Logo ROOT   6.18/05
Reference Guide
TRFunctionExport.h
Go to the documentation of this file.
1// @(#)root/r:$Id$
2// Author: Omar Zapata 16/06/2013
3
4
5/*************************************************************************
6 * Copyright (C) 2013-2015, Omar Andres Zapata Mesa *
7 * All rights reserved. *
8 * *
9 * For the licensing terms see $ROOTSYS/LICENSE. *
10 * For the list of contributors see $ROOTSYS/README/CREDITS. *
11 *************************************************************************/
12#ifndef ROOT_R_TRFunctionExport
13#define ROOT_R_TRFunctionExport
14
15#include <TRInternalFunction.h>
16
17
18namespace ROOT {
19 namespace R {
20
21 /**
22 \class TRFunctionExport
23
24 This is a class to pass functions from ROOT to R
25 <center><h2>TRFunctionExport class</h2></center>
26 <p>
27 The TRFunctionExport class lets you pass ROOT's functions to R's environment<br>
28 </p>
29 <p>
30 The next example was based in <br>
31 <a href="http://root.cern.ch/root/html/tutorials/fit/NumericalMinimization.C.html">
32 http://root.cern.ch/root/html/tutorials/fit/NumericalMinimization.C.html
33 </a><br>
34 <a href="http://stat.ethz.ch/R-manual/R-devel/library/stats/html/optim.html">
35 http://stat.ethz.ch/R-manual/R-devel/library/stats/html/optim.html</a><br>
36
37 </p>
38
39 Let \f$ f(x,y)=(x-1)^{2} + 100(y-x^{2})^{2} \f$ , which is called the Rosenbrock
40 function.
41
42 It's known that this function has a minimum when \f$ y = x^{2}\f$ , and \f$ x = 1.\f$
43 Let's get the minimum using R's optim package through ROOTR's interface.
44 In the code this function was called "Double_t RosenBrock(const TVectorD xx )", because for
45 optim, the input in your function deļ¬nition must be a single vector.
46
47 The Gradient is formed by
48
49 \f$ \frac{\partial f}{\partial x} = -400x(y - x^{2}) - 2(1 - x) \f$
50
51 \f$ \frac{\partial f}{\partial y} = 200(y - x^{2}); \f$
52
53 The "TVectorD RosenBrockGrad(const TVectorD xx )" function
54 must have a single vector as the argument a it will return a single vetor.
55
56 \code{.cpp}
57 #include<TRInterface.h>
58
59 //in the next function the pointer *double must be changed by TVectorD, because the pointer has no
60 //sense in R's environment.
61 Double_t RosenBrock(const TVectorD xx )
62 {
63 const Double_t x = xx[0];
64 const Double_t y = xx[1];
65 const Double_t tmp1 = y-x*x;
66 const Double_t tmp2 = 1-x;
67 return 100*tmp1*tmp1+tmp2*tmp2;
68 }
69
70 TVectorD RosenBrockGrad(const TVectorD xx )
71 {
72 const Double_t x = xx[0];
73 const Double_t y = xx[1];
74 TVectorD grad(2);
75 grad[0]=-400 * x * (y - x * x) - 2 * (1 - x);
76 grad[1]=200 * (y - x * x);
77 return grad;
78 }
79
80
81 void Minimization()
82 {
83 ROOT::R::TRInterface &r=ROOT::R::TRInterface::Instance();
84 //passing RosenBrock function to R
85 r["RosenBrock"]<<ROOT::R::TRFunctionExport(RosenBrock);
86
87 //passing RosenBrockGrad function to R
88 r["RosenBrockGrad"]<<ROOT::R::TRFunctionExport(RosenBrockGrad);
89
90 //the option "method" could be "Nelder-Mead", "BFGS", "CG", "L-BFGS-B", "SANN","Brent"
91 //the option "control" lets you put some constraints like:
92 //"maxit" The maximum number of iterations
93 //"abstol" The absolute convergence tolerance.
94 //"reltol" Relative convergence tolerance.
