public MnistMLPTrainer(MnistDataMgr mgr, MLP mlp, bool nInput = true, bool nOutput = false)
{
_dataMgr = mgr;
_normalizeInput = nInput;
_normalizeOuput = nOutput;
_network = mlp;
}
public void InitMLP(int[] layerStruct, double learnRate)
{
_normalizeOuput = false;
_normalizeInput = true;
_layerStruct = layerStruct;
_network = new MLP(_layerStruct, learnRate);
}
public MainWindow()
{
InitializeComponent();
Graph = new UserControl1();
this.Content = Graph;
var t = Generate();
// var s = t.Item1.Svd(true);
//var nd = t.Item1.Multiply(s.VT().SubMatrix(0, 8, 0, 8));
var mlp = new MLP(8, 100, 1);
var r = mlp.Train(t.Item1, t.Item2, null, null, 2500);
Graph.Set(r.TrainingSquaredError, r.TrainingError);
DenseMatrix data = new DenseMatrix(new double[,] { { 1, 1 }, { 1, 0 }, { 0, 0 }, { 0, 1 } });
DenseMatrix labels = new DenseMatrix(new double[,] { { 1, 0 }, { 0, 1 }, { 1, 0 }, { 0, 1 } });
//var mlp = new MLP(2, 100, 2);
//var r = mlp.Train(data, labels, data, labels, 5000);
//Graph.Set(r.TrainingSquaredError, r.TrainingError);
}
public MnistMLPTester(MLP mlp, MnistDataMgr dataMgr)
{
_mlp = mlp;
_dataMgr = dataMgr;
_testSetSize = _dataMgr.count;
_mse = 0;
_successRate = 0;
}
public DeepBeliefNet(SerializationInfo info, StreamingContext ctxt)
{
_layers = (List <double[]>)info.GetValue("Layers", typeof(List <double[]>));
_weights = (List <double[, ]>)info.GetValue("Weights", typeof(List <double[, ]>));
_bias = (List <double[]>)info.GetValue("Bias", typeof(List <double[]>));
_learningRate = (double)info.GetValue("LearningRate", typeof(double));
_outputNbr = (int)info.GetValue("OutputNbr", typeof(int));
_rnd = new Random();
_mlp = new MLP(_layers, _weights, _bias, _learningRate);
}
public DeepBeliefNet(double learningRate, LRBM lrbm, int outputNbr)
{
_rnd = new Random();
_learningRate = learningRate;
//_lrbm = lrbm;
_outputNbr = outputNbr;
_layers = new List <double[]>();
_weights = new List <double[, ]>();
_bias = new List <double[]>();
InitLRBM(lrbm);
InitOutput(lrbm);
_mlp = new MLP(_layers, _weights, _bias, _learningRate);
}
public void InitMLP(int nrHidden, double learnRate)
{
int nrInput = _dataMgr.inputNum;
int nrOutput;
if (_normalizeOuput)
{
nrOutput = 1;
}
else
{
nrOutput = 10;
}
_layerStruct = new int[] { nrInput, nrHidden, nrOutput };
_network = new MLP(_layerStruct, learnRate);
}
public DeepBeliefNet(double learningRate, LRBM lrbm, int[] outputStruct)
{
_rnd = new Random();
_learningRate = learningRate;
//_lrbm = lrbm;
int outputDepth = outputStruct.Length;
_outputNbr = outputStruct[outputDepth - 1];
_layers = new List <double[]>();
_weights = new List <double[, ]>();
_bias = new List <double[]>();
InitLRBM(lrbm);
InitOutputStruct(outputStruct, lrbm);
_mlp = new MLP(_layers, _weights, _bias, _learningRate);
}
void Init()
{
int [] netStruct = new int[] { _visibleNbr, _hiddenNbr, _visibleNbr };
_mlp = new MLP(netStruct, _learningRate);
}
