public void ResetAutoTest()
{
int[] layers = new int[] { 4, 5, 1 };
MultilayerPerceptron nn = new MLPGenerator().Create(layers, 1, new Sigmoid());
double[] randomSelected = new double[] {
nn.Structure.Input[1].Next[1].Weight.Value,
nn.Structure.Input[1].Next[2].Weight.Value,
nn.Structure.Input[1].Next[2].Next.Next[0].Weight.Value,
nn.Structure.Output[0].Previous[1].Weight.Value
};
double[] inp = new double[] { -0.978, 2.34, 0.2, -0.33 };
double[] pulsed = nn.Pulse(inp);
nn.Reset(0, 1);
double[] randomSelected2 = new double[] {
nn.Structure.Input[1].Next[1].Weight.Value,
nn.Structure.Input[1].Next[2].Weight.Value,
nn.Structure.Input[1].Next[2].Next.Next[0].Weight.Value,
nn.Structure.Output[0].Previous[1].Weight.Value
};
for (int i = 0; i < randomSelected.Length; i++)
{
Assert.AreNotEqual(randomSelected[i], randomSelected2[i]);
}
double[] pulsed2 = nn.Pulse(inp);
for (int i = 0; i < pulsed.Length; i++)
{
Assert.AreNotEqual(pulsed[i], pulsed2[i]);
}
}
private void button2_Click(object sender, EventArgs e)
{
int cycles = int.Parse(textBox1.Text);
int hiddenNeurons = int.Parse(textBox2.Text);
double learnRate = double.Parse(textBox3.Text);
double momentum = double.Parse(textBox4.Text);
int[] layers = new int[] { 1, hiddenNeurons, 1 };
MultilayerPerceptron ann = new MLPGenerator().Create(layers, 1, new Sigmoid(2));
ann.Reset(-1, 1);
ann.Momentum = momentum;
ann.LearnFactor = learnRate;
List <TrainingData> trainingList = new List <TrainingData>();
for (int i = 0; i < Map.Count; i++)
{
double[] input = new double[] { (double)Map[i].Item1 / (double)pictureBox1.Width };
double[] output = new double[] { (double)Map[i].Item2 / (double)pictureBox1.Height };
trainingList.Add(new TrainingData(input, output));
}
var res = ann.BP(new BPRequest(trainingList.ToArray(), cycles));
for (int i = 0; i < pictureBox1.Width; i++)
{
int x = i;
double yd = ann.Pulse(new double[] { (double)i / (double)pictureBox1.Width })[0];
int y = Convert.ToInt32(yd * pictureBox1.Height);
Point p = new Point(x, y);
Graph.DrawRectangle(new Pen(new SolidBrush(Color.Orange)), p.X, p.Y, 1, 1);
}
pictureBox1.Image = Bitmap;
}
public void BPMaxMSETest()
{
int[] layers = new int[] { 4, 5, 1 };
MultilayerPerceptron nn = new MLPGenerator().Create(layers, 1, new Sigmoid());
var result = nn.BP(new BPRequest(TestVec, 1000, 0, 0.7));
Assert.IsTrue(result.Epochs.Length < 1000);
}
public void BPMaxCyclesTest()
{
int[] layers = new int[] { 4, 5, 1 };
MultilayerPerceptron nn = new MLPGenerator().Create(layers, 1, new Sigmoid());
var result = nn.BP(new BPRequest(TestVec, 5));
Assert.AreEqual(result.Epochs.Length, 5);
}
public void ExplicitBPTimeTest()
{
int[] layers = new int[] { 4, 10, 1 };
MultilayerPerceptron nn = new MLPGenerator().Create(layers, 1, new Sigmoid());
var result = nn.BP(new BPRequest(TestVec, 1500 * 10));
throw new NotImplementedException("BP Total ms=" + result.BPTime.TotalMilliseconds.ToString());
}
public void ReasonableBPTimeTest()
{
int[] layers = new int[] { 4, 5, 1 };
MultilayerPerceptron nn = new MLPGenerator().Create(layers, 1, new Sigmoid());
var result = nn.BP(new BPRequest(TestVec, 10000));
Assert.IsTrue(result.BPTime.TotalMilliseconds < 2000);
}
public void ResetManualTest()
{
int[] layers = new int[] { 4, 5, 1 };
MultilayerPerceptron nn = new MLPGenerator().Create(layers, 1, new Sigmoid());
nn.Reset(() => 0.7);
