def _getAutoBinning(self):
'Get binning that is uniform around that peak area and non-uniform\n'
'in the tails. It guaranties that the number of entries per bin is\n'
'in a given range [minEntriesPerBin, maxEntriesPerBin] by merging\n'
'neighboring bins in the tails. This is useful for calculation of\n'
'chi2 statistic that obeys the chi2 PDF.\n'
'Optional features:\n'
' * the bin width is a pretty number, e.g. 1, 0.5, 0.2, etc.\n'
' * calculate the median for each bin. (->better looking plot)\n'
## Get the data as an array of doubles pointed by a tree
entries = self.data.tree().Draw(self.x.GetName(), '', 'goff')
## Create the DataDrivenBinning object
bins = DataDrivenBinning(entries, self.data.tree().GetV1(),
self.binContentMin, self.binContentMax)
return bins
## Build the model
# model = w.factory('Gaussian::gaus(x[-5,5],mean[0,-6,6],sigma[1,0.5,2])')
model = w.factory('RooSechPdf::sech(x[-5,5],mean[0,-6,6],sigma[1,0.5,2])')
x = w.factory('x')
for itoy in range(ntoys):
## Use the model to generate a toy dataset
data = model.generate(ROOT.RooArgSet(x), nentries, roofit.NumCPU(numCPU))
## Fit the model to the generated data
model.fitTo(data, roofit.NumCPU(numCPU), roofit.Verbose(False))
## Get custom binning
nevents = data.tree().Draw(x.GetName(), '', 'goff')
bins = DataDrivenBinning(nevents, data.tree().GetV1(), minBinContent,
maxBinContent)
binning = bins.binning(ROOT.RooBinning())
ubinning = bins.uniformBinning(ROOT.RooUniformBinning())
## Make sure that there are at least 1./epsilonPerBin curve points
## in each bin. The RooChi2Calculator fails if there is none perhaps
## due to a bug in RooCurve::average? Reason is not understood yet.
epsilon = epsilonPerBin * ubinning.binWidth(0) / bins.length()
## Make frames
log_plot = x.frame()
log_plot_bins = x.frame()
log_plot_medians = x.frame()
bins.setSigmaLevel(3)
lin_plot = x.frame(roofit.Range(-3,3))
## Plot the data and the model overlaid
plot = x.frame(nbins)
plot.SetTitle('Default Binning')
data.plotOn(plot)
model.plotOn(plot)
## Display the plot
canvas1 = canvases.next('Data_Driven_Binning')
canvas1.Divide(2,2)
canvas1.cd(1)
plot.Draw()
## Create the DataDrivenBinning object
n = data.tree().Draw(x.GetName(), '', 'goff')
bins = DataDrivenBinning(n, data.tree().GetV1(), minBinContent, maxBinContent)
## Use DataDrivenBinning to define a RooBinning
binning = bins.binning(ROOT.RooBinning())
ubinning = bins.uniformBinning(ROOT.RooUniformBinning())
## Make a second plot for comparison.
plot2 = x.frame()
plot2.SetTitle('Data-Driven Binning')
## First plot the data with uniform binning to set the per bin density
data.plotOn(plot2, ROOT.RooFit.Binning(ubinning), ROOT.RooFit.Invisible())
## Plot the data with the custom binning.
data.plotOn(plot2, ROOT.RooFit.Binning(binning))
## Build the model
# model = w.factory('Gaussian::gaus(x[-5,5],mean[0,-6,6],sigma[1,0.5,2])')
model = w.factory('RooSechPdf::sech(x[-5,5],mean[0,-6,6],sigma[1,0.5,2])')
x = w.factory('x')
for itoy in range(ntoys):
## Use the model to generate a toy dataset
data = model.generate(ROOT.RooArgSet(x), nentries, roofit.NumCPU(numCPU))
## Fit the model to the generated data
model.fitTo(data, roofit.NumCPU(numCPU), roofit.Verbose(False))
## Get custom binning
nevents = data.tree().Draw(x.GetName(), '', 'goff')
bins = DataDrivenBinning(nevents,
data.tree().GetV1(), minBinContent, maxBinContent)
binning = bins.binning(ROOT.RooBinning())
ubinning = bins.uniformBinning(ROOT.RooUniformBinning())
## Make sure that there are at least 1./epsilonPerBin curve points
## in each bin. The RooChi2Calculator fails if there is none perhaps
## due to a bug in RooCurve::average? Reason is not understood yet.
epsilon = epsilonPerBin * ubinning.binWidth(0) / bins.length()
## Make frames
log_plot = x.frame()
log_plot_bins = x.frame()
log_plot_medians = x.frame()
bins.setSigmaLevel(3)
lin_plot = x.frame(roofit.Range(-3, 3))
