Belle II Software development
TOPLocalCalFitter.cc
1/**************************************************************************
2 * basf2 (Belle II Analysis Software Framework) *
3 * Author: The Belle II Collaboration *
4 * *
5 * See git log for contributors and copyright holders. *
6 * This file is licensed under LGPL-3.0, see LICENSE.md. *
7 **************************************************************************/
8
9#include <top/calibration/TOPLocalCalFitter.h>
10
11#include "TH2F.h"
12#include "TFile.h"
13#include "TMath.h"
14#include "TF1.h"
15#include "TROOT.h"
16#include "TTree.h"
17#include "math.h"
18#include <limits>
19#include <framework/logging/Logger.h>
20#include <top/dbobjects/TOPCalChannelT0.h>
21
22
23//using namespace std;
24using namespace Belle2;
25using namespace TOP;
26
27double crystalball_function(double x, double alpha, double n, double sigma, double mean)
28{
29 // evaluate the crystal ball function
30 if (sigma < 0.) return 0.;
31 double z = (x - mean) / sigma;
32 if (alpha < 0) z = -z;
33 double abs_alpha = std::abs(alpha);
34 if (z > - abs_alpha)
35 return std::exp(- 0.5 * z * z);
36 else {
37 double nDivAlpha = n / abs_alpha;
38 double AA = std::exp(-0.5 * abs_alpha * abs_alpha);
39 double B = nDivAlpha - abs_alpha;
40 double arg = nDivAlpha / (B - z);
41 return AA * std::pow(arg, n);
42 }
43}
44
45double crystalball_pdf(double x, double alpha, double n, double sigma, double mean)
46{
47 // evaluation of the PDF ( is defined only for n >1)
48 if (sigma < 0.) return 0.;
49 if (n <= 1) return std::numeric_limits<double>::quiet_NaN(); // pdf is not normalized for n <=1
50 double abs_alpha = std::abs(alpha);
51 double C = n / abs_alpha * 1. / (n - 1.) * std::exp(-alpha * alpha / 2.);
52 double D = std::sqrt(M_PI / 2.) * (1. + erf(abs_alpha / std::sqrt(2.)));
53 double N = 1. / (sigma * (C + D));
54 return N * crystalball_function(x, alpha, n, sigma, mean);
55}
56
57
58
59// Two gaussians to model the TTS of the MCP-PMTs.
60double TTSPDF(double x, double time, double deltaT, double sigma1, double sigma2, double f1)
61{
62 return f1 * TMath::Gaus(x, time, sigma1, kTRUE) + (1 - f1) * TMath::Gaus(x, time + deltaT, sigma2, kTRUE) ;
63}
64
65
66// Full PDF to fit the laser, made of:
67// 2 TTSPDF
68// 1 crystal ball PDF for the extra peak at +1 ns we don't understand
69// 1 gaussian to help modelling the tail
70// cppcheck-suppress constParameter
71double laserPDF(double* x, double* p)
72{
73 double time = p[0];
74 double sigma = p[1];
75 double fraction = p[2];
76 double deltaTLaser = p[3];
77 double sigmaRatio = p[4];
78 double deltaTTS = p[5];
79 double f1 = p[6];
80 double norm = p[7];
81
82 double deltaTExtra = p[8];
83 double sigmaExtra = p[9];
84 double normExtra = p[10];
85
86 double deltaTBkg = p[11];
87 double sigmaBkg = p[12];
88 double bkg = p[13];
89
90 double alpha = p[14];
91 double n = p[15];
92
93 double mainPeak = fraction * TTSPDF(x[0], time, deltaTTS, sigma, TMath::Sqrt(sigma * sigma + sigmaRatio * sigmaRatio), f1);
94 double secondaryPeak = (1. - fraction) * TTSPDF(x[0], time + deltaTLaser, deltaTTS, sigma,
95 TMath::Sqrt(sigma * sigma + sigmaRatio * sigmaRatio), f1);
96 double extraPeak = crystalball_pdf(x[0], alpha, n, sigmaExtra, time + deltaTExtra);
97 double background = TMath::Gaus(x[0], time + deltaTBkg + deltaTTS, sigmaBkg, kTRUE);
98
99 return norm * (mainPeak + secondaryPeak) + normExtra * extraPeak + bkg * background;
100}
101
102
104{
106 "Perform the fit of the laser and pulser runs"
107 );
108
109}
110
111
112
114{
115 m_inputTTS = TFile::Open(m_TTSData.c_str());
116 m_inputConstraints = TFile::Open(m_fitConstraints.c_str());
117
118 B2INFO("Getting the TTS parameters from " << m_TTSData);
119 m_inputTTS->cd();
120 m_inputTTS->GetObject("tree", m_treeTTS);
