com.intel.analytics.bigdl.nn

Graph

abstract class Graph[T] extends Container[Activity, Activity, T] with MklInt8Convertible

A graph container. The modules in the container are connected as a directed Graph. Each module can output one tensor or multiple tensors(as table). The edges between modules in the graph define how these tensors are passed. For example, if a module outputs two tensors, you can pass these two tensors together to its following module, or pass only one of them to its following module. If a tensor in the module output is connected to multiple modules, in the back propagation, the gradients from multiple connection will be accumulated. If multiple edges point to one module, the tensors from these edges will be stack as a table, then pass to that module. In the back propagation, the gradients will be splited based on how the input tensors stack.

The graph container has multiple inputs and multiple outputs. The order of the input tensors should be same with the order of the input nodes when you construct the graph container. In the back propagation, the order of the gradients tensors should be the same with the order of the output nodes.

If there's one output, the module output is a tensor. If there're multiple outputs, the module output is a table, which is actually an sequence of tensor. The order of the output tensors is same with the order of the output modules.

All inputs should be able to connect to outputs through some paths in the graph. It is allowed that some successors of the inputs node are not connect to outputs. If so, these nodes will be excluded in the computation.

T

Numeric type. Only support float/double now

Annotations
@SerialVersionUID( 2896121321564992779L )
Linear Supertypes
MklInt8Convertible, Container[Activity, Activity, T], AbstractModule[Activity, Activity, T], InferShape, Serializable, Serializable, AnyRef, Any
Known Subclasses
Ordering
  1. Alphabetic
  2. By inheritance
Inherited
  1. Graph
  2. MklInt8Convertible
  3. Container
  4. AbstractModule
  5. InferShape
  6. Serializable
  7. Serializable
  8. AnyRef
  9. Any
  1. Hide All
  2. Show all
Learn more about member selection
Visibility
  1. Public
  2. All

Instance Constructors

  1. new Graph(inputs: Seq[ModuleNode[T]], outputs: Seq[ModuleNode[T]], variables: Option[(Array[Tensor[T]], Array[Tensor[T]])] = scala.None)(implicit arg0: ClassTag[T], ev: TensorNumeric[T])

    inputs

    input nodes

    outputs

    output nodes

    variables

    an Array of tensor containing all the weights and biases of this graph, used when different nodes of this graph may share the same weight or bias.

Abstract Value Members

  1. abstract def populateModules(): Unit

    Attributes
    protected
  2. abstract def updateGradInput(input: Activity, gradOutput: Activity): Activity

    Computing the gradient of the module with respect to its own input.

    Computing the gradient of the module with respect to its own input. This is returned in gradInput. Also, the gradInput state variable is updated accordingly.

    input
    gradOutput
    returns

    Definition Classes
    AbstractModule
  3. abstract def updateOutput(input: Activity): Activity

    Computes the output using the current parameter set of the class and input.

    Computes the output using the current parameter set of the class and input. This function returns the result which is stored in the output field.

    input
    returns

    Definition Classes
    AbstractModule

Concrete 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. def accActivity(activity: Activity, other: Activity): Activity

    Attributes
    protected
    Annotations
    @inline()
  7. def accGradParameters(input: Activity, gradOutput: Activity): Unit

    Computing the gradient of the module with respect to its own parameters.

    Computing the gradient of the module with respect to its own parameters. Many modules do not perform this step as they do not have any parameters. The state variable name for the parameters is module dependent. The module is expected to accumulate the gradients with respect to the parameters in some variable.

    input
    gradOutput

    Definition Classes
    AbstractModule
  8. def addZeroTensorToMissingGradOutput(output: Table, gradOutput: Table): Unit

    For a multi-tensor output module, some output tensors may not contributed to the final forward result.

    For a multi-tensor output module, some output tensors may not contributed to the final forward result. So in the back propagation, the gradient on these positions are missing. And we use zero tensor to populate.

    output
    gradOutput

    Attributes
    protected
  9. def apply(name: String): Option[AbstractModule[Activity, Activity, T]]

    Find a module with given name.

