Logo ROOT   6.18/05
Reference Guide
HistoToWorkspaceFactoryFast.cxx
Go to the documentation of this file.
1// @(#)root/roostats:$Id: cranmer $
2// Author: Kyle Cranmer, Akira Shibata
3/*************************************************************************
4 * Copyright (C) 1995-2008, Rene Brun and Fons Rademakers. *
5 * All rights reserved. *
6 * *
7 * For the licensing terms see $ROOTSYS/LICENSE. *
8 * For the list of contributors see $ROOTSYS/README/CREDITS. *
9 *************************************************************************/
10
11////////////////////////////////////////////////////////////////////////////////
12
13/** \class RooStats::HistFactory::HistoToWorkspaceFactoryFast
14 * \ingroup HistFactory
15 */
16
17
18#ifndef __CINT__
19#include "RooGlobalFunc.h"
20#endif
21
22#include "RooDataSet.h"
23#include "RooRealVar.h"
24#include "RooConstVar.h"
25#include "RooAddition.h"
26#include "RooProduct.h"
27#include "RooProdPdf.h"
28#include "RooAddPdf.h"
29#include "RooGaussian.h"
30#include "RooPoisson.h"
31#include "RooExponential.h"
32#include "RooRandom.h"
33#include "RooCategory.h"
34#include "RooSimultaneous.h"
35#include "RooMultiVarGaussian.h"
36#include "RooNumIntConfig.h"
37#include "RooMinuit.h"
38#include "RooNLLVar.h"
39#include "RooProfileLL.h"
40#include "RooFitResult.h"
41#include "RooDataHist.h"
42#include "RooHistFunc.h"
43#include "RooHistPdf.h"
44#include "RooRealSumPdf.h"
45#include "RooProduct.h"
46#include "RooWorkspace.h"
47#include "RooCustomizer.h"
48#include "RooPlot.h"
49#include "RooMsgService.h"
55
56#include "TH2F.h"
57#include "TH3F.h"
58#include "TFile.h"
59#include "TCanvas.h"
60#include "TH1.h"
61#include "TLine.h"
62#include "TTree.h"
63#include "TMarker.h"
64#include "TStopwatch.h"
65#include "TROOT.h"
66#include "TStyle.h"
67#include "TVectorD.h"
68#include "TMatrixDSym.h"
69
70// specific to this package
75#include "Helper.h"
76
77#include <algorithm>
78#include <utility>
79
80#define VERBOSE
81
82#define alpha_Low "-5"
83#define alpha_High "5"
84#define NoHistConst_Low "0"
85#define NoHistConst_High "2000"
86
87// use this order for safety on library loading
88using namespace RooFit ;
89using namespace RooStats ;
90using namespace std ;
91
93
94namespace RooStats{
95namespace HistFactory{
96
98 fNomLumi(1.0), fLumiError(0),
99 fLowBin(0), fHighBin(0)
100 {}
101
103 }
104
106 fSystToFix( measurement.GetConstantParams() ),
107 fParamValues( measurement.GetParamValues() ),
108 fNomLumi( measurement.GetLumi() ),
109 fLumiError( measurement.GetLumi()*measurement.GetLumiRelErr() ),
110 fLowBin( measurement.GetBinLow() ),
111 fHighBin( measurement.GetBinHigh() ) {
112
113 // Set Preprocess functions
115
116 }
117
118 void HistoToWorkspaceFactoryFast::ConfigureWorkspaceForMeasurement( const std::string& ModelName, RooWorkspace* ws_single, Measurement& measurement ) {
119
120 // Configure a workspace by doing any
121 // necessary post-processing and by
122 // creating a ModelConfig
123
124 // Make a ModelConfig and configure it
125 ModelConfig * proto_config = (ModelConfig *) ws_single->obj("ModelConfig");
126 if( proto_config == NULL ) {
127 std::cout << "Error: Did not find 'ModelConfig' object in file: " << ws_single->GetName()
128 << std::endl;
129 throw hf_exc();
130 }
131
132 std::vector<std::string> poi_list = measurement.GetPOIList();
133 if( poi_list.size()==0 ) {
134 std::cout << "Warining: No Parametetrs of interest are set" << std::endl;
135 }
136
137 cout << "Setting Parameter(s) of Interest as: ";
138 for(unsigned int i = 0; i < poi_list.size(); ++i) {
139 cout << poi_list.at(i) << " ";
140 }
141 cout << endl;
142
143 RooArgSet params;
144 for( unsigned int i = 0; i < poi_list.size(); ++i ) {
145 std::string poi_name = poi_list.at(i);
146 RooRealVar* poi = (RooRealVar*) ws_single->var( poi_name.c_str() );
147 if(poi){
148 params.add(*poi);
149 }
150 else {
151 std::cout << "WARNING: Can't find parameter of interest: " << poi_name
152 << " in Workspace. Not setting in ModelConfig." << std::endl;
153 //throw hf_exc();
154 }
155 }
156 proto_config->SetParametersOfInterest(params);
157
158 // Name of an 'edited' model, if necessary
159 std::string NewModelName = "newSimPdf"; // <- This name is hard-coded in HistoToWorkspaceFactoryFast::EditSyt. Probably should be changed to : std::string("new") + ModelName;
160
161#ifdef DO_EDIT_WS
162 // Activate Additional Constraint Terms
163 if( measurement.GetGammaSyst().size() > 0
164 || measurement.GetUniformSyst().size() > 0
165 || measurement.GetLogNormSyst().size() > 0
166 || measurement.GetNoSyst().size() > 0) {
167 HistoToWorkspaceFactoryFast::EditSyst( ws_single, (ModelName).c_str(),
168 measurement.GetGammaSyst(),
169 measurement.GetUniformSyst(),
170 measurement.GetLogNormSyst(),
171 measurement.GetNoSyst());
172
173 proto_config->SetPdf( *ws_single->pdf( "newSimPdf" ) );
174 }
175#endif
176
177 // Set the ModelConfig's Params of Interest
178 RooAbsData* expData = ws_single->data("asimovData");
179 if( !expData ) {
180 std::cout << "Error: Failed to find dataset: " << expData
181 << " in workspace" << std::endl;
182 throw hf_exc();
183 }
184 if(poi_list.size()!=0){
185 proto_config->GuessObsAndNuisance(*expData);
186 }
187
188 // Now, let's loop over any additional asimov datasets
189 // that we need to make
190
191 // Get the pdf
192 // Notice that we get the "new" pdf, this is the one that is
193 // used in the creation of these asimov datasets since they
194 // are fitted (or may be, at least).
195 RooAbsPdf* pdf = ws_single->pdf(NewModelName.c_str());
196 if( !pdf ) pdf = ws_single->pdf( ModelName.c_str() );
197 const RooArgSet* observables = ws_single->set("observables");
198
199 // Create a SnapShot of the nominal values
200 std::string SnapShotName = "NominalParamValues";
201 ws_single->saveSnapshot(SnapShotName.c_str(), ws_single->allVars());
202
203 for( unsigned int i=0; i<measurement.GetAsimovDatasets().size(); ++i) {
204
205 // Set the variable values and "const" ness with the workspace
206 RooStats::HistFactory::Asimov& asimov = measurement.GetAsimovDatasets().at(i);
207 std::string AsimovName = asimov.GetName();
208
209 std::cout << "Generating additional Asimov Dataset: " << AsimovName << std::endl;
210 asimov.ConfigureWorkspace(ws_single);
211 RooDataSet* asimov_dataset =
213
214 std::cout << "Importing Asimov dataset" << std::endl;
215 bool failure = ws_single->import(*asimov_dataset, Rename(AsimovName.c_str()));
216 if( failure ) {
217 std::cout << "Error: Failed to import Asimov dataset: " << AsimovName
218 << std::endl;
219 delete asimov_dataset;
220 throw hf_exc();
221 }
222
223 // Load the snapshot at the end of every loop iteration
224 // so we start each loop with a "clean" snapshot
225 ws_single->loadSnapshot(SnapShotName.c_str());
226
227 // we can now deleted the data set after having imported it
228 delete asimov_dataset;
229
230 }
231
232 // Cool, we're done
233 return; // ws_single;
234 }
235
236
237 // We want to eliminate this interface and use the measurment directly
239
240 // This is a pretty light-weight wrapper function
241 //
242 // Take a fully configured measurement as well as
243 // one of its channels
244 //
245 // Return a workspace representing that channel
246 // Do this by first creating a vector of EstimateSummary's
247 // and this by configuring the workspace with any post-processing
248
249 // Get the channel's name
250 string ch_name = channel.GetName();
251
252 // Create a workspace for a SingleChannel from the Measurement Object
253 RooWorkspace* ws_single = this->MakeSingleChannelWorkspace(measurement, channel);
254 if( ws_single == NULL ) {
255 std::cout << "Error: Failed to make Single-Channel workspace for channel: " << ch_name
256 << " and measurement: " << measurement.GetName() << std::endl;
257 throw hf_exc();
258 }
259
260 // Finally, configure that workspace based on
261 // properties of the measurement
262 HistoToWorkspaceFactoryFast::ConfigureWorkspaceForMeasurement( "model_"+ch_name, ws_single, measurement );
263
264 return ws_single;
265
266 }
267
269
270 // This function takes a fully configured measurement
271 // which may contain several channels and returns
272 // a workspace holding the combined model
273 //
274 // This can be used, for example, within a script to produce
275 // a combined workspace on-the-fly
276 //
277 // This is a static function (for now) to make
278 // it a one-liner
279
280 // First, we create an instance of a HistFactory
281 HistoToWorkspaceFactoryFast factory( measurement );
282
283 // Loop over the channels and create the individual workspaces
284 vector<RooWorkspace*> channel_workspaces;
285 vector<string> channel_names;
286
287 for( unsigned int chanItr = 0; chanItr < measurement.GetChannels().size(); ++chanItr ) {
288
289 HistFactory::Channel& channel = measurement.GetChannels().at( chanItr );
290
291 if( ! channel.CheckHistograms() ) {
292 std::cout << "MakeModelAndMeasurementsFast: Channel: " << channel.GetName()
293 << " has uninitialized histogram pointers" << std::endl;
294 throw hf_exc();
295 }
296
297 string ch_name = channel.GetName();
298 channel_names.push_back(ch_name);
299
300 // GHL: Renaming to 'MakeSingleChannelWorkspace'
301 RooWorkspace* ws_single = factory.MakeSingleChannelModel( measurement, channel );
302
303 channel_workspaces.push_back(ws_single);
304
305 }
306
307
308 // Now, combine the individual channel workspaces to
309 // form the combined workspace
310 RooWorkspace* ws = factory.MakeCombinedModel( channel_names, channel_workspaces );
311
312
313 // Configure the workspace
315
316 // Delete channel workspaces
317 for (vector<RooWorkspace*>::iterator iter = channel_workspaces.begin() ; iter != channel_workspaces.end() ; ++iter) {
318 delete *iter ;
319 }
320
321 // Done. Return the pointer
322 return ws;
323
324 }
325
327 string prefix, string productPrefix,
328 string systTerm ) {
329 if(hist) {
330 cout << "processing hist " << hist->GetName() << endl;
331 } else {
332 cout << "hist is empty" << endl;
333 R__ASSERT(hist != 0);
334 return;
335 }
336
337 /// require dimension >=1 or <=3
338 if (fObsNameVec.empty() && !fObsName.empty()) { fObsNameVec.push_back(fObsName); }
339 R__ASSERT( fObsNameVec.size()>=1 && fObsNameVec.size()<=3 );
340
341 /// determine histogram dimensionality
342 unsigned int histndim(1);
343 std::string classname = hist->ClassName();
344 if (classname.find("TH1")==0) { histndim=1; }
345 else if (classname.find("TH2")==0) { histndim=2; }
346 else if (classname.find("TH3")==0) { histndim=3; }
347 R__ASSERT( histndim==fObsNameVec.size() );
348
349 /// create roorealvar observables
350 RooArgList observables;
351 std::vector<std::string>::iterator itr = fObsNameVec.begin();
352 for (int idx=0; itr!=fObsNameVec.end(); ++itr, ++idx ) {
353 if ( !proto->var(itr->c_str()) ) {
354 const TAxis* axis(0);
355 if (idx==0) { axis = hist->GetXaxis(); }
356 if (idx==1) { axis = hist->GetYaxis(); }
357 if (idx==2) { axis = hist->GetZaxis(); }
358 Int_t nbins = axis->GetNbins();
359 Double_t xmin = axis->GetXmin();
360 Double_t xmax = axis->GetXmax();
361 // create observable
362 proto->factory(Form("%s[%f,%f]",itr->c_str(),xmin,xmax));
363 proto->var(itr->c_str())->setBins(nbins);
364 }
365 observables.add( *proto->var(itr->c_str()) );
366 }
367
368 RooDataHist* histDHist = new RooDataHist((prefix+"nominalDHist").c_str(),"",observables,hist);
369 RooHistFunc* histFunc = new RooHistFunc((prefix+"_nominal").c_str(),"",observables,*histDHist,0) ;
370
371 proto->import(*histFunc);
372
373 /// now create the product of the overall efficiency times the sigma(params) for this estimate
374 proto->factory(("prod:"+productPrefix+"("+prefix+"_nominal,"+systTerm+")").c_str() );
375
376 delete histDHist;
377 delete histFunc;
378
379 }
380
381 void HistoToWorkspaceFactoryFast::AddMultiVarGaussConstraint(RooWorkspace* proto, string prefix,int lowBin, int highBin, vector<string>& constraintTermNames){
382 // these are the nominal predictions: eg. the mean of some space of variations
383 // later fill these in a loop over histogram bins
384
385 TVectorD mean(highBin); //-lowBin); // MB: fix range
386 cout << "a" << endl;
387 for(Int_t i=lowBin; i<highBin; ++i){
388 std::stringstream str;
389 str<<"_"<<i;
390 RooRealVar* temp = proto->var((prefix+str.str()).c_str());
391 mean(i) = temp->getVal();
392 }
393
394 TMatrixDSym Cov(highBin-lowBin);
395 for(int i=lowBin; i<highBin; ++i){
396 for(int j=0; j<highBin-lowBin; ++j){
397 if(i==j) { Cov(i,j) = sqrt(mean(i)); } // MB : this doesn't make sense to me if lowBin!=0 (?)
398 else { Cov(i,j) = 0; }
399 }
400 }
401
402 // can't make MultiVarGaussian with factory yet, do it by hand
403 RooArgList floating( *(proto->set(prefix.c_str() ) ) );
404 RooMultiVarGaussian constraint((prefix+"Constraint").c_str(),"",
405 floating, mean, Cov);
406
407 proto->import(constraint);
408
409 constraintTermNames.push_back(constraint.GetName());
410 }
411
413 std::vector<HistoSys> histoSysList,
414 string prefix, string productPrefix,
415 string systTerm,
416 vector<string>& constraintTermNames){
417
418 // these are the nominal predictions: eg. the mean of some space of variations
419 // later fill these in a loop over histogram bins
420
421 // require dimension >=1 or <=3
422 if (fObsNameVec.empty() && !fObsName.empty()) { fObsNameVec.push_back(fObsName); }
423 R__ASSERT( fObsNameVec.size()>=1 && fObsNameVec.size()<=3 );
424
425 // determine histogram dimensionality
426 unsigned int histndim(1);
427 std::string classname = nominal->ClassName();
428 if (classname.find("TH1")==0) { histndim=1; }
429 else if (classname.find("TH2")==0) { histndim=2; }
430 else if (classname.find("TH3")==0) { histndim=3; }
431 R__ASSERT( histndim==fObsNameVec.size() );
432 // cout <<"In LinInterpWithConstriants and histndim = " << histndim <<endl;
433
434 // create roorealvar observables
435 RooArgList observables;
436 std::vector<std::string>::iterator itr = fObsNameVec.begin();
437 for (int idx=0; itr!=fObsNameVec.end(); ++itr, ++idx ) {
438 if ( !proto->var(itr->c_str()) ) {
439 const TAxis* axis(nullptr);
440 if (idx==0) { axis = nominal->GetXaxis(); }
441 else if (idx==1) { axis = nominal->GetYaxis(); }
442 else if (idx==2) { axis = nominal->GetZaxis(); }
443 else {
444 std::cout << "Error: Too many observables. "
445 << "HistFactory only accepts up to 3 observables (3d) "
446 << std::endl;
447 throw hf_exc();
448 }
449 Int_t nbins = axis->GetNbins();
450 Double_t xmin = axis->GetXmin();
451 Double_t xmax = axis->GetXmax();
452 // create observable
453 proto->factory(Form("%s[%f,%f]",itr->c_str(),xmin,xmax));
454 proto->var(itr->c_str())->setBins(nbins);
455 }
456 observables.add( *proto->var(itr->c_str()) );
457 }
458
459 RooDataHist* nominalDHist = new RooDataHist((prefix+"nominalDHist").c_str(),"",observables,nominal);
460 RooHistFunc* nominalFunc = new RooHistFunc((prefix+"nominal").c_str(),"",observables,*nominalDHist,0) ;
461
462 // make list of abstract parameters that interpolate in space of variations
463 RooArgList params( ("alpha_Hist") );
464 // range is set using defined macro (see top of the page)
465 string range=string("[")+alpha_Low+","+alpha_High+"]";
466
467 // Loop over the HistoSys list
468 for(unsigned int j=0; j<histoSysList.size(); ++j){
469 std::stringstream str;
470 str<<"_"<<j;
471
472 HistoSys& histoSys = histoSysList.at(j);
473 string histoSysName = histoSys.GetName();
474
475 RooRealVar* temp = (RooRealVar*) proto->var(("alpha_" + histoSysName).c_str());
476 if(!temp){
477
478 temp = (RooRealVar*) proto->factory(("alpha_" + histoSysName + range).c_str());
479
480 // now add a constraint term for these parameters
481 string command=("Gaussian::alpha_"+histoSysName+"Constraint(alpha_"+histoSysName+",nom_alpha_"+histoSysName+"[0.,-10,10],1.)");
482 cout << command << endl;
483 constraintTermNames.push_back( proto->factory( command.c_str() )->GetName() );
484 proto->var(("nom_alpha_"+histoSysName).c_str())->setConstant();
485 const_cast<RooArgSet*>(proto->set("globalObservables"))->add(*proto->var(("nom_alpha_"+histoSysName).c_str()));
486 }
487 params.add(* temp );
488 }
489
490 // now make function that linearly interpolates expectation between variations
491 // get low/high variations to interpolate between
492 vector<double> low, high;
493 RooArgSet lowSet, highSet;
494 //ES// for(unsigned int j=0; j<lowHist.size(); ++j){
495 for(unsigned int j=0; j<histoSysList.size(); ++j){
496 std::stringstream str;
497 str<<"_"<<j;
498
499 HistoSys& histoSys = histoSysList.at(j);
500 RooDataHist* lowDHist = new RooDataHist((prefix+str.str()+"lowDHist").c_str(),"",observables, histoSys.GetHistoLow());
501 RooDataHist* highDHist = new RooDataHist((prefix+str.str()+"highDHist").c_str(),"",observables, histoSys.GetHistoHigh());
502 RooHistFunc* lowFunc = new RooHistFunc((prefix+str.str()+"low").c_str(),"",observables,*lowDHist,0) ;
503 RooHistFunc* highFunc = new RooHistFunc((prefix+str.str()+"high").c_str(),"",observables,*highDHist,0) ;
504 lowSet.add(*lowFunc);
505 highSet.add(*highFunc);
506 }
507
508 // this is sigma(params), a piece-wise linear interpolation
509 PiecewiseInterpolation interp(prefix.c_str(),"",*nominalFunc,lowSet,highSet,params);
510 interp.setPositiveDefinite();
511 interp.setAllInterpCodes(4); // LM: change to 4 (piece-wise linear to 6th order polynomial interpolation + linear extrapolation )
512 // KC: interpo codes 1 etc. don't have proper analytic integral.
