/// <summary>
/// <para>
/// Adds the contents of another <see cref="xpidea.neuro.net.patterns.PatternsCollection" /> to the end of the
/// collection.
/// </para>
/// </summary>
/// <param name="value">
/// A <see cref="xpidea.neuro.net.patterns.PatternsCollection" /> containing the objects to add to the collection.
/// </param>
/// <seealso cref="xpidea.neuro.net.patterns.PatternsCollection.Add" />
public void AddRange(PatternsCollection value)
{
for (var i = 0; (i < value.Count); i = (i + 1))
{
Add(value[i]);
}
}
public static void BuildEvenCircle(PatternsCollection aData)
{
BuildEvenSquare(aData);
for (var i = aData.Count - 1; i >= 0; i--)
{
var x = aData[i].Input[0];
var y = aData[i].Input[1];
if ((x*x + y*y) > 0.8)
aData.RemoveAt(i);
}
}
public PatternsCollection CreateTrainingPatterns(Font font)
{
var result = new PatternsCollection(aCharsCount, aMatrixDim*aMatrixDim, aCharsCount);
for (var i = 0; i < aCharsCount; i++)
{
var aBitMatrix = CharToBitArray(Convert.ToChar(aFirstChar + i), font, aMatrixDim, 0);
for (var j = 0; j < aMatrixDim*aMatrixDim; j++)
result[i].Input[j] = aBitMatrix[j];
result[i].Output[i] = 1;
}
return result;
}
public static void BuildEvenSquare(PatternsCollection aData)
{
var c = (int) Math.Round(Math.Sqrt(aData.Count));
var s = 2/(double) c;
double x = -1;
double y = -1;
for (var i = 0; i < c; i++)
{
for (var j = 0; j < c; j++)
{
aData[i*c + j].Input[0] = x;
aData[i*c + j].Input[1] = y;
y = y + s;
}
y = -1;
x = x + s;
}
}
public override void Train(PatternsCollection patterns)
{
int iteration = 0;
if (patterns != null)
{
double error = 0;
int good = 0;
while (good < patterns.Count) // Train until all patterns are correct
{
if (Form1.IsTerminated) return;
error = 0;
owner.progressBar1.Value = good;
owner.label16.Text = "Training progress: " + ((good * 100)/owner.progressBar1.Maximum).ToString() + "%";
good = 0;
for (int i = 0; i<patterns.Count; i++)
{
for (int k = 0; k<NodesInLayer(0); k++)
nodes[k].Value = patterns[i].Input[k];
AddNoiseToInputPattern(owner.trackBar3.Value);
this.Run();
for (int k = 0;k< this.OutputNodesCount;k++)
{
error += Math.Abs(this.OutputNode(k).Error);
this.OutputNode(k).Error = patterns[i].Output[k];
}
this.Learn();
if (BestNodeIndex == OutputPatternIndex(patterns[i]))
good++;
iteration ++;
Application.DoEvents();
}
foreach (NeuroLink link in links) ((EpochBackPropagationLink)link).Epoch(patterns.Count);
if ((iteration%2) == 0)
owner.label17.Text = "AVG Error: " + (error / OutputNodesCount).ToString() + " Iteration: " + iteration.ToString();
}
owner.label17.Text = "AVG Error: " + (error / OutputNodesCount).ToString() + " Iteration: " + iteration.ToString();
}
}
private static void Main(string[] args)
{
BidirectionalAssociativeMemorySystem BAMSystem;
var patterns = new PatternsCollection(4, 4, 4);
SetPattern(patterns[0], -1, 1, -1, 1, 1, -1, 1, -1); //invert
SetPattern(patterns[1], -1, -1, 1, 1, 1, 1, -1, -1);
SetPattern(patterns[2], 1, 1, -1, -1, -1, -1, 1, 1);
SetPattern(patterns[3], -1, -1, -1, -1, 1, 1, 1, 1);
BAMSystem = new BidirectionalAssociativeMemorySystem(4, 4);
BAMSystem.Train(patterns);
BAMSystem.SaveToFile("test.net");
//We didn't exposed following pattens to BAM, but we'd like to see what BAM will produce based on
//previous experiance....