    Find a module with given name. If there is no module with given name, it will return None. If there are multiple modules with the given name, an exception will be thrown.

    name
    returns

    Definition Classes
    ContainerAbstractModule
  10. final def asInstanceOf[T0]: T0

    Definition Classes
    Any
  11. def backward(input: Activity, gradOutput: Activity): Activity

    Performs a back-propagation step through the module, with respect to the given input.

    Performs a back-propagation step through the module, with respect to the given input. In general this method makes the assumption forward(input) has been called before, with the same input. This is necessary for optimization reasons. If you do not respect this rule, backward() will compute incorrect gradients.

    input

    input data

    gradOutput

    gradient of next layer

    returns

    gradient corresponding to input data

    Definition Classes
    AbstractModule
  12. var backwardGraph: DirectedGraph[AbstractModule[Activity, Activity, T]]

    Attributes
    protected
  13. var backwardNodes: Array[Node[AbstractModule[Activity, Activity, T]]]

    Attributes
    protected
  14. var backwardTime: Long

    Attributes
    protected
    Definition Classes
    AbstractModule
  15. def canEqual(other: Any): Boolean

    Definition Classes
    ContainerAbstractModule
  16. final def checkEngineType(): Graph.this.type

    get execution engine type

    get execution engine type

    Definition Classes
    ContainerAbstractModule
  17. def clearState(): Graph.this.type

    Clear cached activities to save storage space or network bandwidth.

    Clear cached activities to save storage space or network bandwidth. Note that we use Tensor.set to keep some information like tensor share

    The subclass should override this method if it allocate some extra resource, and call the super.clearState in the override method

    returns

    Definition Classes
    ContainerAbstractModule
  18. final def clone(deepCopy: Boolean): AbstractModule[Activity, Activity, T]

    Clone the module, deep or shallow copy

    Clone the module, deep or shallow copy

    deepCopy
    returns

    Definition Classes
    AbstractModule
  19. def clone(): AnyRef

    Attributes
    protected[java.lang]
    Definition Classes
    AnyRef
    Annotations
    @throws( ... )
  20. final def cloneModule(): Graph.this.type

    Clone the model

    Clone the model

    returns

    Definition Classes
    AbstractModule
  21. val dummyOutput: Node[AbstractModule[Activity, Activity, T]]

    Attributes
    protected
  22. var dummyOutputGrad: ModuleNode[T]

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

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

    Definition Classes
    ContainerAbstractModule → AnyRef → Any
  25. final def evaluate(): Graph.this.type

    Set the module to evaluate mode

    Set the module to evaluate mode

    returns

    Definition Classes
    ContainerAbstractModule
  26. final def evaluate(dataSet: LocalDataSet[MiniBatch[T]], vMethods: Array[_ <: ValidationMethod[T]]): Array[(ValidationResult, ValidationMethod[T])]

    use ValidationMethod to evaluate module on the given local dataset

    use ValidationMethod to evaluate module on the given local dataset

    dataSet
    vMethods
    returns

    Definition Classes
    AbstractModule
  27. final def evaluate(dataset: RDD[MiniBatch[T]], vMethods: Array[_ <: ValidationMethod[T]]): Array[(ValidationResult, ValidationMethod[T])]

    use ValidationMethod to evaluate module on the given rdd dataset

    use ValidationMethod to evaluate module on the given rdd dataset

    dataset
    vMethods
    returns

    Definition Classes
    AbstractModule
  28. final def evaluate(dataset: RDD[Sample[T]], vMethods: Array[_ <: ValidationMethod[T]], batchSize: Option[Int] = None): Array[(ValidationResult, ValidationMethod[T])]

    use ValidationMethod to evaluate module on the given rdd dataset

    use ValidationMethod to evaluate module on the given rdd dataset

    dataset

    dataset for test

    vMethods

    validation methods

    batchSize

    total batchsize of all partitions, optional param and default 4 * partitionNum of dataset