513 RooArgSet observableSet(observables);
514 interp.setBinIntegrator(observableSet);
515 interp.forceNumInt();
516
517 proto->import(interp); // individual params have already been imported in first loop of this function
518
519 // now create the product of the overall efficiency times the sigma(params) for this estimate
520 proto->factory(("prod:"+productPrefix+"("+prefix+","+systTerm+")").c_str() );
521
522 }
523
524 // GHL: Consider passing the NormFactor list instead of the entire sample
525 string HistoToWorkspaceFactoryFast::AddNormFactor(RooWorkspace* proto, string& channel, string& sigmaEpsilon, Sample& sample, bool doRatio){
526 string overallNorm_times_sigmaEpsilon ;
527 string prodNames;
528
529 vector<NormFactor> normList = sample.GetNormFactorList();
530 vector<string> normFactorNames, rangeNames;
531
532 if(normList.size() > 0){
533
534 for(vector<NormFactor>::iterator itr = normList.begin(); itr != normList.end(); ++itr){
535
536 NormFactor& norm = *itr;
537
538 string varname;
539 if(!prodNames.empty()) prodNames += ",";
540 if(doRatio) {
541 varname = norm.GetName() + "_" + channel;
542 }
543 else {
544 varname=norm.GetName();
545 }
546
547 // GHL: Check that the NormFactor doesn't already exist
548 // (it may have been created as a function expression
549 // during preprocessing)
550 std::stringstream range;
551 range << "[" << norm.GetVal() << "," << norm.GetLow() << "," << norm.GetHigh() << "]";
552
553 if( proto->obj(varname.c_str()) == NULL) {
554 cout << "making normFactor: " << norm.GetName() << endl;
555 // remove "doRatio" and name can be changed when ws gets imported to the combined model.
556 proto->factory((varname + range.str()).c_str());
557 }
558
559 if(norm.GetConst()) {
560 // proto->var(varname.c_str())->setConstant();
561 // cout <<"setting " << varname << " constant"<<endl;
562 cout << "WARNING: Const attribute to <NormFactor> tag is deprecated, will ignore." <<
563 " Instead, add \n\t<ParamSetting Const=\"True\">" << varname << "</ParamSetting>\n" <<
564 " to your top-level XML's <Measurment> entry" << endl;
565 }
566 prodNames+=varname;
567 rangeNames.push_back(range.str());
568 normFactorNames.push_back(varname);
569 }
570
571 overallNorm_times_sigmaEpsilon = sample.GetName() + "_" + channel + "_overallNorm_x_sigma_epsilon";
572 proto->factory(("prod::" + overallNorm_times_sigmaEpsilon + "(" + prodNames + "," + sigmaEpsilon + ")").c_str());
573 }
574
575 unsigned int rangeIndex=0;
576 for( vector<string>::iterator nit = normFactorNames.begin(); nit!=normFactorNames.end(); ++nit){
577 if( count (normFactorNames.begin(), normFactorNames.end(), *nit) > 1 ){
578 cout <<"WARNING: <NormFactor Name =\""<<*nit<<"\"> is duplicated for <Sample Name=\""
579 << sample.GetName() << "\">, but only one factor will be included. \n Instead, define something like"
580 << "\n\t<Function Name=\""<<*nit<<"Squared\" Expresion=\""<<*nit<<"*"<<*nit<<"\" Var=\""<<*nit<<rangeNames.at(rangeIndex)
581 << "\"> \nin your top-level XML's <Measurment> entry and use <NormFactor Name=\""<<*nit<<"Squared\" in your channel XML file."<< endl;
582 }
583 ++rangeIndex;
584 }
585
586 if(!overallNorm_times_sigmaEpsilon.empty())
587 return overallNorm_times_sigmaEpsilon;
588 else
589 return sigmaEpsilon;
590 }
591
593 string interpName,
594 std::vector<OverallSys>& systList,
595 vector<string>& constraintTermNames,
596 vector<string>& totSystTermNames) {
597
598 // add variables for all the relative overall uncertainties we expect
599 // range is set using defined macro (see top of the page)
600
601 string range=string("[0,")+alpha_Low+","+alpha_High+"]";
602 totSystTermNames.push_back(prefix);
603
604 RooArgSet params(prefix.c_str());
605 vector<double> lowVec, highVec;
606
607 std::map<std::string, double>::iterator itconstr;
608 for(unsigned int i = 0; i < systList.size(); ++i) {
609
610 OverallSys& sys = systList.at(i);
611 std::string strname = sys.GetName();
612 const char * name = strname.c_str();
613
614 // case of no systematic (is it possible)
615 if (meas.GetNoSyst().count(sys.GetName()) > 0 ) {
616 std::cout << "HistoToWorkspaceFast::AddConstraintTerm - skip systematic " << sys.GetName() << std::endl;
617 continue;
618 }
619 // case systematic is a gamma constraint
620 if (meas.GetGammaSyst().count(sys.GetName()) > 0 ) {
621 double relerr = meas.GetGammaSyst().find(sys.GetName() )->second;
622 if (relerr <= 0) {
623 std::cout << "HistoToWorkspaceFast::AddConstraintTerm - zero uncertainty assigned - skip systematic " << sys.GetName() << std::endl;
624 continue;
625 }
626 double tauVal = 1./(relerr*relerr);
627 double sqtau = 1./relerr;
628 RooAbsArg * beta = proto->factory(TString::Format("beta_%s[1,0,10]",name) );
629 // the global observable (y_s)
630 RooAbsArg * yvar = proto->factory(TString::Format("nom_%s[%f,0,10]",beta->GetName(),tauVal)) ;
631 // the rate of the gamma distribution (theta)
632 RooAbsArg * theta = proto->factory(TString::Format("theta_%s[%f]",name,1./tauVal));
633 // find alpha as function of beta
634 RooAbsArg* alphaOfBeta = proto->factory(TString::Format("PolyVar::alphaOfBeta_%s(beta_%s,{%f,%f})",name,name,-sqtau,sqtau));
635
636 // add now the constraint itself Gamma_beta_constraint(beta, y+1, tau, 0 )
637 // build the gamma parameter k = as y_s + 1
638 RooAbsArg * kappa = proto->factory(TString::Format("sum::k_%s(%s,1.)",name,yvar->GetName()) );
639 RooAbsArg * gamma = proto->factory(TString::Format("Gamma::%sConstraint(%s, %s, %s, 0.0)",beta->GetName(),beta->GetName(), kappa->GetName(), theta->GetName() ) );
640 alphaOfBeta->Print("t");
641 gamma->Print("t");
642 constraintTermNames.push_back(gamma->GetName());
643 // set global observables
644 RooRealVar * gobs = dynamic_cast<RooRealVar*>(yvar); assert(gobs);
645 gobs->setConstant(true);
646 const_cast<RooArgSet*>(proto->set("globalObservables"))->add(*yvar);
647
648 // add alphaOfBeta in the list of params to interpolate
649 params.add(*alphaOfBeta);
650 std::cout << "Added a gamma constraint for " << name << std::endl;
651
652 }
653 else {
654
655 // add the Gaussian constraint part
656
657 // case systematic is uniform (asssume they are like a gauaaian bbut with a large width
658 // (100 instead of 1)
659 double gaussSigma = 1;
660 if (meas.GetUniformSyst().count(sys.GetName()) > 0 ) {
661 gaussSigma = 100;
662 std::cout << "Added a uniform constraint for " << name << " as a gaussian constraint with a very large sigma " << std::endl;
663 }
664
665 // add Gaussian constraint terms (normal + log-normal case)
666 RooRealVar* alpha = (RooRealVar*) proto->var((prefix + sys.GetName()).c_str());
667 if(!alpha) {
668
669 alpha = (RooRealVar*) proto->factory((prefix + sys.GetName() + range).c_str());
670 RooAbsArg * nomAlpha = proto->factory(TString::Format("nom_%s[0.,-10,10]",alpha->GetName() ) );
671 RooAbsArg * gausConstraint = proto->factory(TString::Format("Gaussian::%sConstraint(%s,%s,%f)",alpha->GetName(),alpha->GetName(), nomAlpha->GetName(), gaussSigma) );
672 //cout << command << endl;
673 constraintTermNames.push_back( gausConstraint->GetName() );
674 proto->var(("nom_" + prefix + sys.GetName()).c_str())->setConstant();
675 const_cast<RooArgSet*>(proto->set("globalObservables"))->add(*nomAlpha);
676 }
677
678
679 // add constraint in terms of bifrucated gauss with low/high as sigmas
680 //std::stringstream lowhigh;
681
682 // check if exists a log-normal constraint
683 if (meas.GetLogNormSyst().count(sys.GetName()) == 0 && meas.GetGammaSyst().count(sys.GetName()) == 0 ) {
684 // just add the alpha for the parameters of the FlexibleInterpVar function
685 params.add(*alpha);
686 }
687 // case systematic is a log-normal constraint
688 if (meas.GetLogNormSyst().count(sys.GetName()) > 0 ) {
689 // log normal constraint for parameter
690 double relerr = meas.GetLogNormSyst().find(sys.GetName() )->second;
691 double tauVal = 1./relerr;
692 std::string tauName = "tau_" + sys.GetName();
693 proto->factory(TString::Format("%s[%f]",tauName.c_str(),tauVal ) );
694 double kappaVal = 1. + relerr;
695 std::string kappaName = "kappa_" + sys.GetName();
696 proto->factory(TString::Format("%s[%f]",kappaName.c_str(),kappaVal ) );
697 const char * alphaName = alpha->GetName();
698
699 std::string alphaOfBetaName = "alphaOfBeta_" + sys.GetName();
700 RooAbsArg * alphaOfBeta = proto->factory(TString::Format("expr::%s('%s*(pow(%s,%s)-1.)',%s,%s,%s)",alphaOfBetaName.c_str(),
701 tauName.c_str(),kappaName.c_str(),alphaName,
702 tauName.c_str(),kappaName.c_str(),alphaName ) );
703
704 std::cout << "Added a log-normal constraint for " << name << std::endl;
705 alphaOfBeta->Print("t");
706 params.add(*alphaOfBeta);
707 }
708
709 }
710 // add low/high vectors
711 double low = sys.GetLow();
712 double high = sys.GetHigh();
713 lowVec.push_back(low);
714 highVec.push_back(high);
715
716 } // end sys loop
717
718 if(systList.size() > 0){
719 // this is epsilon(alpha_j), a piece-wise linear interpolation
720 // LinInterpVar interp( (interpName).c_str(), "", params, 1., lowVec, highVec);
721
722 assert( params.getSize() > 0);
723 assert(int(lowVec.size()) == params.getSize() );
724
725 FlexibleInterpVar interp( (interpName).c_str(), "", params, 1., lowVec, highVec);
726 interp.setAllInterpCodes(4); // LM: change to 4 (piece-wise linear to 6th order polynomial interpolation + linear extrapolation )
727 //interp.setAllInterpCodes(0); // simple linear interpolation
728 proto->import(interp); // params have already been imported in first loop of this function
729 } else{
730 // some strange behavior if params,lowVec,highVec are empty.
