private double GetR(IIrtRegression regression, bool includeOutliers)
{
var r = IrtRegression.R(regression);
if (!double.IsNaN(r))
{
return(r);
}
var pointSet = !includeOutliers?RegularPoints.ToArray() : RegularPoints.Concat(OutlierPoints).ToArray();
var statsX = new Statistics(pointSet.Select(point => point.X));
var statsY = new Statistics(pointSet.Select(point => point.Y));
return(statsY.R(statsX));
}
public void getAllPoints()
{
if (!Initialized)
{
return;
}
List <float> x = getVectorByName(TS_anomalyFlight, Selected);
List <float> y = getVectorByName(TS_anomalyFlight, Correlated[Selected]);
Tuple <string, string> p = new Tuple <string, string>(Selected, Correlated[Selected]);
float x_min = x.Min();
float x_max = x.Max();
if (currentLinePlaying % 300 == 0 && currentLinePlaying != 0)
{
this.regularPoint.Clear();
this.AnomalyPoints.Clear();
for (int i = currentLinePlaying - 300; i < currentLinePlaying; i++)
{
KeyValuePair <float, float> point = new KeyValuePair <float, float>(x[i], y[i]);
if (Anomalys.ContainsKey(p) && Anomalys[p].Contains(i))
{
AnomalyPoints.Add(point);
}
else
{
RegularPoints.Add(point);
}
}
}
// add linear reg line:
if (!Switched)
{
linearReg.Clear();
if (lin_reg_eq.ContainsKey(p))
{
Tuple <float, float> current = lin_reg_eq[p];
float a = current.Item1;
float b = current.Item2;
float y_min = a * x_min + b;
float y_max = a * x_max + b;
linearReg.Add(new KeyValuePair <float, float>(x_min, y_min));
linearReg.Add(new KeyValuePair <float, float>(x_max, y_max));
}
}
}
