sPhenix code displayed by LXR master/​macros/​calibrations/​mbd/​recal_mbd_mip.C	Source navigation Diff markup Identifier search General search
	Version: master Revert   Change

File indexing completed on 2025-08-03 08:20:22

 //
 // Do a recalibration of the mip from the saved histograms
 //
 #include <TSpectrum.h>
 #include <mbd/MbdCalib.h>
 
 #include "get_runstr.h"
 
 
 #if defined(__CLING__)
 //R__LOAD_LIBRARY(libmbd_io.so)
 R__LOAD_LIBRARY(libmbd.so)
 #endif
 
 //using namespace MBDRUNS;
 
 Double_t qmin = 25.;
 Double_t qmax = 1600;
 TF1 *bkgf[128]{nullptr};
 
 double minrej = 2000.;
 double peak = 2000.;
 double maxrej = 5000.;
 Double_t powerlaw(Double_t *x, Double_t *par)
 {
   Float_t xx =x[0];
   if ( xx>minrej && xx<maxrej )
   {
     TF1::RejectPoint();
     return 0;
   }
   Double_t f = par[0]*pow((par[1]/(par[1]+xx)),par[2]); 
   return f;
 }
 
 
 // Two gaussians
 // [0] = ampl, peak 1
 // [1] = mean
 // [2] = sigma
 // [3] = ampl, peak 2
 Double_t gaus2(Double_t *x, Double_t *par)
 {
   Double_t xx =x[0];
   Double_t f = par[0]*TMath::Gaus(xx,par[1],par[2]) + par[3]*TMath::Gaus(xx,2.0*par[1],sqrt(2)*par[2]); 
   return f;
 }
 
 Double_t woodssaxon(Double_t *x, Double_t *par)
 {
   Double_t xx =x[0];
   Double_t e = par[0]/(1.0+exp((xx-par[1])/par[2]));
   return e;
 }
 
 // Two gaussians + expo
 // [0] = ampl, peak 1
 // [1] = mean
 // [2] = sigma
 // [3] = ampl, peak 2
 // [4] = ampl, woods-saxon
 // [5] = a, woods-saxon 
 // [6] = slope, expo
 Double_t gaus2ws(Double_t *x, Double_t *par)
 {
   Double_t xx =x[0];
   Double_t f = par[0]*TMath::Gaus(xx,par[1],par[2]) + par[3]*TMath::Gaus(xx,2.0*par[1],sqrt(2)*par[2]); 
   //Double_t e = par[4]*TMath::Exp(-par[5]*xx);
   Double_t e = par[4]/(1.0+exp((xx-par[5])/par[6]));
   return f + e;
 }
 
 
 Double_t landau2(Double_t *x, Double_t *par)
 {
   Double_t xx =x[0];
   Double_t f = par[0]*TMath::Landau(xx,par[1],par[2]) + par[3]*TMath::Landau(xx,2.0*par[1],par[4]); 
   return f;
 }
 
 Double_t langaufun(Double_t *x, Double_t *par)
 {
   //Fit parameters:
   //par[0]=Width (scale) parameter of Landau density
   //par[1]=Most Probable (MP, location) parameter of Landau density
   //par[2]=Total area (integral -inf to inf, normalization constant)
   //par[3]=Width (sigma) of convoluted Gaussian function
   //
   //In the Landau distribution (represented by the CERNLIB approximation),
   //the maximum is located at x=-0.22278298 with the location parameter=0.
   //This shift is corrected within this function, so that the actual
   //maximum is identical to the MP parameter.
 
