public LinearLeastSquares(Matrix a, Matrix b, Method method)
{
if (a == null || b == null)
throw new ArgumentNullException();
if (a.Rows != b.Rows)
throw new ArgumentException();
if (b.Columns != 1)
throw new ArgumentException("b must be a vector!");
this.n = a.Columns;
if (method == Method.MGS)
leastSquare = new MGS(a);
else if (method == Method.Householder)
leastSquare = new Householder(a);
leastSquare.Orthogonalization();
Console.WriteLine("Q");
Console.WriteLine(leastSquare.GetQ().ToString(4));
Console.WriteLine("R");
Console.WriteLine(leastSquare.GetR().ToString(4));
this.b = b;
this.x = new Matrix(a.Columns, 1);
}
public static void Main()
{
var time = new double[] { 1, 2, 3, 4, 6, 9, 10, 13, 15 };
var activity = new double[] { 117, 100, 88, 72, 53, 29.5, 25.2, 15.2, 11.1 };
vector timeVec = new vector(time.Length);
vector activityVec = new vector(activity.Length);
vector dyVec = new vector(activity.Length);
var fs = new Func <double, double>[] { x => 1, x => x };
for (int i = 0; i < activity.Length; i++)
{
timeVec[i] = time[i];
activityVec[i] = Math.Log(activity[i]);
dyVec[i] = activity[i] / 20;
}
var fitter = new LeastSquare();
fitter.fit(fs, timeVec, activityVec, dyVec);
var ExpData = new Data(time, activity);
File.WriteAllLines("ExpData.txt", ExpData.X.Select((x, i) => $"{x} {ExpData.Y[i]} {dyVec[i]}"));
File.WriteAllLines("Fit.txt", ExpData.X.Select((x, i) => $"{x} {func(x, Math.Exp(fitter.c[0]), fitter.c[1])}"));
for (int i = 0; i < fitter.c.size; i++)
{
System.Console.WriteLine(fitter.c[i]);
}
fitter.S.print();
System.Console.WriteLine("The half-life is : " + $"{-Math.Log(2)/fitter.c[1]}+-{Math.Abs(-Math.Log(2)/Math.Sqrt(fitter.S[1,1]))}");
}
public void Start()
{
if (m_ResetHighscore)
{
PlayerPrefs.DeleteAll();
}
Helper.ResetRandomNumber(m_EvolutionaryRandomSeed);
m_Population = PopulationRandomInitialize();
m_ResetCoroutine = ResetGame(false);
StartCoroutine(m_ResetCoroutine);
m_lSquare = new LeastSquare();
}
private static void Main(string[] args)
{
var path2 = "/home/wyk/RiderProjects/HuberLossWithGrad/GradTest/hxjydq20200203.csv";
//var b = new HuberLossSolver(args[0]);
var b = new LeastSquare(args[0]);
//var theta = b.VectorHuberLossSolver();
b.LeastSquareSolver();
//var r = Outliers.find(args[0], theta);
/*foreach (var t in r)
* {
* Console.WriteLine(t);
* }*/
//寻找离群点之后，可以将该点去除后进行再一次的回归
}
