def train_factory(self, outputfile):
'''Train using TMVA::Factory.'''
# Make the DataLoader.
self.prepare()
# Make the factory.
factory = TMVA.Factory(self.name, outputfile, str(self.factoryoptions))
factory.SetVerbose(self.verbose)
methods = self.book_methods(factory)
# Train MVAs
factory.TrainAllMethods()
# Test MVAs
factory.TestAllMethods()
# Evaluate MVAs
factory.EvaluateAllMethods()
# Save the output.
outputfile.Close()
print '=== wrote root file {0}\n'.format(outputfile.GetName())
print '=== TMVAClassification is done!\n'
weightsfiles = dict((m, self.weights_file(m)) for m in methods)
classfiles = dict(
(m, self.weights_file(m, '.class.C')) for m in methods)
return weightsfiles, classfiles
def runJob():
TMVA.Tools.Instance()
TMVA.PyMethodBase.PyInitialize()
output = TFile.Open('TMVA.root', 'RECREATE')
factory = TMVA.Factory('TMVAClassification', output,
'!V:!Silent:Color:DrawProgressBar:Transformations=D,G:AnalysisType=Classification')
dataloader = TMVA.DataLoader('dataset')
def runJob():
TMVA.Tools.Instance()
TMVA.PyMethodBase.PyInitialize()
output = TFile.Open('TMVA_SSSF.root', 'RECREATE')
factory = TMVA.Factory(
'TMVAClassification', output,
'!V:!Silent:Color:DrawProgressBar:AnalysisType=Classification')
# factory = TMVA.Factory('TMVAClassification', output, '!V:!Silent:Color:DrawProgressBar:Transformations=D,G:AnalysisType=Classification')
dataloader = TMVA.DataLoader('datasetSSSF04Feb')
for br in config.mvaVariables:
dataloader.AddVariable(br)
for sampleName, sample in config.samples.items():
if config.structure[sampleName]['isData'] == 1:
continue
print sampleName
sample['tree'] = TChain("Events")
for f in sample['name']:
sample['tree'].Add(f)
if config.structure[sampleName]['isSignal'] == 1:
dataloader.AddSignalTree(sample['tree'], 1.0)
else:
dataloader.AddBackgroundTree(sample['tree'], 1.0)
# output_dim += 1
dataloader.PrepareTrainingAndTestTree(
TCut(config.cut), 'SplitMode=Random:NormMode=NumEvents:!V')
# factory.BookMethod(dataloader, TMVA.Types.kBDT, "BDTG4C2", "!H:!V:NTrees=500:MinNodeSize=1.5%:BoostType=Grad:Shrinkage=0.05:UseBaggedBoost:GradBaggingFraction=0.5:nCuts=200:MaxDepth=2" );
factory.BookMethod(
dataloader, TMVA.Types.kBDT, "BDTG4C1",
"!H:!V:NTrees=500:MinNodeSize=1.5%:BoostType=Grad:Shrinkage=0.05:UseBaggedBoost:GradBaggingFraction=0.5:nCuts=100:MaxDepth=2"
)
factory.BookMethod(
dataloader, TMVA.Types.kBDT, "BDTG4C05",
"!H:!V:NTrees=500:MinNodeSize=1.5%:BoostType=Grad:Shrinkage=0.05:UseBaggedBoost:GradBaggingFraction=0.5:nCuts=50:MaxDepth=2"
)
factory.BookMethod(
dataloader, TMVA.Types.kBDT, "BDTG4500",
"!H:!V:NTrees=500:MinNodeSize=1.5%:BoostType=Grad:Shrinkage=0.05:UseBaggedBoost:GradBaggingFraction=0.5:nCuts=500:MaxDepth=2"
)
factory.BookMethod(
dataloader, TMVA.Types.kBDT, "BDTG4750",
"!H:!V:NTrees=500:MinNodeSize=1.5%:BoostType=Grad:Shrinkage=0.05:UseBaggedBoost:GradBaggingFraction=0.5:nCuts=750:MaxDepth=2"
)
# Run training, test and evaluation
factory.TrainAllMethods()
factory.TestAllMethods()
factory.EvaluateAllMethods()
output.Close()
def TMVANN(filenameSTRING, outputDOTrootSTRING, sigtreeSTRINGS, bkgtreeSTRING,
variablesSTRING):
NNfile = R.TFile(outputDOTrootSTRING, "recreate")
NNfile.Close()
for sigtreeSTRING in sigtreeSTRINGS:
from ROOT import TMVA
file = R.TFile(filenameSTRING) #importing the datasetsignalslowlevel
signaltree = file.Get(sigtreeSTRING) #setting signaltree
backgroundtree = file.Get(bkgtreeSTRING) #setting backgroundtree)
sigweights = file.Get(sigtreeSTRING + weights)
bkgweights = file.Get(bkgtreeSTRING + weights)
TMVA.Tools.Instance()
NNfile = R.TFile(
outputDOTrootSTRING,
"update") #Writing the root file required for the TMVA factory
TMVAfactory = TMVA.Factory(
"TMVANN", NNfile,
"V:!Silent:Color:DrawProgressBar:Transformations=I;D;P;G,D:AnalysisType=Classification"
)
TMVAfactory.SetVerbose(False) #Setting extra info (verbose) to false
datasetsignalslowlevel = TMVA.DataLoader(
"datasetsignalslowlevel") #Instantiating a datasetsignalslowlevel
datasetsignalslowlevel.AddSignalTree(signaltree, 1.) #adding signal
datasetsignalslowlevel.AddBackgroundTree(backgroundtree,
1.) #adding background
print(sigweights)
datasetsignalslowlevel.SetSignalWeightExpression(weights)
datasetsignalslowlevel.SetBackgroundWeightExpression(weights)
for i in variablesSTRING: #adding our training variables to the TMVA
datasetsignalslowlevel.AddVariable(i)
signalcut = R.TCut("") #Variables are already cut
backgroundcut = R.TCut("")
datasetsignalslowlevel.PrepareTrainingAndTestTree(
signalcut, backgroundcut,
"nTrain_Signal= 0:nTrain_Background=0:Splitmode=Random:NormMode=NumEvents:!V"
)
TMVAfactory.BookMethod(
datasetsignalslowlevel, TMVA.Types.kMLP,
"LowLevelNN_3layer25,20,10_100Epoch_tanhNeuron" + sigtreeSTRING,
"H:!V:NeuronType=tanh:VarTransform=N:NCycles=100:HiddenLayers=25,20,10:TestRate=5"
)
TMVAfactory.TrainAllMethods()
TMVAfactory.TestAllMethods()
TMVAfactory.EvaluateAllMethods()
NNfile.Close()
NNfile = R.TFile(outputDOTrootSTRING, "update")
def runJob():
TMVA.Tools.Instance()
TMVA.PyMethodBase.PyInitialize()
dataloader = TMVA.DataLoader('dataset_8Feb')
output = TFile.Open('TMVA16.root', 'RECREATE')
factory = TMVA.Factory(
'TMVAClassification', output,
'!V:!Silent:Color:DrawProgressBar:AnalysisType=Classification')
for br in config.mvaVariables:
dataloader.AddVariable(br)
for sampleName, sample in config.samples.items():
if config.structure[sampleName]['isData'] == 1:
continue
sample['tree'] = TChain("Events")
for f in sample['name']:
sample['tree'].Add(f)
if config.structure[sampleName]['isSignal'] == 1:
dataloader.AddSignalTree(sample['tree'], 1.0)
else:
dataloader.AddBackgroundTree(sample['tree'], 1.0)
# output_dim += 1
dataloader.PrepareTrainingAndTestTree(
TCut(config.cut),
'SplitMode=Random::SplitSeed=10:NormMode=EqualNumEvents')
factory.BookMethod(
dataloader, TMVA.Types.kBDT, "BDT",
"!H:!V:NTrees=500:MinNodeSize=0.5%:MaxDepth=3:BoostType=AdaBoost:AdaBoostBeta=0.1:SeparationType=GiniIndex:nCuts=500"
)
# factory.BookMethod(dataloader, TMVA.Types.kBDT, "BDT8","!H:!V:NTrees=1500:MinNodeSize=0.5%:MaxDepth=1:BoostType=AdaBoost:AdaBoostBeta=0.75:SeparationType=GiniIndex:nCuts=1000" );
# factory.BookMethod(dataloader, TMVA.Types.kBDT, "BDT2", "!H:!V:NTrees=1200:MinNodeSize=0.5%:MaxDepth=3:BoostType=AdaBoost:AdaBoostBeta=0.1:SeparationType=GiniIndex:nCuts=800" );
# factory.BookMethod(dataloader, TMVA.Types.kBDT, "BDT3", "!H:!V:NTrees=800:MinNodeSize=0.5%:MaxDepth=3:BoostType=AdaBoost:AdaBoostBeta=0.2:SeparationType=GiniIndex:nCuts=500" );
# factory.BookMethod(dataloader, TMVA.Types.kBDT, "BDT4", "!H:!V:NTrees=700:MinNodeSize=0.5%:MaxDepth=3:BoostType=AdaBoost:AdaBoostBeta=0.5:SeparationType=GiniIndex:nCuts=500" );
# factory.BookMethod(dataloader, TMVA.Types.kBDT, "BDTG4D3", "!H:!V:NTrees=500:MinNodeSize=1.5%:BoostType=Grad:Shrinkage=0.05:UseBaggedBoost:GradBaggingFraction=0.5:nCuts=500:MaxDepth=3" );
# factory.BookMethod(dataloader, TMVA.Types.kBDT, "BDTG4C3", "!H:!V:NTrees=500:MinNodeSize=1.5%:BoostType=Grad:Shrinkage=0.05:UseBaggedBoost:GradBaggingFraction=0.5:nCuts=300:MaxDepth=2" );
# factory.BookMethod(dataloader, TMVA.Types.kBDT, "BDTG4SK01", "!H:!V:NTrees=500:MinNodeSize=1.5%:BoostType=Grad:Shrinkage=0.01:UseBaggedBoost:GradBaggingFraction=0.5:nCuts=500:MaxDepth=2" );
# factory.BookMethod(dataloader, TMVA.Types.kBDT, "BDTG4F07" , "!H:!V:NTrees=500:MinNodeSize=1.5%:BoostType=Grad:Shrinkage=0.05:UseBaggedBoost:GradBaggingFraction=0.7:nCuts=500:MaxDepth=2" );
# factory.BookMethod(dataloader, TMVA.Types.kBDT, "BDTG4SK01F07", "!H:!V:NTrees=500:MinNodeSize=1.5%:BoostType=Grad:Shrinkage=0.01:UseBaggedBoost:GradBaggingFraction=0.7:nCuts=500:MaxDepth=2" );
# Run training, test and evaluation
factory.TrainAllMethods()
factory.TestAllMethods()
factory.EvaluateAllMethods()
output.Close()
def TMVANN (filenameSTRING,outputDOTrootSTRING,sigtreeSTRINGS,bkgtreeSTRING,variablesSTRING):
for sigtreeSTRING in sigtreeSTRINGS:
from ROOT import TMVA
file = R.TFile(filenameSTRING) #importing the dataset
signaltree = file.Get(sigtreeSTRING) #setting signaltree
backgroundtree = file.Get(bkgtreeSTRING) #setting backgroundtree)
sigweights = file.Get(sigtreeSTRING+weights)
bkgweights = file.Get(bkgtreeSTRING+weights)
TMVA.Tools.Instance()
NNfile = R.TFile(outputDOTrootSTRING,"recreate") #Writing the root file required for the TMVA factory
TMVAfactory = TMVA.Factory("TMVANN",NNfile,"V:!Silent:Color:DrawProgressBar:Transformations=I;D;P;G,D:AnalysisType=Classification")
TMVAfactory.SetVerbose(False) #Setting extra info (verbose) to false
dataset = TMVA.DataLoader("dataset") #Instantiating a dataset
dataset.AddSignalTree(signaltree,1.) #adding signal
dataset.AddBackgroundTree(backgroundtree ,1.) #adding background
print(sigweights)
#dataset.SetSignalWeightExpression(weights)
#dataset.SetBackgroundWeightExpression(weights)
for i in variablesSTRING: #adding our training variables to the TMVA
dataset.AddVariable(i)
signalcut = R.TCut("") #Variables are already cut
backgroundcut = R.TCut("")
dataset.PrepareTrainingAndTestTree(signalcut,backgroundcut,"nTrain_Signal= 0:nTrain_Background=0:Splitmode=Random:NormMode=NumEvents:!V")
#using all signal and background points to train, random selection, normalised to summed event weights = number of events for each tree, no verbose
#Booking some methods
#TMVAfactory.BookMethod(dataset,TMVA.Types.kMLP,"ArtificialNeuralNetwork_1Layer_500Epoch_tanhNeuron","H:!V:NeuronType=tanh:VarTransform=N:NCycles=500:HiddenLayers=8:TestRate=5") #Artifical Neural Network 1 layers 500 epoch
TMVAfactory.BookMethod(dataset,TMVA.Types.kFisher, "FisherMethod","H:!V:Fisher:CreateMVAPdfs:PDFInterpolMVAPdf=Spline2:NbinsMVAPdf=60:NsmoothMVAPdf=10") #Fisher Method
TMVAfactory.BookMethod(dataset,TMVA.Types.kLikelihood, "BayesLikelihood","H:!V:TransformOutput:PDFInterpol=Spline2:NSmoothSig[0]=20:NSmoothBkg[0]=20:NSmoothBkg[1]=10:NSmooth=1:NAvEvtPerBin=60") #Bayes likleihood
TMVAfactory.BookMethod(dataset,TMVA.Types.kMLP,"ArtificialNeuralNetwork_4Layer8Node_500Epoch_tanhNeuron","H:!V:NeuronType=tanh:VarTransform=N:NCycles=500:HiddenLayers=8,8,8,8:TestRate=5") #Artifical Neural Network 8 layers 500 epoch
TMVAfactory.BookMethod(dataset,TMVA.Types.kMLP,"ArtificialNeuralNetwork_1Layer6Node_500Epoch_reluNeuron","H:!V:NeuronType=tanh:VarTransform=N:NCycles=500:HiddenLayers=6:TestRate=5") #Artifical Neural Network 8 layers 500 epoch
TMVAfactory.BookMethod(dataset,TMVA.Types.kMLP,"ArtificialNeuralNetwork_1Layer3Node_100Epoch_reluNeuron","H:!V:NeuronType=tanh:VarTransform=N:NCycles=100:HiddenLayers=3:TestRate=5") #Artifical Neural Network 8 layers 500 epoch
TMVAfactory.BookMethod(dataset,TMVA.Types.kMLP,"ArtificialNeuralNetwork_2Layer5Node_500Epoch_tanhNeuron","H:!V:NeuronType=tanh:VarTransform=N:NCycles=500:HiddenLayers=5,5:TestRate=5") #Artifical Neural Network 8 layers 500 epoch
TMVAfactory.TrainAllMethods()
TMVAfactory.TestAllMethods()
TMVAfactory.EvaluateAllMethods()
NNfile.Close()
print ('TMVANN Ran & made ROOT file ' + outputDOTrootSTRING+sigtreeSTRING)
def Evaluate(outdir):
sys.stdout = open(outdir + '/tmva.log', 'w')
# Output file
output = TFile(outdir + '/tmva.root', 'RECREATE')
# Create instance of TMVA factory (see TMVA/macros/TMVAClassification.C for more factory options)
# All TMVA output can be suppressed by removing the "!" (not) in
# front of the "Silent" argument in the option string
factory = TMVA.Factory("TMVARuleFit", output, "!V:!Silent:Color" )
# Set the variables use for the analysis
input = open(outdir + '/inputvars.txt')
for variable in input.readlines():
factory.AddVariable(variable[:-1], 'F')
# Set the weight directory
TMVA.gConfig().GetIONames().fWeightFileDir = outdir + "/weights"
# Limit the creation of correlation plots
TMVA.gConfig().GetVariablePlotting().fMaxNumOfAllowedVariablesForScatterPlots = 20
# Set the input file with signal and background events
factory.SetInputTrees(
outdir + '/signals.txt',
outdir + '/backgrounds.txt'
)
cutsig = TCut('')
cutbkg = TCut('')
factory.PrepareTrainingAndTestTree( cutsig, cutbkg, "SplitMode=Random:NormMode=NumEvents:!V" )
factory.BookMethod( TMVA.Types.kRuleFit, "RuleFit",
"H:!V:RuleFitModule=RFTMVA:Model=ModRuleLinear:MinImp=0.00001:RuleMinDist=0.001:NTrees=20:fEventsMin=0.01:fEventsMax=0.5:GDTau=-1.:GDTauPrec=0.01:GDStep=0.01:GDNSteps=10000:GDErrScale=1.02" )
# Train MVAs
factory.TrainAllMethods()
# Test MVAs
factory.TestAllMethods()
# Evaluate MVAs
factory.EvaluateAllMethods()
# Save the output.
output.Close()
def __init__(self, name, n_vars, n_targets=1,
method='BDT', task='Classification'):
self.name = name
self.n_vars = n_vars
self.n_targets = n_targets
self.method = method
self.task = task
self.tmpdir = tempfile.mkdtemp()
self.output = TFile(os.path.join(self.tmpdir, 'tmva_output.root'),
'recreate')
self.factory = TMVA.Factory(name, self.output,
'AnalysisType={0}:Silent'.format(task))
for n in range(n_vars):
self.factory.AddVariable('X_{0}'.format(n), 'F')
if task == 'Regression':
for n in range(n_targets):
self.factory.AddTarget('y_{0}'.format(n), 'F')
def train_and_test_MVA(name, signal_tree, background_tree, output_file_name,
n_sig, n_bgk):
outfile = root_open(output_file_name, 'recreate')
factory = TMVA.Factory(name, outfile, "!V:!Silent:Color:DrawProgressBar")
# signal_tree.SetBranchStatus('*', 0)
# background_tree.SetBranchStatus('*', 0)
for var in variables:
# signal_tree.SetBranchStatus(var, 1)
# background_tree.SetBranchStatus(var, 1)
factory.AddVariable(var, 'F')
factory.AddSignalTree(signal_tree)
factory.AddBackgroundTree(bkg_tree)