    returns

    Definition Classes
    AbstractModule
  29. final def evaluateImage(imageFrame: ImageFrame, vMethods: Array[_ <: ValidationMethod[T]], batchSize: Option[Int] = None): Array[(ValidationResult, ValidationMethod[T])]

    use ValidationMethod to evaluate module on the given ImageFrame

    use ValidationMethod to evaluate module on the given ImageFrame

    imageFrame

    ImageFrame for valudation

    vMethods

    validation methods

    batchSize

    total batch size of all partitions

    returns

    Definition Classes
    AbstractModule
  30. def fetchModelGradInput(): Activity

    Attributes
    protected
  31. def finalize(): Unit

    Attributes
    protected[java.lang]
    Definition Classes
    AnyRef
    Annotations
    @throws( classOf[java.lang.Throwable] )
  32. def findGradOutput(curNode: ModuleNode[T], gradOutput: Activity): Activity

    Attributes
    protected
  33. def findInput(node: ModuleNode[T], input: Activity): Activity

  34. def findModules(moduleType: String): ArrayBuffer[AbstractModule[_, _, T]]

    Definition Classes
    Container
  35. final def forward(input: Activity): Activity

    Takes an input object, and computes the corresponding output of the module.

    Takes an input object, and computes the corresponding output of the module. After a forward, the output state variable should have been updated to the new value.

    input

    input data

    returns

    output data

    Definition Classes
    AbstractModule
  36. val forwardGraph: DirectedGraph[AbstractModule[Activity, Activity, T]]

    Attributes
    protected
  37. val forwardNodes: Array[Node[AbstractModule[Activity, Activity, T]]]

    Attributes
    protected
  38. var forwardTime: Long

    Attributes
    protected
    Definition Classes
    AbstractModule
  39. def freeze(names: Array[String]): Graph.this.type

    set an array of layers that match the given names to be "freezed", i.

    set an array of layers that match the given names to be "freezed", i.e. their parameters(weight/bias, if exists) are not changed in training process

    names

    an array of layer names

    returns

    current graph model

  40. def freeze(names: String*): Graph.this.type

    freeze the module, i.

    freeze the module, i.e. their parameters(weight/bias, if exists) are not changed in training process if names is not empty, set an array of layers that match the given names to be "freezed",

    names

    an array of layer names

    returns

    current graph model

    Definition Classes
    ContainerAbstractModule
  41. final def getClass(): Class[_]

    Definition Classes
    AnyRef → Any
  42. def getExtraParameter(): Array[Tensor[T]]

    Get extra parameter in this module.

    Get extra parameter in this module. Extra parameter means the trainable parameters beside weight and bias. Such as runningMean and runningVar in BatchNormalization.

    The subclass should override this method if it has some parameters besides weight and bias.

    returns

    an array of tensor

    Definition Classes
    ContainerAbstractModule
  43. def getForwardExecutions(): Array[Node[AbstractModule[Activity, Activity, T]]]

    Get forward executions, the dummy node will be filtered.

    Get forward executions, the dummy node will be filtered.

    This method will output an unsorted executions.

    returns

  44. def getInput(node: Node[AbstractModule[Activity, Activity, T]], input: Activity): Activity

    Attributes
    protected
  45. def getInputDimMask(): Int

    Get dimension mask of input

    Get dimension mask of input

    returns

    inputDimMask field which stores value of input dimension mask

    Definition Classes
    MklInt8Convertible
  46. def getInputScales(): Array[Array[Float]]

    Get input scales

    Get input scales

    returns

    field which stores value of input scales

    Definition Classes
    MklInt8Convertible
  47. final def getInputShape(): Shape

    Return the inputShape for the current Layer and the first dim is batch.

    Return the inputShape for the current Layer and the first dim is batch.

