| 1 | /* ======================================================================== *\
|
|---|
| 2 | ! $Name: not supported by cvs2svn $:$Id: MRanForestCalc.cc,v 1.28 2007-08-23 10:25:08 tbretz Exp $
|
|---|
| 3 | ! --------------------------------------------------------------------------
|
|---|
| 4 | !
|
|---|
| 5 | ! *
|
|---|
| 6 | ! * This file is part of MARS, the MAGIC Analysis and Reconstruction
|
|---|
| 7 | ! * Software. It is distributed to you in the hope that it can be a useful
|
|---|
| 8 | ! * and timesaving tool in analysing Data of imaging Cerenkov telescopes.
|
|---|
| 9 | ! * It is distributed WITHOUT ANY WARRANTY.
|
|---|
| 10 | ! *
|
|---|
| 11 | ! * Permission to use, copy, modify and distribute this software and its
|
|---|
| 12 | ! * documentation for any purpose is hereby granted without fee,
|
|---|
| 13 | ! * provided that the above copyright notice appear in all copies and
|
|---|
| 14 | ! * that both that copyright notice and this permission notice appear
|
|---|
| 15 | ! * in supporting documentation. It is provided "as is" without express
|
|---|
| 16 | ! * or implied warranty.
|
|---|
| 17 | ! *
|
|---|
| 18 | !
|
|---|
| 19 | !
|
|---|
| 20 | ! Author(s): Thomas Hengstebeck 2/2005 <mailto:hengsteb@physik.hu-berlin.de>
|
|---|
| 21 | ! Author(s): Thomas Bretz 8/2005 <mailto:tbretz@astro.uni-wuerzburg.de>
|
|---|
| 22 | !
|
|---|
| 23 | ! Copyright: MAGIC Software Development, 2000-2006
|
|---|
| 24 | !
|
|---|
| 25 | !
|
|---|
| 26 | \* ======================================================================== */
|
|---|
| 27 |
|
|---|
| 28 | /////////////////////////////////////////////////////////////////////////////
|
|---|
| 29 | //
|
|---|
| 30 | // MRanForestCalc
|
|---|
| 31 | //
|
|---|
| 32 | //
|
|---|
| 33 | ////////////////////////////////////////////////////////////////////////////
|
|---|
| 34 | #include "MRanForestCalc.h"
|
|---|
| 35 |
|
|---|
| 36 | #include <TF1.h>
|
|---|
| 37 | #include <TFile.h>
|
|---|
| 38 | #include <TGraph.h>
|
|---|
| 39 | #include <TVector.h>
|
|---|
| 40 |
|
|---|
| 41 | #include "MHMatrix.h"
|
|---|
| 42 |
|
|---|
| 43 | #include "MLog.h"
|
|---|
| 44 | #include "MLogManip.h"
|
|---|
| 45 |
|
|---|
| 46 | #include "MData.h"
|
|---|
| 47 | #include "MDataArray.h"
|
|---|
| 48 |
|
|---|
| 49 | #include "MRanForest.h"
|
|---|
| 50 | #include "MParameters.h"
|
|---|
| 51 |
|
|---|
| 52 | #include "MParList.h"
|
|---|
| 53 | #include "MTaskList.h"
|
|---|
| 54 | #include "MEvtLoop.h"
|
|---|
| 55 | #include "MRanForestGrow.h"
|
|---|
| 56 | #include "MFillH.h"
|
|---|
| 57 |
|
|---|
| 58 | ClassImp(MRanForestCalc);
|
|---|
| 59 |
|
|---|
| 60 | using namespace std;
|
|---|