# passes selection (currently marked as all variables are defined.
cut1 = Cut('absolute_eta > 0')
cut2 = Cut('angle_bl > 0')
cut3 = Cut('M3 > 0')
cut = cut1 & cut2 & cut3
training_options = "nTrain_Signal=%d:nTrain_Background=%d:nTest_Signal=%d:nTest_Background=%d:!V" % (
n_sig, n_bgk, n_sig, n_bgk)
factory.PrepareTrainingAndTestTree(cut, cut, training_options)
# methods are
# PDE - RS method (PDERS)
# K-Nearest Neighbour classifier (KNN)
# Linear discriminant (LD)
factory.BookMethod(TMVA.Types.kLikelihood, "Likelihood",
"!V:NAvEvtPerBin=50")
# factory.BookMethod( TMVA.Types.kMLP, "MLP", "!V:NCycles=50:HiddenLayers=10,10:TestRate=5" )
#
# factory.BookMethod( TMVA.Types.kBDT, "BDT", "!V:BoostType=Grad:nCuts=20:NNodesMax=5" );
# Train MVAs using the set of training events
factory.TrainAllMethods()
# ---- Evaluate all MVAs using the set of test events
factory.TestAllMethods()
# ----- Evaluate and compare performance of all configured MVAs
factory.EvaluateAllMethods()
outfile.close()
bkg_file.close()
def add_variables(data_loader, good_variables):
for variable in good_variables:
if variable != "Bc_M":
data_loader.AddVariable(variable, "F")
return data_loader
if __name__ == '__main__':
decay_name = decay_names[decay_nb]
output_file = TFile("~/TMVA/TMVAoutput" + decay_name + str(run) + ".root",
"RECREATE")
factory = TMVA.Factory("TMVA_" + decay_name, output_file,
"DrawProgressBar=True")
data_loader = TMVA.DataLoader("dataloader")
moca_tree = dhand.combine_trees(run, decay_name, True)
data_tree = dhand.combine_trees(run, decay_name, False)
branches_to_keep = dhand.branch_selection(data_tree, branches, [])
moca_tree = dhand.activate_branches(moca_tree, branches_to_keep)
data_tree = dhand.activate_branches(data_tree, branches_to_keep)
add_variables(data_loader, branches_to_keep)
sgcut_test = TCut("runNumber%5==" + kfold +
"&& (Bc_M > 5200 && Bc_M < 5400)")
sgcut_train = TCut("runNumber%5!=" + kfold +
"&& (Bc_M > 5200 && Bc_M < 5400)")
bgcut_test = TCut("runNumber%5==" + kfold + "&& Bc_M > 5400")
bgcut_train = TCut("runNumber%5!=" + kfold + "&& Bc_M > 5400")
from ROOT import TMVA, TFile, TTree, TCut, ObjectProxy
from subprocess import call
from os.path import isfile
import ROOT
from keras.models import Sequential
from keras.layers import Dense, Activation
from keras.regularizers import l2
from keras.optimizers import SGD
# Setup TMVA
TMVA.Tools.Instance()
TMVA.PyMethodBase.PyInitialize()
output = TFile.Open('TMVA.root', 'RECREATE')
factory = TMVA.Factory('TMVAMulticlass', output,
'!V:!Silent:Color:DrawProgressBar:Transformations=D,G:AnalysisType=multiclass')
# Load data
data = TFile.Open("/net/data_cms1b/user/dmeuser/top_analysis/output/ttbar_res100.0_new.root")
t_all = data.Get("ttbar_res100.0/ttbar_res_dilepton_CP5")
dataloader = TMVA.DataLoader('dataset')
# ~for var in ["PuppiMET","METunc_Puppi","MET","HT","nJets","n_Interactions","Lep1_pt","Lep1_phi","Lep1_eta","Lep1_E","Lep1_flavor","Lep2_pt","Lep2_phi","Lep2_eta","Lep2_E","Lep2_flavor","Jet1_pt","Jet1_phi","Jet1_eta","Jet1_E","Jet2_pt","Jet2_phi","Jet2_eta","Jet2_E"]:
for var in ["PuppiMET","METunc_Puppi","HT","nJets","n_Interactions","Lep1_pt","Lep1_phi","Lep1_eta","Lep1_E","Lep1_flavor","Lep2_pt","Lep2_phi","Lep2_eta","Lep2_E","Lep2_flavor","Jet1_pt","Jet1_phi","Jet1_eta","Jet1_E","Jet2_pt","Jet2_phi","Jet2_eta","Jet2_E"]:
dataloader.AddVariable(var)
dataloader.AddTree(t_all, 'Bin1')
dataloader.AddTree(t_all, 'Bin2')
dataloader.AddTree(t_all, 'Bin3')
dataloader.AddTree(t_all, 'Bin4')
X = np.concatenate([class_0, class_1, class_2])
y = np.ones(X.shape[0])
w = RNG.randint(1, 10, n_events * 3)
y[:class_0.shape[0]] *= 0
y[-class_2.shape[0]:] *= 2
permute = RNG.permutation(y.shape[0])
X = X[permute]
y = y[permute]
# Split into training and test datasets
X_train, y_train, w_train = X[:n_events], y[:n_events], w[:n_events]
X_test, y_test, w_test = X[n_events:], y[n_events:], w[n_events:]
output = TFile('tmva_output.root', 'recreate')
factory = TMVA.Factory('classifier', output,
'AnalysisType=Multiclass:'
'!V:Silent:!DrawProgressBar')
if ROOT_VERSION >= '6.07/04':
data = TMVA.DataLoader('.')
else:
data = factory
for n in range(2):
data.AddVariable('f{0}'.format(n), 'F')
# Call root_numpy's utility functions to add events from the arrays
add_classification_events(data, X_train, y_train, weights=w_train)
add_classification_events(data, X_test, y_test, weights=w_test, test=True)
# The following line is necessary if events have been added individually:
data.PrepareTrainingAndTestTree(TCut('1'), 'NormMode=EqualNumEvents')
def main():
try:
# retrive command line options
shortopts = "m:M:C:B:i:t:o:vh?"
opts, args = getopt.getopt(sys.argv[1:], shortopts)
except getopt.GetoptError:
# print help information and exit:
print "ERROR: unknown options in argument %s" % sys.argv[1:]
usage()
sys.exit(1)
infname = DEFAULT_INFNAME
methods = DEFAULT_METHODS
mass = DEFAULT_MASS
cat = DEFAULT_CAT
outfname = DEFAULT_OUTFNAME
treeNameSig = DEFAULT_TREESIG
treeNameBkg = DEFAULT_TREEBKG
bkg_method = DEFAULT_BACKGROUND
width = DEFAULT_WIDTH
verbose = False
for o, a in opts:
if o in ("-?", "-h", "--help", "--usage"):
usage()
sys.exit(0)
elif o in ("-m", "--methods"):
methods = a
elif o in ("-M", "--mass"):
mass = int(a)
elif o in ("-C", "--cat"):
cat = int(a)
elif o in ("-B", "--background"):
bkg_method = int(a)
elif o in ("-i", "--inputfile"):
infname = a
elif o in ("-o", "--outputfile"):
outfname = a
elif o in ("-t", "--inputtrees"):
a.strip()
trees = a.rsplit(' ')
trees.sort()
trees.reverse()
if len(trees) - trees.count('') != 2:
print "ERROR: need to give two trees (each one for signal and background)"
print trees
sys.exit(1)
treeNameSig = trees[0]
treeNameBkg = trees[1]
elif o in ("-v", "--verbose"):
verbose = True
if (width == 0.02): width_str = "_2pt"
elif (width == 0.07): width_str = "_7pt"
mass_str = "_" + str(mass)
cat_str = "_" + str(cat)
if cat < 0:
cat_str = "_all"
outfname = outfname + mass_str + cat_str + ".root"
#treeNameSig = treeNameSig + mass_str
#treeNameBkg = treeNameBkg + mass_str
# Print methods
mlist = methods.replace(' ', ',').split(',')
print "=== TMVAClassification: use method(s)..."
for m in mlist:
if m.strip() != '':
print "=== - <%s>" % m.strip()
# Import ROOT classes
from ROOT import gSystem, gROOT, gApplication, TFile, TTree, TCut
# check ROOT version, give alarm if 5.18
if gROOT.GetVersionCode() >= 332288 and gROOT.GetVersionCode() < 332544:
print "*** You are running ROOT version 5.18, which has problems in PyROOT such that TMVA"
print "*** does not run properly (function calls with enums in the argument are ignored)."
print "*** Solution: either use CINT or a C++ compiled version (see TMVA/macros or TMVA/examples),"
print "*** or use another ROOT version (e.g., ROOT 5.19)."
sys.exit(1)
# Logon not automatically loaded through PyROOT (logon loads TMVA library)
# load also GUI
gROOT.SetMacroPath("./")
#gROOT.Macro ( "./TMVAlogon.C" )
#gROOT.LoadMacro ( "./TMVAGui.C" )
# Import TMVA classes from ROOT
from ROOT import TMVA
# Output file
outputFile = TFile(outfname, 'RECREATE')
# Create instance of TMVA factory (see TMVA/macros/TMVAClassification.C for
# more factory options)
# All TMVA output can be suppressed by removing the "!" (not) in
# front of the "Silent" argument in the option string
factory = TMVA.Factory(
"TMVAClassification", outputFile,
"!V:!Silent:Color:DrawProgressBar:Transformations=I;D;P;G,D:AnalysisType=Classification"
)
# Set verbosity
factory.SetVerbose(verbose)
# Define the input variables that shall be used for the classifier training
factory.AddVariable("pho1_ptOverM", "P_{T}^{lead} / M_{H}", "", 'F')
factory.AddVariable("pho2_ptOverM", "P_{T}^{sublead} / M_{H}", "", 'F')
factory.AddVariable("pho1_eta", "#eta^{lead}", "", 'F')
factory.AddVariable("pho2_eta", "#eta^{sublead}", "", 'F')
factory.AddVariable("d_phi", "#Delta #phi", "rad", 'F')
#should this be cos delta Phi
factory.AddVariable("H_ptOverM", "P_{T}^{Higgs}/M_{H}", "", 'F')
factory.AddVariable("H_eta", "#eta^{Higgs}", "", 'F')
factory.AddVariable("sigmaMOverM", "#sigmaM_{cor} / M", 'F')
factory.AddVariable("sigmaMOverM_wrongVtx", "#sigmaM_{wrong} / M", 'F')
factory.AddVariable("vtx_prob", "P_{vertex}", "", 'F')
#factory.AddVariable( "cos_theta_star","cos(#theta)*", "", 'F' );
#factory.AddVariable( "max_eta","max(#eta^{lead},#eta^{sub.})", "", 'F' );
#factory.AddVariable( "min_r9","min(r9^{lead},r9^{sub.})", "", 'F' );
factory.AddVariable("deltaMOverM", "#DeltaM / M_{Hypth}.", 'F')
input = TFile.Open(infname)
# Get the signal and background trees for training
signal_train = input.Get(treeNameSig + "_train" + mass_str + ".0")
signal_test = input.Get(treeNameSig + "_test" + mass_str + ".0")
background_train = input.Get(treeNameBkg + "_train" + width_str +
mass_str + ".0")
background_test = input.Get(treeNameBkg + "_test" + width_str + mass_str +
".0")
# Global event weights (see below for setting event-wise weights)
signalWeight = 1.0
backgroundWeight = 1.0
# ====== register trees ====================================================
factory.AddSignalTree(signal_train, signalWeight, "train")
factory.AddBackgroundTree(background_train, backgroundWeight, "train")
factory.AddSignalTree(signal_test, signalWeight, "test")
factory.AddBackgroundTree(background_test, backgroundWeight, "test")
# Set individual event weights (the variables must exist in the original
# TTree)
factory.SetBackgroundWeightExpression("wt")
factory.SetSignalWeightExpression("wt")
# Apply additional cuts on the signal and background sample.
# example for cut: mycut = TCut( "abs(var1)<0.5 && abs(var2-0.5)<1" )
mycut = TCut("fabs(deltaMOverM)<=" + str(width)) #
# Here, the relevant variables are copied over in new, slim trees that are
# used for TMVA training and testing
factory.PrepareTrainingAndTestTree(
mycut, mycut,
"nTrain_Signal=0:nTrain_Background=0:NormMode=NumEvents:!V")
# Boosted Decision Trees
# NEW PARAMETERS
factory.BookMethod(
TMVA.Types.kBDT, "BDT_ada" + mass_str + cat_str,
"!H:!V:NTrees=400:nEventsMin=150:MaxDepth=3:BoostType=AdaBoost:AdaBoostBeta=0.05:SeparationType=GiniIndex:nCuts=50:PruneMethod=NoPruning"
)
factory.BookMethod(
TMVA.Types.kBDT, "BDT_grad" + mass_str + cat_str,
"!H:!V:NTrees=500:BoostType=Grad:Shrinkage=0.10:UseBaggedGrad:GradBaggingFraction=0.6:SeparationType=GiniIndex:nCuts=50:NNodesMax=5"
)
#test
# --------------------------------------------------------------------------------------------------
# ---- Now you can tell the factory to train, test, and evaluate the MVAs.
# Train MVAs
#factory.OptimizeAllMethods()
factory.TrainAllMethods()
# Test MVAs
factory.TestAllMethods()
# Evaluate MVAs
factory.EvaluateAllMethods()
# Save the output.
outputFile.Close()
print "=== wrote root file %s\n" % outfname
print "=== TMVAClassification is done!\n"
from os.path import isfile
import sys
import numpy as np
from keras.models import Sequential
from keras.layers import Dense, Conv2D, Flatten, Activation, Reshape
from keras.regularizers import l2
from keras.optimizers import SGD
# Setup TMVA
TMVA.Tools.Instance()
TMVA.PyMethodBase.PyInitialize()
output = TFile.Open('TMVA_CNN_PyKeras.root', 'RECREATE')
factory = TMVA.Factory(
'TMVAClassification', output,
'!V:!Silent:Color:DrawProgressBar:Transformations=None:AnalysisType=Classification'
)
############################Loading the data file
data = TFile.Open("/home/jui/Desktop/tmva/sample_images_32x32.root")
# print(data.ls())
signal = data.Get('sig_tree;2')
background = data.Get('bkg_tree;2')
dataloader = TMVA.DataLoader('dataset_evaltest')
imgSize = 1024
dataloader.AddVariablesArray("vars", imgSize)
dataloader.AddSignalTree(signal, 1.0)
from root_numpy import ROOT_VERSION
from ROOT import TMVA, TFile, TCut
from array import array
plt.style.use('ggplot')
RNG = np.random.RandomState(1)
# Create an example regression dataset
X = np.linspace(0, 6, 100)[:, np.newaxis]
y = np.sin(X).ravel() + \
np.sin(6 * X).ravel() + \
RNG.normal(0, 0.1, X.shape[0])
# Fit a regression model
output = TFile('tmva_output.root', 'recreate')
factory = TMVA.Factory('regressor', output, 'AnalysisType=Regression:'
'!V:Silent:!DrawProgressBar')
if ROOT_VERSION >= '6.07/04':
data = TMVA.DataLoader('.')
else:
data = factory
data.AddVariable('x', 'F')
data.AddTarget('y', 'F')
add_regression_events(data, X, y)
add_regression_events(data, X, y, test=True)
# The following line is necessary if events have been added individually:
data.PrepareTrainingAndTestTree(TCut('1'), '')
if ROOT_VERSION >= '6.07/04':
BookMethod = factory.BookMethod
#ROOT.gROOT.SetBatch(True)
# Setup TMVA
TMVA.Tools.Instance()
TMVA.PyMethodBase.PyInitialize()
#r = ROOT.TRInterface.Instance()
#TMVA.RMethodBase.Initialize()
#MethodRXGB.Init()
output = TFile.Open(
'Output_Classification_BsPhiJpsiEE_noPhiM_pTcuts_BOBest.root', 'RECREATE')
#factory = TMVA.Factory('TMVAClassification_BsPhiEE', output,
# '!V:ROC:!Silent:Color:DrawProgressBar:Transformations=D,G:AnalysisType=Classification')
factory = TMVA.Factory(
'TMVAClassification_BsPhiJpsiEE_noPhiM_pTcuts_BOBest', output,
'!V:ROC:!Silent:Color:DrawProgressBar:AnalysisType=Classification')
bkg_name = "BsPhiJpsiEE_MVATraining_Bkg_pTcuts.root"
sig_name = "BsPhiJpsiEE_MVATraining_Sig_pTcuts.root"
#bkg_name = "BsPhiJpsiEE_MVATraining_Bkg.root"
#sig_name = "BsPhiJpsiEE_MVATraining_Sig.root"
#branches = ['elePtLead', 'elePtSublead', 'kaonPtLead', 'kaonPtSublead', 'jpsiPt', 'phiPt', 'bsPt', 'eledR', 'kaondR', 'jpsiPhidR', 'svProb', 'svCosine', 'svLxySig', 'eleD0Lead', 'eleD0Sublead', 'eleDzLead', 'eleDzSublead', 'kaonD0Lead', 'kaonD0Sublead', 'kaonDzLead', 'kaonDzSublead', 'kaonNormChi2Lead', 'kaonNormChi2Sublead']
#branches = ['elePtLead', 'elePtSublead', 'kaonPtLead', 'kaonPtSublead', 'bsPt', 'eledR', 'kaondR', 'jpsiPhidR', 'svProb', 'svCosine', 'svLxySig', 'eleD0Lead', 'eleD0Sublead', 'eleDzLead', 'eleDzSublead', 'kaonD0Lead', 'kaonD0Sublead', 'kaonDzLead', 'kaonDzSublead', 'kaonNormChi2Lead', 'kaonNormChi2Sublead']
branches = [
'elePtLead', 'elePtSublead', 'kaonPtLead', 'kaonPtSublead', 'bsPt',
'svProb', 'svCosine', 'svLxySig', 'eleD0Lead', 'eleD0Sublead', 'eleDzLead',
'eleDzSublead', 'kaonD0Lead', 'kaonD0Sublead', 'kaonDzLead',
'kaonDzSublead', 'kaonNormChi2Lead', 'kaonNormChi2Sublead'
]
loader.SetBackgroundWeightExpression(weightStrB)
mycutSig = TCut(cutStrS)
mycutBkg = TCut(cutStrB)
loader.PrepareTrainingAndTestTree(
mycutSig, mycutBkg,
"nTrain_Signal=" + str(NSIG) +\
":nTrain_Background=" + str(NBKG) +\
":nTest_Signal=" + str(NSIG_TEST) +\
":nTest_Background=" + str(NBKG_TEST) +\
":SplitMode=Random:NormMode=NumEvents:!V"
)
factory = TMVA.Factory(
"Optimization",
'!V:!ROC:!Silent:Color:!DrawProgressBar:Transformations=I;:AnalysisType=Classification'
)
(TMVA.gConfig().GetIONames()).fWeightFileDir = '/weights'
kerasSetting = '!H:!V:VarTransform=G:FilenameModel=' + modelName +\
':SaveBestOnly=true' +\
':NumEpochs=' + str(epochs) +\
':BatchSize=' + str(batchSize) +\
':TriesEarlyStopping=' + str(patience)
factory.BookMethod(loader, TMVA.Types.kPyKeras, "PyKeras", kerasSetting)
factory.TrainAllMethods()
factory.TestAllMethods()
factory.EvaluateAllMethods()
Use = {
'BDT': 1,
'BDTG': 0,
'BDTRT': 0,
'BDTB': 0,
'BDTD': 0,
'BDTF': 0,
'MLP': 0,
'MLPBFGS': 0,
'MLPBNN': 0,
'CFMlpANN': 0,
'TMlpANN': 0
}
factory = t.Factory(
'vbf_bdt_combined_james_current', outputFile,
'!V:!Silent:Color:DrawProgressBar:Transformations=I;D;G,D:AnalysisType=Classification'
)
factory.Print()
t.gConfig().GetIONames().fWeightFileDir = outputWeightsDir
dataloader = t.DataLoader(".")
dataloader.AddVariable('dijetDEta', 'dijetDEta', 'dijetDEta', 'F')
dataloader.AddVariable('dijetDPhi', 'dijetDPhi', 'dijetDPhi', 'F')
dataloader.AddVariable('llgJJDPhi', 'llgJJDPhi', 'llgJJDPhi', 'F')
dataloader.AddVariable('jPhotonDRMin', 'jPhotonDRMin', 'jPhotonDRMin', 'F')
dataloader.AddVariable('ptt', 'ptt', 'ptt', 'F')
dataloader.AddVariable('jetOnePt', 'jetOnePt', 'jetOnePt', 'F')
dataloader.AddVariable('jetTwoPt', 'jetTwoPt', 'jetTwoPt', 'F')
dataloader.AddVariable('kin_bdt_james', 'kin_bdt_james', 'kin_bdt_james',
'F')
Authors: A. Ramirez-Morales and J. Salmon-Gamboa
---------------------------------------------------------------
'''
# tmva module
import ROOT
from ROOT import TFile, TMVA, TCut
m_outputFile = TFile("TMVA.root", 'RECREATE')
m_inputFile = TFile.Open("./data/cancer.root")
TMVA.Tools.Instance()
m_factory = TMVA.Factory(
"TMVAClassification",
m_outputFile #this is optional
,
"!V:Color:DrawProgressBar:Transformations=I;D;P;G,D:AnalysisType=Classification"
)
m_loader = TMVA.DataLoader("dataset")
m_loader.AddVariable("age", 'F')
m_loader.AddVariable("menopause", 'F')
m_loader.AddVariable("tumorSize", 'F')
m_loader.AddVariable("invNodes", 'F')
m_loader.AddVariable("degMalig", 'F')
# m_loader.AddVariable("nodeCaps", 'F')
# m_loader.AddVariable("breast", 'I')
# m_loader.AddVariable("breastQuad")
# m_loader.AddVariable("irradiat")
traintree_name = treename
trainfilename="proctrain.csv.root"
trainfile = TFile.Open(trainfilename,"read")
traintree = trainfile.Get(traintree_name)
TMVA.Tools.Instance()
# create the tmva output file, which will be full of details about the training
fout = TFile("tmvatest.root","RECREATE")
# use the default factory
factory = TMVA.Factory("TMVAClassification", fout)
dataloader = TMVA.DataLoader("dataset")
# build the list of variables
al=traintree.GetListOfBranches()
varlist=[]
for i in range(al.GetEntries()):
varlist+=[al[i].GetName()]
if debug:
print "all variables of ",trainfile, " ", varlist
print "now stripping EventId Weight and Label "
# these three variables should not be used for training
def main():
NTupName = "JetTree"
alg = "TruthRawTrim"
cuts = ["eta>-1.2","eta<1.2","pt>300","pt<5000","m>61","m<85"]
vars = ["Tau21","T2jet"]
#NTupName = "varTree"
verbose = True
#alg = "AK10LCTRIMF5R20"
spectators = ["m"]
#cuts = ["eta>-1.2","eta<1.2","pt>200","pt<350","m>61","m<85","TruthRecoMatch==1"]
#vars = ["TauWTA2TauWTA1","ZCUT12","Dip23","TJetVol","ActiveArea","PullC10","Angularity"]
methods = "Likelihood"
print "Starting and getting arguments:"
allargs = sys.argv[1:]
if len(allargs)<5:
print "You input these args"
print allargs
print "Not enough args, please try again"
return 1
else:
alg = allargs[0]
spectators = allargs[1].split(",")
cuts = allargs[2].split(",")
vars = allargs[3].split(",")
methods = allargs[4]
print "Running with args:"
print " alg = ",alg
print " spectators = ",spectators
print " cuts = ",cuts
print " vars = ",vars
print " methods = ",methods
# Print methods
mlist = methods.replace(' ',',').split(',')
print "=== TMVAClassification: use method(s)..."
