]>
Commit | Line | Data |
---|---|---|
9e4cc50d | 1 | |
7c679be0 | 2 | // ************************************************************************** |
3 | // * Copyright(c) 1998-1999, ALICE Experiment at CERN, All rights reserved. * | |
4 | // * * | |
5 | // * Author: The ALICE Off-line Project. * | |
6 | // * Contributors are mentioned in the code where appropriate. * | |
7 | // * * | |
8 | // * Permission to use, copy, modify and distribute this software and its * | |
9 | // * documentation strictly for non-commercial purposes is hereby granted * | |
10 | // * without fee, provided that the above copyright notice appears in all * | |
11 | // * copies and that both the copyright notice and this permission notice * | |
12 | // * appear in the supporting documentation. The authors make no claims * | |
13 | // * about the suitability of this software for any purpose. It is * | |
14 | // * provided "as is" without express or implied warranty. * | |
15 | // ************************************************************************** | |
16 | ||
139cbd96 | 17 | /* $Id$ */ |
18 | ||
7c679be0 | 19 | //----------------------------------------------------------------------------------- |
20 | // Jet finder based on Deterministic Annealing | |
21 | // For further informations about the DA working features see: | |
22 | // Phys.Lett. B601 (2004) 56-63 (http://arxiv.org/abs/hep-ph/0407214) | |
23 | // Author: Davide Perrino (davide.perrino@ba.infn.it, davide.perrino@cern.ch) | |
139cbd96 | 24 | // alexandre.shabetai@cern.ch (Modification of the input object (reader/finder splitting)) |
25 | // ** 2011 | |
26 | // Modified accordingly to reader/finder splitting and new handling of neutral information | |
7c679be0 | 27 | //----------------------------------------------------------------------------------- |
28 | ||
29 | #include <TMath.h> | |
386b2e2f | 30 | #include <TRandom2.h> |
139cbd96 | 31 | |
32 | #include "AliAODJet.h" | |
7c679be0 | 33 | #include "AliDAJetHeader.h" |
34 | #include "AliDAJetFinder.h" | |
35 | ||
7c679be0 | 36 | ClassImp(AliDAJetFinder) |
37 | ||
139cbd96 | 38 | /////////////////////////////////////////////////////////////////////// |
7c679be0 | 39 | |
7c679be0 | 40 | AliDAJetFinder::AliDAJetFinder(): |
139cbd96 | 41 | AliJetFinder(), |
42 | fAlpha(1.01), | |
43 | fDelta(1e-8), | |
44 | fAvDist(1e-6), | |
45 | fEps(1e-4), | |
46 | fEpsMax(1e-2), | |
47 | fNloopMax(100), | |
48 | fBeta(0.1), | |
49 | fNclustMax(0), | |
50 | fNin(0), | |
51 | fNeff(0) | |
7c679be0 | 52 | { |
139cbd96 | 53 | // Constructor |
7c679be0 | 54 | } |
55 | ||
56 | //----------------------------------------------------------------------------------- | |
57 | AliDAJetFinder::~AliDAJetFinder() | |
58 | { | |
139cbd96 | 59 | // Destructor |
7c679be0 | 60 | } |
61 | ||