//inputs //expected values, not shown to BAM
SetPattern(patterns[0], 1, 1, 1, 1, -1, -1, -1, -1);
SetPattern(patterns[1], 1, 1, 1, -1, -1, -1, -1, 1);
SetPattern(patterns[2], 1, -1, 1, -1, -1, 1, -1, 1);
SetPattern(patterns[3], -1, 1, -1, 1, 1, -1, 1, -1);
Console.Out.WriteLine("Input pattern: BAM output: Expected output: ");
foreach (var p in patterns)
{
BAMSystem.SetValuesFromPattern(p);
BAMSystem.Run();
foreach (var d in p.Input)
Console.Out.Write(d + ",");
Console.Out.Write(" ");
for (var i = 0; i < BAMSystem.OutputNodesCount; i++)
Console.Out.Write(BAMSystem.OutputNode(i).Value + ",");
Console.Out.Write(" ");
foreach (var d in p.Output)
Console.Out.Write(d + ",");
Console.Out.WriteLine(" ");
}
Console.In.ReadLine();
}
private void SetPattern(PatternsCollection aPatterns, int i, double x, double y)
{
aPatterns[i - 1].Input[0] = x;
aPatterns[i - 1].Input[1] = y;
}
/// <summary>
/// </summary>
/// <param name="mappings"></param>
public CustomPatternEnumerator(PatternsCollection mappings)
{
temp = mappings;
baseEnumerator = temp.GetEnumerator();
}
/// <summary>
/// </summary>
/// <param name="mappings"></param>
public CustomPatternEnumerator(PatternsCollection mappings)
{
temp = mappings;
baseEnumerator = temp.GetEnumerator();
}
/// <summary>
/// <para>
/// Initializes a new instance of <see cref="xpidea.neuro.net.patterns.PatternsCollection" /> based on another
/// <see cref="xpidea.neuro.net.patterns.PatternsCollection" />.
/// </para>
/// </summary>
/// <param name='value'>
/// A <see cref="xpidea.neuro.net.patterns.PatternsCollection" /> from which the contents are copied
/// </param>
public PatternsCollection(PatternsCollection value)
{
AddRange(value);
}
public override void Train(PatternsCollection patterns)
{
var iteration = 0;
if (patterns != null)
{
double error = 0;
var good = 0;
while (good < patterns.Count) // Train until all patterns are correct
{
if (IsTerminated) return;
error = 0;
owner.progressBar1.Value = good;
owner.label16.Text = "Training progress: " + ((good*100)/owner.progressBar1.Maximum) + "%";
good = 0;
for (var i = 0; i < patterns.Count; i++)
{
for (var k = 0; k < NodesInLayer(0); k++)
nodes[k].Value = patterns[i].Input[k];
AddNoiseToInputPattern(owner.trackBar3.Value);
Run();
var idx = (int) patterns[i].Output[0];
for (var k = 0; k < OutputNodesCount; k++)
{
error += Math.Abs(OutputNode(k).Error);
if (k == idx)
OutputNode(k).Error = 1;
else
OutputNode(k).Error = 0;
}
Learn();
if (BestNodeIndex == idx)
good++;
iteration ++;
Application.DoEvents();
}
foreach (var link in links) ((EpochBackPropagationLink) link).Epoch(patterns.Count);
if ((iteration%2) == 0)
owner.label17.Text = "AVG Error: " + (error/OutputNodesCount) + " Iteration: " + iteration;
}
owner.label17.Text = "AVG Error: " + (error/OutputNodesCount) + " Iteration: " + iteration;
}
}
/// <summary>
/// <para>
/// Adds the contents of another <see cref="xpidea.neuro.net.patterns.PatternsCollection" /> to the end of the
/// collection.
