rf514_RooCustomizer.C File Reference

Detailed Description

Using the RooCustomizer to create multiple PDFs that share a lot of properties, but have unique parameters for each category.

As an extra complication, some of the new parameters need to be functions of a mass parameter.

 
[#0] WARNING:InputArguments -- The parameter 'sigmaG' with range [-1e+30, 1e+30] of the RooGaussian 'gauss' exceeds the safe range of (0, inf). Advise to limit its range.
The proto model before customisation:
0x7ffd331ff3f8 RooAddPdf::model = 750.5/1 [Auto,Clean] 
  0x7ffd33200c00/V- RooGaussian::gauss = 0 [Auto,Dirty] 
    0x7ffd331ff010/V- RooRealVar::Energy = 1500
    0x7ffd331fec28/V- RooRealVar::meanG = 100
    0x7ffd332002d8/V- RooRealVar::sigmaG = 3
  0x7ffd331fe840/V- RooRealVar::yieldSig = 1
  0x7ffd332006c0/V- RooPolynomial::linear = 1501 [Auto,Dirty] 
    0x7ffd331ff010/V- RooRealVar::Energy = 1500
    0x7ffd331ffef0/V- RooRealVar::pol1 = 1
  0x7ffd331ffb08/V- RooRealVar::yieldBkg = 1
 
PDF 1 with a yield depending on M:
0x55cb1f101170 RooAddPdf::model_Sample1 = 1156.8/1 [Auto,Clean] 
  0x7ffd332006c0/V- RooPolynomial::linear = 1501 [Auto,Dirty] 
    0x7ffd331ff010/V- RooRealVar::Energy = 1500
    0x7ffd331ffef0/V- RooRealVar::pol1 = 1
  0x7ffd331ffb08/V- RooRealVar::yieldBkg = 1
  0x55cb1f1c5280/V- RooGaussian::gauss_Sample1 = 0 [Auto,Dirty] 
    0x7ffd331ff010/V- RooRealVar::Energy = 1500
    0x7ffd332002d8/V- RooRealVar::sigmaG = 3
    0x55cb1f1de620/V- RooRealVar::meanG_Sample1 = 100
  0x7ffd331fd8d8/V- RooFormulaVar::yieldSig_Sample1 = 0.29755 [Auto,Clean] 
    0x7ffd331fe458/V- RooRealVar::M = 1
 
PDF 2 with a yield depending on M:
0x55cb1f0ff4e0 RooAddPdf::model_Sample2 = 1000.67/1 [Auto,Clean] 
  0x7ffd332006c0/V- RooPolynomial::linear = 1501 [Auto,Dirty] 
    0x7ffd331ff010/V- RooRealVar::Energy = 1500
    0x7ffd331ffef0/V- RooRealVar::pol1 = 1
  0x7ffd331ffb08/V- RooRealVar::yieldBkg = 1
  0x55cb1f0fb580/V- RooGaussian::gauss_Sample2 = 0 [Auto,Dirty] 
    0x7ffd331ff010/V- RooRealVar::Energy = 1500
    0x7ffd332002d8/V- RooRealVar::sigmaG = 3
    0x55cb1f11ad70/V- RooRealVar::meanG_Sample2 = 100
  0x7ffd331fd4a8/V- RooFormulaVar::yieldSig_Sample2 = 0.5 [Auto,Clean] 
    0x7ffd331fe458/V- RooRealVar::M = 1
 
PDF 3 with a free yield:
0x55cb1f127ff0 RooAddPdf::model_Sample3 = 750.5/1 [Auto,Clean] 
  0x7ffd332006c0/V- RooPolynomial::linear = 1501 [Auto,Dirty] 
    0x7ffd331ff010/V- RooRealVar::Energy = 1500
    0x7ffd331ffef0/V- RooRealVar::pol1 = 1
  0x7ffd331ffb08/V- RooRealVar::yieldBkg = 1
  0x55cb1f1c7090/V- RooGaussian::gauss_Sample3 = 0 [Auto,Dirty] 
    0x7ffd331ff010/V- RooRealVar::Energy = 1500
    0x7ffd332002d8/V- RooRealVar::sigmaG = 3
    0x55cb1f11a700/V- RooRealVar::meanG_Sample3 = 100
  0x55cb1f1c38f0/V- RooRealVar::yieldSig_Sample3 = 1
 
The following leafs have been created automatically while customising:
  1) RooRealVar::    meanG_Sample1 = 100
  2) RooRealVar::    meanG_Sample2 = 100
  3) RooRealVar::    meanG_Sample3 = 100
  4) RooRealVar:: yieldSig_Sample3 = 1
 
The following leafs have been used while customising
   (partial overlap with the set of automatically created leaves.
   a new customiser for a different PDF could reuse them if necessary.):
  1) RooFormulaVar:: yieldSig_Sample1 = 0.29755
  2) RooFormulaVar:: yieldSig_Sample2 = 0.5
  3) RooRealVar::    meanG_Sample1 = 200
  4) RooRealVar::    meanG_Sample2 = 300
  5) RooRealVar::    meanG_Sample3 = 100
  6) RooRealVar:: yieldSig_Sample3 = 1

