org.deeplearning4j.graph.models.deepwalk

Class DeepWalk.Builder<V,E>

java.lang.Object

org.deeplearning4j.graph.models.deepwalk.DeepWalk.Builder<V,E>

Enclosing class:

DeepWalk<V,E>

public static class DeepWalk.Builder<V,E>
extends java.lang.Object

Constructor Summary

Constructors

Constructor and Description
Builder()

Method Summary

Modifier and Type	Method and Description
DeepWalk<V,E>	build()
DeepWalk.Builder<V,E>	learningRate(double learningRate) Set the learning rate
DeepWalk.Builder<V,E>	seed(long seed) Seed for random number generation (used for repeatability).
DeepWalk.Builder<V,E>	vectorSize(int vectorSize) Sets the size of the vectors to be learned for each vertex in the graph
DeepWalk.Builder<V,E>	windowSize(int windowSize) Sets the window size used in skipgram model

Methods inherited from class java.lang.Object

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

Constructor Detail

Builder

public Builder()

Method Detail

vectorSize

public DeepWalk.Builder<V,E> vectorSize(int vectorSize)

Sets the size of the vectors to be learned for each vertex in the graph

learningRate

public DeepWalk.Builder<V,E> learningRate(double learningRate)

Set the learning rate

windowSize

public DeepWalk.Builder<V,E> windowSize(int windowSize)

Sets the window size used in skipgram model

seed

public DeepWalk.Builder<V,E> seed(long seed)

Seed for random number generation (used for repeatability). Note however that parallel/async gradient descent might result in behaviour that is not repeatable, in spite of setting seed

build

public DeepWalk<V,E> build()