/// </para>
/// </summary>
/// <param name="value">
/// A <see cref="xpidea.neuro.net.patterns.PatternsCollection" /> containing the objects to add to the collection.
/// </param>
/// <seealso cref="xpidea.neuro.net.patterns.PatternsCollection.Add" />
public void AddRange(PatternsCollection value)
{
for (var i = 0; (i < value.Count); i = (i + 1))
{
Add(value[i]);
}
}
public override void Train(PatternsCollection patterns)
{
this.patterns = patterns;
if (patterns != null)
for (var i = 0; i < trainingIterations; i++)
{
for (var j = 0; j < patterns.Count; j++)
{
SetValuesFromPattern(patterns[j]);
Run();
Learn();
}
Epoch(0);
if (PatternStyles.IsTerminated)
return;
if ((i%3) == 0)
ShowMeTheMovie();
}
ShowMeTheMovie();
}
/// <summary>
/// Overridden.Trains the network (makes the network learn the patterns).
/// </summary>
/// <param name="patterns">Training patterns.</param>
public override void Train(PatternsCollection patterns)
{
//This method implementation is for reference only -
//You may want to implement your own method by overriding this one.
if (patterns != null)
{
var good = 0;
var tolerance = 0.2;
while (good < patterns.Count) // Train until all patterns are correct
{
good = 0;
for (var i = 0; i < patterns.Count; i++)
{
for (var k = 0; k < NodesInLayer(0); k++) nodes[k].Value = (patterns[i]).Input[k];
for (var j = firstMiddleNode; j < NodesCount; j++) nodes[j].Run();
for (var k = firstOutputNode; k < NodesCount; k++)
nodes[k].Error = (patterns[i]).Output[k - firstOutputNode];
for (var j = NodesCount - 1; j >= firstMiddleNode; j--)
nodes[j].Learn();
var InRange = true;
for (var k = 0; k < OutputNodesCount; k++)
{
if (Math.Abs(OutputNode(k).Value - (patterns[i]).Output[k]) >= tolerance) InRange = false;
//Console.Out.WriteLine(this.OutputNode(k).Value.ToString()+" " +this.OutputNode(k).Error.ToString());
}
if (InRange) good++;
}
Epoch(patterns.Count);
}
}
}
public override void Train(PatternsCollection patterns)
{
//This method implementation is for reference only -
//You may want to implement your own method by overriding this one.
var iteration = 0;
if (patterns != null)
{
double error = 0;
var good = 0;
while (good < patterns.Count) // Train until all patterns are correct
{
if (IsTerminated) return;
error = 0;
owner.progressBar1.Value = good;
owner.label16.Text = "Training progress: " + ((good*100)/owner.progressBar1.Maximum) + "%";
good = 0;
for (var i = 0; i < patterns.Count; i++)
{
for (var k = 0; k < NodesInLayer(0); k++)
nodes[k].Value = patterns[i].Input[k];
AddNoiseToInputPattern(owner.trackBar3.Value);
Run();
for (var k = 0; k < OutputNodesCount; k++)
{
error += Math.Abs(OutputNode(k).Error);
OutputNode(k).Error = patterns[i].Output[k];
}
Learn();
if (BestNodeIndex == OutputPatternIndex(patterns[i]))
good++;
iteration ++;
Application.DoEvents();
}
if ((iteration%2) == 0)
owner.label17.Text = "AVG Error: " + (error/OutputNodesCount) + " Iteration: " + iteration;
}
owner.label17.Text = "AVG Error: " + (error/OutputNodesCount) + " Iteration: " + iteration;
}
}
/// <summary>
/// Overridden.Performs network training. Here you write the code to train your network.
/// </summary>
/// <param name="patterns">Set of the patterns that will be exposed to a network during the training.</param>
/// <remarks>
/// This method implementation is for reference only -
/// You may want to implement your own method by overriding this one.