731 //cout << "WARNING: No OverallSyst terms" << endl;
732 RooConstVar interp( (interpName).c_str(), "", 1.);
733 proto->import(interp); // params have already been imported in first loop of this function
734 }
735
736 // std::cout << "after creating FlexibleInterpVar " << std::endl;
737 // proto->Print();
738
739 }
740
741
743 vector<string>& syst_x_expectedPrefixNames,
744 vector<string>& normByNames){
745
746 // for ith bin calculate totN_i = lumi * sum_j expected_j * syst_j
747 string command;
748 string coeffList="";
749 string shapeList="";
750 string prepend="";
751
752 if (fObsNameVec.empty() && !fObsName.empty()) { fObsNameVec.push_back(fObsName); }
753
754 double binWidth(1.0);
755 std::string obsNameVecStr;
756 std::vector<std::string>::iterator itr = fObsNameVec.begin();
757 for (; itr!=fObsNameVec.end(); ++itr) {
758 std::string obsName = *itr;
759 binWidth *= proto->var(obsName.c_str())->numBins()/(proto->var(obsName.c_str())->getMax() - proto->var(obsName.c_str())->getMin()) ; // MB: Note: requires fixed bin sizes
760 if (obsNameVecStr.size()>0) { obsNameVecStr += "_"; }
761 obsNameVecStr += obsName;
762 }
763
764 //vector<string>::iterator it=syst_x_expectedPrefixNames.begin();
765 for(unsigned int j=0; j<syst_x_expectedPrefixNames.size();++j){
766 std::stringstream str;
767 str<<"_"<<j;
768 // repatative, but we need one coeff for each term in the sum
769 // maybe can be avoided if we don't use bin width as coefficient
770 command=string(Form("binWidth_%s_%d[%e]",obsNameVecStr.c_str(),j,binWidth));
771 proto->factory(command.c_str());
772 proto->var(Form("binWidth_%s_%d",obsNameVecStr.c_str(),j))->setConstant();
773 coeffList+=prepend+"binWidth_"+obsNameVecStr+str.str();
774
775 command="prod::L_x_"+syst_x_expectedPrefixNames.at(j)+"("+normByNames.at(j)+","+syst_x_expectedPrefixNames.at(j)+")";
776 /*RooAbsReal* tempFunc =(RooAbsReal*) */
777 proto->factory(command.c_str());
778 shapeList+=prepend+"L_x_"+syst_x_expectedPrefixNames.at(j);
779 prepend=",";
780
781 // add to num int to product
782 // tempFunc->specialIntegratorConfig(kTRUE)->method1D().setLabel("RooBinIntegrator") ;
783 // tempFunc->forceNumInt();
784
785 }
786
787 proto->defineSet("coefList",coeffList.c_str());
788 proto->defineSet("shapeList",shapeList.c_str());
789 // proto->factory(command.c_str());
790 RooRealSumPdf tot(totName.c_str(),totName.c_str(),*proto->set("shapeList"),*proto->set("coefList"),kTRUE);
791 tot.specialIntegratorConfig(kTRUE)->method1D().setLabel("RooBinIntegrator") ;
792 tot.specialIntegratorConfig(kTRUE)->method2D().setLabel("RooBinIntegrator") ;
793 tot.specialIntegratorConfig(kTRUE)->methodND().setLabel("RooBinIntegrator") ;
794 tot.forceNumInt();
795
796 // for mixed generation in RooSimultaneous
797 tot.setAttribute("GenerateBinned"); // for use with RooSimultaneous::generate in mixed mode
798 // tot.setAttribute("GenerateUnbinned"); // we don't want that
799
800 /*
801 // Use binned numeric integration
802 int nbins = 0;
803 if( fObsNameVec.size() == 1 ) {
804 nbins = proto->var(fObsNameVec.at(0).c_str())->numBins();
805
806 cout <<"num bis for RooRealSumPdf = "<<nbins <<endl;
807 //int nbins = ((RooRealVar*) allVars.first())->numBins();
808 tot.specialIntegratorConfig(kTRUE)->getConfigSection("RooBinIntegrator").setRealValue("numBins",nbins);
809 tot.forceNumInt();
810
811 } else {
812 cout << "Bin Integrator only supports 1-d. Will be slow." << std::endl;
813 }
814 */
815
816
817 proto->import(tot);
818
819 }
820
821 void HistoToWorkspaceFactoryFast::AddPoissonTerms(RooWorkspace* proto, string prefix, string obsPrefix, string expPrefix, int lowBin, int highBin,
822 vector<string>& likelihoodTermNames){
823 /////////////////////////////////
824 // Relate observables to expected for each bin
825 // later modify variable named expPrefix_i to be product of terms
826 RooArgSet Pois(prefix.c_str());
827 for(Int_t i=lowBin; i<highBin; ++i){
828 std::stringstream str;
829 str<<"_"<<i;
830 //string command("Poisson::"+prefix+str.str()+"("+obsPrefix+str.str()+","+expPrefix+str.str()+")");
831 string command("Poisson::"+prefix+str.str()+"("+obsPrefix+str.str()+","+expPrefix+str.str()+",1)");//for no rounding
832 RooAbsArg* temp = (proto->factory( command.c_str() ) );
833
834 // output
835 cout << "Poisson Term " << command << endl;
836 ((RooAbsPdf*) temp)->setEvalErrorLoggingMode(RooAbsReal::PrintErrors);
837 //cout << temp << endl;
838
839 likelihoodTermNames.push_back( temp->GetName() );
840 Pois.add(* temp );
841 }
842 proto->defineSet(prefix.c_str(),Pois); // add argset to workspace
843 }
844
845 void HistoToWorkspaceFactoryFast::SetObsToExpected(RooWorkspace* proto, string obsPrefix, string expPrefix, int lowBin, int highBin){
846 /////////////////////////////////
847 // set observed to expected
848 TTree* tree = new TTree();
849 Double_t* obsForTree = new Double_t[highBin-lowBin];
850 RooArgList obsList("obsList");
851
852 for(Int_t i=lowBin; i<highBin; ++i){
853 std::stringstream str;
854 str<<"_"<<i;
855 RooRealVar* obs = (RooRealVar*) proto->var((obsPrefix+str.str()).c_str());
856 cout << "expected number of events called: " << expPrefix << endl;
857 RooAbsReal* exp = proto->function((expPrefix+str.str()).c_str());
858 if(obs && exp){
859
860 //proto->Print();
861 obs->setVal( exp->getVal() );
862 cout << "setting obs"+str.str()+" to expected = " << exp->getVal() << " check: " << obs->getVal() << endl;
863
864 // add entry to array and attach to tree
865 obsForTree[i] = exp->getVal();
866 tree->Branch((obsPrefix+str.str()).c_str(), obsForTree+i ,(obsPrefix+str.str()+"/D").c_str());
867 obsList.add(*obs);
868 }else{
869 cout << "problem retrieving obs or exp " << obsPrefix+str.str() << obs << " " << expPrefix+str.str() << exp << endl;
870 }
871 }
872 tree->Fill();
873 RooDataSet* data = new RooDataSet("expData","", tree, obsList); // one experiment
874
875 delete tree;
876 delete [] obsForTree;
877
878 proto->import(*data);
879
880 delete data;
881
882 }
883
884 //////////////////////////////////////////////////////////////////////////////
885
887 map<string,double> gammaSyst,
888 map<string,double> uniformSyst,
889 map<string,double> logNormSyst,
890 map<string,double> noSyst) {
891 string pdfName(pdfNameChar);
892
893 ModelConfig * combined_config = (ModelConfig *) proto->obj("ModelConfig");
894 if( combined_config==NULL ) {
895 std::cout << "Error: Failed to find object 'ModelConfig' in workspace: "
896 << proto->GetName() << std::endl;
897 throw hf_exc();
898 }
899 // const RooArgSet * constrainedParams=combined_config->GetNuisanceParameters();
900 // RooArgSet temp(*constrainedParams);
901 string edit="EDIT::newSimPdf("+pdfName+",";
902 string editList;
903 string lastPdf=pdfName;
904 string precede="";
905 unsigned int numReplacements = 0;
906 unsigned int nskipped = 0;
907 map<string,double>::iterator it;
908
909
910 // add gamma terms and their constraints
911 for(it=gammaSyst.begin(); it!=gammaSyst.end(); ++it) {
912 //cout << "edit for " << it->first << "with rel uncert = " << it->second << endl;
913 if(! proto->var(("alpha_"+it->first).c_str())){
914 //cout << "systematic not there" << endl;
915 nskipped++;
916 continue;
917 }
918 numReplacements++;
919
920 double relativeUncertainty = it->second;
921 double scale = 1/sqrt((1+1/pow(relativeUncertainty,2)));
922
923 // this is the Gamma PDF and in a form that doesn't have roundoff problems like the Poisson does
924 proto->factory(Form("beta_%s[1,0,10]",it->first.c_str()));
925 proto->factory(Form("y_%s[%f]",it->first.c_str(),1./pow(relativeUncertainty,2))) ;
926 proto->factory(Form("theta_%s[%f]",it->first.c_str(),pow(relativeUncertainty,2))) ;
927 proto->factory(Form("Gamma::beta_%sConstraint(beta_%s,sum::k_%s(y_%s,one[1]),theta_%s,zero[0])",
928 it->first.c_str(),
929 it->first.c_str(),
930 it->first.c_str(),
931 it->first.c_str(),
932 it->first.c_str())) ;
933
934 /*
935 // this has some problems because N in poisson is rounded to nearest integer
936 proto->factory(Form("Poisson::beta_%sConstraint(y_%s[%f],prod::taub_%s(taus_%s[%f],beta_%s[1,0,5]))",
937 it->first.c_str(),
938 it->first.c_str(),
939 1./pow(relativeUncertainty,2),
940 it->first.c_str(),
941 it->first.c_str(),
942 1./pow(relativeUncertainty,2),
943 it->first.c_str()
944 ) ) ;
945 */
946 // combined->factory(Form("expr::alphaOfBeta('(beta-1)/%f',beta)",scale));
947 // combined->factory(Form("expr::alphaOfBeta_%s('(beta_%s-1)/%f',beta_%s)",it->first.c_str(),it->first.c_str(),scale,it->first.c_str()));
948 proto->factory(Form("PolyVar::alphaOfBeta_%s(beta_%s,{%f,%f})",it->first.c_str(),it->first.c_str(),-1./scale,1./scale));
949
950 // set beta const status to be same as alpha
951 if(proto->var(Form("alpha_%s",it->first.c_str()))->isConstant()) {
952 proto->var(Form("beta_%s",it->first.c_str()))->setConstant(true);
953 }
954 else {
955 proto->var(Form("beta_%s",it->first.c_str()))->setConstant(false);
956 }
957 // set alpha const status to true
958 // proto->var(Form("alpha_%s",it->first.c_str()))->setConstant(true);
959
960 // replace alphas with alphaOfBeta and replace constraints
961 editList+=precede + "alpha_"+it->first+"Constraint=beta_" + it->first+ "Constraint";
962 precede=",";
963 editList+=precede + "alpha_"+it->first+"=alphaOfBeta_"+ it->first;
964
965 /*
966 if( proto->pdf(("alpha_"+it->first+"Constraint").c_str()) && proto->var(("alpha_"+it->first).c_str()) )
967 cout << " checked they are there" << proto->pdf(("alpha_"+it->first+"Constraint").c_str()) << " " << proto->var(("alpha_"+it->first).c_str()) << endl;
968 else
969 cout << "NOT THERE" << endl;
970 */
971
972 // EDIT seems to die if the list of edits is too long. So chunck them up.
973 if(numReplacements%10 == 0 && numReplacements+nskipped!=gammaSyst.size()){
974 edit="EDIT::"+lastPdf+"_("+lastPdf+","+editList+")";
975 lastPdf+="_"; // append an underscore for the edit
976 editList=""; // reset edit list
977 precede="";
978 cout << "Going to issue this edit command\n" << edit<< endl;
979 proto->factory( edit.c_str() );
980 RooAbsPdf* newOne = proto->pdf(lastPdf.c_str());
981 if(!newOne)
982 cout << "\n\n ---------------------\n WARNING: failed to make EDIT\n\n" << endl;
983
984 }
985 }
986
987 // add uniform terms and their constraints
988 for(it=uniformSyst.begin(); it!=uniformSyst.end(); ++it) {
989 cout << "edit for " << it->first << "with rel uncert = " << it->second << endl;
990 if(! proto->var(("alpha_"+it->first).c_str())){
991 cout << "systematic not there" << endl;
992 nskipped++;
993 continue;
994 }
995 numReplacements++;
996
997 // this is the Uniform PDF
998 proto->factory(Form("beta_%s[1,0,10]",it->first.c_str()));
999 proto->factory(Form("Uniform::beta_%sConstraint(beta_%s)",it->first.c_str(),it->first.c_str()));
1000 proto->factory(Form("PolyVar::alphaOfBeta_%s(beta_%s,{-1,1})",it->first.c_str(),it->first.c_str()));
1001
1002 // set beta const status to be same as alpha
1003 if(proto->var(Form("alpha_%s",it->first.c_str()))->isConstant())
1004 proto->var(Form("beta_%s",it->first.c_str()))->setConstant(true);
1005 else
1006 proto->var(Form("beta_%s",it->first.c_str()))->setConstant(false);
1007 // set alpha const status to true
1008 // proto->var(Form("alpha_%s",it->first.c_str()))->setConstant(true);
1009
1010 // replace alphas with alphaOfBeta and replace constraints
1011 cout << "alpha_"+it->first+"Constraint=beta_" + it->first+ "Constraint" << endl;
1012 editList+=precede + "alpha_"+it->first+"Constraint=beta_" + it->first+ "Constraint";
1013 precede=",";
1014 cout << "alpha_"+it->first+"=alphaOfBeta_"+ it->first << endl;
1015 editList+=precede + "alpha_"+it->first+"=alphaOfBeta_"+ it->first;
1016
1017 if( proto->pdf(("alpha_"+it->first+"Constraint").c_str()) && proto->var(("alpha_"+it->first).c_str()) )
1018 cout << " checked they are there" << proto->pdf(("alpha_"+it->first+"Constraint").c_str()) << " " << proto->var(("alpha_"+it->first).c_str()) << endl;
1019 else
1020 cout << "NOT THERE" << endl;
1021
1022 // EDIT seems to die if the list of edits is too long. So chunck them up.
1023 if(numReplacements%10 == 0 && numReplacements+nskipped!=gammaSyst.size()){
1024 edit="EDIT::"+lastPdf+"_("+lastPdf+","+editList+")";
1025 lastPdf+="_"; // append an underscore for the edit
1026 editList=""; // reset edit list
1027 precede="";
1028 cout << edit<< endl;
1029 proto->factory( edit.c_str() );
1030 RooAbsPdf* newOne = proto->pdf(lastPdf.c_str());
1031 if(!newOne)
1032 cout << "\n\n ---------------------\n WARNING: failed to make EDIT\n\n" << endl;
1033
1034 }
1035 }
1036
1037 /////////////////////////////////////////
1038 ////////////////////////////////////
1039
1040
1041 // add lognormal terms and their constraints
1042 for(it=logNormSyst.begin(); it!=logNormSyst.end(); ++it) {
1043 cout << "edit for " << it->first << "with rel uncert = " << it->second << endl;
1044 if(! proto->var(("alpha_"+it->first).c_str())){
1045 cout << "systematic not there" << endl;
1046 nskipped++;
1047 continue;
1048 }
1049 numReplacements++;
1050
1051 double relativeUncertainty = it->second;
1052 double kappa = 1+relativeUncertainty;
1053 // when transforming beta -> alpha, need alpha=1 to be +1sigma value.
1054 // the P(beta>kappa*\hat(beta)) = 16%
1055 // and \hat(beta) is 1, thus
1056 double scale = relativeUncertainty;
1057 //double scale = kappa;
1058
1059 const char * cname = it->first.c_str();
1060
1061 // this is the LogNormal
1062 proto->factory(TString::Format("beta_%s[1,0,10]",cname));
1063 proto->factory(TString::Format("nom_beta_%s[1]",cname));
1064 proto->factory(TString::Format("kappa_%s[%f]",cname,kappa));
1065 proto->factory(TString::Format("Lognormal::beta_%sConstraint(beta_%s,nom_beta_%s,kappa_%s)",
1066 cname, cname, cname, cname)) ;
1067 proto->factory(TString::Format("PolyVar::alphaOfBeta_%s(beta_%s,{%f,%f})",cname,cname,-1./scale,1./scale));
1068
1069
1070 // set beta const status to be same as alpha
1071 if(proto->var(TString::Format("alpha_%s",cname))->isConstant())
1072 proto->var(TString::Format("beta_%s",cname))->setConstant(true);
1073 else
1074 proto->var(TString::Format("beta_%s",cname))->setConstant(false);
1075 // set alpha const status to true
1076 // proto->var(TString::Format("alpha_%s",cname))->setConstant(true);
1077
1078 // replace alphas with alphaOfBeta and replace constraints
1079 cout << "alpha_"+it->first+"Constraint=beta_" + it->first+ "Constraint" << endl;
1080 editList+=precede + "alpha_"+it->first+"Constraint=beta_" + it->first+ "Constraint";
1081 precede=",";
1082 cout << "alpha_"+it->first+"=alphaOfBeta_"+ it->first << endl;
1083 editList+=precede + "alpha_"+it->first+"=alphaOfBeta_"+ it->first;
1084
1085 if( proto->pdf(("alpha_"+it->first+"Constraint").c_str()) && proto->var(("alpha_"+it->first).c_str()) )
1086 cout << " checked they are there" << proto->pdf(("alpha_"+it->first+"Constraint").c_str()) << " " << proto->var(("alpha_"+it->first).c_str()) << endl;
1087 else
1088 cout << "NOT THERE" << endl;
1089
1090 // EDIT seems to die if the list of edits is too long. So chunck them up.
1091 if(numReplacements%10 == 0 && numReplacements+nskipped!=gammaSyst.size()){
1092 edit="EDIT::"+lastPdf+"_("+lastPdf+","+editList+")";
1093 lastPdf+="_"; // append an underscore for the edit
1094 editList=""; // reset edit list
1095 precede="";
1096 cout << edit<< endl;
1097 proto->factory( edit.c_str() );
1098 RooAbsPdf* newOne = proto->pdf(lastPdf.c_str());
1099 if(!newOne)
1100 cout << "\n\n ---------------------\n WARNING: failed to make EDIT\n\n" << endl;
1101
1102 }
1103 // add global observables
1104 const RooArgSet * gobs = proto->set("globalObservables");
1105 RooArgSet gobsNew(*gobs);
1106 gobsNew.add(*proto->var(TString::Format("nom_beta_%s",cname)) );
1107 proto->removeSet("globalObservables");
1108 proto->defineSet("globalObservables",gobsNew);
1109 gobsNew.Print();
1110
1111 }
1112
1113 /////////////////////////////////////////
1114
1115 // MB: remove a systematic constraint
1116 for(it=noSyst.begin(); it!=noSyst.end(); ++it) {
1117
1118 cout << "remove constraint for parameter" << it->first << endl;
1119 if(! proto->var(("alpha_"+it->first).c_str()) || ! proto->pdf(("alpha_"+it->first+"Constraint").c_str()) ) {
1120 cout << "systematic not there" << endl;
1121 nskipped++;
1122 continue;
1123 }
1124 numReplacements++;
1125
1126 // dummy replacement pdf
1127 if ( !proto->var("one") ) { proto->factory("one[1.0]"); }
1128 proto->var("one")->setConstant();
1129
1130 // replace constraints
1131 cout << "alpha_"+it->first+"Constraint=one" << endl;
1132 editList+=precede + "alpha_"+it->first+"Constraint=one";
1133 precede=",";
1134
1135 // EDIT seems to die if the list of edits is too long. So chunck them up.
