using System;
using System.Drawing;
using System.Collections;
using System.ComponentModel;
using System.Windows.Forms;
using System.Data;
using SharpUtils;
using LogViewer;
using Neural_Net_Library;
using System.IO;
using System.Text;
using System.Threading;
using System.Xml.Serialization;
using System.Xml;
namespace Neural_Net_Test
{
/// <summary>
/// Summary description for Form1.
/// </summary>
public class Form1 : System.Windows.Forms.Form
{
private DebugLevel debugLevel;
private System.Windows.Forms.TabControl tabControl1;
private System.Windows.Forms.TabPage tabPage1;
private System.Windows.Forms.TabPage tabPage2;
private System.Windows.Forms.RichTextBox netWorkText;
private LogViewer.LogViewer logViewer2;
private System.Windows.Forms.MainMenu mainMenu1;
private System.Windows.Forms.MenuItem FileItem;
private Logger log;
/// threading stuff
private ThreadStart threadStart;
private Thread thread = null;
private System.Windows.Forms.MenuItem menuItem2;
private System.Windows.Forms.MenuItem menuItem3;
private System.Windows.Forms.MenuItem menuItem4;
private System.Windows.Forms.MenuItem menuItem7;
private System.Windows.Forms.MenuItem menuItem8;
/// <summary>
/// Adaline variables ( Adaline 1 )
/// </summary>
private static int nNumberOfItemsInAdalineTrainingFile = 250;
private int nNumberOfItemsInAdalineOneTrainingFile;
private double dLearningRateOfAdalineOne;
/// <summary>
/// Adaline word variables ( Adaline 2 )
/// </summary>
private static int nNumberOfItemsInAdalineWordFile = 100;
private double dLearningRateOfAdalineTwo;
private int nNumberOfWordTestsToGenerate = 250;
private bool bAdalineOneUseBias = false;
private bool bAdalineTwoUseBias = false;
/// <summary>
/// BackPropagation variables ( BackPropagation 1 )
/// </summary>
private int nNumberOfBackPropagationTests = 250;
private double dBackPropagationTolerance = 0.4;
private double dBackPropagationLearningRate = 0.45;
private double dBackPropagationMomentum = 0.9;
private bool bBackPropagationOneUseBias = false;
/// <summary>
/// BackPropagationWord Variables ( Backpropagation 2 )
/// </summary>
private bool bBackPropagationTwoUseBias = false;
private double dBackPropagationWordLearningRate = 0.45;
private double dBackPropagationWordMomentum = 0.9;
private double dBackPropagationWordTolerance = 0.4;
private string strBackPropagationWordOne = "and";
private string strBackPropagationWordTwo = "the";
private string strBackPropagationWordRunFile = "origin-of-species.TXT";
private string strBackPropagationWordTrainingFile = "originpart.txt";
/// <summary>
/// Self Organizing Network One variables
/// </summary>
private int nNumberOfItemsInSelfOrganizingNetworkTrainingFile = 300;
private bool bGenerateNewSelfOrganizingTrainingFile = true;
private bool bTrainingSelfOrganizingNetwork = true;
private double dInitialSonOneLearningRate = 0.5;
private double dFinalSonOneLearningRate = 0.01;
private int nInitialSonOneNeighborhoodSize = 5;
private int nFinalSonOneNeighborhoodSize = 5;
private int nNeighborhoodDecrementInterval = 100;
private int nNumberOfSonOneIterations = 500;
/// <summary>
/// Self Organizing Network two variables
/// </summary>
private double dInitialSonTwoLearningRate = 0.5;
private double dFinalSonTwoLearningRate = 0.01;
private int nInitialSonTwoNeighborhoodSize = 5;
private int nFinalSonTwoNeighborhoodSize = 1;
private int nSonTwoNeighborhoodDecrementInterval = 100;
private int nNumberOfSonTwoIterations = 500;
private System.Windows.Forms.MenuItem menuItem5;
private System.Windows.Forms.MenuItem menuItem9;
private System.Windows.Forms.MenuItem menuItem10;
private System.Windows.Forms.MenuItem menuItem1;
private System.Windows.Forms.MenuItem menuItem6;
private System.Windows.Forms.MenuItem menuItem11;
private System.Windows.Forms.MenuItem menuItem12;
private System.Windows.Forms.MenuItem menuItem13;
private System.Windows.Forms.MenuItem menuItem14;
private System.Windows.Forms.MenuItem menuItem15;
private System.Windows.Forms.MenuItem menuItem16;
private System.Windows.Forms.MenuItem menuItem17;
private System.Windows.Forms.MenuItem menuItem18;
private System.Windows.Forms.MenuItem menuItem19;
private System.Windows.Forms.MenuItem menuItem20;
private System.Windows.Forms.MenuItem menuItem21;
private System.Windows.Forms.MenuItem menuItem22;
private System.Windows.Forms.MenuItem menuItem23;
private System.Windows.Forms.MenuItem menuItem24;
private System.Windows.Forms.MenuItem menuItem26;
private System.Windows.Forms.MenuItem menuItem27;
private System.Windows.Forms.MenuItem menuItem28;
private System.Windows.Forms.MenuItem menuItem29;
private System.Windows.Forms.MenuItem menuItem30;
private System.Windows.Forms.MenuItem menuItem31;
private System.Windows.Forms.MenuItem menuItem32;
private System.Windows.Forms.MenuItem menuItem33;
private System.Windows.Forms.MenuItem menuItem34;
private System.Windows.Forms.MenuItem menuItem35;
private System.Windows.Forms.MenuItem menuItem36;
private System.Windows.Forms.MenuItem menuItem37;
private System.Windows.Forms.MenuItem menuItem38;
private System.Windows.Forms.MenuItem menuItem39;
/// <summary>
/// Required designer variable.
/// </summary>
private System.ComponentModel.Container components = null;
public Form1()
{
//
// Required for Windows Form Designer support
//
InitializeComponent();
//
// TODO: Add any constructor code after InitializeComponent call
//
nNumberOfItemsInAdalineOneTrainingFile = nNumberOfItemsInAdalineTrainingFile;
dLearningRateOfAdalineOne = 0.45;
dLearningRateOfAdalineTwo = 0.223333;
debugLevel = new DebugLevel( DebugLevelSet.Errors );
logViewer2 = new LogViewer.LogViewer();
this.tabPage2.Controls.AddRange(new System.Windows.Forms.Control[] {
this.logViewer2});
this.tabPage2.Location = new System.Drawing.Point(4, 22);
this.tabPage2.Name = "tabPage2";
this.tabPage2.Size = new System.Drawing.Size(592, 358);
this.tabPage2.TabIndex = 1;
this.tabPage2.Text = "Log Viewer";
//
// logViewer2
//
this.logViewer2.Anchor = (((System.Windows.Forms.AnchorStyles.Top | System.Windows.Forms.AnchorStyles.Bottom)
| System.Windows.Forms.AnchorStyles.Left)
| System.Windows.Forms.AnchorStyles.Right);
this.logViewer2.EventLog = null;
this.logViewer2.FileName = "";
this.logViewer2.Name = "logViewer2";
this.logViewer2.Size = new System.Drawing.Size(592, 360);
this.logViewer2.TabIndex = 0;
}
/// <summary>
/// Clean up any resources being used.
/// </summary>
protected override void Dispose( bool disposing )
{
if( disposing )
{
if (components != null)
{
components.Dispose();
}
}
base.Dispose( disposing );
}
#region Windows Form Designer generated code
/// <summary>
/// Required method for Designer support - do not modify
/// the contents of this method with the code editor.
/// </summary>
private void InitializeComponent()
{
System.Resources.ResourceManager resources = new System.Resources.ResourceManager(typeof(Form1));
this.tabControl1 = new System.Windows.Forms.TabControl();
this.tabPage1 = new System.Windows.Forms.TabPage();
this.netWorkText = new System.Windows.Forms.RichTextBox();
this.tabPage2 = new System.Windows.Forms.TabPage();
this.mainMenu1 = new System.Windows.Forms.MainMenu();
this.FileItem = new System.Windows.Forms.MenuItem();
this.menuItem7 = new System.Windows.Forms.MenuItem();
this.menuItem8 = new System.Windows.Forms.MenuItem();
this.menuItem2 = new System.Windows.Forms.MenuItem();
this.menuItem3 = new System.Windows.Forms.MenuItem();
this.menuItem1 = new System.Windows.Forms.MenuItem();
this.menuItem21 = new System.Windows.Forms.MenuItem();
this.menuItem22 = new System.Windows.Forms.MenuItem();
this.menuItem29 = new System.Windows.Forms.MenuItem();
this.menuItem30 = new System.Windows.Forms.MenuItem();
this.menuItem31 = new System.Windows.Forms.MenuItem();
this.menuItem32 = new System.Windows.Forms.MenuItem();
this.menuItem4 = new System.Windows.Forms.MenuItem();
this.menuItem5 = new System.Windows.Forms.MenuItem();
this.menuItem6 = new System.Windows.Forms.MenuItem();
this.menuItem19 = new System.Windows.Forms.MenuItem();
this.menuItem20 = new System.Windows.Forms.MenuItem();
this.menuItem24 = new System.Windows.Forms.MenuItem();
this.menuItem33 = new System.Windows.Forms.MenuItem();
this.menuItem34 = new System.Windows.Forms.MenuItem();
this.menuItem38 = new System.Windows.Forms.MenuItem();
this.menuItem9 = new System.Windows.Forms.MenuItem();
this.menuItem10 = new System.Windows.Forms.MenuItem();
this.menuItem11 = new System.Windows.Forms.MenuItem();
this.menuItem18 = new System.Windows.Forms.MenuItem();
this.menuItem12 = new System.Windows.Forms.MenuItem();
this.menuItem23 = new System.Windows.Forms.MenuItem();
this.menuItem27 = new System.Windows.Forms.MenuItem();
this.menuItem28 = new System.Windows.Forms.MenuItem();
this.menuItem37 = new System.Windows.Forms.MenuItem();
this.menuItem13 = new System.Windows.Forms.MenuItem();
this.menuItem14 = new System.Windows.Forms.MenuItem();
this.menuItem15 = new System.Windows.Forms.MenuItem();
this.menuItem16 = new System.Windows.Forms.MenuItem();
this.menuItem17 = new System.Windows.Forms.MenuItem();
this.menuItem26 = new System.Windows.Forms.MenuItem();
this.menuItem35 = new System.Windows.Forms.MenuItem();
this.menuItem36 = new System.Windows.Forms.MenuItem();
this.menuItem39 = new System.Windows.Forms.MenuItem();
this.tabControl1.SuspendLayout();
this.tabPage1.SuspendLayout();
this.SuspendLayout();
//
// tabControl1
//
this.tabControl1.Anchor = ((System.Windows.Forms.AnchorStyles)((((System.Windows.Forms.AnchorStyles.Top | System.Windows.Forms.AnchorStyles.Bottom)
| System.Windows.Forms.AnchorStyles.Left)
| System.Windows.Forms.AnchorStyles.Right)));
this.tabControl1.Controls.Add(this.tabPage1);
this.tabControl1.Controls.Add(this.tabPage2);
this.tabControl1.Location = new System.Drawing.Point(0, 40);
this.tabControl1.Name = "tabControl1";
this.tabControl1.SelectedIndex = 0;
this.tabControl1.Size = new System.Drawing.Size(600, 368);
this.tabControl1.TabIndex = 0;
//
// tabPage1
//
this.tabPage1.Controls.Add(this.netWorkText);
this.tabPage1.Location = new System.Drawing.Point(4, 22);
this.tabPage1.Name = "tabPage1";
this.tabPage1.Size = new System.Drawing.Size(592, 342);
this.tabPage1.TabIndex = 0;
this.tabPage1.Text = "Net Work Output";
//
// netWorkText
//
this.netWorkText.Anchor = ((System.Windows.Forms.AnchorStyles)((((System.Windows.Forms.AnchorStyles.Top | System.Windows.Forms.AnchorStyles.Bottom)
| System.Windows.Forms.AnchorStyles.Left)
| System.Windows.Forms.AnchorStyles.Right)));
this.netWorkText.Location = new System.Drawing.Point(0, 0);
this.netWorkText.Name = "netWorkText";
this.netWorkText.Size = new System.Drawing.Size(600, 344);
this.netWorkText.TabIndex = 0;
this.netWorkText.Text = "";
//
// tabPage2
//
this.tabPage2.Location = new System.Drawing.Point(4, 22);
this.tabPage2.Name = "tabPage2";
this.tabPage2.Size = new System.Drawing.Size(592, 342);
this.tabPage2.TabIndex = 1;
this.tabPage2.Text = "Log Viewer";
//
// mainMenu1
//
this.mainMenu1.MenuItems.AddRange(new System.Windows.Forms.MenuItem[] {
this.FileItem,
this.menuItem2,
this.menuItem4,
this.menuItem9,
this.menuItem13});
//
// FileItem
//
this.FileItem.Index = 0;
this.FileItem.MenuItems.AddRange(new System.Windows.Forms.MenuItem[] {
this.menuItem7,
this.menuItem8});
this.FileItem.Text = "&File";
//
// menuItem7
//
this.menuItem7.Index = 0;
this.menuItem7.Text = "Stop";
this.menuItem7.Click += new System.EventHandler(this.OnStop);
//
// menuItem8
//
this.menuItem8.Index = 1;
this.menuItem8.Text = "Exit";
this.menuItem8.Click += new System.EventHandler(this.OnExit);
//
// menuItem2
//
this.menuItem2.Index = 1;