62 | //----------------------------------------------------------------------------------- | |
63 | void AliDAJetFinder::FindJets() | |
64 | { | |
139cbd96 | 65 | // Find the jets in current event |
66 | // | |
67 | Float_t betaStop=100.; | |
68 | fDebug = fHeader->GetDebug(); | |
69 | ||
70 | Double_t dEtSum=0; | |
71 | Double_t *xData[2]; | |
72 | TVectorD *vPx = new TVectorD(); | |
73 | TVectorD *vPy = new TVectorD(); | |
74 | TMatrixD *mPyx= new TMatrixD(); | |
75 | TMatrixD *mY = new TMatrixD(); | |
76 | InitDetAnn(dEtSum,xData,vPx,vPy,mPyx,mY); | |
77 | if (fNin < fNclustMax){ | |
78 | delete [] xData[0], delete [] xData[1]; | |
79 | delete vPx; | |
80 | delete vPy; | |
81 | delete mPyx; | |
82 | delete mY; | |
83 | return; | |
84 | } | |
85 | Int_t nc=1, nk; | |
86 | DoubleClusters(nc,nk,vPy,mY); | |
87 | do{ //loop over beta | |
88 | fBeta*=fAlpha; | |
89 | Annealing(nk,xData,vPx,vPy,mPyx,mY); | |
90 | NumCl(nc,nk,vPy,mPyx,mY); | |
91 | }while((fBeta<betaStop || nc<4) && nc<fNclustMax); | |
7c679be0 | 92 | |
139cbd96 | 93 | Int_t *xx=new Int_t[fNeff]; |
94 | for (Int_t i = 0; i < fNeff; i++) xx[i] = 0; | |
95 | ||
96 | EndDetAnn(nk,xData,xx,dEtSum,vPx,vPy,mPyx,mY); | |
97 | StoreJets(nk,xData,xx,mY); | |
98 | delete [] xx; | |
99 | ||
100 | delete [] xData[0], delete [] xData[1]; | |
101 | delete mPyx; | |
102 | delete mY; | |
103 | delete vPx; | |
104 | delete vPy; | |
36b5cc43 | 105 | |
7c679be0 | 106 | } |
107 | ||
108 | //----------------------------------------------------------------------------------- | |
36b5cc43 | 109 | void AliDAJetFinder::InitDetAnn(Double_t &dEtSum,Double_t **xData,TVectorD *vPx,TVectorD *vPy,TMatrixD *mPyx,TMatrixD *mY) |
7c679be0 | 110 | { |
139cbd96 | 111 | //Initialise the variables used by the algorithm |
112 | fBeta=0.1; | |
113 | fNclustMax = ((AliDAJetHeader*)fHeader)->GetFixedCl() ? | |
114 | ((AliDAJetHeader*)fHeader)->GetNclustMax() : | |
115 | TMath::Max((Int_t)TMath::Sqrt(fNin),5); | |
116 | Float_t etaEff = ((AliDAJetHeader*)fHeader)->GetEtaEff(); | |
117 | ||
118 | fNin=0; | |
119 | for (Int_t iTr=0; iTr<GetCalTrkEvent()->GetNCalTrkTracks(); iTr++) if (GetCalTrkEvent()->GetCalTrkTrack(iTr)->GetCutFlag()==1) fNin++; | |
120 | ||
121 | fNeff = ((AliDAJetHeader*)fHeader)->GetNeff(); | |
122 | fNeff = TMath::Max(fNeff,fNin); | |
123 | Double_t *xEta = new Double_t[fNeff]; | |
124 | Double_t *xPhi = new Double_t[fNeff]; | |
125 | xData[0]=xEta; xData[1]=xPhi; | |
126 | vPx->ResizeTo(fNeff); | |
127 | Int_t iIn=0; | |
128 | ||
129 | for (Int_t iTr=0; iTr<GetCalTrkEvent()->GetNCalTrkTracks(); iTr++){ | |
130 | AliJetCalTrkTrack* ctT = GetCalTrkEvent()->GetCalTrkTrack(iTr); | |
131 | if (ctT->GetCutFlag()==0) continue; | |
132 | xEta[iIn] = ctT->GetEta(); | |
133 | xPhi[iIn] = ctT->GetPhi()<0 ? ctT->GetPhi() + 2*TMath::Pi() : ctT->GetPhi(); | |
