMatrixFullyConnected |
new_objc() |
static boolean |
resolveClassMethod(org.moe.natj.objc.SEL sel) |
static boolean |
resolveInstanceMethod(org.moe.natj.objc.SEL sel) |
void |
setAlpha(double value)
[@property] alpha
|
void |
setNeuronTypeParameterAParameterBParameterC(int neuronType,
float parameterA,
float parameterB,
float parameterC)
Specifies a neuron activation function to be used.
|
void |
setSourceInputFeatureChannels(long value)
[@property] sourceInputFeatureChannels
|
void |
setSourceNumberOfFeatureVectors(long value)
[@property] sourceNumberOfFeatureVectors
|
void |
setSourceOutputFeatureChannels(long value)
[@property] sourceOutputFeatureChannels
|
static void |
setVersion_static(long aVersion) |
long |
sourceInputFeatureChannels()
[@property] sourceInputFeatureChannels
|
long |
sourceNumberOfFeatureVectors()
[@property] sourceNumberOfFeatureVectors
|
long |
sourceOutputFeatureChannels()
[@property] sourceOutputFeatureChannels
|
static org.moe.natj.objc.Class |
superclass_static() |
static boolean |
supportsSecureCoding() |
static long |
version_static() |
batchSize, batchStart, primarySourceMatrixOrigin, resultMatrixOrigin, secondarySourceMatrixOrigin, setBatchSize, setBatchStart, setPrimarySourceMatrixOrigin, setResultMatrixOrigin, setSecondarySourceMatrixOrigincopyWithZone, device, encodeWithCoder, label, options, setLabel, setOptionsaccessibilityActivate, accessibilityActivationPoint, accessibilityAssistiveTechnologyFocusedIdentifiers, accessibilityAttributedHint, accessibilityAttributedLabel, accessibilityAttributedUserInputLabels, accessibilityAttributedValue, accessibilityContainerType, accessibilityCustomActions, accessibilityCustomRotors, accessibilityDecrement, accessibilityDragSourceDescriptors, accessibilityDropPointDescriptors, accessibilityElementAtIndex, accessibilityElementCount, accessibilityElementDidBecomeFocused, accessibilityElementDidLoseFocus, accessibilityElementIsFocused, accessibilityElements, accessibilityElementsHidden, accessibilityFrame, accessibilityHint, accessibilityIncrement, accessibilityLabel, accessibilityLanguage, accessibilityNavigationStyle, accessibilityPath, accessibilityPerformEscape, accessibilityPerformMagicTap, accessibilityRespondsToUserInteraction, accessibilityScroll, accessibilityTextualContext, accessibilityTraits, accessibilityUserInputLabels, accessibilityValue, accessibilityViewIsModal, addObserverForKeyPathOptionsContext, attemptRecoveryFromErrorOptionIndex, attemptRecoveryFromErrorOptionIndexDelegateDidRecoverSelectorContextInfo, autoContentAccessingProxy, awakeAfterUsingCoder, awakeFromNib, class_objc, classForCoder, classForKeyedArchiver, copy, dealloc, debugDescription, description, dictionaryWithValuesForKeys, didChangeValueForKey, didChangeValueForKeyWithSetMutationUsingObjects, didChangeValuesAtIndexesForKey, doesNotRecognizeSelector, fileManagerShouldProceedAfterError, fileManagerWillProcessPath, finalize_objc, forwardingTargetForSelector, forwardInvocation, hash, indexOfAccessibilityElement, isAccessibilityElement, isEqual, isKindOfClass, isMemberOfClass, isProxy, methodForSelector, methodSignatureForSelector, mutableArrayValueForKey, mutableArrayValueForKeyPath, mutableCopy, mutableOrderedSetValueForKey, mutableOrderedSetValueForKeyPath, mutableSetValueForKey, mutableSetValueForKeyPath, observationInfo, observeValueForKeyPathOfObjectChangeContext, performSelector, performSelectorInBackgroundWithObject, performSelectorOnMainThreadWithObjectWaitUntilDone, performSelectorOnMainThreadWithObjectWaitUntilDoneModes, performSelectorOnThreadWithObjectWaitUntilDone, performSelectorOnThreadWithObjectWaitUntilDoneModes, performSelectorWithObject, performSelectorWithObjectAfterDelay, performSelectorWithObjectAfterDelayInModes, performSelectorWithObjectWithObject, prepareForInterfaceBuilder, provideImageDataBytesPerRowOrigin_Size_UserInfo, removeObserverForKeyPath, removeObserverForKeyPathContext, replacementObjectForCoder, replacementObjectForKeyedArchiver, respondsToSelector, self, setAccessibilityActivationPoint, setAccessibilityAttributedHint, setAccessibilityAttributedLabel, setAccessibilityAttributedUserInputLabels, setAccessibilityAttributedValue, setAccessibilityContainerType, setAccessibilityCustomActions, setAccessibilityCustomRotors, setAccessibilityDragSourceDescriptors, setAccessibilityDropPointDescriptors, setAccessibilityElements, setAccessibilityElementsHidden, setAccessibilityFrame, setAccessibilityHint, setAccessibilityLabel, setAccessibilityLanguage, setAccessibilityNavigationStyle, setAccessibilityPath, setAccessibilityRespondsToUserInteraction, setAccessibilityTextualContext, setAccessibilityTraits, setAccessibilityUserInputLabels, setAccessibilityValue, setAccessibilityViewIsModal, setIsAccessibilityElement, setNilValueForKey, setObservationInfo, setShouldGroupAccessibilityChildren, setValueForKey, setValueForKeyPath, setValueForUndefinedKey, setValuesForKeysWithDictionary, shouldGroupAccessibilityChildren, superclass, validateValueForKeyError, validateValueForKeyPathError, valueForKey, valueForKeyPath, valueForUndefinedKey, willChangeValueForKey, willChangeValueForKeyWithSetMutationUsingObjects, willChangeValuesAtIndexesForKeyprotected MPSMatrixFullyConnected(org.moe.natj.general.Pointer peer)
public static boolean accessInstanceVariablesDirectly()
public static MPSMatrixFullyConnected alloc()
public static MPSMatrixFullyConnected allocWithZone(org.moe.natj.general.ptr.VoidPtr zone)
public double alpha()
The scale factor to apply to the product. Specified in double precision. Will be converted to the appropriate precision in the implementation subject to rounding and/or clamping as necessary. Defaults to 1.0 at initialization time.
public static boolean automaticallyNotifiesObserversForKey(java.lang.String key)
public static void cancelPreviousPerformRequestsWithTarget(java.lang.Object aTarget)
public static void cancelPreviousPerformRequestsWithTargetSelectorObject(java.lang.Object aTarget,
org.moe.natj.objc.SEL aSelector,
java.lang.Object anArgument)
public static NSArray<java.lang.String> classFallbacksForKeyedArchiver()
public static org.moe.natj.objc.Class classForKeyedUnarchiver()
public MPSMatrixFullyConnected copyWithZoneDevice(org.moe.natj.general.ptr.VoidPtr zone, MTLDevice device)
copyWithZoneDevice in class MPSKernelzone - The NSZone in which to allocate the objectdevice - The device for the new MPSKernel. If nil, then use
self.device.public static java.lang.String debugDescription_static()
public static java.lang.String description_static()
public void encodeToCommandBufferInputMatrixWeightMatrixBiasVectorResultMatrix(MTLCommandBuffer commandBuffer, MPSMatrix inputMatrix, MPSMatrix weightMatrix, MPSVector biasVector, MPSMatrix resultMatrix)
Encodes the operation to the specified command buffer. resultMatrix must be large enough to hold a MIN(sourceNumberOfInputs, inputMatrix.rows - primarySourceMatrixOrigin.x) x MIN(sourceOutputFeatureChannels, weightMatrix.columns - secondarySourceMatrixOrigin.y) array.
The bias vector must contain at least MIN(sourceOutputFeatureChannels, weightMatrix.columns - secondarySourceMatrixOrigin.y) elements.
commandBuffer - A valid MTLCommandBuffer to receive the encoded kernel.inputMatrix - A valid MPSMatrix object which specifies the input array.weightMatrix - A valid MPSMatrix object which specifies the weight array.biasVector - A valid MPSVector object which specifies the bias values, or
a null object to indicate that no bias is to be applied.resultMatrix - A valid MPSMatrix object which specifies the output array.public static long hash_static()
public MPSMatrixFullyConnected init()
init in class MPSMatrixBinaryKernelpublic MPSMatrixFullyConnected initWithCoder(NSCoder aDecoder)
NSCodinginitWithCoder in interface NSCodinginitWithCoder in class MPSMatrixBinaryKernelpublic MPSMatrixFullyConnected initWithCoderDevice(NSCoder aDecoder, java.lang.Object device)
See @ref MPSKernel#initWithCoder.