121 m_treeTTS->SetBranchAddress("mean2", &m_mean2);
122 m_treeTTS->SetBranchAddress("sigma1", &m_sigma1);
123 m_treeTTS->SetBranchAddress("sigma2", &m_sigma2);
124 m_treeTTS->SetBranchAddress("fraction1", &m_f1);
125 m_treeTTS->SetBranchAddress("fraction2", &m_f2);
126 m_treeTTS->SetBranchAddress("pixelRow", &m_pixelRow);
127 m_treeTTS->SetBranchAddress("pixelCol", &m_pixelCol);
128
129 if (m_fitterMode == "MC")
130 std::cout << "Running in MC mode, not constraints will be set" << std::endl;
131 else {
132 B2INFO("Getting the laser fit parameters from " << m_fitConstraints);
133 m_inputConstraints->cd();
134 m_inputConstraints->GetObject("fitTree", m_treeConstraints);
135 m_treeConstraints->SetBranchAddress("peakTime", &m_peakTimeConstraints);
136 m_treeConstraints->SetBranchAddress("deltaT", &m_deltaTConstraints);
137 m_treeConstraints->SetBranchAddress("fraction", &m_fractionConstraints);
138 if (m_fitterMode == "monitoring") {
139 m_treeConstraints->SetBranchAddress("timeExtra", &m_timeExtraConstraints);
140 m_treeConstraints->SetBranchAddress("sigmaExtra", &m_sigmaExtraConstraints);
141 m_treeConstraints->SetBranchAddress("alphaExtra", &m_alphaExtraConstraints);
142 m_treeConstraints->SetBranchAddress("nExtra", &m_nExtraConstraints);
143 m_treeConstraints->SetBranchAddress("timeBackground", &m_timeBackgroundConstraints);
144 m_treeConstraints->SetBranchAddress("sigmaBackground", &m_sigmaBackgroundConstraints);
145 }
146 }
147 return;
148}
149
150
152{
153 m_histFile = new TFile(m_output.c_str(), "recreate");
154 m_histFile->cd();
155 m_fitTree = new TTree("fitTree", "fitTree");
156 m_fitTree->Branch<short>("channel", &m_channel);
157 m_fitTree->Branch<short>("slot", &m_slot);
158 m_fitTree->Branch<short>("row", &m_row);
159 m_fitTree->Branch<short>("col", &m_col);
160 m_fitTree->Branch<float>("peakTime", &m_peakTime);
161 m_fitTree->Branch<float>("peakTimeErr", &m_peakTimeErr);
162 m_fitTree->Branch<float>("deltaT", &m_deltaT);
163 m_fitTree->Branch<float>("deltaTErr", &m_deltaTErr);
164 m_fitTree->Branch<float>("sigma", &m_sigma);
165 m_fitTree->Branch<float>("sigmaErr", &m_sigmaErr);
166 m_fitTree->Branch<float>("fraction", &m_fraction);
167 m_fitTree->Branch<float>("fractionErr", &m_fractionErr);
168 m_fitTree->Branch<float>("yieldLaser", &m_yieldLaser);
169 m_fitTree->Branch<float>("yieldLaserErr", &m_yieldLaserErr);
170 m_fitTree->Branch<float>("timeExtra", &m_timeExtra);
171 m_fitTree->Branch<float>("sigmaExtra", &m_sigmaExtra);
172 m_fitTree->Branch<float>("nExtra", &m_nExtra);
173 m_fitTree->Branch<float>("alphaExtra", &m_alphaExtra);
174 m_fitTree->Branch<float>("yieldLaserExtra", &m_yieldLaserExtra);
175 m_fitTree->Branch<float>("timeBackground", &m_timeBackground);
176 m_fitTree->Branch<float>("sigmaBackground", &m_sigmaBackground);
177 m_fitTree->Branch<float>("yieldLaserBackground", &m_yieldLaserBackground);
178
179 m_fitTree->Branch<float>("fractionMC", &m_fractionMC);
180 m_fitTree->Branch<float>("deltaTMC", &m_deltaTMC);
181 m_fitTree->Branch<float>("peakTimeMC", &m_peakTimeMC);
182 m_fitTree->Branch<float>("firstPulserTime", &m_firstPulserTime);
183 m_fitTree->Branch<float>("firstPulserSigma", &m_firstPulserSigma);
184 m_fitTree->Branch<float>("secondPulserTime", &m_secondPulserTime);
185 m_fitTree->Branch<float>("secondPulserSigma", &m_secondPulserSigma);
186 m_fitTree->Branch<short>("fitStatus", &m_fitStatus);
187
188 m_fitTree->Branch<float>("chi2", &m_chi2);
189 m_fitTree->Branch<float>("rms", &m_rms);
190
191
193 m_timewalkTree = new TTree("timewalkTree", "timewalkTree");
194
195 m_timewalkTree->Branch<float>("binLowerEdge", &m_binLowerEdge);
196 m_timewalkTree->Branch<float>("binUpperEdge", &m_binUpperEdge);
197 m_timewalkTree->Branch<short>("channel", &m_channel);
198 m_timewalkTree->Branch<short>("slot", &m_slot);
