oracle.pgx.api.mllib

Class SupervisedGraphWiseModel

java.lang.Object

oracle.pgx.api.mllib.Model<ModelType>

oracle.pgx.api.mllib.GraphWiseModel<SupervisedGraphWiseModelConfig,oracle.pgx.api.internal.mllib.SupervisedGraphWiseModelMetadata,SupervisedGraphWiseModel>

oracle.pgx.api.mllib.SupervisedGraphWiseModel

All Implemented Interfaces:

java.lang.AutoCloseable

public class SupervisedGraphWiseModel
extends GraphWiseModel<SupervisedGraphWiseModelConfig,oracle.pgx.api.internal.mllib.SupervisedGraphWiseModelMetadata,SupervisedGraphWiseModel>

A variant of GraphSAGE. See SupervisedGraphWiseModelBuilder for documentation of the hyperparameters.

Since:

19.4

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static class	SupervisedGraphWiseModel.SupervisedGraphWiseInferenceType

Field Summary

Fields

Modifier and Type	Field and Description
static java.lang.String	ALGORITHM_NAME

Constructor Summary

Constructors

Constructor and Description
SupervisedGraphWiseModel(PgxSession session, oracle.pgx.api.internal.Core core, java.util.function.Supplier<java.lang.String> keystorePathSupplier, java.util.function.Supplier<char[]> keystorePasswordSupplier, java.util.function.BiFunction<PgxSession,oracle.pgx.api.internal.Graph,PgxGraph> graphConstructor, oracle.pgx.api.internal.mllib.ModelMetadata modelMetadata)
SupervisedGraphWiseModel(PgxSession session, oracle.pgx.api.internal.Core core, java.util.function.Supplier<java.lang.String> keystorePathSupplier, java.util.function.Supplier<char[]> keystorePasswordSupplier, java.util.function.BiFunction<PgxSession,oracle.pgx.api.internal.Graph,PgxGraph> graphConstructor, oracle.pgx.api.internal.mllib.SupervisedGraphWiseModelMetadata modelMetadata) This constructor should never be used to get a model.

Method Summary

Modifier and Type	Method and Description
<ID> PgxFrame	evaluateLabels(PgxGraph graph, java.lang.Iterable<PgxVertex<ID>> vertices) Blocking version of evaluateLabelsAsync(PgxGraph, Iterable).
<ID> PgxFrame	evaluateLabels(PgxGraph graph, java.lang.Iterable<PgxVertex<ID>> vertices, float threshold) Blocking version of evaluateLabelsAsync(PgxGraph, Iterable, float).
<ID> PgxFuture<PgxFrame>	evaluateLabelsAsync(PgxGraph graph, java.lang.Iterable<PgxVertex<ID>> vertices) Evaluates (macro averaged) classification performance statistics for the specified vertices.
<ID> PgxFuture<PgxFrame>	evaluateLabelsAsync(PgxGraph graph, java.lang.Iterable<PgxVertex<ID>> vertices, float threshold) Evaluates (macro averaged) classification performance statistics for the specified vertices.
PgxFuture<java.lang.Double>	fitAsync(PgxGraph graph) Trains the GraphWise model on the input graph.
java.util.Map<?,java.lang.Float>	getClassWeights() Gets the class weights
LossFunction	getLossFunctionClass() Gets the loss function
GraphWisePredictionLayerConfig[]	getPredictionLayerConfigs() Gets the configuration objects for the prediction layers
java.lang.String	getVertexTargetPropertyName() Gets the target property name
<ID> SupervisedGnnExplanation<ID>	inferAndGetExplanation(PgxGraph graph, PgxVertex<ID> vertex) Blocking version of inferAndGetExplanation(PgxGraph, PgxVertex).
<ID> SupervisedGnnExplanation<ID>	inferAndGetExplanation(PgxGraph graph, PgxVertex<ID> vertex, float threshold) Blocking version of inferAndGetExplanation(PgxGraph, PgxVertex).
<ID> PgxFuture<SupervisedGnnExplanation<ID>>	inferAndGetExplanationAsync(PgxGraph graph, PgxVertex<ID> vertex) Performs inference on the specified vertex and generates an explanation that contains scores of how important each property and each vertex in the computation graph is for the prediction.
<ID> PgxFuture<SupervisedGnnExplanation<ID>>	inferAndGetExplanationAsync(PgxGraph graph, PgxVertex<ID> vertex, float threshold) Performs inference on the specified vertex and generates an explanation that contains scores of how important each property and each vertex in the computation graph is for the prediction.
<ID> PgxFuture<PgxFrame>	inferEmbeddingsAsync(PgxGraph graph, java.lang.Iterable<PgxVertex<ID>> vertices) Infers the embeddings for the specified vertices.
<ID> PgxFrame	inferLabels(PgxGraph graph, java.lang.Iterable<PgxVertex<ID>> vertices) Blocking version of inferLabelsAsync(PgxGraph, Iterable).
<ID> PgxFrame	inferLabels(PgxGraph graph, java.lang.Iterable<PgxVertex<ID>> vertices, float threshold) Blocking version of inferLabelsAsync(PgxGraph, Iterable, float).
<ID> PgxFuture<PgxFrame>	inferLabelsAsync(PgxGraph graph, java.lang.Iterable<PgxVertex<ID>> vertices) Infers the labels for the specified vertices.
<ID> PgxFuture<PgxFrame>	inferLabelsAsync(PgxGraph graph, java.lang.Iterable<PgxVertex<ID>> vertices, float threshold) Infers the labels for the specified vertices.
<ID> PgxFrame	inferLogits(PgxGraph graph, java.lang.Iterable<PgxVertex<ID>> vertices) Blocking version of inferLogitsAsync(PgxGraph, Iterable).
<ID> PgxFuture<PgxFrame>	inferLogitsAsync(PgxGraph graph, java.lang.Iterable<PgxVertex<ID>> vertices) Infers the prediction logits for the specified vertices.
void	store(java.lang.String path, java.lang.String key) Blocking version of storeAsync(String, String).
void	store(java.lang.String path, java.lang.String key, boolean overwrite) Blocking version of storeAsync(String, String).
PgxFuture<java.lang.Void>	storeAsync(java.lang.String path, java.lang.String key) Stores the GraphWise model in the specified path, with encryption.
PgxFuture<java.lang.Void>	storeAsync(java.lang.String path, java.lang.String key, boolean overwrite) Stores the GraphWise model in the specified path, with encryption.

