10 #ifndef EIGEN_MATRIX_FUNCTION 11 #define EIGEN_MATRIX_FUNCTION 29 template <
typename MatrixType>
34 typedef typename MatrixType::Scalar
Scalar;
46 MatrixType
compute(
const MatrixType&
A);
52 template <
typename MatrixType>
57 const MatrixType
N = MatrixType::Identity(rows, rows) - A;
58 VectorType e = VectorType::Ones(rows);
59 N.template triangularView<Upper>().solveInPlace(e);
60 return e.cwiseAbs().maxCoeff();
63 template <
typename MatrixType>
69 Index rows = A.rows();
71 MatrixType Ashifted = A - avgEival * MatrixType::Identity(rows, rows);
73 MatrixType F =
m_f(avgEival, 0) * MatrixType::Identity(rows, rows);
74 MatrixType P = Ashifted;
76 for (Index
s = 1;
s < 1.1 * rows + 10;
s++) {
77 Fincr =
m_f(avgEival, static_cast<int>(
s)) * P;
79 P =
Scalar(RealScalar(1.0/(
s + 1))) * P * Ashifted;
82 const RealScalar F_norm = F.cwiseAbs().rowwise().sum().maxCoeff();
83 const RealScalar Fincr_norm = Fincr.cwiseAbs().rowwise().sum().maxCoeff();
86 RealScalar rfactorial = 1;
87 for (Index r = 0; r < rows; r++) {
89 for (Index i = 0; i < rows; i++)
92 rfactorial *= RealScalar(r);
93 delta = (
std::max)(delta, mx / rfactorial);
95 const RealScalar P_norm = P.cwiseAbs().rowwise().sum().maxCoeff();
108 template <
typename Index,
typename ListOfClusters>
111 typename std::list<Index>::iterator j;
112 for (
typename ListOfClusters::iterator i = clusters.begin(); i != clusters.end(); ++i) {
113 j = std::find(i->begin(), i->end(), key);
117 return clusters.end();
131 template <
typename EivalsType,
typename Cluster>
135 typedef typename EivalsType::RealScalar RealScalar;
136 for (Index i=0; i<eivals.rows(); ++i) {
139 if (qi == clusters.end()) {
142 clusters.push_back(l);
148 for (Index j=i+1; j<eivals.rows(); ++j) {
149 if (
abs(eivals(j) - eivals(i)) <= RealScalar(matrix_function_separation)
150 && std::find(qi->begin(), qi->end(), j) == qi->end()) {
152 if (qj == clusters.end()) {
155 qi->insert(qi->end(), qj->begin(), qj->end());
164 template <
typename ListOfClusters,
typename Index>
167 const Index numClusters =
static_cast<Index>(clusters.size());
168 clusterSize.
setZero(numClusters);
169 Index clusterIndex = 0;
170 for (
typename ListOfClusters::const_iterator cluster = clusters.begin(); cluster != clusters.end(); ++cluster) {
171 clusterSize[clusterIndex] = cluster->size();
177 template <
typename VectorType>
180 blockStart.resize(clusterSize.rows());
183 blockStart(i) = blockStart(i-1) + clusterSize(i-1);
188 template <
typename EivalsType,
typename ListOfClusters,
typename VectorType>
192 eivalToCluster.resize(eivals.rows());
193 Index clusterIndex = 0;
194 for (
typename ListOfClusters::const_iterator cluster = clusters.begin(); cluster != clusters.end(); ++cluster) {
195 for (Index i = 0; i < eivals.rows(); ++i) {
196 if (std::find(cluster->begin(), cluster->end(), i) != cluster->end()) {
197 eivalToCluster[i] = clusterIndex;
205 template <
typename DynVectorType,
typename VectorType>
209 DynVectorType indexNextEntry = blockStart;
210 permutation.resize(eivalToCluster.rows());
211 for (Index i = 0; i < eivalToCluster.rows(); i++) {
212 Index cluster = eivalToCluster[i];
213 permutation[i] = indexNextEntry[cluster];
214 ++indexNextEntry[cluster];
219 template <
typename VectorType,
typename MatrixType>
223 for (Index i = 0; i < permutation.rows() - 1; i++) {
225 for (j = i; j < permutation.rows(); j++) {
226 if (permutation(j) == i)
break;
229 for (Index k = j-1; k >= i; k--) {
231 rotation.