this.menuItem2.MenuItems.AddRange(new System.Windows.Forms.MenuItem[] {
this.menuItem3,
this.menuItem1,
this.menuItem21,
this.menuItem22,
this.menuItem29,
this.menuItem30,
this.menuItem31,
this.menuItem32});
this.menuItem2.Text = "Generate";
//
// menuItem3
//
this.menuItem3.Index = 0;
this.menuItem3.Text = "Generate Adaline One Training File";
this.menuItem3.Click += new System.EventHandler(this.OnGenerateAdaline);
//
// menuItem1
//
this.menuItem1.Index = 1;
this.menuItem1.Text = "Generate Adaline Two Working File";
this.menuItem1.Click += new System.EventHandler(this.OnGenerateAdalineWord);
//
// menuItem21
//
this.menuItem21.Index = 2;
this.menuItem21.Text = "-";
//
// menuItem22
//
this.menuItem22.Index = 3;
this.menuItem22.Text = "Generate Backpropagation Working File";
this.menuItem22.Click += new System.EventHandler(this.OnGenerateBackPropagationWorkingFile);
//
// menuItem29
//
this.menuItem29.Index = 4;
this.menuItem29.Text = "-";
//
// menuItem30
//
this.menuItem30.Index = 5;
this.menuItem30.Text = "Generate Self Organizing One Training File";
this.menuItem30.Click += new System.EventHandler(this.OnGenerateSelfOrganizingTrainingFile);
//
// menuItem31
//
this.menuItem31.Index = 6;
this.menuItem31.Text = "Generate Self Organizing One Working File";
this.menuItem31.Click += new System.EventHandler(this.OnGenerateSelfOrganizingNetworkWorkingFile);
//
// menuItem32
//
this.menuItem32.Index = 7;
this.menuItem32.Text = "Generate Self Organizing One Test File";
this.menuItem32.Click += new System.EventHandler(this.OnGenerateSelfOrganizingTestFile);
//
// menuItem4
//
this.menuItem4.Index = 2;
this.menuItem4.MenuItems.AddRange(new System.Windows.Forms.MenuItem[] {
this.menuItem5,
this.menuItem6,
this.menuItem19,
this.menuItem20,
this.menuItem24,
this.menuItem33,
this.menuItem34,
this.menuItem38});
this.menuItem4.Text = "Run";
//
// menuItem5
//
this.menuItem5.Index = 0;
this.menuItem5.Text = "Load and Run Adaline 1";
this.menuItem5.Click += new System.EventHandler(this.OnLoadAndRunAdalineOne);
//
// menuItem6
//
this.menuItem6.Index = 1;
this.menuItem6.Text = "Load and Run Adaline 2";
this.menuItem6.Click += new System.EventHandler(this.OnLoadAndRunAdalineWord);
//
// menuItem19
//
this.menuItem19.Index = 2;
this.menuItem19.Text = "-";
//
// menuItem20
//
this.menuItem20.Index = 3;
this.menuItem20.Text = "Load And Run BackPropagation 1";
this.menuItem20.Click += new System.EventHandler(this.OnLoadAdnRunBackpropagationOne);
//
// menuItem24
//
this.menuItem24.Index = 4;
this.menuItem24.Text = "Load And Run BackPropagation 2";
this.menuItem24.Click += new System.EventHandler(this.OnRunBackPropagtion2);
//
// menuItem33
//
this.menuItem33.Index = 5;
this.menuItem33.Text = "-";
//
// menuItem34
//
this.menuItem34.Index = 6;
this.menuItem34.Text = "Load And Run Self Organizing Network 1";
this.menuItem34.Click += new System.EventHandler(this.OnLoadAndRunSelfOrganizingNetworkOneMenu);
//
// menuItem38
//
this.menuItem38.Index = 7;
this.menuItem38.Text = "Load And Run Self Organizing Network 2";
this.menuItem38.Click += new System.EventHandler(this.OnLoadAndRunSelfOrgainzingNetworktwoClick);
//
// menuItem9
//
this.menuItem9.Index = 3;
this.menuItem9.MenuItems.AddRange(new System.Windows.Forms.MenuItem[] {
this.menuItem10,
this.menuItem11,
this.menuItem18,
this.menuItem12,
this.menuItem23,
this.menuItem27,
this.menuItem28,
this.menuItem37});
this.menuItem9.Text = "Train";
//
// menuItem10
//
this.menuItem10.Index = 0;
this.menuItem10.Text = "Adaline 1";
this.menuItem10.Click += new System.EventHandler(this.AdalineOneClick);
//
// menuItem11
//
this.menuItem11.Index = 1;
this.menuItem11.Text = "Adaline 2";
this.menuItem11.Click += new System.EventHandler(this.OnTrainAdaline2);
//
// menuItem18
//
this.menuItem18.Index = 2;
this.menuItem18.Text = "-";
//
// menuItem12
//
this.menuItem12.Index = 3;
this.menuItem12.Text = "BackPropagtion 1";
this.menuItem12.Click += new System.EventHandler(this.OnBackPropagation1);
//
// menuItem23
//
this.menuItem23.Index = 4;
this.menuItem23.Text = "BackPropagation 2";
this.menuItem23.Click += new System.EventHandler(this.OnBackPropagation2Click);
//
// menuItem27
//
this.menuItem27.Index = 5;
this.menuItem27.Text = "-";
//
// menuItem28
//
this.menuItem28.Index = 6;
this.menuItem28.Text = "Self Organizing Network 1";
this.menuItem28.Click += new System.EventHandler(this.OnSelfOrganisingNetwork1);
//
// menuItem37
//
this.menuItem37.Index = 7;
this.menuItem37.Text = "Self Organizing Network 2";
this.menuItem37.Click += new System.EventHandler(this.OnSelfOrganizingNetworkTwoClick);
//
// menuItem13
//
this.menuItem13.Index = 4;
this.menuItem13.MenuItems.AddRange(new System.Windows.Forms.MenuItem[] {
this.menuItem14,
this.menuItem15,
this.menuItem16,
this.menuItem17,
this.menuItem26,
this.menuItem35,
this.menuItem36,
this.menuItem39});
this.menuItem13.Text = "Options";
//
// menuItem14
//
this.menuItem14.Index = 0;
this.menuItem14.Text = "Adaline 1 Options";
this.menuItem14.Click += new System.EventHandler(this.OnAdalineOneOptions);
//
// menuItem15
//
this.menuItem15.Index = 1;
this.menuItem15.Text = "Adaline 2 Options";
this.menuItem15.Click += new System.EventHandler(this.OnAdalineTwoOptions);
//
// menuItem16
//
this.menuItem16.Index = 2;
this.menuItem16.Text = "-";
//
// menuItem17
//
this.menuItem17.Index = 3;
this.menuItem17.Text = "BackPropagation 1 Options";
this.menuItem17.Click += new System.EventHandler(this.OnBackPoropagationOneOptions);
//
// menuItem26
//
this.menuItem26.Index = 4;
this.menuItem26.Text = "BackPropagation 2 Options";
this.menuItem26.Click += new System.EventHandler(this.OnBackPropagationTwoOptions);
//
// menuItem35
//
this.menuItem35.Index = 5;
this.menuItem35.Text = "-";
//
// menuItem36
//
this.menuItem36.Index = 6;
this.menuItem36.Text = "Self Organizing Network 1 Options";
this.menuItem36.Click += new System.EventHandler(this.OnSelfOrganizingNetworkOneOptions);
//
// menuItem39
//
this.menuItem39.Index = 7;
this.menuItem39.Text = "Self Orgainizing Network 2 Options";
this.menuItem39.Click += new System.EventHandler(this.OnSelfOrganizingNetworkTwoOptions);
//
// Form1
//
this.AutoScaleBaseSize = new System.Drawing.Size(5, 13);
this.ClientSize = new System.Drawing.Size(600, 406);
this.Controls.Add(this.tabControl1);
this.Icon = ((System.Drawing.Icon)(resources.GetObject("$this.Icon")));
this.Menu = this.mainMenu1;
this.Name = "Form1";
this.Text = "Neural Net Tester";
this.tabControl1.ResumeLayout(false);
this.tabPage1.ResumeLayout(false);
this.ResumeLayout(false);
}
#endregion
/// <summary>
/// The main entry point for the application.
/// </summary>
[STAThread]
static void Main()
{
try
{
Application.Run(new Form1());
}
/// catch exceptions thrown by explorer pages if not shown
catch( System.NullReferenceException )
{
}
}
/// <summary>
/// early test code
/// </summary>
/// <param name="sender"></param>
/// <param name="e"></param>
private void OnTestXMLStuff(object sender, System.EventArgs e)
{
/* log.Log( DebugLevelSet.Note, "This is a note", ClassName );
log.Log( DebugLevelSet.Important, "This is important", ClassName );
log.Log( DebugLevelSet.HighLight, "This is a highlight", ClassName );
*/
}
/*****************************************************************************************************
* ADALINE ONE STUFF
*/
/// <summary>
/// start the adaline one thread
/// </summary>
/// <param name="sender"></param>
/// <param name="e"></param>
private void AdalineOneClick(object sender, System.EventArgs e)
{
threadStart = new ThreadStart( DoAdalineOne );
thread = new Thread( threadStart );
thread.Start();
}
/// <summary>
/// load the adaline training file
/// </summary>
/// <returns></returns>
private ArrayList LoadAdalineTrainingFile()
{
ArrayList patterns = new ArrayList();
StreamReader stream = new StreamReader( "lin2var.trn" );
StringBuilder strString = new StringBuilder( "" );
StringBuilder strNumber = new StringBuilder( "" );
/// load the data from the training file
for( int i=0; i<nNumberOfItemsInAdalineOneTrainingFile; i++ )
{
AdalinePattern aPattern = new AdalinePattern( log, 2, 1 );
strString.Remove( 0, strString.Length );
strString.Append( stream.ReadLine() );
/// Look Ma No pointers
int n=0;
char ch = strString[ n ];
while( strString[ n ] != ' ' )
n++;
while( strString[ n ] == ' ' )
n++;
while( strString[ n ] != ' ' )
strNumber.Append( strString[ n++ ] );
aPattern.SetInValue( 0, Double.Parse( strNumber.ToString() ) );
strNumber.Remove( 0, strNumber.Length );
while( strString[ n ] == ' ' )
n++;
while( strString[ n ] != ' ' )
strNumber.Append( strString[ n++ ] );
aPattern.SetInValue( 1, Double.Parse( strNumber.ToString() ) );
strNumber.Remove( 0, strNumber.Length );
while( strString[ n ] == ' ' )
n++;
while( n != strString.Length && strString[ n ] != ' ' )
strNumber.Append( strString[ n++ ] );
aPattern.SetOutValue( 0, Double.Parse( strNumber.ToString() ) );
strNumber.Remove( 0, strNumber.Length );
netWorkText.AppendText( aPattern.Data() + "\n" );
patterns.Add( aPattern );
}
stream.Close();
return patterns;
}
/// <summary>
/// run the adaline 1 program
/// </summary>
private void DoAdalineOne()
{
/// delete the old log first
FileInfo info = new FileInfo( "Neural Network Tester.xml" );
if( info.Exists == true )
{
info.Delete();
}
log = new Logger( "Neural Network Tester.xml", "NeuralNetworkTester", true );
ArrayList patterns = LoadAdalineTrainingFile();
/// create the adaline network
AdalineNeuron neuron = new AdalineNeuron( log, new BasicNode( log ), new BasicNode( log ), new AdalineNode( log, dLearningRateOfAdalineOne ) );
/// train the adaline network
int nIteration = 0;
int nGood = 0;
while( nGood < nNumberOfItemsInAdalineOneTrainingFile )
{
nGood = 0;
for( int i=0; i<nNumberOfItemsInAdalineTrainingFile; i++ )
{
neuron.InputNodeOne.SetValue( Values.NodeValue, ( double )( ( Pattern )patterns[ i ] ).InSet[ 0 ] );
neuron.InputNodeTwo.SetValue( Values.NodeValue, ( double )( ( Pattern )patterns[ i ] ).InSet[ 1 ] );
if( bAdalineOneUseBias == false )
{
neuron.Node.Run( Values.NodeValue );
}
else
{
neuron.Node.Run( Values.NodeValue, neuron.Node.GetBias.GetOriginalBias() );
}
/// if the output value generated by run is not the same as the output value
/// in the training file then it is an error
if( ( ( Pattern )patterns[ i ] ).OutputValue( 0 ) != neuron.Node.GetValue( Values.NodeValue ) )
{
/// run the learn function
log.Log( DebugLevelSet.Errors, "Learn called at number " + i.ToString() + " Pattern value = " + ( ( Pattern )patterns[ i ] ).OutputValue( 0 ).ToString() + " Neuron value = " + neuron.Node.GetValue( Values.NodeValue ), "Form1" );
netWorkText.AppendText( "Learn called at number " + i.ToString() + " Pattern value = " + ( ( Pattern )patterns[ i ] ).OutputValue( 0 ).ToString() + " Neuron value = " + neuron.Node.GetValue( Values.NodeValue ) + "\n" );
neuron.Node.Learn();
break;
}
else
nGood++;
}
log.Log( DebugLevelSet.Progress, "Iteration number " + nIteration.ToString() + " produced " + nGood.ToString() + " Good values out of 250 ", "Form1" );
netWorkText.AppendText( "Iteration number " + nIteration.ToString() + " produced " + nGood.ToString() + " Good values out of 250 \n" );
nIteration++;
}
/// Default xml serialization wont work here.
/// I can't decide if this is because the classes are in a seperate dll or because they are returning
/// arraylists. an I'm not in the mood to rewrite it just to find out it's because of something else
/// so I'll write the xml serialisation for each class the old fashioned way.
/* The following will work though so I'll look into it at a later date.