1136 if(numReplacements%10 == 0 && numReplacements+nskipped!=gammaSyst.size()){
1137 edit="EDIT::"+lastPdf+"_("+lastPdf+","+editList+")";
1138 lastPdf+="_"; // append an underscore for the edit
1139 editList=""; // reset edit list
1140 precede="";
1141 cout << edit << endl;
1142 proto->factory( edit.c_str() );
1143 RooAbsPdf* newOne = proto->pdf(lastPdf.c_str());
1144 if(!newOne) { cout << "\n\n ---------------------\n WARNING: failed to make EDIT\n\n" << endl; }
1145 }
1146 }
1147
1148 /////////////////////////////////////////
1149
1150 // commit last bunch of edits
1151 edit="EDIT::newSimPdf("+lastPdf+","+editList+")";
1152 cout << edit<< endl;
1153 proto->factory( edit.c_str() );
1154 // proto->writeToFile(("results/model_"+fRowTitle+"_edited.root").c_str());
1155 RooAbsPdf* newOne = proto->pdf("newSimPdf");
1156 if(newOne){
1157 // newOne->graphVizTree(("results/"+pdfName+"_"+fRowTitle+"newSimPdf.dot").c_str());
1158 combined_config->SetPdf(*newOne);
1159 }
1160 else{
1161 cout << "\n\n ---------------------\n WARNING: failed to make EDIT\n\n" << endl;
1162 }
1163 }
1164
1166 // Change-> Now a static utility
1167
1168 FILE* covFile = fopen ((filename).c_str(),"w");
1169
1170 TIter iti = params->createIterator();
1171 TIter itj = params->createIterator();
1172 RooRealVar *myargi, *myargj;
1173 fprintf(covFile," ") ;
1174 while ((myargi = (RooRealVar *)iti.Next())) {
1175 if(myargi->isConstant()) continue;
1176 fprintf(covFile," & %s", myargi->GetName());
1177 }
1178 fprintf(covFile,"\\\\ \\hline \n" );
1179 iti.Reset();
1180 while ((myargi = (RooRealVar *)iti.Next())) {
1181 if(myargi->isConstant()) continue;
1182 fprintf(covFile,"%s", myargi->GetName());
1183 itj.Reset();
1184 while ((myargj = (RooRealVar *)itj.Next())) {
1185 if(myargj->isConstant()) continue;
1186 cout << myargi->GetName() << "," << myargj->GetName();
1187 fprintf(covFile, " & %.2f", result->correlation(*myargi, *myargj));
1188 }
1189 cout << endl;
1190 fprintf(covFile, " \\\\\n");
1191 }
1192 fclose(covFile);
1193
1194 }
1195
1196
1197 ///////////////////////////////////////////////
1199
1200 // check inputs (see JIRA-6890 )
1201
1202 if (channel.GetSamples().empty()) {
1203 Error("MakeSingleChannelWorkspace",
1204 "The input Channel does not contain any sample - return a nullptr");
1205 return 0;
1206 }
1207
1208 const TH1* channel_hist_template = channel.GetSamples().front().GetHisto();
1209 if (channel_hist_template == nullptr) {
1210 channel.CollectHistograms();
1211 channel_hist_template = channel.GetSamples().front().GetHisto();
1212 }
1213 if (channel_hist_template == nullptr) {
1214 std::ostringstream stream;
1215 stream << "The sample " << channel.GetSamples().front().GetName()
1216 << " in channel " << channel.GetName() << " does not contain a histogram. This is the channel:\n";
1217 channel.Print(stream);
1218 Error("MakeSingleChannelWorkspace", "%s", stream.str().c_str());
1219 return 0;
1220 }
1221
1222 if( ! channel.CheckHistograms() ) {
1223 std::cout << "MakeSingleChannelWorkspace: Channel: " << channel.GetName()
1224 << " has uninitialized histogram pointers" << std::endl;
1225 throw hf_exc();
1226 }
1227
1228
1229
1230 // Set these by hand inside the function
1231 vector<string> systToFix = measurement.GetConstantParams();
1232 bool doRatio=false;
1233
1234 // to time the macro
1235 TStopwatch t;
1236 t.Start();
1237 //ES// string channel_name=summary[0].channel;
1238 string channel_name = channel.GetName();
1239
1240 /// MB: reset observable names for each new channel.
1241 fObsNameVec.clear();
1242
1243 /// MB: label observables x,y,z, depending on histogram dimensionality
1244 /// GHL: Give it the first sample's nominal histogram as a template
1245 /// since the data histogram may not be present
1246 if (fObsNameVec.empty()) { GuessObsNameVec(channel_hist_template); }
1247
1248 for ( unsigned int idx=0; idx<fObsNameVec.size(); ++idx ) {
1249 fObsNameVec[idx] = "obs_" + fObsNameVec[idx] + "_" + channel_name ;
1250 }
1251
1252 if (fObsNameVec.empty()) {
1253 fObsName= "obs_" + channel_name; // set name ov observable
1254 fObsNameVec.push_back( fObsName );
1255 }
1256
1257 R__ASSERT( fObsNameVec.size()>=1 && fObsNameVec.size()<=3 );
1258
1259 cout << "\n\n-------------------\nStarting to process " << channel_name << " channel with " << fObsNameVec.size() << " observables" << endl;
1260
1261 //
1262 // our main workspace that we are using to construct the model
1263 //
1264 RooWorkspace* proto = new RooWorkspace(channel_name.c_str(), (channel_name+" workspace").c_str());
1265 auto proto_config = make_unique<ModelConfig>("ModelConfig", proto);
1266 proto_config->SetWorkspace(*proto);
1267
1268 // preprocess functions
1269 vector<string>::iterator funcIter = fPreprocessFunctions.begin();
1270 for(;funcIter!= fPreprocessFunctions.end(); ++funcIter){
1271 cout <<"will preprocess this line: " << *funcIter <<endl;
1272 proto->factory(funcIter->c_str());
1273 proto->Print();
1274 }
1275
1276 RooArgSet likelihoodTerms("likelihoodTerms"), constraintTerms("constraintTerms");
1277 vector<string> likelihoodTermNames, constraintTermNames, totSystTermNames, syst_x_expectedPrefixNames, normalizationNames;
1278
1279 vector< pair<string,string> > statNamePairs;
1280 vector< pair<const TH1*, const TH1*> > statHistPairs; // <nominal, error>
1281 std::string statFuncName; // the name of the ParamHistFunc
1282 std::string statNodeName; // the name of the McStat Node
1283 // Constraint::Type statConstraintType=Constraint::Gaussian;
1284 // Double_t statRelErrorThreshold=0.0;
1285
1286 string prefix, range;
1287
1288 /////////////////////////////
1289 // shared parameters
1290 // this is ratio of lumi to nominal lumi. We will include relative uncertainty in model
1291 std::stringstream lumiStr;
1292 // lumi range
1293 lumiStr<<"["<<fNomLumi<<",0,"<<10.*fNomLumi<<"]";
1294 proto->factory(("Lumi"+lumiStr.str()).c_str());
1295 cout << "lumi str = " << lumiStr.str() << endl;
1296
1297 std::stringstream lumiErrorStr;
1298 lumiErrorStr << "nominalLumi["<<fNomLumi << ",0,"<<fNomLumi+10*fLumiError<<"]," << fLumiError ;
1299 proto->factory(("Gaussian::lumiConstraint(Lumi,"+lumiErrorStr.str()+")").c_str());
1300 proto->var("nominalLumi")->setConstant();
1301 proto->defineSet("globalObservables","nominalLumi");
1302 //likelihoodTermNames.push_back("lumiConstraint");
1303 constraintTermNames.push_back("lumiConstraint");
1304 cout << "lumi Error str = " << lumiErrorStr.str() << endl;
1305
1306 //proto->factory((string("SigXsecOverSM[1.,0.5,1..8]").c_str()));
1307 ///////////////////////////////////
1308 // loop through estimates, add expectation, floating bin predictions,
1309 // and terms that constrain floating to expectation via uncertainties
1310 // GHL: Loop over samples instead, which doesn't contain the data
1311 vector<Sample>::iterator it = channel.GetSamples().begin();
1312 for(; it!=channel.GetSamples().end(); ++it) {
1313
1314 //ES// string overallSystName = it->name+"_"+it->channel+"_epsilon";
1315 Sample& sample = (*it);
1316 string overallSystName = sample.GetName() + "_" + channel_name + "_epsilon";
1317
1318 string systSourcePrefix = "alpha_";
1319
1320 // constraintTermNames and totSystTermNames are vectors that are passed
1321 // by reference and filled by this method
1322 AddConstraintTerms(proto,measurement, systSourcePrefix, overallSystName,
1323 sample.GetOverallSysList(), constraintTermNames , totSystTermNames);
1324
1325 // GHL: Consider passing the NormFactor list instead of the entire sample
1326 overallSystName = AddNormFactor(proto, channel_name, overallSystName, sample, doRatio);
1327
1328 // Create the string for the object
1329 // that is added to the RooRealSumPdf
1330 // for this channel
1331 string syst_x_expectedPrefix = "";
1332
1333 // get histogram
1334 //ES// TH1* nominal = it->nominal;
1335 const TH1* nominal = sample.GetHisto();
1336
1337 // MB : HACK no option to have both non-hist variations and hist variations ?
1338 // get histogram
1339 // GHL: Okay, this is going to be non-trivial.
1340 // We will loop over histosys's, which contain both
1341 // the low hist and the high hist together.
1342
1343 // Logic:
1344 // - If we have no HistoSys's, do part A
1345 // - else, if the histo syst's don't match, return (we ignore this case)
1346 // - finally, we take the syst's and apply the linear interpolation w/ constraint
1347
1348 if(sample.GetHistoSysList().size() == 0) {
1349
1350 // If no HistoSys
1351 cout << sample.GetName() + "_" + channel_name + " has no variation histograms " << endl;
1352 string expPrefix = sample.GetName() + "_" + channel_name; //+"_expN";
1353 syst_x_expectedPrefix = sample.GetName() + "_" + channel_name + "_overallSyst_x_Exp";
1354
1355 ProcessExpectedHisto(sample.GetHisto(), proto, expPrefix, syst_x_expectedPrefix,
1356 overallSystName);
1357 }
1358 else {
1359 // If there ARE HistoSys(s)
1360 // name of source for variation
1361 string constraintPrefix = sample.GetName() + "_" + channel_name + "_Hist_alpha";
1362 syst_x_expectedPrefix = sample.GetName() + "_" + channel_name + "_overallSyst_x_HistSyst";
1363 // constraintTermNames are passed by reference and appended to,
1364 // overallSystName is a std::string for this sample
1365
1367 constraintPrefix, syst_x_expectedPrefix, overallSystName,
1368 constraintTermNames);
1369 }
1370
1371 ////////////////////////////////////
1372 // Add StatErrors to this Channel //
1373 ////////////////////////////////////
1374
1375 if( sample.GetStatError().GetActivate() ) {
1376
1377 if( fObsNameVec.size() > 3 ) {
1378 std::cout << "Cannot include Stat Error for histograms of more than 3 dimensions."
1379 << std::endl;
1380 throw hf_exc();
1381 } else {
1382
1383 // If we are using StatUncertainties, we multiply this object
1384 // by the ParamHistFunc and then pass that to the
1385 // RooRealSumPdf by appending it's name to the list
1386
1387 std::cout << "Sample: " << sample.GetName() << " to be included in Stat Error "
1388 << "for channel " << channel_name
1389 << std::endl;
1390
1391 /*
1392 Constraint::Type type = channel.GetStatErrorConfig().GetConstraintType();
1393 statConstraintType = Constraint::Gaussian;
1394 if( type == Constraint::Gaussian) {
1395 std::cout << "Using Gaussian StatErrors" << std::endl;
1396 statConstraintType = Constraint::Gaussian;
1397 }
1398 if( type == Constraint::Poisson ) {
1399 std::cout << "Using Poisson StatErrors" << std::endl;
1400 statConstraintType = Constraint::Poisson;
1401 }
1402 */
1403
1404 //statRelErrorThreshold = channel.GetStatErrorConfig().GetRelErrorThreshold();
1405
1406 // First, get the uncertainty histogram
1407 // and push it back to our vectors
1408
1409 //if( sample.GetStatError().GetErrorHist() ) {
1410 //statErrorHist = (TH1*) sample.GetStatError().GetErrorHist()->Clone();
1411 //}
1412 //if( statErrorHist == NULL ) {
1413
1414 // We need to get the *ABSOLUTE* uncertainty for use in Stat Uncertainties
1415 // This can be done in one of two ways:
1416 // - Use the built-in Errors in the TH1 itself (they are aboslute)
1417 // - Take the supplied *RELATIVE* error and multiply by the nominal
1418 string UncertName = syst_x_expectedPrefix + "_StatAbsolUncert";
1419 TH1* statErrorHist = NULL;
1420
1421 if( sample.GetStatError().GetErrorHist() == NULL ) {
1422 // Make the absolute stat error
1423 std::cout << "Making Statistical Uncertainty Hist for "
1424 << " Channel: " << channel_name
1425 << " Sample: " << sample.GetName()
1426 << std::endl;
1427 statErrorHist = MakeAbsolUncertaintyHist( UncertName, nominal );
1428 } else {
1429 // clone the error histograms because in case the sample has not error hist
1430 // it is created in MakeAbsolUncertainty
1431 // we need later to clean statErrorHist
1432 statErrorHist = (TH1*) sample.GetStatError().GetErrorHist()->Clone();
1433 // We assume the (relative) error is provided.
1434 // We must turn it into an absolute error
1435 // using the nominal histogram
1436 std::cout << "Using external histogram for Stat Errors for "
1437 << " Channel: " << channel_name
1438 << " Sample: " << sample.GetName()
1439 << std::endl;
1440 std::cout << "Error Histogram: " << statErrorHist->GetName() << std::endl;
1441 // Multiply the relative stat uncertainty by the
1442 // nominal to get the overall stat uncertainty
1443 statErrorHist->Multiply( nominal );
1444 statErrorHist->SetName( UncertName.c_str() );
1445 }
1446
1447 // Save the nominal and error hists
1448 // for the building of constraint terms
1449 statHistPairs.push_back( std::make_pair(nominal, statErrorHist) );
1450
1451 // To do the 'conservative' version, we would need to do some
1452 // intervention here. We would probably need to create a different
1453 // ParamHistFunc for each sample in the channel. The would nominally
1454 // use the same gamma's, so we haven't increased the number of parameters
1455 // However, if a bin in the 'nominal' histogram is 0, we simply need to
1456 // change the parameter in that bin in the ParamHistFunc for this sample.
1457 // We also need to add a constraint term.
1458 // Actually, we'd probably not use the ParamHistFunc...?
1459 // we could remove the dependence in this ParamHistFunc on the ith gamma
1460 // and then create the poisson term: Pois(tau | n_exp)Pois(data | n_exp)
1461
1462
1463 // Next, try to get the ParamHistFunc (it may have been
1464 // created by another sample in this channel)
1465 // or create it if it doesn't yet exist:
1466 statFuncName = "mc_stat_" + channel_name;
1467 ParamHistFunc* paramHist = (ParamHistFunc*) proto->function( statFuncName.c_str() );
1468 if( paramHist == NULL ) {
1469
1470 // Get a RooArgSet of the observables:
1471 // Names in the list fObsNameVec:
1472 RooArgList observables;
1473 std::vector<std::string>::iterator itr = fObsNameVec.begin();
1474 for (int idx=0; itr!=fObsNameVec.end(); ++itr, ++idx ) {
1475 observables.add( *proto->var(itr->c_str()) );
1476 }
1477
1478 // Create the list of terms to
1479 // control the bin heights:
1480 std::string ParamSetPrefix = "gamma_stat_" + channel_name;
1481 Double_t gammaMin = 0.0;
1482 Double_t gammaMax = 10.0;
1483 RooArgList statFactorParams = ParamHistFunc::createParamSet(*proto,
1484 ParamSetPrefix.c_str(),
1485 observables,
1486 gammaMin, gammaMax);
1487
1488 ParamHistFunc statUncertFunc(statFuncName.c_str(), statFuncName.c_str(),
1489 observables, statFactorParams );
1490
1491 proto->import( statUncertFunc, RecycleConflictNodes() );
1492
1493 paramHist = (ParamHistFunc*) proto->function( statFuncName.c_str() );
1494
1495 } // END: If Statement: Create ParamHistFunc
1496
1497 // Create the node as a product
1498 // of this function and the
1499 // expected value from MC
1500 statNodeName = sample.GetName() + "_" + channel_name + "_overallSyst_x_StatUncert";
1501
1502 RooAbsReal* expFunc = (RooAbsReal*) proto->function( syst_x_expectedPrefix.c_str() );
1503 RooProduct nodeWithMcStat(statNodeName.c_str(), statNodeName.c_str(),
1504 RooArgSet(*paramHist, *expFunc) );
1505
1506 proto->import( nodeWithMcStat, RecycleConflictNodes() );
1507
1508 // Push back the final name of the node
1509 // to be used in the RooRealSumPdf
1510 // (node to be created later)
1511 syst_x_expectedPrefix = nodeWithMcStat.GetName();
1512
1513 }
1514 } // END: if DoMcStat
1515
1516
1517 ///////////////////////////////////////////
1518 // Create a ShapeFactor for this channel //
1519 ///////////////////////////////////////////
1520
1521 if( sample.GetShapeFactorList().size() > 0 ) {
1522
1523 if( fObsNameVec.size() > 3 ) {
1524 std::cout << "Cannot include Stat Error for histograms of more than 3 dimensions."
1525 << std::endl;
1526 throw hf_exc();
1527 } else {
1528
1529 std::cout << "Sample: " << sample.GetName() << " in channel: " << channel_name
1530 << " to be include a ShapeFactor."