makeGivens(T(k, k+1), T(k+1, k+1) - T(k, k));
232 T.applyOnTheLeft(k, k+1, rotation.
adjoint());
233 T.applyOnTheRight(k, k+1, rotation);
234 U.applyOnTheRight(k, k+1, rotation);
235 std::swap(permutation.coeffRef(k), permutation.coeffRef(k+1));
246 template <
typename MatrixType,
typename AtomicType,
typename VectorType>
249 fT.setZero(T.rows(), T.cols());
251 fT.block(blockStart(i), blockStart(i), clusterSize(i), clusterSize(i))
252 = atomic.compute(T.block(blockStart(i), blockStart(i), clusterSize(i), clusterSize(i)));
278 template <
typename MatrixType>
289 typedef typename MatrixType::Scalar
Scalar;
295 for (Index i = m - 1; i >= 0; --i) {
296 for (Index j = 0; j < n; ++j) {
316 X(i,j) = (C(i,j) - AX - XB) / (A(i,i) + B(j,j));
328 template <
typename MatrixType,
typename VectorType>
332 typedef typename MatrixType::Scalar
Scalar;
334 static const int RowsAtCompileTime = Traits::RowsAtCompileTime;
335 static const int ColsAtCompileTime = Traits::ColsAtCompileTime;
336 static const int Options = MatrixType::Options;
339 for (Index k = 1; k < clusterSize.rows(); k++) {
340 for (Index i = 0; i < clusterSize.rows() - k; i++) {
342 DynMatrixType
A = T.block(blockStart(i), blockStart(i), clusterSize(i), clusterSize(i));
343 DynMatrixType B = -T.block(blockStart(i+k), blockStart(i+k), clusterSize(i+k), clusterSize(i+k));
344 DynMatrixType C = fT.block(blockStart(i), blockStart(i), clusterSize(i), clusterSize(i))
345 * T.block(blockStart(i), blockStart(i+k), clusterSize(i), clusterSize(i+k));
346 C -= T.block(blockStart(i), blockStart(i+k), clusterSize(i), clusterSize(i+k))
347 * fT.block(blockStart(i+k), blockStart(i+k), clusterSize(i+k), clusterSize(i+k));
348 for (Index m = i + 1; m < i + k; m++) {
349 C += fT.block(blockStart(i), blockStart(m), clusterSize(i), clusterSize(m))
350 * T.block(blockStart(m), blockStart(i+k), clusterSize(m), clusterSize(i+k));
351 C -= T.block(blockStart(i), blockStart(m), clusterSize(i), clusterSize(m))
352 * fT.block(blockStart(m), blockStart(i+k), clusterSize(m), clusterSize(i+k));
354 fT.block(blockStart(i), blockStart(i+k), clusterSize(i), clusterSize(i+k))
375 template <typename MatrixType, int IsComplex = NumTraits<typename internal::traits<MatrixType>::Scalar>::IsComplex>
388 template <
typename AtomicType,
typename ResultType>
389 static void run(
const MatrixType&
A, AtomicType& atomic, ResultType &result);
398 template <
typename MatrixType>
401 template <
typename MatA,
typename AtomicType,
typename ResultType>
402 static void run(
const MatA&
A, AtomicType& atomic, ResultType &result)
405 typedef typename Traits::Scalar
Scalar;
406 static const int Rows = Traits::RowsAtCompileTime, Cols = Traits::ColsAtCompileTime;
407 static const int MaxRows = Traits::MaxRowsAtCompileTime, MaxCols = Traits::MaxColsAtCompileTime;
409 typedef std::complex<Scalar> ComplexScalar;
412 ComplexMatrix CA = A.template cast<ComplexScalar>();
413 ComplexMatrix Cresult;
415 result = Cresult.real();
422 template <
typename MatrixType>
425 template <
typename MatA,
typename AtomicType,
typename ResultType>
426 static void run(
const MatA&
A, AtomicType& atomic, ResultType &result)
432 MatrixType T = schurOfA.
matrixT();
433 MatrixType U = schurOfA.
matrixU();
436 std::list<std::list<Index> > clusters;
462 result = U * (fT.template triangularView<Upper>() * U.adjoint());
502 template <
typename ResultType>
503 inline void evalTo(ResultType& result)
const 508 static const int RowsAtCompileTime = Traits::RowsAtCompileTime;
509 static const int ColsAtCompileTime = Traits::ColsAtCompileTime;
510 typedef std::complex<typename NumTraits<Scalar>::Real> ComplexScalar;
514 AtomicType atomic(
m_f);
519 Index
rows()
const {
return m_A.rows(); }
520 Index
cols()
const {
return m_A.cols(); }
528 template<
typename Derived>
539 template <
typename Derived>
546 template <
typename Derived>
554 template <
typename Derived>
562 template <
typename Derived>
570 template <
typename Derived>
580 #endif // EIGEN_MATRIX_FUNCTION Derived::PlainObject ReturnType
void matrix_function_compute_permutation(const DynVectorType &blockStart, const DynVectorType &eivalToCluster, VectorType &permutation)
Compute permutation which groups ei'vals in same cluster together.
