def test_toys2():
"""Perform toys-study for possible fit bias and correct uncertainty evaluation
- generate `nToys` pseudoexperiments with some PDF `gen_pdf`
- fit teach experiment with the PDF `fit_pdf`
- store fit results
- fill distributions of fit results
"""
logger = getLogger('test_toys2')
results, stats = Toys.make_toys2(gen_pdf=gen_gauss,
fit_pdf=fit_gauss,
nToys=1000,
data=[mass],
gen_config={
'nEvents': 200,
'sample': True
},
fit_config={'silent': True},
gen_pars={
'mean_GG': 0.4,
'sigma_GG': 0.1
},
fit_pars={
'mean_GF': 0.4,
'sigma_GF': 0.1
},
silent=True,
progress=True)
for p in stats:
logger.info("Toys: %-20s : %s" % (p, stats[p]))
## make histos
h_mean = ROOT.TH1F(hID(), 'mean of Gauss ', 50, 0, 0.80)
h_sigma = ROOT.TH1F(hID(), 'sigma of Gauss', 50, 0.05, 0.15)
for r in results['mean_FG']:
h_mean.Fill(r)
for r in results['sigma_FG']:
h_sigma.Fill(r)
for h in (h_mean, h_sigma):
h.draw()
logger.info("%s :\n%s" % (h.GetTitle(), h.dump(30, 10)))
time.sleep(1)
def test_significance_toys():
"""Perform toy-study for significance of the signal
- generate `nToys` pseudoexperiments using background-only hypothesis
- fit each experiment with signal+background hypothesis
- store fit results
- fill distributions for fit results
"""
logger = getLogger('test_significance_toys')
## only background hypothesis
bkg_only = Models.Bkg_pdf("BKG", xvar=mass, power=0, tau=0)
signal = Models.Gauss_pdf('S', xvar=mass, mean=0.5, sigma=0.1)
signal.mean.fix(0.4)
signal.sigma.fix(0.1)
## signal + background hypothesis
model = Models.Fit1D(signal=signal, background=1)
model.background.tau.fix(0)
results, stats = Toys.make_toys2(
gen_pdf=bkg_only,
fit_pdf=model,
nToys=1000,
data=[mass],
gen_config={
'nEvents': 100,
'sample': True
},
fit_config={'silent': True},
gen_pars={'tau_BKG': 0.}, ## initial values for generation
fit_pars={
'B': 100,
'S': 10,
'phi0_Bkg_S': 0.0
}, ## initial fit values for parameters
silent=True,
progress=True)
for p in stats:
logger.info("Toys: %-20s : %s" % (p, stats[p]))
h_S = ROOT.TH1F(hID(), '#S', 60, 0, 60)
for r in results['S']:
h_S.Fill(r)
for h in (h_S, ):
h.draw()
logger.info("%s :\n%s" % (h.GetTitle(), h.dump(30, 10)))
time.sleep(1)
def test_parallel_toys():
"""Perform toys-study for possible fit bias and correct uncertainty evaluation
- generate `nToys` pseudoexperiments with some PDF `pdf`
- fit teach experiment with the same PDF
- store fit results
- calculate statistics of pulls
- fill distributions of fit results
- fill distributions of pulls
"""
results, stats = Toys.parallel_toys(pdf=gen_gauss,
nToys=1000,
nSplit=20,
data=[mass],
gen_config={'nEvents': 200},
fit_config={'silent': True},
init_pars={
'mean_GG': 0.4,
'sigma_GG': 0.1
},
silent=True,
progress=False)
for p in stats:
logger.info("Toys: %-20s : %s" % (p, stats[p]))
## make histos:
h_mean = ROOT.TH1F(hID(), 'mean of Gauss ', 100, 0, 0.80)
h_sigma = ROOT.TH1F(hID(), 'sigma of Gauss', 100, 0.05, 0.15)
for r in results['mean_GG']:
h_mean.Fill(r)
for r in results['sigma_GG']:
h_sigma.Fill(r)
for h in (h_mean, h_sigma):
h.draw()
logger.info("%s :\n%s" % (h.GetTitle(), h.dump(30, 10)))
time.sleep(1)
def test_frame1 ( ) :
logger = getLogger ( 'test_frame1' )
if root_info < ( 6 , 16 ) :
logger.warning ( "Test is disabled for this version of ROOT %s" % str ( root_info ) )
return
frame = DataFrame ( tname , fname )
tree = Tree ( name = tname , file = fname ).chain
h1 = tree .draw ( 'b1' , '1/b1' )
h2 = frame.draw ( 'b1' , '1/b1' )
h1 = ROOT.TH1D( hID() , '' , 100 , 0 , 1000 )
h2 = ROOT.TH1D( hID() , '' , 100 , 0 , 1000 )
tree.project ( h1 , 'b1' , '1.0/b1' )
frame_project ( tree , h2 , 'b1' , '2.0/b1' )
with wait ( 3 ), use_canvas ( 'test_frame1' ) :
h1.red ()
h2.blue ()
h1.draw ()
h2.draw ( 'same hist' )
## reload data
data = Data('DATA_tree', testdata)
mc = Data(tag_mc, testdata)
wsum = mc.chain.sumVar('w')
wvar = '%d*w/%s' % (len(data.chain), wsum.value())
nn = '%s' % (len(data.chain) * 1.0 / len(mc.chain))
for phi in vrange(0, 2 * math.pi, 10):
dvar = '%.5f*x+%.5f*y' % (math.cos(phi), math.sin(phi))
mn, mx = data.chain.statVar(dvar).minmax()
h1 = ROOT.TH1D(hID(), '', 100, *axis_range(mn, mx, delta=0.05))
h2 = h1.clone()
h3 = h1.clone()
data.chain.project(h1, dvar) ## data
mc.chain.project(h2, dvar, nn) ## original (non-weighted) MC
mc.chain.project(h3, dvar, wvar) ## weighted MC
mn, mx = h1.minmax()
mn, mx = axis_range(0, mx, delta=0.7)
h1.SetMaximum(mx)
h1.blue()
h2.green()
h3.red()
h1.draw('')
def test_gbreweight():
logger = getLogger("test_gbreweight")
try:
from ostap.tools.reweighter import Reweighter
rw = Reweighter()
except ImportError:
logger.error('GBReweighter is not available!')
