VwSparseMultilabelPredictor

Creates a VW multi-label predictor plugin for MultilabelModel.

K: the label or class type.
modelSource: a specification for the underlying Cost Sensitive One Against All VW model with label dependent features. VW flag --csoaa_ldf mc or --wap_ldf mc is expected. For more information, see the VW CSOAA wiki page. Also see the Cost-Sensitive Multiclass Classification section of Hal Daume's On Multiclass Classification in VW page. This model specification will be materialized in this class.
defaultNs: The list of indices into the features sequence that does not have an exist in any value of the namespaces map.
namespaces: Mapping from namespace name to indices in the features sequence passed to the apply method. There should be no empty namespaces, meaning key-value pairs appearing in the map should not values that are empty sequences. This is a requirement.

Since: 9/8/2017

Linear Supertypes

Serializable, Serializable, Product, Equals, Closeable, AutoCloseable, (SparseFeatures, Labels[K], LabelIndices, SparseLabelDepFeatures) ⇒ Try[Map[K, Double]], AnyRef, Any

Instance Constructors

new VwSparseMultilabelPredictor(modelSource: ModelSource, defaultNs: List[Int], namespaces: List[(String, List[Int])], numLabelsInTrainingSet: Int)

modelSource
a specification for the underlying Cost Sensitive One Against All VW model with label dependent features. VW flag --csoaa_ldf mc or --wap_ldf mc is expected. For more information, see the VW CSOAA wiki page. Also see the Cost-Sensitive Multiclass Classification section of Hal Daume's On Multiclass Classification in VW page. This model specification will be materialized in this class.
defaultNs
The list of indices into the features sequence that does not have an exist in any value of the namespaces map.
namespaces
Mapping from namespace name to indices in the features sequence passed to the apply method. There should be no empty namespaces, meaning key-value pairs appearing in the map should not values that are empty sequences. This is a requirement.

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 apply(features: IndexedSeq[Sparse], labels: IndexedSeq[K], indices: IndexedSeq[Int], labelDependentFeatures: IndexedSeq[IndexedSeq[Sparse]]): Try[Map[K, Double]]

Given the input, form a VW example, and delegate to the underlying CSOAA LDF VW model.
Given the input, form a VW example, and delegate to the underlying CSOAA LDF VW model.
features
(non-label dependent) features shared across all labels.
labels
labels for which the VW learner should produce predictions.
indices
the indices labels into the sequence of all labels encountered during training.
labelDependentFeatures
Any label dependent features. This is not yet utilized and is currently ignored.
returns
a Map from label to prediction.

Definition Classes
VwSparseMultilabelPredictor → Function4
final def asInstanceOf[T0]: T0

Definition Classes
Any
def clone(): AnyRef

Attributes
protected[java.lang]
Definition Classes
AnyRef
Annotations
@throws( ... )
def close(): Unit

Definition Classes
VwSparseMultilabelPredictor → Closeable → AutoCloseable
def curried: (SparseFeatures) ⇒ (Labels[K]) ⇒ (LabelIndices) ⇒ (SparseLabelDepFeatures) ⇒ Try[Map[K, Double]]

Definition Classes
Function4
Annotations
@unspecialized()
val defaultNs: List[Int]

The list of indices into the features sequence that does not have an exist in any value of the namespaces map.
final def eq(arg0: AnyRef): Boolean

Definition Classes
AnyRef
def finalize(): Unit

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

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

Definition Classes
Any
val modelSource: ModelSource

a specification for the underlying Cost Sensitive One Against All VW model with label dependent features.
a specification for the underlying Cost Sensitive One Against All VW model with label dependent features. VW flag --csoaa_ldf mc or --wap_ldf mc is expected. For more information, see the VW CSOAA wiki page. Also see the Cost-Sensitive Multiclass Classification section of Hal Daume's On Multiclass Classification in VW page. This model specification will be materialized in this class.
val namespaces: List[(String, List[Int])]

Mapping from namespace name to indices in the features sequence passed to the apply method.
Mapping from namespace name to indices in the features sequence passed to the apply method. There should be no empty namespaces, meaning key-value pairs appearing in the map should not values that are empty sequences. This is a requirement.
final def ne(arg0: AnyRef): Boolean

Definition Classes
AnyRef
final def notify(): Unit

Definition Classes
AnyRef
final def notifyAll(): Unit

Definition Classes
AnyRef
val numLabelsInTrainingSet: Int
final def synchronized[T0](arg0: ⇒ T0): T0

Definition Classes
AnyRef
def toString(): String

Definition Classes
Function4 → AnyRef → Any
def tupled: ((SparseFeatures, Labels[K], LabelIndices, SparseLabelDepFeatures)) ⇒ Try[Map[K, Double]]

Definition Classes
Function4
Annotations
@unspecialized()
def vwParams(): String

Get the VW parameters used to invoke the underlying VW model.
Get the VW parameters used to invoke the underlying VW model.
returns
VW parameters.
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( ... )

Related Docs: object VwSparseMultilabelPredictor | package multilabel

case class VwSparseMultilabelPredictor[K](modelSource: ModelSource, defaultNs: List[Int], namespaces: List[(String, List[Int])], numLabelsInTrainingSet: Int) extends SparseMultiLabelPredictor[K] with Closeable with Product with Serializable

Instance Constructors

new VwSparseMultilabelPredictor(modelSource: ModelSource, defaultNs: List[Int], namespaces: List[(String, List[Int])], numLabelsInTrainingSet: Int)

Value Members

final def !=(arg0: Any): Boolean

final def ##(): Int

final def ==(arg0: Any): Boolean

def apply(features: IndexedSeq[Sparse], labels: IndexedSeq[K], indices: IndexedSeq[Int], labelDependentFeatures: IndexedSeq[IndexedSeq[Sparse]]): Try[Map[K, Double]]

final def asInstanceOf[T0]: T0

def clone(): AnyRef

def close(): Unit

def curried: (SparseFeatures) ⇒ (Labels[K]) ⇒ (LabelIndices) ⇒ (SparseLabelDepFeatures) ⇒ Try[Map[K, Double]]

val defaultNs: List[Int]

final def eq(arg0: AnyRef): Boolean

def finalize(): Unit

final def getClass(): Class[_]

final def isInstanceOf[T0]: Boolean

val modelSource: ModelSource

val namespaces: List[(String, List[Int])]

final def ne(arg0: AnyRef): Boolean

final def notify(): Unit

final def notifyAll(): Unit

val numLabelsInTrainingSet: Int

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

def toString(): String

def tupled: ((SparseFeatures, Labels[K], LabelIndices, SparseLabelDepFeatures)) ⇒ Try[Map[K, Double]]

def vwParams(): String

final def wait(): Unit

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

final def wait(arg0: Long): Unit

Inherited from Serializable

Inherited from Serializable

Inherited from Product

Inherited from Equals

Inherited from Closeable

Inherited from AutoCloseable

Inherited from (SparseFeatures, Labels[K], LabelIndices, SparseLabelDepFeatures) ⇒ Try[Map[K, Double]]

Inherited from AnyRef

Inherited from Any

Ungrouped