4 // script to correct the multiplicity spectrum + helpers
9 gSystem->Load("libPWG0base");
10 AliMultiplicityCorrection::SetQualityRegions(kFALSE);
13 void draw(const char* fileName = "multiplicity.root", const char* folder = "Multiplicity")
17 AliMultiplicityCorrection* mult = new AliMultiplicityCorrection(folder, folder);
19 TFile::Open(fileName);
20 mult->LoadHistograms();
21 mult->DrawHistograms();
23 TH2* hist = (TH2*) gROOT->FindObject("fCorrelation2_zy");
24 canvas = new TCanvas("c1", "c1", 600, 500);
25 canvas->SetTopMargin(0.05);
26 hist->SetStats(kFALSE);
28 hist->SetTitle(";true multiplicity in |#eta| < 1.5;measured multiplicity in |#eta| < 1.5");
29 hist->GetYaxis()->SetTitleOffset(1.1);
30 gPad->SetRightMargin(0.12);
33 canvas->SaveAs("responsematrix.eps");
38 gSystem->Load("libTree");
39 gSystem->Load("libVMC");
41 gSystem->Load("libSTEERBase");
42 gSystem->Load("libANALYSIS");
43 gSystem->Load("libANALYSISalice");
44 gSystem->Load("libPWG0base");
47 void correct(const char* fileNameMC = "multiplicityMC.root", const char* folder = "Multiplicity", const char* fileNameESD = "multiplicityESD.root", Bool_t chi2 = kTRUE, Int_t histID = 2, Bool_t fullPhaseSpace = kFALSE, Float_t beta = 1e3, Int_t eventType = 0 /* AliMultiplicityCorrection::kTrVtx */)
51 AliMultiplicityCorrection* mult = new AliMultiplicityCorrection(folder, folder);
52 TFile::Open(fileNameMC);
53 mult->LoadHistograms();
55 AliMultiplicityCorrection* esd = new AliMultiplicityCorrection(folder, folder);
56 TFile::Open(fileNameESD);
57 esd->LoadHistograms();
59 TH2F* hist = esd->GetMultiplicityESD(histID);
60 TH2F* hist2 = esd->GetMultiplicityMC(histID, eventType);
62 mult->SetMultiplicityESD(histID, hist);
64 // small hack to get around charge conservation for full phase space ;-)
67 TH1* corr = mult->GetCorrelation(histID + 4);
69 for (Int_t i=2; i<=corr->GetNbinsX(); i+=2)
70 for (Int_t j=1; j<=corr->GetNbinsY(); ++j)
72 corr->SetBinContent(i, j, corr->GetBinContent(i-1, j));
73 corr->SetBinError(i, j, corr->GetBinError(i-1, j));
79 mult->SetRegularizationParameters(AliMultiplicityCorrection::kPol1, beta);
80 //mult->SetCreateBigBin(kFALSE);
81 //mult->SetRegularizationParameters(AliMultiplicityCorrection::kNone, 0); //mult->SetCreateBigBin(kFALSE);
82 //mult->SetRegularizationParameters(AliMultiplicityCorrection::kNone, 0, 125); mult->SetCreateBigBin(kFALSE);
83 //mult->SetRegularizationParameters(AliMultiplicityCorrection::kEntropy, 1e5);
84 //mult->SetRegularizationParameters(AliMultiplicityCorrection::kLog, 1e5);
85 //mult->ApplyMinuitFit(histID, fullPhaseSpace, AliMultiplicityCorrection::kTrVtx, kTRUE, hist2->ProjectionY("mymchist"));
86 mult->ApplyMinuitFit(histID, fullPhaseSpace, eventType, kFALSE); //hist2->ProjectionY("mymchist"));
90 mult->ApplyBayesianMethod(histID, fullPhaseSpace, eventType, 0.2, 100);
93 TFile* file = TFile::Open("unfolded.root", "RECREATE");
94 mult->SaveHistograms();
98 mult->DrawComparison((chi2) ? "MinuitChi2" : "Bayesian", histID, fullPhaseSpace, kTRUE, hist2->ProjectionY("mymchist"));
101 void CompareChi2Bayesian(Int_t histID = 2, const char* chi2File = "chi2.root", const char* bayesianFile = "bayesian.root", const char* label1 = "Chi2", const char* label2 = "Bayesian", const char* mcFile = 0, Float_t simpleCorrect = 0)
103 const char* folder = "Multiplicity";
107 AliMultiplicityCorrection* chi2 = new AliMultiplicityCorrection(folder, folder);
108 TFile::Open(chi2File);
109 chi2->LoadHistograms();
111 AliMultiplicityCorrection* bayesian = new AliMultiplicityCorrection(folder, folder);
112 TFile::Open(bayesianFile);
113 bayesian->LoadHistograms();
115 histRAW = chi2->GetMultiplicityESD(histID)->ProjectionY("raw", 1, chi2->GetMultiplicityESD(histID)->GetNbinsX());
116 histRAW->Scale(1.0 / histRAW->Integral());
118 histC = chi2->GetMultiplicityESDCorrected(histID);
119 histB = bayesian->GetMultiplicityESDCorrected(histID);
121 c = new TCanvas("CompareChi2Bayesian", "CompareChi2Bayesian", 800, 600);
122 c->SetRightMargin(0.05);
123 c->SetTopMargin(0.05);
128 histC->SetTitle(";N;P(N)");
129 histC->SetStats(kFALSE);
130 histC->GetXaxis()->SetRangeUser(0, 100);
132 histC->SetLineColor(1);
133 histB->SetLineColor(2);
134 histRAW->SetLineColor(3);
136 histC->DrawCopy("HISTE");
137 histB->DrawCopy("HISTE SAME");
138 histRAW->DrawCopy("SAME");
140 legend = new TLegend(0.2, 0.2, 0.4, 0.4);
141 legend->SetFillColor(0);
143 legend->AddEntry(histC, label1);
144 legend->AddEntry(histB, label2);
145 legend->AddEntry(histRAW, "raw ESD");
148 if (simpleCorrect > 0)
151 graph->SetMarkerStyle(25);
152 graph->SetFillColor(0);
153 for (Int_t bin=1; bin<=histRAW->GetNbinsX(); bin++)
154 graph->SetPoint(graph->GetN(), histRAW->GetXaxis()->GetBinCenter(bin) * simpleCorrect, histRAW->GetBinContent(bin));
156 graph->Draw("PSAME");
157 legend->AddEntry(graph, "weighting");
159 // now create histogram from graph and normalize
160 histGraph = (TH1*) histRAW->Clone();
162 for (Int_t bin=1; bin<=histGraph->GetNbinsX(); bin++)
165 for (j=1; j<graph->GetN(); j++)
166 if (graph->GetX()[j] > histGraph->GetXaxis()->GetBinCenter(bin))
168 if (j == graph->GetN())
170 if (histGraph->GetXaxis()->GetBinCenter(bin) - graph->GetX()[j] < graph->GetX()[j-1] - histGraph->GetXaxis()->GetBinCenter(bin))
172 histGraph->SetBinContent(bin, graph->GetY()[j]);
175 Printf("Integral = %f", histGraph->Integral());
176 histGraph->Scale(1.0 / histGraph->Integral());
178 histGraph->SetLineColor(6);
179 histGraph->DrawCopy("SAME");
180 legend->AddEntry(histGraph, "weighting normalized");
185 AliMultiplicityCorrection* mc = new AliMultiplicityCorrection(folder, folder);
187 mc->LoadHistograms();
189 histMC = mc->GetMultiplicityVtx(histID)->ProjectionY("mc", 1, mc->GetMultiplicityVtx(histID)->GetNbinsX());
191 histMC->Scale(1.0 / histMC->Integral());
193 histMC->Draw("HISTE SAME");
194 histMC->SetLineColor(4);
195 legend->AddEntry(histMC, "MC");
200 c->SaveAs(Form("%s.png", c->GetName()));
201 c->SaveAs(Form("%s.eps", c->GetName()));
208 c = new TCanvas("CompareChi2BayesianRatio", "CompareChi2BayesianRatio", 800, 600);
209 c->SetRightMargin(0.05);
210 c->SetTopMargin(0.05);
214 for (Int_t bin=1; bin<=histC->GetNbinsX(); bin++)
216 if (histMC->GetBinContent(bin) > 0)
218 histC->SetBinContent(bin, histC->GetBinContent(bin) / histMC->GetBinContent(bin));
219 histB->SetBinContent(bin, histB->GetBinContent(bin) / histMC->GetBinContent(bin));
224 histGraph->SetBinContent(bin, histGraph->GetBinContent(bin) / histMC->GetBinContent(bin));
225 histGraph->SetBinError(bin, 0);
230 histC->SetBinError(bin, 0);
231 histB->SetBinError(bin, 0);
235 histC->GetYaxis()->SetRangeUser(0.5, 2);
236 histC->GetYaxis()->SetTitle("Unfolded / MC");
239 histB->Draw("HIST SAME");
242 histGraph->Draw("HIST SAME");
246 if (simpleCorrect > 0)
249 graph2->SetMarkerStyle(25);
250 graph2->SetFillColor(0);
251 for (Int_t i=0; i<graph->GetN(); i++)
253 Float_t mcValue = histMC->GetBinContent(histMC->FindBin(graph->GetX()[i]));
254 Float_t mcError = histMC->GetBinError(histMC->FindBin(graph->GetX()[i]));
256 graph2->SetPoint(graph2->GetN(), graph->GetX()[i], graph->GetY()[i] / mcValue);
258 graph2->Draw("PSAME");
262 c->SaveAs(Form("%s.png", c->GetName()));
263 c->SaveAs(Form("%s.eps", c->GetName()));
267 void CompareMC(Int_t histID, Int_t eventType, const char* file1, const char* file2, const char* label1, const char* label2)
269 const char* folder = "Multiplicity";
273 AliMultiplicityCorrection* mc1 = new AliMultiplicityCorrection(folder, folder);
275 mc1->LoadHistograms();
276 histMC1 = mc1->GetMultiplicityMC(histID, eventType)->ProjectionY("mc1", 1, mc1->GetMultiplicityMC(histID, eventType)->GetNbinsX());
277 histMC1->Scale(1.0 / histMC1->Integral());
279 AliMultiplicityCorrection* mc2 = new AliMultiplicityCorrection(folder, folder);
281 mc2->LoadHistograms();
282 histMC2 = mc2->GetMultiplicityMC(histID, eventType)->ProjectionY("mc2", 1, mc2->GetMultiplicityMC(histID, eventType)->GetNbinsX());
283 histMC2->Scale(1.0 / histMC2->Integral());
285 c = new TCanvas("CompareMC", "CompareMC", 800, 600);
286 c->SetRightMargin(0.05);
287 c->SetTopMargin(0.05);
290 histMC1->SetTitle(";N;P(N)");
291 histMC1->SetStats(kFALSE);
292 histMC1->GetXaxis()->SetRangeUser(0, 100);
294 histMC1->SetLineColor(1);
295 histMC2->SetLineColor(2);
298 histMC2->Draw("SAME");
300 legend = new TLegend(0.2, 0.2, 0.4, 0.4);
301 legend->SetFillColor(0);
303 legend->AddEntry(histMC1, label1);
304 legend->AddEntry(histMC2, label2);
308 c->SaveAs(Form("%s.gif", c->GetName()));
309 c->SaveAs(Form("%s.eps", c->GetName()));
312 void* fit2Step(const char* fileNameMC = "multiplicityMC_2M.root", const char* fileNameESD = "multiplicityMC_1M_3.root", Int_t histID = 3, Bool_t fullPhaseSpace = kFALSE)
