Package eu.h2020.helios_social.modules.socialgraphmining.GNN

Class ContextTrainingExampleData

java.lang.Object

eu.h2020.helios_social.modules.socialgraphmining.GNN.ContextTrainingExampleData

public class ContextTrainingExampleData
extends java.lang.Object

This class provides a storage structure that organizes a list of TrainingExample data to be stored in the contextual ego network's contexts. It is indented to be used as a dynamically created instance on contexts (which are cross module components) by calling context.getOrCreateInstance(GNNNodeData.class) to either retrieve of create it.

Author:

Emmanouil Krasanakis

Field Summary

Fields

Modifier and Type	Field	Description
mklab.JGNN.core.Tensor	transformToDstEmbedding
mklab.JGNN.core.Tensor	transformToSrcEmbedding

Constructor Summary

Constructors

Constructor	Description
ContextTrainingExampleData()

Method Summary

Modifier and Type	Method	Description
void	addTrainingExample(TrainingExample example)
void	degrade(double factor, double removalThreshold)	Calls the TrainingExample.degrade(double) operation for each TrainingExample in the data (e.g.
java.util.ArrayList<TrainingExample>	getTrainingExampleList()	Grants direct access to a list of training examples to traverse or edit.

Methods inherited from class java.lang.Object

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

Field Detail

transformToSrcEmbedding

public mklab.JGNN.core.Tensor transformToSrcEmbedding

transformToDstEmbedding

public mklab.JGNN.core.Tensor transformToDstEmbedding

Constructor Detail

ContextTrainingExampleData

public ContextTrainingExampleData()

Method Detail

addTrainingExample

public void addTrainingExample(TrainingExample example)

getTrainingExampleList

public java.util.ArrayList<TrainingExample> getTrainingExampleList()

Grants direct access to a list of training examples to traverse or edit.

Returns:

An array list of training examples.

degrade

public void degrade(double factor,
                    double removalThreshold)

Calls the TrainingExample.degrade(double) operation for each TrainingExample in the data (e.g. to reduce all weights) and removes those that end up on a very small weight.

Parameters:

factor - The degrading factor passed on to the degrade operation.

removalThreshold - The threshold weight under which training examples are removed.

See Also:

TrainingExample.getWeight()