CMS 3D CMS Logo

SiPixelCompareTrackSoA.cc
Go to the documentation of this file.
1 // -*- C++ -*-
2 // Package: SiPixelCompareTrackSoA
3 // Class: SiPixelCompareTrackSoA
4 //
7 //
8 // Author: Suvankar Roy Chowdhury
9 //
19 // DQM Histograming
25 // for string manipulations
26 #include <fmt/printf.h>
27 
28 namespace {
29  // same logic used for the MTV:
30  // cf https://github.com/cms-sw/cmssw/blob/master/Validation/RecoTrack/src/MTVHistoProducerAlgoForTracker.cc
32 
33  void setBinLog(TAxis* axis) {
34  int bins = axis->GetNbins();
35  float from = axis->GetXmin();
36  float to = axis->GetXmax();
37  float width = (to - from) / bins;
38  std::vector<float> new_bins(bins + 1, 0);
39  for (int i = 0; i <= bins; i++) {
40  new_bins[i] = TMath::Power(10, from + i * width);
41  }
42  axis->Set(bins, new_bins.data());
43  }
44 
45  void setBinLogX(TH1* h) {
46  TAxis* axis = h->GetXaxis();
47  setBinLog(axis);
48  }
49  void setBinLogY(TH1* h) {
50  TAxis* axis = h->GetYaxis();
51  setBinLog(axis);
52  }
53 
54  template <typename... Args>
55  dqm::reco::MonitorElement* make2DIfLog(DQMStore::IBooker& ibook, bool logx, bool logy, Args&&... args) {
56  auto h = std::make_unique<TH2I>(std::forward<Args>(args)...);
57  if (logx)
58  setBinLogX(h.get());
59  if (logy)
60  setBinLogY(h.get());
61  const auto& name = h->GetName();
62  return ibook.book2I(name, h.release());
63  }
64 } // namespace
65 
66 template <typename T>
68 public:
70 
72  ~SiPixelCompareTrackSoA() override = default;
73  void bookHistograms(DQMStore::IBooker& ibooker, edm::Run const& iRun, edm::EventSetup const& iSetup) override;
74  void analyze(const edm::Event& iEvent, const edm::EventSetup& iSetup) override;
75  static void fillDescriptions(edm::ConfigurationDescriptions& descriptions);
76 
77 private:
81  const bool useQualityCut_;
83  const float dr2cut_;
104  //1D differences
111 
112  //for matching eff vs region: derive the ratio at harvesting
117 };
118 
119 //
120 // constructors
121 //
122 
123 template <typename T>
125  : tokenSoATrackCPU_(consumes<PixelTrackSoA>(iConfig.getParameter<edm::InputTag>("pixelTrackSrcCPU"))),
126  tokenSoATrackGPU_(consumes<PixelTrackSoA>(iConfig.getParameter<edm::InputTag>("pixelTrackSrcGPU"))),
127  topFolderName_(iConfig.getParameter<std::string>("topFolderName")),
128  useQualityCut_(iConfig.getParameter<bool>("useQualityCut")),
129  minQuality_(pixelTrack::qualityByName(iConfig.getParameter<std::string>("minQuality"))),
130  dr2cut_(iConfig.getParameter<double>("deltaR2cut")) {}
131 
132 //
133 // -- Analyze
134 //
135 template <typename T>
137  using helper = TracksUtilities<T>;
138  const auto& tsoaHandleCPU = iEvent.getHandle(tokenSoATrackCPU_);
139  const auto& tsoaHandleGPU = iEvent.getHandle(tokenSoATrackGPU_);
140  if (not tsoaHandleCPU or not tsoaHandleGPU) {
141  edm::LogWarning out("SiPixelCompareTrackSoA");
142  if (not tsoaHandleCPU) {
143  out << "reference (cpu) tracks not found; ";
144  }
145  if (not tsoaHandleGPU) {
146  out << "target (gpu) tracks not found; ";
147  }
148  out << "the comparison will not run.";
149  return;
150  }
151 
152  auto const& tsoaCPU = *tsoaHandleCPU;
153  auto const& tsoaGPU = *tsoaHandleGPU;
154  auto maxTracksCPU = tsoaCPU.view().metadata().size(); //this should be same for both?
155  auto maxTracksGPU = tsoaGPU.view().metadata().size(); //this should be same for both?
156  auto const* qualityCPU = tsoaCPU.view().quality();
157  auto const* qualityGPU = tsoaGPU.view().quality();
158  int32_t nTracksCPU = 0;
159  int32_t nTracksGPU = 0;
160  int32_t nLooseAndAboveTracksCPU = 0;
161  int32_t nLooseAndAboveTracksCPU_matchedGPU = 0;
162  int32_t nLooseAndAboveTracksGPU = 0;
163 
164  //Loop over GPU tracks and store the indices of the loose tracks. Whats happens if useQualityCut_ is false?