    Definition Classes
    InferShape
  48. final def getName(): String

    Get the module name, default name is className@namePostfix

    Get the module name, default name is className@namePostfix

    returns

    Definition Classes
    AbstractModule
  49. final def getNumericType(): TensorDataType

    Get numeric type of module parameters

    Get numeric type of module parameters

    returns

    Definition Classes
    AbstractModule
  50. def getOutputDimMask(): Int

    Get dimension mask of output

    Get dimension mask of output

    returns

    outputDimMask field which stores value of output dimension mask

    Definition Classes
    MklInt8Convertible
  51. def getOutputScales(): Array[Array[Float]]

    Get output scales

    Get output scales

    returns

    field which stores value of output scales

    Definition Classes
    MklInt8Convertible
  52. final def getOutputShape(): Shape

    Return the outputShape for the current Layer and the first dim is batch.

    Return the outputShape for the current Layer and the first dim is batch.

    Definition Classes
    InferShape
  53. def getParametersTable(): Table

    This function returns a table contains ModuleName, the parameter names and parameter value in this module.

    This function returns a table contains ModuleName, the parameter names and parameter value in this module.

    The result table is a structure of Table(ModuleName -> Table(ParameterName -> ParameterValue)), and the type is Table[String, Table[String, Tensor[T]]].

    For example, get the weight of a module named conv1: table[Table]("conv1")[Tensor[T]]("weight").

    The names of the parameters follow such convention:

    1. If there's one parameter, the parameter is named as "weight", the gradient is named as "gradWeight"

    2. If there're two parameters, the first parameter is named as "weight", the first gradient is named as "gradWeight"; the second parameter is named as "bias", the seconcd gradient is named as "gradBias"

    3. If there're more parameters, the weight is named as "weight" with a seq number as suffix, the gradient is named as "gradient" with a seq number as suffix

    Custom modules should override this function the default impl if the convention doesn't meet the requirement.

    returns

    Table

    Definition Classes
    ContainerAbstractModule
  54. final def getPrintName(): String

    Attributes
    protected
    Definition Classes
    AbstractModule
  55. final def getScaleB(): Double

    Get the scale of gradientBias

    Get the scale of gradientBias

    Definition Classes
    AbstractModule
  56. final def getScaleW(): Double

    Get the scale of gradientWeight

    Get the scale of gradientWeight

    Definition Classes
    AbstractModule
  57. def getSortedForwardExecutions(): Array[ModuleNode[T]]

    Get forward executions, the dummy nodes and control dependency nodes will be filtered.

    Get forward executions, the dummy nodes and control dependency nodes will be filtered.

    This method will output a sorted executions. If the graph contains loop, it will throw an exception

    returns

  58. def getStopGradientLayers(): HashSet[String]

  59. def getTimes(): Array[(AbstractModule[_ <: Activity, _ <: Activity, T], Long, Long)]

    Get the forward/backward cost time for the module or its submodules

    Get the forward/backward cost time for the module or its submodules

    returns

    Definition Classes
    GraphContainerAbstractModule
  60. final def getTimesGroupByModuleType(): Array[(String, Long, Long)]

    Get the forward/backward cost time for the module or its submodules and group by module type.

    Get the forward/backward cost time for the module or its submodules and group by module type.

    returns

    (module type name, forward time, backward time)

    Definition Classes
    AbstractModule
  61. def getWeightDimMask(): Int

    Get dimension mask of weight

    Get dimension mask of weight

    returns

    weightDimMask which stores value of weight mask

    Definition Classes
    MklInt8Convertible
  62. def getWeightScales(): Array[Array[Float]]

    Get weight scales

    Get weight scales

    returns

    field which stores value of weight scales

    Definition Classes
    MklInt8Convertible
  63. final def getWeightsBias(): Array[Tensor[T]]

    Get weight and bias for the module

    Get weight and bias for the module

    returns

    array of weights and bias

    Definition Classes
    AbstractModule
  64. var gradInput: Activity

    The cached gradient of activities.

    The cached gradient of activities. So we don't compute it again when need it

    Definition Classes
    AbstractModule
  65. final def hasName: Boolean

    Whether user set a name to the module before

    Whether user set a name to the module before

    returns

    Definition Classes
    AbstractModule
  66. def hashCode(): Int

    Definition Classes
    ContainerAbstractModule → AnyRef → Any
  67. var inputDimMask: Int

    Attributes
    protected
    Definition Classes
    MklInt8Convertible
  68. def inputs(first: (ModuleNode[T], Int), nodesWithIndex: (ModuleNode[T], Int)*): ModuleNode[T]

    Build graph: some other modules point to current module

    Build graph: some other modules point to current module

    first

    distinguish from another inputs when input parameter list is empty

    nodesWithIndex

    upstream module nodes and the output tensor index. The start index is 1.

    returns

    node containing current module

    Definition Classes
    AbstractModule
  69. def inputs(nodes: Array[ModuleNode[T]]): ModuleNode[T]

    Build graph: some other modules point to current module

    Build graph: some other modules point to current module

    nodes

    upstream module nodes in an array

    returns

    node containing current module

    Definition Classes
    AbstractModule
  70. def inputs(nodes: ModuleNode[T]*): ModuleNode[T]

    Build graph: some other modules point to current module

    Build graph: some other modules point to current module

    nodes

    upstream module nodes

    returns

    node containing current module

    Definition Classes
    AbstractModule
  71. val inputs: Seq[ModuleNode[T]]

    input nodes

  72. final def isInstanceOf[T0]: Boolean

    Definition Classes
    Any
  73. def isStopGradient(module: AbstractModule[_ <: Activity, _ <: Activity, T]): Boolean

    whether stop propagating gradInput back

    whether stop propagating gradInput back

    returns

    Attributes
    protected
  74. final def isTraining(): Boolean

    Check if the model is in training mode

    Check if the model is in training mode

    returns

    Definition Classes
    AbstractModule
  75. var line: String

    Attributes
    protected
    Definition Classes
    AbstractModule
  76. final def loadModelWeights(srcModel: Module[Float], matchAll: Boolean = true): Graph.this.type

    copy weights from another model, mapping by layer name

    copy weights from another model, mapping by layer name

    srcModel

    model to copy from

    matchAll

    whether to match all layers' weights and bias,

    returns

    current module

    Definition Classes
    AbstractModule
  77. final def loadWeights(weightPath: String, matchAll: Boolean = true): Graph.this.type

    load pretrained weights and bias to current module

    load pretrained weights and bias to current module

    weightPath

    file to store weights and bias

    matchAll

    whether to match all layers' weights and bias, if not, only load existing pretrained weights and bias

    returns

    current module

    Definition Classes
    AbstractModule
  78. val modules: ArrayBuffer[AbstractModule[Activity, Activity, T]]

    Definition Classes
    Container
  79. final def ne(arg0: AnyRef): Boolean

    Definition Classes
    AnyRef
  80. def node(name: String): ModuleNode[T]

    Return the corresponding node has the given name.

    Return the corresponding node has the given name. If the given name doesn't match any node, NoSuchElementException will be thrown

    name
    returns

  81. final def notify(): Unit

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

    Definition Classes
    AnyRef
  83. var output: Activity

    The cached output.

    The cached output. So we don't compute it again when need it

    Definition Classes
    AbstractModule
  84. var outputDimMask: Int

    Attributes
    protected
    Definition Classes
    MklInt8Convertible
  85. def parameters(): (Array[Tensor[T]], Array[Tensor[T]])

    This function returns two arrays.

    This function returns two arrays. One for the weights and the other the gradients Custom modules should override this function if they have parameters

    returns

    (Array of weights, Array of grad)

    Definition Classes
    GraphContainerAbstractModule
  86. final def predict(dataset: RDD[Sample[T]], batchSize: Int = 1, shareBuffer: Boolean = false): RDD[Activity]

    module predict, return the probability distribution

    module predict, return the probability distribution

    dataset

    dataset for prediction

    batchSize

    total batchSize for all partitions. if -1, default is 4 * partitionNumber of datatset

    shareBuffer

    whether to share same memory for each batch predict results

    Definition Classes
    AbstractModule
  87. final def predictClass(dataset: RDD[Sample[T]], batchSize: Int = 1): RDD[Int]

    module predict, return the predict label

    module predict, return the predict label

    dataset

    dataset for prediction

    batchSize

    total batchSize for all partitions. if -1, default is 4 * partitionNumber of dataset