| 61 |
|
|---|
| 62 | const TString MRanForestCalc::gsDefName = "MRanForestCalc";
|
|---|
| 63 | const TString MRanForestCalc::gsDefTitle = "RF for energy estimation";
|
|---|
| 64 |
|
|---|
| 65 | const TString MRanForestCalc::gsNameOutput = "RanForestOut";
|
|---|
| 66 |
|
|---|
| 67 | MRanForestCalc::MRanForestCalc(const char *name, const char *title)
|
|---|
| 68 | : fData(0), fRFOut(0), fTestMatrix(0), fFunc("Function", "x"),
|
|---|
| 69 | fNumTrees(-1), fNumTry(-1), fNdSize(-1), fNumObsoleteVariables(1),
|
|---|
| 70 | fLastDataColumnHasWeights(kFALSE),
|
|---|
| 71 | fNameOutput(gsNameOutput), fDebug(kFALSE), fEstimationMode(kMean)
|
|---|
| 72 | {
|
|---|
| 73 | fName = name ? name : gsDefName.Data();
|
|---|
| 74 | fTitle = title ? title : gsDefTitle.Data();
|
|---|
| 75 |
|
|---|
| 76 | gROOT->GetListOfFunctions()->Remove(&fFunc);
|
|---|
| 77 |
|
|---|
| 78 | // FIXME:
|
|---|
| 79 | fNumTrees = 100; //100
|
|---|
| 80 | fNumTry = 0; //3 0 means: in MRanForest estimated best value will be calculated
|
|---|
| 81 | fNdSize = 1; //1
|
|---|
| 82 | }
|
|---|
| 83 |
|
|---|
| 84 | MRanForestCalc::~MRanForestCalc()
|
|---|
| 85 | {
|
|---|
| 86 | fEForests.Delete();
|
|---|
| 87 | }
|
|---|
| 88 |
|
|---|
| 89 | // --------------------------------------------------------------------------
|
|---|
| 90 | //
|
|---|
| 91 | // Set a function which is applied to the output of the random forest
|
|---|
| 92 | //
|
|---|
| 93 | Bool_t MRanForestCalc::SetFunction(const char *func)
|
|---|
| 94 | {
|
|---|
| 95 | return !fFunc.Compile(func);
|
|---|
| 96 | }
|
|---|
| 97 |
|
|---|
| 98 | // --------------------------------------------------------------------------
|
|---|
| 99 | //
|
|---|
| 100 | // ver=0: One yes/no-classification forest is trained for each bin.
|
|---|
| 101 | // the yes/no classification is done using the grid
|
|---|
| 102 | // ver=1: One classification forest is trained. The last column contains a
|
|---|
| 103 | // value which is turned into a classifier by rf itself using the grid
|
|---|
| 104 | // ver=2: One classification forest is trained. The last column already contains
|
|---|
| 105 | // the classifier
|
|---|
| 106 | // ver=3: A regression forest is trained. The last column contains the
|
|---|
| 107 | // classifier
|
|---|
| 108 | //
|
|---|
| 109 | Int_t MRanForestCalc::Train(const MHMatrix &matrixtrain, const TArrayD &grid, Int_t ver)
|
|---|
| 110 | {
|
|---|
| 111 | gLog.Separator("MRanForestCalc - Train");
|
|---|
| 112 |
|
|---|
| 113 | if (!matrixtrain.GetColumns())
|
|---|
| 114 | {
|
|---|
| 115 | *fLog << err << "ERROR - MHMatrix does not contain rules... abort." << endl;
|
|---|
| 116 | return kFALSE;
|