for m in mlist:
if m.strip() != '':
print "=== - <%s>" % m.strip()
#===============================
#Read training and test data
#===============================
#InputDir = "../gen_20170529/"
InputDir="../Ana_EventGeneration/"
#InputDir="~/Downloads/"
print "Getting inputs from: ",InputDir
#s1 = TFile(InputDir+"ntuple_ttbar_2000.root");
#b1 = TFile(InputDir+"ntuple_dijet_800_1400.root");
s1 = TFile(InputDir+"ntuple_tt_test10000.root");
b1 = TFile(InputDir+"ntuple_dijet_test10000.root");
# Output file
OutFileName="testout.root"
outputFile = TFile( OutFileName, 'RECREATE' )
# Create instance of TMVA factory (see TMVA/macros/TMVAClassification.C for more factory options)
# All TMVA output can be suppressed by removing the "!" (not) in
# front of the "Silent" argument in the option string
factory = TMVA.Factory( "TMVAClassification", outputFile,
"!V:!Silent:Color:DrawProgressBar:Transformations=I;D;P;G,D:AnalysisType=Classification" )
# Set verbosity
factory.SetVerbose( verbose )
# data loader
dataloader=TMVA.DataLoader("dataset")
# weight=""
# weight+="pass_selection*EventWeight*CrossSection*("
# weight+=alg+"_pt>"+pt1+" && "
# weight+=alg+"_pt<"+pt2
# if m1!="0":
# weight+=" && "+alg+"_m>"+m1+" && "
# weight+=alg+"_m<"+m2
# weight+=")"
#
# #Get signal and background histograms
# if variable=="mass":
# histname = "akt10"+alg+"_trim_"+variable
# else:
# histname = alg+"_"+variable
#======================================
#Predefined cuts - for isntance on M(j1)
#======================================
mycuts = "1.0"
mycutb = "1.0"
for cut in cuts:
placecut=cut
if cut[:2]=="pt" or cut[:3]=="eta" or cut[:4]=="m":
placecut = "* ("+alg+"_"+cut+")"
else:
placecut="* ("+cut+") "
mycuts += placecut
mycutb += placecut
print "MyCutsSig: ",mycuts
print "MyCutsBkg: ",mycutb
#===================================
#Spectator variables from tree
#=====================================
for spec in spectators:
dataloader.AddSpectator( spec, 'F' )
#===================================
#MVA variables from tree
#=====================================
for var in vars:
dataloader.AddVariable( var , 'F' )
#===============================
#Read training and test data
#===============================
print "Getting trees ... ",NTupName
st1 = s1.Get(NTupName)
bt1 = b1.Get(NTupName)
#=========================================
# global event weights per tree (see below for setting event-wise weights)
#=========================================
ws1 = 1.0
wb1 = 1.0
#=========================================
# You can add an arbitrary number of signal or background trees
#=========================================
dataloader.AddSignalTree ( st1, ws1 );
#SCHSU
#dataloader.SetSignalWeightExpression("EventWeight*CrossSection");
dataloader.AddBackgroundTree( bt1, wb1 );
#dataloader.SetBackgroundWeightExpression("EventWeight*CrossSection");
# Here, the relevant variables are copied over in new, slim trees that are
# used for TMVA training and testing
# "SplitMode=Random" means that the input events are randomly shuffled before
# splitting them into training and test samples
mycutSig = TCut(mycuts)
mycutBkg = TCut(mycutb)
dataloader.PrepareTrainingAndTestTree( mycutSig, mycutBkg,
"nTrain_Signal=0:nTrain_Background=0:SplitMode=Random:NormMode=NumEvents:!V" )
# --------------------------------------------------------------------------------------------------
# ---- Book MVA methods
#
# please lookup the various method configuration options in the corresponding cxx files, eg:
# src/MethoCuts.cxx, etc, or here: http:#tmva.sourceforge.net/optionRef.html
# it is possible to preset ranges in the option string in which the cut optimisation should be done:
# "...:CutRangeMin[2]=-1:CutRangeMax[2]=1"...", where [2] is the third input variable
# Cut optimisation
if "Cuts" in mlist:
factory.BookMethod( dataloader, TMVA.Types.kCuts, "Cuts",
"!H:!V:FitMethod=MC:EffSel:SampleSize=200000:VarProp=FSmart" )
if "CutsD" in mlist:
factory.BookMethod( dataloader, TMVA.Types.kCuts, "CutsD",
"!H:!V:FitMethod=MC:EffSel:SampleSize=200000:VarProp=FSmart:VarTransform=Decorrelate" )
if "CutsPCA" in mlist:
factory.BookMethod( dataloader, TMVA.Types.kCuts, "CutsPCA",
"!H:!V:FitMethod=MC:EffSel:SampleSize=200000:VarProp=FSmart:VarTransform=PCA" )
if "CutsGA" in mlist:
factory.BookMethod( dataloader, TMVA.Types.kCuts, "CutsGA",
"H:!V:FitMethod=GA:CutRangeMin[0]=-10:CutRangeMax[0]=10:VarProp[1]=FMax:EffSel:Steps=30:Cycles=3:PopSize=400:SC_steps=10:SC_rate=5:SC_factor=0.95" )
if "CutsSA" in mlist:
factory.BookMethod( dataloader, TMVA.Types.kCuts, "CutsSA",
"!H:!V:FitMethod=SA:EffSel:MaxCalls=150000:KernelTemp=IncAdaptive:InitialTemp=1e+6:MinTemp=1e-6:Eps=1e-10:UseDefaultScale" )
# Likelihood ("naive Bayes estimator")
if "Likelihood" in mlist:
factory.BookMethod( dataloader, TMVA.Types.kLikelihood, "Likelihood",
"H:!V:!TransformOutput:PDFInterpol=Spline2:NSmoothSig[0]=20:NSmoothBkg[0]=20:NSmoothBkg[1]=10:NSmooth=1:NAvEvtPerBin=50" )
# Decorrelated likelihood
if "LikelihoodD" in mlist:
factory.BookMethod( dataloader, TMVA.Types.kLikelihood, "LikelihoodD",
"!H:!V:TransformOutput:PDFInterpol=Spline2:NSmoothSig[0]=20:NSmoothBkg[0]=20:NSmooth=5:NAvEvtPerBin=50:VarTransform=Decorrelate" )
# PCA-transformed likelihood
if "LikelihoodPCA" in mlist:
factory.BookMethod( dataloader, TMVA.Types.kLikelihood, "LikelihoodPCA",
"!H:!V:!TransformOutput:PDFInterpol=Spline2:NSmoothSig[0]=20:NSmoothBkg[0]=20:NSmooth=5:NAvEvtPerBin=50:VarTransform=PCA" )
# Use a kernel density estimator to approximate the PDFs
if "LikelihoodKDE" in mlist:
factory.BookMethod( dataloader, TMVA.Types.kLikelihood, "LikelihoodKDE",
"!H:!V:!TransformOutput:PDFInterpol=KDE:KDEtype=Gauss:KDEiter=Adaptive:KDEFineFactor=0.3:KDEborder=None:NAvEvtPerBin=50" )
# Use a variable-dependent mix of splines and kernel density estimator
if "LikelihoodMIX" in mlist:
factory.BookMethod( dataloader, TMVA.Types.kLikelihood, "LikelihoodMIX",
"!H:!V:!TransformOutput:PDFInterpolSig[0]=KDE:PDFInterpolBkg[0]=KDE:PDFInterpolSig[1]=KDE:PDFInterpolBkg[1]=KDE:PDFInterpolSig[2]=Spline2:PDFInterpolBkg[2]=Spline2:PDFInterpolSig[3]=Spline2:PDFInterpolBkg[3]=Spline2:KDEtype=Gauss:KDEiter=Nonadaptive:KDEborder=None:NAvEvtPerBin=50" )
# Test the multi-dimensional probability density estimator
# here are the options strings for the MinMax and RMS methods, respectively:
# "!H:!V:VolumeRangeMode=MinMax:DeltaFrac=0.2:KernelEstimator=Gauss:GaussSigma=0.3" );
# "!H:!V:VolumeRangeMode=RMS:DeltaFrac=3:KernelEstimator=Gauss:GaussSigma=0.3" );
if "PDERS" in mlist:
# DEFAULT
# factory.BookMethod( dataloader, TMVA.Types.kPDERS, "PDERS",
# "!H:!V:NormTree=T:VolumeRangeMode=Adaptive:KernelEstimator=Gauss:GaussSigma=0.3:NEventsMin=400:NEventsMax=600" )
# CHOOSE RIGID VOLUME SO IT DOESNT TAKE SO LONG
factory.BookMethod( dataloader, TMVA.Types.kPDERS, "PDERS",
"!H:!V:NormTree=T:VolumeRangeMode=Unscaled:KernelEstimator=Gauss:GaussSigma=0.3:NEventsMin=400:NEventsMax=600" )
if "PDERSD" in mlist:
factory.BookMethod( dataloader, TMVA.Types.kPDERS, "PDERSD",
"!H:!V:VolumeRangeMode=Adaptive:KernelEstimator=Gauss:GaussSigma=0.3:NEventsMin=400:NEventsMax=600:VarTransform=Decorrelate" )
if "PDERSPCA" in mlist:
factory.BookMethod( dataloader, TMVA.Types.kPDERS, "PDERSPCA",
"!H:!V:VolumeRangeMode=Adaptive:KernelEstimator=Gauss:GaussSigma=0.3:NEventsMin=400:NEventsMax=600:VarTransform=PCA" )
# Multi-dimensional likelihood estimator using self-adapting phase-space binning
if "PDEFoam" in mlist:
factory.BookMethod( dataloader, TMVA.Types.kPDEFoam, "PDEFoam",
"!H:!V:SigBgSeparate=F:TailCut=0.001:VolFrac=0.0666:nActiveCells=500:nSampl=2000:nBin=5:Nmin=100:Kernel=None:Compress=T" )
if "PDEFoamBoost" in mlist:
factory.BookMethod( dataloader, TMVA.Types.kPDEFoam, "PDEFoamBoost",
"!H:!V:Boost_Num=30:Boost_Transform=linear:SigBgSeparate=F:MaxDepth=4:UseYesNoCell=T:DTLogic=MisClassificationError:FillFoamWithOrigWeights=F:TailCut=0:nActiveCells=500:nBin=20:Nmin=400:Kernel=None:Compress=T" )
# K-Nearest Neighbour classifier (KNN)
if "KNN" in mlist:
factory.BookMethod( dataloader, TMVA.Types.kKNN, "KNN",
"H:nkNN=20:ScaleFrac=0.8:SigmaFact=1.0:Kernel=Gaus:UseKernel=F:UseWeight=T:!Trim" )
# H-Matrix (chi2-squared) method
if "HMatrix" in mlist:
factory.BookMethod( dataloader, TMVA.Types.kHMatrix, "HMatrix", "!H:!V" )
# Linear discriminant (same as Fisher discriminant)
if "LD" in mlist:
factory.BookMethod( dataloader, TMVA.Types.kLD, "LD", "H:!V:VarTransform=None:CreateMVAPdfs:PDFInterpolMVAPdf=Spline2:NbinsMVAPdf=50:NsmoothMVAPdf=10" )
# Fisher discriminant (same as LD)
if "Fisher" in mlist:
factory.BookMethod( dataloader, TMVA.Types.kFisher, "Fisher", "H:!V:Fisher:CreateMVAPdfs:PDFInterpolMVAPdf=Spline2:NbinsMVAPdf=50:NsmoothMVAPdf=10" )
# Fisher with Gauss-transformed input variables
if "FisherG" in mlist:
factory.BookMethod( dataloader, TMVA.Types.kFisher, "FisherG", "H:!V:VarTransform=Gauss" )
# Composite classifier: ensemble (tree) of boosted Fisher classifiers
if "BoostedFisher" in mlist:
factory.BookMethod( dataloader, TMVA.Types.kFisher, "BoostedFisher",
"H:!V:Boost_Num=20:Boost_Transform=log:Boost_Type=AdaBoost:Boost_AdaBoostBeta=0.2" )
# Function discrimination analysis (FDA) -- test of various fitters - the recommended one is Minuit (or GA or SA)
if "FDA_MC" in mlist:
factory.BookMethod( dataloader, TMVA.Types.kFDA, "FDA_MC",
"H:!V:Formula=(0)+(1)*x0+(2)*x1+(3)*x2+(4)*x3:ParRanges=(-1,1)(-10,10);(-10,10);(-10,10);(-10,10):FitMethod=MC:SampleSize=100000:Sigma=0.1" );
if "FDA_GA" in mlist:
factory.BookMethod( dataloader, TMVA.Types.kFDA, "FDA_GA",
"H:!V:Formula=(0)+(1)*x0+(2)*x1+(3)*x2+(4)*x3:ParRanges=(-1,1)(-10,10);(-10,10);(-10,10);(-10,10):FitMethod=GA:PopSize=300:Cycles=3:Steps=20:Trim=True:SaveBestGen=1" );
if "FDA_SA" in mlist:
factory.BookMethod( dataloader, TMVA.Types.kFDA, "FDA_SA",
"H:!V:Formula=(0)+(1)*x0+(2)*x1+(3)*x2+(4)*x3:ParRanges=(-1,1)(-10,10);(-10,10);(-10,10);(-10,10):FitMethod=SA:MaxCalls=15000:KernelTemp=IncAdaptive:InitialTemp=1e+6:MinTemp=1e-6:Eps=1e-10:UseDefaultScale" );
if "FDA_MT" in mlist:
factory.BookMethod( dataloader, TMVA.Types.kFDA, "FDA_MT",
"H:!V:Formula=(0)+(1)*x0+(2)*x1+(3)*x2+(4)*x3:ParRanges=(-1,1)(-10,10);(-10,10);(-10,10);(-10,10):FitMethod=MINUIT:ErrorLevel=1:PrintLevel=-1:FitStrategy=2:UseImprove:UseMinos:SetBatch" );
if "FDA_GAMT" in mlist:
factory.BookMethod( dataloader, TMVA.Types.kFDA, "FDA_GAMT",
"H:!V:Formula=(0)+(1)*x0+(2)*x1+(3)*x2+(4)*x3:ParRanges=(-1,1)(-10,10);(-10,10);(-10,10);(-10,10):FitMethod=GA:Converger=MINUIT:ErrorLevel=1:PrintLevel=-1:FitStrategy=0:!UseImprove:!UseMinos:SetBatch:Cycles=1:PopSize=5:Steps=5:Trim" );
if "FDA_MCMT" in mlist:
factory.BookMethod( dataloader, TMVA.Types.kFDA, "FDA_MCMT",
"H:!V:Formula=(0)+(1)*x0+(2)*x1+(3)*x2+(4)*x3:ParRanges=(-1,1)(-10,10);(-10,10);(-10,10);(-10,10):FitMethod=MC:Converger=MINUIT:ErrorLevel=1:PrintLevel=-1:FitStrategy=0:!UseImprove:!UseMinos:SetBatch:SampleSize=20" );
# TMVA ANN: MLP (recommended ANN) -- all ANNs in TMVA are Multilayer Perceptrons
if "MLP" in mlist:
factory.BookMethod( dataloader, TMVA.Types.kMLP, "MLP", "H:!V:NeuronType=tanh:VarTransform=N:NCycles=600:HiddenLayers=N+5:TestRate=5:!UseRegulator" )
if "MLPBFGS" in mlist:
factory.BookMethod( dataloader, TMVA.Types.kMLP, "MLPBFGS", "H:!V:NeuronType=tanh:VarTransform=N:NCycles=600:HiddenLayers=N+5:TestRate=5:TrainingMethod=BFGS:!UseRegulator" )
if "MLPBNN" in mlist:
factory.BookMethod( dataloader, TMVA.Types.kMLP, "MLPBNN", "H:!V:NeuronType=tanh:VarTransform=N:NCycles=600:HiddenLayers=N+5:TestRate=5:TrainingMethod=BFGS:UseRegulator" ) # BFGS training with bayesian regulators
# CF(Clermont-Ferrand)ANN
if "CFMlpANN" in mlist:
factory.BookMethod( dataloader, TMVA.Types.kCFMlpANN, "CFMlpANN", "!H:!V:NCycles=2000:HiddenLayers=N+1,N" ) # n_cycles:#nodes:#nodes:...
# Tmlp(Root)ANN
if "TMlpANN" in mlist:
factory.BookMethod( dataloader, TMVA.Types.kTMlpANN, "TMlpANN", "!H:!V:NCycles=200:HiddenLayers=N+1,N:LearningMethod=BFGS:ValidationFraction=0.3" ) # n_cycles:#nodes:#nodes:...