199 m_timewalkTree->Branch<short>("row", &m_row);
200 m_timewalkTree->Branch<short>("col", &m_col);
201 m_timewalkTree->Branch<float>("histoIntegral", &m_histoIntegral);
202 m_timewalkTree->Branch<float>("peakTime", &m_peakTime);
203 m_timewalkTree->Branch<float>("peakTimeErr", &m_peakTimeErr);
204 m_timewalkTree->Branch<float>("deltaT", &m_deltaT);
205 m_timewalkTree->Branch<float>("deltaTErr", &m_deltaTErr);
206 m_timewalkTree->Branch<float>("sigma", &m_sigma);
207 m_timewalkTree->Branch<float>("sigmaErr", &m_sigmaErr);
208 m_timewalkTree->Branch<float>("fraction", &m_fraction);
209 m_timewalkTree->Branch<float>("fractionErr", &m_fractionErr);
210 m_timewalkTree->Branch<float>("yieldLaser", &m_yieldLaser);
211 m_timewalkTree->Branch<float>("yieldLaserErr", &m_yieldLaserErr);
212 m_timewalkTree->Branch<float>("timeExtra", &m_timeExtra);
213 m_timewalkTree->Branch<float>("sigmaExtra", &m_sigmaExtra);
214 m_timewalkTree->Branch<float>("nExtra", &m_nExtra);
215 m_timewalkTree->Branch<float>("alphaExtra", &m_alphaExtra);
216 m_timewalkTree->Branch<float>("yieldLaserExtra", &m_yieldLaserExtra);
217 m_timewalkTree->Branch<float>("timeBackground", &m_timeBackground);
218 m_timewalkTree->Branch<float>("sigmaBackground", &m_sigmaBackground);
219 m_timewalkTree->Branch<float>("yieldLaserBackground", &m_yieldLaserBackground);
220
221 m_timewalkTree->Branch<float>("fractionMC", &m_fractionMC);
222 m_timewalkTree->Branch<float>("deltaTMC", &m_deltaTMC);
223 m_timewalkTree->Branch<float>("peakTimeMC", &m_peakTimeMC);
224 m_timewalkTree->Branch<float>("firstPulserTime", &m_firstPulserTime);
225 m_timewalkTree->Branch<float>("firstPulserSigma", &m_firstPulserSigma);
226 m_timewalkTree->Branch<float>("secondPulserTime", &m_secondPulserTime);
227 m_timewalkTree->Branch<float>("secondPulserSigma", &m_secondPulserSigma);
228 m_timewalkTree->Branch<short>("fitStatus", &m_fitStatus);
229
230 m_timewalkTree->Branch<float>("chi2", &m_chi2);
231 m_timewalkTree->Branch<float>("rms", &m_rms);
232
233 }
234
235 return;
236}
237
238
239
240
241void TOPLocalCalFitter::fitChannel(short iSlot, short iChannel, TH1* h_profile)
242{
243
244 // loads the TTS infos and the fit constraint for the given channel and slot
245 if (m_fitterMode == "monitoring")
246 m_treeConstraints->GetEntry(iChannel + 512 * iSlot);
247 else if (m_fitterMode == "calibration") // The MC-based constraint file has only slot 1 at the moment
248 m_treeConstraints->GetEntry(iChannel);
249
250 m_treeTTS->GetEntry(iChannel + 512 * iSlot);
251 // finds the maximum of the hit timing histogram and adjust the histogram range around it (3 ns window)
252 double maxpos = h_profile->GetBinCenter(h_profile->GetMaximumBin());
253 h_profile->GetXaxis()->SetRangeUser(maxpos - 1, maxpos + 2.);
254
255 // gets the histogram integral to give a starting value to the fitter
256 double integral = h_profile->Integral();
257
258 // creates the fit function
259 TF1 laser = TF1("laser", laserPDF, maxpos - 1, maxpos + 2., 16);
260
261 // par[0] = peakTime
262 laser.SetParameter(0, maxpos);
263 laser.SetParLimits(0, maxpos - 0.06, maxpos + 0.06);
264
265 // par[1] = sigma
266 laser.SetParameter(1, 0.1);
267 laser.SetParLimits(1, 0.05, 0.25);
268 if (m_fitterMode == "MC") {
269 laser.SetParameter(1, 0.02);
270 laser.SetParLimits(1, 0., 0.04);
271 }
272 // par[2] = fraction of the main peak respect to the total
273 laser.SetParameter(2, m_fractionConstraints);
274 laser.SetParLimits(2, 0.5, 1.);
275
276 // par[3]= time difference between the main and secondary path. fixed to the MC value
277 laser.FixParameter(3, m_deltaTConstraints);
278
279 // This is an hack: in some channels the MC sees one peak only, while in the data there are clearly
280 // two well distinguished peaks. This will disappear if we'll ever get a better laser simulation.