314 gSystem->Load("libPWG0base");
316 AliMultiplicityCorrection* mult = new AliMultiplicityCorrection("Multiplicity", "Multiplicity");
318 TFile::Open(fileNameMC);
319 mult->LoadHistograms("Multiplicity");
321 TFile::Open(fileNameESD);
322 TH2F* hist = (TH2F*) gFile->Get(Form("Multiplicity/fMultiplicityESD%d", histID));
323 TH2F* hist2 = (TH2F*) gFile->Get(Form("Multiplicity/fMultiplicityVtx%d", ((fullPhaseSpace) ? 4 : histID)));
324 //hist2 = (TH2F*) gFile->Get(Form("Multiplicity/fMultiplicityINEL%d", histID));
326 mult->SetMultiplicityESD(histID, hist);
328 // small hack to get around charge conservation for full phase space ;-)
331 TH1* corr = mult->GetCorrelation(histID + 4);
333 for (Int_t i=2; i<=corr->GetNbinsX(); i+=2)
334 for (Int_t j=1; j<=corr->GetNbinsY(); ++j)
336 corr->SetBinContent(i, j, corr->GetBinContent(i-1, j));
337 corr->SetBinError(i, j, corr->GetBinError(i-1, j));
341 mult->SetRegularizationParameters(AliMultiplicityCorrection::kPol1, 10000);
342 mult->ApplyMinuitFit(histID, fullPhaseSpace, AliMultiplicityCorrection::kTrVtx, kFALSE);
343 mult->DrawComparison("MinuitChi2", histID, fullPhaseSpace, kTRUE, hist2->ProjectionY("mymchist"));
345 TH1* result = (TH1*) mult->GetMultiplicityESDCorrected((fullPhaseSpace) ? 4 : histID))->Clone("firstresult");
347 mult->SetRegularizationParameters(AliMultiplicityCorrection::kEntropy, 100000);
348 mult->ApplyMinuitFit(histID, fullPhaseSpace, AliMultiplicityCorrection::kTrVtx, kFALSE, result);
349 mult->DrawComparison("MinuitChi2_Step2", histID, fullPhaseSpace, kTRUE, hist2->ProjectionY("mymchist"));
354 const char* GetRegName(Int_t type)
358 case AliMultiplicityCorrection::kNone: return "None"; break;
359 case AliMultiplicityCorrection::kPol0: return "Pol0"; break;
360 case AliMultiplicityCorrection::kPol1: return "Pol1"; break;
361 case AliMultiplicityCorrection::kCurvature: return "TotalCurvature"; break;
362 case AliMultiplicityCorrection::kEntropy: return "Reduced cross-entropy"; break;
363 case AliMultiplicityCorrection::kLog : return "Log"; break;
368 const char* GetEventTypeName(Int_t type)
372 case AliMultiplicityCorrection::kTrVtx: return "trigger, vertex"; break;
373 case AliMultiplicityCorrection::kMB: return "minimum bias"; break;
374 case AliMultiplicityCorrection::kINEL: return "inelastic"; break;
379 void EvaluateBayesianMethodIterationsSmoothing(const char* fileNameMC = "multiplicityMC.root", const char* fileNameESD = "multiplicityMC.root", const char* targetDir, Int_t histID = 3)
381 gSystem->mkdir(targetDir);
383 gSystem->Load("libPWG0base");
385 AliMultiplicityCorrection* mult = new AliMultiplicityCorrection("Multiplicity", "Multiplicity");
386 TFile::Open(fileNameMC);
387 mult->LoadHistograms("Multiplicity");
389 AliMultiplicityCorrection* multESD = new AliMultiplicityCorrection("MultiplicityESD", "MultiplicityESD");
390 TFile::Open(fileNameESD);
391 multESD->LoadHistograms("Multiplicity");
392 mult->SetMultiplicityESD(histID, multESD->GetMultiplicityESD(histID));
394 Int_t count = 0; // just to order the saved images...
396 TFile* graphFile = TFile::Open(Form("%s/EvaluateBayesianMethodIterationsSmoothing.root", targetDir), "RECREATE");
398 Int_t colors[3] = {1, 2, 4};
399 Int_t markers[20] = {20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 3, 4, 5, 6};
401 for (AliMultiplicityCorrection::EventType type = AliMultiplicityCorrection::kTrVtx; type <= AliMultiplicityCorrection::kTrVtx; ++type)
402 //for (AliMultiplicityCorrection::EventType type = AliMultiplicityCorrection::kTrVtx; type <= AliMultiplicityCorrection::kINEL; ++type)
405 tmp.Form("EvaluateBayesianMethodIterationsSmoothing_%s", GetEventTypeName(type));
407 TCanvas* canvas = new TCanvas(tmp, tmp, 800, 600);
409 for (Int_t i = 1; i <= 5; i++)
411 Int_t iterArray[5] = {5, 20, 50, 100, -1};
412 //Int_t iter = i * 40 - 20;
413 Int_t iter = iterArray[i-1];
415 TGraph* fitResultsMC[3];
416 for (Int_t region=0; region<AliMultiplicityCorrection::kQualityRegions; ++region)
418 fitResultsMC[region] = new TGraph;
419 fitResultsMC[region]->SetTitle(Form("%d iter. - reg. %d", iter, region+1));
420 fitResultsMC[region]->GetXaxis()->SetTitle("smoothing parameter #alpha");
421 fitResultsMC[region]->GetYaxis()->SetTitle(Form("P_{1} in region %d", region));
422 fitResultsMC[region]->SetName(Form("%s_MC_%d", tmp.Data(), i * AliMultiplicityCorrection::kQualityRegions + region - 2));
423 fitResultsMC[region]->SetFillColor(0);
424 //fitResultsMC[region]->SetMarkerStyle(markers[(i-1) * AliMultiplicityCorrection::kQualityRegions + region]);
425 fitResultsMC[region]->SetMarkerStyle(markers[(i-1)]);
426 fitResultsMC[region]->SetLineColor(colors[region]);
429 TGraph* fitResultsRes = new TGraph;
430 fitResultsRes->SetTitle(Form("%d iterations", iter));
431 fitResultsRes->SetName(Form("%s_Res_%d", tmp.Data(), i));
432 fitResultsRes->GetXaxis()->SetTitle("smoothing parameter");
433 fitResultsRes->GetYaxis()->SetTitle("P_{2}");
435 fitResultsRes->SetFillColor(0);
436 fitResultsRes->SetMarkerStyle(19+i);
437 fitResultsRes->SetMarkerColor(1);
438 fitResultsRes->SetLineColor(1);
440 for (Float_t weight = 0.0; weight < 1.01; weight += 0.2)
443 Float_t residuals = 0;
445 mult->ApplyBayesianMethod(histID, kFALSE, type, weight, iter, 0, kFALSE);
446 mult->DrawComparison(Form("%s/BayesianIterSmooth_%03d_%d_%d_%f", targetDir, count++, type, iter, weight), histID, kFALSE, kTRUE, multESD->GetMultiplicityMC(histID, type)->ProjectionY());
447 mult->GetComparisonResults(&chi2MC, 0, &residuals);
449 for (Int_t region=0; region<AliMultiplicityCorrection::kQualityRegions; ++region)
450 fitResultsMC[region]->SetPoint(fitResultsMC[region]->GetN(), weight, mult->GetQuality(region));
452 fitResultsRes->SetPoint(fitResultsRes->GetN(), weight, residuals);
456 for (Int_t region=0; region<AliMultiplicityCorrection::kQualityRegions; ++region)
457 fitResultsMC[region]->Write();
459 fitResultsRes->Write();
466 void EvaluateDrawResult(const char* targetDir, Int_t type = 0, Bool_t plotRes = kTRUE)
468 gSystem->Load("libPWG0base");
471 TFile* graphFile = 0;
474 name = "EvaluateChi2Method";
475 graphFile = TFile::Open(Form("%s/EvaluateChi2Method.root", targetDir));
479 name.Form("EvaluateBayesianMethodIterationsSmoothing_%s", GetEventTypeName(type));
480 graphFile = TFile::Open(Form("%s/EvaluateBayesianMethodIterationsSmoothing.root", targetDir));
483 TCanvas* canvas = new TCanvas(name, name, 800, 500);
489 canvas->SetTopMargin(0.05);
493 TLegend* legend = new TLegend(0.8, 0.15, 0.98, 0.98);
494 legend->SetFillColor(0);
504 Float_t yMinRegion[3];
505 for (Int_t i=0; i<AliMultiplicityCorrection::kQualityRegions; ++i)
506 yMinRegion[i] = 1e20;
508 TString xaxis, yaxis;
512 TGraph* mc = (TGraph*) graphFile->Get(Form("%s_MC_%d", name.Data(), count));
513 TGraph* res = (TGraph*) graphFile->Get(Form("%s_Res_%d", name.Data(), count));
518 xaxis = mc->GetXaxis()->GetTitle();
519 yaxis = mc->GetYaxis()->GetTitle();
526 xMin = TMath::Min(xMin, mc->GetXaxis()->GetXmin());
527 yMin = TMath::Min(yMin, mc->GetYaxis()->GetXmin());
529 xMax = TMath::Max(xMax, mc->GetXaxis()->GetXmax());
530 yMax = TMath::Max(yMax, mc->GetYaxis()->GetXmax());
534 xMin = TMath::Min(xMin, res->GetXaxis()->GetXmin());
535 yMin = TMath::Min(yMin, res->GetYaxis()->GetXmin());
537 xMax = TMath::Max(xMax, res->GetXaxis()->GetXmax());
538 yMax = TMath::Max(yMax, res->GetYaxis()->GetXmax());
541 for (Int_t i=0; i<mc->GetN(); ++i)
542 yMinRegion[(count-1) % 3] = TMath::Min(yMinRegion[(count-1) % 3], mc->GetY()[i]);
547 for (Int_t i=0; i<AliMultiplicityCorrection::kQualityRegions; ++i)
548 Printf("Minimum for region %d is %f", i, yMinRegion[i]);
552 xaxis = "smoothing parameter";
556 xaxis = "weight parameter";
559 //yaxis = "P_{1} (2 <= t <= 150)";
561 printf("%f %f %f %f\n", xMin, xMax, yMin, yMax);
563 TGraph* dummy = new TGraph;
564 dummy->SetPoint(0, xMin, yMin);
565 dummy->SetPoint(1, xMax, yMax);
566 dummy->SetTitle(Form(";%s;%s", xaxis.Data(), yaxis.Data()));
568 dummy->SetMarkerColor(0);
570 dummy->GetYaxis()->SetMoreLogLabels(1);
576 TGraph* mc = (TGraph*) graphFile->Get(Form("%s_MC_%d", name.Data(), count));
577 TGraph* res = (TGraph*) graphFile->Get(Form("%s_Res_%d", name.Data(), count));
579 //printf("%s_MC_%d %p %p\n", name.Data(), count, mc, res);
584 printf("Loaded %d sucessful.\n", count);
588 legend->AddEntry(mc, Form("Eq. (%d) - reg. %d", 10 + (count-1) / 3, 1+ (count-1) % 3));
591 legend->AddEntry(mc);
597 legend->AddEntry(res);
598 res->Draw("SAME PC");
606 canvas->SaveAs(Form("%s/%s.gif", targetDir, canvas->GetName()));
607 canvas->SaveAs(Form("%s/%s.eps", targetDir, canvas->GetName()));
