public NeuralRobot(BasicNetwork network, bool track, Position source, Position destination)
{
_hStats = new NormalizedField(NormalizationAction.Normalize, "Heading", 359, 0, .9, -.9);
_CanGoStats = new NormalizedField(NormalizationAction.Normalize, "CanGo", 1, 0, 0.9, -0.9);
_track = track;
_network = network;
sim = new RobotSimulator(source, destination);
}
/// <summary>
/// Construct the training generator.
/// </summary>
/// <param name="thePath">The path to use.</param>
public GenerateTraining(string thePath)
{
_path = thePath;
_trainingFile = FileUtil.CombinePath(new FileInfo(_path), Config.FilenameTrain);
_fieldDifference = new NormalizedField(NormalizationAction.Normalize, "diff", Config.DiffRange,
-Config.DiffRange, 1, -1);
_fieldOutcome = new NormalizedField(NormalizationAction.Normalize, "out", Config.PipRange, -Config.PipRange,
1, -1);
}
public NeuralPilot(BasicNetwork network, bool track)
{
_fuelStats = new NormalizedField(NormalizationAction.Normalize, "fuel", 200, 0, -0.9, 0.9);
_altitudeStats = new NormalizedField(NormalizationAction.Normalize, "altitude", 10000, 0, -0.9, 0.9);
_velocityStats = new NormalizedField(NormalizationAction.Normalize, "velocity",
LanderSimulator.TerminalVelocity, -LanderSimulator.TerminalVelocity,
-0.9, 0.9);
_track = track;
_network = network;
}
public void Execute(IExampleInterface app)
{
// Normalize values with an actual range of (0 to 100) to (-1 to 1)
var norm = new NormalizedField(NormalizationAction.Normalize,
null, 100, 0, 1, -1);
double x = 5;
double y = norm.Normalize(x);
Console.WriteLine(x + @" normalized is " + y);
double z = norm.DeNormalize(y);
Console.WriteLine(y + @" denormalized is " + z);
}
public double[] Process(double[] inputArray)
{
double num;
double[] numArray;
int num2;
int num3;
this._x24d1ebc88ca906aa = new NormalizedField();
goto Label_0097;
Label_0010:
if (num2 >= inputArray.Length)
{
goto Label_0085;
}
Label_0023:
numArray[num2] = this._x24d1ebc88ca906aa.Normalize(inputArray[num2]);
num2++;
if ((((uint) num2) - ((uint) num3)) <= uint.MaxValue)
{
goto Label_0010;
}
Label_0085:
if (((uint) num3) <= uint.MaxValue)
{
return numArray;
}
Label_0097:
this._x24d1ebc88ca906aa.NormalizedHigh = this._x891507b50bbab0f9;
if ((((uint) num) - ((uint) num)) > uint.MaxValue)
{
goto Label_0023;
}
this._x24d1ebc88ca906aa.NormalizedLow = this._x136bfff0efb12047;
double[] numArray2 = inputArray;
num3 = 0;
while (true)
{
if (num3 >= numArray2.Length)
{
numArray = new double[inputArray.Length];
num2 = 0;
goto Label_0010;
}
num = numArray2[num3];
this._x24d1ebc88ca906aa.Analyze(num);
num3++;
}
}
/// <summary>
/// Construct the indicator.
/// </summary>
/// <param name="theMethod">The machine learning method to use.</param>
/// <param name="thePath">The path to use.</param>
public MyInd(IMLRegression theMethod, string thePath)
: base(theMethod != null)
{
_method = theMethod;
_path = thePath;
RequestData("CLOSE[1]");
RequestData("SMA(10)[" + Config.InputWindow + "]");
RequestData("SMA(25)[" + Config.InputWindow + "]");
_fieldDifference = new NormalizedField(NormalizationAction.Normalize, "diff", Config.DiffRange,
-Config.DiffRange, 1, -1);
_fieldOutcome = new NormalizedField(NormalizationAction.Normalize, "out", Config.PipRange, -Config.PipRange,
1, -1);
}