Merge pull request #47055 from cms-trackeralign/mm_dev_add_general_tk_modules_in_all-in-one

Introduce a `Generic` validation tool for dataset validation in the alignment all-in-one tool

Author: cmsbuild

Alignment/OfflineValidation/README.md

@@ -117,5 +117,11 @@ For details read [`README_JetHT.md`](https://github.com/cms-sw/cmssw/blob/master
 ## MTS validation
 For details read [`README_MTS.md`](https://github.com/cms-sw/cmssw/blob/master/Alignment/OfflineValidation/README_MTS.md)
 
+## Pixel BaryCenter
+For details read [`README_MTS.md`](https://github.com/cms-sw/cmssw/blob/master/Alignment/OfflineValidation/README_PixBary.md)
+
+## Generic validation (dataset validation)
+For details read [`README_Generic.md`](https://github.com/cms-sw/cmssw/blob/master/Alignment/OfflineValidation/README_Generic.md)
+
 ## General info about IOV/run arguments
 For details read [`README_IOV.md`](https://github.com/cms-sw/cmssw/blob/master/Alignment/OfflineValidation/README_IOV.md)

Alignment/OfflineValidation/README_Generic.md

@@ -0,0 +1,39 @@
+# Generic Validation
+
+## General info
+```
+validations:
+    Generic:
+        <step_type>:
+            <step_name>: 
+                <options>
+```
+
+Generic validation runs in 2 possible types of steps:
+ - single (validation analysis by GenericV_cfg.py)
+ - (optional) merge (GenericVmerge macro)
+Step name is and arbitrary string which will be used as a reference for consequent steps.
+Merge job awill only start if all corresponding single jobs are done.
+Merge jobs can run in parallel.
+
+## Single Generic jobs
+Single jobs can be specified per run (IoV as well). In case of MC, IoV is specified to arbitrary 1.  
+
+Variable | Default value | Explanation/Options
+-------- | ------------- | --------------------
+IOV | None | List of IOVs/runs defined by integer value. IOV 1 is reserved for MC.
+Alignments | None | List of alignments. Will create separate directory for each.
+dataset | See defaultInputFiles_cff.py | Path to txt file containing list of datasets to be used. If file is missing at EOS or is corrupted - job will eventually fail (most common issue).
+goodlumi | cms.untracked.VLuminosityBlockRange() | Path to json file containing lumi information about selected IoV - must contain list of runs under particular IoV with lumiblock info. Format: `IOV_Vali_{}.json`
+maxevents | 1 | Maximum number of events before cmsRun terminates.
+trackcollection | "generalTracks" | Track collection to be specified here, e.g. "ALCARECOTkAlMuonIsolated" or "ALCARECOTkAlMinBias" ... 
+tthrbuilder | "WithAngleAndTemplate" | Specify TTRH Builder
+
+## Merge Generic job
+Its name do not need to match single job name but option `singles` must list all single jobs to be merged.
+Generic merged plot style can be adjusted from global plotting style.
+
+Variable | Default value | Explanation/Options
+-------- | ------------- | --------------------
+singles | None | List of strings matching single job names to be merged in one plot.
+customrighttitle | "" | Top right title. Reserved word "IOV" will be replaced for given IOV/run in the list.

Alignment/OfflineValidation/bin/BuildFile.xml

@@ -18,6 +18,7 @@
 <bin name="SplitVmerge" file="SplitVmerge.cc,Options.cc" />
 <bin name="Zmumumerge" file="Zmumumerge.cc,Options.cc" />
 <bin name="DiMuonVmerge" file="DiMuonVmerge.cc,Options.cc" />
+<bin name="GenericVmerge" file="GenericVmerge.cc,Options.cc" />
 <bin name="MTSmerge" file="MTSmerge.cc,Options.cc" />
 <bin name="haddws" file="haddws.C" />
 <bin name="jetHtPlotter" file="jetHtPlotter.cc,JetHtPlotConfiguration.cc,Options.cc" />