134 | (*vPx)(iIn)=ctT->GetPt(); | |
135 | dEtSum+=(*vPx)(iIn); | |
136 | iIn++; | |
137 | } | |
138 | ||
139 | TRandom2 r; | |
140 | r.SetSeed(0); | |
141 | for (iIn=fNin; iIn<fNeff; iIn++){ | |
142 | xEta[iIn]=r.Uniform(-1*etaEff,etaEff); | |
143 | xPhi[iIn]=r.Uniform(0.,2*TMath::Pi()); | |
144 | (*vPx)(iIn)=r.Uniform(0.01,0.02); | |
145 | dEtSum+=(*vPx)(iIn); | |
146 | } | |
147 | for (iIn=0; iIn<fNeff; iIn++) (*vPx)(iIn)=(*vPx)(iIn)/dEtSum; | |
148 | ||
149 | Int_t njdim=2*fNclustMax+1; | |
150 | mPyx->ResizeTo(fNeff,njdim); | |
151 | mY->ResizeTo(4,njdim); | |
152 | vPy->ResizeTo(njdim); | |
153 | mY->Zero();mPyx->Zero();vPy->Zero(); | |
154 | (*vPy)(0)=1; | |
155 | TMatrixDColumn(*mPyx,0)=1; | |
156 | Double_t ypos=0,xpos=0; | |
157 | for (iIn=0; iIn<fNeff; iIn++){ | |
158 | (*mY)(0,0)+=(*vPx)(iIn)*xEta[iIn]; | |
159 | ypos+=(*vPx)(iIn)*TMath::Sin(xPhi[iIn]); | |
160 | xpos+=(*vPx)(iIn)*TMath::Cos(xPhi[iIn]); | |
161 | } | |
162 | (*mY)(1,0)=(atan2(ypos,xpos)>0) ? atan2(ypos,xpos) : atan2(ypos,xpos)+2*TMath::Pi(); | |
163 | ||
7c679be0 | 164 | } |
165 | ||
166 | //----------------------------------------------------------------------------------- | |
139cbd96 | 167 | void AliDAJetFinder::DoubleClusters(Int_t nc,Int_t &nk, TVectorD *vPy, TMatrixD *mY) const |
7c679be0 | 168 | { |
139cbd96 | 169 | // Return double clusters |
170 | for(Int_t iClust=0; iClust<nc; iClust++){ | |
171 | (*vPy)(iClust)=(*vPy)(iClust)/2; | |
172 | (*vPy)(nc+iClust)=(*vPy)(iClust); | |
173 | for(Int_t iComp=0; iComp<3; iComp++) (*mY)(iComp,nc+iClust)=(*mY)(iComp,iClust); | |
174 | } | |
175 | nk=2*nc; | |
176 | ||
7c679be0 | 177 | } |
178 | ||
179 | //----------------------------------------------------------------------------------- | |
139cbd96 | 180 | void AliDAJetFinder::Annealing(Int_t nk,Double_t **xData, const TVectorD *vPx, TVectorD *vPy, TMatrixD *mPyx, TMatrixD *mY) |
7c679be0 | 181 | { |
139cbd96 | 182 | // Main part of the algorithm |
183 | const Double_t pi=TMath::Pi(); | |
184 | TVectorD *py = new TVectorD(nk); | |
185 | TVectorD *p = new TVectorD(nk); | |
186 | TMatrixD *y = new TMatrixD(4,nk); | |
187 | TMatrixD *y1 = new TMatrixD(4,nk); | |
188 | TMatrixD *ry = new TMatrixD(2,nk); | |
189 | Double_t *xEta = xData[0]; | |
190 | Double_t *xPhi = xData[1]; | |
191 | Double_t Dist(TVectorD,TVectorD); | |
192 | ||
193 | Double_t df[2]={((AliDAJetHeader*)fHeader)->GetFiducialEtaMax(),pi}; | |
194 | TVectorD vPart(2); | |
195 | Double_t *m = new Double_t[nk]; | |
196 | Double_t chi,chi1; | |
197 | do{ | |
198 | Int_t nloop=0; | |
199 | for (Int_t iClust=0; iClust<nk; iClust++){ | |
200 | for (Int_t i=0; i<3; i++)(*y1)(i,iClust)=(*mY)(i,iClust); | |
201 | (*py)(iClust)=(*vPy)(iClust); | |
202 | } | |
203 | //perturbation of codevectors | |
204 | Double_t seed=1000000*gRandom->Rndm(24); | |
