def main():
'''
Main entry point of execution.
'''
## Mitigate numerical noise paying lower performance.
if debuglevel == 0:
ROOT.RooAbsReal.defaultIntegratorConfig().setEpsAbs(1e-9)
ROOT.RooAbsReal.defaultIntegratorConfig().setEpsRel(1e-9)
## Assemble the job name
name = '_'.join([em, src, cat])
f = ModeAndEffSigmaFitter(name, debuglevel)
f.run()
## Store RooFit objects in a rootfile
w = ROOT.RooWorkspace('w')
for item in [f.data, f.data_half_odd, f.data_half_even, f.model]:
w.Import(item)
w.Import(f.fit_data, f.name + '_fit_data')
w.Import(f.train_data, f.name + '_train_data')
w.Import(f.fit_result, f.name + '_fit_result')
w.writeToFile(output_filename)
## Store canvases in a rootfile
outfile = ROOT.TFile.Open(output_filename, 'UPDATE')
outfile.mkdir('Canvases').cd()
for c in canvases.canvases:
if c:
c.Write(c.GetName())
## Make the plots
canvases.make_plots('eps png C'.split())
def main():
'''
Main entry point of execution.
'''
## Mitigate numerical noise paying lower performance.
if debuglevel == 0:
ROOT.RooAbsReal.defaultIntegratorConfig().setEpsAbs(1e-9)
ROOT.RooAbsReal.defaultIntegratorConfig().setEpsRel(1e-9)
## Assemble the job name
name = '_'.join([em, src, cat])
f = ModeAndEffSigmaFitter(name, debuglevel)
f.run()
## Store RooFit objects in a rootfile
w = ROOT.RooWorkspace('w')
for item in [f.data, f.data_half_odd, f.data_half_even, f.model]:
w.Import(item)
w.Import(f.fit_data, f.name + '_fit_data')
w.Import(f.train_data, f.name + '_train_data')
w.Import(f.fit_result, f.name + '_fit_result')
w.writeToFile(output_filename)
## Store canvases in a rootfile
outfile = ROOT.TFile.Open(output_filename, 'UPDATE')
outfile.mkdir('Canvases').cd()
for c in canvases.canvases:
if c:
c.Write(c.GetName())
## Make the plots
canvases.make_plots('eps png C'.split())
## 1: Use limited statistics for testing
## 0: Use full stats for final results
debuglevel = 0
## Mitigate numerical noise paying lower performance.
if debuglevel == 0:
ROOT.RooAbsReal.defaultIntegratorConfig().setEpsAbs(1e-9)
ROOT.RooAbsReal.defaultIntegratorConfig().setEpsRel(1e-9)
fitters = []
for em in 'pho ele'.split():
for src in 'data mc'.split():
for icat in range(4):
cat = 'cat%d' % icat
print '+++ ', em, src, cat
fitter = ModeAndEffSigmaFitter(name = '_'.join([em, src, cat]),
debuglevel = debuglevel)
fitter.run()
fitters.append(fitter)
for icalcat in range(8):
cat = 'calcat%d' % icalcat
print '+++ ', em, src, cat
fitter = ModeAndEffSigmaFitter(name = '_'.join([em, src, cat]),
debuglevel = debuglevel)
fitter.run()
fitters.append(fitter)
## Store RooFit objects in a rootfile
w = ROOT.RooWorkspace('w')
for f in fitters:
for item in [f.data, f.data_half_odd, f.data_half_even, f.model]:
w.Import(item)
## 1: Use limited statistics for testing
## 0: Use full stats for final results
debuglevel = 0
## Mitigate numerical noise paying lower performance.
if debuglevel == 0:
ROOT.RooAbsReal.defaultIntegratorConfig().setEpsAbs(1e-9)
ROOT.RooAbsReal.defaultIntegratorConfig().setEpsRel(1e-9)
fitters = []
for em in 'pho ele'.split():
for src in 'data mc'.split():
for icat in range(4):
cat = 'cat%d' % icat
print '+++ ', em, src, cat
fitter = ModeAndEffSigmaFitter(name='_'.join([em, src, cat]),
debuglevel=debuglevel)
fitter.run()
fitters.append(fitter)
for icalcat in range(8):
cat = 'calcat%d' % icalcat
print '+++ ', em, src, cat
fitter = ModeAndEffSigmaFitter(name='_'.join([em, src, cat]),
debuglevel=debuglevel)
fitter.run()
fitters.append(fitter)
## Store RooFit objects in a rootfile
w = ROOT.RooWorkspace('w')
for f in fitters:
for item in [f.data, f.data_half_odd, f.data_half_even, f.model]:
w.Import(item)