610 void EvaluateDrawResultRegions(const char* targetDir, Int_t type = 0)
612 gSystem->Load("libPWG0base");
615 TFile* graphFile = 0;
618 name = "EvaluateChi2Method";
619 graphFile = TFile::Open(Form("%s/EvaluateChi2Method.root", targetDir));
623 name.Form("EvaluateBayesianMethodIterationsSmoothing_%s", GetEventTypeName(type));
624 graphFile = TFile::Open(Form("%s/EvaluateBayesianMethodIterationsSmoothing.root", targetDir));
627 TCanvas* canvas = new TCanvas(name, name, 800, 1200);
628 //canvas->Divide(1, AliMultiplicityCorrection::kQualityRegions, 0, 0);
629 canvas->Range(0, 0, 1, 1);
632 pad[0] = new TPad(Form("%s_pad1", name.Data()), "", 0.02, 0.05, 0.98, 0.35);
633 pad[1] = new TPad(Form("%s_pad2", name.Data()), "", 0.02, 0.35, 0.98, 0.65);
634 pad[2] = new TPad(Form("%s_pad3", name.Data()), "", 0.02, 0.65, 0.98, 0.95);
636 Float_t yMinRegion[3];
637 for (Int_t i=0; i<AliMultiplicityCorrection::kQualityRegions; ++i)
638 yMinRegion[i] = 1e20;
640 for (Int_t region = 1; region <= AliMultiplicityCorrection::kQualityRegions; region++)
643 pad[region-1]->Draw();
645 pad[region-1]->SetRightMargin(0.05);
648 pad[region-1]->SetBottomMargin(0);
649 if (region != AliMultiplicityCorrection::kQualityRegions)
650 pad[region-1]->SetTopMargin(0);
654 pad[region-1]->SetLogx();
655 pad[region-1]->SetLogy();
657 pad[region-1]->SetTopMargin(0.05);
658 pad[region-1]->SetGridx();
659 pad[region-1]->SetGridy();
661 TLegend* legend = new TLegend(0.8, 0.15, 0.98, 0.98);
662 legend->SetFillColor(0);
672 TString xaxis, yaxis;
678 TGraph* mc = (TGraph*) graphFile->Get(Form("%s_MC_%d", name.Data(), count));
682 if (TString(mc->GetTitle()).Contains(Form("reg. %d", region)) == kFALSE)
685 xaxis = mc->GetXaxis()->GetTitle();
686 yaxis = mc->GetYaxis()->GetTitle();
690 xMin = TMath::Min(xMin, mc->GetXaxis()->GetXmin());
691 yMin = TMath::Min(yMin, mc->GetYaxis()->GetXmin());
693 xMax = TMath::Max(xMax, mc->GetXaxis()->GetXmax());
694 yMax = TMath::Max(yMax, mc->GetYaxis()->GetXmax());
696 for (Int_t i=0; i<mc->GetN(); ++i)
697 yMinRegion[(count-1) % 3] = TMath::Min(yMinRegion[(count-1) % 3], mc->GetY()[i]);
702 xaxis = "smoothing parameter";
706 xaxis = "weight parameter";
709 yaxis = "P_{1}"; // (2 <= t <= 150)";
711 printf("%f %f %f %f\n", xMin, xMax, yMin, yMax);
713 TGraph* dummy = new TGraph;
714 dummy->SetPoint(0, xMin, yMin);
715 dummy->SetPoint(1, xMax, yMax);
716 dummy->SetTitle(Form(";%s;%s", xaxis.Data(), yaxis.Data()));
718 dummy->SetMarkerColor(0);
720 dummy->GetYaxis()->SetMoreLogLabels(1);
728 TGraph* mc = (TGraph*) graphFile->Get(Form("%s_MC_%d", name.Data(), count));
731 if (TString(mc->GetTitle()).Contains(Form("reg. %d", region)) == kFALSE)
734 printf("Loaded %d sucessful.\n", count);
738 legend->AddEntry(mc, Form("Eq. (%d) - reg. %d", 10 + (count-1) / 3, 1+ (count-1) % 3));
741 legend->AddEntry(mc);
750 for (Int_t i=0; i<AliMultiplicityCorrection::kQualityRegions; ++i)
751 Printf("Minimum for region %d is %f", i, yMinRegion[i]);
754 canvas->SaveAs(Form("%s/%s.gif", targetDir, canvas->GetName()));
755 canvas->SaveAs(Form("%s/%s.eps", targetDir, canvas->GetName()));
758 void EvaluateChi2Method(const char* fileNameMC = "multiplicityMC_2M.root", const char* fileNameESD = "multiplicityMC_1M_3.root", const char* targetDir, Int_t histID = 3)
760 gSystem->mkdir(targetDir);
762 gSystem->Load("libPWG0base");
764 AliMultiplicityCorrection* mult = new AliMultiplicityCorrection("Multiplicity", "Multiplicity");
766 TFile::Open(fileNameMC);
767 mult->LoadHistograms("Multiplicity");
769 TFile::Open(fileNameESD);
770 TH2F* hist = (TH2F*) gFile->Get(Form("Multiplicity/fMultiplicityESD%d", histID));
771 TH2F* hist2 = (TH2F*) gFile->Get(Form("Multiplicity/fMultiplicityVtx%d", histID));
773 mult->SetMultiplicityESD(histID, hist);
775 Int_t count = 0; // just to order the saved images...
776 Int_t colors[3] = {1, 2, 4};
777 Int_t markers[12] = {20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 3};
779 TGraph* fitResultsRes = 0;
781 TFile* graphFile = TFile::Open(Form("%s/EvaluateChi2Method.root", targetDir), "RECREATE");
783 for (AliMultiplicityCorrection::RegularizationType type = AliMultiplicityCorrection::kPol0; type <= AliMultiplicityCorrection::kEntropy; ++type)
784 // for (AliMultiplicityCorrection::RegularizationType type = AliMultiplicityCorrection::kEntropy; type <= AliMultiplicityCorrection::kEntropy; ++type)
785 //for (AliMultiplicityCorrection::RegularizationType type = AliMultiplicityCorrection::kPol1; type <= AliMultiplicityCorrection::kPol1; ++type)
787 TGraph* fitResultsMC[3];
788 for (Int_t region=0; region<AliMultiplicityCorrection::kQualityRegions; ++region)
790 fitResultsMC[region] = new TGraph;
791 fitResultsMC[region]->SetTitle(Form("Eq. (%d) - reg. %d", type+9, region+1));
792 fitResultsMC[region]->GetXaxis()->SetTitle("weight parameter #alpha");
793 fitResultsMC[region]->GetYaxis()->SetTitle(Form("P_{1} in region %d", region));
794 fitResultsMC[region]->SetName(Form("EvaluateChi2Method_MC_%d", type * AliMultiplicityCorrection::kQualityRegions + region - 2));
795 fitResultsMC[region]->SetFillColor(0);
796 fitResultsMC[region]->SetMarkerStyle(markers[(type-1) * AliMultiplicityCorrection::kQualityRegions + region]);
797 fitResultsMC[region]->SetLineColor(colors[region]);
800 fitResultsRes = new TGraph;
801 fitResultsRes->SetTitle(Form("%s residual chi2", GetRegName(type)));
802 fitResultsRes->SetName(Form("EvaluateChi2Method_Res_%d", type));
803 fitResultsRes->GetXaxis()->SetTitle("Weight Parameter");
805 fitResultsRes->SetFillColor(0);
806 fitResultsRes->SetMarkerStyle(23+type);
807 fitResultsRes->SetMarkerColor(kRed);
808 fitResultsRes->SetLineColor(kRed);
810 for (Int_t i=0; i<7; ++i)
812 Float_t weight = TMath::Power(TMath::Sqrt(10), i+6);
813 //Float_t weight = TMath::Power(10, i+2);
815 //if (type == AliMultiplicityCorrection::kEntropy) weight = 1e4 * (i+1) * 1.5;
818 Float_t residuals = 0;
819 Float_t chi2Limit = 0;
822 runName.Form("MinuitChi2_%02d_%d_%f", count++, type, weight);
824 mult->SetRegularizationParameters(type, weight);
825 Int_t status = mult->ApplyMinuitFit(histID, kFALSE, AliMultiplicityCorrection::kTrVtx);
826 mult->DrawComparison(Form("%s/%s", targetDir, runName.Data()), histID, kFALSE, kTRUE, hist2->ProjectionY());
829 printf("MINUIT did not succeed. Skipping...\n");
833 mult->GetComparisonResults(&chi2MC, 0, &residuals);
834 TH1* result = mult->GetMultiplicityESDCorrected(histID);
835 result->SetName(runName);
838 for (Int_t region=0; region<AliMultiplicityCorrection::kQualityRegions; ++region)
839 fitResultsMC[region]->SetPoint(fitResultsMC[region]->GetN(), weight, mult->GetQuality(region));
841 fitResultsRes->SetPoint(fitResultsRes->GetN(), weight, residuals);
845 for (Int_t region=0; region<AliMultiplicityCorrection::kQualityRegions; ++region)
846 fitResultsMC[region]->Write();
847 fitResultsRes->Write();
853 void EvaluateChi2MethodAll()
855 EvaluateChi2Method("multiplicityMC_3M.root", "multiplicityMC_3M.root", "eval-3M-3M");
856 EvaluateChi2Method("multiplicityMC_2M.root", "multiplicityMC_1M_3.root", "eval-2M-1M");
857 EvaluateChi2Method("multiplicityMC_3M.root", "multiplicityMC_3M_NBD.root", "eval-3M-NBD");
858 EvaluateChi2Method("multiplicityMC_2M_smoothed.root", "multiplicityMC_1M_3.root", "eval-2MS-1M");
859 EvaluateChi2Method("multiplicityMC_2M_smoothed.root", "multiplicityMC_3M_NBD.root", "eval-2MS-NBD");
862 void EvaluateBayesianMethodAll()
864 EvaluateBayesianMethod("multiplicityMC_3M.root", "multiplicityMC_3M.root", "eval-3M-3M");
865 EvaluateBayesianMethod("multiplicityMC_2M.root", "multiplicityMC_1M_3.root", "eval-2M-1M");
866 EvaluateBayesianMethod("multiplicityMC_3M.root", "multiplicityMC_3M_NBD.root", "eval-3M-NBD");
867 EvaluateBayesianMethod("multiplicityMC_2M_smoothed.root", "multiplicityMC_1M_3.root", "eval-2MS-1M");
868 EvaluateBayesianMethod("multiplicityMC_2M_smoothed.root", "multiplicityMC_3M_NBD.root", "eval-2MS-NBD");
871 void CompareMethods(const char* fileNameMC = "multiplicityMC.root", const char* fileNameESD = "multiplicityMC.root", const char* targetDir, Int_t histID = 3)
873 gSystem->mkdir(targetDir);
875 gSystem->Load("libPWG0base");
877 AliMultiplicityCorrection* mult = new AliMultiplicityCorrection("Multiplicity", "Multiplicity");
879 TFile::Open(fileNameMC);
880 mult->LoadHistograms("Multiplicity");
882 TFile::Open(fileNameESD);
883 AliMultiplicityCorrection* multESD = new AliMultiplicityCorrection("MultiplicityESD", "MultiplicityESD");
884 multESD->LoadHistograms("Multiplicity");
886 mult->SetMultiplicityESD(histID, multESD->GetMultiplicityESD(histID));
888 TCanvas* canvas = new TCanvas("CompareMethods", "CompareMethods", 1200, 1200);
889 canvas->Divide(3, 3);
893 for (AliMultiplicityCorrection::EventType type = AliMultiplicityCorrection::kTrVtx; type <= AliMultiplicityCorrection::kTrVtx; ++type)
895 TH1* mc = multESD->GetMultiplicityMC(histID, type)->ProjectionY("mymc");
898 mult->SetRegularizationParameters(AliMultiplicityCorrection::kPol1, 10000);
899 mult->ApplyMinuitFit(histID, kFALSE, type);
900 mult->DrawComparison(Form("%s/CompareMethods_%d_MinuitChi2", targetDir, type), histID, kFALSE, kTRUE, mc);
901 TH1* chi2Result = (TH1*) mult->GetMultiplicityESDCorrected(histID)->Clone("chi2Result");
903 mult->ApplyBayesianMethod(histID, kFALSE, type, 0.1);
904 mult->DrawComparison(Form("%s/CompareMethods_%d_Bayesian", targetDir, type), histID, kFALSE, kTRUE, mc);
905 TH1* bayesResult = (TH1*) mult->GetMultiplicityESDCorrected(histID)->Clone("bayesResult");
907 mc->GetXaxis()->SetRangeUser(0, 150);
908 chi2Result->GetXaxis()->SetRangeUser(0, 150);
910 /* // skip errors for now
911 for (Int_t i=1; i<=chi2Result->GetNbinsX(); ++i)
913 chi2Result->SetBinError(i, 0);
914 bayesResult->SetBinError(i, 0);
918 mc->SetFillColor(kYellow);
920 chi2Result->SetLineColor(kRed);
921 chi2Result->DrawCopy("SAME");
922 bayesResult->SetLineColor(kBlue);
923 bayesResult->DrawCopy("SAME");
926 canvas->cd(count + 3);
927 chi2ResultRatio = (TH1*) chi2Result->Clone("chi2ResultRatio");
928 bayesResultRatio = (TH1*) bayesResult->Clone("bayesResultRatio");
929 chi2ResultRatio->Divide(chi2Result, mc, 1, 1, "");
930 bayesResultRatio->Divide(bayesResult, mc, 1, 1, "");
932 chi2ResultRatio->GetYaxis()->SetRangeUser(0.5, 1.5);
934 chi2ResultRatio->DrawCopy("HIST");
935 bayesResultRatio->DrawCopy("SAME HIST");
937 canvas->cd(count + 6);
938 chi2Result->Divide(chi2Result, bayesResult, 1, 1, "");
939 chi2Result->GetYaxis()->SetRangeUser(0.5, 1.5);
940 chi2Result->DrawCopy("HIST");
943 canvas->SaveAs(Form("%s/%s.gif", targetDir, canvas->GetName()));
944 canvas->SaveAs(Form("%s/%s.C", targetDir, canvas->GetName()));
947 void StatisticsPlot(const char* fileNameMC = "multiplicityMC_2M.root", Int_t histID = 3)
949 gSystem->Load("libPWG0base");
951 AliMultiplicityCorrection* mult = new AliMultiplicityCorrection("Multiplicity", "Multiplicity");
953 TFile::Open(fileNameMC);
954 mult->LoadHistograms("Multiplicity");
956 const char* files[] = { "multiplicityMC_100k_1.root", "multiplicityMC_200k.root", "multiplicityMC_400k.root", "multiplicityMC_600k.root", "multiplicityMC_800k.root" };
958 TGraph* fitResultsChi2[3];
959 TGraph* fitResultsBayes[3];
961 for (Int_t region=0; region<AliMultiplicityCorrection::kQualityRegions; ++region)
963 fitResultsChi2[region] = new TGraph;
964 fitResultsChi2[region]->SetTitle(";Nevents;Chi2");
965 fitResultsChi2[region]->SetName(Form("fitResultsChi2_%d", region));
966 fitResultsChi2[region]->SetMarkerStyle(20+region);
968 fitResultsBayes[region] = new TGraph;
969 fitResultsBayes[region]->SetTitle(";Nevents;Chi2");
970 fitResultsBayes[region]->SetName(Form("fitResultsBayes_%d", region));
971 fitResultsBayes[region]->SetMarkerStyle(20+region);
972 fitResultsBayes[region]->SetMarkerColor(2);
975 TGraph* fitResultsChi2Limit = new TGraph; fitResultsChi2Limit->SetTitle(";Nevents;Multiplicity reach");
976 TGraph* fitResultsBayesLimit = new TGraph; fitResultsBayesLimit->SetTitle(";Nevents;Multiplicity reach");
977 TGraph* fitResultsChi2Res = new TGraph; fitResultsChi2Res->SetTitle(";Nevents;Chi2");
978 TGraph* fitResultsBayesRes = new TGraph; fitResultsBayesRes->SetTitle(";Nevents;Chi2");
980 fitResultsChi2Limit->SetName("fitResultsChi2Limit");
981 fitResultsBayesLimit->SetName("fitResultsBayesLimit");
982 fitResultsChi2Res->SetName("fitResultsChi2Res");
983 fitResultsBayesRes->SetName("fitResultsBayesRes");
985 TCanvas* canvas = new TCanvas("StatisticsPlot", "StatisticsPlot", 1200, 600);
986 canvas->Divide(5, 2);
991 TFile* file = TFile::Open("StatisticsPlot.root", "RECREATE");
994 for (Int_t i=0; i<5; ++i)
996 TFile::Open(files[i]);
997 AliMultiplicityCorrection* multESD = new AliMultiplicityCorrection("MultiplicityESD", "MultiplicityESD");
998 multESD->LoadHistograms("Multiplicity");
1000 mult->SetMultiplicityESD(histID, multESD->GetMultiplicityESD(histID));
1001 Int_t nEntries = multESD->GetMultiplicityESD(histID)->GetEntries();
1002 TH1* mc = multESD->GetMultiplicityMC(histID, AliMultiplicityCorrection::kTrVtx)->ProjectionY(Form("mc_%d", i));
1004 mult->SetRegularizationParameters(AliMultiplicityCorrection::kPol1, 10000);
1005 mult->ApplyMinuitFit(histID, kFALSE, AliMultiplicityCorrection::kTrVtx);
1006 mult->DrawComparison(Form("StatisticsPlot_%d_MinuitChi2", i), histID, kFALSE, kTRUE, mc);
1008 Int_t chi2MCLimit = 0;
1009 Float_t chi2Residuals = 0;
1010 mult->GetComparisonResults(0, &chi2MCLimit, &chi2Residuals);
1011 for (Int_t region=0; region<AliMultiplicityCorrection::kQualityRegions; ++region)
1013 fitResultsChi2[region]->SetPoint(fitResultsChi2[region]->GetN(), nEntries, mult->GetQuality(region));
1014 min = TMath::Min(min, mult->GetQuality(region));
1015 max = TMath::Max(max, mult->GetQuality(region));
1017 fitResultsChi2Limit->SetPoint(fitResultsChi2Limit->GetN(), nEntries, chi2MCLimit);
1018 fitResultsChi2Res->SetPoint(fitResultsChi2Res->GetN(), nEntries, chi2Residuals);
1020 TH1* chi2Result = (TH1*) mult->GetMultiplicityESDCorrected(histID)->Clone(Form("chi2Result_%d", i));
1022 mult->ApplyBayesianMethod(histID, kFALSE, AliMultiplicityCorrection::kTrVtx, 1, 100, 0, kFALSE);
1023 mult->DrawComparison(Form("StatisticsPlot_%d_Bayesian", i), histID, kFALSE, kTRUE, mc);
1024 TH1* bayesResult = (TH1*) mult->GetMultiplicityESDCorrected(histID)->Clone(Form("bayesResult_%d", i));
1025 mult->GetComparisonResults(0, &chi2MCLimit, &chi2Residuals);
1026 for (Int_t region=0; region<AliMultiplicityCorrection::kQualityRegions; ++region)
1028 fitResultsBayes[region]->SetPoint(fitResultsBayes[region]->GetN(), nEntries, mult->GetQuality(region));
1029 min = TMath::Min(min, mult->GetQuality(region));
1030 max = TMath::Max(max, mult->GetQuality(region));
1032 fitResultsBayesLimit->SetPoint(fitResultsBayesLimit->GetN(), nEntries, chi2MCLimit);
1033 fitResultsBayesRes->SetPoint(fitResultsBayesRes->GetN(), nEntries, chi2Residuals);
1035 mc->GetXaxis()->SetRangeUser(0, 150);
1036 chi2Result->GetXaxis()->SetRangeUser(0, 150);
1038 // skip errors for now
1039 for (Int_t j=0; j<=chi2Result->GetNbinsX(); ++j)
1041 chi2Result->SetBinError(j, 0);
1042 bayesResult->SetBinError(j, 0);
1046 mc->SetFillColor(kYellow);
1048 chi2Result->SetLineColor(kRed);
1049 chi2Result->DrawCopy("SAME");
1050 bayesResult->SetLineColor(kBlue);
1051 bayesResult->DrawCopy("SAME");
1055 chi2Result->Divide(chi2Result, mc, 1, 1, "B");
1056 bayesResult->Divide(bayesResult, mc, 1, 1, "B");
1058 // skip errors for now
1059 for (Int_t j=0; j<=chi2Result->GetNbinsX(); ++j)
1061 chi2Result->SetBinError(j, 0);
1062 bayesResult->SetBinError(j, 0);
1065 chi2Result->SetTitle("Ratios;Npart;unfolded measured/MC");
1066 chi2Result->GetYaxis()->SetRangeUser(0.5, 1.5);
1068 chi2Result->DrawCopy("");
1069 bayesResult->DrawCopy("SAME");
1071 TFile* file = TFile::Open("StatisticsPlot.root", "UPDATE");
1073 chi2Result->Write();
1074 bayesResult->Write();
1078 canvas->SaveAs(Form("%s.gif", canvas->GetName()));
1079 canvas->SaveAs(Form("%s.C", canvas->GetName()));
1081 TCanvas* canvas2 = new TCanvas("StatisticsPlot2", "StatisticsPlot2", 800, 400);
1082 canvas2->Divide(2, 1);
1086 for (Int_t region=0; region<AliMultiplicityCorrection::kQualityRegions; ++region)
1088 fitResultsChi2[region]->GetYaxis()->SetRangeUser(0.5 * min, 1.5 * max);
1089 fitResultsChi2[region]->Draw(((region == 0) ? "AP" : "P SAME"));
1091 fitResultsBayes[region]->Draw("P SAME");
1097 fitResultsChi2Limit->SetMarkerStyle(20);
1098 fitResultsChi2Limit->GetYaxis()->SetRangeUser(0.9 * TMath::Min(fitResultsChi2Limit->GetYaxis()->GetXmin(), fitResultsBayesLimit->GetYaxis()->GetXmin()), 1.1 * TMath::Max(fitResultsChi2Limit->GetYaxis()->GetXmax(), fitResultsBayesLimit->GetYaxis()->GetXmax()));
1099 fitResultsChi2Limit->Draw("AP");
1101 fitResultsBayesLimit->SetMarkerStyle(3);
1102 fitResultsBayesLimit->SetMarkerColor(2);
1103 fitResultsBayesLimit->Draw("P SAME");
1105 canvas2->SaveAs(Form("%s.gif", canvas2->GetName()));
1106 canvas2->SaveAs(Form("%s.C", canvas2->GetName()));
1108 TFile* file = TFile::Open("StatisticsPlot.root", "UPDATE");
1110 for (Int_t region=0; region<AliMultiplicityCorrection::kQualityRegions; ++region)
1112 fitResultsChi2[region]->Write();
1113 fitResultsBayes[region]->Write();
1115 fitResultsChi2Limit->Write();
1116 fitResultsBayesLimit->Write();
1117 fitResultsChi2Res->Write();
1118 fitResultsBayesRes->Write();
1122 void StartingConditions(const char* fileNameMC = "multiplicityMC_2M.root", Int_t histID = 3)
