def get_scan_arrays_brm(self, inputHistograms, scan, bunch, scanPlane, algos):
'''
Returns an array for a single bunch in a single
scan. Array contains mu values versus separation
values and their associated errors.
'''
muVdm = l.dict_map(algos, [{} for i in range(len(algos))])
muErr = l.dict_map(algos, [{} for i in range(len(algos))])
delta = []
deltaErr = []
preSeparation = 0.
dipTime = 0.
orbit = 11246
for row in self._scanData[int(scan)-1]:
if row['BUNCHES'] == '' or row['IPSCAN'] != '5' or row['SCAN_FLAG'] == 0: continue
dipTime = r.TDatime(row['DT']).Convert(r.kTRUE)
separation = float(row['SEP'])
plane = row['PLANE']
time = dipTime - self._offsets['PLT']
#print time, dipTime, row['DT']
if plane is not scanPlane or separation == 0.: continue
### Beam parameters ###
delta.append(separation)
### GET RATES ###
prob = {}
prob['OR'] = inputHistograms['OR'].GetBinContent(inputHistograms['OR'].FindBin(time*1e3))/orbit
prob['XOR1'] = inputHistograms['XOR1'].GetBinContent(inputHistograms['XOR1'].FindBin(time*1e3))/orbit
prob['XOR2'] = inputHistograms['XOR2'].GetBinContent(inputHistograms['XOR2'].FindBin(time*1e3))/orbit
prob['AND'] = inputHistograms['AND'].GetBinContent(inputHistograms['AND'].FindBin(time*1e3))/orbit
prob['0'] = 1 - prob['OR']
if prob['OR'] == 0: continue
#print '{}, {}'.format(prob['OR'], time)
### MU-CORRECTIONS ###
mu = {}
mu['OR'] = -1*r.TMath.log(prob['0'])
mu['XOR1'] = r.TMath.log(1 + prob['XOR1']/prob['0'])
mu['XOR2'] = r.TMath.log(1 + prob['XOR2']/prob['0'])
mu['AND'] = -1*r.TMath.log(prob['0']*(1 + prob['XOR1']/prob['0'])*(1 + prob['XOR2']/prob['0']))
mu['AND_ALT'] = -1*r.TMath.log(1 - prob['AND'])
if bunch is not 'Total':
beamTime = int((dipTime - self._offsets['intensity'])/60.)
#print beamTime, dipTime, self._offsets['intensity']
intensity1 = 1. #float(self._bunchData[0][beamTime][int(bunch)])*1e-11
intensity2 = 1. #float(self._bunchData[1][beamTime][int(bunch)])*1e-11
IxI = intensity1*intensity2
for algo in algos:
if prob[algo] == 0: continue
if separation != preSeparation:
muVdm[algo][separation] = 0
muErr[algo][separation] = 0
error = self.calculate_sigma_mu(prob, 11246., algo)
muVdm[algo][separation] += (mu[algo]/IxI)/pow(error/IxI, 2)
muErr[algo][separation] += pow(IxI/error, 2)
preSeparation = separation
# Prepare arrays with scan data for use in fits
delta = sorted(list(set(delta)))
for algo in algos:
for separation in delta:
muVdm[algo][separation] = muVdm[algo][separation]/muErr[algo][separation]
muErr[algo][separation] = math.sqrt(1/muErr[algo][separation])
deltaErr.append(1e-7)
return (muVdm, muErr, delta, deltaErr)
def get_scan_arrays_lumi(self, inputHistograms, scan, bunch, scanPlane, algos):
'''
Uses CSV files provided by lumi group. The files contain
times for the beginning and end of each scan separation.
Returns an array for a single bunch in a single scan.
Array contains mu values versus separation values and
their associated errors.
'''
muVdm = l.dict_map(algos, [{} for i in range(len(algos))])
muErr = l.dict_map(algos, [{} for i in range(len(algos))])
delta = []
deltaErr = []
preSeparation = 0.
dipTime = 0.
orbit = 11246.
for row in self._scanData[int(scan)-1]:
time_i = int(row['TIME_I']) - self._offsets['PLT']
time_f = int(row['TIME_F']) - self._offsets['PLT']
separation = float(row['SEP'])
plane = row['PLANE']
if plane is not scanPlane or separation == 0.: continue
### Beam parameters ###
delta.append(separation)
for time in range(time_i, time_f, 2):
### PROBABILITIES ###
prob = {}
prob['OR'] = inputHistograms['OR'].GetBinContent(inputHistograms['OR'].FindBin(time*1e3))/(2*orbit)
prob['XOR1'] = inputHistograms['XOR1'].GetBinContent(inputHistograms['XOR1'].FindBin(time*1e3))/(2*orbit)
prob['XOR2'] = inputHistograms['XOR2'].GetBinContent(inputHistograms['XOR2'].FindBin(time*1e3))/(2*orbit)
prob['AND'] = inputHistograms['AND'].GetBinContent(inputHistograms['AND'].FindBin(time*1e3))/(2*orbit)
prob['0'] = 1 - prob['OR']
if prob['OR'] == 0: continue
#print '{}, {}'.format(prob['OR'], time)
### MU-CORRECTIONS ###
mu = {}
mu['OR'] = -1*r.TMath.log(prob['0'])
mu['XOR1'] = r.TMath.log(1 + prob['XOR1']/prob['0'])
mu['XOR2'] = r.TMath.log(1 + prob['XOR2']/prob['0'])
mu['AND'] = -1*r.TMath.log(prob['0']*(1 + prob['XOR1']/prob['0'])*(1 + prob['XOR2']/prob['0']))
mu['AND_ALT'] = -1*r.TMath.log(1 - prob['AND'])
if bunch is not 'Total':
iRow = int((int(row['TIME_I']) + time - self._offsets['intensity'])/60.)