# Support Vector Machine
if "SVM" in mlist:
factory.BookMethod( dataloader, TMVA.Types.kSVM, "SVM", "Gamma=0.25:Tol=0.001:VarTransform=Norm" )
# Boosted Decision Trees
if "BDTG" in mlist:
factory.BookMethod( dataloader, TMVA.Types.kBDT, "BDTG",
"!H:!V:NTrees=1000:MinNodeSize=1.5%:BoostType=Grad:Shrinkage=0.10:UseBaggedBoost:BaggedSampleFraction=0.5:nCuts=20:MaxDepth=2" )
if "BDT" in mlist:
factory.BookMethod( dataloader, TMVA.Types.kBDT, "BDT",
"!H:!V:NTrees=850:MinNodeSize=2.5%:MaxDepth=3:BoostType=AdaBoost:AdaBoostBeta=0.5:UseBaggedBoost:BaggedSampleFraction=0.5:SeparationType=GiniIndex:nCuts=20" )
if "BDTB" in mlist:
factory.BookMethod( dataloader, TMVA.Types.kBDT, "BDTB",
"!H:!V:NTrees=400:BoostType=Bagging:SeparationType=GiniIndex:nCuts=20" )
if "BDTD" in mlist:
factory.BookMethod( dataloader, TMVA.Types.kBDT, "BDTD",
"!H:!V:NTrees=400:MinNodeSize=5%:MaxDepth=3:BoostType=AdaBoost:SeparationType=GiniIndex:nCuts=20:VarTransform=Decorrelate" )
# RuleFit -- TMVA implementation of Friedman's method
if "RuleFit" in mlist:
factory.BookMethod( dataloader, TMVA.Types.kRuleFit, "RuleFit",
"H:!V:RuleFitModule=RFTMVA:Model=ModRuleLinear:MinImp=0.001:RuleMinDist=0.001:NTrees=20:fEventsMin=0.01:fEventsMax=0.5:GDTau=-1.0:GDTauPrec=0.01:GDStep=0.01:GDNSteps=10000:GDErrScale=1.02" )
# --------------------------------------------------------------------------------------------------
# ---- Now you can tell the factory to train, test, and evaluate the MVAs.
# Train MVAs
factory.TrainAllMethods()
# Test MVAs
factory.TestAllMethods()
# Evaluate MVAs
factory.EvaluateAllMethods()
# Save the output.
outputFile.Close()
print "=== wrote root file %s\n" % OutFileName
print "=== TMVAClassification is done!\n"
def main():
try:
# retrive command line options
shortopts = "m:i:n:d:k:l:t:o:vh?"
longopts = ["methods=", "inputfile=", "nTrees=", "maxDepth=", "mass=", "varListKey=", "inputtrees=", "outputfile=", "verbose", "help", "usage"]
opts, args = getopt.getopt( sys.argv[1:], shortopts, longopts )
except getopt.GetoptError:
print "ERROR: unknown options in argument %s" % sys.argv[1:]
usage()
sys.exit(1)
infname = DEFAULT_INFNAME
treeNameSig = DEFAULT_TREESIG
treeNameBkg = DEFAULT_TREEBKG
outfname = DEFAULT_OUTFNAME
methods = DEFAULT_METHODS
nTrees = DEFAULT_NTREES
mDepth = DEFAULT_MDEPTH
mass = DEFAULT_MASS
varListKey = DEFAULT_VARLISTKEY
verbose = True
for o, a in opts:
if o in ("-?", "-h", "--help", "--usage"):
usage()
sys.exit(0)
elif o in ("-m", "--methods"):
methods = a
elif o in ("-d", "--maxDepth"):
mDepth = a
elif o in ("-k", "--mass"):
mass = a
elif o in ("-l", "--varListKey"):
varListKey = a
elif o in ("-i", "--inputfile"):
infname = a
elif o in ("-n", "--nTrees"):
nTrees = a
elif o in ("-o", "--outputfile"):
outfname = a
elif o in ("-t", "--inputtrees"):
a.strip()
trees = a.rsplit( ' ' )
trees.sort()
trees.reverse()
if len(trees)-trees.count('') != 2:
print "ERROR: need to give two trees (each one for signal and background)"
print trees
sys.exit(1)
treeNameSig = trees[0]
treeNameBkg = trees[1]
elif o in ("-v", "--verbose"):
verbose = True
varList = varsList.varList[varListKey]
nVars = str(len(varList))+'vars'
Note=''+methods+'_'+varListKey+'_'+nVars+'_mDepth'+mDepth
outfname = "dataset/weights/TMVA_"+Note+".root"
# Print methods
mlist = methods.replace(' ',',').split(',')
print "=== TMVAClassification: use method(s)..."
for m in mlist:
if m.strip() != '':
print "=== - <%s>" % m.strip()
# Import ROOT classes
# check ROOT version, give alarm if 5.18
if gROOT.GetVersionCode() >= 332288 and gROOT.GetVersionCode() < 332544:
print "*** You are running ROOT version 5.18, which has problems in PyROOT such that TMVA"
print "*** does not run properly (function calls with enums in the argument are ignored)."
print "*** Solution: either use CINT or a C++ compiled version (see TMVA/macros or TMVA/examples),"
print "*** or use another ROOT version (e.g., ROOT 5.19)."
sys.exit(1)
# Output file
outputFile = TFile( outfname, 'RECREATE' )
factory = TMVA.Factory( "TMVAClassification", outputFile,
"!V:!Silent:Color:DrawProgressBar:Transformations=I;:AnalysisType=Classification" )
factory.SetVerbose( verbose )
(TMVA.gConfig().GetIONames()).fWeightFileDir = "weights/"+Note
dataloader = TMVA.DataLoader('dataset')
for iVar in varList:
if iVar[0]=='NJets_singleLepCalc': dataloader.AddVariable(iVar[0],iVar[1],iVar[2],'I')
else: dataloader.AddVariable(iVar[0],iVar[1],iVar[2],'F')
inputDir = varsList.inputDir
infname = "TTTT_TuneCP5_PSweights_13TeV-amcatnlo-pythia8_hadd.root"
iFileSig = TFile.Open(inputDir+infname)
sigChain = iFileSig.Get("ljmet")
dataloader.AddSignalTree(sigChain)
bkg_list = []
bkg_trees_list = []
hist_list = []
weightsList = []
for i in range(len(varsList.bkg)):
bkg_list.append(TFile.Open(inputDir+varsList.bkg[i]))
print inputDir+varsList.bkg[i]
bkg_trees_list.append(bkg_list[i].Get("ljmet"))
bkg_trees_list[i].GetEntry(0)
if bkg_trees_list[i].GetEntries() == 0:
continue
dataloader.AddBackgroundTree( bkg_trees_list[i], 1)
signalWeight = 1
dataloader.SetSignalWeightExpression( weightStrS )
dataloader.SetBackgroundWeightExpression( weightStrB )
mycutSig = TCut( cutStrS )
mycutBkg = TCut( cutStrB )
dataloader.PrepareTrainingAndTestTree( mycutSig, mycutBkg, "nTrain_Signal=0:nTrain_Background=0:SplitMode=Random:NormMode=NumEvents:!V" )
kerasSetting = 'H:!V:VarTransform=G:FilenameModel=model.h5:NumEpochs=10:BatchSize=1028'
model = Sequential()
model.add(Dense(100, activation='relu', input_dim=53))
model.add((Dense(100, activation="relu")))
model.add((Dense(100, activation="relu")))
model.add((Dense(100, activation="relu")))
model.add((Dense(2, activation="sigmoid")))
# Set loss and optimizer
model.compile(loss='categorical_crossentropy', optimizer=Adam(), metrics=['accuracy',])
# Store model to file
model.save('model.h5')
model.summary()
if methods=="Keras": factory.BookMethod(dataloader, TMVA.Types.kPyKeras, "PyKeras",kerasSetting)
factory.TrainAllMethods()
factory.TestAllMethods()
factory.EvaluateAllMethods()
outputFile.Close()
# save plots:
print "DONE"
def main():
try:
# retrive command line options
shortopts = "m:i:n:d:k:l:t:o:vh?"
longopts = [
"methods=", "inputfile=", "nTrees=", "maxDepth=", "mass=",
"varListKey=", "inputtrees=", "outputfile=", "verbose", "help",
"usage"
]
opts, args = getopt.getopt(sys.argv[1:], shortopts, longopts)
except getopt.GetoptError:
# print help information and exit:
print "ERROR: unknown options in argument %s" % sys.argv[1:]
usage()
sys.exit(1)
infname = DEFAULT_INFNAME
treeNameSig = DEFAULT_TREESIG
treeNameBkg = DEFAULT_TREEBKG
outfname = DEFAULT_OUTFNAME
methods = DEFAULT_METHODS
nTrees = DEFAULT_NTREES
mDepth = DEFAULT_MDEPTH
mass = DEFAULT_MASS
varListKey = DEFAULT_VARLISTKEY
verbose = True
for o, a in opts:
if o in ("-?", "-h", "--help", "--usage"):
usage()
sys.exit(0)
elif o in ("-m", "--methods"):
methods = a
elif o in ("-d", "--maxDepth"):
mDepth = a
elif o in ("-k", "--mass"):
mass = a
elif o in ("-l", "--varListKey"):
varListKey = a
elif o in ("-i", "--inputfile"):
infname = a
elif o in ("-n", "--nTrees"):
nTrees = a
elif o in ("-o", "--outputfile"):
outfname = a
elif o in ("-t", "--inputtrees"):
a.strip()
trees = a.rsplit(' ')
trees.sort()
trees.reverse()
if len(trees) - trees.count('') != 2:
print "ERROR: need to give two trees (each one for signal and background)"
print trees
sys.exit(1)
treeNameSig = trees[0]
treeNameBkg = trees[1]
elif o in ("-v", "--verbose"):
verbose = True
varList = varsList.varList[varListKey]
nVars = str(len(varList)) + 'vars'
Note = methods + '_' + varListKey + '_' + nVars + '_mDepth' + mDepth + '_M' + mass
outfname = "weights/TMVA_" + Note + ".root"
# Print methods
mlist = methods.replace(' ', ',').split(',')
print "=== TMVAClassification: use method(s)..."
for m in mlist:
if m.strip() != '':
print "=== - <%s>" % m.strip()
# Import ROOT classes
from ROOT import gSystem, gROOT, gApplication, TFile, TTree, TCut
# check ROOT version, give alarm if 5.18
if gROOT.GetVersionCode() >= 332288 and gROOT.GetVersionCode() < 332544:
print "*** You are running ROOT version 5.18, which has problems in PyROOT such that TMVA"
print "*** does not run properly (function calls with enums in the argument are ignored)."
print "*** Solution: either use CINT or a C++ compiled version (see TMVA/macros or TMVA/examples),"
print "*** or use another ROOT version (e.g., ROOT 5.19)."
sys.exit(1)
# Logon not automatically loaded through PyROOT (logon loads TMVA library) load also GUI
# gROOT.SetMacroPath( "./" )
# gROOT.Macro ( "./TMVAlogon.C" )
# gROOT.LoadMacro ( "./TMVAGui.C" )
# Import TMVA classes from ROOT
from ROOT import TMVA
# Output file
outputFile = TFile(outfname, 'RECREATE')
# Create instance of TMVA factory (see TMVA/macros/TMVAClassification.C for more factory options)
# All TMVA output can be suppressed by removing the "!" (not) in
# front of the "Silent" argument in the option string
# factory = TMVA.Factory( "TMVAClassification", outputFile,
# "!V:!Silent:Color:DrawProgressBar:Transformations=I;D;P;G,D:AnalysisType=Classification" )
factory = TMVA.Factory(
"TMVAClassification", outputFile,
"!V:!Silent:Color:DrawProgressBar:Transformations=I;:AnalysisType=Classification"
)
# Set verbosity
factory.SetVerbose(verbose)
# If you wish to modify default settings
# (please check "src/Config.h" to see all available global options)
# gConfig().GetVariablePlotting()).fTimesRMS = 8.0
(TMVA.gConfig().GetIONames()).fWeightFileDir = "weights/" + Note
# Define the input variables that shall be used for the classifier training
# note that you may also use variable expressions, such as: "3*var1/var2*abs(var3)"
# [all types of expressions that can also be parsed by TTree::Draw( "expression" )]
for iVar in varList:
if iVar[0] == 'NJets_singleLepCalc':
factory.AddVariable(iVar[0], iVar[1], iVar[2], 'I')
else:
factory.AddVariable(iVar[0], iVar[1], iVar[2], 'F')
# You can add so-called "Spectator variables", which are not used in the MVA training,
# but will appear in the final "TestTree" produced by TMVA. This TestTree will contain the
# input variables, the response values of all trained MVAs, and the spectator variables
inputDir = varsList.inputDir
print 'mass point ' + mass
infname = "ChargedHiggs_HplusTB_HplusToTB_M-%s_13TeV_amcatnlo_pythia8_hadd.root" % (
mass)
iFileSig = TFile.Open(inputDir + infname)
sigChain = iFileSig.Get("ljmet")
# os.exits(1)
#BDT machinary
factory.AddSignalTree(sigChain)
bkg_list = []
bkg_trees_list = []
hist_list = []
weightsList = []
for i in range(len(varsList.bkg)):
bkg_list.append(TFile.Open(inputDir + varsList.bkg[i]))
print inputDir + varsList.bkg[i]
bkg_trees_list.append(bkg_list[i].Get("ljmet"))
bkg_trees_list[i].GetEntry(0)
if bkg_trees_list[i].GetEntries() == 0:
continue
factory.AddBackgroundTree(bkg_trees_list[i], 1)
signalWeight = 1 #0.0159/sigChain.GetEntries() #xs (pb)
# ====== register trees ====================================================
# To give different trees for training and testing, do as follows:
# factory.AddSignalTree( signalTrainingTree, signalTrainWeight, "Training" )
# factory.AddSignalTree( signalTestTree, signalTestWeight, "Test" )
# Use the following code instead of the above two or four lines to add signal and background
# training and test events "by hand"
# NOTE that in this case one should not give expressions (such as "var1+var2") in the input
# variable definition, but simply compute the expression before adding the event
#
# # --- begin ----------------------------------------------------------
#
# ... *** please lookup code in TMVA/macros/TMVAClassification.C ***
#
# # --- end ------------------------------------------------------------
#
# ====== end of register trees ==============================================
# Set individual event weights (the variables must exist in the original TTree)
# for signal : factory.SetSignalWeightExpression ("weight1*weight2");
# for background: factory.SetBackgroundWeightExpression("weight1*weight2");
#factory.SetBackgroundWeightExpression( "weight" )
factory.SetSignalWeightExpression(weightStrS)
factory.SetBackgroundWeightExpression(weightStrB)
# Apply additional cuts on the signal and background sample.
# example for cut: mycut = TCut( "abs(var1)<0.5 && abs(var2-0.5)<1" )
mycutSig = TCut(cutStrS)
mycutBkg = TCut(cutStrB)
# Here, the relevant variables are copied over in new, slim trees that are
# used for TMVA training and testing
# "SplitMode=Random" means that the input events are randomly shuffled before
# splitting them into training and test samples
factory.PrepareTrainingAndTestTree(
mycutSig,
mycutBkg,
# "nTrain_Signal=0:nTrain_Background=0:nTest_Signal=10:nTest_Background=100:SplitMode=Random:NormMode=NumEvents:!V" )
"nTrain_Signal=0:nTrain_Background=0:SplitMode=Random:NormMode=NumEvents:!V"
)
# --------------------------------------------------------------------------------------------------
# ---- Book MVA methods
#
# please lookup the various method configuration options in the corresponding cxx files, eg:
# src/MethoCuts.cxx, etc, or here: http://tmva.sourceforge.net/optionRef.html
# it is possible to preset ranges in the option string in which the cut optimisation should be done:
# "...:CutRangeMin[2]=-1:CutRangeMax[2]=1"...", where [2] is the third input variable
# Cut optimisation
# bdtSetting for "BDT"
bdtSetting = '!H:!V:NTrees=%s:MaxDepth=%s' % (nTrees, mDepth)
bdtSetting += ':MinNodeSize=2.5%:BoostType=AdaBoost:AdaBoostBeta=0.5:UseBaggedBoost:BaggedSampleFraction=0.5:SeparationType=GiniIndex:nCuts=20'
bdtSetting += ':IgnoreNegWeightsInTraining=True'
# bdtSetting for "BDTMitFisher"
bdtFSetting = '!H:!V:NTrees=%s' % nTrees
bdtFSetting += ':MinNodeSize=2.5%:UseFisherCuts:MaxDepth=3:BoostType=AdaBoost:AdaBoostBeta=0.5:SeparationType=GiniIndex:nCuts=20'
bdtFSetting += ':IgnoreNegWeightsInTraining=True'
# bdtSetting for "BDTG"
bdtGSetting = '!H:!V:NTrees=%s:MaxDepth=%s' % (nTrees, mDepth)
bdtGSetting += ':MinNodeSize=2.5%:BoostType=Grad:Shrinkage=0.10:UseBaggedBoost:BaggedSampleFraction=0.5:nCuts=20'
bdtGSetting += ':Pray' #Pray takes into account the effect of negative bins in BDTG
#bdtGSetting += ':IgnoreNegWeightsInTraining=True'
# bdtSetting for "BDTB"
bdtBSetting = '!H:!V:NTrees=%s' % nTrees
bdtBSetting += ':MinNodeSize=2.5%:BoostType=Bagging:SeparationType=GiniIndex:nCuts=20'
bdtBSetting += ':IgnoreNegWeightsInTraining=True'
# bdtSetting for "BDTD"
bdtDSetting = '!H:!V:NTrees=%s' % nTrees
bdtDSetting += ':MinNodeSize=2.5%:MaxDepth=3:BoostType=AdaBoost:SeparationType=GiniIndex:nCuts=20:VarTransform=Decorrelate'
bdtDSetting += ':IgnoreNegWeightsInTraining=True'
#Note also that explicitly setting *nEventsMin* so far OVERWRITES the option recomeded ^[[0m
#BOOKING AN ALGORITHM
if methods == "BDT": factory.BookMethod(TMVA.Types.kBDT, "BDT", bdtSetting)
if methods == "BDTG":
factory.BookMethod(TMVA.Types.kBDT, "BDTG", bdtGSetting)
if methods == "BDTMitFisher":
factory.BookMethod(TMVA.Types.kBDT, "BDTMitFisher", bdtFSetting)
if methods == "BDTB":
factory.BookMethod(TMVA.Types.kBDT, "BDTB", bdtBSetting)
if methods == "BDTD":
factory.BookMethod(TMVA.Types.kBDT, "BDTD", bdtDSetting)
# --------------------------------------------------------------------------------------------------
# ---- Now you can tell the factory to train, test, and evaluate the MVAs.
# Train MVAs
factory.TrainAllMethods()
# Test MVAs
factory.TestAllMethods()
# Evaluate MVAs
factory.EvaluateAllMethods()