1531 << std::endl;
1532
1533 std::vector<ParamHistFunc*> paramHistFuncList;
1534 std::vector<std::string> shapeFactorNameList;
1535
1536 for(unsigned int i=0; i < sample.GetShapeFactorList().size(); ++i) {
1537
1538 ShapeFactor& shapeFactor = sample.GetShapeFactorList().at(i);
1539
1540 std::string funcName = channel_name + "_" + shapeFactor.GetName() + "_shapeFactor";
1541 ParamHistFunc* paramHist = (ParamHistFunc*) proto->function( funcName.c_str() );
1542 if( paramHist == NULL ) {
1543
1544 RooArgList observables;
1545 std::vector<std::string>::iterator itr = fObsNameVec.begin();
1546 for (int idx=0; itr!=fObsNameVec.end(); ++itr, ++idx ) {
1547 observables.add( *proto->var(itr->c_str()) );
1548 }
1549
1550 // Create the Parameters
1551 std::string funcParams = "gamma_" + shapeFactor.GetName();
1552
1553 // GHL: Again, we are putting hard ranges on the gamma's
1554 // We should change this to range from 0 to /inf
1555 RooArgList shapeFactorParams = ParamHistFunc::createParamSet(*proto,
1556 funcParams.c_str(),
1557 observables, 0, 1000);
1558
1559 // Create the Function
1560 ParamHistFunc shapeFactorFunc( funcName.c_str(), funcName.c_str(),
1561 observables, shapeFactorParams );
1562
1563 // Set an initial shape, if requested
1564 if( shapeFactor.GetInitialShape() != NULL ) {
1565 TH1* initialShape = static_cast<TH1*>(shapeFactor.GetInitialShape()->Clone());
1566 std::cout << "Setting Shape Factor: " << shapeFactor.GetName()
1567 << " to have initial shape from hist: "
1568 << initialShape->GetName()
1569 << std::endl;
1570 shapeFactorFunc.setShape( initialShape );
1571 }
1572
1573 // Set the variables constant, if requested
1574 if( shapeFactor.IsConstant() ) {
1575 std::cout << "Setting Shape Factor: " << shapeFactor.GetName()
1576 << " to be constant" << std::endl;
1577 shapeFactorFunc.setConstant(true);
1578 }
1579
1580 proto->import( shapeFactorFunc, RecycleConflictNodes() );
1581 paramHist = (ParamHistFunc*) proto->function( funcName.c_str() );
1582
1583 } // End: Create ShapeFactor ParamHistFunc
1584
1585 paramHistFuncList.push_back(paramHist);
1586 shapeFactorNameList.push_back(funcName);
1587
1588 } // End loop over ShapeFactor Systematics
1589
1590 // Now that we have the right ShapeFactor,
1591 // we multiply the expected function
1592
1593 //std::string shapeFactorNodeName = syst_x_expectedPrefix + "_x_" + funcName;
1594 // Dynamically build the name as a long product
1595 std::string shapeFactorNodeName = syst_x_expectedPrefix;
1596 for( unsigned int i=0; i < shapeFactorNameList.size(); ++i) {
1597 shapeFactorNodeName += "_x_" + shapeFactorNameList.at(i);
1598 }
1599
1600 RooAbsReal* expFunc = (RooAbsReal*) proto->function( syst_x_expectedPrefix.c_str() );
1601 RooArgSet nodesForProduct(*expFunc);
1602 for( unsigned int i=0; i < paramHistFuncList.size(); ++i) {
1603 nodesForProduct.add( *paramHistFuncList.at(i) );
1604 }
1605 //RooProduct nodeWithShapeFactor(shapeFactorNodeName.c_str(),
1606 // shapeFactorNodeName.c_str(),
1607 //RooArgSet(*paramHist, *expFunc) );
1608 RooProduct nodeWithShapeFactor(shapeFactorNodeName.c_str(),
1609 shapeFactorNodeName.c_str(),
1610 nodesForProduct );
1611
1612 proto->import( nodeWithShapeFactor, RecycleConflictNodes() );
1613
1614 // Push back the final name of the node
1615 // to be used in the RooRealSumPdf
1616 // (node to be created later)
1617 syst_x_expectedPrefix = nodeWithShapeFactor.GetName();
1618
1619 }
1620 } // End: if ShapeFactorName!=""
1621
1622
1623 ////////////////////////////////////////
1624 // Create a ShapeSys for this channel //
1625 ////////////////////////////////////////
1626
1627 if( sample.GetShapeSysList().size() != 0 ) {
1628
1629 if( fObsNameVec.size() > 3 ) {
1630 std::cout << "Cannot include Stat Error for histograms of more than 3 dimensions."
1631 << std::endl;
1632 throw hf_exc();
1633 } else {
1634
1635 // List of ShapeSys ParamHistFuncs
1636 std::vector<string> ShapeSysNames;
1637
1638 for( unsigned int i = 0; i < sample.GetShapeSysList().size(); ++i) {
1639
1640 // Create the ParamHistFunc's
1641 // Create their constraint terms and add them
1642 // to the list of constraint terms
1643
1644 // Create a single RooProduct over all of these
1645 // paramHistFunc's
1646
1647 // Send the name of that product to the RooRealSumPdf
1648
1649 RooStats::HistFactory::ShapeSys& shapeSys = sample.GetShapeSysList().at(i);
1650
1651 std::cout << "Sample: " << sample.GetName() << " in channel: " << channel_name
1652 << " to include a ShapeSys." << std::endl;
1653
1654 std::string funcName = channel_name + "_" + shapeSys.GetName() + "_ShapeSys";
1655 ShapeSysNames.push_back( funcName );
1656 ParamHistFunc* paramHist = (ParamHistFunc*) proto->function( funcName.c_str() );
1657 if( paramHist == NULL ) {
1658
1659 //std::string funcParams = "gamma_" + it->shapeFactorName;
1660 //paramHist = CreateParamHistFunc( proto, fObsNameVec, funcParams, funcName );
1661
1662 RooArgList observables;
1663 std::vector<std::string>::iterator itr = fObsNameVec.begin();
1664 for(; itr!=fObsNameVec.end(); ++itr ) {
1665 observables.add( *proto->var(itr->c_str()) );
1666 }
1667
1668 // Create the Parameters
1669 std::string funcParams = "gamma_" + shapeSys.GetName();
1670 RooArgList shapeFactorParams = ParamHistFunc::createParamSet(*proto,
1671 funcParams.c_str(),
1672 observables, 0, 10);
1673
1674 // Create the Function
1675 ParamHistFunc shapeFactorFunc( funcName.c_str(), funcName.c_str(),
1676 observables, shapeFactorParams );
1677
1678 proto->import( shapeFactorFunc, RecycleConflictNodes() );
1679 paramHist = (ParamHistFunc*) proto->function( funcName.c_str() );
1680
1681 } // End: Create ShapeFactor ParamHistFunc
1682
1683 // Create the constraint terms and add
1684 // them to the workspace (proto)
1685 // as well as the list of constraint terms (constraintTermNames)
1686
1687 // The syst should be a fractional error
1688 const TH1* shapeErrorHist = shapeSys.GetErrorHist();
1689
1690 // Constraint::Type shapeConstraintType = Constraint::Gaussian;
1691 Constraint::Type systype = shapeSys.GetConstraintType();
1692 if( systype == Constraint::Gaussian) {
1693 systype = Constraint::Gaussian;
1694 }
1695 if( systype == Constraint::Poisson ) {
1696 systype = Constraint::Poisson;
1697 }
1698
1699 Double_t minShapeUncertainty = 0.0;
1700 RooArgList shapeConstraints = createStatConstraintTerms(proto, constraintTermNames,
1701 *paramHist, shapeErrorHist,
1702 systype,
1703 minShapeUncertainty);
1704
1705 } // End: Loop over ShapeSys vector in this EstimateSummary
1706
1707 // Now that we have the list of ShapeSys ParamHistFunc names,
1708 // we create the total RooProduct
1709 // we multiply the expected functio
1710
1711 std::string NodeName = syst_x_expectedPrefix;
1712 RooArgList ShapeSysForNode;
1713 RooAbsReal* expFunc = (RooAbsReal*) proto->function( syst_x_expectedPrefix.c_str() );
1714 ShapeSysForNode.add( *expFunc );
1715 for( unsigned int i = 0; i < ShapeSysNames.size(); ++i ) {
1716 std::string ShapeSysName = ShapeSysNames.at(i);
1717 ShapeSysForNode.add( *proto->function(ShapeSysName.c_str()) );
1718 NodeName = NodeName + "_x_" + ShapeSysName;
1719 }
1720
1721 // Create the name for this NEW Node
1722 RooProduct nodeWithShapeFactor(NodeName.c_str(), NodeName.c_str(), ShapeSysForNode );
1723 proto->import( nodeWithShapeFactor, RecycleConflictNodes() );
1724
1725 // Push back the final name of the node
1726 // to be used in the RooRealSumPdf
1727 // (node to be created later)
1728 syst_x_expectedPrefix = nodeWithShapeFactor.GetName();
1729
1730 } // End: NumObsVar == 1
1731
1732 } // End: GetShapeSysList.size() != 0
1733
1734 // Append the name of the "node"
1735 // that is to be summed with the
1736 // RooRealSumPdf
1737 syst_x_expectedPrefixNames.push_back(syst_x_expectedPrefix);
1738
1739 // GHL: This was pretty confusing before,
1740 // hopefully using the measurement directly
1741 // will improve it
1742 if( sample.GetNormalizeByTheory() ) {
1743 normalizationNames.push_back( "Lumi" );
1744 }
1745 else {
1746 TString lumiParamString;
1747 lumiParamString += measurement.GetLumi();
1748 lumiParamString.ReplaceAll(' ', TString());
1749 normalizationNames.push_back(lumiParamString.Data());
1750 }
1751
1752 } // END: Loop over EstimateSummaries
1753 // proto->Print();
1754
1755 // If a non-zero number of samples call for
1756 // Stat Uncertainties, create the statFactor functions
1757 if( statHistPairs.size() > 0 ) {
1758
1759 // Create the histogram of (binwise)
1760 // stat uncertainties:
1761 TH1* fracStatError = MakeScaledUncertaintyHist( statNodeName + "_RelErr", statHistPairs );
1762 if( fracStatError == NULL ) {
1763 std::cout << "Error: Failed to make ScaledUncertaintyHist for: "
1764 << statNodeName << std::endl;
1765 throw hf_exc();
1766 }
1767
1768 // Using this TH1* of fractinal stat errors,
1769 // create a set of constraint terms:
1770 ParamHistFunc* chanStatUncertFunc = (ParamHistFunc*) proto->function( statFuncName.c_str() );
1771 std::cout << "About to create Constraint Terms from: "
1772 << chanStatUncertFunc->GetName()
1773 << " params: " << chanStatUncertFunc->paramList()
1774 << std::endl;
1775
1776 // Get the constraint type and the
1777 // rel error threshold from the (last)
1778 // EstimateSummary looped over (but all
1779 // should be the same)
1780
1781 // Get the type of StatError constraint from the channel
1782 Constraint::Type statConstraintType = channel.GetStatErrorConfig().GetConstraintType();
1783 if( statConstraintType == Constraint::Gaussian) {
1784 std::cout << "Using Gaussian StatErrors in channel: " << channel.GetName() << std::endl;
1785 }
1786 if( statConstraintType == Constraint::Poisson ) {
1787 std::cout << "Using Poisson StatErrors in channel: " << channel.GetName() << std::endl;
1788 }
1789
1790 double statRelErrorThreshold = channel.GetStatErrorConfig().GetRelErrorThreshold();
1791 RooArgList statConstraints = createStatConstraintTerms(proto, constraintTermNames,
1792 *chanStatUncertFunc, fracStatError,
1793 statConstraintType,
1794 statRelErrorThreshold);
1795
1796
1797 // clean stat hist pair (need to delete second histogram)
1798 for (unsigned int i = 0; i < statHistPairs.size() ; ++i )
1799 delete statHistPairs[i].second;
1800
1801 statHistPairs.clear();
1802 //delete also histogram of stat uncertainties created in MakeScaledUncertaintyHist
1803 delete fracStatError;
1804
1805 } // END: Loop over stat Hist Pairs
1806
1807
1808 ///////////////////////////////////
1809 // for ith bin calculate totN_i = lumi * sum_j expected_j * syst_j
1810 //MakeTotalExpected(proto,channel_name+"_model",channel_name,"Lumi",fLowBin,fHighBin,
1811 // syst_x_expectedPrefixNames, normalizationNames);
1812 MakeTotalExpected(proto, channel_name+"_model", //channel_name,"Lumi",fLowBin,fHighBin,
1813 syst_x_expectedPrefixNames, normalizationNames);
1814 likelihoodTermNames.push_back(channel_name+"_model");
1815
1816 //////////////////////////////////////
1817 // fix specified parameters
1818 for(unsigned int i=0; i<systToFix.size(); ++i){
1819 RooRealVar* temp = proto->var((systToFix.at(i)).c_str());
1820 if(temp) {
1821 // set the parameter constant
1822 temp->setConstant();
1823
1824 // remove the corresponding auxiliary observable from the global observables
1825 RooRealVar* auxMeas = NULL;
1826 if(systToFix.at(i)=="Lumi"){
1827 auxMeas = proto->var("nominalLumi");
1828 } else {
1829 auxMeas = proto->var(TString::Format("nom_%s",temp->GetName()));
1830 }
1831
1832 if(auxMeas){
1833 const_cast<RooArgSet*>(proto->set("globalObservables"))->remove(*auxMeas);
1834 } else{
1835 cout << "could not corresponding auxiliary measurement "
1836 << TString::Format("nom_%s",temp->GetName()) << endl;
1837 }
1838 } else {
1839 cout << "could not find variable " << systToFix.at(i)
1840 << " could not set it to constant" << endl;
1841 }
1842 }
1843
1844 //////////////////////////////////////
1845 // final proto model
1846 for(unsigned int i=0; i<constraintTermNames.size(); ++i){
1847 RooAbsArg* proto_arg = (proto->arg(constraintTermNames[i].c_str()));
1848 if( proto_arg==NULL ) {
1849 std::cout << "Error: Cannot find arg set: " << constraintTermNames.at(i)
1850 << " in workspace: " << proto->GetName() << std::endl;
1851 throw hf_exc();
1852 }
1853 constraintTerms.add( *proto_arg );
1854 // constraintTerms.add(* proto_arg(proto->arg(constraintTermNames[i].c_str())) );
1855 }
1856 for(unsigned int i=0; i<likelihoodTermNames.size(); ++i){
1857 RooAbsArg* proto_arg = (proto->arg(likelihoodTermNames[i].c_str()));
1858 if( proto_arg==NULL ) {
1859 std::cout << "Error: Cannot find arg set: " << likelihoodTermNames.at(i)
1860 << " in workspace: " << proto->GetName() << std::endl;
1861 throw hf_exc();
1862 }
1863 likelihoodTerms.add( *proto_arg );
1864 }
1865 proto->defineSet("constraintTerms",constraintTerms);
1866 proto->defineSet("likelihoodTerms",likelihoodTerms);
1867 // proto->Print();
1868
1869 // list of observables
1870 RooArgList observables;
1871 std::string observablesStr;
1872
1873 std::vector<std::string>::iterator itr = fObsNameVec.begin();
1874 for(; itr!=fObsNameVec.end(); ++itr ) {
1875 observables.add( *proto->var(itr->c_str()) );
1876 if (!observablesStr.empty()) { observablesStr += ","; }
1877 observablesStr += *itr;
1878 }
1879
1880 // We create two sets, one for backwards compatability
1881 // The other to make a consistent naming convention
1882 // between individual channels and the combined workspace
1883 proto->defineSet("observables", TString::Format("%s",observablesStr.c_str()));
1884 proto->defineSet("observablesSet", TString::Format("%s",observablesStr.c_str()));
1885
1886 // Create the ParamHistFunc
1887 // after observables have been made
1888 cout <<"-----------------------------------------"<<endl;
1889 cout <<"import model into workspace" << endl;
1890
1891 auto model = make_unique<RooProdPdf>(
1892 ("model_"+channel_name).c_str(), // MB : have changed this into conditional pdf. Much faster for toys!
1893 "product of Poissons accross bins for a single channel",
1894 constraintTerms, Conditional(likelihoodTerms,observables));
1895 proto->import(*model,RecycleConflictNodes());
1896
1897 proto_config->SetPdf(*model);
1898 proto_config->SetObservables(observables);
1899 proto_config->SetGlobalObservables(*proto->set("globalObservables"));
1900 // proto->writeToFile(("results/model_"+channel+".root").c_str());
1901 // fill out nuisance parameters in model config
1902 // proto_config->GuessObsAndNuisance(*proto->data("asimovData"));
1903 proto->import(*proto_config,proto_config->GetName());
1904 proto->importClassCode();
1905
1906 ///////////////////////////
1907 // make data sets
1908 // THis works and is natural, but the memory size of the simultaneous dataset grows exponentially with channels
1909 const char* weightName="weightVar";
1910 proto->factory(TString::Format("%s[0,-1e10,1e10]",weightName));
1911 proto->defineSet("obsAndWeight",TString::Format("%s,%s",weightName,observablesStr.c_str()));
1912
1913 /* Old code for generating the asimov
1914 Asimov generation is now done later...
1915
1916 RooAbsData* asimov_data = model->generateBinned(observables,ExpectedData());
1917
1918 /// Asimov dataset
1919 RooDataSet* asimovDataUnbinned = new RooDataSet("asimovData","",*proto->set("obsAndWeight"),weightName);
1920 for(int i=0; i<asimov_data->numEntries(); ++i){
1921 asimov_data->get(i)->Print("v");
1922 //cout << "GREPME : " << i << " " << data->weight() <<endl;
1923 asimovDataUnbinned->add( *asimov_data->get(i), asimov_data->weight() );
1924 }
1925 proto->import(*asimovDataUnbinned);
1926 */
1927
1928 // New Asimov Generation: Use the code in the Asymptotic calculator
1929 // Need to get the ModelConfig...