#print iRow, time, int(row['TIME_I']), self._offsets['intensity']
intensity1 = float(self._bunchData[0][iRow][int(bunch)])*1e-11
intensity2 = float(self._bunchData[1][iRow][int(bunch)])*1e-11
IxI = intensity1*intensity2
for algo in algos:
if prob[algo] == 0: continue
if separation != preSeparation:
muVdm[algo][separation] = 0
muErr[algo][separation] = 0
error = self.calculate_sigma_mu(prob, 11246., algo)
muVdm[algo][separation] += (mu[algo]/IxI)/pow(error/IxI, 2)
muErr[algo][separation] += pow(IxI/error, 2)
# Prepare arrays with scan data for use in fits
delta = sorted(list(set(delta)))
for algo in algos:
for separation in delta:
muVdm[algo][separation] = muVdm[algo][separation]/muErr[algo][separation]
muErr[algo][separation] = math.sqrt(1/muErr[algo][separation])
deltaErr.append(1e-7)
return (muVdm, muErr, delta, deltaErr)
def multi_scan(self, scans, bcids, pltData, doCorrFit, savePath):
'''
Analyzes multiple scans
'''
scanResults = []
scanResults2D = []
for scan in scans:
fitResults = l.dict_map(['OR', 'AND', 'XOR1', 'XOR2'], [{}, {}, {}, {}])
fitResults2D = l.dict_map(['OR', 'AND', 'XOR1', 'XOR2'], [{}, {}, {}, {}])
resultFile = open(savePath+'/fill'+str(self._fill)+'/scan'+scan+'_fitSummary.txt', 'w')
for bcid in bcids:
bunch = str(bcid)
histograms = {
'OR':pltData.GetDirectory('BCID_'+bunch).Get('p1_RateVsTime_OR'),
'AND': pltData.GetDirectory('BCID_'+bunch).Get('p1_RateVsTime_AND'),
'XOR1': pltData.GetDirectory('BCID_'+bunch).Get('p1_RateVsTime_XOR1'),
'XOR2': pltData.GetDirectory('BCID_'+bunch).Get('p1_RateVsTime_XOR2')
}
fitResult = {'X':[], 'Y':[]}
fitGraphs = {}
mu2D = {'X':[], 'Y':[], 'Z':[]}
mu2DErr = {'X':[], 'Y':[], 'Z':[]}
for plane in ['X', 'Y']:
if self._dataFormat == 'brm':
fitInput = self.get_scan_arrays_brm(histograms, scan, bunch, plane, self._algos)
elif self._dataFormat == 'lumi':
fitInput = self.get_scan_arrays_lumi(histograms, scan, bunch, plane, self._algos)
muOr = (
[value for (key, value) in sorted(zip(fitInput[0]['OR'].keys(), fitInput[0]['OR'].values()))],
[value for (key, value) in sorted(zip(fitInput[1]['OR'].keys(), fitInput[1]['OR'].values()))]
)
vdmGraph = r.TGraphErrors(len(fitInput[2]), array('f', fitInput[2]), array('f', muOr[0]), array('f', fitInput[3]), array('f', muOr[1]))
vdmGraph.SetTitle(plane+'-scan BCID '+bunch+' (BCM1F);#Delta '+plane+' (mm);')
fitResult[plane] = t.Fit(vdmGraph, 'doubleGaussian', True, self._fill, '1', savePath+'/fill'+str(self._fill)+'/scan'+scan+'/vdmScan_1D_OR_BCM1F_BCID-'+bunch+'_'+plane)
# Collect data for correlated fits
fitGraphs[plane] = vdmGraph
if plane == 'X':
mu2D['X'].extend(fitInput[2])
mu2DErr['X'].extend(fitInput[3])
mu2D['Y'].extend([0. for n in range(len(fitInput[2]))])
mu2DErr['Y'].extend([0. for n in range(len(fitInput[2]))])
elif plane == 'Y':
mu2D['X'].extend([0. for n in range(len(fitInput[2]))])
mu2DErr['X'].extend([0. for n in range(len(fitInput[2]))])
mu2D['Y'].extend(fitInput[2])
mu2DErr['Y'].extend(fitInput[3])
mu2D['Z'].extend(muOr[0])
mu2DErr['Z'].extend(muOr[1])
graph2D = r.TGraph2DErrors(len(mu2D['X']), \
array('d', mu2D['X']), array('d', mu2D['Y']), array('d', mu2D['Z']),\
array('d', mu2DErr['X']), array('d', mu2DErr['Y']), array('d', mu2DErr['Z']))
fitResult2D = t.Fit2D(graph2D, fitGraphs['X'], fitGraphs['Y'], True, self._fill, fitResult['X'], fitResult['Y'], savePath+'/fill'+str(self._fill)+'/scan'+scan+'/vdmScan_2D_OR_BCM1F_BCID-'+bunch)
graph2D.Clear()
sigmaEff = r.TMath.Pi()*fitResult['X'][0]*fitResult['Y'][0]*(fitResult['X'][2] + fitResult['Y'][2])/1e30
