org.deeplearning4j.models.sequencevectors.graph.walkers.impl

Class PopularityWalker.Builder<T extends SequenceElement>

java.lang.Object

org.deeplearning4j.models.sequencevectors.graph.walkers.impl.RandomWalker.Builder<T>

org.deeplearning4j.models.sequencevectors.graph.walkers.impl.PopularityWalker.Builder<T>

Enclosing class:

PopularityWalker<T extends SequenceElement>

public static class PopularityWalker.Builder<T extends SequenceElement>
extends RandomWalker.Builder<T>

Field Summary

Fields

Modifier and Type	Field and Description
protected PopularityMode	popularityMode
protected SpreadSpectrum	spectrum
protected int	spread

Fields inherited from class org.deeplearning4j.models.sequencevectors.graph.walkers.impl.RandomWalker.Builder

alpha, noEdgeHandling, seed, sourceGraph, walkDirection, walkLength

Constructor Summary

Constructors

Constructor and Description
Builder(IGraph<T,?> sourceGraph)

Method Summary

Modifier and Type	Method and Description
PopularityWalker<T>	build() This method builds PopularityWalker object with previously specified params
PopularityWalker.Builder<T>	setNoEdgeHandling(NoEdgeHandling handling) This method defines walker behavior when it gets to node which has no next nodes available Default value: RESTART_ON_DISCONNECTED
PopularityWalker.Builder<T>	setPopularityMode(PopularityMode popularityMode) This method defines which nodes should be taken in account when choosing next hope: maximum popularity, lowest popularity, or average popularity.
PopularityWalker.Builder<T>	setPopularitySpread(int topN) This method defines, how much nodes should take place in next hop selection.
PopularityWalker.Builder<T>	setRestartProbability(double alpha) This method defines a chance for walk restart Good value would be somewhere between 0.03-0.07
PopularityWalker.Builder<T>	setSeed(long seed) This method specifies random seed.
PopularityWalker.Builder<T>	setSpreadSpectrum(SpreadSpectrum spectrum) This method allows you to define, if nodes within popularity spread should have equal chances to be picked for next hop, or they should have chances proportional to their popularity.
PopularityWalker.Builder<T>	setWalkDirection(WalkDirection direction) This method defines next hop selection within walk
PopularityWalker.Builder<T>	setWalkLength(int walkLength) This method specifies output sequence (walk) length

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Field Detail

popularityMode

protected PopularityMode popularityMode

spread

protected int spread

spectrum

protected SpreadSpectrum spectrum

Constructor Detail

Builder

public Builder(IGraph<T,?> sourceGraph)

Method Detail

setPopularityMode

public PopularityWalker.Builder<T> setPopularityMode(@NonNull
                                                     PopularityMode popularityMode)

This method defines which nodes should be taken in account when choosing next hope: maximum popularity, lowest popularity, or average popularity. Default value: MAXIMUM

Parameters:

popularityMode -

Returns:

setPopularitySpread

public PopularityWalker.Builder<T> setPopularitySpread(int topN)

This method defines, how much nodes should take place in next hop selection. Something like top-N nodes, or bottom-N nodes. Default value: 10

Parameters:

topN -

Returns:

setSpreadSpectrum

public PopularityWalker.Builder<T> setSpreadSpectrum(@NonNull
                                                     SpreadSpectrum spectrum)

This method allows you to define, if nodes within popularity spread should have equal chances to be picked for next hop, or they should have chances proportional to their popularity. Default value: PLAIN

Parameters:

spectrum -

Returns:

setNoEdgeHandling

public PopularityWalker.Builder<T> setNoEdgeHandling(@NonNull
                                                     NoEdgeHandling handling)

This method defines walker behavior when it gets to node which has no next nodes available Default value: RESTART_ON_DISCONNECTED

Overrides:

setNoEdgeHandling in class RandomWalker.Builder<T extends SequenceElement>

Parameters:

handling -

Returns:

setSeed

public PopularityWalker.Builder<T> setSeed(long seed)

This method specifies random seed.

Overrides:

setSeed in class RandomWalker.Builder<T extends SequenceElement>

Parameters:

seed -

Returns:

setWalkDirection

public PopularityWalker.Builder<T> setWalkDirection(@NonNull
                                                    WalkDirection direction)

This method defines next hop selection within walk

Overrides:

setWalkDirection in class RandomWalker.Builder<T extends SequenceElement>

Parameters:

direction -

Returns:

setWalkLength

public PopularityWalker.Builder<T> setWalkLength(int walkLength)

This method specifies output sequence (walk) length

Overrides:

setWalkLength in class RandomWalker.Builder<T extends SequenceElement>

Parameters:

walkLength -

Returns:

setRestartProbability

public PopularityWalker.Builder<T> setRestartProbability(double alpha)

This method defines a chance for walk restart Good value would be somewhere between 0.03-0.07

Overrides:

setRestartProbability in class RandomWalker.Builder<T extends SequenceElement>

Parameters:

alpha -

Returns:

build

public PopularityWalker<T> build()

This method builds PopularityWalker object with previously specified params

Overrides:

build in class RandomWalker.Builder<T extends SequenceElement>

Returns: