Logo ROOT  
Reference Guide
 
Loading...
Searching...
No Matches
RooAbsPdf.cxx
Go to the documentation of this file.
1/*****************************************************************************
2 * Project: RooFit *
3 * Package: RooFitCore *
4 * @(#)root/roofitcore:$Id$
5 * Authors: *
6 * WV, Wouter Verkerke, UC Santa Barbara, verkerke@slac.stanford.edu *
7 * DK, David Kirkby, UC Irvine, dkirkby@uci.edu *
8 * *
9 * Copyright (c) 2000-2005, Regents of the University of California *
10 * and Stanford University. All rights reserved. *
11 * *
12 * Redistribution and use in source and binary forms, *
13 * with or without modification, are permitted according to the terms *
14 * listed in LICENSE (http://roofit.sourceforge.net/license.txt) *
15 *****************************************************************************/
16
17//////////////////////////////////////////////////////////////////////////////
18/** \class RooAbsPdf
19 \ingroup Roofitcore
20 \brief Abstract interface for all probability density functions.
21
22## RooAbsPdf, the base class of all PDFs
23
24RooAbsPdf is the base class for all probability density
25functions (PDFs). The class provides hybrid analytical/numerical
26normalization for its implementations, error tracing, and a Monte Carlo
27generator interface.
28
29### A Minimal PDF Implementation
30
31A minimal implementation of a PDF class derived from RooAbsPdf
32should override the `evaluate()` function. This function should
33return the PDF's value (which does not need to be normalised).
34
35
36#### Normalization/Integration
37
38Although the normalization of a PDF is an integral part of a
39probability density function, normalization is treated separately
40in RooAbsPdf. The reason is that a RooAbsPdf object is more than a
41PDF: it can be a building block for a more complex composite PDF
42if any of its variables are functions instead of variables. In
43such cases, the normalization of the composite PDF may not simply be
44integral over the dependents of the top-level PDF: these are
45functions with potentially non-trivial Jacobian terms themselves.
46\note Therefore, no explicit attempt should be made to normalize the
47function output in evaluate(). In particular, normalisation constants
48can be omitted to speed up the function evaluations, and included later
49in the integration of the PDF (see below), which is rarely called in
50comparison to the `evaluate()` function.
51
52In addition, RooAbsPdf objects do not have a static concept of what
53variables are parameters, and what variables are dependents (which
54need to be integrated over for a correct PDF normalization).
55Instead, the choice of normalization is always specified each time a
56normalized value is requested from the PDF via the getVal()
57method.
58
59RooAbsPdf manages the entire normalization logic of each PDF with
60the help of a RooRealIntegral object, which coordinates the integration
61of a given choice of normalization. By default, RooRealIntegral will
62perform an entirely numeric integration of all dependents. However,
63PDFs can advertise one or more (partial) analytical integrals of
64their function, and these will be used by RooRealIntegral, if it
65determines that this is safe (i.e., no hidden Jacobian terms,
66multiplication with other PDFs that have one or more dependents in
67common, etc).
68
69#### Implementing analytical integrals
70To implement analytical integrals, two functions must be implemented. First,
71
72```
73Int_t getAnalyticalIntegral(const RooArgSet& integSet, RooArgSet& anaIntSet)
74```
75should return the analytical integrals that are supported. `integSet`
76is the set of dependents for which integration is requested. The
77function should copy the subset of dependents it can analytically
78integrate to `anaIntSet`, and return a unique identification code for
79this integration configuration. If no integration can be
80performed, zero should be returned. Second,
81
82```
83double analyticalIntegral(Int_t code)
84```
85
86implements the actual analytical integral(s) advertised by
87`getAnalyticalIntegral()`. This function will only be called with
88codes returned by `getAnalyticalIntegral()`, except code zero.
89
90The integration range for each dependent to be integrated can
91be obtained from the dependent's proxy functions `min()` and
92`max()`. Never call these proxy functions for any proxy not known to
93be a dependent via the integration code. Doing so may be
94ill-defined, e.g., in case the proxy holds a function, and will
95trigger an assert. Integrated category dependents should always be
96summed over all of their states.
97
98
99
100### Direct generation of observables
101
102Distributions for any PDF can be generated with the accept/reject method,
103but for certain PDFs, more efficient methods may be implemented. To
104implement direct generation of one or more observables, two
105functions need to be implemented, similar to those for analytical
106integrals:
107
108```
109Int_t getGenerator(const RooArgSet& generateVars, RooArgSet& directVars)
110```
111and
112```
113void generateEvent(Int_t code)
114```
115
116The first function advertises observables, for which distributions can be generated,
117similar to the way analytical integrals are advertised. The second
118function implements the actual generator for the advertised observables.
119
120The generated dependent values should be stored in the proxy
121objects. For this, the assignment operator can be used (i.e. `xProxy
122= 3.0` ). Never call assign to any proxy not known to be a dependent
123via the generation code. Doing so may be ill-defined, e.g. in case
124the proxy holds a function, and will trigger an assert.
126
127### Batched function evaluations (Advanced usage)
128
129To speed up computations with large numbers of data events in unbinned fits,
130it is beneficial to override `doEval()`. Like this, large spans of
131computations can be done, without having to call `evaluate()` for each single data event.
132`doEval()` should execute the same computation as `evaluate()`, but it
133may choose an implementation that is capable of SIMD computations.
134If doEval is not implemented, the classic and slower `evaluate()` will be
135called for each data event.
136*/
137
138#include "RooAbsPdf.h"
139
140#include "FitHelpers.h"
142#include "RooMsgService.h"
143#include "RooArgSet.h"
144#include "RooArgProxy.h"
145#include "RooRealProxy.h"
146#include "RooRealVar.h"
147#include "RooGenContext.h"
148#include "RooBinnedGenContext.h"
149#include "RooPlot.h"
150#include "RooCurve.h"
151#include "RooCategory.h"
152#include "RooNameReg.h"
153#include "RooCmdConfig.h"
154#include "RooGlobalFunc.h"
155#include "RooRandom.h"
156#include "RooNumIntConfig.h"
157#include "RooProjectedPdf.h"
158#include "RooParamBinning.h"
159#include "RooNumCdf.h"
160#include "RooFitResult.h"
161#include "RooNumGenConfig.h"
162#include "RooCachedReal.h"
163#include "RooRealIntegral.h"
164#include "RooWorkspace.h"
165#include "RooNaNPacker.h"
166#include "RooFitImplHelpers.h"
167#include "RooHelpers.h"
168#include "RooFormulaVar.h"
169#include "RooDerivative.h"
170
171#include "ROOT/StringUtils.hxx"
172#include "TMath.h"
173#include "TPaveText.h"
174#include "TMatrixD.h"
175#include "TMatrixDSym.h"
176
177#include <algorithm>
178#include <iostream>
179#include <string>
180#include <cmath>
181#include <stdexcept>
182
183namespace {
184
185inline double getLog(double prob, RooAbsReal const *caller)
186{
187
188 if (prob < 0) {
189 caller->logEvalError("getLogVal() top-level p.d.f evaluates to a negative number");
191 }
192
193 if (std::isinf(prob)) {
194 oocoutW(caller, Eval) << "RooAbsPdf::getLogVal(" << caller->GetName()
195 << ") WARNING: top-level pdf has an infinite value" << std::endl;
196 }
197
198 if (prob == 0) {
199 caller->logEvalError("getLogVal() top-level p.d.f evaluates to zero");
200
201 return -std::numeric_limits<double>::infinity();
202 }
203
204 if (TMath::IsNaN(prob)) {
205 caller->logEvalError("getLogVal() top-level p.d.f evaluates to NaN");
206
207 return prob;
208 }
209
210 return std::log(prob);
211}
212
213} // namespace
214
215using std::endl, std::string, std::ostream, std::vector, std::pair, std::make_pair;
216
218
219
220
221
224
225////////////////////////////////////////////////////////////////////////////////
226/// Default constructor
227
228RooAbsPdf::RooAbsPdf() : _normMgr(this, 10) {}
229
230////////////////////////////////////////////////////////////////////////////////
231/// Constructor with name and title only
232
233RooAbsPdf::RooAbsPdf(const char *name, const char *title) :
234 RooAbsReal(name,title), _normMgr(this,10), _selectComp(true)
235{
237 setTraceCounter(0) ;
238}
239
240
241
242////////////////////////////////////////////////////////////////////////////////
243/// Constructor with name, title, and plot range
244
245RooAbsPdf::RooAbsPdf(const char *name, const char *title,
246 double plotMin, double plotMax) :
247 RooAbsReal(name,title,plotMin,plotMax), _normMgr(this,10), _selectComp(true)
248{
250 setTraceCounter(0) ;
251}
252
253
254
255////////////////////////////////////////////////////////////////////////////////
256/// Copy constructor
257
260 _normMgr(other._normMgr,this), _selectComp(other._selectComp), _normRange(other._normRange)
261{
263 setTraceCounter(other._traceCount) ;
264
265 if (other._specGeneratorConfig) {
266 _specGeneratorConfig = std::make_unique<RooNumGenConfig>(*other._specGeneratorConfig);
267 }
268}
269
270
271
272////////////////////////////////////////////////////////////////////////////////
273/// Destructor
274
278
279
281
282 if (normVal < 0. || (normVal == 0. && rawVal != 0)) {
283 //Unreasonable normalisations. A zero integral can be tolerated if the function vanishes, though.
284 const std::string msg = "p.d.f normalization integral is zero or negative: " + std::to_string(normVal);
285 logEvalError(msg.c_str());
287 return RooNaNPacker::packFloatIntoNaN(-normVal + (rawVal < 0. ? -rawVal : 0.));
288 }
289
290 if (rawVal < 0.) {
291 std::stringstream ss;
292 ss << "p.d.f value is less than zero (" << rawVal << "), trying to recover";
293 logEvalError(ss.str().c_str());
296 }
297
298 if (TMath::IsNaN(rawVal)) {
299 logEvalError("p.d.f value is Not-a-Number");
301 return rawVal;
302 }
303
304 return (rawVal == 0. && normVal == 0.) ? 0. : rawVal / normVal;
305}
306
307
308////////////////////////////////////////////////////////////////////////////////
309/// Return current value, normalized by integrating over
310/// the observables in `nset`. If `nset` is 0, the unnormalized value
311/// is returned. All elements of `nset` must be lvalues.
312///
313/// Unnormalized values are not cached.
314/// Doing so would be complicated as `_norm->getVal()` could
315/// spoil the cache and interfere with returning the cached
316/// return value. Since unnormalized calls are typically
317/// done in integration calls, there is no performance hit.
318
319double RooAbsPdf::getValV(const RooArgSet* nset) const
320{
321
322 // Special handling of case without normalization set (used in numeric integration of pdfs)
323 if (!nset) {
324 RooArgSet const* tmp = _normSet ;
325 _normSet = nullptr ;
326 double val = evaluate() ;
327 _normSet = tmp ;
328
329 return TMath::IsNaN(val) ? 0. : val;
330 }
331
332
333 // Process change in last data set used
334 bool nintChanged(false) ;
335 if (!isActiveNormSet(nset) || _norm==nullptr) {
337 }
338
339 // Return value of object. Calculated if dirty, otherwise cached value is returned.
341
342 // Evaluate numerator
343 const double rawVal = evaluate();
344
345 // Evaluate denominator
346 const double normVal = _norm->getVal();
347
349
351 }
352
353 return _value ;
354}
355
356
357////////////////////////////////////////////////////////////////////////////////
358/// Analytical integral with normalization (see RooAbsReal::analyticalIntegralWN() for further information).
359///
360/// This function applies the normalization specified by `normSet` to the integral returned
361/// by RooAbsReal::analyticalIntegral(). The passthrough scenario (code=0) is also changed
362/// to return a normalized answer.
363
364double RooAbsPdf::analyticalIntegralWN(Int_t code, const RooArgSet* normSet, const char* rangeName) const
365{
366 cxcoutD(Eval) << "RooAbsPdf::analyticalIntegralWN(" << GetName() << ") code = " << code << " normset = " << (normSet?*normSet:RooArgSet()) << std::endl ;
367
368
369 if (code==0) return getVal(normSet) ;
370 if (normSet) {
372 } else {
373 return analyticalIntegral(code,rangeName) ;
374 }
375}
376
377
378
379////////////////////////////////////////////////////////////////////////////////
380/// Check that passed value is positive and not 'not-a-number'. If
381/// not, print an error, until the error counter reaches its set
382/// maximum.
383
385{
386 // check for a math error or negative value
387 bool error(false) ;
388 if (TMath::IsNaN(value)) {
389 logEvalError(Form("p.d.f value is Not-a-Number (%f), forcing value to zero",value)) ;
390 error=true ;
391 }
392 if (value<0) {
393 logEvalError(Form("p.d.f value is less than zero (%f), forcing value to zero",value)) ;
394 error=true ;
395 }
396
397 // do nothing if we are no longer tracing evaluations and there was no error
398 if(!error) return error ;
399
400 // otherwise, print out this evaluations input values and result
401 if(++_errorCount <= 10) {
402 cxcoutD(Tracing) << "*** Evaluation Error " << _errorCount << " ";
403 if(_errorCount == 10) cxcoutD(Tracing) << "(no more will be printed) ";
404 }
405 else {
406 return error ;
407 }
408
409 Print() ;
410 return error ;
411}
412
413
414////////////////////////////////////////////////////////////////////////////////
415/// Get normalisation term needed to normalise the raw values returned by
416/// getVal(). Note that `getVal(normalisationVariables)` will automatically
417/// apply the normalisation term returned here.
418/// \param nset Set of variables to normalise over.
419double RooAbsPdf::getNorm(const RooArgSet* nset) const
420{
421 if (!nset) return 1 ;
422
423 syncNormalization(nset,true) ;
424 if (_verboseEval>1) cxcoutD(Tracing) << ClassName() << "::getNorm(" << GetName() << "): norm(" << _norm << ") = " << _norm->getVal() << std::endl ;
425
426 double ret = _norm->getVal() ;
427 if (ret==0.) {
428 if(++_errorCount <= 10) {
429 coutW(Eval) << "RooAbsPdf::getNorm(" << GetName() << ":: WARNING normalization is zero, nset = " ; nset->Print("1") ;
430 if(_errorCount == 10) coutW(Eval) << "RooAbsPdf::getNorm(" << GetName() << ") INFO: no more messages will be printed " << std::endl ;
431 }
432 }
433
434 return ret ;
435}
436
437
438
439////////////////////////////////////////////////////////////////////////////////
440/// Return pointer to RooAbsReal object that implements calculation of integral over observables iset in range
441/// rangeName, optionally taking the integrand normalized over observables nset
442
443const RooAbsReal* RooAbsPdf::getNormObj(const RooArgSet* nset, const RooArgSet* iset, const TNamed* rangeName) const
444{
445 // Check normalization is already stored
446 CacheElem* cache = static_cast<CacheElem*>(_normMgr.getObj(nset,iset,nullptr,rangeName)) ;
447 if (cache) {
448 return cache->_norm.get();
449 }
450
451 // If not create it now
454
455 // Normalization is always over all pdf components. Overriding the global
456 // component selection temporarily makes all RooRealIntegrals created during
457 // that time always include all components.
459 RooAbsReal* norm = std::unique_ptr<RooAbsReal>{createIntegral(depList,*nset, *getIntegratorConfig(), RooNameReg::str(rangeName))}.release();
460
461 // Store it in the cache
463
464 // And return the newly created integral
465 return norm ;
466}
467
468
469
470////////////////////////////////////////////////////////////////////////////////
471/// Verify that the normalization integral cached with this PDF
472/// is valid for given set of normalization observables.
473///
474/// If not, the cached normalization integral (if any) is deleted
475/// and a new integral is constructed for use with 'nset'.
476/// Elements in 'nset' can be discrete and real, but must be lvalues.
477///
478/// For functions that declare to be self-normalized by overloading the
479/// selfNormalized() function, a unit normalization is always constructed.
480
482{
483 setActiveNormSet(nset);
484
485 // Check if data sets are identical
486 CacheElem* cache = static_cast<CacheElem*>(_normMgr.getObj(nset)) ;
487 if (cache) {
488
489 bool nintChanged = (_norm!=cache->_norm.get()) ;
490 _norm = cache->_norm.get();
491
492 // In the past, this condition read `if (nintChanged && adjustProxies)`.
493 // However, the cache checks if the nset was already cached **by content**,
494 // and not by RooArgSet instance! So it can happen that the normalization
495 // set object is different, but the integral object is the same, in which
496 // case it would be wrong to not adjust the proxies. They always have to be
497 // adjusted when the nset changed, which is always the case when
498 // `syncNormalization()` is called.
499 if (adjustProxies) {
500 // Update dataset pointers of proxies
501 const_cast<RooAbsPdf*>(this)->setProxyNormSet(nset) ;
502 }
503
504 return nintChanged ;
505 }
506
507 // Update dataset pointers of proxies
508 if (adjustProxies) {
509 const_cast<RooAbsPdf*>(this)->setProxyNormSet(nset) ;
510 }
511
513 getObservables(nset, depList);
514
515 if (_verboseEval>0) {
516 if (!selfNormalized()) {
517 cxcoutD(Tracing) << ClassName() << "::syncNormalization(" << GetName()
518 << ") recreating normalization integral " << std::endl ;
519 depList.printStream(ccoutD(Tracing),kName|kValue|kArgs,kSingleLine) ;
520 } else {
521 cxcoutD(Tracing) << ClassName() << "::syncNormalization(" << GetName() << ") selfNormalized, creating unit norm" << std::endl;
522 }
523 }
524
525 // Destroy old normalization & create new
526 if (selfNormalized() || !dependsOn(depList)) {
527 auto ntitle = std::string(GetTitle()) + " Unit Normalization";
528 auto nname = std::string(GetName()) + "_UnitNorm";
529 _norm = new RooRealVar(nname.c_str(),ntitle.c_str(),1) ;
530 } else {
531 const char* nr = (_normRangeOverride.Length()>0 ? _normRangeOverride.Data() : (_normRange.Length()>0 ? _normRange.Data() : nullptr)) ;
532
533// std::cout << "RooAbsPdf::syncNormalization(" << GetName() << ") rangeName for normalization is " << (nr?nr:"<null>") << std::endl ;
535 {
536 // Normalization is always over all pdf components. Overriding the global
537 // component selection temporarily makes all RooRealIntegrals created during
538 // that time always include all components.
540 normInt = std::unique_ptr<RooAbsReal>{createIntegral(depList,*getIntegratorConfig(),nr)}.release();
541 }
542 static_cast<RooRealIntegral*>(normInt)->setAllowComponentSelection(false);
543 normInt->getVal() ;
544// std::cout << "resulting normInt = " << normInt->GetName() << std::endl ;
545
546 const char* cacheParamsStr = getStringAttribute("CACHEPARAMINT") ;
548
549 std::unique_ptr<RooArgSet> intParams{normInt->getVariables()} ;
550
552
553 if (!cacheParams.empty()) {
554 cxcoutD(Caching) << "RooAbsReal::createIntObj(" << GetName() << ") INFO: constructing " << cacheParams.size()
555 << "-dim value cache for integral over " << depList << " as a function of " << cacheParams << " in range " << (nr?nr:"<default>") << std::endl ;
556 std::string name = normInt->GetName() + ("_CACHE_[" + cacheParams.contentsString()) + "]";
558 cachedIntegral->setInterpolationOrder(2) ;
559 cachedIntegral->addOwnedComponents(*normInt) ;
560 cachedIntegral->setCacheSource(true) ;
561 if (normInt->operMode()==ADirty) {
562 cachedIntegral->setOperMode(ADirty) ;
563 }
565 }
566
567 }
568 _norm = normInt ;
569 }
570
571 // Register new normalization with manager (takes ownership)
572 cache = new CacheElem(*_norm) ;
573 _normMgr.setObj(nset,cache) ;
574
575// std::cout << "making new object " << _norm->GetName() << std::endl ;
576
577 return true ;
578}
579
580
581
582////////////////////////////////////////////////////////////////////////////////
583/// Reset error counter to given value, limiting the number
584/// of future error messages for this pdf to 'resetValue'
585
591
592
593
594////////////////////////////////////////////////////////////////////////////////
595/// Reset trace counter to given value, limiting the
596/// number of future trace messages for this pdf to 'value'
597
599{
600 if (!allNodes) {
602 return ;
603 } else {
606 for(auto * pdf : dynamic_range_cast<RooAbsPdf*>(branchList)) {
607 if (pdf) pdf->setTraceCounter(value,false) ;
608 }
609 }
610
611}
612
613
614
615
616////////////////////////////////////////////////////////////////////////////////
617/// Return the log of the current value with given normalization
618/// An error message is printed if the argument of the log is negative.
619
620double RooAbsPdf::getLogVal(const RooArgSet* nset) const
621{
622 return getLog(getVal(nset), this);
623}
624////////////////////////////////////////////////////////////////////////////////
625/// This function returns the penalty term.
626/// Penalty terms modify the likelihood,during model parameter estimation.This penalty term is usually
627// a function of the model parameters
629{
630 return 0;
631}
632
633////////////////////////////////////////////////////////////////////////////////
634/// Check for infinity or NaN.
635/// \param[in] inputs Array to check
636/// \return True if either infinity or NaN were found.
637namespace {
638template<class T>
639bool checkInfNaNNeg(const T& inputs) {
640 // check for a math error or negative value
641 bool inf = false;
642 bool nan = false;
643 bool neg = false;
644
645 for (double val : inputs) { //CHECK_VECTORISE
646 inf |= !std::isfinite(val);
647 nan |= TMath::IsNaN(val); // Works also during fast math
648 neg |= val < 0;
649 }
650
651 return inf || nan || neg;
652}
653}
654
655
656////////////////////////////////////////////////////////////////////////////////
657/// Scan through outputs and fix+log all nans and negative values.
658/// \param[in,out] outputs Array to be scanned & fixed.
659/// \param[in] begin Begin of event range. Only needed to print the correct event number
660/// where the error occurred.
661void RooAbsPdf::logBatchComputationErrors(std::span<const double>& outputs, std::size_t begin) const {
662 for (unsigned int i=0; i<outputs.size(); ++i) {
663 const double value = outputs[i];
664 if (TMath::IsNaN(outputs[i])) {
665 logEvalError(Form("p.d.f value of (%s) is Not-a-Number (%f) for entry %zu",
666 GetName(), value, begin+i));
667 } else if (!std::isfinite(outputs[i])){
668 logEvalError(Form("p.d.f value of (%s) is (%f) for entry %zu",
669 GetName(), value, begin+i));
670 } else if (outputs[i] < 0.) {
671 logEvalError(Form("p.d.f value of (%s) is less than zero (%f) for entry %zu",
672 GetName(), value, begin+i));
673 }
674 }
675}
676
677
678void RooAbsPdf::getLogProbabilities(std::span<const double> pdfValues, double * output) const {
679 for (std::size_t i = 0; i < pdfValues.size(); ++i) {
680 output[i] = getLog(pdfValues[i], this);
681 }
682}
683
684////////////////////////////////////////////////////////////////////////////////
685/// Return the extended likelihood term (\f$ N_\mathrm{expect} - N_\mathrm{observed} \cdot \log(N_\mathrm{expect} \f$)
686/// of this PDF for the given number of observed events.
687///
688/// For successful operation, the PDF implementation must indicate that
689/// it is extendable by overloading `canBeExtended()`, and must
690/// implement the `expectedEvents()` function.
691///
692/// \param[in] sumEntries The number of observed events.
693/// \param[in] nset The normalization set when asking the pdf for the expected
694/// number of events.
695/// \param[in] sumEntriesW2 The number of observed events when weighting with
696/// squared weights. If non-zero, the weight-squared error
697/// correction is applied to the extended term.
698/// \param[in] doOffset Offset the extended term by a counterterm where the
699/// expected number of events equals the observed number of events.
700/// This constant shift results in a term closer to zero that is
701/// approximately chi-square distributed. It is useful to do this
702/// also when summing multiple NLL terms to avoid numeric precision
703/// loss that happens if you sum multiple terms of different orders
704/// of magnitude.
705///
706/// The weight-squared error correction works as follows:
707/// adjust poisson such that
708/// estimate of \f$N_\mathrm{expect}\f$ stays at the same value, but has a different variance, rescale
709/// both the observed and expected count of the Poisson with a factor \f$ \sum w_{i} / \sum w_{i}^2 \f$
710/// (the effective weight of the Poisson term),
711/// i.e., change \f$\mathrm{Poisson}(N_\mathrm{observed} = \sum w_{i} | N_\mathrm{expect} )\f$
712/// to \f$ \mathrm{Poisson}(\sum w_{i} \cdot \sum w_{i} / \sum w_{i}^2 | N_\mathrm{expect} \cdot \sum w_{i} / \sum w_{i}^2 ) \f$,
713/// weighted by the effective weight \f$ \sum w_{i}^2 / \sum w_{i} \f$ in the likelihood.
714/// Since here we compute the likelihood with the weight square, we need to multiply by the
715/// square of the effective weight:
716/// - \f$ W_\mathrm{expect} = N_\mathrm{expect} \cdot \sum w_{i} / \sum w_{i}^2 \f$ : effective expected entries
717/// - \f$ W_\mathrm{observed} = \sum w_{i} \cdot \sum w_{i} / \sum w_{i}^2 \f$ : effective observed entries
718///
719/// The extended term for the likelihood weighted by the square of the weight will be then:
720///
721/// \f$ \left(\sum w_{i}^2 / \sum w_{i}\right)^2 \cdot W_\mathrm{expect} - (\sum w_{i}^2 / \sum w_{i})^2 \cdot W_\mathrm{observed} \cdot \log{W_\mathrm{expect}} \f$
722///
723/// aund this is using the previous expressions for \f$ W_\mathrm{expect} \f$ and \f$ W_\mathrm{observed} \f$:
724///
725/// \f$ \sum w_{i}^2 / \sum w_{i} \cdot N_\mathrm{expect} - \sum w_{i}^2 \cdot \log{W_\mathrm{expect}} \f$
726///
727/// Since the weights are constants in the likelihood we can use \f$\log{N_\mathrm{expect}}\f$ instead of \f$\log{W_\mathrm{expect}}\f$.
728///
729/// See also RooAbsPdf::extendedTerm(RooAbsData const& data, bool weightSquared, bool doOffset),
730/// which takes a dataset to extract \f$N_\mathrm{observed}\f$ and the
731/// normalization set.
732double RooAbsPdf::extendedTerm(double sumEntries, RooArgSet const* nset, double sumEntriesW2, bool doOffset) const
733{
734 return extendedTerm(sumEntries, expectedEvents(nset), sumEntriesW2, doOffset);
735}
736
737double RooAbsPdf::extendedTerm(double sumEntries, double expected, double sumEntriesW2, bool doOffset) const
738{
739 // check if this PDF supports extended maximum likelihood fits
740 if(!canBeExtended()) {
741 coutE(InputArguments) << GetName() << ": this PDF does not support extended maximum likelihood"
742 << std::endl;
743 return 0.0;
744 }
745
746 if(expected < 0.0) {
747 coutE(InputArguments) << GetName() << ": calculated negative expected events: " << expected
748 << std::endl;
749 logEvalError("extendedTerm #expected events is <0 return a NaN");
750 return TMath::QuietNaN();
751 }
752
753
754 // Explicitly handle case Nobs=Nexp=0
755 if (std::abs(expected)<1e-10 && std::abs(sumEntries)<1e-10) {
756 return 0.0;
757 }
758
759 // Check for errors in Nexpected
760 if (TMath::IsNaN(expected)) {
761 logEvalError("extendedTerm #expected events is a NaN") ;
762 return TMath::QuietNaN() ;
763 }
764
765 double extra = doOffset
766 ? (expected - sumEntries) - sumEntries * (std::log(expected) - std::log(sumEntries))
767 : expected - sumEntries * std::log(expected);
768
769 if(sumEntriesW2 != 0.0) {
770 extra *= sumEntriesW2 / sumEntries;
771 }
772
773 return extra;
774}
775
776////////////////////////////////////////////////////////////////////////////////
777/// Return the extended likelihood term (\f$ N_\mathrm{expect} - N_\mathrm{observed} \cdot \log(N_\mathrm{expect} \f$)
778/// of this PDF for the given number of observed events.
779///
780/// This function is a wrapper around
781/// RooAbsPdf::extendedTerm(double, RooArgSet const *, double, bool) const,
782/// where the number of observed events and observables to be used as the
783/// normalization set for the pdf is extracted from a RooAbsData.
784///
785/// For successful operation, the PDF implementation must indicate that
786/// it is extendable by overloading `canBeExtended()`, and must
787/// implement the `expectedEvents()` function.
788///
789/// \param[in] data The RooAbsData to retrieve the set of observables and
790/// number of expected events.
791/// \param[in] weightSquared If set to `true`, the extended term will be scaled by
792/// the ratio of squared event weights over event weights:
793/// \f$ \sum w_{i}^2 / \sum w_{i} \f$.
794/// Intended to be used by fits with the `SumW2Error()` option that
795/// can be passed to RooAbsPdf::fitTo()
796/// (see the documentation of said function to learn more about the
797/// interpretation of fits with squared weights).
798/// \param[in] doOffset See RooAbsPdf::extendedTerm(double, RooArgSet const*, double, bool) const.
799
800double RooAbsPdf::extendedTerm(RooAbsData const& data, bool weightSquared, bool doOffset) const {
801 double sumW = data.sumEntries();
802 double sumW2 = 0.0;
803 if (weightSquared) {
804 sumW2 = data.sumEntriesW2();
805 }
806 return extendedTerm(sumW, data.get(), sumW2, doOffset);
807}
808
809
810/** @fn RooAbsPdf::createNLL()
811 *
812 * @brief Construct representation of -log(L) of PDF with given dataset.
813 *
814 * If dataset is unbinned, an unbinned likelihood is constructed.
815 * If the dataset is binned, a binned likelihood is constructed.
816 *
817 * @param data Reference to a RooAbsData object representing the dataset.
818 * @param cmdArgs Variadic template arguments representing optional command arguments.
819 * You can pass either an arbitrary number of RooCmdArg instances
820 * or a single RooLinkedList that points to the RooCmdArg objects.
821 * @return An owning pointer to the created RooAbsReal NLL object.
822 *
823 * @tparam CmdArgs_t Template types for optional command arguments.
824 * Can either be an arbitrary number of RooCmdArg or a single RooLinkedList.
825 *
826 * \note This front-end function should not be re-implemented in derived PDF types.
827 * If you mean to customize the NLL creation routine,
828 * you need to override the virtual RooAbsPdf::createNLLImpl() method.
829 *
830 * The following named arguments are supported:
831 *
832 * <table>
833 * <tr><th> Type of CmdArg <th> Effect on NLL
834 * <tr><td> `ConditionalObservables(Args_t &&... argsOrArgSet)` <td> Do not normalize PDF over listed observables.
835 * Arguments can either be multiple RooRealVar or a single RooArgSet containing them.
836 * <tr><td> `Extended(bool flag)` <td> Add extended likelihood term, off by default.
837 * <tr><td> `Range(const char* name)` <td> Fit only data inside range with given name. Multiple comma-separated range names can be specified.
838 * In this case, the unnormalized PDF \f$f(x)\f$ is normalized by the integral over all ranges \f$r_i\f$:
839 * \f[
840 * p(x) = \frac{f(x)}{\sum_i \int_{r_i} f(x) dx}.
841 * \f]
842 * <tr><td> `Range(double lo, double hi)` <td> Fit only data inside given range. A range named "fit" is created on the fly on all observables.
843 * <tr><td> `SumCoefRange(const char* name)` <td> Set the range in which to interpret the coefficients of RooAddPdf components
844 * <tr><td> `NumCPU(int num, int istrat)` <td> Parallelize NLL calculation on num CPUs. (Currently, this setting is ignored with the **cpu** Backend.)
845 * <table>
846 * <tr><th> Strategy <th> Effect
847 * <tr><td> 0 = RooFit::BulkPartition - *default* <td> Divide events in N equal chunks
848 * <tr><td> 1 = RooFit::Interleave <td> Process event i%N in process N. Recommended for binned data with
849 * a substantial number of zero-bins, which will be distributed across processes more equitably in this strategy
850 * <tr><td> 2 = RooFit::SimComponents <td> Process each component likelihood of a RooSimultaneous fully in a single process
851 * and distribute components over processes. This approach can be beneficial if normalization calculation time
852 * dominates the total computation time of a component (since the normalization calculation must be performed
853 * in each process in strategies 0 and 1. However beware that if the RooSimultaneous components do not share many
854 * parameters this strategy is inefficient: as most minuit-induced likelihood calculations involve changing
855 * a single parameter, only 1 of the N processes will be active most of the time if RooSimultaneous components
856 * do not share many parameters
857 * <tr><td> 3 = RooFit::Hybrid <td> Follow strategy 0 for all RooSimultaneous components, except those with less than
858 * 30 dataset entries, for which strategy 2 is followed.
859 * </table>
860 * <tr><td> `EvalBackend(std::string const&)` <td> Choose a likelihood evaluation backend:
861 * <table>
862 * <tr><th> Backend <th> Description
863 * <tr><td> **cpu** - *default* <td> New vectorized evaluation mode, using faster math functions and auto-vectorisation (currently on a single thread).
864 * Since ROOT 6.23, this is the default if `EvalBackend()` is not passed, succeeding the **legacy** backend.
865 * If all RooAbsArg objects in the model support vectorized evaluation,
866 * likelihood computations are 2 to 10 times faster than with the **legacy** backend (each on a single thread).
867 * - unless your dataset is so small that the vectorization is not worth it.
868 * The relative difference of the single log-likelihoods with respect to the legacy mode is usually better than \f$10^{-12}\f$,
869 * and for fit parameters it's usually better than \f$10^{-6}\f$. In past ROOT releases, this backend could be activated with the now deprecated `BatchMode()` option.
870 * <tr><td> **cuda** <td> Evaluate the likelihood on a GPU that supports CUDA.
871 * This backend re-uses code from the **cpu** backend, but compiled in CUDA kernels.
872 * Hence, the results are expected to be identical, modulo some numerical differences that can arise from the different order in which the GPU is summing the log probabilities.
873 * This backend can drastically speed up the fit if all RooAbsArg object in the model support it.
874 * <tr><td> **legacy** <td> The original likelihood evaluation method.
875 * Evaluates the PDF for each single data entry at a time before summing the negative log probabilities.
876 * It supports multi-threading, but you might need more than 20 threads to maybe see about 10% performance gain over the default cpu-backend (that runs currently only on a single thread).
877 * <tr><td> **codegen** <td> **Experimental** - Generates and compiles minimal C++ code for the NLL on-the-fly and wraps it in the returned RooAbsReal.
878 * Also generates and compiles the code for the gradient using Automatic Differentiation (AD) with [Clad](https://github.com/vgvassilev/clad).
879 * This analytic gradient is passed to the minimizer, which can result in significant speedups for many-parameter fits,
880 * even compared to the **cpu** backend. However, if one of the RooAbsArg objects in the model does not support the code generation,
881 * this backend can't be used.
882 * <tr><td> **codegen_no_grad** <td> **Experimental** - Same as **codegen**, but doesn't generate and compile the gradient code and use the regular numerical differentiation instead.
883 * This is expected to be slower, but useful for debugging problems with the analytic gradient.
884 * </table>
885 * <tr><td> `Optimize(bool flag)` <td> Activate constant term optimization (on by default)
886 * <tr><td> `SplitRange(bool flag)` <td> Use separate fit ranges in a simultaneous fit. Actual range name for each subsample is assumed to
887 * be `rangeName_indexState`, where `indexState` is the state of the master index category of the simultaneous fit.
888 * Using `Range("range"), SplitRange()` as switches, different ranges could be set like this:
889 * ```
890 * myVariable.setRange("range_pi0", 135, 210);
891 * myVariable.setRange("range_gamma", 50, 210);
892 * ```
893 * <tr><td> `Constrain(const RooArgSet&pars)` <td> For p.d.f.s that contain internal parameter constraint terms (that is usually product PDFs, where one
894 * term of the product depends on parameters but not on the observable(s),), only apply constraints to the given subset of parameters.
895 * <tr><td> `ExternalConstraints(const RooArgSet& )` <td> Include given external constraints to likelihood by multiplying them with the original likelihood.
896 * <tr><td> `GlobalObservables(const RooArgSet&)` <td> Define the set of normalization observables to be used for the constraint terms.
897 * If none are specified the constrained parameters are used.
898 * <tr><td> `GlobalObservablesSource(const char* sourceName)` <td> Which source to prioritize for global observable values.
899 * Can be either:
900 * - `data`: to take the values from the dataset,
901 * falling back to the pdf value if a given global observable is not available.
902 * If no `GlobalObservables` or `GlobalObservablesTag` command argument is given, the set
903 * of global observables will be automatically defined to be the set stored in the data.
904 * - `model`: to take all values from the pdf and completely ignore the set of global observables stored in the data
905 * (not even using it to automatically define the set of global observables
906 * if the `GlobalObservables` or `GlobalObservablesTag` command arguments are not given).
907 * The default option is `data`.
908 * <tr><td> `GlobalObservablesTag(const char* tagName)` <td> Define the set of normalization observables to be used for the constraint terms by
909 * a string attribute associated with pdf observables that match the given tagName.
910 * <tr><td> `Verbose(bool flag)` <td> Controls RooFit informational messages in likelihood construction
911 * <tr><td> `CloneData(bool flag)` <td> Use clone of dataset in NLL (default is true).
912 * \warning Deprecated option that is ignored. It is up to the implementation of the NLL creation method if the data is cloned or not.
913 * <tr><td> `Offset(std::string const& mode)` <td> Likelihood offsetting mode. Can be either:
914 * <table>
915 * <tr><th> Mode <th> Description
916 * <tr><td> **none** - *default* <td> No offsetting.
917 * <tr><td> **initial** <td> Offset likelihood by initial value (so that starting value of FCN in minuit is zero).
918 * This can improve numeric stability in simultaneous fits with components with large likelihood values.
919 * <tr><td> **bin** <td> Offset likelihood bin-by-bin with a template histogram model based on the obersved data.
920 * This results in per-bin values that are all in the same order of magnitude, which reduces precision loss in the sum,
921 * which can drastically improve numeric stability.
922 * Furthermore, \f$2\cdot \text{NLL}\f$ defined like this is approximately chi-square distributed, allowing for goodness-of-fit tests.
923 * </table>
924 * <tr><td> `IntegrateBins(double precision)` <td> In binned fits, integrate the PDF over the bins instead of using the probability density at the bin centre.
925 * This can reduce the bias observed when fitting functions with high curvature to binned data.
926 * - precision > 0: Activate bin integration everywhere. Use precision between 0.01 and 1.E-6, depending on binning.
927 * Note that a low precision such as 0.01 might yield identical results to 1.E-4, since the integrator might reach 1.E-4 already in its first
928 * integration step. If lower precision is desired (more speed), a RooBinSamplingPdf has to be created manually, and its integrator
929 * has to be manipulated directly.
930 * - precision = 0: Activate bin integration only for continuous PDFs fit to a RooDataHist.
931 * - precision < 0: Deactivate.
932 * \see RooBinSamplingPdf
933 * <tr><td> `ModularL(bool flag)` <td> Enable or disable modular likelihoods, which will become the default in a future release.
934 * This does not change any user-facing code, but only enables a different likelihood class in the back-end. Note that this
935 * should be set to true for parallel minimization of likelihoods!
936 * Note that it is currently not recommended to use Modular likelihoods without any parallelization enabled in the minimization, since
937 * some features such as offsetting might not yet work in this case.
938 * </table>
939 */
940
941
942/** @brief Protected implementation of the NLL creation routine.
943 *
944 * This virtual function can be overridden in case you want to change the NLL creation logic for custom PDFs.
945 *
946 * \note Never call this function directly. Instead, call RooAbsPdf::createNLL().
947 */
948
949std::unique_ptr<RooAbsReal> RooAbsPdf::createNLLImpl(RooAbsData &data, const RooLinkedList &cmdList)
950{
951 return RooFit::FitHelpers::createNLL(*this, data, cmdList);
952}
953
954
955/** @fn RooAbsPdf::fitTo()
956 *
957 * @brief Fit PDF to given dataset.
958 *
959 * If dataset is unbinned, an unbinned maximum likelihood is performed.
960 * If the dataset is binned, a binned maximum likelihood is performed.
961 * By default the fit is executed through the MINUIT commands MIGRAD, HESSE in succession.
962 *
963 * @param data Reference to a RooAbsData object representing the dataset.
964 * @param cmdArgs Variadic template arguments representing optional command arguments.
965 * You can pass either an arbitrary number of RooCmdArg instances
966 * or a single RooLinkedList that points to the RooCmdArg objects.
967 * @return An owning pointer to the created RooAbsReal NLL object.
968 * @return RooFitResult with fit status and parameters if option Save() is used, `nullptr` otherwise. The user takes ownership of the fit result.
969 *
970 * @tparam CmdArgs_t Template types for optional command arguments.
971 * Can either be an arbitrary number of RooCmdArg or a single RooLinkedList.
972 *
973 * \note This front-end function should not be re-implemented in derived PDF types.
974 * If you mean to customize the likelihood fitting routine,
975 * you need to override the virtual RooAbsPdf::fitToImpl() method.
976 *
977 * The following named arguments are supported:
978 *
979 * <table>
980 * <tr><th> Type of CmdArg <th> Options to control construction of -log(L)
981 * <tr><td> <td> All command arguments that can also be passed to the NLL creation method.
982 * \see RooAbsPdf::createNLL()
983 *
984 * <tr><th><th> Options to control flow of fit procedure
985 * <tr><td> `Minimizer("<type>", "<algo>")` <td> Choose minimization package and optionally the algorithm to use. Default is MINUIT/MIGRAD through the RooMinimizer interface,
986 * but others can be specified (through RooMinimizer interface).
987 * <table>
988 * <tr><th> Type <th> Algorithm
989 * <tr><td> Minuit <td> migrad, simplex, minimize (=migrad+simplex), migradimproved (=migrad+improve)
990 * <tr><td> Minuit2 <td> migrad, simplex, minimize, scan
991 * <tr><td> GSLMultiMin <td> conjugatefr, conjugatepr, bfgs, bfgs2, steepestdescent
992 * <tr><td> GSLSimAn <td> -
993 * </table>
994 *
995 * <tr><td> `InitialHesse(bool flag)` <td> Flag controls if HESSE before MIGRAD as well, off by default
996 * <tr><td> `Optimize(bool flag)` <td> Activate constant term optimization of test statistic during minimization (on by default)
997 * <tr><td> `Hesse(bool flag)` <td> Flag controls if HESSE is run after MIGRAD, on by default
998 * <tr><td> `Minos(bool flag)` <td> Flag controls if MINOS is run after HESSE, off by default
999 * <tr><td> `Minos(const RooArgSet& set)` <td> Only run MINOS on given subset of arguments
1000 * <tr><td> `Save(bool flag)` <td> Flag controls if RooFitResult object is produced and returned, off by default
1001 * <tr><td> `Strategy(Int_t flag)` <td> Set Minuit strategy (0 to 2, default is 1)
1002 * <tr><td> `MaxCalls(int n)` <td> Change maximum number of likelihood function calls from MINUIT (if `n <= 0`, the default of 500 * #%parameters is used)
1003 * <tr><td> `EvalErrorWall(bool flag=true)` <td> When parameters are in disallowed regions (e.g. PDF is negative), return very high value to fitter
1004 * to force it out of that region. This can, however, mean that the fitter gets lost in this region. If
1005 * this happens, try switching it off.
1006 * <tr><td> `RecoverFromUndefinedRegions(double strength)` <td> When PDF is invalid (e.g. parameter in undefined region), try to direct minimiser away from that region.
1007 * `strength` controls the magnitude of the penalty term. Leaving out this argument defaults to 10. Switch off with `strength = 0.`.
1008 *
1009 * <tr><td> `SumW2Error(bool flag)` <td> Apply correction to errors and covariance matrix.
1010 * This uses two covariance matrices, one with the weights, the other with squared weights,
1011 * to obtain the correct errors for weighted likelihood fits. If this option is activated, the
1012 * corrected covariance matrix is calculated as \f$ V_\mathrm{corr} = V C^{-1} V \f$, where \f$ V \f$ is the original
1013 * covariance matrix and \f$ C \f$ is the inverse of the covariance matrix calculated using the
1014 * squared weights. This allows to switch between two interpretations of errors:
1015 * <table>
1016 * <tr><th> SumW2Error <th> Interpretation
1017 * <tr><td> true <td> The errors reflect the uncertainty of the Monte Carlo simulation.
1018 * Use this if you want to know how much accuracy you can get from the available Monte Carlo statistics.
1019 *
1020 * **Example**: Simulation with 1000 events, the average weight is 0.1.
1021 * The errors are as big as if one fitted to 1000 events.
1022 * <tr><td> false <td> The errors reflect the errors of a dataset, which is as big as the sum of weights.
1023 * Use this if you want to know what statistical errors you would get if you had a dataset with as many
1024 * events as the (weighted) Monte Carlo simulation represents.
1025 *
1026 * **Example** (Data as above):
1027 * The errors are as big as if one fitted to 100 events.
1028 * </table>
1029 * \note If the `SumW2Error` correction is enabled, the covariance matrix quality stored in the RooFitResult
1030 * object will be the minimum of the original covariance matrix quality and the quality of the covariance
1031 * matrix calculated with the squared weights.
1032 * <tr><td> `AsymptoticError()` <td> Use the asymptotically correct approach to estimate errors in the presence of weights.
1033 * This is slower but more accurate than `SumW2Error`. See also https://arxiv.org/abs/1911.01303).
1034 This option even correctly implements the case of extended likelihood fits
1035 (see this [writeup on extended weighted fits](https://root.cern/files/extended_weighted_fits.pdf) that complements the paper linked before).
1036 * <tr><td> `PrefitDataFraction(double fraction)`
1037 * <td> Runs a prefit on a small dataset of size fraction*(actual data size). This can speed up fits
1038 * by finding good starting values for the parameters for the actual fit.
1039 * \warning Prefitting may give bad results when used in binned analysis.
1040 *
1041 * <tr><th><th> Options to control informational output
1042 * <tr><td> `Verbose(bool flag)` <td> Flag controls if verbose output is printed (NLL, parameter changes during fit).
1043 * <tr><td> `Timer(bool flag)` <td> Time CPU and wall clock consumption of fit steps, off by default.
1044 * <tr><td> `PrintLevel(Int_t level)` <td> Set Minuit print level (-1 to 3, default is 1). At -1 all RooFit informational messages are suppressed as well.
1045 * See RooMinimizer::PrintLevel for the meaning of the levels.
1046 * <tr><td> `Warnings(bool flag)` <td> Enable or disable MINUIT warnings (enabled by default)
1047 * <tr><td> `PrintEvalErrors(Int_t numErr)` <td> Control number of p.d.f evaluation errors printed per likelihood evaluation.
1048 * A negative value suppresses output completely, a zero value will only print the error count per p.d.f component,
1049 * a positive value will print details of each error up to `numErr` messages per p.d.f component.
1050 * <tr><td> `Parallelize(Int_t nWorkers)` <td> Control global parallelization settings. Arguments 1 and above enable the use of RooFit's parallel minimization
1051 * backend and uses the number given as the number of workers to use in the parallelization. -1 also enables
1052 * RooFit's parallel minimization backend, and sets the number of workers to the number of available processes.
1053 * 0 disables this feature.
1054 * In case parallelization is requested, this option implies `ModularL(true)` in the internal call to the NLL creation method.
1055 * <tr><td> `ParallelGradientOptions(bool enable=true, int orderStrategy=0, int chainFactor=1)` <td> **Experimental** - Control gradient parallelization settings. The first argument
1056 * only disables or enables gradient parallelization, this is on by default.
1057 * The second argument determines the internal partial derivative calculation
1058 * ordering strategy. The third argument determines the number of partial
1059 * derivatives that are executed per task package on each worker.
1060 * <tr><td> `ParallelDescentOptions(bool enable=false, int splitStrategy=0, int numSplits=4)` <td> **Experimental** - Control settings related to the parallelization of likelihoods
1061 * outside of the gradient calculation but in the minimization, most prominently
1062 * in the linesearch step. The first argument this disables or enables likelihood
1063 * parallelization. The second argument determines whether to split the task batches
1064 * per event or per likelihood component. And the third argument how many events or
1065 * respectively components to include in each batch.
1066 * <tr><td> `TimingAnalysis(bool flag)` <td> **Experimental** - Log timings. This feature logs timings with NewStyle likelihoods on multiple processes simultaneously
1067 * and outputs the timings at the end of a run to json log files, which can be analyzed with the
1068 * `RooFit::MultiProcess::HeatmapAnalyzer`. Only works with simultaneous likelihoods.
1069 * </table>
1070 */
1071
1072
1073/** @brief Protected implementation of the likelihood fitting routine.
1074 *
1075 * This virtual function can be overridden in case you want to change the likelihood fitting logic for custom PDFs.
1076 *
1077 * \note Never call this function directly. Instead, call RooAbsPdf::fitTo().
1078 */
1079
1080std::unique_ptr<RooFitResult> RooAbsPdf::fitToImpl(RooAbsData& data, const RooLinkedList& cmdList)
1081{
1082 return RooFit::FitHelpers::fitTo(*this, data, cmdList, false);
1083}
1084
1085
1086////////////////////////////////////////////////////////////////////////////////
1087/// Print value of p.d.f, also print normalization integral that was last used, if any
1088
1089void RooAbsPdf::printValue(ostream& os) const
1090{
1091 // silent warning messages coming when evaluating a RooAddPdf without a normalization set
1093
1094 getVal() ;
1095
1096 if (_norm) {
1097 os << getVal() << "/" << _norm->getVal() ;
1098 } else {
1099 os << getVal();
1100 }
1101}
1102
1103
1104
1105////////////////////////////////////////////////////////////////////////////////
1106/// Print multi line detailed information of this RooAbsPdf
1107
1108void RooAbsPdf::printMultiline(ostream& os, Int_t contents, bool verbose, TString indent) const
1109{
1110 RooAbsReal::printMultiline(os,contents,verbose,indent);
1111 os << indent << "--- RooAbsPdf ---" << std::endl;
1112 os << indent << "Cached value = " << _value << std::endl ;
1113 if (_norm) {
1114 os << indent << " Normalization integral: " << std::endl ;
1115 auto moreIndent = std::string(indent.Data()) + " " ;
1117 }
1118}
1119
1120
1121
1122////////////////////////////////////////////////////////////////////////////////
1123/// Return a binned generator context
1124
1126{
1127 return new RooBinnedGenContext(*this,vars,nullptr,nullptr,verbose) ;
1128}
1129
1130
1131////////////////////////////////////////////////////////////////////////////////
1132/// Interface function to create a generator context from a p.d.f. This default
1133/// implementation returns a 'standard' context that works for any p.d.f
1134
1136 const RooArgSet* auxProto, bool verbose) const
1137{
1138 return new RooGenContext(*this,vars,prototype,auxProto,verbose) ;
1139}
1140
1141
1142////////////////////////////////////////////////////////////////////////////////
1143
1145 bool verbose, bool autoBinned, const char* binnedTag) const
1146{
1147 if (prototype || (auxProto && !auxProto->empty())) {
1148 return genContext(vars,prototype,auxProto,verbose);
1149 }
1150
1151 RooAbsGenContext *context(nullptr) ;
1152 if ( (autoBinned && isBinnedDistribution(vars)) || ( binnedTag && strlen(binnedTag) && (getAttribute(binnedTag)||string(binnedTag)=="*"))) {
1153 context = binnedGenContext(vars,verbose) ;
1154 } else {
1155 context= genContext(vars,nullptr,nullptr,verbose);
1156 }
1157 return context ;
1158}
1159
1160
1161
1162////////////////////////////////////////////////////////////////////////////////
1163/// Generate a new dataset containing the specified variables with events sampled from our distribution.
1164/// Generate the specified number of events or expectedEvents() if not specified.
1165/// \param[in] whatVars Choose variables in which to generate events. Variables not listed here will remain
1166/// constant and not be used for event generation.
1167/// \param[in] arg1,arg2,arg3,arg4,arg5,arg6 Optional RooCmdArg() to change behaviour of generate().
1168/// \return RooDataSet *, owned by caller.
1169///
1170/// Any variables of this PDF that are not in whatVars will use their
1171/// current values and be treated as fixed parameters. Returns zero
1172/// in case of an error.
1173///
1174/// <table>
1175/// <tr><th> Type of CmdArg <th> Effect on generate
1176/// <tr><td> `Name(const char* name)` <td> Name of the output dataset
1177/// <tr><td> `Verbose(bool flag)` <td> Print informational messages during event generation
1178/// <tr><td> `NumEvents(int nevt)` <td> Generate specified number of events
1179/// <tr><td> `Extended()` <td> If no number of events to be generated is given,
1180/// use expected number of events from extended likelihood term.
1181/// This evidently only works for extended PDFs.
1182/// <tr><td> `GenBinned(const char* tag)` <td> Use binned generation for all component pdfs that have 'setAttribute(tag)' set
1183/// <tr><td> `AutoBinned(bool flag)` <td> Automatically deploy binned generation for binned distributions (e.g. RooHistPdf, sums and products of
1184/// RooHistPdfs etc)
1185/// \note Datasets that are generated in binned mode are returned as weighted unbinned datasets. This means that
1186/// for each bin, there will be one event in the dataset with a weight corresponding to the (possibly randomised) bin content.
1187///
1188///
1189/// <tr><td> `AllBinned()` <td> As above, but for all components.
1190/// \note The notion of components is only meaningful for simultaneous PDFs
1191/// as binned generation is always executed at the top-level node for a regular
1192/// PDF, so for those it only mattes that the top-level node is tagged.
1193///
1194/// <tr><td> ProtoData(const RooDataSet& data, bool randOrder)
1195/// <td> Use specified dataset as prototype dataset. If randOrder in ProtoData() is set to true,
1196/// the order of the events in the dataset will be read in a random order if the requested
1197/// number of events to be generated does not match the number of events in the prototype dataset.
1198/// \note If ProtoData() is used, the specified existing dataset as a prototype: the new dataset will contain
1199/// the same number of events as the prototype (unless otherwise specified), and any prototype variables not in
1200/// whatVars will be copied into the new dataset for each generated event and also used to set our PDF parameters.
1201/// The user can specify a number of events to generate that will override the default. The result is a
1202/// copy of the prototype dataset with only variables in whatVars randomized. Variables in whatVars that
1203/// are not in the prototype will be added as new columns to the generated dataset.
1204///
1205/// </table>
1206///
1207/// #### Accessing the underlying event generator
1208/// Depending on the fit model (if it is difficult to sample), it may be necessary to change generator settings.
1209/// For the default generator (RooFoamGenerator), the number of samples or cells could be increased by e.g. using
1210/// myPdf->specialGeneratorConfig()->getConfigSection("RooFoamGenerator").setRealValue("nSample",1e4);
1211///
1212/// The foam generator e.g. has the following config options:
1213/// - nCell[123N]D
1214/// - nSample
1215/// - chatLevel
1216/// \see rf902_numgenconfig.C
1217
1219 const RooCmdArg& arg3,const RooCmdArg& arg4, const RooCmdArg& arg5,const RooCmdArg& arg6)
1220{
1221 // Select the pdf-specific commands
1222 RooCmdConfig pc("RooAbsPdf::generate(" + std::string(GetName()) + ")");
1223 pc.defineObject("proto","PrototypeData",0,nullptr) ;
1224 pc.defineString("dsetName","Name",0,"") ;
1225 pc.defineInt("randProto","PrototypeData",0,0) ;
1226 pc.defineInt("resampleProto","PrototypeData",1,0) ;
1227 pc.defineInt("verbose","Verbose",0,0) ;
1228 pc.defineInt("extended","Extended",0,0) ;
1229 pc.defineInt("nEvents","NumEvents",0,0) ;
1230 pc.defineInt("autoBinned","AutoBinned",0,1) ;
1231 pc.defineInt("expectedData","ExpectedData",0,0) ;
1232 pc.defineDouble("nEventsD","NumEventsD",0,-1.) ;
1233 pc.defineString("binnedTag","GenBinned",0,"") ;
1234 pc.defineMutex("GenBinned","ProtoData") ;
1235 pc.defineMutex("Extended", "NumEvents");
1236
1237 // Process and check varargs
1239 if (!pc.ok(true)) {
1240 return nullptr;
1241 }
1242
1243 // Decode command line arguments
1244 RooDataSet* protoData = static_cast<RooDataSet*>(pc.getObject("proto",nullptr)) ;
1245 const char* dsetName = pc.getString("dsetName") ;
1246 bool verbose = pc.getInt("verbose") ;
1247 bool randProto = pc.getInt("randProto") ;
1248 bool resampleProto = pc.getInt("resampleProto") ;
1249 bool extended = pc.getInt("extended") ;
1250 bool autoBinned = pc.getInt("autoBinned") ;
1251 const char* binnedTag = pc.getString("binnedTag") ;
1252 Int_t nEventsI = pc.getInt("nEvents") ;
1253 double nEventsD = pc.getInt("nEventsD") ;
1254 //bool verbose = pc.getInt("verbose") ;
1255 bool expectedData = pc.getInt("expectedData") ;
1256
1257 double nEvents = (nEventsD>0) ? nEventsD : double(nEventsI);
1258
1259 // Force binned mode for expected data mode
1260 if (expectedData) {
1261 binnedTag="*" ;
1262 }
1263
1264 if (extended) {
1265 if (nEvents == 0) nEvents = expectedEvents(&whatVars);
1266 } else if (nEvents==0) {
1267 cxcoutI(Generation) << "No number of events specified , number of events generated is "
1268 << GetName() << "::expectedEvents() = " << expectedEvents(&whatVars)<< std::endl ;
1269 }
1270
1271 if (extended && protoData && !randProto) {
1272 cxcoutI(Generation) << "WARNING Using generator option Extended() (Poisson distribution of #events) together "
1273 << "with a prototype dataset implies incomplete sampling or oversampling of proto data. "
1274 << "Set randomize flag in ProtoData() option to randomize prototype dataset order and thus "
1275 << "to randomize the set of over/undersampled prototype events for each generation cycle." << std::endl ;
1276 }
1277
1278
1279 // Forward to appropriate implementation
1280 std::unique_ptr<RooDataSet> data;
1281 if (protoData) {
1282 data = std::unique_ptr<RooDataSet>{generate(whatVars,*protoData,Int_t(nEvents),verbose,randProto,resampleProto)};
1283 } else {
1284 data = std::unique_ptr<RooDataSet>{generate(whatVars,nEvents,verbose,autoBinned,binnedTag,expectedData, extended)};
1285 }
1286
1287 // Rename dataset to given name if supplied
1288 if (dsetName && strlen(dsetName)>0) {
1289 data->SetName(dsetName) ;
1290 }
1291
1292 return RooFit::makeOwningPtr(std::move(data));
1293}
1294
1295
1296
1297
1298
1299
1300////////////////////////////////////////////////////////////////////////////////
1301/// \note This method does not perform any generation. To generate according to generations specification call RooAbsPdf::generate(RooAbsPdf::GenSpec&) const
1302///
1303/// Details copied from RooAbsPdf::generate():
1304/// --------------------------------------------
1305/// \copydetails RooAbsPdf::generate(const RooArgSet&,const RooCmdArg&,const RooCmdArg&,const RooCmdArg&,const RooCmdArg&,const RooCmdArg&,const RooCmdArg&)
1306
1308 const RooCmdArg& arg1,const RooCmdArg& arg2,
1309 const RooCmdArg& arg3,const RooCmdArg& arg4,
1310 const RooCmdArg& arg5,const RooCmdArg& arg6)
1311{
1312
1313 // Select the pdf-specific commands
1314 RooCmdConfig pc("RooAbsPdf::generate(" + std::string(GetName()) + ")");
1315 pc.defineObject("proto","PrototypeData",0,nullptr) ;
1316 pc.defineString("dsetName","Name",0,"") ;
1317 pc.defineInt("randProto","PrototypeData",0,0) ;
1318 pc.defineInt("resampleProto","PrototypeData",1,0) ;
1319 pc.defineInt("verbose","Verbose",0,0) ;
1320 pc.defineInt("extended","Extended",0,0) ;
1321 pc.defineInt("nEvents","NumEvents",0,0) ;
1322 pc.defineInt("autoBinned","AutoBinned",0,1) ;
1323 pc.defineString("binnedTag","GenBinned",0,"") ;
1324 pc.defineMutex("GenBinned","ProtoData") ;
1325
1326
1327 // Process and check varargs
1329 if (!pc.ok(true)) {
1330 return nullptr ;
1331 }
1332
1333 // Decode command line arguments
1334 RooDataSet* protoData = static_cast<RooDataSet*>(pc.getObject("proto",nullptr)) ;
1335 const char* dsetName = pc.getString("dsetName") ;
1336 Int_t nEvents = pc.getInt("nEvents") ;
1337 bool verbose = pc.getInt("verbose") ;
1338 bool randProto = pc.getInt("randProto") ;
1339 bool resampleProto = pc.getInt("resampleProto") ;
1340 bool extended = pc.getInt("extended") ;
1341 bool autoBinned = pc.getInt("autoBinned") ;
1342 const char* binnedTag = pc.getString("binnedTag") ;
1343
1345
1346 return new GenSpec(cx,whatVars,protoData,nEvents,extended,randProto,resampleProto,dsetName) ;
1347}
1348
1349
1350////////////////////////////////////////////////////////////////////////////////
1351/// If many identical generation requests
1352/// are needed, e.g. in toy MC studies, it is more efficient to use the prepareMultiGen()/generate()
1353/// combination than calling the standard generate() multiple times as
1354/// initialization overhead is only incurred once.
1355
1357{
1358 //Int_t nEvt = spec._extended ? RooRandom::randomGenerator()->Poisson(spec._nGen) : spec._nGen ;
1359 //Int_t nEvt = spec._extended ? RooRandom::randomGenerator()->Poisson(spec._nGen==0?expectedEvents(spec._whatVars):spec._nGen) : spec._nGen ;
1360 //Int_t nEvt = spec._nGen == 0 ? RooRandom::randomGenerator()->Poisson(expectedEvents(spec._whatVars)) : spec._nGen;
1361
1362 double nEvt = spec._nGen == 0 ? expectedEvents(spec._whatVars) : spec._nGen;
1363
1364 std::unique_ptr<RooDataSet> ret{generate(*spec._genContext,spec._whatVars,spec._protoData, nEvt,false,spec._randProto,spec._resampleProto,
1365 spec._init,spec._extended)};
1366 spec._init = true ;
1367 return RooFit::makeOwningPtr(std::move(ret));
1368}
1369
1370
1371
1372
1373
1374////////////////////////////////////////////////////////////////////////////////
1375/// Generate a new dataset containing the specified variables with
1376/// events sampled from our distribution.
1377///
1378/// \param[in] whatVars Generate a dataset with the variables (and categories) in this set.
1379/// Any variables of this PDF that are not in `whatVars` will use their
1380/// current values and be treated as fixed parameters.
1381/// \param[in] nEvents Generate the specified number of events or else try to use
1382/// expectedEvents() if nEvents <= 0 (default).
1383/// \param[in] verbose Show which generator strategies are being used.
1384/// \param[in] autoBinned If original distribution is binned, return bin centers and randomise weights
1385/// instead of generating single events.
1386/// \param[in] binnedTag
1387/// \param[in] expectedData Call setExpectedData on the genContext.
1388/// \param[in] extended Randomise number of events generated according to Poisson(nEvents). Only useful
1389/// if PDF is extended.
1390/// \return New dataset. Returns zero in case of an error. The caller takes ownership of the returned
1391/// dataset.
1392
1393RooFit::OwningPtr<RooDataSet> RooAbsPdf::generate(const RooArgSet &whatVars, double nEvents, bool verbose, bool autoBinned, const char* binnedTag, bool expectedData, bool extended) const
1394{
1395 if (nEvents==0 && extendMode()==CanNotBeExtended) {
1396 return RooFit::makeOwningPtr(std::make_unique<RooDataSet>("emptyData","emptyData",whatVars));
1397 }
1398
1399 // Request for binned generation
1400 std::unique_ptr<RooAbsGenContext> context{autoGenContext(whatVars,nullptr,nullptr,verbose,autoBinned,binnedTag)};
1401 if (expectedData) {
1402 context->setExpectedData(true) ;
1403 }
1404
1405 std::unique_ptr<RooDataSet> generated;
1406 if(nullptr != context && context->isValid()) {
1407 generated = std::unique_ptr<RooDataSet>{context->generate(nEvents, false, extended)};
1408 }
1409 else {
1410 coutE(Generation) << "RooAbsPdf::generate(" << GetName() << ") cannot create a valid context" << std::endl;
1411 }
1412 return RooFit::makeOwningPtr(std::move(generated));
1413}
1414
1415
1416
1417
1418////////////////////////////////////////////////////////////////////////////////
1419/// Internal method
1420
1421std::unique_ptr<RooDataSet> RooAbsPdf::generate(RooAbsGenContext& context, const RooArgSet &whatVars, const RooDataSet *prototype,
1422 double nEvents, bool /*verbose*/, bool randProtoOrder, bool resampleProto,
1423 bool skipInit, bool extended) const
1424{
1425 if (nEvents==0 && (prototype==nullptr || prototype->numEntries()==0)) {
1426 return std::make_unique<RooDataSet>("emptyData","emptyData",whatVars);
1427 }
1428
1429 std::unique_ptr<RooDataSet> generated;
1430
1431 // Resampling implies reshuffling in the implementation
1432 if (resampleProto) {
1434 }
1435
1436 if (randProtoOrder && prototype && prototype->numEntries()!=nEvents) {
1437 coutI(Generation) << "RooAbsPdf::generate (Re)randomizing event order in prototype dataset (Nevt=" << nEvents << ")" << std::endl ;
1438 Int_t* newOrder = randomizeProtoOrder(prototype->numEntries(),Int_t(nEvents),resampleProto) ;
1439 context.setProtoDataOrder(newOrder) ;
1440 delete[] newOrder ;
1441 }
1442
1443 if(context.isValid()) {
1444 generated = std::unique_ptr<RooDataSet>{context.generate(nEvents,skipInit,extended)};
1445 }
1446 else {
1447 coutE(Generation) << "RooAbsPdf::generate(" << GetName() << ") do not have a valid generator context" << std::endl;
1448 }
1449 return generated;
1450}
1451
1452
1453
1454
1455////////////////////////////////////////////////////////////////////////////////
1456/// Generate a new dataset using a prototype dataset as a model,
1457/// with values of the variables in `whatVars` sampled from our distribution.
1458///
1459/// \param[in] whatVars Generate for these variables.
1460/// \param[in] prototype Use this dataset
1461/// as a prototype: the new dataset will contain the same number of
1462/// events as the prototype (by default), and any prototype variables not in
1463/// whatVars will be copied into the new dataset for each generated
1464/// event and also used to set our PDF parameters. The user can specify a
1465/// number of events to generate that will override the default. The result is a
1466/// copy of the prototype dataset with only variables in whatVars
1467/// randomized. Variables in whatVars that are not in the prototype
1468/// will be added as new columns to the generated dataset.
1469/// \param[in] nEvents Number of events to generate. Defaults to 0, which means number
1470/// of event in prototype dataset.
1471/// \param[in] verbose Show which generator strategies are being used.
1472/// \param[in] randProtoOrder Randomise order of retrieval of events from proto dataset.
1473/// \param[in] resampleProto Resample from the proto dataset.
1474/// \return The new dataset. Returns zero in case of an error. The caller takes ownership of the
1475/// returned dataset.
1476
1478 Int_t nEvents, bool verbose, bool randProtoOrder, bool resampleProto) const
1479{
1480 std::unique_ptr<RooAbsGenContext> context{genContext(whatVars,&prototype,nullptr,verbose)};
1481 if (context) {
1483 }
1484 coutE(Generation) << "RooAbsPdf::generate(" << GetName() << ") ERROR creating generator context" << std::endl ;
1485 return nullptr;
1486}
1487
1488
1489
1490////////////////////////////////////////////////////////////////////////////////
1491/// Return lookup table with randomized order for nProto prototype events.
1492
1494{
1495 // Make output list
1496 Int_t* lut = new Int_t[nProto] ;
1497
1498 // Randomly sample input list into output list
1499 if (!resampleProto) {
1500 // In this mode, randomization is a strict reshuffle of the order
1501 std::iota(lut, lut + nProto, 0); // fill the vector with 0 to nProto - 1
1502 // Shuffle code taken from https://en.cppreference.com/w/cpp/algorithm/random_shuffle.
1503 // The std::random_shuffle function was deprecated in C++17. We could have
1504 // used std::shuffle instead, but this is not straight-forward to use with
1505 // RooRandom::integer() and we didn't want to change the random number
1506 // generator. It might cause unwanted effects like reproducibility problems.
1507 for (int i = nProto-1; i > 0; --i) {
1508 std::swap(lut[i], lut[RooRandom::integer(i+1)]);
1509 }
1510 } else {
1511 // In this mode, we resample, i.e. events can be used more than once
1512 std::generate(lut, lut + nProto, [&]{ return RooRandom::integer(nProto); });
1513 }
1514
1515
1516 return lut ;
1517}
1518
1519
1520
1521////////////////////////////////////////////////////////////////////////////////
1522/// Load generatedVars with the subset of directVars that we can generate events for,
1523/// and return a code that specifies the generator algorithm we will use. A code of
1524/// zero indicates that we cannot generate any of the directVars (in this case, nothing
1525/// should be added to generatedVars). Any non-zero codes will be passed to our generateEvent()
1526/// implementation, but otherwise its value is arbitrary. The default implementation of
1527/// this method returns zero. Subclasses will usually implement this method using the
1528/// matchArgs() methods to advertise the algorithms they provide.
1529
1530Int_t RooAbsPdf::getGenerator(const RooArgSet &/*directVars*/, RooArgSet &/*generatedVars*/, bool /*staticInitOK*/) const
1531{
1532 return 0 ;
1533}
1534
1535
1536
1537////////////////////////////////////////////////////////////////////////////////
1538/// Interface for one-time initialization to setup the generator for the specified code.
1539
1541{
1542}
1543
1544
1545
1546////////////////////////////////////////////////////////////////////////////////
1547/// Interface for generation of an event using the algorithm
1548/// corresponding to the specified code. The meaning of each code is
1549/// defined by the getGenerator() implementation. The default
1550/// implementation does nothing.
1551
1553{
1554}
1555
1556
1557
1558////////////////////////////////////////////////////////////////////////////////
1559/// Check if given observable can be safely generated using the
1560/// pdfs internal generator mechanism (if that existsP). Observables
1561/// on which a PDF depends via more than route are not safe
1562/// for use with internal generators because they introduce
1563/// correlations not known to the internal generator
1564
1566{
1567 // Arg must be direct server of self
1568 if (!findServer(arg.GetName())) return false ;
1569
1570 // There must be no other dependency routes
1571 for (const auto server : _serverList) {
1572 if(server == &arg) continue;
1573 if(server->dependsOn(arg)) {
1574 return false ;
1575 }
1576 }
1577
1578 return true ;
1579}
1580
1581
1582////////////////////////////////////////////////////////////////////////////////
1583/// Generate a new dataset containing the specified variables with events sampled from our distribution.
1584/// \param[in] whatVars Choose variables in which to generate events. Variables not listed here will remain
1585/// constant and not be used for event generation
1586/// \param[in] arg1,arg2,arg3,arg4,arg5,arg6 Optional RooCmdArg to change behaviour of generateBinned()
1587/// \return RooDataHist *, to be managed by caller.
1588///
1589/// Generate the specified number of events or expectedEvents() if not specified.
1590///
1591/// Any variables of this PDF that are not in whatVars will use their
1592/// current values and be treated as fixed parameters. Returns zero
1593/// in case of an error. The caller takes ownership of the returned
1594/// dataset.
1595///
1596/// The following named arguments are supported
1597/// | Type of CmdArg | Effect on generation
1598/// |---------------------------|-----------------------
1599/// | `Name(const char* name)` | Name of the output dataset
1600/// | `Verbose(bool flag)` | Print informational messages during event generation
1601/// | `NumEvents(int nevt)` | Generate specified number of events
1602/// | `Extended()` | The actual number of events generated will be sampled from a Poisson distribution with mu=nevt. This can be *much* faster for peaked PDFs, but the number of events is not exactly what was requested.
1603/// | `ExpectedData()` | Return a binned dataset _without_ statistical fluctuations (also aliased as Asimov())
1604///
1605
1607 const RooCmdArg& arg3,const RooCmdArg& arg4, const RooCmdArg& arg5,const RooCmdArg& arg6) const
1608{
1609
1610 // Select the pdf-specific commands
1611 RooCmdConfig pc("RooAbsPdf::generate(" + std::string(GetName()) + ")");
1612 pc.defineString("dsetName","Name",0,"") ;
1613 pc.defineInt("verbose","Verbose",0,0) ;
1614 pc.defineInt("extended","Extended",0,0) ;
1615 pc.defineInt("nEvents","NumEvents",0,0) ;
1616 pc.defineDouble("nEventsD","NumEventsD",0,-1.) ;
1617 pc.defineInt("expectedData","ExpectedData",0,0) ;
1618
1619 // Process and check varargs
1621 if (!pc.ok(true)) {
1622 return nullptr;
1623 }
1624
1625 // Decode command line arguments
1626 double nEvents = pc.getDouble("nEventsD") ;
1627 if (nEvents<0) {
1628 nEvents = pc.getInt("nEvents") ;
1629 }
1630 //bool verbose = pc.getInt("verbose") ;
1631 bool extended = pc.getInt("extended") ;
1632 bool expectedData = pc.getInt("expectedData") ;
1633 const char* dsetName = pc.getString("dsetName") ;
1634
1635 if (extended) {
1636 //nEvents = (nEvents==0?Int_t(expectedEvents(&whatVars)+0.5):nEvents) ;
1637 nEvents = (nEvents==0 ? expectedEvents(&whatVars) :nEvents) ;
1638 cxcoutI(Generation) << " Extended mode active, number of events generated (" << nEvents << ") is Poisson fluctuation on "
1639 << GetName() << "::expectedEvents() = " << nEvents << std::endl ;
1640 // If Poisson fluctuation results in zero events, stop here
1641 if (nEvents==0) {
1642 return nullptr ;
1643 }
1644 } else if (nEvents==0) {
1645 cxcoutI(Generation) << "No number of events specified , number of events generated is "
1646 << GetName() << "::expectedEvents() = " << expectedEvents(&whatVars)<< std::endl ;
1647 }
1648
1649 // Forward to appropriate implementation
1651
1652 // Rename dataset to given name if supplied
1653 if (dsetName && strlen(dsetName)>0) {
1654 data->SetName(dsetName) ;
1655 }
1656
1657 return data;
1658}
1659
1660
1661
1662
1663////////////////////////////////////////////////////////////////////////////////
1664/// Generate a new dataset containing the specified variables with
1665/// events sampled from our distribution.
1666///
1667/// \param[in] whatVars Variables that values should be generated for.
1668/// \param[in] nEvents How many events to generate. If `nEvents <=0`, use the value returned by expectedEvents() as target.
1669/// \param[in] expectedData If set to true (false by default), the returned histogram returns the 'expected'
1670/// data sample, i.e. no statistical fluctuations are present.
1671/// \param[in] extended For each bin, generate Poisson(x, mu) events, where `mu` is chosen such that *on average*,
1672/// one would obtain `nEvents` events. This means that the true number of events will fluctuate around the desired value,
1673/// but the generation happens a lot faster.