281 if (m_deltaTConstraints > -0.001) {
282 laser.SetParameter(3, -0.3);
283 laser.SetParLimits(3, -0.4, -0.2);
284 }
285
286 // par[4] is the quadratic difference of the sigmas of the two TTS gaussians (tail - core)
287 laser.FixParameter(4, TMath::Sqrt(m_sigma2 * m_sigma2 - m_sigma1 * m_sigma1));
288 // par[5] is the position of the second TTS gaussian w/ respect to the first one
289 laser.FixParameter(5, m_mean2);
290 // par[6] is the relative contribution of the second TTS gaussian
291 laser.FixParameter(6, m_f1);
292 if (m_fitterMode == "MC")
293 laser.FixParameter(6, 0);
294
295 // par[7] is the PDF normalization, = integral*bin width
296 laser.SetParameter(7, integral * 0.005);
297 laser.SetParLimits(7, 0.2 * integral * 0.005, 2.*integral * 0.005);
298
299 // par[8-10] are the relative position, the sigma and the integral of the extra peak
300 laser.SetParameter(8, 1.);
301 laser.SetParLimits(8, 0.3, 2.);
302 laser.SetParameter(9, 0.2);
303 laser.SetParLimits(9, 0.08, 1.);
304 laser.SetParameter(10, 0.1 * integral * 0.005);
305 laser.SetParLimits(10, 0., 0.2 * integral * 0.005);
306 // par[14-15] are the tail parameters of the crystal ball function used to describe the extra peak
307 laser.SetParameter(14, -2.);
308 laser.SetParameter(15, 2.);
309 laser.SetParLimits(15, 1.01, 20.);
310
311 // par[11-13] are relative position, sigma and integral of the broad gaussian added to better describe the tail at high times
312 laser.SetParameter(11, 1.);
313 laser.SetParLimits(11, 0.1, 5.);
314 laser.SetParameter(12, 0.8);
315 laser.SetParLimits(12, 0., 5.);
316 laser.SetParameter(13, 0.01 * integral * 0.005);
317 laser.SetParLimits(13, 0., 0.2 * integral * 0.005);
318
319 // if it's a monitoring fit, fix a buch more parameters.
320 if (m_fitterMode == "monitoring") {
321 laser.FixParameter(2, m_fractionConstraints);
322 laser.FixParameter(3, m_deltaTConstraints);
323 laser.FixParameter(8, m_timeExtraConstraints);
324 laser.FixParameter(9, m_sigmaExtraConstraints);
325 laser.FixParameter(14, m_alphaExtraConstraints);
326 laser.FixParameter(15, m_nExtraConstraints);
327 laser.FixParameter(11, m_timeBackgroundConstraints);
328 laser.FixParameter(12, m_sigmaBackgroundConstraints);
329 }
330
331
332 // if it's a MC fit, fix a buch more parameters.
333 if (m_fitterMode == "MC") {
334 laser.SetParameter(2, 0.8);
335 laser.SetParLimits(2, 0., 1.);
336 laser.SetParameter(3, -0.1);
337 laser.SetParLimits(3, -0.4, -0.);
338 // The following are just random reasonable number, only to pin-point the tail components to some value and remove them form the fit
339 laser.FixParameter(8, 0);
340 laser.FixParameter(9, 0.1);
341 laser.FixParameter(14, -2.);
342 laser.FixParameter(15, 2);
343 laser.FixParameter(11, 1.);
344 laser.FixParameter(12, 0.1);
345 laser.FixParameter(13, 0.);
346 laser.FixParameter(10, 0.);
347 }
348
349
350
351 // make the plot of the fit function nice setting 2000 sampling points
352 laser.SetNpx(2000);
353
354
355 // do the fit!
