Class

toolkit.neuralnetwork.function

SpatialConvolution

Related Doc: package function

Permalink

case class SpatialConvolution(input: DifferentiableField, weights: DifferentiableField, border: libcog.BorderPolicy = BorderValid, stride: Int = 1) extends DifferentiableField with Product with Serializable

A convolutional transformation that applies a filter bank to a signal in the space domain.

input

The signal node

weights

The filter bank input, typically a TrainableState field

stride

The factor by which the output with be downsampled after the BorderValid convolution

Linear Supertypes
Serializable, Serializable, Product, Equals, DifferentiableField, GradientPropagation, DifferentiableFieldOps, BasicOps, AnyRef, Any
Ordering
  1. Alphabetic
  2. By inheritance
Inherited
  1. SpatialConvolution
  2. Serializable
  3. Serializable
  4. Product
  5. Equals
  6. DifferentiableField
  7. GradientPropagation
  8. DifferentiableFieldOps
  9. BasicOps
  10. AnyRef
  11. Any
  1. Hide All
  2. Show all
Visibility
  1. Public
  2. All

Instance Constructors

  1. new SpatialConvolution(input: DifferentiableField, weights: DifferentiableField, border: libcog.BorderPolicy = BorderValid, stride: Int = 1)

    Permalink

    input

    The signal node

    weights

    The filter bank input, typically a TrainableState field

    stride

    The factor by which the output with be downsampled after the BorderValid convolution

Value Members

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

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

    Permalink
    Definition Classes
    AnyRef → Any
  3. def *(that: Float): DifferentiableField

    Permalink
    Definition Classes
    DifferentiableFieldOps
  4. def *(that: DifferentiableField): DifferentiableField

    Permalink
    Definition Classes
    DifferentiableFieldOps
  5. def +(that: Float): DifferentiableField

    Permalink
    Definition Classes
    DifferentiableFieldOps
  6. def +(that: DifferentiableField): DifferentiableField

    Permalink
    Definition Classes
    DifferentiableFieldOps
  7. def -(that: Float): DifferentiableField

    Permalink
    Definition Classes
    DifferentiableFieldOps
  8. def -(that: DifferentiableField): DifferentiableField

    Permalink
    Definition Classes
    DifferentiableFieldOps
  9. def /(that: Float): DifferentiableField

    Permalink
    Definition Classes
    DifferentiableFieldOps
  10. def /(that: DifferentiableField): DifferentiableField

    Permalink
    Definition Classes
    DifferentiableFieldOps
  11. final def ==(arg0: Any): Boolean

    Permalink
    Definition Classes
    AnyRef → Any
  12. def activateSGD(initField: libcog.Field = ScalarField(1f), invokeCallbacks: Boolean = true): Unit

    Permalink
    Definition Classes
    GradientPropagation
  13. def add(input: DifferentiableField, c: Float): DifferentiableField

    Permalink
    Definition Classes
    BasicOps
  14. def add(left: DifferentiableField, right: DifferentiableField): DifferentiableField

    Permalink
    Definition Classes
    BasicOps
  15. final def asInstanceOf[T0]: T0

    Permalink
    Definition Classes
    Any
  16. var backward: Option[libcog.Field]

    Permalink
    Definition Classes
    DifferentiableField
  17. def backwardCallback(back: libcog.Field): Unit

    Permalink
    Definition Classes
    DifferentiableField
  18. val batchSize: Int

    Permalink
  19. val border: libcog.BorderPolicy

    Permalink
  20. def clone(): AnyRef

    Permalink
    Attributes
    protected[java.lang]
    Definition Classes
    AnyRef
    Annotations
    @throws( ... )
  21. def divide(input: DifferentiableField, c: Float): DifferentiableField

    Permalink
    Definition Classes
    BasicOps
  22. def divide(left: DifferentiableField, right: DifferentiableField): DifferentiableField

    Permalink
    Definition Classes
    BasicOps
  23. final def eq(arg0: AnyRef): Boolean

    Permalink
    Definition Classes
    AnyRef
  24. def finalize(): Unit

    Permalink
    Attributes
    protected[java.lang]
    Definition Classes
    AnyRef
    Annotations
    @throws( classOf[java.lang.Throwable] )
  25. val forward: libcog.Field

    Permalink
  26. final def getClass(): Class[_]

    Permalink
    Definition Classes
    AnyRef → Any
  27. var gradientBinding: Option[GradientBinding]

    Permalink
    Definition Classes
    DifferentiableField
  28. val gradientConsumer: Boolean

    Permalink
    Definition Classes
    DifferentiableField
  29. val input: DifferentiableField

    Permalink

    The signal node

  30. val inputs: Map[Symbol, GradientPort]

    Permalink
  31. final def isInstanceOf[T0]: Boolean

    Permalink
    Definition Classes
    Any
  32. def multiply(input: DifferentiableField, c: Float): DifferentiableField

    Permalink
    Definition Classes
    BasicOps
  33. def multiply(left: DifferentiableField, right: DifferentiableField): DifferentiableField

    Permalink
    Definition Classes
    BasicOps
  34. final def ne(arg0: AnyRef): Boolean

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

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

    Permalink
    Definition Classes
    AnyRef
  37. def pow(input: DifferentiableField, n: Float): DifferentiableField

    Permalink

    Raise a node to a fixed power.

    Raise a node to a fixed power. Cog has two pow() function signatures corresponding to both integer and non-integer powers. The integer case is detected here and special-cased (instead of having a separate PowN node for this).

    If the power n is anything other than a positive integer, make sure the inputs are always positive or NaNs will result.

    input

    the input signal

    n

    the power to raise the input to

    Definition Classes
    BasicOps
  38. val stride: Int

    Permalink

    The factor by which the output with be downsampled after the BorderValid convolution

  39. def subtract(input: DifferentiableField, c: Float): DifferentiableField

    Permalink
    Definition Classes
    BasicOps
  40. def subtract(left: DifferentiableField, right: DifferentiableField): DifferentiableField

    Permalink
    Definition Classes
    BasicOps
  41. final def synchronized[T0](arg0: ⇒ T0): T0

    Permalink
    Definition Classes
    AnyRef
  42. def totalDerivative(): libcog.Field

    Permalink
    Definition Classes
    GradientPropagation
  43. final def wait(): Unit

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

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

    Permalink
    Definition Classes
    AnyRef
    Annotations
    @throws( ... )
  46. val weights: DifferentiableField

    Permalink

    The filter bank input, typically a TrainableState field

Inherited from Serializable

Inherited from Serializable

Inherited from Product

Inherited from Equals

Inherited from DifferentiableField

Inherited from GradientPropagation

Inherited from DifferentiableFieldOps

Inherited from BasicOps

Inherited from AnyRef

Inherited from Any

Ungrouped