Class

com.github.nearbydelta.deepspark.layer

VectorRBFLayer

Related Doc: package layer

Permalink

class VectorRBFLayer extends TransformLayer

Layer : An Radial Basis Function Layer, with its radial basis.

Note

This is a RBF layer, mainly the same with 3-phrase RBF in paper Three learning phrases for radial-basis-function networks

Linear Supertypes
TransformLayer, Layer[DataVec, DataVec], KryoSerializable, Serializable, Serializable, AnyRef, Any
Ordering
  1. Alphabetic
  2. By inheritance
Inherited
  1. VectorRBFLayer
  2. TransformLayer
  3. Layer
  4. KryoSerializable
  5. Serializable
  6. Serializable
  7. AnyRef
  8. Any
  1. Hide All
  2. Show all
Visibility
  1. Public
  2. All

Instance Constructors

  1. new VectorRBFLayer()

    Permalink

Value Members

  1. final def !=(arg0: Any): Boolean

    Permalink
    Definition Classes
    AnyRef → Any
  2. final def ##(): Int

    Permalink
    Definition Classes
    AnyRef → Any
  3. final def ==(arg0: Any): Boolean

    Permalink
    Definition Classes
    AnyRef → Any
  4. var NIn: Int

    Permalink

    Size of input *

    Size of input *

    Definition Classes
    Layer
  5. var NOut: Int

    Permalink

    Size of output *

    Size of output *

    Definition Classes
    Layer
  6. var act: RBFActivation

    Permalink

    RBF Activation *

  7. def apply(x: DataVec): DataVec

    Permalink

    Apply this layer (forward computation)

    Apply this layer (forward computation)

    returns

    Output computation information.

    Definition Classes
    VectorRBFLayerLayer
  8. def apply(in: RDD[(Long, DataVec)]): RDD[(Long, DataVec)]

    Permalink

    Apply RDD of vector.

    Apply RDD of vector.

    in

    RDD of (ID, Value), ID is Long value.

    returns

    RDD of (ID, Vector), with same ID for input.

    Definition Classes
    Layer
  9. final def asInstanceOf[T0]: T0

    Permalink
    Definition Classes
    Any
  10. def backprop(seq: ParSeq[((DataVec, DataVec), DataVec)]): (ParSeq[DataVec], ParSeq[() ⇒ Unit])

    Permalink

    Backward computation

    Backward computation

    seq

    Sequence of entries to be used for backward computation.

    returns

    Error sequence, to backpropagate into previous layer.

    Definition Classes
    VectorRBFLayerLayer
    Note

    For the computation, we only used denominator layout. (cf. Wikipedia Page of Matrix Computation) For the computation rules, see "Matrix Cookbook" from MIT.

  11. def backward(error: ParSeq[DataVec]): (ParSeq[DataVec], ParSeq[() ⇒ Unit])

    Permalink

    Backward computation using propagated error.

    Backward computation using propagated error.

    error

    Propagated error sequence.

    returns

    Error sequence for back propagation.

    Definition Classes
    Layer
  12. val centers: Weight[Matrix]

    Permalink

    Centroids *

  13. def clone(): AnyRef

    Permalink
    Attributes
    protected[java.lang]
    Definition Classes
    AnyRef
    Annotations
    @throws( ... )
  14. val epsilon: Weight[Matrix]

    Permalink

    Epsilon values vector *

  15. final def eq(arg0: AnyRef): Boolean

    Permalink
    Definition Classes
    AnyRef
  16. def equals(arg0: Any): Boolean

    Permalink
    Definition Classes
    AnyRef → Any
  17. implicit val evidenceI: ClassTag[DataVec]

    Permalink

    ClassTag for input *

    ClassTag for input *

    Attributes
    protected
    Definition Classes
    TransformLayerLayer
  18. def finalize(): Unit

    Permalink
    Attributes
    protected[java.lang]
    Definition Classes
    AnyRef
    Annotations
    @throws( classOf[java.lang.Throwable] )
  19. def forward(in: ParSeq[DataVec]): ParSeq[DataVec]

    Permalink

    Apply Parallel Sequence of vector.

    Apply Parallel Sequence of vector.

    in

    Parallel Sequence of Input

    returns

    Parallel Sequence of Vector

    Definition Classes
    Layer
  20. final def forward(in: RDD[(Long, DataVec)]): RDD[(Long, DataVec)]

    Permalink

    Apply RDD of vector.

    Apply RDD of vector.

    in

    RDD of (ID, Value), ID is Long value.

    returns

    RDD of (ID, Vector), with same ID for input.