1674/// Especially if the PDF is sharply peaked, the multinomial event generation necessary to generate *exactly* `nEvents` events can
1675/// be very slow.
1676///
1677/// The binning used for generation of events is the currently set binning for the variables.
1678/// It can e.g. be changed using
1679/// ```
1680/// x.setBins(15);
1681/// x.setRange(-5., 5.);
1682/// pdf.generateBinned(RooArgSet(x), 1000);
1683/// ```
1684///
1685/// Any variables of this PDF that are not in `whatVars` will use their
1686/// current values and be treated as fixed parameters.
1687/// \return RooDataHist* owned by the caller. Returns `nullptr` in case of an error.
1689{
1690 // Create empty RooDataHist
1691 auto hist = std::make_unique<RooDataHist>("genData","genData",whatVars);
1692
1693 // Scale to number of events and introduce Poisson fluctuations
1694 if (nEvents<=0) {
1695 if (!canBeExtended()) {
1696 coutE(InputArguments) << "RooAbsPdf::generateBinned(" << GetName() << ") ERROR: No event count provided and p.d.f does not provide expected number of events" << std::endl ;
1697 return nullptr;
1698 } else {
1699
1700 // Don't round in expectedData or extended mode
1701 if (expectedData || extended) {
1702 nEvents = expectedEvents(&whatVars) ;
1703 } else {
1704 nEvents = std::round(expectedEvents(&whatVars));
1705 }
1706 }
1707 }
1708
1709 // Sample p.d.f. distribution
1710 fillDataHist(hist.get(),&whatVars,1,true) ;
1711
1712 vector<int> histOut(hist->numEntries()) ;
1713 double histMax(-1) ;
1714 Int_t histOutSum(0) ;
1715 for (int i=0 ; i<hist->numEntries() ; i++) {
1716 hist->get(i) ;
1717 if (expectedData) {
1718
1719 // Expected data, multiply p.d.f by nEvents
1720 double w=hist->weight()*nEvents ;
1721 hist->set(i, w, sqrt(w));
1722
1723 } else if (extended) {
1724
1725 // Extended mode, set contents to Poisson(pdf*nEvents)
1726 double w = RooRandom::randomGenerator()->Poisson(hist->weight()*nEvents) ;
1727 hist->set(w,sqrt(w)) ;
1728
1729 } else {
1730
1731 // Regular mode, fill array of weights with Poisson(pdf*nEvents), but to not fill
1732 // histogram yet.
1733 if (hist->weight()>histMax) {
1734 histMax = hist->weight() ;
1735 }
1736 histOut[i] = RooRandom::randomGenerator()->Poisson(hist->weight()*nEvents) ;
1737 histOutSum += histOut[i] ;
1738 }
1739 }
1740
1741
1742 if (!expectedData && !extended) {
1743
1744 // Second pass for regular mode - Trim/Extend dataset to exact number of entries
1745
1746 // Calculate difference between what is generated so far and what is requested
1747 Int_t nEvtExtra = std::abs(Int_t(nEvents)-histOutSum) ;
1748 Int_t wgt = (histOutSum>nEvents) ? -1 : 1 ;
1749
1750 // Perform simple binned accept/reject procedure to get to exact event count
1751 std::size_t counter = 0;
1752 bool havePrintedInfo = false;
1753 while(nEvtExtra>0) {
1754
1755 Int_t ibinRand = RooRandom::randomGenerator()->Integer(hist->numEntries()) ;
1756 hist->get(ibinRand) ;
1757 double ranY = RooRandom::randomGenerator()->Uniform(histMax) ;
1758
1759 if (ranY<hist->weight()) {
1760 if (wgt==1) {
1761 histOut[ibinRand]++ ;
1762 } else {
1763 // If weight is negative, prior bin content must be at least 1
1764 if (histOut[ibinRand]>0) {
1765 histOut[ibinRand]-- ;
1766 } else {
1767 continue ;
1768 }
1769 }
1770 nEvtExtra-- ;
1771 }
1772
1773 if ((counter++ > 10*nEvents || nEvents > 1.E7) && !havePrintedInfo) {
1774 havePrintedInfo = true;
1775 coutP(Generation) << "RooAbsPdf::generateBinned(" << GetName() << ") Performing costly accept/reject sampling. If this takes too long, use "
1776 << "extended mode to speed up the process." << std::endl;
1777 }
1778 }
1779
1780 // Transfer working array to histogram
1781 for (int i=0 ; i<hist->numEntries() ; i++) {
1782 hist->get(i) ;
1783 hist->set(histOut[i],sqrt(1.0*histOut[i])) ;
1784 }
1785
1786 } else if (expectedData) {
1787
1788 // Second pass for expectedData mode -- Normalize to exact number of requested events
1789 // Minor difference may be present in first round due to difference between
1790 // bin average and bin integral in sampling bins
1791 double corr = nEvents/hist->sumEntries() ;
1792 for (int i=0 ; i<hist->numEntries() ; i++) {
1793 hist->get(i) ;
1794 hist->set(hist->weight()*corr,sqrt(hist->weight()*corr)) ;
1795 }
1796
1797 }
1798
1799 return RooFit::makeOwningPtr(std::move(hist));
1800}
1801
1802
1803
1804////////////////////////////////////////////////////////////////////////////////
1805/// Special generator interface for generation of 'global observables' -- for RooStats tools
1806
1811
1812namespace {
1813void removeRangeOverlap(std::vector<std::pair<double, double>>& ranges) {
1814 //Sort from left to right
1815 std::sort(ranges.begin(), ranges.end());
1816
1817 for (auto it = ranges.begin(); it != ranges.end(); ++it) {
1818 double& startL = it->first;
1819 double& endL = it->second;
1820
1821 for (auto innerIt = it+1; innerIt != ranges.end(); ++innerIt) {
1822 const double startR = innerIt->first;
1823 const double endR = innerIt->second;
1824
1825 if (startL <= startR && startR <= endL) {
1826 //Overlapping ranges, extend left one
1827 endL = std::max(endL, endR);
1828 *innerIt = make_pair(0., 0.);
1829 }
1830 }
1831 }
1832
1833 auto newEnd = std::remove_if(ranges.begin(), ranges.end(),
1834 [](const std::pair<double,double>& input){
1835 return input.first == input.second;
1836 });
1837 ranges.erase(newEnd, ranges.end());
1838}
1839}
1840
1841
1842////////////////////////////////////////////////////////////////////////////////
1843/// Plot (project) PDF on specified frame.
1844/// - If a PDF is plotted in an empty frame, it
1845/// will show a unit-normalized curve in the frame variable. When projecting a multi-
1846/// dimensional PDF onto the frame axis, hidden parameters are taken are taken at
1847/// their current value.
1848/// - If a PDF is plotted in a frame in which a dataset has already been plotted, it will
1849/// show a projection integrated over all variables that were present in the shown
1850/// dataset (except for the one on the x-axis). The normalization of the curve will
1851/// be adjusted to the event count of the plotted dataset. An informational message
1852/// will be printed for each projection step that is performed.
1853/// - If a PDF is plotted in a frame showing a dataset *after* a fit, the above happens,
1854/// but the PDF will be drawn and normalised only in the fit range. If this is not desired,
1855/// plotting and normalisation range can be overridden using Range() and NormRange() as
1856/// documented in the table below.
1857///
1858/// This function takes the following named arguments (for more arguments, see also
1859/// RooAbsReal::plotOn(RooPlot*,const RooCmdArg&,const RooCmdArg&,const RooCmdArg&,const RooCmdArg&,
1860/// const RooCmdArg&,const RooCmdArg&,const RooCmdArg&,const RooCmdArg&,const RooCmdArg&,
1861/// const RooCmdArg&) const )
1862///
1863///
1864/// <table>
1865/// <tr><th> Type of argument <th> Controlling normalisation
1866/// <tr><td> `NormRange(const char* name)` <td> Calculate curve normalization w.r.t. specified range[s].
1867/// See the tutorial rf212_plottingInRanges_blinding.C
1868/// \note Setting a Range() by default also sets a NormRange() on the same range, meaning that the
1869/// PDF is plotted and normalised in the same range. Overriding this can be useful if the PDF was fit
1870/// in limited range[s] such as side bands, `NormRange("sidebandLeft,sidebandRight")`, but the PDF
1871/// should be drawn in the full range, `Range("")`.
1872///
1873/// <tr><td> `Normalization(double scale, ScaleType code)` <td> Adjust normalization by given scale factor.
1874/// Interpretation of number depends on code:
1875/// `RooAbsReal::Relative`: relative adjustment factor
1876/// `RooAbsReal::NumEvent`: scale to match given number of events.
1877///
1878/// <tr><th> Type of argument <th> Misc control
1879/// <tr><td> `Name(const chat* name)` <td> Give curve specified name in frame. Useful if curve is to be referenced later
1880/// <tr><td> `Asymmetry(const RooCategory& c)` <td> Show the asymmetry of the PDF in given two-state category
1881/// \f$ \frac{F(+)-F(-)}{F(+)+F(-)} \f$ rather than the PDF projection. Category must have two
1882/// states with indices -1 and +1 or three states with indices -1,0 and +1.
1883/// <tr><td> `ShiftToZero(bool flag)` <td> Shift entire curve such that lowest visible point is at exactly zero.
1884/// Mostly useful when plotting -log(L) or \f$ \chi^2 \f$ distributions
1885/// <tr><td> `AddTo(const char* name, double_t wgtSelf, double_t wgtOther)` <td> Create a projection of this PDF onto the x-axis, but
1886/// instead of plotting it directly, add it to an existing curve with given name (and relative weight factors).
1887/// <tr><td> `Components(const char* names)` <td> When plotting sums of PDFs, plot only the named components (*e.g.* only
1888/// the signal of a signal+background model).
1889/// <tr><td> `Components(const RooArgSet& compSet)` <td> As above, but pass a RooArgSet of the components themselves.
1890///
1891/// <tr><th> Type of argument <th> Projection control
1892/// <tr><td> `Slice(const RooArgSet& set)` <td> Override default projection behaviour by omitting observables listed
1893/// in set from the projection, i.e. by not integrating over these.
1894/// Slicing is usually only sensible in discrete observables, by e.g. creating a slice
1895/// of the PDF at the current value of the category observable.
1896/// <tr><td> `Slice(RooCategory& cat, const char* label)` <td> Override default projection behaviour by omitting the specified category
1897/// observable from the projection, i.e., by not integrating over all states of this category.
1898/// The slice is positioned at the given label value. Multiple Slice() commands can be given to specify slices
1899/// in multiple observables, e.g.
1900/// ```{.cpp}
1901/// pdf.plotOn(frame, Slice(tagCategory, "2tag"), Slice(jetCategory, "3jet"));
1902/// ```
1903/// <tr><td> `Project(const RooArgSet& set)` <td> Override default projection behaviour by projecting
1904/// over observables given in set, completely ignoring the default projection behavior. Advanced use only.
1905/// <tr><td> `ProjWData(const RooAbsData& d)` <td> Override default projection _technique_ (integration). For observables
1906/// present in given dataset projection of PDF is achieved by constructing an average over all observable
1907/// values in given set. Consult RooFit plotting tutorial for further explanation of meaning & use of this technique
1908/// <tr><td> `ProjWData(const RooArgSet& s, const RooAbsData& d)` <td> As above but only consider subset 's' of
1909/// observables in dataset 'd' for projection through data averaging
1910/// <tr><td> `ProjectionRange(const char* rn)` <td> When projecting the PDF onto the plot axis, it is usually integrated
1911/// over the full range of the invisible variables. The ProjectionRange overrides this.
1912/// This is useful if the PDF was fitted in a limited range in y, but it is now projected onto x. If
1913/// `ProjectionRange("<name of fit range>")` is passed, the projection is normalised correctly.
1914///
1915/// <tr><th> Type of argument <th> Plotting control
1916/// <tr><td> `LineStyle(Int_t style)` <td> Select line style by ROOT line style code, default is solid
1917/// <tr><td> `LineColor(Int_t color)` <td> Select line color by ROOT color code, default is blue
1918/// <tr><td> `LineWidth(Int_t width)` <td> Select line with in pixels, default is 3
1919/// <tr><td> `FillStyle(Int_t style)` <td> Select fill style, default is not filled. If a filled style is selected,
1920/// also use VLines() to add vertical downward lines at end of curve to ensure proper closure
1921/// <tr><td> `FillColor(Int_t color)` <td> Select fill color by ROOT color code
1922/// <tr><td> `Range(const char* name)` <td> Only draw curve in range defined by given name. Multiple comma-separated ranges can be given.
1923/// An empty string "" or `nullptr` means to use the default range of the variable.
1924/// <tr><td> `Range(double lo, double hi)` <td> Only draw curve in specified range
1925/// <tr><td> `VLines()` <td> Add vertical lines to y=0 at end points of curve
1926/// <tr><td> `Precision(double eps)` <td> Control precision of drawn curve w.r.t to scale of plot, default is 1e-3. A higher precision will
1927/// result in more and more densely spaced curve points. A negative precision value will disable
1928/// adaptive point spacing and restrict sampling to the grid point of points defined by the binning
1929/// of the plotted observable (recommended for expensive functions such as profile likelihoods)
1930/// <tr><td> `Invisible(bool flag)` <td> Add curve to frame, but do not display. Useful in combination AddTo()
1931/// <tr><td> `VisualizeError(const RooFitResult& fitres, double Z=1, bool linearMethod=true)`
1932/// <td> Visualize the uncertainty on the parameters, as given in fitres, at 'Z' sigma.
1933/// The linear method is fast but may not be accurate in the presence of strong correlations (~>0.9) and at Z>2 due to linear and Gaussian approximations made.
1934/// Intervals from the sampling method can be asymmetric, and may perform better in the presence of strong correlations, but may take (much) longer to calculate
1935/// \note To include the uncertainty from the expected number of events,
1936/// the Normalization() argument with `ScaleType` `RooAbsReal::RelativeExpected` has to be passed, e.g.
1937/// ```{.cpp}
1938/// pdf.plotOn(frame, VisualizeError(fitResult), Normalization(1.0, RooAbsReal::RelativeExpected));
1939/// ```
1940///
1941/// <tr><td> `VisualizeError(const RooFitResult& fitres, const RooArgSet& param, double Z=1, bool linearMethod=true)`
1942/// <td> Visualize the uncertainty on the subset of parameters 'param', as given in fitres, at 'Z' sigma
1943/// </table>
1944
1946{
1947
1948 // Pre-processing if p.d.f. contains a fit range and there is no command specifying one,
1949 // add a fit range as default range
1950 std::unique_ptr<RooCmdArg> plotRange;
1951 std::unique_ptr<RooCmdArg> normRange2;
1952 if (getStringAttribute("fitrange") && !cmdList.FindObject("Range") &&
1953 !cmdList.FindObject("RangeWithName")) {
1954 plotRange.reset(static_cast<RooCmdArg*>(RooFit::Range(getStringAttribute("fitrange")).Clone()));
1955 cmdList.Add(plotRange.get());
1956 }
1957
1958 if (getStringAttribute("fitrange") && !cmdList.FindObject("NormRange")) {
1959 normRange2.reset(static_cast<RooCmdArg*>(RooFit::NormRange(getStringAttribute("fitrange")).Clone()));
1960 cmdList.Add(normRange2.get());
1961 }
1962
1963 if (plotRange || normRange2) {
1964 coutI(Plotting) << "RooAbsPdf::plotOn(" << GetName() << ") p.d.f was fitted in a subrange and no explicit "
1965 << (plotRange?"Range()":"") << ((plotRange&&normRange2)?" and ":"")
1966 << (normRange2?"NormRange()":"") << " was specified. Plotting / normalising in fit range. To override, do one of the following"
1967 << "\n\t- Clear the automatic fit range attribute: <pdf>.removeStringAttribute(\"fitrange\");"
1968 << "\n\t- Explicitly specify the plotting range: Range(\"<rangeName>\")."
1969 << "\n\t- Explicitly specify where to compute the normalisation: NormRange(\"<rangeName>\")."