|---|
| 117 | }
|
|---|
| 118 |
|
|---|
| 119 | const Int_t ncols = matrixtrain.GetM().GetNcols();
|
|---|
| 120 | const Int_t nrows = matrixtrain.GetM().GetNrows();
|
|---|
| 121 | if (ncols<=0 || nrows <=0)
|
|---|
| 122 | {
|
|---|
| 123 | *fLog << err << "ERROR - No. of columns or no. of rows of matrixtrain equal 0 ... abort." << endl;
|
|---|
| 124 | return kFALSE;
|
|---|
| 125 | }
|
|---|
| 126 |
|
|---|
| 127 | // rules (= combination of image par) to be used for energy estimation
|
|---|
| 128 | TFile fileRF(fFileName, "recreate");
|
|---|
| 129 | if (!fileRF.IsOpen())
|
|---|
| 130 | {
|
|---|
| 131 | *fLog << err << "ERROR - File to store RFs could not be opened... abort." << endl;
|
|---|
| 132 | return kFALSE;
|
|---|
| 133 | }
|
|---|
| 134 |
|
|---|
| 135 | // The number of columns which have to be removed for the training
|
|---|
| 136 | // The last data column may contain weight which also have to be removed
|
|---|
| 137 | const Int_t nobs = fNumObsoleteVariables + (fLastDataColumnHasWeights?1:0); // Number of obsolete columns
|
|---|
| 138 |
|
|---|
| 139 | const MDataArray &dcol = *matrixtrain.GetColumns();
|
|---|
| 140 |
|
|---|
| 141 | // Make a copy of the rules for accessing the train-data
|
|---|
| 142 | MDataArray usedrules;
|
|---|
| 143 | for (Int_t i=0; i<ncols; i++)
|
|---|
| 144 | if (i<ncols-nobs) // -3 is important!!!
|
|---|
| 145 | usedrules.AddEntry(dcol[i].GetRule());
|
|---|
| 146 | else
|
|---|
| 147 | *fLog << inf << "Skipping " << dcol[i].GetRule() << " for training" << endl;
|
|---|
| 148 |
|
|---|
| 149 | // In the case of regression store the rule to be regessed in the
|
|---|
| 150 | // last entry of your rules
|
|---|
| 151 | MDataArray rules(usedrules);
|
|---|
| 152 | rules.AddEntry(ver<3?"Classification.fVal":dcol[ncols-1].GetRule().Data());
|
|---|
| 153 |
|
|---|
| 154 | // prepare train-matrix finally used
|
|---|
| 155 | TMatrix mat(matrixtrain.GetM());
|
|---|
| 156 |
|
|---|
| 157 | // Resize it such that the obsolete columns are removed
|
|---|
| 158 | mat.ResizeTo(nrows, ncols-nobs+1);
|
|---|
| 159 |
|
|---|
| 160 | if (fDebug)
|
|---|
| 161 | gLog.SetNullOutput(kTRUE);
|
|---|
| 162 |
|
|---|
| 163 | // In the case one independant RF is trained for each bin (e.g.
|
|---|
| 164 | // energy-bin) train all of them
|
|---|
| 165 | const Int_t nbins = ver>0 ? 1 : grid.GetSize()-1;
|
|---|
| 166 | for (Int_t ie=0; ie<nbins; ie++)
|
|---|
| 167 | {
|
|---|
| 168 | // In the case weights should be used initialize the
|
|---|
| 169 | // corresponding array
|
|---|
| 170 | TArrayF weights(nrows);
|
|---|
| 171 | if (fLastDataColumnHasWeights)
|
|---|
| 172 | for (Int_t j=0; j<nrows; j++)
|
|---|