    Definition Classes
    AbstractModule
  88. final def predictImage(imageFrame: ImageFrame, outputLayer: String = null, shareBuffer: Boolean = false, batchPerPartition: Int = 4, predictKey: String = ImageFeature.predict, featurePaddingParam: Option[PaddingParam[T]] = None): ImageFrame

    model predict images, return imageFrame with predicted tensor, if you want to call predictImage multiple times, it is recommended to use Predictor for DistributedImageFrame or LocalPredictor for LocalImageFrame

    model predict images, return imageFrame with predicted tensor, if you want to call predictImage multiple times, it is recommended to use Predictor for DistributedImageFrame or LocalPredictor for LocalImageFrame

    imageFrame

    imageFrame that contains images

    outputLayer

    if outputLayer is not null, the output of layer that matches outputLayer will be used as predicted output

    shareBuffer

    whether to share same memory for each batch predict results

    batchPerPartition

    batch size per partition, default is 4

    predictKey

    key to store predicted result

    featurePaddingParam

    featurePaddingParam if the inputs have variant size

    returns

    Definition Classes
    AbstractModule
  89. def processInputs(first: (ModuleNode[T], Int), nodesWithIndex: (ModuleNode[T], Int)*): ModuleNode[T]

    Attributes
    protected
    Definition Classes
    AbstractModule
  90. def processInputs(nodes: Seq[ModuleNode[T]]): ModuleNode[T]

    Attributes
    protected
    Definition Classes
    AbstractModule
  91. final def quantize(): Module[T]

    Quantize this module, which reduces the precision of the parameter.

    Quantize this module, which reduces the precision of the parameter. Get a higher speed with a little accuracy cost.

    returns

    Definition Classes
    AbstractModule
  92. def release(): Unit

    if the model contains native resources such as aligned memory, we should release it by manual.

    if the model contains native resources such as aligned memory, we should release it by manual. JVM GC can't release them reliably.

    Definition Classes
    ContainerAbstractModule
  93. def reset(): Unit

    Reset module parameters, which is re-initialize the parameter with given initMethod

    Reset module parameters, which is re-initialize the parameter with given initMethod

    Definition Classes
    GraphContainerAbstractModule
  94. def resetModules(): Unit

    Clear the original module and reset with module in the graph

  95. def resetTimes(): Unit

    Reset the forward/backward record time for the module or its submodules

    Reset the forward/backward record time for the module or its submodules

    returns

    Definition Classes
    ContainerAbstractModule
  96. final def saveCaffe(prototxtPath: String, modelPath: String, useV2: Boolean = true, overwrite: Boolean = false): Graph.this.type

    Save this module to path in caffe readable format

    Save this module to path in caffe readable format

    prototxtPath
    modelPath
    useV2
    overwrite
    returns

    Definition Classes
    AbstractModule
  97. final def saveDefinition(path: String, overWrite: Boolean = false): Graph.this.type

    Save this module definition to path.

    Save this module definition to path.

    path

    path to save module, local file system, HDFS and Amazon S3 is supported. HDFS path should be like "hdfs://[host]:[port]/xxx" Amazon S3 path should be like "s3a://bucket/xxx"

    overWrite

    if overwrite

    returns

    self

    Definition Classes
    AbstractModule
  98. def saveGraphTopology(logPath: String, backward: Boolean = false): Graph.this.type

    Save current model graph to a folder, which can be display in tensorboard by running tensorboard --logdir logPath

    Save current model graph to a folder, which can be display in tensorboard by running tensorboard --logdir logPath

    logPath
    backward

    Draw backward graph instead of forward

    returns

  99. final def saveModule(path: String, weightPath: String = null, overWrite: Boolean = false): Graph.this.type

