Logo ROOT  
Reference Guide
TFumiliMinimizer.cxx
Go to the documentation of this file.
1// @(#)root/fumili:$Id$
2// Author: L. Moneta Wed Oct 25 16:28:55 2006
3
4/**********************************************************************
5 * *
6 * Copyright (c) 2006 LCG ROOT Math Team, CERN/PH-SFT *
7 * *
8 * *
9 **********************************************************************/
10
11// Implementation file for class TFumiliMinimizer
12
13#include "TFumiliMinimizer.h"
14#include "Math/IFunction.h"
15#include "Math/Util.h"
16#include "TError.h"
17
18#include "TFumili.h"
19
20#include <iostream>
21#include <cassert>
22#include <algorithm>
23#include <functional>
24
25
26// setting USE_FUMILI_FUNCTION will use the Derivatives provided by Fumili
27// instead of what proided in FitUtil::EvalChi2Residual
28// t.d.: use still standard Chi2 but replace model function
29// with a gradient function where gradient is computed by TFumili
30// since TFumili knows the step size can calculate it better
31// Derivative in FUmili are very fast (1 extra call for each parameter)
32// + 1 function evaluation
33//
34//#define USE_FUMILI_FUNCTION
35#ifdef USE_FUMILI_FUNCTION
36bool gUseFumiliFunction = true;
37//#include "FumiliFunction.h"
38// fit method function used in TFumiliMinimizer
39
42#include "Fit/Chi2FCN.h"
43#include "TF1.h"
44#include "TFumili.h"
45
46template<class MethodFunc>
47class FumiliFunction : public ROOT::Math::FitMethodFunction {
48
50
51public:
52 FumiliFunction(TFumili * fumili, const ROOT::Math::FitMethodFunction * func) :
53 ROOT::Math::FitMethodFunction(func->NDim(), func->NPoints() ),
54 fFumili(fumili),
55 fObjFunc(0)
56 {
57 fObjFunc = dynamic_cast<const MethodFunc *>(func);
58 assert(fObjFunc != 0);
59
60 // create TF1 class from model function
61 fModFunc = new TF1("modfunc",ROOT::Math::ParamFunctor( &fObjFunc->ModelFunction() ) );
62 fFumili->SetUserFunc(fModFunc);
63 }
64
65 ROOT::Math::FitMethodFunction::Type_t Type() const { return fObjFunc->Type(); }
66
67 FumiliFunction * Clone() const { return new FumiliFunction(fFumili, fObjFunc); }
68
69
70 // recalculate data elemet using Fumili stuff
71 double DataElement(const double * /*par */, unsigned int i, double * g) const {
72
73 // parameter values are inside TFumili
74
75 // suppose type is bin likelihood
76 unsigned int npar = fObjFunc->NDim();
77 double y = 0;
78 double invError = 0;
79 const double *x = fObjFunc->Data().GetPoint(i,y,invError);
80 double fval = fFumili->EvalTFN(g,const_cast<double *>( x));
81 fFumili->Derivatives(g, const_cast<double *>( x));
82
83 if ( fObjFunc->Type() == ROOT::Math::FitMethodFunction::kLogLikelihood) {
84 double logPdf = y * ROOT::Math::Util::EvalLog( fval) - fval;
85 for (unsigned int k = 0; k < npar; ++k) {
86 g[k] *= ( y/fval - 1.) ;//* pdfval;
87 }
88
89 // std::cout << "x = " << x[0] << " logPdf = " << logPdf << " grad";
90// for (unsigned int ipar = 0; ipar < npar; ++ipar)
91// std::cout << g[ipar] << "\t";
92// std::cout << std::endl;
93
94 return logPdf;
95 }
96 else if (fObjFunc->Type() == ROOT::Math::FitMethodFunction::kLeastSquare ) {
97 double resVal = (y-fval)*invError;
98 for (unsigned int k = 0; k < npar; ++k) {
99 g[k] *= -invError;
100 }
101 return resVal;
102 }
103
104 return 0;
105 }
106
107
108private:
109
110 double DoEval(const double *x ) const {
111 return (*fObjFunc)(x);
112 }
113
114 TFumili * fFumili;
115 const MethodFunc * fObjFunc;
116 TF1 * fModFunc;
117
118};
119#else
121#endif
122//______________________________________________________________________________
123//
124// TFumiliMinimizer class implementing the ROOT::Math::Minimizer interface using
125// TFumili.
126// This class is normally instantiates using the plug-in manager
127// (plug-in with name Fumili or TFumili)
128// In addition the user can choose the minimizer algorithm: Migrad (the default one), Simplex, or Minimize (combined Migrad + Simplex)
129//
130//__________________________________________________________________________________________
131
132// initialize the static instances
133
137
138
140
141
143 fDim(0),
144 fNFree(0),
145 fMinVal(0),
146 fEdm(-1),
147 fFumili(0)
148{
149 // Constructor for TFumiliMinimier class
150
151 // construct with npar = 0 (by default a value of 25 is used in TFumili for allocating the arrays)
152#ifdef USE_STATIC_TMINUIT
153 // allocate here only the first time
154 if (fgFumili == 0) fgFumili = new TFumili(0);
156#else
157 if (fFumili) delete fFumili;
158 fFumili = new TFumili(0);
160#endif
161
162}
163
164
166{
167 // Destructor implementation.
168 if (fFumili) delete fFumili;
169}
170
172 Minimizer()
173{
174 // Implementation of copy constructor (it is private).
175}
176
178{
179 // Implementation of assignment operator (private)
180 if (this == &rhs) return *this; // time saving self-test
181 return *this;
182}
183
184
185
187 // Set the objective function to be minimized, by passing a function object implement the
188 // basic multi-dim Function interface. In this case the derivatives will be
189 // calculated by Fumili
190
191 // Here a TFumili instance is created since only at this point we know the number of parameters
192 // needed to create TFumili
193 fDim = func.NDim();
195
196 // for Fumili the fit method function interface is required
197 const ROOT::Math::FitMethodFunction * fcnfunc = dynamic_cast<const ROOT::Math::FitMethodFunction *>(&func);
198 if (!fcnfunc) {
199 Error("SetFunction","Wrong Fit method function type used for Fumili");
200 return;
201 }
202 // assign to the static pointer (NO Thread safety here)
203 fgFunc = const_cast<ROOT::Math::FitMethodFunction *>(fcnfunc);
204 fgGradFunc = 0;
206
207#ifdef USE_FUMILI_FUNCTION
208 if (gUseFumiliFunction) {
213 }
214#endif
215
216}
217
219 // Set the objective function to be minimized, by passing a function object implement the
220 // multi-dim gradient Function interface. In this case the function derivatives are provided
221 // by the user via this interface and there not calculated by Fumili.
222
223 fDim = func.NDim();
225
226 // for Fumili the fit method function interface is required
227 const ROOT::Math::FitMethodGradFunction * fcnfunc = dynamic_cast<const ROOT::Math::FitMethodGradFunction *>(&func);
228 if (!fcnfunc) {
229 Error("SetFunction","Wrong Fit method function type used for Fumili");
230 return;
231 }
232 // assign to the static pointer (NO Thread safety here)
233 fgFunc = 0;
234 fgGradFunc = const_cast<ROOT::Math::FitMethodGradFunction *>(fcnfunc);
236
237}
238
239void TFumiliMinimizer::Fcn( int & , double * g , double & f, double * x , int /* iflag */) {
240 // implementation of FCN static function used internally by TFumili.
241 // Adapt IMultiGenFunction interface to TFumili FCN static function
242 f = TFumiliMinimizer::EvaluateFCN(const_cast<double*>(x),g);
243}
244
245// void TFumiliMinimizer::FcnGrad( int &, double * g, double & f, double * x , int iflag ) {
246// // implementation of FCN static function used internally by TFumili.
247// // Adapt IMultiGradFunction interface to TFumili FCN static function in the case of user
248// // provided gradient.