TextWriter textWriter = new StreamWriter( "Network.xml" );
XmlSerializer xmlSerial = new XmlSerializer( typeof( TestSerializer ) );
TestSerializer test = new TestSerializer();
xmlSerial.Serialize( textWriter, );
*/
FileStream xmlstream = new FileStream( "adalinenetworkone.xml", FileMode.Create, FileAccess.Write, FileShare.ReadWrite, 8, true );
XmlWriter xmlWriter = new XmlTextWriter( xmlstream, System.Text.Encoding.UTF8 );
xmlWriter.WriteStartDocument();
neuron.Save( xmlWriter );
xmlWriter.WriteEndDocument();
xmlWriter.Close();
/// now load the file
FileStream readStream = new FileStream( "adalinenetworkone.xml", FileMode.Open, FileAccess.Read, FileShare.ReadWrite, 8, true );
XmlReader xmlReader = new XmlTextReader( readStream );
/// create the adaline network
AdalineNeuron neuron2 = new AdalineNeuron( log, new BasicNode( log ), new BasicNode( log ), new AdalineNode( log, dLearningRateOfAdalineOne ) );
neuron2.Load( xmlReader );
xmlReader.Close();
/// rerun the adaline should get 250 out of 250
/// could use the pattern count here but I'll be slack
for( int i=0; i<nNumberOfItemsInAdalineOneTrainingFile; i++ )
{
neuron2.InputNodeOne.SetValue( Values.NodeValue, ( double )( ( Pattern )patterns[ i ] ).InSet[ 0 ] );
neuron2.InputNodeTwo.SetValue( Values.NodeValue, ( double )( ( Pattern )patterns[ i ] ).InSet[ 1 ] );
if( bAdalineOneUseBias == false )
{
neuron2.Node.Run( Values.NodeValue );
}
else
{
neuron2.Node.Run( Values.NodeValue, neuron2.Node.GetBias.GetOriginalBias() );
}
netWorkText.AppendText( "Pattern " + i.ToString() + " Input = ( " + ( ( Pattern )patterns[ i ] ).InSet[ 0 ].ToString() + "," + ( ( Pattern )patterns[ i ] ).InSet[ 1 ].ToString() +
" ) Adaline = " + neuron2.Node.GetValue( Values.NodeValue ) + " Actual = " + ( ( Pattern )patterns[ i ] ).OutputValue( Values.NodeValue ) + "\n" );
}
log.Close();
patterns.Clear();
}
/// <summary>
/// Generate the training file for the adaline one network
/// </summary>
/// <param name="sender"></param>
/// <param name="e"></param>
private void OnGenerateAdaline(object sender, System.EventArgs e)
{
netWorkText.AppendText( "Generating Adaline File... Please Wait\n" );
Cursor.Current = Cursors.WaitCursor;
double x = 0.0;
double y = 0.0;
double dOutput = 0.0;
FileInfo info = new FileInfo( "Lin2Var.trn" );
if( info.Exists == true )
info.Delete();
FileStream stream = new FileStream( "Lin2Var.trn", FileMode.Create, FileAccess.Write );
StreamWriter writer = new StreamWriter( stream, Encoding.UTF8 );
for( int i=0; i<nNumberOfItemsInAdalineOneTrainingFile; i++ )
{
x = Values.Random( -1, 1 );
y = Values.Random( -1, 1 );
if( y<x )
dOutput = -1.0;
else
dOutput = 1.0;
writer.WriteLine( i.ToString() + " " + x.ToString() + " " + y.ToString() + " " + dOutput.ToString() );
}
writer.Close();
Cursor.Current = Cursors.Default;
netWorkText.AppendText( "Adaline File Generated\n" );
}
/// <summary>
/// stop the current running thread
/// </summary>
/// <param name="sender"></param>
/// <param name="e"></param>
private void OnStop(object sender, System.EventArgs e)
{
log.Close();
thread.Suspend();
}
/// <summary>
/// exit the application
/// </summary>
/// <param name="sender"></param>
/// <param name="e"></param>
private void OnExit(object sender, System.EventArgs e)
{
if( thread != null && thread.IsAlive == true )
thread.Suspend();
Application.Exit();
}
/// <summary>
/// start the adaline one run code
/// </summary>
/// <param name="sender"></param>
/// <param name="e"></param>
private void OnLoadAndRunAdalineOne(object sender, System.EventArgs e)
{
threadStart = new ThreadStart( LoadAndRunAdaline );
thread = new Thread( threadStart );
thread.Start();
}
/// <summary>
/// Load a saved adaline network and run the code
/// </summary>
private void LoadAndRunAdaline()
{
log = new Logger( "Neural Network Tester Load And run Adaline One.xml", "NeuralNetworkTester", true );
Cursor.Current = Cursors.WaitCursor;
OnGenerateAdaline( this, EventArgs.Empty );
Cursor.Current = Cursors.Default;
FileStream readStream = new FileStream( "adalinenetworkone.xml", FileMode.Open, FileAccess.Read, FileShare.ReadWrite, 8, true );
XmlReader xmlReader = new XmlTextReader( readStream );
ArrayList patterns = LoadAdalineTrainingFile();
/// create the adaline network
AdalineNeuron neuron = new AdalineNeuron( log, new BasicNode( log ), new BasicNode( log ), new AdalineNode( log, 0.45 ) );
neuron.Load( xmlReader );
/// could use the pattern count here but I'll be slack
for( int i=0; i<nNumberOfItemsInAdalineTrainingFile; i++ )
{
neuron.InputNodeOne.SetValue( Values.NodeValue, ( double )( ( Pattern )patterns[ i ] ).InSet[ 0 ] );
neuron.InputNodeTwo.SetValue( Values.NodeValue, ( double )( ( Pattern )patterns[ i ] ).InSet[ 1 ] );
if( bAdalineOneUseBias == false )
{
neuron.Node.Run( Values.NodeValue );
}
else
{
neuron.Node.Run( Values.NodeValue, neuron.Node.GetBias.GetOriginalBias() );
}
netWorkText.AppendText( "Pattern " + i.ToString() + " Input = ( " + ( ( Pattern )patterns[ i ] ).InSet[ 0 ].ToString() + "," + ( ( Pattern )patterns[ i ] ).InSet[ 1 ].ToString() +
" ) Adaline = " + neuron.Node.GetValue( Values.NodeValue ) + " Actual = " + ( ( Pattern )patterns[ i ] ).OutputValue( Values.NodeValue ) + "\n" );
}
log.Close();
patterns.Clear();
}
/// <summary>
/// start the training adaline 2 thread
/// </summary>
/// <param name="sender"></param>
/// <param name="e"></param>
private void OnTrainAdaline2(object sender, System.EventArgs e)
{
threadStart = new ThreadStart( OnDoAdaline2 );
thread = new Thread( threadStart );
thread.Start();
}
/// <summary>
/// load the adaline 2 training file
/// </summary>
/// <returns></returns>
private ArrayList LoadAdaline2TrainingFile()
{
ArrayList patterns = new ArrayList();
StreamReader stream = new StreamReader( "adword.trn" );
StringBuilder strString = new StringBuilder( "" );
StringBuilder strNumber = new StringBuilder( "" );
/// load the data from the training file
for( int i=0; i<nNumberOfItemsInAdalineWordFile; i++ )
{
AdalineWordPattern aPattern = new AdalineWordPattern( log, 2, 1 );
strString.Remove( 0, strString.Length );
strString.Append( stream.ReadLine() );
/// Look Ma No pointers
int n=0;
char ch = strString[ n ];
while( strString[ n ] != ' ' )
n++;
while( strString[ n ] == ' ' )
n++;
while( strString[ n ] != ' ' )
strNumber.Append( strString[ n++ ] );
aPattern.SetInValue( 0, strNumber.ToString() );
strNumber.Remove( 0, strNumber.Length );
while( strString[ n ] == ' ' )
n++;
while( strString[ n ] != ' ' )
strNumber.Append( strString[ n++ ] );
aPattern.SetInValue( 1, strNumber.ToString() );
strNumber.Remove( 0, strNumber.Length );
while( strString[ n ] == ' ' )
n++;
while( n != strString.Length && strString[ n ] != ' ' )
strNumber.Append( strString[ n++ ] );
aPattern.SetOutValue( 0, Double.Parse( strNumber.ToString() ) );
strNumber.Remove( 0, strNumber.Length );
netWorkText.AppendText( aPattern.Data() + "\n" );
patterns.Add( aPattern );
}
stream.Close();
return patterns;
}
/// <summary>
/// load the adaline 2 working file ( for the the run functionality )
/// </summary>
/// <returns></returns>
private ArrayList LoadAdaline2WorkingFile()
{
ArrayList patterns = new ArrayList();
StreamReader stream = new StreamReader( "adalinewordWorkingFile.wrk" );
StringBuilder strString = new StringBuilder( "" );
StringBuilder strNumber = new StringBuilder( "" );
/// load the data from the training file
for( int i=0; i<nNumberOfWordTestsToGenerate; i++ )
{
AdalineWordPattern aPattern = new AdalineWordPattern( log, 2, 1 );
strString.Remove( 0, strString.Length );
strString.Append( stream.ReadLine() );
/// Look Ma No pointers
int n=0;
char ch = strString[ n ];
while( strString[ n ] != ' ' )
n++;
while( strString[ n ] == ' ' )
n++;
while( strString[ n ] != ' ' )
strNumber.Append( strString[ n++ ] );
aPattern.SetInValue( 0, strNumber.ToString() );
strNumber.Remove( 0, strNumber.Length );
while( strString[ n ] == ' ' )
n++;
while( strString[ n ] != ' ' )
strNumber.Append( strString[ n++ ] );
aPattern.SetInValue( 1, strNumber.ToString() );
strNumber.Remove( 0, strNumber.Length );
while( strString[ n ] == ' ' )
n++;
while( n != strString.Length && strString[ n ] != ' ' )
strNumber.Append( strString[ n++ ] );
aPattern.SetOutValue( 0, Double.Parse( strNumber.ToString() ) );
strNumber.Remove( 0, strNumber.Length );
netWorkText.AppendText( aPattern.Data() + "\n" );
patterns.Add( aPattern );
}
stream.Close();
return patterns;
}
/// <summary>
/// Run the adaline 2 code
/// </summary>
private void OnDoAdaline2()
{
/// delete the old log first
FileInfo info = new FileInfo( "Neural Network Tester.xml" );
if( info.Exists == true )
{
info.Delete();
}
log = new Logger( "Neural Network Tester.xml", "NeuralNetworkTester", true );
ArrayList patterns = LoadAdaline2TrainingFile();
/// create the adaline network
AdalineWordNeuron neuron = new AdalineWordNeuron( log, new BasicNode( log ), new BasicNode( log ), new AdalineWordNode( log, dLearningRateOfAdalineTwo ) );
/// train the adaline network
int nIteration = 0;
int nGood = 0;
while( nGood < nNumberOfItemsInAdalineWordFile )
{
nGood = 0;
for( int i=0; i<nNumberOfItemsInAdalineWordFile; i++ )
{
netWorkText.AppendText( "Setting the Node Data to, Pattern " + i.ToString() + " word 1 = " + ( ( AdalineWordPattern )patterns[ i ] ).InputValue( 0 ) +
" value = " + ( ( AdalineWordPattern )patterns[ i ] ).GetInSetAt( 0 ).ToString() + " word 2 = " + ( ( AdalineWordPattern )patterns[ i ] ).InputValue( 1 ) +
" value = " + ( ( AdalineWordPattern )patterns[ i ] ).GetInSetAt( 1 ).ToString() + " output value = " + ( ( AdalineWordPattern )patterns[ i ] ).OutSet[ 0 ].ToString() + "\n" );
neuron.InputNodeOne.SetValue( Values.NodeValue, ( ( AdalineWordPattern )patterns[ i ] ).GetInSetAt( 0 ) );
neuron.InputNodeTwo.SetValue( Values.NodeValue, ( ( AdalineWordPattern )patterns[ i ] ).GetInSetAt( 1 ) );
if( bAdalineTwoUseBias == false )
{
neuron.Node.Run( Values.NodeValue );
}
else
{
neuron.Node.Run( Values.NodeValue, neuron.Node.GetBias.GetOriginalBias() );
}
/// if the output value generated by run is not the same as the output value
/// in the training file then it is an error
if( ( ( Pattern )patterns[ i ] ).OutputValue( 0 ) != neuron.Node.GetValue( Values.NodeValue ) )
{
/// run the learn function
log.Log( DebugLevelSet.Errors, "Learn called at number " + i.ToString() + " Pattern value = " + ( ( Pattern )patterns[ i ] ).OutputValue( 0 ).ToString() + " Neuron value = " + neuron.Node.GetValue( Values.NodeValue ), "Form1" );
netWorkText.AppendText( "Learn called at number " + i.ToString() + " Pattern value = " + ( ( Pattern )patterns[ i ] ).OutputValue( 0 ).ToString() + " Neuron value = " + neuron.Node.GetValue( Values.NodeValue ) + "\n" );
neuron.Node.Learn();
break;
}
else
nGood++;
}
log.Log( DebugLevelSet.Progress, "Iteration number " + nIteration.ToString() + " produced " + nGood.ToString() + " Good values out of " + nNumberOfItemsInAdalineWordFile.ToString(), "Form1" );
netWorkText.AppendText( "Iteration number " + nIteration.ToString() + " produced " + nGood.ToString() + " Good values out of " + nNumberOfItemsInAdalineWordFile.ToString() + "\n" );
nIteration++;
}
/// run the training sample
for( int i=0; i<nNumberOfItemsInAdalineWordFile; i++ )
{
neuron.InputNodeOne.SetValue( Values.NodeValue, ( ( AdalineWordPattern )patterns[ i ] ).GetInSetAt( 0 ) );
neuron.InputNodeTwo.SetValue( Values.NodeValue, ( ( AdalineWordPattern )patterns[ i ] ).GetInSetAt( 1 ) );
if( bAdalineTwoUseBias == false )
{
neuron.Node.Run( Values.NodeValue );
}
else
{
neuron.Node.Run( Values.NodeValue, neuron.Node.GetBias.GetOriginalBias() );
}
netWorkText.AppendText( "Pattern " + i.ToString() + " Input = ( " + ( string )( ( AdalineWordPattern )patterns[ i ] ).InSet[ 0 ] + "," + ( string )( ( AdalineWordPattern )patterns[ i ] ).InSet[ 1 ] +
" ) Adaline = " + neuron.Node.GetValue( Values.NodeValue ) + " Actual = " + ( ( AdalineWordPattern )patterns[ i ] ).OutSet[ 0 ].ToString() + "\n" );
}
FileStream xmlstream = new FileStream( "adalinewordnetwork.xml", FileMode.Create, FileAccess.Write, FileShare.ReadWrite, 8, true );
XmlWriter xmlWriter = new XmlTextWriter( xmlstream, System.Text.Encoding.UTF8 );
xmlWriter.WriteStartDocument();
neuron.Save( xmlWriter );
xmlWriter.WriteEndDocument();
xmlWriter.Close();
/// now load the file
FileStream readStream = new FileStream( "adalinewordnetwork.xml", FileMode.Open, FileAccess.Read, FileShare.ReadWrite, 8, true );
XmlReader xmlReader = new XmlTextReader( readStream );
/// create the adaline network
AdalineWordNeuron neuron2 = new AdalineWordNeuron( log, new BasicNode( log ), new BasicNode( log ), new AdalineWordNode( log ) );
neuron2.Load( xmlReader );
xmlReader.Close();
for( int i=0; i<nNumberOfItemsInAdalineWordFile; i++ )
{