205 | ry->Randomize(-0.5,0.5,seed); | |
206 | for (Int_t i=0; i<2; i++){ | |
207 | for (Int_t iClust=0; iClust<nk/2; iClust++) | |
208 | (*y1)(i,iClust)+=((*ry)(i,iClust)+TMath::Sign(0.5,(*ry)(i,iClust)))*fDelta*df[i]; | |
209 | for (Int_t iClust=nk/2; iClust<nk; iClust++) | |
210 | (*y1)(i,iClust)-=((*ry)(i,iClust-nk/2)+TMath::Sign(0.5,(*ry)(i,iClust-nk/2)))*fDelta*df[i]; | |
211 | } | |
212 | do{ | |
213 | //recalculate conditional probabilities | |
214 | nloop++; | |
215 | for (Int_t iIn=0; iIn<fNeff; iIn++){ | |
216 | vPart(0)=xEta[iIn]; vPart(1)=xPhi[iIn]; | |
217 | for(Int_t iClust=0; iClust<nk; iClust++){ | |
218 | (*mPyx)(iIn,iClust)=-log((*py)(iClust))+fBeta*Dist(vPart,TMatrixDColumn(*y1,iClust)); | |
219 | m[iClust]=(*mPyx)(iIn,iClust); | |
7c679be0 | 220 | } |
139cbd96 | 221 | Double_t pyxNorm=0; |
222 | Double_t minPyx=TMath::MinElement(nk,m); | |
223 | for (Int_t iClust=0; iClust<nk; iClust++){ | |
224 | (*mPyx)(iIn,iClust)-=minPyx; | |
225 | (*mPyx)(iIn,iClust)=exp(-(*mPyx)(iIn,iClust)); | |
226 | pyxNorm+=(*mPyx)(iIn,iClust); | |
227 | } | |
228 | for (Int_t iClust=0; iClust<nk; iClust++) (*mPyx)(iIn,iClust)/=pyxNorm; | |
229 | } | |
230 | p->Zero(); | |
231 | y->Zero(); | |
232 | //recalculate codevectors | |
233 | for (Int_t iClust=0; iClust<nk; iClust++){ | |
234 | Double_t xpos=0,ypos=0,pxy; | |
235 | for (Int_t iIn=0; iIn<fNeff; iIn++) (*p)(iClust)+=(*vPx)(iIn)*(*mPyx)(iIn,iClust); | |
236 | for (Int_t iIn=0; iIn<fNeff; iIn++){ | |
237 | pxy=(*vPx)(iIn)*(*mPyx)(iIn,iClust)/(*p)(iClust); | |
238 | ypos+=pxy*TMath::Sin(xPhi[iIn]); | |
239 | xpos+=pxy*TMath::Cos(xPhi[iIn]); | |
240 | (*y)(0,iClust)+=pxy*xEta[iIn]; | |
241 | } | |
242 | (*y)(1,iClust)=(atan2(ypos,xpos)>0) ? atan2(ypos,xpos) : atan2(ypos,xpos)+2*pi; | |
243 | } | |
244 | //verify codevectors' stability | |
245 | chi=0; | |
246 | for (Int_t iClust=0; iClust<nk; iClust++){ | |
247 | chi1=TMath::CosH((*y1)(0,iClust)-(*y)(0,iClust))-TMath::Cos((*y1)(1,iClust)-(*y)(1,iClust)); | |
248 | chi1/=(2*TMath::CosH((*y1)(0,iClust))*TMath::CosH((*y)(0,iClust))); | |
249 | chi1*=chi1; | |
250 | if (chi1>chi) chi=chi1; | |
251 | } | |
252 | chi=TMath::Sqrt(chi); | |
253 | for (Int_t iClust=0; iClust<nk; iClust++){ | |
254 | for (Int_t i=0; i<2; i++) (*y1)(i,iClust)=(*y)(i,iClust); | |
255 | (*py)(iClust)=(*p)(iClust); | |
256 | } | |
257 | if (nloop>fNloopMax){ | |
258 | if (chi<fEpsMax || nloop>500) break; | |
259 | } | |
260 | }while (chi>fEps); | |
261 | }while (chi>fEpsMax); | |
262 | for (Int_t iClust=0; iClust<nk; iClust++){ //set codevectors and probability equal to those calculated | |
263 | for (Int_t i=0; i<2; i++) (*mY)(i,iClust)=(*y)(i,iClust); | |
264 | (*vPy)(iClust)=(*p)(iClust); | |
265 | } | |
266 | delete py; | |
267 | delete p; | |
268 | delete y; | |
269 | delete y1; | |
270 | delete ry; | |