249// ROOT::Math::IMultiGradFunction * gFunc = dynamic_cast<ROOT::Math::IMultiGradFunction *> ( fgFunc);
250
251// assert(gFunc != 0);
252// f = gFunc->operator()(x);
253
254// // calculates also derivatives
255// if (iflag == 2) gFunc->Gradient(x,g);
256// }
257
258double TFumiliMinimizer::EvaluateFCN(const double * x, double * grad) {
259 // function callaed to evaluate the FCN at the value x
260 // calculates also the matrices of the second derivatives of the objective function needed by FUMILI
261
262
263 //typedef FumiliFCNAdapter::Function Function;
264
265
266
267 // reset
268// assert(grad.size() == npar);
269// grad.assign( npar, 0.0);
270// hess.assign( hess.size(), 0.0);
271
272 double sum = 0;
273 unsigned int ndata = 0;
274 unsigned int npar = 0;
275 if (fgFunc) {
276 ndata = fgFunc->NPoints();
277 npar = fgFunc->NDim();
279 }
280 else if (fgGradFunc) {
281 ndata = fgGradFunc->NPoints();
282 npar = fgGradFunc->NDim();
284 }
285
286 // eventually store this matrix as static member to optimize speed
287 std::vector<double> gf(npar);
288 std::vector<double> hess(npar*(npar+1)/2);
289
290 // reset gradients
291 for (unsigned int ipar = 0; ipar < npar; ++ipar)
292 grad[ipar] = 0;
293
294
295 //loop on the data points
296//#define DEBUG
297#ifdef DEBUG
298 std::cout << "=============================================";
299 std::cout << "par = ";
300 for (unsigned int ipar = 0; ipar < npar; ++ipar)
301 std::cout << x[ipar] << "\t";
302 std::cout << std::endl;
303 if (fgFunc) std::cout << "type " << fgFunc->Type() << std::endl;
304#endif
305
306
307 // assume for now least-square
308 // since TFumili doet not use errodef I must diveide chi2 by 2
311
312 double fval = 0;
313 for (unsigned int i = 0; i < ndata; ++i) {
314 // calculate data element and gradient
315 // DataElement returns (f-y)/s and gf is derivatives of model function multiplied by (-1/sigma)
316 if (gUseFumiliFunction) {
317 fval = fgFunc->DataElement( x, i, &gf[0]);
318 }
319 else {
320 if (fgFunc != 0)
321 fval = fgFunc->DataElement(x, i, &gf[0]);
322 else
323 fval = fgGradFunc->DataElement(x, i, &gf[0]);
324 }
325
326 // t.b.d should protect for bad values of fval
327 sum += fval*fval;
328
329 // to be check (TFumili uses a factor of 1/2 for chi2)
330
331 for (unsigned int j = 0; j < npar; ++j) {
332 grad[j] += fval * gf[j];
333 for (unsigned int k = j; k < npar; ++ k) {
334 int idx = j + k*(k+1)/2;
335 hess[idx] += gf[j] * gf[k];
336 }
337 }
338 }
339 }
342
343
344
345 double fval = 0;
346
347 //std::cout << "\t x " << x[0] << " " << x[1] << " " << x[2] << std::endl;
348
349 for (unsigned int i = 0; i < ndata; ++i) {
350
351 if (gUseFumiliFunction) {
352 fval = fgFunc->DataElement( x, i, &gf[0]);
353 }
354 else {
355 // calculate data element and gradient
356 if (fgFunc != 0)
357 fval = fgFunc->DataElement(x, i, &gf[0]);
358 else
359 fval = fgGradFunc->DataElement(x, i, &gf[0]);
360 }
361
362 // protect for small values of fval
363 // std::cout << i << " " << fval << " log " << " grad " << gf[0] << " " << gf[1] << " " << gf[2] << std::endl;
364// sum -= ROOT::Math::Util::EvalLog(fval);
365 sum -= fval;
366
367 for (unsigned int j = 0; j < npar; ++j) {
368 double gfj = gf[j];// / fval;
369 grad[j] -= gfj;
370 for (unsigned int k = j; k < npar; ++ k) {
371 int idx = j + k*(k+1)/2;
372 hess[idx] += gfj * gf[k];// / (fval );
373 }
374 }
375 }
376 }
377 else {
378 Error("EvaluateFCN"," type of fit method is not supported, it must be chi2 or log-likelihood");
379 }
380
381 // now TFumili excludes fixed prameter in second-derivative matrix
382 // ned to get them using the static instance of TFumili
383 double * zmatrix = fgFumili->GetZ();
384 double * pl0 = fgFumili->GetPL0(); // parameter limits
385 assert(zmatrix != 0);
386 assert(pl0 != 0);
387 unsigned int k = 0;
388 unsigned int l = 0;
389 for (unsigned int i = 0; i < npar; ++i) {
390 for (unsigned int j = 0; j <= i; ++j) {
391 if (pl0[i] > 0 && pl0[j] > 0) { // only for non-fixed parameters
392 zmatrix[l++] = hess[k];
393 }
394 k++;
395 }
396 }
397
398#ifdef DEBUG
399 std::cout << "FCN value " << sum << " grad ";
400 for (unsigned int ipar = 0; ipar < npar; ++ipar)
401 std::cout << grad[ipar] << "\t";
402 std::cout << std::endl << std::endl;
403#endif
404
405
406 return 0.5*sum; // fumili multiply then by 2
407
408}
409
410
411
412bool TFumiliMinimizer::SetVariable(unsigned int ivar, const std::string & name, double val, double step) {
413 // set a free variable.
414 if (fFumili == 0) {
415 Error("SetVariableValue","invalid TFumili pointer. Set function first ");
416 return false;
417 }
418#ifdef DEBUG
419 std::cout << "set variable " << ivar << " " << name << " value " << val << " step " << step << std::endl;
420#endif
421
422 int ierr = fFumili->SetParameter(ivar , name.c_str(), val, step, 0., 0. );
423 if (ierr) {
424 Error("SetVariable","Error for parameter %d ",ivar);
425 return false;
426 }
427 return true;
428}
429
430bool TFumiliMinimizer::SetLimitedVariable(unsigned int ivar, const std::string & name, double val, double step, double lower, double upper) {
431 // set a limited variable.
432 if (fFumili == 0) {
433 Error("SetVariableValue","invalid TFumili pointer. Set function first ");
434 return false;
435 }
436#ifdef DEBUG
437 std::cout << "set limited variable " << ivar << " " << name << " value " << val << " step " << step << std::endl;
438#endif
439 int ierr = fFumili->SetParameter(ivar, name.c_str(), val, step, lower, upper );
440 if (ierr) {
441 Error("SetLimitedVariable","Error for parameter %d ",ivar);
442 return false;
443 }
444 return true;
445}
446#ifdef LATER
447bool Fumili2Minimizer::SetLowerLimitedVariable(unsigned int ivar , const std::string & name , double val , double step , double lower ) {
448 // add a lower bounded variable as a double bound one, using a very large number for the upper limit
449 double s = val-lower;
450 double upper = s*1.0E15;
451 if (s != 0) upper = 1.0E15;
452 return SetLimitedVariable(ivar, name, val, step, lower,upper);
453}
454#endif
455
456
457bool TFumiliMinimizer::SetFixedVariable(unsigned int ivar, const std::string & name, double val) {
458 // set a fixed variable.
459 if (fFumili == 0) {
460 Error("SetVariableValue","invalid TFumili pointer. Set function first ");
461 return false;
462 }
463
464
465 int ierr = fFumili->SetParameter(ivar, name.c_str(), val, 0., val, val );
466 fFumili->FixParameter(ivar);
467
468#ifdef DEBUG
469 std::cout << "Fix variable " << ivar << " " << name << " value " << std::endl;
470#endif
471
472 if (ierr) {
473 Error("SetFixedVariable","Error for parameter %d ",ivar);
474 return false;
475 }
476 return true;
477}
478
479bool TFumiliMinimizer::SetVariableValue(unsigned int ivar, double val) {
480 // set the variable value
481 if (fFumili == 0) {
482 Error("SetVariableValue","invalid TFumili pointer. Set function first ");
483 return false;
484 }
486 double oldval, verr, vlow, vhigh = 0;
487 int ierr = fFumili->GetParameter( ivar, &name[0], oldval, verr, vlow, vhigh);
488 if (ierr) {
489 Error("SetVariableValue","Error for parameter %d ",ivar);
490 return false;
491 }
492#ifdef DEBUG
493 std::cout << "set variable " << ivar << " " << name << " value "
494 << val << " step " << verr << std::endl;
495#endif
496
497 ierr = fFumili->SetParameter(ivar , name , val, verr, vlow, vhigh );
498 if (ierr) {
499 Error("SetVariableValue","Error for parameter %d ",ivar);
500 return false;
501 }
502 return true;
503}
504
506 // perform the minimization using the algorithm chosen previously by the user
507 // By default Migrad is used.
508 // Return true if the found minimum is valid and update internal chached values of
509 // minimum values, errors and covariance matrix.
510
511 if (fFumili == 0) {
512 Error("SetVariableValue","invalid TFumili pointer. Set function first ");
513 return false;
514 }
515
516 // need to set static instance to be used when calling FCN
518
519
520 double arglist[10];
521
522 // error cannot be set in TFumili (always the same)
523// arglist[0] = ErrorUp();
524// fFumili->ExecuteCommand("SET Err",arglist,1);
525
526 int printlevel = PrintLevel();
527 // not implemented in TFumili yet
528 //arglist[0] = printlevel - 1;
529 //fFumili->ExecuteCommand("SET PRINT",arglist,1,ierr);
530
531 // suppress warning in case Printlevel() == 0
532 if (printlevel == 0) fFumili->ExecuteCommand("SET NOW",arglist,0);
533 else fFumili->ExecuteCommand("SET WAR",arglist,0);
534
535
536 // minimize: use ExecuteCommand instead of Minimize to set tolerance and maxiter
537
538 arglist[0] = MaxFunctionCalls();
539 arglist[1] = Tolerance();
540
541 if (printlevel > 0)
542 std::cout << "Minimize using TFumili with tolerance = " << Tolerance()
543 << " max calls " << MaxFunctionCalls() << std::endl;
544
545 int iret = fFumili->ExecuteCommand("MIGRAD",arglist,2);
546 fStatus = iret;
547 //int iret = fgFumili->Minimize();
548
549 // Hesse and IMP not implemented
550// // run improved if needed
551// if (ierr == 0 && fType == ROOT::Fumili::kMigradImproved)
552// fFumili->mnexcm("IMPROVE",arglist,1,ierr);
553
554// // check if Hesse needs to be run
555// if (ierr == 0 && IsValidError() ) {
556// fFumili->mnexcm("HESSE",arglist,1,ierr);
557// }
558
559
560 int ntot;
561 int nfree;
562 double errdef = 0; // err def is not used by Fumili
563 fFumili->GetStats(fMinVal,fEdm,errdef,nfree,ntot);
564
565 if (printlevel > 0)
566 fFumili->PrintResults(printlevel,fMinVal);
567
568
569 assert (static_cast<unsigned int>(ntot) == fDim);
570 assert( nfree == fFumili->GetNumberFreeParameters() );
571 fNFree = nfree;
572
573
574 // get parameter values and correlation matrix
575 // fumili stores only lower part of diagonal matrix of the free parameters
576 fParams.resize( fDim);
577 fErrors.resize( fDim);
578 fCovar.resize(fDim*fDim);
579 const double * cv = fFumili->GetCovarianceMatrix();
580 unsigned int l = 0;
581 for (unsigned int i = 0; i < fDim; ++i) {
582 fParams[i] = fFumili->GetParameter( i );
583 fErrors[i] = fFumili->GetParError( i );
584
585 if ( !fFumili->IsFixed(i) ) {
586 for (unsigned int j = 0; j <=i ; ++j) {
587 if ( !fFumili->IsFixed(j) ) {
588 fCovar[i*fDim + j] = cv[l];
589 fCovar[j*fDim + i] = fCovar[i*fDim + j];
590 l++;
591 }
592 }
593 }
594 }
595
596 return (iret==0) ? true : false;
597}
598
599
600// } // end namespace Fit
601
602// } // end namespace ROOT
603
#define f(i)
Definition: RSha256.hxx:104
#define g(i)
Definition: RSha256.hxx:105
#define ClassImp(name)
Definition: Rtypes.h:365
void Error(const char *location, const char *msgfmt,...)
bool gUseFumiliFunction
char name[80]
Definition: TGX11.cxx:109
Chi2FCN class for binnned fits using the least square methods.
Definition: Chi2FCN.h:49
class evaluating the log likelihood for binned Poisson likelihood fits it is template to distinguish ...
FitMethodFunction class Interface for objective functions (like chi2 and likelihood used in the fit) ...
virtual Type_t Type() const
return the type of method, override if needed
Type_t
enumeration specyfing the possible fit method types
virtual double DataElement(const double *x, unsigned int i, double *g=0) const =0
method returning the data i-th contribution to the fit objective function For example the residual fo...
virtual unsigned int NPoints() const
return the number of data points used in evaluating the function
FunctionType::BaseFunc BaseFunction
virtual unsigned int NDim() const
Number of dimension (parameters) .
virtual void UpdateNCalls() const
update number of calls
Documentation for the abstract class IBaseFunctionMultiDim.
Definition: IFunction.h:62
virtual unsigned int NDim() const =0
Retrieve the dimension of the function.
Interface (abstract class) for multi-dimensional functions providing a gradient calculation.
Definition: IFunction.h:327
virtual unsigned int NDim() const=0
Retrieve the dimension of the function.
double Tolerance() const
absolute tolerance
Definition: Minimizer.h:420
unsigned int MaxFunctionCalls() const
max number of function calls
Definition: Minimizer.h:414
int PrintLevel() const
minimizer configuration parameters
Definition: Minimizer.h:411
Param Functor class for Multidimensional functions.
Definition: ParamFunctor.h:274
1-Dim function class
Definition: TF1.h:211
TFumiliMinimizer class: minimizer implementation based on TFumili.
~TFumiliMinimizer()
Destructor (no operations)
virtual bool Minimize()
method to perform the minimization
static ROOT::Math::FitMethodFunction * fgFunc
TFumiliMinimizer(int dummy=0)
Default constructor (an argument is needed by plug-in manager)
virtual bool SetVariableValue(unsigned int ivar, double val)
set the value of an existing variable
std::vector< double > fParams
virtual bool SetFixedVariable(unsigned int, const std::string &, double)
set fixed variable (override if minimizer supports them )
static double EvaluateFCN(const double *x, double *g)
implementation of FCN for Fumili when user provided gradient is used
virtual bool SetVariable(unsigned int ivar, const std::string &name, double val, double step)
set free variable
virtual bool SetLimitedVariable(unsigned int ivar, const std::string &name, double val, double step, double, double)
set upper/lower limited variable (override if minimizer supports them )
static ROOT::Math::FitMethodGradFunction * fgGradFunc
TFumiliMinimizer & operator=(const TFumiliMinimizer &rhs)
Assignment operator.
static TFumili * fgFumili
virtual void SetFunction(const ROOT::Math::IMultiGenFunction &func)
set the function to minimize
unsigned int fNFree
std::vector< double > fErrors
std::vector< double > fCovar
static void Fcn(int &, double *, double &f, double *, int)
implementation of FCN for Fumili
virtual Int_t GetNumberFreeParameters() const
Return the number of free parameters.
Definition: TFumili.cxx:805
virtual Double_t GetParError(Int_t ipar) const
Return error of parameter ipar.
Definition: TFumili.cxx:817
virtual void PrintResults(Int_t k, Double_t p) const
Prints fit results.
Definition: TFumili.cxx:1474
virtual Int_t ExecuteCommand(const char *command, Double_t *args, Int_t nargs)
Execute MINUIT commands.
Definition: TFumili.cxx:382
virtual void FixParameter(Int_t ipar)
Fixes parameter number ipar.
Definition: TFumili.cxx:765
Double_t * GetPL0() const
Definition: TFumili.h:95
void SetParNumber(Int_t ParNum)
Definition: TFumili.h:112
virtual Int_t GetStats(Double_t &amin, Double_t &edm, Double_t &errdef, Int_t &nvpar, Int_t &nparx) const
Return global fit parameters.
Definition: TFumili.cxx:895
Double_t * GetZ() const
Definition: TFumili.h:102
virtual Bool_t IsFixed(Int_t ipar) const
Return kTRUE if parameter ipar is fixed, kFALSE otherwise)
Definition: TFumili.cxx:1051
virtual Double_t * GetCovarianceMatrix() const
Return a pointer to the covariance matrix.
Definition: TFumili.cxx:775
virtual Int_t SetParameter(Int_t ipar, const char *parname, Double_t value, Double_t verr, Double_t vlow, Double_t vhigh)
Sets for parameter number ipar initial parameter value, name parname, initial error verr and limits v...
Definition: TFumili.cxx:1632
virtual Double_t GetParameter(Int_t ipar) const
Return current value of parameter ipar.
Definition: TFumili.cxx:826
virtual const char * GetParName(Int_t ipar) const
Return name of parameter ipar.
Definition: TFumili.cxx:863
Basic string class.
Definition: TString.h:131
virtual void SetFCN(void(*fcn)(Int_t &, Double_t *, Double_t &f, Double_t *, Int_t))
To set the address of the minimization objective function called by the native compiler (see function...
Double_t y[n]
Definition: legend1.C:17
Double_t x[n]
Definition: legend1.C:17
Namespace for new Math classes and functions.
T EvalLog(T x)
safe evaluation of log(x) with a protections against negative or zero argument to the log smooth line...
Definition: Util.h:64
BasicFitMethodFunction< ROOT::Math::IMultiGenFunction > FitMethodFunction
Definition: Fitter.h:40
VSD Structures.
Definition: StringConv.hxx:21
RooCmdArg Minimizer(const char *type, const char *alg=0)
static constexpr double s
auto * l
Definition: textangle.C:4
static long int sum(long int i)
Definition: Factory.cxx:2276