1124 gSystem->Load("libPWG0base");
1126 AliMultiplicityCorrection* mult = new AliMultiplicityCorrection("Multiplicity", "Multiplicity");
1128 TFile::Open(fileNameMC);
1129 mult->LoadHistograms("Multiplicity");
1131 const char* files[] = { "multiplicityMC_1M_3.root", "multiplicityMC_100k_1.root", "multiplicityMC_100k_2.root", "multiplicityMC_100k_3.root", "multiplicityMC_100k_4.root" }
1133 // this one we try to unfold
1134 TFile::Open(files[0]);
1135 AliMultiplicityCorrection* multESD = new AliMultiplicityCorrection("MultiplicityESD", "MultiplicityESD");
1136 multESD->LoadHistograms("Multiplicity");
1137 mult->SetMultiplicityESD(histID, multESD->GetMultiplicityESD(histID));
1138 TH1* mc = multESD->GetMultiplicityMC(histID, AliMultiplicityCorrection::kTrVtx)->ProjectionY("mc");
1140 TGraph* fitResultsChi2 = new TGraph;
1141 fitResultsChi2->SetTitle(";Input Dist ID;Chi2");
1142 TGraph* fitResultsBayes = new TGraph;
1143 fitResultsBayes->SetTitle(";Input Dist ID;Chi2");
1144 TGraph* fitResultsChi2Limit = new TGraph;
1145 fitResultsChi2Limit->SetTitle(";Input Dist ID;Multiplicity reach");
1146 TGraph* fitResultsBayesLimit = new TGraph;
1147 fitResultsBayesLimit->SetTitle(";Input Dist ID;Multiplicity reach");
1149 TCanvas* canvas = new TCanvas("StartingConditions", "StartingConditions", 1200, 600);
1150 canvas->Divide(8, 2);
1152 TCanvas* canvas3 = new TCanvas("StartingConditions3", "StartingConditions3", 1000, 400);
1153 canvas3->Divide(2, 1);
1159 TH1* firstBayesian = 0;
1160 TH1* startCond = multESD->GetMultiplicityESD(histID)->ProjectionY("startCond");
1162 TLegend* legend = new TLegend(0.7, 0.7, 1, 1);
1164 TFile* file = TFile::Open("StartingConditions.root", "RECREATE");
1168 for (Int_t i=0; i<8; ++i)
1172 startCond = (TH1*) mc->Clone("startCond2");
1176 TFile::Open(files[i-1]);
1177 AliMultiplicityCorrection* multESD2 = new AliMultiplicityCorrection("MultiplicityESD2", "MultiplicityESD2");
1178 multESD2->LoadHistograms("Multiplicity");
1179 startCond = multESD2->GetMultiplicityESD(histID)->ProjectionY("startCond");
1183 func = new TF1("nbd", "[0] * TMath::Binomial([2]+TMath::Nint(x)-1, [2]-1) * pow([1] / ([1]+[2]), TMath::Nint(x)) * pow(1 + [1]/[2], -[2])", 0, 50);
1184 func->SetParNames("scaling", "averagen", "k");
1185 func->SetParLimits(0, 1e-3, 1e10);
1186 func->SetParLimits(1, 0.001, 1000);
1187 func->SetParLimits(2, 0.001, 1000);
1189 func->SetParameters(1, 10, 2);
1190 for (Int_t j=2; j<=startCond->GetNbinsX(); j++)
1191 startCond->SetBinContent(j, func->Eval(j-1));
1195 for (Int_t j=1; j<=startCond->GetNbinsX(); j++)
1196 startCond->SetBinContent(j, 1);
1199 mult->SetRegularizationParameters(AliMultiplicityCorrection::kPol1, 10000);
1200 mult->ApplyMinuitFit(histID, kFALSE, AliMultiplicityCorrection::kTrVtx, kFALSE, startCond);
1201 mult->DrawComparison(Form("StartingConditions_%d_MinuitChi2", i), histID, kFALSE, kTRUE, mc);
1204 Int_t chi2MCLimit = 0;
1205 mult->GetComparisonResults(&chi2MC, &chi2MCLimit, 0);
1206 fitResultsChi2->SetPoint(fitResultsChi2->GetN(), i, chi2MC);
1207 fitResultsChi2Limit->SetPoint(fitResultsChi2Limit->GetN(), i, chi2MCLimit);
1208 min = TMath::Min(min, chi2MC);
1209 max = TMath::Max(max, chi2MC);
1211 TH1* chi2Result = (TH1*) mult->GetMultiplicityESDCorrected(histID)->Clone(Form("chi2Result_%d", i));
1213 firstChi = (TH1*) chi2Result->Clone("firstChi");
1215 mult->ApplyBayesianMethod(histID, kFALSE, AliMultiplicityCorrection::kTrVtx, 1, 100, startCond);
1216 mult->DrawComparison(Form("StartingConditions_%d_Bayesian", i), histID, kFALSE, kTRUE, mc);
1217 TH1* bayesResult = (TH1*) mult->GetMultiplicityESDCorrected(histID)->Clone(Form("bayesResult_%d", i));
1219 firstBayesian = (TH1*) bayesResult->Clone("firstBayesian");
1221 mult->GetComparisonResults(&chi2MC, &chi2MCLimit, 0);
1222 fitResultsBayes->SetPoint(fitResultsBayes->GetN(), i, chi2MC);
1223 fitResultsBayesLimit->SetPoint(fitResultsBayesLimit->GetN(), i, chi2MCLimit);
1225 TFile* file = TFile::Open("StartingConditions.root", "UPDATE");
1226 chi2Result->Write();
1227 bayesResult->Write();
1230 min = TMath::Min(min, chi2MC);
1231 max = TMath::Max(max, chi2MC);
1232 mc->GetXaxis()->SetRangeUser(0, 150);
1233 chi2Result->GetXaxis()->SetRangeUser(0, 150);
1235 // skip errors for now
1236 for (Int_t j=0; j<=chi2Result->GetNbinsX(); ++j)
1238 chi2Result->SetBinError(j, 0);
1239 bayesResult->SetBinError(j, 0);
1243 TH1* tmp = (TH1*) chi2Result->Clone("tmp");
1244 tmp->SetTitle("Difference to best initial conditions;Npart;Ratio");
1245 tmp->Divide(firstChi);
1246 tmp->GetYaxis()->SetRangeUser(0.5, 1.5);
1247 tmp->GetXaxis()->SetRangeUser(0, 200);
1248 tmp->SetLineColor(i+1);
1249 legend->AddEntry(tmp, Form("%d", i));
1250 tmp->DrawCopy((i > 0) ? "SAME HIST" : "HIST");
1253 tmp = (TH1*) bayesResult->Clone("tmp");
1254 tmp->SetTitle("Difference to best initial conditions;Npart;Ratio");
1255 tmp->Divide(firstBayesian);
1256 tmp->SetLineColor(i+1);
1257 tmp->GetYaxis()->SetRangeUser(0.5, 1.5);
1258 tmp->GetXaxis()->SetRangeUser(0, 200);
1259 tmp->DrawCopy((i > 0) ? "SAME HIST" : "HIST");
1262 mc->SetFillColor(kYellow);
1264 chi2Result->SetLineColor(kRed);
1265 chi2Result->DrawCopy("SAME");
1266 bayesResult->SetLineColor(kBlue);
1267 bayesResult->DrawCopy("SAME");
1271 chi2Result->Divide(chi2Result, mc, 1, 1, "B");
1272 bayesResult->Divide(bayesResult, mc, 1, 1, "B");
1274 // skip errors for now
1275 for (Int_t j=0; j<=chi2Result->GetNbinsX(); ++j)
1277 chi2Result->SetBinError(j, 0);
1278 bayesResult->SetBinError(j, 0);
1281 chi2Result->SetTitle("Ratios;Npart;unfolded measured/MC");
1282 chi2Result->GetYaxis()->SetRangeUser(0.5, 1.5);
1284 chi2Result->DrawCopy("");
1285 bayesResult->DrawCopy("SAME");
1291 canvas->SaveAs(Form("%s.gif", canvas->GetName()));
1293 TCanvas* canvas2 = new TCanvas("StartingConditions2", "StartingConditions2", 800, 400);
1294 canvas2->Divide(2, 1);
1297 fitResultsChi2->SetMarkerStyle(20);
1298 fitResultsChi2->GetYaxis()->SetRangeUser(0.5 * min, 1.5 * max);
1299 fitResultsChi2->Draw("AP");
1301 fitResultsBayes->SetMarkerStyle(3);
1302 fitResultsBayes->SetMarkerColor(2);
1303 fitResultsBayes->Draw("P SAME");
1308 fitResultsChi2Limit->SetMarkerStyle(20);
1309 fitResultsChi2Limit->GetYaxis()->SetRangeUser(0.9 * TMath::Min(fitResultsChi2Limit->GetYaxis()->GetXmin(), fitResultsBayesLimit->GetYaxis()->GetXmin()), 1.1 * TMath::Max(fitResultsChi2Limit->GetYaxis()->GetXmax(), fitResultsBayesLimit->GetYaxis()->GetXmax()));
1310 fitResultsChi2Limit->Draw("AP");
1312 fitResultsBayesLimit->SetMarkerStyle(3);
1313 fitResultsBayesLimit->SetMarkerColor(2);
1314 fitResultsBayesLimit->Draw("P SAME");
1316 canvas2->SaveAs(Form("%s.gif", canvas2->GetName()));
1317 canvas3->SaveAs(Form("%s.gif", canvas3->GetName()));
1320 void DifferentSamples(const char* fileNameMC = "multiplicityMC_2M.root", Int_t histID = 3)
1322 gSystem->Load("libPWG0base");
1324 AliMultiplicityCorrection* mult = new AliMultiplicityCorrection("Multiplicity", "Multiplicity");
1326 TFile::Open(fileNameMC);
1327 mult->LoadHistograms("Multiplicity");
1329 const char* files[] = { "multiplicityMC_100k_1.root", "multiplicityMC_100k_2.root", "multiplicityMC_100k_3.root", "multiplicityMC_100k_4.root", "multiplicityMC_100k_5.root", "multiplicityMC_100k_6.root", "multiplicityMC_100k_7.root", "multiplicityMC_100k_8.root" };
1331 TGraph* fitResultsChi2 = new TGraph;
1332 fitResultsChi2->SetTitle(";Input Dist ID;Chi2");
1333 TGraph* fitResultsBayes = new TGraph;
1334 fitResultsBayes->SetTitle(";Input Dist ID;Chi2");
1335 TGraph* fitResultsChi2Limit = new TGraph;
1336 fitResultsChi2Limit->SetTitle(";Input Dist ID;Multiplicity reach");
1337 TGraph* fitResultsBayesLimit = new TGraph;
1338 fitResultsBayesLimit->SetTitle(";Input Dist ID;Multiplicity reach");
1340 TCanvas* canvasA = new TCanvas("DifferentSamplesA", "DifferentSamplesA", 1200, 600);
1341 canvasA->Divide(4, 2);
1343 TCanvas* canvasB = new TCanvas("DifferentSamplesB", "DifferentSamplesB", 1200, 600);
1344 canvasB->Divide(4, 2);
1346 TCanvas* canvas4 = new TCanvas("DifferentSamples4", "DifferentSamples4", 1000, 400);
1347 canvas4->Divide(2, 1);
1349 TCanvas* canvas3 = new TCanvas("DifferentSamples3", "DifferentSamples3", 1000, 400);
1350 canvas3->Divide(2, 1);
1356 TH1* firstBayesian = 0;
1358 TLegend* legend = new TLegend(0.7, 0.7, 1, 1);
1360 TFile* file = TFile::Open("DifferentSamples.root", "RECREATE");
1363 for (Int_t i=0; i<8; ++i)
1365 TFile::Open(files[i]);
1366 AliMultiplicityCorrection* multESD = new AliMultiplicityCorrection("MultiplicityESD2", "MultiplicityESD2");
1367 multESD->LoadHistograms("Multiplicity");
1368 mult->SetMultiplicityESD(histID, multESD->GetMultiplicityESD(histID));