Alignment/OfflineValidation/bin/GenericVmerge.cc

@@ -0,0 +1,73 @@
+#include <cstdlib>
+#include <string>
+#include <iostream>
+#include <numeric>
+#include <functional>
+
+#include "exceptions.h"
+#include "toolbox.h"
+#include "Options.h"
+
+#include "boost/filesystem.hpp"
+#include "boost/property_tree/ptree.hpp"
+#include "boost/property_tree/json_parser.hpp"
+#include "boost/optional.hpp"
+
+#include "TString.h"
+#include "TASImage.h"
+#include "TGraph.h"
+
+#include "Alignment/OfflineValidation/macros/loopAndPlot.C"
+#include "Alignment/OfflineValidation/interface/TkAlStyle.h"
+
+using namespace std;
+using namespace AllInOneConfig;
+
+namespace pt = boost::property_tree;
+
+int merge(int argc, char* argv[]) {
+  // parse the command line
+
+  Options options;
+  options.helper(argc, argv);
+  options.parser(argc, argv);
+
+  //Read in AllInOne json config
+  pt::ptree main_tree;
+  pt::read_json(options.config, main_tree);
+
+  pt::ptree alignments = main_tree.get_child("alignments");
+  pt::ptree validation = main_tree.get_child("validation");
+
+  TString filesAndLabels;
+  for (const auto& childTree : alignments) {
+    // Print node name and its attributes
+    // std::cout << "Node: " << childTree.first << std::endl;
+    // for (const auto& attr : childTree.second) {
+    //   std::cout << "  Attribute: " << attr.first << " = " << attr.second.data() << std::endl;
+    // }
+
+    std::string file = childTree.second.get<string>("file");
+    std::cout << file << std::endl;
+    std::cout << childTree.second.get<string>("title") << std::endl;
+
+    // Check if the file contains "/eos/cms/" and add the prefix accordingly
+    std::string prefixToAdd = file.find("/eos/cms/") != std::string::npos ? "root://eoscms.cern.ch/" : "";
+    std::string toAdd = prefixToAdd + file + "/GenericValidation.root=" + childTree.second.get<string>("title") + ",";
+    filesAndLabels += toAdd;
+  }
+
+  if (filesAndLabels.Length() > 0) {
+    filesAndLabels.Remove(filesAndLabels.Length() - 1);  // Remove the last character
+  }
+
+  std::cout << "filesAndLabels: " << filesAndLabels << std::endl;
+
+  loopAndPlot(filesAndLabels);
+
+  return EXIT_SUCCESS;
+}
+
+#ifndef DOXYGEN_SHOULD_SKIP_THIS
+int main(int argc, char* argv[]) { return exceptions<merge>(argc, argv); }
+#endif

Alignment/OfflineValidation/plugins/GeneralPurposeVertexAnalyzer.cc

@@ -52,6 +52,50 @@
 
 using reco::TrackCollection;
 
+namespace gpVertexAnalyzer {
+  void setBinLog(TAxis *axis) {
+    int bins = axis->GetNbins();
+    float from = axis->GetXmin();
+    float to = axis->GetXmax();
+    float width = (to - from) / bins;
+    std::vector<float> new_bins(bins + 1, 0);
+    for (int i = 0; i <= bins; i++) {
+      new_bins[i] = TMath::Power(10, from + i * width);
+    }
+    axis->Set(bins, new_bins.data());
+  }
+
+  void setBinLogX(TH1 *h) {
+    TAxis *axis = h->GetXaxis();
+    setBinLog(axis);
+  }
+
+  void setBinLogY(TH1 *h) {
+    TAxis *axis = h->GetYaxis();
+    setBinLog(axis);
+  }
+
+  template <typename... Args>
+  TProfile *makeProfileIfLog(const edm::Service<TFileService> &fs, bool logx, bool logy, Args &&...args) {
+    auto prof = fs->make<TProfile>(std::forward<Args>(args)...);
+    if (logx)
+      setBinLogX(prof);
+    if (logy)
+      setBinLogY(prof);
+    return prof;
+  }
+
+  template <typename... Args>
+  TH1D *makeTH1IfLog(const edm::Service<TFileService> &fs, bool logx, bool logy, Args &&...args) {
+    auto h1 = fs->make<TH1D>(std::forward<Args>(args)...);
+    if (logx)
+      setBinLogX(h1);
+    if (logy)
+      setBinLogY(h1);
+    return h1;
+  }
+}  // namespace gpVertexAnalyzer
+
 class GeneralPurposeVertexAnalyzer : public edm::one::EDAnalyzer<edm::one::SharedResources> {
 public:
   explicit GeneralPurposeVertexAnalyzer(const edm::ParameterSet &);
@@ -73,9 +117,9 @@ class GeneralPurposeVertexAnalyzer : public edm::one::EDAnalyzer<edm::one::Share
   struct IPMonitoring {
     std::string varname_;
     float pTcut_;
-    TH1D *IP_, *IPErr_;
-    TProfile *IPVsPhi_, *IPVsEta_;
-    TProfile *IPErrVsPhi_, *IPErrVsEta_;
+    TH1D *IP_, *IPErr_, *IPPull_;
+    TProfile *IPVsPhi_, *IPVsEta_, *IPVsPt_;
+    TProfile *IPErrVsPhi_, *IPErrVsEta_, *IPErrVsPt_;
     TProfile2D *IPVsEtaVsPhi_, *IPErrVsEtaVsPhi_;
 