| 173 | {
|
|---|
| 174 | weights[j] = matrixtrain.GetM()(j, ncols-nobs);
|
|---|
| 175 | if (j%250==0)
|
|---|
| 176 | cout << weights[j] << " ";
|
|---|
| 177 | }
|
|---|
| 178 |
|
|---|
| 179 | // Setup the matrix such that the last comlumn contains
|
|---|
| 180 | // the classifier or the regeression target value
|
|---|
| 181 | switch (ver)
|
|---|
| 182 | {
|
|---|
| 183 | case 0: // Replace last column by a classification which is 1 in
|
|---|
| 184 | // the case the event belongs to this bin, 0 otherwise
|
|---|
| 185 | {
|
|---|
| 186 | Int_t irows=0;
|
|---|
| 187 | for (Int_t j=0; j<nrows; j++)
|
|---|
| 188 | {
|
|---|
| 189 | const Double_t value = matrixtrain.GetM()(j,ncols-1);
|
|---|
| 190 | const Bool_t inside = value>grid[ie] && value<=grid[ie+1];
|
|---|
| 191 |
|
|---|
| 192 | mat(j, ncols-nobs) = inside ? 1 : 0;
|
|---|
| 193 |
|
|---|
| 194 | if (inside)
|
|---|
| 195 | irows++;
|
|---|
| 196 | }
|
|---|
| 197 | if (irows==0)
|
|---|
| 198 | *fLog << warn << "WARNING - Skipping";
|
|---|
| 199 | else
|
|---|
| 200 | *fLog << inf << "Training RF for";
|
|---|
| 201 |
|
|---|
| 202 | *fLog << " bin " << ie << " (" << grid[ie] << ", " << grid[ie+1] << ") " << irows << "/" << nrows << endl;
|
|---|
| 203 |
|
|---|
| 204 | if (irows==0)
|
|---|
| 205 | continue;
|
|---|
| 206 | }
|
|---|
| 207 | break;
|
|---|
| 208 |
|
|---|
| 209 | case 1: // Use last column as classifier or for regression
|
|---|
| 210 | case 2:
|
|---|
| 211 | case 3:
|
|---|
| 212 | for (Int_t j=0; j<nrows; j++)
|
|---|
| 213 | mat(j, ncols-nobs) = matrixtrain.GetM()(j,ncols-1);
|
|---|
| 214 | break;
|
|---|
| 215 | }
|
|---|
| 216 |
|
|---|
| 217 | MHMatrix matrix(mat, &rules, "MatrixTrain");
|
|---|
| 218 |
|
|---|
| 219 | MParList plist;
|
|---|
| 220 | MTaskList tlist;
|
|---|
| 221 | plist.AddToList(&tlist);
|
|---|
| 222 | plist.AddToList(&matrix);
|
|---|
| 223 |
|
|---|
| 224 | MRanForest rf;
|
|---|
| 225 | rf.SetNumTrees(fNumTrees);
|
|---|
| 226 | rf.SetNumTry(fNumTry);
|
|---|
| 227 | rf.SetNdSize(fNdSize);
|
|---|
| 228 | rf.SetClassify(ver<3 ? kTRUE : kFALSE);
|
|---|
| 229 | if (ver==1)
|
|---|
| 230 | rf.SetGrid(grid);
|
|---|
| 231 | if (fLastDataColumnHasWeights)
|
|---|
| 232 | rf.SetWeights(weights);
|
|---|
| 233 |
|
|---|
| 234 | plist.AddToList(&rf);
|
|---|
| 235 |
|
|---|
| 236 | MRanForestGrow rfgrow;
|
|---|
| 237 | tlist.AddToList(&rfgrow);
|
|---|
| 238 |
|
|---|
| 239 | MFillH fillh("MHRanForestGini");
|
|---|
| 240 | tlist.AddToList(&fillh);
|
|---|
| 241 |
|
|---|
| 242 | MEvtLoop evtloop(fTitle);
|
|---|
| 243 | evtloop.SetParList(&plist);
|
|---|
| 244 | evtloop.SetDisplay(fDisplay);
|
|---|
| 245 | evtloop.SetLogStream(fLog);