271 | delete [] m; | |
272 | ||
7c679be0 | 273 | } |
274 | ||
275 | //----------------------------------------------------------------------------------- | |
139cbd96 | 276 | void AliDAJetFinder::NumCl(Int_t &nc,Int_t &nk,TVectorD *vPy, TMatrixD *mPyx,TMatrixD *mY) |
7c679be0 | 277 | { |
139cbd96 | 278 | // Number of clusters |
279 | static Bool_t growcl=true; | |
7c679be0 | 280 | |
139cbd96 | 281 | if (nk==2) growcl=true; |
282 | if (growcl){ | |
283 | //verify if two codevectors are equal within fAvDist | |
284 | Int_t *nSame = new Int_t[nk]; | |
285 | Int_t **iSame = new Int_t*[nk]; | |
286 | Int_t **cont = new Int_t*[nk]; | |
287 | for (Int_t iClust=0; iClust<nk; iClust++) { | |
288 | cont[iClust] =new Int_t[nk]; | |
289 | iSame[iClust]=new Int_t[nk]; | |
290 | } | |
291 | ||
292 | for (Int_t iClust=0; iClust<nk; iClust++){ | |
293 | iSame[iClust][iClust]=1; | |
294 | for (Int_t iClust1=iClust+1; iClust1<nk; iClust1++){ | |
295 | Double_t eta = (*mY)(0,iClust) ; Double_t phi = (*mY)(1,iClust); | |
296 | Double_t eta1 = (*mY)(0,iClust1); Double_t phi1 = (*mY)(1,iClust1); | |
297 | Double_t distCl=(TMath::CosH(eta-eta1)-TMath::Cos(phi-phi1))/(2*TMath::CosH(eta)*TMath::CosH(eta1)); | |
298 | if (distCl < fAvDist) iSame[iClust][iClust1]=iSame[iClust1][iClust]=1; | |
299 | else iSame[iClust][iClust1]=iSame[iClust1][iClust]=0; | |
300 | } | |
301 | } | |
302 | ReduceClusters(iSame,nk,nc,cont,nSame); | |
303 | if (nc >= fNclustMax) growcl=false; | |
304 | //recalculate the nc distinct codevectors | |
305 | TMatrixD *pyx = new TMatrixD(fNeff,2*nc); | |
306 | TVectorD *py = new TVectorD(nk); | |
307 | TMatrixD *y1 = new TMatrixD(3,nk); | |
308 | for (Int_t iClust=0; iClust<nc; iClust++){ | |
309 | for(Int_t j=0; j<nSame[iClust]; j++){ | |
310 | Int_t iClust1 = cont[iClust][j]; | |
311 | for (Int_t iIn=0; iIn<fNeff; iIn++) (*pyx)(iIn,iClust)+=(*mPyx)(iIn,iClust1); | |
312 | (*py)(iClust)+=(*vPy)(iClust1); | |
313 | for (Int_t i=0; i<2; i++) (*y1)(i,iClust)+=(*mY)(i,iClust1); | |
314 | } | |
315 | for (Int_t i=0; i<2; i++) (*y1)(i,iClust)/=nSame[iClust]; | |
316 | } | |
317 | for (Int_t iClust=0; iClust<nk; iClust++) delete [] cont[iClust], delete [] iSame[iClust]; | |
318 | delete [] iSame; | |
319 | delete [] cont; | |
320 | delete [] nSame; | |
321 | if (nc > nk/2){ | |
322 | for (Int_t iClust=0; iClust<nc; iClust++){ | |
323 | for (Int_t iIn=0; iIn<fNeff; iIn++) (*mPyx)(iIn,iClust)=(*pyx)(iIn,iClust); | |
324 | for (Int_t iComp=0; iComp<2; iComp++) (*mY)(iComp,iClust)=(*y1)(iComp,iClust); | |
325 | (*vPy)(iClust)=(*py)(iClust); | |
326 | } | |
327 | nk=nc; | |
328 | if (growcl) DoubleClusters(nc,nk,vPy,mY); | |
329 | } | |
330 | delete pyx; | |
331 | delete py; | |
332 | delete y1; | |
333 | } | |
7c679be0 | 334 | |
335 | } | |
336 | ||
