public override List <ChartData> OnPlot(ReportNode rp)
{
PlotType = this.Name;
List <ChartData> cds = new List <ChartData>();
ChartData cd = ChartData.GetInstance();
List <double> values = rp.GetPlot(this.Name);
//Set bin count
histCount = values.Count / 100;
if (histCount < 10)
{
histCount = 10;
}
//Find Min and Max
double max = double.MinValue;
double min = double.MaxValue;
foreach (double d in values)
{
if (d > max)
{
max = d;
}
if (d < min)
{
min = d;
}
}
if (min == max)
{
return(cds);
}
;
//Place values into bins
double[] vals = new double[histCount + 1];
double step = (max - min) / ((double)histCount);
double vmax = double.MinValue;
foreach (double d in values)
{
int v = ((int)Math.Round((d - min) / step));
vals[v] += 1;
if (vals[v] > vmax)
{
vmax = vals[v];
}
}
List <double> xval = new List <double>(histCount + 1);
for (int i = 0; i < (histCount + 1); i++)
{
xval.Add(min + step * i);
}
cd.TitleX = this.Name;
cd.X = xval.ToArray();
if (histType == HistogramType.WithNormalDist)
{
double um, sigma;
MathStatistics.MeanStandardDeviation(values, out um, out sigma);
double[] norm_vals = new double[histCount + 1];
double vl_max = double.MinValue;
cd.TitlesY = new string[] { "Count (u=" + um.ToString("e2") + ",std=" + sigma.ToString("e2") + ")", "Normal distribution" };
for (int i = 0; i < (histCount + 1); i++)
{
double vl = 1 / (sigma * Math.Sqrt(2 * Math.PI)) * Math.Exp(-Math.Pow(xval[i] - um, 2) / (2 * Math.Pow(sigma, 2)));
norm_vals[i] = vl;
if (vl > vl_max)
{
vl_max = vl;
}
}
for (int i = 0; i < (histCount + 1); i++)
{
norm_vals[i] = norm_vals[i] * vmax / vl_max;
}
cd.Y = new double[][] { vals, norm_vals };
}
else
{
double um, sigma;
MathStatistics.MeanStandardDeviation(values, out um, out sigma);
cd.ChartType = ChartType.Stem;
cd.TitlesY = new string[] { this.Name + " ( u=" + um.ToString("e2") + ",std=" + sigma.ToString("e2") + ") " };
cd.Y = new double[][] { vals };
}
cds.Add(cd);
return(cds);
}
