Network

Abstract Value Members

abstract def backward(err: ParSeq[DataVec]): ArrayBuffer[() ⇒ Unit]

Execute error backpropagation
Execute error backpropagation
err
Sequence of error to be backprop.
abstract def forward(in: In): DataVec

Apply this network.
Apply this network.
in
input
returns
Output vector.
abstract def forward(in: ParSeq[In]): ParSeq[DataVec]

Apply this network.
Apply this network.
in
Parallel Sequence of input
returns
Sequence of Output vector.
abstract def forward(in: RDD[(Long, In)]): RDD[(Long, DataVec)]

Apply this network.
Apply this network.
in
RDD of (ID, Value).
returns
RDD of (ID, Output vector).

Concrete Value Members

final def !=(arg0: Any): Boolean

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

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

Definition Classes
AnyRef → Any
def NOut: Int

Size of output
Size of output
returns
Output size
final def add(layer: Layer[DataVec, DataVec]): Network.this.type

Add new top layer.
Add new top layer.
layer
Layer to be added on top of current stack.
returns
self
final def asInstanceOf[T0]: T0

Definition Classes
Any
final def backwardSeq(err: ParSeq[DataVec]): (ParSeq[DataVec], ArrayBuffer[() ⇒ Unit])

Backpropagate errors
Backpropagate errors
err
Top-level error to be propagated backward.
returns
lowest-level error sequence.

Attributes
protected
def broadcast(sc: SparkContext): Unit

Broadcast resources of input layer.
Broadcast resources of input layer.
sc
Spark Context
var builder: WeightBuilder

Weight builder for this network.
Weight builder for this network. *

Attributes
protected
def clone(): AnyRef

Attributes
protected[java.lang]
Definition Classes
AnyRef
Annotations
@throws( ... )
final def eq(arg0: AnyRef): Boolean

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

Definition Classes
AnyRef → Any
def finalize(): Unit

Attributes
protected[java.lang]
Definition Classes
AnyRef
Annotations
@throws( classOf[java.lang.Throwable] )
final def forwardRDD(in: RDD[(Long, DataVec)], layers: Seq[Layer[DataVec, DataVec]] = layerSeq): RDD[(Long, DataVec)]

Apply by layers.
Apply by layers.
in
RDD of (ID, Vector)
layers
Sequence of layers to be applied. Default value is layer sequence.
returns
RDD of final output (ID, Output Vector)

Attributes
protected
Annotations
@tailrec()
final def forwardSeq(in: ParSeq[DataVec]): ParSeq[DataVec]

Apply by layers.
Apply by layers.
in
Parallel sequence of Vector
returns
Parallel sequence of final output

Attributes
protected
final def forwardSingle(in: DataVec, layers: Seq[Layer[DataVec, DataVec]] = layerSeq): DataVec

Apply by layers.
Apply by layers.
in
Vector
layers
Sequence of layers to be applied. Default value is layer sequence.
returns
Final output Output Vector

Attributes
protected
Annotations
@tailrec()
final def getClass(): Class[_]

Definition Classes
AnyRef → Any
def hashCode(): Int

Definition Classes
AnyRef → Any
def initiateBy(builder: WeightBuilder): Network.this.type

Initiate this network by given builder
Initiate this network by given builder
builder
Weight builder
returns
self
final def isInstanceOf[T0]: Boolean

Definition Classes
Any
val layerSeq: ListBuffer[Layer[DataVec, DataVec]]

Seqeunce of hidden/output layer to be applied.
Seqeunce of hidden/output layer to be applied. *

Attributes
protected
def layers: Seq[Layer[DataVec, DataVec]]

Sequence of hidden/output layers
Sequence of hidden/output layers
returns
Sequence of hidden/output layer
def loss: Double

Weight loss of layers
Weight loss of layers
returns
Weight loss
final def ne(arg0: AnyRef): Boolean

Definition Classes
AnyRef
final def notify(): Unit

Definition Classes
AnyRef
final def notifyAll(): Unit

Definition Classes
AnyRef
def predictSoft(in: In): DataVec

Apply this network.
Apply this network.
in
input
returns
Output vector.
def predictSoft(in: RDD[In])(implicit arg0: ClassTag[DataVec]): RDD[DataVec]

Apply this network.
Apply this network.
in
RDD of Value.
returns
RDD of Output vector.
def read(kryo: Kryo, input: Input): Unit

Definition Classes
Network → KryoSerializable
def saveTo(file: String): Unit

Save this network.
Save this network.
file
Save path of this network.
def setUpdatable(bool: Boolean): Network[In, Out]

Set this network can be used in backpropagation.
Set this network can be used in backpropagation.
bool
True if this should prepare backpropagation
returns
self
final def synchronized[T0](arg0: ⇒ T0): T0

Definition Classes
AnyRef
def toString(): String

Definition Classes
AnyRef → Any
def unbroadcast(): Unit

Unpersist resources of input layer.
final def wait(): Unit

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

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

Definition Classes
AnyRef
Annotations
@throws( ... )
def write(kryo: Kryo, output: Output): Unit

Definition Classes
Network → KryoSerializable

Related Docs: object Network | package network

trait Network[In, Out] extends Serializable with KryoSerializable

Abstract Value Members

abstract def backward(err: ParSeq[DataVec]): ArrayBuffer[() ⇒ Unit]

abstract def forward(in: In): DataVec

abstract def forward(in: ParSeq[In]): ParSeq[DataVec]

abstract def forward(in: RDD[(Long, In)]): RDD[(Long, DataVec)]

Concrete Value Members

final def !=(arg0: Any): Boolean

final def ##(): Int

final def ==(arg0: Any): Boolean

def NOut: Int

final def add(layer: Layer[DataVec, DataVec]): Network.this.type

final def asInstanceOf[T0]: T0

final def backwardSeq(err: ParSeq[DataVec]): (ParSeq[DataVec], ArrayBuffer[() ⇒ Unit])

def broadcast(sc: SparkContext): Unit

var builder: WeightBuilder

def clone(): AnyRef

final def eq(arg0: AnyRef): Boolean

def equals(arg0: Any): Boolean

def finalize(): Unit

final def forwardRDD(in: RDD[(Long, DataVec)], layers: Seq[Layer[DataVec, DataVec]] = layerSeq): RDD[(Long, DataVec)]

final def forwardSeq(in: ParSeq[DataVec]): ParSeq[DataVec]

final def forwardSingle(in: DataVec, layers: Seq[Layer[DataVec, DataVec]] = layerSeq): DataVec

final def getClass(): Class[_]

def hashCode(): Int

def initiateBy(builder: WeightBuilder): Network.this.type

final def isInstanceOf[T0]: Boolean

val layerSeq: ListBuffer[Layer[DataVec, DataVec]]

def layers: Seq[Layer[DataVec, DataVec]]

def loss: Double

final def ne(arg0: AnyRef): Boolean

final def notify(): Unit

final def notifyAll(): Unit

def predictSoft(in: In): DataVec

def predictSoft(in: RDD[In])(implicit arg0: ClassTag[DataVec]): RDD[DataVec]

def read(kryo: Kryo, input: Input): Unit

def saveTo(file: String): Unit

def setUpdatable(bool: Boolean): Network[In, Out]

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

def toString(): String

def unbroadcast(): Unit

final def wait(): Unit

final def wait(arg0: Long, arg1: Int): Unit

final def wait(arg0: Long): Unit

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

Inherited from KryoSerializable

Inherited from Serializable

Inherited from Serializable

Inherited from AnyRef

Inherited from Any

Ungrouped