    Save this module to path with protobuf format

    Save this module to path with protobuf format

    path

    path to save module, local file system, HDFS and Amazon S3 is supported. HDFS path should be like "hdfs://[host]:[port]/xxx" Amazon S3 path should be like "s3a://bucket/xxx"

    weightPath

    where to store weight

    overWrite

    if overwrite

    returns

    self

    Definition Classes
    AbstractModule
  100. final def saveTF(inputs: Seq[(String, Seq[Int])], path: String, byteOrder: ByteOrder = ByteOrder.LITTLE_ENDIAN, dataFormat: TensorflowDataFormat = TensorflowDataFormat.NHWC): Graph.this.type

    Save this module to path in tensorflow readable format

    Save this module to path in tensorflow readable format

    inputs
    path
    byteOrder
    dataFormat
    returns

    Definition Classes
    AbstractModule
  101. final def saveTorch(path: String, overWrite: Boolean = false): Graph.this.type

    Save this module to path in torch7 readable format

    Save this module to path in torch7 readable format

    path
    overWrite
    returns

    Definition Classes
    AbstractModule
  102. final def saveWeights(path: String, overWrite: Boolean): Unit

    save weights and bias to file

    save weights and bias to file

    path

    file to save

    overWrite

    whether to overwrite or not

    Definition Classes
    AbstractModule
  103. var scaleB: Double

    Attributes
    protected
    Definition Classes
    AbstractModule
  104. var scaleW: Double

    The scale of gradient weight and gradient bias before gradParameters being accumulated.

    The scale of gradient weight and gradient bias before gradParameters being accumulated.

    Attributes
    protected
    Definition Classes
    AbstractModule
  105. final def setExtraParameter(extraParam: Array[Tensor[T]]): Graph.this.type

    Set extra parameter to this module.

    Set extra parameter to this module. Extra parameter means the trainable parameters beside weight and bias. Such as runningMean and runningVar in BatchNormalization.

    returns

    this

    Definition Classes
    AbstractModule
  106. def setInputDimMask(mask: Int): Unit

    Set dimension mask of input

    Set dimension mask of input

    mask

    value of input dimension mask to be set

    returns

    Unit

    Definition Classes
    MklInt8Convertible
  107. def setInputScales(inScales: Array[Array[Float]]): Unit

    Set input scales Clear existing buffer of input scales, and place updated scales into the cleared buffer

    Set input scales Clear existing buffer of input scales, and place updated scales into the cleared buffer

    inScales

    value of input scales to be set

    returns

    Unit

    Definition Classes
    MklInt8Convertible
  108. final def setLine(line: String): Graph.this.type

    Set the line separator when print the module

    Set the line separator when print the module

    line
    returns

    Definition Classes
    AbstractModule
  109. final def setName(name: String): Graph.this.type

    Set the module name

    Set the module name

    name
    returns

    Definition Classes
    AbstractModule
  110. def setOutputDimMask(mask: Int): Unit

    Set dimension mask of output

    Set dimension mask of output

    mask

    value of output dimension mask to be set

    returns

    Unit

    Definition Classes
    MklInt8Convertible
  111. def setOutputScales(outScales: Array[Array[Float]]): Unit

    Set output scales Clear existing buffer of output scales, and place updated scales into the cleared buffer

    Set output scales Clear existing buffer of output scales, and place updated scales into the cleared buffer

    outScales

    value of output scales to be set

    returns

    Unit

    Definition Classes
    MklInt8Convertible
  112. def setScaleB(b: Double): Graph.this.type

    Set the scale of gradientBias

    Set the scale of gradientBias

    b

    the value of the scale of gradientBias

    returns

    this

    Definition Classes
    ContainerAbstractModule
  113. def setScaleW(w: Double): Graph.this.type

    Set the scale of gradientWeight

    Set the scale of gradientWeight

    w

    the value of the scale of gradientWeight

    returns

    this

    Definition Classes
    ContainerAbstractModule
  114. def setWeightDimMask(mask: Int): Unit

    Set dimension mask of weight

    Set dimension mask of weight

    mask

    value of weight mask to be set

    returns

    Unit

    Definition Classes
    MklInt8Convertible
  115. def setWeightScales(weightScales: Array[Array[Float]]): Unit

    Set weight scales Clear existing buffer of weight scales, and place updated scales into the cleared buffer