neuron2.InputNodeOne.SetValue( Values.NodeValue, ( ( AdalineWordPattern )patterns[ i ] ).GetInSetAt( 0 ) );
neuron2.InputNodeTwo.SetValue( Values.NodeValue, ( ( AdalineWordPattern )patterns[ i ] ).GetInSetAt( 1 ) );
if( bAdalineTwoUseBias == false )
{
neuron2.Node.Run( Values.NodeValue );
}
else
{
neuron2.Node.Run( Values.NodeValue, neuron2.Node.GetBias.GetOriginalBias() );
}
netWorkText.AppendText( "Pattern " + i.ToString() + " Input = ( " + ( string )( ( AdalineWordPattern )patterns[ i ] ).InSet[ 0 ] + "," + ( string )( ( AdalineWordPattern )patterns[ i ] ).InSet[ 1 ] +
" ) Adaline = " + neuron2.Node.GetValue( Values.NodeValue ) + " Actual = " + ( ( AdalineWordPattern )patterns[ i ] ).OutSet[ 0 ].ToString() + "\n" );
}
log.Close();
patterns.Clear();
}
/// <summary>
/// generate an adaline word file
/// </summary>
/// <param name="sender"></param>
/// <param name="e"></param>
private void OnGenerateAdalineWord(object sender, System.EventArgs e)
{
threadStart = new ThreadStart( GenerateAdalineWord );
thread = new Thread( threadStart );
thread.Start();
}
/// <summary>
/// Generate the working file for the adaline word network
/// </summary>
/// <param name="sender"></param>
/// <param name="e"></param>
private void GenerateAdalineWord()
{
netWorkText.AppendText( "Generating Adaline Word File... Please Wait\n" );
Cursor.Current = Cursors.WaitCursor;
ArrayList arrayWords = new ArrayList();
FileInfo info = new FileInfo( "adalinewordWorkingFile.wrk" );
if( info.Exists == true )
info.Delete();
FileStream stream = new FileStream( "adalinewordWorkingFile.wrk", FileMode.Create, FileAccess.Write );
StreamWriter writer = new StreamWriter( stream, Encoding.UTF8 );
FileInfo words = new FileInfo( "AdalineWordFile.dat" );
if( words.Exists == false )
{
MessageBox.Show( "Error something has gone horribly wrong and the word file no longer exists" );
return;
}
FileStream readerStream = new FileStream( words.FullName, FileMode.Open, FileAccess.Read );
StreamReader reader = new StreamReader( readerStream );
while( reader.Peek() > -1 )
{
arrayWords.Add( reader.ReadLine().TrimEnd( ' ', '\n', '\r' ) );
}
int nNumber = 0;
double dTemp1Value = 0;
double dTemp2Value = 0;
string strTemp1;
string strTemp2;
int nDecisionValue = 0;
for( int i=0; i<nNumberOfWordTestsToGenerate; i++ )
{
nNumber = ( int )Values.Random( 0, arrayWords.Count );
strTemp1 = arrayWords[ nNumber ].ToString();
nNumber = ( int )Values.Random( 0, arrayWords.Count );
strTemp2 = arrayWords[ nNumber ].ToString();
for( int n=0; n<strTemp1.Length; n++ )
{
dTemp1Value += strTemp1[ n ] *( n+1 );
}
for( int n=0; n<strTemp2.Length; n++ )
{
dTemp2Value += strTemp2[ n ] *( n+1 );
}
dTemp1Value = dTemp1Value / 10000;
dTemp2Value = dTemp2Value / 10000;
if( dTemp1Value > dTemp2Value )
nDecisionValue = -1;
else
nDecisionValue = 1;
dTemp1Value = 0;
dTemp2Value = 0;
writer.WriteLine( i.ToString() + " " + strTemp1 + " " + strTemp2 + " " + nDecisionValue.ToString() );
}
reader.Close();
writer.Close();
Cursor.Current = Cursors.Default;
netWorkText.AppendText( "Adaline File Generated\n" );
}
/// <summary>
/// start the thread for loading and running the adaline 2 code
/// </summary>
/// <param name="sender"></param>
/// <param name="e"></param>
private void OnLoadAndRunAdalineWord(object sender, System.EventArgs e)
{
threadStart = new ThreadStart( LoadAndRunAdalineWord );
thread = new Thread( threadStart );
thread.Start();
}
/// <summary>
/// run the adaline 2 code
/// </summary>
private void LoadAndRunAdalineWord()
{
log = new Logger( "Neural Network Tester Load And run Adaline Word Network.xml", "NeuralNetworkTester", true );
Cursor.Current = Cursors.WaitCursor;
GenerateAdalineWord();
Cursor.Current = Cursors.Default;
FileStream readStream = new FileStream( "adalinewordnetwork.xml", FileMode.Open, FileAccess.Read, FileShare.ReadWrite, 8, true );
XmlReader xmlReader = new XmlTextReader( readStream );
ArrayList patterns = LoadAdaline2WorkingFile();
/// create the adaline network
AdalineWordNeuron neuron = new AdalineWordNeuron( log, new BasicNode( log ), new BasicNode( log ), new AdalineWordNode( log ) );
neuron.Load( xmlReader );
/// could use the pattern count here but I'll be slack
for( int i=0; i<nNumberOfWordTestsToGenerate; i++ )
{
neuron.InputNodeOne.SetValue( Values.NodeValue, ( ( AdalineWordPattern )patterns[ i ] ).GetInSetAt( 0 ) );
neuron.InputNodeTwo.SetValue( Values.NodeValue, ( ( AdalineWordPattern )patterns[ i ] ).GetInSetAt( 0 ) );
if( bAdalineTwoUseBias == false )
{
neuron.Node.Run( Values.NodeValue );
}
else
{
neuron.Node.Run( Values.NodeValue, neuron.Node.GetBias.GetOriginalBias() );
}
netWorkText.AppendText( "Pattern " + i.ToString() + " Input = ( " + ( ( AdalineWordPattern )patterns[ i ] ).InputValue( 0 ) + "," + ( ( AdalineWordPattern )patterns[ i ] ).InputValue( 1 ) +
" ) Adaline = " + neuron.Node.GetValue( Values.NodeValue ) + " Actual = " + ( ( AdalineWordPattern )patterns[ i ] ).OutputValue( Values.NodeValue ) + "\n" );
}
log.Close();
patterns.Clear();
}
/*******************************************************************************************************
* BACKPROPAGATION ONE STUFF
*/
/// <summary>
/// start the thread to run bp 1
/// </summary>
/// <param name="sender"></param>
/// <param name="e"></param>
private void OnBackPropagation1(object sender, System.EventArgs e)
{
threadStart = new ThreadStart( OnBackPropagationOne );
thread = new Thread( threadStart );
thread.Start();
}
/// <summary>
/// Train and run the first backpropagation example
/// </summary>
private void OnBackPropagationOne()
{
FileInfo info = new FileInfo( "Neural Network Tester.xml" );
if( info.Exists == true )
{
info.Delete();
}
log = new Logger( "Neural Network Tester.xml", "NeuralNetworkTester", true );
/// create a trainign set to learn the xor problem
ArrayList patterns = new ArrayList();
patterns.Add( new Pattern( log, 0.0, 0.0, 0.0 ) );
patterns.Add( new Pattern( log, 0.0, 1.0, 1.0 ) );
patterns.Add( new Pattern( log, 1.0, 0.0, 1.0 ) );
patterns.Add( new Pattern( log, 1.0, 1.0, 0.0 ) );
/// create a backpropagation network with a learning rate of 0.45
/// a momentum of 0.9 with 2 nodes in the input layer, 3 nodes in the middle layer
/// and one output node.
BackPropagationNetwork bpNetwork = new BackPropagationNetwork( log, dBackPropagationLearningRate, dBackPropagationMomentum, 2, 2, 1 );
bpNetwork.UseBias = bBackPropagationOneUseBias;
int nGood = 0;
double dTotalError = 0.0;
double dTest = 0.0;
while( nGood < 4 )
{
nGood = 0;
dTotalError = 0.0;
for( int i=0; i<4; i++ )
{
/// set the input values
bpNetwork.SetValue( ( ( Pattern )patterns[ i ] ) );
/// run the network
bpNetwork.Run();
/// set the desired output
bpNetwork.SetOutputError( ( ( Pattern )patterns[ i ] ) );
/// run learn anyway
bpNetwork.Learn();
dTest = ( Math.Abs( bpNetwork.GetOutputValue( 0 ) ) - ( ( Pattern )patterns[ i ] ).OutputValue( 0 ) );
netWorkText.AppendText( "Network out put value = " + ( double )bpNetwork.GetOutputValue( 0 ) + " Pattern out put value = " + ( ( Pattern )patterns[ i ] ).OutputValue( 0 ) + "\n" );
netWorkText.AppendText( "Absolute output value = " + dTest.ToString() + "\n" );
if( dTest < dBackPropagationTolerance )
{
if( debugLevel.TestDebugLevel( DebugLevelSet.Progress ) == true )
{
log.Log( DebugLevelSet.Progress, "Test falls within tolerance levels", ClassName );
}
netWorkText.AppendText( "Test = " + dTest.ToString() + " Tolerance = " + dBackPropagationTolerance.ToString() + " Test falls within tolerance levels\n" );
nGood++;
}
dTotalError += Math.Abs( bpNetwork.OutputError( 0 ) );
if( debugLevel.TestDebugLevel( DebugLevelSet.Progress ) == true )
{
log.Log( DebugLevelSet.Progress, "Back Propagation training run " + i.ToString() + " completed, total error = " + dTotalError.ToString(), ClassName );
}
netWorkText.AppendText( "Pattern " + ( ( Pattern )patterns[ i ] ).InputValue( 0 ).ToString() + "," + ( ( Pattern )patterns[ i ] ).InputValue( 1 ).ToString() +
" run through the network " + " output error at 0 = " + bpNetwork.OutputError( 0 ).ToString() +
" total error = " + dTotalError.ToString() + "\n" );
}
}
netWorkText.AppendText( "Saving the BackPropagation Network\n" );
FileStream xmlstream = new FileStream( "backpropagationnetworkone.xml", FileMode.Create, FileAccess.Write, FileShare.ReadWrite, 8, true );
XmlWriter xmlWriter = new XmlTextWriter( xmlstream, System.Text.Encoding.UTF8 );
xmlWriter.WriteStartDocument();
bpNetwork.Save( xmlWriter );
xmlWriter.WriteEndDocument();
xmlWriter.Close();
netWorkText.AppendText( "Loading the BackPropagation Network\n" );
/// now load the file
FileStream readStream = new FileStream( "backpropagationnetworkone.xml", FileMode.Open, FileAccess.Read, FileShare.ReadWrite, 8, true );
XmlReader xmlReader = new XmlTextReader( readStream );
/// create a backpropagation network with a learning rate of 0.45
/// a momentum of 0.9 with 2 nodes in the input layer, 3 nodes in the middle layer
/// and one output node.
BackPropagationNetwork bpNetwork2 = new BackPropagationNetwork( log, dBackPropagationLearningRate, dBackPropagationMomentum, 2, 2, 1 );
bpNetwork2.Load( xmlReader );
bpNetwork2.UseBias = bBackPropagationOneUseBias;
xmlReader.Close();
double dTemp = 0.0;
for( int i=0; i<4; i++ )
{
bpNetwork2.SetValue( ( Pattern )patterns[ i ] );
bpNetwork2.Run();
/// first output value is output value 0
dTemp = bpNetwork2.GetOutputValue( 0 );
netWorkText.AppendText( " Pattern Values = " + ( ( Pattern )patterns[ i ] ).InputValue( 0 ).ToString() + ", " + ( ( Pattern )patterns[ i ] ).InputValue( 1 ).ToString() +
" Pattern Output Value = " + ( ( Pattern )patterns[ i ] ).OutputValue( 0 ).ToString() + " Net work Output Value = " + bpNetwork2.GetCalculatedOutputValue( 0, ( ( Pattern )patterns[ i ] ).OutputValue( 0 ) ).ToString() + "\n" );
}
log.Close();
patterns.Clear();
}
/// <summary>
/// start the thread to generate a new file for the back propagtion network
/// </summary>
/// <param name="sender"></param>
/// <param name="e"></param>
private void OnGenerateBackPropagationWorkingFile(object sender, System.EventArgs e)
{
threadStart = new ThreadStart( GenerateBackPropagationWorkingFile );
thread = new Thread( threadStart );
thread.Start();
}
/// <summary>
/// generate a new working file for the back propagtion network
/// </summary>
private void GenerateBackPropagationWorkingFile()
{
netWorkText.AppendText( "Generating Backpropagation working File... Please Wait\n" );
Cursor.Current = Cursors.WaitCursor;
FileInfo info = new FileInfo( "BackpropagationWorkingFile.wrk" );
if( info.Exists == true )
info.Delete();
FileStream stream = new FileStream( "BackpropagationWorkingFile.wrk", FileMode.Create, FileAccess.Write );
StreamWriter writer = new StreamWriter( stream, Encoding.UTF8 );
double dTemp1Value = 0;
double dTemp2Value = 0;
StringBuilder strTemp1 = new StringBuilder();
StringBuilder strTemp2 = new StringBuilder();
bool bOne = false;
for( int i=0; i<nNumberOfBackPropagationTests; i++ )
{
bOne = false;
dTemp1Value = Values.Random( 0, 1 );
dTemp2Value = Values.Random( 0, 1 );
strTemp1.Remove( 0, strTemp1.Length );
strTemp2.Remove( 0, strTemp2.Length );
if( dTemp1Value <= 0.5 )
strTemp1.Append( "0" );
else
{
strTemp1.Append( "1" );
bOne = true;
}
if( dTemp2Value <= 0.5 )
strTemp2.Append( "0" );
else
{
strTemp2.Append( "1" );
if( bOne == false )
bOne = true;
else
bOne = false;
}
if( bOne == true )
writer.WriteLine( i.ToString() + " " + strTemp1.ToString() + " " + strTemp2.ToString() + " 1" );
else
writer.WriteLine( i.ToString() + " " + strTemp1.ToString() + " " + strTemp2.ToString() + " 0" );
}
writer.Close();
Cursor.Current = Cursors.Default;
netWorkText.AppendText( "Backpropagation Working File Generated\n" );
}
/// <summary>
/// start the thread for running the back propagation one code
/// </summary>
/// <param name="sender"></param>
/// <param name="e"></param>
private void OnLoadAdnRunBackpropagationOne(object sender, System.EventArgs e)
{
threadStart = new ThreadStart( LoadAndRunBackpropagationOne );
thread = new Thread( threadStart );
thread.Start();
}
/// <summary>
/// load a saved back propagattion network and run it
/// </summary>
private void LoadAndRunBackpropagationOne()
{
log = new Logger( "Neural Network Tester Load And Run Backpropagation One Network.xml", "NeuralNetworkTester", true );
Cursor.Current = Cursors.WaitCursor;
GenerateBackPropagationWorkingFile();
Cursor.Current = Cursors.Default;
FileStream readStream = new FileStream( "backpropagationnetworkone.xml", FileMode.Open, FileAccess.Read, FileShare.ReadWrite, 8, true );
XmlReader xmlReader = new XmlTextReader( readStream );
ArrayList patterns = LoadBackpropagationWorkingFile();
/// create a backpropagation network with a learning rate of 0.45
/// a momentum of 0.9 with 2 nodes in the input layer, 3 nodes in the middle layer
/// and one output node.