# Save the output.
outputFile.Close()
#
# print "=== wrote root file %s\n" % outfname
# print "=== TMVAClassification is done!\n"
# save plots:
os.chdir('weights/' + Note)
#TMVA.mvaeffs( "../../"+outfname ) #Classifier Cut Efficiencies
gROOT.SetBatch(1)
TMVA.efficiencies(
"../../" + outfname
) #Classifier Background Rejection vs Signal Efficiency (ROC curve)
#TMVA.efficiencies( "weights/TMVA_BDTG_APR9_33vars_mDepth3_MLow.root", 3 ) #Classifier 1/(Backgr. Efficiency) vs Signal Efficiency (ROC curve)
TMVA.mvas("../../" + outfname,
0) #Classifier Output Distributions (test sample)
TMVA.correlations(
"../../" + outfname) #Input Variable Linear Correlation Coefficients
TMVA.variables("../../" + outfname) #Input variables (training sample)
#TMVA.mvas( "../../"+outfname ) #Classifier Output Distributions (test and training samples superimposed)
#gROOT.ProcessLine( ".x ../../mvas.C(\"%s\",3)" % ("../../"+outfname) ) #Classifier Output Distributions (test and training samples superimposed)
if not gROOT.IsBatch(): TMVA.TMVAGui("../../" + outfname)
# os.chdir('plots')
# try: os.system('convert CorrelationMatrixS.eps CorrelationMatrixS_'+Note+'.png')
# except: pass
# try: os.system('convert CorrelationMatrixB.eps CorrelationMatrixB_'+Note+'.png')
# except: pass
# #try: os.system('convert invBeffvsSeff.eps invBeffvsSeff_'+Note+'.png')
# #except: pass
# try: os.system('convert mva_'+Note.split('_')[0]+'.eps mva_'+Note+'.png')
# except: pass
# try: os.system('convert mvaeffs_'+Note.split('_')[0]+'.eps mvaeffs_'+Note+'.png')
# except: pass
# try: os.system('convert overtrain_'+Note.split('_')[0]+'.eps overtrain_'+Note+'.png')
# except: pass
# try: os.system('convert rejBvsS.eps rejBvsS_'+Note+'.png')
# except: pass
# try: os.system('convert variables_id_c1.eps variables_id_c1_'+Note+'.png')
# except: pass
# try: os.system('convert variables_id_c2.eps variables_id_c2_'+Note+'.png')
# except: pass
# try: os.system('convert variables_id_c3.eps variables_id_c3_'+Note+'.png')
# except: pass
# try: os.system('convert variables_id_c4.eps variables_id_c4_'+Note+'.png')
# except: pass
# try: os.system('convert variables_id_c5.eps variables_id_c5_'+Note+'.png')
# except: pass
# try: os.system('convert variables_id_c6.eps variables_id_c6_'+Note+'.png')
# except: pass
# os.system('rm *.eps')
print "DONE"
def main(o, args):
# Import TMVA classes from ROOT
from ROOT import TMVA, TFile, TCut
print o
# Output file
outputFile = TFile(o.outfile % {"label": o.label}, 'RECREATE')
atype = "Classification"
if hasattr(o, "type"):
atype = str(o.type)
factory = TMVA.Factory(
"TMVAClassification", outputFile,
"!V:!Silent:!Color:!DrawProgressBar:Transformations=I:AnalysisType=%s"
% atype)
# Set verbosity
factory.SetVerbose(o.verbose)
TMVA.Config.Instance().GetIONames().fWeightFileDir = o.weightsdir
# variables
if type(o.variables) == str:
o.variables = [
v.lstrip().rstrip() for v in o.variables.split(":") if v != ""
]
allvars = ""
for v in o.variables:
factory.AddVariable(str(v))
if allvars != "": allvars += ":"
allvars += v.split(":=")[0].lstrip(" ").rstrip(" ")
print "variables %s" % allvars
print o.spectators
for s in o.spectators:
if not s in o.variables:
factory.AddSpectator(str(s))
# categories and sub categories
categories = []
subcategories = []
if hasattr(o, "subcategories") and len(o.subcategories) > 0:
subcategories = o.subcategories[0]
for sc in o.subcategories[1:]:
subcategories = map(
lambda x: (TCut(x[0][0]) * TCut(x[1][0]), "%s_%s" %
(x[0][1], x[1][1])),
itertools.product(subcategories, sc))
for cut, name, vars in o.categories:
myvars = allvars
if vars != "":
for v in vars.split(":"):
myvars = myvars.replace(v, "").replace("::", ":")
myvars = myvars.rstrip(":")
vars = str(myvars)
print vars
if len(subcategories) > 0:
for subcut, subname in subcategories:
if subname == "":
subname = subname.replace(" ", "").replace(
">", "_gt_").replace("<", "_lt_").replace(
"=", "_eq_").replace("&", "_and_")
fullname = "%s_%s" % (name, subname)
categories.append(
(TCut(cut) * TCut(subcut), str(fullname), vars))
else:
categories.append((TCut(cut), str(name), vars))
# load tree
selection = TCut(o.selection)
for evclass, info in o.classes.iteritems():
samples = info["samples"]
for name, weight, cut, ttype in samples:
tcut = TCut(cut) * selection
factory.AddTree(mkChain(getListOfFiles(o.indir, o.files), name),
str(evclass), float(weight), tcut, int(ttype))
# weights
if "weight" in info:
weight = info["weight"]
factory.AddSpectator(str("%s_wei := %s" % (evclass, weight)))
factory.SetWeightExpression(str(weight), str(evclass))
else:
factory.SetWeightExpression("1.", str(evclass))
# "SplitMode=Random" means that the input events are randomly shuffled before
# splitting them into training and test samples
factory.PrepareTrainingAndTestTree(
TCut(""), "SplitMode=Random:NormMode=NumEvents:!V")
# --------------------------------------------------------------------------------------------------
# Fisher discriminant (same as LD)
defaultSettings = {
"BDT":
"!H:!V:!CreateMVAPdfs:BoostType=Grad:UseBaggedGrad"
":GradBaggingFraction=0.6:SeparationType=GiniIndex:nCuts=20:NNodesMax=5"
":Shrinkage=0.3:NTrees=1000",
"Cuts":
"!H:!V:FitMethod=MC:EffSel:SampleSize=200000:VarProp=FSmart"
}
if "FisherD" in o.methods:
mname = "FisherD%s" % o.label
fcats = factory.BookMethod(TMVA.Types.kCategory, mname)
for cut, name, vars in categories:
print "booking sub-category classifier : %s %s %s" % (cut, name,
vars)
fcats.AddMethod(cut, vars, TMVA.Types.kFisher,
"%s_%s" % (mname, name),
"!H:!V:Fisher:!CreateMVAPdfs:VarTransform=D")
if "Fisher" in o.methods:
mname = "Fisher%s" % o.label
fcats = factory.BookMethod(TMVA.Types.kCategory, mname)
for cut, name, vars in categories:
print "booking sub-category classifier : %s %s %s" % (cut, name,
vars)
fcats.AddMethod(cut, vars, TMVA.Types.kFisher,
"%s_%s" % (mname, name),
"!H:!V:Fisher:!CreateMVAPdfs")
if "Likelihood" in o.methods:
mname = "Likelihood%s" % o.label
fcats = factory.BookMethod(TMVA.Types.kCategory, mname)
for cut, name, vars in categories:
print "booking sub-category classifier : %s %s %s" % (cut, name,
vars)
fcats.AddMethod(
cut, vars, TMVA.Types.kLikelihood, "%s_%s" % (mname, name),
"!H:!V:!CreateMVAPdfs:!TransformOutput:PDFInterpol=KDE:KDEtype=Gauss:KDEiter=Adaptive:KDEFineFactor=0.3:KDEborder=None:NAvEvtPerBin=150"
)
if "LikelihoodD" in o.methods:
mname = "LikelihoodD%s" % o.label
fcats = factory.BookMethod(TMVA.Types.kCategory, mname)
for cut, name, vars in categories:
print "booking sub-category classifier : %s %s %s" % (cut, name,
vars)
fcats.AddMethod(
cut, vars, TMVA.Types.kLikelihood, "%s_%s" % (mname, name),
"!H:!V:!CreateMVAPdfs:!TransformOutput:VarTransform=D:PDFInterpol=KDE:KDEtype=Gauss:KDEiter=Adaptive:KDEFineFactor=0.3:KDEborder=None:NAvEvtPerBin=150"
)
if "BDT" in o.methods:
mname = str("BDT%s" % o.label)
settings = defaultSettings["BDT"]
if hasattr(o, "settings") and "BDT" in o.settings:
settings = str(o.settings["BDT"])
print mname, settings
if len(categories) == 0:
cats = factory.BookMethod(TMVA.Types.kBDT, mname, settings)
else:
cats = factory.BookMethod(TMVA.Types.kCategory, mname)
for cut, name, vars in categories:
print "booking sub-category classifier : %s %s %s" % (
cut, name, vars)
cats.AddMethod(cut, vars, TMVA.Types.kBDT,
"%s_%s" % (mname, name), settings)
if "Cuts" in o.methods:
mname = "Cuts%s" % o.label
settings = defaultSettings["Cuts"]
if hasattr(o, "settings") and "Cuts" in o.settings:
settings = str(o.settings["Cuts"])
if len(categories) == 0:
cats = factory.BookMethod(TMVA.Types.kCuts, mname, settings)
else:
cats = factory.BookMethod(TMVA.Types.kCategory, mname)
for cut, name, vars in categories:
print "booking sub-category classifier : %s %s %s" % (
cut, name, vars)
cats.AddMethod(cut, vars, TMVA.Types.kCuts,
"%s_%s" % (mname, name), settings)
# ---- Now you can tell the factory to train, test, and evaluate the MVAs.
if o.optimize:
print "Optimizing?"
factory.OptimizeAllMethods()
factory.TrainAllMethods()
factory.TestAllMethods()
factory.EvaluateAllMethods()
# Save the output.
outputFile.Close()
def main():
try:
# retrive command line options
shortopts = "w:m:i:j:f:g:t:o:a:vgh?"
longopts = ["weight_fold=", "methods=", "inputfilesig=", "inputfilebkg=", "friendinputfilesig=", "friendinputfilebkg=", "inputtrees=", "outputfile=", "verbose", "gui", "help", "usage"]
opts, args = getopt.getopt( sys.argv[1:], shortopts, longopts )
except getopt.GetoptError:
# print help information and exit:
print "ERROR: unknown options in argument %s" % sys.argv[1:]
usage()
sys.exit(1)
infnameSig = DEFAULT_INFNAMESIG
infnameBkg = DEFAULT_INFNAMEBKG
friendfnameSig = DEFAULT_FRIENDNAMESIG
friendfnameBkg = DEFAULT_FRIENDNAMEBKG
treeNameSig = DEFAULT_TREESIG
treeNameBkg = DEFAULT_TREEBKG
outfname = DEFAULT_OUTFNAME
methods = DEFAULT_METHODS
weight_fold = "weights"
verbose = False
gui = False
addedcuts = ""
for o, a in opts:
if o in ("-?", "-h", "--help", "--usage"):
usage()
sys.exit(0)
elif o in ("-w", "--weight_fold"):
weight_fold = a
elif o in ("-m", "--methods"):
methods = a
elif o in ("-i", "--inputfilesig"):
infnameSig = a
elif o in ("-j", "--inputfilebkg"):
infnameBkg = a
elif o in ("-f", "--friendinputfilesig"):
friendfnameSig = a
elif o in ("-g", "--friendinputfilebkg"):
friendfnameBkg = a
elif o in ("-o", "--outputfile"):
outfname = a
elif o in ("-a", "--addedcuts"):
addedcuts = a
elif o in ("-t", "--inputtrees"):
a.strip()
trees = a.rsplit( ' ' )
trees.sort()
trees.reverse()
if len(trees)-trees.count('') != 2:
print "ERROR: need to give two trees (each one for signal and background)"
print trees
sys.exit(1)
treeNameSig = trees[0]
treeNameBkg = trees[1]
elif o in ("-v", "--verbose"):
verbose = True
elif o in ("-g", "--gui"):
gui = True
# Print methods
mlist = methods.replace(' ',',').split(',')
print "=== TMVAClassification: use method(s)..."
for m in mlist:
if m.strip() != '':
print "=== - <%s>" % m.strip()
# Print the file
print "Using file " + infnameSig + " for signal..."
print "Using file " + infnameBkg + " for background..."
# Import ROOT classes
from ROOT import gSystem, gROOT, gApplication, TFile, TTree, TCut
# check ROOT version, give alarm if 5.18
print "ROOT version is " + str(gROOT.GetVersionCode())
if gROOT.GetVersionCode() >= 332288 and gROOT.GetVersionCode() < 332544:
print "*** You are running ROOT version 5.18, which has problems in PyROOT such that TMVA"
print "*** does not run properly (function calls with enums in the argument are ignored)."
print "*** Solution: either use CINT or a C++ compiled version (see TMVA/macros or TMVA/examples),"
print "*** or use another ROOT version (e.g., ROOT 5.19)."
sys.exit(1)
# Logon not automatically loaded through PyROOT (logon loads TMVA library) load also GUI
gROOT.SetMacroPath( "./" )
## SO I TAKE DEFAULT FORM ROOT# gROOT.Macro ( "./TMVAlogon.C" )
#! gROOT.LoadMacro ( "./TMVAGui.C" )
# Import TMVA classes from ROOT
from ROOT import TMVA
# Output file
outputFile = TFile( outfname, 'RECREATE' )
# Create instance of TMVA factory (see TMVA/macros/TMVAClassification.C for more factory options)
# All TMVA output can be suppressed by removing the "!" (not) in
# front of the "Silent" argument in the option string
factory = TMVA.Factory( "TMVAClassification", outputFile,
"!V:!Silent:Color:DrawProgressBar:Transformations=I;D;P;G,D:AnalysisType=Classification" )
# Set verbosity
factory.SetVerbose( verbose )
# If you wish to modify default settings
# (please check "src/Config.h" to see all available global options)
# gConfig().GetVariablePlotting()).fTimesRMS = 8.0
(TMVA.gConfig().GetIONames()).fWeightFileDir = weight_fold;
# Define the input variables that shall be used for the classifier training
# note that you may also use variable expressions, such as: "3*var1/var2*abs(var3)"
# [all types of expressions that can also be parsed by TTree::Draw( "expression" )]
factory.AddVariable( "dR_l1l2", "dR_l1l2", "", 'F' )
factory.AddVariable( "dR_b1b2", "dR_b1b2", "", 'F' )
factory.AddVariable( "dR_bl", "dR_bl", "", 'F' )
factory.AddVariable( "dR_l1l2b1b2", "dR_l1l2b1b2", "", 'F' )
factory.AddVariable( "MINdR_bl", "MINdR_bl", "", 'F' )
factory.AddVariable( "dphi_l1l2b1b2", "dphi_l1l2b1b2", "", 'F' )
factory.AddVariable( "mass_l1l2", "mass_l1l2", "", 'F' )
factory.AddVariable( "mass_b1b2", "mass_b1b2", "", 'F' )
factory.AddVariable( "mass_trans", "mass_trans", "", 'F' )
factory.AddVariable( "MT2", "MT2", "", 'F' )
factory.AddVariable( "pt_b1b2", "pt_b1b2", "", 'F' )
#factory.AddVariable( "MMC_h2mass_MaxBin", "MMC_h2mass_MaxBin", "", 'F' )
#factory.AddVariable( "MMC_h2mass_RMS", "MMC_h2mass_RMS", "", 'F' )
#factory.AddVariable( "MMC_h2mass_prob", "MMC_h2mass_prob", "", 'F' )
# You can add so-called "Spectator variables", which are not used in the MVA training,
# but will appear in the final "TestTree" produced by TMVA. This TestTree will contain the
# input variables, the response values of all trained MVAs, and the spectator variables
# factory.AddSpectator( "spec1:=var1*2", "Spectator 1", "units", 'F' )
# factory.AddSpectator( "spec2:=var1*3", "Spectator 2", "units", 'F' )
# Read input data
if gSystem.AccessPathName( infnameSig ) != 0 or gSystem.AccessPathName( infnameBkg ): gSystem.Exec( "wget http://root.cern.ch/files/" + infname )
inputSig = TFile.Open( infnameSig )
inputBkg = TFile.Open( infnameBkg )
# Get the signal and background trees for training
signal = inputSig.Get( treeNameSig )
background = inputBkg.Get( treeNameBkg )
##signal.AddFriend( "eleIDdir/isoT1 = eleIDdir/T1", friendfnameSig )
##background.AddFriend( "eleIDdir/isoT1 = eleIDdir/T1", friendfnameBkg )
# Global event weights (see below for setting event-wise weights)
signalWeight = 1.
backgroundWeight = 1.
#I don't think there's a general answer to this. The safest 'default'
#is to use the envent weight such that you have equal amounts of signal
#and background
#for the training, otherwise for example: if you look for a rare
#signal and you use the weight to scale the number of events according
#to the expected ratio of signal and background
#according to the luminosity... the classifier sees hardly any signal
#events and "thinks" .. Oh I just classify everything background and do
#a good job!
#
#One can try to 'optimize' the training a bit more in either 'high
#purity' or 'high efficiency' by choosing different weights, but as I
#said, there's no fixed rule. You'd have
#to 'try' and see if you get better restults by playing with the weights.
# ====== register trees ====================================================
#
# the following method is the prefered one:
# you can add an arbitrary number of signal or background trees
factory.AddSignalTree ( signal, signalWeight )
factory.AddBackgroundTree( background, backgroundWeight )
# To give different trees for training and testing, do as follows:
# factory.AddSignalTree( signalTrainingTree, signalTrainWeight, "Training" )
# factory.AddSignalTree( signalTestTree, signalTestWeight, "Test" )
# Use the following code instead of the above two or four lines to add signal and background
# training and test events "by hand"
# NOTE that in this case one should not give expressions (such as "var1+var2") in the input
# variable definition, but simply compute the expression before adding the event
#
# # --- begin ----------------------------------------------------------
#
# ... *** please lookup code in TMVA/macros/TMVAClassification.C ***
#
# # --- end ------------------------------------------------------------
#
# ====== end of register trees ==============================================
# Set individual event weights (the variables must exist in the original TTree)
# for signal : factory.SetSignalWeightExpression ("weight1*weight2");
# for background: factory.SetBackgroundWeightExpression("weight1*weight2");
# Apply additional cuts on the signal and background sample.
# example for cut: mycut = TCut( "abs(var1)<0.5 && abs(var2-0.5)<1" )
#mycutSig = TCut( "nu1and2_diBaxis_t>-900 && met_diBaxis_t>-900&& hasb1jet && hasb2jet && hasMET && hasGenMET && hasdRljet && hastwomuons" )
mycutSig = TCut( addedcuts )
#mycutBkg = TCut( "event_n%2!=0 && " + addedcuts )
mycutBkg = TCut( addedcuts )
#mycutBkg = TCut( "nu1and2_diBaxis_t>-900 && met_diBaxis_t>-900&& hasb1jet && hasb2jet && hasMET && hasGenMET && hasdRljet && hastwomuons" )
print mycutSig
# Here, the relevant variables are copied over in new, slim trees that are
# used for TMVA training and testing
# "SplitMode=Random" means that the input events are randomly shuffled before
# splitting them into training and test samples
factory.PrepareTrainingAndTestTree( mycutSig, mycutBkg,
"nTrain_Signal=0:nTrain_Background=0:SplitMode=Random:NormMode=NumEvents:!V" )
# --------------------------------------------------------------------------------------------------
# ---- Book MVA methods
#
# please lookup the various method configuration options in the corresponding cxx files, eg:
# src/MethoCuts.cxx, etc, or here: http://tmva.sourceforge.net/optionRef.html
# it is possible to preset ranges in the option string in which the cut optimisation should be done:
# "...:CutRangeMin[2]=-1:CutRangeMax[2]=1"...", where [2] is the third input variable
# Cut optimisation
if "Cuts" in mlist:
factory.BookMethod( TMVA.Types.kCuts, "Cuts",
"!H:!V:FitMethod=MC:EffSel:VarProp[0]=FMax:VarProp[1]=FMin" )
if "CutsD" in mlist:
factory.BookMethod( TMVA.Types.kCuts, "CutsD",
"!H:!V:FitMethod=MC:EffSel:SampleSize=200000:VarProp=FSmart:VarTransform=Decorrelate" )
if "CutsPCA" in mlist:
factory.BookMethod( TMVA.Types.kCuts, "CutsPCA",
"!H:!V:FitMethod=MC:EffSel:SampleSize=200000:VarProp=FSmart:VarTransform=PCA" )
if "CutsGA" in mlist:
factory.BookMethod( TMVA.Types.kCuts, "CutsGA",
"H:!V:FitMethod=GA:EffSel:Steps=30:Cycles=3:PopSize=400:SC_steps=10:SC_rate=5:SC_factor=0.95:VarProp[0]=FMin:VarProp[1]=FMax" )
if "CutsSA" in mlist:
factory.BookMethod( TMVA.Types.kCuts, "CutsSA",
"!H:!V:FitMethod=SA:EffSel:MaxCalls=150000:KernelTemp=IncAdaptive:InitialTemp=1e+6:MinTemp=1e-6:Eps=1e-10:UseDefaultScale" )
# Likelihood ("naive Bayes estimator")
if "Likelihood" in mlist:
factory.BookMethod( TMVA.Types.kLikelihood, "Likelihood",
"H:!V:!TransformOutput:PDFInterpol=Spline2:NSmoothSig[0]=20:NSmoothBkg[0]=20:NSmoothBkg[1]=10:NSmooth=1:NAvEvtPerBin=50" )
# Decorrelated likelihood
if "LikelihoodD" in mlist:
factory.BookMethod( TMVA.Types.kLikelihood, "LikelihoodD",
"!H:!V:TransformOutput:PDFInterpol=Spline2:NSmoothSig[0]=20:NSmoothBkg[0]=20:NSmooth=5:NAvEvtPerBin=50:VarTransform=Decorrelate" )
# PCA-transformed likelihood
if "LikelihoodPCA" in mlist:
factory.BookMethod( TMVA.Types.kLikelihood, "LikelihoodPCA",
"!H:!V:!TransformOutput:PDFInterpol=Spline2:NSmoothSig[0]=20:NSmoothBkg[0]=20:NSmooth=5:NAvEvtPerBin=50:VarTransform=PCA" )
# Use a kernel density estimator to approximate the PDFs
if "LikelihoodKDE" in mlist:
factory.BookMethod( TMVA.Types.kLikelihood, "LikelihoodKDE",
"!H:!V:!TransformOutput:PDFInterpol=KDE:KDEtype=Gauss:KDEiter=Adaptive:KDEFineFactor=0.3:KDEborder=None:NAvEvtPerBin=50" )
# Use a variable-dependent mix of splines and kernel density estimator
if "LikelihoodMIX" in mlist:
factory.BookMethod( TMVA.Types.kLikelihood, "LikelihoodMIX",
"!H:!V:!TransformOutput:PDFInterpolSig[0]=KDE:PDFInterpolBkg[0]=KDE:PDFInterpolSig[1]=KDE:PDFInterpolBkg[1]=KDE:PDFInterpolSig[2]=Spline2:PDFInterpolBkg[2]=Spline2:PDFInterpolSig[3]=Spline2:PDFInterpolBkg[3]=Spline2:KDEtype=Gauss:KDEiter=Nonadaptive:KDEborder=None:NAvEvtPerBin=50" )
# Test the multi-dimensional probability density estimator
# here are the options strings for the MinMax and RMS methods, respectively:
# "!H:!V:VolumeRangeMode=MinMax:DeltaFrac=0.2:KernelEstimator=Gauss:GaussSigma=0.3" );
# "!H:!V:VolumeRangeMode=RMS:DeltaFrac=3:KernelEstimator=Gauss:GaussSigma=0.3" );
if "PDERS" in mlist:
factory.BookMethod( TMVA.Types.kPDERS, "PDERS",
"!H:!V:NormTree=T:VolumeRangeMode=Adaptive:KernelEstimator=Gauss:GaussSigma=0.3:NEventsMin=400:NEventsMax=600" )
if "PDERSD" in mlist:
factory.BookMethod( TMVA.Types.kPDERS, "PDERSD",
"!H:!V:VolumeRangeMode=Adaptive:KernelEstimator=Gauss:GaussSigma=0.3:NEventsMin=400:NEventsMax=600:VarTransform=Decorrelate" )
if "PDERSPCA" in mlist:
factory.BookMethod( TMVA.Types.kPDERS, "PDERSPCA",
"!H:!V:VolumeRangeMode=Adaptive:KernelEstimator=Gauss:GaussSigma=0.3:NEventsMin=400:NEventsMax=600:VarTransform=PCA" )
# Multi-dimensional likelihood estimator using self-adapting phase-space binning
if "PDEFoam" in mlist:
factory.BookMethod( TMVA.Types.kPDEFoam, "PDEFoam",
"!H:!V:SigBgSeparate=F:TailCut=0.001:VolFrac=0.0666:nActiveCells=500:nSampl=2000:nBin=5:Nmin=100:Kernel=None:Compress=T" )
if "PDEFoamBoost" in mlist:
factory.BookMethod( TMVA.Types.kPDEFoam, "PDEFoamBoost",
"!H:!V:Boost_Num=30:Boost_Transform=linear:SigBgSeparate=F:MaxDepth=4:UseYesNoCell=T:DTLogic=MisClassificationError:FillFoamWithOrigWeights=F:TailCut=0:nActiveCells=500:nBin=20:Nmin=400:Kernel=None:Compress=T" )
# K-Nearest Neighbour classifier (KNN)
if "KNN" in mlist:
factory.BookMethod( TMVA.Types.kKNN, "KNN",
"H:nkNN=20:ScaleFrac=0.8:SigmaFact=1.0:Kernel=Gaus:UseKernel=F:UseWeight=T:!Trim" )
# H-Matrix (chi2-squared) method
if "HMatrix" in mlist:
factory.BookMethod( TMVA.Types.kHMatrix, "HMatrix", "!H:!V" )
# Linear discriminant (same as Fisher discriminant)
if "LD" in mlist:
factory.BookMethod( TMVA.Types.kLD, "LD", "H:!V:VarTransform=None:CreateMVAPdfs:PDFInterpolMVAPdf=Spline2:NbinsMVAPdf=50:NsmoothMVAPdf=10" )
# Fisher discriminant (same as LD)
if "Fisher" in mlist:
factory.BookMethod( TMVA.Types.kFisher, "Fisher", "H:!V:Fisher:CreateMVAPdfs:PDFInterpolMVAPdf=Spline2:NbinsMVAPdf=50:NsmoothMVAPdf=10" )
# Fisher with Gauss-transformed input variables
if "FisherG" in mlist:
factory.BookMethod( TMVA.Types.kFisher, "FisherG", "H:!V:VarTransform=Gauss" )
# Composite classifier: ensemble (tree) of boosted Fisher classifiers
if "BoostedFisher" in mlist:
factory.BookMethod( TMVA.Types.kFisher, "BoostedFisher",
"H:!V:Boost_Num=20:Boost_Transform=log:Boost_Type=AdaBoost:Boost_AdaBoostBeta=0.2" )
# Function discrimination analysis (FDA) -- test of various fitters - the recommended one is Minuit (or GA or SA)
if "FDA_MC" in mlist:
factory.BookMethod( TMVA.Types.kFDA, "FDA_MC",
"H:!V:Formula=(0)+(1)*x0+(2)*x1+(3)*x2+(4)*x3:ParRanges=(-1,1)(-10,10);(-10,10);(-10,10);(-10,10):FitMethod=MC:SampleSize=100000:Sigma=0.1" );
if "FDA_GA" in mlist:
factory.BookMethod( TMVA.Types.kFDA, "FDA_GA",
"H:!V:Formula=(0)+(1)*x0+(2)*x1+(3)*x2+(4)*x3:ParRanges=(-1,1)(-10,10);(-10,10);(-10,10);(-10,10):FitMethod=GA:PopSize=300:Cycles=3:Steps=20:Trim=True:SaveBestGen=1" );
if "FDA_SA" in mlist:
factory.BookMethod( TMVA.Types.kFDA, "FDA_SA",
"H:!V:Formula=(0)+(1)*x0+(2)*x1+(3)*x2+(4)*x3:ParRanges=(-1,1)(-10,10);(-10,10);(-10,10);(-10,10):FitMethod=SA:MaxCalls=15000:KernelTemp=IncAdaptive:InitialTemp=1e+6:MinTemp=1e-6:Eps=1e-10:UseDefaultScale" );
if "FDA_MT" in mlist:
factory.BookMethod( TMVA.Types.kFDA, "FDA_MT",
"H:!V:Formula=(0)+(1)*x0+(2)*x1+(3)*x2+(4)*x3:ParRanges=(-1,1)(-10,10);(-10,10);(-10,10);(-10,10):FitMethod=MINUIT:ErrorLevel=1:PrintLevel=-1:FitStrategy=2:UseImprove:UseMinos:SetBatch" );
if "FDA_GAMT" in mlist:
factory.BookMethod( TMVA.Types.kFDA, "FDA_GAMT",
"H:!V:Formula=(0)+(1)*x0+(2)*x1+(3)*x2+(4)*x3:ParRanges=(-1,1)(-10,10);(-10,10);(-10,10);(-10,10):FitMethod=GA:Converger=MINUIT:ErrorLevel=1:PrintLevel=-1:FitStrategy=0:!UseImprove:!UseMinos:SetBatch:Cycles=1:PopSize=5:Steps=5:Trim" );
if "FDA_MCMT" in mlist:
factory.BookMethod( TMVA.Types.kFDA, "FDA_MCMT",
"H:!V:Formula=(0)+(1)*x0+(2)*x1+(3)*x2+(4)*x3:ParRanges=(-1,1)(-10,10);(-10,10);(-10,10);(-10,10):FitMethod=MC:Converger=MINUIT:ErrorLevel=1:PrintLevel=-1:FitStrategy=0:!UseImprove:!UseMinos:SetBatch:SampleSize=20" );
# TMVA ANN: MLP (recommended ANN) -- all ANNs in TMVA are Multilayer Perceptrons
if "MLP" in mlist:
factory.BookMethod( TMVA.Types.kMLP, "MLP", "H:!V:NeuronType=tanh:VarTransform=N:NCycles=600:HiddenLayers=N+5:TestRate=5:!UseRegulator" )
if "MLPBFGS" in mlist:
factory.BookMethod( TMVA.Types.kMLP, "MLPBFGS", "H:!V:NeuronType=tanh:VarTransform=N:NCycles=600:HiddenLayers=N+5:TestRate=5:TrainingMethod=BFGS:!UseRegulator" )
if "MLPBNN" in mlist:
factory.BookMethod( TMVA.Types.kMLP, "MLPBNN", "H:!V:NeuronType=tanh:VarTransform=N:NCycles=600:HiddenLayers=N+5:TestRate=5:TrainingMethod=BFGS:UseRegulator" ) # BFGS training with bayesian regulators
# CF(Clermont-Ferrand)ANN
if "CFMlpANN" in mlist:
factory.BookMethod( TMVA.Types.kCFMlpANN, "CFMlpANN", "!H:!V:NCycles=2000:HiddenLayers=N+1,N" ) # n_cycles:#nodes:#nodes:...
# Tmlp(Root)ANN
if "TMlpANN" in mlist:
factory.BookMethod( TMVA.Types.kTMlpANN, "TMlpANN", "!H:!V:NCycles=200:HiddenLayers=N+1,N:LearningMethod=BFGS:ValidationFraction=0.3" ) #n_cycles:#nodes:#nodes:...
# Support Vector Machine
if "SVM" in mlist:
factory.BookMethod( TMVA.Types.kSVM, "SVM", "Gamma=0.25:Tol=0.001:VarTransform=Norm" )
# Boosted Decision Trees
if "BDTG" in mlist:
factory.BookMethod( TMVA.Types.kBDT, "BDTG",
"!H:!V:NTrees=1000:BoostType=Grad:Shrinkage=0.30:UseBaggedGrad:GradBaggingFraction=0.6:SeparationType=GiniIndex:nCuts=20:NNodesMax=5" )
if "BDT" in mlist:
factory.BookMethod( TMVA.Types.kBDT, "BDT",
"!H:!V:NTrees=850:nEventsMin=150:MaxDepth=3:BoostType=AdaBoost:AdaBoostBeta=0.5:SeparationType=GiniIndex:nCuts=20:PruneMethod=NoPruning" )
if "BDTB" in mlist:
factory.BookMethod( TMVA.Types.kBDT, "BDTB",
"!H:!V:NTrees=400:BoostType=Bagging:SeparationType=GiniIndex:nCuts=20:PruneMethod=NoPruning" )
if "BDTD" in mlist:
factory.BookMethod( TMVA.Types.kBDT, "BDTD",
"!H:!V:NTrees=400:nEventsMin=400:MaxDepth=3:BoostType=AdaBoost:SeparationType=GiniIndex:nCuts=20:PruneMethod=NoPruning:VarTransform=Decorrelate" )
# RuleFit -- TMVA implementation of Friedman's method
if "RuleFit" in mlist:
factory.BookMethod( TMVA.Types.kRuleFit, "RuleFit",
"H:!V:RuleFitModule=RFTMVA:Model=ModRuleLinear:MinImp=0.001:RuleMinDist=0.001:NTrees=20:fEventsMin=0.01:fEventsMax=0.5:GDTau=-1.0:GDTauPrec=0.01:GDStep=0.01:GDNSteps=10000:GDErrScale=1.02" )
# --------------------------------------------------------------------------------------------------
# ---- Now you can tell the factory to train, test, and evaluate the MVAs.
# Train MVAs
factory.TrainAllMethods()
# Test MVAs
factory.TestAllMethods()
# Evaluate MVAs
factory.EvaluateAllMethods()
# Save the output.
outputFile.Close()
print "=== wrote root file %s\n" % outfname
print "=== TMVAClassification is done!\n"
# open the GUI for the result macros
if( gui ):
gROOT.ProcessLine( "TMVAGui(\"%s\")" % outfname )
# keep the ROOT thread running
gApplication.Run()
def _setFactory(self, outFileName):
self._fout = TFile(outFileName, "RECREATE")
self._factory = TMVA.Factory(self._options['factory']['name'],
self._fout,
self._options['factory']['options'])
def main():
print "\n", "=" * 80
print "\tDESY 2017 - classification with TMVA"
print "=" * 80
# summary root file
summaryFilename = 'TMVA.root'
# results directory
resultsDir = 'results'
os.system('mkdir -p %s' % resultsDir)
#------------------------------------------------------------------
# get signal file and associated Root tree
sigFilename = "../public/ntuple_HZZ4L.root"
sigFile, sigTree = getTree(sigFilename)
# get background file and associated Root tree
bkgFilename = "../public/ntuple_ZZ4L.root"
bkgFile, bkgTree = getTree(bkgFilename)
#------------------------------------------------------------------
# create a factory for booking machine learning methods
#------------------------------------------------------------------
outputFile = TFile("TMVA.root", "recreate")
options = '''
!V
Color
!Silent
DrawProgressBar
AnalysisType=Classification
Transformations=I;D
'''
factory = TMVA.Factory("Z1massZ2mass", outputFile, formatOptions(options))
#------------------------------------------------------------------
# set up data set for training and testing
#------------------------------------------------------------------
dataLoader = TMVA.DataLoader(resultsDir)
# define variables to be used
dataLoader.AddVariable("Z1mass", 'D')
dataLoader.AddVariable("Z2mass", 'D')
# define from which trees data are to be taken
# from and the global weights to be assigned to
# the training data
sigWeight = 1.0
dataLoader.AddSignalTree(sigTree, sigWeight)
dataLoader.SetSignalWeightExpression("weight")
bkgWeight = 1.0
dataLoader.AddBackgroundTree(bkgTree, bkgWeight)
dataLoader.SetBackgroundWeightExpression("weight")
# you can apply cuts, if needed
cut = TCut("")
options = '''
SplitMode=Random
NormMode=EqualNumEvents
nTrain_Signal=2500
nTest_Signal=2500
nTrain_Background=2500
nTest_Background=2500
!V
'''
dataLoader.PrepareTrainingAndTestTree(
cut, # signal cut
cut, # background cut
formatOptions(options))
#------------------------------------------------------------------
# ok, almost done, define machine learning methods to be run
#------------------------------------------------------------------
options = '''
!H
!V
VarTransform=None
'''
factory.BookMethod(dataLoader, TMVA.Types.kFisher, "Fisher",
formatOptions(options))
options = '''
!H
!V
BoostType=AdaBoost
NTrees=200
nEventsMin=100
nCuts=50
'''
factory.BookMethod(dataLoader, TMVA.Types.kBDT, "BDT",
formatOptions(options))
options = '''
!H
!V
NCycles=500
VarTransform=N
HiddenLayers=5
TrainingMethod=BFGS
'''
factory.BookMethod(dataLoader, TMVA.Types.kMLP, "MLP",
formatOptions(options))
#------------------------------------------------------------------
# ok, let's go!
#------------------------------------------------------------------
factory.TrainAllMethods()
factory.TestAllMethods()
factory.EvaluateAllMethods()
outputFile.Close()
def main():
try:
# retrive command line options
shortopts = "m:i:t:b:s:a:o:vh?"
longopts = [
"methods=", "inputfile=", "inputtrees=", "bkgList=", "sigList=",
"anaType=", "outputfile=", "verbose", "help", "usage"
]
opts, args = getopt.getopt(sys.argv[1:], shortopts, longopts)
except getopt.GetoptError:
# print help information and exit:
print "ERROR: unknown options in argument %s" % sys.argv[1:]
usage()
sys.exit(1)
infname = DEFAULT_INFNAME
bkgList = DEFAULT_BKGLIST
sigList = DEFAULT_SIGLIST
anaType = DEFAULT_ANATYPE
treeNameSig = DEFAULT_TREESIG
treeNameBkg = DEFAULT_TREEBKG
outfname = DEFAULT_OUTFNAME
methods = DEFAULT_METHODS
verbose = False
for o, a in opts:
if o in ("-?", "-h", "--help", "--usage"):
usage()
sys.exit(0)
elif o in ("-m", "--methods"):
methods = a
elif o in ("-i", "--inputfile"):
infname = a
elif o in ("-b", "--bkgList"):
bkgList = a
elif o in ("-s", "--sigList"):
sigList = a
elif o in ("-a", "--anaType"):
anaType = a
elif o in ("-o", "--outputfile"):
outfname = a
elif o in ("-t", "--inputtrees"):
a.strip()
trees = a.rsplit(' ')
trees.sort()
trees.reverse()
if len(trees) - trees.count('') != 2:
print "ERROR: need to give two trees (each one for signal and background)"
print trees
sys.exit(1)
treeNameSig = trees[0]
treeNameBkg = trees[1]
elif o in ("-v", "--verbose"):
verbose = True
# Print methods
mlist = methods.replace(' ', ',').split(',')
print "=== TMVAClassification: use method(s)..."
for m in mlist:
if m.strip() != '':
print "=== - <%s>" % m.strip()
# Import ROOT classes
from ROOT import gSystem, gROOT, gApplication, TFile, TTree, TCut
# check ROOT version, give alarm if 5.18
if gROOT.GetVersionCode() >= 332288 and gROOT.GetVersionCode() < 332544:
print "*** You are running ROOT version 5.18, which has problems in PyROOT such that TMVA"
print "*** does not run properly (function calls with enums in the argument are ignored)."
print "*** Solution: either use CINT or a C++ compiled version (see TMVA/macros or TMVA/examples),"
print "*** or use another ROOT version (e.g., ROOT 5.19)."
sys.exit(1)
#Edited
setupTMVA(pathToTMVA)
# Import TMVA classes from ROOT
from ROOT import TMVA
# Output file
outputFile = TFile(outfname, 'RECREATE')
# Create instance of TMVA factory (see TMVA/macros/TMVAClassification.C for more factory options)
# All TMVA output can be suppressed by removing the "!" (not) in
# front of the "Silent" argument in the option string
suffix = outfname.split("/")[-1].split(".")[0]
factory = TMVA.Factory(
"TMVAClassification_%s" % suffix, outputFile,
"!V:!Silent:Color:DrawProgressBar:Transformations=I;D;P:AnalysisType=Classification"
)
# Set verbosity
factory.SetVerbose(verbose)
# If you wish to modify default settings
# (please check "src/Config.h" to see all available global options)
# gConfig().GetVariablePlotting()).fTimesRMS = 8.0
# gConfig().GetIONames()).fWeightFileDir = "myWeightDirectory"
# Define the input variables that shall be used for the classifier training
# note that you may also use variable expressions, such as: "3*var1/var2*abs(var3)"
# [all types of expressions that can also be parsed by TTree::Draw( "expression" )]
#Edited
#follow 2LSS note Ch5.2: Disciminating variables
factory.AddVariable("mT2 := sig.mT2", 'F')
factory.AddVariable("pt := l12.pt", 'F')
factory.AddVariable("MET := sig.MetRel", 'F')
factory.AddVariable("Ht := Sum$(jets.pt) + Sum$(leps.pt)", 'F')
factory.AddVariable("mTl1 := leps.mT[0]", 'F')
factory.AddVariable("mTl2 := leps.mT[1]", 'F')
factory.AddVariable("ll_dPhi:= l12.dPhi", 'F')
factory.AddVariable(
"l12m := (int(abs(leps.ID[0]))!=int(abs(leps.ID[1])))*100 + l12.m",
'F')
#ISR region
if (anaType == "doISR"):
factory.AddVariable("JetMET_dPhi := jets.MET_dPhi[0]", 'F')
factory.AddVariable("MET_JetPt_R := sig.MetRel/jets.pt[0]", 'F')
factory.AddVariable("l1Pt_JetPt_R := leps.pt[0]/jets.pt[0]", 'F')
#factory.AddSpectator( "pt1 := leps.pt[0]" , 'F' )
#factory.AddSpectator( "pt2 := leps.pt[1]" , 'F' )
#factory.AddSpectator( "ID1 := int(leps.ID[0])" , 'I' )
#factory.AddSpectator( "ID2 := int(leps.ID[1])" , 'I' )
#factory.AddSpectator( "nCentralJets := Sum$(jets.pt>20 && abs(jets.eta)<2.4)" , 'I' )
#FIXME
setupXsecDB(pathToSUSYTools)
from ROOT.SUSY import CrossSectionDB
xsecDB = CrossSectionDB(pathToSUSYTools + "data/mc15_13TeV/")
#read in training data
openedInFileList = []
# Read input sig
sigList = open(sigList, "r")
for infname in sigList:
inFile = TFile.Open(infname[:-1])
openedInFileList.append(inFile)
hCutFlow = inFile.FindObjectAny("hCutFlow")
mcEntry = hCutFlow.GetBinContent(1)
#FIXME: hard coded extract runNum from filePath
m = re.match(".*\.([0-9]{6})\..*", infname)
runNum = int(m.groups()[0])
xSECxEff = xsecDB.xsectTimesEff(runNum, 125) + xsecDB.xsectTimesEff(
runNum, 127) #125,127 is channel no.
