Class MPSGraphSingleGateRNNDescriptor

  • All Implemented Interfaces:
    NSCopying, NSObject

    public class MPSGraphSingleGateRNNDescriptor
    extends MPSGraphObject
    implements NSCopying
    This class defines parameters for a single gate (vanilla) RNN operation. Use this descriptor with the following ``MPSGraph`` methods: - ``MPSGraph/singleGateRNNWithSourceTensor:recurrentWeight:initState:descriptor:name:``, - ``MPSGraph/singleGateRNNWithSourceTensor:recurrentWeight:inputWeight:bias:initState:descriptor:name:``, - ``MPSGraph/singleGateRNNWithSourceTensor:recurrentWeight:inputWeight:bias:initState:mask:descriptor:name:``, - ``MPSGraph/singleGateRNNGradientsWithSourceTensor:recurrentWeight:sourceGradient:zState:initState:descriptor:name:``, - ``MPSGraph/singleGateRNNGradientsWithSourceTensor:recurrentWeight:sourceGradient:zState:inputWeight:bias:initState:descriptor:name:``, - ``MPSGraph/singleGateRNNGradientsWithSourceTensor:recurrentWeight:sourceGradient:zState:inputWeight:bias:initState:mask:descriptor:name:``, and - ``MPSGraph/singleGateRNNGradientsWithSourceTensor:recurrentWeight:sourceGradient:zState:stateGradient:inputWeight:bias:initState:mask:descriptor:name:``. API-Since: 15.4
    • Constructor Detail

      • MPSGraphSingleGateRNNDescriptor

        protected MPSGraphSingleGateRNNDescriptor​(org.moe.natj.general.Pointer peer)
    • Method Detail

      • accessInstanceVariablesDirectly

        public static boolean accessInstanceVariablesDirectly()
      • activation

        public long activation()
        A parameter which defines the activation function to use with the RNN operation. Default value: `MPSGraphRNNActivationRelu`.
      • automaticallyNotifiesObserversForKey

        public static boolean automaticallyNotifiesObserversForKey​(@NotNull
                                                                   @NotNull java.lang.String key)
      • bidirectional

        public boolean bidirectional()
        A parameter that defines a bidirectional RNN layer. If set to `YES` then the input sequence is traversed in both directions and the two results are concatenated together on the channel-axis. Default value: `NO`.
      • cancelPreviousPerformRequestsWithTarget

        public static void cancelPreviousPerformRequestsWithTarget​(@NotNull
                                                                   @NotNull java.lang.Object aTarget)
      • cancelPreviousPerformRequestsWithTargetSelectorObject

        public static void cancelPreviousPerformRequestsWithTargetSelectorObject​(@NotNull
                                                                                 @NotNull java.lang.Object aTarget,
                                                                                 @NotNull
                                                                                 @NotNull org.moe.natj.objc.SEL aSelector,
                                                                                 @Nullable
                                                                                 @Nullable java.lang.Object anArgument)
      • classFallbacksForKeyedArchiver

        @NotNull
        public static @NotNull NSArray<java.lang.String> classFallbacksForKeyedArchiver()
      • classForKeyedUnarchiver

        @NotNull
        public static @NotNull org.moe.natj.objc.Class classForKeyedUnarchiver()
      • copyWithZone

        @NotNull
        public @NotNull java.lang.Object copyWithZone​(@Nullable
                                                      @Nullable org.moe.natj.general.ptr.VoidPtr zone)
        Specified by:
        copyWithZone in interface NSCopying
      • debugDescription_static

        public static java.lang.String debugDescription_static()
      • description_static

        public static java.lang.String description_static()
      • hash_static

        public static long hash_static()
      • instanceMethodSignatureForSelector

        public static NSMethodSignature instanceMethodSignatureForSelector​(org.moe.natj.objc.SEL aSelector)
      • instancesRespondToSelector

        public static boolean instancesRespondToSelector​(org.moe.natj.objc.SEL aSelector)
      • isSubclassOfClass

        public static boolean isSubclassOfClass​(org.moe.natj.objc.Class aClass)
      • keyPathsForValuesAffectingValueForKey

        @NotNull
        public static @NotNull NSSet<java.lang.String> keyPathsForValuesAffectingValueForKey​(@NotNull
                                                                                             @NotNull java.lang.String key)
      • resolveClassMethod

        public static boolean resolveClassMethod​(org.moe.natj.objc.SEL sel)
      • resolveInstanceMethod

        public static boolean resolveInstanceMethod​(org.moe.natj.objc.SEL sel)
      • reverse

        public boolean reverse()
        A parameter that defines time direction of the input sequence. If set to `YES` then the input sequence is passed in reverse time order to the layer. Note: Ignored when `bidirectional = YES`. Default value: `NO`.
      • setActivation

        public void setActivation​(long value)
        A parameter which defines the activation function to use with the RNN operation. Default value: `MPSGraphRNNActivationRelu`.
      • setBidirectional

        public void setBidirectional​(boolean value)
        A parameter that defines a bidirectional RNN layer. If set to `YES` then the input sequence is traversed in both directions and the two results are concatenated together on the channel-axis. Default value: `NO`.
      • setReverse

        public void setReverse​(boolean value)
        A parameter that defines time direction of the input sequence. If set to `YES` then the input sequence is passed in reverse time order to the layer. Note: Ignored when `bidirectional = YES`. Default value: `NO`.
      • setTraining

        public void setTraining​(boolean value)
        A parameter that makes the RNN layer support training. If set to `YES` then the layer will produce training state tensor as a secondary output. Default value: `NO`.
      • setVersion_static

        public static void setVersion_static​(long aVersion)
      • superclass_static

        public static org.moe.natj.objc.Class superclass_static()
      • training

        public boolean training()
        A parameter that makes the RNN layer support training. If set to `YES` then the layer will produce training state tensor as a secondary output. Default value: `NO`.
      • version_static

        public static long version_static()
      • useStoredAccessor

        @Deprecated
        public static boolean useStoredAccessor()
        Deprecated.