   // Numeric constants
   Double_t invsq2pi = 0.3989422804014;   // (2 pi)^(-1/2)
   Double_t mpshift  = -0.22278298;       // Landau maximum location
 
   // Control constants
   Double_t np = 100.0;      // number of convolution steps
   Double_t sc =   5.0;      // convolution extends to +-sc Gaussian sigmas
 
   // Variables
   Double_t xx;
   Double_t mpc;
   Double_t fland;
   Double_t sum = 0.0;
   Double_t xlow,xupp;
   Double_t step;
   Double_t i;
 
 
   // MP shift correction
   mpc = par[1] - mpshift * par[0];
 
   // Range of convolution integral
   xlow = x[0] - sc * par[3];
   xupp = x[0] + sc * par[3];
 
   step = (xupp-xlow) / np;
 
   // Convolution integral of Landau and Gaussian by sum
   for(i=1.0; i<=np/2; i++) {
     xx = xlow + (i-.5) * step;
     fland = TMath::Landau(xx,mpc,par[0]) / par[0];
     sum += fland * TMath::Gaus(x[0],xx,par[3]);
 
     xx = xupp - (i-.5) * step;
     fland = TMath::Landau(xx,mpc,par[0]) / par[0];
     sum += fland * TMath::Gaus(x[0],xx,par[3]);
   }
 
   return (par[2] * step * sum * invsq2pi / par[3]);
 }
 
 
 // find the threshold
 // may also want to consider getting threshold from turn-on curves
 void FindThreshold(TH1 *h, double& threshold)
 {
   threshold = 0.;
   double absolute_min = 10.;
   double absolute_max = 70.;    // max adc where the threshold could be
   int absminbin = h->FindBin( absolute_min );
   int absmaxbin = h->FindBin( absolute_max );
   double absmaxval = h->GetBinContent( h->GetMaximumBin() );
   int maxbin = 0;
   double maxval = 0.;
   double prev_val = 0.;
   int maxjumpbin = 0;
   double maxratio = 0.;
 
   for (int ibin=absminbin; ibin<=absmaxbin; ibin++)
   {
     double val = h->GetBinContent(ibin);
     if ( val > maxval )
     {
       maxval = val;
       maxbin = ibin;
     }
 
     if ( val>0. && prev_val>0. )
     {
       double ratio = val/prev_val;
       if ( val>(absmaxval*0.25) )
       {
         //cout << ibin << "\t" << ratio << endl;
         if ( ratio>maxratio )
         {
           maxratio = ratio;
           maxjumpbin = ibin;
         }
       }
     }
 
     prev_val = val;
   }
 
   if ( maxbin==absmaxbin )
   {
     // no exponential before mip peak, use max jump to find threshold
     // (Note: should raise gain for this channel)
     threshold = h->GetBinLowEdge( maxjumpbin );
   }
   else
   {
     // use max bin in threshold range
     threshold = h->GetBinLowEdge( maxbin );
   }
   //cout << threshold << endl;
 }
 
 
 // xmin and xmax are the min and max range of the peak
 // This version uses TSpectrum to find the peaks
 /*
 void FindPeakRange(TH1 *h, double& xmin, double& peak, double& xmax, double threshold)
 {
   int verbose = 1;    // Peak finder
   if ( verbose )
   {
     static TCanvas *ac = new TCanvas("cpeak","peakfinder",800,600);
     ac->cd();
     h->GetXaxis()->SetRangeUser( qmin, qmax );
     h->Draw();
   }
 
   const Int_t maxPeaks = 1;
   TSpectrum spectrum(maxPeaks);
 
   // sigma: minimum expected width of a peak in bins (e.g., sigma=2 means width ≈ 4 bins)
   Double_t sigma = 5;
 
   // Threshold = minimum relative height (0.1 = 10% of max)
   Int_t nPeaks = spectrum.Search(h, sigma, "", 0.1);
 
   std::cout << "Found " << nPeaks << " peaks:\n";
 
   // Get peak positions
   Double_t* peaksX = spectrum.GetPositionX();
   for (Int_t i = 0; i < nPeaks; ++i) {
     Double_t x = peaksX[i];
     Double_t y = h->GetBinContent(h->FindBin(x));
     std::cout << "  Peak at x = " << x << ", height = " << y << "\n";
   }
 
   if ( verbose )
   {
     gPad->Modified();
     gPad->Update();
     std::string junk;
     std::cout << "? ";
     std::cin >> junk;
   }
 }
 */
 
 void FindPeakRange(TH1 *h, double& xmin, double& peak, double& xmax, double threshold)
 {
   int verbose = 1;
 
   int bin = h->FindBin( threshold );
   int maxbin = h->FindBin( qmax );
   double ymin = 1e12; // the minimum y val
   int nabove = 0;     // num points above the min
 