165  std::vector<int32_t> looseTrkidxGPU;
166  for (int32_t jt = 0; jt < maxTracksGPU; ++jt) {
167  if (helper::nHits(tsoaGPU.view(), jt) == 0)
168  break; // this is a guard
169  if (!(tsoaGPU.view()[jt].pt() > 0.))
170  continue;
171  nTracksGPU++;
172  if (useQualityCut_ && qualityGPU[jt] < minQuality_)
173  continue;
174  nLooseAndAboveTracksGPU++;
175  looseTrkidxGPU.emplace_back(jt);
176  }
177 
178  //Now loop over CPU tracks//nested loop for loose gPU tracks
179  for (int32_t it = 0; it < maxTracksCPU; ++it) {
180  int nHitsCPU = helper::nHits(tsoaCPU.view(), it);
181 
182  if (nHitsCPU == 0)
183  break; // this is a guard
184 
185  float ptCPU = tsoaCPU.view()[it].pt();
186  float etaCPU = tsoaCPU.view()[it].eta();
187  float phiCPU = helper::phi(tsoaCPU.view(), it);
188  float zipCPU = helper::zip(tsoaCPU.view(), it);
189  float tipCPU = helper::tip(tsoaCPU.view(), it);
190 
191  if (!(ptCPU > 0.))
192  continue;
193  nTracksCPU++;
194  if (useQualityCut_ && qualityCPU[it] < minQuality_)
195  continue;
196  nLooseAndAboveTracksCPU++;
197  //Now loop over loose GPU trk and find the closest in DeltaR//do we need pt cut?
198  const int32_t notFound = -1;
199  int32_t closestTkidx = notFound;
200  float mindr2 = dr2cut_;
201 
202  for (auto gid : looseTrkidxGPU) {
203  float etaGPU = tsoaGPU.view()[gid].eta();
204  float phiGPU = helper::phi(tsoaGPU.view(), gid);
205  float dr2 = reco::deltaR2(etaCPU, phiCPU, etaGPU, phiGPU);
206  if (dr2 > dr2cut_)
207  continue; // this is arbitrary
208  if (mindr2 > dr2) {
209  mindr2 = dr2;
210  closestTkidx = gid;
211  }
212  }
213 
214  hpt_eta_tkAllCPU_->Fill(etaCPU, ptCPU); //all CPU tk
215  hphi_z_tkAllCPU_->Fill(phiCPU, zipCPU);
216  if (closestTkidx == notFound)
217  continue;
218  nLooseAndAboveTracksCPU_matchedGPU++;
219 
220  hchi2_->Fill(tsoaCPU.view()[it].chi2(), tsoaGPU.view()[closestTkidx].chi2());
221  hCharge_->Fill(helper::charge(tsoaCPU.view(), it), helper::charge(tsoaGPU.view(), closestTkidx));
222  hnHits_->Fill(helper::nHits(tsoaCPU.view(), it), helper::nHits(tsoaGPU.view(), closestTkidx));
223  hnLayers_->Fill(tsoaCPU.view()[it].nLayers(), tsoaGPU.view()[closestTkidx].nLayers());
224  hpt_->Fill(ptCPU, tsoaGPU.view()[closestTkidx].pt());
225  hptLogLog_->Fill(ptCPU, tsoaGPU.view()[closestTkidx].pt());
226  heta_->Fill(etaCPU, tsoaGPU.view()[closestTkidx].eta());
227  hphi_->Fill(phiCPU, helper::phi(tsoaGPU.view(), closestTkidx));
228  hz_->Fill(zipCPU, helper::zip(tsoaGPU.view(), closestTkidx));
229  htip_->Fill(tipCPU, helper::tip(tsoaGPU.view(), closestTkidx));
230  hptdiffMatched_->Fill(ptCPU - tsoaGPU.view()[closestTkidx].pt());
231  hCurvdiffMatched_->Fill((helper::charge(tsoaCPU.view(), it) / tsoaCPU.view()[it].pt()) -
232  (helper::charge(tsoaGPU.view(), closestTkidx) / tsoaGPU.view()[closestTkidx].pt()));
233  hetadiffMatched_->Fill(etaCPU - tsoaGPU.view()[closestTkidx].eta());
234  hphidiffMatched_->Fill(reco::deltaPhi(phiCPU, helper::phi(tsoaGPU.view(), closestTkidx)));
235  hzdiffMatched_->Fill(zipCPU - helper::zip(tsoaGPU.view(), closestTkidx));
236  htipdiffMatched_->Fill(tipCPU - helper::tip(tsoaGPU.view(), closestTkidx));
237  hpt_eta_tkAllCPUMatched_->Fill(etaCPU, tsoaCPU.view()[it].pt()); //matched to gpu
238  hphi_z_tkAllCPUMatched_->Fill(etaCPU, zipCPU);
239  }
240  hnTracks_->Fill(nTracksCPU, nTracksGPU);
241  hnLooseAndAboveTracks_->Fill(nLooseAndAboveTracksCPU, nLooseAndAboveTracksGPU);