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

destroy, destroyAsync, fit, getBatchSize, getConfig, getConvLayerConfigs, getEdgeInputFeatureDim, getEdgeInputPropertyNames, getEmbeddingDim, getInputFeatureDim, getLearningRate, getNumEpochs, getSeed, getTrainingLoss, getVertexInputPropertyNames, inferEmbeddings, isFitted

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

close, export

Methods inherited from class java.lang.Object

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

Field Detail

ALGORITHM_NAME

public static final java.lang.String ALGORITHM_NAME

See Also:

Constant Field Values

Constructor Detail

SupervisedGraphWiseModel

public SupervisedGraphWiseModel(PgxSession session,
                                oracle.pgx.api.internal.Core core,
                                java.util.function.Supplier<java.lang.String> keystorePathSupplier,
                                java.util.function.Supplier<char[]> keystorePasswordSupplier,
                                java.util.function.BiFunction<PgxSession,oracle.pgx.api.internal.Graph,PgxGraph> graphConstructor,
                                oracle.pgx.api.internal.mllib.ModelMetadata modelMetadata)

SupervisedGraphWiseModel

public SupervisedGraphWiseModel(PgxSession session,
                                oracle.pgx.api.internal.Core core,
                                java.util.function.Supplier<java.lang.String> keystorePathSupplier,
                                java.util.function.Supplier<char[]> keystorePasswordSupplier,
                                java.util.function.BiFunction<PgxSession,oracle.pgx.api.internal.Graph,PgxGraph> graphConstructor,
                                oracle.pgx.api.internal.mllib.SupervisedGraphWiseModelMetadata modelMetadata)

This constructor should never be used to get a model. Use SupervisedGraphWiseModelBuilder instead.

Parameters:

session - PgxSession to which the model is connected

core - Core to which the model is connected

graphConstructor - Constructor for a PgxGraph

modelMetadata - Metadata concerning the different hyper-parameters of the GraphWise Model

Since:

19.4

Method Detail

evaluateLabels

public <ID> PgxFrame evaluateLabels(PgxGraph graph,
                                    java.lang.Iterable<PgxVertex<ID>> vertices)

Blocking version of evaluateLabelsAsync(PgxGraph, Iterable). Evaluates (macro averaged) classification performance statistics for the specified vertices.

Parameters:

graph - the input graph

vertices - the vertices to evaluate the model on

Returns:

PgxFrame containing the metrics

Since:

19.4

evaluateLabels

public <ID> PgxFrame evaluateLabels(PgxGraph graph,
                                    java.lang.Iterable<PgxVertex<ID>> vertices,
                                    float threshold)

Blocking version of evaluateLabelsAsync(PgxGraph, Iterable, float). Evaluates (macro averaged) classification performance statistics for the specified vertices.