1369 TH1* mc = multESD->GetMultiplicityMC(histID, AliMultiplicityCorrection::kTrVtx)->ProjectionY(Form("mc_%d", i));
1372 mult->SetRegularizationParameters(AliMultiplicityCorrection::kPol1, 10000);
1373 mult->ApplyMinuitFit(histID, kFALSE, AliMultiplicityCorrection::kTrVtx, kFALSE);
1374 mult->DrawComparison(Form("DifferentSamples_%d_MinuitChi2", i), histID, kFALSE, kTRUE, mc);
1377 Int_t chi2MCLimit = 0;
1378 mult->GetComparisonResults(&chi2MC, &chi2MCLimit, 0);
1379 fitResultsChi2->SetPoint(fitResultsChi2->GetN(), i, chi2MC);
1380 fitResultsChi2Limit->SetPoint(fitResultsChi2Limit->GetN(), i, chi2MCLimit);
1381 min = TMath::Min(min, chi2MC);
1382 max = TMath::Max(max, chi2MC);
1384 TH1* chi2Result = (TH1*) mult->GetMultiplicityESDCorrected(histID)->Clone(Form("chi2Result_%d", i));
1386 firstChi = (TH1*) chi2Result->Clone("firstChi");
1388 mult->ApplyBayesianMethod(histID, kFALSE, AliMultiplicityCorrection::kTrVtx, 1, 100);
1389 mult->DrawComparison(Form("DifferentSamples_%d_Bayesian", i), histID, kFALSE, kTRUE, mc);
1390 TH1* bayesResult = (TH1*) mult->GetMultiplicityESDCorrected(histID)->Clone(Form("bayesResult_%d", i));
1392 firstBayesian = (TH1*) bayesResult->Clone("firstBayesian");
1394 TFile* file = TFile::Open("DifferentSamples.root", "UPDATE");
1396 chi2Result->Write();
1397 bayesResult->Write();
1400 mult->GetComparisonResults(&chi2MC, &chi2MCLimit, 0);
1401 fitResultsBayes->SetPoint(fitResultsBayes->GetN(), i, chi2MC);
1402 fitResultsBayesLimit->SetPoint(fitResultsBayesLimit->GetN(), i, chi2MCLimit);
1404 min = TMath::Min(min, chi2MC);
1405 max = TMath::Max(max, chi2MC);
1406 mc->GetXaxis()->SetRangeUser(0, 150);
1407 chi2Result->GetXaxis()->SetRangeUser(0, 150);
1409 // skip errors for now
1410 for (Int_t j=0; j<=chi2Result->GetNbinsX(); ++j)
1412 chi2Result->SetBinError(j, 0);
1413 bayesResult->SetBinError(j, 0);
1417 TH1* tmp = (TH1*) chi2Result->Clone("tmp");
1418 tmp->SetTitle("Unfolded/MC;Npart;Ratio");
1420 tmp->GetYaxis()->SetRangeUser(0.5, 1.5);
1421 tmp->GetXaxis()->SetRangeUser(0, 200);
1422 tmp->SetLineColor(i+1);
1423 tmp->DrawCopy((i > 0) ? "SAME HIST" : "HIST");
1426 tmp = (TH1*) bayesResult->Clone("tmp");
1427 tmp->SetTitle("Unfolded/MC;Npart;Ratio");
1429 tmp->SetLineColor(i+1);
1430 tmp->GetYaxis()->SetRangeUser(0.5, 1.5);
1431 tmp->GetXaxis()->SetRangeUser(0, 200);
1432 tmp->DrawCopy((i > 0) ? "SAME HIST" : "HIST");
1435 TH1* tmp = (TH1*) chi2Result->Clone("tmp");
1436 tmp->SetTitle("Ratio to first result;Npart;Ratio");
1437 tmp->Divide(firstChi);
1438 tmp->GetYaxis()->SetRangeUser(0.5, 1.5);
1439 tmp->GetXaxis()->SetRangeUser(0, 200);
1440 tmp->SetLineColor(i+1);
1441 legend->AddEntry(tmp, Form("%d", i));
1442 tmp->DrawCopy((i > 0) ? "SAME HIST" : "HIST");
1445 tmp = (TH1*) bayesResult->Clone("tmp");
1446 tmp->SetTitle("Ratio to first result;Npart;Ratio");
1447 tmp->Divide(firstBayesian);
1448 tmp->SetLineColor(i+1);
1449 tmp->GetYaxis()->SetRangeUser(0.5, 1.5);
1450 tmp->GetXaxis()->SetRangeUser(0, 200);
1451 tmp->DrawCopy((i > 0) ? "SAME HIST" : "HIST");
1460 mc->SetFillColor(kYellow);
1462 chi2Result->SetLineColor(kRed);
1463 chi2Result->DrawCopy("SAME");
1464 bayesResult->SetLineColor(kBlue);
1465 bayesResult->DrawCopy("SAME");
1475 chi2Result->Divide(chi2Result, mc, 1, 1, "B");
1476 bayesResult->Divide(bayesResult, mc, 1, 1, "B");
1478 // skip errors for now
1479 for (Int_t j=0; j<=chi2Result->GetNbinsX(); ++j)
1481 chi2Result->SetBinError(j, 0);
1482 bayesResult->SetBinError(j, 0);
1485 chi2Result->SetTitle("Ratios;Npart;unfolded measured/MC");
1486 chi2Result->GetYaxis()->SetRangeUser(0.5, 1.5);
1488 chi2Result->DrawCopy("");
1489 bayesResult->DrawCopy("SAME");
1495 canvasA->SaveAs(Form("%s.gif", canvasA->GetName()));
1496 canvasB->SaveAs(Form("%s.gif", canvasB->GetName()));
1498 TCanvas* canvas2 = new TCanvas("DifferentSamples2", "DifferentSamples2", 800, 400);
1499 canvas2->Divide(2, 1);
1502 fitResultsChi2->SetMarkerStyle(20);
1503 fitResultsChi2->GetYaxis()->SetRangeUser(0.5 * min, 1.5 * max);
1504 fitResultsChi2->Draw("AP");
1506 fitResultsBayes->SetMarkerStyle(3);
1507 fitResultsBayes->SetMarkerColor(2);
1508 fitResultsBayes->Draw("P SAME");
1513 fitResultsChi2Limit->SetMarkerStyle(20);
1514 fitResultsChi2Limit->GetYaxis()->SetRangeUser(0.9 * TMath::Min(fitResultsChi2Limit->GetYaxis()->GetXmin(), fitResultsBayesLimit->GetYaxis()->GetXmin()), 1.1 * TMath::Max(fitResultsChi2Limit->GetYaxis()->GetXmax(), fitResultsBayesLimit->GetYaxis()->GetXmax()));
1515 fitResultsChi2Limit->Draw("AP");
1517 fitResultsBayesLimit->SetMarkerStyle(3);
1518 fitResultsBayesLimit->SetMarkerColor(2);
1519 fitResultsBayesLimit->Draw("P SAME");
1521 canvas2->SaveAs(Form("%s.gif", canvas2->GetName()));
1522 canvas3->SaveAs(Form("%s.gif", canvas3->GetName()));
1523 canvas4->SaveAs(Form("%s.gif", canvas4->GetName()));
1526 void Merge(Int_t n, const char** files, const char* output)
1528 // const char* files[] = { "multiplicityMC_100k_1.root", "multiplicityMC_100k_2.root", "multiplicityMC_100k_3.root", "multiplicityMC_100k_4.root", "multiplicityMC_100k_5.root", "multiplicityMC_100k_6.root", "multiplicityMC_100k_7.root", "multiplicityMC_100k_8.root" };
1531 gSystem->Load("libPWG0base");
1533 AliMultiplicityCorrection** data = new AliMultiplicityCorrection*[n];
1535 for (Int_t i=0; i<n; ++i)
1537 TString name("Multiplicity");
1539 name.Form("Multiplicity%d", i);
1541 TFile::Open(files[i]);
1542 data[i] = new AliMultiplicityCorrection(name, name);
1543 data[i]->LoadHistograms("Multiplicity");
1548 data[0]->Merge(&list);
1550 //data[0]->DrawHistograms();
1552 TFile::Open(output, "RECREATE");
1553 data[0]->SaveHistograms();
1557 void testMethod(Int_t caseNo, const char* fileName = "multiplicityMC.root")
1559 gSystem->Load("libPWG0base");
1561 AliMultiplicityCorrection* mult = new AliMultiplicityCorrection("Multiplicity", "Multiplicity");
1563 TFile::Open(fileName);
1564 mult->LoadHistograms("Multiplicity");
1570 func = new TF1("nbd", "[0] * TMath::Binomial([2]+TMath::Nint(x)-1, [2]-1) * pow([1] / ([1]+[2]), TMath::Nint(x)) * pow(1 + [1]/[2], -[2])", 0, 500);
1571 func->SetParNames("scaling", "averagen", "k");
1576 case 0: func = new TF1("flat", "1000"); break;
1577 case 1: func = new TF1("flat", "501-x"); break;
1578 case 2: func = new TF1("flat", "1000 * 1/(x+1)"); break;
1579 case 3: func = new TF1("flat", "1000 * TMath::Landau(x, 10, 5)"); break;
1580 case 4: func->SetParameters(1e7, 10, 2); break;
1581 case 5: func->SetParameters(1, 13, 7); break;
1582 case 6: func->SetParameters(1e7, 30, 4); break;
1583 case 7: func->SetParameters(1e7, 30, 2); break; // ***
1584 case 8: func = new TF1("testlaszlo", "10*1000*x*exp(-0.1*x)"); break;
1592 mult->SetGenMeasFromFunc(func, 3);
1594 TFile::Open("out.root", "RECREATE");
1595 mult->SaveHistograms();
1597 new TCanvas; mult->GetMultiplicityESD(3)->ProjectionY()->DrawCopy();
1598 new TCanvas; mult->GetMultiplicityVtx(3)->ProjectionY()->DrawCopy();
1600 //mult->ApplyBayesianMethod(2, kFALSE);
1601 //mult->ApplyMinuitFit(2, kFALSE);
1602 //mult->ApplyGaussianMethod(2, kFALSE);
1603 //mult->ApplyLaszloMethod(2, kFALSE, AliMultiplicityCorrection::kTrVtx);
1606 void smoothCorrelationMap(const char* fileName = "multiplicityMC.root", Int_t corrMatrix = 2)
1608 gSystem->Load("libPWG0base");
1610 AliMultiplicityCorrection* mult = new AliMultiplicityCorrection("Multiplicity", "Multiplicity");
1612 TFile::Open(fileName);
1613 mult->LoadHistograms("Multiplicity");
1615 // empty under/overflow bins in x, otherwise Project3D takes them into account
1616 TH3* corr = mult->GetCorrelation(corrMatrix);
1617 for (Int_t j=1; j<=corr->GetYaxis()->GetNbins(); ++j)
1619 for (Int_t k=1; k<=corr->GetZaxis()->GetNbins(); ++k)
1621 corr->SetBinContent(0, j, k, 0);
1622 corr->SetBinContent(corr->GetXaxis()->GetNbins()+1, j, k, 0);
1626 TH2* proj = (TH2*) corr->Project3D("zy");
1628 // normalize correction for given nPart
1629 for (Int_t i=1; i<=proj->GetNbinsX(); ++i)
1631 Double_t sum = proj->Integral(i, i, 1, proj->GetNbinsY());
1635 for (Int_t j=1; j<=proj->GetNbinsY(); ++j)
1638 proj->SetBinContent(i, j, proj->GetBinContent(i, j) / sum);
1639 proj->SetBinError(i, j, proj->GetBinError(i, j) / sum);
1644 proj->DrawCopy("COLZ");
1646 TH1* scaling = proj->ProjectionY("scaling", 1, 1);
1648 scaling->SetMarkerStyle(3);
1649 //scaling->GetXaxis()->SetRangeUser(0, 50);
1650 TH1* mean = (TH1F*) scaling->Clone("mean");
1651 TH1* width = (TH1F*) scaling->Clone("width");
1653 TF1* lognormal = new TF1("lognormal", "[0]*exp(-(log(x)-[1])^2/(2*[2]^2))/(x*[2]*TMath::Sqrt(2*TMath::Pi()))", 0.01, 500);
1654 lognormal->SetParNames("scaling", "mean", "sigma");
1655 lognormal->SetParameters(1, 1, 1);
1656 lognormal->SetParLimits(0, 1, 1);
1657 lognormal->SetParLimits(1, 0, 100);
1658 lognormal->SetParLimits(2, 1e-3, 1);
1660 TF1* nbd = new TF1("nbd", "[0] * TMath::Binomial([2]+TMath::Nint(x)-1, [2]-1) * pow([1] / ([1]+[2]), TMath::Nint(x)) * pow(1 + [1]/[2], -[2])", 0, 50);
1661 nbd->SetParNames("scaling", "averagen", "k");
1662 nbd->SetParameters(1, 13, 5);
1663 nbd->SetParLimits(0, 1, 1);
1664 nbd->SetParLimits(1, 1, 100);
1665 nbd->SetParLimits(2, 1, 1e8);
1667 TF1* poisson = new TF1("poisson", "[0] * exp(-(x+[2])) * (x+[2])**[1] / TMath::Factorial([1])", 0.01, 50);
1668 poisson->SetParNames("scaling", "k", "deltax");
1669 poisson->SetParameters(1, 1, 0);
1670 poisson->SetParLimits(0, 0, 10);
1671 poisson->SetParLimits(1, 0.01, 100);
1672 poisson->SetParLimits(2, 0, 10);
1674 TF1* mygaus = new TF1("mygaus", "[0] * exp(-(x-[1])**2 / 2 / [2] - [3] * log(x + [4]) / [5])", 0.01, 50);
1675 mygaus->SetParNames("scaling", "mean", "width", "scale2log", "logmean", "logwidth");
1676 mygaus->SetParameters(1, 0, 1, 1, 0, 1);
1677 mygaus->SetParLimits(2, 1e-5, 10);
1678 mygaus->SetParLimits(4, 1, 1);
1679 mygaus->SetParLimits(5, 1e-5, 10);
1681 //TF1* sqrt = new TF1("sqrt", "[0] + [1] * sqrt((x + [3]) * [2])", 0, 50);
1682 TF1* sqrt = new TF1("sqrt", "[0] + (x + [1])**[2]", 0, 50);
1683 sqrt->SetParNames("ydelta", "exp", "xdelta");
1684 sqrt->SetParameters(0, 0, 1);
1685 sqrt->SetParLimits(1, 0, 10);
1687 const char* fitWith = "gaus";
1689 for (Int_t i=1; i<=150; ++i)
1691 printf("Fitting %d...\n", i);
1693 TH1* hist = proj->ProjectionY(Form("proj%d", i), i, i, "e");
1695 //hist->GetXaxis()->SetRangeUser(0, 50);
1696 //lognormal->SetParameter(0, hist->GetMaximum());
1697 hist->Fit(fitWith, "0 M", "");
1699 TF1* func = hist->GetListOfFunctions()->FindObject(fitWith);
1701 if (0 && (i % 5 == 0))
1705 func->Clone()->Draw("SAME");
1709 scaling->Fill(i, func->GetParameter(0));
1710 mean->Fill(i, func->GetParameter(1));
1711 width->Fill(i, func->GetParameter(2));
1714 TF1* log = new TF1("log", "[0] + [1] * log([2] * x)", 0.01, 500);
1715 log->SetParameters(0, 1, 1);
1716 log->SetParLimits(1, 0, 100);
1717 log->SetParLimits(2, 1e-3, 10);
1719 TF1* over = new TF1("over", "[0] + [1] / (x+[2])", 0.01, 500);
1720 over->SetParameters(0, 1, 0);
1721 //over->SetParLimits(0, 0, 100);
1722 over->SetParLimits(1, 1e-3, 10);
1723 over->SetParLimits(2, 0, 100);
1725 c1 = new TCanvas("fitparams", "fitparams", 1200, 400);
1731 //TF1* scalingFit = new TF1("mypol0", "[0]");
1732 TF1* scalingFit = over;
1733 scaling->Fit(scalingFit, "", "", 3, 140);
1734 scalingFit->SetRange(0, 200);
1735 scalingFit->Draw("SAME");
1740 //TF1* meanFit = log;
1741 TF1* meanFit = new TF1("mypol1", "[0]+[1]*x");
1742 mean->Fit(meanFit, "", "", 3, 140);
1743 meanFit->SetRange(0, 200);
1744 meanFit->Draw("SAME");
1749 //TF1* widthFit = over;
1750 TF1* widthFit = new TF1("mypol", "[0]+[1]*TMath::Sqrt([2]*x)");
1751 widthFit->SetParLimits(2, 1e-5, 1e5);
1752 width->Fit(widthFit, "", "", 5, 140);
1753 widthFit->SetRange(0, 200);
1754 widthFit->Draw("SAME");
1756 // build new correction matrix
1757 TH2* new = (TH2*) proj->Clone("new");
1760 for (Int_t i=1; i<=new->GetXaxis()->GetNbins(); i+=1)
1762 TF1* func = (TF1*) gROOT->FindObject(fitWith);
1763 x = new->GetXaxis()->GetBinCenter(i);
1767 func->SetParameters(scalingFit->Eval(x), meanFit->Eval(x), widthFit->Eval(x));
1768 printf("%f %f %f %f\n", x, scalingFit->Eval(x), meanFit->Eval(x), widthFit->Eval(x));
1770 for (Int_t j=1; j<=new->GetYaxis()->GetNbins(); j+=1)
1774 // leave bins 1..20 untouched
1775 new->SetBinContent(i, j, corr->Integral(1, corr->GetNbinsX(), i, i, j, j));
1779 y = new->GetYaxis()->GetBinCenter(j);
1782 if (func->Eval(y) > 1e-4)
1783 new->SetBinContent(i, j, func->Eval(y));
1788 // fill 0 multiplicity bins, this cannot be done with the function because it does not accept 0
1789 // we take the values from the old response matrix
1790 //for (Int_t i=1; i<=new->GetXaxis()->GetNbins(); i+=1)
1791 // new->SetBinContent(i, 1, proj->GetBinContent(i, 1));
1793 //for (Int_t j=1; j<=new->GetYaxis()->GetNbins(); j+=1)
1794 // new->SetBinContent(1, j, proj->GetBinContent(1, j));
1796 // normalize correction for given nPart
1797 for (Int_t i=1; i<=new->GetNbinsX(); ++i)
1799 Double_t sum = new->Integral(i, i, 1, proj->GetNbinsY());
1803 for (Int_t j=1; j<=new->GetNbinsY(); ++j)
1806 new->SetBinContent(i, j, new->GetBinContent(i, j) / sum);
1807 new->SetBinError(i, j, new->GetBinError(i, j) / sum);
1814 TH2* diff = (TH2*) new->Clone("diff");
1815 diff->Add(proj, -1);
1819 diff->SetMinimum(-0.05);
1820 diff->SetMaximum(0.05);
1824 for (Int_t i=1; i<=new->GetNbinsX(); ++i)
1825 for (Int_t j=1; j<=new->GetNbinsY(); ++j)
1826 corr->SetBinContent(corr->GetXaxis()->GetNbins() / 2, i, j, new->GetBinContent(i, j));
1829 corr->Project3D("zy")->Draw("COLZ");
1831 TFile::Open("out.root", "RECREATE");
1832 mult->SaveHistograms();
1834 TH1* proj1 = proj->ProjectionY("proj1", 36, 36);
1835 TH1* proj2 = new->ProjectionY("proj2", 36, 36);
1836 proj2->SetLineColor(2);
1840 proj2->Draw("SAME");
1843 void buildCorrelationMap(const char* fileName = "multiplicityMC_2M.root", Int_t corrMatrix = 3)
1845 gSystem->Load("libPWG0base");
1847 AliMultiplicityCorrection* mult = new AliMultiplicityCorrection("Multiplicity", "Multiplicity");
1849 TFile::Open(fileName);
1850 mult->LoadHistograms("Multiplicity");
1852 TH3F* new = mult->GetCorrelation(corrMatrix);
1855 TF1* func = new TF1("func", "gaus(0)");
1857 Int_t vtxBin = new->GetNbinsX() / 2;
1862 for (Int_t i=1; i<=new->GetYaxis()->GetNbins(); i+=1)
1864 Float_t x = new->GetYaxis()->GetBinCenter(i);
1865 func->SetParameters(1, x * 0.8, sigma);
1866 //func->SetParameters(1, x, sigma);
1868 for (Int_t j=1; j<=new->GetZaxis()->GetNbins(); j+=1)
1870 Float_t y = new->GetYaxis()->GetBinCenter(j);
1873 if (TMath::Abs(y-x*0.8) < sigma)
1874 new->SetBinContent(vtxBin, i, j, func->Eval(y));
1876 // test only bin 40 has smearing
1878 // new->SetBinContent(vtxBin, i, j, (i == j));
1883 new->Project3D("zy")->DrawCopy("COLZ");
1885 TFile* file = TFile::Open("out.root", "RECREATE");
1886 mult->SetCorrelation(corrMatrix, new);
1887 mult->SaveHistograms();
1891 void GetCrossSections(const char* fileName)
1893 gSystem->Load("libPWG0base");
1895 AliMultiplicityCorrection* mult = new AliMultiplicityCorrection("Multiplicity", "Multiplicity");
1897 TFile::Open(fileName);
1898 mult->LoadHistograms("Multiplicity");
1900 TH1* xSection2 = mult->GetCorrelation(3)->Project3D("y")->Clone("xSection2");
1902 xSection2->Scale(1.0 / xSection2->Integral());
1904 TH1* xSection15 = mult->GetCorrelation(2)->Project3D("y")->Clone("xSection15");
1905 xSection15->Sumw2();
1906 xSection15->Scale(1.0 / xSection15->Integral());
1908 TFile::Open("crosssection.root", "RECREATE");
1910 xSection15->Write();
1914 void AnalyzeSpeciesTree(const char* fileName)
1917 // prints statistics about fParticleSpecies
1920 gSystem->Load("libPWG0base");
1922 TFile::Open(fileName);
1923 TNtuple* fParticleSpecies = (TNtuple*) gFile->Get("fParticleSpecies");
1925 const Int_t nFields = 8;
1927 for (Int_t i=0; i<nFields; i++)
1930 for (Int_t i=0; i<fParticleSpecies->GetEntries(); i++)
1932 fParticleSpecies->GetEvent(i);
1934 Float_t* f = fParticleSpecies->GetArgs();
1936 for (Int_t j=0; j<nFields; j++)
1937 totals[j] += f[j+1];
1940 for (Int_t i=0; i<nFields; i++)
1941 Printf("%d --> %ld", i, totals[i]);
1944 void BuildResponseFromTree(const char* fileName, const char* target)
1947 // builds several response matrices with different particle ratios (systematic study)
1950 gSystem->Load("libPWG0base");
1952 TFile::Open(fileName);
1953 TNtuple* fParticleSpecies = (TNtuple*) gFile->Get("fParticleSpecies");
1955 TFile* file = TFile::Open(target, "RECREATE");
1958 Int_t tracks = 0; // control variables
1960 Int_t secondaries = 0;
1961 Int_t doubleCount = 0;
1963 for (Int_t num = 0; num < 7; num++)
1965 AliMultiplicityCorrection* fMultiplicity = new AliMultiplicityCorrection(Form("Multiplicity_%d", num), Form("Multiplicity_%d", num));
1967 Float_t ratio[4]; // pi, K, p, other
1968 for (Int_t i = 0; i < 4; i++)
1973 case 1 : ratio[1] = 0.5; break;
1974 case 2 : ratio[2] = 0.5; break;
1975 case 3 : ratio[1] = 1.5; break;
1976 case 4 : ratio[2] = 1.5; break;
1977 case 5 : ratio[1] = 0.5; ratio[2] = 0.5; break;
1978 case 6 : ratio[1] = 1.5; ratio[2] = 1.5; break;
1981 for (Int_t i=0; i<fParticleSpecies->GetEntries(); i++)
1983 fParticleSpecies->GetEvent(i);
1985 Float_t* f = fParticleSpecies->GetArgs();
1990 for (Int_t j = 0; j < 4; j++)
1992 gene += ratio[j] * f[j+1];
1993 meas += ratio[j] * f[j+1+4];
1997 // add the ones w/o label
2001 // secondaries are already part of meas!