   // look for 1st min
   int ibin = bin;
   while ( ibin<=maxbin )
   {
     double val = h->GetBinContent( ibin );
     if ( val < ymin )
     {
       ymin = val;
       nabove = 0; // new min, reset nabove
     }
     else
     {
       nabove++;
     }
 
     if ( verbose )
     {
       double x = h->GetBinCenter( ibin );
       std::cout << "bin x y nabove " << ibin << "\t" << x << "\t" << val << "\t" << nabove << std::endl;
     }
 
     // if we see this many above the min, the signal is rising
     if ( nabove==20 )
     {
       xmin = h->GetBinCenter( ibin-20 );
       break;
     }
 
     ibin++;
   }
 
   // now look for peak after first min
   double ymax = 0; // the minimum y val
   int nbelow = 0;     // num points below the max
   while ( ibin<=maxbin )
   {
     double val = h->GetBinContent( ibin );
     if ( val > ymax )
     {
       ymax = val;
       nbelow = 0; // new max, reset nbelow
     }
     else
     {
       nbelow++;
     }
 
     // if we see this many below the max, the signal is falling
     if ( nbelow==20 )
     {
       peak = h->GetBinCenter( ibin-20 );
       break;
     }
 
     ibin++;
   }
 
   xmax = peak + 4*(peak - xmin);
 
 }
 
 
 // type0: auau200
 // type1: pp200
 // method0:  TSpectrum bkg + 2 gaus fit
 void recal_mbd_mip(const char *tfname = "DST_MBDUNCAL-00020869-0000.root", const int pass = 3, const int type = 0, const int method = 0)
 {
   cout << "recal_mbd_mip(), type method " << type << " " << method << endl;
   cout << "tfname " << tfname << endl;
 
   const int NUM_PMT = 128;
   //const int NUM_PMT = 2;
   const int NUM_ARMS = 2;
 
   // Create new TFile
   int runnumber = get_runnumber(tfname);
   TString dir = "results/";
   dir += runnumber;
   dir += "/";
   TString name = "mkdir -p "; name += dir;
   gSystem->Exec( name );
 
   name = dir; name += "calmbdpass2.3_q-"; name += runnumber; name += ".root";
 
   // Read in TFile with h_q
   TFile *oldfile = new TFile(name,"READ");
 
   TH1 *h_q[NUM_PMT];
   TH1 *h_tq[NUM_PMT];
 
   TString title;
   for (int ipmt=0; ipmt<NUM_PMT; ipmt++)
   {
     name = "h_q"; name += ipmt;
     title = "q"; title += ipmt;
     //h_q[ipmt] = new TH1F(name,title,15100/4,-100,15000);
     h_q[ipmt] = (TH1*)oldfile->Get(name);
     if ( type == MBDRUNS::AUAU200 )
     {
       //h_q[ipmt]->Rebin(2);
     }
     else if ( type == MBDRUNS::PP200 )
     {
       //h_q[ipmt]->Rebin(4);  // b-off
       //h_q[ipmt]->Rebin(2);
     }
 
     name = "h_tq"; name += ipmt;
     title = "tq"; title += ipmt;
     //h_tq[ipmt] = new TH1F(name,title,7000,-150,31*17.76);
     h_tq[ipmt] = (TH1*)oldfile->Get(name);
   }
   //TH2 *h2_tq = new TH2F("h2_tq","ch vs tq",900,-150,150,NUM_PMT,-0.5,NUM_PMT-0.5);
   TH2 *h2_tq = (TH2*)oldfile->Get("h2_tq");
 
   name = dir; 
   name += "recalmbd_pass"; name += pass; name += ".root";
   cout << name << endl;
 