356 h_profile->Fit("laser", "R L Q");
357
358
359 // Add by hand the different fit components to the histogram, mostly for debugging/presentation purposes
360 TF1* peak1 = new TF1("peak1", laserPDF, maxpos - 1, maxpos + 2., 16);
361 TF1* peak2 = new TF1("peak2", laserPDF, maxpos - 1, maxpos + 2., 16);
362 TF1* extra = new TF1("extra", laserPDF, maxpos - 1, maxpos + 2., 16);
363 TF1* background = new TF1("background", laserPDF, maxpos - 1, maxpos + 2., 16);
364 for (int iPar = 0; iPar < 16; iPar++) {
365 peak1->FixParameter(iPar, laser.GetParameter(iPar));
366 peak2->FixParameter(iPar, laser.GetParameter(iPar));
367 extra->FixParameter(iPar, laser.GetParameter(iPar));
368 background->FixParameter(iPar, laser.GetParameter(iPar));
369 }
370 peak1->FixParameter(2, 0.);
371 peak1->FixParameter(7, (1 - laser.GetParameter(2))*laser.GetParameter(7));
372 peak1->FixParameter(10, 0.);
373 peak1->FixParameter(13, 0.);
374
375 peak2->FixParameter(2, 1.);
376 peak2->FixParameter(7, laser.GetParameter(2)*laser.GetParameter(7));
377 peak2->FixParameter(10, 0.);
378 peak2->FixParameter(13, 0.);
379
380 extra->FixParameter(7, 0.);
381 extra->FixParameter(13, 0.);
382
383 background->FixParameter(7, 0.);
384 background->FixParameter(10, 0.);
385
386 h_profile->GetListOfFunctions()->Add(peak1);
387 h_profile->GetListOfFunctions()->Add(peak2);
388 h_profile->GetListOfFunctions()->Add(extra);
389 h_profile->GetListOfFunctions()->Add(background);
390
391
392 // save the results in the variables linked to the tree branches
393 m_channel = iChannel;
396 m_slot = iSlot + 1;
397 m_peakTime = laser.GetParameter(0);
398 m_peakTimeErr = laser.GetParError(0);
399 m_deltaT = laser.GetParameter(3);
400 m_deltaTErr = laser.GetParError(3);
401 m_sigma = laser.GetParameter(1);
402 m_sigmaErr = laser.GetParError(1);
403 m_fraction = laser.GetParameter(2);
404 m_fractionErr = laser.GetParError(2);
405 m_yieldLaser = laser.GetParameter(7) / 0.005;
406 m_yieldLaserErr = laser.GetParError(7) / 0.005;
407 m_timeExtra = laser.GetParameter(8);
408 m_sigmaExtra = laser.GetParameter(9);
409 m_yieldLaserExtra = laser.GetParameter(10) / 0.005;
410 m_alphaExtra = laser.GetParameter(14);
411 m_nExtra = laser.GetParameter(15);
412 m_timeBackground = laser.GetParameter(11);
413 m_sigmaBackground = laser.GetParameter(12);
414 m_yieldLaserBackground = laser.GetParameter(13) / 0.005;
415 m_chi2 = laser.GetChisquare() / laser.GetNDF();
416
417 // copy some MC information to the output tree
421
422 return;
423}
424
425
426void TOPLocalCalFitter::fitPulser(TH1* h_profileFirstPulser, TH1* h_profileSecondPulser)
427{
428 float maxpos = h_profileFirstPulser->GetBinCenter(h_profileFirstPulser->GetMaximumBin());
429 h_profileFirstPulser->GetXaxis()->SetRangeUser(maxpos - 1, maxpos + 1.);
430 if (h_profileFirstPulser->Integral() > 1000) {
431 TF1 pulser1 = TF1("pulser1", "[0]*TMath::Gaus(x, [1], [2], kTRUE)", maxpos - 1, maxpos + 1.);
432 pulser1.SetParameter(0, 1.);
433 pulser1.SetParameter(1, maxpos);
434 pulser1.SetParameter(2, 0.05);
435 h_profileFirstPulser->Fit("pulser1", "R Q");
436 m_firstPulserTime = pulser1.GetParameter(1);
437 m_firstPulserSigma = pulser1.GetParameter(2);
438 h_profileFirstPulser->Write();
439 } else {
440 m_firstPulserTime = -999;
441 m_firstPulserSigma = -999;
442 }
443
444 maxpos = h_profileSecondPulser->GetBinCenter(h_profileSecondPulser->GetMaximumBin());
445 h_profileSecondPulser->GetXaxis()->SetRangeUser(maxpos - 1, maxpos + 1.);
446 if (h_profileSecondPulser->Integral() > 1000) {
447 TF1 pulser2 = TF1("pulser2", "[0]*TMath::Gaus(x, [1], [2], kTRUE)", maxpos - 1, maxpos + 1.);
448 pulser2.SetParameter(0, 1.);
449 pulser2.SetParameter(1, maxpos);
450 pulser2.SetParameter(2, 0.05);
451 h_profileSecondPulser->Fit("pulser2", "R Q");
452 m_secondPulserTime = pulser2.GetParameter(1);
453 m_secondPulserSigma = pulser2.GetParameter(2);
454 h_profileSecondPulser->Write();
455 } else {
456 m_secondPulserTime = -999;
457 m_secondPulserSigma = -999;
458 }
459 return;
460}
461
462
463
465{
466 if (m_chi2 < 4 && m_sigma < 0.2 && m_yieldLaser > 1000) {
467 m_fitStatus = 0;
468 } else {
469 m_fitStatus = 1;
470 }
471 return;
472}
473
474
476{
477 int nEntries = m_fitTree->GetEntries();
478 if (nEntries != 8192) {
479 B2ERROR("fitTree does not contain an entry wit a fit result for each channel. Found " << nEntries <<
480 " instead of 8192. Perhaps you tried to run the commonT0 calculation before finishing the fitting?");
481 return;
482 }
483
484 // Create and fill the TOPCalChannelT0 object.
