def divideScreener(country, process_num = CONFIG.NUM_MULTIPROCESSING):
df_screener = Collector.readFile(country, "Screener")
unit_size = int(len(df_screener) / process_num)
for i in range(process_num):
start_num = i * unit_size
end_num = (i+1) * unit_size
if i != process_num - 1:
df = df_screener.iloc[start_num:end_num, :]
else:
df = df_screener.iloc[start_num:, :]
Collector.saveFile(country, df, "Screener_" + str(i+1))
def __init__(self,
skipHadd=False,
haddAllSamples=False,
haddDataSamples=False,
skipComponentHadd=False):
self.moduleList = []
self.collector = Collector()
self.hadder = Hadder()
self.skipHadd = skipHadd
self.haddAllSamples = haddAllSamples
self.haddDataSamples = haddDataSamples
self.allSampleName = "AllSample"
self.allDataName = "Data"
self.skipComponentHadd = skipComponentHadd
rebin = 3
# ____________________________________________________________________________________________________________________________________________ ||
parser = argparse.ArgumentParser()
parser.add_argument("--inputDir",action="store")
parser.add_argument("--outputDir",action="store")
parser.add_argument("--verbose",action="store_true")
option = parser.parse_args()
# ____________________________________________________________________________________________________________________________________________ ||
outputInfo.outputDir = option.inputDir
outputInfo.TFileName = "ClosureTest.root"
# ____________________________________________________________________________________________________________________________________________ ||
collector = Collector()
collector.makeSampleList(componentList)
collector.makeMergedSampleList(componentList,mergeSampleDict)
collector.openFiles(collector.samples+collector.mergeSamples,outputInfo)
# ____________________________________________________________________________________________________________________________________________ ||
if not os.path.exists(os.path.abspath(option.outputDir)):
os.makedirs(os.path.abspath(option.outputDir))
c = ROOT.TCanvas()
for var in variables:
if option.verbose: print "*"*100
dataHist = collector.getObj(dataSample,var)
if option.verbose: print "Data: ",dataHist.Integral()
outHist = dataHist.Clone(var+"_shape")
outHist.Rebin(rebin)
for ibin in range(1,outHist.GetNbinsX()+1):
def getScreener(country):
collector = Collector()
collector.selectCountry(country)
end_page = collector.getHowManyPages()
collector.getStocksBasicInfoByRange(1,end_page,country)
collector.driver.quit()
def startCrawling(country, process_num):
collector = Collector()
done = False
while done == False:
done = collector.crawlingStart(country, process_num)
option = parser.parse_args()
# ____________________________________________________________________________________________________________________________________________ ||
lnSystFilePath = "/home/lucien/UF-PyNTupleRunner/RA5/StatFW/Config/CommonSyst.txt"
allSampleDir = "AllSample"
fileName = "StatInput.root"
setDataToMC = True
addMCStat = True
sigModel = "SMS-T1qqqqL_1500" if not option.sigModel else option.sigModel
fixWidthStr = "%15s"
fixWidthFloat = "%15.2f"
if option.inputDir: outputInfo.outputDir = option.inputDir
# ____________________________________________________________________________________________________________________________________________ ||
collector = Collector()
collector.makeSampleList(componentList)
collector.makeMergedSampleList(componentList,mergeSampleDict)
collector.openFiles(collector.samples+collector.mergeSamples,outputInfo)
# ____________________________________________________________________________________________________________________________________________ ||
# syst
lnSystReader = LogNormalSystReader()
commonLnSystematics = lnSystReader.makeLnSyst(lnSystFilePath)
if not os.path.exists(os.path.abspath(option.outputDir)):
os.makedirs(os.path.abspath(option.outputDir))
binDict = {}
inputFileAll = ROOT.TFile(os.path.join(option.inputDir,allSampleDir,fileName))
for k in inputFileAll.GetListOfKeys():
class EndSequence(BaseModule):
def __init__(self,
skipHadd=False,
haddAllSamples=False,
haddDataSamples=False,
skipComponentHadd=False):
self.moduleList = []
self.collector = Collector()
self.hadder = Hadder()
self.skipHadd = skipHadd
self.haddAllSamples = haddAllSamples
self.haddDataSamples = haddDataSamples
self.allSampleName = "AllSample"
self.allDataName = "Data"
self.skipComponentHadd = skipComponentHadd
def run(self,
inputInfo,
componentList,
mergeSampleDict={},
mergeSigSampleDict={}):
self.collector.makeSampleList(componentList)
self.collector.makeMergedSampleList(componentList, mergeSampleDict,
mergeSigSampleDict)
for sampleName in self.collector.mergeSamples + self.collector.mergeSigSamples:
self.makedirs(inputInfo.outputDir + sampleName)
if not self.skipHadd:
if not self.skipComponentHadd:
for sampleName in self.collector.samples:
print "Hadding " + sampleName
self.makedirs(inputInfo.outputDir + sampleName)
self.hadder.makeHaddScript(
inputInfo.outputDir + sampleName, [sampleName],
inputInfo)
self.hadder.haddSampleDir(inputInfo.outputDir + sampleName)
if self.haddAllSamples:
print "Hadding " + self.allSampleName
self.makedirs(inputInfo.outputDir + self.allSampleName)
self.hadder.makeHaddScript(
inputInfo.outputDir + self.allSampleName,
self.collector.samples, inputInfo)
self.hadder.haddSampleDir(inputInfo.outputDir +
self.allSampleName)
if self.haddDataSamples and self.collector.dataSamples:
print "Hadding " + self.allDataName
self.makedirs(inputInfo.outputDir + self.allDataName)
self.hadder.makeHaddScript(
inputInfo.outputDir + self.allDataName,
self.collector.dataSamples, inputInfo)
self.hadder.haddSampleDir(inputInfo.outputDir +
self.allDataName)
for sampleName in self.collector.mergeSamples + self.collector.mergeSigSamples:
print "Hadding (and merging) " + sampleName
if sampleName in self.collector.mergeSamples:
tmpDict = self.collector.mergeSampleDict
elif sampleName in self.collector.mergeSigSamples:
tmpDict = self.collector.mergeSigSampleDict
self.hadder.makeHaddScript(inputInfo.outputDir + sampleName,
tmpDict[sampleName], inputInfo)
self.hadder.haddSampleDir(inputInfo.outputDir + sampleName)
self.collector.openFiles(self.collector.samples, inputInfo)
self.collector.openFiles(self.collector.mergeSamples, inputInfo)
self.collector.openFiles(self.collector.mergeSigSamples, inputInfo)
if self.haddAllSamples:
self.collector.openFiles([
self.allSampleName,
], inputInfo)
if self.haddDataSamples:
self.collector.openFiles([
self.allDataName,
], inputInfo)
for module in self.moduleList:
module(self.collector)
self.collector.closeFiles()
def add(self, module):
self.moduleList.append(module)
def __len__(self):
return len(self.moduleList)
def __getitem__(self, index):
if index >= len(self):
raise IndexError
else:
return self.moduleList[index]