Parameters:

graph - the input graph

vertices - the vertices to evaluate the model on

threshold - decision threshold

Returns:

PgxFrame containing the labels for each vertex.

Since:

21.4

evaluateLabelsAsync

public <ID> PgxFuture<PgxFrame> evaluateLabelsAsync(PgxGraph graph,
                                                    java.lang.Iterable<PgxVertex<ID>> vertices)

Evaluates (macro averaged) classification performance statistics for the specified vertices.

Parameters:

graph - the input graph

vertices - the vertices to evaluate the model on

Returns:

PgxFrame containing the metrics

Since:

19.4

evaluateLabelsAsync

public <ID> PgxFuture<PgxFrame> evaluateLabelsAsync(PgxGraph graph,
                                                    java.lang.Iterable<PgxVertex<ID>> vertices,
                                                    float threshold)

Evaluates (macro averaged) classification performance statistics for the specified vertices.

Parameters:

graph - the input graph

vertices - the vertices to evaluate the model on

threshold - the decision threshold

Returns:

PgxFrame containing the labels for each vertex.

Since:

21.4

fitAsync

public PgxFuture<java.lang.Double> fitAsync(PgxGraph graph)

Trains the GraphWise model on the input graph.

Specified by:

fitAsync in class GraphWiseModel<SupervisedGraphWiseModelConfig,oracle.pgx.api.internal.mllib.SupervisedGraphWiseModelMetadata,SupervisedGraphWiseModel>

Parameters:

graph - input graph to fit on.

Since:

19.4

getClassWeights

public java.util.Map<?,java.lang.Float> getClassWeights()

Gets the class weights

Returns:

class weights

Since:

19.4

getLossFunctionClass

public LossFunction getLossFunctionClass()

Gets the loss function

Returns:

loss function

Since:

21.4

getPredictionLayerConfigs

public GraphWisePredictionLayerConfig[] getPredictionLayerConfigs()

Gets the configuration objects for the prediction layers

Returns:

configurations

Since:

19.4

getVertexTargetPropertyName

public java.lang.String getVertexTargetPropertyName()

Gets the target property name

Returns:

target property name

Since:

19.4

inferAndGetExplanation

public <ID> SupervisedGnnExplanation<ID> inferAndGetExplanation(PgxGraph graph,
                                                                PgxVertex<ID> vertex)

Blocking version of inferAndGetExplanation(PgxGraph, PgxVertex). Performs inference on the specified vertex and generates an explanation that contains scores of how important each property and each vertex in the computation graph is for the prediction.

Parameters:

graph - the input graph

vertex - the vertex

Returns:

explanation containing feature importance and vertex importance.

Since:

21.3

inferAndGetExplanation

public <ID> SupervisedGnnExplanation<ID> inferAndGetExplanation(PgxGraph graph,
                                                                PgxVertex<ID> vertex,
                                                                float threshold)

Blocking version of inferAndGetExplanation(PgxGraph, PgxVertex). Performs inference on the specified vertex and generates an explanation that contains scores of how important each property and each vertex in the computation graph is for the prediction.

Parameters:

graph - the input graph

vertex - the vertex

threshold - threshold

Returns:

explanation containing feature importance and vertex importance.

Since:

21.4

inferAndGetExplanationAsync

public <ID> PgxFuture<SupervisedGnnExplanation<ID>> inferAndGetExplanationAsync(PgxGraph graph,
                                                                                PgxVertex<ID> vertex)

Performs inference on the specified vertex and generates an explanation that contains scores of how important each property and each vertex in the computation graph is for the prediction.

Parameters:

graph - the input graph

vertex - the vertex

Returns:

explanation containing feature importance and vertex importance.

Since:

21.3

inferAndGetExplanationAsync

public <ID> PgxFuture<SupervisedGnnExplanation<ID>> inferAndGetExplanationAsync(PgxGraph graph,
                                                                                PgxVertex<ID> vertex,
                                                                                float threshold)

Performs inference on the specified vertex and generates an explanation that contains scores of how important each property and each vertex in the computation graph is for the prediction.

Parameters:

graph - the input graph

vertex - the vertex

threshold - decision threshold

Returns:

explanation containing feature importance and vertex importance.

Since:

21.4

inferEmbeddingsAsync

public <ID> PgxFuture<PgxFrame> inferEmbeddingsAsync(PgxGraph graph,
                                                     java.lang.Iterable<PgxVertex<ID>> vertices)

Infers the embeddings for the specified vertices.