 
 
#include "RooRealVar.h"
#include "RooGaussian.h"
#include "RooPolynomial.h"
#include "RooAddPdf.h"
#include "RooCustomizer.h"
#include "RooCategory.h"
#include "RooFormulaVar.h"
#include <iostream>
 
void rf514_RooCustomizer() {
 
  // Define a proto model that will be used as the template for each category
  // ---------------------------------------------------------------------------
 
  RooRealVar E("Energy","Energy",0,3000);
 
  RooRealVar meanG("meanG","meanG", 100., 0., 3000.);
  RooRealVar sigmaG("sigmaG","sigmaG", 3.);
  RooGaussian gauss("gauss", "gauss", E, meanG, sigmaG);
 
  RooRealVar pol1("pol1", "Constant of the polynomial", 1, -10, 10);
  RooPolynomial linear("linear", "linear", E, pol1);
 
  RooRealVar yieldSig("yieldSig", "yieldSig", 1, 0, 1.E4);
  RooRealVar yieldBkg("yieldBkg", "yieldBkg", 1, 0, 1.E4);
 
  RooAddPdf model("model", "S + B model",
      RooArgList(gauss,linear),
      RooArgList(yieldSig, yieldBkg));
 
  std::cout << "The proto model before customisation:" << std::endl;
  model.Print("T"); // "T" prints the model as a tree
 
 
  // Build the categories
  RooCategory sample("sample","sample");
  sample["Sample1"] = 1;
  sample["Sample2"] = 2;
  sample["Sample3"] = 3;
 
 
  // Start to customise the proto model that was defined above.
  // ---------------------------------------------------------------------------
 
  // We need two sets for bookkeeping of PDF nodes:
  RooArgSet newLeafs;           // This set collects leafs that are created in the process.
  RooArgSet allCustomiserNodes; // This set lists leafs that have been used in a replacement operation.
 
 
  // 1. Each sample should have its own mean for the gaussian
  // The customiser will make copies of `meanG` for each category.
  // These will all appear in the set `newLeafs`, which will own the new nodes.
  RooCustomizer cust(model, sample, newLeafs, &allCustomiserNodes);
  cust.splitArg(meanG, sample);
 
 
  // 2. Each sample should have its own signal yield, but there is an extra complication:
  // We need the yields 1 and 2 to be a function of the variable "mass".
  // For this, we pre-define nodes with exacly the names that the customiser would have created automatically,
  // that is, "<nodeName>_<categoryName>", and we register them in the set of customiser nodes.
  // The customiser will pick them up instead of creating new ones.
  // If we don't provide one (e.g. for "yieldSig_Sample3"), it will be created automatically by cloning `yieldSig`.
  RooRealVar mass("M", "M", 1, 0, 12000);
  RooFormulaVar yield1("yieldSig_Sample1", "Signal yield in the first sample", "M/3.360779", mass);
  RooFormulaVar yield2("yieldSig_Sample2", "Signal yield in the second sample", "M/2", mass);
  allCustomiserNodes.add(yield1);
  allCustomiserNodes.add(yield2);
 
  // Instruct the customiser to replace all yieldSig nodes for each sample:
  cust.splitArg(yieldSig, sample);
 
 
  // Now we can start building the PDFs for all categories:
  auto pdf1 = cust.build("Sample1");
  auto pdf2 = cust.build("Sample2");
  auto pdf3 = cust.build("Sample3");
 
  // And we inspect the two PDFs
  std::cout << "\nPDF 1 with a yield depending on M:" << std::endl;
  pdf1->Print("T");
  std::cout << "\nPDF 2 with a yield depending on M:" << std::endl;
  pdf2->Print("T");
  std::cout << "\nPDF 3 with a free yield:" << std::endl;
  pdf3->Print("T");
 
  std::cout << "\nThe following leafs have been created automatically while customising:" << std::endl;
  newLeafs.Print("V");
 
 
  // If we needed to set reasonable values for the means of the gaussians, this could be done as follows:
  auto& meanG1 = static_cast<RooRealVar&>(allCustomiserNodes["meanG_Sample1"]);
  meanG1.setVal(200);
  auto& meanG2 = static_cast<RooRealVar&>(allCustomiserNodes["meanG_Sample2"]);
  meanG2.setVal(300);
 
  std::cout << "\nThe following leafs have been used while customising"
    << "\n\t(partial overlap with the set of automatically created leaves."
    << "\n\ta new customiser for a different PDF could reuse them if necessary.):" << std::endl;
  allCustomiserNodes.Print("V");
 
 
}

Author

Stephan Hageboeck, CERN

Definition in file rf514_RooCustomizer.C.