public double[] Train()
{
var startingThetas = new double[this.featuresCount + 1]; // add one column for intercept
var o = new QuasiNewtonOptimizer();
return(o.Minimize(this.CostFunction, startingThetas));
}
private Array generateEval(int rennAbteilungsAnzahl)
{
Random rtmp = new Random();
double[] digit = new double[rennAbteilungsAnzahl];
for (int i = digit.Length - 1; i >= 0; i--)
{
digit[i] = rtmp.Next(1, maxValue);
}
Extreme.Mathematics.Vector <double> initialGuess = Extreme.Mathematics.Vector.Create(digit);
Func <Extreme.Mathematics.Vector <double>, double> f = evalFunctionfromVector;
// Which method is used, is specified by a constructor
// parameter of type QuasiNewtonMethod:
var bfgs = new QuasiNewtonOptimizer(QuasiNewtonMethod.Bfgs);
bfgs.InitialGuess = initialGuess;
bfgs.ExtremumType = ExtremumType.Minimum;
// Set the ObjectiveFunction:
bfgs.ObjectiveFunction = f;
// The FindExtremum method does all the hard work:
bfgs.FindExtremum();
Console.WriteLine("BFGS Method:");
Console.WriteLine(" Solution: {0}", bfgs.Extremum);
Console.WriteLine(" Estimated error: {0}", bfgs.EstimatedError);
Console.WriteLine(" # iterations: {0}", bfgs.IterationsNeeded);
// Optimizers return the number of function evaluations
// and the number of gradient evaluations needed:
Console.WriteLine(" # function evaluations: {0}", bfgs.EvaluationsNeeded);
Console.WriteLine(" # gradient evaluations: {0}", bfgs.GradientEvaluationsNeeded);
return(digit);
}