1970 << "\n\tThe default (full) range can be denoted with Range(\"\") / NormRange(\"\")."<< std::endl ;
1971 }
1972
1973 // Sanity checks
1974 if (plotSanityChecks(frame)) return frame ;
1975
1976 // Select the pdf-specific commands
1977 RooCmdConfig pc("RooAbsPdf::plotOn(" + std::string(GetName()) + ")");
1978 pc.defineDouble("scaleFactor","Normalization",0,1.0) ;
1979 pc.defineInt("scaleType","Normalization",0,Relative) ;
1980 pc.defineSet("compSet","SelectCompSet",0) ;
1981 pc.defineString("compSpec","SelectCompSpec",0) ;
1982 pc.defineObject("asymCat","Asymmetry",0) ;
1983 pc.defineDouble("rangeLo","Range",0,-999.) ;
1984 pc.defineDouble("rangeHi","Range",1,-999.) ;
1985 pc.defineString("rangeName","RangeWithName",0,"") ;
1986 pc.defineString("normRangeName","NormRange",0,"") ;
1987 pc.defineInt("rangeAdjustNorm","Range",0,0) ;
1988 pc.defineInt("rangeWNAdjustNorm","RangeWithName",0,0) ;
1989 pc.defineMutex("SelectCompSet","SelectCompSpec") ;
1990 pc.defineMutex("Range","RangeWithName") ;
1991 pc.allowUndefined() ; // unknowns may be handled by RooAbsReal
1992
1993 // Process and check varargs
1994 pc.process(cmdList) ;
1995 if (!pc.ok(true)) {
1996 return frame ;
1997 }
1998
1999 // Decode command line arguments
2000 ScaleType stype = (ScaleType) pc.getInt("scaleType") ;
2001 double scaleFactor = pc.getDouble("scaleFactor") ;
2002 const RooAbsCategoryLValue* asymCat = static_cast<const RooAbsCategoryLValue*>(pc.getObject("asymCat")) ;
2003 const char* compSpec = pc.getString("compSpec") ;
2004 const RooArgSet* compSet = pc.getSet("compSet");
2005 bool haveCompSel = ( (compSpec && strlen(compSpec)>0) || compSet) ;
2006
2007 // Suffix for curve name
2008 std::stringstream nameSuffix;
2009 if (compSpec && strlen(compSpec)>0) {
2010 nameSuffix << "_Comp[" << compSpec << "]";
2011 } else if (compSet) {
2012 nameSuffix << "_Comp[" << compSet->contentsString() << "]";
2013 }
2014
2015 // Remove PDF-only commands from command list
2016 RooCmdConfig::stripCmdList(cmdList,"SelectCompSet,SelectCompSpec") ;
2017
2018 // Adjust normalization, if so requested
2019 if (asymCat) {
2020 RooCmdArg cnsuffix("CurveNameSuffix", 0, 0, 0, 0, nameSuffix.str().c_str(), nullptr, nullptr, nullptr);
2021 cmdList.Add(&cnsuffix);
2022 return RooAbsReal::plotOn(frame, cmdList);
2023 }
2024
2025 // More sanity checks
2026 double nExpected(1) ;
2027 if (stype==RelativeExpected) {
2028 if (!canBeExtended()) {
2029 coutE(Plotting) << "RooAbsPdf::plotOn(" << GetName()
2030 << "): ERROR the 'Expected' scale option can only be used on extendable PDFs" << std::endl ;
2031 return frame ;
2032 }
2033 frame->updateNormVars(*frame->getPlotVar()) ;
2035 }
2036
2037 if (stype != Raw) {
2038
2039 if (frame->getFitRangeNEvt() && stype==Relative) {
2040
2041 bool hasCustomRange(false);
2042 bool adjustNorm(false);
2043
2044 std::vector<pair<double,double> > rangeLim;
2045
2046 // Retrieve plot range to be able to adjust normalization to data
2047 if (pc.hasProcessed("Range")) {
2048
2049 double rangeLo = pc.getDouble("rangeLo") ;
2050 double rangeHi = pc.getDouble("rangeHi") ;
2051 rangeLim.push_back(make_pair(rangeLo,rangeHi)) ;
2052 adjustNorm = pc.getInt("rangeAdjustNorm") ;
2054
2055 coutI(Plotting) << "RooAbsPdf::plotOn(" << GetName() << ") only plotting range ["
2056 << rangeLo << "," << rangeHi << "]" ;
2057 if (!pc.hasProcessed("NormRange")) {
2058 ccoutI(Plotting) << ", curve is normalized to data in " << (adjustNorm?"given":"full") << " range" << std::endl ;
2059 } else {
2060 ccoutI(Plotting) << std::endl ;
2061 }
2062
2063 nameSuffix << "_Range[" << rangeLo << "_" << rangeHi << "]";
2064
2065 } else if (pc.hasProcessed("RangeWithName")) {
2066
2067 for (const std::string& rangeNameToken : ROOT::Split(pc.getString("rangeName", "", false), ",")) {
2068 const char* thisRangeName = rangeNameToken.empty() ? nullptr : rangeNameToken.c_str();
2069 if (thisRangeName && !frame->getPlotVar()->hasRange(thisRangeName)) {
2070 coutE(Plotting) << "Range '" << rangeNameToken << "' not defined for variable '"
2071 << frame->getPlotVar()->GetName() << "'. Ignoring ..." << std::endl;
2072 continue;
2073 }
2074 rangeLim.push_back(frame->getPlotVar()->getRange(thisRangeName));
2075 }
2076 adjustNorm = pc.getInt("rangeWNAdjustNorm") ;
2078
2079 coutI(Plotting) << "RooAbsPdf::plotOn(" << GetName() << ") only plotting range '" << pc.getString("rangeName", "", false) << "'" ;
2080 if (!pc.hasProcessed("NormRange")) {
2081 ccoutI(Plotting) << ", curve is normalized to data in " << (adjustNorm?"given":"full") << " range" << std::endl ;
2082 } else {
2083 ccoutI(Plotting) << std::endl ;
2084 }
2085
2086 nameSuffix << "_Range[" << pc.getString("rangeName") << "]";
2087 }
2088 // Specification of a normalization range override those in a regular range
2089 if (pc.hasProcessed("NormRange")) {
2090 rangeLim.clear();
2091 for (const auto& rangeNameToken : ROOT::Split(pc.getString("normRangeName", "", false), ",")) {
2092 const char* thisRangeName = rangeNameToken.empty() ? nullptr : rangeNameToken.c_str();
2093 if (thisRangeName && !frame->getPlotVar()->hasRange(thisRangeName)) {
2094 coutE(Plotting) << "Range '" << rangeNameToken << "' not defined for variable '"
2095 << frame->getPlotVar()->GetName() << "'. Ignoring ..." << std::endl;
2096 continue;
2097 }
2098 rangeLim.push_back(frame->getPlotVar()->getRange(thisRangeName));
2099 }
2100 adjustNorm = true ;
2102 coutI(Plotting) << "RooAbsPdf::plotOn(" << GetName() << ") p.d.f. curve is normalized using explicit choice of ranges '" << pc.getString("normRangeName", "", false) << "'" << std::endl ;
2103
2104 nameSuffix << "_NormRange[" << pc.getString("rangeName") << "]";
2105 }
2106
2107 if (hasCustomRange && adjustNorm) {
2108 // If overlapping ranges were given, remove them now
2109 const std::size_t oldSize = rangeLim.size();
2111
2112 if (oldSize != rangeLim.size() && !pc.hasProcessed("NormRange")) {
2113 // User gave overlapping ranges. This leads to double-counting events and integrals, and must
2114 // therefore be avoided. If a NormRange has been given, the overlap is already gone.
2115 // It's safe to plot even with overlap now.
2116 coutE(Plotting) << "Requested plot/integration ranges overlap. For correct plotting, new ranges "
2117 "will be defined." << std::endl;
2118 auto plotVar = dynamic_cast<RooRealVar*>(frame->getPlotVar());
2119 assert(plotVar);
2120 std::string rangesNoOverlap;
2121 for (auto it = rangeLim.begin(); it != rangeLim.end(); ++it) {
2122 std::stringstream rangeName;
2123 rangeName << "Remove_overlap_range_" << it - rangeLim.begin();
2124 plotVar->setRange(rangeName.str().c_str(), it->first, it->second);
2125 if (!rangesNoOverlap.empty())
2126 rangesNoOverlap += ",";
2127 rangesNoOverlap += rangeName.str();
2128 }
2129
2130 auto rangeArg = static_cast<RooCmdArg*>(cmdList.FindObject("RangeWithName"));
2131 if (rangeArg) {
2132 rangeArg->setString(0, rangesNoOverlap.c_str());
2133 } else {
2134 plotRange = std::make_unique<RooCmdArg>(RooFit::Range(rangesNoOverlap.c_str()));
2135 cmdList.Add(plotRange.get());
2136 }
2137 }
2138
2139 double rangeNevt(0) ;
2140 for (const auto& riter : rangeLim) {
2141 double nevt= frame->getFitRangeNEvt(riter.first, riter.second);
2142 rangeNevt += nevt ;
2143 }
2144
2145 scaleFactor *= rangeNevt/nExpected ;
2146
2147 } else {
2148 // First, check if the PDF *can* be extended.
2149 if (this->canBeExtended()) {
2150 // If it can, get the expected events.
2151 const double nExp = expectedEvents(frame->getNormVars());
2152 if (nExp > 0) {
2153 // If the prediction is valid, use it for normalization.
2154 scaleFactor *= nExp / nExpected;
2155 } else {
2156 // If prediction is not valid (e.g. 0), fall back to data.
2157 scaleFactor *= frame->getFitRangeNEvt() / nExpected;
2158 }
2159 } else {
2160 // If the PDF can't be extended, just use the data.
2161 scaleFactor *= frame->getFitRangeNEvt() / nExpected;
2162 }
2163 }
2164 } else if (stype==RelativeExpected) {
2165 scaleFactor *= nExpected ;
2166 } else if (stype==NumEvent) {
2167 scaleFactor /= nExpected ;
2168 }
2169 scaleFactor *= frame->getFitRangeBinW() ;
2170 }
2171 frame->updateNormVars(*frame->getPlotVar()) ;
2172
2173 // Append overriding scale factor command at end of original command list
2174 RooCmdArg tmp = RooFit::Normalization(scaleFactor,Raw) ;
2175 tmp.setInt(1,1) ; // Flag this normalization command as created for internal use (so that VisualizeError can strip it)
2176 cmdList.Add(&tmp) ;
2177
2178 // Was a component selected requested
2179 if (haveCompSel) {
2180
2181 // Get complete set of tree branch nodes
2184
2185 // Discard any non-RooAbsReal nodes
2186 for (const auto arg : branchNodeSet) {
2187 if (!dynamic_cast<RooAbsReal*>(arg)) {
2188 branchNodeSet.remove(*arg) ;
2189 }
2190 }
2191
2192 // Obtain direct selection
2193 std::unique_ptr<RooArgSet> dirSelNodes;
2194 if (compSet) {
2195 dirSelNodes = std::unique_ptr<RooArgSet>{branchNodeSet.selectCommon(*compSet)};
2196 } else {
2197 dirSelNodes = std::unique_ptr<RooArgSet>{branchNodeSet.selectByName(compSpec)};
2198 }
2199 if (!dirSelNodes->empty()) {
2200 coutI(Plotting) << "RooAbsPdf::plotOn(" << GetName() << ") directly selected PDF components: " << *dirSelNodes << std::endl ;
2201
2202 // Do indirect selection and activate both
2204 } else {
2205 if (compSet) {
2206 coutE(Plotting) << "RooAbsPdf::plotOn(" << GetName() << ") ERROR: component selection set " << *compSet << " does not match any components of p.d.f." << std::endl ;
2207 } else {
2208 coutE(Plotting) << "RooAbsPdf::plotOn(" << GetName() << ") ERROR: component selection expression '" << compSpec << "' does not select any components of p.d.f." << std::endl ;
2209 }
2210 return nullptr ;
2211 }
2212 }
2213
2214 RooCmdArg cnsuffix("CurveNameSuffix", 0, 0, 0, 0, nameSuffix.str().c_str(), nullptr, nullptr, nullptr);
2215 cmdList.Add(&cnsuffix);
2216
2218
2219 // Restore selection status ;
2220 if (haveCompSel) plotOnCompSelect(nullptr) ;
2221
2222 return ret ;
2223}
2224
2225
2226//_____________________________________________________________________________
2227/// Plot oneself on 'frame'. In addition to features detailed in RooAbsReal::plotOn(),
2228/// the scale factor for a PDF can be interpreted in three different ways. The interpretation
2229/// is controlled by ScaleType
2230/// ```
2231/// Relative - Scale factor is applied on top of PDF normalization scale factor
2232/// NumEvent - Scale factor is interpreted as a number of events. The surface area
2233/// under the PDF curve will match that of a histogram containing the specified
2234/// number of event
2235/// Raw - Scale factor is applied to the raw (projected) probability density.
2236/// Not too useful, option provided for completeness.
2237/// ```
2238// coverity[PASS_BY_VALUE]
2240{
2241
2242 // Sanity checks
2243 if (plotSanityChecks(frame)) return frame ;
2244
2245 // More sanity checks
2246 double nExpected(1) ;
2247 if (o.stype==RelativeExpected) {
2248 if (!canBeExtended()) {
2249 coutE(Plotting) << "RooAbsPdf::plotOn(" << GetName()
2250 << "): ERROR the 'Expected' scale option can only be used on extendable PDFs" << std::endl ;
2251 return frame ;
2252 }
2253 frame->updateNormVars(*frame->getPlotVar()) ;
2255 }
2256
2257 // Adjust normalization, if so requested
2258 if (o.stype != Raw) {
2259
2260 if (frame->getFitRangeNEvt() && o.stype==Relative) {
2261 // If non-default plotting range is specified, adjust number of events in fit range
2262 o.scaleFactor *= frame->getFitRangeNEvt()/nExpected ;
2263 } else if (o.stype==RelativeExpected) {
2264 o.scaleFactor *= nExpected ;
2265 } else if (o.stype==NumEvent) {
2266 o.scaleFactor /= nExpected ;
2267 }
2268 o.scaleFactor *= frame->getFitRangeBinW() ;
2269 }
2270 frame->updateNormVars(*frame->getPlotVar()) ;
2271
2272 return RooAbsReal::plotOn(frame,o) ;
2273}
2274
2275
2276
2277
2278////////////////////////////////////////////////////////////////////////////////
2279/// The following named arguments are supported
2280/// <table>
2281/// <tr><th> Type of CmdArg <th> Effect on parameter box
2282/// <tr><td> `Parameters(const RooArgSet& param)` <td> Only the specified subset of parameters will be shown. By default all non-constant parameters are shown.
2283/// <tr><td> `ShowConstants(bool flag)` <td> Also display constant parameters
2284/// <tr><td> `Format(const char* what,...)` <td> Parameter formatting options.
2285/// | Parameter | Format
2286/// | ---------------------- | --------------------------
2287/// | `const char* what` | Controls what is shown. "N" adds name (alternatively, "T" adds the title), "E" adds error, "A" shows asymmetric error, "U" shows unit, "H" hides the value
2288/// | `FixedPrecision(int n)`| Controls precision, set fixed number of digits
2289/// | `AutoPrecision(int n)` | Controls precision. Number of shown digits is calculated from error + n specified additional digits (1 is sensible default)
2290/// <tr><td> `Label(const chat* label)` <td> Add label to parameter box. Use `\n` for multi-line labels.
2291/// <tr><td> `Layout(double xmin, double xmax, double ymax)` <td> Specify relative position of left/right side of box and top of box.
2292/// Coordinates are given as position on the pad between 0 and 1.
2293/// The lower end of the box is calculated automatically from the number of lines in the box.
2294/// </table>
2295///
2296///
2297/// Example use:
2298/// ```
2299/// pdf.paramOn(frame, Label("fit result"), Format("NEU",AutoPrecision(1)) ) ;
2300/// ```
2301///
2302
2304 const RooCmdArg& arg3, const RooCmdArg& arg4, const RooCmdArg& arg5,
2305 const RooCmdArg& arg6, const RooCmdArg& arg7, const RooCmdArg& arg8)
2306{
2307 // Stuff all arguments in a list
2309 cmdList.Add(const_cast<RooCmdArg*>(&arg1)) ; cmdList.Add(const_cast<RooCmdArg*>(&arg2)) ;
2310 cmdList.Add(const_cast<RooCmdArg*>(&arg3)) ; cmdList.Add(const_cast<RooCmdArg*>(&arg4)) ;
2311 cmdList.Add(const_cast<RooCmdArg*>(&arg5)) ; cmdList.Add(const_cast<RooCmdArg*>(&arg6)) ;
2312 cmdList.Add(const_cast<RooCmdArg*>(&arg7)) ; cmdList.Add(const_cast<RooCmdArg*>(&arg8)) ;
2313
2314 // Select the pdf-specific commands
2315 RooCmdConfig pc("RooAbsPdf::paramOn(" + std::string(GetName()) + ")");
2316 pc.defineString("label","Label",0,"") ;
2317 pc.defineDouble("xmin","Layout",0,0.65) ;
2318 pc.defineDouble("xmax","Layout",1,0.9) ;
2319 pc.defineInt("ymaxi","Layout",0,Int_t(0.9*10000)) ;
2320 pc.defineInt("showc","ShowConstants",0,0) ;
2321 pc.defineSet("params","Parameters",0,nullptr) ;
2322 pc.defineInt("dummy","FormatArgs",0,0) ;
2323
2324 // Process and check varargs
2325 pc.process(cmdList) ;
2326 if (!pc.ok(true)) {
2327 return frame ;
2328 }
2329
2330 auto formatCmd = static_cast<RooCmdArg const*>(cmdList.FindObject("FormatArgs")) ;
2331
2332 const char* label = pc.getString("label") ;
2333 double xmin = pc.getDouble("xmin") ;
2334 double xmax = pc.getDouble("xmax") ;
2335 double ymax = pc.getInt("ymaxi") / 10000. ;
2336 int showc = pc.getInt("showc") ;
2337
2338 // Decode command line arguments
2339 std::unique_ptr<RooArgSet> params{getParameters(frame->getNormVars())} ;
2340 if(RooArgSet* requestedParams = pc.getSet("params")) {
2341 params = std::unique_ptr<RooArgSet>{params->selectCommon(*requestedParams)};
2342 }
2343 paramOn(frame,*params,showc,label,xmin,xmax,ymax,formatCmd);
2344
2345 return frame ;
2346}
2347
2348
2349////////////////////////////////////////////////////////////////////////////////
2350/// Add a text box with the current parameter values and their errors to the frame.
2351/// Observables of this PDF appearing in the 'data' dataset will be omitted.
2352///
2353/// An optional label will be inserted if passed. Multi-line labels can be generated
2354/// by adding `\n` to the label string. Use 'sigDigits'
2355/// to modify the default number of significant digits printed. The 'xmin,xmax,ymax'
2356/// values specify the initial relative position of the text box in the plot frame.
2357
2358RooPlot* RooAbsPdf::paramOn(RooPlot* frame, const RooArgSet& params, bool showConstants, const char *label,
2359 double xmin, double xmax ,double ymax, const RooCmdArg* formatCmd)
2360{
2361
2362 // parse the options
2363 bool showLabel= (label != nullptr && strlen(label) > 0);
2364
2365 // calculate the box's size, adjusting for constant parameters
2366
2367 double ymin(ymax);
2368 double dy(0.06);
2369 for (const auto param : params) {
2370 auto var = static_cast<RooRealVar*>(param);
2371 if(showConstants || !var->isConstant()) ymin-= dy;
2372 }
2373
2374 std::string labelString = label;
2375 unsigned int numLines = std::count(labelString.begin(), labelString.end(), '\n') + 1;
2376 if (showLabel) ymin -= numLines * dy;
2377
2378 // create the box and set its options
2379 TPaveText *box= new TPaveText(xmin,ymax,xmax,ymin,"BRNDC");
2380 if(!box) return nullptr;
2381 box->SetName((std::string(GetName()) + "_paramBox").c_str());
2382 box->SetFillColor(0);
2383 box->SetBorderSize(0);
2384 box->SetTextAlign(12);
2385 box->SetTextSize(0.04F);
2386 box->SetFillStyle(0);
2387
2388 for (const auto param : params) {
2389 auto var = static_cast<const RooRealVar*>(param);
2390 if(var->isConstant() && !showConstants) continue;
2391
2392 box->AddText((formatCmd ? var->format(*formatCmd) : var->format(2, "NELU")).c_str());
2393 }
2394
2395 // add the optional label if specified
2396 if (showLabel) {
2397 for (const auto& line : ROOT::Split(label, "\n")) {
2398 box->AddText(line.c_str());
2399 }
2400 }
2401
2402 // Add box to frame
2403 frame->addObject(box) ;
2404
2405 return frame ;
2406}
2407
2408
2409
2410
2411////////////////////////////////////////////////////////////////////////////////
2412/// Return expected number of events from this p.d.f for use in extended
2413/// likelihood calculations. This default implementation returns zero
2414
2416{
2417 return 0 ;
2418}
2419
2420
2421
2422////////////////////////////////////////////////////////////////////////////////
2423/// Change global level of verbosity for p.d.f. evaluations
2424
2426{
2427 _verboseEval = stat ;
2428}
2429
2430
2431
2432////////////////////////////////////////////////////////////////////////////////
2433/// Return global level of verbosity for p.d.f. evaluations
2434
2436{
2437 return _verboseEval ;
2438}
2439
2440
2441
2442////////////////////////////////////////////////////////////////////////////////
2443/// Destructor of normalization cache element. If this element
2444/// provides the 'current' normalization stored in RooAbsPdf::_norm
2445/// zero _norm pointer here before object pointed to is deleted here
2446
2448{
2449 // Zero _norm pointer in RooAbsPdf if it is points to our cache payload
2450 if (_owner) {
2451 RooAbsPdf* pdfOwner = static_cast<RooAbsPdf*>(_owner) ;
2452 if (pdfOwner->_norm == _norm.get()) {
2453 pdfOwner->_norm = nullptr ;
2454 }
2455 }
2456}
2457
2458
2459
2460////////////////////////////////////////////////////////////////////////////////
2461/// Return a p.d.f that represent a projection of this p.d.f integrated over given observables
2462
2464{
2465 // Construct name for new object
2466 std::string name = std::string{GetName()} + "_Proj[" + RooHelpers::getColonSeparatedNameString(iset, ',') + "]";
2467
2468 // Return projected p.d.f.