initWithCoderDevice in class MPSMatrixBinaryKernelaDecoder - The NSCoder subclass with your serialized MPSMatrixFullyConnecteddevice - The MTLDevice on which to make the MPSMatrixFullyConnected object.public MPSMatrixFullyConnected initWithDevice(java.lang.Object device)
MPSKernelinitWithDevice in class MPSMatrixBinaryKerneldevice - The device that the filter will be used on. May not be NULL.public static NSObject.Function_instanceMethodForSelector_ret instanceMethodForSelector(org.moe.natj.objc.SEL aSelector)
public static NSMethodSignature instanceMethodSignatureForSelector(org.moe.natj.objc.SEL aSelector)
public static boolean instancesRespondToSelector(org.moe.natj.objc.SEL aSelector)
public static boolean isSubclassOfClass(org.moe.natj.objc.Class aClass)
public static NSSet<java.lang.String> keyPathsForValuesAffectingValueForKey(java.lang.String key)
public float neuronParameterA()
public float neuronParameterB()
public float neuronParameterC()
public int neuronType()
public static MPSMatrixFullyConnected new_objc()
public static boolean resolveClassMethod(org.moe.natj.objc.SEL sel)
public static boolean resolveInstanceMethod(org.moe.natj.objc.SEL sel)
public void setAlpha(double value)
The scale factor to apply to the product. Specified in double precision. Will be converted to the appropriate precision in the implementation subject to rounding and/or clamping as necessary. Defaults to 1.0 at initialization time.
public void setNeuronTypeParameterAParameterBParameterC(int neuronType,
float parameterA,
float parameterB,
float parameterC)
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 MPSMatrixFullyConnected kernel is initialized with a default neuron function of MPSCNNNeuronTypeNone.
neuronType - Type of neuron activation function. For full list see MPSCNNNeuronType.hparameterA - 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.public void setSourceInputFeatureChannels(long value)
The input size to to use in the operation. This is equivalent to the number of columns and the number of rows in the primary (input array) and secondary (weight array) source matrices respectively. This property is modifiable and defaults to NSUIntegerMax. At encode time the larger of this property or the available input size is used. The value of NSUIntegerMax thus indicates that all available columns in the input array (beginning at primarySourceMatrixOrigin.y) and all available rows in the weight array (beginning at secondarySourceMatrixOrigin.x) should be considered. Note: The value used in the operation will be MIN(MIN(inputMatrix.columns - primarySourceMatrixOrigin.y, weightMatrix.rows - secondarySourceMatrixOrigin.x), sourceInputFeatureChannels)
public void setSourceNumberOfFeatureVectors(long value)
The number of input vectors which make up the input array. This is equivalent to the number of rows to consider from the primary source matrix. This property is modifiable and defaults to NSUIntegerMax. At encode time the larger of this property or the available number of inputs is used. The value of NSUIntegerMax thus indicates that all available input rows (beginning at primarySourceMatrixOrigin.x) should be considered.
public void setSourceOutputFeatureChannels(long value)
The output size to to use in the operation. This is equivalent to the number of columns to consider in the weight array, or the secondary source matrix. This property is modifiable and defaults to NSUIntegerMax. At encode time the larger of this property or the available output size is used. The value of NSUIntegerMax thus indicates that all available columns in the weight array (beginning at secondarySourceMatrixOrigin.y) should be considered.
public static void setVersion_static(long aVersion)
public long sourceInputFeatureChannels()
The input size to to use in the operation. This is equivalent to the number of columns and the number of rows in the primary (input array) and secondary (weight array) source matrices respectively. This property is modifiable and defaults to NSUIntegerMax. At encode time the larger of this property or the available input size is used. The value of NSUIntegerMax thus indicates that all available columns in the input array (beginning at primarySourceMatrixOrigin.y) and all available rows in the weight array (beginning at secondarySourceMatrixOrigin.x) should be considered. Note: The value used in the operation will be MIN(MIN(inputMatrix.columns - primarySourceMatrixOrigin.y, weightMatrix.rows - secondarySourceMatrixOrigin.x), sourceInputFeatureChannels)
public long sourceNumberOfFeatureVectors()
The number of input vectors which make up the input array. This is equivalent to the number of rows to consider from the primary source matrix. This property is modifiable and defaults to NSUIntegerMax. At encode time the larger of this property or the available number of inputs is used. The value of NSUIntegerMax thus indicates that all available input rows (beginning at primarySourceMatrixOrigin.x) should be considered.
public long sourceOutputFeatureChannels()
The output size to to use in the operation. This is equivalent to the number of columns to consider in the weight array, or the secondary source matrix. This property is modifiable and defaults to NSUIntegerMax. At encode time the larger of this property or the available output size is used. The value of NSUIntegerMax thus indicates that all available columns in the weight array (beginning at secondarySourceMatrixOrigin.y) should be considered.
public static org.moe.natj.objc.Class superclass_static()
public static boolean supportsSecureCoding()
public boolean _supportsSecureCoding()
NSSecureCoding_supportsSecureCoding in interface NSSecureCoding_supportsSecureCoding in class MPSMatrixBinaryKernelpublic static long version_static()