| Package | Description |
|---|---|
| oracle.pgx.api.mllib |
This package contains graph machine learning tools for use with PGX.
|
| oracle.pgx.config.mllib |
| Modifier and Type | Method and Description |
|---|---|
GraphWisePredictionLayerConfig |
GraphWisePredictionLayerConfigBuilder.build()
Build the configuration
|
GraphWisePredictionLayerConfig[] |
SupervisedGraphWiseModel.getPredictionLayerConfigs()
Gets the configuration objects for the prediction layers
|
| Modifier and Type | Method and Description |
|---|---|
SupervisedGraphWiseModelBuilder |
SupervisedGraphWiseModelBuilder.setPredictionLayerConfigs(GraphWisePredictionLayerConfig... layerConfigs)
Set the prediction layer configurations (See
GraphWisePredictionLayerConfig. |
| Modifier and Type | Field and Description |
|---|---|
static GraphWisePredictionLayerConfig[] |
SupervisedGraphWiseModelConfig.DEFAULT_PREDICTION_LAYER_CONFIGS
one default initialized config (See
GraphWisePredictionLayerConfig) |
| Modifier and Type | Method and Description |
|---|---|
GraphWisePredictionLayerConfig[] |
SupervisedGraphWiseModelConfig.getPredictionLayerConfigs() |
| Modifier and Type | Method and Description |
|---|---|
void |
SupervisedGraphWiseModelConfig.setPredictionLayerConfigs(GraphWisePredictionLayerConfig... predictionLayerConfigs) |
| Constructor and Description |
|---|
SupervisedGraphWiseModelConfig(int batchSize, int numEpochs, double learningRate, int embeddingDim, java.lang.Integer seed, GraphWiseConvLayerConfig[] convLayerConfigs, boolean standardize, boolean shuffle, java.util.List<java.lang.String> vertexInputPropertyNames, java.util.List<java.lang.String> edgeInputPropertyNames, boolean fitted, double trainingLoss, int inputFeatureDim, int edgeInputFeatureDim, oracle.pgx.config.mllib.SupervisedGraphWiseModelConfig.LossFunction lossFunction, LossFunction lossFunctionClass, BatchGenerator batchGenerator, GraphWisePredictionLayerConfig[] predictionLayerConfigs, boolean normalize, java.lang.String vertexTargetPropertyName, LabelMaps labelMaps, GraphWiseModelConfig.Backend backend) |