485 // This part is mostly copy-pasted from the DB importer used up to Jan 2020
486 auto* channelT0 = new TOPCalChannelT0();
487 short nCal[16] = {0};
488 for (int i = 0; i < nEntries; i++) {
489 m_fitTree->GetEntry(i);
491 if (m_fitStatus == 0) {
492 nCal[m_slot - 1]++;
493 } else {
494 channelT0->setUnusable(m_slot, m_channel);
495 }
496 }
497
498 // Normalize the constants
499 channelT0->suppressAverage();
500
501 // create the localDB
502 saveCalibration(channelT0);
503
504 short nCalTot = 0;
505 B2INFO("Summary: ");
506 for (int iSlot = 1; iSlot < 17; iSlot++) {
507 B2INFO("--> Number of calibrated channels on Slot " << iSlot << " : " << nCal[iSlot - 1] << "/512");
508 B2INFO("--> Cal on ch 1, 256 and 511: " << channelT0->getT0(iSlot, 0) << ", " << channelT0->getT0(iSlot,
509 257) << ", " << channelT0->getT0(iSlot, 511));
510 nCalTot += nCal[iSlot - 1];
511 }
512
513 B2RESULT("Channel T0 calibration constants imported to database, calibrated channels: " << nCalTot << "/ 8192");
514
515
516 // Loop again on the output tree to save the constants there too, adding two more branches.
517 TBranch* channelT0Branch = m_fitTree->Branch<float>("channelT0", &m_channelT0);
518 TBranch* channelT0ErrBranch = m_fitTree->Branch<float>("channelT0Err", &m_channelT0Err);
519
520 for (int i = 0; i < nEntries; i++) {
521 m_fitTree->GetEntry(i);
522 m_channelT0 = channelT0->getT0(m_slot, m_channel);
523 m_channelT0Err = channelT0->getT0Error(m_slot, m_channel);
524 channelT0Branch->Fill();
525 channelT0ErrBranch->Fill();
526 }
527 return ;
528
529}
530
531
532
534{
535 gROOT->SetBatch();
536
538
539
541
542 // Loads the tree with the hits
543 auto hitTree = getObjectPtr<TTree>("hitTree");
544 TH2F* h_hitTime = new TH2F("h_hitTime", " ", 512 * 16, 0., 512 * 16, 22000, -70, 40.); // 5 ps bins
545
546 float amplitude, hitTime;
547 short channel, slot;
548 bool refTimeValid;
549 hitTree->SetBranchAddress("amplitude", &amplitude);
550 hitTree->SetBranchAddress("hitTime", &hitTime);
551 hitTree->SetBranchAddress("channel", &channel);
552 hitTree->SetBranchAddress("slot", &slot);
553 hitTree->SetBranchAddress("refTimeValid", &refTimeValid);
554
555 // An attempt to speed things up looping over the tree only once.
