public static Network CreateNetworkMaxout(GPUModule module, int minibatchSize)
{
var net = new Network(module, minibatchSize: minibatchSize);
net.AddInputLayer(Constants.TOTAL_VALUE_COUNT, sparseDataSize: minibatchSize * RawRecord.FEATURE_COUNT * 2);
net.AddLabelLayer(1);
var fc1 = net.AddFullyConnectedLayer(128, "FC1");
fc1.Weights.InitValuesUniformCPU(0.1f);
fc1.L2Regularization = 0.00001f;
fc1.RegularizationRatio = 10;
net.AddMaxoutLayer("MAXOUT1", groupsize: 4);
var fc2 = net.AddFullyConnectedLayer(256, "FC2");
fc2.Weights.InitValuesUniformCPU(0.1f);
net.AddMaxoutLayer("MAXOUT2", groupsize: 2);
net.AddDropoutLayer();
var sm = net.AddSoftmaxLayer(2, "SMAX");
sm.Weights.InitValuesUniformCPU(0.1f);
return(net);
}
public void CNNTest()
{
network.AddInputLayer(7, 1);
network.AddConvolutionLayer(2, 2);
network.AddPoolingLayer(2);
network.AddActivationLayer(ActivatorType.Relu);
network.AddConvolutionLayer(3, 2);
network.AddPoolingLayer(2);
network.AddActivationLayer(ActivatorType.Relu);
network.AddFlattenLayer();
network.AddDenseLayer(5, true);
network.AddActivationLayer(ActivatorType.Relu);
network.AddDenseLayer(3, false);
network.AddSoftMaxLayer();
TestNetwork(network, "test1");
}
private static Network CreateModel()
{
var network = new Network(LossFunctionType.CrossEntropy, new Flat(0.001), NumberOfClasses);
network.AddInputLayer(28, 1);
network.AddConvolutionLayer(16, 5);
network.AddActivationLayer(ActivatorType.Relu);
network.AddPoolingLayer(2);
network.AddConvolutionLayer(32, 5);
network.AddActivationLayer(ActivatorType.Relu);
network.AddPoolingLayer(2);
network.AddFlattenLayer();
network.AddDenseLayer(1024, true);
network.AddActivationLayer(ActivatorType.Relu);
network.AddDenseLayer(NumberOfClasses, false);
network.AddSoftMaxLayer();
network.RandomizeWeights(0.1);
return(network);
}