return
if not os.path.exists(testdata):
with timing("Prepare input data", logger=logger):
prepare_data()
# =========================================================================
## Input data/mc samples
# =========================================================================
data = Data('DATA_tree', testdata)
mc = Data(tag_mc, testdata)
ddata, wdata = data.chain.slice('x y', transpose=True)
dmc, wmc = mc.chain.slice('x y', transpose=True)
## train BDT
rw.reweight(original=dmc, target=ddata)
## new weights
wnew = rw.weight(original=dmc)
mc.chain.add_new_branch('w', wnew)
## reload data
data = Data('DATA_tree', testdata)
mc = Data(tag_mc, testdata)
wsum = mc.chain.sumVar('w')
wvar = '%d*w/%s' % (len(data.chain), wsum.value())
nn = '%s' % (len(data.chain) * 1.0 / len(mc.chain))
for phi in vrange(0, 2 * math.pi, 10):
dvar = '%.5f*x+%.5f*y' % (math.cos(phi), math.sin(phi))
mn, mx = data.chain.statVar(dvar).minmax()
h1 = ROOT.TH1D(hID(), '', 100, *axis_range(mn, mx, delta=0.05))
h2 = h1.clone()
h3 = h1.clone()
data.chain.project(h1, dvar) ## data
mc.chain.project(h2, dvar, nn) ## original (non-weighted) MC
mc.chain.project(h3, dvar, wvar) ## weighted MC
mn, mx = h1.minmax()
mn, mx = axis_range(0, mx, delta=0.7)
h1.SetMaximum(mx)
h1.blue()
h2.green()
h3.red()
h1.draw('')
h2.draw('same hist')
h3.draw('same')
time.sleep(2)
def ex_func(particle, dataset, plots=None, verbose=False):
"""The example of the actual function that build histos
In this example it builds two histograms:
- accepted events
- rejected events
For dataset structure and variable names see:
https://twiki.cern.ch/twiki/bin/view/LHCb/PIDCalibPackage
The efficiency historgam can be later build in Ostap as :
>>> h_acc = ...
>>> h_rej = ...
>>> eff = 1/(1 + h_rej/h_acc)
"""
## we need ROOT and Ostap machinery!
import ROOT
from ostap.core.pyrouts import hID
# 1) define PID-cut and its negation
# For dataset structure and variable names see:
# https://twiki.cern.ch/twiki/bin/view/LHCb/PIDCalibPackage
accepted = 'Pi_ProbNNpi>0.5' ## note variable names
rejected = 'Pi_ProbNNpi<0.5' ## note variable names
# 2) prepare the histogtrams
hA = ROOT.TH2D(hID(), 'Accepted(%s)' % accepted, 15, 0, 150000, 10, 2, 5)
h1.Sumw2()
hR = ROOT.TH2D(hID(), 'Rejected(%s)' % rejected, 15, 0, 150000, 10, 2, 5)
h2.Sumw2()
# 3) fill the historgams with 'accepted' and 'rejected' events
# For dataset structure and variable names see:
# https://twiki.cern.ch/twiki/bin/view/LHCb/PIDCalibPackage
vlst = ROOT.RooArgList()
vlst.add(dataset.Pi_P) ## note variable names
vlst.add(dataset.Pi_Eta) ## note variable name
hA = DataSet.fillHistogram(hA, vlst, accepted) ## fill histo
hR = DataSet.fillHistogram(hR, vlst, rejected) ## fill histo
#
## and now update the output
#
if not plots:
hA.SetName(hA.GetTitle())
hR.SetName(hR.GetTitle())
plots = [hA, hR] ## "Accepted" & "Rejected" histograms
else:
plots[0] += hA ## "Accepted" histogram
plots[1] += hR ## "Rejected" histogram
hA.Delete()
hR.Delete()
if hA: del hA
if hR: del hR
if verbose:
logger.info('Accepted histo: %s' % plots[0].stat())
logger.info('Rejected histo: %s' % plots[1].stat())
return plots
def __init__(
self,
category, ## accessor to category
N, ## number of categories
methods, ## list of TMVA methods
variables, ## list of variables
signal, ## signal tree
background, ## background tree
signal_cuts='', ## signal cuts
background_cuts='', ## background cuts
spectators=[],
bookingoptions="Transformations=I;D;P;G,D",
configuration="nTrain_Signal=0:nTrain_Background=0:SplitMode=Random:NormMode=NumEvents:!V",
signal_weight=None,
background_weight=None,
name='TMVAChopper', ## the name
verbose=False, ## verbose ?