BackPropagationNetwork bpNetwork2 = new BackPropagationNetwork( log, dBackPropagationLearningRate, dBackPropagationMomentum, 2, 2, 1 );
bpNetwork2.Load( xmlReader );
bpNetwork2.UseBias = bBackPropagationOneUseBias;
xmlReader.Close();
double dTemp = 0.0;
double dCalculatedTemp = 0.0;
StringBuilder strTemp = new StringBuilder();
for( int i=0; i<nNumberOfBackPropagationTests; i++ )
{
bpNetwork2.SetValue( ( Pattern )patterns[ i ] );
bpNetwork2.Run();
/// first output value is output value 0
dTemp = bpNetwork2.GetOutputValue( 0 );
dCalculatedTemp = bpNetwork2.GetCalculatedOutputValue( 0, ( ( Pattern )patterns[ i ] ).OutputValue( 0 ) );
strTemp.Remove( 0, strTemp.Length );
if( dCalculatedTemp < dBackPropagationTolerance )
{
strTemp.Append( "Output value is within tolerance level of " + dBackPropagationTolerance.ToString() + " of 0 " );
}
else if( dCalculatedTemp > ( 1 - dBackPropagationTolerance ) )
{
strTemp.Append( "Output value is within tolerance level of " + dBackPropagationTolerance.ToString() + " of 1 " );
}
else
{
strTemp.Append( "The Network is officially retarded " );
}
netWorkText.AppendText( " Pattern Values = " + ( ( Pattern )patterns[ i ] ).InputValue( 0 ).ToString() + ", " + ( ( Pattern )patterns[ i ] ).InputValue( 1 ).ToString() +
" Pattern Output Value = " + ( ( Pattern )patterns[ i ] ).OutputValue( 0 ).ToString() + " Net work Output Value = " + dCalculatedTemp + " " + strTemp.ToString() + "\n" );
}
log.Close();
patterns.Clear();
}
/// <summary>
/// load the working or run file for the backpropagation network
/// </summary>
/// <returns></returns>
private ArrayList LoadBackpropagationWorkingFile()
{
ArrayList patterns = new ArrayList();
StreamReader stream = new StreamReader( "BackpropagationWorkingFile.wrk" );
StringBuilder strString = new StringBuilder( "" );
StringBuilder strNumber = new StringBuilder( "" );
/// load the data from the training file
for( int i=0; i<nNumberOfBackPropagationTests; i++ )
{
strString.Remove( 0, strString.Length );
strString.Append( stream.ReadLine() );
Pattern pattern = new Pattern( log, 2, 1 );
/// Look Ma No pointers
int n=0;
char ch = strString[ n ];
while( strString[ n ] != ' ' )
n++;
while( strString[ n ] == ' ' )
n++;
while( strString[ n ] != ' ' )
strNumber.Append( strString[ n++ ] );
pattern.SetInValue( 0, Double.Parse( strNumber.ToString() ) );
strNumber.Remove( 0, strNumber.Length );
while( strString[ n ] == ' ' )
n++;
while( strString[ n ] != ' ' )
strNumber.Append( strString[ n++ ] );
pattern.SetInValue( 1, Double.Parse( strNumber.ToString() ) );
strNumber.Remove( 0, strNumber.Length );
while( strString[ n ] == ' ' )
n++;
while( n != strString.Length && strString[ n ] != ' ' )
strNumber.Append( strString[ n++ ] );
pattern.SetOutValue( 0, Double.Parse( strNumber.ToString() ) );
strNumber.Remove( 0, strNumber.Length );
netWorkText.AppendText( "Setting pattern " + i.ToString() + " to, Input One = " + pattern.InputValue( 0 ).ToString() + " Input Two = " + pattern.InputValue( 1 ).ToString() + " Output Value = " + pattern.OutputValue( 0 ).ToString() + "\n" );
patterns.Add( pattern );
}
stream.Close();
return patterns;
}
/***********************************************************************************************************
* OPTIONS STUFF
*/
/// <summary>
/// Set the Adaline one options
/// </summary>
/// <param name="sender"></param>
/// <param name="e"></param>
private void OnAdalineOneOptions(object sender, System.EventArgs e)
{
AdalineOneOptions options = new AdalineOneOptions( nNumberOfItemsInAdalineOneTrainingFile, dLearningRateOfAdalineOne, bAdalineOneUseBias );
if( options.ShowDialog( this ) == DialogResult.OK )
{
nNumberOfItemsInAdalineOneTrainingFile = options.NumberOfItems;
dLearningRateOfAdalineOne = options.LearningRate;
bAdalineOneUseBias = options.UseBias;
}
}
/// <summary>
/// set the adaline two options
/// </summary>
/// <param name="sender"></param>
/// <param name="e"></param>
private void OnAdalineTwoOptions(object sender, System.EventArgs e)
{
AdalineTwoOptions options = new AdalineTwoOptions( nNumberOfItemsInAdalineWordFile, dLearningRateOfAdalineTwo, bAdalineTwoUseBias );
if( options.ShowDialog( this ) == DialogResult.OK )
{
nNumberOfItemsInAdalineWordFile = options.NumberOfWords;
dLearningRateOfAdalineTwo = options.LearningRate;
bAdalineTwoUseBias = options.UseBias;
}
}
/// <summary>
/// set the back propagation one options
/// </summary>
/// <param name="sender"></param>
/// <param name="e"></param>
private void OnBackPoropagationOneOptions(object sender, System.EventArgs e)
{
BackPropagationOneOptions options = new BackPropagationOneOptions( nNumberOfBackPropagationTests, dBackPropagationTolerance, dBackPropagationMomentum, dBackPropagationLearningRate, bBackPropagationOneUseBias );
if( options.ShowDialog( this ) == DialogResult.OK )
{
nNumberOfBackPropagationTests = options.NumberOfTests;
dBackPropagationTolerance = options.Tolerance;
dBackPropagationLearningRate = options.LearningRate;
dBackPropagationMomentum = options.Momentum;
bBackPropagationOneUseBias = options.UseBias;
}
}
/// <summary>
/// set the back propagation two options
/// </summary>
/// <param name="sender"></param>
/// <param name="e"></param>
private void OnBackPropagationTwoOptions(object sender, System.EventArgs e)
{
BackPropagationTwoOptions options = new BackPropagationTwoOptions( bBackPropagationTwoUseBias, dBackPropagationWordLearningRate, dBackPropagationWordMomentum, dBackPropagationWordTolerance, strBackPropagationWordOne, strBackPropagationWordTwo, strBackPropagationWordRunFile, strBackPropagationWordTrainingFile );
if( options.ShowDialog( this ) == DialogResult.OK )
{
bBackPropagationTwoUseBias = options.UseBias;
dBackPropagationWordLearningRate = options.LearningRate;
dBackPropagationWordMomentum = options.Momentum;
dBackPropagationWordTolerance = options.Tolerance;
strBackPropagationWordOne = options.WordOne;
strBackPropagationWordTwo = options.WordTwo;
strBackPropagationWordRunFile = options.RunFile;
strBackPropagationWordTrainingFile = options.TrainingFile;
}
}
/// <summary>
/// Set the self organizing network one options
/// </summary>
/// <param name="sender"></param>
/// <param name="e"></param>
private void OnSelfOrganizingNetworkOneOptions(object sender, System.EventArgs e)
{
SelfOrganizingNetworkOneOptions options = new SelfOrganizingNetworkOneOptions( nNumberOfItemsInSelfOrganizingNetworkTrainingFile, dInitialSonOneLearningRate, dFinalSonOneLearningRate, nInitialSonOneNeighborhoodSize, nNeighborhoodDecrementInterval, nNumberOfSonOneIterations );
if( options.ShowDialog( this ) == DialogResult.OK )
{
nNumberOfItemsInSelfOrganizingNetworkTrainingFile = options.NumberOfItems;
dInitialSonOneLearningRate = options.InitialLearningRate;
dFinalSonOneLearningRate = options.FinalLearningRate;
nInitialSonOneNeighborhoodSize = options.InitialNeighborhoodSize;
nNeighborhoodDecrementInterval = options.NeighborhoodDecrementInterval;
nNumberOfSonOneIterations = options.NumberOfIterations;
}
}
private void OnSelfOrganizingNetworkTwoOptions(object sender, System.EventArgs e)
{
SelfOrganizingNetworkTwoOptions options = new SelfOrganizingNetworkTwoOptions( dInitialSonTwoLearningRate, dFinalSonTwoLearningRate, nInitialSonTwoNeighborhoodSize, nFinalSonTwoNeighborhoodSize, nSonTwoNeighborhoodDecrementInterval, nNumberOfSonTwoIterations );
if( options.ShowDialog( this ) == DialogResult.OK )
{
dInitialSonTwoLearningRate = options.InitialLearningRate;
dFinalSonTwoLearningRate = options.FinalLearningRate;
nInitialSonTwoNeighborhoodSize = options.InitialNeighborhoodSize;
nFinalSonTwoNeighborhoodSize = options.FinalNeighborhoodSize;
nSonTwoNeighborhoodDecrementInterval = options.NeighborhoodDecrementInterval;
nNumberOfSonTwoIterations = options.NumberOfIterations;
}
}
public string ClassName
{
get
{
return this.Name;
}
}
/******************************************************************************************************
* BACK PROPAGATION 2 STUFF
*/
/// <summary>
/// start the thread for training the back propagation two network
/// </summary>
/// <param name="sender"></param>
/// <param name="e"></param>
private void OnBackPropagation2Click(object sender, System.EventArgs e)
{
threadStart = new ThreadStart( OnBackPropagation2Train );
thread = new Thread( threadStart );
thread.Start();
}
/// <summary>
/// load the back propagation two training file
/// </summary>
/// <returns></returns>
private ArrayList LoadBackPropagationWordTrainingFile()
{
netWorkText.AppendText( "Loading Back Propagation Training File please wait ... \n" );
ArrayList patterns = new ArrayList();
StreamReader stream = new StreamReader( strBackPropagationWordTrainingFile );
StringBuilder strString = new StringBuilder( "" );
StringBuilder strWord = new StringBuilder( "" );
int nCount = 0;
int n = 0;
BackPropagationWordPattern pattern = new BackPropagationWordPattern( log );
StringBuilder strTemp = new StringBuilder( " " );
bool bEndOfSentence = false;
/// load the data from the training file
while( stream.Peek() > -1 )
{
strString.Remove( 0, strString.Length );
strString.Append( stream.ReadLine() );
n=0;
while( n != strString.Length )
{
bEndOfSentence = false;
strWord.Remove( 0, strWord.Length );
while( n != strString.Length && strString[ n ] == ' ' )
n++;
while( n != strString.Length && strString[ n ] != ' ' )
{
/// end of sentence
if( strString[ n ] == '.' )
{
pattern.SetInValue( nCount, strWord.ToString() );
if( strWord.ToString().CompareTo( strBackPropagationWordOne ) == 0 )
pattern.SetOutValue( nCount, 1.0 );
else if( strWord.ToString().CompareTo( strBackPropagationWordTwo ) == 0 )
pattern.SetOutValue( nCount, 1.0 );
else
pattern.SetOutValue( nCount, 0.0 );
patterns.Add( pattern );
nCount = 0;
n++;
strTemp.Remove( 0, strTemp.Length );
/// get the full string and calculate the output value as we go
/// can't use string search functions such as indexof as this
/// will find part words
int nWordCountOne = 0;
int nWordCountTwo = 0;
for( int i=0; i<pattern.InputSize(); i++ )
{
strTemp.Append( pattern.InputValue( i ) + " " );
if( pattern.InputValue( i ).CompareTo( strBackPropagationWordOne ) == 0 )
nWordCountOne++;
if( pattern.InputValue( i ).CompareTo( strBackPropagationWordTwo ) == 0 )
nWordCountTwo++;
}
if( nWordCountOne == 1 && nWordCountTwo == 0 )
pattern.SetOutValue( 0, 1.0 );
else if( nWordCountTwo == 1 && nWordCountOne == 0 )
pattern.SetOutValue( 0, 1.0 );
else
pattern.SetOutValue( 0, 0.0 );
netWorkText.AppendText( "Setting pattern to, " + strTemp.ToString() + "\n" );
pattern = new BackPropagationWordPattern( log );
bEndOfSentence = true;
}
else /// save the word
{
strWord.Append( strString[ n++ ] );
}
}
if( bEndOfSentence == false )
{
pattern.SetInValue( nCount, strWord.ToString() );
}
nCount++;
}
}
stream.Close();
return patterns;
}
/// <summary>
/// train the back propagation two network
/// </summary>
private void OnBackPropagation2Train()
{
FileInfo info = new FileInfo( "Neural Network Tester.xml" );
if( info.Exists == true )
{
info.Delete();
}
log = new Logger( "Neural Network Tester.xml", "NeuralNetworkTester", true );
ArrayList patterns = LoadBackPropagationWordTrainingFile();
/// create a backpropagation network with a learning rate of 0.45
/// a momentum of 0.9 with 2 nodes in the input layer, 3 nodes in the middle layer
/// and one output node.
