com.intel.analytics.bigdl.nn

Graph

object Graph extends GraphSerializable with Serializable

Linear Supertypes
Serializable, Serializable, GraphSerializable, ContainerSerializable, ModuleSerializable, Savable, Loadable, AnyRef, Any
Ordering
  1. Alphabetic
  2. By inheritance
Inherited
  1. Graph
  2. Serializable
  3. Serializable
  4. GraphSerializable
  5. ContainerSerializable
  6. ModuleSerializable
  7. Savable
  8. Loadable
  9. AnyRef
  10. Any
  1. Hide All
  2. Show all
Learn more about member selection
Visibility
  1. Public
  2. All

Type Members

  1. type ModuleNode[T] = Node[AbstractModule[Activity, Activity, T]]

    Node for graph container.

    Node for graph container. The module should have a tensor/table input while a tensor output

Value Members

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

    Definition Classes
    AnyRef
  2. final def !=(arg0: Any): Boolean

    Definition Classes
    Any
  3. final def ##(): Int

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

    Definition Classes
    AnyRef
  5. final def ==(arg0: Any): Boolean

    Definition Classes
    Any
  6. var _copyWeightAndBias: Boolean

    Attributes
    protected
    Definition Classes
    ModuleSerializable
  7. def apply[T](input: ModuleNode[T], output: ModuleNode[T])(implicit arg0: ClassTag[T], ev: TensorNumeric[T]): Graph[T]

    Build a single input, single output graph container

    Build a single input, single output graph container

    input

    input nodes

    output

    output nodes

    returns

    a graph container

  8. def apply[T](input: Array[ModuleNode[T]], output: ModuleNode[T])(implicit arg0: ClassTag[T], ev: TensorNumeric[T]): Graph[T]

    Build a multiple inputs, single output graph container

    Build a multiple inputs, single output graph container

    input

    input nodes

    output

    output node

    returns

    a graph container

  9. def apply[T](input: ModuleNode[T], output: Array[ModuleNode[T]])(implicit arg0: ClassTag[T], ev: TensorNumeric[T]): Graph[T]

    Build a single input, multiple outputs graph container

    Build a single input, multiple outputs graph container

    input

    input node

    output

    output nodes

    returns

    a graph container

  10. def apply[T](preprocessor: Module[T], trainable: Module[T])(implicit arg0: ClassTag[T], ev: TensorNumeric[T]): Graph[T]

  11. def apply[T](input: Array[ModuleNode[T]], output: Array[ModuleNode[T]], variables: Option[(Array[Tensor[T]], Array[Tensor[T]])] = None)(implicit arg0: ClassTag[T], ev: TensorNumeric[T]): Graph[T]

    Build multiple inputs, multiple outputs graph container.

    Build multiple inputs, multiple outputs graph container.

    input

    input node

    output

    output node

    returns

    a graph container

  12. final def asInstanceOf[T0]: T0

    Definition Classes
    Any
  13. def attrValueToFloatArray(attr: AttrValue): Array[Float]

    Convert Attr Value object to Array of Float

    Convert Attr Value object to Array of Float

    attr
    returns

    Array[Float]

    Attributes
    protected
    Definition Classes
    ModuleSerializable
  14. def checkVersion[T](module: BigDLModule)(implicit arg0: ClassTag[T], ev: TensorNumeric[T]): Unit

    Attributes
    protected
    Definition Classes
    ModuleSerializable
  15. def clone(): AnyRef

    Attributes
    protected[java.lang]
    Definition Classes
    AnyRef
    Annotations
    @throws( ... )
  16. def copy2BigDL[T](context: DeserializeContext, module: ModuleData[T])(implicit arg0: ClassTag[T], ev: TensorNumeric[T]): Unit

    copy serialized data (weight and bias if exist) to BigDL module

    copy serialized data (weight and bias if exist) to BigDL module

    context

    deserialized context

    module

    bigDL Module with relationships

    Attributes
    protected
    Definition Classes
    ModuleSerializable
  17. def copyFromBigDL[T](context: SerializeContext[T], modelBuilder: Builder)(implicit arg0: ClassTag[T], ev: TensorNumeric[T]): Unit

    copy BigDL module data (weight and bias if exist) to BigDL Model to be persisted

    copy BigDL module data (weight and bias if exist) to BigDL Model to be persisted

    context

    serialization context

    modelBuilder

    serialized module builder

    Attributes
    protected
    Definition Classes
    ModuleSerializable
  18. def copyWeightAndBias(): Boolean

    Attributes
    protected
    Definition Classes
    ModuleSerializable
  19. def createBigDLModule[T](context: DeserializeContext, module: AbstractModule[Activity, Activity, T])(implicit arg0: ClassTag[T], ev: TensorNumeric[T]): ModuleData[T]

    Re-create BigDL module by deserializing protobuf context.

