/// <summary>
/// Given a 2D plot, map to a 1D plot assuming radial distances.
/// </summary>
/// <param name="ctx"></param>
/// <param name="plot"></param>
/// <returns>1D plot of the 2D plot</returns>
/// <remarks>
/// - Radial conversion assumes that both coordinates are the same units. For example, if they are x and y distances.
/// - Overflow & underflow bins are not handled.
/// - Errors are not properly handled.
/// </remarks>
public static ROOTNET.Interface.NTH1 asRadial(IScopeContext ctx, ROOTNET.Interface.NTH2 plot)
{
// First, determine the distances
var xmin = plot.Xaxis.GetBinLowEdge(1);
var ymin = plot.Yaxis.GetBinLowEdge(1);
var xmax = plot.Xaxis.GetBinUpEdge(plot.Xaxis.Nbins);
var ymax = plot.Yaxis.GetBinUpEdge(plot.Yaxis.Nbins);
var rmin = Math.Sqrt(xmin * xmin + ymin * ymin);
var rmax = Math.Sqrt(xmax * xmax + ymax * ymax);
var nbin = Math.Max(plot.Xaxis.Nbins, plot.Yaxis.Nbins);
var result = new ROOTNET.NTH1F(plot.Name, plot.Title, nbin, rmin, rmax);
// Loop over, adding everything in.
for (int i_x = 1; i_x <= plot.Xaxis.Nbins; i_x++)
{
for (int i_y = 1; i_y <= plot.Yaxis.Nbins; i_y++)
{
var x = plot.Xaxis.GetBinCenter(i_x);
var y = plot.Yaxis.GetBinCenter(i_y);
var r = Math.Sqrt(x * x + y * y);
result.Fill(r, plot.GetBinContent(i_x, i_y));
}
}
return(result);
}
/// <summary>
/// Given a 2D plot, turn it into an efficiency plot. You can control the cut on each axis ( greater than or less than ).
/// Each bin in the plot is the cut efficiency if you made the cut at those value.
/// </summary>
/// <param name="ctx"></param>
/// <param name="plot">The 2D plot to convert to an efficiency map</param>
/// <param name="xCutGreaterThan">True if the cut is greater than along the x axis</param>
/// <param name="yCutGreaterThan">True if the cut is greater than along the y axis</param>
/// <returns>The efficiency map</returns>
/// <remarks>
/// The overflow and underflow bins are taken into account for the total calculations, so they will contain
/// sensible results.
///
/// This works on a histogram, and thus a bin. So we have to define what the cut means. Since the cut is only has a real
/// value at the bin edges, we must define it as such.
/// - For a ">" cut, the low edge of the bin is the cut value.
/// - for a "<" cut, the high edge of the bin is the cut value.
///
/// </remarks>
public static ROOTNET.Interface.NTH2 asEfficiency(IScopeContext ctx, ROOTNET.Interface.NTH2 plot, bool xCutGreaterThan = true, bool yCutGreaterThan = true)
{
var result = plot.Clone() as ROOTNET.Interface.NTH2;
Tags.CopyTags(ctx, plot, result);
var xBins = plot.Xaxis.Nbins;
var yBins = plot.Yaxis.Nbins;
var xBinRange = BinOrdering(xCutGreaterThan, xBins);
var yBinRange = BinOrdering(yCutGreaterThan, yBins);
// We must now build a cumulative 2D matrix with the sizes for each.
foreach (var ixBin in xBinRange)
{
foreach (var iyBin in yBinRange)
{
var binSum = SumBinArea(plot,
BinOrdering(xCutGreaterThan, xBins, ixBin),
BinOrdering(yCutGreaterThan, yBins, iyBin));
result.SetBinContent(ixBin, iyBin, binSum);
}
}
// Turn it into an efficiency
var totalSum = SumBinArea(plot, BinOrdering(true, xBins), BinOrdering(true, yBins));
foreach (var ixBin in xBinRange)
{
foreach (var iyBin in yBinRange)
{
result.SetBinContent(ixBin, iyBin, result.GetBinContent(ixBin, iyBin) / totalSum);
}
}
// Set up for good display
result.Maximum = 1.0;
return(result);
}