95 r<<"result <- optim( c(0.01,0.01), RosenBrock,method='BFGS',control = list(maxit = 1000000) )";
96
97 //Getting results from R
98 TVectorD min=r.Eval("result$par");
99
100 std::cout.precision(8);
101 //printing results
102 std::cout<<"-----------------------------------------"<<std::endl;
103 std::cout<<"Minimum x="<<min[0]<<" y="<<min[1]<<std::endl;
104 std::cout<<"Value at minimum ="<<RosenBrock(min)<<std::endl;
105
106 //using the gradient
107 r<<"optimHess(result$par, RosenBrock, RosenBrockGrad)";
108 r<<"hresult <- optim(c(-1.2,1), RosenBrock, NULL, method = 'BFGS', hessian = TRUE)";
109 //getting the minimum calculated with the gradient
110 TVectorD hmin=r.Eval("hresult$par");
111
112 //printing results
113 std::cout<<"-----------------------------------------"<<std::endl;
114 std::cout<<"Minimization with the Gradient"<<endl;
115 std::cout<<"Minimum x="<<hmin[0]<<" y="<<hmin[1]<<std::endl;
116 std::cout<<"Value at minimum ="<<RosenBrock(hmin)<<std::endl;
117
118 }
119 \endcode
120
121 Output
122 \code
123 Processing Minimization.C...
124 -----------------------------------------
125 Minimum x=0.99980006 y=0.99960016
126 Value at minimum =3.9974288e-08
127 -----------------------------------------
128 Minimization with the Gradient
129 Minimum x=0.99980443 y=0.99960838
130 Value at minimum =3.8273828e-08
131 \endcode
132 <h2>Users Guide </h2>
133 <a href="http://oproject.org/tiki-index.php?page=ROOT+R+Users+Guide"> http://oproject.org/tiki-index.php?page=ROOT+R+Users+Guide</a><br>
134 <a href="https://root.cern.ch/drupal/content/how-use-r-root-root-r-interface"> https://root.cern.ch/drupal/content/how-use-r-root-root-r-interface</a>
135
136 @ingroup R
137 */
138
139
140 class TRInterface;
141 class TRFunctionExport: public TObject {
142 friend class TRInterface;
143 friend SEXP Rcpp::wrap<TRFunctionExport>(const TRFunctionExport &f);
144 protected:
145 TRInternalFunction *f; //Internar Function to export
146 public:
147 /**
148 Default TRFunctionExport constructor
149 */
151
152 /**
153 Default TRFunctionExport destructor
154 */
156 {
157 if (f) delete f;
158 }
159 /**
160 TRFunctionExport copy constructor
161 \param fun other TRFunctionExport
162 */
164
165 /**
166 TRFunctionExport template constructor that supports a lot of function's prototypes
167 \param fun supported function to be wrapped by Rcpp
168 */
169 template<class T> TRFunctionExport(T fun)
170 {
171 f = new TRInternalFunction(fun);
172 }
173
174 /**
175 function to assign function to export,
176 template method that supports a lot of function's prototypes
177 \param fun supported function to be wrapped by Rcpp
178 */
179 template<class T> void SetFunction(T fun)
180 {
181 f = new TRInternalFunction(fun);
182 }
183
185 };
186 }
187}
188
189
190
191#endif
#define R(a, b, c, d, e, f, g, h, i)
Definition: RSha256.hxx:110
#define ClassDef(name, id)
Definition: Rtypes.h:326
This is a class to pass functions from ROOT to R.
~TRFunctionExport()
Default TRFunctionExport destructor.
TRInternalFunction * f
void SetFunction(T fun)
function to assign function to export, template method that supports a lot of function's prototypes
TRFunctionExport(T fun)
TRFunctionExport template constructor that supports a lot of function's prototypes.
TRFunctionExport()
Default TRFunctionExport constructor.
ROOT R was implemented using the R Project library and the modules Rcpp and RInside
Definition: TRInterface.h:137
Mother of all ROOT objects.
Definition: TObject.h:37
double T(double x)
Definition: ChebyshevPol.h:34
Rcpp::TRInternalFunction_Impl< Rcpp::PreserveStorage > TRInternalFunction
Namespace for new ROOT classes and functions.
Definition: StringConv.hxx:21