1930 unique_ptr<RooAbsData> asimov_dataset(AsymptoticCalculator::GenerateAsimovData(*model, observables));
1931 proto->import(dynamic_cast<RooDataSet&>(*asimov_dataset), Rename("asimovData"));
1932
1933 // GHL: Determine to use data if the hist isn't 'NULL'
1934 if(channel.GetData().GetHisto() != NULL) {
1935
1936 Data& data = channel.GetData();
1937 TH1* mnominal = data.GetHisto();
1938 if( !mnominal ) {
1939 std::cout << "Error: Data histogram for channel: " << channel.GetName()
1940 << " is NULL" << std::endl;
1941 throw hf_exc();
1942 }
1943
1944 // THis works and is natural, but the memory size of the simultaneous dataset grows exponentially with channels
1945 auto obsDataUnbinned = make_unique<RooDataSet>("obsData","",*proto->set("obsAndWeight"),weightName);
1946
1947
1948 ConfigureHistFactoryDataset( obsDataUnbinned.get(), mnominal,
1949 proto, fObsNameVec );
1950
1951 /*
1952 //ES// TH1* mnominal = summary.at(0).nominal;
1953 TH1* mnominal = data.GetHisto();
1954 TAxis* ax = mnominal->GetXaxis();
1955 TAxis* ay = mnominal->GetYaxis();
1956 TAxis* az = mnominal->GetZaxis();
1957
1958 for (int i=1; i<=ax->GetNbins(); ++i) { // 1 or more dimension
1959 Double_t xval = ax->GetBinCenter(i);
1960 proto->var( fObsNameVec[0].c_str() )->setVal( xval );
1961 if (fObsNameVec.size()==1) {
1962 Double_t fval = mnominal->GetBinContent(i);
1963 obsDataUnbinned->add( *proto->set("obsAndWeight"), fval );
1964 } else { // 2 or more dimensions
1965 for (int j=1; j<=ay->GetNbins(); ++j) {
1966 Double_t yval = ay->GetBinCenter(j);
1967 proto->var( fObsNameVec[1].c_str() )->setVal( yval );
1968 if (fObsNameVec.size()==2) {
1969 Double_t fval = mnominal->GetBinContent(i,j);
1970 obsDataUnbinned->add( *proto->set("obsAndWeight"), fval );
1971 } else { // 3 dimensions
1972 for (int k=1; k<=az->GetNbins(); ++k) {
1973 Double_t zval = az->GetBinCenter(k);
1974 proto->var( fObsNameVec[2].c_str() )->setVal( zval );
1975 Double_t fval = mnominal->GetBinContent(i,j,k);
1976 obsDataUnbinned->add( *proto->set("obsAndWeight"), fval );
1977 }
1978 }
1979 }
1980 }
1981 }
1982 */
1983
1984 proto->import(*obsDataUnbinned);
1985 } // End: Has non-null 'data' entry
1986
1987
1988 for(unsigned int i=0; i < channel.GetAdditionalData().size(); ++i) {
1989
1990 Data& data = channel.GetAdditionalData().at(i);
1991 std::string dataName = data.GetName();
1992 TH1* mnominal = data.GetHisto();
1993 if( !mnominal ) {
1994 std::cout << "Error: Additional Data histogram for channel: " << channel.GetName()
1995 << " with name: " << dataName << " is NULL" << std::endl;
1996 throw hf_exc();
1997 }
1998
1999 // THis works and is natural, but the memory size of the simultaneous dataset grows exponentially with channels
2000 auto obsDataUnbinned = make_unique<RooDataSet>(dataName.c_str(), dataName.c_str(),
2001 *proto->set("obsAndWeight"), weightName);
2002
2003 ConfigureHistFactoryDataset( obsDataUnbinned.get(), mnominal,
2004 proto, fObsNameVec );
2005
2006 /*
2007 //ES// TH1* mnominal = summary.at(0).nominal;
2008 TH1* mnominal = data.GetHisto();
2009 TAxis* ax = mnominal->GetXaxis();
2010 TAxis* ay = mnominal->GetYaxis();
2011 TAxis* az = mnominal->GetZaxis();
2012
2013 for (int i=1; i<=ax->GetNbins(); ++i) { // 1 or more dimension
2014 Double_t xval = ax->GetBinCenter(i);
2015 proto->var( fObsNameVec[0].c_str() )->setVal( xval );
2016 if (fObsNameVec.size()==1) {
2017 Double_t fval = mnominal->GetBinContent(i);
2018 obsDataUnbinned->add( *proto->set("obsAndWeight"), fval );
2019 } else { // 2 or more dimensions
2020 for (int j=1; j<=ay->GetNbins(); ++j) {
2021 Double_t yval = ay->GetBinCenter(j);
2022 proto->var( fObsNameVec[1].c_str() )->setVal( yval );
2023 if (fObsNameVec.size()==2) {
2024 Double_t fval = mnominal->GetBinContent(i,j);
2025 obsDataUnbinned->add( *proto->set("obsAndWeight"), fval );
2026 } else { // 3 dimensions
2027 for (int k=1; k<=az->GetNbins(); ++k) {
2028 Double_t zval = az->GetBinCenter(k);
2029 proto->var( fObsNameVec[2].c_str() )->setVal( zval );
2030 Double_t fval = mnominal->GetBinContent(i,j,k);
2031 obsDataUnbinned->add( *proto->set("obsAndWeight"), fval );
2032 }
2033 }
2034 }
2035 }
2036 }
2037 */
2038
2039 proto->import(*obsDataUnbinned);
2040
2041 } // End: Has non-null 'data' entry
2042
2043 proto->Print();
2044 return proto;
2045 }
2046
2047
2049 TH1* mnominal,
2051 std::vector<std::string> ObsNameVec) {
2052
2053 // Take a RooDataSet and fill it with the entries
2054 // from a TH1*, using the observable names to
2055 // determine the columns
2056
2057 if (ObsNameVec.empty() ) {
2058 Error("ConfigureHistFactoryDataset","Invalid input - return");
2059 return;
2060 }
2061
2062 //ES// TH1* mnominal = summary.at(0).nominal;
2063 // TH1* mnominal = data.GetHisto();
2064 TAxis* ax = mnominal->GetXaxis();
2065 TAxis* ay = mnominal->GetYaxis();
2066 TAxis* az = mnominal->GetZaxis();
2067
2068 for (int i=1; i<=ax->GetNbins(); ++i) { // 1 or more dimension
2069
2070 Double_t xval = ax->GetBinCenter(i);
2071 proto->var( ObsNameVec[0].c_str() )->setVal( xval );
2072
2073 if(ObsNameVec.size()==1) {
2074 Double_t fval = mnominal->GetBinContent(i);
2075 obsDataUnbinned->add( *proto->set("obsAndWeight"), fval );
2076 } else { // 2 or more dimensions
2077
2078 for(int j=1; j<=ay->GetNbins(); ++j) {
2079 Double_t yval = ay->GetBinCenter(j);
2080 proto->var( ObsNameVec[1].c_str() )->setVal( yval );
2081
2082 if(ObsNameVec.size()==2) {
2083 Double_t fval = mnominal->GetBinContent(i,j);
2084 obsDataUnbinned->add( *proto->set("obsAndWeight"), fval );
2085 } else { // 3 dimensions
2086
2087 for(int k=1; k<=az->GetNbins(); ++k) {
2088 Double_t zval = az->GetBinCenter(k);
2089 proto->var( ObsNameVec[2].c_str() )->setVal( zval );
2090 Double_t fval = mnominal->GetBinContent(i,j,k);
2091 obsDataUnbinned->add( *proto->set("obsAndWeight"), fval );
2092 }
2093 }
2094 }
2095 }
2096 }
2097 }
2098
2100 {
2101 fObsNameVec.clear();
2102
2103 // determine histogram dimensionality
2104 unsigned int histndim(1);
2105 std::string classname = hist->ClassName();
2106 if (classname.find("TH1")==0) { histndim=1; }
2107 else if (classname.find("TH2")==0) { histndim=2; }
2108 else if (classname.find("TH3")==0) { histndim=3; }
2109
2110 for ( unsigned int idx=0; idx<histndim; ++idx ) {
2111 if (idx==0) { fObsNameVec.push_back("x"); }
2112 if (idx==1) { fObsNameVec.push_back("y"); }
2113 if (idx==2) { fObsNameVec.push_back("z"); }
2114 }
2115 }
2116
2117
2118 RooWorkspace* HistoToWorkspaceFactoryFast::MakeCombinedModel(vector<string> ch_names, vector<RooWorkspace*> chs)
2119 {
2120
2121
2122 // check first the inputs (see JIRA-6890)
2123 if (ch_names.empty() || chs.empty() ) {
2124 Error("MakeCombinedModel","Input vectors are empty - return a nullptr");
2125 return 0;
2126 }
2127 if (chs.size() < ch_names.size() ) {
2128 Error("MakeCombinedModel","Input vector of workspace has an invalid size - return a nullptr");
2129 return 0;
2130 }
2131
2132 //
2133 /// These things were used for debugging. Maybe useful in the future
2134 //
2135
2136 map<string, RooAbsPdf*> pdfMap;
2137 vector<RooAbsPdf*> models;
2138 stringstream ss;
2139
2140 RooArgList obsList;
2141 for(unsigned int i = 0; i< ch_names.size(); ++i){
2142 ModelConfig * config = (ModelConfig *) chs[i]->obj("ModelConfig");
2143 obsList.add(*config->GetObservables());
2144 }
2145 cout <<"full list of observables:"<<endl;
2146 obsList.Print();
2147
2148 RooArgSet globalObs;
2149 for(unsigned int i = 0; i< ch_names.size(); ++i){
2150 string channel_name=ch_names[i];
2151
2152 if (ss.str().empty()) ss << channel_name ;
2153 else ss << ',' << channel_name ;
2154 RooWorkspace * ch=chs[i];
2155
2156 RooAbsPdf* model = ch->pdf(("model_"+channel_name).c_str());
2157 if(!model) cout <<"failed to find model for channel"<<endl;
2158 // cout << "int = " << model->createIntegral(*obsN)->getVal() << endl;;
2159 models.push_back(model);
2160 globalObs.add(*ch->set("globalObservables"));
2161
2162 // constrainedParams->add( * ch->set("constrainedParams") );
2163 pdfMap[channel_name]=model;
2164 }
2165 //constrainedParams->Print();
2166
2167 cout << "\n\n------------------\n Entering combination" << endl;
2168 RooWorkspace* combined = new RooWorkspace("combined");
2169 // RooWorkspace* combined = chs[0];
2170
2171
2172 RooCategory* channelCat = (RooCategory*) combined->factory(("channelCat["+ss.str()+"]").c_str());
2173 RooSimultaneous * simPdf= new RooSimultaneous("simPdf","",pdfMap, *channelCat);
2174 ModelConfig * combined_config = new ModelConfig("ModelConfig", combined);
2175 combined_config->SetWorkspace(*combined);
2176 // combined_config->SetNuisanceParameters(*constrainedParams);
2177
2178 combined->import(globalObs);
2179 combined->defineSet("globalObservables",globalObs);
2180 combined_config->SetGlobalObservables(*combined->set("globalObservables"));
2181
2182
2183 ////////////////////////////////////////////
2184 // Make toy simultaneous dataset
2185 cout <<"-----------------------------------------"<<endl;
2186 cout << "create toy data for " << ss.str() << endl;
2187
2188
2189 // now with weighted datasets
2190 // First Asimov
2191 //RooDataSet * simData=NULL;
2192 combined->factory("weightVar[0,-1e10,1e10]");
2193 obsList.add(*combined->var("weightVar"));
2194
2195 // Loop over channels and create the asimov
2196 /*
2197 for(unsigned int i = 0; i< ch_names.size(); ++i){
2198 cout << "merging data for channel " << ch_names[i].c_str() << endl;
2199 RooDataSet * tempData=new RooDataSet(ch_names[i].c_str(),"", obsList, Index(*channelCat),
2200 WeightVar("weightVar"),
2201 Import(ch_names[i].c_str(),*(RooDataSet*)chs[i]->data("asimovData")));
2202 if(simData){
2203 simData->append(*tempData);
2204 delete tempData;
2205 }else{
2206 simData = tempData;
2207 }
2208 }
2209
2210 if (simData) combined->import(*simData,Rename("asimovData"));
2211 */
2212 RooDataSet* asimov_combined = (RooDataSet*) AsymptoticCalculator::GenerateAsimovData(*simPdf,
2213 obsList);
2214 if( asimov_combined ) {
2215 combined->import( *asimov_combined, Rename("asimovData"));
2216 }
2217 else {
2218 std::cout << "Error: Failed to create combined asimov dataset" << std::endl;
2219 throw hf_exc();
2220 }
2221 delete asimov_combined;
2222
2223 // Now merge the observable datasets across the channels
2224 if(chs[0]->data("obsData") != NULL) {
2225 MergeDataSets(combined, chs, ch_names, "obsData", obsList, channelCat);
2226 }
2227
2228 /*
2229 if(chs[0]->data("obsData") != NULL){
2230 RooDataSet * simData=NULL;
2231 //simData=NULL;
2232
2233 // Loop through channels, get their individual datasets,
2234 // and add them to the combined dataset
2235 for(unsigned int i = 0; i< ch_names.size(); ++i){
2236 cout << "merging data for channel " << ch_names[i].c_str() << endl;
2237
2238 RooDataSet* obsDataInChannel = (RooDataSet*) chs[i]->data("obsData");
2239 RooDataSet * tempData = new RooDataSet(ch_names[i].c_str(),"", obsList, Index(*channelCat),
2240 WeightVar("weightVar"),
2241 Import(ch_names[i].c_str(),*obsDataInChannel));
2242 // *(RooDataSet*) chs[i]->data("obsData")));
2243 if(simData) {
2244 simData->append(*tempData);
2245 delete tempData;
2246 }
2247 else {
2248 simData = tempData;
2249 }
2250 } // End Loop Over Channels
2251
2252 // Check that we successfully created the dataset
2253 // and import it into the workspace
2254 if(simData) {
2255 combined->import(*simData, Rename("obsData"));
2256 }
2257 else {
2258 std::cout << "Error: Unable to merge observable datasets" << std::endl;
2259 throw hf_exc();
2260 }
2261
2262 } // End 'if' on data != NULL
2263 */
2264
2265 // Now, create any additional Asimov datasets that
2266 // are configured in the measurement
2267
2268
2269 // obsList.Print();
2270 // combined->import(obsList);
2271 // combined->Print();
2272
2273 obsList.add(*channelCat);
2274 combined->defineSet("observables",obsList);
2275 combined_config->SetObservables(*combined->set("observables"));
2276
2277 combined->Print();
2278
2279 cout << "\n\n----------------\n Importing combined model" << endl;
2280 combined->import(*simPdf,RecycleConflictNodes());
2281 //combined->import(*simPdf, RenameVariable("SigXsecOverSM","SigXsecOverSM_comb"));
2282 // cout << "check pointer " << simPdf << endl;
2283 // cout << "check val " << simPdf->getVal() << endl;
2284
2285 std::map< std::string, double>::iterator param_itr = fParamValues.begin();
2286 for( ; param_itr != fParamValues.end(); ++param_itr ){
2287 // make sure they are fixed
2288 std::string paramName = param_itr->first;
2289 double paramVal = param_itr->second;
2290
2291 RooRealVar* temp = combined->var( paramName.c_str() );
2292 if(temp) {
2293 temp->setVal( paramVal );
2294 cout <<"setting " << paramName << " to the value: " << paramVal << endl;
2295 } else
2296 cout << "could not find variable " << paramName << " could not set its value" << endl;
2297 }
2298
2299
2300 for(unsigned int i=0; i<fSystToFix.size(); ++i){
2301 // make sure they are fixed
2302 RooRealVar* temp = combined->var((fSystToFix.at(i)).c_str());
2303 if(temp) {
2304 temp->setConstant();
2305 cout <<"setting " << fSystToFix.at(i) << " constant" << endl;
2306 } else
2307 cout << "could not find variable " << fSystToFix.at(i) << " could not set it to constant" << endl;
2308 }
2309
2310 ///
2311 /// writing out the model in graphViz
2312 ///
2313 // RooAbsPdf* customized=combined->pdf("simPdf");
2314 //combined_config->SetPdf(*customized);
2315 combined_config->SetPdf(*simPdf);
2316 // combined_config->GuessObsAndNuisance(*simData);
2317 // customized->graphVizTree(("results/"+fResultsPrefixStr.str()+"_simul.dot").c_str());
2318 combined->import(*combined_config,combined_config->GetName());
2319 combined->importClassCode();
2320 // combined->writeToFile("results/model_combined.root");
2321
2322 //clean up
2323 delete combined_config;
2324 delete simPdf;
2325
2326 return combined;
2327 }
2328
2329
2331 std::vector<RooWorkspace*> wspace_vec,
2332 std::vector<std::string> channel_names,
2333 std::string dataSetName,
2334 RooArgList obsList,
2335 RooCategory* channelCat) {
2336
2337 // Create the total dataset
2338 RooDataSet* simData=NULL;
2339
2340 // Loop through channels, get their individual datasets,
2341 // and add them to the combined dataset
2342 for(unsigned int i = 0; i< channel_names.size(); ++i){
2343
2344 // Grab the dataset for the existing channel
2345 std::cout << "Merging data for channel " << channel_names[i].c_str() << std::endl;
2346 RooDataSet* obsDataInChannel = (RooDataSet*) wspace_vec[i]->data(dataSetName.c_str());
2347 if( !obsDataInChannel ) {
2348 std::cout << "Error: Can't find DataSet: " << dataSetName
2349 << " in channel: " << channel_names.at(i)
2350 << std::endl;
2351 throw hf_exc();
2352 }
2353
2354 // Create the new Dataset
2355 RooDataSet * tempData = new RooDataSet(channel_names[i].c_str(),"",
2356 obsList, Index(*channelCat),
2357 WeightVar("weightVar"),
2358 Import(channel_names[i].c_str(),*obsDataInChannel));
2359 if(simData) {
2360 simData->append(*tempData);
2361 delete tempData;
2362 }
2363 else {
2364 simData = tempData;
2365 }
2366 } // End Loop Over Channels
2367
2368 // Check that we successfully created the dataset
2369 // and import it into the workspace
2370 if(simData) {
2371 combined->import(*simData, Rename(dataSetName.c_str()));
2372 delete simData;
2373 simData = (RooDataSet*) combined->data(dataSetName.c_str() );
2374 }
2375 else {
2376 std::cout << "Error: Unable to merge observable datasets" << std::endl;
2377 throw hf_exc();
2378 }
2379
2380 return simData;
2381
2382 }
2383
2384
2386
2387 // Take a nominal TH1* and create
2388 // a TH1 representing the binwise
2389 // errors (taken from the nominal TH1)
2390
2391 TH1* ErrorHist = (TH1*) Nominal->Clone( Name.c_str() );
2392 ErrorHist->Reset();
2393
2394 Int_t numBins = Nominal->GetNbinsX()*Nominal->GetNbinsY()*Nominal->GetNbinsZ();
2395 Int_t binNumber = 0;
2396
2397 // Loop over bins
2398 for( Int_t i_bin = 0; i_bin < numBins; ++i_bin) {
2399
2400 binNumber++;
2401 // Ignore underflow / overflow
2402 while( Nominal->IsBinUnderflow(binNumber) || Nominal->IsBinOverflow(binNumber) ){
2403 binNumber++;
2404 }
2405
2406 Double_t histError = Nominal->GetBinError( binNumber );
2407
2408 // Check that histError != NAN
2409 if( histError != histError ) {
2410 std::cout << "Warning: In histogram " << Nominal->GetName()
2411 << " bin error for bin " << i_bin
2412 << " is NAN. Not using Error!!!"
