Package tv.floe.metronome.classification.neuralnetworks.networks

Examples of tv.floe.metronome.classification.neuralnetworks.networks.MultiLayerPerceptronNetwork

62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
    c.setConfValue("neuronType", Neuron.class);
    c.setConfValue("networkType", NeuralNetwork.NetworkType.MULTI_LAYER_PERCEPTRON);
    c.setConfValue("layerNeuronCounts", "2,3,1" );
    c.parse(null);
   
    MultiLayerPerceptronNetwork mlp_network = new MultiLayerPerceptronNetwork();
   
   
   
//    int[] neurons = { 2, 3, 1 };
//    c.setLayerNeuronCounts( neurons );
   
    mlp_network.buildFromConf(c);   
   
    return mlp_network;
  }
View Full Code Here
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
    vec.set(1,  2);
   
    Config c = new Config();
    c.parse(null); // default layer: 2-3-2
   
    MultiLayerPerceptronNetwork mlp_network = new MultiLayerPerceptronNetwork();
   
    mlp_network.buildFromConf(c);
   

    mlp_network.setInputVector(vec);
   
    //mlp_network.
   
    Layer l0 = mlp_network.getLayerByIndex(0);
   
    assertEquals( 2, l0.getNeurons().size() );
   
    assertEquals(0, l0.getNeuronAt(0).getInConnections().size() );
    assertEquals(0, l0.getNeuronAt(1).getInConnections().size() );

    assertEquals(3, l0.getNeuronAt(0).getOutConnections().size() );
    assertEquals(3, l0.getNeuronAt(1).getOutConnections().size() );

   
   
    Layer l1 = mlp_network.getLayerByIndex(1);

    assertEquals( 3, l1.getNeurons().size() );
   
    assertEquals(2, l1.getNeuronAt(0).getInConnections().size() );
    assertEquals(2, l1.getNeuronAt(1).getInConnections().size() );
    assertEquals(2, l1.getNeuronAt(2).getInConnections().size() );

    assertEquals(2, l1.getNeuronAt(0).getOutConnections().size() );
    assertEquals(2, l1.getNeuronAt(1).getOutConnections().size() );
    assertEquals(2, l1.getNeuronAt(2).getOutConnections().size() );
   
   
   
    Layer l2 = mlp_network.getLayerByIndex(2);
   
    assertEquals( 2, l2.getNeurons().size() );
   
    assertEquals(3, l2.getNeuronAt(0).getInConnections().size() );
    assertEquals(3, l2.getNeuronAt(1).getInConnections().size() );
View Full Code Here
43
44
45
46
47
48
49
50
51
52
53
  }
 
 
  public static MultiLayerPerceptronNetwork loadModelFromDisk() {
   
    MultiLayerPerceptronNetwork nnet = null;
   
    try {

      Path out = new Path("/tmp/nn.model");
      FileSystem fs =
View Full Code Here
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
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
 
 
  public static void scoreIrisNeuralNetworkModel(String modelFileLocation) throws Exception {
   
    // load model
    MultiLayerPerceptronNetwork mlp = loadModelFromDisk();
   
    //Utils.PrintNeuralNetwork(  mlp );
   
    //System.out.println("Layers: " + mlp.getLayersCount());
   
    MetronomeRecordFactory rec_factory = new MetronomeRecordFactory("i:4 | o:3");
 
     
      String recs = "src/test/resources/data/uci/iris/iris_data_normalised.mne";
      Vector v_in_0 = new RandomAccessSparseVector(rec_factory.getInputVectorSize());
      Vector v_out_0 = new RandomAccessSparseVector(rec_factory.getOutputVectorSize());
     
     
     
      int total_recs = 0;
      int correct = 0;
     
      BufferedReader br = new BufferedReader(new FileReader(recs));
      String line;
      while ((line = br.readLine()) != null) {
       
       

        rec_factory.vectorizeLine( line, v_in_0, v_out_0 );
       
        if (isCorrect(mlp, v_in_0, v_out_0) ) {
          correct++;
        }
        total_recs++;
       
        //System.out.println("rec > " + line);
        //System.out.println("out > " + v_out_0.toString());
       
      }
      br.close();   
     
      BackPropogationLearningAlgorithm bp = ((BackPropogationLearningAlgorithm)mlp.getLearningRule());
     
      System.out.println("Avg RMSE: " + bp.getMetrics().getLastRMSE());
      System.out.println("Total: " + total_recs);
      System.out.println("Correct: " + correct);
   
View Full Code Here
68
69
70
71
72
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
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
    c.setConfValue("neuronType", Neuron.class);
    c.setConfValue("networkType", NeuralNetwork.NetworkType.MULTI_LAYER_PERCEPTRON);
    c.setConfValue("layerNeuronCounts", "2,3,1" );
   
   
    MultiLayerPerceptronNetwork mlp_network = new MultiLayerPerceptronNetwork();
   
   
   