|
|---|
| 246 |
|
|---|
| 247 | if (!evtloop.Eventloop())
|
|---|
| 248 | return kFALSE;
|
|---|
| 249 |
|
|---|
| 250 | if (fDebug)
|
|---|
| 251 | gLog.SetNullOutput(kFALSE);
|
|---|
| 252 |
|
|---|
| 253 | if (ver==0)
|
|---|
| 254 | {
|
|---|
| 255 | // Calculate bin center
|
|---|
| 256 | const Double_t E = (TMath::Log10(grid[ie])+TMath::Log10(grid[ie+1]))/2;
|
|---|
| 257 |
|
|---|
| 258 | // save whole forest
|
|---|
| 259 | rf.SetUserVal(E);
|
|---|
| 260 | rf.SetName(Form("%.10f", E));
|
|---|
| 261 | }
|
|---|
| 262 |
|
|---|
| 263 | rf.Write();
|
|---|
| 264 | }
|
|---|
| 265 |
|
|---|
| 266 | // save rules
|
|---|
| 267 | usedrules.Write("rules");
|
|---|
| 268 |
|
|---|
| 269 | return kTRUE;
|
|---|
| 270 | }
|
|---|
| 271 |
|
|---|
| 272 | Int_t MRanForestCalc::ReadForests(MParList &plist)
|
|---|
| 273 | {
|
|---|
| 274 | TFile fileRF(fFileName, "read");
|
|---|
| 275 | if (!fileRF.IsOpen())
|
|---|
| 276 | {
|
|---|
| 277 | *fLog << err << dbginf << "File containing RFs could not be opened... aborting." << endl;
|
|---|
| 278 | return kFALSE;
|
|---|
| 279 | }
|
|---|
| 280 |
|
|---|
| 281 | fEForests.Delete();
|
|---|
| 282 |
|
|---|
| 283 | TIter Next(fileRF.GetListOfKeys());
|
|---|
| 284 | TObject *o=0;
|
|---|
| 285 | while ((o=Next()))
|
|---|
| 286 | {
|
|---|
| 287 | MRanForest *forest=0;
|
|---|
| 288 | fileRF.GetObject(o->GetName(), forest);
|
|---|
| 289 | if (!forest)
|
|---|
| 290 | continue;
|
|---|
| 291 |
|
|---|
| 292 | forest->SetUserVal(atof(o->GetName()));
|
|---|
| 293 |
|
|---|
| 294 | fEForests.Add(forest);
|
|---|
| 295 | }
|
|---|
| 296 |
|
|---|
| 297 | // Maybe fEForests[0].fRules could be used instead?
|
|---|
| 298 | if (fData->Read("rules")<=0)
|
|---|
| 299 | {
|
|---|
| 300 | *fLog << err << "ERROR - Reading 'rules' from file " << fFileName << endl;
|
|---|
| 301 | return kFALSE;
|
|---|
| 302 | }
|
|---|
| 303 |
|
|---|
| 304 | if (fileRF.GetListOfKeys()->FindObject("Function"))
|
|---|
| 305 | fFunc.Read("Function");
|
|---|
| 306 |
|
|---|
| 307 | return kTRUE;
|
|---|
| 308 | }
|
|---|
| 309 |
|
|---|
| 310 | Int_t MRanForestCalc::PreProcess(MParList *plist)
|
|---|
| 311 | {
|
|---|
| 312 | fRFOut = (MParameterD*)plist->FindCreateObj("MParameterD", fNameOutput);
|
|---|
| 313 | if (!fRFOut)
|
|---|
| 314 | return kFALSE;
|
|---|
| 315 |
|
|---|
| 316 | fData = (MDataArray*)plist->FindCreateObj("MDataArray");
|
|---|
| 317 | if (!fData)
|
|---|
| 318 | return kFALSE;
|
|---|
| 319 |
|
|---|
| 320 | if (!ReadForests(*plist))
|
|---|
| 321 | {
|
|---|
| 322 | *fLog << err << "Reading RFs failed... aborting." << endl;
|
|---|
| 323 | return kFALSE;
|
|---|
| 324 | }
|
|---|
| 325 |
|
|---|