    Re-create BigDL module by deserializing protobuf context.

    T
    context

    Deserialization context

    module

    The BigDL module to be re-created

    ev
    returns

    Tuple3 contains information of current module and modules adjacent to it

    Attributes
    protected
    Definition Classes
    ModuleSerializable
  20. def createSerializeBigDLModule[T](modelBuilder: Builder, context: SerializeContext[T])(implicit arg0: ClassTag[T], ev: TensorNumeric[T]): SerializeResult

    Attributes
    protected
    Definition Classes
    ModuleSerializable
  21. def doLoadModule[T](context: DeserializeContext)(implicit arg0: ClassTag[T], ev: TensorNumeric[T]): AbstractModule[Activity, Activity, T]

    Default deserialization using reflection

    Default deserialization using reflection

    context

    deserialize context

    returns

    BigDL module

    Definition Classes
    GraphSerializableContainerSerializableModuleSerializable
  22. def doSerializeModule[T](context: SerializeContext[T], graphBuilder: Builder)(implicit arg0: ClassTag[T], ev: TensorNumeric[T]): Unit

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

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

    Definition Classes
    AnyRef → Any
  25. def finalize(): Unit

    Attributes
    protected[java.lang]
    Definition Classes
    AnyRef
    Annotations
    @throws( classOf[java.lang.Throwable] )
  26. final def getClass(): Class[_]

    Definition Classes
    AnyRef → Any
  27. def getLock: AnyRef

    Attributes
    protected
    Definition Classes
    ModuleSerializable
  28. def getTypes(context: DeserializeContext): (Array[ClassTag[_]], Array[TensorNumeric[_]])

    Attributes
    protected
    Definition Classes
    ModuleSerializable
  29. def hashCode(): Int

    Definition Classes
    AnyRef → Any
  30. final def isInstanceOf[T0]: Boolean

    Definition Classes
    Any
  31. def loadModule[T](context: DeserializeContext)(implicit arg0: ClassTag[T], ev: TensorNumeric[T]): ModuleData[T]

    Default deserialization to provide the template

    Default deserialization to provide the template

    returns

    BigDL module instance with linkages with other modules

    Definition Classes
    ModuleSerializableLoadable
  32. def loadSubModules[T](context: DeserializeContext, module: AbstractModule[Activity, Activity, T])(implicit arg0: ClassTag[T], ev: TensorNumeric[T]): Unit

    Attributes
    protected
    Definition Classes
    ContainerSerializable
  33. val lock: AnyRef

    Attributes
    protected
    Definition Classes
    ModuleSerializable
  34. final def ne(arg0: AnyRef): Boolean

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

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

    Definition Classes
    AnyRef
  37. def saveMklInt8Attr[T](module: MklInt8Convertible, modelBuilder: Builder)(implicit arg0: ClassTag[T], ev: TensorNumeric[T]): Unit

    Serialize and save MKL DNN INT8 attributes into BigDL Model of protobuf definition

    Serialize and save MKL DNN INT8 attributes into BigDL Model of protobuf definition

    module
    modelBuilder

    serialized module builder

    Attributes
    protected
    Definition Classes
    ModuleSerializable
  38. def serializeModule[T](context: SerializeContext[T])(implicit arg0: ClassTag[T], ev: TensorNumeric[T]): SerializeResult

    Default serialization skeleton using reflection

    Default serialization skeleton using reflection

    context

    Serialization context

    returns

    serialized protobuf module instace

    Definition Classes
    ModuleSerializableSavable
  39. def serializeSubModules[T](context: SerializeContext[T], containerBuilder: Builder)(implicit arg0: ClassTag[T], ev: TensorNumeric[T]): Unit

    Attributes
    protected
    Definition Classes
    ContainerSerializable
  40. def setCopyWeightAndBias(copyWeightAndBias: Boolean): Graph.this.type

    Definition Classes
    ModuleSerializable
  41. def setDataTypes[T](context: SerializeContext[T], bigDLModelBuilder: Builder)(implicit arg0: ClassTag[T], ev: TensorNumeric[T]): Unit

    Attributes
    protected
    Definition Classes
    ModuleSerializable
  42. def setVersion[T](modelBuilder: Builder)(implicit arg0: ClassTag[T], ev: TensorNumeric[T]): Unit

    Attributes
    protected
    Definition Classes
    ModuleSerializable
  43. final def synchronized[T0](arg0: ⇒ T0): T0

    Definition Classes
    AnyRef
  44. def toString(): String

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

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

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

    Definition Classes
    AnyRef
    Annotations
    @throws( ... )

Inherited from Serializable

Inherited from Serializable

Inherited from GraphSerializable

Inherited from ContainerSerializable

Inherited from ModuleSerializable

Inherited from Savable

Inherited from Loadable

Inherited from AnyRef

Inherited from Any

Ungrouped