337 | //----------------------------------------------------------------------------------- | |
c0a5117c | 338 | void AliDAJetFinder::ReduceClusters(Int_t **iSame,Int_t nc,Int_t &ncout,Int_t **cont,Int_t *nSameOut) const |
7c679be0 | 339 | { |
139cbd96 | 340 | // Reduction step |
341 | Int_t *nSame = new Int_t[nc]; | |
342 | Int_t *iperm = new Int_t[nc]; | |
343 | Int_t *go = new Int_t[nc]; | |
344 | for (Int_t iCl=0; iCl<nc; iCl++){ | |
345 | nSame[iCl]=0; | |
346 | for (Int_t jCl=0; jCl<nc; jCl++) nSame[iCl]+=iSame[iCl][jCl], cont[iCl][jCl]=0; | |
347 | iperm[iCl]=iCl; | |
348 | go[iCl]=1; | |
349 | } | |
350 | TMath::Sort(nc,nSame,iperm,true); | |
351 | Int_t l=0; | |
352 | for (Int_t iCl=0; iCl<nc; iCl++){ | |
353 | Int_t k=iperm[iCl]; | |
354 | if (go[k] == 1){ | |
355 | Int_t m=0; | |
356 | for (Int_t jCl=0; jCl<nc; jCl++){ | |
357 | if (iSame[k][jCl] == 1){ | |
358 | cont[l][m]=jCl; | |
359 | go[jCl]=0; | |
360 | m++; | |
7c679be0 | 361 | } |
139cbd96 | 362 | } |
363 | nSameOut[l]=m; | |
364 | l++; | |
365 | } | |
366 | } | |
367 | ncout=l; | |
368 | delete [] nSame; | |
369 | delete [] iperm; | |
370 | delete [] go; | |
371 | ||
7c679be0 | 372 | } |
373 | ||
374 | //----------------------------------------------------------------------------------- | |
139cbd96 | 375 | void AliDAJetFinder::EndDetAnn(Int_t &nk,Double_t **xData,Int_t *xx,Double_t etx,const TVectorD *vPx,TVectorD *vPy,TMatrixD *mPyx,TMatrixD *mY) |
7c679be0 | 376 | { |
139cbd96 | 377 | //now assign each particle to only one cluster |
378 | Double_t *clusters=new Double_t[nk]; | |
379 | for (Int_t iIn=0; iIn<fNeff; iIn++){ | |
380 | for (Int_t iClust=0; iClust<nk; iClust++) clusters[iClust]=(*mPyx)(iIn,iClust); | |
381 | xx[iIn]=TMath::LocMax(nk,clusters); | |
382 | } | |
383 | delete [] clusters; | |
7c679be0 | 384 | |
139cbd96 | 385 | //recalculate codevectors, having all p(y|x)=0 or 1 |
386 | Double_t *xEta = xData[0]; | |
387 | Double_t *xPhi = xData[1]; | |
388 | mY->Zero(); | |
389 | mPyx->Zero(); | |
390 | vPy->Zero(); | |
391 | for (Int_t iIn=0; iIn<fNin; iIn++){ | |
392 | Int_t iClust=xx[iIn]; | |
393 | (*mPyx)(iIn,iClust)=1; | |
394 | (*vPy)(iClust)+=(*vPx)(iIn); | |
395 | (*mY)(0,iClust)+=(*vPx)(iIn)*xEta[iIn]; | |
396 | (*mY)(3,iClust)+=(*vPx)(iIn)*etx; | |
397 | } | |
398 | Int_t k=0; | |
399 | for (Int_t iClust=0; iClust<nk; iClust++){ | |
400 | if ((*vPy)(iClust)>0){ | |
401 | Double_t xpos=0,ypos=0,pxy; | |
402 | for (Int_t iIn=0; iIn<fNin; iIn++){ | |
403 | pxy=(*vPx)(iIn)*(*mPyx)(iIn,iClust)/(*vPy)(iClust); | |
404 | ypos+=pxy*TMath::Sin(xPhi[iIn]); | |
405 | xpos+=pxy*TMath::Cos(xPhi[iIn]); | |
406 | if (xx[iIn]==iClust) xx[iIn]=k; | |
407 | } | |
408 | (*mY)(0,k)=(*mY)(0,iClust)/(*vPy)(iClust); | |
409 | (*mY)(1,k)=(atan2(ypos,xpos)>0) ? atan2(ypos,xpos) : atan2(ypos,xpos)+2*TMath::Pi(); | |
410 | (*mY)(3,k)=(*mY)(3,iClust); | |