BackPropagationNetwork bpNetwork = new BackPropagationNetwork( log, dBackPropagationWordLearningRate, dBackPropagationWordMomentum, 20, 20, 1 );
bpNetwork.UseBias = bBackPropagationTwoUseBias;
double dTemp = 0.0;
double dCalculatedTemp = 0.0;
StringBuilder strTemp = new StringBuilder( " " );
StringBuilder strSentence = new StringBuilder( " " );
int nGood = 0;
int nIteration = 1;
while( nGood < patterns.Count )
{
nGood = 0;
for( int i=0; i<patterns.Count; i++ )
{
for( int n=0; n<20; n++ )
{
( ( BasicNode )bpNetwork.Nodes[ n ] ).SetValue( Values.NodeValue, ( ( BackPropagationWordPattern )patterns[ i ] ).GetInSetAt( n ) );
}
bpNetwork.Run();
/// set the desired output
bpNetwork.SetOutputError( ( ( BackPropagationWordPattern )patterns[ i ] ) );
/// run learn anyway
bpNetwork.Learn();
/// first output value is output value 0
dTemp = bpNetwork.GetOutputValue( 0 );
dCalculatedTemp = bpNetwork.GetCalculatedOutputValue( 0, ( ( BackPropagationWordPattern )patterns[ i ] ).OutputValue( 0 ) );
strTemp.Remove( 0, strTemp.Length );
netWorkText.AppendText( "Absolute output value = " + dCalculatedTemp.ToString() + "\n" );
if( dCalculatedTemp < dBackPropagationWordTolerance )
{
strTemp.Append( "Output value is within tolerance level of " + dBackPropagationWordTolerance.ToString() + " of 0 " );
nGood++;
}
else if( dCalculatedTemp > ( 1 - dBackPropagationWordTolerance ) )
{
strTemp.Append( "Output value is within tolerance level of " + dBackPropagationWordTolerance.ToString() + " of 1 " );
nGood++;
}
else
{
strTemp.Append( "Output value is outside tolerance levels\n" );
}
strTemp.Append( " " );
for( int n=0; n<( ( BackPropagationWordPattern )patterns[ i ] ).InputSize(); n++ )
{
strTemp.Append( ( ( BackPropagationWordPattern )patterns[ i ] ).InputValue( n ) + " " );
}
strTemp.Append( "\n Output Value = " + ( ( BackPropagationWordPattern )patterns[ i ] ).OutputValue( Values.NodeValue ) );
netWorkText.AppendText( "Test = " + dCalculatedTemp.ToString() + " Tolerance = " + dBackPropagationWordTolerance.ToString() + " " + strTemp.ToString() + "\n" );
}
netWorkText.AppendText( "\nNetwork iteration " + nIteration.ToString() + " produced " + nGood.ToString() + " good results out of " + patterns.Count.ToString() + "\n" );
nIteration++;
}
netWorkText.AppendText( "Saving the BackPropagation Network\n" );
FileStream xmlstream = new FileStream( "backpropagationnetworktwo.xml", FileMode.Create, FileAccess.Write, FileShare.ReadWrite, 8, true );
XmlWriter xmlWriter = new XmlTextWriter( xmlstream, System.Text.Encoding.UTF8 );
xmlWriter.WriteStartDocument();
bpNetwork.Save( xmlWriter );
xmlWriter.WriteEndDocument();
xmlWriter.Close();
netWorkText.AppendText( "\nTraining Done - Reloading Network and Testing against the training set\n" );
/// now load the file
FileStream readStream = new FileStream( "backpropagationnetworktwo.xml", FileMode.Open, FileAccess.Read, FileShare.ReadWrite, 8, true );
XmlReader xmlReader = new XmlTextReader( readStream );
/// create a backpropagation network with a learning rate of 0.45
/// a momentum of 0.9 with 2 nodes in the input layer, 3 nodes in the middle layer
/// and one output node.
netWorkText.AppendText( "Building Network\n" );
BackPropagationNetwork bpNetwork2 = new BackPropagationNetwork( log, dBackPropagationWordLearningRate, dBackPropagationWordMomentum, 20, 20, 1 );
netWorkText.AppendText( "Loading Network\n" );
bpNetwork2.Load( xmlReader );
bpNetwork2.UseBias = bBackPropagationTwoUseBias;
xmlReader.Close();
netWorkText.AppendText( "Testing Network\n" );
ArrayList bpPatterns = LoadBackPropagationWordTrainingFile();
for( int i=0; i<bpPatterns.Count; i++ )
{
for( int n=0; n<20; n++ )
{
( ( BasicNode )bpNetwork.Nodes[ n ] ).SetValue( Values.NodeValue, ( ( BackPropagationWordPattern )patterns[ i ] ).GetInSetAt( n ) );
}
bpNetwork2.Run();
/// first output value is output value 0
dTemp = bpNetwork2.GetOutputValue( 0 );
dCalculatedTemp = bpNetwork2.GetCalculatedOutputValue( 0, ( ( BackPropagationWordPattern )bpPatterns[ i ] ).OutputValue( 0 ) );
strTemp.Remove( 0, strTemp.Length );
if( dCalculatedTemp > ( 1 - dBackPropagationWordTolerance ) )
{
netWorkText.AppendText( "Absolute output value = " + dCalculatedTemp.ToString() + "\n" );
strTemp.Append( "Output value is within tolerance level of " + dBackPropagationWordTolerance.ToString() + " of 1 " );
strTemp.Append( " " );
for( int n=0; n<( ( BackPropagationWordPattern )bpPatterns[ i ] ).InputSize(); n++ )
{
strTemp.Append( ( ( BackPropagationWordPattern )bpPatterns[ i ] ).InputValue( n ) + " " );
}
strTemp.Append( "\n Output Value = " + ( ( BackPropagationWordPattern )bpPatterns[ i ] ).OutputValue( Values.NodeValue ) );
netWorkText.AppendText( "Test = " + dCalculatedTemp.ToString() + " Tolerance = " + dBackPropagationWordTolerance.ToString() + " " + strTemp.ToString() + "\n" );
}
}
netWorkText.AppendText( "Test Finished\n" );
log.Close();
patterns.Clear();
bpPatterns.Clear();
}
/// <summary>
/// load the working or the run file for the back propagtion two network
/// </summary>
/// <returns>an array of patterns</returns>
private ArrayList LoadBackPropagationWordWorkingFile()
{
netWorkText.AppendText( "Loading Back Propagation Working File please wait ... \n" );
ArrayList patterns = new ArrayList();
StreamReader stream = new StreamReader( strBackPropagationWordRunFile );
StringBuilder strString = new StringBuilder( "" );
StringBuilder strWord = new StringBuilder( "" );
int nCount = 0;
int n = 0;
BackPropagationWordPattern pattern = new BackPropagationWordPattern( log );
StringBuilder strTemp = new StringBuilder( " " );
bool bEndOfSentence = false;
/// load the data from the training file
while( stream.Peek() > -1 )
{
strString.Remove( 0, strString.Length );
strString.Append( stream.ReadLine() );
n=0;
while( n != strString.Length )
{
bEndOfSentence = false;
strWord.Remove( 0, strWord.Length );
while( n != strString.Length && strString[ n ] == ' ' )
n++;
while( n != strString.Length && strString[ n ] != ' ' )
{
/// end of sentence
if( strString[ n ] == '.' )
{
pattern.SetInValue( nCount, strWord.ToString() );
if( strWord.ToString().CompareTo( strBackPropagationWordOne ) == 0 )
pattern.SetOutValue( nCount, 1.0 );
else if( strWord.ToString().CompareTo( strBackPropagationWordOne ) == 0 )
pattern.SetOutValue( nCount, 1.0 );
else
pattern.SetOutValue( nCount, 0.0 );
patterns.Add( pattern );
nCount = 0;
n++;
strTemp.Remove( 0, strTemp.Length );
/// get the full string and calculate the output value as we go
/// can't use string search functions such as indexof as this
/// will find part words
int nWordCountOne = 0;
int nWordCountTwo = 0;
for( int i=0; i<pattern.InputSize(); i++ )
{
strTemp.Append( pattern.InputValue( i ) + " " );
if( pattern.InputValue( i ).CompareTo( strBackPropagationWordOne ) == 0 )
nWordCountOne++;
if( pattern.InputValue( i ).CompareTo( strBackPropagationWordTwo ) == 0 )
nWordCountTwo++;
}
if( nWordCountOne == 1 && nWordCountTwo == 0 )
pattern.SetOutValue( 0, 1.0 );
else if( nWordCountTwo == 1 && nWordCountOne == 0 )
pattern.SetOutValue( 0, 1.0 );
else
pattern.SetOutValue( 0, 0.0 );
netWorkText.AppendText( "Setting pattern to, " + strTemp.ToString() + "\n" );
pattern = new BackPropagationWordPattern( log );
bEndOfSentence = true;
}
else /// save the word
{
strWord.Append( strString[ n++ ] );
}
}
if( bEndOfSentence == false )
{
pattern.SetInValue( nCount, strWord.ToString() );
}
nCount++;
}
}
stream.Close();
return patterns;
}
/// <summary>
/// start the thread to run the back propagation two network
/// </summary>
/// <param name="sender"></param>
/// <param name="e"></param>
private void OnRunBackPropagtion2( object sender, System.EventArgs e )
{
threadStart = new ThreadStart( OnBackPropagation2Run );
thread = new Thread( threadStart );
thread.Start();
}
/// <summary>
/// run the back propagation two network
/// </summary>
private void OnBackPropagation2Run()
{
FileInfo info = new FileInfo( "Neural Network Tester.xml" );
if( info.Exists == true )
{
info.Delete();
}
log = new Logger( "Load And Run BackPropagation 2.xml", "NeuralNetworkTester", true );
double dTemp = 0.0;
double dCalculatedTemp = 0.0;
StringBuilder strTemp = new StringBuilder( " " );
StringBuilder strSentence = new StringBuilder( " " );
FileStream readStream = new FileStream( "backpropagationnetworktwo.xml", FileMode.Open, FileAccess.Read, FileShare.ReadWrite, 8, true );
XmlReader xmlReader = new XmlTextReader( readStream );
/// create a backpropagation network with a learning rate of 0.45
/// a momentum of 0.9 with 2 nodes in the input layer, 3 nodes in the middle layer
/// and one output node.
netWorkText.AppendText( "Building Network\n" );
BackPropagationNetwork bpNetwork2 = new BackPropagationNetwork( log, dBackPropagationWordLearningRate, dBackPropagationWordMomentum, 20, 20, 1 );
netWorkText.AppendText( "Loading Network\n" );
bpNetwork2.Load( xmlReader );
bpNetwork2.UseBias = bBackPropagationTwoUseBias;
xmlReader.Close();
netWorkText.AppendText( "Testing Network\n" );
ArrayList bpPatterns = LoadBackPropagationWordWorkingFile();
for( int i=0; i<bpPatterns.Count; i++ )
{
for( int n=0; n<20; n++ )
{
( ( BasicNode )bpNetwork2.Nodes[ n ] ).SetValue( Values.NodeValue, ( ( BackPropagationWordPattern )bpPatterns[ i ] ).GetInSetAt( n ) );
}
bpNetwork2.Run();
/// first output value is output value 0
dTemp = bpNetwork2.GetOutputValue( 0 );
dCalculatedTemp = bpNetwork2.GetCalculatedOutputValue( 0, ( ( BackPropagationWordPattern )bpPatterns[ i ] ).OutputValue( 0 ) );
strTemp.Remove( 0, strTemp.Length );
if( dCalculatedTemp > ( 1 - dBackPropagationWordTolerance ) )
{
netWorkText.AppendText( "Absolute output value = " + dCalculatedTemp.ToString() + "\n" );
strTemp.Append( "Output value is within tolerance level of " + dBackPropagationWordTolerance.ToString() + " of 1 " );
strTemp.Append( " " );
for( int n=0; n<( ( BackPropagationWordPattern )bpPatterns[ i ] ).InputSize(); n++ )
{
strTemp.Append( ( ( BackPropagationWordPattern )bpPatterns[ i ] ).InputValue( n ) + " " );
}
strTemp.Append( "\n Output Value = " + ( ( BackPropagationWordPattern )bpPatterns[ i ] ).OutputValue( Values.NodeValue ) );
netWorkText.AppendText( "Test = " + dCalculatedTemp.ToString() + " Tolerance = " + dBackPropagationWordTolerance.ToString() + " " + strTemp.ToString() + "\n" );
}
}
netWorkText.AppendText( "Test Finished\n" );
log.Close();
bpPatterns.Clear();
}
private void OnAbout(object sender, System.EventArgs e)
{
}
/*****************************************************************************************************
* SELF ORGANISING NETWORK ONE STUFF
*/
private void OnSelfOrganisingNetwork1(object sender, System.EventArgs e)
{
threadStart = new ThreadStart( OnTrainSelfOrganizingNetwork1 );
thread = new Thread( threadStart );
thread.Start();
}
/// <summary>
/// load the self organizing network training file
/// </summary>
/// <returns></returns>
private ArrayList LoadSelfOrganizingNetworkFile( string strFileName )
{
ArrayList patterns = new ArrayList();
StreamReader stream = new StreamReader( strFileName );
StringBuilder strString = new StringBuilder( "" );
StringBuilder strNumber = new StringBuilder( "" );
/// load the data from the training file
for( int i=0; i<nNumberOfItemsInSelfOrganizingNetworkTrainingFile; i++ )
{
AdalinePattern aPattern = new AdalinePattern( log, 2, 1 );
strString.Remove( 0, strString.Length );
strString.Append( stream.ReadLine() );
/// Look Ma No pointers
int n=0;
char ch = strString[ n ];
while( strString[ n ] != ' ' )
n++;
while( strString[ n ] == ' ' )
n++;
while( strString[ n ] != ' ' )
strNumber.Append( strString[ n++ ] );
aPattern.SetInValue( 0, Double.Parse( strNumber.ToString() ) );
strNumber.Remove( 0, strNumber.Length );
while( strString[ n ] == ' ' )
n++;
while( n != strString.Length && strString[ n ] != ' ' )
strNumber.Append( strString[ n++ ] );
aPattern.SetInValue( 1, Double.Parse( strNumber.ToString() ) );
strNumber.Remove( 0, strNumber.Length );
netWorkText.AppendText( aPattern.Data() + "\n" );
patterns.Add( aPattern );
}
stream.Close();
return patterns;
}
/// <summary>
/// Generate the files for the self organizing network
/// </summary>
/// <param name="sender"></param>
/// <param name="e"></param>
private void OnGenerateSelfOrganizingNetworkFileFunction( bool bGenerateTestFile )
{
netWorkText.AppendText( "Generating Self Organizing Network File... Please Wait\n" );
Cursor.Current = Cursors.WaitCursor;
double x = 0.0;
double y = 0.0;
FileInfo info;
if( bTrainingSelfOrganizingNetwork == true )
{
info = new FileInfo( "SelfOrganizingNetworkOne.trn" );
}
else if( bGenerateTestFile == true )
{
info = new FileInfo( "SelfOrganizingNetworkOneTest.tst" );
}
else
info = new FileInfo( "SelfOrganizingNetworkOne.wrk" );
if( info.Exists == true )
info.Delete();
FileStream stream;
if( bTrainingSelfOrganizingNetwork == true )
{
stream = new FileStream( "SelfOrganizingNetworkOne.trn", FileMode.Create, FileAccess.Write );
}
else if( bGenerateTestFile == true )
{
stream = new FileStream( "SelfOrganizingNetworkOneTest.tst", FileMode.Create, FileAccess.Write );
}
else
stream = new FileStream( "SelfOrganizingNetworkOne.wrk", FileMode.Create, FileAccess.Write );
StreamWriter writer = new StreamWriter( stream, Encoding.UTF8 );
if( bGenerateTestFile == true )
{
x = Values.Random( -5, 5 );
y = Values.Random( -5, 5 );
}