   TFile *savefile = new TFile(name,"RECREATE");
 
   // Load in calib constants
 
   TCanvas *ac[100];
   int cvindex = 0;
 
   //q
   ac[cvindex] = new TCanvas("cal_q","q",425*1.5,550*1.5);
   if ( pass>0 )
   {
     ac[cvindex]->Divide(1,2);
     ac[cvindex]->cd(1);
   }
 
   ofstream cal_mip_file;
   TString calfname;
   if ( pass==3 ) 
   {
     calfname = dir; calfname += "mbd_qfit.calib";
     cal_mip_file.open( calfname );
     std::cout << "Writing to " << calfname << std::endl;
   }
 
   // Fit ranges
   // Run23AuAu
   qmin = 25.;
   qmax = 1600;
 
   if ( type==MBDRUNS::PP200 )
   {
     cout << "setting up for pp200" << endl;
     //qmin = 200;
     //qmax = 10000;       // Run24pp boff
     //qmin = 100;
     //qmin = 50;
     qmin = 10;
     qmax = 2000;       // Run24pp bon
   }
 
   if ( type==MBDRUNS::SIMAUAU200 || type==MBDRUNS::SIMPP200 )
   {
     qmin = 0.25;
     qmax = 6.;
   }
 
   TF1 *mipfit[128];
 
   TH1 *h_bkg[NUM_PMT];  // background histogram
   TH1 *h_mip[NUM_PMT];  // mip signal histogram
   TH1 *h_bkgmip[NUM_PMT];  // bkg + fit histogram
 
   //TF1 *fws = new TF1("fws",woodssaxon,0,1000,3);
   //fws->SetParameters(-85.27,559.8,99.77);
 
   // Set up output pdf to save plots
   TString pdfname = dir; pdfname += "calmbdpass2."; pdfname += pass; pdfname += "_mip-"; pdfname += runnumber; pdfname += ".pdf";
   cout << pdfname << endl;
   ac[cvindex]->Print( pdfname + "[" );
 
   for (int ipmt=0; ipmt<NUM_PMT; ipmt++)
   {
     if (pass>0)
     {
       h_bkg[ipmt] = (TH1*)h_q[ipmt]->Clone();
       name = h_q[ipmt]->GetName(); name.ReplaceAll("q","bkg");
       h_bkg[ipmt]->SetName( name );
       h_bkg[ipmt]->SetTitle( name );
       h_bkg[ipmt]->SetLineColor(2);
       double threshold = qmin + 2.;
       FindThreshold( h_q[ipmt], threshold );
       cout << "threshold " << ipmt << "\t" << threshold << endl;
       h_q[ipmt]->GetXaxis()->SetRangeUser( threshold, qmax );
       h_q[ipmt]->Sumw2();
 
       double sigma = 20;
       double seedmean = 0;
       TSpectrum s{};
       if ( method==0 )
       {
         h_bkg[ipmt] = s.Background( h_q[ipmt] );
         //h_bkg[ipmt]->Add( fws );
 
         h_mip[ipmt] = (TH1*)h_q[ipmt]->Clone();
         name = h_q[ipmt]->GetName(); name.ReplaceAll("q","mip");
         h_mip[ipmt]->SetName( name );
         h_mip[ipmt]->SetTitle( name );
         h_mip[ipmt]->Add( h_bkg[ipmt], -1.0 );
 
         /*
            Int_t nfound = s.Search(h_q[ipmt],sigma,"",0.1);
            Double_t *xpeaks = s.GetPositionX();
 
         //h_bkg[ipmt] = s.Background( h_q[ipmt] );
 
         double best_peak = xpeaks[0];
         if ( best_peak < 50. )
         {
         best_peak = xpeaks[1];
         }
 
         cout << "peaks\t" << ipmt << "\t" << nfound << "\t" << best_peak << endl;
         */
 