411 | k++; | |
412 | } | |
413 | } | |
414 | nk=k; | |
415 | ||
7c679be0 | 416 | } |
417 | ||
418 | //----------------------------------------------------------------------------------- | |
139cbd96 | 419 | void AliDAJetFinder::StoreJets(Int_t nk, Double_t **xData, const Int_t *xx, const TMatrixD *mY) |
7c679be0 | 420 | { |
139cbd96 | 421 | //evaluate significant clusters properties |
422 | const Double_t pi=TMath::Pi(); | |
423 | Float_t dFidEtaMax = ((AliDAJetHeader*)fHeader)->GetFiducialEtaMax(); | |
424 | Float_t dFidEtaMin = ((AliDAJetHeader*)fHeader)->GetFiducialEtaMin(); | |
425 | Float_t dFiducialEta= dFidEtaMax - dFidEtaMin; | |
426 | Double_t *xEta = xData[0]; | |
427 | Double_t *xPhi = xData[1]; | |
428 | Int_t nEff = 0; | |
429 | for (Int_t i=0; i<fNeff; i++) if (xEta[i]<dFidEtaMax && xEta[i]>dFidEtaMin) nEff++; | |
430 | Double_t dMeanDist=0.; | |
431 | if (nEff > 0) | |
432 | dMeanDist=TMath::Sqrt(2*dFiducialEta*pi/nEff); | |
433 | Bool_t *isJet = new Bool_t[nk]; | |
434 | Double_t *etNoBg= new Double_t[nk]; | |
435 | Double_t *dDeltaEta=new Double_t[nk]; | |
436 | Double_t *dDeltaPhi=new Double_t[nk]; | |
437 | Double_t *surf = new Double_t[nk]; | |
438 | Double_t *etDens= new Double_t[nk]; | |
439 | for (Int_t iClust=0; iClust<nk; iClust++){ | |
440 | isJet[iClust]=false; | |
441 | Double_t dEtaMin=10.,dEtaMax=-10.,dPhiMin=10.,dPhiMax=0.; | |
442 | for (Int_t iIn=0; iIn<fNeff; iIn++){ | |
443 | if (xx[iIn]!=iClust || xEta[iIn]>dFidEtaMax || xEta[iIn]<dFidEtaMin) continue; | |
444 | if (xEta[iIn] < dEtaMin) dEtaMin=xEta[iIn]; | |
445 | if (xEta[iIn] > dEtaMax) dEtaMax=xEta[iIn]; | |
446 | Double_t dPhi=xPhi[iIn]-(*mY)(1,iClust); | |
447 | if (dPhi > pi ) dPhi-=2*pi; | |
448 | else if (dPhi < (-1)*pi) dPhi+=2*pi; | |
449 | if (dPhi < dPhiMin) dPhiMin=dPhi; | |
450 | else if (dPhi > dPhiMax) dPhiMax=dPhi; | |
451 | } | |
452 | dDeltaEta[iClust]=dEtaMax-dEtaMin+dMeanDist; | |
453 | dDeltaPhi[iClust]=dPhiMax-dPhiMin+dMeanDist; | |
454 | surf[iClust]=0.25*pi*dDeltaEta[iClust]*dDeltaPhi[iClust]; | |
455 | etDens[iClust]=(*mY)(3,iClust)/surf[iClust]; | |
456 | } | |
7c679be0 | 457 | |
139cbd96 | 458 | if (((AliDAJetHeader*)fHeader)->GetSelJets()){ |
459 | for (Int_t iClust=0; iClust<nk; iClust++){ | |
460 | if (!isJet[iClust] && (*mY)(0,iClust)<dFidEtaMax && (*mY)(0,iClust)>dFidEtaMin){ | |
461 | Double_t etDensMed=0.; | |
462 | Double_t etDensSqr=0.; | |
463 | Int_t norm=0; | |
464 | for (Int_t iClust1=0; iClust1<nk; iClust1++){ | |
465 | if(iClust1!=iClust && (*mY)(0,iClust)<dFidEtaMax && (*mY)(0,iClust)>dFidEtaMin){ | |
466 | etDensMed+=etDens[iClust1]; | |
467 | etDensSqr+=TMath::Power(etDens[iClust1],2); | |
468 | norm++; | |
469 | } | |
7c679be0 | 470 | } |
139cbd96 | 471 | etDensMed/=TMath::Max(norm,1); |
472 | etDensSqr/=TMath::Max(norm,1); | |