/// This implementation will
/// complete network training only after the network will produce
/// correct output for all input patterns.
/// Be advised that in this example network training will never complete if input patterns
/// have non-linear character.
/// </remarks>
public override void Train(PatternsCollection patterns)
{
int Good, i;
if (patterns != null)
{
Good = 0;
while (Good < patterns.Count)
{
Good = 0;
for (i = 0; i < patterns.Count; i++)
{
SetValuesFromPattern(patterns[i]);
AdalineNode.Run();
if ((patterns[i]).Output[0] != AdalineNode.Value)
{
AdalineNode.Learn();
break;
}
Good++;
}
}
}
}
private static void Main(string[] args)
{
Console.Out.WriteLine(" BACKPROPAGATION neural network demo.");
Console.Out.WriteLine("");
Console.Out.WriteLine(" Copyright(C) XP Idea.com 2001-2004 ");
Console.Out.WriteLine("");
Console.Out.WriteLine("The purpose of this demo is to show learning abilities of BACKPROP network.");
Console.Out.WriteLine("The BACKPROP network is able to learn much more complex data patterns, than");
Console.Out.WriteLine("Adaline network (please see OCR demo application). ");
Console.Out.WriteLine("This example simple shows that the Backprop network is able to learn ");
Console.Out.WriteLine("an 'exclusive OR' (XOR) operation, but the Adaline network is not able to do so.");
Console.Out.WriteLine("");
Console.Out.WriteLine(" false XOR false = false");
Console.Out.WriteLine(" true XOR false = true");
Console.Out.WriteLine(" false XOR true = true");
Console.Out.WriteLine(" true XOR true = false");
Console.Out.WriteLine("");
Console.Out.WriteLine(" As result of the training, the network will produce output ‘0’");
Console.Out.WriteLine("corresponding to logical ‘false’ or ‘1’ corresponding to logical ‘true’ value.");
Console.Out.WriteLine("");
Console.Out.WriteLine("PLEASE HIT ENTER TO CONTINUE");
Console.In.ReadLine();
Console.Out.WriteLine("");
Console.Out.WriteLine("During this demo you will be prompted to enter input values");
Console.Out.WriteLine("for the network. Then network will perform “XOR” operation on ");
Console.Out.WriteLine("the entered values and result will be displayed to you. ");
Console.Out.WriteLine("Please enter any values in range from 0 to 1 and hit [ENTER] when prompted. ");
Console.Out.WriteLine("");
Console.Out.WriteLine("NOW THE NETWORK IS READY TO LEARN FOLLOWING PATTERNS");
Console.Out.WriteLine("");
Console.Out.WriteLine(" false XOR false = false;");
Console.Out.WriteLine(" true XOR false = true;");
Console.Out.WriteLine(" false XOR true = true;");
Console.Out.WriteLine(" true XOR true = false;");
Console.Out.WriteLine("PLEASE HIT ENTER TO BEGIN TRAINING");
Console.In.ReadLine();
Console.Out.Write("TRAINING....");
double d;
BackPropagationNetwork BackPropNet;
var patterns = new PatternsCollection(TrainingSets, 2, 1);
SetPattern(patterns[0], 0, 0, 0);
SetPattern(patterns[1], 0, 1, 1);
SetPattern(patterns[2], 1, 0, 1);
SetPattern(patterns[3], 1, 1, 0);
//Network(0.55,0.6,
BackPropNet = new BackPropagationNetwork(0.55, 0.6, new int[3] {2, 3, 1});
BackPropNet.Train(patterns);
Console.Out.WriteLine("DONE!");
Console.Out.WriteLine("");
//BackPropNet.SaveToFile("test.net");
while (true)
{
try
{
Console.Out.Write("Enter 1st value: ");
d = double.Parse(Console.In.ReadLine());
BackPropNet.InputNode(0).Value = d;
Console.Out.Write("Enter 2nd value: ");
d = double.Parse(Console.In.ReadLine());
BackPropNet.InputNode(1).Value = d;
BackPropNet.Run();
Console.Out.WriteLine("Result: " + Math.Round(BackPropNet.OutputNode(0).Value));
Console.Out.WriteLine("");
}
catch
{
return;
}
}
}
/// <summary>
/// Overridden.Trains the network.