# Get the trees for training
signal = inFile.Get("Data_")
# Global event weights (see below for setting event-wise weights)
#signalWeight = getXSECxEff(xsecDB, infname) * tarLumi / mcEntry
#signalWeight = xSECxEff * tarLumi / mcEntry
signalWeight = 1.0 * tarLumi / mcEntry #treat diff SUSY scenario with equal weight
if signalWeight <= 0:
print "Encounter <=0 weight sample %s , skipped" % infname
continue
print "mc sig ", runNum, mcEntry, xSECxEff
factory.AddSignalTree(signal, signalWeight)
sigList.close()
# Read input bkg
bkgList = open(bkgList, "r")
for infname in bkgList:
inFile = TFile.Open(infname[:-1])
openedInFileList.append(inFile)
if "physics" in infname:
#its real data
print "data bkg", infname[:-1]
background = inFile.Get("CFlip_")
if background: factory.AddBackgroundTree(background, 1.0)
background = inFile.Get("FakeLep_")
if background: factory.AddBackgroundTree(background, 1.0)
else:
#its MC data
hCutFlow = inFile.FindObjectAny("hCutFlow")
mcEntry = hCutFlow.GetBinContent(1)
#FIXME: hard coded extract runNum from filePath
m = re.match(".*\.([0-9]{6})\..*", infname)
runNum = int(m.groups()[0])
xSECxEff = xsecDB.xsectTimesEff(runNum)
# Get trees for training
background = inFile.Get("Data_")
# Global event weights (see below for setting event-wise weights)
backgroundWeight = xSECxEff * tarLumi / mcEntry
if backgroundWeight <= 0:
print "Encounter <=0 weight sample %s , skipped" % infname
print "mc bkg ", runNum, mcEntry, xSECxEff
factory.AddBackgroundTree(background, backgroundWeight)
bkgList.close()
# event-wise weights
#factory.SetSignalWeightExpression( "weight" )
#factory.SetBackgroundWeightExpression( "weight" )
factory.SetSignalWeightExpression("ElSF*MuSF")
factory.SetBackgroundWeightExpression(
"(CFlipWeight0*FakeLepWeight0)!=1.0 ? CFlipWeight0*FakeLepWeight0 : !TMath::IsNaN(weight)? ElSF*MuSF*weight: 0.0"
)
# Apply additional cuts on the signal and background sample.
# example for cut: mycut = TCut( "abs(var1)<0.5 && abs(var2-0.5)<1" )
# trigCut = "sig.trigCode!=0"
#"HLT_mu24_iloose_L1MU15" for mumu emu, "HLT_e24_lhmedium_iloose_L1EM20VH" for ee
trigCut = "((nMu>0) && (sig.trigCode & (1<<2))) || ((nMu==0) && (sig.trigCode & (1<<26)))"
grlCut = "evtInfo.passGRL==1"
wCut = "weight>0 && weight<1e9"
tauCut = "1" # "nTau==0" FIXME nTau not properly filled in NTUP yet..
bjetCut = "Sum$(jets.isBJet)==0"
cosmicCut = "Sum$(leps.isCosmic)==0"
htCut = "(Sum$(jets.pt) + Sum$(leps.pt))>40"
posWCut = "FakeLepWeight0>0"
isrCut = "Sum$(jets.pt>20 && abs(jets.eta)<2.4) %s" % (
">0" if anaType == "doISR" else "==0") #nCentralJets>0 or ==0
zMassCut = "!(int(abs(leps.ID[0])) == int(abs(leps.ID[1])) && fabs(l12.m - 91.1876)<=5)"
#commonCut = "&&".join(["(%s)"%cut for cut in [trigCut , grlCut , bjetCut, cosmicCut]])
commonCut = "&&".join([
"(%s)" % cut for cut in
[trigCut, grlCut, wCut, zMassCut, isrCut, tauCut, bjetCut, cosmicCut]
])
commonCut = TCut(commonCut)
sigCut = "&&".join([
"(%s)" % cut for cut in
[trigCut, grlCut, zMassCut, isrCut, tauCut, bjetCut, cosmicCut]
])
sigCut = TCut(sigCut)
bkgCut = "&&".join([
"(%s)" % cut for cut in [
trigCut, grlCut, wCut, zMassCut, isrCut, tauCut, bjetCut,
cosmicCut, posWCut
]
])
bkgCut = TCut(bkgCut)
# Here, the relevant variables are copied over in new, slim trees that are
# used for TMVA training and testing
# "SplitMode=Random" means that the input events are randomly shuffled before
# splitting them into training and test samples
factory.PrepareTrainingAndTestTree(
sigCut, bkgCut,
"nTrain_Signal=0:nTrain_Background=0:nTest_Background=0:SplitMode=Random:NormMode=EqualNumEvents:!V"
)
#"nTrain_Signal=0:nTrain_Background=2000:SplitMode=Random:NormMode=EqualNumEvents:!V" )
# --------------------------------------------------------------------------------------------------
# ---- Book MVA methods
#
# please lookup the various method configuration options in the corresponding cxx files, eg:
# src/MethoCuts.cxx, etc, or here: http://tmva.sourceforge.net/optionRef.html
# it is possible to preset ranges in the option string in which the cut optimisation should be done:
# "...:CutRangeMin[2]=-1:CutRangeMax[2]=1"...", where [2] is the third input variable
# Cut optimisation
if "Cuts" in mlist:
factory.BookMethod(
TMVA.Types.kCuts, "Cuts",
"!H:!V:FitMethod=MC:EffSel:SampleSize=200000:VarProp=FSmart")
if "CutsD" in mlist:
factory.BookMethod(
TMVA.Types.kCuts, "CutsD",
"!H:!V:FitMethod=MC:EffSel:SampleSize=200000:VarProp=FSmart:VarTransform=Decorrelate"
)
if "CutsPCA" in mlist:
factory.BookMethod(
TMVA.Types.kCuts, "CutsPCA",
"!H:!V:FitMethod=MC:EffSel:SampleSize=200000:VarProp=FSmart:VarTransform=PCA"
)
if "CutsGA" in mlist:
factory.BookMethod(
TMVA.Types.kCuts, "CutsGA",
"H:!V:FitMethod=GA:VarProp=FSmart:EffSel:Steps=30:Cycles=3:PopSize=400:SC_steps=10:SC_rate=5:SC_factor=0.95"
)
if "CutsSA" in mlist:
factory.BookMethod(
TMVA.Types.kCuts, "CutsSA",
"!H:!V:FitMethod=SA:EffSel:MaxCalls=150000:KernelTemp=IncAdaptive:InitialTemp=1e+6:MinTemp=1e-6:Eps=1e-10:UseDefaultScale"
)
# Likelihood ("naive Bayes estimator")
if "Likelihood" in mlist:
factory.BookMethod(
TMVA.Types.kLikelihood, "Likelihood",
"H:!V:!TransformOutput:PDFInterpol=Spline2:NSmoothSig[0]=20:NSmoothBkg[0]=20:NSmoothBkg[1]=10:NSmooth=1:NAvEvtPerBin=50"
)
# Decorrelated likelihood
if "LikelihoodD" in mlist:
factory.BookMethod(
TMVA.Types.kLikelihood, "LikelihoodD",
"!H:!V:TransformOutput:PDFInterpol=Spline2:NSmoothSig[0]=20:NSmoothBkg[0]=20:NSmooth=5:NAvEvtPerBin=50:VarTransform=Decorrelate"
)
# PCA-transformed likelihood
if "LikelihoodPCA" in mlist:
factory.BookMethod(
TMVA.Types.kLikelihood, "LikelihoodPCA",
"!H:!V:!TransformOutput:PDFInterpol=Spline2:NSmoothSig[0]=20:NSmoothBkg[0]=20:NSmooth=5:NAvEvtPerBin=50:VarTransform=PCA"
)
# Use a kernel density estimator to approximate the PDFs
if "LikelihoodKDE" in mlist:
factory.BookMethod(
TMVA.Types.kLikelihood, "LikelihoodKDE",
"!H:!V:!TransformOutput:PDFInterpol=KDE:KDEtype=Gauss:KDEiter=Adaptive:KDEFineFactor=0.3:KDEborder=None:NAvEvtPerBin=50"
)
# Use a variable-dependent mix of splines and kernel density estimator
if "LikelihoodMIX" in mlist:
factory.BookMethod(
TMVA.Types.kLikelihood, "LikelihoodMIX",
"!H:!V:!TransformOutput:PDFInterpolSig[0]=KDE:PDFInterpolBkg[0]=KDE:PDFInterpolSig[1]=KDE:PDFInterpolBkg[1]=KDE:PDFInterpolSig[2]=Spline2:PDFInterpolBkg[2]=Spline2:PDFInterpolSig[3]=Spline2:PDFInterpolBkg[3]=Spline2:KDEtype=Gauss:KDEiter=Nonadaptive:KDEborder=None:NAvEvtPerBin=50"
)
# Test the multi-dimensional probability density estimator
# here are the options strings for the MinMax and RMS methods, respectively:
# "!H:!V:VolumeRangeMode=MinMax:DeltaFrac=0.2:KernelEstimator=Gauss:GaussSigma=0.3" );
# "!H:!V:VolumeRangeMode=RMS:DeltaFrac=3:KernelEstimator=Gauss:GaussSigma=0.3" );
if "PDERS" in mlist:
factory.BookMethod(
TMVA.Types.kPDERS, "PDERS",
"!H:!V:NormTree=T:VolumeRangeMode=Adaptive:KernelEstimator=Gauss:GaussSigma=0.3:NEventsMin=400:NEventsMax=600"
)
if "PDERSD" in mlist:
factory.BookMethod(
TMVA.Types.kPDERS, "PDERSD",
"!H:!V:VolumeRangeMode=Adaptive:KernelEstimator=Gauss:GaussSigma=0.3:NEventsMin=400:NEventsMax=600:VarTransform=Decorrelate"
)
if "PDERSPCA" in mlist:
factory.BookMethod(
TMVA.Types.kPDERS, "PDERSPCA",
"!H:!V:VolumeRangeMode=Adaptive:KernelEstimator=Gauss:GaussSigma=0.3:NEventsMin=400:NEventsMax=600:VarTransform=PCA"
)
# Multi-dimensional likelihood estimator using self-adapting phase-space binning
if "PDEFoam" in mlist:
factory.BookMethod(
TMVA.Types.kPDEFoam, "PDEFoam",
"!H:!V:SigBgSeparate=F:TailCut=0.001:VolFrac=0.0666:nActiveCells=500:nSampl=2000:nBin=5:Nmin=100:Kernel=None:Compress=T"
)
if "PDEFoamBoost" in mlist:
factory.BookMethod(
TMVA.Types.kPDEFoam, "PDEFoamBoost",
"!H:!V:Boost_Num=30:Boost_Transform=linear:SigBgSeparate=F:MaxDepth=4:UseYesNoCell=T:DTLogic=MisClassificationError:FillFoamWithOrigWeights=F:TailCut=0:nActiveCells=500:nBin=20:Nmin=400:Kernel=None:Compress=T"
)
# K-Nearest Neighbour classifier (KNN)
if "KNN" in mlist:
factory.BookMethod(
TMVA.Types.kKNN, "KNN",
"H:nkNN=20:ScaleFrac=0.8:SigmaFact=1.0:Kernel=Gaus:UseKernel=F:UseWeight=T:!Trim"
)
# H-Matrix (chi2-squared) method
if "HMatrix" in mlist:
factory.BookMethod(TMVA.Types.kHMatrix, "HMatrix", "!H:!V")
# Linear discriminant (same as Fisher discriminant)
if "LD" in mlist:
factory.BookMethod(
TMVA.Types.kLD, "LD",
"H:!V:VarTransform=None:CreateMVAPdfs:PDFInterpolMVAPdf=Spline2:NbinsMVAPdf=50:NsmoothMVAPdf=10"
)
# Fisher discriminant (same as LD)
if "Fisher" in mlist:
factory.BookMethod(
TMVA.Types.kFisher, "Fisher",
"H:!V:Fisher:CreateMVAPdfs:PDFInterpolMVAPdf=Spline2:NbinsMVAPdf=50:NsmoothMVAPdf=10"
)
# Fisher with Gauss-transformed input variables
if "FisherG" in mlist:
factory.BookMethod(TMVA.Types.kFisher, "FisherG",
"H:!V:VarTransform=Gauss")
# Composite classifier: ensemble (tree) of boosted Fisher classifiers
if "BoostedFisher" in mlist:
factory.BookMethod(
TMVA.Types.kFisher, "BoostedFisher",
"H:!V:Boost_Num=20:Boost_Transform=log:Boost_Type=AdaBoost:Boost_AdaBoostBeta=0.2"
)
# Function discrimination analysis (FDA) -- test of various fitters - the recommended one is Minuit (or GA or SA)
if "FDA_MC" in mlist:
factory.BookMethod(
TMVA.Types.kFDA, "FDA_MC",
"H:!V:Formula=(0)+(1)*x0+(2)*x1+(3)*x2+(4)*x3:ParRanges=(-1,1)(-10,10);(-10,10);(-10,10);(-10,10):FitMethod=MC:SampleSize=100000:Sigma=0.1"
)
if "FDA_GA" in mlist:
factory.BookMethod(
TMVA.Types.kFDA, "FDA_GA",
"H:!V:Formula=(0)+(1)*x0+(2)*x1+(3)*x2+(4)*x3:ParRanges=(-1,1)(-10,10);(-10,10);(-10,10);(-10,10):FitMethod=GA:PopSize=300:Cycles=3:Steps=20:Trim=True:SaveBestGen=1"
)
if "FDA_SA" in mlist:
factory.BookMethod(
TMVA.Types.kFDA, "FDA_SA",
"H:!V:Formula=(0)+(1)*x0+(2)*x1+(3)*x2+(4)*x3:ParRanges=(-1,1)(-10,10);(-10,10);(-10,10);(-10,10):FitMethod=SA:MaxCalls=15000:KernelTemp=IncAdaptive:InitialTemp=1e+6:MinTemp=1e-6:Eps=1e-10:UseDefaultScale"
)
if "FDA_MT" in mlist:
factory.BookMethod(
TMVA.Types.kFDA, "FDA_MT",
"H:!V:Formula=(0)+(1)*x0+(2)*x1+(3)*x2+(4)*x3:ParRanges=(-1,1)(-10,10);(-10,10);(-10,10);(-10,10):FitMethod=MINUIT:ErrorLevel=1:PrintLevel=-1:FitStrategy=2:UseImprove:UseMinos:SetBatch"
)
if "FDA_GAMT" in mlist:
factory.BookMethod(
TMVA.Types.kFDA, "FDA_GAMT",
"H:!V:Formula=(0)+(1)*x0+(2)*x1+(3)*x2+(4)*x3:ParRanges=(-1,1)(-10,10);(-10,10);(-10,10);(-10,10):FitMethod=GA:Converger=MINUIT:ErrorLevel=1:PrintLevel=-1:FitStrategy=0:!UseImprove:!UseMinos:SetBatch:Cycles=1:PopSize=5:Steps=5:Trim"
)
if "FDA_MCMT" in mlist:
factory.BookMethod(
TMVA.Types.kFDA, "FDA_MCMT",
"H:!V:Formula=(0)+(1)*x0+(2)*x1+(3)*x2+(4)*x3:ParRanges=(-1,1)(-10,10);(-10,10);(-10,10);(-10,10):FitMethod=MC:Converger=MINUIT:ErrorLevel=1:PrintLevel=-1:FitStrategy=0:!UseImprove:!UseMinos:SetBatch:SampleSize=20"
)
# TMVA ANN: MLP (recommended ANN) -- all ANNs in TMVA are Multilayer Perceptrons
if "MLP" in mlist:
factory.BookMethod(
TMVA.Types.kMLP, "MLP",
"H:!V:NeuronType=tanh:VarTransform=N:NCycles=600:HiddenLayers=N+5:TestRate=5:!UseRegulator"
)
if "MLPBFGS" in mlist:
factory.BookMethod(
TMVA.Types.kMLP, "MLPBFGS",
"H:!V:NeuronType=tanh:VarTransform=N:NCycles=600:HiddenLayers=N+5:TestRate=5:TrainingMethod=BFGS:!UseRegulator"
)
if "MLPBNN" in mlist:
factory.BookMethod(
TMVA.Types.kMLP, "MLPBNN",
"H:!V:NeuronType=tanh:VarTransform=N:NCycles=600:HiddenLayers=N+5:TestRate=5:TrainingMethod=BFGS:UseRegulator"
) # BFGS training with bayesian regulators
# CF(Clermont-Ferrand)ANN
if "CFMlpANN" in mlist:
factory.BookMethod(TMVA.Types.kCFMlpANN, "CFMlpANN",
"!H:!V:NCycles=2000:HiddenLayers=N+1,N"
) # n_cycles:#nodes:#nodes:...
# Tmlp(Root)ANN
if "TMlpANN" in mlist:
factory.BookMethod(
TMVA.Types.kTMlpANN, "TMlpANN",
"!H:!V:NCycles=200:HiddenLayers=N+1,N:LearningMethod=BFGS:ValidationFraction=0.3"
) # n_cycles:#nodes:#nodes:...
# Support Vector Machine
if "SVM" in mlist:
factory.BookMethod(TMVA.Types.kSVM, "SVM",
"Gamma=0.25:Tol=0.001:VarTransform=Norm")
# Boosted Decision Trees
if "BDTG" in mlist:
factory.BookMethod(
TMVA.Types.kBDT, "BDTG",
"!H:!V:NTrees=1000:MinNodeSize=1.5%:BoostType=Grad:Shrinkage=0.10:UseBaggedBoost:BaggedSampleFraction=0.5:nCuts=20:MaxDepth=2"
)
if "BDT" in mlist:
factory.BookMethod(
TMVA.Types.kBDT, "BDT",
"!H:!V:NTrees=850:MinNodeSize=2.5%:MaxDepth=3:BoostType=AdaBoost:AdaBoostBeta=0.5:UseBaggedBoost:BaggedSampleFraction=0.5:SeparationType=GiniIndex:nCuts=20"
)
if "BDTB" in mlist:
factory.BookMethod(
TMVA.Types.kBDT, "BDTB",
"!H:!V:NTrees=400:BoostType=Bagging:SeparationType=GiniIndex:nCuts=20"
)
if "BDTD" in mlist:
factory.BookMethod(
TMVA.Types.kBDT, "BDTD",
"!H:!V:NTrees=400:MinNodeSize=5%:MaxDepth=2:BoostType=AdaBoost:SeparationType=GiniIndex:nCuts=20:VarTransform=Decorrelate"
)
# RuleFit -- TMVA implementation of Friedman's method
if "RuleFit" in mlist:
factory.BookMethod(
TMVA.Types.kRuleFit, "RuleFit",
"H:!V:RuleFitModule=RFTMVA:Model=ModRuleLinear:MinImp=0.001:RuleMinDist=0.001:NTrees=20:fEventsMin=0.01:fEventsMax=0.5:GDTau=-1.0:GDTauPrec=0.01:GDStep=0.01:GDNSteps=10000:GDErrScale=1.02"
)
# --------------------------------------------------------------------------------------------------
# ---- Now you can tell the factory to train, test, and evaluate the MVAs.
# Train MVAs
factory.TrainAllMethods()
# Test MVAs
factory.TestAllMethods()
# Evaluate MVAs
factory.EvaluateAllMethods()
# Save the output.
outputFile.Close()
print "=== wrote root file %s\n" % outfname
print "=== TMVAClassification is done!\n"
def main():
try:
# retrive command line options
shortopts = "m:i:t:o:vh?"
longopts = [
"methods=", "inputfile=", "inputtrees=", "outputfile=", "verbose",
"help", "usage"
]
opts, args = getopt.getopt(sys.argv[1:], shortopts, longopts)
except getopt.GetoptError:
# print help information and exit:
print "ERROR: unknown options in argument %s" % sys.argv[1:]
usage()
sys.exit(1)
infname = DEFAULT_INFNAME
treeNameSig = DEFAULT_TREESIG
treeNameBkg = DEFAULT_TREEBKG
outfname = DEFAULT_OUTFNAME
methods = DEFAULT_METHODS
verbose = False
for o, a in opts:
if o in ("-?", "-h", "--help", "--usage"):
usage()
sys.exit(0)
elif o in ("-m", "--methods"):
methods = a
elif o in ("-i", "--inputfile"):
infname = a
elif o in ("-o", "--outputfile"):
outfname = a
elif o in ("-t", "--inputtrees"):
a.strip()
trees = a.rsplit(' ')
trees.sort()
trees.reverse()
if len(trees) - trees.count('') != 2:
print "ERROR: need to give two trees (each one for signal and background)"
print trees
sys.exit(1)
treeNameSig = trees[0]
treeNameBkg = trees[1]
elif o in ("-v", "--verbose"):
verbose = True
# Print methods
mlist = methods.replace(' ', ',').split(',')
print "=== TMVAClassification: use method(s)..."
for m in mlist:
if m.strip() != '':
print "=== - <%s>" % m.strip()
# Import ROOT classes
from ROOT import gSystem, gROOT, gApplication, TFile, TTree, TCut
# check ROOT version, give alarm if 5.18
if gROOT.GetVersionCode() >= 332288 and gROOT.GetVersionCode() < 332544:
print "*** You are running ROOT version 5.18, which has problems in PyROOT such that TMVA"
print "*** does not run properly (function calls with enums in the argument are ignored)."
print "*** Solution: either use CINT or a C++ compiled version (see TMVA/macros or TMVA/examples),"
print "*** or use another ROOT version (e.g., ROOT 5.19)."