242  hnLooseAndAboveTracks_matched_->Fill(nLooseAndAboveTracksCPU, nLooseAndAboveTracksCPU_matchedGPU);
243 }
244 
245 //
246 // -- Book Histograms
247 //
248 template <typename T>
250  edm::Run const& iRun,
251  edm::EventSetup const& iSetup) {
252  iBook.cd();
253  iBook.setCurrentFolder(topFolderName_);
254 
255  // clang-format off
256  std::string toRep = "Number of tracks";
257  // FIXME: all the 2D correlation plots are quite heavy in terms of memory consumption, so a as soon as DQM supports THnSparse
258  // these should be moved to a less resource consuming format
259  hnTracks_ = iBook.book2I("nTracks", fmt::sprintf("%s per event; CPU; GPU",toRep), 501, -0.5, 500.5, 501, -0.5, 500.5);
260  hnLooseAndAboveTracks_ = iBook.book2I("nLooseAndAboveTracks", fmt::sprintf("%s (quality #geq loose) per event; CPU; GPU",toRep), 501, -0.5, 500.5, 501, -0.5, 500.5);
261  hnLooseAndAboveTracks_matched_ = iBook.book2I("nLooseAndAboveTracks_matched", fmt::sprintf("%s (quality #geq loose) per event; CPU; GPU",toRep), 501, -0.5, 500.5, 501, -0.5, 500.5);
262 
263  toRep = "Number of all RecHits per track (quality #geq loose)";
264  hnHits_ = iBook.book2I("nRecHits", fmt::sprintf("%s;CPU;GPU",toRep), 15, -0.5, 14.5, 15, -0.5, 14.5);
265 
266  toRep = "Number of all layers per track (quality #geq loose)";
267  hnLayers_ = iBook.book2I("nLayers", fmt::sprintf("%s;CPU;GPU",toRep), 15, -0.5, 14.5, 15, -0.5, 14.5);
268 
269  toRep = "Track (quality #geq loose) #chi^{2}/ndof";
270  hchi2_ = iBook.book2I("nChi2ndof", fmt::sprintf("%s;CPU;GPU",toRep), 40, 0., 20., 40, 0., 20.);
271 
272  toRep = "Track (quality #geq loose) charge";
273  hCharge_ = iBook.book2I("charge",fmt::sprintf("%s;CPU;GPU",toRep),3, -1.5, 1.5, 3, -1.5, 1.5);
274 
275  hpt_ = iBook.book2I("pt", "Track (quality #geq loose) p_{T} [GeV];CPU;GPU", 200, 0., 200., 200, 0., 200.);
276  hptLogLog_ = make2DIfLog(iBook, true, true, "ptLogLog", "Track (quality #geq loose) p_{T} [GeV];CPU;GPU", 200, log10(0.5), log10(200.), 200, log10(0.5), log10(200.));
277  heta_ = iBook.book2I("eta", "Track (quality #geq loose) #eta;CPU;GPU", 30, -3., 3., 30, -3., 3.);
278  hphi_ = iBook.book2I("phi", "Track (quality #geq loose) #phi;CPU;GPU", 30, -M_PI, M_PI, 30, -M_PI, M_PI);
279  hz_ = iBook.book2I("z", "Track (quality #geq loose) z [cm];CPU;GPU", 30, -30., 30., 30, -30., 30.);
280  htip_ = iBook.book2I("tip", "Track (quality #geq loose) TIP [cm];CPU;GPU", 100, -0.5, 0.5, 100, -0.5, 0.5);
281  //1D difference plots
282  hptdiffMatched_ = iBook.book1D("ptdiffmatched", " p_{T} diff [GeV] between matched tracks; #Delta p_{T} [GeV]", 60, -30., 30.);
283  hCurvdiffMatched_ = iBook.book1D("curvdiffmatched", "q/p_{T} diff [GeV] between matched tracks; #Delta q/p_{T} [GeV]", 60, -30., 30.);
284  hetadiffMatched_ = iBook.book1D("etadiffmatched", " #eta diff between matched tracks; #Delta #eta", 160, -0.04 ,0.04);
285  hphidiffMatched_ = iBook.book1D("phidiffmatched", " #phi diff between matched tracks; #Delta #phi", 160, -0.04 ,0.04);
286  hzdiffMatched_ = iBook.book1D("zdiffmatched", " z diff between matched tracks; #Delta z [cm]", 300, -1.5, 1.5);
287  htipdiffMatched_ = iBook.book1D("tipdiffmatched", " TIP diff between matched tracks; #Delta TIP [cm]", 300, -1.5, 1.5);
288  //2D plots for eff
289  hpt_eta_tkAllCPU_ = iBook.book2I("ptetatrkAllCPU", "Track (quality #geq loose) on CPU; #eta; p_{T} [GeV];", 30, -M_PI, M_PI, 200, 0., 200.);
