oracle.pgx.api.mllib

Class SupervisedGraphWiseModelBuilder

java.lang.Object

oracle.pgx.api.mllib.GraphWiseModelBuilder<SupervisedGraphWiseModel,oracle.pgx.config.mllib.SupervisedGraphWiseModelConfig,SupervisedGraphWiseModelBuilder>

oracle.pgx.api.mllib.SupervisedGraphWiseModelBuilder

public class SupervisedGraphWiseModelBuilder
extends GraphWiseModelBuilder<SupervisedGraphWiseModel,oracle.pgx.config.mllib.SupervisedGraphWiseModelConfig,SupervisedGraphWiseModelBuilder>

Builder for SupervisedGraphWiseModel The builder can be used to set the configuration of the model and create the model object.

Since:

19.4

Constructor Summary

Constructors

Constructor and Description
SupervisedGraphWiseModelBuilder(PgxSession session, oracle.pgx.api.internal.Core core, java.util.function.Supplier<java.lang.String> keystorePathSupplier, java.util.function.Supplier<char[]> keystorePasswordSupplier)

Method Summary

Modifier and Type	Method and Description
SupervisedGraphWiseModel	build() Builds the SupervisedGraphWise model with the specified parameters.
SupervisedGraphWiseModelBuilder	setClassWeights(java.util.Map<?,java.lang.Float> classWeights) Set the class weights to be used in the loss function.
SupervisedGraphWiseModelBuilder	setLossFunction(oracle.pgx.config.mllib.SupervisedGraphWiseModelConfig.LossFunction lossFunction) Sets the loss function for the algorithm.
SupervisedGraphWiseModelBuilder	setNormalize(boolean normalize) Whether to l2 normalize the output of the convolutional layers.
SupervisedGraphWiseModelBuilder	setPredictionLayerConfigs(oracle.pgx.config.mllib.GraphWisePredictionLayerConfig... layerConfigs) Set the prediction layer configurations (See GraphWisePredictionLayerConfig.
SupervisedGraphWiseModelBuilder	setVertexTargetPropertyName(java.lang.String propertyName) Set the target (labels) for the algorithm in the form of a property name of the graph.

Methods inherited from class oracle.pgx.api.mllib.GraphWiseModelBuilder

setBatchSize, setConvLayerConfigs, setEmbeddingDim, setLearningRate, setNumEpochs, setSeed, setShuffle, setStandardize, setVertexInputPropertyNames, setVertexInputPropertyNames

Methods inherited from class java.lang.Object

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

Constructor Detail

SupervisedGraphWiseModelBuilder

public SupervisedGraphWiseModelBuilder(PgxSession session,
                                       oracle.pgx.api.internal.Core core,
                                       java.util.function.Supplier<java.lang.String> keystorePathSupplier,
                                       java.util.function.Supplier<char[]> keystorePasswordSupplier)

Method Detail

build

public SupervisedGraphWiseModel build()
                               throws java.lang.InterruptedException,
                                      java.util.concurrent.ExecutionException

Builds the SupervisedGraphWise model with the specified parameters.

Specified by:

build in class GraphWiseModelBuilder<SupervisedGraphWiseModel,oracle.pgx.config.mllib.SupervisedGraphWiseModelConfig,SupervisedGraphWiseModelBuilder>

Returns:

SupervisedGraphWise model

Throws:

java.lang.InterruptedException

java.util.concurrent.ExecutionException

Since:

19.4

setClassWeights

public SupervisedGraphWiseModelBuilder setClassWeights(java.util.Map<?,java.lang.Float> classWeights)

Set the class weights to be used in the loss function. The loss for the corresponding class will be multiplied by the factor given in this map. If null, uniform class weights will be used.

default: SupervisedGraphWiseModelConfig.DEFAULT_CLASS_MAP

Parameters:

classWeights - map from classes to weights

Returns:

this

Since:

19.4

setLossFunction

public SupervisedGraphWiseModelBuilder setLossFunction(oracle.pgx.config.mllib.SupervisedGraphWiseModelConfig.LossFunction lossFunction)

Sets the loss function for the algorithm. See SupervisedGraphWiseModelConfig.LossFunction

default: SupervisedGraphWiseModelConfig.LossFunction#SOFTMAX_CROSS_ENTROPY

Parameters:

lossFunction - loss function

Returns:

this

Since:

19.4

setNormalize

public SupervisedGraphWiseModelBuilder setNormalize(boolean normalize)

Whether to l2 normalize the output of the convolutional layers.

default: SupervisedGraphWiseModelConfig.DEFAULT_NORMALIZE

Parameters:

normalize - whether to normalize

Returns:

this

Since:

19.4

setPredictionLayerConfigs

public SupervisedGraphWiseModelBuilder setPredictionLayerConfigs(oracle.pgx.config.mllib.GraphWisePredictionLayerConfig... layerConfigs)

Set the prediction layer configurations (See GraphWisePredictionLayerConfig. You must pass at least one prediction layer config (or leave it as default).

Note that for the last layer:

• the activation function will be ignored (and replaced with the activation function of the loss function, e.g. softmax or sigmoid).
• the hidden dimension will be ignored (and replaced with the number of classes)

default: SupervisedGraphWiseModelConfig.DEFAULT_PREDICTION_LAYER_CONFIGS

Parameters:

layerConfigs - layer configs

Returns:

this

Since:

19.4

setVertexTargetPropertyName

public SupervisedGraphWiseModelBuilder setVertexTargetPropertyName(java.lang.String propertyName)

Set the target (labels) for the algorithm in the form of a property name of the graph.

Supported property types are given by SupervisedGraphWiseModelConfig.SUPPORTED_LABEL_TYPES.

Parameters:

propertyName - property name

Returns:

this

Since:

19.4