       }
       else if ( method==1 )
       {
         FindPeakRange( h_bkg[ipmt], minrej, peak, maxrej, threshold );
         cout << "peak range\t" << minrej << "\t" << peak << "\t" << maxrej << endl;
         sigma = peak-minrej;
         seedmean = peak;
 
         // Fit background
         name = "bkgf"; name += ipmt;
         bkgf[ipmt] = new TF1(name,powerlaw,qmin,qmax,3);
         bkgf[ipmt]->SetParameter(0,240e-5);
         bkgf[ipmt]->SetParameter(1,1240);
         bkgf[ipmt]->SetParameter(2,2);
         bkgf[ipmt]->SetLineColor(3);
         h_bkg[ipmt]->Draw();
         //h_bkg[ipmt]->Fit(bkgf[ipmt],"MR");
         h_bkg[ipmt]->Fit(bkgf[ipmt],"R");
 
         // get mip histogram
         h_mip[ipmt] = (TH1*)h_q[ipmt]->Clone();
         name = h_q[ipmt]->GetName(); name.ReplaceAll("q","mip");
         h_mip[ipmt]->SetName( name );
         h_mip[ipmt]->SetTitle( name );
 
         double orig_minrej = minrej;
         double orig_maxrej = maxrej;
         minrej = 0.;
         maxrej = 0.;
         h_mip[ipmt]->Add( bkgf[ipmt], -1.0 );
       }
 
       // Fit the mip peak after background subtraction
       name = "mipfit"; name += ipmt;
       /*
       // Laudau fit
       mipfit[ipmt] = new TF1(name,"landau",qmin,qmax);
       mipfit[ipmt]->SetLineColor(4);
       mipfit[ipmt]->SetParameter( 0, 5.0*h_mip[ipmt]->GetMaximum() );
       Double_t seedmean = h_mip[ipmt]->GetBinCenter( h_mip[ipmt]->GetMaximumBin() );
       cout << "SEEDMEAN " << seedmean << endl;
       mipfit[ipmt]->SetParameter( 1, seedmean );
       mipfit[ipmt]->SetParameter( 2, 20. );
       */
       // Two gaussian fit
       mipfit[ipmt] = new TF1(name,gaus2,qmin,qmax,4);
       //mipfit[ipmt] = new TF1(name,gaus2expo,qmin,qmax,7);
       //mipfit[ipmt] = new TF1(name,"gaus",qmin,qmax,4);
 
       //mipfit[ipmt] = new TF1(name,langaufun,qmin,qmax,4);
 
       seedmean = h_mip[ipmt]->GetBinCenter( h_mip[ipmt]->GetMaximumBin() );
       cout << "SEEDMEAN " << seedmean << endl;
       Double_t seedsigma = sigma;
       if ( type==MBDRUNS::SIMAUAU200 )
       {
         seedsigma = 0.2;
       }
       else if ( type==MBDRUNS::PP200 )
       {
         seedsigma = seedmean*(116./719.);  // b-on
         //seedsigma = 50;  // b-on
         //seedsigma = 200;  // b-off
       }
 
       mipfit[ipmt]->SetLineColor(4);
       // for langaus
       /*
          mipfit[ipmt]->SetParameter( 0, seedsigma );
          mipfit[ipmt]->SetParameter( 1, seedmean );
          mipfit[ipmt]->SetParameter( 2, mipfit[ipmt]->GetParameter(0)*0.1 );
          mipfit[ipmt]->SetParameter( 3, 0.12*seedsigma );
          */
       // for gaus2
       mipfit[ipmt]->SetParameter( 0, 5.0*h_mip[ipmt]->GetMaximum() );
       mipfit[ipmt]->SetParameter( 1, seedmean );
       mipfit[ipmt]->SetParameter( 2, seedsigma );
       mipfit[ipmt]->SetParameter( 3, mipfit[ipmt]->GetParameter(0)*0.1 );
       //mipfit[ipmt]->SetParameter( 4, mipfit[ipmt]->GetParameter(0)*0.01 );
       //mipfit[ipmt]->SetParameter( 5, 200.);
       //mipfit[ipmt]->SetParameter( 6, 20.);
       //mipfit[ipmt]->SetParameter( 4, mipfit[ipmt]->GetParameter(1) );
 