Specified by:

inferEmbeddingsAsync in class GraphWiseModel<SupervisedGraphWiseModelConfig,oracle.pgx.api.internal.mllib.SupervisedGraphWiseModelMetadata,SupervisedGraphWiseModel>

Parameters:

graph - the input graph

vertices - the vertices to produce embeddings to

Returns:

PgxFrame containing the embeddings for each vertex.

Since:

19.4

inferLabels

public <ID> PgxFrame inferLabels(PgxGraph graph,
                                 java.lang.Iterable<PgxVertex<ID>> vertices)

Blocking version of inferLabelsAsync(PgxGraph, Iterable). Infers the labels for the specified vertices.

Parameters:

graph - the input graph

vertices - the vertices to produce labels to

Returns:

PgxFrame containing the labels for each vertex.

Since:

19.4

inferLabels

public <ID> PgxFrame inferLabels(PgxGraph graph,
                                 java.lang.Iterable<PgxVertex<ID>> vertices,
                                 float threshold)

Blocking version of inferLabelsAsync(PgxGraph, Iterable, float). Infers the labels for the specified vertices.

Parameters:

graph - the input graph

vertices - the vertices to produce labels to

Returns:

PgxFrame containing the labels for each vertex.

Since:

21.4

inferLabelsAsync

public <ID> PgxFuture<PgxFrame> inferLabelsAsync(PgxGraph graph,
                                                 java.lang.Iterable<PgxVertex<ID>> vertices)

Infers the labels for the specified vertices.

Parameters:

graph - the input graph

vertices - the vertices to produce labels to

Returns:

PgxFrame containing the labels for each vertex.

Since:

19.4

inferLabelsAsync

public <ID> PgxFuture<PgxFrame> inferLabelsAsync(PgxGraph graph,
                                                 java.lang.Iterable<PgxVertex<ID>> vertices,
                                                 float threshold)

Infers the labels for the specified vertices.

Parameters:

graph - the input graph

vertices - the vertices to produce labels to

threshold - the decision threshold

Returns:

PgxFrame containing the labels for each vertex.

Since:

21.4

inferLogits

public <ID> PgxFrame inferLogits(PgxGraph graph,
                                 java.lang.Iterable<PgxVertex<ID>> vertices)

Blocking version of inferLogitsAsync(PgxGraph, Iterable). Infers the prediction logits for the specified vertices.

Parameters:

graph - the input graph

vertices - the vertices to produce logits to

Returns:

PgxFrame containing the logits for each vertex.

Since:

19.4

inferLogitsAsync

public <ID> PgxFuture<PgxFrame> inferLogitsAsync(PgxGraph graph,
                                                 java.lang.Iterable<PgxVertex<ID>> vertices)

Infers the prediction logits for the specified vertices.

Parameters:

graph - the input graph

vertices - the vertices to produce logits to

Returns:

PgxFrame containing the logits for each vertex.

Since:

19.4

store

public void store(java.lang.String path,
                  java.lang.String key)
           throws java.util.concurrent.ExecutionException,
                  java.lang.InterruptedException

Blocking version of storeAsync(String, String). Calls storeAsync(String, String) and waits for the returned PgxFuture to complete.

Throws:

java.lang.InterruptedException - if the caller thread gets interrupted while waiting for completion.

java.util.concurrent.ExecutionException - if any exception occurred during asynchronous execution. The actual exception will be nested.

Since:

19.4

store

public void store(java.lang.String path,
                  java.lang.String key,
                  boolean overwrite)
           throws java.util.concurrent.ExecutionException,
                  java.lang.InterruptedException

Blocking version of storeAsync(String, String). Calls storeAsync(String, String) and waits for the returned PgxFuture to complete.

Throws:

java.lang.InterruptedException - if the caller thread gets interrupted while waiting for completion.

java.util.concurrent.ExecutionException - if any exception occurred during asynchronous execution. The actual exception will be nested.

Since:

21.1

storeAsync

public PgxFuture<java.lang.Void> storeAsync(java.lang.String path,
                                            java.lang.String key)
                                     throws java.util.concurrent.ExecutionException,
                                            java.lang.InterruptedException

Stores the GraphWise model in the specified path, with encryption.

Parameters:

path - path to store the model

key - the encryption key, or null if no encryption should be used.

Throws:

java.util.concurrent.ExecutionException

java.lang.InterruptedException

Since:

19.4

storeAsync

public PgxFuture<java.lang.Void> storeAsync(java.lang.String path,
                                            java.lang.String key,
                                            boolean overwrite)

Stores the GraphWise model in the specified path, with encryption.

Parameters:

path - path to store the model

key - the encryption key, or null if no encryption should be used.

Since:

21.1