556 std::vector<TH2F*> h_hitTimeLaserHistos = {};
557 for (int iLowerEdge = 0; iLowerEdge < (int)m_binEdges.size() - 1; iLowerEdge++) {
558 TH2F* h_hitTimeLaser = new TH2F(("h_hitTimeLaser_" + std::to_string(iLowerEdge + 1)).c_str(), " ", 512 * 16, 0., 512 * 16, 14000,
559 -70, 0.); // 5 ps bins
560 h_hitTimeLaserHistos.push_back(h_hitTimeLaser);
561 }
562
563 for (unsigned int i = 0; i < hitTree->GetEntries(); i++) {
564 auto onepc = (unsigned int)(hitTree->GetEntries() / 100);
565 if (i % onepc == 0)
566 std::cout << "processing hit " << i << " of " << hitTree->GetEntries() << " (" << i / (onepc * 10) << " %)" << std::endl;
567 hitTree->GetEntry(i);
568 auto it = std::lower_bound(m_binEdges.cbegin(), m_binEdges.cend(), amplitude);
569 int iLowerEdge = std::distance(m_binEdges.cbegin(), it) - 1;
570
571 if (iLowerEdge >= 0 && iLowerEdge < static_cast<int>(m_binEdges.size()) - 1 && refTimeValid)
572 h_hitTimeLaserHistos[iLowerEdge]->Fill(channel + (slot - 1) * 512, hitTime);
573 if (amplitude > 80. && refTimeValid)
574 h_hitTime->Fill(channel + (slot - 1) * 512, hitTime);
575 }
576
577 m_histFile->cd();
578 h_hitTime->Write();
579
580 for (short iSlot = 0; iSlot < 16; iSlot++) {
581 std::cout << "fitting slot " << iSlot + 1 << std::endl;
582 for (short iChannel = 0; iChannel < 512; iChannel++) {
583 TH1D* h_profile = h_hitTime->ProjectionY(("profile_" + std::to_string(iSlot + 1) + "_" + std::to_string(iChannel)).c_str(),
584 iSlot * 512 + iChannel + 1, iSlot * 512 + iChannel + 1);
585 if (m_fitterMode == "MC")
586 h_profile->GetXaxis()->SetRangeUser(-10, -10);
587 else
588 h_profile->GetXaxis()->SetRangeUser(-65,
589 -5); // if you will even change it, make sure not to include the h_hitTime overflow bins in this range
590 fitChannel(iSlot, iChannel, h_profile);
591
593
594 // Now let's fit the pulser
595 TH1D* h_profileFirstPulser = h_hitTime->ProjectionY(("profileFirstPulser_" + std::to_string(iSlot + 1) + "_" + std::to_string(
596 iChannel)).c_str(), iSlot * 512 + iChannel + 1, iSlot * 512 + iChannel + 1);
597 TH1D* h_profileSecondPulser = h_hitTime->ProjectionY(("profileSecondPulser_" + std::to_string(iSlot + 1) + "_" + std::to_string(
598 iChannel)).c_str(), iSlot * 512 + iChannel + 1, iSlot * 512 + iChannel + 1);
599 h_profileFirstPulser->GetXaxis()->SetRangeUser(-10, 10);
600 h_profileSecondPulser->GetXaxis()->SetRangeUser(10, 40);
601
602 fitPulser(h_profileFirstPulser, h_profileSecondPulser);
603
604 m_fitTree->Fill();
605
606 h_profile->Write();
607 h_profileFirstPulser->Write();
608 h_profileSecondPulser->Write();
609 }
610
611 h_hitTime->Write();
612 }
613
615
616 m_fitTree->Write();
617
618
620 std::cout << "Fitting in bins" << std::endl;
621 for (int iLowerEdge = 0; iLowerEdge < (int)m_binEdges.size() - 1; iLowerEdge++) {
622 m_binLowerEdge = m_binEdges[iLowerEdge];
623 m_binUpperEdge = m_binEdges[iLowerEdge + 1];
624 std::cout << "Fitting the amplitude interval (" << m_binLowerEdge << ", " << m_binUpperEdge << " )" << std::endl;
625
626 for (short iSlot = 0; iSlot < 16; iSlot++) {
627 std::cout << " Fitting slot " << iSlot + 1 << std::endl;
628 for (short iChannel = 0; iChannel < 512; iChannel++) {
629 TH1D* h_profile = h_hitTimeLaserHistos[iLowerEdge]->ProjectionY(("profile_" + std::to_string(iSlot + 1) + "_" + std::to_string(
630 iChannel) + "_" + std::to_string(iLowerEdge)).c_str(),
631 iSlot * 512 + iChannel + 1, iSlot * 512 + iChannel + 1);
632 if (m_fitterMode == "MC")
633 h_profile->GetXaxis()->SetRangeUser(-10, -10);
634 else
635 h_profile->GetXaxis()->SetRangeUser(-65,
636 -5); // if you will even change it, make sure not to include the h_hitTime overflow bins in this range
637 fitChannel(iSlot, iChannel, h_profile);
638 m_histoIntegral = h_profile->Integral();
640
641 m_timewalkTree->Fill();
642 h_profile->Write();
643 }
644 }
645 }
646
647 m_timewalkTree->Write();
648 }
649
650 m_histFile->Close();
651
652 return c_OK;
653}
Base class for calibration algorithms.
void saveCalibration(TClonesArray *data, const std::string &name)
Store DBArray payload with given name with default IOV.
void setDescription(const std::string &description)
Set algorithm description (in constructor)
EResult
The result of calibration.
@ c_OK
Finished successfully =0 in Python.