2469 return new RooProjectedPdf(name.c_str(),name.c_str(),*this,iset) ;
2470}
2471
2472
2473
2474////////////////////////////////////////////////////////////////////////////////
2475/// Create a cumulative distribution function of this p.d.f in terms
2476/// of the observables listed in iset. If no nset argument is given
2477/// the c.d.f normalization is constructed over the integrated
2478/// observables, so that its maximum value is precisely 1. It is also
2479/// possible to choose a different normalization for
2480/// multi-dimensional p.d.f.s: eg. for a pdf f(x,y,z) one can
2481/// construct a partial cdf c(x,y) that only when integrated itself
2482/// over z results in a maximum value of 1. To construct such a cdf pass
2483/// z as argument to the optional nset argument
2484
2489
2490
2491
2492////////////////////////////////////////////////////////////////////////////////
2493/// Create an object that represents the integral of the function over one or more observables listed in `iset`.
2494/// The actual integration calculation is only performed when the return object is evaluated. The name
2495/// of the integral object is automatically constructed from the name of the input function, the variables
2496/// it integrates and the range integrates over
2497///
2498/// The following named arguments are accepted
2499/// | Type of CmdArg | Effect on CDF
2500/// | ---------------------|-------------------
2501/// | SupNormSet(const RooArgSet&) | Observables over which should be normalized _in addition_ to the integration observables
2502/// | ScanNumCdf() | Apply scanning technique if cdf integral involves numeric integration [ default ]
2503/// | ScanAllCdf() | Always apply scanning technique
2504/// | ScanNoCdf() | Never apply scanning technique
2505/// | ScanParameters(Int_t nbins, Int_t intOrder) | Parameters for scanning technique of making CDF: number of sampled bins and order of interpolation applied on numeric cdf
2506
2508 const RooCmdArg& arg3, const RooCmdArg& arg4, const RooCmdArg& arg5,
2509 const RooCmdArg& arg6, const RooCmdArg& arg7, const RooCmdArg& arg8)
2510{
2511 // Define configuration for this method
2512 RooCmdConfig pc("RooAbsReal::createCdf(" + std::string(GetName()) + ")");
2513 pc.defineSet("supNormSet","SupNormSet",0,nullptr) ;
2514 pc.defineInt("numScanBins","ScanParameters",0,1000) ;
2515 pc.defineInt("intOrder","ScanParameters",1,2) ;
2516 pc.defineInt("doScanNum","ScanNumCdf",0,1) ;
2517 pc.defineInt("doScanAll","ScanAllCdf",0,0) ;
2518 pc.defineInt("doScanNon","ScanNoCdf",0,0) ;
2519 pc.defineMutex("ScanNumCdf","ScanAllCdf","ScanNoCdf") ;
2520
2521 // Process & check varargs
2523 if (!pc.ok(true)) {
2524 return nullptr ;
2525 }
2526
2527 // Extract values from named arguments
2528 const RooArgSet* snset = pc.getSet("supNormSet",nullptr);
2529 RooArgSet nset ;
2530 if (snset) {
2531 nset.add(*snset) ;
2532 }
2533 Int_t numScanBins = pc.getInt("numScanBins") ;
2534 Int_t intOrder = pc.getInt("intOrder") ;
2535 Int_t doScanNum = pc.getInt("doScanNum") ;
2536 Int_t doScanAll = pc.getInt("doScanAll") ;
2537 Int_t doScanNon = pc.getInt("doScanNon") ;
2538
2539 // If scanning technique is not requested make integral-based cdf and return
2540 if (doScanNon) {
2541 return createIntRI(iset,nset) ;
2542 }
2543 if (doScanAll) {
2544 return createScanCdf(iset,nset,numScanBins,intOrder) ;
2545 }
2546 if (doScanNum) {
2547 std::unique_ptr<RooAbsReal> tmp{createIntegral(iset)} ;
2548 Int_t isNum= !static_cast<RooRealIntegral&>(*tmp).numIntRealVars().empty();
2549
2550 if (isNum) {
2551 coutI(NumericIntegration) << "RooAbsPdf::createCdf(" << GetName() << ") integration over observable(s) " << iset << " involves numeric integration," << std::endl
2552 << " constructing cdf though numeric integration of sampled pdf in " << numScanBins << " bins and applying order "
2553 << intOrder << " interpolation on integrated histogram." << std::endl
2554 << " To override this choice of technique use argument ScanNone(), to change scan parameters use ScanParameters(nbins,order) argument" << std::endl ;
2555 }
2556
2558 }
2559 return nullptr ;
2560}
2561
2563{
2564 string name = string(GetName()) + "_NUMCDF_" + integralNameSuffix(iset,&nset).Data() ;
2565 RooRealVar* ivar = static_cast<RooRealVar*>(iset.first()) ;
2566 ivar->setBins(numScanBins,"numcdf") ;
2567 auto ret = std::make_unique<RooNumCdf>(name.c_str(),name.c_str(),*this,*ivar,"numcdf");
2568 ret->setInterpolationOrder(intOrder) ;
2569 return RooFit::makeOwningPtr<RooAbsReal>(std::move(ret));
2570}
2571
2572
2573
2574
2575////////////////////////////////////////////////////////////////////////////////
2576/// This helper function finds and collects all constraints terms of all component p.d.f.s
2577/// and returns a RooArgSet with all those terms.
2578
2579std::unique_ptr<RooArgSet>
2581{
2582 RooArgSet constraints;
2584
2585 std::unique_ptr<RooArgSet> comps(getComponents());
2586 for (const auto arg : *comps) {
2587 auto pdf = dynamic_cast<const RooAbsPdf*>(arg) ;
2588 if (pdf && !constraints.find(pdf->GetName())) {
2589 if (auto compRet = pdf->getConstraints(observables,constrainedParams, pdfParams)) {
2590 constraints.add(*compRet,false) ;
2591 }
2592 }
2593 }
2594
2596
2597 // Strip any constraints that are completely decoupled from the other product terms
2598 auto finalConstraints = std::make_unique<RooArgSet>("AllConstraints");
2599 for(auto * pdf : static_range_cast<RooAbsPdf*>(constraints)) {
2600
2601 RooArgSet tmp;
2602 pdf->getParameters(nullptr, tmp);
2603 conParams.add(tmp,true) ;
2604
2605 if (pdf->dependsOnValue(pdfParams) || !stripDisconnected) {
2606 finalConstraints->add(*pdf) ;
2607 } else {
2608 coutI(Minimization) << "RooAbsPdf::getAllConstraints(" << GetName() << ") omitting term " << pdf->GetName()
2609 << " as constraint term as it does not share any parameters with the other pdfs in product. "
2610 << "To force inclusion in likelihood, add an explicit Constrain() argument for the target parameter" << std::endl ;
2611 }
2612 }
2613
2614 // Now remove from constrainedParams all parameters that occur exclusively in constraint term and not in regular pdf term
2615
2617 conParams.selectCommon(constrainedParams, cexl);
2618 cexl.remove(pdfParams,true,true) ;
2619 constrainedParams.remove(cexl,true,true) ;
2620
2621 return finalConstraints ;
2622}
2623
2624
2625////////////////////////////////////////////////////////////////////////////////
2626/// Returns the default numeric MC generator configuration for all RooAbsReals
2627
2632
2633
2634////////////////////////////////////////////////////////////////////////////////
2635/// Returns the specialized integrator configuration for _this_ RooAbsReal.
2636/// If this object has no specialized configuration, a null pointer is returned
2637
2642
2643
2644
2645////////////////////////////////////////////////////////////////////////////////
2646/// Returns the specialized integrator configuration for _this_ RooAbsReal.
2647/// If this object has no specialized configuration, a null pointer is returned,
2648/// unless createOnTheFly is true in which case a clone of the default integrator
2649/// configuration is created, installed as specialized configuration, and returned
2650
2652{
2654 _specGeneratorConfig = std::make_unique<RooNumGenConfig>(*defaultGeneratorConfig()) ;
2655 }
2656 return _specGeneratorConfig.get();
2657}
2658
2659
2660
2661////////////////////////////////////////////////////////////////////////////////
2662/// Return the numeric MC generator configuration used for this object. If
2663/// a specialized configuration was associated with this object, that configuration
2664/// is returned, otherwise the default configuration for all RooAbsReals is returned
2665
2667{
2668 const RooNumGenConfig *config = specialGeneratorConfig();
2669 return config ? config : defaultGeneratorConfig();
2670}
2671
2672
2673
2674////////////////////////////////////////////////////////////////////////////////
2675/// Set the given configuration as default numeric MC generator
2676/// configuration for this object
2677
2679{
2680 _specGeneratorConfig = std::make_unique<RooNumGenConfig>(config);
2681}
2682
2683
2684
2685////////////////////////////////////////////////////////////////////////////////
2686/// Remove the specialized numeric MC generator configuration associated
2687/// with this object
2688
2693
2695
2696
2697////////////////////////////////////////////////////////////////////////////////
2698
2700 bool extended, bool randProto, bool resampleProto, TString dsetName, bool init) :
2701 _genContext(context), _whatVars(whatVars), _protoData(protoData), _nGen(nGen), _extended(extended),
2702 _randProto(randProto), _resampleProto(resampleProto), _dsetName(dsetName), _init(init)
2703{
2704}
2705
2706
2707namespace {
2708
2709void sterilizeClientCaches(RooAbsArg & arg) {
2710 auto const& clients = arg.clients();
2711 for(std::size_t iClient = 0; iClient < clients.size(); ++iClient) {
2712
2713 const std::size_t oldClientsSize = clients.size();
2714 RooAbsArg* client = clients[iClient];
2715
2716 for(int iCache = 0; iCache < client->numCaches(); ++iCache) {
2717 if(auto cacheMgr = dynamic_cast<RooObjCacheManager*>(client->getCache(iCache))) {
2718 cacheMgr->sterilize();
2719 }
2720 }
2721
2722 // It can happen that the objects cached by the client are also clients of
2723 // the arg itself! In that case, the position of the client in the client
2724 // list might have changed, and we need to find the new index.
2725 if(clients.size() != oldClientsSize) {
2726 auto clientIter = std::find(clients.begin(), clients.end(), client);
2727 if(clientIter == clients.end()) {
2728 throw std::runtime_error("After a clients caches were cleared, the client was gone! This should not happen.");
2729 }
2730 iClient = std::distance(clients.begin(), clientIter);
2731 }
2732 }
2733}
2734
2735} // namespace
2736
2737
2738////////////////////////////////////////////////////////////////////////////////
2739
2741{
2743
2744 // the stuff that the clients have cached may depend on the normalization range
2745 sterilizeClientCaches(*this);
2746
2747 if (_norm) {
2749 _norm = nullptr ;
2750 }
2751}
2752
2753
2754////////////////////////////////////////////////////////////////////////////////
2755
2757{
2759
2760 // the stuff that the clients have cached may depend on the normalization range
2761 sterilizeClientCaches(*this);
2762
2763 if (_norm) {
2765 _norm = nullptr ;
2766 }
2767}
2768
2769
2770////////////////////////////////////////////////////////////////////////////////
2771/// Hook function intercepting redirectServer calls. Discard current
2772/// normalization object if any server is redirected
2773
2775 bool nameChange, bool isRecursiveStep)
2776{
2777 // If servers are redirected, the cached normalization integrals and
2778 // normalization sets are most likely invalid.
2780
2781 // Object is own by _normCacheManager that will delete object as soon as cache
2782 // is sterilized by server redirect
2783 _norm = nullptr ;
2784
2785 // Similar to the situation with the normalization integral above: if a
2786 // server is redirected, the cached normalization set might not point to
2787 // the right observables anymore. We need to reset it.
2788 setActiveNormSet(nullptr);
2790}
2791
2792
2793std::unique_ptr<RooAbsArg>
2795{
2796 if (normSet.empty() || selfNormalized()) {
2798 }
2799 std::unique_ptr<RooAbsPdf> pdfClone(static_cast<RooAbsPdf *>(this->Clone()));
2801
2802 auto newArg = std::make_unique<RooFit::Detail::RooNormalizedPdf>(*pdfClone, normSet);
2803
2804 // The direct servers are this pdf and the normalization integral, which
2805 // don't need to be compiled further.
2806 for (RooAbsArg *server : newArg->servers()) {
2807 ctx.markAsCompiled(*server);
2808 }
2809 ctx.markAsCompiled(*newArg);
2810 newArg->addOwnedComponents(std::move(pdfClone));
2811 return newArg;
2812}
2813
2814/// Returns an object that represents the expected number of events for a given
2815/// normalization set, similar to how createIntegral() returns an object that
2816/// returns the integral. This is used to build the computation graph for the
2817/// final likelihood.
2818std::unique_ptr<RooAbsReal> RooAbsPdf::createExpectedEventsFunc(const RooArgSet * /*nset*/) const
2819{
2820 std::stringstream errMsg;
2821 errMsg << "The pdf \"" << GetName() << "\" of type " << ClassName()
2822 << " did not overload RooAbsPdf::createExpectedEventsFunc()!";
2823 coutE(InputArguments) << errMsg.str() << std::endl;
2824 return nullptr;
2825}
#define e(i)
Definition RSha256.hxx:103
#define coutI(a)
#define cxcoutI(a)
#define cxcoutD(a)
#define coutP(a)
#define oocoutW(o, a)
#define coutW(a)
#define coutE(a)
#define ccoutI(a)
#define ccoutD(a)
int Int_t
Signed integer 4 bytes (int)
Definition RtypesCore.h:59
static void indent(ostringstream &buf, int indent_level)
ROOT::Detail::TRangeCast< T, true > TRangeDynCast
TRangeDynCast is an adapter class that allows the typed iteration through a TCollection.
Option_t Option_t TPoint TPoint const char GetTextMagnitude GetFillStyle GetLineColor GetLineWidth GetMarkerStyle GetTextAlign GetTextColor GetTextSize void data
Option_t Option_t TPoint TPoint const char GetTextMagnitude GetFillStyle GetLineColor GetLineWidth GetMarkerStyle GetTextAlign GetTextColor GetTextSize void input
Option_t Option_t TPoint TPoint const char GetTextMagnitude GetFillStyle GetLineColor GetLineWidth GetMarkerStyle GetTextAlign GetTextColor GetTextSize void value
char name[80]
Definition TGX11.cxx:110
float xmin
float ymin
float xmax
float ymax
char * Form(const char *fmt,...)
Formats a string in a circular formatting buffer.
Definition TString.cxx:2496
const_iterator begin() const
const_iterator end() const
Common abstract base class for objects that represent a value and a "shape" in RooFit.
Definition RooAbsArg.h:76
void clearValueAndShapeDirty() const
Definition RooAbsArg.h:535
void Print(Option_t *options=nullptr) const override
Print the object to the defaultPrintStream().
Definition RooAbsArg.h:238
bool dependsOn(const RooAbsCollection &serverList, const RooAbsArg *ignoreArg=nullptr, bool valueOnly=false) const
Test whether we depend on (ie, are served by) any object in the specified collection.
RooFit::OwningPtr< RooArgSet > getParameters(const RooAbsData *data, bool stripDisconnected=true) const
Create a list of leaf nodes in the arg tree starting with ourself as top node that don't match any of...
RooFit::OwningPtr< RooArgSet > getObservables(const RooArgSet &set, bool valueOnly=true) const
Given a set of possible observables, return the observables that this PDF depends on.
const Text_t * getStringAttribute(const Text_t *key) const
Get string attribute mapped under key 'key'.
virtual std::unique_ptr< RooAbsArg > compileForNormSet(RooArgSet const &normSet, RooFit::Detail::CompileContext &ctx) const
RooFit::OwningPtr< RooArgSet > getComponents() const
Create a RooArgSet with all components (branch nodes) of the expression tree headed by this object.
bool getAttribute(const Text_t *name) const
Check if a named attribute is set. By default, all attributes are unset.
RooAbsCache * getCache(Int_t index) const
Return registered cache object by index.
const RefCountList_t & clients() const
List of all clients of this object.
Definition RooAbsArg.h:137
bool isValueDirty() const
Definition RooAbsArg.h:356
void setProxyNormSet(const RooArgSet *nset)
Forward a change in the cached normalization argset to all the registered proxies.
void branchNodeServerList(RooAbsCollection *list, const RooAbsArg *arg=nullptr, bool recurseNonDerived=false) const
Fill supplied list with all branch nodes of the arg tree starting with ourself as top node.
TObject * Clone(const char *newname=nullptr) const override
Make a clone of an object using the Streamer facility.
Definition RooAbsArg.h:88
RefCountList_t _serverList
Definition RooAbsArg.h:564
Int_t numCaches() const
Return number of registered caches.
RooAbsArg * findServer(const char *name) const
Return server of this with name name. Returns nullptr if not found.
Definition RooAbsArg.h:147
RooAbsArg * _owner
! Pointer to owning RooAbsArg
Abstract base class for objects that represent a discrete value that can be set from the outside,...
Abstract container object that can hold multiple RooAbsArg objects.
virtual bool add(const RooAbsArg &var, bool silent=false)
Add the specified argument to list.
RooAbsArg * find(const char *name) const
Find object with given name in list.
void Print(Option_t *options=nullptr) const override
This method must be overridden when a class wants to print itself.
Abstract base class for binned and unbinned datasets.
Definition RooAbsData.h:57
Abstract base class for generator contexts of RooAbsPdf objects.
virtual RooDataSet * generate(double nEvents=0, bool skipInit=false, bool extendedMode=false)
Generate the specified number of events with nEvents>0 and and return a dataset containing the genera...
bool isValid() const
virtual void setProtoDataOrder(Int_t *lut)
Set the traversal order of prototype data to that in the lookup tables passed as argument.
Normalization set with for above integral.
Definition RooAbsPdf.h:318
std::unique_ptr< RooAbsReal > _norm
Definition RooAbsPdf.h:323
~CacheElem() override
Destructor of normalization cache element.
Abstract interface for all probability density functions.
Definition RooAbsPdf.h:32
virtual bool syncNormalization(const RooArgSet *dset, bool adjustProxies=true) const
Verify that the normalization integral cached with this PDF is valid for given set of normalization o...
double getNorm(const RooArgSet &nset) const
Get normalisation term needed to normalise the raw values returned by getVal().
Definition RooAbsPdf.h:191
std::unique_ptr< RooAbsArg > compileForNormSet(RooArgSet const &normSet, RooFit::Detail::CompileContext &ctx) const override
RooObjCacheManager _normMgr
Definition RooAbsPdf.h:325
std::unique_ptr< RooNumGenConfig > _specGeneratorConfig
! MC generator configuration specific for this object
Definition RooAbsPdf.h:336
double getValV(const RooArgSet *set=nullptr) const override
Return current value, normalized by integrating over the observables in nset.
virtual std::unique_ptr< RooFitResult > fitToImpl(RooAbsData &data, const RooLinkedList &cmdList)
Protected implementation of the likelihood fitting routine.
virtual void generateEvent(Int_t code)
Interface for generation of an event using the algorithm corresponding to the specified code.