Assert.AreEqual(nn.Structure.Input[1].Next[2].Weight.Value, 0.7); // random selected
}
public void MLPSerialization()
{
MultilayerPerceptron mlp = new MLPGenerator().Create(new int[] { 2, 2, 1 });
mlp.Reset(0, 1);
MLPXMLSerializer serializer = new MLPXMLSerializer();
XDocument xdoc = serializer.Serialize(mlp);
MultilayerPerceptron mlp2 = serializer.Deserialize(xdoc);
Assert.AreEqual(mlp2.Structure.Elements.Length, mlp.Structure.Elements.Length);
Assert.AreEqual(mlp2.Structure.Elements[3].GetDescription(), mlp.Structure.Elements[3].GetDescription());
Assert.AreEqual(mlp2.Structure.Elements[0].Next[0].Weight.Value, mlp.Structure.Elements[0].Next[0].Weight.Value);
Assert.AreEqual(((NeuronBase)mlp2.Structure.Elements[4]).Previous[0].Weight.Value, ((NeuronBase)mlp.Structure.Elements[4]).Previous[0].Weight.Value);
Assert.AreEqual(((NeuronBase)mlp2.Structure.Elements[3]).Func.GetType(), ((NeuronBase)mlp.Structure.Elements[3]).Func.GetType());
}
public void PulseManySameResultsTest()
{
MLPGenerator gen = new MLPGenerator();
int[] layers = new int[] { 4, 3, 2 };
MultilayerPerceptron nn = gen.Create(layers, 1, new Sigmoid());
double[] inp = new double[] { -0.978, 2.34, 0.2, -0.33 };
double[] res = nn.Pulse(inp);
double[] res2 = nn.Pulse(inp);
double[] res3 = nn.Pulse(inp);
Assert.AreEqual(res.Length, res2.Length);
Assert.AreEqual(res.Length, res3.Length);
for (int i = 0; i < res.Length; i++)
{
Assert.AreEqual(res[i], res2[i]);
Assert.AreEqual(res[i], res3[i]);
}
}
static void Main(string[] args)
{
TrainingData[] xor=new TrainingData[] {
new TrainingData(new double[] {1,0}, new double[] {1}),
new TrainingData(new double[] {0,1}, new double[] {1}),
new TrainingData(new double[] {1,1}, new double[] {0}),
new TrainingData(new double[] {0,0}, new double[] {0}),
};
int[] layers = new int[] { 2,2,1 };
MultilayerPerceptron mlp=new MLPGenerator().Create(layers,1,new Sigmoid());
mlp.Inspector=new SignalInspector();
//shows the neurons processing in VS Output window
mlp.Inspector.Debug=true;
mlp.Inspector.TraceLevel=3;
mlp.LearnFactor=0.9;
var response=mlp.BP(new BPRequest(xor,1000));
Console.ReadKey();
}
static void Main(string[] args)
{
TrainingData[] xor = new TrainingData[] {
new TrainingData(new double[] { 1, 0 }, new double[] { 1 }),
new TrainingData(new double[] { 0, 1 }, new double[] { 1 }),
new TrainingData(new double[] { 1, 1 }, new double[] { 0 }),
new TrainingData(new double[] { 0, 0 }, new double[] { 0 }),
};
int[] layers = new int[] { 2, 2, 1 };
MultilayerPerceptron mlp = new MLPGenerator().Create(layers, 1, new Sigmoid());
mlp.Inspector = new SignalInspector();
//shows the neurons processing in VS Output window
mlp.Inspector.Debug = true;
mlp.Inspector.TraceLevel = 3;
mlp.LearnFactor = 0.9;
var response = mlp.BP(new BPRequest(xor, 1000));
Console.ReadKey();
}
public void FullNetworkScanTest()
{
int[] layers = new int[] { 2, 3, 1 };
MultilayerPerceptron nn = new MLPGenerator().Create(layers, 1, new Sigmoid());
//check input and output layer
Assert.AreEqual(nn.Structure.Input.Length, 2);
Assert.AreEqual(nn.Structure.Output.Length, 1);
int count = 0;
foreach (var i in nn.Structure.Input)
{
foreach (var j in i.Next)
{
CheckConnection(i, j.Next, ref count);
}
}
Assert.AreEqual(count, 12);
//check bias
Assert.AreEqual(nn.Structure.Bias.Next.Length, 4);
Assert.IsTrue(nn.Structure.Bias.Next.All(x => x.Next is Neuron));
}