Channel T0 calibration constants for all 512 channels of 16 modules.
short m_fitStatus
Fit quality flag, propagated to the constants.
float m_yieldLaserErr
Statistical error on yield.
float m_binLowerEdge
Lower edge of the amplitude bin in which this fit is performed.
void fitChannel(short, short, TH1 *)
Fits the laser light on one channel.
float m_nExtraConstraints
parameter n of the tail of the extra peak
float m_chi2
Reduced chi2 of the fit.
float m_timeExtraConstraints
Position of the guassian used to describe the extra peak on the timing distribution tail.
float m_f1
Fraction of the first gaussian on the TTS parametrization.
float m_fraction
Fraction of events in the secondary peak.
float m_sigmaExtra
Gaussian sigma of the extra peak in the timing tail
float m_secondPulserSigma
Time resolution from the fit of the first electronic pulse, from a Gaussian fit.
float m_yieldLaserBackground
Integral of the background gaussian.
float m_sigma
Gaussian time resolution, fitted.
float m_peakTimeMC
Time of the main peak in teh MC simulation, i.e.
std::string m_output
Name of the output file.
float m_timeBackground
Position of the gaussian used to describe the background, w/ respect to peakTime.
void loadMCInfoTrees()
loads the TTS parameters and the MC truth info
float m_deltaTMC
Time difference between the main peak and the secondary peak in the MC simulation.
float m_peakTimeErr
Statistical error on peakTime.
short m_channel
Channel number (0-511)
float m_channelT0Err
Statistical error on channelT0.
std::string m_TTSData
File with the Fit constraints and MC info.
std::vector< float > m_binEdges
Amplitude bins.
float m_sigmaBackground
Sigma of the gaussian used to describe the background.
float m_peakTime
Fitted time of the main (i.e.
float m_fractionErr
Statistical error on fraction.
bool m_isFitInAmplitudeBins
Enables the fit in amplitude bins.
void calculateChennelT0()
Calculates the commonT0 calibration after the fits have been done.
float m_alphaExtra
alpha parameter of the tail of the extra peak.
float m_rms
RMS of the histogram used for the fit.
float m_mean2
Position of the second gaussian of the TTS parametrization with respect to the first one.
std::string m_fitConstraints
File with the TTS parametrization.
TTree * m_timewalkTree
Output of the fitter.
float m_yieldLaser
Total number of laser hits from the fitting function integral.
TFile * m_inputTTS
File containing m_treeTTS.
float m_nExtra
parameter n of the tail of the extra peak
float m_alphaExtraConstraints
alpha parameter of the tail of the extra peak.
float m_channelT0
Raw, channelT0 calibration, defined as peakTime-peakTimeMC.
float m_sigmaErr
Statistical error on sigma.
std::string m_fitterMode
Fit mode.
float m_binUpperEdge
Upper edge of the amplitude bin in which this fit is performed.
TFile * m_histFile
Output of the fitter.
TTree * m_treeConstraints
Input to the fitter.
float m_sigmaExtraConstraints
Width of the guassian used to describe the extra peak on the timing distribution tail.
float m_firstPulserSigma
Time resolution from the fit of the first electronic pulse, from a Gaussian fit.
float m_deltaT
Time difference between the main peak and the secondary peak.
TTree * m_fitTree
Output of the fitter.
float m_f2
Fraction of the second gaussian on the TTS parametrization.
void determineFitStatus()
determines if the constant obtained by the fit are good or not
float m_secondPulserTime
Average time of the second electronic pulse respect to the reference pulse, from a gaussian fit.
float m_peakTimeConstraints
Time of the main laser peak in the MC simulation (aka MC correction)
float m_sigma1
Width of the first gaussian on the TTS parametrization.
float m_fractionMC
Fraction of events in the secondary peak form the MC simulation.
float m_sigmaBackgroundConstraints
Sigma of the gaussian used to describe the background.
virtual EResult calibrate() override
Runs the algorithm on events.
TTree * m_treeTTS
Input to the fitter.
void setupOutputTreeAndFile()
prepares the output tree
float m_timeBackgroundConstraints
Position of the gaussian used to describe the background, w/ respect to peakTime.
TFile * m_inputConstraints
File containing m_treeConstraints.
float m_deltaTErr
Statistical error on deltaT.
float m_firstPulserTime
Average time of the first electronic pulse respect to the reference pulse, from a Gaussian fit.
float m_histoIntegral
Integral of the fitted histogram.
float m_yieldLaserExtra
Integral of the extra peak.
float m_sigma2
Width of the second gaussian on the TTS parametrization.
void fitPulser(TH1 *, TH1 *)
Fits the two pulsers.
float m_timeExtra
Position of the extra peak seen in the timing tail, w/ respect to peakTime.
float m_fractionConstraints
Fraction of the main peak.
float m_deltaTConstraints
Distance between the main and the secondary laser peak.
Abstract base class for different kinds of events.