sys.exit(1)
# Logon not automatically loaded through PyROOT (logon loads TMVA library) load also GUI
gROOT.SetMacroPath("./")
gROOT.Macro("./TMVAlogon.C")
gROOT.LoadMacro("./TMVAGui.C")
# Import TMVA classes from ROOT
from ROOT import TMVA
# Output file
outputFile = TFile(outfname, 'RECREATE')
# Create instance of TMVA factory (see TMVA/macros/TMVAClassification.C for more factory options)
# All TMVA output can be suppressed by removing the "!" (not) in
# front of the "Silent" argument in the option string
factory = TMVA.Factory(
"TMVAClassification", outputFile,
"!V:!Silent:Color:DrawProgressBar:Transformations=I;D;P;G,D:AnalysisType=Classification"
)
# Set verbosity
factory.SetVerbose(verbose)
# If you wish to modify default settings
# (please check "src/Config.h" to see all available global options)
# gConfig().GetVariablePlotting()).fTimesRMS = 8.0
# gConfig().GetIONames()).fWeightFileDir = "myWeightDirectory"
# Define the input variables that shall be used for the classifier training
# note that you may also use variable expressions, such as: "3*var1/var2*abs(var3)"
# [all types of expressions that can also be parsed by TTree::Draw( "expression" )]
#varList = ['svMass', 'mJJ', 'met/fMass','pZ - pZV']
# varList = ['svMass', 'fMass', 'dRTauTau', 'dRJJ', 'svPt', 'dRhh', 'met', 'mJJReg',
# 'metTau1DPhi', 'metTau2DPhi', 'metJ1DPhi', 'metJ2DPhi', 'metTauPairDPhi', 'metSvTauPairDPhi', 'metJetPairDPhi','CSVJ1', 'CSVJ2']
# varList = ['svMass', 'dRTauTau', 'svPt', 'dRhh', 'met', 'mJJReg',
# 'metTau1DPhi', 'metTau2DPhi', 'metJ2DPhi', 'metJetPairDPhi','CSVJ1', 'CSVJ2']
varList = [
'svMass', 'dRTauTau', 'dRJJ', 'svPt', 'dRhh', 'met', 'mJJReg',
'metTau1DPhi', 'metTau2DPhi', 'metJ1DPhi', 'metJ2DPhi',
'metTauPairDPhi', 'metSvTauPairDPhi', 'metJetPairDPhi', 'CSVJ1',
'CSVJ2'
]
for iVar in varList:
factory.AddVariable(iVar, 'F')
#factory.AddVariable( "NBTags",'I' )
# You can add so-called "Spectator variables", which are not used in the MVA training,
# but will appear in the final "TestTree" produced by TMVA. This TestTree will contain the
# input variables, the response values of all trained MVAs, and the spectator variables
# factory.AddSpectator( "fMass")
# factory.AddSpectator( "spec2:=var1*3", "Spectator 2", "units", 'F' )
# Read input data
# sigChain = r.TChain("ttTreeFinal/eventTree")
# bkg1Chain = r.TChain("ttTreeFinal/eventTree")
# bkg2Chain = r.TChain("ttTreeFinal/eventTree")
# Get the signal and background trees for training
iFileSig = TFile.Open("/scratch/zmao/relaxed_regression/%s" % (infname))
iFileBkg = TFile.Open(
"/scratch/zmao/relaxed_regression/trainSample_relaxedsamebTag.root")
sigChain = iFileSig.Get("eventTree")
bkgChain = iFileBkg.Get("eventTree")
# Global event weights (see below for setting event-wise weights)
signalWeight = 1 #0.0159/sigChain.GetEntries() #xs (pb)
# ====== register trees ====================================================
#
# the following method is the prefered one:
# you can add an arbitrary number of signal or background trees
factory.AddSignalTree(sigChain, signalWeight)
factory.AddBackgroundTree(bkgChain, 1)
factory.SetSignalWeightExpression('triggerEff')
# To give different trees for training and testing, do as follows:
# factory.AddSignalTree( signalTrainingTree, signalTrainWeight, "Training" )
# factory.AddSignalTree( signalTestTree, signalTestWeight, "Test" )
# Use the following code instead of the above two or four lines to add signal and background
# training and test events "by hand"
# NOTE that in this case one should not give expressions (such as "var1+var2") in the input
# variable definition, but simply compute the expression before adding the event
#
# # --- begin ----------------------------------------------------------
#
# ... *** please lookup code in TMVA/macros/TMVAClassification.C ***
#
# # --- end ------------------------------------------------------------
#
# ====== end of register trees ==============================================
# Set individual event weights (the variables must exist in the original TTree)
# for signal : factory.SetSignalWeightExpression ("weight1*weight2");
# for background: factory.SetBackgroundWeightExpression("weight1*weight2");
#factory.SetBackgroundWeightExpression( "weight" )
# Apply additional cuts on the signal and background sample.
# example for cut: mycut = TCut( "abs(var1)<0.5 && abs(var2-0.5)<1" )
mycutSig = TCut(
"iso1<1.5 && iso2<1.5 && CSVJ1 > 0.679 && CSVJ2 > 0.244 && abs(eta1)<2.1 && abs(eta2)<2.1 && charge1 + charge2 == 0"
)
mycutBkg = TCut("")
# Here, the relevant variables are copied over in new, slim trees that are
# used for TMVA training and testing
# "SplitMode=Random" means that the input events are randomly shuffled before
# splitting them into training and test samples
factory.PrepareTrainingAndTestTree(
mycutSig, mycutBkg,
"nTrain_Signal=0:nTrain_Background=0:SplitMode=Random:NormMode=NumEvents:!V"
)
# --------------------------------------------------------------------------------------------------
# ---- Book MVA methods
#
# please lookup the various method configuration options in the corresponding cxx files, eg:
# src/MethoCuts.cxx, etc, or here: http://tmva.sourceforge.net/optionRef.html
# it is possible to preset ranges in the option string in which the cut optimisation should be done:
# "...:CutRangeMin[2]=-1:CutRangeMax[2]=1"...", where [2] is the third input variable
# Cut optimisation
# Fisher discriminant (same as LD)
# if "Fisher" in mlist:
#factory.BookMethod( TMVA.Types.kFisher, "Fisher", "H:!V:Fisher:CreateMVAPdfs:PDFInterpolMVAPdf=Spline2:NbinsMVAPdf=50:NsmoothMVAPdf=10" )
# factory.BookMethod( TMVA.Types.kFisher, "Fisher")
if "BDT" in mlist:
factory.BookMethod(
TMVA.Types.kBDT, "BDT",
"!H:!V:NTrees=150:MinNodeSize=2.5%:MaxDepth=3:BoostType=AdaBoost:AdaBoostBeta=0.5:UseBaggedBoost:BaggedSampleFraction=0.5:SeparationType=GiniIndex:nCuts=-1"
)
# --------------------------------------------------------------------------------------------------
# ---- Now you can tell the factory to train, test, and evaluate the MVAs.
# Train MVAs
factory.TrainAllMethods()
# Test MVAs
factory.TestAllMethods()
# Evaluate MVAs
factory.EvaluateAllMethods()
# Save the output.
outputFile.Close()
print "=== wrote root file %s\n" % outfname
print "=== TMVAClassification is done!\n"
# open the GUI for the result macros
# gROOT.ProcessLine( "TMVAGui(\"%s\")" % outfname )
ChangeWeightName = 'mv /afs/hep.wisc.edu/home/zmao/CMSSW_5_3_15/src/TMVA-v4.2.0/test/weights/TMVAClassification_BDT.weights.xml /afs/hep.wisc.edu/home/zmao/CMSSW_5_3_15/src/TMVA-v4.2.0/test/weights/TMVAClassification_BDT.weights_QCD_%i.xml' % len(
varList)
os.system(ChangeWeightName)
def main():
try:
# retrive command line options
shortopts = "m:p:M:C:B:i:t:T:o:vh?"
opts, args = getopt.getopt(sys.argv[1:], shortopts)
except getopt.GetoptError:
# print help information and exit:
print "ERROR: unknown options in argument %s" % sys.argv[1:]
usage()
sys.exit(1)
infname = DEFAULT_INFNAME
methods = DEFAULT_METHODS
mass = DEFAULT_MASS
cat = DEFAULT_CAT
phil = DEFAULT_PHIL
outfname = DEFAULT_OUTFNAME
treeNameSig = DEFAULT_TREESIG
treeNameBkg = DEFAULT_TREEBKG
bkg_method = DEFAULT_BACKGROUND
width = DEFAULT_WIDTH
verbose = False
test = False
testType = DEFAULT_TEST_TYPE
methTest = False
testMethod = DEFAULT_TEST_METHOD
for o, a in opts:
if o in ("-?", "-h", "--help", "--usage"):
usage()
sys.exit(0)
elif o in ("-m", "--methods"):
methods = a
elif o in ("-M", "--mass"):
mass = int(a)
elif o in ("-C", "--cat"):
cat = int(a)
elif o in ("-p", "--philosophy"):
phil = a
elif o in ("-B", "--background"):
bkg_method = int(a)
elif o in ("-i", "--inputfile"):
infname = a
elif o in ("-o", "--outputfile"):
outfname = a
elif o in ("-T", "--test"):
test = True
temp = a.split('_')
if len(temp) == 1:
testType = temp[0]
if testType != "ada" or testType != "grad":
print "ERROR: testType must be ada or grad not", testType
elif len(temp) - temp.count('') == 2:
methTest = True
testType = temp[0]
testMethod = temp[1]
checkTestType(testType, testMethod)
else:
print "ERROR: need to give one or two test options"
print temp
sys.exit(1)
elif o in ("-t", "--inputtrees"):
a.strip()
trees = a.rsplit(' ')
trees.sort()
trees.reverse()
if len(trees) - trees.count('') != 2:
print "ERROR: need to give two trees (each one for signal and background)"
print trees
sys.exit(1)
treeNameSig = trees[0]
treeNameBkg = trees[1]
elif o in ("-v", "--verbose"):
verbose = True
if (width == 0.02): width_str = "_2pt"
elif (width == 0.07): width_str = "_7pt"
mass_str = "_" + str("%3.1f" % mass)
cat_str = "_" + str(cat)
if cat < 0:
cat_str = "_all"
if test:
if methTest:
outfname = "TMVAStuff/" + outfname + "_" + phil + cat_str + "_test_" + testType + "_" + testMethod + ".root"
else:
outfname = "TMVAStuff/" + outfname + "_" + phil + cat_str + "_test_" + testType + ".root"
else:
outfname = "TMVAStuff/" + outfname + "_" + phil + cat_str + ".root"
#treeNameSig = treeNameSig + mass_str
#treeNameBkg = treeNameBkg + mass_str
# Print methods
mlist = methods.replace(' ', ',').split(',')
print "=== TMVAClassification: use method(s)..."
for m in mlist:
if m.strip() != '':
print "=== - <%s>" % m.strip()
# Import ROOT classes
from ROOT import gSystem, gROOT, gApplication, TFile, TTree, TCut
# check ROOT version, give alarm if 5.18
if gROOT.GetVersionCode() >= 332288 and gROOT.GetVersionCode() < 332544:
print "*** You are running ROOT version 5.18, which has problems in PyROOT such that TMVA"
print "*** does not run properly (function calls with enums in the argument are ignored)."
print "*** Solution: either use CINT or a C++ compiled version (see TMVA/macros or TMVA/examples),"
print "*** or use another ROOT version (e.g., ROOT 5.19)."
sys.exit(1)
# Logon not automatically loaded through PyROOT (logon loads TMVA library)
# load also GUI
gROOT.SetMacroPath("/vols/cms03/mk1009/h2g/MVA/tmvaMacros/")
gROOT.Macro("/vols/cms03/mk1009/h2g/MVA/tmvaMacros/TMVAlogon.C")
gROOT.LoadMacro("/vols/cms03/mk1009/h2g/MVA/tmvaMacros/TMVAGui.C")
# Import TMVA classes from ROOT
from ROOT import TMVA
# Output file
outputFile = TFile(outfname, 'RECREATE')
# Create instance of TMVA factory (see TMVA/macros/TMVAClassification.C for
# more factory options)
# All TMVA output can be suppressed by removing the "!" (not) in
# front of the "Silent" argument in the option string
factory = TMVA.Factory(
"TMVAClassification", outputFile,
"!V:!Silent:Color:DrawProgressBar:Transformations=I;D;P;G,D:AnalysisType=Classification"
)
# Set verbosity
factory.SetVerbose(verbose)
factory.AddVariable("bdtoutput", "BDT Output", 'F')
factory.AddVariable("deltaMOverM", "#DeltaM / M_{Hypth}.", 'F')
input = TFile.Open(infname)
# Get the signal and background trees for training
signal_train = input.Get(treeNameSig + "_train" + mass_str)
signal_test = input.Get(treeNameSig + "_test" + mass_str)
background_train = input.Get(treeNameBkg + "_train" + width_str + mass_str)
background_test = input.Get(treeNameBkg + "_test" + width_str + mass_str)
# Global event weights (see below for setting event-wise weights)
signalWeight = 1.0
backgroundWeight = 1.0
# ====== register trees ====================================================
factory.AddSignalTree(signal_train, signalWeight, "train")
factory.AddBackgroundTree(background_train, backgroundWeight, "train")
factory.AddSignalTree(signal_test, signalWeight, "test")
factory.AddBackgroundTree(background_test, backgroundWeight, "test")
# Set individual event weights (the variables must exist in the original
# TTree)
factory.SetBackgroundWeightExpression("wt")
factory.SetSignalWeightExpression("wt")
# Apply additional cuts on the signal and background sample.
# example for cut: mycut = TCut( "abs(var1)<0.5 && abs(var2-0.5)<1" )
mycut = TCut("fabs(deltaMOverM)<=" + str(width) +
" && bdtoutput > -0.5") #
# Here, the relevant variables are copied over in new, slim trees that are
# used for TMVA training and testing
factory.PrepareTrainingAndTestTree(
mycut, mycut,
"nTrain_Signal=0:nTrain_Background=0:NormMode=NumEvents:!V")
# Boosted Decision Trees
# NEW PARAMETERS
if (not test):
# Likelihood
factory.BookMethod(
TMVA.Types.kLikelihood, "Likelihood" + phil,
"H:!V:TransformOutput:PDFInterpol=Spline2:NSmoothSig[0]=20:NSmoothBkg[0]=20:NSmoothBkg[1]=10:NSmooth=1:NAvEvtPerBin=50"
)
factory.BookMethod(
TMVA.Types.kLikelihood, "LikelihoodD" + phil,
"!H:!V:TransformOutput:PDFInterpol=Spline2:NSmoothSig[0]=20:NSmoothBkg[0]=20:NSmooth=5:NAvEvtPerBin=50:VarTransform=Decorrelate"
)
#factory.BookMethod( TMVA.Types.kPDERS, "MultiLikelihood"+phil,"!H:!V:NormTree=T:VolumeRangeMode=Adaptive:KernelEstimator=Gauss:GaussSigma=0.3:NEventsMin=400:NEventsMax=600" );
# BDT
factory.BookMethod(
TMVA.Types.kBDT, "BDTada" + phil,
"!H:!V:NTrees=200:nEventsMin=150:MaxDepth=3:BoostType=AdaBoost:AdaBoostBeta=1.0:SeparationType=GiniIndex:nCuts=50:PruneMethod=NoPruning"
)
factory.BookMethod(
TMVA.Types.kBDT, "BDTgrad" + phil,
"!H:!V:NTrees=200:MaxDepth=3:BoostType=Grad:Shrinkage=0.5:UseBaggedGrad:GradBaggingFraction=1.0:SeparationType=GiniIndex:nCuts=50:NNodesMax=10"
)
else: #test
# BDT ada
if testType == "ada":
#if testMethod=="nTrees":
for nTrees in [10, 50, 100, 200, 500]:
for depth in [2, 3]:
factory.BookMethod(
TMVA.Types.kBDT, "BDT_ada" + str(phil) + "_" +
str(nTrees) + "t_" + str(depth) + "d",
"!H:!V:NTrees=" + str(nTrees) +
":nEventsMin=150:MaxDepth=" + str(depth) +
":BoostType=AdaBoost:AdaBoostBeta=1:SeparationType=GiniIndex:nCuts=50:PruneMethod=NoPruning"
)
# if testMethod=="depth":
# for depth in [2,3]:
# factory.BookMethod( TMVA.Types.kBDT, "BDT_ada"+str(phil)+"_200t_"+str(depth)+"d_0.05b_50c","!H:!V:NTrees=200:nEventsMin=150:MaxDepth="+str(depth)+":BoostType=AdaBoost:AdaBoostBeta=0.05:SeparationType=GiniIndex:nCuts=50:PruneMethod=NoPruning")
#if testMethod=="nCuts":
# for nCuts in [5,10,20,50,100,200]:
# factory.BookMethod( TMVA.Types.kBDT, "BDT_ada"+str(phil)+"_200t_50d_0.05b_"+str(nCuts)+"c","!H:!V:NTrees=200:nEventsMin=150:MaxDepth=50:BoostType=AdaBoost:AdaBoostBeta=0.05:SeparationType=GiniIndex:nCuts="+str(nCuts)+":PruneMethod=NoPruning")
#if testMethod=="beta":
# for beta in [0.05,0.5,1.]:
# factory.BookMethod( TMVA.Types.kBDT, "BDT_ada"+str(phil)+"_200t_50d_"+str(beta)+"b_50c","!H:!V:NTrees=200:nEventsMin=150:MaxDepth=50:BoostType=AdaBoost:AdaBoostBeta="+str(beta)+":SeparationType=GiniIndex:nCuts=50:PruneMethod=NoPruning")
# BDT grad
if testType == "grad":
if testMethod == "nTrees":
for nTrees in [10, 50, 100, 200, 500]:
for depth in [2, 3]:
for shrinkage in [0.05, 0.5, 1.]:
factory.BookMethod(
TMVA.Types.kBDT, "BDT_grad" + str(phil) + "_" +
str(nTrees) + "t_" + str(depth) + "d_" +
str(shrinkage) + "s", "!H:!V:NTrees=" +
str(nTrees) + ":MaxDepth=" + str(depth) +
":BoostType=Grad:Shrinkage=" + str(shrinkage) +
":UseBaggedGrad:GradBaggingFraction=1:SeparationType=GiniIndex:nCuts=50:NNodesMax=10"
)
#if testMethod=="depth":
# for depth in [2,3]:
# factory.BookMethod( TMVA.Types.kBDT, "BDT_ada"+str(phil)+"_200t_"+str(depth)+"d_0.05b_50c","!H:!V:NTrees=200:nEventsMin=150:MaxDepth="+str(depth)+":BoostType=AdaBoost:AdaBoostBeta=0.05:SeparationType=GiniIndex:nCuts=50:PruneMethod=NoPruning")
#if testMethod=="shrinkage":
# for shrinkage in [0.05,0.1,0.5,1.]:
# factory.BookMethod( TMVA.Types.kBDT, "BDT_grad"+str(phil)+"_200t_"+str(shrinkage)+"s_1gb_50c_10nm","!H:!V:NTrees=200:BoostType=Grad:Shrinkage="+str(shrinkage)+":UseBaggedGrad:GradBaggingFraction=1:SeparationType=GiniIndex:nCuts=50:NNodesMax=10")
#if testMethod=="bagFrac":
# for bagFrac in [0.05,0.1,0.5,1.]:
# factory.BookMethod( TMVA.Types.kBDT, "BDT_grad"+str(phil)+"_200t_1s_"+str(bagFrac)+"gb_50c_10nm","!H:!V:NTrees=200:BoostType=Grad:Shrinkage=1:UseBaggedGrad:GradBaggingFraction="+str(bagFrac)+":SeparationType=GiniIndex:nCuts=50:NNodesMax=10")
#if testMethod=="nCuts":
# for nCuts in [5,10,20,50,100,200]:
# factory.BookMethod( TMVA.Types.kBDT, "BDT_grad"+str(phil)+"_200t_1s_1gb_"+str(nCuts)+"c_10nm","!H:!V:NTrees=200:BoostType=Grad:Shrinkage=1:UseBaggedGrad:GradBaggingFraction=1:SeparationType=GiniIndex:nCuts="+str(nCuts)+":NNodesMax=10")
#if testMethod=="nNM":
# for nNM in [10,100,500,1000,10000]:
# factory.BookMethod( TMVA.Types.kBDT, "BDT_grad"+str(phil)+"_200t_1s_1gb_50c_"+str(nNM)+"nm","!H:!V:NTrees=200:BoostType=Grad:Shrinkage=1:UseBaggedGrad:GradBaggingFraction=1:SeparationType=GiniIndex:nCuts=50:NNodesMax"+str(nNM))
# --------------------------------------------------------------------------------------------------
# ---- Now you can tell the factory to train, test, and evaluate the MVAs.
# Train MVAs
#factory.OptimizeAllMethods()
factory.TrainAllMethods()
# Test MVAs
factory.TestAllMethods()
# Evaluate MVAs
factory.EvaluateAllMethods()
# Save the output.
outputFile.Close()
print "=== wrote root file %s\n" % outfname
print "=== TMVAClassification is done!\n"