473 | Double_t deltaEtDens=TMath::Sqrt(etDensSqr-TMath::Power(etDensMed,2)); | |
474 | if ((*mY)(3,iClust) > (etDensMed+deltaEtDens)*surf[iClust]) isJet[iClust]=kTRUE; | |
475 | etNoBg[iClust]=(*mY)(3,iClust)-etDensMed*surf[iClust]; | |
476 | } | |
477 | } | |
478 | for (Int_t iClust=0; iClust<nk; iClust++){ | |
479 | if (isJet[iClust]){ | |
480 | Double_t etDensMed=0; | |
481 | Double_t extSurf=2*dFiducialEta*pi; | |
482 | for (Int_t iClust1=0; iClust1<nk; iClust1++){ | |
483 | if (!isJet[iClust1]) etDensMed+=(*mY)(3,iClust1); | |
484 | else extSurf-=surf[iClust1]; | |
7c679be0 | 485 | } |
139cbd96 | 486 | etDensMed/=extSurf; |
487 | etNoBg[iClust]=(*mY)(3,iClust)-etDensMed*surf[iClust]; | |
488 | if (etNoBg[iClust]<((AliDAJetHeader*)fHeader)->GetEtMin()){ | |
489 | isJet[iClust]=kFALSE; | |
490 | iClust=-1; | |
491 | } | |
492 | } | |
493 | } | |
494 | } else { | |
495 | for (Int_t iClust=0; iClust<nk; iClust++){ | |
496 | isJet[iClust]=true; | |
497 | etNoBg[iClust]=(*mY)(3,iClust); | |
498 | } | |
499 | } | |
500 | delete [] etDens; | |
501 | delete [] surf; | |
502 | ||
503 | //now add selected jets to the list | |
504 | Int_t *iSort = new Int_t[nk]; | |
505 | TMath::Sort(nk,etNoBg,iSort,true); | |
506 | Int_t iCl = 0; | |
507 | ||
508 | for (Int_t iClust=0; iClust<nk; iClust++){ //clusters loop | |
509 | iCl=iSort[iClust]; | |
510 | if (isJet[iCl]){ //choose cluster | |
511 | Float_t px,py,pz,en; | |
512 | px = (*mY)(3,iCl)*TMath::Cos((*mY)(1,iCl)); | |
513 | py = (*mY)(3,iCl)*TMath::Sin((*mY)(1,iCl)); | |
514 | pz = (*mY)(3,iCl)/TMath::Tan(2.0 * TMath::ATan(TMath::Exp(-(*mY)(0,iCl)))); | |
515 | en = TMath::Sqrt(px * px + py * py + pz * pz); | |
516 | AliAODJet jet(px, py, pz, en); | |
517 | Int_t iIn=0; | |
518 | Int_t nTr = GetCalTrkEvent()->GetNCalTrkTracks(); | |
519 | for (Int_t iTr=0; iTr<nTr; iTr++){ | |
520 | AliJetCalTrkTrack* ctT = GetCalTrkEvent()->GetCalTrkTrack(iTr); | |
521 | if (ctT->GetCutFlag()==0) continue; | |
522 | if (xx[iIn]==iCl) jet.AddTrack(ctT->GetTrackObject()); | |
523 | iIn++; | |
524 | } | |
525 | AddJet(jet); | |
526 | if (fDebug > 0) printf("jet %d, Eta: %f, Phi: %f, Et: %f\n",iCl,jet.Eta(),jet.Phi(),jet.Pt()); | |
527 | } | |
528 | } | |
529 | delete [] dDeltaEta; delete [] dDeltaPhi; | |
530 | delete [] etNoBg; | |
531 | delete [] isJet; | |
532 | delete [] iSort; | |
533 | ||
7c679be0 | 534 | } |
535 | ||
536 | //----------------------------------------------------------------------------------- | |
537 | Double_t Dist(TVectorD x,TVectorD y) | |
538 | { | |
139cbd96 | 539 | // Squared distance |
540 | const Double_t pi=TMath::Pi(); | |
541 | Double_t dphi=TMath::Abs(x(1)-y(1)); | |
542 | if (dphi > pi) dphi=2*pi-dphi; | |
543 | Double_t dist=pow(x(0)-y(0),2)+pow(dphi,2); | |
544 | return dist; | |
545 | ||
7c679be0 | 546 | } |