       h_mip[ipmt]->Fit( mipfit[ipmt], "RM" );
 
       double integ = mipfit[ipmt]->GetParameter(0);
       double best_peak = mipfit[ipmt]->GetParameter(1);
       double width = mipfit[ipmt]->GetParameter(2);
       double integerr = mipfit[ipmt]->GetParError(0);
       double best_peakerr = mipfit[ipmt]->GetParError(1);
       double widtherr = mipfit[ipmt]->GetParError(2);
       double chi2 = mipfit[ipmt]->GetChisquare();
       double ndf = mipfit[ipmt]->GetNDF();
 
       cal_mip_file << ipmt << "\t" << integ << "\t" << best_peak << "\t" << width << "\t"
         << integerr << "\t" << best_peakerr << "\t" << widtherr << "\t"
         << chi2/ndf << endl;
       cout << ipmt << "\t" << integ << "\t" << best_peak << "\t" << width << "\t"
         << integerr << "\t" << best_peakerr << "\t" << widtherr << "\t"
         << chi2/ndf << endl;
 
       // Get full fit
       h_bkgmip[ipmt] = (TH1*)h_bkg[ipmt]->Clone();
       name = h_q[ipmt]->GetName(); name.ReplaceAll("q","bkgmip");
       h_bkgmip[ipmt]->SetName( name );
       h_bkgmip[ipmt]->SetTitle( name );
       h_bkgmip[ipmt]->Add( mipfit[ipmt] );
       h_bkgmip[ipmt]->SetLineColor( 8 );
 
       /*
          gPad->Modified();
          gPad->Update();
       //mipfit->SetRange( xpeaks[0]-2.5*sigma, 600 );
       mipfit->SetRange( qmin, 600 );
       mipfit->SetParameter( 0, 1000 );
       mipfit->SetParameter( 1, best_peak );
       mipfit->SetParameter( 2, sigma );
 
       mipfit->SetParameter( 8, f_expo->GetParameter(0) );
       mipfit->SetParameter( 9, f_expo->GetParameter(1) );
       h_q[ipmt]->Fit( mipfit, "R" );
       */
 
 
       /*
          int npar = f_bkg->GetNpar();
          for (int ipar=0; ipar<npar; ipar++)
          {
          f_bkg->SetParameter( ipar, mipfit->GetParameter(ipar+3) );
          }
          */
 
       // Now draw the full dist, plus fit
       ac[cvindex]->cd(1);
       gPad->SetLogy(1);
       h_q[ipmt]->GetXaxis()->SetRangeUser( qmin, qmax );
       //h_q[ipmt]->SetMinimum(10.);
       h_q[ipmt]->Draw();
       h_bkg[ipmt]->Draw("same");
       h_bkgmip[ipmt]->Draw("same");
 
       gPad->Modified();
       gPad->Update();
 
       ac[cvindex]->cd(2);
       h_mip[ipmt]->Draw();
       gPad->Modified();
       gPad->Update();
 
       /*
          string junk;
          cout << "? ";
          cin >> junk;
          */
 
       //name = dir + "h_qfit"; name += ipmt; name += ".png";
       title = "h_qfit"; title += ipmt;
       //cout << pdfname << " " << title << endl;
       ac[cvindex]->Print( pdfname, title );
     }
 
   }
 
   ac[cvindex]->Print( pdfname + "]" );
   ++cvindex;
 
   if ( pass==3 )
   {
     cal_mip_file.close();
 
     // Convert to CDB format
     MbdCalib mcal;
     mcal.Download_Gains( calfname.Data() );
     calfname.ReplaceAll(".calib",".root");
     mcal.Write_CDB_Gains( calfname.Data() );
   }
 
   savefile->Write();
   savefile->Close();
 }
 

This page was automatically generated by the 2.3.7 LXR engine. The LXR team