Package com.github.neuralnetworks.samples.mnist

Examples of com.github.neuralnetworks.samples.mnist.MnistInputProvider

30
31
32
33
34
35
36
37
38
39
40
41
42
43
    @Test
    public void testSigmoidBP() {
  NeuralNetworkImpl mlp = NNFactory.mlpSigmoid(new int[] { 784, 10 }, true);

  MnistInputProvider trainInputProvider = new MnistInputProvider("train-images.idx3-ubyte", "train-labels.idx1-ubyte", 1, 1, new MnistTargetMultiNeuronOutputConverter());
  trainInputProvider.addInputModifier(new ScalingInputFunction(255));
  MnistInputProvider testInputProvider = new MnistInputProvider("t10k-images.idx3-ubyte", "t10k-labels.idx1-ubyte", 1000, 1, new MnistTargetMultiNeuronOutputConverter());
  testInputProvider.addInputModifier(new ScalingInputFunction(255));

  BackPropagationTrainer<?> bpt = TrainerFactory.backPropagation(mlp, trainInputProvider, testInputProvider, new MultipleNeuronsOutputError(), new NNRandomInitializer(new MersenneTwisterRandomInitializer(-0.01f, 0.01f)), 0.02f, 0.5f, 0f, 0f);

  bpt.addEventListener(new LogTrainingListener(Thread.currentThread().getStackTrace()[1].getMethodName(), false, true));
View Full Code Here
52
53
54
55
56
57
58
59
60
61
62
63
64
65
    @Ignore
    @Test
    public void testSigmoidHiddenBP() {
  NeuralNetworkImpl mlp = NNFactory.mlpSigmoid(new int[] { 784, 300, 100, 10 }, true);

  MnistInputProvider trainInputProvider = new MnistInputProvider("train-images.idx3-ubyte", "train-labels.idx1-ubyte", 1, 2, new MnistTargetMultiNeuronOutputConverter());
  trainInputProvider.addInputModifier(new ScalingInputFunction(255));
  MnistInputProvider testInputProvider = new MnistInputProvider("t10k-images.idx3-ubyte", "t10k-labels.idx1-ubyte", 1000, 1, new MnistTargetMultiNeuronOutputConverter());
  testInputProvider.addInputModifier(new ScalingInputFunction(255));

  BackPropagationTrainer<?> bpt = TrainerFactory.backPropagation(mlp, trainInputProvider, testInputProvider, new MultipleNeuronsOutputError(), new NNRandomInitializer(new MersenneTwisterRandomInitializer(-0.01f, 0.01f)), 0.01f, 0.5f, 0f, 0f);

  bpt.addEventListener(new LogTrainingListener(Thread.currentThread().getStackTrace()[1].getMethodName(), false, true));
View Full Code Here
72
73
74
75
76
77
78
79
80
81
82
83
84
85
    }

    @Test
    public void testRBM() {
  RBM rbm = NNFactory.rbm(784, 10, false);
  MnistInputProvider trainInputProvider = new MnistInputProvider("train-images.idx3-ubyte", "train-labels.idx1-ubyte", 1, 1, new MnistTargetMultiNeuronOutputConverter());
  trainInputProvider.addInputModifier(new ScalingInputFunction(255));
  MnistInputProvider testInputProvider = new MnistInputProvider("t10k-images.idx3-ubyte", "t10k-labels.idx1-ubyte", 1000, 1, new MnistTargetMultiNeuronOutputConverter());
  testInputProvider.addInputModifier(new ScalingInputFunction(255));

  AparapiCDTrainer t = TrainerFactory.cdSigmoidTrainer(rbm, trainInputProvider, testInputProvider,  new MultipleNeuronsOutputError(), new NNRandomInitializer(new MersenneTwisterRandomInitializer(-0.01f, 0.01f)), 0.01f, 0.5f, 0f, 0f, 1, false);

  t.addEventListener(new LogTrainingListener(Thread.currentThread().getStackTrace()[1].getMethodName(), false, true));
  Environment.getInstance().setExecutionMode(EXECUTION_MODE.CPU);
View Full Code Here
92
93
94
95
96
97
98
99
100
101
102
103
104
105
    @Test
    public void testAE() {
  Autoencoder nn = NNFactory.autoencoderSigmoid(784, 10, true);

  MnistInputProvider trainInputProvider = new MnistInputProvider("train-images.idx3-ubyte", "train-labels.idx1-ubyte", 1, 1, new MnistTargetMultiNeuronOutputConverter());
  trainInputProvider.addInputModifier(new ScalingInputFunction(255));
  MnistInputProvider testInputProvider = new MnistInputProvider("t10k-images.idx3-ubyte", "t10k-labels.idx1-ubyte", 1000, 1, new MnistTargetMultiNeuronOutputConverter());
  testInputProvider.addInputModifier(new ScalingInputFunction(255));