/// </summary>
/// <param name="patterns"></param>
public override void Train(PatternsCollection patterns)
{
if (patterns != null)
for (var i = 0; i < trainingIterations; i++)
{
for (var j = 0; j < patterns.Count; j++)
{
SetValuesFromPattern(patterns[j]);
Run();
Learn();
}
Epoch(0);
}
}
/// <summary>
/// Overridden.Trains the network to recognize specific patterns. Employs
/// <see cref="xpidea.neuro.net.bam.BidirectionalAssociativeMemoryOutputNode.Run" /> and
/// <see cref="xpidea.neuro.net.bam.BidirectionalAssociativeMemoryOutputNode.Learn" />
/// to teach the network.
/// </summary>
/// <param name="patterns">Training patterns.</param>
public override void Train(PatternsCollection patterns)
{
if (patterns != null)
for (var i = 0; i < patterns.Count; i++)
{
SetValuesFromPattern(patterns[i]);
Learn();
}
}
private static void Main(string[] args)
{
Console.Out.WriteLine(" ADALINE neural network demo.");
Console.Out.WriteLine("");
Console.Out.WriteLine(" Copyright(C) XP Idea.com 2001-2004 ");
Console.Out.WriteLine("");
Console.Out.WriteLine("The purpose of this demo is to show classification ability of ADALINE network.");
Console.Out.WriteLine("The ADALINE network is able to classify LINEARY SEPARATABLE data into two");
Console.Out.WriteLine("categories. Such data could be presented by linear function in form of");
Console.Out.WriteLine("");
Console.Out.WriteLine(" y=a*x+b;");
Console.Out.WriteLine("");
Console.Out.WriteLine(" or any other similar function, including logical “OR” operation:");
Console.Out.WriteLine("");
Console.Out.WriteLine(" false OR false = false;");
Console.Out.WriteLine(" true OR false = true;");
Console.Out.WriteLine(" false OR true = true;");
Console.Out.WriteLine(" true OR true = true;");
Console.Out.WriteLine("");
Console.Out.WriteLine("This example will show how the Adaline network can perform logical “OR” ");
Console.Out.WriteLine("operation. As result of this operation, the network will produce output ‘-1’");
Console.Out.WriteLine("corresponding to logical ‘false’ or ‘1’ corresponding to logical true ‘true’. ");
Console.Out.WriteLine("");
Console.Out.WriteLine("PLEASE HIT ENTER TO CONTINUE");
Console.In.ReadLine();
Console.Out.WriteLine("");
Console.Out.WriteLine("During this demo you will be prompted to enter input values");
Console.Out.WriteLine("for the network. Then network will perform “OR” operation on ");
Console.Out.WriteLine("the entered values and result will be displayed to you. ");
Console.Out.WriteLine("Please enter any values in range from 0 to 1 and hit [ENTER] when prompted. ");
Console.Out.WriteLine("");
Console.Out.WriteLine("NOW THE NETWORK IS READY TO LEARN FOLLOWING PATTERNS");
Console.Out.WriteLine("");
Console.Out.WriteLine(" false OR false = false;");
Console.Out.WriteLine(" true OR false = true;");
Console.Out.WriteLine(" false OR true = true;");
Console.Out.WriteLine(" true OR true = true;");
Console.Out.WriteLine("PLEASE HIT ENTER TO BEGIN TRAINING");
Console.In.ReadLine();
Console.Out.Write("TRAINING....");
double d;
AdalineNetwork AdalineNet;
var patterns = new PatternsCollection(TrainingSets, 2, 1);
SetPattern(patterns[0], 0, 0, -1);
SetPattern(patterns[1], 0, 1, 1);
SetPattern(patterns[2], 1, 0, 1);
SetPattern(patterns[3], 1, 1, 1);
AdalineNet = new AdalineNetwork(2, 0.1);
AdalineNet.Train(patterns);
