Package net.maswag.falcaun

Class SAEQOracle

java.lang.Object

net.maswag.falcaun.AbstractSelectEQOracle

net.maswag.falcaun.HillClimbingEQOracle

net.maswag.falcaun.SAEQOracle

All Implemented Interfaces:

EquivalenceOracle<net.automatalib.automaton.transducer.MealyMachine<?,String,?,String>,String,net.automatalib.word.Word<String>>, EquivalenceOracle.MealyEquivalenceOracle<String,String>, EvaluationCountable, EvaluationCountable.MealyEquivalenceOracle<String,String>

public class SAEQOracle
extends HillClimbingEQOracle

Answer the Equivalence query by simulated annealing

Author:

Masaki Waga

Nested Class Summary

Nested classes/interfaces inherited from interface de.learnlib.oracle.EquivalenceOracle

EquivalenceOracle.DFAEquivalenceOracle<I extends Object>, EquivalenceOracle.MealyEquivalenceOracle<I extends Object,O extends Object>, EquivalenceOracle.MooreEquivalenceOracle<I extends Object,O extends Object>

Nested classes/interfaces inherited from interface net.maswag.falcaun.EvaluationCountable

EvaluationCountable.MealyEquivalenceOracle<I,O>, EvaluationCountable.Sum

Field Summary

Fields

Modifier and Type	Field	Description
private double	alpha
private double	iteration
private int	maxIteration

Fields inherited from class net.maswag.falcaun.AbstractSelectEQOracle

childrenSize, generationSize, memOracle, random, resetWord, symbolList

Constructor Summary

Constructors

Constructor	Description
SAEQOracle(NumericMembershipOracleCost memOracle, int length, Random random, int maxTests, int generationSize, int childrenSize, boolean resetWord, double alpha)	Constructor for SAEQOracle.
SAEQOracle(NumericMembershipOracleCost memOracle, int length, Random random, int maxTests, int generationSize, int childrenSize, boolean resetWord, double alpha, PropertyOracle.MealyPropertyOracle<String,String,String> ltlOracle)

Method Summary

Modifier and Type	Method	Description
protected List<net.automatalib.word.Word<String>>	createNextGeneration(List<net.automatalib.word.Word<String>> goodSamples)	Generates the next set of test cases based on the provided list of successful (good) samples.
private double	onlyEval(net.automatalib.word.Word<String> input)
private boolean	shouldPick(double srcCost, double dstCost)

Methods inherited from class net.maswag.falcaun.HillClimbingEQOracle

neighborhoodStream

Methods inherited from class net.maswag.falcaun.AbstractSelectEQOracle

findCounterExample

Methods inherited from class java.lang.Object

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

Methods inherited from interface net.maswag.falcaun.EvaluationCountable

getEvaluateCount

Field Details

alpha

private double alpha

maxIteration

private int maxIteration

iteration

private double iteration

Constructor Details

SAEQOracle

public SAEQOracle(NumericMembershipOracleCost memOracle,
 int length,
 Random random,
 int maxTests,
 int generationSize,
 int childrenSize,
 boolean resetWord,
 double alpha)

Constructor for SAEQOracle.

Parameters:

memOracle - a NumericMembershipOracleCost object.

length - a int.

random - a Random object.

maxTests - a int.

generationSize - a int.

childrenSize - a int.

resetWord - a boolean.

alpha - a double.

SAEQOracle

SAEQOracle(NumericMembershipOracleCost memOracle,
 int length,
 Random random,
 int maxTests,
 int generationSize,
 int childrenSize,
 boolean resetWord,
 double alpha,
 PropertyOracle.MealyPropertyOracle<String,String,String> ltlOracle)

Method Details

createNextGeneration

protected List<net.automatalib.word.Word<String>> createNextGeneration(List<net.automatalib.word.Word<String>> goodSamples)

Generates the next set of test cases based on the provided list of successful (good) samples. This method is responsible for creating new test cases that are derived from the current set of successful samples, ensuring a diverse and evolving set of test cases.

Overrides:

createNextGeneration in class HillClimbingEQOracle

Parameters:

goodSamples - A list of successful test cases used as the basis for generating the next generation of test cases.

Returns:

A list of newly generated test cases for the next evaluation cycle.

shouldPick

private boolean shouldPick(double srcCost,
 double dstCost)

Parameters:

srcCost - the cost of src

dstCost - the cost of dst

Returns:

if we should transit from src to dst. We followed the algorithm in https://ja.wikipedia.org/wiki/%E7%84%BC%E3%81%8D%E3%81%AA%E3%81%BE%E3%81%97%E6%B3%95 .

onlyEval

private double onlyEval(net.automatalib.word.Word<String> input)