     void bookIPMonitor(const edm::ParameterSet &, const edm::Service<TFileService> fs);
@@ -128,7 +172,7 @@ class GeneralPurposeVertexAnalyzer : public edm::one::EDAnalyzer<edm::one::Share
   TH1D *type[2];
   TH1D *bsX, *bsY, *bsZ, *bsSigmaZ, *bsDxdz, *bsDydz, *bsBeamWidthX, *bsBeamWidthY, *bsType;
 
-  TH1D *sumpt, *ntracks, *weight, *chi2ndf, *chi2prob;
+  TH1D *trackpt, *sumpt, *ntracks, *weight, *chi2ndf, *chi2prob;
   TH1D *dxy2;
   TH1D *phi_pt1, *eta_pt1;
   TH1D *phi_pt10, *eta_pt10;
@@ -157,6 +201,10 @@ void GeneralPurposeVertexAnalyzer::IPMonitoring::bookIPMonitor(const edm::Parame
   double EtaMin = config.getParameter<double>("EtaMin");
   double EtaMax = config.getParameter<double>("EtaMax");
 
+  int PtBin = config.getParameter<int>("PtBin");
+  double PtMin = config.getParameter<double>("PtMin") * pTcut_;
+  double PtMax = config.getParameter<double>("PtMax") * pTcut_;
+
   IP_ = fs->make<TH1D>(fmt::format("d{}_pt{}", varname_, pTcut_).c_str(),
                        fmt::format("PV tracks (p_{{T}} > {} GeV) d_{{{}}} (#mum)", pTcut_, varname_).c_str(),
                        VarBin,
@@ -169,6 +217,13 @@ void GeneralPurposeVertexAnalyzer::IPMonitoring::bookIPMonitor(const edm::Parame
                           0.,
                           (varname_.find("xy") != std::string::npos) ? 2000. : 10000.);
 
+  IPPull_ = fs->make<TH1D>(
+      fmt::format("d{}Pull_pt{}", varname_, pTcut_).c_str(),
+      fmt::format("PV tracks (p_{{T}} > {} GeV) d_{{{}}}/#sigma_{{d_{{{}}}}}", pTcut_, varname_, varname_).c_str(),
+      100,
+      -5.,
+      5.);
+
   IPVsPhi_ =
       fs->make<TProfile>(fmt::format("d{}VsPhi_pt{}", varname_, pTcut_).c_str(),
                          fmt::format("PV tracks (p_{{T}} > {}) d_{{{}}} VS track #phi", pTcut_, varname_).c_str(),
@@ -193,6 +248,21 @@ void GeneralPurposeVertexAnalyzer::IPMonitoring::bookIPMonitor(const edm::Parame
   IPVsEta_->SetXTitle("PV track (p_{T} > 1 GeV) #eta");
   IPVsEta_->SetYTitle(fmt::format("PV tracks (p_{{T}} > {} GeV) d_{{{}}} (#mum)", pTcut_, varname_).c_str());
 