2413 << std::endl;
2414 throw hf_exc();
2415 //histError = sqrt( histContent );
2416 //histError = 0;
2417 }
2418
2419 // Check that histError ! < 0
2420 if( histError < 0 ) {
2421 std::cout << "Warning: In histogram " << Nominal->GetName()
2422 << " bin error for bin " << binNumber
2423 << " is < 0. Setting Error to 0"
2424 << std::endl;
2425 //histError = sqrt( histContent );
2426 histError = 0;
2427 }
2428
2429 ErrorHist->SetBinContent( binNumber, histError );
2430
2431 }
2432
2433 return ErrorHist;
2434
2435 }
2436
2437 TH1* HistoToWorkspaceFactoryFast::MakeScaledUncertaintyHist( const std::string& Name, std::vector< std::pair<const TH1*, const TH1*> > HistVec ) {
2438
2439 // Take a list of < nominal, absolError > TH1* pairs
2440 // and construct a single histogram representing the
2441 // total fractional error as:
2442
2443 // UncertInQuad(bin i) = Sum: absolUncert*absolUncert
2444 // Total(bin i) = Sum: Value
2445 //
2446 // TotalFracError(bin i) = Sqrt( UncertInQuad(i) ) / TotalBin(i)
2447
2448
2449 unsigned int numHists = HistVec.size();
2450
2451 if( numHists == 0 ) {
2452 std::cout << "Warning: Empty Hist Vector, cannot create total uncertainty" << std::endl;
2453 return NULL;
2454 }
2455
2456 const TH1* HistTemplate = HistVec.at(0).first;
2457 Int_t numBins = HistTemplate->GetNbinsX()*HistTemplate->GetNbinsY()*HistTemplate->GetNbinsZ();
2458
2459 // Check that all histograms
2460 // have the same bins
2461 for( unsigned int i = 0; i < HistVec.size(); ++i ) {
2462
2463 const TH1* nominal = HistVec.at(i).first;
2464 const TH1* error = HistVec.at(i).second;
2465
2466 if( nominal->GetNbinsX()*nominal->GetNbinsY()*nominal->GetNbinsZ() != numBins ) {
2467 std::cout << "Error: Provided hists have unequal bins" << std::endl;
2468 return NULL;
2469 }
2470 if( error->GetNbinsX()*error->GetNbinsY()*error->GetNbinsZ() != numBins ) {
2471 std::cout << "Error: Provided hists have unequal bins" << std::endl;
2472 return NULL;
2473 }
2474 }
2475
2476 std::vector<double> TotalBinContent( numBins, 0.0);
2477 std::vector<double> HistErrorsSqr( numBins, 0.0);
2478
2479 Int_t binNumber = 0;
2480
2481 // Loop over bins
2482 for( Int_t i_bins = 0; i_bins < numBins; ++i_bins) {
2483
2484 binNumber++;
2485 while( HistTemplate->IsBinUnderflow(binNumber) || HistTemplate->IsBinOverflow(binNumber) ){
2486 binNumber++;
2487 }
2488
2489 for( unsigned int i_hist = 0; i_hist < numHists; ++i_hist ) {
2490
2491 const TH1* nominal = HistVec.at(i_hist).first;
2492 const TH1* error = HistVec.at(i_hist).second;
2493
2494 //Int_t binNumber = i_bins + 1;
2495
2496 Double_t histValue = nominal->GetBinContent( binNumber );
2497 Double_t histError = error->GetBinContent( binNumber );
2498 /*
2499 std::cout << " Getting Bin content for Stat Uncertainty"
2500 << " Nom name: " << nominal->GetName()
2501 << " Err name: " << error->GetName()
2502 << " HistNumber: " << i_hist << " bin: " << binNumber
2503 << " Value: " << histValue << " Error: " << histError
2504 << std::endl;
2505 */
2506
2507 if( histError != histError ) {
2508 std::cout << "Warning: In histogram " << error->GetName()
2509 << " bin error for bin " << binNumber
2510 << " is NAN. Not using error!!"
2511 << std::endl;
2512 throw hf_exc();
2513 //histError = 0;
2514 }
2515
2516 TotalBinContent.at(i_bins) += histValue;
2517 HistErrorsSqr.at(i_bins) += histError*histError; // Add in quadrature
2518
2519 }
2520 }
2521
2522 binNumber = 0;
2523
2524 // Creat the output histogram
2525 TH1* ErrorHist = (TH1*) HistTemplate->Clone( Name.c_str() );
2526 ErrorHist->Reset();
2527
2528 // Fill the output histogram
2529 for( Int_t i = 0; i < numBins; ++i) {
2530
2531 // Int_t binNumber = i + 1;
2532 binNumber++;
2533 while( ErrorHist->IsBinUnderflow(binNumber) || ErrorHist->IsBinOverflow(binNumber) ){
2534 binNumber++;
2535 }
2536
2537 Double_t ErrorsSqr = HistErrorsSqr.at(i);
2538 Double_t TotalVal = TotalBinContent.at(i);
2539
2540 if( TotalVal <= 0 ) {
2541 std::cout << "Warning: Sum of histograms for bin: " << binNumber
2542 << " is <= 0. Setting error to 0"
2543 << std::endl;
2544
2545 ErrorHist->SetBinContent( binNumber, 0.0 );
2546 continue;
2547 }
2548
2549 Double_t RelativeError = sqrt(ErrorsSqr) / TotalVal;
2550
2551 // If we otherwise get a NAN
2552 // it's an error
2553 if( RelativeError != RelativeError ) {
2554 std::cout << "Error: bin " << i << " error is NAN" << std::endl;
2555 std::cout << " HistErrorsSqr: " << ErrorsSqr
2556 << " TotalVal: " << TotalVal
2557 << std::endl;
2558 throw hf_exc();
2559 }
2560
2561 // 0th entry in vector is
2562 // the 1st bin in TH1
2563 // (we ignore underflow)
2564
2565 ErrorHist->SetBinContent( binNumber, RelativeError );
2566
2567 std::cout << "Making Total Uncertainty for bin " << binNumber
2568 << " Error = " << sqrt(ErrorsSqr)
2569 << " Val = " << TotalVal
2570 << " RelativeError = " << RelativeError
2571 << std::endl;
2572
2573 }
2574
2575 return ErrorHist;
2576
2577}
2578
2579
2580
2582 createStatConstraintTerms( RooWorkspace* proto, vector<string>& constraintTermNames,
2583 ParamHistFunc& paramHist, const TH1* uncertHist,
2584 Constraint::Type type, Double_t minSigma ) {
2585
2586
2587 // Take a RooArgList of RooAbsReal's and
2588 // create N constraint terms (one for
2589 // each gamma) whose relative uncertainty
2590 // is the value of the ith RooAbsReal
2591 //
2592 // The integer "type" controls the type
2593 // of constraint term:
2594 //
2595 // type == 0 : NONE
2596 // type == 1 : Gaussian
2597 // type == 2 : Poisson
2598 // type == 3 : LogNormal
2599
2600 RooArgList ConstraintTerms;
2601
2602 RooArgList paramSet = paramHist.paramList();
2603
2604 // Must get the full size of the TH1
2605 // (No direct method to do this...)
2606 Int_t numBins = uncertHist->GetNbinsX()*uncertHist->GetNbinsY()*uncertHist->GetNbinsZ();
2607 Int_t numParams = paramSet.getSize();
2608 // Int_t numBins = uncertHist->GetNbinsX()*uncertHist->GetNbinsY()*uncertHist->GetNbinsZ();
2609
2610 // Check that there are N elements
2611 // in the RooArgList
2612 if( numBins != numParams ) {
2613 std::cout << "Error: In createStatConstraintTerms, encountered bad number of bins" << std::endl;
2614 std::cout << "Given histogram with " << numBins << " bins,"
2615 << " but require exactly " << numParams << std::endl;
2616 throw hf_exc();
2617 }
2618
2619 Int_t TH1BinNumber = 0;
2620 for( Int_t i = 0; i < paramSet.getSize(); ++i) {
2621
2622 TH1BinNumber++;
2623
2624 while( uncertHist->IsBinUnderflow(TH1BinNumber) || uncertHist->IsBinOverflow(TH1BinNumber) ){
2625 TH1BinNumber++;
2626 }
2627
2628 RooRealVar& gamma = (RooRealVar&) (paramSet[i]);
2629
2630 std::cout << "Creating constraint for: " << gamma.GetName()
2631 << ". Type of constraint: " << type << std::endl;
2632
2633 // Get the sigma from the hist
2634 // (the relative uncertainty)
2635 Double_t sigma = uncertHist->GetBinContent( TH1BinNumber );
2636
2637 // If the sigma is <= 0,
2638 // do cont create the term
2639 if( sigma <= 0 ){
2640 std::cout << "Not creating constraint term for "
2641 << gamma.GetName()
2642 << " because sigma = " << sigma
2643 << " (sigma<=0)"
2644 << " (TH1 bin number = " << TH1BinNumber << ")"
2645 << std::endl;
2646 gamma.setConstant(kTRUE);
2647 continue;
2648 }
2649
2650 // set reasonable ranges for gamma parameters
2651 gamma.setMax( 1 + 5*sigma );
2652 // gamma.setMin( TMath::Max(1. - 5*sigma, 0.) );
2653 gamma.setMin( 0. );
2654
2655 // Make Constraint Term
2656 std::string constrName = string(gamma.GetName()) + "_constraint";
2657 std::string nomName = string("nom_") + gamma.GetName();
2658 std::string sigmaName = string(gamma.GetName()) + "_sigma";
2659 std::string poisMeanName = string(gamma.GetName()) + "_poisMean";
2660
2661 if( type == Constraint::Gaussian ) {
2662
2663 // Type 1 : RooGaussian
2664
2665 // Make sigma
2666
2667 RooConstVar constrSigma( sigmaName.c_str(), sigmaName.c_str(), sigma );
2668 //proto->import( constrSigma, RecycleConflictNodes() );
2669 //proto->import( constrSigma );
2670
2671 // Make "observed" value
2672 RooRealVar constrNom(nomName.c_str(), nomName.c_str(), 1.0,0,10);
2673 constrNom.setConstant( true );
2674
2675 // Make the constraint:
2676 RooGaussian gauss( constrName.c_str(), constrName.c_str(),
2677 constrNom, gamma, constrSigma );
2678
2679 proto->import( gauss, RecycleConflictNodes() );
2680 //proto->import( gauss );
2681
2682 } else if( type == Constraint::Poisson ) {
2683
2684 Double_t tau = 1/sigma/sigma; // this is correct Poisson equivalent to a Gaussian with mean 1 and stdev sigma
2685
2686 // Make nominal "observed" value
2687 RooRealVar constrNom(nomName.c_str(), nomName.c_str(), tau);
2688 constrNom.setMin(0);
2689 constrNom.setConstant( true );
2690
2691 // Make the scaling term
2692 std::string scalingName = string(gamma.GetName()) + "_tau";
2693 RooConstVar poissonScaling( scalingName.c_str(), scalingName.c_str(), tau);
2694
2695 // Make mean for scaled Poisson
2696 RooProduct constrMean( poisMeanName.c_str(), poisMeanName.c_str(), RooArgSet(gamma, poissonScaling) );
2697 //proto->import( constrSigma, RecycleConflictNodes() );
2698 //proto->import( constrSigma );
2699
2700 // Type 2 : RooPoisson
2701 RooPoisson pois(constrName.c_str(), constrName.c_str(), constrNom, constrMean);
2702 pois.setNoRounding(true);
2703 proto->import( pois, RecycleConflictNodes() );
2704
2705 } else {
2706
2707 std::cout << "Error: Did not recognize Stat Error constraint term type: "
2708 << type << " for : " << paramHist.GetName() << std::endl;
2709 throw hf_exc();
2710 }
2711
2712 // If the sigma value is less
2713 // than a supplied threshold,
2714 // set the variable to constant
2715 if( sigma < minSigma ) {
2716 std::cout << "Warning: Bin " << i << " = " << sigma
2717 << " and is < " << minSigma
2718 << ". Setting: " << gamma.GetName() << " to constant"
2719 << std::endl;
2720 gamma.setConstant(kTRUE);
2721 }
2722
2723 constraintTermNames.push_back( constrName );
2724 ConstraintTerms.add( *proto->pdf(constrName.c_str()) );
2725
2726 // Add the "observed" value to the
2727 // list of global observables:
2728 RooArgSet* globalSet = const_cast<RooArgSet*>(proto->set("globalObservables"));
2729
2730 RooRealVar* nomVarInWorkspace = proto->var(nomName.c_str());
2731 if( ! globalSet->contains(*nomVarInWorkspace) ) {
2732 globalSet->add( *nomVarInWorkspace );
2733 }
2734
2735 } // end loop over parameters
2736
2737 return ConstraintTerms;
2738
2739}
2740
2741} // namespace RooStats
2742} // namespace HistFactory
2743
#define alpha_Low
#define alpha_High
int Int_t
Definition: RtypesCore.h:41
double Double_t
Definition: RtypesCore.h:55
const Bool_t kTRUE
Definition: RtypesCore.h:87
#define ClassImp(name)
Definition: Rtypes.h:365
#define R__ASSERT(e)
Definition: TError.h:96
char name[80]
Definition: TGX11.cxx:109
int type
Definition: TGX11.cxx:120
float xmin
Definition: THbookFile.cxx:93
float xmax
Definition: THbookFile.cxx:93
double pow(double, double)
double sqrt(double)
double exp(double)
char * Form(const char *fmt,...)
const char * proto
Definition: civetweb.c:16604
A class which maps the current values of a RooRealVar (or a set of RooRealVars) to one of a number of...
Definition: ParamHistFunc.h:28
void setConstant(bool constant)
const RooArgList & paramList() const
Definition: ParamHistFunc.h:39
static RooArgList createParamSet(RooWorkspace &w, const std::string &, const RooArgList &Vars)
Create the list of RooRealVar parameters which represent the height of the histogram bins.
void setShape(TH1 *shape)
Bool_t setBinIntegrator(RooArgSet &allVars)
void setPositiveDefinite(bool flag=true)
RooAbsArg is the common abstract base class for objects that represent a value (of arbitrary type) an...
Definition: RooAbsArg.h:70
virtual void Print(Option_t *options=0) const
Print TNamed name and title.
Definition: RooAbsArg.h:272
void setAttribute(const Text_t *name, Bool_t value=kTRUE)
Set (default) or clear a named boolean attribute of this object.
Definition: RooAbsArg.cxx:256
Bool_t isConstant() const
Definition: RooAbsArg.h:311
Int_t getSize() const
Bool_t contains(const RooAbsArg &var) const
virtual Bool_t add(const RooAbsArg &var, Bool_t silent=kFALSE)
Add the specified argument to list.
virtual void Print(Option_t *options=0) const
This method must be overridden when a class wants to print itself.
