HoeffdingTree

The Hoeffding tree is an incremental decision tree learner for large data streams, that assumes that the data distribution is not changing over time. It grows incrementally a decision tree based on the theoretical guarantees of the Hoeffding bound (or additive Chernoff bound). A node is expanded as soon as there is sufficient statistical evidence that an optimal splitting feature exists, a decision based on the distribution-independent Hoeffding bound. The model learned by the Hoeffding tree is asymptotically nearly identical to the one built by a non-incremental learner, if the number of training instances is large enough.

It is controlled by the following options:

numeric observer to use (-n); for the moment, only Gaussian approximation is supported; class of type FeatureClassObserver;
number of examples a leaf should observe before a split attempt (-g);
Number of examples a leaf should observe before applying NaiveBayes (-q);
split criterion to use (-s), an object of type SplitCriterionType;
allowable error in split decision (-c);
threshold of allowable error in breaking ties (-t);
allow only binary splits (-b), boolean flag;
disable poor attributes (-r);
allow growth (-o);
disable pre-pruning (-p);
leaf prediction to use (-l), either MajorityClass (0), NaiveBayes (1) or adaptive NaiveBayes (2, default);
enable splitting at all leaves (-a); the original algorithm can split at any time, but in Spark Streaming one can only split once per RDD; the option controls whether to split at leaf of the last Example of the RDD, or at every leaf.

Linear Supertypes

Classifier, Learner, Serializable, Configurable, Serializable, AnyRef, Any

Instance Constructors

new HoeffdingTree()

Type Members

type T = HoeffdingTreeModel

Definition Classes
HoeffdingTree → Learner

Value Members

final def !=(arg0: AnyRef): Boolean

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

Definition Classes
Any
final def ##(): Int

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

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

Definition Classes
Any
final def asInstanceOf[T0]: T0

Definition Classes
Any
val binaryOnlyOption: FlagOption
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
var espec: ExampleSpecification
def finalize(): Unit

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

Definition Classes
AnyRef → Any
def getModel: HoeffdingTreeModel

Gets the current Model used for the Learner.
Gets the current Model used for the Learner.
returns
the Model object used for training

Definition Classes
HoeffdingTree → Learner
val growthAllowedOption: FlagOption
def hashCode(): Int

Definition Classes
AnyRef → Any
def init(exampleSpecification: ExampleSpecification): Unit

Init the model based on the algorithm implemented in the learner.
Init the model based on the algorithm implemented in the learner.
exampleSpecification
the ExampleSpecification of the input stream.

Definition Classes
HoeffdingTree → Learner
final def isInstanceOf[T0]: Boolean

Definition Classes
Any
val learningNodeOption: IntOption
var model: HoeffdingTreeModel
val nbThresholdOption: IntOption
final def ne(arg0: AnyRef): Boolean

Definition Classes
AnyRef
val noPrePruneOption: FlagOption
final def notify(): Unit

Definition Classes
AnyRef
final def notifyAll(): Unit

Definition Classes
AnyRef
val numGraceOption: IntOption
val numericObserverTypeOption: IntOption
def predict(input: DStream[Example]): DStream[(Example, Double)]

Definition Classes
HoeffdingTree → Classifier
val removePoorFeaturesOption: FlagOption
val splitAllOption: FlagOption
val splitConfidenceOption: FloatOption
val splitCriterionOption: ClassOption
final def synchronized[T0](arg0: ⇒ T0): T0

Definition Classes
AnyRef
val tieThresholdOption: FloatOption
def toString(): String

Definition Classes
AnyRef → Any
def train(input: DStream[Example]): Unit

Train the model based on the algorithm implemented in the learner, from the stream of Examples given for training.
Train the model based on the algorithm implemented in the learner, from the stream of Examples given for training.
input
a stream of Examples

Definition Classes
HoeffdingTree → Learner
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( ... )

class HoeffdingTree extends Classifier

Instance Constructors

new HoeffdingTree()

Type Members

type T = HoeffdingTreeModel

Value Members

final def !=(arg0: AnyRef): Boolean

final def !=(arg0: Any): Boolean

final def ##(): Int

final def ==(arg0: AnyRef): Boolean

final def ==(arg0: Any): Boolean

final def asInstanceOf[T0]: T0

val binaryOnlyOption: FlagOption

def clone(): AnyRef

final def eq(arg0: AnyRef): Boolean

def equals(arg0: Any): Boolean

var espec: ExampleSpecification

def finalize(): Unit

final def getClass(): Class[_]

def getModel: HoeffdingTreeModel

val growthAllowedOption: FlagOption

def hashCode(): Int

def init(exampleSpecification: ExampleSpecification): Unit

final def isInstanceOf[T0]: Boolean

val learningNodeOption: IntOption

var model: HoeffdingTreeModel

val nbThresholdOption: IntOption

final def ne(arg0: AnyRef): Boolean

val noPrePruneOption: FlagOption

final def notify(): Unit

final def notifyAll(): Unit

val numGraceOption: IntOption

val numericObserverTypeOption: IntOption

def predict(input: DStream[Example]): DStream[(Example, Double)]

val removePoorFeaturesOption: FlagOption

val splitAllOption: FlagOption

val splitConfidenceOption: FloatOption

val splitCriterionOption: ClassOption

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

val tieThresholdOption: FloatOption

def toString(): String

def train(input: DStream[Example]): Unit

final def wait(): Unit

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

final def wait(arg0: Long): Unit

Inherited from Classifier

Inherited from Learner

Inherited from Serializable

Inherited from Configurable

Inherited from Serializable

Inherited from AnyRef

Inherited from Any

Ungrouped