RooFit::OwningPtr< RooAbsReal > createScanCdf(const RooArgSet &iset, const RooArgSet &nset, Int_t numScanBins, Int_t intOrder)
void setGeneratorConfig()
Remove the specialized numeric MC generator configuration associated with this object.
virtual void resetErrorCounters(Int_t resetValue=10)
Reset error counter to given value, limiting the number of future error messages for this pdf to 'res...
std::unique_ptr< RooArgSet > getAllConstraints(const RooArgSet &observables, RooArgSet &constrainedParams, bool stripDisconnected=true) const
This helper function finds and collects all constraints terms of all component p.d....
static int verboseEval()
Return global level of verbosity for p.d.f. evaluations.
RooFit::OwningPtr< RooAbsReal > createCdf(const RooArgSet &iset, const RooArgSet &nset=RooArgSet())
Create a cumulative distribution function of this p.d.f in terms of the observables listed in iset.
bool isActiveNormSet(RooArgSet const *normSet) const
Checks if normSet is the currently active normalization set of this PDF, meaning is exactly the same ...
Definition RooAbsPdf.h:295
virtual double expectedEvents(const RooArgSet *nset) const
Return expected number of events to be used in calculation of extended likelihood.
virtual RooAbsGenContext * binnedGenContext(const RooArgSet &vars, bool verbose=false) const
Return a binned generator context.
TString _normRange
Normalization range.
Definition RooAbsPdf.h:338
virtual bool isDirectGenSafe(const RooAbsArg &arg) const
Check if given observable can be safely generated using the pdfs internal generator mechanism (if tha...
Int_t * randomizeProtoOrder(Int_t nProto, Int_t nGen, bool resample=false) const
Return lookup table with randomized order for nProto prototype events.
void setNormRange(const char *rangeName)
~RooAbsPdf() override
Destructor.
RooArgSet const * _normSet
Normalization integral (owned by _normMgr)
Definition RooAbsPdf.h:316
RooPlot * plotOn(RooPlot *frame, const RooCmdArg &arg1={}, const RooCmdArg &arg2={}, const RooCmdArg &arg3={}, const RooCmdArg &arg4={}, const RooCmdArg &arg5={}, const RooCmdArg &arg6={}, const RooCmdArg &arg7={}, const RooCmdArg &arg8={}, const RooCmdArg &arg9={}, const RooCmdArg &arg10={}) const override
Helper calling plotOn(RooPlot*, RooLinkedList&) const.
Definition RooAbsPdf.h:116
RooNumGenConfig * specialGeneratorConfig() const
Returns the specialized integrator configuration for this RooAbsReal.
virtual bool selfNormalized() const
Shows if a PDF is self-normalized, which means that no attempt is made to add a normalization term.
Definition RooAbsPdf.h:203
void printMultiline(std::ostream &os, Int_t contents, bool verbose=false, TString indent="") const override
Print multi line detailed information of this RooAbsPdf.
virtual double getCorrection() const
This function returns the penalty term.
Int_t _traceCount
Number of traces remaining to print.
Definition RooAbsPdf.h:331
bool canBeExtended() const
If true, PDF can provide extended likelihood term.
Definition RooAbsPdf.h:214
RooAbsReal * _norm
Definition RooAbsPdf.h:315
GenSpec * prepareMultiGen(const RooArgSet &whatVars, const RooCmdArg &arg1={}, const RooCmdArg &arg2={}, const RooCmdArg &arg3={}, const RooCmdArg &arg4={}, const RooCmdArg &arg5={}, const RooCmdArg &arg6={})
Prepare GenSpec configuration object for efficient generation of multiple datasets from identical spe...
void setTraceCounter(Int_t value, bool allNodes=false)
Reset trace counter to given value, limiting the number of future trace messages for this pdf to 'val...
Int_t _errorCount
Number of errors remaining to print.
Definition RooAbsPdf.h:330
@ CanNotBeExtended
Definition RooAbsPdf.h:208
virtual std::unique_ptr< RooAbsReal > createExpectedEventsFunc(const RooArgSet *nset) const
Returns an object that represents the expected number of events for a given normalization set,...
Int_t _negCount
Number of negative probabilities remaining to print.
Definition RooAbsPdf.h:332
virtual RooPlot * paramOn(RooPlot *frame, const RooCmdArg &arg1={}, const RooCmdArg &arg2={}, const RooCmdArg &arg3={}, const RooCmdArg &arg4={}, const RooCmdArg &arg5={}, const RooCmdArg &arg6={}, const RooCmdArg &arg7={}, const RooCmdArg &arg8={})
Add a box with parameter values (and errors) to the specified frame.
RooFit::OwningPtr< RooDataSet > generate(const RooArgSet &whatVars, Int_t nEvents, const RooCmdArg &arg1, const RooCmdArg &arg2={}, const RooCmdArg &arg3={}, const RooCmdArg &arg4={}, const RooCmdArg &arg5={})
See RooAbsPdf::generate(const RooArgSet&,const RooCmdArg&,const RooCmdArg&,const RooCmdArg&,...
Definition RooAbsPdf.h:49
virtual const RooAbsReal * getNormObj(const RooArgSet *set, const RooArgSet *iset, const TNamed *rangeName=nullptr) const
Return pointer to RooAbsReal object that implements calculation of integral over observables iset in ...
void setActiveNormSet(RooArgSet const *normSet) const
Setter for the _normSet member, which should never be set directly.
Definition RooAbsPdf.h:280
double analyticalIntegralWN(Int_t code, const RooArgSet *normSet, const char *rangeName=nullptr) const override
Analytical integral with normalization (see RooAbsReal::analyticalIntegralWN() for further informatio...
void setNormRangeOverride(const char *rangeName)
virtual RooFit::OwningPtr< RooDataSet > generateSimGlobal(const RooArgSet &whatVars, Int_t nEvents)
Special generator interface for generation of 'global observables' – for RooStats tools.
double normalizeWithNaNPacking(double rawVal, double normVal) const
virtual RooAbsGenContext * autoGenContext(const RooArgSet &vars, const RooDataSet *prototype=nullptr, const RooArgSet *auxProto=nullptr, bool verbose=false, bool autoBinned=true, const char *binnedTag="") const
const RooNumGenConfig * getGeneratorConfig() const
Return the numeric MC generator configuration used for this object.
virtual void initGenerator(Int_t code)
Interface for one-time initialization to setup the generator for the specified code.
virtual ExtendMode extendMode() const
Returns ability of PDF to provide extended likelihood terms.
Definition RooAbsPdf.h:212
RooAbsPdf()
Default constructor.
virtual RooFit::OwningPtr< RooDataHist > generateBinned(const RooArgSet &whatVars, double nEvents, const RooCmdArg &arg1, const RooCmdArg &arg2={}, const RooCmdArg &arg3={}, const RooCmdArg &arg4={}, const RooCmdArg &arg5={}) const
As RooAbsPdf::generateBinned(const RooArgSet&, const RooCmdArg&,const RooCmdArg&, const RooCmdArg&,...
Definition RooAbsPdf.h:102
bool traceEvalPdf(double value) const
Check that passed value is positive and not 'not-a-number'.
static RooNumGenConfig * defaultGeneratorConfig()
Returns the default numeric MC generator configuration for all RooAbsReals.
bool redirectServersHook(const RooAbsCollection &newServerList, bool mustReplaceAll, bool nameChange, bool isRecursiveStep) override
The cache manager.
void printValue(std::ostream &os) const override
Print value of p.d.f, also print normalization integral that was last used, if any.
virtual std::unique_ptr< RooAbsReal > createNLLImpl(RooAbsData &data, const RooLinkedList &cmdList)
Protected implementation of the NLL creation routine.
void logBatchComputationErrors(std::span< const double > &outputs, std::size_t begin) const
Scan through outputs and fix+log all nans and negative values.
virtual RooAbsGenContext * genContext(const RooArgSet &vars, const RooDataSet *prototype=nullptr, const RooArgSet *auxProto=nullptr, bool verbose=false) const
Interface function to create a generator context from a p.d.f.
void getLogProbabilities(std::span< const double > pdfValues, double *output) const
static TString _normRangeOverride
Definition RooAbsPdf.h:339
static Int_t _verboseEval
Definition RooAbsPdf.h:310
double extendedTerm(double sumEntries, double expected, double sumEntriesW2=0.0, bool doOffset=false) const
virtual Int_t getGenerator(const RooArgSet &directVars, RooArgSet &generateVars, bool staticInitOK=true) const
Load generatedVars with the subset of directVars that we can generate events for, and return a code t...
virtual RooAbsPdf * createProjection(const RooArgSet &iset)
Return a p.d.f that represent a projection of this p.d.f integrated over given observables.
virtual double getLogVal(const RooArgSet *set=nullptr) const
Return the log of the current value with given normalization An error message is printed if the argum...
bool hasRange(const char *name) const override
Check if variable has a binning with given name.
std::pair< double, double > getRange(const char *name=nullptr) const
Get low and high bound of the variable.
Abstract base class for objects that represent a real value and implements functionality common to al...
Definition RooAbsReal.h:63
RooDataHist * fillDataHist(RooDataHist *hist, const RooArgSet *nset, double scaleFactor, bool correctForBinVolume=false, bool showProgress=false) const
Fill a RooDataHist with values sampled from this function at the bin centers.
void plotOnCompSelect(RooArgSet *selNodes) const
Helper function for plotting of composite p.d.fs.
double getVal(const RooArgSet *normalisationSet=nullptr) const
Evaluate object.
Definition RooAbsReal.h:107
bool plotSanityChecks(RooPlot *frame) const
Utility function for plotOn(), perform general sanity check on frame to ensure safe plotting operatio...
void printMultiline(std::ostream &os, Int_t contents, bool verbose=false, TString indent="") const override
Structure printing.
bool redirectServersHook(const RooAbsCollection &newServerList, bool mustReplaceAll, bool nameChange, bool isRecursiveStep) override
Function that is called at the end of redirectServers().
double _value
Cache for current value of object.
Definition RooAbsReal.h:539
virtual double analyticalIntegral(Int_t code, const char *rangeName=nullptr) const
Implements the actual analytical integral(s) advertised by getAnalyticalIntegral.
TString integralNameSuffix(const RooArgSet &iset, const RooArgSet *nset=nullptr, const char *rangeName=nullptr, bool omitEmpty=false) const
Construct string with unique suffix name to give to integral object that encodes integrated observabl...
virtual double evaluate() const =0
Evaluate this PDF / function / constant. Needs to be overridden by all derived classes.
void logEvalError(const char *message, const char *serverValueString=nullptr) const
Log evaluation error message.
const RooNumIntConfig * getIntegratorConfig() const
Return the numeric integration configuration used for this object.
virtual bool isBinnedDistribution(const RooArgSet &) const
Tests if the distribution is binned. Unless overridden by derived classes, this always returns false.
Definition RooAbsReal.h:352
RooFit::OwningPtr< RooAbsReal > createIntRI(const RooArgSet &iset, const RooArgSet &nset={})
Utility function for createRunningIntegral.
virtual RooPlot * plotOn(RooPlot *frame, const RooCmdArg &arg1={}, const RooCmdArg &arg2={}, const RooCmdArg &arg3={}, const RooCmdArg &arg4={}, const RooCmdArg &arg5={}, const RooCmdArg &arg6={}, const RooCmdArg &arg7={}, const RooCmdArg &arg8={}, const RooCmdArg &arg9={}, const RooCmdArg &arg10={}) const
Plot (project) PDF on specified frame.
RooFit::OwningPtr< RooAbsReal > createIntegral(const RooArgSet &iset, const RooCmdArg &arg1, const RooCmdArg &arg2={}, const RooCmdArg &arg3={}, const RooCmdArg &arg4={}, const RooCmdArg &arg5={}, const RooCmdArg &arg6={}, const RooCmdArg &arg7={}, const RooCmdArg &arg8={}) const
Create an object that represents the integral of the function over one or more observables listed in ...
RooArgList is a container object that can hold multiple RooAbsArg objects.
Definition RooArgList.h:22
RooArgSet is a container object that can hold multiple RooAbsArg objects.
Definition RooArgSet.h:24
Efficient implementation of the generator context specific for binned pdfs.
Int_t setObj(const RooArgSet *nset, T *obj, const TNamed *isetRangeName=nullptr)
Setter function without integration set.
T * getObj(const RooArgSet *nset, Int_t *sterileIndex=nullptr, const TNamed *isetRangeName=nullptr)
Getter function without integration set.
Implementation of RooAbsCachedReal that can cache any external RooAbsReal input function provided in ...
Named container for two doubles, two integers two object points and three string pointers that can be...
Definition RooCmdArg.h:26
Configurable parser for RooCmdArg named arguments.
void defineMutex(const char *head, Args_t &&... tail)
Define arguments where any pair is mutually exclusive.
bool process(const RooCmdArg &arg)
Process given RooCmdArg.
bool hasProcessed(const char *cmdName) const
Return true if RooCmdArg with name 'cmdName' has been processed.
double getDouble(const char *name, double defaultValue=0.0) const
Return double property registered with name 'name'.
bool defineDouble(const char *name, const char *argName, int doubleNum, double defValue=0.0)
Define double property name 'name' mapped to double in slot 'doubleNum' in RooCmdArg with name argNam...
static void stripCmdList(RooLinkedList &cmdList, const char *cmdsToPurge)
Utility function that strips command names listed (comma separated) in cmdsToPurge from cmdList.
RooArgSet * getSet(const char *name, RooArgSet *set=nullptr) const
Return RooArgSet property registered with name 'name'.
bool defineSet(const char *name, const char *argName, int setNum, const RooArgSet *set=nullptr)
Define TObject property name 'name' mapped to object in slot 'setNum' in RooCmdArg with name argName ...
bool ok(bool verbose) const
Return true of parsing was successful.
bool defineObject(const char *name, const char *argName, int setNum, const TObject *obj=nullptr, bool isArray=false)
Define TObject property name 'name' mapped to object in slot 'setNum' in RooCmdArg with name argName ...
const char * getString(const char *name, const char *defaultValue="", bool convEmptyToNull=false) const
Return string property registered with name 'name'.
bool defineString(const char *name, const char *argName, int stringNum, const char *defValue="", bool appendMode=false)
Define double property name 'name' mapped to double in slot 'stringNum' in RooCmdArg with name argNam...
bool defineInt(const char *name, const char *argName, int intNum, int defValue=0)
Define integer property name 'name' mapped to integer in slot 'intNum' in RooCmdArg with name argName...
void allowUndefined(bool flag=true)
If flag is true the processing of unrecognized RooCmdArgs is not considered an error.
int getInt(const char *name, int defaultValue=0) const
Return integer property registered with name 'name'.
TObject * getObject(const char *name, TObject *obj=nullptr) const
Return TObject property registered with name 'name'.
Container class to hold unbinned data.
Definition RooDataSet.h:32
void markAsCompiled(RooAbsArg &arg) const
void compileServers(RooAbsArg &arg, RooArgSet const &normSet)
Implements a universal generator context for all RooAbsPdf classes that do not have or need a special...
Switches the message service to a different level while the instance is alive.
Definition RooHelpers.h:37
Collection class for internal use, storing a collection of RooAbsArg pointers in a doubly linked list...
static const char * str(const TNamed *ptr)
Return C++ string corresponding to given TNamed pointer.
Definition RooNameReg.h:39
Holds the configuration parameters of the various numeric integrators used by RooRealIntegral.
static RooNumGenConfig & defaultConfig()
Return reference to instance of default numeric integrator configuration object.
Implementation of a RooCacheManager<RooAbsCacheElement> that specializes in the storage of cache elem...
void sterilize() override
Clear the cache payload but retain slot mapping w.r.t to normalization and integration sets.
Plot frame and a container for graphics objects within that frame.
Definition RooPlot.h:43
void addObject(TObject *obj, Option_t *drawOptions="", bool invisible=false)
Add a generic object to this plot.
Definition RooPlot.cxx:325
double getFitRangeNEvt() const
Return the number of events in the fit range.
Definition RooPlot.h:139
const RooArgSet * getNormVars() const
Definition RooPlot.h:146
RooAbsRealLValue * getPlotVar() const
Definition RooPlot.h:137
void updateNormVars(const RooArgSet &vars)
Install the given set of observables are reference normalization variables for this frame.
Definition RooPlot.cxx:310
double getFitRangeBinW() const
Return the bin width that is being used to normalise the PDF.
Definition RooPlot.h:142
virtual void printStream(std::ostream &os, Int_t contents, StyleOption style, TString indent="") const
Print description of object on ostream, printing contents set by contents integer,...
A RooAbsPdf implementation that represent a projection of a given input p.d.f and the object returned...
static UInt_t integer(UInt_t max, TRandom *generator=randomGenerator())
Return an integer uniformly distributed from [0,n-1].
Definition RooRandom.cxx:95
static TRandom * randomGenerator()
Return a pointer to a singleton random-number generator implementation.
Definition RooRandom.cxx:47
Performs hybrid numerical/analytical integrals of RooAbsReal objects.
const RooArgSet & numIntRealVars() const
Variable that can be changed from the outside.
Definition RooRealVar.h:37
The TNamed class is the base class for all named ROOT classes.
Definition TNamed.h:29
const char * GetName() const override
Returns name of object.
Definition TNamed.h:49
const char * GetTitle() const override
Returns title of object.
Definition TNamed.h:50
virtual const char * ClassName() const
Returns name of class to which the object belongs.
Definition TObject.cxx:227
A Pave (see TPave) with text, lines or/and boxes inside.
Definition TPaveText.h:21
Basic string class.
Definition TString.h:138
Ssiz_t Length() const
Definition TString.h:425
const char * Data() const
Definition TString.h:384
TLine * line
void box(Int_t pat, Double_t x1, Double_t y1, Double_t x2, Double_t y2)
Definition fillpatterns.C:1
RooCmdArg SupNormSet(const RooArgSet &nset)
RooCmdArg NormRange(const char *rangeNameList)
RooCmdArg Range(const char *rangeName, bool adjustNorm=true)
RooCmdArg Normalization(double scaleFactor)
std::vector< std::string > Split(std::string_view str, std::string_view delims, bool skipEmpty=false)
Splits a string at each character in delims.
T * OwningPtr
An alias for raw pointers for indicating that the return type of a RooFit function is an owning point...
Definition Config.h:35
OwningPtr< T > makeOwningPtr(std::unique_ptr< T > &&ptr)
Internal helper to turn a std::unique_ptr<T> into an OwningPtr.
Definition Config.h:40
RooArgSet selectFromArgSet(RooArgSet const &, std::string const &names)
std::string getColonSeparatedNameString(RooArgSet const &argSet, char delim=':')
Bool_t IsNaN(Double_t x)
Definition TMath.h:898
Double_t QuietNaN()
Returns a quiet NaN as defined by IEEE 754.
Definition TMath.h:908
static double packFloatIntoNaN(float payload)
Pack float into mantissa of a NaN.
static void output()