void matrix_function_compute_map(const EivalsType &eivals, const ListOfClusters &clusters, VectorType &eivalToCluster)
Compute mapping of eigenvalue indices to cluster indices.
std::complex< typename NumTraits< Scalar >::Real > ComplexScalar
EIGEN_DEVICE_FUNC Matrix< _Scalar, _Rows, _Cols, _Options, _MaxRows, _MaxCols > & setZero(Index size)
MatrixType::Scalar Scalar
void makeGivens(const Scalar &p, const Scalar &q, Scalar *z=0)
MatrixType compute(const MatrixType &A)
Compute matrix function of atomic matrix.
internal::ref_selector< Derived >::type DerivedNested
static int f(const TensorMap< Tensor< int, 3 > > &tensor)
static void run(const MatA &A, AtomicType &atomic, ResultType &result)
Rotation given by a cosine-sine pair.
Holds information about the various numeric (i.e. scalar) types allowed by Eigen. ...
EIGEN_DEVICE_FUNC EIGEN_STRONG_INLINE const AbsReturnType abs() const
static void run(const MatA &A, AtomicType &atomic, ResultType &result)
const MatrixFunctionReturnValue< Derived > sinh() const
Proxy for the matrix function of some matrix (expression).
NumTraits< typename MatrixType::Scalar >::Real matrix_function_compute_mu(const MatrixType &A)
const MatrixFunctionReturnValue< Derived > cos() const
EIGEN_STRONG_INLINE EIGEN_DEVICE_FUNC half() max(const half &a, const half &b)
void matrix_function_compute_block_start(const VectorType &clusterSize, VectorType &blockStart)
Compute start of each block using clusterSize.
MatrixFunctionReturnValue(const Derived &A, StemFunction f)
Constructor.
void matrix_function_compute_block_atomic(const MatrixType &T, AtomicType &atomic, const VectorType &blockStart, const VectorType &clusterSize, MatrixType &fT)
Compute block diagonal part of matrix function.
void matrix_function_partition_eigenvalues(const EivalsType &eivals, std::list< Cluster > &clusters)
Partition eigenvalues in clusters of ei'vals close to each other.
ListOfClusters::iterator matrix_function_find_cluster(Index key, ListOfClusters &clusters)
Find cluster in clusters containing some value.
const MatrixFunctionReturnValue< Derived > cosh() const
EIGEN_DEFAULT_DENSE_INDEX_TYPE Index
The Index type as used for the API.
Class for computing matrix functions.
JacobiRotation adjoint() const
const ComplexMatrixType & matrixT() const
Returns the triangular matrix in the Schur decomposition.
void matrix_function_compute_cluster_size(const ListOfClusters &clusters, Matrix< Index, Dynamic, 1 > &clusterSize)
Compute size of each cluster given a partitioning.
MatrixFunctionAtomic(StemFunction f)
Constructor.
void matrix_function_permute_schur(VectorType &permutation, MatrixType &U, MatrixType &T)
Permute Schur decomposition in U and T according to permutation.
internal::stem_function< Scalar >::type StemFunction
void evalTo(ResultType &result) const
Compute the matrix function.
static const float matrix_function_separation
Maximum distance allowed between eigenvalues to be considered "close".
void matrix_function_compute_above_diagonal(const MatrixType &T, const VectorType &blockStart, const VectorType &clusterSize, MatrixType &fT)
Compute part of matrix function above block diagonal.
static void run(const MatrixType &A, AtomicType &atomic, ResultType &result)
Compute the matrix function.
Helper class for computing matrix functions of atomic matrices.
const MatrixFunctionReturnValue< Derived > matrixFunction(StemFunction f) const
stem_function< Scalar >::type StemFunction
The matrix class, also used for vectors and row-vectors.
void run(Expr &expr, Dev &dev)
MatrixType matrix_function_solve_triangular_sylvester(const MatrixType &A, const MatrixType &B, const MatrixType &C)
Solve a triangular Sylvester equation AX + XB = C.
const ComplexMatrixType & matrixU() const
Returns the unitary matrix in the Schur decomposition.
void swap(mpfr::mpreal &x, mpfr::mpreal &y)
const MatrixFunctionReturnValue< Derived > sin() const