| 326 | *fLog << inf << "RF read from " << fFileName << endl;
|
|---|
| 327 |
|
|---|
| 328 | if (fTestMatrix)
|
|---|
| 329 | return kTRUE;
|
|---|
| 330 |
|
|---|
| 331 | fData->Print();
|
|---|
| 332 |
|
|---|
| 333 | if (!fData->PreProcess(plist))
|
|---|
| 334 | {
|
|---|
| 335 | *fLog << err << "PreProcessing of the MDataArray failed... aborting." << endl;
|
|---|
| 336 | return kFALSE;
|
|---|
| 337 | }
|
|---|
| 338 |
|
|---|
| 339 | return kTRUE;
|
|---|
| 340 | }
|
|---|
| 341 |
|
|---|
| 342 | Double_t MRanForestCalc::Eval() const
|
|---|
| 343 | {
|
|---|
| 344 | TVector event;
|
|---|
| 345 | if (fTestMatrix)
|
|---|
| 346 | *fTestMatrix >> event;
|
|---|
| 347 | else
|
|---|
| 348 | *fData >> event;
|
|---|
| 349 |
|
|---|
| 350 | // --------------- Single Tree RF -------------------
|
|---|
| 351 | if (fEForests.GetEntriesFast()==1)
|
|---|
| 352 | {
|
|---|
| 353 | MRanForest *rf = static_cast<MRanForest*>(fEForests.UncheckedAt(0));
|
|---|
| 354 | return rf->CalcHadroness(event);
|
|---|
| 355 | }
|
|---|
| 356 |
|
|---|
| 357 | // --------------- Multi Tree RF -------------------
|
|---|
| 358 | static TF1 f1("f1", "gaus");
|
|---|
| 359 |
|
|---|
| 360 | Double_t sume = 0;
|
|---|
| 361 | Double_t sumh = 0;
|
|---|
| 362 | Double_t maxh = 0;
|
|---|
| 363 | Double_t maxe = 0;
|
|---|
| 364 |
|
|---|
| 365 | Double_t max = -1e10;
|
|---|
| 366 | Double_t min = 1e10;
|
|---|
| 367 |
|
|---|
| 368 | TIter Next(&fEForests);
|
|---|
| 369 | MRanForest *rf = 0;
|
|---|
| 370 |
|
|---|
| 371 | TGraph g;
|
|---|
| 372 | while ((rf=(MRanForest*)Next()))
|
|---|
| 373 | {
|
|---|
| 374 | const Double_t h = rf->CalcHadroness(event);
|
|---|
| 375 | const Double_t e = rf->GetUserVal();
|
|---|
| 376 |
|
|---|
| 377 | g.SetPoint(g.GetN(), e, h);
|
|---|
| 378 |
|
|---|
| 379 | sume += e*h;
|
|---|
| 380 | sumh += h;
|
|---|
| 381 |
|
|---|
| 382 | if (h>maxh)
|
|---|
| 383 | {
|
|---|
| 384 | maxh = h;
|
|---|
| 385 | maxe = e;
|
|---|
| 386 | }
|
|---|
| 387 | if (e>max)
|
|---|
| 388 | max = e;
|
|---|
| 389 | if (e<min)
|
|---|
| 390 | min = e;
|
|---|
| 391 | }
|
|---|
| 392 |
|
|---|
| 393 | switch (fEstimationMode)
|
|---|
| 394 | {
|
|---|
| 395 | case kMean:
|
|---|
| 396 | return sume/sumh;
|
|---|
| 397 | case kMaximum:
|
|---|
| 398 | return maxe;
|
|---|
| 399 | case kFit:
|
|---|
| 400 | f1.SetParameter(0, maxh);
|
|---|
| 401 | f1.SetParameter(1, maxe);
|
|---|
| 402 | f1.SetParameter(2, 0.125);
|
|---|
| 403 | g.Fit(&f1, "Q0N");
|
|---|
| 404 | return f1.GetParameter(1);
|
|---|
| 405 | }
|
|---|
| 406 |
|
|---|
| 407 | return 0;
|
|---|
| 408 | }
|
|---|
| 409 |
|
|---|
| 410 | Int_t MRanForestCalc::Process()
|
|---|
| 411 | {