  Trainer<?> t = TrainerFactory.backPropagationAutoencoder(nn, trainInputProvider, testInputProvider,  new MultipleNeuronsOutputError(), new NNRandomInitializer(new MersenneTwisterRandomInitializer(-0.01f, 0.01f)), 0.01f, 0.5f, 0f, 0f, 0f);

  t.addEventListener(new LogTrainingListener(Thread.currentThread().getStackTrace()[1].getMethodName(), false, true));
  Environment.getInstance().setExecutionMode(EXECUTION_MODE.CPU);
View Full Code Here
116
117
118
119
120
121
122
123
124
125
126
127
128
129
  NeuralNetworkImpl nn = NNFactory.convNN(new int[][] { { 28, 28, 1 }, { 5, 5, 20, 1 }, { 2, 2 }, { 5, 5, 50, 1 }, { 2, 2 }, {512}, {10} }, true);
  nn.setLayerCalculator(NNFactory.lcSigmoid(nn, null));
  NNFactory.lcMaxPooling(nn);

  // Mnist dataset provider
  MnistInputProvider trainInputProvider = new MnistInputProvider("train-images.idx3-ubyte", "train-labels.idx1-ubyte", 1, 1, new MnistTargetMultiNeuronOutputConverter());
  trainInputProvider.addInputModifier(new ScalingInputFunction(255));
  MnistInputProvider testInputProvider = new MnistInputProvider("t10k-images.idx3-ubyte", "t10k-labels.idx1-ubyte", 1000, 1, new MnistTargetMultiNeuronOutputConverter());
  testInputProvider.addInputModifier(new ScalingInputFunction(255));

  // Backpropagation trainer that also works for convolutional and subsampling layers
  BackPropagationTrainer<?> bpt = TrainerFactory.backPropagation(nn, trainInputProvider, testInputProvider, new MultipleNeuronsOutputError(), new NNRandomInitializer(new MersenneTwisterRandomInitializer(-0.01f, 0.01f), 0.5f), 0.01f, 0.5f, 0f, 0f);

  // log data
View Full Code Here
147
148
149
150
151
152
153
154
155
156
157
158
159
160
  NeuralNetworkImpl nn = NNFactory.convNN(new int[][] { { 28, 28, 1 }, { 2, 2 }, {10} }, true);
  nn.setLayerCalculator(NNFactory.lcSigmoid(nn, null));
  NNFactory.lcMaxPooling(nn);

  // MNIST dataset
  MnistInputProvider trainInputProvider = new MnistInputProvider("train-images.idx3-ubyte", "train-labels.idx1-ubyte", 1, 1, new MnistTargetMultiNeuronOutputConverter());
  trainInputProvider.addInputModifier(new ScalingInputFunction(255));
  MnistInputProvider testInputProvider = new MnistInputProvider("t10k-images.idx3-ubyte", "t10k-labels.idx1-ubyte", 1, 1, new MnistTargetMultiNeuronOutputConverter());
  testInputProvider.addInputModifier(new ScalingInputFunction(255));

  // Backpropagation trainer that also works for convolutional and subsampling layers
  BackPropagationTrainer<?> bpt = TrainerFactory.backPropagation(nn, trainInputProvider, testInputProvider, new MultipleNeuronsOutputError(), new NNRandomInitializer(new MersenneTwisterRandomInitializer(-0.01f, 0.01f)), 0.02f, 0.5f, 0f, 0f);

  // log data
View Full Code Here
182
183
184
185
186
187
188
189
190
191
192
193
194
195
  NeuralNetworkImpl nn = NNFactory.convNN(new int[][] { { 28, 28, 1 }, { 5, 5, 6, 1 }, {2, 2}, {10} }, true);
  nn.setLayerCalculator(NNFactory.lcSigmoid(nn, null));
  NNFactory.lcMaxPooling(nn);

  // MNIST dataset
  MnistInputProvider trainInputProvider = new MnistInputProvider("train-images.idx3-ubyte", "train-labels.idx1-ubyte", 1, 1, new MnistTargetMultiNeuronOutputConverter());
  trainInputProvider.addInputModifier(new ScalingInputFunction(255));
  MnistInputProvider testInputProvider = new MnistInputProvider("t10k-images.idx3-ubyte", "t10k-labels.idx1-ubyte", 1, 1, new MnistTargetMultiNeuronOutputConverter());
  testInputProvider.addInputModifier(new ScalingInputFunction(255));