+  IPVsPt_ = gpVertexAnalyzer::makeProfileIfLog(
+      fs,
+      true,  /* x-axis */
+      false, /* y-axis */
+      fmt::format("d{}VsPt_pt{}", varname_, pTcut_).c_str(),
+      fmt::format("PV tracks (p_{{T}} > {}) d_{{{}}} VS track p_{{T}}", pTcut_, varname_).c_str(),
+      PtBin,
+      log10(PtMin),
+      log10(PtMax),
+      VarMin,
+      VarMax,
+      "");
+  IPVsPt_->SetXTitle("PV track (p_{T} > 1 GeV) p_{T} [GeV]");
+  IPVsPt_->SetYTitle(fmt::format("PV tracks (p_{{T}} > {} GeV) d_{{{}}} (#mum)", pTcut_, varname_).c_str());
+
   IPErrVsPhi_ =
       fs->make<TProfile>(fmt::format("d{}ErrVsPhi_pt{}", varname_, pTcut_).c_str(),
                          fmt::format("PV tracks (p_{{T}} > {}) d_{{{}}} error VS track #phi", pTcut_, varname_).c_str(),
@@ -217,6 +287,21 @@ void GeneralPurposeVertexAnalyzer::IPMonitoring::bookIPMonitor(const edm::Parame
   IPErrVsEta_->SetXTitle("PV track (p_{T} > 1 GeV) #eta");
   IPErrVsEta_->SetYTitle(fmt::format("PV tracks (p_{{T}} > {} GeV) d_{{{}}} error (#mum)", pTcut_, varname_).c_str());
 
+  IPErrVsPt_ = gpVertexAnalyzer::makeProfileIfLog(
+      fs,
+      true,  /* x-axis */
+      false, /* y-axis */
+      fmt::format("d{}ErrVsPt_pt{}", varname_, pTcut_).c_str(),
+      fmt::format("PV tracks (p_{{T}} > {}) d_{{{}}} error VS track p_{{T}}", pTcut_, varname_).c_str(),
+      PtBin,
+      log10(PtMin),
+      log10(PtMax),
+      VarMin,
+      VarMax,
+      "");
+  IPErrVsPt_->SetXTitle("PV track (p_{T} > 1 GeV) p_{T} [GeV]");
+  IPErrVsPt_->SetYTitle(fmt::format("PV tracks (p_{{T}} > {} GeV) d_{{{}}} error (#mum)", pTcut_, varname_).c_str());
+
   IPVsEtaVsPhi_ = fs->make<TProfile2D>(
       fmt::format("d{}VsEtaVsPhi_pt{}", varname_, pTcut_).c_str(),
       fmt::format("PV tracks (p_{{T}} > {}) d_{{{}}} VS track #eta VS track #phi", pTcut_, varname_).c_str(),
@@ -294,6 +379,7 @@ GeneralPurposeVertexAnalyzer::GeneralPurposeVertexAnalyzer(const edm::ParameterS
       bsBeamWidthX(nullptr),
       bsBeamWidthY(nullptr),
       bsType(nullptr),
+      trackpt(nullptr),
       sumpt(nullptr),
       ntracks(nullptr),
       weight(nullptr),
@@ -411,6 +497,8 @@ void GeneralPurposeVertexAnalyzer::pvTracksPlots(const reco::Vertex &v) {
     }
 
     const float pt = t->pt();
+    trackpt->Fill(pt);
+
     if (pt < 1.f) {
       continue;
     }
@@ -435,22 +523,28 @@ void GeneralPurposeVertexAnalyzer::pvTracksPlots(const reco::Vertex &v) {
     phi_pt1->Fill(phi);
     eta_pt1->Fill(eta);
 
+    // dxy pT>1
+
     dxy_pt1.IP_->Fill(Dxy);
     dxy_pt1.IPVsPhi_->Fill(phi, Dxy);
     dxy_pt1.IPVsEta_->Fill(eta, Dxy);
     dxy_pt1.IPVsEtaVsPhi_->Fill(eta, phi, Dxy);
 
     dxy_pt1.IPErr_->Fill(DxyErr);
+    dxy_pt1.IPPull_->Fill(Dxy / DxyErr);
     dxy_pt1.IPErrVsPhi_->Fill(phi, DxyErr);
     dxy_pt1.IPErrVsEta_->Fill(eta, DxyErr);
     dxy_pt1.IPErrVsEtaVsPhi_->Fill(eta, phi, DxyErr);
 