TIterator * createIterator(Bool_t dir=kIterForward) const R__SUGGEST_ALTERNATIVE("begin()
TIterator-style iteration over contained elements.
RooAbsData is the common abstract base class for binned and unbinned datasets.
Definition: RooAbsData.h:37
void setConstant(Bool_t value=kTRUE)
RooAbsReal is the common abstract base class for objects that represent a real value and implements f...
Definition: RooAbsReal.h:53
Double_t getVal(const RooArgSet *normalisationSet=nullptr) const
Evaluate object.
Definition: RooAbsReal.h:81
RooNumIntConfig * specialIntegratorConfig() const
Returns the specialized integrator configuration for this RooAbsReal.
virtual void forceNumInt(Bool_t flag=kTRUE)
Definition: RooAbsReal.h:127
RooArgList is a container object that can hold multiple RooAbsArg objects.
Definition: RooArgList.h:21
RooArgSet is a container object that can hold multiple RooAbsArg objects.
Definition: RooArgSet.h:28
virtual Bool_t add(const RooAbsCollection &col, Bool_t silent=kFALSE)
Add a collection of arguments to this collection by calling add() for each element in the source coll...
Definition: RooArgSet.h:88
RooCategory represents a fundamental (non-derived) discrete value object.
Definition: RooCategory.h:24
virtual Bool_t setLabel(const char *label, Bool_t printError=kTRUE)
Set value by specifying the name of the desired state If printError is set, a message will be printed...
RooConstVar represent a constant real-valued object.
Definition: RooConstVar.h:25
The RooDataHist is a container class to hold N-dimensional binned data.
Definition: RooDataHist.h:40
RooDataSet is a container class to hold unbinned data.
Definition: RooDataSet.h:31
virtual void add(const RooArgSet &row, Double_t weight=1.0, Double_t weightError=0)
Add a data point, with its coordinates specified in the 'data' argset, to the data set.
void append(RooDataSet &data)
Add all data points of given data set to this data set.
RooFitResult is a container class to hold the input and output of a PDF fit to a dataset.
Definition: RooFitResult.h:40
Double_t correlation(const RooAbsArg &par1, const RooAbsArg &par2) const
Definition: RooFitResult.h:117
Plain Gaussian p.d.f.
Definition: RooGaussian.h:25
RooHistFunc implements a real-valued function sampled from a multidimensional histogram.
Definition: RooHistFunc.h:29
Multivariate Gaussian p.d.f.
RooCategory & method2D()
RooCategory & methodND()
RooCategory & method1D()
Poisson pdf.
Definition: RooPoisson.h:19
void setNoRounding(bool flag=kTRUE)
Definition: RooPoisson.h:33
A RooProduct represents the product of a given set of RooAbsReal objects.
Definition: RooProduct.h:32
The class RooRealSumPdf implements a PDF constructed from a sum of functions:
Definition: RooRealSumPdf.h:24
RooRealVar represents a fundamental (non-derived) real valued object.
Definition: RooRealVar.h:36
void setMin(const char *name, Double_t value)
Set minimum of name range to given value.
Definition: RooRealVar.cxx:416
virtual void setVal(Double_t value)
Set value of variable to 'value'.
Definition: RooRealVar.cxx:233
RooSimultaneous facilitates simultaneous fitting of multiple PDFs to subsets of a given dataset.
static RooAbsData * GenerateAsimovData(const RooAbsPdf &pdf, const RooArgSet &observables)
generate the asimov data for the observables (not the global ones) need to deal with the case of a si...
TODO Here, we are missing some documentation.
Definition: Asimov.h:23
std::string GetName()
Definition: Asimov.h:32
void ConfigureWorkspace(RooWorkspace *)
Definition: Asimov.cxx:22
This class encapsulates all information for the statistical interpretation of one experiment.
Definition: Channel.h:26
std::vector< RooStats::HistFactory::Data > & GetAdditionalData()
retrieve vector of additional data objects
Definition: Channel.h:60
void Print(std::ostream &=std::cout)
Definition: Channel.cxx:73
HistFactory::StatErrorConfig & GetStatErrorConfig()
get information about threshold for statistical uncertainties and constraint term
Definition: Channel.h:67
RooStats::HistFactory::Data & GetData()
get data object
Definition: Channel.h:55
std::string GetName()
get name of channel
Definition: Channel.h:39
std::vector< RooStats::HistFactory::Sample > & GetSamples()
get vector of samples for this channel
Definition: Channel.h:71
std::string GetName()
Definition: Data.h:33
Configuration for a constrained, coherent shape variation of affected samples.
Definition: Systematics.h:205
void ConfigureHistFactoryDataset(RooDataSet *obsData, TH1 *nominal, RooWorkspace *proto, std::vector< std::string > obsNameVec)
static void EditSyst(RooWorkspace *proto, const char *pdfNameChar, std::map< std::string, double > gammaSyst, std::map< std::string, double > uniformSyst, std::map< std::string, double > logNormSyst, std::map< std::string, double > noSyst)
RooWorkspace * MakeSingleChannelModel(Measurement &measurement, Channel &channel)
RooWorkspace * MakeSingleChannelWorkspace(Measurement &measurement, Channel &channel)
void ProcessExpectedHisto(const TH1 *hist, RooWorkspace *proto, std::string prefix, std::string productPrefix, std::string systTerm)
void SetObsToExpected(RooWorkspace *proto, std::string obsPrefix, std::string expPrefix, int lowBin, int highBin)
void SetFunctionsToPreprocess(std::vector< std::string > lines)
RooDataSet * MergeDataSets(RooWorkspace *combined, std::vector< RooWorkspace * > wspace_vec, std::vector< std::string > channel_names, std::string dataSetName, RooArgList obsList, RooCategory *channelCat)
TH1 * MakeAbsolUncertaintyHist(const std::string &Name, const TH1 *Hist)
static void ConfigureWorkspaceForMeasurement(const std::string &ModelName, RooWorkspace *ws_single, Measurement &measurement)
RooArgList createStatConstraintTerms(RooWorkspace *proto, std::vector< std::string > &constraintTerms, ParamHistFunc &paramHist, const TH1 *uncertHist, Constraint::Type type, Double_t minSigma)
void AddPoissonTerms(RooWorkspace *proto, std::string prefix, std::string obsPrefix, std::string expPrefix, int lowBin, int highBin, std::vector< std::string > &likelihoodTermNames)
static void PrintCovarianceMatrix(RooFitResult *result, RooArgSet *params, std::string filename)
void MakeTotalExpected(RooWorkspace *proto, std::string totName, std::vector< std::string > &syst_x_expectedPrefixNames, std::vector< std::string > &normByNames)
void AddMultiVarGaussConstraint(RooWorkspace *proto, std::string prefix, int lowBin, int highBin, std::vector< std::string > &likelihoodTermNames)
std::string AddNormFactor(RooWorkspace *proto, std::string &channel, std::string &sigmaEpsilon, Sample &sample, bool doRatio)
TH1 * MakeScaledUncertaintyHist(const std::string &Name, std::vector< std::pair< const TH1 *, const TH1 * > > HistVec)
void AddConstraintTerms(RooWorkspace *proto, Measurement &measurement, std::string prefix, std::string interpName, std::vector< OverallSys > &systList, std::vector< std::string > &likelihoodTermNames, std::vector< std::string > &totSystTermNames)
RooWorkspace * MakeCombinedModel(std::vector< std::string >, std::vector< RooWorkspace * >)
void LinInterpWithConstraint(RooWorkspace *proto, const TH1 *nominal, std::vector< HistoSys >, std::string prefix, std::string productPrefix, std::string systTerm, std::vector< std::string > &likelihoodTermNames)
The RooStats::HistFactory::Measurement class can be used to construct a model by combining multiple R...
Definition: Measurement.h:30
std::map< std::string, double > & GetGammaSyst()
Definition: Measurement.h:121
std::map< std::string, double > & GetLogNormSyst()
Definition: Measurement.h:123
std::map< std::string, double > & GetNoSyst()
Definition: Measurement.h:124
std::vector< std::string > & GetPOIList()
get vector of PoI names
Definition: Measurement.h:51
std::map< std::string, double > & GetUniformSyst()
Definition: Measurement.h:122
std::vector< std::string > & GetConstantParams()
get vector of all constant parameters
Definition: Measurement.h:60
std::vector< RooStats::HistFactory::Channel > & GetChannels()
Definition: Measurement.h:105
std::vector< RooStats::HistFactory::Asimov > & GetAsimovDatasets()
get vector of defined Asimov Datasets
Definition: Measurement.h:79
double GetLumi()
retrieve integrated luminosity
Definition: Measurement.h:88
std::vector< std::string > GetPreprocessFunctions()
Returns a list of defined preprocess function expressions.
Configuration for an un- constrained overall systematic to scale sample normalisations.
Definition: Systematics.h:77
std::string GetName() const
Definition: Systematics.h:84
Configuration for a constrained overall systematic to scale sample normalisations.
Definition: Systematics.h:49
std::string GetName() const
Definition: Systematics.h:56
std::vector< RooStats::HistFactory::OverallSys > & GetOverallSysList()
Definition: Sample.h:109
std::string GetName() const
get name of sample
Definition: Sample.h:83
const TH1 * GetHisto() const
Definition: Sample.cxx:99
RooStats::HistFactory::StatError & GetStatError()
Definition: Sample.h:118
std::vector< RooStats::HistFactory::ShapeFactor > & GetShapeFactorList()
Definition: Sample.h:116
std::vector< RooStats::HistFactory::NormFactor > & GetNormFactorList()
Definition: Sample.h:110
std::vector< RooStats::HistFactory::HistoSys > & GetHistoSysList()
Definition: Sample.h:112
bool GetNormalizeByTheory() const
does the normalization scale with luminosity
Definition: Sample.h:79
std::vector< RooStats::HistFactory::ShapeSys > & GetShapeSysList()
Definition: Sample.h:115
*Un*constrained bin-by-bin variation of affected histogram.
Definition: Systematics.h:267
const TH1 * GetInitialShape() const
Definition: Systematics.h:283
Constrained bin-by-bin variation of affected histogram.
Definition: Systematics.h:225
Constraint::Type GetConstraintType() const
Definition: Systematics.h:257
const TH1 * GetErrorHist() const
Definition: Systematics.h:249
Constraint::Type GetConstraintType() const
Definition: Systematics.h:379
const TH1 * GetErrorHist() const
Definition: Systematics.h:349
ModelConfig is a simple class that holds configuration information specifying how a model should be u...
Definition: ModelConfig.h:30
virtual void SetObservables(const RooArgSet &set)
specify the observables
Definition: ModelConfig.h:140
virtual void SetWorkspace(RooWorkspace &ws)
Definition: ModelConfig.h:66
virtual void SetParametersOfInterest(const RooArgSet &set)
Definition: ModelConfig.h:99
virtual void SetGlobalObservables(const RooArgSet &set)
specify the global observables
Definition: ModelConfig.h:166
const RooArgSet * GetObservables() const
get RooArgSet for observables (return NULL if not existing)
Definition: ModelConfig.h:243
void GuessObsAndNuisance(const RooAbsData &data)
guesses Observables and ParametersOfInterest if not already set
Definition: ModelConfig.cxx:50
virtual void SetPdf(const RooAbsPdf &pdf)
Set the Pdf, add to the the workspace if not already there.
Definition: ModelConfig.h:81
The RooWorkspace is a persistable container for RooFit projects.
Definition: RooWorkspace.h:43
RooAbsData * data(const char *name) const
Retrieve dataset (binned or unbinned) with given name. A null pointer is returned if not found.
RooArgSet allVars() const
Return set with all variable objects.
Bool_t importClassCode(const char *pat="*", Bool_t doReplace=kFALSE)
Inport code of all classes in the workspace that have a class name that matches pattern 'pat' and whi...
void Print(Option_t *opts=0) const
Print contents of the workspace.
Bool_t defineSet(const char *name, const RooArgSet &aset, Bool_t importMissing=kFALSE)
Define a named RooArgSet with given constituents.
Bool_t saveSnapshot(const char *name, const char *paramNames)
Save snapshot of values and attributes (including "Constant") of parameters 'params' If importValues ...
Bool_t loadSnapshot(const char *name)
Load the values and attributes of the parameters in the snapshot saved with the given name.
RooRealVar * var(const char *name) const
Retrieve real-valued variable (RooRealVar) with given name. A null pointer is returned if not found.
Bool_t import(const RooAbsArg &arg, const RooCmdArg &arg1=RooCmdArg(), const RooCmdArg &arg2=RooCmdArg(), const RooCmdArg &arg3=RooCmdArg(), const RooCmdArg &arg4=RooCmdArg(), const RooCmdArg &arg5=RooCmdArg(), const RooCmdArg &arg6=RooCmdArg(), const RooCmdArg &arg7=RooCmdArg(), const RooCmdArg &arg8=RooCmdArg(), const RooCmdArg &arg9=RooCmdArg())
Import a RooAbsArg object, e.g.
RooFactoryWSTool & factory()
Return instance to factory tool.
const RooArgSet * set(const char *name)
Return pointer to previously defined named set with given nmame If no such set is found a null pointe...
TObject * obj(const char *name) const
Return any type of object (RooAbsArg, RooAbsData or generic object) with given name)
RooAbsPdf * pdf(const char *name) const
Retrieve p.d.f (RooAbsPdf) with given name. A null pointer is returned if not found.
Class to manage histogram axis.
Definition: TAxis.h:30
virtual Double_t GetBinCenter(Int_t bin) const
Return center of bin.
Definition: TAxis.cxx:464
Double_t GetXmax() const
Definition: TAxis.h:134
Double_t GetXmin() const
Definition: TAxis.h:133
Int_t GetNbins() const
Definition: TAxis.h:121
The TH1 histogram class.
Definition: TH1.h:56
TAxis * GetZaxis()
Definition: TH1.h:318
virtual Bool_t Multiply(TF1 *f1, Double_t c1=1)
Performs the operation:
Definition: TH1.cxx:5634
virtual Int_t GetNbinsY() const
Definition: TH1.h:293
virtual Double_t GetBinError(Int_t bin) const
Return value of error associated to bin number bin.
Definition: TH1.cxx:8476
virtual Int_t GetNbinsZ() const
Definition: TH1.h:294
virtual void Reset(Option_t *option="")
Reset this histogram: contents, errors, etc.
Definition: TH1.cxx:6700
TAxis * GetXaxis()
Get the behaviour adopted by the object about the statoverflows. See EStatOverflows for more informat...
Definition: TH1.h:316
TObject * Clone(const char *newname=0) const
Make a complete copy of the underlying object.
Definition: TH1.cxx:2664
virtual Int_t GetNbinsX() const
Definition: TH1.h:292
TAxis * GetYaxis()
Definition: TH1.h:317
Bool_t IsBinUnderflow(Int_t bin, Int_t axis=0) const
Return true if the bin is underflow.
Definition: TH1.cxx:5035
Bool_t IsBinOverflow(Int_t bin, Int_t axis=0) const
Return true if the bin is overflow.
Definition: TH1.cxx:5003
virtual void SetBinContent(Int_t bin, Double_t content)
Set bin content see convention for numbering bins in TH1::GetBin In case the bin number is greater th...
Definition: TH1.cxx:8635
virtual void SetName(const char *name)
Change the name of this histogram.
Definition: TH1.cxx:8374
virtual Double_t GetBinContent(Int_t bin) const
Return content of bin number bin.
Definition: TH1.cxx:4882
TObject * Next()
Definition: TCollection.h:249
void Reset()
Definition: TCollection.h:252
virtual const char * GetName() const
Returns name of object.
Definition: TNamed.h:47
virtual const char * GetName() const
Returns name of object.
Definition: TObject.cxx:357
virtual const char * ClassName() const
Returns name of class to which the object belongs.
Definition: TObject.cxx:128
virtual void Error(const char *method, const char *msgfmt,...) const
Issue error message.
Definition: TObject.cxx:880
Stopwatch class.
Definition: TStopwatch.h:28
void Start(Bool_t reset=kTRUE)
Start the stopwatch.
Definition: TStopwatch.cxx:58
Basic string class.
Definition: TString.h:131
const char * Data() const
Definition: TString.h:364
TString & ReplaceAll(const TString &s1, const TString &s2)
Definition: TString.h:687
static TString Format(const char *fmt,...)
Static method which formats a string using a printf style format descriptor and return a TString.
Definition: TString.cxx:2311
A TTree represents a columnar dataset.
Definition: TTree.h:71
double beta(double x, double y)
Calculates the beta function.
const Double_t sigma
double gamma(double x)
Template specialisation used in RooAbsArg:
RooCmdArg WeightVar(const char *name, Bool_t reinterpretAsWeight=kFALSE)
RooCmdArg RecycleConflictNodes(Bool_t flag=kTRUE)
RooCmdArg Index(RooCategory &icat)
RooCmdArg Rename(const char *suffix)
RooCmdArg Import(const char *state, TH1 &histo)
RooCmdArg Conditional(const RooArgSet &pdfSet, const RooArgSet &depSet, Bool_t depsAreCond=kFALSE)
Namespace for the RooStats classes.
Definition: Asimov.h:20
static constexpr double second
static constexpr double gauss
Definition: tree.py:1
const char * Name
Definition: TXMLSetup.cxx:66
void ws()
Definition: ws.C:66