    Set weight scales Clear existing buffer of weight scales, and place updated scales into the cleared buffer

    weightScales

    value of weight scales to be set

    returns

    Unit

    Definition Classes
    MklInt8Convertible
  116. final def setWeightsBias(newWeights: Array[Tensor[T]]): Graph.this.type

    Set weight and bias for the module

    Set weight and bias for the module

    newWeights

    array of weights and bias

    returns

    Definition Classes
    AbstractModule
  117. def stopGradient(names: Array[String]): Graph.this.type

    stop the input gradient of layers that match the given names their input gradient are not computed.

    stop the input gradient of layers that match the given names their input gradient are not computed. And they will not contributed to the input gradient computation of layers that depend on them.

    names

    an array of layer names

    returns

    current graph model

  118. final def synchronized[T0](arg0: ⇒ T0): T0

    Definition Classes
    AnyRef
  119. def toGraph(startNodes: ModuleNode[T]*): Graph[T]

    Generate graph module with start nodes

    Generate graph module with start nodes

    startNodes
    returns

    Definition Classes
    GraphAbstractModule
  120. def toString(): String

    Definition Classes
    AbstractModule → AnyRef → Any
  121. var train: Boolean

    Module status.

    Module status. It is useful for modules like dropout/batch normalization

    Attributes
    protected
    Definition Classes
    AbstractModule
  122. final def training(): Graph.this.type

    Set the module to training mode

    Set the module to training mode

    returns

    Definition Classes
    ContainerAbstractModule
  123. def unFreeze(names: String*): Graph.this.type

    "unfreeze" module, i.

    "unfreeze" module, i.e. make the module parameters(weight/bias, if exists) to be trained(updated) in training process if names is not empty, unfreeze layers that match given names

    names

    array of module names to unFreeze

    Definition Classes
    ContainerAbstractModule
  124. def updateInputScales(scale: Array[Float], index: Int): Unit

    Update input scales at specific index with provided new scale

    Update input scales at specific index with provided new scale

    scale

    the new scale

    index

    the index of which the scale need to be updated

    returns

    Unit

    Definition Classes
    MklInt8Convertible
  125. def updateOutputScales(scale: Array[Float], index: Int): Unit

    Update output scales at specific index with provided new scale

    Update output scales at specific index with provided new scale

    scale

    the new scale

    index

    the index of which the scale need to be updated

    returns

    Unit

    Definition Classes
    MklInt8Convertible
  126. def updateWeightScales(scale: Array[Float], index: Int): Unit

    Update weight scales at specific index with provided new scale

    Update weight scales at specific index with provided new scale

    scale

    the new scale

    index

    the index of which the scale need to be updated

    returns

    Unit

    Definition Classes
    MklInt8Convertible
  127. final def wait(): Unit

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

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

    Definition Classes
    AnyRef
    Annotations
    @throws( ... )
  130. var weightDimMask: Int

    Attributes
    protected
    Definition Classes
    MklInt8Convertible
  131. def zeroGradParameters(): Unit

    If the module has parameters, this will zero the accumulation of the gradients with respect to these parameters.

    If the module has parameters, this will zero the accumulation of the gradients with respect to these parameters. Otherwise, it does nothing.

    Definition Classes
    AbstractModule

Deprecated Value Members

  1. final def save(path: String, overWrite: Boolean = false): Graph.this.type

    Save this module to path.

    Save this module to path.

    path

    path to save module, local file system, HDFS and Amazon S3 is supported. HDFS path should be like "hdfs://[host]:[port]/xxx" Amazon S3 path should be like "s3a://bucket/xxx"

    overWrite

    if overwrite

    returns

    self

    Definition Classes
    AbstractModule
    Annotations
    @deprecated
    Deprecated

    (Since version 0.3.0) please use recommended saveModule(path, overWrite)

Inherited from MklInt8Convertible

Inherited from Container[Activity, Activity, T]

Inherited from AbstractModule[Activity, Activity, T]

Inherited from InferShape

Inherited from Serializable

Inherited from Serializable

Inherited from AnyRef

Inherited from Any

Ungrouped