Examples of EvolutionaryAlgorithm

Examples of com.heatonresearch.aifh.evolutionary.train.EvolutionaryAlgorithm

73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
        Population pop = new BasicPopulation();
        pop.setGenomeFactory(new IntegerArrayGenomeFactory(5));

        // Create a trainer with a very simple score function.  We do not care
        // about the calculation of the score, as they will never be calculated.
        EvolutionaryAlgorithm train = new BasicEA(pop, new ScoreFunction() {
            @Override
            public double calculateScore(MLMethod method) {
                return 0;
            }

            @Override
            public boolean shouldMinimize() {
                return false;
            }
        });

        // Create a shuffle operator.  Use it 1.0 (100%) of the time.
        MutateShuffle opp = new MutateShuffle();
        train.addOperation(1.0, opp);

        // Create a single parent, the genes are set to 1,2,3,4,5.
        IntegerArrayGenome[] parents = new IntegerArrayGenome[1];
        parents[0] = (IntegerArrayGenome) pop.getGenomeFactory().factor();
        for (int i = 1; i <= 5; i++) {
View Full Code Here

Examples of com.heatonresearch.aifh.evolutionary.train.EvolutionaryAlgorithm

124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
        Population pop = new BasicPopulation();
        pop.setGenomeFactory(new DoubleArrayGenomeFactory(5));

        // Create a trainer with a very simple score function.  We do not care
        // about the calculation of the score, as they will never be calculated.
        EvolutionaryAlgorithm train = new BasicEA(pop, new ScoreFunction() {
            @Override
            public double calculateScore(MLMethod method) {
                return 0;
            }

            @Override
            public boolean shouldMinimize() {
                return false;
            }
        });


        MutatePerturb opp = new MutatePerturb(0.1);
        train.addOperation(1.0, opp);


        // Create a peterb operator.  Use it 1.0 (100%) of the time.
        DoubleArrayGenome[] parents = new DoubleArrayGenome[1];
        parents[0] = (DoubleArrayGenome) pop.getGenomeFactory().factor();
View Full Code Here

Examples of com.heatonresearch.aifh.evolutionary.train.EvolutionaryAlgorithm

129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
        GenerateRandom rnd = new MersenneTwisterGenerateRandom();
        EvaluateExpression eval = new EvaluateExpression(rnd);
        Population pop = initPopulation(rnd, eval);
        ScoreFunction score = new ScoreSmallExpression(training,30);

        EvolutionaryAlgorithm genetic = new BasicEA(pop, score);
        genetic.addOperation(0.3, new MutateTree(3));
        genetic.addOperation(0.7, new CrossoverTree());
        genetic.setShouldIgnoreExceptions(false);


        int sameSolutionCount = 0;
        int iteration = 1;
        double lastSolution = Double.MAX_VALUE;
        StringBuilder builder = new StringBuilder();

        while (sameSolutionCount < MAX_SAME_SOLUTION && iteration<1000) {
            genetic.iteration();

            double thisSolution = genetic.getLastError();

            builder.setLength(0);
            builder.append("Iteration: ");
            builder.append(iteration++);
            builder.append(", Current error = ");
            builder.append(thisSolution);
            builder.append(", Best Solution Length = ");
            builder.append(genetic.getBestGenome().size());

            System.out.println(builder.toString());

            if (Math.abs(lastSolution - thisSolution) < 1.0) {
                sameSolutionCount++;
            } else {
                sameSolutionCount = 0;
            }

            lastSolution = thisSolution;
        }

        System.out.println("Good solution found:");
        TreeGenome best = (TreeGenome) genetic.getBestGenome();
        System.out.println(eval.displayExpressionNormal(best.getRoot()));
        genetic.finishTraining();
    }
View Full Code Here

Examples of com.heatonresearch.aifh.evolutionary.train.EvolutionaryAlgorithm

75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
        Population pop = new BasicPopulation();
        pop.setGenomeFactory(new IntegerArrayGenomeFactory(10));

        // Create a trainer with a very simple score function.  We do not care
        // about the calculation of the score, as they will never be calculated.
        EvolutionaryAlgorithm train = new BasicEA(pop, new ScoreFunction() {
            @Override
            public double calculateScore(MLMethod method) {
                return 0;
            }

            @Override
            public boolean shouldMinimize() {
                return false;
            }
        });

        // Create a splice operator, length = 5.  Use it 1.0 (100%) of the time.
        Splice opp = new Splice(5);
        train.addOperation(1.0, opp);