  // Backpropagation trainer that also works for convolutional and subsampling layers
  BackPropagationTrainer<?> bpt = TrainerFactory.backPropagation(nn, trainInputProvider, testInputProvider, new MultipleNeuronsOutputError(), new NNRandomInitializer(new MersenneTwisterRandomInitializer(-0.01f, 0.01f)), 0.02f, 0.5f, 0f, 0f);

  // log data
View Full Code Here
27
28
29
30
31
32
33
34
35
36
37
38
39
40
    public void testSigmoidBP() {
  Environment.getInstance().setUseDataSharedMemory(false);
  Environment.getInstance().setUseWeightsSharedMemory(false);
  NeuralNetworkImpl mlp = NNFactory.mlpSigmoid(new int[] { 784, 10 }, true);

  MnistInputProvider trainInputProvider = new MnistInputProvider("train-images.idx3-ubyte", "train-labels.idx1-ubyte");
  trainInputProvider.addInputModifier(new ScalingInputFunction(255));
  MnistInputProvider testInputProvider = new MnistInputProvider("t10k-images.idx3-ubyte", "t10k-labels.idx1-ubyte");
  testInputProvider.addInputModifier(new ScalingInputFunction(255));

  BackPropagationTrainer<?> bpt = TrainerFactory.backPropagation(mlp, trainInputProvider, testInputProvider, new MultipleNeuronsOutputError(), new NNRandomInitializer(new MersenneTwisterRandomInitializer(-0.01f, 0.01f)), 0.02f, 0.5f, 0f, 0f, 0f, 1, 1000, 1);

  bpt.addEventListener(new LogTrainingListener(Thread.currentThread().getStackTrace()[1].getMethodName(), false, true));
View Full Code Here
49
50
51
52
53
54
55
56
57
58
59
60
61
62
    @Ignore
    @Test
    public void testSigmoidHiddenBP() {
  NeuralNetworkImpl mlp = NNFactory.mlpSigmoid(new int[] { 784, 300, 100, 10 }, true);

  MnistInputProvider trainInputProvider = new MnistInputProvider("train-images.idx3-ubyte", "train-labels.idx1-ubyte");
  trainInputProvider.addInputModifier(new ScalingInputFunction(255));
  MnistInputProvider testInputProvider = new MnistInputProvider("t10k-images.idx3-ubyte", "t10k-labels.idx1-ubyte");
  testInputProvider.addInputModifier(new ScalingInputFunction(255));

  BackPropagationTrainer<?> bpt = TrainerFactory.backPropagation(mlp, trainInputProvider, testInputProvider, new MultipleNeuronsOutputError(), new NNRandomInitializer(new MersenneTwisterRandomInitializer(-0.01f, 0.01f)), 0.01f, 0.5f, 0f, 0f, 0f, 1, 1000, 2);

  bpt.addEventListener(new LogTrainingListener(Thread.currentThread().getStackTrace()[1].getMethodName(), false, true));
View Full Code Here
77
78
79
80
81
82
83
84
85
86
87
88
89
90
  NeuralNetworkImpl nn = NNFactory.convNN(new int[][] { { 28, 28, 1 }, { 5, 5, 20, 1 }, { 2, 2 }, { 5, 5, 50, 1 }, { 2, 2 }, {512}, {10} }, true);
  nn.setLayerCalculator(NNFactory.lcSigmoid(nn, null));
  NNFactory.lcMaxPooling(nn);

  // Mnist dataset provider
  MnistInputProvider trainInputProvider = new MnistInputProvider("train-images.idx3-ubyte", "train-labels.idx1-ubyte");
  trainInputProvider.addInputModifier(new ScalingInputFunction(255));
  MnistInputProvider testInputProvider = new MnistInputProvider("t10k-images.idx3-ubyte", "t10k-labels.idx1-ubyte");
  testInputProvider.addInputModifier(new ScalingInputFunction(255));

  // Backpropagation trainer that also works for convolutional and subsampling layers
  BackPropagationTrainer<?> bpt = TrainerFactory.backPropagation(nn, trainInputProvider, testInputProvider, new MultipleNeuronsOutputError(), new NNRandomInitializer(new MersenneTwisterRandomInitializer(-0.01f, 0.01f), 0.5f), 0.01f, 0.5f, 0f, 0f, 0f, 1, 1000, 1);

  // log data
View Full Code Here
TOP

Related Classes of com.github.neuralnetworks.samples.mnist.MnistInputProvider

  • com.github.neuralnetworks.samples.test.MnistTest
  • com.github.neuralnetworks.training.TrainingInputData
  • com.github.neuralnetworks.training.TrainingInputDataImpl
  • com.github.neuralnetworks.util.Matrix
  • java.util.Random
  • java.io.RandomAccessFile
    • 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.