for( int i=0; i<nNumberOfItemsInSelfOrganizingNetworkTrainingFile; i++ )
{
if( bGenerateTestFile == false )
{
x = Values.Random( -5, 5 );
y = Values.Random( -5, 5 );
}
writer.WriteLine( i.ToString() + " " + x.ToString() + " " + y.ToString() );
}
writer.Close();
Cursor.Current = Cursors.Default;
netWorkText.AppendText( "Self Organizing Network File Generated\n" );
}
/// <summary>
/// Train the self organizing network
/// </summary>
public void OnTrainSelfOrganizingNetwork1()
{
/// delete the old log first
FileInfo info = new FileInfo( "Neural Network Tester.xml" );
if( info.Exists == true )
{
info.Delete();
}
log = new Logger( "Neural Network Tester.xml", "NeuralNetworkTester", true );
if( bGenerateNewSelfOrganizingTrainingFile == true )
OnGenerateSelfOrganizingNetworkFileFunction( false );
ArrayList patterns = LoadSelfOrganizingNetworkFile( "SelfOrganizingNetworkOne.trn" );
netWorkText.AppendText( "Creating the network\n" );
/// create the network
SelfOrganizingNetwork son = new SelfOrganizingNetwork( log, 2, 10, 10, dInitialSonOneLearningRate, dFinalSonOneLearningRate, nInitialSonOneNeighborhoodSize, nFinalSonOneNeighborhoodSize, nNeighborhoodDecrementInterval, nNumberOfSonOneIterations );
netWorkText.AppendText( "Network Created\n" );
/// train the self organizing network
int nIteration = 0;
for( nIteration=0; nIteration<nNumberOfSonOneIterations; nIteration++ )
{
for( int i=0; i<nNumberOfItemsInSelfOrganizingNetworkTrainingFile; i++ )
{
son.SetValue( ( Pattern )patterns[ i ] );
son.Run();
netWorkText.AppendText( "." );
}
son.Learn();
son.Epoch();
log.Log( DebugLevelSet.Progress, "Iteration number " + nIteration.ToString() + " produced a winning node at " + son.WinningHorizontalPos + " Horizontal and " + son.WinningVerticalPos + " vertical, winning node value = " + son.GetWinningNodeValue( son.WinningHorizontalPos, son.WinningVerticalPos ) + "\n", ClassName );
netWorkText.AppendText( "\nIteration number " + nIteration.ToString() + " produced a winning node at " + son.WinningHorizontalPos + " Horizontal and " + son.WinningVerticalPos + " vertical, winning node value = " + son.GetWinningNodeValue( son.WinningHorizontalPos, son.WinningVerticalPos ) + "\n" );
}
netWorkText.AppendText( "Saving the network" );
FileStream xmlstream = new FileStream( "selforganizingnetworkone.xml", FileMode.Create, FileAccess.Write, FileShare.ReadWrite, 8, true );
XmlWriter xmlWriter = new XmlTextWriter( xmlstream, System.Text.Encoding.UTF8 );
xmlWriter.WriteStartDocument();
son.Save( xmlWriter );
xmlWriter.WriteEndDocument();
xmlWriter.Close();
/// now load the file
FileStream readStream = new FileStream( "selforganizingnetworkone.xml", FileMode.Open, FileAccess.Read, FileShare.ReadWrite, 8, true );
XmlReader xmlReader = new XmlTextReader( readStream );
SelfOrganizingNetwork sonTest = new SelfOrganizingNetwork( log );
sonTest.Load( xmlReader );
xmlReader.Close();
netWorkText.AppendText( "Testing against the test file the following out put should be identical for the test\n" );
ArrayList testPatterns = this.LoadSelfOrganizingNetworkFile( "SelfOrganizingNetworkOneTest.tst" );
for( int i=0; i<nNumberOfItemsInSelfOrganizingNetworkTrainingFile; i++ )
{
sonTest.SetValue( ( Pattern )testPatterns[ i ] );
sonTest.Run();
netWorkText.AppendText( "Run called at " + i.ToString() + " Network Values are :- Composite Value = " + sonTest.GetPosition( Values.Composite ) + ", Horizontal Value = " + sonTest.GetPosition( Values.Row ) + ", Vertical Value = " + sonTest.GetPosition( Values.Column ) + ", Current Winning Horizontal Position = " + sonTest.WinningHorizontalPos + ", Current Winning Vertical Position " + sonTest.WinningVerticalPos + ", Inputs = " + ( ( BasicNode )sonTest.Nodes[ 0 ] ).NodeValues[ Values.NodeValue ].ToString() + "," + ( ( BasicNode )sonTest.Nodes[ 1 ] ).NodeValues[ Values.NodeValue ].ToString() + ", Winning Node Value = " + sonTest.GetWinningNodeValue( sonTest.WinningHorizontalPos, sonTest.WinningVerticalPos ) + "\n" );
}
testPatterns.Clear();
StringBuilder strDataDisplay = new StringBuilder( "" );
ArrayList arrayOutput = new ArrayList();
SelfOrganizingNetworkData data;
netWorkText.AppendText( "Completed the test ... now reprocessing the orginal data through the loaded network\n " );
for( int i=0; i<nNumberOfItemsInSelfOrganizingNetworkTrainingFile; i++ )
{
sonTest.SetValue( ( Pattern )patterns[ i ] );
sonTest.Run();
strDataDisplay.Remove( 0, strDataDisplay.Length );
strDataDisplay.Append( "Run Called at " + i.ToString() + " Network Values are :- Composite Value = " + sonTest.GetPosition( Values.Composite ) + ", Horizontal Value = " + sonTest.GetPosition( Values.Row ) + ", Vertical Value = " + sonTest.GetPosition( Values.Column ) + ", Current Winning Horizontal Position = " + sonTest.WinningHorizontalPos + ", Current Winning Vertical Position " + sonTest.WinningVerticalPos + ", Inputs = " + ( ( BasicNode )sonTest.Nodes[ 0 ] ).NodeValues[ Values.NodeValue ].ToString() + "," + ( ( BasicNode )sonTest.Nodes[ 1 ] ).NodeValues[ Values.NodeValue ].ToString() + ", Winning Node Value = " + sonTest.GetWinningNodeValue( sonTest.WinningHorizontalPos, sonTest.WinningVerticalPos ) + "\n" );
netWorkText.AppendText( strDataDisplay.ToString() );
data = new SelfOrganizingNetworkData();
data.CompositeValue = ( int )sonTest.GetPosition( Values.Composite );
data.Data = strDataDisplay.ToString();
arrayOutput.Add( data );
}
netWorkText.AppendText( "\n Completed the run through the training data ... processing the answers\n" );
int nItemCount = 0;
SelfOrganizingNetworkData dataTest;
bool bDataValid = false;
for( int i=0; i<arrayOutput.Count; i++ )
{
/// first value is always valid
if( i==0 )
{
nItemCount = 0;
data = ( SelfOrganizingNetworkData )arrayOutput[ i ];
bDataValid = true;
}
else
{
bool bFound = false;
data = ( SelfOrganizingNetworkData )arrayOutput[ i ];
for( int n=0; n<i; n++ )
{
dataTest = ( SelfOrganizingNetworkData )arrayOutput[ n ];
if( dataTest.CompositeValue == data.CompositeValue )
{
bFound = true;
n=i;
}
}
if( bFound == false )
{
nItemCount = 0;
data = ( SelfOrganizingNetworkData )arrayOutput[ i ];
bDataValid = true;
}
}
if( bDataValid == true )
{
for( int n=0; n<arrayOutput.Count; n++ )
{
dataTest = ( SelfOrganizingNetworkData )arrayOutput[ n ];
if( dataTest.CompositeValue == data.CompositeValue )
{
nItemCount++;
}
}
netWorkText.AppendText( "\n\nThere are " + nItemCount.ToString() + " items out of " + arrayOutput.Count.ToString() + " That have the Composite Value " + data.CompositeValue.ToString() + "\n" );
for( int n=0; n<arrayOutput.Count; n++ )
{
dataTest = ( SelfOrganizingNetworkData )arrayOutput[ n ];
if( dataTest.CompositeValue == data.CompositeValue )
{
netWorkText.AppendText( dataTest.Data + "\n" );
}
}
bDataValid = false;
}
}
netWorkText.AppendText( "Training for Self Organizing One network finished\n " );
log.Close();
patterns.Clear();
}
private void OnGenerateSelfOrganizingTrainingFile(object sender, System.EventArgs e)
{
bTrainingSelfOrganizingNetwork = true;
threadStart = new ThreadStart( OnGenerateSelfOrganizingNetworkFile );
thread = new Thread( threadStart );
thread.Start();
}
private void OnGenerateSelfOrganizingNetworkWorkingFile(object sender, System.EventArgs e)
{
bTrainingSelfOrganizingNetwork = false;
threadStart = new ThreadStart( OnGenerateSelfOrganizingNetworkFile );
thread = new Thread( threadStart );
thread.Start();
}
private void OnGenerateSelfOrganizingNetworkFile()
{
OnGenerateSelfOrganizingNetworkFileFunction( false );
}
private void OnGenerateSelfOrganizingTestFile(object sender, System.EventArgs e)
{
bTrainingSelfOrganizingNetwork = false;
threadStart = new ThreadStart( OnGenerateSelfOrganizingTestNetworkFile );
thread = new Thread( threadStart );
thread.Start();
}
private void OnGenerateSelfOrganizingTestNetworkFile()
{
OnGenerateSelfOrganizingNetworkFileFunction( true );
}
private void OnLoadAndRunSelfOrganizingNetworkOneMenu(object sender, System.EventArgs e)
{
bTrainingSelfOrganizingNetwork = false;
threadStart = new ThreadStart( OnLoadAndRunSelfOrganizingNetworkOne );
thread = new Thread( threadStart );
thread.Start();
}
private void OnLoadAndRunSelfOrganizingNetworkOne()
{
/// delete the old log first
FileInfo info = new FileInfo( "Neural Network Tester.xml" );
if( info.Exists == true )
{
info.Delete();
}
log = new Logger( "Load And Run Self Organizing Network One.xml", "NeuralNetworkTester", true );
if( bGenerateNewSelfOrganizingTrainingFile == true )
OnGenerateSelfOrganizingNetworkFileFunction( false );
ArrayList patterns = LoadSelfOrganizingNetworkFile( "SelfOrganizingNetworkOne.trn" );
/// now load the file
FileStream readStream = new FileStream( "selforganizingnetworkone.xml", FileMode.Open, FileAccess.Read, FileShare.ReadWrite, 8, true );
XmlReader xmlReader = new XmlTextReader( readStream );
SelfOrganizingNetwork sonTest = new SelfOrganizingNetwork( log );
sonTest.Load( xmlReader );
xmlReader.Close();
StringBuilder strDataDisplay = new StringBuilder( "" );
ArrayList arrayOutput = new ArrayList();
SelfOrganizingNetworkData data;
netWorkText.AppendText( "Completed the test ... now reprocessing the orginal data through the loaded network\n " );
for( int i=0; i<nNumberOfItemsInSelfOrganizingNetworkTrainingFile; i++ )
{
sonTest.SetValue( ( Pattern )patterns[ i ] );
sonTest.Run();
strDataDisplay.Remove( 0, strDataDisplay.Length );
strDataDisplay.Append( "Run Called at " + i.ToString() + " Network Values are :- Composite Value = " + sonTest.GetPosition( Values.Composite ) + ", Horizontal Value = " + sonTest.GetPosition( Values.Row ) + ", Vertical Value = " + sonTest.GetPosition( Values.Column ) + ", Current Winning Horizontal Position = " + sonTest.WinningHorizontalPos + ", Current Winning Vertical Position " + sonTest.WinningVerticalPos + ", Inputs = " + ( ( BasicNode )sonTest.Nodes[ 0 ] ).NodeValues[ Values.NodeValue ].ToString() + "," + ( ( BasicNode )sonTest.Nodes[ 1 ] ).NodeValues[ Values.NodeValue ].ToString() + ", Winning Node Value = " + sonTest.GetWinningNodeValue( sonTest.WinningHorizontalPos, sonTest.WinningVerticalPos ) + "\n" );
netWorkText.AppendText( strDataDisplay.ToString() );
data = new SelfOrganizingNetworkData();
data.CompositeValue = ( int )sonTest.GetPosition( Values.Composite );
data.Data = strDataDisplay.ToString();
arrayOutput.Add( data );
}
netWorkText.AppendText( "\n Completed the run through the data ... processing the answers\n" );
int nItemCount = 0;
SelfOrganizingNetworkData dataTest;
bool bDataValid = false;
for( int i=0; i<arrayOutput.Count; i++ )
{
/// first value is always valid
if( i==0 )
{
nItemCount = 0;
data = ( SelfOrganizingNetworkData )arrayOutput[ i ];
bDataValid = true;
}
else
{
bool bFound = false;
data = ( SelfOrganizingNetworkData )arrayOutput[ i ];
for( int n=0; n<i; n++ )
{
dataTest = ( SelfOrganizingNetworkData )arrayOutput[ n ];
if( dataTest.CompositeValue == data.CompositeValue )
{
bFound = true;
n=i;
}
}
if( bFound == false )
{
nItemCount = 0;
data = ( SelfOrganizingNetworkData )arrayOutput[ i ];
bDataValid = true;
}
}
if( bDataValid == true )
{
for( int n=0; n<arrayOutput.Count; n++ )
{
dataTest = ( SelfOrganizingNetworkData )arrayOutput[ n ];
if( dataTest.CompositeValue == data.CompositeValue )
{
nItemCount++;
}
}
netWorkText.AppendText( "\n\nThere are " + nItemCount.ToString() + " items out of " + arrayOutput.Count.ToString() + " That have the Composite Value " + data.CompositeValue.ToString() + "\n" );
for( int n=0; n<arrayOutput.Count; n++ )
{
dataTest = ( SelfOrganizingNetworkData )arrayOutput[ n ];
if( dataTest.CompositeValue == data.CompositeValue )
{
netWorkText.AppendText( dataTest.Data + "\n" );
}
}
bDataValid = false;
}
}
netWorkText.AppendText( "Load And Run for Self Organizing One network finished\n " );
log.Close();
patterns.Clear();
}
/////////////////////////////////////////////////////////////////////////////////////////////////
/// SELF ORGANIZING NETWORK TWO STUFF
///
private void OnSelfOrganizingNetworkTwoClick(object sender, System.EventArgs e)
{
threadStart = new ThreadStart( OnSelfOrganizingNetworkTwoTrain );
thread = new Thread( threadStart );
thread.Start();
}
private void OnSelfOrganizingNetworkTwoTrain()
{
FileInfo info = new FileInfo( "Neural Network Tester.xml" );
if( info.Exists == true )
{
info.Delete();
}
log = new Logger( "Neural Network Tester.xml", "NeuralNetworkTester", true );
/// rather than cut and paste the code and pretend its different just use earlier function
ArrayList patterns = LoadBackPropagationWordTrainingFile();
netWorkText.AppendText( "Creating Network\n" );
/// note 20 input nodes ( there are gonna be a lot of links )
SelfOrganizingNetworkWordNetwork soNetwork = new SelfOrganizingNetworkWordNetwork( log, 20, 10, 10, dInitialSonTwoLearningRate, dFinalSonTwoLearningRate, nInitialSonTwoNeighborhoodSize, nFinalSonTwoNeighborhoodSize, nSonTwoNeighborhoodDecrementInterval, nNumberOfSonTwoIterations );
netWorkText.AppendText( "Network Created\n" );
/// train the self organizing network
int nIteration = 0;
for( nIteration=0; nIteration<nNumberOfSonTwoIterations; nIteration++ )
{
for( int i=0; i<patterns.Count; i++ )
{
for( int n=0; n<20; n++ )
{
/// Note because I use the LoadBackPropagationWord training file function the pattern should be cast to an adaline word pattern which
/// contains the required function to get the values from the words.