        // Create two parents, the genes are set to 1,2,3,4,5,7,8,9,10
        // and 10,9,8,7,6,5,4,3,2,1.
        IntegerArrayGenome[] parents = new IntegerArrayGenome[2];
        parents[0] = (IntegerArrayGenome) pop.getGenomeFactory().factor();
View Full Code Here

Examples of com.heatonresearch.aifh.evolutionary.train.EvolutionaryAlgorithm

131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
        Population pop = new BasicPopulation();
        pop.setGenomeFactory(new IntegerArrayGenomeFactory(10));

        // Create a trainer with a very simple score function.  We do not care
        // about the calculation of the score, as they will never be calculated.
        EvolutionaryAlgorithm train = new BasicEA(pop, new ScoreFunction() {
            @Override
            public double calculateScore(MLMethod method) {
                return 0;
            }

            @Override
            public boolean shouldMinimize() {
                return false;
            }
        });

        // Create a splice (no repeat) operator, length = 5.  Use it 1.0 (100%) of the time.
        SpliceNoRepeat opp = new SpliceNoRepeat(5);
        train.addOperation(1.0, opp);

        // Create two parents, the genes are set to 1,2,3,4,5,7,8,9,10
        // and 10,9,8,7,6,5,4,3,2,1.
        IntegerArrayGenome[] parents = new IntegerArrayGenome[2];
        parents[0] = (IntegerArrayGenome) pop.getGenomeFactory().factor();
View Full Code Here

Examples of com.heatonresearch.aifh.evolutionary.train.EvolutionaryAlgorithm

79
80
81
82
83
84
85
86
87
88
89
        GenerateRandom rnd = new MersenneTwisterGenerateRandom();

        // Create a trainer with a very simple score function.  We do not care
        // about the calculation of the score, as they will never be calculated.
        // We only care that we are maximizing.
        EvolutionaryAlgorithm train = new BasicEA(pop, new ScoreFunction() {
            @Override
            public double calculateScore(MLMethod method) {
                return 0;
            }
View Full Code Here

Examples of org.encog.ml.ea.train.EvolutionaryAlgorithm

55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
    pop.reset();

    CalculateScore score = new TrainingSetScore(trainingSet);
    // train the neural network
   
    final EvolutionaryAlgorithm train = NEATUtil.constructNEATTrainer(pop,score);
   
    do {
      train.iteration();
      System.out.println("Epoch #" + train.getIteration() + " Error:" + train.getError()+ ", Species:" + pop.getSpecies().size());
    } while(train.getError() > 0.01);

    NEATNetwork network = (NEATNetwork)train.getCODEC().decode(train.getBestGenome());

    // test the neural network
    System.out.println("Neural Network Results:");
    EncogUtility.evaluate(network, trainingSet);
   
View Full Code Here

Examples of org.encog.ml.ea.train.EvolutionaryAlgorithm

121
122
123
124
125
126
127
128
129
130
131
132
   
    // reload the method
    method = null;
   
    if( trainer instanceof EvolutionaryAlgorithm ) {
      EvolutionaryAlgorithm ea = (EvolutionaryAlgorithm)trainer;
      method = ea.getPopulation();
    }
   
    if( method==null ) {
      method = trainer.getMethod(); 
    }
View Full Code Here

Examples of org.encog.ml.ea.train.EvolutionaryAlgorithm

55
56
57
58
59
60
61
62
63
64
65
    CalculateScore score = new TrainingSetScore(trainingSet);
    // train the neural network
    ActivationStep step = new ActivationStep();
    step.setCenter(0.5);

    EvolutionaryAlgorithm train = NEATUtil.constructNEATTrainer(
        score, 2, 1, 10);
    //train.setOutputActivationFunction(step);
   
    return (NEATPopulation)train.getPopulation();
  }
View Full Code Here

Examples of org.encog.ml.ea.train.EvolutionaryAlgorithm

91
92
93
94
95
96
97
98
99
    Assert.assertEquals(0.2,pop.getSurvivalRate());
   
    // see if the population can actually be used to train
    MLDataSet trainingSet = new BasicMLDataSet(XOR.XOR_INPUT, XOR.XOR_IDEAL);   
    CalculateScore score = new TrainingSetScore(trainingSet);
    EvolutionaryAlgorithm train = NEATUtil.constructNEATTrainer(pop, score);
    train.iteration();

  }
View Full Code Here
TOP
Copyright © 2018 www.massapi.com. All rights reserved.
All source code are property of their respective owners. Java is a trademark of Sun Microsystems, Inc and owned by ORACLE Inc. Contact coftware#gmail.com.