+    // dz pT>1
+
     dz_pt1.IP_->Fill(Dz);
     dz_pt1.IPVsPhi_->Fill(phi, Dz);
     dz_pt1.IPVsEta_->Fill(eta, Dz);
     dz_pt1.IPVsEtaVsPhi_->Fill(eta, phi, Dz);
 
     dz_pt1.IPErr_->Fill(DzErr);
+    dz_pt1.IPPull_->Fill(Dz / DzErr);
     dz_pt1.IPErrVsPhi_->Fill(phi, DzErr);
     dz_pt1.IPErrVsEta_->Fill(eta, DzErr);
     dz_pt1.IPErrVsEtaVsPhi_->Fill(eta, phi, DzErr);
@@ -459,22 +553,28 @@ void GeneralPurposeVertexAnalyzer::pvTracksPlots(const reco::Vertex &v) {
       phi_pt10->Fill(phi);
       eta_pt10->Fill(eta);
 
+      // dxy pT>10
+
       dxy_pt10.IP_->Fill(Dxy);
       dxy_pt10.IPVsPhi_->Fill(phi, Dxy);
       dxy_pt10.IPVsEta_->Fill(eta, Dxy);
       dxy_pt10.IPVsEtaVsPhi_->Fill(eta, phi, Dxy);
 
       dxy_pt10.IPErr_->Fill(DxyErr);
+      dxy_pt10.IPPull_->Fill(Dxy / DxyErr);
       dxy_pt10.IPErrVsPhi_->Fill(phi, DxyErr);
       dxy_pt10.IPErrVsEta_->Fill(eta, DxyErr);
       dxy_pt10.IPErrVsEtaVsPhi_->Fill(eta, phi, DxyErr);
 
+      // dz pT>10
+
       dz_pt10.IP_->Fill(Dz);
       dz_pt10.IPVsPhi_->Fill(phi, Dz);
       dz_pt10.IPVsEta_->Fill(eta, Dz);
       dz_pt10.IPVsEtaVsPhi_->Fill(eta, phi, Dz);
 
       dz_pt10.IPErr_->Fill(DzErr);
+      dz_pt10.IPPull_->Fill(Dz / DzErr);
       dz_pt10.IPErrVsPhi_->Fill(phi, DzErr);
       dz_pt10.IPErrVsEta_->Fill(eta, DzErr);
       dz_pt10.IPErrVsEtaVsPhi_->Fill(eta, phi, DzErr);
@@ -649,6 +749,8 @@ void GeneralPurposeVertexAnalyzer::beginJob() {
 
   dxy2 = book<TH1D>("dxyzoom", fmt::sprintf("%s d_{xy} (#mum)", s_1).c_str(), dxyBin_, dxyMin_ / 5., dxyMax_ / 5.);
 
+  trackpt = gpVertexAnalyzer::makeTH1IfLog(
+      fs_, true, false, "pt_track", "PV tracks p_{T};PV tracks p_{T} [GeV];#tracks", 49, log10(1.), log10(50.));
   phi_pt1 =
       book<TH1D>("phi_pt1", fmt::sprintf("%s #phi; PV tracks #phi;#tracks", s_1).c_str(), phiBin_, phiMin_, phiMax_);
   eta_pt1 =
@@ -701,6 +803,9 @@ void GeneralPurposeVertexAnalyzer::fillDescriptions(edm::ConfigurationDescriptio
   desc.add<int>("EtaBin2D", 8);
   desc.add<double>("EtaMin", -2.7);
   desc.add<double>("EtaMax", 2.7);
+  desc.add<int>("PtBin", 49);
+  desc.add<double>("PtMin", 1.);
+  desc.add<double>("PtMax", 50.);
   descriptions.addWithDefaultLabel(desc);
 }