( ( BasicNode )soNetwork.Nodes[ n ] ).SetValue( Values.NodeValue, ( ( AdalineWordPattern )patterns[ i ] ).GetInSetAt( n ));
}
soNetwork.Run();
netWorkText.AppendText( "." );
}
soNetwork.Learn();
soNetwork.Epoch();
log.Log( DebugLevelSet.Progress, "Iteration number " + nIteration.ToString() + " produced a winning node at " + soNetwork.WinningHorizontalPos + " Horizontal and " + soNetwork.WinningVerticalPos + " vertical, winning node value = " + soNetwork.GetWinningNodeValue( soNetwork.WinningHorizontalPos, soNetwork.WinningVerticalPos ) + "\n", ClassName );
netWorkText.AppendText( "\nIteration number " + nIteration.ToString() + " produced a winning node at " + soNetwork.WinningHorizontalPos + " Horizontal and " + soNetwork.WinningVerticalPos + " vertical, winning node value = " + soNetwork.GetWinningNodeValue( soNetwork.WinningHorizontalPos, soNetwork.WinningVerticalPos ) + "\n" );
}
netWorkText.AppendText( "Saving the network\n" );
FileStream xmlstream = new FileStream( "selforganizingnetworktwo.xml", FileMode.Create, FileAccess.Write, FileShare.ReadWrite, 8, true );
XmlWriter xmlWriter = new XmlTextWriter( xmlstream, System.Text.Encoding.UTF8 );
xmlWriter.WriteStartDocument();
soNetwork.Save( xmlWriter );
xmlWriter.WriteEndDocument();
xmlWriter.Close();
/// now load the file
FileStream readStream = new FileStream( "selforganizingnetworktwo.xml", FileMode.Open, FileAccess.Read, FileShare.ReadWrite, 8, true );
XmlReader xmlReader = new XmlTextReader( readStream );
netWorkText.AppendText( "Loading the network\n" );
SelfOrganizingNetworkWordNetwork sonTest = new SelfOrganizingNetworkWordNetwork( log );
sonTest.Load( xmlReader );
xmlReader.Close();
StringBuilder strDataDisplay = new StringBuilder( "" );
ArrayList arrayOutput = new ArrayList();
SelfOrganizingNetworkData data;
netWorkText.AppendText( "Completed the test ... now reprocessing the orginal data through the loaded network\n " );
for( int i=0; i<patterns.Count; i++ )
{
for( int n=0; n<20; n++ )
{
/// Note because I use the LoadBackPropagationWord training file function the pattern should be cast to an adaline word pattern which
/// contains the required function to get the values from the words.
( ( BasicNode )sonTest.Nodes[ n ] ).SetValue( Values.NodeValue, ( ( AdalineWordPattern )patterns[ i ] ).GetInSetAt( n ) );
}
sonTest.Run();
strDataDisplay.Remove( 0, strDataDisplay.Length );
strDataDisplay.Append( "Run Called at " + i.ToString() + " Network Values are :- Composite Value = " + sonTest.GetPosition( Values.Composite ) + ", Horizontal Value = " + sonTest.GetPosition( Values.Row ) + ", Vertical Value = " + sonTest.GetPosition( Values.Column ) + ", Current Winning Horizontal Position = " + sonTest.WinningHorizontalPos + ", Current Winning Vertical Position " + sonTest.WinningVerticalPos + ", Inputs = " + ( ( BasicNode )sonTest.Nodes[ 0 ] ).NodeValues[ Values.NodeValue ].ToString() + "," + ( ( BasicNode )sonTest.Nodes[ 1 ] ).NodeValues[ Values.NodeValue ].ToString() + ", Winning Node Value = " + sonTest.GetWinningNodeValue( sonTest.WinningHorizontalPos, sonTest.WinningVerticalPos ) + "\n" );
strDataDisplay.Append( " String Data :- " );
for( int n=0; n<( ( AdalineWordPattern )patterns[ i ] ).InputSize(); n++ )
{
strDataDisplay.Append( ( ( AdalineWordPattern )patterns[ i ] ).InputValue( n ) + " " );
}
netWorkText.AppendText( strDataDisplay.ToString() );
data = new SelfOrganizingNetworkData();
data.CompositeValue = ( int )sonTest.GetPosition( Values.Composite );
data.Data = strDataDisplay.ToString();
arrayOutput.Add( data );
}
netWorkText.AppendText( "\n Completed the run through the training data ... processing the answers\n" );
int nItemCount = 0;
SelfOrganizingNetworkData dataTest;
bool bDataValid = false;
for( int i=0; i<arrayOutput.Count; i++ )
{
/// first value is always valid
if( i==0 )
{
nItemCount = 0;
data = ( SelfOrganizingNetworkData )arrayOutput[ i ];
bDataValid = true;
}
else
{
bool bFound = false;
data = ( SelfOrganizingNetworkData )arrayOutput[ i ];
for( int n=0; n<i; n++ )
{
dataTest = ( SelfOrganizingNetworkData )arrayOutput[ n ];
if( dataTest.CompositeValue == data.CompositeValue )
{
bFound = true;
n=i;
}
}
if( bFound == false )
{
nItemCount = 0;
data = ( SelfOrganizingNetworkData )arrayOutput[ i ];
bDataValid = true;
}
}
if( bDataValid == true )
{
for( int n=0; n<arrayOutput.Count; n++ )
{
dataTest = ( SelfOrganizingNetworkData )arrayOutput[ n ];
if( dataTest.CompositeValue == data.CompositeValue )
{
nItemCount++;
}
}
netWorkText.AppendText( "\n\nThere are " + nItemCount.ToString() + " items out of " + arrayOutput.Count.ToString() + " That have the Composite Value " + data.CompositeValue.ToString() + "\n" );
for( int n=0; n<arrayOutput.Count; n++ )
{
dataTest = ( SelfOrganizingNetworkData )arrayOutput[ n ];
if( dataTest.CompositeValue == data.CompositeValue )
{
netWorkText.AppendText( dataTest.Data + "\n" );
}
}
bDataValid = false;
}
}
netWorkText.AppendText( "Training for Self Organizing Two network finished\n " );
netWorkText.AppendText( "Test Finished\n" );
log.Close();
patterns.Clear();
}
private void OnLoadAndRunSelfOrgainzingNetworktwoClick(object sender, System.EventArgs e)
{
threadStart = new ThreadStart( OnSelfOrganizingNetworkTwoRun );
thread = new Thread( threadStart );
thread.Start();
}
private void OnSelfOrganizingNetworkTwoRun()
{
FileInfo info = new FileInfo( "Load And Run Self Organizing Network Two.xml" );
if( info.Exists == true )
{
info.Delete();
}
log = new Logger( "Load And Run Self Organizing Network Two.xml", "NeuralNetworkTester", true );
/// rather than cut and paste the code and pretend its different just use earlier function
ArrayList patterns = LoadBackPropagationWordWorkingFile();
/// now load the file
FileStream readStream = new FileStream( "selforganizingnetworktwo.xml", FileMode.Open, FileAccess.Read, FileShare.ReadWrite, 8, true );
XmlReader xmlReader = new XmlTextReader( readStream );
SelfOrganizingNetworkWordNetwork sonTest = new SelfOrganizingNetworkWordNetwork( log );
netWorkText.AppendText( "Loading the network\n" );
sonTest.Load( xmlReader );
xmlReader.Close();
StringBuilder strDataDisplay = new StringBuilder( "" );
ArrayList arrayOutput = new ArrayList();
SelfOrganizingNetworkData data;
for( int i=0; i<patterns.Count; i++ )
{
for( int n=0; n<20; n++ )
{
/// Note because I use the LoadBackPropagationWord training file function the pattern should be cast to an adaline word pattern which
/// contains the required function to get the values from the words.
( ( BasicNode )sonTest.Nodes[ n ] ).SetValue( Values.NodeValue, ( ( AdalineWordPattern )patterns[ i ] ).GetInSetAt( n ) );
}
sonTest.Run();
strDataDisplay.Remove( 0, strDataDisplay.Length );
strDataDisplay.Append( "Run Called at " + i.ToString() + " Network Values are :- Composite Value = " + sonTest.GetPosition( Values.Composite ) + ", Horizontal Value = " + sonTest.GetPosition( Values.Row ) + ", Vertical Value = " + sonTest.GetPosition( Values.Column ) + ", Current Winning Horizontal Position = " + sonTest.WinningHorizontalPos + ", Current Winning Vertical Position " + sonTest.WinningVerticalPos + ", Inputs = " + ( ( BasicNode )sonTest.Nodes[ 0 ] ).NodeValues[ Values.NodeValue ].ToString() + "," + ( ( BasicNode )sonTest.Nodes[ 1 ] ).NodeValues[ Values.NodeValue ].ToString() + ", Winning Node Value = " + sonTest.GetWinningNodeValue( sonTest.WinningHorizontalPos, sonTest.WinningVerticalPos ) + "\n" );
strDataDisplay.Append( " String Data :- " );
for( int n=0; n<( ( AdalineWordPattern )patterns[ i ] ).InputSize(); n++ )
{
strDataDisplay.Append( ( ( AdalineWordPattern )patterns[ i ] ).InputValue( n ) + " " );
}
netWorkText.AppendText( strDataDisplay.ToString() );
data = new SelfOrganizingNetworkData();
data.CompositeValue = ( int )sonTest.GetPosition( Values.Composite );
data.Data = strDataDisplay.ToString();
arrayOutput.Add( data );
}
netWorkText.AppendText( "\n Completed the run through the data ... processing the answers\n" );
int nItemCount = 0;
SelfOrganizingNetworkData dataTest;
bool bDataValid = false;
for( int i=0; i<arrayOutput.Count; i++ )
{
/// first value is always valid
if( i==0 )
{
nItemCount = 0;
data = ( SelfOrganizingNetworkData )arrayOutput[ i ];
bDataValid = true;
}
else
{
bool bFound = false;
data = ( SelfOrganizingNetworkData )arrayOutput[ i ];
for( int n=0; n<i; n++ )
{
dataTest = ( SelfOrganizingNetworkData )arrayOutput[ n ];
if( dataTest.CompositeValue == data.CompositeValue )
{
bFound = true;
n=i;
}
}
if( bFound == false )
{
nItemCount = 0;
data = ( SelfOrganizingNetworkData )arrayOutput[ i ];
bDataValid = true;
}
}
if( bDataValid == true )
{
for( int n=0; n<arrayOutput.Count; n++ )
{
dataTest = ( SelfOrganizingNetworkData )arrayOutput[ n ];
if( dataTest.CompositeValue == data.CompositeValue )
{
nItemCount++;
}
}
netWorkText.AppendText( "\n\nThere are " + nItemCount.ToString() + " items out of " + arrayOutput.Count.ToString() + " That have the Composite Value " + data.CompositeValue.ToString() + "\n" );
for( int n=0; n<arrayOutput.Count; n++ )
{
dataTest = ( SelfOrganizingNetworkData )arrayOutput[ n ];
if( dataTest.CompositeValue == data.CompositeValue )
{
netWorkText.AppendText( dataTest.Data + "\n" );
}
}
bDataValid = false;
}
}
netWorkText.AppendText( "Load And Run for Self Organizing Two network finished\n " );
netWorkText.AppendText( "Test Finished\n" );
log.Close();
patterns.Clear();
}
}
/// <summary>
/// structure to help storing and structuring the self organizing network output
/// </summary>
public struct SelfOrganizingNetworkData
{
private int nCompositeValue;
private string strData;
public int CompositeValue
{
get
{
return nCompositeValue;
}
set
{
nCompositeValue = value;
}
}
public string Data
{
get
{
return strData;
}
set
{
strData = value;
}
}
}
public class TestSerializer
{
public int nOne = 1;
}
}
Submit an article or tip