sigmaErr = (r.TMath.Pi()/1e30)*math.sqrt((pow(fitResult['X'][0]*fitResult['Y'][1], 2) \
+ pow(fitResult['X'][1]*fitResult['Y'][0], 2))*pow(fitResult['X'][2] + fitResult['Y'][2], 2) \
+ pow(fitResult['X'][0]*fitResult['X'][1], 2)*(fitResult['X'][3]*fitResult['X'][3] + fitResult['Y'][3]*fitResult['Y'][3]))
fitResults['OR'][bunch] = (fitResult['X'], fitResult['Y'], (sigmaEff*1e30, sigmaErr*1e30))
fitResults2D['OR'][bunch] = (fitResult2D, (sigmaEff*1e30, sigmaErr*1e30))
subprocess.call('pdftk ' + savePath + '/fill' + str(self._fill) + '/scan' + scan + '/vdmScan_1D*.pdf output ' + savePath + '/fill' + str(self._fill) + '/scan' + scan + '_1D.pdf', shell = True)
subprocess.call('pdftk ' + savePath + '/fill' + str(self._fill) + '/scan' + scan + '/vdmScan_2D*.pdf output ' + savePath + '/fill' + str(self._fill) + '/scan' + scan + '_2D.pdf', shell = True)
l.print_vdm_tables('OR', bcids, fitResults, resultFile, False, False)
scanResults.append(fitResults)
scanResults2D.append(fitResults2D)
return scanResults, scanResults2D
def __init__(self, timeStep = 25, nIter = 72, sourceFile= 'histograms/adcHistograms.root', doEff = False, \
channels = [str(i) for i in range(8)], \
pList = [0.5 for i in range(8)], \
sList = [45 for i in range(8)], \
gList = [(10,12.5) for i in range(8)]
):
# configuration parameters
self._sourceFile = r.TFile(sourceFile, 'OPEN')
self._timeStep = timeStep
self._nObjects = len(channels)
self._nIter = nIter
self._doEff = doEff
# initialize generators, counters
self._rndGen = r.TRandom3(0)
self._counter = 0.
# Specify the list of outcomes that are being tested (diamond, algo, ?)
self._channels = channels
self._algos = ['OR', 'XOR+', 'XOR-', 'AND']
# Channel status booleans
self._isLive = l.dict_map(self._channels, [True for i in range(len(channels))])
self._isTriggered = l.dict_map(self._channels, [[False, False] for i in range(len(channels))]) #[<hit>, <recorded hit>]
self._isOvershoot = l.dict_map(self._channels, [False for i in range(len(channels))])
# Saves pulse characteristics for evolving (for now, just TOT)
self._pulseStatus = l.dict_map(self._channels, [[0, 0, 0] for i in range(len(channels))])
# Input parameters for detector modelling
self._fillPattern = [True for i in range(nIter)] # list of bools saying whether bunch is colliding or not
self._probabilities = l.dict_map(self._channels, pList) # diamond-by-diamond probabilities
self._gate = l.dict_map(self._channels, gList) # gate width and offset w.r.t. center of hit distribution
self._overshoot = l.dict_map(self._channels, sList) # saturationn parameters
#self._tosParams = l.dict_map(self._channels, osList)
# Spectra for modelling detector effects
self._pulseSpectrum = l.dict_map(self._channels, []) # 2D hist for producing pulse height and tot
self._osSpectrum = l.dict_map(self._channels, [r.TF1('f_osSpectrum', 'exp([0]*(x - [1]))', -5000, 0) for i in range(len(channels))])
self._tosSpectrum = l.dict_map(self._channels, [r.TF1('f_tosSpectrum', '[0] + [1]*atan([2]*(x - [3]))', 0, 2000) for i in range(len(channels))])
self._bunchSpectrum = r.TF1('f_bunchSpectrum', 'gaus', 0, 25) # for modelling distribution of colliding hits in time per bunch
self._albedoSpectrum = r.TF1('f_albedoSpectrum', 'pol0', 25, 50) # for modelling distribution of albedo hits in time per bunch
# Outputs
self._hits = l.dict_map(self._channels, [[False for i in range(self._nIter)] for j in range(self._nObjects)])
self._coins = l.dict_map(self._algos, [[False for i in range(self._nIter)] for j in range(len(self._algos))])