2002 secondaries += f[10];
2004 // double counted are already part of meas!
2005 doubleCount += f[11];
2007 // ones w/o label are added without weighting to allow comparison to default analysis. however this is only valid when their fraction is low enough!
2010 //Printf("%.f %.f %.f %.f %.f", f[5], f[6], f[7], f[8], f[9]);
2012 fMultiplicity->FillCorrection(f[0], gene, gene, gene, gene, 0, meas, meas, meas, meas);
2013 fMultiplicity->FillGenerated(f[0], kTRUE, kTRUE, gene, gene, gene, gene, 0);
2014 fMultiplicity->FillMeasured(f[0], meas, meas, meas, meas);
2017 //fMultiplicity->DrawHistograms();
2019 TFile* file = TFile::Open(target, "UPDATE");
2020 fMultiplicity->SaveHistograms();
2025 Printf("%d total tracks, %d w/o label = %.2f %%, %d double counted = %.2f %%, secondaries = %.2f %%", tracks, noLabel, 100.0 * noLabel / tracks, doubleCount, 100.0 * doubleCount / tracks, 100.0 * secondaries / tracks);
2026 if ((Float_t) noLabel / tracks > 0.02)
2027 Printf("WARNING: More than 2%% of tracks without label, this might bias the study!");
2032 void MergeModifyCrossSection(const char* output)
2034 const char* files[] = { "multiplicityMC_400k_syst_nd.root", "multiplicityMC_400k_syst_sd.root", "multiplicityMC_400k_syst_dd.root" };
2036 gSystem->Load("libPWG0base");
2038 TFile::Open(output, "RECREATE");
2041 for (Int_t num=0; num<7; ++num)
2043 AliMultiplicityCorrection* data[3];
2049 case 0: ratio[0] = 1.0; ratio[1] = 1.0; ratio[2] = 1.0; break;
2050 case 1: ratio[0] = 1.0; ratio[1] = 1.5; ratio[2] = 1.0; break;
2051 case 2: ratio[0] = 1.0; ratio[1] = 0.5; ratio[2] = 1.0; break;
2052 case 3: ratio[0] = 1.0; ratio[1] = 1.0; ratio[2] = 1.5; break;
2053 case 4: ratio[0] = 1.0; ratio[1] = 1.0; ratio[2] = 0.5; break;
2054 case 5: ratio[0] = 1.0; ratio[1] = 1.5; ratio[2] = 1.5; break;
2055 case 6: ratio[0] = 1.0; ratio[1] = 0.5; ratio[2] = 0.5; break;
2059 for (Int_t i=0; i<3; ++i)
2062 name.Form("Multiplicity_%d", num);
2064 name.Form("Multiplicity_%d_%d", num, i);
2066 TFile::Open(files[i]);
2067 data[i] = new AliMultiplicityCorrection(name, name);
2068 data[i]->LoadHistograms("Multiplicity");
2071 for (Int_t j=0; j<AliMultiplicityCorrection::kMCHists; j++)
2073 data[i]->GetMultiplicityVtx(j)->Scale(ratio[i]);
2074 data[i]->GetMultiplicityMB(j)->Scale(ratio[i]);
2075 data[i]->GetMultiplicityINEL(j)->Scale(ratio[i]);
2078 for (Int_t j=0; j<AliMultiplicityCorrection::kESDHists; j++)
2079 data[i]->GetMultiplicityESD(j)->Scale(ratio[i]);
2081 for (Int_t j=0; j<AliMultiplicityCorrection::kCorrHists; j++)
2082 data[i]->GetCorrelation(j)->Scale(ratio[i]);
2088 printf("Case %d, %s: Entries in response matrix 3: ND: %.2f SD: %.2f DD: %.2f", num, data[0]->GetName(), data[0]->GetCorrelation(3)->Integral(), data[1]->GetCorrelation(3)->Integral(), data[2]->GetCorrelation(3)->Integral());
2090 data[0]->Merge(&list);
2092 Printf(" Total: %.2f", data[0]->GetCorrelation(3)->Integral());
2094 TFile::Open(output, "UPDATE");
2095 data[0]->SaveHistograms();
2100 for (Int_t i=0; i<3; ++i)
2105 void Rebin(const char* fileName = "multiplicityMC_3M.root", Int_t corrMatrix = 3)
2107 // rebins MC axis of correlation map, MC and histogram for corrected (for evaluation of effect of regularization)
2108 // rebin does not exist for 3D hists, so we convert to 2D and then back to 3D (loosing the vertex information)
2110 Printf("WARNING: Vertex information is lost in this process. Use result only for evaluation of errors.");
2112 gSystem->Load("libPWG0base");
2114 AliMultiplicityCorrection* mult = new AliMultiplicityCorrection("Multiplicity", "Multiplicity");
2116 TFile::Open(fileName);
2117 mult->LoadHistograms("Multiplicity");
2119 // rebin correlation
2120 TH3* old = mult->GetCorrelation(corrMatrix);
2122 // empty under/overflow bins in x, otherwise Project3D takes them into account
2123 for (Int_t y=1; y<=old->GetYaxis()->GetNbins(); ++y)
2125 for (Int_t z=1; z<=old->GetZaxis()->GetNbins(); ++z)
2127 old->SetBinContent(0, y, z, 0);
2128 old->SetBinContent(old->GetXaxis()->GetNbins()+1, y, z, 0);
2132 TH2* response = (TH2*) old->Project3D("zy");
2133 response->RebinX(2);
2135 TH3F* new = new TH3F(old->GetName(), old->GetTitle(),
2136 old->GetXaxis()->GetNbins(), old->GetXaxis()->GetBinLowEdge(1), old->GetXaxis()->GetBinUpEdge(old->GetXaxis()->GetNbins()),
2137 old->GetYaxis()->GetNbins() / 2, old->GetYaxis()->GetBinLowEdge(1), old->GetYaxis()->GetBinUpEdge(old->GetYaxis()->GetNbins()),
2138 old->GetZaxis()->GetNbins(), old->GetZaxis()->GetBinLowEdge(1), old->GetZaxis()->GetBinUpEdge(old->GetZaxis()->GetNbins()));
2141 Int_t vtxBin = new->GetNbinsX() / 2;
2145 for (Int_t i=1; i<=new->GetYaxis()->GetNbins(); i+=1)
2146 for (Int_t j=1; j<=new->GetZaxis()->GetNbins(); j+=1)
2147 new->SetBinContent(vtxBin, i, j, response->GetBinContent(i, j));
2149 // rebin MC + hist for corrected
2150 for (AliMultiplicityCorrection::EventType eventType = AliMultiplicityCorrection::kTrVtx; eventType <= AliMultiplicityCorrection::kINEL; eventType++)
2151 mult->GetMultiplicityMC(corrMatrix, eventType)->RebinY(2);
2153 mult->GetMultiplicityESDCorrected(corrMatrix)->Rebin(2);
2155 // recreate measured from correlation matrix to get rid of vertex shift effect
2156 TH2* newMeasured = (TH2*) old->Project3D("zx");
2157 TH2* esd = mult->GetMultiplicityESD(corrMatrix);
2160 // transfer from TH2D to TH2F
2161 for (Int_t i=0; i<=new->GetXaxis()->GetNbins()+1; i+=1)
2162 for (Int_t j=0; j<=new->GetYaxis()->GetNbins()+1; j+=1)
2163 esd->SetBinContent(i, j, newMeasured->GetBinContent(i, j));
2166 new->Project3D("zy")->DrawCopy("COLZ");
2168 TFile* file = TFile::Open("out.root", "RECREATE");
2169 mult->SetCorrelation(corrMatrix, new);
2170 mult->SaveHistograms();
2174 void EvaluateRegularizationEffect(Int_t step, const char* fileNameRebinned = "multiplicityMC_3M_rebinned.root", const char* fileNameNormal = "multiplicityMC_3M.root", Int_t histID = 3)
2176 // due to some static members in AliMultiplicityCorrection, the session has to be restarted after changing the number of parameters, to be fixed
2177 // that is why this is done in 2 steps
2179 gSystem->Load("libPWG0base");
2181 Bool_t fullPhaseSpace = kFALSE;
2185 // first step: unfold without regularization and rebinned histogram ("N=M")
2186 AliMultiplicityCorrection* mult = new AliMultiplicityCorrection("Multiplicity", "Multiplicity");
2187 TFile::Open(fileNameRebinned);
2188 mult->LoadHistograms();
2190 mult->SetRegularizationParameters(AliMultiplicityCorrection::kNone, 0, 125);
2191 mult->SetCreateBigBin(kFALSE);
2193 mult->ApplyMinuitFit(histID, fullPhaseSpace, AliMultiplicityCorrection::kTrVtx, kFALSE);
2194 mult->DrawComparison("MinuitChi2", histID, fullPhaseSpace, kTRUE, mult->GetMultiplicityVtx(histID)->ProjectionY("mymchist"));
2196 TFile* file = TFile::Open("EvaluateRegularizationEffect1.root", "RECREATE");
2197 mult->SaveHistograms();
2202 // second step: unfold with regularization and normal histogram
2203 AliMultiplicityCorrection* mult2 = new AliMultiplicityCorrection("Multiplicity", "Multiplicity");
2204 TFile::Open(fileNameNormal);
2205 mult2->LoadHistograms();
2207 mult2->SetRegularizationParameters(AliMultiplicityCorrection::kPol1, 1e4);
2208 mult2->SetCreateBigBin(kTRUE);
2209 mult2->ApplyMinuitFit(histID, fullPhaseSpace, AliMultiplicityCorrection::kTrVtx, kFALSE);
2210 mult2->DrawComparison("MinuitChi2", histID, fullPhaseSpace, kTRUE, mult2->GetMultiplicityVtx(histID)->ProjectionY("mymchist"));
2212 TH1* result2 = mult2->GetMultiplicityESDCorrected(histID);
2214 AliMultiplicityCorrection* mult = new AliMultiplicityCorrection("Multiplicity", "Multiplicity");
2215 TFile* file = TFile::Open("EvaluateRegularizationEffect1.root");
2216 mult->LoadHistograms();
2218 TH1* result1 = mult->GetMultiplicityESDCorrected(histID);
2221 TCanvas* canvas = new TCanvas("EvaluateRegularizationEffect", "EvaluateRegularizationEffect", 1000, 800);
2222 canvas->Divide(2, 2);
2225 result1->SetLineColor(1);
2226 result1->DrawCopy();
2227 result2->SetLineColor(2);
2228 result2->DrawCopy("SAME");
2232 result1->Scale(1.0 / result1->Integral());
2233 result2->Scale(1.0 / result2->Integral());
2236 result1->DrawCopy();
2237 result2->DrawCopy("SAME");
2240 TH1* diff = (TH1*) result1->Clone("diff");
2241 diff->Add(result2, -1);
2244 diff->DrawCopy("HIST");
2247 diff->Divide(result1);
2248 diff->GetYaxis()->SetRangeUser(-0.3, 0.3);
2249 diff->DrawCopy("HIST");
2252 for (Int_t i=1; i<=diff->GetNbinsX(); i++)
2253 chi2 += diff->GetBinContent(i) * diff->GetBinContent(i);
2255 Printf("Chi2 is %e", chi2);
2257 canvas->SaveAs(Form("%s.eps", canvas->GetName()));