Console.Out.WriteLine("DONE!");
Console.Out.WriteLine("");
// Console.Out.WriteLine("Saving the network to the file 'test.net'");
// AdalineNet.SaveToFile("test.net");
Console.Out.WriteLine("");
//AdalineNet = new AdalineNetwork( "test.net");
while (true)
{
try
{
Console.Out.Write("Enter 1st value: ");
d = double.Parse(Console.In.ReadLine());
AdalineNet.InputNode(0).Value = d;
Console.Out.Write("Enter 2nd value: ");
d = double.Parse(Console.In.ReadLine());
AdalineNet.InputNode(1).Value = d;
AdalineNet.AdalineNode.Run();
Console.Out.WriteLine("Result: " + AdalineNet.AdalineNode.Value);
Console.Out.WriteLine("");
}
catch
{
return;
}
}
}
private void button1_Click(object sender, EventArgs e)
{
{
var data = new PatternsCollection(PatternsCount, InputsCount, OutputsCount);
var checkedItem = GetCheckedIndex();
if (checkedItem > 6)
switch (checkedItem)
{
case 7:
{
PatternStyles.BuildEvenSquare(data);
break;
}
case 8:
{
PatternStyles.BuildEvenCircle(data);
break;
}
}
else
{
for (var i = 1; i <= PatternsCount; i++)
{
double x = 0;
double y = 0;
switch (checkedItem)
{
case 0:
{
PatternStyles.BuildSquare(out x, out y);
break;
}
case 1:
{
PatternStyles.BuildDimond(out x, out y);
break;
}
case 2:
{
PatternStyles.BuildPlus(out x, out y);
break;
}
case 3:
{
PatternStyles.BuildRing(out x, out y);
break;
}
case 4:
{
PatternStyles.BuildCircle(out x, out y);
break;
}
case 5:
{
PatternStyles.BuildTwoCircles(out x, out y);
break;
}
case 6:
{
PatternStyles.ChessBoard(out x, out y);
break;
}
default:
throw new ApplicationException("Please select pattern styly!");
}
SetPattern(data, i, x, y);
}
}
var LearningRateStart = 0.7;
var LearningRateEnds = 0.05;
var InitialNeighborhoodSize = 5;
var NeighborhoodDecrRate = 5;
var Iterations = 2000;
var a = new MySelfOrganizingNetwork(InputsCount, MapSize, MapSize, LearningRateStart, LearningRateEnds,
InitialNeighborhoodSize, Iterations/NeighborhoodDecrRate, Iterations);
a.control = panel1;
a.Train(data);
data = null;
a = null;
}
}
/// <summary>
/// Overridden.Performs network training. Here you write the code to train your network.
/// </summary>
/// <param name="patterns">Set of the patterns that will be exposed to a network during the training.</param>
public override void Train(PatternsCollection patterns)
{
//This method implementation is for reference only -
//You may want to implement your own method by overriding this one.
if (patterns != null)
{
var good = 0;
double tolerance = 0.2F;
while (good < patterns.Count) // Train until all patterns are correct
{
good = 0;
for (var i = 0; i < patterns.Count; i++)
{
for (var k = 0; k < NodesInLayer(0); k++)
nodes[k].Value = patterns[i].Input[k];
Run();
for (var k = 0; k < OutputNodesCount; k++)
OutputNode(k).Error = patterns[i].Output[k];
Learn();
var InRange = true;
for (var k = 0; k < OutputNodesCount; k++)
{
if (Math.Abs(OutputNode(k).Value - patterns[i].Output[k]) >= tolerance) InRange = false;
// Console.Out.WriteLine(this.OutputNode(k).Value.ToString()+" " +this.OutputNode(k).Error.ToString());
// InRange = Math.Round(nodes[k].Value) == Math.Round((patterns[i]).Output[k - firstOutputNode]);
}
if (InRange)
good++;
}
}
}
}
/// <summary>
/// <para>
/// Initializes a new instance of <see cref="xpidea.neuro.net.patterns.PatternsCollection" /> based on another
/// <see cref="xpidea.neuro.net.patterns.PatternsCollection" />.
