1 // This file is part of Eigen, a lightweight C++ template library
2 // for linear algebra.
3 //
4 // Copyright (C) 2009 Gael Guennebaud <gael.guennebaud@inria.fr>
5 //
6 // This Source Code Form is subject to the terms of the Mozilla
7 // Public License v. 2.0. If a copy of the MPL was not distributed
8 // with this file, You can obtain one at http://mozilla.org/MPL/2.0/.
9 
10 #ifndef EIGEN_AUTODIFF_JACOBIAN_H
11 #define EIGEN_AUTODIFF_JACOBIAN_H
12 
13 namespace Eigen
14 {
15 
16 template<typename Functor> class AutoDiffJacobian : public Functor
17 {
18 public:
AutoDiffJacobian()19   AutoDiffJacobian() : Functor() {}
AutoDiffJacobian(const Functor & f)20   AutoDiffJacobian(const Functor& f) : Functor(f) {}
21 
22   // forward constructors
23   template<typename T0>
AutoDiffJacobian(const T0 & a0)24   AutoDiffJacobian(const T0& a0) : Functor(a0) {}
25   template<typename T0, typename T1>
AutoDiffJacobian(const T0 & a0,const T1 & a1)26   AutoDiffJacobian(const T0& a0, const T1& a1) : Functor(a0, a1) {}
27   template<typename T0, typename T1, typename T2>
AutoDiffJacobian(const T0 & a0,const T1 & a1,const T2 & a2)28   AutoDiffJacobian(const T0& a0, const T1& a1, const T2& a2) : Functor(a0, a1, a2) {}
29 
30   enum {
31     InputsAtCompileTime = Functor::InputsAtCompileTime,
32     ValuesAtCompileTime = Functor::ValuesAtCompileTime
33   };
34 
35   typedef typename Functor::InputType InputType;
36   typedef typename Functor::ValueType ValueType;
37   typedef typename Functor::JacobianType JacobianType;
38   typedef typename JacobianType::Scalar Scalar;
39   typedef typename JacobianType::Index Index;
40 
41   typedef Matrix<Scalar,InputsAtCompileTime,1> DerivativeType;
42   typedef AutoDiffScalar<DerivativeType> ActiveScalar;
43 
44 
45   typedef Matrix<ActiveScalar, InputsAtCompileTime, 1> ActiveInput;
46   typedef Matrix<ActiveScalar, ValuesAtCompileTime, 1> ActiveValue;
47 
operator()48   void operator() (const InputType& x, ValueType* v, JacobianType* _jac=0) const
49   {
50     eigen_assert(v!=0);
51     if (!_jac)
52     {
53       Functor::operator()(x, v);
54       return;
55     }
56 
57     JacobianType& jac = *_jac;
58 
59     ActiveInput ax = x.template cast<ActiveScalar>();
60     ActiveValue av(jac.rows());
61 
62     if(InputsAtCompileTime==Dynamic)
63       for (Index j=0; j<jac.rows(); j++)
64         av[j].derivatives().resize(this->inputs());
65 
66     for (Index i=0; i<jac.cols(); i++)
67       ax[i].derivatives() = DerivativeType::Unit(this->inputs(),i);
68 
69     Functor::operator()(ax, &av);
70 
71     for (Index i=0; i<jac.rows(); i++)
72     {
73       (*v)[i] = av[i].value();
74       jac.row(i) = av[i].derivatives();
75     }
76   }
77 protected:
78 
79 };
80 
81 }
82 
83 #endif // EIGEN_AUTODIFF_JACOBIAN_H
84