34 template <
typename FUNCTOR,
int M,
int N1,
int N2>
57 double rowMajor1[
M * N1] = {}, rowMajor2[
M * N2] = {};
58 double* jacobians[] = {rowMajor1, rowMajor2};
59 success = AutoDiff<FUNCTOR, double, N1, N2>::Differentiate(
60 f, parameters,
M, result.
data(), jacobians);
73 throw std::runtime_error(
74 "AdaptAutoDiff: function call resulted in failure");
Eigen::Matrix< double, M, 1 > VectorT
VectorT operator()(const Vector1 &v1, const Vector2 &v2, OptionalJacobian< M, N1 > H1={}, OptionalJacobian< M, N2 > H2={})
Matrix< RealScalar, Dynamic, Dynamic > M
EIGEN_DEVICE_FUNC EIGEN_STRONG_INLINE const Scalar * data() const
A matrix or vector expression mapping an existing array of data.
Eigen::Matrix< double, M, N1, Eigen::RowMajor > RowMajor1
Eigen::Matrix< double, N1, 1 > Vector1
static ConjugateGradientParameters parameters
Eigen::Matrix< double, N2, 1 > Vector2
Special class for optional Jacobian arguments.
The matrix class, also used for vectors and row-vectors.
Eigen::Matrix< double, M, N2, Eigen::RowMajor > RowMajor2