chop_signal=False, ## chop the signal ?
chop_background=True): ## chop the background ?
"""Create TMVA ``chopping'' trainer
>>> N = 11
>>> trainer = Trainer (
... category = '137*evt+813*run' ,
... N = N ,
... methods = [ # type name configuration
... ( ROOT.TMVA.Types.kMLP , 'MLP' , 'H:!V:EstimatorType=CE:VarTransform=N:NCycles=200:HiddenLayers=N+3:TestRate=5:!UseRegulator' ) ,
... ( ROOT.TMVA.Types.kBDT , 'BDTG' , 'H:!V:NTrees=100:MinNodeSize=2.5%:BoostType=Grad:Shrinkage=0.10:UseBaggedBoost:BaggedSampleFraction=0.5:nCuts=20:MaxDepth=2' ) ,
... ( ROOT.TMVA.Types.kCuts , 'Cuts' , 'H:!V:FitMethod=MC:EffSel:SampleSize=200000:VarProp=FSmart' ) ,
... ( ROOT.TMVA.Types.kFisher , 'Fisher' , 'H:!V:Fisher:VarTransform=None:CreateMVAPdfs:PDFInterpolMVAPdf=Spline2:NbinsMVAPdf=50:NsmoothMVAPdf=10' ),
... ( ROOT.TMVA.Types.kLikelihood , 'Likelihood' , 'H:!V:TransformOutput:PDFInterpol=Spline2:NSmoothSig[0]=20:NSmoothBkg[0]=20:NSmoothBkg[1]=10:NSmooth=1:NAvEvtPerBin=50' ) ] ,
... variables = [ 'var1' , 'var2' , 'var3' ] , ## Variables to use in the training
... signal = signal_tree , ## TTree/TChain with ``signal'' sample
... background = background_tree , ## TTree/TChain with ``background'' sample
... name = 'TMVAChopper' ,
... verbose = False )
"""
assert isinstance(
N, (int, long)) and 1 < N, "Invalid number of categories"
self.__chop_signal = True if chop_signal else False
self.__chop_background = True if chop_background else False
assert self.__chop_signal or self.__chop_background, "Neither signal nor background chopping"
self.__category = category
self.__N = N
self.__signal = signal
self.__background = background
self.__methods = tuple(methods)
self.__signal_weight = signal_weight
self.__signal_cuts = ROOT.TCut(signal_cuts)
self.__background_weight = background_weight
self.__background_cuts = ROOT.TCut(background_cuts)
self.__variables = tuple(variables)
self.__spectators = tuple(spectators)
self.__bookingoptions = bookingoptions
self.__configuration = configuration
self.__name = name
self.__verbose = True if verbose else False
self.__sig_histos = ()
self.__bkg_histos = ()
cat = '(%s)%%%d' % (self.category, self.N)
if self.chop_signal:
hs1 = ROOT.TH1F(hID(), 'Signal categories', self.N * 5, -0.5,
self.N - 1)
hs2 = h1_axis([-0.5 + i for i in range(self.N + 1)],
title=hs1.GetTitle())
self.signal.project(hs1, cat, self.signal_cuts)
self.signal.project(hs2, cat, self.signal_cuts)
self.__sig_histos = hs1, hs2
st = hs2.stat()
if 0 >= st.min():
logger.warning("Some signal categories are empty!")
logger.info('Signal category population mean/rms: %s/%s' %
(st.mean(), st.rms()))
if self.chop_background:
hb1 = ROOT.TH1F(hID(), 'Background categories', self.N * 5, -0.5,
self.N - 1)
hb2 = h1_axis([-0.5 + i for i in range(self.N + 1)],
title=hb1.GetTitle())
self.background.project(hb1, cat, self.background_cuts)
self.background.project(hb2, cat, self.background_cuts)
self.__bkg_histos = hb1, hb2
##
st = hb2.stat()
if 0 >= st.min():
logger.warning("Some background categories are empty!")
logger.info('Background category population mean/rms: %s/%s' %
(st.mean(), st.rms()))
## trick to please Kisa
from ostap.trees.trees import Chain
self.__signal = Chain(signal)
self.__background = Chain(background)
## book the trainers
self.__trainers = ()
self.__weights_files = []
self.__class_files = []
self.__output_files = []
self.__tar_file = None
self.__log_file = None
