Package oracle.pgx.api.mllib
Class GraphWiseDominantLayerConfigBuilder
- java.lang.Object
-
- oracle.pgx.api.mllib.GraphWiseDominantLayerConfigBuilder
-
public class GraphWiseDominantLayerConfigBuilder extends java.lang.Object
Builder forGraphWiseDominantLayerConfig
.- Since:
- 23.2
-
-
Constructor Summary
Constructors Constructor Description GraphWiseDominantLayerConfigBuilder()
-
Method Summary
All Methods Instance Methods Concrete Methods Modifier and Type Method Description GraphWiseDominantLayerConfig
build()
Build the configurationGraphWiseDominantLayerConfigBuilder
setAlpha(double alpha)
Set the alpha weight, the proportion of importance of feature reconstruction over graph structure reconstructionGraphWiseDominantLayerConfigBuilder
setDecoderLayerConfigs(GraphWisePredictionLayerConfig... layerConfigs)
Set the decoder layer configurations (SeeGraphWisePredictionLayerConfig
.
-
-
-
Method Detail
-
setDecoderLayerConfigs
public GraphWiseDominantLayerConfigBuilder setDecoderLayerConfigs(GraphWisePredictionLayerConfig... layerConfigs)
Set the decoder layer configurations (SeeGraphWisePredictionLayerConfig
. You must pass at least one prediction layer config (or leave it as default).Note that an additional layer will be inserted in the end for which:
- the hidden dimension will be equal to the input feature dimension
- the weight initialization scheme will be copied from the previous layer
default: null
- Parameters:
layerConfigs
- layer configs- Returns:
- this
- Since:
- 23.2
-
setAlpha
public GraphWiseDominantLayerConfigBuilder setAlpha(double alpha)
Set the alpha weight, the proportion of importance of feature reconstruction over graph structure reconstruction- Parameters:
alpha
- feature reconstruction weights- Returns:
- this
- Since:
- 23.2
-
build
public GraphWiseDominantLayerConfig build()
Build the configuration- Returns:
- built configuration
- Since:
- 23.2
-
-