|
|---|
| 412 | const Double_t val = Eval();
|
|---|
| 413 |
|
|---|
| 414 | fRFOut->SetVal(fFunc.Eval(val));
|
|---|
| 415 | fRFOut->SetReadyToSave();
|
|---|
| 416 |
|
|---|
| 417 | return kTRUE;
|
|---|
| 418 | }
|
|---|
| 419 |
|
|---|
| 420 | void MRanForestCalc::Print(Option_t *o) const
|
|---|
| 421 | {
|
|---|
| 422 | *fLog << all;
|
|---|
| 423 | *fLog << GetDescriptor() << ":" << endl;
|
|---|
| 424 | *fLog << " - Forest ";
|
|---|
| 425 | switch (fEForests.GetEntries())
|
|---|
| 426 | {
|
|---|
| 427 | case 0: *fLog << "not yet initialized." << endl; break;
|
|---|
| 428 | case 1: *fLog << "is a single tree forest." << endl; break;
|
|---|
| 429 | default: *fLog << "is a multi tree forest." << endl; break;
|
|---|
| 430 | }
|
|---|
| 431 | /*
|
|---|
| 432 | *fLog << " - Trees: " << fNumTrees << endl;
|
|---|
| 433 | *fLog << " - Trys: " << fNumTry << endl;
|
|---|
| 434 | *fLog << " - Node Size: " << fNdSize << endl;
|
|---|
| 435 | *fLog << " - Node Size: " << fNdSize << endl;
|
|---|
| 436 | */
|
|---|
| 437 | *fLog << " - FileName: " << fFileName << endl;
|
|---|
| 438 | *fLog << " - NameOutput: " << fNameOutput << endl;
|
|---|
| 439 | }
|
|---|
| 440 |
|
|---|
| 441 | // --------------------------------------------------------------------------
|
|---|
| 442 | //
|
|---|
| 443 | //
|
|---|
| 444 | Int_t MRanForestCalc::ReadEnv(const TEnv &env, TString prefix, Bool_t print)
|
|---|
| 445 | {
|
|---|
| 446 | Bool_t rc = kFALSE;
|
|---|
| 447 | if (IsEnvDefined(env, prefix, "FileName", print))
|
|---|
| 448 | {
|
|---|
| 449 | rc = kTRUE;
|
|---|
| 450 | SetFileName(GetEnvValue(env, prefix, "FileName", fFileName));
|
|---|
| 451 | }
|
|---|
| 452 | if (IsEnvDefined(env, prefix, "Debug", print))
|
|---|
| 453 | {
|
|---|
| 454 | rc = kTRUE;
|
|---|
| 455 | SetDebug(GetEnvValue(env, prefix, "Debug", fDebug));
|
|---|
| 456 | }
|
|---|
| 457 | if (IsEnvDefined(env, prefix, "NameOutput", print))
|
|---|
| 458 | {
|
|---|
| 459 | rc = kTRUE;
|
|---|
| 460 | SetNameOutput(GetEnvValue(env, prefix, "NameOutput", fNameOutput));
|
|---|
| 461 | }
|
|---|
| 462 | if (IsEnvDefined(env, prefix, "EstimationMode", print))
|
|---|
| 463 | {
|
|---|
| 464 | TString txt = GetEnvValue(env, prefix, "EstimationMode", "");
|
|---|
| 465 | txt = txt.Strip(TString::kBoth);
|
|---|
| 466 | txt.ToLower();
|
|---|
| 467 | if (txt==(TString)"mean")
|
|---|
| 468 | fEstimationMode = kMean;
|
|---|
| 469 | if (txt==(TString)"maximum")
|
|---|
| 470 | fEstimationMode = kMaximum;
|
|---|
| 471 | if (txt==(TString)"fit")
|
|---|
| 472 | fEstimationMode = kFit;
|
|---|
| 473 | rc = kTRUE;
|
|---|
| 474 | }
|
|---|
| 475 | return rc;
|
|---|
| 476 | }
|
|---|