    Definition Classes
    Layer
  21. final def forward(in: DataVec): DataVec

    Permalink

    Apply this layer (forward computation)

    Apply this layer (forward computation)

    in

    Input value

    returns

    Output Vector

    Definition Classes
    Layer
  22. final def getClass(): Class[_]

    Permalink
    Definition Classes
    AnyRef → Any
  23. def hashCode(): Int

    Permalink
    Definition Classes
    AnyRef → Any
  24. def initiateBy(builder: WeightBuilder): VectorRBFLayer.this.type

    Permalink

    Set weight builder for this layer.

    Set weight builder for this layer.

    builder

    Weight builder to be applied

    returns

    self

    Definition Classes
    VectorRBFLayerLayer
  25. var inoutSEQ: ParSeq[(DataVec, DataVec)]

    Permalink

    Sequence for backpropagation.

    Sequence for backpropagation. Stores output values. *

    Attributes
    protected
    Definition Classes
    Layer
  26. final def isInstanceOf[T0]: Boolean

    Permalink
    Definition Classes
    Any
  27. var isUpdatable: Boolean

    Permalink

    True if this layer affected by backward propagation *

    True if this layer affected by backward propagation *

    Attributes
    protected
    Definition Classes
    Layer
  28. def loss: Double

    Permalink

    Weight Loss of this layer

    Weight Loss of this layer

    returns

    Weight loss

    Definition Classes
    VectorRBFLayerLayer
  29. final def ne(arg0: AnyRef): Boolean

    Permalink
    Definition Classes
    AnyRef
  30. final def notify(): Unit

    Permalink
    Definition Classes
    AnyRef
  31. final def notifyAll(): Unit

    Permalink
    Definition Classes
    AnyRef
  32. final val outVecOf: (DataVec) ⇒ DataVec

    Permalink

    Output converter from OutInfo to Vector *

    Output converter from OutInfo to Vector *

    Definition Classes
    TransformLayerLayer
  33. def read(kryo: Kryo, input: Input): Unit

    Permalink
    Definition Classes
    VectorRBFLayerLayer → KryoSerializable
  34. def setUpdatable(bool: Boolean): Unit

    Permalink

    Assign whether this layer updatable or not.

    Assign whether this layer updatable or not. value.

    bool

    True if this layer used in backpropagation.

    Definition Classes
    Layer
  35. final def synchronized[T0](arg0: ⇒ T0): T0

    Permalink
    Definition Classes
    AnyRef
  36. def toString(): String

    Permalink
    Definition Classes
    AnyRef → Any
  37. final def wait(): Unit

    Permalink
    Definition Classes
    AnyRef
    Annotations
    @throws( ... )
  38. final def wait(arg0: Long, arg1: Int): Unit

    Permalink
    Definition Classes
    AnyRef
    Annotations
    @throws( ... )
  39. final def wait(arg0: Long): Unit

    Permalink
    Definition Classes
    AnyRef
    Annotations
    @throws( ... )
  40. def withActivation(act: RBFActivation): VectorRBFLayer.this.type

    Permalink

    Set the activation function.

    Set the activation function.

    act

    Activation function.

    returns

    self

  41. def withCenters(centroid: Seq[DataVec], minDistances: Int = 3, alpha: Double = 1.0): VectorRBFLayer.this.type

    Permalink

    Set centroids

    Set centroids

    centroid

    Sequence of Centroids to be set.

    minDistances

    Minimum number of distances will be used in epsilon initialization.

    alpha

    Scaling paramater alpha, when determining epsilon.

    returns

    self

  42. def write(kryo: Kryo, output: Output): Unit

    Permalink
    Definition Classes
    VectorRBFLayerLayer → KryoSerializable

Deprecated Value Members

  1. def withInput(in: Int): VectorRBFLayer.this.type

    Permalink

    Set input size

    Set input size

    in

    Size of input

    returns

    self

    Definition Classes
    VectorRBFLayerLayer
    Annotations
    @deprecated
    Deprecated

    Input size automatically set when centroid attached

  2. def withOutput(out: Int): VectorRBFLayer.this.type

    Permalink

    Set output size

    Set output size

    out

    Size of output

    returns

    self

    Definition Classes
    VectorRBFLayerLayer
    Annotations
    @deprecated
    Deprecated

    Output size automatically set when centroid attached

Inherited from TransformLayer

Inherited from Layer[DataVec, DataVec]

Inherited from KryoSerializable

Inherited from Serializable

Inherited from Serializable

Inherited from AnyRef

Inherited from Any

Ungrouped