/// </para>
/// </summary>
/// <param name='value'>
/// A <see cref="xpidea.neuro.net.patterns.PatternsCollection" /> from which the contents are copied
/// </param>
public PatternsCollection(PatternsCollection value)
{
AddRange(value);
}
private void button1_Click(object sender, System.EventArgs e)
{
trainingPatterns = CreateTrainingPatterns(label5.Font);
tabControl1.SelectedTab = tabPage2;
}
private void button1_Click(object sender, EventArgs e)
{
aFirstChar = textBox1.Text[0];
aLastChar = textBox2.Text[0];
aCharsCount = aLastChar - aFirstChar + 1;
label5.Text = "";
var chCnt = aCharsCount;
if (aCharsCount > 50)
chCnt = 50;
for (var i = 0; i < chCnt; i++)
label5.Text += Convert.ToChar(aFirstChar + i) + " ";
trainingPatterns = CreateTrainingPatterns(label5.Font);
tabControl1.SelectedTab = tabPage2;
}
/// <summary>
/// Performs network training. Here you write the code to train your network.
/// </summary>
/// <param name="patterns">Set of the patterns that will be exposed to a network during the training.</param>
/// <remarks>
/// <p>
/// There are several major paradigms, or approaches, to neural network learning. These include
/// <i>supervised, unsupervised</i>, and <i>reinforcement</i> learning. How the training data is processed is a
/// major aspect of these learning paradigms.
/// </p>
/// <p>
/// <i>Supervised</i> learning is the most common form of learning and is sometimes called programming by example.
/// The neural network is trained by showing it examples of the problem state or attributes along with the desired
/// output or action. The neural network makes a prediction based on the inputs and if the output differs from
/// the desired out put, then the network is adjusted or adapted to produce the correct output. This process is
/// repeated over and over until the agent learns to make accurate classifications or predictions. Historical data
/// from databases, sensor logs, or trace logs is often used as the training or example data.
/// </p>
/// <p>
/// <i>Unsupervised</i> learning is used when the neural network needs to recognize similarities between inputs or
/// to identify features in the input data. The data is presented to the network, and it adapts so that it
/// partitions the data into groups. The clustering or segmenting process continues until the neural network places
/// the
/// same data into the same group on successive passes over the data. An unsupervised learning algorithm performs a
/// type of feature detection where important common attributes in the data are extracted. The Kohonen map will be
/// a good example of the network using unsupervised learning.
/// </p>
/// <p>
/// <i>Reinforcement</i> learning is a type of supervised learning used when explicit input/ output pairs of
/// training data are not available. It can be used in cases where there is a sequence of inputs arid the desired
/// output is only known after the specific sequence occurs. This process of identifying the relationship between a
/// series of input values and a later output value is called temporal credit assignment. Because we provide less
/// specific error information, reinforcement learning usually takes longer than supervised learning and is less
/// efficient. However, in many situations, having exact prior information about the desired outcome is not
/// possible. In many ways,
/// reinforcement learning is the most realistic form of learning.
/// </p>
/// </remarks>
public virtual void Train(PatternsCollection patterns)
{
}
public void DrawPatterns(PatternsCollection pat, Graphics g)
{
foreach (var ptn in pat)
{
var pen = new Pen(Color.Black, 4);
var pt = new Point(p(ptn.Input[0]), p(ptn.Input[1]));
var pt2 = pt;
pt2.Offset(2, 2);
g.DrawLine(pen, pt, pt2);
pen.Dispose();
}
}