290  hpt_eta_tkAllCPUMatched_ = iBook.book2I("ptetatrkAllCPUmatched", "Track (quality #geq loose) on CPU matched to GPU track; #eta; p_{T} [GeV];", 30, -M_PI, M_PI, 200, 0., 200.);
291 
292  hphi_z_tkAllCPU_ = iBook.book2I("phiztrkAllCPU", "Track (quality #geq loose) on CPU; #phi; z [cm];", 30, -M_PI, M_PI, 30, -30., 30.);
293  hphi_z_tkAllCPUMatched_ = iBook.book2I("phiztrkAllCPUmatched", "Track (quality #geq loose) on CPU; #phi; z [cm];", 30, -M_PI, M_PI, 30, -30., 30.);
294 
295 }
296 
297 template<typename T>
299  // monitorpixelTrackSoA
301  desc.add<edm::InputTag>("pixelTrackSrcCPU", edm::InputTag("pixelTracksSoA@cpu"));
302  desc.add<edm::InputTag>("pixelTrackSrcGPU", edm::InputTag("pixelTracksSoA@cuda"));
303  desc.add<std::string>("topFolderName", "SiPixelHeterogeneous/PixelTrackCompareGPUvsCPU");
304  desc.add<bool>("useQualityCut", true);
305  desc.add<std::string>("minQuality", "loose");
306  desc.add<double>("deltaR2cut", 0.04);
307  descriptions.addWithDefaultLabel(desc);
308 }
309 
313 
constexpr double deltaPhi(double phi1, double phi2)
Definition: deltaPhi.h:26
void addWithDefaultLabel(ParameterSetDescription const &psetDescription)
Definition: helper.py:1
MonitorElement * hphi_z_tkAllCPUMatched_
SiPixelCompareTrackSoA(const edm::ParameterSet &)
virtual void setCurrentFolder(std::string const &fullpath)
Definition: DQMStore.cc:36
Quality qualityByName(std::string const &name)
const edm::EDGetTokenT< PixelTrackSoA > tokenSoATrackGPU_
ALPAKA_FN_HOST_ACC ALPAKA_FN_INLINE constexpr float zip(ConstView const &tracks, int32_t i)
Definition: TracksSoA.h:90
void analyze(const edm::Event &iEvent, const edm::EventSetup &iSetup) override
T eta() const
Definition: PV3DBase.h:73
const std::string topFolderName_
MonitorElement * hnLooseAndAboveTracks_matched_
void bookHistograms(DQMStore::IBooker &ibooker, edm::Run const &iRun, edm::EventSetup const &iSetup) override
dqm::reco::DQMStore DQMStore
int iEvent
Definition: GenABIO.cc:224
The Signals That Services Can Subscribe To This is based on ActivityRegistry and is current per Services can connect to the signals distributed by the ActivityRegistry in order to monitor the activity of the application Each possible callback has some defined which we here list in angle e< void, edm::EventID const &, edm::Timestamp const & > We also list in braces which AR_WATCH_USING_METHOD_ is used for those or
Definition: Activities.doc:12
#define DEFINE_FWK_MODULE(type)
Definition: MakerMacros.h:16
#define M_PI
MonitorElement * hpt_eta_tkAllCPUMatched_
constexpr auto deltaR2(const T1 &t1, const T2 &t2) -> decltype(t1.eta())
Definition: deltaR.h:16
static void fillDescriptions(edm::ConfigurationDescriptions &descriptions)
static const GlobalPoint notFound(0, 0, 0)
const pixelTrack::Quality minQuality_
HLT enums.
MonitorElement * hnLooseAndAboveTracks_
MonitorElement * book1D(TString const &name, TString const &title, int const nchX, double const lowX, double const highX, FUNC onbooking=NOOP())
Definition: DQMStore.h:98
TupleMultiplicity< TrackerTraits > const *__restrict__ uint32_t nHits
The Signals That Services Can Subscribe To This is based on ActivityRegistry h
Helper function to determine trigger accepts.
Definition: Activities.doc:4
MonitorElement * book2I(TString const &name, TString const &title, int nchX, double lowX, double highX, int nchY, double lowY, double highY, FUNC onbooking=NOOP())
Definition: DQMStore.h:305
~SiPixelCompareTrackSoA() override=default
const edm::EDGetTokenT< PixelTrackSoA > tokenSoATrackCPU_
Definition: Run.h:45