//    mlp_network.setInputVector(vec);
   
    int[] neurons = { 2, 3, 1 };
    c.setLayerNeuronCounts( neurons );
   
    mlp_network.buildFromConf(c);
       
   
    for ( int x = 0; x < 40000; x++ ) {
     
     
         

      mlp_network.train(v0_out, v0);
      mlp_network.train(v1_out, v1);
      mlp_network.train(v2_out, v2);
      mlp_network.train(v3_out, v3);

     
      int total_records = 0;
      int number_correct = 0;
         
      total_records++;
     
     
      mlp_network.setInputVector( v0 );
      mlp_network.calculate();
            Vector networkOutput = mlp_network.getOutputVector();

            System.out.println( "> out: 0 =? " + networkOutput.get(0) );
               
           
           
      mlp_network.setInputVector( v1 );
      mlp_network.calculate();
            Vector networkOutput_1 = mlp_network.getOutputVector();

            System.out.println( "> out: 1 =? " + networkOutput_1.get(0) );
                   

      mlp_network.setInputVector( v2 );
      mlp_network.calculate();
            Vector networkOutput_2 = mlp_network.getOutputVector();

            System.out.println( "> out: 1 =? " + networkOutput_2.get(0) );


      mlp_network.setInputVector( v3 );
      mlp_network.calculate();
            Vector networkOutput_3 = mlp_network.getOutputVector();

            System.out.println( "> out: 0 =? " + networkOutput_3.get(0) );
           
           
     
View Full Code Here
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
    c.setConfValue("neuronType", Neuron.class);
    c.setConfValue("networkType", NeuralNetwork.NetworkType.MULTI_LAYER_PERCEPTRON);
    c.setConfValue("layerNeuronCounts", "2,3,1" );
    c.parse(null);
   
    MultiLayerPerceptronNetwork mlp_network = new MultiLayerPerceptronNetwork();
   
   
   
//    int[] neurons = { 2, 3, 1 };
//    c.setLayerNeuronCounts( neurons );
   
    mlp_network.buildFromConf(c);   
   
    return mlp_network;
  }   
View Full Code Here
36
37
38
39
40
41
42
43
44
45
46
  }
 
 
  public static MultiLayerPerceptronNetwork loadModelFromDisk(String model_path) {
   
    MultiLayerPerceptronNetwork nnet = null;
   
    try {

      Path out = new Path(model_path);
      FileSystem fs =
View Full Code Here
88
89
90
91
92
93
94
95
96
97
98
  }
 
  public static void scoreIrisNeuralNetworkModel(String modelFileLocation) throws Exception {
   
    // load model
    MultiLayerPerceptronNetwork mlp = loadModelFromDisk(modelFileLocation);
   
    //Utils.PrintNeuralNetwork(  mlp );
   
    //System.out.println("Layers: " + mlp.getLayersCount());
   
View Full Code Here
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
    c.setConfValue("neuronType", Neuron.class);
    c.setConfValue("networkType", NeuralNetwork.NetworkType.MULTI_LAYER_PERCEPTRON);
    c.setConfValue("layerNeuronCounts", "2,3,1" );
    c.parse(null);
   
    MultiLayerPerceptronNetwork mlp_network = new MultiLayerPerceptronNetwork();
   
   
   
//    int[] neurons = { 2, 3, 1 };
//    c.setLayerNeuronCounts( neurons );
   
    mlp_network.buildFromConf(c);   
   
    return mlp_network;
  } 
View Full Code Here
46
47
48
49
50
51
52
53
54
55
56
57
58
59
    c.setConfValue("neuronType", Neuron.class);
    c.setConfValue("networkType", NeuralNetwork.NetworkType.MULTI_LAYER_PERCEPTRON);
    c.setConfValue("layerNeuronCounts", "2,3,1" );
    c.parse(null);
   
    NeuralNetwork nn = new MultiLayerPerceptronNetwork();
   
    try {
      nn.buildFromConf(c);
    } catch (Exception e) {
      // TODO Auto-generated catch block
      e.printStackTrace();
    }
   
View Full Code Here
TOP

Related Classes of tv.floe.metronome.classification.neuralnetworks.networks.MultiLayerPerceptronNetwork

  • tv.floe.metronome.classification.neuralnetworks.core.Layer
  • tv.floe.metronome.classification.neuralnetworks.core.neurons.BiasNeuron
  • tv.floe.metronome.classification.neuralnetworks.core.TestLayerToVectorSerde
  • tv.floe.metronome.classification.neuralnetworks.core.TestNeuralNetwork
  • tv.floe.metronome.classification.neuralnetworks.eval.ModelEvaluator
  • tv.floe.metronome.classification.neuralnetworks.iterativereduce.abalone.AbaloneModelTestingTools
  • tv.floe.metronome.classification.neuralnetworks.iterativereduce.iris.IrisDatasetUtils
  • tv.floe.metronome.classification.neuralnetworks.iterativereduce.MasterNode
  • tv.floe.metronome.classification.neuralnetworks.iterativereduce.NeuralNetworkUtil
  • tv.floe.metronome.classification.neuralnetworks.iterativereduce.TestNeuralNetworkUtil
    • Copyright © 2018 www.massapicom. 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.